Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands
(USC Thesis Other)
Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MODULATION OF TRANSCRIPTION AND RECEPTOR
FUNCTION WITH SYNTHETIC SMALL MOLECULES
AND MULTI-FUNCTIONAL INTEGRIN LIGANDS
by
Ramin Dubey
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)
December 2012
ii
Dedication
This dissertation is dedicated to all my family and friends.
iii
Acknowledgments
I would like to thank my Ph.D. advisor Professor Bogdan Olenyuk for his encouragement
and support throughout my tenure in his lab. I am truly grateful for his intellectual input
and guidance in all my research projects. With his encouragement I got opportunity to
work on multidisciplinary projects that allowed me develop skills in chemistry as well as
translational science. I would also like to thank all my former and present group members
for their positive attitude and support in my research. I would like to acknowledge Dr.
Nathan Polaske for help with organocatalytic sulfenylation project and Swati Kushal for
collaboration on integrin ligands project. Special thanks are due to Dr. Csaba Laszlo and
Dr. Lajos Szabo for help with synthetic bis-ETP project. I would like to thank our
collaborators Dr. Schnitzer’s lab, PRISM, San Diego for help with intravital microscopy
and Dr. Arora’s group, NYU for collaboration in HIF regulation projects.
I would like to thank my dissertation committee members Prof. Julio Camarero, Prof.
Matthew Pratt and my graduate committee members Prof. Clay Wang and Prof. Curtis
Okamoto. I would like to take this opportunity to thank Vandana Lamba, my sister and
brother-in-law Sonal and Manish Sharma, my parents Dr. Mohan and Manorama Dubey,
and rest of my family for their help and moral support.
iv
Table of Contents
Dedication ........................................................................................................................... ii
Acknowledgments.............................................................................................................. iii
Table of Contents ............................................................................................................... iv
List of Tables ..................................................................................................................... ix
List of Figures ..................................................................................................................... x
Abstract ........................................................................................................................... xvii
Chapter 1: Introduction....................................................................................................... 1
1.1 The role of hypoxia in cancer biology ................................................................. 2
1.2 Chetomin: small molecule inhibitor of HIF-1α- p300/CBP complex .................. 9
1.3 Biologically relevant epidithiodiketopiperazines (ETPs) .................................. 11
1.4 Hypoxia-inducible transcription of VEGF, Met and other downstream genes . 17
1.4.1 The role of VEGF in tumor progression and angiogenesis. ............................. 17
1.4.2 Role of c-Met in cancers and its regulation by HIF-1α. ............................. 21
1.4.3 Other HIF-1α inducible genes..................................................................... 24
1.5 Targeting alpha-V beta-3 (α
v
β
3
) integrin receptors ............................................ 28
Chapter 2: Development of facile synthetic strategies en route to compounds bearing
ETPs and indoles............................................................................................................... 35
2.1 Introduction to synthetic strategies for construction of ETP rings.......................... 36
2.2 Brief overview of important developments in the field of sulfenylation and
organocatalytic sulfenylation ........................................................................................ 40
2.2.1 Approaches towards asymmetrical sulfenylation. ............................................ 40
2.2.2 Organocatalytic sulfenylation reactions. .......................................................... 43
2.3 Organocatalytic sulfenylation of diketopiperazines ........................................... 51
2.3.1 Initial efforts of organocatalytic sulfenylation of diester-diketopiperazines. ... 51
2.3.2 Synthesis of diketopiperazine substrates. ......................................................... 53
2.3.3 Organocatalytic sulfenylation of monoester-diketopiperazines. ...................... 59
2.3.4 Novel sulfenylating reagent 1-phenylsulfanyl[1,2,4]triazole and its
organocatalytic sulfenylation of DKPs. ..................................................................... 63
v
2.3.5 X-ray crystal structure of tert-Butyl-1,4-dimethyl-3,6-dioxo-2-(phenylthio)
piperazine-2-carboxylate (27). ................................................................................... 66
2.3.6 Enantioselectivity of organocatalytic sulfenylation. ........................................ 67
2.3.7 Proposed Mechanism for Organocatalysis of DKPs leading to sulfenylation. . 71
2.4 Organocatalytic coupling of diketopiperazines with indoles .................................. 72
2.4.1 Gramine as an alkylation reagent. .................................................................... 72
2.4.2 Use of quinine for alkylation of diketopiperazines with gramine. ................... 76
2.4.3 Vinylogous iminium ion intermediates generated from arylsulfonyl indoles add
to aldehydes. .............................................................................................................. 79
2.4.4 Novel strategy of organocatalytic coupling of diketopiperazines with indoles.81
2.5 Conclusions and future directions. ...................................................................... 86
Chapter 3: Design of Dimeric ETPs by Biomimetic Rational Approach and Investigation
of their in vitro Biophysical Properties ............................................................................. 89
3.1 Structural basis of p300 - HIF-1α binding. ............................................................. 90
3.2.1 Physical basis of disruption of p300 and HIF-1α complex by chetomin. ........ 91
3.2.2 Basis for design of novel bis-ETPs................................................................... 92
3.3.1 Expression and purification of p300-CH1 domain for biophysical assays. ..... 97
3.3.2 Expression of
15
N labeled GST-p300-CH1 fusion protein. .............................. 99
3.3.3 Cleavage of p300-CH1 domain from GST. .................................................... 100
3.4.1 Binding study of ETPs toward CH1 domain of p300 with surface plasmon
resonance (SPR). ..................................................................................................... 100
3.4.2 SPR analysis of chetomin binding to CH1 domain of p300. .......................... 101
3.4.3 Direct binding of ETP-3 and ETP-5 to CH1 domain of p300 in SPR assays. 103
3.4.4 SPR analysis of the binding of control DKP. ................................................. 105
3.4.5 Binding affinity of two stereoisomers of ETP-3 analyzed by SPR. ............... 108
3.5.1 Fluorescence polarization assays to study disruption of TAD-C HIF-1α and
CH1-p300 complex by ETPs. .................................................................................. 110
3.5.2 Saturation binding fluorescence polarization assays for GST-CH1-p300 and
CH1-p300 with TAD-C HIF-1α. ............................................................................. 111
3.5.3 Fluorescence polarization assays to monitor the disruption of TAD-C HIF-1α
and CH1 p300 binding by chetomin. ....................................................................... 114
3.5.4 Fluorescence polarization assays with ETP-5 to monitor the disruption of
TAD-C HIF-1α and CH1 p300 binding................................................................... 115
vi
3.5.5
15
N NMR of CH1-p300 domain and CH1-p300 complex with HIF-1α TAD-C:
................................................................................................................................. 116
3.6 Conclusion. ............................................................................................................ 117
Chapter 4: Transcriptional Regulation of HIF-1α Inducible Genes with Designed
Dimeric Epidithiodiketopiperazines. .............................................................................. 119
4.1 Regulation of transcription of hypoxia inducible genes........................................ 120
4.2 Inhibition of HIF-1α inducible gene expression by chetomin and ETP-3. ........... 121
4.3 Evaluation of ETP-5 by HIF inducible luciferase reporter assay.......................... 124
4.4 Cytotoxicity of ETP-5 in MCF7 breast cancer cell line and A549 lung epithelial
adenocarcinoma cell line. ............................................................................................ 125
4.5 Transcriptional regulation of HIF-1α genes in MCF7 cells treated with ETPs. ... 129
4.6 Transcriptional changes in MDA-MB-231 cell line upon treatment with ETP-5. 135
4.7 Modulation of HIF-1α inducible genes transcription levels with ETP-5 in A549
lung adenocarcinoma line............................................................................................ 136
4.8 HIF-1α protein levels are unchanged in A549 cells treated with ETP-5 under
hypoxia and normoxia. ................................................................................................ 145
4.9 VEGF and c-Met protein levels in MCF7 and MDA-MB-231 cell lines upon
treatment with ETPs. ................................................................................................... 146
4.10 Gene expression profiling and microarray analysis. ........................................... 148
4.11 In vivo study of the efficacy of ETP-5 in mouse tumor xenografts model using
intravital microscopy. .................................................................................................. 153
4.12 Conclusion ........................................................................................................... 155
Chapter 5: Integrin Ligand – Boron Cluster Dendrimer Conjugates for Imaging and
Targeted Delivery of Boron to Tumors .......................................................................... 156
5.1 Introduction ........................................................................................................... 157
5.2 Targeted delivery of antitumor cargo to α
v
β
3
integrin receptors. .......................... 159
5.3 Applications of boron-rich compounds in boron MRI and BNCT. ...................... 160
5.4 Building blocks for targeted boron delivery. ........................................................ 165
5.5 Structure and photophysical properties of α
v
β
3
integrin ligand conjugates. ......... 167
5.5.1 DIL-1: α
v
β
3
integrin ligand – cyan dye conjugate. ......................................... 167
5.5.2 DILB-2: α
v
β
3
integrin ligand – cyan dye – carborane dendritic wedge
conjugate. ................................................................................................................. 169
5.5.3 DB-3: Cyan 40 dye – carborane dendritic wedge conjugate. ......................... 171
vii
5.6 Confocal microscopy imaging in carcinoma cell lines expressing α
v
β
3
integrin
receptors. ..................................................................................................................... 174
5.6.1 Confocal imaging with DIL-1 in WM115 cells. ............................................. 174
5.6.2 Confocal microscopy imaging with DILB-2 in WM115 and MCF7 cells. .... 176
5.6.3 Confocal imaging with DB-3 in WM115 cells. .............................................. 179
5.7 Flow cytometry in WM115 and MCF7 cells. ....................................................... 179
5.8 Intravital microscopy imaging in murine subcutaneous and in ectopic-orthotopic
tumor xenograft models with DILB-1 and DB-3. ....................................................... 180
5.9 Conclusion. ............................................................................................................ 182
Chapter 6: Experimental Sections .................................................................................. 183
6.1 Introduction to Experimental Section. .................................................................. 184
6.2 Experimental Section for Chapter 2 ...................................................................... 184
6.2.1 General Methods............................................................................................. 184
6.2.2 Synthetic procedures and analytical data for diketopiperazines. .................... 185
6.2.3 Synthesis procedures and analytical data for sulfenylated compounds. ......... 195
6.2.4 Synthesis procedures and analytical data for compounds prepared by alkylation
ofdiketopiperazines with indoles. ............................................................................ 201
6.3 Experimental Section for Chapter 3 ...................................................................... 206
6.3.1 Protein Expression. ......................................................................................... 206
6.3.2
15
N-labeled CH1-p300-GST expression. ........................................................ 208
6.3.3 Thrombin-mediated CH1-p300-GST fusion protein cleavage. ...................... 208
6.3.4 Immobilization of GST-CH1-p300 on SPR chip. .......................................... 209
6.3.5 SPR binding assays of ETPs to immobilized GST-CH1-p300. ...................... 210
6.3.6 Determination of K
D
from the saturation binding curves between CH1-p300-
GST/CH1-p300 and HIF-1α-Fl in fluorescence polarization competition assays. .. 210
6.3.7 Fluorescence polarization competition assays to measure the disruption of
CH1-p300-GST/CH1-p300 and HIF-1α-Fl binding with bis-ETPs. ....................... 211
6.4 Experimental Section for Chapter 4 ...................................................................... 212
6.4.1 Cell lines. ........................................................................................................ 212
6.4.2 Cell culture. .................................................................................................... 212
6.4.3 Luciferase assays. ........................................................................................... 212
6.4.4 Cell viability assay.......................................................................................... 213
viii
6.4.5 Isolation of mRNA. ........................................................................................ 214
6.4.6 Analysis of gene expression. .......................................................................... 215
6.4.7 Western blot analysis of VEGF and c-Met protein levels. ............................. 216
6.4.8 Western blot analysis of HIF-1α levels. ......................................................... 216
6.4.9 Animal use. ..................................................................................................... 217
6.4.10 Fluorescent tumor cell lines. ......................................................................... 217
6.4.11 Mouse xenograft tumor models. ................................................................... 218
6.4.12 Tumor Growth. ............................................................................................. 219
6.5 Experimental Section for Chapter 5 ...................................................................... 219
6.5.1 Absorbance spectrum for DIL-1, DILB-2 and DB-3. .................................... 219
6.5.2 Fluorescence spectra for DIL-1, DILB-2 and DB-3. ...................................... 220
6.5.3 Confocal Microscopy with DIL-1, DILB-2 and DB-3. .................................. 220
6.5.4 Intravital microscopy with DILB-2 and DB-3. .............................................. 220
Bibliography ................................................................................................................... 222
Appendix A ..................................................................................................................... 242
Appendix B ..................................................................................................................... 299
Appendix C ..................................................................................................................... 309
Appendix D ..................................................................................................................... 310
Appendix E ..................................................................................................................... 320
ix
List of Tables
Table 2.1 Organocatalytic α-sulfenylation of N1 substituted monoester DKPs with 1-
benzylsulfanyl[1,2,4]triazole 22a and cinchona alkaloids 9a–d as organocatalysts.. ...... 61
Table 2.2 Optimization of the reaction conditions for sulfenylation of 3-(tert-
butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione with 1-phenylsulfanyl[1,2,4]triazole
(22b) .................................................................................................................................. 64
Table 2.3 Scope of α-sulfenylation of trisubstituted piperazine-2,5-diones with reagent
22b and 10 mol % of organocatalysts 9a–c.. .................................................................... 65
Table 2.4 Optimization studies of sulfenylation of DKPs with different organocatalysts
and sulfenylating reagents 22a and 22b............................................................................ 69
Table 2.5 Organocatalytic enantioselective α-sulfenylation of substituted piperazine-2,5-
diones with sulfenylating reagents 22a and 22b catalyzed by quinine 9c (10 mol %) ..... 70
Table 2.6 Effect of substrate and temperature on quinine-promoted alkylation of
substituted diketopiperazines with gramine.. .................................................................... 78
Table 2.7 Optimization and scope of α-alkylation of diketopiperazines with arylsulfonyl
indoles catalyzed by cinchona alkaloids.. ......................................................................... 84
Table 2.8 Optimization of catalysts and solvents in coupling of indole derivative to 2-
oxocyclo-pentanecarboxylate with β-ketoester................................................................. 86
Table 3.1 Amino acid sequence and corresponding DNA sequence for CH1-p300 domain.
…………………………………………………………………………………98
Table 3.2 Amino acid sequence of the Fusion GST-p300-CH1 domain………………..98
Table 4.1 List of important for cancer progression HIF-1α inducible genes that are
downregulated under hypoxic induction with DFO (300 µM) by treatment with ETP-5
(400 nM).. ……………………………………………………………………...151
Table 4.2 List of Solute Carrier (SLC) family genes that are down-regulated in MCF7
cells under hypoxic induction with DFO (300 µM) upon treatment with ETP-5 (400
nM)……………………………………………………………………………………..152
x
List of Figures
Figure 1.1 Prodrugs that selectively target hypoxic cells and can be used as adjuvant in
radiation therapy to target otherwise resistant hypoxic tumor cells. .................................. 5
Figure 1.2 Schematic representation of HIF-1α signaling pathway and its dependence on
normoxia and hypoxia......................................................................................................... 6
Figure 1.3 Domain map of HIF-1α and p300. ................................................................... 8
Figure 1.4 Structure of chetomin, having ETPs (epidithiodiketopiperazines) within its
north and south fragment and a substituted cyclotryptophan motif. ................................... 9
Figure 1.5 Epidithiodiketopiperazine (ETP) ring is a core fragment of the natural
products called sporidesmins. ........................................................................................... 11
Figure 1.6 Representative structures of bis-ETPs, containing an ETP ring within a north
and a south fragment. ........................................................................................................ 12
Figure 1.7 Structures of gliotoxin, sporidesmin A and hyalodendrin. ............................. 13
Figure 1.8 Structure of ETP-3, a synthetic bis-ETP. The two ETP rings are connected by
an aromatic linker. ............................................................................................................ 14
Figure 1.9 Structures of the ETP bearing synthetic compounds SC-5 to SC-9 developed
by Schofield and coworkers. ............................................................................................. 15
Figure 1.10 Structure of compounds SC-10, SC-11 and SC-12 developed by Schofield &
coworkers. ......................................................................................................................... 15
Figure 1.11 Structure of ditryptophenaline, WIN 64821, WIN 64745 and Leptosin S. .. 16
Figure 1.12 Interaction between VEGF dimers with different combinations of the three
VEGFRs dimers. ............................................................................................................... 19
Figure 1.13 Schematic representation of VEGF pathway showing VEGFR2 and some of
the important cascades that emanate from it. .................................................................... 20
Figure 1.14 Schematic diagram of Met receptor having β and α chain in the extracellular
domain and intracellular tyrosine kinase domain. ............................................................ 22
Figure 1.15 Met dimer activation by HGF. ...................................................................... 23
Figure 1.16 Integrin-mediated signaling pathway ........................................................... 29
Figure 1.17 Structures of three RGD based peptides that have been designed by different
Pharma Companies. .......................................................................................................... 32
Figure 1.18 Non-peptidic small molecule antagonists that bind integrins. ...................... 33
Figure 2.1 Classic strategies for synthesis of the ETP core. ............................................ 37
Figure 2.2 Total synthesis of Sporidesmin A by Kishi and coworkers. ........................... 38
xi
Figure 2.3 Total synthesis of dideoxyverticillin by Movassaghi and coworkers............. 39
Figure 2.4 Chetomin structure and unique structural features. ........................................ 40
Figure 2.5 Oxazolidinone approach for asymmetric sulfur addition. .............................. 41
Figure 2.6 Thiazolidine-derived reagent for sulfenylation of ketone derivatives. ........... 42
Figure 2.7 An optically active hydrazine is condensed with a ketone followed by
stereoselective sulfenylation with alkyl disulfide. ............................................................ 42
Figure 2.8 Use of optically active ketone is condensed with α-amino ester followed by
sulfenylation. ..................................................................................................................... 43
Figure 2.9 Pyrrolidine-based catalyst for sulfenylation of ketones. ................................ 43
Figure 2.10 Different pyrrolidine based organocatalysts used for catalytic sulfenylation.
........................................................................................................................................... 44
Figure 2.11 S-benzyl based sulfenylating reagents. ......................................................... 45
Figure 2.12 Model reaction for the catalyst optimization study. ..................................... 45
Figure 2.13 Sulfenylation strategy with 1-benzylsulfanyl-1,2,4-triazole. ....................... 46
Figure 2.14 Commercially available cinchona alkaloids used as organocatalysts. ......... 47
Figure 2.15 β-Ketoester as a model substrate sulfenylated by 1-
benzylsulfanyl[1,2,4]triazole with a cinchona alkaloid as a catalyst. ............................... 48
Figure 2.16 Example of cyclic β-ketoesters and lactams that were substrates for
organocatalytic sulfenylation. ........................................................................................... 48
Figure 2.17 Ti[TADDOL(ato)]-complex used as a catalyst in sulfenylation reactions with
β-ketoesters using phenylsulfenyl chloride as an electrophilic sulfenylating reagent. ..... 49
Figure 2.18 Sulfenylation of β-keto phosphonates using pyrrolidine-based organocatalyst
and N-(phenyl-thio)phthalimide as sulfenylating reagent................................................. 50
Figure 2.19 Organocatalytic sulfenylation of diethylester-DKP substrate using
(DHQD)2PYR as catalyst. ................................................................................................ 51
Figure 2.20 Schematic representation of the mechanism leading to anti-additions of
sulfur groups on the DKP ring. ......................................................................................... 52
Figure 2.21 Piperazine-2,5-dione is a suitable starting material for synthesis of
substituted DKPs. .............................................................................................................. 53
Figure 2.22 Synthesis of 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione............ 54
Figure 2.23 Two-step novel approach for synthesis of 3-(ethoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione ........................................................................................... 54
Figure 2.24 Synthesis of compound 14. ........................................................................... 55
xii
Figure 2.25 Acylation of 14 with acetyl chloride, benzoyl chloride and tosyl chloride
yielding DKPs 15, 16 and 17, respectively. ...................................................................... 57
Figure 2.26 Synthesis of 3-(tert-butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione from
sarcosine anhydride. .......................................................................................................... 58
Figure 2.27 Synthesis of 3-(ethoxycarbonyl)-1,4-diethylpiperazine-2,5-dione and 3-
(ethoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione. ....................................................... 58
Figure 2.28 Synthesis of 3-(tert-butoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione. .... 59
Figure 2.29 Organocatalytic sulfenylation of monoester-DKP 3-(ethoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione with 1-benzylsulfanyl[1,2,4]triazole. ............................... 59
Figure 2.30 Sulfenylation of 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione with 1-
benzylsulfanyl[1,2,4]triazole ............................................................................................ 60
Figure 2.31 Cinchona alkaloids used as organocatalysts in this study. ........................... 62
Figure 2.32 Synthesis of novel sulfenylating reagent 1-phenylsulfanyl[1,2,4]triazole. .. 63
Figure 2.33 ORTEP view of the structure of tert-Butyl-1,4-dimethyl-3,6-dioxo-2-
(phenylthio)piperazine-2-carboxylate (27) ....................................................................... 67
Figure 2.34 Structure of chiral lanthanide shift reagent Eu(hfc)
3
. ................................... 68
Figure 2.35 Proposed catalytic cycle with quinine leading to sulfenylation of
diketopiperazines by an aryl triazole reagent. ................................................................... 71
Figure 2.36 Mechanism of gramine coupling with alkyl/aryl anion via methylene
indolenine intermediate. .................................................................................................... 73
Figure 2.37 Total synthesis of Brevianamide E involving the key step of coupling of
substituted gramine with substituted diketopiperazine. .................................................... 74
Figure 2.38 Coupling of a substituted gramine and a substituted diketopiperazine in the
course of stereoselective total synthesis of paraherquamide A. ....................................... 75
Figure 2.39 Coupling of gramine with 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione
in the presence of n-tributylphosphine in acetonitrile under reflux. ................................. 76
Figure 2.40 Coupling of gramine with 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione
in the presence of quinine as a Lewis base in acetonitrile under reflux. .......................... 77
Figure 2.41 Coupling of gramine with 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-
dione in the presence of quinine ....................................................................................... 77
Figure 2.42 Reaction with N-Boc gramine and gramine N-oxide where no product was
obtained. ............................................................................................................................ 79
Figure 2.43 Coupling of 3-(1-arylsulfonylalkyl)indole with aldehyde in the presence of
KF/alumina with L-proline as a catalyst. .......................................................................... 80
xiii
Figure 2.44 Proposed mechanism of coupling of the aryl sulfones to aldehydes. ........... 81
Figure 2.45 Coupling of 2-methyl-3-(phenyl(tosyl)methyl)-1H-indole with 3-(t-
butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione ......................................................... 82
Figure 2.46 Coupling of indole and diketopiperazine in the presence of (DHQD)2PHAL
as organocatalyst. .............................................................................................................. 83
Figure 2.47 Coupling of 3-(phenyl(tosyl)methyl)-1H-indole with 3-(t-butoxycarbonyl)-
1,4-dimethylpiperazine-2,5-dione. .................................................................................... 83
Figure 2.48 Coupling of 3-(tosylmethyl)-1H-indole with 3-(t-butoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione ........................................................................................... 85
Figure 2.49 Summary of sulfenylation and indole-DKP coupling strategies. ................. 87
Figure 2.50 Schematic illustration of future application of the sulfenylation strategy to
obtain an ETP molecule from syn-hemithioaminal. ......................................................... 88
Figure 3.1 Structure of p300-CH1 domain in complex with HIF-1α TAD-C. ................ 90
Figure 3.2 Structure of chetmoin ..................................................................................... 91
Figure 3.3 Possible modes of chelation of zinc ion by chetomin. ................................... 92
Figure 3.4 Structures of designed bis-ETPs ETP-3 and ETP-5. ...................................... 94
Figure 3.5 Scheme representing synthesis of ETP-3. ...................................................... 95
Figure 3.6 Synthesis of ETP-5. ........................................................................................ 96
Figure 3.7 SPR sensorgram for chetomin binding with the immobilized GST-CH1-p300.
......................................................................................................................................... 102
Figure 3.8 SPR data for direct binding of ETP-3 to immobilized GST-CH1-p300. ...... 103
Figure 3.9 SPR data for direct binding of ETP-5 to immobilized GST-CH1-p300. ...... 104
Figure 3.10 Structure of NP-481 control compound...................................................... 105
Figure 3.11 SPR sensogram for NP-481. ....................................................................... 106
Figure 3.12 SPR analysis of ETP-3 in two different reducing conditions. .................... 107
Figure 3.13 The equilibrium between the disulfide and dithiol forms of ETP-3. .......... 108
Figure 3.14 CD spectra for the enantiomers of ETP-3 viz. ent1-3 and ent2-3. ............. 109
Figure 3.15 SPR sensorgrams for stereoisomers of ETP-3. ........................................... 110
Figure 3.16 Saturation binding curve for HIF-1α-Fl and GST-CH1-p300 obtained by FP
assay. ............................................................................................................................... 112
Figure 3.17 Saturation binding curve for the cleaved CH1-p300 domain and HIF-1α-Fl,
obtained by FP assay. ...................................................................................................... 113
xiv
Figure 3.18 Fluorescence polarization competition assay shows the disruption of the
complex between HIF-1α TAD-C and GST-CH1-p300 by chetomin. ........................... 114
Figure 3.19 Fluorescence polarization disruption assay to assess the disruption in binding
of HIF-1α and GST-CH1-p300 binding by ETP-5. ........................................................ 115
Figure 3.20 Data from FP assay showing disruption of a complex between HIF-1α TAD-
C and cleaved CH1 domain of p300 with ETP-5. .......................................................... 116
Figure 3.21 HSQC-NMR spectra of p300-CH1 domain and p300-CH1 bound to HIF-1α
TAD-C. ........................................................................................................................... 117
Figure 4.1 Mechanism of downregulation of HIF-1α inducible genes with bis-ETPs. . 120
Figure 4.2 Luciferase assay data for chetomin, ETP-3, and NP-481 in MDA-231-HRE-
Luc cell line..................................................................................................................... 122
Figure 4.3 Downregulation of the expression of VEGF gene in MCF7 cells treated with
chetomin, ETP-3 and NP-481. ........................................................................................ 123
Figure 4.4 LOX mRNA levels in MCF7 cell line after treatment with chetomin, ETP-3
and NP-481. .................................................................................................................... 124
Figure 4.5 Results from the luciferase reporter assay with ETP-5. ............................... 125
Figure 4.6 MTT cytotoxicity assay for ETP-5 in MCF7 cells. ...................................... 126
Figure 4.7 MTT cytotoxicity assay data for chetomin in A549 cell line. ...................... 127
Figure 4.8 MTT cytotoxicity assay data for ETP-5 in A549 cell line. .......................... 128
Figure 4.9 Quantification of VEGF transcriptional levels in MCF-7 cell lines after
treatment with ETPs chetomin, ETP-5 and LS75. .......................................................... 130
Figure 4.10 Quantification of c-Met transcriptional levels in MCF7 cells. ................... 131
Figure 4.11 VEGF and c-Met mRNA levels in MCF7 cells grown in serum-free media
under hypoxia.................................................................................................................. 132
Figure 4.12 Optimization of three control genes viz. β-glucuronidase, RPL32 and β-actin
for normalization of c-Met mRNA levels. ...................................................................... 133
Figure 4.13 Glut1 mRNA levels in MCF7 cells after treatment with ETP-5. ............... 134
Figure 4.14 In MDA-MB-231 cell line ETP-5 showed inhibition of hypoxia inducible
transcription of c-Met gene. ............................................................................................ 135
Figure 4.15 qRT-PCR of mRNA isolated from A549 lung adenocarcinoma cell line
grown in F-12K medium supplemented with 10% serum. ............................................. 137
Figure 4.16 mRNA levels of three HIF-1α inducible genes: VEGF, c-Met and Glut1 in
A549 cells after treatment with ETP-5. .......................................................................... 138
xv
Figure 4.17 qRT-PCR data for LOX and CXCR4 genes in A549 cells treated with ETP-
5....................................................................................................................................... 139
Figure 4.18 mRNA levels for VEGF in A549 cell line, illustrating the dose response to
ETP-5 treatment at three different concentrations. ......................................................... 141
Figure 4.19 mRNA levels for c-Met in A549 cell line showing dose response to ETP-5 at
concentrations of 100 nM, 400 nM, 1600 nM. ............................................................... 142
Figure 4.20 qRT-PCR data for Glut1mRNA in 85% confluent A549 cells treated with
ETP-5 at three different concentrations. ......................................................................... 143
Figure 4.21 qRT-PCR data for A549 cells where hypoxia was induced in 85% confluent
cells. ................................................................................................................................ 144
Figure 4.22 HIF-1α protein levels are not affected by the treatment with ETP-5. ........ 146
Figure 4.23 ETP-5 down-regulates the protein levels of (a) VEGF in MCF7 cells and (b)
c-Met MDA-MB-231 cells in hypoxic cells. .................................................................. 147
Figure 4.24 Microarray genome analysis of MCF7 cells treated with or without treatment
of ETP-5 and with or without hypoxia induction using DFO (300 µM). ....................... 149
Figure 4.25 Results from the analysis of microarray data. ............................................ 150
Figure 4.26 Intravital microscopy images of murine subcutaneous tumor model of
fluorescent N2O2 cells stably transfected with H2B-GFP construct.............................. 153
Figure 4.27 Change in tumor volume obtained from IVM of mice treated with or without
ETP-5. ............................................................................................................................. 154
Figure 4.28 Transcriptional inhibition of HIF-1α inducible genes with dimeric ETP
through disruption of the binding of HIF-1α and p300/CBP . ........................................ 155
Figure 5.1 Diagram of αvβ3 integrin pathway activation. ............................................. 158
Figure 5.2 Structure of α
v
β
3
specific ligand and it role as a targeting ligand in the
delivery of cargo to tumor cells. ..................................................................................... 159
Figure 5.3 Structures of BSH (borocaptate sodium), BSSB (BSH dimer) and BPA
(boronophenylalanine), the most common reagents for BNCT and boron MRI. ........... 161
Figure 5.4 Fission nuclear reaction of 10B. ................................................................... 162
Figure 5.5 Examples of third generation boron delivery agents. ................................... 163
Figure 5.6 Structures of building blocks of targetable boron delivery system. ............. 166
Figure 5.7 Structure DIL-1 having α
v
β
3
integrin ligand conjugated to cyan 40. ........... 167
Figure 5.8 Fluorescence emission spectra for DIL-1 at 435 nm. ................................... 168
Figure 5.9 Structure of trifunctional conjugate DILB-2. ............................................... 169
xvi
Figure 5.10 Fluorescence emission spectra of DILB-2 at 435 nm in aqueous (dashed line)
and viscous (bold line) media. ........................................................................................ 170
Figure 5.11 Excitation spectra for DILB-2 in viscous and aqueous media measured at
540 nm. ........................................................................................................................... 170
Figure 5.12 Structure of DB-3 control, a cyan dye – carborane dendritic wedge
conjugate. ........................................................................................................................ 171
Figure 5.13 Emission spectra for DB-3 in viscous and aqueous media at 435 nm. ....... 172
Figure 5.14 Excitation spectra for DB-3 in aqueous and viscous media measured at 540
nm. .................................................................................................................................. 173
Figure 5.15 Confocal microscopy images of WM115 cells treated with 100 µM of DIL-
1....................................................................................................................................... 175
Figure 5.16 Images of MCF7 cells treated with DIL-1 at 100 µM concentration. ........ 175
Figure 5.17 Imaging of WM115 and MCF7 cells at 4 °C after treatment with DIL-1. . 176
Figure 5.18 WM115 cells show significantly higher uptake of DILB-2 as compared to
MCF7 cells. ..................................................................................................................... 177
Figure 5.19 Temperature dependence of uptake of DILB-2 in WM115 and MCF7 cells.
......................................................................................................................................... 178
Figure 5.20 Control DB-3 shows no noticeable uptake in WM115 cells. ..................... 179
Figure 5.21 Intravital microscopy (IVM) images showing facile uptake of the DILB-2,
but not control DB-3 conjugate in subcutaneous tumor models. .................................... 181
xvii
Abstract
Angiogenesis, the process of formation of new blood vessels, is a characteristic of
most solid malignancies and is involved in promoting rapid growth of neoplastic
diseases. In cancers, several cytokines are secreted that give rise to angiogenesis, and
facilitate enhanced cell migration, invasion, survival and proliferation. Apart from the
transcriptional changes that result in elevated levels of expression of the oncogenic
transcription factors, the involvement of integrins, heterodimeric transmembrane
receptors that modulate cell-cell and cell-matrix interactions is another key element in
promoting tumorigenesis. This dissertation is focused on two aspects of modulation of
oncogenic pathways: 1) by direct inhibition of the transcription factor complexes that are
responsible for maintenance of oncogenic phenotype and 2) targeting of integrin
receptors that are overexpressed on cancer cells and tumor vasculature with the goal of
developing of novel therapeutics and imaging agents.
Our main focus is on regulation of the activity of hypoxia inducible transcription
factor 1α (HIF-1α), which is overexpressed in cancer cells. High metabolic rate of the
growing tumor coupled with limited supply of oxygen by the nearby blood vessels
quickly results in the state of local hypoxia within the tumor. The hypoxic tumor cells
rapidly stabilize and accumulate the alpha subunit of the HIF-1 (HIF-1α) which is a key
component of the transcription factor complex responsible for overexpression of the key
genes involved in angiogenesis, invasion, and altered energy metabolism. Stabilized
HIF-1α translocates to the nucleus, heterodimerizes with its beta subunit, binds to its
cognate DNA sequence called hypoxia response (HRE) element and recruits cofactor
p300/CBP, resulting in upregulation of hypoxia-inducible genes. This dissertation
xviii
describes development of small molecules that inhibit the hypoxia-inducible transcription
factor complex through the disruption of its binding to coactivator proteins p300/CBP.
Using rational design and inspired by the natural product chetomin, we designed small
molecules that inhibit binding of HIF-1α to p300/CBP. We have shown that the synthetic
small molecules ETP-3 and ETP-5 directly interact with the cysteine-histidine rich region
1 (CH1) of human p300 coactivator and that they also disrupt the complex between HIF-
1α and p300, resulting in a rapid downregulation of hypoxia-inducible genes, such as
VEGF, c-Met, Glut1, LOX, CXCR4. These small molecules are potent inhibitors of HIF-
1α induced signaling in vitro. In addition, ETP-5 has shown remarkably high activity in
mouse tumor xenograft model.
The second part of this dissertation is focused on integrins, bidirectional allosteric
signaling proteins responsible for cell-cell communication, adhesion and interaction with
extracellular matrix. We focused on α
v
β
3
integrins, as these receptors are overexpressed
on vascular endothelial cells that undergo angiogenesis, although they are typically not
found on quiescent cells. This makes them attractive anticancer targets, as they can be
targeted in order to deliver imaging and therapeutic reagents to growing tumor
vasculature. We have designed and synthesized novel α
v
β
3
-specific sensors incorporating
integrin ligand and an environment-sensitive fluorescent cyan 40 dye. Such sensors show
40-80 fold fluorescence enhancement upon binding to their cognate integrin receptors.
Furthermore, we designed α
v
β
3
-specific ligand incorporating therapeutic moieties, such as
boron-rich dendritic wedges that can be used for boron neutron capture therapy and boron
MRI. The integrin ligand conjugates show good affinity and specificity towards cells
xix
with high levels of expression of α
v
β
3
integrins. In murine subcutaneous tumor models,
these conjugates have shown excellent uptake.
This dissertation summarizes work focused on development and evaluation of
small molecules for modulating and targeting cancer cells both on transcriptional and
cellular levels.
1
Chapter 1: Introduction
2
1.1 The role of hypoxia in cancer biology
Cancer is a state where the delicate balance between proliferative and antiproliferative
signals found in a normal cell is disrupted. It is a broad group of diseases that are
manifested by a deviation from homeostasis and is typically leading to an uncontrolled
proliferation of cells.
Hypoxia is one of the most important hallmarks of solid tumors that plays a vital role in
cell proliferation, signaling and growth (Brown and Wilson 2004). A typical neoplasm is
usually devoid of blood vessels in its early stage. The rapidly proliferating cells
contribute to development of hypoxia (Vaupel, Schlenger et al. 1991). Despite the fact
that cell proliferation decreases in those parts of a tumor that are away from blood vessels
(Tannock 1968), they tend to select for more aggressive cellular phenotypes. Moreover, it
has been reported that the hypoxic tissue away from the blood vessels give rise to cells
that have lost sensitivity to p53-mediated apoptosis (Brown and Wilson 2004).
Hypoxia also leads to upregulation of genes involved in drug resistance, such as P-
glycoproteins (Comerford, Wallace et al. 2002; Wartenberg, Ling et al. 2003) in addition
to the fact that lack of adequate blood supply to hypoxic cells severely impairs the
delivery of drug to these cells (Durand 1994; Tannock 1998). Most importantly, from a
transcriptional standpoint, hypoxia results in an upregulation of genes involved in
angiogenesis (Harris 2002) and tumor invasion (Pennacchietti, Michieli et al. 2003)
resulting in more aggressive cancer phenotype (Graeber, Osmanian et al. 1996).
In cells and tissues, response to hypoxia is primarily mediated by the family of hypoxia-
inducible transcription factors, among which HIF-1 plays a major role. HIF-1 is
3
heterodimeric, basic helix-loop-helix protein of Per-ARNT-Sim family, comprised up of
two cognate subunits, HIF-1α and HIF-1β. HIF-1α, whose cellular levels are inversely
proportional to partial pressure of oxygen, is critical component of hypoxic response.
HIF-1α under hypoxia is stabilized and translocates to nucleus where it binds to its
partner HIF-1β/ARNT and the HIF-1 complex binds to the hypoxia response element
(HRE) in the DNA and they recruit transcriptional co-activator of p300/CBP. This
assembly leads to the upregulated transcription of various genes that are induced under
hypoxia (Harris 2002). In many tumor cells where oncogenic mutations in RAS, SRC and
HER2/NEU/ERBB2 are found, high levels of HIF-1 have been detected even under well-
oxygenated condition (Giaccia, Siim et al. 2003).
Under well-oxygenated conditions HIF-1α is nearly undetectable in normal cells and
tissues. The cells in hypoxic tumors and cells where vHL protein is mutated, on the other
hand, show high levels of this protein, potentially presenting an attractive therapeutic
strategy (Subarsky and Hill 2003). Therefore, targeting of HIF-1 in order to suppress its
activity in hypoxic cells could become a very attractive antitumor strategy (Zhong, De
Marzo et al. 1999; Talks, Turley et al. 2000). This can be accomplished either through the
induction of degradation of HIF-1 or by reducing levels of its gene expression.
Geldanamycin has been shown to reduce the HIF-1 levels by vHL-independent
proteasomal degradation pathway (Sun, Kanwar et al. 2001; Mabjeesh, Post et al. 2002).
It has been shown that antisense construct of HIF-1α eradicates in vivo a small
transplanted thymic lymphoma and even increases the efficacy of immunotherapy against
larger tumors (Sun, Kanwar et al. 2001). Small molecule inhibitors of microtubules, such
as 2-methoxyestradiol, vincristine and paclitaxel have been shown to reduce HIF-1α
4
levels in vitro and also reduce tumor growth and vascularization (Mabjeesh, Escuin et al.
2003). However, it is not clearly understood whether the effects shown in tumor growth
reduction is due to microtubule inhibition or reduction of HIF-1α levels.
Hypoxic cells are resistant towards radiation therapy. It has been suggested that the cause
for this resistance is in the primary mechanism of damage by radiation by which it kills
the cells is by ionizing the DNA itself either directly or indirectly by ionizing the
molecules in the close vicinity of DNA. The ionized DNA (DNA·) can either bind to an
oxygen molecule forming DNA-OO· in normoxic cells or grab the hydrogen radical from
thiols (-SH) in hypoxic cells and DNA· is restored to the normal state. In hypoxic cells
lacking oxygen and due to the reducing environment within the cells, the concentration of
thiols (-SH) is high. DNA-OO· leads to permanent DNA damage and eventually leads to
cell death (Brown and Wilson 2004). The concept of prodrugs that target hypoxic cells is
an interesting approach that has been in development for over a decade. Tirapazamine
(TPZ) is one such pro-drug that has been discovered by Brown and Lee (Zeman, Brown
et al. 1986; Brown and Lemmon 1990; Brown 1993). TPZ forms a radical under ionizing
conditions. Oxidizing radical formed by spontaneous decay of the protonated TPZ radical
is considered to be the toxic species. Thus, in hypoxic cells TPZ increases the toxic
effects of ionizing radiation which are otherwise subdued by presence of free thiols under
hypoxia. TPZ itself is not toxic but becomes such when combined with ionizing radiation
giving rise to toxic products (Peters and Brown 2002). NLCQ-1 (Papadopoulou and
Bloomer 2003) is another example of a compound that is used as an adjuvant to radiation
therapy in hypoxic cells. SN 23862 (Siim, Denny et al. 1997) and AQ4N (Patterson 1993;
Patterson, McKeown et al. 2000) are two examples that have shown their efficacy as pro-
5
drugs when used along with radiation therapy on hypoxic cells. These prodrugs mainly
act by inducing DNA damage, although the exact mechanism of action is different for
each molecule.
Figure 1.1 Prodrugs that selectively target hypoxic cells and can be used as adjuvant in
radiation therapy to target otherwise resistant hypoxic tumor cells.
As mentioned above, in most cell lines HIF-1α is virtually undetectable under normoxia
(Wang, Jiang et al. 1995) because it undergoes hydroxylation followed by proteasomal
degradation (Huang, Arany et al. 1996; Salceda and Caro 1997). Under normoxic
condition where pO
2
(extracellular) > pO
2
(intracellular) there is sufficient amount of O
2
present in the cell to activate Fe
2+
containing metallo-enzyme prolyl hydroxylase 2
(PHD2), which subsequently hydroxylates HIF-1α at amino acid positions Pro-402 and
Pro-564. The hydroxylated HIF-1α is then poly-ubiquitinated by von Hippel Lindau
(vHL) tumor suppressor protein. The poly-ubiquitinated HIF-1α undergoes proteasomal
degradation (Ivan, Kondo et al. 2001; Jaakkola, Mole et al. 2001; Masson, Willam et al.
2001). Apart from this mechanism of degradation, a normoxia-independent mechanism of
proteasomal degradation has also recently been suggested which depends on the protein
RACK-1 (Liu, Baek et al. 2007). Another mechanism that acts as a back-up in case the
prolyl hydroxylase mediated degradation is not able to destabilize all the HIF-1α is
hydroxylation of Asn-803 by asparaginyl hydroxylase/FIH (Factor Inhibiting HIF) under
6
normoxic condition. The hydroxylated Asn-803 prevents binding or recruitment
p300/CBP by HIF-1α (Lando, Peet et al. 2002).
Figure 1.2 Schematic representation of HIF-1α signaling pathway and its dependence on
normoxia and hypoxia. The stabilized HIF-1α translocates into the nucleus forming HIF-
1 heterodimer with ARNT/HIF-1β which interacts with p300/CBP coactivator and
activates transcription of hypoxia inducible genes.
Under hypoxic conditions, HIF-1α is stabilized as the limited supply of oxygen in the
cells sharply reduces its hydroxylation levels by PHD2. The stabilized HIF-1α
translocates into the nucleus where it heterodimerizes with aryl hydrocarbon receptor
translocator ARNT/HIF-1β, that is known to bind proteins with bHLH-PAS domain. The
HIF-1 heterodimer interacts with p300/CBP and some members of SRC-1 family co-
activators. This complex binds to hypoxia response element (HRE) on DNA and
transactivates HRE containing promoters and enhancers, leading to transcription of genes
7
involved in angiogenesis and metastasis (Huang and Bunn 2003; Pugh and Ratcliffe
2003). Mapping of the structural requirements for binding of HIF-1α to p300 (Kung,
Wang et al. 2000) through in-vitro transcription assays with the various deletion
fragments of p300 protein highlighted that for interaction of p300 and HIF-1α, the
minimal HIF-1α binding domain comprises of 116 residues of p300 (aa302-418), which
includes CH1 domain and its flanking sequences.
In order to further detail the binding interaction of HIF-1α, various Gal4 fused truncation
mutants of HIF-1α carboxyl terminal domain were generated, as this region of HIF-1α
has been shown to interact with p300/CBP. The binding partner in the in-vitro assays to
these fused proteins was defined as GST-CH1 (a.a. 302-443) region of p300. The C-
terminal transactivation domain (TAD-C) comprised of 41 residues from C-terminal of
HIF-1α is necessary and sufficient for binding to GST-CH1 protein. Further studies
showed that it is imperative for the CH1 domain to bind zinc ion, which stabilizes its
conformation in order to bind HIF-1α.
8
Figure 1.3 Domain map of HIF-1α and p300. a) TAD-C of HIF-1α that is necessary and
sufficient for binding to p300 CH1 domain is from amino acids 786-826. b) Fusion
protein Gal4-TAD-C binds GST-CH1 bind in vitro; similar fusion peptides also attenuate
gene expression in vivo.
Hypoxia-inducible reporter transcriptional activity was downregulated by overexpression
of polypeptides corresponding to TAD-C of HIF-1α and CH1 domain of p300.
Polypeptides for TAD-C were expressed as Gal4 fusions. Retrovirally infected tumor
cells expressing Gal4 fused TAD-C polypeptide in the MDA-MB-435 and HCT-116 cell
carcinomas were then xenografted in nude mice. Tumor growth was significantly reduced
for xenografts treated with the TAD-C polypeptide fusions that were selected based upon
their ability to disrupt complex of CH1 and TAD-C in vitro.
9
1.2 Chetomin: small molecule inhibitor of HIF-1α- p300/CBP complex
In 2004 Livingston and coworkers reported that chetomin, a natural product produced by
filamentous fungus Chaetomium sp. disrupts binding of HIF-1α and p300 (Kung,
Zabludoff et al. 2004). This molecule was identified from a high-throughput screen of a
library of more than 600,000 compounds consisting of both natural and synthetic drug-
like compounds. Chetomin showed inhibition of HIF-1α /p300 interaction at sub-
micromolar concentrations and good in vitro activity, although its mechanism of action
was unclear.
Figure 1.4 Structure of chetomin, having ETPs (epidithiodiketopiperazines) within its
north and south fragment and a substituted cyclotryptophan motif.
Chetomin belongs to the family of natural products called sporidesmins, majority of
which have been isolated from filamentous fungi (Brewer, Archibal.Rm et al. 1972;
Sekita, Yoshihira et al. 1981). Structurally, sporidesmins and chetomin have unique
chemical moiety, epidithiodiketopiperazine (ETP) that is manifested by the disulfide
bridge incorporated into the diketopiperazine ring. No total synthesis of chetomin has
been reported to date and its main source is fungal extracts.
10
Chetomin disrupts binding of full-length HIF-1α and its homolog HIF-2α to GST-CH1 in
vitro. The interaction between TAD-C of HIF-1α (a.a. 776-826) fused to Gal4 DNA
binding domain and CH1 domain of p300 (a.a. 311-528) fused to VP16 activation
domain was measured in Hep3B cells. This cell-based two hybrid assay also showed that
chetomin disrupts interaction between TAD-C of HIF-1α and CH1 domain of p300 in a
dose-dependent manner at low nanomolar concentrations. Further studies have shown
that Cys-800 is essential for the interaction of HIF-1α TAD-C to p300 and that the same
efficacy was found in a Cys800Val mutant for TAD-C suggesting that the mechanism of
action of chetomin is not by its attack at the Cys-800 in TAD-C. In addition, chetomin
showed high specificity towards CH1 domain of p300 as it did not disrupt the
transactivation of other domains of p300, nor global activity of that protein. NMR
experiments demonstrated that chetomin induces unfolding of CH1 domain in a dose-
dependent fashion. A plausible explanation for chetomin and other ETPs that were shown
to disrupt HIF-1α binding to p300 is that the ETPs bind/chelate zinc atoms present in the
CH1 domain of p300. In vitro, chetomin showed dose-dependent attenuation of hypoxia
inducible transcription in luciferase assays utilizing VEGF-Luc, CMV-Luc, 3xHRE-Luc
and Epo-Luc promoter constructs. Real time qRT-PCR assays on hypoxia inducible
genes Eno1, Glut1 and Tf showed downregulation of mRNA levels with chetomin in
preliminary in vivo experiments, chetomin downregulated HIF-1 markers, like serum
erythropoietin (EPO) reduced expression of hypoxia inducible transcription in two hybrid
assays with Gal4-TAD-C and VP16-CH1 and also attenuated tumor growth. Despite the
initial encouraging reports, further design of inhibitors of hypoxia-inducible pathway is
11
needed, because chetomin induced coagulative necrosis, anemia and leukocytosis in
experimental animals.
1.3 Biologically relevant epidithiodiketopiperazines (ETPs)
Epidithiodiketopiperazines, also called epidithiopiperazine-2,5-diones are secondary
fungal metabolites that belong to the same class as alkaloids, cyclic peptides, polyketides
and sesquiterpenoids. ETPs have are structurally diverse and have a rich spectrum of
biological activities (Gardiner, Waring et al. 2005). ETPs are marked by the presence of
epidithiodiketopiperazine ring, which is a six-membered ring having a syn-disulfide
bond. All examples of the ETP natural products also have an aromatic group, thereby
implying involvement of an aromatic amino acid in their biosynthesis. Most often the
aromatic group present in the ETP molecule is an indole group or its derivative,
suggesting that tryptophan is the amino acid most often involved in the biosynthesis of
ETPs.
Figure 1.5 Epidithiodiketopiperazine (ETP) ring is a core fragment of the natural
products called sporidesmins.
The two sulfurs syn to each other are the major factor in imparting biological activity to
these molecules. ETPs can further be divided into two classes, those having only one ETP
ring (mono-ETPs) and those having two ETP rings (bis-ETPs). Apart from chetomin,
12
other important biologically relevant bis-ETPs (Taylor 1971) are chaetocin (Hauser,
Weber et al. 1970), verticillins (Joshi, Gloer et al. 1999; Son, Jensen et al. 1999; Liu, Liu
et al. 2011) and leptosin K(Takahashi, Numata et al. 1994). Most of the bis-ETPs have
anticancer properties (Vigushin, Mirsaidi et al. 2004). Many of them are also antibiotics.
