USF Libraries
USF Digital Collections

Steady-state and theoretical investigations of peptidylglycine α-amidating monooxygenase

MISSING IMAGE

Material Information

Title:
Steady-state and theoretical investigations of peptidylglycine α-amidating monooxygenase
Physical Description:
Book
Language:
English
Creator:
Lowe, Edward W
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Inactivation
Dynamics
Radical
Kinetics
Ab initio
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Approximately 50 percent of all known peptide hormones are post-translationally modified at their C-terminus. These peptide hormones are responsible for cellular functions critical to survival. Peptidylglycine alpha-amidating monooxygenase (PAM) is a bi-functional enzyme which catalyzes the conversion of peptide pro-hormones to peptide hormones. PAM is the only known mammalian enzyme that catalyzes the necessary alpha-amidation to activate these peptide hormones. PAM has previously been found to perform N-dealkylation, as well as O-dealkylation. We report here that a novel chemistry for PAM, S-dealkylation, has now been shown. PAM was able to catalyzes the hydroxylation and subsequent dealkylation for a series of substituted 2- (phenylthio)acetic acid analogs, leaving a product containing a free thiol capable of coordinating to copper(I). A series of cinnamic acid derivatives have been investigated as turnover dependent inactivators of PAM.It was shown that the inactivating compounds contained electron donating substituents. All compounds bound competitively versus substrate, though no catalytic activity was noted when tested as substrates. Although no superscript Dksubscript inact was observed when using perdeuterated cinnamic acid, one cannot rule out hydrogen abstraction from the Cα as this step may not be rate limiting for inactivation. This suggests that the activated oxygen species generated at CuM may be sufficiently reactive to abstract a hydrogen from an alkene to generate a vinyl radical. Substrate activation is believed to be facilitated by a Cu(II)-superoxo complex formed at Cusubscript M. Hydrogen abstraction from the Cα is hypothesized to generate a radical, though this has never been demonstrated spectrometrically.We report here further evidence for the generation of an Cα radical by comparing log(Vsubscript max/Ksubscript O₂) vs δ⁺ for a series of ring-substituted 4-phenyl-3- butenoic acids. Lastly, a computational study was carried out to probe for a possible binding pocket for the reductant, ascorbate. Though crystal structures have argued that reduction of the enzymebound coppers is collisional, kinetic data for inhibitors competitive against ascorbate indicates that a discrete binding pocket may exist. Our study suggests a specific site for binding and provides free energy calculations in agreement with experimental values for binding constants.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Edward W. Lowe.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 189 pages.
General Note:
Includes vita.
General Note:
Non-Latin script record

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002046417
oclc - 496017671
usfldc doi - E14-SFE0002717
usfldc handle - e14.2717
System ID:
SFS0027034:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200421Ka 4500
controlfield tag 001 002046417
005 20100106145635.0
007 cr mnu|||uuuuu
008 100106s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002717
035
(OCoLC)496017671
040
FHM
c FHM
049
FHMM
090
QD31.2 (Online)
1 100
Lowe, Edward W.
0 245
Steady-state and theoretical investigations of peptidylglycine -amidating monooxygenase
h [electronic resource] /
by Edward W. Lowe.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 189 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Non-Latin script record
590
Advisor: David J. Merkler, Ph.D.
520
ABSTRACT: Approximately 50 percent of all known peptide hormones are post-translationally modified at their C-terminus. These peptide hormones are responsible for cellular functions critical to survival. Peptidylglycine alpha-amidating monooxygenase (PAM) is a bi-functional enzyme which catalyzes the conversion of peptide pro-hormones to peptide hormones. PAM is the only known mammalian enzyme that catalyzes the necessary alpha-amidation to activate these peptide hormones. PAM has previously been found to perform N-dealkylation, as well as O-dealkylation. We report here that a novel chemistry for PAM, S-dealkylation, has now been shown. PAM was able to catalyzes the hydroxylation and subsequent dealkylation for a series of substituted 2- (phenylthio)acetic acid analogs, leaving a product containing a free thiol capable of coordinating to copper(I). A series of cinnamic acid derivatives have been investigated as turnover dependent inactivators of PAM.It was shown that the inactivating compounds contained electron donating substituents. All compounds bound competitively versus substrate, though no catalytic activity was noted when tested as substrates. Although no [superscript D]k[subscript inact] was observed when using perdeuterated cinnamic acid, one cannot rule out hydrogen abstraction from the C as this step may not be rate limiting for inactivation. This suggests that the activated oxygen species generated at CuM may be sufficiently reactive to abstract a hydrogen from an alkene to generate a vinyl radical. Substrate activation is believed to be facilitated by a Cu(II)-superoxo complex formed at Cu[subscript M]. Hydrogen abstraction from the C is hypothesized to generate a radical, though this has never been demonstrated spectrometrically.We report here further evidence for the generation of an C radical by comparing log(V[subscript max]/K[subscript O]) vs for a series of ring-substituted 4-phenyl-3- butenoic acids. Lastly, a computational study was carried out to probe for a possible binding pocket for the reductant, ascorbate. Though crystal structures have argued that reduction of the enzymebound coppers is collisional, kinetic data for inhibitors competitive against ascorbate indicates that a discrete binding pocket may exist. Our study suggests a specific site for binding and provides free energy calculations in agreement with experimental values for binding constants.
653
Inactivation
Dynamics
Radical
Kinetics
Ab initio
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2717



PAGE 1

Amidating Monooxygenase (PAM) By Edward W. Lowe, Jr. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: David J. Merkler, Ph.D. Robert Potter, Ph.D. Wayne Guida, Ph.D. Paul C. Kline, Ph.D. Date of Approval: November 14, 2008 Keywords: inactivation dynamics, radical, kinetics, ab initio Copyright 2008, Edward W. Lowe, Jr.

PAGE 2

Dedications I would li ke to dedicate this to my family, but mo st importantly my wife and daughter. Without the support and encouragement of my family I would have never made it through these difficult years in graduate school. To my wife, I would like to thank you for your patience, support, and unwavering faith in me. You and our daughter truly drive me to be all that I can. You are my best friend, and I can t imagine my life without you.

PAGE 3

Acknowledgments I would like to thank Dr. David Merkler for giving me the opportunity to work in his laborato ry. I am thankful for your guidance and for what you have taught me in both life and science. I appreciate all of the opportunities you have given me. I would also like to thank the Merkler lab members. It s been an interesting journey, and I don t know that I could have made it without As the Lab Burns I wish you all the best of luck. Lastly, I d like to thank NRM and SEC for the fun times we shared at Savy Jack s, in lab, and at the gym. I definitely wouldn t have made it through without you two

PAGE 4

i Table of Contents List of Tables iv List of Figures v Abstract x Chapter One: Introduction 1 Peptide Hormones 1 Amidating Enzyme 1 2 4 Peptidylglycine Amidoglycolate Lyase 7 Introduction to Chapters 7 References 9 Chapter Two: Novel Chemistry in PAM: S dealkylation 28 Introduction 28 Materials and Methods 29 Materials 29 Methods 30 di deutero 2(phenylthio)acetic acid 30 Determination of KM,app and VMAX,app values for 2 (phenylthio)acetic D,D), 2 (4 methylphenylthio)acetic acid, 2(4 chlorophenylthio)acetic acid, and 1H indole 2ylcarbonyl)thiol]acetic acid 30 Inhibition of enzyme dependent O2 consumption from N acetylglycine by 2(phenylthio)acetic acid 30 Glyoxylate production from 2(phenylthio)acetic acid derivative substrates 31 RP HPLC separation of 2 (phenylthio)acetic acid and thiophenol 32 Analysis of steady state kinetic dat a 32 In silico docking 33 Computational Chemistry 33 Results and Discussion 34 2(phenylthio)acetic acid as a PAM substrate 34 2(phenylthio)acetic acid derivatives as PAM substrates and inhibitors 34 Product mediated reversible inactivation 35 Modeling of PTAA, analogs, and thiophenol in the PHM active site 36 Coordination of CuM by thiophenol 36 Conclusion 37

PAGE 5

ii References 39 Chapter Three: Cinnamic acid Derivatives as Inactivators of PAM 58 Introduction 58 Materials and Methods 59 Materials 59 Methods 60 Synthesis of Dansyl 4aminocinnamic acid 60 Synthesis of 3 phenyloxirane 2carboxylic acid 60 In silico ligand docking 60 H dissociation energies 61 Inhibition of O2 consumption from N acetylglyc ine by cinnamic acid 61 Inactivation of PAM by cinnamic acid 61 Reversibility of inactivation 62 PAM labeling by Dansyl 4aminocinnamic acid 62 PAM labeling by 14C cinnamic acid 63 PAM modification through inactivation by cinnamic acid 63 PAL inactivation assay 64 Analysis of steady state kinetic data 64 Results 65 Cinnamic acid and analogs as inhibitors of PAM 65 Inactivation of PAM by cinnamic acid 65 Inactivation of PAL activity 66 Irreversible inactivation 66 Non lab eled enzyme 66 Docking of cinnamic acid and analogs 67 Computation results for bond energy 67 Discussion 67 Conclusion 69 References 70 Chapter F our. PAM Inactivation by Phenylbutenoates: Evidence for a radical intermediate 88 Introduction 88 Materials and Methods 89 Materials 89 Methods 89 In silico docking 89 Radical stabilization energies 90 Inactivation of PAM by 4 phenyl 3butenoates at ambient O2 90 Inactivation of PAM by 4 phenyl 3butenoates at variable O2 concentrations 91 Reversibility of inactivation 92 Inactivation of PAL activity by 4 phenyl 3butenoic acid 93 Determination of partition ratios 93 Determination of glyoxylate concentrations 93

PAGE 6

iii Investigation of inactivation through PAM modification 94 Analysis of kinetic data 94 Results 94 Molecular modeling 94 Partition ratios 94 Inactivation of PAM by phenylbutenoates 94 Inactivation of PAL activity 96 Non labeled PAM 97 Radical stabilization energies 97 Discussion 97 Conclusion 99 References 101 Chapter F ive. Computational Elucidation of Reductant Binding Sites in PAM 120 Introduction 120 Methods 122 Equilibrium Molecular Dynamics 122 Steered Molecular Dynamics 123 Analysis of Trajectories 124 Results 126 A scorbate binding simulations 126 Mimosine binding simulations 126 Steered molecular dynamics 127 Discussion 128 References 130 Appendices 151 Appendix A: PCGAMESS Geometry Optimization Settings 151 Appendix B: PCGAMESS Hes sian Calculation Input Files 165 Appendix C: NAMD Configuration Files 168 Appendix D: Monooxygenase X: Homology modeling results 176 Appendix E: Equilibra tion results of oxidized PHM 179 Appendix F: Dopamine b M onooxygenase: Homology model 181 Appendix G: Synthesis of 14C mimosine 183 About the Author End Page

PAGE 7

iv List of Tables Table 2.1 Steady state kinetic constants for 2 (phen ylthio)acetic acid and analogs 54 Table 2.2 PAM inhibitors 55 Table 3.1 Inhibition constants calculated from computer fit of data to equation 1. 81 Table 3.2 Inactivation constants calculated by KitzWilson analysis using the dilution assay method as outline in the m aterials and methods section. 83 Table 3.3 Calculated bond dissociation energies of selected cinnamate analogs By DFT/6 31G* 84 Table 4.1 Partition ratios for ring substitut ed 4 phenyl 3butenoic acids 114 Table 4.2 The kinetic parameters for the inactivation of PHM by ring substituted 4phenyl 3butenoic acids 115 Table 4.3 Radical stabilization energies for ring substituted 4 phenyl 3butenoic aci ds 116

PAGE 8

v List of Figures Figure 1.1 Pro substance P converted to the bioactive substance P through some amidation mechan ism 15 Figure 1.2 PAM reaction scheme showing the roles of both PHM and PAL 16 Figure 1.3 monooxygenase reaction scheme 17 Figure 1.4 18 Figure 1.5 PHM crystal structure (reduced) in licorice representation 19 Figure 1.6 PHM crystal structure shown with bound coppers flanking the active site 20 Figure 1.7 The two copper centers in the PHM crystal structures coordinated to their r espective 3 ligand system 21 Figure 1.8 PHM crystal structure with IYT forming a salt bridge with R240 22 Figure 1.9 PHM crystal structures of reduced (red) and oxidized (blue) forms aligned s how very little differences in structure versus redox states 23 Figure 1.10 The pre catalytic PHM crystal structure with oxygen bound to CuM i n end o n fashion 24 Figure 1.11 The two mechanisms proposed from experimental data are compared with the side 2 above and the end 1 mechanism below 25 Figure 1.12 The two mechanisms proposed by theoreticians comparing the quartet to doublet spin inversion and the triplet to singlet spin inversion 26 Figure 2.1 The reacti ons catalyzed by PHM and PAL 40 Figure 2.2 The CuM domain with the L3 side chains methyl capped and a single water molecule coordinated to Cu1+ 41 Figure 2.3 The CuM domain with the L3 side chains methyl capped and thiophenolate coordinated to Cu1+ 42 Figure 2.4 The CuM domain with L3 side chains methyl capped and coordinated to Cu1+ 43 Figure 2.5 Proposed 2 (phe nylthio)acetic acid mechanism 44

PAGE 9

vi Figure 2.6 The PAM dependent consumption of O2 in the presence of 2 (4 chlorop henyl)acetic acid 45 Figure 2.7 Inhibition of PAM by 2 (benzoylthio)acetic acid 46 Figure 2.8 HPLC analysis of PAM reaction con taining PTAA as the substrate 47 Figure 2.9 Mercury experiment to outcompete thiol coordination of copper 48 Figure 2.10 PHM crystal structure 49 Figure 2.11 PTAA and N benzylglycine docked i nto the PHM crystal structure 50 Figure 2.12 L3Cu1+ geometry optimized us ing the LanL2dz+ECP basis set 51 Figure 2.13 L3Cu1+H2O geometry optimized us ing the LanL2dz+ECP basis set 52 Figure 2.14 L3Cu1+thiophenolate geometry optimized us ing the LanL2dz+ECP basis set 53 Fi gure 3.1 PAM catalytic scheme 71 Figure 3.2 Illustration of hydrogen bond elongation for the DFT calculation of bond dissociation energies 72 Figure 3.3 Inhibition of PAM by cinnamic acid 73 Figure 3.4 Inactiv ation plots for cinnamic acid 74 Figure 3.5 Ascorbate and O2 dependence of PHM inactivation inves tigated by oxygen consumption 75 Figure 3.6 Cinnamate a nalogs as inactivators of the PAL activity of PAM by the g lyoxylate assay 76 Figure 3.7 Reversibility of inactiv ation of PAM by cinnamic acid 77 Figure 3.8 MALDI TOF overlay of trypsin digested active and cinnamatei nactivated PAM 78 Figure 3.9 Cinnamate docked in the active site of the PHM crystal structur e (1SDW) 79 Figure 3.10 Cinnamate inactivation scheme where cinnamate radical is released prior to hydroxylation leaving the activated Cu alkoxide species 80

PAGE 10

vii Figure 4.1 The react ion catalyzed by PHM and PAL 101 Figure 4.2 The reaction schemes of PHM and 102 Figure 4.3 The hydrogen abstraction and hydroxylation mechanisms for the two most accepted mechanisms 103 Figure 4.4 4phenyl 3butenoate forming a salt bridge with th e guanidin o group of R240 104 Figure 4.5 4phenyl 3butenoate shown docked within the PHM active site showing the H to the CuM 105 Figure 4.6 The time dependent inactivation of PHM activity by 4 (3 c hlorophenyl) 3butenoic acid 106 Figure 4.7 Protection against 4 phenyl 3butenoic acid mediated inactivation of PHM by a known PHM substrate 107 Figure 4.8 Linear free energy plot of kinact/KO2 values as a function of electron Donating and electronwithdrawin g ability of the substituent 108 Figure 4.9 The effect of 4 phenyl 3butenoic acid on the PAL activity of PAM 109 Figure 4.10 The trypsin digested MALDI TOF fingerprint of normal (blue) and 4phenyl 3butenoic a cid inactivation (green) PAM 110 Figure 4.11 Free energy pl ot of Vmax/KO2 as a function of electron withdrawing and electrondonating ability of the substituent 111 Figure 4.12 Radical rearrangement between alpha and gamma positions in 4 phenyl 3butenoic acid leading to two hydroxylated products 112 Figure 4.13 Benzylglycine docked to illustrate the proper hydrogen bonding interaction between N316 and the ami de hydrogen of the substrate 113 Figure 5.1 amid ating monooxygenase reaction 130 Figure 5.2 States A and B in which state A is the starting structure and state B is the final state postSMD simulation 131 Figure 5.3 Methyl capped CuM and CuH domains in the PHM crystal structure pr ior to geometry optimization 132 Figure 5.4 PHM crystal struc ture soaked with ascorbate 133

PAGE 11

viii Figure 5.5 Il lustration of the PHM secondary structure with bound ascorbate and the spring as the force vector applied during the steering molecular dynamics simulations 134 Figure 5.6 Ascorbate and mimosine used to probe for a reductant binding site on PHM and th e carbon atoms in which the force was applied during the steering molecular dynamics simulations 135 Figure 5.7 PHM secondary structure with bound ascorbate and the fixed L336 and N337 136 Figure 5.8 Ascorbate bound to the surface of PHM 137 Figure 5.9 Ascorbate in the proposed binding pocket with hydrogen bonding illustrated as blue or red dashed lines 138 Figure 5.10 Hydrogen bonding between ascorbate and th e proposed PHM binding sites 139 Figure 5.11 Mimosine bound to the surface of PHM 140 Figure 5.12 Mimosine in the proposed binding pocket with hydrogen bond ing illustrated as the dashed lines 141 Figure 5.13 hydrogen bonding between mimosine and t he proposed PHM binding site 142 Figure 5.14 PME electrostatic potential of PHM w ith off and on ascorbate 143 Figure 5.15 PME electrostatic potential of PHM with off and on mimosine 144 Figure 5.16 Force vs extension for both ascorbate and mimosine ov er a 5 reaction coordinate 145 Figure 5.17 Work vs extension for asco rbate and mimosine unb inding 146 Figure 5.18 Potential of mean force vs extension for asco rbate and mimosine unbinding 147 Figure A 1 Monooxygenase X solvated and ionized prior to equilibration using NAMD 173 Figure A 2 Monooxygenase X in ribbon representation of the seconda ry structure a fter equilibration with NAMD 174 Figure A 3 The rmsd versus time graph for the equilibration of m onooxygenase X using NAMD 175

PAGE 12

ix Figure A 4 PHM crystal structure in ribbon format solvated prio r to equilibration with NAMD 176 Figure A 5 The time versus rmsd for the equilibration of oxidized PHM with NA MD 177 Figure A 6 monooxygenase homology model prior to refinement with NAMD 178 Figure A 7 monooxygenase homology model 179 Figure A 8 Mimos ine synthase reaction scheme 180 Figure A 9 The synthetic scheme for Oacetylserine 181 Figure A 10 The synthesis of 3,4dihydroxypyridine 182 Figure A 11 The separation of a mimosi ne standard solution by HPLC 183 Figure A 12 The HPLC separation of 3,4dihydroxypyridine 184 Figure A 13 The enzymatic synthe sis of radiolabeled mimosine 185 Figure A 14 30minute time point for the conversion of Oacetylserine and 3 4dihydroxypyridine to m imosine by mimosine synthase 186

PAGE 13

x S teady State and Theoretical investigations Amidating Monooxygenase Edward W. Lowe, Jr. ABSTRACT Approximately 50% of all known peptide hormones are post translationally modified at their Cterminus. These peptide hormones are responsible for cellular functions critical to amidating monooxygenase (PAM) is a bi functional enzyme which catalyzes the conversion of peptide pro hormones to peptide hormones. PAM is t he only known mammalian enzyme that cat alyzes the necessary aamidation to activate these peptide hormones. PAM has previously been found to perform N dealkylation, as well as O dealkylation. We report here that a novel chemistry for PAM, S dealkylation, has now been shown. PAM was able to catalyzes the hydroxylation and subsequent dealkylation for a series of substituted 2(phenylthio)acetic acid analogs, leaving a product containing a free thiol capable of coordinating to copper(I). A series of cinnamic acid derivatives have been investigated as turnover dependent inactivators of PAM. It was shown that the inactivating compounds contained electron donating substituents. All compounds bound competitively versus substrate, though no catalytic activity wa s noted when tested as substrates. Although no Dkinact was observed when using perdeuterated limiting for inactivation. This suggests that the activated oxygen species generated at CuM m ay be sufficiently reactive to abstract a hydrogen from an alkene to generate a vinyl radical.

PAGE 14

xi Substrate activation is believed to be facilitated by a Cu(II) superoxo complex formed at CuM. Hydrogen abstraction from the never been demonstrated spectrometrically. We report here further evidence for the generation Vmax/KO2+ for a series of ring substituted 4 phenyl 3butenoic a cids. Lastly, a computational study was carried out to probe for a possible binding pocket for the reductant, ascorbate. Though crystal structures have argued that reduction of the enzymebound coppers is collisional, kinetic data for inhibitors competiti ve against ascorbate indicates that a discrete binding pocket may exist. Our study suggests a specific site for binding and provides free energy calculations in agreement with experiment al values for binding constants

PAGE 15

Chapter One: Introduction 1 Peptide Hormones amidated. In amidated. The amino acid prohormone sequence of these peptides contains a C terminal glycine The amidation of the carboxy terminus is essential for activation of these peptide hormones as the glycineextended precursors are ge ne r a l l y >1000amidated peptides [1] Some of these amidated hormones act as amidated peptide hormones have been f ound to be over expressed in human disease states such as substance P in rheumatoid arthritis [2, 3] luteinizing hormone releasing hormone and vasoactive intestinal peptide in cancer [4, 5] and corticotropin releasing factor in anxiety and depression [6] The fact that the precursor sequences for these essential amidated hormones all contained a C terminal glycine suggested that the C terminal amide biosynthesis required the action of a specific enzyme to activate these peptide prohormones through the functionalization of the glycine ( Figure 1. 1). In 1982, the enzyme responsible for the conversion of glycine extended peptides to the corresponding peptide amide was discovered in porcine pituitary [7] Amidating Enzyme amidating monooxygenase (PAM) is a bif unctional copper monooxygenase. PAM is found within neuronal and endocrine cells, neurosecretory vesicles, and the highest concentrations are found in the atrium of the heart [8 11] PAM is primarily responsible in vivo for the activation of peptide hormones through the conversion of the glycine extension to the corresponding amide. PAM also plays a role in the biosynthesis of fatty acid amides [1215] PAM is the only mammalian enzyme known to catalyze this amidation reaction.

