University of South Florida University of South Florida

Scholar Commons Scholar Commons

Graduate Theses and Dissertations Graduate School

3-31-2006

The Theoretical Modeling, Design, and Synthesis of Key Structural The Theoretical Modeling, Design, and Synthesis of Key Structural

Units for Novel Molecular Clamps and Pro-Apoptotic Alpha Helix Units for Novel Molecular Clamps and Pro-Apoptotic Alpha Helix

Peptidomimetics Peptidomimetics

Stephanie Tara Weiss University of South Florida

Follow this and additional works at: https://scholarcommons.usf.edu/etd

Part of the American Studies Commons

Scholar Commons Citation Scholar Commons Citation Weiss, Stephanie Tara, "The Theoretical Modeling, Design, and Synthesis of Key Structural Units for Novel Molecular Clamps and Pro-Apoptotic Alpha Helix Peptidomimetics" (2006). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/2753

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact scholarcommons@usf.edu.

The Theoretical Modeling, Design, and Synthesis of Key Structural Units for Novel

Molecular Clamps and Pro-Apoptotic Alpha Helix Peptidomimetics

by

Stephanie Tara Weiss

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: Mark L. McLaughlin, Ph.D. Srikumar Chellappan, Ph.D.

Abdul Malik, Ph.D. David Merkler, Ph.D. Edward Turos, Ph.D.

Date of Approval: March 31, 2006

Keywords: hydrazine, nitrosamine, substrate-targeted inhibitor, Bax, p53, MDM2

© Copyright 2006 , Stephanie T. Weiss

DEDICATION I dedicate this dissertation to the three people who most profoundly influenced my

decision to study chemistry. One is Paul Scudder, my first organic chemistry teacher.

The second is Wayne Brouillette, my M.S. advisor. Finally, I dedicate it to Mark

McLaughlin, my Ph.D. advisor, without whom this dissertation could not have existed.

ACKNOWLEDGMENTS

Many people have helped me in various ways throughout my time in graduate

school, and it is impossible for me to mention them all by name, but I hope that they all

know how much I appreciate everything that they’ve done for me. First, I thank Mark

McLaughlin, my Ph.D. advisor, for giving me the opportunity to complete my graduate

education. I would also like to thank all of my committee members: Srikumar

Chellappan, Abdul Malik, David Merkler, and Edward Turos, as well as Wayne Guida

for chairing my defense and proofreading this dissertation. Wayne Brouillette, my M.S.

advisor, got me started with chemistry research. I have had many great co-workers

during my time in graduate school, and I would like to thank Umut Oguz, Kiran

Avancha, Tanaji Talele, Mohanraja Kumar, Gabriel Garcia, Ben Davis, and all of my

other labmates for their support and help. I also would like to recognize Emily McIntosh,

Scott Haake, and several other undergraduate lab helpers. Ted Gauthier and Edwin

Rivera have my utmost appreciation for their help with the MS and NMR instruments,

respectively. Shen-Shu Sung performed the molecular modeling experiments described

in Chapter Four, and Jiazhi “George” Sun conducted the biological testing of my

compounds against MDM2 and Bax. I would like to thank all of my friends for their

support, especially David Rabson for proofreading this dissertation. Finally, I want to

recognize my parents Alicia and Donald Weiss, and my sister Shelly Weiss, for their

moral support, because I know it wasn’t easy for you to listen to so much chemistry talk!

i

TABLE OF CONTENTS LIST OF FIGURES iii LIST OF SCHEMES vi LIST OF ABBREVIATIONS viii LIST OF NMR SPECTRA x ABSTRACT xiii CHAPTER ONE: GENERAL INTRODUCTION:

PROTEINS AND THEIR INTERACTIONS 1 1.1: Proteins, Peptides, and Peptidomimetics 1 1.2: Molecular Clamps: Background 6 1.3: Molecular Clamps: A Model 11 1.4: References 19 CHAPTER TWO: SYNTHESIS OF POLAR AND NONPOLAR

HYDRAZINO AMINO ACIDS 22 2.1: Introduction 22 2.1.1: Hydrazines 22 2.1.2: Synthesis of Hydrazines 24 2.1.3: Use of Hydrazines in Peptidomimetics 30 2.1.4: Use of Hydrazines in β-Sheet Mimics 33 2.2: Results and Discussion 35 2.3: Experimental Data 47 2.4: References 69 CHAPTER THREE: SYNTHESIS OF MDM2 INHIBITORS

AND KEY STRUCTURAL UNITS 73 3.1: Introduction 73 3.1.1: The Hallmarks of Cancer 73 3.1.2: Apoptosis and MDM2/p53 75 3.1.3: Inhibiting the MDM2/p53 Interaction with Small Molecules 78 3.1.4: Inhibiting MDM2/p53 with Peptides and Peptidomimetics 83

ii

3.2: Results and Discussion 85 3.3: Experimental Data 92 3.4: References 103 CHAPTER FOUR: DESIGN AND SYNTHESIS OF A BAX PROTEIN

α-HELIX MIMIC 106 4.1: Introduction 106 4.1.1: Pro-Apoptotic and Anti-Apoptotic Members of the Bcl-2 Family. 106 4.1.2: Mimicking the BH3 Domain of Pro-Apoptotic Proteins. 110 4.2: Results and Discussion 116 4.3: Experimental Data 135 4.4: References 143 CHAPTER FIVE: APPENDIX: PROTON AND CARBON NMR SPECTRA 147 ABOUT THE AUTHOR End Page

iii

LIST OF FIGURES

Figure 1.1A: Structure of an Alpha Helix (1). 2 Figure 1.1B: Structure of a Beta Sheet (1). 2 Figure 1.2: Leu-Enkephalin, a Natural Peptide Ligand of Opiate Receptors,

and Morphine, an Opiate Receptor Peptidomimetic (4). 4

Figure 1.3: Schematic of the Substrate-Targeted Inhibitor Concept (8). 7 Figure 1.4: Schematic of the Binding of a Substrate by a MC. 11 Figure 1.5: Dissociation Constants for a Classical System with an

Enzyme-Targeted Inhibitor. 12 Figure 1.6: Dissociation Constants for a MC Substrate-Targeted Inhibitor. 12 Figure 1.7A: Effect of Substrate Concentration on Relative Enzyme Velocity 14 Figure 1.7B: Effect of MC Concentration on Relative Enzyme Velocity. 15 Figure 1.7C: Effect of Km Value on Relative Enzyme Velocity. 16 Figure 1.7D: Effect of KMC Value on Relative Enzyme Velocity. 17 Figure 2.1: Example of an Azapeptide (23). 31 Figure 2.2: Examples of Hydrazides (26-28). 31 Figure 2.3: Examples of N-Amino Amides (26,28). 32 Figure 2.4: Constraining the Extended Conformation of a Peptide to Mimic a Beta Sheet (22). 34 Figure 2.5: Some Constrained Hydrazine-Containing

Peptidomimetics (31,32). 34

iv

Figure 2.6: NMRs of N-Benzyl-(L)-Threonine Methyl Ester (top) and N-Benzyl-(D)-Threonine Methyl Ester. 41 Figure 3.1: Cellular Pathways that Can Be Affected in Cancer Cells (1). 74 Figure 3.2: The Intrinsic (Mitochondrial) and Extrinsic (FAS)

Pathways (3). 76

Figure 3.3: Regulation of p53 by MDM2 (7). 77 Figure 3.4: Small Molecule Inhibitors of MDM2 (13). 80 Figure 3.5: Structure of syc-7, a Non-Peptide Inhibitor of HDM2 (19). 82 Figure 3.6: Design, Model, and Structures of Hamilton’s Terphenyl Helices (21,22). 84 Figure 3.7: Structure and Model of the McLaughlin Piperazine Helix. 85 Figure 3.8: Synthesis of McLaughlin Piperazine Dione α-Helix Mimics. 86 Figure 3.9: MDM2 Small-Molecule Inhibitor Lead Compound

NSC-131734. 90 Figure 4.1: The Bcl-2 Family of Proteins (2). 107 Figure 4.2: Theorized Mechanism of Action of Bax (6). 109 Figure 4.3: Structures of Some Small-Molecule Bcl-2 Family

Inhibitors (8, 9). 111

Figure 4.4: Structures of Two Pro-Apoptotic Natural Products (12,13). 111 Figure 4.5: Chemical and Stereo X-Ray Structure of an Oligoamide (17). 113 Figure 4.6: Terphenyl Derivatives with Binding Affinity for Bcl-xL (21). 114 Figure 4.7: Terephthalamide Derivatives with Binding Affinity for

Bcl-xL (23). 115

Figure 4.8: Structure of the Pro-Apoptotic Bax Protein. 117 Figure 4.9: Structures of Some NCI Compounds Virtually Screened to Fit into the Hydrophobic Pocket of Bax. 118

v

Figure 4.10: Some Iterations to Obtain Compound 24 from NSC 109816. 119 Figure 4.11: Mimicking the Alpha Helix of Bax. 120 Figure 4.12: Compound 24 Docked to Bax in Lieu of the Helix. 121

vi

LIST OF SCHEMES

Scheme 2.1: Reduction of a Hydrazone to a Hydrazine (7). 24 Scheme 2.2: Strecker-Like Synthesis of a Hydrazino Acid (6). 24 Scheme 2.3: Reaction of Hydrazine with an α-Halo Acid (8, 9). 25 Scheme 2.4: Synthesis of a Hydrazine via Rearrangement of a Hydantoic

Acid (10). 26

Scheme 2.5: Attempt to Synthesize a Hydrazine Using a Hydroxylamine Derivative (10). 27 Scheme 2.6: Synthesis of a Hydrazine Via N-Amination (10). 27 Scheme 2.7: Synthesis of a Hydrazine from a Nosyloxy Ester (12). 28 Scheme 2.8: Hydrazine Formation Via Asymmetric Electrophilic Amination (13,14). 28 Scheme 2.9: Synthesis of Hydrazines Using Oxaziridines (17, 20). 29 Scheme 2.10: Synthesis of Hydrazines from Amino Acids Via

Nitrosamines (21). 30

Scheme 2.11: Procedure for the Synthesis of Five Nonpolar Amino Acids (21,22). 36

Scheme 2.12: Synthesis of the Ala (6a), Lys (6b) Ser (6c), Thr (6d), and Tyr (6e) Boc-Protected Hydrazines from the Free Amines (1a-e). 37

Scheme 2.13: Unsuccessful Two-Step Method for Synthesizing Compound 2a 38 Scheme 2.14: Epimerization of Thr Due to 2-Step Benzylation Procedure. 40 Scheme 2.15: Unsuccessful Attempt to Reduce Lys Nitrosamine 4b. 42

vii

Scheme 2.16: Procedure for Synthesizing Compound 5b (Lys). 43 Scheme 2.17: Synthesis of Ser, Thr, and Tyr Methyl Esters (1c-e). 44 Scheme 2.18: Procedure for Synthesizing Compound 1b (Lys). 45 Scheme 2.19: Deprotection of Nonpolar Amino Acid Hydrazines. 46 Scheme 3.1: Synthesis of Alkylated Hydrazino Amino Diacids. 87 Scheme 3.2: Synthesis of Tryptophan Methyl Esters 14b and 14c. 88 Scheme 3.3: Results of the Hydrolysis of 16a with NaOH versus LiOH. 90 Scheme 3.4: Synthesis of MDM2 Inhibitors 23a-c. 91 Scheme 4.1: Retrosynthesis of Compound 24. 122 Scheme 4.2: Synthesis of Indole 25. 123 Scheme 4.3: Failed Attempts to Synthesize 31 via Reductive Amination

of 29. 124

Scheme 4.4: Successful Synthesis of 31 via N, N-Dialkylation. 125 Scheme 4.5: Unsuccessful Attempts to Oxidize 29 and 31 to the

Carboxylic Acids. 126

Scheme 4.6: Unsuccessful Attempt to N,N-Dialkylate a Benzoic Ester. 128 Scheme 4.7: Unsuccessful Attempt to Perform Reductive Amination on 38. 129 Scheme 4.8: Synthesis of Succinimide 41 from Aniline 29. 130 Scheme 4.9: Side Product of the Succinimide Reaction. 131 Scheme 4.10: Proposed Synthesis of 44 from 41. 133 Scheme 4.11: Alternative Proposed Route to Compound 26. 134

viii

LIST OF ABBREVIATIONS

µM: Micromolar AcOAc: Acetic Anhydride AcOH: Acetic Acid ADP: Azadipeptide Unit Ala: Alanine ANT: Adenine Nucleotide Transporter Bcl-2: B-Cell Lymphoma 2 BH3: Bcl-2 Homology Region 3 Boc: tert-Butoxycarbonyl Cbz: Carboxybenzyl conc.: Concentrated Dap: Diaminopimelic Acid DBAD: Di-tert-butyl Azodicarboxylate DCM: Dichloromethane DIEA: Diisopropylethylamine DMF: N,N-Dimethylformamide DMSO: Dimethylsulfoxide DP: Dipeptide DPU: Dipeptide Unit E: Enzyme ESI: Electrospray Ionization HDM2: Human Double Minute 2 HR: High Resolution IC50: 50% Inhibitory Concentration Ile: Isoleucine Ki: Inhibitor-Enzyme Dissociation Constant Km: Enzyme-Substrate Dissociation Constant KMC: Inhibitor-Substrate Dissociation Constant LiOH: Lithium Hydroxide Leu: Leucine Lys: Lysine MC: Molecular Clamp MDM2: Mouse Double Minute 2 MeOH: Methanol MIM: Mitochondrial Inner Membrane mM: Millimolar MOM: Mitochondrial Outer Membrane

ix

MP: Melting Point MS: Mass Spectrum NBS: N-Bromosuccinimide nM: Nanomolar NMR: Nuclear Magnetic Resonance Spectrum P: Product PBP: Penicillin Binding Protein Phe: Phenylalanine pM: Picomolar PTP: Permeability Transition Pore p-TSOH: p-Toluenesulfonic Acid Rf: Retardation Factor S: Substrate Ser: Serine TBAB: Tetrabutylammonium Bromide TEA: Triethylamine TFA: Trifluoroacetic Acid THF: Tetrahydrofuran Thr: Threonine TLC: Thin Layer Chromatography Trp: Tryptophan Tyr: Tyrosine Val: Valine VDAC: Voltage-Dependent Anion Channel

x

LIST OF NMR SPECTRA Spectrum 5.1: 148 Spectrum 5.2: 149 Spectrum 5.3: 150 Spectrum 5.4: 151 Spectrum 5.5: 152 Spectrum 5.6: 153 Spectrum 5.7: 154 Spectrum 5.8: 155 Spectrum 5.9: 156 Spectrum 5.10: 157 Spectrum 5.11: 158 Spectrum 5.12: 159 Spectrum 5.13: 160 Spectrum 5.14: 161 Spectrum 5.15: 162 Spectrum 5.16: 163 Spectrum 5.17: 164 Spectrum 5.18: 165 Spectrum 5.19: 166

xi

Spectrum 5.20: 167 Spectrum 5.21: 168 Spectrum 5.22: 169 Spectrum 5.23: 170 Spectrum 5.24: 171 Spectrum 5.25: 172 Spectrum 5.26: 173 Spectrum 5.27: 174 Spectrum 5.28: 175 Spectrum 5.29: 176 Spectrum 5.30: 177 Spectrum 5.31: 178 Spectrum 5.32: 179 Spectrum 5.33: 180 Spectrum 5.34: 181 Spectrum 5.35: 182 Spectrum 5.36: 183 Spectrum 5.37: 184 Spectrum 5.38: 185 Spectrum 5.39: 186 Spectrum 5.40: 187 Spectrum 5.41: 188

xii

Spectrum 5.42: 189 Spectrum 5.43: 190 Spectrum 5.44: 191 Spectrum 5.45: 192 Spectrum 5.46: 193 Spectrum 5.47: 194 Spectrum 5.48: 195 Spectrum 5.49: 196 Spectrum 5.50: 197 Spectrum 5.51: 198 Spectrum 5.52: 199 Spectrum 5.53: 200 Spectrum 5.54: 201

xiii

The Theoretical Modeling, Design, and Synthesis of Key Structural Units for Novel

Molecular Clamps and Pro-Apoptotic Alpha Helix Peptidomimetics

Stephanie Tara Weiss

ABSTRACT

This dissertation presents the theory and practice of designing, synthesizing and

using peptidomimetics to disrupt protein-protein interactions. Our general strategy is to

design and synthesize peptidomimetics that will mimic peptide secondary structures (α-

helices and β-sheets). Chapter One is a theoretical examination of the feasibility of using

β-sheet mimics called molecular clamps to inhibit substrate-receptor interactions by

blocking the substrate rather than the receptor or enzyme. Several natural and synthetic

examples of this approach are given in support of this concept. We also present the

results of a kinetic modeling study and a consideration of which types of systems would

be the best candidates for a substrate-targeted inhibitor approach. Chapter Two relates a

continuation of previous work in our lab to synthesize five novel β-protected hydrazino

amino acids. These hydrazines are essential precursors for synthesizing constrained β-

strand mimetics. We showed that we could selectively deprotect the α-nitrogen of the

hydrazines, and we synthesized several novel examples of polar β-protected hydrazino

amino acids. Chapter Three discusses the design and synthesis of small-molecule and

peptidomimetic MDM2 inhibitors, including our work on synthesizing a new class of α-

helix mimics that have improved water solubility compared with previously reported

xiv

examples of α-helix mimics. As with the constrained β-strand mimics described in

Chapter Two, the synthesis of novel hydrazino amino acid precursors is a key step in

synthesizing our α-helix mimics. One isoleucine hydrazine derivative was synthesized,

and progress was made toward synthesizing two other hydrazines from tryptophan. In

addition, the synthesis of three potential small-molecule inhibitors of MDM2 is

described. Chapter Four describes the use of the GLIDE program to design and evolve

an α-helix mimic that will interact with the pro-apoptotic protein Bax. Progress toward

the synthesis of this compound is also reported.

1

CHAPTER ONE

GENERAL INTRODUCTION: PROTEINS AND THEIR INTERACTIONS 1.1 Proteins, Peptides, and Peptidomimetics

Proteins and peptides are biopolymers comprised of monomers called amino

acids. There are twenty naturally occurring amino acids, and they are linked together by

amide bonds to form linear chains. Shorter chains of less than 40 amino acids are called

peptides, while longer chains are called polypeptides or proteins (1).

The linear amino acid sequence of a protein is called its primary structure. A

protein’s primary structure is ultimately responsible for the structure and function of that

protein. However, proteins are also organized on additional levels based on the local and

global folding of the polypeptide chain. Local folding gives rise to the protein’s

secondary structures and occurs due to hydrogen bonding among amino acids. Two

common secondary structures are the α-helix and the β-sheet (1).

An α-helix is a spiral shape that occurs due to hydrogen bonding between the

carbonyl group of one amino acid and the amine group of another amino acid located

four positions up the chain. The hydrogen bonds are oriented parallel to the axis of the

helix, while the side chains of the amino acids point out approximately perpendicularly.

See Figure 1.1A below (1). Beta sheets resemble a pleated piece of paper, and are also

held together by hydrogen bonds between the carbonyl and amino groups of different

amino acids. The hydrogen bonds lie within the plane of the pleats, while the amino acid

2

side chains are oriented perpendicular to the peptide backbone both above and below the

plane of the pleats. See Figure 1.1B (1).

Figure 1.1A: Structure of an Alpha Helix (1). Hydrogen bonds in an α-helix are parallel to the helical axis, while the amino acid side chains are oriented perpendicularly.

Figure 1.1B: Structure of a Beta Sheet (1). The peptide backbone forms the pleated sheet, while the amino acid side chains stick out above and below it.

Proteins also have a global folding conformation that comprises their tertiary

structure. The α-helices, β-sheets, and other local folding patterns interact with one

another to form the overall shape of the protein (1). Although infinite possible protein

conformations are possible, each protein must be folded correctly in order to be capable

of performing its biological function. This function often entails interaction with other

proteins (2). The importance of protein-protein interactions has only recently become

appreciated, along with the possibilities for interfering with protein-protein interactions in

order to probe biological systems and develop new therapeutics (3).

There are several ways in which protein-protein interactions might be prevented.

One obvious possibility is to screen for or design a small organic molecule inhibitor.

Small molecule inhibitors are often effective in cases where the binding residues of the

3

protein are clustered together or there is a known natural product inhibitor (3). However,

since the size of a protein is very large compared with a typical natural product or

synthetic organic inhibitor, it may be difficult to block the entire protein surface by using

small molecules. There is also the possibility that there may be no binding pocket on the

surface of a flat protein to which a small organic inhibitor could bind; the surface of

many proteins is flat (3).

A second possibility is to use inhibitors that are actual peptides with the same

sequence as the portions of the protein that interact with other proteins. Such peptides

serve as competitive inhibitors of the proteins whose binding sites they resemble. This

strategy was successfully used to inhibit several examples of protein-protein interactions,

including HIV-1 protease dimerization, herpes simplex virus ribonucleotide reductase

dimerization, and p53/HDM2 interactions (3). However, this strategy tends not to work

well in cases where amino acid residues from noncontiguous parts of the protein’s

primary sequence must come together to form an epitope that is then capable of

interacting with another protein (3). In addition, peptides tend to have limitations in

terms of poor bioavailability and metabolic stability in living organisms (3), as well as

poor pharmacokinetics and potential side effects due to a lack of selectivity (4).

However, the peptide-based strategy is not limited merely to making an exact

copy of the binding peptide portion of the protein of interest. The chemical and

biological properties of inhibitory peptides can be modified by incorporating non-natural

amino acids into their sequence, or by even dispensing with the peptide’s amide-based

backbone altogether (3). Such modifications give rise to yet another possible source of

inhibitors called peptidomimetics.

4

Peptidomimetics have peptide-like structures and activities. However, since they

do not have actual peptide bonds, the pharmacokinetic properties of peptidomimetics are

modified. This allows peptidomimetics to interact with receptors and enzymes

analogously to their parent peptides while avoiding many of the problems that arise from

using actual peptides (4). One of the best-known examples of a peptidomimetic is

morphine, an opiate receptor agonist (4). Unlike most peptide-peptidomimetic pairs, the

peptidomimetic morphine was known long before the natural enkephalin peptide ligands

of the opiate receptors were discovered (4). The structures of Leu-enkephalin and

morphine are shown in Figure 1.2.

OH

NH2

NH

O

NH

O

NH

O

NH

O

COOH

Leu-enkephalin

OHO

OH

N

morphine

Figure 1.2: Leu-Enkephalin, a Natural Peptide Ligand of Opiate Receptors, and Morphine, an Opiate Receptor Peptidomimetic (4). The analgesic activity of morphine has been known for over 2000 years, but the natural peptide enkephalin ligands were not discovered until the 1970s.

Unlike morphine, most peptidomimetics are purposefully designed based on the

structure of their parent peptides (4). Farmer’s Rules describe a systematic method for

converting a peptide into a peptidomimetic (4). These rules include using the simplest

possible active portion of the peptide (the pharmacophore), maintaining the steric

parameters of the parent peptide, beginning with a flexible molecule and then

progressively constraining it through several rounds of structure-activity relationship

5

studies, and mimicking the three-dimensional shape of the parent peptide, particularly

with regard to α-helices and β-sheets (4).

This dissertation presents the theory and practice of designing, synthesizing and

using peptidomimetics to disrupt protein-protein interactions. In general, our strategy for

doing this is to design and synthesize peptidomimetics that will mimic peptide secondary

structures (α-helices and β-sheets). The remaining portion of Chapter One is a

theoretical examination of the feasibility of using β-sheet mimics called molecular

clamps to inhibit substrate-receptor interactions by blocking the substrate rather than the

receptor or enzyme. Several natural and synthetic examples of this approach are given in

support of this concept, and a kinetic modeling study is presented. Chapter Two relates a

continuation of previous work in our lab to synthesize β-protected hydrazino amino

acids. These hydrazines are essential precursors for synthesizing constrained β-strand

mimetics. Chapter Three summarizes the current literature concerning the design and

synthesis of α-helix mimics, to block the interaction between p53 and MDM2. It also

describes our work on synthesizing a new class of α-helix mimics that have improved

water-solubility compared with previously reported examples of α-helix mimics. As

with the constrained β-strand mimics described in Chapter Two, the synthesis of

hydrazino amino acid precursors is a key step in synthesizing our α-helix mimics.

Chapter Four describes the use of molecular modeling and docking software to design an

α-helix mimic that will interact with the pro-apoptotic protein Bax. Also, the ongoing

synthesis of one compound that showed promise in the computer screening is reported.

6

1.2 Molecular Clamps: Background

The canonical strategy for blocking enzyme-substrate or receptor-ligand

interactions is to screen for or design a small molecule or peptide that can bind to the

enzyme or receptor in such a way as to inhibit that enzyme or receptor. This inhibition

can either be competitive, where it is targeted to the active site, or noncompetitive, where

it is targeted to a separate allosteric site. Both types of inhibition share the same general

strategy, however, which is to prevent the enzyme or receptor from conducting its normal

activity. This strategy has lead to multiple successes in drug discovery and continues to

do so, as exemplified by recent reports of retinoic acid analogs that inhibit retinoic acid

receptors (5) and inhibitors of cyclin-dependent kinases (6).

However, this strategy is not suitable in all cases, because many enzymes catalyze

reactions on more than one substrate. Blocking such enzymes leads to the undesired

effect of blocking other pathways that may not be involved in the mechanism whose

modulation is desired. One recent example of the potential bad outcome due to the

inhibition of promiscuous enzymes is farnesyltransferase and geranylgeranyltransferase

inhibitors intended to prevent prenylation of K-Ras (7). Inhibition of K-Ras prenylation

is desirable for cancer therapy; however, use of farnesyltransferase-specific inhibitors

will not prevent prenylation of K-Ras by geranylgeranyltransferase, and use of dual-

purpose prenylation inhibitors is toxic to the point of complete lethality (7).

One possible way to solve this problem is to take the complementary approach,

which entails having the inhibitor bind to the substrate instead of to the enzyme or

receptor. Rather than having the inhibitor compete with the substrate for the enzyme’s

active or allosteric site, the inhibitor competes with the enzyme to bind the substrate,

7

analogous to the way antibodies recognize and neutralize their antigens (8). Thus, the

substrate binds to the inhibitor, and this prevents that particular substrate from being

acted upon by the enzyme. However, the enzyme is left free to continue to perform its

functions on all other substrates (8). Figure 1.3 is a schematic comparing the effects on a

dual-substrate enzyme of using no inhibitor (top panel), a traditional enzyme-targeted

inhibitor (middle panel), or a substrate-targeted inhibitor (bottom panel).

Figure 1.3: Schematic of the Substrate-Targeted Inhibitor Concept (8). The top panel shows two substrates (squiggles) interacting with the same enzyme (block) in the absence of inhibitor. The middle panel shows an inhibitor (triangle) that targets the enzyme, thereby blocking all functions of that enzyme. The bottom panel shows an inhibitor (boomerang) that targets one substrate (left squiggle), leaving the enzyme free to continue catalyzing its reaction with the second substrate (right squiggle).

8

There is good reason to believe that this approach of inhibiting the substrates

rather than the enzymes or receptors would be successful based on the prevalence of both

natural and synthetic examples in the literature. Some of the natural precedents include

not only antibodies, but also antibiotics in the vancomycin family. The vancomycin

group antibiotics are cell wall synthesis inhibitors in bacteria, but their mechanism of

action is quite different from that of the penicillins, which also inhibit cell wall growth

(9). The penicillins accomplish this task by binding to and inhibiting the penicillin-

binding proteins (PBPs), which are a group of enzymes required by the bacterium to

synthesize its cell wall (10). Thus, the penicillins are examples of enzyme-targeted

inhibitors, as shown in the middle panel of Figure 3. The vancomycin group antibiotics,

in contrast, complex with the peptidoglycan precursors needed to synthesize the bacterial

cell wall, but do not affect the PBP enzymes (9). Thus, they function analogously to the

substrate-targeted inhibitor portrayed in the bottom panel of Figure 3. The fundamentally

different strategies for inhibiting cell wall synthesis by the penicillins versus the

vancomycins explains why bacteria that have developed resistance to the penicillins may

still be susceptible to the vancomycins, and vice versa (9).

The use of antibodies themselves in this context is an obvious approach, and

several examples of antibodies are in clinical use (11). While therapeutic use of

antibodies has multiple advantages, including high specificity, long half-life, and lack of

resistance, antibodies also have serious disadvantages in that they can be immunogenic or

toxic, and they have limited ability to cross the blood-brain barrier or penetrate solid

tumors (11,12). Some of these problems may be solved through the use of human

recombinant antibodies rather than murine ones (11,12). In addition, several approaches

9

using small molecules rather than antibodies have been published (8). However, small

molecules have problems of their own, one of which is that they have fewer interactions

with the substrate as compared to polypeptides (8). Thus, several techniques have been

developed that attempt to remedy this by mimicking antibody-antigen interactions and

creating a multivalent inhibitor capable of recognizing multiple epitopes on the target

substrate protein (8).

