University of Southern California Dissertations and Theses

Synthetic small molecules and protein secondary structure mimetics as modulators of hypoxia-inducible transcription and integrin receptors function

(USC Thesis Other)

Synthetic small molecules and protein secondary structure mimetics as modulators of hypoxia-inducible transcription and integrin receptors function

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Synthetic Small Molecules and Protein Secondary Structure
Mimetics as Modulators of Hypoxia-Inducible Transcription
and Integrin Receptors Function

by
Swati Kushal

______________________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)

August 2013

ii
DEDICATION
This dissertation is dedicated to my husband, Venoo Mishra and to my little angel, Aanya
Mishra. I would like to give them heartfelt thanks for their unconditional love and
support. Thanks Aanya for your cute little hugs every day for the past three years.

iii
ACKNOWLEDGEMENTS
The completion of this work has been accomplished due to the contribution of several
people, who with their continued guidance and support kept me motivated. My immense
gratitude goes to my advisor Professor Bogdan Z Olenyuk for his constant motivation,
insightful discussions, and for his valuable scientific inputs. His enthusiasm and
thoughtful feedback kept me focused towards my objectives. I would also like to thank
him for providing me an opportunity to work on the cutting edge multidisciplinary
projects, which over the years have helped me to hone my skills.
I would also like to thank Dr. Hui Wang, Dr. Csaba Laszlo and Dr. Lajos Szabo
for their assistance in the synthesis of ETP project. Thanks also to Dr. Ramin Dubey and
Dr. Nathan Polaske for their contribution in the integrin project. I would like to gratefully
acknowledge Prof. Paramjit Arora and Dr. Laura Henchey at NYU, for their
collaboration on Hydrogen bond surrogate project. In addition, I would like to thank Prof.
Schnitzer at PRISM, San Diego for help with intravital microscopy. I also wish to express
my sincere appreciation to my dissertation committee members Prof. Clay Wang and
Prof. Matthew Pratt and my graduate committee members Prof. Wei-Chiang Shen and
Prof. Curtis Okamoto.
I would like to thank my group members, Ivan Grishagin and John Gallagher, for
providing a conducive work atmosphere and good humor. I would like to take this
opportunity to thank my good friend, Vineela Kadiyala for light hearted talks and keeping
me encouraged.
iv
I owe a sincere thanks to my loving parents and parents-in-law. It wouldn’t have
been possible without their blessings, sacrifices, and encouragement. I thank you for
being there, whenever needed. This venture would not be possible without love and
support of my sister, brother, brother-in-law and sister-in-law.
Everyone needs a support system to thrive in their life, and for me it is my
husband, Venoo Mishra. His support and motivation was the inspiration and driving force
for making me move towards my goal. Thanks for all your support and cheer through all
the years. Finally, a special thanks to my daughter, Aanya Mishra, for putting a smile on
my face every day. I cannot thank you and GOD enough for making you such a precious
part of my life.

v
TABLE OF CONTENTS
DEDICATION .................................................................................................................... ii
ACKNOWLEDGEMENTS ............................................................................................... iii
LIST OF FIGURES .......................................................................................................... xii
LIST OF TABLES ........................................................................................................... xix
LIST OF SCHEMES......................................................................................................... xx
ABSTRACT ..................................................................................................................... xxi
Chapter 1: Introduction ....................................................................................................... 1
PART I: Small Molecules as Transcriptional Modulators .............................................. 2
1.1 Transcription Factors as Therapeutic Targets .................................................. 2
1.2 The role of hypoxia in cancer biology .............................................................. 5
1.3 Physiological and pathophysiological roles of HIF-1α .................................... 8
1.4 Regulation of hypoxia-inducible factor signaling .......................................... 11
1.4.1 Oxygen-independent regulation of hypoxia-inducible factor signaling ....... 15
1.4.2 Therapeutic intervention of hypoxia-inducible factor pathway .................... 17
1.5 Introduction to epipolythiodiketopiperazines (ETPs) ..................................... 20
1.5.1 Biological activities of epipolythiodiketopiperazines ................................... 21
1.5.2 Mode of action of epipolythiodiketopiperazines .......................................... 24
1.6 Chetomin: A modulator of the HIF transcriptional pathway .......................... 26
1.7 Introduction to HIF-1α target genes ............................................................... 27
1.7.1 The role of VEGF in tumor progression ....................................................... 28
1.7.2 Role of LOX in tumor progression .............................................................. 32
1.7.3 Role of Glut-1 in tumor progression ............................................................. 33
1.7.4 The role of CXCR4 in tumor progression..................................................... 34
1.8 Chemical approaches to target HIF pathway .................................................. 36
vi
1.8.1 Allosteric approach ....................................................................................... 36
1.8.2 Orthosteric approach ..................................................................................... 36
1.9 Screening of a small molecule library ............................................................ 38
PART II: Targeting α
v
β
3
integrin receptors .................................................................. 39
1.10 Cell surface receptors ..................................................................................... 39
1.11 Integrin family of adhesion receptors ............................................................. 40
1.12 Integrin signaling and tumorigenesis .............................................................. 43
1.13 Targeting α
v
β
3
integrin receptors .................................................................... 46
Chapter 2: Modulation of Hypoxia-Inducible Transcription by Rationally Designed
Epipolythiodiketopiperazines ........................................................................................... 49
2.1 Structural basis of interaction between HIF-1α and p300 ................................... 50
2.2 Chetomin blocks the interaction between HIF-1α and p300 ............................... 52
2.3 Rationale behind the design of novel monoepipolythiodiketopiperazines
(monoETPs) ............................................................................................................... 53
2.4 Synthesis of mono epipolythiodiketopiperazines (monoETPs) ........................... 56
2.4.1 Synthesis of mono-ETP-1 ............................................................................. 57
2.4.2 Synthesis of mono-ETP-2 ............................................................................. 57
2.4.3 Synthesis of mono-ETP-3 ............................................................................. 58
2.4.4 Synthesis of monoETP-4 .............................................................................. 59
2.4.5 Synthesis of mono-ETP-5 ............................................................................. 60
2.5 Preliminary structure activity relationship (SAR) studies of mono-ETPs .......... 61
2.6 Synthetic mono-ETP-5 disrupts HIF-1 /p300 complex ..................................... 63
2.6.1 Determination of the binding affinity between HIF-1 -flu CTAD and p300-
CH1-GST by fluorescence polarization (FP) saturation binding assay ................. 64
2.6.2 Fluorescence polarization (FP) competition assay with mono-ETP-5, mono-
ETP-4, and chetomin ............................................................................................. 65
vii
2.7 Cytotoxicity of ETP-5, ETP-4, and chetomin in MDA-MB-231-HRE-Luc cell
line ............................................................................................................................. 68
2.8 Evaluation of mono-ETP-5 as transcriptional modulator of hypoxia-inducible
pathway ...................................................................................................................... 70
2.9 Synthetic mono-ETP-5 downregulates levels of secreted VEGF ........................ 72
2.10 Effect of mono-ETP-5 on the global fold of p300-CH1 .................................... 74
2.11 SAR studies with bis-epidithiodiketopiperazines .............................................. 75
2.11.1 Rationale behind the design of bis-epidithiodiketopiperazines (bis-ETPs) 75
2.11.3 Synthetic bis-ETPs inhibits the activity of VEGF promoter ...................... 78
2.13 Conclusion ......................................................................................................... 83
Chapter 3: Novel Hydrogen Bond Surrogates as Orthosteric Modulators of Hypoxia
Inducible Transcription ..................................................................................................... 86
3.1 Helical protein interfaces and their critical role in transcription ......................... 87
3.2 Structural determinants of transcription factor HIF-1 and its coactivator p30088
3.3 Approaches to stabilize or mimic α-helical fold .................................................. 89
3.4 Hydrogen bond surrogate (HBS) approach to stabilize short α-helices .............. 91
3.4.1 General synthesis of HBS helices ................................................................. 93
3.4.2 Structural characterization of HBS α-helices ................................................ 95
3.4.3 The potential of HBS helices as modulators of protein-protein interactions 98
3.5 The rationale behind the design of first generation of αA-helix mimics of HIF-1α
................................................................................................................................. 100
3.6 Results of the first generation of αA-helix mimetics of HIF-1α ....................... 101
3.7 Rationale behind the design of the second generation of αA-helix mimics of HIF-
1α ............................................................................................................................. 102
3.8 Characterization of second generation of αA-helix mimics of HIF-1α HBS by
circular dichroism spectroscopy .............................................................................. 104
3.9 Affinity of the second generation of an αA-helix mimics toward p300 CH1
domain ..................................................................................................................... 106
viii
3.10 Cytotoxicity of HBS 4 and peptide 3 in HeLa cells ........................................ 109
3.11 Transcriptional regulation of HIF-1α genes in HeLa cells treated with HBS 4,
Peptide 3 and chetomin............................................................................................ 110
3.12 The rationale behind the design of αB-helix mimics of HIF-1α ..................... 114
3.13 Characterization of secondary structure of a small library of HBS αB mimics of
HIF-1α ..................................................................................................................... 116
3.14 HBS αB helices disrupt the HIF-1α/p300 complex in vitro ............................ 117
3.15 Impact of GST tag on the binding affinities of HBS αB helices in fluorescence
polarization assays ................................................................................................... 120
3.16 Determination of direct binding of HBS 9 to GST-p300-CH1 by fluorescence
polarization assay .................................................................................................... 123
3.17 Cytotoxicity of HBS 6 and Chetomin in HeLa cells ....................................... 125
3.18 Transcriptional regulation of HIF pathway by designed HBS αB helices ...... 126
3.19 HIF-1α levels remain unchanged upon treatment with HBS 6........................ 136
3.20 HBS 6 downregulates the endogenous level of VEGF protein ....................... 137
3.21 Conclusion ....................................................................................................... 138
Chapter 4: Multifunctional integrin-selective small molecules for tumor targeting ....... 139
4.1 Integrins and their relevance in the regulation of signaling pathways .............. 140
4.2 Targeting α
v
β
3
integrin receptors as a therapeutic strategy for tumorigenesis .. 142
4.3 Active targeting of tumors by boron neutron capture therapy (BNCT) ............ 146
4.4 Design and synthesis of α
v
β
3
ligand-dye conjugates for optical imaging of
integrin receptors ..................................................................................................... 150
4.5 Photophysical properties of α
v
β
3
integrin ligand conjugates, ABL-1 and ABL-2
................................................................................................................................. 155
4.6 Cytotoxicity of integrin-ligand dye conjugate (ABL-1) in melanoma and breast
cancer cell lines ....................................................................................................... 160
4.7 Cellular uptake of integrin-ligand dye conjugates ............................................. 161
ix
4.7.1 Investigation of uptake and internalization of ABL-1 by confocal microscopy
.............................................................................................................................. 162
4.7.2 Distribution of integrin-ligand dye conjugate, ABL-1 in WM115 cells by
flow cytometry ..................................................................................................... 167
4.7.3 Measurement of the apoptosis rate of WM115 cells in the presence of ABL-
1 by flow cytometry ............................................................................................. 169
4.8 Integrin-ligand dye conjugate ABL-1 disrupts the cellular adhesion to vitronectin
................................................................................................................................. 170
4.9 Intravital microscopy imaging of ABL-1 in murine subcutaneous and in ectopic-
orthotopic tumor xenograft models with ................................................................. 174
4.10 Design and biological properties of trifunctional carborane-integrinligand dye
conjugate .................................................................................................................. 175
4.11 Conclusion ....................................................................................................... 181
Chapter 5: Targeting monoamine oxidase A (MAOA) in prostate cancer ..................... 183
5.1 Introduction to monoamine oxidase (MAOA) .................................................. 184
5.2 Monoamine oxidase inhibitors (MAOI) ........................................................... 186
5.3 The role of monoamine oxidase A (MAOA) in prostate cancer ....................... 188
5.4 Near infrared dyes (NIR) for optical imaging and cancer targeting .................. 191
5.5 Design and biological evaluation of NIR dye-MAOA inhibitor conjugate....... 193
5.5.1 Effect of MHI-Clg on cell viability ............................................................ 194
5.5.2 Assessment of MHI-Clg uptake in C4-2B cells by confocal microscopy .. 197
5.5.3 Uptake and accumulation of MHI-Clg in tumors in live mice ................... 198
5.6 Conclusion ......................................................................................................... 199
Chapter 6: Experimental Section .................................................................................... 200
6.1 Introduction to Experimental Section ................................................................ 201
6.2 Experimental section of Chapter 2 .................................................................... 203
6.2.1 Synthesis of mono-ETP compounds ........................................................... 203
x
6.2.2 Protein expression ....................................................................................... 209
6.2.3. Purification of fusion protein p300-CH1-GST .......................................... 210
6.2.4. Fluorescence polarization (FP) saturation binding assay for the
determination of K
d
(affinity constant) between HIF1a-flu CTAD and p300-CH1-
GST ...................................................................................................................... 211
6.2.5 Fluorescence polarization (FP) competition binding assay for the
determination of binding affinity of mono-ETPs and bis-ETPs towards the p300-
CH1-GST and HIF-1α-flu CTAD complex ......................................................... 211
6.2.6 Effect of mono-ETP on the structure of p300-CH1-GST by circular
dichroism (CD) spectroscopy .............................................................................. 212
6.2.7 Determination of cell viability by cell titer blue cytotoxicity assay ........... 213
6.2.8 Luciferase reporter assays ........................................................................... 214
6.2.9 Analysis of secreted VEGF by Enzyme-linked immunosorbent assay
(ELISA)................................................................................................................ 215
6.2.10 Luciferase reporter assay for screening library of small molecules ......... 216
6.3 Experimental Section of Chapter 3 ................................................................... 217
6.3.1 Fluorescence polarization (FP) saturation binding assay for the determination
of K
d
(affinity constant) between HIF1a-flu CTAD and p300-CH1 ................... 217
6.3.2 Fluorescence polarization (FP) competition binding assay for the
determination of binding affinity of HBS helices towards the p300-CH1 (with and
without GST tag) and HIF-1α-flu CTAD complex ............................................. 217
6.3.3 Purification of fusion protein p300-CH1-GST at higher concentration ..... 218
6.3.4 Fluorescence polarization assay for the direct binding of HBS helix to p300-
CH1-GST ............................................................................................................. 219
6.3.5 Circular Dichroism Spectroscopy ............................................................... 219
6.3.6 Determination of binding affinity by Isothermal titration Calorimeter (ITC)
.............................................................................................................................. 220
6.3.7 Cell density and population doubling assay................................................ 221
6.3.8 Cell viability assay ...................................................................................... 221
xi
6.3.9 Plating and dosing for isolation of mRNA αA- Helices ............................. 222
6.3.10 Isolation of mRNA .................................................................................... 224
6.3.11 Analysis of gene expression...................................................................... 224
6.3.12 Analysis of secreted VEGF by Enzyme-linked immunosorbent assay
(ELISA)................................................................................................................ 225
6.3.13 Analysis of HIF-1α by Western blotting................................................... 226
6.3.14 Luciferase assay with hypoxia induction using anaerobic pouch ............. 227
6.4 Experimental Section of Chapter 4 .................................................................... 228
6.4.1 Cell viability assay for integrin compounds ............................................... 228
6.4.2 Absorption spectra measurement ................................................................ 229
6.4.3 Fluorescence excitation and emission measurement .................................. 229
6.4.4 Confocal microscopy of integrin compounds ............................................. 230
6.4.5 Flow cytometry of ABL-1, CIL-1 and CD-1 .............................................. 231
6.4.6 Adhesion assay............................................................................................ 231
6.4.7 Intravital microscopy with ABL-1 .............................................................. 232
6.5 Experimental Section of Chapter 5 ................................................................... 233
6.5.1 MTS assay with MHI-Clg and MHI-148 .................................................... 233
6.5.2 Confocal microscopy .................................................................................. 233
6.5.3 Near Infrared Imaging (NIR) imaging ........................................................ 234
BIBLIOGRAPHY ........................................................................................................... 235
APPENDIX A ................................................................................................................. 261
APPENDIX B ................................................................................................................. 277
APPENDIX C ................................................................................................................. 284

xii
LIST OF FIGURES
Figure 1.1 Central dogma of molecular biology. ............................................................... 3
Figure 1.2 Structural domains of HIF transcription factor family. .................................. 12
Figure 1.3 Regulation of HIF-1α signaling pathway. ...................................................... 14
Figure 1.4 Oxygen-independent regulation of HIF. ......................................................... 15
Figure 1.5 Natural product inhibitors of HIF-1 pathway. ............................................... 17
Figure 1.6 Small molecule inhibitors of HIF-1 pathway. ................................................ 18
Figure 1.7 Biologically active monoETP compounds. ................................................... 22
Figure 1.8 Biologically active bis-ETP compounds. ...................................................... 23
Figure 1.9 Proposed mechanism of action of ETPs. ........................................................ 26
Figure 1.10 Family of VEGF receptors and ligands. ...................................................... 29
Figure 1.11 Schematic representation of VEGF pathway............................................... 30
Figure 1.12 Schematic representation of CXCR4 pathway. ........................................... 35
Figure 1.13 Allosteric and orthosteric approach to target HIF-1α and p300 interface. .. 37
Figure 1.14 Classification of family of integrin receptors. ............................................. 40
Figure 1.15 Schematic representation of extracellular domain of α and β subunit......... 42
Figure 1.16 Schematic representation of integrin signaling. .......................................... 44
Figure 1.17 Examples of Cyclic RGD peptide. .............................................................. 47
Figure 2.1 Domain structures of HIF-1α and p300/CBP ................................................ 50
Figure 2.2 Solution structure of CH1 domain of p300 with CTAD domain of HIF-1α. 51
Figure 2.3 Structure of chetomin.. .................................................................................. 52
Figure 2.4 Structure of natural products, chetomin and leptosin I. ................................. 54
Figure 2.5 Structures of monomeric ETPs. ..................................................................... 55
Figure 2.6 Structures of designed monoETPs for the structure activity relationships. ... 56
xiii
Figure 2.7 Luciferase reporter assay in MDA-MB-231-HRE-Luc cell line with
chetomin, synthetic monoETPs, and gliotoxin .......................................................... 62
Figure 2.8 Luciferase reporter assay for dose respose studies with monoETP-4 and
gliotoxin. .................................................................................................................... 63
Figure 2.9 Amino acid sequence of GST-p300-CH1 domain ......................................... 64
Figure 2.10 Synthetic monoETP-5 disrupts the complex of CTAD HIF-1α and p300-
CH1.. .......................................................................................................................... 66
Figure 2.11 FP assay with synthetic monoETP-4 ........................................................... 67
Figure 2.12 Chetomin disrupt the complex of CTAD HIF-1α and p300-CH1 with an
IC
50
value of 633 nM.. ............................................................................................... 68
Figure 2.13 Cell titer blue assay for viability with monoETP-5 in mDA-MB-231-HRE-
Luc. ............................................................................................................................ 69
Figure 2.14 Cell titer blue assay for viability with monoETP-4 in mDA-MB-231-HRE-
Luc ............................................................................................................................. 70
Figure 2.15 Inhibition of HIF-1α inducible promoter activity in a stably transfected
MDA-MB-231-HRE-Luc cell line by synthetic monoETPs.. ................................... 71
Figure 2.16 Analysis of secreted levels of VEGF protein with monoETP-5.. ................ 73
Figure 2.17 Effect of monoETP-5 on the global fold of CH1 domain of p300 by CD
spectroscopy.. ............................................................................................................ 74
Figure 2.18 Structures of chetomin and designed dimeric ETP molecules.. .................. 76
Figure 2.19 Synthetic bisETP-1 disrupt the complex of CTAD HIF-1α and p300-CH1 77
Figure 2.20 Synthetic bisETP-2 disrupt the complex of CTAD HIF-1α and p300-CH1.
................................................................................................................................... 78
Figure 2.21 Dose-dependent inhibition of HIF-1α inducible promoter activity by
bisETP-1 and bisETP-2 in MDA-MB-231-HRE-Luc cell line.. ............................... 79
Figure 2.22 Two enantiomers of bisETP-1, ent1-bisETP-1 and ent2-bisETP-2, showed
similar inhibition of HIF-1α inducible promoter activity in the MDA-MB-231-HRE-
Luc cell line. .............................................................................................................. 80
Figure 2.23 Pictorial presentation of screening of a library of small molecules in a
luciferase reporter assay. ........................................................................................... 81
xiv
Figure 2.24 Inhibition of HIF-1α inducible promoter activity in a stably transfected
MDA-MB-231-HRE-Luc cell line by small molecules. ........................................... 82
Figure 2.25 Targeting HIF pathway by small molecules ................................................ 83
Figure 3.1 NMR structure of complex between C-terminal activation domain (CTAD) of
HIF-1α and cysteine-histidine rich region 1(CH1) of p300/CBP ............................. 89
Figure 3.2 Different approaches to stabilize the α-helical conformation in peptides or
mimicking it with non-natural scaffolds.. .................................................................. 90
Figure 3.3 Hydrogen bond surrogate (HBS) approach. .................................................. 92
Figure 3.4 General scheme for the synthesis of hydrogen bond surrogate (HBS) based α-
helices ........................................................................................................................ 94
Figure 3.5 CD spectra of constrained HBS α-helix and its unconstrained analog.. ........ 96
Figure 3.6 NMR derived structureof HBS Bak-BH3 HBS helix. ................................... 97
Figure 3.7 Lactum bridged approach and HBS based helices fortargeting Bak-BcL-xL.
................................................................................................................................... 99
Figure 3.8 Circular dichroism spectra of designed HBS α-helices and unconstrained
peptide ..................................................................................................................... 104
Figure 3.9 ITC raw data of HBS 4 into a solution of GST-p300-CH1 ......................... 107
Figure 3.10 Cell density and cell population doubling assay with HBS 4, peptide 3 and
with chetomin .......................................................................................................... 109
Figure 3.11 Inhibition of VEGF mRNA levels measured in HeLa cells by qRT-PCR..
................................................................................................................................. 111
Figure 3.12 Inhibition of Glut1 mRNA levels as measured in HeLa cells by qRT-PCR.
................................................................................................................................. 113
Figure 3.13 Effect of HBS 4 on the global fold of p300-CH1 domain.. ....................... 114
Figure 3.14 Saturation binding curve for GST-p300-CH1 and HIF-1α-flu (786-826) as
measured by fluorescence polarization assay.. ........................................................ 117
Figure 3.15 Fluorescence polarization competition assay showing disruption of CTAD
domain of HIF-1α and CH1 domain of p300 by designed HBS αB helices. .......... 119
Figure 3.16 Combined FP competition assay results with designed αB helices. .......... 120
xv
Figure 3.17 Saturation binding curve for GST removed p300-CH1 domain and HIF-1α-
flu (786-826) as measured by fluorescence polarization assay.. ............................. 121
Figure 3.18 Fluorscence polarization competition assay exhibiting the disruption of GST
cleaved CH1 domain of p300 and CTAD domain of HIF-1α with Chetomin and HBS
8 ............................................................................................................................... 122
Figure 3.19 Graph showing combined FP competition assay results with designed αB
helices in presence of the complex of GST cleaved p300 and CTAD domain of HIF-
1α. ............................................................................................................................ 123
Figure 3.20 Direct binding of HBS 9-flu to the CH1 domain of p300 as determined by
fluorescence polarization assay. .............................................................................. 124
Figure 3.21 Cell viability of HBS 6 in HeLa cells as determined by MTT assay. ....... 125
Figure 3.22 Inhibition of VEGF mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6, HBS 7 and peptide 4 n 10% serum. ..................... 126
Figure 3.23 Inhibition of Glut1 mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6, HBS 7 and peptide 4 in 10% serum. .................... 128
Figure 3.24 Inhibition of LOX mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6, HBS 7 and peptide 4 in 10% serum. .................... 129
Figure 3.25 Inhibition of CXCR4 mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6 in 10% serum. ........................................................ 130
Figure 3.26 VEGF mRNA levels as measured in HeLa cells by qRT-PCR in 0.2 %
serum. ...................................................................................................................... 132
Figure 3.27 LOX mRNA levels as measured in HeLa cells by qRT-PCR in 0.2 % serum
................................................................................................................................. 133
Figure 3.28 Glut1 and CXCR4 mRNA levels as measured in HeLa cells by qRT-PCR in
0.2% serum. ............................................................................................................. 134
Figure 3.29 c-Met mRNA levels as measured in HeLa cells by qRT-PCR.. ................ 135
Figure 3.30 Analysis of HIF-1α levels by western blotting. ......................................... 136
Figure 3.31 Analysis of VEGF protein levels in HeLa cells ........................................ 137

xvi
Figure 4.1 Interaction between growth factor receptor VEGFR2 and integrin receptor
α
v
β
3
. ................................................................................................................................. 143
Figure 4.2 Non-peptidic small molecule antagonists of α
v
β
3
.. ...................................... 145
Figure 4.3 Schematic of boron neutron capture reaction. ............................................. 147
Figure 4.4 Structure of first and second generation boron compound. ......................... 148
Figure 4.5 Structures of ortho, meta and para carborane clusters. ................................ 149
Figure 4.6 Structure of α
v
β
3
specific integrin-ligand dye conjugate, ABL-1................ 151
Figure 4.7 Structure of sulfur (ABL-1) and selenium (ABL-2) based α
v
β
3
specific
integrin-ligand dye conjugates. ................................................................................ 155
Figure 4.8 Uv-vis absorption spectra of ABL-1 in aqueous buffer and 90% glycerol
v/v/water system. ..................................................................................................... 156
Figure 4.9 Uv-vis absorption spectra of ABL-2 in 90% glycerol v/v/water system. .... 156
Figure 4.10 Fluorescence emission spectra of ABL-1 in aqueous buffer and 90%
glycerol v/v/water system ........................................................................................ 157
Figure 4.11 Fluorescence emission spectra of ABL-2 in aqueous buffer and 90%
glycerol v/v/water system ........................................................................................ 158
Figure 4.12 Fluorescence emission spectra of ABL-1 in PBS and with WM115 cells 159
Figure 4.13 Fluorescence emission spectra of ABL-2 in PBS and with WM115 cells 159
Figure 4.14 Cell titer blue cytotoxicity assay with ABL-1 in WM115 and MCF-7 cells
................................................................................................................................. 160
Figure 4.15 MTT cytotoxicity assay with ABL-1 in WM115 and MCF-7 cells. ......... 161
Figure 4.16 Confocal microscopy images of WM115 treated with 5 µM of ABL-1 at 37
°C ............................................................................................................................ 162
Figure 4.17 Confocal microscopy image of WM115 cells treated with 100 µM of ABL-1
for 3 h at 37 °C. ....................................................................................................... 163
Figure 4.18 Confocal microscopy image of WM115 cells treated with 5 µM of ABL-1
and MCF-7 cells treated with 5 µM of ABL-1 ........................................................ 164
Figure 4.19 Confocal microscopy image of WM115 cells treated with 100 µM of ABL-2
and 100 µM of ABL-1 ............................................................................................. 165
xvii
Figure 4.20 Confocal microscopy images of WM115 cells treated with 5 µM of ABL-1
for 2 h at 4 °C and MCF-7 cells treated with 5 µM of ABL-1 for 2 h at 4 °C ........ 166
Figure 4.21 Confocal images of WM115 cells and MCF7 cells after incubation with 20
µM of ABL-1 and 200 µM of 1 in cell culture media ............................................. 167
Figure 4.22 Distribution and uptake of ABL-1 in WM115 cells by flow cytometry. .. 168
Figure 4.23 WM115 cells treated with Annexin-V and increasing concentration of ABL-
1 ............................................................................................................................. 169
Figure 4.24 WM115 cells treated with 7-AAD and increasing concentration of ABL-1
................................................................................................................................. 170
Figure 4.25 Structure of negative control ABL-3. ........................................................ 171
Figure 4.26 Adhesion disruption by ABL-1 and Cyclo-(RGDfV) ............................... 172
Figure 4.27 Combined graph of adhesion adhesion assay. ........................................... 173
Figure 4.28 Effect of ABL-1 and cyclo-(RGDfV) on the adhesion process of WM115 to
fibronectin. ............................................................................................................... 174
Figure 4.29 Intravital microscopy (IVM) images showing uptake of the ABL-1 in
subcutaneous tumor models and ectopic-orthotopic model. ................................... 175
Figure 4.30 Structure of carborane integrin-ligand dye conjugate, CIL-1.................... 176
Figure 4.31 Structure of carborane -dye conjugate, CD-1. ........................................... 177
Figure 4.32 Confocal images of WM115 cells and MCF-7 cells treated with CIL-1.. 178
Figure 4.33 Confocal images of WM115 cells and MCF-7 cells treated with CD-1 .. 178
Figure 4.34 WM115 cells treated with CIL-1and CD-1 at 4 °C for 2h. ....................... 179
Figure 4.35 Distribution and uptake of CIL-1, ABL-1 and CD-1 by flow cytometry .. 180
Figure 5.1 Mechanism of action of MAO and MAOI …… …………………………..186
Figure 5.2 Structures of MAO inhibitors. ..................................................................... 187
Figure 5.3 Structure of NIR dyes. ................................................................................. 191
Figure 5.4 Structure of MHI-Clg conjugate.. ................................................................ 193
Figure 5.5 Cytotoxicity of MHI-Clg in LNCaP cells. ................................................... 195
xviii
Figure 5.6 Cytotoxicity of MHI-Clg in C4-2B cells. .................................................... 195
Figure 5.7 Cytotoxicity of MHI-Clg in PC-3 cells. ...................................................... 196
Figure 5.8 Uptake of MHI-Clg by C4-2B cells ............................................................ 197
Figure 5.9 In vivo NIR imaging of MHI-Clg ................................................................ 198

xix
LIST OF TABLES
Table 1 Summary of results of first generation of αA-helix mimetics of HIF-1α ........ 101
Table 2 Second generation of αA-helix mimics of HIF-1α .......................................... 103
Table 3 Tabular presentation of percentage helicity of second generation of HBS α-helix
mimics as measured by circular dichroiam spectroscopy. ...................................... 105
Table 4 Tabular summary of key biophysical data of second generation of HBS α-helix
mimics ..................................................................................................................... 108
Table 5 Summary of results of biophysical and transcriptional inhibition assays of
second generation of HBS αA helices. .................................................................... 112
Table 6 Tabular presentation of all designed αB HBS helices ...................................... 115
Table 7 Change in percentage helicity of designed HBS αB helices and unconstrained
peptide 4. ................................................................................................................. 116

xx
LIST OF SCHEMES
Scheme 2.1 Synthetic scheme for monoETP-1. ............................................................... 57
Scheme 2.2 Synthetic scheme for monoETP-2. .............................................................. 58
Scheme 2.3 Synthetic scheme for monoETP-3. .............................................................. 58
Scheme 2.4 Synthetic scheme for monoETP-4. .............................................................. 59
Scheme 2.5 Synthetic scheme for monoETP-5. .............................................................. 60
Scheme 4.1 Synthesis of the fluorescent probe integrin ligand, 1 ................................. 152
Scheme 4.2 Synthesis of monomethine cyanine dye, cyan-39. ..................................... 154
Scheme 4.3 Synthesis of integrin-ligand dye conjugate, ABL-1.. ................................ 154

xxi
ABSTRACT
Transcription factors are the key regulators of cancer gene expression and play a critical
role in every aspect of tumorigenesis. Despite their relevance in tumor progression, the
shallow binding surfaces and absence of the dependence on enzymatic activity makes
them challenging therapeutic targets. However, now the fields of chemical biology,
genetics, cancer biology, and biotechnology have evolved to a point where targeting
transcription factors has become tractable. Direct modulation of the activity of
transcription factors has become a promising, broadly applicable strategy in drug
discovery and in biology, because a limited number of oncogenic transcription factors, as
compared to signaling kinases are involved in progression of certain disease states, such
as cancer, which elects transcription factors as cogent targets. A second broad family of
targets is cell surface receptors which, in addition to transcription factors, either
independently or in collaboration with the growth factor receptors, modulate downstream
signals. The work described in this Dissertation is focused on chemical strategies in
targeting oncogenic transcription factors and cell adhesion receptors involved in cancer
progression, as a prerequisite for the development of novel anticancer therapeutics.
Chronic hypoxia is a hallmark of solid tumors and is associated with
aggressiveness and rapid progression of the disease. In tumors under hypoxia, a specific
microenvironment is created, which is very different from that of normal tissues. Under
these conditions, activation and stabilization of the α subunit of a transcription factor,
termed hypoxia-inducible factor 1 (HIF-1α) is achieved, and its interaction with the
coactivator p300/CBP elevates the expression of a number of genes involved in
angiogenesis, invasion, altered energy metabolism and other proliferative mechanisms
xxii
that promote tumor growth. The abundance of the expressed HIF-1α in most solid
tumors makes it an attractive therapeutic target. We report the two complementary
strategies in targeting of the interface of transcription factor HIF-1α and its coactivator
p300/CBP: allosteric and orthosteric. In Chapter 2 of this dissertation, an allosteric
approach for targeting of HIF-1α is described, which is accomplished through a novel
class of an epipolythiodiketopiperazine (ETP) transcriptional antagonists that disrupt
hypoxia-inducible transcription. In Chapter 3, we describe the design of orthosteric
transcriptional antagonists termed hydrogen bond surrogates (HBSs) - stable α-helices
that mimic key interacting domain of HIF-1α that interacts with p300/CBP coactivator.
Through the HBS approach we explore the structural roles of the two critical α-helices of
HIF-1α C-terminal transactivation domain (C-TAD) that are responsible for the
recognition of p300/CBP. The stabilization of the secondary structure in HBS via an α-
helical motif results in a short, stable α-helix mimics that disrupt the interaction between
the HIF-1α CTAD and coactivator complex, resulting in a downregulation of hypoxia-
inducible genes, such as VEGF, LOX, Glut1, and CXCR4. Through the use of the
orthosteric and allosteric approaches, we demonstrated the potential of designed small
molecules and protein secondary structure mimetics to become new research tools for
disruption of protein-protein interfaces, regulation of transcription and, ultimately, as
leads in the discovery of novel therapeutics.
In cancers, the elevated levels of certain integrin receptors have been linked to
tumor progression and poor overall prognosis for the patients. Of these, the α
v
β
3
-
receptors are overexpressed on activated endothelial cells undergoing angiogenesis as
well as on proliferating tumor cells, but show only low levels of expression on quiescent
xxiii
cells. This makes α
v
β
3
integrin an attractive molecular target for both diagnosis and
anticancer therapy. In the Chapter 4 of this Dissertation, we report the design and
biological evaluation of α
v
β
3
specific conjugates comprised of high affinity, α
v
β
3
-
selective integrin ligand, a carborane moiety, and a fluorescent probe that exhibits
fluorescence enhancement upon binding to its target. Our in vitro as well as in vivo
studies indicate selectivity toward tumorigenic cells and active uptake of the designed
bifunctional and trifunctional ligands.
Monoamine oxidase A (MAOA) is a mitochondrial membrane-bound enzyme that
catalyzes the oxidative deamination of dietary amines to corresponding aldehydes. This
process produces H
2
O
2
, a major source of reactive oxygen species, which predisposes
cancer cells to DNA damage and promotes tumor initiation and progression. It was
demonstrated that knock-out and pharmacological inhibition of MAOA in prostate cancer
cells slowed down cancer progression. In the Chapter 5 of this Dissertation, we explore
the connection between the activity of MAOA, the levels of H
2
O
2
and its connection with
the elevated levels of HIF-1  and the expression of its downstream genes. Clorgyline, a
selective MAO A inhibitor used as an anti-depressant, and its conjugate NIR-clorgyline,
significantly reduced tumor growth in animal tumor xenograft models. The presence of
the NIR dye moiety renders selective uptake of NIR-clorgyline conjugate to cancer cells
and increases its in vivo efficacy. The preliminary results from this study suggest that
MAOA could become a novel target obligatory to human prostate cancer growth and
metastasis. Through the induction of H
2
O
2
,

MAOA could elevate the levels of HIF-1α
resulting in a rapid vascularization, invasiveness of the tumor and a poor prognosis for
prostate cancer patients. Targeting MAOA in prostate cancer holds the promise of
xxiv
achieving combination therapeutic effects by its synergistic inhibition of multiple
downstream tumor growth promoting factors (e.g. H
2
O
2
and HIF-1α), which could help to
more effectively treat prostate cancer and to overcome the accompanying chemo-
resistance.
1

Chapter 1: Introduction

2
PART I: Small Molecules as Transcriptional Modulators
1.1 Transcription Factors as Therapeutic Targets
Transcription is the process of creating a complementary RNA copy of a sequence of
DNA and is the first step in a complex sequence of events during the process of gene
expression. Aberrant gene expression has been implicated in the onset of many diseases,
such as cancer, inflammatory diseases and heart diseases (Engelkamp and vanHeyningen
1996; Latchman 1996). Modulation of transcription factor activity plays a crucial role in
controlling the tumor-specific changes (genetic and epigenetic), which are responsible
either for repressing tumor suppressor genes, or for activating oncogenes. Tumor-specific
and disease-specific deregulation of transcriptional machinery, responsible for disrupting
cell signaling pathways leads to a number of genetic disorders. Therefore, the ability to
regulate transcription in order to reverse disease-causing changes in gene expression
makes transcription factors (TFs) as important therapeutic targets. Transcription is the
first step of the information transfer from gene to a protein, and hence the ability to
selectively modulate transcription provides a powerful means to control signal
transduction pathways (Figure 1.1). Transcription factors involved in cancer are divided
into three broad groups: 1) steroid hormone receptors (estrogen receptor in breast cancer
and androgen receptor in prostate cancer) act as transcription factors upon activation by
ligand, translocate to nucleus and control gene expression levels by binding to a specific
DNA sequence (Shirinsky and Shirinsky 2011); 2) resident nuclear proteins which are
activated by serine kinase cascade and reside in the nucleus independent of activation
state, and 3) latent cytoplasmic factors, activation of which is triggered by the receptor-
3
ligand interaction at the cell surface and then eventually they translocate into nucleus
(Darnell 2002).
The critical role of transcription factors in the regulation of gene expression
makes them both promising and desirable targets in cancer therapy. A number of
oncogenes and tumor suppressor genes encode TFs, and any functional irregularity, such
as mutation in one of the domains of TF, leads to the aberrant expression of genes
associated with tumor development and progression (Dolores Delgado and León 2006) .

Figure 1.1 The central dogma of molecular biology.
TFs are implicated in a complex web of interactions with their multiple partner
proteins and DNA, through which they regulate gene expression. However, modulation
of protein-protein or protein-DNA interfaces of transcription factors by small molecules
is challenging, as most of the protein-protein or protein-DNA interfaces consist of large
shallow surfaces lacking well-defined structural features, such as clefts or pockets
4
(Jochim and Arora 2009). Nuclear localization of transcription factors further
complicates targeting of TF activity. For a long time, TFs were considered as
undruggable targets; however today the fields of chemical biology and genetics and a
general understanding of transcription process have evolved to a point where this task has
become tractable.
In signal transduction, transcription factors lie downstream of many protein
kinases and activated cell surface receptors. Therefore, targeting kinases or cell surface
receptors could be an effective way of modulating the transcription factor (Bill, Nicholas
et al. 2012). Most of the activated TFs regulate the gene expression levels by binding to
DNA as a dimer (homodimer or heterodimer) and this feature of TFs was also explored
towards the development of inhibitors that can specifically inhibit dimer formation. For
example, STAT3 dimerization,Myc-max dimerization (Berg, Cohen et al. 2002;
Siddiquee, Zhang et al. 2007; Brennan, Donev et al. 2008). Taking advantage of the
sequence selectivity by which DNA binds to the transcription factor, several nucleic acid-
based strategies such as, antisense, triple helix-forming oligonucleotides, and peptide
nucleic acids (PNAs) were explored for the regulation of transcription and translation
(Braasch and Corey 2002; Opalinska and Gewirtz 2002). However, the use of
oligonucleotides is often limited by their poor cell permeability. siRNA and shRNA also
have potential to regulate transcription, however immunological effects, such as off-
target binding to toll-like receptors, restrict their use as drugs in the clinic (Schlee,
Hornug et al. 2006). Sequence-specific DNA minor groove binding polyamides target
predetermined sequence of DNA and have the ability to regulate transcription (Olenyuk,
Zhang et al. 2004), however their uptake and pharmacokinetics are limited by their
5
composition, and in cell-based experiments they target rather short DNA sequences (~6-8
bp). As a result, the problem of selectivity in the context of the entire genome remains to
be solved.
In order to activate gene expression, a transcription factor needs to translocate
into the nucleus where it interacts with a diverse family of coactivators, leads to
chromatin remodeling and, eventually, upregulation of transcription and is responsible for
tumor growth and progression. Therefore, interaction of TFs with the coactivator proteins
provides another potential interface for targeting of a signaling pathway. Chronic hypoxia
is a hallmark of almost all solid tumors which is associated with aggressiveness and rapid
progression of cancers and certain ischemic diseases. Because of the rapid proliferation
of cells, the cells which are farthest from the capillary become hypoxic and this low
partial pressure of oxygen in tissues create a tumor-specific microenvironment which is
very different from that of the normal tissues (Ruan, Song et al. 2009). The adaptation to
hypoxic environment drives cancer cells towards certain phenotypes that confer a
significant growth advantage to these cells and promotes the invasiveness and metastasis
of a tumor (Hanahan and Weinberg 2000). The major underlying mechanism that
mediates the hypoxic response lies in the regulation of transcription by hypoxia inducible
factor-1 (HIF-1) (Semenza and Wang 1992; Semenza 2007).

1.2 The role of hypoxia in cancer biology
Oxygen homeostasis is essential for all metazoans. Every individual cell can sense the
limited oxygen supply and accommodates itself by altering the levels of expression of
hypoxia-inducible genes. Most of the solid tumors could become hypoxic when oxygen is
6
able to diffuse only to a certain distance, 100-120 µm, from the capillary and thus, the
cells which are further away will not be exposed to the same oxygen levels (Tredan,
Galmarini et al. 2007). Hypoxia is implicated in many aspects of tumor biology which
includes genomic instability, suppression of apoptosis, altered metabolism, angiogenesis,
invasion, metastasis, and also confers resistance to conventional treatments, such as
radiotherapy and chemotherapy (Bristow and Hill 2008).
Gradient in cell proliferation rate, regions of hypoxia, anoxia and acidity
contribute towards the heterogeneity of the tumor microenvironment and typically select
for cells with more aggressive phenotype (Shay and Simon 2012). Hypoxic cells are
known to become chemoresistant through several different mechanisms, for example, for
any anticancer drug to reach at the site of action in lethal dose, it has to be distributed
throughout the tumor vasculature, cross the vessel walls and then traverse the tumor
tissue. However, because of the tumor heterogeneity and abnormal tumor vasculature
(disorganized), only very small amount of the drug is able to reach at the site of action. It
is also reported that proper oxygen concentration is required for the action of many
chemotherapeutic agents which works by generating superoxide and thus contributes to
cytotoxicity (Tredan, Galmarini et al. 2007). Under hypoxic conditions, increased
production of nucleophile like glutathione, may compete with the target DNA for
alkylation and thus reduces the efficacy of cytotoxic agents (Vaupel, Thews et al. 2001;
Brown 2007). Most of the chemotherapeutic drugs exert their effects on rapidly
proliferating cells and thereby showed reduced efficacy under hypoxia because of slow
cell growth cycle. Unfortunately, most of the chemotherapeutic drugs also target normal,
non-tumorigenic cells underscoring the critical need of a phenotype-specific anticancer
7
therapy. The mode of action of certain chemotherapeutic drugs also relies directly on the
molecular oxygen concentration, for example, Bleomycin A2, which first chelates with
the metal ion and then reacts with molecular oxygen to produce superoxide and
hydroxide radical that cleave the DNA strand. In the absence of oxygen, Bleomycin A2
efficacy is reduced because of the reduced generation of free radicals (Cunningham,
Ringrose et al. 1984). Hypoxic condition in tumor cells is also reported to have
developed indirect resistance towards chemotherapeutic agents. It has been reported that
hypoxia increases the production of metallothioneins, which have high affinity towards
various heavy metals and is considered as one of the reasons for cisplatin resistance
(Murphy, Laderoute et al. 1994). Because of the increased genomic instability, hypoxia
selects for cells which are either p53 deficient or have p53 mutations and become
resistant to p53 mediated apoptosis (Graeber, Peterson et al. 1994). Low oxygen
concentration in tumor also activates certain genes which are involved in angiogenesis,
cell proliferation, metastasis, metabolism and others (Bunn and Poyton 1996; Cameron,
Harding et al. 2008). Hypoxia is also reported to amplify the expression of genes
encoding P-glycoproteins and dihydrofolate reductase and further conferring resistance
towards the chemotherapeutic agents (Rice, Hoy et al. 1986; Comerford, Wallace et al.
2002).
It was realized in 1950s that the effect of ionizing radiation diminishes in the
absence of oxygen and that is why hypoxic tumor cells become resistant towards
radiotherapy. The primary mechanism by which ionizing radiations kills the cell is by
inducing ionization at the genomic DNA site and producing free radicals. Free radicals of
DNA then bind with molecular oxygen (oxygen has high affinity for these radicals) and
8
cause the permanent damage to the DNA strand. On the other hand, in the absence of
oxygen and under reducing environment, DNA can be reduced to its normal state either
due to lack of generation of free radical or by accepting a hydrogen atom from sulfhydryl
group of glutathione and thus the damage wouldn’t be severe (Gray, Conger et al. 1953).

1.3 Physiological and pathophysiological roles of HIF-1α
The delicate balance of oxygen concentration is disrupted in a number of pathological
conditions and is primarily mediated by hypoxia inducible factor-1 (HIF-1). HIF-1 is a
heterodimeric transcription factor and is involved in key physiological and
pathophysiological processes. Among other members of HIF-1, HIF-2α which is also
termed as HIF related factor or HRF, has an extensive sequence similarity with HIF-1α
(Ema, Taya et al. 1997) and also has similar biochemical properties, like
heterodimerization with HIF-1β (Lau, Tian et al. 2007). In comparison to the
ubiquitously expressed HIF-1α, HIF-2α expression is tissue-specific (Tian, McKnight et
al. 1997). Another member of this family is HIF-3α, also called as Inhibitory-Per-ARNT-
Sim (IPAS), interacts with the amino terminal of HIF-1α and act as a dominant negative
regulator of HIF-1α (Makino, Cao et al. 2001). HIF-3α (IPAS) dimerizes with HIF-1β
and sequesters dimerization of HIF-1β with HIF-1α and thus prevents the transcriptional
activation of HIF-pathway. HIF-1 is implicated in essential developmental processes and
also in different disease states, such as cancer, myocardial ischemia, cerebral ischemia
and others (Semenza 2000). Metabolic shift of hypoxic cells towards glycolysis not only
gives cancer cells pro-survival advantage but also drives the transition of embryonic stem
cells (ESC) to epiblast stem cells (EpiSC) during embryonic development (Zhou, Choi et
9
al. 2012). ESC and EpiSC are two pluripotency stages in embryonic development and it
has been shown recently by Zhou et al. that the more mature EpiSC cells have lower
mitochondrial respiration and higher glycolytic rate, implicating the role of HIF-1α in the
metabolism of stem cell populations. Glycolysis is the energetically less favorable
process of ATP generation, but still, it is believed that it would be favorable for ESC to
EPiSC transition. This metabolic shift increases the production of metabolic
intermediates, promotes the cell proliferation rate and the extracellular acidification
exerted by production of lactic acid might help in the implantation of blastocyst into the
uterine wall. The upregulation of HIF-1α is also suggested to have a significant role in
immune response by production of pro-inflammatory cytokines, such as IL-1α, IL-1β,
TNF-α (Peyssonnaux, Datta et al. 2005; Jantsch, Chakravortty et al. 2008; Nizet and
Johnson 2009). HIF-1α knockout embryos showed neural tube defects (Iyer, Kotch et al.
1998; Ryan, Lo et al. 1998) and mice with one mutant HIF-1α allele showed impaired
physiological responses to chronic hypoxia (Yu, Shimoda et al. 1999). HIF-1α is the
master regulator of oxygen homeostasis and plays a critical role in developmental
processes as well as in the progression of diseases (tumorigenesis, ischemia) which is
tightly regulated by a number of specific mitogens, among which vascular endothelial
growth factor (VEGF) and its receptor play a key role. A tumor cannot grow beyond 1-2
mm
3
unless it receives the oxygen supply and nutrients from the newly developed blood
vessels, and hence, VEGF is one of the key growth factors that support the growth of
newly formed vessel (Zhong, De Marzo et al. 1999).
Chronic hypoxia is a hallmark of almost all solid tumors; hypoxic environment
triggers the activation of HIF-1 inducible transcriptional pathway which results in the up-
10
regulation of VEGF and other mitogens that promote tumor growth and progression
(Pugh and Ratcliffe 2003). In certain cardiovascular diseases, such as ischemia, HIF-1
activation up-regulates the expression of VEGF, thereby triggering the protection
mechanism that leads to an increase in the blood flow. Under these conditions, HIF-1
activation plays an important role in improving myocardial function (Loor and
Schumacker 2008). Mice implanted with ARNT deficient (impaired HIF-1α
dimerization) hepatoma cells (C4-cells) showed reduced growth of tumors in comparison
to the wild type cells (Jiang, Agani et al. 1997; Maxwell, Dachs et al. 1997), tumor
growth in nude mice implanted with HIF-1α deficient embryonic stem cells (Ryan, Lo et
al. 1998) is diminished due to the lack of proper vascularization. HIF-1α targeted with
adenovirus delivered siRNA showed, enhanced hypoxia induced apoptosis in HeLa,
HCT116 cells (Zhang, Kon et al. 2004) and in vivo inhibition of subcutaneous tumor
growth when cells with attenuated HIF-1α expression were used. HIF-1α regulates the
expression of hypoxia-inducible genes involved in angiogenesis, glucose metabolism,
cell proliferation and other critical cellular processes. These genes are also responsible
for the formation of several growth factors (IGF2, TGFβ), anti-apoptotic (BcL-xL) and
angiogenic factors (VEGF). These growth factors activate their respective receptors and
initiate a signaling cascade, which exert overexpression of HIF-1α and also activate the
pro-survival PI3K-Akt pathway (Semenza 2003; Zhang, Gozal et al. 2003).
Overexpression of HIF-1α is also associated with negative prognosis and increased
mortality in cancer patients. Moreover, there are number of pathways intrinsically linked
to HIF-1α that are critical for tumor growth and progression, making it an attractive
therapeutic target.
11

1.4 Regulation of hypoxia-inducible factor signaling
HIF-1 was first discovered by the identification of a hypoxia response element (HRE)
present in the 3’-enhancer region of erythropoietin (EPO), a hormone which stimulates
erythrocyte proliferation (Goldberg, Dunning et al. 1988; Semenza, Nejfelt et al. 1991).
Later, it was found as a transcription factor which binds to HRE as a heterodimer under
hypoxic conditions. HIF-1 is a heterodimeric, DNA-binding, basic helix-loop-helix, per-
arnt-sim (PAS) type transcription factor which consists of an oxygen-sensitive α-subunit
(HIF-1α) and a constitutively expressed β subunit (HIF-1β) (Wang, Jiang et al. 1995).
HIF-1β subunit is also called as aryl hydrocarbon receptor nuclear translocator, ARNT.
Both subunits of HIF-1 belong to the bHLH/PAS (basic helix–loop–helix/Per-Arnt-Sim
homology) family of transcription factors. Similar to HIF-1, ARNT also has three
paralogues, ARNT-1, ARNT-2, ARNT-3 (Lisy and Peet 2008). Both, HIF-1α and HIF-
1β, have one basic-helix-loop-helix (bHLH) domain and two Per-ARNT-Sim (PAS-A,
PAS-B) domains. bHLH and PAS domains are required for the heterodimerization of
HIF-1α and HIF-1β, to make the heterodimer as an active transcription complex (Crews
1998). The aa sequence (1-390) is responsible for the optimal binding of HIF-1α to HRE,
DNA sequence (Jiang, Rue et al. 1996). It has one oxygen dependent degradation domain
which mediates the oxygen-dependent stability of HIF-1 (Pugh, Orourke et al. 1997). The
C-terminal half of HIF-1 contains two transactivation domains, an N-terminal activation
domain (N-TAD) and C-terminal activation domain (C-TAD).
CTAD domain interacts with coactivator p300/CBP under hypoxia to activate the
transcription of hypoxia inducible genes (Lando, Peet et al. 2002). HIF-3α has similar
12
structural makeup as of HIF-1α/2α (bHLH and PAS domains) but it lacks the CTAD
(Figure 1.2). Under well-oxygenated conditions, HIF-1α is hydroxylated at proline
residues 402 and 564, present in the ODDD of HIF-1α by enzymes, proline hydroxylases
(PHD1-3).

