Picomolar Inhibitors as Transition-State Probes of 5 = -Methylthioadenosine Nucleosidases Jemy A. Gutierrez †,¶ , Minkui Luo †,¶ , Vipender Singh † , Lei Li † , Rosemary L. Brown ‡ , Gillian E. Norris ‡ , Gary B. Evans § , Richard H. Furneaux § , Peter C. Tyler § , Gavin F. Painter § , Dirk H. Lenz § , and Vern L. Schramm †, * † Department of Biochemistry, Albert Einstein College of Medicine, Bronx, New York, 10461, ‡ Institute of Molecular Biosciences, Massey University, Private Bag 11222, Palmerston North, New Zealand, and § Carbohydrate Chemistry Team, Industrial Research Ltd., Lower Hutt, New Zealand. ¶ Contributed equally to this work. E nzymes accelerate reactions by lowering the ac- tivation barriers for the formation of transition states. Transition-state theory suggests that the catalytic acceleration provided by an enzyme is propor- tional to the free energy released upon binding of transition-state analogues ( 1, 2). Stable transition-state analogue inhibitors that ideally mimic geometric and electrostatic features of a transition state are proposed to bind to the enzyme more tightly than substrate by a factor approaching the enzyme-imposed rate accelera- tion. Most enzymes have substrate Michaelis constant (K m ) values in the millimolar to micromolar range and catalytic accelerations on the order of 10 10 -10 15 . Therefore, perfect transition-state analogues are ex- pected to show dissociation constants ( K d ) of 10 -14 - 10 -23 M(3). The elucidation of transition-state struc- tures therefore provides a blueprint for generating po- tent, tight-binding inhibitors. Conversely, probing en- zymes with structurally diverse transition-state ana- logues can be a powerful tool to differentiate closely re- lated isozymes by revealing the variations at their transi- tion states. Isozymes are now known to exhibit charac- teristic transition-state structures, and comparing the structural differences among these transition states can provide valuable information for designing target- specific inhibitors ( 4 –7). Enzymatic transition states are difficult to probe be- cause of their short lifetimes, the complications of en- zyme matrix, and rate-limiting steps that are not in- volved with bond breaking or formation. The use of linear free energy relationships with an array of sub- *Corresponding author, vern@aecom.yu.edu. Received for review August 6, 2007 and accepted October 5, 2007. Published online November 16, 2007 10.1021/cb700166z CCC: $37.00 © 2007 American Chemical Society ABSTRACT Transition states can be predicted from an enzyme’s affinity to re- lated transition-state analogues. 5=-Methylthioadenosine nucleosidases (MTANs) are involved in bacterial quorum sensing pathways and thus are targets for anti- bacterial drug design. The transition-state characteristics of six MTANs are com- pared by analyzing dissociation constants (K d ) with a small array of representa- tive transition-state analogues. These inhibitors mimic early or late dissociative transition states with K d values in the picomolar range. Our results indicate that the K d ratio for mimics of early and late transition states are useful in distinguishing between these states. By this criterion, the transition states of Neisseria meningiti- des and Helicobacter pylori MTANs are early dissociative, whereas Escherichia coli, Staphylococcus aureus, Streptococcus pneumoniae, and Klebsiella pneu- moniae MTANs have late dissociative characters. This conclusion is confirmed in- dependently by the characteristic [1=- 3 H] and [1=- 14 C] kinetic isotope effects (KIEs) of these enzymes. Large [1=- 3 H] and unity [1=- 14 C] KIEs are observed for late disso- ciative transition states, whereas early dissociative states showed close-to-unity [1=- 3 H] and significant [1=- 14 C] KIEs. K d values of various MTANs for individual transition-state analogues provide tentative information about transition-state structures due to varying catalytic efficiencies of enzymes. Comparing K d ratios for mimics of early and late transition states removes limitations inherent to the en- zyme and provides a better predictive tool in discriminating between possible transition-state structures. A RTICLE www.acschemicalbiology.org VOL.2 NO.11 • ACS CHEMICAL BIOLOGY 725