One such technique involves using double synthetic peptides linked together to

form “molecular forceps” that were shown to be specific for the carboxy terminus of

hRas (8,13). These forceps consisted of two tetrapeptides connected by either a flexible

or a rigid linker (13). The peptides on each arm of the forceps were identical, increasing

the collective binding affinity of the conglomerate for the substrate analogous to an

antibody’s two identical epitope-binding sites (8). Large numbers of potential forceps

were synthesized and screened using combinatorial chemistry procedures previously

developed (13). In addition, active forceps were found to contain multiple Lys residues,

either as part of the linker or as part of the forceps sequence (13). Testing for binding of

other Ras sequences showed that the molecular forceps were specific for the particular

hRas sequence; other substrates continued to be farnesylated in the presence of the

forceps (13). It was found that binding resolution as sensitive as a single amino acid

could be obtained using molecular forceps, precluding the need for complex

macromolecular structures (13). However, the IC50 values for the best molecular forceps

were on the order of 100 µM, and improvement in their affinity is necessary before they

could be used in a therapeutic context (13). The authors also were concerned that their

forceps might not be capable of penetrating cell membranes (13).

10

A second group used a similar approach where multiple identical pentapeptides

were attached to library beads, which were then used to screen for a dopamine D2

receptor target sequence that was attached to magnetic beads (8,14). This approach had

the additional advantage that the bound inhibitor-substrate complexes could be easily

isolated using a magnet to pull them to the side of the tube, while unbound ligand beads

would simply settle to the bottom (8). This was the first example where small peptide

ligands were found that bind to a small peptide target molecule (14). Binding affinity

was better for this bead-to-bead technique, with the best Kd values on the order of 100

nM (14).

Molecular clamps (MCs) function slightly differently than these previously

reported strategies. They prevent substrates from being posttranslationally modified by

binding only to the specific substrate sequence upon which an enzyme acts, thus

preventing the active site of the enzyme from being able to interact with and catalyze that

substrate. Each MC consists of 12 amino acid residues connected by a peptidomimetic

linker. Functioning as a β–turn mimic, the MC itself undergoes a change in conformation

from a random coil when unbound, to form a small, stabile β–sheet upon binding the

target sequence. MCs can be designed to have affinity for any known substrate site of

posttranslational modification. The approach has the added advantage of only requiring

knowledge of the enzyme’s activity and the substrate sequence that the enzyme modifies,

but not the structure of the enzyme or its active site. A schematic of the mechanism of

MC activity is shown in Figure 1.4. The remainder of this chapter describes the

applications and considerations needed to use the MC strategy.

11

ksub1

ksub-1

molecular clamp

ksub2

ksub-2

substratesequence

A B C Figure 1.4: Schematic of the Binding of a Substrate by a MC. The substrate protein has multiple sequences of amino acids, but only one sequence that is specific for the MC (A). Binding of the MC to the substrate to form a three-stranded β-sheet occurs in a two-step process, where first one side of the clamp binds (B), followed by rapid binding of the second side (C).

1.3 Molecular Clamps: A Model

Classical enzyme kinetics systems are comprised of an enzyme (E) and a substrate

(S). The enzyme and substrate react to form an enzyme-substrate complex (E·S). The

E·S complex can either dissociate back to enzyme and substrate, or it can go on to form

the product (P). Km, the Michaelis-Menton constant, describes the propensity for E·S

complex dissociation to occur. If an enzyme-targeted inhibitor (I) is present in the

system, it will react with the enzyme to form an enzyme-inhibitor (E·I) complex, thereby

tying up some of the enzyme and giving rise to an E·I dissociation constant called Ki. All

of these relationships are shown in Figure 1.5.

The Michaelis-Menton equation describes the rate (v) at which an enzymatic

reaction will occur for a given concentration of substrate:

Vmax [S]

Km + [S]v =

The reaction rate also depends on the maximum rate at which the enzyme can catalyze

the reaction (Vmax), as well as on Km.

12

Ki

E + S E + P +I

E S

E I

Km

Figure 1.5: Dissociation Constants for a Classical System with an Enzyme-Targeted Inhibitor. Ki is the dissociation constant for the enzyme-inhibitor complex, and Km is the dissociation constant for the enzyme-substrate complex. See the text for the definitions of the remaining variables.

However, the presence of a substrate-targeted inhibitor like the MC complicates

this scenario considerably. This is because the enzyme can only act upon the free

substrate ([S]free), but the MC would sequester some of the substrate and make it

unavailable for binding. The enzyme binds to the remaining free substrate similarly to

the previous scenario. However, rather than binding to the enzyme like the classical

inhibitor does, the MC binds directly to the substrate instead, giving rise to a new

dissociation constant (KMC). These relationships are shown in Figure 1.6.

KMC

E + S E + P +

MC

E S

MC S

Km

Figure 1.6: Dissociation Constants for a MC Substrate-Targeted Inhibitor. Km is the dissociation constant for the enzyme-substrate complex, and KMC is the dissociation constant for the inhibitor-substrate complex. See the text for the definitions of the remaining variables.

13

The total amount of substrate is equal to the sum of the free substrate and the MC-

substrate complex:

[S]total = [S]free + [S•MC]

The concentration of free substrate depends on the concentration of MC, as well as on the

binding affinity of the MC for the substrate:

[S]free = ½[{([MC]total - [S]total + KMC)2 + 4KMC[S]total}0.5 - ([MC]total - [S]total + KMC)]

Since only the free substrate is available to be catalyzed by the enzyme, [S]free must be

substituted for [S] in the Michaelis-Menton equation:

Vmax [S]

Km + [S]freev =

free

Substituting the formula for [S]free into the Michaelis-Menton equation gives a revised

version of the Michaelis-Menton equation that can be used to study the kinetics of a

system comprised of an enzyme, a substrate, and a MC. This equation was then

manipulated to study the effects of the MC on enzyme kinetics.

There are four potential variables that can be studied using this model: the

concentration of substrate ([S]), the concentration of MC inhibitor ([MC]), the binding

constant of the enzyme for the substrate (Km), and the binding constant of the MC for the

substrate (KMC). Changing each of these variables affects the expected outcome of the

model in predictable and logical ways. In each graph, the reaction velocity was plotted as

the percentage of the actual velocity compared to the maximum velocity.

If the concentration of substrate is increased beyond a certain point, it is expected

that the ability of the MC to sequester the substrate would be overpowered; this

conclusion is supported by the model. See Figure 1.7A. When all other variables are

14

held constant, a high enough concentration of substrate (> 1 mM) will allow the enzyme

to reach its maximum velocity, assuming that the enzyme has a Km in the low micromolar

range and the MC is also present in a low micromolar concentration. The model suggests

that once the substrate and MC are present in approximately a 1:1 ratio, the usefulness of

the MC begins to diminish. This is true even in a scenario like the one shown in Figure

1.7A, where the MC has a much higher binding affinity for the substrate than the enzyme

does. In contrast, if the [MC] is at least an order of magnitude greater than the total [S],

then the MC will be effective at removing most or all of the free substrate from the

system and inhibiting formation of the E·S complex.

Effect of Substrate Concentration on Relative Enzyme Velocity

0

10

20

30

40

50

60

70

80

90

100

1E-151E-131E-110.000000001

0.00000010.000010.0010.1

Total [S]

% V

max

Figure 1.7A: Effect of Substrate Concentration on Relative Enzyme Velocity. A sufficient concentration of substrate will overpower the MC. [MC] = 5 µM, Km = 30 µM, KMC = 50 nM.

15

Effect of Molecular Clamp Concentration on Relative Enzyme Velocity

0

10

20

30

40

50

60

70

80

90

100

0.00E+00 1.00E-06 2.00E-06 3.00E-06 4.00E-06 5.00E-06 6.00E-06

[MC]

% V

max

Figure 1.7B: Effect of Molecular Clamp Concentration on Relative Enzyme Velocity. A sufficient concentration of MC will sequester the substrate and prevent catalysis. [S] = 1 nM, Km = 30 µM, KMC = 50 nM.

Increasing the concentration of MC was predicted to decrease the relative enzyme

velocity, and this was indeed found to be the case. If the substrate/MC ratio is

approximately 1:1, inhibition is relatively low, and the enzyme reaction velocity is

relatively close to normal. It is not until the [MC] gets about three orders of magnitude

higher than [S] that catalysis is essentially blocked. Again, this scenario assumes that the

MC has a higher affinity for the substrate than the enzyme does.

The activity of an enzyme that binds tightly to its substrates is considerably less

affected by the presence of the MC compared with an enzyme that does not, assuming

that excessive concentration of substrate is not present. This can clearly be seen by

comparing the velocities of several enzymes with different Km values while holding the

16

concentrations of substrate and MC and the KMC constant. If the Km of the enzyme were

sub-nanomolar, the MC would likely not be able to effectively compete with it, as noted

previously, unless its own binding constant was also sub-nanomolar. See Figure 1.7C.

Effect of Km Value on Relative Enzyme Velocity

0

10

20

30

40

50

60

70

80

90

100

1E-151E-131E-111E-090.00000010.000010.0010.1

Km

% V

man

Figure 1.7C: Effect of Km Value on Relative Enzyme Velocity. As the enzyme’s affinity for the substrate increases, it is able to out-compete the MC for the substrate. [MC] = 5 µM, [S] = 1 nM, KMC = 50 nM

Finally, the binding affinity of the MC also plays an important role, with the

effectiveness of the inhibition increasing as the affinity of the MC for the substrate

increased when all other factors were held constant. A molecular clamp present at high

nM concentrations with a KMC of 1 mM was not able to affect the rate of substrate

catalysis by the enzyme. Similarly, the MC with a KMC of 1 µM had only a minimal

effect. However, when the KMC dropped to 1 nM, catalysis was dramatically decreased

17

even at these low concentrations of inhibitor; a KMC of 1 pM essentially prevented

catalysis altogether, as shown in Figure 1.7D. These results were also in line with our

expectations.

Effect of KMC Value on Relative Enzyme Velocity

0

10

20

30

40

50

60

70

80

90

100

1.00E-12 1.00E-11 1.00E-10 1.00E-09 1.00E-08 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03

KMC (M)

% v

max

Figure 1.7D: Effect of KMC Value on Relative Enzyme Velocity. As the MC’s binding affinity improves, the MC becomes more effective at sequestering the substrate and preventing catalysis. [S] = 1 nM, Km = 30 µM, [MC] = 95.9 nM.

Based on these model results, it is reasonable to conclude that MCs would be

most useful in systems where the concentration of the substrate is low, the affinity of the

enzyme for the substrate is low (high Km), and the affinity of the MC for the substrate is

high (low KMC). It is also beneficial to raise the concentration of MC as high as is

feasible within pharmacokinetic and toxicological limits. Consideration must be given to

the concentration of the substrate that is naturally present in cells in order to select the

18

substrate candidates likeliest to be successfully inhibited using MCs. If the substrate to

be inhibited is normally present at low concentrations, then the amount of MC needed to

inhibit it will also be low, and within an achievable range. However, if the substrate is

present in large amounts in cells, then the concentration of MC needed to inhibit that

substrate may be prohibitively high or toxic. Thus, substrates present in large amounts

are not likely to be good candidates for inhibition by MCs.

Since MCs are essentially peptides with a small peptidomimetic region, one

possible problem that might limit their use is membrane impermeability. Proteins and

peptides are not generally able to penetrate cell membranes. However, it is also known

that there are various techniques that can be used to allow peptides to enter cells, and that

certain peptide sequences are themselves readily transferred across cell membranes along

with any proteins attached to them (12,16,17). The most important feature of these

peptides appears to be the inclusion of multiple basic residues in them, particularly

arginine (16,17). Such guanidine-containing sequences could easily be incorporated into

MCs as well, facilitating their uptake into cells. Successful penetration of fusion proteins

has already been demonstrated with a number of other peptide examples (17). Membrane

permeability can also be increased through palmitoylation, myristoylation, or other

addition of long-chain hydrophobic groups to the peptide (18).

There is also the issue of peptide stability to proteases. As with the issue of cell

membrane penetration, chemical modification of the MC by conjugating it with a

polymer is one possibility (19). Such conjugation has the added advantage of reducing

the immunogenicity of the peptide (19). It is also possible to increase the peptidomimetic

character of the MCs as needed if degradation of its peptide bonds is a major problem.

19

Finally, the biochemistry and pharmacology of MCs are not known. Since the

intracellular MC-bound substrate will not necessarily be degraded, such substrates could

function as undesired inhibitors of other metabolic pathways. It is also unknown whether

MCs will be toxic to the kidneys or other organs where they might accumulate. These

questions require answers before the MC strategy could be used in a clinical scenario.

In conclusion, we have considered the pros and cons of using MCs as drugs, as

well as established some guidelines to determine which enzymes would be the best

candidates for the use of MCs. We also performed some virtual kinetics studies on a

hypothetical system where we varied the MC concentration, the substrate concentration,

the enzyme-substrate dissociation constant, and the MC-substrate dissociation constant to

see how each of these changes would affect the usefulness of the MC. Future work on

this project would involve designing and synthesizing some actual MCs that target a

carefully selected substrate, as well as performing biological tests to increase our

knowledge of the behavior of MCs with respect to their ability to penetrate cells and their

pharmacological properties.

1.3 References 1. Garrett, R. H. and Grisham, C. M. Biochemistry, 2nd Ed.; Saunders College

Publishing: Fort Worth, 1999. 2. Branden, C. and Tooze, J. Introduction to Protein Structure, 2nd Ed.; Garland

Publishing: New York City, 1999. 3. Thorsten, B. (2003) Modulation of Protein-Protein Interactions with Small Organic

Molecules, Angewandte Chemie, Int. Ed., 42, 2462-2481. 4. Luthman, K. and Hacksell, U. Peptides and Peptidomimetics. In Textbook of Drug

Design and Discovery, 3rd Ed.; Krogsgaard-Larsen, P., Liljefors, T. and Madsen, U., Ed.; Taylor and Francis: New York, 2002; pp. 459-485.

20

5. Patel, J. B.; Huynh, C. K.; Handratta, V. D.; Gediya, L. K.; Brodie, A. M. H.; Goloubeva, O. G.; Clement, O. O.; Nanne, I. P.; Soprano, D. R.; and Njar, V. C. O. (2004) Novel Retinoic Acid Metabolism Blocking Agents Endowed with Multiple Biological Activities Are Efficient Growth Inhibitors of Human Breast and Prostate Cancer Cells in Vitro and a Human Breast Tumor Xenograft in Nude Mice, J. Med. Chem. ASAP article

6. Markwalder, J. A.; Arnone, M. R.; Benfield, P. A.; Boisclair, M.; Burton, C. R.;

Chang, C. H.; Cox, S. S.; Czerniak, P. M.; Dean, C. L.; Doleniak, D.; Grafstrom, R.; Harrison, B. A.; Kaltenbach, R. F.; Nugiel, D. A.; Rossi, K. A.; Sherk, S. R.; Sisk, L. M.; Stouten, P.; Trainor, G. L.; Worland, P.; and Seitz, S. P. (2004) Synthesis and Biological Evaluation of 1-Aryl-4, 5-dihydro-1H-pyrazolo[3,4-d]pyrimidin-4-one Inhibitors of Cyclin-Dependent Kinases. J. Med. Chem., 47, 5894-5911.

7. deSolms, S. J.; Ciccarone, T. M.; MacTough, S. C.; Shaw, A. W.; Buser, C. A.; Ellis-

Hutchings, M.; Fernandes, C.; Hamilton, K. A.; Huber, H. E.; Kohl, N. E.; Lobell, R. E.; Robinson, R. G.,; Tsou, N. N.; Walsh, E. S.; Graham, S. L.; Beese, L. S.; and Taylor, J. S. (2003) Dual Protein Farnesyltransferase-Geranylgeranyltransferase I Inhibitors as Potential Cancer Chemotherapeutic Agents, J. Med. Chem. 46, 2973-2984.

8. Kodadek, T. (2002) Inhibition of Proteolysis and Other Posttranslational

Modifications with Substrate-Targeted Inhibitors, Biopolymers, 66, 134-140. 9. Barna, J. C. J. and Williams, D. H. (1984) The Structure and Mode of Action of

Glycopeptide Antibiotics of the Vancomycin Group, Annu. Rev. Microbiol., 38, 339-357.

10. Waxman, D. J. and Strominger, J. L. (1983) Penicillin-Binding Proteins and the

Mechanism of Action of β-Lactam Antibiotics, Ann. Rev. Biochem., 52, 825-869. 11. Roskos, L. K.; Davis, C. G.; and Schwab, G. M. (2004) The Clinical Pharmacology

of Therapeutic Monoclonal Antibodies, Drug Development Research, 61, 108-120. 12. Sanz, L.; Blanco, B.; and Alverez-Vallina, L. (2004) Antibodies and Gene Therapy:

Teaching Old “Magic Bullets” New Tricks, Trends in Immunology, 25, 85-91. 13. Dong, D. L.; Liu, R.; Sherlock, R.; Wigler, M. H.; and Nestler, H. P. (1999)

Molecular forceps from combinatorial libraries prevent the farnesylation of Ras by binding to its carboxyl terminus, Chemistry & Biology, 6, 133-141.

14. Sasaki, S.; Takagi, M.; Tanaka, Y.; and Maeda, M. (1996) A New Application of a

Peptide Library to Identify Selective Interaction Between Small Peptides in an Attempt to Develop Recognition Molecules Toward Protein Surfaces, Tetrahedron Letters, 37, 85-88.

21

15. Segel, I.H., Enzyme Kinetics, John Wiley: New York, 1975. 16. Lundberg, P. and Langel, U. (2003) A brief introduction to cell-penetrating peptides,

J. Molecular Recognition, 16, 227-233. 17. Tung, C. H. and Weissleder, R. (2003) Arginine containing peptides as delivery

vectors, Advanced Drug Delivery Reviews, 55, 281-294. 18. Martin, M. L. and Busconi, L. (2000) Membrane localization of a rice calcium-

dependent protein kinase (CDPK) is mediated by myristoylation and palmitoylation, The Plant Journal, 24, 429-435.

19. Torchilin, V. P. and Lukyanov, A. N. (2003) Peptide and protein drug delivery to

and into tumors: challenges and solutions, Drug Discovery Today, 8, 259-286.

22

CHAPTER TWO

SYNTHESIS OF POLAR AND NONPOLAR HYDRAZINO AMINO ACIDS 2.1 Introduction 2.1.1 Hydrazines

Hydrazines are compounds that contain a nitrogen-nitrogen single bond. The first

hydrazine synthesis to be reported in the chemical literature was for phenylhydrazine in

1875 by Emil Fischer (1). Ordinary hydrazine was synthesized about a decade later by

Theodor Curtius (1). The physical, chemical, and biological properties of hydrazine and

its derivatives are well documented. These compounds are surprisingly common in

nature, as well as from synthetic sources such as agriculture, industry, military and space

vehicle fuels, and even medicinal drugs (1).

The use of over thirty hydrazines as drugs is especially interesting in light of the

fact that the majority of hydrazines studied thus far are known to be toxic, teratogenic,

and carcinogenic (1,2). However, it is not unprecedented for antineoplastic agents to also

be carcinogens. Mustard gas, alkylating compounds, cyclophosphamide, and several

other antineoplastic agents are also known to both induce and inhibit cancer (2). The

exact mechanism of action of the antineoplastic hydrazines is not known, but they are

postulated to be cytotoxic, interfere with glycolysis and gluconeogenesis, chelate

essential trace metals, inhibit mitosis, and/or degrade DNA (2).

23

Most hydrazine-containing drugs are heterocyclic compounds, and they exhibit a

wide range of activities (1). A sample of hydrazine-containing drugs includes several

anti-hypertensives (pheniprazine, hydralazine), anti-depressants (nialamide, marplan),

cancer chemotherapeutics (procarbazine and proresid, along with hydrazine itself),

Parkinson’s disease chemotherapeutics (carbidopa), anti-viral drugs (methiazone), anti-

tuberculotics (isoniazid, phtivazid), and antibiotics (sulfaphenazole, sulfamethizole),

among others (1,2).

Hydrazine derivatives have also been tested as inhibitors of enzymes that

metabolize α-amino acids (3,4). Some of these compounds have been shown to be

potent, specific inhibitors of such enzymes in bacteria, giving rise to a class of hydrazines

with antimicrobial activities. One well-studied example is a class of diaminopimelic acid

(DAP) analogs. These compounds were found to inhibit the enzyme meso-DAP D-

dehydrogenase in B. sphaericus and the enzyme LL-DAP epimerase in E. coli, thereby

preventing these bacteria from synthesizing the amino acid lysine and interfering with

their cell wall synthesis (4).

More recently, a series of hydrazine-containing peptides were reported that inhibit

the enzyme DAP aminotransferase in E. coli (5). The authors were able to observe good

in vivo antimicrobial activity from some of their compounds when they coupled their

hydrazines to normal amino acids to make short peptides. They speculated that the

presence of the natural amino acids would allow their compounds to be transported into

the cells as peptide prodrugs that would then be cleaved intracellularly via a peptide

transporter. However, they found that some of the transportable peptides were not active,

24

which suggests that the regular amino acids may actually play a role in the inhibitory

properties of the active compounds (5).

2.1.2 Synthesis of Hydrazines There are several methods to synthesize hydrazines that have been reported in the

literature. Before 1970, there were basically three methods that had been reported:

reduction of the hydrazone of an α-keto acid, functionalization of a carbonyl compound

(analogous to the Strecker synthesis), and reaction of hydrazine with an α-halo acid (6).

Reduction of hydrazones was achieved using sodium and mercury to give the free

hydrazine (7). This procedure is shown in Scheme 2.1. The Strecker-like synthesis,

shown in Scheme 2.2, involves the use of hydrazine and potassium cyanide to give the

cyanohydrazine. This can then be hydrolyzed to the hydrazino acid (6).

N

H

O

OHO

N

H

OHO

NHNH2Na(Hg)NH2NH2

N

H

NNH2

OO

NH2NH3

Scheme 2.1: Reduction of a Hydrazone to a Hydrazine (7). This procedure gives the racemate of the hydrazine.

O

HOO

NH2NH2

KCN

O

HONHNH2

CNO

HONHNH2

COOHH3O

Scheme 2.2: Strecker-Like Synthesis of a Hydrazino Acid (6). This procedure also gives the racemate of the hydrazine.

25

These two methods could not be used to form optically pure isomers; only the

reaction of hydrazine with an α-halo acid procedure was useful in synthesizing α-

hydrazino acids (6). This method involves reacting the α-hydrazino acid with hydrazine

via a bimolecular nucleophilic substitution reaction. Since the α-carbon is already

functionalized in this case, an optically active product can be obtained (8, 9). Two

examples of this reaction are shown in Scheme 2.3.

N

N

COOH

Cl

H

NH2NH2 N

N

COOH

HNHNH2

ZHNCOOH

BrNH2NH2

ZHNCOOH

NHNH2

Ref. 8

Ref. 9

Scheme 2.3: Reaction of Hydrazine with an α-Halo Acid (8,9). This procedure can be used to obtain an optically active hydrazine. Z denotes the Cbz protecting group.

Subsequently, it was discovered that hydantoic acids could be rearranged into

hydrazines using sodium hypochlorite (10). Substituted hydroxylamines had been used

previously to form nitrogen-nitrogen bonds, but since excess amine was required, the

purification of the products was difficult (10). The synthesis route involved reaction of

the amino acid with potassium cyanate to form the hydantoic acid. The hydantoic acid

26

was then reacted with sodium hypochlorite to form the hydrazine (10). However, the

yield of the rearrangement reaction was relatively low, between 20-30% (11). More

recently, it was found that using potassium hypochlorite instead of sodium hypochlorite

could raise the yield of this method to around 70% (11). The synthesis of hydrazines via

rearrangement of hydantoic acids is depicted in Scheme 2.4.

O

HONH2

COOHKCNO O

HONH

COOH

NH2O

NaOCl O

HONHNH2

COOH

Scheme 2.4: Synthesis of a Hydrazine via Rearrangement of a Hydantoic Acid (10). This procedure gives an optically pure product.

These authors had also attempted to make hydrazines via two other methods (10).

One procedure that they tried was to react an amino acid with hydroxylamine-O-sulfonic

acid, which theoretically should have given the hydrazino acid directly (10). However,

they were not able to separate the hydrazine product from the amine starting material; this

was a major problem because the parent amine had to be used in excess (10), as shown in

Scheme 2.5. They also tried N-aminating an optically active acetylated amino cyanide

(10). This approach was successful, giving the optically active acetylated hydrazino

cyanide, which could then be hydrolyzed to the hydrazino acid (10). This reaction is

shown in Scheme 2.6.

Another group found that they were able to synthesize optically active hydrazines

from 2-(((4-nitrobenzene)sulfonyl)oxy) (nosyloxy) esters via a bimolecular nucleophilic

substitution mechanism (12). The nosyloxy compounds were made by reacting 2-

27

hydroxy esters with nosyloxy chloride, followed with Boc-protected hydrazine, to give

the Boc-protected hydrazine (12). This procedure is shown in Scheme 2.7.

NH2OSO3NaO

HONH2

COOH X O

HONHNH2

COOH

Scheme 2.5: Attempt to Synthesize a Hydrazine Using a Hydroxylamine Derivative (10). The hydrazine product was obtained, but it could not be readily purified from the parent amine.

O

HONH

CN

O

NaHO

HON

CN

O

ClNH2

O

HONNH2

CN

O

HCl O

HONHNH2

COOH

Scheme 2.6: Synthesis of a Hydrazine Via N-Amination (10). This method gives an optically active product.

Several groups successfully synthesized optically active hydrazines via

asymmetric electrophilic amination (13-16). This is a complementary approach to

several of the previous procedures described where the hydrazine or its synthon served as

the nucleophile (14). In contrast, asymmetric electrophilic amination uses the carbon

skeleton as the nucleophile and the nitrogen-containing compound as the electrophile

28

(13-16). The electrophilic hydrazine synthon used was di-tert-butyl azodicarboxylate

(DBAD) (13,14); this particular azodicarboxylate was chosen because it was found that a

bulkier alkyl group improves the diasteroselectivity of the reaction (15). In addition,

DBAD is stable and commercially available, and the tert-butoxy group can be removed

from the hydrazine by conditions that will not affect the N-N hydrazine bond (16). Some

examples of hydrazine formation via electrophilic amination are shown in Scheme 2.8.

CO2CH3

OH

NsClTEA

CO2CH3

ONs

BocNHNH2

MeCN

CO2CH3

NHNHBoc

Scheme 2.7: Synthesis of a Hydrazine from a Nosyloxy Ester (12). The nosyloxy esters are made from chiral hydroxy esters.

Ph O

NMe2

OSiMe3

R

TiCl4

O

O

NO

O

N

DCM

Ph O

NMe2

O

R

NN

H

Boc

Boc

Ref. 13

NO

OOR

Bn

NO

OOR

BnBocNH-BocN

O

O

NO

O

N

LiNR21.

2.

Ref. 14 Scheme 2.8: Hydrazine Formation Via Asymmetric Electrophilic Amination (13, 14). This procedure is complementary to previous procedures that used hydrazine as the nucleophile.

29

A few groups have also synthesized hydrazines by reacting amino acids with

oxaziridines (17-20). The oxaziridines themselves are produced by reacting a Schiff base

with a peracid like oxone (18-20), or by electrophilic amination of a carbonyl compound

(17). Although N-amination is needed to produce hydrazines from oxaziridines,

oxaziridines are well-known aminating agents that can also be used to attach nitrogen

atoms to nucleophilic carbon atoms, sulfur atoms, oxygen atoms, and even phosphorus

atoms (17,19). In addition, the synthesis of oxaziridines with protecting groups that are

amenable to peptide synthetic methods made this methodology useful for the synthesis of

peptides and peptidomimetics (18). Oxaziridines have been synthesized with Boc (18-

20), Cbz (19), Fmoc (19), and other protecting groups on them. Some examples of

protected hydrazines synthesized by reacting an amine with an oxaziridine are shown in

Scheme 2.9.

NH

MeO

MeO

NHO

1.

2. 2N HCl N

MeO

MeO NH2 HClRef. 17

O

NHO

NCl3C Boc

O

NNHBoc

Ref. 20 Scheme 2.9: Synthesis of Hydrazines Using Oxaziridines (17,20). Several stable oxaziridines like the second example shown have been synthesized.

30

Recently, we published a method to synthesize hydrazines from amino acids by

the reduction of nitrosamines (21,22). The amino acid is first converted to a nitrosamine,

and this is then reduced using Zn/conc. HCl at –78 °C. The free hydrazine is reactive and

has a tendency to self-condense, so we protected the free β-nitrogen using Boc anhydride

(21,22). This procedure was very successful for the synthesis of the Boc-protected

hydrazines of several nonpolar amino acids (21,22).

NO

OH O N O

NO

ONO

NO

ONH2Zn/HCl

-78 C

Scheme 2.10: Synthesis of Hydrazines from Amino Acids Via Nitrosamines (21). After formation of the hydrazine, it was immediately reacted with Boc anhydride to protect it as the stable Boc derivative.

