Identification of Keratinocyte-specific Markers Using Phage Display and Mass Spectrometry
2003; Elsevier BV; Volume: 2; Issue: 2 Linguagem: Inglês
10.1074/mcp.m200049-mcp200
ISSN1535-9484
AutoresKim B. Jensen, Ole N. Jensen, Peter Ravn, Brian F.C. Clark, Peter Kristensen,
Tópico(s)Protein purification and stability
ResumoSpecific molecular markers for various normal and pathogenic cell states and cell types provide knowledge of basic biological systems and have a direct application in targeted therapy. We describe a proteomic method based on the combination of new and improved phage display antibody technologies and mass spectrometry that allows identification of cell type-specific protein markers. The most important features of the method are (i) reduction of experimental noise originating from background binding of phage particles and (ii) isolation of affinity binders after a single round of selection, which assures a high diversity of binders. The method demonstrates, for the first time, the ability to detect, identify, and analyze both secreted and membrane-associated extracellular proteins as well as a variety of different cellular structures including proteins and carbohydrates. The optimized phage display method was applied to analysis of human skin keratinocytes resulting in the isolation of a panel of antibodies. Fourteen of these antibodies were further characterized, half of which predominantly recognized keratinocytes in a screen of a range of different cell types. Three cognate keratinocyte antigens were subsequently identified by mass spectrometry as laminin-5, plectin, and fibronectin. The combination of phage display technology with mass spectrometry methods for protein identification is a general and promising approach for proteomic analysis of cell surface complexity. Specific molecular markers for various normal and pathogenic cell states and cell types provide knowledge of basic biological systems and have a direct application in targeted therapy. We describe a proteomic method based on the combination of new and improved phage display antibody technologies and mass spectrometry that allows identification of cell type-specific protein markers. The most important features of the method are (i) reduction of experimental noise originating from background binding of phage particles and (ii) isolation of affinity binders after a single round of selection, which assures a high diversity of binders. The method demonstrates, for the first time, the ability to detect, identify, and analyze both secreted and membrane-associated extracellular proteins as well as a variety of different cellular structures including proteins and carbohydrates. The optimized phage display method was applied to analysis of human skin keratinocytes resulting in the isolation of a panel of antibodies. Fourteen of these antibodies were further characterized, half of which predominantly recognized keratinocytes in a screen of a range of different cell types. Three cognate keratinocyte antigens were subsequently identified by mass spectrometry as laminin-5, plectin, and fibronectin. The combination of phage display technology with mass spectrometry methods for protein identification is a general and promising approach for proteomic analysis of cell surface complexity. The epidermis has very important functions such as protecting the organism against environmental hazards, e.g. microbes and stress, and keeping fluids inside the body. The main cell type found in the epidermis is the keratinocyte, which makes up more than 80% of the cells (1.Holbrook K.A. Ultrastructure of the Epidermis. 1st Ed. Cambridge University Press, Cambridge, UK1994Google Scholar). Although the understanding of keratinocyte proliferation and differentiation is increasing, additional markers for single cell states are always required (2.Watt F.M. Stem cell fate and patterning in mammalian epidermis.Curr. Opin. Genet. Dev. 2001; 11: 410-417Google Scholar). A variety of different methods, e.g. antibody technology (3.Shusta E.V. Boado R.J. Pardridge W.M. Vascular proteomics and subtractive antibody expression cloning.Mol. Cell. Proteomics. 2002; 1: 75-82Google Scholar), two-dimensional PAGE, mass spectrometry (4.Gygi S.P. Rist B. Aebersold R. Measuring gene expression by quantitative proteome analysis.Curr. Opin. Biotechnol. 