Dissecting Antibodies with Regards to Linear and Conformational Epitopes

March Dissecting Antibodies with Regards to Linear and Conformational Epitopes Bjrn Forsstrm 0 1 Barbara Bisawska Axns 0 1 Johan Rockberg 0 1 Hanna Danielsson 0 1 Anna Bohlin 0 1 Mathias Uhlen 0 1 0 1 Science for Life Laboratory, KTH-Royal Institute of Technology , SE-171 21 Stockholm , Sweden , 2 Department of Proteomics, School of Biotechnology, AlbaNova University Center, Royal Institute of Technology (KTH) , Stockholm , Sweden 1 Academic Editor: Nicholas J Mantis, New York State Dept. Health , UNITED STATES An important issue for the performance and specificity of an antibody is the nature of the binding to its protein target, including if the recognition involves linear or conformational epitopes. Here, we dissect polyclonal sera by creating epitope-specific antibody fractions using a combination of epitope mapping and an affinity capture approach involving both synthesized peptides and recombinant protein fragments. This allowed us to study the relative amounts of antibodies to linear and conformational epitopes in the polyclonal sera as well as the ability of each antibody-fraction to detect its target protein in Western blot assays. The majority of the analyzed polyclonal sera were found to have most of the targetspecific antibodies directed towards linear epitopes and these were in many cases giving Western blot bands of correct molecular weight. In contrast, many of the antibodies towards conformational epitopes did not bind their target proteins in the Western blot assays. The results from this work have given us insights regarding the nature of the antibody response generated by immunization with recombinant protein fragments and has demonstrated the advantage of using antibodies recognizing linear epitopes for immunoassay involving wholly or partially denatured protein targets. - Competing Interests: The authors have declared that no competing interests exist. Antibodies are invaluable tools for the study of proteins and antibody-based methods are widely used both in research and diagnostics [1]. In addition, antibodies have become one of the largest growing fields in therapeutics with a large number of antibody biopharmaceuticals introduced into the clinic during the last years for treatment of e.g. cancer, autoimmunity and inflammation [2,3]. The number of publicly available antibodies is growing with a fast pace as exemplified by the fact that there are more than a million antibodies available towards human targets as part of various antibody listing sites, such as the Antibodypedia portal [4] www. antibodypedia.com. A key to understanding the varying performance of antibodies is to determine the binding sites, or epitopes, they recognize. Epitopes are generally divided in two categories, linear epitopes where a stretch of continuous amino acids are sufficient for binding and conformational epitopes where key amino acid residues are brought together by protein folding [5]. Conformational epitopes might be preferred for applications involving protein targets in their native state, such as therapeutic applications or flow cytometry. On the other hand, linear epitopes might be preferred for applications in which the protein target is wholly or partially denatured during the sample preparation prior to the immuno assay, such as in Western blot (WB), immunohistochemistry (IHC) or immunofluorescence-based confocal microscopy. The choice of immunogen used for immunization can greatly influence the performance of the resulting antibodies. Immunogens vary from short peptides of only 1020 amino acids coupled to carrier proteins [7,8], longer protein fragments [9] and up to using the entire full-length proteins [10]. Using full-length proteins for generation of antibodies has the drawback of possibly creating antibodies that are cross-reactive to other proteins sharing sequence similarities with the protein target. Using short peptides or protein fragments can be a way to circumvent this problem, since they can be chosen to cover a unique part of the amino acids sequence. However, anti-peptide antibodies often lack the ability to bind the native proteins, due to the unstructured nature of the peptide [11]. An attractive alternative is therefore to generate antibodies towards protein fragments covering 50150 amino acids of a unique sequence region of the target protein as compared to other proteins from this species [9]. This strategy of using recombinant protein fragments, referred to as Protein Epitope Signature Tags (PrESTs), as immunogens have been used to generate more than 16,000 validated polyclonal antibodies towards human protein targets within the framework of the Human Protein Atlas (www. proteinatlas.org) project [12,13]. The aim of this study was to characterize polyclonal sera from immunizations with the above-mentioned recombinant protein fragments to determine if this strategy evokes an antibody response targeting mainly linear or conformational epitopes. Here, we describe how epitope-specific fractionation of polyclonal sera based on epitope mapping and affinity capture on chromatography columns can be used to determine the ratio of antibodies targeting linear and conformational epitopes. This approach also allowed us to investigate the performance of antibody fractions targeting linear and conformational epitopes in the most frequently used immunoassay in life science, namely Western blot. The software PRESTIGE [14] was used to design protein fragments, 95149 amino acids long, with low sequence similarity to other proteins. Gene fragments were amplified from a pool of human RNA and cloned into an Escherichia coli vector. The antigens were expressed as fusions of a His6-ABP tag and the protein fragments, purified on IMAC columns, validated and used for immunization of New Zeeland White rabbits as described elsewhere [15]. N-terminally biotin-tagged overlapping synthetic 15-mer peptides with five amino acids lateral shift (PEPscreen, SigmaAldrich, St Louis, MO) covering the antigen sequences were dissolved in 80% DMSO to a concentration of 10 mg/ml. An aliquot of 2 L was diluted in 198 L PBS-B (1PBS, 1% BSA, pH 7.4) and the rest was stored at -80C until further use. Color-coded microsphere (MagPlex Microspheres, Luminex-Corp., Austin, TX) previously coated with Neutravidin (Pierce, Rockford, IL), according to the manufacturers instructions, were distributed in a half-area 96-well plate (approx. 19000 per bead-ID) and diluted peptides were added to a final concentration of 50 g/mL in 100 L PBS-B. The plate was sealed and incubated for 1 hour on a plate shaker in the dark and then washed three times with 100 L PBS-T (1PBS, 0.1% Tween20, pH 7.4) using a magnetic microplate washer (BioTek, Winooski, VT) before resuspending the beads in 50 L PBS-T. Re-suspended beads with peptides corresponding to the same antigens were pooled to create the different bead arrays and they were stored in storage buffer (BRE, Blocking Reagent for ELISA, Roche supp (...truncated)