Transition state stabilization in enzyme catalysis: insights into mechanism from QM/MM modelling AdrianJ. Mulholland Centre for Computational Chemistry, School of Chemistry, University of Bristol, Bristol BS8 1TS, UK GST: Mutations and stereospecificity M2-2 isoenzyme produces only (S,S) product - why? Model effects of mutations (e.g. Ser209Ala, Asn8Asp and Thr13Ala represent differences between M1-1 and M2-2 GST isoenzymes) Calculate average changes in barrier and reaction energy through dynamics Asn8Asp mutation identified as determinant of stereospecificity (increases tbarrier for forming the (9R,10R) product more than barrier for (9S,10S) Agrees with the fact that the M2-2 isoenzyme preferentially forms (9S,10S) Also identified role for Tyr115 in TS stabilization: important for catalysis Simulations can help understand and predict effects of polymorphism on GST specificity and stereospecificity Introduction References Acknowledgements 'Insights into enzyme catalysis from QM/MM modelling: transition state stabilization in chorismate mutase' K.E. Ranaghan, L. Ridder, B. Szefczyk, W.A. Sokalski, J.C. Hermann & A.J. Mulholland, Mol. Phys., in press (2003). 'Identification of Glu166 as the General Base in the Acylation Reaction of Class A beta-Lactamases through QM/MM Modeling' Hermann, J. C.; Ridder, L.; Mulholland, A. J.; Holtje, H.-D.; J. Am. Chem. Soc.; in press (2003). 'Ab initio QM/MM modeling of the hydroxylation step in p-hydroxybenzoate hydroxylase' L. Ridder, J.N. Harvey, I. Rietjens, J. Vervoort & A.J. Mulholland, J. Phys. Chem. B, 107; 2118-2126 (2003). 'Modeling biotransformation reactions by combined quantum mechanical/molecular mechanical approaches: from structure to activity' L. Ridder & A.J. Mulholland, Current Topics in Medicinal Chemistry, 3, 1241-1256, (2003). 'QM/MM Free Energy Simulations of the Glutathione S-Transferase (M1-1) Reaction with Phenanthrene 9,10-Oxide' L. Ridder, I.Rietjens, J. Vervoort & A.J. Mulholland, J. Am. Chem. Soc., 124, 9926-9936 (2002). 'QM/MM Methods in Medicinal Chemistry', F. Perruccio, L. Ridder & A.J. Mulholland in 'Medicinal Quantum Chemistry', F. Alber & P. Carloni, Eds., Wiley, New York, (2003). 'The QM/MM Approach to Enzymatic Reactions' A.J. Mulholland, pp. 597-653 in 'Theoretical Biochemistry', L.Eriksson, Ed., Elsevier, Amsterdam, (2001). 'A QM/MM Study of the Hydroxylation of Phenol and Halogenated Derivatives by Phenol Hydroxylase' L. Ridder, A.J. Mulholland, I. Rietjens & J. Vervoort, J. Am. Chem. Soc.,122, 8728-8738 (2000). 'Ab Initio QM/MM Study of the Citrate Synthase Mechanism. A Low-Barrier Hydrogen Bond is Not Involved' A.J. Mulholland, P.D. Lyne & M. Karplus, J. Am. Chem. Soc., 122, 534-535 (2000). Role of Protein Flexibility in the Catalytic Cycle of p-Hydroxybenzoate Hydroxylase Elucidated by the Pro293Ser MutantB. A. Palfey, R. Basu, K.K. Frederick, K. B. Entsch, D.P. Ballou, Biochemistry, 41, 8438 (2002) 0 2 4 6 8 10 12 14 16 -3 -2 -1 0 1 2 3 Reaction coordinate (Angstrom) QM/MM stabilization energy (kcal/mol, AM1/CHARMM) TS -40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0 -2.5 -1.5 -0.5 0.5 1.5 Reaction coordinate (A) Energy (kcal/mol) Reaction in solution (MP2/6- 31+G(d)/PCM) Reaction in enzyme (MP2/6-31+G(d)//6- 31G(d) QM/MM) Substrate, catalytic residues treated by QM Glutathione S-Transferase (GST) Catalyses the addition of the tripeptide glutathione to endogenous and xenobiotic substrates Plays an important role in the detoxification and metabolism of xenobiotics (e.g. drugs) Exist as five classes (a, m, p, k, and q) within which various isoenzymes are present Epoxide ring-opening of phenanthrene 9,10-oxide in M1-1 isoenzyme Two possible alternative diastereomeric producs can be formed (R,R) and (S,S) An important model reaction for understanding determinants of stereospecificity in GSTs Two possible alternative diastereomeric producs can be formed (R,R) and (S,S) An important model reaction for understanding determinants of stereospecificity in GSTs Protein, solvent treated MM;interacts with QM Free energy barrier is ~20 kcal/mol, compares well with experimental activation energy of ~18 Kcal/mol M1-1 isoenzyme is not stereospecific: produces 50:50 mixture of (R,R) and (S,S) products (experiment again agrees with simulation findings) Combined quantum mechanics/molecular mechanics (QM/MM) methods allow modelling of reactions within enzymes Substrate and glutathione modelled QM with