2000; 11: 396-401Google Scholar), DNA arrays (5.Diehn M. Eisen M.B. Botstein D. Brown P.O. Large-scale identification of secreted and membrane-associated gene products using DNA microarrays.Nat. Genet. 2000; 25: 58-62Google Scholar), and protein arrays (6.Zhu H. Bilgin M. Bangham R. Hall D. Casamayor A. Bertone P. Lan N. Jansen R. Bidlingmaier S. Houfek T. Mitchell T. Miller P. Dean R.A. Gerstein M. Snyder M. Global analysis of protein activities using proteome chips.Science. 2001; 293: 2101-2105Google Scholar), have has been used either alone or in combination to analyze protein expression. However, the majority of these methods have difficulties analyzing high molecular weight, transmembrane, and extracellular proteins as well as posttranslational modifications such as glycosylations and proteolytically processed proteins. Increased knowledge of these components is vital for understanding biological processes such as those involved in cell communication. Furthermore, an intrinsic problem of these methods is that the average cellular distribution in a cell population is analyzed, hindering the investigation of minor protein fractions with subtle localization. Analysis of cell-specific protein expression using conventional antibody technology has utilized immunization with preparations of the cell type or tissue and concomitant analysis of antibody specificities (3.Shusta E.V. Boado R.J. Pardridge W.M. Vascular proteomics and subtractive antibody expression cloning.Mol. Cell. Proteomics. 2002; 1: 75-82Google Scholar). Recently phage display has played an important role in the generation of such cell type-specific antibodies and the ensuing identification of cell markers (for reviews, see Refs. 7.Winter G. Griffiths A.D. Hawkins R.E. Hoogenboom H.R. Making antibodies by phage display technology.Annu. Rev. Immunol. 1994; 12: 433-455Google Scholar, 8.Hoogenboom H.R. de Bruine A.P. Hufton S.E. Hoet R.M. Arends J.W. Roovers R.C. Antibody phage display technology and its applications.Immunotechnology. 1998; 4: 1-20Google Scholar, 9.Kristensen P. Ravn P. Jensen K.B. Jensen K.H. Applying phage display technology in aging research.Biogerontology. 2000; 1: 67-78Google Scholar). Several approaches have been reported for identification of cell markers such as selection using activated cell sorting (10.de Kruif J. Terstappen L. Boel E. Logtenberg T. Rapid selection of cell subpopulation-specific human monoclonal antibodies from a synthetic phage antibody library.Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3938-3942Google Scholar, 11.Sawyer C. Embleton J. Dean C. Methodology for selection of human antibodies to membrane proteins from a phage-display library.J. Immunol. Methods. 1997; 204: 193-203Google Scholar), selection directly on cells in suspension (12.Marks J.D. Ouwehand W.H. Bye J.M. Finnern R. Gorick B.D. Voak D. Thorpe S.J. Hughes Jones N.C. Winter G. Human antibody fragments specific for human blood group antigens from a phage display library.Biotechnology (N. Y.). 1993; 11: 1145-1149Google Scholar), selections directly on adherent cells (13.Cai X. Garen A. Anti-melanoma antibodies from melanoma patients immunized with genetically modified autologous tumor cells: selection of specific antibodies from single-chain Fv fusion phage libraries.Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 6537-6541Google Scholar, 14.Pereira S. Van Belle P. Elder D. Maruyama H. Jacob L. Sivanandham M. Wallack M. Siegel D. Herlyn D. Combinatorial antibodies against human malignant melanoma.Hybridoma. 1997; 16: 11-16Google Scholar), and selection on tumor tissue sections (15.Tordsson J. Abrahmsen L. Kalland T. Ljung C. Ingvar C. Brodin T. Efficient selection of scFv antibody phage by adsorption to in situ expressed antigens in tissue sections.J. Immunol. Methods. 1997; 210: 11-23Google Scholar). Recently another approach has been described enabling the identification of cell surface components capable of internalizing binding ligands (16.Becerril B. Poul M.A. Marks J.D. Toward selection of internalizing antibodies from phage libraries.Biochem. Biophys. Res. Commun. 