PAGE 16

Chapter One: Introduction 2 PAM (E.C. 1.14.17.3) consists of two hetero hydroxylating monooxygenase (PHM) and peptidylglycine amidoglycolate lyase (PAL) ( Figure 1. 2). The PHM domain has been deeply studied primarily due to its similarity in both reaction and structure to another copper monooxygenase [16, 17] of dopamine to norepinephrine ( Figure 1. 3) [16, 18] This tetrameric protein exists in both membrane and soluble forms within neurosecretory vesicles of the sympathetic nervous system [19] While dopamine is a neurotransmitter in the central nervous system, its functionalization to in the sympathetic nervous system which consists of neuronal axons that interact with smooth muscle and the catecholamine secreting cells within the adrenal medullae [20] sequence similarity in the catalytic core [17, 21] Both contain a large, solvent accessible active site flanked by two noncoupled copper atoms [2224] PHM catalyzes the copper O2, and ascorbatehe copper O2, and ascorbatedependent hydroxylation of a benzyl ic carbon ( Figure 1. 4) [16, 22, 25 27] While [28] The crystal structure of PHM reveals it to be a prolate ellipsoid composed of two 9 sandwich domains ( Figure 1. 5). The domains are approximately of equivalent size and are held together by a 500 2 interface. The active site is a large, solvent accessible cleft residing at the interdomain interface with two bound coppers, one on each side of the cleft ( Figure 1. 6). Each domain is centered around their respecti ve copper atom. Each of the two coppers has a different ligand set. One copper center,

PAGE 17

Chapter One: Introduction 3 CuM, has two histidine N ligands and a methionine sulfur ligand (H242, H244, M314). The remaining copper center, CuH, has three histidine N ligands (H107, H108, H172) ( Figure 1. 7). This is also the case in D M as one copper is coordinated to three histidines and the other to two histidines and a methionine. The enzyme was co crystallized with N acetyl 3,5diiodotyrosyl threonine, a poor substrate for PHM. The substrate appears to form a salt bridge between the carboxy terminus and the guanidin o group of arginine 240 ( Figure 1. 8). This places the substrate in close proximity to the CuM wher e O2 has been found to bind and where activation of O2 is thought to occur. Hydrogen bonding between the glycyl amide hydrogen and the side chain oxygen of N316 also plays a role in the proper positioning of the substrate. Crystal structures for oxidized reduced, and pre catalytic forms of PHM with substrate bound have been solved [2830] The crystal structures are surprisingly identical with no differences seen between the oxidized and reduced form and an rmsd value of 0.27 ( Figure 1. 9). This suggests that movement or closure of the active site during catalysis does not occur. These structures also revealed that the copper atoms are ~ 10.6 apart which is in good agreement with previous extended X ray than 4 apart [2 1, 28, 30 32] The s e data along with electron paramagnetic resonance (EPR) coupled di copper enzymes or non blue/type II enzymes [22, 33, 34] The geometries of the two copper sites were found to change upon reduction. CuM is observed to have a square pyramidal geometry while in the oxidized form while CuH is square planar. Upon reduction, CuM becomes tetrahedral and CuH becomes T shaped. The bond length of CuMSM314 has been shown to tighten in the reduced form compared to oxidized enzyme according to EXAFS data suggesting that the oxidized form is more distorted

PAGE 18

Chapter One: Introduction 4 tetrahedral than square planar [34] EXAFS data also suggests that CuH is coordinated to only two of the three histidine residues in the reduced form. hydrogen abstraction from a substrate by some activated Cu O species followed by the subsequent stereospecific hydroxylation of the proinvolves the abstraction of hydrogen, but does so from a benzylic carbon followed by pro R hydroxylation. In both cases, the mechanism is believed to involve electron transfer from CuH to CuM. PHM/D M Postulated Mech anisms The first step involved for catalysis is the irreversible reduction of both enzyme bound coppers by ascorbate allowing di oxygen and substrate to bind to the reduced enzyme [16, 21, 35, 36] This reduction is a pingpong mechanism. Burst ph ase kinetic experiments w i t h D M have shown that pre reduced enzyme is capable of hydroxylating substrate with the amplitude of the pre steady state burst being equivalent to the concentration of enzyme [3739] This indicates that the chemical step is much faster than product release and also allowed for reoxidation with product release to be observed. A temperature dependence on primary and secondary intrinsic isotop H cleavage has also been observed for PAM. This nonclassical behavior suggests that quantum tunneling is involved in hydrogen transfer [40] 18O kinetic isotope effects have been used to probe the nature of the activated Cu/O provide information up to and includ ing the O O cleavage step due to the quantum mechanical nature of hydrogen transfer [41] When coupled with substrate deuterium isotope effects, the

PAGE 19

Chapter One: Introduction 5 H bond cleavage occurs prior to that of O O necessary for hydroxylation [4245] This suggests that a Cu(II) hydroperoxo is being formed during hydrogen abstraction but lends no evidence as to what the hydroxylating species may be. Rapid freeze quench EPR data also sugges ts that a copper superoxo species is formed, as O O and C H cleavage are tightly coupled [46] The data suggest that the substrate intermediate resulting from Cu ( II)superoxo abstraction of hydrogen would be a radical intermediate. Evidence for a radical [4749] This evidence has been applied towards the PHM mechanism though it has not been proven directly for PHM. Two of these competing mechanisms both support a copper superoxo nucleophile for hydrogen abstraction. While the nature of the hydrogen acceptor is in agreement, the initial geometry of the di oxygen species to copper is in question. Spectroscopic data h as indicated a side 2 [50, 51] When compared to the end 1 spec ies, the side on Cu(II)superoxo was determined computationally to be more thermodynamically favorable [52, 53] However, crystallographic evidence of a pre catalytic state of PHM with good resolution (1.85 ) has found the end1 species present ( Figure 1. 10) [28] Upon forming eith er disputed Cusuperoxo species, both mechanisms proceed to abstract hydrogen to form a Cu(II) hydroperoxo species. This hydroperoxo species is coordinated to copper in an end1 geometry. The competing mechanisms again diverge as the side 2 mecha nism predicts direct hydroxylation of the resulting substrate radical followed by radical recombination. The net effect is reduction of the copper (II)hydroperoxo species and release of the hydroxylated product. Electron transfer would then occur from CuH to complete the reaction.

PAGE 20

Chapter One: Introduction 6 On the other hand, the end 1 mechanism postulates that the Cu(II) hydroperoxo species is reduced by an electron transfer event prior to hydroxylation of the substrate radical. This electron provided by the CuH site allows for homolysis of the Cu(II)O OH yielding a Cu(II)O species. This newly formed Cu/O would then recombine with the substrate radical intermediate resulting in a covalent, inner sphere alcohol intermediate. Product release then occurs via h ydrolysis of this intermediate. The proposed mechanisms are compared in Figure 1. 11. While experimentalists agree that the hydrogen abstracting species is, indeed, a copper superoxo, several theoreticians have propos ed that a different activated Cu/O species is responsible for this chemistry. Using a series of molecular dynamics (MD) and mixed quantum mechanics molecular mechanics (QM/MM) calculations, two separate mechanisms have been proposed. The species proposed as responsible for hydrogen abstraction is a reduced copper oxo. This suggests that di oxygen reduction and substrate activation are uncoupled which is in direct contrast to experimental data. QM/MM simulations suggest that Cu(III) oxide/Cu(II) oxyl spe cies are thermodynamically favored over the superoxo/hydroperoxo species proposed by experimentalists [5456] These theoretical mechanisms suggest that electron transfer precedes hydrogen abstraction allowing Cu(II) superoxo to acquire two protons from solvent. Water is then released from the Cu( II)O OH2 species leaving either a Cu(II) oxyl quartet or a Cu(II) oxyl triplet species to abstract hydrogen from the substrate. The quartet is proposed to concertedly abstract the substrate hydrogen and hydroxylate the radical intermediate with a spin inversion t o the doublet ground state occurring upon substrate oxidation [54] This suggests that subst rate oxidation and product release happen simultaneously. A water molecule would then bind to the remaining Cu(II) to restore proper geometry in the oxidized state. The triplet Cu(II) oxyl is

PAGE 21

Chapter One: Introduction 7 believed to undergo a spin inversion event to the singlet ground state upon substrate C H oxidation driving concerted product release [55, 56] These two theoretical mechanisms are compared in Figu re 1. 12. Peptidylglycine Amidoglycolate Lyase hydroxyglycine PHM product resulting in the corresponding amide and glyoxylate is the PAL domain. This 33 kDa monomer is zinc calc ium, and iron dependent and bound to the C terminus of the PHM domain. Although the complete PAL mechanism is unknown, it is believed to proceed through a zinc hydrolase type reaction [57] The iron is involved in a tyrosine bridged Zn(II)Fe(III) complex. Mutation of this tyrosine (Y564) results in PAL inactivation and the inability to bind iron. No substrate channeling between PHM and PAL has been observed. While the inclusion of the PAL domain in PAM provides the key difference N dealkylation reactions solely through hydroxylation like that of benzylic N substituted analogues [58] have also both demonstrated sulfoxidation while PAM has demonstrated O dealkylation [59] I n t r od u c t i on t o C h ap t e r s PAM has previously been found to perform N dealkylation, as well as O dealkylation. We report here that a novel chemistry for PAM, S dealkylation, has now been shown. PAM was able to catalyzes the hydroxylation and subsequent dealkylation for a series of substituted 2(phenylthio)acetic acid analogs, leaving a product containing a free thiol capable of coordinating to copper(I).

PAGE 22

Chapter One: Introduction 8 A series of cinnamic acid derivatives have been investigated as turnover dependent inactivators of PAM. It was shown that the inactivating compounds con tained electron donating substituents. All compounds bound competitively versus substrate, though no catalytic activity was noted when tested as substrates. A l t h o ugh n o Dki n a c t w a s o b s e r v e d w h e n us i ng pe r de ut e r a t e d c i nn a m i c a c i d, o n e c a nn o t r ul e o ut hy dr o ge n a b s t r a c t i o n f r o m t h e C a s t hi s s t e p m a y n o t b e r a t e l i m i t i n g f o r i na c t i v a t i o n T hi s s ugge s t s t h a t t h e activated oxygen species generated at CuM m a y be sufficiently reactive to abstract a hydrogen from an alkene to generate a vinyl radical. Substrate activation is believed to be facilitated by a Cu(II) superoxo complex formed at CuM. Hy never been demonstrated spectrometrically. We report here further evidence for the generation Vmax/KO2+ for a series of ring substituted 4 phenyl 3butenoic acids. Lastly, a computational study was carried out to probe for a possible binding pocket for the reductant, ascorbate. Though crystal structures have argued that reduction of the enzymebound copper s is collisional, kinetic data for inhibitors competitive against ascorbate indicates that a discrete binding pocket may exist. Our study suggests a specific site for binding and provides free energy calculations in agreement with experimental values for binding constants.

PAGE 23

Chapter One: Introduction 9 References 1. Merkler, D.J., C terminal amidated peptides: production by the in vitro enzymatic amidation of glycine extended peptides and the importance of the amide to bioactivity. En zyme Microb Technol, 1994. 16 (6): p. 4506. 2. O'Connor, T.M., et al., The role of substance P in inflammatory disease. J Cell Physiol, 2004. 201(2): p. 167 80. 3. Bauer, J.D., et al., Anti inflammatory effects of 4 phenyl 3butenoic acid and 5(acetylamino) 4oxo 6phenyl 2hexenoic acid methyl ester, potential inhibitors of neuropeptide bioactivation. J Pharmacol Exp Ther, 2007. 320(3): p. 11717. 4. Tan, S.H. and A.C. Wolff, Luteinizing hormone releasing hormone agonists in premenopausal hor mone receptor positive breast cancer. Clin Breast Cancer, 2007. 7(6): p. 45564. 5. Engel, J.B. and A.V. Schally, Drug Insight: clinical use of agonists and antagonists of luteinizing hormone releasing hormone. Nat Clin Pract Endocrinol Metab, 2007. 3 (2): p. 15767. 6. Holsboer, F., The rationale for corticotropinreleasing hormone receptor (CRH R) antagonists to treat depression and anxiety. J Psychiatr Res, 1999. 33(3): p. 181 214. 7. Bradbury, A.F., M.D. Finnie, and D.G. Smyth, Mechanism of C terminal am ide formation by pituitary enzymes. Nature, 1982. 298 (5875): p. 6868. 8. Bolkenius, F.N. and A.J. Ganzhorn, Peptidylglycine alphaamidating monooxygenase: neuropeptide amidation as a target for drug design. Gen Pharmacol, 1998. 31 (5): p. 6559.

PAGE 24

Chapter One: Introduction 10 9. Braas, K.M., et al., Expression of peptidylglycine alphaamidating monooxygenase: an in situ hybridization and immunocytochemical study. Endocrinology, 1992. 130(5): p. 2778 88. 10. Ouafik, L., et al., Developmental regulation of peptidylglycine alphaamid ating monooxygenase (PAM) in rat heart atrium and ventricle. Tissue specific changes in distribution of PAM activity, mRNA levels, and protein forms. J Biol Chem, 1989. 264(10): p. 5839 45. 11. Eipper, B.A., V. May, and K.M. Braas, Membrane associated pept idylglycine alphaamidating monooxygenase in the heart. J Biol Chem, 1988. 263(17): p. 8371 9. 12. Merkler, D.J., et al., Oleic acid derived metabolites in mouse neuroblastoma N18TG2 cells. Biochemistry, 2004. 43(39): p. 12667 74. 13. Merkler, K.A., et al. A pathway for the biosynthesis of fatty acid amides. Adv Exp Med Biol, 1999. 469: p. 51925. 14. Wilcox, B.J., et al., N acylglycine amidation: implications for the biosynthesis of fatty acid primary amides. Biochemistry, 1999. 38(11): p. 323545. 15. Me rkler, D.J., et al., Fatty acid amide biosynthesis: a possible new role for peptidylglycine alphaamidating enzyme and acyl coenzyme A: glycine N acyltransferase. Arch Biochem Biophys, 1996. 330 (2): p. 430 4. 16. Klinman, J.P., The copper enzyme family of dopamine betamonooxygenase and peptidylglycine alphahydroxylating monooxygenase: resolving the chemical pathway for substrate hydroxylation. J Biol Chem, 2006. 281(6): p. 30136.

PAGE 25

Chapter One: Introduction 11 17. Southan, C. and L.I. Kruse, Sequence similarity between dopamine betahydroxylase and peptide alphaamidating enzyme: evidence for a conserved catalytic domain. FEBS Lett, 1989. 255(1): p. 116 20. 18. Klinman, J.P., Mechanisms Whereby Mononuclear Copper Proteins Functionalize Organic Substrates. Chem Rev, 1996. 96 (7): p. 25412562. 19. Rush, R.A. and L.B. Geffen, Dopamine betahydroxylase in health and disease. Crit Rev Clin Lab Sci, 1980. 12(3): p. 24177. 20. Carmichael, S.W. and H. Winkler, The adrenal chromaffin cell. Sci Am, 1985. 253(2): p. 409. 21. Kulathila, R., et al ., Bifunctional peptidylglcine alphaamidating enzyme requires two copper atoms for maximum activity. Arch Biochem Biophys, 1994. 311(1): p. 191 5. 22. Klinman, J.P., et al., Evidence for two copper atoms/subunit in dopamine betamonooxygenase catalysis. J Biol Chem, 1984. 259(6): p. 3399402. 23. Ash, D.E., et al., Kinetic and spectroscopic studies of the interaction of copper with dopamine betahydroxylase. J Biol Chem, 1984. 259(6): p. 3395 8. 24. Tamburini, P.P. and S.D. Young, J Am Chem Soc, 1989. 111: p. 1933. 25. Merkler, D.J. and S.D. Young, Recombinant type A rat 75kDa alphaamidating enzyme catalyzes the conversion of glycine extended peptides to peptide amides via an alphahydroxyglycine intermediate. Arch Biochem Biophys, 1991. 289(1): p. 1926. 26. Merkler, D.J., et al., 18O isotopic 13C NMR shift as proof that bifunctional peptidylglycine alphaamidating enzyme is a monooxygenase. Biochemistry, 1992. 31(32): p. 7282 8.

PAGE 26

Chapter One: Introduction 12 27. Prigge, S.T., et al., Dioxygen binds endon to mononuclear copper in a precatalytic enzyme complex. Science, 2004. 304 (5672): p. 864 7. 28. Prigge, S.T., et al., Amidation of bioactive peptides: the structure of peptidylglycine alphahydroxylating monooxygenase. Science, 1997. 278(5341): p. 1300 5. 29. Prigge, S.T., et al., Substrate mediated electron transfer in peptidylglycine alpha hydroxylating monooxygenase. Nat Struct Biol, 1999. 6(10): p. 976 83. 30. Prigge, S.T., et al., New insights into copper monooxygenases and peptide amidation: structure, mechanism and function. Cell Mol Life Sci, 2000. 57 (8 9): p. 123659. 31. Blackburn, N.J., et al., Copper K extended xray absorption fine structure studies of oxidized and reduced dopamine betahydroxylase. Confirmation of a sulfur ligand to copper(I) in the reduced enzyme. J B iol Chem, 1991. 266(34): p. 23120 7. 32. Blumberg, W.E., et al., X ray absorption spectroscopic study of the active copper sites in dopamine betahydroxylase. J Biol Chem, 1989. 264(11): p. 6029 32. 33. Jaron, S. and N.J. Blackburn, Characterization of a half apo derivative of peptidylglycine monooxygenase. Insight into the reactivity of each active site copper. Biochemistry, 2001. 40(23): p. 6867 75. 34. Rhames, F.C., et al., Isocyanide binding to the copper(I) centers of the catalytic core of peptidylglyc ine monooxygenase (PHMcc). J Biol Inorg Chem, 2001. 6(5 6): p. 567 77. 35. Glembotski, C.C., The characterization of the ascorbic acidmediated alphaamidation of alphamelanotropin in cultured intermediate pituitary lobe cells. Endocrinology, 1986. 118(4) : p. 1461 8. 36. Glembotski, C.C., The role of ascorbic acid in the biosynthesis of the neuroendocrine peptides alphaMSH and TRH. Ann N Y Acad Sci, 1987. 498: p. 54 62.

PAGE 27

Chapter One: Introduction 13 37. Brenner, M.C., C.J. Murray, and J.P. Klinman, Rapid freeze and chemical quench st udies of dopamine betamonooxygenase: comparison of pre steady state and steady state parameters. Biochemistry, 1989. 28(11): p. 4656 64. 38. Klinman, J.P. and M. Brenner, Role of copper and catalytic mechanism in the copper monooxygenase, dopamine betahy droxylase (D beta H). Prog Clin Biol Res, 1988. 274: p. 22748. 39. Brenner, M.C. and J.P. Klinman, Correlation of copper valency with product formation in single turnovers of dopamine betamonooxygenase. Biochemistry, 1989. 28(11): p. 466470. 40. Francis co, W.A., et al., Hydrogen tunneling in peptidylglycine alphahydroxylating monooxygenase. J Am Chem Soc, 2002. 124 (28): p. 81945. 41. Roth, J.P., Advances in studying bioinorganic reaction mechanisms: isotopic probes of activated oxygen intermediates in metalloenzymes. Curr Opin Chem Biol, 2007. 11(2): p. 14250. 42. Miller, S.M. and J.P. Klinman, Magnitude of intrinsic isotope effects in the dopamine beta monooxygenase reaction. Biochemistry, 1983. 22(13): p. 3091 6. 43. Francisco, W.A., N.J. Blackburn, and J.P. Klinman, Oxygen and hydrogen isotope effects in an active site tyrosine to phenylalanine mutant of peptidylglycine alphahydroxylating monooxygenase: mechanistic implications. Biochemistry, 2003. 42 (7): p. 18139. 44. Tian, G., J.A. Berry, and J.P Klinman, Oxygen 18 kinetic isotope effects in the dopamine beta monooxygenase reaction: evidence for a new chemical mechanism in nonheme metallomonooxygenases. Biochemistry, 1994. 33(1): p. 22634.

PAGE 28

Chapter One: Introduction 14 45. Francisco, W.A., et al., Kinetic mechanism and intr insic isotope effects for the peptidylglycine alphaamidating enzyme reaction. Biochemistry, 1998. 37(22): p. 824452. 46. Evans, J.P., K. Ahn, and J.P. Klinman, Evidence that dioxygen and substrate activation are tightly coupled in dopamine betamonooxyge nase. Implications for the reactive oxygen species. J Biol Chem, 2003. 278(50): p. 496918. 47. Miller, S.M. and J.P. Klinman, Secondary isotope effects and structure reactivity correlations in the dopamine betamonooxygenase reaction: evidence for a chemi cal mechanism. Biochemistry, 1985. 24(9): p. 211427. 48. Fitzpatrick, P.F., D.R. Flory, Jr., and J.J. Villafranca, 3Phenylpropenes as mechanism based inhibitors of dopamine betahydroxylase: evidence for a radical mechanism. Biochemistry, 1985. 24(9): p. 210814. 49. Fitzpatrick, P.F. and J.J. Villafranca, Mechanism based inhibitors of dopamine betahydroxylase. Arch Biochem Biophys, 1987. 257(2): p. 23150. 50. Chen, P. and E.I. Solomon, Frontier molecular orbital analysis of Cu(n) O(2) reactivity. J Ino rg Biochem, 2002. 88(3 4): p. 368 74. 51. Chen, P., et al., Spectroscopic and electronic structure studies of the diamagnetic side on CuII superoxo complex Cu(O2)[HB(3R 5iPrpz)3]: antiferromagnetic coupling versus covalent delocalization. J Am Chem Soc, 2003. 125 (2): p. 466 74. 52. Chen, P. and E.I. Solomon, O2 activation by binuclear Cu sites: noncoupled versus exchange coupled reaction mechanisms. Proc Natl Acad Sci U S A, 2004. 101(36): p. 1310510.

PAGE 29

Chapter One: Introduction 15 53. Chen, P., et al., Oxygen activation by the noncoupled binuclear copper site in peptidylglycine alphahydroxylating monooxygenase. Spectroscopic definition of the resting sites and the putative CuIIM OOH intermediate. Biochemistry, 2004. 43(19): p. 573547. 54. Crespo, A., et al., The catalytic mechanism of peptidylglycine alphahydroxylating monooxygenase investigated by computer simulation. J Am Chem Soc, 2006. 128(39): p. 1281728. 55. Kamachi, T., et al., Computational exploration of the catalytic mechanism of dopamine beta monooxygenase: modeling of i ts mononuclear copper active sites. Inorg Chem, 2005. 44(12): p. 4226 36. 56. Yoshizawa, K., et al., Catalytic mechanism of dopamine betamonooxygenase mediated by Cu(III)oxo. Inorg Chem, 2006. 45(7): p. 303441. 57. Bell, J., et al., Structural and functional investigations on the role of zinc in bifunctional rat peptidylglycine alphaamidating enzyme. Biochemistry, 1997. 36(51): p. 16239 46. 58. Padgette, S.R., et al., Olefin oxygenation and N dealkylation by dopamine beta monooxygenase: catalysis and mechanism based inhibition. Biochemistry, 1985. 24(21): p. 582639. 59. Katopodis, A.G. and S.W. May, Novel substrates and inhibitors of peptidylglycine alpha amidating monooxygenase. Biochemistry, 1990. 29 (19): p. 4541 8.

PAGE 30

Chapter One: Introduction 16 List of Figures Figure 1. 1. Pro substance P converted to the bioactive substance P through some amidation mechanism. A r g P r o L ys P r o G l n G l n P he P he G l y L e u M e t G l y A r g P r o L ys P r o G l n G l n P he P he G l y L e u M e t N H2 ?

PAGE 31

Chapter One: Introduction 17 Figure 1. 2. PAM reaction scheme showing the roles of both PHM and PAL. H N O H O 2A s c 2 C u ( I I ) O2H2O O H O O H P A M P H M 2 C u ( I I ) 2A s c P A LZ n ( I I ) / F e ( I I I ) / C a R O H N O H O R O O H N H2 R O 2S D A 2S D A

PAGE 32

Chapter One: Introduction 18 Figure 1. 3 monooxygenase reaction scheme. N H3 H O H O N H3 H O H O O H

PAGE 33

Chapter One: Introduction 19 Figure 1. 4. The reaction schemes of PHM (left) and D M (right). N H3H O H O N H3H O H O O H1 8O2 + 2 A sco rb a t e H2 1 8O + 2 Se mi d e h yd ro a sco rb a t e 2 C u (I I )1 8 R N H C O O O R N H C O O O O H1 8O2 + 2 A sco rb a t e H2 1 8O + 2 Se mi d e h yd ro a sco rb a t e1 82 C u (I I ) PHM D M

PAGE 34

Chapter One: Introduction 20 Figure 1. 5. PHM crystal structure (reduced) in licorice representation. Note, carbons are teal, oxygens are red, nitrogens are blue, sulfurs are yellow, and hydrogens are not shown for clarity.

PAGE 35

Chapter One: Introduction 21 Figure 1. 6. PHM crystal structure shown with bound coppers flanking the active site.

PAGE 36

Chapter One: Introduction 22 Figure 1. 7. The two copper centers in the PHM crystal structure coordinated to their respective 3 ligand system. Note, only the side chains of the residues are shown methyl capped.

PAGE 37

Chapter One: Introduction 23 Figure 1. 8. PHM crystal structure with IYT forming a salt bridge with R240.

PAGE 38

Chapter One: Introduction 24 Figure 1. 9. PHM crystal structures of reduced (red) and oxidized (blue) forms align ed show very little differences in structure versus redox state ( Rmsd = 0.27 )

PAGE 39

Chapter One: Introduction 25 Figure 1. 10. The pre catalytic PHM crystal structure with oxygen bound to CuM in end on fashion.