Figure 1.2 Structural domains of the HIF transcription factor family.
The activity of proline hydroxylases is tightly regulated by the presence of
oxygen, Fe(II) and 2-oxoglutarate (Ivan, Kondo et al. 2001; Jaakkola, Mole et al. 2001).
Proline hydroxylases catalyze the process of hydroxylation by transferring one oxygen
atom to proline and other to 2-oxoglutarate to form succinate and carbon dioxide (Dann
and Bruick 2005). Von Hippel-Lindau tumor suppressor protein (pVHL) recognizes the
hydroxylated HIF-1α which is then subsequently ubiquitinylated by the complex
comprising of elongin B, elongin C/cullin 2/ring-box 1 ubiquitin ligase and tag HIF-1α
for polyubiquitylation and rapid proteosomal degradation (Maxwell, Wiesener et al.
1999). Under normal oxygen levels (normoxia) the half-life of HIF-1α is less than 1
13
minute because of its rapid degradation (Yu, Frid et al. 1998). It has been shown that
under normoxia, mutation of one of the conserved proline residues causes partial
stabilization of HIF-1α and mutation of both prolines disrupt the interaction between
pVHL and HIF-1α (Masson, Willam et al. 2001). Mutation in pVHL gene results in the
stabilization of HIF-1α under normoxic conditions and activates the expression of HIF-
inducible genes. In addition to pVHL-dependent pathway, pVHL-independent pathways
are also reported that can cause degradation of HIF-1α and are discussed in the next
section.
The lysine residue (K532) present in the ODDD domain is acetylated by arrest-
defective-1 (ARD1) enzyme and is reported to be responsible for the degradation of HIF-
1α. Acetylated lysine promotes the interaction of HIF-1α with pVHL and thus
destabilizes HIF-1α (Jeong, Bae et al. 2002). Although the acetyltransferases (ARD1)
activity is not directly influenced by oxygen, and it may acetylate K532 of HIF-1α even
in the absence of oxygen, but the mRNA and protein levels of ARD1 decreases under
hypoxia which results in the reduced acetylation and thus stabilization of HIF-1α.
Under hypoxic conditions, PHD activity is inhibited by the limited supply of
oxygen and is unable to hydroxylate HIF-1α (Figure1.3). As a result, pVHL is no longer
recognizes HIF-1α for proteosomal degradation, resulting in an accumulation of HIF-1α
(Semenza 1998). Upon stabilization, HIF-1α translocates into the nucleus, where it
dimerizes with its β subunit (ARNT) and binds to a consensus DNA sequence,
5’G/ACGTG 3’, called as hypoxia response element (HRE) present in the promoter
region of its target genes (Semenza, Jiang et al. 1996). To date, more than 100 hypoxia-
inducible genes that are direct targets of HIF-1α (Manalo, Rowan et al. 2005) have been
14
identified, with many of them having crucial role in tumor progression such as
angiogenesis, glucose metabolism, metastasis and others (Semenza 2002). Eventually, the
heterodimer complex recruits the coactivator protein p300/CBP and activates the
transcription of hypoxia inducible genes (Semenza 2003) like VEGF, Glut1, LOX, etc. As
a complementary mechanism, transcriptional activity of HIF-1α is also controlled by
another oxygen, Fe (II) and 2-oxoglutarate dependent enzyme, Factor Inhibiting HIF-1,
FIH-1.

Figure 1.3 Regulation of HIF-1α signaling pathway.
FIH-1 is an asparagine hydroxylase that serves as a second checkpoint before HIF
can induce the transcription of hypoxia inducible genes (Mahon, Hirota et al. 2001;
Lando, Peet et al. 2002). FIH-1 hydroxylates the asparagine residue (N803) of HIF-1α
under normal oxygen concentration and inhibits the recruitment of transcriptional
15
coactivator, p300-CBP. Hypoxia leads to the reduced activity of both PHD and FIH and
thus results in an elevated expression of HIF-1 inducible genes.
1.4.1 Oxygen-independent regulation of hypoxia-inducible factor signaling
As discussed earlier, pVHL dependent pathway is not the only pathway responsible for
the HIF-1α degradation. Murine double minute-2 (mdm-2) can cause
polyubiquitinylation of HIF-1α and degrades it in a p53-dependent manner.

Figure 1.4 Oxygen-independent regulation of hypoxia-inducible factor signaling.
It has been shown that p53 inhibits the activity of HIF-1 by targeting HIF-1α for
proteasomal degradation. Also, p53 deficient cells have shown higher levels of HIF-1α
and overexpression of HIF-inducible genes. Recently it has been shown that p53
regulated microRNA-107 (miR-107) inhibits hypoxic signaling by controlling the
expression of HIF-1β (Yamakuchi, Lotterman et al. 2010). HIF-1α levels also increase in
16
response to the growth factor stimulation. Activation of PI3K-Akt pathway promotes the
nuclear accumulation of HIF-1α and thus mimics the hypoxia conditions for the
transactivation of hypoxia inducible genes (Zhong, Chiles et al. 2000). It has been shown
that mammalian target of rapamycin (mTOR) increases the translation of HIF-1α by
phosphorylating the eukaryotic initiation factor-4E (elF-4E) (Hudson, Liu et al. 2002). In
hepatocellular carcinoma cells (HCC), PI3K inhibitor (LY294002) is able to inhibit the
translation of HIF-1α which implies the role of PI3K-Akt activation in regulating the
HIF-pathway (Jiao and Nan 2012). In contrast to hypoxia, stimulation of HIF pathway by
activation of PI3K-Akt signaling seems to be cell type specific (Shafee, Kaluz et al.
2009). HSP90, a heat shock chaperone protein, also associates with HIF-1α and induces
some conformational changes that promote its dimerization with HIF-1β (Figure 1.4).
Geldanamycin, a HSP90 inhibitor, has been shown to induce the degradation of HIF-1α
even in cell lines lacking functional pVHL (renal carcinoma cell line) and also showed
decreased transcriptional activity by downregulating the levels of VEGF. Disruption of
HSP90 induces proteasomal degradation of HIF-1α by an oxygen independent
mechanism (Isaacs, Jung et al. 2002). Ras-Raf pathway is also known to be regulating the
HIF pathway at both transcriptional and translational levels. The downstream effector of
this pathway, ERK, phosphorylates the coactivator protein (p300) and transactivates the
expression of hypoxia inducible genes (Hur, Chang et al. 2001). Intracellular secondary
messengers, calcium ion (Hui, Bauer et al. 2006; Jung, Kim et al. 2010) and reactive
oxygen species (Lopez-Lazaro 2006) are also known to promote the synthesis of HIF-1α
protein and thus regulate the transcriptional activity of HIF-1.

17
1.4.2 Therapeutic intervention of hypoxia-inducible factor pathway
Hypoxia renders cancer cells resistant toward a range of chemotherapeutic agents and
also allows them to acquire aggressive phenotype which enhances their invasive and
metastatic potential. Central role of HIF in multiple pathways, responsible for the tumor
growth and progression, makes it an interesting and desirable therapeutic target for the
development of antineoplastic agents. Efforts have been made in the past to develop
inhibitors that act at different checkpoints of HIF pathway. Several natural product
inhibitors affecting HIF-1α protein levels have been identified from the activity-guided
fractionation study. Moracin O and moracin P isolated from Morus species have shown to
inhibit the accumulation of HIF-1α under hypoxia in a dose dependent manner (Xia, Jin
et al. 2011).

Figure 1.5 Natural product inhibitors of HIF-1 pathway.
18
Isolated from Saururus cernuus, Manassantin A, showed decrease in the VEGF secretion
and downregulation of HIF-1α levels in 4T1 (mammary carcinoma) cells (Kasper, Moon
et al. 2009). Another natural product curcumin, a component of commonly used spice,
has been shown to inhibit the HIF activity (Figure 1.5) and the expression of HIF
inducible genes by proteasomal degradation of HIF-1β (Bae, Kim et al. 2006; Choi, Chun
et al. 2006). In addition to natural product based inhibitors, small molecule inhibitors
were also studied that can affect the HIF-1α protein levels. PI3K inhibitor, wortmannin
and LY294002, can downregulate the synthesis of HIF-1α (Jiang, Jiang et al. 2001).

Figure 1.6 Small molecule inhibitors of HIF-1 pathway: wortmannin, a PI3K inhibitor;
YC-1, prevents recruitment of p300; glendamycin: a HSP90 inhibitor.

mTOR inhibitors, Rad001 and CCI-779, have also been shown to attenuate the
HIF-1α protein synthesis (Majumder, Febbo et al. 2004; Wan, Shen et al. 2006) SIRT1, a
histone deacetylase (HDAC), binds to HIF-1α and deacetylates it at K674, which inhibits
the interaction of the coactivator protein p300/CBP with HIF-1α. Under hypoxia, SIRT1
was downregulated and activates the transcription of HIF-inducible genes (Lim, Lee et al.
2010). HDAC inhibitor FK228 inhibits transcriptional activity of HIF-1 (Lee, Kim et al.
19
2003). YC-1, is also known to prevent the recruitment of p300 and rescues HIF-1α
mediated transcription (Chun, Yeo et al. 2001; Li, Shin et al. 2008). HSP90 inhibitor,
Glendamycin induces ubiquitinylation and proteasomal degradation (Figure 1.6) of HIF-
1α (Mabjeesh, Post et al. 2002). Sequence-specific DNA binding polyamides have been
designed by creating a code of aromatic rings, for example, N-methylpyrrole: N-
methylhydroxypyrrole for A: T recognition and N-methylpyrrole : N-methyimidazole for
C:G recognition. These sequences bind to the minor groove of DNA and inhibit the
binding of HIF-1α/HIF-1β heterodimer to HRE and thus regulate HIF-1 transcription
(Olenyuk, Zhang et al. 2004).
Cell based reporter assays have also identified a few small molecule inhibitors of
HIF pathway. Camptothecin derivatives and topotecan (topoisomerase inhibitor) are
known to inhibit the translation of HIF-1α (Rapisarda, Uranchimeg et al. 2004; Lou,
Chua et al. 2010). Topotecan is currently used as a chemotherapeutic agent in the
treatment of small cell lung cancer and ovarian cancer (Ardizzoni 2004). Inspite of the
difficulties of targeting the shallow surfaces of transcription factors, sufficient efforts
have been made to develop small molecule inhibitors that can selectively target the HIF
pathway. In 2004, Kung and coworkers identified chetomin, an epidithiodiketopiperazine
based small molecule, from high-throughput screening as a transcriptional regulator of
HIF pathway (Kung, Zabludoff et al. 2004). A proteasome inhibitor, bortezomib, impairs
the p300-HIF-1α interaction by promoting the interaction of FIH to HIF-1α. The
interaction of the α-subunit of HIF-1 with different protein partners in the HIF-1
signaling pathway can become a point of intervention by the tools of chemical genetics.
These include: 1) interaction of HIF-1α with pVHL, 2) interaction of HIF-1α with its β-
20
subunit (ARNT, heterodimerization), 3) interaction of HIF-1α with p300/CBP
coactivators, and 4) interaction of HIF-1α with FIH. First part of this thesis is focused on
using a chemical biology approach to design highly specific small molecules and protein
ligands that can regulate the hypoxia inducible transcription pathway. The main
hypothesis of the research reported in this Dissertation is that the rationally designed
small molecules and low molecular weight α-helix mimetics could serve as inhibitors of
the critical protein-protein interfaces involved in the regulation of HIF-1
transcriptional activity.

1.5 Introduction to epipolythiodiketopiperazines (ETPs)
Epipolythiodiketopiperazines are an interesting class of structurally diverse biologically
active fungal secondary metabolites (Hino and Nakagawa 1989; Wrobel and
Wojtasiewicz 1992; Rezanka, Sobotka et al. 2006). These metabolites are characterized
by the presence of diketopiperazine rings and sensitive epi-polysulfide
bridges. Biosynthetic pathways of ETP production are not very well characterized; still
with the isolation of certain key intermediates, its synthesis can be accomplished.
Labeling and feeding experiments indicated amino acids, phenylalanine and serine as
precursors of ETP core in gliotoxin (Suhadolnik and Chenoweth 1958; Winstead and
Suhadolnik 1960). Recently, a P450 monooxygenase, GliC, and a specialized glutathione
transferase, GliG, have been shown to play a critical role in the formation of C-S bond
formation in gliotoxin (Scharf, Remme et al. 2011). In another paper from the same
group, GliT gene, a pyridine dinucleotide dependent oxidoreductase, is found to be
involved in the formation of disulfide bridge in gliotoxin (Scharf, Remme et al. 2010).
21
ETPs production in fungal species is found to be discontinuous. For example, Aspergillus
fumigatus produces gliotoxin and another closely related species of Aspergillus does not.
Similarly, gliotoxin is produced by Aspergillus fumigatus and also by unrelated species
Trichoderma virens and Candida albicans (Weindling and Emerson 1936; Shah and
Larsen 1991).
1.5.1 Biological activities of epipolythiodiketopiperazines
Because of structural diversity and presence of internal di or polysulfide bridges, ETPs
have shown broad range of biological activities including antibacterial, antitumor,
antimicrobial, antiviral, apoptotic and antiallergic (Waring and Beaver 1996). Given the
interesting biological properties of these compounds (ETPs) and paucity of their
quantities acquired from natural products, chemical synthesis would facilitate the process
of biological evaluation of ETPs. The presence of structural complexity and relatively
labile nature of ETP ring renders their synthesis difficult, and hence only a few successful
syntheses of ETPs were reported over the past few decades.
Gliotoxin, the first well-characterized mono-ETP, was isolated in 1936 from
Gliocladium fimbriatum and is shown to have antiviral (McDougal.Jk 1969), antifungal
and antibacterial (Li, Kim et al. 2006) properties. It also possesses profound
immunosuppressive effects in vivo. Mice treated with intraperitoneal injection of
gliotoxin showed delay in the recovery of immune cells (Pahl, Krauss et al. 1996). The
first total synthesis of (±) gliotoxin was reported by Fukuyama and Kishi in 1976. Their
strategy was based on the three key steps: 1) installing thioacetal moiety at an early stage
of synthesis 2) functionalization of C-3 and C-6 carbon of ETP ring and 3) regeneration
of disulfide bridge at the end of synthesis (Fukuyama and Kishi 1976). Sporidesmin A, a
four-ring fused ETP metabolite was isolated from Pithomyces chartarum (Wróbel 1985)
22
and has shown antibacterial, antineoplastic and antiviral activities (Jordan and Cordiner
1987; Waring and Beaver 1996). Acetylaranotin isolated from Aspergillus terreus
(Nagaraja.R, Neuss et al. 1968; Neuss, Boeck et al. 1968), MPC1001B isolated from
Cladorrhinum sp. KY4922 (Onodera, Hasegawa et al. 2004; Tsumagari, Nakai et al.
2004) and emethallicin A isolated from Aspergillus heterothallicus (Kawahara, Nakajima
et al. 1989) belong to one of the structural subgroup of ETPs which contains a seven
membered dihydrooxepine ring (Figure1.7).

Figure 1.7 Biologically active monoETP compounds.

All these compounds exhibited the antiproliferative and apoptotic properties in various
cancer cell lines. An epoxide containing mono-ETP, scabrosin ester isolated from
23
Xanthoparmelia scabrosa (Ernst-Russell, Chai et al. 1999) has shown anticancer activitiy
with IC
50
value of 0.5 µM against P815 mastocytoma cancer cell line, which is approx. 6
times effective as compare to gliotoxin (Ernst-Russell, Chai et al. 1999; Chai, Elix et al.
2004). Apart from mono-ETPs, several examples of bis-ETP molecules are reported in
the past with antitumor activities. For example, 11,11’-Dideoxyverticillin, a symmetric
bis-ETP alkaloid, was first isolated from a marine derived fungus of Penicillium sp. (Son,
Jensen et al. 1999). This ETP has shown cytotoxic, antiangiogenic and tyrosine kinase-
inhibitory activities (Figure 1.8).

Figure 1.8 Biologically active bis-ETP compounds.

24
Bis-ETP, 11,11’-dideoxyverticillin, inhibited the proliferation of human
umbilical vein endothelial cells (HUVECs) and antagonized the antiapoptotic effects of
VEGF on serum-deprived HUVECs. Upon treatment with 11,11’-Dideoxyverticillin,
decreased levels of VEGF secretion in breast cancer cell line and considerable
suppression of VEGF-induced tyrosine phosphorylation was observed in HUVECs
(Chen, Zhang et al. 2005).
Despite of the isolation of ETP molecules several decades ago, the first total
synthesis of a dimeric ETP was achieved by Movassaghi et al. in 2009, highlighting the
complexity involved in the synthesis of these molecules (Kim, Ashenhurst et al. 2009).
Chaetocin A, also a bis-ETP molecule, was isolated from Chaetonium minutum (Sekita,
Yoshihira et al. 1981). It is reported to have antibacterial, antiangiogenic, cytostatic
properties and also acts as a potent inhibitor of lysine-specific histone methyltransferases
(Iwasa, Hamashima et al. 2010). Chetomin, leptosin, verticillin B are some examples of
biologically relevant asymmetric bis-ETP molecules whose total synthesis is not reported
to date. Chetomin has shown the cytotoxic and antiangiogenic properties (Kung,
Zabludoff et al. 2004). Leptosin (F and C) and verticillin A have shown cytotoxic and
antibacterial properties (Katagiri, Sato et al. 1970; Yanagihara, Sasaki-Takahashi et al.
2005; Liu, Liu et al. 2011).
1.5.2 Mode of action of epipolythiodiketopiperazines
Several studies suggest that biological activity of ETPs is due to the presence of the
disulfide bridge, because removal of the sulfurs or presence of reducing agents
completely abrogates their activity (Chai and Waring 2000; Cook, Hilton et al. 2009). It
is believed that ETPs can exert their effect by three different ways:

25
1) The formation of a mixed disulfide between cysteine residues and ETPs: ETPs can
form covalent complex with the cysteine residues present in the protein and thus make it
nonfunctional. For example, gliotoxin forms a covalent complex with the cysteine
residues of alcohol dehydrogenase (Waring, Sjaarda et al. 1995). Another example is
transcription factor NF-κB. It is believed that thiol residue of NF-κB may interact with
the disulfides of gliotoxin which would be responsible for the immunosuppressive effects
of gliotoxin (Pahl, Krauss et al. 1996) as NF-κB controls the expression of many
cytokines. It can also catalyze the formation of internal disulfide bond between the two
physical close cysteine residues, e.g., creatine kinase. Reaction of ETPs with thiol residue
of plasma membrane calcium channel suggested that calcium influx can also be one of
the potential mechanisms of oxidative damage responsible for the necrosis of thymocytes
in the presence of gliotoxin (Hurne, Chai et al. 2002).
2) The generation of superoxide anion radical by redox cycling: The presence of
glutathione (GSH) in cells can reduce ETPs to the free thiol form. This reduced form of
ETP is reactive towards the molecular oxygen and generates superoxide anion radical
(O
2
•˗
), which further produces the reactive oxygen species (ROS), such as hydroxyl
radical or hydrogen peroxide (Munday 1982; Bernardo, Brasch et al. 2003). These ROS
species are the source of damage to many biomolecules, such as DNA or proteins
(Figure 1.9 A).
3) Ejection of the metal ions from the active site of proteins: Sulfurs can make stable
complexes with transition metal ions, for example, zinc, cadmium and mercury
complexes with sporidesmin (Woodcock, Henderson et al. 2001; Cook, Hilton et al.
2009). This particular reactivity of sulfurs suggests that disulfides of ETPs can form
26
stable complexes with the metal ions present in the active site and disrupt the global fold
of the protein (Figure 1.9 B).

Figure 1.9 Proposed mechanism of action of ETPs. A) Redox cycling between the
reduced and oxidized form of ETPs (Gliotoxin) B) Zinc ejection from protein active site
by ETPs (Cook, Hilton et al. 2009).

1.6 Chetomin: A modulator of the HIF transcriptional pathway
Chetomin (Figure1.8) was originally isolated in 1944 (Geiger, Conn et al. 1944) from a
fungus Chaetomium cochliodes and showed a potent antibacterial activity towards gram-
positive bacteria. Chemical and spectroscopic studies on chetomin structure revealed the
presence of two epidithiodiketopiperazine rings (McInnes, Taylor et al. 1976).
27
In 2004, Kung and coworkers have identified chetomin as a transcriptional
modulator of the HIF pathway by performing a high-throughput screening (HTS) on a
natural and synthetic compound library that consisted of more than 600,000 compounds
(Kung, Zabludoff et al. 2004). After HIF-1α translocation to nucleus under hypoxia, the
major interaction responsible for the activation of hypoxia inducible genes is between the
CTAD domain of HIF-1α (786-826) and CH1 domain of p300 (302-423) (Kung, Wang et
al. 2000). A time-resolved fluorescence assay identified chetomin as an inhibitor of HIF-
1α/p300 interaction at submicromolar concentrations. In a cell-based luciferase reporter
system, chetomin showed dose-dependent attenuation of hypoxia inducible genes such as,
VEGF, EPO and Glut1. Further, systemic administration of chetomin in mice inhibited
hypoxia-inducible transcription within tumors and also inhibited tumor growth. Although
mechanism by which chetomin disrupts the interaction between HIF-1α and p300 is
unclear, still is certain evidence that it disrupts the global fold of the CH1 domain of
p300. For example, NMR analysis of CH1 domain of p300 in the presence of chetomin
showed loss of proton peak as well as broadening of spectral line width, suggesting a loss
of the structural integrity of the CH1 domain.

1.7 Introduction to HIF-1α target genes
HIF-1α is the master regulator of transcription of oxygen dependent genes. More than 70
genes are reported, which are characterized by the presence of cis-acting hypoxia
response element. These genes regulate a number of cellular processes involved in tumor
angiogenesis, metabolism, metastasis, invasion and other critical processes of tumor
growth and progression.
28
1.7.1 The role of VEGF in tumor progression
Vascular endothelial growth factor (VEGF) is the most important component of the
vascular system responsible for both, vasculogenesis and angiogenesis. Vasculogenesis is
the de novo formation of blood vessels from hematopoietic precursor cells and
angiogenesis is the formation (sprouting) of the new blood vessels from the pre-existing
vasculature (Hoeben, Landuyt et al. 2004). In higher organisms, there are five distinct
members of VEGF family (Holmes and Zachary 2005), VEGFA, VEGFB, VEGFC,
VEGFD and PLGF (placenta growth factor). These ligands occur in different splice
variants and they bind with mainly three receptor tyrosine kinases, VEGFR1, VEGFR2
and VEGFR3 (Olsson, Dimberg et al. 2006).
VEGFA, a 46 kDa dimeric glycoprotein is transcribed from a single gene and
exists in nine isoforms with different biological activities (Tischer, Mitchell et al. 1991).
The most predominant form of VEGFA is VEGFA-165 (number corresponds to the
amino acids) followed by VEGFA-189 and VEGFA-121. Other major isoforms are
VEGFA-206 and VEGFA-145. VEGFA is a potent mitogen and survival factor for
vascular endothelial cell. VEGFB exists in two isoforms, VEGFB-167 and VEGFB-186.
VEGFC and VEGFD both are synthesized as prepro-protein and bind to VEGFR3 and
are critical in lymphogenesis (Olofsson, Pajusola et al. 1996; Robinson and Stringer
2001).
VEGFA binds to VEGFR1 homodimer, VEGFR2 homodimer and with VEGFR1-
VEGFR2 heterodimer. VEGFB and PLGF bind to VEGFR1 homodimer (Figure 1.10).
VEGFC and VEGFD bind to VEGFR2 homodimer, VEGFR3 homodimer and VEGFR2-
VEGFR3 heterodimer (Ferrara and DavisSmyth 1997; Olofsson, Korpelainen et al.
1998).
29

Figure 1.10 The family of VEGF receptors and ligands.

VEGF receptors have an extracellular domain (organized into seven immunoglobin-
like folds), a transmembrane domain, a juxtamembrane domain, a split tyrosine kinase
domain and a small C-terminal tail (Roskoski 2008). Binding of ligand initiates the
dimerization of receptors and activates the tyrosine activity which eventually leads to
autophosphorylation of the receptors. The phosphorylated receptors activate other
proteins and initiate a signaling cascade (Figure 1.11).
Several knockout studies revealed the importance of VEGF/VEGFR family in
cardiovascular, haematopoietic and lymphatic system development. VEGFA
-/-
in mice
was embryonic lethal, E9.5-10.5 with severe defects in vascular development (Carmeliet,
Ferreira et al. 1996). Single allele inactivation of VEGFA
+/-
is also embryonically lethal
30
with deficient endothelial cell development and less developed vasculature (Ferrara,
CarverMoore et al. 1996).

Figure 1.11 Schematic representation of VEGF pathway.

VEGFR1
-/-
and VEGF2
-/-
mice were also embryonic lethal with disorganized
vasculature and defective vasculogenesis, respectively. VEGFC
-
/
-
embryos died
prenatally because of edema and lack of lymphatic vessels (Karkkainen, Haiko et al.
2004). The corresponding receptor, VEGF3
-/-
knockout animals were also embryonically
lethal because of cardiovascular failure (Dumont, Jussila et al. 1998). An extremely
critical feature for continuous growth and progression of tumor is the accessibility to
oxygen and nutrients. A tumor cannot grow beyond 1-2 mm
3
in the absence of proper
31
supply of oxygen and thus gaining access to oxygen supply from host vasculature is the
rate limiting step for tumor growth. In order to satisfy these needs, tumor relies on tumor
angiogenesis. A number of pro and anti-angiogenic factors work coherently and induction
of angiogenesis depends on how heavily the balance is in favor of pro-angiogetic factors.
Out of several pro-angiogenic factors known, VEGF (also known as vascular
permeability factor) is the most powerful angiogenic factor that enables a tumor to gain
access to oxygen and nutrients for continuous growth and progression. Abnormal
vasculature (leaky and disorganized) of tumor associated vessels makes angiogenic
switch a critical step towards the metastatic progression of cancer. VEGFA with its
receptor VEGFR2 makes the predominant pair responsible for angiogenesis and its
expression is primarily regulated by a transcription factor HIF-1α, under hypoxic
conditions (Mazure, Brahimi-Horn et al. 2004). Analysis of human VEGF gene promoter
revealed the presence of a single hypoxia response element (Roskoski 2007) from
nucleotide position -975 to -968 (5’-TACGTGGG-3’). Under hypoxic conditions, HIF-
1α recognizes the hypoxia response element on VEGF gene and up-regulates its
expression which is further responsible for increased angiogenesis.
The absolute requirement of tumor progression on angiogenesis makes it an
important target in cancer therapy. Several strategies have been designed to target VEGF,
VEGF receptors or its tyrosine kinase activity. By far, the most successful and clinically
relevant strategy is the use of a VEGF-neutralizing antibody, Bevacizumab (Avastin)
along with chemotherapeutic agents (Los, Roodhart et al. 2007). Targeting HIF-1α,
primary mediator of downstream production of VEGF under hypoxia, seems to be an
32
interesting and effective way of inhibiting angiogenesis and, eventually, tumor
progression.
1.7.2 Role of LOX in tumor progression
LOX (lysyl oxidase) is a 32 kDa, copper-dependent enzyme which catalyzes the
formation of aldehydes from lysine and hydroxylysine present in elastin and collagen,
respectively (Krawetz 1994). It plays a critical role in crosslinking the collagen and
elastin which is essential for the stability of extracellular matrix (Coral, Angayarkanni et
al. 2008). Structural and computational studies on LOX revealed that a single divalent
copper ion is present as a part of octahedral coordination complex co-ordinated with three
histidine nitrogens (Krebs and Krawetz 1993). LOX is crucial for the developmental
processes as LOX knockout mice died perinatally due to a cardiovascular dysfunction
(Maki, Sormunen et al. 2005).
Tumor microenvironment plays an important role in cancer progression and a
major component of this milieu is extracellular matrix (Henning, Kraus et al. 2004).
Extracellular matrix (ECM) has an interesting role in maintaining the proper functioning
of cells beyond just providing the support structures. Multiple mechanisms tightly
regulate the proper functioning of ECM and any disruption in these processes results in
the imbalance of organ homeostasis. The major reason of mortality in cancer patients is
metastasis of tumor to other organs and a number of evidence now linked the
overexpression of LOX (component of ECM) with metastasis (Baker, Cox et al. 2011;
Chen, Tu et al. 2012). It has been observed that in cancers, such as breast cancer, tissues
are quite stiffer than normal tissue and this stiffness can be attributed to the increased
cross-linking due to the overexpression of LOX (Lu, Weaver et al. 2012). Recent studies
in mouse models have shown that overexpression of the LOX leads to increased
33
invasiveness and also, mouse models which were engineered to produce less LOX
showed decreased metastatic potential (Levental, Yu et al. 2009). The ovexpression of
LOX is also implicated in ERK and PI3K signaling leading to tumor progression. In
addition, hypoxic cells showed increased expression of LOX. The gene microarray data
with hypoxic human tumor cells revealed the overexpression of LOX and LOX-like
proteins. LOX promoter also contains the hypoxia response element and its induction
under hypoxia is mediated by transcription factor HIF-1α (Erler and Giaccia 2006;
Higgins, Kimura et al. 2007). Hypoxia induced invasion is shown to be decreased by
using LOX shRNA, antisense oligonucleotides in breast, head and neck and other types
of cancer. Tumor hypoxia associated invasion and metastasis is always considered as
poor prognosis for patients and thus inhibiting hypoxia-induced LOX overexpression by
targeting HIF-1 pathways has a therapeutic potential.
1.7.3 Role of Glut-1 in tumor progression
Glut (solute carrier or glucose transporter) facilitates the transport of glucose across the
plasma membrane. Glut-1 is the first glucose transporter isolated with twelve highly
hydrophobic transmembrane α-helices (Carruthers, DeZutter et al. 2009). Glut-1 deficient
homozygous transgenic mice were lethal during gestation and Glut-1 deficiency seems to
play an important role in embryonic malformations and hyperglycemia (Heilig, Saunders
et al. 2003). Glut-1 was ubiquitously expressed by normal tissues and overexpressed
by tumor tissues (Airley, Loncaster et al. 2001). Overexpression of Glut-1 is reported in
several cancers, such as colorectal cancer, breast cancer and others (Younes, Brown et al.
1995; Chung, Huang et al. 2009). Around 90% of the ATP production in normal cells
relies upon mitochondria and only 10% uses glycolysis. On the other hand, tumor cells
showed increased dependence on glycolysis even in the presence of sufficient oxygen
34
called as aerobic glycolysis (Warburg 1956; Gatenby and Gillies 2004). The altered
metabolism of tumor cells which showed greater dependence on aerobic glycolysis is
supported by increase in glucose transport and glucose consumption (Macheda, Rogers et
al. 2005). Activation of HIF-1α stimulates the glycolytic energy production by increasing
the expression of Glut1 and by increasing the expression of genes involved in the
breakdown of glucose to pyruvate (Kim, Tchernyshyov et al. 2006). HIF also
downregulates the oxidative phosphorylation by activating genes, such as PDK1 and
MXI1 (Zhang, Gao et al. 2007).
1.7.4 The role of CXCR4 in tumor progression
CXCR4 is a G protein-coupled chemokine receptor of type 4 and has been implicated in
immune system and development (Busillo and Benovic 2007). These receptors induce
migration of cells towards a concentration gradient of a cytokine. CXCR4 is an SDF-1
specific chemokine receptor, which plays crucial role in hematopoiesis, vasculogenesis
and immune cell trafficking. Crystal structure of CXCR4 with antagonist revealed the
presence of homodimer with ligand binding site closer to the extracellular surface (Wu,
Chien et al. 2010). CXCR4 is also associated with a number of pathological conditions,
its relevance was first observed as a co-receptor for entry of HIV virus into CD4
+
T cells
(Feng, Broder et al. 1996) and later its role in cancer progression became evident
(Darash-Yahana, Pikarsky et al. 2004; Dubrovska, Elliott et al. 2012). SDF-1 binding to
CXCR4 results in the activation of multiple G- protein dependent signaling pathways
involved in migration, survival and transcriptional activation. CXCR4/SDF-1 axis
promotes tumor metastasis by migrating CXCR4 positive tumor cells to the organs where
stromal cells secrete SDF-1 (Figure 1.12).
35
It is quite evident from several reports that CXCR4 is the major chemokine
receptor expressed by cells and responsible for metastatic characteristic of a tumor, such
as small cell lung cancer (Na, Scheibenbogen et al. 2008). Hypoxia is known to
upregulate the expression of CXCR4 through HIF-1α transcriptional pathway and thus its
inhibition has a potential to diminish metastatic potential of cancer cells (Sun, Wei et al.
2010).

Figure 1.12 Schematic representation of CXCR4 pathway.

36
1.8 Chemical approaches to target HIF pathway
Considering the central role of hypoxia and HIF in tumor growth and progression,
targeting HIF pathway could present a novel approach for cancer therapy. The interaction
between CTAD domain of HIF-1α and CH1 domain of its coactivator protein p300/CBP
is quite critical for the transactivation of hypoxia-inducible genes. My first two projects
are focused on the use of chemical tools to explore their potential in the regulation of
transcription of HIF pathway. We have applied two different approaches, allosteric and
orthosteric, to modulate the hypoxia-induced transcription.
1.8.1 Allosteric approach
The contact surface area between HIF-1α and p300/CBP is quite extensive - it spans
approximately 3393 Å and, hence, it is difficult to design small molecules which could
disrupt this interaction. To circumvent this issue, a more effective way would be to
disrupt the global fold of one of the binding partner and thus inhibiting the complex
formation. Even with a great potential of chetomin as a transcriptional modulator of HIF-
1 pathway, it showed coagulative necrosis, anemia, and leukocytosis in experimental
animals. Given the complexity in the structure and unsymmetrical nature of chetomin, no
total synthesis of chetomin is reported yet. Keeping in mind the biological activity of
chetomin, we have designed a series of mono-ETP and bis-ETP inhibitors and studied
their structure-activity relationship (Figure 1.13). Our design is mainly inspired by
leptosin I (Takahashi, Numata et al. 1994) and chetomin. Leptosin I is a mono-ETP
which has a tetrasulfide motif within its south fragment.
1.8.2 Orthosteric approach
Structural details of solution structure of HIF-1α CTAD (786-826) and CH1 domain of
p300 reveals two short α-helical domains from HIF-1α as key determinants for its
37
recognition by p300. Computational analysis revealed that even though protein-protein
interfaces are large, there are certain residues called “hotspots”, whose free energy of
interaction is responsible for the binding affinity of a particular protein complex.

Figure 1.13 Allosteric and orthosteric approach to target HIF-1α and p300 interface.

We reasoned, that stable mimics of these domains can potentially inhibit the
interaction between HIF-1α and p300/CBP and further downregulate the expression of
hypoxia-inducible genes (Figure 1.13). In collaboration with the Arora group at NYU, we
have proposed a novel approach for stabilizing the α-helices present in the HIF-1α
domain and target the transcription factor HIF-1α–coactivator p300/CBP complex
orthosterically.

38
1.9 Screening of a small molecule library
Hypoxia and overexpression of HIF-1 seem to be the general characteristics of most solid
tumors and thus identification of inhibitors of HIF pathway, with novel mechanism of
action may lead to the development of drugs which would be applicable towards solid
malignancies. We have used a cell based luciferase reporter system and screened a library
of small molecules. During initial screening we have identified three potent inhibitors of
HIF pathway. The mechanistic studies of these inhibitors are currently underway and
will determine the critical point of intervention along the HIF pathway.

39
PART II: Targeting α
v
β
3
integrin receptors
1.10 Cell surface receptors
Cell surface receptors are transmembrane proteins that constitute an important class of
biological molecules that plays a crucial role in cellular communication processes.
Binding of a paracrine factor to these receptors initiates a signaling cascade and leads to
different biological outcomes, such as, cell migration, cell shape and spreading,
proliferation, apoptosis and others. Based on structure and functional similarities, these
receptors are classified into three major classes (Yokoyama, Higashida 1988):
a) Ion channel-linked receptors. These receptors converts chemical signal into
electrical signals, chemical signals in the form of neurotransmitter binds to these
receptors, alter their conformation by opening or closing the ion channels and maintain
the flow of Na
+
, Ca
2+
and Cl
-
ions across the cell membrane.
b) G-protein coupled receptors. These receptors traverse through the cell
membrane seven times and thus also called as 7-TM protein receptors. GPCRs transduce
extracellular signals by activating the heterotrimeric G-protein. G-protein remains in a
GDP bound inactive state however, upon ligand binding, GPCRs activates the G-protein
and transfer signals to the target protein (Wess 1997).
c) Enzyme-linked receptors. Upon ligand binding, these receptors either activate
their intracellular catalytic domain or associate with an enzyme for their activity.
Activated receptors then traverse the conformational change through the transmembrane
helix and initiate a signaling cascade that plays an important role in maintenance of
numerous cellular processes. Receptor tyrosine kinase, receptor serine/threonine kinases
40
and integrin receptors are a few examples of receptors belong to this class. Unlike
GPCRs, these receptors pass only once through the membrane and are heterodimeric.

1.11 Integrin family of adhesion receptors
As mentioned above, integrins are heterodimeric cell surface adhesion receptors that
mediate the interaction between extracellular matrix and cytoskeleton. They are
composed of two type I transmembrane glycoproteins, α subnunit and β subunit. To this
date, 18 α and 8β subunits have been identified that form 24 distinct integrin
heterodimers and are involved in specific non-redundant functions (Fig 1.14) (Hynes
2002; Takada, Ye et al. 2007).
Figure 1.14 Classification of integrin receptors.
They are further classified into four classes based on their evolutionary
characteristics, ligand specificity or cellular expression levels. Laminin, collagen,
41
fibronectin, vitronectin are a few high molecular weight proteins of the extracellular
matrix which provides support structure and also acts as a physical barrier for the cell
(Barczyk, Carracedo et al. 2010). The αv, α5, α8, αIIb subunits of integrins recognize the
RGD sequence present in fibronectin, vitronectin and are thus classified as RGD
receptors (van der Flier and Sonnenberg 2001). Integrins containing α1, α2, α10, α11
binds to collagen and integrins with α3, α6, α7 binds to laminin and are classified as
collagen and laminin receptors, respectively (Velling, Kusche-Gullberg et al. 1999).
Because of the restricted expression of β2, β7 integrins on white blood cells, they are
categorized as leukocyte-specific receptors.
Both α and β subunits of an integrin are composed of an extracellular domain, a
single transmembrane domain and a short cytoplasmic tail. The N-terminal extracellular
motif of an α-subunit comprises of a seven-bladed β-propeller domain, thigh, calf-1, and
calf-2 domains (Xiong, Stehle et al. 2001). Half of the α-subunits of integrins have an
inserted I-domain comprising of the nearly of 200 amino acids, inserted between the
second and third blade of the β-propeller (Springer 1997). Structural study of (I) domain
revealed the presence of a metal ion coordination site which also influences the ligand
binding activity and thus referred to as a metal-ion dependent adhesion site (MIDAS).
Thigh, calf-1 and calf-2 domains comprise the three β-sandwich domains. The interface
between the β-propeller and a thigh domain and between thigh and calf-1 domain
provides flexibility to the integrin structure and is also suggested to have a critical role in
the bidirectional signaling of an integrin. The β-subunit of integrin has more complex
structure and is composed of seven domains. The N-terminal region contains a β-I
domain, a hybrid domain, a psi domain followed by the four cysteine-rich repeats of
42
epidermal growth factor-like folds and a small β-tail (Figure1.15). The interface between
the β-I domain of β-subunit and a β-propeller of an -subunit provides the ligand-binding
site to the integrins lacking the I-domain within their α-subunit. Although the cytoplasmic
domains of integrins are very small, but there is no question that they play an
indispensable role in the various signal transduction pathways.

Figure 1.15 Schematic representation of the extracellular domain of integrin α and β
subunit.

However, the flexibility of these structures renders the information inaccessible
on the structural details of the integrin’s cytoplasmic tails. Several studies indicate that
43
cytoplasmic tails of α and β subunits interact with each other and control the activation
state of integrins. For example, αIIbβ
3
(a major platelet integrin) should remain in the
inactive state in the resting platelets; otherwise it will bind to its ligands in the
extracellular matrix and will lead to aggregation and, as a result, thrombosis. It has been
shown by Ginsberg and coworkers that deletion of the short cytoplasmic tail of the αIIb
integrin generates a constitutively active receptor (Hughes, Otoole et al. 1995). It has
been also proposed that the binding of a ligand ligand to the receptor is coupled with the
conformational changes within the I-domain or β-I domain. These changes further
transmit into other domains and are responsible for the open and close conformation or
active and inactive state of these receptors (Liddington and Ginsberg 2002).

1.12 Integrin signaling and tumorigenesis
Beyond adhesion, integrins are also known to regulate a diverse set of cellular processes,
such as differentiation, migration, proliferation, cell survival, gene expression and several
others (Streuli 2009). Activation of integrin receptor upon ligand binding results in
clustering of the integrins in the plane of the cell membrane and recruits multi-protein
complexes that link the extracellular matrix with cytoskeletal proteins and promotes actin
assembly (Giancotti and Ruoslahti 1999). Integrin engagement further leads to the
recruitment and activation of the focal adhesion kinase (FAK) which is a non-receptor
tyrosine kinase. Activation of FAK, in turn, autophosphorylates Tyr397 and creates a
docking site for SH2 domain that contains Src family of cytoplasmic tyrosine kinases
(Schlaepfer and Hunter 1996). Phosphorylated Src kinase then further mediate
phosphorylation at the other sites on FAK and allows binding of other SH2 domain-
44
containing signaling proteins. Activation of the FAK-Src pathway then initiates a
downstream signaling cascade which regulates cell migration, spreading, proliferation
and survival (Figure 1.16). Therefore, integrins are signaling receptors that transmit
signals in both direction, that is, signals from the ECM to inside of cell and also from
inside to the ECM.

Figure 1.16 Schematic example of the integrin signaling.

A significant amount of data in the literature has implicated the role of integrins in
tumor cell proliferation, migration and survival. Although 24 different integrin
45
heterodimers are known, α
v
β
3
, α
5
β
1
, α
6
β
4
, α
4
β
1
and α
v
β
6
receptors are overexpressed in a
number of cancers and thus are extensively studied integrins in cancer biology
(Desgrosellier and Cheresh 2010). For example, α
v
β
3
and α
5
β
1
are overexpressed in
melanoma and high expression of α
5
β
1
in non-small cell lung cancer is associated with
the poor prognosis and decreased survival rate for patients (Albelda, Mette et al. 1990;
Adachi, Taki et al. 2000).
Overexpression of α
v
β
3
in prostate and pancreatic cancer leads to high metastatic
potential and may have a possible role in invasion in glioblastoma. Most of the
chemotherapeutic drugs that are used as a first-line treatment in various cancers target
tumor cells as well as normal cells and that is why there is an increasing need for
therapies that target integrins. The overexpression of specific integrin receptors,
especially, α
v
β
3
, α
5
β
1
and α
v
β
6
in cancer also may provide an effective means of targeted
delivery of chemotherapeutic agents. The involvement of integrins in tumor cells
proliferation is well-supported by different studies and now there is growing body of
evidence that suggests that integrins on the surface of tumor-associated host cells can
significantly influence the tumor progression by inducing angiogenesis,
lymphangiogenesis and inflammation (Avraamides, Garmy-Susini et al. 2008). In
addition to their growth factor-independent signaling, integrins can also cooperate with
the growth factor receptors and can regulate the responses to other receptors. For
example, the crosstalk between α
v
β
3
and fibroblast growth factor receptor results in the
phosphorylation of Ser338 and Ser339 of Raf and protects cells from apoptosis
(Mahabeleshwar, Feng et al. 2007). There are also studies which indicate that α
v
β
3
also
46
activates the VEGFR2 and thus has crucial role in promoting angiogenesis (Hood,
Frausto et al. 2003).

1.13 Targeting α
v
β
3
integrin receptors
Among all integrins, α
v
β
3
receptor is the most widely studied member of the integrin
family. It is well-established that α
v
β
3
integrins are overexpressed on the surface of
various cancers, such as, melanoma, breast, prostate, pancreatic, ovarian, cervical and
glioblastoma. The α
v
β
3
integrin receptors are ubiquitously overexpressed on the surface of
endothelial cells undergoing angiogenesis and are almost absent on the surface of the
quiescent cells (Brooks, Stromblad et al. 1996; Eliceiri and Cheresh 1999). This
overexpression makes α
v
β
3
receptors an attaractive target both for cancer imaging and
therapy.
The α
v
β
3
receptor contains a 125 KDa α
v
and a 105 KDa β
3
subunit. The arginine-
glycine-aspartic acid (RGD) is the most common recognition sequence for this type of
integrin receptors (Liu, Wang et al. 2008). Despite the fact that half of the integrin
receptors recognize the RGD sequence in the different ECM proteins, ligand-integrin
specificity can be achieved because of the difference in the conformation of the RGD site
and the nature of its flanking amino acids. This sequence was first identified in
fibronectin and then later in vitronectin, and thrombospondin (Ruoslahti and
Pierschbacher 1987). Vitronectin is the major ligand for the α
v
β
3
receptor and many
ligands based on the RGD sequence have been designed and developed for targeting α
v
β
3

receptor for imaging as well as for the therapeutic purpose. For the past decade, several
47
cyclic RGD peptide-based ligands with increased specificity and also bioavailability have
been developed (Figure 1.17).

Figure 1.17 Examples of cyclic RGD peptides.

Because of the promiscuity in the RGD sequence recognition, efforts have been
made to design ligands that can specifically recognize one group of receptors over the
other. An in vitro screen of a library of conformationally constrained, bicyclic, lactam-
based Arg-Gly-Asp–containing pseudopeptides identified ST-1646 as a highly specific
ligand for α
v
β
3
/ α
v
β
5
with almost no affinity for α
5
β
1
(Belvisi, Riccioni et al. 2005). It is
1000-fold more potent inhibitor of α
v
β
3
/α
v
β
5
integrin as compared to the α
5
β
1
integrin,
which has substantial sequence similarity with the α
v
β
3
/α
v
β
5
receptors.Several non-cyclic
RGD based peptidomimetics that target α
v
β
3
/α
v
β
5
receptors have also been developed
(Marchini, Mingozzi et al. 2012) . The main goal in developing of these inhibitors or
ligands was to increase the proteolytic resistance and bioavailability. The sequence
similarity in α
v
β
3
and α
v
β
5
receptors makes it challenging to develop ligands which can
specifically target one integrin over another. The overexpression of α
v
β
3
receptors on
tumor cells and activated endothelial cells of the tumor vasculature offers an opportunity
48
for non-invasive imaging and delivery of integrin-drug conjugates to α
v
β
3
-positive
tumors. Until recently, most in vivo imaging studies that utilize α
v
β
3
integrin ligands
were performed with radiopharmaceuticals, mailny due to the high sensitivity of these
methods. However, use of nuclear imaging technique requires special facility and
handling of the materials is difficult due to the issues associated with exposure to
radioisotopes.
In the second part of the thesis we describe the design, synthesis, and biological
properties of a conjugate comprised of high affinity, α
v
β
3
-selective integrin ligand, and a
fluorescent probe that exhibits fluorescence enhancement upon binding to target. Given
the high specificity of this ligand towards α
v
β
3
receptors, this system has a potential for
applications in optical imaging and, through further modification with a cytotoxic
modality, as a therapeutic agent, for treatment of solid tumors.

49

Chapter 2: Modulation of Hypoxia-Inducible Transcription by Rationally Designed
Epipolythiodiketopiperazines

50
2.1 Structural basis of interaction between HIF-1α and p300
Over a decade ago, Kung and coworkers reported that the interaction between HIF-1α
and p300 and the transactivation of hypoxia inducible genes is almost exclusively
dependent on the interaction between the C-terminal transactivation domain (CTAD) of
HIF-1α and cysteine-histidine rich (CH1) domain of the coactivator protein p300 or it
ortholog, CREB-binding protein (CBP) (Kung, Wang et al. 2000). Like HIF-1α,
p300/CBP (Figure 2.1) is a multi-domain protein with three cysteine-histidine rich
domains (CH1, CH2, and CH3), a nuclear hormone receptor-binding domain (Nu), a
kinase inducible domain of CREB interacting domain (KIX), bromodomain (Br), a
histone acetyltransferase domain (HAT), a glutamine-rich domain (Q), and an IRF-3-
binding domain (I) (Freedman, Sun et al. 2002).

Figure 2.1 Domain structures of HIF-1α (Top) and p300/CBP (bottom). Amino acid
residues which undergo hydroxylation under normoxic conditions are labeled with OH
and lysine residue labeled as Ac for acetylation. Three zinc dependent domains of p300
were labeled with Zn
2+
. CH1 and CH3 domains are structurally homologous and also
termed as TAZ1 and TAZ2, respectively.
The CH1 and CH3 domains are structurally homologous, yet there is only a small overlap
in the regions where transcription factors interact (Figure 2.1). Although it is reported
51
that N-terminal transactivation domain (NTAD) of HIF-1α interacts with the CH3
domain of p300/CBP, it transactivates less effectively (Ruas, Berchner-Pfannschmidt et
al. 2010).
The three zinc ions are present at the vertices of this triangle and coordinates with three
cysteine and one histidine residues with a small C-terminal helix α4 completing the
coordination of the third zinc ion (Figure 2.2). The CTAD domain of HIF-1α is mainly
composed of two α-helices (αA and αB) that clamps around the triangular structure of the
CH1 domain (Freedman, Sun et al. 2002).

