ABSTRACT Design and Synthesis of Small-Molecule Inhibitors Of

ABSTRACT

Design and Synthesis of Small-Molecule Inhibitors of Cathepsin L Bearing the Thiosemicarbazone Warhead as Potential Anti-Metastatic Agents

Erica N. Parker, Ph.D.

Mentor: Kevin G. Pinney, Ph.D.

Metastasis is a devastating component associated with tumor progression which claims the majority of lives lost to cancer. Upregulation of cathepsin L, a powerful cysteine protease capable of degrading components in the extracellular matrix, is correlated with increased invasiveness of cancer cells and poor prognosis. Consequently, cathepsin L has emerged as a target for the development of new anticancer agents designed to mitigate the invasion and metastasis of malignant cancer cells. The development of a focused, small library of thiosemicarbazone analogues was pursued in an effort to discover potent and selective inhibitors of cathepsin L, which possess the potential to bind to the active site of cathepsin L through a proposed covalent interaction with the catalytic amino acid residue Cys25. Incorporation of the thiosemicarbazone moiety with a variety of aryl based molecular scaffolds led to a functionally diverse series of thiosemicarbazone analogues that were evaluated, through collaborative efforts, for their ability to inhibit cathepsin L and thus to potentially function as anti-metastatic agents. Benzophenone, benzoylbenzophenone, di-2-naphthalenylmethanone, 1,3-

diphenylpropan-2-one, dibenzylideneacetone, and 1,5-diphenyl-3-pentanone aryl-based

molecular scaffolds were all explored. From the thirty-five compounds in this

thiosemicarbazone based library, fifteen analogues were determined to be potent

inhibitors of cathepsin L (IC50 < 10 μM). 1,3-Bis(2-fluorobenzoyl)-5-bromobenzene

thiosemicarbazone (KGP312) demonstrated the greatest inhibitory activity against

cathepsin L with an IC50 value of 8.1 nM. Furthermore, KGP312 inhibited the invasion

and migration of both MDA-MB-231 breast cancer cells and PC-3ML prostate cancer

cells. In addition to the design and synthesis of new thiosemicarbazone analogues,

improvement of the synthetic route for our previously discovered lead cathepsin L

inhibitor, KGP94, was achieved. Derivatization designed to increase aqueous solubility

and potentially oral bioavailability was performed and ultimately led to the preparation of

KGP420, the phosphate prodrug of KGP94. Owing to high cathepsin L inhibitory activity

and favorable outcomes in initial biological studies, on-going collaborative efforts have

been aimed towards the advancement of these preclinical candidates as potential anti- metastatic agents.

Design and Synthesis of Small-Molecule Inhibitors of Cathepsin L Bearing the Thiosemicarbazone Warhead as Potential Anti-Metastatic Agents

Erica N. Parker, B.S.

A Dissertation

Approved by the Department of Chemistry and Biochemistry

Patrick J. Farmer, Ph.D., Chairperson

Submitted to the Graduate Faculty of Baylor University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

Approved by the Dissertation Committee

Kevin G. Pinney, Ph.D., Chairperson

Charles M. Garner, Ph.D.

Robert R. Kane, Ph.D.

Mary L. Trawick, Ph.D.

Bessie W. Kebaara, Ph.D.

Accepted by the Graduate School December 2015

J. Larry Lyon, Ph.D., Dean

Page bearing signatures is kept on file in the Graduate School.

TABLE OF CONTENTS

LIST OF FIGURES ...... vii LIST OF SCHEMES...... ix LIST OF TABLES ...... x ACKNOWLEDGMENTS ...... xi DEDICATION ...... xiv CHAPTER ONE ...... 1 Introduction ...... 1 Cancer ...... 1 Structure of Cathepsins ...... 1 Normal Physiological Roles of Cathepsins ...... 5 Cathepsins and Cancer ...... 9 Inhibitors of Cathepsins ...... 11 Thiosemicarbazones ...... 18 CHAPTER TWO ...... 22 Synthesis and biochemical evaluation of benzoylbenzophenone thiosemicarbazone analogues as potent and selective inhibitors of cathepsin L ...... 22 1. Introduction ...... 24 2. Results and discussion ...... 27 2.1 Design and synthetic chemistry ...... 27 2.2 Cathepsin Inhibition Studies ...... 32 2.3 Molecular modeling studies ...... 37 2.4 Invasion and migration studies ...... 39 2.5 Growth Inhibition of Mammary Carcinoma ...... 41 2.6 Cytotoxicity ...... 42 3. Conclusions ...... 43 4. Experimental Section ...... 44 4.1 Chemistry ...... 44 4.2 Biology ...... 74 Acknowledgments ...... 79

v Supplementary data ...... 80 References ...... 80 CHAPTER THREE ...... 88 Synthesis and Biological Evaluation of a Water Soluble Phosphate Prodrug Salt and Structural Analogues of KGP94, a Lead Inhibitor of Cathepsin L ...... 88 Abstract ...... 88 Introduction ...... 89 Results and Discussion ...... 93 Biological evaluation ...... 99 Conclusion ...... 101 Experimental Section ...... 102 Acknowledgements ...... 125 CHAPTER FOUR ...... 126 Design and Synthesis of Cathepsin Inhibitors: Symmetrical Thiosemicarbazone Analogues ...... 126 Introduction ...... 126 Symmetrical Benzophenone Thiosemicarbazones ...... 126 Dibenzylideneacetone and 1,5-diphenyl-3-pentanone thiosemicarbazones ...... 131 General Synthetic Methods: Synthesis of Symmetrical Thiosemicarbazone Analogues ...... 136 CHAPTER FIVE ...... 162 Conclusions ...... 162 APPENDICES ...... 164 Appendix A ...... 166 Appendix B ...... 365 Appendix C ...... 504 REFERENCES ...... 668

LIST OF FIGURES

Figure 1.1. Structural features of cathepsin L, a papain-like protease ...... 2

Figure 1.2. Catalytic mechanism of peptide hydrolysis in cysteine cathepsin……………………………………………..…………………………………………………………………...…...3

Figure 1.3. Schechter and Berger nomenclature: Defining subsites of proteases...... 4

Figure 1.4. Cathepsins and the MHC class II antigen-presentation pathway ...... 5

Figure 1.5. Cathepsin K and bone resorption ...... 7

Figure 1.6. Involvement of cathepsins in invasion and migration of tumor cells ...... 11

Figure 1.7. Cathepsin inhibitors in clinical trials and advanced pre-clinical studies ...... 13

Figure 1.8. Cathepsin inhibitors from natural sources ...... 14

Figure 1.9. Inhibitors of cathepsins described in literature ...... 16

Figure 1.10. Thiosemicarbazone based inhibitors of cysteine proteases ...... 19

Figure 1.11. Proposed cathepsin L inhibition mechanism with thiosemicarbazone based inhibitors ...... 21

Figure 2.1 Overview of cysteine cathepsins in invasion and metastasis...... 24

Figure 2.2 Representative potent inhibitors of cathepsin L utilizing various warheads. ...26

Figure 2.3. Activity of selected benzophenone thiosemicarbazone analogues against cathepsin L ...... 27

Figure 2.4. Isomerization of thiosemicarbazone analogue 33 in DMSO-d6 ...... 32

Figure 2.6. Molecular docking of analogues 1 and 32 within the active site of cathepsin L ...... 37

Figure 2.7. Effect of cathepsin L inhibitors on tumor cell invasion...... 39

vii Figure 2.8. Inhibition of invasion and migration of MDA-MB-231 breast cancer cells by thiosemicarbazone analogues 1, 8, and 32 ...... 40

Figure 2.9. 3-Benzoylbenzophenone thiosemicarbazone (1) effects on the growth of C3H mammary carcinomas implanted in the right rear foot of female CDF1 mice...... 41

Figure 3.1. Cathepsin inhibitors in the pharmaceutical pipeline...... 90

Figure 3.2 Representative sampling of potent covalent inhibitors of cathepsin L ...... 91

Figure 3.3. Sub-set of previously described thiosemicarbazone based inhibitors of cathepsin L ...... 92

Figure 3.4. ROESY and COSY correlations for analogue 13 as a representative example of the major isomer observed in DMSO-d6...... 97

Figure 3.5. KGP420 Incubated with 1.2 Units ALP...... 100

Figure 4.1. Previously synthesized benzophenone thiosemicarbazone analogues ...... 126

Figure 4.2 X-ray crystal structure of compound 37 ...... 135

viii

LIST OF SCHEMES

Scheme 2.1. Synthesis of benzoylbenzophenone thiosemicarbazone analogues...... 28

Scheme 2.2. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing Friedel-Crafts acylation...... 29

Scheme 2.3. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing isophthaloyl dichloride ...... 30

Scheme 2.4. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing 1,3,5-tribromobenzene...... 31

Scheme 3.1. Previously reported synthetic route towards KGP94 ...... 94

Scheme 3.2. Improved synthetic route towards KGP94 (11) and analogues 12-13...... 95

Scheme 3.3. Synthesis of dimethylresorcinol and resorcinol analogues 19-22 ...... 96

Scheme 3.4. Prodrug derivatization of KGP94 (11) to form water-soluble salt KGP420...... 97

Scheme 4.1. Synthesis of thiosemicarbazones with a symmetrical benzophenone scaffolds ...... 128

Scheme 4.2. Synthesis of the 1,3-diphenylpropan-2-one thiosemicarbazone ...... 129

Scheme 4.3. Synthesis of di(naphthalen-2-yl)methanone thiosemicarbazone ...... 129

Scheme 4.4. Synthesis of dibenzylideneacetone and 1,5-diphenyl-3-pentanone thiosemicarbazone analogues ...... 131

Scheme 4.5. Synthesis of (E)-7-(4-fluorobenzylidene)-3-(4-fluorophenyl)- 7a-hydroxy-3a,4,5,6,7,7a-hexahydro-1H-indazole-1-carbothioamide ...... 134

LIST OF TABLES

Table 1.1 Cathepsin L processing of proneuropeptides ...... 8

Table 1.2 Cathepsin Inhibitors in the pharmaceutical pipeline ...... 12

Table 2.1 Inhibitory activity of benzoylbenzophenone thiosemicarbazone analogues...... 34

Table 2.2. Evaluation of cytotoxicity against HUVECs ...... 42

Table 3.1. Isomerization of benzophenone thiosemicarbazone analogues in

DMSO-d6 ...... 98

Table 3.2. Inhibitory activity of benzoylbenzophenone thiosemicarbazone Analogues ...... 99

Table 3.3. Cytotoxicity against HUVECs ...... 101

Table 4.1. Inhibitory activity of benzophenone symmetrical thiosemicarbazones and extended analogues 12 and 15...... 130

Table 4.2. Reduction of dibenzylideneacetone ...... 132

Table 4.3. Inhibitory activity of dibenzylideneacetone and 1,5-diphenyl-3-pentanone thiosemicarbazone analogues ...... 133

ACKNOWLEDGMENTS

I am grateful to many who have been influential in my life and helped mold the individual I have become. I am especially thankful to those who have in the past several

years offered so much support and encouragement during the journey towards my doctoral degree. I acknowledge, with gratitude, my debt of thanks to my advisor Dr.

Kevin G. Pinney for his mentorship and guidance throughout the time of my graduate career. I admire his immense passion for the development of anti-cancer therapeutics and greatly appreciate his unwavering patience as a leader, endless enthusiasm, and dedication. My experience as a graduate student in Dr. Pinney’s Laboratory has influenced me to pursue a career in medicinal chemistry.

I am thankful to Dr. Mary Lynn Trawick, an exceptional committee member and collaborator, who has offered encouragement over the years and kindly and enthusiastically strengthened my knowledge concerning the biological aspects of the cathepsin project through insightful discussions.

I am thankful to Dr. Charles M. Garner for his encouragement and appreciate his generosity, enthusiasm, and suggestions in helping me understand research based questions.

I express my appreciation and gratitude to my dissertation committee Dr. Kevin

G. Pinney, Dr. Charles M. Garner, Dr. Robert R. Kane, Dr. Mary Lynn Trawick, and Dr.

Bessie W. Kebaara who have given their valuable time in reviewing material for oral examinations and the dissertation defense as well as insightful comments and encouragement.

xi I would like to thank friends and colleagues in the Pinney Group and Trawick

Group. I would especially like to thank my classmates Chen-Ming and Laxman who have shared the same journey in pursuit of our doctoral degrees. We ardently encouraged each other along the way, provided valuable advice to each other, and I deeply cherish the friendship we have developed over the years.

I am thankful to Nancy Kallus, Adonna Cook, Virginia Hynek, and Andrea

Johnson for their hard work and help during these years. I am thankful to the Department of Chemistry and Biochemistry, Baylor University, and OXiGENE Inc. for their generous financial support.

I am extremely thankful to my family and friends for their love, encouragement, and support. From a very early age, my Mom encouraged my innate curiosity and emboldened me to pursue my dreams. Both my Mom and my Dad through pursuit of their own goals have demonstrated that hard-work, dedication, and perseverance are key elements for achieving personal dreams. For their infinite love and encouragement, I am thankful to my, Mom, Dad, Adele, Matt, Paw, Granny, and Meemaw. I am grateful to my sisters, Jessica and Julia, and truly treasured friends Anna, Sarah, Melissa, Kelly, and

Morgan for acceptance, understanding, laughter, consolation, motivation, and encouragement. I am thankful to my friend Anna who has been very much like a sister to me; I greatly appreciate her advice and honesty as well as her enduring encouragement in both social and intellectual realms. I express appreciation and gratitude to my friend

Sarah, a rock star in her own rights, who so often instills strength, courage, and a fiery passion in others through her encouragement. I am thankful to Vikas Kumar who has

xii provided me with vast amount of support, understanding, and encouragement over the past few years.

I am immensely privileged to have mentors, colleagues, family, and friends who

have given me an enormous amount of strength during this journey.

xiii

DEDICATION

To My Parents and Grandparents

xiv

CHAPTER ONE

Introduction

Cancer

Over 90% of fatalities resulting from cancer are attributed to metastasis. A number of changes occur within the tumor microenvironment which facilitates detachment of a tumor cell from the local, invasion into the surrounding tissue of the primary tumor site, migration to a distant location, and establishment of a secondary

tumor. The availability of anti-metastatic agents remains largely an unmet need.1,2

Inhibition of cysteine cathepsins has emerged in recent years as a potential anti-metastatic

target owing to their ability to degrade components of the extracellular matrix aiding in

cancer cell invasion migration and the correlation between their upregulation in various

cancers and overall poorer prognosis.

Structure of Cathepsins

Cathepsins are comprised of a diverse class of proteases including the serine

cathepsins A and G, the aspartyl cathepsins D and E, the metalloprotease cathepsin III,

and the cysteine cathepsins B, L, K, S, V, F, W, H, X, C, and O.3,4 The cysteine

cathepsins belong to the C1 family, also known as the papain family, of the Clan A

cysteine peptidases.5 Cysteine cathepsins, with the exception of cathepsin C, are

monomeric proteins comprised of a single polypeptide chain containing between 214 and

260 amino acids in their mature forms. Two domains, a left domain containing three α

helices and a right domain containing a β barrel motif, forming a V-shaped active site

cleft define the general tertiary structure of the mature enzymes (Figure 1.1). Hydrolysis

1 of peptide bonds are facilitated by the amino acid residues Cys25 (left domain), His159

(right domain), and Asn175 (right domain) which form the catalytic triad.6,7

Figure 1.1. Structural features of cathepsin L, a papain-like protease (Reprinted from Biochimica et Biophysica Acta (BBA)-Proteins and Proteomics, Vol 1824, Vito Turk, Veronika Stoka, Olga Vasiljeva, Miha Renko, Tao Sun, Boris Turk, Dusan Turk, Cysteine cathepsins: From structure, function and regulation to new frontiers, 66-88, 2012, with permission from Elsevier.)8

In papain-like enzymes, the Cys25 and His159 residues exist as a thiolate/imidazolium ion pair between the approximate pH range of 4-8.5.6,9 Analogous to the aspartic acid residue in serine proteases, the asparagine residue helps stabilize the active site thiolate/imidazolium ion pair through hydrogen bonding.6,10 The hydrogen bond between the His159 imidazolium NH and Asn175 side chain carbonyl is shielded from solvent molecules by the aromatic side chain of Trp177. Substrate proteolysis is initiated by the attack of the nucleophilic thiolate ion on the amide carbonyl carbon of the substrate to form a thiohemiketal intermediate (Figure 1.2).

Figure 1.2. Catalytic mechanism of peptide hydrolysis in cysteine cathepsins

The tetrahedral intermediate is stabilized in the oxyanion hole through hydrogen bonding with the side chain NH of Gln19 and the backbone NH of Cys25.6,11 Proton

transfer from the imidazolium ion NH to the substrate NH followed by collapse of the

tetrahedral intermediate releases the C-terminus of the peptide and generates a thioester

intermediate. Deacylation of the enzyme is initiated by the nucleophilic attack of an

activated water molecule to form a tetrahedral intermediate which subsequently collapses

to release the N-terminus of the substrate and regenerate the enzyme active site.6

The active sites of proteases can be further described from the manner in which

amino acid residues of natural substrates bind within the active site of the enzyme prior to hydrolysis. According to Schechter and Berger nomenclature, 12,13 binding regions of the

3 enzyme are designated as subsites, S, and amino acid residues of the substrate are assigned positions, P. Numbering of the positions of the amino acid residues in the substrate begins with the cleaved peptide bond with the C-terminus residues labeled as primed and the N-terminus residues labeled as non-primed. Numbering of the enzyme binding regions, subsites, directly correlates to assigned P values for the amino acid residues of the substrate (Figure 1.3).

The mature form of cathepsin L shares 79.5% sequence identity with cathepsin

V,15 the closest human homolog, 60% sequence identity with cathepsin K,16 and 57% sequence identity with cathepsin S.17 The S2 subsite of papain enzymes forms a binding pocket and is generally considered as the specificity binding pocket. The S2 pocket of cathepsin L, an endopeptidase, is large and hydrophobic and prefers amino acids with aromatic side chains. Cathepsin V, an endopeptidase, showed preference for hydrophobic aromatic amino acid residues at the P2 position. Cathepsin S and Cathepsin K are both endopeptidases which prefer non-aromatic hydrophobic residues in the S2 binding pocket. Cathepsin B, and endopeptidase as well as exopeptidase, prefers large aromatic residues as well as amino acids with basic side chains such as arginine in the S2 binding

4 pocket. Cathepsins L, V, K, S, and B prefer arginine, lysine and glycine at the S1 subsite.18,19

In addition to protein turnover, cysteine cathepsins have been implicated in specialized physiological roles such as antigen presentation, bone remodeling, neuropeptide and hormone processing. Cathepsins L and S play significant roles in the major histocompatibility complex (MHC) class II antigen presentation pathway (Figure

1.4). In antigen presenting cells, the invariant chain acts as a chaperone molecule of the

MHC class II αβ heterodimer as well as prevents premature binding of antigen peptides.20–23

Normal Physiological Roles of Cathepsins

5 Endosomal cathepsin L24 (present in cortical thymic epithelial cells and

macrophages) and cathepsin S25,26 (present in dendritic cells, B cells, and macrophages)

participate in the late-stage degradation of the invariant chain leaving a smaller peptide

fragment CLIP, which ultimately undergoes exchange with an antigen peptide. In

addition to processing of the invariant chain, cathepsins degrade antigens to generate

epitopes which can bind to the MHC class II αβ heterodimer.27,28 Once the antigen peptide is bound to the MHC class II αβ heterodimer and expressed on the cell surface, the MHC class II complex binds to the CD4+ protein on a T-helper cell allowing the

antigen to bind to the T-cell receptor which subsequently activates the adaptive immune

response.20

Discovery of a physiological role for cathepsin K emerged from the link between pycnodysostosis, a disease characterized by increased bone density resulting from

abnormal accumulation of collagen in osteoclasts, and a mutation in the cathepsin K gene

leading to deficient translation of the lysosomal protease.29,30 Physiological bone

remodeling involves bone resorption performed by osteoblasts followed by bone

formation performed by osteoclasts.31–33 Bone matrix consists of the inorganic bone

matrix which is primarily comprised of hydroxyapatite and calcium carbonate, and the organic bone matrix of which 90% is type I collagen.33–37 During bone resorption osteoclasts are recruited to the bone surface and acidify the extracellular lacunae which facilitate bone demineralization.38,39 Secretion of cathepsin K, a powerful collagenolytic

enzyme, into the resorption lacunae facilitates type I collagen degradation (Figure 1.5).40

Cathepsin K degrades collagen in the telopeptide region as well as the helical region

6 when in an oligomeric complex with glycosaminoglycans. The collagen fragments are endocytosed and subsequently released at the cell’s antiresorptive surface.41

Cathepsin L, also known as prohormone thiol protease (PTP), 42 is involved in processing proneuropeptides into neuropeptides and peptide hormones (Table 1.1).43

Cathepsin L knockout models revealed substantial dependence for the presence of this protease in secretory vesicles for processing a number of proneuropeptides including proenkephalin,42 proneuropeptide Y,44 proopiomelanocortin,45 prodynorphin,46 and prochlocystokinin.47 Cleavage of proneuropeptides by cathepsin L produces

intermediates which subsequently undergo removal of terminal basic amino acid residues

7 by the appropriate aminopeptidase or carboxypeptidase to generate the active neuropeptide. Cathepsin L facilitates processing of pro-enkephalin, proopiomelanocortin, and pro-dynorphin to generate Met-enkephalin, β-endorphin, and dynorphin A/B which are ligands for opioid receptors that mediate analgesic effects.48

Proteolytic processing of proneuropeptide Y generates neuropeptide Y which has implications in stress induced obesity49 and regulation of blood pressure.50

Cholecystokinin octapeptide (CCK8) generated from proteolytic processing of procholecystokinin suppresses hunger51 and induces anxiety.52

Table 1.1 Cathepsin L processing of proneuropeptides a

Proneuropeptide Neuropeptide/ Regulatory Percent hormone function reduction in CTSL KO model

Proenkephalin (Met)enkephalin Analgesia 50%42

Proneuropeptide Y Neuropeptide Y Blood pressure and 80-90% obesity

Proopiomelanocortin ACTH Steroid production 77%

Β-endorphin Analgesia 82%

α-MSH Skin pigmentation 93%

Prodynorphin Dynorphin A Analgesia 75%

Dynorphin B Analgesia 83%

α-Neoendorphin Analgesia 90% procholecystokinin CCK8 Anxiety, cognition, 75% analgesia a Adapted from L. Funkelstein et al43

8 Cathepsins and Cancer

Dysregulation of normal physiological processes contributes to numerous disease

states. For reviews discussing the involvement of cathepsins in cancer metastasis, see the

cited references.53–56 A strong association between cathepsins and cancer developed when

the major excreted protein (MEP) of transformed cells, which exhibited a 200-fold

increase in extracellular levels compared to non-transformed , was identified as cathepsin

L;57–60 moreover, variance of cathepsin expression of tumor cells coincides with the

metastatic state of the tumor cells.61,62

In A375SM melanoma cells, hypomethylation of a GpG island in the promoter

region of the gene encoding for cathepsin L leads to increased translation and subsequent

synthesis of the cysteine protease.63 Regulation of cathepsin proteolytic activity occurs

through auxiliary mechanisms beyond translation and protein synthesis. The prodomain

occupies the active site in a backwards binding orientation and prevents unwarranted proteolysis. Binding affinity of the pro-domain within the active sites of cathepsin are pH

dependent.64–66 Cathepsins are autocatalytically activated within endosomes or lysosomes

under low pH conditions.15,67,68 Acidosis of the tumor microenvironment contributes to

increased cathepsin activity through activation of proforms. Proteolytic potential of the

mature enzymes are governed by pH dependent activity and stability in which acidic tumor microenvironments provide favorable conditions.15 Additionally, hypoxic and

acidic conditions inherent in the tumor microenvironment lead to trafficking of lysosomes to the plasma membrane and subsequent exocytosis.69,70

Proteolytic activity of cathepsins are regulated by the endogenous proteins type 1

cystatins (stefins), type 2 cystatins, and type 3 cystatins (kininogens). Stefins A and B

9 primarily exhibit intracellular localization. Type 2 cystatins (C, D, E/M, F, S, SA, and

SN) contain a signal peptide for extracellular translocation which allows for the regulation of the proteolytic activity of secreted proteases. High molecular weight and low molecular weight kininogens, stefin A, stefin B, and cystatin C are potent inhibitors of cathepsins L, B, K, H, and S with Ki values in the picomolar to nanomolar ranges. 71

Cystatins D, F, SA, SN, L-kininogen, and H-kininogen strongly inhibit cathepsin L.71 In tumor cells, increased expression of cathepsins often coincides with downregulation of endogenous inhibitors.72

Upregulation of cathepsin activity occurs through increased expression, secretion

into the extracellular space, acidification of extracellular space, and downregulation of

endogenous inhibitors. The proteolytic potential of cathepsins in the degradation of extracellular matrix proteins including fibronectin,73–76 laminin,74–76 collagen70,75–80 and

participation in proteolytic cascades,81 facilitates invasion and migration of malignant

tumor cells (Figure 1.6).8

Figure 1.6. Involvement of cathepsins in invasion and migration of tumor cells. (Reprinted from Bioorganic and Medicinal Chemistry, Vol xx, Parker, E. N.; Song, J.; Kishore Kumar, G. D.; Odutola, S. O.; Chavarria, G. E.; Charlton-Sevcik, A. K.; Strecker, T. E.; Barnes, A. L.; Sudhan, D. R.; Wittenborn, T. R.; Siemann, D. W.; Horsman, M. R.; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. Synthesis and Biochemical Evaluation of Benzoylbenzophenone Thiosemicarbazone Analogues as Potent and Selective Inhibitors of Cathepsin L, 6974-6992, Copyright 2015, with permission from Elsevier).