Chaetocin is a potent anticancer molecule, antiangiogenic and inhibitor of thioredoxin
reductase. Verticillins A, B and C are reported to have activity of antibiotics. Leptosins
show pronounced cytotoxicity towards different cell lines.
Figure 1.6 Representative structures of bis-ETPs, containing an ETP ring within a north
and a south fragment.
Despite the fact that many bis-ETPs were first isolated over 40 years ago, their total
syntheses were not reported until recently, underlining the complexity and difficulty
involved in the preparation of these compounds.
13
11,11'-Dideoxyverticillin A and chaetocin (A) both show antiangiogenic activities and
both of them are among the few recent examples of bis-ETPs whose total synthesis have
been achieved, with the first total synthesis of 11, 11'-dideoxyverticillin A reported by
Movassaghi (Kim, Ashenhurst et al. 2009) and total synthesis of chaetocin - by Sodeoka
(Iwasa, Hamashima et al. 2010). Despite these encouraging reports, there is a need to
further develop facile and rapid synthesis of unnatural bis-ETPs with improved biological
activities.
Gliotoxin, sporidesmin A and hyalodendrin are three examples of biologically active
mono-ETPs (Gardiner, Waring et al. 2005). Gliotoxin, for example, has antiangiogenic
and cytotoxic properties. However, in cell culture it induces apoptosis at lower
concentrations and necrosis at higher concentrations. Sporidesmin A (Di Menna, Smith et
al. 2009) have been reported to cause facial eczema in animals and hyalodendrin is
fungitoxic (Strunz, Kakushima et al. 1973).
Figure 1.7 Structures of gliotoxin, sporidesmin A and hyalodendrin.
ETPs have been shown to undergo redox cycle in the cells. The disulfide bond in an ETP
inside a cell is reduced into dithiol and is oxidized back to a disulfide by oxygen
molecule (Hurne, Chai et al. 2002; Gardiner, Waring et al. 2005). This redox cycle is
important but does not explain other biological activities shown by ETPs in general. The
redox cycle is believed to generate reactive oxygen species (ROS) which could be the one
14
possible cause of the biological activity of gliotoxin. However, ROS scavengers did not
diminish the activity of gliotoxin thereby implying that apart from the redox cycle there
are other factors also that impart biological activity/toxicity to gliotoxin.
Recent reports by us and others highlighted the mechanism of action of the ETPs under
hypoxia conditions. In case of chetomin it was explicitly shown that a major cause for the
action of chetomin is through the disruption of interaction between CH1 of p300 and
TAD-C of HIF-1α (Kung, Zabludoff et al. 2004). Recently a synthetic ETPs have been
reported that target this very HIF-1α inducible transcription by recruitment of p300/CBP.
Our laboratory (Block, Wang et al. 2009) synthesized a bis-ETP that showed direct
binding to the CH1 domain of p300 by SPR assays and in vitro disruption of GST-CH1
and TAD-C (a.a. 786-826) of HIF-1α by fluorescence polarization assay. In cell based
assays the ETP-3 showed attenuation of transcriptional activity of promoter containing 5
repeats of HRE sequent (TACGTGGG) fused to luciferase and showed dose-dependent
inhibition of mRNA levels of hypoxia-inducible genes VEGF and LOX.
Figure 1.8 Structure of ETP-3, a synthetic bis-ETP. The two ETP rings are connected by
an aromatic linker.
15
Schofield and coworkers (Cook, Hilton et al. 2009) reported synthesis and biological
activity of a several mono-ETPs under hypoxic conditions. In the first set of ETPs the
substituent on one of the nitrogen of the ETP ring were varied to obtain compounds SC-5
to SC-9 (Figure 1.10). Compounds SC-5 to SC-8 showed good inhibition of binding
between GST-CH1 and biotinylated TAD-C of HIF-1α.
Figure 1.9 Structures of the ETP bearing synthetic compounds SC-5 to SC-9 developed
by Schofield and coworkers.
A second set of compounds with varying substituents on the sulfurs (Figure 1.11) were
synthesized and their ability to disrupt p300 and HIF-1α binding was investigated.
Compound SC-11, having single sulfur in the form of a thiol group and compound SC-12
with two sulfurs being methylated showed reduced activity toward disruption of the
binding between HIF-1α TAD-C and GST-CH1.
Figure 1.10 Structure of compounds SC-10, SC-11 and SC-12 developed by Schofield &
coworkers.
16
Compound SC-10 with two sulfurs as thiols in syn-orientation showed good inhibition of
TAD-C and CH1 binding. A zinc ejection was postulated as the mechanism for disruption
of in vitro binding of TAD-C of HIF-1α and GST-CH1. Cell-based assays showed low to
medium micromolar range IC
50
values for secreted levels of VEGF for these compounds
while cell viability was still maintained.
Diketopiperazines (2,5-piperazinediones) are precursors to ETPs in many biosynthetic
pathways, which are also structural components of natural products. Several examples of
biologically active diketopiperazine natural products are shown in Figure 1.12.
Figure 1.11 Structure of ditryptophenaline, WIN 64821, WIN 64745 and Leptosin S.
Ditryptophenaline is a moderate P-substance antagonist (Springer, Buchi et al. 1977).
Whereas WIN 64821 (Barrow, Cai et al. 1993; Popp, Musza et al. 1994) is a potent P-
substance inhibitor suggesting that the stereochemistry at the north and south fragment
linkage plays an important part in mechanism of P-substance inhibition. Compounds
17
WIN 64821 and WIN 64745 apart from being potent P-substance antagonist for human
neurokinin-1 are also antagonist to cholecystokinin B receptors (Hiramoto 1994). These
targets make them potential therapeutics in treatments of arthritis, asthma and other
inflammatory diseases. Leptosin S has shown cytotoxicity towards 39 cancer cell lines
(Yamada, Iwamoto et al. 2004).
As precursors in the biosynthesis of ETPs, DKPs (diketopiperazines) are important class
of compounds that on their own have potential to be used as therapeutics. Therefore, it is
highly desirable to not only synthesize new ETPs and DKPs but also to develop novel
strategies to synthesize building blocks and important linkages en route to their synthesis.
1.4 Hypoxia-inducible transcription of VEGF, Met and other downstream genes
HIF-1α is the most abundant HIF protein that responds to oxygen levels in cells and
tissues and induces transcription of hypoxia-inducible genes, such as VEGF,Met, Glut1,
LOX, CXCR4 etc. Although more than one pathway/factor can be implicated in
transcriptional regulation of these genes under normoxia, under hypoxia HIF-1α is the
most important transcription factor implicated regulating the mRNA levels of these
genes.
1.4.1 The role of VEGF in tumor progression and angiogenesis. Vascular endothelial
growth factor (VEGF) protein and its receptors VEGFRs (vascular endothelial growth
factors receptors) are integral part of the signaling mechanism involved in angiogenesis
and tumor progression. VEGFs and VEGFRs are regulators of vasculogenesis during
18
embryogenesis (Carmeliet, Ferreira et al. 1996; Ferrara, CarverMoore et al. 1996) and
angiogenesis in adults (Kubo and Alitalo 2003; Schatteman and Awad 2004).
In mammalian cells VEGF protein is found in five isoforms: VEGFA-VEGFD and PLGF
(placenta growth factor). Each of these five factors also have different splice variants.
The VEGFs are cytokines, which act as ligands for their receptors VEGFRs proteins that
belong to receptor tyrosine kinase (RTK) family. There are three types of VEGFR; both
VEGFs and VEGFRs act as dimers, with VEGFs typically forming homodimers in anti-
parallel fashion. VEGFRs, on the other hand, form both homo- and heterodimers.
VEGFA, VEGFB and PLGF bind to VEGFR1. VEGFA also binds to VEGFR2. VEGFC
and VEGFD, as such, bind to VEGFR3 but after proteolytic processing VEGFC and D
may bind to VEGFR2. VEGFs are known to undergo alternative splicing, giving rise to
different isoforms with different biological activities (Olsson, Dimberg et al. 2006).
19
Figure 1.12 Interaction between VEGF dimers with different combinations of the three
VEGFRs dimers.
In humans VEGFA has five isoforms (Olsson, Dimberg et al. 2006) which are denoted as
VEGFA121, VEGFA145, VEGFA165, VEGFA189 and VEGFA206; the number at the
end of the denotation refers to the number of amino acid residues in the VEGFs. The
VEGFRs have the extracellular domain consisting of seven immunoglobulin (Ig) like
folds in VEGFR1 and VEGFR2. In VEGFR3 the fifth Ig domain is substituted by a
disulfide linkage. This 750 amino acid residues long extracellular domain is followed by
a juxta-membrane domain that lies in the cell membrane. The intracellular region of
VEGFRs has a split tyrosine kinase domain. VEGFRs on being activated by the ligands
belonging to VEGF family initially auto-phosphorylates itself. After VEGFRs are auto-
20
phosphorylated at the required tyrosine residues, they in turn recruit different proteins
thereby propagating the signaling cascade.
Figure 1.13 Schematic representation of VEGF pathway showing VEGFR2 and some of
the important cascades that emanate from it.
VEGFR2 is one of the most important VEGFR that is responsible for most of the
signaling and physiological manifestation of VEGF binding to VEGFRs. Vascular
permeability, cell migration and actin remodeling are some of the key effects of VEGF
pathway (Figure 1.14). These effects are part of the mechanism that ultimately leads to
angiogenesis in adults and vasculogenesis in embryos.
Hypoxia regulates the levels of VEGFA (Pugh and Ratcliffe 2003)and VEGFRs (Gerber,
Condorelli et al. 1997). HIFs, including HIF-1, are directly responsible for the regulation
of VEGFA and VEGFRs. VEGFA and VEGFR2 are the most important pair of ligand
21
and receptor respectively which is directly regulated by HIF-1α. VEGFR1 and VEGFR2
are known to affect the function of each other. VEGFR1 negatively affects VEGFR2
(Rahimi, Dayanir et al. 2000; Zeng, Dvorak et al. 2001). This is believed to be due to the
capture of VEGFA by VEGFR1 as result the effective amount VEGFA available for
VEGFR2 is reduced thereby affecting the signaling by VEGFR2 which is the main
pathway responsible for vasculogenesis and vascular permeability. Until recently,
targeting VEGF and VEGFR2 interaction has been main strategy in combating
angiogenesis in treatment of tumors. Some drugs that have been approved by FDA (Jain,
Duda et al. 2006) or are hopeful candidates to do so are bevacizumab (Avastin) which is
a monoclonal anti-VEGF antibody is approved by FDA. VEGF neutralizing aptamer,
Pegaptanib and Sunitinib along with Sorafenib which are multitargeted kinase inhibitors
including VEGFR2 kinase response are also FDA approved drugs. All these drugs inhibit
the interaction of VEGF with VEGFR2. These anti-VEGF drugs are used along with
other treatments and there use in isolation to other therapies can be counterproductive.
The cell treated with these drugs become more resistant to radiation therapy and tend to
metastasize more (Hayden 2009). A transcriptional regulation of VEGF and VEGFR2 by
HIF-1α may prove to be a very effective approach because transcriptional regulation by
HIF-1α not only inhibits VEGF but also other genes involved in metastasis.
1.4.2 Role of c-Met in cancers and its regulation by HIF-1α. c-Met is a proto-
oncogene which is mainly involved in invasiveness and metastasis of cancers
(Pennacchietti, Michieli et al. 2003). It is a receptor for hepatocyte growth factor (HGF).
Met is a single pass receptor having a disulfide linked α/β heterodimer. Its α chain is
completely extra-cellular whereas β chain is partially extracellular and has the trans-
22
membrane subunit followed by the tyrosine kinase domain (Trusolino and Comoglio
2002). The β chain has a sema domain which is about 500 amino acid long and has an
eight-cysteine peptide module called MRS (Met-related sequence). Sema domain also has
three glycine-proline rich (G-P) repeats. The tyrosine kinase domain of met has two key
tyrosine residue, viz. Tyr1234 and Tyr1235 within the catalytic site whose
phosphorylation leads to positive modulation of the enzymatic activity of Met. On the
contrary the phosphorylation of Ser975 in the juxtamembrane domain downregulates the
kinase activity of Met. Phosphorylation of Tyr1349 and Tyr1356 in the carboxy-terminal
domain forms the docking site for several signal transducers and adaptors (Boccaccio and
Comoglio 2006).
Figure 1.14 Schematic diagram of Met receptor having β and α chain in the extracellular
domain and intracellular tyrosine kinase domain.
Met protein overexpression and its gene amplification are found in many aggressive
carcinomas. The deregulation of Met's catalytic activity is implicated in its tumorigenic
23
activity. Constitutive activation of Met activity (Trusolino and Comoglio 2002) is
obtained by all three ways of activation of a tyrosine kinase receptors, viz. a) Ligand-
receptor autocrine circuits; b) Receptor overexpression leading to reciprocal activation in
the absence of the ligands, and c) Structural alteration in the receptor that keep it
functioning continuously (Hanahan and Weinberg 2000). Most of the mutations found in
the Met protein are in the catalytic domain leading to increased tyrosine kinase activity.
HGF availability is crucial for the Met dependent tumorigenicity in cancer cells where
Met is overexpressed (Michieli, Basilico et al. 1999).
Figure 1.15 Met dimer activation by HGF. Binding of catalytic tyrosine kinase domain
of activated Met to signaling proteins activates pathways that lead to regulation of cell
survival, motility, proliferation and morphogenesis.
HGF/Met overexpression reduces the aggregation of cancer cells. This is done by
affecting the adherens junction (Shibamoto, Hayakawa et al. 1994). The expression of
24
cadherins is reduced along with altered proteolytic activity of the carcinoma cells in order
to escape from the neighboring cells and extracellular matrix. Phosphorylation of
Tyr1349 and Tyr1356 of Met receptor allows it to promiscuously bind to several
different proteins that include PI3K (Ponzetto, Bardelli et al. 1994), SRC, GRB2, SHC
(Pelicci, Giordano et al. 1995), GAB1 (Weidner, DiCesare et al. 1996), PLCγ (Gual,
Giordano et al. 2000), SHP2 (Maroun, Naujokas et al. 2000; Schaeper, Gehring et al.
2000)and STAT3 (Boccaccio, Ando et al. 1998). These pathways lead to enhanced cell
proliferation, motility and survival (Figure 1.16).
Hypoxia induces expression of Met in several cell lines (Pennacchietti, Michieli et al.
2003). Human lung carcinoma cell line A549, osteosarcoma cell line U2-OS,
hepatocellular carcinoma cell line HepG2 and breast epithelium B5/589 are a few
examples of important cell lines that show overexpression of Met under hypoxia. The
promoter region of Met has many hypoxia response elements (HRE) which are the
putative binding sites for HIF-1 heterodimer. It has been reported that knockdown of Met
by RNA interference abrogates the hypoxia-induced invasive tumor growth.
1.4.3 Other HIF-1α inducible genes. Many other important genes are induced by
hypoxia in a HIF-1α dependent manner. These genes can be divided according to their
functions. In cancer pathology, three genes that play significant role in metastasis and
cancer cell proliferation are Glut1, LOX and CXCR4.
Role of Glut1 in cancer: Glut1 is a facilitative glucose transporter protein expressed by
cells (Maxwell, Dachs et al. 1997; Wenger 2000). It is expressed ubiquitously in normal
cells but it has been found to be overexpressed in hypoxic cancer cells especially those
25
that are prone to metastasis (Brown 1993; Mellanen, Minn et al. 1994; Kurata, Oguri et
al. 1999). Glut1 expression correlates with hypoxia in tumors and predicts metastasis-free
survival in cervical cancer. In renal cell carcinoma the cells are dependent on aerobic
glycolysis for ATP production. This phenomenon is called Warburg effect (Warburg
1956). Glut1 aids in the induction of glycolysis by these cancer cells.
In a recent report (Chan, Sutphin et al. 2011) it is shown that a class of molecules that
bind to Glut1 thereby inhibiting the intake of glucose by the cells selectively kill the
cancer cells in presence of normal cells. The STF-31 compound was shown to directly
bind to Glut1 protein and in vivo experiments showed no toxicity to the animal but was
efficacious in killing the renal carcinoma cells.
This exemplifies the importance of Glut1 inhibition as a therapeutic target in cancers.
Glut1 is overexpressed in hypoxic cells and HIF-1 is shown to be the factor that regulates
hypoxic upregulation of Glut1 (Behrooz and IsmailBeigi 1997).
Role of lysyl oxidase (LOX) in cancer: Lysyl oxidase (LOX) plays a very key role in
the formation and repair of extracellular matrix by oxidizing lysine residue in elastin and
collagen. This leads to formation of crosslinks which strengthens these proteins toward
mechanical forces (Smith-Mungo and Kagan 1998). Lysyl oxidase requires copper ion as
a cofactor along with a carbonyl compound for its catalytic activity (Dove and Klinman
2001). Apart from this basic role of LOX it also plays very important role in cancers. In
breast cancer LOX mRNA levels are known to be upregulated, especially in highly
invasive and metastatic cell lines like MDA-MB-231 and Hs578T LOX mRNA levels are
upregulated (Kirschmann, Seftor et al. 1999; Kirschmann, Seftor et al. 2002). It is
26
expected since metastatic cells are required to produce proteins that are able to modify
extra cellular matrix so as to aid the cells in escaping from the tissue. It has been shown
that LOX overexpression enhances cell migration in collagen IV/laminin/gelatin matrix
and the LOX inhibitors like β-aminopropionitrile decrease the invasiveness of breast
cancer cells.
It has been shown that LOX is essential for hypoxia-induced metastasis (Denko, Fontana
et al. 2003; Erler, Bennewith et al. 2006). The role of LOX in tumorigenesis is largely
dependent on cell type, cell location and tumor cell transformation status. In all hypoxic
tumor cells LOX levels are elevated. LOX expression is regulated by HIF-1 under
hypoxia. Thus in hypoxic tumor cells LOX secretion is essential for invasiveness of the
cells. LOX overexpression is essential for metastasis in breast, neck and head tumors and
is an important therapeutic target.
Role of CXCR4 in cancers: CXCR4 is a G-protein coupled receptor which along with
α-chemokine stromal-derived factor (SDF-1) is a key component in regulation of cell
trafficking of many different type of cells including normal hematopoietic stem cells
(HSCs) and nonhematopoietic tissue committed stem/progenitor cells (TSCs). Thus in
metastatic cancer cells SDF-1-CXCR4 axis plays key role in cell trafficking (Darash-
Yahana, Pikarsky et al. 2004; Kucia, Reca et al. 2005).
CXCR4 has been shown to be upregulated in hypoxia and its expression is regulated by
HIF-1 pathway (Schioppa, Uranchimeg et al. 2003; Ceradini, Kulkarni et al. 2004). In
many cancers including prostate cancer CXCR4 overexpression is a marker for
metastasis and it accelerates tumor growth (Taichman, Cooper et al. 2002). CXCR4 is
27
also implicated in enhanced proliferation of tumor cells and VEGF secretion(Kijowski,
Baj-Krzyworzeka et al. 2001). Along with VEGF secretion CXCR4 has been shown to
promote prostate tumor vascularization. In breast cancers overexpression of CXCR4 is
found on mammary epithelium cells whereas in the normal mammary epithelium CXCR4
is undetectable (Dontu, Al-Hajj et al. 2003). CXCR4 has been shown to mediate actin
polymerization, pseudopod formation, and invasiveness. Moreover the theory of
chemotaxis of metastatic cell was proved with the observation that the organ to which
breast cancer metastasizes secretes high levels of SDF-1. In lung cancer in vivo
experiments in animal models where anti-CXCR4 monoclonal antibody was used
inhibited metastasis of lung cancer (Muller, Homey et al. 2001; Gangadhar, Nandi et al.
2010). Similar results of inhibition of invasiveness were also observed in pancreatic
cancers where CXCR4 were inhibited (Mori, Doi et al. 2004). Thus in many tumors the
metastatic cells are often hypoxic and this leads to HIF-1 regulation expression of
CXCR4. Therefore, CXCR4 is an important therapeutic target that can be regulated by
targeting HIF-1 pathway.
Therefore, these genes can be used as a benchmark in order to evaluate the regulation of
HIF-1α inducible transcription. Our goal is to design and synthesize small molecules that
are capable of disrupting the interaction between HIF-1α and p300 proteins inside the
nucleus of a cell thereby inhibiting the transcription of these downstream genes.
28
1.5 Targeting alpha-V beta-3 (α
v
β
3
) integrin receptors
The effects of extracellular matrix on cells are mediated by integrins. Integrins form a
large family of glycoproteins that are cell surface receptors which bind to components in
extracellular matrix and hence mediate cell adhesion, cytoskeleton organization and
activate intracellular signaling pathway (Stromblad and Cheresh 1996).
Each integrin consists of two transmembrane subunits - α and β. Currently, there are 18α
and 8β different subunits reported in mammals. These subunits associate in many
different combinations and in total 24 different combinations of α and β subunits are
known that bind to distinct subsets of ligands (Hynes 2002; Cai and Chen 2006).
Integrins upon binding to its ligand can initiate both chemical and mechanical signaling.
Most integrin signaling involves activation of focal adhesion kinase (FAK) and Src
family kinases (SFK) which eventually leads to phosphorylation of paxillin and
pI30CAS. α
v
β
3
along with α
1
β
1
, α
5
β
1
activates signaling pathway with help of SHC
adaptor protein. Integrin signaling regulates cell proliferation and migration in response
to cytokines and soluble growth factors (Parsons and Parsons 1997; Schlaepfer and
Hunter 1998; Cary, Han et al. 1999; Guo and Giancotti 2004).
Cancer cells become less dependent on growth factors and cytokines present in the
extracellular matrix for proliferation and migration. Despite that it has been shown that
expression of those integrins that increase the proliferation and migration of cells is
enhanced. Whereas those integrins that lower the proliferation of cells are decreased.
Integrin signaling is implicated in tumor growth and angiogenesis and also in metastasis
and invasiveness (Liu, Wang et al. 2008).
29
Endothelial cell activation is essential for tumor angiogenesis. The activation of
endothelial cell leads to dissolution of extracellular matrix, increased endothelial cell
migration and proliferation. This leads to tube formation, vessel formation and pruning
which forms vascular network. One of the most important factors in angiogenesis is
VEGFR2 and its ligand VEGF.
Figure 1.16 Integrin-mediated signaling pathway: α and β subunits of the integrins are
present on the cell membrane. The αβ integrins tend to cluster upon receiving signals in
the form of cytokines present in ECM (extracellular matrix). Since integrins themselves
do not have kinase domains so they use different proteins like Vinculin, Talin, Paxillin
which mediate the signaling pathway via FAK (Focal Adhesion Kinase). Different
integrins binds to their preferred ligands and mediate cell survival, cytoskeleton
reorganization, cell invasion and migration either directly through different kinase
pathways or inducing the transcription of different proteins needing to carry out the
changes.
30
Apart from this the regulation of cell adhesion also plays a very important role in
angiogenesis. Integrins which are heterodimeric, allosteric transmembrane glycoproteins
have ligand specificity with ligand overlap. α
v
β
3
integrins are receptors for vitronectin
found in extra cellular matrix. The α
4
β
1
and α
5
β
1
integrins are fibronectin receptors. α
1
β
1
and α
2
β
1
are collagen receptors while α
3
β
1,
α
6
β
1
and α
6
β
4
are laminin receptors and α
9
β
1
is
the osteopontin receptor (Hood and Cheresh 2002).
Of all integrins α
v
β
3
receptor is the most studied receptor (Brooks, Clark et al. 1994;
Wilder 2002; Kumar 2003). In proliferating vascular endothelial cells α
v
β
3
integrins are
overexpressed, indicating the positive correlation between angiogenesis and α
v
β
3
activity.
Since α
v
β
3
binds to vitronectin therefore another integrin α
v
β
5
was also thought to be
involved in angiogenesis. This led to a search for antagonists of α
v
β
3
and α
v
β
5
(Hodivala-
Dilke 2008). These antagonists inhibited the binding of α
v
β
3
and α
v
β
5
integrins to
vitronectin in in vitro assays and also inhibited neovascularization in in vivo animal
models including tumor angiogenesis and retinal angiogenesis induced by hypoxia.
Monoclonal antibody Vitaxin and RGD-mimetic Cilengitide are two drugs under clinical
trials that are β
3
antagonists (Gutheil, Campbell et al. 2000; Nabors, Mikkelsen et al.
2007). Interestingly these agents did not inhibit angiogenesis in tumor tissues. Cilengitide
did not work through the desired pathway but is does affect cell survival of glioma cells.
Although initial studies showed the importance of α
v
integrins in angiogenesis but some
recent studies have shown that the genetic ablation or inhibition of αv integrin alone is
not sufficient to suppress angiogenesis. Surprisingly, knockout of β
3
integrin also did not
suppress angiogenesis; on the contrary, an increase in the angiogenesis rate was observed.
According to these findings it seems that α
v
β
3
is as such not required for pathological
31
angiogenesis but in cases where β3 is knocked out proangiogenic effect is seen (Bader,
Rayburn et al. 1998; Reynolds, Wyder et al. 2002; Lacy-Hulbert, Smith et al. 2007).
Recent reports suggest that α
v
β
3
integrin have both proangiogenic and antiangiogenic
effects. α
v
β
3
integrin can indeed bind to different types of receptors some of which are
proangiogenic and others are antiangiogenic. Proangiogenic factors that bind to α
v
β
3
integrins are vitronectin, VEGFR2, fibronectin, Del1, ANGPTL3, CYR61, thrombin and
bone sialoprotein. Factors that have antiangiogenic effect and are known to bind to α
v
β
3
integrins are angiostatin, thrombospondin and tumstatin.
The reported study involving β
3
null mice was an interesting example of dual function of
the integrins where β
3
null mice showed enhancement of angiogenesis. This phenomenon
was traced to the overexpression of VEGFR2 observed in β
3
null mice which more than
compensated for the antiangiogenic effect (Reynolds, Reynolds et al. 2004;
Mahabeleshwar, Feng et al. 2006). Therefore, in case when α
v
β
3
antagonists were used
antiangiogenic effect was observed because the inhibition of α
v
β
3
did not increase the
protein levels of VEGFR2. Interaction and crosstalk between VEGFR2 and α
v
β
3
is very
important. A mutant mice model where two tyrosines (Tyr747 and Tyr459) in β
3
were
replaced with phenylalanines to obtain DiYF mutant showed inhibition of angiogenesis
similar to the use of antagonists (Mahabeleshwar, Feng et al. 2006). This shows that
phosphorylation of the two tyrosines are important for interaction of α
v
β
3
integrins with
VEGFR2 and other factors.
Integrins recycle themselves and VEGFR2 from cell surface to endocytic compartments
and back to cell surface. This endocytic recycling of integrins is believed to be another
32
factor that may be responsible for the dual function of α
v
β
3
integrins both as pro and anti-
angiogenic factor (Mahabeleshwar, Feng et al. 2007). Due to this “moody” nature of α
v
β
3
integrins it is less fruitful to find inhibitors from them. A better approach is to find
ligands for the α
v
β
3
integrins and use them to deliver the payload of a drug or other
desired molecule.
Since vitronectin and other ligands that bind to integrins have an RGD tripeptide as a part
of their binding domain so many peptides and peptidomimetics have been designed in
order to work as integrin antagonists. All integrins share certain degree of homology and
some integrins share more homology with each other than others. The α
v
β
3
integrins share
maximum homology with α
v
β
5
integrins. Some important short peptides having good
binding to integrins and specifically to α
v
β
3
integrins are shown in Figure 1.18.
Figure 1.17 Structures of three RGD based peptides that have been designed by different
Pharma Companies. All three structures are cyclic peptides in order to improve their
stability in vivo towards proteolytic damage.
33
The RGD based cyclic peptides shown in Figure 1.18 are much more stable towards
proteolytic degradation in vivo as compared to their non-cyclic counterparts. The
conformational restriction obtained by ring closure along with the use of D-
phenylalanines (f stand for D-phenylalanine) not only increases the in vivo half-life but
also improves the bioavailability and binding of the cyclic peptides to the integrins (Eble
and Haier 2006). Cilengitide is an α
v
specific inhibitor used to target integrins as
antagonists. It is a drug under trial for glioblastoma (Liu, Wang et al. 2008). As such
antagonists for α
v
β
3
integrins are potential therapeutics for cancer, osteoporosis, diabetic
retinopathy and restenosis.
Figure 1.18 Non-peptidic small molecule antagonists that bind integrins.
Another important class of integrin antagonists is non-peptide small molecules that
mimic the biological actions of the natural parent peptide. A few important examples of
such peptidomimetics are shown in Figure 1.19 (Nicolaou, Trujillo et al. 1998; Liu,
Wang et al. 2008). The benefit of peptidomimetics over cyclic peptides lies in their better
stability towards proteolysis and hence much improved oral bioavailability. SC-68448 is
34
an α
v
β
3
specific integrin antagonist from Monsanto. It is 100 fold more potent inhibitor of
α
v
β
3
integrin as compared to α
IIB
β
3
integrin which is very similar structurally to α
v
β
3
integrin (Carron, Meyer et al. 2000). Duggan et al. (Duggan, Fisher et al. 1996) and
Merck reported an α
v
β
3
specific antagonist (Figure 1.19) which showed more than 50 fold
specificity towards α
v
β
3
compared to its homologically closest integrin α
v
β
5
.
Using these ligands to target cells and tissues having over expression of integrins,
specifically α
v
β
3
integrins is an attractive strategy for targeted delivery of drugs and or
imaging reagents like dyes, contrast reagents etc. We endeavored to design and
synthesize α
v
β
3
integrin specific conjugates that deliver a payload of boron atoms to the
cells and tissues that express α
v
β
3
integrins so as to perform boron neutron capture
therapy (BNCT) and Boron-MRI in vivo.
35
Chapter 2: Development of facile synthetic strategies en route to compounds bearing
ETPs and indoles
36
2.1 Introduction to synthetic strategies for construction of ETP rings
As discussed in Chapter 1, ETPs (epidithiodiketopiperazines) belong to an important
class of molecules which show promise as potential therapeutic agents for several
pathologies. Thus, their facile and efficient synthesis is of prime importance. The
presence of a torsionally-strained disulfide bond in a bicyclic frame makes preparation of
ETPs rather challenging. Although several ETP-bearing natural products, known as
sporidesmins, were isolated around the 1950s (Waksman and Bugie 1944), the first non-
stereoselective total synthesis of a sporidesmin A was reported only in 1973 by Kishi et
al. (Kishi, Nakatsuk.S et al. 1973).
Figure 2.1 Classic strategies for synthesis of the ETP core. In both the Trown and the
Hino approaches, the disulfide bond is formed in the last step.
37
the ETP core has been prepared earlier by Trown's (Trown 1968) or Hino's (Hino and
Sato 1971) groups (Figure 2.1), but introduction of the disulfside bridge made the
molecule unstable under reducing, oxidative or basic conditions and any further
synthetic manipulations became very difficult. In order to circumvent this problem, Kishi
and coworkers protected the thiols as the anysaldehyde thioacetal, (Figure 2.2).
An unsymmetrical diketopiperazine is the starting point of the synthesis. Initial
bromination of the α-position on the diketopiperazine ring is followed by SN2
displacement by thioacetate anion. Next, a trithiane derivative is used to introduce the
second sulfur, which is bridged with the first sulfur in the syn orientation by anisaldehyde
in presence borontrifluoride-etherate. The dithioacetal protecting group survives the
subsequent steps, introducing a substituted indole and finally forming four fused rings of
sporidesmin A. m-Chloro perbenzoic acid and BF
3
-etherate are used in the final step in
order to remove the thioacetal group and obtain the syn disulfide bridge.
38
Figure 2.2 Total synthesis of Sporidesmin A by Kishi and coworkers.
The synthesis of sporidesmin A was a remarkable feat in development of a strategy to
obtain ETPs. After this report, a few other approaches to sporidesmins have been
reported by the same and other groups. The first total synthesis of sporidesmin A,
although very elegant, was non-stereoselective and low yielding. The first stereoselective
39
total synthesis of biologically relevant bis-ETP compound, dideoxyverticillin, was first
reported in 2009 by Movassaghi group (Kim, Ashenhurst et al. 2009) (Figure 2.3).
Figure 2.3 Total synthesis of dideoxyverticillin by Movassaghi and coworkers.
In thier approach, hydroxyl groups were used as precursors for sulfurs which were
introduced enantioselectively with K
2
CS
3
-TFA using the inherent facial bias of the
crowded core of the molecule. This approach would fail if the diketopiperazines ring is
40
attached to the core by a flexible linker, which is the case for one of the rings in chetomin
(Figure 2.4)
Figure 2.4 Chetomin structure and unique structural features. Chetomin as a
representative example of sporidesmins that have a common feature of two sulfurs in syn
orientation forming a disulfide bond in the ETP ring (shown in red) and most
sporidesmins also have an indole system attached to the ETP ring (shown in blue).
To overcome this problem, novel strategies for facile, efficient and stereoselective
introduction of sulfurs and indoles into the ETP rings are needed, of which catalytic ones
are perhaps the most highly desired.
2.2 Brief overview of important developments in the field of sulfenylation and
organocatalytic sulfenylation
2.2.1 Approaches towards asymmetrical sulfenylation. Formation of a heteroatom-
carbon bond, and especially the sulfur-carbon bond, plays a very important role in
organic synthesis (Trost 1978; Trost 1978). Like with any other bond formation, the
ability to form the sulfur-carbon linkage stereoselectively is very much desirable. One of
the classical approaches to asymmetric bond formation is the use of the Evan's
oxazolidinone auxiliary to direct the spatial orientation of the sulfur addition. In a report
41
from Paterson et al (Alexander and Paterson 1985), valine-oxazolidinone derivatives
were converted into enolates and reacted with phenylsulfenyl chloride and phenyl
disulfide to give sulfenylated products in decent chemical and diastereoselective yield
(Figure 2.5).
Figure 2.5 Oxazolidinone approach for asymmetric sulfur addition.
The oxazolidinone approach makes use of the fact that there is a facial bias for sulfur
addition due to the presence of a chiral oxazolidinone ring. Another approach using a
similar concept of inherent facial bias is to use a chiral sulfenylating reagent. 3-
Phenylsulfenyl-2-(N-cyanoimino) thiazolidine and its optically active sulfenylating
analog provided for a stereoselective sulfenylation of at least some substrates (Figure 2.6)
(Tanaka, Azuma et al. 2000).
42
Figure 2.6 Thiazolidine-derived reagent for sulfenylation of ketone derivatives.
Yields in excess of 80% were observed when derivatives of 3-Phenylsulfenyl-2-(N-
cyanoimino)thiazolidine were used as sulfenylating regent. A major drawback of this
sulfenylation strategy is that the sulfenylating reagent is very tedious to synthesize. Chiral
hydrazones have also been used to direct the sulfenylating reagent to obtain asymmetrical
addition of sulfur (Figure 2.7) (Hendrix, Seftor et al. 2003).
Figure 2.7 An optically active hydrazine is condensed with a ketone followed by
stereoselective sulfenylation with alkyl disulfide. Oxidation or acidolysis converts the
hydrazone back into ketone.
A similar approach using an amine instead of a hydrazine was reported where an
optically active ketone was condensed with an α-amino ester followed by sulfenylation
(Figure 2.8) (Ruoslahti 2002). This approach has yields in excess of 50% and de above
80%.
43
Figure 2.8 Use of optically active ketone is condensed with α-amino ester followed by
sulfenylation. Optically active ketone is condensed with amino ester and sulfenylated by
methyl disulfide under conditions in which sulfenylation occurred of the α-amino ester
rather than ketone.
The main drawback of the above mentioned approaches is poor atom economy due to the
use of equimolar chiral auxiliaries.
2.2.2 Organocatalytic sulfenylation reactions. Recently, several approaches that make
use of organocatalysts for sulfenylation of substrates, have been reported. As the first
example of organocatalytic sulfenylation, a method of sulfenylation of ketones and
aldehydes using pyrrolidine trifluoromethane sulfonamide as the catalyst, was reported
by Wang et al (Wang, Li et al. 2004).
Figure 2.9 Pyrrolidine-based catalyst for sulfenylation of ketones. Cyclohexanone is
sulfenylated with N-(phenylthio) phthalimide or diphenyldisulfide or dimethyldisulfide in
the presence of 20-30 mol% of pyrrolidine trifluoromethane sulfonamide.
N-(Phenylthio)phthalimide sulfenylating reagent showed the best results while
dimethyldisulfide and diphenyldisulfide gave poor yields. Various pyrrolidine-based
44
organocatalysts were studied as potential catalysts for sulfenylation. Pyrrolidine-
trifluoromethane-sulfonamide is the catalyst that showed maximum yields along with N-
(phenylthio)phthalimide sulfenylating reagent.
Figure 2.10 Different pyrrolidine based organocatalysts used for catalytic sulfenylation.
a) Four examples of pyrrolidine based organocatalysts studied for sulfenylation of
cyclohexanone. Out of these, pyrrolidine-trifluoromethane-sulfonamide gave the highest
yields. b) Various ketones and aldehydes (R1 = H) gave decent yields of sulfenylated
products upon reaction with N-(phenylthio)phthalimide sulfenylating reagent and
pyrrolidine-trifluoromethane-sulfonamide as the catalyst.
Different ketones and aldehydes were probed in order to find out their reactivity towards
the sulfenylating reagent and organocatalysts (Figure 2.10). Cyclic ketones showed more
reactivity towards the sulfenylating conditions as compared to aldehydes. In the case of
aldehydes, multiple α-sulfenylations could be observed.
Jorgensen et al. reported organocatalyzed α-sulfenylation of aldehydes (Marigo, Wabnitz
et al. 2005). They synthesized an array of sulfenylating reagent all of which had S-benzyl
45
groups as the sulfur bearing moiety (Figure 2.11). The imidazole derivative turned out to
be very unstable and the phthalimide and succinimide reagents reacted very slowly.
However, 1-benzylsulfanyl-1,2,4-triazole (S5) turned out to be the most reactive yet
sufficiently stable.
Figure 2.11 S-benzyl based sulfenylating reagents. The imidazole-based sulfenylating
reagent is very unstable, whereas the 1,2,4-triazole-based sulfenylating reagent turned out
to be most reactive sulfenylating reagent under the reaction conditions.
The catalysts used in this study were the L-proline-derived or pyrrolidine-based catalysts.
Authors first carried out a study in order to find a suitable catalyst from eight different
pyrrolidine-based catalyst. In a model system, isovaleraldehyde was used as the substrate
and 1-benzylsulfanyl-1,2,4-triazole was used as the sulfenylating reagent.
Figure 2.12 Model reaction for the catalyst optimization study.
46
Under the optimized conditions a number of aldehydes were reacted with triazole-derived
sulfur donor reagent and the catalyst-of-choice (Figure 2.13). In Figure 2.13a, the R
group was chosen to be i-Pr and Bn in order to obtain higher yield and enantioselectivity.
To facilitate work-up, the aldehydes could be reduced into alcohols with sodium
borohydride. Figures 2.13b and 2.13c show the versatility of this strategy. Reductive
amination was carried out with dibenzylamine under the reducing condition of
triacetoxyborohydride.
Figure 2.13 Sulfenylation strategy with 1-benzylsulfanyl-1,2,4-triazole. a) Different
aldehydes were sulfenylated under the optimized conditions. R’ as H gave higher yields.
R group as i-Pr and Bn were optimal for good enantioselectivity. b) reductive amination
of the sulfenylated product showed the versatility of the strategy. c) Cleavage of S-Bn
group opens the possibility of obtaining a thiol.
47
This strategy can also yield thiol after reduction of the aldehyde into alcohol
subsequently protection with tert-butyldimethylsilyl (TBDMS) group. Treatment with
Na/NH
3
cleaves the benzyl group yielding the thiol-bearing molecule.
Another report by Jorgensen et al. was focused on the use of cinchona alkaloids as
organocatalysts (Sobhani, Fielenbach et al. 2005). The mechanism of action for the
cinchona alkaloids is believed to be different from that of the pyrrolidine-based catalysts.
Pyrrolidine-based catalysts are thought to form an imine adduct in-situ at the carbonyl
group of the substrate, catalyzing the addition of the alkyl or sulfur group from a specific
spatial orientation. Cinchona alkaloids, on the other hand, act as chiral bases which form
an adduct after deprotonating the substrate. This deprotonated substrate adduct imparts
the facial selectivity of the addition.
Figure 2.14 Commercially available cinchona alkaloids used as organocatalysts.
48
For their study Jorgensen et al. utilized different aryl-sulfanyl[1,2,4]triazoles as
sulfenylating reagents. In their system, 1-benzylsulfanyl[1,2,4]triazole showed the best
results.
Figure 2.15 β-Ketoester as a model substrate sulfenylated by 1-
benzylsulfanyl[1,2,4]triazole with a cinchona alkaloid as a catalyst.
Various cinchona alkaloids were tested as organocatalysts (Figure 2.14). In order to find
the most suitable catalyst and appropriate solvent and temperature for sulfenylation, β-
ketoester was used as a model system. Three different alkyl 2-oxocyclopentane-
carboxylate substrates were used for optimizing the reaction conditions. It was found that
the bulkier t-Bu group as R is suitable in order to achieve enantioselectivity.
Dichloromethane as a solvent gave superior yields while toluene was suitable for
achieving higher enantioselectivity of the reaction. Authors noticed that (DHQD)
2
PYR
was a better catalyst in order to achieve higher enantioselectivity.
Figure 2.16 Example of cyclic β-ketoesters and lactams that were substrates for
organocatalytic sulfenylation.
A series of different cyclic β-ketoesters and lactams were sulfenylated using three
different triazole based sulfenylating reagents. Good yields and enantioselectivity was
achieved through the use of this methodology.
49
Figure 2.17 Ti[TADDOL(ato)]-complex used as a catalyst in sulfenylation reactions with
β-ketoesters using phenylsulfenyl chloride as an electrophilic sulfenylating reagent.
Togni et al. reported the use of Ti[TADDOL(ato)]-complex as a catalyst in sulfenylation
of β-ketoesters (Srisailam and Togni 2006). Ti[TADDOL(ato)]-complexes had been
showed in the past by the authors to catalyze halogenation of substrates (Hintermann and
Togni 2000). Non-cyclic β-ketoesters were used as substrates with variable R
1
, R
2
, R
3
groups. In the course of optimization studies, the R1 group was the most important
substituent which affected the yield. The less bulky group like methyl as R
1
gave high
yields, whereas more sterically bulky groups deduced yields, presumably by imparting
greater facial bias of the transition for the approaching sulfenylating reagent. The
mechanism of action of the Ti[TADDOL(ato)]-complex was explained by its transition
where a bidentate complex with the two oxygens of the β-ketoester with titanium is
responsible for reactivity and stereoselectivity towards the sulfenylating reagent. By
employing this strategy and readily available phenylsulfenyl chloride the authors were
able to achieve high yields up to 95% and e.e. up to 88%.
50
Figure 2.18 Sulfenylation of β-keto phosphonates using pyrrolidine-based organocatalyst
and N-(phenyl-thio)phthalimide as sulfenylating reagent.
Recently Cheng et al. reported α-sulfenylation of β-keto phosphonates using N-(phenyl-
thio)phthalimide as the source of electrophilic sulfur and pyrrolidine-based
organocatalysts α,α-diaryl-L-prolinols (Lin, Fang et al. 2011). High yields and e.e. were
obtained when catalyst bearing 3,5-(CH
3
)
2
C
6
H
5
substitution were chosen. The substrates
used were aromatic cyclic β-keto phosphonates. Reaction was done at 0 °C and gave high
stereoselectivity. Using this method the sterically bulky proloinol catalysts were able to
form the C-S bond stereoselectively with good yields.
The reports mentioned above are key recent works that have been carried out in the field
of organocatalytic formation of C-S bond. Most of these methodologies rely on relatively
reactive model substrate, such as aldehydes, ketones, β-keto esters and β-keto
phosphonates. Nevertheless, they provide good insights into the recent applications of
organocatalysis in sulfenylationof various substrates.
51
2.3 Organocatalytic sulfenylation of diketopiperazines
2.3.1 Initial efforts of organocatalytic sulfenylation of diester-diketopiperazines. The
use of organocatalysis for formation of heteroatom-carbon bond is a relatively new field.
The examples discussed above are important developments in the field of organocatalytic
sulfenylation. They all follow the use of model substrates like aldehydes, ketones, β-keto
carbonyls etc. Although these examples are of great methodological importance, they are
not applicable to sulfenylation of much less reactive diketopiperazines. The low reactivity
of diketopiperazine is due to stereoelectronic effects that increase the pKa of the protons
at the alpha position. These factors make the formation of anion at the α position of the
DKP ring and subsequent addition of electrophilic sulfur difficult.
Figure 2.19 Organocatalytic sulfenylation of diethylester-DKP substrate using
(DHQD)
2
PYR as catalyst. The two sulfurs in the product are in the anti-orientation with
respect to each other.
Our initial attempts towards organocatalytic sulfenylation were focused on diester-
diketopiperazine. This diketopiperazine is relatively easily synthesized by condensation
of 2-methylamine-diethylmalonate. Cinchona alkaloids were used as organocatalysts
52
instead of proline-based compounds, based on the considerations of reactivity (vide
supra, p. 51).
Figure 2.20 Schematic representation of the mechanism leading to anti-additions of
sulfur groups on the DKP ring. The first -SPMB group blocks the access of one of the
face of the DKP as result the second -SPMB addition takes place from the other face of
the DKP ring.
Therefore in our initial study we employed the use of (DHQD)
2
PYR to carry out
sulfenylation of the diethylester-diketopiperazine. The two sulfenylating reagents used
were 1-benzylsulfanyl-[1,2,4]-triazole and 1-(p-methoxybenzyl)sulfanyl succinimide.
Although the sulfur addition was done in good yields both sulfurs added anti- with regard
to each other. This result was undesirable as our goal was to obtain the sulfur addition in
a syn-fashion in order to convert the adduct to ETP. One plausible explanation for anti-
53
orientation is that after addition of the first sulfur the aryl group on one face of the DKP
is blocked. As a result, the second sulfur will essentially be added from the opposite face
regardless of the stereochemistry of the first addition. When the second sulfur-bearing
group is added the stereoselectivity is lost and an achiral molecule is obtained.
Therefore, in order to develop a stereoselective syn sulfenylation strategy that may lead to
ETP synthesis it was decided to use monoester-diketopiperazines as substrates for the
subsequent studies. Monoester-DKPs, being unsymmetrical molecules are more
challenging to synthesize as compared to their diester counterparts.
2.3.2 Synthesis of diketopiperazine substrates. In order to prepare monosubstituted
diketopiperazine, piperazine-2,5-dione (glycine anhydride) was the reactant of choice.
The nitrogens on the glycine anhydride which is a diketopiperazine can be alkylated or
arylated, thereby introducing the substitutuent of choice.
Figure 2.21 Piperazine-2,5-dione is a suitable starting material for synthesis of
substituted DKPs. The nitrogens at positions N1 and N4 can be readily alkylated or
arylated to give N-substituted DKP.
Two new methods were developed in order to synthesize the monoethylester-DKP. The
first method involved converting 1,4-dimethylpiperazine-2,5-dione to 3-
(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione by reacting it with diethyl carbonate
in presence of potassium t-butoxide in THF at -78 ˚C under nitrogen.
54
Figure 2.22 Synthesis of 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione from 1,4-
dimethylpiperazine-2,5-dione using diethyl carbonate as the acylating reagent.
The second method involves formation of ethyl-2-(N-methyl, N-benzylamino)-3-[(2-
benzyloxycarbonyl)methylamino]-3-oxopropanoate from reaction between (N-methyl, N-
benzylamino)malonate and benzyl sarcosinate p-toluenesulfonate with N,N
diisopropylethylamine (DIEA) as a base in toluene under reflux. In the second step upon
treatment with palladium on activated charcoal under hydrogen atmosphere ethyl-2-
(methylbenzylamino)-3-[(2-benzyloxycarbonyl)methylamino]-3-oxopropanoate is
converted to 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione 11.
Figure 2.23 Two-step novel approach for synthesis of 3-(ethoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione from (N-methyl, N-benzylamino)malonate and benzyl
sarcosinate p-toluenesulfonate.
55
This result was somewhat surprising because the reducing conditions of the reaction was
expected to only cleave the benzyl group of the amine but serendipitous cyclization along
with benzyl cleavage afforded DKP ring, saving a step that would have been required for
cyclization. The process is high yielding and benzyl cleavage along with cyclization
occurs in 92% yield.
The N1 position is near the α position of the DKP ring. Thus, it plays an important role in
determining the reactivity at the α position as the substituent at N1 can impart a very
important steric and electronic effect in sulfenylation reaction. Thus, the first goal was to
synthesize a DKP without any substituent at N1 position and the strategy developed for
the synthesis of 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione 14 following the
method for preparation of 11.
Figure 2.24 Synthesis of compound 14. Synthesis of 3-(ethoxycarbonyl)-1-
methylpiperazine-2,5-dione from diethyl (dibenzylamino)-malonate and benzyl
sarcosinate p-toluenesulfonate followed by one step benzyl cleavage cyclization and
removal of second benzyl group.
Instead of (N-methyl, N-benzylamino)malonate, (dibenzylamino)malonate was used with
the hope that after the cyclization of ethyl-2-(dibenzylamino)-3-[(2-benzyloxy-carbonyl)
56
methylamino]-3-oxopropanoate the second benzyl will also cleave under reducing
conditions yielding unsubstituted at N1 DKP. The first step was formation of the
intermediate ethyl-2-(dibenzylamino)-3-[(2-benzyloxy-carbonyl) methylamino]-3-
oxopropanoate from diethyl (dibenzylamino)malonate and benzyl sarcosinate p-
toluenesulfonate with DIEA as base in toluene under reflux. The second step involves a
benzyl cleavage, cyclization and cleavage of second benzyl group all in one step under
reducing conditions of Pd/C in ethanol under nitrogen at room temperature. Ethyl-2-
(dibenzylamino)-3-[(2-benzyloxy carbonyl) methylamino]-3-oxopropanoate (14) was
obtained in good overall yield.
Alkylation of the N1 on 14 was not feasible as any attempt lead to alkylation on α
position. Thus it was decided to try acylating this position with different groups in order
to study the stereoelectronic effect of a carbonyl group at this position. When 14 reacted
with acetyl chloride under anhydrous conditions in presence of DMAP, the N1 position
was acetylated.