Figure 2.2 Solution structure of the CH1 domain of p300 in complex with CTAD
domain of HIF-1α. The CH1 domain (aa 323-423) shown in violet with three major
helices (α1,α2, α3) and one small α-helix (α4). The CTAD domain is shown in red with
two α-helices (αA, αB). Zinc ions are represented as green spheres at the vertices of
violet ribbon triangular structure of CH1 domain. Structure was created from Pdb ID:
1L3E.
Other than PHD2, which degrades HIF-1α under normoxic conditions, cellular
machinery offers another checkpoint in a form of asparagine hydroxylase. This enzyme
52
hydroxylates asparagine (Asn803) of HIF-1α under well-oxygenated conditions and
prevents the recruitment of p300/CBP. The Asn803 hydroxylation creates steric clash
and places the hydroxyl group in a hydrophobic environment without any hydrogen-
bonding partner. These unfavorable interactions seem to be responsible for the
degradation of HIF-1α under normoxic conditions.

2.2 Chetomin blocks the interaction between HIF-1α and p300
Realizing the importance of the HIF-1α in tumor progression and development of
resistance towards chemo- and radiotherapy, significant efforts have been made to
identify inhibitors of HIF-1 signaling. A number of small molecule inhibitors of HIF-1
pathway were identified by high-throughput screen or by rational design.

Figure 2.3 Structure of chetomin. Chetomin contains two epidithiodiketpiperazine
motifs with rigid side chains and cyclotryptophan motifs.
These inhibitors are mainly classified as 1) inhibitors of HIF-1α protein synthesis; 2)
inhibitors of HIF-1 DNA binding; 3) agonists of enzymes responsible for HIF-1α
degradation, and 4) inhibitors of HIF-1α transcriptional activity (Xia, Choi et al. 2012).
From a 600,000-member library, Kung and coworkers identified a small molecule,
53
chetomin, containing bis-epidithiodiketopiperazine (bis-ETP) motif as a putative
transcriptional modulator of the HIF pathway (Kung, Zabludoff et al. 2004). Chetomin
was isolated as a secondary metabolite from a fungus Chaetomium cochliodes and also
showed antifungal and antibacterial activities. As mentioned above, chetomin is a dimeric
epidithiodiketopiperazine (ETP) that hasthe two ETP rings are connected via rigid indole-
based scaffold (Figure 2.3). It has been shown through an in vitro interaction assay that
chetomin disrupts the high-affinity complex between the full-length HIF-1α or HIF-2α
and CH1 domain of p300/CBP. Although at that time the mechanism of action remained
unclear, NMR spectral shifts of the resonances of CH1 domain in the presence of
chetomin were found to be similar to those of CH1 in the presence of EDTA, suggesting
an unfolding of the CH1 domain by chetomin. These findings and also absence of any
covalent modifications to p300/CBP, suggested that chetomin is an allosteric inhibitor of
the p300 coactivator that is likely coordinates with zinc ions of the CH1 domain, thereby
disrupting the global fold of the CH1 domain. Recent mechanistic studies of ETP
molecules also implicated the redox environment of cancer cells as likely facilitating the
uptake of the ETP compounds. The sulfurs of the disulfide core are reduced to dithiol and
coordinates with the zinc ion of the CH1 domain. This coordination is primarily
responsible for disrupting the proper folding of the CH1 domain and thus inhibits its
interaction with the CTAD domain of HIF-1α.

2.3 Rationale behind the design of novel monoepipolythiodiketopiperazines
(monoETPs)
The interaction surface area between the two interacting domains of HIF-1α and p300 is
extensive - it spans approximately 3300 Å and thus, direct blockage by a small molecule
54
would be difficult. Instead, inducing a structural change in one of the binding partners
would be more effective approach to disrupt the complex. Our design was inspired by the
natural products chetomin, dimeric epipolythiodiketopiperazine (ETP) and leptosin I
(Takahashi, Numata et al. 1994), a monomeric ETP with the tetrasulfide motif within its
south fragment (Figure 2.4). Despite the ability of chetomin to disrupt binding between
the transcription factor HIF-1α and its coactivator p300 complex its use in vivo is
severely limited, as it induced coagulative necrosis, anemia, and leukocytosis in
experimental animals.

Figure 2.4 Structure of natural products chetomin and leptosin I. Chetomin contains two
ETP moieties with disulfide bridges in north and south fragments, whereas leptosin I has
one ETP moiety containing tetrasulfide core within its south fragment.
Given the complexity of the structure of chetomin, no total synthesis of chetomin
is reported to date. It has been observed by our group as well as by others that the ETP
core and its bridging disulfides is both necessary and sufficient for the activity of
chetomin and synthetic ETPs (Block, Wang et al. 2009; Cook, Hilton et al. 2009). For
instance, Schofield and coworkers have published a study with designed monomeric ETP
compounds (Figure 2.5) that contain a disulfide core (1), thiol (2), dithiol (3), and thiol
55
ethers (4). They have shown that compounds having monothiol and thiol ethers were not
able to disrupt the HIF-1 / p300 complex in vitro. The lack of activity of these
compounds suggests the importance of disulfide/dithiols in the ETP core.

Figure 2.5 Structures of monomeric ETPs. Compound 3 and d 4 were reported to be
inactive in vitro.
In order to facilitate the synthesis, study the structure-activity relationship of ETP
and to diminish the non-specific in vivo toxicities of the ETP compounds, we designed
compounds with a single ETP core having varied side chain functionalities at C-3 and C-
6 positions. The mono-ETP-1 was designed with a single disulfide bridge and two
flexible side chains of benzyloxy methyl at C-3 and C-6 position. To study the effect of a
single flexible side chain, mono-ETP-2 and mono-ETP-3 were proposed. Furthermore,
mono-ETP-4 and gliotoxin were included in the study (Figure 2.6). It was proposed that
ETP compounds could disrupt HIF-1α -p300 complex by a zinc ejection mechanism.
The torsionally-strained disulfide bridges can be readily reduced to cysteine thiolates and
together with cysteines in CH1 domain forms covalent bond with zinc ions, resulting in
the formation of a protein-ETP complex. This transient complex then rearranges to form
disulfide in the protein and the ejected Zn-ETP complex further binds another ETP core
to stabilize the zinc complex (Cook, Hilton et al. 2009). Considering this mechanism, it
was shown that both kinetic and thermodynamic considerations make the torsionally
56
strained bridging disulfides of the ETP core that are particularly suitable for disrupting
zinc-binding sites.

Figure 2.6 Structures of designed mono-ETPs designed for the SAR studies.
After analyzing the structure of chetomin, we explored the north fragment of the
chetomin containing diketopiperazine tetrasulfide motif and synthesized mono-ETP-5
(Figure2.6). The presence of a tetrasulfide bridge would induce additional torsional strain
and placement of the indole side chain may also result in the favorable interaction
between the mostly hydrophobic contact surfaces of CTAD of HIF-1α and CH1 domain
of p300/CBP (Freedman, Sun et al. 2003).

2.4 Synthesis of mono epipolythiodiketopiperazines (monoETPs)
All mono-ETP compounds were synthesized using previously published procedure with
certain modifications (Fukuyama, Nakatsuka et al. 1981). The synthesis of compounds
involved three key steps: 1) protection of sulfurs in the form of dithioacetal; 2)
57
functionalization at C-3 and C-6 positions via carbanion chemistry, and 3) deprotection
of dithioacetal and formation of the polysulfide bridge. Protection of the polysulfide in
the form of dithioacetal is important, as it would allow for the functionalization of the
bridgehead carbons via carbanion chemistry. Synthesis of mono-ETP compounds was
accomplished in collaboration with Dr. Hui Wang and Dr. Lajos Z. Szabo.
2.4.1 Synthesis of mono-ETP-1
The thioacetal 5 was obtained from sarcosine anhydride 1 by the method reported by
Kishi et al. (Fukuyama, Nakatsuka et al. 1981). The regioselective deprotection of 5 at
the bridgehead carbon with a strong base, n-butyl lithium, and its subsequent reaction
with benzyloxymethyl chloride afforded disubstituted dithioacetal 9 in good yield.
Scheme 2.1 Synthetic scheme for mono-ETP-1.
Further, conversion of the protected dithioacetal into mono-ETP-1 was carried
out by initial oxidation with m-chloroperbenzoic acid, followed by the treatment with
70% perchloric acid in dimethyl sulfide (Scheme 2.1).
2.4.2 Synthesis of mono-ETP-2
Mono-ETP-2 was synthesized by following the same procedure as above for the
synthesis of mono-ETP-1, except that instead of 2.2 equivalents of n-butyl lithium, only 1
equivalent of the reactant was used. Deprotection of the bridgehead carbon with n-butyl
lithium afforded monosubstituted dithioacetal 6 in 53% yield (Scheme 2.2). Finally,
58
removal of dithioacetal with m-chloroperbenzoic acid and perchloric acid yielded mono-
ETP-2 in 50% yield.

Scheme 2.2 Synthetic scheme for mono-ETP-2.
2.4.3 Synthesis of mono-ETP-3
Removal of benzyloxy methyl groups from disubstituted dithioacetal 9 was achieved by
treating 9 with boron trichloride in DCM and afforded dihydroxy dithioacetal 10 in 93%
yield.

Scheme 2.3 Synthetic scheme for mono-ETP-3.

59
Acetyl groups were placed at the C-3 and C-6 position by treating 10 with acetic
anhydride and pyridine, resulting in the formation of 11 in 54% yield (Scheme 2.3).
Finally, removal of the dithioacetal and regeneration of the disulfide bridges was
accomplished by treatment of 11 with mCPBA and dimethyl sulfide in perchloric acid
and afforded mono-ETP-3 in 50% yield.
2.4.4 Synthesis of monoETP-4
Mono-ETP-4 was synthesized from commercially available 1.4-dimethyl-2,5-
piperazinedione by following the modified procedure of Kishi and coworkers. Briefly,
dithioacetal, 5 was treated with strong base and subsequently treated with benzyloxy
methyl chloride to afford monosubstituted benzyloxy methyl dithioacetal 6 in good yield.
.
Scheme 2.4 Synthetic scheme for mono-ETP-4.
60
Next, monosubstituted benzyloxy methyl dithioacetal 6 was treated with benzyl bromide
in presence of n-butyl lithium and gave disubstituted dithioacetal 7 in 57% yield (Scheme
2.4). Further, benzyl group was removed by boron trichloride and gave 8, followed by the
installation of disulfide bridges with mCPBA and dimethyl sulfide in perchloric acid and
afforded mono-ETP-4.
2.4.5 Synthesis of mono-ETP-5
Monosubstituted dithioacetal 6 was deprotonated at the bridgehead position with
LHMDS and treated with tert-butyl-3-bromomethyl-indole-1-carboxylate 13 giving the
disubstituted dithioacetal 12 in 46% yield.

Scheme 2.5 Synthetic scheme for mono-ETP-5.
Removal of the Boc group from the indole moiety and subsequent installation of
the polysulfide bridge was accomplished by treatment of 12 with mCPBA and dimethyl
sulfide in perchloric acid and afforded the intermediate ETP-5a/ETP-5b. This
61
intermediate was a mixture of the tetrasulfide with some amount of the disulfide and
monosulfide byproducts. It has been previously reported that the rapid sulfur exchange
between disulfide and reduced dithiols results in the formation of a mixture of disulfides
and tetrasulfides. Removal of the benzyl group in ETP-5b with one equivalent of boron
trichloride afforded mono-ETP-5 (Scheme.2.5), albeit in low overall yield from 12.

2.5 Preliminary structure activity relationship (SAR) studies of mono-ETPs
In order to determine the efficacy of mono-ETP compounds in regulation of the
expression of hypoxia inducible genes, we tested the activity of these compounds
in a cell culture-based luciferase assay. For these studies, a stably
transfected MDA-MB-231 cell line containing five repeats of the HRE
sequence (TACGTGGG) from the VEGF promoter region was used (Shibata, Giaccia
et al. 2000). In preliminary experiments, the newlysynthesized ETPs were first tested
with chetomin as positive control and vehicle as a negative control. Cells were plated in a
24-well plate and dosed with mono-ETP-1, mono-ETP-2, mono-ETP-3, mono-ETP-4,
chetomin and gliotoxin (Sigma) at 200 nM concentration for 24 h. Six hours after dosing
hypoxia was induced chemically with desferroxamine mesylate (DFO). DFO is an iron
chelator, which stabilizes HIF-1α by diminishing the availability of Fe
2+
for PHD2
enzyme. Luciferase activity was measured by a luminometer and inhibition was reported
in reference to vehicle (Figure 2.7). At 200 nM concentration, no significant decrease of
the luciferase activity was observed with designed ETPs having flexible side chains. On
the other hand, gliotoxin showed 30% decrease at 200 nM concentration.

62

Figure 2.7 Luciferase reporter assay in MDA-MB-231-HRE-Luc cell line with chetomin,
synthetic mono-ETPs, and gliotoxin. Cells were maintained in DMEM with 10% FBS
and geneticin at working concentration of 0.4 g/L. Each compound was tested at 200 nM
concentration. Vehicle sample had only media with 0.1% DMSO. Gliotoxin showed
inhibition of HIF-1α promoter activity by 30%. Experiment was performed in triplicates.

Next, we evaluated the dose response with mono-ETP-4 (hyalodendrin) at 200
and 600 nM and gliotoxin at 50 and 200 nM. At higher concentration, mono-ETP-4
showed 37% decrease in activity (Figure 2.8). Based on these preliminary results, we
decided to synthesize the second generation mono-ETP-5. Both mono-ETP-4 and
gliotoxin were considered as controls, however, due to high toxicity of gliotoxin in most
cell lines available to us, we opted to exclude it from the study.
63

Figure 2.8 Luciferase reporter assay for dose response studies with mono-ETP-4 and
gliotoxin. Assay was carried out in a stably transfected breast cancer cell line, MDA-MB-
231. Chetomin (CTM) was used as a positive control at 200 nM concentration, mono-
ETP-4 was tested at 200 and 600 nM and gliotoxin (GTX) was tested at 50 and 200 nM
concentrations.

2.6 Synthetic mono-ETP-5 disrupts HIF-1/p300 complex
As discussed earlier, the triangular structure of CH1 domain acts as a scaffold for the
proper folding of the CTAD domain of HIF-1α. In the absence of CH1 domain, CTAD
domain remains unstructured. This suggests that in order to act as transactivator, CTAD
domain of HIF-1α needs to bind tightly to the CH1 domain of p300. For determining the
affinity of ETPs, we expressed the CH1 domain of p300 as a fusion with glutathione-S-
transferase (GST). The gene encoding aa residues 323 to 423 of the CH1 domain of p300
(Figure 2.9), was cloned in pGEX4T-2 vector along with the GST tag.
64

In brief, the gene corresponding to human p300 CH1 domain (aa323-423) along with
GST tag in puc57 was obtained from Genscript Inc. The plasmid was subcloned into the
expression vector pGEX 4T-2 between BamH1 and EcoR1 restriction sites.

Figure 2.9 Amino acid sequence of GST-p300-CH1 domain. GST tag is shown in black,
thrombin cleavage site is shown in blue, CH1 domain of p300 is shown in green, and a
tail of residual amino acids is shown in red.

After ligation, the pGEX 4T-2-p300 fusion vector was transformed into BL21
DE3 pLys competent E. coli. The presence of the desired insert (GST-p300-CH1) was
confirmed by DNA sequencing. Further, 39 kDa GST-p300-CH1 protein was purified, as
discussed in the Experimental Section 6.2.2. For binding assays, fluorescein-labeled
CTAD domain of HIF-1α (aa 786-826) was obtained from Prof. Paramjit Arora, NYU.
2.6.1 Determination of the binding affinity between HIF-1-flu CTAD and p300-CH1-
GST by fluorescence polarization (FP) saturation binding assay
The basic underlying principle of this technique is that when a fluorescent molecule is
excited by a plane polarized light it will emit light in the same plane, if the molecule
remains stationary (Perrin 1926). If the fluorescent molecule rotates or tumbles during the
excitation, it will emit light in the different plane than the plane of the excited light.
65
The FP assay is a powerful tool to study bimolecular interactions (Jolley, Stroupe et al.
1981) and it was used extensively in our binding study. First, the binding affinity of CH1
domain of p300 towards the CTAD domain of HIF-1α was determined. Fluorescein-
tagged CTAD of HIF-1α was taken as a limiting component and maintained at a constant
concentration of 15 nM in an FP buffer. The concentration of GST-p300-CH1 varied
between 1 to 2000 nM. Each concentration of p300 was made in FP buffer (50 mM Tris,
150 mM NaCl, 100 μM ZnCl2, 0.1% NP-40, 1 mM DTT and 10% glycerol) with 1%
DMSO and 0.1% pluronic acid. HIF-1α-flu CTAD was added to each concentration of
p300-CH1 and incubated at room temperature before reading the polarization values in a
multiwell microplate reader. The K
d
values obtained from this assay allow to determine
the final concentration of p300-CH1 to be used for competition assays. The K
d
value of
55 nM was obtained from the saturation binding curve. To increase the dynamic range of
the absolute value of polarization, we chose to use a slightly higher concentration of
GST-p300-CH1 for FP-based competition assays.
2.6.2 Fluorescence polarization (FP) competition assay with mono-ETP-5, mono-ETP-
4, and chetomin
If the interaction between HIF-1α and p300 is inhibited, the complex between CTAD
HIF-1α and CH1-p300 is disrupted with ETPs at a range of concentrations that is
comparable to the K
d
value obtained from saturation binding curve. This led us to use 75
nM of GST-p300-CH1 and 15 nM of HIF-1α-flu CTAD for the competition assay. The
principle behind the competition experiment is that an unbound fluorophore, excited by a
polarized light, tumbles rapidly and will have low polarization value. On the other hand,
in the presence of the bound protein, the fluorophore will rotate slowly and give higher
polarization value.
66
From the competition assay, we aimed to see if the polarization value obtained from the
saturation binding assay (75 nM of GST-p300-CH1 and 15 nM of HIF-1α-flu CTAD)
would decrease in the presence of the mono-ETP compounds.

Figure 2.10 Synthetic mono-ETP-5 disrupts the complex of CTAD HIF-1α and p300-
CH1. Fluorescence polarization competition assay was performed with a fluorescein
labeled CTAD HIF-1α and GST tagged p300-CH1. The IC
50
value obtained from the data
was 3.8 ×10
-5
M. These data points are average values with the error bars representing ±
s.d. (standard deviation) of experiments performed in triplicate.
Next, we used this assay with mono-ETP-5, mono-ETP-4 and chetomin to
analyze their effect on the stability of the HIF-1α - p300 complex. The fluorescein-tagged
CTAD of HIF-1α was incubated with GST-p300-CH1 before the addition of increasing
concentrations of the test compounds (Figure 2.10). The results indicate that synthetic
mono-ETP-5 has a high affinity for p300 and disrupts the HIF-1α and p300 complex with
an IC
50
value of 3.8 x 10
-5
M.
67
Next, the log of mono-ETP-4 concentration was plotted against the percentage change in
the FP values and showed no disruption of the complex at concentrations up to 100 µM
with only a partial dissociation of the complex at 300 µM concentration, indicating that
its IC
50
value is above 3.3 x 10
-4
M (Figure 2.11).

Figure 2.11 FP assay with synthetic mono-ETP-4. No disruption of CTAD HIF-1α and
p300-CH1. Fluorescence polarization competition assay was performed with a
fluorescein labeled CTAD HIF-1α and GST tagged p300-CH1. These data points are
average values with the error bars representing ± s.d. (standard deviation) of experiments
performed in triplicate.
The higher affinity of mono-ETP-5 as compared to mono-ETP-4 supports our
hypothesis that higher torsional strain of polysulfide bridge makes it higher affinity
ligand and, potentially, more effective allosteric inhibitor of HIF-1α and p300 than ETP-4
under our FP assay conditions.
68
Chetomin, the bis-ETP, was used as a positive control and showed high affinity
towards the complex (Figure 2.12). However, as stated above, the high toxicity of
chetomin diminished its potential for in vivo applications.

Figure 2.12 Chetomin disrupts the complex of CTAD HIF-1α and p300-CH1 with an
IC
50
value of 633 nM. These data points are average values with the error bars
representing ± s.d. (standard deviation) of experiments performed in triplicate.

2.7 Cytotoxicity of ETP-5, ETP-4, and chetomin in MDA-MB-231-HRE-Luc cell line
In order to test these compounds in cell-based assays, it was important to evaluate their
cytotoxicity profile. In addition, ETP compounds are shown to have glutathione-
dependent uptake, often resulting in a final concentration in cells that is much higher than
the actual concentration in the cell culture medium (Bernardo, Brasch et al. 2003). In this
experiment, cells were plated in a 96-well plate and dosed with increasing concentrations
of mono-ETP-5, mono-ETP-4 and chetomin and incubated for 24 h. We employed a cell
titer blue viability assay which is based on the ability of live cells to convert a non-
69
fluorescent redox dye resazurin into fluorescent dye resorufin. On the other hand, dead
cells due to their low metabolic activity produce low fluorescent signal. Mono-ETP-5
showed the GI
50
value of 850 nM (Figure 2.13) and mono-ETP-4 – the value of 1.1 µM
(Figure 2.14).

Figure 2. 13 Cell titer blue assay with mono-ETP-5 in MDA-MB-231-HRE-Luc cells.
Cells were treated with increasing concentration of monoETP-5 and incubated for 24 h.
Fluorescent signal produced was then recorded by multiplate reader with excitation and
emission wavelength of 560 nm and 590 nm, respectively. Error bars represent ± s.d.
(standard deviation) of experiments performed in quadruplicate. Cells were maintained in
DMEM with 10% FBS and geneticin at working concentration of 0.4 g/L.
Interestingly, chetomin showed a GI
50
value of 200 nM, indicating high
cytotoxicity (Appendix A, Figure A.1). This data implies that both of the mono-ETP
compounds are less cytotoxic than bis-ETP chetomin. Based on the results of our
viability assays, we chose to measure the effect of mono-ETP-5 and mono-ETP-4 at a
concentrations not exceeding 600 nM.
70

Figure 2.14 Cell titer blue assay with mono-ETP-4 in mDA-MB-231-HRE-Luc.Cells
were treated with increasing concentration of monoETP-4 and incubated for 24 h.
Fluorescent signal produced was then recorded by multiplate reader with excitation and
emission wavelength of 560 nm and 590 nm, respectively. Error bars represent ± s.d.
(standard deviation) of experiments performed in quadruplicate.

2.8 Evaluation of mono-ETP-5 as transcriptional modulator of hypoxia-inducible
pathway
Our binding assays showed that ETP-5 disrupts the complex of HIF-1α and p300 in vitro.
Next, we evaluated the potential of this compound to disrupt the transcriptional activity in
a luciferase reporter system. For this, we used a stably transfected MDA-MB-231 breast
cancer cell line, a gift from Prof. Robert Gillies. This cell line contains five repeats of
hypoxia response element (HRE, TACGTGGG) present in the VEGF promoter to drive
the luciferase expression. Cells were plated and dosed with mono-ETP-5 at three
different concentrations, 50, 200, and 600 nM with a final concentration of 0.1% DMSO.
Hypoxia was induced by DFO for 18 h with total incubation time of 24 h with
71
compounds. Treatment with mono-ETP-5 at 200 nM concentration resulted in 37%
decrease of the levels of HIF-1 inducible transcription and at 600 nM reduced the levels
of transcription by 75% (Figure 2.15).

Figure 2.15 Inhibition of HIF-1α inducible promoter activity in a stably transfected
MDA-MB-231-HRE-Luc cell line by synthetic mono-ETPs. Cells were treated with
mono-ETP-5 at 50, 200 and 600 nM and with monoETP-4 at 200 and 600 nM.
monoETP-5 showed 75% decrease in the promoter activity. Error bars represent ± s.e.m.
of experiments performed in triplicate. ***P < 0.001, **P< 0.01, t test.

The sub-micromolar activity of mono-ETP-5 in cell based assays, as compared to
the FP assay is likely due to its active uptake in live cells. In contrast, mono-ETP-4 at 200
nM was not able to show any significant decrease and at increasing concentration of 600
nM, it showed decrease of 37%. Although mono-ETP-4 showed decrease at higher
concentration, mono-ETP-5 is significantly more effective than control mono-ETP-4 in
reducing the luciferase activity in hypoxic cells at lower concentrations. The reduced
72
activity of mono-ETP-4 correlates well with its lower affinity toward the HIF-1α-p300
complex in FP binding assays (Kushal, Wang et al. 2011).

2.9 Synthetic mono-ETP-5 downregulates levels of secreted VEGF
After testing the efficacy of mono-ETP-5 in downregulating the expression of HIF
inducible genes, we evaluated the ability of this compound to reduce the levels of
secreted VEGF protein. VEGF plays a critical role during embryogenesis, as well as in
tumor angiogenesis. It is one of the key downstream target proteins of hypoxia inducible
transcription and is a powerful mitogen. The observed downregulation of the VEGF
promoter activity in luciferase assay should correleate with decreased levels of its protein
and the rate of neovascularization.
For this experiment, MCF-7 and a triple negative breast cancer cell line MDA-
MB-231 were used to determine the secreted levels of VEGF by ELISA. Cells were
plated in 24 well plates and treated with compounds under both normoxic and hypoxic
conditions for 24 h. Treatment with mono-ETP-5 at 400 nM showed a 21% decrease in
MDA-MB-231 cells (Figure 2.16A) and 24% decrease in MCF-7 cells, respectively
(Figure 2.16B). On the other hand, no significant decrease was observed with mono-
ETP-4. This data demonstrates that compensatory cellular stress response mechanisms
such as those affecting internal ribosome entry sites (IRES) or mechanisms enhancing
protein translation do not override the observed downregulation in expression. Moreover,
this data also parallels with the in vitro FP assays. This result indicates that mono-ETP-5
has the ability to inhibit secreted levels of VEGF, and,potentially, to reduce its mitogenic
effect on vasculature.
73

Figure 2.16 Analysis of secreted levels of VEGF protein with mono-ETP-5. A) MDA-
MB-231 and B) MCF7 cells treated with chetomin, mono-ETP-5, and mono-ETP-4 by
ELISA. Cells were treated under both normoxic (DFO,-) and hypoxic (DFO,+)
conditions . All compounds were added at a final concentration of 400 nM. Error bars
represent ±s.e.m of experiments performed in quadruplicate. **P< 0.01, t test.
74

2.10 Effect of mono-ETP-5 on the global fold of p300-CH1
As shown previously that chetomin alters the global fold of the CH1 domain of p300, we
utilized the circular dichroism spectroscopy to analyze if synthetic mono-ETP-5 is
showing the similar effect.

Figure 2.17 Effect of monoETP-5 on the global fold of CH1 domain of p300 by CD
spectroscopy. Blue curve represents the p300-CH1 domain and green cure represents the
p300-CH1 domain in the presence of 1 µM of monoETP-5.

NMR studies on chetomin showed that it does not modify CH1 domain covalently
and also chemical shifts observed were similar to that of EDTA treatment. The p300-CH1
fusion protein was exposed to increasing concentration of mono-ETP-5 and showed
significant change in the CD spectra, indicating that mono-ETP-5 alters the fold of the
CH1 domain (Figure 2.17).

75
2.11 SAR studies with bis-epidithiodiketopiperazines
In parallel with our evaluation of mono-ETPs as inhibitors of hypoxia-inducible
transcription, we also explored the potential of bis-ETPs as transcriptional modulators.
Chetomin, a bis-ETP fungal metabolite, is shown to downregulate the expression of HIF
inducible genes in vitro as well as in vivo. As discussed above, it has shown high
affinity toward HFI-1α-p300 complex; however its reported in vivo toxicity has limited
its further development in vivo. In order to better understand the effect of dimeric ring
system in bis-ETPs on affinity and activity, we designed a series of the linked ETP
dimers and explored their binding properties and biological effects.
2.11.1 Rationale behind the design of bis-epidithiodiketopiperazines (bis-ETPs)
It has been shown that under physiological conditions ETPs exist in either oxidized
disulfide form or reduced dithiol form, which is essential for their biological activity. The
results from our experiments with mono-ETPs indicate that the chain functionalities do
contribute towards the biological activity of this class of natural products. However, the
two properly spaced ETP rings of chetomin could create a multivalent interaction that is
capable of significant contribution toward the increase of the binding affinity. Hence, a
dimeric ETP that is mimicking the placement of the two ETP rings in chetomin could
have binding affinity comparable to chetomin. In order to test that hypothesis, the first
dimeric ETP (bis-ETP-1) was designed to match the conformationally-averaged distance
between the bridging disulfide groups in chetomin, which is approximately 10 Å, and
linked with an aromatic linker instead of indole side chains. To ascertain the relevance of
the disulfide bridge, a diketopiperazine (DKP-1) lacking the disulfide bridges was
synthesized as a control compound (Figure 2.18). Since the two enantiomers could bind
differently to the CH1 domain of p300 and may have different activity; the racemic bis-
76
ETP-1 was subjected to chiral separation and binding affinity and activity of each
enantiomer of bis-ETP-1 were tested (Block, Wang et al. 2009).

Figure 2.18 Structures of chetomin and designed dimeric ETPs. bisETP-1 and bisETP-2
are the designed dimeric ETP compounds, DKP-1, is a negative control.

Examination of the backbone resonances of p300 CH1 region by NMR showed
that adding chetomin induces a conformational change in that region. Schofield and
coworkers further examined the mechanism and suggested that ejection of zinc ions from
the CH1 domain of p300 is primarily responsible for the disruption of the global fold of
CH1 domain. The second dimeric ETP (bis-ETP-2) was designed to mimic the
conformationally-averaged distance between the two zinc ions present within the
structure of p300-CH1, which is approximately 20 Å (Figure 2.23). This was achieved by
placing one additional aromatic linker between the ETP rings with the averaged distance
between the two disulfide bridges of 20 Å.

77
2.11.2 Synthetic bis-ETP disrupts the formation of a complex between HIF-1α and
p300
Next, we evaluated the ability of synthetic bis-ETPs to inhibit the formation of complex
between HIF-1α and p300. After obtaining a saturation binding curve as described above
(see Chapter 6, Experimental Section), we employed a fluorescence polarization
competition assay to access the disruption of fluorescein labeled CTAD of HIF-1α and
CH1 of p300 by bis-ETP-1 and bis-ETP-2.

Figure 2.19 Synthetic bis-ETP-1 disrupts the complex of CTAD HIF-1α and p300-CH1
with an IC
50
value of 1.5 µM. Fluorescence polarization competition assay was
performed with a fluorescein labeled CTAD HIF-1α and GST tagged p300-CH1. These
data points are average values with the error bars representing ± s.d. (standard deviation)
of experiments performed in triplicate.

Our competition binding assays confirmed that both bis-ETP-1 and bis-ETP-2 are
capable of disrupting the p300-CH1-GST/CTAD HIF-1α complex. The calculated IC
50

for bis-ETP-1 was 1.5 µM (Figure 2.19) and bis-ETP-2 was 1.9 µM (Figure 2.20) under
our experimental conditions.
78
On the other hand, DKP-1 did not bind the CH1 region at concentrations up to 80 M.

Figure 2.20 Synthetic bis-ETP-2 disrupts the complex of CTAD HIF-1α and p300-CH1
with an IC
50
value of 1.9 µM. Fluorescence polarization competition assay was
performed with a fluorescein labeled CTAD HIF-1α and GST tagged p300-CH1. These
data points are average values with the error bars representing ± s.e.m (standard
deviation) of experiments performed in triplicate.

2.11.3 Synthetic bis-ETPs inhibits the activity of VEGF promoter
After observing that synthetic bis-ETPs disrupts the complex formation of CTAD HIF-1α
and GST-p300-CH1, we conducted a preliminary evaluation of these molecules in a
luciferase reporter system in order to test their ability to reduce HIF-1α inducible
promoter activity. As discussed above in section 2.9, MDA-MB-231-HRE-Luc breast
79
cancer cells were plated and dosed with bis-ETP-1, bis-ETP-2 and control compound
DKP-1 at 200 nM concentration.

Figure 2.21 Dose-dependent inhibition of HIF-1α inducible promoter activity by bis-
ETP-1 and bis-ETP-2 in MDA-MB-231-HRE-Luc cell line. Bis-ETP-1 was tested at 200
and 600 nM, bis-ETP-2 was tested at 200, 400 and 600 nM. DKP-1 is used as a negative
control and tested at 200 and 600 nM. Chetomin is a positive control and is added at 200
nM concentration. All samples were maintained at a final concentration of 0.1% DMSO.
Hypoxia was induced with 300 µM DFO. Error bars represent ±s.e.m. of experiments
performed in triplicate. **P< 0.001, t test.

Chetomin at 200 nM was used as a positive control. Furthermore, these
compounds were evaluated for dose response by treating cells with different
concentrations of bis-ETP-1 and bis-ETP-2. Hypoxia was chemically mimicked by DFO
and cells were incubated for 24 h. After 24 h, cell lysate was collected and used for the
determination of the luciferase activity. Based on the GI
50
values for bis-ETP-1 and bis-
ETP-2, a maximum concentration of 600 nM was chosen.
80
Treatment with bis-ETP-1 reduced the luciferase expression by 85% at 200 nM
and by 94% at 600 nM. Bis-ETP-2 at the same concentrations showed a more significant
decrease (90% and 96% at 200 nM and 600 nM, respectively, Figure 2.21).

Figure 2.22 Two enantiomers of bis-ETP-1, ent1-bis-ETP-1 and ent2-bis-ETP-2,
showed similar inhibition of HIF-1α inducible promoter activity in the MDA-MB-231-
HRE-Luc cell line and additive effect was observed when applied as a racemate. . Error
bars represent ± s.e.m. of experiments performed in triplicate. **P < 0.01, *P< 0.05, t
test.

Our control, DKP-1, did not show any significant inhibition of VEGF promoter
activity. Finally, we evaluated the activity of each of the individual enantiomer of bis-
ETP-1 (ent-1-bis-ETP-1 and ent-2-bis-ETP-1). Both the enantiomers showed identical
decrease at 100 nM concentration and the corresponding decrease by the racemate
confirm the dose dependence (Figure 2.22). These results also correlate well with our
81
SPR studies (Block, Wang et al. 2009), suggesting that both enantiomers have similar
binding affinity and activity in cell-based assays.

2.12 Screening of small molecule library for novel inhibitors of HIF-1 pathway
The involvement of HIF-1 in multiple pathways responsible for tumor progression makes
HIF-1 a promising molecular target. In an effort to identify new inhibitors of hypoxia-
inducible pathway, we screened a library of small molecules in collaboration with Prof.
Neamati at the University of Sothern California, School of Pharmacy (Figure 2.23).

Figure 2.23 Pictorial presentation of screening of a library of small molecules in a
luciferase reporter assay.

For this project, the previously discussed cell-based luciferase reporter assay was
employed. MDA-MB-231 cells which stably express luciferase reporter construct under
the control of five repeats of hypoxia response element were plated and dosed with 110
compounds at 10 µM concentration for 24 h. Hypoxia was mimicked by placing the 96
82
well plates in an anaerobic pouch system (GasPak EZ) and the corresponding luciferase
activity was recorded by a luminometer. The data were normalized to total protein
concentration determined from the BCA assay.

Figure 2.24 Representative set of small molecules A-C identified from the library screen
that inhibit HIF-1α inducible promoter activity in a stably transfected MDA-MB-231-
HRE-Luc cell line. Cells were treated with compound A, B, and C at 10 µM
concentration. Error bars represent ±s.e.m of experiments performed in triplicate. ***P <
0.0001 Hypoxia was mimicked by hypoxic bag.
We have identified three potent inhibitors (A, B, and C) which have
downregulated the hypoxia inducible VEGF promoter activity (Figure 2.24). Compound
A downregulated luciferase activity by 55%, compound B showed downregulation by
more than 95% and compound C showed inhibition by 82%. Further, study of dose-
83
dependent response and detailed investigation of the mechanism of action of these
inhibitors is currently underway.
2.13 Conclusion
Targeting transcription factor - coactivator interactions by designed small molecules is an
attractive approach for chemical regulation of gene expression. In the first part of this
Chapter, we discussed design and synthesis of novel monomeric and dimeric ETPs
transcriptional antagonists that act as allosteric inhibitors of the interaction between HIF-
1  CTAD and p300/CBP CH1 domain (Figure 2.25).

Figure 2.25 Targeting hypoxia-inducible pathway by small molecules. Right: depicts
the inhibition of transcription factor-coactivator interaction by
epipolythiodiketopiperazine based molecules. Left and center: (the potential targets of
small molecules that inhibit the HIF pathway.

84
These compounds bind to their target p300/CBP CH1 domain with micromolar
and submicromolar affinities and disrupt the formation of CTAD HIF-1α-p300/CBP
complex in vitro. From our structural activity relationships studies, we found that
increase of the strain of the polysulfide bridge and introduction of the indole ring as a
side chain results in an increase of the binding affinity of the ETP toward the p300 CH1
domain and also enhances its activity in cell-based assays as an inhibitor of the HIF-1α
inducible transcription. Control mono-ETP-4 showed reduced binding affinity and also
low activity in cell-based assays. On the other hand, mono-ETP-5 binds to its target with
ten-fold higher affinity as compared to control, mono-ETP-4. DKP-1 showed no
significant decrease in the luciferase activity indicating that disulfide bridges are essential
for the biological activity of ETP based compounds. A comparative assessment of the
enantiomers of bis-ETP-1 suggested that stereochemistry of disulfide bridges does not
impact their binding affinity and activity in luciferase assays.
In other part of this Chapter we described a library screen of small molecules in a
luciferase reporter system in order to identify novel, potent inhibitors of the hypoxia-
inducible pathway. In the course of this screen we identified three inhibitors, compound
A, B, and C. Under hypoxic conditions treatment with compound B at 10 µM in MDA-
MB-231-HRE-Luc cell line showed inhibition below the normoxic levels which reflects
the non-specific cytotoxicity associated with this compound. To ensure the specificity, it
is important to conduct the cytotoxicity assays followed by dose response curves. In
keeping with the depicted signaling pathway, (Figure 2.25) small molecules can disrupt
the transcription, translation or increase the rate of degradation of HIF-1α, they can act as
agonist of the enzyme PHD2 or they might disrupt the interaction between HIF-1α and its
85
protein partner, HIF-1β/ARNT. Further mechanistic and in vivo studies of these
inhibitors are currently underway.

86

Chapter 3: Novel Hydrogen Bond Surrogates as Orthosteric Modulators of
Hypoxia Inducible Transcription

87
3.1 Helical protein interfaces and their critical role in transcription
For a long time surfaces of many proteins were not considered as druggable targets
mainly due to their large sizes and shallow contact surfaces. The lack of chemical
approaches that could be broadly applicable to disruption of any intracellular protein-
protein interaction also complicates the process of targeting these interfaces. However,
one cannot ignore the fundamental role they play in regulating a number of cellular
processes. In the complex machinery of a living cell, transcription factors are especially
critical, as they are involved in an intricate web of interactions with other partner proteins
and regulate critical downstream signaling pathways. The lack of fidelity of these
interactions is often associated with different disease states and thus, selective modulation
of protein-protein interfaces holds a great promise for the development of synthetic
inhibitors of protein function.
Proteins secondary structures, such as α-helices, play a crucial role in mediating
critical protein-protein interactions. Recently, Arora and coworkers analyzed the
interaction surfaces of multiprotein complexes available in the protein data bank (PDB).
Of these complexes, 62% possess helical protein interfaces. Out of this pool, 20% are
involved in the regulation of gene expression (Guarracino, Bullock et al. 2011).
Although proteins typically display large interaction surfaces, only a subset of few
residues (also called as hotspot residues) often bring a significant contribution toward the
free energy of binding during the complex formation. These hotspot residues were
determined by alanine mutation and defined as the residues that upon mutation decrease
the binding energy by a threshold value of ∆∆G ≥ 1 Kcal mol
-1
. Based on the ∆∆G
contributions, helix interface protein-protein interactions were classified in three
88
categories: 1) receptor possessing clefts for helix binding; 2) receptors with extended
surfaces requiring contact from two to five turn helices, and 3) receptors possessing
weaker interactions. Role of transcription factor is implicated in various fundamental
processes, for e.g., cell growth, cell proliferation, gene expression and others. Because of
the essential role these factors play in the disease progression, targeting multiprotein
complexes of transcription factors could become an interesting point of intervention for
the given signaling pathway. The p53-mdm2 and Notch-CSL-MAML1 complexes are the
examples of interactions targetable by the helix mimetics that could have an effect on
transcriptional machinery (Chen, Yin et al. 2005; Moellering, Cornejo et al. 2009).

3.2 Structural determinants of transcription factor HIF-1 and its coactivator p300
As discussed earlier, in the nucleus, HIF-1α interacts with its coactivator protein
p300/CBP and transactivates the expression of key hypoxia inducible genes involved in
tumorigenesis and cancer progression. The critical interaction primarily responsible for
the induction of HIF-inducible genes is between the CTAD domain of HIF-1α and CH1
domain of p300.
Although the contact surface area of these two proteins is quite extensive, a few
amino acid residues contribute considerably towards the favorable free energy of binding.
Interestingly, -solution structure of HIF-1α CTAD (786-826) and CH1 domain of p300
(323-423) reveals two short α-helical domains (Figure 3.1) from HIF-1α as key
determinants for its recognition by p300 (Freedman, Sun et al. 2002). Stable mimics of
the helical domains have potential for selective targeting of such protein interfaces as
they can make contact over a large surface area and interact with their cognate protein
89
with higher specificity. We reasoned, that stable mimics of these domains (αA and αB)
could potentially inhibit the interaction between HIF-1α and p300/CBP and further
downregulate the expression of hypoxia-inducible genes.

Figure 3.1 NMR structure of complex between C-terminal activation domain (CTAD) of
HIF-1α (Red) and cysteine-histidine rich region 1(CH1) of p300/CBP (Violet). CH1
domain has a triangular structure with three zinc ions (Green) present at the vertices of
the triangle. Structure was created from Pdb ID: 1L3E.

3.3 Approaches to stabilize or mimic α-helical fold
Generally α-helices function as a part of a large protein structure. Once excised from the
protein environment, they are unable to bind to their target with high affinity (Henchey,
Jochim et al. 2008). Peptides have poor cell permeability and also degrade quickly in the
blood stream by the action of proteases. Structural mimicking or preorganization of these
90
peptides into an -helical state could lead to the development of functional inhibitors.
Several approaches have been developed to either stabilize this conformation in peptides
or to mimic it with non-natural scaffolds (Figure 3.2) (Patgiri, Jochim et al. 2008). These
approaches are broadly divided into three main categories: 1) helical foldamers, 2) helical
surface mimetics, and 3) helix stabilization by cross-linking.

Figure 3.2 Different approaches to stabilize the α-helical conformation in peptides or
mimicking it with non-natural scaffolds. Some of the successful strategies are β-peptides
helices, peptoids helices, side chain cross-linked helices, terphenyl mimicking helical
domain. Violet spheres represent the amino acid side chains.

Helical foldamers use the non-natural scaffolds as attachment points for the side
chains. Examples include peptoids, foldamers: N-substituted polyglycine analogs and β-
91
peptides: composed of β-amino acid residues. β-peptides, which are designed by
mimicking the activation domain of p53, were shown to regulate the cellular levels of
p53, p21 and hdm2 (Harker and Schepartz 2009). Helical surface mimetics were
developed by mimicking the hydrophobic face of the α-helix. These mimetics were first
designed on a conformationally restricted scaffold of terphenyl with attached functional
groups that resemble i, i+4 and i+7 residues of the α-helix. Other helical surface mimetics
were also designed by using the skeletal structures of indanes, oligophenyls,
benzodiazepine and trisubstituted imidazoles (Davis, Tsou et al. 2007).
Helix stabilization is another attractive approach to stabilize the helical segments.
The standard strategy to stabilize the α-helical conformation is to introduce a covalent
bond between i and i + 4 or i and i + 7 side chain groups. For example, formation of a
disulfide bridge and lactam bridge (between the Lys and Asp or Lys and Glu amino acid
residues) at i and i + 4 position of the helix. Verdine and coworkers (Schafmeister, Po et
al. 2000) stapled the side chains present on the same face of the helix. This reduces the
maximum linker length between the two residues and makes the α-helix formation more
favorable.

3.4 Hydrogen bond surrogate (HBS) approach to stabilize short α-helices
At the protein-protein interaction surface, the average length of the helical domain is
quite small and spans around two to three helical turns (Barlow and Thornton 1988).
According to helix-coil transition theory, organization of three consecutive amino acids
into the helical orientation is an energetically demanding process and makes stabilization
of short peptides a challenging venture. In an α-helix, the hydrogen bond between the
92
C=O of i
th
residue and the NH of (i+4)
th
residue, stabilizes and nucleates the helix
structure. Preorganization of amino acid residues in an α-helical turn could decrease the
barrier of nucleation and initiate the helix formation. In this very interesting approach,
our collaborators at NYU proposed a novel method for the stabilization of short α-helical
segments. This method involves the replacement of the main chain hydrogen bond,
between the C=O of i
th
residue and the NH of (i+4)
th
residue, with a carbon-carbon
covalent bond and termed it as a hydrogen bond surrogate (HBS) α-helices (Figure
3.3).

Figure 3.3 Hydrogen bond surrogate (HBS) strategy involves the replacement of the
main chain hydrogen bond between i
th
and i+4
th
amino acid residues with a covalent
bond. Dashed lines represent the intramolecular hydrogen bonds between the C=O and
NH. Pink line represents the internal crosslink stabilizing the α-helix conformation.
Violet spheres represent the amino acid side chains.

HBS is an attractive strategy to stabilize α-helices and offers significant
advantages: 1) presence of an internal cross link does not block the solvent exposed
93
molecular recognition surfaces; 2) have all the side chain functionalities available for the
binding; 3) increased proteolytic resistance, and 4) preorganized α-turn reduces the
entropic cost of folding and thus is expected to bind to its partner with high affinity.
3.4.1 General synthesis of HBS helices
Synthesis of HBS helices is primarily comprised of two major steps; first, standard Fmoc
solid phase peptide synthesis and second, ring closing olefin metathesis reaction. These
syntheses were performed by our collaborators at NYU. The resin-bound bis-olefin is
synthesized by conventional Fmoc chemistry using rink amide or Knorr resin. The three
fundamental blocks for the synthesis of bis-olefin are Fmoc amino acids, N-allyl
dipeptide and 4-pentenoic acid. First, the 5 equivalents of appropriate Fmoc monomers
are activated by HBTU chemistry for 10 minutes and then coupled with the resin bound
free amine for 45 minutes. Next, N-allyl dipeptide with proper substitution of R-group is
introduced at the appropriate position and coupled with the resin bound peptide chain.
Finally to create another arm of the bis-olefin, 4-pentenoic acid is placed at the C-
terminal end of the polypeptide chain. After each coupling step, the efficiency of the
reaction is monitored by ninhydrin test and Fmoc was removed with 20% piperidine in
NMP (Figure 3.4). Eventually, after thorough washes with DMF and DCM bis-olefin is
dried under vacuum and stored at 4 °C. During the synthesis of HBS helices, the critical
and the challenging step is the formation of the macrocycle that nucleates the α-helices.
Although different approaches were explored for the synthesis of a macrocycle, factoring
the compatibility with almost all amino acids and solid phase chemistry, our collaborators
opted for the ring closing metathesis (RCM) reaction.
94

Figure 3.4 General scheme for the synthesis of hydrogen bond surrogate (HBS) based α-
helices. Orange spheres represent the resin used for solid phase peptide synthesis.

After a few attempts with different reaction conditions and varied RCM catalysts, it was
observed that second generation Grubbs catalyst with microwave irradiation would
provide the HBS helices in high yields and shorter reaction times (Chapman and Arora
95
2006). For the synthesis of HBS helices specifically used for this project, resin bound bis-
olefin (0.1 mmol) was treated with 15 mol% of Hoveyda-Grubbs II catalyst and irradiated
in CEM discover microwave reactor for 10 minutes with holding time of 5 minutes. Next,
reaction was cooled and resin was washed thoroughly with DCM, 10% 1,3-
bis(diphenylphosphino)propane in DCM & DCM respectively. The constrained α-helical
peptide was cleaved from the resin with 95% TFA: 2.5% triisopropylsilane : 2.5% water
for 90 min and then purified by HPLC. Purified product was analyzed by mass
spectrometry. Unconstrained peptides, the negative controls, were synthesized by the
standard Fmoc solid-phase peptide chemistry.
3.4.2 Structural characterization of HBS α-helices
Structural characterization of designed HBS helices was performed by 2D-NMR and
circular dichroism (CD) spectroscopy. CD spectroscopy is - readily used to analyze the
conformation of protein secondary structures, especially, α-helices.
The well-folded α-helices have a signature spectrum that not only confirm the
helical structure, but is also used to calculate the fraction of molecules present in -
helical conformation. Arora et al. studied the conformation of two designed HBS α-
helices, one 10-mer peptide derived from Bak-BH3 and another one from c-Jun coiled
coil domains (Patgiri, Jochim et al. 2008). They have designed both the constrained as
well as the unconstrained analog of Bak-BH3 10-mer sequence (QVARQLAEIY). CD
spectroscopy of both the analogs was recorded and upon comparison, the unconstrained
analog remain almost completely unstructured while the constrained HBS analog showed
the typical characteristic spectra of an α-helix with a double minima at 208 nm and 222
nm and a maximum at 190 nm with a 70% helical content (Figure 3.5).
96
Further structural characterization was done by a combination of 2D NMR
spectroscopic techniques. The constrained HBS analog of Bak-BH3 peptide exhibited
the medium range NOEs between dNN (i, i+1), dαN (i, i+1), dαN (i, i+3), dαN (i, i+4)
and dCαCβ (i, i+3).

Figure 3.5 CD spectra of constrained HBS α-helix and its unconstrained analog recorded
in 10% TFE in PBS. HBS helix curve (XQV*ARQLAEIY) is shown in red and
unconstrained analog AcQVARQLAEIY-NH
2
) is shown in black. .

Detection of NOE signals at the very last C-terminal residue supported the fact
that the stability in the helical structure achieved by HBS strategy is quite rigid (Figure
3.6a). The X-ray crystal structure of an HBS α-helix at 1.15 Å resolution also supported a
97
fully hydrogen-bonded short α-helix (Figure 3.6b). The chemical shift of an amide proton
is also indicative of the extent of hydrogen bonding. Analysis of the rate of exchange of
amide protons of constrained HBS helix revealed that almost all of these protons showed
the value of temperature coefficient more positive than -4.5ppb/°K, which is indicative of
hydrogen bonding. Of all residues, only valine which is present at position 2 showed the
temperature coefficient between 8 and 10 ppb/K suggesting no intramolecular hydrogen
bond for this amide proton.

Figure 3.6 A) NMR derived structureof Bak-BH3 HBS helix. Carbon (green), oxygen
(red), nitrogen(blue), Carbon-carbon covalent bond (magenta). B) X-ray crystal structure
at 1.15 Å resolution of Bak-BH3 HBS helix.