Inhibitors of Cathepsins

Cathepsin inhibitors as promising candidates for FDA approval are currently in the pipeline of several pharmaceutical companies (Table 1.2). Recently, GSK-

2793660,82,83 a cathepsin C inhibitor, was in phase I clinical trials for the treatment of bronchiectasis. Past and present clinical trials involving inhibition of cathepsin K include

Odanacatib84,85 and MIV-71186 which are primarily aimed towards decreasing bone

11 resorption in individuals with osteoporosis. Cathepsin S inhibitors in clinical trials

include VBY-37687,88 for the treatment of liver fibrosis and therapeutic agents for the

mediation of overactive immune responses in psoriasis (VBY-89189,90 and RWJ-

44538091), rheumatoid arthritis (RWJ-445380)92, and neuropathic pain (VBY-03693,94 and

MIV-24795). Disclosed structures of cathepsin inhibitors in the pharmaceutical pipeline

are included in Figure 1.7.

Table 1.2. Cathepsin inhibitors in the pharmaceutical pipeline

Inhibitor Targeted Therapeutic Clinical Trial Pharmaceutical Cathepsin Indication Phase Company VBY-825 Pan-Cysteine Primary Biliary Pre-Clinical Virobay Cirrhosis GSK-2793660 Cathepsin C Bronchiectasis Phase I GlaxoSmithKline VBY-376 Cathepsin B Liver Fibrosis Phase I Virobay Odanacatib Cathepsin K Osteoporosis Phase III Merck MIV-711 Cathepsin K Osteoporosis Phase I Medivir VBY-891 Cathepsin S Psoriasis Phase I Virobay and LEO Pharma VBY-036 Cathepsin S Neuropathic Phase I Virobay Pain LY3000328 Cathepsin S Abdominal Phase I Eli Lily Aortic Aneurysm RWJ-445380 Cathepsin S Rheumatoid Phase II Johnson & Arthritis, Johnson psoriasis MIV-247 Cathepsin S Neuropathic Pre-Clinical Medivir Pain

12 VBY-82596,97 (Virobay) has been implicated for the treatment of primary biliary

cirrhosis and inhibits cathepsins S, L, V, B, K, and F with Ki values in the picomolar to

nanomolar range. VBY-825 contains a carbonyl electrophilic which reversibly interacts

with the catalytic cysteine residue via a hemiothioketal intermediate. Odanacatib84,85

(Merck), also known as MK-0822, targets osteoporosis through inhibition of cathepsin K.

The nitrile based inhibitor reversibly and selectively inhibits cathepsin K. LY300032898,99

(Eli Lily) targets the treatment of abdominal aortic aneurysm through potent and selective

inhibition of cathepsin S (IC50 = 7.7 nM). A co-crystal structure of cathepsin S with

LY3000328 revealed that inhibition occurred through non-covalent interactions. To date,

cathepsin L specific inhibitors have not yet entered clinical trials.

Figure 1.7. Cathepsin inhibitors in clinical trials and advanced pre-clinical studies

Small-molecule inhibitors of cysteine cathepsins incorporate electrophilic

warheads which form covalent bonds with the Cys25 thiolate in the catalytic triad.

Several natural products have been discovered to be inhibitors of cysteine proteases

(Figure 1.8). Nonpeptidic natural products 3-epiursolic acid (I),100 isolated from Myrcia

linga, and the prenylated dihydrochalcone (II),101 isolated from Metrodorea stipularis,

displayed reasonable activity towards cathepsin L with IC50 values in the low micromolar

range.

Figure 1.8. Cathepsin inhibitors from natural sources

E-64 (III),102–105 isolated from the mold Aspergillus japonicus, is an

epoxysuccinyl peptide which demonstrates specificity towards cysteine proteases

compared to serine proteases. E-64 (III) irreversibly inhibits several cysteine proteases

including actinidin, papain, cathepsin L, cathepsin B, and cathepsin K. Leupeptin

(IV),106–109 isolated from actinobacteria within the Streptomyces genus, displays broad specificity against serine, cysteine, and threonine proteases. An aldehyde electrophilic warhead lies within its argininal peptide molecular framework. Leupeptin (IV) reversibly inhibits cathepsins L and B with IC50 values of 6.6 nM and 9.4 nM, respectively.

Tokaramide A (V),110 another argininal peptide aldehyde inhibitor, was isolated from the

marine sponge Theonella aff. Mirabilis in 1999. Evaluation against cathepsin B revealed

Tokaramide A (V) to inhibit 50% of enzyme activity at 61 nM. Miraziridine A (VI),

14 isolated from the same species of marine sponge as Tokaramide A (V), irreversibly inhibits cathepsin L and cathepsin B. Incorporated into the framework are an aziridine ring, a statine, and an αβ-unsaturated ester as electrophilic centers available for interaction with nucleophilic amino acid residues within protease active sites. Analysis of truncated miraziridine A (VI)111,112 analogues suggested that the aziridine ring contributes

to the inhibitory activity against cathepsin B. Initial biological evaluation of gallinamide

A (VII),113,114 isolated from marine cyanobacteria from a Schizothrix species, showed

that the natural product possessed potent antimalarial activity against Plasmodium

falciparum, Leishmania donovani, and Trypanasoma cruzi. Gallinamide A (VII)

irreversibly inhibits cathepsin L with a low IC50 value of 5.0 nM (30 min pre-incubation

time). Additionally, inhibition is 28-fold more selective for cathepsin L than cathepsin V

(also referred as cathepsin L2) which share 79.5% sequence identity in their mature

forms.

Based on the natural product E-64, Katanuma et al incorporated the

epoxysuccinate warhead into a new series of cathepsin inhibitors named CLIK.115

Emerging from this series of inhibitors was CLIK-148 (VIII) which displayed selective inhibition of cathepsin L compared to cathepsins B, C, K, and S. Inhibition of cathepsin L by vinyl sulfone IX116 occurs irreversibly through formation of an irreversible thioether

bond with the major catalytic amino acid residue (Cys25) of cathepsin L (Figure 1.9).

Inhibition occurs with a second order rate constant of 181.420 s-1 M-1. Molecular

modeling suggests that the phenyl substituent extending from the sulfone resides in the

S2 pocket, the biphenyl group resides in the S1 subsite, and the 4-methylpiperazinyl

group extends towards the S2’ subsite.

Figure 1.9. Inhibitors of cathepsins described in literature

KD-1 (X),117a peptide based vinyl sulfonate, irreversibly inhibited cathepsin L with a second order rate constant of 4.3 x 106 M-1 s-1 and was selective for cathepsin L

compared to cathepsin S by 13-fold, cathepsin K by 100-fold, and cathepsin B by 44,000-

fold. KD-1 inhibited proteolytic activity of cathepsin L towards collagen type 1 as well as

intracellular cathepsin L activity of MDA-MB-231 breast cancer cells. The natural

product leupeptin inspired the design of potent peptidyl aldehyde cathepsin L and

cathepsin B inhibitors. Among the described inhibitors, the tryptophanal derivative XI109

showed selectivity towards cathepsin L compared to cathepsin B and inhibited in vivo

118 bone resorption. The peptidomimetic α-ketoamide XII inhibited cathepsin S (IC50 =

5.8 nM) and cell migration of CL1-5 lung cancer cells and Human umbilical vein endothelial cells (HUVECs). Azepanone based inhibitors of cathepsins L, K, S, and B first reported in 2001 by Marquis and co-workers119 interact with the catalytic cysteine

16 residue through formation of a reversible hemithioketal. Modification of the backbone

template to incorporate a β-naphthylalanine moiety in the P2 position and a naphthylene-

1-carboxamide moiety in the P3 position led to the design of XIII120 which selectively

inhibited cathepsin L (Ki, app = 0.43 nM) compared to cathepsin K (Ki, app = > 10,000 nM)

by >10,000-fold, cathepsin S (Ki, app = 15.6 nM) by 36-fold, and cathepsin B (Ki, app = 150 nM) by 348-fold. Comparison of crystallographic and molecular modeling studies of azepanone based inhibitors attributed the increase in selectivity for cathepsin L compared to cathepsin K to disruption of π- π stacking interactions of the P3 substituent with Tyr67 in the S3 subsite of cathepsin K and increased steric bulk of the P2 β-naphthylalanine moiety in the S2 pocket of cathepsin K which cathepsin L can more easily

120 121–123 accommodate. Thiocarbazate XVI potently inhibited cathepsin L (IC50 = 19 nM)

and possessed selectivity over cathepsins S (530 nM) and B (3250 nM). Molecular

modeling studies with similar analogues suggest that the indole ring occupies the S2

pocket while the thiocarbazate carbonyl interacts with Cys25. Triazine nitrile XV124 inhibits rhodesain (Ki = 13 nM), a cysteine protease found within Trypanosoma brucei,

and cathepsin L (Ki = 3 nM). Co-crystal structure of a triazine nitrile XV with cathepsin

L revealed that the cyclohexyl portion of the molecule resides deep within the S2 pocket and a thioimidate intermediate is formed between the Cys25 thiolate and nitrile of XV which is stabilized through hydrogen bonding with the side chain of Gln19 in the oxyanion hole. Optimization through interchanging substituents on the triazine ring lead to analogues with increased selectivity for rhodesain compared to cathepsin L.

125 Nonpeptidic cyanamide XVI inhibited cathepsin K (IC50 = 50 nM) and cathepsin L

17 (IC50 = 80 nM). Inhibitors of this class of molecules were found to be reversible time-

dependent inhibitors of cathepsin K.

Thiosemicarbazones

Thiosemicarbazone based inhibitors have been designed for a number of cysteine proteases of the papain family found in a number of species including Homo

sapiens,69,126–133,134 Trypanosoma cruzi and Trypanosoma brucei,135–140 Plasmodium

140–142 143 144 falciparum, Leishmania Mexicana, and Eimeria tenella (Figure 1.10). Prior to

the utilization of thiosemicarbazone based protease inhibitors closely related

semicarbazone analogues were shown to be bioisosteric replacements for dipeptides,145 and evaluated for inhibitory activity against cathepsin K.146 As an interesting note, semicarbazones are used as protecting groups for the purification of peptidyl aldehydes.147

In 2002, Du et al published a manuscript describing inhibitors of cruzain, a

cysteine protease in Trypanosoma cruzi which potentiates Chagas disease, with a variety

of scaffolds incorporating the thiosemicarbazone moiety including 3′-

bromopropiophenone thiosemicarbazone XVII (cruzain IC50 = 100 nM) and 5-(3-

chlorophenyl)-2-furancarboxaldehyde thiosemicarbazone XVIII (cruzain IC50 = 140

nM)136 (Figure 1.10). In 2004, Fujii et al described thiosemicarbazone based inhibitors of

cathepsin-L like cysteine proteases found in Trypanosoma brucei (rhodesain) and

Trypanosoma cruzi (cruzain) incorporating the 3,4-dichlorophenyl functionality on one

side of the imine bond and a variety of groups on the other side of the imine bond

139 including analogue XIX (cruzain IC50 = 40 nM, rhodesain IC50 = 38 nM).

Figure 1.10. Thiosemicarbazone based inhibitors of cysteine proteases

Drawing inspiration from these studies the Pinney Laboratory in collaboration with the Trawick Laboratory, described (in 2006) a series of thiosemicarbazone based cruzain inhibitors with tetrahydronaphthalene, benzophenone, and propiophenone molecular frameworks. 135 m-Bromo substituted benzophenone analogue XX (Cruzain

IC50 = 24 nM) and 7-bromo tetrahydronaphthalene analogue XXI (Cruzain IC50 = 17 nM)

emerged as lead inhibitors of cruzain.135 Selectivity screening for inhibitors of cruzain

against human cathepsins L, B, and K revealed analogues with potent inhibition against

these cysteine proteases. Owing to the upregulation of cathepsins in several diseases

including cancer, collaborative research in the Pinney Laboratory and the Trawick

Laboratory gradually channeled to the design and synthesis of cathepsin inhibitors as

anti-metastatic agents. Explored molecular frameworks (Pinney and Trawick

19 collaboration) incorporating the thiosemicarbazone warhead for cathepsin inhibition in

the include benzophenone, pyridine, thiophene, fluorene, thiochromanone, benzothiepine,

dihydroquinoline, and benzoylbenzophenone. 69,126–133 Potent thiosemicarbazone based

cathepsin L inhibitors of these series include (3-bromo-3'-hydroxybenzophenone)

128,129,69 thiosemicarbazone XXII (cathepsin L IC50 = 131.4 nM), (3-bromo-2'-

127 fluorobenzophenone) thiosemicarbazone XXIII (cathepsin L IC50 = 30.5 nM), and 6-

131 nitrothio-4-chromanone thiosemicarbazone XXIV (cathepsin L IC50 = 68 nM). In 2014,

Raghav et al described benzaldehyde thiosemicarbazone analogues including o-

chlororbenzadehyde thiosemicarbazone XXV, which inhibited cathepsin B with a Ki

value of 1.48 x10-5 M which was 50 times greater than the value observed for the

corresponding semicarbazone benzaldehyde derivative.134 The indole based

thiosemicarbazone XXVI discovered in a high throughput screening inhibited CPB, a

cathepsin L-like protease found in the protozoan Leishmania Mexicana, with an IC50 value of 60 nM.143 High throughput screening which led the identification of CPB

inhibitors also led to inhibitors of the cathepsin-B like protease EtCatB found in Eimeria

tenella. 144 Pryazole thiosemicarbazone analogue XXVII moderately inhibited EtCatB

(EtCatB IC50 = 940 nM) and showed selectivity against bovine cathepsin B (IC50 > 30).

Isatin based thiosemicarbazone analogues including XXVIII (falcipain-2 = 6070 nM)141 and dihydro-arteminsinin derivatives including XXIX (falcipain-2 = 520 nM)142

moderately inhibited the cysteine protease falcipain-2 found in the malaria causing

protozoan Plasmodium falciparum.

The proposed mechanism of inhibition, supported by molecular docking and

kinetic studies,129,133 of papain-like proteases with thiosemicarbazone inhibitors based on

20 the benzophenone and benzoylbenzophenone scaffolds occurs through formation of a

transient covalent bond between the thiocarbonyl of the inhibitor and the Cys25 thiolate

of the enzyme (Figure 1.11). Kinetic studies have shown that cathepsin inhibition with

these analogues is time-dependent and reversible.129,133

Figure 1.11. Proposed cathepsin L inhibition mechanism with thiosemicarbazone based inhibitors

CHAPTER TWO

Synthesis and biochemical evaluation of benzoylbenzophenone thiosemicarbazone analogues as potent and selective inhibitors of cathepsin L

This chapter is reprinted from Bioorganic and Medicinal Chemistry, Vol xx, Parker, E. N.; Song, J.; Kishore Kumar, G. D.; Odutola, S. O.; Chavarria, G. E.; Charlton-Sevcik, A. K.; Strecker, T. E.; Barnes, A. L.; Sudhan, D. R.; Wittenborn, T. R.; Siemann, D. W.; Horsman, M. R.; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. Synthesis and Biochemical Evaluation of Benzoylbenzophenone Thiosemicarbazone Analogues as Potent and Selective Inhibitors of Cathepsin L, 6974-6992, Copyright 2015, with permission from Elsevier. The author Erica N. Parker played a significant role in the preparation of the manuscript. Also, Erica N. Parker contributed to this manuscript through the synthesis, purification, and characterization of final analogues 1, 2, 4, 9, 11, 13, 20, 22, 31, 32, 33 and their corresponding intermediates. The author Jiangli Song contributed to this manuscript through the synthesis, purification, and characterization of final analogues 8, 14 and their corresponding intermediates. The author G.D.K. Kumar contributed to this manuscript through the synthesis, purification, and characterization of final analogues 1, 2, 8 and their corresponding intermediates. The author Ashleigh L. Barnes contributed to this manuscript through the synthesis, purification, and characterization of final analogue 10 and corresponding intermediates. Abbreviations used

HUVECs, human umbilical vein endothelial cells; ECM, extracellular matrix; TSC, thiosemicarbazide; SAR, structure-activity relationship; Z-FR-AMC, N-carbobenzyloxy- L-Phe-L-Arg-7-amino-4-methylcoumarin; FBS, fetal bovine serum; SRB, sulforhodamine B.

Keywords

Small-molecule synthesis, thiosemicarbazone warhead, cathepsin L inhibitors, inhibition of cancer cell invasion and migration, anti-metastatic agents

Note: Citations which appear outside of parenthesis correlate with the bibliography at the end of this chapter and references which appear inside of the parenthesis correlate with the bibliography at the end of the dissertation.

22 Abstract

Upregulation of cathepsin L in a variety of tumors and its ability to promote

cancer cell invasion and migration through degradation of the extracellular matrix

suggest that cathepsin L is a promising biological target for the development of anti-

metastatic agents. Based on encouraging results from studies on benzophenone

thiosemicarbazone cathepsin inhibitors, a series of fourteen benzoylbenzophenone

thiosemicarbazone analogues were designed, synthesized, and evaluated for their

inhibitory activity against cathepsins L and B. Thiosemicarbazone inhibitors 3-

benzoylbenzophenone thiosemicarbazone 1, 1,3-bis(4-fluorobenzoyl)benzene

thiosemicarbazone 8, and 1,3-bis(2-fluorobenzoyl)-5-bromobenzene thiosemicarbazone

32 displayed the greatest potency against cathepsin L with low IC50 values of 9.9 nM,

14.4 nM, and 8.1 nM, respectively. The benzoylbenzophenone thiosemicarbazone analogues evaluated were selective in their inhibition of cathepsin L compared to

cathepsin B. Thiosemicarbazone analogue 32 inhibited invasion through Matrigel of

MDA-MB-231 breast cancer cells by 70% at 10 µM. Thiosemicarbazone analogue 8

significantly inhibited the invasive potential of PC-3ML prostate cancer cells by 92% at 5

µM. The most active cathepsin L inhibitors from this benzoylbenzophenone

thiosemicarbazone series (1, 8, and 32) displayed low cytotoxicity toward normal

primary cells [in this case human umbilical vein endothelial cells (HUVECs)]. In an

initial in vivo study, 3-benzoylbenzophenone thiosemicarbazone (1) was well-tolerated in

a CDF1 mouse model bearing an implanted C3H mammary carcinoma, and showed

efficacy in tumor growth delay. Low cytotoxicity, inhibition of cell invasion, and in vivo

tolerability are desirable characteristics for anti-metastatic agents functioning through an

23 inhibition of cathepsin L. Active members of this structurally diverse group of

benzoylbenzophenone thiosemicarbazone cathepsin L inhibitors show promise as

potential anti-metastatic, pre-clinical drug candidates.

1. Introduction

In various human cancers cathepsins L, B, K, H, S, and X display increased

activity and play significant mechanistic roles in the invasion and migration of malignant

cells.1-5 (148,149,8,54,55) Cysteine protease cathepsins degrade proteolytic targets predominately within lysosomes under normal physiological conditions.1-5 (148,149,8,54,55)

Figure 2.1 Overview of cysteine cathepsins in invasion and metastasis. Compared to normal physiological conditions, certain cathepsins are upregulated in the tumor microenvironment. Additionally, changes occur within the tumor microenvironment such as downregulation of endogenous inhibitors including the cystatins, extracellular acidification increasing the proteolytic capability of cathepsins, and extracellular secretion of cathepsins. Degradation of the extracellular matrix through the direct proteolysis of matrix and cell-adhesion proteins such as laminin, fibronectin, collagen, E- cadherin facilitates tumor cell invasion from the primary tumor site and ultimately migration to a secondary tumor site. Small-molecule cathepsin inhibitors have the potential to inhibit the invasive nature of malignant cells and may provide increased effectiveness in combination with known chemotherapeutics.17 (151)

24 In cancer, cathepsins promote metastasis through phagocytosis as well as

extracellulary through multiple pathways including direct proteolysis of E-cadherin6 (150)

and components in the extracellular matrix (ECM) such as fibronectin,7-10 (73–76) laminin,8-

10 (74–76) collagen,9-15 (70,75–80) and indirectly through the activation of other proteases16 (81) resulting in the degradation of the ECM (Figure 2.1).

Elevated levels of cathepsin L, a ubiquitous endopeptidase, and cathepsin B, an endopeptidase as well as a carboxypeptidase, are associated with poor prognosis in breast cancer, colorectal cancer, lung cancer, brain tumors, melanoma, and head and neck cancer.18, 19 (152,153) Cathepsin K, an endopeptidase involved in bone resorption, shows

increased activity in breast cancer, cervical cancer, and lung cancer.20 (154) The cathepsin

K inhibitor Odanacatib (MK-0822) displayed promise in a recent clinical trial by

reducing bone resorption in individuals with metastatic bone disease.21-22 (155,156)

In addition to metastasis, cathepsins are involved in other pathological processes.

Drug candidates currently in the pipeline of pharmaceutical companies include the

cathepsin K inhibitors Odanacatib (Merck)21-24 (155,156,85,84) and MIV-711 (Medivir)25 (86)

for the treatment of osteoporosis, the cathepsin B inhibitor VBY-376 (Virobay)26-27 (87,88) and the pan cysteine protease inhibitor VBY-825 (Virobay)28-29 (96,97) for the treatment of

liver fibrosis, the cathepsin C inhibitor GSK2793660 (GlaxoSmithKline) for the

treatment of bronchiectasis,30-31 (82,83) and the cathepsin S inhibitor MIV-247 (Medivir)32

(157) for the treatment of neuropathic pain. Although selective inhibitors of cathepsin L have not yet reached clinical trials, several have been described in the literature (Figure

2.2).

25 Historically, molecules incorporating the thiosemicarbazone functionality have

been used as antiviral and antibacterial agents in the treatment of smallpox41-43 (161–163) and tuberculosis.44 (164)

Figure 2.2 Representative potent inhibitors of cathepsin L utilizing various warheads. Notable selective inhibitors of cathepsin L include the natural product gallinamide A I,33 (114) the epoxysuccinamide inhibitor Clik 148 II,34 (115) the oxocarbazate III,35 (158) the vinyl sulfonate IV,36 (117) the dipeptidyl nitrile V,37 (159) and the azepanone VI.38 (120) Potent but non-selective inhibitors of cathepsin L include the nonpeptidic cyanamide VII,39 (125) the azadipeptide nitrile VIII,40 (160) and the diketone VBY-825 IX.28-29 (96,97)

In the investigation of thiosemicarbazone inhibitors in our group45-50 (69,127–131) several cathepsin L inhibitors based on the benzophenone, thiochromanone, thiochromanone sulfone, and dihydroquinoline scaffolds including (3'-bromo-3- hydroxybenzophenone) thiosemicarbazone X,46-48 (69,128,129) (3'-bromo-2-

fluorobenzophenone) thiosemicarbazone XI, 45 (127) and 6-nitrothio-4-chromanone

thiosemicarbazone XVI49 (130) emerged as promising lead compounds (Figure 2.3).

Figure 2.3. Activity of selected benzophenone thiosemicarbazone analogues against cathepsin L

2. Results and discussion

2.1 Design and synthetic chemistry

Functional group considerations based on our initial benzophenone thiosemicarbazone inhibitors included the incorporation of a benzoyl moiety which, in

this study, led to the discovery that appropriately functionalized benzoylbenzophenone

thiosemicarbazone analogues demonstrated potent inhibition of cathepsin L. In addition

to incorporation of a benzoyl group, structural motifs previously identified in potent

benzophenone thiosemicarbazone inhibitors were integrated into the design of this new

series of benzoylbenzophenone thiosemicarbazone analogues.51

27 O O OH O c

R R R=H 3:R=H R=Bz

aorb a

S NH2 S NH2 NH NH O N OH N

R R 1:R=H 4:R=H 2: R=Bz

Scheme 2.1. Synthesis of benzoylbenzophenone thiosemicarbazone analogues. Reagents and conditions: (a) Thiosemicarbazide (TSC), TsOH, THF, reflux; (b) TSC, TsOH, o MeOH, reflux; (c) EtOH, NaBH4, 0 C.

Previously,45-46 (127,128) incorporation of the 3-bromo functionality in

benzophenone thiosemicarbazone analogues proved critical in providing analogues which

displayed potent inhibitory activity against cathepsin L as demonstrated by the large

differences in IC50 values found for X and XI compared to the non-brominated analogues

XIII and XIV, respectively (Figure 2.3). Additionally, the 4-bromo analogue XII displayed reduced activity towards cathepsin L compared to the 3-bromo analogue XI

further emphasizing the importance of the 3-bromo functionality. Drawing on previous

SAR studies based on the benzophenone scaffold,45-46 (127,128) extension of key structural

components of the potent cathepsin L benzophenone thiosemicarbazone inhibitors X and

XI led to the design of benzoylbenzophenone thiosemicarbazone inhibitors 31 and 32

which incorporated the 3-bromo functionality in the central aromatic ring. In order to

avoid potential complications that could arise due to the difficulty of separating

thiosemicarbazone regioisomers, only symmetrical diketones were utilized in the

28 synthesis of the new target compounds. Various synthetic methodologies (Schemes 2.1-

2.4) were employed to assemble each benzoylbenzophenone molecular scaffold.

Benzoylbenzophenone thiosemicarbazones 1, 2 and benzoylbenzhydrol

thiosemicarbazone 4 were prepared from commercially available diketones as illustrated

in Scheme 2.1.

