University of Southern California Dissertations and Theses

Small molecule modulators of HIF1α signaling

(USC Thesis Other)

Small molecule modulators of HIF1α signaling

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content University of Southern California Los Angeles, California June 2014 Small Molecule Modulators of HIF1α Signaling Dissertation presented by Ivan V. Grishagin to the Department of Pharmacology and Pharmaceutical Sciences in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in the subject of PHARMACEUTICAL SCIENCES ii Table of Contents List of Figures ................................................................................................................ v List of Tables................................................................................................................. ix List of Schemes .............................................................................................................. x Chapter I. Introduction. .............................................................................................. 1 I.1. Transcription Factors as Therapeutic Targets. ..................................................... 2 I.2. Classification of Transcription Factors and Modulation of Their Activity. ......... 6 I.3. Hypoxia in Cancer. ............................................................................................. 10 I.4. HIF Structure and Regulation of Activity. ......................................................... 11 I.4.1. Oxygen-Dependent HIF Regulation. ........................................................... 13 I.4.2. Oxygen-Independent HIF Regulation. ........................................................ 16 I.5. The HIF Family of Transcription Factors. ......................................................... 21 I.6. HIF Function. ..................................................................................................... 24 I.6.1. Development. ............................................................................................... 24 I.6.2. Ischemia. ...................................................................................................... 25 I.6.3. Inflammation. .............................................................................................. 26 I.6.4. Apoptosis. .................................................................................................... 27 I.6.5. Cancer. ......................................................................................................... 28 I.7. Key HIF-Dependent Genes. ............................................................................... 31 I.7.1. VEGF ........................................................................................................... 32 I.7.2. MET ............................................................................................................. 38 I.7.3. LOX .............................................................................................................. 41 I.7.4. SLC2A1 ........................................................................................................ 43 I.7.5. CXCR4 ......................................................................................................... 46 I.8. Hypoxia and Anti-Cancer Treatment. ................................................................ 49 I.9. Therapeutic Targeting of HIF Pathway. ............................................................. 51 I.9.1. Natural Products Targeting HIF Pathway. .................................................. 51 I.9.2. Synthetic Small Molecules Targeting the HIF Pathway. ............................ 53 I.9.3. Transcriptional Regulation of HIF Pathway. ............................................... 56 I.10. Vascularization of Mouse Tumor Xenograft Models. ..................................... 60 iii Chapter II. Topographical Helix Mimetics as In Vivo Modulators of Hypoxia- Inducible Signaling. ..................................................................................................... 65 II.1. Introduction. ...................................................................................................... 66 II.2. Results. .............................................................................................................. 69 II.2.1. Design and Synthesis of Topographical HIF1α Mimics. ........................... 69 II.2.2. Binding Affinities of OHMs for the p300–CH1 Domain. ......................... 73 II.2.3. Designed Mimetics Down-regulate Hypoxia-Inducible Gene Expression. 77 II.2.4. Gene Expression Profiling of OHM in A549 Cells.................................... 87 II.2.5. In Vivo Assessment of the Efficacy of OHM in a Mouse Tumor Xenograft Model. .................................................................................................................... 92 II.2.6. MDA-MB-231 Mouse Tumor Xenografts. ................................................ 93 II.2.7. 786-O Mouse Tumor Xenografts. .............................................................. 96 II.3. Discussion. ........................................................................................................ 98 II.4. Conclusion. ..................................................................................................... 100 II.5. Experimental Section for Chapter II. .............................................................. 101 Chapter III. In vivo Anti-Cancer Activity of Rhomboidal Pt(II) Metallacyclic Assemblies. ................................................................................................................. 119 III.1. Introduction. .................................................................................................. 120 III.2. Results. .......................................................................................................... 123 III.2.1. Preparation and evaluation of Pt-based SCCs. ....................................... 123 III.2.2. Uptake and toxicity of SCCs in cells. ..................................................... 124 III.2.3. In Vivo Assessment of the Efficacy of SCCs in a Mouse Tumor Xenograft Model. .................................................................................................................. 127 III.3. Conclusion. .................................................................................................... 131 III.4. Experimental Section for Chapter III. ........................................................... 132 Chapter IV. Synergistic Effect of Simultaneous Inhibition of MAOA and Hypoxia Inducible Transcription in Prostate Cancer. .......................................................... 137 IV.1. Introduction. .................................................................................................. 138 IV.2. Results. .......................................................................................................... 140 IV.3. Conclusion. .................................................................................................... 150 IV.4. Experimental Section for Chapter IV. ........................................................... 151 Chapter V. Synthesis of Near-Infrared Dye MHI148. .......................................... 156 V.1. Introduction..................................................................................................... 157 V.3. Discussion of Synthetic Steps. ........................................................................ 160 iv V.3. Analysis. ......................................................................................................... 162 V.4. Conclusion. ..................................................................................................... 167 V.5. Experimental Section for Chapter V. .............................................................. 168 V.6. Notes. .............................................................................................................. 170 Chapter VI. A Scalable Procedure for the Synthesis of 3-(4-methoxyphenyl)-6,8- dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane-7,9-dione. .............................. 183 VI.1. Introduction. .................................................................................................. 184 VI.2. Results and Discussion. ................................................................................. 185 VI.3. Experimental Section for Chapter VI. ........................................................... 188 VI.4. Notes. ............................................................................................................. 192 Chapter VII. Automatic Cell Counting with ImageJ. .......................................... 205 VII.1. Introduction. ................................................................................................. 206 VII.2. Hardware. ..................................................................................................... 210 VII.3. Software. ...................................................................................................... 212 VII.4. Results. ......................................................................................................... 214 VII.5. Conclusion. .................................................................................................. 220 VII.6. Experimental Section for Chapter VII. ........................................................ 221 VII.7. Instructions. .................................................................................................. 223 VII.8. Code. ............................................................................................................ 224 VII.8.1. Macros 1. ............................................................................................... 224 VII.8.2. Macros 2. ............................................................................................... 225 References .................................................................................................................. 227 v List of Figures Figure I.1. Schematic representation of the “central dogma of molecular biology”, showing tentative relationship between fundamental biopolymers. ................................... 2 Figure I.2. Role of DNA methylation in cancer. .............................................................. 4 Figure I.3. Oxygen-dependent regulation of HIF1α signaling pathway. ........................ 15 Figure I.4. Oxygen-independent regulation of HIF1α signaling. ................................... 20 Figure I.5. Domain structure of HIF proteins. ................................................................ 21 Figure I.6. HIF2α signaling. ........................................................................................... 22 Figure I.7. Genes involved in tumor onset and progression that are activated by HIF. . 32 Figure I.8. The family of VEGF receptors and ligands and schematic representation of VEGF pathway.................................................................................................................. 35 Figure I.9. MET signaling pathway. ............................................................................... 40 Figure I.10. SDF-1-CXCR4 signaling. ........................................................................... 47 Figure I.11. Natural product inhibitors of the HIF1 pathway. ......................................... 52 Figure I.12. Synthetic small molecule inhibitors of the HIF1 pathway. ......................... 54 Figure I.13. Synthetic analogs of chetomin. ................................................................... 57 Figure I.14. HBS helix mimetics of the key α-helices of HIF1α. ................................... 59 Figure I.10. Schematic representation of the vascular system. ....................................... 60 Figure II.1. Design of HIF1α mimetics to modulate hypoxia inducible gene expression. ........................................................................................................................................... 68 Figure II.2. OHM derivatives – positive and negative controls – designed to inhibit the target complex. .................................................................................................................. 70 Figure II.3. Binding between HIF1α CTAD and p300/CBP CH1 domain. ................... 73 vi Figure II.4. Binding affinities of designed compounds for p300–CH1. .......................... 74 Figure II.5. 1 H- 15 N HSQC titration spectra. ................................................................... 76 Figure II.6-1. OHMs exhibit low cytotoxicity as evaluated in an MTT assay. .............. 78 Figure II.6-2. OHMs exhibit low cytotoxicity as evaluated by MTT assay. .................. 79 Figure II.7-1. Regulation of hypoxia inducible promoter activity by helix mimetics. ... 80 Figure II.7-2. Regulation of hypoxia inducible promoter activity by helix mimetics.. .. 81 Figure II.7-3. Regulation of hypoxia inducible promoter activity by helix mimetics. ... 82 Figure II.7-4. Transcriptional regulation of hypoxia inducible genes by helix mimetics. ........................................................................................................................................... 84 Figure II.7-5. Transcriptional regulation of hypoxia inducible genes by helix mimetics. ........................................................................................................................................... 85 Figure II.8. Western blot analysis of HIF1α levels in the whole cell extracts of A549 cells. .................................................................................................................................. 86 Figure II.9. Results from gene expression profiling obtained with Affymetrix Human Gene ST 1.0 arrays. ........................................................................................................... 89 Figure II.10. Results from gene expression profiling obtained with Affymetrix Human Gene ST 1.0 arrays. ........................................................................................................... 90 Figure II.11. Effect of OHM 1 treatment on the weight of BALB/c mice. .................... 93 Figure II.12. Effect of OHM 1 treatment on tumor growth rate in MDA-MB-231 xenografts. ......................................................................................................................... 95 Figure II.13. Histopathology data. .................................................................................. 96 Figure II.14. Effect of OHM 1 treatment on tumor growth rate in 786-O xenografts. ... 97 Figure III.1. Organoplatinum rhomboid SCCs have low cytotoxicity. ........................ 125 vii Figure III.2. LSCM images of SCCs localized in live cells. ........................................ 126 Figure III.3. LSCM images of SCCs localized in live cells. ........................................ 126 Figure III.4. Effect of SCC 4 treatment on tumor growth rate in MDA-MB-231 xenografts. ....................................................................................................................... 129 Figure III.5. Histopathology data. ................................................................................ 130 Figure IV.1. Baseline MAOA activity in normoxia. .................................................... 140 Figure IV.2. MAOA is induced by hypoxia in PC3 cells. ............................................ 141 Figure IV.3. MAOA and HIF1α are induced by hypoxia in DU145 cells. ................... 142 Figure IV.4. In C42B cells, HIF1α is abundant in normoxia, and further induced by hypoxia. ........................................................................................................................... 142 Figure IV.5. Prostate cancer is sensitive to transcriptional antagonist of hypoxia- inducible genes LS72. ..................................................................................................... 143 Figure IV.6. LS72 treatment decreases hypoxia-induced transcription levels of HIF- dependent genes (VEGFA, LOX, GLUT1) in PC3 and C42B cells. ............................... 145 Figure IV.7. MAOA is down-regulated by LS72 in a dose dependent manner............. 146 Figure IV.8. MAOA is up-regulated by LS72 in PC3 cells. .......................................... 146 Figure IV.9. In C42B cells, MAOA inhibitor clorgyline and LS72 down-regulate transcription levels of (a) MAOA and (b) HIF-dependent gene LOX, as determined by qRT-PCR......................................................................................................................... 147 Figure IV.10. In C42B cells, MAOA inhibitor clorgyline and LS72 down-regulate transcription levels of (a) MAOA, HIF-dependent genes (b) LOX and (c) GLUT1, as determined by qRT-PCR................................................................................................. 149 viii Figure IV.11. Schematic representation of existing and proposed interactions between prostate cancer, MAOA, and HIF1α. .............................................................................. 150 Figure V.1. ¹H peak assignment for V.1. ...................................................................... 163 Figure V.2. ¹³C peak assignment for V.1. ..................................................................... 163 Figure V.3. 3D graph of the HPLC trace of V.1. .......................................................... 164 Figure V.4. Products of V.1 fragmentation/methylation in mass spectrometer. ........... 