2.1.3 Use of Hydrazines in Peptidomimetics

Hydrazines are the key structural units of both azapeptide and hydrazinopeptide

peptidomimetics. Azapeptides are isosteres of peptides where the alpha carbon has been

replaced with a nitrogen atom. The first reported azapeptide, an analog of the peptide

hormone angiotensin II, is shown in Figure 2.1 (23). The authors found that this

compound was about 100-fold less active than angiotensin II, but its duration of activity

was twice as long (23). Another azapeptide with an unnatural N-cyclohexyl substituent

was found to be an excellent inhibitor of HIV protease (24). More recently, azapeptides

were studied as anti-cancer agents (25).

31

H2NN

NN N

NN

NOH

O

O

O O

O O

O

HOOCH

H

H

H H

HN

HN NH2 HO

N NH

Figure 2.1: Example of an Azapeptide (23). This is the first reported example in the literature of an azapeptide isostere. The azapeptide portion of the molecule is boxed.

H2NN

NN

NN

NN

NNH2

O

O

O

O

O O

O

O

H

H

H

H

H

H

H

H

NH2

O

HO

S

NBoc

NN

N

O O

H

H

H

NN

NOH

O

H O

H

H O

Ref. 27

Ref. 26 Ref. 28 Figure 2.2: Examples of Hydrazides (26-28). In a hydrazide, the hydrazine is coupled to the rest of the peptoid by its β-nitrogen. The hydrazines in each molecule are boxed.

32

Azapeptides are easier to synthesize in comparison with hydrazinopeptides. One

reason is that the chiral alpha carbon has been replaced with an achiral nitrogen atom. In

addition, there are fewer problems with regioselectivity in coupling the nitrogens to make

an azapeptide (26). However, hydrazinopeptides, where the normal α-amino acid is

replaced with an α-hydrazino amino acid, more closely mimic the three-dimensional

structure of real peptides. There are two groups of hydrazinopeptides that are classified

based on how the hydrazine subunits are coupled to the rest of the molecule (26).

Hydrazides are coupled by the β-nitrogen, whereas N-amino amides are coupled by the

α-nitrogen (26). Some examples of hydrazides and N-amino amides are shown in

Figures 2.2 and 2.3, respectively (26-29).

NBoc

N

O

NH2

N

O

H

NN

O

H O

NH2

OH

O

Ref. 26 Ref. 28

Figure 2.3: Examples of N-Amino Amides (26, 28). In N-amino amides, the hydrazine is coupled to the rest of the peptoid by its α-nitrogen. The hydrazines in each molecule are boxed.

33

2.1.4 Use of Hydrazines in β-Sheet Mimics

Proteins are ubiquitous in biology, and there is consequently a great deal of

interest in interfering with their interactions as well as mimicking their secondary

structures. The use of peptidomimetics is also considered to be a way to overcome the

disadvantages inherent in using peptides as drugs, such as poor bioavailability, enzymatic

degradation, antigenicity, and high cost of production (29). The hydrazines described in

this chapter can be used as structural subunits for conformationally restricted peptide

secondary structures like α-helices and β-sheets. This section briefly describes the

construction of β-sheet mimics, while α-helix mimics are described in Chapter Three.

Peptide ligands must adopt a specific conformation in order to interact with their

receptors. There is a tremendous loss of entropy that occurs when a flexible ligand in

solution is forced to interact in a single conformation with a protein such as a receptor

(30). In general, processes where a large number of torsional motions become restricted

upon binding are unfavorable due to entropic factors, even if they are enthalpically

favored (30). Thus, it would be advantageous to conformationally restrict such peptide

ligands in order to reduce entropic losses, as this would be expected to increase their

binding affinity.

To this end, our group has been working on the construction of two units that

mimic the extended conformation of a peptide by constraining the peptide as part of a

six-membered ring (22). The first diagram in Figure 2.4 shows a regular peptide that is

free to rotate around its single backbone bonds. There are five such dipeptide (DP)

bonds, numbered DP1 to DP5. Forming a six-membered dipeptide unit (DPU) ring as in

the middle diagram constrains two of these DP bonds and prevents any rotation around

34

them. Beyond this, using a hydrazine as in the right-hand azadipeptide (ADP) structure

forms two six-membered rings; now there are four constrained DP bonds. We are

currently working on the synthesis and testing of constrained ADP-type peptidomimetics

using our amino hydrazines.

Figure 2.4: Constraining the Extended Conformation of a Peptide to Mimic a Beta Sheet. (22) The first figure shows a regular peptide in an extended conformation. All of the dipeptide (DP) bonds are bolded. The middle figure shows a partially constrained dipeptide unit (DPU), where a six-membered ring prevents the peptide from rotating around DP bonds 2 and 3. The figure on the right shows a completely constrained azadipeptide (ADP) unit, where the peptide is prevented from rotating around DP bonds 2, 3, 4, and 5.

NN

O

N HO

HO

O

H2N

HO

N

OH

H

N

O

H

NHBn

O

HO

Ref. 31

NN

Bzl

CO2CH3

NO2O2N

Ref. 32 Figure 2.5: Some Constrained Hydrazine-Containing Peptidomimetics (31, 32). The ring size can be six-membered, as in the second example, or much larger.

There are some examples in the literature where hydrazines have been used to

construct conformationally restricted peptidomimetics. One is a mimic of the CD4 loop

O

NN

O

N

H O H H

R1

R2

DP2 DP3

O

NN

N O

N

H O H H

R1

R2

O

H

DP2 DP3 DP4 DP5DPU ADP

O

N

R1N

H O

N

H OR2H

H

DP1 DP2 DP3 DP4 DP5

HA

35

region that prevents the HIV glycoprotein gp120 from binding to CD4 on human

lymphocytes (31). Another group synthesized a six-membered ring with a hydrazine

(32). These two structures are shown in Figure 2.5.

2.2 Results and Discussion

This work builds on our group’s previous work to synthesize the hydrazines of

five nonpolar amino acids: valine, isoleucine, leucine, methionine, and phenylalanine

(21,22). Initially, the synthesis of compounds 6a-6e, shown in Scheme 2.12, was

attempted using our previously developed methodology (21,22), shown in Scheme 2.11.

However, we ran into trouble at several key points during the synthesis, such that two

new procedures were necessary. One change was to use a single-step reductive

amination rather than a two-step procedure to benzylate the amino groups. Second, the

successful synthesis of Lys compound 6b required the use of oxaziridine 10 rather than

the nitrosamine reduction pathway used for the other four amino acids reported here.

Our original procedure begins with the protection of the methyl ester of the amino

acid as the benzyl amino ester. This was done via a two-step process that involved

forming the imine using benzaldehyde and the amino acid methyl ester, isolating the

imine, and then reducing it to the benzyl amine using sodium borohydride. The final

compounds were made by converting each benzyl amine to the corresponding

nitrosamine using tert-butylnitrite, reducing the nitrosamine with zinc and concentrated

HCl at –78 °C, and protecting the β-nitrogen of the hydrazine by reacting it with Boc

anhydride neat (21,22). The original procedure is depicted in Scheme 2.11.

36

H2NO

O

R

HCl

O

H TEA MgSO4 THF1.

2. NaBH4 MeOHN

O

O

R

H

NO

O

R

NH

OO

O N O1.

2.

DCM

Zn/HCl MeOH -78 C

Boc2O

SR =

3.

Scheme 2.11: Procedure for the Synthesis of Five Protected Nonpolar Amino Acids (21, 22). This work was the starting basis for the synthesis of the four polar amino acid and alanine hydrazines described in this chapter.

The general synthetic procedure used to synthesize the analogous alanine, lysine,

serine, threonine, and tyrosine hydrazines is shown in Scheme 2.12. This synthesis

begins with the reductive amination of the free amine 1a-e using benzaldehyde,

triethylamine, and sodium cyanoborohydride to give compounds 2a-e. The benzyl-

protected amines were then treated with tert-butyl nitrite to give nitrosamine compounds

3a-e in high yield and purity. The nitrosamines were reduced using Zn/conc. HCl to give

the reactive intermediate hydrazines 4a-e, which were immediately reacted with Boc

anhydride to give the stable, Boc-protected hydrazines 5a-e. Finally, the benzyl

protecting group was removed from the α-nitrogen via hydrogenolysis to give

compounds 6a-e. Lysine compound 5b could not be obtained by this method, and was

synthesized by an alternative procedure as discussed below.

37

O N O

DCM94- 10 0%

NO

N

R

O

O

3a-e

con c. HCl

MeO H

Zn

NO

H

R

O

2a-e40- 90 %

NOH

H

R

O

1a-e

NO

NH2

R

O

4a-e

MeO H

O

H

Et3N

NaBH3CN

67- 10 0%38- 10 0%(0 % for Lys)

NO

N

R

OH Boc

5a-e

NOH

N

R

O

BocH

6a-e

H2

Pd/ C

MeO H

Boc2O

CH3CN

a b c d e

R = CH3

N

O

O

OO

O

Boc = O

O

Scheme 2.12: Synthesis of the Ala (6a), Lys (6b) Ser (6c), Thr (6d), and Tyr (6e) Boc-Protected Hydrazines from the Free Amines (1a-e). Compound 1a was commercially available, while compounds 1b-1e were synthesized as described below. Compound 5b could not be obtained by this procedure.

38

.

1a 2a

NHO

OCH3

H2NHClO

OCH3

2. NaBH4

90-95 %

1. H

O

X

Scheme 2.13: Unsuccessful Two-Step Method for Synthesizing Compound 2a. Unlike the other nonpolar amino acids published previously, the Ala compound 2a could not be obtained using this two-step method.

The two-step benzylation method shown in Scheme 2.11 was successfully used

for all five nonpolar amino acids attempted previously. Therefore, we began our

synthesis of the new hydrazine examples by attempting to protect the corresponding free

amines with a benzyl group using the two-step process (21,22). However, we ran into

some difficulties with the synthesis of several of the new benzyl amino acids. One

problem was that we were unable to obtain the alanine benzyl amine 2a using the two-

step procedure, even after multiple attempts, as depicted in Scheme 2.13. Instead, only

benzyl alcohol was isolated as evidenced by TLC and NMR.

Amino acid imine stability studies have shown that bulkier side chain amino acids

like Leu and Ile provide greater protection against imine hydrolysis when compared with

smaller groups (33). This would explain why the two-step procedure worked for the

larger nonpolar amino acids but not for Ala. However, since reduction of the imine

apparently occurs faster than hydrolysis during the one-step reduction, we were able to

successfully obtain 2a by that route. Further conversion of 2a to the protected hydrazine

6a was performed by the usual procedures.

39

Second, we had trouble benzyl-protecting the polar amino acids due to the

racemization of the alpha carbon, as depicted in Scheme 2.14. Initially, we did not know

about this problem because in most cases, racemization would produce a pair of

enantiomers, which would not be detectable. However, when the imine of threonine was

reduced to the benzyl amine, we were clearly able to detect two stereoisomers. This is

due to the presence of a second chiral center in the side chain of threonine that does not

exist in any of the other four amino acids. Thus, when the alpha carbon of threonine

racemized, we could clearly detect two diastereomers by TLC, and they could be

separated by column chromatography.

An examination of the NMRs of the two diastereomers shows that they are very

similar, but the groups of protons have slightly different coupling constants. This can be

seen in Figure 2.6. The aromatic protons in 2d, the (L)-isomer, show up as a multiplet

ranging from 7.22-7.38 ppm. In contrast, the aromatic protons of the (D)-isomer 13b are

more closely overlapping, with the multiplet for the aromatic protons ranging from 7.24-

7.35 ppm. The benzyl CH2 protons appear as a multiplet ranging from 3.92-3.99 ppm for

2d, versus 3.82-3.94 ppm for 13b. The alpha proton shows up as a doublet at 3.11 ppm

with a coupling constant of 3.5 Hz for 2d, while the same doublet shows up at 3.25 ppm

and has a coupling constant of 6.0 Hz for 13b. Finally, the methyl group on the side

chain of threonine shows up 1.24 ppm for 2d with a coupling constant of 6.2 Hz, versus

1.19 ppm for 13b, with the same coupling constant. It is clear that these two spectra are

of the same compound, with very slight differences due to the spatial orientation of the

groups that are attached to the two chiral centers in each molecule.

40

NOH

H O

O

1d

O

H

NaBH4

THF MeOH

NO

O

OH

13a

13b

NO

H O

O

18%2d36%

NO

H O

O

Scheme 2.14: Epimerization of Thr Due to 2-Step Benzylation Procedure. The (L) and (D) isomers 2d and 13 were formed in a 2:1 ratio due to the acidity of the imine intermediate’s alpha proton.

Our nitrosamine reduction procedure has several advantages, in that the starting

materials are all cheap and readily available, and that it can be used to synthesize

hydrazino amino acids from the corresponding amino acids without a loss of chirality at

the α-carbon. However, there are also some problems: the tert-butylnitrite reagent is

toxic and carcinogenic, the highly acidic conditions limit the types of side chain groups

and protecting groups that can be present, and there is a tendency to over-reduce the

nitrosamine back to the parent amine rather than stopping at the hydrazine. We had the

most trouble with the synthesis of the benzyl lysine hydrazine 5b. Although we were

able to selectively protect, deprotect, and react each of the two amino groups of lysine, as

well as form the nitrosamine 3b, we were not able to reduce 3b to the hydrazine 4b

cleanly, as depicted in Scheme 2.15. Rather, we obtained a multitude of decomposition

products, as evidenced by TLC and proton NMR.

41

Figure 2.6: NMRs of N-Benzyl-(L)-Threonine Methyl Ester (top) and N-Benzyl-(D)-Threonine Methyl Ester. The two diastereomers could clearly be differentiated by proton NMR, as well as by TLC.

N-Benzyl α-(L)-Threonine

p p m ( t 1 )1 . 02 . 03 . 04 . 05 . 06 . 07 . 08 . 0

7.35

7

7.31

1

7.28

2

7.26

0

7.23

3

4.01

7

4.00

3

3.99

2

3.97

7

3.96

8

3.95

3

3.94

3

3.92

3

3.71

8

3.63

0

3.57

7

3.18

3

3.16

8

3.13

1

2.17

3

1.25

4

1.22

9

1.13

1

1.07

9

2.10

3.001.13

1.04

0.93

4.71

9.203.40

NO

H O

O

N-Benzyl α-(D)-Threonine

p p m ( t1 )1 .02 .03 .04 .05 .06 .07 .08 .0

7.34

8

7.28

5

7.27

5

7.26

6

7.25

5

7.23

27.

329

7.31

6

7.30

9

3.94

1

3.88

7

3.85

4

3.83

0

3.80

5

3.67

0

3.61

6

3.25

8

3.23

4

3.72

8

2.04

9

1.20

2

1.17

7

1.12

6

1.00

4.40

1.241.152.821.26

1.19

3.458.70

NO

H O

O

42

3b

XZn/HCl

MeOHN

O

O

NO

N O

O

4bN

O

O

NO

NH2 O

Scheme 2.15: Unsuccessful Attempt to Reduce Lys Nitrosamine 3b. Unlike the other amino acids we have reported, compound 3b gave a complex mixture of products upon repeated attempts to reduce it using Zn/HCl.

This problem was solved by reacting lysine benzyl amine 2b with oxaziridine 10,

as shown in Scheme 2.16, to directly form the desired benzyl lysine hydrazine 5b in good

yield. Oxaziridine 10 was synthesized in our lab following a literature procedure (20,34),

and the yield and ease of using it to synthesize a Boc-protected hydrazine were an

improvement over our own nitrosamine method. 10 also has the advantage of being

much less carcinogenic and volatile compared with tert-butylnitrite. However, the

procedure to synthesize 10 is highly laborious and time-consuming, and it requires the

use of chloral, which is a controlled substance in the United States that is not readily

available for purchase from commercial sources.

Unlike the nonpolar amino acids, the polar amino acids were not readily available

as the hydrochloride salts of the methyl esters. Thus, we obtained compounds 11c–e as

the free acids with their amino groups protected by a Cbz. 11 was treated with potassium

carbonate and methyl iodide to give methyl ester 12, which was then deprotected on the

amine by hydrogenolysis to give starting material 1c–1e. This series of protection and

deprotection steps is depicted in Scheme 2.17.

43

Boc = O

O

XPd/C

H2AcOH

MeOH

40%MeOH

O

H

Et3N

NaBH3CN

1b

NOH

H O

N OO

H

CH3CN

67%

NO

H O

N OO

2b 10

N

O

Boc

Cl

Cl

Cl

5b

NO

N OH Boc

N

O

O

NOH

N OH Boc

N

O

O6b

Scheme 2.16: Procedure for Synthesizing Compound 6b (Lys). The nitrosamine of Lys 5b could not be reduced to hydrazine 6b using Zn/HCl, but compound 6b was successfully synthesized using the Boc-protected oxaziridine 10 according to literature procedure (20,34). The final hydrogenation step resulted in multiple products. The lysine starting material 1b was also formed from the free acid, but a few

additional steps were necessary in order to selectively protect the alpha and epsilon

nitrogens. We began with compound 11b, which was orthogonally protected by a Boc

group on the alpha nitrogen and a Cbz group on the epsilon nitrogen. This was reacted

with methyl iodide and potassium carbonate to form methyl ester 12b as before. The Cbz

44

group was then removed from the epsilon nitrogen by hydrogenolysis, and compound 7

was immediately reacted with phthalic anhydride 8 to form the lysine phthalimide 9 in

excellent yield. Finally, the alpha nitrogen of 9 was deprotected using a 1:1 mixture of

TFA and DCM to give starting material 1b. This procedure is shown in Scheme 2.18.

Cbz = O

O

12c-e

NOHZ

H

R

O

11c-e

NOZ

H

R

OK2CO3CH3I

DMF91-100%

H2/Pd50 psi

MeOH92-100%

NOH

H

R

O

1c-e

R =

c d e

OO

O

Scheme 2.17: Synthesis of Ser, Thr, and Tyr Methyl Esters (1c-e). The three tert-butyl ether-protected alcohol side chain amino esters were synthesized from the commercially available Cbz-protected acids in high yield.

45

Boc = O

O

Cbz = O

O

NOH

H O

N OO

H

1b9

NOBoc

H O

N OO

7

NOBoc

H O

NH H

H

Pd/C

H2AcOH

MeOH

12b

NOBoc

H O

NH Cbz

K2CO3

CH3IDMF

11b

NO

HBoc

H O

NH Cbz

TFA

DCMEt3N

OO

100%

O

O

O

8

Scheme 2.18: Procedure for Synthesizing Compound 1b (Lys). The amino group on the side chain of Lys was protected as a phthalimide in order to selectively form the hydrazine on the alpha amino group.

46

Boc = O

O

53- 100%

NO

N

R

OH Boc

5f-i

NOH

N

R

OBocH

6f-i

H2

Pd/C

MeOH

i

S

f g h

R =

Scheme 2.19: Deprotection of Nonpolar Amino Acid Hydrazines. Compounds 5f – 5i were synthesized as described previously by our group (21, 22). Hydrogenolysis of the benzylated hydrazines gave compounds 6f - 6i.

Finally, all of the benzyl-protected hydrazines that were synthesized previously

by our group were deprotected via hydrogenolysis. Previously, only the phenylalanine

example had been attempted and published (21). The procedure for synthesizing 6f–6i

was analogous to that described for deprotection of the polar amino acid hydrazines

above, and is shown in Scheme 2.19. It was difficult to deprotect the methionine benzyl

hydrazine 5i due to the sulfur poisoning the catalyst. We attempted to overcome this

problem by adding several drops of acetic acid, as well as by adding excess Pd/C. The

47

use of excess Pd/C was more helpful than adding acetic acid, but even with this change to

the procedure, the yield was still significantly lower than it was for the other amino acids.

The final lysine compound 6b could not be obtained; repeated attempts to hydrogenate

benzyl-protected hydrazine 5b gave either starting material or multiple decomposition

products as evidenced by TLC and NMR.

In conclusion, we have successfully deprotected the nonpolar hydrazino amino

acids that were published by our group previously (21). We have also synthesized the

hydrazine derivatives of five additional amino acids using modified procedures. Future

work on this project will entail continued efforts to successfully deprotect the lysine

benzyl hydrazine 5b, as well as the synthesis of the hydrazino amino acid derivatives of

the remaining ten natural amino acids.

2.3 Experimental Data

NO

O

OH

13a

Experimental Procedure to Synthesize Methyl tert-Butoxy-N-Benzylidene-(R)-

Threoninate (13a). A 100 mL Schlenk flask was placed in an ice bath with a stir bar and

flushed with argon. The methyl amino acid 1d (2.55 mmol) was dissolved in the flask in

a few mL of anhydrous THF, and this mixture was stirred for ten minutes at 0 °C. Then

benzaldehyde (5.10 mmol) and TEA (2.55 mmol) were added, along with a small scoop

of MgSO4 to keep the reaction dry. The reaction was allowed to come to room

48

temperature and stirred overnight. The MgSO4 was then removed by filtering the

reaction mixture through a powder funnel and rinsing the MgSO4 several times with

THF. The crude product 13a was obtained by evaporating the THF under reduced

pressure, and was carried on to the next step without purification due to the instability of

the imine to column chromatography. Rf = 0.56 (3:1 hexanes/EtOAc); 1H NMR (250

MHz, MeOD) δ 2.17 (s, 9H), 2.26 (d, J = 7.6 Hz, 1H), 2.39 (d, J = 6.3 Hz, 1H), 2.56

(quin, J = 6.3 Hz), 2.75 (quin, J = 6.2 Hz, 1H), 3.33 (s, 3H), 5.83-5.91 (m, 3H), 6.22-6.25

(m, 2H), 6.67 (s, 1H), 6.75 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 19.48, 20.44, 28.54,

28.72, 52.21, 69.26, 75.42, 80.13, 80.90, 129.44, 132.10, 136.60, 166.58, 166.77, 172.68

(The peak intensities were approximately the same for the both isomers, so the signals

reported are for both isomers); ESI-MS (0.1% AcOH in MeOH) 278.1 (M+H)+.

Experimental Procedure to Synthesize (S)- and (R)-Methyl tert-Butoxy-N-Benzyl-

Threoninate (2d and 13b). Compound 13a (2.55 mmol) was dissolved in several mL of

dry MeOH, flushed with argon, and stirred at room temperature. Sodium borohydride

(10.20 mmol) was added slowly in three portions over the course of ten minutes due to

vigorous bubbling in the reaction. The reaction was stirred at room temperature for two

hours, and then worked up with 1M NaOH. This was extracted three times with ethyl

acetate, and the three organic layers were combined and washed with brine, dried over

MgSO4, and evaporated under reduced pressure to give the benzyl amines 13b and 2d.

The crude products were chromatographed in 9:1 hexanes/ethyl acetate to give the pure

products in a 2:1 (S):(R) ratio in an overall yield of 54.5%.

49

NO

H O

O

2d

Methyl tert-Butoxy-N-Benzyl-(S)-Threoninate (2d). Rf = 0.56 (3:1 hexanes/EtOAc);

1H NMR (250 MHz, CDCl3) δ 1.12 (s, 9H), 1.24 (d, J = 6.2 Hz, 3H), 2.23 (broad, 1H),

3.12 (d, J = 3.5 Hz, 1H), 3.57-3.62 (m, 1H), 3.70 (s, 3H), 3.92-3.99 (m, 2H), 7.22-7.38

(m, 5H); 13C NMR (75 MHz, CDCl3) δ 20.57, 28.29, 51.52, 52.08, 65.87, 68.50, 73.65,

126.85, 128.20, 128.27, 140.09, 174.17; ESI-MS (0.1% AcOH in MeOH) 280.1 (M+H)+.

13b

NO

H O

O

Methyl tert-Butoxy-N-Benzyl-(R)-Threoninate (13b). Rf = 0.44 (3:1 hexanes/

EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.13 (s, 9H), 1.19 (d, J = 6.2 Hz, 3H), 2.36

(broad, 1H), 3.24 (d, J = 5.8 Hz, 1H), 3.64-3.67 (m, 1H), 3.73 (s, 3H), 3.84 (quin, J = 5.2

Hz, 1H), 3.90-3.94 (m, 1H), 7.24-7.26 (m, 1H), 7.31-7.35 (m, 4H); 13C NMR (100 MHz,

CDCl3) δ 19.83, 28.56, 51.79, 52.45, 66.38, 69.02, 74.31, 127.22, 128.50, 128.55, 139.80,

174.47; ESI-MS (0.1% AcOH in MeOH) 280.1 (M+H)+.

50

Experimental Procedure for Reductive Amination (2a-e). The commercially

available HCl salt of the alanine methyl ester 1a (10.7 mmol) or the previously

synthesized methyl ester 1b-e was dissolved in dry MeOH (50 mL), and cooled to 0 °C.

TEA (10.7 mmol) was added, followed by benzaldehyde (21.5 mmol). The reaction was

allowed to come to room temperature and stirred for 20 minutes. Sodium

cyanoborohydride (43.0 mmol) was then added to the reaction, which was stirred

overnight. The reaction mixture was quenched with 1M NaOH (30 mL), saturated with

NaCl, and extracted three times with ethyl acetate. Then, the combined ethyl acetate

layers were extracted with saturated NaCl solution, dried over MgSO4, and evaporated

under reduced pressure. The pure product 2a-e was isolated by column chromatography

(4:1 hexanes:ethyl acetate) in 40-90% yield.

2a

NHO

OCH3

Methyl N-Benzyl-(S)-Alaninate (2a). Rf = 0.31 (3:1 hexanes/EtOAc); 1H NMR (400

MHz, CDCl3) δ 1.69 (d, J=7.2 Hz, 3H), 3.69-3.67 (m, 1H), 3.78 (s, 3H), 4.16 (d, J=12.8

Hz, 1H), 4.25 (d, J=12.8 Hz, 1H), 7.40-7.35 (m, 3H), 7.64-7.62 (m, 2H); 13C NMR (100

MHz, CDCl3) δ 15.08, 48.92, 53.13, 53.30, 129.16, 129.61, 129.74, 130.86, 169.06; ESI-

MS (0.1% AcOH in MeOH) 194.1 (M+H)+.

51

NOBnz

H O

N OO

O

H

NaBH3CNMeOH

TEA

NOH

H O

N OO

H

9

NOBoc

H O

N OO

1b 2b

TFA

DCM

Methyl N’-Phthalyl-N-Benzyl-(S)-Lysinate (2b) via Methyl N’-Phthalyl-(S)-

Lysinate (1b). Boc-protected amine 9 (0.383 mmol) was dissolved in a large excess of

1:1 DCM/TFA. The reaction mixture was stirred under argon at room temperature for

ten minutes, and then the DCM/TFA was removed under reduced pressure to give crude

1b as the trifluoroacetate salt. 1b was immediately carried on as described above to give

2b in 66.7% yield. Rf = 0.13 (3:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.36-

1.44 (m, 2H), 1.60-1.65 (m, 4H), 1.89 (broad, 1H), 3.22 (t, J = 6.8 Hz, 1H), 3.56-3.70 (m,

6H), 3.73-3.78 (m, 1H), 7.18-7.26 (m, 5H), 7.66 (dd, J = 5.4 Hz, 2.8 Hz, 2H), 7.79 (dd, J

= 5.4 Hz, 2.8 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 23.29, 28.47, 33.15, 37.89, 51.90,

52.32, 60.62, 123.36, 123.38, 127.23, 128.42, 128.55, 128.57, 132.33, 134.08, 139.98,

168.55, 176.02; HR ESI-MS (0.1% AcOH in MeOH) 381.18056 (M + H)+

NO

H O

O2c

Methyl tert-Butoxy-N-Benzyl-(S)-Serinate (2c). Rf = 0.43 (3:1 hexanes/EtOAc); 1H

NMR (250 MHz, CDCl3) δ 1.15 (s, 9H), 2.23 (broad, 1H), 3.43-3.47 (m, 1H), 3.58-3.62

52

(m, 2H), 3.70-3.75 (m, 1H), 3.73 (s, 3H), 3.89-3.95 (m, 1H), 7.24-7.37 (m, 5H); 13C

NMR (75 MHz, CDCl3) δ 27.31, 51.65, 51.93, 61.00, 63.11, 73.14, 126.96, 128.25,

128.31, 139.76, 173.82; HR ESI-MS (0.1% AcOH in MeOH) 266.17505 (M+H)+.

NO

H O

O

2d

Methyl tert-Butoxy-N-Benzyl-(S)-Threoninate (2d). Rf = 0.61 (3:1 hexanes/EtOAc);

1H NMR (250 MHz, CDCl3) δ 1.13 (s, 9H), 1.24 (d, J = 6.2 Hz, 3H), 2.16 (broad, 1H),

3.12 (d, J = 3.6 Hz, 1H), 3.58-3.63 (m, 1H), 3.72 (s, 3H), 3.92-3.99 (m, 2H), 7.23-7.38

(m, 5H); 13C NMR (75 MHz, CDCl3) δ 20.51, 28.31, 51.49, 52.13, 66.01, 68.53, 73.66,

126.83, 128.19, 128.26, 140.13, 174.19; ESI-MS (0.1% AcOH in MeOH) 280.2 (M+H)+.