PAGE 40

Chapter One: Introduction 26 Figure 1. 11. The two mechanisms proposed from experimental data are compared with the side on/n2 mechanism above and the endon/n1 mechanism below. EC uI I HC uI I M i ) 2e-i i ) O2EC uI HC uI I MO O N H C O OH H H R EC uI HC uI I MO O N H C OOH H H R EC uI HC uI I MO O HN H C OOH H R EC uI HC uI I MO O HN H C O OH H R H2OO H2 EC uI HC uI I MO O HN H C OOH H R O H2 EC uI HC uI I MO N H C OOH H R O H2 OH 1 e-EC uI I HC uI I MH O N H C OOH H R O H H2O H+H2O EC uI I HC uI I M i ) 2e-i i ) O2EC uI HC uI I MO O N H C O OH H H R EC uI HC uI I MO N H C OO H H R EC uI HC uI I MO O HN H C OO H H R EC uI I HC uI I MO N H C O OH H R H+EC uI HC uI I MO N H C OO H H R EC uI I HC uI I M N H C OOH H R H2O OH O H 1 e-H2O H2O

PAGE 41

Chapter One: Introduction 27 Figure 1. 12. The two mechanisms proposed by theoreticians are compared with the quartet to doublet spin inversion above and the tr iplet to singlet spin inversion below. EC uII HC uII M i ) 2e-i i ) O2EC uI HC uII MO O N H C O O H H H R EC uII HC uII MO N H C O O H H R EC uII HC uII MO N H C O O H H R EC uII HC uI MH O N H C O O H H R EC uII HC uII MO HN H C O O H H R O H 1 eH2O H+ H+ H2O H H H2O s pi n i n ve r s i on s i ngl e tEC uII HC uII M i ) 2e-i i ) O2EC uI HC uII MO O N H C O O H H H R EC uII HC uII MO N H C O O H H R O H 1 eH+ H+ H2OH EC uII HC uII MO N H C O O H H R EC uII HC uII MH O N H C O O H H R EC uII HC uII MO HN H C O O H H R H2OH H2O s p i n i nve r s i on dou bl e t + 2 + 2 + 2

PAGE 42

Chapter Two: Novel Chemistry in PAM: S dealkylation 28 Introduction amidating monooxygenase (PAM) catalyzes the oxidative cleavage of C terminal glycine extended peptide hormones [1] fatty acid precursors [2 5] and bile acids [6] to the corresponding amides and glyoxylate. This two step amidation proceeds via a copper O2, and ascorbate, iron and calcium dependent dealkylation of the carbinolamide intermediate ( Figure 2.1). This bifunctional nature is attributable to the peptidylglycine hydroxylating monooxygenase (PHM) and peptidylglycine amidoglycolate lyase (PAL) domains, respectively [7, 8] amide moiety is a requisite for most active peptide hormones, with the amidated derivatives generally displaying >1000fold more potency than the glycine extended precursor [1] PAM is responsible for the activation of many i mportant neuropeptides such as s ubstance P and oxytocin [9 11] While this is the primary in vivo function of PAM, alternate catalytic functions have been reported including sulfoxidation, and O dealkylation [12] amidated amidated peptides implicated in disease are luteinizing hormone releasing hormone (cancer), vasoactive intestinal peptide (cancer), substance P (rheumatoid arthritis), and corticotropin releasing factor (anxiety and depression). Because prevention of amidation often reduces the efficacy of peptide hormones by ~3 orders of magnitude, broad inhibi tion of PAM may be overly toxic by blocking the conversion of essential amidated peptides. Toxicity may, however, prove useful in the possible development of insecticides. Insects possess separate PHM and PAL monomers and lack the bi functio nal PAM enzyme [13, 14]

PAGE 43

Chapter Two: Novel Chemistry in PAM: S dealkylation 29 Elimination of expression in Drosophila melanogaster has been demonstrated to be lethal [14] Because the mammalian PHM sequence and Drosophila P HM sequence share 41% identity and 52% similarity, sufficient differences may exist to target only the insect enzyme. Several (2 phenylthio)acetic acid (PTAA) derivatives were investigated as inhibitors of PAM activity. PTAA displays characteristics of a turnover dependent inactivator of PAM. We now report that PAM can catalyze Sdealkylation. This Sdealkylation leaves a free thiol capable of in situ coordination of the active site CuM leading to reversible inactivation of PAM. This discovery may give a new direction for the design of PAM inhibitors. The Sfor the design of PAM specific pro drugs that rely on PAM turnover for drug activity. MATERIALS AND METHODS Materials. [(1 H indole 2ylcarbonyl)thiol]acetic acid and 2(phenylthio)acetic acid were purchased from Sigma, 2 (4 chloroph enylthio)acetic acid was purchased from Pfaltz & Bauer, Inc., 2 (4 methylphenylthio)acetic acid was purchased from Aldrich Chem. Co., 2 (benzoylthio)acetic acid, 2 (2 nitrophenylthio)acetic acid, N benzoyl D alanylthioglycolic acid were purchased from Ebur on Organics N.V., and bovine catalase was from Worthington. Recombinant rat PAM was a gift from Unigene Laboratories, Inc. All other reagents were of the highest quality from commercial suppliers.

PAGE 44

Chapter Two: Novel Chemistry in PAM: S dealkylation 30 di deutero 2(phenylthio)acetic acid Thiophenol (6.1 mL, 60 mmol) in aqueous sodium hydroxide (20 mL 6M) was reacted D,D) bromoacetic acid (12.5 gm, 90 mmol) for 2 hr at 50 oC. The reaction was then acidified with dilute HCl(aq), precipitate collected, and recrystallized twice in hot ethanol. Determination of KM,app and VMAX,app values for 2D,D), 2 (4 methylphenylthio)acetic acid, 2(4 chlorophenylthio)acetic acid, and 1H indole 2ylcarbonyl) thiol]acetic acid. Reactions at 37.0 0.1 C were initiated by the addition of PAM (2050 g) into 2.0 ml of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascor bate, and 0.1 6.0 mM of the oxidizable substrate. Initial rates were measured by following the PAM dependent consumption of O2 using a Yellow Springs Instrument Model 5300 oxygen monitor interfaced with a personal computer using a Dataq Instruments anal ogue/digital converter (model DI 158UP). VMAX,app values were normalized to controls performed at 11.0 mM N acetylglycine. Inhibition of O2 consumption from N acetylglycine by 2 (phenylthio)acetic acid Reactions at 37.0 0.1 C were initiated by the add ition of PAM (35 g) into 2.0 ml of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, 8 mM N acetylglycine, and 0 5 mM 2 (phenylthio)acetic acid. Initial ra tes were measured by following the PAM dependent consumption of O2 using a Yellow Springs Instrument Model 5300 oxygen monitor.

PAGE 45

Chapter Two: Novel Chemistry in PAM: S dealkylation 31 Glyoxylate production from 2 (phenylthio)acetic acid derivative substrates To first determine if the PHM hydroxylated 2(phe nylthio)acetic acid was a substrate for PAL, 20 g of PAM was added to a 2.0 mL of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/mL catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 10 mM 2 (phenylthio)aceti c acid. After incubation for 2 hr. at 37C, the reaction was terminated by the addition 500 L of 6% (v/v) TFA and duplicate 500 L aliquots were removed for glyoxylate analysis. Initial rates of glyoxylate formation from 2 (phenylthio)acetic acid were d etermined by adding 20 g of PAM to a 3.0 mL solution containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/mL catalase, 1.0 M Cu(NO3)2, 5 mM sodium ascorbate, and 10 mM 2 (phenylthio)acetic acid. At 30 min. i ntervals for 150 min., a 500 L aliquot was removed, added to a vial containing 100 L of 6% (v/v) TFA to terminate the reaction, and the concentration of glyoxylate formed measured in the acidified samples. hydroxyhippurate, a known PAL substrate, as a function of the 2 (phenylthio)acetic acid concentration were determined by adding 20 g of PAM to a 1.5 mL solution containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1. 0% (v/v) ethanol, 0.001% (v/v) Trition X hydroxyhippurate, and 010.0 mM 2(phenylthio)acetic acid. At 10 min. intervals from 10 to 60 min., a 100 L aliquot was removed, added to a vial containing 20 L of 6% (v/v) TFA to terminate the rea ction, and the concentration of glyoxylate formed measured in the acidified samples. Glyoxylate was determined by the spectrophotometric method of Christman et al [15] as modified by Katopodis and May [12] Standard curves of [glyoxylate] vs. A520 were constructed in the appropriate buffers using a glyoxylate solution that had been standardized by measuring

PAGE 46

Chapter Two: Novel Chemistry in PAM: S dealkylation 32 the glyoxylate dependent oxidation of NADH ( 340 = 6.22 x 103 M1 cm-1) as catalyzed by lactate dehydrogenase. RP HPLC s eparation of 2 (phenylthio)acetic acid and thiophenol. HPLC analyses were performed using a Hewlett Packard 1100 liquid chromatography system equipped with an auto sampler and quaternary pump system Aliquots (10 L) of the reaction mixture were analyzed using a Phenomenex Luna 5 C18 (250 x 4.60 mm) column. The mobile phase was a linear 70 100% gradient of 50 mM sodium acetate pH 6/acetonitrile (85/15) in acetonitrile. Analytes were detected at 254 nm. Analysis of steady state kinetic data Stead y state kinetic parameters were obtained by Kaleidagraph fit of the initial velocity (v) vs. substrate concentration, ([S]), data to (1) Wher e KM,app is the apparent Michaelis constant for the oxidizable substrate at fixed [ascorbate] and [O2] concentrations and VMAX,app is the apparent maximum initial velocity at saturating [S]. Initial rate studies that resulted in competitive inhibition were also analyzed by Kaleidagraph fit of v vs. [S] as a function of inhibitor concentration to equation 2. (2)

PAGE 47

Chapter Two: Novel Chemistry in PAM: S dealkylation 33 In silico docking hydroxylating monooxygenase (PHM) was obtained from the Protein Data Bank (htt p://www.rcsb.org/pdb/, 1SDW) [16] All co crystallized species determined to be redundant for ligand binding were removed (nickel, water, glycerol, and substrate). Formal charges for enzymebound copper ions and bond orders were corrected, and hydrogens were added using Maestro (www.schrodinger.com). Further receptor refinements were carried out utilizing ProteinPrep from within Maestro. Glide and Q site from the FirstDiscovery 3.0 suite (www.schrodinger.com) were used for quantum polarized ligand docking (qpld) to generate highly accurate ligand binding modes [17, 18] Computational Chemistry The top pose from the in silico docking studies of thiophenolate was used as input for the geometry optimization of the L3Cu thiophenolate system. The CuM ligand set, H242, H244, and M314, were truncated to methyl capped side chains to lessen the computational expense of geometry optimizing the system. Three systems were thus created. System 1 included the previously mentioned methyl capped side chains, Cu1+, and a single water molecule coordinated to the copper ( Figure 2.2). System 2 included, once again, the L3Cu1+ system, but the coord inated water was replaced by the qpld docked thiophenolate ( Figure 2.3). The final system, System 3, contained only the three residue CuM coordination site along with Cu1+ ( Figure 2.4). The geometries of these two systems were optimized using density functional theory with B3LYP hybrid exchange correlation and the LanL2DZ+ECP basis set [1921]

PAGE 48

Chapter Two: Novel Chemistry in PAM: S dealkylation 34 RESULTS AND DISCUSSION 2(phenylthio)acetic acid as a PAM substrate The addition of PAM to solutions containing pCl PTAA resulted in the consumption of O2 (Figure 2.5) The dependence of the initial rate of O2 cons umption on the initial concentration of PTAA, at a fixed concentration of reductant and O2, is described by equation 1 yielding the steady state kinetic values for KM,app, VMAX,app, and (V/K)app included in Table 2. 1. Investigation d,dPTAA gave a D( V/K)app H bond cleavage was occuring Incubation of 10 g/mL PAM with 10 mM PTAA for 1 hr at 37 oC yielded 96 M glyoxylate verifying that PTAA is indeed a substrate and alluding to the inactivating s pecies, thiophenol ( Figure 2. 6) 2(phenylthio)acetic acid derivatives as PAM substrates and inhibitors The pattern of O2 consumption observed for all PTAA related substrates appeared characteristic of that of a m echanism based inhibitor indicating turnover followed by enzyme inactivation. Because this inactivation was slow, analyses of initial rates in obtaining the steady state kinetic parameters is valid. Glyoxylate was also produced in all instances. Those c ompounds in which glyoxylate was not produced, however, neither consumed O2 nor inactivated PAM. The derivatives which did not consume O2 when tested as substrates did inhibit O2 consumption from N acetylglycine ( Figur e 2. 7) yielding the steady state kinetic values as described by equation 2 (Table 2. 2). LC/MS of tryps i ndigested PAM after incubation with PTAA showed no labeling (data not shown) suggesting that the enzyme was not being covalently modified as is typica l of a mechanismbased inhibitor. LC/MS analysis did, however, indicate

PAGE 49

Chapter Two: Novel Chemistry in PAM: S dealkylation 35 that thiophenol was being produced. An HPLC separation was then developed and thiophenol was successful ly found in the reaction mixture ( Figure 2. 8 ) Product mediated reversible inactivation Thiophenol is a known Cu1+ chelator and interacts more weakly with Cu2+. It was hypothesized that the thiophenol product either coordinate to active site copper, forming an E 2Cu(II)/(I) inhibitor complex, or removes the active site copper, forming a copper free E and separate Cu(II)/(I) inhibitor complexes. The copper free form of PAM is inactive [22, 23] This type of thiol mediated inhibition is welldocumented in both PAM and related enzyme dopamine dependent monooxygenases [2427] The inhibition of PAM by homocysteine extended peptides [28] and by captopril [29] has been attributed to sulfur attributed to in situ chela tion of the enzyme bound coppers [30, 31] If in situ chelation were occurring, out competing the S Cu interaction leaving the enzyme bound copper free to once again support catalysis should be possible. Thus, a reaction was initiated with 20 g of PAM under the previously mentioned condition for monitoring O2 consumption and 5 mM 2 (4 chlorophenylthio)acetic acid. Upon observing the inactivation of PAM, with the rate of O2 consumption matching that of the backgr ound prior to initiating the reaction with enzyme, the reaction was spiked with HgCl2 to a final Hg2+ concentration of 5 M. This caused a resumption of O2 consumption of 2 M, leading to the supposition that the stronger Hg S interaction was relieving th e enzymebound copper of the proposed chelating species, thiophenol ( Figure 2. 9)

PAGE 50

Chapter Two: Novel Chemistry in PAM: S dealkylation 36 Modeling of PTAA, analogs, and thiophenol in the PHM active site The X ray structures of oxidized and reduced PHM, the subunit responsible for catalyzing the copper bound coppers are ~11 apart flanking a solvent accessible active site cleft [16, 24] CuH and CuM, named for their respe ctive ligands of three histidines (CuH), and two histidines and a methionine (CuM), have different roles in catalysis as CuH is involved in electron transfer while CuM is the site of O2 activation and substrate hydroxylation. Utilizing the crystal structu re of reduced PHM in a precatalytic state with O2 coordinated to CuM and a substrate analog in the active site ( Figure 2.10) we were afforded an excellent opportunity to model the sulfur containing compounds and the proposed product responsible for in situ chelation, thiophenol. The reduced PHM crystal structure, 1SDW [16] was chosen for the docking studies. All ligands were docked along with the proposed thiophenol product and the thiophenolate species (C6H5S-), L and Lrespectively. This thiolate species should account for ~20% of the product under reaction conditions, pH 6.0. All of the analogs bound similarly to the reduced PHM crystal structure, displaying a salt b ridge of the carboxy terminus to the guanidine group of R240 similar to that of the natural glycine extended substrates of PHM [24] The greatest dissimilarity in binding modes with that of the biological substrate is the absence of the glycyl amide hydrogen bondin g to the oxygen of N316 [24] This interaction pl ays a large role in proper H for H abstraction ( Figure 2. 11) Coordination of CuM by thiophenol The binding mode from the qpld docking of thiophenolate was used as the starting structure for the geometry optimization as indicated in the Materials and Methods section. The

PAGE 51

Chapter Two: Novel Chemistry in PAM: S dealkylation 37 optimized geometries for the coordination of Cu1+ to the three residue coordination site in the CuM domain were compared between the L3Cu1+H2O system, the L3Cu1+ system, and the L3Cu1+thiophenolate system to investigate coordination stability and to determine whether the coordination of the thiophenolate would disrupt the L3Cu1+ stability. The L3Cu1+ uncoordinated system is as expected with coordination distances of 2.01, 1.98, and 2.43 to NH242,NH244, and SM314 respectively (Figure 2. 12). The L3Cu1+H2O geometry optimized system displaced a tetrahedral geometry with coordination distances of 2.05, 2.01, 2.52, and 2.19 for NH242, NH244, SM314, and H2O respectively (Figure 2. 13). This geometry is in good agreement with experimental values for this coordination state in the CuM domain. The geometry optimization for the final system revealed something unexpected. The thiophenolate anion disrupted the geometry of the CuM and changed the coordination number of the system. The geometry shifts from the typical slightly distorted tetrahedral from the water coordinated system to a trigonal pyramidal 3 coordinate system with the SM314 4.75 away from CuM (Figure 2. 14). The other coordination distances are 2.13, 2.08, and 2.38 for NH242, NH244, and S thiophenolate respectively. This change in coordination number and geometry would gr e a t l y r e duc e t h e c h a n c e o f oxygen a ctivation Conclusion In conclusion, we have shown that PAM can catal yze the dealkylation of the R S C moiety leaving a free thiol/thiolate capable of in situ coordination to enzyme bound copper. Recent work by the Merkler group has demonstrated that the inclusion of a sulfur atom in small molecule inhibitors of PAM incre ases binding affinity regardless of oxidation state ( in press). The present work suggests that a pro drug could be developed allowing PAM to catalyze the

PAGE 52

Chapter Two: Novel Chemistry in PAM: S dealkylation 38 production of the free thiol thus greatly reducing the possibility for unintentional free thiol inter actions with other metallo enzyme prior to contact with PAM.

PAGE 53

Chapter Two: Novel Chemistry in PAM: S dealkylation 39 REFERENCES 1. Merkler, D.J., Cterminal amidated peptides: production by the in vitro enzymatic amidation of glycine extended peptides and the importance of the amide to bioactivity. Enzyme Microb Technol, 1994. 16(6): p. 4506. 2. Merkler, D.J., et al., Oleic acid derived metabolites in mouse neuroblastoma N18TG2 cells. Biochemistry, 2004. 43(39): p. 1266774. 3. Merkler, K.A., et al., A pathway for the biosy nthesis of fatty acid amides. Adv Exp Med Biol, 1999. 469: p. 51925. 4. Wilcox, B.J., et al., N acylglycine amidation: implications for the biosynthesis of fatty acid primary amides. Biochemistry, 1999. 38(11): p. 323545. 5. Merkler, D.J., et al., Fatty acid amide biosynthesis: a possible new role for peptidylglycine alphaamidating enzyme and acyl coenzyme A: glycine N acyltransferase. Arch Biochem Biophys, 1996. 330(2): p. 4304. 6. King, L., 3rd, et al., The enzymatic formation of novel bile acid primary amides. Arch Biochem Biophys, 2000. 374(2): p. 10717. 7. Bradbury, A.F., M.D. Finnie, and D.G. Smyth, Mechanism of C terminal amide formation by pituitary enzymes. Nature, 1982. 298(5875): p. 6868. 8. Eipper, B.A., R.E. Mains, and C.C. Glembotski, Ide ntification in pituitary tissue of a peptide alphaamidation activity that acts on glycine extended peptides and requires molecular oxygen, copper, and ascorbic acid. Proc Natl Acad Sci U S A, 1983. 80(16): p. 51448. 9. Sheldrick, E.L. and A.P. Flint, Pos t translational processing of oxytocinneurophysin prohormone in the ovine corpus luteum: activity of peptidyl glycine alphaamidating mono oxygenase and concentrations of its cofactor, ascorbic acid. J Endocrinol, 1989. 122(1): p. 31322.

PAGE 54

Chapter Two: Novel Chemistry in PAM: S dealkylation 40 10. Prigge, S.T. et al., New insights into copper monooxygenases and peptide amidation: structure, mechanism and function. Cell Mol Life Sci, 2000. 57(8 9): p. 123659. 11. Oldham, C.D., et al., Amidative peptide processing and vascular function. Am J Physiol, 1997. 273( 6 Pt 1): p. C190814. 12. Katopodis, A.G. and S.W. May, Novel substrates and inhibitors of peptidylglycine alphaamidating monooxygenase. Biochemistry, 1990. 29(19): p. 45418. 13. Zabriskie, T.M., et al., Peptide amidation in an invertebrate: purification, characterization, and inhibition of peptidylglycine alphahydroxylating monooxygenase from the heads of honeybees (Apis mellifera). Arch Insect Biochem Physiol, 1994. 26(1): p. 2748. 14. Kolhekar, A.S., et al., Neuropeptide amidation in Drosophila: separate genes encode the two enzymes catalyzing amidation. J Neurosci, 1997. 17(4): p. 136376. 15. Christman, A.A., P.W. Foster, and M.B. Esterer, The allantoin content of blood. J Biol Chem, 1944. 155: p. 161171. 16. Prigge, S.T., et al., Dioxygen binds end on to mononuclear copper in a precatalytic enzyme complex. Science, 2004. 304(5672): p. 8647. 17. Friesner, R.A., et al., Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J Med Chem, 2004. 47(7 ): p. 173949. 18. Cho, A.E., et al., Importance of accurate charges in molecular docking: quantum mechanical/molecular mechanical (QM/MM) approach. J Comput Chem, 2005. 26 (9): p. 91531. 19. Hay, P.J. and W.R. Wadt, J. Chem. Phys., 1985. 82: p. 270283. 2 0. Hay, P.J. and W.R. Wadt, J. Chem. Phys., 1985. 82: p. 284298. 21. Hay, P.J. and W.R. Wadt, J. Chem. Phys., 1985. 82: p. 299310. 22. Freeman, J.C., J.J. Villafranca, and D.J. Merkler, Redox cycling of enzyme bound copper during peptide amidation. J Am Chem Soc, 1993. 115: p. 4923 4924.

PAGE 55

Chapter Two: Novel Chemistry in PAM: S dealkylation 41 23. Klinman, J.P. and M. Brenner, Role of copper and catalytic mechanism in the copper monooxygenase, dopamine betahydroxylase (D beta H). Prog Clin Biol Res, 1988. 274 : p. 22748. 24. Prigge, S.T., et al., Amidation of bioactive peptides: the structure of peptidylglycine alphahydroxylating monooxygenase. Science, 1997. 278(5341): p. 13005. 25. Kulathila, R., et al., Bifunctional peptidylglcine alphaamidating enzyme requires two copper atoms for maximum activity. Arch Biochem Biophys, 1994. 311(1): p. 1915. 26. Ash, D.E., et al., Kinetic and spectroscopic studies of the interaction of copper with dopamine beta hydroxylase. J Biol Chem, 1984. 259(6): p. 33958. 27. Klinman, J.P., et al., Evidence for two copper atoms/su bunit in dopamine betamonooxygenase catalysis. J Biol Chem, 1984. 259(6): p. 3399402. 28. Erion, M.D., et al., Inhibition of peptidylglycine alphaamidating monooxygenase by N substituted homocysteine analogs. J Med Chem, 1994. 37(26): p. 44307. 29. Mue ller, S.A., W.J. Driscoll, and G.P. Mueller, Captopril inhibits peptidylglycine alpha hydroxylating monooxygenase: implications for therapeutic effects. Pharmacology, 1999. 58 (5): p. 27080. 30. Nagatsu, T., H. Kuzuya, and H. Hidaka, Inhibition of dopamine beta hydroxylase by sulfhydryl compounds and the nature of the natural inhibitors. Biochim Biophys Acta, 1967. 139(2): p. 31927. 31. Palatini, P., F. Dabbeni Sala, and P. Finotti, Inhibition of dopamine betahydroxylase by captopril. Biochem Pharmacol, 1989. 38(6): p. 10113.

PAGE 56

Chapter Two: Novel Chemistry in PAM: S dealkylation 42 List of Figures Figure 2. 1. The reactions catalyzed by PHM and PAL. Bifunctional PAM is comp osed of the two monofunctional enzymes. The possible role of Fe(III) in PAL catalysis is unclea r. R N H O O O H 2 A s c or ba t e + O22 S e m i de hydr oa s c or ba t e + H2O HSHR R N H O O O H O H HR HR O H O R N H2 O O 2 C u ( I I ) F e ( I I I ) / C a2 +/ Z n ( I I )P H M P A L

PAGE 57

Chapter Two: Novel Chemistry in PAM: S dealkylation 43 Figure 2. 2. The CuM domain with the L3 side chains methyl capped and a single water molecule coordinated to Cu1+. Note, the carbons are displa yed in teal, the nitrogens in blue, the hydrogens in white, the oxygen in red, the sulfur in yellow, and the copper in orange.

PAGE 58

Chapter Two: Novel Chemistry in PAM: S dealkylation 44 Figure 2. 3. The CuM doma in with the L3 side chains methyl capped and thiophenolate coordinated to Cu1+. Note, the carbons are displayed in teal, the nitrogens in blue, the hydrogens in white, the oxygen in red, the sulfur in yellow, and the copper in orange.

PAGE 59

Chapter Two: Novel Chemistry in PAM: S dealkylation 45 Figure 2. 4. The CuM domain with the L3 side chains methyl capped and coordinated to Cu1+. Note, the carbons are displayed in teal, the nitrogens in blue, the hydrogens in white, the oxygen in red, the sulfur in yellow, and the copper in orange.

PAGE 60

Chapter Two: Novel Chemistry in PAM: S dealkylation 46 Figur e 2. 5. The PAM Dependent Consumption of O2 in the Presence of 2 (4 chlorophenylthio)acetic acid. O2 of PAM (A) to a 2.0 mL solution containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol 0.001% (v/v) Triton X ascorbate, and 5 mM 2 (4 chlorophenylthio)acetic acid. O2 consumption was measured as described in the materials and methods section. Note that the background rate has been removed for clarity 200 202 204 206 208 210 212 0 50 100 150 200 250 [O 2 ],(M) Time (s)

PAGE 61

Chapter Two: Novel Chemistry in PAM: S dealkylation 47 Figure 2. 6. Proposed 2(phenylthio)acetic acid c l e avage b y P A M hydroxylation (PHM) and subsequent Sglyoxylate. Thiophenolate is a known Cu1+ coordinator. S O O H 2 A s c or ba t e + O22 S e m i de hydr oa s c or ba t e + H2O S H O O H 2 C u ( I I ) F e ( I I I ) / C a2 +/ Z n ( I I )P A M O H

PAGE 62

Chapter Two: Novel Chemistry in PAM: S dealkylation 48 Figure 2. 7. Inhibition of PAM by 2 (benzoylthio)acetic acid. Initial rates were determined at 37 C as described in the Materials and Methods section. The points are the experimentally determined initial rates and the curve was drawn using the c onstants obtained by computer fit to eq. 2. The error bars represent the standard deviation of duplicate measurements of the initial rates.