Utilization of Friedel-Crafts acylation chemistry facilitated the synthesis of para

substituted benzoylbenzophenone analogues as depicted in Scheme 2.2. Reaction of

isophthaloyl dichloride and the appropriately substituted benzene ring with aluminum

chloride afforded para substituted 1,3-dibenzoylbenzene analogues 5-7 which underwent

condensation with thiosemicarbazide to yield target thiosemicarbazone analogues 8-11.

Demethylation of 1,3-bis(4-methoxybenzoyl)benzene 7 followed by condensation with

thiosemicarbazide afforded thiosemicarbazone 13 and reaction of diketone 12 with

isopropyl bromide followed by condensation with thiosemicarbazide afforded the p-

isopropoxy derivative 14.

Scheme 2.2. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing Friedel-Crafts acylation. Reagents and conditions: (a) AlCl3 (neat), reflux; (b) AlCl3, CH2Cl2, reflux; (c) TSC, TsOH, THF, microwave irradiation; (d) TSC, TsOH, THF, o reflux; (e) BF3·SMe2, CH2Cl2, rt; (f) isopropyl bromide, K2CO3, DMF, 90 C.

29 In order to incorporate meta substituents in the outermost rings of the benzoylbenzophenone scaffold (Scheme 2.3), isophthaloyl dichloride starting materials were used as precursors to Weinreb amides 16 and 17. Diketones 18 and 19 were synthesized from an intermediate organolithium reagent and Weinreb amides 16 and 17

(separately). Condensation of the resulting meta substituted 1,3-dibenzoylbenzene analogues with thiosemicarbazide and removal of any protecting groups afforded benzoylbenzophenone thiosemicarbazone analogues 20 and 22.

Scheme 2.3. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing isophthaloyl chloride. Reagents and conditions: (a) Me(OMe)NH ·HCl, NEt3, CH2Cl2, 0 oC -rt; (b) n-BuLi, THF, -78 oC; (c) THF, -78 oC; (d) TSC, TsOH, MeOH, reflux; (e) TSC, TsOH, THF, microwave irradiation, 90 oC; (f) TBAF, THF, rt.

In our previous studies, benzophenone thiosemicarbazone analogues45-48 (69,127–129)

containing a meta-bromo substituent were among those with the highest inhibitory activity against cathepsin L. Incorporation of meta-bromo substituents on the central aromatic ring in the benzoylbenzophenone scaffold was achieved by utilizing 1,3,5- tribromobenzene as a starting material. Halogen-metal exchange of 1,3,5- tribromobenzene with t-BuLi followed by the addition of two equivalents of either aldehyde 23 or Weinreb amides 24 and 25 (separately) allowed for the incorporation of a

30 meta-bromo substituent on the central ring of diketone analogues 27-29 (Scheme 2.4).

Subsequent condensation of the diketones 27-29 with thiosemicarbazide yielded the benzoylbenzophenone thiosemicarbazone analogues 30, 32-33, and deprotection of 30 afforded the m-hydroxy substituted analogue 31.

Scheme 2.4. Synthesis of benzoylbenzophenone thiosemicarbazone analogues utilizing 1,3,5-tribromobenzene. Reagents and conditions: (a) Me(OMe)NH ·HCl, NEt3, CH2Cl2, 0 oC -rt; (b) TBSCl, DMF, imidazole, 0 oC -rt; (c) (i) t-BuLi, ether, -78 oC; (ii) 23, 24, or 25 o o in ether, -78 C; (d) PCC, celite, CH2Cl2, 0 C -rt; (e) TSC, TsOH, THF, microwave o irradiation, 90 C; (f) TSC, TsOH, THF, reflux; (g) TSC, Ti(OiPr)4, THF, reflux (h) TBAF, THF, rt.

Moderate to low yields were observed for the condensation reaction to form the

mono thiosemicarbazone analogues due to the competing formation of the bis-

thiosemicarbazone products. For some of the desired mono-thiosemicarbazone products a

lower equivalency of thiosemicarbazide proved effective at minimizing formation of the

bis derivative and maximizing yield, but not in all cases. With the exception of 1,3-bis(2-

fluorobenzoyl)-5-bromobenzene thiosemicarbazone 32, which was isolated as the Z

isomer and did not isomerize after standing in acetone for one week, the final products

31 exist as an inseparable mixture of E/Z isomers in solution (As evidenced by 1H NMR).

Imines including thiosemicarbazones are well-known for their propensity to isomerize in solution under catalyzed62-63 (165,166) and non-catalyzed64-65 (167,168) conditions or from heating while in the solid state.66 (169)

Figure 2.4. Isomerization of thiosemicarbazone analogue 33 in DMSO-d6. The red ellipses indicate regions in the 1H NMR spectra where additional peaks emerge due to the presence of the other geometrical isomer. (a) 1H NMR of compound 33 after 0 hours in 1 1 DMSO-d6. (b) H NMR of compound 33 after 24 hours in DMSO-d6. (c) H NMR of compound 33 after 16 days in DMSO-d6.

As an example, benzoylbenzophenone thiosemicarbazone 33, after purification and drying, was isolated as a single isomer; however, the compound slowly isomerized in solution (Figure 2.4). After 16 days of standing in DMSO at room temperature, thiosemicarbazone 33 isomerized to a 1:1 mixture of E/Z isomers.

2.2 Cathepsin Inhibition Studies

The thiosemicarbazone derivative of commercially available 1,3- dibenzoylbenzene (3-benzoylbenzophenone thiosemicarbazone 1) displayed pronounced

32 inhibitory activity against cathepsin L with an IC50 value of 9.9 nM (Table 2.1).

Interestingly, the analogous benzophenone thiosemicarbazone XV (Figure 3) was

46 (127) inactive (IC50 >10000 nM) against cathepsin L. Since thiosemicarbazone 1

displayed pronounced activity against cathepsin L, several analogues were synthesized

including compounds which incorporated previously reported molecular scaffolds (Figure

2.3) known to be effective in terms of providing compounds with strong inhibitory

activity against cathepsin L.45-48 (69,127–129)

In addition, certain structural modifications of the unsubstituted analogue 1 such

as the incorporation of a structurally demanding benzoyl substituent on the central

aromatic ring, exemplified by tribenzoylbenzophenone thiosemicarbazone analogue 2

(IC50 = 56.0 nM), and the incorporation of a secondary alcohol in place of the carbonyl,

exemplified by 3-benzoylbenzhydrol thiosemicarbazone 4 (IC50 = 23.8 nM), were well

tolerated. Various para substituted analogues were evaluated against cathepsin L. The p-

fluoro analogue 8 had comparable activity to the extremely potent unsubstituted analogue

1. The activity against cathepsin L diminished with increasing steric hindrance in the

para-substituted series (p-hydroxy 13, p-methoxy 11, p-isopropoxy 14).

Thiosemicarbazone analogues 9, 22, and 33 each with p-bromo or m-bromo substituents on the outermost rings of the benzoylbenzophenone molecular template, although still active, showed a significant reduction in regard to inhibitory activity against cathepsin L compared to the other compounds. Additionally, the p-bromo substituted bis- thiosemicabazone analogue 10 showed no activity against cathepsin L at a concentration of 10,000 nM.

33 Table 2.1. Inhibitory activity of benzoylbenzophenone thiosemicarbazone analogues

a IC50 Values (nM)

Cmpd R1 R2 R3 R4 X Cat L Cat B 1 H H H H C=O 9.9 >10000

2 H H H Bz C=O 56.0 >10000 4 H H H H CH(OH) 23.8 >10000 8 F H H H C=O 14.4 >10000 9 Br H H H C=O 1522 >10000

10 Br H H H C=NNHC(S)NH2 >10000 >10000

11 OCH3 H H H C=O 5117 >10000 13 OH H H H C=O 340 >10000

14 OCH(CH3)2 H H H C=O >10000 >10000

20 H CH3 H H C=O 654 >10000 22 H Br H OH C=O ~10000b >10000 31 H OH H Br C=O 71.6 >10000 32 H H F Br C=O 8.1 >10000 33 H Br H Br C=O 10347 >10000 aThese values are averages of a minimum of a triplicate of experiments. Each assay utilized 2% DMSO with a 5 min pre-incubation period. Standard error values can be found in the Supplementary data. b Compound 22 inhibited cathepsin L activity by 56.9% at 10000 nM.

Compared to the parent benzophenone thiosemicarbazone XI (Figure 2.3) with an

IC50 value of 30.5 nM, 1,3-bis(2-fluorobenzoyl)-5-bromobenzene thiosemicarbazone 32

exhibited a greater than three fold increase in activity against cathepsin L with an IC50

34 value of 8.1 nM. Incorporation of the benzophenone thiosemicarbazone X (Figure 2.3)

molecular design into the benzoylbenzophenone scaffold led to two thiosemicarbazone

analogues; compound 31 with meta-hydroxy substituents on the outermost rings and

compound 22 with meta-bromo substituents on the outermost rings. Thiosemicarbazone

31 showed a nearly two fold increase in activity against cathepsin L with an IC50 value of

71.6 nM compared to the parent benzophenone thiosemicarbazone X (IC50 = 131.4 nM)

while analogue 22 showed a significant decrease in activity against cathepsin L with an

IC50 value of ~10000 nM. Overall, increasing steric bulk on the outermost rings likely resulted in benzoylbenzophenone analogues with decreased inhibitory activity.

Interestingly, all active benzoylbenzophenone thiosemicarbazone analogues were selective against cathepsin L compared to cathepsin B and thiosemicarbazone analogues

(31 and 32) of the parent benzophenone thiosemicarbazones X and XI (which do not have bulky groups on the outermost rings) displayed increased activity against cathepsin

L. These studies have demonstrated that the benzoylbenzophenone scaffold coupled with the thiosemicarbazone electrophilic moiety represents an effective molecular design leading to low nM inhibitors of cathepsin L and additionally allows for a variety of structural modifications in the core scaffold while maintaining activity against cathepsin

Kinetic analysis of the closely related thiosemicarbazone analogue X (Figure 2.3) showed that it was a time-dependent and slowly reversible inhibitor of cathepsin L.47 (129)

The reaction progress curves for benzoylbenzophenone thiosemicarbazone analogues 1 and 32 (Figure 2.5) were comparable to those previously observed for (3'-bromo-3- hydroxybenzophenone) thiosemicarbazone X. The cathepsin L catalyzed reaction was

35 carried out with N-carbobenzyloxy-L-Phe-L-Arg-7-amino-4-methylcoumarin (Z-FR-

AMC) as substrate in the presence of varying concentrations (0–10 μM) of

thiosemicarbazone analogues 1 and 32.

Figure 2.5. Representative progress curves of cathepsin L (1 nM) activity using Z-FR- AMC (10 μM) as substrate and increasing concentrations of (A) inhibitor 1 and (B) inhibitor 32 (0–10 μM). Reactions were initiated by the addition of enzyme with no pre- incubation with compounds. Release of the fluorescent product AMC was followed as a function of time.

The progress curves (Figure 2.5) demonstrated that these compounds are not classical, readily reversible inhibitors. The increase in inhibition as a function of time, as observed from a reduction in the slopes of the curves over time compared to untreated control, indicated that both compounds are time-dependent inhibitors of cathepsin L.

Separate experiments demonstrated that analogues 1 and 32 were slowly reversible inhibitors of cathepsin L (see Supplementary data).

36 2.3 Molecular modeling studies

Figure 2.6. Molecular docking of analogues 1 and 32 within the active site of cathepsin L. (A) Analogue 1 and (B) analogue 32 are shown in ball and stick mode (C, gray; H, white; O, red; N, blue; S, yellow; F, light blue; Br, brown). Cathepsin L is shown in ribbon mode with a transparent molecular surface (Green) and enzyme active site amino acid residues are labeled and shown in stick mode (C, gray; H, white; O, red; N, blue; S, yellow).

Thiosemicarbazone analogues 1 and 32 which displayed the most pronounced

activity against cathepsin L were selected for molecular docking studies. The co-crystal

structure of cathepsin L with a nitrile inhibitor (PDB: 2XU1)67 (170) within the active site

of cathepsin L was chosen as a starting model for the docking studies which were

performed with Discovery Studio 4.1 (Accelrys). A number of different binding modes

with similar interaction energies were obtained from docking analogues 1 and 32. In each

case, several of the top binding orientations placed the thiocarbonyl carbon atom of the

inhibitor in close proximity to the Cys25 thiolate in a similar manner to that observed

with benzophenone thiosemicarbazones X47 (129) and XI.45 (127) One of the top binding

orientations for each of benzoylbenzophenone thiosemicarbazone analogues 1 and 32 which positions the Cys25 thiolate within 5 Å of the thiocarbonyl carbon atom is illustrated in Figure 2.6. For analogue 1, the benzophenone fragment of the inhibitor

37 resides deep within the S2 pocket of cathepsin L which is considered an important subsite for binding among the papain family of cysteine proteases.67-68 (18,170) The remaining outermost ring of analogue 1 is oriented towards the S3 subsite. A hydrogen bond between the carbonyl oxygen of the inhibitor and the backbone NH of His163 and a hydrogen bond between the terminal NH2 of the thiosemicarbazone and the backbone carbonyl of Asp162 is observed for analogue 1. Molecular docking studies for analogue

32 orient the aryl rings of the inhibitor within the active site of cathepsin L in a similar fashion to analogue 1 with the benzophenone fragment occupying the S2 pocket and the opposing ring oriented toward the S3 subsite. For comparison, molecular docking studies were carried out with analogue 14, which was inactive against cathepsin L. Molecular modeling indicated that analogue 14 did not bind in a manner that would facilitate formation of a covalent bond. The benzophenone portion resides in the S2 pocket of cathepsin L in a similar fashion observed for active cathepsin L inhibitors 1 and 32.

However, the remaining outermost ring resides in the S1 subsite instead of being oriented towards the S3 subsite and the thiosemicarbazone moiety is oriented towards the solvent instead of residing in the S1 subsite. In each of the top binding orientations, the thiocarbonyl carbon atom of analogue 14 was not in close proximity to the Cys25 thiolate ion of cathepsin L. Active cathepsin L inhibitors 1 and 32 bind in such a manner to facilitate formation of a transient covalent bond between the enzyme and inhibitor. This comparison emphasizes that formation of a transient covalent bond is necessary for activity against cathepsin L in this series of compounds (see Supplementary data).

38 2.4 Invasion and migration studies

A hallmark of invasive cancer cells is their ability to invade through the ECM.

Cell invasion through Matrigel as a model for the ECM69 (171) of the highly metastatic

prostate cancer cell line PC-3ML48,70 (69,172) and the highly invasive breast cancer cell line

MDA-MB-23171-72 (173,174) was inhibited by selected benzoylbenzophenone

thiosemicarbazone inhibitors. Additionally, selected thiosemicarbazone analogues were

evaluated for their ability to inhibit cell migration. To separate anti-tumorigenic from

anti-metastatic effects, non-cytotoxic doses of evaluated thiosemicarbazones were used.

The results showed that treatment with thiosemicarbazone analogue 8 significantly

impaired the invasive capacities of prostate PC-3ML cancer cells by 59% and 92% at 1

µM and 5µM, respectively (Figure 2.7). The unsubstituted analogue 1 inhibited cell invasion of PC-3ML prostate cancer cells by 37% and 53% at 10 µM and 25 µM, respectively.

Figure 2.7. Effect of cathepsin L inhibitors on tumor cell invasion. PC-3ML prostate cancer cell invasion in the presence of cathepsin inhibitors compound 1 (A) or compound 8 (B). Data are mean and standard error values of 3-5 independent experiments.

Figure 2.8. Inhibition of invasion and migration of MDA-MB-231 breast cancer cells by thiosemicarbazone analogues 1, 8, and 32. (a) Inhibition of MDA-MB-231 breast cancer cell invasion through Matrigel at 10 µM concentration of inhibitor. (b) Inhibition of MDA-MB-231 breast cancer cell invasion through Matrigel at 25 µM concentration of inhibitor. (c) Inhibition of MDA-MB-231 breast cancer cell migration at 10 µM concentration of inhibitor. (d) Inhibition of MDA-MB-231 breast cancer cell migration at 25 µM concentration of inhibitor. All values are normalized to 2% DMSO vehicle and are from a minimum of a triplicate of experiments (* p < 0.05, ** p < 0.01, *** p < 0.001).

Thiosemicarbazone 32 displayed the greatest effect on inhibition of invasion of

MDA-MB-231 breast cancer cells with 70 % inhibition at 10 µM, slightly better than

60% inhibition found for the irreversible pan cysteine protease inhibitor E-6473 (175) which was used as a positive control for direct comparison of the effectiveness of the inhibitors

(Figure 2.8). A significant change in inhibition of cell invasion was not observed for analogue 32 at 25 µM compared to 10 µM. However, the p-fluoro analogue 8 demonstrated an increase from 15% to 60% inhibition of invasion of MDA-MB-231 cells at 10 µM compared to 25 µM. The unsubstituted analogue 1, which inhibited cell

40 invasion by 30% at 25 µM, did not impact invasion of MDA-MB-231 cells as significantly as analogues 8 and 32.

All three benzoylbenzophenone analogues when evaluated at 10 µM and 25 µM concentrations, inhibited MDA-MB-231 cell migration; 3-benzoylbenzophenone thiosemicarbazone 1 showed the greatest effect with 80% inhibition of migration at 25

µM which represented a greater than 2-fold increase in inhibition compared with the positive control E-64 (Figure 2.8). In addition, 1,3-bis(2-fluorobenzoyl)-5-bromobenzene thiosemicarbazone 32 and 1,3-bis(4-fluorobenzoyl)benzene thiosemicarbazone 8 also inhibited migration greater than 50% at 25 µM.

2.5 Growth Inhibition of Mammary Carcinoma

Figure 2.9. 3-Benzoylbenzophenone thiosemicarbazone (1) effects on the growth of C3H mammary carcinomas implanted in the right rear foot of female CDF1 mice. Mice were given 5 daily intraperitoneal injections from the day of tumour implant (as indicated by the arrows). Results show means + 1 S.E. from 5-8 mice/group. [Legend:  Control (10% Tween80)  Analogue 1 (20 mg/kg)]

41 To investigate the effect of 3-benzoylbenzophenone thiosemicarbazone (1) (20

mg/kg) on tumor initiation, animals were observed on a daily basis and tumor volume

determined, as described in Figure 2.9, when tumors were measureable. From these data

3 the endpoint of TGT500 (time taken in days for tumors to reach 500 mm ) was

determined. For control animals injected with vehicle (10% Tween80) for 5 days from

the time of tumor implant the mean TGT500 (+ 1 S.E.) was found to be 20.1 days (+ 0.6

days) and this was significantly (Student’s T-test; significance level of p<0.05) increased

to 22.1 days (+ 0.5 days) when mice were given 5 daily treatments with 3-

benzoylbenzophenone thiosemicarbazone (1). This initial in vivo study demonstrated that analogue 1 was well tolerated in this mouse model and showed modest, statistically relevant efficacy in tumor growth delay. Cathepsin L inhibitors, functioning as anti- metastatic agents, are not anticipated to impart dramatic decreases in tumor growth when utilized as single agents; however, it is anticipated that their full value will be realized in combination therapy with standard chemotherapy and/or radiation.17 (151)

2.6 Cytotoxicity

Table 2.2. Evaluation of cytotoxicity against HUVECs

Compound Doxorubicin Paclitaxel 1 8 31 32

Cytotoxicity GI50 0.0268 0.00148 53.3 >126 13.3 >85.9 (µM)

Benzoylbenzophenone thiosemicarbazone analogues 1, 8, 31, and 32 were

evaluated for cytotoxicity against normal endothelial cells (HUVECs). Compared to

known chemotherapeutics, doxorubicin and paclitaxel, none of the compounds displayed

42 significant cytotoxicity against HUVECs (Table 2.2). Compounds 8 and 32 were the least toxic against HUVECs with GI50 values greater than 126 µM and 85.9 µM, respectively.

Low cytotoxicity to normal cells (HUVECs in this case) is a desirable feature since the

development of cathepsin L inhibitors as anti-metastatic agents would likely involve

chronic dosing to achieve efficacy.

3. Conclusions

Several benzoylbenzophenone thiosemicarbazone analogues were synthesized and

evaluated for their inhibitory activity against cathepsins L and B. Thiosemicarbazone

inhibitors 1, 8 and 32 displayed the highest activity against cathepsin L with low IC50 values of 9.9 nM, 14.4 nM and 8.1 nM, respectively. The benzoylbenzophenone thiosemicarbazone analogues represent a class of potent inhibitors that can tolerate structural changes and maintain activity against cathepsin L. The high versatility of the core molecular scaffold offers potential for optimization of the benzoylbenzophenone thiosemicarbazone analogues as inhibitors of cathepsin L. Both thiosemicarbazone analogues 1 and 32 demonstrated time-dependent inhibition of cathepsin L. Molecular docking studies of analogues 1 and 32 revealed that the benzophenone fragment of the benzoylbenzophenone thiosemicarbazone inhibitors is capable of filling the lipophilic S2 pocket of cathepsin L thus placing the thiosemicarbazone moiety in close proximity to the Cys25 thiolate. In contrast, molecular modeling of analogue 14 (inactive against cathepsin L) indicated that it did not bind to the enzyme in a conformation that would promote formation of a covalent bond. Kinetic studies combined with molecular docking studies indicate that the formation of a covalent bond between the enzyme and the most potent of this subset of inhibitors represents an important contribution to their efficacy.

43 The most potent inhibitor of cathepsin L from this series, 1,3-bis(2-fluorobenzoyl)-5-

bromobenzene thiosemicarbazone 32 significantly inhibited the ability of MDA-MB-231

cells to invade through Matrigel by 70%. Additionally, the p-fluoro analogue 8 inhibited

cell invasion of PC-3ML cells by 92% at 5 µM. 3-Benzoylbenzophenone

thiosemicarbazone (1) delayed growth in vivo of recently implanted tumors in a C3H mouse mammary carcinoma model. Cathepsin L inhibitors 1, 8, 31, and 32 demonstrated very low cytotoxicity toward HUVECs. The high potency against cathepsin L and ability to inhibit the invasion of tumor cells in vitro coupled with the desirable property of low

cytotoxicity to normal cells positions these compounds as viable pre-clinical candidates

for further development.

4. Experimental Section

4.1 Chemistry

4.1.1 General Synthetic Protocols.

All reactions were carried out under an inert atmosphere of nitrogen unless

otherwise noted. Anhydrous solvents were used for carrying out reactions. Unless

otherwise noted all reagents and solvents were purchased from commercial suppliers and

used without further purification. Hexanes used for column chromatography were

distilled from 4 Å molecular sieves prior to use. Anhydrous tetrahydrofuran and

dichloromethane were purchased from commercial suppliers or dried using a VAC

(Vacuum Atmospheres Co.) solvent purification system. Reactions were monitored by

SiliaPlate™ silica gel thin layer chromatography (TLC) plates (250 μm, F-254, 60 Å).

Manual flash chromatography and automated flash chromatography was carried out with

silica gel purchased from either Silicycle Inc. (230-400 mesh) or Biotage (40-65

44 microns). Proton (1H), Carbon (13C), and Fluorine (19F) NMR spectra were recorded

using a Varian Inova 500 MHz NMR system, Bruker Avance III HD 600 MHz NMR

system, or a Bruker Avance III HD 400 MHz NMR system. Chemical shifts are reported

in ppm (δ), coupling constants (J) are reported in hertz (Hz), and peak patterns are

reported as broad (br), singlet (s), doublet (d), triplet (t), quartet (q) and multiplet (m).

High resolution mass spectra (HRMS) were obtained in the Baylor University Mass

Spectrometry Core Facility on a Thermo Scientific LTQ Orbitrap Discovery using

electrospray ionization (ESI). Purity of final compounds was analyzed using an Agilent

Technologies 1200 series HPLC system with a Diode Array and Multiple Wavelength

Detector SL, equipped with a Zorbax reliance cartridge guard-column; Agilent Eclipse

XDB-C18 column (4.6 mm ID X 250 mm, 5 μm particle size, 80 Å pore size); T = 25 oC; flow rate 1.0 mL/min; injection volume 20 µL; monitored at 254 nm, 300 nm and 320 nm. Two HPLC methods were used in the purity analysis of final compounds: Method A: water:acetonitrile, gradient 50:50 to 10:90 from 0-25 min and isocratic 10:90 from 25-30 min ; Method B: water:acetonitrile, gradient 70:30 to 10:90 from 0-25 min and isocratic

10:90 from 25-30 min.

4.1.2 3-Benzoylbenzophenone thiosemicarbazone (1).51 (132)

p-Toluenesulfonic acid monohydrate (0.026 g, 0.136 mmol) was added to a

solution of 1,3-dibenzoylbenzene (2.00 g, 6.98 mmol) in anhydrous methanol (20 mL).

After stirring at reflux for 10 min, thiosemicarbazide (0.476 g, 5.235 mmol) was added to

the flask. After 26 h, methanol was removed under reduced pressure and water (50 mL)

was then added. The products were extracted with ethyl acetate (2 x 50 mL) and the

combined organic phases were washed with brine, dried over anhydrous sodium sulfate,

45 and the solvent was removed under reduced pressure. Purification using flash

chromatography (silica gel, hexanes:ethyl acetate, gradient 90:10 to 52:48) afforded 3- benzoylbenzophenone thiosemicarbazone (0.829 g, 5.24 mmol, 44% yield) as a light

1 yellow solid. H NMR (500 MHz, CDCl3): δ 8.71 (1H, s), 8.00-7.99 (1H, m), 7.85-7.83

(1H, m), 7.77-7.67 (3H, m), 7.60-7.29 (10H, m), 6.61-6.57 (1H, m). 13C NMR (125 MHz,

CDCl3): δ 195.97, 195.29, 179.04, 178.99, 149.85, 149.70, 139.13, 137.95, 136.99,

136.94, 136.70, 136.01, 132.97, 132.81, 132.28, 131.84, 131.47, 131.45, 131.34, 130.68,

130.56, 130.48, 130.11, 130.09, 130.05, 128.86, 128.62, 128.58, 128.39, 128.34, 127.76.