NO

H O

O

2e

Methyl tert-Butoxy-N-Benzyl-(S)-Tyrosinate (2e). Rf = 0.46 (3:1 hexanes/EtOAc);

1H NMR (250 MHz, CDCl3) δ 1.35 (s, 9H), 1.96 (broad, 1H), 2.93 (d, J = 7.0 Hz, 2 H),

3.52 (t, J = 7.0 Hz, 1H), 3.61-3.66 (m, 1H), 3.63 (s, 3H), 4.70 (s, 2H), 6.92 (d, J = 8.4 Hz,

2H), 7.07 (d, J = 8.4 Hz, 2H), 7.20-7.39 (m, 5H); 13C NMR (75 MHz, CDCl3) δ 28.80,

39.10, 51.64, 51.93, 62.09, 65.23, 78.36, 124.21, 126.95, 127.02, 127.58, 128.09, 128.33,

53

128.52, 129.59, 132.17, 139.46, 153.94, 175.17; HR ESI-MS (0.1% AcOH in MeOH)

342.20687 (M+H)+.

Experimental Procedure for Nitrosamine Formation (3a-e). The benzyl protected

amine 2 (1.14 mmol) was dissolved in DCM (10 mL), cooled to 0 °C, and tert-

butylnitrite (1.25 mmol) in DCM was added drop-wise from an addition funnel. The

reaction was refluxed for 4 hours, cooled to RT, and stirred overnight. The solvent was

evaporated under reduced pressure to give nitrosamine 3 as yellow oil in a 100 % yield.

Nitrosamines were typically pure and did not require chromatography.

NO

N O

O

3a

Methyl N-Benzyl-N-Nitroso-(S)-Alaninate (3a). Rf = 0.33 (3:1 hexanes/EtOAc); 1H

NMR (400 MHz, CDCl3) δ 1.69 (d, J = 7.2 Hz, 3H), 2.48 (d, J = 6.8 Hz, 3H), 3.57 (s,

3H), 3.67 (s, 3H), 4.54 (q, J = 7.6 Hz, 1H), 4.72 (d, J = 15.2 Hz, 1H), 4.94 (d, J = 15.2

Hz, 1H), 5.21 (q, J = 7.6 Hz, 1H), 5.39 (s, 2H), 7.08-7.09 (m, 2H), 7.27-7.30 (m, 3H),

7.37-7.39 (m, 5H); 13C NMR (100 MHz, CDCl3) δ 13.14, 16.47, 46.64, 52.35, 52.72,

52.96, 55.42, 59.96, 127.78, 127.98, 128.40, 128.61, 128.73, 128.91, 134.32, 134.61,

168.94, 170.87 (The peak intensities were approximately the same for the both isomers,

so the signals reported are for both isomers); ESI-MS (0.1% AcOH in MeOH) 223.1

(M+H)+.

54

NO

N O

N OO

O

3b

Methyl N’-Phthalyl-N-Benzyl-N-Nitroso-(S)-Lysinate (3b). Rf = 0.09, 0.15 (3:1

hexanes/EtOAc; two nitrosamine stereoisomers); 1H NMR (400 MHz, CDCl3) δ 1.17-

1.29 (m, 2H), 1.44-1.60 (m, 2H), 2.09-2.24 (m, 2H), 3.46 (s, 3H), 3.47-3.62 (m, 2H),

4.64-4.74 (m, 1H), 4.90-5.03 (m, 1H), 5.28-5.38 (m, 1H), 7.03-7.10 (m, 1H), 7.20-7.42

(m, 4H), 7.71-7.73 (m, 2H), 7.82-7.83 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 23.42,

28.02, 28.14, 29.50, 37.48, 46.94, 52.38, 52.80, 55.66, 56.15, 64.65, 123.37, 128.06,

128.16, 128.67, 128.80, 128.85, 128.94, 129.01, 129.84, 132.21, 134.16, 134.20, 134.48,

168.44, 170.40; (The spectrum is complex as with all of the nitrosamines; several of the

peaks split for both proton and carbon NMRs. No MS was taken of this compound due to

its toxicity and carcinogenicity.)

NO

N O

O

O

3c

Methyl tert-Butoxy-N-Benzyl-N-Nitroso-(S)-Serinate (3c). Rf = 0.43, 0.48 (3:1

hexanes/ EtOAc; two nitrosamine stereoisomers); 1H NMR (250 MHz, CDCl3) δ 1.05 (s,

55

9H), 1.10 (s, 9H), 3.58 (s, 3H), 3.71 (s, 3H), 3.73-3.77 (m, 2H), 3.92 (dd, J = 4.1 Hz, 9.7

Hz, 1H), 4.00-4.07 (m, 1H), 4.45-4.50 (m, 1H), 4.75-4.81 (m, 1H), 5.05-5.11 (m, 1H),

5.20-5.26 (m, 1H), 5.34-5.39 (m, 1H), 5.74-5.80 (m, 1H), 7.09-7.11 (m, 1H), 7.26-7.28

(m, 2H), 7.36-7.40 (m, 2H); 13C NMR (75 MHz, CDCl3) δ 26.98, 27.19, 47.91, 52.08,

52.50, 56.87, 58.65, 60.84, 65.07, 73.80, 127.27, 127.64, 128.32, 128.65, 134.71, 168.88

(The peak intensities were approximately the same for the both isomers, so the signals

reported are for both isomers. No MS was taken of this compound due to its toxicity and

carcinogenicity.)

NO

N O

O

O

3d

Methyl tert-Butoxy-N-Benzyl-N-Nitroso-(S)-Threoninate (3d). Rf = 0.37, 0.44 (3:1

hexanes/EtOAc; two nitrosamine stereoisomers); 1H NMR (250 MHz, CDCl3) δ 1.15 (s,

9H), 1.16 (d, J = 6.2 Hz, 3H), 1.20 (s, 9H), 3.50 (s, 3H), 3.64 (s, 3H), 4.35-4.36 (m, 1H),

4.53-4.58 (m, 1H), 4.75-4.81 (m, 1H), 5.14-5.25 (m, 2H), 5.79-5.85 (m, 1H), 7.17-7.47

(m, 5H); 13C NMR (75 MHz, CDCl3) δ 20.88, 22.01, 28.55, 28.69, 49.52, 51.84, 52.27,

57.53, 61.59, 66.05, 68.16, 70.41, 74.81, 127.13, 128.21, 128.40, 129.31, 134.73, 135.04,

168.82 (The peak intensities were approximately in a 2:1 ratio for the two isomers, so the

signals are reported for both isomers. No MS was taken of this compound due to its

toxicity and carcinogenicity.)

56

NO

N O

O

O

3e

Methyl tert-Butoxy-N-Benzyl-N-Nitroso-(S)-Tyrosinate (3e). Rf = 0.45 (3:1

hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.32 (s, 9H), 1.35 (s, 9H), 3.01-3.06 (m,

1H), 3.24-3.29 (m, 2H), 3.40-3.49 (m, 2H), 3.51 (s, 3H), 3.62 (s, 3H), 4.15-4.23 (m, 3H),

4.52-4.82 (m, 4H), 4.80-4.84 (m, 1H), 5.40-5.46 (m, 1H), 6.85 (d, J = 8.2 Hz, 2H), 6.95

(d, J = 8.4 Hz, 2H), 7.16-7.22 (m, 5H), 7.30-7.36 (m, 5H); 13C NMR (75 MHz, CDCl3) δ

28.78, 32.63, 35.86, 47.60, 52.22, 52.70, 56.98, 59.60, 65.11, 65.37, 78.58, 124.29,

124.61, 126.90, 127.45, 127.78, 128.38, 128.60, 128.68, 129.10, 129.54, 130.74, 131.61,

133.28, 133.50, 154.32, 167.59, 169.80 (The peak intensities were approximately the

same for the both isomers, so the signals reported are for both isomers. No MS was taken

of this compound due to its toxicity and carcinogenicity.)

Experimental Procedure for Boc-Protected α-Hydrazino Esters via Hydrazine

Formation (5a-e). Zinc was activated by washing with 10% HCl, dI water, methanol,

and diethyl ether. The nitrosamine 3 (1.60 mmol) was dissolved in dry MeOH (15 mL)

under argon in a 3-necked flask. A powder addition funnel was filled with activated Zn

(12.8 mmol) and fitted to one neck, while the other two necks were sealed with rubber

septa. The reaction system was then cooled to -78 °C, and activated Zn was slowly

added to the stirred suspension, followed by addition of conc. HCl (12.8 mmol) via

57

syringe. The reaction was then stirred under argon at -78 °C for 4 hours. To work up the

reaction, the excess Zn was filtered while the reaction mixture was still cold, and the

filtrate was treated with cold 6N KOH until it was strongly basic (pH 12-13). It was then

extracted with 3 equal portions of cold ethyl acetate. The combined ethyl acetate layers

were dried over Na2SO4, and evaporated under reduced pressure from a RT water bath, to

give the hydrazine. The neat oil will oligomerize, so the hydrazines were immediately

protected using Boc anhydride with no characterization. Melted tert-Boc2O (4.32 mmol)

was added directly to 4 (3.92 mmol) and they were stirred together for 30 min at room

temperature neat. After 30 min, a very small amount of acetonitrile was added to prevent

the reaction from solidifying, and the reaction mixture was stirred overnight. The

acetonitrile was removed under reduced pressure at RT, and the product 5 was isolated by

column chromatography (4:1 hexanes:ethyl acetate) in 38-100% yield.

NO

N OH Boc

5a

Methyl N-Amino-N-Benzyl-N’-tert-butoxycarbonyl-(S)-Alaninate (5a). Rf = 0.45

(3:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.32-1.44 (m, 12H), 3.62-3.64 (m,

1H), 3.71 (s, 3H), 3.95-4.05 (m, 2H), 6.60 (broad, 1H), 7.24-7.30 (m, 3H), 7.31-7.40 (m,

2H); 13C NMR (100 MHz, CDCl3) δ 16.21, 28.19, 51.55, 60.08, 60.65, 79.52, 127.41,

128.20, 129.27, 136.89, 155.37, 175.50; HR ESI-MS (0.1% AcOH in MeOH) 331.18090

(M + H)+

58

NOBnz

O

N OO

NH Boc

5b

Methyl N’-Phthalyl-N-Amino-N-Benzyl-N’’-tert-Butoxycarbonyl-(S)-Lysinate (5b).

Rf = 0.29 (3:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.33 (s, 9H), 1.56-1.60

(m, 2H), 1.70-1.78 (m, 4H), 3.35-3.39 (m, 1H), 3.63-3.65 (m, 2H), 3.88-3.91 (m, 1H),

3.98-4.00 (m, 1H), 6.70 (broad, 1H), 7.20-7.37 (m, 5H), 7.70 (d, J = 2.8 Hz, 2H), 7.82 (d,

J = 2.8 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 23.59, 28.33, 29.86, 37.96, 51.81, 61.32,

64.23, 79.94, 95.90, 123.37, 127.74, 128.46, 129.65, 130.09, 132.41, 134.05, 136.89,

155.37, 168.62, 174.38; HR ESI-MS (0.1% AcOH in MeOH) 496.24433 (M + H)+

NO

N OH Boc

O

5c

Methyl tert-Butoxy-N-Amino-N-Benzyl-N’-tert-Butoxycarbonyl-(S)-Serinate (5c).

Rf = 0.40 (3:1 hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.12 (s, 9H), 1.36 (s, 9H),

3.65-3.81 (m, 3H), 3.73 (s, 3H), 3.94-4.04 (m, 2H), 6.69 (broad, 1H), 7.23-7.31 (m, 3H),

7.39-7.41 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 27.25, 27.37, 27.47, 27.54, 28.47,

51.68, 61.24, 61.88, 65.33, 65.71, 73.47, 79.80, 127.65, 128.39, 128.59, 129.29, 129.60,

137.00, 172.49; HR ESI-MS (0.1% AcOH in MeOH) 403.22016 (M + Na)+

59

NO

N OH Boc

O

5d

Methyl tert-Butoxy-N-Amino-N-Benzyl-N’-tert-Butoxycarbonyl-(S)-Threoninate (5d). Rf

= 0.54 (3:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.10 (s, 9H), 1.27 (d, J = 6.3

Hz), 1.38 (s, 9H), 3.44-3.45 (m, 1H), 3.72 (s, 3H), 3.92-3.98 (m, 1H), 4.17-4.22 (m, 2H),

6.94 (broad, 1H), 7.22-7.42 (m, 5H); 13C NMR (100 MHz, CDCl3) δ 20.24, 28.32, 51.30,

61.62, 68.63, 69.93, 73.74, 79.10, 127.25, 128.09, 129.17, 137.49, 155.05, 172.37; HR

ESI-MS (0.1% AcOH in MeOH) 395.25477 (M + H)+

NO

N O

O

BocH

5e

Methyl tert-Butoxy-N-Amino-N-Benzyl-N’-tert-Butoxycarbonyl-(S)-Tyrosinate (5e). Rf =

0.46 (3:1 hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.34 (s, 9H), 1.38 (s, 1H),

2.89-3.01 (m, 1H), 3.06-3.12 (m, 1H), 3.62 (s, 3H), 3.67-3.71 (m, 1H), 3.92-3.99 (m,

2H), 6.64 (broad, 1H), 6.90 (d, J = 8.4 Hz, 2H), 7.14 (d, J = 8.4 Hz, 2H), 7.22-7.29 (m,

5H); 13C NMR (75 MHz, CDCl3) δ 28.18, 28.80, 35.55, 51.37, 66.73, 78.13, 79.76,

107.88, 123.93, 127.41, 128.08, 129.30, 129.66, 132.73, 136.47, 147.68, 151.93, 152.90,

153.81, 173.38; HR ESI-MS (0.1% AcOH in MeOH) 479.25147 (M + Na)+

60

General Procedure for Hydrogenolysis of Boc-Protected α-Hydrazino Esters (6a-

i). Compound 5 (1.32 mmol) was dissolved in MeOH (8 mL) and 10% Pd/C (0.280 g)

was added. The solution was hydrogenated at 45 psi H2 in a Paar Hydrogenator for 3

hours. The reaction mixture was filtered through a Celite cake, which was rinsed with

MeOH. Evaporation of the MeOH yielded 6 in 67-100% yield.

NOH

N OBocH

6a

Methyl N-Amino-N’-tert-Butoxycarbonyl-(S)-Alaninate (6a). Rf = 0.15 (3:1

hexanes/ EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.33 (d, J = 7.2 Hz, 3H), 1.45 (s, 9H),

3.71-3.75 (m, 4H), 4.17 (broad, 1H), 6.34 (broad, 1H); 13C NMR (100 MHz, CDCl3) δ

16.00, 28.30, 52.11, 58.47, 80.68, 156.36, 174.30; HR ESI-MS (0.1% AcOH in MeOH)

241.11595 (M + Na)+

NOH

N OBocH

O

6c

Methyl tert-Butoxy-N-Amino -N’-tert-Butoxycarbonyl-(S)-Serinate (6c). Rf = 0.36

(3:1 hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.14 (s, 9H), 1.44 (s, 9H), 2.69

61

(broad, 1H), 3.68-3.73 (m, 2H), 3.75 (s, 3H), 6.39 (broad, 1H), 6.64 (broad, 1H); 13C

NMR (100 MHz, CDCl3) δ 27.45, 28.52, 44.23, 51.66, 52.21, 61.47, 63.88, 68.01, 73.51,

73.62, 80.60, 156.25, 172.19; HR ESI-MS (0.1% AcOH in MeOH) 313.17306 (M + Na)+

NOH

N OBocH

O

6d

Methyl tert-Butoxy-N-Amino- N’-tert-Butoxycarbonyl-(S)-Threoninate (6d). Rf =

0.20 (3:1 hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.10 (s, 9H), 1.30 (d, J = 6.2

Hz, 3H), 1.42 (s, 9H), 3.43 (broad, 1H), 3.72 (s, 3H), 3.98-4.02 (m, 1H), 6.22 (broad,

1H); 13C NMR (75 MHz, CDCl3) δ 20.64, 28.26, 51.85, 67.49, 69.38, 73.87, 80.12,

155.93, 172.84; HR ESI-MS (0.1% AcOH in MeOH) 327.18926 (M + Na)+

NOH

N O

O

BocH

6e

Methyl tert-Butoxy-N-Amino -N’-tert-Butoxycarbonyl-(S)-Tyrosinate (6e). MP =

48-49 °C; Rf = 0.39 (3:1 hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.32 (s, 9H),

1.41 (s, 9H), 2.93-3.04 (m, 2H), 3.67 (s, 3H), 3.91 (t, J = 6.6 Hz, 1H), 4.18 (broad, 1H),

6.18 (broad, 1H), 6.91 (d, J= 8.4 Hz, 2H), 7.10 (d, J = 8.4 Hz, 2H); 13C NMR (75 MHz,

62

CDCl3) δ 28.22, 28.77, 36.42, 51.93, 64.25, 80.60, 124.23, 129.53, 131.17, 154.18,

156.08, 173.15; HR ESI-MS (0.1% AcOH in MeOH) 389.20463 (M + Na)+

NOH

N OBocH

6f

Methyl N-Amino-N’-tert-Butoxycarbonyl-(S)-Valinate (6f). Rf = 0.28 (3:1 hexanes/

EtOAc); 1H NMR (250 MHz, CDCl3) δ 0.97-1.02 (m, 6H), 1.44 (s, 9H), 2.01-2.06 (m, 1

H), 3.48 (d, J = 4.4 Hz, 1 H), 3.75 (s, 3H), 4.17 (broad, 1H), 6.19 (broad, 1H); 13C NMR

(75 MHz, CDCl3) δ 18.93, 19.40, 28.69, 30.34, 52.30, 69.74, 81.09, 156.70, 173.95; HR

ESI-MS (0.1% AcOH in MeOH) 269.14784 (M + Na)+

NOH

N OBocH

6g

Methyl N-Amino-N’-tert-Butoxycarbonyl-(S)-Isoleucinate (6g). Rf = 0.37 (3:1

hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 0.76-0.86 (m, 6H), 1.13-1.18 (m, 1H),

1.32 (s, 9H), 1.37-1.44 (m, 1H), 1.66-1.68 (m, 1H), 3.41 (d, J = 4.9 Hz, 1H), 3.62 (s, 3H),

4.09 (broad, 1H), 6.48 (broad, 1H); 13C NMR (75 MHz, CDCl3) δ 11.86, 12.09, 15.18,

15.77, 25.95, 26.43, 28.52, 34.86, 36.97, 51.90, 52.05, 67.80, 68.46, 80.52, 156.73,

173.80, 174.20; HR ESI-MS (0.1% AcOH in MeOH) 283.16317 (M + Na)+

63

NOH

N OBocH

6h

Methyl N-Amino-N’-tert-Butoxycarbonyl-(S)-Leucinate (6h). Rf = 0.40 (3:1

hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 0.85 (d, J = 7.3 Hz, 6H), 1.35 (s, 9H),

1.38-1.44 (m, 1H), 1.69-1.75 (m, 1H), 3.57 (m, 1H), 3.64 (s, 3H), 4.08 (broad, 1H), 6.48

(broad, 1H); 13C NMR (75 MHz, CDCl3) δ 22.44, 23.24, 25.22, 28.64, 39.99, 52.36,

62.26, 80.79, 156.64, 174.95; HR ESI-MS (0.1% AcOH in MeOH) 283.16326 (M + Na)+

NOH

N OBocH

S6i

Methyl N-Amino-N’-tert-Butoxycarbonyl-(S)-Methioninate (6i). Rf = 0.14 (3:1

hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.44 (s, 9H), 2.00-2.08 (m, 2H), 2.09 (s,

3H), 2.65 (t, J = 7.8 Hz, 2H,) 3.68-3.72 (m, 1H), 3.75 (s, 3H), 4.25 (broad, 1H), 6.25

(broad, 1H); 13C NMR (125 MHz, CDCl3) δ 15.64, 28.49, 30.06, 30.58, 52.45, 62.42,

80.93, 156.51, 173.79; HR ESI-MS (0.1% AcOH in MeOH) 301.12002 (M + Na)+

64

9

NOBoc

H O

N OO

Experimental Procedure to Synthesize Methyl N’-Phthalyl-N-tert-Butoxycarbonyl-

(S)-Lysinate (9). Compound 12b (2.54 mmol) was dissolved in ~10 mL methanol, to

which was added a few drops of glacial AcOH and 0.1 g 10% Pd/C. The mixture was

shaken under H2 at 50 psi for one hour, filtered through a Celite cake to remove the Pd/C,

and evaporated under reduced pressure to give the free amine 7. Deprotected amine 7

(2.54 mmol) was immediately dissolved in 20 mL dry dioxane in a three-necked flask.

Phthalic anhydride 8 (2.80 mmol) and TEA (2.80 mmol) were added, the reaction was

fitted with a condenser and drying tube, and it was then refluxed overnight. The reaction

was cooled after 16 hours and diluted with EtOAc. Then the organic layer was washed

with dI water (3x) and brine (2x), dried over MgSO4, and evaporated under reduced

pressure. The pure product 9 was isolated in 100% yield with no purification needed. Rf

= 0.15 (3:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.43-1.44 (m, 11H), 1.65-

1.72 (m, 3H), 1.82-1.85 (m, 1H), 3.66-3.70 (m, 2H), 3.73 (s, 3H), 4.28-4.29 (m, 1H), 5.08

(broad, 1H), 7.72 (dd, J = 5.4 Hz, 3.2 Hz, 2H), 7.84 (dd, J = 5.2 Hz, 3.2 Hz, 2H); 13C

NMR (100 MHz, CDCl3) δ 22.75, 28.28, 28.48, 32.34, 37.66, 52.44, 53.42, 67.25, 80.01,

123.39, 132.28, 134.07, 155.56, 168.57, 173.41; HR ESI-MS (0.1% AcOH in MeOH)

413.16832 (M + Na)+

65

Experimental Procedure for Methyl Esterification (12b-12e). The commercially

available amino acid 11b-11e (3.49 mmol) was dissolved in dry DMF (20 mL) and

cooled to 0 °C. Anhydrous potassium carbonate (3.84 mmol) was added, and the reaction

was flushed with argon and stirred at 0 °C for 10 minutes. Methyl iodide (6.98 mmol)

was then added drop-wise via syringe, and the reaction was allowed to warm up to room

temperature. After stirring for 4.5 hours, the reaction was diluted with ethyl acetate (50

mL) and washed twice with saturated sodium bicarbonate solution. The organic layer

was then washed once with saturated NaCl solution, dried over MgSO4, and evaporated

under reduced pressure. The pure product 12c-12e was isolated by filtering through a

small pad of silica using 3:1 hexanes:ethyl acetate in 91-100% yield.

12b

NOBoc

H O

NH Cbz

Methyl N-tert-Butoxycarbonyl-N’-Carboxybenzyl-(S)-Lysinate (12b). Rf = 0.29

(3:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.36-1.41 (m, 2H), 1.43 (s, 9H),

1.49-1.56 (m, 2H), 1.60-1.64 (m, 1H), 1.76-1.83 (m, 1H), 3.16-3.21 (m, 2H), 3.73 (s,

3H), 4.24-4.29 (m, 1H), 4.87 (broad, 1H), 5.09 (s, 2H), 5.12 (broad, 1H), 7.31-7.36 (m,

5H); 13C NMR (100 MHz, CDCl3) δ 14.41, 22.61, 28.52, 29.59, 30.54, 32.58, 40.84,

52.49, 53.38, 66.85, 80.14, 128.31, 128.34, 128.72, 136.83, 156.70, 171.36, 173.48; ESI-

MS (0.1% AcOH in MeOH) 417.1 (M + Na)+

66

NOZ

H O

O

12c

Methyl tert-Butoxy-N-Carboxybenzyl-(S)-Serinate (12c). Rf = 0.35 (3:1 hexanes/

EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.12 (s, 9H), 3.57 (dd, J = 3.1 Hz, 9.0 Hz, 1H),

3.74 (s, 3H), 3.81 (dd, J = 2.9 Hz, 9.1 Hz, 1H), 4.45-4.49 (m, 1H), 5.13 (s, 2H), 5.68 (d, J

= 8.8 Hz, 1H), 7.31-7.37 (m, 5H); 13C NMR (75 MHz, CDCl3) δ 27.21, 52.33, 54.59,

61.94, 66.94, 73.38, 128.12, 128.49, 136.25, 156.11, 171.11; HR ESI-MS (0.1% AcOH in

MeOH) 310.16486 (M + H)+

NOZ

H O

O

12d

Methyl tert-Butoxy-N-Carboxybenzyl-(S)-Threoninate (12d). Rf = 0.48 (3:1

hexanes/EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.10 (s, 9 H), 1.22 (d, J = 6.3 Hz, 3H),

3.72 (s, 3H), 4.22-4.26 (m, 2H), 5.14 (s, 2H), 5.60 (d, J = 9.6 Hz, 1H), 7.32-7.41 (m, 5H);

13C NMR (75 MHz, CDCl3) δ 21.00, 28.24, 52.20, 59.80, 67.01, 67.28, 74.00, 128.13,

128.52, 136.25, 156.76, 171.61; HR ESI-MS (0.1% AcOH in MeOH) 346.16278 (M +

Na)+

67

12e

NOZ

H O

O

Methyl tert-Butoxy-N-Carboxybenzyl-(S)-Tyrosinate (12e). Rf = 0.36 (3:1 hexanes/

EtOAc); 1H NMR (250 MHz, CDCl3) δ 1.33 (s, 9H), 3.07 (d, J = 5.9 Hz, 2H), 3.70 (s,

3H), 4.64 (q, J = 8.1 Hz, 1H), 5.10 (s, 2H), 5.28 (d, J = 8.2 Hz, 1H), 6.90 (d, J = 8.4 Hz,

2H), 7.00 (d, J = 8.5 Hz, 2H), 7.32-7.35 (m, 5H); 13C NMR (75 MHz, CDCl3) δ 28.79,

37.56, 52.26, 54.83, 66.94, 78.43, 124.25, 128.11, 128.19, 128.52, 129.67, 130.43,

136.16, 154.39, 155.58, 172.02; HR ESI-MS (0.1% AcOH in MeOH) 408.17881 (M +

Na)+

Experimental Procedure for Cbz Deprotection (1c-1e). The methyl ester 12c-12e

(3.04 mmol) was dissolved in MeOH (10 mL), and the mixture was slightly acidified by

adding 10 drops of glacial acetic acid. Then 10% Pd/C (0.4 g) was added, and the

reaction was placed under a H2 environment at 50 psi on a Paar hydrogenator. After

shaking for 1 hour, the product was filtered through a small pad of Celite and evaporated

under reduced pressure. No purification was necessary to obtain the pure product 1c-1e

in 92-100% yield.

68

NOH

H O

O

1c

Methyl tert-Butoxy-(S)-Serinate (1c). Rf = 0.10 (100% EtOAc); 1H NMR (250 MHz,

CDCl3) δ 0.96 (s, 9H), 3.41-3.48 (m, 3H), 3.53 (s, 3H), 3.75 (broad, 2H); 13C NMR (75

MHz, CDCl3) δ 27.09, 51.81, 54.47, 63.00, 72.91, 173.67; HR ESI-MS (0.1% AcOH in

MeOH) 176.12745 (M+H)+.

NOH

H O

O

1d

Methyl tert-Butoxy-(S)-Threoninate (1d). Rf = 0.16 (100% EtOAc); 1H NMR (250

MHz, CDCl3) δ 1.11 (s, 9H), 1.22 (d, J = 6.2 Hz, 3H), 3.32-3.33 (m, 1H), 3.53 (broad,

2H), 3.70 (s, 3H), 3.99-4.04 (m, 1H); 13C NMR (75 MHz, CDCl3) δ 20.82, 28.36, 51.93,

60.08, 68.02, 73.72, 174.58; HR ESI-MS (0.1% AcOH in MeOH) 190.14360 (M+H)+.

1e

NOH

H O

O

69

Methyl tert-Butoxy-(S)-Tyrosinate (1e). Rf = 0.26 (100% EtOAc); 1H NMR (250

MHz, CDCl3) δ 1.32 (s, 9H), 2.33 (broad, 2H), 2.80-2.88 (m, 1H), 3.04 (dd, J = 5.3 Hz,

13.6 Hz, 1H), 3.70 (s, 3H), 3.72-3.75 (m, 1H), 6.92 (d, J = 8.4 Hz, 2H), 7.07 (d, J = 8.4

Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 28.77, 40.25, 51.96, 55.63, 78.37, 124.27, 129.64,

131.75, 154.16, 175.29; HR ESI-MS (0.1% AcOH in MeOH) 252.15986 (M+H)+.

