The interaction between HIV-1 protease and inhibitors has been studied with optical biosensor technology. Optimized experimental procedures and mathematical analysis permitted determination of association and dissociation rate constants. A sensor surface with native enzyme was unstable and exhibited a drift that was influenced by the binding of inhibitor. This was hypothesized to be due to a specific mechanism involving autoproteolysis and/or dimer dissociation. The use of a mutant predicted to be less susceptible to autoproteolysis (Q7K) than wild-type enzyme resulted in a minor effect on surface stability, while a completely stable surface was obtained by treatment of the immobilized enzyme with N-ethyl-N'-(dimethylaminopropyl)-carbodiimide and N-hydroxysuccinimide; the most stable surface was achieved by chemically modifying the Q7K enzyme. The stabilized surface was enzymatically active and the interaction with inhibitors was similar to that for native enzyme. Several of the inhibitors had very high association rates, and estimation of kinetic constants was therefore performed with a binding equation accounting for limited mass transport. Of the clinical inhibitors studied, saquinavir had the highest affinity for the enzyme, a result of the lowest dissociation rate. Although the dissociation rate for ritonavir was sixfold faster, the affinity was only twofold lower than that for saquinavir since the association rate was threefold faster. Nelfinavir and indinavir exhibited lower affinities relative to the other inhibitors, a consequence of a slower association for nelfinavir and a relatively fast dissociation for indinavir. These results show that biosensor-based interaction studies can resolve affinity into association and dissociation rates, and that these are characteristic parameters for the interaction between enzymes and inhibitors.
Copyright 2001 Academic Press.