PAGE 63

Chapter Two: Novel Chemistry in PAM: S dealkylation 49 Figure 2. 8. R P HPLC analysis of PAM reaction containing PTAA as the substr ate. The mobile phase was a linear 70100% gradient of 50 mM sodium acetate pH 6/acetonitrile (85/15) in acetonitrile. The chromatograms shown are a thiophenol standard, a 2(phenylthio)acetic acid standard, and a reaction mixture after treatment with PA M, respectively. Analytes were detected at 254 nm.

PAGE 64

Chapter Two: Novel Chemistry in PAM: S dealkylation 50 Figure 2. 9. E f f e c t s of m ercury c om p e t i t i on w i t h t h i ol f o r c op p e r on O2 c on s u m p t i on The reaction was prepared as indicated in the Materials and Methods section for monitoring oxygen consumption. Upon initiation of the reaction with the addition of PAM (A), the reaction went to completion which is indicated by the slope of oxygen cons umption matching that of the background rate prior to the addition of enzyme. The sample was then spiked with HgCl2 to a final concentration of 5 M (B). Oxygen consumption then resumed until 1.2 M of additional oxygen had been consumed. Note that the b ackground rate has been removed for clarity. 200 202 204 206 208 210 212 0 100 200 300 400 500 [O 2 ], (M) Time (s)

PAGE 65

Chapter Two: Novel Chemistry in PAM: S dealkylation 51 Figure 2. 10. PHM crystal structure. The 1SDW PHM crystal structure with the secondary s tructure rendered as a ribbon. The copper is depicted in orange, the bound molecular oxygen in red, and the substrate IYT rendered in tube format colored by atom (red=oxygen, green=iodine, teal=carbon, blue=nitrogen).

PAGE 66

Chapter Two: Novel Chemistry in PAM: S dealkylation 52 Figure 2.11. PTAA and N benzylglycine dock ed into the PHM crystal structure. (A) PTAA was docked into the PHM crystal structure (1SDW). The carboxylate forms a salt bridge with the R240 guanidino as expected. However, the absence of an amide hydrogen available for hydrogen bonding to the N316 oxygen allows for a great deal of movement H for abstraction by an activated oxygen species. (B) N benzylglycine docked for comparison to illustrate proper hydrogen bonding and positioning. Non amide hydrogens wer e omitted for clarity.

PAGE 67

Chapter Two: Novel Chemistry in PAM: S dealkylation 53 Figure 2.12. L3Cu1+ geometry optimized using the LanL2dz+ECP basis set.

PAGE 68

Chapter Two: Novel Chemistry in PAM: S dealkylation 54 Figure 2.13. L3Cu1+H2O system geometry optimized with the LanL2dz+ECP basis set.

PAGE 69

Chapter Two: Novel Chemistry in PAM: S dealkylation 55 Figure 2.14. L3Cu1+thiophenolate system geometry optimized with the LanL2dz+ECP basis set.

PAGE 70

Chapter Two: Novel Chemistry in PAM: S dealkylation 56 TABLES Table 2.1. Steady s tate kinetic c onstants for 2 (phenylthio)acetic acid and a nalogs Name Structure K M, app (mM) V max, app (s-1) ( V MAX / K M ) app (mM-1 s-1) [(1-H -Indol -2ylcarbonyl)thio]acetic acid 2.3 0.6 0.4 0.01 0.17 0.05 2-(phenylthio)acetic acid 0.95 0.13 1.5 0.2 1.6 0.3 D D 2 (phenylthio)acetic acid 3.1 0.6 2.6 0.2 0.84 0.17 2 (4 chlorophenylthio)acetic acid 0.53 0.11 1.8 0.1 3.4 0.7 2-(4-methylphen ylthio)acetic acid 4.2 0.2 2.3 0.4 0.55 0.10 H N S C O O H O S C O O H S C O O H D D S C O O H C l S C O O H

PAGE 71

Chapter Two: Novel Chemistry in PAM: S dealkylation 57 Table 2.2. PAM Inhibitors. Name Structure K is,APP (mM) 2 (benzoylthio)acetic acid S O C O O H 1.9 0.1 2 (2 nitropheylthio)acetic acid S C O O H N O2 0.14 0.02 N benzoyl D alanylthioglycolic acid S H O O C H N O O 0.89 0.01

PAGE 72

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 58 Introduction hydroxylating monooxygenase (PHM) is a copper oxygen and ascorbatedependent enzyme responsible for activating glycine extended peptide prohormones through an oxidative cleavage yielding the corresponding peptide amide [1 6] This reaction This b hydroxylating monooxygenase (PHM) and the zinc calcium and irondependent peptidylglycine amidoglycolate lyase (PAL) domains, responsible for the hydrogen abstraction and hydroxylation, and the hydro lysis of the amide Figure 3.1) [7, 8] monooxygenase, have been studied extensively and are mechanistically simila r [9] The products of these reactions are essential signal ing molecules which are stored in secretory granules and used for intercellular communication. PAM activates several inflammatory hormones such as s ubstance P. Previous studies have targeted PAM using mechanism based inactivators to inhibit carrageenan i nduced edema as well as all three phases of adjuvant induced polyarthritis in rats [10] Substance P is also found in the spinal fluid of f ibromyalgia patients, making PAM a possible drug target for such inflammat ory diseases. Genetically engineered mice and Drosophila lacking a functioning PHM gene generally die as embryos [1113] Thus, simply targeting PAM or PHM activity would be lethal. However, with advances in tis sue specific targeted drug delivery, this will become less of an obstacle.

PAGE 73

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 59 Cinnamic acid and its derivatives are found in fruits, vegetables, and flowers. Cinnamic acid is an intermediate in the shikimate pathway, the pathway which links carbohydrate met abolism to the biosynthesis of aromatic compounds in plants. Recent studies on the utility of cinnamic acid derivatives as insulin releasing agents have proven certain ring substituted cinnamic acids very effective in lowering plasma glucose levels [14, 15] It has been suggested that cinnamic acid derivatives may be effective as treatment of diabetes mellitus for the regulation of blood glucose levels through the stimulation of insulin secretion [14] Here we report on the investigation of several inactivators of PAM related to cinnamic acid. These compounds, whose derivatives are used as food and fragrance additives, are structura lly similar to a known suicide substrate, 4 phenyl 3butenoic acid [16] Our results demonstrate that while the inactivation is turnover dependent, the inactivator does not undergo any detectible chemistry as is typical of a suicide substrate. This finding merits further study into the effects of cinnamic acid inactivation of PAM in vivo as a potential drawback of using these molecules as diabetes therapeutics. Materials and Methods Materials Cinnamic acid, 2 trifluorocinnamic acid, 3 (3 pyridyl)acrylic acid, phenylpropiolic aci d, 3,4methylenedioxycinnamic acid, N,N dimethylaminocinnamic acid, maleamic acid, N phenylmaleamic acid, Urocanic acid, 4 aminocinnamic acid, and perdeuterated cinnamic acid were from Sigma. Bovine catalase was from Worthington and 4 anilino 4oxobut 2e noic acid was from Enamine. Recombinant rat PAM was a gift from Unigene Laboratories, Inc. All other reagents were of the highest quality commercially available.

PAGE 74

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 60 Synthesis Dansyl 4aminocinnamic acid A solution of dansyl chloride (400 mg, 1.5 mmol) in a minimal dry pyridine (~2 mL) was added dropwise to a N2purged solution o f 4aminocinnamic acid (500 mg, 3.1 mmol) in 50 mL of dry pyridine at 60 C. After 24 hours, the reaction was diluted then extracted with Et2O (100 mL x 3), yielding a yellow oil. Crystallization of the final dansyl 4aminocinnamic acid was done with methanol/ddH2O (285 mg, 48%). 3phenyloxirane 2carboxylic acid Cinnamic acid (5 g, 34 mmol) dissolved in a 30 mM NaOH(aq) (~100 mL) was combined with an oxone (30 g, 98 mmol) solution in H2O ( 100 mL ) A constant pH of six was maintained over six hours through the addition of NaOH then stirred for an additional 16 hours. Acid extraction (pH 1 ) of the reaction was performed with Et2O (500 mL) by vigorously stirring the bi phasic solution for ~2 hours. The organic layer was then separated and removed under reduced pressure. The resulting residue was crystallized with ethanol/H2O (1.6 gm, 29%). In silico ligand docking The crystal structure for reduced peptidylglycine hydroxylating monooxygenase (PHM) was obtained from the Protein Data Bank (http://www.rcsb.org/pdb/, 1SDW) [17] All co crystallized species determined to be redundant for ligand binding were removed (nickel, w ater, glycerol, and substrate). Formal charges for enzymebound copper ions and bond orders were corrected, and hydrogens were added using Maestro (www.schrodinger.com). Further receptor refinements were carried out utilizing ProteinPrep from within Maes tro. Investigation

PAGE 75

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 61 of inhibitor binding modes were performed using Glide [18] and Qsite [19] jointly for quantum polarized ligand docking. This method generates highly accurate binding modes by quantum mechanically calculating the partial atomic charges of the docked ligand using B3LYP/ 631G* within the receptor and subsequently re docking the ligand [20, 21] H dissocia tion energies H for all cinnamic acid analogs were calculated using Jaguar [22] DFT calculations were performed with the B3LYP hybrid exchange correlation functional and the 6 31G* basis set. The geometries were optimized for all H bond was then increased from 1.08 3.0 the mole cule rigid ( Figure 3.2), and single point energies were calculated Inhibition of O2 consumption from N acetylglycine by cinnamic acid Reactions at 37.0 0.1 C were initiated by the addition of PAM (35 g) into 2.0 ml of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, 1.0 45 mM N acetylglycine, and 0 9 mM cinnamic acid. Initial rates were measured by following the PAM dependent consumption of O2 using a Yellow Springs Instrument Model 5300 oxygen monitor. Inactivation of PAM by cinnamic acid Inactivation reactions of 100 L containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 0 3 mM cinnamic acid were initiated by the addition of enzyme and incubated at 37 C. Aliquots of 15 L were withdrawn at various intervals and diluted into 2.0

PAGE 76

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 62 mL reactions containin g 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 20 mM N acetylglycine and monitored for O2 consumption. Reversibility of inactivation An inactiva tion reaction of 250 L containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 7 mM cinnamic acid was initiated by the addition of enzyme and incubated at 37 C for 2 hours. The reaction was then extensively dialyzed and concentrated to 50 L. The concentrate was then used to initiate a 1.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, and 10 mM hippurate. At various time intervals, a 100 L aliquot was removed, added to a vial containing 20 L of 6% (v/v) TFA to terminate the reaction, and the concentration of glyoxylate formed measured in the acidified samples to test for the recovery of activity. PAM labeling by dansyl 4aminocinnamic acid Covalent modification of PAM was investigated by using dansyl 4aminocinnamic acid, a fluorescent molecule, as an inactivator. A 0.5 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 1.0 mM dansyl 4aminocinnamic acid was initiated by the addition of 50 g of enzyme and incubated at 37 C for 3 hours. The reaction mixture was di alyzed against 100 mL of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, and 0.001% (v/v) Triton X 100 for 4 hours changing the dialysis buffer every hour. The reaction mixture was then concentrated by ultrafiltration to ~100 L. This sample was t hen

PAGE 77

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 63 trypsin digested and the peptide fragments were analyzed by RP HPLC utilizing a fluorescence detector. PAM labeling by 14C cinnamic acid Enzyme modification was also examined using radio labeled cinnamic acid. A 250 L reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 3 Ci of 14C cinnamic acid was initiated by the addition of 18 52 g of PAM. The reaction was allowed to in cubate at 37 C for 3 hours before ultrafiltration was performed. The reaction was then washed with 200 L of 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, and 0.001% (v/v) Triton X 100 and ultra filtration repeated. The underside of the filtrat ion membrane was then washed with poly(ethylene glycol) to remove excess nonenzyme bound radio labeled cinnamic acid. Counts per minute were then compared using a scintillation counter. PAM modification through inactivation by cinnamic acid M o di f i c a t i o n of the PAM active site by t h e r e a c t i ve C u/ O s pe c i e s was also investigated as a possible means of cinnamic acid mediated inactivation. A 100 L reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorb ate, and 3 mM cinnamic acid was initiated with 20 g of PAM and incubated at 37 C for 12 hours. The enzyme was then dialyzed against 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100 for 4 hours and the volume reduced to ~40 L by ultra filtration. The enzyme was then analyzed by MALDI TOF against control for any modification. The reaction was then repeated as previously stated

PAGE 78

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 64 followed by trypsin digestion. The reaction was then analyzed by LC/MS, LC/MS/MS, and MALDI TOF. PAL inactivation assay To determine if the cinnamic acid was also inactivating PAL activity by PAM, initial rates of glyoxylate fo hydroxyhippurate, a known PAL substrate, were monitored. Reactions of 20 g of PAM in a 100 L solution containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, and a concentration of cinnamic acid or an alog equal to 5times KI of PHM related inhibition for that particular inhibitor were incubated for 2 hours. A 20 L aliquot was removed and used to initiate a 1.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, and 10 mM hydroxyhippurate. At 10 minute intervals from 10 to 60 min., a 100 L aliquot was removed, added to a vial containing 20 L of 6% (v/v) TFA to terminate the reaction, and the concentration of glyoxylate formed measured in the acidified samples. Glyoxyla te was determined by the spectrophotometric method of Christman et al [23] as modified by Katopodis and May [24] Standard curves of [glyoxylate] vs. A520 were constructed in the appropriate buffers using a glyoxylate solutio n that had been calibrated by measuring the glyoxylate dependent oxidation of NADH ( 340 = 6.22 x 103 M1 cm-1) as catalyzed by lactate dehydrogenase. Analysis of steady state kinetic data Initial rate studies that resulted in competitive inhibition were also analyzed by Kaleidagraph fit of v vs. [S] as a function of inhibitor concentration to equation 1.

PAGE 79

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 65 (1) Where KM,app is the apparent Michaeli s constant for the oxidizable substrate at fixed [ascorbate] and ambient [O2] concentrations and VMAX,app is the apparent maximum initial velocity at saturating [S]. Inactivation kinetics were analyzed by Kaleidagraph fit of 1/kobs vs 1/[ I ] to equati on 2. (2) Where kobs is the observed rate of inactivation, and kinact is the intrinsic rate of inactivation of enzyme. Results Cinnamic acid and analogs as inhibitors of PAM Cinnamic acid and several analogs were initially investigated as small molecule inhibitors of PAM. Cinnamic acid was shown to inhibit the consumption of O2 from N acetylglycine in a competitive manner ( Figure 3.3) yielding the steady state kinetic values as described by equation 1 (Table 3. 1). All cinnamic acid analogs were assumed competitive and also analyzed according to equation 1. Inactivation of PAM by cinnamic acid Cinnamic acid was inve stigated as an inactivator of PAM by the dilution method, a common method for determining the kinetic parameters of timedependent inactivators [25]

PAGE 80

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 66 Inactivation experiments performed with cinnamic acid in dicated that it is a timedependent inactivator of PAM ( Figure 3.4). The inactivation was p s e udo first order and concentration dependent, as well as O2and ascorbate dependent ( Figure 3. 5). Substrate was also found to protect against inactivation of PAM by cinnamic acid. Perdeuterated cinnamic acid was also investigated in search of an isotope effect on kinact as evidence for H abstraction, though one was not observed (Table 3. 2). In activation of PAL activity Cinnamic acid and analogs were tested as PAL inactivators as well to investigate whether or not the PAM activation occurring was specific to the PHM domain. The experiment indicates that PAL is completely unaffected by cinnamic acid and the various analogs tested and retained 100% activity when compared to the control reaction ( Figure 3.6). Irreversible inactivation Cinnamic acid was able to irreversibly inactivate PAM. Extensive dialysi s of a reaction mixture incubated with cinnamic acid yielded dead enzyme incapable of producing glyoxylate from hippuric acid ( Figure 3.7). Non labeled PAM Multiple experiments were performed in an attempt to verify covalent linkage of cinnamic acid to PAM. Both the fluorescent labeling with dansyl 4aminocinnamic acid, and the radio labeling experiments indicated that cinnamic acid was not covalently linking to the enzyme. Multiple methods of mass analysis were pe rformed in an attempt to elucidate any auto oxidation or hydroxylation of the PAM active site as has been hypothesized in earlier studies

PAGE 81

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 67 with 4 phenyl 3butenoic acid [26] However, we were unable to detect any modification of the enzyme after exhausting all efforts and resources ( Figure 3.8). Docking of cinnamic acid and analogs Docking of the investiga ted compounds yielded poses in agreement with the competitive nature of inhibition observed experimentally. The compounds bind forming a salt bridge between the carboxy terminus and the guanidino group of R240 as has been previously shown with glycine ext ended substrates [27] H in close proximity to CuM ( Figure 3.9). However, the lack of glycyl amide hydrogen to H bond with the N316 oxygen coupled with the compounds small size preventing any interaction w ith the nearby hydrophobic pocket would allow for a great deal of movement within the active site making hydrogen abstraction difficult. Computation results for bond energy The ab initio calculations performed to estimate the bond dissociation energies of H bond for cinnamic acid and the investigated analogs indicate that more than 110 kcal/mol are needed (Table 3. 3). Discussion Cinnamic acid is a structural analogue of 4phenyl 3butenoic acid (PBA), a well documented mechanism based, irrevers ible inhibitor (inactivator) of the PHM domain as both molecules contain an olefin moiety [16, 24] While PBA has been suggested to be hydroxylated [26] none of the compounds tested within this study displayed oxygen consumption when screened for activity even though inactivation was O2and

PAGE 82

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 68 ascorbatedependent. Fur ther evidence that hydroxylation was absent with cinnamic acid was the lack of a deuterium kinetic isotope effect for kinact o r ki n a c t/ KI. Although the presence of an absence may suggest that the magnitude of Dkinact may be suppressed by much faster steps in the mechanism. The lack of PHM inactivation with 3 phenyloxirane 2carboxylic acid suggests that C H cleavage is required to generate the inactivating olefin int ermediate species in the hydroxylation pathway. The C H bond dissociation energies, displayed in Table 3. 3, display enthalpies for each conjugated species to be very close to each other ( a di f f e r e n c e o f o nl y 13 kcal/mole). Therefore, if the inactivation species were ultimate ly dependent only on a C H cleavage step, 3 phenyloxirane 2carboxylic acid would be expected to inactivate PHM. T hi s s ugge s t s t h a t t h e i na c t i v a t i o n i s de pe n de n t o n t h e pr e s e n c e o f a n o l e f i n w hi l e t h e l a c k o f i na c t i v a t i o n by p h e nyl pr o pi o l i c a c i d, t h e a l k y ne de r i v a t i v e s ugge s t s t h a t t h e C H i s a l s o a pr e r e qu i s i t e f o r i na c t i va t i o n T hi s m a y s ugge s t t h a t t h e i n a c t i va t i n g s pe c i e s i s a n i n t e r m e d i a t e a l o n g t h e hy dr o xy l a t i o n pa t h w a y a f t e r t h e a b s t r a c t i o n o f t h e C H a n d t h a t t h e i n a c t i v a t i o n i s de pe n d e n t o n t h e o l e f i n. T hi s s pe c i e s m a y be a vi n yl r a d i c a l pr e s e n t o n t h e C a s t hi s r a d i c a l w o ul d n o t de l o c a l i z e The instrumental nature of the olefin radical generated upon hydrogen abstraction further suggest that dynamical freedom of this small inactivator may also be important to this inact ivation mechanism. The modeling performed suggests that there are few interactions to hold cinnamic acid in the active site other than the salt bridge with R240. This is also reflected in the poor KI value for cinnamic acid (3.6 mM, Table 1). This may b e directly attributed to the H and properly position it for abstraction along with the inability of the phenyl ring to bury itself into the hydrophobic pocket to further stabilize the

PAGE 83

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 69 molecule. This results in a sign ificant increase in the conformational sampling and a decrease in the probability of proper wave function overlap between donor and acceptor [28] Upon successful hydrogen abstraction, the radical character carbons. Assuming the Cu alkoxide intermediate, hydroxylation of such a mobile species with delocalized radical character would prove difficult. Thus, the reactive cinnamic acid species may leave the active site al lowing for hydroxylation to occur by the addition of water to the radical intermediate. This would leave the activated Cu oxo species to either modify the enzyme in a way undetectable by mass analysis or perhaps linger in a trapped dynamic state unable to return to the free enzyme state ( Figure 3.10). Conc lusion In conclusion, we have demonstrated that potential diabetes drug candidates are capable of PAM inactivation. Because of the importance of PAM in the biosynt hesis of amidated peptide hormones, these results are of great pharmaco logical significance. As the development of small paid to the possible inactivation of PAM. These findings are also therapeutically relevant as PAM is over expressed in both small cell lung cancer and prostate cancer [2931] making it a possible drug target.

PAGE 84

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 70 References 1 Tamburini, P.P. and S.D. Young, J Am Chem Soc, 1989. 111: p. 1933. 2. Merkler, D.J. and S.D. Young, Recombinant type A rat 75 kDa alphaamidating enzyme catalyzes the conversion of glycine extended peptides to peptide amides via an alphahydroxyglycine i ntermediate. Arch Biochem Biophys, 1991. 289(1): p. 1926. 3. Merkler, D.J., et al., 18O isotopic 13C NMR shift as proof that bifunctional peptidylglycine alphaamidating enzyme is a monooxygenase. Biochemistry, 1992. 31(32): p. 72828. 4. Bradbury, A.F., M.D. Finnie, and D.G. Smyth, Mechanism of C terminal amide formation by pituitary enzymes. Nature, 1982. 298(5875): p. 6868. 5. Eipper, B.A., R.E. Mains, and C.C. Glembotski, Identification in pituitary tissue of a peptide alphaamidation activity that ac ts on glycine extended peptides and requires molecular oxygen, copper, and ascorbic acid. Proc Natl Acad Sci U S A, 1983. 80(16): p. 51448. 6. Kulathila, R., et al., Bifunctional peptidylglcine alphaamidating enzyme requires two copper atoms for maximum activity. Arch Biochem Biophys, 1994. 311(1): p. 1915. 7. Bell, J., et al., Structural and functional investigations on the role of zinc in bifunctional rat peptidylglycine alphaamidating enzyme. Biochemistry, 1997. 36(51): p. 1623946. 8. De, M., et al. Role for an essential tyrosine in peptide amidation. J Biol Chem, 2006. 281(30): p. 2087382. 9. Klinman, J.P., The copper enzyme family of dopamine betamonooxygenase and peptidylglycine alphahydroxylating monooxygenase: resolving the chemical pathway for substrate hydroxylation. J Biol Chem, 2006. 281(6): p. 30136. 10. Bauer, J.D., et al., Anti inflammatory effects of 4 phenyl 3 butenoic acid and 5 (acetylamino) 4 oxo 6 phenyl 2 hexenoic acid methyl ester, potential inhibitors of neuropeptide bioactiv ation. J Pharmacol Exp Ther, 2007. 320(3): p. 11717.

PAGE 85

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 71 11. Jiang, N., et al., PHM is required for normal developmental transitions and for biosynthesis of secretory peptides in Drosophila. Dev Biol, 2000. 226(1): p. 11836. 12. Czyzyk, T.A., et al., Targete d mutagenesis of processing enzymes and regulators: implications for development and physiology. J Neurosci Res, 2003. 74(3): p. 44655. 13. Thomas, S.A., A.M. Matsumoto, and R.D. Palmiter, Nature, 2002. 374: p. 643646. 14. Adisakwattana, S., P. Moonsan, and S. Yibchok Anun, Insulin releasing properties of a series of cinnamic acid derivatives in vitro and in vivo. J Agric Food Chem, 2008. 56 (17): p. 783844. 15. Yibchok anun, S., et al., Insulinsecretagogue activity of pmethoxycinnamic acid in rats, per fused rat pancreas and pancreatic betacell line. Basic Clin Pharmacol Toxicol, 2008. 102(5): p. 47682. 16. Rhodes, C.H. and C. Honsinger, Structure activity relationships among inhibitors of peptidylglycine amidating monooxygenase. Ann N Y Acad Sci, 1993 689: p. 6636. 17. Prigge, S.T., et al., Dioxygen binds endon to mononuclear copper in a precatalytic enzyme complex. Science, 2004. 304(5672): p. 8647. 18. Schrodinger, GLIDE 2000: Portland, OR. 19. Schrodinger, Qsite 2000: Portland, OR. 20. Friesne r, R.A., et al., Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J Med Chem, 2004. 47(7): p. 173949. 21. Cho, A.E., et al., Importance of accurate charges in molecular docking: quantum mechanica l/molecular mechanical (QM/MM) approach. J Comput Chem, 2005. 26 (9): p. 91531. 22. Jaguar, version 6.5, Schrodinger, LLC, New York, NY, 2006 23. Christman, A.A., P.W. Foster, and M.B. Esterer, The allantoin content of blood. J Biol Chem, 1944. 155: p. 161171. 24. Katopodis, A.G. and S.W. May, Novel substrates and inhibitors of peptidylglycine alphaamidating monooxygenase. Biochemistry, 1990. 29(19): p. 45418.