1999; 255: 386-393Google Scholar). In the present study we applied a protease-sensitive helper phage (17.Kristensen P. Winter G. Proteolytic selection for protein folding using filamentous bacteriophages.Fold. Des. 1998; 3: 321-328Google Scholar) in cell surface selections on cultured human keratinocytes. The protease-sensitive helper phage has previously been used successfully in the generation of phage display-derived antibodies by increasing the fraction of target-specific antibodies (18.Ravn P. Kjær S. Jensen K.H. Wind T. Jensen K.B. Kristensen P. Brosh R.M. Orren D.K. Bohr V.A. Clark B.F. Identification of phage antibodies toward the Werner protein by selection on Western blots.Electrophoresis. 2000; 21: 509-516Google Scholar, 19.Goletz S. Christensen P.A. Kristensen P. Blohm D. Tomlinson I. Winter G. Karsten U. Selection of large diversities of antiidiotypic antibody fragments by phage display.J. Mol. Biol. 2002; 315: 1087-1097Google Scholar, 20.de Wildt R.M. Mundy C.R. Gorick B.D. Tomlinson I.M. Antibody arrays for high-throughput screening of antibody-antigen interactions.Nat. Biotechnol. 2000; 18: 989-994Google Scholar). The antibody repertoires used in the present study are conceptually different, namely a semisynthetic repertoire, the Griffin library, 1www.mrc-cpe.cam.ac.uk/winter-hp.php?menu=1808. and two single framework repertoires, the Tomlinson I and Tomlinson J (20.de Wildt R.M. Mundy C.R. Gorick B.D. Tomlinson I.M. Antibody arrays for high-throughput screening of antibody-antigen interactions.Nat. Biotechnol. 2000; 18: 989-994Google Scholar). One of the most important features of the established method is that it enabled screening of monoclonal phage antibodies after a single round of selection. This ensured a high diversity of binding antibodies, some of which recognized markers specifically produced by the target. These markers were either membrane-associated or soluble. Other antibodies had cross-reactivity to epitopes presented by other cell types. Three antigens were subsequently isolated, and their identities were determined using mass spectrometry. This study demonstrates the potential of applying a new helper phage in complex selections since it reduces the number of antibody-displaying phage retrieved nonspecifically. Furthermore, it exemplifies the powerful combination of mass spectrometry and phage antibody technology in analyses of cell type-specific expression profiles. Primary human keratinocytes were isolated from mammary tissue and cultured according to Norsgaard et al. (22.Norsgaard H. Glenting M. Kristensen P. Clark B.F.C. Rattan S.I.S. Ageing, differentiation and apoptosis in human epidermal keratinocytes in culture.Gerontol. Geriatr. Suppl. Arch. 1994; 4: 185-196Google Scholar). Cells other than keratinocytes were cultured in Dulbecco's modified Eagle's medium (BioWhittaker) containing 10% fetal calf serum, 1 mm glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin. Furthermore, the cells were incubated at 37 °C in 5% CO2 and 99% humidity. The amber codon suppressing Escherichia coli strain TG1 (supE hsdD5Δ(lac-proAB)thi F′ {traD36 pro AB+ lacIq lacZΔ}) was used for amplification of phage displaying antibody fragments. The Griffin repertoire as well as the Tomlinson I and J repertoires are described elsewhere (20.de Wildt R.M. Mundy C.R. Gorick B.D. Tomlinson I.M. Antibody arrays for high-throughput screening of antibody-antigen interactions.Nat. Biotechnol. 2000; 18: 989-994Google Scholar). 1www.mrc-cpe.cam.ac.uk/winter-hp.php?menu=1808. Antibody repertoires were rescued either with the protease-sensitive helper phage KM13 or the traditional helper phage M13K07 (Stratagene) as described in Refs. 17.Kristensen P. Winter G. Proteolytic selection for protein folding using filamentous bacteriophages.Fold. Des. 1998; 3: 321-328Google Scholar and 18.Ravn P. Kjær S. Jensen K.H. Wind T. Jensen K.B. Kristensen P. Brosh R.M. Orren D.K. Bohr V.A. Clark B.F. Identification of phage antibodies toward the Werner protein by selection on Western blots.Electrophoresis. 