2.4: References 1. Toth, B. (2000) A Review of the Natural Occurrence, Synthetic Production, and Use

of Carcinogenic Hydrazines and Related Chemicals, In Vivo, 14, 299-319. 2. Toth, B. (1996) A Review of the Antineoplastic Action of Certain Hydrazines and

Hydrazine-Containing Natural Products, In Vivo, 10, 65-96. 3. Scaman, C. H., Palcic, M. M., McPhalen, C., Gore, M. P., Lam, L. K. P., Vederas, J.

C. (1991) Inhibition of Cytoplasmic Aspartate Aminotransferase from Porcine Heart by R and S Isomers of Aminooxysuccinate and Hydrazinosuccinate, Journal of Biological Chemistry, 266, 5525-5533.

4. Lister, L. K. P., Arnold, L. D., Kalantar, T. H., Kelland, J. G., Lane-Bell, P. M.,

Palcic, M. M., Pickard, M. A., Vederas, J. C. (1988) Analogs of Diaminopimelic Acid as Inhibitors of meso-Diaminopimelate Dehydrogenase and LL-Diaminopimelate Epimerase, Journal of Biological Chemistry, 263, 11814-11819.

5. Cox, R. J., Jenkins, H., Schouten, J. A., Stentiford, R. A., Wareing, K. J. (2000)

Synthesis and in vitro enzyme activity of peptide derivatives of bacterial cell wall biosynthesis inhibitors, Journal of the Chemical Society, Perkin Transactions 1, 2023-2036.

6. Kerady, S., Ly, M. G., Pines, S. H., Sletzinger, M. (1971) Synthesis of D- and L-α-

(3,4-Dihydroxybenzyl)-α-hydrazinopropionic Acid via Resolution, Journal of Organic Chemistry, 36, 1946-1948.

7. Glamkowski, E. J., Gal, G., Sletzinger, M., Porter, C. C., Watson, L. S. (1967) A

New Class of Potent Decarboxylase Inhibitors. β-(3-Indolyl)-α-hydrazinopropionic Acids, Journal of Medicinal Chemistry, 10, 852-855.

8. Sletzinger, M., Firestone, R. A., Reinhold, D. F., Rooney, C. S. (1968) The α-

Hydrazino Analog of Histidine, Journal of Medicinal Chemistry, 11, 261-263.

70

9. Sawayama, T., Kinugasa, H., Nishimura, H. (1976) Syntheses of Ornithine Decarboxylase Inhibitors: D- and DL-α-Hydrazinoornithine, Chemical and Pharmaceutical Bulletin, 24, 326-329.

10. Kerady, S., Ly, M. G., Pines, S. H., Sletzinger, M. (1971) Synthesis of L-α-(3,4-

Dihydroxybenzyl)-α-hydrazinopropionic Acid from Optically Active Precursors by N-Homologization, Journal of Organic Chemistry, 36, 1949-1951.

11. Viret, J., Gabard, J., Collet, A. (1987) Practical Synthesis of Optically Active α-

Hydrazino Acids from α-Amino Acids, Tetrahedron, 43, 891-894. 12. Hoffman, R. B., Kim, H. O. (1990) The preparation of 2-Hydrazinyl Esters in High

Optical Purity from 2-Sulfonyloxy Esters, Tetrahedron Letters, 31, 2953-2956. 13. Gennari, C., Colombo, L., Bertolini, G. (1986) Asymmetric Electrophilic

Amination: Synthesis of α-Amino and α-Hydrazino Acids with High Optical Purity, Journal of the American Chemical Society, 108, 6394-6395.

14. Evans, D. A., Britton, T. C., Dorow, R. L., Dellaria, J. F. (1986) Stereoselective

Amination of Chiral Enolates. A New Approach to the Asymmetric Synthesis of a-Hydrazino and a-Amino Acid Derivatives, Journal of the American Chemical Society, 108, 6395-6397.

15. Trimble, L. A., Vederas, J. C. (1986) Amination of Chiral Enolates by Dialkyl

Azodiformates. Synthesis of α-Hydrazino Acids and α-Amino Acids, Journal of the American Chemical Society, 108, 6397-6399.

16. Evans, D. A., Britton, T. C., Dorow, R. L., Dellaria, J. F. (1988) The Asymmetric

Synthesis of α-Amino and α-Hydrazino Acid Derivatives via the Stereoselective Amination of Chiral Enolates with Azodicarboxylate Esters, Tetrahedron, 44, 5525-5540.

17. Andrae, S., Schmitz, E. (1991) Electrophilic Aminations with Oxaziridines,

Synthesis, 327-341. 18. Vidal, J., Guy, L., Sterin, S., Collet, A. (1993) Electrophilic Amination: Preparation

and Use of N-Boc-3-(4-cyanophenyl)oxaziridine, a New Reagent that Transfers a N-Boc Group to N- and C-Nucleophiles, Journal of Organic Chemistry, 58, 4791-4793.

19. Vidal, J., Damestoy, S., Guy, L., Hannachi, J. C., Aubry, A., Collet, A. (1997) N-

Alkyloxycarbonyl-3-aryloxaziridines: Their Preparation, Structure, and Utilization as Electrophilic Amination Reagents, Chemistry: A European Journal, 3, 1691-1709.

71

20. Vidal, J., Hannachi, J. C., Hourdin, G., Mulatier, J. C., Collet, A. (1998) N-Boc-3-trichloromethyloxaziridine: a new powerful reagent for electrophilic amination, Tetrahedron Letters, 39, 8845-8848.

21. Oguz, U., Guilbeau, G., McLaughlin, L. (2002) A Facile Stereospecific Synthesis of

α-Hydrazino Esters, Tetrahedron Letters, 43, 2873-2875. 22. Oguz, U. (2003) Design and Synthesis of Constrained Dipeptide Units for Use as β-

Sheet Promoters, Dissertation, Louisiana State University. 23. Hess, H. J., Moreland, W. T., Laubach, G. D. (1963) N-[2-Isopropyl-3-(L-aspartyl-

L-arinyl)-carbazoyl]-L-tyrosyl-L-valyl-L-histidyl-L-prolyl-L-phenylalanine, an Isostere of Bovine Angiotensin II, Journal of the American Chemical Society, 85, 4040-4041.

24. Fasler, A., Bold, G., Capraro, H. G., Cozens, R., Mestan, J., Poncioni, B., Rosel, J.,

Tintelnot-Blomley, M., Lang, M. (1996) Aza-Peptide Analogs as Potent Human Immunodeficiency Virus Type-1 Protease Inhibitors with Oral Bioavailability, J. Med. Chem., 39, 3203-3216.

25. Bouget, K., Aubin, S., Delcros, J. G., Arlot-Bonemains, Y., Baudy-Floc’h, M. (2003)

Hydrazino-Aza and N-Azapeptoids with Therapeutic Potential as Anticancer Agents, Bioorganic & Medicinal Chemistry, 11, 4881-4889.

26. Lecoq, A., Marraud, M., Aubry, A. (1991) Hydrazino and N-Amino Peptides.

Chemical and Structural Aspects, Tetrahedron Letters, 32, 2765-2768. 27. Niedrich, H. (1967) Synthese des Eledoisin-(4-11)-Octapeptides Lys-Asp(NH2)-Ala-

Phe-Ile-Gly-Leu-Met-NH2 und seines Heterologen mit Hydrazinoessigsaure statt Glycin, Chemische Berichte, 100, 3273-3282.

28. Killian, J. A., Van Cleve, M. D., Shayo, Y. F., Hecht, S. M. (1998) Ribosome-

Mediated Incorporation of Hydrazinophenylalanine into Modified Peptide and Protein Analogs, Journal of the American Chemical Society, 120, 3032-3042.

29. Guy, L., Vidal, J., Collet, A. (1998) Design and Synthesis of Hydrazinopeptides and

Their Evaluation as Human Leukocyte Elastase Inhibitors J. Med. Chem., 41, 4833-4843.

30. Chen, S., Chrusciel, R. A., Nakanishi, H., Raktabutr, A., Johnson, M. E., Sato, A.,

Weiner, D., Hoxie, J., Saragovi, H. U., Greene, M. I., Kahn, M. (1992) Design and synthesis of a CD4 β-turn mimetic that inhibits human immunodeficiency virus envelope glycoprotein gp120 binding and infection of human lymphocytes, Proc. Natl. Acad. Sci., 89, 5872-5876.

72

31. Mammen, M., Shakhnovich, E. I., Whitesides, G. M. (1998) Using a Convenient, Quantitative Model for Torsional Entropy to Establish Qualitative Trends for Molecular Processes That Restrict Conformational Freedom, Journal of Organic Chemistry, 63, 3168-3175.

32. Hannachi, J. C., Vidal, J., Mulatier, J. C., Collet, A., (2004) Electrophilic Amination

of Amino Acids with N-Boc-oxaziridines: Efficient Preparation of N-Orthogonally Diprotected Hydrazino Acids and Piperazic Acid Derivatives, J. Org. Chem, 69, 2367-2373.

33. Vazquez, M. A., Munoz, F., Donoso, J. (1990) Influence of the Side Chain on the

Stability of Schiff-Bases Formed Between Pyridoxal 5’-Phosphate and Amino Acids, International Journal of Chemical Kinetics, 22, 905-914.

34. Avancha, K. (2006) Design and Synthesis of Novel HIV-1 Protease Inhibitors and

Synthesis and Biological Activity of Novel 20S Proteasome Inhibitors, Dissertation, University of South Florida.

73

CHAPTER THREE

SYNTHESIS OF MDM2 INHIBITORS AND KEY STRUCTURAL UNITS 3.1 Introduction 3.1.1 The Hallmarks of Cancer

Cancer is fundamentally a genetic disease (1). Our current best understanding

suggests that cancer arises from multiple genetic transformations, and that the

development of normal cells into cancerous cells is a multi-step process (1). While no

one genetic transformation is likely to be sufficient to cause cancer, a combination of

several such transformations has been experimentally and clinically observed to lead to

tumorigenesis (1). In contrast to normal cells, which have tightly controlled cell growth

cycles, cancer cells have some defect in the regulation of their cell cycles that allows

them to proliferate in an uncontrolled manner (1).

Although there are literally dozens of known types of cancers in various tissues,

cancer cells tend to have several common characteristics that relate to their abnormal

proliferative ability. These include self-sufficiency in producing growth signals, lack of

response to anti-growth signals, evasion of programmed cell death (apoptosis), no limit to

cell reproduction, the induction of new blood vessels to support tumor growth

(angiogenesis), and the ability to invade other tissues (metastasis) (1). These six defining

characteristics are thought to be present in most if not all cancers; there may be additional

common factors such as senescence as well (1).

74

Figure 3.1: Cellular Pathways that Can Be Affected in Cancer Cells (1). These pathways are involved in cellular growth and death. Bax and p53 are both depicted on the right side of the drawing as components of the intrinsic apoptotic pathway.

There are several pathways in cells that are involved with cellular growth and

death (1). An examination of Figure 3.1 emphasizes how complex and redundant these

pathways are, as well as their inter-relatedness. MDM2, which is an inhibitor of the pro-

apoptotic tumor suppressor gene p53, and Bax, which is a promoter of apoptosis, can

both be located at the right-hand side of the figure. These two proteins and many others

involved with apoptosis are potential targets for disrupting protein-protein interactions in

75

order to promote apoptosis. This chapter will describe the design and synthesis of three

potential small-molecule inhibitors of MDM2, as well as three key structural components

needed to synthesize inhibitory MDM2 α-helix mimics. Chapter 4 will focus on the

design and synthesis of an α-helix mimic that is intended to stabilize the active form of

the Bax protein.

3.1.2 Apoptosis and MDM2/p53

The cells in a multicellular organism exist in a precarious balance between life

and programmed cell death, or apoptosis. Apoptosis is the process by which unwanted

cells that have genetic abnormalities, physical damage, or parasitic infection can be

removed from the body (2,3). In normal cells, the decision to undergo apoptosis is

governed by the relative levels of pro-apoptotic and anti-apoptotic proteins. Abnormal

expression of many of these proteins has been implicated in the transformation of normal

cells into cancer cells, and in fact is a common characteristic of cancer cells (1,3).

Apoptosis is a controlled process that is often contrasted to necrosis, where cells

die off in an uncontrolled manner (2). However, it is probably more correct to think of

necrosis and apoptosis as two extremes on a continuum of types of cell death (2).

Apoptosis can be recognized based on specific morphological features, including nuclear

fragmentation, chromatin condensation, cell shrinkage, and “blebbing” of the cell

membrane (4).

Traditionally, apoptosis was considered as only being effected by a family of

proteases called caspases, but there are also non-caspase dependent forms of controlled

cell death known (2). The caspases are activated by two primary pathways, which are

illustrated in Figure 3.2. The intrinsic pathway involves the mitochondria, while the

76

extrinsic pathway involves signaling via death receptors (3,4). Intrinsic pathway

apoptosis occurs due to the release of cytochrome c from the mitochondria following

depolarization of the mitochondrial proton gradient (3,4). The intrinsic pathway is

discussed further in the Introduction section of Chapter Four.

Figure 3.2: The Intrinsic (Mitochondrial) and Extrinsic (FAS) Pathways (3). Both p53/MDM2 and Bax (discussed in Chapter 4) are members of the intrinsic pathway. The members of this pathway function by either promoting or inhibiting cytochrome c release from the mitochondria.

One method by which cells become cancerous is due to the abnormal expression

of mutated oncogenes, which code for proteins that affect the cell cycle, or tumor

suppression genes, which block cell cycle progression (1). p53 is a tumor suppressor

77

protein that detects damage to a cell’s DNA and either prevents the cell from dividing

until after the DNA has been repaired, or causes the cell to undergo apoptosis if the

damage to the DNA is severe enough (1,3,5). p53 is regulated by MDM2 or HDM2,

which are oncogene proteins in mice and humans respectively (3). The remainder of this

Introduction will use the term MDM2 to refer to both proteins.

MDM2 physically inhibits p53, and it also promotes the ubiquitination of p53;

p53 ubiquitination leads to the degradation of p53 by the proteasome (3,5,6). In normal

cells, p53 is unstable and has a short half-life due to MDM2-mediated degradation (6).

Prevention of the MDM2-p53 interaction lengthens p53’s half-life considerably, from

minutes to hours (6).

Figure 3.3: Regulation of p53 by MDM2 (7). This drawing shows the auto-regulatory feedback loop between p53 and MDM2.

Figure 3.3 is a schematic showing the regulation of p53 by MDM2 (7). MDM2

and p53 are tied together in an auto-regulatory feedback loop, where p53 stimulates the

expression of MDM2, and MDM2 inhibits p53 (7). However, the pathway is very

78

complex and involves multiple other proteins. For example, one protein called ARF is

responsible for inhibiting MDM2 (7). There are also multiple splice variants of MDM2,

as well as related proteins with similar functions such as MDMX (7). MDMX and

MDM2 are known to interact with each other, and this interaction is also important in the

regulation of p53 (8).

Since apoptosis pathways are abnormal in many cancer cells and are not abnormal

in regular cells, it is theoretically possible to develop pro-apoptotic drugs that selectively

target cancer cells (1,3). For example, it is estimated that up to half of human cancers

have a disrupted or mutated p53 pathway, making p53 and its protein-protein interactions

an important target of anti-cancer chemotherapy (3).

Although p53 itself can be under-expressed due to mutation, its turnover can also

be increased due to abnormally high levels of MDM2 (3,4). For this reason, inhibitors of

MDM2 would be expected to restore the apoptotic pathway in cancer cells that over-

express this protein (7). This assumes that the cells still have normal expression of p53,

which is typically the case (7). There are certain types of cancers that are more likely to

have higher levels of MDM2 over-expression, including soft tissue tumors,

osteosarcomas, and gliomas (8).

3.1.3 Inhibiting the MDM2/p53 Interaction with Small Molecules MDM2 is an attractive target for several reasons (9). One of them is that the p53

regulatory function of MDM2 has been well characterized at the biological level (9).

Another reason is that the crystal structure of the MDM2/p53 binding complex has been

known for the past ten years (9). (The crystal structure of the MDM2/p53 complex is

shown in panel A of Figure 3.6 below.) And finally, experiments with site-directed

79

mutagenesis and peptide inhibitors have established a clearly-defined pharmacophore,

such that the binding requirements of any anti-MDM2 drug are well understood (9).

Several compounds have been reported that inhibit the p53-MDM2 interaction; in

general, they mimic p53 as opposed to MDM2 (6). This is because MDM2 (but not p53)

has a well-defined binding site, and p53 (but not MDM2) has an interface comprised of a

single, short stretch of amino acids arranged in an alpha helix secondary structure (6,7).

The MDM2/p53 interaction has also been studied using molecular modeling in order to

analyze the movements of both proteins when they are bound together versus unbound

(10). It was found that the conformation of MDM2 changes very little upon association

with the p53 helix, suggesting that a good inhibitor of MDM2 can be developed (10).

The first reported p53 inhibitors were phage-display peptide libraries that showed

greater affinity for MDM2 than even p53 itself did (6). The use of unnatural amino acids

and conformational restrictions improved peptide potency even more, and also provided

the proof of concept that inhibitors of MDM2 could block its interaction with p53 (6).

Unfortunately, peptides do not make good drugs for many reasons, including their lack of

oral bioavailability and their susceptibility to proteases (11). Thus, several groups are

working to make small-molecule inhibitors that can block MDM2/p53 interactions that

would hopefully show improved drug-like properties in comparison to peptides.

Until quite recently, it was believed that using small molecules to inhibit protein-

protein interactions would be difficult at best and impossible at worst (12). However,

there have now been several examples of small-molecule protein-protein inhibitors

published. These compounds can either act directly such that they inhibit the proteins

from interacting, or indirectly at an allosteric site (12). Although many protein-protein

80

interactions are difficult to disrupt using small-molecule inhibitors due to the lack of a

well-defined binding pocket, the p53/MDM2 interaction is particularly attractive in this

regard because the interface between the two molecules is relatively small, and it is

known that p53 does bury itself into a pocket on MDM2 (7).

Figure 3.4: Small Molecule Inhibitors of MDM2 (13). Panel A is an example of a chalcone, and Panel B is a boronic chalcone. Chlorofusin is shown in Panel C. Panel D shows nutlin 2, the best currently known small-molecule inhibitor of MDM2.

The first small-molecule inhibitors of MDM2 to be identified were chalcones,

which are pictured in panel A of Figure 3.4 (13). These compounds are weak MDM2

81

inhibitors with IC50 values in the high micromolar range (13). Chalcones were known to

have anti-tumor properties, and NMR studies showed that they bind to a portion of the

p53-binding site on MDM2 (14). In addition, biochemical experiments established that

chalcones can disrupt MDM2/p53 interactions (14). Boron-containing chalcone

derivatives have also been described (13); an example is shown in panel B of Figure 3.4.

Another MDM2 inhibitor is chlorofusin, a cyclic peptide derivative that was

isolated and characterized from a microorganism associated with a Fusarium genus

fungus (13,15). The structure of chlorofusin is shown in Panel C of Figure 3.4. It is

comprised of a cyclic peptide with a non-peptide chromophore attached to it (15). An

examination of this structure shows that it is too large, complex, and peptide-like to be a

good drug candidate. The cyclic peptide portion of chlorofusin was synthesized and

tested alone; it was found not to inhibit the p53-MDM2 interaction (16). This suggests

that the chromophore portion of the molecule may be needed for the compound’s activity,

which is also in the micromolar range (13).

Recently, the nutlins, a class of potent, orally bioavailable small-molecule

inhibitors of the MDM2/p53 interaction were reported (17). The nutlins were found to

bind MDM2 in the p53 pocket, activate p53 in tumor cells, and induce apoptosis (17).

They are active at a submicromolar level (17). The structure of nutlin-2 is shown in

panel D of Figure 3.4. Nutlin-2’s crystal structure with MDM2 showed that nutlin-2

closely mimics the binding properties of p53 (17). In particular, three binding residues of

p53 (Phe, Trp, and Leu) were all mimicked by nutlin-2, in contrast to previous inhibitors

like the chalcones that only replaced a portion of the p53 alpha helix (13).

82

Biological tests with another compound, nutlin-3, showed that nutlin-3 selectively

induces the p53 pathway, promoting cell cycle arrest and apoptosis in vivo as well as in

vitro (18). Cancer cells with over-expressed or even normal levels of MDM2 both

showed sensitivity to nutlin-3, but only if p53 expression was normal (18). Fortunately,

it appears that most tumors that over-express MDM2 do not also have mutations of p53

as well (18). It seems reasonable that tumor cells that over-express MDM2 would most

likely have intact p53 signaling, because further attenuation of p53 would not be

necessary for them to avoid undergoing apoptosis (18).

OH

OOO

NH

HN

O

Figure 3.5: Structure of syc-7, a Non-Peptide Inhibitor of HDM2 (19). This compound and some of its derivatives were able to inhibit HDM2/p53 interactions and induce apoptosis in some tumor cell lines.

Finally, there have also been a series of small-molecule inhibitors of the

HDM2/p53 association reported (19). These compounds, called syc compounds, were

conceived using computer-aided design, and then synthesized (19). One compound in

particular, syc-7, showed an ability to inhibit HDM2-p53 interactions, and also could be

seen to activate the p53 pathway and induce apoptosis in four different tumor cell lines

(19). However, the concentrations of syc-7 needed to induce apoptosis were

83

prohibitively high for a drug, in the low millimolar range (19). The structure of syc-7 is

shown in Figure 3.5.

3.1.4 Inhibiting MDM2/p53 with Peptides and Peptidomimetics As mentioned previously, the first inhibitors of the MDM2/p53 interaction were

peptides and modified peptides obtained by screening phage libraries, reported in 1996

(6,20). These researchers originally used peptides to establish the p53 pharmacophore,

and they discovered that the hexapeptide Thr-Phe-Ser-Asp-Leu-Trp was the consensus

sequence (20). However, this peptide had fairly low binding affinity that was in the

micromolar range (20). Somewhat better results were obtained by screening a phage

display peptide library, which improved the binding affinity by 28-fold (20). Starting

with an octapeptide lead, the binding affinity of this peptide for HDM2 was improved

even more to the low nanomolar range by replacing non-pharmacophore amino acids

with α,α-disubstituted amino acids (20). It is known that the use of disubstituted amino

acids can stabilize the α-helix secondary structure of a short peptide (20).

More recently, Andy Hamilton’s group at Yale has made a variety of α-helix-like

peptidomimetics based on a terphenyl scaffold (21,22). The substituents on the benzene

rings in these structures are too close together to allow the benzene rings to be coplanar,

which gives the molecule a twisted or helical shape. Placing substituents at the ortho

positions on each benzene ring orients these substituents so that they mimic the

orientation of the amino acid side chains on an alpha helix (21,22). The relationship

between Hamilton’s helices and the p53 α-helix can be seen in Figure 3.6. Panel A

shows the crystal structure of p53 interacting with MDM2. The orientation of the side

84

chains at the i, i+4, and i+7 positions can be seen in p53 versus a terphenyl helix in Panel

B. (These are the essential Phe, Trp, and Leu residues identified previously.)

Figure 3.6: Design, Model, and Structures of Hamilton’s Terphenyl Helices (21, 22). Panel A shows the x-ray crystal structure of p53 (backbone) complexed with MDM2 (space-filling). Panel B shows how the physical topography of an α-helix is mimicked by the Hamilton helices. Several examples of Hamilton helices that were screened against MDM2 are shown in Panel C.

Hamilton’s group discovered that the best inhibition occurred with the use of an

isobutyl or benzyl group on the first phenyl ring, mimicking Phe (22). In addition, it was

helpful to have an extra methyl side chain on the second phenyl ring, since this made the

compound more rigid and better able to induce activation of p53 (22). Fortuitously, it

was found that several of the terphenyl compounds were membrane-permeable, and they

could induce p53 activation and accumulation in tumor cells (22). Several examples of

85

p53-like terphenyl helices are shown in Panel C. Other terphenyl helices have been

developed as mimics of the Bak BH3 domain; this work will be discussed further in

Chapter Four.

3.2 Results and Discussion

Although Hamilton’s helices seem like a promising idea for p53 mimics, they do

have the disadvantage of being nonpolar and therefore not very water-soluble. They are

also difficult to synthesize. Accordingly, we have designed a new set of p53 helices

based on hydrazino subunits that can be assembled into a set of piperazine dione rings, as

shown in the top panel of Figures 3.7. As with Hamilton’s terphenyl helices, the

McLaughlin compounds are also predicted to form a twisted shape based on molecular

models. The model is shown in the bottom panel of Figure 3.7.

Figure 3.7: Structure and Model of the McLaughlin Piperazine Helix. Not only does this helix twist like the Hamilton Helix does, it has the additional advantage of being more polar and water-soluble. The building blocks of the helix can be synthesized from ordinary natural or unnatural amino acids, allowing for a great diversity of side chains.

86

The chemistry to make piperazine diones is well known, and there are several

examples in the medicinal chemistry literature (23). These compounds can be made from

ethylene diamine tetraacetic acid tetraethyl ester scaffolds (23). Members of several

classes of drugs contain piperazine diones, including topoisomerase inhibitors, tricyclic

anti-depressants, and antihistamines (23). In addition, these compounds can be made

using solid-phase synthesis, as well as in solution (24). As far as we are aware, there are

currently no examples in the literature of hydrazino piperazine diones from amino acids.

XO

NBocHN R1

HO

X R N N

O

O

R1

NHBoc

H1) 95% TFA 2) TEA 3) couple

XO

NBocHN R2

HO

X

4) 95% TFA 5) TEA 6) couple

Resin N N

O

O

R1

N

H

N

O

O

NHBoc

R2H

R N N

O

O

R1

N

H

N

O

O

N

R2H

N

O

O

R3H

NHBocrepeat

N NH

O

O

R1

N

H

NH

O

O

N

R2H

NH

O

O

R3H

NH

A1 A2

ResinO O

NHBoc

HO

O O

OH

O

10) 95% TFA 11) TEA

12)

CO2BnX

O 13) HBr, AcOH

for A3

Figure 3.8: Synthesis of McLaughlin Piperazine Dione α-Helix Mimics. The hydrazino units will be incorporated into piperazine dione peptidomimetics via solid-phase synthesis. X represents an activated carboxylic acid; R is an amino acid side chain.

The McLaughlin helices have the advantage of being amenable to solid-phase

peptide synthesis, thus facilitating their construction. They also have the advantage that

multiple natural and unnatural amino acid side chains can be inserted into the

peptidomimetic. Figure 3.8 shows how the hydrazino subunits can be combined into the

piperazine dione compound. Compounds 17a–c and other alkylated hydrazino amino

87

acids can be incorporated into the piperazine dione alpha helix mimics described above.

This is currently being done in our lab via solid-phase synthesis through a series of

deprotection, activation, and coupling steps. The synthesis of hydrazines 17a-c is

discussed below.

OBr

O

ON

O

O O

R

H

ON

O

O O

R

N

OO

H

HON

OH

O O

R

N

OO

H

H2N

O

O

R

ClH

LiOH

CH3OH

DIEA

CH3CN

CH3OH

ON

Cl3CO

O10

15a-c14a-c

16a-c 17a-c

a b c

R =

N HN

Scheme 3.1: Synthesis of Alkylated Hydrazino Amino Diacids. N-amination of the three alkylated amino acids with oxaziridine 10 (25,26) was the key step in this synthesis. Hydrolysis of hydrazine 16a gave product 17a. The synthesis of compounds 16b-c and 17b-c was not completed.

88

18

NaH

MeIDMF

H2/Pd

MeOHAcOH

K2CO3CH3I

DMF91-100%

19a

19b 14c 14b

Cbz = O

O

H2/Pd

MeOHAcOH

NO

HCbz

H O

N H

NOCbz

H O

N H

NOH

H O

N

NOH

H O

N H

NOCbz

H O

N

Scheme 3.2: Synthesis of Tryptophan Methyl Esters 14b and 14c. Both compounds were prepared via a series of protection and deprotection steps from commercially available Cbz-tryptophan. Each of these steps proceeds with approximately quantitative yields. Hydrazino amino acid derivative 17a was synthesized to be used as a subunit in

the construction of the McLaughlin helices. It was synthesized using the procedure

depicted in Scheme 3.1. The methyl ester of the amino acid 14a was first reacted with

89

ethyl bromoacetate in the presence of diisopropylethylamine to form secondary amine

15a. This compound was then reacted with oxaziridine 10 (25, 26) to give the Boc-

protected hydrazine 16a. Oxaziridine 10 here is the same compound that was used

previously to synthesize compound 5b, as described in Chapter Two. The final step in

the synthesis was to hydrolyze both esters using lithium hydroxide to give 17a. The

synthesis of two other hydrazino amino acid derivatives (17b-c) was partially completed.

The isoleucine methyl ester 14a was commercially available as the hydrochloride

salt, but the two tryptophan methyl esters 14b and 14c had to be synthesized from

commercially available compound 18 as shown in Scheme 3.2. Compound 18 is the Cbz-

protected amino acid tryptophan. The first step was to make the methyl ester 19a of 18

using potassium carbonate and methyl iodide. Compound 19a was then split into two

portions. One portion was hydrogenated to directly form compound 14b, which was then

used to form the hydrazine diacid 17b as shown in Scheme 3.1. The other portion was N-

alkylated on the indole nitrogen to give 19b, and compound 19b was then also

hydrogenated to form the alkylated indole tryptophan ester 14c. Finally, compound 14c

was converted into hydrazine diacid 17c, also shown in Scheme 3.1.