165 Figure V.5. MHI148 obtained via the modified procedure crosses blood-brain barrier (data courtesy of Dr. Jason Wu, Cedars-Sinai Medical Center). .................................... 166 Figure VII.1. Application of the automatic cell counting methods. ............................. 216 Figure VII.2. Comparison of manual and automatic counting. .................................... 217 Figure VII.3. Normalized manual counting results. ..................................................... 218 Figure VII.4. Normalized automatic counting results for preferred fields. .................. 219 Figure VII.5. U251 cells prior to lysis. ......................................................................... 221 ix List of Tables Table II.1. Computational alanine scanning mutagenesis energies calculated with Rosetta(289) ver. 3.3. ....................................................................................................... 72 Table II.2. Number of affected genes in cells treated with the inhibitors of hypoxia- inducible transcription factor complex. ............................................................................ 92 Table III.1. Molar absorption coefficients and emission band maxima for compounds 1,2,4,5.............................................................................................................................. 124 Table IV.1. GI50 values of LS72 in prostate cancer cell lines. ...................................... 144 Table VII.1. Commercially available cell counters. ..................................................... 207 Table VII.2. Results of manual and automatic cell counting on the 1×1 mm area and total manual cell counting and picture taking time. ........................................................ 214 x List of Schemes Scheme II.1. Solid-phase synthesis of oxopiperazine dimers. ........................................ 72 Scheme II.2. Mechanism of hypoxia induction by DFO. ............................................... 79 Scheme III.1. Synthesis of SCCs 4 and 5. .................................................................... 123 Scheme V.1. General structure of indocyanine polymethine dyes. ............................... 157 Scheme V.2. Synthesis of MHI148 V.1. ....................................................................... 161 Scheme VI.1. Synthesis of the dithioacetal VI.1. ......................................................... 186 1 Chapter I. Introduction. 2 I.1. Transcription Factors as Therapeutic Targets. The onset of many diseases, including inflammatory, cardiovascular, and cancer, is frequently attributed to abnormal gene expression.(1-3) In turn, regulation of the gene expression can be performed virtually at any level: transcription initiation and progression, post-transcriptional RNA editing and modification, including splicing, translation, and post-translational protein modification – as outlined by the “central dogma of molecular biology”, introduced by Crick in 1958, and reinforced in 1970 (Figure I.1).(4) Figure I.1. Schematic representation of the “central dogma of molecular biology”, showing tentative relationship between fundamental biopolymers. (A) Reproduced from reference (4). Solid arrows show general (omnipresent) transfers; dotted arrows show special (rare) transfers. (B) Current more elaborate analysis with hierarchical subdivision from reference (5). Jacob and Monod established the details of the genetic regulatory mechanisms of the protein synthesis between 1959 and 1965.(6-12) They introduced the concept of the three fundamental modulators of gene regulation, as particular DNA sequences susceptible to the effector agents that specifically alter, positively or negatively, the rate of the protein 3 synthesis. First, RNA polymerases can recognize promoters, DNA sequences that contain all the information necessary for initiation of transcription. Second, repressor proteins bind to the operators and inhibit transcription by stabilizing the polymerase structure in the initiation phase and preventing it from proceeding into the elongation phase.(5) Third, activator proteins interact with positive control (often called response) elements that stimulate transcription from the promoter. On the most fundamental level, DNA methylation, the cis-regulator of DNA-protein interactions, is a well-recognized mechanism of gene regulation, performed by only one known methyltransferase in mammals;(13, 14) however, little is known about DNA demethylation. The activity of methyltransferase peaks during the S-phase of the cell cycle, but it also performs de novo methylation during oogenesis, spermatogenesis, and in embryonic stem cells. DNA methylation is frequently associated with the lack of gene expression activity,(15) and appears to be wide-spread in large genomes as compared to small ones. DNA methylation plays a genetic as well as an epigenetic role in cancer (Figure I.2).(13) Hypomethylation of c-myc, raf, c-fos, c-H-ras, and c-k-ras genes, implicated in oncogenesis, correlates with neoplasia, however the causal relationship is not clear in these cancers. On the other hand, hypermethylation of genes responsible for the maintenance of normal phenotype may lead to their silencing and the onset of the tumorigenesis. The disproportionately high mutation rate of 5-methylcytosine residues, transforming into thymine residues via deamination at carbon 4, is believed to contribute to tumorigenesis. In particular, such mutation in palindromic sequence 5´-CpG-3´ is prevalent in the p53 tumor-suppressor gene, and is characteristic of many human cancers. 4 Figure I.2. Role of DNA methylation in cancer. Reproduced from reference (13). Adjustment of the degree of DNA supercoiling (DNA over- or under-winding) is another way to achieve the differential regulation of promoters. Topoisomerases can bind to sites on the gene other than the promoter, unwind the DNA, and thereby facilitate the access of the RNA polymerase.(16) Negatively supercoiled genome regions appear to contain active genes, while the positively supercoiled ones incorporate silent genes.(17) Many tumors are also known to overexpress topoisomerases, and such misregulation may lead to chromosomal aberrations and greater mutation rate.(18) DNA-dependent RNA polymerase performs the transcription of the genes stored in the form of DNA into mRNA, by recognizing specific promoter sequences in DNA and denaturing them.(5) The most attractive property of transcription is its regulation by a vast variety of activator and repressor proteins. The function of repressors and activators can be modulated by native or artificial physiological conditions, thereby permitting regulation of the expression of corresponding genes. Additionally, RNA polymerases bind directly to the specific sequences of the promoters immediately upstream of the initiation site only 5 in prokaryotes. In contrast, in eukaryotes, the initiation of transcription is much more complex, and the RNA polymerases do not bind directly to the core promoter sequences. Moreover, the transcription of eukaryotic genes requires three RNA polymerases: RNA polymerase I transcribes the ribosomal genes, while RNA polymerase III – genes for small RNAs, including tRNAs. The basic transcription machinery for the protein-encoding genes is a ribosome-sized entity with two basic components, transcription factor II D (TFIID) and the polymerase II (Pol II) holoenzyme, which consists of the core RNA polymerase II, basal transcription factors, and other related proteins. TFIID binds specifically to TATA boxes (promoter elements), while the Pol II holoenzyme binds to the TFIID and thus interacts specifically with promoters. The core promoter is essentially inactive in its ground state, and almost all eukaryotic genes require multiple activators upstream and downstream of the promoter for efficient transcription.(19) Thus, transcription is the only upstream element of gene expression that allows for extensive regulation and consequently precise control, and as such, it is the most promising target for the modulation of gene expression and, potentially, a particular disease. In particular, transcription factors control genetic and epigenetic deregulation inherent to cancer, responsible for the establishment of the micromilieu, rich in oncogenes and poor in tumor suppressors.(20) Hence, specific focus in the regulation of transcription of particular cancer-related genes is shifting towards corresponding transcription factors, making them important therapeutic targets. 6 I.2. Classification of Transcription Factors and Modulation of Their Activity. Transcription factors involved in cancer can be loosely subdivided into three groups. First group, steroid hormone receptors (estrogen receptors in breast cancer and androgen receptors in prostate cancer) upon activation by a ligand, translocate into the nucleus and control gene expression levels by binding to a specific DNA sequence.(21) Second group, resident constitutively expressed nuclear proteins (bZip and ETS families), activated by serine kinase cascades.(22) Third group, latent cytoplasmic factors (STATs, NF-kB, Notch, and WNT-β-catenin signaling pathway), activated by various kinases at the cell surface, and then translocated into the nucleus (23). Another classification separates the transcription factors based on their role as oncogenes: overexpression, contribution of a DNA-binding domain through fusion with other proteins, or mutation. Some transcription factors exhibit most (MYC) or all of these properties (ETS family) simultaneously.(23) A number of oncogenes and tumor suppressor genes encode transcription factors, and any functional abnormality in their expression leads to the aberrant expression of downstream genes associated with tumor progression.(24) Despite the fact that the versatility of the roles of transcription factors in the regulation of cancer progression makes them both promising and desirable targets, they were considered “undruggable” targets for a long time. Only recently the general understanding of transcription and advances in computational modeling, biochemistry, and genetics have evolved to a point when this challenge could be addressed. Modulation of protein-protein or protein-DNA interfaces of transcription factors by small molecules is difficult, as most 7 of these interfaces employ large, shallow surfaces often lacking well-defined structural features, such as clefts or pockets suitable for binding of small molecules.(25). Many transcription factors are parts of signaling cascades initiated by cell surface receptors and mediated by protein kinases. Therefore, targeting kinases or cell surface receptors could be an effective way of modulating corresponding downstream transcription factors. This has been demonstrated in a report where the regulation of a family of basic helix-loop- helix (bHLH) transcription factors involved in cell differentiation and apoptosis, has been performed via the inhibition of BMP cell surface receptor 2.(26) On the other hand, these upstream pathways are at times rather ubiquitous. For example, numerous overactive receptor tyrosine kinases (RTKs), ligands, non-receptor tyrosine kinases (NRTKs) and some downstream serine kinases lead to serine phosphorylation and activation of c-JUN, and their inhibition may cause severe off-target effects.(23) As such, the approaches that target transcription factors themselves, or their immediate binding partners, are considerably more precise. A possible point of interference with the transcription factor function is its DNA binding site, often called response element. Several approaches have been developed employing nucleic acid-based strategies. Furthermore, outstanding specificity in vitro has been reported for antisense oligonucleotides (ASO) targeting mRNA, although their poor cell permeability is frequently an issue.(27) Multiple patents have been filed reporting the use of ASO for the inhibition of the mRNA encoding transcription factors in cancer, including STAT3 activity in multiple myeloma,(28) modulation of expression of STAT1,(29) hypoxia-inducible factor 1α (HIF1α),(30) serum response factor accessory protein-1 (SAP- 1),(31) Sp1 and Sp3 in pancreatic cancer and fibrosarcomas.(32) 8 Peptide nucleic acids (PNA) developed by Nielsen in 1991 are true mimics of DNA in terms of sequence recognition and thus exhibit excellent specificity; they also form a stronger bond with DNA, than found in the DNA homodimer, but often suffer from poor water solubility.(33, 34) Small interfering RNAs (siRNAs) and short hairpin siRNAs (shRNAs) emerged as a powerful tool for a specific gene knockdown in cell culture, however their application is limited, as they need to be introduced in a lentiviral vector to facilitate transfection. Furthermore, they often trigger an innate antiviral immune response, and elicit an off-target effect by stimulating the interferon (IFN)-inducible genes via induction of type I IFN and binding to toll-like receptors 3 and 7.(35, 36) Polyamides containing N-methylimidazole and N-methylpyrrole amino acids are designed to target specific predetermined sequences of DNA via localization in the minor groove and formation of a network of hydrogen bonds with both DNA strands of the region of interest.(37) Such polyamides have the ability to regulate transcription by directly interfering with the binding of transcription factors to their response elements,(38) however their cellular uptake and pharmacokinetics present a problem due to their structure often violating Lipinski’s rule. In order to improve the uptake, polyamides are often made short, targeting 6-8 bp which raises questions of the genome-wide specificity and off-target effects.(39) Upon translocation into the nucleus, most of the activated transcription factors regulate the gene expression by binding to DNA as a homo- or heterodimer, and this property was exploited in the development of specific dimerization inhibitors, in particular, for 9 STAT3,(40, 41) and MYC-MAX systems.(42-44) Additionally, transcription factors interact with a diverse family of coactivators that have activity of histone acetylase and lead to chromatin remodeling, thus providing another potential point of interference with their function. In particular, the large coactivator p300 and its ortholog CREB binding protein (CBP) are of particular interest as they are key mediators of the response to hypoxia in metazoan cells; they act through recruitment of their biding partners, hypoxia-inducible factors (HIFs).(45) 10 I.3. Hypoxia in Cancer. Ionizing radiation induces apoptosis via production of free radicals that damage genomic DNA. Molecular oxygen, trapped by these free radicals of DNA, causes permanent oxidative damage to the DNA strands, and it was found that lack of oxygen diminishes the response to the ionizing radiation. Anoxic tissue has been shown to be from 2 to as many as 12 times less responsive to radiation compared to the cells in the atmosphere of pure oxygen.(46) Maintenance of the oxygen homeostasis in mammalian cells is the key process to achieve a desired balance between the oxygen consumption for the essential metabolic reactions, and the prevention of the oxidative damage to cellular biomolecules. Oxygen is completely consumed by the time it diffuses as far as 100–180 µm away from the end of the nearest capillary to cells.(47) The proliferation rate of cancer cells exceeds the rate of the vascular supply growth, and thus results in a poorly vascularized micromilieu, featuring oxygen deprivation, nutrient starvation, and low pH.(48) Chronic hypoxia, defined as the condition where the partial pressure of O2 in the tissue is less than 2%,(49) is a hallmark of almost all solid tumors. In order to survive in such hypoxic microenvironment, tumor cells implement an adaptive switch to glycolysis, express angiogenic factors to promote neovascularization, up-regulate invasion and migration related factors, become invisible to the immune system, and replicate limitlessly,(50, 51) resulting in resistance to conventional treatments, such as radiotherapy and chemotherapy.(52) Adaptive responses to such reduced oxygen levels are primarily mediated by the family of hypoxia-inducible factors, which therefore play an essential role in the maintenance of oxygen homeostasis in metazoan organisms.(53, 54) 11 I.4. HIF Structure and Regulation of Activity. Hypoxia-inducible factors (HIFs) are heterodimeric transcription factors involved in key physiological and pathophysiological processes related to a hypoxic response. In 1991, Semenza and coworkers first characterized the promoter sequence of hypoxia-dependent genes, termed hypoxia response element (HRE). It is located in the 3´-enhancer region of erythropoietin (EPO), a hormone, which stimulates erythrocyte proliferation and which is responsible for stimulating the proliferation and differentiation of erythrocytic progenitors in bone marrow.(55) HIF1 was then discovered in another study, where Semenza et al. have demonstrated that the 50-nt hypoxia-inducible enhancer of EPO binds several nuclear factors, one of which is constitutively expressed and another induced by hypoxia via de novo protein synthesis.(56) HIF1 is therefore a heterodimer under hypoxia, consisting of an oxygen-sensitive α-subunit (HIF1α) and a constitutively expressed β subunit (HIF1β), also known as aryl hydrocarbon receptor nuclear translocator, ARNT.(57) Importantly, HIF1 exists as a heterodimer before binding to DNA.