+ + HRMS (ESI) calculated for C21H17N3OSH (M+H) 360.11651, found 360.11667. HPLC retention time (Method A): 7.50 min. (Obtained as a mixture of E/Z isomers)

4.1.3 1,3,5-Trisbenzoylbenzene thiosemicarbazone (2). 51 (132)

1,3,5–Tribenzoylbenzene (2.34 g, 6.00 mmol) (purified by flash chromatography

before using) and p-toluenesulfonic acid monohydrate (11.4 mg, 0.06 mmol) were

dissolved in anhydrous THF (20 mL) and heated to reflux. Thiosemicarbazide (546 mg,

6.00 mmol) was then added to the flask and the reaction mixture was stirred at reflux for

35 h. The reaction mixture was concentrated under reduced pressure and the products

were extracted from water (100 mL) with ethyl acetate (2 x 50 mL). The combined

organic phases were washed with brine, dried over anhydrous sodium sulfate, and the

solvent was removed under reduced pressure. Purification using flash chromatography

(silica gel, hexanes: dichloromethane: ethyl acetate, gradient 80:10:10 to 40:30:30)

afforded 1,3,5-trisbenzoylbenzene thiosemicarbazone (0.961 g, 2.07 mmol, 35% yield) as

1 a yellow solid. H NMR (500 MHz, acetone-d6): δ 9.40 (0.2H, s), 8.72 (0.8H, s), 8.33

(.2H, t, J = 1.5 Hz), 8.28 (1.6H, d, J = 1.5 Hz), 8.21 (0.8H, s), 8.13 (0.2H, s), 8.07 (0.8H,

46 t, J = 1.5 Hz), 8.05 (0.4H, d, J = 1.5 Hz), 7.96-7.95 (0.8H, m), 7.85-7.84 (3.2H, m), 7.78

13 (0.8H, s), 7.73-7.52 (10.4 H, m), 7.43-7.37 (0.8H, m). C NMR (125 MHz, acetone-d6): δ

194.47, 194.33, 180.08, 179.66, 147.62, 147.43, 139.25, 138.02, 137.94, 136.77, 136.74,

136.68, 133.68, 133.01, 132.99, 132.59, 131.83, 131.62, 131.37, 130.94, 130.47, 130.06,

130.05, 130.00, 129.99, 129.90, 128.67, 128.62, 128.53, 128.42, 127.85. X-ray

crystallographic data obtained for 1,3,5-trisbenzoylbenzene thiosemicarbazone has been

deposited in the Cambridge Crystallographic Data Centre and was allocated the

+ deposition number CDCC 1042567. HRMS (ESI) calculated for C28H21O2N3SH

(M+H)+ 464.14272, found 464.14280. HPLC retention time (Method A): 10.53, 10.81 min. (Obtained as a mixture of E/Z isomers)

[Note: The use of 0.8H and 0.2H for 1H NMR integral values relate to isomer ratios. This

system is use throughout.]

4.1.4 3-benzoylbenzyhdrol (3). 51,74 (132,176)

1,3-Dibenzoylbenzene (2.00 g, 6.98 mmol) was dissolved in anhydrous ethanol

(20 mL). The reaction mixture was cooled to 0 oC followed by the addition of sodium

borohydride (0.090g, 2.094 mmol). The reaction mixture was stirred for 4 h and

quenched by the addition of a small amount of water. The reaction mixture was

concentrated under reduced pressure and the products were extracted from water with

ethyl acetate (2 x 50 mL). The combined organic phases were washed with brine, dried

over anhydrous sodium sulfate, and the solvent was removed under reduced pressure.

Purification using flash chromatography (silica gel, hexanes: ethyl acetate, gradient 93:7

to 30:70) afforded 3-benzoylbenzhydrol (0.601 g, 2.084 mmol, 30% yield). 1H NMR

(500 MHz, acetone-d6): δ 7.90 (1H, t, J = 1.7 Hz), 7.77-7.74 (2H, m), 7.72-7.69 (1H, m),

47 7.67-7.62 (2H, m), 7.55-5.53 (2H, m), 7.51-7.48 (1H, m), 7.46-7.44 (2H, m), 7.34-7.30

(2H, m), 7.25-7.21 (1H, m), 5.95 (1H, s), 5.04 (1H, brs). 13C NMR (125 MHz, acetone- d6): δ 196.47, 146.94, 146.03, 138.59, 139.40, 133.24, 131.36, 130.58, 129.24, 129.23,

+ + 129.11, 128.54, 127.96, 127.36, 75.71. HRMS (ESI) calculated for C20 H16O2H (M+H)

289.12231, found 289.12266.

4.1.5 3-Benzoylbenzhydrol thiosemicarbazone (4). 51 (132)

p-Toluenesulfonic acid monohydrate (0.007 g, 0.03 mmol) was added to a

solution of 3-benzoylbenzyhdrol (0.396 g, 1.09 mmol) in anhydrous tetrahydrofuran (10

mL). After stirring at reflux for 10 min, thiosemicarbazide (0.199 g, 2.19 mmol) was

added to the reaction mixture. After 2 days, tetrahydrofuran was removed under reduced

pressure and water (50 mL) was added. The products were extracted with ethyl acetate (2 x 20 mL) and the combined organic phases were washed with brine, dried over anhydrous sodium sulfate, and the solvent was removed under reduced pressure.

Purification using flash chromatography (silica gel, hexanes:ethyl acetate, gradient 88:12 to 0:100) afforded 3-benzoylbenzhydrol thiosemicarbazone (0.228 g, 0.631 mmol, 57%

1 yield) as a white solid. H NMR (500 MHz, acetone-d6): δ 8.51 (0.5H, brs), 8.50 (0.5H,

brs), 8.08 (0.5H, brs), 8.00 (0.5H, brs), 7.82 (0.5H, t, J = 1.7 Hz), 7.74-7.61 (4.5H m),

7.49-7.19 (10H, m), 5.97 (0.5H, d, J = 3.9 Hz), 5.82 (0.5H, d, J = 3.9 Hz), 5.10 (0.5H, d,

13 J = 3.9 Hz), 4.87 (0.5H, d, J = 3.9 Hz). C NMR (125 MHz, acetone-d6): δ 179.44,

179.41, 149.48, 149.42, 147.57, 145.86, 145.21, 144.96, 136.76, 136.61, 131.66, 131.35,

130.05, 129.81, 129.72, 129.70, 128.46, 128.29, 128.28, 128.19, 128.07, 127.96, 127.61,

127.07, 126.89, 126.55, 126.39, 126.35, 126.24, 125.55, 74.99, 74.48. HRMS (ESI)

48 + + calculated for C21H19N3OSNa (M+Na) 384.11410, found 384.11447. HPLC retention time (Method B): 11.73, 12.32. (Obtained as a mixture of E/Z isomers)

4.1.6 1,3-Bis(4-fluorobenzoyl)benzene (5).51,75 (132,177)

Aluminum trichloride (6.53 g, 49.5 mmol) was added to a solution of

isophthaloyl dichloride (5.00 g, 24.8 mmol) in dichloromethane (100 mL). After heating

at reflux for 30 min, fluorobenzene (3.41 g, 37.3 mmol) was added. After stirring for 12

h, the reaction mixture was poured onto crushed ice, neutralized with 10% NaOH (100

mL), and extracted with dichloromethane (3 × 100 mL). The combined organic layers

were washed with water, followed by brine and dried over anhydrous sodium sulfate.

After the organic layer was concentrated under reduced pressure, purification using flash

chromatography (silica gel, hexanes: ethyl acetate, 60:40) afforded 1,3-bis-(4-

fluorobenzoyl)benzene (3.37 g, 10.5 mmol, 56% yield) as a white solid. 1H NMR (500

MHz, CDCl3): δ 8.13 (1H, t, J = 1.7 Hz), 8.00 (2H, dd, J = 7.7 Hz, 1.7 Hz), 7.87 (4H, dd,

13 J = 8.9 Hz, 5.4 Hz), 7.64 (1H, t, J = 7.7 Hz), 7.18 (4H, m). C NMR (125 MHz, CDCl3):

δ 194.24, 165.65 (d, J = 255 Hz), 137.81, 133.32, 133.17 (d, J = 3.2 Hz), 132.71 (d, J =

19 9.3 Hz), 130.79, 128.66, 155.75 (d, J = 22 Hz). F NMR (470 MHz, CDCl3): δ 104.91

+ + (1F, tt, J = 8.3 Hz, 5.4 Hz). HRMS (ESI) calculated for C20H12F2O2H (M+H)

323.08781, found 323.08817.

4.1.7 1,3-Bis(4-bromobenzoyl)benzene (6).51,76 (132,178)

To a flask containing aluminum trichloride (1.36 g, 10.2 mmol) under nitrogen

was added bromobenzene (15 mL, 142.8 mmol). Isophthaloyl dichloride (1.00 g, 4.88

mmol) was dissolved in a minimal amount of bromobenzene and added to the reaction

flask. The reaction stirred at reflux for 6 h, after which it was stirred at room temperature

49 for 12 h, followed by another 3 h at reflux. The reaction was quenched with water (20

mL). The resulting mixture was then added to 1M HCl (50 mL) cooled in an ice bath.

The mixture was extracted with ethyl acetate (3 × 10 mL) and the combined organic

layers were washed sequentially with deionized water, dilute HCl, deionized water, and

brine. Crude product crashed out of the organic layers as a white solid and was collected

via filtration. Recrystallization of the solid from ethyl acetate afforded 1,3-bis(4- bromobenzoyl)benzene (0.671 g, 1.51 mmol, 31% yield). 1H NMR (500 MHz, DMSO-

d6): δ 8.04 (2H, dd, J = 7.8 Hz, 1.7 Hz), 7.98 (t, 1H, J = 1.7 Hz), 7.82-7.78 (4H, d, J =

8.6 Hz), 7.77 (t, 1H, J = 7.7 Hz), 7.74-7.71 (4H, d, J = 8.6 Hz). 13C NMR (125 MHz,

DMSO-d6): δ 194.08, 136.78, 135.52, 133.51, 131.75, 131.72, 130.44, 129.24, 121.10.

+ + HRMS (ESI) calculated for C20H12Br2O2H (M+H) 442.92768, found 442.92792.

4.1.8 1,3-Bis(4-methoxybenzoyl)benzene (7).51,77 (132,179)

Aluminum trichloride (5.60 g, 42.0 mmol) was added to a solution of isophthaloyl

dichloride (4.00 g, 19.7 mmol) in dichloromethane (50 mL) and heated to reflux for 30

min. Anisole (2.58 mL, 23.7 mmol) was added and the reaction mixture was refluxed for

20 h. The reaction mixture was poured onto the crushed ice. The resulting solution was

neutralized with 10% NaOH (100 mL), and extracted with dichloromethane (3 × 100

mL). The combined organic layers were washed with water, followed by brine, and dried

over anhydrous sodium sulfate. After the organic layer was concentrated under reduced

pressure, the purification using flash chromatography (silica gel, hexanes: ethyl acetate,

60:40) afforded 1,3-bis-(4-methoxy benzoyl)benzene (2.60 g, 11.9 mmol, 63% yield) as

1 a white solid. H NMR (500 MHz, CDCl3): δ 8.09 (1H, td, J = 1.7 Hz, 0.4Hz), 7.95 (2H, dd, J = 7.7 Hz, 1.7 Hz), 7.84 (4H, d, J = 8.9 Hz), 7.61 (1H, td, J = 7.7 Hz, 0.4 Hz), 6.97

50 13 (4H, d, J = 8.9 Hz), 3.88 (s, 6H). C NMR (125 MHz, CDCl3): δ 194.71, 163.50, 138.36,

132.72, 132.60, 130.61, 129.62, 128.38, 113.74, 55.54. HRMS (ESI) calculated for

+ + C22H18O4H (M+H) 347.12779, found 347.12817.

4.1.9 1,3-Bis(4-fluorobenzoyl)benzene thiosemicarbazone (8).51(132)

1,3-Bis-(4-fluorobenzoyl)benzene (0.347 g, 1.08 mmol) was dissolved in

anhydrous methanol (50 mL). The solution was heated to reflux for 15 min followed by

the addition of thiosemicarbazide (0.049 g, 0.54 mmol) and a catalytic amount of p-

toluenesulfonic acid monohydrate (0.020 g, 0.11 mmol). After 10 h at reflux, the solution

was concentrated under reduced pressure. Purification using flash chromatography (silica

gel, hexanes: ethyl acetate, 70:30) afforded the desired product (0.110 g, 0. 278 mmol,

1 52% yield) as a white solid. H NMR (500 MHz, CDCl3): δ 8.66 (1H, brs), 7.99-7.97

(0.5H, m), 7.95 (0.5H, m), 7.90 (1H, dd, J = 8.9 Hz, 5.4 Hz), 7.83 (1H, dd, J = 8.9 Hz,

5.4 Hz), 7.77-7.73 (1H, m), 7.69-7.66 (1H, m), 7.54-7.52 (0.5H, m), 7.51-7.46 (1.5H, m),

7.38 (1H, brs), 7.32-7.30 (2H, m), 7.23-7.20 (1H, m), 7.18-7.14 (1H, m), 7.07-7.04 (1H,

13 m), 6.35 (1H, brs). C NMR (125 MHz, CDCl3): δ 194.44, 193.71, 179.16, 179.08,

165.74 (d, J = 256 Hz), 165.63 (d, J = 255 Hz), 164.16 (d, J = 252 Hz), 163.74 (d, J =

253 Hz), 148.67, 139.19, 138.02, 137.00, 133.31 (d, J = 3.0 Hz), 132.90 (d, J = 3.0 Hz),

132.81 (d, J = 9.2 Hz), 132.71 (d, J = 9.1 Hz), 132.28, 132.19 (d, J = 3.2 Hz), 131.83,

131.37, 131.26, 130.73 (d, J = 8.5 Hz ), 130.34, 129.82, 129.75 (d, J = 8.6 Hz), 128.59,

128.49, 126.44 (d, J = 3.7 Hz), 117.56 (d, J = 22 Hz), 115.84 (d, J = 22 Hz), 115.81 (d, J

19 = 22 Hz), 115.65 (d, J = 22 Hz). F NMR (565 MHz, CDCl3): δ -104.48 (0.5 F, tt, J =

8.3 Hz, 5.4 Hz), -104.80 (0.5 F, tt, J = 8.3 Hz, 5.4 Hz), -108.02 (0.5 F, tt, J = 8.0 Hz, 5.5

+ Hz), -109.13 (0.5 F, tt, J = 8.2 Hz, 5.4 Hz). HRMS (ESI) calculated for C21H15F2N3OSH

51 (M-H)+ 396.09767, found 396.09741. HPLC retention time (Method A): 8.09, 8.37 min.

(Obtained as a mixture of E/Z isomers)

4.1.10 1,3-Bis(4-bromobenzoyl)benzene thiosemicarbazone (9).

1,3-Bis(4-bromobenzoyl)benzene (106 mg, 0.238 mmol) was dissolved in

anhydrous ethanol (2 mL) followed by the addition of p-toluenesulfonic acid

monohydrate (1.0 mg, 0.005 mmol) and thiosemicarbazide (21.7 mg, 0.238 mmol). The

reaction was carried out at 100 oC for 30 min under microwave irradiation. The solvent

was removed under reduced pressure and the crude reaction mixture was purified using

flash chromatography (silica gel, hexanes: dichloromethane: ethyl acetate, gradient,

50:50:0 to 40:50:10) to afford 1,3-bis(4-bromobenzoyl)benzene thiosemicarbazone (17.5

1 mg, 0.034 mmol, 14% yield). H NMR (500 MHz, DMSO-d6): δ 9.45 (0.5H, s), 9.11

(0.5H, s), 8.58 (0.5H, s), 8.56 (0.5H, s), 8.37 (0.5H, s), 8.32 (0.5H, s), 8.05 (0.5H, d, J =

8.1 Hz), 7.91 (0.5H, m), 7.82-7.69 (5H, m), 7.66-7.53 (5H, m), 7.34-7.31 (1H, m). 13C

NMR (125 MHz, DMSO-d6): δ 194.37, 194.28, 146.81, 146.80, 137.78, 136.86, 136.82,

135.70, 135.68, 135.58, 132.98, 132.69, 131.90, 131.72, 131.66, 131.64, 131.51, 131.32,

131.28, 130.91, 130.76, 130.56, 130.45, 130.12, 130.00, 129.58, 128.70, 128.48, 127.04,

+ + 126.97, 123.52, 123.29. HRMS (ESI) calculated for C21H15Br2N3OSNa (M+Na)

537.91948, found 537.91980. HPLC retention time (Method A): 13.78, 14.50 min. The

purity for the mono thiosemicarbazone product (9) is less than 95%. The major impurity is the bis-thiosemicarbazone product (10) which was inactive against cathepsin L and B.

The combined mono and bis thiosemicarbazone products are 94.8% at 254 nm and

greater than 95% at 320 nm (254 nm: mono thiosemicarbazone (9) 88.21% and bis

52 thiosemicarbazone (10) 6.56%. 320 nm: mono thiosemicarbazone (9) 93.99% and bis thiosemicarbazone (10) 5.63%). (Obtained as a mixture of E/Z isomers)

4.1.11 1,3-Bis(4-bromobenzoyl)benzene bis-thiosemicarbazone (10).51 (132)

1,3-Bis(4-bromobenzoyl)benzene (0.550 g, 1.21 mmol) was dissolved in dry THF

(20 mL). The solution was heated at reflux for 15 min, and thiosemicarbazide (0.220 g,

2.42 mmol) and a catalytic amount of p-toluenesulfonic acid monohydrate (0.023 g,

0.121 mmol) were added to the reaction mixture. After 24 h at reflux, the result solution

was concentrated under reduced pressure and the residue was dissolved in

dichloromethane (25 mL). The organic layer was washed with deionized water (15 mL),

dried over anhydrous sodium sulfate, and concentrated. The crude product was purified

via column chromatography to give the double condensation product in <1% isolated

yield (6.7 mg, 11.3 µmol). The 1H NMR for the major isomer is reported. The 13C NMR for the isomer mixture is reported. Three isomers exist for the bis thiosemicarbazones.1H

NMR (500 MHz, DMSO-d6): δ 9.87 (2H, s), 8.46 (2H, s), 8.29 (2H, s), 7.84 (1H, t, J =

7.5 Hz), 7.64 (4H, d, J = 8.5 Hz), 7.56 (4H, d, J = 7.5 Hz), 7.48 (2H, dd, J = 7.5 Hz, 1.5

13 Hz), 7.20 (1H, s). C NMR (125 MHz, DMSO-d6): δ 178.73, 147.42, 146.67, 146.45,

137.67, 136.03, 135.55, 133.37, 132.76, 132.36, 131.25, 131.17, 130.90, 130.76, 130.25,

129.93, 129.86, 129.56, 129.50, 129.43, 129.38, 129.32, 126.59, 123.53, 123.27, 123.09.

X-ray crystallographic data obtained for 1,3,5-trisbenzoylbenzene thiosemicarbazone has been deposited in the Cambridge Crystallographic Data Centre and was allocated the

+ deposition number CDCC 1046061. HRMS (ESI) calculated for C22H18Br2N6S2Na

(M+Na)+ 610.92933, found 610.92822. HPLC retention time (Method A): 10.19, 11.76,

12.16 min. (Obtained as a mixture of isomers)

53 4.1.12 1,3-Bis(4-methoxybenzoyl)benzene thiosemicarbazone (11).

1,3-Bis(4-methoxybenzoyl)benzene (0.346 g, 1.00 mmol), thiosemicarbazide

(0.091 g, 1.00 mmol) and p-toluenesulfonic acid monohydrate (0.010 g, 0.050 mmol)

were dissolved in tetrahydrofuran (10 mL) and the reaction was carried out at 90 oC for

45 min under microwave irradiation. The solvent was removed under reduced pressure

and the crude reaction mixture was purified using flash chromatography (silica gel,

hexanes: ethyl acetate, gradient, 88:12 to 40:60) to afford 1,3-bis(4-

methoxybenzoyl)benzene thiosemicarbazone (0.189 mg, 0.410 mmol, 45% yield). 1H

NMR (500 MHz, CDCl3): δ 8.807 (0.6H, s), 8.667 (0.4H, s), 7.973-7.952 (1H, m), 7.869

(0.8H, d, J = 8.8 Hz), 7.813 (1.2H, d, J = 8.9 Hz), 7.741-7.700 (1H, m), 7.677-7.654 (1H, m), 7.501 (0.4H, dt, J = 7.7 Hz, 1.5 Hz), 7.460-7.393 (2.4H, m), 7.245 (1.2H, d, J = 8.8

Hz), 7.082 (1.2H, d, J = 8.8 Hz), 7.008 (0.8H, J = 8.9 Hz), 6.958 (1.2H, d, J = 8.9 Hz),

6.864 (0.8H, d, J = 8.9 Hz), 6.353 (1H, s), 3.897 (1.8H, s), 3.894 (3H, s), 3.827 (1.2H, s).

13 C NMR (125 MHz, CDCl3): δ 194.85, 194.11, 178.87, 178.69, 163.64, 163.50, 161.55,

161.11, 150.18, 149.93, 139.83, 138.65, 137.24, 132.68, 132.61, 131.80, 131.49, 131.33,

131.15, 131.10, 130.07, 130.02, 129.78, 129.61, 129.39, 129.31, 128.71, 128.56, 128.20,

122.39, 115.48, 114.00, 113.93, 113.65, 55.58, 55.54, 55.50, 55.43. HRMS (ESI)

+ + calculated for C23H21N3O3SNa (M+Na) 442.11958, found 442.11960. HPLC retention

time (Method A): 7.21, 7.62 min (Obtained as a mixture of E/Z isomers)

4.1.13 1,3-Bis(4-hydroxybenzoyl)benzene (12).51,78 (132,180)

To a well-stirred solution of 1,3-bis-(4-methoxybenzoyl)benzene (1.80 g, 5.20

mmol) in dichloromethane (85 mL) was added boron trifluoride dimethyl sulfide

complex (BF3·SMe2, 15 mL). The mixture was stirred for 27 h. After the reaction was

54 quenched by water, the mixture was extracted with ethyl acetate (3 × 80 mL). The

combined organic layer was washed with brine, dried over anhydrous sodium sulfate, and

concentrated under vacuum. The resulting solid was further purified using flash

chromatography (silica gel, hexanes: ethyl acetate, 50:50) to afford the pure product

1 (0.620 g, 1.95 mmol, 38% yield) as a white solid. H NMR (500 MHz, acetone-d6): δ

9.28 (2H, s), 8.04 (1H, td, J = 1.7 Hz, 0.4 Hz), 7.98 (2H, dd, J = 7.7 Hz, 1.7 Hz), 7.80

(4H, m), 7.73 (1H, td, J = 7.7 Hz, 0.4 Hz), 7.00 (4H, m). 13C NMR (125 MHz, acetone-

d6): δ 194.46, 162.74, 139.45, 133.49, 133.13, 130.96, 129.71, 128.38, 116.11. HRMS

+ + (ESI) calculated for C20H14O4H (M+H) 319.09649, found 319.09671.

4.1.14 1,3-Bis(4-hydroxybenzoyl)benzene thiosemicarbazone (13).

1,3-Bis(4-hydroxybenzoyl)benzene (0.318 g, 1.00 mmol) was dissolved in

anhydrous ethanol (10 mL) followed by the addition of p-toluenesulfonic acid

monohydrate (0.005 g, 0.025 mmol) and thiosemicarbazide (0.910 g, 1.00 mmol). The

reaction was carried out at 100 oC for 45 min under microwave irradiation. The solvent

was removed under reduced pressure and the crude reaction mixture was purified using

flash chromatography (silica gel, dichloromethane: ethyl acetate, gradient, 90:10 to

60:40) to afford 1,3-bis(4-hydroxybenzoyl)benzene thiosemicarbazone (0.160 mg, 0.410

1 mmol, 50% yield). H NMR (600 MHz, DMSO-d6): δ 10.49 (0.4H, s), 10.48 (0.6H, s),

10.08 (0.6H, s), 9.90 (0.4H, s), 9.03 (0.4H, s), 8.62 (0.6H, s), 8.56 (0.6H, s), 8.45 (0.4H,

s), 8.40 (0.6H, s), 8.17 (0.4H, s), 8.06 (0.6H, d, J = 7.9 Hz), 7.85 (0.4H, dt, J = 7.8 Hz,

1.4 Hz), 7.78-7.74 (1.8H, m), 7.65-7.63 (1.8H, m), 7.56-7.51 (1H, m), 7.49-7.46 (1.2H, m), 7.20 (1.2H, d, J = 8.5 Hz), 7.00 (1.2H, d, J = 8.5 Hz), 6.91 (0.8H, d, J = 8.7 Hz), 6.87

13 (1.2H, d, J = 8.7 Hz), 6.75 (0.8H, d, J = 8.8 Hz). C NMR (150 MHz, DMSO-d6): δ

55 193.87, 193.65, 177.88, 177.72, 162.23, 162.19, 159.27, 158.78, 148.80, 148.71, 139.04,

138.27, 137.11, 132.83, 132.66, 131.96, 130.50, 130.22, 130.10, 130.03, 129.84, 129.50,

129.33, 128.40, 128.27, 127.60, 127.59, 127.31, 120.89, 116.57, 115.39, 115.31, 115.2.