Initially, our synthetic procedure called for using sodium hydroxide to hydrolyze

the esters rather than lithium hydroxide. This procedure worked fine for some of the

first amino acids tried, which included phenylalanine and leucine. However, the hindered

amino acids valine and isoleucine did not hydrolyze to the diacid with NaOH. Instead,

treatment of isoleucine hydrazine diester 16a with NaOH consistently yielded monoacid

20 instead of the desired diacid 17a. We tried refluxing 16a for 48 hours using NaOH,

but we still could not obtain the diacid. However, we finally were able to obtain 17a

90

when we used lithium hydroxide. Subsequent attempts at hydrolysis of 16a with NaOH

were finally successful, but doing this required harsh conditions that might lead to the

racemization of the alpha carbon. However, we did not observe any such racemization.

ON

O

O ON

OO

H

HON

OH

O ON

OO

HLiOH

CH3OH

16a 17a

ON

O

O ON

OO

H

HON

O

O ON

OO

H

16a 20

NaOH

CH3OH

Scheme 3.3: Results of the Hydrolysis of 16a with NaOH versus LiOH. Unlike the less hindered amino acids, use of NaOH to hydrolyze the Ile diester led to the monoacid product 20 rather than the desired diacid 17a. Compound 17a was ultimately obtained by hydrolysis of 16a with LiOH.

N

OH

O

O

N H

NSC-131734

Gly-465

Figure 3.9: MDM2 Small-Molecule Inhibitor Lead Compound NSC-131734. Our analogs 23 attempted to mimic the quinoline and anthracene systems of this compound, and all three compounds could be synthesized easily in a single step.

91

N

H

Cl

NaH

DMF

N

N

H

Cl

NaH

DMF

N

N

H

Cl

NaH

DMF

N

21a

21b

21c

22

22

22

23a

23b

23c Scheme 3.4: Synthesis of MDM2 Inhibitors 23a-c. All three compounds were synthesized in a single step from 1-(chloromethyl)naphthalene and the appropriate indole or carbazole derivative.

We have also been working on the design and synthesis of a small library of

small-molecule inhibitors of MDM2. One compound that showed modest MDM2

inhibitory activity was NCI compound NSC-131734, shown in Figure 3.9. Three mimics

23 of NSC-131734 were easily synthesized using similar chemistry as previously

described for the tryptophan alkylation. This procedure is shown in Scheme 3.4. 2,3-

dimethylindole (21a), carbazole (21b), and 1,2,3,4-tetrahydracarbazole (21c) were each

treated with sodium hydride and then 1-(chloromethyl)-naphthalene (22) to give N-

alkylated products 23.

92

All three of these compounds were tested as inhibitors of MDM2 and were found

to be inactive. This may be due to their hydrophobicity, as all three compounds are

insoluble in water and are significantly less polar than the model NCI compound. It may

also be due to the molecules being too small to adequately cover the entire p53 binding

pocket on MDM2.

In conclusion, we successfully carried out the synthesis of isoleucine hydrazino

amino acid derivative 17a. We also made progress toward the synthesis of two

tryptophan hydrazino acid derivatives 17b-c. Finally, we synthesized three potential

small-molecule inhibitors (23a-c) of MDM2 that were modeled after a screened NCI

compound. Further work on this project will entail completion of the syntheses of 17b-c,

assembly of the hydrazino amino acid derivatives into piperazine diones, and further

biological testing of both the piperazine diones and the small-molecule inhibitors 23.

3.3 Experimental Data

Experimental Procedure for N-Alkylation. The free amine 14 (5.50 mmol) was

dissolved in acetonitrile (10 mL). Diisopropylethylamine (11.01 mmol) and ethyl

bromoacetate (16.51 mmol) were added via syringe. The reaction was stirred overnight

in an argon atmosphere at room temperature. The resulting product was washed with 5%

citric acid, the organic layer was separated, and the aqueous layer was extracted with

ethyl acetate (3 x 20 mL). The combined organic layers were washed with brine (1 x 20

mL), dried with magnesium sulfate and evaporated under reduced pressure, yielding a

yellow oil. The crude product was purified by column chromatography (4:1

hexane/EtOAc) to give pure 15, a colorless oil.

93

ON

O

O OH

15a

Methyl N-(2-Ethoxy-2-Oxoethyl)-(S)-Isoleucinate (15a). Rf = 0.62 (3:1 hexanes

/EtOAc); 1H NMR (400 MHz, CDCl3) δ 0.82-0.86 (m, 6 H), 1.12-1.22 (m, 4 H), 1.45-

1.51 (m, 1 H), 1.63-1.69 (m, 1 H), 1.96 (s, 1 H), 3.08 (d, J = 6.0 Hz, 1 H), 3.24 (d, J =

17.2 Hz, 1 H), 3.34 (d, J = 17.2 Hz, 1 H), 3.65 (s, 3 H), 4.11 (q, J = 7.2 Hz, 2 H); 13C

NMR (100 MHz, CDCl3) δ 11.62, 14.33, 15.72 25.63, 38.48, 49.94, 51.65, 60.90, 65.76,

171.96, 174.83 ppm; MS (GC/MS) m/z 232, 199, 172, 158, 146, 130, 114, 100, 86, 69,

56.

15b

NO

OH

N H

O

O

Methyl N-(2-Ethoxy-2-Oxoethyl)-(S)-Tryptophanate (15b). Rf = 0.73 (100%

EtOAc), Rf = 0.04 (3:1 hexanes /EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.19 (t, J = 6.8

Hz, 3H), 2.37 (broad, 1H), 3.14-3.19 (m, 1H), 3.26 (dd, J = 6.0 Hz, 14.4 Hz, 1H), 3.32-

3.36 (m, 1H), 3.45-3.49 (m, 1H), 3.66 (s, 3H), 3.72 (t, J = 6.0 Hz, 1H), 4.10 (q, J = 6.8

Hz, 2H), 7.10-7.13 (m, 2H), 7.16-7.20 (m, 1H), 7.34 (d, J = 8.0 Hz, 1H), 7.60 (d, J = 7.6

Hz, 1H), 8.27 (broad, 1H); 13C NMR (100 MHz, CDCl3) δ 14.32, 29.33, 49.45, 52.17,

61.11, 61.38, 110.94, 111.48, 118.87, 119.68, 122.31, 123.33, 127.60, 136.48, 171.84,

174.59; HR ESI-MS (0.1% AcOH in MeOH) 327.13188 (M + Na)+

94

15c

NO

OH

N

O

O

Methyl N’-Methyl-N-(2-Ethoxy-2-Oxoethyl)-(S)-Tryptophanate (15c). Rf = 0.81

(100% EtOAc), Rf = 0.12 (3:1 hexanes /EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.22 (t, J

= 7.2 Hz, 3H), 2.17 (broad, 1H), 3.14 (dd, J = 7.2 Hz, 14.8 Hz, 1H), 3.24 (dd, J = 5.6 Hz,

14.4 Hz, 1H), 3.31-3.36 (m, 1H), 3.44-3.3.50 (m, 1H), 3.66 (s, 3H), 3.70 (t, J = 5.6 Hz,

1H), 3.75 (s, 3H), 4.11 (q, J = 7.6 Hz, 2H), 6.98 (s, 1H), 7.09-7.13 (m, 1H), 7.20-7.24 (m,

1H), 7.26-7.30 (m, 2H), 7.59 (d, J = 8.0 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 14.30,

28.70, 32.96, 49.03, 52.40, 61.24, 61.44, 108.66, 109.57, 118.91, 119.30, 122.00, 127.98,

128.27, 137.27, 170.92, 172.76; HR ESI-MS (0.1% AcOH in MeOH) 341.14739 (M +

Na)+

Experimental Procedure for Hydrazine Formation (16a-c). Compound 15 (5.75

mmol) was dissolved in methanol (12 mL) in a round bottom flask and cooled to –78 °C

under an argon atmosphere. Oxaziridine 10 (6.55 mmol, Ref. 25,26) was dissolved in

methanol (10 mL) and added to the reaction mixture slowly via syringe, and the reaction

was allowed to warm to room temperature and stirred overnight. The reaction was then

washed with water, the organic layer was separated, and the aqueous layer was extracted

with ethyl acetate (3 x 20 mL). The combined organic layers were washed with brine (1

x 25 mL), dried with magnesium sulfate and evaporated under reduced pressure. The

95

crude product was purified by column chromatography (9:1 hexane/EtOAc) to give pure

16, a colorless oil.

ON

O

O ONH

O O

16a

Methyl N”-tert-Butoxycarbonyl-N-(2-Ethoxy-2-Oxoethyl)-(S)-Isoleucinate (16a). Rf

= 0.53 (3:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 0.83-0.94 (m, 6 H), 1.16-1.22

(m, 1H), 1.23 (t, 3 H), 1.44 (s, 9 H), 1.74-1.78 (m, 1 H), 1.95-2.01 (m, 1H), 3.16-3.21 (m,

1H), 3.57-3.61 (m, 1 H), 3.69-3.76 (m, 4 H), 4.15 (q, J = 7.2, 2H), 7.21 (broad, 1H); 13C

NMR (100 MHz, CDCl3) δ 11.03, 11.67, 14.33, 14.37, 15.47, 15.74, 25.71, 28.18, 28.41,

28.47, 34.91, 38.47, 49.92, 51.47, 51.75, 56. 55, 60.93, 61.03, 65.78, 71.99, 80.15,

170.52, 171.91, 172.89, 174.76 ppm; MS (GC/MS) m/z 346, 287, 273, 246, 231, 217,

189, 173, 157, 145, 129, 113, 101, 87, 69, 57; HR ESI-MS (0.1% AcOH in MeOH)

369.19977 (M + Na)+

20

HON

O

O ONH

O O

96

Procedure for N”-tert-Butoxycarbonyl-N-(2-Ethoxy-2-Oxoethyl)-(S)-Isoleucinate

Synthesis (20). Compound 16a (0.75 mmol) was dissolved in methanol (3 mL), and 1.87

mL of a 1.0 solution of sodium hydroxide was added to it by syringe. The reaction was

stirred for four hours at RT. The resulting product was washed with potassium bisulfate.

The organic layer was separated and the aqueous layer was extracted with ethyl acetate (3

x 10 mL). The combined organic layers were washed with brine (1 x 15 mL), dried with

magnesium sulfate and dried under reduced pressure, yielding 20 (0.21 g, 79%), a

colorless oil; Rf = 0.69 (9:1 DCM/MeOH); 1H NMR (400 MHz, CDCl3) δ 0.80-0.86 (m, 6

H), 1.14-1.19 (m, 1 H), 1.40 (s, 9 H), 1.73-1.76 (m, 1 H), 1.77-1.89 (m, 1H), 3.16 (d, J =

9.6 Hz, 1 H), 3.45-3.54 (m, 2 H), 3.71 (s, 3 H), 7.21 (broad, 1H); 13C NMR (100 MHz,

CDCl3) δ 11.14, 15.70, 26.26, 28.36, 28.41, 35.50, 52.12, 72.74, 82.78, 158.00, 171.17

ppm; ESI-MS (0.1% AcOH in MeOH) 317.1 (M - H)-

Procedure for Diacid Synthesis. Compound 16a (0.45 g, 1.3 mmol) was dissolved in

methanol (13 mL). 0.5 M lithium hydroxide (9.29 mL) was added, the flask was heated

to 50 °C and stirred for 24 hours. The resulting product was washed with potassium

bisulfate. The organic layer was separated and the aqueous layer was extracted with

ethyl acetate (3 x 10 mL). The combined organic layers were washed with brine (1 x 15

mL), dried with magnesium sulfate and dried under reduced pressure. The crude product

was recrystallized (DCM/hexane) yielding 17a (0.17g, 45%), a white solid.

97

17a

HON

OH

O ONH

O O

N”-tert-Butoxycarbonyl-N-Carboxymethyl-(S)-Isoleucinate (17a). MP = 98-100 °C;

Rf = 0.13 (9:1 DCM/MeOH); 1H NMR (400 MHz, CD3OD) δ 0.89-0.95 (m, 6 H), 1.17-

1.30 (m, 1 H), 1.46 (s, 9 H), 1.71 -1.87 (m, 2 H), 3.34-3.36 (m, 1 H), 3.64-3.69 (m, 2 H);

13C NMR (100 MHz, CD3OD) δ 9.87, 10.41, 10.60, 14.55, 14.86, 15.38, 25.12, 25.61,

26.03, 27.40, 34.38, 35.11, 35.45, 53.63, 57.82, 70.56, 71.82, 72.19, 80.83, 157.12,

172.26, 174.07, 174.18, 174.74 ppm; ESI-MS (0.1% AcOH in MeOH) 303.1 (M - H)-

19a

NOCbz

H O

N H

Experimental Procedure for Methyl Esterification of N-Carboxybenzyl-(S)-

Tryptophanate to Methyl N-Carboxybenzyl-(S)-Tryptophanate (19a). The

commercially available Cbz- protected Trp 18 (1.48 mmol) was dissolved in dry DMF

(20 mL) and cooled to 0 °C. Anhydrous potassium carbonate (1.63 mmol) was added,

and the reaction was flushed with argon and stirred at 0 °C for 10 minutes. Methyl iodide

98

(2.96 mmol) was then added drop-wise via syringe, and the reaction was allowed to warm

up to room temperature. After stirring for 4.5 hours, the reaction was diluted with ethyl

acetate (50 mL) and washed twice with saturated sodium bicarbonate solution. The

organic layer was then washed once with saturated NaCl solution, dried over MgSO4, and

evaporated under reduced pressure. The pure product 19a was isolated by filtering

through a small pad of silica using 3:1 hexanes:ethyl acetate in 100% yield. Rf = 0.59

(1:1 hexanes/EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.31 (d, J = 5.6 Hz, 2H), 3.68 (s,

3H), 4.71-4.75 (m, 1H), 5.07-5.15 (m, 2H), 5.42 (d, J = 8 Hz, 1H), 6.92 (s, 1H), 7.10 (t, J

= 7.2 Hz, 1H), 7.19 (t, J = 7.6 Hz, 1H), 7.32-7.35 (m, 6H), 7.53 (d, J = 8.0 Hz, 1H), 8.42

(broad, 1H); 13C NMR (100 MHz, CD3OD) δ 28.18, 52.61, 54.81, 67.19, 109.87, 111.59,

118.78, 119.84, 122.39, 123.24, 127.76, 128.36, 128.41, 128.77, 136.44, 136.55, 156.12,

172.74; ESI-MS (0.1% AcOH in MeOH) 353.1 (M + H)+

19b

NOCbz

H O

N

Experimental Procedure to Form Methyl N’-Methyl-N-Carboxybenzyl-(S)-

Tryptophanate (19b). The Trp methyl ester 19a (1.48 mmol) was dissolved in 25 mL

anhydrous DMF and the reaction mixture was cooled to 0 °C. Unwashed 60% sodium

hydride was added slowly, and the reaction was stirred for fifteen minutes until all

effervescence had ceased. Then methyl iodide (1.63 mmol) was added, and the reaction

99

was allowed to warm to room temperature and stirred for three hours. It was quenched

with water, forming an oily white precipitate that could not be filtered. The reaction

mixture was acidified with 10% KHSO4 to pH = 3, and the aqueous layer was extracted

three times with DCM. The combined organic layer was then extracted twice with dI

water and once with brine, dried over MgSO4, and evaporated under reduced pressure to

give the pale yellow oil 19b in 77.8% yield. Rf = 0.67 (1:1 hexanes/EtOAc); 1H NMR

(400 MHz, CDCl3) δ 3.33 (d, J = 5.6 Hz, 2H), 3.71 (s, 3H), 3.72 (s, 3H), 4.73-4.75 (m,

1H), 5.10-5.18 (m, 2H), 5.40 (d, J = 8.0 Hz, 1H), 6.84 (s, 1H), 7.10-7.14 (m, 1H), 7.22-

7.37 (m, 7H), 7.54 (d, J = 8.0 Hz, 1H); 13C NMR (100 MHz, CD3OD) δ 28.04, 32.94,

52.59, 54.81, 67.15, 108.45, 109.59, 118.98, 119.44, 122.05, 127.79, 128.20, 128.30,

128.41, 128.78, 136.64, 137.19, 156.07, 172.69; ESI-MS (0.1% AcOH in MeOH) 367.1

(M + H)+

Experimental Procedure to Remove the Cbz Protecting Group (14b-c). The

methylated tryptophan derivatives (0.426 mmol) were dissolved in a few mL of

methanol/THF, to which was added 0.1 g 10% Pd/C and a couple of drops of glacial

acetic acid. This mixture was then placed in a hydrogen atmosphere at approximately 50

psi and shaken for one hour, after which the Pd/C was removed by filtering the dissolved

product through a Celite bed. The product was obtained as a white solid in 100% yield.

100

14b

HN

O

OH

N H

Methyl (S)-Tryptophanate (14b). MP = 101-103 °C; Rf = 0.08 (100 % EtOAc); 1H

NMR (400 MHz, CD3OD) δ 3.59 (s, 3H), 4.00-4.03 (m, 1H), 6.89-6.91 (m, 1H), 6.95-

6.97 (m, 1H), 7.01 (s, 1H), 7.22 (d, J = 8.4 Hz, 1H), 7.37 (d, J = 8.0 Hz); 13C NMR (100

MHz, CD3OD) δ 27.93, 52.60, 54.26, 107.63, 111.90, 118.22, 119.49, 122.13, 124.64,

127.63, 137.55, 171.75; HR ESI-MS 219.11305 (0.1% AcOH in MeOH) (M + H)+

14c

HN

O

OH

N

Methyl N’-Methyl -(S)-Tryptophanate (14c). MP = 105-106 °C; Rf = 0.15 (100%

EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.07-3.13 (m, 1H), 3.29 (dd, J = 4.4 Hz, 14.4Hz,

1H), 3.85-3.88 (m, 1H), 3.73 (s, 3H), 3.76 (s, 3H), 4.96 (broad, 2H), 6.94 (s, 1H), 7.11 (t,

J = 8.0 Hz, 1H), 7.21-7.26 (m, 1H), 7.30 (d, J = 8.4 Hz, 1H), 7.58 (d, J = 7.6 Hz, 1H); 13C

NMR (100 MHz, CDCl3) δ 30.27, 32.93, 52.35, 54.75, 109.27, 109.55, 119.03, 119.32,

122.02, 128.05, 128.14, 137.30, 175.48; HR ESI-MS (0.1% AcOH in MeOH) 233.12831

(M + H)+

101

Procedure to Synthesize MDM2 Inhibitors. 60% sodium hydride (0.897 mmol) was

washed two times with dry DMF to remove the oil, and then suspended in approximately

ten mL dry DMF. The reaction was cooled to 0° C, and then the indole or carbazole

derivative (0.598 mmol) was added slowly. The reaction mixture was then stirred for

fifteen minutes at 0° C, after which 1-chloromethylnaphthalene (0.658 mmol) was added

dropwise. The reaction was allowed to warm up to room temperature and stirred at room

temperature for five hours. Ethyl acetate was then used to dilute the reaction, and it was

washed three times with dI water and once with brine and dried over magnesium sulfate.

Evaporation under reduced pressure yielded the crude product, which was recrystallized

in ethyl acetate mixed with a few drops of hexanes to give the pure product in 66-91%

yield.

N

23a

1-(1’-Methylnaphthyl)-2,3-Dimethylindole (23a). MP = 150-152 °C; Rf = 0.77 (3:1

hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 2.29 (s, 3H), 2.38 (s, 3H), 5.76 (s, 2H),

6.37 (d, J = 6.8 Hz, 1H), 7.11-7.17 (m, 3H), 7.21-7.26 (m, 2H), 7.59-7.63 (m, 3H), 7.74

(d, J = 8.4 Hz, 1H), 7.94 (d, J = 8.0 Hz, 1H), 8.14 (d, J = 8.4 Hz, 1H); 13C NMR (100

MHz, CDCl3) δ 9.24, 10.21, 44.41, 107.40, 109.10, 118.32, 119.20, 121.10, 122.42,

122.91, 126.06, 126.11, 126.62, 127.77, 129.02, 129.27, 130.53, 132.94, 133.39, 133.79,

136.76; HR ESI-MS (0.1% AcOH in CH3CN) 286.1590 (M + H)+.

102

N

23b

9-(1’-Methylnaphthyl)-Carbazole (23b). MP = 146-148 °C; Rf = 0.73 (3:1

hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 5.94 (s, 2H), 6.68 (d, J = 7.2 Hz, 1H),

7.20 (t, J = 7.6 Hz, 1 H), 7.30-7.32 (m, 2H), 7.34-7.38 (m, 2H), 7.43-7.48 (m, 2H), 7.64-

7.71 (m, 2H), 7.79 (d, J = 8.4 Hz, 1H), 7.99 (d, J = 8.8 Hz, 1H), 8.17 (d, J = 8.4 Hz, 1H),

8.28 (d, J = 7.6 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 44.57, 109.31, 119.68, 120.78,

122.61, 123.29, 123.44, 125.98, 126.25, 126.28, 126.80, 128.11, 129.41, 130.94, 131.94,

134.04, 141.18; HR ESI-MS (0.1% AcOH in CH3CN) 308.1429 (M + H)+.

N

23c

9-(1’-Methylnaphthyl)-1,2,3,4-Tetrahydrocarbazole. MP = 162-164 °C; Rf = 0.76

(3:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.92-1.93 (m, 4H), 2.62 (s, 2H), 2.86

(s, 2H), 5.73 (s, 2H), 6.45 (d, J = 6.4 Hz, 1H), 7.10-7.17 (m, 3H), 7.23-7.27 (m, 1H),

7.59-7.65 (m, 3H), 7.75 (d, J = 8.4 Hz, 1H), 7.94 (d, J = 7.6 Hz, 1H), 8.12 (d, J = 8.8 Hz,

1H); 13C NMR (100 MHz, CDCl3) δ 21.45, 22.16, 23.46, 23.52, 44.08, 109.25, 110.33,

118.10, 119.22, 121.10, 122.46, 123.02, 126.04, 126.09, 126.61, 127.76, 127.82, 129.26,

130.58, 133.47, 133.79, 136.15, 136.99; HR ESI-MS (0.1% AcOH in CH3CN) 312.1743

(M + H)+.

103

3.4 References 1. Hanahan, D., Weinberg, R. A. (2000) The Hallmarks of Cancer, Cell, 100, 57-70. 2. Lockshin, R. A., Zakeri, Z. (2002) Caspase-Independent Cell Deaths, Current

Opinion in Cell Biology, 14, 727-733. 3. Fesik, S. W. (2005) Promoting apoptosis as a strategy for cancer drug discovery.

Nature Reviews Cancer, 5(11), 876-885. 4. Senderowicz, A. M. (2004) Targeting cell cycle and apoptosis for the treatment of

human malignancies, Current Opinion in Cell Biology, 16, 670-678. 5. Shmueli, A., Oren, M. (2004) Regulation of p53 by Mdm2: Fate is in the numbers.

Molecular Cell, 13(1), 4-5. 6. Moll, U. M., Petrenko, O. (2003) The MDM2-p53 Interaction, Molecular Cancer

Research, 1, 1001-1008. 7. Chene, P. (2003) Inhibiting the p53-MDM2 Interaction: An Important Target for

Cancer Therapy, Nature Reviews: Cancer, 3, 102-109. 8. Levav-Cohen, Y, Haupt, S., Haupt, Y., (2005) MDM2 in growth signaling and

cancer, Growth Function, 23, 183-192. 9. Chene, P. (2004) Inhibition of the p53-MDM2 Interaction: Targeting a Protein-

Protein Interface, Molecular Cancer Research, 2, 20-26. 10. Barrett, C. P., Hall, B. A., Noble, M. E. M. (2004) Dynamite: a simple way to gain

insight into protein motions. Biological Crystallography, 60 (1), 2280-2287. 11. Guy, L., Vidal, J., Collet, A. (1998) Design and Synthesis of Hydrazinopeptides and

Their Evaluation as Human Leukocyte Elastase Inhibitors Journal of Medicinal Chemistry, 41, 4833-4843.

12. Pagliaro, L., Felding, J., Audouze, K., Nielsen, S. J., Terry, R. B., Krog-Jenson, C.,

Butcher, S. (2004) Emerging classes of protein-protein interaction inhibitors and new tools for their development, Current Opinions in Chemical Biology, 8, 442-449.

13. Klein, C., Vassilev, L. T. (2004) Targeting the p53-MDM2 interaction to treat

cancer, British Journal of Cancer, 91, 1415-1419.

104

14. Stoll, R., Renner, C., Hansen, S., Palme, S., Klein, C., Belling, A., Zeslawski, W., Kamionka, M., Rehm, T., Muhlhahn, P., Schumacher, R., Hesse, F., Kaluza, B., Voelter, W., Engh, R. A., Holak, T. A. (2001) Chalcone Derivatives Antagonize Interactions between the Human Oncoprotein MDM2 and p53, Biochemistry, 40, 336-344.

15. Duncan, S. J., Gruschow, S., Williams, D. H., McNicholas, C., Purewal, R., Hajek,

M., Gerlitz, M., Martin, S., Wrigley, S. K., Moore, M. (2001) Isolation and Structure Elucidation of Chlorofusin, a Novel p53-MDM2 Antagonist from a Fusarium sp., Journal of the American Chemical Society, 123, 554-560.

16. Malkinson, J. P., Zloh, M., Kadom, M., Errington, R., Smith, P. J., Searcey, M.

(2003) Solid Phase Synthesis of the Cyclic Peptide Portion of Chlorofusin, an Inhibitor of p53-MDM2 Interactions, Organic Letters, 5, 5051-5054.

17. Vassilev, L. T., Vu, B. T., Graves, B., Carvajal, D., Podlaski, F., Filipovic, Z., Kong,

N., Kammlott, U., Lukacs, C., Klein, C., Fotouhi, N., Liu, E. A. (2004) In Vivo Activation of the p53 Pathway by Small-Molecule Antagonists of MDM2, Science, 303, 844-848.

18. Tovar, C., Rosinski, J., Filipovic, Z., Higgins, B., Kolinsky, K., Hilton, H., Zhao, X.,

Vu, B. T., Qing, W., Packman, K., Myklebost, O., Heimbrook, D. C., Vassilev, L. T. (2006) Small-molecule MDM2 antagonists reveal aberrant p53 signaling in cancer: Implications for therapy, Proceedings of the National Academy of Sciences, 103, 1888-1893.

19. Zhao, J., Wang, M., Chen, J., Luo, A., Wang, X., Wu, M., Yin, D., Liu, Z. (2002)

The initial evaluation of non-peptide small-molecule HDM2 inhibitors based on p53-HDM2 complex structure, Cancer Letters, 183, 69-77.

20. Garcia-Echeverria, C., Chene, P., Blommers, M. J. J., Furet, P. (2000) Discovery of

Potent Antagonists of the Interaction between Human Double Minute 2 and Tumor Suppressor p53, Journal of Medicinal Chemistry, 43, 3205-3208.

21. Yin, H., Lee, G., Park, H. S., Payne, G. A., Rodriguez, J., Sebti, S. M., Hamilton, A.

D. (2005) Terphenyl-Based Helical Mimetics That Disrupt the p53/HDM2 Interaction, Angewandte Chemie International Edition, 44, 2704-2707.

22. Chen, L., Yin, H., Farooqi, B., Sebti, S., Hamilton, A. D., Chen, J. (2005) p53 α-

Helix mimetics antagonize p53/MDM2 interaction and activate p53, Molecular Cancer Therapeutics, 4, 1019-1025.

23. Insaf, S. S., Witiak, D. T. (2000) Synthesis of All Distinct α-Methyl-Substituted

Isomers of Amino Bis(2,2’-Ethanoic Acid) Diethyl Ester and Ethylene Diamine Tetraacetic Acid Tetraethyl Ester Scaffolds, Tetrahedron, 56, 2359-2367.

105

24. Perotta, E., Altamura, M., Barani, T., Bindi, S., Giannotti, Harmat, N. J. S., Nannicini, R., Maggi, C. A. (2001) 2,6-Diketopiperazines from Amino Acids, from Solution-Phase to Solid-Phase Organic Synthesis, Journal of Combinatorial Chemistry, 3, 453-460.

25. Vidal, J., Hannachi, J. C., Hourdin, G., Mulatier, J. C., Collet, A. (1998) N-Boc-3-

trichloromethyloxaziridine: a new powerful reagent for electrophilic amination, Tetrahedron Letters, 39, 8845-8848.

26. Avancha, K. (2006) Design and Synthesis of Novel HIV-1 Protease Inhibitors and

Synthesis and Biological Activity of Novel 20S Proteasome Inhibitors, Dissertation, University of South Florida.

