Bound Tris confounds the identification of binding site residues in a paraquat single chain antibody

We produced an anti-paraquat single chain antibody (scFv) to investigate its potential use in immunotherapy for paraquat poisoning. However, this scFv was expressed in an insoluble form and only displayed moderate binding affinity. An earlier examination of the pH dependence of antigen binding by the parent paraquat-specific mAb (7D7-3) suggested that the electrostatic effects of a tyrosine residue were important. The aims of the current study were to obtain expression of a soluble scFv (D10) and to increase its binding affinity. The former was achieved by expression in a phagemid vector. Site-directed mutagenesis of tyrosine residues in CDR H3 did not result in improved affinity for paraquat, suggesting that the original pH dependence required re-examination. Nuclear magnetic resonance studies of 7D7-3 Fab revealed that the original observation of the pH-dependent paraquat binding with a mid-point of ~pH 8.9 was due to tightly bound Tris. It appears that as Tris is titrated to a neutral species the energetically unfavourable juxtaposition of its positive charge with that of paraquat is reduced. These findings have broad implications in the interpretation of the pH or salt dependence of any antibody-antigen interaction which should be made cautiously and with regard to the possible interference of buffer components introduced during the preparation of the antibody