57
Figure 2.25 Acylation of 14 with acetyl chloride, benzoyl chloride and tosyl chloride
yielding DKPs 15, 16 and 17, respectively.
Similarly, reaction of 14 with benzoyl chloride resulted in benzoylation of N1 in 14. Both
acylations occurred with more than 80% yield. Under these conditions tosylation of N1 of
14 was also achieved, although the yield was less than 50% probably due to steric
reasons.
The type of ester group present at the α position is another important consideration.
Hence, methyl and t-butyl esters were also prepared and tested in addition to ethyl ester.
58
Figure 2.26 Synthesis of 3-(tert-butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione from
sarcosine anhydride.
The t-butyl analog 3-(tert-butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione was
prepared from sarcosine anhydride by acylating it with Boc anhydride in THF at -78 ˚C.
with LHMDS used as the base. The alkyl groups on the nitrogens N1 and N4 varied
between methyl, ethyl and benzyl. For example, 3-(ethoxycarbonyl)-1,4-
diethylpiperazine-2,5-dione was prepared from 1,4-diethylpiperazine-2,5-dione by
reacting it with diethyl carbonate and LHMDS as a base.
Figure 2.27 Synthesis of 3-(ethoxycarbonyl)-1,4-diethylpiperazine-2,5-dione and 3-
(ethoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione.
Similarly, 3-(ethoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione was prepared by reacting
1,4-dibenzylpiperazine-2,5-dione with diethyl carbonate and LHMDS as a base. A
59
sterically bulky DKP was prepared with benzyl groups at the N1 and N4 positions and
the t-Butyl ester at the α position. The synthesis of 3-(tert-butoxycarbonyl)-1,4-
dibenzylpiperazine-2,5-dione was achieved by reacting 1,4-dibenzylpiperazine-2,5-dione
with Boc anhydride in THF at -78 ˚C and LHMDS as a base.
Figure 2.28 Synthesis of 3-(tert-butoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione.
2.3.3 Organocatalytic sulfenylation of monoester-diketopiperazines. Initial attempts of
sulfenylation was done with 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione and 1-
benzylsulfanyl[1,2,4]triazole. The reaction resulted in 46% yield in the presence of
quinine (15 mol%) in dichloromethane at room temperature for 120 h. The first goal was
to analyze and increase the efficiency of sulfenylation and reduce reaction time. Also,
prior to optimization of stereoselectivity it was of prime importance to obtain high yields
of sulfenylation.
Figure 2.29 Organocatalytic sulfenylation of monoester-DKP 3-(ethoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione with 1-benzylsulfanyl[1,2,4]triazole.
60
In order to reach the upper limit on reactivity of the DKP substrate under the typical
conditions of sulfenylation, we first utilized N1 unsubstituted monoester-DKP 3-
(ethoxycarbonyl)-1-methylpiperazine-2,5-dione that gave product in high yield (93%).
Figure 2.30 Sulfenylation of 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione with 1-
benzylsulfanyl[1,2,4]triazole in dichloromethane with quinine as the organocatalyst.
This result suggested that sterics and electronics of DKP ring impact its reactivity
towards sulfur electrophile. Next, an optimization study was done using a panel of
monoester-DKPs with different substituent at N1 position, i.e., acetyl, benzoyl and tosyl
group along with the methylated and unsubstituted DKP. The effects of temperature,
solvent and catalyst type and loading on the yield of the reaction were also assessed.
When N1 is methylated the yields are low regardless of the type of solvent used. The use
of sulfenylating reagent 22a has been shown to give high yields when used with ketones,
aldehydes etc. Hovewer, with N1 alkylated DKPs the yields were moderate to low. In
order to obtain higher yields larger excess of sulfenylating reagent was used. We started
with (DHQD)
2
PYR as the catalyst of choice based on the recent reports of high yields
and good stereoselectivity in the reactions that involved sulfur donor electrophiles.
Initially, when 10 equivalents of 22a were used along with sterically less crowded
catalyst dihydroquinine (9d) the yields increased marginally to 55%. When the N1
unsubstituted DKP was used the yields obtained were 90% even when only 2 equivalents
61
of sulfenylating reagent 22a were used in dichloromethane as a solvent. Moreover, in
dichloromethane reaction was complete in 24 hours as opposed to 120 hours required for
N1 methylated DKP. Interestingly, when toluene was used the higher yields of 95% were
obtained, but reaction required 48 hours. Reactions with N1 unsubstituted DKPs were
also explored at lower temperatures.
Table 2.1 Organocatalytic α-sulfenylation of N1 substituted monoester DKPs with 1-
benzylsulfanyl[1,2,4]triazole 22a and cinchona alkaloids 9a–d as organocatalysts.
R
substituent
Catalyst Solvent Temp. (˚C) Time (h) Equiv. of 22
Yield
(%)
a
Me 9a CH
2
Cl
2
RT 120 5.0 46
Me 9a Toluene RT 120 5.0 44
Me 9d CH
2
Cl
2
RT 120 10.0 55
H 9d CH
2
Cl
2
RT 48 2.0 93
H 9c Toluene RT 24 2.0 95
H 9c Toluene -10 24 1.5 90
H 9c Toluene -78 24 1.5 20
H 9a Toluene -10 24 2.0 94
H 9b Toluene -10 24 2.0 91
CH
3
CO 9a Toluene RT 72 2.6 10
b
CH
3
CO 9a Toluene -10 72 2.6 10
b
PhCO 9a Toluene RT 72 2.6 10
b
Ts 9a Toluene RT 72 2.6 c
a Yield of purified product after chromatographic separation. b Compound 3-
Benzylsulfanyl-3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione (25) was obtained as
the product. c No product was formed, as observed by
1
H-NMR.
62
At -10 ˚C in toluene 90% yield was achieved with only 1.5 equivalents of sulfenylating
reagent in 24 hours. The increased stability of the sulfenylating reagent against
decomposition at lower temperature permitted to reduce its excess. When the temperature
of the reactions was decreased to -78 ˚C the yields decreased to 20% suggesting that at
this temperature reactivity of the substrate is too low. Quinine (9c) was the organocatalyst
that gave best yields under these conditions.
When the N1-acylated and N1-benzoylated DKPs (R = CH
3
CO, PhCO) were used the R
group in the product was removed, giving the same products as N1 unsubstituted DKP.
The thiolate anion formed in trace amounts during the reaction was one possible
explanation of this result. Subsequent attack on the acetyl or benzoyl group on the N1
cleaved the acyl group. The yields obtained were low (~10%) and hence, this reaction
was not pursued further. In the case where tosyl group was present in the N1 position no
product was formed.
Figure 2.31 Cinchona alkaloids used as organocatalysts in this study.
63
2.3.4 Novel sulfenylating reagent 1-phenylsulfanyl[1,2,4]triazole and its
organocatalytic sulfenylation of DKPs. Based on the analysis of the transition state it
was acknowledged that benzylsulfanyl[1,2,4]triazole may not be the best reagent due to
steric and electronic reasons. It was decided to prepare another sulfenylating reagent 1-
phenylsulfanyl[1,2,4]triazole where the phenyl group will be in place of the benzyl
group. Although a few substituted phenylsulfanyl[1,2,4]triazoles have been reported, no
synthesis of unsubstituted sulfur donating reagent 1-phenylsulfanyl[1,2,4]triazole was
known. The synthesis of 1-phenylsulfanyl[1,2,4]triazole was achieved by reacting 1,2,4-
triazole with diphenyl disulfide and sulfuryl chloride in presence of triethylamine at 0 ˚C.
Figure 2.32 Synthesis of novel sulfenylating reagent 1-phenylsulfanyl[1,2,4]triazole.
The new sulfenylating reagent 22b gave products in high yield even with the bulky t-
butyl ester DKP substrate. The yields of sulfenylation were 79% in dichloromethane
with quinine as the catalyst. Interestingly, when toluene was used as the solvent the yield
increased to 91%. Even bulkier catalysts (DHQD)
2
PYR (9a) and (DHQD)
2
PHAL (9b)
gave products in good the yields ranging from 60% to 65% at room temperature in
toluene. At lower temperatures yields were greater than 95% .using quinine as catalyst
and toluene and benzene-toluene mixture (3:1) as the solvent. The high stability of the
reagent within the wide range of temperatures is one possible reason for achieving higher
yields. Lowering reaction temperature further aids in reducing decomposition of the
sulfenylating reagent.
64
Table 2.2 Optimization of the reaction conditions for sulfenylation of 3-(tert-
butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione with 1-phenylsulfanyl[1,2,4]triazole
(22b)
Solvent Temp (˚C) Catalyst Equiv of 22b Yield (%)
a
CH
2
Cl
2
RT 9c 2.0 79
Toluene RT 9a 2.0 60
Toluene RT 9b 2.0 65
Toluene RT 9c 2.0 91
Toluene 0 9c 3.0 97
PhH/PhMe (3:1) -10 9c 3.0 96
a Yield of purified product after chromatographic separation.
In order to assess the substrate scope, different diketopiperazine substrates were used.
Table 2.3 illustrates different examples where the substituent on N-1 (R
1
), N-4 (R
2
) and
ester group (R
3
) is systematically varied. The novel sulfenylating reagent 22b gave high
yields of >90% when R
3
was varied between methyl, ethyl and t-butyl in benzene-toluene
mixture (3:1) at -10 ˚C. The substituents on R
1
and R
2
were also varied to see the effect
of different groups on sulfenylation. It was noticed that on changing the groups on N-1
and N-4 from methyl to ethyl and benzyl the yields decreased when 3.0 equivalents of
sulfur reagent were used. When 8.0 equivalents of sulfur reagents were used the yields
increased to more than 90% at room temperature in toluene. The N-1 unsubstituted
diketopiperazines were obtained in good yields.
65
Table 2.3 Scope of α-sulfenylation of trisubstituted piperazine-2,5-diones with reagent
22b and 10 mol % of organocatalysts 9a–c.
R1 R2 R3 Catalyst Solvent
Temp
(˚C)
Equiv. of
22b
Time
(h)
Yiel
d
(%)
a
Me Me Me 9c
PhH/PhMe
(3:1)
-10 3.0 18 99
Me Me Et 9c
PhH/PhMe
(3:1)
-10 3.0 30 99
Me Me
t-
Bu
9c
PhH/PhMe
(3:1)
-10 3.0 48 96
Et Et Et 9c Toluene RT 3.0 60 77
Et Et Et 9c Toluene RT 8.0 48 94
Bn Bn Et 9c Toluene RT 3.0 60 75
Bn Bn Et 9c Toluene RT 8.0 48 95
H Me Et 9c Toluene RT 3.0 16 81
Bn Bn
t-
Bu
9c Toluene RT 10.0 72 52
Bn Bn
t-
Bu
9a Toluene RT 2.0 60 5
Bn Bn
t-
Bu
9b Toluene RT 2.0 60 5
a Yield of purified product after chromatographic separation.
In order to determine the efficiency of sulfenylating reagent 22b the most sterically
hindered diketopiperazine with bulky benzyl groups on N-1 and N-4 and t-butyl ester at
the C-3 position was tested. With quinine as the organocatalyst, 52% yield was achieved.
It is worth noting that no product was observed with the DKP when 22a sulfenylating
reagent was used.
66
The data in Table 2.3 shows that the novel sulfenylating reagent 1-
phenylsulfanyl[1,2,4]triazole (22b) is superior in efficiency as compared to 1-
benzylsulfanyl[1,2,4]triazole (22a).
2.3.5 X-ray crystal structure of tert-Butyl-1,4-dimethyl-3,6-dioxo-2-(phenylthio)
piperazine-2-carboxylate (27). In order to better understand the structural basis of the
reactivity of alkylated DKPs in sulfenylation reactions, an x-ray crystallographic analysis
was performed. However, the process of crystallization of several sulfur containing DKP
products turned out to be very challenging as most of the aprotic solvents like ether,
chloroform, ethyl acetate etc. did not result in crystallization both using vapor diffusion
and gradual solvent evaporation techniques. Hovewer, dissolution of tert-Butyl-1,4-
dimethyl-3,6-dioxo-2-(phenylthio)piperazine-2-carboxylate (27) in ethanol followed by
rapid evaporation of the some solvent at 45 ˚C for 30 seconds and thereafter subjecting
the vessel to ice cold conditions resulted in a formation of the x-ray quality crystals. The
ORTEP diagram for compound 27 with thermal ellipsoids drawn at 50% probability level
is shown in Figure 2.32.
67
Figure 2.33 ORTEP view of the structure of tert-Butyl-1,4-dimethyl-3,6-dioxo-2-
(phenylthio)piperazine-2-carboxylate (27) with anisotropic displacement parameters at
30% probability level. The R enantiomer is shown.
The crystal structure shows that one face of the DKP is blocked by the S-phenyl group.
This phenyl group has limited degree of rotation about the C10-S bond. The compound
has crystallized in the centrosymmetric orthorhombic space group Pbca and the crystal
packing shows efficient stacking of the phenyl and diketopiperazine rings in the solid
state. It is hypothesized that this stacking may provide additional stabilization in the
transition state of the sulfenylation reaction.
2.3.6 Enantioselectivity of organocatalytic sulfenylation. After optimization of the
reaction conditions in order to obtain high yields we proceeded to study the
enantioselectivity of the process. Enantiomeric excess (e.e.) was measured
simultaneously for each reaction carried out by using chiral lanthanide shift reagent
Eu(hfc)
3
.
68
Figure 2.34 Structure of chiral lanthanide shift reagent Eu(hfc)
3
.
It a typical experiment, chiral lanthanide shift reagent Eu(hfc)
3
was added to the NMR
tube having the sulfenylated DKP and in 1H NMR the peaks of the ester group would
split into two peaks. The area of the two peaks have the relative ratio of two different
enantiomers present in the sample. Optimization studies to study the enantioselectivity
were done to in order to find the best conditions including the best catalyst and best sulfur
donating reagent. As seen from the Table 2.4 the sulfenylating reagent 22a gives better
enantioselectivity but low yield whereas reagent 22b gives low enantioselectivity and
high yield. Overall, the bulkier ester t-butyl groups tend to give higher enantioselectivity
than ethyl groups.
We found that the mixture of benzene-toluene (3:1) mixture at -10 ˚C gave best
enantioselectivity. Benzene is another suitable solvent but it freezes at -10 ˚C, therefore
3:1 mixture of benzene-toluene is used as this mixture does not freeze at this temperature.
Quinine, surprisingly, was found to be the best organocatalyst in order to obtain both high
yields and high enantioselectivity. Bulky catalysts, like (DHQD)
2
PYR and
(DHQD)
2
PHAL which have been show to give very high e.e. on sulfur addition to simple
and ketones and aldehydes failed to give good e.es in sulfenylation reactions that involve
DKPs.
69
After realization that quinine is the best organocatalyst, a comprehensive study was done
with many different substituted DKPs. The N-1 unsubstituted DKP which gave 95%
yield with sulfur reagent 22a gave only 10% e.e. As expected, the enantioselectivity of
sulfenylation increased with increasing size of ester group. Methyl ester gave least e.e. of
40% followed by ethyl ester e.e. of 62% and t-butyl ester gave e.e. of 75%.
Table 2.4 Optimization studies of sulfenylation of DKPs with different organocatalysts
and sulfenylating reagents 22a and 22b.
R
1
Sulfur reagent Catalyst Solvent
Temp.
(˚C)
Time
(h)
Yield
(%)
a
e.e.
Et 22a 9a PhH/PhMe (3:1) -10 120 70 62
Et 22a 9b PhH/PhMe (3:1) -10 120 52 15
Et 22a 9c PhH/PhMe (3:1) -10 120 55 22
t-
Bu
22b 9c PhH/PhMe (3:1) -10 48 96 35
t-
Bu
22b 9a PhH/PhMe (3:1) -10 48 72 18
t-
Bu
22a 9a PhH RT 72 32 30
t-
Bu
22a 9c PhH RT 72 47 60
a Yield of purified product after chromatographic separation.
Sulfur reagent 22a has shown very low reactivity toward DKPs. In the case where the
DKP has ethyl and benzyl on both N-1 and N-4 the reagent 22a was not able to
sulfenylate the DKPs, presumably due to the steric hindrance introduced by the relatively
bulky groups at N-1 and N-4. The novel reagent 22b gave good yields with the ethyl and
70
benzyl N-1 and N-4 substituted DKPs but very low e.es. 3-(tert-butoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione gave 35% e.e. on sulfenylation with 22b in benzene-
toluene (3:1) at -10 ˚C and 3-(tert-butoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione
gave 42% e.e. with 22b in toluene at room temperature. The highest enantioselectivity
was achieved with 1-benzylsulfanyl[1,2,4]triazole with 3-(tert-butoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione in benzene-toluene (3:1) at -10˚C and the product e.e. was
found to be 75%
Table 2.5 Organocatalytic enantioselective α-sulfenylation of substituted piperazine-2,5-
diones with sulfenylating reagents 22a and 22b catalyzed by quinine 9c (10 mol %)
R
1
R
2
R
3
Sulfur reagent
(equiv.)
Solvent
Temp.
(˚C)
Time
(h)
Yield
(%)
a
e.e.
H Me Et 22a (2.0) Toluene RT 24 95 10
H Me Et 22a (1.5) Toluene -10 24 90 12
Me Me
t-
Bu
22a (2.6) Benzene 4 72 27 70
Me Me Me 22a (2.6) Toluene RT 24 67 40
Me Me
t-
Bu
22a (10.0)
PhH/PhMe
(3:1)
-10 120 68 75
Et Et Et 22a (2.6) Benzene RT 72 NR -
Bn Bn Et 22a (2.6) Benzene RT 72 NR -
Me Me
t-
Bu
22b (3.0)
PhH/PhMe
(3:1)
-10 48 96 35
Et Et Et 22b (3.0) Toluene RT 60 77 15
Bn Bn Et 22b (3.0) Toluene RT 60 75 20
Bn Bn
t-
Bu
22b (10.0) Toluene RT 72 52 42
a Yield of purified product after chromatographic separation.
71
2.3.7 Proposed Mechanism for Organocatalysis of DKPs leading to sulfenylation. In
Figure 2.35 illustrates the proposed mechanism for organocatalytic sulfenylation of DKPs
with quinine and a 1,2,4-triazole based sulfenylating reagent. Quinine and other cinchona
alkaloids used in this study are chiral tertiary amines that act as Lewis bases. In the
transition state A of Figure 2.35, quinine extracts a proton on the α carbon of the
diketopiperazine next to the ester and a transition state is achieved where the proton is
shared between the two oxygens of the DKP and amine of the quinine.
Figure 2.35 Proposed catalytic cycle with quinine leading to sulfenylation of
diketopiperazines by an aryl triazole reagent.
72
The sulfenylating reagent which in this case is a 1,2,4-triazole after protonation can
readily donate the sulfur group as an electrophile. Thus, the nucleophilic deprotonated
DKP can attack the sulfur group, forming a C-S bond. The complex between DKP and
quinine is weak. The orientation of quinine is such that it blocks one particular face of the
DKP thereby allowing the sulfur reagent to approach from the other face. In the case of
bulky substrate due to the steric effects the quinine-DKP complex is not tightly bound the
transition state. The facial selectivity of the sulfur attack is significantly decreased,
thereby reducing the stereoselectivity of the reaction.
2.4 Organocatalytic coupling of diketopiperazines with indoles
2.4.1 Gramine as an alkylation reagent. Gramine, (1-(1H-indol-3-yl)-N,N-
dimethylmethanamine) is a suitable compound for alkylation of different substrates and
it was known since 1980 that reflux of gramine or gramine derivatives in an appropriate
solvent, such as acetonitrile (MeCN), and a trialkylphosphine creates an indolenine cation
which undergoes a coupling reaction with an anion of an α-carbonyl compound
(Kametani, Kanaya et al. 1980).
73
Figure 2.36 Mechanism of gramine coupling with alkyl/aryl anion via methylene
indolenine intermediate. Gramine derivatives in the presence of n-tributylphosphine
undergoes decomposition to form methylene indolenine type cation which is believed to
be in equilibrium with the phosphine adduct. Both these transition states can undergo
coupling with alkyl/aryl anion of an α-keto compound.
Due to the aromatic character of the indole the cationic intermediate methylene
indolenine type is stabilized. The phosphine can also react with this intermediate as
shown in Figure 2.36. Both intermediates can react with a suitable alkyl/aryl α-carbonyl
anion intermediate produced in-situ to give the coupled product.
This process, known as Somei-Kametani reaction, is a classic way of coupling a
methylene-indole group to different substrates in the synthesis of important bioactive
natural products. Usually, the yield of this reaction is low to moderate. In the case of
dicarbonyl compounds the yields could be high but for diketopiperazines the reported
yields are usually low to moderate (~70%) when N-1 of DKP is unsubtituted.
74
Figure 2.37 Total synthesis of Brevianamide E involving the key step of coupling of
substituted gramine with substituted diketopiperazine.
In 1980 Kametani et al. pioneered the use of this reaction in his published asymmetric
total synthesis of Brevianamide E (Kametani, Kanaya et al. 1980). There he reported
coupling of substituted gramine, 3-dimethylaminomethyl-2-(1',1'-dimethylallyl)indole
with a substituted diketopiperazine as a key reaction. The reaction was done under reflux
in dimethyl formamide in the presence of sodium hydride. After formation of the coupled
product, the subsequent steps afforded Brevianamide E.
In 1981 Somei et al. reported the use of n-tributyl phosphine for coupling of substituted
gramines with different α-carbonyl compounds (Figure 2.38) (Somei, Karasawa et al.
1981). The yields obtained for unhindered acyclic dicarbonyl compounds were high.
Thus, the strategy of using trialkyl phosphine under reflux in acetonitrile as a solvent
became the reaction of choice for coupling of gramines to various structurally complex
substrates, including diketopiperazines.
75
Figure 2.38 Coupling of a substituted gramine and a substituted diketopiperazine in the
course of stereoselective total synthesis of paraherquamide A.
Williams et al.. reported total synthesis of many alkaloids and other natural products
where a key reaction was coupling of a substituted gramine with a complex
diketopiperazine (Cushing, Sanzcervera et al. 1993; Cushing, SanzCervera et al. 1996;
Williams, Cao et al. 2003). The total synthesis of Paraherquamide A is an especially
cogent example of this transformation (Williams, Cao et al. 2003).
In the course of synthetic studies Williams et al. reported total synthesis of alkaloid type
natural products and emphasized the role of different substituent on N-1 of the DKP ring.
When the nitrogen was substituted by p-methoxy benzyl group the yield of the reaction
was around 50%. When the N-1 substituent is changed to methoxy carbonyl group the
yield increased to 70%. Thus, an electron withdrawing group increases the yield of
Somei-Kametani reaction whereas an electron releasing groups on N-1 of the ring
decrease the product yield.
76
Thus, the low to moderate yields and the narrow substrate scope N-1 alkyl DKPs were
main limitations of this strategy. Thus, there was unmet need to develop a more general
strategy that could give higher yields and have broad scope of DKP substrates.
2.4.2 Use of quinine for alkylation of diketopiperazines with gramine. After the initial
success of the application of cinchona alkaloids in α-substitution it was decided to
explore organocatalytic strategy in the coupling of methylene-indole substrates with
substituted piperazine-2,5-diones. First, Somei-Kametani reaction was attempted with
DKP 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione in acetonitrile solvent in reflux in
presence on n-tributylphosphine.
Figure 2.39 Coupling of gramine with 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione
in the presence of n-tributylphosphine in acetonitrile under reflux.
The yield of the reaction was moderate (42%) and the formation of the n-
tribultylphosphine oxide as a byproduct significantly comkplicated product isolation.
Thus, we envisaged the possibility of using quinine as a catalyst in a new approach to
couple gramine and DKPs.
Indeed, the use of catalytic amounts of quinine in place of n-tributylphosphine gave the
desired product, albeit in low yield, which increased 92% when one equivalent of quinine
was used.
77
Figure 2.40 Coupling of gramine with 3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione
in the presence of quinine as a Lewis base in acetonitrile under reflux.
After success with N-1 unsubstituted DKP the N-1 methylated DKP was tried with
quinine, to give product in 81% yield.
Figure 2.41 Coupling of gramine with 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-
dione in the presence of quinine yielding ethyl 2-((1H-indol-3-yl)methyl)-1,4-dimethyl-
3,6-dioxopiperazine-2-carboxylate.
Generally, in the classic Somei-Kametani diketopiperazines substituted at N-1 by an
electron withdrawing acyl group give high yields. In the case where the N-1 is
unsubstituted or when an alkyl group is present the yields are low. Using quinine in
refluxing acetonitrile afforded high yields for DKPs regardless of their substitution at N-1
position.
78
Table 2.6 Effect of substrate and temperature on quinine-promoted alkylation of
substituted diketopiperazines with gramine.
R
1
R
2
Equiv. of gramine Time (h) Temp. (˚C) Yield (%)
a
Et H 2.0 24 RT NR
Et Me 2.0 24 RT NR
Et H 2.0 24 50 10
Et Me 2.0 24 50 <5
Et H 2.0 24 82 92
Et Me 3.5 48 82 81
t-Bu Ne 4.0 48 82 72
a Yield of purified product after chromatographic separation.
Table 2.6 shows the results after optimization of the reaction. This strategy shows
promising results even in the case of hindered ester group as 72% yield was achieved
after 48 hours of reflux with 4.0 equivalents of gramine and 1.0 equivalents of quinine.
As expected, the DKPs did not couple with gramine at room temperature, presumably
due to the fact that the gramine gives methylene-indolenine type intermediate only under
refluxing conditions. Even at moderate temperatures (50 ˚C) the yields were low. These
results show the importance of using high temperature (acetonitrile at reflux) that
facilitates removal of the formed dimethylamine in order to generate the reactive
intermediate.
79
Figure 2.42 Reaction with N-Boc gramine and gramine N-oxide where no product was
obtained.
Despite the use of chiral catalyst, the gramine-DKP coupling reactions gave low
enantioselectivity on the order of 10-25% e.e. This result was not unexpected because the
reaction occurs at high temperatures of reflux. In order to study the effect of the gramine
substitution, N-Boc gramine and gramine N-oxide were tested. No desired product was
obtained. In case of N-Boc gramine, as expected, the formation of methylene indolenine
intermediate is disfavored by the electron withdrawing Boc group. Gramine N-oxide is
insoluble in acetonitrile at room temperature as shown in Figure 2.42 and hence gives no
product at room temperature although under reflux it gives yield similar to gramine itself.
2.4.3 Vinylogous iminium ion intermediates generated from arylsulfonyl indoles add to
aldehydes. In 2008, Melchiorre et al. reported the use of 3-(1-arylsulfonylalkyl)indoles as
suitable electrophilic precursors to couple the vinylogous iminium ions with aldehydes
(Shaikh, Mazzanti et al. 2008).
80
Figure 2.43 Coupling of 3-(1-arylsulfonylalkyl)indole with aldehyde in the presence of
KF/alumina with L-proline as a catalyst.
In 3-(1-arylsulfonylalkyl)indoles the sulfonyl moiety is a leaving group, thus under acidic
or basic conditions the removal of the sulfonyl group gives rise to electrophilic
indolenine intermediate. KF on activated alumina was found to be superior reagent to
promote the removal of the sulfonyl group.
A range of different R
2
groups were tested in order to determine the impact of the
substrate on the yield and stereoselectivity of the reaction. All the R
2
variants tried where
substituted or unsubstituted phenyl rings. The best yields and stereoselectivities were
achieved when R
2
was a simple phenyl group without any substitution. R
3
was group was
varied between hydrogen, methyl and phenyl. The best results were obtained when R
3
was methyl group. In case when R
3
was hydrogen, low stereoselectivity was achieved
and when R
3
phenyl group was used both the yields and stereoselectivity were low.
Different pyrollidine-based catalysts were used but in their system the catalyst which
gave best yields was L-proline.
The proposed mechanism of coupling of the aryl sulfones to aldehydes is depicted in
Figure 2.44.
81
Figure 2.44 Proposed mechanism of coupling of the aryl sulfones to aldehydes. After
treating 3-(phenyl(tosyl)methyl)-1H-indole with KF/alumina the -SO
2
Tol group leaves
producing an electrophile which is stabilized by aromaticity of indole ring and phenyl
group.
KF/alumina is a strong base which extracts a proton from the indole, formin an
electrophile after the arylsulfonyl group leaves. The subsequent proton transfer from the
reaction mixture generates substituted indolenine and the two stabilized resonance forms
of the electrophile are shown in Figure 2.44.
2.4.4 Novel strategy of organocatalytic coupling of diketopiperazines with indoles. In
the course of our studies of the organocatalytic sulfenylation we developed an efficient
way of generating a nucleophile from the substituted DKP by deprotonation of its α-
carbon. Based on these studies, we envisaged a novel strategy for the construction of the
DKP-indole linkage when electrophilic indolenine intermediate (vinylogous iminium
indole ions) formed from arylsulfonyl indoles could be used to alkylate DKP in a process
promoted by cinchona alkaloids.
Although the addition of the vinylogous iminium indole ions to nucleophiles generated
from aldehydes, was documented (Shaikh, Mazzanti et al. 2008), the nucleophiles
generated by deprotonation of an α-carbons of DKPs are much less reactive. Despite
that, when 2-methyl-3-(phenyl(tosyl)methyl)-1H-indole with KF/alumina was reacted
with 3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione and (DHQD)
2
PYR as catalyst
82
in dichloromethane at room temperature the product was formed, albeit in moderate yield
(Figure 2.45).
Figure 2.45 Coupling of 2-methyl-3-(phenyl(tosyl)methyl)-1H-indole with 3-(t-
butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione in the presence of KF/alumina as
heterogeneous base and (DHQD)
2
PYR as an organocatalyst.
The reaction gave good diastereoselectivity of 5:1 and was chosen as a starting point for
optimization of conditions.
Interestingly, the organocatalyst used (DHQD)
2
PYR is bulky and gave low yields in
sulfenylation reactions. In this case, however, this catalyst gave moderate yields with
only 3.0 equivalents of diketopiperazine. Next, (DHQD)
2
PHAL was tested, giving
increased yield to 72% whereas the diastereoselectivity achieved was 4.5:1 which is close
the value obtained with (DHQD)
2
PYR (Figure 2.46).
Figure 2.46 Coupling of indole and diketopiperazine in the presence of (DHQD)
2
PHAL
as organocatalyst.
83
The amount of KF/alumina is critical in order to obtain high yields. In the reaction
showed in Figure 2.45 and Figure 2.46 80 mg per 0.1 mol of arylsulfone indole of
KF/alumina produced best results.
After coupling with 2-methylated arylsulfonyl indole, 3-(phenyl(tosyl)methyl)-1H-indole,
which is lacking substitution at indole position 2 was coupled (Figure 2.47).
Figure 2.47 Coupling of 3-(phenyl(tosyl)methyl)-1H-indole with 3-(t-butoxycarbonyl)-
1,4-dimethylpiperazine-2,5-dione.
The reaction gave 75% yield of the with 3.0 equivalents of 3-(t-butoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione and (DHQD)
2
PHAL as organocatalyst in dichloromethane.
The diastereoselectivity was 3:1, lower than the diastereoselectivity achieved with 2-
methyl-3-(phenyl(tosyl)methyl)-1H-indole.
The variability of the indole substrates and organocatalysts show the versatility of the
coupling strategy. In order to determine effect of solvent, temperature and catalysts on
yiels and diastereoselectivity, optimization study was performed. Different of conditions
and coupling substrates were used along with a variety catalysts to gain insight into the
substrate scope and versatility of this reaction.
When R
1
was hydrogen instead of methyl at position 2 of the indole the yields increased
with (DHQD)
2
PYR to 70% in 72 h from 55% in 48 h but the diastereoselectivity ratio
84
decreased from 5:1 to 2.6:1. (DHQD)
2
PHAL is a better organocatalyst than
(DHQD)
2
PYR, giving higher yields. But when toluene was used as the solvent in place of
dichloromethane with (DHQD)
2
PHAL as the catalyst the yields decreased to 46% in 120
h from 72% in 48 h. There was also a marginal decrease in diastereoselectivity to 3.8:1 in
case of toluene as the solvent. Toluene-dichloromethane at 0 ˚C did not yield any
product. Similar trends were seen at lower temperature for dichloromethane where the
yields drastically decreased at 10 ˚C and 0 ˚C.
Table 2.7 Optimization and scope of α-alkylation of diketopiperazines with arylsulfonyl
indoles catalyzed by cinchona alkaloids.
R
1
R
2
Solvent Catalyst Temp. (˚C) Time (h) Yield (%)
a
d.r.
Me t-Bu CH
2
Cl
2
(DHQD)
2
PYR RT 48 55 5:1
H t-Bu CH
2
Cl
2
(DHQD)
2
PYR RT 72 70 2.6:1
Me t-Bu CH
2
Cl
2
(DHQD)
2
PHAL RT 48 72 4.5:1
Me t-Bu Toluene (DHQD)
2
PHAL 0 120 NR NR
Me t-Bu CH
2
Cl
2
(DHQD)
2
PHAL 10 120 22 3:1
Me t-Bu CH
2
Cl
2
(DHQD)
2
PHAL 0 120 8 2.5:1
Me t-Bu Toluene (DHQD)
2
PHAL RT 120 46 3.8:1
H t-Bu CH
2
Cl
2
(DHQD)
2
PHAL RT 72 75 3:1
Me t-Bu CH
2
Cl
2
(DHQD)
2
PHAL RT 72 89 4.5:1
Me Et CH
2
Cl
2
Quinine RT 48 71 1:1
Me Et Toluene Quinine RT 48 54 1:1
Me t-Bu CH
2
Cl
2
Quinine RT 48 70 2.7:1
H t-Bu CH
2
Cl
2
Quinine RT 72 70 2.5:1
a Yield of purified product after chromatographic separation.
85
When quinine was used an organocatalyst with the t-butyl ester DKP, the yield wasthe
same as that achieved by (DHQD)
2
PHAL, however, the diastereoselective ratio obtained
by quinine was lower than that obtained by (DHQD)
2
PHAL. Thus (DHQD)
2
PHAL
became the catalyst of choice, giving good yield and stereoselectivity. Apart from 3-(t-
butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione, the ethyl ester DKP 3-
(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione was also tested. The yields obtained
were good but with quinine no diastereoselective bias was observed.
All reactions in Table 2.7 were done with optimized amount of arylsulfone indole. In the
arylsulfone indoles used in this novel strategy the examples in Table 2.7 have phenyl
group at the methylene position of the indole group. Although the absence of phenyl
group makes the indolenine electrophile less stable, the substrate without phenyl group
was also tested in order to verify the versatility of the reaction. To our pleasant surprise,
the coupling occurred in very high yield (90%, Figure 2.48).
Figure 2.48 Coupling of 3-(tosylmethyl)-1H-indole with 3-(t-butoxycarbonyl)-1,4-
dimethylpiperazine-2,5-dione in the presence of KF/alumina and (DHQD)
2
PHAL as
organocatalyst.
The broad substrate scope of this reaction makes it applicabile in coupling of a wide
range of indoles and diketopiperazines in the course of the synthesis of alkaloids and
other natural products.
86
To further test generality of this method, the same strategy used for DKPs was applied to
couple 2-methyl-3-(phenyl(tosyl)methyl)-1H-indole with a cyclic β keto ester, ethyl 2-
oxocyclopentanecarboxylate.
Table 2.8 Optimization of catalysts and solvents in coupling of indole derivative to 2-
oxocyclo-pentanecarboxylate with -ketoester.
Entry Solvent Catalyst Temp. (°C) Time (h) Yield (%)
a
d.r.
1 CH
2
Cl
2
9a rt 24 81 2:1
2 Toluene 9a rt 48 74 1.5:1
3 CH
2
Cl
2
9b rt 48 78 1.6:1
4 CH
2
Cl
2
9c rt 48 75 2.2:1
As shown in Table 2.8, under given conditions the arylsulfonyl indoles readily coupled
with ethyl 2-oxocyclopentanecarboxylate thereby showing that this strategy can be used
for other substrates having carbonyl functionality apart from DKPs. Different catalysts
(9a-9c) and solvents were used for this strategy showed its robustness by giving decent
yields under these different conditions.
2.5 Conclusions and future directions. In conclusion, two novel strategies were devised
for organocatalytic sulfenylation of diketopiperazines and indole diketopiperazine
coupling.
87
Figure 2.49 Summary of sulfenylation and indole-DKP coupling strategies. A)
Organocatalytic sulfenylation of diketopiperazines with electrophilic sulfenylating
reagent phenyl/benzyl [1,2,4] triazole. B) Coupling between diketopiperazines and
indoles with gramine and arylsulfonyl indole.
The future goal of the sulfenylation strategy is to develop a method for stereoselective
placement of the two sulfurs on the diketopiperazine ring syn with regard to each other.
As mentioned in Chapter 1 syn-sulfurs and redox-active disulfide bridge on a DKP ring
plays a crucial role in biological activity of ETPs. In addition, the DKP molecule with
syn-hemithioaminal function may aslo have biological activity when the two sulfur
88
bearing groups are syn to each other and are capable of binding to biologically mobile
metal ions or reacting with metalloproteins.
Figure 2.50 Schematic illustration of future application of the sulfenylation strategy to
obtain an ETP molecule from syn-hemithioaminal.
The future applications of our indole-diketopiperazine coupling strategy are numerous
and may include total syntheses of natural products containing cyclotryptophan motif,
such as chetomin and chaetocin. This coupling strategy may also provide very facile and
high yielding routes to synthetic drug-like molecules that contain indole-diketopiperazine
linkages.
89
Chapter 3: Design of Dimeric ETPs by Biomimetic Rational Approach and
Investigation of their in vitro Biophysical Properties
90
3.1 Structural basis of p300 - HIF-1α binding. The structural basis for the recruitment of
p300 coactivator by HIF-1α was reported by Wright and coworkers and Livingston and
coworkers (Freedman, Sun et al. 2002; De Guzman, Wojciak et al. 2005). In these
structural studies the p300-CH1 domain (a.a. 323-423) and HIF-1α TAD-C (786-826)
were co-expressed in E. Coli and the zinc-bearing complex was purified.
Figure 3.1 Structure of p300-CH1 domain in complex with HIF-1α TAD-C. NMR
structure of p300-CH1 domain (a.a. 323-423, green) in complex with HIF-1α TAD-C
(a.a. 786-826, red). Zinc ions are shown as blue spheres. Structure was obtained from
pdb.org, accession number 1L3E.
The p300-CH1 region consists of four α-helices and has a zinc trinuclear site i.e., three
Zn
2+
ions are present in the fully folded domain. All the three Zn
2+
are complexed by the
amino acid residues that represent Cys
3
His motif. The zinc ions lie at the vertices of the
triangle formed by the four α-helices. The packing of the helices exposes large
hydrophobic surface of the protein that may interact with other proteins. HIF-1α TAD-C
domain that interacts with p300-CH1 has four components, with two α-helices, a loop
91
region between the helices and an N-terminal extended region. The two α-helices are
parallel to each other in the complex with p300-CH1, so that they are embedded in the
hydrophobic pockets surrounding α-helix 3 of p300-CH1.
3.2.1 Physical basis of disruption of p300 and HIF-1α complex by chetomin. Zinc ions
are critical for mainataining of the global fold of p300-CH1 region. It has been shown by
Wright et al.. that either excess or absence of zinc ions results in misfolding of p300-CH1
(De Guzman, Wojciak et al. 2005). In fact, NMR data shows the binding between HIF-1α
and p300-CH1 decreases upon addition of a chelating reagent, such as EDTA.
Thus, one possible explanation of the mode of action of chetomin in inhibiting the HIF-
1α - p300-CH1 interaction is the disruption of the global fold due to the removal of one
or more zinc from the CH1 domain of p300.
The four sulfurs in chetomin, two in each ETP ring
are believed to for a chelate with one or more zinc
ions in the CH1 domain of p300 (Kung, Zabludoff et
al. 2004; Cook, Hilton et al. 2009). It is not clear
whether eventually all the four sulfurs on the chetomin
would chelate to one zinc ion or whether only two
sulfurs from one molecule chelate to a zinc ion and
after ejection of zinc from the site another molecule of
chetomin forms two more sulfur coordinate bond with the zinc ion. The two possibilities
are shown in Figure 3.3.
Figure 3.2 Structure of
chetmoin
92
Figure 3.3 Possible modes of chelation of zinc ion by chetomin. A) Schematic
representation of chelation of zinc by four sulfurs (left) and the MM2 minimized model
of this complex (right). B) Chelation of zinc by two chetomin molecules; two sulfurs
from each of the two chetomin are involved in formation of coordination complex with
zinc ion (left), MM2 minimized model of this complex (right).
The model A, is energetically less favorable as compared to model B because of torsional
strain. But if torsional strain is relaxed by changing the structure of the bis-ETP the
chelation of zinc by one molecule might become more favorable as compared to the
chelation complex that involves two different molecules.
3.2.2 Basis for design of novel bis-ETPs. The idea that optimizing of the spatial
placement of sulfurs and relaxing the torsional strain in the molecule after formation of
complex with zinc ion may improve biophysical properties and/or transcriptional
93
regulation by the molecule led us to design novel bis-ETPs. Such compounds should
fulfill a few criteria, outlined below:
a) Designed bis-ETPs should have more relaxed linker between them so that the torsional
strain after formation of complex with zinc ions should be lesser as compared to
chetomin.
b) Natural bis-ETPs, like chetomin, chaetocin, verticillin etc., have aromatic/indole based
scaffold. But it is believed that this indole scaffold may also be a reason which
exacerbates the cytotoxicity of these natural bis-ETPs. The aromaticity of the indole,
hovewer, may be important for its high-affinity binding to the target p300-CH1 region.
Apart from the ETP core, the other functionalities and moieties present in the molecule
are also important in dtermining its efficacy .
c) No total synthesis of chetomin has been reported to date, therefore another important
criterion in the design of new bis-ETP should be its facile synthesis.
Therefore, to maintain the aromatic character of the bis-ETP but at the same time in order
to avoid incorporation of the indole group, it was decided to use a non-indole aromatic
linker.
94
Figure 3.4 Structures of designed bis-ETPs ETP-3 and ETP-5. Chetomin displays
distance of 10 Å between the centers of disulfide bridges of the two ETP rings. ETP-3
also has 10 Å of distance between sulfurs on two ETP rings. ETP-5 has more flexible 1,2-
di-p-tolylethane linker. ETP-5 has 19 Å distance between the sulfurs of the two ETP
rings.
Based on the criteria discussed above, two ETPs were designed having non-indole
aromatic linker. Synthesis of ETP-3 along with its biological activity data was published
by our group in 2009 (Block, Wang et al. 2009). It has a p-xylene linker joining the two
ETP rings. The key feature in the synthesis is protection of the two sulfurs via formation
of a thioacetal early in the synthesis. α,α' - Diiodoxylene was used to install the linker and
join the two thioacetal ETPs in one step. Deprotection of thioacetal and benzyloxy methyl
groups yielded ETP-3.
95
Figure 3.5 Scheme representing synthesis of ETP-3.
The meso-8 compound was removed from the reaction mixture prior to final steps via
HPLC purification. In order to study the difference between the biophysical properties of
the two enantiomers of ETP-3, the two enantiomers were also separated via HPLC using
chiral columns.
96
Figure 3.6 Synthesis of ETP-5.
97
The synthesis of ETP-5 is shown in Figure 3.6. The coupling of the two ETP rings was
achieved in two steps by using α,α'-dibromo-p-xylene to install aromatic linker between
the two ETP rings. Boron trichloride is the reagent used to deprotect benzyloxy methyl
group by cleaving the benzyl group off to obtain an alcohol. In the final step, the
thioacetal group was removed by mCPBA followed by the treatment with perchloric acid
in the presence of dimethyl sulfide. Trifluoroacetic acid was also used instead of
perchloric acid in an optimized one pot reaction to get a combined yield of about 65% for
the final two steps of ETP-5 synthesis.
3.3.1 Expression and purification of p300-CH1 domain for biophysical assays. The first
set of experiments conducted in order to evaluate the activity of ETP-3 and ETP-5 were
the biophysical assays to determine the binding of these molecules to the CH1 domain of
p300 and their ability to disrupt the binding interaction between p300-CH1 domain and
HIF-1α TAD-C.
We first used direct binding assay employing surface plasmon resonance (SPR) method
(Liedberg, Nylander et al. 1983). This technique involves immobilization of the binding
protein, in this case CH1 domain of p300, to the SPR chip and flowing the solution of
bis-ETPs through the channels on the chip. This method allows to obtain the binding and
dissociation constants for the small molecule flowing through the channels..
For the second set of experiments fluorescence polarization competition assays (Jolley,
Stroupe et al. 1981; Jameson and Ross 2010) were used, in order to study the disruption
of binding between p300-CH1 domain and HIF-1α TAD-C. The two binding partners
98
were p300-CH1 domain and the essential segment of HIF-1α TAD-C (aa. 786-826)
conjugated to fluorescein dye.
The plasmid encoding p300-CH1 domain (aa. 323-423) was cloned into the PGEX-4T2
vector immediately following its coding for the GST tag.
Table 3.1 Amino acid sequence and corresponding DNA sequence for CH1-p300 domain.
Amino acid sequence for CH1 domain of p300 (a.a. 323-423):
MGSGAHTADPEKRKLIQQQLVLLLHAHKCQRREQANGEVRQCNLPHCRTMKNVLN
HMTHCQSGKSCQVAHCASSRQIISHWKNCTRHDCPVCLPLKNAGDK
Optimized DNA Sequence for CH1-p300: Length: 315, GC%:56.19 :
GGCAGCGGCGCGCATACCGCCGATCCGGAAAAACGTAAACTGATTCAGCAGC
AGCTGGTGCTGCTGCTGCATGCGCATAAATGCCAGCGCCGTGAACAGGCGAA
TGGCGAAGTTCGTCAGTGCAATCTGCCGCATTGCCGCACCATGAAAAACGTGC
TGAACCATATGACCCATTGTCAGAGCGGTAAAAGCTGCCAGGTTGCCCATTGC
GCGAGCAGCCGCCAGATTATTAGCCACTGGAAAAACTGCACCCGCCATGATT
GCCCGGTGTGCCTGCCGCTGAAAAACGCGGGCGATAAA
Table 3.2 Amino acid sequence of the Fusion GST-p300-CH1 domain.
MSPILGYWKIKGLVQPTRLLLEYLEEKYEEHLYERDEGDKWRNKKFELGLEFPNLPYYIDG
DVKLTQSMAIIRYIADKHNMLGGCPKERAEISMLEGAVLDIRYGVSRIAYSKDFETLKVDFLS
KLPEMLKMFEDRLCHKTYLNGDHVTHPDFMLYDALDVVLYMDPMCLDAFPKLVCFKKRIE
AIPQIDKYLKSSKYIAWPLQGWQATFGGGDHPPKSDLVPRGSMGSGAHTADPEKRKLIQQ
QLVLLLHAHKCQRREQANGEVRQCNLPHCRTMKNVLNHMTHCQSGKSCQVAHCASSRQII
SHWKNCTRHDCPVCLPLKNAGDKEFPGRLERPHRD
Black - GST tag, green - thrombin cleavage site, red - p300 CH1 domain, blue - tail
AAs
This p300-CH1-GST fusion protein was cloned, expressed and purified in a procedure
similar to the one described by Livingston and coworkers (Kung, Wang et al. 2000). The
99
DNA fragment for p300-CH1 protein domain shown in Table 3.1, inserted in puc57 was
synthesized by GenScript, Inc. This p300-CH1 insert was then sub-cloned into the
PGEX-4T2 and transformed into E. Coli BL21 (DE3) .
As mentioned above, zinc chloride was added to the expression media, to achieve the
final concentration of 10 µM zinc ions. The lysis buffer E. Coli had 100 µM zinc
chloride. Zinc was added during the expression and lysis of cells because for one
equivalent of CH1 domain requires three equivalents of zinc ions. It was noted that after
the expression of GST-p300-CH1 protein there needs to be some zinc ions present in the
buffer to maintain the zinc in the proper sites of the CH1 domain and to maintain the
proper folding of the protein. Protein purification was done with glutathione sepharose
beads (GE Healthcare). Addition of glycerol is essential for stability of protein in the
storage conditions of -80 ˚C.
3.3.2 Expression of
15
N labeled GST-p300-CH1 fusion protein.
15
N isotope protein was
expressed in order to use NMR to study the in vitro disruption of GST-p300-CH1 and
HIF-1α complex with ETPs and other HIF-1α mimics (Henchey, Kushal et al. 2012).
In order to express
15
N labeled GST-p300-CH1 the E. coli transformed with the plasmids
for this protein was grown to 0.6 O.D. in 1 L media after which the culture was
centrifuged in order to obtain pellets of the bacteria. The pellets were washed and
resuspended in 1 L minimal media having
15
N ammonium chloride (
15
NH
4
Cl). No other
source of nitrogen was present in the minimal media. Other ingredients in the minimal
media were α-MEM vitamins, glucose and NaCl. After resuspension of pellets in the
minimal media the protein expression was induced by IPTG. After five hours, the cells
100
were pelleted, lysed and purification was done as mentioned above. In case of
15
N labeled
GST-p300-CH1 the usual yields were approximately 15 mg/L of culture as compared to
25-30 mg of unlabeled GST-p300-CH1 protein obtained using the standard procedure.
The method used for expression was
15
N labeled protein is derived from a widely used
method by researchers and is reported by Bracken et al.. (Marley, Lu et al. 2001).
Preliminary investigation by NMR of recombinant protein showed a similar resonance
pattern in 2D NMR as is reported in literature.
3.3.3 Cleavage of p300-CH1 domain from GST. For certain set of biophysical assays
like fluorescence polarization (FP) competition assays and NMR analysis of protein
complex, it was desirable to cleave the p300-CH1 domain from the GST tag. For
example, the NMR analysis could become more complicated if the spectra included
signals of the ~28 kDa GST tag. Therefore, the recombinant protein was cleaved from the
GST tag via its a thrombin cleavage site, LVPRGS, located between GST and CH1
domain.
The thrombin cleavage was done right after the protein was expressed and immobilized
on glutathione sepharose beads. For
15
N labeled GST-CH1-p300 protein thrombin
cleavage was done in NMR buffer and the concentration of the obtained
15
N p300-CH1
was around ~300 µM (APPENDIX C, S3.1).
3.4.1 Binding study of ETPs toward CH1 domain of p300 with surface plasmon
resonance (SPR). Surface plasmon resonance is a powerful technique that can be used to
determine direct binding of two components. In such an experiment, one protein is
immobilized on the SPR chip either directly to the dextrose layer or indirectly, via
101
antibody. Surface plasmons are electromagnetic waves present on the interface of metal
and non-metallic surface (Barnes, Dereux et al. 2003). They are very sensitive to the
depth of the interface. Any change in the interface depth due to absorption or binding of
other molecules to the non-metallic (organic) layer causes change in the surface dielectric
constant. With the change in dielectric constant the surface plasmons change and the
angle of reflection of the incident resonating light changes. The Biacore SPR instrument
(Malmqvist 1999) measures rate of association of the free molecule in the flowing buffer
to the immobilized molecule/protein on the SPR chip and dissociation rate of the two
molecules upon flow of empty buffer. The binding constant K
D
is determined by
calculating the ratio of rate of dissociation (off rate) to the rate of association (on rate).
Binding Constant, K
D
= off rate/on rate = k
d
/k
a
The SPR chip has many channels with gold surface. A dextrose layer is present on the
gold surface through which the protein/antibody is immobilized on the chip using NHS
ester coupling. The protein being immobilized is flowed on to the channels having
covalently linked antibody. Upon immobilization of the protein on the antibody layer in
the channels of the chip the experiments can be performed either immediately or the chip
can be stored at 4 ˚C.
3.4.2 SPR analysis of chetomin binding to CH1 domain of p300. From the 1 mg/mL
stock solution of chetomin in DMSO different concentration solutions of chetomin were
made. The buffer to make the dilution was as follows: 10 mM Tris, 100 mM NaH
2
PO
4
,
500 µM DTT, 100 µM ZnCl
2
, and 0.1% NP-40 at pH 8.0. Six concentrations for
chetomin were made 50 nM, 200 nM, 500 nM, 1 µM, 10 µM and 50 µM.
102
Each sample and buffer had 5% DMSO. DMSO was added because chetomin is not very
soluble in buffer without DMSO. The data was double-referenced with a channel having
only anti-GST antibody and no GST-CH1-p300. Each sample was also run through this
channel having no GST-CH1-p300. The 5% DMSO buffer was injected to establish the
baseline for the blank.
Figure 3.7 SPR sensorgram for chetomin binding with the immobilized GST-CH1-p300.
Each sample was injected at a rate of 30 µL/min for 3.3 minutes followed by buffer wash
for 3.3 minutes at 30 µL/min. After each sample the chip was regenerated by injecting 10
mM H
3
PO
4
onto the chip for 1 minute at the rate of 30 µL/min followed by buffer wash.
The measurement of each sample was done in triplicate.
For chetomin, the measured the rate of association of chetomin to the CH1 domain of
p300 was obtained for the six different dilutions. The "on-rate" of the rate of association
k
a
measured for chetomin is 2.93 x 10
3
± 32.3. The rate of dissociation ("off-rate")
obtained during the wash step just after the association step for the six different samples
103
(k
d
) is 3.54x10
-3
± 4.8x10
-5
. The binding constant for chetomin to the immobilized GST-
CH1-p300 was calculated as k
d
/k
a
givin a value of K
D
= 1.21 µM.
3.4.3 Direct binding of ETP-3 and ETP-5 to CH1 domain of p300 in SPR assays. After
measurement of direct binding between chetomin and CH1 domain of p300 by SPR we
tested the designed ETPs ETP-3 and ETP-5 for direct binding to the CH1 domain.
Figure 3.8 SPR data for direct binding of ETP-3 to immobilized GST-CH1-p300.
Binding constant of 1.12 µM was obtained for ETP-3 - CH1 domain.
For ETP-3 six dilutions of 50 nM, 200 nM, 500 nM, 1 µM, 10 µM and 50 µM were made
from the DMSO stock solution of 1 mg/mL ETP-3. The buffers and samples were
degassed and same procedure and rates of injections were used as for chetomin. The rate
of association k
a
obtained for ETP-3 was 12 ± 4.35 and rate of dissociation k
d
obtained
for ETP-3 in the wash step following the association step was 1.34x10
-5
± 1.05x10
-5
.
Therefore the binding constant measured by SPR analysis for ETP-3 was 1.12 µM.
104
Figure 3.9 SPR data for direct binding of ETP-5 to immobilized GST-CH1-p300.
Binding constant of 3.62 µM was obtained for ETP-5 - CH1 domain.
For the bis-ETP ETP-5 six dilutions of 50 nM, 200 nM, 500 nM, 1 µM, 10 µM and 50
µM were made from the 1 mg/mL of DMSO stock. The rate of association k
a
obtained by
SPR analysis for ETP-5 was is 4.25x10
3
± 85.8 while in the following wash step the rate
of dissociation k
d
obtained was 1.54x10
-2
± (1.36x10
-3
). Thus the binding constant
obtained for ETP-5 was 3.62 µM.
To summarize, chetomin and ETP-3 and ETP-5 bind to CH1 domain of p300 in vitro
with binding constants in low micromolar range. Although ETP-3 and ETP-5 have
similar binding constants the ETP-5, which has longer linker length and as a result has
greater spatial distance of 19 Å between the sulfurs on has slightly higher K
D
of 3.62 µM.
This higher binding constant, or slightly lower affinity towards CH1 domain for ETP-5
may be due to the entropic cost associated with longer linker length.
105
3.4.4 SPR analysis of the binding of control DKP. Sulfurs are thought to play a very
important role in high-affinity binding of ETPs to the CH1 domain of p300. In order to
test this, a desthio-ETP3 was synthesized. The compound NP-481 has similar framework
and linker structure as ETP-3, but is lacking disulfide bridges.
Figure 3.10 Structure of NP-481 control compound. Structure of control desthio-ETP-3
NP-481 for binding and in vitro cell culture experiments.
Six different dilutions of 50 nM, 200 nM, 500 nM, 1 µM, 10 µM and 50 µM for NP-481
were made and SPR analysis was conducted. The SPR data revealed that NP-481
practically does not bind to the GST-CH1-p300 protein. As can be seen from the SPR
data in Figure 3.12, NP-481 shows very low affinity toward the immobilized protein. The
numerical values obtained on calculation of k
a
and k
d
gave binding constant K
D
above 50
µM.
106
Figure 3.11 SPR sensogram for NP-481. SPR data for NP-481 shows that K
D
is >50 µM.
SPR data for NP-481 underscored the importance of disulfide bridges in high-affinity
binding of ETPs toward the CH1 domain of p300. Since NP-481 does not have any
sulfurs thus it is not able to bind to the CH1 domain.
Next, we investigate dthe redox state of sulfurs in ETPs that interact with the CH1
domain. In ETPs the sulfurs are a part of a disulfide bridge, which can be readily reduced
to dithiols inside a mammalian cells. In order to study the difference in binding of ETPs
forming oxidized and reduced frorm, SPR experiments were designed where in one case
the six different dilutions for ETP-3 were prepared in a buffer containing 500 µM DTT
and in another case the six different dilutions were prepared in a buffer without DTT.
107
Figure 3.12 SPR analysis of ETP-3 in two different reducing conditions. A) Binding in a
buffer that contains 500 µM DTT give K
D
of 1.09 µM. B) Binding in a buffer without
DTT gived K
D
of 27.1 µM..
The SPR data of these two different cases clearly show that when the reducing agent
DTT is present the ETP-3 gave K
D
of 1.09 µM which is about 27 times higher than K
D
of
27.1 µM obtained when no DTT is present in the solution. Thus this data shows that the
dithiol form of ETP is a much more potent binder of CH1 domain as compared to the
disulfide form of ETP.
108
Figure 3.13 The equilibrium between the disulfide and dithiol forms of ETP-3. The
equilibrium between the disulfide and dithiol forms of the ETP is shifted to the right hand
side (dithiol form) in presence of DTT.
3.4.5 Binding affinity of two stereoisomers of ETP-3 analyzed by SPR. Due to the fact
that racemic building blocks are used in the synthesis, the product ETP-3 forms three
stereoisomers, two of which enantiomers and one is a meso compound. The meso-
compound was resolved in the penultimate step of the synthesis via HPLC. In order to
separate the enantiomers, Chiralcel OD-H chiral column was used. With this technique
the two enantiomers of ETP-3 were separated. Apart from chiral HPLC peaks and mass
spectrometry analysis of the two enantiomers ent1-3 and ent2-3, in order to confirm
opposite spatial configuration of the two enantiomers, CD spectra was conducted. Figure
3.15 shows the superimposed CD spectra for the two enantiomers. CD spectrum shows a
mirror image of the CD spectra plot for the two enantiomers.
109
Figure 3.14 CD spectra for the enantiomers of ETP-3 viz. ent1-3 and ent2-3. CD spectra
shows opposite phase for the two enantiomers i.e. the two spectra are mirror images of
each other.
The amount of the recovered enantiomers after the chiral separation was insufficient for
measurements of optical activity.
In order to determine the binding affinity for the two enantiomers and meso compound
SPR analysis was done. Six different dilutions of 50 nM, 200 nM, 500 nM, 1 µM, 10 µM
and 50 µM were made for ent1-3, ent2-3 and meso-3 compounds.
110
Figure 3.15 SPR sensorgrams for stereoisomers of ETP-3. SPR sensorgrams for three
stereoisomers of ETP-3 i.e. the two enantiomers and meso-compound.
The SPR data sussgest that the two enantiomers and meso compound have very similar
binding affinities toward the CH1 domain. This was not unexpected, because the main
contact is not a non-covalent hydrophobic interaction but a coordinate bond with a metal
ion.
3.5.1 Fluorescence polarization assays to study disruption of TAD-C HIF-1α and CH1-
p300 complex by ETPs. Fluorescence polarization competition assays were used to study
disruption of the complex between p300-CH1 and HIF-1 TAD-C domains. It is based
on the phenomenon that the polarization of the emitted light from a fluorophore attached
to a molecule increases when it is bound to another macromolecule. The polarization of
emitted light is lost/decreased in fluorophore that freely rotates as opposed to the case
when it is bound to another molecule (Jolley, Stroupe et al. 1981; Jameson and Ross
2010).
111
The first step in fluorescence polarization (FP) assays involves determination of the
quantities and ratios of the two proteins. A saturation binding curve is obtained by
keeping the fluorophore-bearing molecule constant and varying the concentration of the
other protein. A saturation binding curve, depicting percent change in fluorescence
polarization is obtained. The concentration of the other protein at which FP (%) change
measured is 50%, representing that 50% molecules of fluorophore-bearing protein is
bound to the other protein, is the K
D
of the complex. Thus, when FP assays are done to
find the disruption of the protein-protein binding, the amount of the other protein is taken
somewhat greater than its concentration at a 50% fractional saturation, but lower than the
saturation value so that the protein does not out-compete the inhibitor.
3.5.2 Saturation binding fluorescence polarization assays for GST-CH1-p300 and
CH1-p300 with TAD-C HIF-1α. In order to perform fluorescence polarization a
fluorescein attached TAD-C (a.a. 786-826) HIF1-α peptide (HIF-1α-Fl) was used.
According to the saturation limit of the BioTek Synergy HT multiplate reader used for
the FP assays the amount of HIF-1α-Fl peptide was kept constant at 15 nM for all the
experiments. In order to determine the optimum amount of the GST-CH1-p300 protein
for the FP assays the saturation binding curve for the HIF-1α-Fl and GST-CH1-p300 was
found out by measuring the FP values on a 96 well plate for different wells having
different concentrations of GST-CH1-p300 and fixed concentration of HIF-1α-Fl at 15
nM.
112
Figure 3.16 Saturation binding curve for HIF-1α-Fl and GST-CH1-p300 obtained by FP
assay.
The log of GST-CH1-p300 concentration was plotted against the percentage of change in
the FP value to obtain the saturation binding curve (Figure 3.17). The K
D
obtained from
the saturation binding curve of HIF-1α-Fl and GST-CH1-p300 was 45 nM. Thus, for the
FP assays, to monitor the effect of an inhibitor, the amount of GST-CH1-p300 should be
higher than 45 nM to see appreciable change in FP values but should be lower than the
saturation value so that the inhibitor is not competed out.
In order to exclude the interference of GST in binding of the CH1 domain to HIF-1α, a
GST-cleaved protein was also used.
113
Figure 3.17 Saturation binding curve for the cleaved CH1-p300 domain and HIF-1α-Fl,
obtained by FP assay.
Just as in the case of GST-CH1-p300, the concentration of HIF-1α-Fl was kept constant
at 15 nM. Different concentrations of CH1-p300 protein were present in different wells
of the plate in triplicate. The log of CH1-p300 concentration was plotted against the
percentage change in FP value to obtain the saturation binding curve as seen in Figure
3.18. The value of K
D
obtained for CH1-p300 and HIF-1α-Fl binding is 544 nM. This
result is somewhat surprising and a possible reason for this higher K
D
could be that the
GST tag helps the CH1 domain to maintain proper folding. Therefore, in the absence of
GST, the binding between HIF-1α and CH1 domain becomes less structured. Therefore,
for the FP assays with tagless CH1-p300, the concentration of CH1-p300 was taken
higher than 544 nM, but lower than the saturation concentration.
114
3.5.3 Fluorescence polarization assays to monitor the disruption of TAD-C HIF-1α
and CH1 p300 binding by chetomin. FP assays were conducted to monitor whether
chetomin is able to disrupt the binding of HIF-1α-Fl and GST-CH1-p300. The
concentration of HIF-1α-Fl was kept at 15 nM and GST-CH1-p300 concentration was 75
nM for all the samples. 75 nM for GST-CH1-p300 was chosen as it is higher than 45 nM
but lower than the saturation value. To each well having 15 nM HIF-1α-Fl and 75 nM
GST-CH1-p300 a different concentration of chetomin was added.
Plotting the log of chetomin concentration against the percentage change in the FP values
gave graph shown in Figure 3.19. The plot clearly shows that chetomin does disrupt the
binding of HIF-1α-Fl and GST-CH1-p300.
Figure 3.18 Fluorescence polarization competition assay shows the disruption of the
complex between HIF-1α TAD-C and GST-CH1-p300 by chetomin.
Chetomin gave an IC
50
value of 578 nM in the FP assay. For all the FP assays we used
buffer that contains 50 mM Tris, 150 mM NaCl, NP-40 0.1%, ZnCl
2
100 µM, DTT 1
115
mM, and 10% glycerol was an important component. Pluronic acid at 0.1% of final
concentration was also added to each sample.
3.5.4 Fluorescence polarization assays with ETP-5 to monitor the disruption of TAD-C
HIF-1α and CH1 p300 binding.
Figure 3.19 Fluorescence polarization disruption assay to assess the disruption in binding
of HIF-1α and GST-CH1-p300 binding by ETP-5.
In order to test the ability of ETP-5 to disrupt the HIF-1α and p300 binding, fluorescence
polarization assays were conducted in the same FP buffer was used as used in case of
chetomin. The fluorescence polarization data shows dose-dependent inhibition of HIF-1α
binding to CH1-p300 by ETP-5. The IC
50
calculated for the protein-protein binding
disruption with ETP-5 is 655 nM.
116
Fluorescence polarization assays are good biophysical methods that allow to evaluate the
ability of drugs or compound to inhibitory the chosen protein-protein interaction, but it is
often difficult to choose a better of two candidates based on FP assay.
Figure 3.20 Data from FP assay showing disruption of a complex between HIF-1α TAD-
C and cleaved CH1 domain of p300 with ETP-5. The IC
50
obtained with cleaved CH1
domain is two times higher as compared to the complex of HIF-1α TAD-C and GST-
CH1-p300.
The fluorescence polarization data for chetomin, ETP-3 and ETP-5 show that their IC
50
values toward the protein complex are similar to each other. This suggests, that another
assay is needed to more accurately predict the efficacy of these molecules in vitro. .
3.5.5
15
N NMR of CH1-p300 domain and CH1-p300 complex with HIF-1α TAD-C:
The
15
N labeled GST-p300-CH1 labeled protein was cloned, expressed and purified using
conditions similar to the described by Freedman et al.. (Dial, Sun et al. 2003) also
obtained NMR in similar conditions. According to the authors the tertiary structure of
117
p300-CH1 domain changes between unbound state and upon complexing with HIF-1α.
Similar results were obtained in our preliminary HSQC experiments. Figure 3.22a and
3.22b shows HSQC spectra for the unbound p300-CH1 and p300-CH1 complex with
HIF-1α TAD-C. The spectral changes between the two states are similar fashion to those
seen by Freedman et al. It should be noted that the primary sequence of the p300-CH1
expressed by us, as shown in Table 3.2, was slightly different from that used by
Freedman et al.
Figure 3.21 HSQC-NMR spectra of p300-CH1 domain and p300-CH1 bound to HIF-1α
TAD-C. a) and b) are HSQC spectra obtained by joint efforts of our lab and Arora lab,
for p300-CH1 domain and p300-CH1 complexed with HIF-1α TAD-C.
This collaborative work with Dr. Arora’s lab is still in progress. Conditions are being
optimized to obtain consistent HSQC spectra for p300-CH1 and HIF-1α complex and
future work will include addition of small molecule inhibitors, onorder to study their
impact on the structure of the complex.
3.6 Conclusion. Two dimeric ETPs were designed and synthesized by a biomimetic
approach starting with the structure of chetomin. The two main considerations for
118
designing these compounds were the distance between the disulfide bridges of the two
ETP rings and the flexibility of the linker connecting the two ETPs.
Our SPR assays have shown that both ETP-3 and ETP-5 bind to the CH1 domain of p300
with low micromolar affinity. It was also demonstrated that ETPs in a reduced form
interact with ~30-fold higher affinity as compared to the non-reduced form. The control
DKP compound NP-481 lacking the disulfide bridges is a poor binder toward the CH1
domain. This underscores the importance of ETP sulfurs under reducing conditions in
mediating high-affinity interaction with the protein.
Fluorescence polarization competition assays confirmed the ability of ETP-3 and ETP-5
to disrupt the complex between HIF-1α TAD-C and p300-CH1. Whereas SPR and FP
assays suggest that designed ETPs interact with the CH1 domain of p300, our CD
experiments indicate that this interaction resuls in a disruption of a global folding of this
domain, thereby altering its function.
119
Chapter 4: Transcriptional Regulation of HIF-1α Inducible Genes with Designed
Dimeric Epidithiodiketopiperazines.
120
4.1 Regulation of transcription of hypoxia inducible genes. After establishing the
feasibility of the inhibition of the complex between CH1-p300-GST fusion protein and
fluorescently labeled C-TAD fragment of HIF-1α with dimeric ETPs (Chapter 3, vide
supra) and measuring thermodynamic parameters of direct binding of bis-ETPs to the
CH1 domain of p300 by SPR, we next set out to establish the efficacy of these molecules
in vitro and to study their impact on transcription of hypoxia-inducible genes. An
important factor that underpins the ability of these molecules to carry out their biological
function in cells is their uptake and nuclear localization where they will carry out their
function.
Figure 4.1 Mechanism of downregulation of HIF-1α inducible genes with bis-ETPs.
Under hypoxia, HIF-1α is in the nucleus and binds to its cofactor p300/CBP, whose
histone deacetylase activity results in chromatin remodeling and elevated levels of
transcription of hypoxia responsive genes. The bis-ETPs chetomin, ETP-3 and ETP-5 can
inhibit the interaction of HIF-1α transactivating domain with p300/CBP and
downregulate the transcription of hypoxia inducible genes.
Under physiological conditions, the bridged disulfide moiety can exist either in disulfide
or dithiol forms and is thought to be essential for uptake of this class of compounds. This
121
hypothesis is supported by our preliminary results and by the work of Bernardo and
Waring who have shown that only the natural (oxidized) form of
epidithiodiketopiperazine is actively concentrated in live cells in a glutathione-dependent
manner. Intracellular levels of the ETP can be up to 1500-fold greater than the applied
concentration, and ETP in the cells exists almost exclusively in the reduced form
(Bernardo, Brasch et al. 2003).
These molecules after entering the cells must enter the nucleus and recruit the p300
coactivator by interacting with its CH1 domain. The coactivator p300 or its homolog
CBP are histone deacetylases that are involved in uncoiling of the genomic DNA from
histones (Ogryzko, Schiltz et al. 1996). Their association inside the nucleus leads to the
transcription of genes with hypoxia response elements (HRE) sequence in their
promoters. The designed bis-ETPs and chetomin are expected to bind to the CH1 domain
of p300 and disrupt its association with HIF-1α, leading to the downregulation of hypoxia
inducible genes that contain HRE (Figure 4.1).
4.2 Inhibition of HIF-1α inducible gene expression by chetomin and ETP-3. In the first
set of experiments we employed MDA-MB-231 cell line having a 5x tandem repeat of
HRE sequence in the promoter region fused to luciferase gene (Brasier, Tate et al. 1989).
This MDA-MB-231-Luc cell line overexpresses luciferase under hypoxic conditions that
could be either induced physiologically or chemically. The ratio of the luciferase levels at
hypoxia to the luciferase levels at normoxia represents the inducibility of the promoter
under specified treatment conditions. When the MDA-MB-231-Luc cells were treated
with chetomin, ETP-3 and NP-481 (control compound) under hypoxia and normoxia, the
ratio of luciferase induction decreased by >95% for chetomin and >80% for ETP-3 as
122
compared to the values vehicle Treatment with NP-481 control compound resulted only
in a slight inhibition of the promoter activity. ETP-3 shows dose dependent inhibition in
the luciferase activity with higher inhibition obtained at 600 nM as compared to 200 nM.
Desferroxamine (DFO) was used to induce hypoxia (Wang and Semenza 1993). Based on
our experience, DFO is a suitable mimetic of the hypoxic state in many cancer cells lines.
The mechanism of DFO induction of chemical hypoxia is based on chelation of the Fe
2+
ions and, as a result, inactivation of the prolyl hydroxylases (Guo, Song et al. 2006; Woo,
Lee et al. 2006).
Figure 4.2 Luciferase assay data for chetomin, ETP-3, and NP-481 in MDA-231-HRE-
Luc cell line. Chetomin and ETP-3 showed significant reduction in the activity of the
hypoxia-inducible promoter. Error bars are ± s.e.m. of experiments performed in
triplicate. *** P < 0.001, ** P < 0.01, * P < 0.05, t test.
123
After obtaining encouraging results with the luciferase reporter assays, the transcriptional
levels of endogenous VEGF gene were measured upon treatment with chetomin, ETP-3
and NP-481. VEGF is the gene encoding vascular endothelial growth factor, a
proangiogenic factor secreted by endothelial cells under hypoxia. Treatment with ETP-3
showed dose dependent inhibition of VEGF mRNA levels at 200 and 600 nM
concentrations. Control compound NP-481 showed no decrease of VEGF mRNA levels
(Block, Wang et al. 2009).
Figure 4.3 Downregulation of the expression of VEGF gene in MCF7 cells treated with
chetomin, ETP-3 and NP-481. mRNA levels of VEGF gene in MCF7 after treatment with
chetomin (200 nM) and ETP-3 and NP-481 at 200 nM and 600 nM were quantified by
real time qRT-PCR. Data shows that chetomin and ETP-3 downregulate the levels of
VEGF under hypoxia whereas NP-481 does not show any effect on the transcription of
124
VEGF. Error bars are ± s.e.m. of experiments performed in triplicate. *** P < 0.001, * P
< 0.05, t test.
These experiments clearly suggest that bis-ETP 3 is an effective inhibitor of the HIF-1α
inducible transcription factor complex.
Figure 4.4 LOX mRNA levels in MCF7 cell line after treatment with chetomin, ETP-3
and NP-481. At hypoxia, significant transcriptional downregulation is observed with
chetomin and ETP-3, whereas the control compound NP-481 did not have significant
effect on the LOX mRNA levels. Error bars are ± s.e.m. of experiments performed in
triplicate. *** P < 0.001, t test.
Lysyl oxidase (LOX) is an important gene involved in invasion and metastasis in cancer
cells which is induced by HIF dependent pathway. Chetomin and ETP-3 downregulate
the transcriptional levels of LOX gene under hypoxia (Block, Wang et al. 2009).
4.3 Evaluation of ETP-5 by HIF inducible luciferase reporter assay. After the
successful testing of our first designed synthetic ETP-3 we next designed optimized
ETP-5 having where the linker between the two ETP rings was optimized, with
125
conformationally averaged distances between the bridging disulfide that match the
distances between the zinc ions within the CH1 domain of p300.
Figure 4.5 Results from the luciferase reporter assay with ETP-5. Relative luciferase
activity data for chetomin (200 nM), ETP-3 (200 nM & 600 nM), ETP-5 (200 nM & 600
nM), and NP-481 (200 nM & 600 nM) in MDA-231-HRE-Luc cell line. ETP-5 showed
better effect than ETP-3 and was very similar in efficacy to chetomin (>95% inhibition).
Error bars are ± s.e.m. of experiments performed in triplicate. *** P < 0.001, ** P < 0.01,
* P < 0.05, t test.
In our luciferase reporter activity assays ETP-5 showed more pronounced inhibition of
hypoxia inducible transcription as compared to ETP-3. At both 200 and 600 nM
concentrations, ETP-5 significantly reduced the luciferase expression levels in a dose
dependent manner and was more efficacious than ETP-3.
4.4 Cytotoxicity of ETP-5 in MCF7 breast cancer cell line and A549 lung epithelial
adenocarcinoma cell line. One potential issue that arises with the use of ETPs as
126
transcriptional inhibitors is their cytotoxicity. Therefore, careful assessment of the
cytotoxicity is crucial for every small molecule that acts as a transcriptional inhibitor in
order to rule out non-specific, global effects on transcriptional machinery.
Figure 4.6 MTT cytotoxicity assay for ETP-5 in MCF7 cells. Cells were maintained in
RPMI-1640 media supplemented with 10% FBS. Cells were treated with different
concentrations of ETP-5 for 24 hours and the amount of purple formazan formed was
determined via a UV spectrophotometry (Denizot and Lang 1986).
We performed cytotoxicity experiments in order to obtain the EC
50
values of ETP-5 in
MCF7 breast cancer cell line and A549 lung adenocarcinoma cell line. The goal was to
determine the window of viable concentrations and perform our transcription inhibition
experiments at concentrations significantly below the EC
50
values in these cell lines.
In our previous work (Block, Wang et al. 2009) we reported the EC
50
value for chetomin
in MCF7 cells to be 180 nM.. We found he newly designed ETP-5 to be much less
127
cytotoxic towards MCF7 cells as compared to chetomin. In the MTT cell cytotoxicity
assay in MCF7 cells, the EC
50
value obtained for was 547 nM after 24 h treatment with
ETP-5. Based on this EC
50
value of ETP-5 in MCF7 cells we chose to measure its effect
on HIF inducible transcription at a maximum concentration of 400 nM. This is important
in order to minimize the nonspecific effects on mRNA levels due to reduction in cell
viability.
Cell line A549 is lung epithelial adenocarcinoma of non-small cell type that is known to
exhibit significant up-regulation of key HIF-1α inducible genes, such as c-Met, VEGF
and Glut1 under hypoxic conditions. In our viability assays in this cell line ETPs showed
less cytotoxicity as compared to MCF7 cell line. After 24 h treatment with both chetomin
and ETP-5 in Kahn modified F-12 media, an EC
50
of >10 M as observed. Therefore the
treatment of the cells was extended to 48 h in order to better determine its cytotoxicity.
Figure 4.7 MTT cytotoxicity assay data for chetomin in A549 cell line. The cells were
treated with different concentrations of compound for 48 hours in serum free F-12K
medium.
128
The EC
50
value obtained for chetomin was 0.9 µM in A549 cell line after treatment for 48
hours. For comparison, treatment of MCF7 cell line for 24 hours with chetomin chetomin
gave EC
50
of 0.2 µM, indicating significantly higher cytotoxicity of the compound in that
cell line.
Figure 4.8 MTT cytotoxicity assay data for ETP-5 in A549 cell line. The cells were
treated with different concentrations of compound for 48 hours in serum free F-12K
medium.
MTT cytotoxicity assay was carried out for ETP-5 in A549 cell line for 48 h under
similar conditions. The EC
50
for ETP-5 obtained from this assay after 48 h treatment was
2.8 µM. This value is about five times higher than the EC
50
value obtained in MCF7 cell
line after 24 h treatment. These data suggest that A549 cell line is much more robust
toward treatment as compared to MCF7 cell line. In addition, ETP-5 is clearly much less
toxic to cells as compared to chetomin. Since ETP motifs are common in both ETP-5 and
chetomin, we could only speculate that higher toxicity of chetomin may be due to its
cyclotryptophan motif, that is absent in ETP-5.
129
4.5 Transcriptional regulation of HIF-1α genes in MCF7 cells treated with ETPs. We
next evaluated the potential of ETP-5 to inhibit transcription of hypoxia-inducible genes.
MCF7 cell line is a non-metastatic tumorigenic line which shows moderate up-regulation
of those genes. Two genes were analyzed by qRT-PCR: VEGF and c-Met (Maxwell,
Dachs et al. 1997; Pennacchietti, Michieli et al. 2003).
The MCF7 cells were treated with 100 µM of CoCl
2
to chemically induce hypoxia. Six
hours prior to induction of hypoxia, the required amounts of ETP-5, chetomin and control
mon-ETP LS75 were added to media. The final concentration was 200 nM for chetomin
and 200 nM and 400 nM for ETP-5. Similarly, for the control compound LS75 was at
two concentrations of 200 nM and 400 nM.
In MCF7 cells, VEGF gene is induced ~3-fold at hypoxia. Both chetomin and ETP-5
downregulated the VEGF mRNA in hypoxic cells essentially to its normoxic levels.
Moreover, for ETP-5, the observed inhibition of VEGF transcription was in a dose-
dependent manner. The control compound mono-ETP LS75 was, as expected, less
effective than bis-ETPs.
130
Figure 4.9 Quantification of VEGF transcriptional levels in MCF-7 cell lines after
treatment with ETPs chetomin, ETP-5 and LS75. Both normoxic and hypoxic mRNA
levels are shown. Hypoxia was induced chemically by 100 µM CoCl
2.
Under hypoxia
VEGF levels are induced more than 3-fold. Both chetomin and ETP-5 show efficacy in
inhibiting of VEGF transcription under hypoxia. LS75 show reduced efficacy as
compared to bis-ETPs chetomin and ETP-5. Error bars are ± s.e.m. of experiments
performed in triplicate. *** P < 0.001, ** P < 0.01, * P < 0.05, t test.
Surprisingly, chetomin also slightly inhibits the VEGF levels at normoxia, suggesting that
higher toxicity of chetomin may play a role in inhibition of basal levels of hypoxia-
inducible gene expression.
c-Met gene is another important downstream gene target of hypoxia-inducible
transcription factor system. It has five repeats of HRE sequence in its promoter region
and Comoglio et al. have shown that HRE 4 and HRE 5 are mainly responsible for the
hypoxia inducible transcription of c-Met gene. Mutation or deletion of HRE 4 and HRE 5
in the promoter sequence of c-Met gene significantly diminishes the hypoxia inducible
induction of its transcription. The mRNA as well as the protein levels of c-Met are
significantly upregulated under hypoxia in many cancer cell lines and most of these
131
cancer cell lines are typically metastatic in nature. In our system with MCF7 cells grown
in serum- containing media, very low induction of c-Met transcription was achieved.
Since the c-Met induction is low in MCF7 cells therefore modulation of c-Met mRNA
levels could not be effectively achieved upon treatment with ETP-5. That’s being said,
chetomin lowers the c-Met levels both in normoxia and hypoxia and no significant
difference in mRNA levels was observed in the two states.
Figure 4.10 Quantification of c-Met transcriptional levels in MCF7 cells. MCF7 cells
under normoxia and hypoxia after treatment with bis-ETPs chetomin (150 nM) and ETP-
5 (200 nM and 400 nM). Conditions: 10% serum in the media and 100 µM CoCl
2
to
induce hypoxia. Error bars are ± s.e.m. of experiments performed in triplicate. ** P <
0.01, t test.
In order to achieve higher induction of hypoxia-inducible transcription of c-Met and
VEGF genes different conditions were tested. Of these, DFO (desferroxamine) was found
132
to be the best method of HIF-1α dependent induction of VEGF and c-Met genes in this
cell line.
Figure 4.11 VEGF and c-Met mRNA levels in MCF7 cells grown in serum-free media
under hypoxia and normoxia and treated with ETP-5 for 24 h. Hypoxia was induced with
300 µM DFO. VEGF level showed ~5 fold increase under these conditions while c-Met
also showed noticeable induction under hypoxia. Error bars are ± s.e.m. of experiments
performed in triplicate. ** P < 0.01, * P < 0.05, t test.
After finding the best method of hypoxia induction we assessed how different media and
serum amounts impact the levels of hypoxia-inducible gene expression. We found that in
serum free conditions the hypoxic induction was much more pronounced as compared to
media containing the regular amount (10%) of serum.
Therefore, the conditions that consistently gave better hypoxic induction in MCF7 cells
are induction with DFO in serum-free media. Under these conditions more than five fold
induction was achieved for VEGF whereas for c-Met consistently 1.5- to 2-fold induction
was observed. Since the cells growth rate is reduced and uptake of ETPs is increased in
serum-free conditions, lower concentration of 100 nM for ETP-5 was used to treat the
133
cells for 24 h. At this concentration both VEGF and c-Met showed inhibition of
transcription.
Figure 4.12 Optimization of three control genes viz. β-glucuronidase, RPL32 and β-actin
for normalization of c-Met mRNA levels. In order to confirm that β-glucuronidase gene is
the correct control gene for c-Met normalization two more genes was carried out with β-
actin and RPL32 as housekeeping genes. The c-Met mRNA levels normalized against the
three different genes at normoxia and hypoxia showed similar results in all three cases. In
order to rule out the variation in control genes upon treatment with bis-ETPs the mRNA
levels in cells treated with chetomin (150 nM) and ETP-5 (400 nM) were quantified with
normalization to the three control genes. Both β-glucuronidase and RPL32 showed
similar results and were better than β-actin gene as a control. Error bars are ± s.e.m. of
experiments performed in triplicate. # P < 0.1, t test.
Under these conditions, c-Met did not show high induction levels and therefore other cell
lines were later tested in order to devise a good model system to study the ETPs.
One important factor that needs to be taken care of while monitoring the transcriptional
levels of different genes is validation of the housekeeping gene. The control gene in the
134
data presented above is β-glucuronidase (de Kok, Roelofs et al. 2004). Although this gene
has been reported previously as a suitable housekeeping gene for studying hypoxic
response it was necessary to validate this gene again under our conditions of serum-free
media. To accomplish that, we tested it in MCF7 cells under our specified treatment
conditions (chetomin (150 nM), ETP-5 (400 nM) at normoxia and at hypoxia induced
with 100 µM CoCl
2
) by measuring three different control genes viz. β-glucuronidase, β-
actin (de Kok, Roelofs et al. 2004), and RPL32 (Kriegova, Arakelyan et al. 2008). All the
three genes are well established housekeeping genes and β-glucuronidase and RPL32
have been used by many to monitor mRNA levels in hypoxia. β-Actin, on the other hand,
has shown variation under hypoxic conditions and is, therefore, not a suitable control
gene. Our data (Figure 4.12) confirm that under our conditions of treatment β-
glucuronidase and RPL32 do not change their expression levels and, therefore are
suitable as reporter genes whereas the levels of β-actin have been variable. This result has
validated both β-glucuronidase and RPL32 as controls for our in vitro system.
Figure 4.13 Glut1 mRNA levels in MCF7 cells after treatment with ETP-5. Glut1 mRNA
levels in MCF7 cells after treatment with ETP-5 (200 nM & 400 nM) in hypoxia showed
dose-dependent inhibition of transcription. Error bars are ± s.e.m. of experiments
performed in triplicate. ** P < 0.01, t test.
135
Facilitated glucose transporter 1 (Glut1) is another key gene that is up-regulated in
chronic hypoxia (Behrooz and IsmailBeigi 1997). This gene is involved in glucose
metabolism and is overexpressed in many cancer cells. Under hypoxic conditions, Glut1
is induced about five fold in MCF7 cells. Upon treatment with ETP-5 (200 nM and 400
nM) for 24 h the mRNA levels are downregulated in a dose-dependent manner.
4.6 Transcriptional changes in MDA-MB-231 cell line upon treatment with ETP-5. In
order to obtain higher c-Met induction and possible inhibition of hypoxic levels of its
transcription with ETPs, a different cell line, MDA-MB-231 breast cancer cell line was
used. This unlike MCF7 is a highly metastatic cell line. In this cell line, under hypoxic
induction with CoCl
2
no c-Met induction was observed, but upon treatment with ETP-5 at
three different concentrations of 50 nM, 200 nM and 400 nM under hypoxia, a dose
dependent inhibition of c-Met transcription was observed.
Figure 4.14 In MDA-MB-231 cell line ETP-5 showed inhibition of hypoxia inducible
transcription of c-Met gene. At 400 nM the ETP-5 showed better inhibition as compared
to chetomin. Error bars are ± s.e.m. of experiments performed in triplicate. ** P < 0.01, *
P < 0.05, t test.
136
This suggests that although no net increase of c-Met mRNA levels takes place upon
induction of hypoxia but the mechanism of transcription of c-Met shifts to HIF-1α
pathway and thus the ETP-5 is able to inhibit it.
Under normoxia the c-Met gene is known to undergo transcription with the help of ETS-1
protein (Gambarotta, Pistoi et al. 1994). There is an ETS-1 putative binding site in the
promoter region of c-Met gene. Most of the hypoxia inducible genes that have HRE in the
promoter region also have other putative binding sites in their promoter region that bind
to transcriptional factors like ETS-1, AP1 etc. which are involved in the transcription of
these genes under normoxic conditions (Dittmer 2003). c-Met gene has an ETS-1 and an
AP-1 binding site in its promoter region. It is extremely important to note that ETS-1
interacts with CH1 domain of CBP/p300 (Yang, Shapiro et al. 1998). ETS-1 is also
known to be involved in transcription of many hypoxia inducible genes (Oikawa, Abe et
al. 2001; Salnikow, Aprelikova et al. 2008). It is likely that in hypoxic conditions the
mechanism of transcription of c-Met predominantly shifts from ETS-1 and AP-1
regulated transcription to HIF-1α regulated transcription.
4.7 Modulation of HIF-1α inducible genes transcription levels with ETP-5 in A549
lung adenocarcinoma line. Finding of a good in vitro model that consistently displays
high transcriptional activation of hypoxia-inducible genes turned out to be a challenging
task. After evaluating several cell lines we focused our attention on A549 cells, a non-
small cell lung adenocarcinoma cell line. It has been reported that A549 cell line
produces robust upregulation of key HIF-inducible genes under hypoxia conditions.
Specifically, Comoglio et al. (Pennacchietti, Michieli et al. 2003) reported that under
hypoxia c-Met mRNA level is significantly upregulated in A549 cell line.
137
Initially, the cells were reared in F-12K media with 10% serum and hypoxia was induced
by CoCl
2
(100 µM). mRNA levels of VEGF and c-Met genes were measured by qRT-
PCR. For VEGF, very little hypoxia induction has been observed, while, surprisingly, for
c-Met the mRNA levels were lower than that in normoxia. In both cases the reduction of
hypoxic levels of VEGF and c-Met, was observed after treatment of the cells with ETP-5
for 24 h (Figure 4.15).
Figure 4.15 qRT-PCR of mRNA isolated from A549 lung adenocarcinoma cell line
grown in F-12K medium supplemented with 10% serum. Hypoxia was induced
chemically with CoCl
2
(100 µM). Cells were treated with ETP-5 for 24 h with or without
induction of hypoxia. A) mRNA levels of VEGF in normoxia, hypoxia, normoxia with
200 nM ETP-5 and hypoxia with 200 nM ETP-5. B) c-Met mRNA levels in the same four
conditions. Error bars are ± s.e.m. of experiments performed in triplicate. ** P < 0.01, t
test.
138
Different serum levels were tried in growth media. Cells were grown in 5% and 2% fetal
bovine serum in F-12K media and hypoxia was induced with CoCl
2
(100 µM) for 48 h.
These conditions did not show satisfactory induction levels either for VEGF or c-Met.
After tests with various serum levels in the media and hypoxia induction methods, the
conditions that worked remarkably and consistently well for the induction of HIF-1α
dependent genes were to keep A549 cells in 2% serum followed by treating cells with
compound or control in the media with 0.2% serum for 48 h.
Figure 4.16 mRNA levels of three HIF-1α inducible genes: VEGF, c-Met and Glut1 in
A549 cells after treatment with ETP-5. Data from qRT-PCR experiments showing
mRNA levels of three HIF-1α inducible genes, VEGF, c-Met and Glut1 in A549 after
treatment of the cells in a medium with 0.2% serum with ETP-5. Hypoxia was induced
300 µM by DFO . Error bars are ±sem for the experiments performed in quadruplicate.
Error bars are ± s.e.m. of experiments performed in triplicate. *** P < 0.001, ** P < 0.01,
t test.
139
Under these conditions while hypoxia bag was the best option, especially for induction of
the LOX gene, whereas the best hypoxic response leading to up-regulation of many other
HIF-1α inducible genes was treatment with 300 µM DFO.
Figure 4.16 shows effect of treatment with ETP-5 on the levels of three important genes
VEGF, Glut1 and c-Met which are known to be up-regulated in many solid tumors under
hypoxic conditions. Under our conditions hypoxia is induced in A549 cells with DFO
(300 µM) in the media 0.2% with serum for 48 h. We observed significant up-regulation
of c-Met, which is more than three-fold induction, is seen as compared to normoxic
levels.
Figure 4.17 qRT-PCR data for LOX and CXCR4 genes in A549 cells treated with ETP-5.
Hypoxia was induced by hypoxia bag. Cells were maintained in F-12K medium with 2%
serum. After reaching 65% confluency the cells were grown in serum free media and
treated with ETP-5. Hypoxia was induced with DFO (300 µM) for 48 h. Error bars are ±
s.e.m. of experiments performed in triplicate. ** P < 0.01, t test.
140
The cells were initially grown in 2% serum for two weeks while subculturing them from
time to time upon reaching 100% confluency. After two weeks the cells were plated in a
media with 0.2% serum.
Under these conditions, high induction levels of c-Met VEGF and Glut1 mRNA were
observed. VEGF mRNA showed more than 50 fold increase while Glut1 mRNA levels
showed more than six fold increase in the course of induction. Treatment with ETP-5
resulted in a significant reduction in the hypoxic response of VEGF, Glut1 and c-Met
genes. VEGF levels were reduced by 50%, whereas Glut1 mRNA levels were reduced
more than 60%. c-Met was also significantly downregulated, essentially reaching its
normoxic levels (Figure 4.16).
LOX (lysyl oxidase) is another gene that is up-regulated under hypoxia and the protein is
involved in regulating the extracellular matrix during invasive behavior and metastasis of
cancer tissue (Erler, Bennewith et al. 2006). LOX gene showed better induction with
hypoxia bag after 48 h and showed significant down-regulation in the transcriptional
activity after treatment with ETP-5. CXCR4 is a gene that is essential for chemotaxis of
stem cells and progenitor cells during healing of an injury that is also implicated in
cancer stem cells migration (Darash-Yahana, Pikarsky et al. 2004). SDF1-CXCR4 axis
leads to chemotaxis of progenitor and stem cells to the cancer tissue or wound followed
by their differentiation. In our model system with of A549 cells CXCR4 is also up-
regulated more than 100-fold after chemical induction of hypoxia with DFO or hypoxia
bag. Upon treatment with ETP-5 at 400 nM concentration, excellent inhibition of the
transcriptional activity was observed for CXCR4 gene (Figure 4.17).
141
Overall, A549 cell line under the conditions mentioned above became a very good model
for studying HIF-1α inducible gene expression. All the five genes mentioned above not
only showed high up-regulation of HIF-1α inducible transcription of many key genes
involved in tumorigenesis but also under the given conditions showed very little change
in the transcriptional activity under normoxia in the presence of ETP-5.
Figure 4.18 mRNA levels for VEGF in A549 cell line, illustrating the dose response to
ETP-5 treatment at three different concentrations. qRT-PCR assays were performed in
order to determine the mRNA levels for VEGF in A549 cell line treated with ETP-5 at
concentrations: 100 nM, 400 nM, 1600 nM. Hypoxia was induced by DFO (300 µM).
Error bars are ±sem for the experiments performed in quadruplicate. Error bars are ±
s.e.m. of experiments performed in triplicate. ** P < 0.01, * P < 0.05, t test.
After obtaining great transcriptional induction for the five genes VEGF, c-Met, Glut1,
LOX and CXCR4 which are up-regulated by HIF-1α transcriptional; system and are
142
downregulated upon treatment with 400 nM of ETP-5, the next logical step was to study
the drug dose response. The modulation of HIF-1α inducible transcription with ETP-5
was studied at three different concentrations of 100 nM, 400 nM and 1600 nM.
The hypoxia induction was done at a confluency of 85% cells in serum free F-12K
medium with 300 µM DFO. Under these conditions the VEGF up-regulation under
hypoxia was similar to that achieved when hypoxia was induced in cells at 65% of
confluency.
Figure 4.19 mRNA levels for c-Met in A549 cell line showing dose response to ETP-5 at
concentrations of 100 nM, 400 nM, 1600 nM. qRT-PCR was used to determine the
mRNA levels of c-Met Hypoxia was induced with DFO (300 µM). Error bars are ±sem
for the experiments performed in quadruplicate. Error bars are ± s.e.m. of experiments
performed in triplicate. *** P < 0.001, ** P < 0.01, # P < 0.1, t test.
143
All samples were collected in triplicates and at three different concentrations of 100 nM,
400 nM, and 1600 nM. For each concentration of ETP-5 control samples were also
present, where the cells were treated with ETP-5 but without induction of hypoxia. The
controls showed that at the three different concentrations of ETP-5 in normoxia the
VEGF transcription levels were not changed significantly, underlining the fact that in
these conditions ETP-5 did not show increase or decrease in the transcriptional levels due
to stress or some other pathway. Under hypoxia ETP-5 showed dose dependent decrease
in the HIF-1α inducible transcription of VEGF gene.
Figure 4.20 qRT-PCR data for Glut1 mRNA in 85% confluent A549 cells treated with
ETP-5 at three different concentrations. Hypoxia was induced with DFO (300 µM).
Error bars are ±sem for the experiments performed in quadruplicate. Error bars are ±
s.e.m. of experiments performed in triplicate. *** P < 0.001, ** P < 0.01, t test.
144
c-Met gene under these conditions of hypoxic induction to highly confluent cells showed
enhanced upregulation of its transcription. c-Met mRNA was up-regulated more than 5
folds in hypoxia. Dose-dependent decrease in transcriptional upregulation was observed
for c-Met upon treatment with ETP-5.
Glut1 which showed 10-fold induction in highly confluent cells. Glut1 also shows dose
dependent decrease of hypoxic transcriptional up-regulation at 100 nM, 400 nM and 1600
nM of ETP-5.
Figure 4.21 qRT-PCR data for A549 cells where hypoxia was induced in 85% confluent
cells. More than 650 fold induction of CXCR4 transcription was observed with DFO at
300 µM concentration. ETP-5 at two different concentrations downregulatedd the CXCR4
mRNA levels in a dose dependent manner. Error bars are ±sem for the experiments
performed in quadruplicate. Error bars are ± s.e.m. of experiments performed in
triplicate. ** P < 0.01, # P < 0.1, t test.
CXCR4 is a g-protein coupled receptor that is upregulated under hypoxic conditions. We
chose A549 cells and induced the hypoxia with DFO and found that in 85% confluent
145
cells the levels of CXCR4 gene were transcriptionally overexpressed more than 650-fold.
Upon treatment with ETP-5 at 400 nM and 1600 nM concentration a dose-dependent
decrease in mRNA levels of CXCR4 could be observed. These finding not only show that
CXCR4 is induced transcriptionally in A549 cells under hypoxic conditions but that it can
be downregulated with small molecules targeting HIF-1α pathway.
An interesting observation was made when Monoamine oxidase A (MAO-A) mRNA
levels were found in A549 cells by. Monoamine oxidase A (MAO-A) mRNA levels have
been reported to be down-regulated in ~95% of cancers (Rybaczyk, Bashaw et al. 2008).
A notable exception is prostate cancer where MAO-A levels have been shown to be up-
regulated (Peehl, Coram et al. 2008; Flamand, Zhao et al. 2010). When the cells were
treated with ETP-5 the MAO-A levels were increased in a dose dependent manner which
shows that as the HIF-1α pathway was being down-regulated there seems to be an
increase in the MAO-A levels (APPENDIX C, S4.1). Thus in A549 cells MAO-A shows
inverse relationship with HIF-1α pathway.
4.8 HIF-1α protein levels are unchanged in A549 cells treated with ETP-5 under
hypoxia and normoxia. In the study of transcriptional regulation of HIF-1α pathway with
ETP-5 the proposed mechanism is the inhibition of interaction between HIF-1α and
p300/CBP. Hence, it is important to study the changes in the HIF-1α protein levels upon
hypoxia induction as well as upon the treatment with ETP-5. In order to rule out the
possibility that ETP-5 can also alter the levels of HIF-1α under hypoxia, we performed
western blot in order to determine the levels of HIF-1α in cells treated with ETP-5 at a
concentration of 1600 nM, the highest concentration used to study transcriptional and
compare it to the levels of HIF-1α in control cells at normoxia and hypoxia,. As seen in
146
Figure 4.22, HIF-1α levels are practically absent under normoxia in cells treated with or
without ETP-5 and under hypoxia similar increase in HIF-1α protein levels are seen in
cells treated with or without ETP-5.
Figure 4.22 HIF-1α protein levels are not affected by the treatment with ETP-5. A549
cells were treated with ETP-5 (1600 nM) in presence or absence of DFO (300 µM) to
mimic hypoxia. Lamin-A/C protein levels were used as control.
Thus, the data shows that the treatment with ETP-5 does not affect the HIF-1α protein
levels and the observed transcriptional inhibition of HIF-1α pathway is not due to the
change in HIF-1α stability.
4.9 VEGF and c-Met protein levels in MCF7 and MDA-MB-231 cell lines upon
treatment with ETPs. We analyzed the effect of ETP 5 and controls on the levels of
VEGF and c-Met proteins, which are downstream targets of hypoxia inducible
transcription factor complex. To determine VEGF protein levels, MCF7 cells were
treated with chetomin (200 nM), ETP-5 (400 nM) and LS75 (400 nM). Western blot
analysis shows significant downregulation of VEGF protein levels upon treatment with
ETP-5 (Figure 4.23a).
147
Figure 4.23 ETP-5 down-regulates the protein levels of (a) VEGF in MCF7 cells and (b)
c-Met MDA-MB-231 cells in hypoxic cells. Data from the western blot assay under
normoxia (DFO (-)) and hypoxia (DFO (+)) conditions upon treatment with chetomin,
ETP-5 and LS75. CTM (chetomin) was at 200 nM, ETP-5 and LS75 were at 400 nM,
respectively. Hypoxia was mimicked with 300 µM DFO (a) or 100 µM CoCl
2
(b),
respectively. Error bars are ± s.e.m. of experiments performed in triplicate. *P < 0.05, #P
< 0.1, t test.
To study c-Met tyrosine kinase levels, MDA-MB-231 cell line was used. MDA-MB-231
cells were treated with chetomin (200 nM), ETP-5 (400 nM) and LS75 (400 nM). Both
ETP-5 and chetomin showed significant down-regulation of c-Met protein levels,
148
whereas control monomeric ETP LS75 did not result in a decrease of the levels of that
protein (Figure 4.23b).
4.10 Gene expression profiling and microarray analysis. Since the target proteins p300
and CBP are pleiotropic multidomain coactivators, their CH1 regions contain binding
sites for multiple transcription factors. One potential concern of the use of ETPs for gene
regulation is specificity, because inhibiting the interaction between CBP/p300 and
transcription factors other than HIF-1α may result in large numbers of affected genes. To
rule out nonspecific genome-wide effects of ETPs, we conducted in vitro gene expression
profiling experiments with ETP-5 using Affymetrix Human Gene ST 1.0 Arrays
containing oligonucleotide sequences representing 28,869 transcripts (Pradervand,
Paillusson et al. 2008; Affymetrix 2012).
Figure 4.24 shows agglomerative clustering of genes under different conditions of
hypoxia and treatment of ETP-5 (400 nM). The clustering shows that in many genes the
effect of ETP-5 under hypoxia is to nullify the effect of hypoxia such that many genes
transcriptional levels behave similar to that as seen in vehicle i.e. normoxia without ETP-
5 treatment.
In order to interrogate cellular genome for global effects, MCF7 cells treated with
ETP-5 at 400 nM were used (Figure 4.25). Treatment of cells with ETP-5 at a
concentration of 400 nM affected the expression of only 178 genes at > 2.0 fold levels.
By comparison, treatment with DFO alone changed levels of 329 genes > 2.0 fold. Of
these, 88 genes were downregulated > 2.0 fold and 90 – upregulated by > 2.0 fold,
respectively. In cells treated with ETP-5 under DFO-induced hypoxia conditions, we
149
identified 190 genes were affected by this compound. Clustering analysis was performed
to identify similarities in the expression profiles between the different treatments.
Figure 4.24 Microarray genome analysis of MCF7 cells treated with or without treatment
of ETP-5 and with or without hypoxia induction using DFO (300 µM). Clustering
analysis was done to see the similar trends in genes among different conditions. The
analysis shows that the MCF7 cells under hypoxia and treated with ETP-5 (400 nM)
shows similar trends as seen in vehicle which suggests that ETP-5 works towards
nullifying the effect of hypoxia on global transcriptional levels.
The expression profile of cells treated with ETP-5 under DFO-induced hypoxia is
largely different from the profile under DFO alone. However, the profiles of the cells
treated with ETP-5 under DFO-induced hypoxia and cells under normoxia conditions are
150
showing distinct regions of similarity. This suggests, that treatment with ETP-5 reduces
the effect of DFO treatment on certain group of genes, as expected for the transcriptional
inhibitor that affects hypoxia-inducible genes. It is not entirely surprising that there is
some overlap in genes affected by both ETP-5 and DFO given the complexity of cellular
signaling pathways involved in the hypoxic response. The results also clearly
demonstrate the high specificity of ETP-5 in its effect on hypoxia-inducible within the
context of the genome.
Figure 4.25 Results from the analysis of microarray data. Green Venn diagram shows
genes that are downregulated in vehicle (left green circle) i.e. genes that are
downregulated in vehicle hypoxia as compared to vehicle normoxia and the right green
circle shows the number of genes downregulated (> 2.0 fold) in hypoxia treated with
ETP-5 (400 nM) as compared to genes in normoxia treated with ETP-5. The red diagram
shows the genes that are up-regulated in same conditions as explained for green Venn
diagram. The blue diagram shows the overall effect of increase or decrease of genes (>
2.0 fold) under the conditions mentioned above.
151
Table 4.1 lists important genes that are down-regulated under hypoxia with 400 nM ETP-
5 treatment in MCF7 cells. The data extracted from the lists of genes that show >2-fold
change.
Interestingly, many genes that belong to solute carrier (SLC) family of proteins were
down-regulated under hypoxia with ETP-5. They are listed in Table 4.2. This shows that
under hypoxia solute carrier proteins are up-regulated to facilitate higher uptake and
secretion of molecules in the cells and ETP-5 has reversed this trend.
Table 4.1 List of important for cancer progression HIF-1α inducible genes that are
downregulated under hypoxic induction with DFO (300 µM) by treatment with ETP-5
(400 nM).
Symbol Entrez Gene ID ETP-5
TGFB3 7043 -1.5
TFRC 7037 -1.5
LOXL2 4017 -1.4
CAV1 857 -1.3
MET 4233 -2.2
SLC35D1 23169 -3.1
152
Table 4.2 List of Solute Carrier (SLC) family genes that are down-regulated in MCF7
cells under hypoxic induction with DFO (300 µM) upon treatment with ETP-5 (400 nM).
Symbol Entrez Gene ID ETP-5
SLC35D1 23169 -3.1
SLC5A8 160728 -1.4
SLC25A15 10166 -1.5
SLC9A3R1 9368 -1.6
SLC39A11 201266 -1.9
SLC9A2 6549 -1.8
SLC5A6 8884 -2.4
SLC25A12 8604 -1.8
SLC26A2 1836 -1.4
SLC38A9 153129 -1.5
SLC35A1 10559 -2.2
SLC7A2 6542 -2.5
SLC27A4 10999 -1.4
153
4.11 In vivo study of the efficacy of ETP-5 in mouse tumor xenografts model using
intravital microscopy. Tumor spheroids from N2O2 (breast carcinoma) were prepared
and implanted subcutaneously into the nude mice. Tumors were allowed to vascularize
for 10-14 days after which mice were injected on Day 0 with 1 mg/kg of ETP-5 via tail
vein. From Day 8 to Day 13 mice were daily injected with 2 mg/kg of ETP-5. (For
details see Chapter 6.4.9 – 6.4.12). Intravital microscopy (IVM) images, obtained on
specified days are shown in Figure 4.26.
Figure 4.26 Intravital microscopy images of murine subcutaneous tumor model of
fluorescent N2O2 cells stably transfected with H2B-GFP construct. Mice with N202
H2B-GFP tumors were injected intravenously on day 0 with 1 mg/kg of ETP-5
compound followed by daily injections after day 8 and imaged over 2 weeks.
Fluorescence IVM images of tumors taken on days indicated post-treatment.
Figure 4.27 is the quantification of the tumor volume obtained from the IVM images. The
data clearly shows that in mice #1 and #2, injected with ETP-5 the tumor vasculature and
tumor growth are significantly suppressed. In the course of these experiments, ETP-5
showed very low toxicity to mice, as confirmed by observation of the behavior of the
animals and monitoring of their body weights. This low toxicity of our designed bis-ETP
154
is giving it a significant advantage in vivo over the natural bis-ETP chetomin, which is
reported to be toxic and even lethal to animals, because mice treated with chetomin do
not survive after five days of consecutive injection.
Figure 4.27 Change in tumor volume obtained from IVM of mice treated with or without
ETP-5. Graphs show the quantitative difference between the tumor volumes as seen in
IVM images in Figure 4.26. Vehicle mouse (-■-) and mice treated with ETP-5, #1 (-▲-)
and #2 (-♦-).
In our study mice treated with ETP-5 survived the 14-day treatment and did not show any
signs of acute toxicity. This study validates, efficacy of ETP-5 as an inhibitor of HIF-1
inducible gene expression in cancer cell lines in vitro and tumor growth in mouse
xenograft model in vivo. ETP-5 is significantly less toxic than chetomin within the tested
range of therapeutic concentrations sufficient to maintain the inhibition of tumor growth
in vivo.
155
4.12 Conclusion
This Chapter describes design of optimized dimeric ETP-5 and study of its efficacy as
inhibitor of HIF-1α inducible transcription factor complex. Our biophysical, in vitro and
genome-wide study demonstrated that bis-ETPs are potent regulators of HIF-1α inducible
transcription.
Figure 4.28 Transcriptional inhibition of HIF-1α inducible genes with dimeric ETP
through disruption of the binding of HIF-1α and p300/CBP .
In hypoxic breast carcinoma cell lines MCF7 and MDA-MB-231 the designed dimeric
ETP-5 shows significant downregulation of HIF-1α inducible transcription of VEGF and
c-Met genes and their protein products. In lung adenocarcinoma cell line A549 five key
genes VEGF, c-Met, Glut1, LOX and CXCR4 have been significantly downregulated with
ETP-5 in a dose-dependent manner. Our gene expression profiling experiments provided
important insights into the global genomic effects of ETP-5 under hypoxia conditions.
The number and type of genes affected by ETP-5 is consistent with our previous results
suggesting that this compound is highly specific transcriptional inhibitor with well-
defined pharmacogenomic profile. Further structural, genomic and in vivo studies of this
novel class of designed transcription-based inhibitors are underway.
156
Chapter 5: Integrin Ligand – Boron Cluster Dendrimer Conjugates for Imaging and
Targeted Delivery of Boron to Tumors
157
5.1 Introduction
Integrins are transmembrane receptors involved in cell adhesion and interaction between
cells and extracellular matrix. Structurally, integrins are heterodimers having an α and a β
subunits that span the cell membrane. There are 18α and 8β subunits known in mammals.
Endothelial cells express many different types of integrins and out of them α
v
β
3
appear to
be the most important integrin for angiogenesis and therefore by implication are a
primary target in cancer therapy to control tumor growth (Cai and Chen 2006). Among
endothelial cells, α
v
β
3
integrins are overexpressed in activated cells or cells that belong to
newly formed vasculature. They are not over-expressed and in fact, are very rare on
resting endothelial cells. This unique distribution of α
v
β
3
integrins make them an
attractive target to selectively cut of supply lines to tumors with minimal side effects.
The α
v
β
3
integrin, also known as vitronectin receptor, is a heterodimer of α
v
(125 kDa)
and β
3
(103 kDa) subunits. α
v
β
3
integrin binds a range of ligands in the extracellular
matrix that have RGD (Arg-Gly-Asp) triple peptide motif. Molecules like vitronectin,
fibronectin, laminin and proteolyzed collagen have RGD motif and bind α
v
β
3
integrin
with sub-micromolar affinity. Integrin α
5
β
1
, which is structurally very close to α
v
β
3
integrin, binds with high affinity only to fibronectin.
Among integrins, α
v
β
3
receptor is the most important one for angiogenesis. It also plays a
role in metastasis and osteoclast-mediated bone resorption. Apart from vitronectin, α
v
β
3
integrin is also known to bind VEGF receptors, MMP-2, PDGF and insulin receptors
(Kumar 2003). These interactions lead to cell proliferation, invasion and reduction in
apoptosis. Interaction between α
v
β
3
integrins and VEGF or PDGF receptors is not
158
necessarily physical. Recent reports have shown that they can influence each other
function by augmenting or interfering in each other’s signaling pathways (Somanath,
Malinin et al. 2009; Ishigaki, Imanaka-Yoshida et al. 2011).
Figure 5.1 Diagram of α
v
β
3
integrin pathway activation. Factors in extracellular matrix,
like vitronectin and other receptors such as VEGF, PDGF and insulin receptors interact
with α
v
β
3
integrins and eventually lead to cell proliferation, increase in invasiveness and
downregulation of apoptosis.
Inhibitors for α
v
β
3
integrin include RGD peptide antagonists, peptidomimetics and
monoclonal antibodies. They all have been shown to inhibit angiogenesis and induce cell
apoptosis in endothelial cells. This suggests that inhibitors of α
v
β
3
integrins have a
potential for treatment of tumors in cancers, since the α
v
β
3
integrins are overexpressed in
tumor neovasculature. In essence, antagonists and ligands of α
v
β
3
integrins can be used to
selectively target tumors and tumor vasculature without affecting other tissues.
159
5.2 Targeted delivery of antitumor cargo to α
v
β
3
integrin receptors. In order to target
α
v
β
3
integrins overexpressed on tumor cells and on vascular endothelial cells in tumors, a
ligand is required that binds to α
v
β
3
integrin receptor with high affinity in the presence of
other structurally related integrins. In our work we focused on small molecule α
v
β
3
specific ligands as opposed to proteins and monoclonal antibodies. Since our ultimate
aim was to make a delivery system in vivo we focused on peptidomimetics and non-
peptidic small molecules. In our first-generation design, a α
v
β
3
integrin selective non-
peptidic small molecule reported by Duggan and coworkers (Duggan, Fisher et al. 1996)
was used as a targeting ligand. This compound (Figure 5.2) is 50 times more specific
towards α
v
β
3
as compared to its homologically closest integrin α
v
β
5
integrin (Duggan,
Fisher et al. 1996; Duggan, Duong et al. 2000).
Figure 5.2 Structure of α
v
β
3
specific ligand and it role as a targeting ligand in the
delivery of cargo to tumor cells.
Therefore, we decided to utilize this α
v
β
3
-specific ligand for targeted delivery of cargo to
cells that overexpress α
v
β
3
integrins. A suitable imaging or therapeutic cargo could be
conjugated to the ligand via its aliphatic linker directly attached to sulfonamido group.
Among the numerous choices of cargo boron clusters (carboranes) appeared as especially
attractive choice because the same cluster could be utilized as a contrast agent in boron
160
MRI and as a therapeutic modality in boron neutron capture therapy (BNCT). In addition,
due to orthogonal nature of the chemical transformation involved, we could incorporate
fluorescent dyes to provide additional means of optical imaging in order to monitor
uptake and localization of the conjugate within tumor cells and monitor the degree of
tumor tissue penetration with confocal microscopy and intravital microscopy imaging.
5.3 Applications of boron-rich compounds in boron MRI and BNCT. Boron is present
in nature as two isotopes,
11
B with 80% natural abundance and
10
B with 20% natural
abundance.
11
B nucleus has spin of 3/2 and
10
B nucleus has spin of 3. Gyromagnetic
ratios for
11
B and
10
B compared to
1
H are 0.32 for
11
B and 0.107 for
10
B. For NMR
spectroscopy or MRI, both choices of
11
B and
10
B are available (Hermanek 1992; Bendel
2005). For boron neutron capture therapy (BNCT), generally
10
B enriched reagents are
used and in this case it is preferable to use clusters suitable for
10
B MRI. As such,
10
B
nucleus has T
2
relaxation time that is much longer than that of
11
B, thereby giving it some
advantage in detection. On the other hand, the natural abundance of
11
B isotope is higher
(80%), therefore NMR/MRI using this isotope may be more suitable economically.
Bradshaw and coworkers reported the use of
11
B MRI using BSH (B
12
H
12
S), a polyhedral
boron compound, in human patients (Bradshaw, Schweizer et al. 1995).
10
B is a special
nucleus since it has integral spin of 3. Its T
2
relaxation is longer and
10
B NMR/MRI and
can be used for
10
B-
1
H double resonance experiments. BSH and its dimer BSSB have
been extensively used in experiments and in clinical trials. These molecules lack carbon
atoms, therefore their conjugation with organic ligand is a challenge. A new motif to
circumvent this issue was needed. Boron clusters, such as carboranes, are polyhedral
molecules containing ten boron and two carbon atoms, can be chemically derivatized
161
while still having high boron content. Therefore, they became the compounds of choice for
design of our conjugates and subsequent in vitro and in vivo experiments.
Figure 5.3 Structures of BSH (borocaptate sodium), BSSB (BSH dimer) and BPA
(boronophenylalanine), the most common reagents for BNCT and boron MRI.
Boron neutron capture therapy is a strategy to target malignant cells with agents that are
capable of becoming therapeutics after capturing of epithermal neutrons by boron nuclei .
The physics behind this phenomenon is that the
10
B nucleus upon neutron capture
converts into the high energy intermediate [
11
B] (Figure 5.4) followed by the fission of
[
11
B], which results in the release of an α particle with high linear velocity from the
process termed “linear energy transfer”; the process also produces “recoiling”
7
Li nuclei.
Since the high linear energy transfer (LET) particles have limited path length, hence their
destructive effect is limited to the area that is approximately equal to the diameter of a
cell. . The amount of
10
B atoms that needs to be present in cells/tissue for successful
162
BNCT is critical parameter and is estimated to about 10
9
atoms per cell or 20 µg/g of
tissue or cells.
Figure 5.4 Fission nuclear reaction of
10
B. Neutron capture by
10
B forms metastable [
11
B]
which undergoes disintegration giving an alpha (α) particle, lithium and energy. The
principle of BNCT is based on the capture of a neutron by
10
B in the tissue and energy
release via the shown mechanism, damaging the surrounding tissue.
BPA (boronophenylalanine) is the first generation of boron delivery agent. BSH and its
derivative that have a polyhedral borane moiety are considered to be second generation
boron delivery reagents. The third generation BNCT compounds have a hydrolytically
stable linker that bind the borane polyhedral moiety to a targeting ligand. Figure 5.5
shows examples of two third generation boron delivery agents (Barth, Coderre et al.
2005).
Boron delivery reagents should have three characteristics: a) they should have low
systemic toxicity, b) they should have low normal tissue uptake but high tumor tissue
uptake with 20 µg
10
B atoms per gram of tumor tissue considered to be acceptable limit
of an uptake amount for success in BNCT and/or boron MRI, and c) they should have
rapid clearance in normal tissue but not in tumor tissue.
163
Figure 5.5 Examples of third generation boron delivery agents. a) β-5-o-carboranyl-2′-
deoxyuridine is a precursor that can readily get phosphorylated inside a tumor cell into
corresponding nucleotide. Due to higher rate of phosphorylation inside a tumor cell the
corresponding nucleotides are trapped and retained in a higher amount inside a tumor
cell. b) Trimethoxyindole derivative of a boron compound is a DNA binder and is shown
to undergo good uptake in rapidly dividing cancer cells.
Two examples of third generation boron delivery reagents are shown in Figure 5.5. A
common chemical feature of these compounds is the presence of carborane moiety.
Carborane is a structure having boron atoms clustered in a form of polyhedron with two
additional carbon atoms that complete the closed icosahedral shape. The carbons are
often used to connect the polyhedron to a short aliphatic linker. The linker gives an
opportunity to conjugate the carborane to a targeting moiety. The example a and b in
Figure 5.5 shows an example of nucleotide precursor and a DNA binder respectively,
attached to a carborane (Tietze, Griesbach et al. 2002; Al-Madhoun, Johnsamuel et al.
2004; Barth, Yang et al. 2004). The nucleotide precursor has low systemic toxicity while
DNA binding indole derivative has slightly higher toxicity. These two compounds belong
to a class of low molecular weight boron delivery agents. Apart from these, another
164
important class is high molecular weight boron delivery agents. Monoclonal antibodies
and their fragments can be used as boron delivery agents, as boron compounds can be
readily conjugated to these. A major issue that hampers the use of these large delivery
agents in vivo is their rapid clearance by reticuloendothelial system (macrophage system)
(Faillot, Magdelenat et al. 1996). Thus, the most successful strategy to deliver boron
agents to tumors is through the direct injection into the tumor site itself. This is especially
important in the case of brain tumors where the boron compounds are not able to cross
the blood-brain barrier. In cases like this, intracranial or intratumoral injections are given
(Barth, Yang et al. 2004). Tumor cells/tissue undergoes rapid proliferation and
neovascularization therefore they have overexpression of growth receptors like VEGFR
and EGFR. Conjugation of boron bearing compounds to growth factors, like EGF and
VEGF, provides targeted delivery to tumor cells (Yang, Barth et al. 2002).
Apart from the strategies discussed above, another approach is the use of monoclonal
antibodies that target caveolae. Many high molecular weight agents enter tissue and cells
of tumor via caveolar transport, therefore conjugation of boron compounds to caveolar
antibody can, in principle, deliver the boron payload to tumors (Schnitzer, Liu et al.
1995).
Nuclear reactors and accelerators are two types neutron beam sources for boron neutron
capture therapy. Nuclear reactors with special modifications, such as Li barriers, are the
most common neutron sources for BNCT (Riley, Binns et al. 2003). Neutrons are
classified according to their energies with the following classification: thermal neutrons
with E
n
< 0.5 eV, epithermal neutrons with 0.5 eV < E
n
< 10 keV and fast neutrons with
E
n
> 10 keV. Thermal neutrons are best type of neutrons for
10
B capture, however since
165
they are low in energy, their penetration depth is low. Thus, epithermal neutrons are often
used as they can penetrate deeper into the tissue and in doing so loose energy and convert
into thermal neutrons (Harling and Riley 2003).
5.4 Building blocks for targeted boron delivery. As mentioned above, we decided to use
the α
v
β
3
integrin ligand reported by Duggan et al. in our design of integrin ligand
conjugates, because it shows very high specificity towards α
v
β
3
integrin in the presence
of the closely related integrins α
v
β
5
and α
IIB
β
3
. In fact, it has been shown to have 50-fold
stronger affinity toward α
v
β
3
receptors as compared to α
v
β
5
(Duggan, Fisher et al. 1996;
Duggan, Duong et al. 2000).
In order to first evaluate uptake and distribution of the targeted boron cluster conjugate,
fluorescent dye was required. An unusual choice of cyan 39/40 dye was made
(Ohulchanskyy, Pudavar et al. 2003) due to its ability to show enhanced fluorescence
upon binding the target and favorable photophysical properties. As seen in Figure 5.6 the
precursor, shown in green, reacts with primary amines, producing conjugated cyan 40
dye. The main property of this dye is that intensity of its fluorescence is dependent on the
type of its environment, such as the viscosity or local dielectric constant of the medium
which can be altered by binding of the dye to its target. The dye has a network of
conjugated double bonds and therefore the planarity of the molecule is a pre-requisite for
its fluorescence efficiency. When the intramolecular rotation of the dye molecule around
one of its single bonds in free then the fluorescence is low. On the other hand, if due to
higher viscosity or other factors this intramolecular rotation is hindered, its fluorescence
is enhanced. Thus, the use of this dye has an added advantage that it will minimize the
166
background fluorescence of the unbound conjugate and so the imaging of the cells that
uptake conjugates of this dye will be more convenient due to better contrast.
As discussed above, we chose carborane motif for intratumoral delivery of boron. The
benefit of carborane is in the facile conjugation of the polyhedron to a ligand via covalent
linkage. In order to achieve the requisite number of boron atoms per molecule, three
carboranes (B
10
C
2
H
12
) were linked into a dendritic wedge. Two such dendritic wedges
could be conjugated to an integrin ligand.
Figure 5.6 Structures of building blocks of targetable boron delivery system. α
v
β
3
integrin ligand reported by Duggan et al. is shown in blue. Cyan 39 (X = O) and Cyan 40
(X = NH) shown in green. Dendritic wedge shown with three carborane motif joined by a
tetradentate linker. Carborane motif, B
10
C
2
H
12
, is a polyhedron having 10 boron atoms
and two carbon atoms. The linkage to the tetradentate linker is via one of the carbon
atoms in the polyhedron. Author thanks Swati Kushal, Dr. Nathan Polaske and Dr.
Vladimir Neschadimenko for synthesis of the dye.
167
5.5 Structure and photophysical properties of α
v
β
3
integrin ligand conjugates.
5.5.1 DIL-1: α
v
β
3
integrin ligand – cyan dye conjugate. This was the first molecule
designed to study the photophysical properties and biological activity of α
v
β
3
integrin
ligand conjugated to a fluorescent dye (Figure 5.7). The synthesis, photophysical
properties and its behavior in vitro assays was investigated by the author in collaboration
with Swati Kushal and detailed in the attached manuscript (see Appendix D).
Figure 5.7 Structure DIL-1 having α
v
β
3
integrin ligand conjugated to cyan 40.
In 90% glycerol, fluorescence spectra for DIL-1 clearly showed ~40 fold increase in
fluorescence emission as compared to the fluorescence in aqueous medium (Figure 5.9).
168
Figure 5.8 Fluorescence emission spectra for DIL-1 at 435 nm. Emission spectra for
DIL-1 compound at 100 µM concentration in aqueous (dashed line) and viscous (bold
line) media is shown. In viscous medium the emission spectra shows a maximum at 475
nm while in aqueous medium there are two peaks for maxima at 475 nm and 515 nm. The
experiment clearly shows that the fluorescence emission is much higher in viscous
medium as compared to aqueous medium due to hindrance of intramolecular rotation.
169
5.5.2 DILB-2: α
v
β
3
integrin ligand – cyan dye – carborane dendritic wedge conjugate.
A trifunctional conjugate with integrin ligand, cyan 40 dye and a carborane dendritic
wedge carrying 60 boron atoms was designed and synthesized.
Figure 5.9 Structure of trifunctional conjugate DILB-2. The conjugate consists of two
dendritic wedges having six carborane units and carrying 60 boron atoms in total. The
α
v
β
3
integrin ligand and cyan 40 dye are attached via an aliphatic linker.
To study the photophysical properties of DILB-2, fluorescence emission spectra were
recorded in viscous and aqueous media at 435 nm. In viscous media (90% glycerol, 5%
water and 5% methanol) DILB-2 at 100 µM concentration showed 8-fold higher
fluorescence emission intensity as compared to aqueous media (95% water and 5%
methanol). This is important due to the fact that this molecule is substantially more
sterically crowded as compared to DIL-1 and hence the cyan 40 dye can be expected to
170
be more hindered in its intramolecular rotation. Despite this, DILB-2 still showed
enhanced fluorescence emission in viscous media.
Figure 5.10 Fluorescence emission spectra of DILB-2 at 435 nm in aqueous (dashed line)
and viscous (bold line) media. DILB-2 compound was at 100 µM concentration. The two
maxima were at 475 nm and 515 nm.
Figure 5.11 Excitation spectra for DILB-2 in viscous and aqueous media measured at
540 nm. Excitation spectra were measured in viscous medium (bold line) and aqueous
medium (dashed line) in a range between 375 nm and 520 nm. The emission was
measured at 540 nm.
Excitation spectra for DILB-2 were measured in aqueous and viscous media at 540 nm.
Similar to DILB-1, the excitation spectra showed substantially higher intensity in viscous
medium (bold line) as compared to aqueous medium (dashed line).
171
5.5.3 DB-3: Cyan 40 dye – carborane dendritic wedge conjugate. A control compound
was designed and synthesized which contains cyan 40 dye attached to the cluster of
carborane dendritic wedge but lacked the α
v
β
3
integrin ligand. This compound was
designed to rule out the non-specific binding and uptake due to the effect of carborane
dendritic wedges and cyan 40. The main reason to suspect non-specific uptake by cells is
due to the relative hydrophobic carborane motifs. In both DILB-2 and DB-3 the
carborane clusters form dendritic wedges. To rule out impact of spacing between wedges,
the linker structure in DB-3 was designed to be comparable in size and molecular weight
to DILB-2.
Figure 5.12 Structure of DB-3 control, a cyan dye – carborane dendritic wedge
conjugate. DB-3 lacks α
v
β
3
integrin ligand and is a conjugate of cyan 40 dye linked to the
cluster of two carborane dendritic wedges of the same structure as in DILB-2.
172
To study photophysical property of DB-3, emission spectra were measured in viscous
medium and aqueous medium at 435 nm at 100 µM concentration. Emission spectra in
both the media showed that in viscous media the intensity of emitted light is substantially
higher than that in aqueous media. The maximum for DB-3 emission spectrum in viscous
media was around 485 nm and in aqueous medium was around 520 nm.
Figure 5.13 Emission spectra for DB-3 in viscous and aqueous media at 435 nm. DB-3 at
concentration was 100 µM the excitation maximum was 435 nm. The emission spectra
were measured in viscous (bold line) and aqueous media (dashed line).
An interesting observation is that the fold increase in intensity of emitted light for DILB-
2 in viscous medium as compared to aqueous medium is higher as compared to the
increase seen for DB-3.
There are two possible reasons for this phenomenon. , First, DB-3 is a much more
hydrophobic molecule as compared to DILB-2 which has a hydrophilic integrin ligand.
Due to the hydrophobic nature of DB-3, its intramolecular movement will be hindered in
aqueous and glycerol media to a similar extent. Second reason could be that due to the
173
larger distance between the cyan 40 dye and its neighboring groups like integrin ligand
and carborane clusters in DILB-2 the cyan 40 dye has freedom to undergo intramolecular
rotations as compared to dye DB-3.
Excitation spectra for DB-3 were measured at a concentration of 100 µM for emitted
light at a wavelength of 540 nm. At this wavelength, viscous medium showed little
increase in intensity as compared to aqueous medium. The plausible reasons for the
observed decrease in enhancement of emitted light at 540 nm upon excitation from 375
nm to 520 nm can be the same as mentioned above in the explanation for decreased
enhancement in the emission spectra. In both aqueous and viscous media the maxima for
excitation spectra was around 445 nm.
Figure 5.14 Excitation spectra for DB-3 in aqueous and viscous media measured at 540
nm. The excitation spectra for DB-3 at 100 µM show maxima for aqueous and viscous
media around 445 nm. Interestingly, the excitation spectra measured at 540 nm showed
very little increase in intensity in viscous medium (bold line) as compared to aqueous
medium (dashed line).
174
All the three designed conjugates having cyan 40 dye showed good photophysical
properties with enhancement in the intensity of the emitted light upon increase in
viscosity of the medium. The next sets of experiments were conducted to determine the
ability of these molecules to bind with cells that express α
v
β
3
integrins both in vitro and
in vivo using optical imaging with cyan 40 dye.
5.6 Confocal microscopy imaging in carcinoma cell lines expressing α
v
β
3
integrin
receptors.
5.6.1 Confocal imaging with DIL-1 in WM115 cells. Initial imaging experiments were
carried out with DIL-1 molecule in WM115 cells, which is highly metastatic melanoma
cell line that has been shown to express high levels of >100,000 of α
v
β
3
integrin.
Confocal microscope imaging with DIL-1 was done at different concentrations ranging
from 100 nM to 100 µM. Cellular uptake of DIL-1 was readily noticeable at 5 µM
concentration with 100 µM concentration giving good contrast.
The images shown in Figure 5.18a is taken at 100 µM DIL-1 in WM115 cells. The
cellular morphology of the cells treated with DIL-1 showed change to spherical shape
from the typical elongated form, presumably due to the disruption of their adhesion to the
dish. The images of cells fixed with 25% ethanol are shown in Figure 5.18b. The
experiment clearly distinguishes uptake levels of DIL-1 cell in live and fixed cells. In
MCF7 breast carcinoma cell line that have much lower expression levels of α
v
β
3
integrins, DIL-1 also shows substantial uptake at 100 µM concentration (Figure 5.19).
175
Figure 5.15 Confocal microscopy images of WM115 cells treated with 100 µM of DIL-
1. a) Live and b) ethanol-fixed WM115 cells treated with DIL-1 in DMEM media with
10% FBS. Both images were taken after 15 min of incubation. The top left quadrant
shows cyan 40 dye emission (green), top right quadrant shows bright field image of the
cells and bottom left quadrant shows overlay of cyan 40 dye emission with bright field
image.
Figure 5.16 Images of MCF7 cells treated with DIL-1 at 100 µM concentration. MCF7
cells treated with DIL-1 show significant uptake into cells along with change of cell
morphology. The top left quadrant shows cyan 40 dye emission (green), top right
quadrant shows bright field image of the cells and bottom left quadrant shows overlay of
cyan 40 dye emission with bright field image.
176
Based on these data, DIL-1 undergoes internalization into cells very rapidly, presumably
by following multiple endocytotic pathways. Even in cell lines such as MCF7 with the
low count of α
v
β
3
this enhanced endocytosis results in an efficient uptake of DIL-1.
Figure 5.17 Imaging of WM115 and MCF7 cells at 4 °C after treatment with DIL-1. a)
WM115 and b) MCF7 cells were treated with 5 µM DIL-1 at 4 °C for 2 h in cell culture
media supplemented with 10% FBS. The top left quadrant shows cyan 40 dye emission
(green), top right quadrant shows bright field image of the cells and bottom left quadrant
shows overlay of cyan 40 dye emission with bright field image.
Next, to study the intake of DIL-1 into the cells 4 °C experiments were carried out. In
these experiments WM115 and MCF7 cells were treated with 5 µM DIL-1 for 2 h at 4 °C
and confocal images were taken. No noticeable uptake of DIL-1 was observed in both
cell lines. This shows that the uptake of DIL-1 happens via active process whose rate
greatly diminishes at 4 °C.
5.6.2 Confocal microscopy imaging with DILB-2 in WM115 and MCF7 cells. To assess
cellular uptake and localization of DILB-2, WM115 cells in DMEM media with 10%
177
FBS were treated with the compound at different concentrations. DILB-2 provided good
contrast when imaged at 25 µM concentrations (Figure 5.21a and b).
Figure 5.18 WM115 cells show significantly higher uptake of DILB-2 as compared to
MCF7 cells. a) Confocal image of WM115 cells in DMEM media having 10% FBS
treated with DILB-2 at a final concentration of 25 µM . b) Confocal image of MCF7 cells
treated with 25 µM DILB-2 in RPMI media having 10% FBS. The top left quadrant
shows cyan 40 dye emission (green), top right quadrant shows bright field image of the
cells and bottom left quadrant shows overlay of cyan 40 dye emission with bright field
image.
The images show significantly higher uptake of DILB-2 by WM115 cells as compared to
MCF7 cells. Under these conditions DILB-2 showed higher binding and uptake by
WM115 cells that have > 100,000 α
v
β
3
integrins/cell as compared to MCF7 cells that
display ~10,000 α
v
β
3
receptors per cell.
178
Figure 5.19 Temperature dependence of uptake of DILB-2 in WM115 and MCF7 cells.
a) WM115 cells incubated with 25 µM DILB-2 in DMEM medium having 10% FBS for
2 h at 37 °C. b) WM115 cells incubated with 25 µM DILB-2 in DMEM medium having
10% FBS for 2 h at 4 °C. c) MCF7 cells incubated with 25 µM DILB-2 in RPMI medium
having 10% FBS for 2 h at 37 °C. d) MCF7 cells incubated with 25 µM DILB-2 in
DMEM medium having 10% FBS for 2 h at 4 °C. The top left quadrant shows cyan 40
dye emission (green), top right quadrant shows bright field image of the cells and bottom
left quadrant shows overlay of cyan 40 dye emission with bright field image.
Next, to establish the active nature of the uptake of DILB-2 into the cells, WM115 and
MCF7 cells were treated with 25 µM DILB-2 for 2 h at before a series of confocal
images were taken and compared to those taken at 37 °C. The images clearly show that
at the uptake of DILB-2 is nearly absent, suggesting uptake via endocytotic pathways that
require substantial amount of energy and, therefore absent at 4 °C.
179
5.6.3 Confocal imaging with DB-3 in WM115 cells. Control compound DB-3 that is
lacking α
v
β
3
integrin ligand was imaged in WM115 cells after 2 h of treatment with DB-3
at (Figure 5.23a) and at 37 °C (Figure 5.23b). In both cases no significant uptake was
detected, suggesting that control compound is not internalized by the cells.
Figure 5.20 Control DB-3 shows no noticeable uptake in WM115 cells. a) WM115 cells
at 4 °C and b) WM115 cells at 37 °C were treated with 20 µM DB-3 for 2 h and 14 h,
respectively in cell culture media supplemented with 10% FBS. The top left quadrant
shows cyan 40 dye emission (green), top right quadrant shows bright field image of the
cells and bottom left quadrant shows overlay of cyan 40 dye emission with bright field
image.
5.7 Flow cytometry in WM115 and MCF7 cells. In order to statistically quantify the
uptake of DIL-1 and DILB-2 into α
v
β
3
integrin expressing cells we conducted flow
cytometry experiments. These experiments were performed in collaboration and with
substantial input from Swati Kushal and presented in the joint manuscript (see Appendix
180
D). The results confirmed our confocal imaging data, suggesting enhanced uptake of
DIL-1 and DILB-2, but not DB-3 into cells.
5.8 Intravital microscopy imaging in murine subcutaneous and in ectopic-orthotopic
tumor xenograft models with DILB-1 and DB-3. To study the in vivo applicability of
designed α
v
β
3
integrin ligand – carborane conjugates for imaging and retention in tumor
tissue, mouse tumor xenograft models were used and intravital microscopy was
performed (Vajkoczy, Ullrich et al. 2000; Koehl, Gaumann et al. 2009; Hak, Reitan et al.
2010; Lohela and Werb 2010). The experiments were carried out at the imaging facility ,
at Proteogenomics Institute for Systems Medicine (PRISM) in San Diego, by our
collaborators Prof. Jan Schnitzer and Dr. Philip Oh who used both subcutaneous and
more advanced ectopic orthotopic tumor models. In the subcutaneous (SQ) tumor model,
a window chamber is installed on the dorsal side of nude mice and the cells are injected
directly into the tissue there they form a tumor. In an ectopic-orthotopic (EO) tumor
model, a window is installed on the dorsal side of mice where ectopically an extenuous
tissue or tumor cells that form a spheroid are implanted into the mammary fat pad. .
181
Figure 5.21 Intravital microscopy (IVM) images showing facile uptake of the DILB-2,
but not control DB-3 conjugate in subcutaneous tumor models. Top row shows images
from subcutaneous tumor model after injection of DILB-2 intravenously on Day 1. The
substantial amount of compound DILB-2 is retained in the tumor for atleast 3 days and is
visible in significant amount on Day 5 and Day 7. Second row from the top presents the
ectopic orthotopic tumor model injected with DILB-2. In this tumor model the
vasculature is normalized and is non-leaky. Therefore the entry and/or retention of DILB-
2 was not observed. The bottom two rows are for subcutaneous and ectopic-orthotopic
tumor models respectively where DB-3 was injected intravenously. In both the cases no
entry or retention of DB-3 was observed.
Apart from ectopic orthotopic model a second model of subcutaneous tumor was also
used. In the subcutaneous (SQ) tumors with leaky vasculature, such as melanomas ,
(Ruoslahti 2002; Hendrix, Seftor et al. 2003) DILB-2 is readily accumulated and remains
clearly detectable for up to 7 days. Figure 5.26 shows that the control compound DB-3
which lacks α
v
β
3
integrin is unable to enter either the EO or SQ tumor, while DILB-2 is
able to enter SQ tumor and is retained for several days. However DILB-2 is unable to
enter EO tumor. This result shows that the macromolecule DILB-2 needs leaky tumor
182
vasculature for entry into tumor. At the same time it also shows that even in leaky tumor
for entry α
v
β
3
integrin ligand plays important role in uptake and retention of the
compound as the control compound DB-3 is unable to enter the SQ tumor (Figure 5.26).
5.9 Conclusion. We designed, synthesized and performed in vitro and in vivo testing of
the new class of tumor-targeting macromolecules based on integrin ligand - boron cluster
dendrimer conjugates for applications in cancer imaging and therapy. Our designed
conjugates DIL-1, DILB-2 and DB-3 show the desired photophysical properties as all of
them show enhancement in fluorescence intensity upon binding to cells expressing α
v
β
3
integrins. Confocal microscopy reveals uptake of DIL-1 and DILB-2 by cells that express
α
v
β
3
integrins while no noticeable uptake was seen by the control compound DB-3. In
vivo, SQ mouse tumor xenograft models show enhanced uptake and retention of DILB-2
as opposed to of DB-3.
Future direction of this research should include continuing development of conjugates
based on DILB-2 for application in boron MRI and BNCT. At the present time, the
uptake of such conjugates is mainly limited to SQ tumors with leaky vasculature,
however studies the ongoing collaborative work of our group and PRISM is focused on
overcoming the vascular endothelial barrier found in EO models by designing the
conjugates that take advantage of the active transport via caveoli. In such tumor models
it should be determined whether the molecule DILB-2 or its analog is able to deliver the
critical amount of boron atoms to the tumor tissue for its successful application in boron
MRI and BNCT .
183
Chapter 6: Experimental Sections
184
6.1 Introduction to Experimental Section. This section contains descriptions of synthetic
methodologies, analytical data and experimental procedures for each chapter.
6.2 Experimental Section for Chapter 2
6.2.1 General Methods
All reagents and solvents were obtained from commercial sources and were used as
received unless otherwise stated. All reactions involving moisture-sensitive reagents
were conducted under a dry N
2
atmosphere with anhydrous solvent and flame dried
glassware. Gravity chromatography was performed on silica gel (230-400 mesh) using
reagent grade solvents. Nuclear Magnetic Resonance (NMR) spectra were collected on
Varian Unity 300 MHz, or Bruker 250 MHz, 500 MHz or 600 MHz instruments in the
indicated solvents. Mass spectra were obtained from the Mass Spectrometry Laboratory
in the Department of Chemistry at the University of Arizona. Infrared spectra (IR) were
collected on polyethylene spotted with the compound in the indicated solvent and
recorded in cm
-1
. Melting points were measured on Mel-Temp capillary melting point
apparatus and all were uncorrected.
185
6.2.2 Synthetic procedures and analytical data for diketopiperazines.
3-(Ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (11)
A suspension of sarcosine anhydride (2.00 g, 14.1 mmol) in dry THF (40 mL) under a N
2
atmosphere was cooled to –78
o
C. KOt-Bu (1.0 M solution in THF) (35.2 mL, 35.2
mmol) was added dropwise with stirring. Once the addition was complete, the mixture
was allowed to stir for 5 min. Then, Et
2
CO
3
(5.00 g, 5.13 mL, 42.3 mmol) was added
dropwise. The temperature was maintained for 15 min, and then the reaction was
warmed to room temperature and stirred for an additional 3 h. A saturated NH
4
Cl
solution (40 mL) was added and stirring was continued for 10 min. Most of the THF was
removed under reduced pressure, and the remaining residue was extracted with CH
2
Cl
2
(3
x 50 mL). The organic extracts were combined, dried over MgSO
4
, filtered, and
concentrated. The crude oil was purified by column chromatography (silica-gel,
CH
2
Cl
2
/acetone (4:1)) to afford 11 (2.33 g, 77% yield) as a white solid. mp. 69-71
o
C.
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1738, 1665, 1410, 1264, 1032.
1
H NMR (500 MHz, CDCl
3
) δ
4.55 (1H, s), 4.33-4.25 (2H, m), 4.18 (1H, d, J=17 Hz), 3.85 (1H, d, J=17 Hz), 2.99 (3H,
s), 2.96 (3H, s), 1.33 (3H, dt, J=6 Hz and J=2 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 166.4,
164.5, 160.2, 66.13, 63.08, 51.78, 33.90, 32.91, 14.11. R
f
at 20% acetone in CH
2
Cl
2
: 0.4.
HRMS (EI) calculated for C
9
H
14
N
2
O
4
M
+
: 214.0954, found: 214.0959.
186
Ethyl-2-(dibenzylamino)-3-[(2-benzyloxycarbonyl)methylamino]-3-oxopropanoate
(13)
To a solution of diethyl (dibenzylamino)malonate (15.0 g, 42.2 mmol) in 50 mL toluene
were added N,N diisopropylethylamine(5.5 mL) and benzyl sarcosinate p-
toluenesulfonate (5.0 g, 14.2 mmol). The reaction was stirred under reflux for 8 h.
Another batch of benzyl sarcosinate p-toluenesulfonate (5.0 g, 14.2 mmol) and N,N-
diisopropylethylamine (5.5 mL) was added to the reaction mixture and refluxed for 8 h.
The final batch of benzyl sarcosinate p-toluenesulfonate (5.0 g, 14.2 mmol) and N,N-
diisopropylethylamine (5.5 ml) was added along with 20 mL of toluene and the reaction
mixture was refluxed for additional 8 h. The reaction mixture was cooled to room
temperature and the solvent and excess N,N-diisopropylethylamine were removed under
reduced pressure. The crude product was purified by column chromatography (silica gel,
hexane/EtOAc (8:2)) to give 13 (17.7 g, 86%) as a colorless viscous oil. IR (acetone)
ν
max
(cm
-1
) 3029, 1745, 1656, 1401, 1194, 1136, 1027, 969, 746, 705.
1
H NMR
(250MHz, CDCl
3
) δ 7.40-7.17 (16H, m), 5.1-3.61 (10H, m), 2.89 (keto); 2.85 (enol) (3H,
two s), 1.32 (keto); 1.31 (enol) (3H, two t, J=7 Hz and J=7 Hz).
13
C NMR (75 MHz,
CDCl
3
) δ 168.7, 168.63, 168.4, 168.4, 168.4, 168.3, 138.8, 138.3, 135.2, 129.20, 129.0,
128.5, 128.4, 128.3, 128.2, 128.1, 127.3, 127.2, 66.86, 66.7, 64.57, 63.97, 60.74, 60.66,
187
55.4, 55.15, 50.77, 49.61, 36.38, 34.84, 14.22, 14.17. R
f
at 20% EtOAc in hexanes: 0.3.
HRMS (FAB) calculated for C
29
H
33
N
2
O
5
[M+H]
+
: 489.2390, found: 489.2404.
3-(Ethoxycarbonyl)-1-methylpiperazine-2,5-dione (14)
Compound 6 (12.0 g, 24.6 mmol) was dissolved in 60 mL of ethanol and 10% palladium
on carbon (2.5 g) was added to the solution. The walls of the flask were washed with
additional 10 mL of ethanol and the reaction mixture was stirred under hydrogen
atmosphere at room temperature for 72 h. The reaction mixture was filtered and the
filtrate was concentrated under reduced pressure. The crude product was purified by
column chromatography (silica gel, CH
2
Cl
2
/acetone (1:1)) to yield 14 (3.84 g, 78%) as a
white solid. mp. 93-95
o
C. IR (acetone) ν
max
(cm
-1
) 1741, 1677, 1259, 1199, 1025.
1
H
NMR (600 MHz, CDCl
3
) δ 7.49 (1H, s), 4.68 (1H, d, J=2 Hz), 4.30-4.17 (2H, m), 4.19
(1H, d, J=17 Hz), 3.84 (1H, d, J=17 Hz), 3.00 (3H, s), 1.32 (3H, dt, J=7 Hz and J=1 Hz).
13
C NMR (150 MHz, CDCl
3
) δ 167.3, 166.8, 160.4, 62.98, 59.23, 51.57, 34.20, 13.95. R
f
at 50% acetone in CH
2
Cl
2
: 0.45. HRMS (FAB) calculated for C
8
H
13
N
2
O
4
[M+H]
+
:
201.0876, found: 201.0883.
188
3-(Ethoxycarbonyl)-1-methyl-4-acetylpiperazine-2,5-dione (15)
To a solution of 1b (210 mg, 1.0 mmol) in CH
2
Cl
2
(2 mL) was added 4-
(dimethylamino)pyridine (183 mg, 1.5 mmol). Acetyl chloride (110 µL, 1.5 mmol) was
added dropwise to the reaction mixture in an ice bath over the period of 15 minutes. The
reaction mixture was stirred at room temperature for 6 h. The reaction mixture was
concentrated and the product isolated by column chromatography (silica gel,
CH
2
Cl
2
/acetone (95:5)) to yield 15 (208 mg, 86% yield) as a white solid. mp. 125-127
o
C. IR (acetone) ν
max
(cm
-1
) 1727, 1670, 1378, 1317, 1244, 1191, 1016.
1
H NMR (500
MHz, CDCl
3
) δ 5.52 (1H, s), 4.29 (1H, d, J=18 Hz), 4.30-4.17 (2H, m) 2.97 (3H, s), 2.54
(3H, s), 1.28 (3H, t, J=7 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 171.2, 165.9, 165.7, 159.9,
63.18, 59.77, 53.11, 33.72, 26.57, 13.90. R
f
at 5% acetone in CH
2
Cl
2
: 0.5. HRMS (FAB)
calculated for C
10
H
15
N
2
O
5
[M+H]
+
: 243.0981, found: 243.0973.
3-(Ethoxycarbonyl)-1-methyl-4-benzoylpiperazine-2,5-dione (16)
Compound 1b (210 mg, 1 mmol) was dissolved in CH
2
Cl
2
(2 mL) and 4-
(dimethylamino)pyridine (183 mg, 1.5 mmol) was added to the solution. Then, benzoyl
189
chloride (174 µL, 1.5 mmol) was added to the reaction mixture over 10 minutes at room
temperature and the reaction mixture was stirred for 12 h. The reaction mixture was
concentrated and the product isolated by column chromatography (silica gel,
CH
2
Cl
2
/acetone (95:5)) to yield 16 (249 mg, 82%) as a white solid. mp. 139-141
o
C. IR
(acetone) ν
max
(cm
-1
) 1739, 1689, 1376, 1265, 1192, 1020.
1
H NMR (500 MHz, CDCl
3
) δ
7.67-7.41 (5H, m), 5.27 (1H, s), 4.30 (1H, d, J=19 Hz), 3.97 (1H, d, J=19 Hz), 3.1 (3H,
s), 1.37 (3H, t, J=7 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 171.2, 166.4, 166.2, 160.2,
133.5, 132.9, 128.9, 128.2, 63.32, 61.84, 52.84, 33.79, 13.97. R
f
at 5% acetone in
CH
2
Cl
2
: 0.6. HRMS (FAB) calculated for C
15
H
17
N
2
O
5
[M+H]
+
: 305.1138, found:
305.1143.
3-(Ethoxycarbonyl)-1-methyl-4-tosylpiperazine-2,5-dione (17)
To a solution of 1b (210 mg, 1.0 mmol) in CH
2
Cl
2
(3 mL) was added 4-
(dimethylamino)pyridine (244 mg, 2.0 mmol). Tosyl chloride (381 mg, 2.0 mmol) was
then added to the reaction. After stirring for 3 d, the reaction mixture was concentrated
under reduced pressure and the product was isolated by column chromatography (silica
gel, CH
2
Cl
2
/acetone (95:5)) to yield 17 (145 mg, 41%) as a colorless oil. IR (acetone)
ν
max
(cm
-1
) 1740, 1692, 1598, 1366, 1174, 1089, 1023, 917, 672, 538.
1
H NMR (250
MHz, CDCl
3
) δ 7.90 (2H, dd, J=9 Hz and J=2 Hz), 7.33 (2H, d, J=8 Hz), 5.64 (1H, s),
4.19 (1H, d, J=18 Hz), 4.38-4.13 (2H, m), 3.77 (1H, d, J=18 Hz), 2.96 (3H, s), 2.66 (3H,
190
s), 1.31 (3H, t, J=7 Hz).
13
C NMR (75 MHz, CDCl
3
) δ 165.75, 163.43, 159.36, 146.1,
134.1, 129.48, 129.37, 63.57, 61.70, 52.7, 34.13, 21.76, 13.98. R
f
at 5% acetone in
CH
2
Cl
2
: 0.65. HRMS (FAB) calculated for C
15
H
19
N
2
O
6
S [M+H]
+
: 355.0965, found:
355.0954.
3-(tert-butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (18)
LHMDS (1.0 M solution in THF) (35.2 mL, 35.2 mmol) was added dropwise to a
suspension of sarcosine anhydride (2.00 g, 14.1 mmol) in dry THF (40 mL) under a N
2
atmosphere was cooled to –78
o
C. Once the addition was complete, the mixture was
allowed to stir for 5 min. Then, Boc
2
O (9.23 g, 42.3 mmol) in dry THF (10 mL) was
added dropwise. The temperature was maintained for 15 min, and then the reaction was
warmed to room temperature and stirred for an additional 3 h. A saturated NH
4
Cl
solution (40 mL) was added and stirring was continued for 10 min. Most of the THF was
removed under reduced pressure, and the remaining residue was extracted with CH
2
Cl
2
(3
x 50 mL). The organic extracts were combined, dried over MgSO
4
, filtered, and
concentrated. The crude oil was purified by column chromatography (silica-gel,
CH
2
Cl
2
/acetone (85:15)) to afford 18 (2.03 g, 60% yield) as a white solid. mp. 71-73
o
C.
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1737, 1678, 1156, 1030.
1
H NMR (500 MHz, CDCl
3
) δ 4.39
(1H, s), 4.13 (1H, d, J=17 Hz), 3.80 (1H, d, J=17 Hz), 2.95 (3H, s), 2.91 (3H, s), 1.46
(9H, s).
13
C NMR (125 MHz, CDCl
3
) δ 165.4, 164.5, 160.6, 84.35, 66.86, 51.71, 33.68,
191
32.66, 27.78. R
f
at 15% acetone in CH
2
Cl
2
: 0.4. HRMS (FAB) calculated for
C
11
H
18
N
2
O
4
[M+H]
+
: 243.1346, found: 243.1353.
1,4-Diethylpiperazine-2,5-dione (12)
Dry DMF (50 mL) was added to 2,5-piperazinedione (2.50 g, 21.9 mmol) in under a N
2
atmosphere was cooled to 0
o
C in an ice bath. NaH (60% in mineral oil) (2.63 g, 65.8
mmol) was then added portionwise over the course of 5 min, and the mixture was
allowed to stir for 15 min. Then, iodoethane (7.23 g, 3.71 mL, 46.0 mmol) was added
dropwise with stirring. The reaction was warmed to room temperature and stirred
overnight. MeOH (25 mL) was then added, and the solvent was removed under reduced
pressure. The residue was extracted with CH
2
Cl
2
(3 x 50 mL). The organic extracts were
combined, dried over MgSO
4
, filtered, and concentrated. The crude solid was purified by
column chromatography (silica-gel, CH
2
Cl
2
/acetone (7:3)) to afford 12 (2.49 g, 67%
yield) as a white solid. mp. 125-127
o
C. IR (acetone) ν
max
(cm
-1
) 1723, 1673, 1405,
1327, 1256, 1017.
1
H NMR (500 MHz, CDCl
3
) δ 3.89 (4H, s), 3.39 (4H, q, J=7 Hz), 1.10
(6H, t, J=7 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 162.8, 49.16, 40.58, 11.65. R
f
at 30%
acetone in CH
2
Cl
2
: 0.4. HRMS (EI) calculated for C
8
H
14
N
2
O
2
M
+
: 170.1055, found:
170.1060.
192
3-(Ethoxycarbonyl)-1,4-diethylpiperazine-2,5-dione (19)
A suspension of 1,4-diethyl-2,5-piperazinedione 5 (500 mg, 2.94 mmol) in dry THF (10
mL) under a N
2
atmosphere was cooled to –78
o
C. LHMDS (1.0 M solution in THF)
(7.35 mL, 7.35 mmol) was added dropwise with stirring. Once the addition was
complete, the mixture was allowed to stir for 5 min. Then, Et
2
CO
3
(1.04 g, 1.10 mL, 8.82
mmol) was added dropwise. The temperature was maintained for 15 min, and then the
reaction was warmed to room temperature and stirred for an additional 3 h. A saturated
NH
4
Cl solution (10 mL) was added and stirring was continued for 10 min. Most of the
THF was removed under reduced pressure, and the remaining residue was extracted with
CH
2
Cl
2
(3 x 25 mL). The organic extracts were combined, dried over MgSO
4
, filtered,
and concentrated. The crude oil was purified by column chromatography (silica-gel,
CH
2
Cl
2
/acetone (85:15)) to afford 19 (624 g, 88% yield) as a colorless oil. IR (CH
2
Cl
2
)
ν
max
(cm
-1
) 1742, 1677, 1350, 1302, 1263, 1202, 1052, 1020.
1
H NMR (500 MHz,
CDCl
3
) δ 4.52 (1H, s), 4.26-4.16 (2H, m), 4.14 (1H, d, J=17 Hz), 3.75 (1H, d, J=17 Hz),
3.61 (1H, sextet, J=7 Hz), 3.50 (1H, sextet, J=7 Hz), 3.29 (1H, sextet, J=7 Hz), 3.20 (1H,
sextet, J=7 Hz), 1.24 (3H, t, J=7 Hz), 1.13-1.07 (6H, m).
13
C NMR (125 MHz, CDCl
3
) δ
167.0, 164.4, 160.1, 64.06, 62.79, 49.45, 41.26, 40.74, 13.93, 12.17, 11.67. R
f
at 10%
acetone in CH
2
Cl
2
: 0.5. HRMS (EI) calculated for C
11
H
18
N
2
O
4
M
+
: 242.1267, found:
242.1256.
193
3-(Ethoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione (20)
To a suspension of 1,4-dibenzyl-2,5-piperazinedione (750 mg, 2.54 mmol) in dry THF
(15 mL) under a N
2
atmosphere at –78
o
C was added LHMDS (1.0 M solution in THF)
(6.36 mL, 6.36 mmol) dropwise with stirring. Once the addition was complete, the
mixture was allowed to stir for 5 min. Then, Et
2
CO
3
(900 mg, 0.923 mL, 7.26 mmol)
was added dropwise. The temperature was maintained for 15 min, and then the reaction
was warmed to room temperature and stirred for an additional 3 h. A saturated NH
4
Cl
solution (40 mL) was added and stirring was continued for 10 min. Most of the THF was
removed under reduced pressure, and the remaining residue was extracted with CH
2
Cl
2
(3
x 50 mL). The organic extracts were combined, dried over MgSO
4
, filtered, and
concentrated. The crude oil was purified by column chromatography (silica gel,
hexane/CH
2
Cl
2
/acetone (5:4:1)) to afford 20 (725 mg, 78% yield) as a white solid. mp.
136-137
o
C. IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1728, 1672, 1380, 1082.
1
H NMR (500 MHz,
CDCl
3
) δ 7.35-7.20 (10H, m), 4.90 (1H, d, J=15 Hz), 4.78 (1H, d, J=15 Hz), 4.58 (1H, s),
4.39 (1H, d, J=15 Hz), 4.31 (1H, d, J=15 Hz), 4.12-4.07 (3H, m), 3.83 (1H, d, J=17 Hz),
1.18 (3H, t, J=7 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 166.5, 164.8, 160.6, 134.7, 134.4,
128.9, 128.8, 128.3, 128.2, 128.1, 63.46, 62.88, 49.77, 49.57, 48.84. 13.95. R
f
at
hexane/CH
2
Cl
2
/acetone (5:4:1): 0.45. HRMS (FAB) calculated for C
21
H
22
N
2
O
4
[M+H]
+
:
367.1658, found: 367.1658.
194
3-(tert-butoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione (21)
To a suspension of 1,4-dibenzyl-2,5-piperazinedione (2.0, 6.77 mmol) in dry THF (30
mL) under argon atmosphere at -78 ˚C was added LHMDS (1.0 M solution in THF) (17.0
mL, 17.0 mmol) dropwise with stirring. Once the addition was complete, the mixture
was allowed to stir for 10 min. Then, Boc
2
O (4.62 g, 21.2 mmol) in dry THF (10 mL)
was added dropwise. The temperature was maintained for 15 min, and then the reaction
was warmed to room temperature and stirred for an additional 3 h. A saturated NH
4
Cl
solution (35 mL) was added and stirring was continued for 20 min. Most of the THF was
removed under reduced pressure, and the remaining residue was extracted with CH
2
Cl
2
(3
x 75 mL). The organic extracts were combined, dried over MgSO
4
, filtered, and
concentrated. The crude product was purified by column chromatography (silica gel,
EtOAc/hexane (7:3)) to afford 21 (1.93 g, 72% yield) as a white solid. mp. 142-144
o
C.
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1735, 1668, 1461, 1327, 1262, 1156, 1070
1
H NMR (500 MHz,
CDCl
3
) δ 7.37-7.24 (10H, m), 5.03 (1H, d, J = 15 Hz), 4.94 (1H, d, J = 15 Hz), 4.50 (1H,
s), 4.29 (1H, d, J = 15 Hz), 4.21 (1H, d, J = 15 Hz), 4.11 (1H, d, J = 17 Hz), 3.85 (1H, d, J
= 17 Hz), 1.37 (9H, s)
13
C NMR (125 MHz, CDCl
3
) 165.29, 164.92, 160.94, 134.92,
134.58, 128.98, 128.88, 128.87, 128.27, 128.17, 128.14, 84.28, 64.48, 49.86, 49.84,
48.85, 27.71. R
f
in 30% Hexanes in EtOAc: 0.4. HRMS (EI) calculated for C
23
H
27
N
2
O
4
[M+H]
+
: 395.1965, found: 395.1961.
195
6.2.3 Synthesis procedures and analytical data for sulfenylated compounds.
1-phenylsulfanyl[1,2,4]triazole (22b)
Sulfuryl chloride (3.24 mL, 40 mmol) was added dropwise to a solution of diphenyl
sulfide (9.86 g, 40 mmol) in CH
2
Cl
2
(40 mL). Initially the addition was done at room
temperature but after initiation of the reaction as the solution turned dark orange, further
addition was done at 0 ºC. After 30 minutes the resulting solution was added drop wise to
a solution of 1,2,4-triazole (6.9 g, 100 mmol) an triethylamine (13.93 mL, 100 mmol) in
CH
2
Cl
2
(40 mL) at 0 ºC . The reaction mixture was stirred for 30 minutes and the solvent
was evaporated in vacuum at 25 ºC. The product was extracted from the reaction mixture
using pentane/CH
2
Cl
2
(3:7) mixture (3 x 100 mL). The solvents were evaporated in vacuo
at 25 ºC. The crude product was purified by column chromatography (diethyl
ether/pentane (1:1) followed by pure diethyl ether). The product 22b was stored under
argon at -20 ºC as it decomposes at room temperature in air. IR (CH
2
Cl
2
) ν
max
(cm
-1
)
3108, 1581, 1473, 1368, 1331, 1268, 1198, 1141, 1100
1
H NMR (500 MHz, CDCl
3
) δ
8.39 (1H, s), 8.07 (1H, s), 7.36-7.29 (5H, m)
13
C NMR (125 MHz, CDCl
3
) δ 153.61,
150.58, 134.32, 129.14, 129.05, 128.73. R
f
in 5% acetone in CH
2
Cl
2
: 0.4. HRMS (EI)
calculated for C
8
H
8
N
3
S [M+H]
+
: 178.0433, found: 178.0432.
General procedure for sulfenylation. A flame-dried flask purged with N
2
and equipped
with a magnetic stirring bar was charged with diketopiperazine, catalyst, and solvent.
This mixture was allowed to reach the desired temperature, followed by addition of the
sulfenylating reagent. The mixture was stirred for the appropriate period of time. Then,
196
the reaction was quenched with saturated solution of NH
4
Cl (same volume as solvent)
and extracted three times with CH
2
Cl
2
. The organic extracts were combined, dried over
anhydrous MgSO
4
, filtered, and concentrated in vacuo. The crude oils were purified by
column chromatography (vide supra).
3-Benzylsulfanyl-3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (24)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1745, 1668, 1460, 1390, 1263, 1048.
1
H NMR (500 MHz,
CDCl
3
) δ 7.37-7.26 (5H, m), 4.40-4.30 (2H, m), 3.91-3.77 (4H, m), 3.00 (3H, s), 2.76
(3H, s), 1.34 (3H, t, J=7 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 165.0, 163.2, 160.7, 135.6,
128.6, 128.5, 127.5, 76.79, 63.75, 50.89, 34.61, 33.79, 30.51, 13.93. R
f
in 5% acetone in
CH
2
Cl
2
: 0.45. HRMS (FAB) calculated for C
16
H
20
N
2
O
4
S [M+H]
+
: 337.1222, found:
337.1236.
3-Benzylsulfanyl-3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione (25)
197
IR (acetone) ν
max
(cm
-1
) 3219,3107, 1746, 1680, 1403, 1239, 1039.
1
H NMR (600 MHz,
CDCl
3
) δ 7.96 (1H, s), 7.33 (5H, m), 4.30-3.96 (4H, m), 4.95 (1H, d, J=18 Hz), 3.87 (1H,
d, J=18 Hz), 2.82 (3H, s), 1.31 (3H, t, J=7 Hz).
13
C NMR (150 MHz, CDCl
3
) δ 165.9,
165.6, 160.3, 135.6, 128.9, 128.4, 127.3, 68.54, 63.61, 51.53, 35.26, 34.24, 13.70. R
f
at
5% acetone in CH
2
Cl
2
: 0.35. HRMS (FAB) calculated for C
15
H
19
N
2
O
4
S [M+H]
+
:
323.1066, found: 323.1078.
3-Phenylsulfanyl-3-(methoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (26)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1753, 1677, 1439, 1392, 1337, 1247, 1173, 1049
1
H NMR (500
MHz, CDCl
3
) δ 7.52-7.35 (5H, m), 3.90(3H, t), 3.40 (1H, d, J = 18 Hz), 3.07 (3H, s), 2.72
(3H, s), 2.06 (1H, d, J = 18 Hz)
13
C NMR (125 MHz, CDCl
3
) δ 165.29, 162.84, 160.60,
138.00, 131.11, 129.12, 128.21, 81.20, 54.36, 50.02, 33.55, 30.22. R
f
in 10% acetone in
CH
2
Cl
2
: 0.45. HRMS (EI) calculated for C
14
H
17
N
2
O
4
S [M+H]
+
: 309.0904, found:
309.0901.
3-Phenylsulfanyl-3-(tert-butoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (27)
198
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1744, 1677, 1463, 1387, 1336, 1258, 1157, 1045
1
H NMR (500
MHz, CDCl
3
) δ 7.53-7.36 (5H, m), 3.41 (1H, d, J = 18 Hz), 3.14 (3H, s), 2.75 (3H, s),
2.04 (1H, d, 18 Hz), 1.57 (9H, s)
13
C NMR (125 MHz, CDCl
3
) δ 163.49, 163.10, 161.03,
138.02, 130.92, 129.03, 128.75, 85.58, 81.74, 50.07, 33.47, 30.00, 27.70. R
f
in 5%
acetone in CH
2
Cl
2
: 0.45. HRMS (EI) calculated for C
17
H
22
N
2
O
4
SNa
+
: 373.1192, found:
373.1189.
3-Phenylsulfanyl-3-(ethoxycarbonyl)-1,4-diethylpiperazine-2,5-dione (28)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1748, 1673, 1460, 1414, 1357, 1294, 1248, 1165, 1069, 1019
1
H
NMR (500 MHz, CDCl
3
) δ 7.51-7.33 (5H, m), 4.44-4.31 (2H, m), 3.93-3.88 (1H, m),
3.44 (1H, d, J = 18 Hz), 3.33-3.22 (3H, m), 2.20 (1H, d, J = 18 Hz), 1.36 (3H, t, J = 7
Hz), 1.28 (3H, t, J = 7 Hz), 1.0 (3H, t, J = 7 Hz)
13
C NMR(125 MHz, CDCl
3
) δ 165.25,
163.44, 160.67, 138.22, 130.88, 129.15, 128.11, 81.30, 63.75, 47.72, 41.56, 41.34, 13.86,
12.72, 11.41. R
f
in 5% acetone in CH
2
Cl
2
: 0.6. HRMS (EI) calculated for
C
17
H
22
N
2
O
4
SNa
+
: 373.1192, found: 373.1204.
199
3-Phenylsulfanyl-3-(ethoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione (29)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1749, 1675, 1465, 1245, 1172, 1025
1
H NMR (500 MHz, CDCl
3
)
δ 7.40-7.09 (15H, m), 5.46 (1H, d, J = 16 Hz), 4.58 (1H, d, J = 15 Hz), 4.46 (1H, d, J = 16
Hz), 4.13 (1H, d, J = 15 Hz), 3.93-3.89 (1H, m), 3.52 (1H, d, J = 18 Hz), 3.17-3.13 (1H,
m), 2.19 (1H, d, J = 18 Hz), 0.86 (3H, t, J = 7 Hz)
13
C NMR (125 MHz, CDCl
3
) δ 164.56,
164.11, 161.41, 138.02, 135.79, 134.22, 130.85, 129.09, 129.01, 128.82, 128.63, 128.31,
128.17, 127.71, 127.56, 81.13, 63.18, 50.12, 47.69, 47.07, 13.20. R
f
in 5% acetone in
CH
2
Cl
2
: 0.65. HRMS (EI) calculated for C
27
H
26
N
2
O
4
SNa
+
: 497.1505, found: 497.1501.
3-Phenylsulfanyl-3-(tert-butoxycarbonyl)-1,4-dibenzylpiperazine-2,5-dione (30)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1746, 1680, 1462, 1402, 1238, 1038
1
H NMR (500 MHz, CDCl
3
)
δ 7.46-7.18 (15H, m), 5.3 (1H, d, J = 16 Hz), 4.86 (1H, d, J = 14 Hz), 4.63 (1H, d, J = 16
Hz), 3.94 ( J = 14 Hz), 3.56 (1H, d, J = 18 Hz), 2.17 (1H, d, J = 18 Hz)
13
C NMR (125
MHz, CDCl
3
) δ 164.26, 163.38, 161.77, 138.06, 135.38, 134.38, 130.79, 129.03, 128.77,
128.63, 128.20, 128.17, 128.01, 127.64, 127.21, 85.55, 82.49, 49.70, 47.96, 47.52, 26.99.
R
f
in 30% Hexanes in EtOAc: 0.5. HRMS (EI) calculated for C
29
H
31
N
2
O
4
S [M+H]
+
:
503.1999, found: 503.2.
200
3-Phenylsulfanyl-3-(ethoxycarbonyl)-1,4-dimethylpiperazine-2,5-dione (31)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1749, 1677, 1440, 1391, 1337, 1242, 1173, 1046
1
H NMR (500
MHz, CDCl
3
) δ 7.51-7.35 (5H, m), 4.42-4.31 (2H, m), 3.40(1H, d, J = 18 Hz), 3.08 (3H,
s), 2.72 (3H, ds, J = 5 Hz), 2.05 (1H, d, J = 18 Hz), 1.34 (3H, t, J = 7 Hz)
13
C NMR (125
MHz, CDCl
3
) δ 164.58, 162.82, 160.53, 137.90, 130.95, 128.98, 128.22, 81.15, 63.83,
49.91, 33.41, 30.03, 13.87. R
f
in 10% acetone in CH
2
Cl
2
: 0.55. HRMS (EI) calculated for
C
15
H
19
N
2
O
4
S [M+H]
+
: 323.1060, found: 323.1058.
3-Phenylsulfanyl-3-(ethoxycarbonyl)-1-methylpiperazine-2,5-dione (32)
IR (CH
2
Cl
2
) ν
max
(cm
-1
) 1749, 1679, 1464, 1403, 1245, 1181, 1046
1
H NMR (500 MHz,
CDCl
3
) δ 7.61-7.29 (6H, m), 4.41-4.38(2H, m), 3.52 (1H, d, 18 Hz), 2.82 (3H, s), 2.52
(1H, dd, J = 18 Hz and J = 2 Hz), 1.39 (3H, t, J = 7 Hz)
13
C NMR(125 MHz, CDCl
3
) δ
165.23, 164.17, 160.31, 138.30, 131.06, 129.26, 128.03, 73.40, 64.20, 50.74, 34.24,
14.00. R
f
in 15% acetone in CH
2
Cl
2
: 0.4. HRMS (EI) calculated for C
14
H
16
N
2
O
4
SNa
+
:
331.0723, found: 331.0728.
201
6.2.4 Synthesis procedures and analytical data for compounds prepared by alkylation
ofdiketopiperazines with indoles.
Ethyl 2-((1H-indol-3-yl)methyl)-4-methyl-3,6-dioxopiperazine-2-carboxylate (33)
To a suspension of gramine (523 mg, 3.0 mmol) in acetonitrile (5.0 mL) was added
compound 14 (300 mg, 1.498 mmol). The mixture was stirred under reflux for 24 h. The
reaction mixture was concentrated under vacuum. The crude reaction mixture was
purified by column chromatography (silica gel, CH
2
Cl
2
/acetone (70:30)) to yield 33 in
92% yield (454 mg, 1.378 mmol). IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3375, 1665.
1
H NMR (500
MHz, CDCl
3
) δ 8.21 (1H, s), 7.62-7.09 (5H, m), 6.11 (1H,s), 4.34-4.27 (2H, m), 3.85
(1H, d, J=18 Hz), 3.74 (1h, d, J=15 Hz), 3.66 (1H, d, J=15 Hz), 3.29 (1H, d, J=18 Hz),
2.86 (3H, s), 1.33 (3H, t, J=4 Hz).
13
C NMR (125 MHz, CDCl
3
) δ 168.45, 165.77, 163.03,
127.44, 124.40, 122.67, 120.26, 118.77, 111.25, 107.70, 66.70, 63.12, 51.80, 34.38,
31.90, 14.11. R
f
at 30% acetone in CH
2
Cl
2
: 0.45. HRMS (ESI) calculated for
C
17
H
19
N
3
NaO
4
[M+Na]
+
: 352.1268, measured: 352.1262.
202
Ethyl 2-((1H-indol-3-yl)methyl)-1,4-dimethyl-3,6-dioxopiperazine-2-carboxylate (34)
To a suspension of gramine (488 mg, 2.80 mmol) in acetonitrile (5.0 mL) was added
compound 11 (300 mg, 1.40 mmol). The mixture was stirred under reflux for 48 h. The
reaction mixture was concentrated under vacuum. The crude reaction mixture was
purified by column chromatography (silica gel, CH
2
Cl
2
/acetone (75:25)) to yield 34 in
81% yield (389 mg, 1.134 mmol). IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3314, 1750, 1668, 1246.
1
H
NMR (500 MHz, CDCl
3
) δ 8.48 (1H, s), 7.56-6.90 (5H, m), 4.43-4.31 (2H, m), 3.91 (1H,
d, J=15 Hz), 3.44 (1H, d, J=15 Hz), 3.35 (1H, d, J=17 Hz), 3.0 (3H, s), 2.46 (3H, s), 2.20
(1H, d, J=17 Hz), 1.37 (3H, m).
13
C NMR (125 MHz, CDCl
3
) δ 167.54, 164.086, 163.82,
135.90, 126.92, 124.22, 122.50, 119.87, 119.17, 111.10, 107.72, 72.45, 62.70, 50.48,
33.43, 29.95, 29.43, 14.07. R
f
at 25% acetone in CH
2
Cl
2
: 0.45. HRMS (ESI) calculated
for C
18
H
21
N
3
NaO
4
[M+Na]
+
: 366.1424, measured: 366.1427.
Tert-2 butyl-((1H-indol-3-yl)methyl)-1,4-dimethyl-3,6-dioxopiperazine-2-carboxylate
(35)
203
To a suspension of gramine (432 mg, 2.476 mmol) in acetonitrile (5.0 mL) was added
compound 18 (300 mg, 1.238 mmol). The mixture was stirred under reflux for 48 h. The
reaction mixture was concentrated under vacuum. The crude reaction mixture was
purified by column chromatography (silica gel, CH
2
Cl
2
/acetone (80:20)) to yield 35 in
72% yield (332 mg, 0.891 mmol). IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3314, 1745, 1668, 1258, 1160.
1
H NMR (500MHz, CDCl
3
) δ 8.91 (1H, s), 7.55-6.85 (5H, m), 3.85 (1H, d, 15 Hz), 3.37
(1H, d, 15 Hz), 3.32 (1H, d, 17 Hz), 3.03 (3H, s), 2.43 (3H, s), 2.16 (1H, d, 17 Hz), 1.55
(9H, s).
13
C NMR (125 MHz, CDCl
3
) δ 166.40, 164.19, 135.94, 126.92, 124.20, 122.26,
119.62, 119.07, 111.12, 107.68, 83.49, 72.85, 50.45, 33.31, 29.80, 29.29, 27.83. R
f
at
20% acetone in CH
2
Cl
2
: 0.40. HRMS (ESI) calculated for C
20
H
25
N
3
NaO
4
[M+Na]
+
:
394.1737, measured: 394.1738.
Ethyl 1-((2-methyl-1H-indol-3-yl)(phenyl)methyl)-2-oxocyclopentanecarboxylate
(36)
To a solution of 40 (75 mg, 0.2 mmol) in a solvent (2.0 mL) was added compound 5 (94
mg, 0.6 mmol). To the reaction mixture 15 mol% (with respect to 5) of the cinchona
alkaloid organocatalyst was added. Subsequently 80 mg KF/alumina was added and the
reaction mixture was stirred at room temperature. The crude reaction mixture was
purified with column chromatography (silica gel, CH
2
Cl
2
/acetone (95:05)) to obtain 7 in
204
the mentioned yields. IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3393, 1717, 1639, 1467, 1230.
1
H NMR
(500MHz, CDCl
3
) δ 7.82-6.86 (10H, m), 5.50 (1H, d, 24 Hz), 4.16-3.79 (2H, m), 3.3-1.2
(6H, m), 2.45 (3H, s), 1.15-0.76 (3H, m).
13
C NMR (125 MHz, CDCl
3
) δ 129.30, 128.15,
127.92, 126.30, 125.69, 120.78, 120.31, 119.30, 110.1, 61.6, 61.34, 46.42, 38.55, 31.95,
30.96, 30.64, 20.12, 19.63, 13.66, 13.44, 12.7. R
f
at 5% acetone in CH
2
Cl
2
: 0.65. HRMS
(ESI) calculated for C
24
H
25
NNaO
3
[M+Na]
+
: 398.1727, measured: 398.1731.
Ethyl 1,4-dimethyl-2-(-(2-methyl-1H-indol-3-yl)(phenyl)methyl)-3,6-
dioxopiperazine-2-carboxylate (37)
To a solution of 40 (150 mg, 0.4 mmol) in a solvent (3.0 mL) was added compound 11
(257 mg, 1.2 mmol). To the reaction mixture 15 mol% (with respect to 11) of the
cinchona alkaloid organocatalyst was added. Subsequently 160 mg KF/alumina was
added and the reaction mixture was stirred at room temperature. The crude reaction
mixture was purified with column chromatography (silica gel, CH
2
Cl
2
/acetone (85:15)) to
obtain 37 in the mentioned yields. IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3322, 1753, 1664, 1403, 1237,
1050.
1
H NMR (500MHz, CDCl
3
) δ 8.26-6.91 (10H, m), 5.78; 4.93 (1H, ds), 4.40-4.24
(2H, m), 3.52 (d, J=17 Hz), 3.33-2.35 (11H, m), 1.36-1.22 (3H, m).
13
C NMR (125 MHz,
CDCl
3
) δ 167.30, 166.97, 164.19, 164.17, 163.65, 162.1, 140.32, 139.46, 135.22, 135.16,
132.16, 130.97, 129.69, 127.41, 127.33, 127.08, 126.89, 125.81, 121.36, 120.92, 120.44,
120.42, 119.92, 119.56, 110.64, 110.23, 109.23, 109.21, 108.97, 62.69, 62.45, 51.44,
205
50.75, 48.83, 48.37, 36.07, 33.78, 33.33, 30.63, 14.06, 13.91, 12.60, 12.28. R
f
at 15%
acetone in CH
2
Cl
2
: 0.45. HRMS (ESI) calculated for C
25
H
27
N
3
NaO
4
[M+Na]
+
: 456.1894,
measured: 456.1896.
Tert-butyl 1,4-dimethyl-2-((2-methyl-1H-indol-3-yl)(phenyl)methyl)-3,6-
dioxopiperazine-2-carboxylate (38)
To a solution of 40 (150 mg, 0.4 mmol) in a solvent (3.0 mL) was added compound 18
(291 mg, 1.2 mmol). To the reaction mixture 15 mol% (with respect to 18) of the
cinchona alkaloid organocatalyst was added. Subsequently 160 mg KF/alumina was
added and the reaction mixture was stirred at room temperature. The crude reaction
mixture was purified with column chromatography (silica gel, CH
2
Cl
2
/acetone (90:10)) to
obtain 38 in the mentioned yields. IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3386, 1647, 729.
1
H NMR
(500MHz, CDCl
3
) δ 8.08-6.88 (10H, m), 5.74; 4.83 (1H, ds), 3.51-2.0 (11H, m), 1.54-
1.40 (9H, m).
13
C NMR (125 MHz, CDCl
3
) δ 166.0, 164.26, 139.73, 135.17, 135.09,
131.0, 129.77, 128.04, 127.36, 127.0, 120.94, 120.53, 119.95, 110.54, 109.36, 83.41,
51.40, 48.56, 36.14, 33.74, 33.32, 27.84, 27.61, 12.70. R
f
at 10% acetone in CH
2
Cl
2
:
0.40. HRMS (ESI) calculated for C
27
H
31
N
3
NaO
4
[M+Na]
+
: 484.2207, measured: 484.22.
206
Tert-butyl 2-((1H-indol-3-yl)(phenyl)methyl)-1,4-dimethyl-3,6-dioxopiperazine-2-
carboxylate (39)
To a solution of 41 (145 mg, 0.4 mmol) in a solvent (3.0 mL) was added compound 18
(291 mg, 1.2 mmol). To the reaction mixture 15 mol% (with respect to 18) of the
cinchona alkaloid organocatalyst was added. Subsequently 160 mg KF/alumina was
added and the reaction mixture was stirred at room temperature. The crude reaction
mixture was purified with column chromatography (silica gel, CH
2
Cl
2
/acetone (90:10)) to
obtain 39 in the mentioned yields. IR (CH
2
Cl
2
) ν
max
(cm
-1
) 3398, 1749, 1660, 1255, 1159.
1
H NMR (500MHz, CDCl
3
) δ 8.18-7.01 (11H, m), 5.12 (1H, s), 3.41 (1H, s), 3.41 (1H, d,
J=18Hz), 3.15(3H, s), 2.63 (3H, s), 2.17 (1H, d, J=18Hz), 1.37 (9H, s).
13
C NMR (125
MHz, CDCl
3
) δ 165.40, 164.49, 163.23, 138.60, 135.07, 130.19, 128.06, 127.79, 127.63,
124.0, 121.83, 119.25, 118.51, 113.67, 110.77, 83.57, 50.53, 47.26, 33.43, 31.56, 27.62.
R
f
at 10% acetone in CH
2
Cl
2
: 0.45. HRMS (ESI) calculated for C
26
H
29
N
3
NaO
4
[M+Na]
+
:
470.2050, measured: 470.2053.6.3
6.3 Experimental Section for Chapter 3
6.3.1 Protein Expression. CH1 domain (a.a. 323-423) was cloned into puc57. Using
EcoR1 and BamH1 the CH1 DNA sequence was subcloned into pGEX 4T-2 vector. The
pGEX 4T-2 fusion vector was transformed into BL21 DE3 pLys competent E. coli
207
(Novagen). Production of the desired p300-CH1-GST fusion product was verified by
SDS-PAGE and confirmed by sequencing.
For protein expression, BL21 DE3 pLys competent E. coli with pGEX 4T-2 fusion vector
were grown in 10 mL LB media overnight. Overnight culture was added to 1 L LB media
and allowed to grow to ~ 0.7 O.D. after which IPTG at final concentration of 200 µM
was added to induce protein expression. After 6 h E. Coli were centrifuged at 4400 rpm
and pellets were resuspended in 30 ml lysis buffer (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 (70% solution, Sigma), 50 mg/mL RNase A (Sigma), and 50
mg/mL DNase A (Sigma) at pH 8.0). Lysate was frozen and thawed once, followed by
sonication to lyse the E. Coli bacteria. Sonicate was centrifuged at 18,000 x g for 45 min
at 4 °C. To the supernatant was added glutathione sepharose beads 4B (GE Health Care).
After 2 h at 4 °C and 2 h at RT the beads were washed with PBS (10 mL, 2 times),
protein-buffer-A (10 mL, 4 times). Protein-buffer-A is 50 mM Tris (70% solution,
Sigma), 150 mM NaCl (Fisher), 100 μM ZnCl2 (Sigma), 1 mM DTT (Fisher), 0.1%
NP40 (Sigma) at pH 8.0. Two washing were given with 10 mL of 10 mM 1, 10
Phenanthroline buffer in protein buffer. This was followed by four washings with
protein-buffer-A (10 mL). Elution of CH1-p300-GST protein was accomplished with
reduced glutathione in protein buffer at the following concentration and amount: 2.5 mM
GSH (4 x 2 mL), 5 mM GSH (5 x 2 mL) and 10 mM GSH (5 x 2 mL). For each elution
step incubation of 5 min was given at RT. Collected fractions were assayed by SDS
PAGE gel; pooled fractions were treated with protease inhibitor cocktail (Sigma) and
dialyzed against protein-buffer-A with 10% to ensure proper folding and removal of
208
glutathione and protease inhibitors. 10% glycerol was introduced so that the dialyzed
protein can be stored at -80 °C. After 12 h of dialysis with three buffer changes the
protein was aliquoted and stored at -80 °C.
6.3.2
15
N-labeled CH1-p300-GST expression. Expression of CH1-p300-GST with
15
N
atoms was achieved in two different ways with the idea of using minimum amounts of
expensive labeling reagents. The main goal was to grow E. Coli with pGEX-4T2 fusion
vector in 1 L LB media to an O.D. of ~0.8 and centrifuging the bacteria at 4400 rpm. In
the first method the pellets were washed with ice cold M9 salts solution and resuspended
in minimal media having
15
NH
4
Cl. Minimal media 1 L composition is 12.8 g
Na
2
HPO
4
·7H
2
O, 3 g KH
2
PO
4
, 0.5 g NaCl, 1 g
15
NH
4
Cl, 4 g glucose, 2 mL of 1 M
MgSO
4
, 100 µL of 1 M CaCl
2
, 1 µM ZnCl
2
, and 10 mL 100x MEM vitamins at pH 7.4.
After 1 h protein expression was induced with IPTG at final concentration of 200 µM.
After 5 h cells were centrifuged and resuspended in lysis buffer having the same
composition as mentioned above. Subsequent steps were similar as explained above for
normal protein expression. The second method for
15
N labeled protein expression is
similar to the first method and the difference is that in this method the bacteria are grown
in 4 L LB media instead of 1 L LB media to an O.D. of ~0.8 and centrifuged at 4400 rpm.
The pellets from the 4 L batch are resuspended in 1 L minimal media. Therefore, the
density of bacteria in minimal media is four times in this case as compared to the first
method. The second method gives approximately four times the yield of protein as
compared to the first method.
6.3.3 Thrombin-mediated CH1-p300-GST fusion protein cleavage. A general protocol
for thrombin cleavage used for cleaving CH1-p300-GST protein through the thrombin
209
cleavage site LVPRGS involved doing the cleavage on the glutathione sepharose beads.
The main benefit of cleaving surface bound GST fusion protein is that after the cleavage
the GST portion remains bound to the surface and the solution phase has the cleaved CH1
domain. Thrombin cleavage was done by suspending 300 µL of beads having
immobilized CH1-p300-GST protein, in 700 µL of protein-buffer-A having 10%
glycerol. To this suspension was added 15 U of thrombin. The suspension is mixed gently
for 16 h at room temperature after which the beads were centrifuged at 500 x g and
supernatant is collected. SDS PAGE is used check for the purity of cleaved protein.
Concentration of protein was determined by UV absorption and Bradford assay.
6.3.4 Immobilization of GST-CH1-p300 on SPR chip. For our experiments Biacore 2000
instrument (GE Healthcare) was used. The CM5 chip was used which consists of 4
channels having gold surface to which carboxymethylated dextran polymer is covalently
attached. Goat anti-GST antibody (50 µg/mL, Pharmacia) was covalently coupled to the
dextran layer. 1-ethyl-3-(3-dimethylpropyl)-carbodiimide (0.4 M) was mixed with 0.1 M
N-hydroxysuccinimide in a 1:1 (v/v) ratio and flowed over the CM5 chip to activate the
dextran layer. Anti-GST antibody was flowed over the channel for coupling reaction at
the rate of 15 µL/min for 10 minutes. This was followed by injection of ethanolamine-
HCl (1 M) at pH 8.5 at a flow rate of 5 µL/min for 10 minutes. After coupling of anti-
GST antibody to the chip GST-CH1-p300 was immobilized on the chip by allowing a 1
mg/mL solution to flow through the channels at the rate of 10 µL/min for 10 minutes.
The step of immobilization by injection of the protein solution was repeated several times
before desired level of immobilization of GST-CH1-p300 was achieved.
210
6.3.5 SPR binding assays of ETPs to immobilized GST-CH1-p300. All compounds were
assayed at each of the following concentrations of 50 μM, 10 μM, 1 μM, 500 nM, 200
nM, and 50 nM. Samples were prepared from DMSO stock solutions in a buffer
containing 10 mM Tris, 100 mM NaH
2
PO
4
, 500 μM DTT, and 100 μM ZnCl
2
at pH 8.0.
This buffer was used for all SPR assays and was filtered through a 0.2 μm filter and
thoroughly degassed prior to use or sample preparation. All experiments were performed
in triplicates. All samples contained a final concentration of 5% DMSO. For each
experiment, a sample containing 5% DMSO in buffer was run for each compound tested
to establish a baseline for the blank. All runs were also double-referenced against a flow
cell containing only immobilized anti-GST antibody but not p300-CH1-GST to account
for any nonspecific binding. A run consisted of 3.3 min with a 100 μL sample injection
(flow rate was 30 μL/min) followed by 3.3 min buffer flow as a wash. After each
injection, the chip surface was regenerated with a 1 min injection (flow rate of 30
μL/min) of 10 mM H
3
PO
4
following the buffer wash.
6.3.6 Determination of K
D
from the saturation binding curves between CH1-p300-
GST/CH1-p300 and HIF-1α-Fl in fluorescence polarization competition assays.
Different concentrations ranging from 1 nM to 4000 nM of CH1-p300-GST/CH1-p300
protein were made in the following buffer: 50 mM Tris, 150 mM NaCl, 10% glycerol, 1
mM DTT, 0.1% NP-40, 100 µM ZnCl
2
and 1% pluronic acid and pH of the buffer was
8.0. In the same buffer 30 nM HIF-1α-Fl solution was made. In an opaque (black)
multiwell plate 60 µL of HIF-1α-Fl solution was added to each well. To these wells 60
µL of different dilutions of CH1-p300-GST/CH1-p300 protein were added so that the
final volume in each well is 120 µL and final concentration of HIF-1α-Fl is 15 nM. Each
211
sample was done in triplicate. The multiwell plate was incubated at 25 ºC for 1h. The
fluorescence polarization measurement was done with BioTek HT multiplate reader. The
fluorescence polarization data obtained was normalized in terms of percentage change in
fluorescence polarization value and the data was plotted using sigma plot as 4 parameter
sigmoidal curve.
6.3.7 Fluorescence polarization competition assays to measure the disruption of CH1-
p300-GST/CH1-p300 and HIF-1α-Fl binding with bis-ETPs. To each well in an opaque
(black) multiwall plate 60 µL mixture having 150 nM CH1-p300-GST and 30 nM HIF-
1α-Fl –or- 1100 nM CH1-p300 and 30 nM HIF-1α-Fl in the following buffer: 50 mM
Tris, 150 mM NaCl, 10% glycerol, 1 mM DTT, 0.1% NP-40, 100 µM ZnCl
2
and 1%
pluronic acid and pH of the buffer was 8.0. Different dilutions of bis-ETPs (Chetomin,
ETP-3 or ETP-5) ranging from 0.001 µM to 20 µM were made in the same buffer. After
1 h of incubation given to the CH1-p300-GST/CH1-p300 and HIF-1α-Fl mixture 60 µL
of bis-ETP were added. The final concentration of CH1-p300-GST is 75 nM and for
CH1-p300 is 550 nM. The final concentration for HIF-1α-Fl was 15 nM. Each sample
was done in triplicate. After 1 h of incubation at 25 ºC done after addition of bis-ETPs the
fluorescence polarization values were measured on the BioTek HT multiplate reader. The
fluorescence polarization data obtained was normalized in terms of percentage change in
fluorescence polarization value and the data was plotted using sigma plot as 4 parameter
sigmoidal curve.
212
6.4 Experimental Section for Chapter 4
6.4.1 Cell lines. Human breast carcinoma (MCF7 and MDA-MB-231) and human
epithelial lung carcinoma (A549) cell lines were obtained from ATCC (accession
numbers CCL-2 and HTB-22). Aggressive human breast carcinoma stably transfected
with an HRE luciferase construct (MDA-MB-231-HRE-Luc) was a gift of Dr. Robert
Gillies.
6.4.2 Cell culture. MCF7 cells were maintained in RPMI 1640 media (Sigma)
supplemented with 10% fetal bovine serum, penicillin (50 U/mL) and streptomycin (50
µg/mL). MDA-MB-231-HRE-Luc cells were grown in high glucose DMEM
supplemented with 10% fetal bovine serum and 0.4 g/L geneticin (RPI). A549 cells were
grown in F-12K medium supplemented with 10% fetal bovine serum, penicillin (50
U/mL) and streptomycin (50 µg/mL). All cells were incubated at 37
o
C in a humidified
atmosphere with 5% CO
2
. Cell growth and morphology were monitored by phase-
contrast microscopy.
6.4.3 Luciferase assays. MDA-MB-231-HRE-Luc cells were plated in 24-well dishes
(BD Falcon) at a density of 6.5 × 10
4
cells/mL. After attachment, cells were treated with
1 mL of fresh media containing chetomin (150 nM), ETP-5 and NP-481 at 200 and 600
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, hypoxia was induced , and cells were incubated for another 18 h.
Whole cell lysate was isolated by washing the cells twice with ice-cold PBS and then
adding 150 μL of Cell Culture Lysis Reagent (CCLR) from the Luciferase Assay Kit
213
(Promega). Further steps were carried out according to the manufacturer’s instructions.
Relative light intensity was measured using a Turner TD-20e luminometer and the results
were normalized to total protein content determined by Bradford assay. Briefly, 50 μL of
cell lysate was added to 200 μL of Bradford reagent and 750 μL of water. Absorbance
was measured at 595 nm using a DU-800 spectrophotometer and normalized to protein
standards (1-10 μg/mL of BSA solution).
6.4.4 Cell viability assay. MCF7 cells were plated in opaque 96-well plates (Greinier) at a
density of 10,000 cells/well in 200 µL RPMI medium supplemented with 10% fetal
bovine serum. A549 cells were plated in opaque 96-well plates (Greinier) at a density of
5,000 cells/well in 200 µL serum-free F-12K. Both MCF7 and A549 cell lines were
grown to 70% confluency and were treated with 100 μL of fresh media containing
chetomin or ETP 5 at concentrations ranging from 0 to 1.5 µM and 0 to 7 µM for
chetomin in MCF7 and A549 cells respectively and for ETP 5 from 0 to 10 µM and 0 to
12 µM in MCF7 and A549 cells respectively. All samples contained a final concentration
of 0.1% to 0.5% DMSO. Vehicle samples were treated with cell culture media containing
0.1% DMSO. MCF7 cells were incubated with compounds for a total of 24 h. Once the
incubation was complete, 11 μL of MTT (Promega) stock solution (5 mg/mL in PBS)
was added to each well, and plates were incubated at 37 °C and 5% CO2 for 3.5 h. The
110 µL of media was removed and 100 µL of DMSO was added to each well. Plates were
incubated at 37 °C for 5 minutes to dissolve the MTT formazan crystals. Absorbance for
the plate was measured at 570 nm to quantify the amount of MTT formazan and
reference was absorbance at 690 nm. Synergy 2 Multi-Mode Microplate Reader (BioTek)
214
was used to read the plate. EC
50
(GI
50
) curves were plotted using Sigma Plot version 11.0
software (Systat Software, Inc.).
6.4.5 Isolation of mRNA. MCF7 cells were plated in 6-well plates (BD Falcon) in 2 mL
of media at a density of 0.75 × 10
5
cells/mL. After 48 h the attached cells were at 80%
confluency. Cells were treated with fresh media containing chetomin, ETP-5 or LS75 at
desired concentrations. All samples contained a final concentration of 0.1% DMSO;
vehicle samples were treated with cell culture media containing 0.1% DMSO. After 6 h,
hypoxia was induced with cobalt chloride to a final concentration of 150 µM and cells
were incubated for another 18 hours. Cells were washed with ice-cold PBS. Total RNA
was isolated with an RNeasy kit (Qiagen) according to the manufacturer’s instructions
and quantified by UV absorbance. The RNA was further treated with DNase I (Ambion,
DNAfree kit) to remove any remaining genomic DNA. Reverse transcription was
performed with SuperScript III Reverse Transcriptase (Invitrogen) as recommended by
the manufacturer. Each sample was done in triplicate.
MDA-MB-231 cells were plated in 6-well plates in 2 mL of DMEM media
supplemented with 10% fetal bovine serum at a density of 1.25 × 10
5
cells/mL. Cells
were treated with fresh media containing chetomin or ETP 5 at desired concentrations.
Further steps were carried out as described for MCF7 cells.
A549 cells were grown in low serum F-12K containing 2% fetal bovine serum, penicillin
(50 U/mL) and streptomycin (50 µg/mL) for 1 week. A549 cells were plated in 6-well
plates in 2 mL of serum-free media (0.2% fetal bovine serum) at a density of 1.25 × 10
5
cells/mL. Cells were treated with fresh serum-free media containing ETP-5 at desired
215
concentrations. After 6 h hypoxia was induced by adding desferroxamine mesylate (DFO,
Sigma) to a final concentration of 300 µM. Further steps were carried out as described for
MCF7 cells.
6.4.6 Analysis of gene expression. Real-time qRT-PCR was used to determine the effect
of ETPs on VEGF, c-Met, Glut1, LOX and CXCR4 genes in the MCF7, MDA-MB-231
and A549 cell lines. Compounds were examined under both normoxic and hypoxic
conditions. For VEGF, the forward primer 5’-AGG CCA GCA CAT AGG AGA 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 c-Met gene the following
primer pair were used: 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. For the Glut1
gene the following primers were used: forward 5'-AGT ATG TGG AGC AAC TGT
GTG G-3', reverse 5'-CGG CCT TTA GTC TCA GGA AC-3' to yield a product of 106
bp. For LOX, we employed the following primer pair: 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 were 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. RNA levels were standardized by quantification of the -
glucuronidase as the housekeeping gene with forward primer 5'-CTC ATT TGG AAT
TTT GCC GAT T-3' and reverse primer 5'- CCG AGT GAA GAT CCC CTT TTT A-3'.
The experiments were performed with Applied Biosystems SYBR Green RT-PCR master
mix. Temperature cycling and detection of the SYBR green emission were performed
216
with an ABI 7900HT Fast Real-Time PCR System. Data were analyzed with Applied
Biosystems Sequence Detection System, version 2.4.
6.4.7 Western blot analysis of VEGF and c-Met protein levels. MCF7 and MDA-MB-
231 cells were plated in 60 mm diameter cell culture dishes (BD Falcon) to a density of
1.0 × 10
6
cells/mL. After attachment, they were treated with media containing chetomin
(200 nM), ETP-5 and LS75 (400 nM). All samples contained a final concentration 0.1-
0.2% v/v of DMSO. After a 6 hour incubation period, hypoxia was induced with 300 µM
DFO in MCF7 and with 150 µM CoCl
2
in MDA-MB-231 cells. Samples were incubated
for an additional 18 hours. Total cellular proteins were extracted from the cells using cell
lysis buffer according to manufacturer’s protocol (Cell Signaling). Protein concentrations
were measured with BCA Protein assay kit (Pierce/Thermo Scientific). Equal amounts of
protein samples were subjected to SDS-PAGE and electroblotted to PVDF membrane
(Bio-Rad). These were probed first with an anti-VEGF mouse monoclonal (sc-57496,
Santa Cruz Biotechnology) or anti c-Met rabbit polyclonal antibody (sc-10, Santa Cruz
Biotechnology), stripped with Restore Western Blot Stripping Buffer (Pierce/Thermo
Scientific) and re-probed with a rabbit polyclonal anti- -actin antibody (4867, Cell
Signaling).
After washing with tris-buffered saline – Tween 20 (TBST) solution, the membranes
were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies
(Santa Cruz Biotechnology). Signals were detected by using SuperSignal
chemiluminescent kit (Pierce/Thermo Scientific).
6.4.8 Western blot analysis of HIF-1α levels. A549 cells were plated in T75 cell culture
flasks (BD Falcon) to a density of 2.0 × 10
5
cells/mL. After the cells were 80%
217
confluent, they were treated with media containing ETP-5 (1600 nM). All samples
contained a final concentration 0.1% v/v of DMSO. After a 6 hour incubation period,
hypoxia was induced with 300 µM DFO. Samples were incubated for an additional 42
hours. Cells were washed twice with ice cold PBS buffer. Total cellular proteins were
extracted from the cells using 0.5 mL RIPA cell lysis buffer (Promega) per T75 flask.
Protein concentrations were measured with BCA Protein assay kit (Pierce/Thermo
Scientific). Equal amounts of protein samples were subjected to SDS-PAGE and
electroblotted to PVDF membrane (Bio-Rad). These were probed first with a monoclonal
mouse anti-human HIF-1α antibody (BD Transduction Laboratories) stripped with
Restore Western Blot Stripping Buffer (Pierce/Thermo Scientific) and re-probed with a
goat polyclonal anti-Lamin A/C antibody (sc-6215, Santa Cruz Biotechnology).
After washing with tris-buffered saline – Tween 20 (TBST) solution, 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).
6.4.9 Animal use. Animal experiments were done in accordance with federal guidelines
following review and approval by the PRISM institutional animal care and use
committees (IACUC). Athymic nude mice were purchased from Harlan at the age of 8–9
weeks.
6.4.10 Fluorescent tumor cell lines. N202 (gift from Joseph Lustgarten, Mayo Clinic,
Scottsdale, AZ) were maintained in DMEM High Glucose supplemented with L-
Glutamine (2mM), Penicillin (100 U/ml), Streptomycin (100 U/ml), Sodium Pyruvate (1
mM) (Invitrogen, Carlsbad, CA) and 10% heat inactivated FBS (Omega Scientific,
218
Tarzana, CA) at 37
o
C in 5% CO
2
in air. The histone H2B-GFP was subcloned into the
SalI/HpaI sites in the LXRN vector (Clontech, Palo Alto, CA) using SalI and blunted
NotI sites from the BOSH2BGFPN1 vector.(Kanda, Sullivan et al. 1998) N202 were
transduced with the viable virus to stably incorporate the H2B-GFP gene. The transduced
cells were FACs sorted twice to ensure 100% of the cells stably expressed the H2B-GFP
protein.
6.4.11 Mouse xenograft tumor models. Classic IVM tumor model (Frost, Lustgarten et
al. 2005) was used with minor modifications. The mice, usually athymic nude mice (25-
30 g body weight), were anesthetized (7.3 mg ketamine hydrochloride and 2.3 mg
xylazine per 100 g body weight, intraperitoneal injection) and placed on a heating pad. A
titanium frame was placed onto the dorsal skinfold of mice to sandwich the extended
double layer of skin. A 15mm diameter full-thickness circular layer of skin was then
excised. The superficial fascia on top of the remaining skin is carefully removed to
expose the underlying muscle and subcutaneous tissue which is then covered with
another titanium frame with a glass cover slip to form the window chamber. After a
recovery period of 1-2 days, tumor spheroids were implanted.
Tumor spheroids were formed by plating 50,000 N202 cells onto 1% agar-coated 96-well
non-tissue culture treated flat bottom dishes (20 µl cells in 100 µl medium) and
centrifuging 4 times at 2000 rpm for 15 min, rotating the dish after every centrifugation.
The cells were incubated an additional 3-7 days (depending on cell type) at 37
o
C in 5%
CO
2
in air to form tight spheroids.
219
The tumor spheroids were implanted directly onto the dorsal skin in the window chamber
alone. Tumors were allowed to vascularize over 10-14 days before the injection of 1
mg/kg of ETP-5 compound at Day 0, followed by the daily administration at 2 mg/kg at
Days 8-13.
6.4.12 Tumor Growth. Analysis of tumor growth with IVM. Tumors were imaged by
intravital fluorescence microscopy, as described (Oh, Borgstrom et al. 2007) . Tumor
growth was analyzed off-line from the recorded, digital, grayscale 0-to-256 images using
Image-Pro Plus (Media Cybernetics, Bethesda, MD). Tumor growth was determined by
quantifying the cumulative fluorescence signal for the tumor over time. The cumulative
tumor fluorescence signal was measured by signal summation of all pixels over 75. All
growth curves are normalized to the tumor on day 0 after treatment.
6.5 Experimental Section for Chapter 5
6.5.1 Absorbance spectrum for DIL-1, DILB-2 and DB-3. Beckmann spectrophotometer
was used to measure absorbance spectrum for DIL-1. DIL-1 was dissolved in solvent
mixture having 95% water and 5% methanol at concentration of 100 µM. Incident light
intensity of adjusted to obtain maximum absorbance below 1. Similarly, absorbance
spectrum for DIL-1 (100 µM) was also taken in viscous solvent mixture having 90%
glycerol, 5% water and 5% methanol.
For absorbance spectra of DILB-2 and DB-3 both the compounds were dissolved in the
solvent mixture of 95% water and 5% methanol at concentration of 100 µM. All the
spectra were measured from 200 nm to 800 nm wavelength range.
220
6.5.2 Fluorescence spectra for DIL-1, DILB-2 and DB-3. Instrument from Photon
Technology International - Model 814 was used to measure fluorescence spectra for the
three compounds. Compounds were dissolved in aqueous media mixture of 95% water
and 5% methanol and viscous media of 90% glycerol, 5% water and 5% methanol. The
concentration for all the three compounds was 100 µM. For emission spectra the
excitation wavelength of 435 nm was chosen and fluorescence emission intensity was
measured from 450 nm to 550 nm at 90 ° angle from the incident (excitation) light.
Measurement was made was made at each wavelength from 450-550 nm. For excitation
spectra the incident light was varied from 375 nm to 520 nm and the emission was
measured at 540 nm at 90 ° angle from the incident (excitation) light. In this case also the
incident (excitation) light was measured at each wavelength from 375-520 nm.
6.5.3 Confocal Microscopy with DIL-1, DILB-2 and DB-3. For confocal microscopy
Mat Tek 35 mm diameter dishes with 14 mm glass bottom were used. A general method
for confocal microscopy consisted of plating 30,000 cells (MCF7 or WM115) in glass
bottom dish with 300 µL medium. After 24 h the dishes with cells were treated with
compounds DIL-1, DILB-2 and DB-3 at desired concentration and keeping DMSO final
concentration ≤ 1%. The dishes with cells and compounds are incubated for desired time
at 37 °C in 5% CO
2
. In general, for DIL-1 2 h incubation was given and for DILB-2 and
DB-3 14 h incubation was given. For, 4 °C experiments 2 h incubation was given in
plastic wrap to avoid evaporation of media in fridge.
6.5.4 Intravital microscopy with DILB-2 and DB-3. The experiments were conducted by
collaborators at Schnitzer Lab, PRISM, San Diego. Two murine tumor models viz.
subcutaneous and ectopic orthotopic models. Subcutaneous model is achieved by
221
implanting extraneous tumor cells (N2O2) in the subcutaneous region while ectopic
orthotopic model is achieved by first implanting a tissue of a specific organ on the dorsal
side of the animal (ectopic) after which tumor cells belonging to that organ are implanted
in the tissue (orthotopic). In both the animal models the compound DIL-1 and DB-3 were
injected intravenously via the tail vein at concentration of 1 mg/kg. The injection had the
desired compound dissolved in saline and 1% DMSO.
In both the cases, a window was implanted in the dorsal side of the animal through which
intravital microscopy could be done daily in order to see the distribution of the
compounds in the tumor tissue. The fluorescence images were taken at wavelength
determined by the excitation and emission wavelength of the cyan 40 dye of the
conjugates.
222
Bibliography
Affymetrix (2012). from http://www.affymetrix.com/estore/browse/products
Al-Madhoun, A. S., J. Johnsamuel, et al. (2004). "Evaluation of human thymidine kinase
1 substrates as new candidates for boron neutron capture therapy." Cancer Research
64(17): 6280-6286.
Alexander, R. P. and I. Paterson (1985). "Asymmetric alkylation and sulfenylation of
chiral o-silylated imide enolates." Tetrahedron Lett. 26(43): 5339-5340.
Aplin, A. E., A. Howe, et al. (1998). "Signal transduction and signal modulation by cell
adhesion receptors: The role of integrins, cadherins, immunoglobulin-cell adhesion
molecules, and selectins." Pharmacol. Rev. 50(2): 197-263.
Bader, B. L., H. Rayburn, et al. (1998). "Extensive vasculogenesis, angiogenesis, and
organogenesis precede lethality in mice lacking all alpha v integrins." Cell 95(4): 507-
519.
Barnes, W. L., A. Dereux, et al. (2003). "Surface plasmon subwavelength optics." Nature
424(6950): 824-830.
Barrow, C. J., P. Cai, et al. (1993). "Win 64821, a new competitive antagonist to
substance-p, isolated from an aspergillus species - structure determination and solution
conformation." J. Org. Chem. 58(22): 6016-6021.
Barth, R. F., J. A. Coderre, et al. (2005). "Boron neutron capture therapy of cancer:
Current status and future prospects." Clin. Cancer Res. 11(11): 3987-4002.
Barth, R. F., W. L. Yang, et al. (2004). "Boron-containing nucleosides as potential
delivery agents for neutron capture therapy of brain tumors." Cancer Research 64(17):
6287-6295.
Behrooz, A. and F. IsmailBeigi (1997). "Dual control of glut1 glucose transporter gene
expression by hypoxia and by inhibition of oxidative phosphorylation." J. Biol. Chem.
272(9): 5555-5562.
223
Bendel, P. (2005). "Biomedical applications of b-10 and b-11 nmr." NMR Biomed.
18(2): 74-82.
Bernardo, P. H., N. Brasch, et al. (2003). "A novel redox mechanism for the glutathione-
dependent reversible uptake of a fungal toxin in cells." J. Biol. Chem. 278(47): 46549-
46555.
Block, K. M., H. Wang, et al. (2009). "Direct inhibition of hypoxia-inducible
transcription factor complex with designed dimeric epidithiodiketopiperazine." J. Am.
Chem. Soc. 131(50): 18078-18088.
Boccaccio, C., M. Ando, et al. (1998). "Induction of epithelial tubules by growth factor
hgf depends on the stat pathway." Nature 391(6664): 285-288.
Boccaccio, C. and P. M. Comoglio (2006). "Opinion - invasive growth: A met-driven
genetic programme for cancer and stem cells." Nature Reviews Cancer 6(8): 637-645.
Boehm, T., J. Folkman, et al. (1997). "Antiangiogenic therapy of experimental cancer
does not induce acquired drug resistance." Nature 390(6658): 404-407.
Bradshaw, K. M., M. P. Schweizer, et al. (1995). "Bsh distributions in the canine head
and a human patient using b-11 mri (vol 34, pg 48, 1995)." Magn.Reson.Med. 34(4):
645-645.
Brasier, A. R., J. E. Tate, et al. (1989). "Optimized use of the firefly luciferase assay as a
reporter gene in mammalian cell lines." BioTechniques 7(10): 1116-1122.
Brewer, D., Archibal.Rm, et al. (1972). "Ovine ill-thrift in nova-scotia .5. Production and
toxicology of chetomin, a metabolite of chaetomium spp." Can. J. Microbiol. 18(7):
1129-&.
Brooks, P. C., R. A. F. Clark, et al. (1994). "Requirement of vascular integrin
alpha(v)beta(3) for angiogenesis." Science 264(5158): 569-571.
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.
224
Brown, J. M. (1993). "Sr-4233 (tirapazamine) - a new anticancer drug exploiting hypoxia
in solid tumors." Br. J. Cancer 67(6): 1163-1170.
Brown, J. M. and M. J. Lemmon (1990). "Potentiation by the hypoxic cytotoxin sr4233
of cell killing produced by fractionated-irradiation of mouse-tumors." Cancer Research
50(24): 7745-7749.
Brown, J. M. and W. R. Wilson (2004). "Exploiting tumour hypoxia in cancer treatment."
Nature Reviews Cancer 4(6): 437-447.
Cai, W. and X. Chen (2006). "Anti-angiogenic cancer therapy based on integrin
alpha(v)beta(3) antagonism." Anti-Cancer Agents in Medicinal Chemistry 6(5): 407-428.
Carlson, C. B., P. Mowery, et al. (2007). "Selective tumor cell targeting using low-
affinity, multivalent interactions." ACS Chem. Biol. 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.
Carron, C. P., D. M. Meyer, et al. (2000). "Peptidomimetic antagonists of alpha v beta 3
inhibit bone resorption by inhibiting osteoclast bone resorptive activity, not osteoclast
adhesion to bone." J. Endocrinol. 165(3): 587-598.
Cary, L. A., D. C. Han, et al. (1999). "Integrin-mediated signal transduction pathways."
Histol. Histopath. 14(3): 1001-1009.
Ceradini, D. J., A. R. Kulkarni, et al. (2004). "Progenitor cell trafficking is regulated by
hypoxic gradients through hif-1 induction of sdf-1." Nat. Med. 10(8): 858-864.
Chan, D. A., P. D. Sutphin, et al. (2011). "Targeting glut1 and the warburg effect in renal
cell carcinoma by chemical synthetic lethality." Sci. Transl. Med. 3(94).
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." J. Biol. Chem. 284(39): 26831-26838.
225
Cushing, T. D., J. F. Sanzcervera, et al. (1993). "Stereocontrolled total synthesis of (+)-
paraherquamide-b." J. Am. Chem. Soc. 115(20): 9323-9324.
Cushing, T. D., J. F. SanzCervera, et al. (1996). "Stereocontrolled total synthesis of (+)-
paraherquamide b." J. Am. Chem. Soc. 118(3): 557-579.
Darash-Yahana, M., E. Pikarsky, et al. (2004). "Role of high expression levels of cxcr4 in
tumor growth, vascularization, and metastasis." Faseb J. 18(9): 1240-+.
De Guzman, R. N., J. M. Wojciak, et al. (2005). "Cbp/p300 taz1 domain forms a
structured scaffold for ligand binding." Biochemistry 44(2): 490-497.
de Kok, J. B., R. W. Roelofs, et al. (2004). "Normalization of gene expression
measurements in tumor tissues: Comparison of 13 endogenous control genes." Lab Invest
85(1): 154-159.
Denizot, F. and R. Lang (1986). "Rapid colorimetric assay for cell growth and survival:
Modifications to the tetrazolium dye procedure giving improved sensitivity and
reliability." Journal of Immunological Methods 89(2): 271-277.
Denko, N. C., L. A. Fontana, et al. (2003). "Investigating hypoxic tumor physiology
through gene expression patterns." Oncogene 22(37): 5907-5914.
Di Menna, M. E., B. L. Smith, et al. (2009). "A history of facial eczema
(pithomycotoxicosis) research." N. Z. J. Agric. Res. 52(4): 345-376.
Dial, R., Z. Y. J. Sun, et al. (2003). "Three conformational states of the p300 ch1 domain
define its functional properties." Biochemistry 42(33): 9937-9945.
Dittmer, J. (2003). "The biology of the ets1 proto-oncogene." Molecular Cancer 2(1): 29.
Dontu, G., M. Al-Hajj, et al. (2003). "Stem cells in normal breast development and breast
cancer." Cell Prolif. 36: 59-72.
Dove, J. E. and J. P. Klinman (2001). "Trihydroxyphenylalanine quinone (tpq) from
copper amine oxidases and lysyl tyrosylquinone (ltq) from lysyl oxidase." Adv.Protein
Chem. 58: 141-174.
226
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." J. Med. Chem. 43(20): 3736-3745.
Duggan, M. E., J. E. Fisher, et al. (1996). "Design and evaluation of potent nonpeptide
ligands of alpha-v-beta-3 as inhibitors of bone resorption." Abstracts of Papers American
Chemical Society 211(1-2): 234.
Durand, R. E. (1994). "The influence of microenvironmental factors during cancer
therapy." In Vivo (Attiki) 8(5): 691-702.
Dvorak, H. F., L. F. Brown, et al. (1995). "Vascular-permeability factor vascular
endothelial growth-factor, microvascular hyperpermeability, and angiogenesis." Am. J.
Pathol. 146(5): 1029-1039.
Eble, J. A. and J. Haier (2006). "Integrins in cancer treatment." Curr. Cancer Drug
Targets 6(2): 89-105.
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." J.
Clin. Invest. 103(9): 1227-1230.
Erler, J. T., K. L. Bennewith, et al. (2006). "Lysyl oxidase is essential for hypoxia-
induced metastasis." Nature 440(7088): 1222-1226.
Faillot, T., H. Magdelenat, et al. (1996). "A phase i study of an anti-epidermal growth
factor receptor monoclonal antibody for the treatment of malignant gliomas."
Neurosurgery 39(3): 478-483.
Ferrara, N., K. CarverMoore, et al. (1996). "Heterozygous embryonic lethality induced
by targeted inactivation of the vegf gene." Nature 380(6573): 439-442.
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. (2002). "Structural basis for recruitment of cbp/p300
by hypoxia-inducible factor-1 alpha." Proc. Natl. Acad. Sci. U. S. A. 99(8): 5367-5372.
227
Frost, G. I., J. Lustgarten, et al. (2005). "Novel syngeneic pseudo-orthotopic prostate
cancer model: Vascular, mitotic and apoptotic responses to castration." Microvasc Res
69(1-2): 1-9.
Gambarotta, G., S. Pistoi, et al. (1994). "Structure and inducible regulation of the human
met promoter." J. Biol. Chem. 269(17): 12852-12857.
Gangadhar, T., S. Nandi, et al. (2010). "The role of chemokine receptor cxcr4 in lung
cancer." Cancer Biol. Ther. 9(6).
Gardiner, D. M., P. Waring, et al. (2005). "The epipolythiodioxopiperazine (etp) class of
fungal toxins: Distribution, mode of action, functions and biosynthesis." Microbiology-
(UK) 151: 1021-1032.
Gerber, H. P., F. Condorelli, et al. (1997). "Differential transcriptional regulation of the
two vascular endothelial growth factor receptor genes - flt-1, but not flk-1/kdr, is up-
regulated by hypoxia." J. Biol. Chem. 272(38): 23659-23667.
Giaccia, A., B. G. Siim, et al. (2003). "Hif-1 as a target for drug development." Nat. Rev.
Drug Discov. 2(10): 803-811.
Gottschalk, K. E. and H. Kessler (2002). "The structures of integrins and integrin-ligand
complexes: Implications for drug design and signal transduction." Angew. Chem.-Int.
Edit. 41(20): 3767-3774.
Graeber, T. G., C. Osmanian, et al. (1996). "Hypoxia-mediated selection of cells with
diminished apoptotic potential in solid tumours." Nature 379(6560): 88-91.
Gual, P., S. Giordano, et al. (2000). "Sustained recruitment of phospholipase c-gamma to
gab1 is required for hgf-induced branching tubulogenesis." Oncogene 19(12): 1509-1518.
Guo, M., L. P. Song, et al. (2006). "Hypoxia-mimetic agents desferrioxamine and cobalt
chloride induce leukemic cell apoptosis through different hypoxia-inducible factor-1α
independent mechanisms." Apoptosis 11(1): 67-77.
Guo, W. J. and F. G. Giancotti (2004). "Integrin signalling during tumour progression."
Nat. Rev. Mol. Cell Biol. 5(10): 816-826.
228
Gutheil, J. C., T. N. Campbell, et al. (2000). "Targeted antiangiogenic therapy for cancer
using vitaxin: A humanized monoclonal antibody to the integrin alpha(nu)beta(3)." Clin.
Cancer Res. 6(8): 3056-3061.
Hak, S., N. K. Reitan, et al. (2010). "Intravital microscopy in window chambers: A
unique tool to study tumor angiogenesis and delivery of nanoparticles." Angiogenesis
13(2): 113-130.
Hanahan, D. and R. A. Weinberg (2000). "The hallmarks of cancer." Cell 100(1): 57-70.
Harling, O. K. and K. J. Riley (2003). "Fission reactor neutron sources for neutron
capture therapy - a critical review." J. Neuro-Oncol. 62(1): 7-17.
Harris, A. L. (2002). "Hypoxia - a key regulatory factor in tumour growth." Nature
Reviews Cancer 2(1): 38-47.
Hauser, D., H. P. Weber, et al. (1970). "Isolation and structure elucidation of chaetocin."
Helv. Chim. Acta 53(5): 1061-&.
Hayden, E. C. (2009). "Cutting off cancer's supply lines." Nature 458(7239): 686-687.
Henchey, L. K., S. Kushal, et al. (2012). "Inhibition of hypoxia inducible factor 1-
transcription coactivator interaction by a hydrogen bond surrogate alpha-helix (vol 132,
pg 941, 2010)." J. Am. Chem. Soc. 134(18): 8000-8000.
Hendrix, M. J. C., E. A. Seftor, et al. (2003). "Vasculogenic mimicry and tumour-cell
plasticity: Lessons from melanoma." Nat Rev Cancer 3(6): 411-421.
Hermanek, S. (1992). "B-11 nmr-spectra of boranes, main-group heteroboranes, and
substituted derivatives - factors influencing chemical-shifts of skeletal atoms." Chem.
Rev. 92(2): 325-362.
Hino, T. and T. Sato (1971). "Synthesis of 3,6-diethoxycarbonyl3,6-epipolythia-2,5-
piperazinedione derviatives." Tetrahedron Lett.(33): 3127-&.
Hintermann, L. and A. Togni (2000). "Catalytic enantioselective chlorination and
bromination of β-keto esters." Helv. Chim. Acta 83(9): 2425-2435.
229
Hiramoto, M. (1994). Tennen Yuki Kagobutsu Toronkai Koen Yoshishu 36: 557.
Hodivala-Dilke, K. (2008). "Alpha v beta 3 integrin and angiogenesis: A moody integrin
in a changing environment." Curr. Opin. Cell Biol. 20(5): 514-519.
Hood, J. D., M. Bednarski, et al. (2002). "Tumor regression by targeted gene delivery to
the neovasculature." Science 296(5577): 2404-2407.
Huang, L. E., Z. Arany, et al. (1996). "Activation of hypoxia-inducible transcription
factor depends primarily upon redox-sensitive stabilization of its alpha subunit." J. Biol.
Chem. 271(50): 32253-32259.
Huang, L. E. and H. F. Bunn (2003). "Hypoxia-inducible factor and its biomedical
relevance." J. Biol. Chem. 278(22): 19575-19578.
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." J. Biol. Chem. 277(35): 31631-31638.
Hynes, R. O. (2002). "Integrins: Bidirectional, allosteric signaling machines." Cell
110(6): 673-687.
Ishigaki, T., K. Imanaka-Yoshida, et al. (2011). "Tenascin-c enhances crosstalk signaling
of integrin αvβ3/pdgfr-β complex by src recruitment promoting pdgf-induced
proliferation and migration in smooth muscle cells." Journal of Cellular Physiology
226(10): 2617-2624.
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.
Iwasa, E., Y. Hamashima, et al. (2010). "Total synthesis of (+)-chaetocin and its
analogues: Their histone methyltransferase g9a inhibitory activity." J. Am. Chem. Soc.
132(12): 4078-+.
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.
230
Jain, R. K., D. G. Duda, et al. (2006). "Lessons from phase iii clinical trials on anti-vegf
therapy for cancer." Nat. Clin. Pract. Oncol. 3(1): 24-40.
Jameson, D. M. and J. A. Ross (2010). "Fluorescence polarization/anisotropy in
diagnostics and imaging." Chem. Rev. 110(5): 2685-2708.
Jolley, M. E., S. D. Stroupe, et al. (1981). "Fluorescence polarization immunoassay. Iii.
An automated system for therapeutic drug determination." Clinical Chemistry 27(9):
1575-1579.
Joshi, B. K., J. B. Gloer, et al. (1999). "New verticillin and glisoprenin analogues from
gliocladium catenulatum, a mycoparasite of aspergillus flavus sclerotia." J. Nat. Prod.
62(5): 730-733.
Kametani, T., N. Kanaya, et al. (1980). "Asymmetric total synthesis of brevianamide-e."
J. Am. Chem. Soc. 102(11): 3972-3975.
Kanda, T., K. F. Sullivan, et al. (1998). "Histone-gfp fusion protein enables sensitive
analysis of chromosome dynamics in living mammalian cells." Curr Biol 8(7): 377-385.
Kerbel, R. and J. Folkman (2002). "Clinical translation of angiogenesis inhibitors."
Nature Reviews Cancer 2(10): 727-739.
Kijowski, J., M. Baj-Krzyworzeka, et al. (2001). "The sdf-1-cxcr4 axis stimulates vegf
secretion and activates integrins but does not affect proliferation and survival in
lymphohematopoietic cells." Stem Cells 19(5): 453-466.
Kim, J., J. A. Ashenhurst, et al. (2009). "Total synthesis of (+)-11,11 '-dideoxyverticillin
a." Science 324(5924): 238-241.
Kirschmann, D. A., E. A. Seftor, et al. (2002). "A molecular role for lysyl oxidase in
breast cancer invasion." Cancer Research 62(15): 4478-4483.
Kirschmann, D. A., E. A. Seftor, et al. (1999). "Differentially expressed genes associated
with the metastatic phenotype in breast cancer." Breast Cancer Res. Treat. 55(2): 127-
136.
231
Kishi, Y., Nakatsuk.S, et al. (1973). "Total synthesis of sporidesmin-a." J. Am. Chem.
Soc. 95(19): 6493-6495.
Koehl, G. E., A. Gaumann, et al. (2009). "Intravital microscopy of tumor angiogenesis
and regression in the dorsal skin fold chamber: Mechanistic insights and preclinical
testing of therapeutic strategies." Clin. Exp. Metastasis 26(4): 329-344.
Koivunen, E., B. C. Wang, et al. (1995). "Phage libraries displaying cyclic-peptides with
different ring sizes - ligand specificities of the rgd-directed integrins." Bio-Technology
13(3): 265-270.
Kriegova, E., A. Arakelyan, et al. (2008). "Psmb2 and rpl32 are suitable denominators to
normalize gene expression profiles in bronchoalveolar cells." BMC Molecular Biology
9(1): 69.
Kubo, H. and K. Alitalo (2003). "The bloody fate of endothelial stem cells." Genes Dev.
17(3): 322-329.
Kucia, M., R. Reca, et al. (2005). "Trafficking of normal stem cells and metastasis of
cancer stem cells involve similar mechanisms: Pivotal role of the sdf-1-cxcr4 axis." Stem
Cells 23(7): 879-894.
Kumar, C. C. (2003). "Integrin alpha v beta 3 as a therapeutic target for blocking tumor-
induced angiogenesis." Curr. Drug Targets 4(2): 123-131.
Kung, A. L., S. Wang, et al. (2000). "Suppression of tumor growth through disruption of
hypoxia-inducible transcription." Nat. Med. 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.
Kurata, T., T. Oguri, et al. (1999). "Differential expression of facilitative glucose
transporter (glut) genes in primary lung cancers and their liver metastases." Jpn. J. Cancer
Res. 90(11): 1238-1243.
Lacy-Hulbert, A., A. M. Smith, et al. (2007). "Ulcerative colitis and autoirnmunity
induced by loss of myeloid alpha v integrins." Proc. Natl. Acad. Sci. U. S. A. 104(40):
15823-15828.
232
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 Dev. 16(12):
1466-1471.
Liedberg, B., C. Nylander, et al. (1983). "Surface-plasmon resonance for gas-detection
and biosensing." Sensors and Actuators 4(2): 299-304.
Lin, A. J., L. Fang, et al. (2011). "Organocatalytic enantioselective sulfenylation of beta-
keto phosphonates: A convenient approach to construct hetero-quaternary stereocenters."
Adv. Synth. Catal. 353(4): 545-549.
Liu, F. Y., Q. 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, Y. V., J. H. Baek, et al. (2007). "Rack1 competes with hsp90 for binding to hif-1
alpha and is required for o-2-independent and hsp90 inhibitor-induced degradation of hif-
1 alpha." Mol. Cell 25(2): 207-217.
Liu, Z. F., F. Wang, et al. (2008). "Integrin alpha(v)beta(3)-targeted cancer therapy."
Drug Dev. Res. 69(6): 329-339.
Lohela, M. and Z. Werb (2010). "Intravital imaging of stromal cell dynamics in tumors."
Curr. Opin. Genet. Dev. 20(1): 72-78.
Mabjeesh, N. J., D. Escuin, et al. (2003). "2me2 inhibits tumor growth and angiogenesis
by disrupting microtubules and dysregulating hif." Cancer Cell 3(4): 363-375.
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.
Mahabeleshwar, G. H., W. Y. Feng, et al. (2006). "Integrin signaling is critical for
pathological angiogenesis." J. Exp. Med. 203(11): 2495-2507.
Mahabeleshwar, G. H., W. Y. Feng, et al. (2007). "Mechanisms of integrin-vascular
endothelial growth factor receptor cross-activation in angiogenesis." Circ.Res. 101(6):
570-580.
233
Malmqvist, M. (1999). "Biacore: An affinity biosensor system for characterization of
biomolecular interactions." Biochem. Soc. Trans. 27(2): 335-340.
Marigo, M., T. C. Wabnitz, et al. (2005). "Enantioselective organocatalyzed a
sulfenylation of aldehydes." Angew. Chem.-Int. Edit. 44(5): 794-797.
Marley, J., M. Lu, et al. (2001). "A method for efficient isotopic labeling of recombinant
proteins." J. Biomol. NMR 20(1): 71-75.
Maroun, C. R., M. A. Naujokas, et al. (2000). "The tyrosine phosphatase shp-2 is
required for sustained activation of extracellular signal-regulated kinase and epithelial
morphogenesis downstream from the met receptor tyrosine kinase." Mol. Cell. Biol.
20(22): 8513-8525.
Masson, N., C. Willam, et al. (2001). "Independent function of two destruction domains
in hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation." Embo J.
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." Proc.
Natl. Acad. Sci. U. S. A. 94(15): 8104-8109.
Mellanen, P., H. Minn, et al. (1994). "Expression of glucose transporters in head-and-
neck tumors." Int. J. Cancer 56(5): 622-629.
Michieli, P., C. Basilico, et al. (1999). "Mutant met-mediated transformation is ligand-
dependent and can be inhibited by hgf antagonists." Oncogene 18(37): 5221-5231.
Mori, T., R. Doi, et al. (2004). "Cxcr4 antagonist inhibits stromal cell-derived factor 1-
induced migration and invasion of human pancreatic cancer." Mol. Cancer Ther. 3(1): 29-
37.
Muller, A., B. Homey, et al. (2001). "Involvement of chemokine receptors in breast
cancer metastasis." Nature 410(6824): 50-56.
Nabors, L. B., T. Mikkelsen, et al. (2007). "Phase i and correlative biology study of
cilengitide in patients with recurrent malignant glioma." J. Clin. Oncol. 25(13): 1651-
1657.
234
Nicolaou, K. C., J. I. Trujillo, et al. (1998). "Design, synthesis and biological evaluation
of nonpeptide integrin antagonists." Bioorg. Med. Chem. 6(8): 1185-1208.
Ogryzko, V. V., R. L. Schiltz, et al. (1996). "The transcriptional coactivators p300 and
cbp are histone acetyltransferases." Cell 87(5): 953-959.
Oh, P., P. Borgstrom, et al. (2007). "Live dynamic imaging of caveolae pumping targeted
antibody rapidly and specifically across endothelium in the lung." Nat Biotechnol 25(3):
327-337.
Ohulchanskyy, T. Y., H. E. Pudavar, et al. (2003). "A monomethine cyanine dye cyan 40
for two-photon–excited fluorescence detection of nucleic acids and their visualization in
live cells¶." Photochemistry and Photobiology 77(2): 138-145.
Oikawa, M., M. Abe, et al. (2001). "Hypoxia induces transcription factor ets-1 via the
activity of hypoxia-inducible factor-1." Biochem. Biophys. Res. Commun. 289(1): 39-43.
Olsson, A. K., A. Dimberg, et al. (2006). "Vegf receptor signalling - in control of
vascular function." Nat. Rev. Mol. Cell Biol. 7(5): 359-371.
Papadopoulou, M. V. and W. D. Bloomer (2003). "Nlcq-1 (nsc 709257): Exploiting
hypoxia with a weak DNA-intercalating bioreductive drug." Clin. Cancer Res. 9(15):
5714-5720.
Parsons, J. T. and S. J. Parsons (1997). "Src family protein tyrosine kinases: Cooperating
with growth factor and adhesion signaling pathways." Curr. Opin. Cell Biol. 9(2): 187-
192.
Patterson, L. H. (1993). "Rationale for the use of aliphatic n-oxides of cytotoxic
anthraquinones as prodrug DNA-binding agents - a new class of bioreductive agent."
Cancer Metastasis Rev. 12(2): 119-134.
Patterson, L. H., S. R. McKeown, et al. (2000). "Enhancement of chemotherapy and
radiotherapy of murine tumours by aq4n, a bioreductively activated anti-tumour agent."
Br. J. Cancer 82(12): 1984-1990.
Peehl, D. M., M. Coram, et al. (2008). "The significance of monoamine oxidase-a
expression in high grade prostate cancer." The Journal of Urology 180(5): 2206-2211.
235
Pelicci, G., S. Giordano, et al. (1995). "The motogenic and mitogenic responses to hgf are
amplified by the shc adapter protein." Oncogene 10(8): 1631-1638.
Pennacchietti, S., P. Michieli, et al. (2003). "Hypoxia promotes invasive growth by
transcriptional activation of the met protooncogene." Cancer Cell 3(4): 347-361.
Peters, K. B. and J. M. Brown (2002). "Tirapazamine: A hypoxia-activated
topoisomerase ii poison." Cancer Research 62(18): 5248-5253.
Ponzetto, C., A. Bardelli, et al. (1994). "A multifunctional docking site mediates
signaling and transformation by the hepatocyte growth-factor scatter factor-receptor
family." Cell 77(2): 261-271.
Popp, J. L., L. L. Musza, et al. (1994). "Win-64821, a novel neurokinin antagonist
produced by an aspergillus sp .3. Biosynthetic analogs (vol 47, pg 411, 1994)." J.
Antibiot. 47(8): C5-C5.
Pradervand, S., A. Paillusson, et al. (2008). "Affymetrix whole-transcript human gene 1.0
st array is highly concordant with standard 3 ' expression arrays." BioTechniques 44(6):
759-762.
Pugh, C. W. and P. J. Ratcliffe (2003). "Regulation of angiogenesis by hypoxia: Role of
the hif system." Nat. Med. 9(6): 677-684.
Rahimi, N., V. Dayanir, et al. (2000). "Receptor chimeras indicate that the vascular
endothelial growth factor receptor-1 (vegfr-1) modulates mitogenic activity of vegfr-2 in
endothelial cells." J. Biol. Chem. 275(22): 16986-16992.
Reynolds, A. R., L. E. Reynolds, et al. (2004). "Elevated flk1 (vascular endothelial
growth factor receptor 2) signaling mediates enhanced angiogenesis in beta(3)-integrin-
deficient mice." Cancer Research 64(23): 8643-8650.
Reynolds, L. E., L. Wyder, et al. (2002). "Enhanced pathological angiogenesis in mice
lacking beta(3) integrin or beta(3) and beta(5) integrins." Nat. Med. 8(1): 27-34.
Riley, K. J., P. J. Binns, et al. (2003). "Performance characteristics of the mit fission
converter based epithermal neutron beam." Physics in Medicine and Biology 48(7): 943.
236
Ruoslahti, E. (2002). "Specialization of tumour vasculature." Nature reviews. Cancer
2(2): 83-90.
Rybaczyk, L., M. Bashaw, et al. (2008). "An indicator of cancer: Downregulation of
monoamine oxidase-a in multiple organs and species." BMC Genomics 9(1): 134.
Salceda, S. and J. Caro (1997). "Hypoxia-inducible factor 1 alpha (hif-1 alpha) protein is
rapidly degraded by the ubiquitin-proteasome system under normoxic conditions - its
stabilization by hypoxia depends on redox-induced changes." J. Biol. Chem. 272(36):
22642-22647.
Salnikow, K., O. Aprelikova, et al. (2008). "Regulation of hypoxia-inducible genes by
ets1 transcription factor." Carcinogenesis 29(8): 1493-1499.
Schaeper, U., N. H. Gehring, et al. (2000). "Coupling of gab1 to c-met, grb2, and shp2
mediates biological responses." J. Cell Biol. 149(7): 1419-1432.
Schatteman, G. C. and O. Awad (2004). "Hemangioblasts, angioblasts, and adult
endothelial cell progenitors." Anat. Rec. Part A 276A(1): 13-21.
Schioppa, T., B. Uranchimeg, et al. (2003). "Regulation of the chemokine receptor cxcr4
by hypoxia." J. Exp. Med. 198(9): 1391-1402.
Schlaepfer, D. D. and T. Hunter (1998). "Integrin signalling and tyrosine
phosphorylation: Just the faks?" Trends Cell Biol. 8(4): 151-157.
Schnitzer, J. E., J. Liu, et al. (1995). "Endothelial caveolae have the molecular transport
machinery for vesicle budding, docking, and fusion including vamp, nsf, snap, annexins,
and gtpases." J. Biol. Chem. 270(24): 14399-14404.
Sekita, S., K. Yoshihira, et al. (1981). "Myco-toxin production by chaetomium spp and
related fungi." Can. J. Microbiol. 27(8): 766-772.
Shaikh, R. R., A. Mazzanti, et al. (2008). "Proline-catalyzed asymmetric formal alpha-
alkylation of aldehydes via vinylogous iminium ion intermediates generated from
arylsulfonyl indoles." Angew. Chem.-Int. Edit. 47(45): 8707-8710.
237
Shibamoto, S., M. Hayakawa, et al. (1994). "Tyrosine phosphorylation of beta-catenin
and plakoglobin enhanced by hepatocyte growth-factor and epidermal growth-factor in
human carcinoma-cells." Cell Adhes. Commun. 1(4): 295-305.
Siim, B. G., W. A. Denny, et al. (1997). "Nitro reduction as an electronic switch for
bioreductive drug activation." Oncol. Res. 9(6-7): 357-369.
Smith-Mungo, L. I. and H. M. Kagan (1998). "Lysyl oxidase: Properties, regulation and
multiple functions in biology." Matrix Biol. 16(7): 387-398.
Sobhani, S., D. Fielenbach, et al. (2005). "Direct organocatalytic asymmetric alpha-
sulfenylation of activated c-h bonds in lactones, lactams, and beta-dicarbonyl
compounds." Chem.-Eur. J. 11(19): 5689-5694.
Somanath, P., N. Malinin, et al. (2009). "Cooperation between integrin
α<sub>ν</sub>β<sub>3</sub> and vegfr2 in angiogenesis."
Angiogenesis 12(2): 177-185.
Somei, M., Y. Karasawa, et al. (1981). "The chemistry of indoles .15. Selective mono-
alkylation of carbon nucleophiles with gramine." Heterocycles 16(6): 941-949.
Son, B. W., P. R. Jensen, et al. (1999). "New cytotoxic epidithiodioxopiperazines related
to verticillin a from a marine isolate of the fungus penicillium." Nat. Prod. Lett. 13(3):
213-222.
Springer, J. P., G. Buchi, et al. (1977). "Structure of ditryptophenaline - new metabolite
of aspergillus-flavus." Tetrahedron Lett.(28): 2403-2406.
Srisailam, S. K. and A. Togni (2006). "Phthalimide-n-sulfenyl chloride in the ti-catalyzed
asymmetric sulfenylation of beta-keto esters." Tetrahedron: Asymmetry 17(18): 2603-
2607.
Stromblad, S. and D. A. Cheresh (1996). "Integrins, angiogenesis and vascular cell
survival." Chem. Biol. 3(11): 881-885.
Strunz, G. M., M. Kakushima, et al. (1973). "Hyalodendrin: A new fungitoxic
epidithiodioxopiperazine produced by a hyalodendron species." Journal of the Chemical
Society, Perkin Transactions 1: 2600-2602.
238
Subarsky, P. and R. P. Hill (2003). "The hypoxic tumour microenvironment and
metastatic progression." Clin. Exp. Metastasis 20(3): 237-250.
Sun, X., J. R. Kanwar, et al. (2001). "Gene transfer of antisense hypoxia inducible factor-
1 alpha enhances the therapeutic efficacy of cancer immunotherapy." Gene Ther. 8(8):
638-645.
Taichman, R. S., C. Cooper, et al. (2002). "Use of the stromal cell-derived factor-1/cxcr4
pathway in prostate cancer metastasis to bone." Cancer Research 62(6): 1832-1837.
Takahashi, C., A. Numata, et al. (1994). "Leptosin-i and leptosin-j, cytotoxic substances
produced by a leptosphaeria sp - physicochemical properties and structures." J. Antibiot.
47(11): 1242-1249.
Talks, K. L., H. Turley, et al. (2000). "The expression and distribution of the hypoxia-
inducible factors hif-1 alpha and hif-2 alpha in normal human tissues, cancers, and
tumor-associated macrophages." Am. J. Pathol. 157(2): 411-421.
Tanaka, T., T. Azuma, et al. (2000). "A new sulfenylation reagent, 3-phenylsulfenyl-2-(n-
cyanoimino)thiazolidine, and its optically active version." Synlett(1): 33-36.
Tannock, I. F. (1968). "Relation between cell proliferation and vascular system in a
transplanted mouse mammary tumour." Br. J. Cancer 22(2): 258-&.
Tannock, I. F. (1998). "Conventional cancer therapy: Promise broken or promise
delayed?" The Lancet 351: SII9-SII16.
Taylor, A. (1971). The toxicology of sporidesmins and other epi poly thiadioxo
piperazines.
Tietze, L. F., U. Griesbach, et al. (2002). "Novel carboranes with a DNA binding unit for
the treatment of cancer by boron neutron capture therapy." Chembiochem 3(2-3): 219-
225.
Trost, B. M. (1978). "Alpha-sulfenylated carbonyl-compounds in organic-synthesis."
Chem. Rev. 78(4): 363-382.
239
Trost, B. M. (1978). "Some aspects of organosulfur-mediated synthetic methods."
Accounts Chem. Res. 11(12): 453-461.
Trown, P. W. (1968). "Antiviral activity of n,n'-dimethyl-epidithiapiperazinedione a
synthetic compound related to gliotoxins ll-s88alpha and beta chetomin and
sporidesmins." Biochem. Biophys. Res. Commun. 33(3): 402-&.
Trusolino, L. and P. M. Comoglio (2002). "Scatter-factor and semaphorin receptors: Cell
signalling for invasive growth." Nature Reviews Cancer 2(4): 289-300.
Vajkoczy, P., A. Ullrich, et al. (2000). "Intravital fluorescence videomicroscopy to study
tumor angiogenesis and microcirculation." Neoplasia 2(1-2): 53-61.
Vaupel, P., K. Schlenger, et al. (1991). "Oxygenation of human tumors: Evaluation of
tissue oxygen distribution in breast cancers by computerized o2 tension measurements."
Cancer Research 51(12): 3316-3322.
Vigushin, D. M., N. Mirsaidi, et al. (2004). "Gliotoxin is a dual inhibitor of
farnesyltransferase and geranylgeranyltransferase i with antitumor activity against breast
cancer in vivo." Med. Oncol. 21(1): 21-30.
Waksman, S. A. and E. Bugie (1944). "Chaetomin, a new antibiotic substance produced
by chaetomium cochliodes i. Formation and properties." J. Bacteriol. 48(5): 527-530.
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." Proc. Natl. Acad. Sci. U. S. A.
92(12): 5510-5514.
Wang, G. L. and G. L. Semenza (1993). "Desferrioxamine induces erythropoietin gene-
expression and hypoxia-inducible factor-i DNA-binding activity - implications for
models of hypoxia signal-transduction." Blood 82(12): 3610-3615.
Wang, W., H. Li, et al. (2004). "Direct, organocatalytic alpha-sulfenylation of aldehydes
and ketones." Tetrahedron Lett. 45(44): 8229-8231.
Warburg, O. (1956). "Respiratory impairment in cancer cells." Science 124(3215): 269-
270.
240
Wartenberg, M., F. C. Ling, et al. (2003). "Regulation of the multidrug resistance
transporter p-glycoprotein in multicellular tumor spheroids by hypoxiainducible factor-1
and reactive oxygen species." Faseb J. 17(1): 503-+.
Weidner, K. M., S. DiCesare, et al. (1996). "Interaction between gab1 and the c-met
receptor tyrosine kinase is responsible for epithelial morphogenesis." Nature 384(6605):
173-176.
Wenger, R. H. (2000). "Mammalian oxygen sensing, signalling and gene regulation." J.
Exp. Biol. 203(8): 1253-1263.
Wilder, R. L. (2002). "Integrin alpha v beta 3 as a target for treatment of rheumatoid
arthritis and related rheumatic diseases." Ann. Rheum. Dis. 61: 96-99.
Williams, R. M., J. H. Cao, et al. (2003). "Asymmetric, stereocontrolled total synthesis of
paraherquamide a." J. Am. Chem. Soc. 125(40): 12172-12178.
Woo, K. J., T.-J. Lee, et al. (2006). "Desferrioxamine, an iron chelator, enhances hif-1α
accumulation via cyclooxygenase-2 signaling pathway." Biochem. Biophys. Res.
Commun. 343(1): 8-14.
Yamada, T., C. Iwamoto, et al. (2004). "Leptosins o-s, cytotoxic metabolites of a strain of
leptosphaeria sp isolated from a marine alga." Heterocycles 63(3): 641-653.
Yang, C., L. H. Shapiro, et al. (1998). "A role for creb binding protein and p300
transcriptional coactivators in ets-1 transactivation functions." Mol. Cell. Biol. 18(4):
2218-2229.
Yang, W. L., R. F. Barth, et al. (2002). "Convection-enhanced delivery of boronated
epidermal growth factor for molecular targeting of egf receptor-positive gliomas." Cancer
Research 62(22): 6552-6558.
Yarmoluk, S. M., A. M. Kostenko, et al. (2000). "Interaction of cyanine dyes with nucleic
acids. Part 19: New method for the covalent labeling of oligonucleotides with pyrylium
cyanine dyes." Bioorg. Med. Chem. Lett. 10(19): 2201-2204.
Yarmoluk, S. M., D. B. Kryvorotenko, et al. (2001). "Interaction of cyanine dyes with
nucleic acids. Xx. New methods for the preparation of fluorescent probes based on
benzothiazol-4-[2,6-dimethylpyridinium] cyanine dyes." Dyes Pigment. 48(3): 165-172.
241
Zeman, E. M., J. M. Brown, et al. (1986). "Sr-4233 - a new bioreductive agent with high
selective toxicity for hypoxic mammalian-cells." Int. J. Radiat. Oncol. Biol. Phys. 12(7):
1239-1242.
Zeng, H. Y., H. F. Dvorak, et al. (2001). "Vascular permeability factor (vpf)/vascular
endothelial growth factor (vegf) receptor-1 down-modulates vpf/vegf receptor-2-
mediated endothelial cell proliferation, but not migration, through phosphatidylinositol 3-
kinase-dependent pathways." J. Biol. Chem. 276(29): 26969-26979.
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.
242
Appendix A
1
H and
13
C NMRs for Chapter 2
243
Figure N1.
1
H NMR spectrum of 11 (600 MHz, CDCl
3
).
244
Figure N2.
13
C NMR spectrum of 11 (150 MHz, CDCl
3
).
245
Figure N3.
1
H NMR spectrum of 13 (600 MHz, CDCl
3
).
246
Figure N4.
13
C NMR spectrum of 13 (75 MHz, CDCl
3
).
247
Figure N5.
1
H NMR spectrum of 14 (600 MHz, CDCl
3
).
248
Figure N6.
13
C NMR spectrum of 14 (150 MHz, CDCl
3
).
249
Figure N7.
1
H NMR spectrum of 15 (600 MHz, CDCl
3
).
250
Figure N8.
13
C NMR spectrum of 15 (125 MHz, CDCl
3
).
251
Figure N9.
1
H NMR spectrum of 16 (500 MHz, CDCl
3
).
252
Figure N10.
13
C NMR spectrum of 16 (125 MHz, CDCl
3
).
253
Figure N11.
1
H NMR spectrum of 17 (250 MHz, CDCl
3
).
254
Figure N12.
13
C NMR spectrum of 17 (75 MHz, CDCl
3
).
255
Figure N13.
1
H NMR spectrum of 18 (600 MHz, CDCl
3
).
256
Figure N14.
13
C NMR spectrum of 18 (150 MHz, CDCl
3
).
257
Figure N15.
1
H NMR spectrum of 12 (600 MHz, CDCl
3
).
258
Figure N16.
13
C NMR spectrum of 12 (150 MHz, CDCl
3
).
259
Figure N17.
1
H NMR spectrum of 19 (600 MHz, CDCl
3
).
260
Figure N18.
13
C NMR spectrum of 19 (150 MHz, CDCl
3
).
261
Figure N19.
1
H NMR spectrum of 20 (600 MHz, CDCl
3
).
262
Figure N20.
13
C NMR spectrum of 20 (150 MHz, CDCl
3
).
263
Figure N21.
1
H NMR spectrum of 21 (500 MHz, CDCl
3
).
264
Figure N22.
13
C NMR spectrum of 21 (125 MHz, CDCl
3
).
265
Figure N23.
1
H NMR spectrum of 22b (500 MHz, CDCl
3
).
266
Figure N24.
13
C NMR spectrum of 22b (125 MHz, CDCl
3
).
267
Figure N25.
1
H NMR spectrum of 24 (600 MHz, CDCl
3
).
268
Figure N26.
13
C NMR spectrum of 24 (150 MHz, CDCl
3
).
269
Figure N27.
1
H NMR spectrum of 25 (600 MHz, CDCl
3
).
270
Figure N28.
13
C NMR spectrum of 25 (150 MHz, CDCl
3
).
271
Figure N29.
1
H NMR spectrum of 26 (500 MHz, CDCl
3
).
272
Figure N30.
13
C NMR spectrum of 26 (125 MHz, CDCl
3
).
273
Figure N31.
1
H NMR spectrum of 27 (500 MHz, CDCl
3
).
274
Figure N32.
13
C NMR spectrum of 27 (125 MHz, CDCl
3
).
275
Figure N33.
1
H NMR spectrum of 28 (500 MHz, CDCl
3
).
276
Figure N34.
13
C NMR spectrum of 28 (125 MHz, CDCl
3
).
277
Figure N35.
1
H NMR spectrum of 29 (500 MHz, CDCl
3
).
278
Figure N36.
13
C NMR spectrum of 29 (125 MHz, CDCl
3
).
279
Figure N37.
1
H NMR spectrum of 30 (500 MHz, CDCl
3
).
280
Figure N38.
13
C NMR spectrum of 30 (125 MHz, CDCl
3
).
281
Figure N39.
1
H NMR spectrum of 31 (500 MHz, CDCl
3
).
282
Figure N40.
13
C NMR spectrum of 31 (125 MHz, CDCl
3
).
283
Figure N41.
1
H NMR spectrum of 32 (500 MHz, CDCl
3
).
284
Figure N42.
13
C NMR spectrum of 32 (125 MHz, CDCl
3
).
285
Figure N43.
1
H NMR spectrum of 33 (500 MHz, CDCl
3
).
286
Figure N44.
13
C NMR spectrum of 33 (125 MHz, CDCl
3
).
287
Figure N45.
1
H NMR spectrum of 34 (500 MHz, CDCl
3
).
288
Figure N46.
13
C NMR spectrum of 34 (125 MHz, CDCl
3
).
289
Figure N47.
1
H NMR spectrum of 35 (500 MHz, CDCl
3
).
290
Figure N48.
13
C NMR spectrum of 35 (125 MHz, CDCl
3
).
291
Figure N49.
1
H NMR spectrum of 36 (500 MHz, CDCl
3
).
292
Figure N50.
13
C NMR spectrum of 36 (125 MHz, CDCl
3
).
293
Figure N51.
1
H NMR spectrum of 37 (500 MHz, CDCl
3
).
294
Figure N52.
13
C NMR spectrum of 37 (125 MHz, CDCl
3
).
295
Figure N53.
1
H NMR spectrum of 38 (500 MHz, CDCl
3
).
296
Figure N54.
13
C NMR spectrum of 38 (125 MHz, CDCl
3
).
297
Figure N55.
1
H NMR spectrum of 39 (500 MHz, CDCl
3
).
298
Figure N56.
13
C NMR spectrum of 39 (125 MHz, CDCl
3
).
299
Appendix B
X-ray Crystallographic Experimental Details for 27.
Figure S1. Structure of tert-Butyl-1,4-dimethyl-3,6-dioxo-2-(phenylthio)piperazine-2-
carboxylate 27.
Compound 27 crystallized as large colorless prisms suitable for analysis by single crystal X-
ray diffraction. Data for compound 27 were collected using graphite-monochromated MoKα
radiation. In all cases the crystal temperature was maintained at 120 °K using an Oxford
Cryosystems Cryostream. The diffractometer control program SMART
1
was used for data
collection, SAINT
2
was used for data integration and reduction. A wavelength of 0.7749 Å was
used and data were measured using a Bruker APEXII CCD detector, controlled with the APEXII
3
software suite. Face-indexed absorption corrections were applied using SADABS.
4
The structures
were solved by direct methods and refined using SHELXTL.
5
Molecular diagrams were
produced using ORTEP 3
6
for Windows and Mercury 1.5.
7
Hydrogen atoms were freely refined.
Crystal data and structure refinement for 27 are provided in Table S1.
300
The molecular structure of 27 is shown in Figure S1. The compound has adopted a compact
conformation with the phenyl and N-methylpiperazinedione rings roughly aligned. Molecular
dimensions are shown in Table S3. The central piperazine-2,5-dione ring adopts a slight boat
conformation (Cremer-Pople puckering parameter Q = 0.2375(19)Å)
8
with a mean plane fitted
through all ring non-hydrogen atoms having an rms deviation of 0.1111Å.
The crystal packing contains no π…π interactions with only two weak C–H…O hydrogen
bonds (Table S7) forming an ) 6 (
2
1
R motif and linking the molecules into a chain parallel with the
a-axis (Figure S2).
Figure S2. A b-axis plot showing weak hydrogen bonding which link adjacent molecules into a
chain parallel with the a axis.
Table S1. Crystal data and structure refinement for 27.
301
Chemical formula (moiety) C
17
H
22
N
2
O
4
S
Chemical formula (total) C
17
H
22
N
2
O
4
S
Formula weight 350.43
Temperature 220(2) K
Radiation, wavelength MoK
Crystal system, space group orthorhombic, Pbca
Unit cell parameters a = 11.4784(15) Å = 90°
b = 16.628(2) Å = 90°
c = 18.778(2) Å = 90°
Cell volume, V 3584.1(8) Å
3
Z 8
Calculated density 1.299 g/cm
3
Absorption coefficient 0.203 mm
1
F(000) 1488
Crystal colour and size colorless, 0.72 0.47 0.33 mm
3
Reflections for cell refinement 7824 ( range 2.2 to 28.1°)
Data collection method Bruker SMART 1000 CCD diffractometer
thin-
range for data collection 2.2 to 27.1°
Index ranges h 14 to 14, k 21 to 21, l 24 to 24
Completeness to = 27.1° 99.9 %
Reflections collected 28216
Independent reflections 3954 (R
int
= 0.0321)
Reflections with F
2
2960
Absorption correction numerical
Min. and max. transmission 0.8675 and 0.9360
302
Structure solution direct methods
Refinement method Full-matrix least-squares on F
2
Weighting parameters a, b 0.0406, 2.5145
Data / restraints / parameters 3954 / 0 / 305
Final R indices [F
2
R1 = 0.0399, wR2 = 0.0928
R indices (all data) R1 = 0.0624, wR2 = 0.1114
Goodness-of-fit on F
2
1.070
Largest and mean shift/su 0.000 and 0.000
Largest diff. peak and hole 0.28 and 0.26 e Å
3
303
Table S2. Atomic coordinates and equivalent isotropic displacement parameters (Å
2
) for 27. U
eq
is defined as one third of the trace of the orthogonalized U
ij
tensor.
x y z U
eq
S 0.48573(4) 0.00759(3) 0.35161(3) 0.03451(14)
O(1) 0.09666(12) 0.04164(10) 0.41048(8) 0.0471(4)
O(2) 0.45547(11) 0.18376(8) 0.28255(7) 0.0398(3)
O(3) 0.56986(12) 0.13134(9) 0.43811(8) 0.0470(4)
O(4) 0.41207(11) 0.21273(8) 0.44461(7) 0.0338(3)
N(1) 0.28846(13) 0.07330(9) 0.41234(8) 0.0296(3)
N(2) 0.27044(13) 0.13785(9) 0.27489(8) 0.0316(3)
C(1) 0.39321(15) 0.09612(10) 0.37495(9) 0.0279(4)
C(2) 0.18276(16) 0.06740(11) 0.37996(10) 0.0318(4)
C(3) 0.17305(17) 0.09333(13) 0.30401(11) 0.0342(4)
C(4) 0.37414(15) 0.14422(10) 0.30613(9) 0.0283(4)
C(5) 0.47146(16) 0.14873(11) 0.42344(10) 0.0323(4)
C(6) 0.46909(17) 0.27578(12) 0.48937(10) 0.0366(4)
C(7) 0.3735(3) 0.33630(18) 0.50043(19) 0.0624(8)
C(8) 0.5677(3) 0.3123(2) 0.44735(19) 0.0699(9)
C(9) 0.5079(4) 0.24001(18) 0.55890(15) 0.0668(8)
C(10) 0.38293(17) 0.04420(11) 0.29779(11) 0.0352(4)
C(11) 0.3042(2) 0.09693(13) 0.32883(13) 0.0452(5)
C(12) 0.2197(2) 0.13357(15) 0.28770(18) 0.0618(7)
C(13) 0.2132(3) 0.11744(16) 0.21607(18) 0.0666(8)
C(14) 0.2927(3) 0.06704(17) 0.18483(14) 0.0599(7)
304
C(15) 0.3792(2) 0.03066(14) 0.22486(12) 0.0452(5)
C(16) 0.3026(2) 0.04103(15) 0.48432(11) 0.0412(5)
C(17) 0.2519(2) 0.17053(16) 0.20327(12) 0.0456(5)
Table S3. Bond lengths [Å] and angles [°] for 27.
S–C(1) 1.8674(18) S–C(10) 1.776(2)
O(1)–C(2) 1.220(2) O(2)–C(4) 1.225(2)
O(3)–C(5) 1.198(2) O(4)–C(5) 1.325(2)
O(4)–C(6) 1.495(2) N(1)–C(1) 1.443(2)
N(1)–C(2) 1.361(2) N(1)–C(16) 1.463(2)
N(2)–C(3) 1.448(2) N(2)–C(4) 1.331(2)
N(2)–C(17) 1.466(2) C(1)–C(4) 1.536(2)
C(1)–C(5) 1.549(2) C(2)–C(3) 1.494(3)
C(3)–H(3A) 0.96(2) C(3)–H(3B) 0.97(2)
C(6)–C(7) 1.503(3) C(6)–C(8) 1.508(3)
C(6)–C(9) 1.502(3) C(7)–H(7A) 0.99(3)
C(7)–H(7B) 0.94(4) C(7)–H(7C) 0.97(3)
C(8)–H(8A) 0.92(4) C(8)–H(8B) 0.93(3)
C(8)–H(8C) 1.00(4) C(9)–H(9A) 0.96(3)
C(9)–H(9B) 0.92(3) C(9)–H(9C) 0.98(3)
C(10)–C(11) 1.388(3) C(10)–C(15) 1.389(3)
C(11)–H(11) 0.97(2) C(11)–C(12) 1.382(3)
C(12)–H(12) 0.87(3) C(12)–C(13) 1.374(4)
C(13)–H(13) 0.98(3) C(13)–C(14) 1.371(4)
C(14)–H(14) 0.95(3) C(14)–C(15) 1.384(3)
C(15)–H(15) 0.98(2) C(16)–H(16A) 0.95(3)
C(16)–H(16B) 0.94(3) C(16)–H(16C) 0.93(3)
C(17)–H(17A) 0.96(3) C(17)–H(17B) 0.97(3)
C(17)–H(17C) 0.96(3)
C(1)–S–C(10) 97.93(8) C(5)–O(4)–C(6) 120.46(14)
C(1)–N(1)–C(2) 123.00(15) C(1)–N(1)–
C(16)
116.94(16)
C(2)–N(1)–C(16) 119.04(16) C(3)–N(2)–C(4) 124.37(15)
C(3)–N(2)–C(17) 115.07(16) C(4)–N(2)–
C(17)
120.31(17)
S–C(1)–N(1) 112.38(12) S–C(1)–C(4) 107.10(11)
S–C(1)–C(5) 104.68(12) N(1)–C(1)–C(4) 115.31(14)
N(1)–C(1)–C(5) 110.22(14) C(4)–C(1)–C(5) 106.45(14)
O(1)–C(2)–N(1) 122.53(18) O(1)–C(2)–C(3) 119.29(17)
N(1)–C(2)–C(3) 118.18(16) N(2)–C(3)–C(2) 116.77(16)
N(2)–C(3)–H(3A) 110.3(13) N(2)–C(3)–
H(3B)
108.5(13)
C(2)–C(3)–H(3A) 106.4(13) C(2)–C(3)–
H(3B)
106.4(13)
H(3A)–C(3)–H(3B) 108.1(18) O(2)–C(4)–N(2) 124.40(17)
O(2)–C(4)–C(1) 118.38(16) N(2)–C(4)–C(1) 117.19(15)
O(3)–C(5)–O(4) 127.59(18) O(3)–C(5)–C(1) 123.05(17)
O(4)–C(5)–C(1) 109.36(15) O(4)–C(6)–C(7) 103.17(16)
O(4)–C(6)–C(8) 108.49(18) O(4)–C(6)–C(9) 109.93(18)
C(7)–C(6)–C(8) 110.5(3) C(7)–C(6)–C(9) 111.2(2)
C(8)–C(6)–C(9) 113.1(3) C(6)–C(7)–H(7A) 108.1(16)
C(6)–C(7)–H(7B) 114(2) C(6)–C(7)–H(7C) 109(2)
H(7A)–C(7)–H(7B) 111(3) H(7A)–C(7)–H(7C) 108(3)
H(7B)–C(7)–H(7C) 107(3) C(6)–C(8)–H(8A) 107(3)
305
C(6)–C(8)–H(8B) 107.3(19) C(6)–C(8)–H(8C) 112(2)
H(8A)–C(8)–H(8B) 111(3) H(8A)–C(8)–H(8C) 110(3)
H(8B)–C(8)–H(8C) 109(3) C(6)–C(9)–H(9A) 113.9(19)
C(6)–C(9)–H(9B) 107.8(19) C(6)–C(9)–H(9C) 107.8(17)
H(9A)–C(9)–H(9B) 103(3) H(9A)–C(9)–H(9C) 115(3)
H(9B)–C(9)–H(9C) 108(3) S–C(10)–C(11) 120.00(16)
S–C(10)–C(15) 120.21(17) C(11)–C(10)–C(15) 119.8(2)
C(10)–C(11)–H(11) 117.4(14) C(10)–C(11)–C(12) 120.1(2)
H(11)–C(11)–C(12) 122.5(14) C(11)–C(12)–H(12) 119(2)
C(11)–C(12)–C(13) 120.0(3) H(12)–C(12)–C(13) 121.4(19)
C(12)–C(13)–H(13) 116.6(19) C(12)–C(13)–C(14) 120.1(3)
H(13)–C(13)–C(14) 123.2(19) C(13)–C(14)–H(14) 122.5(18)
C(13)–C(14)–C(15) 120.8(3) H(14)–C(14)–C(15) 116.7(18)
C(10)–C(15)–C(14) 119.2(2) C(10)–C(15)–H(15) 121.3(14)
C(14)–C(15)–H(15) 119.5(14) N(1)–C(16)–H(16A) 109.1(16)
N(1)–C(16)–H(16B) 110.5(16) N(1)–C(16)–H(16C) 110.9(18)
H(16A)–C(16)–H(16B) 104(2) H(16A)–C(16)–H(16C) 110(2)
H(16B)–C(16)–H(16C) 112(2) N(2)–C(17)–H(17A) 109.6(15)
N(2)–C(17)–H(17B) 109(2) N(2)–C(17)–H(17C) 108.4(16)
H(17A)–C(17)–H(17B) 114(2) H(17A)–C(17)–H(17C) 107(2)
H(17B)–C(17)–H(17C) 109(3)
Table S4. Anisotropic displacement parameters (Å
2
) for 27. The anisotropic displacement
factor exponent takes form:
2
[h
2
a*
2
U
11
+ ...+ 2hka*b*U
12
]
U
11
U
22
U
33
U
23
U
13
U
12
S 0.0312(2) 0.0359(3) 0.0364(3) 0.0036(2) 0.00170(19) 0.00845(19)
O(1) 0.0337(8) 0.0606(10) 0.0468(9) 0.0005(7) 0.0093(7) 0.0075(7)
O(2) 0.0333(7) 0.0445(8) 0.0415(8) 0.0059(6) 0.0030(6) 0.0057(6)
O(3) 0.0332(8) 0.0530(9) 0.0549(9) 0.0129(7) 0.0148(7) 0.0083(7)
O(4) 0.0308(7) 0.0341(7) 0.0367(7) 0.0071(6) 0.0041(6) 0.0002(5)
N(1) 0.0290(8) 0.0331(8) 0.0265(7) 0.0007(6) 0.0020(6) 0.0008(6)
N(2) 0.0312(8) 0.0337(8) 0.0297(8) 0.0022(6) 0.0037(6) 0.0005(6)
C(1) 0.0254(8) 0.0296(9) 0.0286(9) 0.0025(7) 0.0005(7) 0.0038(7)
C(2) 0.0285(9) 0.0317(9) 0.0352(10) 0.0045(8) 0.0027(8) 0.0006(7)
C(3) 0.0277(9) 0.0383(11) 0.0366(10) 0.0027(8) 0.0038(8) 0.0004(8)
C(4) 0.0289(9) 0.0271(8) 0.0289(9) 0.0022(7) 0.0017(7) 0.0029(7)
C(5) 0.0292(9) 0.0382(10) 0.0294(9) 0.0005(8) 0.0028(7) 0.0001(8)
C(6) 0.0385(11) 0.0357(10) 0.0357(10) 0.0077(8) 0.0051(8) 0.0050(8)
C(7) 0.0556(16) 0.0514(15) 0.080(2) 0.0286(15) 0.0154(15) 0.0108(13)
C(8) 0.071(2) 0.0668(19) 0.072(2) 0.0182(16) 0.0159(16) 0.0349(17)
C(9) 0.106(3) 0.0533(16) 0.0413(14) 0.0063(12) 0.0234(16) 0.0012(18)
C(10) 0.0366(10) 0.0316(9) 0.0374(10) 0.0067(8) 0.0004(8) 0.0084(8)
C(11) 0.0502(13) 0.0330(10) 0.0525(14) 0.0025(9) 0.0008(11) 0.0028(9)
C(12) 0.0580(16) 0.0372(12) 0.090(2) 0.0101(13) 0.0073(15) 0.0060(12)
C(13) 0.0672(17) 0.0461(14) 0.087(2) 0.0317(14) 0.0290(16) 0.0099(13)
C(14) 0.0724(18) 0.0605(16) 0.0467(14) 0.0207(12) 0.0160(13) 0.0208(14)
C(15) 0.0504(13) 0.0472(12) 0.0379(11) 0.0082(9) 0.0012(10) 0.0111(10)
C(16) 0.0455(13) 0.0493(13) 0.0287(10) 0.0034(9) 0.0033(9) 0.0020(11)
306
C(17) 0.0508(13) 0.0505(13) 0.0354(11) 0.0107(10) 0.0104(10) 0.0038(11)
Table S5. Hydrogen coordinates and isotropic displacement parameters (Å
2
) for 27.
x y z U
H(3A) 0.1596(18) 0.0454(13) 0.2767(11) 0.040(6)
H(3B) 0.104(2) 0.1266(14) 0.3009(12) 0.046(6)
H(7A) 0.405(2) 0.3813(18) 0.5290(16) 0.075(8)
H(7B) 0.306(3) 0.315(2) 0.5223(19) 0.109(13)
H(7C) 0.349(3) 0.357(2) 0.4545(19) 0.098(12)
H(8A) 0.626(3) 0.275(2) 0.445(2) 0.117(15)
H(8B) 0.592(3) 0.358(2) 0.4718(17) 0.084(9)
H(8C) 0.543(3) 0.328(2) 0.398(2) 0.122(14)
H(9A) 0.568(3) 0.200(2) 0.5541(16) 0.084(10)
H(9B) 0.446(3) 0.2120(18) 0.5776(16) 0.070(10)
H(9C) 0.527(2) 0.2845(18) 0.5913(16) 0.075(9)
H(11) 0.3098(19) 0.1050(14) 0.3798(13) 0.048(6)
H(12) 0.171(2) 0.1666(17) 0.3083(15) 0.070(9)
H(13) 0.148(3) 0.1415(18) 0.1903(16) 0.085(9)
H(14) 0.291(2) 0.0542(18) 0.1355(16) 0.079(9)
H(15) 0.434(2) 0.0060(14) 0.2017(13) 0.050(7)
H(16A) 0.232(3) 0.0167(17) 0.4988(13) 0.065(8)
H(16B) 0.314(2) 0.0829(16) 0.5173(14) 0.058(7)
H(16C) 0.363(3) 0.0036(19) 0.4860(16) 0.081(10)
H(17A) 0.323(2) 0.1937(16) 0.1859(14) 0.062(8)
H(17B) 0.187(3) 0.208(2) 0.2046(17) 0.095(11)
H(17C) 0.233(2) 0.1270(17) 0.1720(14) 0.069(8)
Table S6. Torsion angles [°] for 27.
C(2)–N(1)–C(1)–S 100.25(17) C(2)–N(1)–C(1)–
C(4)
22.9(2)
C(2)–N(1)–C(1)–
C(5)
143.41(17) C(16)–N(1)–C(1)–S 68.06(19)
C(16)–N(1)–C(1)–
C(4)
168.80(16) C(16)–N(1)–C(1)–
C(5)
48.3(2)
C(10)–S–C(1)–
N(1)
60.08(14) C(10)–S–C(1)–C(4) 67.54(13)
C(10)–S–C(1)–
C(5)
179.70(12) C(1)–N(1)–C(2)–
O(1)
173.71(17)
C(1)–N(1)–C(2)–
C(3)
6.0(3) C(16)–N(1)–C(2)–
O(1)
5.6(3)
C(16)–N(1)–C(2)–
C(3)
174.10(18) C(4)–N(2)–C(3)–
C(2)
12.8(3)
C(17)–N(2)–C(3)–
C(2)
172.94(18) O(1)–C(2)–C(3)–
N(2)
168.07(18)
307
N(1)–C(2)–C(3)–
N(2)
12.2(3) C(3)–N(2)–C(4)–
O(2)
177.80(18)
C(3)–N(2)–C(4)–
C(1)
4.4(3) C(17)–N(2)–C(4)–
O(2)
8.2(3)
C(17)–N(2)–C(4)–
C(1)
169.58(17) S–C(1)–C(4)–O(2) 73.78(18)
S–C(1)–C(4)–N(2) 104.15(15) N(1)–C(1)–C(4)–
O(2)
160.34(16)
N(1)–C(1)–C(4)–
N(2)
21.7(2) C(5)–C(1)–C(4)–
O(2)
37.8(2)
C(5)–C(1)–C(4)–
N(2)
144.30(16) C(6)–O(4)–C(5)–
O(3)
3.2(3)
C(6)–O(4)–C(5)–
C(1)
177.08(15) S–C(1)–C(5)–O(3) 1.4(2)
S–C(1)–C(5)–O(4) 178.38(12) N(1)–C(1)–C(5)–
O(3)
122.4(2)
N(1)–C(1)–C(5)–
O(4)
57.32(19) C(4)–C(1)–C(5)–
O(3)
111.9(2)
C(4)–C(1)–C(5)–
O(4)
68.40(18) C(5)–O(4)–C(6)–
C(7)
179.5(2)
C(5)–O(4)–C(6)–
C(8)
62.3(3) C(5)–O(4)–C(6)–
C(9)
61.8(3)
C(1)–S–C(10)–
C(11)
87.35(17) C(1)–S–C(10)–C(15) 91.11(17)
S–C(10)–C(11)–
C(12)
176.11(18) C(15)–C(10)–C(11)–
C(12)
2.4(3)
C(10)–C(11)–
C(12)–C(13)
0.4(4) C(11)–C(12)–C(13)–
C(14)
2.1(4)
C(12)–C(13)–
C(14)–C(15)
1.2(4) C(13)–C(14)–C(15)–
C(10)
1.5(3)
S–C(10)–C(15)–
C(14)
175.18(17) C(11)–C(10)–C(15)–
C(14)
3.3(3)
Table S7. Hydrogen bonds for 27 [Å and °].
D–H...A d(D–H) d(H...A) d(D...A) <(DHA)
C(3)–H(3B)...O(2A) 0.97(2) 2.50(2) 3.338(2) 144(2)
C(17)–H(17B)...O(2A) 0.97(3) 2.70(3) 3.420(3) 132(3)
Symmetry operations for equivalent atoms
308
A x 1/2,y, z+1/2
References for Appendix B:
1. SMART. Bruker AXS Inc.: Madison, WI, USA (2007).
2. SAINT. Bruker AXS Inc.: Madison, WI, USA (2007).
3. APEX II. Bruker AXS Inc.: Madison, WI, USA (2008).
4. Sheldrick, G. M.. SADABS 2007/4. University of Göttingen: Göttingen, Germany
(1996).
5. Sheldrick, G. M.. Acta Crystallogr. Sect. A, 2008, 64, 112.
6. Farrugia, L. J. J. Appl. Cryst. 1997, 30, 565.
7. Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Shields, G. P.; Taylor,
R., Towler, M.; van de Streek, J. J. Appl. Cryst. 2006, 39, 453.
8. Cremer, D.; Pople, J. A. J. Am. Chem. Soc. 1975, 97, 1354
309
Appendix C
Figure S3. sds-PAGE gel showing
15
N labeled cleaved CH1-p300. sds-PAGE gel
showing relatively pure
15
N labeled CH1-p300 from three different eppis after cleavage
from the GST tag by thrombin.
Figure S4. MAO-A levels are increased in a dose dependent manner upon treatment with
ETP-5. Monoamine oxidase A (MAO-A) levels are reduced under hypoxia induced by
DFO (300 µM). MAO-A levels are upregulated in a dose-dependent manner upon
treatment of ETP-5 at 100 nM, 400 nM and 1600 nM of ETP-5.
310
Appendix D
Part 1
Direct Fluorescence Imaging of Integrin Receptors with Small Molecule
Sensor
Swati Kushal
†
, Nathan W. Polaske
†
,
Ramin Dubey
†
,
Alexis Mollard
#
,
Katherine M.
Block
†‡
,
and Bogdan Z. Olenyuk*
†§
Department of Chemistry and Biochemistry, The University of Arizona, 1306 E University Blvd,
Tucson, AZ 85721, College of Pharmacy, The University of Arizona, 1295 N. Martin St., Tucson,
AZ 85721, Arizona Cancer Center, 1515 North Campbell Ave., Tucson, AZ 85724, Department of
Chemistry, The University of Utah, 315 S 1400 E, Salt Lake City, UT 84112
1
RECEIVED DATE (automatically inserted by publisher); olenyuk@email.arizona.edu
†
Department of Chemistry and Biochemistry ‡
College of Pharmacy
§
Arizona Cancer Center
#
Department of Chemistry, The University of Utah
311
Angiogenesis, the induction of new
blood vessels in cells and tissues, is
critical for growth and metastatic spread
of solid tumors. This process is generally
initiated by degradation of the basement
membrane underlying an existing blood
vessel. The newly created networks of
highly permeable blood vessels provide
efficient exit routes by which tumorigenic
cells enter the bloodstream.(Dvorak,
Brown et al. 1995) Fortuitously,
experimental therapies targeting this
newly formed tumor vasculature typically
do not result in acquired drug
resistance.(Kerbel and Folkman
2002)
,
(Boehm, Folkman et al. 1997)
Furthermore, due to the unique
expression of adhesion receptors such as
integrins on the surface of angiogenic
endothelial cells, such therapies allow
selective targeting of the blood supply in
tumors without affecting normal blood
vessels.
Integrins are heterodimeric
transmembrane receptors that mediate the
interactions between endothelial cells and
the extracellular matrix.(Hynes 2002)
They are involved in a large number of
fundamental intracellular processes, such
as cell-cell and cell-matrix adhesion,
differentiation, stress response and
apoptosis.
(Aplin, Howe et al. 1998)
This diverse
family contains at least 18 and 8
subunits that can dimerize in more than
24 different combinations to yield surface
receptors capable of recognizing one or
more components of extracellular
matrix.(Gottschalk and Kessler 2002)
H
N
N
NH
O
O
HN
NH O
O
-
SO
2
N
N
S
2
O
HN
O
H
N
N
NH
O
OH NH
SO
2
H
2
N
1
312
The resulting intracellular signaling is
thought to control cell survival,
proliferation, and migration. Although 9
out of 24 integrin heterodimers have been
implicated in blood vessel formation to
date,
v
3
receptors are ubiquitously
overexpressed in endothelial cells
undergoing angiogenesis, but are not
typically found on quiescent
cells.(Brooks, Stromblad et al. 1996;
Eliceiri and Cheresh 1999) This renders
them attractive antitumor targets, since
antagonists of this receptor that
effectively compete with its natural
ligand cause apoptosis in proliferating
vessels. As a result, one possible strategy
of modulating angiogenesis is to block
the function of the integrin receptors by
administration of high affinity ligands.
Based on considerations of their
bioactivity, low molecular weight
fluorescent ligands for
v
3
integrin
receptors could prove valuable as probes
for optical imaging of
v
3
positive
tumors and developing tumor
neovasculature. Until recently, most in
vivo imaging studies were performed with
radiopharmaceuticals due to high
sensitivity of the nuclear imaging
methods and because the use of small
monoatomic isotopes does not interfere
with the uptake and biological activity of
the ligands. Despite these advantages,
nuclear imaging requires the use of
complex equipment and is typically
performed in specialized centers because
of handling and regulatory issues
associated with the use of radioactive
materials. Optical imaging, on the other
hand, is an alternative and
complementary noninvasive method that
could be used to interrogate the biological
process of interest both in vitro and in
vivo.
Chart 1. Structures of v 3 specific integrin ligand 1 and
fluorescent probe 2.
313
Optical imaging methods typically rely
on the presence of the chromophore
system that can emit light within the
visible or near-infrared spectra with the
wavelength between 400-1500 nm. The
monitoring of the propagation of the light
in tissues is usually done by a CCD
camera or point source detectors. Optical
imaging in the diagnosis of disease states
with molecular probes is attractive,
because less complex imaging equipment
and less stringent regulatory procedures
are required. However, such procedures
require the development of bioavailable
fluorescent molecular probes or sensors
with the ability to bind their targets and
generate detectable signal upon binding
under physiological range of
concentrations. In addition, such probes
should not be toxic to cells and have
typical biodistribution of a small
molecule drug. Herein we report the
design of low molecular weight
v
3
–
selective
integrin ligand capable of
altering the intensity of the detectable
signal upon binding to its receptors.
Integrin-specific ligand 1 (Chart
1),
(Duggan, Duong et al. 2000)
was selected
because it represents a non-peptidic RGD
mimetic that exhibits high specificity
both in preliminary cell adhesion
experiments and in receptor-binding
studies with a mean inhibitory
concentration (IC
50
) of 40 nM for
v
3
versus 5.5 µM for
v
5
and 2.1 µM for
α
IIb
β
3
receptors.(Hood, Bednarski et al.
2002) The 140-fold difference in IC
50
between
v
3
and
v
5
is remarkable
since many RGD-type integrin ligands do
not differentiate well between the two
receptors.(Koivunen, Wang et al. 1995)
The free amine group in 1 is not essential
to its biological activity, but could be
utilized as a reactive site for the
straightforward coupling of the
314
photoactive moiety in the preparation of
fluorescent probe 2.
Figure 1. UV-vis absorbance (A), fluorescence emission (B,
ex=380 nm) and fluorescence excitation spectra (C, em=540
nm) for 2 in aqueous buffer (dotted line) and in 90% v/v
glycerol/water (solid line).
Our synthetic route to 1 was designed
to be rapid and efficient based on known
syntheses of structurally similar
compounds (see Supporting Information,
Scheme S1).
9
The precursor 10 is a
“protected” version of 1, which allows for
easier synthetic manipulation and greater
flexibility in the choosing of different
conjugation procedures. The synthesis of
1 is shown in Scheme S1 (Supporting
Information). First, N-Boc-ethanolamine
and methyl-4-hydroxybenzoate were
coupled under Mitsunobu conditions,
yielding desired ether 3 as a crude
mixture. Deprotection of the Boc group
yielded amine 4 (56% yield over two
steps). The pyrimidine ring was then
installed by the reaction with 2-
bromopyrimidine, producing methyl ester
5 in moderate yield (62%). Hydrolysis of
5 with base resulted in 6 (68% yield)
which was coupled to 3-amino-L-alanine
derivative 7 under standard peptide
coupling conditions, giving amide 8 in
60% yield. Hydrolysis of the Boc group
315
with TFA followed by reaction of the
deprotected amine with 9 provided the
protected intermediate 10 in 51% yield
over two steps (for this step compound 9
must be prepared in-situ and can be
obtained from taurine). Synthesis of the
chromophore moiety cyan 39 (11) was
carried out from commercially available
2-methylbenzothiazole as described in the
literature.(Yarmoluk, Kostenko et al.
2000; Yarmoluk, Kryvorotenko et al.
2001) Reduction of 10 with Pd/H
2
,
followed by the removal of Boc group
with TFA furnished 1. The final step was
coupling of the chromophore 11 and 1 to
obtain fluorescent probe 2.
Figure 1A shows absorption spectra for
the probe 2 in aqueous buffer and in 90%
v/v glycerol/water. The ratio of the peaks
in both solvents is nearly 1:1 with slightly
lower absorbance in 90% v/v
glycerol/water mixture, indicating a
change in the environment of the
Figure 2. Confocal images of WM115 cells (A and B) and
MCF7 cells (C and D) after incubation with 5 M of 2 in cell
culture media with 1% DMSO at 37 C (A and C), and 4 C (B
and D), respectively. Left image: fluorescence signal from
conjugate; middle: overlay of fluorescence signal with visible
light image; right: visible light image.
chromophore. Fluorescence
spectroscopy was used next to analyze
the emission properties of 2. As shown in
Figure 1B, the probe shows low
fluorescence in aqueous buffer when
excited at 380 nm. This is consistent with
the low fluorescence quantum yields
associated with unsymmetrical cyanine
dyes, likely due to torsional motions that
rapidly deactivate the excited state. In
contrast, when the probe is placed in a
glycerol/water mixture, a 35-fold
enhancement of the fluorescence intensity
is observed due to reduced torsional
motions in this more viscous solvent.
316
Similar fluorescence increase was
observed when the probe was incubated
with WM115 cells: the fluorescence of
the probe was enhanced by binding to
integrin receptors and subsequent
internalization of the probe. These results
indicate that the probe 2 exhibits
fluorescence enhancement in viscous
media and in the presence of cells
displaying
v
3
integrins (Supporting
Information, Figure S1). Corresponding
excitation spectra were also recorded,
with the emission wavelength being set at
540 nm for 2 in aqueous buffer and in the
90% v/v glycerol/water mixture. As with
the emission spectra, enhancement of the
intensity of the excitation peaks was
observed.
To determine the ability of the
fluorescent probe to bind selectively to
cells, we required cells displaying both
high and low levels of
v
3
integrin.
Based on recent work by Kiessling and
co-workers,(Carlson, Mowery et al. 2007)
we selected WM115 cells, derived from
human melanoma, that are displaying on
average 110,000 of
v
3
integrins and
MCF7 breast cancer cells with a 10-fold
reduced copy number (~11,000) of these
receptors.
In order to ensure that the prepared
conjugate was not cytotoxic, we tested
the ability of WM115 and MCF7 cells to
maintain their metabolic activity the
presence of increased concentrations of 2.
We employed the Cell Titer Blue assay
(Promega) that is based on the ability of
living cells to convert a redox dye
(resazurin) into a fluorescent end product
(resorufin). Non-viable cells rapidly lose
metabolic capacity and thus do not
generate a fluorescent signal. The assay
involved adding the reagent directly to
cells cultured in medium in the presence
of 2 and 0.5% aqueous DMSO. After the
incubation step, data were recorded with
317
fluorescent microplate reader. We found
that conjugate 2 is essentially non-
cytotoxic to cells after incubation for 1.5
- 3 h at a concentration range 1 nM – 10
M and shows low cytotoxicity at
concentrations of 10 M-100 M
(Supporting Information, Figure S2).
To assess the binding and
internalization properties of the
conjugate, we incubated 2 with WM115
and MCF7 cells for the period of 2 h,
both at 37 °C and 4 °C. The distribution
and cellular localization of the conjugate
was then analyzed by laser-scanning
confocal microscopy. Figure 2 shows
probe 2 internalized within WM115 and
MCF7 cells at 37 °C (Figure 2A and 2C),
although the extent of internalization for
each cell line is different. Images show
mature cells with well-defined nuclei and
slightly round morphology. In contrast,
the conjugate remained completely extra-
cellular when cells were incubated with
the probe at 4 °C (Figure 2B and 2D). To
further assess the mechanism of
internalization, we performed
competition experiments by co-
incubating the cells with probe 2 at a
concentration of 20 M in the presence of
a 10-fold excess of non-fluorescent ligand
1 at 37 °C for 1 h. No fluorescence was
observed within cells (Supporting
Information, Figure S3), indicating that
unlabeled ligand 1 is blocking both the
binding and internalization of 2. Taken
together, the results from the imaging
experiments support the view that 2 is
likely to enter the cells through
endocytosis.
Flow cytometry was used to analyze
the distribution and uptake of the
conjugate with the population of cells.
The probe 2 was applied at
concentrations of 5 M, 25 M and 100
M. In parallel, Annexin V staining was
used to ascertain the low apoptosis rate of
318
the cells (Supporting Information, Figure
S4). Upon excitation at 488 nm, both
WM115 and MCF7cells treated with 2
displayed high emission intensities. The
results from flow cytometry assays
parallel the results from confocal
microscopy experiments, indicating
active uptake and internalization of the
probe 2 within cells.
In conclusion, a new low molecular
weight fluorescent sensor
selective for
v
3
receptors has been designed. It
exhibits 35-50 fold fluorescence
enhancement upon binding to the cell
surface displaying its target receptors,
shows low cytotoxicity within the large
range of concentrations and has good
water solubility. Through the ability to
target
v
3
receptors it could be used in
visible-light imaging of tumor cells that
display these receptors as well as newly
formed tumor vasculature. This opens
new avenues for the application of this
ligand in integrin-targeted fluorescence-
based optical imaging. We continue
expanding our set of available
chromophores with 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 a receptor-specific ligand
conjugated to cyanine dyes forms a viable
platform with potential for future cancer
diagnostics.
Acknowledgments. Financial support by the US
National Science Foundation (CHE-0748838), the
National Institute of Health (R21 CA129388) and by
the University of Arizona are gratefully acknowledged.
Supporting Information Available. Synthetic
procedures and characterization of compounds 1-10,
additional fluorescence emission and excitation data for 2,
data from cell viability and flow cytometry assays.
319
ABSTRACT FOR WEB PUBLICATION
Designed ligands that target newly formed blood vessels could offer new tools for cancer imaging,
diagnostics, and potentially, aid drug discovery efforts for the treatment of solid tumors. We report design,
synthesis, photophysical properties and biological evaluation of a cell receptor-specific small molecule that
binds to v 3 integrins overexpressed on the surface of v 3-positive tumor cells. Its fluorescence is strongly
enhanced in the bound state or in viscous solvent due to conformational constraint of the excited state.
Staining of WM115 cells that express large number of v 3 integrins results in the efficient internalization of
the ligand.
320
Appendix E
Supporting Information of
Direct Fluorescence Imaging of Integrin
Receptors with Small Molecule Sensor
Swati Kushal,
†
Nathan W. Polaske,
†
Ramin Dubey,
†
Alexis Mollard,
#
Katherine M.
Block,
†‡
and Bogdan Z. Olenyuk*
†§
†
Department of Chemistry and Biochemistry, The University of Arizona, 1306 E University Blvd,
Tucson, AZ 85721
‡
College of Pharmacy, The University of Arizona, 1295 N. Martin St., Tucson, AZ 85721
§
Arizona Cancer Center, 1515 North Campbell Ave., Tucson, AZ 85724
#
Department of Chemistry, The University of Utah, 315 S 1400 E, Salt Lake City, UT84112
*Corresponding author, olenyuk@email.arizona.edu
Contents
General Methods S1-S7
NMR Spectra S8-S17
Scheme S1 S18
Figure S5 S19
Figure S6 S20
Figure S7 S21
Figure S8 S22
321
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
322
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
323
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), 3.66-3.53
324
(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 DMF (3.6 μL, 46 mol) in
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
325
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.
Cell lines. Human breast carcinoma (MCF7) and human malignant melanoma (WM115)
cell were obtained from ATCC (accession numbers HTB-22 and CRL-1675).
Cell culture. MCF7 cells were maintained in RPMI 1640 media (Sigma) supplemented
with 10% fetal bovine serum (Irvine Scientific). WM115 cells were grown in -MEM
supplemented with 10% fetal bovine serum and 1% of total volume of Pen/Strep (Sigma).
All cells were incubated at 37
o
C in a humidified atmosphere with 5% CO
2
. Cell growth
and morphology were monitored by phase-contrast microscopy.
Cell viability assay. WM115 or MCF7 cells were plated in opaque 96-well plates
(Greinier) at a density of 10,000 cells/well (50,000 cells/mL). After attachment, cells
were treated with 100 µL of fresh media containing compound 2 at concentrations from 1
nM to 100 M. All samples contained a final concentration of 0.5% DMSO or less.
326
Vehicle samples were treated with cell culture media containing 0.5% DMSO. Cells
were incubated with compound for a total of 3 hours. Once the incubation was complete,
20 µL of Cell Titer Blue reagent (Promega) was added to each well, and plates were
incubated at 37
o
C and 5% CO
2
for 2 hr before reading fluorescence on a BioTek Synergy
2 microplate reader. GI
50
curves were plotted using Prism (GraphPad, Inc) software.
Confocal Microscopy. MCF7 cells were detached from the surface using standard
trypsinization while WM115 cells were resuspended with accutase (Sigma). The cells
were plated in 35 mm glass-bottom dishes (MatTek) at a density of 2.97 × 10
5
cells/mL
for imaging experiment. Cells were cultured with 300 μL of media per dish for 18 hrs.
Next, cells were treated with compound 2 at concentrations of 5 μM or 20 μM or, in
competition assays, with unlabelled control 1 at 200 μM concentration and incubated for
3 h. Before imaging cells were washed with PBS and media was replaced with fresh
media. Images were captured with Zeiss LSM 510 laser scanning confocal microscope
using ×63 oil-immersion objective lens and excitation wavelength of 488 nM.
Flow Cytometry Assays. WM115 or MCF7 cells were plated in a 25 mL
cell culture
flask. After attachment, cells were treated for 2 hr with compound 2. Next, cells were
trypsinized and resuspended in the binding buffer at a concentration of 1 × 10
6
cells/mL.
Cells in binding buffer were incubated with 5 µL of Annexin V-PE and 400 µL of
binding buffer that contained compound 2 at final concentration of 100 µM at room
temperature. For control experiment cells were incubated with Annexin V-PE only. Once
the incubation was complete, cells were analyzed for uptake of 2 and apoptosis rate by
flow cytometer.
327
NMR Spectra
1
H NMR spectrum of compound 4.
328
13
C NMR spectrum of compound 4.
329
1
H NMR spectrum of compound 5.
330
13
C NMR spectrum of compound 5.
331
1
H NMR spectrum of compound 6.
332
13
C NMR spectrum of compound 6.
333
1
H NMR spectrum of compound 8.
334
1
H NMR spectrum of compound 10.
335
13
C NMR spectrum of compound 10.
336
1
H NMR spectrum of compound 2.
337
Scheme S1: Synthesis of the fluorescent probe 2. (a) PPh3, 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) 6, HBTU, DIEA, DMF, RT, 18 h. (f) TFA, CH
2
Cl
2
, RT, 30 min. (g) 9, DMAP, CH2Cl2, 0 °C to RT, 18
h. (h) Pd black, H
2
, EtOH, RT, 3 h. (i) TFA, CH
2
Cl
2
, RT, 30 min. (j) 1, DIEA, DMF, RT, 1 h.
338
Figure S5. Fluorescence emission (A,
ex
= 435 nm) and fluorescence excitation spectra (B,
em
=
540 nm) for 2 in aqueous buffer (dotted line) and in 90% v/v glycerol/water (solid line).
339
Figure S6. Cell viability values for WM115 (A) MCF7 (B) cells treated by compound 2 as
determined by Cell Titer Blue Assay. Cells were incubated with probe 2 at concentrations of 1
nM – 100 M at 37 °C for 3 h.
340
Figure S7. Confocal images of WM115 cells (A) and MCF7 cells (B) after incubation with 20 M
of 2 and 200 M 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.
341
Figure S8. Flow cytometry data for WM115 cells (A) and MCF7 cells (B) treated with Annexin V-
PE and cellular uptake of 2 in WM115 (C) and in MCF7 (D) as measured by flow cytometry. The x
axis represents fluorescence intensity of the cells resulting from the uptake of probe 2, the y axis
– Annexin V-PE apoptosis stain. The cells were incubated with 100 M of 2 for 2 hr.
Abstract (if available)
Abstract
Angiogenesis, the process of formation of new blood vessels, is a characteristic of most solid malignancies and is involved in promoting rapid growth of neoplastic diseases. In cancers, several cytokines are secreted that give rise to angiogenesis, and facilitate enhanced cell migration, invasion, survival and proliferation. Apart from the transcriptional changes that result in elevated levels of expression of the oncogenic transcription factors, the involvement of integrins, heterodimeric transmembrane receptors that modulate cell-cell and cell-matrix interactions is another key element in promoting tumorigenesis. This dissertation is focused on two aspects of modulation of oncogenic pathways: 1) by direct inhibition of the transcription factor complexes that are responsible for maintenance of oncogenic phenotype and 2) targeting of integrin receptors that are overexpressed on cancer cells and tumor vasculature with the goal of developing of novel therapeutics and imaging agents. ❧ Our main focus is on regulation of the activity of hypoxia inducible transcription factor 1α (HIF-1α), which is overexpressed in cancer cells. High metabolic rate of the growing tumor coupled with limited supply of oxygen by the nearby blood vessels quickly results in the state of local hypoxia within the tumor. The hypoxic tumor cells rapidly stabilize and accumulate the alpha subunit of the HIF-1 (HIF-1α) which is a key component of the transcription factor complex responsible for overexpression of the key genes involved in angiogenesis, invasion, and altered energy metabolism. Stabilized HIF-1α translocates to the nucleus, heterodimerizes with its beta subunit, binds to its cognate DNA sequence called hypoxia response (HRE) element and recruits cofactor p300/CBP, resulting in upregulation of hypoxia-inducible genes. This dissertation describes development of small molecules that inhibit the hypoxia-inducible transcription factor complex through the disruption of its binding to coactivator proteins p300/CBP. Using rational design and inspired by the natural product chetomin, we designed small molecules that inhibit binding of HIF-1α to p300/CBP. We have shown that the synthetic small molecules ETP-3 and ETP-5 directly interact with the cysteine-histidine rich region 1 (CH1) of human p300 coactivator and that they also disrupt the complex between HIF-1α and p300, resulting in a rapid downregulation of hypoxia-inducible genes, such as VEGF, c-Met, Glut1, LOX, CXCR4. These small molecules are potent inhibitors of HIF-1α induced signaling in vitro. In addition, ETP-5 has shown remarkably high activity in mouse tumor xenograft model. ❧ The second part of this dissertation is focused on integrins, bidirectional allosteric signaling proteins responsible for cell-cell communication, adhesion and interaction with extracellular matrix. We focused on αvβ3 integrins, as these receptors are overexpressed on vascular endothelial cells that undergo angiogenesis, although they are typically not found on quiescent cells. This makes them attractive anticancer targets, as they can be targeted in order to deliver imaging and therapeutic reagents to growing tumor vasculature. We have designed and synthesized novel αvβ3-specific sensors incorporating integrin ligand and an environment-sensitive fluorescent cyan 40 dye. Such sensors show 40-80 fold fluorescence enhancement upon binding to their cognate integrin receptors. Furthermore, we designed αvβ3-specific ligand incorporating therapeutic moieties, such as boron-rich dendritic wedges that can be used for boron neutron capture therapy and boron MRI. The integrin ligand conjugates show good affinity and specificity towards cells with high levels of expression of αvβ3 integrins. In murine subcutaneous tumor models, these conjugates have shown excellent uptake. ❧ This dissertation summarizes work focused on development and evaluation of small molecules for modulating and targeting cancer cells both on transcriptional and cellular levels.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Synthetic small molecules and protein secondary structure mimetics as modulators of hypoxia-inducible transcription and integrin receptors function
PDF
Small molecule modulators of HIF1α signaling
PDF
Mechanism of repression of RNA polymerase III-dependent transcription under hypoxia
PDF
Discovery of novel small molecules targeting cancer cell metabolism
PDF
Characterization of a fragment in secreted Hsp90α with potential therapeutic benefits in wound healing and cancer
PDF
Regulation of mitochondrial bioenergetics via PTEN (phosphatase and tensin homolog deleted on chromosome 10)/estrogen-related receptor alpha (ERRα) signaling
PDF
Development of novel small molecules targeting mitochondrial and oxidative stress signaling pathways for pancreatic cancer therapy
PDF
Identification of small-molecules targeting CXCR2 function and signaling
PDF
Design of novel anticancer agents targeting cellular stress response pathways
PDF
Role of integrin α4 in drug resistant acute lymphoblastic leukemia
PDF
Adventures in medicinal chemistry: design and synthesis of small molecule biological modulators
PDF
Understanding anti-angiogenic signaling and treatment for cancer through mechanistic modeling
PDF
Modulation of Runx proteins by steroid hormone receptors
PDF
Discovery of novel small molecules for ovarian cancer treatment
PDF
Transcriptional regulation of IFN-γ and PlGF in response to Epo and VEGF in erythroid cells
PDF
Placental growth factor mediated transcriptional and post-transcriptional regulation of hemeoxygenase-1
PDF
Cell penetrating peptide-based drug delivery system for targeting mildly acidic pH
PDF
Using cyclotides as a bioscaffold to target intracellular protein-protein interactions
PDF
Physical activity optimizes circuit-specific cellular metabolism in neuroplasticity: a role for hypoxia-inducible factor-1 and its downstream targets
PDF
Using novel small molecule modulators as a tool to elucidate the role of the Myocyte Enhancer Factor 2 (MEF2) family of transcription factors in leukemia
Asset Metadata
Creator
Dubey, Ramin
(author)
Core Title
Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publication Date
05/29/2013
Defense Date
07/25/2012
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
angiogenesis,drug discovery,hypoxia,inhibitor,integrin,OAI-PMH Harvest,Transcription
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Olenyuk, Bogdan Z. (
committee chair
), Camarero, Julio A. (
committee member
), Pratt, Matthew R. (
committee member
)
Creator Email
ramindub@usc.edu,rdubey@email.arizona.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-124862
Unique identifier
UC11292389
Identifier
usctheses-c3-124862 (legacy record id)
Legacy Identifier
etd-DubeyRamin-1373.pdf
Dmrecord
124862
Document Type
Dissertation
Rights
Dubey, Ramin
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
Tags
angiogenesis
drug discovery
hypoxia
inhibitor
integrin