+ + HRMS (ESI) calculated for C21H17N3O3SNa (M+Na) 414.08828, found 414.08841.

HPLC retention time (Method B): 6.33, 6.75 min. Analysis by 1H NMR shows that

compound 13 contains 1.42 wt% ethyl acetate (6 mol% ethyl acetate). Ethyl acetate was found to have an IC50 value >>10000 nM. (Obtained as a mixture of E/Z isomers)

4.1.15 1,3-Bis(4-isopropoxybenzoyl)benzene (intermediate for compound 14).51 (132)

Reactions were conducted using a commercially available microwave reactor

(Biotage). In a microwave vial, 1,3-bis-(4-hydroxybenzoyl)benzene (0.310 g, 0.975

mmol), isopropyl bromide (0.851 g, 6.92 mmol) and potassium carbonate (0.955 g, 6.91

mmol) were added to DMF (10 mL), with a magnetic stir bar. The vial was capped tightly

and the reaction mixture was heated from r.t. to 90 oC for 2 h. After the reaction mixture

was cooled to room temperature, the vial was opened and the mixture was transferred to a round bottom flask. The mixture was quenched with water (50 mL) and extracted with ether (2 × 50 mL). The organic layer was washed with brine, dried over anhydrous sodium sulfate, and concentrated under reduced pressure. The crude product was purified

using flash chromatography (silica gel, hexanes: ethyl acetate, 50:50) to afford the pure

product (0.23 g, 0.57 mmol, 59% yield) as a white solid. 1H NMR (500 MHz, acetone-

d6): δ 8.07 (1H, s), 7.99 (2H, dd, J = 7.6 Hz, 1.4 Hz), 7.84 (4H, d, J = 8.7 Hz), 7.71 (1H, t, J = 7.6 Hz), 7.05 (4H, d, J = 8.7), 4.76 (2H, Septet, J = 6.0 Hz), 1.35 (12H, d, J = 6.0

13 Hz). C NMR (125 MHz, acetone-d6): δ 193.51, 162.03, 138.41, 132.40, 130.25, 129.22,

56 + + 128.57, 115.07, 69.96, 21.33. HRMS (ESI) calculated for C26H26O4H (M+H)

403.19039, found 403.19055.

4.1.16 1,3-Bis(4-isopropoxybenzoyl)benzene thiosemicarbazone (14).51 (132)

1,3-Bis-(4-isopropoxybenzoyl)benzene (0.14 g, 0.39 mmol) was dissolved in anhydrous methanol (20 mL). The solution was heated at reflux for 15 min followed by the addition of thiosemicarbazide (0.043 g, 0.47 mmol) and a catalytic amount of p- toluenesulfonic acid monohydrate. After 10 h at reflux, the solvent was removed under vacuum and the resulting solid was further purified using flash chromatography (silica gel, hexanes: ethyl acetate, 50:50) to afford the desired product as a white solid (0.050

1 g, 0. 105 mmol, 27% yield). H NMR (500 MHz, CDCl3): δ 8.84 (0.6H, brs), 8.66 (0.4H,

brs), 7.96-7.94 (1H, m), 7.84 (0.8H, d, J = 8.9 Hz), 7.78 (1.2H, d, J = 8.9 Hz), 7.74-7.65

(2H, m), 7.49 (0.4H, m), 7.45-7.40 (2.4H, m), 7.21 (1.2H, d, J = 8.8 Hz), 7.03 (1.2H, d, J

= 8.8 Hz), 6.98-6.95 (0.8H, d, J = 8.9 Hz), 6.91 (1.2H, d, J = 8.9 Hz), 6.83 (0.8H, d, J =

8.9 Hz), 6.46 (1H, brs), 4.69-4.54 (2H, m), 1.40 (3.6H, d, J = 6.1 Hz), 1.38-1.37 (6H, m),

13 1.33 (2.4H, d, J = 6.1 Hz). C NMR (125 MHz, CDCl3): δ 194.80, 194.08, 178.87,

178.66, 162.19, 162.04, 159.98, 159.59, 150.32, 150.04, 139.91, 138.71, 137.28, 132.74,

132.66, 131.72, 131.40, 131.34, 131.09, 131.03, 130.09, 129.97, 129.77, 129.41, 129.15,

128.84, 128.79, 128.17, 128.16, 121.94, 116.79, 115.53, 115.263, 114.98, 70.21, 70.19,

+ + 70.17, 70.03, 22.01, 21.91. HRMS (ESI) calculated for C27H29N3O3SH (M+H)

476.20024, found 476.20059. HPLC retention time (Method A): 14.82, 15.31 min

(Obtained as a mixture of E/Z isomers)

57 4.1.17 5-(tert-butyldimethylsilyloxy)isophthalic acid (intermediate for compound 15).79

(181)

3-hydroxyisophthalic acid (5.46 g, 30.0 mmol), tert-butyldimethylchlorosilane

(22.50 g, 150.0 mmol), and imidazole (12.24 g, 180.0 mmol) were dissolved in N,N-

dimethylformamide (150 mL) and the reaction mixture was stirred for 11 h between 50-

60 oC. The reaction mixture was diluted with water (150 mL) and extracted with hexanes

(4 x 100 mL). The organic phase was dried over sodium sulfate and the solvent was

removed under reduced pressure. The crude product was dissolved in tetrahydrofuran (30

mL) followed by the addition of glacial acetic acid (90 mL), and distilled water (30 mL).

After stirring for 3 h at room temperature, the reaction mixture was diluted with distilled

water (60 mL) and cooled to 0 oC. The precipitated product was filtered. The filtrate was

concentrated and cooled again to 0 oC. The precipitated product was filtered and the

combined solids were dried to yield 5-(tert-butyldimethylsilyloxy)isophthalic acid (6.55 g,

1 22.1 mmol ,74% yield). H NMR (600 MHz, DMSO-d6): δ 13.34 (2H, s), 8.10 (1H, t, J =

1.5 Hz), 7.56 (2H, d, J = 1.5 Hz), 0.96 (9H, s), 0.22 (6H, s). 13C NMR (150 MHz,

DMSO-d6): δ 166.28, 155.35, 132.80, 124.46, 123.21, 25.49, 17.98, -4.60. HRMS (ESI)

+ + calculated for C14H20O5SiNa (M+Na) 319.09722, found 319.09741.

4.1.18 5-(tert-butyldimethylsilyloxy)isophthaloyl chloride (15). 79 (181)

Oxalyl chloride (2.04 mL, 27.8 mmol) was added dropwise to a solution of 5-

(tert-butyldimethylsilyloxy)isophthalic acid (2.82 g, 9.51 mmol) in dichloromethane (90

mL). N,N-Dimethylformamide (0.073 mL, 0.095 mmol) was added dropwise. The

reaction mixture was allowed to stir until the solution was transparent. After 2 h, the

solvent was removed under reduced pressure. The product was dissolved in anhydrous

58 dichloromethane (20 mL) and the solvent was removed under reduced pressure. This was

repeated two times to afford 5-(tert-butyldimethylsilyloxy)isophthaloyl chloride as yellow

solid. The product was used immediately without further purification. 1H NMR (500

MHz, CDCl3): δ 8.46 (1H, t, J = 1.6 Hz), 7.82 (2H, d, J = 1.6 Hz), 1.02 (9H, s), 0.28 (6H,

13 s). C NMR (125 MHz, CDCl3): δ 167.21, 156.95, 135.55, 128.34, 126.72, 25.67, 18.39,-

4.30.

4.1.19 N1,N3-dimethoxy-N1,N3-dimethyl isophthalamide (16).51(132)

Triethylamine (5.61 mL, 40.0 mmol) was added dropwise to a solution of N,O- dimethylhydroxylamine hydrochloride (2.93 g, 30.0 mmol) in anhydrous dichloromethane (35 mL) at 0 oC. After stirring for 10 min, a solution of isophthaloyl

dichloride (2.03 g, 10 mmol) dissolved in anhydrous dichloromethane (6 mL) was added

dropwise. The reaction mixture was returned to room temperature and stirred for 5 h. The

reaction was quenched with water (50 mL) and the products were extracted with

dichloromethane (2 x 50 mL). The combined organic phases were washed with brine,

dried over anhydrous sodium sulfate, and the solvent was removed under reduced

pressure. Purification using flash chromatography (silica gel, dichloromethane:methanol,

95:5) afforded N1,N3-dimethoxy-N1,N3-dimethyl isophthalamide (2.47 g, 9.77 mmol, 97%

1 yield). H NMR (500 MHz, CDCl3): δ 7.98 (1H, s), 7.76 (2H, dd, J = 7.8 Hz, 1.6 Hz),

13 7.44 (1H, t, J = 7.8 Hz), 3.53 (6H, s), 3.35 (6H, s). C NMR (125 MHz, CDCl3): δ

169.15, 133.96, 130.55, 128.20, 128.05, 61.27, 33.79. HRMS (ESI) calculated for

+ + C12H16N2O4H (M+H) 253.11828, found 253.11859.

59 4.1.20 5-((tert-butyldimethylsilyl)oxy)-N1,N3-dimethoxy-N1,N3-dimethyl isophthalamide

(17).

Triethylamine (5.35 mL, 38.1 mmol) was added dropwise to a solution of N,O- dimethylhydroxylamine hydrochloride (2.78 g, 28.5 mmol) in dichloromethane (50 mL) at 0 oC. A solution of 5-(tert-butyldimethylsilyloxy)isophthaloyl chloride in

dichloromethane (10 mL) was added dropwise and the reaction was allowed to return to

room temperature. After 6 h, the reaction was quenched with water (90 mL) and extracted

with dichloromethane (3 x 60 mL). The combined organic phases were dried over

anhydrous sodium sulfate and the solvent was removed under reduced pressure.

Purification using flash chromatography (silica gel, dichloromethane:methanol, gradient,

100:0 to 90:10) afforded 5-((tert-butyldimethylsilyl)oxy)-N1,N3-dimethoxy-N1,N3-

dimethyl isophthalamide as a white solid (3.40 g, 8.89 mmol, 93% yield over two steps).

1 H NMR (500 MHz, DMSO-d6): δ 7.38 (1H, t, J = 1.5 Hz), 7.15 (2H, d, J = 1.5 Hz), 3.52

13 (6 H, s), 3.26 (6H, s), 0.96 (9H, s), 0.21 (6H, s). C NMR (125 MHz, DMSO-d6): δ

167.52, 154.16, 135.50, 120.98, 120.26, 60.80, 25.45, 17.94, -4.66. HRMS (ESI)

+ + calculated for C18H30 N2O5SiH (M+H) 383.19968, found 383.20117.

4.1.21 1,3-Bis(3-methylbenzoyl)benzene (18).51,80 (132,182)

3-Bromotoluene (6.684 g, 39.08 mmol) was dissolved in anhydrous tetrahydrofuran (30 mL). The solution was stirred for 5 min and cooled to -78 oC followed by the dropwise addition of n-butyllithium in hexanes (2.5 M, 13.6 mL). After stirring for 3.5 h, a solution of N1,N3-dimethoxy-N1,N3-dimethyl isophthalamide

tetrahydrofuran (5 mL) was added dropwise and the solution was stirred for 1 h at -78 oC.

The reaction mixture was quenched with 1 M HCl (40 mL) and the products were

60 extracted with dichloromethane (2 x 50 mL). The combined organic phases were washed

with aqueous saturated aqueous sodium bicarbonate, dried over anhydrous sodium

sulfate, and the solvent was removed under reduced pressure. Purification using flash

chromatography (silica gel, hexanes:ethyl acetate, gradient, 95:5 to 70:30) afforded 1,3- bis(3-methylbenzoyl)benzene (2.47 g, 9.77 mmol, 72% yield). 1H NMR (500 MHz,

CDCl3): δ 8.17-8.15 (1H, m), 8.02 (2H, dd, J = 7.7 Hz, 1.7 Hz), 7.66-7.61 (3H, m), 7.59

(2H, d, J = 7.5 Hz), 7.42 (2H, d, J = 7.5 Hz), 7.37 (2H, t, J = 7.5 Hz), 2.43 (6H, s). 13C

NMR (125 MHz, CDCl3): δ 192.22, 138.55, 138.06, 137.20, 133.79, 133.54, 131.30,

+ + 130.59, 128.63, 128.40, 127.55, 21.53. HRMS (ESI) calculated for C22H18O2H (M+H)

315.13796, found 315.13833.

4.1.22 1,3-Bis(3-bromobenzoyl)-5-tert-butyldimethylsilyloxybenzene (19).

n-Butyllithium in hexanes (2.5M, 4.0 mL, 10 mmol) was added dropwise to a

solution of 1,3-dibromobenzne (2.53 mL, 20.9 mmol) in tetrahydrofuran (50 mL) cooled

to -78 oC. After 30 min, a solution of 5-((tert-butyldimethylsilyl)oxy)-N1,N3-dimethoxy-

N1,N3-dimethyl isophthalamide (2.00g, 5.23 mmol) in tetrahydrofuran (10 mL) was added

dropwise to the reaction mixture. After stirring for 2 h, the reaction mixture was

quenched with 1 M HCl (30 mL). The product was extracted with ethyl acetate (3 x 50

mL). The combined organic phases were dried over anhydrous sodium sulfate and the

solvent was removed under reduced pressure. Purification using flash chromatography

(silica gel, hexane:ethyl acetate, gradient, 98:2 to 90:10) afforded 1,3-bis(3-

bromobenzoyl)-5-tert-butyldimethylsilyloxybenzene as a yellow oil (1.63 g, 2.84 mmol,

1 57% yield). H NMR (500 MHz, CDCl3): δ 7.95 (2H, t, J = 1.7 Hz), 7.75-7.70 (4H, m),

7.68 (1H, t, J = 1.5 Hz), 7.45 (2H, d, J = 1.5 Hz), 7.38 (2H, t, J = 7.8 Hz), 1.0 (9H, s),

61 13 0.26 (6H, s). C NMR (125 MHz, CDCl3): δ 193.76, 156.14, 138.75, 138.57, 135.77,

132.76, 130.04, 128.50, 125.17, 124.18, 122.79, 25.80, 18.24, -4.35. HRMS (ESI)

+ + calculated for C26H26Br2 O3SiH (M+H) 573.00907, found 573.00909.

4.1.23 1,3-Bis(3-methylbenzoyl)benzene thiosemicarbazone (20).51(132)

p-Toluenesulfonic acid monohydrate (0.001 g, 0.01 mmol) was added to a

solution of 1,3-bis(3-methylbenzoyl)benzene (0.154 g, 0.480 mmol) in anhydrous methanol (10 mL). After stirring at reflux for 10 min, thiosemicarbazide (0.022 g, 0.24 mmol) was added to the reaction mixture and stirred for 7 h under an inert atmosphere of nitrogen gas. After 7 h, methanol was removed under reduced pressure and 10 mL of water was then added. The products were extracted with ethyl acetate and the combined

organic phases were washed with brine, dried over anhydrous sodium sulfate, and the

solvent was removed under reduced pressure. Purification using flash chromatography

(silica gel, hexanes:ethyl acetate, gradient 90:10 to 40:60) afforded 1,3-bis(3-

methylbenzoyl)benzene thiosemicarbazone (0.057 g, 0.015 mmol, 64% yield) as a

1 yellow solid. H NMR (500 MHz, DMSO-d6): δ 9.17 (0.4H, s), 8.66 (0.6H, s), 8.55

(0.4H, s), 8.48 (0.6H, s), 8.43 (0.6H, s), 8.27 (0.4H, s), 8.21 (0.6H, d, J = 7.8 Hz), 7.93

(0.4H, dt, J = 7.8 Hz, 1.5 Hz), 7.81 (0.4H, t, J = 7.6 Hz), 7.74-7.71 (1.2 H, m), 7.64-7.39

(7H, m), 7.33-7.30 (0.4 H, m), 7.27-7.17 (2H, m), 2.40 (1.8 H, s), 2.38 (1.2H, s), 2.36

13 (1.8H, s), 2.31 (1.2H, s). C NMR (125 MHz, DMSO-d6): δ 195.32, 195.31, 148.25,

148.14, 139.47, 138.18, 138.08, 137.98, 137.63, 137.27, 136.74, 136.70, 136.59, 136.34,

133.58, 133.51, 132.61, 131.91, 130.93, 130.78, 130.76, 130.64, 130.60, 130.46, 130.02,

129.99, 129.92, 129.75, 129.73, 128.66, 128.62, 128.50, 128.37, 128.21, 127.83, 127.18,

127.01, 125.27, 125.16, 20.97, 20.93, 20.88, 20.82. HRMS (ESI) calculated for

62 + + C23H21N3OSNa (M+Na) 410.12975, found 410.13013. HPLC retention time (Method

A): 11.38 min. (Obtained as a mixture of E/Z isomers)

4.1.24 1,3-Bis(3-bromobenzoyl)-5-tert-butyldimethylsilyloxybenzene thiosemicarbazone

(21).

1,3-Bis(3-bromobenzoyl)-5-tert-butyldimethylsilyloxybenzene (1.39 g, 2.43

mmol), thiosemicarbazide (0.166 g, 1.82 mmol), and p-toluenesulfonic acid monohydrate

(0.010 g, 0.05 mmol) were dissolved in tetrahydrofuran (15 mL). The reaction was carried out at 90 oC for 30 min under microwave irradiation. The solvent was removed

under reduced pressure and the crude reaction mixture was purified using flash

chromatography (silica gel, hexanes:ethyl acetate, gradient, 93:7 to 50:50) to afford 1,3-

bis(3-bromobenzoyl)-5-tert-butyldimethylsilyloxybenzene thiosemicarbazone (0.418 g,

1 0.650 mmol, 36% yield). H NMR (500 MHz, DMSO-d6): δ 9.35 (0.7H, s), 9.12 (0.3H,

s), 8.57 (1H, m), 8.44 (0.7H, s), 8.34 (0.3H, s), 8.04 (0.7H, t, J = 1.7 Hz), 7.94 (0.7H, t, J

= 1.7 Hz), 7.89-7.85 (1H, m), 7.83-7.80 (1H, m), 7.78 (0.3H, ddd, J = 8Hz, 2Hz, 1Hz),

7.70-7.68 (0.3H, m), 7.62-7.60 (0.3H, t, J = 1.7 Hz), 7.59-7.57 (1H, m), 7.56-7.48 (1.3H, m), 7.47-7.44 (0.7H, m), 7.39 (0.3H, dt, J = 7.6 Hz, 1.2 Hz), 7.33-7.29 (1.7H, m), 7.20

(0.7H, t, J = 1.5 Hz), 7.13-7.11 (1H, m), 0.96 (6H, s), 0.91 (3H, s), 0.24 (4H, s), 0.15(2H,

13 s). C NMR (125 MHz, DMSO-d6): δ 193.42, 193.38, 178.53, 155.99, 155.02, 146.05,

145.55, 139.32, 138.75, 138.66, 138.65, 138.44, 138.02, 135.57, 135.51, 133.66, 132.97,

132.90, 132.28, 132.02, 131.85, 131.69, 131.26, 130.80, 130.72, 130.36, 129.48, 128.90,

128.68, 127.68, 126.89, 124.47, 123.04, 122.84, 122.75, 122.12, 122.07, 121.85, 121.82,

121.80, 121.48, 25.52, 25.51, 18.00, 17.96, -4.59. HRMS (ESI) calculated for

+ + C27H29Br2N3O2SSiH (M+H) 646.01893, found 646.01995.

63 4.1.25 1,3-Bis(3-bromobenzoyl)-5-hydroxybenzene thiosemicarbazone (22).

1,3-Bis(3-bromobenzoyl)-5-hydroxybenzene (0.388 g, 0.600 mmol) was

dissolved in tetrahydrofuran (5 mL) followed by the dropwise addition of a 1M solution

of tetra-butylammonium fluoride in tetrahydrofuran (1.20 mL, 1.20 mmol) at room temperature. After 1 h, the reaction mixture was diluted with ethyl acetate (30 mL) and washed with brine (15 mL). The combined organic phases were dried over anhydrous sodium sulfate and the solvent was removed under reduced pressure. Purification using flash chromatography (silica gel, gradient, hexane:ethyl acetate: methanol, 77.5:20:2.5 to

57.5:40:2.5) afforded 1,3-bis(3-bromobenzoyl)-5-hydroxybenzene thiosemicarbazone as

1 a yellow solid (0.277 g, 0.520 mmol, 87% yield). H NMR (500 MHz, DMSO-d6): δ

10.39 (0.6H, s), 9.98 (0.4H, s), 9.24 (0.6H, s), 9.07 (0.4H, s), 8.62-8.54 (1H, m), 8.46

(0.6H, s), 8.28 (0.4H, s), 8.09 (0.6H, s), 7.94 (0.6H, t, J = 1.8 Hz), 7.88-7.84 (1H, m),

7.83-7.79 (1H, m), 7.77 (0.4H, ddd, J = 8.1 Hz, 2.0 Hz, 1.0 Hz), 7.69-7.66 (0.4H, m),

7.61-7.47 (2.4 H, m), 7.44 (0.6H, d, J = 7.8 Hz), 7.39-7.35 (0.8H, m), 7.34-7.27 (1.6H, m), 7.12 (0.4H, dd, J = 2.2 Hz, 1.0 Hz), 7.00-6.99 (0.6H, m), 6.97 (0.6H, dd, J = 2.2 Hz,

13 1.5 Hz). C NMR (125 MHz, DMSO-d6): δ 193.74, 193.73, 178.41, 158.46, 157.38,

146.50, 146.10, 139.12, 138.98, 138.91, 138.58, 138.16, 137.65, 135.43, 135.33, 133.81,

132.85, 132.59, 132.32, 131.91, 131.72, 132.71, 131.17, 130.75, 130.65, 130.40, 129.36,

128.86, 128.66, 127.71, 127.04, 122.82, 122.17, 121.87, 121.82, 120.40, 119.89, 119.52,

+ + 118.49, 117.60, 117.29. HRMS (ESI) calculated for C21H15Br2N3O2SNa (M+Na)

553.91439, found 553.91436. HPLC retention time (Method B): 15.75, 16.88 min.

(Obtained as a mixture of E/Z isomers)

64 4.1.26 3-(tert-butyldimethylsilyloxy)benzaldehyde (23).81 (182)

3-Hydroxybenzaldehyde (2.00 g, 16.4 mmol) was dissolved in N,N-

dimethylformamide (50 mL) followed by the addition of imidazole (2.23 g, 32.8 mmol).

The reaction mixture was cooled to 0 oC and tert-butyldimethylchlorosilane (3.69 g, 24.6

mmol) was added. The reaction mixture was returned to room temperature and stirred for

3 h. After reaction completion, the reaction mixture was quenched with saturated aqueous

sodium bicarbonate (50 mL) and the product was extracted with hexanes (2 X 50 mL).

The organic extracts were dried over anhydrous sodium sulfate and concentrated under

reduced pressure. The crude mixture was purified using flash chromatography (silica gel,

hexanes:ethyl acetate, gradient, 100:00 to 90:10) to afford 3-(tert-

butyldimethylsilyloxy)benzaldehyde (3.57, 15.1 mmol , 92% yield). 1H NMR (500 MHz,

CDCl3): δ 9.95 (1H, s), 7.47 (1H, dt, J = 7.3 Hz, 1.2 Hz), 7.40 (1H, t, J = 7.8 Hz), 7.34-

7.32 (1H, m), 7.12-7.09 (1H, m), 1.0 (9H, m), 0.23 (6H, m). 13C NMR (125 MHz,

CDCl3): δ 192.23, 156.54, 138.07, 130.21, 126.68, 123.69, 120.01, 25.76, 18.34, -4.28.

+ + HRMS (ESI) calculated for C13H20O2SiH (M+H) 237.13053, found 237.13066.

4.1.27 2-Fluoro-N-methoxy-N-methyl-benzamide (24).51,82 (132,183)

Triethylamine (5.32 mL, 37.8 mmol) was added dropwise to a solution of N,O- dimethylhydroxylamine hydrochloride (2.77 g, 28.4 mmol) in anhydrous dichloromethane (45 mL) at 0 oC. After stirring for 10 min, 2-flurobenzoyl chloride

(0.303 mL, 2.52 mmol) in anhydrous dichloromethane (15 mL) was added dropwise. The

reaction mixture was returned to room temperature and stirred for 5 h. The reaction

mixture was quenched with water (60 mL) and the products were extracted with

dichloromethane (2 x 60 mL). The combined organic phases were washed with brine,

65 dried over anhydrous sodium sulfate, and the solvent was removed under reduced

pressure. Purification using flash chromatography (silica gel, hexanes:ethyl acetate, gradient 93:7 to 60:40) afforded 2-Fluoro-N-methoxy-N-methyl-benzamide (3.12 g, 17.0

1 mmol, 90% yield). H NMR (500 MHz, CDCl3): δ 7.44-7.38 (2H, m), 7.19 (1H, t, J = 7.5

Hz), 7.10 (1H, t, J = 8.9 Hz), 3.55 (3H, brs), 3.35 (3H, brs). 13C NMR (125 MHz,

CDCl3): δ 166.40, 158.66, (d, J = 249 Hz), 131.50, 128.90, 124.11, 123.52 (d, J = 17 Hz),

19 - 115.69 (d, J = 21 Hz), 61.21, 32.31. F NMR (470 MHz, CDCl3): δ 114.04 (1F, s).