106

CHAPTER FOUR

DESIGN AND SYNTHESIS OF A BAX PROTEIN α-HELIX MIMIC

4.1 Introduction 4.1.1 Pro-Apoptotic and Anti-Apoptotic Members of the Bcl-2 Family

Apoptosis was discussed in the Introduction section of Chapter Three regarding

the MDM2-p53 interaction. As shown in Figure 3.2, there are two apoptotic pathways:

the extrinsic (FAS) pathway, which involves death receptors; and the intrinsic

(mitochondrial) pathway, which involves cytochrome c release from the mitochondria as

a result of action by the Bcl-2 family of proteins (1). This chapter will discuss some of

the Bcl-2 family members, as well as ideas for manipulating them in order to promote

apoptosis in cancer cells.

The Bcl-2 family of proteins is comprised of three groups of pro-apoptotic and

anti-apoptotic members (2,3), several of which are shown in Figure 4.1. The top panel

shows the Bcl-2 group, which includes Bcl-xL and Bcl-2; these proteins are anti-apoptotic

(2). They share a sequence homology to Bcl-2 in four regions that are designated Bcl-2

homology one through four (BH1-BH4), as shown in Figure 4.1 (2). The pro-apoptotic

Bax group is shown in the middle panel; these proteins have three Bcl-2 homologous

regions numbered from BH1-BH3 (2). The Bik group members, shown at the bottom of

Figure 4.1, are another set of pro-apoptotic proteins that only share homology in the BH3

region (2).

107

Figure 4.1: The Bcl-2 Family of Proteins (2). The Bcl-2 group proteins are anti-apoptotic, while the Bax and Bik groups are both pro-apoptotic.

It is the interplay of various Bcl-2 family members that determines the

mitochondrial response, and ultimately the cell’s fate (3,4). The Bcl-2 family proteins

can form heterodimers between pro-apoptotic and anti-apoptotic members; this inhibits

the biological activity of their partners (2-4). Heterodimerization occurs via the BH3

region of a pro-apoptotic protein being inserted into a hydrophobic cleft comprised of the

BH1, BH2, and BH3 regions of an anti-apoptotic protein (2). The BH1, BH2, and BH4

regions are all necessary for anti-apoptotic activity, but BH3 alone is sufficient for pro-

apoptotic activity (2). Some of the pro-apoptotic family members like Bax and Bak can

also form homo-oligomers that allow for outer mitochondrial membrane permeabilization

in response to a loss of the mitochondrial proton gradient (3).

108

Originally, it was thought that any of the pro-apoptotic proteins could pair up with

any of the anti-apoptotic ones (4). However, it is now known that several of the Bcl-2

family members have specificities for certain partners and not for others (4). It appears

that many of the BH3-only pro-apoptotic family members interact with certain anti-

apoptotic partners, and that once these anti-apoptotic proteins are neutralized, the pro-

apoptotic Bax group of proteins are free to permeabilize the mitochondria and other

organelles (4). There are, however, some BH3-only pro-apoptotic proteins that can pair

with any of the anti-apoptotic proteins; one consequence of this differential specificity is

that the BH3-only proteins with more specificity are weaker cell killers in comparison to

the broader-spectrum BH3-only proteins (4). It has been suggested that the more specific

binding properties of some proteins may be due to their greater rigidity, while the more

general binding properties of other proteins occur because they are more flexible (4).

Several members of the Bcl-2 family can form channels in the mitochondrial

membrane, including Bax, Bcl-2, and Bcl-xL (5). The pro-apoptotic Bax presumably

forms channels to permit the escape of cytochrome c from the mitochondria, but it is not

known what the purpose is of the channels formed by the anti-apoptotic Bcl-2 proteins

(5). Presumably, their channel-forming activity relates to promoting cell survival (5).

It is not precisely known how pro-apoptotic family members like Bax are able to

cause mitochondrial rupture and cytochrome c release, but some of the details have been

worked out. One thing that is known is that the ability of Bax to form channels in the

mitochondria is pH-dependent (5). A proposed mechanism for the Bax channel-forming

process is shown in Figure 4.2 (6). There are several steps that are necessary before

apoptosis can occur.

109

Under normal, non-apoptotic conditions, a mitochondrion is characterized by a

proton gradient across its inner membrane. The cytochrome c molecules are located in

the intermembrane space on the mitochondrial inner membrane (MIM). Bax is soluble in

the cytoplasm of the cell. A protein called the permeability transition pore (PTP), which

spans both the MIM and the mitochondrial outer membrane (MOM) in its active form, is

inactive under normal conditions. This scenario is depicted in Panel A of Figure 4.2 (6).

Figure 4.2: Theorized Mechanism of Action of Bax (6). Panel A shows the normal mitochondrial membrane, with Bax solubilized in the cytoplasm and a proton gradient across the mitochondrial membrane. In Panel B, the proton gradient has been dissipated, which causes a conformational change in Bax and allows it to insert its 9th alpha helix into the mitochondrial membrane (Panel C). The Bax helices form channels that allow the cytochromes to escape into the cytoplasm (Panel D), thereby killing the cell.

110

The first thing that must happen in order for Bax-induced apoptosis to occur is

that the mitochondrial proton gradient must be lost. There are two proteins that comprise

the PTP, called the voltage-dependent anion channel (VDAC) and the adenine nucleotide

transporter (ANT). VDAC, which is in the MOM, and ANT, which is in the MIM,

associate to form the PTP, which spans both membranes (2). The PTP opens, allowing

protons to move down their electrochemical gradient and enter the mitochondrial matrix.

The proton gradient is thus dissipated (6), as shown in Panel B of Figure 4.2.

The loss of the mitochondrial proton gradient causes a conformational change in

Bax. Individual Bax proteins begin to insert their C-terminal α-helices into the MOM

(6). This is shown in Panel C of Figure 4.2. As the Bax proteins continue to insert into

the mitochondrion, they begin to oligomerize and form channels in the MOM. These

channels allow the cytochrome c proteins in the intermembrane space to pass through the

MOM into the cytoplasm, where they recruit the caspases that ultimately cause cellular

death (6). Panel D depicts the Bax oligomer channels allowing the cytochrome c

molecules to escape from the mitochondrion into the cytoplasm.

4.1.2 Mimicking the BH3 Domain of Pro-Apoptotic Proteins To date, there have been several examples of small molecules, natural products,

antisense nucleotides, peptides, and peptidomimetics that inhibit the Bcl-2 family protein

interactions reported in the literature (7-23). Generally, these compounds interfere with

the interactions between the pro-apoptotic and anti-apoptotic family members. For

example, the Bcl-2 antisense nucleotide ODN 2009 was found to induce apoptosis in

leukemia cells, and to work synergistically with the cytotoxic compound chlorambucil

(7). Several small molecules have been reported that inhibit the interactions between pro-

111

apoptotic BH3 domains and Bcl-xL (8) and Bcl-2 (9-11). These compounds were also

found to promote apoptosis in tumor cells; their structures are shown in Figure 4.3.

S

N

O

S

HO OC

Br

BH3I-1 (Ref. 8)

BrOH

ClO

NH

Cl

SO

O

Cl

BH3I-2 (Ref. 8)

O NH2

O

O

ON

Br

HA14-1 (Ref. 9)

Figure 4.3: Structures of Some Small-Molecule Bcl-2 Family Inhibitors (8, 9). The first two compounds inhibit Bcl-xL, and the third compound inhibits Bcl-2.

Alisol B acetate (Ref. 13)

NH

H

O

OMe O

NH

OO

OO O

O

2-Methoxyantimycin A (Ref. 12)

Figure 4.4: Structures of Two Pro-Apoptotic Natural Products (12, 13). These compounds are two natural products that were found to promote apoptosis.

There are also natural products that have been reported to promote apoptosis (12-

13). One is a methoxy derivative of antimycin A, shown on the right side of Figure 4.4.

This compound competes with the Bak BH3 peptide to bind to Bcl-2 and Bcl-xL (12).

112

Another pro-apoptotic compound that was recently reported is alisol B acetate, shown on

the left side of Figure 4.4. This compound was found to induce Bax up-regulation and

the insertion of Bax into the nuclear membrane (13). Alisol B acetate does not affect the

levels of the anti-apoptotic proteins Bcl-2 and Bcl-xL (13), suggesting that its mechanism

of action (up-regulation of Bax) is different than that of most of the previous compounds

discussed, which function by inhibiting the anti-apoptotic proteins.

It is known that the interface between the anti-apoptotic proteins and the pro-

apoptotic BH3 helices is not flat; in fact, the BH3 helices bind in a groove on their anti-

apoptotic partners (14). This fact makes the normally difficult process of finding a small-

molecule inhibitor of a protein-protein interaction seem more hopeful. However, all of

the small-molecule and natural product apoptosis promoters described thus far are active

in the low micromolar concentration range, which is considerably higher in comparison

to the natural BH3 helix ligands (14). In addition, it is known that the BH3 domain is

unstructured in solution, but it takes on a helical conformation when it binds to its anti-

apoptotic partner (14). Thus, it is reasonable to propose that rigid, α-helix-mimicking

compounds should have higher affinity and possibly higher specificity than the natural

ligands do, since their binding would not result in as large of an entropic penalty due to

the loss of conformational degrees of freedom (14).

One group reported a protein grafting method where the active BH3 amino acid

side chains were strategically placed on a rigid helical scaffold (15). This design was

then followed by a peptide selection process using phage display libraries where the

peptides that were the most stable and had the best binding were selected (15). The best

proteins were found to have dissociation constants in the nanomolar range that were even

113

superior to the BH3 helix itself (15). These authors also reported that the BH3-binding

pockets on Bcl-2 versus Bcl-xL are different enough to make it possible to develop

inhibitors specific to one protein over the other (14,15).

More recently, a hydrocarbon stapling strategy was reported where BH3 peptides

that had been chemically constrained into the helical binding shape were used to inhibit

Bcl-2 (16). These compounds were found to also be resistant to proteases and permeable

to cell membranes, as well as having improved affinity for Bcl-2 in comparison with the

native BH3 peptide (16). As with the grafted peptides, these compounds were able to

activate apoptosis in cancer cells (16).

Finally, it is also possible to inhibit the anti-apoptotic proteins using

peptidomimetics that are similar to the BH3 peptides, but that have entirely different

scaffolds (17-23). Several scaffolds have been reported, including oligoamide foldamers

(17), terphenyl derivatives (18-21), and terephthalamide derivatives (21-22). These

compounds are shown in Figures 4.5, 4.6, and 4.7 respectively.

Figure 4.5: Chemical and Stereo X-Ray Structure of an Oligoamide (17). This compound is flat, but the isopropyl ether side chains twist out of the plane at a 45° angle.

114

The oligoamide compounds were shown to have a flat backbone, but their side

chains twist out of the plane at a 45° angle, as shown in Figure 4.5 (17). These

compounds were found to have a low micromolar affinity for Bcl-xL, which is about an

order of magnitude worse than the binding affinity of the native BH3 peptide ligand (17).

Attempts to lengthen the backbone to four units were unsuccessful due to the insolubility

of the compound, and hydrolyzing the ester to improve solubility did not improve the

affinity of the longer peptidomimetic (17).

Figure 4.6: Terphenyl Derivatives with Binding Affinity for Bcl-xL (21). Panel A shows a schematic of how the terphenyl mimics the side chain orientation on the BH3 helix. The crystal structure of the terphenyl is overlaid with an α-helix in Panel B.

The use of a terphenyl backbone was even more successful (18-21). These so-

called “proteomimetics” have several advantages over the molecules previously

115

described, including their modular synthesis from subunits, their ability to project the side

chains in the same orientation as the native BH3 peptide, and their twisted shape that

mimics an α-helix (18). The phenyl rings in the backbone are able to rotate around the

single bonds between them, but the bulky side chains prevent them from being able to lie

flat (19). These compounds had binding constants for Bcl-xL in the high nanomolar

range (20). In the minimum energy conformation, the ortho substitution pattern on the

terphenyl backbone mimics the i, i+4, and i+7 side chains on the BH3 helix; these are the

residues that are important for BH3 to be able to interact with Bcl-xL (21). The structure

of these compounds is shown in Figure 4.6.

Figure 4.7: Terephthalamide Derivatives with Binding Affinity for Bcl-xL (23). Panel A shows the general structure of the terephthalamide derivatives. The overlap of a terephthalamide with an α-helix is shown in Panel B, while Panel C shows the overlap of a terphenyl compound with a terephthalamide. The authors note that the terphenyl derivatives were too hydrophobic to be water-

soluble and that they were difficult to synthesize (22). Thus, they developed a new

116

proteomimetic based on a terephthalamide scaffold, which is shown in Figure 4.7.

Basically, the two terminal phenyl rings of the terphenyl were replaced with carboxamide

groups, which also have a restricted geometry due to the planar amide bonds they contain

(22). This compound maintained the side chain geometry of the terphenyl compounds,

while increasing the hydrophilicity of the backbone (22). The terephthalamides were

tested against Bcl-xL and found to be high nanomolar inhibitors, similar to their terphenyl

predecessors (23).

All of the peptidomimetics described thus far function by affecting the pairings of

pro-apoptotic BH3 proteins with anti-apoptotic Bcl-2 family proteins. These compounds

affect the apoptotic pathway upstream of the Bax family members, which form the

membrane channels that release cytochrome c from the mitochondria. A search of the

literature did not turn up any compounds that directly affect Bax itself with the possible

exception of alisol B. The remainder of this chapter will discuss the effort to design and

synthesize an α-helix mimic that could bind to Bax itself, thus forcing Bax to adopt its

active, channel-forming conformation and ultimately inducing apoptosis.

4.2 Results and Discussion

The Bax protein is comprised of nine α-helices (24). The first eight helices

resemble the anti-apoptotic protein Bcl-xL, while the ninth C-terminal helix is involved

with Bax dimerization and mitochondrial pore formation (24). When Bax is in its

inactive conformation, the ninth helix is tucked into a hydrophobic pocket formed by the

rest of the protein (24). Early in the apoptosis process, helix 9 inserts itself into

mitochondrial membranes and forms the channels that permit cytochrome c to escape into

the cytoplasm (24). The structure of Bax is shown in Figure 4.8.

117

It is known that disruption of the inactive structure of Bax by detergents will

cause oligomer formation (24). Presumably, these oligomers would lead to membrane

penetration, cytochrome release, and apoptosis. We therefore propose a novel compound

that is hypothesized to disrupt the inactive conformation of Bax, leaving the hydrophobic

helix 9 free to insert into the mitochondrial membranes.

Figure 4.8: Structure of the Pro-Apoptotic Bax Protein. The apoptosis-inducing α-helix 9 can be seen at the top of the picture, horizontal and nearest to the viewer.

118

We began by virtually screening a library of 1981 NCI compounds (~2500 3D

models) in an effort to fit them into the hydrophobic binding site of helix 9 using the

GLIDE automated docking program. Since the binding site of the helix is relatively

large, we obtained compounds that docked into the C-terminal, N-terminal, and middle

portions of the Bax molecule. A sample of the compounds that were found to dock to

Bax is shown in Figure 4.9. An examination of these molecules shows that they have

several characteristics in common, including that they tend to be relatively symmetrical,

contain aromatic rings, and unsurprisingly tend to be relatively hydrophobic.

N N

O

HHO

NSC109816

N

N

O OH

O

HO

Cl

Cl

NSC210627

NN N

N NN

Cl

ClNSC343227

N

NN

NOH

NSC231643

Figure 4.9: Structures of Some NCI Compounds Virtually Screened to Fit into the Hydrophobic Pocket of Bax. The hit compounds tend to be relatively symmetrical, nonpolar, and aromatic.

119

We then began a series of iterations to build up these small molecules into larger

molecules that could occupy the entire helix 9 binding site. This required several rounds

of design and screening. Although we initially built upon all four of the compounds

shown in Figure 4.9, we discovered that the ligands with the best docking scores were

derivatives of compound NSC109816. We also found that the best compound 24

contained an additional central carbonyl, and that molecular symmetry was apparently

helpful. Some of the intermediate compounds that evolved between the starting

compound NSC109816 and the final compound 24 are shown in Figure 4.10.

Figure 4.10: Some Iterations to Obtain Compound 24 from NSC 109816. Each successive group of compounds was virtually screened, and the results were used to design the next set. Eventually, this process resulted in compound 24, the synthesis of which is discussed below. Based on the GLIDE docking simulations, compound 24 was predicted to have a

similar shape and binding orientation in comparison to helix 9, and to fit into the same

N N

O

H

HO

NSC109816

OH

OH

CH3

O

NHN

N

NHNH

NH

O

O

ONH

N

NH

O

NN

H H

NNO O

N N

N N

24

120

hydrophobic binding pocket. Although the compound structure appears to be flat, the

model predicts that 24 will exist in a twisted conformation, making its shape more

analogous to an α-helix. The structure of 24, its orientation according to the GLIDE

model, and helix 9 are all shown in Figure 4.11.

O

NN

H H

NNO O

N N

N N

24

Figure 4.11: Mimicking the Alpha Helix of Bax. The bottom panel shows the alpha helix from the Bax protein. Analogously, target compound 24 twists like an alpha helix, shown in the middle panel. The chemical structure of 24 is shown in the top panel.

A superimposition of compound 24 with the Bax protein minus helix 9 is shown

in Figure 4.12. This diagram shows that 24 is large enough to occupy most of the

121

hydrophobic helix 9 binding pocket. It also shows the twisted conformation of

compound 24.

Figure 4.12: Compound 24 Docked to Bax in Lieu of the Helix. Computer simulations predict that 24 will bind Bax in the same orientation and place where helix 9 normally folds.

122

The remainder of this discussion presents our progress toward the synthesis of

compound 24. Scheme 4.1 shows the retrosynthesis of 24. Since 24 is a symmetrical

compound, only one half of the molecule must be synthesized, and the two identical

pieces can then be joined together as a urea to make the entire compound. Each half of

24 was further broken down into an indole portion 25 and a benzoic acid portion 26.

These two pieces are theoretically accessible from available starting materials 27 and 29.

O

NN

H H

NNO O

N N

N N

24

N

NH2

HO

O

2

25 26

2N

NH2

Scheme 4.1: Retrosynthesis of Compound 24. 24 is a dimer requiring two pieces: indole derivate 25, and benzoic acid derivative 26. We began with the synthesis of 25 because it is the simpler piece of the molecule;

these reactions are shown in Scheme 4.2. The first step was the N-alkylation of 5-

nitromethylindole using sodium hydride and methyl iodide. This reaction is basically the

123

same as in our previous syntheses of N-methyl tryptophan and the small molecule

MDM2 inhibitors described in Chapter Three. The next step would be to reduce the nitro

group to an amine via hydrogenation. This is a well-known literature procedure (25-27),

but we have not yet performed it on compound 28 due to potential difficulties with

storing it.

25

N

NH2

28

N

NO2

27

N

H

NO2 NaH

CH3I

DMF

H2/Pd/C

MeOH

100%

AcOH

Scheme 4.2: Synthesis of Indole 25. Compound 25 could readily be synthesized from commercially available nitroindole 27. We did not complete the final step of the synthesis while we were attempting the synthesis of 26. We then turned our attention to synthesizing 26, the other half of the molecule.

Our initial plan was to begin with 2-methyl-5-nitroaniline 29 and reductively aminate it

using 2,5-dimethoxytetrahydrofuran 30 to form tertiary aniline 31. Compound 30 is a

diacetal that we planned to open by adding acid, after which we would proceed with our

normal reductive amination procedures as in Chapter Two. We would then oxidize the

124

methyl group up to the carboxylic acid, and reduce the nitro to the amine. Unfortunately,

several attempts at conducting a reductive amination were all unsuccessful, as shown in

Scheme 4.3. We initially tried using acetic acid, a weak acid. When that reaction failed,

we attempted to use p-toluenesulfonic acid. This reaction also failed, as did a third

attempt using 2N HCl.

NH2

NO2

29

OOO X

N

NO2

AcOH

1. MeOH

2. NaBH3CN

30 31

NH2

NO2

29

OOO X

N

NO2

1. MeOH

2. NaBH3CN

30 31

p-TsOH

NH2

NO2

29

OOO X

N

NO2

1. MeOH

2. NaBH3CN

30 31

2N HCl

Scheme 4.3: Failed Attempts to Synthesize 31 via Reductive Amination of 29. Several acids were tried, including acetic, catalytic p-toluenesulfonic, stoichiometric p-toluenesulfonic, and hydrochloric acids.

125

We next attempted to synthesize compound 31 via an N,N-dialkylation using 1,4-

dibromobutane. Our first attempt used DIEA in toluene following a literature procedure

(28), which was not successful in our hands. This failure is probably due to the high

electron-withdrawing capacity of the nitro group on the benzene ring of 29, which greatly

reduces the nucleophilicity of the aniline nitrogen. The reported literature examples

either had electron-donating groups or weaker electron-withdrawing groups attached to

the benzene ring, but no examples of aniline dialkylation with a nitrobenzene were

reported (28). However, when we attempted the reaction again using potassium

carbonate and the phase transfer catalyst TBAB (29), we were able to successfully obtain

31. These reactions are both shown in Scheme 4.4.

NH2

NO2

29

XN

NO231

BrBr

DIEA

NH2

NO2

29

N

NO231

BrBr

K2CO3

TBAB

59.3%

Scheme 4.4: Successful Synthesis of 31 via N, N-Dialkylation. The first attempt to synthesize 31 gave only starting materials, but addition of TBAB, a phase transfer catalyst, and use of potassium carbonate gave 31 as a bright orange solid.

126

N

NO2

31

XN

NO2

HO

O

32

KMnO4

H2O

34

29

NH2

NO2

X

33

KMnO4

H2O

NH2

NO2

HO

O

Cl

KMnO4

H2O

Cl

O OH

35

71.0%

Scheme 4.5: Unsuccessful Attempts to Oxidize 29 and 31 to the Carboxylic Acids. Regardless of whether the amine was tertiary or primary, the methyl groups in 29 and 31 could not be oxidized up to the corresponding acids in the presence of the amine. However, the oxidation reaction did proceed normally on toluene derivative 34, which lacks the amino group, to give acid 35. Having obtained the tertiary aniline 31, we next attempted to oxidize the methyl

group up to carboxylic acid 32 using potassium permanganate. This reaction is also a

127

common, well-known literature procedure (30, 31). However, we were not able to

successfully obtain product 32 after several attempts at oxidizing 31. Attempting the

procedure on the initial starting material primary aniline 29 was also unsuccessful. We

were, however, able to successfully oxidize p-chlorotoluene 34 to the corresponding

benzoic acid. This suggested that our reagents were fine, and that somehow the presence

of the aniline group was problematic. We were furthermore able to rule out the nitro

group as being the problem because one literature procedure reported the successful

oxidation of a nitrotoluene to the corresponding nitrobenzoic acid (29). These reactions

are shown in Scheme 4.5.

Since we were unable to oxidize the methyl group in the presence of the amine, it

was necessary to find an alternative route to synthesize 32. We found another literature

procedure where the methyl group had been oxidized in the presence of an acetylated

amine (32). Thus, our next strategy was to acetylate the aniline first, and then oxidize the

methyl to a carboxylic acid. This could be protected as an ester, after which the aniline

could be deacetylated and then reacted with 1,4-dibromobutane to give 32.

We were able to successfully synthesize the acetylated aniline 36 using acetic

anhydride and acetic acid (33,34), and oxidation of the methyl group with potassium

permanganate gave benzoic acid 37. We then protected 37 as the methyl ester using

methyl iodide to give 38, and hydrolyzed the amide back to the free aniline 39a using

HCl (35). However, when we attempted to perform the dialkylation reaction on 39a to

give tertiary aniline 39b, we found that no reaction would occur, even in the presence of

TBAB. Again, this is likely to be due to the presence of an additional electron-

withdrawing group on the benzene ring. Before the methyl group was oxidized, it was

128

mildly activating. However, once it was oxidized to the acid or ester, it became

deactivating. This series of reactions is shown in Scheme 4.6.

X

29

NH2

NO2

AcOAc

AcOH

N

NO2

H

O

3663.6%

KMnO4

H2O

MgSO4

N

NO2

H

O

O

OH

37

75.5%

K2CO3CH3I

100%DMF

N

NO2

H

O

O

O

38

NH2

NO2

O

O

39a

conc. HCl

77.8%EtOH

N

NO2

O

O

39b

BrBr

K2CO3

TBAB

Scheme 4.6: Unsuccessful Attempt to N,N-Dialkylate a Benzoic Ester. Acetylation of the amine allowed oxidation of the methyl group with KMnO4, but N,N-dialkylation of the ester was not successful due to the extremely low nucleophilicity of the aniline nitrogen.

129

At this point, we attempted to perform another reductive amination of starting

compound 29 using 2.5 M sulfuric acid and sodium borohydride in THF and methanol,

following a literature procedure (36). These reaction conditions were powerful enough to

allow the reductive amination of 29 to give compound 31 in good yield. However, when

we attempted to repeat this procedure on aniline 38, there was no reaction. Again, this is

likely to be due to the low nucleophilicity of 38, which possesses two powerful electron-

withdrawing groups attached to the benzene ring in contrast to the weakly electron-

donating methyl group that is present on 29. The procedure for reductive amination

using sulfuric acid is shown in Scheme 4.7.

X

29

NH2

NO2

NH2

NO2

O

O

38

N

NO2

O

O

39

OOO

30

THF/MeOH

2.5M H2SO4NaBH4

N

NO231

OOO

30THF/MeOH

2.5M H2SO4NaBH4

59.5%

Scheme 4.7: Unsuccessful Attempt to Perform Reductive Amination on 38. Although several earlier attempts to do reductive amination had failed, the reaction was finally performed successfully on 29 using 2.5M H2SO4. However, the reaction failed when it was attempted on compound 38.

130

X

29

NH2

NO2

15.2%

OOO

40

N

NO2

OO

41

dioxane

TEA

29

NH2

NO2

ClCl

O

O TEA N

NO2

OO

4142

reflux

DCM

29

NH2

NO2

OOO

40

N

NO2

OO

41

AcOH

30.3%

Scheme 4.8: Synthesis of Succinimide 41 from Aniline 29. Using succinyl chloride was messy and led to multiple products, but a small amount of product 41 could be obtained. Succinic anhydride gave the product if refluxed, along with a side product that was initially thought to be the incomplete cyclization product. This side product could not be cyclized using peptide coupling agents, and it was later determined to be the acetylated aniline 36. Our next strategy was to attempt to synthesize the succinimide 41, which would

be followed by oxidation of the methyl group to the carboxylic acid and then reduction of

131

the imide and nitro groups. We predicted that the succinimide, like the acetylated aniline,

should not interfere with the permanganate oxidation of the methyl group into the

carboxylic acid. Our first attempt to synthesize the succinimide by refluxing succinic

anhydride in dioxane was not successful. We then attempted to synthesize 41 using

succinyl chloride; this reaction did produce a small amount of product, but the yield was

poor and the reaction was messy and difficult to purify. Finally, we followed a literature

procedure (37) to make 41 from succinic anhydride by refluxing it in acetic acid. This

reaction turned out to be the most successful, and all of these procedures are shown in

Scheme 4.8.

reflux

29

NH2

NO2

OOO

40

N

NO2

OO

41

AcOH

30.3%

36

NH

NO2

O

~30%

Scheme 4.9: Side Product of the Succinimide Reaction. We concluded by NMR and TLC that the other major product of this reaction was acetanilide 36.

132

However, even the second succinic anhydride procedure was still relatively low-

yielding, and TLC showed the presence of a second major spot present in about the same

proportion as the desired product 41. NMR and TLC tests allowed us to conclude that

this side product was actually 36, the acetylated aniline that we had already synthesized

previously. This is not ultimately surprising, since the succinimide-forming conditions

are similar to the acetylating conditions (33,34). The literature yield for this reaction was

also relatively low at 36% (37). The reaction and both products formed are depicted in

Scheme 4.9.

At this point, we had obtained sufficient quantities of 41 to allow us to continue

with the synthesis. The separation of 41 from 36 is relatively difficult since the two

products run fairly close together on TLC, but we were able to separate them by

chromatographing them in dichloromethane. 41 has a higher Rf compared with 36, and

by using 100% DCM as the eluent, it is possible to collect 41 while 36 remains on the

silica. 36 can be isolated by adding a small percentage of methanol to the DCM. 41 was

then oxidized to carboxylic acid 42. This reaction proceeded successfully as we expected

based on our previous experience with oxidizing the acetylated aniline.

The next step in the synthesis is the reduction of the acid, imide, and nitro groups

using lithium aluminum hydride to the primary alcohol, tertiary amine, and primary

amine, respectively. Then, the alcohol will be oxidized back to the carboxylic acid using

Jones reagent to give 26, which can then be combined with 25 to give 44, the

symmetrical half of 24. These reactions are shown in Scheme 4.10. Finally, the free

amines on 44 can be connected together as a urea to give target molecule 24, as discussed

previously.

