Sign Up to BioTechniques free email alert service to receive content updates.
Multiplexed fluid array screening of phage displayed anti-ricin single domain antibodies for rapid assessment of specificity
 
George P. Anderson1, Russell Matney1, Jinny L. Liu1, Andrew Hayhurst2, Ellen R. Goldman1
1, U.S. Naval Research Laboratory, Washington, DC
2, Southwest Foundation for Biomedical Research, San Antonio, TX, USA
BioTechniques, Vol. 43, No. 6, December 2007, pp. 806–811
Full Text (PDF)
Abstract

Phage display is a well-known technique that facilitates the selection of peptides or proteins that bind to a desired target. Using this tool, binding elements contained in the natural immune repertoires of the source animal or from a synthetically generated collection may be selected. The unpaired variable domain of the llama's heavy-chain-only classes of immunoglobulins represents an ideal source of genetic material to create phage display libraries. Initial panning of a semi-synthetic llama library yielded only one binder to the toxin ricin. In an effort to increase the number of monoclonal phage binders selected, the Luminex xMAP technology (Luminex, Austin, TX, USA) was used in addition to the enzyme-linked immunosorbent assay (ELISA) to screen clonal populations of phage after three rounds of selection. The xMAP technology detected phage displayed single domain antibody (sdAb) bound to ricin immobilized on the surface of microspheres under equilibrium conditions. This enhanced capability led directly to the identification of additional single domain antibodies of interest. The selected sdAbs were expressed, purified, and then evaluated for their specificity as well as enhanced thermal stability in comparison to conventional immunoglobulin G (IgG). We determined equilibrium dissociation constants and demonstrated their use as effective capture molecules in sandwich immunoassays.

Introduction

A variety of methods have been developed for the presentation and subsequent selection of binding elements from a library of candidates. These include techniques such as phage display (1,2,3), yeast display (4,5), Escherichia coli display (6), baculovirus/insect cell display (7), and ribosome display (8). While all useful, our group chose phage display, a common method for selection of binders from single domain antibody (sdAb) libraries (9,10,11,12,13). sdAbs, the smallest functional antibody binding unit, originated with the discovery that the entire Camelidae family possesses unique classes of heavy chain immunoglobulins (14). These antibodies, a result of a fortuitous genetic mutation, lack light chains. The variable domain of the heavy chain has several altered surface amino acids to increase its hydophilicity to compensate for the lack of the light chain and forms effective high-affinity binding sites for a wide variety of target molecules.

Many sdAbs have been derived using phage display libraries created from immunized camels or llamas, which allow the animal to enrich the phage pool for high-affinity binders (11,13). Others have focused on creating naive phage display libraries from which it is theoretically possible to extract binders for a nearly boundless number of targets (15,16); however, naive libraries may possess only low-affinity binders. While most applications require high-affinity binders, lower affinity binders may provide the starting point for affinity maturation. Affinity maturation has resulted in the attainment of binders having hundreds-to thousands-fold higher affinity than the starting antibody fragment (17,18). Unfortunately, identifying low-affinity clones using the standard enzyme-linked immunosorbent assay (ELISA) methodology can be highly problematic; since numerous wash steps are involved, low-affinity binders can be easily lost. Thus, the ideal selection method would avoid wash steps and allow one to evaluate ligand-target interactions under equilibrium binding conditions. This would increase the likelihood of isolating sdAbs that possess the selectivity and stability desired, while absolute affinity could be addressed downstream. Recently, approaches that offer this avoidance of the washing steps required in the ELISA format have been demonstrated; namely, the use of surface plasmon resonance and standard flow cytometry (19,20). The method we chose to explore is a multiplexed flow cytometric approach using the Luminex 100 platform and xMAP technology. Using this specialized system, 100 different microsphere sets at a time can simultaneously be evaluated, thereby allowing us to rapidly examine the various phage display sdAb for their binding to our target, ricin, as well as to monitor for nonspecific binding to a number of nontarget proteins, in a high-throughput manner.

Materials and Methods

Materials

Ricin, ricin A chain, and ricin B chain were from Vector (Burlingame, CA, USA), cholera toxin (CTX) from Calbiochem (San Diego, CA, USA), and staphylococcal enterotoxin B (SEB) from Toxin Technology (Sarasota, FL, USA). Phosphate-buffered saline (PBS) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Polyclonal rabbit anti-ricin antibody and mouse monoclonal antibodies specific for ricin (MAb Ric-03-A-G1, MAb Ric-07-A-G1) were the kind gifts of Robert Bull (Naval Medical Research Center, Silver Spring, MD, USA). The anti-M13 antibody was purchased from GE Healthcare (Piscataway, NJ, USA). This antibody was fluorescently labeled using bis functional NHS-Cy3 ester (GE Healthcare) as described by the manufacturer.

ELISA and xMAP Monoclonal Phage Selections

The sdAb library was constructed and panned for ricin binders as described previously (9). The phage display vector is based upon an ampicillin-resistant derivative of the low-level expression phagemid pAK100 (9,21,22) with sdAb fused to full-length g3p with no intervening amber codon. Individual colonies of E. coli were randomly selected following the third round of panning. These were then grown in wells of a microtiter plate; afterwards crude phage preparations were prepared in the same wells. Meanwhile, wells of high-binding ELISA plates were coated overnight with ricin or 3 µg/mL bovine serum albumin (BSA). The next morning, wells were blocked with PBS with 2% low-fat dry milk for 1 h, then each phage was added to both target-and BSA-coated wells and allowed to incubate for 1 h. Excess phage was washed off with PBS with 0.5% Tween-20 (PBST) and the binding was detected using an anti-M13-horseradish peroxidase (HRP) conjugate (GE Healthcare). Excess conjugate was washed off with PBST/PBS, and HRP activity was determined by colorimetric o-phenylenediamine dihydrochloride (OPD) substrate (Sigma-Aldrich) ((Figure 1), lower panel).

  1    2    3    4  



Back to top