Characterization of monoclonal antibodies to aflatoxin M1 and molecular modeling studies of related aflatoxins.

Aflatoxin M1 (AFM1) and seven structural analogs were used to investigate the correlation between antibody binding and the conformational and electronic properties of these molecules. Mice were immunized with AFM1-BSA and hybridomas secreting anti-AFM1 antibodies were isolated and characterized. The cross-reactivities of seven structurally similar aflatoxins were determined by competition enzyme-linked immunosorbent assay (cELISA). In an effort to correlate antibody binding with three-dimensional properties of the analogs, all of the aflatoxins (and the immunogen) were modeled and global energy minima were determined using molecular, mechanical and quantum mechanical methods. The results demonstrate that, for these molecules, loss of optimum structure and introduction of steric hindrance in the portion of the molecule that would fit into the antibody binding site are more important to binding than simply loss of a determinant group. Molecular computational techniques can give reasons for the wide variation in IC50 values observed between structural analogs and can be used as a tool for determining which conformational and electronic properties of molecules are most important for antibody binding.

Characterization of monoclonal anti-furosemide antibodies and molecular modeling studies of cross-reactive compounds.

Four mouse monoclonal antibodies directed against furosemide have been isolated and characterized. The cross-reactivity of the antibodies with eight compounds which are structurally and/or functionally related to furosemide was determined using a competition ELISA. All of the compounds, including furosemide, were then modeled using molecular mechanical and quantum mechanical methods in an attempt to correlate antibody binding with the conformational and electronic properties of the molecules. The results of these experiments demonstrated that all of the cross-reactivity observed could be readily explained using these techniques. Furthermore, these results should allow for more accurate prediction of unexpected cross-reactivities with these antibodies when they are used in immunoassays for determination of furosemide.

Immunoassays for des-Arg9-bradykinin.

Splenocytes from a female, BALB/c mouse immunized with bradykinin conjugated to ovalbumin with toluene diisocyanate were fused with mouse myeloma cells, X63/Ag8.653, using polyethylene glycol. Seventy-nine hybridomas were identified by ELISA to be making kinin reactive antibodies. In preliminary specificity studies it was determined that all of these hybridomas were producing antibodies more reactive with des-Arg9-bradykinin than with bradykinin. ELISAs were developed with the five clones that displayed the highest affinities for des-Arg9-bradykinin. Radioimmunoassays were developed for 3 of these 5 clones as well as with 5 monoclonal antibodies previously described (Odya and Lee 1990). The most sensitive des-Arg9-bradykinin assay developed was a radioimmunoassay in which carboxypeptidase B-treated [Tyr5]-bradykinin was the labeled antigen, clone OLNBK-5 was the antibody, and dextran-coated charcoal was used to separate bound from free radioactivity. The concentration of des-Arg9-bradykinin that inhibited 50% of the radioactive peptide binding was 0.08 +/- 0.03 nM. The relative specificity of this assay (des-Arg9-bradykinin = 100%) was: 29% bradykinin and about 1% with each of the following: lysyl-bradykinin, methionyl-lysyl-bradykinin, des-Arg1-bradykinin and des-Phe8-Arg9-bradykinin.

Analysis of hapten-carrier protein conjugates by nondenaturing gel electrophoresis.

Hapten-carrier protein conjugates were made using five different small haptens (MW < 1000), two carrier proteins and two methods of conjugation. Nondenaturing agarose gel electrophoresis was used to demonstrate that when as few as two molecules of a small hapten are attached to the carrier, the conjugate band migrates differently from that of the carrier alone or of the coupling reagent-treated carrier. Furthermore, the direction of the change in migration of each conjugate correlates with the change in charge which occurs upon attachment of the hapten to the carrier.

Elicitation and immunological characterization of monoclonal anti-idiotypic antibodies reactive with the ligand binding sites of monoclonal kinin antibodies.

Ten monoclonal anti-idiotypic antibodies (mAB2s) were obtained from fusions of myeloma cells, X63/Ag8.653, and splenocytes from mice immunized with one of two monoclonal kinin antibodies (mAB1s). The interactions of these mAB2s, with four different mAB1s, which have similar kinin binding specificities, was examined. Five of the ten mAB2s cross-reacted with similar affinities, with all four mAB1s. In addition, these five mAB2s were able to inhibit biotinylated-kallidin binding to the mAB1s. This indicated that these mAB2s interact with the mAB1s at, or near, their ligand binding sites. These immunological results are consistent with these five mAB2s being "internal image" beta type anti-idiotypic antibodies.

Monoclonal ligand binding site related anti-idiotypic antibodies elicited with a polyclonal kinin antibody.

Splenocytes from mice immunized with homogenous, polyclonal, rabbit kinin antibody (BK21) were fused using polyethylene glycol with the mouse myeloma cell line SP2/o. Eleven monoclonal antibodies, whose binding to BK21 could be inhibited by bradykinin, were obtained from 3 fusions. All of these anti-idiotypic antibodies were of the IgG1k isotype, except for one, which was an IgG2ak. An IgMk, auto-anti-idiotypic antibody, reactive with BK21 was obtained from a fusion of SP2/o cells and splenocytes from a mouse immunized with bradykinin conjugated with carbodiimide to keyhole limpet hemocyanin. Bradykinin could completely inhibit the binding of all of the anti-idiotypic antibodies to BK21 in an enzyme-linked immunosorbent assay. This result is consistent with the anti-idiotypic antibodies being reactive with the ligand binding sites of BK21. It was possible to separate the anti-idiotypic antibodies into 2 groups. The first group, 10 of the 12 antibodies tested, was more sensitive to inhibition by bradykinin than the second group and was not readily inhibited by des-Arg9-bradykinin. The second group was about 7 times more sensitive to inhibition by des-Arg9-bradykinin than by bradykinin. Further experiments will be needed to determine whether or not these monoclonal anti-idiotypic antibodies are "internal image" antibodies.