Surface plasmon resonance (SPR), using BIAcore2000 instruments, is utilized by our laboratory for studying protein-ligand interactions. For many years our laboratory has studied the interaction between antigens and antibodies and we have developed a novel experimental design using the surface plasmon resonance to resolve time-dependent kinetic behavior which is consistent with the formation of a complex, first as an encounter complex and changing into a more stable docked complex, evidencing at least two-states. This is the same model which recently has been proposed for time-dependent binding of the 90kDa heat shock protein (Hsp90) to inhibitors of the genaldenamycin (GA) family. Our previous experience with the antibody-antigen system therefore informs the study of new, uncharacterized systems such as Hsp90-GA interaction. The unique protocols we have developed in the long-term project on antibody-protein interactions will be the basis of several new initiatives addressing targeted Receptor-Ligand interactions important to cancer biology. The development of reliable methodology to time-dependent residency times is important to understanding pharmacological profiles of drugs many drugs. We have begun experiments to examine the time-dependent complex formation using SPR, and to develop general methodology for identifying time-dependent binding in other drug-target interactions. The first initiative is the use of SPR to identify time-dependent drug-target interactions leading to long residence times and selectvity of inhibitors. In vitro drug-target interaction assays, classically quantified in terms of binding parameters such as IC50 or Kd., do not always predict in vivo efficicacy. Hsp90 interaction with the natural product GA is an example, where cellular activity is much higher than in vitro affinities would predict. Hsp90 is emerging as an important target for chemotherapeutic agents. The GA binds to Hsp90, inhibits its ATPase activity, and decreases levels of many of its client proteins implicated in cancer cell survival. Derivatives of this natural product are currently in multiple clinical trials. Although binding assays show affinities of the analogues to be in the micromolar range, they exhibit cellular effects in low nanomolar ranges. Time-dependent formation of a tight complex is one recently proposed mechanism to explain this phenomena, and may account for the accumulation of the drug in tissues and its pharmacological effects (Gooljarsingh et al, Proc natl Acad Sci 103:7625,2006). In parallel experiments, we have recently shown, using our model antibody system, that a mutant antigen with apparently lower binding affinity is actually a better inhibitor in SPR of antibody binding than the original antigen when examined in a long term competitive binding assay: binding inhibition correlates with length of preincubation time with the inhibitor. We have developed unique protocols for measuring binding kinetics and thermodynamics and thermodynamics whichshow that the underlying mechanism is the slow, time-dependent formation of a very tight complex by the mutant inhibitor. This is the first reported demonstration of time-dependent inhibition efficicacy. We are planning experiments to also demonstrate this by isothermal titration calorimetry (ITC), which will require the development of additional novel ITC protocols. This experimental design can now be applied to drug-target interactions such as the Hsp90-GS interaction. The time-dependence of binding leading to long residence time of such inhibitors is likely to significantly impact their pharmalogical profile in patients, such as pharmacological effect and target selectivity. Thus development of reliable methodology to study this characteristic is likely of high impact A second new initiative is study of the interaction of anti-tubulin drugs with tubulin. Microtubule-targeted peptides and depsipeptides are receiving considerable attention as possible anti-cancer agents, and new synthetic analogues have fewer side effects than natural compounds such as dolstatin-15 isolated from marine organisms. They inhibit binding of vinblastin to tubulin, and bind to what has been termed the Vinca domain of tubulin. Extensive biochemical, cellular, and whole animal structure activity relationship (SAR) data for dolastatin-10 and dolastatin-15, in combination with modeling studies, have led to a preliminary model of the dolastatin binding site to the Vinca domain on tubulin (Hamel & Covell, Curr. Med. Chem. 2:19,2002). Potent cytotoxicity has not been observed in the absence of a strong tubulin-binding activity. However, the absence of a quantitative binding assay has limited the precision with which a pharmacaphore can be defined. SPR has high potential for investigation of tubulin-based assays. The compound SJ-81, a synthetic analogue of dolstatin-10, has been shown to inhibit tubulin polymerization in cellular and biochemical assays (Tarasova & Hamel, personal communication). Attachment of a linker to the amino terminus does not interfere with binding. Previous attempts to develop an SPR tubulin-binding assay, by capturing tubulin with anti-tubulin and using SJ-81 as the analyte did not yield results (personal communication, E. Hamel & R.Fischer). A maleimidocapronic linker has been has covalently attached to the amino terminus of SJ-81. This can be readily covalently coupled to a cysteine-derived surface Biacore chip, producing a stable interaction surface for binding assays. By covalently coupling the stable drug to the chip and using tubulin as the analyte, we can better control the tubulin state, and binding of the larger analyte will be more readily observable by the SPR technology. New methodology for determining binding kinetics and thermodynamics will be developed as necessary to study to biological and physical factors controlling tubulin polymerization. Tubulin protein will be provided by E. Hamel (SRB,TP, NCI) whose expertise will also inform the experimental design of the SPR experiements. Development of a quantitative binding assay will provide a basis for more detailed SAR studies, characterization of the ligand binding site(s) on tubulin, and more precise definition of a pharmacaphore. Understanding the details of the interaction may also give insight into a better experimental design for screening multiple drug interactions with immobilized tubulin. Finally, it will serve as a prototype for investigation of the molecular details of novel anti-tubulin agents with tubulin.