It is expected as valine at the N-terminus does not have any intramolecular
hydrogen-bonding partner. Conformational stability of HBS helices was analyzed by
98
monitoring the effect of increasing temperature on the CD spectra of the constrained
helix. With the increasing temperature, unwinding of the designed helix was evident from
the slow increase in the mean residue ellipticity at 222 nm. However, even at high
temperature of 85 °C, the constrained peptide retains the 60-70 % of its room temperature
helicity, indicating rigidity and thermal stability. In comparison to unconstrained peptide,
relatively slow rate of exchange (H/D exchange) of amide protons in constrained HBS
helix also implicates that these protons are protected from hydrogen bonding.
3.4.3 The potential of HBS helices as modulators of protein-protein interactions
The elegant approach of affording stable two to three turn α-helices by HBS strategy
seems to have a potential to probe specific protein-protein interactions involved in a
particular disease state. The concession in the entropic penalty gave a large advantage to
HBS helices to bind to their receptor with high affinity as compared to the unconstrained
peptide of the same sequence. With increased proteolytic stability in the constrained
helical conformation, HBS approach with intramolecular hydrogen bonding is also
expected to increase the cellular uptake of helical peptides by reducing the hydration of
amide bonds. Initially the potential of HBS helices to target specific protein-protein
interfaces was studied by designing HBS helices of the α-helical BH3 domain of Bak
(Wang, Liao et al. 2005). Bak is a proapoptotic protein whose α-helical BH3 domain
interacts with an antiapoptotic protein Bcl-xL and regulates the process of cell death.
BH3 domain upon dimerization interferes with the death suppressor function of Bcl-xL
and regulate the process of apoptosis (Holinger, Chittenden et al. 1999). The interface of
Bak BH3 and Bcl-xL was also explored by other approaches that utilize stabilized α-
helical fold. The α-helical mimetics of the BH3 domain designed by lactam-based side
chain cross-linking strategy were not able to bind to Bcl-xL. There is a possibility that the
99
side chain cross linking approach presents the steric clash and would not be able to reach
the site of action. However, fluorescence anisotropy studies on helices designed by HBS
approach showed that Bak BH3 HBS helix binds with Bcl-xL with high affinity (Figure
3.7), supporting fact that internal cross-link in HBS helices can access the deep
hydrophobic pocket of Bcl-xL which might not be possible with side chain constrained
approach.

Figure 3.7 Lactam bridged approach was not able to target the Bcl-xL and HBS based
Bak-BH3 α-helix reconized the Bcl-xL with high affinity. Blue ribbon represent the Bcl-
xL and red ribbon represent the Bak-BH3 peptide (Pdb ID 1BXL). This structure was
published by Fesik and coworkers.
HBS approach has been successfully applied and tested in a cell free, split-protein
reassembly assay to inhibit the p53-MDM2 interaction (Henchey, Porter et al. 2010).
Seven HBS helices were designed and synthesized by mimicking the p53 (17-31) helical
domain and compared with a modified unconstrained analog (AcEAFSDLWKLLPENNV).
Luciferase protein was splitted into two halves (N-Fluc and C-Fluc) and both halves were
Lactum bridged
stabilized α-helix
Bak-BH3 bound to Bcl-xL Bak-BH3 HBS α-helix
100
conjugated with protein partners. As readout of the interaction, the system will produce
light in the presence of luciferin, if the two split component proteins interact with each
other.. In the presence of an inhibitor, protein complex formation reduces the
bioluminescence. Modified p53 HBS, XQE*GFSDLWKLLS, is shown to be quite
specific in targeting the p53-MDM2 interaction.

3.5 The rationale behind the design of first generation of αA-helix mimics of HIF-1α
The solution structure of the HIF-1α and p300 shows that they share a common
hydrophobic core where two helices of the CTAD domain of HIF-1α, αA and αB, clamps
around the four α-helices of the CH1 domain of p300. The idea here is to rationally
design the HBS α-helical mimetics of the αA domain of HIF-1α, which can bind to its
target p300 with high affinity and thus act as orthosteric transcription modulator of HIF
pathway. We started our design by mimicking the αA domain of HIF-1α which
corresponds to a sequence of 9 amino acids,
796
TSYDCEVNA
804.
We have designed
several constrained HBS and the corresponding unconstrained peptides to analyze their
potential in downregulating the expression of hypoxia- inducible genes.
All HBS helices were designed based on the high-resolution structure of CTAD
domain of HIF-1α in complex with the CH1 domain of p300. Structural analysis of
CTAD-HIF-1α and CH1-p300 revealed that serine 797 does not play a critical role in the
binding. In order to facilitate the synthesis of helices, the first HBS helix was synthesized
as a direct mimic of the αA helix, except serine 797 was mutated to alanine, (HBS 1,
XTA*YDCEVNA-NH
2
). To ensure that the modified HBS helix is not disrupting the
binding properties of the complex, two unconstrained analogs (AcTSYDCEVNA-NH
2

101
and AcTAYDCEVNA-NH
2
) were synthesized by the solid phase peptide synthesis and
compared for the difference in the binding affinity. Cysteine 800 is considered as a
crucial residue for the binding but it was also observed that this cysteine targets a
hydrophobic pocket on p300, so replacing cysteine with a more hydrophobic residue
would lead to enhanced binding of the complex. Keeping this in mind, (HBS 2,
XTA*YDVEVNA-NH
2
) was synthesized by replacing cysteine 800 with valine.
3.6 Results of the first generation of αA-helix mimetics of HIF-1α
Aim of designing HBS helices is to determine their potential as orthosteric inhibitor of
hypoxia-inducible pathway and the very first proof of the appropriate design would be to
investigate the solution conformation of the designed helices.
Compound Sequence % Helicity K
d
, nM Transcription
inhibition, %
WT HIF-1αA
796
TSYDCEVNA
804
- - -
HBS 1 XTA*YDCEVNA-NH
2
44 540 ± 40 0
HBS 2 XTA*YDVEVNA-NH
2
16 690 ± 60 0
Peptide 1 AcTSYDCEVNA-NH
2
14 1350 ± 50 10
Peptide 2 AcTAYDCEVNA-NH
2
15 1220 ± 80 15

Table 1 Summary of results of first generation of αA-helix mimetics of HIF-1α. X
represents the pentenoic acid residue in the HBS macrocycle and * represents the
macrocycle linkage by a carbon-carbon covalent bond.
The increase in the helicity of the HBS compounds was determined by the CD
spectroscopy. All constrained and unconstrained peptides were dissolved in 10% TFE in
PBS at pH 6.3. HBS 1 showed increased percent helicity of 44% and as expected the
unconstrained analog of HBS 1 was only 15% helical. To confirm, that the replacement
102
of serine 797 residue with alanine is not perturbing the binding affinity, K
d
values of two
unconstrained analogs (AcTSYDCEVNA-NH
2
/peptide 1 and AcTAYDCEVNA-
NH
2
/peptide 2) were determined and compared. Both peptides exhibited comparable
binding affinity in an isothermal titration calorimetry experiment (Table 1). On the other
hand, HBS 1, which is significantly more helical than its unconstrained analog showed a
K
d
value of 540 nM towards p300 while HBS 2, binds to p300 with slightly lower affinity
indicating valine is not a proper replacement of cysteine. Although HBS 1 and HBS 2,
both were able to show considerable binding to p300, each of these HBS helices were
failed to inhibit the transcription of HIF inducible genes in a luciferase reporter assay.

3.7 Rationale behind the design of the second generation of αA-helix mimics of HIF-
1α
The failure of the first generation of HBS helices to downregulate the expression of HIF
inducible genes in cell culture led us to speculate that these compounds are unable to
cross the cell membrane as they have overall negative charge at physiological pH and cell
penetrating peptides are often rich in cationic residues (Schmidt, Mishra et al. 2010).
Based on this hypothesis, we decided to design our second generation of HBS αA-helix
mimic (Table 2) by reducing the overall negative charge (Henchey, Kushal et al. 2010).
HBS 3 was designed and its sequence was derived from the parent αA-helix
(
796
TSYDCEVNA
804
). Mutagenesis data suggest that cysteine-800 and asparagine-803
are very important for binding. Further structural analysis revealed that serine-797 and
tyrosine-798 are not directly involved at the binding surface; these amino acids were
replaced by alanine to facilitate the synthesis of HBS 3, XTA*ADCEYNA-NH
2
. Valine
residue at 802 position is not expected to be involved in binding and that is why it was
103
mutated to tyrosine residue. Tyrosine and tryptophan residues are generally included for
the determination of HBS concentrations by UV-Vis. The next helix HBS 4 was designed
by keeping the same sequence as of HBS 3 except it contains a C-terminal arginine
residue (HBS 4, XTA*ADCEYNAR-NH
2
). Incorporation of arginine residue decreases
the overall negative charge on the peptide which is expected to enhance the permeability
of HBS peptide.
Compound Sequence
WT HIF-1αA
796
TSYDCEVNA
804

HBS 3 XTA*ADCEYNA-NH
2

HBS 4 XTA*ADCEYNAR-NH
2

HBS 5 XTA*ADREYNAR-NH
2

Peptide 3 AcTAADCEYNAR-NH
2

Table 2 Second generation of αA-helix mimics of HIF-1α. . X represents the pentenoic
acid residue in the HBS macrocycle and * represents the macrocycle linkage by a carbon-
carbon covalent bond. WT is the wild-type sequence of αA helix.
Placement of arginine is also expected to increases the helical content of the HBS
because of the stabilization the helix macrodipole and potential formation of the ionic
interaction between i and i+4 side chain group of arginine and glutamic acid residues.
Next, a negative control HBS 5, XTA*ADREYNAR-NH
2
was synthesized by replacing
cysteine-800, which is known to be quite critical for binding and is present in the
hydrophobic pocket. To verify the proof of principle, an unconstrained analog (Peptide 3)
of HBS 4 was also synthesized.
104
3.8 Characterization of second generation of αA-helix mimics of HIF-1α HBS by
circular dichroism spectroscopy
The solution conformation of designed HBS helices (HBS 3, HBS 4 and HBS 5) and of
unconstrained peptide (peptide 3) was analyzed by circular dichroism (CD) spectroscopy.
CD spectroscopy is a technique which measures the difference between the absorption of
the right and left circularly polarized light by an asymmetric molecule as a function of
wavelength. This technique is very sensitive and helpful in the determination of
secondary structure of polypeptides and proteins. Secondary structures such as, α-helix,
β-sheet and random coil each give rise to a distinct type of characteristic CD spectra
measured in the far UV region of 170 to 250 nm.

Figure 3.8 Circular dichroism spectra of designed HBS α-helices and unconstrained
peptide in 10% TFE in PBS. HBS 4 (53%) and HBS 5 (51%) showed enhancement in the
helicity with characteristic two minimas at 208 nm and 222 nm.

HBS helices and peptide 3 were dissolved in 10 % TFE in PBS at a concentration
of 50 to 100 µM. Wavelength scans were obtained at 1 nm increment at 25 °C with an
105
integration time of 2 s. The CD spectra obtained from all the designed HBS helices is
consistent with the signature spectra of an α-helix displaying double minima at 208 and
222 nm and a maxima at 190 nm. The helix content of each peptide was determined from
the mean residue ellipticity at 222 nm, [θ] 222 (deg cm2 dmol-1) corrected for the
number of amino acids. As expected, all the HBS helices were more helical then their
unconstrained analog, HBS 4 is more helical (53%) than HBS 3 (40%) (Figure 3.8).

Compound Sequence % Helicity
WT HIF-1αA
796
TSYDCEVNA
804
-
HBS 3 XTA*ADCEYNA-NH
2
40
HBS 4 XTA*ADCEYNAR-NH
2
53
HBS 5 XTA*ADREYNAR-NH
2
51
Peptide 3 AcTAADCEYNAR-NH
2
15

Table 3 Tabular presentation of percentage helicity of second generation of HBS α-helix
mimics as measured by circular dichroism spectroscopy.

The increased helicity of HBS 4 than HBS 3 seems to be an effect of salt bridge
formation between arginine and glutamic acid residue. On the other hand, peptide 3
remains largely unstructured. This data supports our design firmly, according to which
placement of an internal crosslink confers high degree of preorganization in HBS helices
and provides a stable α-helical structure.

106
3.9 Affinity of the second generation of an αA-helix mimics toward p300 CH1 domain
Before undergoing any rigorous biological characterization of αA-helix mimetics of HIF-
1α it is important to validate the design on the basis of its thermodynamic binding
properties toward its target, p300 CH1 domain. We conducted an isothermal titration
calorimetry (ITC) experiments with HBS helices and unconstrained peptide to determine
the direct binding of compounds to the CH1 domain of p300. ITC is a powerful
technique that can directly and precisely measure the binding energetics of protein-
protein, protein-ligand and other biologically relevant interactions. The basic principle
behind this technique is that an amount of heat evolved or absorbed when a ligand or
small molecule binds to its target. The heat difference is then used to calculate the
thermodynamic parameters, such as, binding constants, Gibbs free energy, enthalpy and
entropy of a bimolecular interaction. Before the addition of the ligand, both reference cell
and sample cell were maintained at constant temperature difference which is generally set
close to zero. Each injection of the ligand solution to the molecule elicits the binding
reaction and evolves or absorbs heat depending on the binding affinity and ligand
concentration. Addition of each ligand causes a difference in the temperature of reference
cell and sample cell and the feedback power then again equilibrate the difference. Once
the macromolecule in sample cell becomes saturated, the heat signal diminishes and
attains value close to the background value.
For this experiment, GST tagged p300-CH1 protein was cloned, expressed and
purified as discussed in the previous chapter. HBS helices at a concentration of 100 µM
were titrated into a solution containing the GST-p300-CH1 (10 µM) and the heat
generated or absorbed during the binding reaction is measured as a function of HBS helix
107
concentration. First injection of 10 μL was followed by 20 injections of the same volume
until a molar ratio of 2.5 was obtained.

Figure 3. 9 A) ITC raw data of titration of HBS 4 into a solution of GST-p300-CH1 B)
Integration of the signals, showing binding of HBS 4 to p300 CH1 domain.
108
Each peak represents the amount of heat evolved during the formation of complex
between p300-CH1 and HBS 4. Integration of the baseline-subtracted thermograms
yielded binding isotherms that were fit to a model of one-site interaction. Addition of the
HBS helices (HBS 3, HBS 4, HBS 5) and peptide 3 showed that HBS 4 binds to its target
protein, p300-CH1 domain, at a submicromolar affinity with K
d
value of 420 nM (Figure
3.9) while HBS 3 binds with lower affinity with Kd value of 950 nM. This data suggested
that incorporation of arginine is not detrimental to the binding of the HBS 4 towards
GST-p300-CH1. As expected, HBS 5 with cysteine mutated to an arginine is a very weak
binder and is supporting cysteine as an important residue at the binding interface. On the
other hand, peptide 3 which is the unconstrained analog of HBS 4, binds with very low
affinity as compare to the HBS 4 (Table 4).

Compound Sequence % Helicity K
d
, nM
WT HIF-1αA
796
TSYDCEVNA
804
- -
HBS 3 XTA*ADCEYNA-NH
2
40 950 ± 90
HBS 4 XTA*ADCEYNAR-NH
2
53 420 ± 35
HBS 5 XTA*ADREYNAR-NH
2
51 >>>2200
Peptide 3 AcTAADCEYNAR-NH
2
15 825 ± 50

Table 4 Tabular summary of key biophysical data of second generation of HBS α-helix
mimics. X represents the pentenoic acid residue in the HBS macrocycle and * represents
the macrocycle linkage by a carbon-carbon covalent bond.

After analyzing the CD and ITC results, we decided to proceed with HBS 4 as our
lead compound for its further biological evaluation.
109
3.10 Cytotoxicity of HBS 4 and peptide 3 in HeLa cells
Prior to the evaluation of HBS 4 in a cell culture assay, it is important to ensure that the
observed biological effect is specific and not due to the inherent toxicity of the
compound. We carried out the cell viability assay by using trypan blue stain. This dye is
used to differentiate between the viable and non-viable cells. When the cell membrane is
compromised or damaged by the action of certain agents then trypan blue traverse
through the membrane of the dead cells and stain it blue while is not absorbed by the
healthy live cell.

Figure 3.10 Cell density and cell population doubling assay, performed with HBS 4,
peptide 3 at 1 µM concentration and with chetomin at 200 nm for a period of 72h.
Control represents only media and vehicle represents media with 0.1% DMSO.

110
In this assay, HeLa cells were plated at the density of 35000 cells/ well in a 24-
well plate and after the monolayer formation, treated with HBS 4, peptide 3, at 1 µM
concentration, vehicle (media with 0.1 % DMSO), control (only media) and chetomin at
200 nM concentration. The cell density and the rate of population doubling were
monitored for up to 72 h. Unlike chetomin, which shows a continuous decrease in the
number of viable cells with increase in time, HBS 4 and peptide 3 does not showed
significant cytotoxicity (Figure 3.10). Number of viable cells upon treatment with HBS 4
and peptide 3 remain quite comparable to that of vehicle and control for the mentioned
period of time.

3.11 Transcriptional regulation of HIF-1α genes in HeLa cells treated with HBS 4,
Peptide 3 and chetomin
Next, we evaluated the potential of HBS 4 to downregulate the transcription of HIF
inducible genes. Angiogenesis is an important process in tumor progression; it supports
the process of dissemination of cancer cells from one tissue to another. It also helps
cancer cells in transiting from cellular multiplication state to an uncontrolled proliferating
state. Vascular endothelial growth factor (VEGF) is a potent mitogen which is primarily
responsible for the initiation and ongoing process of angiogenesis. Inhibition of VEGFA
with a monoclonal antibody targeting VEGFA showed almost complete suppression due
to decreased vascular density (Roskoski 2007). VEGFA genetic deletion in intestinal
epithelial cells also shows significant reduction in tumor growth (Korsisaari, Kasman et
al. 2007).
Hypoxia stimulates angiogenesis through the binding of hypoxia inducible factor
to the hypoxia response element (HRE) in the promoter region of the VEGF gene.
111
Considering the significance of VEGF in tumor growth and progression, we decided to
measure the mRNA level upon treatment with HBS helices under both normoxic and
hypoxic conditions. HeLa cells were incubated with designed HBS 4, Peptide 3 at 1 µM,
Chetomin at 200 nM and vehicle for total of 12 h and 24 h.

Figure 3.11 Inhibition of VEGF mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 4 (1 µM), peptide 3 (1 µM) and chetomin (200 nM) for A)
12h, and B) 24 h under both normoxic and hypoxic conditions. Hypoxia was mimicked
with 300 µM DFO. Error bars represent ± s.d. of experiments performed in triplicate.
*P< 0.05, t test.

Hypoxia was mimicked by adding 300 µM of desferroxamine (DFO). DFO is an
iron chelator that coordinates with the Fe
2+
present in the PHD2 and other iron-containing
enzymes and thus inactivates their function. The cells were incubated for 6 h at 37 °C
and 5% CO
2
and then hypoxia was induced with DFO and cells were incubated for
112
another 6 h or 18h. Total RNA was isolated by RNeasy kit (Qiagen) and reverse
transcribed by using oligo-(dT)
18
, random hexamers and Superscript III reverse
transcriptase. In HeLa cells, the induction of VEGF gene under hypoxia is about 3 fold.
Treatment of cells with HBS 4 at 1 µM downregulates the expression of the VEGF gene
at levels comparable to that of chetomin. Results after 24 h also showed similar decrease
of roughly 50% (Figure 3.11). As per our expectation, the unconstrained peptide 3 was
not able to achieve any significant reduction in expression, which is consistent with the
fact that the unconstrained peptide is more prone to the action of proteases and thus is not
stable enough under the cell culture conditions (Table 5).
Compound Sequence % Helicity Kd, nM Transcription
inhibition, %
WT HIF-1αA
796
TSYDCEVNA
804
- - -
HBS 3 XTA*ADCEYNA-NH
2
40 950 ± 90 0
HBS 4 XTA*ADCEYNAR-NH
2
53 420 ± 35 45
HBS 5 XTA*ADREYNAR-NH
2
51 >>>2200 2
Peptide 3 AcTAADCEYNAR-NH
2
15 825 ± 50 8

Table 5 Summary of results of biophysical and transcriptional inhibition assays of
second generation of HBS αA helices.

Next, we looked at the level of another important hypoxia inducible gene, Glut1.
Under hypoxic conditions, overexpression of Glut1 promotes the transport of the glucose
and shifts the metabolism of hypoxic cells from oxidative phosphorylation to aerobic
glycolysis. This adaptation of cancer cells is y responsible for the high survival rate and
also provides these cells with growth advantage under poorly oxygenated conditions.

113

HBS 4 showed significant inhibition of the transcription of Glut1 at 1 µM
concentration, whereas peptide 3 had negligible effect (Figure 3.12).

Figure 3.12 Inhibition of Glut1 mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 4(1 µM), peptide 3(1 µM) and chetomin (200 nM) for A)
12h and B) 24 h under both normoxic and hypoxic conditions. Hypoxia was mimicked by
300 µM DFO. Error bars represent ± s.d. of experiments performed in triplicate. *P<
0.05, t test.

Similar results were obtained after 24 h with almost 50% reduction in the mRNA levels
of Glut-1. The levels of β-glucuronidase were used as an endogenous control, as it
remains unchanged during experimental conditions. These results have shown that the
designed HBS helices (mimicking αA sequences) are targeting the interaction between
HIF-1α and p300 and were able to reduce the expression of HIF inducible genes. A few
114
small molecules, such as chetomin and other ETP based transcriptional modulators, are
shown to inhibit the interaction of HIF-1α and p300 by disrupting the global fold of p300.

Figure 3.13 Effect of HBS 4 on the global fold of p300-CH1 domain. HBS 4 (10 µM)
was added to 2.5 µM solution of p300-CH1 protein in Tris buffer (50 mM Tris, 150 mM
NaCl, 100 uM ZnCl2, 1mM DTT, 10% glycerol, 0.2% NP-40) at pH 8.0. The solid black
line represents the CD spectra of GST-p300-CH1 and dashed black line represents the
CD spectra of GST-p300-CH1 with HBS 4.
To exclude this possibility, we also checked the effect of HBS 4 on p300 structure
by CD (Figure 3.13). No significant change in CD spectra of p300 was observed,
suggesting that HBS 4 does not denature the folded structure of p300.

3.12 The rationale behind the design of αB-helix mimics of HIF-1α
The CTAD domain of HIF-1α is comprised of two helices, αA and αB. Both of these
helices interact strongly with the CH1 domain of p300 and stabilize the complex; HIF-1α
115
remains unstructured in the absence of properly folded CH1 domain. Inspired by our
results with αA helices, and based on in silico modeling data we decided to explore the
possibility of the use of αB helix mimetics in the disruption of the HIF1 /p300 complex.

Compound Sequence
WT HIF-1αB
817
ELLRALDQ
824

HBS 6 XEL*ARALDQ-NH
2

HBS 7 XEL*ARAADQ-NH
2

HBS 8 XEL*ARALDQR-NH
2

HBS 9 XEL*ARALDQC-Flu
Peptide 4 AcELARALDQ-NH
2

Table 6 Tabular presentation of all designed αB HBS helices. WT is the wild type
sequence of αB helix of HIF-1α, X represents the pentenoic acid residue in the HBS
macrocycle and * represents the macrocycle linkage by a carbon-carbon covalent bond

The αB sequence of HIF-1α corresponds to the sequence of 8 amino acids
817
ELLRALDQ
824.
Mutagenesis and structural data suggested that leucine 818, leucine
822 and glutamine 824 of HIF-1α are the three critical residues required for the binding
of p300/CBP. Keeping the structural details in mind, first HBS was synthesized by
replacing leucine 819 by alanine (HBS 6, XEL*ARALDQ) as it was not interacting
directly with the coactivator surface. An unconstrained analog of HBS 6 (peptide 4,

AcELARALDQ) was also synthesized for the control studies (Table 6). A negative
control was synthesized by replacing leucine 822 by alanine (HBS 7, XEL*ARAADQ)
116
and another HBS was synthesized by placing an arginine at the C-terminal end (HBS 8,
XEL*ARALDQR). A fluorescent analog (HBS 9-flu, XEL*ARALDQC-Flu) was also
synthesized for binding and cellular uptake studies. General scheme of synthesis of αB-
helices is shown in appendix (Appendix B, Figure B.1).

3.13 Characterization of secondary structure of a small library of HBS αB mimics of
HIF-1α
HBS mimics of αB sequence (ELLRALDQ) were synthesized by following the same
synthetic strategy as that of αA. The solution conformation of HBS αB mimics was
investigated by CD spectroscopy.

Compound Sequence % Helicity
WT HIF-1αB
817
ELLRALDQ
824
-
HBS 6 XEL*ARALDQ-NH
2
23
HBS 7 XEL*ARAADQ-NH
2
32
HBS 8 XEL*ARALDQR-NH
2
29
HBS 9 XEL*ARALDQC-Flu -
Peptide 4 AcELARALDQ-NH
2
8

Table 7 Change in the percentage helicity of designed HBS αB helices and unconstrained
peptide 4.

HBS 6, HBS7, HBS 8 and peptide 4 were dissolved in 10% TFE in PBS at a
concentration of 100 µM and CD spectrum was recorded as a function of wavelength. As
117
expected, designed HBS helices were more helical than their unconstrained analog. The
optimized helices were calculated to be 23 to 30 % helical, while the unconstrained
analog is essentially unstructured with 8% helicity (Table 7).

3.14 HBS αB helices disrupt the HIF-1α/p300 complex in vitro
HBS helices were designed as the orthosteric inhibitors of HIF-1α and p300 interaction.
For testing their ability to disrupt the complex between these two proteins we employed a
fluorescence polarization assay (FP). For this assay, as described earlier in the Chapter 2,
we first measured the interaction between the CTAD domain of HIF-1α and CH1 domain
of GST-tagged p300.

Figure 3.14 Saturation binding curve for GST-p300-CH1 and HIF-1α-flu (786-826) as
measured by the fluorescence polarization assay. X-axis is labeled as a log concentration
of p300-CH1 and Y-axis is showed the % bound fluorescein labeled HIF-1α.
118
The K
d
value obtained from the saturation binding curve was then used to
determine the concentration of GST-p300-CH1 required for the competition assay. For
the saturation binding assay, the fluorescently tagged CTAD domain of HIF-1α (HIF-1α-
flu CTAD) was maintained at a constant concentration of 15 nM while concentration of
other protein (GST-p300-CH1) varied in the concentration range between 1 - 2000 nM.
In order to avoid the crosstalk between wells and to reduce the background, this assay
was carried out in a 96-well black opaque plate. The experiment was conducted in
triplicates with each triplicate having constant concentration of HIF-1α-flu and increasing
concentration of GST-p300-CH1. Next, the plate was incubated for 1 h at room
temperature and polarization values were recorded by BioTek Synergy HT microplate
reader. The K
d
value of 55 nM was obtained from the saturation binding curve (Figure
3.14). To increase the dynamic range of the absolute value difference, we chose to use a
little higher concentration of GST-p300-CH1 for the FP-based competition assays.
Next, we evaluated the ability of designed HBS αB helices to disrupt the complex
between HIF-1α and p300. This experiment was conducted by maintaining the constant
concentration of both partner proteins, but varying the concentration of HBS helices.
Complex of HIF-1α-flu CTAD (15 nM) and GST-p300-CH1 (75 nM) was preincubated
for 1 h at room temperature in a 96-well black opaque plate. After 1 h different
concentrations of HBS helices were added to the preincubated complex and further
incubated for 1 h at room temperature. HBS 6 was varied in a concentration range of 0 to
1 mM, while HBS 7 and HBS 8 were varied from 0 to 100 µM. Log of the HBS helices
concentration on the X-axis was plotted against the percentage change in the polarization
119
value on the Y-axis. Chetomin was used as a positive control for each experiment and, as
shown in Figure 3.15D, chetomin does disrupt the complex with an IC
50
value of 633 nM.

Figure 3.15 Fluorescence polarization competition assay showing disruption of CTAD
domain of HIF-1α and CH1 domain of p300 by A) HBS 6, B) HBS 8, C) Peptide 4 and
D) Chetomin (CTM). The unconstrained peptide was unable to disrupt the complex.
HBS 6 was able to disrupt the complex with a K
d2
value of approx. 200 µM (Figure
3.15A, while HBS 8 showed the disruption of the complex with a K
d2
value of 8 µM
(Figure 3.15B). As expected, both HBS 7, which was designed as a negative control by
120
replacing a very important leucine residue, and peptide 4 (the unconstrained analog of
HBS 6) were not able to bind or disrupt the complex in the concentration range tested
(Figure 3.15C)(Figure 3.16). For all FP assay, same buffer was used as discussed in
chapter 2.

Figure 3.16 Graph showing combined FP competition assay results with designed αB
helices (HBS 6, HBS 7, HBS 8 and peptide 4).

3.15 Impact of GST tag on the binding affinities of HBS αB helices in fluorescence
polarization assays
We were a little surprised by observing a large difference in the K
d2
values for HBS 6 and
HBS 8 as these two sequences are having a difference of only one residue at the C-
terminal end. For all previous FP based competition assays, we have used the GST tagged
121
p300-CH1 domain. In order to ensure that the effect we are observing in K
d
values is not
because of the presence of GST tag, we decided to analyze the outcome of p300-CH1
domain without GST on the binding affinities of HBS helices. The fusion protein has a
thrombin cleavage site, LVPRGS, between GST and CH1 domain of p300. As per
manufactures instructions, GST tag was removed by adding 15 U of thrombin to the
protein immobilized on the glutathione sepharose beads.

Figure 3.17 Saturation binding curve for GST removed p300-CH1 domain and HIF-1α-
flu (786-826) as measured by fluorescence polarization assay. X-axis labeled as log
concentration of p300-CH1 and Y-axis showed the % bound fluorescein labeled HIF-1α.
Next, a saturation binding assay was performed to calculate the concentration of
the GST free p300-CH1 domain to be used in the competition assays. HIF-1α-flu CTAD
(15 nM) was incubated with increasing concentration of p300-CH1 in a concentration
range of 0 to 6000 nM. The K
d
value obtained from the saturation binding curve was 520
122
nM (Figure 3.17),, which indicates weaker interaction between p300-CH1 and CTAD
HIF-1α as compared to GST-tagged p300 CH1 domain. This can be explained by
assuming that GST tag might be facilitating the global fold and proper conformation of
the p300-CH1 domain and thus making a higher-affinity complex with a low K
d
value of
55 nM.
For FP based competition assays, p300-CH1 was used at a concentration higher
than 520 nM, but lower than the saturation value. As discussed above in section 3.14, FP
completion assay was conducted by incubating the complex of HIF-1α-flu CTAD (15
nM) and p300-CH1 (800 nM) for 1 h at room temperature,then adding the increasing
concentration of both HBS 6 and HBS 7 ( from 0 to 100 µM).

Figure 3.18 Fluorescence polarization competition assay showing the disruption of
GST-cleaved CH1 domain of p300 and CTAD domain of HIF-1α with A) Chetomin and
B) HBS 8.
Chetomin was also tested under these conditions which showed an IC
50
value of
145 nM (Figure 3.18). However, no difference in the K
d
values of HBS 6 and HBS 8 was
observed within the entire concentration range tested for this assay (Figures 3.18 and
123
3.19). One explanation for this behavior could be that HBS helices are binding at such a
position on p300-CH1 domain which has no influence of GST tag. On the other hand, in
absence of GST tag, chetomin occupies a position that was earlier masked by the GST
tag. Also, chetomin is a small molecule which has a different mode of action and
allosterically disrupts the complex between CTAD domain of HIF-1α and p300 CH1
domain.

Figure 3.19 Graph showing results from combined FP competition assay with designed
αB helices (HBS 6, HBS 8) in presence of the complex of GST cleaved p300 and CTAD
domain of HIF-1α.

3.16 Determination of direct binding of HBS 9 to GST-p300-CH1 by fluorescence
polarization assay

As these helices are designed as orthosteric inhibitors of HIF-1α-CTAD and p300-CH1
interaction by mimicking the CTAD domain of αB of HIF-1α, they are supposed to bind
124
at the CH1 domain of p300. In order to study the direct binding of these helices, we
designed a fluorescent analog (HBS 9-flu) which had the same sequence as that of HBS 6
except for a cysteine residue at the C-terminal end for the attachment of fluorescein. For
FP assays, HBS 9-flu was used in place of HIF-1α-flu CTAD at a constant concentration
of 30 nM.
Figure 3.20 Direct binding of HBS 9-flu to the CH1 domain of p300 as determined by
the fluorescence polarization assay.
Experiment was conducted in triplicates and each set was treated with an
increasing concentration of GST-p300-CH1 ranging from 0 to 200 µM. Complex of HBS
9-flu and GST-p300-CH1 was incubated for 1 h at room temperature and then
polarization values were recorded by BioTek Synergy HT plate reader. Log concentration
125
of GST-p300-CH1 was plotted against the percentage change in the polarization value
and as shown in Figure 3.20 HBS 9-flu showed a Kd value of 80 µM.
3.17 Cytotoxicity of HBS 6 and Chetomin in HeLa cells
Although from the previous observation with αA mimetics toxicity studies, it can be
anticipated that these helices are not cytotoxic within the range of biologically relevant
concentrations, but it is important to evaluate the toxic effects of the sequence
particularly involved in the αB mimics. First, a cell count study experiment was
performed to optimize the number of cells required for plating.

Figure 3.21 Cell viability in HeLa cells as determined by MTT assay. Cells were treated
with A) Chetomin (CTM) and B) HBS 6. CTM showed IC
50
of 150 nM and no
cytotoxicity was observed with HBS 6 at concentrations up to 100 µM.

Next, HeLa cells were incubated with HBS 6 at a concentration range of 1 to 100
µM and chetomin with maximum concentration of 2 µM. No cytotoxicity was observed
in MTT cytotoxicity assay after 24 h treatment with HBS 6 at concentrations up to 100
µM (Figure 3.21B) whereas chetomin showed toxicity to cells with a GI
50
value of 150
nM (Figure 3.21A). Based on the GI
50
value obtained from the MTT assay, we decided to
126
measure the effect of HBS helices on hypoxia-inducible transcription in HeLa cells
within the concentration range of 1 to 50 µM, with more emphasis on pharmacologically
relevant concentration of 10 µM.

3.18 Transcriptional regulation of HIF pathway by designed HBS αB helices
After analyzing the potential of αB helices in biophysical assays, we next investigated
their impact on the modulation of HIF inducible transcription in cell culture assays. As
with αA helices, we tested αB mimics in HeLa cells.

Figure 3.22 Inhibition of VEGF mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6 (10 & 50 µM), HBS 7 (1 & 50 µM) and peptide 4 (10
µM and 50 µM ) 24 h under both normoxic and hypoxic conditions. Hypoxia was
mimicked by 300 µM DFO. Hela cells were supplemented with 10% serum. Error bars
represent ± s.e.m of experiments performed in triplicate. **P< 0.01,*P< 0.05, t test.

127
Considering the complex nature of transcriptional regulation, we tested the efficacy of αB
helices under different sets of conditions. In first set of experiments, HeLa cells were
maintained in 10% serum and treated with designed αB helices for 24 h.
It is well known that almost all solid tumors need high vascularization and out of
all proangiogenic factors, VEGF is considered as most powerful mitogen responsible for
the induction of blood vessels for continuous supply of oxygen and nutrients for the
constant growth and development of a tumor. We used quantitative real time PCR (qRT-
PCR) to determine the relative mRNA levels of VEGF. To observe the dose response,
cells were treated with two different concentrations of HBS 6, HBS 7 and peptide 4. Six
hours after dosing, hypoxia was induced by DFO and cells were incubated for further 18
h at 37 °C and 5% CO
2
. Under these experimental conditions, HeLa cells showed
approximately 3-fold induction of VEGF gene with DFO. HBS 6 at 10 µM effectively
downregulated the levels of VEGF mRNA and it has also shown a dose response at 50
µM with almost 60% reduction (Figure 3.22). This was a surprising observation, as this
helix was not that effective in our biophysical protein-protein interactions assays. This
result suggested that in a biological system, in the presence of other interactions, the
global fold of p300 CH1 could be different from the one observed in in vitro assays. With
all complex pathways interacting at multiple levels, it is very difficult to predict the
biological potential of molecules based on only in vitro assays. HBS 7, which was a
negative control of HBS 6 with an important substitution of leucine residue, was not able
to show any substantial degree of inhibition of the VEGF transcription, though there was
a slight decrease at 50 µM, but that was not significant. As expected, the unconstrained
control peptide 4 was completely inactive. HBS 8, which has arginine at the C-terminal
128
end and disrupted the complex between HIF-1α and p300 with K
d2
value of 8 µM,
showed only minimal downregulation of VEGF levels (Appendix B, Figure B.2).

Figure 3.23 Inhibition of Glut1 mRNA levels as measured in HeLa cells by qRT-PCR.
Cells were treated with HBS 6 (10 & 50 µM), HBS 7 (1 & 50 µM) and peptide 4 (10
µM and 50 µM ) 24 h under both normoxic and hypoxic conditions. Hypoxia was
mimicked by 300 µM DFO. HeLa cells medium was supplemented with 10% serum.
Error bars represent ± s.e.m. of experiments performed in triplicate. **P, 0.01, *P< 0.05,
t test.

Next, we measured the effect of these helices on another very important gene
involved in the cancer cells metabolism , Glut1. HBS 6 showed the similar amount of
downregulation as was observed with VEGF (Figure 3.23). No significant reduction was
observed with HBS 7 and peptide 4 at 10 µM. With HBS 8 only a minimal decrease was
observed (Appendix B, Figure B.3). Lysyl oxidase (LOX) is another HIF target gene
129
which recently has been shown to have a critical role in breast cancer metastasis. In
HeLa cells ~13fold induction of LOX was observed under hypoxia conditions.

Figure 3.24 Inhibition of LOX mRNA levels in HeLa cells as measured by qRT-PCR.
Cells were treated with HBS 6 (10 & 50 µM), HBS 7 (1 & 50 µM) and peptide 4 (10
µM and 50 µM) 24 h under both normoxic and hypoxic conditions. Hypoxia was
mimicked by 300 µM DFO. HeLa cells were maintained ain a medium supplemented
with 10% serum. Error bars represent ± s.e.m. of experiments performed in triplicate.
***P<0.0001 **P, 0.01, t test.

HBS 6 exhibited drastic decrease in the LOX mRNA levels with dose response (Figure
3.24). On the other hand, HBS 7 and peptide 4 were not able to downregulate LOX
mRNA levels to any substantial extent. With LOX gene, HBS 8 showed slight decrease,
130
approximately 30 % (Appendix B, Figure B.4) at 50 µM, but no measurable decrease at
10 µM.
Recently, hypoxic dependent activation of CXCR4 is being shown in several
cancers and direct correlation between HIF-1α and CXCR4 has also been established. It is
a chemokine receptor and is specific for SDF-1 ligand.

Figure 3.25 Inhibition of CXCR4 mRNA levels in HeLa cells as measured by qRT-PCR.
Cells were treated with HBS 6 at 10 µM for 24 h under both normoxic and hypoxic
conditions. Hypoxia was mimicked by 300 µM DFO. HeLa cells were maintained in a
medium supplemented with 10% serum. Error bars represent ± s.e.m. of experiments
performed in triplicate. **P, 0.01 t test.

The overexpression of CXCR4 is reported in a number of malignant cell types
such as breast, colon, prostate, pancreatic and other cancers. Upregulation of this

131
chemokine receptor on tumor cells facilitates the migration of cancer cells towards the
organs or tissues that secrete high amount of SDF-1. Blockade of CXCR4/SDF-1 axis has
been shown to prevent the migration and invasion of pancreatic cancer cells (Li, Ma et al.
2012). We evaluated the potential of one of our lead compound, HBS 6, to downregulate
the levels of CXCR4 gene, which is also a downstream target of HIF transcriptional
pathway. Gratifyingly, treatment with HBS 6 at 10 µM concentration resulted in a
significant reduction of gene expression levels under these conditions (Figure 3.25).
It is believed that under conditions of serum starvation cells have reduced basal
activity and make the population more homogenous as they prefer to remain in G0/G1
phase. Although serum-supplemented media is used to provide the optimal conditions
for cell growth, it also adds a certain variability to the assay which can have additional
impact on the experimental results (Pirkmajer and Chibalin 2011). Moreover, addition of
different batches of serum can also lead to variable experimental results. In keeping with
all these points, we decided to test the potential of HBS helices in minimal serum
conditions. Based on the qRT-PCR results discussed above , the efficacy of HBS 6 in
modulation of the HIF induced transcription was tested further.HeLa cells were
maintained in 2% media followed by the treatment of cells with HBS 6 at 1 µM and 10
µM in 0.2% media for 48 h. Six hours after dosing, hypoxia was induced with DFO and
cells were further incubated for 42 h. In another set of conditions, hypoxia was induced in
half of the wells by placing the six-well plate in a hypoxic bag (BD GasPak EZ pouch).
First, we measured the mRNA levels of VEGF under two different conditions of hypoxic
induction. Although the 9-fold enhancement in VEGF expression was achieved with the
132
hypoxic bag, upon treatment with HBS 6 at 10 µM, VEGF was downregulated to almost
same levels as observed with the previous experiment (Figures 3.26 A and B).

Figure 3.26 VEGF mRNA levels in HeLa cells as measured by qRT-PCR. Cells were
treated with HBS 6 at 1 and 10 µM for 48h under both normoxic and hypoxic conditions.
Hypoxia was mimicked by A) 300 µM DFO and also induced by B) hypoxic bag (HB).
Hela cells were in the medium supplemented with 0.2% serum. Error bars represent ±
s.e.m. of experiments performed in triplicate. ***P<0.0001,**P, 0.01, t test.

Upon treatment with HBS 6 normoxic levels of VEGF remained at the basal
levels, indicating there is no non-specific effects on treatment with HBS 6. Although by
comparison, under these conditions hypoxic bag showed the best induction values, a
significant downregulation of HIF inducible genes was observed with DFO. Figure 3.27
and Figure 3.28, shows the effect of treatment with HBS 6 on the levels of three
important hypoxia inducible genes (LOX, CXCR4 and Glut1) with DFO used as a
hypoxia mimicking agent. LOX and CXCR4 mRNA levels were reduced to a significant
amount and a moderate reduction was achieved for Glut1 mRNA expression levels.
133

Figure 3.27 LOX mRNA levels in HeLa cells as measured by qRT-PCR. Cells were
treated with HBS 6 at 1 and 10 µM for 48h under both normoxic and hypoxic conditions.
Hypoxia was mimicked by A) 300 µM DFO and also induced by B) hypoxic bag (HB).
Hela cells were supplemented with 0.2% serum. Error bars represent ± s.e.m. of
experiments performed in triplicate. **P, 0.01, *P< 0.05, t test.

With hypoxic bag, induction levels with all the genes were remarkably high as
compared to DFO, but HBS helices had only modest effect on the downregulation of
tested genes (30% inhibition). On the other hand, chetomin showed the expected decrease
in mRNA levels of all genes under both conditions, exceptfor CXCR4 gene, where only
modest inhibition was observed in DFO induced samples (Appendix B, Figure B.5&
B.6).
134

Figure 3. 28 A) Glut1 and B) CXCR4 mRNA levels in HeLa cells as measured by qRT-
PCR. Cells were treated with HBS 6 at 1 and 10 µM for 48h under both normoxic and
hypoxic conditions. Hypoxia was mimicked by 300 µM DFO and also induced by
hypoxic bag (HB). HeLa cells were in a media supplemented with 0.2% serum. Error
bars represent ± s.e.m. of experiments performed in triplicate. **P, 0.01, *P< 0.05, t test.
Vehicle
HBS 6
135

c-Met is another key downstream target gene of the HIF-inducible transcriptional
system. The promoter region of this gene has 5 repeats of HRE and a number of
metastatic cancer cell lines have shown to have upregulated levels of c-Met. A549, a lung
carcinoma cell line, is reported to have high levels of key hypoxia inducible genes under
hypoxic conditions including c-Met.

Figure 3.29 c-Met mRNA levels in HeLa cells as measured by qRT-PCR. Cells were
treated with HBS 6 at 1 and 10 µM for 24 h under both normoxic and hypoxic conditions.
Hypoxia was mimicked by 300 µM DFO. HeLa cells were in a medium supplemented
with 10% serum. Error bars represent ± s.e.m. of experiments performed in triplicate. *P<
0.05, t test.

The A549 cells were cultured in 10% serum for 24 h and efficacy of HBS 6 was
measured.Unfortunately, HBS 6 did not show any reduction in the mRNA levels of
136
VEGF, Glut1, CXCR4 and LOX. On the other hand, 20% reduction was observed in c-
Met mRNA level (Figure 3.29). This data suggests that HBS helices might not be cell
permeable in A549 as compared to HeLa cells.

3.19 HIF-1α levels remain unchanged upon treatment with HBS 6
As per the proposed design, HBS helices are mimics of the CTAD domain of HIF-1α and
thus act as orthosteric inhibitors of HIF-1α and p300 interaction. To further support our
design hypothesis, it was important to analyze the impact of HBS helices on the levels of
HIF-1α protein. In this assay, HeLa cells were treated with HBS 6 at 10 µM, under
normoxic as well as hypoxic conditions and HIF-1α levels were analyzed by western
blot.

Figure 3.30 Analysis of HIF-1α levels by western blotting. HeLa cells were treated for
24 h with HBS 6 and hypoxia was induced with 300 µM DFO. Endogenous levels of
HIF-1α remained unchanged upon treatment with HBS 6 at 10 µM.

As shown in Figure 3.30, levels of HIF-1α remained completely unchanged under
hypoxic conditions (vehicle vs. HBS 6) whereas, under normoxic conditions, the levels of
HIF-1α were below the detection limit (Figure 3.30). This data indicates that the
inhibition mechanism of HBS 6 does not affect the stability of HIF-1α.
137
3.20 HBS 6 downregulates the endogenous level of VEGF protein
We wanted to ensure that no compensatory mechanism, downstream of transcription, is
reversing the observed downregulation of genes. Therefore, we analyzed the secreted
levels of VEGF protein in the presence of HBS 6 under both normoxic and hypoxic
conditions. HeLa cells were treated with different concentration of HBS 6 and the treated
culture media was analyzed for secreted VEGF by ELISA.
.

Figure 3.31 Analysis of VEGF protein levels in HeLa cells after 24 h of treatment with
HBS 4 at 1, 10 and 20 µM under normoxic (-DFO) and hypoxic (+DFO) conditions.
Hypoxia was induced with 300 µM DFO. Error bars represent ± s.e.m. of experiments
performed in triplicate. *P< 0.05, t test.

Under hypoxic conditions, upon treatment with HBS 6 a readily detectable
decrease was observed in the VEGF protein levels (Figure 3.31). This data clearly
138
parallels with the observed decrease at the mRNA levels and suggests that transcriptional
control could be an effective way of modulating the levels of expression of hypoxia-
inducible genes and their protein products.

3.21 Conclusion
Targeting transcriptional machinery of HIF-1α and thus regulating the expression of its
downstream target genes is an attractive and promising therapeutic approach for many
cancers. Stabilizing α-helices have a great potential as inhibitor of protein-protein
interaction. Along with the obvious improvement in the proteolytic stability and cellular
permeability, HBS offers significant advantage of not blocking the solvent exposed
molecular recognition surfaces of the molecule and also does not engage the important
side chain functionalities. With this novel approach in hand, in this Chapter we explored
the therapeutic potential of HBS mimetics in regulating the activity of the hypoxia-
inducible transcription factor complex.
Our data shows that designed HBS mimics of αA and αB of CTAD domain of
HIF-1α can bind to their target (p300-CH1) and disrupt the formation of the complex
between transcription factor, HIF-1α and its coactivator, p300. In cell culture HBS
helices showed the dose-dependent downregulation of four important HIF inducible
genes, VEGF, Glut1, LOX and CXCR4 in HeLa cells. Further, genome-wide effects and
in vivo studies of these designed HBS helices are currently underway.

139

Chapter 4: Multifunctional integrin-selective small molecules for tumor targeting

140
4.1 Integrins and their relevance in the regulation of signaling pathways
Several decades ago, it was believed that integrins are just adhesive macromolecules that
promote the adhesion of cells to other cells and to the extracellular matrix. Now, it is
increasingly apparent that integrins itself are cell signaling molecules and like growth
factor receptors, upon activation, they can also initiate signaling cascade and affect a
number of downstream signaling pathways. Integrins play an important role in
maintaining certain key cellular processes such as cell motility, cell migration, cell shape,
and cell proliferation. They are involved in the T-cell activation, lymphocyte adhesion to
endothelial cells and also regulate apoptotic processes by a mechanism called integrin-
mediated cell death (Panes, Perry et al. 1999; Stupack and Cheresh 2002). After binding
of extracellular ligands, integrins lead to the formation of focal adhesion complexes.
These complexes then further recruit various cytoskeletal proteins such as focal adhesion
kinases (FAK), paxillin, and vinculin. Activated FAK and Src protein phosphorylates
paxillin and p
130
Cas. Phosphorylated p
130
Cas recruits the small GTPase which further
activates Rac and contributes towards the cell motility. Activated FAK-Src complex is
also shown to activate the PI3K/Akt pathway which leads to cell survival (Mitra and
Schlaepfer 2006). Another protein, integrin linked kinase (ILK), also activates PI3K/Akt
pathway and supports the cell survival (Stupack and Cheresh 2002). Activation of Src is
also responsible for cell differentiation and proliferation through Ras-Erk pathway.
Another very interesting twist in the integrin signaling is that they collaborate with
different growth factor receptors and activates critical downstream signaling. Different
mechanisms of crosstalk between growth factor receptors and integrin receptors have
been discussed in literature (Eliceiri 2001; Ross 2004). For example, integrin receptors
141
upon ligand binding can activate growth factor receptor even in the absence of a growth
factor and is referred as direct activation. Adherent cells require integrins to create an
environment that is suitable for them to respond to growth factors and thus, defined as
collaborative activation (Ivaska and Heino 2011). Sometimes both receptors activates the
same pathway, for example, PI3K-Akt and Ras-ERK pathways. Integrins are distinct kind
of cell signaling proteins as they participate in bidirectional signaling across the plasma
membrane, that is, outside-in signaling as well as inside-out signaling. The long-range
conformational changes seem to be responsible for the bidirectional signaling. There are
several evidences which suggest that integrin remain in a bent form, an inactive state,
which hides the epitopes present on the I-EGF domain. While upon activation by ligand
(outside-in signaling) or by interaction of talin or other proteins to the β-cytoplamic
domain, integrin go under long range conformational changes (Hynes 2002). These
changes separate the two subunits and in this process, expose the epitopes present on the
I-EGF domain, which lead to the strong binding of the ligands and thus activate the
integrin signaling. In vertebrates, integrin family is composed of 18α and 8β subunits that
assemble to form 24 integrin heterodimers. Integrins are important in diverse array of
biological processes and generally, each heterodimer has a specific and non-redundant
function. Given the importance of integrins, it is believed that lack of integrin expression
or mutation in their genes would exhibit wide range of phenotypes. Integrin β1, for
example, dimerizes with a number of α-subunits and its knockout is embryonic lethal
because of the complete blockage of peri-implantation development. Similarly, knockout
of α
v
gene results in extensive brain and intestinal hemorrhaging, indicating the defect in
blood vessel integrity and essential role of αv in blood vessel development (Eliceiri and
142
Cheresh 1999). Studies with α
v
antagonists showed regression in tumor xenografts and
reduction in the blood vessel formation in CAM assay indicating the important role of α
v

in angiogenesis. Knockout of α4 and α5 were also embryonic lethal due to defects in
placenta and vascular development, respectively (BeauvaisJouneau and Thiery 1997).

