COMPUTATIONAL INVESTIGATIONS OF CYTOCHROME P450 AROMATASE CATALYSIS AND BIOLOGICAL EVALUATION OF ISOFLAVONE AROMATASE INHIBITORS DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By John C Hackett, B.Sc. Pharm., M. S. * * * * * The Ohio State University 2004 Dissertation Committee: Approved by Professor Robert W. Brueggemeier, Adviser Professor Paul Blower _______________________________ Adviser Professor Pui-Kai Li Graduate Program in Pharmacy Professor Karl Werbovetz ABSTRACT Dioxygen-containing P450 catalytic intermediates including the reduced dioxygen and hydroperoxo species beyond the dioxygen-ligated enzyme have eluded direct structural observation. In this study, structures, vibrations, and relative energetics of low and intermediate spin states of these species are characterized with unrestricted density functional theory using gradient-corrected and hybrid exchange correlation potentials. All functionals tested produce quality geometries compared to available experimental data, and the dioxygen vibrational frequencies are in reasonable agreement with data from resonance Raman and infrared spectroscopic results. The red shifts which have been observed in the UV/Visible spectrum of wild-type and D251N- cytochrome P450cam from radiolytic reduction of these enzymes are reproduced well using RI-J time- dependent density functional theory calculations. Interestingly, the computed spectral patterns for the reduced dioxygen and hydroperoxo model systems are quite similar. B3LYP computations of 14N and 1H hyperfine coupling constants revealed that it is a competent level of theory to reproduce the experimental ENDOR values of the hydroperoxo species. It has been postulated that a species provisionally assigned as the reduced dioxygen intermediate has the same 14N hyperfine coupling constant as the hydroperoxo species due to the spin density distortion effects of protein active-site hydrogen-bond donors. Interestingly, even in the presence of hydrogen-bond donors ii which should mimic a gross excess of hydrogen-bonding potential, the spin density on iron never recovers to give rise to the experimentally observed hyperfine coupling constant. This observation raises some controversy about the true identity of the species assigned as the reduced dioxygen intermediate. B3LYP density functional theory calculations are used to unravel the mysterious third step of aromatase catalysis. The feasibility of mechanisms in which the reduced ferrous dioxygen intermediate mediates androgen aromatization are explored and determined to be unlikely. Mechanisms for the aromatization/deformylation sequence which are initiated by 1β-hydrogen atom abstraction by P450 Compound I are considered. 1β- Hydrogen atom abstraction from substrates in the presence of the 2,3-enol encounters strikingly low barriers (5.3-7.8 kcal/mol), whereas barriers for this same process rise to 17.0-27.1 kcal/mol in the keto tautomer. Transition states for 1β-hydrogen atom abstraction from enolized substrates in the presence of the 19-gem-diol decayed directly to the experimentally observed products. If the C19 aldehyde remains unhydrated, aromatization occurs with concomitant decarbonylation, and therefore does not support dehydration of the C19 aldehyde prior to the final catalytic step. On the doublet surface, the transition state connects to a potentially labile 1(10) dehydrogenated product, which may undergo rapid aromatization, as well as formic acid. As the reaction vectors indicated, ab initio molecular dynamics on the Born-Oppenheimer potential energy hypersurface confirmed that the 1β-hydrogen atom abstraction and deformylation or decarbonylation occur in a non-synchronous, coordinated manner. These calculations iii support a dehydrogenase behavior of aromatase in the final catalytic step, which can be summarized by 1β-hydrogen atom abstraction followed by gem-diol deprotonation. Aromatase, a cytochrome P450 hemoprotein that is responsible for estrogen biosynthesis by conversion of androgens into estrogens, has been an attractive target in the treatment of hormone-dependent breast cancer. As a result, a number of synthetic steroidal or nonsteroidal aromatase inhibitors have been successfully developed. In addition, there are several classes of natural products that exert potent activities in aromatase inhibition, with the flavonoids being most prominent. Previous studies have exploited flavone and flavanone scaffolds for the development of new aromatase inhibitors. In this dissertation, we describe design, synthesis, and biological evaluation of a novel series of 2-(4’- pyridylmethyl)thioisoflavones as the first example of synthetic isoflavone-based aromatase inhibitors. The biological evaluation