+ + HRMS (ESI) calculated for C9H10FNO2H (M+H) 184.07683, found 184.07702.

4.1.28 3-Bromo-N-methoxy-N-methyl-benzamide (25).83 (184)

To a well stirred suspension of N,O-dimethylhydroxylamine hydrochloride (5.33

g, 54.7 mmol) in dichloromethane (120 mL) was added triethylamine (10.2 mL, 72.9

mmol) dropwise at 0 oC. After stirring for a few minutes, 3-bromobenzoyl chloride (4.81

mL, 36.4 mmol) in dichloromethane (20 mL) was added dropwise. The reaction mixture

was allowed to warm to room temperature and stirred for 4 h. The reaction mixture was

quenched with water (150 mL) and extracted with dichloromethane (3 x 100 mL). The

combined organic phases were dried over anhydrous sodium sulfate and concentrated.

Purification using flash chromatography (silica gel, hexane: ethyl acetate, gradient, 90:10

to 60:40) afforded 3-bromo-N-methoxy-N-methyl-benzamide as a colorless oil (8.60 g,

1 35.2 mmol, 97% yield). H NMR (500 MHz, CDCl3): 7.83 (1H, t, J = 1.8 Hz), 7.61 (1H,

dt, J = 7.7 Hz, 1.3 Hz), 7.59 (1H, ddd, J = 8.0 Hz, 2.1 Hz, 1.1 Hz), 7.28 (1H, m), 3.35

13 (3H, s), 3.36 (3H, s). C NMR (125 MHz, CDCl3): 168.32, 136.06, 133.69, 131.36,

+ 129.74, 126.93, 122.14, 61.33, 33.71. HRMS (ESI) calculated for C9H10BrNO2H

(M+H)+ 243.99677, found 243.99704.

66 4.1.29 1,3-Bis[(3-tert-butyldimethylsilyloxyphenyl)hydroxymethyl]-5-bromobenzene (26).

Tert-butyllithium in pentane (1.7M, 11.76 mL) was added dropwise to a solution of 1,3,5 tribromobenzene (1.57 g, 5.00 mmol) in anhydrous diethyl ether (20 mL) cooled to -78 oC. The mixture was sonicated at every 30 min interval. After 2 h, a solution of 3-

(tert-butyldimethylsilyloxy)benzaldehyde (2.37 g, 10.0 mmol) in diethyl ether (10 mL) was added dropwise and the reaction mixture was allowed to slowly warm to room temperature over a period of 19 h. The reaction was quenched with 1 M HCl (30 mL) and extracted with diethyl ether (3 X 30 mL). The combined organic extracts were washed with saturated aqueous sodium bicarbonate, followed by brine and dried over anhydrous sodium sulfate. After removing the solvent under reduced pressure, the crude mixture was purified using flash chromatography (silica gel, hexanes:ethyl acetate, gradient,

100:00 to 40:60) to afford 1,3-bis[(3-tert-butyldimethylsilyloxyphenyl)hydroxymethyl]-

1 5-bromobenzene (1.16 g, 1.85 mmol, 37% yield). H NMR (600 MHz, CDCl3): δ 7.411-

7.390 (2H, m), 7.357 (0.5H, s), 7.327 (0.5H, s), 7.206-7.169 (2H, m), 6.905-6.881 (2H, m), 6.836-6.881 (2H, m), 6.768-6.741 (2H, m), 5.707-5.704 (2H, m), 2.196 (2H, s), 0.964

13 (9H, s), 0.963 (9H, s), 0.168 (6H, s), 0.166 (6H, s). C NMR (150 MHz, CDCl3): δ

155.89, 145.94, 145.89, 144.59, 144.56, 129.68, 129.67, 128.62, 128.51, 123.21, 122.71,

122.65 119.58, 119.56, 119.45, 119.43, 118.27, 118.23, 75.39, 25.68, 18.21, -4.40.

+ + HRMS (ESI) calculated for C32H45BrO4Si2Na (M+Na) 651.19320, found 651.19128.

4.1.30 1,3-Bis[(3-tert-butyldimethylsilyloxy)benzoyl]-5-bromobenzene (27).

1,3-Bis[(3-tert-butyldimethylsilyloxyphenyl)hydroxymethyl]-5-bromobenzene

(1.10 g, 1.75 mmol) in dichloromethane (5 mL) was added dropwise to a solution of pyridinium chlorochromate (1.13 g, 5.24 mmol), celite (1.00 g), and dichloromethane (5

67 mL) at 0 oC. The reaction mixture was returned to room temperature. After 4.5 h, the

reaction mixture was filtered over a pad of celite. The celite was rinsed several times with

dichloromethane and the filtrate was concentrated. The solvent was removed under

reduced pressure and the crude reaction mixture was purified using flash chromatography

(silica gel, hexanes:ethyl acetate, gradient, 90:08 to 20:80) to afford 1,3-bis[(3-tert-

butyldimethylsilyloxy)benzoyl]-5-bromobenzene (1.02 g, 1.63 mmol, 93% yield). 1H

NMR (500 MHz, CDCl3): δ 8.12 (2H, d, J = 1.5 Hz), 8.08 (1H, t, J = 1.5 Hz), 7.38-7.34

(4H, m), 7.26-7.25 (2H, m), 7.11-7.08 (2H, m), 0.98 (18H, s), 0.21 (12H, s). 13C NMR

(125 MHz, CDCl3): δ 193.91, 155.94, 139.51, 137.70, 135.96, 129.68, 129.51, 125.21,

123.25, 122.65, 121.14, 25.63, 18.22, -4.39. HRMS (ESI) calculated for

+ + C32H41BrO4Si2H (M+H) 625.17995, found 625.17969.

4.1.31 1,3-Bis(2-fluorobenzoyl)-5-bromobenzene (28).51 (132)

Tert-butyllithium in pentane (1.6M, 7.72 mL) was added dropwise to a solution of

1,3,5-tribromobenzene (0.972 g, 3.09 mmol) in anhydrous ether (30 mL) at -78 oC under a flow of nitrogen gas. After 2 h, 2-fluoro-N-methoxy-N-methyl-benzamide (1.132 g,

6.18 mmol) dissolved in anhydrous ether (5 mL) was added dropwise and the reaction

mixture was allowed to slowly come to room temperature and stirred for 24 h. After 24

h, the reaction mixture was quenched with water (30 mL) and the products were extracted

with ether (2 x 50 mL). The combined organic phases were washed with brine, dried

over anhydrous sodium sulfate, and the solvent was removed under reduced pressure.

Purification using flash chromatography (silica gel, hexanes:ethyl acetate, gradient 95:5

to 60:40) afforded 1,3-bis(3-fluorobenzoyl)-5-bromobenzene (0.617 g, 6.18 mmol, 50%

1 yield). H NMR (500 MHz, CDCl3): δ 8.16 (2H, m), 8.12 (1H, pentet, J = 1.4 Hz), 7.61-

68 7.56 (4H, m), 7.30 (2H, td, J = 7.5 Hz, 1.0Hz), 7.20-7.16 (2H, m). 13C NMR (125 MHz,

CDCl3): δ 191.23, 160.40 (d, J = 254 Hz), 139.67, 133.61, 134.29 (d, J = 8.6 Hz), 131.12

(d, J = 1.9 Hz), 129.31, 125.79 (d, J = 13.8 Hz), 124.79 (d, J = 3.8 Hz), 123.18, 116.70

19 - - (d, J = 21 Hz). F NMR (470 MHz, CDCl3): δ 109.73 - 109.78 (1F, m). HRMS (ESI)

+ + calculated for C20H11 BrF2O2H (M+H) 400.99833, found 400.99857.

4.1.32 1,3-Bis(3-bromobenzoyl)-5-bromobenzene (29).

Tert-butyllithium (1.7 M, 11.8 mL) was added dropwise to a solution of 1,3,5- tribromobenzene in diethyl ether (25 mL) at -78o C. The mixture was sonicated for 0.5

min at 20 min intervals to help dissolve the starting material. After 1 h, a solution of 3-

bromo-N-methoxy-N-methyl-benzamide (2.44 g, 10.0 mmol) in tetrahydrofuran (15 mL) was added to the reaction mixture and allowed to slowly come to room temperature overnight. The reaction was quenched with 1 M HCl (30 mL) and extracted with diethyl ether (3 x 30 mL). The combined organic phases were washed with saturated aqueous sodium bicarbonate and concentrated. Crystallization of the crude mixture in acetone followed by recrystallization in ethanol afforded 1,3-bis(3-bromobenzoyl)-5- bromobenzene as a white solid (1.77 g, 0.338 mmol, 68% yield). 1H NMR (500 MHz,

DMSO-d6): δ 8.17 (2H, d, J = 1.5 Hz), 7.96 (2H, t, J = 1.7 Hz), 7.92-7.98 (3H, m), 7.81-

13 7.79 (2H, m), 7.55 (2H, t, J = 7.9 Hz). C NMR (125 MHz, DMSO-d6): δ 192.28,

138.60, 138.18, 135.88, 135.54, 131.98, 130.86, 129.61, 128.93, 122.31, 122.04. HRMS

+ + (ESI) calculated for C20H11 Br3O2H (M+H) 520.83819, found 520.83820.

69 4.1.33 1,3-Bis[(3-tert-butyldimethylsilyloxy)benzoyl]-5-bromobenzene thiosemicarbazone

(30).

1,3-Bis[(3-tert-butyldimethylsilyloxy)benzoyl]-5-bromobenzene (0.890 g, 1.42

mmol ), thiosemicarbazide (0.970 g, 1.06 mmol) and p-toluenesulfonic acid monohydrate

(0.135 g, 0.0710 mmol) were dissolved in tetrahydrofuran (10 mL) and the reaction was

carried out at 90 oC for 30 min under microwave irradiation. The solvent was removed

under reduced pressure and the crude reaction mixture was purified using flash

chromatography (silica gel, hexanes:ethyl acetate, gradient, 90:10 to 20:80) to afford 1,3-

bis[(3-tert-butyldimethylsilyloxy)benzoyl]-5-bromobenzene thiosemicarbazone as a

1 yellow solid (0.329 g, 0.471 mmol , 33% yield). H NMR (500 MHz, CDCl3): δ 8.77

(1H, s), 7.87 (2H, d, J = 1.7 Hz), 7.83 (1H, t, J = 1.7 Hz), 7.48-7.44 (1H, m), 7.37-7.32

(2H, m), 7.30 (1H, dt, J = 7.6 Hz, 1.5 Hz), 7.23-7.22 (1H, m), 7.08 (1H, ddd, J = 7.8 Hz,

2.5 Hz, 1.5 Hz), 7.02 (1H, ddd, J = 8.3 Hz, 2.5 Hz, 1.0 Hz), 6.86 (1H, dt, J = 7.6 Hz, 1.0

Hz), 6.71-6.70 (1H, m), 6.44 (1H, s), 0.99 (9H, s), 0.98 (9H, s), 0.23 (6H, s), .22 (6H, s).

13 C NMR (125 MHz, CDCl3): δ 194.25, 179.26, 157.34, 156.02, 148.09, 139.76, 138.89,

137.90, 134.01, 133.88, 131.74, 131.25, 129.70, 127.55, 125.27, 123.35, 122.80, 122.70,

121.28, 120.99, 119.99, 29.84, 25.77, 18.35, -4.17, -4.24. HRMS (ESI) calculated for

+ + C33H44BrN3O3SSi2H (M+H) 698.18981, found 698.18915.

4.1.34 1,3-Bis(3-hydroxybenzoyl)-5-bromobenzene thiosemicarbazone (31).

1,3-Bis[(3-tert-butyldimethylsilyloxy)benzoyl]-5-bromobenzene

thiosemicarbazone (0.325 g, 0.465 mmol) was dissolved in tetrahydrofuran (10 mL)

followed by the addition of tetra-butylammonium fluoride trihydrate (0.604 g, 1.91 mmol) at room temperature. After 1.5 h, the reaction mixture was diluted with ethyl

70 acetate (50 mL), washed with brine (50 mL), and dried over anhydrous sodium sulfate.

After concentrating the reaction mixture under reduced pressure, the crude product was

purified using flash chromatography (silica gel, hexanes: dichloromethane, gradient,

50:50 to 0:100 followed by dichloromethane: ethyl acetate, gradient 100:0 to 60:40) to afford 1,3-bis(3-hydroxy-benzoyl)-5-bromobenzene thiosemicarbazone as a yellow solid

1 (0.158 mg, 0.336 mmol, 72% yield). H NMR (500 MHz, DMSO-d6) : δ 9.99 (0.8H, s),

9.92 (0.2H, s), 9.88 (0.2H, s), 9.86 (0.8H, s), 9.47 (0.2H, s), 8.72 (0.8H, s), 8.70 (0.8H, s),

8.53 (0.8H, s), 8.48 (0.8H, s), 8.45 (0.2H, s), 8.09 (0.2H,s), 7.98 (0.2H, t, J = 1.6 Hz),

7.81-7.79 (1H, m), 7.62 (0.8H, J = 1.5 Hz), 7.47-7.44 (1H, m), 7.37 (0.2H, t, J = 7.8 Hz),

7.34-7.31 (0.8H, m), 7.27-7.24 (0.4H, m), 7.17 (0.2H, t, J = 7.9 Hz), 7.13-7.04 (2.8H, m),

6.98 (0.8H, ddd, J = 8.3 Hz, 2.4 Hz, 0.65 Hz), 6.89 (0.2H, t, J = 2.0 Hz), 6.80 (0.2H, ddd,

J = 7.9 Hz, 2.4 Hz, 1.1 Hz), 6.78-6.76 (0.8H, m), 6.72-6.70 (0.8H, m). 13C NMR (125

MHz, DMSO-d6): δ 193.79, 193.70, 177.89, 158.50, 157.55, 157.48, 157.23, 146.51,

146.47, 140.20, 139.30, 138.72, 137.60, 137.38, 137.16, 134.93, 134.80, 132.63, 132.49,

132.46, 131.38, 131.25, 129.77, 129.70, 129.37, 128.84, 127.44, 122.68, 122.10, 121.38,

120.38, 120.50, 120.46, 188.48, 117.36, 116.81, 115.84, 115.79, 114.55. HRMS (ESI)

+ + calculated for C21H16BrN3O3SNa (M+Na) 491.99880, found 491.99903. HPLC

retention time (Method B): 9.97, 10.34 min. (Obtained as a mixture of E/Z isomers)

4.1.35 1,3-Bis(2-fluorobenzoyl)-5-bromobenzene thiosemicarbazone (32).51 (132)

p-Toluenesulfonic acid monohydrate (0.006 g, 0.03 mmol) was added to a

solution of 1,3-bis(3-fluorobenzoyl)-5-bromobenzene (0.190 g, 0.473 mmol) in

anhydrous tetrahydrofuran (15 mL). After stirring at reflux for 10 min, thiosemicarbazide

(0.088g, 0.97 mmol) was added to the reaction mixture and stirred for 28 h under an inert

71 atmosphere of nitrogen gas. After 28 h, tetrahydrofuran was removed under reduced

pressure and 10 mL of water was then added. The products were extracted with ethyl

acetate (2 x 50 mL) and the combined organic phases were washed with brine, dried over

anhydrous sodium sulfate, and the solvent was removed under reduced pressure.

Purification using flash chromatography (silica gel, hexanes:ethyl acetate, gradient 90:11

to 30:70) afforded 1,3-bis(3-fluorobenzoyl)-5-bromobenzene thiosemicarbazone (0.068

1 g, 0.143 mmol, 30% yield). H NMR (500 MHz, acetone-d6): δ 8.92 (1H, brs), 8.44 (1H, t, J = 1.8 Hz), 8.33 (1H, brs), 7.92 (1H, q, J = 1.5 Hz), 7.86 (1H, brs), 7.74-7.64 (3H, m),

7.59 (1H, td, J = 7.4 Hz, 1.8 Hz), 7.54-7.48 (2H, m), 7.46-7.41 (1H, m), 7.36 (1H, td, J =

7.5 Hz, 1.0 Hz), 7.27 (1H, ddd, J = 10 Hz, 8.4 Hz, 1.0 Hz). 1H NMR (400 MHz, acetone-

19 d6, F decoupled) 8.92 (1H, brs), 8.44 (1H, t, J = 1.8 Hz), 8.33 (1H, brs), 7.92 (1H, t, J =

1.7 Hz), 7.85 (1H, brs), 7.74-7.64 (3H, m), 7.59 (1H, dd, J = 7.7 Hz, 1.8 Hz), 7.57-7.47

(2H, m), 7.43 (1H, d, J = 8.4 Hz), 7.36 (1H, td, J = 7.5 Hz, 1.0 Hz), 7.27 (1H, dd, J = 8.4

13 Hz, 1.0 Hz). C NMR (125 MHz, acetone-d6): δ 191.49, 180.88, 160.87 (d, J = 251 Hz),

160.34 (d, J = 248 Hz), 141.81, 140.39, 140.01, 134.98 (d, J = 8.6), 134.13 (d, J = 8.3

Hz), 134.01, 133.79 (d, J = 1.4 Hz), 131.69 (d, J = 2.5 Hz), 131.59 (d, J = 3.1 Hz),

127.98 (d, J = 0.83 Hz), 126.83 (d, J = 3.5 Hz), 126.65 (d, J = 14 Hz), 125.59 (d, J = 3.5

Hz), 123.57, 118.89 (d, J = 17 Hz), 117.82 (d, J = 21 Hz), 117.11 (d, J = 22 Hz). ). 19F

- - - - NMR (470 MHz, acetone-d6): δ 112.74 - 112.79 (1F, m), 113.57 - 113.67 (1F, m).

+ + HRMS (ESI) calculated for C21H14BrF2 N3OSH (M+H) 474.00818, found 474.00845.

HPLC retention time (Method A): 10.19 min.

72 4.1.36 1,3-Bis(3-bromobenzoyl)-5-bromobenzene thiosemicarbazone (33).

1,3-Bis(3-bromobenzoyl)-5-bromobenzene (0.523 g, 1.00 mmol) was dissolved in

tetrahydrofuran (10 mL) followed by the dropwise addition of titanium isopropoxide

(1.18 mL, 4.00 mmol). The reaction was stirred for 10 min followed by the addition of

thiosemicarbazide (0.182 g, 2.00 mmol). After refluxing for 2 h, the reaction mixture was

quenched with 1 M HCl (20 mL) and extracted with ethyl acetate (3 x 40 mL). The combined organic phases were washed saturated aqueous sodium bicarbonate (20 mL) and dried over anhydrous sodium sulfate. Purification using flash chromatography (silica gel, hexane: ethyl acetate, gradient, 90:10 to 20:80) afforded 1,3-bis(3-bromobenzoyl)-5- bromobenzene thiosemicarbazone (0.140 g, mmol, 0.234 mmol, 23% yield). The product isomerizes in DMSO. The 1H NMR and 13C NMR for the isomer mixture is reported. 1H

NMR (500 MHz, DMSO-d6): δ 10.02 (0.5H, s), 9.14 (0.5H, s), 8.66 (0.5H, s), 8.60 (0.5H, s), 8.55-8.48 (1H, m), 8.40 (0.5H, s), 8.07 (0.5H, t, J = 1.7 Hz), 8.01 (0.5H, t, J = 1.7 Hz),

7.98 (0.5H, t, J = 1.7Hz), 7.91-7.84 (2.5H, m), 7.83 (0.5H, t, J = 1.7 Hz), 7.70 (0.5H, ddd,

J = 8.01 Hz, 2.0 Hz, 1.0 Hz), 7.67 (0.5H, ddd, J = 7.7 Hz, 1.6 Hz, 1.0 Hz), 7.65 (0.5H, t,

J = 1.7 Hz), 7.61-7.47 (3H, m), 7.42-7.36 (1H, m), 7.33-7.27 (0.5H, t, J = 8.0 Hz). 13C

NMR (125 MHz, DMSO-d6): δ 192.38, 178.49, 144.58, 138.87, 138.30, 138.14, 135.71,

133.08, 132.97, 132.92, 132.58, 131.83, 131.23, 130.71, 128.82, 127.85, 127.74, 122.96,

122.42, 121.94. X-ray crystallographic data obtained for 1,3-bis(3-bromobenzoyl)-5- bromobenzene thiosemicarbazone has been deposited in the Cambridge Crystallographic

Data Centre and was allocated the deposition number CDCC 1042568. HRMS (ESI)

+ + calculated for C21H14Br3N3OSH (M+H) 593.84805, found 593.84797. HPLC retention

time (Method A): 17.56. (Exist as a mixture of E/Z isomers in solution)

73 4.2 Biology

4.2.1 Cathepsin L Inhibition Assay

Enzyme assays for cathepsins L and B were modified from procedures originally described by Barrett and Kirschke.84 (185) Assays to determine the effects of inhibitors on human liver cathepsin L (Sigma Aldrich) activity were carried out using a fluorogenic peptide substrate N-carbobenzyloxy-L-Phe-L-Arg-7-amino-4-methylcoumarin (Z-FR-

AMC) (BACHEM). It should be noted that IC50 values were only determined for

compounds that, at a concentration of 10 µM, demonstrated greater than 50% inhibition

of enzyme activity. Cathepsin L was pre- incubated with inhibitors at various

concentrations (or with vehicle) for 5 min at 25˚C. The assay was initiated by the addition

of substrate Z-FR-AMC. The final assay conditions were 100 mM NaOAc buffer, pH 5.5,

1 mM EDTA, 3 mM DTT, 0.01% Brij 35 (Sigma), 1 nM cathepsin L, 2% DMSO

(Sigma) and 50 µM of Z-FR-AMC, in a total reaction volume of 200 μl. Inhibitors were

diluted to include a final concentration range of 10 µM to 10 pM. The release of AMC

from the substrate was monitored fluorometrically at 15 second intervals for 5 min at 25

oC using black 96 well Corning 3686 assay microplates with a Thermo Fluoroskan

Ascent FL microplate reader at excitation and emission filter wavelengths of 355 nm and

460 nm, respectively. Data were analyzed and IC50 values calculated utilizing GraphPad

Prism 5.0 software. IC50 +S.E (see Supplementary data) values represent the average

results from a minimum of three separate experiments.

4.2.2 Cathepsin B Inhibition Assay

Assays to determine the effects of inhibitors on cathepsin B activity were carried

out using the fluorogenic peptide substrate N-carbobenzyloxy-L-Arg-L-Arg-7-amino-4-

74 methylcoumarin (Z-RR-AMC), with IC50 values determined for compounds that demonstrated greater than 50% inhibition of enzyme activity at 10 µM of inhibitor concentration. Human liver Cathepsin B (EMD Millipore) was pre-incubated with inhibitors at various concentrations for 5 min at 37 oC. The assay was initiated by the addition of substrate Z-RR-AMC (EMD Millipore) and the final assay conditions were 1 nM cathepsin B, 60 µM Z-RR-AMC, 120 mM sodium potassium phosphate buffer pH

6.0, 1 mM EDTA, 3 mM DTT, 0.01% Brij 35, and 2% DMSO. Inhibitors were diluted to include a final concentration range of 10 µM to 10 pM. The reaction was monitored fluorometrically for 5 min at 37 oC using black 96 well Corning 3686 assay microplates with a Thermo Fluoroskan Ascent FL microplate reader at excitation and emission filter wavelengths of 355 nm and 460 nm, respectively. Data were analyzed and IC50 values calculated utilizing GraphPad Prism 5.0 software. Experiments were performed in triplicate.

4.2.3 Progress Curves

Various concentrations of compounds (final concentrations were 10, 5, 1, 0.5, 0.1,

0.01 and 0.005 μM) were mixed with Z-FR-AMC solution (final concentration 10 μM).

Reactions were initiated by the addition of cathepsin L in assay buffer with no pre- incubation time with inhibitor. Readings were taken every 30 sec for 50 min using a fluorescence microplate reader as described above.

4.2.4 Cell Culture

Human umbilical vein endothelial cell culture conditions:

Human umbilical vein endothelial cells (HUVECs) from pooled donors

(Invitrogen) were grown on rat tail collagen-1 coated flasks (CELLCOAT®) in M200

75 medium (Invitrogen) supplemented with 1% gentamycin sulfate (Teknova), 1% amphotericin B (Corning) and low serum growth factor supplement kit (Invitrogen).

HUVECs were passaged using Trypsin EDTA/Trypsin Neutralizer solutions (Invitrogen) as per manufacturer recommendations. Cells were maintained at 37 oC in a humidified atmosphere of 5% CO2. HUVECs were not used beyond passage 5 in these experiments.

PC-3ML cell culture conditions:

PC-3ML are highly metastatic sublines isolated from PC-3 cells.70(172) Cell were cultured in appropriate media supplemented with 10% FBS at 37 oC in a humidified atmosphere of

5% CO2 in air.

MDA-MB-231 cell culture conditions:

MDA-MB-231 cells (ATCC) were cultured in DMEM (Corning) supplemented with 10% fetal bovine serum (Gibco One Shot®) obtained from Invitrogen and 1% gentamycin sulfate, and passaged using trypsin solution (Mediatech) in T-75 culture flasks (Corning).

4.2.5 Cell Invasion Assay

PC-3ML

Invasion assays were performed using Matrigel coated invasion inserts from BD

Biosciences (354480). Cells suspended in serum free media were seeded into the inserts.