4.2 Targeting α
v
β
3
integrin receptors as a therapeutic strategy for tumorigenesis
Angiogenesis, the process of formation of new blood vessels, promotes the tumor growth
and metastasis. Process of angiogenesis involves four critical steps; 1) activation of
endothelial cells by angiogenic factors, such as basic fibroblast growth factor (bFGF),
vascular endothelial growth factor (VEGF), and others 2) degradation of basal lamina by
release of proteinases from activated endothelial cells and 3) proliferation and migration
of endothelial cells and 4) tube formation. Release of growth factors activates the
expression of integrins like, α1β1, α
5
β
1
, α
v
β
5
, and α
v
β
3
on blood vessels and leads to the
formation of new sprouting vessels (Avraamides, Garmy-Susini et al. 2008). These newly
formed vessels then support the migration of endothelial cells and promote the process of
tumor growth and metastasis by providing proper oxygen and nutrient supply. Different
cell types express different subtypes of integrins on their cell surfaces, for example,
endothelial cells express α
v
β
3
, α
v
β
5
, α
1
β
1
, α
2
β
1
, α
4
β
1,
α
5
β
1
, α
6
β
1
,α
6
β
4
and α
9
β
1
(Alghisi,
Ponsonnet et al. 2009). Of these, α
v
β
3
and α
v
β
5
are the most studied integrins involved in
angiogenesis. α
v
β
3
is the most abundantly expressed on the active angiogenic endothelial
cell and remain almost absent on the surface of quiescent cells (Brooks, Stromblad et al.
1996). These receptors bind with both proangiogenoc and antiangiogenic factors. For
example, it interacts with VEGFR2, vitronectin, Del1 to stimulate the process of
143
angiogenesis and also binds with antiangiogenic factors such as angiostatin and
thrombospondin (Niu and Chen 2011).
Due to the overexpression of α
v
β
3
receptors on the surface of the vascular endothelial
cells, activation of vascular tyrosine kinase receptors is observed. Activation of VEGFR2
by VEGFA recruits the c-src kinase, which in turn phosphorylates the cytoplasmic tail of
the β3 subunits at Y747 and Y759 and results in the formation of VEGFR2/α
v
β
3
complex
(Figure 4.1), which further triggers the conformational changes and activation of integrin
(Serini, Napione et al. 2008).

Figure 4.1 Interaction between growth factor receptor VEGFR2 and integrin receptor
α
v
β
3
. Upon activation by VEGFA, VEGFR2 triggers the phosphorylation events and
activates c-src which eventually phosphorylates the cytoplasmic tail of the β3 subunit.
These signaling events increase the affinity of integrin receptor toward the extracellular
matrix proteins.

Mutation of the two tyrosine residues (Y747 and Y759) in mouse models showed
inhibition of angiogenesis, underscoring the importance of VEGFR2 and α
v
β
3
complex
formation during the process of vascularization. Several small molecule antagonists and
144
antibodies have been developed to target α
v
β
3
and are shown to be effective inhibitors of
neovascularization in vivo (Brooks, Montgomery et al. 1994; Mulgrew, Kinneer et al.
2006). The overexpression of α
v
β
3
is observed in a number of solid tumors, including
prostate, skin, lung, and breast. A humanized monoclonal antibody, Abegrin, specific for
α
v
β
3
receptors is in phase II clinical trials for the treatment of colorectal cancer and is
shown to be effective as antiangiogenic and antitumor compound (Veeravagu, Liu et al.
2008). The α
v
β
3
adhesion receptor belongs to the RGD (arginine, glycine, aspartic acid)
family of receptors where they recognize the RGD triad in the extracellular matrix
proteins, further facilitating adhesion and downstream signaling. The first small molecule
antagonist was a cyclic peptide compound c(RGDfV) which selectively blocks the
activation of α
v
β
3
and α
v
β
5
receptors (Aumailley, Gurrath et al. 1991). Later an expanded
sequence c(RGDf(NMe)V) was constructed that is now is in the Phase III clinical trial for
glioma and Phase II clinical trial for several other solid tumors overexpressing α
v
β
3
and
α
v
β
5
receptors. Its preclinical and clinical studies have shown that the sequnce
c(RGDf(NMe)V) is a potent inhibitor of angiogenesis and also inducer of apoptosis
(Mas-Moruno, Rechenmacher et al. 2010).
In last few years, several non-peptidic small molecule high affinity α
v
β
3

antagonists have also been developed. All these non-peptidic small molecules contain a
free carboxylate group from an aspartic acid, a guanidine group, and a central
hydrophobic group attached to a non-cyclic scaffold (Figure 4.2). Because of the
selective expression of α
v
β
3
receptors on the surface of the endothelial cells undergoing
angiogenesis, these receptors have been explored for the development of antiangiogenic
145
inhibitors as well as because a focus for new strategies for thargeting overexpression of
α
v
β
3
receptors for imaging and to deliver cytotoxic therapeutic agents to tumor cells.

Figure 4.2 Non-peptidic small molecule antagonists of α
v
β
3
receptorsshare three key
features: carboxylate group, hydrophobic group and guanidinium-like moiety.

It has been showed by Murphy and coworkers that RGD-based nanoparticles
selectively accumulate within the tumor-associated blood vessels and show very little
affinity toward other normal vascular areas (Murphy, Majeti et al. 2008). Hence, it was
suggested that the use of integrin targeting for selective delivery of therapeutics to cancer
cells could reduce the effective dose and, consequently, diminish the unwanted toxic side
effects of a chemotherapeutic agent to normal tissues.
Recently, Kiessling and coworkers have exploited the bifunctional ligand to target
cancer cells using integrins as molecular targets. They conjugated the immunogenic
carbohydrate α-Gal to a RGD based α
v
β
3
antagonist (Owen, Carlson et al. 2007).
146
Although α-Gal itself is not selective toward the tumor cells, it can be delivered to tumor
area by α
v
β
3
antagonist, where they could trigger an immune response, thereby destroying
tumor cells without damage to normal, healthy tissue. In another report, Burkhart and
coworkers utilized the bifunctional approach to deliver an active metabolite of
doxorubicin under physiological conditions (Burkhart, Kalet et al. 2004). They designed
two conjugates of the prodrug, doxaliform, with RGD-based α
v
β
3
-targeting peptides,
CDCRGDCFC (RGD-4C) and cyclic-(N-Me-VRGDf) and showed high affinity of the
conjugate toward vitronectin, the natural ligand of α
v
β
3
.
In this chapter we describe the design, synthesis, and biological properties of a
fluorescent conjugate comprised of high affinity, α
v
β
3
-selective integrin ligand and a
fluorescent probe that exhibits fluorescence enhancement upon binding. Such low
molecular weight conjugates can find application in cancer imaging and therapy .

4.3 Active targeting of tumors by boron neutron capture therapy (BNCT)
The best case scenario for cancer treatment would be a therapy where tumor cells are
selectively targeted without affecting the normal tissues. The limitation of the current
therapies, such as selectivity, toxicity and the recurrence of the tumors have led many
efforts in the development of binary therapies that involve the two-component systems.
In such system each component itself is not cytotoxic but when combined lethal effects to
cells are produced. Among binary therapies, neutron capture therapy is increasingly
gaining attention in tumor-selective delivery of chemotherapeutics and offers a promising
alternative to classical or traditional treatments. It relies on the ability of certain nuclei to
capture thermal neutrons and release high-energy particles that can damage or destroy
147
cancer cells. The importance of this technique for the treatment of cancer was first
realized by G.L. Locher in 1936 (Locher 1936). The most actively used nuclide for
neutron capture therapy is boron-10 and thus this referred as boron neutron capture
therapy (BNCT).

Figure 4.3 Schematic illustration of boron neutron capture.
10
B nuclei capture neutrons
and convert into unstable species [
11
B] which further disintegrate into helium-4 nuclei (α-
particles) and lithium.
In BNCT, upon irradiation by thermal neutrons, the non-toxic
10
B nucleus is
converted into an unstable species
11
B. The resulting
11
B then subsequently decay to high
linear energy α-particle and
7
Li (Figure 4.3). The high energy particles are damaging and
their flight path confined to the cell size (5-9 µm), as their path length is roughly equal to
a cell diameter (McDevitt, Ma et al. 2001). In addition to boron, other elements have also
been studied as possible nuclides for the neutron capture therapy. In the initial phase of
development, radioactive
235
U was considered as potential nuclide. The fission particles
generated by the neutron capture are of desired size and energy so that the radiation
would be confined to the cell. However, use of uranium is restricted due to the high
toxicity associated with these compounds. In general, NCT concentrated more on the use
of non-radioactive nuclides as energy produced by radioactive nuclide, unless restricted,
148
can damage normal cells and its supporting structures. Next,
6
Li being non-toxic and non-
radioactive, was proposed as a potential nuclide. The fission reaction products of
6
Li also
meet the criteria of high energy particles, which can be constrained to a distance
comparable with the length of the cell and destroy it selectively. The major limitation
with the use of these particles is the stability of alkali metal lithium. Under physiological
conditions,
6
Li converts into lithium cation and as a result, it is difficult to form stable
compounds with
6
Li (Zahl, Cooper et al. 1940). Other than these elements, recent studies
have been performed with
157
Gd because of its high neutron capture cross-section.
However, use of
157
Gd is limited due to the nature of the radiation emitted upon neutron
capture; one of the fission products of this reaction is γ-rays which are not confined to the
tumor cell and thus are less selective. (Martin, Dcunha et al. 1988).

Figure 4.4 Structure of the first and second-generation boron compounds for BNCT.
Borax (1
st
generation), BPA and BSH (2
nd
generation).
Therefore, nonradioactive
10
B is the most preferred nuclide for the neutron capture
therapy. Boron exists in two forms,
10
B and
11
B, out of which
10
B comprise the 20% of
the total natural boron. Along with the advantage of producing high energy particles upon
149
irradiation which are confined to the cell, the small size of boron atom also allows it to be
incorporated as a the replacement of carbon in many compounds.
Many boron compounds can be synthesized with hydrolytically stable linkers between
boron and other elements, like carbon, nitrogen and oxygen (Barth, Coderre et al. 2005).
Although,
10
B has a relatively high neutron capture cross-sectional area, high
concentration of nitrogen and hydrogen in tissues can hinder its efficacy by absorbing the
neutron beam upon exposure.
Figure 4. 5 Structures of ortho-, meta- and para-carborane clusters.
In order to minimize this effect and to make BNCT successful, the concentration of boron
atoms inside or near tumor cells must be high (Hawthorne 1993) (≥10
9
-
10
B atoms/cell,
≈20-35 µg
10
B/cell) while it must remain as low as possible in the normal tissue.
Therefore, tumor to normal tissue concentration ratio ≥ 3:1 is preferred. In an effort to
achieve this goal, several boron compounds have been investigated. In 1950’s and early
1960’s the first generation, simple boron compounds (Figure 4.4), boric acid and sodium
borate were tried as boron delivery agents (Farr, Sweet et al. 1954). However, due to their
low retention in the tumor tissue and inability to penetrate blood-brain barrier restricted
150
their use in clinical trials. These compounds were mainly tested for BNCT in brain
cancer, head and neck cancer and melanoma (Kato, Ono et al. 2004).
Next, the second generation compounds (Snyder, Reedy et al. 1958), [(L)-4-
dihydroxyborylphenylalanine], (BPA) based on arylboronic acids and polyhedral borane
anion, sodium mercaptoundecahydro-closo-dodecaborate (BSH) were designed (Figure
4.4). These second-generation compounds were more water soluble, had longer retention
time in the tumor tissues and also had low toxicities. The third-generation boron delivery
agents mainly consist of stable boron clusters called carboranes (Figure 4.5), with
hydrolytically stable linkers attached to tumor-targeting moieties (Hawthorne and Lee
2003). Carboranes are very stable, show low toxicity and are able to deliver high number
of boron atoms per molecule. As discussed earlier, α
v
β
3
is over-expressed almost
exclusively on the surface of angiogenic endothelial cells involved in the growth of tumor
vasculature, while being nearly absent on the surface of normal cells. In this Chapter we
discuss design of a multifunctional carborane cluster to target α
v
β
3
positive tumors
throuhht the use of a BNCT approach.

4.4 Design and synthesis of α
v
β
3
ligand-dye conjugates for optical imaging of integrin
receptors
Optical imaging is a sensitive and non-invasive molecular imaging tool which could be
used to interrogate a multitude of biological processes both in vitro and in vivo. Unlike
radiopharmaceuticals,optical imaging does not require special handling equipment and is
safer in its applications in the clinic. The relatively low cost, high sensitivity, rapid
imaging times and no issues of handling radioisotopes made optical imaging a preferred
151
diagnostic technique. As discussed earlier, the arginine-glycine-aspartic acid (RGD) is
the most common recognition sequence for integrins and half of the integrin receptors
recognize the RGD sequence in different ECM proteins. This promiscuity in ligand
recognition makes it challenging to design or to identify a specific ligand for a given
integrin receptor. Several high-affinity α
v
β
3
integrin ligands have been reported in
literature, for our purpose we chose a non-peptidic integrin ligand (Figure 4.6) which was
originally reported by Duggan et al (Duggan, Duong et al. 2000). This ligand exhibited
very high specificity towards α
v
β
3
integrin in the presence of the closely related
integrins,such as α
v
β
5
and αIIbβ3.

Figure 4.6 Structure of an α
v
β
3
specific integrin-ligand dye conjugate, ABL-1. The
structure of monomethine cyanine dye is in green and integrin ligand is in black.

Both cell adhesion assay and receptor binding assay with the ligand showed its high
specificity with mean inhibitory concentration of 40 nM for α
v
β
3
as opposed to 5.5 µM
for α
v
β
5
and 2.1 µM for α
IIb
β
3
receptors (Hood, Bednarski et al. 2002). The 140-fold
152
specificity for α
v
β
3
receptors over

α
v
β
5
is quite remarkable, considering that not many
natural integrin ligands can differentiate between these two receptors.
Another interesting feature of this ligand is the presence of the free amine group
which can be utilized for the conjugation purposes.

Scheme 4.1 Synthesis of the fluorescent probe integrin ligand, 1. (a) PPh
3
, DIEA, THF,
0 °C to RT, 16 h. (b) HCl, THF, 60 °C, 18 h. (c) 2-bromopyrimidine, DIEA, dioxane,
reflux, 48 h. (d) NaOH, EtOH, 60 °C, 2 h. (e) 7, HBTU, DIEA, DMF, RT, 18 h. (f) TFA,
CH
2
Cl
2
, RT, 30 min. (g) 9, DMAP, CH
2
Cl
2
, 0 °C to RT, 18 h. (h) Pd Black, H
2
, EtOH,
RT, 3 h. (i) TFA, CH
2
Cl
2
, RT, 30 min.
153
The synthesis of integrin ligand - cyanine dye conjugate, ABL-1 and ABL-2, (Figure 4.6)
was done in collaboration with Nathan W. Polaske and was designed and performed
based on the known synthesis of a structurally similar compound. The first part of the
synthesis requires the formation of a precursor 10 in a protected from which allows for
easy further synthetic manipulations and flexibility for the attachment of fluorescent dye
and other conjugate procedures. First, N-Boc-ethanolamine and methyl-4-
hydroxybenzoate were coupled under Mitsunobu conditions, yielding desired ether 3 as a
crude mixture. Next, Boc group was deprotected in the presence of HCl to yield the free
amine 4 . The pyrimidine ring was then installed by the reaction of 4 with 2-
bromopyrimidine, producing methyl ester 5 in a moderate yield. Hydrolysis of 5 with
NaOH as a base resulted in 6, which was then coupled to 3-amino-L-alanine derivative 7
under standard peptide coupling conditions and provided amide 8 in good yield (Scheme
4.1). Finally, hydrolysis of the Boc group with TFA followed by reaction of the
deprotected amine with 9 provided the protected intermediate 10 in 51% yield over two
steps.
A lesser known monomethine cyanine dye cyan 39 (11) was used as a fluorescent
probe for the optical imaging (Scheme 4.2). This dye has a property of being fluorescent
only when the free rotation around the single bond that connects the two rings is hindered
and remains virtually non-fluorescent in the environment where such free rotation is
possible (Yarmoluk, Kovalska et al. 2001).
154

Scheme 4.2 Synthesis of the monomethine cyanine dye, cyan-39.
In the free, non-bound state or in the non-viscous environment, the dye remains non-
fluorescent. In the viscous media or when the dye is in the bound state, the rotation
around the intramolecular bond is restricted, resulting in an enhancement of the
fluorescence intensity. This property of the cyanine dye gave us an additional advantage
of having minimal background fluorescence due to the unbound conjugate being non-
fluorescent. The synthesis of cyan-39 dye was carried out from the commercially
available 2-methylbenzothiazole 12 as described in the literature (Kostenko, Dmitrieva et
al. 2002).

Scheme 4.3 Synthesis of the integrin-ligand dye conjugate, ABL-1. (j) 11, DIEA, DMF,
RT, 1 h.
155
Further reduction of 10 with Pd/H
2
, followed by the removal of Boc group with TFA
furnished 1. In the final step chromophore 11 was coupled to 1 to produce fluorescent
integrin ligand conjugate probe ABL-1 (Scheme 4.3).
We have also synthesized a second integrin ligand-dye conjugate ABL-2 (Figure
4.7) by following the same procedure. ABL-2 contains benzoselenazole instead of
benzothiazole ring in its chromophore.

Figure 4.7 Structure of sulfur (ABL-1) and selenium (ABL-2) based α
v
β
3
specific
integrin-ligand dye conjugates tested in biological system.

4.5 Photophysical properties of α
v
β
3
integrin ligand conjugates, ABL-1 and ABL-2
Absorption spectra of ABL-1, recorded in both aqueous buffer and in 90% v/v
glycerol/water, show similar profiles. Slightly lower absorbance was observed in 90% v/v
glycerol/water system,indicating change in the chromophore environment with
absorption maxima at 448-455 nm (Figure 4.8). Uv-vis spectra of ABL-2 also showed
similar absorption maxima recorded in 90% v/v glycerol-water mixture (Figure 4.9).

156

Figure 4.8 Uv-vis absorption spectra of ABL-1 in aqueous buffer and 90% v/v glycerol -
water system. The dotted black line represents aqueous buffer and the solid black line
represents glycerol-water mixture. The absorption maxima were observed at 448-455 nm.

Figure 4.9 Uv-vis absorption spectra of ABL-2 in 90% v/v glycerol-water system. The
absorption maximum was observed at approx. 448 nm.
157

Next, to analyze the photophysical properties of both integrin-ligand fluorescent
conjugates, fluorescence spectroscopy was employed. In order to measure the emission
of our fluorescent probe in free and bound state, the fluorescence emission spectra of
ABL-1 and ABL-2 were recorded in both aqueous buffer and in 90% v/v glycerol-water
system (Figure 4.10 and Figure 4.11). Both showed low fluorescence in aqueous buffer
when photoexcited at 380 nm while in 90% v/v glycerol-water both the fluorescent probe
showed nearly 35-50 fold increase in the fluorescence emission intensity.

Figure 4.10 Fluorescence emission spectra of ABL-1 in aqueous buffer and 90% v/v
glycerol-water system. The dotted black line represents aqueous buffer and solid black
line represents glycerol-water mixture.
We have also analyzed the the fluorescent enhancement for the probes upon their
addition to WM115, a melanoma cell line with high number of the α
v
β
3
receptors.
158

Figure 4.11 Fluorescence emission spectra of ABL-2 (λex=380 nm) in aqueous buffer
and 90% v/v glycerol-water system. The dotted black line represents aqueous buffer and
solid black line represents glycerol-water mixture.
ABL-1 (Figure 4.12) and ABL-2 (Figure 4.13) were incubated with 2 million cells at a
concentration of 100 µM for 2h at 37 °C and in 5% CO
2.
Fluorescence emission of
fluorescent probes was measured in presence or absence of cells at 435 nm excitation
wavelength. As observed with glycerol water system enhancement in the fluorescence
was observed with cells. ABL-1 showed approximately five times enhancement and
ABL-2 showed enhancement two times in fluorescence as compared to the free probe in
PBS. These results indicate that the probes, ABL-1 and ABL-2, exhibit fluorescence
enhancement in the viscous media and in the presence of cells displaying α
v
β
3
integrins.
Corresponding excitation spectra were also recorded with the emission wavelength being
set at 540 nm in aqueous buffer and in the 90% v/v glycerol-water mixture. As with the
emission spectra, the increase of the peak intensity in the excitation spectra was observed.

159

Figure 4.12 Fluorescence emission spectra of ABL-1 in PBS and with WM115 cells.
The dotted black line represents ABL-1 in PBS and solid black line represents 100 µM
ABL-1 in WM115 cells. λex=435 nm, 5-fold enhancement was observed with cells.

Figure 4.13 Fluorescence emission spectra of ABL-2 (λex=435 nm) in PBS and with
WM115 cells. The dotted black line represents ABL-2 in PBS and solid black line
represents 100 µM ABL-2 in WM115 cells.,
160
4.6 Cytotoxicity of integrin-ligand dye conjugate (ABL-1) in melanoma and breast
cancer cell lines
Before analyzing the integrin-ligand dye conjugate in biological assays, we tested the
impact of ABL-1 on the overall cell viability. ABL-1 was tested in two different cell
lines, WM115 (melanoma) and MCF-7 (breast cancer). The metabolic activity of cells
was measured by two different cytotoxicity assays, cell titer blue and MTT. In these
assays, 10,000 cells/well were plated in a 96-well plate and allowed to form a monolayer.
After attachment, cells were treated with increasing concentration of ABL-1, ranging
from 1 nM to 100 µM.

Figure 4.14 Cell titer blue cytotoxicity assay with ABL-1 in A) WM115 and B) MCF-7
cells. Cells were incubated with the fluorescent probe for 3 h at 37 °C and in 5% CO
2

atmosphere.
In one experiment cells were incubated with compound for 3 h (Figure 4.14) and in
another experiment cells were incubated with compounds for a total of 24 h (Figure
4.15). Cell titer blue is a fluorometric method to count the number of viable cells. It
measures the rate of conversion of resazurin, a non-fluorescent compound, to a
fluorescent compound resorufin which emits at 590 nm upon excitation at 560 nm. This
rate of conversion is directly proportional to the metabolic activity of the cells. Further
161
steps were performed as discussed in experimental section. Under our conditions no
significant cytotoxicity was observed at concentrations from 1 nM to 10 µM and a slight
toxicity in the concentration range of 10-100 µM. The results remained the same in both
cell lines.

Figure 4. 15 MTT cytotoxicity assay with ABL-1 in A) WM115 and B) MCF-7 cells.
Cells were incubated with the fluorescent probe for 24 h at 37 °C and in 5% CO
2

atmosphere.
In another experiment cells were treated with ABL-1 for 24 h and 3-(4, 5-
dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide, an MTT reagent, was used to
measure the metabolic activity of the cells. In both assays, approximately70% of the cells
were live at a concentration of 100 µM in WM115 while MCF-7 cells showed slightly
higher toxicity with 60% cells live at 100 µM.

4.7 Cellular uptake of integrin-ligand dye conjugates
The cellular uptake of ABL-1 is primarily examined by two complementary techniques,
confocal laser scanning microscopy and flow cytometry. First, confocal imaging was
carried out in order to determine the internalization and subcellular localization of the
162
conjugate, ABL-1. Next, the distribution of the conjugate and its ability to induce
apoptosis were studied by flow cytometry.
4.7.1 Investigation of uptake and internalization of ABL-1 by confocal microscopy
The typical cell permeabilization techniques often use fixed cells for imaging and can
cause relocalization or diffusion of small molecule imaging agents. Therefore, to avoid
these artifacts we chose live cell imaging in our analysis of the cellular uptake and
localization of the fluorescent probe, ABL-1. Initial in vitro imaging experiments were
performed with melanoma cells, WM115. Based on the recent work by Kiessling and co-
workers (Carlson, Mowery et al. 2007), we selected this highly metastatic melanoma cell
line as they display on an average 100,000 of α
v
β
3
integrin receptors. WM115 cells were
plated and allowed to adhere overnight in MatTek dishes.

Figure 4.16 Confocal microscopy images of WM115 treated with 5 µM of ABL-1 at 37
°C . WM115 cells shows active uptake of ABL-1 treated at A) 5 µM concentration for 30
min B) 5 µM concentration for 3 h. The top left quadrant shows conjugate emission
(green), top right quadrant shows bright field image of the cells and bottom left quadrant
shows overlay of conjugate emission with brightfield image.

163
Next, cells were incubated with ABL-1 at different concentrations, ranging from 1 µM to
100 µM at 37 °C and in a 5% CO
2
atmosphere for 30 minutes for 1 h and 2 h,
respectively. Imaging was performed on a Zeiss LSM 510 inverted laser-scanning
confocal microscope equipped with an 63× oil-immersion objective lens at an excitation
of 488 nm and an emission wavelength of 510 nm. Similar experiments were carried out
with MCF7 cells, which have lower count of α
v
β
3
integrins. It has been reported that α
v
β
3
receptors are internalized and transported through the early endosomes and reach the
perinuclear recycling compartment in about 30 minutes (Roberts, Barry et al. 2001).

Figure 4.17 Confocal microscopy images of WM115 cells treated with 100 µM of ABL-
1 for 3 h at 37 °C. The top left quadrant shows integrin-ligand-dye emission (green), top
right quadrant shows bright field image of the cells and bottom left quadrant shows
overlay of dye emission with bright field image.
This is consistent with our observation, as shown in Figure 4.16A, where
significant accumulation of ABL-1 at 5 µM concentration was found after incubation of
164
30 minutes at 37 °C. With increased incubation time more fluorescence was observed at
the periphery of the cells, possibly indicating the recycling of the α
v
β
3
receptors (Figure
4.16 B). We also tested the uptake of ABL-1 at the highest concentration of 100 µM. At
this concentration, after 3 h of incubation, a dramatic increase in the extent of
internalization of ABL-1 was observed in WM115 cells (Figure4.17). CF-7 cells also
showed substantial uptake of the probe under our experimental conditions (Figure
4.18B).

Figure 4.18 Confocal microscopy images of A) WM115 cells treated with 5 µM of
ABL-1 for 30 min at 37 °C B) MCF-7 cells treated with 5 µM of ABL-1 for 30 min at 37
°C.
165

Figure 4.19 Confocal microscopy images of WM115 cells treated with A) 100 µM of
ABL-2 for 3 h B) 100 µM of ABL-1 for 3 h at 37 °C. The top left quadrant shows
integrin-ligand-dye emission (green), top right quadrant shows bright field image of the
cells and bottom left quadrant shows overlay of dye emission with brightfield image.
166
Similar imaging experiments were performed with ABL-2 in WM115 cells. ABL-
2 did not exhibit the expected a similar enhancement in the fluorescence intensity and
was excluded from the further studies (Figure 4.19).
To test the mechanism of internalization, we incubated cells in the presence of 5
µM of ABL-1 at 4 °C for 2 h before imaging. At this low temperature, the process of
endocytotic internalization is slow and as expected, the probe remained completely non-
cellular at 4 °C (Figure4.20).

Figure 4.20 Confocal microscopy images of A) WM115 cells treated with 5 µM of
ABL-1 for 2 h at 4 °C B) MCF-7 cells treated with 5 µM of ABL-1 for 2 h at 4 °C. No
fluorescence was observed at this temperature in both cell lines. The top left quadrant
shows integrin-ligand-dye emission (green), top right quadrant shows bright field image
of the cells and bottom left quadrant shows overlay of dye emission with brightfield
image.
To testthe specificity of our fluorescent probe, a competition experiments were
performed by co-incubating the WM115 cells with ABL-1 at a concentration of 20 µM in
the presence of a 10-fold excess of the non-fluorescent ligand 1 at 37 °C for 1 h. No
fluorescence was observed, indicating that unlabeled ligand is competitively inhibiting
the binding and internalization of ABL-1 (Figure 4.21).
167

Figure 4.21 Confocal images of WM115 cells (A) and MCF7 cells (B) after incubation
with 20 µM of ABL-1 and 200 µM of 1 in cell culture media with at 37 °C. Left image:
fluorescence signal from conjugate; middle: overlay of fluorescence signal with visible
light image; right: visible light image.

4.7.2 Distribution of integrin-ligand dye conjugate, ABL-1 in WM115 cells by flow
cytometry
Flow cytometry is a powerful technique which can measure multiple properties of a
single cell in a heterogeneous population. It can be used for cell counting, cell sorting and
determining of the biomarker distribution. We performed this assay to analyze the
distribution and uptake of the conjugate within the population of WM115 cells.
168

Figure 4.22 Distribution and uptake of ABL-1 in WM115 cells by flow cytometry.
WM115 cell were treated with increasing concentration of ABL-1 for 2 h at 37 °C. A) 1.
WM115 cells; 2. WM115 cells with 5 µM ABL-1; 3. WM115 cells with 25 µM of ABL-
1. The y-axis corresponds to the emission wavelength of cyanine dye. B) Histogram
presentation of WM115 with increasing concentration of ABL-1.Shift in. population
along the x-axis indicates the uptake of ABL-1.

WM115 cells were incubated with increasing concentration of the conjugate, 5
µM and 25 µM at 37 °C for 2 h and the uptake was measured by BD LSR II Flow
Cytometer. The shift in the population of cells with increasing concentration along the y-
axis indicates the active uptake of the conjugate (Figure4.22). The results from flow
cytometry assays parallel the results from confocal microscopy experiments, indicating
active uptake and internalization of the probe ABL-1 by cells.

169
4.7.3 Measurement of the apoptosis rate of WM115 cells in the presence of ABL-1 by
flow cytometry
Upon treatment with ABL-1, cells showed round morphology, presumably because of the
interference of ABL-1 with the adhesion process. In order to ascertain that the observed
change in morphology is not due to the apoptosis, apoptosis rate was analyzed by flow
cytometry with annexin V-PE as an early apoptosis marker and 7-AAD,a late apoptosis
marker.

Figure 4.23 WM115 cells treated with annexin-V and increasing concentration of ABL-
1; A) WM115 cells; B) WM115 cells with annexin-V and 5 µM ABL-1; C) WM115 cells
with annexin-V and 25 µM ABL-1. Lower right quadrant showed the annexin-V positive
cells and upper right quadrant shows cells positive for both annexin-V and ABL-1s. The
shift in the population of cells along the y-axis is for cells positive for ABL-1.

Degradation of plasma membrane and exposure of phospholipid
phosphatidylserine to the external cellular environment is the initial step during
apoptosis. Annexin-V is a calcium-dependent phospholipid-binding protein that has a
high affinity for phosphatidylserine and can be labeled with phycoerythrin. The early
stage of apoptosis is then recognized with the help of labeled annexin-V. Loss of
membrane integrity in the later stages of apoptosis can be analyzed by 7-AAD which
stains cells that have undergone apoptosis or necrosis. WM115 cells were treated with
increasing concentration of conjugate, 5 µM and 25 µM at 37 °C for 2 h and then treated
170
with early and late apoptosis marker,per manufacturer’s protocol. The low apoptosis rate
at 5 µM and a small increase in apoptosis at 25 µM was observed (Figure4.23 and
Figure4.24) which is consistent with the results from our cytotoxicity experiment that
also showed a slight increase in the cytotoxicity in a concentration range 10-100 µM.

Figure 4.24 WM115 cells treated with 7-AAD and increasing concentration of ABL-1.
A) WM115 cells; B) WM115 cells with 7-AAD and 5 µM ABL-1; C) WM115 cells with
7-AAD and 25 µM ABL-1. The upper left quadrant showed the 7-AAD positive cells and
upper right quadrant shows cells positive for both 7-AAD and ABL-1. The shift in the
population of cell along the X-axis is the cells positive for ABL-1.

4.8 Integrin-ligand dye conjugate ABL-1 disrupts the cellular adhesion to vitronectin
Integrins are a large family of adhesion receptors which are primarily responsible for the
attachment to different extracellular matrix components (ECM) such as vitronectin,
fibronectin, collagen and others. Cells bind to vitronectin mainly by α
v
β
3
receptors.
Vitronectin is a 75 KDa glycoprotein found in the extracellular matrix that promotes cell
adhesion and spreading.
Incubation with ABL-1, WM115 cells showed dramatic change in morphology,
from elongated and well-spread to round. Based on this observation, we speculated that
ABL-1 is disrupting the process of adhesion of cells to its extracellular matrix proteins
and is responsible for the round morphology of cells. To validate our conjecture, we
decided to measure the adhesion of highly overexpressed α
v
β
3
integrins that express
171
WM115 cells to its major ECM ligand, vitronectin. For this experiment, a negative
control ABL-3 (Figure 4.25) was designed which contains guanidinium moiety in a fully
oxidized ring. The compound was synthesized by Vladimir Neschadimenko.

Figure 4.25 Structure of negative control ABL-3.

Cyclo-(RGDfV), a known integrin α
v
β
3
antagonist that inhibits the cell adhesion events
mediated by α
v
β
3
receptor

was employed as a positive control. WM115 cells were plated
in a vitronectin coated 96-well plate (R & D Systems) and incubated with ABL-1, Cyclo-
(RGDfV) and ABL-3 at increasing concentrations from 10 pm to 100 M for 2 h at 37 °C
and in 5% CO
2
.

172

Figure 4.26 Adhesion assays with A) ABL-1 and B) Cyclo-(RGDfV). WM115 cells
were plated on vitronectin coated plate and treated with increasing concentration of ABL-
1 and cyclo-(RGDfV).

173
After 2 h of incubation, adhesion was measured via a change of metabolic activity
by adding MTT (3-(4, 5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide). ABL-1
showed disruption of adhesion with an IC
50
value of 0.26 µM and cyclo-(RGDfV)
showed inhibition with 2.6 µM (Figure 4.26). In contrast, the negative control ABL-3 did
not interfere with the adhesion at concentrations up to 10 µM (Figure 4.27).

Figure 4.27 Combined data from adhesion assays. WM115 cells were plated on
vitronectin-coated plate and treated with the increasing concentration of ABL-1 (Red),
cyclo-(RGDfV) (Green) and ABL-3 (Blue).

To confirm the specificity of ABL-1 toward vitronectin-based receptors, it was
also tested against another abundant ECM protein, fibronectin. WM115 cells were plated
on a fibronectin-coated plate and incubated with ABL-1 and cyclo-(RGDfV) for 2 h at 37
174
°C in 5% CO
2.
Both compounds were practically ineffective in disrupting the adhesion of
cells to fibronectin (Figure 4.28), although a slight disruption was observed at 20 µM
concentration. These results confirm specificity of ABL-1 toward vitronectin receptors.

Figure 4.28 Effect of ABL-1 and cyclo-(RGDfV) on the adhesion of WM115 to
fibronectin. WM115 cells were plated on fibronectin-coated plate and treated with the
increasing concentration of ABL-1 (Red) and cyclo-(RGDfV) (Green).

4.9 Intravital microscopy imaging of ABL-1 in murine subcutaneous and in ectopic-
orthotopic tumor xenograft models with
In order to visualize the uptake and retention of fluorescent probe in real time we used
intravital microscopy (Lohela and Werb 2010) for in vivo imaging. These experiments
were performed by our collaborators, Prof. Jan Schnitzer and Dr. Philip Oh at the
Proteogenomics Institute for Systems Medicine (PRISM),San Diego. They used
subcutaneous (SQ) subcutaneous tumor model derived from human breast carcinoma
cells. In subcutaneous model, N2O2 breast carcinoma cells were injected directly into the
175
tissue and allowed to form tumor. A window chamber was installed on the dorsal side of
nude mice to facilitate imaging.

Figure 4.29 Intravital microscopy (IVM) images showing uptake of the ABL-1 in
subcutaneous tumor models. Upper images shows subcutaneous tumor after injection of
ABL-1 from day 1 to day 3. The substantial amount of ABL-1 is accumulated in the
tumor on day 1 and is retained for up to 3 days.
After tumor formation, ABL-1 was injected intravenously at a concentration of 1
mg/kg and via the tail vein. The fluorescence images were taken at excitation and
emission wavelength of the cyan 39 dye of the conjugate. In subcutaneous tumor model
with the leaky vasculature ABL-1 was able to accumulate and is detectable in significant
amount up to 3 days (Figure 4.29). This result suggests that active accumulation of ABL-
1 in tumor requires leaky vasculature.
4.10 Design and biological properties of trifunctional carborane-integrinligand dye
conjugate
Given the high specificity of ABL-1 ligand toward α
v
β
3
receptors, we proposed a
trifunctional system which may have a potential as a therapeutic agent for treatment of
solid tumors. Our approach relies on the application of boron neutron capture therapy
(BNCT) to target tumor cells. We have designed a bifunctional ligand featuring a
176
targeting moiety specific for α
v
β
3
receptors and capable of selectively delivering a boron
payload specifically to neoplastic cells.

Figure 4.30 Structure of carborane integrin-ligand dye conjugate, CIL-1. Blue arrow
represents the integrin ligand, Gray line represents the linker, Green represents the
cyanine dye and pink represents the boron clusters in the form of o-carboranes.
For the successful application of the triifunctional ligand we need a system which carries
a large number of boron atoms in a biocompatible form. As discussed in section 4.3, we
conjugated o-carboranes to our integrin-ligand dye conjugate (ABL-1) and made it a
trifunctional system (CIL-1) that can be used for optical imaging as well as for the
delivery of sufficient quantity of
10
B into the proximity of a cancerous cell (Figure 4.30).
This work has been conducted in collaboration with Ramin Dubey and Nathan W.
Polaske.
177

Figure 4.31 Structure of the conjugate, CD-1. Gray line represents the linker, green
represents the cyanine dye and pink represents the boron clusters in the form of o-
carboranes.

To ensure the specificity of ABL-1 and CIL-1, a negative control compound, CD-
1 lacking integrin targeting moiety was prepared (Figure 4.31). Similarly to ABL-1,
fluorescence emission spectra of CIL-1 and CD-1 were measured in both aqueous buffer
and 90% v/v glycerol/water system. As expected, both the compounds showed
enhancement in the fluorescence intensity in viscous media. The uptake of CIL-1 was
studied in both WM115 and MCF-7 cell lines by confocal laser scanning microscopy.
178

Figure 4. 32 Confocal images of A) WM115 cells B) MCF-7 cells treated with CIL-1 at
37 °C at 25 µM for 2h.

Figure 4.33 Confocal images of A) WM115 cells B) MCF-7 cells treated with CD-1 at
37 °C at 25 µM for 2 h.
Both cells were incubated with 25 µM of CIL-1 and CD-1 at 37 °C and in 5% CO
2.

WM115 cells showed a significant amount of uptake of CIL-1 while MCF-7, cells with
low α
v
β
3
count, exhibit minimal fluorescence (Figure 4.32). On the other hand, CD-1 was
179
not detectable at this concentration in either cell lines (Figure 4.33). Similarly to ABL-1,
no accumulation of CIL-1 was observed at 4 °C in WM115 cells (Figure 4.34).

Figure 4.34 WM115 cells treated with A) CIL-1 and B) CD-1 at 4 °C for 2h. The top
left quadrant shows integrin-ligand-dye emission (green), top right quadrant shows bright
field image of the cells and bottom left quadrant shows overlay of dye emission with
bright field image.

Distribution of CIL-1 and CD-1 with the population of cells was measured by
flow cytometry. The shift in the population of the cells indicates active uptake of CIL-1
and ABL-1, but not CD-1 (Figure 4.35). The absence of active uptake of CD-1 by
WM115 cells suggests that it is the integrin-ligand that is mediating the uptake of the
conjugate. This observation is also consistent with the intravital microscopy studies
performed at PRISM with CIL-1 and CD-1 in mouse subcutaneous tumor xenografts
models.
180

Figure 4.35 Distribution and uptake of CIL-1, ABL-1 and CD-1 by flow cytometry. A)
WM115 cells; B) WM115 cells with CIL-1; C) WM115 cells with ABL-1, and D)
WM115 cells with CD-1.

181
4.11 Conclusion
We designed bifunctional and trifunctional integrin BNCT ligand conjugates which hold
considerable promise for future applications for cancer imaging and therapy. In the first
part of this Chapter we described the design and biological application of a novel low
molecular weight fluorescent sensor,

ABL-1, selective for α
v
β
3
receptors. It exhibits 5-
fold fluorescence enhancement upon binding to the cell surface. It also shows low
cytotoxicity within the large range of concentrations and has good water solubility.
Confocal microscopy and flow cytometry experiments showed active uptake of ABL-1
and its selectivity towards vitronectin, a natural ligand of α
v
β
3
receptors. Our designed
α
v
β
3
-selective fluorescent probe, ABL-1, showed a noticeable accumulation in a
subcutaneous tumor xenografts models. Due to its ability to bind specifically to α
v
β
3

receptors, it could be used in visible-light imaging of tumor cells that display these
receptors as well as of newly formed tumor vasculature. This opens new avenues for
application of this ligand in integrin-targeted fluorescence-based optical imaging. The
future work is aimed to continue expanding our set of available fluorescent dyes capable
of emission in the red and near-infrared regions as well as to new monomeric and
multivalent ligands targeting other cell surface receptors. The structural framework of
integrin-specific ligand conjugated to cyanine dye forms a viable platform with potential
for future cancer diagnostics. The second part of this Chapter describes the design and
preliminary biological evaluation of the trifuctional ligand, CIL-1 that can be used in
BNCT. Our observed significant accumulation of CIL-1 in the in vivo tumor models and
the absence of the observed accumulation of CD-1 indicate the selectivity of the targeting
ligand. The success of the application of our trifunctional system in BNCT will depend
182
upon the availability of the neutron-generating equipment and our ability to choose
proper animal models of the disease.

183

Chapter 5: Targeting monoamine oxidase A (MAOA) in prostate cancer

184
5.1 Introduction to monoamine oxidase (MAOA)
Monoamine oxidases (MAO) are the flavin-containing, mitochondrial membrane-bound
enzymes that catalyze the oxidative deamination of the dietary amines to corresponding
aldehydes. They were first discovered in 1928 by Mary Bernheim and termed as tyramine
oxidases (Hare 1928). The aldehydes formed during the oxidative process will, in turn,
rapidly oxidize to the corresponding acids by another enzyme, aldehyde dehydrogenase
(Tipton, Boyce et al. 2004). The primary substrates for MAOs are monoamine
neurotransmitters, such as serotonin (5-HT), dopamine (DA), epinephrine (EP),
norepinephrine (NE) and β-phenylethyl amine (PEA). MAO exists in two isoforms,
MAOA and MAOB, with 70 % sequence similarity between each isoform. Both isoforms
differ in their substrate specificity and inhibitor sensitivities (Tsugeno and Ito 1997).
Norepinephrine (NE) and serotonin (5-HT) are the major substrates for MAOA and β-
phenylethyl amine is for MAOB. It has been shown by point mutation, that when a single
amino acid, Ile335 in MAOA and Tyr326 in MAOB, were interchanged reciprocally then
the substrate and inhibitor sensitivities were reversed (Geha, Rebrin et al. 2001). These
enzymes differ in their substrate selectivity, which varies with the concentration of the
substrate, as well as the turnover rate of the enzyme and the concentration of enzyme.
Sometimes in the absence of MAOA enzyme, MAOB will oxidize the MAOA-specific
substrate. Both enzymes require flavin adenine dinucleotide (FAD) as a cofactor which
makes a covalent bond with the cysteine of the common pentapeptide sequence via a
thioether linkage. In the brain MAOA is predominantly present in catecholaminergic
neurons and MAOB is present in the histaminergic and serotoninergic neurons. During
the transmission of electric signals between neurons, MAO inactivates the excess
185
neurotransmitters by flushing them out of the synaptic gap. Insufficient amount of MAO
would lead to the excess concentration of serotonin in the synaptic cleft, leading to the
transmission of uncontrolled signals. On the other hand, the excess of MAO would not
leave enough serotonin for proper communication, which would lead to depression or
Parkinson’s disease (Riederer and Laux 2011). MAO oxidizes the amines present in the
blood and protects the body from its toxic effects. Similarly, in the brain, it oxidizes
neurotransmitters and normalizes the levels of hormones required for normal activity.
The major substrate of MAOA activity, 5-HT or serotonin regulates the mood, sleep and
memory-related functions (Figure5.1). In the MAOAdeficient mouse pups, the levels of
serotonin were nine-fold higher than the wild type pups (Shih, Chen et al. 1999).
Correlation of the deficiency in MAOA activity and aggressive behavior has been
reported in several studies. This abnormal aggression in pups is consistent with the
aggressive behavior of males of a Dutch family having complete deficiency of MAOA
because of a point deletion in MAOA gene. In the same way, excess of MAOA is
responsible for depression in humans. Though not important for survival, it has been
shown that MAOA is essential during the developmental phase. MAOA knockout
studies showed increased levels of 5-HT and serotonin which is responsible for the
aggressive phenotype. These studies also showed the importance of maintained levels of
5-HT in the normal development of thalamocortical axons and the aggregation of neurons
to form barrels (Rebsam, Seif et al. 2005).
Given the important role MAOA plays in the behavior, several inhibitors were
developed and identified as anti-depressants, which can inhibit the activity of this enzyme
and maintain the relevant levels of neurotransmitters. Due to the difference in the
186
substrate specificity, it is possible to develop inhibitors which can selectively inhibit the
activity of MAOA or MAOB. For example, chlorgyline is found to be a specific inhibitor
of MAOA and deprenyl is identified as MAOB inhibitor.

Figure 5.1 Mechanism of action of MAO and MAOI. MAO regulates the levels of
neurotransmitters by oxidizing amines to aldehyde and further to acid. While in presence
of MAOI, neurotransmitters were not degraded and remain active for longer time.
5.2 Monoamine oxidase inhibitors (MAOI)
Monoamine oxidase inhibitors work by inhibiting the activity of MAO and thereby
increasing the concentration of monoamine neurotransmitters. Several reversible and
187
irreversible inhibitors were developed since 1952. The very first non-selective MAO
inhibitor was a hydrazine based compound, iproniazid, which was initially developed for
the treatment of tuberculosis, but later found to have properties of anti-depressant
(Fagervall and Ross 1986). Later, many hydrazine inhibitors were developed as anti-
depressants, but were withdrawn due to the liver toxicity, hemorrhage, hypertensive crisis
and other serious side effects. Further development of the non-hydrazine-based, selective
inhibitors rectified some of the issues of non-specific toxicities. The use of MAOIs,
nevertheless, placed restriction on the consumption of specific kind of foods.

Figure 5.2 Representative structures of MAO inhibitors.
MAOIs inhibit the breakdown of dietary amines and thus in the presence of
MAOI, consumption of tyramine-containing foods could lead to hypertensive crisis (Da
Prada, Zurcher 1988). On the other hand, selective inhibitors of MAOB in right amount
of dosage and reversible inhibitors of MAOA solved this problem. Most of the tyramine
188
is metabolized by MAOA so that the selective inhibition of MAOB does not interfere
with the MAOA activity (Hasan, McCrodden 1988).
Also, reversible inhibitors of MAOA such as, moclobemide and lazabemide
(Youdim, Edmondson et al. 2006), (Figure 5.2) allowed sufficient inhibition of its
activity and hence, tyramine could replace the inhibitor from the enzymes active site for
the active metabolism (Da Prada, Kettler et al. 1990). Several structural studies have been
performed in the past to better understand the substrate and inhibitor selectivity for
MAOA and MAOB. Structural studies of MAOs with its inhibitor adduct revealed that
rat MAOA (rMAO A) and human MAOB (hMAO B) were quite similar, as they both
form dimer while human MAOA (hMAO A) crystallizes as monomer (De Colibus, Li et
al. 2005). Both rMAOA and hMAOB contains a glutamic acid residue at position 151
and is mediating the dimer formation while this residue was replaced by lysine in
hMAOA. Active site of hMAOA is a hydrophobic cavity of 550 Å
3
which is smaller than
the hMAOB (700 Å
3
). Structural data showed that inhibitors covalently bind to the N5 of
the cofactor FAD which is situated between two tyrosine residues. MAOA active site has
more extended conformation as compared to MAOB and major alteration was observed
in the loop region bearing amino acid residues from 210 to 216. This difference in the
loop conformation is primarily mediating the selectivity in the substrate and inhibitor
recognition by the two enzymes.

5.3 The role of monoamine oxidase A (MAOA) in prostate cancer
In the United States, prostate cancer is the second most common leading cause of death
from cancer in men (http://www.cdc.gov/cancer/prostate/). The severity of prostate
189
cancer is recognized by the degree of tumor differentiation or in other terms, by Gleason
scoring. High levels of prostate specific antigen (PSA) in blood represent the disruption
of the normal prostate structure. Conventional treatments of prostate cancer involve the
surgical removal of prostate and, in most advanced cancer; it is followed by androgen
deprivation therapy. Although the androgen deprivation therapy or hormonal therapy
initially reduced the levels of PSA, in most cases it is followed by the recurrence of the
disease (Shen and Abate-Shen 2010). Androgen independent or castration resistance
tumors pose a great challenge in the prognosis of treatment of prostate cancer. Recently,
few reports suggested that PSA screening is not sufficient for distinguishing between low
Gleason grade and Gleason 4/5 grade prostate tumors (Wolf, Wender et al. 2010). There
is an unmet need of biomarkers to be identified that can distinguish between low and high
grade tumors for the successful treatment of prostate cancer. Another challenge
associated with this disease is its fast progression to bone. Most of the tumors rarely
metastasize to bone, while high grade (Gleason score 4/5) prostate tumor metastasize to
bone and is the major reason for the patient mortality dealing with prostate cancer
(Logothetis and Lin 2005).
A cDNA microarray analysis of Gleason grade 3 and 4/5 prostate tumor identified
monoamine oxidase A (MAOA) as one of the highly expressed enzyme in Gleason4/5
tumor cells and they remain at lower level in low grade prostate tumors (True, Coleman
et al. 2006). MAOA is also overexpressed in basal normal prostatic epithelium but not in
secretory cells. Work by Peehl and coworkers showed that MAO inhibition induces the
differentiation of basal prostatic epithelial towards secretory cells (Peehl, Coram et al.
2008). Same group also showed that MAOA inhibition reduces the growth of prostate
190
cancer (PCa) cells in vitro as well as in vivo. They showed that an irreversible inhibitor of
MAOA, chlorgyline, induces differentiation and inhibits several oncogenic pathways
such as, MAPK, beta-catenin in castration-resistant and more advanced-stage VCaP cells
(Flamand, Zhao et al. 2010). It has also been shown by our collaborators that inhibition or
silencing of MAOA significantly reduces the growth of prostate cancer in both androgen
sensitive and androgen insensitive cells. They also reported the role of MAOA in
inducing epithelial-to-mesenchymal transition (EMT) in human PCa cells.
Reactive oxygen species (ROS) are the common mediators of oxidative stress
which can induce irreversible damage to the mitochondrial DNA and to other
biomolecules of the cell. The major by-product of the MAO activity, hydrogen peroxide,
is a potential source of reactive oxygen species (Bortolato, Chen et al. 2008). A growing
evidence suggests the connection of ROS and tumor initiation and progression. The ROS
acts as second messenger and can initiate downstream signaling cascades which are
associated with different oncogenic phenotypes. Prostate cancer has shown potential
association with increased ROS levels (Kumar, Koul et al. 2008). Blocking the ROS
activity by using ROS scavengers attenuated the cell migration and invasion in LNCaP
cells.
Given the important role MAOA plays in the human prostate cancer growth and
metastasis targeting it specifically, especially in castration-resistant and bone metastatic
advanced prostate cancer, would provide us with improved prostate cancer therapy. In an
effort to target MAOA, in collaboration with Prof. Jean Shih at USC, we have designed a
NIR dye-MAOA inhibitor conjugate which can specifically target prostate tumors and
determined its uptake, in vitro and in vivo.
191
5.4 Near infrared dyes (NIR) for optical imaging and cancer targeting
Recently, much attention has been paid in the development of imaging modalities that are
non-invasive and require relatively simple equipment for imaging. The design and
development of multifunctional near infrared dyes (NIR) is a significant improvement
towards the optical imaging and cancer targeting.