133

N

NH2

HO

26

N

NO2

OO

41

KMnO4

H2O

N

NO2

OO

O

HO

43

25

N

H2N

44

N

NH2

O

N

N

H

THF

LiAlH4 Jones

reagent

N

NH2

HO

O

26a

63.6%

Scheme 4.10: Proposed Synthesis of 44 from 41. Compound 44 would then be used to synthesize target compound 24. The synthesis of 43 was successfully completed.

Should this series of reactions not be successful, we propose to attempt an

alternative route where we would form the pyrrolidine either by dialkylating the aniline

ring of 29 using 1,4-dibromobutane or by reductive amination. We would then convert

134

the methyl group to the primary alcohol using wet NBS, followed by oxidation of alcohol

45 to acid 46 with Jones reagent. 46 could then be hydrogenated to give intermediate 26,

as shown in Scheme 4.11. The first reaction in this procedure to form 31 has already

been successfully accomplished as described previously.

NO2

NH2

29

BrBr

NO2

NNBS

CHCl3H2O

NO2

N

HO

Jones

reagent

NO2

N

HO

O

H2/Pd/C

NH2

N

HO

O

31

45 46

26

Scheme 4.11: Alternative Proposed Route to Compound 26. This route will be attempted next if the proposed route in Scheme 4.10 is unsuccessful.

135

In conclusion, we have successfully used GLIDE docking simulations to virtually

screen a library of NCI compounds, and we developed target compound 24 based on

multiple screening iterations. We then made significant progress on the synthesis of 24.

Further work will entail completing the synthesis described above, as well as performing

biological testing on 24.

Experimental Data:

28

N

NO2

Procedure for Synthesis of N-Methyl-5-Nitroindole (28). Sodium hydride (0.925

mmol) was washed twice with dry DMF, and then suspended in a third aliquot of dry

DMF. The reaction mixture was cooled to 0° C, and 5-nitroindole (0.671 mmol) was

added slowly. The reaction was then warmed up to room temperature and stirred for 15

minutes under Ar, after which methyl iodide (0.678 mmol) was added dropwise via

syringe. The reaction was stirred for thirty minutes at RT, and TLC then showed that it

was complete. Ethyl acetate was added to the reaction, and it was then washed with dI

water. The water was back-extracted with ethyl acetate, and the combined ethyl acetate

layers were washed once with water and twice with brine, dried over MgSO4, and

evaporated under reduced pressure to give 28, a yellow solid, in 100% yield. MP = 160-

162 °C; Rf = 0.32 (3:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.85 (s, 3H), 6.66

(d, J = 2.8 Hz, 1H), 7.20 (d, J = 2.8 Hz, 1H), 7.32 (d, J = 8.8 Hz, 1H), 8.10 (d, J = 9.2 Hz,

136

1H), 8.56 (s, 1H); 13C NMR (100 MHz, CDCl3) δ 33.43, 103.95, 109.22, 117.33, 118.25,

127.74, 132.18, 139.55, 141.67; HR ESI-MS (0.1% AcOH in CH3CN) 177.06553 (M +

H)+; (The compound begins to decompose above 100 °C.)

N

NO231

Procedure for Dialkylation Synthesis of 2-Pyrollidino-4-Nitrotoluene (31). 1-

methyl-5-nitroaniline (0.657 mmol) was dissolved in 5 mL toluene, and 1,4-

dibromobutane (0.657 mmol), potassium carbonate (0.657 mmol), and TBAB (0.066

mmol) were added. The reaction was then refluxed overnight and worked up by washing

it with 2N HCl and back-extracting the aqueous layer with DCM. The DCM was back-

extracted with 2N HCl three times, and then all of the aqueous layers were combined and

basified with solid NaOH. The aqueous layer was extracted five times with DCM, and

these five organic layers were combined, dried over magnesium sulfate, and evaporated

under reduced pressure. Column chromatography using 9:1 hexanes/ethyl acetate gave

31, an orange solid, in 59.3% yield. MP = 73-74 °C; Rf = 0.68 (3:1 hexane/EtOAc); 1H

NMR (400 MHz, CDCl3) δ 1.96-2.00 (m, 4H), 2.40 (s, 3H), 3.27-3.29 (m, 4H), 7.18 (d, J

= 8.0 Hz, 1H), 7.62-7.64 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 21.48, 25.48, 109.98,

114.58, 132.22, 135.52, 147.16, 150.35; ESI-MS (0.1% AcOH in CH3CN) 207.1 (M +

H)+.

137

N

NO231

Procedure for Reductive Amination Synthesis of 2-Pyrollidino-4-Nitrotoluene (31).

2,5-dimethoxytetrahydrofuran (2.32 mmol) was dissolved in THF, and reacted with 2.5

M sulfuric acid (4.50 mmol). The reaction mixture was cooled to 0 °C and stirred for 15

minutes. In a separate flask, 1-methyl-5-nitroaniline (1.77 mmol) was dissolved in a 1:1

methanol/THF mixture and cooled to 0 °C. The acidic acetal solution was then placed

into an addition funnel, and this solution was added to the aniline reaction mixture

dropwise. The combined reaction mixture was stirred at 0 °C for 15 minutes, and then

sodium borohydride was added from a powder addition funnel over twenty minutes. The

reaction was then warmed up to room temperature and stirred for an hour, after which it

was quenched with NaOH and water, and extracted with DCM. Chromatography in 9:1

hexanes/ethyl acetate gave 31 as a bright orange powdery solid in 59.5% yield. MP = 74-

75 °C; Rf = 0.63 (3:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 1.96-1.99 (m, 4H),

2.40 (s, 3H), 3.27-3.30 (m, 4H), 7.17 (d, J = 8.0 Hz, 1H), 7.60 (s, 1H), 7.62 (d, J = 8.0

Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 21.48, 25.48, 51.30, 109.99, 114.58, 132.22,

135.52, 150.35; ESI-MS (0.1% AcOH in CH3CN) 207.1 (M + H)+.

138

N

NO2

H

O

36

Procedure for Synthesis of N-Acetyl-2-Amino-4-Nitrotoluene (36). Aniline 29

(0.0329 mol) was dissolved in 5 mL acetic acid, to which was added 8 mL acetic

anhydride. The flask was placed in a water bath at 45 °C, causing it to solidify. 25 mL

chloroform was added, and the reaction was stirred at 45 °C for 10 minutes, until the SM

had dissolved to form a reddish-brown solution. The chloroform was removed under

reduced pressure, and the product was recrystallized from a 1:1 solution of ethanol:water

to give white, flowery crystals in a 68.6% yield. MP = 149-150.5 °C; Rf = 0.56 (3:1

hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 2.27 (s, 3H), 2.37 (s, 3H), 7.07 (broad,

1H), 7.34 (d, J = 8.8 Hz, 1H), 7.94 (d, J = 8.0 Hz, 1H), 8.78 (s, 1H); 13C NMR (100

MHz, CDCl3) δ 18.12, 24.31, 118.05, 119.78, 130.97, 136.35, 136.48, 146.76, 168.57;

ESI-MS (0.1% AcOH in CH3CN) 193.0 (M - H)-.

N

NO2

H

O

O

OH

37

Procedure for Synthesis of N-Acetyl-2-Amino-4-Nitrobenzoic Acid (37). Acetyl

aniline 36 (0.476 mmol) was suspended in 10 mL dI water, to which magnesium sulfate

139

(0.297 mmol) was added. The reaction flask was fitted with a condenser, and heated to

80 °C, after which potassium permanganate (1.19 mmol) was added. The reaction was

stirred at 85 °C for one hour, then cooled to RT. Small portions of sodium bisulfite and

concentrated sulfuric acid were added until the solid manganese dioxide byproduct

dissolved and the product 37 precipitated out of solution as a pale peachy yellow powder

in 75.5 % yield. MP = 197-198 °C; Rf = 0.22 (100% EtOAc); 1H NMR (400 MHz,

CD3OD) δ 2.24 (s, 3H), 7.90-7.93 (m, 1H), 8.23-8.26 (m, 1H), 9.44 (s, 1H); 13C NMR

(100 MHz, CD3OD) δ 23.84, 114.55, 116.60, 119.97, 131.13, 132.69, 142.10, 169.53;

ESI-MS (0.1% AcOH in CH3CN) 223.0 (M - H)-.

N

NO2

H

O

O

O

38

Procedure for Synthesis of Methyl N-Acetyl-2-Amino-4-Nitrobenzoate (38).

Benzoic acid 37 (0.416 mmol) was dissolved in 5 mL DMF, and the reaction was cooled

to 0 °C. Potassium carbonate (0.458 mmol) was added, and the reaction was stirred for

ten minutes at 0 °C before adding methyl iodide (0.833 mmol). Then the reaction was

allowed to warm to RT and stirred for four hours at RT. The reaction was diluted with

ethyl acetate and washed with saturated sodium bicarbonate solution, dI water, and brine

solution to give 37 as a yellow oil in 76.9% yield. Rf = 0.31 (3:1 hexane/EtOAc); 1H

NMR (400 MHz, CDCl3) δ 2.29 (s, 3H), 4.00 (s, 3H), 7.87 (dd, J = 2.4 Hz, 8.8 Hz, 1H),

140

8.19 (d, J = 8.8 Hz, 1H), 9.59 (d, J = 2.0 Hz, 1H); 11.12 (broad, 1H); 13C NMR (100

MHz, CDCl3) δ 25.45, 53.12, 115.26, 116.50, 119.05, 131.94, 142.40, 151.26, 167.36,

169.28.

NH2

NO2

O

O

39a

Procedure for Synthesis of Methyl 2-Amino-4-Nitrobenzoate (39a). Acetyl aniline

38 (0.323 mmol) was suspended in 7 mL absolute ethanol, to which was added 1.4 mL

concentrated HCl. The reaction was refluxed for 18 hours at 79 °C, after which it was

soluble in ethanol. The ethanol was removed under reduced pressure, and the crude

product was suspended in dI water and basified with 1M NaOH until the pH was 6 to

give a gelatinous yellow solid in 77.8 % yield. Rf = 0.00 (3:1 hexane/EtOAc); 1H NMR

(400 MHz, CDCl3) δ 3.92 (s, 3H), 6.06 (broad, 2H), 7.40 (d, J = 7.6 Hz, 1H), 7.51 (s,

1H), 8.01 (d, J = 7.8 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 52.42, 110.30, 111.35,

115.16, 133.05, 150.91, 151.58, 167.50; ESI-MS (0.1% AcOH in CH3CN) 197.0 (M +

H)+.

N

NO2

OO

41

141

Procedure for Succinyl Chloride Synthesis of 2-Succinimido-4-Nitrotoluene (41).

Aniline 29 (3.29 mmol) was dissolved in 25 mL DCM, to which was added succinyl

chloride (3.61 mmol) and TEA (6.57 mmol). The reaction was stirred at RT for one hour,

after which the solid byproduct formed was filtered off, and the supernatant was

chromatographed on silica in 2:1 hexanes/EtOAc to give 41 as pale yellow crystals in a

15.2% yield. MP = 214-217 °C; Rf = 0.29 (1:1 hexane/EtOAc); 1H NMR (400 MHz,

CDCl3) δ 2.27 (s, 3H), 2.99 (s, 4H), 7.51 (d, J = 8.8 Hz, 1H), 8.02 (d, J = 2.4 Hz, 1H),

8.22 (dd, J = 2.4 Hz, 8.4 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 18.49, 28.90, 124.11,

124.48, 129.06, 131.14, 132.18, 144.08, 175.56.

Procedure for Succinic Anhydride Synthesis of 2-Succinimido-4-Nitrotoluene (41)

and Acetylated Side Product (36). Aniline 29 (3.29 mmol) and succinic anhydride

(6.57 mmol) were dissolved in glacial acetic acid, and the reaction was refluxed for two

hours. It was then cooled to RT, and extracted with saturated bicarbonate solution. The

crude product was chromatographed on silica in 1:1 hexanes/EtOAc to give succinimide

product 41 as pale yellow crystals in 30.3% yield, and side product acetyl aniline 36 as a

white powdery solid in 30% yield.

N

NO2

OO

41

142

2-Succinimido-4-Nitrotoluene (41). MP = 215-218 °C; Rf = 0.30 (9:1 DCM/MeOH),

Rf = 0.33 (1:1 hexane/EtOAc); 1H NMR (400 MHz, CDCl3) δ 2.27 (s, 3H), 2.99 (s, 4H),

7.52 (d, J = 8.8 Hz, 1H), 8.02 (d, J = 2.4 Hz, 1H), 8.22 (dd, J = 2.4 Hz, 8.4 Hz); 13C

NMR (100 MHz, CDCl3) δ 18.47, 28.90, 124.11, 124.47, 132.17, 144.10, 146.99, 175.56.

N

NO2

H

O

36

N-Acetyl-2-Amino-4-Nitrotoluene (36). MP = 149-150 °C; Rf = 0.12 (9:1 DCM

/MeOH); 1H NMR (400 MHz, CDCl3) δ 2.26 (s, 3H), 2.37 (s, 3H), 7.16 (broad, 1H), 7.34

(d, J = 8.8 Hz, 1H), 7.93 (d, J = 8.8 Hz, 1H), 8.75 (s, 1H); 13C NMR (100 MHz, CDCl3) δ

18.33, 24.57, 118.20, 120.00, 131.18, 136.26, 136.71, 147.10, 168.58; ESI-MS (0.1%

AcOH in CH3CN) 193.0 (M - H)-.

N

NO2

OO

O

HO

43

2-Succinimido-4-Nitrobenzoic Acid (43). MP = 145-146 °C; Rf = 0.00 (100% DCM);

1H NMR (400 MHz, DMSO) δ 2.56-2.59 (m, 2H), 2.67-2.70 (m, 2H), 7.95 (dd, J = 2.0

Hz, 8.4 Hz, 1H), 8.20 (d, J = 8.4 Hz, 1H), 9.30 (d, J = 2.4 Hz, 1H), 11.22 (broad, 1H);

143

13C NMR (100 MHz, DMSO) δ 29.25, 32.78, 114.97, 117.52, 124.54, 133.31, 141.79,

150.82, 168.77, 171.74, 174.19.

4.3 References 1. Fesik, S. W. (2005) Promoting apoptosis as a strategy for cancer drug discovery.

Nature Reviews Cancer, 5(11), 876-885. 2. Tsujimoto, Y., Shimizu, S. (2000) Bcl-2 family: Life-or-death switch, FEBS Letters,

466, 6-10. 3. Breckenridge, D. G., Xue, D. (2004) Regulation of mitochondrial membrane

permeabilization by Bcl-2 family proteins and caspases, Current Opinion in Cell Biology, 16, 647-652.

4. Chen, L., Willis, S. N., Wei, A., Smith, B. J., Fletcher, J. I., Hinds, M. G., Colman, P.

M., Day, C. L., Adams, J. M., Huang, D., C., S. (2005) Differential Targeting of Pro-Survival Bcl-2 Proteins by Their BH3-Only Ligands Allows Complementary Apoptotic Function, Molecular Cell, 17, 393-403.

5. Antonsson, B., Conti, F., Ciavatta, A. M., Montessuit, S., Lewis, S., Martinou, I.,

Bernasconi, L., Bernard, A., Mermod, J. J., Mazzei, G., Maundrell, K., Gambale, F., Sadoul, R., Martinou, J. C. (1997) Inhibition of Bax Channel-Forming Activity by Bcl-2, Science, 277, 370-372.

6. de Giorgi, F., Lartigue, L., Bauer, M. K. A., Schubert, A., Grimm, S., Hanson, G. T.,

Remington, J. S., Youle, R. J., Ichas, F. (2002) The permeability transition pore signals apoptosis by directing Bax translocation and multimerization, The FASEB Journal, 16, 607-610.

7. Pepper, C., Hooper, K., Thomas, A., Hoy, T., Bentley, P. (2001) Bcl-2 Antisense

Oligonucleotides Enhance the Cytotoxicity of Chlorambucil in B-Cell Chronic Lymphocytic Leukaemia Cells, Leukemia and Lymphoma, 42, 491-498.

8. Degterev, A., Lugovskoy, A., Cardone, M., Mulley, B., Wagner, G., Mitchison, T.,

Yuan, J. (2001) Identification of small-molecule inhibitors of interaction between the BH3 domain and Bcl-xL, Nature Cell Biology, 3, 173-182.

9. Wang, J. L., Liu, D., Zhang, Z. J., Shan, S., Han, X., Srinivasula, S. M., Croce, C. M.,

Alnermri, E. S. Huang, Z. (2000) Structure-based discovery of an organic compound that binds Bcl-2 protein and induces apoptosis of tumor cells, Proceedings of the National Academy of Science, 97, 7124-7129.

144

10. Chen, J., Freeman, A., Liu, J., Dai, Q., Lee, R. M. (2002) The Apoptotic Effect of HA14-1, a Bcl-2-interacting Small Molecular Compound, Requires Bax Translocation and is Enhanced by PK11195, Molecular Cancer Therapeutics, 1, 981-987.

11. Sinicrope, F. A., Penington, R. C., Tang, X. M. (2004) Tumor Necrosis Factor-

Related Apoptosis-Inducing Ligand-Induced Apoptosis Is Inhibited by Bcl-2 but Restored by the Small Molecule Bcl-2 Inhibitor, HA14-1, in Human Colon Cancer Cells, Clinical Cancer Research, 10, 8284-8292.

12. Tzung, S. P., Kim, K. M., Basanez, G., Giedt, C., D., Simon, J., Zimmerberg, J.,

Zhang, K. Y. J., Hockenbery, D., M. (2001) Antimycin A mimics a cell-death-inducing Bcl-2 homology domain 3, Nature Cell Biology, 3, 183-191.

13. Huang, Y. T., Huang, D. M., Chueh, S. C., Teng, C. M., Guh, J. H. (2006) Alisol B

acetate, a triterpene from Alismatis rhizoma, induces Bax nuclear translocation and apoptosis in human hormone-resistant prostate cancer PC-3 cells, Cancer Letters, 231, 270-278.

14. Rutledge, S. E., Chin, J. W., Schepartz, A. (2002) A view to a kill: ligands for Bcl-2

family proteins, Current Opinion in Chemical Biology, 6, 479-485. 15. Chin, J. W., Schepartz, A. (2001) Design and Evolution of a Miniature Bcl-2

Binding Protein, Angewandte Chemie International Edition, 40, 3806-3809. 16. Walensky, L. D., Kung, A. L, Escher, I., Malia, T. J., Barbuto, S., Wright, R. D.,

Wagner, G., Verdine, G. L., Korsmeyer, S. J. (2004) Activation of Apoptosis in Vivo by a Hydrocarbon-Stapled BH3 Helix, Science, 305, 1466-1470.

17. Ernst, J. T., Becerril, J., Park, H. S., Yin, H. Hamilton, A. D. (2003) Design and

Application of an α-Helix-Mimetic Scaffold Based on an Oligoamide-Foldamer Strategy: Antagonism of the Bak BH3/Bcl-xL Complex, Angewandte Chemie International Edition, 42, 535-539.

18. Orner, B. P., Ernst, J. T., Hamilton, A. D. (2001) Toward Proteomimetics:

Terphenyl Derivatives as Structural and Functional Mimics of Extended Regions of an α-Helix, Journal of the American Chemical Society, 123, 5382-5383.

19. Ernst, J. T., Kutzki, O., Debnath, A. K. , Jiang, S., Lu, H., Hamilton, A. (2002)

Design of a Protein Surface Antagonist Based on α-Helix Mimicry: Inhibition of gp41 Assembly and Viral Fusion, Angewandte Chemie International Edition, 41, 278-281.

145

20. Kutzki, O., Park, H. S., Ernst, J. T., Orner, B. P., Yin, H., Hamilton, A. D. (2002) Development of a Potent Bcl-xL Antagonist Based on α-Helix Mimicry, Journal of the American Chemical Society, 124, 11838-11839.

21. Yin, H., Lee, G., Sedey, K. A., Kutzki, O., Park, H. S., Ornder, B. P., Ernst, J. T.,

Wang, H. G., Sebti, S. M., Hamilton, A. D. (2005) Terphenyl-Based Bak BH3 α-Helical Proteomimetics as Low-Molecular-Weight Antagonists of Bcl-xL, J. Am. Chem. Soc., 127, 10191-10196.

22. Yin, H., Hamilton, A. D. (2004) Terephthalamide derivatives as mimetics of the

helical region of Bak peptide target Bcl-xL protein, Bioorganic and Medicinal Chemistry Letters, 14, 1375-1379.

23. Yin, H., Lee, G., Sedey, K. A., Rodriguez, J., Wang, H. G., Sebti, S. M., Hamilton, A.

D. (2005) Terephthalamide Derivatives as Mimetics of Helical Peptides: Disruption of the Bcl-xL/Bak Interaction. J. Am. Chem. Soc., 127, 5463-5468.

24. Suzuki, M., Youle, R. J., Tjandra, N. (2000) Structure of Bax: Coregulation of

Dimer Formation and Intracellular Localization, Cell, 103, 645-654. 25. Matassa, V. G., Maduskuie, T. P., Shapiro, H. S., Hesp, B., Snyder, D. W., Aharony,

D., Krell, R. D., Keith, R. A. (1990) Journal of Medicinal Chemistry, 33, 1781-1790.

26. Tietze, L. F., Haunert, F., Feurstein, T., Herzig, T. (2003) A Concise and Efficient

Synthesis of seco-Duocarmycin SA, European Journal of Organic Chemistry, 562-566.

27. Baraldi, P. G., Romagnoli, R., Beria, I., Cozzi, P., Geroni, C., Mongelli, N., Bianchi,

N., Mischiati, C., Gambari, R. (2000) Synthesis and Antitumor Activity of New Benzoheterocyclic Derivatives of Distamycin A, Journal of Medicinal Chemistry, 43, 2675-2684.

28. Verboom, W., Lammerink. B. H. M., Egberink. J. M., Reinhoudt, D. N., Harkema, S.

(1985) Synthesis of Mitomycin C Analogues. 2. Introduction of a Leaving Group at C-1 and Oxidation of the Aromatic Ring in 2,3,9,9a-Tetrahydro-1H-pyrrolo[1,2-a]indoles, Journal of Organic Chemistry, 50, 3797-3806.

29. Huang, Y. L., Lin, C. F., Lee, Y. J., Li, W. W., Chao, T. C., Bacherikov, V. A., Chen,

K. T., Chen, C. M., Su, T. L. (2003) Non-classical Antifolates, 5-(N-Phenylpyrrolidin-3-yl)-2,4,6-triaminopyrimidines and 2,4-Diamino-6(5H)-oxopyrimidines, Synthesis and Antitumor Studies, Bioorganic and Medicinal Chemistry, 11, 145-157.

146

30. Leonard, N. J., Boyd, S. N. (1946) Cinnolines. I. Synthesis of Aminoacetophenones and Aminopropiophenones, Journal of Organic Chemistry, 11, 405-418.

31. Adams, R., Gold, M. H. (1940) The Synthesis of 1,3-

Dephenyldihydroisobenzofurans, 1,3-Diphenylisobenzofurans, and o-Dibenzoylbenzenes from the Diene Addition Products to Dibenzoylethylene, Journal of the American Chemical Society, 62, 56-61.

32. Rogister, F., Laeckmann, D., Plasman, P. O., Van Eylen, F., Ghyoot, M., Maggetto,

C., Liegeois, J. F., Geczy, J., Herchuelz, A., Delarge, J., Masereel, B. (2001) Novel inhibitors of the sodium-calcium exchanger: benzene ring analogues of N-guanidino substituted amiloride derivatives, European Journal of Medicinal Chemistry, 36, 597-614.

33. Glinsukon, T., Weisburger, E. K., Benjamin, T., Roller, P. P. (1975) Preparation and

Spectra of Some Acetyl Derivatives of 2,4-Toluenediamine, Journal of Chemical and Engineering Data, 20, 207-209.

34. DeRuiter, J., Swearingen, B. E., Wandrekar, V., Mayfield, C. A. (1989) Synthesis

and in Vitro Aldose Reductase Inhibitory Activity of Compounds Containing an N-Acylglycine Moiety, Journal of Medicinal Chemistry, 32, 1033-1038.

35. Zhou, Z. L., Kher, S. M., Cai, S. X., Whittemore, E. R., Espitia, S. A., Hawkinson, J.

E., Tran, M., Woodward, R. M., Weber, E., Keana, J. F. W. (2003) Synthesis and SAR of Novel Di- and Trisubstituted 1,4-Dihydroquinoxaline-2,3-diones Related to Licostinel (Acea 1021) as NMDA/Glycine Site Antagonists, Bioorganic and Medicinal Chemistry, 11, 1769-1780.

36. Verardo, G., Dolce, A., Toniutti, N. (1999) Reductive One Batch Synthesis of N-

Substituted Pyrrolidines from Primary Amines and 2,5-Demethoxytetrahydrofuran, Synthesis, 1, 74-79.

37. Marshalkin, V. N., Brovko, D. A., Samet, A. V., Semenov, V. V. (2001) Reaction of

Polynitrotoluenes with Phthalic Anhydride, Russian Journal of Organic Chemistry, 37, 1244-1248.

147

CHAPTER FIVE

APPENDIX: PROTON AND CARBON NMR SPECTRA

148

Spectrum 5.1

NOH

H O

O

1c

149

Spectrum 5.2

NOH

H O

O

1d

150

Spectrum 5.3

1e

NOH

H O

O

151

Spectrum 5.4

2a

NH

O

OCH3

152

Spectrum 5.5

NO

H O

N OO

2b

153

Spectrum 5.6

NO

H O

O2c

154

Spectrum 5.7

NO

H O

O

2d

155

Spectrum 5.8

NO

H O

O

2e

156

Spectrum 5.9

NO

N O

O

3a

157

Spectrum 5.10

NO

N O

N OO

O

3b

158

Spectrum 5.11

NO

N O

O

O

3c

159

Spectrum 5.12

NO

N O

O

O

3d

160

Spectrum 5.13

NO

N O

O

O

3e

161

Spectrum 5.14

NO

N O

H Boc

5a

162

Spectrum 5.15

NOBnz

O

N OO

NH Boc

5b

163

Spectrum 5.16

NO

N O

H Boc

O

5c

164

Spectrum 5.17

NO

N O

H Boc

O

5d

165

Spectrum 5.18

NO

N O

O

BocH

5e

166

Spectrum 5.19

NOH

N O

BocH

6a

167

Spectrum 5.20

NOH

N O

BocH

O

6c

168

Spectrum 5.21

NOH

N O

BocH

O

6d

169

Spectrum 5.22

NOH

N O

O

BocH

6e

170

Spectrum 5.23

NOH

N O

BocH

6f

171

Spectrum 5.24

NOH

N O

BocH

6g

172

Spectrum 5.25

NOH

N O

BocH

6h

173

Spectrum 5.26

NOH

N O

BocH

S6i

174

Spectrum 5.27

9

NOBoc

H O

N OO

175

Spectrum 5.28

12b

NOBoc

H O

NH Cbz

176

Spectrum 5.29

NOZ

H O

O

12c

177

Spectrum 5.30

NOZ

H O

O

12d

178

Spectrum 5.31

12e

NOZ

H O

O

179

Spectrum 5.32

NO

O

OH

13a

180

Spectrum 5.33

13b

NO

H O

O

181

Spectrum 5.34

14b

HN

O

OH

N H

182

Spectrum 5.35

14c

HN

O

OH

N

183

Spectrum 5.36

ON

O

O OH

15a

184

Spectrum 5.37

15b

NO

OH

N H

O

O

185

Spectrum 5.38

15c

NO

OH

N

O

O

186

Spectrum 5.39

ON

O

O ONH

O O

16a

187

Spectrum 5.40

17a

HON

OH

O ONH

O O

188

Spectrum 5.41

19a

NOCbz

H O

N H

189

Spectrum 5.42

19b

NOCbz

H O

N

190

Spectrum 5.43

20

HON

O

O ONH

O O

191

Spectrum 5.44

N

23a

192

Spectrum 5.45

N

23b

193

Spectrum 5.46

N

23c

194

Spectrum 5.47

28

N

NO2

195

Spectrum 5.48

N

NO2

31

196

Spectrum 5.49

N

NO2

H

O

36

197

Spectrum 5.50

N

NO2

H

O

O

OH

37

198

Spectrum 5.51

N

NO2

H

O

O

O

38

199

Spectrum 5.52

NH2

NO2

O

O

39a

200

Spectrum 5.53

N

NO2

OO

41

201

Spectrum 5.54

N

NO2

OO

O

HO

43

ABOUT THE AUTHOR

Stephanie Tara Weiss was born in Detroit, Michigan and grew up in Plantation,

Florida. After graduating from South Plantation High School in 1993, she attended New

College of Florida, where she majored in natural sciences and Spanish. She earned her

B.A. there in 1997, and then went to the University of Alabama at Birmingham, where in

2001 she earned her M.S. in pharmaceutical design and medicinal chemistry under the

supervision of Dr. Wayne Brouillette. In the spring of 2002, she began her graduate

studies at the University of South Florida, where she joined the lab of Professor Mark L.

McLaughlin at the Moffitt Cancer Center and Research Institute. She will receive her

Ph.D. in organic chemistry with a focus on peptidomimetics in May 2006.