Media supplemented with 10% FBS was added to the bottom chamber. The cells were incubated under desired conditions and 24h later, cells that invaded to the underside of the membrane were stained and counted under a microscope.

76 MDA-MB-231

The effects of compounds 1, 8, and 32 on the invasive ability of MDA-MB-231 cells were analyzed using a cell invasion assay with BD Bioscience Matrigel™ invasion kits. MDA-MB-231 cells were cultured and passaged as previously described. Cells were trypsinized and removed from the culture flasks when they were 80% confluent. Cell density was determined with a Beckman Coulter Z-Coulter cell counter. Stock solutions of compounds 1, 8, and 32 in DMSO were prepared in addition to E-64, a known generic cysteine protease inhibitor which was used as a positive control. BD BioCoat Matrigel® invasion chambers, which contained 8 micron pore size PET membranes with a layer of

Matrigel, were used for the experiment. The experiment was initiated by adding DMEM supplemented with 10% FBS (which functions as a chemoattractant) and gentamicin to a

24-well microplate (i.e. lower chamber). Then, the inserts were carefully placed on top of the wells containing the chemoattractant. Cell and compound solutions were added such that the final conditions per well were: 2% DMSO, 50,000 cells in DMEM and 25 or 10

μM of the compounds. Final conditions for untreated (controls) cells were: 2% DMSO and 50,000 cells. The 24-well plates containing the invasion chambers were placed in an

o incubator with a 5% CO2 environment for 24 hours at 37 C. The experiments were terminated, invaded cells were fixed with methanol, stained using a Diff staining kit

(IMEB Inc), and rinsed with deionized water. Samples were air-dried and membranes were removed and placed on glass slides. Each sample was observed with a Zeiss

Axiovert 40 CFL inverted microscope to perform manual cell counting (ten fields were observed under a 40x objective). Experiments were performed in triplicate.

77 4.2.6 Cell Migration Assay

Inhibition of MDA-MB-231 cell migration was determined with an assay similar

to that described for the cell invasion assay. However, the membrane inserts had 8µm

pores but no Matrigel layer. Experiments were performed in triplicate.

4.2.7 Growth Inhibition of C3H Mammary Carcinoma

Experiments were performed using 10-14-week-old female CDF1 mice, in which

a C3H mammary carcinoma was implanted in the right rear foot. This tumor model is an

anaplastic adenocarcinoma that arose spontaneously in a C3H mouse at Aarhus

University Hospital and was originally designated as HB;85 (186) the name was changed to

C3H mammary carcinoma when it was grown in the more stable CDF1 mouse variant.86

(187) C3H mammary carcinomas do not grow in culture, thus experimental tumors were

produced following sterile dissection of large flank tumors as previously described.87 (188)

Basically, macroscopically viable tumor tissue was minced with scissors and 5-10 µl of this material implanted into the foot. Treatments were started at the time of tumor implantation; tumor volume was determined by the formula D1 x D2 x D3 x π/6 (where the D values represent three orthogonal diameters). All animal studies were conducted according to the animal welfare policy of Aarhus University (http://dyrefaciliteter.au.dk), with the Danish Animal Experiments Inspectorate’s approval. 3-Benzoylbenzophenone thiosemicarbazone (1) was supplied by Baylor University (Waco, Texas, USA). It was freshly prepared before each experiment by dissolving in Tween80 and then diluted to

10% using saline before injection. Stock solutions were kept cold and protected from light. 3-Benzoylbenzophenone thiosemicarbazone (1) was injected intraperitoneally (i.p.)

78 in a volume of 0.02 ml/g mouse body weight for 5 consecutive days starting either on the

day of tumor implantation or when tumors had reached 200 mm3.

4.2.8 Cytotoxicity Assay

The sulforhodamine B (SRB) assay was used to assess inhibition of human cell

line growth as previously described.88-90 (189–191) Briefly, HUVECs were passaged using

normal conditions and plated at 9,000 cells/well in 96-well plates (Corning) and

incubated for 24 h. Ten-fold serial dilutions of the compounds to be tested were then

added to the wells. After 48 h, treated and control cells were fixed with 10%

trichloroacetic acid, stained with 0.4% sulforhodamine B (Acid Red 52) (TKI) for 30 minutes, and subsequently washed 4 times with 1% acetic acid in water. The plates were air dried, and the SRB dye was solubilized with 200 µL of 10 mM Tris base and read at

540 nm with an automated Biotek Elx800 plate reader (Biotek). Values were normalized

91 (192) to 630 nm to account for background absorbance. A growth inhibition of 50% (GI50

or the drug concentration causing a 50% reduction in net protein absorbance relative to

controls) was calculated from a minimum of two replicates and averaged for a minimum

of three experiments with Excel software.

Acknowledgments

Financial support of this research was generously provided by OXiGENE, Inc.

(grant to KGP and MLT) and the NIH-United States (grant R01 CA 169300 to DWS).

The authors express their appreciation to Dr. Kevin K. Klausmeyer and Marissa Penney

for X-ray crystallography studies, Dr. Nicole Kruse (Applications Scientist at Bruker

BioSpin) for acquiring data on the 400 MHz NMR, Dr. Alejandro Ramirez (Mass

Spectrometry Core Facility, Baylor University) for assistance with mass spectrometry

79 analysis, Dr. Michelle Nemec (Director of the Molecular Biosciences Center at Baylor

University) for use of shared facilities, and Mr. Kahler Low for technical assistance. For

the in vivo experiments, the authors thank Ms. Dorthe Grand and Ms.Inger Marie

Horsman for technical assistance, and the Danish Cancer Society and the Danish Council

for Independent Research: Medical Sciences, for financial support.

Supplementary data

Supplementary data including X-ray crystallography, 1H NMR, 19F NMR, 13C

NMR, HRMS, HPLC analysis, molecular modeling studies, enzyme kinetic studies, and

Lipinski rule of five analysis associated with this article can be found, in the online version, at http://www.sciencedirect.com/science/article/pii/S0968089615300614.

References

1. Fonović, M.; Turk, B. Biochim. Biophys. Acta 2014, 1840, 2560-2570.

2. Mohamed, M. M.; Sloane, B.F. Nature Reviews Cancer 2006, 6, 764-775.

3. Gocheva, V.; Joyce, J. A. Cell Cycle 2007, 6, 60-64.

4. Lankelma, J. M.; Vorrend, D. M.; Barwari, T.; Koetsveld, J.; Van der Spek, A. H.; De Porto, A. P.; Van Rooijen, G.;Van Noorden, C. J. Life Sci. 2010, 86, 225- 233.

5. Brömme, D.; Wilson, S. Role of Cysteine Cathepsins in Extracellular Proteolysis. In Extracellular Matrix Degradation; Parks, W. C.; Mecham, R. P., Eds.; Biology of Extracellular Matrix 2; Springer Berlin Heidelberg, 2011; pp 23-51.

6. Gocheva, V.; Zeng, W.; Ke, D.; Klimstra, D.; Reinheckel, T.; Peters, C.; Hanahan, D.; Joyce, J. A. Genes Dev. 2006, 543-546.

7. Blondeau, X.; Vidmar, S. L.; Emod, I.; Pagano, M.; Turk, V.; Keil-Dloua, V. Biol. Chem. Hoppe-Seyler 1993, 374, 651-656.

8. Ishidoh, K.; Kominami, E. Biochem. Biophys. Res. Commun. 1995, 217, 624-631.

80 9. Guinec, N.; Dalet-Fumeron, V.; Pagano, M. Biol. Chem. Hoppe-Seyler 1993, 374, 1135-1146.

10. Buck, M. R.; Karustis, D. G.; Day, N. A.; Honn, K.V.; Sloane, B. F. Biochem J. 1992, 282, 273-278.

11. Kirschke, H.; Kembhavi, A. A.; Bohley, P.; Barret, A. J. Biochem. J. 1982, 201, 367-372.

12. Maciewicz, R. A.; Wotton, S. F.; Etherington, D. J.; Duance, V. C. FEBS Lett. 1990, 269, 189-193.

13. Garnero, P.; Borel, O.; Byrjalsen, I.; Ferreras, M.; Drake, F. H.; McQueney, M. S.; Foged, N. T.; Delmas, P. D.; Delaissé, J.-M. J. Biol. Chem. 1998, 273, 32347- 32352.

14. Sameni, M.; Dosescu, J.; Moin, K.; Sloane, B. F. Mol. Imaging 2003, 2, 159-175.

15. Rothberg, J. M.; Bailey, K. M.; Wojtkowiak, J. W.; Ben-Nun, Y.; Bogyo, M.; Weber, E.; Moin, K.; Blum, G.; Mattingly, R. R.; Gillies, R. J.; Sloane, B. F. Neoplasia 2013, 15, 1125-1137.

16. Manson, S. D.; Joyce, J. A. Trends in Cell Biol. 2011, 21, 228-237.

17. Bell-McGuinn, K. M.; Garfall, A. L.; Bogyo, M.; Hanahan, D.; Joyce, J. A. Can. Res. 2007, 65, 7378-7385.

18. Berdowska, I. Clin. Chim. Acta 2004, 342, 41-69.

19. Ullah, M. F.; Aatif, M. Cancer Treat. Rev. 2009, 35, 193-200.

20. Chen, B.; Platt, M. O. J. Transl. Med. 2011, 9, 109-120.

21. Jensen, A. B.; Wynne, C.; Ramirez, G.; He, W.; Song, Y.; Berd, Y.; Wang, H.; Mehta, A.; Lombardi, A. Clin. Breast Cancer 2010, 10, 452-458.

22. Clezardin, P. Breast Cancer Res. 2011, 13, 207.

23. Stoch, S. A.; Zajic, S.; Stone, J. A.; Miller, D. L.; van Bortel, L.; Lasseter, K. C.; Pramanik, B.; Cilissen, C. ; Liu, Q.; Liu, L.; Boyd, B. S.; Paneblanco, D.; Ding, Yu, Gottesdiener, K.; Wagner, J. A. Br. J. Clin. Pharmacol. 2012, 75, 1240-1254.

24. Gauthier, J. Y.; Chauret, N.; Cromlish, W.; Desmarais, S.; Duong, L. T.; Falgueyret, J.-P.; Kimmel, D. B.; Lamontagne, S.; Léger, S.; LeRiche, T.; Li, C. S.; Massé, F.; McKay, D. J.; Nicoll-Griffith, D. A.; Oballa, R. M.; Palmer, J. T.; Percival, D.; Riendeau, D.; Robichaud, J.; Rodan, G. A.; Rodan, S. B.; Seto, C.;

81 Thérien, M.; Truong, V.-L.; Venuti, M. C.; Wesolowski, G.; Young, R. N.; Zamboni, R.; Black, W. C. Bioorg. Med. Chem. Lett. 2008, 18, 923-928.

25. Lindstrom, E.; Grabowska, U.; Jerling, M.; Edenius, C.; MIV-711, A Highly Selective Cathepsin K Inhibitor, Reduces Biomarkers of Bone Resorption and Cartilage Degradation in Healthy Subjects. Proceedings of the 2014 OASRI World Congress, Paris, France, April 24-27, 2014. Osteoarthritis and Cartilage, 2014, S197.

26. About Cathepsin B and Liver Disease. http://www.virobayinc.com/liver- fibrosis.php (Accessed on March 20, 2014).

27. Evaluation of Safety and Pharmacokinetics of Single Doses of VBY-376 in Healthy Adults. http://clinicaltrials.gov/ct2/show/NCT00557583 (Accessed on October 1, 2014).

28. Pan cathepsin inhibitor: VBY-825. http://www.virobayinc.com/VBY-825.php (Accessed on October 1, 2014)

29. Elie, B. T.; Gocheva, V.; Shree, T.; Darlrymple, S. A.; Holsinger, L. J.; Joyce, J. A. Biochimie 2010, 92, 1618-1624.

30. GSK Product development pipeline February 2014. http://www.gsk.com/media/280387/product-pipeline-2014.pdf (Accessed on September, 8, 2014)

31. A Study to Evaluate the Safety, Tolerability, Pharmacokinetics (PK), Pharmacodynamics (PD) and Food Effect of Single or Repeat Doses of GSK2793660 in Healthy Subjects. https://clinicaltrials.gov/ct2/show/NCT02058407 (Accessed on September, 8, 2014).

32. Rizoska, B; Edenius, C.; Göhlin, K.; Hewitt, E.; Lindström, E.; Malcangio, M.; Pitcher, T.; Tunblad, K. MIV-247 A Cathepsin Protease Inhibitor Developed as a Novel Therapy for Neuropathic Pain. Proceedings of the 15th World Congress on Pain 2014, Buenos Aires, Argentina, October 6-11, 2014.

33. Miller, B.; Friedman, A. J.; Choi, H.; Hogan, J.; McCammon, J. A.; Hook, V.; Gerwick, W. H. J. Nat. Prod. 2014, 77, 92-99.

34. Katunuma, N.; Murata, E.; Kakegawa, H.; Matsui, A.; Tsuzuki, H.; Tsuge, H.; Turk, D.; Turk, V.; Fukushima, M.; Yada, Y.; Asao, T. FEBS Lett. 1999, 458, 6- 10.

82 35. Shah, P. P.; Wang, T.; Kaletsky, R. L.; Myers, M. C.; Purvis, J. E.; Jing, H.; Huryn, D. M.; Greenbaum, D. C.; Smith III, A. B.; Bates, P.; Diamond, S. L. Mol. Pharmacol. 2010, 78, 319-324.

36. Dana, D.; De, S.; Rathod, P.; Davalos, A. R.; Novoa, D. A.; Paroly, S.; Torres, V. M. Afzal, N.; Lankalapalli, R. S.; Rotenberg, S. A.; Chang, E. J. Subramaniam, G.; Kumar, S. Chem. Commun. 2014, 50, 10875-10878.

37. Assad, N.; Bethel, P. A.; Coulson, M. D.; Dawson, J. E.; Ford, S. J.; Gerhardt, S.; Grist, M.; Hamlin, G. A.; James, M. J.; Jones, E. V.; Karoutchi, G. I.; Kenny, P. W.; Morley, A. D.; Oldham, K.; Rankine, N.; Ryan, D.; Wells, S. L.; Wood, L.; Augustin, M.; Krapp, S.; Simader, H.; Steinbacher, S. Bioorg. Med. Chem. Lett. 2009, 19, 4280-4283.

38. Marquis, R. W.; James, I.; Zeng, J.; Trout, R. E. L.; Thompson, S.; Rahman, A.; Yamashita, D. S.; Xie, R.; Ru, Y.; Gress, C. J.; Blake, S.; Lark, M. A.; Hwang, S.- M.; Tomaszek, T.; Offen, P.; Head, M. S.; Cummings, M. D.; Veber, D. F. J. Med. Chem. 2005, 48, 6870-6878.

39. Falgueyret, J.-P.; Oballa, R. M.; Okamoto, O.; Wesolowski, G.; Aubin, Y.; Rydzewski, R. M.; Prasit, P.; Riendeau, D.; Rodan, S. B.; Percival M. D. J. Med. Chem. 2001, 44, 94-104.

40. Frizler, M.; Lohr, F.; Furtmann, N.; Kläs, J.; Gütschow, M. J. Med. Chem. 2011, 54, 396-400.

41. Bauer, D. J. Clinical experience with the antiviral drug Marboran® (1-methylisatin 3-thiosemicarbazone) Ann. N. Y. Acad. Sci. 1965, 130, 110–117.

42. Bauer D. J.; St Vincent, L.; Kempe, C.; Young, P. A.; Downie, A.W. Am. J. Epidemiol.I 1969, 90, 130-145.

43. Weiss, M. M.; Weiss, P. D.; Mathisen, G.; Guze, P. Clin. Infect. Dis. 2004, 39, 1668-1673.

44. Caminero, J. A.; Sotgiu, G.; Zumla, A.; Migliori, G. B. Lancet Infect. Dis. 2010, 10, 621-629.

45. Kishore Kumar, G. D.; Chavarria, G. E.; Charlton-Sevcik, A. K.; Arispe, W. M.; MacDonough, M. T.; Strecker, T. E.; Chen, S.-E.; Siim, B. G.; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. Bioorg. Med. Chem. Lett. 2010, 20, 1415−1419.

46. Kishore Kumar, G. D.; Chavarria, G. E.; Charlton-Sevcik, A. K.; Yoo, G. K.; Song, J.; Strecker, T. E.; Siim, B. G.; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. Bioorg. Med. Chem. Lett. 2010, 20, 6610−6615.

83 47. Chavarria, G. E.; Horsman, M. R.; Arispe, W. M.; Kishore Kumar, G. D; Chen, S.-E.; Strecker, T. E.; Parker, E. N.; Chaplin, D. J.; Pinney, K.G.; Trawick, M.L. Eur. J. Med. Chem. 2012, 58, 568-572.

48. Sudhan, D. R.; Siemann, D. W. Clin. Exp. Metastasis 2013, 30, 891-902.

49. Song, J.; Jones, L. M.; Kishore Kumar, G. D.; Conner, Liela Bayeh; E. S.; Chavarria, G. E.; Charlton-Sevcik, A. K.; Shen-En Chen; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. ACS Med. Chem. Lett. 2012, 3,450-453.

50. Song, J.; Jones, L. M.; Chavarria, G. E.; Charlton-Sevcik, A.K.; Jantz, A.; Johansen, A.; Bayeh, L.; Soeung, V.; Snyder, L.K.; Lade, S.D.; Chaplin, J.; Trawick, M.L.; Pinney, K.G. Bioorganic and Medicinal Chemistry Letters 2013, 23, 2801-2807.

51. Chaplin, D. J.; Gaddale Devanna, K.K; Parker, E.; Pinney, K. G.; Jiangli, S.; Trawick, M. L. PCT Int. Appl., WO 2013142038 A2 2013092, September 26, 2013.

52. Raghav, N.; Kaur, R. Med. Chem. Res. 2014, 23, 4669-4679.

53. Siles, R.; Chen. S.-E.; Zhou, M.; Pinney, K. G.; Trawick, M. L. Bioorg. Med. Chem. Lett. 2006, 16, 4405-4409.

54. Du, X.; Guo, C.; Hansell, E.; Doyle, P. S.; Caffey, C. R.; Holler, T. P.; McKerrow, J. H.; J. Med. Chem. 2002, 45, 2695-2707.

55. Mallari, J. P.; Shelat, A.; Kosinski, A.; Caffrey, C. R.; Connelly, M.; Zhu, F.; McKerrow, J. H.; Guy, R. K. Bioorg. Med. Chem Lett. 2008, 18, 2883-2885.

56. Caputto, M. E.; Fabian, L. E.; Benítez, D.; Merlino, A.; Ríos, N.; Cerecetto, H.; Moltrasio, G. Y.; Moglioni, A. G.; González, M.; Finkielsztein, L. M. Bioorg. Med. Chem. 2011, 19, 6818-6826.

57. Greenbaum, D. C.; Mackey, Z.; Hansell, E.; Doyle, P.; Gut, J.; Caffrey, C. R.; Lehrman, J.; Rosenthal, P. J.; McKerrow, J. H.; Chibale, K. J. Med. Chem. 2004, 47, 3212-3219.

58. Chiyanzu, I.; Clarkson, C.; Smith, P. J.; Lehman, J.; Gut, J.; Rosenthal, P. J.; Chibale, K. Bioorg. Med. Chem. 2005, 13, 3249-3261.

59. Liu, Y.; Cui, K.; Lu, W.; Luo, W.; Wang, J.; Huang, J.; Guo, C. Molecules 2011, 16, 4527-4538.

60. Schröder, J.; Noack, S.; Marhöfer, R. J.; Mottram. J. C.; Coombs, G. H.; Selzer, P. M. PLoS ONE [Online] 2013, 8, e77460.

84 61. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0077460 (Accessed on September, 8, 2014).

62. Schaeffer, M.; Schroeder, J.; Heckeroth, A. R.; Noack, S.; Gassel, M.; Mottram, J. C.; Selzer, P. M.; Coombs, G. H. Antimicrob. Agents Chemother. 2012, 56, 1190- 1201.

63. Johnson, J. E.; Morales, N. M.; Gorczyca, A. M.; Dolliver, D. D.; McAllister, M. A. J. Org. Chem. 2001, 66, 7979-7985.

64. Karabatsos, G. J.; Vane, F. M.; Taller, R. A.; Hsi, N. J. Am. Chem. Soc. 1964, 86, 3351-3357.

65. Kerek, F.; Ostrogovich, G.; Simon, Z. J. Chem. Soc. B 1971, 541-544.

66. Ostrogovich, G.; Simon, Z.; Kerek, F. Revue Tomaine de Chimie, 1970, 15, 1453- 1459.

67. Curtin, D. Y.; Grubbs, E. J.; McCarty, C. G. J. Am. Chem. Soc. 1966, 88, 2775- 2786.

68. Hardegger, L. A.; Kuhn, B.; Spinnler, B.; Anselm, L.; Ecabert, R.; Stihle, M.; Gsell, B.; Thoma, R.; Diez, J.; Benz, J.; Plancher, J.-M.; Hartmann, G.; Banner, D. W.; Happ, W.; Diderich, F. Angew. Chem. Int. Ed. 2011, 50, 314-318.

69. Choe, Y.; Leonetti, F.; Greenbaum, D. C.; Lecaille, F.; Bogyo, M.; Bromme, D.; Ellman, J. A.; Craik, C. S. J. Biol. Chem. 2006, 281, 12824–12832.

70. Albini, A.; Iwamoto, Y.; Kleinman, H. K; Martin, G. R.; Aaronson, S. A.; Kozlowski, J. M.; McEwan, R. N. Cancer Res. 1987, 47, 3239-3245.

71. Wang, M.; Stearns, M. E. Differentiation 1991, 48, 115-125.

72. Neve, R. M,; Chin, K.; Fridlyand, J.; Yeh, J.; Baehner, F. L.; Fevr, T.; Clark, L.; Bayani, N.; Coppe, J.; Tong, F.; Speed, T.; Spellman, P. T.; DeVries, S.; Lapuk, A.; Wang, N.J.; Kuo, W.-L.; Stilwell, J. L.; Pinkel, D.; Albertson, D. G.; Waldman, F. M.; McCormick, F.; Dickson, R. B.; Johnson, M. D.; Lippman, M.; Ethier, S.; Gazdar, A.; Gray, J. W.. Cancer Cell 2006, 10, 515-527.

73. Zajc, I.; Sever, N.; Bervar, A.; Lah, T. T. Cancer. Lett. 2002, 187, 185-190.

74. Tamai, M.; Yokoo, C.; Murata, M.; Oguma, K.; Sota, K.; Sato, E.; Kanaoka, Y. Chem. Pharm. Bull. 1987, 35, 1098-1104.

85 75. Mitchell, D.; Lukeman, M.; Lehnherr, D.; Wan, P. Formal intramolecular photoredox chemistry of meta-substituted benzophenones. Org. Lett. 2005, 7, 3387-3389.

76. Zhao, W.; Carreira, E. M. Org. Lett. 2006, 8, 99-102.

77. Ye S.; Chen, J.; Di, C.; Liu, Y.; Lu, K.; Wu, W.; Du, C.; Liu, Y.; Shuai, Z.; Yu, G. J. Mater. Chem. 2010, 20, 3186-3194.

78. Sharghi, H.; Jokar, M.; Doroodmand, M. M.; Khalifeh, R. Adv. Synth. Catal. 2010, 352, 3031-3044.

79. Blick, F. F.; Patelski, R. A. J. Am. Chem. Soc. 1938, 60, 2283-2285.

80. Miller, T. M.; Kwock, E. W.; Neenan, T. X. Macromolecules 1992, 25, 3143- 3148.

81. Teki, Y.; Takui, T.; Itoh, K.; Iwamura, H.; Kobayashi, K. J. Am. Chem. Soc. 1986, 108, 2147-2156.

82. Ronald, S.; Michael, S. PCT Int. Appl. WO 9907336 A1 19990218, February 18, 1999.

83. Knudsen, S. M.; Pettersson, I.; Lau, J.; Behrens, C.; Petersen, A. K.; Garibay, P. W. PCT Int. Appl. WO 2006003096 A1 20060112, January 12, 2006.

84. Chang, L. L. PCT Int. Appl. WO 9822108 A1 19980528, May 28, 1998.

85. Barrett, A. J.; Kirschke, H. Method. Enzymol. 1981, 80, Pt C, 535-561.

86. Overgaard, K.; Overgaard, J. Eur. J. Cancer 1972, 8, 65-78.

87. Overgaard, J. Int. J. Radiat. Oncol. Biol. Phys. 1980, 6, 1507-1515.

88. Horsman, M.R.; Murata, R.; Breidahl, T.; Nielsen, F.U.; Maxwell, R.J.; Stødkilde- Jørgensen, H.; Overgaard, J. Adv. Exp. Med. Biol. 2000, 476, 311-323.

89. Monks, A.; Scudiero, D.; Skehan, P.; Shoemaker, R.; Paull, K.; Vistica, D.; Hose, C.; Langley, J.; Cronise, P.; Vaigro-Wolff, A.; Gray-Goodrich, M.; Campbell, H.; Mayo, J.; Boyd, M. J. Natl. Cancer Inst. 1991, 83, 757-766.

90. Vichai, V.; Kirtikara, K. Nat. Protoc. 2006, 1, 1112-1116.

91. MacDonough, M. T.; Strecker, T. E.; Hamel, E.; Hall, J. J.; Chaplin, D. J.; Trawick, M. L.; Pinney, K. G. Bioorg. Med. Chem. 2013, 21, 6831-6843.

86 92. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1107-1112.