Figure 5.3 Structure of NIR dyes.
NIR dyes allow the fluorescence imaging in 700-1000 nm range and efforts are being
directed toward the design of these dyes so that they show minimum inference with
absorption and emission spectral range. Initially, NIR dyes were developed as
nanoparticles with inorganic molecules and used for the imaging of biological samples;
however toxicity of heavy metals restricted their frequent use. On the other hand, NIR-
192
emitting fluorescent probe based on organic molecules showed enhanced improvement in
chemical and photophysical properties with a capability of conjugation with targeting
moieties. Figure 5.3 shows examples of several NIR dyes and their spectral
characteristics.
Cyanine dyes are small organic molecules with two aromatic nitrogen-containing
heterocycles linked by a polymethine bridge. Trimethine cyanine dyes generally show
absorption in the visible region, but with increasing number of methine groups (CH=),
the absorption shifts toward 700-1000 nm range with a molar absorption coefficient
reaching approx. 200,000 mol
-1
cm
_1
L. Squaraine dyes have a oxocyclobutenolate core
with aromatic or sometimes heterocyclic components at both ends of the molecules. They
showed high photophysical and chemical stability. BODIPY dyes are another examples
of NIR dyes, which depending on their structure show absorption in the range 650-800
nm range (Luo, Zhang et al. 2011). These dyes have a very interesting property of
selective uptake and retention by cancer cells. The mechanism by which these dyes show
preferential accumulation in tumor cells is not very clear. However, it has been suggested
that the higher mitochondrial membrane potential of tumor cells as compared to normal
cells facilitates the uptake of delocalized lipophilic cation and it has been shown that
these multifunctional NIR dyes, which are lipophilic cationic molecules, selectively
accumulate in the mitochondria of tumor cells (Zhang, Zhang et al. 2011). Another theory
supports the selective uptake of NIR dyes due to the overexpression of some specific
transporter in tumor cells but not in normal cells.

193
5.5 Design and biological evaluation of NIR dye-MAOA inhibitor conjugate
The broad aim of this project, which is beyond the scope of this Thesis, is to understand
the role of MAOA in the progression of prostate cancer and bone metastasis. As a part of
this project, we along with our collaborators, proposed to design a novel NIR dye-MAOA
inhibitor conjugate which target specifically PCa cells without affecting normal cells.
This bifunctional system has the NIR imaging capability as well as has therapeutic
potential to target prostate cancer.

Figure 5.4 Structure of MHI-Clg conjugate. MHI-148 was conjugated to clorgyline, a
potent inhibitor of MAOA.
Recently, Chung and coworkers reported the preferential uptake and retention of two
hepatamethine cyanine dyes, IR-783 and MHI-148, in tumor cells and tissues as compare
to normal cells (Yang, Shi et al. 2010). They showed that both of the dyes have a great
potential to specifically target cancer cells in vitro as well as in vivo. These dyes can be
194
used to detect metastasis and cancer cells in blood with high sensitivity (10 cells per 1
mL of blood). To explore the potential of the bifunctional system, we chose clorgyline, a
potent inhibitor of MAOA and conjugated it with a NIR dye MHI-148 (Figure 5.4). In
this Chapter, I will discuss the preliminary studies conducted on the MHI-chlorgyline
(MHI-Clg) conjugate. The detailed biological studies are currently underway.
5.5.1 Effect of MHI-Clg on cell viability
As we are interested in investigating the role of MAOA in the prostate cancer
progression, it is legitimate to see the effect of MHI-Clg in androgen–sensitive as well as
in androgen-insensitive cells. For this study, we chose three different prostate cancer cell
lines: LNCap, C4-2B and PC-3. LNCaP is an androgen-sensitive prostate cancer cell line
which was established from a metastatic lesion of human prostatic adenocarcinoma and
has very low tumorigenicity. C4-2B is a bone metastatic subline which was derived from
LNCaP and is androgen-insensitive. PC-3 is also an androgen-insensitive cell line which
has high rate of tumorigenicity. To measure the effect of MHI-Clg on the viability of
these cells we performed a colorimetric based MTS assay. This assay is composed of a
tetrazolium compound (3-(4,5-dimethylthiazole-2-yl)-5-(3-carboxymethoxyphenyl)-2(4-
sulphophenyl)-2H-tetrazolium salt and an electron coupling reagent, phenazine
methosulfate, which measures the metabolic activity of cells. Before dosing cells with
MHI-Clg, a cell count study was performed to ensure the appropriate number of cells for
MTS assay.
195

Figure 5.5 Cytotoxicity of MHI-Clg in LNCaP cells.

Figure 5.6 Cytotoxicity of MHI-Clg in C4-2B cells.
196
Based on the cell count study, 4000 cells/well of LNCaP, 8000 cells/well of C4-
2B and 2000 cells/well of PC-3 were plated in 96 well plate. Cells were allowed to form a
monolayer and then dosed with increasing concentration of MHI-Clg from 10 nM to 12
µM for 96 h at 37 °C and in 5% CO
2
atmosphere. After 4 days or 96 h incubation, MTS
reagent was added to the cell and further incubated for 3 h. Number of viable cells was
measured by reading the absorbance of formazan at 490 nm. MHI-Clg showed a IC
50
value of 5.10 µM in LNCap cells (Figure5.5), 5.61 µM in C4-2B (Figure5.6) and 6.13
µM in PC-3 cells (Figure5.7).

Figure 5.7 Cytotoxicity of MHI-Clg in PC-3 cells.

197
To ensure that fluorescent dye is non-toxic to cells at biologically relevant
concentrations, we measured the effect of MHI-148 alone on cells by MTS assay. No
significant toxicity was observed in the tested concentration range.

5.5.2 Assessment of MHI-Clg uptake in C4-2B cells by confocal microscopy
The in vitro imaging experiments were performed with C4-2B prostate cancer cells. Cells
were plated and allowed to adhere overnight in MatTek dishes. Next, cells were exposed
to MHI-Clg and MHI-148 for 30 min at 37 °C in a 5% CO
2
atmosphere.

Figure 5.8 Uptake of MHI-Clg by C4-2B cells. Top left quadrant showed the uptake of
Mitotracker, middle quadrant is the bright field image, top right quadrant showed the
uptake of MHI-Clg. Bottom left quadrant showed the staining of nucleus by DAPI and
bottom right quadrant is the overlay of all images.
The uptake of dye-conjugate was observed by Zeiss confocal laser scanning microscope
using 63x oil-immersion objective (Figure 5.8). This system was equipped with a 37 °C
198
stage warmer and constant CO
2
perfusion. The 633 nm laser line was set at 10% power
for the excitation of MHI-148. Since this dye has shown to have preferential uptake in
mitochondria of cancer cells, we used a mitochondrial tracking dye (Mito Tracker green)
to determine the localization of the dye-conjugate. After 40 minutes of incubation, there
was an active uptake of the dye-conjugate in C4-2B cells at 5 µM concentration. The co-
localization of MHI-Clg and Mitotracker green supports the preferential uptake of MHI-
Clg by mitochondria.
5.5.3 Uptake and accumulation of MHI-Clg in tumors in live mice
To determine the uptake and accumulation of MHI-Clg in vivo, imaging was performed
on nude tumor-bearing mice. This experiment was done at the Molecular imaging center
(MIC), Keck School of Medicine, USC.

Figure 5.9 In vivo NIR imaging of MHI-Clg. MHI-Clg was injected at a concentration
of 0.5 mg/kg.
199
The human prostate cancer cells, C4-2B, were implanted subcutaneously in 4 to 5 week
old mice and when the tumor size reached between 2-5 mm, MHI-Clg was injected i.t at a
concentration of 0.5 mg/kg. Due to its selective uptake, MHI-Clg rapidly accumulated in
the tumor tissue with a very low background signal.

5.6 Conclusion
MAOA, a mitochondrial enzyme, is a novel target in prostate cancer progression. In this
chapter, we along with our collaborators proposed a novel NIR dye-MAOA inhibitor
conjugate for the specific imaging and treatment of prostate cancer. From our preliminary
studies, this conjugate has shown promising capability to image and target MAOA
overexpressed prostate cancer cells, in vitro as well as in vivo. The overexpression of
MAOA is also presumed to have role in the stabilization of HIF-1α, which could result in
the overexpression of HIF-inducible genes and, therefore, accelerated progression of
prostate cancer and poor prognosis for patients. Detailed mechanistic studies concerning
the role of MAOA in the progression and metastasis of prostate cancer are currently
underway.

200

Chapter 6: Experimental Section
201
6.1 Introduction to Experimental Section

Materials and Methods

Cell Lines
MDA-MB-231, MCF-7, HeLa, MDA-231-HRE-Luc, A549, PC3, LNCaP, C4-2B.
MDA-MB-231, MCF-7, HeLa and A549 cells were obtained from ATCC.
MDA-231-HRE-Luc was a generous gift from Prof. Robert Gillies.
PC3, LNCap and C4-2B cells are obtained from Prof. Leland Chung at the Cedars Sinai
Medical Center.

All cell lines were split and reseeded every three or four days. This process involves
washing cells with PBS, treatment with 1× trypsin in EDTA and then resuspending cells
in fresh media. Cells were collected by centrifugation and then split in the desired ratio
with fresh media in a T-75 flask with final volume of 15 mL. After three passages, one
portion of cells were saved and stored in cryostorage for future use.

General synthetic methods
All reactions involving moisture-sensitive reagents were conducted under argon
atmosphere and flame-dried glassware. Hygroscopic liquids were transferred into
reaction vessels via a syringe through rubber septa. All reagents and solvents were
obtained from commercial sources and were used as received unless otherwise stated.
Dry DCM was freshly distilled with calcium hydride and dry THF was obtained by
distillation with sodium and benzophenone. Column chromatography was performed on
202
silica gel (230–400 mesh) using reagent grade solvents. Analytical thin-layer
chromatography(TLC) was performed on glass-backed, precoated plates (0.25 mm, silica
gel 60, F-254, EM Science).Reaction product solutions were concentrated using a rotary
evaporator at 30–150 mmHg. Nuclear magnetic resonance (NMR) spectra were collected
on Bruker 500 MHz and Varian unity 300 MHz instrument in the indicated solvents.
Mass spectra were obtained from the Mass Spectrometry Laboratory in the Department of
Chemistry, University of Arizona.

203
6.2 Experimental section of Chapter 2
6.2.1 Synthesis of mono-ETP compounds
ETP intermediates 1 to 5 were synthesized as mentioned in the literature: Fukuyama, T.;
Nakatsuka, S.; Kishi, Y. Tetrahedron 1981, 37, 2045–2078.

1,5-bis((benzyloxy)methyl)-3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-
diazabicyclo[3.2.2]nonane-7,9-dione (9)

A solution of 3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-
diazabicyclo[3.2.2]nonane-7,9-dione 5 (227 mg, 0.70 mmol) and benzyloxymethyl
chloride (1 mL, 3.5 mmol) was dissolved in dry THF and was cooled to -78 °C. To the
stirred solution of 5 in THF, n-butyllithium (1 mL, 1.54 mmol) in hexane was added
dropwise over a period of 10 min. Reaction mixture was then allowed to stir at room
temperature for another 30 min. A saturated NaCl solution was added into the reaction
mixture and crude product was extracted with DCM (30 mL, 3×). Organic extract was
then dried over MgSO
4
and evaporated under reduced pressure. Product 9, was then
purified by column chromatography on silica gel with DCM: EtOAc (8:2) in 47 % yield.
1
H NMR (300 MHz, CDCl
3
) δ 3.16 (3H, s), 3.25 (3H, s), 3.78 (3H, s), 3.79 (1H, d, J=6
Hz), 3.81 (1H, d, J=6 Hz), 4.26 (1H, d, J=16 Hz), 4.45 (1H, d, J=16 Hz), 4.54-4.78 (4H,
m), 4.95 (1H, s), 6.84 (2H, m), 7.29 (12H, m). MS (FAB) calculated for C
30
H
32
N
2
O
2
S
2
:
564.1753, found: 564.937.
1,4-bis((benzyloxy)methyl)-5,7-dimethyl-2,3-dithia-5,7-diazabicyclo[2.2.2]octane-
6,8-dione (mono-ETP1)
m-Chloroperbenzoic acid (13 mg, 77%, 0.0586 mmol) was added to an ice-cold solution
of 9 (24 mg, 0.0425 mmol) in anhydrous DCM (6 mL) with stirring. After 10 min of
stirring at 0 °C, 10 µLof dimethyl sulfide was added, followed by treatment with 20 µL
solution of 70% perchloric acid in methanol (1:5). The solution was allowed to stand at 0
204
°C for 18 h and then the reaction mixture was evaporated under vacuum. The solid
residue was dissolved in MeOH and purified by HPLC to afford ETP-1 in 25 % yield.
1
H NMR (300 MHz, CDCl
3
) δ 3.14 (6H, s), 4.24 (4H, d), 4.71 (4H, d), 7.33 (10H, m).
13
C
NMR (125 MHz, CDCl
3
) 28.11, 67.24, 74.41, 74.72, 127.95, 128.17, 128.57, 136.75,
164.86. HRMS (FAB) calculated for [C
22
H
25
N
2
O
4
S
2
+]
: 445.1177 found: 445.11.
1-((benzyloxy)methyl)-3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-
diazabicyclo[3.2.2]nonane-7,9-dione (6)
A solution of 5 (1.9 g,, 5.86 mmol) in dry THF (200 mL) was cooled to -78 °C. 2.5 M of
n-BuLi in hexane (3.6 mL, 9 mmol, ) was added dropwise over a period of 5 min. Next,
benzyloxymethyl chloride (3.32 mL, 24 mmol, 4 eq, 60 % reagent) in dry THF was
added to the resulting mixture. Reaction mixture was stirred for additional 20 min and
reaction progress was monitored by HPLC. The mixture was then allowed to warm to
room temperature for 2 h. A saturated solution of NaCl was added and stirring was
continued for another 10 min. Most of the THF was removed under reduced pressure and
the residue was dissolved in DCM. Organic extracts were combined, dried over MgSO
4
and evaporated under reduced pressure. Crude product was obtained by precipitation with
hexane: diethyl ether (1:1) mixture. Compound 6 was then purified by column
chromatography on silica gel with DCM:hexane: ethyl acetate (5:4:1) in 53 % yield.
1
H
NMR (500 MHz, CDCl
3
): 3.10 (3H,s), 3.22 (3H, s), 3.78 (3H, s), 3.81 (1H, d, J = 10.5
Hz,), 4.22 (1H , d, J = 10.5 Hz), 4.54 (1H, d, J = 11.2 Hz), 4.74 (1H, d , J = 11.2 Hz),
5.04 (1H, s), 5.11 (1H, s), 6.84 (2H, d, J = 8.7 Hz), 7.31 (7H, m).

205
1-((benzyloxy)methyl)-5,7-dimethyl-2,3-dithia-5,7-diazabicyclo[2.2.2]octane-6,8-
dione (mono-ETP-2)
Meta-Chloroperbenzoic acid (5.5 mg, 77%, 0.0248 mmol) was added to an ice-cold
solution of 6 (9.2 mg, 0.0207 mmol) in anhydrous DCM (4 mL) with stirring. After 10
min of stirring at 0 °C, 4.6 µL of dimethyl sulfide was added, followed by treatment with
9 µLsolution of 70% perchloric acid in methanol (1:5). The rest of the steps were carried
out as discussed in the experimental details of ETP-1 synthesis. ETP-2 was obtained in
50 % yield.
1
H NMR (300 MHz, CDCl
3
) δ 3.12 (3H, s), 3.13 (3H, s), 4.22 (2H, d) 4.72
(2H, d), 5.65 (1H, s), 7.37 (5H, m). HRMS (ESI) calculated for C
14
H
16
N
2
O
3
S
2
Na
+
:
347.0500 found: 347.1073.
1,5-bis((hydroxy)methyl)-3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-
diazabicyclo[3.2.2]nonane-7,9-dione (10)
A solution of 9 (40.0 mg, 0.084 mmol) in dichloromethane was cooled to 0 °C. To the
stirred solution 1M boron trichloride (150 µL, 0.151 mmol) in DCM was added dropwise
over a period of 30 sec. Reaction mixture was allowed to stir at 0 °C for 15 min and then
poured into ice water. The aqueous phase was extracted with DCM (50 mL, 3×). Organic
extracts were combined, dried over MgSO4 and evaporated under vacuum and purified
by HPLC to afford 10 in 93 % yield.
1
H NMR (300 MHz, CDCl
3
) δ 3.21 (3H , s), 3.33
(3H, s), 3.79 (3H, s), 3.83 (1H, d, J=12.6 Hz), 4.00 (1H, d, J=12.6 Hz) , 4.32 (1H, d,
J=12.6 Hz), 4.64 (1H, d, J=12.6 Hz), 5.01 (1H, s), 6.85 (2H, m), 7.29 (2H, m).

206
(3-(4-methoxyphenyl)-6,8-dimethyl-7,9-dioxo-2,4-dithia-6,8-
diazabicyclo[3.2.2]nonane-1,5-diyl)-bis(methylene)diactetate (11)
Compound 10 (3mg, 0.0078 mmol) was dissolved in 500 µl of dichloromethane. To the
stirred solution, 100 µL of pyridine and 100µL of acetic anhydride were added and the
reaction mixture was stirred at room temperature for 16 hours. Reaction was monitored
by HPLC. After completion, ice cold water was added and stirred for another 2 h. The
organic layer was washed with a saturated solution of NaHCO
3
. Organic extracts were
combined, dried over MgSO
4
, evaporated under vacuum and purified by HPLC to afford
11 in 54% yield.
1
H NMR (300 MHz, CDCl3) δ 2.11 (6H, s), 3.10 (3H, s), 3.25 (3H, s),
3.79 (3H, s), 4.43 (2H, d, J=12.3 Hz), 4.92 (1H, d, J=11.7 Hz), 5.00 (1H, s), 5.03 (2H, d,
J=12.3 Hz), 6.86 (2H, m), 7.30 (2H, m).
(5,7-dimethyl-6,8-dioxo-2,3-dithia-5,7-diazabicyclo[2.2.2]octane-1,4-diyl)-
bis(methylene)diactetate (mono-ETP-3)
Meta-Chloroperbenzoic acid (17.0 mg, 77%, 0.077 mmol) was added to an ice-cold
solution of 11 (30.0 mg, 0.064 mmol) in anhydrous dichloromethane (8 mL) with stirring.
After 10 min of stirring at 0 °C, dimethyl sulfide (10 µL) was added, followed by
treatment with 20 µL solution of 70% perchloric acid in methanol (1:5). The rest of the
steps were carried out as discussed in the experimental details of ETP-1/LS52 synthesis.
ETP-3 was obtained in 50 % yield.
1
H NMR (300 MHz, CDCl3) δ 2.16 (6H, s), 3.13
(6H, s,), 4.75 (2H, d, J=12.6 Hz), 4.97 (2H, d, J=12.6 Hz).
13
C NMR (125 MHz, CDCl
3
) δ
20.61, 28.05, 60.46, 73.70, 163.90, 169.52. MS (FAB) calculated for C
12
H
17
N
2
O
6
S
2
+
:
349.0528 found: 349.11.

207
Tert-butyl 3-((5-((benzyloxy)methyl)-3-(4-methoxyphenyl)-6,8-dimethyl-7,9-dioxo-
2,4-dithia-6,8-diazabicyclo[3.2.2]-nonan-1-yl)-1H-indole-1-carboxylate (12)
1 M solution of LHMDS in THF (1.0 mL, 1.0 mmol) was added dropwise over a period
of 2 min to a solution of 6 (0.30 g, 0.68 mmol) in dry THF (15 mL) at -78 °C with
constant stirring. The tert-butyl-3-bromomethyl-indole-1-carboxylate 13 (178 mg, 0.57
mmol), dissolved in 2 mL of THF, was then added dropwise into the reaction mixture and
the solution was allowed to warm up to room temperature for 3 h. Water was added into
the reaction and the mixture was extracted with dichloromethane (3×, 50 mL). Organic
extracts were combined, dried over anhydrous MgSO4 and concentrated under reduced
pressure to obtain crude product 12. The product was purified by column
chromatography on silica gel with hexane: DCM: ethyl acetate (6:3:1). Yield 214 mg of
12 (46%).
1
H NMR (500 MHz, CDCl3) δ 1.63 (9H, s), 3.04 (3H, s), 3.28 (1H, s), 3.33
(3H, s), 3.82 (3H, s), 3.87 (1H, d), 4.15 (1H, d, J = 11Hz), 4.32 (1H, d, J = 11 Hz), 4.57
(1H, d, J = 12 Hz), 5.12 (1H,s), 4.78 (1H, d, J = 12 Hz), 6.87 (2H, d, J = 9 Hz), 7.38 (9H
,m), 7.52 (1H, d, J = 7 Hz), 8.15 (1H, s).
(1,6)-1-((1H-indol-3-yl)methyl)-6-(benzyloxy)methyl)-7,9-dimethyl-2,3,4,5-tetrathia-
7,9-diazabicyclo[4.2.2]decane-8,10-dione (ETP-5a/ETP5b)
Meta-Chloroperbenzoic acid (100 mg, 77%, 0.45 mmol) was added to an ice-cold
solution of 12 (255 mg, 0.38 mmol) in anhydrous dichloromethane with constant stirring.
After 10 min of stirring, dimethyl sulfide (100 L) was added followed by treatment with
200 µL of a solution of 70% perchloric acid in methanol (1:5). The solution was allowed
to stand at room temperature for 9 h. A saturated solution of NaHCO
3
was poured into the
reaction mixture and was extracted with dichloromethane (3×, 30 mL). Organic extracts
208
were dried over anhydrous MgSO
4
, filtered and concentrated under reduced pressure to
give crude ETP-5 as a mixture of a disulfide ETP-5a tetrasulfide ETP-5b and a small
amount of monosulfide product. Mixture was separated by column chromatography on
silica gel using hexane: DCM: ethyl acetate( 6:3:1) to afford ETP-5b.
1
H NMR (DMSO-
D6) δ: 3.8 (1H, d, J = 15 Hz), 4.05 (1H, d, J = 10 Hz), ), 4.07 (1H, d, J = 15 Hz), 4.50
(1H, d, J = 10 Hz), 4.53 (2H, dd, J = 10 Hz, 15 Hz), 6.34 (1H,s), 6.74 (1H, d, J = 5 Hz),
7.06 (1H, d, J = 3 Hz), 7.08 (1H, d, J = 3 Hz), 7.12 (1H, d, J = 10 Hz), 7.40 (2H, d, J = 5
Hz), 7.55 (1H, d, J = 5 Hz), 7.50 (3H, m).
1-((1H-indol-3-yl)methyl)-6-(hydroxymethyl)-7,9-dimethyl-2,3,4,5-tetrathia-7,9-
diazabicyclo[4.2.2]decane-8,10-dione (mono-ETP-5)
1M solution of boron trichloride in DCM (110 µ L, 0.11 mmol) was added dropwise to a
solution of ETP-5 (20 mg, 0.038 mmol) in DCM (5 mL) at 0 °C. Reaction mixture was
stirred for additional 10 min and then the mixture was poured into ice-cold water (5 mL)
and extracted with dichloromethane (25 mL). Organic extracts were combined, dried over
MgSO
4
and evaporated under reduced pressure to obtain crude product as a white solid.
The crude product was purified by semi-preparative thin layer chromatography on silica
gel with hexane: DCM: ethyl acetate (5:4:1) to afford 3. Yield 12% over 3 steps from 12.
1
H NMR (CDCl3) δ: 3.11 (3H, s), 3.15 (3H, s), 3.51 (1H, d, J = 16 Hz), 3.97 (1H, d, J =
12 Hz), 4.10 (1H, d, J = 16 Hz), 4.35 (1H, d, J =12 Hz), 7.07 (1H, m), 7.13 (2H, m),
7.33 (1H, d, J =5 Hz), 7.54 (1H, s), 7.57 (1H, d, J = 10 Hz). MS (ESI): calculated for
C
16
H
17
NaN
3
O
3
S
+
4
[M+Na]
+
: 450.0, found 449.9.

209
6.2.2 Protein expression
Single stranded oligonucleotide sequence corresponding to human p300 CH1 domain
(aa323-423) alongwith GST tag was designed and cloned (by Genscript Inc) into puc57
plasmid. Plasmid was transformed into JM109 competent cells (Promega) and the gene
sequence was subcloned into the expression vector pGEX 4T-2 between BamH1 and
EcoR1 restriction sites. After the ligation, the pGEX 4T-2-p300 fusion vector was
transformed into BL21 DE3 pLys competent E. coli (Novagen). Transformation into
JM109 and BL21 DE3 pLys competent E. coli cells was performed at 41.5 °C for 50 sec.
Cells were spread onto the agar plates containing ampicillin and placed in the incubator at
37 °C for 12-14 hrs. Presence of the desired insert (p300-CH1-GST) was confirmed by
DNA sequencing.
For protein expression, a single colony was picked from the E.coli competent
cells with an autoclaved toothpick and was allowed to grow for 12-14 hrs in a shaker at
200 rpm at 37 °C in 20 mL of Luria-Bertani (LB) media. After overnight, 20 mL culture
was then added to the 1L of previously autoclaved LB media and allowed to grow until
the optical density (OD) of 0.7 was reached. After which isopropyl β-D-1-
thiogalactopyranoside (IPTG, 200 µM) was added to induce the protein expression. The
culture with the LB media was then centrifuged at 4400 rpm and pellet was resuspended
in the 50 mL of lysis buffer with 50 mM Tris (Sigma), 150 mM NaCl (Fisher), 100 μM
ZnCl
2
(Sigma), 1 mM EDTA (Fisher), 10 mM MgCl
2
(Fisher), 1 mM DTT (Fisher), 0.1%
NP40 (Sigma), 50 μg/mL RNase A (Sigma), and 50 μg/mL DNase A (Sigma) at pH 8.0.
After one freeze- thaw cycle, the lysate was sonicated at 40% amplitude for 8 minutes,
divided into 20 second pulses. The sonicated lysate was then centrifuged in a Nalgene
tube at 18,000 x g for 45 min at 4 °C and supernatant was stored at -80 °C for O/N. Next,
210
1.33 mL of glutathione sepharose beads 4B (GSH) were added to the supernatant and
rotated o/n at 4 °C. Glutathione sepharose beads 4B (GE Health Care) were prepared
according to manufactures protocol.

6.2.3. Purification of fusion protein p300-CH1-GST
Fusion protein p300-CH1-GST was purified by using increasing concentrations of
reduced glutathione in the range of 2.5 mM to 10 mM. First, supernatant-GSH bead
mixture was passed through the column and flow through was collected. Then beads
were washed three times with ice cold PBS, 10 mL per wash, followed by the elution
buffer (FP buffer) with increasing concentration of GSH. With each and every elution 10
min of incubation was given to the samples at the room temperature with occasional
shaking. Fractions were collected into different tubes and presence of p300-CH1-GST
(37 KDa protein) was verified by SDS-PAGE gel. Pooled fractions were treated with
protease inhibitor cocktail (Thermoscientific) and dialyzed twice against a buffer
containing 10 mM Tris, 30 mM NaCl
,
10% glycerol, 1mM DTT (Fisher), and 100 μM
ZnCl
2
at pH 8.0 to ensure proper folding.
After complete dialysis, protein was recovered from the cassette and protein
concentration was determined by bradford assay. In brief, bovine serum albumin (BSA)
standard of varying concentration were made and 200 µl of Bradford reagent was added
to each cuvette. Absorbance of each standard and protein was measured at 595 nm by
UV-Vis spectrophotometer. Dialyzed protein was then stored at -80 °C for further
experiments.

211
6.2.4. Fluorescence polarization (FP) saturation binding assay for the determination of
K
d
(affinity constant) between HIF1a-flu CTAD and p300-CH1-GST
Fluorescein-labeled 41-mer (aa 786–826) HIF-1α-CTAD was dissolved in FP assay
buffer (50 mM Tris, 150 mM NaCl, 10% glycerol, 1 mM DTT, 0.1% NP-40, 100 μM
ZnCl
2,
pH 8.0) diluted to a concentration of 30 nM. Saturation binding curve was
obtained by dissolving increasing concentrations of p300-CH1-GST ranging from 0 to
4000 nM in FP buffer with 2 % of pluronic acid. P300-CH1 was expressed as GST
fusion and purified as discussed in previous experiments. In an opaque 96 well black
plate, 60 µl of each and every concentration of p300-CH1-GST was added. To these
wells, 60 µl of 30 nM solution of HIF-1α-flu CTAD was added so that the final
concentration of p300-CH1-GST ranges from 0 to 2000 nM and 15 nM of HIF-1α-flu
CTAD. Every sample was done in triplicate. Plate was incubated for 1h at room
temperature in dark and polarization of the fluorescent probe was recorded by a BioTek
synergy 2 microplate reader with excitation and emission wavelengths of 485 and 525
nm, respectively. The IC
50
value derived from this binding curve was used to obtain the
affinity constant (K
d
) for the fluorescein-labeled HIF-1α CTAD and p300-CH1-GST
complex for the competition assays. Data was plotted by SigmaPlot 12.0 using sigmoidal
equation, parameter 5.
6.2.5 Fluorescence polarization (FP) competition binding assay for the determination
of binding affinity of mono-ETPs and bis-ETPs towards the p300-CH1-GST and HIF-
1α-flu CTAD complex
As determined from the saturation binding assay, a solution of 150 nM of p300-CH1-
GST and 30 nM of HIF-1α-flu CTAD was incubated in FP assay buffer (50 mM Tris, 150
mM NaCl, 10% glycerol, 1 mM DTT, 0.1% NP-40, 100 μM ZnCl
2,
2 % pluronic acid, pH
8.0 ) for 1h at room temperature. After 1h, 60 µl of different concentrations of mono-
212
ETPs, mono-ETP-4 ranging from 0 to 600 µM, mono-ETP-5 ranging from 0 to 200 µM
and bis-ETPs (bis-ETP-1, bis-ETP-2 and Chetomin) ranging from 0 to 16 µM were added
to an opaque 96 well black plate. To these wells, 60 µl of the pre-incubated p300-CH1-
GST and HIF-1α-flu CTAD complex was added so that the final concentration of p300-
CH1-GST is 75 nM (corresponding to 60 % of saturation binding), HIF-1α-flu CTAD is
15 nM , mono-ETPs concentration is from 0 to 100 µM and bis-ETPs is from 0 to 8 µM.
For all the samples, final concentration of DMSO was maintained at 1%. Amount of
dissociated fluorescent probe was determined by the BioTek Synergy 2 microplate reader
with excitation and emission wavelengths of 485 and 525 nm, respectively. The Kd
2
values were determined for each compound by fitting the averages of three individual
measurements to a sigmoidal dose response curve using nonlinear regression model with
SigmaPlot 12.0 software.
6.2.6 Effect of mono-ETP on the structure of p300-CH1-GST by circular dichroism
(CD) spectroscopy
CD spectra was recorded on Olis DSM20 double beam spectrophotometer (The
University of Arizona) equipped with temperature controller using 1 mm length cells and
a scan speed of 40 nm/sec. Baseline calibrations were generated from the wavelength
scans obtained with FP assay buffer (50 mM Tris, 150 mM NaCl, 10% glycerol, 1 mM
DTT, 0.1% NP-40, 100 μM ZnCl
2
). Wavelength scans were obtained at 1 nm increment
at 25 °C with an integration time of 2 s. mono-ETP-5 stock was prepared in methanol
and diluted to final concentration of 1 µM and 10 µM in FP assay buffer. First, the CD
spectrum of expressed, purified and natively folded p300-CH1-GST fusion protein was
recorded and then the fusion protein at 1 µM concentration was exposed to increasing
213
concentrations of mono-ETP-5. Data obtained was plotted by SigmaPlot 12.0 using
smoothing technique with bisquare weighting and polynomial regression.

6.2.7 Determination of cell viability by cell titer blue cytotoxicity assay
Stably transfected MDA-MB-231-HRE-Luc breast cancer cells were maintained in
Dulbecco’s Modified Eagle Media (DMEM) supplemented with 10 % Fetal Bovine
Serum (FBS) and 0.4 g/L G418 antibiotic. Cells were plated in an opaque 96 well plate
(Greiner) at a density of 10,000 cells/ well (50,000 cells/mL) and allowed to form
monolayer before dosing with the compounds. After attachment the media was replaced
by 100 µl of fresh media containing increasing concentration of mono-ETP-4, mono-
ETP-5 and chetomin ranging from 0 to 2000 nM. All samples were maintained at the
final percentage of 0.1 to 0.3 % of DMSO. Vehicle sample were treated with complete
cell culture media containing 0.1 % DMSO. Cells were treated with compounds for total
of 24 h. After 24 h of incubation 20 µL of Cell Titer Blue (CTB) reagent from promega
was added into each and every well and cells were incubated at 37 °C and 5% CO
2
for 2
h before taking the reading. Every sample was done in quadruplets. CTB assay is based
on the ability of live cells to convert a non-fluorescent redox dye resazurin into
fluorescent dye resorufin. On the other hand, non-viable cells lose their metabolic
activity and will not produce fluorescent signal. Fluorescent signal was measured by
BioTek Synergy 2 microplate reader with excitation and emission wavelength of 560 nm
and 590 nm, respectively. GI
50
curves were obtained by plotting data using SigmaPlot
10.0, parameter 3.

214
6.2.8 Luciferase reporter assays
Stably transfected MDA-MB-231-HRE-Luc cells were plated in 24-well dishes (BD-
Falcon) at seeding density of 65,000 cells/well in 2 mL of DMEM media, supplemented
with 10 % FBS and 0.4 g/L G418. Cells were allowed to adhere and allowed to form
monolayer before dosing with the compounds (approx. 70% confluent). After attachment,
cells were treated with 1 mL of fresh media containing mono-ETP-4 at 200 and 600 nM
concentration and mono-ETP-5 at 50, 200 and 600 nM concentration. For bis-ETP
experiment, cells were treated with bis-ETP-1, bis-ETP-2 and control bis-ETP compound
in a concentration range of 50 nM to 600 nM. All samples contained a final concentration
of 0.1% DMSO, whereas the vehicle samples were treated with cell culture media having
0.1% DMSO. Cells were incubated for 6 h at 37 °C and 5% CO
2
and then hypoxia was
induced with 300 µM of DFO or using anaerobe pouch system (BD GasPak EZ pouch),
and cells were incubated for another 18 h. DFO solution was made at a stock
concentration of 30 mM in autoclaved DI water and then 10 µl of stock was spiked into 1
mL of media into each well. The whole cell lysates were isolated by washing the cells
twice with ice-cold PBS and then adding 150 µl of cell culture lysis reagent (Promega).
Prior to collecting cell lysate, halt protease inhibitor cocktail (Thermoscientific) was
added to the cell culture lysis reagent in order to ensure the stability of proteins. Cell
lysate was collected in 1.5 mL low adhesion pre-chilled eppendorf tubes (UA Scientific)
and centrifuged at 13,000 rpm for 5 min at 4 °C. Supernatant was collected in another set
of pre-chilled eppendorf tubes and pellet was discarded. Relative light intensity was
measured by reacting 20 µL of cell lysate with 100 µL of luciferase assay
reagent(Promega) using a Turner TD-20e luminometer and the results were normalized to
total protein content determined by BCA assay. Briefly, 10 µL of cell lysate was added to
215
200 µl of BCA reagent. Absorbance was measured at 562 nm using a BioTek Synergy 2
microplate reader and normalized to protein standards (125 µg/mL to 2000 µg/mL of
BSA solution).

6.2.9 Analysis of secreted VEGF by Enzyme-linked immunosorbent assay (ELISA)
Parent MDA-MB-231 and MCF7 breast cancer cells were grown in DMEM media and
RPMI media, respectively. Both cells were maintained in 10% FBS and 0.5% Pen-Strep.
For ELISA, assay cells were plated in 24-well plates (BD-Falcon) at 50% confluency in 2
mL of respective media. Cells were allowed to adhere for overnight (approx. 70%
confluent) before dosing with the compound. After approx. 24 h, old media was replaced
by fresh media containing mono-ETP-4, mono-ETP-5 at 400 nM concentration and
chetomin at 200 nM concentration. In order to avoid any interference because of high
percentage of serum, fresh media with compounds was supplemented with reduced serum
media (2% FBS). All samples contained a final concentration of 0.1% DMSO; vehicle
samples were treated with cell culture media containing 0.1% DMSO. Cells were
incubated for 6 h at 37 °C and 5% CO
2
and then hypoxia was induced in half of the wells
with 300 µM of DFO, and cells were incubated for another 18 h. After 24 h of
incubation supernatant was collected from each well into their respective labeled tubes
and stored at -20 °C. Secreted levels of VEGF in the supernatant were measured with the
Quantikine Human VEGF kit (R&D Systems) according to the manufacturer’s protocol.
In brief, VEGF standard in varying concentration (concentration range from15.6 pg/mL
to 1000 pg/mL) was made from a stock of 2000 pg/mL. To each well, 50 µL of assay
diluent and 200 µl of standard, control and sample was added and incubated for 2h at 37
°C and 5% CO
2
. After 2h, all wells were washed with the wash buffer (3×) and 200 µl of
216
VEGF conjugate and incubated for further 2h followed by washing and aspiration (3×).
After last aspiration, substrate solution was added and incubated for 20 min followed by
addition of stop solution. Absorbance was measured at 450 nm by a BioTek Synergy 2
microplate reader. Readings were normalized by total protein concentration.

6.2.10 Luciferase reporter assay for screening library of small molecules
Stably transfected MDA-MB-231-HRE-Luc cells were plated in 96-well plates (Greiner
BioOne) at seeding density of 10,000 cells/well in 200 µl of DMEM media,
supplemented with 10 % FBS and 0.4 g/L G418. Cells were allowed to adhere and
allowed to form monolayer before dosing with the compounds (approx. 70% confluent).
After attachment, cells were treated with 100 µl of fresh media containing small
molecules from library at 8 µM concentration. All samples contained a final
concentration of 0.1% DMSO; vehicle samples were treated with cell culture media
containing 0.1% DMSO. Cells were incubated for 6 h at 37 °C and 5% CO
2
and then
hypoxia was induced by placing half of the plates in anaerobe pouch system (BD GasPak
EZ pouch), and cells were incubated for another 18hThe whole cell lysates were isolated
by washing the cells twice with ice-cold PBS (2×) and then adding 30 µl of cell culture
lysis reagent (Promega). Prior to collecting cell lysate, halt protease inhibitor cocktail
(Thermo Scientific) was added to the cell culture lysis reagent in order to ensure the
stability of proteins. Cell lysate was centrifuged at 13,000 rpm for 5 min at 4 °C.
Relative light intensity was measured by reacting 50 µl of luciferase assay reagent
(Promega) to 10 µl of cell lysate using a Turner TD-20e luminometer and the results were
normalized to total protein content determined by BCA assay.

217
6.3 Experimental Section of Chapter 3
6.3.1 Fluorescence polarization (FP) saturation binding assay for the determination of
K
d
(affinity constant) between HIF1a-flu CTAD and p300-CH1
The fusion protein p300-CH1-GST was expressed and purified as mentioned in the
previous experiments (6.2.1, 6.2.2). A standard protocol of thrombin cleavage was used
to cleave GST tag from the fusion protein (p300-CH1-GST). SDS-PAGE was used to
check the completion of cleavage reaction. Saturation binding curve was obtained by
dissolving increasing concentrations of p300-CH1 ranging from 0 to 12000 nM in FP
buffer with 2 % of pluronic acid. Fluorescein-labeled 41 mer (aa 786–826) HIF-1α-
CTAD was dissolved in FP assay buffer and diluted to a concentration of 30 nM. In an
opaque 96 well black plate, 60 µl of different concentrations of p300-CH1 was added.
To these wells, 60 µl of 30 nM solution of HIF-1α-flu CTAD was added so that the final
concentration of p300-CH1 ranges from 0 to 6000 nM and 15 nM of HIF-1α-flu CTAD.
Every sample was done in triplicate. Plate was incubated for 1h at room temperature in
dark and polarization of the fluorescent probe was recorded by a BioTek synergy 2
microplate reader with excitation and emission wavelengths of 485 and 525 nm,
respectively. The IC
50
value derived from this binding curve was used to obtain the
affinity constant (K
d
) for the fluorescein-labeled HIF-1α CTAD and p300-CH1-GST
complex for the competition assays. Data was plotted by SigmaPlot 12.0 using sigmoidal
equation, parameter 4.
6.3.2 Fluorescence polarization (FP) competition binding assay for the determination
of binding affinity of HBS helices towards the p300-CH1 (with and without GST tag)
and HIF-1α-flu CTAD complex
As determined from the saturation binding assay, a solution of 150 nM of p300-CH1-
GST and 30 nM of HIF-1α-flu CTAD or 800 nM of p300-CH1 and 30 nM of HIF-1α-flu
218
CTAD was incubated in FP assay buffer for 1h at room temperature. After 1h, 60 µl of
different concentrations of HBS helices (HBS-5, HBS-6, HBS-8 and peptide-7) ranging
from 0 to 200 µM were added to a 96 well black opaque plate. To these wells, 60 µl of
the pre-incubated p300-CH1-GST and HIF-1α-flu CTAD complex or p300-CH1 and
HIF-1α-flu CTAD complex was added. Final concentration of p300-CH1-GST is 75 nM
(corresponding to 60 % of saturation binding), p300-CH1 is 400 nM, HIF-1α-flu CTAD
is 15 nM and HBS helices concentration is from 0 to 100 µM. For all the samples, final
concentration of DMSO was maintained at 1%. Amount of dissociated fluorescent probe
was determined by the BioTek Synergy 2 microplate reader with excitation and emission
wavelengths of 485 and 525 nm, respectively. The Kd
2
values were determined for each
compound by fitting the averages of three individual measurements to a sigmoidal dose
response curve using nonlinear regression model with SigmaPlot 12.0 software.

6.3.3 Purification of fusion protein p300-CH1-GST at higher concentration
Fusion protein (p300-CH1-GST) was expressed as discussed previously in experiment
6.2.1. In order to get protein at higher concentration for direct binding assay, p300-CH1-
GST was purified with 10 mM reduced glutathione. First, supernatant-GSH bead mixture
was passed through the column and flow through was collected. Then beads were
washed three times with ice cold PBS, 10 mL per wash, followed by the elution buffer at
10 mM GSH. With each and every elution, 10 min of incubation was given to the
samples at the room temperature with occasional shaking. Fractions were collected into
different tubes and assayed by SDS-PAGE. Pooled fractions were treated with protease
inhibitor cocktail (Thermoscientific) and dialyzed twice against a buffer containing 10
mM Tris, 30 mM NaCl
,
10% glycerol, 1mM DTT (Fisher), and 100 μM ZnCl
2
at pH 8.0
219
to ensure proper folding. After complete dialysis, protein was recovered from the cassette
and protein concentration was determined by bradford assay. Absorbance of each
standard and protein was measured at 595 nm by UV-Vis spectrophotometer. Dialyzed
protein was then stored at -80 °C for further experiments.

6.3.4 Fluorescence polarization assay for the direct binding of HBS helix to p300-
CH1-GST
The fusion protein p300-CH1-GST was expressed and purified as mentioned in the
previous experiments (6.2.1, 6.3.3). Direct binding of HBS-9-Flu towards p300-CH1-
GST was measured by dissolving increasing concentrations of p300-CH1-GST ranging
from 0 to 200 µM in FP buffer with 1% of pluronic acid. Fluorescein-labeled HBS-9-flu
(αB-helix) was dissolved in FP assay buffer and diluted to a concentration of 36 nM. In
an opaque 96 well black plate, 115 µl of different concentrations of p300-CH1-GST was
added. To these wells, 5 µl of 36 nM solution of HBS-9-flu was added and incubated for
1h at room temoerature in dark. Each sample was done in triplicate. Polarization of the
fluorescent probe was recorded by a BioTek synergy 2 microplate reader with excitation
and emission wavelengths of 485 and 525 nm, respectively. The affinity constant
(K
d
) value was derived by plotting data using SigmaPlot 12.0 and fitted in sigmoidal
equation, parameter 3.

6.3.5 Circular Dichroism Spectroscopy
CD spectra was recorded on AVIV 202SF CD spectrometer using 1 mm length cell and a
scan speed of 5 nm/min. Baseline calibrations were generated from the wavelength scans
obtained with assay buffer. HBS αA-helix (HBS-1, HBS-2, HBS-3 and peptide-4) and
220
HBS αB-helix (HBS-5, HBS-6, HBS-8 and peptide-7) samples were dissolved in 0.1×
phosphate buffered saline (13.7 mM NaCl, 1 mM phosphate, 0.27 mM KCl, pH 6.3),
containing 10% trifluoroethanol with final concentration of peptide between 50 µM to
100 µM. Wavelength scans were obtained at 1 nm increment at 25 °C with an integration
time of 2 s. The helix content of each peptide was determined from the mean residue CD
at 222 nm, [θ] 222 (deg cm2 dmol-1) corrected for the number of amino acids. Percent
helicity was calculated from the ratio [θ] 222/[θ]max, where [θ]max = (–44000 + 250T)(1
–k/n), with k = 4.0 and n = number of residues. Data obtained was plotted by SigmaPlot
12.0 using smoothing technique with bisquare weighting and polynomial regression.

6.3.6 Determination of binding affinity by Isothermal titration Calorimeter (ITC)
Isothermal titration calorimeter (ITC) experiment was performed on VP-ITC titration
calorimeter from MicroCal Inc. at 25 °C. HBS αA-helices (HBS-1, HBS-2, HBS-3) and
peptide-4 was titrated against p300-CH1-GST. All HBS helices, peptides and protein
were dialyzed in the ITC buffer (50 mM Tris (pH 7.0), 50 mM NaCl) followed by
filtration and degassing to minimize background. The concentration of p300-CH1-GST
in the ITC cell was 10-20 μM and that of HBS helices and peptide in the syringe was 10-
fold over that of protein (100-200 μM). Typically, first injection of 10 μL was followed
by 20 injections of the same volume until a molar ratio of 2.5 was obtained. Integration
of the baseline-substracted thermograms yielded binding isotherms that were fit to a
model of one-site interaction. ITC data analysis was performed with Origin 5.0 software
(MicroCal Inc.). The first and two last data points were excluded from analysis. A non-
linear least squares fit of the data was used to determine the equilibrium association
221
constants Kd. Only Kd values were reported (together with S.D. from triplicate
experiments), since H values had higher uncertainties and were therefore not interpreted.
The stoichiometric ratios obtained from the curves fit consistently to 1:1 stoichiometry
within 5-8% error.

6.3.7 Cell density and population doubling assay
Hela cells were plated in 24-well dishes (BD-Falcon) at seeding density of 35,000
cells/well in 2 mLof DMEM media, supplemented with 10 % FBS and 0.5 % pen-strep.
Cells were allowed to adhere and allowed to form monolayer before dosing with the
compounds and control (approx. 70% confluent). After attachment, cells were treated
with 1 mL of fresh media containing HBS-2, peptide-4 at 1 µM concentrations and
chetomin at 200 nM concentration. All samples were maintained at a final concentration
of 0.1% DMSO except control, which has only fresh media.Cells were harvested at time
points 12, 24, 36, 48, 72 h by trypsinization using 0.5% trypsin in EDTA (Gibco). Cell
suspension was then centrifuged at 1200 rpm for 5 min at 4 °C and resuspended into 300
µl of PBS prior to trypan blue staining. 50 µl of aliquots were then mixed with trypan
blue in 1:1 ratio and incubated for 2 min at room temperature. Cells were then counted by
haemocytometer (Beckman). All experiments were performed in triplicates.
6.3.8 Cell viability assay
HeLa cells were maintained in Dulbecco’s Modified Eagle Media (DMEM)
supplemented with 10 % Fetal Bovine Serum (FBS) and 0.5% Pen-strep. First, a cell
count study was performed to determine the appropriate number of cells that need to be
plated for the cell viability assay. Serial dilutions of cells from 2500 cells/mL to 100,000
222
cells/mL were made in cell culture media. The number of cells used in the viability assay
lies within the linear portion of the plot and has absorbance value between 0.75- 1.25. As
determined from the cell count study, 6,000 cells/well were plated in a 96-well plate and
allowed to form monolayer before dosing with the compounds. After attachment, media
was replaced by 100 µl of fresh media containing increasing concentration of HBS-5
ranging from 0 to 100 µM and chetomin ranging from 0-2000 nM. All samples were
maintained at the final percentage of 0.1 % of DMSO. Vehicle sample was treated with
complete cell culture media containing 0.1 % DMSO. Cells were treated with
compounds for total of 24 h. After 24 h of incubation, 11 µl of MTT (3-(4, 5-
dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide) reagent, stock concentration 5
mg/mL in PBS was added to each well and incubated at 37 °C and 5 % CO
2
for 3h.
Every sample was done in quadruplets. MTT assay measures the metabolic activity of
cells by reducing the yellow tetrazolium compound to purple formazan crystal by the
action of dehydrogenase enzymes. Non-viable cells will lose their metabolic activity and
will not be able to form purple formazan crystals. After 3h incubation, media with MTT
was removed and purple crystals were dissolved in 100 µl of dimethyl sulfoxide
(DMSO). Absorbance was measured at 570 nm to quantify the amount of formazan
crystals with a correction at 690 nm. GI
50
curves were obtained by plotting data using
SigmaPlot 10.0, parameter 4.

