Water channel in the binding site of a high affinity anti-methotrexate antibody.

In the present study, we report the structure of the free and drug-bound Fab fragment of a high affinity anti-methotrexate antibody and perform a thermodynamic analysis of the binding process. The anti-methotrexate Fab fragment features a remarkably rigid tunnel-like binding site that extends into a water channel serving as a specialized route to move solvent out and into the site upon ligand binding and dissociation. This new finding in antibody structure-function relationships directly relates to the fast association (1 × 107 M⁻¹ s⁻¹) and slow dissociation (4 × 10⁻5 s⁻¹) rates determined for mAb ADD056, resulting in a very strong binding with a K(D) ~ 3.6 pM at 20 °C. As follows from the X-ray data analysis, the methotrexate-antibody complex is stabilized by an extended network of hydrogen bonds and stacking interactions. The analysis also shows structural involvement of the CDR H3 in formation of the water channel revealing another important role of this hypervariable region. This suggests a new direction in natural affinity maturation and opens a new possibility in antibody engineering. Methotrexate is a widely used therapeutic agent for many malignant diseases and inflammatory disorders. Unfortunately, it may also interfere with central aspects of metabolism and thereby cause inevitable side effects. Therefore, methotrexate therapy requires careful monitoring of drug blood levels, which is traditionally done by immunoassays. An understanding of the structure-function properties of antibodies selected for drug monitoring substantiates the performance and robustness of such tests.

Supporting immunoassay design with biophysical tools.

In this article, we demonstrate how the application of biophysical tools facilitates the design of robust immunoassays. The binding characteristics of the reagents used in an immunoassay determine the assay response to the analyte concentrations. We applied several biophysical methods to obtain pertinent equilibrium and kinetic coefficients and used this information in the design of a microparticle-based immunoassay for detection of neutrophil gelatinase-associated lipocalin (NGAL), which is a new diagnostic marker of acute kidney injury (AKI). We characterized the conformational stability of recombinant human NGAL and the solution phase binding properties of six monoclonal antibodies. A preferred antibody pair was selected on the basis of the affinities of the antibodies and their sandwich pairing capabilities. One of the antibodies was coated on magnetic microparticles, and the second antibody was conjugated with a reporter group. The apparent kinetic rates of the immobilized and conjugated antibodies were measured and used to compute the assay calibration plot for the target concentration range of the analyte at specific incubation times. The experimental assay results were found to be in good agreement with the computed data, confirming that applying biophysical tools provides a solid foundation for immunoassay design and optimization.

Development of an Abbott ARCHITECT cyclosporine immunoassay without metabolite cross-reactivity.

OBJECTIVE: We investigated the mechanism by which the ARCHITECT cyclosporine (CsA) chemiluminescent microparticle immunoassay (CMIA) eliminates cross-reactivity to CsA metabolites AM1 and AM9, despite its use of a monoclonal antibody which shows cross-reactivity in fluorescence polarization immunoassays. DESIGN AND METHODS: The CMIA was accomplished by incubating an extracted blood sample with magnetic microparticles coated with a very low amount of anti-CsA antibody. After a wash step the microparticles were incubated with a chemiluminescent CsA tracer, followed by a second wash step and measurement of chemiluminescence. The reagent concentrations of salt and detergent were optimized to maximize CsA binding and minimize metabolite interference. RESULTS: Elimination of CsA metabolite cross-reactivity was shown using purified metabolites and blood samples containing native CsA metabolites. The CMIA demonstrated precision and sensitivity acceptable for use in a clinical setting. CONCLUSION: We conclude that it is possible to eliminate CsA metabolite immuno-cross-reactivity by careful assay design.

Rapid determination of antigenic epitopes in human NGAL using NMR.