(58) Both subunits of HIF1 belong to the bHLH/PAS family of transcription factors, and as such, each of them has one basic- helix-loop-helix (bHLH) domain and two PER-ARNT-SIM (PAS-A, PAS-B) domains. The sequences similar to those of the PAS domains, were originally found in PER (a circadian rhythm factor), ARNT, and SIM (an analog of HIF1 in D. melanogaster) proteins. The HLH domains mediate dimerization of HIF1α and HIF1β, necessary for their basic domains to bind to the core sequence 5´-(A/G)CGTG-3´ in the HRE.(59) Unlike in the case of most other dimers of bHLH/PAS proteins, PAS domains of both HIF 1 subunits participate in the dimerization: while the amino acids 1-166 encompassing domains bHLH 12 and PAS-A are sufficient for heterodimerization, binding improves considerably when amino acids 1-390 participate (bHLH, PAS-A, and PAS-B).(58) The C-terminal half of HIF1 contains two transactivation domains, an N-terminal activation domain (NTAD) and C-terminal activation domain (CTAD), loosely spanning amino acids 531-584 and 776-786, respectively.(60) The adenovirus E1A-binding protein p300 and CREB-binding protein (CBP) are homologous transcriptional adaptor proteins often called coactivators or cofactors that are active in multiple transcriptional events. They act, at least in part, as signaling conduits between specific DNA-bound transcription factors and the basal transcriptional machinery.(45) HIF1α CTAD domain directly interacts with the cysteine-histidine rich 1 (CH1) domain of the central integrating coactivator p300/CBP hypoxia to activate the transcription of hypoxia inducible genes, while recruiting accessory coactivators SRC-1 and TIF-2 that interact with HIF1α to enhance its hypoxia-dependent transactivation function.(60) Additionally, in hypoxic cell extracts, HIF1 DNA binding was reversibly abolished by sulfhydryl oxidation, while thioredoxin-dependent redox factor Ref-1 was demonstrated to remedy this process and therefore potentiate the HIF1α-dependent transcription under hypoxia, underscoring the critical importance of redox chemistry at Cys800 for the HIF1α activation and its interaction with p300/CBP and SRC-1.(61) Regarding details of the protein-protein interaction between HIF1α and coactivators, four residues in the CTAD of HIF1α, Leu795, Cys800, Leu818, and Leu822, have been identified as crucial for HIF1α transactivation in mammalian systems. Moreover, data from residue substitution experiments indicate the stringent necessity of leucine and hydrophobic cysteine for C-terminal activation domain function. In the CH1 domain of 13 p300, residues Leu344, Leu345, Cys388, and Cys393 have been demonstrated to be essential for the functional interaction. The HIF1α-p300 interaction is deemed to rely upon a leucine-rich hydrophobic interface that is regulated by the hydrophilic and hydrophobic sulfhydryls of HIF1α (Cys800) and p300 (Cys388 and Cys393).(62) I.4.1. Oxygen-Dependent HIF Regulation. HIF1α was found to have an exceptionally short half-life (ca. 5 min) under normoxia (63) due to an oxygen-dependent degradation domain (ODDD), incorporating proline residues Pro402 and Pro564.(64, 65) In mammalian cells, three prolyl hydroxylase (PHD) isoforms, PHD1, PHD2, and PHD3, have been identified and shown to hydroxylate in vitro these key proline residues,(66) however only one of them, PHD2, was demonstrated to be key to setting the low steady-state levels of HIF1α under normoxia.(67) Prolyl hydroxylases belong to the Fe(II)- and 2-oxoglutarate-dependent dioxygenase family. They catalyze the process of hydroxylation by transferring one oxygen atom to 2-oxoglutarate yielding succinate and carbon dioxide, and another to Fe(II) forming a putative Fe(IV)-oxo intermediate. This intermediate in turn abstracts a hydrogen atom from the target proline residue, producing a hydroxylated Fe(III) and a proline radical. The proline radical is then oxidized by the Fe(III) hydroxide, restoring Fe(II) and yielding hydroxyproline.(68) Oxidized HIF1α is recognized by von Hippel-Lindau tumor suppressor protein (pVHL),(69) which subsequently forms a complex with elongins B and C, cullin 2 (Cu2), and finger protein Rbx1 leading to the hUbc5a ubiquitination of HIF1α and its subsequent rapid proteasomal degradation.(70) 14 HIF1α CTAD is subject to regulation by hydroxylation as well. Asparaginyl hydroxylase factor inhibiting HIF (FIH1), a Fe(II)-dependent enzyme that uses molecular O2 to modify its substrate, interacts with both VHL and the HIF1α CTAD to mediate transcription repression via recruitment of histone deacetylases. Hypoxic induction via the HIF1α CTAD occurs through abrogation of hydroxylation of a conserved asparagine Asn803 in the CTAD.(71, 72) Hydroxylation at Asn803 prevents the recruitment of p300/CBP coactivator.(73) HIF1α can also be acetylated at Lys532 out of the six available lysine residues by the arrest- defective 1 (ARD1) protein. Acetylation positively contributes to the binding between HIF1α and pVHL, and ultimately promotes HIF1α ubiquitination. Thus, acetylation of HIF1α by ARD1 is also critical to its proteasomal degradation. The activity of acetyltransferases does not depend on oxygen; hence, ARD1 may be active and acetylate HIF1α regardless of oxygen conditions. However, expression of ARD1 is repressed by hypoxia, which may diminish the rate of HIF1α acetylation in hypoxia as compared to normoxia.(74) The mitochondrial deacetylase sirtuin‐3 (SIRT3) has a tumor suppressor function, as SIRT3 −/− cells were reported to have an aberrant intracellular metabolism, decreased mitochondrial integrity, and a tumorigenic phenotype relative to wild‐type cells.(75) SIRT3 counteracts the switch to anaerobic metabolism under normoxia (Warburg effect), promoted by ROS generation and HIF1α stabilization, by the virtue of suppression of ROS and succinate production resulting in PHD-dependent HIF1α degradation. The likely modes of SIRT3 action is deacetylation and activation of the antioxidant enzyme 15 superoxide dismutase, interference with some electron transport chain (ETC) components, and activation of the tricarboxylic acid (TCA) cycle enzymes.(76) To summarize, HIF1 mediates the essential homeostatic response to low oxygen levels in all metazoans. Normal oxygen levels stimulate rapid oxidative elimination of its alpha subunit by enabling the hydroxylation of Pro402 and Pro564 by the prolyl hydroxylase PHD2 and Asn803 by FIH1, resulting in the recruitment of pVHL, polyubiquitination, and subsequent proteasomal degradation of HIF1α (Figure I.3). Under hypoxia, HIF1α translocates into the nucleus, where it forms a dimer with its beta subunit HIF1β (ARNT), and binds to a hypoxia response element (HRE), a DNA sequence 5´-CGTG-3´ present in the promoter region of the hypoxia-dependent genes.(77) Figure I.3. Oxygen-dependent regulation of HIF1α signaling pathway. Adapted from references (60), (78), and (79). 16 I.4.2. Oxygen-Independent HIF Regulation. Oxygen- and pVHL-dependent pathway are not uniquely responsible for the HIF1α degradation.(80) Posttranslational phosphorylation of HIF1α is conducted by mitogen activated protein kinases (MAPK) p42 and p44 at residues Ser641 and Ser643,(81) and it is sufficient to promote HIF1α nuclear accumulation and transcriptional activity by blocking its major exportin CRM1-dependent nuclear export activation.(82) The phosphorylated HIF1α is the major form that binds ARNT, and ectopically expressed ARNT enhances HIF1α phosphorylation in a concentration-dependent manner. In contrast, the dephosphorylated HIF1α primarily binds to p53. Depletion of the dephosphorylated HIF1α suppresses p53 induction, subsequent apoptosis, and hypoxia-induced nuclear accumulation of HDM2, a negative regulator of p53. As a result, functions of HIF1α vary with its phosphorylation status and dephosphorylated HIF1α mediates apoptosis by binding to and stabilizing p53, while phosphorylated HIF1α exerts an anti-apoptotic function.(83) Glycogen synthase kinase 3 (GSK3), which consists of two isoforms (α and β), is phosphorylated and inactivated by AKT. GSK3β overexpression results in HIF1α phosphorylation and prolyl-hydroxylation-independent and pVHL-independent HIF1α ubiquitination and proteasomal degradation.(84) Insulin and insulin-growth factors IGF-1 and IGF-2 (most highly up-regulated gene in colon cancer) induce expression of HIF1α, which is required for expression of genes encoding IGF-2 and IGF-binding proteins IGFBP-2 and IGFBP-3, thus contributing to an autocrine growth factor loop.(85) Expression of HIF1-dependent gene encoding EPO is inhibited by the proinflammatory cytokines interleukin-1β (IL-1β) and tumor necrosis factor-α (TNFα). However, these cytokines do not affect expression of another HIF1- 17 dependent gene encoding vascular endothelial growth factor (VEGF) or the mRNA levels of HIF1α, but do increase levels of luciferase reporter expression in hypoxia. As such, cytokine-induced inhibition of EPO production is not related to impairment of HIF1 function, suggesting HIF1 involvement in the immune responses via the induction of its DNA-binding activity by IL-1β and TNFα.(86) Exposure of human osteosarcoma cells to NiCl2 leads to marked induction of HIF1,(87) and activation of phosphatidylinositol 3-kinase (PI3K), AKT, and p70 S6 kinase (p70 S6k ). Furthermore, inhibition of PI3K and AKT impairs nickel-induced HIF1 transactivation, whereas inhibition of p70 S6k does not seem to affect HIF1 transactivation induced by nickel compounds. Consistent with HIF1 transactivation, inhibition of the PI3K/AKT pathway causes dramatic inhibition of nickel-induced expression of Cap43 protein, necessary for p53-mediated caspase activation and apoptosis. As such, nickel compounds induce HIF1 transactivation and Cap43 protein expression through a PI3K/AKT-dependent and p70 S6k - independent pathway.(88) However, PI3K–AKT pathway is intricately linked to HIF1 regulation also by mediating the response to insulin and IL-1, and through the regulation of HIF1α protein degradation. Thus, the modulation of the PI3K–AKT pathway and its role in HIF1α regulation during hypoxia remains controversial and is highly context dependent. It has been suggested that the PI3K pathway might be activated by short-term hypoxia but inhibited when hypoxia is maintained for prolonged time.(89-91) Molecular mechanism that controls oxygen-independent degradation of HIF1α has remained elusive until recently, when the role of receptor of activated protein-C kinase-1 (RACK1) was established. HSP90, a heat shock chaperone protein, associates with HIF1α and induces conformational changes, stabilizing HIF1α and promoting its dimerization 18 with HIF1β. Geldanamycin, a HSP90 inhibitor, has been shown to induce the degradation of HIF1α even in cell lines lacking functional pVHL (renal carcinoma cell line). RACK1 competes with HSP90 for binding to HIF1α, and then recruits Elongin C and other components of E3 ubiquitin ligases, thus facilitating HIF1α ubiquitination and degradation in an oxygen-independent manner.(92) Human rhomboid family-1 (RHBDF1) protein is a member of a large family of nonprotease rhomboids, whose function is largely unknown. RHBDF1 expression was shown to be highly elevated in breast cancer, and strongly correlated with further progression of a disease, metastasis, poor response to chemotherapy, and poor prognosis. RHBDF1 was demonstrated to promote HIF1α stabilization resulting in its increased nuclear accumulation. It turns out, RHBDF1 interacts with RACK1 in breast cancer cells and prevents RACK1-assisted oxygen- independent HIF1α degradation.(93) Murine double minute-2 (MDM2) oncoprotein can cause polyubiquitination of HIF1α, followed by its binding to p53 and subsequent degradation. Hence, p53 directly interacts with and limits hypoxia-induced expression of HIF1α by promoting MDM2-mediated ubiquitination and proteasomal degradation under hypoxic conditions.(94) However, there is much controversy regarding the role of MDM2, as it has been demonstrated to participate in HIF1α stabilization as well.(95) Such MDM2 behavior is likely to be context- dependent, similar to the frequently observed reversal of function performed by the members of the HIF family (vide infra). Downstream target of Akt, mammalian target of rapamycin (mTOR) facilitates translation of HIF1α by phosphorylating the eukaryotic initiation factor 4E binding proteins (4E-BP), 19 which in turn decreases their affinity to eukaryotic initiation factor 4E (eIF-4E), an mRNA cap binding protein. Then eIF-4E binds to HIF1α mRNA, and initiates translation.(96) To summarize, oxygen-independent regulation of HIF1α implies several intricate pathways and is not entirely understood, but considerable progress has been made toward elucidation of this pathway (Figure I.4). For instance, GSK3β phosphorylates HIF1α leading to its polyubiquitination, as well as RACK1, which binds to HIF1α as a dimer when HSP90 is inhibited and recruits components of the E3 ligase complex through a process facilitated by SSAT1. GSK3β is inactivated by the PI3K pathway, which is itself activated by growth factors and Ni 2+ ions, and inhibited by serum withdrawal or the phosphatase and tensin homolog (PTEN). Calcineurin A inhibits RACK1-dependent degradation of HIF1α in a calcium-dependent manner by inhibiting RACK1 dimerization; RHBDF1 protein interacts with RACK1 as well, and inhibits its binding to HIF1α. Additionally, DNA-binding activity of HIF1α is positively regulated by IL-1β and TNFα. 20 Figure I.4. Oxygen-independent regulation of HIF1α signaling. Adapted from references (91, 92). 21 I.5. The HIF Family of Transcription Factors. Figure I.5. Domain structure of HIF proteins. Adapted from reference (97). Another member of the HIF family, HIF2α, also known as endothelial PAS 1 (EPAS1) protein, HIF-like factor (HLF), HIF-related factor (HRF), and member of the PAS superfamily 2 (MOP2), is closely related to HIF1α. HIF2α shares 48% of amino acid sequence identity with HIF1α, and consequently a number of structural and biochemical similarities: bHLH, PAS, ODD, and CTAD domains (Figure I.5), heterodimerization with HIF1β, and binding to HREs. HIF2α is, however, predominantly expressed in the lung, endothelium, and carotid body, unlike the ubiquitously expressed HIF1α.(98, 99) HIF2a was found to have an opposite effect compared to HIF1α in renal cell carcinoma (RCC). Overexpression of HIF2α enhances the growth of experimental tumours derived from RCC 22 cells, while overexpression of HIF1α retards the growth of similar RCC-derived experimental tumours. In clinic, RCC exhibits an unusual preference for the HIF2α over HIF1α expression.(100) However, in hepatocellular carcinoma (HCC), HIF2α was found to exert a tumor-suppressing effect: inhibition of HCC cell and tumor growth and induction of high levels of apoptosis and pro-apoptotic proteins. HIF2α inhibits transcription factor dimerization partner 3 (TFDP3), which in turn was found to bind with E2F transcription factor 1 (E2F1), a fundamental biological regulator, responsible for the induction of p53- dependent and independent apoptosis (Figure I.6A).(101) As such, HIF1α and HIF2α have opposing and occasionally overlapping roles in tumor cells and tumor-associated macrophages (TAMs) (Figure I.6B).(102) Figure I.6. HIF2α signaling. (A) Hallmarks. Adapted from reference (101). (B) Schematic representation of the differences between HIF1α and HIF2α signaling. Adapted from reference (102). By contrast, the third member of the HIF family HIF3α, also known as inhibitory PAS domain protein (IPAS), acts as a dominant negative regulator of HIF1α- and HIF2α- mediated transcription. It is found in high levels in the thymus, cerebellar Purkinje cells, 23 the corneal epithelium of the eye, and also found in lung, heart, and kidney. HIF3α open reading frame encodes a 662-amino acid protein with a predicted molecular weight of 73 kDa. The N-terminal bHLH-PAS domains of HIF3α share a 57% and 53% amino acid sequence identity with those of HIF1α and HIF2α, respectively. The C-terminus of HIF3α contains only 36 amino acids, 61% identical to the ODD domain of HIF1α. HIF3α also forms a dimer with ARNT, and binds to HRE,(80) albeit it has no known role as a transcription factor. HIF3α contains no endogenous transactivation function, however it shows dominant negative regulation of HIF1α function possibly by sequestration of HIF1β from HIF1α, manifested, in particular, in down-regulation of both basal and hypoxia- induced expression levels of VEGF.(49, 73, 103) Similar to the HIF alpha subunit, ARNT also has three paralogs, ARNT1, ARNT2, ARNT3.(73) ARNT2 is very similar to ARNT, both structurally and functionally. Both isoforms are mutually interchangeable, but ARNT2 is overexpressed in neurons and kidney cells, while ARNT expression is ubiquitous. As such, ARNT2/HIF1α heterodimers mediate transcriptional responses to oxygen deprivation in the central nervous system.(104) ARNT3 is also a structural and functional analog of both ARNT and ARNT2; it can form functional dimers with HIF1α, and bind to HRE. ARNT3 expression is limited to brain, skeletal muscle, and certain stages of embryogenesis. Notably, ARNT3 contains an activation domain at the C-terminal region and a repression domain between the PAS- A and PAS-B regions, which serves as an intramolecular repressor of the activation domain. Albeit ARNT also has an activation domain in its C-terminal region (identical to the one found in ARNT2), the two sequences are very different from each other. (105) 24 I.6. HIF Function. HIF is a master regulator of cellular and developmental O2 homeostasis, and it is implicated in essential developmental processes and in various disease states, including cancer, myocardial and cerebral ischemia, and chronic obstructive pulmonary disease.