6.3.9 Plating and dosing for isolation of mRNA αA- Helices
HeLa cells were plated in six-well dishes (BD-Falcon) at the seeding density of 150,000
cells/well in 2 mL of DMEM media, supplemented with 10 % FBS and 0.5 % pen-strep.
Cells were allowed to adhere and form monolayer before dosing with the compounds
223
(approx. 70% confluent). After attachment, cells were treated with 1.5 mL of fresh media
containing HBS-2 (αA-helix), peptide-4 at 1 µM, chetomin at 200 nM concentrations. All
samples contained a final concentration of 0.1% DMSO; vehicle samples were treated
with cell culture media containing 0.1% DMSO. Cells were incubated for 6 h at 37 °C
and 5% CO
2
and then hypoxia was induced with 300 µM of DFO and cells were
incubated for another 12h or 18h. Every experiment was performed in quadruplets.
αB-Helices
a) HeLa cells
HeLa cells were plated in six-well dishes (BD-Falcon) at the seeding density of 300,000
cells/well in 2 mL of DMEM media, supplemented with 10 % FBS and 0.5 % pen-strep.
After attachment, cells were treated with 1.5 mL of fresh media containing HBS-5, HBS-
6, HBS-8 and peptide-7 in a concentration range of 1 µM to 50 µM. All samples
contained a final concentration of 0.1% DMSO; vehicle samples were treated with cell
culture media containing 0.1% DMSO. Cells were incubated for 6 h at 37 °C and 5%
CO
2
before hypoxia was induced with 300 µM of DFO and cells were further incubated
for another 18h.
For another experiment HeLa cells were maintained in 2% FBS and while dosing
was replaced by 0.2 % FBS. Prior to inducing hypoxia, either with 300 µM of DFO or by
anaerobe pouch system (BD GasPak EZ pouch), cells were incubated for 6 h at 37 °C and
5% CO
2
and further incubated for another 42h. Every experiment was performed in
quadruplets.

224
b) A549 cells
A549 cells were plated in six-well dishes (BD-Falcon) at the seeding density of 250,000
cells/well in 2 mL of F-12K media, supplemented with 10 % FBS and 0.5 % pen-strep.
After attachment, cells were treated with 1.5 mL of fresh media containing HBS-5, HBS-
6, HBS-8 and peptide-7 in a concentration range of 1 µM to 50 µM. All samples
contained a final concentration of 0.1% DMSO; vehicle samples were treated with cell
culture media containing 0.1% DMSO. Cells were incubated for 6 h at 37 °C and 5%
CO
2
and then hypoxia was induced with 300 µM of DFO and cells were incubated for
another 18h.

6.3.10 Isolation of mRNA
Once the incubation was done, cells were washed twice with ice cold PBS and lysed by
RLT buffer having 1 % of β-mercaptoethanol. Total RNA was isolated with RNeasy kit
(Qiagen) according to manufacturer’s instructions and quantified by UV
spectrophotometer. Further, RNA was treated with DNaseI (Ambion) to remove any
genomic DNA present. Reverse transcription was performed by using Superscript III
reverse transcriptase (Invitrogen) as per manufacturer’s instructions. Each experiment
was performed in triplicates.

6.3.11 Analysis of gene expression
Real-time qRT-PCR was used to determine the effect of HBS helices (HBS-2,pepride-4,
HBS-5, HBS-6, HBS-8 and peptide-7) on the level of expression of VEGF, GLUT1,
LOX,CXCR4 and c-Met genes in HeLa and A549 cells under both normoxic and hypoxic
conditions. For VEGF analysis, the forward primer 5’-AGG CCA GCA CAT AGG AGA
225
GA-3’ and reverse primer 5’-TTT CCC TTT CCT CGA ACT GA-3’ were used to
amplify a 104-bp fragment from the 3’-translated region of the gene. For GLUT1 gene
forward primer is 5'-AGT ATG TGG AGC AAC TGT GTG G-3' and reverse primer is
5'-CGG CCT TTA GTC TCA GGA AC-3' to yield a product of 106 bp. For LOX, the
following primer pair was used, forward: 5'-ATG AGT TTA GCC ACT ATG ACC TGC
TT-3' and reverse: 5'-AAA CTT GCT TTG TGG CCT TCA- 3' to amplify a 73 bp
product. For CXCR4, following primer pair was used: forward 5'- GAA GCT GTT GGC
TGA AAA GG-3', reverse 5'-CTC ACT GAC GTT GGC AAA GA-3' to yield a product
of 94 bp. For c-Met gene we have used the following pair, forward: 5'-GGA AGA GGG
CAT TTT GGT TG-3', reverse: 5'-TTG GGA AAC TTC TCC TAT GTC A-3' to yield a
product of 117 bp. mRNA levels were normalized by expression levels of a
housekeeping gene, β-glucuronidase.For β-glucuronidase the following primer was
designed and used, forward: 5’-CTC ATT TGG AAT TTT GCC GAT T-3’ and reverse:
5’- CCG AGT GAA GAT CCC CTT TTT A-3’. Reaction was performed by Fast
SYBR(R) Green Master Mix (Applied Biosystems). Temperature cycling and detection
of the SYBR green emission were performed with an ABI 7900HT Fast Real-Time PCR
instrument. Data were analyzed with Applied Biosystems Sequence Detection System,
version 2.4.

6.3.12 Analysis of secreted VEGF by Enzyme-linked immunosorbent assay (ELISA)
HeLa cells were grown in DMEM media in 10% FBS and 0.5% Pen-Strep. For ELISA,
assay cells were plated in 24-well dishes (BD-Falcon) at a seeding density of 70000 cells/
well. Cells were allowed to adhere for overnight (approx. 70% confluent) before dosing
with the compound. After approx. 24 h, old media was replaced by fresh media with 2%
226
FBS, containing HBS-5 in a concentration range from 1 µM to 20 µM and chetomin at
130 nM concentration. All samples contained a final concentration of 0.1% DMSO;
vehicle samples were treated with cell culture media containing 0.1% DMSO. Cells were
incubated for 6 h at 37 °C and 5% CO
2
and then hypoxia was induced in half of the wells
with 300 µM of DFO, and cells were incubated for another 18 h. After 24 h of
incubation supernatant was collected from each well into their respective labeled tubes
and stored at -20 °C. Secreted levels of VEGF in the supernatant were measured with the
Quantikine Human VEGF kit (R&D Systems) as discussed in the previous experiment
6.2.8. Every experiment is performed in quadruplets.

6.3.13 Analysis of HIF-1α by Western blotting
HeLa cells were plated in four of 75cm2 culture flask at a confluency of 70%. Once cells
were attached, they were treated with vehicle, HBS-5 (10 µM) in cell culture complete
media enriched with 10% FBS. All samples contained a final concentration of 0.1%
DMSO .Cells were incubated for 6 h at 37 °C and 5% CO
2
and then hypoxia was induced
in half of the flasks with 300 µM of DFO. After 24 h incubation, cells were lysed and
cytoplasmic and nuclear proteins were collected by using NE-PER kit (Pierce) according
to manufacturer’s protocol. To ensure equal loading, protein concentration was
determined by BCA assay (ThermoScientific). A 1.5 mm 10 % SDS-PAGE gel was
casted and 30 µg of protein from each sample was loaded into separate wells. Proteins
were then electroblotted onto the PVDF (BioRad) membrane from the SDS-PAGE gel.
After transfer, membrane was washed with 1× TBST buffer (2×) and then incubated with
5 % milk for 1h to avoid non-specific binding. Membrane was then probed for HIF-1α by
a monoclonal mouse anti-human HIF-1α antibody (BD Transduction Laboratories).
227
Membrane was further washed with 1× TBST for 5 min (4×). After washing, the
membrane was incubated with horseradish peroxidase (HRP)-conjugated secondary
antibody (Santa Cruz Biotechnology). Signals were detected by using SuperSignal
chemiluminescent kit (Pierce/Thermo Scientific)

6.3.14 Luciferase assay with hypoxia induction using anaerobic pouch
Stably transfected MDA-MB-231-HRE-Luc cells were plated in 24-well plates (BD
Falcon) at seeding density of 70,000 cells/well in 2 mL of DMEM media, supplemented
with 10 % FBS and 0.4 g/L G418. Cells were allowed to adhere and allowed to form
monolayer before dosing with the compounds (approx. 70% confluent). After
attachment, cells were treated with 1 mL of fresh media containing HBS-5, HBS-6,
peptide-7 and chetomin. All samples contained a final concentration of 0.1% DMSO;
vehicle samples were treated with cell culture media containing 0.1% DMSO. Cells were
incubated for 6 h at 37 °C and 5% CO
2
and then hypoxia was induced by placing one
plate in anaerobe pouch system (BD GasPak EZ pouch), and cells were incubated for
another 18h. Further steps were performed as discussed in experiment 6.2.7.

228

6.4 Experimental Section of Chapter 4
Experimental details of integrin ligand, ABL-1, is attached in the appendix.
6.4.1 Cell viability assay for integrin compounds
a) MTT assay
MCF7 cells and WM115 cells were maintained in RPMI-1640 (Gibco) and
minimum essential Media- alpha (MEM-alpha from Gibco and cellgro), respectively.
Both media were supplemented with 10 % Fetal Bovine Serum (FBS) and 0.5% Pen-
strep. 10,000 cells/well were plated in a 96-well plate and allowed to form monolayer
before dosing with the compounds. After attachment, media was replaced by 100 µl of
fresh media containing increasing concentration of ABL-1 ranging from 0 to 100 µM.
Cells were treated with compounds for total of 24 h. After 24 h of incubation, 11 µl of
MTT (3-(4, 5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide) reagent, stock
concentration 5 mg/mL in PBS was added to each well and incubated at 37 °C and 5 %
CO
2
for 3h. Every sample was done in quadruplets. All further steps were performed as
discussed in experiment 6.3.8.
b) CTB assay
10,000 cells/well (MCF7 and WM115) were plated in a 96-well plate and
allowed to form monolayer before dosing with the compounds. After attachment,
media was replaced by 100 µl of fresh media containing increasing concentration of
ABL-1 ranging from 0 to 100 µM. Cells were treated with compounds for total of 24
229
h. After 24 h of incubation, 20 µl of CTB reagent was added to each well and further
steps were performed as discussed in experiment 6.2.6.
6.4.2 Absorption spectra measurement
Absorption spectra were recorded on a DU800 spectrophotometer (Beckmann). ABL-1
and ABL-2 were dissolved in the solvent mixture having 95% water and 5% methanol at
a concentration of 100 µM, 50 µM and 50 µM, respectively. Absorption spectra for all
the samples were recorded in the wavelength range of 200-800 nm. Baseline calibration
was done by running a background scan using a blank sample. Blank sample was
prepared in the same solvent which was used to prepare the samples.
CIL-1 was also dissolved in the above said solvent mixture at a concentration of
100 µM and absorbance was measured at each wavelength between 200-800 nm.
Acquired scan data was used to determine the λ
max
(wavelength of maximum
absorbance).

6.4.3 Fluorescence excitation and emission measurement
a) Fluorescence excitation and emission measurement in aqueous and glycerol:water
system
ABL-1 and ABL-2 were dissolved in aqueous solution of 95 % water and 5 % methanol
at a concentration of 100 µM. An excitation spectrum was recorded in the wavelength
range of 375 nm to 520 nm. The emission was fixed at 540 nm and excitation was
recorded at each wavelength from 375 to 520 nm. For emission spectrum measurement,
the excitation wavelength of 435 nm was used and the fluorescence intensity was
230
measured in the wavelength range of 450 nm to 550 nm at right angle to the excitation
beam. Measurement was recorded at each wavelength from 450 to 550 nm.
For glycerol:water system, ABL-1 and ABL-2 were dissolved in 90% (v/v)
glycerol:water solvent mixture and fluorescence excitation and emission spectra were
recorded in the same way as for the samples in aqueous mixture.
b) Fluorescence excitation and emission measurement in PBS and with cells
ABL-1 and ABL-2 were dissolved in aqueous solution of PBS at a concentration of 100
µM. Excitation spectra were recorded from 375 nm to 520 nm. The emission was fixed
at 540 nm and excitation was recorded at each wavelength from 375 to 520 nm. For
emission spectrum measurement, the excitation wavelength of 435 nm was used and the
fluorescence intensity was measured in the wavelength range of 450 nm to 550 nm at
right angle to the excitation beam. Measurement was recorded at each wavelength from
450 to 550 nm.
WM115 cells at a density of 2 × 10
6
cells were incubated with ABL-1 and ABL-2
for 2 hours at 37 °C and 5% CO
2
. Cells were then washed with PBS (2×) and dissolved
at a concentration of 1 × 10
6
cells/mL in PBS. Fluorescence excitation and emission
spectra was recorded as discussed above. All the measurements were performed by using
a photomultiplier detection system – Model 814 from photon technology international.

6.4.4 Confocal microscopy of integrin compounds
WM115 and MCF7 cells were maintained in their respective media supplemented with
10% FBS and 0.5% pen-strep. For confocal microscopy, Mat Tek 35 mm diameter dishes
231
with 14 mm glass bottom were used. 30,000 cells/dish (MCF7 or WM115) with 300 μL
of media were plated and incubated at 37 °C and 5% CO
2
for overnight. Once the cells
were attached, the dishes with cells were treated with compounds ABL-1, ABL-2, CIL-1
and CD-1 at desired concentrations and incubated for desired time at 37 °C in 5% CO2.
For 4 °C experiments, all the compounds were incubated for 2 hours. Confocal imaging
was then performed on 510 Ziess confocal microscope at excitation and emission
wavelength of 488 nm and 510 nm, respectively.
6.4.5 Flow cytometry of ABL-1, CIL-1 and CD-1
WM115 cells were maintained in Modified Eagle Media alpha (MEM-α) supplemented
with 10 % Fetal Bovine Serum (FBS) and 0.5% Pen-strep. 500,000 cells were plated in
six-well dishes and allowed to establish a monolayer for overnight. Next, cells were
treated with ABL-1, CIL-1 and CD-1 at desired concentrations and incubated for 2h at 37
°C and 5 % CO2. After 2h of incubation, cells were washed with PBS and dissolved in
1× binding buffer at a concentration of 1 ×106 cells/mL. 100 µl of cell suspension was
transferred to a 5 mL culture tube followed by addition of 5 µl of PE Annexin V and 5 µl
of 7-AAD. Cell culture tube was incubated for 15 min at room temperature in the dark
and then 400 µl of 1× binding buffer with desired concentration of compounds was added
to the tubes. Samples were then analyzed by BD LSR II Flow Cytometer.

6.4.6 Adhesion assay
WM115 cells were maintained in Modified Eagle Media alpha (MEM-α) supplemented
with 10 % Fetal Bovine Serum (FBS) and 0.5% Pen-strep. As determined from the cell
count study, 60,000 cells/well along with increasing concentration of ABL-1 and ABL-3
were plated in a 96-well, vitronectin and fibronectin coated, plates from R & D and
232
incubated for 2h at 37 °C and 5 % CO
2.
After 2h of incubation, plates were washed 3
times with PBS and 11 µl of MTT (3-(4, 5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium
bromide) reagent, stock concentration 5 mg/mL in PBS was added to each well and
further incubated at 37 °C and 5 % CO
2
for 3h. Every sample was done in quadruplets.
After 3h incubation, media with MTT was removed and purple crystals were dissolved in
100 µl of dimethyl sulfoxide (DMSO). Absorbance was measured at 570 nm to quantify
the amount of formazan crystals with a correction at 690 nm. GI
50
curves were obtained
by plotting data using SigmaPlot 12.0, parameter 3.

6.4.7 Intravital microscopy with ABL-1
Intravital microscopy experiments were conducted by our collaborators at PRISM, San
Diego.. For subcutaneous model, breast carcinoma cells (N2O2) were implanted in the
subcutaneous region. ABL-1 was dissolved in saline and 1% DMSO at a concentration of
1 mg/kg and injected intravenously via the tail vein. A window was implanted in the
dorsal side of the animal through which intravital microscopy was performed and
distribution of the compound was observed in the tumor tissue. The fluorescence images
were taken at excitation and emission wavelength of the cyan 40 dye of the conjugate.

233
6.5 Experimental Section of Chapter 5
6.5.1 MTS assay with MHI-Clg and MHI-148
LNCaP cells were maintained in RPMI media with 10% FBS and 0.5% Pen-strep. C4-2B
and PC-3 cells were maintained in T-media and were supplemented with 10 % Fetal
Bovine Serum (FBS) and 0.5% Pen-strep. Desired numbers of cells (optimized by cell
cont study) were plated in a 96-well plate and were allowed to form a monolayer before
dosing with the compound. After attachment, media was replaced by 100 µl of fresh
media containing increasing concentration of MHI-Clg and MHI-148 ranging from 0 to
12 µM. Cells were treated with compounds for total of 96 h. After 96 h of incubation,
20 µl of MTS/PMS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
sulfophenyl)-2H-tetrazolium, inner salt; MTS/(phenazine methosulfate; PMS ) reagent,
was added to each well and incubated at 37 °C and 5 % CO
2
for 3h.Every sample was
done in quadruplets. After 3h of incubation, absorbance was recorded by a plate reader at
490 nm.

6.5.2 Confocal microscopy
C4-2B cells were maintained in T media supplemented with 10% FBS and 0.5% pen-
strep. For confocal microscopy, Mat Tek 35 mm diameter dishes with 14 mm glass
bottom were used. 30,000 cells/dish were plated and incubated overnight at 37°C and 5%
CO
2
. Once the cells were attached the dishes with cells were treated with MHI-Clg,
mitotracker green and DAPI at desired concentrations and incubated for 40 minutes at 37
°C in 5% CO
2
. Confocal imaging was then performed on 510 Zeiss confocal microscope
at excitation wavelengths of 488 nm, 633 nm and790 nm.
234
6.5.3 Near Infrared Imaging (NIR) imaging
C4-2B cancer cells were implanted subcutaneously in 4 to 5 week old athymic nude
mice. MHI-Clg was injected i.t at a dose of 0.5 mg/kg and the whole body imaging was
performed by Xenogen IVIS 200 Imaging Series equipped with fluorescent filter sets
with excitation and emission filter sets of 800 and 850 nm, respectively. During imaging
mice were mainained in anesthetized state by using a built-in system to keep animals in
anesthetized for the duration of experiment. All animal studies were performed by
following the Institutional Animal Care and Use Committee (IACUC) protocol.

235
BIBLIOGRAPHY
Adachi, M., T. Taki, et al. (2000). "Significance of integrin alpha 5 gene expression as a prognostic
factor in node-negative non-small cell lung cancer." Clinical Cancer Research 6(1): 96-101.

Airley, R., J. Loncaster, et al. (2001). "Glucose transporter Glut-1 expression correlates with
tumor hypoxia and predicts metastasis-free survival in advanced carcinoma of the cervix."
Clinical Cancer Research 7(4): 928-934.

Albelda, S. M., S. A. Mette, et al. (1990). "INTEGRIN DISTRIBUTION IN MALIGNANT-MELANOMA -
ASSOCIATION OF THE BETA-3-SUBUNIT WITH TUMOR PROGRESSION." Cancer Research 50(20):
6757-6764.

Alghisi, G. C., L. Ponsonnet, et al. (2009). "The Integrin Antagonist Cilengitide Activates alpha V
beta 3, Disrupts VE-Cadherin Localization at Cell Junctions and Enhances Permeability in
Endothelial Cells." Plos One 4(2).

Ardizzoni, A. (2004). "Topotecan in the treatment of recurrent small cell lung cancer: An
update." Oncologist 9: 4-13.

Aumailley, M., M. Gurrath, et al. (1991). "ARG-GLY-ASP CONSTRAINED WITHIN CYCLIC
PENTAPEPTIDES - STRONG AND SELECTIVE INHIBITORS OF CELL-ADHESION TO VITRONECTIN
AND LAMININ FRAGMENT-P1." Febs Letters 291(1): 50-54.

Avraamides, C. J., B. Garmy-Susini, et al. (2008). "Integrins in angiogenesis and
lymphangiogenesis." Nature Reviews Cancer 8(8): 604-617.

Bae, M. K., S. H. Kim, et al. (2006). "Curcumin inhibits hypoxia-induced angiogenesis via down-
regulation of HIF-1." Oncology Reports 15(6): 1557-1562.

Baker, A.-M., T. R. Cox, et al. (2011). "The Role of Lysyl Oxidase in SRC-Dependent Proliferation
and Metastasis of Colorectal Cancer." Journal of the National Cancer Institute 103(5): 407-424.

Barczyk, M., S. Carracedo, et al. (2010). "Integrins." Cell and Tissue Research 339(1): 269-280.

Barlow, D. J. and J. M. Thornton (1988). "HELIX GEOMETRY IN PROTEINS." Journal of Molecular
Biology 201(3): 601-619.

Barth, R. F., J. A. Coderre, et al. (2005). "Boron neutron capture therapy of cancer: Current
status and future prospects." Clinical Cancer Research 11(11): 3987-4002.
236

BeauvaisJouneau, A. and J. P. Thiery (1997). "Multiple roles for integrins during development."
Biology of the Cell 89(1): 5-11.

Belvisi, L., T. Riccioni, et al. (2005). "Biological and molecular properties of a new
alpha(v)beta(3)/alpha(v)beta(5) integrin antagonist." Molecular Cancer Therapeutics 4(11):
1670-1680.

Berg, T., S. B. Cohen, et al. (2002). "Small-molecule antagonists of Myc/Max dimerization inhibit
Myc-induced transformation of chicken embryo fibroblasts." Proceedings of the National
Academy of Sciences of the United States of America 99(6): 3830-3835.

Bernardo, P. H., N. Brasch, et al. (2003). "A novel redox mechanism for the glutathione-
dependent reversible uptake of a fungal toxin in cells." Journal of Biological Chemistry 278(47):
46549-46555.

Bill, M. A., C. Nicholas, et al. (2012). "Structurally Modified Curcumin Analogs Inhibit STAT3
Phosphorylation and Promote Apoptosis of Human Renal Cell Carcinoma and Melanoma Cell
Lines." Plos One 7(8).

Block, K. M., H. Wang, et al. (2009). "Direct Inhibition of Hypoxia-Inducible Transcription Factor
Complex with Designed Dimeric Epidithiodiketopiperazine." Journal of the American Chemical
Society 131(50): 18078-18088.

Bortolato, M., K. Chen, et al. (2008). "Monoamine oxidase inactivation: From pathophysiology to
therapeutics." Advanced Drug Delivery Reviews 60(13-14): 1527-1533.

Braasch, D. A. and D. R. Corey (2002). "Novel antisense and peptide nucleic acid strategies for
controlling gene expression." Biochemistry 41(14): 4503-4510.

Brennan, P., R. Donev, et al. (2008). "Targeting transcription factors for therapeutic benefit."
Molecular Biosystems 4(9): 909-919.

Bristow, R. G. and R. P. Hill (2008). "Hypoxia, DNA repair and genetic instability." Nature Reviews
Cancer 8(3): 180-192.

Brooks, P. C., A. M. P. Montgomery, et al. (1994). "INTEGRIN ALPHA(V)BETA(3) ANTAGONISTS
PROMOTE TUMOR-REGRESSION BY INDUCING APOPTOSIS OF ANGIOGENIC BLOOD-VESSELS."
Cell 79(7): 1157-1164.

237
Brooks, P. C., S. Stromblad, et al. (1996). "Localization of matrix metalloproteinase MMP-2 to the
surface of invasive cells by interaction with integrin alpha v beta 3." Cell 85(5): 683-693.

Brown, J. M. (2007). Tumor Hypoxia in Cancer Therapy. Methods in Enzymology. S. Helmut and
B. Bernhard, Academic Press. Volume 435: 295-321.

Bunn, H. F. and R. O. Poyton (1996). "Oxygen sensing and molecular adaptation to hypoxia."
Physiological Reviews 76(3): 839-885.

Burkhart, D. J., B. T. Kalet, et al. (2004). "Doxorubicin-formaldehyde conjugates targeting
alpha(v)beta(3) integrin." Molecular Cancer Therapeutics 3(12): 1593-1604.

Busillo, J. M. and J. L. Benovic (2007). "Regulation of CXCR4 signaling." Biochimica Et Biophysica
Acta-Biomembranes 1768(4): 952-963.

Cameron, C. M., F. Harding, et al. (2008). "Activation of hypoxic response in human embryonic
stem cell-derived embryoid bodies." Experimental Biology and Medicine 233(8): 1044-1057.

Carlson, C. B., P. Mowery, et al. (2007). "Selective tumor cell targeting using low-affinity,
multivalent interactions." Acs Chemical Biology 2(2): 119-127.

Carmeliet, P., V. Ferreira, et al. (1996). "Abnormal blood vessel development and lethality in
embryos lacking a single VEGF allele." Nature 380(6573): 435-439.

Carruthers, A., J. DeZutter, et al. (2009). "Will the original glucose transporter isoform please
stand up!" American Journal of Physiology-Endocrinology and Metabolism 297(4): E836-E848.

Chai, C. L. L., J. A. Elix, et al. (2004). "Scabrosin esters and derivatives: chemical derivatization
studies and biological evaluation." Bioorganic & Medicinal Chemistry 12(22): 5991-5995.

Chai, C. L. L. and P. Waring (2000). "Redox sensitive epidithiodioxopiperazines in biological
mechanisms of toxicity." Redox Report 5(5): 257-264.

Chapman, R. N. and P. S. Arora (2006). "Optimized synthesis of hydrogen-bond surrogate
helices: Surprising effects of microwave heating on the activity of grubbs catalysts." Organic
Letters 8(25): 5825-5828.

238
Chen, L.-C., S.-H. Tu, et al. (2012). "Human breast cancer cell metastasis is attenuated by lysyl
oxidase inhibitors through down-regulation of focal adhesion kinase and the paxillin-signaling
pathway." Breast Cancer Research and Treatment 134(3): 989-1004.

Chen, L. H., H. Yin, et al. (2005). "P53 alpha-helix mimetics antagonize p53/MDM2 interaction
and activate p53." Molecular Cancer Therapeutics 4(6): 1019-1025.

Chen, Y., Y. X. Zhang, et al. (2005). "Antiangiogenic activity of 11,11'-dideoxyverticillin, a natural
product isolated from the fungus Shiraia bambusicola." Biochemical and Biophysical Research
Communications 329(4): 1334-1342.

Choi, H., Y.-S. Chun, et al. (2006). "Curcumin inhibits hypoxia-inducible factor-1 by degrading aryl
hydrocarbon receptor nuclear translocator: A mechanism of tumor growth inhibition."
Molecular Pharmacology 70(5): 1664-1671.

Chun, Y. S., E. J. Yeo, et al. (2001). "Inhibitory effect of YC-1 on the hypoxic induction of
erythropoietin and vascular endothelial growth factor in Hep3B cells." Biochemical
Pharmacology 61(8): 947-954.

Chung, F.-Y., M.-Y. Huang, et al. (2009). "GLUT1 gene is a potential hypoxic marker in colorectal
cancer patients." Bmc Cancer 9.

Comerford, K. M., T. J. Wallace, et al. (2002). "Hypoxia-inducible factor-1-dependent regulation
of the multidrug resistance (MDR1) gene." Cancer Research 62(12): 3387-3394.

Cook, K. M., S. T. Hilton, et al. (2009). "Epidithiodiketopiperazines Block the Interaction between
Hypoxia-inducible Factor-1 alpha (HIF-1 alpha) and p300 by a Zinc Ejection Mechanism." Journal
of Biological Chemistry 284(39): 26831-26838.

Coral, K., N. Angayarkanni, et al. (2008). "Lysyl Oxidase Activity in the Ocular Tissues and the
Role of LOX in Proliferative Diabetic Retinopathy and Rhegmatogenous Retinal Detachment."
Investigative Ophthalmology & Visual Science 49(11): 4746-4752.

Crews, S. T. (1998). "Control of cell lineage-specific development and transcription by bHLH-PAS
proteins." Genes & Development 12(5): 607-620.

Cunningham, M. L., P. S. Ringrose, et al. (1984). "INHIBITION OF THE GENOTOXICITY OF
BLEOMYCIN BY SUPEROXIDE-DISMUTASE." Mutation Research 135(3): 199-202.

239
Da Prada, M., R. Kettler, et al. (1990). FROM MOCLOBEMIDE TO RO 19-6327 AND RO 41-1049
THE DEVELOPMENT OF A NEW CLASS OF REVERSIBLE SELECTIVE MAO-A AND MAO-B
INHIBITORS. Youdim, M. B. H. And K. F. Tipton: 279-292.

Dann, C. E. and R. K. Bruick (2005). "Dioxygenases as O-2-dependent regulators of the hypoxic
response pathway." Biochemical and Biophysical Research Communications 338(1): 639-647.

Darash-Yahana, M., E. Pikarsky, et al. (2004). "Role of high expression levels of CXCR4 in tumor
growth, vascularization, and metastasis." Faseb Journal 18(9): 1240-+.

Darnell, J. E. (2002). "Transcription factors as targets for cancer therapy." Nature Reviews
Cancer 2(10): 740-749.

Davis, J. M., L. K. Tsou, et al. (2007). "Synthetic non-peptide mimetics of alpha-helices." Chemical
Society Reviews 36(2): 326-334.

De Colibus, L., M. Li, et al. (2005). "Three-dimensional structure of human monoamine oxidase A
(MAO A): Relation to the structures of rat MAO A and human MAO B." Proceedings of the
National Academy of Sciences of the United States of America 102(36): 12684-12689.

Desgrosellier, J. S. and D. A. Cheresh (2010). "Integrins in cancer: biological implications and
therapeutic opportunities." Nature Reviews Cancer 10(1): 9-22.

Dolores Delgado, M. and J. León (2006). "Gene expression regulation and cancer." Clinical and
Translational Oncology 8(11): 780-787.

Dubrovska, A., J. Elliott, et al. (2012). "CXCR4 Expression in Prostate Cancer Progenitor Cells."
Plos One 7(2).

Duggan, M. E., L. T. Duong, et al. (2000). "Nonpeptide alpha(v)beta(3) antagonists. 1.
Transformation of a potent, integrin-selective alpha(IIb)beta(3) antagonist into a potent
alpha(v)beta(3) antagonist." Journal of Medicinal Chemistry 43(20): 3736-3745.

Dumont, D. J., L. Jussila, et al. (1998). "Cardiovascular failure in mouse embryos deficient in
VEGF receptor-3." Science 282(5390): 946-949.

Eliceiri, B. P. (2001). "Integrin and growth factor receptor crosstalk." Circulation Research 89(12):
1104-1110.

240
Eliceiri, B. P. and D. A. Cheresh (1999). "The role of alpha v integrins during angiogenesis:
insights into potential mechanisms of action and clinical development." Journal of Clinical
Investigation 103(9): 1227-1230.

Ema, M., S. Taya, et al. (1997). "A novel bHLH-PAS factor with close sequence similarity to
hypoxia-inducible factor 1α regulates the VEGF expression and is potentially involved in lung and
vascular development." Proceedings of the National Academy of Sciences 94(9): 4273-4278.

Engelkamp, D. and V. vanHeyningen (1996). "Transcription factors in disease." Current Opinion
in Genetics & Development 6(3): 334-342.

Erler, J. T. and A. J. Giaccia (2006). "Lysyl oxidase mediates hypoxic control of metastasis."
Cancer Research 66(21): 10238-10241.

Ernst-Russell, M. A., C. L. L. Chai, et al. (1999). "Structure revision and cytotoxic activity of the
scabrosin esters, epidithiopiperazinediones from the lichen Xanthoparmelia scabrosa."
Australian Journal of Chemistry 52(4): 279-283.

Fagervall, I. and S. B. Ross (1986). "INHIBITION OF MONOAMINE-OXIDASE IN MONOAMINERGIC
NEURONS IN THE RAT-BRAIN BY IRREVERSIBLE INHIBITORS." Biochemical Pharmacology 35(8):
1381-1387.

Farr, L. E., W. H. Sweet, et al. (1954). "NEUTRON CAPTURE THERAPY WITH BORON IN THE
TREATMENT OF GLIOBLASTOMA MULTIFORME." American Journal of Roentgenology 71(2): 279-
293.

Feng, Y., C. C. Broder, et al. (1996). "HIV-1 entry cofactor: Functional cDNA cloning of a seven-
transmembrane, G protein-coupled receptor." Science 272(5263): 872-877.

Ferrara, N., K. CarverMoore, et al. (1996). "Heterozygous embryonic lethality induced by
targeted inactivation of the VEGF gene." Nature 380(6573): 439-442.

Ferrara, N. and T. DavisSmyth (1997). "The biology of vascular endothelial growth factor."
Endocrine Reviews 18(1): 4-25.

Flamand, V., H. Zhao, et al. (2010). "Targeting monoamine oxidase A in advanced prostate
cancer." Journal of Cancer Research and Clinical Oncology 136(11): 1761-1771.

Freedman, S. J., Z. Y. J. Sun, et al. (2003). "Structural basis for negative regulation of hypoxia-
inducible factor-1 alpha by CITED2." Nature Structural Biology 10(7): 504-512.
241

Freedman, S. J., Z. Y. J. Sun, et al. (2002). "Structural basis for recruitment of CBP/p300 by
hypoxia-inducible factor-1 alpha." Proceedings of the National Academy of Sciences of the
United States of America 99(8): 5367-5372.

Fukuyama, T. and Y. Kishi (1976). "TOTAL SYNTHESIS OF GLIOTOXIN." Journal of the American
Chemical Society 98(21): 6723-6724.

Fukuyama, T., S. Nakatsuka, et al. (1981). "TOTAL SYNTHESIS OF GLIOTOXIN,
DEHYDROGLIOTOXIN AND HYALODENDRIN." Tetrahedron 37(11): 2045-2078.

Gatenby, R. A. and R. J. Gillies (2004). "Why do cancers have high aerobic glycolysis?" Nature
Reviews Cancer 4(11): 891-899.

Geha, R. M., I. Rebrin, et al. (2001). "Substrate and inhibitor specificities for human monoamine
oxidase A and B are influenced by a single amino acid." Journal of Biological Chemistry 276(13):
9877-9882.

Geiger, W. B., J. E. Conn, et al. (1944). "Chaetomin, a new antibiotic substance produced by
chaetomium cochliodes II. Isolation and concentration." Journal of Bacteriology 48(5): 531-536.

Giancotti, F. G. and E. Ruoslahti (1999). "Transduction - Integrin signaling." Science 285(5430):
1028-1032.

Goldberg, M. A., S. P. Dunning, et al. (1988). "REGULATION OF THE ERYTHROPOIETIN GENE -
EVIDENCE THAT THE OXYGEN SENSOR IS A HEME PROTEIN." Science 242(4884): 1412-1415.

Graeber, T. G., J. F. Peterson, et al. (1994). "HYPOXIA INDUCES ACCUMULATION OF P53
PROTEIN, BUT ACTIVATION OF A G(1)-PHASE CHECKPOINT BY LOW-OXYGEN CONDITIONS IS
INDEPENDENT OF P53 STATUS." Molecular and Cellular Biology 14(9): 6264-6277.

Gray, L. H., A. D. Conger, et al. (1953). "THE CONCENTRATION OF OXYGEN DISSOLVED IN
TISSUES AT THE TIME OF IRRADIATION AS A FACTOR IN RADIOTHERAPY." British Journal of
Radiology 26(312): 638-648.

Guarracino, D. A., B. N. Bullock, et al. (2011). "Protein-Protein Interactions in Transcription: A
Fertile Ground for Helix Mimetics." Biopolymers 95(1): 1-7.

Hanahan, D. and R. A. Weinberg (2000). "The hallmarks of cancer." Cell 100(1): 57-70.
242

Hare, M. L. C. (1928). "Tyramine oxidase. I. A new enzyme system in liver." Biochemical Journal
22(4): 968-979.

Harker, E. A. and A. Schepartz (2009). "Cell-Permeable beta-Peptide Inhibitors of p53/hDM2
Complexation." Chembiochem 10(6): 990-993.

Hawthorne, M. F. (1993). "THE ROLE OF CHEMISTRY IN THE DEVELOPMENT OF BORON
NEUTRON-CAPTURE THERAPY OF CANCER." Angewandte Chemie-International Edition in English
32(7): 950-984.

Hawthorne, M. F. and M. W. Lee (2003). "A critical assessment of boron target compounds for
boron neutron capture therapy." Journal of Neuro-Oncology 62(1): 33-45.

Heilig, C. W., T. Saunders, et al. (2003). "Glucose transporter-1-deficient mice exhibit impaired
development and deformities that are similar to diabetic embryopathy." Proceedings of the
National Academy of Sciences of the United States of America 100(26): 15613-15618.

Henchey, L. K., A. L. Jochim, et al. (2008). "Contemporary strategies for the stabilization of
peptides in the alpha-helical conformation." Current Opinion in Chemical Biology 12(6): 692-
697.

Henchey, L. K., S. Kushal, et al. (2010). "Inhibition of Hypoxia Inducible Factor 1-Transcription
Coactivator Interaction by a Hydrogen Bond Surrogate alpha-Helix." Journal of the American
Chemical Society 132(3): 941-+.

Henchey, L. K., J. R. Porter, et al. (2010). "High Specificity in Protein Recognition by Hydrogen-
Bond-Surrogate alpha-Helices: Selective Inhibition of the p53/MDM2 Complex." Chembiochem
11(15): 2104-2107.

Henning, T., M. Kraus, et al. (2004). "Relevance of tumor microenvironment for progression,
therapy and drug development." Anti-Cancer Drugs 15(1): 7-14.

Higgins, D. F., K. Kimura, et al. (2007). "Hypoxia promotes fibrogenesis in vivo via HIF-1
stimulation of epithelial-to-mesenchymal transition." Journal of Clinical Investigation 117(12):
3810-3820.

Hino, T. and M. Nakagawa (1989). Chapter 1 Chemistry and Reactions of Cyclic Tautomers of
Tryptamines and Tryptophans. The Alkaloids: Chemistry and Pharmacology. B. Arnold, Academic
Press. Volume 34: 1-75.
243

Hoeben, A., B. Landuyt, et al. (2004). "Vascular endothelial growth factor and angiogenesis."
Pharmacological Reviews 56(4): 549-580.

Holinger, E. P., T. Chittenden, et al. (1999). "Bak BH3 peptides antagonize Bcl-x(L) function and
induce apoptosis through cytochrome c-independent activation of caspases." Journal of
Biological Chemistry 274(19): 13298-13304.

Holmes, D. I. R. and I. Zachary (2005). "The vascular endothelial growth factor (VEGF) family:
angiogenic factors in health and disease." Genome Biology 6(2).

Hood, J. D., M. Bednarski, et al. (2002). "Tumor regression by targeted gene delivery to the
neovasculature." Science 296(5577): 2404-2407.

Hood, J. D., R. Frausto, et al. (2003). "Differential alpha v integrin-mediated Ras-ERK signaling
during two pathways of angiogenesis." Journal of Cell Biology 162(5): 933-943.

Hudson, C. C., M. Liu, et al. (2002). "Regulation of hypoxia-inducible factor 1 alpha expression
and function by the mammalian target of rapamycin." Molecular and Cellular Biology 22(20):
7004-7014.

Hughes, P. E., T. E. Otoole, et al. (1995). "THE CONSERVED MEMBRANE-PROXIMAL REGION OF
AN INTEGRIN CYTOPLASMIC DOMAIN SPECIFIES LIGAND-BINDING AFFINITY." Journal of
Biological Chemistry 270(21): 12411-12417.

Hui, A. S., A. L. Bauer, et al. (2006). "Calcium signaling stimulates translation of HIF-alpha during
hypoxia." Faseb Journal 20(3): 466-475.

Hur, E., K. Y. Chang, et al. (2001). "Mitogen-activated protein kinase kinase inhibitor PD98059
blocks the trans-activation but not the stabilization or DNA binding ability of hypoxia-inducible
factor-1 alpha." Molecular Pharmacology 59(5): 1216-1224.

Hurne, A. M., C. L. L. Chai, et al. (2002). "Influx of calcium through a redox-sensitive plasma
membrane channel in thymocytes causes early necrotic cell death induced by the
epipolythiodioxopiperazine toxins." Journal of Biological Chemistry 277(35): 31631-31638.

Hynes, R. O. (2002). "Integrins: Bidirectional, allosteric signaling machines." Cell 110(6): 673-687.

244
Isaacs, J. S., Y. J. Jung, et al. (2002). "Hsp90 regulates a von Hippel Lindau-independent hypoxia-
inducible factor-1 alpha-degradative pathway." Journal of Biological Chemistry 277(33): 29936-
29944.

Ivan, M., K. Kondo, et al. (2001). "HIF alpha targeted for VHL-mediated destruction by proline
hydroxylation: Implications for O-2 sensing." Science 292(5516): 464-468.

Ivaska, J. and J. Heino (2011). Cooperation Between Integrins and Growth Factor Receptors in
Signaling and Endocytosis. Annual Review of Cell and Developmental Biology, Vol 27. R.
Schekman, L. Goldstein and R. Lehmann. 27: 291-320.

Iwasa, E., Y. Hamashima, et al. (2010). "Total Synthesis of (+)-Chaetocin and its Analogues: Their
Histone Methyltransferase G9a Inhibitory Activity." Journal of the American Chemical Society
132(12): 4078-+.

Iyer, N. V., L. E. Kotch, et al. (1998). "Cellular and developmental control of O-2 homeostasis by
hypoxia-inducible factor 1 alpha." Genes & Development 12(2): 149-162.

Jaakkola, P., D. R. Mole, et al. (2001). "Targeting of HIF-alpha to the von Hippel-Lindau
ubiquitylation complex by O-2-regulated prolyl hydroxylation." Science 292(5516): 468-472.

Jantsch, J., D. Chakravortty, et al. (2008). "Hypoxia and hypoxia-inducible factor-1 alpha
modulate lipopolysaccharide-induced dendritic cell activation and function." Journal of
Immunology 180(7): 4697-4705.

Jeong, J. W., M. K. Bae, et al. (2002). "Regulation and destabilization of HIF-1 alpha by ARD1-
mediated acetylation." Cell 111(5): 709-720.

Jiang, B. H., F. Agani, et al. (1997). "V-SRC induces expression of hypoxia-inducible factor 1 (HIF-
1) and transcription of genes encoding vascular endothelial growth factor and enolase 1:
Involvement of HIF-1 in tumor progression." Cancer Research 57(23): 5328-5335.

Jiang, B. H., G. Q. Jiang, et al. (2001). "Phosphatidylinositol 3-kinase signaling controls levels of
hypoxia-inducible factor." Cell Growth & Differentiation 12(7): 363-369.

Jiang, B. H., E. Rue, et al. (1996). "Dimerization, DNA binding, and transactivation properties of
hypoxia-inducible factor 1." Journal of Biological Chemistry 271(30): 17771-17778.

245
Jiao, M. and K.-J. Nan (2012). "Activation of PI3 kinase/Akt/HIF-1 alpha pathway contributes to
hypoxia-induced epithelial-mesenchymal transition and chemoresistance in hepatocellular
carcinoma." International Journal of Oncology 40(2): 461-468.

Jochim, A. L. and P. S. Arora (2009). "Assessment of helical interfaces in protein-protein
interactions." Molecular Biosystems 5(9): 924-926.

Jolley, M. E., S. D. Stroupe, et al. (1981). "FLUORESCENCE POLARIZATION IMMUNOASSAY .3. AN
AUTOMATED-SYSTEM FOR THERAPEUTIC DRUG DETERMINATION." Clinical Chemistry 27(9):
1575-1579.

Jordan, T. W. and S. J. Cordiner (1987). "FUNGAL EPIPOLYTHIODIOXOPIPERAZINE TOXINS HAVE
THERAPEUTIC POTENTIAL AND ROLES IN DISEASE." Trends in Pharmacological Sciences 8(4):
144-149.

Jung, H. J., J. H. Kim, et al. (2010). "A Novel Ca2+/Calmodulin Antagonist HBC Inhibits
Angiogenesis and Down-regulates Hypoxia-inducible Factor." Journal of Biological Chemistry
285(33): 25867-25874.

Karkkainen, M. J., P. Haiko, et al. (2004). "Vascular endothelial growth factor C is required for
sprouting of the first lymphatic vessels from embryonic veins." Nature Immunology 5(1): 74-80.

Kasper, A. C., E. J. Moon, et al. (2009). "Analysis of HIF-1 inhibition by manassantin A and
analogues with modified tetrahydrofuran configurations." Bioorganic & Medicinal Chemistry
Letters 19(14): 3783-3786.

Katagiri, K., K. Sato, et al. (1970). "VERTICILLIN A, A NEW ANTIBIOTIC FROM VERTICILLIUM-SP."
Journal of Antibiotics 23(8): 420-&.

Kato, I., K. Ono, et al. (2004). "Effectiveness of BNCT for recurrent head and neck malignancies."
Applied Radiation and Isotopes 61(5): 1069-1073.

Kawahara, N., S. Nakajima, et al. (1989). "STUDIES ON FUNGAL PRODUCTS .29. STRUCTURE OF A
NOVEL EPIDITHIODIOXOPIPERAZINE, EMETHALLICIN-A, A POTENT INHIBITOR OF HISTAMINE-
RELEASE, FROM EMERICELLA-HETEROTHALLICA." Chemical & Pharmaceutical Bulletin 37(10):
2592-2595.

Kim, J., J. A. Ashenhurst, et al. (2009). "Total Synthesis of (+)-11,11 '-Dideoxyverticillin A."
Science 324(5924): 238-241.

246
Kim, J. W., I. Tchernyshyov, et al. (2006). "HIF-1-mediated expression of pyruvate
dehydrogenase kinase: A metabolic switch required for cellular adaptation to hypoxia." Cell
Metabolism 3(3): 177-185.

Korsisaari, N., I. M. Kasman, et al. (2007). "Inhibition of VEGF-A prevents the angiogenic switch
and results in increased survival of Apc(+/min) mice." Proceedings of the National Academy of
Sciences of the United States of America 104(25): 10625-10630.

Kostenko, O. M., S. Y. Dmitrieva, et al. (2002). "New approach for fluorescent peptide labeling
with pyrylium cyanine dyes." Journal of Fluorescence 12(2): 173-175.

Krawetz, S. A. (1994). "THE ORIGIN OF LYSYL OXIDASE." Comparative Biochemistry and
Physiology B-Biochemistry & Molecular Biology 108(1): 117-119.

Krebs, C. J. and S. A. Krawetz (1993). "LYSYL OXIDASE COPPER-TALON COMPLEX - A MODEL."
Biochimica Et Biophysica Acta 1202(1): 7-12.

Kumar, B., S. Koul, et al. (2008). "Oxidative stress is inherent in prostate cancer cells and is
required for aggressive phenotype." Cancer Research 68(6): 1777-1785.

Kung, A. L., S. Wang, et al. (2000). "Suppression of tumor growth through disruption of hypoxia-
inducible transcription." Nature Medicine 6(12): 1335-1340.

Kung, A. L., S. D. Zabludoff, et al. (2004). "Small molecule blockade of transcriptional
coactivation of the hypoxia-inducible factor pathway." Cancer Cell 6(1): 33-43.

Kushal, S., H. Wang, et al. (2011). "Inhibition of Hypoxia-Inducible Transcription Factor Complex
with Designed Epipolythiodiketopiperazine." Biopolymers 95(1): 8-16.

Lando, D., D. J. Peet, et al. (2002). "FIH-1 is an asparaginyl hydroxylase enzyme that regulates
the transcriptional activity of hypoxia-inducible factor." Genes & Development 16(12): 1466-
1471.

Lando, D., D. J. Peet, et al. (2002). "Asparagine hydroxylation of the HIF transactivation domain:
A hypoxic switch." Science 295(5556): 858-861.

Latchman, D. S. (1996). "Mechanisms of disease - Transcription-factor mutations and disease."
New England Journal of Medicine 334(1): 28-33.

247
Lau, K. W., Y. M. Tian, et al. (2007). "Target gene selectivity of hypoxia-inducible factor-alpha in
renal cancer cells is conveyed by post-DNA-binding mechanisms." British Journal of Cancer
96(8): 1284-1292.

Lee, Y. M., S. H. Kim, et al. (2003). "Inhibition of hypoxia-induced angiogenesis by FK228, a
specific histone deacetylase inhibitor, via suppression of HIF-1 alpha activity." Biochemical and
Biophysical Research Communications 300(1): 241-246.

Levental, K. R., H. Yu, et al. (2009). "Matrix Crosslinking Forces Tumor Progression by Enhancing
Integrin Signaling." Cell 139(5): 891-906.

Li, S. H., D. H. Shin, et al. (2008). "A novel mode of action of YC-1 in HIF inhibition: stimulation of
FIH-dependent p300 dissociation from HIF-1 alpha." Molecular Cancer Therapeutics 7(12): 3729-
3738.

Li, X., S.-K. Kim, et al. (2006). "A new antibacterial dioxopiperazine alkaloid related to gliotoxin
from a marine isolate of the fungus Pseudallescheria." Journal of Antibiotics 59(4): 248-250.

Li, X., Q. Ma, et al. (2012). "SDF-1/CXCR4 signaling induces pancreatic cancer cell invasion and
epithelial-mesenchymal transition in vitro through non-canonical activation of Hedgehog
pathway." Cancer Letters 322(2): 169-176.

Liddington, R. C. and M. H. Ginsberg (2002). "Integrin activation takes shape." Journal of Cell
Biology 158(5): 833-839.

Lim, J.-H., Y.-M. Lee, et al. (2010). "Sirtuin 1 Modulates Cellular Responses to Hypoxia by
Deacetylating Hypoxia-Inducible Factor 1 alpha." Molecular Cell 38(6): 864-878.

Lisy, K. and D. J. Peet (2008). "Turn me on: regulating HIF transcriptional activity." Cell Death and
Differentiation 15(4): 642-+.

Liu, F., Q. Liu, et al. (2011). "Verticillin A Overcomes Apoptosis Resistance in Human Colon
Carcinoma through DNA Methylation-Dependent Upregulation of BNIP3." Cancer Research
71(21): 6807-6816.

Liu, Z., F. Wang, et al. (2008). "Integrin alpha(v)beta(3)-Targeted Cancer Therapy." Drug
Development Research 69(6): 329-339.

Locher, G. L. (1936). "Biological effects and therapeutic possibilities of neutrons." American
Journal of Roentgenology and Radium Therapy 36: 1-13.
248

Logothetis, C. J. and S. H. Lin (2005). "Osteoblasts in prostate cancer metastasis to bone." Nature
Reviews Cancer 5(1): 21-28.

Lohela, M. and Z. Werb (2010). "Intravital imaging of stromal cell dynamics in tumors." Current
Opinion in Genetics & Development 20(1): 72-78.

Loor, G. and P. T. Schumacker (2008). "Role of hypoxia-inducible factor in cell survival during
myocardial ischemia-reperfusion." Cell Death and Differentiation 15(4): 686-690.

Lopez-Lazaro, M. (2006). "HIF-1: hypoxia-inducible factor or dysoxia-inducible factor?" Faseb
Journal 20(7): 828-832.

Los, M., J. M. L. Roodhart, et al. (2007). "Target practice: Lessons from phase III trials with
bevacizumab and vatalanib in the treatment of advanced colorectal cancer." Oncologist 12(4):
443-450.

Lou, J. J. W., Y. L. Chua, et al. (2010). "Inhibition of Hypoxia-Inducible Factor-1 alpha (HIF-1
alpha) Protein Synthesis by DNA Damage Inducing Agents." Plos One 5(5).

Lu, P., V. M. Weaver, et al. (2012). "The extracellular matrix: A dynamic niche in cancer
progression." Journal of Cell Biology 196(4): 395-406.

Luo, S., E. Zhang, et al. (2011). "A review of NIR dyes in cancer targeting and imaging."
Biomaterials 32(29): 7127-7138.