reaction-specific AM1 parameters Umbrella sampling MD along approximate reaction coordinate (r(C-O)-r(S-C)) WHAM (Weighted Histogram Analysis Method) used to give Free Energy Profile GST: QM/MM Dynamics Simulations Model transition states and intermediates; identify and analyse key interactions Calculations with ab initio, semiempirical or density functional QM Molecular dynamics simulations for free energy profiles Test and predict effects of mutations Test mechanistic proposals Small active site region treated QM (e.g 50 atoms) Surrounding protein and solvent (e.g. 25A radius) treated by MM (e.g. CHARMM) Recent applications demonstrate the useful insight QM/MM modelling can provide: glutathione transferase (GST), chorismate mutase, and para-hydroxybenzoate hydroxylase (PHBH) Tyr115 Ser209 Glutathionyl Tyr6 Thr13 Phenan- threne 9,10-oxide Tyr115 Ser209 Glutathionyl Tyr6 Thr13 Phenan- threne 9,10-oxide -20 -15 -10 -5 0 5 10 15 20 25 -3 -2 -1 0 1 Reaction coordinate (A) Free energy(kcal/mol) (9R,10R) (9S,10S) Successive simulations every 0.1A along reaction coordinate (e.g. 540ps total) AJM thanks his co-workers and collaborators in the work described here (see references below). AJM is an EPSRC Advanced Research Fellow, and thanks EPSRC, BBSRC, the WellcomeTrust, the Wolfson Trust, the Royal Society and the European Union for support of this research. Chorismate mutase GST active site (TS complex) Chorismate mutase active site (TS complex) Claisen rearrangement of chorismate to prephenate No covalent catalysis by the enzyme Substrate strain/ the 'NAC' effect in catalysis? Ab initio and semiempirical QM/MM modelling Comparison with reaction in solution 0 0.5 1 1.5 2 2.5 3 3.5 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 reaction coordinate (A) average number of H-bonds (9S,10S) (9R,10R) average GST: solvation of thiolate sulfur along reaction path Para-hydroxybenzoate hydroxylase (PHBH) Biodegradation of aromatic compounds (pollutants etc) Flavoprotein monoxygenase; hydroperoxyflavin intermediate Ab initio and semiempirical QM/MM Electrophilic aromatic substitution mechanism Wide range of halogenated substrates -10 -8 -6 -4 -2 0 2 4 -2 -1.5 -1 -0.5 0 0.5 1 1.5 reaction coordinate (A) energy (kcal/mol) Effect of whole protein Effect of Pro293 carbonyl -80 -60 -40 -20 0 20 40 -2 -1.5 -1 -0.5 0 0.5 1 reaction coordinate (A) QM energy (kcal/mol) HF/3-21G* B3LYP/6-311+G** // HF/3-21G* LMP2/6-31+G* // HF/3-21G* AM1 PHBH active site showing transition state interaction with Pro293 PHBH Calculations identified novel catalytic interaction Hydrogen bond with Pro293 carbonyl Specifically stabilizes transition state (only) Supported by subsequent experiments Now also found in phenol hydroxylase (PH) General feature of flavoprotein oxygenase catalysis? QM/MM tested by higher level calculations PHBH PHBH Calculated GST free energy profiles Correlation between experimental and calculated barriers Validates QM/MM model Identification of mechanism Rate-limiting step Correlation with experimental rates for phenol hydroxylase also 4 5 6 7 8 9 17 18 19 20 21 22 23 COO O F COO O F COO O COO O F F COO O F F F F ln k cat QM/MM energy barrier (kcal/mol) R = 0.91 3 4 5 6 7 23 24 25 26 27 28 29 30 31 QM/MM energy barrier (kcal/mol) ln k cat Phenol hydroxylase (PH): Predict rates for alternative substrates PH Stabilization by the enzyme is greatest at the transition state Enzyme binds TS more strongly than substrate TS stabilization central to catalysis TS is stabilized (relative to substrate) more by the enzyme than it is in solution Calculated rate acceleration consistent with experiment Substrate is distorted: strain/compression may also assist catalysis Chorismate mutase active site is complementary to the transition state Major contributions to TS stabilization from Arg90, Glu78, Arg7, and some bound water molecules Conclusions QM/MM modelling can provide detailed insight into enzyme mechanisms Can compare mechanisms, model transition states Identify and analyse catalytic interactions Results demonstrate the importance of TS stabilization Novel TS stabilizing interaction identified in PHBH and PH Relative stabilities of TSs determine stereospecificity in GST Chorismate mutase active site is exquisitely well-organized to stabilize TS Modelling helps relate enzyme structure to activity OH OH OH O 2 COOH COOH OH OH OH O 2 PHBH: Chorismate mutase: stabilization along reaction path