(106) I.6.1. Development. HIF1α −/− embryonic stem cells (ESCs) and cystic embryoid bodies have a markedly decreased vascular endothelial growth factor mRNA expression even in hypoxia. In embryos, complete deficiency of HIF1α results in developmental arrest and lethality by E11, manifesting in neural tube defects, cardiovascular malformations, and marked cell death within the cephalic mesenchyme. The devopmental defects onset between E8.5 and E9.5, conincident with HIF1α expression increase in HIF1α +/+ embryos.(107) HIF1α regulates the metabolic shift towards glycolysis in cancer, termed Warburg effect, and in pluripotent ESCs during their development into the epiblast stem cells (EpiSCs).(108) ESCs and EpiSCs cells represent two morphologically and expression- wise distinct pluripotency stages of the embryonic development. In mice, ESCs switch between both oxidative phosphorylation and glycolysis, while the survival of EpiSCs entirely depends on the glycolytic metabolic pathway, manifesting in up-regulation of PDK1, LDHA, and PYGL glycolytic genes, and down-regulation of cytochrome c oxidase (COX), responsible for mitochondrial respiration. Another noteworthy observation was made for the hematopoietic stem cells (HSCs). Their adaptation to the hypoxic environment upon transplantation requires a similar significant metabolic shift. This process is modulated by the Meis1, which belongs to the Hox family of homeobox genes 25 that encode DNA-binding transcription factors. Meis1 is expressed in the most primitive hematopoietic populations and is down-regulated upon differentiation. In HSC, Meis1 controls the metabolic shift by performing the transcriptional activation of HIF1α.(109) I.6.2. Ischemia. Atherosclerosis leads to the impairment of the blood flow, lack of an adequate tissue blood supply, and ultimately ischemia. Prolonged exposure of the myocardium to the diminished levels of oxygen and glucose adversely affects myocardial viability, culminating in necrosis and an infarction. In response to acute ischemia and early infarction, mRNA and protein levels of HIF1α are increased in the heart, leading to the up-regulation of VEGF, and the onset of angiogenesis and vascular remodeling.(110) Analogous to myocardium, the blockade of a middle cerebral artery induces the focal cerebral ischemia. HIF1α was demonstrated to be up-regulated in the viable surrounding tissue, resulting in the up-regulation of some of the known HIF1α downstream targets: VEGF, glucose transporter 1, the glycolytic enzymes aldolase A, lactate dehydrogenase A, phosphofructokinase L, and pyruvate kinase M.(111) These data implicate HIF1α dependent neovascularization and switch to glycolytic metabolism as primary mechanisms of neuron survival in ischemia, and correlate well with a model in which hypoxia-induced HIF1α associates with and prevents the degradation of p53 protein, responsible for the induction of the apoptosis of cortical neurons.(112) 26 I.6.3. Inflammation. During inflammation and injury, the influx of the pathogens and the cells governing the response and their active oxygen consumption, frequently results in hypoxia, anoxia, hypoglycemia, acidosis, and abundant presence of ROS. This process is often accompanied by swelling, edema, and vascular insult, amplifying the lack of oxygen and pronounced shift in the microenvironment. Mice with a myeloid cell HIF1α conditional deletion had an impaired inflammatory response in a collagen-induced arthritis model. HIF1α was demonstrated to be activated in the affected tissues of patients with inflammatory disorders, such as rheumatoid arthritis, dermatomyositis, neonatal lupus syndrome, and atherosclerosis.(113) Overexpression of HIF1α in human mast cells leads to the up-regulation of key inflammatory cytokines IL-6, IL-8, IL-12, and TNFα, which are necessary for macrophage activation and phagocytic function, as well as lymphocyte proliferation.(114) Hypoxia activates κB kinase-β (IKKβ), leading to phosphorylation-dependent degradation of nuclear factor of kappa light polypeptide gene enhancer in B cells alpha (IκBα) and subsequent liberation of NF-κB.(115) In turn, NF-κB is a critical transcriptional activator of HIF1α.(116) In pathogen-primed phagocytes, NF-κB can also be activated via p44/42 MAPK signaling cascade, initiated upon the recognition of the pathogen-associated molecular patterns such as lipopolysaccharides by the toll-like receptors (TLRs).(117) Skin cells are hypoxic under normal conditions, and heavily depend on HIF regulation for controlling necrotic infections. HIF1α mediates the regulation of keratinocyte cathelicidin 27 production which is crucial for cutaneous defense against infection by the invasive pathogen group A streptococci.(118) Finally, contrary to action of HIF1α that drives VEGF production and angiogenesis (vide supra), the production of HIF2α in macrophages blocks angiogenesis by inducing the expression of soluble VEGF receptor-1, inactivating biologically active free VEGF. It is speculated that HIF2α might have evolved subsequent to HIF1α to regulate VEGF in response to hypoxia, to regulate the host vascular network.(119) The role of a particular member of the HIF family in the inflammatory response seems to depend on the tissue, as in hypoxic gastrointestinal mucosa and epithelium, HIF1α also has been demonstrated to have a robust anti‐inflammatory effect.(120) I.6.4. Apoptosis. HIF1α is involved in hypoxia induced apoptosis. During environmental stress or DNA damage, p53 induces programmed cell death by regulating proteins such as Bax, or it can cause growth arrest, mediated by p21. HIF1α can prevent apoptosis via binding to the p53 ubiquitin ligase Mdm2, thereby preventing the degradation of p53.(121) Under hypoxia human carcinoma cell lines, endothelial cells, and macrophages overexpress proapoptotic proteins BNIP3 (BCL2/adenovirus E1B 19 kDa interacting protein 3) and its homolog, NIX. BNIP3 induces apoptosis by binding to and inhibiting the antiapoptotic proteins Bcl-2 and Bcl-xL.(122) Transcription of BNIP3 is impaired in HIF1α deficient cells, and since BNIP3 harbors an HRE promoter, it is reasonable to conclude that BNIP3 is the downstream target of HIF1α.(123) 28 However, as mentioned before, when we compared different members of the HIF family, the function of HIF1α sometimes exhibits tissue-specific switch to its complete opposite. In particular, in pancreatic cells HIF1α is up-regulated in normoxia by the activation of PI3K/AKT pathway, and they showed higher resistance to the hypoxia- and hypoglycemia- induced apoptosis compared to the other cell lines with low basal levels of HIF1α.(124) I.6.5. Cancer. Chronic hypoxia is a hallmark of almost all solid tumors, and as such, HIF-inducible transcriptional pathway, being the master regulator of oxygen homeostasis, plays a critical role in the progression of tumorigenesis. It regulates a number of specific mitogens that promote tumor growth and invasion, and the establishment of oncogene-rich, tumor suppressor-poor microenvironment.(125) In the beginning of the development of a solid tumor, the limited O2 perfusion from nearby host vessels limits its growth to a few cubic millimeters as the cell division is balanced by the cell death. Therefore, hypoxic tumor cells overexpress VEGF, an essential factor for the initiation and progression of angiogenesis, and a number of other proteins dependent on HIF1α: nitric oxide synthases involved in governing vascular tone; growth factors such as angiopoietins; fibroblast growth factors and their receptors; matrix metabolism proteins, including matrix metalloproteinases; plasminogen activator receptors and inhibitors. However, the rapid neovascularization results in a dysfunctional network of blood vessels that are unable to adequately perfuse the whole tumor, thus preserving the tumor cells in a hypoxic state.(125, 126) 29 Amongst other HIF-dependent growth factors are platelet-derived growth-factor B chain (PDGFβ) and transforming growth factor-alpha (TGFα). The latter is suspected of being a HIF target because it is induced by hypoxia and down-regulated in the presence of a dominant-negative HIF1α mutant. Epidermal growth-factor receptor (EGFR), which is the receptor for TGFα, is also up-regulated in hypoxia, thereby establishing a potential autocrine loop.(127) HIF up-regulation is a requirement for the predisposition to renal cell carcinoma (RCC), hemangioblastoma, and pheochromocytoma in the VHL disease, caused by a mutation in the VHL tumor-suppressor gene and consequently a dysfunctional pVHL protein.(127) Specifically, it is HIF2α that is responsible for the manifestation of the tumor phenotype in RCC. Conversely, single nucleotide point mutations in HIF2α have been linked with the risk of RCC development as well. On the other hand, deletion of chromosome 14q harboring HIF1α, is frequently observed in RCC and interestingly associated with a poor prognosis, suggesting that HIF1α acts as a tumor suppressor in RCC.(128) As it turns out, such HIF1α behavior is not restricted to RCC, and was observed in embryonic stem cells, where HIF1α can initiate apoptosis by inducing high concentrations of proapoptotic Bcl-2- binding proteins, such as BNIP3 and NIX, inhibiting the antiapoptotic effect of Bcl-2, or by stabilizing wild-type p53.(129) In acute myeloid leukemia, HIF1α interacts with CCAAT/enhancer binding protein alpha (C/EBPα) and Runx1/AML1, two hematopoietic transcription factors, thus increasing their transcriptional activities.(130) Overexpression of both most prominent members of the HIF family, HIF1α and HIF2α, is associated with negative prognosis and increased mortality in cancer patients, however, as mentioned before, in some cancers/tissues these factors exert opposite effects. Poor patient 30 outcomes are more strongly associated with HIF2α than HIF1α overexpression in CNS, colorectal, non-small cell lung, kidney, and head and neck tumors, and the augmented proliferation is speculated to stem from the increased expression of TGFα and Cyclin D1. Additional effects on the tumor phenotype might result from HIF2α-mediated induction of the stem cell factor Oct4 and promotion of c-Myc transcriptional activity. Similarly, the effects of HIF1α and HIF2α gene disruption are substantially different: HIF1α knockout consistently leads to impaired cardiac and vascular development and E10.5 lethality (vide supra),(107) while the loss of HIF2α results in embryonic and perinatal lethality due to the defective phenotypes, including bradycardia and vascular defects, impaired lung maturation, multi-organ failure, and mitochondrial dysfunction.(131) Thus members of the HIF family of proteins regulate the expression of hypoxia-inducible genes involved in angiogenesis (VEGF, NOS2, EGFR), glucose metabolism (GLUT1, GLUT3), cell proliferation (IGF2, TGFβ), cell motility and invasion (cMet, LOX), apoptosis (p53, Bcl-2), and inflammation (IL-6, IL-8, TNFα), in a tissue- and cancer- dependent manner, making them challenging yet attractive therapeutic targets. 31 I.7. Key HIF-Dependent Genes. HIF family transcriptionally activates and is a master regulator of several hundred oxygen dependent genes. HIF regulation has an intricate mechanism, and some of these genes are regulated ubiquitously, whereas others require a particular microenvironment of a certain tissue. Additionally, it is necessary to reinforce yet again that despite their structural and mechanistic similarity, HIF1α and HIF2α tend to regulate different groups of genes in a different and sometimes completely opposing fashion. Both hypoxia and HIF1α up- regulated 245 and down-regulated 325 genes,(132) out of which more than 60 putative direct HIF1 target genes have been identified on the basis of one or more of the following restrictions: identification of a cis-acting hypoxia-response element that contains a HIF1 binding site1;(77) loss of hypoxia-induced gene expression in HIF1α-null cells or cells treated with siRNA that targets HIF1α mRNA; increased expression in von Hippel–Lindau (VHL)-null cells or in cells transfected with a HIF1α expression vector.(133) Additionally, both HIF1α and HIF2α together affected more than 100 of such genes.(134) These genes regulate a number of cellular processes involved in tumor angiogenesis, metabolism, metastasis, invasion and other critical processes of tumor growth and progression (Figure I.7), HIF2α preferentially activating VEGF, TGFα, lysyl oxidase (LOX), Oct4, and Cyclin D1.(135) 32 Figure I.7. Genes involved in tumor onset and progression that are activated by HIF. Adapted from reference (49). I.7.1. VEGF Wound healing, ischemic and inflammatory diseases, and cancer require the formation of new blood vessels (or neovascularization), which proceeds via two distinct stages. “Vasculogenesis”, the initial events of the vascular growth, involve the migration of angioblasts (endothelial cell precursors) to discrete locations, their differentiation and assembly into solid endothelial cords, later forming a branching network with endocardial 33 tubes. The second phase, known as “angiogenesis”, implies the subsequent growth, expansion and remodeling of these primitive vessels into a mature vascular network.(136) A positive correlation was established between tumor vascularization, metastasis, and Gleason’s score in breast, lung, and prostate cancer in human patients.(137) It was also shown that higher density of the blood micro vessels corresponds to lower survival rates in the patients with breast carcinoma, non-small cell lung carcinoma, head-and-neck squamous carcinoma, and prostate carcinoma. Interestingly, microvessel cell density, intratumoral endothelial cell proliferation, and tumor cell proliferation were shown to be completely independent of each other, suggesting these processes could employ separate mechanisms.(138) Angiogenesis is governed primarily by the vascular endothelial growth factor (VEGF, also known as VEGFA or vascular permeability factor),(139) the most potent endothelial- specific mitogen, governing a hypoxic response by directly recruiting endothelial cells into the hypoxic avascular area and stimulating their proliferation.(140) VEGFA is a basic, heparin-binding, homodimeric glycoprotein of 45kDa that belongs to a VEGF growth factor family, which has a total of five known members: placental growth factor (PLGF), VEGFA, VEGFB, VEGFC, and VEGFD (also known as c-Fos-induced growth factor, FIGF),(141) each factor is having several different splice variants. These VEGF ligands bind in an overlapping pattern to the three receptor tyrosine kinases, VEGFR1, VEGFR2, and VEGFR3, and to heparin-sulfate proteoglycan (HSPG) co-receptors. VEGFR1 is a positive regulator of monocyte and macrophage migration, and has been described as a positive and negative regulator of VEGFR2 signaling capacity. Negative regulation is exerted by an alternatively spliced soluble VEGFR1 variant that binds to VEGF, preventing 34 its binding to VEGFR2. VEGFR2 is implicated in all aspects of normal and pathological vascular-endothelial-cell biology, whereas VEGFR3 is important for lymphatic- endothelial-cell development and function.(142) VEGFA, VEGFB, and PLGF bind to VEGFR1, VEGFA and VEGFE bind to VEGFR2, and VEGFC and VEGFD bind to VEGFR3. Proteolytic degradation of the human VEGFC and VEGFD allows for binding to VEGFR2, however, these factors bind to VEGF2 with lower affinity than to VEGFR3 (Figure I.8).(143) The VEGF receptor protein–tyrosine kinases consist of an extracellular component containing seven immunoglobulin-like domains, a single transmembrane segment, a juxtamembrane segment, an intracellular protein–tyrosine kinase domain that contains an insert of about 70 amino acid residues, and a carboxy terminal tail. Binding of growth factors to the ectodomain of their transmembrane receptors leads to receptor dimerization, protein kinase activation, trans-autophosphorylation of one subunit by the other, and initiation of signaling pathways. VEGF binds to the second immunoglobulin domain of VEGFR1 and the second and third immunoglobulin domains of VEGFR2.(144) VEGFR1 (Flt-1, fms-like tyrosine kinase-1) has 50 times higher affinity for VEGF than VEGFR2. VEGF induces a loss of tyrosine kinase activity in VEGFR1 upon binding to it, however, it does not change the expression profile in the endothelial cells. VEGFR2 (Flk- 1/KDR, Fetal liver kinase-1/Kinase insert Domain-containing Receptor) is the predominant mediator of VEGF-stimulated endothelial cell migration, proliferation, survival, and enhanced vascular permeability, and VEGF2R exhibits a robust phosphorylation activity 35 in response to its ligands, including autophosphorylation, which leads to the increase in its activity.(144) Figure I.8. The family of VEGF receptors and ligands and schematic representation of VEGF pathway. Reproduced from reference (143). VEGFA has at least four frequently detected alternative splice isoforms: VEGF121, VEGF165, VEGF189, and VEGF206. These variants are homodimers with the individual chains of varying lengths: 121, 165, 189, and 206 amino acids, respectively.(143, 145) Some of the less frequent splice variants include VEGF145, VEGF183, VEGF162 and 36 VEGF165b. From the functional standpoint, the isoforms primarily vary in their ability to bind to the cell surface HSPG co-receptors. The major isoform is VEGF165, and it defines the VEGFA properties (vide supra). VEGF189 and VEGF206 are more basic and bind to heparin with greater affinity than VEGF165, while VEGF121 is a weakly acidic polypeptide that fails to bind to heparin. Consequently, these properties define the bioavailability of the VEGFA isoforms. Solubility of VEGF165 is limited, and a significant fraction of it remains bound to the cell surface and the extracellular matrix. VEGF189 and VEGF206 are almost completely sequestered in the extracellular matrix, while VEGF121 is a freely soluble protein.(146) VEGF165 binds the co-receptors NRP-1 (neuropilin-1) and NRP-2 (neuropilin-2), whereas VEGF145 binds only NRP-2 (Figure I.8). VEGF isoforms play distinct roles in vascular patterning and arterial development, although the VEGF165 isoform is crucial for vascular development and survival of the new blood vessels.(147) Besides HIF1α transcriptional regulation, VEGF has another mechanism of regulation at the mRNA level. The 5´- and 3´-UTRs of the VEGF confer increased mRNA stability during hypoxia via binding HuR, an AU-rich element binding protein, and PAIP2, polyadenylated-binding protein-interacting protein 2.(148) Additionally, it has been shown that the 5´-UTR of VEGF mRNA contains two functional internal ribosome entry sites that maintain efficient cap-independent translation of VEGF mRNA and allow for translational regulation. As discussed before, in cancer, continuous cell proliferation without adequate oxygen and nutrient supply causes the tumor cells to become apoptotic and necrotic. Therefore, VEGFA is overexpressed in lung, breast, gastrointestinal tract, renal and ovarian solid tumors, in particular in the vicinity of a necrotic pocket, leading to the growth of new blood 37 vessels into a tumor.(140) VEGF depletion by neutralizing antibodies (such as bevacizumab, vide infra), VEGFR2 blockade by a VEGFR tyrosine kinase inhibitor, antisense oligonucleotides, and vaccination reduce tumor angiogenesis and growth in vivo.(149) VEGF also plays a role in carcinogenesis. Overexpression of SV40 Tag oncogene in insulin-producing β-cells in the multistage carcinogenesis Rip1–Tag2 (rat insulin promoter – T-antigen 2) mouse model, leads to the development of hyperplastic islet tumors in the pancreas. Conditional ablation of VEGFA from islet β-cells resulted in sever disruption of angiogenesis, carcinogenesis, and tumor growth, indicating the vital role of VEGFA in these processes.(150) Similar impairment of the tumor angiogenesis and growth was observed when a soluble VEGFR1 receptor(151) or pharmacological inhibitor of VEGFR2(152) were used. Matrix metalloproteinase 9 (MMP9) was found to promote the angiogenic switch by the proteolytic release of VEGFA from the extracellular matrix (ECM), and thus increasing its bioavailability.(153) VEGFA increases invasiveness of the tumor cells. VEGFA was found to disrupt the VE- cadherin/β-catenin complex, resulting in activation of the SRC family kinases that are responsible for the disruption of the endothelial barrier function, leading to extravasation and metastasis of the tumor cells. Inhibition of VEGF, VEGFR2, or SRC stabilizes the endothelial barrier function and suppresses tumor cell extravasation in vivo.(154) VEGFA also induces the disruption of hepatocellular occluding-delineated tight junctions, which may promote tumor invasion.(155) MMP9 was also found to be up-regulated in lung cancer, and involved in lung cancer metastasis via VEGFR1 signaling.(156) 38 Thus VEGF presents a well-defined therapeutic target, and VEGFA inhibitors are being pursued clinically. These inhibitors include small-molecule receptor tyrosine kinase (RTK) inhibitors such as PTK787, soluble receptors such as VEGF-Trap, and anti- VEGFR2 monoclonal antibodies. A significant breakthrough has been achieved with the introduction of bevacizumab (Avastin), a humanized anti-VEGF monoclonal antibody, approved by the FDA as the first-line treatment for colorectal, lung, breast, renal, and brain cancers.(157) I.7.2. MET Metastasis is a multistep process of invasive growth that involves a complex genetic program, in which the tumor cells gain access to the vasculature in the primary tumor, survive the circulation, arrest the microvasculature of the target tissue, exit from the vasculature, and proliferate in the target tissue.(158) MET (cMet) is a proto-oncogene encoding MET tyrosine kinase, a cell surface receptor for hepatocyte growth factor (HGF). MET delivers a combination of pro-migratory, anti- apoptotic, and mitogenic signals. In particular, in a healthy organism, HGF/MET pair is implicated in morphogenic processes during embryo development(159) and wound healing, for instance, myocardial protection(160) and liver regeneration.(161) It is however well established now that gastrulation and wound healing mechanisms are uncontrollably re-activated in cancers, with MET being one of the key regulators of the tumor invasive growth and metastasis.(162) Autocrine activation of cMet results in development of melanoma with a high metastatic potential.(163) 39 MET receptor is a disulfide-linked heterodimer that consists of an extracellular alpha chain and a longer beta chain. The beta chain encompasses the remainder of the MET ectodomain, the transmembrane helix and the cytoplasmic portion. The latter contains the juxtamembrane and kinase domains, and a carboxy terminal tail that is essential for downstream signaling.(164) From the mechanistic standpoint, MET is able to recruit several SH2-domain-containing signal transducers that in turn activate a number of pathways, including the GRB2–SOS– RAS–RAF–MEK–ERK axis, the PI3K–AKT cascade, SRC, STAT3 and Rho-like GTPases such as RAC1 (Figure I.9).(164, 165) MET-triggered pathways are very potent, and their endurance is secured by a network of upstream signaling co-receptors that physically associate with MET at the cell surface: CD44, α6β4 integrin, and plexins of the B family.(165) The v6 splice variant of the hyaluronan receptor CD44 is necessary to link the MET cytoplasmic tail to the actin cytoskeleton and to optimize MET-triggered activation of the RAS–ERK cascade.(166) The laminin receptor α6β4 integrin acts as a supplementary docking platform for the additional recruitment of signaling molecules that amplify activation of PI3K-, RAS- and SRC-dependent pathways.(167) Semaphorin receptors of the plexin B family transactivate MET in HGF-independent manner when stimulated by their semaphorin ligands, providing an alternative way to induce invasive growth and angiogenesis.(168) 40 Figure I.9. MET signaling pathway. Reproduced from reference (165). In various types of tumors, activation of MET is a secondary event that exacerbates the malignant properties of already transformed cells. In these cases, aberrant MET activation usually occurs through transcriptional up-regulation by other oncogenes,(165) in particular hypoxia,(169) as the promoter region of human MET contains several putative HIF1α binding sites (HREs). MET amplification correlates with cell proliferation, and tumor cells overexpressing MET are dependent on the continued activity of this oncogene for maintaining their transformed phenotype. This phenomenon, known as “oncogene addiction” implicates that human tumors up-regulating MET are likely to respond effectively to therapies targeting MET 41 and/or HGF.(165) Several inhibitors targeting the HGF/MET system have been developed in recent years, including HGF and MET biological antagonists, anti-HGF and anti-MET antibodies,(170, 171) and small molecules.(172, 173) I.7.3. LOX Lysyl oxidase (LOX) is a 32 kDa copper-dependent enzyme expressed and secreted by fibrogenic cells, catalyzing the initial steps in covalent crosslinking of specific collagen hydroxylysyl residues and elastin lysyl residues.(174) LOX oxidizes lysine residues of collagen and elastin to peptidyl α-aminoadipic-δ-semialdehyde (AAS). These aldehyde residues can spontaneously condense with vicinal peptidyl aldehydes or with ε-amino groups of peptidyl lysine to generate the covalent cross-links, thus insolubilizing collagen and elastin fibers and stabilizing the ECM and the nucleus.(175) LOX and its substrates (collagen) are up-regulated by HIF1α in hypoxia,(176) a stimulus implicated in the development of fibrosis.(177) Specifically, the putative promoter region of the LOX gene has four HRE sequences, and HIF1α was demonstrated to bind to one of them, thereby up-regulating LOX.(176) Thus LOX is a HIF-dependent gene; however, one novel tumor suppressor Pdcd4 was found to down-regulate LOX in HIF-independent manner in breast cancer.(178) There are five members in the LOX family: LOX, LOXL1, LOXL2, LOXL3, and LOXL4. The C-terminal domain of LOX is highly conserved between the family members, and exerts enzymatic activity. The N-terminal end contains pro-peptide regions with variable sequences. Cysteine-rich scavenger receptor domains, assumed to play an amine oxidase role, are found in LOXL2, LOXL3, and LOXL4.(179) LOX is synthesized as a 50 kDa 42 pro-enzyme that undergoes post-translational modification in the endoplasmic reticulum (ER) and Golgi apparatus, and it is subsequently secreted into the extracellular space, where it is processed to form the 30 kDa mature active enzyme.(180) TGF-β1 up-regulates LOX, and activates PI3K/AKT, Smad and MAPK signaling pathways. Inhibition of the key elements of these pathways, Smad3, p38-MAPK, JNK, and ERK1/2, disrupts the response of LOX transcription to TGF-β1, implicating PI3K/AKT and Smad pathways in the signal transduction between TGF-β1 and LOX.(181) Rac1 GTPase is a mediator of angiotensin II (Ang II) human cardiac fibroblasts. Ang II up-regulates LOX in a Rac1-dependent manner. In turn, connective tissue growth factor (CTGF) is regulated by Rac1 and regulates LOX, as its ablation disrupts Rac1-LOX signal transduction,(182) thereby revealing the Ang II-Rac1-CTGF-LOX signaling cascade. Opposing the effect of the ras oncogene, LOX was once postulated to act as a tumor suppressor,(174) while involvement in the advancement of metastatic phenotype later established its role as a tumor promoter. Breast cancer mortality remains relatively high and is strongly correlated with metastasis, invariably diagnosed in 30% of all patients at some stage. Melanoma and colorectal cancers incur substantial metastatic risks as well.(183) The metastatic process involves multiple steps of interactions between migratory tumor cells, stromal cells, and the extracellular matrix (ECM), ultimately resulting in increased cell motility.(184) Thus to a large extent, metastasis is driven by the tumor microenvironment, and the mechanisms by which tumor cells alter their micromilieu have become novel therapeutic targets for drug development, one such target being LOX. Breast cancer stroma was noticed to be up to 10 times stiffer than healthy tissue, further 43 implicating fibrosis-related machinery, in particular, LOX; its expression has been observed in various cancers, including breast cancer and head and neck cancer, and it is a poor prognostic marker.(185) Overexpression of LOX in mice was demonstrated to increase ECM stiffness, and promote tumor cell invasion.(186) To summarize, LOX, a HIF-dependent oncogene, has been convincingly demonstrated to cause increased tumor cell motility. Hypoxia-associated invasion and metastasis is a hallmark of poor prognosis in cancer. As such, the pursuit of upstream HIF pathway inhibition to bring down levels of LOX as the ultimate therapeutic target, is warranted. I.7.4. SLC2A1 Compared to normal cells, malignant cancer cells exhibit substantial differences in the cell- surface glycoproteins and glycolipids, leading to increase in glucose and amino acids uptake.(187) Glucose uptake is mediated by the sodium-coupled glucose transporter (SGLT) and glucose transporter facilitator (GLUT) proteins, both encoded by the solute carriers SLC5A and SLC2A gene families, respectively.(188) Overexpression of GLUT proteins is a common event in a wide variety of human cancers, such as cervical, pancreatic, lung, liver, ovarian, prostate, and breast.(189) GLUT proteins comprise 12 hydrophobic transmembrane α-helices with both carboxy and amino termini located on the inside of the cell.(190) While the other members of the GLUT family exhibit a tissue-specific expression, GLUT1, encoded by the SLC2A1, is ubiquitously expressed and is responsible for the basic glucose supply to the cells.(191) GLUT1 is especially abundant in endothelial and epithelial-like barriers of the brain, eye, peripheral nerve, placenta and lactating mammary gland, and erythrocytes.(192) GLUT1 44 is essential for development in most cells and tissues, as its deprivation during gestation is lethal, and embryos with low levels of GLUT1 experience developmental deviations and, occasionally, early death.(188) Healthy cells obtain most of their ATP through oxidative phosphorylation in mitochondria, whereas cancer cells are known to experience a metabolic switch from oxidative phosphorylation (respiration), to as much as 100 times less efficient anaerobic glycolysis (fermentation). Irreversible damage to the cellular respiratory system is deemed to be the underlying fundamental cause of the malignancy development. Chronic hypoxia is one of the primary mediators of this process, and the lack of energy that ensues the decrease in oxygen-dependent production of ATP, leads to the permanent destruction of the respiratory system and, importantly, destruction of the internal cell structure, causing dedifferentiation of cancer cells.(193) However, some glioma, hepatoma, and breast cancer cell lines possess functional mitochondria and obtain their ATP mainly from oxidative phosphorylation. Additionally, some other cancer cells can reversibly switch between fermentation and oxidative metabolism, depending on the absence or the presence of glucose and the environmental conditions.(194) As such, glucose transport plays a vital role in the survival of cancer cells under unfavorable conditions. In case if the damage to the respiratory system is permanent, the cells do not engage in oxidative phosphorylation even in the presence of oxygen, a phenomenon known as “aerobic glycolysis”, dubbed the “Warburg effect”, which simply stems from the cells being unable to switch back to respiration from fermentation due to the lack of the appropriate machinery. However, a short-term reversible suppression of respiration, known as the “Crabtree effect”, is also possible under hypoxic conditions.(194) 45 Out of all fourteen known glucose transporters, GLUT1 is also the most widely expressed isoform in human cancers, followed by the fructose transporters GLUT2 and GLUT5, which are two times less abundant.(195) Overexpression of GLUT1 was correlated with increased aggressiveness and poor prognosis in non-small cell lung carcinoma,(196) colorectal cancer,(197) increased lymph node metastasis,(198) and liver vascular tumors.(199) Increase of the GLUT1 levels in cancer has been reported to be mediated by a decrease in the protein degradation rate, with little or no increase in its biosynthesis rate. The opposite phenomenon is also known, with up-regulation of SLC2A1 and lack of a change in GLUT1 degradation rate.(200) SLC2A1 is known to be up-regulated in hypoxia,(201) in particular, through HIF1α interaction with the consensus HRE.(202) Additionally, SLC2A1 is up- regulated by the cellular oncogenes ras and SRC,(203) implicated in the development of many human tumors. However, the mechanism of such intervention may still involve HIF1α as the final member of the cascade. Indeed, it has been demonstrated, that H-ras not only leads to a substantial increase in GLUT1 mRNA levels under normoxic conditions, but additively increases GLUT1 mRNA levels under hypoxia, suggesting that ras mediates its effect through HIF1α,(204) which thus is the final member of the H-ras-MAPK cascade.(200) SLC2A1 was found to participate in the regulation of collagenase metalloproteinase-2 (MMP-2) expression, up-regulating its levels by more than 4-fold.(205) MMP-2 is implicated in the neoplastic invasiveness by the virtue of degrading type IV collagen, the major component of basement membranes, and thus permitting the cancer cell migration into the bloodstream.(206) 46 As such, GLUT1 controls not only the metabolism switch, but invasive growth as well. Given the fundamental nature of both of these cancer hallmarks, with glycolytic switch deemed responsible for the onset of cancer altogether, the regulation of GLUT1 expression comes to light as one of the most important challenges from the therapeutic prospective. I.7.5. CXCR4 Small pro-inflammatory chemoattractant cytokines, called chemokines, and their specific binding partners, G-protein-coupled seven-span transmembrane receptors present on plasma membranes of target cells, are the major regulators of cell trafficking, and occasionally modulate cell survival and growth.(207) Chemokines usually bind to multiple receptors, and the same receptor may bind to more than one chemokine. However, there is one exception to this rule: α-chemokine stromal-derived factor-1 (SDF-1), which binds exclusively to G-protein-coupled CXCR4, and appears to be its only ligand.(208) SDF-1 and CXCR4 play an important and unique role in the regulation of stem/progenitor cell trafficking. SDF-1 regulates the trafficking of CXCR4 + hemato/lymphopoietic cells and their retention in major hemato/lymphopoietic organs, accumulation of CXCR4 + immune cells in tissues affected by inflammation, trafficking of CXCR4 + during embryo/organogenesis and tissue/organ regeneration.(207) However, CXCR4 is expressed on certain cells in almost all types of cancer as well, which enables them to metastasize to the organs that express and secrete SDF-1, such as bones, lymph nodes, lung, and liver.(207) In particular, rhabdomyosarcoma, neuroblastoma, and nephroblastoma can metastasize to the bone marrow and infiltrate the marrow to a substantial extent.(209) As such, levels of CXCR4 expression may be a good indicator of 47 the severity of a metastatic phenotype. Additionally, the malignant phenotype is deemed to originate from the maturation arrest of the progenitor cells, rather than the dedifferentiation of somatic cells, as stem cells are a long lived-population that is capable of accumulating mutations and passing them to their daughter cells.(210) Thus the SDF- 1-CXCR4 axis may be responsible for the recruitment of the cancer stem cells from the hematopoietic organs into tumor sites.(211) Figure I.10. SDF-1-CXCR4 signaling. Reproduced from reference (212). Interaction of SDF-1 with CXCR4 initiates the activation of a cascade of several signal transduction pathways (Figure I.10).(212) In particular, they activate several key cascades, regulating angiogenesis, metastasis, and survival in cancer: calcium flux, focal adhesion components, such as proline-rich kinase-2 (Pyk-2), Crk-associated substrate 48 (p130Cas), focal adhesion kinase, paxilin, Nck, Crk, Crk-L, protein kinase C, phospholipase C-italic gamma (PKC- italic gamma), MAPK p42/44-ELK-1, and PI3K- AKT-NF-κB axes.(211) HIF1α up-regulates the expression of CXCR4, by specifically binding to the HRE region of the CXCR4 promoter.(213) Interestingly, SDF-1 expression is up-regulated in hypoxic endothelial cells by HIF1α as well,(214) providing the fundamental explanation to the observed phenomenon of the recruitment of stem cells by solid tumors. Thus, solid tumors being permanently hypoxic microenvironments, elicit behavior similar to that of an injured tissue, a transiently hypoxic microenvironment, and therefore deserve a label of a “wound that never heals”.(211) SDF-1-CXCR4 axis is responsible for the motility of the stem cells, recruitment of the new cells into the tumor site, and metastasis of the cancer stem cells; implicated in survival and proliferation pathways. Both components of this axis are regulated by HIF1α, and thus add tremendous value to the research directed towards the disruption of its transcriptional activity. 49 I.8. Hypoxia and Anti-Cancer Treatment. Regions of hypoxia with especially low oxygen levels, anoxia (≤0.02% O2), and elevated acidity allow for a heterogeneous tumor microenvironment, and typically incur selection pressure for a more aggressive phenotype.(102) Neovascularization, considerably rushed by the rapidly proliferating starved cells, is not implemented properly, typically resulting in the disorganized abnormal vasculature. As a result, the difference in pressure between arterioles and venules is reduced and viscous and geometric resistance is increased, culminating in a decreased blood flow. Therefore, small molecules do not have access to most of the tumor mass, especially when they rely exclusively on passive uptake. Hypoxia diminishes efficacy of certain chemotherapeutic agents which rely on the generation of superoxides (doxorubicin, bleomycin) and free radicals,(215) and is known to promote the development of indirect drug resistance. In particular, hypoxia increases the production of metallothioneins that possess high affinity towards heavy metal ions and consequently may be responsible for the development of cisplatin, Adriamycin, and micomycin C resistance,(216) and leads to the amplification of the expression of dihydrofolate reductase conferring methotrexate resistance.(217) Nutrient deprivation induces cell cycle arrest, and since most chemotherapeutic drugs are more effective against proliferating than quiescent cells, starved cells in the solid tumors exhibit resistance towards such drugs and agents that target cell proliferation.(218) In particular, an antineoplastic agent bleomycin A2 is known to be most effective in killing cells in M and G2 phases, and its activity is substantially diminished in essentially non-cycling hypoxic cells.(219) Glutathione, overproduced under hypoxia, may compete with the target DNA for alkylation and lead to acquisition of the resistance to alkylating agents.(220) 50 Lack of tumor targeting ability in a vast majority of the chemotherapeutic drugs translates into the diminished tissue specificity. Many anticancer drugs, in particular, rely on passive uptake and preferential accumulation in solid tumors, and resort to non-specific cytotoxicity as their mode of action; as such, they affect healthy cells as well, and cause profound adverse effects, highlighting the critical need of a phenotype-specific anticancer therapy. 51 I.9. Therapeutic Targeting of HIF Pathway. It has been conclusively demonstrated that cancer cells acquire an aggressive phenotype largely due to the activation of HIF-dependent pathways under hypoxia, promoting a switch for a glycolytic metabolism, angiogenesis, and increased invasiveness and motility. Therefore HIF transcription factors, their binding partners, and immediate downstream targets are attractive yet challenging therapeutic targets for the development of anticancer agents, and multiple corresponding approaches have been developed. Transcription factors including HIF1α possess vast shallow surfaces, making them “undruggable” targets. Despite that, multiple small molecules that can target HIF- dependent pathway have been discovered or designed. These circumvent the direct inhibition approach and target the machinery responsible for the destabilization of HIF1α (VHL) or interaction of HIF1α with the HRE region of the DNA promoter, its β-subunit, its cofactor p300/CBP, and FIH. I.9.1. Natural Products Targeting HIF Pathway. A few natural products having anticancer properties, also manifested the interference with HIF pathway (Figure I.11), primarily via stimulation of HIF1α proteasomal degradation. Curcumin, a component of commonly used spice and coloring agent, has a variety of beneficial biological effects, including inhibition of tumor growth. Curcumin blocks hypoxia-stimulated angiogenesis, and in particular inhibits the expression of VEGF and erythropoietin. Such anticancer activity of curcumin stems from the down-regulation of HIF1α protein levels in HepG2, HEK293, and HUVEC cells, possibly involving p53- 52 mediated ubiquitination or activation of the proteases in the proteasomal pathway,(221) and the stimulation of the proteasomal degradation of HIF1β in Hep3B cell line.(222) Figure I.11. Natural product inhibitors of the HIF1 pathway. Manassantin A, isolated from and aquatic plant Saururus cernuus L., prevents the HIF1α protein accumulation under hypoxia without affecting its mRNA levels, resulting in the inhibition of VEGF secretion in 4T1 mammary carcinoma cells,(223) and down-regulation of CDKN1A, VEGF, and GLUT1 HIF-dependent gene expression in T47D breast carcinoma, Hep3B, and MDA-MB-231 cells.(224) It is noteworthy, that effective concentration of manassantin A is far below its GI50. Bioassay-guided fractionation of the chloroform-soluble extracts of Morus species has led to a discovery of several natural product inhibitors affecting HIF1α protein levels.(225) 53 Moracins O and P inhibit the accumulation of HIF1α and secretion of VEGF under hypoxia at nanomolar concentrations and do not exhibit any noticeable toxicity at the concentrations exceeding 30 µM in Hep3B cells.(225, 226) A target-based high-throughput screen of 600,000 compounds identified chetomin, a dithiodiketopiperazine metabolite of the fungus Chaetomium species, as a regulator of HIF pathway. Unlike the vast majority of other HIF pathway inhibitors, reducing HIF protein levels, chetomin instead disrupts the native fold of the p300/CBP cofactor, therefore preventing it from binding to HIF1α and HIF2α, and abolishing the transcription of HIF- dependent genes.(227) I.9.2. Synthetic Small Molecules Targeting the HIF Pathway. Several synthetic small molecule inhibitors of HIF pathway at various levels have been developed (Figure I.12). They range from the mimetics of the known natural products to known drugs and drug-like molecules that found a new application to novel entities obtained strictly via rational design. Histone deacetylase (HDAC) inhibitor romdepsin inhibits transcriptional activity of HIF1α manifesting in blockade of induction of VEGF and angiogenesis in hypoxia in Lewis lung carcinoma model, however the exact mechanism is not known.(228) YC-1 (3-(5´-hydroxymethyl-2´-furyl)-1-benzylindazole), first reported in 1994 as an activator of platelet guanylate cyclase,(229) has received considerable attention as an anticancer agent. It was found to suppress angiogenesis and halt the tumor growth in vivo due to inhibition of HIF1 activity via the prevention of hypoxic accumulation of HIF1α 54 protein.(230) Another mechanism for the YC-1 anticancer activity suggests that YC-1 inactivates CTAD domain of HIF1α (and HIF2α) in a FIH-dependent manner and thus prevents the recruitment of p300/CBP.(231) Figure I.12. Synthetic small molecule inhibitors of the HIF1 pathway. 55 Insulin and EGF increase the protein levels of HIF1α in a PI3K-dependent manner. PI3K- specific inhibitors LY294002 and wortmannin were demonstrated to largely or completely abolish the accumulation of HIF1α protein in PC3 and DU145 cells due to the loss of the AKT-mediated stability.(232) Such AKT-induced prostate epithelial neoplasia phenotype is mTOR-dependent. Therefore, mTOR inhibitors everolimus and temsirolimus, synthetic derivatives of rapamycin (sirolimus), were demonstrated to attenuate HIF1α protein synthesis as well, thus implicating the same mechanism involving the loss of AKT- mediated HIF1α stabilization.(233, 234) Upon its synthesis on a ribosome, HIF1α is stably associated with chaperone HSP90. Disruption of such interaction by the HSP90 inhibitor geldanamycin induces oxygen- independent ubiquitination and proteasomal degradation of HIF1α in PC3 and LNCaP prostate cancer cells.(235) Camptothecin derivative topotecan, a topoisomerase I inhibitor, is known to inhibit the translation of HIF1α; such phenomenon is independent of PI3K-AKT-mTOR pathway and does not affect the HIF1α stability or cell cycle.(236) Topotecan is currently used in clinic as a chemotherapeutic (Hycamtin) in the treatment of recurrent small cell lung cancer(237) and ovarian cancer.(238) Indenopyrazole efficiently suppresses HIF1α activity (IC50 = 14 nM) without affecting either HIF1α protein levels, heterodimerization between HIF1α and ARNT, or recruitment of p300/CBP, suggesting it is a downstream effector. The exact mechanism is not known. Indenopyrazole and its analogs are readily available due to their simplistic structure and lack of stereocenters.(239) 56 Interference with HIF-dependent transcription at the DNA-binding level was achieved via the implementation of the sequence-specific DNA-binding pyrrole-imidazole polyamides. These molecules mimic the arrangement of nucleobases by having N-methylpyrrole, N- methylhydroxypyrrole, and N-methyimidazole in various combinations on the two chains connected with a linker that bind to the minor groove of the DNA. The polyamides bind to the DNA with very high affinity, preventing the binding of transcription factors. In particular, the polyamide designed to bind to the putative HRE region of the VEGF promoter blocks binding of HIF1α to its response element and prevents the further recruitment of the transcription machinery.(240) I.9.3. Transcriptional Regulation of HIF Pathway. All described attempts to target HIF pathway with synthetic molecules of natural products yielded either translational inhibitors of HIF1α, or agents that prevent HIF1α accumulation through the initiation of its destabilization, while the direct inhibition of HIF1α transcriptional activity remained a challenge – until the discovery of chetomin in 2004 by Kung et al. (vide supra).(227) Recapitulating, under hypoxia upon translocation into the nucleus, HIF1α forms a heterodimer with ARNT, and recruits p300/CBP. Specifically, CTAD domain of HIF1α (aa 786-826) binds to the CH1 domain of p300 (aa 302-423).(241) A time-resolved fluorescence assay established chetomin as a submicromolar inhibitor of the interaction between HIF1α and p300/CBP. Chetomin showed dose-dependent down-regulation of the transcription of some hypoxia-inducible genes in Hep3B cells and exhibited significant antitumor efficacy in mice tumor xenografts established with PC3 cells. The mechanism 57 of chetomin activity has not been fully elucidated; however it is known that chetomin targets the CH1 domain of p300, disrupting its global fold and thereby allosterically inhibiting its interaction with HIF1α, as well as other proteins interacting with the CH1domain.(227) However promising, chetomin showed necrosis, anemia, and leukocytosis in mice. Additionally, the structure of chetomin is so complex, that it has not been synthesized to date. It has been suggested and proven by Olenyuk and coworkers that the two epidithiodiketopiperazine (ETP) moieties play the key role in defining the efficacy of chetomin, while the scaffold allows for substantial changes (Figure I.13).(242, 243) The mechanism of ETP action is thought to involve zinc ejection from the cysteine-rich sites within the CH1 domain.(244) Figure I.13. Synthetic analogs of chetomin. For more than a decade, the key protein-protein interactions mediated by alpha-helices have been a primary research subject of many groups trying to engineer small scaffolds that mimic the topography of the key parts of the alpha-helices. The major effort was 58 directed towards the development of competitive (orthosteric) inhibitors of these protein- protein interactions. Substantial progress has been reported in vitro several times, with Orner et al. on terphenyls (245) breaking the ground; however, successful in vivo application has been attained only recently. Insufficient affinity and/or selectivity of targeting might be the reason for such discrepancy. The in silico models of protein-protein interactions revealed that despite the lack of well- defined pockets and the large size of the interfaces, in most cases there is a small and complementary set of “hotspot” residues that contribute the most to the binding free energy of the protein-protein interactions.(246) These hot spots tend to be enriched in tryptophan, tyrosine and arginine, surrounded by energetically less important residues speculated to guard the hot spot from the bulk solvent.(247) The interaction between HIF1α and p300/CBP is in agreement with this theory. Despite the high affinity of HIF1α CTAD to p300/CBP CH1, addition of a single hydroxyl group on Asn803 of CTAD disrupts the complex, indicating that it may be susceptible to competitive inhibition by the small molecules. NMR solution structure of HIF1α CTAD bound to CH1 domain of p300 reveals that two short α-helical domains of HIF1α are requisite for its recognition by p300. Stable mimics of these domains can orthosterically inhibit the interaction between HIF1α and p300/CBP and down-regulate the expression of hypoxia-inducible genes.(248) In 2013 Kushal et al. reported protein domain mimetic targeting the interaction of HIF1α with p300/CBP.(248) They have implemented a hydrogen bond surrogate (HBS) approach to stabilize short peptides in a helical conformation with the goal to mimic the two key α- helical domains of HIF1α (Figure I.14). HBS mimics of HIF1α exhibited remarkable efficacy in suppressing the expression of a variety of HIF-inducible genes and tumor 59 growth in the fully grown mouse tumor xenograft models established with clear cell type renal cell carcinoma (RCC) cell line. Interestingly, such mimics are essentially non-toxic, and possess limited off-target effects, as demonstrated by the relatively minimal perturbation of the unrelated signaling pathways. This highlights the advantages of the rational design approach in targeting hypoxia-inducible transcription factor complex. Figure I.14. HBS helix mimetics of the key α-helices of HIF1α. Adapted from reference (248). To summarize, among all small molecule inhibitors of HIF pathway, many primarily encompass the inhibitors of HIF1α translation and accumulation. However, by the virtue of targeting the effectors of HIF1α higher upstream, these inhibitors inherently possess the potential for a substantial off-target effect, which might explain why very few of these molecules have reached the clinic. Development of the inhibitors of the HIF1α transcriptional activity is further complicated by the inherent complexity of the transcription factor targeting: multiple binding partners and vast and shallow interaction interfaces. Two fundamental approaches have been in the works: allosteric, focused on disrupting the fold of p300/CBP; and orthosteric, aiming at competitive inhibition of the interaction between HIF1α and p300/CBP. These candidates include natural products, mimics of the natural products, and rationally designed molecules. 60 I.10. Vascularization of Mouse Tumor Xenograft Models. Abnormal architecture of the vasculature in solid tumors largely defines, both directly and indirectly, the tumor responsiveness to chemotherapy. Blood vessels in tumors are often dilated and convoluted and, compared with normal tissues, have branching patterns that feature excessive loops and arteriolar–venous shunts (Figure I.10).(249) The vessels in some tumors share structural similarities of arterioles, capillaries, and venules simultaneously. The walls of tumor vessels may have fenestrations, discontinuous or absent basement membranes, and fewer pericytes than walls of normal vessels and may lack perivascular smooth muscle.(250) In addition, cancer cells may be integrated into the vessel wall, yielding a cellular lining composed of disorganized, loosely connected, branched, overlapping or sprouting endothelial cells. These abnormalities tend to make tumor vessels leaky, although their permeability varies both within and among tumors.(251) Figure I.10. Schematic representation of the vascular system. A) Normal tissue. B) Solid tumor. Red: well-oxygenated arterial blood. Blue: poorly oxygenated venous blood. Green: lymphatic vessels. Reproduced from reference (215). 61 Quantification of the vascular density is rarely reported in a standard uniform way in the literature. Some authors prefer to discuss the vascular density by directly referring to the visual side-by-side comparison of the pictures of CD31-stained slides, while others report the number of vessels obtained at a given magnification, rendering these absolute vascular density values almost useless outside the scope of their work. However, the standard does exist. Chalkley vascular count is used to analyze the degree of vascularization. It is the number of grid points that hit stained vessels, taken as the average from the assessment of three hot spots. The three most vascular areas (hot spots) with the highest number of microvessel profiles are chosen subjectively from each tumor section. A 25-point Chalkley eyepiece graticule(252) is applied to each hot-spot area and oriented to permit the maximum number of points to hit on, or within the areas of immunohistochemically highlighted microvessel profiles (Leitz Orthoplan, ×250; Chalkley grid area, 0.196 mm 2 ). The Chalkley count for an individual tumor is taken as the mean value of the three graticule counts. The tertiles of the Chalkley estimates, 5 and 7, were used to describe the malignancy of the tumors. The independent prognostic estimate demonstrated a 57% higher risk of dying when a tumor had a Chalkley count between 5 and 7, and 125% higher risk with Chalkley counts ≥7, compared with the risk associated with tumors showing Chalkley counts ≤5.(253) To date, there is no comprehensive review on the comparison of the vascular density of different xenografts in mice. There are, however, a few publications, that either analyze the vascularization of a given xenograft, or compare the vascularization of the tumor xenografts based on the cell line they were established from, or their type. 62 To test the efficacy of oncolytic adenoviruses, HeLa (cervical carcinoma), LoVo (colon carcinoma), MDA-MB-435 (breast cancer), and Hep3B (hepatoma) ectopic tumor models were established by injecting the cells into the portal vein of immunodeficient mice. These tumor models differ in morphological features and in the accessibility to virus transduction. The LoVo and Hep3 models exhibit extensive tumor vascularization and contact between blood vessels and tumor cells. The HeLa model displays a small number of blood vessels at the tumor periphery that do not reach the center of the tumor, where early necrosis was detected. In MDA-MB-435 model, the blood vessels were visible between the tumor nodules, but they rarely penetrated into the tumor nests. The MDA-MB-435 nests were lined by a continuous basement membrane, a thin sheet of specialized extracellular matrix consisting mainly of type IV collagen, laminin, heparan-sulfate proteoglycans, and nidogen/entactin that blocks contact between tumor cells and blood vessels.(254) Similarly, MDA-MB-231 breast adenocarcinoma xenografts in nude mice produced somewhat vascularized tumors (quantification not available).(255) A stromal component pro-collagen type I peptide (PICP) was found to considerably induce the vascularization of MDA-MB-231 tumors. Soft tissue sarcoma subcutaneous implants frequently fail, presumably due to the poor neovascularization, as the models with intrinsic vascularization, deliberately surgically established around existing blood vessels, are always successful.(256) Subcutaneous pancreatic PancTu I tumor models exhibit a significantly higher vascularization compared to their orthotopic analogues.(257) 63 H1975, A549 NSCLC subcutaneous tumor models, and H441 lung papillary adenocarcinoma orthotopic tumor models exhibit substantial vascularization (microvessel density of 28, 32, and 46 per 0.7386 mm 2 ) and develop resistance to bevacuzimab-induced VEGF blockade, in part, due to increased EGFR signaling.(258) These results are confirmed by another study featuring A549 subcutaneous xenografts.(259) The control tumors were shown to exhibit microvessel density of 10 per 0.155 mm2, which can be translated to 48 per 0.7386 mm2. This number is quite close to the values reported in the previous study. As described by Sun et al. in 2001,(260) stained blood vessels were counted in five blindly chosen random fields (0.155 mm 2 ) at 40x magnification, and the mean of the highest three counts was calculated. Perfused breast cancer cell line ZR75-1 subcutaneous and tissue-isolated tumor models exhibit vascular density of 5.4 vessels per 0.25 mm2. Antiestrogen therapy does not seem to reduce the tumor vascularization. No change in oxygen consumption of ZR75-1 tumors after antiestrogen therapy was found.(261) Caki-2 and 786-O renal cell carcinoma can produce highly vascularized xenografts, as demonstrated by the CD31 staining (quantification not available).(262) Proliferation of VHL-mutant RCC cells lines 786-O and A498, growth and vascularization of corresponding tumor xenografts are substantially inhibited by 2-hydroxyflavanone treatment. In ovarian cancer xenograft models established intraperitoneally even very small tumors, comprised of less than 20,000 fluorescent SKOV3.ip1 cells, are fully vascularized, 64 consistent with a mathematical three-dimensional cellular Potts ovarian tumor model that examines ovarian cancer cell attachment, chemotaxis, growth, and vascularization.(263) Murine renal parenchyma was inoculated with HuH-6 hepatoblastoma cells to produce a well-vascularized tumor.(264) Treatment with anti-VEGF antibody dramatically inhibited the growth and resulted in reduced vascularity of the tumor xenografts. To summarize, understanding the vasculature pattern and quantification of the vascular density is important for the correct assessment of tumor microenvironment, therapeutic efficacy of anti-cancer drugs, and establishment of the realistic animal tumor xenograft models. This rather difficult and quite technical aspect of in vivo work is arguably underappreciated and frequently overlooked, however some substantial progress has been made in the past decade. 65 Chapter II. Topographical Helix Mimetics as In Vivo Modulators of Hypoxia-Inducible Signaling. 66 II.1. Introduction. Protein-protein interactions (PPIs) represent attractive yet largely unexplored groups of targets for drug design,(265) and the development of small molecule inhibitors of PPIs is a fundamental challenge at the interface of chemistry and cancer biology. Despite the importance of PPIs in biological signaling, relatively few small molecule inhibitors have been discovered, underscoring the difficulty in inhibiting large interfaces.(266) Successful methods for design of PPI inhibitors include computational and experimental high- throughput and fragment-based screening strategies to locate small molecule fragments that bind to protein surfaces. An alternative rational design approach seeks to mimic the orientation and disposition of critical binding residues at protein interfaces. Recent analyses suggest that PPIs may be categorized as those that are amenable to inhibition by small molecules and those that will require large molecules.(267-269) Key to these analyses is the recognition that although PPIs encompass larger surface areas than enzyme– substrate complexes, a handful of key residues dominate the binding energy landscape.(270, 271) The design of PPI inhibitors, then, requires mimicry of the relative positioning and disposition of these important residues, termed “hotspot residues”, on synthetic scaffolds.(267, 268, 272-275) In particular, the rigidity of α-helices allows for the design of small molecule scaffolds that could capture their topography. Based on these hypotheses, we describe a rationally designed compound that down-regulates hypoxia- inducible signaling in cell culture by targeting a protein-protein interaction predicted to be suitable for small inhibitors. The designed inhibitor reduces tumor burden in mouse tumor xenografts where hypoxia-inducible proteins are overexpressed. 67 Recuperating the detailed review provided in Chapter I, mammalian cells possess an intricate signaling network that responds to changes in the oxygen tension in their immediate surroundings.(276) Under hypoxia, cells express a family of hypoxia-inducible transcription factors (HIFs), which are heterodimeric proteins composed of a regulatory α and a constitutively expressed β subunits. The C-terminal transactivation domain (CTAD, aa 786–826) of HIF1α interacts with the cysteine-histidine rich 1 (CH1) domain of the coactivator protein p300 (or CREB binding protein, CBP, Figure II.1a).(277, 278) The HIF/p300 complex mediates transactivation of hypoxia-inducible genes, which are important contributors to angiogenesis, invasion and altered energy metabolism in cancer.(279) Inhibitors of hypoxia inducible gene expression could serve as unique tools for dissecting hypoxia signaling in tumors, as well as leads for cancer therapeutics.(240, 242, 243, 248, 280-285) The transcription factor–coactivator interaction presents an intriguing target for controlling hypoxia signaling because it is a critical node directing downstream expression of various genes which work in concert to modulate cancer progression. From a ligand design perspective, transcriptional PPIs are often challenging because of their transient existence and relatively low binding affinities.(286-288) Our work supports the hypothesis that topographical mimics of energetically important residues on protein secondary structures offer a rational approach for discovery of PPI inhibitors.(267, 268, 272-275, 289, 290) 68 Figure II.1. Design of HIF1α mimetics to modulate hypoxia inducible gene expression. (A) Complex of HIF1α and the CH1 domain of p300/CBP (PDB code 1L8C). The key residues Leu818, Leu822, and Gln824 of HIF1α C-TAD (shown in blue) are located in the binding pocket of p300/CBP CH1 domain (depicted in red). Magnified: an overlay of HIF1α helix spanning residues 816-824 (blue) and OHM 1 (green). (B) Oxopiperazine helix mimetics (OHMs) were designed to mimic the key helical region. OHMs feature ethylene bridges between adjacent amino acid residues; the bridges lock the side chain groups in orientations that mimic their topography in α-helices. 69 II.2. Results. II.2.1. Design and Synthesis of Topographical HIF1α Mimics. The C-terminal activation domain of HIF1α utilizes two short α-helices to bind to the CH1 domain of p300/CBP (Figure II.1A). Small molecules that mimic the structural arrangement of the key residues on these helices should afford competitive inhibitors of HIF1α/p300 complex formation.(248, 283) We employed a recently described strategy from our groups to mimic the interacting face of an α-helix on a small molecule oxopiperazine scaffold.(291) Oxopiperazine helix mimetics (OHMs) are assembled from naturally-occurring amino acids with the nitrogen atoms of neighboring backbone amides constrained with ethylene bridges to afford a nonpeptidic chiral scaffold displaying protein- like functionality (Figure II.1B). Oxopiperazines are attractive scaffolds for discovery of PPI inhibitors because of their rich history in drug design.(292) Molecular models suggest that the low energy conformation of oxopiperazine scaffold arrays side chain functionality to mimic the arrangement of the i, i+4, and the i+6 or i+7 residues on canonical α- helices.(291) OHMs add to a growing class of nonpeptidic helix mimetics that have been shown to have biological activity.(272, 273, 275, 289) The main advantage of OHMs is dues to their chiral backbone. A majority of nonpeptidic helix mimetics are based on achiral aromatic scaffolds.(272, 291) Chiral scaffolds are expected to interact with molecular binding pockets with higher specificity. High-resolution structures and computational analyses(293) of the HIF1α/p300 complex reveal that four helical residues from the HIF1α helix816-824 (Leu818, Leu822, Asp823, and Gln824) make close contacts with the CH1 domain of p300/CBP (Table II.1). Three of 70 these residues, Leu818, Leu822 and Gln824, can be mimicked by oxopiperazine dimers consisting of the appropriate building blocks (Figure II.2). We designed and synthesized four analogs of HIF1α CTAD to inhibit its binding with p300/CBP. N N N O N O NH 2 O NH HN NH 2 O H 2 N O HN N N O N O NH 2 O O HN N N O N O NH 2 O O NH 2 O HN N N O N O NH 2 O O NH 2 O HN N N O N O NH 2 O O NH 2 O OHM 1 OHM 2 OHM 3 OHM 4 OHM 5 Figure II.2. OHM derivatives – positive and negative controls – designed to inhibit the target complex. 71 OHM 1 contains projections representing all three residues from HIF1α: R1 as Leu818, R2 as Leu822, and R4 as Gln824. The R3 position of the oxopiperazine scaffold is not predicted to make contacts with the target protein, and an alanine residue was inserted at this position. OHMs 2 and 3 are single mutants of 1 with R4 and R2 positions, respectively, substituted with alanine residues. Based on computational analysis and the relative contributions of Leu822 and Gln824, OHM 2 would be expected to bind p300–CH1 with a much higher affinity than OHM 3. We also prepared OHM 4 containing alanine residues at all four positions as a negative control. Additionally, a longer OHM 5 was constructed based on the extended dimer design, with the native residues at the i, i+2, and i+4 residues. Only the leucine residues at the i and i+4 positions are making contacts, whereas the Arg820 is turned away and out of the binding cleft, and it was introduced to increase water solubility and facilitate cell permeability, while the i+3 and i+7 positions were substituted with alanine residues. OHMs 1-4 were synthesized using standard Fmoc amino acids and coupling reagents on Rink amide resin (Scheme II.1). The oxopiperazine ring is obtained via the application of the Fukuyama–Mitsunobu strategy, which involves activating the amine functionality by reaction with o-nitrobenzenesulfonyl chloride (o-Ns-Cl) followed by alkylation of the neighboring amide with 2-bromoethanol. The formation of the ring is then finalized by alkylating the o-Ns-activated amine with the ethoxy fragment attached to the amide in presence of DBU in THF, and deprotecting the available amine by 2-mercaptoethanol. Typical syntheses are performed at 0.25 mmol scale on standard solid phase matrices, and afford overall yields of 10-20% after HPLC purification. 72 Table II.1. Computational alanine scanning mutagenesis energies calculated with Rosetta(293) ver. 3.3. Scans were performed on the HIF1α/CBP complex (PDB codes 1L8C and 1L3E). HELIX B (817-824): ELLRALDQ Residue Helix B residue ΔΔG (kcal/mol) Leu 818 1.4 Leu 819 0.5 Arg 820 0.1 Ala 821 0.0 Leu 822 1.9 Asp 823 1.4 Gln 824 0.3 Scheme II.1. Solid-phase synthesis of oxopiperazine dimers. 73 Figure II.3. Binding between HIF1α CTAD and p300/CBP CH1 domain. (A) Chemical structure of fluorescein-labeled C-TAD (Flu-HIF1α C-TAD786-826). Mass [M+H] + calculated = 4977.1; found = 4976.8. (B) Binding of Flu-HIF C-TAD to p300-CH1 as measured by a fluorescence polarization assay. II.2.2. Binding Affinities of OHMs for the p300–CH1 Domain. The binding affinities of OHMs for p300-CH1 domain were evaluated using intrinsic tryptophan fluorescence spectroscopy, as described previously in the literature.(248, 294) Because Trp403 is located in the binding cleft of p300/CBP where a native HIF1α816-824 helix binds, it offers a probe for investigating mimetics of this helix. Data from the tryptophan fluorescence spectroscopy gave a dissociation constant, Kd of (3.8 ± 1.4) × 10 - 8 M for HIF1α CTAD786-826 to p300 CH1, which is consistent with the values obtained from a fluorescence polarization assay using fluorescein-labeled HIF1α CTAD (Figure II.3) and those reported in the literature with isothermal titration microcalorimetry.(248, 278) 74 OHM 1 targets CH1 with an affinity of (5.3 ± 1.4) × 10 -7 M (Figure II.4A). OHM 2, which contains the two critical leucine residues but lacks Gln824 binds with a slightly reduced affinity Kd = (6.2 ± 1.1) × 10 -7 M. The binding affinity of OHM 2 confirms the computational prediction that Gln824 is a weak contributor to binding (Table II.1). The negative controls OHM 3 and OHM 4 displayed very weak affinities for p300-CH1, with Kd values of >>1.0 × 10 -5 M in each case. The results signify that the designed scaffolds are able to target the protein of interest in a predetermined manner. Figure II.4. Binding affinities of designed compounds for p300–CH1. (a) The affinity of OHMs 1-4 and HIF1α C-TAD786-826 for CH1 domain was determined by tryptophan fluorescence spectroscopy. (b) Molecular model that depicts the results of a 1 H- 15 N HSQC NMR titration experiment. The p300-CH1 residues undergoing chemical shift perturbations upon addition of OHM 1 are color-mapped, matching the magnitude of the chemical shift changes. The structure of the HIF1α/CH1 complex (PDB code 1L8C) was used to construct the model. 