PAGE 86

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 72 25. Silverman, R.B., Mechanism based enzyme inactivators. Methods Enzymol, 1995. 249: p 24083. 26. Driscoll, W.J., et al., Peptidylglycine alphahydroxylating monooxygenase generates two hydroxylated products from its mechanism based suicide substrate, 4 phenyl 3 butenoic acid. Biochemistry, 2000. 39(27): p. 800716. 27. Prigge, S.T., et a l., Amidation of bioactive peptides: the structure of peptidylglycine alphahydroxylating monooxygenase. Science, 1997. 278(5341): p. 13005. 28. Nagel, Z.D. and J.P. Klinman, Tunneling and dynamics in enzymatic hydride transfer. Chem Rev, 2006. 106(8): p. 3095118. 29. Martinez, A., et al., Expression of peptidyl glycine alpha amidating mono oxygenase (PAM) enzymes in morphological abnormalities adjacent to pulmonary tumors. Am J Pathol, 1996. 149(2): p. 70716. 30. Yang, H.K., et al., Correlation of expre ssion of bombesinlike peptides and receptors with growth inhibition by an anti bombesin antibody in small cell lung cancer cell lines. Lung Cancer, 1998. 21(3): p. 16575. 31. Rocchi, P., et al., Expression of adrenomedullin and peptide amidation activity in human prostate cancer and in human prostate cancer cell lines. Cancer Res, 2001. 61 (3): p. 1196206.

PAGE 87

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 73 Figures and Tables List of Figures Figure 3. 1. PAM catalytic scheme. R N H C O O H O R N H C O O H O O H O2H2O R N H2 O 2 A s c orb a t e 2 S e m i de hy droa s c orb a t e G l yox yl a t eZ n ( I I ) 2 C u( I I )P HM P A L

PAGE 88

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 74 Figure 3. 2. Illustration of carbon hydrogen bond elongation for the DFT calculation of bond dissociation energies.

PAGE 89

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 75 Figure 3. 3. Inhibition of PAM by cinnamic acid. Initial rates were determined at 37 described in the Materials and Methods section. The points are experimentally determined initial rates. The lines were computer fit to the data using equation 1.

PAGE 90

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 76 Figure 3. 4. Inactivation plots for cinnamic acid. Experiments were performed as indicated in the Materials and Methods section and analyzed as indicated in the analysis subsection. 0 5 10 15 20 25 30 35 Control 3mM Cinn 2mM Cinn 1mM Cinn 0.5mM Cinn 0 10 20 30 40 50 60 0 0.5 1 1.5 2 2.5 1/[ cinn ] ( mM ) Kitz Wilson Plot 1/ kobs(min)Time (min)% EnzymeActivity100 10Time dependent Inactivation of PHM by Cinnamate

PAGE 91

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 77 Figure 3. 5. Ascorbate and O2 dependence of PHM inactivation investigated by oxyg en consumption as described in the Materials and Methods section. 1 10 100 0 5 10 15 20 25 30 35 (+) asc (+) cinn Buffer (+) cinn (+) asc ( ) cinn Time (min)% Residual ActivityAscorbate Dependence of Inactivation Oxygen Dependence of Inactivation

PAGE 92

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 78 Figure 3. 6. Cinnamate analogs as inactivators of the PAL activity of PAM by the glyoxylate assay. 0 50 100 150 200 250 300 350 0 10 20 30 40 50 60 70 [Glyoxylate], M Time (min ) PAL Inactivation by cinnamic acid and Derivatives: Glyoxylate Assay Control 1. Cinnamic acid 2. 2 trifjluoro cinnamic acid 3. 4 nitro cinnamic acid 4. 4 anilino 4 oxobut 2 enoic acid 5. 3 (3 pyridyl)acrylic acid 6. phenylpropiolic acid 7. 3,4 methylene dioxy cinnamic acid 9. maleamic acid 10. N phenyl maleamic acid

PAGE 93

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 79 F igure 3. 7. Reversibility of inactivation of PAM by cinnamic acid. BTime (min)[Glyoxylate] ( M) Reversibility of Inactivation by Cinnamic acid: Glyoxylate Assay20 0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 Control 1mM Asc Control 7mM Cinnamic acid BTime (min)[Glyoxylate] ( M) Reversibility of Inactivation by Cinnamic acid: Glyoxylate Assay20 0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 Control 1mM Asc Control 7mM Cinnamic acid BTime (min)[Glyoxylate] ( M) Reversibility of Inactivation by Cinnamic acid: Glyoxylate Assay20 0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 Control 1mM Asc Control 7mM Cinnamic acid Reversibility of Inactivation by Cinnamic acid: Glyoxylate Assay20 0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 Control 1mM Asc Control 7mM Cinnamic acid Control 1mM Asc Control 7mM Cinnamic acid Control 1mM Asc Control 7mM Cinnamic acid

PAGE 94

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 80 Figure 3. 8. MALDI TOF overlay of Trypson digested active (blue) and cinnamate inactivated (green) PAM.

PAGE 95

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 81 Figure 3. 9. Cinnamate docked in the active site of the PHM crystal structure (1SDW).

PAGE 96

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 82 Figure 3. 10. P os t u l at e d c innamate inactivation scheme s where cinnamate radical is released prior to hydroxylation leavin g the activated Cu alkoxide species an d w h e r e ac t i vat e d C u / O s p e c i e s i s u n ab l e t o a b s t r ac t C H at om EC uH IIC uM I I H O O H H O22 A s cEC uH IC uM II O O H O O H H H O O H EC uH IC uM II O OH EC uH IC uM II O OH 1 eH2O H+ H O O H EC uH IIC uM II O H O O H H O O H EC uH IIC uM II O EC uH IIC uM I I H O O H H O22 A s cEC uH IC uM II O O H O O H H H O O H H EC uH IC uM II O O

PAGE 97

83 List of Tables Table 3.1. Inhibition constants calculated from computer f it of data to equation 1. Structure Name KI (mM) O HO Cinnamic acid 3.5 0.2 O HO F3C 2-trifluorocinnamic acid 0.22 0.02 O HO NO2 4-nitrocinnamic acid 0.56 0.04 O OH O N H 4-anilino-4-oxobut-2-enoic acid 3.5 0.3 O HO N 3-(3-pyridyl)acrylic acid 5.9 0.6 O OH Phenylpropiolic acid 2.4 0.3 O OH O O 3.4-methylenedioxycinnamic acid 0.28 0.04

PAGE 98

84 O HO N N,N -dimethylaminocinnamic acid 3.1 0.5 O OH O NH2 Maleamic acid 0.89 0.13 O OH O N H N-phenylmaleamic acid 1.7 0.2 N NH O HO Urocanic acid 10.5 1.6 O O HO 3-phenyloxirane-2-carboxylic acid 1.6 0.1 O HO NH2 4-aminocinnamic acid 0.54 0.06 O HO NH S N OO Dansyl-4-aminocinnamic acid 0.011 0.001

PAGE 99

85 O HO D D D D D D D Perdeuterated cinnamic acid 3.5 0.3

PAGE 100

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 86 Table 3.2. Inactivation constants calculated by KitzWilson analysis using the dilution assay method as outline in the materials and methods section. Compound k inact (min 1 ) k inact / K I (mM 1 min 1 ) Cinnamic acid 0.15 0.02 0.04 0.006 N,N dimethylaminocinnamic acid 0.08 0.01 0.03 0.005 4 aminocinnam ic acid 0.03 0.005 0.06 0.01 Dansyl 4 aminocinnamic acid 0.21 0.03 19.1 3.2 Perdeuterated cinnamic acid 0.15 0.01 0.04 0.004

PAGE 101

Chapter Three: Cinnamic Acid Derivatives as Inactivators of PAM 87 Table 3.3. Calculated bond dissociation energies of selected cinnamate analogs by DFT/6 31G*. Structure Name kcal/mol O H O Cinnamic acid 114 O H O N H S N O O Dns 4 aminocinnamic acid 116 O O H O 3 phenyloxirane 2 carboxylic acid 117 O H O N H2 4 aminocinnamic a cid 123 O H O N O2 4 nitrocinnamic acid 127

PAGE 102

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 88 Introduction amidat ing monooxygenase (PAM) is a bi functional metalloenzyme responsible for the conversion of glycine extended peptide prohormones to the active hormone, peptide amide [1, 2] Cat alysis proceeds through via a reactive Cu oxo species followed by hydroxylation, and finally the cleavage of the N yield glyoxylate and the peptide amide. Two domains in PAM are responsible for this che hydroxylating monooxygenase (PHM) domain is responsible for the hydrogen abstraction and subsequent hydroxylation while the peptidylglycine amidoglycolate lyase (PAL) domain is responsible for the cleavage of the N Figure 4.1) [3] PHM activity is copper ascorbate and O2dependent while PAL catalysis is zinc calcium and iron dependent [4 9] PHM and it r e l a t e d secretory granules and used for intercellular communication [10] They catalyze very similar reactions utilizing a reactive copper oxygen species to hydroxylate their very different respective subst rates ( Figure 4.2 and Drosophila melanogaster lacking a function a l as embryos [1113] mechanistic study. Many of the proposed mechanistic schemes involve a substrate free radical intermediate ( Figure 4.3). A l l of everal [1417]

PAGE 103

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 89 been applied towards th e elucidation of both mechanisms. Here we report experiments that provide evidence for a radical intermediate in PAM catalysis. Ring substituted suicide substrates are used to probe electronic effects to de fine the nature of the r e a c t i ve in termediate. The results provide evidence for a radical mechanism and Materials and Methods Materials 4P henyl 3butenoic acid and N acetylglycine were from S igma Aldrich. 4 (4 c hlorophenyl) 3butenoic acid, 4 (4 methoxyphenyl) 3butenoic acid, 4 (3 chlorophenyl) 3butenoic acid, and 4 (3 methoxyphenyl) 3butenoic acid were a gift from Dr. John Vederas. Bovine catalase was from Worthington. Recombinant rat PAM was a gift from Un igene Laboratories, Inc. All other reagents were of the highest quality commercially available. In silico docking hydroxylating monooxygenase (PHM) was obtained from the Protein Data Bank (http://www.rcsb.org/pdb/, 1SDW) [18] All species deemed superfluous for ligand binding were removed from the pdb file (nickel, water, glycerol, and substrate). Formal charges for enzymebound copper ions were set as 1, bond orders were corrected, and hydrogens were added using Maestro (www.schrodinger.com). Further receptor refinements w ere carried out utilizing ProteinPrep from within Maestro. Investigation of the phenylbutenoate binding modes were performed using Glide [19] and

PAGE 104

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 90 Qsite [20] jointly for quantum polarized ligand docking. This method generates highly accurate binding modes by quantum mechanically calculating the partial atomic charges of the docked ligand using B3LYP/ 631G* within the receptor and subsequently re docking the ligand [21, 22] Radical stabilization energies To investigate the effec ab initio molecular orbital theory and density functional theory calculations were performed using PCGAMESS The geometries of each phenylbutenoate and the corresponding radical along wit h methane and the methane radical were determined at the B3LYP/6 31G(d) level. Single point energy calculations were then carried out on the optimized structures at the MP2 level with 6 311+G(2df,p) basis sets. Radical stabilization energies of PBA radica ls were calculated as energies of the following reaction: (1) Thus, the difference between the C H bond dissociation energy of methane and the C H bond dissociation energy in the phenylbutenoates. (2) A positive radical stabilization energy value would then indicate that the radical is stable relative to the methane radical compared with the corresponding closed shell systems. Inactivation of PAM by 4phenyl 3butenoates at a mbient O2 Inactivation reactions of 50 L containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and various concentrations of phenylbutenoates were i nitiated by the addition of enzyme and incubated at 37 C. Aliquots of 20 L were withdrawn at various intervals and diluted into 2.0 mL reactions containing 100 mM MES/NaOH pH 6.0, 30 mM

PAGE 105

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 91 NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 20 mM N acetylglycine The residual activity was thus measured by the rate of oxygen consumption Because of the high potency of the phenylbutenoates, the inactivation reactions were not performed continuously due to the down time required to prepare the next reaction mixture within the oxygen electrode chamber. Thus, a separate reaction was required for each timepoint at each concentration. Inactivation of PAM by 4phenyl 3butenoates at variabl e O2 concentrations Inactivation stock reactions of 200 L containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and various concentrations of phenyl butenoates were incubated at 37 C for 10 minutes in septum sealed tubes. T he rubber septum was then pierced with a 25 gauge needle as a pressure release and a second needle attached to non permeable rubber tubing. Through this tubing, the desired concen tration of O2 was delivered by means of an A C S c e r t i f i e d O2:N2 mixture. The delivery needle was submerged in the reaction solution to bubble through the gas mixture. Bubbling of the gas mixture through the reaction solution prevented the evaporation of reaction mixt ure that was noted when blowing over the gas mixture while stirring the solution. While the total volume change related to this evaporation was small in absolute terms (20 40 L), it was significant relative to the total reaction volume making the inact ivator concentrations unreliable. The thin gauge needle allowed for sufficient aerat ion of the sample s providing both accurate concentrations of O2. After streaming the gas mixture through the reaction sample for 5 minutes, one empty reaction tube was pr epared with rubber septum for each time point to be taken at that particular inactivator concentration. A typical inactivation st udy by the dilution method is c l a s s i c a l l y a continuous a s s a y H o w e v e r t he time

PAGE 106

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 92 required to prepare the oxy gen electrode in between time points rendered a continuous assay method impossible. T h us e a c h t i m e po i n t r e qu i r e d t h e pr e pa r a t i o n o f a n e w r e a c t i o n m i x t ur e a n d t h e i nc u b a t i o n w a s r e pe a t e d f r o m t i m e = 0. Each empty reaction tube with septum was then purged with the desired O2:N2 reaction mixture and the needles removed. 30 L of the stock reaction was then placed into each purged tube with a 25 gauge needle and syringe. Each reaction was then initiated with enzyme and incubated for the desired length of time before removing a 20 L aliquot and initiating a 2.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 20 mM N acetylglycine and monitoring O2 consumption to examine residual enzyme activity. Reversibility of Inactivation An inactivation reacti on of 100 L containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 100 M 4 phenyl 3butenoic acid was initiated by the addition of enzyme and inc ubated at 37 C for 2 hours. The reaction was then extensively dialyzed and concentrated to ~50 L. The concentrate was then used to initiate a 1.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, and 10 mM hippurate. At various time intervals, a 100 L aliquot was removed, added to a vial containing 20 L of 6% (v/v) TFA to terminate the reaction, and the concentration of glyoxylate formed measured in the acidified samples to test for the recovery of activity.

PAGE 107

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 93 Inactivation of PAL activity by 4 phenyl 3 butenoic acid To determine if 4 phenyl 3butenoic acid was also inactivating PAL activity of PAM, hydroxyhippurate, a known PAL substrate, were monitored. Reactions of 20 g of PAM in a 100 L solution containing 100 mM MES /NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X 100, and 100 M 4 phenyl 3butenoic acid was incubated for 2 hours. A 20 L aliquot was removed and used to initiate a 1.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1.0% (v/v) ethanol, 0.001% (v/v) Triton X hydroxyhippurate. At various time intervals, a 100 L aliquot was removed, added to a vial containing 20 L of 6% (v/v) TFA to terminate the reaction, and the concentration of glyoxylate formed measured in the acidified samples. Determination of partition ratios To determine the partition ratio for product formation versus enzyme inactivation, a 2.0 mL reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 10 g/ml bovine catalase, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 1 mM of the desired 4 phenyl 3butenoic acid derivative. Oxygen consumption was then monitored by oxygen electrode to monitor hydroxylation activity. Determination of glyoxylate conc entration Glyoxylate was determined by the spectrophotometric method of Christman et al [23] as modified by Katopodis and May [24] Standard curves of [glyoxylate] vs. A520 were constructed in the appropriate buffers using a glyoxylate solution that had been calibrated by measuring the glyoxylate dependent oxidation of NADH ( 340 = 6.22 x 103 M1 cm-1) as catalyzed by lactate dehydrogenase.

PAGE 108

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 94 Investigation of inactivation through PAM modification M o di f i c a t i o n of the PAM act ive site by t h e a c t i v a t e d C u/ O s pe c i e s was investigated as a possible means of phenylbutenoate mediated inactivation in addition to covalent labeling by phenylbutenoate. A 100 L reaction containing 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100, 1.0 M Cu(NO3)2, 5.0 mM sodium ascorbate, and 500 M 4 phenyl 3butenoic acid was initiated with 20 g of PAM and incubated at 37 C for 12 hours. The enzyme was then dialyzed against 100 mM MES/NaOH pH 6.0, 30 mM NaCl, 1% (v/v) ethanol, 0.001% (v/v) Triton X 100 for 4 hours and the volume reduced to ~40 L by ultra filtration. The enzyme was then analyzed by MALDI TOF against control for any modification. The reaction was then repeated as previously stated followed by trypsin digestion. The reaction was the n analyzed by both LC/MS and MALDI TOF. Analysis of kinetic data Inactivation kinetics were analyzed by Kaleidagraph fit of 1/kobs vs 1/[ I ] to equation 3. (3) where kobs is the observed rate of inactivation, and kinact is the intrinsic rate of inactivation of enzyme. Results Molecular modeling The ph enylbutenoates of interest were docked into the reduced PHM crystal structure. As expected, the terminal carboxylate formed a salt bridge with R240 ( Figure 4.4 H is

PAGE 109

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 95 in close proximity to the catalytic active site copper, CuM ( Figure 4.5) The phenylbutenoates are not large enough, however, for the phenyl groups to bury themselves into the hydrophobic pocket to restrict movement of the molecule. The small size in additio n to the lack of an amide hydrogen needed for hydrogen bonding to N316 to further limit ligand mobility within the active site allows for a great deal of movement of the phenylbutenoates once within the active site. Molecular dynamics simulations performed with the docked 4 phenyl 3butenoate using NAMD 2.6 demonstrate that there is, indeed, a great deal of mobility for these small compounds without the requisite interactions present when peptide gly substrates are bound to PAM. Partition ratios The parti tion ratios for the ring substituted phenylbutenoates were determined by measuring the total oxygen consumed during inactivation as a stoichiometric amount is O2 is consumed during hydroxylation. Thus, upon complete inactivation the amount of O2 consumed is equivalent to the amount of hydroxylated phenylbutenoate produced. All phenylbutenoates investigated are efficient inactivators with the most efficient giving a partition ratio of 18 turnovers per inactivation (Table 4. 1). Inactivation of PAM by phenylbutenoates Several ring substituted phenylbutenoates were investigated as inactivators of PAM by the dilution method, a common method for determining the kinetic parameters of timedependent inactivators [25] Inactivation experiments performed with these analogs indicated that they are indeed time dependent inactivators of PAM ( Figure 4.6). The inactivation was ps e udo first order and concentration dependent, as well as O2and ascorbatedependent. Substrate was found to protect against inactivation of PAM by 4 phenyl 3butenoic acid ( Figure 4.7). The inactivation was found to be irreversible. Extensive dialysis of a reaction mixture incubated with

PAGE 110

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 96 4phenyl 3butenoic acid yielded dead enzyme incapable of producing glyoxylate from hippuric acid and no oxygen consumption was observed when monitored with the oxygen electrode. The kinetic parameters for inactivation under ambient condit ions are list e d in Table 4. 2. The ring substituted phenylbutenoates were then further examined to investigate the effects of substitution on the partition ratio, Vmax/ kinact, and the kinetics of inactivation. A double reciprocal plot of 1/kobs vs. 1/[PBA ] at fixed O2 concentrations yields a pattern of intersecting lines (3) with a slope of: (4) and the ordinate intercept of: (5) A replot of the intercept vs. 1/[O2] yields a straight line. From the inactivation kinetics at varying concentrations of O2, a linear free energy plot of values can be constructed as a function of electron withdrawing or electron donating ability of the substituent ( Figure 4.8). Inactivation of PAL activity 4P henyl 3butenoic acid was tested as a PAL inactivator as well to investigate whether or not the PAM inactivation occurring was specific to the PHM domai n. Glyoxylate was readily hydroxyhippurate, a known PAL substrate. The experiment indicates that PAL is

PAGE 111

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 97 completely unaffected by 4 phenyl 3butenoic acid and retained 100% activity when compared to the control reaction ( Figure 4.9). Non labeled PAM Multiple methods of mass analysis were performed in an attempt to elucidate any auto oxidation or hydroxylation of the PAM active site as has been hypothesized in earlier studies with 4 phenyl 3butenoic ac id [26] While the hydroxylated product was detected by LC/MS, we were unable to detect any modification of the enzyme by MALDI TOF, or LC/MS and LC/MS/MS on trypsin digested samples ( Figure 4.10). Radical stabilization energies The radical stabilization energies calculated from the isodesmic reaction discussed previously indicate that several of the phen ylbutenoates are capable of forming extremely stable radicals when compared to the methane radical [27] While the radical stabi lities are relative and not absolute, the data indicates that the 4 (4 chlorophenyl) 3butenoic acid and 4 (4 methoxyphenyl) 3butenoic acid derivatives are greater than an order of magnitude more stable than the other phenylbutenoates analogs investigated (Table 4. 3). These two very stable derivatives also had the highest partition ratios. Thus, there exists a direct correlation between radical stability and the ability of PHM to hydroxylate the phenylbutenoates after hydrogen abstraction. Discussion Th e phenylbutenoates investigated meet the criteria for mechanism based inhibitors of PAM. The inactivation is both ascorbateand O2dependent. The phenylbutenoates all act as suicide substrates under turnover conditions inactivating the enzyme. The ina ctivation is protected again st by the presence of substrate

PAGE 112

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 98 The experiments performed were designed to give evidence of a radical intermediate being formed during PAM catalysis. To accomplish this, the kinetics of inactivation and the partition ratios w ere examined using phenylbutenoates with substituents on the phenyl ring of different electron donating and electronwithdrawing effects. The steady state kinetics of PAM with PBA can be described as A minimal mechanism accounting for catalysis and inactivation is described as C u ( I I ) C u ( I I ) C u ( I I ) C u ( I I ) C u ( I ) C u ( I ) C u ( I ) C u ( I )PBAC u ( I ) C u ( I )PBA O2XC u ( I I ) C u ( I I )PC u ( I I ) C u ( I I )Ek1[ As c ]2k2E Ek5[ PBA] k62 As c k3Ek7[ O2] k8E Ek9Ek1 1E ED EA D k1 5k1 3. For this inactivation scheme, the rate constant for the catalytic steps is k9. is then the kinetic parameter that contains the catalytic step and would equal The partition ratio for inactivation for a mechanism based inhibitor is Thus, one can arrive at by simply multiplying the partition ratio by the inactivation kinetic data; this allows the attainment of steady state kinetic data when such data cannot be obtained directly. Therefore, we have multiplied the partition ratios by the res pective values of the phenylbutenoates to calculate the values. A free energy plot was then created with these values and compared to various sigma constants for each substituent ( Figure 4.11). The different sigma constants differ in that they are based on different standard reaction series. When + f 1.03 was obtained.

PAGE 113

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 99 Similar studies have previously been performed on D amino acid oxidase. This fl avoprotein was found to go through a carbanion intermediate [28] Vmax was measured for a series of ring substituted phenylglycines was 5.44 [29] This finding suggests that PAM does not proceed through a carbanion mechanism. That PAM proceeds via a radical mechanism is consistent with the data w hich shows a small absolute magnitude for the +, results characteristic of a radical mechanism [14] Studies on styrene derivatives involving allylic hydrogen abstraction through a radical mechanism have demonstrated similar results, finding 0.82 [30] The partition ratios and relative radical stabilities calculated using ab initio methods also support a radical mechanism. The partition ratios are all within an order of magnitude. The agreement seen between the radical stability of the two most stable phenylbutenoates calculat ed and that they are also the two ligands with the highest partition ratios provides further evidence allows for a greater ease in hydroxylation by the PHM domain in PAM. This leads to more turnovers before inactivation occurs. Conclusion In conclusion, we report here findings that suggest PAM catalysis occurs through a radical mechanism. Other studies have shown that two hydroxylation products are formed during phenylbutenoate catalysis leading to both the alpha and gamma hydroxylated derivatives [26] The most likely mechanism by which this hydroxylation may proceed is through the resonance of a radical generated at the alpha position during hydrogen abstraction ( Figure 4.12). The dynamical freedom allowed the phenylbutenoate from the lack of amide hydro gen to N316 hydrogen bonding, an interaction present in the peptide gly substrates ( Figure 4.13), has been

PAGE 114

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 100 hypothesized to cause the lack of absolute stereospecificity in hydroxylation of the 2 position observed in ear lier studies [26] This mobility coupled with the resonance of the radical may reduce the probability of hydroxylation at the 2position allow ing for release of the PBA radical intermediate leaving an activated oxygen at CuM. This woul d also allow for autoxidation of the active site resulting in inactiv e enzyme. W ater would then be the hydroxylating species for the 4OHPBA species.

PAGE 115

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 101 References 1. Bradbury, A.F., M.D. Finnie, and D.G. Smyth, Mechanism of C terminal amide formation by pituitary enzymes. Nature, 1982. 298(5875): p. 6868. 2. Eipper, B.A., R.E. Mains, and C.C. Glembotski, Identification in pituit ary tissue of a peptide alphaamidation activity that acts on glycine extended peptides and requires molecular oxygen, copper, and ascorbic acid. Proc Natl Acad Sci U S A, 1983. 80(16): p. 51448. 3. Katopodis, A.G., D. Ping, and S.W. May, A novel enzyme f rom bovine neurointermediate pituitary catalyzes dealkylation of alphahydroxyglycine derivatives, thereby functioning sequentially with peptidylglycine alphaamidating monooxygenase in peptide amidation. Biochemistry, 1990. 29(26): p. 611520. 4. Tamburin i, P.P. and S.D. Young, J Am Chem Soc, 1989. 111: p. 1933. 5. Merkler, D.J. and S.D. Young, Recombinant type A rat 75 kDa alphaamidating enzyme catalyzes the conversion of glycine extended peptides to peptide amides via an alphahydroxyglycine intermediat e. Arch Biochem Biophys, 1991. 289(1): p. 1926. 6. Merkler, D.J., et al., 18O isotopic 13C NMR shift as proof that bifunctional peptidylglycine alphaamidating enzyme is a monooxygenase. Biochemistry, 1992. 31(32): p. 72828. 7. Kulathila, R., et al., Bifunctional peptidylglcine alpha amidating enzyme requires two copper atoms for maximum activity. Arch Biochem Biophys, 1994. 311(1): p. 1915. 8. Bell, J., et al., Structural and functional investigations on the role of zinc in bifunctional rat peptidylglycine alpha amidating enzyme. Biochemistry, 1997. 36(51): p. 1623946. 9. De, M., et al., Role for an essential tyrosine in peptide amidation. J Biol Chem, 2006. 281(30): p. 2087382.

PAGE 116

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 102 10. Breakfield, X.O., et al., Localized catecholamine storage associate d with granules in murine neuroblastoma cells. Brain Res, 1975. 92(2): p. 23756. 11. Thomas, S.A., A.M. Matsumoto, and R.D. Palmiter, Nature, 2002. 374: p. 643646. 12. Czyzyk, T.A., et al., Targeted mutagenesis of processing enzymes and regulators: impli cations for development and physiology. J Neurosci Res, 2003. 74(3): p. 44655. 13. Jiang, N., et al., PHM is required for normal developmental transitions and for biosynthesis of secretory peptides in Drosophila. Dev Biol, 2000. 226(1): p. 11836. 14. Fit zpatrick, P.F., D.R. Flory, Jr., and J.J. Villafranca, 3 Phenylpropenes as mechanism based inhibitors of dopamine beta hydroxylase: evidence for a radical mechanism. Biochemistry, 1985. 24(9): p. 210814. 15. Miller, S.M. and J.P. Klinman, Secondary isotop e effects and structure reactivity correlations in the dopamine betamonooxygenase reaction: evidence for a chemical mechanism. Biochemistry, 1985. 24(9): p. 211427. 16. Fitzpatrick, P.F. and J.J. Villafranca, Mechanism based inhibitors of dopamine betah ydroxylase. Arch Biochem Biophys, 1987. 257(2): p. 23150. 17. Wimalasena, K., H.H. Herman, and S.W. May, Effects of dopamine betamonooxygenase substrate analogs on ascorbate levels and norepinephrine synthesis in adrenal chromaffin granule ghosts. J Biol Chem, 1989. 264(1): p. 12430. 18. Prigge, S.T., et al., Dioxygen binds endon to mononuclear copper in a precatalytic enzyme complex. Science, 2004. 304(5672): p. 8647. 19. Schrodinger, GLIDE 2000: Portland, OR. 20. Schrodinger, Qsite 2000: Portland, OR.

PAGE 117

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 103 21. Friesner, R.A., et al., Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J Med Chem, 2004. 47(7): p. 173949. 22. Cho, A.E., et al., Importance of accurate charges in molecular docking: qu antum mechanical/molecular mechanical (QM/MM) approach. J Comput Chem, 2005. 26 (9): p. 91531. 23. Christman, A.A., P.W. Foster, and M.B. Esterer, The allantoin content of blood. J Biol Chem, 1944. 155: p. 161171. 24. Katopodis, A.G. and S.W. May, Novel s ubstrates and inhibitors of peptidylglycine alphaamidating monooxygenase. Biochemistry, 1990. 29(19): p. 45418. 25. Silverman, R.B., Mechanism based enzyme inactivators. Methods Enzymol, 1995. 249: p. 24083. 26. Driscoll, W.J., et al., Peptidylglycine a lpha hydroxylating monooxygenase generates two hydroxylated products from its mechanism based suicide substrate, 4 phenyl 3 butenoic acid. Biochemistry, 2000. 39(27): p. 800716. 27. Barratt, B.J., et al., Inhibition of peptidylglycine alpha amidating mono oxygenase by exploitation of factors affecting the stability and ease of formation of glycyl radicals. J Am Chem Soc, 2004. 126(41): p. 1330611. 28. Ghisla, S., flavins and flavoproteins, 1982: p. 133 142. 29. Neims, A.H., D.C. De Luca, and L. Hellerman, Studies on crystalline D amino acid oxidase. 3. Substrate specificity and sigmarho relationship. Biochemistry, 1966. 5 (1): p. 20312. 30. Sohrabi, H., et al., Nickel(macrocycle) Complexes Immobilized within Montmorillonite and MCM 41 as Catalysts for Epoxidation of Olefins. Journal of inclusion phenomena and macrocyclic chemistry, 2006. 54 : p. 2328.

PAGE 118

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 104 List of Figures Figure 4. 1. The reactions catalyzed by PHM and PAL. Bifunctional PAM is compromised of the two monofun ctional enzymes. R N H O O O H 2 A s c or ba t e + O22 S e m i de hydr oa s c or ba t e + H2O HSHR R N H O O O H O H HR HR O H O R N H2 O O 2 C u ( I I ) F e ( I I I ) / C a2 +/ Z n ( I I )P H M P A L

PAGE 119

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 105 Figure 4. 2 N H3H O H O N H3H O H O O H1 8O2 + 2 A sco rb a t e H2 1 8O + 2 Se mi d e h yd ro a sco rb a t e 2 C u (I I )1 8 R N H C O O O R N H C O O O O H1 8O2 + 2 A sco rb a t e H2 1 8O + 2 Se mi d e h yd ro a sco rb a t e1 82 C u (I I ) PHM DBM

PAGE 120

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 106 Figure 4. 3. The hydrogen abstraction mechanisms for the two most accepted mechanisms of PHM an Klinman while mechanism B has been proposed by Chen et al. EC uH I IC uM I I 1 2 e2 O 2EC uH IC uM I IO O R N H H H O O H ABEC uH IC uM I IO O R N H H H O O H EC uH IC uM I I R N H H H O O H EC uH IC uM I IO O H R N H H O O H O O EC uH IC uM I IR N H H O O H O O H

PAGE 121

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 107 Figure 4. 4. 4 phenyl 3butenoate forming a salt bridge with the guanidino group of R240. The secondary structure is rendered as a transparent ribbon in orange. Note that carbon is teal, nitrogen is blue, oxygen is red, and hydrogen is white.

PAGE 122

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 108 Figure 4. 5. 4 phenyl 3butenoate shown docked within the PHM active sit e showing the H to the CuM. A lack of interaction with N316 should also be noted. Note that carbon is teal, nitrogen is blue, oxygen is red, sulfur is yellow, hydrogen is white, and copper is orange. PBA R240 N316 O O H

PAGE 123

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 109 Figure 4. 6. The time dependent inactivation of PHM activity by 4(3 chlorophenyl) 3butenoic acid. 0 1 2 3 4 5 6 7 8 9 Residual Activity U/mg Time min 100 10 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70

PAGE 124

110 Figure 4.7. Substrate was shown to protect again st inactivation by 4-phenyl-3-butenoic acid. 0102030405060708090100Time, min Control 7.5mM cinnamate 7.5mM cinn + 60uM thiopronin Control 50 M PBA 50 M PBA + 60 M Tiopronin90 75 60 45 30 15Glyoxylate, M

PAGE 125

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 111 Figure 4. 8. Linear free energy plot of kinact/ KO2 values as a function of electron withdrawing or electron donating ability of the substituent. 2 1 0 1 2 3 1 0.5 0 0.5 Log(kinact/KO 2)+

PAGE 126

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 112 Figure 4. 9. The effect of 4 phenyl 3butenoic acid on the PAL activity of PAM. 0 10 20 30 40 50 60 70 Time (min) Control PBA 10 20 30 40 50 60 70

PAGE 127

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 113 Figure 4. 10. The trypsin digested MALDI TOF fingerprint of normal (blue) a nd 4 phenyl 3butenoic acid inactivated (green) PAM.

PAGE 128

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 114 Figure 4. 11. Free energy plot of V max/ KO2 as a function of electron withdrawing and electron donating ability of the substituent. 1 0 1 2 3 4 5 6 1 0.5 0 0.5 Log( Vmax/KO 2)+

PAGE 129

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 115 Figure 4. 12. Radical rearrangement between alpha and gamma positions in 4 phenyl 3butenoic acid leading to two hydroxy lated products.

PAGE 130

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 1 16 Figure 4. 13. Benzylglycine docked to illustrate the proper hydrogen bonding interaction between N316 and the amide hydrogen of the substrate. H N O O H R240 N316 N benzylglycine

PAGE 131

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 117 Tables Table 4.1. Partition ratios for ring substituted 4 phenyl 3butenoic acids Name Partition Ratio 4 phenyl 3 butenoic acid 85 4 (4 chlorophenyl) 3 butenoic acid 94 4 (3 chlorophenyl) 3 butenoic acid 35 4 (4 methoxyphenyl) 3 butenoic acid 143 4 (3 methoxyphenyl) 3 butenoic acid 18

PAGE 132

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 118 Table 4.2. The kinetic parameters for the inactivation of PHM by ring substituted 4 phenyl 3butenoic acids. Name k inact (min 1 ) K I (mM) 4 phenyl 3 butenoic acid 4.18 0.004 4 (4 chlorophenyl) 3 butenoic acid 13.6 0.98 4 (3 chlorophenyl) 3 butenoic acid 1.18 1.45 4 (4 methoxyphenyl) 3 butenoic acid 0.22 0.06 4 (3 methoxyphenyl) 3 butenoic acid 0.10 0.04

PAGE 133

Chapter Four: PAM Inactivation by Phenylbutenoates: E vidence for a radical intermediate 119 Table 4.3. Radical stabilization energies for ring substituted 4 phenyl 3butenoic acids. Name RSE (kcal/mol) 4 phenyl 3 butenoic acid 47 .6 4 (4 chlorophenyl) 3 butenoic acid 1480 4 (3 chlorophenyl) 3 butenoic acid 53.9 4 (4 methoxyphenyl) 3 butenoic acid 1440 4 (3 methoxyphenyl) 3 butenoic acid 49.5

PAGE 134

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 120 Introduction amidating monooxygenase (PAM) is a copper and zinc dependent, bifunctional enzyme that catalyzes the cleavage of glycine extended peptides to the corresponding amides and glyoxylate [1, 2] The sequential action o carbon and then cleaving the carbonhydroxylating monooxygenase (PHM) and peptidylglycine amidoglycolate lyase (PAL) domains, respectively (Fig ure 5. 1) [3 11] PAM is responsible for activating peptide prohormones in vivo Although a considerable body of mechanistic and structure/function data has been generated in an attempt to understand this redox chemistry [1217] there are still crucial unanswered questions regarding electron transfer, dioxygen activation, and radical formation during C H bond cleavage that preclude a comprehensive understanding of glycine hydroxylation. Crystallographers have stated that the reduction of PAM by ascorbate is collisional as they have been unable to identify a binding site for ascorbate. Mimosine, however, has been shown to be a competitive inhibitor ag ainst ascorbate s ugge s t i n g t h a t a binding site must exist [18] This work attempts to elucidate possible binding sites for ascorbate using all atom molecular dynamics simulations. Steered molecular dynamics (SMD) simulations are also employed in an attempt to confirm the computationally discovered binding sites of asc orbate by comparing the experimentally determined KM with the potential of mean force calculated using the Jarzynski relation. In SMD simulations, external forces are applied in a timedependent manner to, for example, a ligand to facilitate unbinding fro m an active site. One can analyze the interactions of the unbinding ligand with the pocket, and the applied forces and ligand position as

PAGE 135

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 121 a function of time yielding important structural information about the binding mechanism and s tructure function relat ionships of the ligand receptor complex [19] By utilizing the Jarzynski relationship, one can relate the average potential of mean force derived from a series of SMD simulations to the average difference in free energy of two equilibrium states. The Jarzynski relatio nship allows for the use of non equilibrium calculations of equilibrium free energies. In 1997, Jarzynski proved that: BT, Wi is the out of equilibrium work done onto the system when going from state A to state B, and the exponential average is done over an equilibrium ensemble only for state A. The only requirements for this equality to work are that the initial ensemble over state A be equilibrated, and that the exponential average be converged. There is, however, no requirement as to how the switch from state A to state B should be done (Fig ure 5. 2). Computationally, this is the power of the Jarzynski relationship as ther e are no requirements for how slowly one must switch from state A to state B. The real advantage of employing the Jarzynski relationship is that one only needs to equilibrate the initial state. This eliminates the necessity for a slow, quasi static transformation from state A to state B. In an era of massively parallel computers available for short spans of time, this type of simulation is ideal where a complete Jarzynski style computation can be submitted in parallel and all data collected in the t ime of one simulated pulling. When small counts of CPUs are available for long periods of time, other methods may be preferred.

PAGE 136

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 122 Methods Equilibrium Molecular Dynamics The PHM crystal structure (1SDW) was first obtained from the Protein Databank, and all catalytically superfluous molecules and ions were removed. The force field parameters for both Cu2+ domains were determined by normal mode analysis in the harmonic approximation. The Hessian was calculated for each methyl cap ped Cucontaining fragment (F igure 5. 3) separately using Density Functional Theory to obtain the bonding and bending force constants. DFT calculations were performed with the B3LYP hybrid exchange correlation functional and the 6 31G* basis set. The equilibrium bond distances were f ound via geometry optimization. The partial charges of the coppers were the formal charges expected to exist prior to reduction by ascorbate. All other residues used potential parameters from the CHARMM27 force field. The PHM system was then solvated i n a 120 3 water box using the TIP3P water model to which Na+ and Clcounter ions were added to a concentration of 100 mM. NAMD was then used to equilibrate the system. All PHM non backbone atoms were fixed and the system was energy minimized for 10 000 steps followed by a minimization of all atoms for 10 000 steps. ps in the NVT ensemble (T=310.15 K), the NP T ensemble (P=1 atm, T=310.15 K) for 2.4 ns (Figure 5. 4). Two separate systems were then created using the equilibrated PHM structure to probe for both ascorbate (ASC) and mimosine (MIM) binding sites. The desired probe (ASC or MIM) was added to the solution surrounding the equilibrated PHM structure to a concentration of ~10 mM. A new set of counter ions was added to a concentration of 100mM.

PAGE 137

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 123 All PHM atoms were fixed and the system was energy minimized for 3 000 s t e ps and dynamics run for 200 ps on the water and counter ions. The ASC or MIM molecules were then released and, with PHM still fixed, the system was minimized for 5 000 steps and dynamics was performed for 500 ps. These steps were performed in the NVT ensemble (T=310.15 K). All atoms were then re leased from constraints, the system minimized for 3 000 steps, and free dynamics in the NPT ensemble (P=1 atm, T=310.15 K) run for 80 ns. All simulations described utilized periodic boundary conditions and the Particle Mesh Ewald (PME) method to calculate electrostatic forces without cutoff [20] Van der Waals interactions were calculated with a cutoff of 12 and a switching distance of 10 Bonded forces, short range non bonded forces (within the cutoff), and long range electrostatics (outside the cutoff) were evaluated every femtosecond dur ing the initial equilibration of the PHM structure. For the binding simulation, a multiple time stepping algorithm [21, 22] was utilized with a 1 fs step for bonded force evaluation, 2 fs for short range nonbond ed forces (within the cutoff), and 4 fs for long range electrostatics (outside the cutoff). Langevin dynamics was used to control the temperature using a damping coefficient of 5 ps1 with hydrogen atoms not coupled to the heat bath. Pressure was regulat ed via the hybrid Nose Hoover [23] Langevin[24] piston method using a piston oscillation period of 100 fs and a damping time scale of 50 fs. Steered Molecular Dynamics

PAGE 138

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 124 Upon binding of ascorbate or mimosine in their respective simulations, ten starting structures from each simulation were co llected for multiple steering molecular dynamics simulations. External steering forces were then applied to pull the ligand, ASC or MIM, out of its binding site. The ligands were pulled under constant velocity in directions predetermined through visuali zation, one pull per each of the 20 collected starting structures (Fig ure 5. 5). Bonded forces, short range non bonded forces (within the cutoff), and long range electrostatics (outside the cutoff) were evaluated every femtosecond for all steered molecular dynamics simulations. The pulling velocity was set at 0.00005 per time step (1 fs). A spring constant of 7 kcal/mol/2 was used to constrain the ligand. This constraint is large enough to allow for the use of the stiff spring approximation [25, 26] For all pulling simulations, the pulling force was applied to the atoms indicated in Fig ure 5. 6. The force was only applied along the pulling direction during the simulation. The ligands we re free from constraint in the plane perpendicular to the pulling vector. Trajectories were saved every 0.1 ps, and steering forces recorded every 10 fs. Analysis of Trajectories The methods used to construct the potential of mean force (PMF) from the steered molecular dynamics simulations were all based on Jarzynskis equality and were independent of the friction coefficient [27] The reaction coordinate was the increasing distance between the ligand and the binding site during the pulling simulation. Because drifting of the protein would change the distance between the ligand and the binding site throughout the simulation, several atoms on the side opposite of the binding site were fixed on th e protein (Fig ure 5. 7).

PAGE 139

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 125 ASC and MIM are sufficient as fewer slower pulling trajectories have been demonstrated to yield more accurate PMF values [25, 26] The 50 /ns pulling velocity is sufficiently slow to produce accurate PMF values from the number of trajectories used here. The pulling force was calculated using the following equation: (1) where F is the pulling force, t is time, k is the spring constant of the constraint, v is the pulling velocity, n is the pulling direction normal, and r ( t ) and r0 are the positions of the ligand at time t and t0. Work was calculated by integrating the force over the distance pulled from the pulling trajectories: (2) The PMF, or freeenergy difference, was derived from work W util izing Jarzynskis equality as follows [26, 28] : (3) where R is the universal gas constant and T is the absolute temperature. Previous work by the Schulten group has demonstrated that work distribution from Langevin dynamics satisfies a Gaussian distribution and can be expressed as a second order cumulant expansion [26] : (4)

PAGE 140

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 126 where is the average work from all trajectories aW is the standard deviation of the work distribution. Results Ascorbate binding simulations The simulation for the ascorbate binding experiment reveals the probe diffusing though the aqueous solution around PHM. Some of the ligands search the surface of PHM and again diffuse. Some ascorbate molecules in the simulation search the surface and bind to it temporarily. One ascorbate molecule in the simulation, however, binds and remains bound for 15 ns. The criteria for determining the legitimacy of a binding site included two important guidelines. The binding site must be a depressed, hydrophobic pocket and not simply hydrogen bonding interactions on the surface of PHM. The binding should last on the order of >10 ns to ensure sufficient interaction strength. The ascorbate simulation was carried out for 79 ns and revealed only one potential binding site for ascorbate (Fig ure 5. 8). The binding site for ascorbate is a depression on the PHM surface that includes residues important for strong hydro gen bonding. These residues include ASP61, GLN198, and LYS152. A frame from the trajectory during the ascorbate on portion of the simulation is rendered below (Fig ure 5. 9). As a means of measuring the amount of time the particular binding ascorbate li gand spent bound to the enzyme, hydrogen bonding was monitored between the ascorbate ligand and the surface of PHM. The data indicates that ascorbate bound to PHM for ~15 ns before diffusing. This meets our criteria for legitimate binding and is depicted in Figure 5. 10. Mimosine binding simulations

PAGE 141

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 127 The simulation for the mimosine binding experiment was performed in the same manner as that of the ascorbate binding experiment. Mimosine diffused freely throughout the solvent during the simulation. The mim osine simulation was carried out for ~82 ns, and only one potential binding site for mimosine was identified (Figure 5. 11). The binding site for mimosine is a depression on the PHM surface much like that of ascorbat and includes residues important for str ong hydrogen bonding. The binding site residues include ASN45, ARG67, ARG100, and LYS98. A frame from the trajectory during the mimosine bound portion of t he simulation is rendered in Fig ure 5. 12. In order to determine the amount of time mimosine was bo und to PHM during the simulations, the hydrogen bonding between the bound ligand and the binding site was investigated. Figure 5. 13 illustrates that mimosine seemed to bind longer than ascorbate perhaps suggesting a stronger interaction which would be con sistent with what has been observed experimentally. Because ascorbate and mimosine were shown to bind in close proximity but different binding pockets, the particle mesh Ewald electrostatic potential of PHM was calculated for the on and off states of both ligands. This illustrates that the electrostatic effects of mimosine binding would indeed affect the binding pocket of ascorbate. Thus, the seemingly competitive experimental data may not rule out slightly differing binding sites within a few ang stroms of one another (Fig ures 5. 14 & 5. 15). Steered Molecular Dynamics The SMD simulations for both ascorbate and mimosine yielded force data that fluctuate between positive and negative values (Fig ure 5. 16). This indicates that the thermal fluctuation of the ligand (ASC or MIM) is larger than the perturbation from the pulling. This shows that the

PAGE 142

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 128 process by which the ligand unbinds is near equilibrium. No conformational changes were seen during the unbinding process with mild peaks in the force graph relating only to the breaking of hydrogen bonds. The unbinding of the two ligands, ascorbate and mimosine, are nonequilibrium processes. The external work was sampled from repeated trajectories in order to calculate the potential of mean force using the Jarzynski equality (Figure 5.17) Jarzynskis equality cannot be used directly unless the deviation of work distribution biological systems of the size simulated here [19] Because standard deviation will grow as the ligand is pulled further from the binding site, the free energy estimated from the Jarzynski equality is dominated by smaller values of work. The cumulant expansion method was developed to solve this problem [25, 26] (Figure 5.18) Utilizing the cumulant expansion method improves the statis tics of work distribution caused by having such limited sampling when calculating the PMF. Discussion The simulations performed in an attempt to elucidate the binding site for ascorbate, the reductant of PAM, have led to the discovery of a possible bindi ng site. The successful elucidation of a binding site will help eliminate some of the proposed mechanisms and help determine where electron transfer occurs prior to the reduction of CuM. The mimosine simulation was used as a control in that an ascorbate binding site would either accommodate mimosine or a mimosine binding site would be very nearby. This proved to be the case in the proposed ascorbate binding site. Mimosine binds within a few angstroms of ascorbate. Another method used as a way of veri fying the proposed binding site was the calculation of the free energy of binding. While the calculated free energy for mimosine can be directly

PAGE 143

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 129 compared with the experimental KI value (4 M) as no chemistry is occurring, that of ascorbate is a KM (223 M ) and cannot be compared directly but is a good estimate. Using the equations for Gibbs free energy: (5) where K is the Kd or KM for mimosine and ascorbate, respectively, calculated from the PMF values for free energy simulated. The calculated PMF values for ascorbate yields a theoretical KM of 200 M. This is sufficiently close to justify consideration of the proposed binding si te as valid. The theoretical KI of mimosine from the simulation results in a value of 1 M which is also consistent with the experimental data within the accepted error for this type of free energy calculation. While initial results are convincing, mor e work is needed to experimentally verify these initial findings. Mutants targeting the proposed binding site should be made and the effects of these mutations tested on reduction of PAM by ascorbate. A quicker method may be to perform in silico docking of a library of commercially available compounds. Hits for binding in the proposed binding site should be tested as competitive inhibitors against the reduction of PAM by ascorbate.

PAGE 144

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 130 References 1. Bradbury, A.F., M.D. Finnie, and D.G. Smyth, Mechanism of C terminal amide formation by pituitary enzymes. Nature, 1982. 298(5875): p. 6868. 2. Eipper, B.A., R.E. Mains, and C.C. Glembotski, Identification in pituitary tissue of a peptide alphaamidation activity that acts on glycine extended peptides and requires molecular oxygen, copper, and ascorbic acid. Proc Natl Acad Sci U S A, 1983. 80(16): p. 51448. 3. Merkler, D.J. and S.D. Young, Recombinant type A rat 75 kDa alphaamidating enzyme catalyzes the conversion of glycine extended peptides to peptide amides via an alphahydroxyglycine intermediate. Arch Biochem Biophys, 1991. 289(1): p. 1926. 4. Ramer, S.E., et al., J Am Chem Soc, 1988. 110 : p. 8526. 5. Tamburini, P.P. and S.D. Young, J Am Chem Soc, 1989. 111: p. 1933. 6. Zabriskie, T.M., H. Cheng, and J.C. Vederas, J. Chem. Soc., Chem. Comm., 1991. 571. 7. Merkler, D.J., et al., 18O isotopic 13C NMR shift as proof that bifunctional peptidylglycine alphaamidating enzyme is a monooxygenase. Biochemistry, 1992. 31(32): p. 72828. 8. Noguchi, M., et al., The source of the oxygen atom in the alphahydroxyglycine intermediate of the peptidylglycine alphaamidating reaction. Biochem J, 1992. 283 ( Pt 3) : p. 8838. 9. Kulathila, R., et al., Bifunc tional peptidylglcine alpha amidating enzyme requires two copper atoms for maximum activity. Arch Biochem Biophys, 1994. 311(1): p. 1915. 10. Bell, J., et al., Structural and functional investigations on the role of zinc in bifunctional rat peptidylglycin e alpha amidating enzyme. Biochemistry, 1997. 36(51): p. 1623946. 11. De, M., et al., Role for an essential tyrosine in peptide amidation. J Biol Chem, 2006. 281(30): p. 2087382.

PAGE 145

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 131 12. Francisco, W.A., et al., Investigation of the pathway for inter copper electron transfer in peptidylglycine alphaamidating monooxygenase. J Am Chem Soc, 2004. 126(41): p. 131689. 13. Francisco, W.A., et al., Kinetic mechanism and intrinsic isotope effects for the peptidylglycine alphaamidating enzyme reaction. Biochemistry 1998. 37(22): p. 824452. 14. Klinman, J.P., The copper enzyme family of dopamine betamonooxygenase and peptidylglycine alphahydroxylating monooxygenase: resolving the chemical pathway for substrate hydroxylation. J Biol Chem, 2006. 281(6): p. 30136. 15. Landymore Lim, A.E., A.F. Bradbury, and D.G. Smyth, The amidating enzyme in pituitary will accept a peptide with C terminal D alanine as substrate. Biochem Biophys Res Commun, 1983. 117(1): p. 28993. 16. Tamburini, P.P., et al., Structure activity rel ationships for glycine extended peptides and the alphaamidating enzyme derived from medullary thyroid CA77 cells. Arch Biochem Biophys, 1988. 267(2): p. 62331. 17. Tamburini, P.P., et al., Peptide substrate specificity of the alphaamidating enzyme isol ated from rat medullary thyroid CA77 cells. Int J Pept Protein Res, 1990. 35(2): p. 1536. 18. Miller, D.A., et al., Characterization of a bifunctional peptidylglycine alphaamidating enzyme expressed in Chinese hamster ovary cells. Arch Biochem Biophys, 1992. 298(2): p. 3808. 19. Xiong, H., et al., Theor Chem Acc, 2006. 116: p. 338346. 20. Darden, T., D.M. York, and L. Pedersen, Particle mesh Ewald. An N*log(N) method for Ewald sums in large systems. J. Chem. Phys., 1993. 98: p. 1008910092. 21. Grubmul ler, H., et al., Generalized Verlet algorithm for efficient molecular dynamics simulations with longrange interactions. Mol Sim, 1991. 6 : p. 121142.

PAGE 146

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 132 22. Schlick, T., et al., Algorithmic challenges in computational molecular biophysics. J Comp Phys, 1999. 151: p. 948. 23. Martyna, G., D.J. Tobias, and M.L. Klein, Constant pressure molecular dynamics algorithms. J. Chem. Phys., 1994. 101: p. 41774189. 24. Feller, S.E., et al., Constant pressure molecular dynamics simulation the Langev in piston method. J. Chem. Phys., 1995. 103: p. 46134621. 25. Park, S., et al., J chem phys, 2003. 119 : p. 35593566. 26. Park, S. and K. Schulten, J Chem Phys, 2004. 120: p. 59465961. 27. Zhang, D., J. Gullingsrud, and J.A. McCammon, Potentials of mean force for acetylcholine unbinding from the alpha7 nicotinic acetylcholine receptor ligandbinding domain. J Am Chem Soc, 2006. 128(9): p. 301926. 28. Jarzynski, C., Phys. Rev. Lett., 1997. 78 : p. 26902693.

PAGE 147

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 133 Figures and Tables Figures Figure 5.1. amidating monooxygenase reaction. Note, Asc represents ascorbate, deAsc represents semidehydroascorbate, PAM is peptidylglycine aamidating hydroxylating monooxygenase, and PAL is peptidylgyci ne amidoglycolate lyase. H N O H O 2A s c 2 C u ( I I ) O2H2O O H O O H P A M P H M 2 C u ( I I ) 2A s c P A LZ n ( I I ) / F e ( I I I ) / C a R O H N O H O R O O H N H2 R O 2S D A 2S D A

PAGE 148

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 134 Figure 5.2. States A and B in which state A is the starting structure and state B is the final state post SMD simulation. Water and ions omitted for clarity.

PAGE 149

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 135 Figure 5.3. Methyl capped CuH and CuM domains in the PHM crysta l structure prior to geometry optimization.

PAGE 150

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 136 Figure 5.4. PHM crystal structure soaked with ascorbate. Note, water and ions not shown for clarity.

PAGE 151

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 137 Figure 5.5. Illustration of the PHM secondary structure with bound ascorbate and the spring as the force vector applied during the steering molecular dynamics simulations.

PAGE 152

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 138 Figure 5.6. Ascorbate and mimosine used to probe for a reductant binding site on PHM and the carbon atoms in which the force was applied during the steered molecular dynamics simulation. k k

PAGE 153

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 139 Figure 5.7. PHM secondary structure with bound ascorbate and the fixed L336 and N337.

PAGE 154

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 140 Figure 5.8. Ascorbate bound to the surface of PHM.

PAGE 155

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 141 Figure 5.9. Ascorbate in the proposed binding pocket with hydrogen bonding illustrated as blue or red dashed lines.

PAGE 156

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 142 Figure 5.10. Hydrogen bonding between ascorbate and the proposed PHM binding site.

PAGE 157

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 143 Figure 5.11. Mimosine bound to the surface of PHM.

PAGE 158

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 144 Figure 5.12. Mimosine in the proposed binding pocket with hydrogen bonding illustrated as blue dashed lines.

PAGE 159

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 145 Figure 5.13. Hydrogen bonding between mimosine and the proposed PHM binding site.

PAGE 160

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 146 Figure 5.14. PME electrostatic potential of PHM with off and on ascorbate.

PAGE 161

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 147 Figure 5.15. PME electrostatic potential of PHM with off and on mimosine.

PAGE 162

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 148 Figure 5.16. Force vs extension for both ascorbate and mimosine over a 5 reaction coordinate.

PAGE 163

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 149 Figure 5.17. Work vs extension for ascorbate and mimosine unbinding.

PAGE 164

Chapter Five: Computational Elucidation of Reductant Binding Sites in PAM 150 Figure 5.18. Potential of mean force vs extension for ascorbate and mimosine unbinding.

PAGE 165

Appendices 151 Appending A PCGAMESS Geometry Optimization Settings File 1. L3Cu(I) thiophenolate geometry optimization (truncated) Geometry optimize the CuM cluster with thiophenolate anion coordinated (froze the carbons of the methyl caps to simulated the protein backbone) lanl2dz basis set ! Jon Belof/Edward W. Lowe Department of Chemistry University of South Florida $CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE MAXIT=200 ECP=READ $END $SYSTEM TIMLIM=9999999999 MWORDS=70 $END $STATPT NSTEP=1000 IFREEZ(1)=10,11,12,13,14,15,16,17,18 $END $DFT DFTTYP=B3LYP $END $GUESS GUESS=HUCKEL $END $ECP CU ECP GEN 10 2 3 ----d potential ----10.0000000 1 511.9951763 72.5548282 2 93.2801074 12.7450231 2 23.2206669 4 ----s d potential ----3.0000000 0 173.1180854 23.8351825 1 185.2419886 473.8930488 2 73.1517847 157.6345823 2 14.6884157 4 ----p d potential ----5.0000000 0 100.7191369 6.4990936 1 130.8345665 351.4605395 2 53.8683720 85.5016036 2 14.0989469 S ECP GEN 10 2 5 ----d potential ----10.0000000 1 532.6685222 85.3593846 2 108.1342248 30.4513290 2 24.5697664

PAGE 166

Appendices 152 10.3745886 2 7.3702438 0.9899295 2 2.3712569 5 ----s d potential ----3.0000000 0 106.3176781 10.6284036 1 100.8245833 223.6360469 2 53.5858472 93.6460845 2 15.3706332 28.7609065 2 3.1778402 6 ----p d potential ----5.0000000 0 101.9709185 6.0969842 1 93.2808973 285.4425500 2 65.1431772 147.1448413 2 24.6347440 53.6569778 2 7.8120535 8.9249559 2 2.3112730 $END $DATA File 2. L3Cu(I) H2O geometry optimization (truncated) Geopt of CuM cluster with frozen methyl caps and water coordinated lanl2dz Edward W. Lowe Jr Merkler Research Group Department of Chemistry University of South Florida $CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE ICHARG=+1 MULT=1 UNITS=ANGS $END $CONTRL INTTYP=HONDO ICUT=10 MAXIT=1000 ECP=READ $END $SYSTEM TIMLIM=9999999999 MWORDS=70 $END $SMP CSMTX=.T. $END $P2P P2P=.T. DLB=.T. $END $SCF DIRSCF=.TRUE. FDIFF=.FALSE. SOSCF=.FALSE. $END $SCF DIIS=.TRUE. DAMP=.T. ETHRSH=2.0 $END $STATPT NSTEP=1000 OPTTOL=0.00001 HESS=CALC $END $STATPT IFREEZ(1)=1 ,2,3,19,20,21,37,38,39 $END $DFT DFTTYP=B3LYP $END $GUESS GUESS=HUCKEL $END $ECP CU ECP GEN 10 2

PAGE 167

Appendices 153 3 ----d potential ----10.0000000 1 511.9951763 72.5548282 2 93.2801074 12.7450231 2 23.2206669 4 ----s d potential ----3.0000000 0 173.1180854 23.8351825 1 185.2419886 473.8930488 2 73.1517847 157.6345823 2 14.6884157 4 ----p d potential ----5.0000000 0 100.7191369 6.4990936 1 130.8345665 351.4605395 2 53.8683720 85.5016036 2 14.0989469 S ECP GEN 10 2 5 ----d potential ----10.0000000 1 532.6685222 85.3593846 2 108.1342248 30.4513290 2 24.5697664 10.3745886 2 7.3702438 0.9899295 2 2.3712569 5 ----s d potential ----3.0000000 0 106.3176781 10.6284036 1 100.8245833 223.6360469 2 53.5858472 93.6460845 2 15.3706332 28.7609065 2 3.1778402 6 ----p d potential ----5.0000000 0 101.9709185 6.0969842 1 93.2808973 285.4425500 2 65.1431772 147.1448413 2 24.6347440 53.6569778 2 7.8120535 8.9249559 2 2.3112730 $END $DATA File 3. L3Cu(I) geometry optimization (truncated) Geometry optimize the CuM cluster with no water coordinated

PAGE 168

Appendices 154 (froze the carbons of the methyl caps to simulated the protein backbone) lanl2dz Jon Belof/Edward W. Lowe Department of Chemistry University of South Florida $CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE ICHARG=1 MAXIT=2 00 ECP=READ $END $SYSTEM TIMLIM=9999999999 MWORDS=235 $END $STATPT NSTEP=200 IFREEZ(1)=10,11,12,13,14,15,16,17,18 $END $DFT DFTTYP=B3LYP $END $GUESS GUESS=HUCKEL $END $ECP CU ECP GEN 10 2 3 ----d potential ----10.0000000 1 511.9951763 72.5548282 2 93.2801074 12.7450231 2 23.2206669 4 ----s d potential ----3.0000000 0 173.1180854 23.8351825 1 185.2419886 473.8930488 2 73.1517847 157.6345823 2 14.6884157 4 ----p d potential ----5.0000000 0 100.7191369 6.4990936 1 130.8345665 351.4605395 2 53.8683720 85.5016036 2 14.0989469 S ECP GEN 10 2 5 ----d potential ----10.0000000 1 532.6685222 85.3593846 2 108.1342248 30.4513290 2 24.5697664 10.3745886 2 7.3702438 0.9899295 2 2.3712569 5 ----s d potential ----3.0000000 0 106.3176781 10.6284036 1 100.8245833 223.6360469 2 53.5858472

PAGE 169

Appendices 155 93.6460845 2 15.3706332 28.7609065 2 3.1778402 6 ----p d potential ----5.0000000 0 101.9709185 6.0969842 1 93.2808973 285.4425500 2 65.1431772 147.1448413 2 24.6347440 53.6569778 2 7.8120535 8.9249559 2 2.3112730 $END $DATA File 4. L3Cu(II) geometry optimization of CuM for NAMD parameterization !Optimization of CuM for the ascorbate binding site probe using NAMD !Edward W. Lowe !Merkler Research Group !University of South Florida !05/04/07 $CONTRL SCFTYP=UHF RUNTYP=OPTIMIZE ICHARG=+2 MULT=2 $END $CONTRL INTTYP=HONDO ICUT=10 MAXIT=75 ECP=SBK $END $SYSTEM KDIAG=0 TIMLIM=999999999 MWORDS=23 $END $SMP CSMTX=.T. $END $P2P P2P=.T. DLB=.T. $END $TRANS AOINTS=DIST $END $SCF DIRSCF=.T. FDIFF=. F. SOSCF=.F. DIIS=.T. ETHRSH=2.0 $END $STATPT NSTEP=1000 HESS=CALC $END $STATPT IFREEZ(1)=13,14,15 $END $STATPT IFREEZ(2)=49,50,51 $END $STATPT IFREEZ(3)=76,77,78 $END $DFT DFTTYP=B3LYP $END $GUESS GUESS=HUCKEL $END $BASIS GBASIS=SBK $END $DATA CuM with water for thiol paper geopt C1 H 1.0 1.40800 6.66500 0.02200 N 7.0 0.45700 3.88900 0.20500 H 1.0 1.36400 3.51400 0.01000

PAGE 170

Appendices 156 C 6.0 0.18600 5.13500 0.72900 C 6.0 1.21300 6.20000 0.90400 H 1.0 2.05800 5.78700 1.49600 H 1.0 0.78100 7.03400 1.50000 N 7.0 1.71100 3.96100 0.53200 C 6.0 1.16200 5.16900 0.92700 H 1.0 1.78700 5.97500 1.29300 C 6.0 0.71100 3.21300 0.11000 H 1.0 0.79400 2.18800 0.26100 H 1.0 2.80000 3.66700 5.08200 N 7.0 5.63800 3.10800 3.17700 H 1.0 6.42700 3.01200 3.78500 C 6.0 4.32400 3.25600 3.56400 C 6.0 3.86900 3.38000 4.98500 H 1.0 4.47300 4.16500 5.49000 H 1.0 4.06600 2.42700 5.52400 N 7.0 4.45300 3.06700 1.32800 C 6.0 3.59600 3.23700 2.41500 H 1.0 2.53000 3.36200 2.27000 C 6.0 5.66700 2.99600 1.83500 H 1.0 6.58300 2.85800 1.25400 H 1.0 4.93600 7.63500 1.24100 C 6.0 4.39500 7.14200 0.41800 H 1.0 3.73600 6.41300 0.94600 H 1.0 3.74900 7.66600 0.32100 C 6.0 5.53200 6.39500 0.26700 H 1.0 6.24300 7.11300 0.72800 H 1.0 6.09500 5.80000 0.48500 S 16.0 4.86300 5.30900 1.55000 C 6.0 6.42100 4.66400 2.29100 H 1.0 6.20300 3.96700 3.12800 H 1.0 7.04600 5.49200 2.69000 H 1.0 7.02400 4.11300 1.53700 Cu 29.0 3.77600 3.54100 0.63000 $END File 5. L3Cu(II) geometry optimization of CuH for NAMD parameterization !Optimization of CuH for the ascorbate binding site probe using NAMD

PAGE 171

Appendices 157 !Edward W. Lowe !Mer kler Research Group !University of South Florida !05/04/07 $CONTRL SCFTYP=UHF RUNTYP=OPTIMIZE ICHARG=+2 MULT=2 $END $CONTRL INTTYP=HONDO ICUT=10 MAXIT=75 ECP=SBK $END $SYSTEM KDIAG=0 TIMLIM=999999999 MWORDS=235 AOINTS=DIST $END $SMP CSMTX=.T. $END $P2P P2P=.T. DLB=.T. $END $SCF DIRSCF=.T. FDIFF=.F. SOSCF=.F. DIIS=.T. ETHRSH=2.0 $END $STATPT NSTEP=1000 HESS=CALC $END $STATPT IFREEZ(1)=13,14,15 $END $STATPT IFREEZ(2)=49,50,51 $END $STATPT IFREEZ(3)=76,77,78 $END $DFT DFTTYP=B3LYP $END $BASIS GBA SIS=SBK $END $GUESS GUESS=HUCKEL $END $DATA CuH rxn center geopt C1 H 1.0 6.119 6.698 0.238 N 7.0 5.674 4.283 3.250 H 1.0 5.965 3.920 4.135 C 6.0 5.797 5.589 2.834 H 1.0 6.243 6.369 3.435 N 7.0 4.798 4.407 1.231 C 6.0 5.252 5.651 1.592 C 6.0 5.067 3.616 2.254 H 1.0 4.824 2.551 2.295 C 6.0 5.148 6.819 0.667 H 1.0 4.962 7.729 1.277 H 1.0 4.279 6.702 0.014 H 1.0 5.795 4.086 2.649 N 7.0 2.530 2.247 3.704 H 1.0 1.878 1.910 4.382 C 6.0 3.893 2.373 3.885 H 1.0 4.401 2. 121 4.805 N 7.0 3.372 2.999 1.816 C 6.0 4.395 2.839 2.714

PAGE 172

Appendices 158 C 6.0 2.272 2.635 2.446 H 1.0 1.273 2.641 2.003 C 6.0 5.816 3.154 2.352 H 1.0 6.449 2.318 2. 719 H 1.0 5.927 3.185 1.248 H 1.0 2.117 7.001 2.768 N 7.0 0.484 6.682 0.781 H 1.0 0.160 7.116 1.410 C 6.0 0.652 7.001 0.549 H 1.0 0.097 7.779 1.053 N 7.0 2.017 5.319 0.023 C 6.0 1.598 6.155 1.030 C 6.0 1.325 5.669 1.046 H 1.0 1.416 5.195 2.026 C 6.0 2.160 6.049 2.414 H 1.0 1.639 5.214 2.931 H 1.0 3.237 5.784 2.373 Cu 29.0 3.637 3.998 0.110 $END File 6. Geometry optimization of ascorbate for NAMD parameterization $CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE ICHARG=0 MULT=1 COORD=ZMT NZVAR=54 $END $CONTRL MAXIT=75 $END $SYSTEM T IMLIM=999999999 MWORDS=235 $END $SCF DIRSCF=.TRUE. $END $STATPT NSTEP=1000 $END $BASIS GBASIS=N31 NGAUSS=6 $END $ELPOT IEPOT=1 WHERE=PDC $END $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $ZMAT IZMAT(1) =1,2,1,1,3,2,1,4,3,1,5,4,1,6,2,1,7,4,1,8,7,1,9,5, 1,10,1,1,11,7,1,12,11,1,13,4,1,14,7,1,15,11,1,16,11, 1,17,12,1,18,8,1,19,10,1,20,9, 2,3,2,1,2,4,3,2,2,5,4,3,2,6,2,3,2,7,4,3,2,8,7,4, 2,9,5,4,2 ,10,1,2,2,11,7,4,2,12,11,7,2,13,4,3,2,14,7,4, 2,15,11,7,2,16,11,7,2,17,12,11,2,18,8,7,2,19,10,1,2,20,7,5, 3,4,3,2,1,3,5,4,3,2,3,6,2,3,4,3,7,4,3,2,3,8,7,4,3,

PAGE 173

Appendices 159 3,9,5,4,3,3,10,1,2,3,3,11,7,4,3,3,12,11,7,4,3,13,4,3,2, 3,14,7,4,3,3,15,11,7,4,3,16,11,7,4,3,17,12,11,7,3,18,8,7,4, 3,19,10,1,2,3,20,9,5,4 $END $DATA ascorbate toppar C1 C O 1 1.3993695 C 2 1.4911748 1 109.3353707 C 3 1.5083526 2 102.8105937 1 0.4558029 0 C 4 1.3447000 3 110.2718397 2 0.5401568 0 C 3 1.5288325 2 108.6056231 1 123.6554732 0 C 6 1.5314014 3 113.7297864 2 172.2854080 0 O 7 1.4620625 6 103.6177438 3 171.2245510 0 O 6 1.4454707 3 105.7709296 2 68.4700166 0 O 1 1.2312353 2 124.4287108 3 179.7704008 0 O 5 1.3768209 4 126.8178161 3 179.5237568 0 O 4 1.3704297 3 119.9144041 2 178.6049028 0 H 3 1.0945449 2 106.7457350 1 119.0405695 0 H 6 1.0991831 3 108.6614051 2 50.6657410 0 H 7 1.0937666 6 111.7872871 3 68.8319724 0 H 7 1.0980365 6 110.1003640 3 52.5684184 0 H 8 0.9758375 7 111.3198901 6 171.5451546 0 H 9 0.9819455 6 107.0046583 3 170.4584470 0 H 11 0.9808199 5 109.9007971 4 178.8505208 0 H 12 0.9799619 4 111.6479499 3 177 .6398852 0 $END File 7. Geometry optimization of mimosine for NAMD parameterization $CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE ICHARG= 1 MULT=1 COORD=ZMT NZVAR=63 $END $CONTRL MAXIT=75 $END $SYSTEM TIMLIM=999999999 MWORDS=235 $END $SCF DIRSCF=.TRUE. $END $STATPT NSTEP=1000 $END $BASIS GBASIS=N31 NGAUSS=6 $END $ELPOT IEPOT=1 WHERE=PDC $END

PAGE 174

Appendices 160 $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $DATA mimosine toppar C1 c c 1 cc2 c 2 cc3 1 ccc3 n 3 nc4 2 ncc4 1 dih4 c 4 cn5 3 cnc5 2 dih5 c 5 cc6 4 ccn6 3 dih6 o 2 oc7 3 occ7 4 dih7 c 4 cn8 3 cnc8 2 dih8 c 8 cc9 4 ccn9 3 d ih9 n 9 nc10 8 ncc10 4 dih10 o 1 oc11 2 occ11 3 dih11 c 9 cc12 8 ccc12 4 dih12 o 12 oc13 9 occ13 8 dih13 o 12 oc14 9 occ14 8 dih14 h 3 hc15 2 hcc15 1 dih15 h 5 hc16 4 hcn16 3 dih16 h 6 hc17 5 hcc17 4 dih17 h 7 ho18 2 hoc18 3 dih18 h 8 hc19 4 hcn19 3 dih19 h 8 hc20 4 hcn20 3 dih20 h 9 hc21 8 hcc21 4 dih21 h 10 hn22 9 hnc22 8 dih22 h 10 hn23 9 hnc23 8 dih23 cc2 1.409701 cc3 1.399166 ccc3 118.654 nc4 1.368604 ncc4 121.037 dih4 0 .201 cn5 1.367139 cnc5 120.411 dih5 0.620 cc6 1.395360

PAGE 175

Appendices 161 ccn6 120.605 dih6 0.665 oc7 1.380548 occ7 118.403 dih7 179.856 cn8 1.472220 cnc8 119.482 dih8 179.991 cc9 1.541563 ccn9 111.386 dih9 71.975 nc10 1.463633 ncc10 109.893 dih10 177.868 oc11 1.350598 occ11 121.154 dih11 179.946 cc12 1.507657 ccc12 110.483 dih12 60.722 oc13 1.225477 occ13 120.220 dih13 73.235 oc14 1.352379 occ14 119.610 dih14 96.438 hc15 1.083728 hcc15 118.813 dih15 179.605 hc16 0.983454 hcn16 120.565 dih16 179.918 hc17 1.083065 hcc17 119.985 dih17 179.661 ho18 0.990257 hoc18 109.377 dih18 179.940 hc19 1.114040

PAGE 176

Appendices 162 hcn19 110.876 dih19 166.298 hc20 1.112321 hcn20 109.145 dih20 49.392 hc21 1.113838 hcc21 109.120 dih21 58.779 hn22 1.046531 hnc22 108.316 dih22 177.274 hn23 1.046012 hnc23 108.372 dih23 62.789 $END File 8. General input for all PBA ge ometry optimizations $CONTRL ICHARG= 1 MULT=1 RUNTYP=OPTIMIZE SCFTYP=RHF COORD=CART $END $CONTROL UNITS=ANGS MAXIT=200 $END $DFT DFTTYP=B3LYP METHOD=GRID $END $BASIS GBASIS=N31 NGAUSS=6 $END $STATPT METHOD=QA NSTEP=1000 OPTTOL=0.0001 HESS=CALC $END $GUESS GUESS=HUCKEL $END $SYSTEM MWORDS=200 $END $P2P P2P=.T. DLB=.T. $END $DATA pba geopt C1 C 6.0 1.21290 1.62180 0.02210 C 6.0 1.98910 2.79560 0.04280 C 6.0 3.38860 2.68230 0.12420 C 6.0 4.00030 1.42460 0.11890 C 6.0 3.22200 0.26740 0.04100 C 6.0 1.82980 0.36600 0.02630 H 1.0 0.13300 1.66350 0.04760 C 6.0 1.37430 4.14610 0.05840 H 1.0 3.95040 3.48720 0.17960

PAGE 177

Appendices 163 H 1.0 5.07890 1.34760 0.17620 H 1.0 3.65420 0.61520 0.03880 H 1.0 1.28160 0.44850 0.07100 C 6.0 0.01850 4.41530 0.37530 C 6.0 0.57530 5.65830 0.28230 C 6.0 1.76220 5.62260 0.63100 O 8.0 1.58680 5.73260 1.96700 O 8.0 2.97600 5.27760 0.14560 H 1.0 1.93810 4.91010 0.32520 H 1.0 0.60170 3.62470 0.83730 H 1.0 0.15140 6.40800 0.10090 H 1.0 0.89630 5.99040 1.29280 $END File 9. General input for PBA single point energy calculations $CONTRL COORD=CART ICHARG= 1 MAXIT=200 MPLEVL=2 MULT=1 RUNTYP=ENERGY SCFTYP=RHF UNITS=ANGS $END $BASIS DIFFS=.true. GBASIS=N311 NDFUNC=2 NFFUNC=1 NGAUSS=6 NPFUNC=1 POLAR=HONDO7 $END $SCF DAMP=. false. DEM=.false. DIIS=.false. DIRSCF=.true. EXTRAP=.true. RSTRCT=.false. SHIFT=.false. SOSCF=.true. $END $STATPT METHOD=QA NSTEP=1000 OPTTOL=0.001 $END $P2P P2P=.T. DLB=.T. $END $FORCE TEMP=0 $END $GUESS GUESS=HUCKEL $END $SYSTEM MWORDS=100 $END $DATA high level spe for RSE

PAGE 178

Appendices 164 C1 C 6.0 1.64210 1.04520 0.01360 C 6.0 1.39220 0.33140 0.06250 C 6.0 2.49530 1.18880 0.02550 C 6.0 3.79150 0.69980 0.04000 C 6.0 4.02160 0.66700 0.07370 C 6.0 2.93620 1.53430 0.04800 H 1.0 0.81830 1.73130 0.00850 C 6.0 0.03280 0.89950 0.14130 H 1.0 2.33020 2.25030 0.05320 H 1.0 4.61830 1.38580 0.06430 H 1.0 5.02340 1.05040 0.12550 H 1.0 3.09770 2.59580 0.08560 C 6.0 1.07700 0.24630 0.47900 C 6.0 2.44780 0.83830 0.54640 C 6.0 3.55620 0.11790 0.02940 O 8.0 4.58990 0.44840 0.40270 O 8.0 3.31650 1.35010 0.14560 H 1.0 0.03150 1.94980 0.09300 H 1.0 1.05380 0.79980 0.72080 H 1.0 2.51460 1.77740 0.01190 H 1.0 2.69920 1.04500 1.58790 $END File 10. General input for PBA radical geometry optimizations (truncated) $CONTRL ICHARG= 1 MULT=2 RUNTYP=OPTIMIZE SCFTYP=UHF COORD=CART $END $CONTROL UNITS=ANGS MAXIT=200 $END $DFT DFTTYP=B3LYP METHOD=GRID $END $BASIS GBASIS=N31 NGAUSS=6 $END $STATPT METHOD=QA NSTEP=1000 OPTTOL=0.0001 HESS=CALC $END $GUESS GUESS=HUCKEL $END $SYSTEM MWORDS=200 $END $P2P P2P=.T. DLB=.T. $END $DATA pba radical geopt File 11. General input for PBA radical single point energy calculations (truncated)

PAGE 179

Appendices 165 $CONTROL RUNTYP=ENERGY COORD=CART ICHARG= 1 MULT=2 $END $CONTROL MAXIT=200 MPLEVL=2 SCFTYP=UHF UNITS=ANGS $END $BASIS GBASIS=N311 NDFUNC=2 NFFUNC=1 NGAUSS=6 NPFUNC=1 $END $STATPT NSTEP=1000 $END $GUESS GUESS=HUCKEL $END $SYSTEM MWORDS=200 $END $DATA high level spe for RSE File 12. General input for cinnamate ana log geometry optimizations (truncated) $CONTRL ICHARG= 1 MULT=1 RUNTYP=OPTIMIZE SCFTYP=RHF COORD=CART $END $CONTRL UNITS=ANGS MAXIT=10000 $END $SYSTEM KDIAG=0 TIMLIM=999999999 AOINTS=DIST $END $SMP CSMTX=.T. $END $P2P P2P=.T. DLB=.T. $END $SCF DIRSCF=.T. FDIFF=.F. SOSCF=.F. DIIS=.T. ETHRSH=2.0 $END $DFT DFTTYP=B3LYP METHOD=GRID $END $BASIS GBASIS=N31 NGAUSS=6 $END $STATPT METHOD=QA NSTEP=1000 OPTTOL=0.0001 HESS=CALC $END $GUESS GUESS=HUCKEL $END $SYSTEM MWORDS=200 $END $P2P P2P=.T. DLB=.T. $END $DATA Appendix B PCGAMESS Hessian Calculation Input Files File 13. Hessian calculation for L3Cu(I) CuM for NAMD parameterization (truncated) $CONTRL SCFTYP=RHF RUNTYP=HESSIAN ICHARG=+1 MULT=1 COORD=ZMT NZVAR=105 $END $CONTRL MAXIT=100 ECP=SBK $END $SYSTEM TIMLIM=9999999999 MWORDS=200 $END $SCF DIRSCF=.TRUE. $END

PAGE 180

Appendices 166 $STATPT NSTEP=1000 HESS=CALC $END $GUESS GUESS=HUCKEL $END $BASIS GBASIS=SBK $END $ELPOT IEPOT=1 WHERE=PDC $END $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $ZMAT IZMAT(1)=1,2,1,1,3,2,1,4,2,1,5,1,1,6,5,1,7,5,1,8,2,1,9,4,1,10,9, 1,11,8,1,12,11,1,13,8,1,14,13,1,15,14,1,16,14,1,17,13,1,18,17,1,19,17, 1,20,14,1,21,16,1,22,21,1,23,20,1,24,23,1,25,24,1,26,25,1,27,26,1,28,26, 1,29,26,1,30,29,1 ,31,29,1,32,29,1,33,32,1,34,33,1,35,33,1,36,33,1,37,20, 2,3,2,1,2,4,2,1,2,5,1,2,2,6,5,1,2,7,5,1,2,8,2,1,2,9,4,2,2,10,9,4,2,11,8,2, 2,12,11,8,2,13,8,2,2,14,13,8,2,15,14,13,2,16,14,13,2,17,13,8,2,18,17,13, 2,19,17,13,2,20,14,13,2,21,16,14,2,22,21,16,2,23,20,14,2,24,23,20,2,25,24,23, 2,26,25,24,2,27,26,25,2,28,26,25,2,29,26,25,2,30,29,26,2,31,29,26,2,32,29,26, 2,33,32,29,2,34,33,32,2,35,33,32,2,36,33,32,2,37,20,14,3,4,2,1,3,3,5,1,2,3,3,6,5,1,2,3,7,5,1,2,3,8, 2,1,3,3,9,4,2,1,3,10,9,4,2,3,11,8,2,1,3,12,11,8,2,3,13,8,2,1,3,14,13,8,2,3,15,14,13,8,3,16,14,13,8, 3,17,13,8,2,3,18,17,13,8,3,19,17,13,8,3,20,14,13,8,3,21,16,14,13,3,22,21,16,14,3,23,20,14,13,3,2 4,23,20,14,3,25,24,23,20,3,26,25,24,23,3,27,26,25,24,3,28,26,25,24,3,29,26,25,24,3,30,29,26,25, 3,31,29,26,25,3,32,29,26,25,3,33,32,29,26,3,34,33,32,29,3,35,33,32,29,3,36,33,32,29,3,37,20,14, 13, $END $DATA File 14. Hessian calculation for L3Cu(II) CuM for NAMD parameterization (truncated) $CONTRL SCFTYP=RHF RUNTY P=HESSIAN ICHARG=+2 MULT=2 COORD=ZMT NZVAR=105 $END $CONTRL MAXIT=100 ECP=SBK $END $SYSTEM TIMLIM=9999999999 MWORDS=200 $END $SCF DIRSCF=.TRUE. $END $STATPT NSTEP=1000 HESS=CALC $END $GUESS GUESS=HUCKEL $END $BASIS GBASIS=SBK $END $ELPOT IEPOT=1 WHE RE=PDC $END $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $ZMAT IZMAT(1)=1,2,1,1,3,2,1,4,2,1,5,1,1,6,5,1,7,5,1,8,2,1,9,4,1,10,9, 1,11,8,1,12,11,1,13,8,1,14,13,1,15,14,1,16,14,1,17,13,1,18,17,1,19,17, 1,20,14,1,21,16,1,22,21,1,23,20,1,24,23,1,25,24,1,26,25,1,27,26,1,28,26,

PAGE 181

Appendices 167 1,29,26,1,30,29,1,31,29,1,32,29,1,33,32,1,34,33,1,35,33,1,36,33,1,37,20, 2,3,2,1,2,4,2,1,2,5,1,2,2,6,5,1,2,7,5,1,2,8,2,1,2,9,4,2,2,10,9,4,2,11,8,2, 2,12,11,8,2,13,8,2,2,14,13,8,2,15,14,13,2 ,16,14,13,2,17,13,8,2,18,17,13, 2,19,17,13,2,20,14,13,2,21,16,14,2,22,21,16,2,23,20,14,2,24,23,20,2,25,24,23, 2,26,25,24,2,27,26,25,2,28,26,25,2,29,26,25,2,30,29,26,2,31,29,26,2,32,29,26, 2,33,32,29,2,34,33,32,2,35,33,32,2,36,33,32,2,37,20,14,3,4,2,1,3,3,5 ,1,2,3,3,6,5,1,2,3,7,5,1,2,3,8, 2,1,3,3,9,4,2,1,3,10,9,4,2,3,11,8,2,1,3,12,11,8,2,3,13,8,2,1,3,14,13,8,2,3,15,14,13,8,3,16,14,13,8, 3,17,13,8,2,3,18,17,13,8,3,19,17,13,8,3,20,14,13,8,3,21,16,14,13,3,22,21,16,14,3,23,20,14,13,3,2 4,23,20,14,3,25,24,23,20,3,26, 25,24,23,3,27,26,25,24,3,28,26,25,24,3,29,26,25,24,3,30,29,26,25, 3,31,29,26,25,3,32,29,26,25,3,33,32,29,26,3,34,33,32,29,3,35,33,32,29,3,36,33,32,29,3,37,20,14, 13, $END $DATA File 15. Hessian Calculation for ascorbic acid for NAMD parameterization (truncated) $CONTRL SCFTYP=RHF RUNTYP=HESSIAN ICHARG=0 MULT=1 COORD=ZMT NZVAR=54 $END $CONTRL MAXIT=75 $END $SYSTEM TIMLIM=999999999 MWORDS=235 $END $SCF DIRSCF=.TRUE. $END $STATPT NSTEP=1000 $END $BASIS GBASIS=N31 NGAUSS=6 $END $ELPOT IEPOT=1 WHER E=PDC $END $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $ZMAT IZMAT(1)=1,2,1,1,3,2,1,4,3,1,5,4,1,6,2,1,7,4,1,8,7,1,9,5,1,10,1, 1,11,7,1,12,11,1,13,4,1,14,7,1,15,11,1,16,11,1,17,12,1,18,8,1,19,10,1,20,9,2,3,2,1,2,4,3,2,2,5,4,3, 2,6,2,3,2,7,4,3,2,8,7,4,2,9,5,4,2,10,1,2,2,11,7,4,2,12,11,7,2,13,4,3,2,14,7,4,2,15,11,7,2,16,11,7,2, 17,12,11,2,18,8,7,2,19,10,1,2,20,7,5,3,4,3,2,1,3,5,4,3,2,3,6,2,3,4,3,7,4,3,2,3,8,7,4,3,3,9,5,4,3,3,10 ,1,2,3,3,11,7,4,3,3,12,11,7,4,3,13 ,4,3,2,3,14,7,4,3,3,15,11,7,4,3,16,11,7,4,3,17,12,11,7,3,18,8,7,4, 3,19,10,1,2,3,20,9,5,4, $END $ZMAT DLC=.TRUE. AUTO=.TRUE. $END $DATA ascorbate toppar File 16. Hessian Calculation for mimosine for NAMD parameterization (truncated)

PAGE 182

Appendices 168 $CONTRL SCFTYP=RHF RUNTYP=HESSIAN ICHARG= 1 MULT=1 COORD=ZMT NZVAR=63 $END $CONTRL MAXIT=75 $END $SYSTEM TIMLIM=999999999 MWORDS=235 $END $SCF DIRSCF=.TRUE. $END $STATPT NSTEP=1000 $END $BASIS GBASIS=N31 NGAUSS=6 $END $ELPOT IEPOT=1 WHERE=PDC $END $PDC PTSEL=CONNOLLY CONSTR=CHARGE $END $DATA mimosine toppar $FORCE PURIFY=.TRUE. PRTIFC=.TRUE. DECOMP=.TRUE. $END $IZMAT IZMAT(1)=2,1,3,2,4,3,5,4,6,5,7,2,8,4,9,8,10,9,11,1,12,9,13,12,14,12, 15,3,16,5,17,6,18,7,19,8,20,8,21,9,22,10,23,10,3,2,1,4,3,2,5 ,4,3,6,5 ,4,7,2,3,8,4,3,9,8,4,10,9,8,11,1,2,12,9,8,13,12,9,14,12,9,15,3,2,16,5,4,17,6,5,18,7,2,19,8,4,20,8,4, 21,9,8,22,10,9,23,10,9,4,3,2,1,5,4,3,2,6,5,4,3,7,2,3,4,8,4,3,2,9,8,4,3,10,9,8,4,11,1,2,3,12,9,8,4,13, 12,9,8,14,12,9,8,15,3,2,1,16,5,4,3,17,6,5,4,18 ,7,2,3,19,8,4,3,20,8,4,3,21,9,8,4,22,10,9,8,23,10,9,8 $END Appendix C NAMD Configuration Files File 17. General configuration file for the rapid equilibration of proteins # Equilibration run # # Monday May 26th, 2008 # # Edward W. Lowe Jr # Merkler Research Group # Department of Chemistry # University of South Florida structure dbm_for_equil.solvated.ionized.psf coordinates dbm_for_equil.solvated.ionized.pdb seed 238897325 binaryoutput on

PAGE 183

Appendices 169 outputname dbm_equilibration DCDfile dbm_equilibration.dcd # output frequency outputenergies 1000 outputtiming 1000 restartfreq 10000 DCDfreq 1000 # integration parameters timestep 1.0 rigidBonds all nonbondedFreq 1 fullElectFrequency 1 stepspercycle 10 # FF paraTypeCharmm on parameters par_all22_prot+Cu+ASC+MIM.inp parameters par_all27_prot_lipid_na.inp exclude scaled1 4 14scaling 1.0 switching on cutoff 12.0 switchdist 10.0 pairlistdist 14.0 # PBC cellBasisVector1 94.0 0.0 0.0 cellBasisVector2 0.0 94.0 0.0 cellBasisVector3 0.0 0.0 94.0 cellOrigin 0.0 0.0 0.0 wrapall on # PME PME on PMEGridSizeX 96 PMEGridSizeY 96 PMEGridSizeZ 96

PAGE 184

Appendices 170 # set temperature to 0 initially temperature 0 # fix the backbone atoms and coppers in DBM fixedAtoms on fixedAtomsForces on fixedAtomsFile dbm_for_equil.solvated.ionized.fixed.pdb fixedAtomsCol B # turn on constraints for all of the alpha carbons in DBM constraints on consRef dbm_for_equil.solvated.ionized.constrained.pdb consKFile dbm_for_equil.so lvated.ionized.constrained.pdb consKCol B # set piston will large damping langevin on langevinDamping 5 langevinTemp 310.15 langevinHydrogen off langevinPiston on langevinPistonTarget 1.01325 langevinPistonPeriod 100 langevinPistonDecay 50 langevinPistonTemp 310.15 useGroupPressure yes # run one step to get into scripting mode minimize 0 # turn off langevin pressure control dynamics langevinPiston off # minimize non backbone atoms minimize 10000 # minimize all atoms fixedAtoms off

PAGE 185

Appendices 171 minimize 10000 # heat with alpha carbons constrained run 10000 # equilibrate volume with alpha carbons constrained (NPT on) langevinPiston on run 10000 # equilibrate volume without constraints (1ns NPT) constraintScaling 0 run 1000000 File 18. General configuration file for long molecular dynamics simulations using multiple stepping algorithm # # PHM domain with bound coppers # NPT 80 ns run for asc binding site # Saturday July 20th, 2007 # # Edward W. Lowe Jr # Merkler Research Group # Department of Chemistry # University of South Florida structure PHM+ox_asc_solvated_ionized.psf coordinates PHM+ox_asc_solvated_ionized.pdb bincoordinates PHM+ox_asc_NPT_WRAPALL_17.restart.coor extendedsystem PHM+ox_asc_NPT_WRAPALL_ 17.restart.xsc binvelocities PHM+ox_asc_NPT_WRAPALL_17.restart.vel seed 238897325 binaryoutput on outputname PHM+ox_asc_NPT_WRAPALL_18 dcdfile PHM+ox_asc_NPT_WRAPALL_18.dcd

PAGE 186

Appendices 172 # output frequency outputenergies 10000 outputtiming 10000 restartfreq 10000 DCDfreq 10000 # integration parameters timestep 1.0 rigidBonds all nonbondedFreq 2 fullElectFrequency 4 stepspercycle 20 # FF paraTypeCharmm on parameters par_all22_prot+Cu+ASC.inp parameters par_all27_prot_lipid_na.inp exclude sca led1 4 14scaling 1.0 switching on cutoff 12.0 switchdist 10.0 pairlistdist 14.0 # PBC cellBasisVector1 120.0 0.0 0.0 cellBasisVector2 0.0 120.0 0.0 cellBasisVector3 0.0 0.0 120.0 cellOrigin 0.0 0.0 0.0 wrapall on # PME PME on PMEGridSizeX 122 PMEGridSizeY 122 PMEGridSizeZ 122 # set piston will large damping

PAGE 187

Appendices 173 langevin on langevinDamping 5 langevinTemp 310.15 langevinHydrogen off langevinPiston on langevinPistonTarget 1.01325 langevinPistonPeriod 100 langevinPi stonDecay 50 langevinPistonTemp 310.15 useGroupPressure yes # run one step to get into scripting mode minimize 0 # MD for 80 ns run 80000000 File 19. General configuration file for steered molecular dynamics simulations # Edward W. Lowe Jr # Merkler Research Group # Department of Chemistry # University of South Florida # # Constant Velocity Pulling of reductant from PHM structure PHM+ox_asc1_bound_solvated_ionized.psf coordinates PHM+ox_asc1_bound_solvated_ionized.pdb bincoordinates 1.coor outputName asc_pulling_1_022408 set temperature 310 firsttimestep 0 # Input paraTypeCharmm on parameters par_all22_prot+Cu+ASC+MIM7.inp parameters par_all27_prot_lipid_na.inp temperature $temperature wra pWater on

PAGE 188

Appendices 174 wrapAll on # Force Field Parameters exclude scaled1 4 14scaling 1.0 cutoff 12. switching on switchdist 10. pairlistdist 14. # Integrator Parameters timestep 1.0 ;# 1fs/step nonbondedFreq 1 fullElectFrequency 1 stepspercycle 10 # Constant Temperature Control langevin off langevinDamping 5 langevinTemp $temperature langevinHydrogen no restartfreq 500 dcdfreq 500 xstFreq 500 outputEnergies 100 outputPressure 100 # Fixed Atoms Constraint if {1} { fixedAtoms on fixedAtomsFile PHM+ox_asc1.SMD.ref fixedAtomsCol B } SMD on SMDFile PHM+ox_asc1.SMD.ref SMDk 7 SMDVel 0.0001

PAGE 189

Appendices 175 SMDDir 0.467180057326 0.0811620777483 0.880429163064 SMDOutputFreq 10 run 1000000

PAGE 190

Appendices 176 Appendix D Monooxygenase X: Homology modeling results List of Figures ( Figure ) Figure A.1. Monooxygenase X solvated and ionized prior to equilibration using NAMD.

PAGE 191

Appendices 177 ( Figure ) Figure A.2. Monooxygenase X in ribbon representation of the secondary structure after equilibration with NAMD.

PAGE 192

Appendices 178 ( Figure ) Figure A.3. The rmsd versus time graph for the equilibration of Monooxygenase X using NAMD. Rmsd = 1.86.

PAGE 193

Appendices 179 Appendix E Equilibration results of oxidized PHM ( Figure ) Figure A.4. PHM crystal structure in ribbon format solvated prior to equilibration with NAMD.

PAGE 194

Appendices 180 ( Figure ) Figure A.5. The time versus rmsd for the equilibration of oxidized PHM with NAMD. Rmsd = 2.057 0 0.5 1 1.5 2 2.5 3 22 11Time, nanosecondsRmsd Anstroms

PAGE 195

Appendices 181 Appendix F Monooxygenase: Homology model ( Figure ) monooxygenase homology model prior to refinement with NAMD.

PAGE 196

Appendices 182 ( Figure ) monooxygenase homology model. Rmsd = 1.86 0 0.5 1 1.5 2 2.5 0 200 400 600 800 1000 1200 Time, picosecondsRmsd Anstroms

PAGE 197

Appendices 183 Appendix G Synthesis of 14C mimosine ( Figure ) Figure A.8. Mimosine synthase reaction scheme. N H2O O O H O o-a c e t yl -l -s e ri ne N H O H O 3 ,4-d i h ydrox ypy ri di ne m i m o s i n e s y n .N H2N H O O O O H l m i m o s i n e

PAGE 198

Appendices 184 ( Figure ) Figure A.9. The synthetic scheme for O acetyl serine

PAGE 199

Appendices 185 ( Figure ) Figure A.10. The s c h e m e of t h e synthesis of dihydroxypyridine.

PAGE 200

Appendices 186 ( Figure ) Figure A.11. The c h r om at ogr a m of t h e separation of a mimosine standard solution by R P HPLC. Mimosine std 2% O -phosphoric acid C18 254nm 25C 1ml/min N H2N H O O O O H m i m os i ne

PAGE 201

Appendices 187 ( Figure ) Figure A.12. The c h r om at ogr a m of t h e R P HPLC separation of dihydroxypyridine. N H O H O 3,4di hydr oxypyr i di ne

PAGE 202

Appendices 188 ( Figure ) Figure A.13. The s c h e m e of t h e enzymatic synthesis of 1 4C radiolabeled mimosine. N ot e t h at = 1 4C N H2O O O H O o-a c e t yl -l -s e ri ne N H O H O 3 ,4-d i h ydrox ypy ri di ne m i m o s i n e s y n .N H2N H O O O O H l m i m o s i n e * * *30mM KPi, pH 8

PAGE 203

Appendices 189 ( Figure ) Figure A.14. T h e c h r om at ogr a m of t h e R P H P L C s e p a r at i on of a 30minute time point for the conversion of O acetylserine and 3,4dihydroxypyridine to mimosine by mimosine synthase. N H O H O 3,4di hydr oxypyr i di ne N H2N H O O O O H m i m os i ne

PAGE 204

About the Author Edward W. Lowe, Jr. (Will) graduated as va ledictorian at A. Crawford Mosley High School in Panama City, FL in 2000. Will relocated to Tampa, FL in the Fall of 2000 to attend the University of South Florida. He gradua ted with Honors receiving a bachelors degree in chemistry in 2003. Will entered the chemistry graduate program at USF in 2003 and received his Ph.D. in the Fall 2008. When not performi ng geometry optimizations or molecular dynamics simulations in his spare time, Will enjoys pow er lifting having won the “Strongest Bull” competition in 2007. Above all, Will enjoys sp ending time with his wife, Kim, and daughter Addison.