2000; 21: 509-516Google Scholar. The selections with KM13-rescued repertoires were performed on cultured keratinocytes in six-well plates (Costar). A total number of 2 × 105 keratinocytes were added to each well 1 day prior to selection to obtain 80% confluence. After overnight incubation in a cell incubator (37 °C, 99% humidity, and 5% CO2) the wells were blocked with 2% MPBS 2The abbreviations used are: MPBS, PBS supplemented with low fat milk powder; PBS, phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; HPLC, high pressure liquid chromatography; MALDI, matrix-assisted laser desorption ionization; TRITC, tetramethylrhodamine isothiocyanate. (PBS supplemented with 2% (w/v) low fat milk powder) for 2 h at room temperature. Meanwhile phage antibody repertoires were separately incubated with 50:50 volume % 4% MPBS and keratinocyte medium (Invitrogen). After blocking the wells were briefly rinsed with PBS and incubated with phage repertoire for 1.5 h at room temperature in 3 ml of 50:50 volume % 4% MPBS and keratinocyte medium. The wells were then washed six times for 5 min with PBS. Bound phage were specifically eluted using trypsin (1 mg/ml in 1 mm CaCl2, 50 mm Tris-HCl, pH 7.4). Trypsin proteolyses the Myc tag that separates the single chain Fv and protein III as well as the protease-sensitive linker in KM13-encoded protein III. A log-phase TG1 culture was infected with eluted phage and plated on TYE (10 g/liter peptone, 5 g/liter yeast extract, 8 g/liter NaCl, 15 g/liter agar) plates containing ampicillin (200 μg/ml) and glucose (1%). The selection with the Griffin repertoire was propagated for a second round of panning by rescue with helper phage KM13 as described in Refs. 18.Ravn P. Kjær S. Jensen K.H. Wind T. Jensen K.B. Kristensen P. Brosh R.M. Orren D.K. Bohr V.A. Clark B.F. Identification of phage antibodies toward the Werner protein by selection on Western blots.Electrophoresis. 2000; 21: 509-516Google Scholar and 23.Marks J.D. Hoogenboom H.R. Bonnert T.P. McCafferty J. Griffiths A.D. Winter G. By-passing immunization. Human antibodies from V-gene libraries displayed on phage.J. Mol. Biol. 1991; 222: 581-597Google Scholar. The control selection with the M13K07 (Stratagene)-rescued Griffin repertoire was performed using the conditions described above and in Marks et al. (23.Marks J.D. Hoogenboom H.R. Bonnert T.P. McCafferty J. Griffiths A.D. Winter G. By-passing immunization. Human antibodies from V-gene libraries displayed on phage.J. Mol. Biol. 1991; 222: 581-597Google Scholar) except for elution of phage with 100 mm triethylamine and subsequent neutralization with 1 m Tris-HCl, pH 7.5 before infection into log-phase TG1. Monoclonal phage antibody clones were rescued overnight with KM13 for ELISA (18.Ravn P. Kjær S. Jensen K.H. Wind T. Jensen K.B. Kristensen P. Brosh R.M. Orren D.K. Bohr V.A. Clark B.F. Identification of phage antibodies toward the Werner protein by selection on Western blots.Electrophoresis. 2000; 21: 509-516Google Scholar). Afterward cultures were cleared by centrifugation, and phage were precipitated by addition of 1:5 volume of 20% polyethylene glycol 6000, 2.5 m NaCl. Phage were pelleted by centrifugation and resuspended in 1:1 volume of PBS. ELISA was performed on the indicated cell type by culturing overnight in 96-well plates (5000 cells/well, Costar) with the appropriate medium. Cell culture medium, either conditioned or not conditioned by cell culturing, was analyzed by overnight coating in ELISA plates (MAXI-sorp™, Nunc, Roskilde, Denmark). Next wells were incubated with 2% MPBS for 2 h before addition of 5 × 1010 monoclonal phage in 100 μl of 2% MPBS and incubation for 1 h. After phage incubation, the wells were washed six times in PBS, and bound phage were detected with a horseradish peroxidase-conjugated mouse anti-phage antibody (Amersham Biosciences). ELISAs were developed with o-phenylenediamine tablets (Dako) according to the manufacturer's instructions. Antibodies were cloned into the novel expression vector pKBJ1 to allow expression of active and soluble antibody (24.Jensen K.B. Larsen M. Pedersen J.S. Christensen P.A. Alvarez-Vallina L. Goletz S. Clark B.F. Kristensen P. Functional improvement of antibody fragments using a novel phage coat protein III fusion system.Biochem. Biophys. Res. Commun. 