Mabjeesh, N. J., D. E. Post, et al. (2002). "Geldanamycin induces degradation of hypoxia-
inducible factor 1 alpha protein via the proteosome pathway in prostate cancer cells." Cancer
Research 62(9): 2478-2482.

Macheda, M. L., S. Rogers, et al. (2005). "Molecular and cellular regulation of glucose
transporter (GLUT) proteins in cancer." Journal of Cellular Physiology 202(3): 654-662.

Mahabeleshwar, G. H., W. Feng, et al. (2007). "Mechanisms of integrin-vascular endothelial
growth factor receptor cross-activation in angiogenesis." Circulation Research 101(6): 570-580.

Mahon, P. C., K. Hirota, et al. (2001). "FIH-1: a novel protein that interacts with HIF-1 alpha and
VHL to mediate repression of HIF-1 transcriptional activity." Genes & Development 15(20):
2675-2686.
249

Majumder, P. K., P. G. Febbo, et al. (2004). "mTOR inhibition reverses Akt-dependent prostate
intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways."
Nature Medicine 10(6): 594-601.

Maki, J. M., R. Sormunen, et al. (2005). "Lysyl oxidase is essential for normal development and
function of the respiratory system and for the integrity of elastic and collagen fibers in various
tissues." American Journal of Pathology 167(4): 927-936.

Makino, Y., R. H. Cao, et al. (2001). "Inhibitory PAS domain protein is a negative regulator of
hypoxia-inducible gene expression." Nature 414(6863): 550-554.

Manalo, D. J., A. Rowan, et al. (2005). "Transcriptional regulation of vascular endothelial cell
responses to hypoxia by HIF-1." Blood 105(2): 659-669.

Marchini, M., M. Mingozzi, et al. (2012). "Cyclic RGD Peptidomimetics Containing Bifunctional
Diketopiperazine Scaffolds as New Potent Integrin Ligands." Chemistry-a European Journal
18(20): 6195-6207.

Martin, R. F., G. Dcunha, et al. (1988). "INDUCTION OF DOUBLE-STRAND BREAKS FOLLOWING
NEUTRON-CAPTURE BY DNA-BOUND GD-157." International Journal of Radiation Biology 54(2):
205-208.

Mas-Moruno, C., F. Rechenmacher, et al. (2010). "Cilengitide: The First Anti-Angiogenic Small
Molecule Drug Candidate. Design, Synthesis and Clinical Evaluation." Anti-Cancer Agents in
Medicinal Chemistry 10(10): 753-768.

Masson, N., C. Willam, et al. (2001). "Independent function of two destruction domains in
hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation." Embo Journal 20(18):
5197-5206.

Maxwell, P. H., G. U. Dachs, et al. (1997). "Hypoxia-inducible factor-1 modulates gene expression
in solid tumors and influences both angiogenesis and tumor growth." Proceedings of the
National Academy of Sciences of the United States of America 94(15): 8104-8109.

Maxwell, P. H., M. S. Wiesener, et al. (1999). "The tumour suppressor protein VHL targets
hypoxia-inducible factors for oxygen-dependent proteolysis." Nature 399(6733): 271-275.

250
Mazure, N. M., M. C. Brahimi-Horn, et al. (2004). "HIF-1: master and commander of the hypoxic
world - A pharmacological approach to its regulation by siRNAs." Biochemical Pharmacology
68(6): 971-980.

McDevitt, M. R., D. S. Ma, et al. (2001). "Tumor therapy with targeted atomic nanogenerators."
Science 294(5546): 1537-1540.

McDougal.Jk (1969). "ANTIVIRAL ACTION OF GLIOTOXIN." Archiv Fur Die Gesamte
Virusforschung 27(2-4): 255-&.

McInnes, A. G., A. Taylor, et al. (1976). "STRUCTURE OF CHETOMIN." Journal of the American
Chemical Society 98(21): 6741-6741.

Mitra, S. K. and D. D. Schlaepfer (2006). "Integrin-regulated FAK-Src signaling in normal and
cancer cells." Current Opinion in Cell Biology 18(5): 516-523.

Moellering, R. E., M. Cornejo, et al. (2009). "Direct inhibition of the NOTCH transcription factor
complex." Nature 462(7270): 182-U157.

Mulgrew, K., K. Kinneer, et al. (2006). "Direct targeting of alpha(v)beta(3) integrin on tumor cells
with a monoclonal antibody, Abegrin (TM)." Molecular Cancer Therapeutics 5(12): 3122-3129.

Munday, R. (1982). "STUDIES ON THE MECHANISM OF TOXICITY OF THE MYCO-TOXIN,
SPORIDESMIN .1. GENERATION OF SUPEROXIDE RADICAL BY SPORIDESMIN." Chemico-Biological
Interactions 41(3): 361-374.

Murphy, B. J., K. R. Laderoute, et al. (1994). "METALLOTHIONEIN IIA IS UP-REGULATED BY
HYPOXIA IN HUMAN A431 SQUAMOUS CARCINOMA-CELLS." Cancer Research 54(22): 5808-
5810.

Murphy, E. A., B. K. Majeti, et al. (2008). "Nanoparticle-mediated drug delivery to tumor
vasculature suppresses metastasis." Proceedings of the National Academy of Sciences of the
United States of America 105(27): 9343-9348.

Na, I.-K., C. Scheibenbogen, et al. (2008). "Nuclear expression of CXCR4 in tumor cells of non-
small cell lung cancer is correlated with lymph node metastasis." Human Pathology 39(12):
1751-1755.

251
Nagaraja.R, N. Neuss, et al. (1968). "ARANOTIN AND RELATED METABOLITES .3.
CONFIGURATION AND CONFORMATION OF ACETYLARANOTIN." Journal of the American
Chemical Society 90(23): 6518-&.

Neuss, N., L. D. Boeck, et al. (1968). "Aranotin and related metabolites from Arachniotus aureus
(Eidam) Schroeter. IV. Fermentation, isolation, structure elucidation, biosynthesis, and antiviral
properties." Antimicrob Agents Chemother (Bethesda) 8: 213-219.

Niu, G. and X. Chen (2011). "Why Integrin as a Primary Target for Imaging and Therapy."
Theranostics 1: 30-47.

Nizet, V. and R. S. Johnson (2009). "Interdependence of hypoxic and innate immune responses."
Nature Reviews Immunology 9(9): 609-617.

Olenyuk, B. Z., G. J. Zhang, et al. (2004). "Inhibition of vascular endothelial growth factor with a
sequence-specific hypoxia response element antagonist." Proceedings of the National Academy
of Sciences of the United States of America 101(48): 16768-16773.

Olofsson, B., E. Korpelainen, et al. (1998). "Vascular endothelial growth factor B (VEGF-B) binds
to VEGF receptor-1 and regulates plasminogen activator activity in endothelial cells."
Proceedings of the National Academy of Sciences of the United States of America 95(20): 11709-
11714.

Olofsson, B., K. Pajusola, et al. (1996). "Genomic organization of the mouse and human genes
for vascular endothelial growth factor B (VEGF-B) and characterization of a second splice
isoform." Journal of Biological Chemistry 271(32): 19310-19317.

Olsson, A. K., A. Dimberg, et al. (2006). "VEGF receptor signalling - in control of vascular
function." Nature Reviews Molecular Cell Biology 7(5): 359-371.

Onodera, H., A. Hasegawa, et al. (2004). "MPC1001 and its analogues: New antitumor agents
from the fungus Cladorrhinum species." Organic Letters 6(22): 4101-4104.

Opalinska, J. B. and A. M. Gewirtz (2002). "Nucleic-acid therapeutics: Basic principles and recent
applications." Nature Reviews Drug Discovery 1(7): 503-514.

Owen, R. M., C. B. Carlson, et al. (2007). "Bifunctional ligands that target cells displaying the
alpha(v)beta(3) integrin." Chembiochem 8(1): 68-82.

252
Pahl, H. L., B. Krauss, et al. (1996). "The immunosuppressive fungal metabolite gliotoxin
specifically inhibits transcription factor NF-kappa B." Journal of Experimental Medicine 183(4):
1829-1840.

Panes, J., M. Perry, et al. (1999). "Leukocyte endothelial cell adhesion: avenues for therapeutic
intervention." British Journal of Pharmacology 126(3): 537-550.

Patgiri, A., A. L. Jochim, et al. (2008). "A Hydrogen Bond Surrogate Approach for Stabilization of
Short Peptide Sequences in alpha-Helical Conformation." Accounts of Chemical Research 41(10):
1289-1300.

Peehl, D. M., M. Coram, et al. (2008). "The Significance of Monoamine Oxidase-A Expression in
High Grade Prostate Cancer." Journal of Urology 180(5): 2206-2211.

Perrin, F. (1926). "The polarisation of flourescence light. Average life of molecules in their
excited state." Journal De Physique Et Le Radium 7: 390-401.

Peyssonnaux, C., V. Datta, et al. (2005). "HIF-1 alpha expression regulates the bactericidal
capacity of phagocytes." Journal of Clinical Investigation 115(7): 1806-1815.

Pirkmajer, S. and A. V. Chibalin (2011). "Serum starvation: caveat emptor." American Journal of
Physiology-Cell Physiology 301(2): C272-C279.

Pugh, C. W., J. F. Orourke, et al. (1997). "Activation of hypoxia-inducible factor-1; Definition of
regulatory domains within the alpha subunit." Journal of Biological Chemistry 272(17): 11205-
11214.

Pugh, C. W. and P. J. Ratcliffe (2003). "Regulation of angiogenesis by hypoxia: role of the HIF
system." Nature Medicine 9(6): 677-684.

Rapisarda, A., B. Uranchimeg, et al. (2004). "Topoisomerase I-mediated inhibition of hypoxia-
inducible factor 1: Mechanism and therapeutic implications." Cancer Research 64(4): 1475-1482.

Rebsam, A., I. Seif, et al. (2005). "Dissociating barrel development and lesion-induced plasticity
in the mouse somatosensory cortex." Journal of Neuroscience 25(3): 706-710.

Rezanka, T., M. Sobotka, et al. (2006). "Pharmacologically active sulfur-containing compounds."
Anti-Infective Agents in Medicinal Chemistry 5(2): 187-224.

253
Rice, G. C., C. Hoy, et al. (1986). "TRANSIENT HYPOXIA ENHANCES THE FREQUENCY OF
DIHYDROFOLATE-REDUCTASE GENE AMPLIFICATION IN CHINESE-HAMSTER OVARY CELLS."
Proceedings of the National Academy of Sciences of the United States of America 83(16): 5978-
5982.

Riederer, P. and G. Laux (2011). "MAO-inhibitors in Parkinson's Disease." Exp Neurobiol 20(1): 1-
17.

Roberts, M., S. Barry, et al. (2001). "PDGF-regulated rab4-dependent recycling of alpha v beta 3
integrin from early endosomes is necessary for cell adhesion and spreading." Current Biology
11(18): 1392-1402.

Robinson, C. J. and S. E. Stringer (2001). "The splice variants of vascular endothelial growth
factor (VEGF) and their receptors." Journal of Cell Science 114(5): 853-865.

Roskoski, R., Jr. (2007). "Vascular endothelial growth factor (VEGF) signaling in tumor
progression." Critical Reviews in Oncology Hematology 62(3): 179-213.

Roskoski, R., Jr. (2008). "VEGF receptor protein-tyrosine kinases: Structure and regulation."
Biochemical and Biophysical Research Communications 375(3): 287-291.

Ross, R. S. (2004). "Molecular and mechanical synergy: cross-talk between integrins and growth
factor receptors." Cardiovascular Research 63(3): 381-390.

Ruan, K., G. Song, et al. (2009). "Role of Hypoxia in the Hallmarks of Human Cancer." Journal of
Cellular Biochemistry 107(6): 1053-1062.

Ruas, J. L., U. Berchner-Pfannschmidt, et al. (2010). "Complex Regulation of the Transactivation
Function of Hypoxia-inducible Factor-1 alpha by Direct Interaction with Two Distinct Domains of
the CREB-binding Protein/p300." Journal of Biological Chemistry 285(4): 2601-2609.

Ruoslahti, E. and M. D. Pierschbacher (1987). "NEW PERSPECTIVES IN CELL-ADHESION - RGD
AND INTEGRINS." Science 238(4826): 491-497.

Ryan, H. E., J. Lo, et al. (1998). "HIF-1 alpha is required for solid tumor formation and embryonic
vascularization." Embo Journal 17(11): 3005-3015.

Schafmeister, C. E., J. Po, et al. (2000). "An all-hydrocarbon cross-linking system for enhancing
the helicity and metabolic stability of peptides." Journal of the American Chemical Society
122(24): 5891-5892.
254

Scharf, D. H., N. Remme, et al. (2011). "A Dedicated Glutathione S-Transferase Mediates Carbon-
Sulfur Bond Formation in Gliotoxin Biosynthesis." Journal of the American Chemical Society
133(32): 12322-12325.

Scharf, D. H., N. Remme, et al. (2010). "Transannular Disulfide Formation in Gliotoxin
Biosynthesis and Its Role in Self-Resistance of the Human Pathogen Aspergillus fumigatus."
Journal of the American Chemical Society 132(29): 10136-10141.

Schlaepfer, D. D. and T. Hunter (1996). "Evidence for in vivo phosphorylation of the Grb2 SH2-
domain binding site on focal adhesion kinase by Src-family protein-tyrosine kinases." Molecular
and Cellular Biology 16(10): 5623-5633.

Schlee, M., V. Hornug, et al. (2006). "siRNA and isRNA: Two edges of one sword." Molecular
Therapy 14(4): 463-470.

Schmidt, N., A. Mishra, et al. (2010). "Arginine-rich cell-penetrating peptides." Febs Letters
584(9): 1806-1813.

Sekita, S., K. Yoshihira, et al. (1981). "MYCO-TOXIN PRODUCTION BY CHAETOMIUM SPP AND
RELATED FUNGI." Canadian Journal of Microbiology 27(8): 766-772.

Semenza, G. L. (1998). "Hypoxia-inducible factor 1: master regulator of O-2 homeostasis."
Current Opinion in Genetics & Development 8(5): 588-594.

Semenza, G. L. (2000). "HIF-1 and human disease: one highly involved factor." Genes &
Development 14(16): 1983-1991.

Semenza, G. L. (2002). "Involvement of hypoxia-inducible factor 1 in human cancer." Internal
Medicine 41(2): 79-83.

Semenza, G. L. (2003). "Targeting HIF-1 for cancer therapy." Nature Reviews Cancer 3(10): 721-
732.

Semenza, G. L. (2007). "Life with oxygen." Science 318(5847): 62-64.

Semenza, G. L., B. H. Jiang, et al. (1996). "Hypoxia response elements in the aldolase A, enolase
1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-
inducible factor 1." Journal of Biological Chemistry 271(51): 32529-32537.
255

Semenza, G. L., M. K. Nejfelt, et al. (1991). "HYPOXIA-INDUCIBLE NUCLEAR FACTORS BIND TO AN
ENHANCER ELEMENT LOCATED 3' TO THE HUMAN ERYTHROPOIETIN GENE." Proceedings of the
National Academy of Sciences of the United States of America 88(13): 5680-5684.

Semenza, G. L. and G. L. Wang (1992). "A NUCLEAR FACTOR INDUCED BY HYPOXIA VIA DENOVO
PROTEIN-SYNTHESIS BINDS TO THE HUMAN ERYTHROPOIETIN GENE ENHANCER AT A SITE
REQUIRED FOR TRANSCRIPTIONAL ACTIVATION." Molecular and Cellular Biology 12(12): 5447-
5454.

Serini, G., L. Napione, et al. (2008). "Besides adhesion: new perspectives of integrin functions in
angiogenesis." Cardiovascular Research 78(2): 213-222.

Shafee, N., S. Kaluz, et al. (2009). "PI3K/Akt activity has variable cell-specific effects on
expression of HIF target genes, CA9 and VEGF, in human cancer cell lines." Cancer Letters
282(1): 109-115.

Shah, D. T. and B. Larsen (1991). "CLINICAL ISOLATES OF YEAST PRODUCE A GLIOTOXIN-LIKE
SUBSTANCE." Mycopathologia 116(3): 203-208.

Shay, J. E. S. and M. C. Simon (2012). "Hypoxia-inducible factors: Crosstalk between
inflammation and metabolism." Seminars in Cell & Developmental Biology 23(4): 389-394.

Shen, M. M. and C. Abate-Shen (2010). "Molecular genetics of prostate cancer: new prospects
for old challenges." Genes & Development 24(18): 1967-2000.

Shibata, T., A. J. Giaccia, et al. (2000). "Development of a hypoxia-responsive vector for tumor-
specific gene therapy." Gene Therapy 7(6): 493-498.

Shih, J. C., K. Chen, et al. (1999). Monoamine oxidase: From genes to behavior. Annual Review of
Neuroscience. W. M. Cowan. 22: 197-217.

Shirinsky, I. V. and V. S. Shirinsky (2011). "Targeting Nuclear Hormone Receptors: PPARα
Agonists as Potential Disease-Modifying Drugs for Rheumatoid Arthritis." International Journal
of Rheumatology 2011.

Siddiquee, K., S. Zhang, et al. (2007). "Selective chemical probe inhibitor of Stat3, identified
through structure-based virtual screening, induces antitumor activity." Proceedings of the
National Academy of Sciences of the United States of America 104(18): 7391-7396.

256
Snyder, H. R., A. J. Reedy, et al. (1958). "SYNTHESIS OF AROMATIC BORONIC ACIDS - ALDEHYDO
BORONIC ACIDS AND A BORONIC ACID ANALOG OF TYROSINE." Journal of the American
Chemical Society 80(4): 835-838.

Son, B. W., P. R. Jensen, et al. (1999). "New cytotoxic epidithiodioxopiperazines related to
verticillin A from a marine isolate of the fungus Penicillium." Natural Product Letters 13(3): 213-
222.

Springer, T. A. (1997). "Folding of the N-terminal, ligand-binding region of integrin alpha-
subunits into a beta-propeller domain." Proceedings of the National Academy of Sciences of the
United States of America 94(1): 65-72.

Streuli, C. H. (2009). "Integrins and cell-fate determination." Journal of Cell Science 122(2): 171-
177.

Stupack, D. G. and D. A. Cheresh (2002). "Get a ligand, get a life: integrins, signaling and cell
survival." Journal of Cell Science 115(19): 3729-3738.

Suhadolnik, R. J. and R. G. Chenoweth (1958). "BIOSYNTHESIS OF GLIOTOXIN .1.
INCORPORATION OF PHENYLALANINE-1-C-14 AND PHENYLALANINE-2-C-14." Journal of the
American Chemical Society 80(16): 4391-4392.

Sun, X., L. Wei, et al. (2010). "CXCR4/SDF1 mediate hypoxia induced chondrosarcoma cell
invasion through ERK signaling and increased MMP1 expression." Molecular Cancer 9.

Takada, Y., X. Ye, et al. (2007). "The integrins." Genome Biology 8(5).

Takahashi, C., A. Numata, et al. (1994). "LEPTOSIN-I AND LEPTOSIN-J, CYTOTOXIC SUBSTANCES
PRODUCED BY A LEPTOSPHAERIA SP - PHYSICOCHEMICAL PROPERTIES AND STRUCTURES."
Journal of Antibiotics 47(11): 1242-1249.

Tian, H., S. L. McKnight, et al. (1997). "Endothelial PAS domain protein 1 (EPAS1), a transcription
factor selectively expressed in endothelial cells." Genes & Development 11(1): 72-82.

Tipton, K. F., S. Boyce, et al. (2004). "Monoamine oxidases: Certainties and uncertainties."
Current Medicinal Chemistry 11(15): 1965-1982.

Tischer, E., R. Mitchell, et al. (1991). "THE HUMAN GENE FOR VASCULAR ENDOTHELIAL
GROWTH-FACTOR - MULTIPLE PROTEIN FORMS ARE ENCODED THROUGH ALTERNATIVE EXON
SPLICING." Journal of Biological Chemistry 266(18): 11947-11954.
257

Tredan, O., C. M. Galmarini, et al. (2007). "Drug resistance and the solid tumor
microenvironment." Journal of the National Cancer Institute 99(19): 1441-1454.

True, L., I. Coleman, et al. (2006). "A molecular correlate to the Gleason grading system for
prostate adenocarcinoma." Proceedings of the National Academy of Sciences of the United
States of America 103(29): 10991-10996.

Tsugeno, Y. and A. Ito (1997). "A key amino acid responsible for substrate selectivity of
monoamine oxidase A and B." Journal of Biological Chemistry 272(22): 14033-14036.

Tsumagari, N., R. Nakai, et al. (2004). "MPC1001, a new antitumor antibiotic produced by
Cladorrhinum sp." Journal of Antibiotics 57(8): 532-534.

van der Flier, A. and A. Sonnenberg (2001). "Function and interactions of integrins." Cell and
Tissue Research 305(3): 285-298.

Vaupel, P., O. Thews, et al. (2001). "Treatment resistance of solid tumors - Role of hypoxia and
anemia." Medical Oncology 18(4): 243-259.

Veeravagu, A., Z. Liu, et al. (2008). "Integrin alpha(v)beta(3)-Targeted Radioimmunotherapy of
Glioblastoma Multiforme." Clinical Cancer Research 14(22): 7330-7339.

Velling, T., M. Kusche-Gullberg, et al. (1999). "cDNA cloning and chromosomal localization of
human alpha(11) integrin - A collagen-binding, I domain-containing, beta(1-)associated integrin
alpha-chain present in muscle tissues." Journal of Biological Chemistry 274(36): 25735-25742.

Wan, X., N. Shen, et al. (2006). "CCl-779 inhibits rhabdomyosarcoma xenograft growth by an
antiangiogenic mechanism linked to the targeting of mTOR/Hif-1 alpha/VEGF signaling."
Neoplasia 8(5): 394-401.

Wang, D. Y., W. Liao, et al. (2005). "Binding properties of artificial a helices derived from a
hydrogen-bond surrogate: Application to Bcl-xL." Angewandte Chemie-International Edition
44(40): 6525-6529.

Wang, G. L., B. H. Jiang, et al. (1995). "HYPOXIA-INDUCIBLE FACTOR-1 IS A BASIC-HELIX-LOOP-
HELIX-PAS HETERODIMER REGULATED BY CELLULAR O-2 TENSION." Proceedings of the National
Academy of Sciences of the United States of America 92(12): 5510-5514.

Warburg, O. (1956). "ORIGIN OF CANCER CELLS." Science 123(3191): 309-314.
258

Waring, P. and J. Beaver (1996). "Gliotoxin and related epipolythiodioxopiperazines." General
Pharmacology 27(8): 1311-1316.

Waring, P., A. Sjaarda, et al. (1995). "GLIOTOXIN INACTIVATES ALCOHOL-DEHYDROGENASE BY
EITHER COVALENT MODIFICATION OR FREE-RADICAL DAMAGE MEDIATED BY REDOX CYCLING."
Biochemical Pharmacology 49(9): 1195-1201.

Weindling, R. and O. H. Emerson (1936). "The isolation of a toxic substance from the culture
filtrate of Trichoderma." Phytopathology 26(11): 1068-1070.

Wess, J. (1997). "G-protein-coupled receptors: Molecular mechanisms involved in receptor
activation and selectivity of G-protein recognition." Faseb Journal 11(5): 346-354.

Winstead, J. A. and R. J. Suhadolnik (1960). "BIOSYNTHESIS OF GLIOTOXIN .2. FURTHER STUDIES
ON THE INCORPORATION OF CARBON-14 AND TRITIUM-LABELED PRECURSORS." Journal of the
American Chemical Society 82(7): 1644-1647.

Wolf, A. M. D., R. C. Wender, et al. (2010). "American Cancer Society Guideline for the Early
Detection of Prostate Cancer Update 2010." Ca-a Cancer Journal for Clinicians 60(2): 70-98.

Woodcock, J. C., W. Henderson, et al. (2001). "Metal complexes of the mycotoxins sporidesmin
A and gliotoxin, investigated by electrospray ionisation mass spectrometry." Journal of Inorganic
Biochemistry 85(2-3): 187-199.

Wróbel, J. T. (1985). Chapter 2 Sulfur-Containing Alkaloids. The Alkaloids: Chemistry and
Pharmacology. B. Arnold, Academic Press. Volume 26: 53-87.

Wrobel, J. T. and K. Wojtasiewicz (1992). Chapter 4 Sulfur-Containing Alkaloids. The Alkaloids:
Chemistry and Pharmacology. A. C. Geoffrey, Academic Press. Volume 42: 249-297.

Wu, B., E. Y. T. Chien, et al. (2010). "Structures of the CXCR4 Chemokine GPCR with Small-
Molecule and Cyclic Peptide Antagonists." Science 330(6007): 1066-1071.

Xia, Y., H.-K. Choi, et al. (2012). "Recent advances in hypoxia-inducible factor (HIF)-1 inhibitors."
European Journal of Medicinal Chemistry 49: 24-40.

Xia, Y., Y. Jin, et al. (2011). "HIF-1 alpha inhibitors: Synthesis and biological evaluation of novel
moracin O and P analogues." European Journal of Medicinal Chemistry 46(6): 2386-2396.
259

Xiong, J. P., T. Stehle, et al. (2001). "Crystal structure of the extracellular segment of integrin
alpha V beta 3." Science 294(5541): 339-345.

Yamakuchi, M., C. D. Lotterman, et al. (2010). "P53-induced microRNA-107 inhibits HIF-1 and
tumor angiogenesis." Proceedings of the National Academy of Sciences of the United States of
America 107(14): 6334-6339.

Yanagihara, M., N. Sasaki-Takahashi, et al. (2005). "Leptosins isolated from marine fungus
Leptoshaeria species inhibit DNA topoisomerases I and/or II and induce apoptosis by
inactivation of Akt/protein kinase B." Cancer Science 96(11): 816-824.

Yang, X., C. Shi, et al. (2010). "Near IR Heptamethine Cyanine Dye-Mediated Cancer Imaging."
Clinical Cancer Research 16(10): 2833-2844.

Yarmoluk, S. M., V. B. Kovalska, et al. (2001). "Interaction of cyanine dyes with nucleic acids.
XXV. Influence of affinity-modifying groups in the structure of benzothiazol-4- 2,6-
dimethylpyridinium dyes on the spectral properties of the dyes in the presence of nucleic acids."
Spectrochimica Acta Part a-Molecular and Biomolecular Spectroscopy 57(7): 1533-1540.

Youdim, M. B. H., D. Edmondson, et al. (2006). "The therapeutic potential of monoamine
oxidase inhibitors." Nature Reviews Neuroscience 7(4): 295-309.

Younes, M., R. W. Brown, et al. (1995). "GLUT1 expression in human breast carcinoma:
Correlation with known prognostic markers." Anticancer Research 15(6B): 2895-2898.

Yu, A. Y., M. G. Frid, et al. (1998). "Temporal, spatial, and oxygen-regulated expression of
hypoxia-inducible factor-1 in the lung." American Journal of Physiology-Lung Cellular and
Molecular Physiology 275(4): L818-L826.

Yu, A. Y., L. A. Shimoda, et al. (1999). "Impaired physiological responses to chronic hypoxia in
mice partially deficient for hypoxia-inducible factor 1 alpha." Journal of Clinical Investigation
103(5): 691-696.

Zahl, P. A., F. S. Cooper, et al. (1940). "Some in vivo effects of localized nuclear disintegration
products on a transplantable mouse sarcoma." Proceedings of the National Academy of Sciences
of the United States of America 26: 589-598.

Zhang, E., C. Zhang, et al. (2011). "Newly developed strategies for multifunctional mitochondria-
targeted agents in cancer therapy." Drug Discovery Today 16(3-4): 140-146.
260

Zhang, H., P. Gao, et al. (2007). "HIF-1 inhibits mitochondrial biogenesis and cellular respiration
in VHL-deficient renal cell carcinoma by repression of C-MYC activity." Cancer Cell 11(5): 407-
420.

Zhang, S. X. L., D. Gozal, et al. (2003). "Hypoxia induces an autocrine-paracrine survival pathway
via platelet-derived growth factor (PDGF)-B/PDGF-beta receptor/phosphatidylinositol 3-
kinase/Akt signaling in RN46A neuronal cells." Faseb Journal 17(10): 1709-+.

Zhang, X. W., T. Kon, et al. (2004). "Enhancement of hypoxia-induced tumor cell death in vitro
and radiation therapy in vivo by use of small interfering RNA targeted to hypoxia-inducible
factor-1 alpha." Cancer Research 64(22): 8139-8142.

Zhong, H., K. Chiles, et al. (2000). "Modulation of hypoxia-inducible factor 1 alpha expression by
the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human
prostate cancer cells: Implications for tumor angiogenesis and therapeutics." Cancer Research
60(6): 1541-1545.

Zhong, H., A. M. De Marzo, et al. (1999). "Overexpression of hypoxia-inducible factor 1 alpha in
common human cancers and their metastases." Cancer Research 59(22): 5830-5835.

Zhou, W., M. Choi, et al. (2012). "HIF1 alpha induced switch from bivalent to exclusively
glycolytic metabolism during ESC-to-EpiSC/hESC transition." Embo Journal 31(9): 2103-2116.

261

APPENDIX A
Chapter 2

262

263
1
H NMR Spectrum of Compound 9 (300 MHz, CDCl
3
, TMS)
264

1
H NMR Spectrum of monoETP-1 (300 MHz, CDCl
3
, TMS)
265

13
C NMR Spectrum of monoETP-1 (CDCl
3
, TMS)

266

1
H NMR Spectrum of Compound 6 (300 MHz, CDCl
3
, TMS)

267

1
H NMR Spectrum of monoETP-2 (300 MHz, CDCl
3
, TMS)

268

1
H NMR Spectrum of compound 10 (300 MHz, CDCl
3
, TMS)
269

1
H NMR Spectrum of compound 11 (300 MHz, CDCl
3
, TMS)
270

1
H NMR Spectrum of monoETP-3 (300 MHz, CDCl
3
, TMS)
271

1
H NMR Spectrum of monoETP-3 (300 MHz, CDCl
3
, TMS)
272

1
H NMR Spectrum of Compound 12 (500 MHz, CDCl
3
, TMS)
273

1
H NMR Spectrum of Compound 5a/5b (500 MHz, CDCl
3
, TMS)
274

1
H NMR Spectrum of Compound monoETP-5 (500 MHz, CDCl
3
, TMS)

275

ESI-MS

Spectrum of Compound monoETP-5 (Thermo Finnigan LCQ). Main Peak
[M+Na]
+
m/z-449.9

276

Figure A.1 Cytotoxicity of Chetomin in MDA-231-HRE-Luc cells. Cells were treated
with compound for 24 h.

277

APPENDIX B
Chapter 3

278

Figure B.1 General scheme of synthesis of αB-HBS helices

279

Figure B.2 mRNA levels of VEGF. HeLa cells were maintained in 10% serum and
treated with HBS 6, HBS 7 and HBS 8 for 24 h. Hypoxia was induced by DFO.

280

Figure B.3 mRNA levels of Glut1. HeLa cells were maintained in 10% serum and treated
with HBS 6, HBS 7 and HBS 8 for 24 h. Hypoxia was induced by DFO.

281

Figure B.4 mRNA levels of LOX. HeLa cells were maintained in 10% serum and treated
with HBS 6, HBS 7 and HBS 8 for 24 h. Hypoxia was induced by DFO.

282

Figure B.5 mRNA levels of VEGF and Glut1. HeLa cells were maintained in 0.2%
serum and treated with HBS 6and chetomin (CTM) for 48 h. Hypoxia was induced by
anaerobic pouch (hypoxic bag).
283

Figure B.6 mRNA levels of LOX and CXCR4. HeLa cells were maintained in 0.2%
serum and treated with HBS 6 and chetomin (CTM) for 48 h. Hypoxia was induced by
anaerobic pouch (hypoxic bag) for LOX and by DFO for CXCR4.
284

APPENDIX C
Chapter 4

285
General Methods
All reagents and solvents were obtained from commercial sources and were used as
received unless otherwise stated. All reactions were conducted under a dry N
2

atmosphere with anhydrous solvent and flame dried glassware. Hygroscopic liquids were
transferred via a syringe and were introduced into reaction vessels through rubber septa.
Crude reaction mixtures were concentrated using a rotary evaporator at 30-150 mm Hg.
Flash chromatography was performed on silica gel (230-400 mesh) using reagent grade
solvents. Analytical thin-layer chromatography was performed on glass-backed, pre-
coated plates (0.25 ram, silica gel 60, F-254, EM Science). Solvent systems are reported
as v/v mixtures. Nuclear Magnetic Resonance (NMR) spectra were collected at ambient
temperature on Bruker (250 MHz, 500 MHz or 600 MHz) or Varian (300 MHz or
500MHz, broadband probe) instruments. Chemical shifts (δ) reported in ppm are relative
to the solvent signals for both
1
H and
13
C NMR spectra. The coupling constants (J) are
reported in Hertz (Hz). The following abbreviations are used: singlet (s), doublet (d),
triplet (t), quartet (q), doublet of doublets (dd), doublet of triplets (dt), broad (br). Mass
spectra were obtained from the Mass Spectrometry Laboratory in the Department of
Chemistry at the University of Arizona. Melting points were measured on Mel-Temp
capillary melting point apparatus and all were uncorrected.
Methyl 4-(2-aminoethoxy)benzoate (4). N-Boc-ethanolamine (1.6 g, 10 mmol), p-
hydroxymethylbenzoate (1.5 g, 10 mmol) and triphenylphosphine (3.3 g, 13 mmol) were
dissolved in anhydrous THF (20 mL). The resulting solution was cooled to 0
o
C and a
solution of DEAD in toluene (40%, 5.4 mL, 13 mmol) was added dropwise over the
course of 10 min. The reaction mixture was allowed to warm to room temperature and
286
stirred for 16 h, after which time the solvent was removed in vacuo. The resulting solid
was purified by flash chromatography (CH
2
Cl
2
:EtOAc 95:5) to yield a viscous oil (2.6 g)
shown by NMR to be a mixture of the desired coupling product 3 and p-
hydroxymethylbenzoate. The oil was subsequently dissolved in THF (25 mL) and HCl
(12M, 2.5 mL) then heated to 60
o
C for 18 h. Removal of the solvent in vacuo yielded a
crude solid which was washed several times with ether (50 mL) to yield 4 (1.3 g, 56%) as
a white solid. mp. 226-228
o
C.
1
H NMR (600 MHz, DMSO-d
6
, δ): 8.29 (3H, br s), 7.93
(2H, d, J=8 Hz), 7.08 (2H, d, J=8 Hz), 4.26 (2H, t, J=5 Hz), 3.81 (3H, s), 3.22 (2H, t, J=5
Hz).
13
C NMR (150 MHz, DMSO-d
6
, δ) 165.8, 161.6, 131.2, 122.4, 114.6, 64.53, 51.84,
38.11. HRMS (ESI) calculated for C
10
H
14
NO
3
[M+H]
+
: 196.0974, found: 196.0974.
Methyl 4-(2-(2-pyrimidinylamino)ethoxy)benzoate (5). DIEA (3.4 mL, 19 mmol) was
added to a suspension of 5 (1.5 g, 6.5 mmol) and 2-bromopyrimidine (1.2 g, 7.8 mmol) in
1,4-dioxane (25 mL). The reaction mixture was refluxed and stirred for 48 h, after which
it was allowed to cool to room temperature and the solvent was removed in vacuo. The
crude material was purified by flash chromatography (CH
2
Cl
2
:acetone 85:15) to yield 5
(1.1 g, 62%) as a light brown solid, mp. 106-107
o
C.
1
H NMR (500 MHz, CDCl
3
, δ):
8.29 (2H, d, J=5 Hz), 7.98 (2H, d, J=9 Hz), 6.92 (2H, d, J=9 Hz), 6.56 (1H, t, J=5 Hz),
5.59 (1H, br s), 4.20 (2H, t, J=5 Hz), 3.88 (5H, m).
13
C NMR (125 MHz, CDCl
3
, δ):
167.2, 162.8, 162.6, 158.5, 132.0, 123.3, 114.5, 111.5, 67.27, 52.32, 41.13. HRMS (ESI)
calculated for C
14
H
16
N
3
O
3
[M+H]
+
: 274.1192, found: 274.1191.
4-(2-(2-pyrimidinylamino)ethoxy)benzoic acid (6). NaOH (1M, 10 mL) was added to
a solution of 5 (1.0 g, 3.7 mmol) in EtOH (10 mL) and the resulting mixture was heated
to 60
o
C for 2 h, after which it was allowed to cool to room temperature. The pH of the
287
solution was adjusted to approximately 4 by dropwise addition of HCl (1M), resulting in
the formation of a yellow precipitate which was isolated, washed several times with water
then dried in vacuo to yield 6 (650 mg, 68%) as a cotton-like yellow solid. mp. 225-227
o
C.
1
H NMR (600 MHz, DMSO-d
6
, δ): 12.58 (1H, br s), 8.28 (2H, d, J=5 Hz), 7.86 (2H,
d, J=9 Hz), 7.28 (1H, t, J=5 Hz), 7.02 (2H, d, J=9 Hz), 6.58 (1H, t, J=5 Hz), 4.16 (2H, t,
J=6 Hz), 3.65 (2H, q, J=6 Hz).
13
C NMR (150 MHz, DMSO-d
6
, δ): 166.9, 162.2, 162.1,
160.0, 131.3, 123.0, 114.2, 110.3, 66.18. HRMS (ESI) calculated for C
13
H
14
N
3
O
3

[M+H]
+
: 260.1035, found: 260.1034.
4-(2-(2-pyrimidinylamino)ethoxy)benzoic acid conjugate (8). HBTU (700 mg, 1.80
mmol) was added to a solution of 6 (425 mg, 1.63 mmol) and DIEA (1.06 mL, 6.08
mmol) in DMF (10 mL). The resulting mixture was stirred for 2 min then added
dropwise to a solution of (N-Boc- -amino)-L-alanine benzyl ester 7 (956 mg, 3.27
mmol) and DIEA (1.06 mL, 6.08 mmol) in DMF (10 mL). The reaction medium was
stirred at room temperature for 18 h after which time the solvent was removed in vacuo.
The residue was then suspended in a solution of HCl in EtOH (5%, 25 mL) and stirred at
40
o
C for 18 h. After cooling, the mixture was diluted with toluene (50 mL) and the
solvent removed under vacuo. The crude material was taken in a saturated solution of
K
2
CO
3
(50 mL) and extracted with CH
2
Cl
2
(3 x 50 mL). The combined organic phases
were dried over MgSO
4
, concentrated, and the crude product was carried through to the
next step. An analytical sample of 8 was obtained as a white solid (425 mg, 60%) after
flash chromatography (CH
2
Cl
2
:MeOH 90:10). mp. 175-177
o
C.
1
H NMR (600 MHz,
DMSO-d
6
, δ): 8.38 (1H, br t), 8.30 (2H, d, J=5 Hz), 7.36-7.28 (7H, m), 7.01 (2H, d, J=9
Hz), 6.61 (1H, t, J=5 Hz), 5.06 (2H, s), 4.27-4.25 (1H, m), 4.16 (2H, t, J=6 Hz), 3.66-
288
3.53 (4H, m), 1.36 (9H, s). HRMS (ESI) calculated for C
28
H
34
N
5
O
6
[M+H]
+
: 536.2509,
found: 536.2487.
Synthesis of protected precursor 10. TFA (2 mL) was added to a solution of 8 (100
mg, 0.231 mmol) in CH
2
Cl
2
(2 mL) and the resulting mixture was stirred for 30 min at
room temperature. Following removal of the solvent in vacuo, the residue was dissolved
in CH
2
Cl
2
(1 mL), cooled to 0
o
C and DMAP (57 mg, 0.462 mmol) was added, followed
dropwise by a freshly prepared mixture of triphosgene (55 mg, 0.185 mmol), N-Boc-
taurine tetrabutylammonium salt (215 mg, 0.462 mmol) and
CH
2
Cl
2
(2.0 mL) stirred for 30 min to afford 9. The combined solutions were allowed to
warm to room temperature and stirred for 18 h, after which the solvent was removed in
vacuo. The crude material was purified by flash chromatography (CH
2
Cl
2
:acetone
65:35), yielding fully protected intermediate 10 (75mg, 51%) as a white solid. mp. 146-
149
o
C.
1
H NMR (600 MHz, CDCl
3
, δ): 8.33 (2H, br s), 7.62 (2H, d, J=8 Hz), 7.39-7.34
(5H, m), 6.90 (2H, d, J=11 Hz), 6.65 (1H, t, J=5 Hz), 6.64 (1H, br t), 6.12 (1H, d, J=7
Hz), 5.26 (1H, d, J=12 Hz), 5.20-5.17 (2H, m), 4.43-4.40 (1H, m), 4.20 (2H, t, J=6 Hz),
3.93-3.83 (4H, m), 3.65-3.50 (2H, m), 3.22-3.17 (2H, m), 1.43 (9H, s).
13
C NMR (150
MHz, CDCl
3
, δ): 170.2, 168.1, 161.5, 155.9, 134.8, 129.0, 128.8, 128.7, 128.6, 126.0,
114.4, 110.3, 79.93, 68.13, 66.35, 56.42, 53.57, 42.39, 40.71, 35.41, 28.32. HRMS (ESI)
calculated for C
30
H
39
N
6
O
8
S [M+H]
+
: 643.2550, found: 643.2551.
Synthesis of probe 2. In a small vial, 10 (15 mg, 0.024 mmol) was dissolved in MeOH
(2 mL), followed by the addition of Pd black (10 mg). The vial was purged with H
2
, and
the solution was stirred under an H
2
atmosphere for 3 h. Once complete, the solids were
filtered off through a plug of cotton. The filtrate was collected and the solvent removed
289
under reduced pressure, followed by addition of CH
2
Cl
2
(2 mL) and TFA (2 mL). The
resulting solution was stirred for 30 min at RT. Then, the solvent was removed under
reduced pressure to yield 1. The residue was dissolved in DMF (1 mL), followed by the
11 (14 mg, 0.048 mmol). After
stirring for 1 h at RT, the solvent was removed under reduced pressure. The residue was
redissolved in MeOH (3 mL) and purified by semi-preparative HPLC to give 2 (10 mg,
58%) as a viscous orange oil.
1
H NMR (600 MHz, CD
3
OD) δ 7.84 (2H, d, J=5 Hz), 7.75
(1H, d, J=8 Hz), 7.52-7.46 (2H, m), 7.30 (1H, t, J=8 Hz), 7.20 (2H, s), 7.00 (2H, d, J=5
Hz), 4.81-4.72 (1H, m), 4.69-4.60 (1H, m), 4.50-4.45 (1H, m), 4.18-4.09 (2H, t, J=9 Hz),
3.95-3.88 (1H, m), 3.79-3.63 (4H, m), 3.53-3.48 (4H, m), 3.40-3.32 (4H, m), 2.70 (6H,
s), 1.96-1.93 (2H, m). UV-Vis (MeOH): 446, 253, 228, and 205 nm. HRMS (ESI)
calculated for C
34
H
42
N
7
O
6
S
2
[M+H]
+
: 708.2638, found: 708.2623.

290

1
H NMR spectrum of compound 4.
291

13
C NMR spectrum of compound 4.
292
1
H NMR spectrum of compound 5.
293

13
C NMR spectrum of compound 5.
294

1
H NMR spectrum of compound 6.
295

13
C NMR spectrum of compound 6.
296

1
H NMR spectrum of compound 8.
297

1
H NMR spectrum of compound 10.
298

13
C NMR spectrum of compound 10.

299

1
H NMR spectrum of compound ABL-1.

Abstract (if available)

Abstract Transcription factors are the key regulators of cancer gene expression and play a critical role in every aspect of tumorigenesis. Despite their relevance in tumor progression, the shallow binding surfaces and absence of the dependence on enzymatic activity makes them challenging therapeutic targets. However, now the fields of chemical biology, genetics, cancer biology, and biotechnology have evolved to a point where targeting transcription factors has become tractable. Direct modulation of the activity of transcription factors has become a promising, broadly applicable strategy in drug discovery and in biology, because a limited number of oncogenic transcription factors, as compared to signaling kinases are involved in progression of certain disease states, such as cancer, which elects transcription factors as cogent targets. A second broad family of targets is cell surface receptors which, in addition to transcription factors, either independently or in collaboration with the growth factor receptors, modulate downstream signals. The work described in this Dissertation is focused on chemical strategies in targeting oncogenic transcription factors and cell adhesion receptors involved in cancer progression, as a prerequisite for the development of novel anticancer therapeutics. ❧ Chronic hypoxia is a hallmark of solid tumors and is associated with aggressiveness and rapid progression of the disease. In tumors under hypoxia, a specific microenvironment is created, which is very different from that of normal tissues. Under these conditions, activation and stabilization of the α subunit of a transcription factor, termed hypoxia-inducible factor 1 (HIF-1α) is achieved, and its interaction with the coactivator p300/CBP elevates the expression of a number of genes involved in angiogenesis, invasion, altered energy metabolism and other proliferative mechanisms that promote tumor growth. The abundance of the expressed HIF-1α in most solid tumors makes it an attractive therapeutic target. We report the two complementary strategies in targeting of the interface of transcription factor HIF-1α and its coactivator p300/CBP: allosteric and orthosteric. In Chapter 2 of this dissertation, an allosteric approach for targeting of HIF-1α is described, which is accomplished through a novel class of an epipolythiodiketopiperazine (ETP) transcriptional antagonists that disrupt hypoxia-inducible transcription. In Chapter 3, we describe the design of orthosteric transcriptional antagonists termed hydrogen bond surrogates (HBSs) - stable α-helices that mimic key interacting domain of HIF-1α that interacts with p300/CBP coactivator. Through the HBS approach we explore the structural roles of the two critical α-helices of HIF-1α C-terminal transactivation domain (C-TAD) that are responsible for the recognition of p300/CBP. The stabilization of the secondary structure in HBS via an α-helical motif results in a short, stable α-helix mimics that disrupt the interaction between the HIF-1α CTAD and coactivator complex, resulting in a downregulation of hypoxia-inducible genes, such as VEGF, LOX, Glut1, and CXCR4. Through the use of the orthosteric and allosteric approaches, we demonstrated the potential of designed small molecules and protein secondary structure mimetics to become new research tools for disruption of protein-protein interfaces, regulation of transcription and, ultimately, as leads in the discovery of novel therapeutics. ❧ In cancers, the elevated levels of certain integrin receptors have been linked to tumor progression and poor overall prognosis for the patients. Of these, the αᵥβ₃-receptors are overexpressed on activated endothelial cells undergoing angiogenesis as well as on proliferating tumor cells, but show only low levels of expression on quiescent cells. This makes αᵥβ₃ integrin an attractive molecular target for both diagnosis and anticancer therapy. In the Chapter 4 of this Dissertation, we report the design and biological evaluation of αᵥβ₃ specific conjugates comprised of high affinity, αᵥβ₃-selective integrin ligand, a carborane moiety, and a fluorescent probe that exhibits fluorescence enhancement upon binding to its target. Our in vitro as well as in vivo studies indicate selectivity toward tumorigenic cells and active uptake of the designed bifunctional and trifunctional ligands. ❧ Monoamine oxidase A (MAOA) is a mitochondrial membrane-bound enzyme that catalyzes the oxidative deamination of dietary amines to corresponding aldehydes. This process produces H₂O₂, a major source of reactive oxygen species, which predisposes cancer cells to DNA damage and promotes tumor initiation and progression. It was demonstrated that knock-out and pharmacological inhibition of MAOA in prostate cancer cells slowed down cancer progression. In the Chapter 5 of this Dissertation, we explore the connection between the activity of MAOA, the levels of H₂O₂ and its connection with the elevated levels of HIF-1α and the expression of its downstream genes. Clorgyline, a selective MAO A inhibitor used as an anti-depressant, and its conjugate NIR-clorgyline, significantly reduced tumor growth in animal tumor xenograft models. The presence of the NIR dye moiety renders selective uptake of NIR-clorgyline conjugate to cancer cells and increases its in vivo efficacy. The preliminary results from this study suggest that MAOA could become a novel target obligatory to human prostate cancer growth and metastasis. Through the induction of H₂O₂, MAOA could elevate the levels of HIF-1α resulting in a rapid vascularization, invasiveness of the tumor and a poor prognosis for prostate cancer patients. Targeting MAOA in prostate cancer holds the promise of achieving combination therapeutic effects by its synergistic inhibition of multiple downstream tumor growth promoting factors (e.g. H₂O₂ and HIF-1α), which could help to more effectively treat prostate cancer and to overcome the accompanying chemo-resistance.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands

PDF

Small molecule modulators of HIF1α signaling

PDF

Modulation of Runx proteins by steroid hormone receptors

PDF

Mechanism of repression of RNA polymerase III-dependent transcription under hypoxia

PDF

Characterization of a fragment in secreted Hsp90α with potential therapeutic benefits in wound healing and cancer

PDF

Monoamine oxidase A inhibitors and androgen receptor antagonists regulate mitochondrial function in prostate cancer cells

PDF

Regulation of mitochondrial bioenergetics via PTEN (phosphatase and tensin homolog deleted on chromosome 10)/estrogen-related receptor alpha (ERRα) signaling

PDF

Mechanism of action of rapamycin and its applications in aging, cancer therapy and metabolism

PDF

RNA polymerase III-dependent transcription is repressed under prolonged hypoxic conditions

PDF

Transcriptional regulation of IFN-γ and PlGF in response to Epo and VEGF in erythroid cells

PDF

Genetic engineering of non-reducing polyketide synthases by domain swap strategy in Aspergillus nidulans

PDF

Physical activity optimizes circuit-specific cellular metabolism in neuroplasticity: a role for hypoxia-inducible factor-1 and its downstream targets

PDF

Characteristics of hydrogen peroxide inducible clone-5 and its potential role as a nuclear receptor coactivator

PDF

Estrogen receptor-β characterization in breast cancer: development of a reliable assay for measuring expression

PDF

Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy

PDF

Identification of small-molecules targeting CXCR2 function and signaling

PDF

The modulation of dynamin and receptor endocytosis machinery using elastin-like polypeptides

PDF

Enhancing the specificity and cytotoxicity of chimeric antigen receptor Natural Killer cells

PDF

Repression of RNA polymerase III-dependent transcription by the tumor suppressors p53 and PTEN

PDF

Using chemical biology approaches to investigate the consequences of protein concentration and activity in cancer cells

Asset Metadata

Creator Kushal, Swati (author)

Core Title Synthetic small molecules and protein secondary structure mimetics as modulators of hypoxia-inducible transcription and integrin receptors function

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 01/08/2014

Defense Date 04/03/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag epidithodiketopiperazines,HIF-1,hydrogen bond surrogates,hypoxia,integrin receptors,OAI-PMH Harvest,transcription factors

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Olenyuk, Bogdan Z. (committee chair), Pratt, Matthew R. (committee member), Wang, Clay C.C. (committee member)

Creator Email kushal@usc.edu,swatimishra27@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-286640

Unique identifier UC11294506

Identifier etd-KushalSwat-1746.pdf (filename),usctheses-c3-286640 (legacy record id)

Legacy Identifier etd-KushalSwat-1746.pdf

Dmrecord 286640

Document Type Dissertation

Format application/pdf (imt)

Rights Kushal, Swati

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA