Parallel Immunizations of Rabbits Using the Same Antigen Yield Antibodies with Similar, but Not Identical, Epitopes

Epitopes. PLoS ONE 7(12): e45817. doi:10.1371/journal.pone.0045817 Parallel Immunizations of Rabbits Using the Same Antigen Yield Antibodies with Similar, but Not Identical, Epitopes Barbara Hjelm 0 Bjo rn Forsstro m 0 John Lo fblom 0 Johan Rockberg 0 Mathias Uhle n 0 Shan Lu, University of Massachusetts Medical Center, United States of America 0 1 School of Biotechnology, AlbaNova University Center, Royal Institute of Technology, Stockholm, Sweden, 2 Science for Life Laboratory, Royal Institute of Technology , Stockholm , Sweden A problem for the generation of polyclonal antibodies is the potential difficulties for obtaining a renewable resource due to batch-to-batch variations when the same antigen is immunized into several separate animals. Here, we have investigated this issue by determining the epitopes of antibodies generated from parallel immunizations of rabbits with recombinant antigens corresponding to ten human protein targets. The epitopes were mapped by both a suspension bead array approach using overlapping synthetic 15-mer peptides and a bacterial display approach using expression of random fragments of the antigen on the surface of bacteria. Both methods determined antibody binding with the aid of fluorescentbased analysis. In addition, one polyclonal antibody was fractionated by peptide-specific affinity capture for in-depth comparison of epitopes. The results show that the same antigen immunized in several rabbits yields polyclonal antibodies with similar epitopes, but with larger differences in the relative amounts of antibodies to the different epitopes. In some cases, unique epitopes were observed for one of the immunizations. The results suggest that polyclonal antibodies generated by repeated immunizations do not display an identical epitope pattern, although many of the epitopes are similar. - Funding: This work was supported by grants from the Knut and Alice Wallenberg Foundation, Swedish Research Council and the VINNOVA ProNova center grant. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. As antibodies have proven to be an exceptional tool to study the proteins of human biology and disease the need to create wellvalidated reagents of this kind is evident [1]. Several initiatives have started to generate antibodies and other affinity reagents in a systematic genome-wide manner, including the Human Protein Atlas project [2], the SH2-consortium [3] and the protein binder consortiums [4]. In addition, various commercial providers have generated several hundred thousands antibodies towards human proteins, and approximately 150,000 of these antibodies are listed in the community-based Antibodypedia portal [5]. The Human Protein Atlas [6] with information on more than 11,000 proteincoding genes, contains tissue profiles for more than 60 human cell types covering 48 tissues and organs, including liver, kidney, heart, different parts of the brain, the gastrointestinal tract etc. which are based on more than 14,000 commercially available antibodies. An important issue in this regard is the renewability of antibodies. Ideally, when results are obtained with well-validated antibodies, the reagent should be available to the scientific community indefinitely for further in-depth functional studies. This is the driving force for efforts trying to generate truly renewable affinity reagents, such as monoclonal antibodies using hybridoma cells or recombinant protein binders, such as antibodyfragments [7], scaffold binders [8] or nucleic acid based binders, such as aptamers and somamers [9]. However, more than 70% of the antibodies in Antibodypedia and 80% of the antibodies in the Human Protein Atlas are polyclonal antibodies [10]. Here, the limited availability of many polyclonal antibodies is a great concern, since there only exist a limited supply from the original immunization. The degree of renewability of polyclonal antibodies has been questioned, due to possible batch-to-batch variations when a follow-up immunization is done to generate new quantities of antibodies. In the diagnostic arena, this problem has been overcome by immunizing large animals, such as sheep or goat, to generate large quantities of antibodies. Alternatively, many animals, such as rabbits, are immunized with the same antigen and the sera from many animals are pooled to generate a large supply of antibodies with the same batch number. However, despite the frequent use of polyclonal antibodies, few studies have been performed in the past to estimate the degree of reproducibility when a new batch of polyclonal antibodies have been generated by immunization of a second animal. Recently, Larsson et al [11] used two recombinant antigens in repeated immunizations and determined the epitopes and immunohistochemistry staining patterns for the obtained antibodies. They concluded that all immunizations detected the correct band in Western blotting, but they rendered different staining patterns in IHC possibly related to their different epitope patterns. In the work by Geysen et al [12], the comparison of seven sera from outbred rabbits immunized with myohemerythrin, showed that no antibody specificity was common in all seven rabbits. Established methods for epitope mapping of antibodies involves chemical synthesis of peptides [13,14] or peptide display on phages [15,16]. Recently, we have described two independent methods for epitope mapping of antibodies [17], as schematically outlined in Figure 1. The first method relies on bacterial surface display on Staphylococcus carnosus in which the gene encoding the target protein is fragmented, cloned into an expression vector and subsequently introduced into S. carnosus host cells (Figure 1A). A library of bacterial cells is created, each member with a small fragment of the original gene expressed on the surface of the cell. The cells are incubated with the antibody to be mapped labeled with a fluorescent dye and the cells are analyzed in a flow cytometer so that cells expressing fragments bound by the antibody can be collected. These cells are grown, the insert of the expression vector DNA sequenced and the insert is mapped back to the original gene sequence. In this way, the amino acid sequence binding to the antibody can be mapped back in an efficient manner. The second method relies on suspension bead arrays with color-coded beads, in which each has a synthetic peptide bound to its surface (Figure 1C). The bead mixture with overlapping peptides spanning the whole antigen sequence is incubated with the fluorescently labeled antibody and the beads are analyzed on a flow sorter capable of identifying each color-coded bead. Both methods will give an apparent affinity of the binding to the corresponding epitope, i.e. a signal corresponding to the amount of bound antibody. However, it (...truncated)