2002; 298: 566-573Google Scholar). Briefly, the vector directs the production of antibodies fused to the N terminus of filamentous coat protein III domain I-II to the periplasmic space of the bacteria. Fusion to protein III domain I-II functionally rescues otherwise inactive phage display-derived antibodies. Medium was obtained from 80% confluent culture flasks after 2 days of culturing and cleared from cell material by centrifugation (3000 × g, 20 min). Conditioned medium was dialyzed in low salt buffer (50 mm Tris-HCl, pH 8.0, 50 mm NaCl) and subjected to Q-Sepharose fractionation applying a gradient of NaCl. Ten microliters of each fraction were coated in ELISA wells and tested for the presence of antigen. Next antigen-containing fractions were concentrated by centrifugation in Centricon 3 (Amicon Bioseparations) and subjected to gel filtration on an TSK-gel G3000 SW column (ToSoHaas) using HPLC (Biotek Instruments). Positive fractions from ion-exchange chromatography were pooled and dialyzed in acetic acid buffer (50 mm sodium acetate, pH 4.8, 50 mm NaCl) leading to protein precipitation. The resulting pellet was resuspended in 50 mm Tris-HCl, pH 8.0, 100 mm NaCl. Confluent keratinocyte cultures were metabolically labeled overnight with a 1:1000 dilution of a Redivue PRO-MIX 35S cell labeling mix (Amersham Biosciences) in keratinocyte medium. After labeling, cultures were washed twice with PBS and incubated with antibody clones 12 and B3, which had been biotinylated using an ECL™ protein biotinylation module (Amersham Biosciences) according to the manufacturer's instructions. Next antigen-antibody complexes were precipitated by incubation with PBS supplemented with 0.5% Triton X-100 before incubation with 1.7 × 107 streptavidin-coated Dyna™ beads (Dynal). Subsequently beads were washed, and bound protein was analyzed using SDS-PAGE. Proteins were isolated by SDS-PAGE. Protein bands were excised and treated with dithiothreitol, S-alkylated with iodoacetamide, and then incubated with trypsin (sequencing grade, modified porcine trypsin, 12 ng/μl, Promega, Madison, WI) as described previously (25.Shevchenko A. Wilm M. Vorm O. Mann M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels.Anal. Chem. 1996; 68: 850-858Google Scholar). Peptides were extracted, dried, and redissolved in 20 μl of 5% formic acid prior to mass analysis. Peptide mixtures were analyzed using a REFLEX MALDI reflector time-of-flight mass spectrometer (Bruker-Daltonics, Bremen, Germany) operated in the positive ion delayed extraction mode. Peptide mixtures were desalted and concentrated using custom-made nanoscale columns (26.Gobom J. Nordhoff E. Mirgorodskaya E. Ekman R. Roepstorff P. Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry.J. Mass Spectrom. 1999; 34: 105-116Google Scholar) and deposited onto the MALDI target by elution with the matrix solution. Peptide mass spectra were mass calibrated by using trypsin autolysis peptide signals (m/z 842.51 and 2211.10) generating a mass error of less than 50 ppm for all peptides (27.Jensen O.N. Podtelejnikov A. Mann M. Delayed extraction improves specificity in database searches by matrix-assisted laser desorption/ionization peptide maps.Rapid Commun. Mass Spectrom. 1996; 10: 1371-1378Google Scholar). Proteins were identified via their peptide mass maps by using the ProFound search engine (Genomic Solutions) and the Mascot search engine (Matrix Science Ltd., London, UK). ELISA was performed on purified recombinant laminin-5, which was a kind gift from Dr. Kaoru Miyazaki, Yokohama City University (28.Kariya Y. Ishida K. Tsubota Y. Nakashima Y. Hirosaki T. Ogawa T. Miyazaki K. Efficient expression system of human recombinant laminin-5.J. Biochem. (Tokyo). 2002; 132: 607-612Google Scholar), and purified human fibronectin (Sigma). One microgram of purified protein was coated overnight at 4 °C in 50 mm NaHCO3, pH 9.6. Phage-displayed antibodies D4 and 12 were incubated with the wells, and the ELISA was performed as described above for monoclonal phage ELISAs. Indirect immunofluorescence studies were performed on keratinocytes fixated and permeabilized using PBS supplemented with 1% formaldehyde and 0.1% Tween 20 for 5 min at room temperature. Residual binding to the glass surfaces of wells was blocked with 2% MPBS before incubating with ∼10 μg/ml biotinylated antibody B3 in the presence of 2% MPBS. Next biotinylated antibodies were detected using a 1:400 dilution of fluorescein isothiocyanate-conjugated Extravidin® (Sigma). A commercially available monoclonal mouse anti-plectin antibody (Sigma) was applied in a 1:100 dilution for colocalization. The