of a series of 2-azole and 2-thioazole isoflavones as potential aromatase inhibitors are described. Differences in inhibitory activity of triazole and imidazole inhibitors are rationalized with density functional theory to expose a key difference in the electronic structure of these molecules. In addition, difference binding spectra of inhibitors to immunoaffinity-purified aromatase produces classical Type II spectra consistent with coordination of the nitrogen lone pair electrons to the P450 heme. iv For my loving wife and son, Rajini and Dev v ACKNOWLEDGMENTS I wish to thank my adviser, Dean Robert W. Brueggemeier for support over the course of my Ph.D. study and provision of unconditional freedom to pursue all of my research ideas, no matter how far the required techniques seemed to deviate from the traditions of the laboratory. If not for this, this thesis could not have materialized. I thank all faculty in the Division and in the Department of Chemistry for excellent and enthusiastic teaching. In particular, I thank Professor Christopher M. Hadad. His outstanding knowledge and enthusiasm for computational chemistry sparked my intense interest for this area of research. Without his guidance, the marriage of computational chemistry and the traditional enzymological approaches undertaken in our laboratory could not have flourished. Thus, the insight gained in this research into the most fundamental processes occuring within cytochrome P450 enzymes would not have been possible. I wish to express my gratitude to all my current and past labmates for their friendship over the years; Jon Baker, Jennifer Whetstone, Trevor Petrel, Jeannette Richards, Surachai Joomprabutra, and Danielle Pellegata, Danyetta Davis, and Edgar Diaz-Cruz. I want to acknowledge Mike Ivers, Kevin Schaefer, and Jason Holton, if not for their v i support and long-lasting friendship, success in the early years of my graduate study would have been far more difficult. I am especially grateful to Dr. Young-Woo Kim who is an amazingly talented scientist, extraordinary collaborator, and friend. I am grateful to Bin Su, for useful discussions about all aspects of my research and his friendship. I am especially indebted to Serena Landini, whose unwavering friendship for myself and family has made our life away from home enjoyable. Special mention must go to my wife, Rajini, for tolerating a change from a stable lifestyle in Florida and many sacrifices to endure this adventure. I cannot express the gratitude I have for her, as her partnership and love have provided the motivation for my intellectual and personal evolution. This research was supported by USAMRC Pre-doctoral Fellowship (DAMD17-02-1- 0529). Computations were carried out at the Ohio Supercomputer Center (OSC) and supported by OSC grant PAS0091. v ii VITA July 11, 1976……………………………...Born – Orlando, Florida 1999……………………………………… B.S. in Pharmacy, University of Florida 1999-2002..……………………………….Graduate Teaching Associate Division of Medicinal Chemistry and Pharmacognosy, The Ohio State University, Columbus, Ohio 2002-2004………………………………...Graduate Research Associate Division of Medicinal Chemistry and Pharmacognosy, The Ohio State University, Columbus, Ohio PUBLICATIONS Research Publication 1. Kim, Y.-W.; Hackett, J. C.; Brueggemeier, R. W. Synthesis and aromatase inhibitory activity of novel pyridine-containing isoflavones. J. Med. Chem. 2004, 47, 4032-4040. viii FIELDS OF STUDY Major Field: Pharmacy Medicinal Chemistry ix TABLE OF CONTENTS Page Abstract .....................................................................................................................................ii Dedication .................................................................................................................................v Acknowledgments ...................................................................................................................vi Vita..........................................................................................................................................vii List of Tables .........................................................................................................................xiv List of Figures......................................................................................................................xviii Chapters: 1. Introduction: Aromatase Catalysis and Inhibition...............................................................1 1.1 Physiology and Biosynthesis of Estrogens ........................................................................1 1.2 The Catalytic Mechanism of Aromatase............................................................................7 1.3 Inhibitors