The recent remarkable rise in biomedical applications of antibodies and their recombinant constructs has shifted the interest in determination of antigenic epitopes in target proteins from the areas of protein science and molecular immunology to the vast fields of modern biotechnology. In this article, we demonstrated that measuring binding induced changes in two-dimensional NMR spectra enables rapid determination of antibody binding footprints on target protein antigens. Such epitopes recognized by six high-affinity monoclonal murine antibodies (mAbs) against human neutrophil gelatinase-associated lipocalin (NGAL) were determined by measuring chemical shifts or broadening of peaks in (1)H-(15)N-TROSY HSQC and (1)H-(13)C HSQC spectra of isotope-labeled NGAL occurring upon its binding to the antibodies. Locations of the epitopes defined by the NMR studies are in good agreement with the results of antibody binding pairing observed by dual-color fluorescence cross-correlation spectroscopy. In all six cases, the antibodies recognize conformational epitopes in regions of relatively rigid structure on the protein. None of the antibodies interact with the more flexible funnel-like opening of the NGAL calyx. All determined epitope areas in NGAL reflect the dimensions of respective antibody binding surface (paratopes) and contain amino acid residues that provide strong interactions. This NMR-based approach offers comprehensive information on antigenic epitopes and can be applied to numerous protein targets of diagnostic or therapeutic interest.

Crystal structure and thermodynamic analysis of diagnostic mAb 106.3 complexed with BNP 5-13 (C10A).

B-type natriuretic peptide (BNP) is a naturally secreted regulatory hormone that influences blood pressure and vascular water retention in human physiology. The plasma BNP concentration is a clinically recognized biomarker for various cardiovascular diseases. Quantitative detection of BNP can be achieved in immunoassays using the high-affinity monoclonal IgG1 antibody 106.3, which binds an epitope spanning residues 5-13 of the mature bioactive peptide. To understand the structural basis of this molecular recognition, we crystallized the Fab fragment complexed with the peptide epitope and determined the three-dimensional structure by X-ray diffraction to 2.1 A resolution. The structure reveals the detailed interactions that five of the complementarity-determining regions make with the partially folded peptide. Thermodynamic measurements using fluorescence spectroscopy suggest that the interaction is enthalpy driven, with an overall change in free energy of binding, DeltaG = -54 kJ/mol, at room temperature. The parameters are interpreted on the basis of the structural information. The kinetics of binding suggest a diffusion-limited mechanism, whereby the peptide easily adopts a bound conformation upon interaction with the antibody. Moreover, comparative analysis with alanine-scanning results of the epitope explains the basis of selectivity for BNP over other related natriuretic peptides.

Interactions of two monoclonal antibodies with BNP: high resolution epitope mapping using fluorescence correlation spectroscopy.

Structure-function studies of antibody-antigen systems include the identification of amino acid residues in the antigen that interact with an antibody and elucidation of their individual contributions to binding affinity. We used fluorescence correlation spectroscopy (FCS) and alanine-scanning mutagenesis to characterize the interactions of brain natriuretic peptide (BNP) with two monoclonal antibodies. Human BNP is a 32 amino acid residue long cyclic polypeptide with the ring structure confined between cysteines in positions 10 and 26. It is an important cardiovascular hormone and a valuable diagnostic cardiac marker. We compare the binding strength of the N-terminus Alexa488-labeled BNP, native cyclic BNP, BNP alanine-substituted mutants, linear BNP, and its short fragments to determine the individual contributions of amino acid residues included in the continuous antigenic epitopes that are recognized by two different monoclonal antibodies raised toward BNP. Implementation of FCS for these studies offers all of the advantages of solution phase measurements, including high sensitivity, simplicity of manipulation with reagents, and elimination of solid phase interferences or separation steps. Significant differences in the molecular masses of the free and antibody bound BNP results in a substantial ( approximately 2.5-times) increase in the diffusion rates. Determination of the binding constants and inhibition effects by measuring the diffusion rates of the ligand at the single molecule level introduces the ultimate opportunity for researching systems where the fluorescence intensity and/or fluorescence anisotropy do not change upon interaction of the ligand with the protein. Monoclonal antibodies 106.3 and BC203 demonstrate high affinities to BNP and bind two distant epitopes forming robust antibody sandwiches. Both antibodies are used in Abbott diagnostic assays on AxSYM, IMx, and Architect platforms.