murine antibody was subsequently detected with a rabbit anti-mouse antibody before incubation with a TRITC-conjugated swine anti-rabbit antibody (Dako). Between each incubation step and before addition of fluorescent mounting medium (Dako) slides were washed four times in PBS supplemented with 0.05% Tween 20 and four times in PBS. Cells were analyzed by confocal laser scanning microscopy using a Leica DM IRBE microscope equipped with TCS system. Images were subsequently prepared using Adobe Photoshop 5.0®. The variable genes from individual clones were amplified by PCR and sequenced as described elsewhere (18.Ravn P. Kjær S. Jensen K.H. Wind T. Jensen K.B. Kristensen P. Brosh R.M. Orren D.K. Bohr V.A. Clark B.F. Identification of phage antibodies toward the Werner protein by selection on Western blots.Electrophoresis. 2000; 21: 509-516Google Scholar). The gene segments were identified using the VBASE directory at the Medical Research Council, Cambridge, UK. 3www.mrc-cpe.cam.ac.uk/DNAPLOT.php?menu=901. Human keratinocytes were chosen as the target for several reasons. First, they constitute a substantial part of our largest organ, the human skin, and second, only limited information exists on the membrane-associated contents. Third, primary cultures of keratinocytes have not undergone any transformational changes as is the case for the cell lines used in previous studies with phage display in direct selections (13.Cai X. Garen A. Anti-melanoma antibodies from melanoma patients immunized with genetically modified autologous tumor cells: selection of specific antibodies from single-chain Fv fusion phage libraries.Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 6537-6541Google Scholar, 14.Pereira S. Van Belle P. Elder D. Maruyama H. Jacob L. Sivanandham M. Wallack M. Siegel D. Herlyn D. Combinatorial antibodies against human malignant melanoma.Hybridoma. 1997; 16: 11-16Google Scholar, 30.Figini M. Obici L. Mezzanzanica D. Griffiths A. Colnaghi M.I. Winter G. Canevari S. Panning phage antibody libraries on cells: isolation of human Fab fragments against ovarian carcinoma using guided selection.Cancer Res. 1998; 58: 991-996Google Scholar, 31.Meulemans E.V. Slobbe R. Wasterval P. Ramaekers F.C. van Eys G.J. Selection of phage-displayed antibodies specific for a cytoskeletal antigen by competitive elution with a monoclonal antibody.J. Mol. Biol. 1994; 244: 353-360Google Scholar, 32.Kupsch J.M. Tidman N.H. Kang N.V. Truman H. Hamilton S. Patel N. Newton Bishop J.A. Leigh I.M. Crowe J.S. Isolation of human tumor-specific antibodies by selection of an antibody phage library on melanoma cells.Clin. Cancer Res. 1999; 5: 925-931Google Scholar). Thus, the cultured cells do to some extend reflect the situation in the skin. Fourth, keratinocytes are grown in a defined medium without the addition of fetal calf serum thereby avoiding selection of antibodies recognizing serum proteins, which is observed when traditional selections are performed on cells grown in medium supplemented with fetal calf serum. 4P. Kristensen, unpublished data. Furthermore, preliminary studies showed that it was not necessary to fix the cells before selection due to the adherent capacity of this particular cell type (data not shown); the antigen conformations consequently will not be modified by applied fixation procedures. Fixation prior to selection has been shown to result in a higher retrieval of specifically binding phage (33.Mutuberria R. Hoogenboom H.R. van der Linden E. de Bruine A.P. Roovers R.C. Model systems to study the parameters determining the success of phage antibody selections on complex antigens.J. Immunol. Methods. 1999; 231: 65-81Google Scholar), but the retrieved antibodies were of limited applicability (34.Roovers R.C. van der Linden E. de Bruine A.P. Arends J. Hoogenboom H.R. Identification of colon tumour-associated antigens by phage antibody selections on primary colorectal carcinoma.Eur. J. Cancer. 2001; 37: 542-549Google Scholar). Phage display using phagemid systems relies on the addition of structural proteins from a replicative unit, a helper phage needed for the production of phage particles, while phage are isolated on the basis of fusion coat protein IIIs encoded by phagemids. The helper phage KM13 (17.Kristensen P. Winter G. Proteolytic selection for protein folding using filamentous bacteriophages.Fold. Des. 1998; 3: 321-328Google Scholar) contains an engineered coat protein III in which a protease-sensitive linker is inserted between the domain anchoring the protein into the phage coat, domain III, and the two domains responsible for infection, domains I and II (35.Crissman J.W. Smith G.P. Gene-III protein of filamentous phages: evidence for a carboxyl-terminal domain with a role in morphogenesis.Virology. 1984; 132: 445-455Google Scholar). Thus, protease treatment renders helper phage-encoded coat protein IIIs non-functional with respect to infection, whereas phagemid-encoded fusion coat protein IIIs remain functional for infection. This is critical when phagemid repertoires are used in selections in general and in cell surface selections in particular because only a small fraction of phage in a phage pool contains a fusion coat protein III. The magnitude of this fraction very much depends on the type of repertoire. In a semisynthetic repertoire, such as the Griffin repertoire, ∼1/1000 phage display an antibody fragment, whereas the display level of single framework repertoires, like the Tomlinson I and J repertoires, is in the order of 1/10 to 1/100 when colony-forming units are determined for unselected repertoires before and after protease treatment as well as by Western blotting experiments with an antibody directed against filamentous phage protein III domain III (data not shown). Thus, between 90 and 99.9% of the phage present during a selection represent noise as these non-displaying phage still contain an antibody gene and therefore need to be analyzed if they are retrieved after a selection. This observation implies that the use of KM13 in selections will lower the background between 20- and 1000-fold when compared with a traditional selection because nonspecific phage are retrieved due to their presence and not based on a displayed antibody fragment. Accordingly the nonspecifically retrieved phage are a mixture of non-displaying and displaying phage representing the initial pool, and the majority are consequently removed by protease treatment. The high fraction of nonspecifically retrieved antibodies has been a considerable obstacle in cell surface selection (33.Mutuberria R. Hoogenboom H.R. van der Linden E. de Bruine A.P. Roovers R.C. Model systems to study the parameters determining the success of phage antibody selections on complex antigens.J. Immunol. Methods. 1999; 231: 65-81Google Scholar). Therefore, we set out to apply the protease-sensitive helper phage in cell surface selection. Furthermore, the presence of a Myc tag between the displayed antibody fragment and protein III in the fusion coat protein as well as the protease-sensitive linker in helper phage-encoded protein III enables elution of phage bound via antibodies and concomitant functional elimination of helper phage-encoded protein III (17.Kristensen P. Winter G. Proteolytic selection for protein folding using filamentous bacteriophages.Fold. Des. 1998; 3: 321-328Google Scholar, 36.Ørum H. Andersen P.S. Øster A. Johansen L.K. Riise E. Bj⊘rnvad M. Svendsen I. Engberg J. Efficient method for constructing comprehensive murine Fab antibody libraries displayed on phage.Nucleic Acids Res. 1993; 21: 4491-4498Google Scholar). To verify the benefits gained by using the novel helper phage KM13 over the traditionally used helper phage M13K07, two cell surface selections were performed in parallel with the Griffin repertoire. A number of antibodies were screened after both the first and second round of selection (Table I). It was observed that background antibodies were selected with M13K07, whereas KM13 directed the selections toward epitopes that were either cell-associated or secreted by keratinocytes. Screening showed that 6% and over 60% of the recovered phage were positive after one and two rounds of selection, respectively. Such high fractions of target-specific antibodies have previously only been reported after selections with phage vector-based repertoires (37.O'Connell D. Becerril B. Roy-Burman A. Daws M. Marks J.D. Phage versus phagemid libraries for generation of human monoclonal antibodies.J. Mol. Biol. 2002; 321: 49-56Google Scholar). However, suc
Referência(s)