CHAPTER THREE Synthesis and Biological Evaluation of a Water Soluble Phosphate Prodrug Salt and Structural Analogues of KGP94, a Lead Inhibitor of Cathepsin L

Erica N. Parker †, Samuel O. Odutola‡, Yifan Wang†, Gustavo E. Chavarria†, Tracy E. Strecker†, David J. Chaplin§, Mary Lynn Trawick†, and Kevin G. Pinney*,‡,† †Department of Chemistry and Biochemistry Baylor University One Bear Place #97348 Waco, TX 76798-7348 ‡Institute of Biomedical Studies Baylor University One Bear Place #97224 Waco, TX 76798-7224 §OXiGENE, Inc. 701 Gateway Blvd, Suite 210 South San Francisco, CA 94080

Note: The present chapter is formulated as a manuscript draft in preparation for

submission to a medicinal chemistry journal.

Abstract

The degree of expression of cathepsin L, often unregulated in the tumor

microenvironment, reciprocally parallels with the invasive and metastatic nature of

certain cancer cell lines. Inhibition of cathepsin L represents and emerging strategy for

the treatment of metastatic cancer. KGP94 ([(3-bromophenyl)-(3-hydroxyphenyl)-ketone]

thiosemicarbazone), a potent inhibitor of cathepsin L, was synthesized by an improved

route in order to eliminate the replacement of bromine by hydrogen as observed in the

previous synthesis. Additional analogues containing the KGP94 motif were synthesized

and evaluated for inhibitory activity against cathepsin L. Due to limited aqueous

solubility of KGP94, a water-soluble phosphate salt (KGP420) was prepared. Enzymatic

hydrolysis assays with alkaline phosphatase demonstrated that the phosphate prodrug,

KGP420, was readily converted to the parent compound, KGP94.

88 Introduction

The discovery and development of effective therapeutic regimens that target cancer metastasis remains an urgent and largely unmet need. Metastasis is a significant contributing factor in over ninety percent of deaths attributed to cancer.1 The sequence of steps involved in the metastatic process associated with tumor cells includes invasion of these cells from the primary tumor into the surrounding tissue, intravasation into the circulatory system, extravasation from the circulatory system, and establishment of a secondary tumor.1,193,194 One promising strategy towards the development of anti- metastatic agents involves targeting one or more members of the Papain family of cysteine cathepsin proteases (comprised of 11 members: B, C, F, H, K, L, O, S, V, W, and X/Z) with small-molecule inhibitors.195,53,54,8,148 Cathepsins aid in the invasion and migration of tumor cells through the degradation of proteins comprising the extracellular matrix including collagen,70,75–80 fibronectin, 73–76 and laminin74–76. Elevated levels of cathepsins L, B, H, X, and S have been detected in several cancer types including breast, prostate, brain, colorectal, and lung cancers. Moreover, increased expression of these cathepsins correlates inversely to overall prognosis in breast, ovarian, colorectal, brain, lung, and head and neck cancers.152,153 Decreased tumor volume and increased survival rates were observed in a mouse model with inclusion of a pan-cathepsin inhibitor with cyclophosphamide, a known chemotherapeutic, as a cancer treatment regimen.196

In addition to the role that these enzymes play in metastasis and migration of cancer cells through degradation of components of the extracellular matrix, cysteine cathepsin proteases have been implicated as targets for osteoporosis, rheumatoid arthritis, atherosclerosis, and diseases of the immune system.195 Cathepsin inhibitors as drug

89 candidates for the treatment of various diseases in the pharmaceutical pipeline include

VBY-825 (Virobay), 96,97 Odanacatib (Merck), 84,85 LY3000328 (Eli Lily), 98,99 (Figure

3.1).

Figure 3.1. Cathepsin inhibitors in the pharmaceutical pipeline.

VBY-825 (Virobay), a pan cysteine protease inhibitor targeting the treatment of

liver fibrosis, incorporates a diketone warhead into a peptide-like backbone allowing for

reversible inhibition of targeted cathepsins. Odanacatib (Merck), a nitrile based inhibitor, targets cathepsin K for suppression of bone resorption in osteoporosis. LY3000328 (Eli

Lily), a noncovalent cathepsin S inhibitor with an IC50 value in the low nM range, targets

the treatment of abdominal aortic aneurysm. Although significant progress has been

achieved towards the development of cathepsin inhibitors as therapeutic treatment

options, no FDA approved drugs currently exist in this pharmaceutical field.

Additionally, small molecular entities that specifically target the inhibition of cathepsin L

for the treatment of pathological processes in clinical trials remains an unmapped frontier

open to exploration.

Figure 3.2 Representative sampling of potent covalent inhibitors of cathepsin L

Small-molecule inhibitors of cathepsin L have been synthesized that incorporate a variety of electrophilic moieties (warheads) capable of interacting with the catalytic site residue Cys-25 (Figure 3.2). Warheads which undergo covalent bonding with the Cys25 thiolate of cathepsin L include the epoxide in Clik 148 (I),158 the carbonyl of thiocarbazate II,123 the nitrile in the purine analogue III,197 the cyclic carbonyl in azepanone IV,120 the nitrile in the triazine analogue V,124 the α,β-unsaturated amide of gallinamide A (VI), 113,114 and the aldehyde of the N-(1-naphthalenylsulfonyl) peptide derivative VII109 (Figure 3.2). Cruzain, is a cysteine protease found in Trypanosoma cruzi, and is targeted for the treatment of Chagas’ disease. Initial studies towards the development of cruzain135,136 inhibitors led to the discovery of cathepsin L inhibitors bearing the thiosemicarbazone moiety which contains an electrophilic thiocarbonyl warhead. Building on these results, we embarked on a structure-activity relationship

(SAR) guided program designed to incorporate the thiosemicarbazone moiety within

91 appropriately functionalized benzophenone, pyridine, thiophene, fluorene,

thiochromanone, benzothiepine, and dihydroquinoline molecular scaffolds. A focused

small library of cathepsin inhibitors resulted from these studies, from which a sub-set of

69,126– molecules (greater than 35) demonstrated IC50 values below 500 nM (Figure 3.3).

133

Figure 3.3. Sub-set of previously described thiosemicarbazone based inhibitors of cathepsin L

A lead compound (referred to as KGP94), which is a slowly reversible, time- dependent inhibitor of cathepsin L, emerged from this small library of structurally diverse

thiosemicarbazone analogues.128,129,69 Low cytotoxicity against human umbilical vein

endothelial cells (HUVECs) and the ability to inhibit the invasive and migratory potential

of both PC-3ML (prostate cancer cell line) and MDA-MB-231 (breast cancer cell line)

has led to the identification of KGP94 as a pre-clinical candidate for potential

development as an anti-metastatic agent, functioning through a potent inhibition of

cathepsin L.69,129 Since KGP94 demonstrates limited solubility in water, it proved

92 desirable to prepare a water-soluble prodrug salt to further the pre-clinical development

of this promising agent. Fortunately, KGP94 bears a phenolic hydroxyl group which

provides a convenient molecular handle for the introduction of a bioreversible promoiety in order to improve aqueous solubility and potentially enhance ADME (adsorption, distribution, metabolism, and excretion) pharmacokinetic properties. Installation of a phosphate prodrug salt for the purpose of increasing aqueous solubility of FDA approved drugs intended for oral or parental administration has been successfully demonstrated by fosfpropofol, fosamprenavir, and fosfluconazole.198

Results and Discussion

Design and Synthetic Chemistry

In addition to the synthesis of a water-soluble phosphate prodrug of KGP94, several analogues of this parent, lead compound were also prepared. Previous structure- activity relationship (SAR) studies related to thiosemicarbazone inhibitors based on the benzophenone molecular scaffold highlighted the importance of the 3-bromophenyl moiety. 127,128The extended series of benzophenone-based thiosemicarbazone inhibitors

incorporate both m-hydroxy and m-bromo substituents. Our initial synthetic route128 to

KGP94 (Scheme 3.1) utilized the addition of 3-bromophenylmagnesium bromide to the corresponding Weinreb amide to afford (3-bromophenyl)-(3-hydroxyphenyl)-methanone, which was condensed with thiosemicarbazide followed by deprotection of the silyl ether.

However, HPLC analysis of the final product revealed that a trace amount of the bromine atom had been replaced by a hydrogen.128 Replacement of bromine by hydrogen likely occurred during the halogen-metal exchange reaction since excess magnesium was used to prepare benzophenone (3-bromophenyl)-(3-hydroxyphenyl)-methanone.

Scheme 3.1. Previously reported synthetic route towards KGP94128

In an effort to avoid these trace impurities, KGP94 and analogues were synthesized utilizing a revised route (Scheme 3.2). Instead of using 1,3- dibromobenzene as the precursor to the organometallic reagent for the synthesis of KGP94 (11), a protected m-bromophenol 1 was reacted with n-butyllithium to form the intermediate organolithium reagent, which was reacted with Weinreb amide 4 to afford the desired functionalized benzophenone 7. Benzophenone intermediates 8 and 10 were synthesized in a similar manner by reacting the appropriately substituted aromatic ring with n- butyllithium followed by the addition of Weinreb amide 5 to afford ketone 8 or the addition of aldehyde 6 followed by oxidation with PCC to afford ketone 10.

Condensation of benzophenones intermediates 7, 8, and 10 (separately) with thiosemicarbazide followed by desilylation with TBAF afforded target thiosemicarbazone analogues KGP94 (11), 12, and 13. HPLC analysis indicated no trace amount of the impurity in which bromine was replaced by hydrogen in the final product for KGP94

(11).

94 Scheme 3.2. Improved synthetic route towards KGP94 (11) and analogues 12-13.

Synthesis of dimethylresorcinol and resorcinol analogues 19-22 utilized

commercially available 1-bromo-3,5-dimethoxy benzene as a starting material to form an

intermediate organolithium reagent which upon reaction with Weinreb 4 or 14afforded

ketones 15 and 16, respectively (Scheme 3.3). Demethylation of these 3,5-

dimethoxybenzophenone intermediates 15 and 16 with boron tribromide afforded 3,5- dihydroxybenzophenones 17 and 18. Subsequent condensation of ketones 15-18 with

thiosemicarbazide (separately) under microwave irradiation generated target

thiosemicarbazone analogues 19-22.

Scheme 3.3. Synthesis of dimethylresorcinol and resorcinol analogues 19-22.

In order to increase the aqueous solubility and possibly the bioavailability of

KGP94 (11), a water-soluble phosphate prodrug salt was prepared through phosphorylation of the phenolic moiety. Thus, 3-bromophenyl-3-hydroxyphenyl methanone 23 was phosphorylated with dibenzyl chlorophosphate (prepared in situ)199 to afford the corresponding dibenzyl phosphate ester 24 (Scheme 3.4). Subsequent deprotection of the benzyl groups with 33% HBr in AcOH generated phosphoric acid ester 25. Successful completion of the synthesis of the phosphate salt of benzophenone

thiosemicarbazone KGP420 (27) was accomplished by the condensation of phosphoric

acid ester 25 with thiosemicarbazide to afford benzophenone thiosemicarbazone 26, which upon reaction with sodium carbonate generated the desired disodium phosphate salt KGP420 (27).

Scheme 3.4. Prodrug derivatization of KGP94 (11) to form water-soluble salt KGP420.

Isomerization

Figure 3.4. ROESY and COSY correlations for analogue 13 as a representative example of the major isomer observed in DMSO-d6.

One challenge inherent to the synthesis of thiosemicarbazone based inhibitors is the propensity for isomerization about the imine bond.166,200,201 As a notable example, 2-

formylpyridine-4’,4’-dimethyl thiosemicarbazone, isolated in the Z configuration,

isomerized in solution generating varying Z/E equilibrium isomeric ratios which were

97 dependent upon on the capability of the particular solvent to disrupt the intramolecular

hydrogen bonding interaction between the pyridine nitrogen and N-H hydrogen of the

thiosemicarbazone.202 Benzophenone thiosemicarbazone analogues reported herein, except KGP420 (27, 60:40 ratio in D2O), exist predominately in the E configuration in

solution (Figure 3.4). Isomerization of KGP94 (11) and benzophenone

thiosemicarbazone analogues in DMSO-d6 occurred over a period of time (Table 3.1).

After standing for one week in DMSO-d6 at room temperature, KGP94 isomerized by the

greatest extent to an equilibrium ratio of 25%. Analogue 22 isomerized the least with

only 9% of the minor isomer present after standing in DMSO-d6 for one week.

Table 3.1. Isomerization of benzophenone thiosemicarbazone analogues in DMSO-d6

Percent Z isomer present a

Compound 0 hours 48 hours 1 week

11 5% 25%b

12 NAc NA NA

13 1.5% 17% 18%

19 4% 12% 20%

20 2% 10% 14%

21 11% 13% 13%

22 0.2% 8% 9%

a Isomerization of KGP94 and analogues were monitored by H NMR in DMSO-d6 as solvent. bReported percent of Z isomer present observed at 24 h cNot applicable

98 Biological Evaluation

Table 3.2. Inhibitory activity of benzoylbenzophenone thiosemicarbazone analogues

a IC50 Values (nM) Percent Inhibition (10 µM) Cmpd R1 R2 R3 R4 CatL CatB CatL

11 H Br H OH 131b >10000 b

12 OH Br OH Br >10000 NDc 40%

13 Br Br H OH 202 ND 94%

19 H Br OCH3 OCH3 >10000 ND 46%

20 Br Br OCH3 OCH3 >10000 ND 23%

21 H Br OH OH ~10000 ND 52%

22 Br Br OH OH >10000 ND 44%

27 H Br H O-P(O)O2Na2 >10000 ND ND

a These values are averages of a minimum of a triplicate of experiments. Each assay utilized 2% DMSO with a 5 min pre-incubation period. b Previously reported203 c Not Determined

(3,5-Dibromophenyl)-(3-hyrdroxyphenyl) ketone thiosemicarbazone (13) exhibited comparable activity to KGP94 (11) against cathepsin L with an IC50 value of

202 nM (Table 3.2). With cathepsin L inhibition of 23% - 52% at 10 μM, activity of the symmetrical, dimethylresorcinol, and resorcinol thiosemicarbazone analogues19-22

99 borders the cutoff threshold (percent inhibition ≤ 50% at 10 μM). While having one m-

hydroxyl group on the aromatic ring opposing the 3-bromophenyl substituent in the

thiosemicarbazone analogues is important for activity against cathepsin L, for this group

of compounds, the presence of two m-hydroxyl or two m-dimethoxy substituents impairs

inhibitory activity against cathepsin L.

Chemical and Enzymatic Hydrolysis

Chromatogram of 18 hours ALP-treated KGP420 Rate Study of KGP420 (from an old batch) with 2% DMSO 2.5 30 2 KGP420, 1.564 min R = 0.9168 2.0

KGP94, 5. 598 min 20 Control 1.5 ALP-treated w/ 2% DMSO 1.0 10 Area (320nm) 0.5

Absorbance (mau, 320nm) (mau, Absorbance 0 0.0 0 2 4 6 8 10 1 2 3 4 5 6 7 Time (min) Time (min)

Figure 3.5. Incubation of KGP420 with 1.2 Units ALP.

The stability of phosphate prodrug KGP420 (27) in aqueous solution was

evaluated with and without the presence of alkaline phosphatase. Significant self-

hydrolysis was not observed for prodrug KGP420 (27) after a 48 hour incubation at 37 oC in glycine buffer solution (pH 8.6) (Figure 3.5). Additionally, KGP420 was not hydrolyzed when stored as a stock solution for one week at 4 oC. Enzymatic cleavage of

prodrug KGP420 (27) in alkaline phosphatase solution occurred with nearly 100%

conversion to the anticipated parent drug KGP94 (11) over the course of 18 hours.

KGP94 (11) was formed at a rate of 0.0832 μM/min (100 µM of KGP420 (27), 0.005

100 units of alkaline phosphatase) and the specific activity of alkaline phosphatase toward

KGP420 was determined to be 16.6 µM/ min.

Cytotoxicity

Table 3.3. Cytotoxicity against HUVECs

Compound Doxorubicin Paclitaxel KGP94 (11) KGP420 (27)

* * a Cytotoxicity GI50 (µM) 0.0268 0.00148 >50 >55.9 a Previously reported204

KGP94 (11) and the corresponding prodrug KGP420 (27) were evaluated for cytotoxicity toward normal primary cells. HUVECs were used a model for normal primary cells. Both KGP94 (11) and KGP420 (27) did not exhibit significant cytotoxicity against HUVECs especially compared to FDA approved cancer therapeutics doxorubicin and paclitaxel (Table 3.3).

Conclusion

Improved methodology resulted in an alternative synthesis of KGP94 which circumvented unwanted byproduct formation. In an effort to prepare thiosemicarbazone based inhibitors which exhibit a lower threshold for isomerization, new analogues with

KGP94 as the core structure were prepared and evaluated for inhibitory activity against cathepsin L. The most potent of these, [(3,5-dibromophenyl)(3-hydroxyphenyl) ketone] thiosemicarbazone exhibited comparable activity to KGP94 with and IC50 value of 202 nM. Advancement of the pre-clinical candidate, KGP94, through phosphate prodrug derivatization to generate KGP420 resulted in a water soluble analogue which is desirable for potentially increasing oral bioavailability. In vitro studies demonstrated that KGP420

101 is hydrolyzed to the parent compound, KGP94, in the presence of alkaline phosphatase.

Additionally, KGP420 favorably displayed low cytotoxicity to HUVECs which were

used as a model for normal cells.

Experimental Section

Chemistry

General Synthetic Protocols

All reactions were performed under inert atmosphere using nitrogen gas unless

otherwise specified. Chemicals and reagents used in the synthetic procedures were

purchased from commercial suppliers and used without further purification. Anhydrous

tetrahydrofuran and dichloromethane were purchased from commercial suppliers or dried

using a VAC (Vacuum Atmospheres Co.) solvent purification system. Reactions were

monitored by normal phase thin layer chromatography (TLC) plates SiliaPlateTM (silica

gel, 250 µM, F-254, 60 Å) and with reverse phase EMD Chemicals Inc. TLC plates (C18,

200-270 µM, F-254, 60 Å). Normal phase chromatographic purification was performed

using manual flash chromatography and automated flash chromatography which was carried out with silica gel purchased from either Silicycle Inc (230–400 mesh) or Biotage

(40–65 microns). Reverse phase chromatographic purification was performed using automated flash chromatography with a C-18 column purchased from Biotage (30g,

SNAP C18-KP-HS). Intermediates and products synthesized were characterized using a

Varian Inova 500 MHz NMR system and/or Bruker Avance III HD 600 MHz NMR system. Deuterated CDCl3 (with 0.03 % TMS as internal standard), acetone-d6, DMSO-

d6, and D2O were used as solvents for recording the NMR. All the chemical shifts are

102 expressed in ppm (δ), coupling constants (J, Hz) and peak patterns are reported as broad

(br), singlet (s), doublet (d), triplet (t), quartet (q) and multiplet (m). High resolution mass spectra (HRMS) were obtained in the Baylor University Mass Spectrometry Core Facility on a Thermo Scientific LTQ Orbitrap Discovery using electrospray ionization ESI. Purity of the final compounds was analyzed using an Agilent Technologies 1200 series HPLC system with a Diode Array and Multiple Wavelength Detector SL, equipped with an

Agilent Eclipse XDB-C18 column (4.6 mm ID x 250 mm, 5 μm particle size, 80 Å pore size). HPLC method parameters: T = 25 oC; 1.0 mL/min; injection volume 20 μL;

monitored at 254 nm, 300 nm, 320 nm. Four different HPLC gradients were used for purity analysis; Method A: acetonitrile/water, gradient 30:70 to 70:30 from 0 to 20 min,

70:30 to 100:0 from 20 to 25 min, isocratic 100:00 from 25 to 30 min, gradient 100:00 to

30:70 from 30-32 min, and isocratic 30:70 from 32 to 40 min; Method B:

water/acetonitrile, gradient 30:70 to 90:10 from 0 to 25 min and isocratic 90:10 from 25

to 30 min; Method C: water/acetonitrile, gradient 50:50 to 90:10 from 0 to 25 min and

isocratic 90:10 from 25 to 30 min; Method D: 0.05 % trifluoroacetic acid in

water/acetonitrile, isocratic 10:90 from 0 to 5 min, gradient 10:90 to 100:0 from 5 to 25

min and isocratic 100:00 from 25 to 30 min.

(3-Bromophenoxy)-tert-butyl-dimethyl-silane (1).

Tert-butyl dimethylsilyl chloride (3.150 g, 21.00 mmol) was added to a solution

of imidazole (1.900 g, 27.94 mmol) and 3-bromophenol (1.520 mL, 14.01 mmol) in

anhydrous DMF (40 mL) at 0o C. The reaction mixture was stirred for 6 hrs. Upon

completion of the reaction, 5% aqueous NaHCO3 (20 ml) was added to the reaction

mixture. The products were extracted with hexanes (2 x 50 mL) and concentrated under

103 reduced pressure. Purification by flash column chromatography (silica gel, hexanes

100%) afforded (3-bromo-phenoxy)-tert-butyl-dimethyl-silane (3.905 g, 13.59 mmol,

1 97% yield) as a colorless oil. H NMR (500 MHz, CDCl3) δ 7.10-7.06 (2H, m), 7.01-7.00

13 (1H, m), 6.78-6.74 (1H, m), 0.98 (9H, s), 0.20 (6H, s). C NMR (125 MHz, CDCl3)

δ156.67, 130.55, 124.61, 123.66, 122.61, 118.96, 25.75, 18.33, -4.32.

(3,5-Dibromophenoxy)-tert-butyldimethylsilane (2).

3,5-dibromophenol (3.78 g, 15.0 mmol) was dissolved in N,N-dimethylformamide

(45 mL) followed by the addition of imidazole (2.04 g, 30.0 mmol). The reaction mixture

was cooled to 0 oC and tert-butyldimethylchlorosilane (3.37 g, 22.5 mmol) was added.

The reaction mixture was returned to room temperature and stirred for 4 h. After reaction

completion, the reaction mixture was quenched with saturated aqueous sodium

bicarbonate (50 mL) and the product was extracted with hexanes (3 X 50 mL). The

organic extracts were dried over anhydrous sodium sulfate and concentrated under

reduced pressure. The crude mixture was purified using flash chromatography (silica gel,

hexanes) to afford (3,5-dibromophenoxy)-tert-butyldimethylsilane (5.38 g, 14.7 mmol ,

1 98%). H NMR (600 MHz, CDCl3): δ 7.26 (1H, t, J = 1.7 Hz), 6.93 (2H, d, J = 1.7 Hz),

13 0.97 (9H, s), 0.21 (6H, s). C NMR (150 MHz, CDCl3): δ 156.98, 127.12, 122.76,

122.35, 25.49, 18.12, -4.53. tert-Butyldimethylsilyl 3-bromo-5-((tert-butyldimethylsilyl)oxy)benzoate (Precursor for

Compound 3).

3-Bromo-5-hydroxybenzoic acid (1.960 g, 9.031 mmol) and imidazole (4.064,

27.09 mmol) were dissolved in anhydrous dichloromethane (35 mL) followed by the

addition of tert-butyldimethylchlorosilane (3.684, 54.18 mmol) and the reaction mixture

104 was refluxed for 5 h. The reaction mixture was allowed to cool to room temperature and

was quenched with water (50 mL) and the product was extracted with dichloromethane (3

X 50 mL). The organic extracts were dried over sodium sulfate and concentrated. . The

crude mixture was purified using flash chromatography (silica gel, hexanes:ethyl acetate,

gradient, 98:02 to 70:30) to afford tert-butyldimethylsilyl 3-bromo-5-((tert-

butyldimethylsilyl)oxy)benzoate (1.587 g, 3.562 mmol, 39%). 1H NMR (600 MHz,

CDCl3): δ 7.84 (1H, t, J = 1.6 Hz), 7.47 (1H, dd, J = 2.3 Hz, 1.4 Hz), 7.23 (1H, dd, J =

2.3 Hz, 1.8 Hz), 0.99 (9H, s), 0.91 (9H, s), 0.24 (6H, s), 0.11 (6H, s). 13C NMR (150

MHz, CDCl3): δ 169.89, 156.54, 131.71, 128.66, 126.11, 122.56, 120.35, 25.62, 25.54,

18.17, 17.98, -3.60, -4.49.

3-Bromo-5-((tert-butyldimethylsilyl)oxy)benzoyl chloride (3).

Oxalyl chloride (359 μL, 4.19 mmol) was added dropwise to a solution of tert-

butyldimethylsilyl 3-bromo-5-((tert-butyldimethylsilyl)oxy)benzoate (1.494 g, 3.353

mmol) in dichloromethane (10 mL). A catalytic amount of DMF (2.6 μL, 0.034 mmol)

was added dropwise and the reaction mixture was stirred for 12 h at room temperature.

After concentration, the reaction mixture was dissolved in dichloromethane and concentrated and repeated to afford 3-bromo-5-((tert-butyldimethylsilyl)oxy)benzoyl

chloride. The product was used immediately without further purification. 1H NMR (600

MHz, CDCl3): δ 7.85 (1H, t, J = 1.7 Hz), 7.47 (1H, dd, J = 2.3 Hz, 1.7 Hz), 7.29 (1H, dd,

13 J = 2.3 Hz, 1.7 Hz), 0.99 (9H, s), 0.25 (6H, s). C NMR (150 MHz, CDCl3): δ 167.01,

156.77, 135.53, 130.03, 127.05, 122.99, 121.21, 25.50, 18.18, -4.49.

105 3-Bromo-N-methoxy-N-methylbenzamide (4).