75 We further characterized the interaction of OHM 1 with p300-CH1 domain using 1 H- 15 N HSQC NMR titration experiments with the uniformly 15 N-labeled CH1.(248) Concentration-dependent resonance shifts of several residues were observed upon addition of OHM 1 to 170 μM CH1 in CH1:OHM 1 ratios of 1:0.6, 1:1.2, 1:3.5 and 1:7 (Figure II.5). Addition of OHM 1 leads to consistent shifts in resonances of residues corresponding to the HIF1α816-824 binding surface, as expected from the design and the tryptophan fluorescence spectroscopy assay. The magnitude of the resonance shifts is consistent with the observation that CH1 has a stable conformation that does not reorganize substantially, at least upon binding of small ligands.(248, 295) Figure II.4B depicts a model of OHM 1 docked to CH1 residues that undergo chemical shift perturbation. The NMR results, along with the fluorescence binding experiments, provide strong evidence that rationally designed topographical mimics of protein α-helices can bind the predicted binding surfaces of their intended targets. 76 Figure II.5. 1 H- 15 N HSQC titration spectra. (A) Spectra of p300–CH1 (blue), CH1:OHM 1 (1:1.2, red), and CH1:OHM 1 (1:7, green) are overlaid. (B) Mean chemical shift difference (∆δNH) plot depicting changes in residues. 77 II.2.3. Designed Mimetics Down-regulate Hypoxia-Inducible Gene Expression. We next assessed the potential of OHMs to modulate the target interaction in cell culture and suppress hypoxia-inducible gene expression. We began by measuring the effect of compounds on the viability of human breast adenocarcinoma (MCF7) and human alveolar adenocarcinoma cells (A549) using the MTT assay. The three compounds expected to bind to p300 well (OHM 1, 2, and 5), showed dose-dependent decrease in MCF7 cell viability with EC50 values of 30–40 μM over 48 h, suggesting low cytotoxicity (Figure II.6-1A). Incubation over 72 h, quite expectedly, brought the EC50 values down (Figure II.6-1B), and, interestingly, highlighted the differences between the compounds. OHM 2 exhibited EC50 of 10 µM, OHM 1 – 20 µM, with the weakest binder of the three OHM 5 being the least toxic at EC50 of 35 µM. In A549 cells over the course of 48 h, both strong (OHMs 1, 2) and weak (OHMs 3, 4) binders of p300 exhibited similar low cytotoxicity, with EC50 values varying between 30 to 40 μM (Figure II.6-2). Next, we determined the efficacy of these compounds in inhibiting HIF1 transcriptional activity using a luciferase-based reporter assay.(248, 296) A triple-negative breast cancer cell line MDA-MB-231 was stably transfected with a construct designed to express a firefly luciferase protein under hypoxic conditions, which were generated by incubation with deferoxamine (DFO) at 300 µM. To our surprise, we did not see any significant decrease in luminescence levels; on the contrary all compounds we tested (OHM 1, 2, and 5) have led to higher levels of luciferase than control (Figure II.7-1). Despite the fact that DFO is widely used to mimic hypoxia, this approach is rather crude as DFO depletes prolyl hydroxylases of Fe 2+ , and as such, has a pronounced off-target effect. At the same time, 78 DFO does not accurately reproduce any of the other effects that natural hypoxia has on a cell. These conclusions compelled us to suspect that the DFO-mediated hypoxia induction might be the culprit, and try to find a better way to mimic hypoxia (Scheme II.2). Figure II.6-1. OHMs exhibit low cytotoxicity as evaluated in an MTT assay. MCF7 cells were treated with OHMs 1, 2, and 5 in the range of concentrations of 0.5 µM and 70 µM for 48 h (A) and 72 h (B). 79 0.01 0.1 1 10 100 0 50 100 OHM 1 OHM 2 OHM 3 OHM 4 Concentration of OHM, μM Relative Absorbance, % Figure II.6-2. OHMs exhibit low cytotoxicity as evaluated by MTT assay. A549 cells were treated with OHMs 1-4 in the range of concentrations of 1 µM and 100 µM for 48 h. Scheme II.2. Mechanism of hypoxia induction by DFO. As such, we attempted the O2 deprivation with a GasPak EZ pouch (BD). Under these conditions, treatments with OHM 1, OHM 2, and OHM 5 resulted in a dose-dependent reduction in the promoter (hypoxia response element) activity (Figure II.7-2). OHM 1 at 20 μM reduces the level of HIF1α transcriptional activity under hypoxia to that observed under normoxia (Figure II.7-2A). OHM 2 is slightly less effective in this assay (Figure II.7-2B), while OHM 5 exhibits an intermediate level of activity (Figure II.7-2C). The designed negative controls OHM 3 and OHM 4 do not cause a decrease in HIF1α activity 80 at 20 μM concentrations (Figure II.7-3). The activity of OHMs 1, 2, and 5 in MDA-MB- 231 is encouraging as this cell line often exhibits confluence-dependent resistance to anticancer drugs.(297) Importantly, the oxopiperazine mimetics did not lead to a decrease in HIF1α protein levels, as measured by Western blots (Figure II.8A), when either hypoxia mimicking condition was used. Figure II.7-1. Regulation of hypoxia inducible promoter activity by helix mimetics. OHM 1 (A), 2 (B), and 5 (C) do not affect hypoxia-induced promoter activity in luciferase assays if hypoxia is induced with DFO. Error bars are ± s.e.m. of four independent experiments. H, hypoxia; N, normoxia. 81 Figure II.7-2. Regulation of hypoxia inducible promoter activity by helix mimetics. OHM 1 (A), 2 (B), and 5 (C) down-regulate hypoxia-induced promoter activity in luciferase assays. Error bars are ± s.e.m. of four independent experiments. *** P < 0.001, ** P < 0.01, * P < 0.05, t-test. H, hypoxia; N, normoxia. 82 Figure II.7-3. Regulation of hypoxia inducible promoter activity by helix mimetics. OHMs 3 (A) and 4 (B) do not show inhibitory activity in luciferase assays. Error bars are ± s.e.m. of four independent experiments. H = hypoxia; N = normoxia. The ability of OHMs to inhibit transcription of the prominent HIF-dependent genes was assessed using real-time quantitative RT-PCR (qRT-PCR) assays in MCF7 (Figure II.7- 4) and A549 cells (Figure II.7-5). OHM 1 at 10 μM concentration significantly down- regulates the mRNA expression level of the critical angiogenesis and neovascularization regulator vascular endothelial growth factor (VEGFA)(298) by 40% (Figure II.7-4A). In comparison, OHMs 2 and 5 did not affect VEGFA mRNA levels at these concentrations. cMet gene encodes the cMet proto-oncogene (also known as MET and HGF) a tyrosine kinase that stimulates cell motility and invasion, conveys protection from apoptosis and resistance to angiogenesis, thereby promoting metastasis.(165) Although in our case cMet was not up-regulated by hypoxia in MCF7 cells (Figure II.7-4B), its levels were significantly reduced by all OHMs: 1 (35% at 5 µM and 40% at 10 µM), 2 (35% at 5 µM and 37% at 10 µM), and 5 (80% at 20 µM). 83 Lysyl oxidase (LOX) is secreted by hypoxic breast cancer cells and is responsible for the remodeling of the extracellular matrix in the lungs, leading to the recruitment of bone marrow-derived cells and facilitation of metastasis.(299) LOX was dramatically down- regulated by all OHMs (Figure II.7-4C): 1 (82% at 5 µM and 93% at 10 µM), 2 (75% at 5 µM and 95% at 10 µM), and 5 (90% at 20 µM). GLUT1, encodes a membrane protein facilitating the transport of glucose across the cell membrane, and as such, plays an important role in the tissues highly dependent on glucose metabolism.(300) Consequently, GLUT1 is of considerable interest to cancer research. GLUT1 was down-regulated by all OHMs we tested (Figure II.7-4D): 2 (25% at 5 µM and 50% at 10 µM), and 5 (10% at 20 µM). Chemokine receptor four (CXCR4), encoded by CXCR4, is a chemokine receptor most commonly expressed in tumors, where it indirectly promotes tumor metastasis by mediating proliferation and migration of tumor cells and enhancing tumor-associated angiogenesis.(301) CXCR4 was consistently down-regulated by the two OHMs 1 (40% at 5 µM and 70% at 10 µM) and 2 (55% at 5 µM and 85% at 10 µM), again in a dose- dependent manner, while OHM 5 exhibited a 40% significant up-regulation of CXCR4 at 20 µM (Figure II.7-4E). Overall, OHM 2 and especially OHM 5 appear to be overall slightly to substantially less active than OHM 1 in the qRT-PCR experiments just as it was predicted by the binding experiments and shown in luciferase assays. 84 Figure II.7-4. Transcriptional regulation of hypoxia inducible genes by helix mimetics. OHM 1, 2, and 5 down-regulate transcription of VEGFA, cMet, LOX, GLUT1, CXCR4 genes in MCF7 cells as measured by real-time qRT-PCR. Error bars are ± s.e.m. of four independent experiments. *** P < 0.001, ** P < 0.01, * P < 0.05, t-test. H = hypoxia; N = normoxia. 85 Figure II.7-5. Transcriptional regulation of hypoxia inducible genes by helix mimetics. OHM 1 and 2 down-regulate transcription of VEGFA, cMet, LOX, GLUT1, CXCR4 genes in A549 cells as measured by real-time qRT-PCR. OHM 3 and 4 show reduced inhibitory activities at the same concentrations. Error bars are ± s.e.m. of four independent experiments. *** P < 0.001, ** P < 0.01, t-test. H = hypoxia; N = normoxia. 86 Figure II.8. Western blot analysis of HIF1α levels in the whole cell extracts of A549 cells. Treatment with OHM 1 does not lead to a decrease in the intracellular levels of HIF1α (A), and causes depletion of cMet proto-oncogene (B). Cells were incubated for a total of 24 h with OHMs: after 6 h upon dosing, hypoxia was mimicked with DFO (300μM) or GasPak EZ pouch (HB) for an additional 18 h. The ability of some of the OHMs to inhibit transcription of these selected HIF target genes VEGFA, cMet, LOX, GLUT1, and CXCR4 was confirmed in A549 cells (Figure II.7-5). OHMs 1 and 2 at 10 μM concentrations down-regulate the mRNA expression levels of VEGFA by 90% and 95%, respectively (Figure II.7-5A). Similar levels of decrease were observed for the expression of cMet (80% and 80%, Figure II.7-5B), LOX (80 and 90%, Figure II.7-5C), and GLUT1 (50% and 70%, Figure II.7-5D). Interestingly, in A549 cells compared to MCF7 cells, cMet was up-regulated in hypoxia by 20%, and the OHM compounds 1 and 2 substantially down-regulated it under normoxia (80 and 65%, Figure II.7-5B) as well as under hypoxia (vide supra). OHM 2 is slightly more active than OHM 87 1 in the qRT-PCR experiments in A549 cells as compared to the luciferase assays and qRT- PCR in MCF7 cells. In particular, such discrepancy manifests clearly in the case of CXCR4 inhibition in A549 cells, as it is nearly unaltered by OHM 1, and down-regulated by OHM 2 by 70%. Such difference in behavior of the two compounds reflects the inherent variance in quantification of gene expression of endogenous genes and transfected constructs by two different methods, and possibly stems from the dissimilarities between the responses of two types of cancer: breast adenocarcinoma (MDA-MB-231 and MCF7) and NSCLC (A549). Control compounds 3 and 4 had no effect on mRNA levels of the indicated genes at these concentrations. We have also confirmed that some of these genomic changes manifest at the protein level. OHM 1 substantially decreases levels of cMet protein in A549 cells, while OHM 4 does not cause any noticeable change (Figure II.8B, C). II.2.4. Gene Expression Profiling of OHM in A549 Cells. The interaction of HIF1α with p300/CBP controls multiple downstream genes beyond the three genes interrogated with qRT-PCR. To comprehensively assess the effect of the mimetics on global gene expression, we used Affymetrix Human Gene ST 1.0 arrays containing oligonucleotide sequences representing over 28,000 transcripts. The OHMs are small scaffolds (MW < 500) with an inherently restricted set of interaction points, and as such, they could have non-specific off-target effects. In addition, p300/CBP are multi- domain coactivator proteins known to interact with many different transcription factors.(302) We sought to assess the genome-wide effects of the two successful inhibitors 88 of the hypoxia-inducible transcription OHM 1 and OHM 2 and compare these to the negative control OHM 4 (Figure II.9). Analysis of oncogenic signaling pathways in cells treated with OHM 1 revealed down- regulation in expression of multiple genes responsible for accelerated tumor progression. Examples of several HIF gene targets involved in glucose and lipid metabolism, and iron transport are shown in Figure II.9B. The targeted genes and the affected pathways can be classified into functional groups commonly attributed to as “hallmarks of cancer” (Figure II.9C).(51, 303) OHM 1 down-regulates multiple genes implicated in angiogenesis, apoptosis, cell proliferation, and invasion, along with several cancer specific markers in A549 NSCLC cell line. These include ceruloplasmin (CP), DNA topoisomerase 2-α (TOP2A), the high mobility group box family member 3 (TOX3) that modifies chromatin structure, Gremlin 1 (GREM1), a 184 amino acid glycoprotein that is important in the survival of cancer stroma and cancer cell proliferation, and fibroblast growth factor receptor 1 (FGFR1), which is overexpressed in squamous cell lung cancers. All these markers were down-regulated in A549 cells treated by OHM 1, suggesting that in addition to suppression of hypoxia- inducible transcription, the oxopiperazine also exerts general anti-tumor growth effect. The microarray data also reveals down-regulation of angiogenesis genes, such as VEGFA, SERPINE1, and genes that encode proteins promoting invasion, such as E-cadherin (CHD1), vimentin (VIM), and MET proto-oncogene. Expression of the anti-apoptosis genes such as BCL2A1, XIAP, MCL1R (Figure II.9B) was also modulated. 89 Figure II.9. Results from gene expression profiling obtained with Affymetrix Human Gene ST 1.0 arrays. (A) Hierarchical agglomerative clustering of transcripts induced or repressed 2-fold or more (one-way ANOVA, P ≤ 0.005) by hypoxia alone (GasPak EZ pouch) under the three specified conditions: -, no treatment; 1, OHM 1 (10 μM); 2, OHM 2 (10 μM); 4, OHM 4 (10 μM). Clustering was based on a Pearson centered correlation of intensity ratios for each treatment compared to hypoxia-induced cells (controls) using average-linkage as a distance. (B) Select tumor-promoting genes affected upon OHM treatment. (C) Schematic representation of genes affected by OHM 1 treatment in panel (B) color-coded in relation to “hallmarks of cancer”. (D) Venn diagrams representing transcripts up- and down-regulated (|fold-change| ≥ 2.0, P ≤ 0.005) by OHMs 1, 2 and 4. Numbers inside the intersections represent hypoxia-induced transcripts affected by corresponding treatments. 90 Figure II.10. Results from gene expression profiling obtained with Affymetrix Human Gene ST 1.0 arrays. (A) Hierarchical agglomerative clustering of transcripts induced or repressed 4-fold or more (one-way ANOVA, P ≤ 0.005) by hypoxia alone (GasPak EZ pouch) under the three specified conditions: -, no treatment; 1, OHM 1 (10 µM); 2, OHM 2 (10 µM); 4, OHM 4 (10 µM). Clustering was based on a Pearson centered correlation of intensity ratios for each treatment compared to hypoxia-induced cells (controls) using average-linkage as a distance. (B) Venn diagrams representing transcripts up- and down- regulated (|fold-change| ≥ 4.0, P ≤ 0.005) by OHMs 1, 2 and 4. Numbers inside the intersections represent hypoxia-induced transcripts affected by corresponding treatments. In A549 cells several genes are up-regulated that constitute a signature of lung cancer. Treatment with OHM 1 resulted in the change of 32 transcripts by at least 4-fold (P ≤ 0.005) and 597 transcripts by at least 2-fold (P ≤ 0.005), 11 of which are validated as HIF1α target genes (Figure II.9C).(304) Similar results were observed for OHM 2, in which 41 transcripts were affected by treatment at least 4-fold (P ≤ 0.005) and 608 transcripts at least 2-fold. OHM 1 and OHM 2 were found to have overlap in the down-regulation of 9 91 transcripts and up-regulation of 10 transcripts by at least 4-fold (Figure II.10). In contrast, no overlap was found between OHM 1 or OHM 2 with negative control OHM 4 for down- regulated genes and only one transcript overlapped for up-regulation by 4-fold. In total, treatment with OHM 4 changed expression of 27 transcripts by 4-fold and 91 transcripts by 2-fold. We compared our microarray results with oxopiperazines with other direct acting inhibitors of hypoxia-inducible transcription reported in the literature. Dimeric epidithiodiketopiperazines (ETPs) are recent examples of allosteric small molecule inhibitors of hypoxia-inducible transcription factor complex that also target the p300 CH1 domain.(242, 243) Treatment of hypoxic MCF7 cells with dimeric ETP 2 resulted in at least 2-fold expression changes in 329 genes (Table II.2).(243) A structurally similar ETP 3 alters the expression of 409 genes by at least 2-fold.(242) A sequence-specific DNA- binding oligomer, polyamide 1, has been reported to change expression levels of 2284 genes by at least 2-fold in hypoxic U251 cells.(281) In contrast, echinomycin, a DNA- binding cyclic peptide, under the same conditions and cell line produces changes in expression of 10918 genes by at least 2-fold. RNA interference directed at HIF1α leads to expression changes in 3194 genes at least 2-fold (Table II.2). Taken together, these data suggest that the designed oxopiperazine α-helix mimetic, despite its low molecular weight and a limited number of contacts with the intended target proteins p300/CBP, shows remarkably high specificity on a genome-wide scale. 92 Table II.2. Number of affected genes in cells treated with the inhibitors of hypoxia- inducible transcription factor complex. Cell Line A549 MCF7 U251 Inhibitor OHM 1 OHM 2 ETP 2 ETP 3 Echinomycin siRNA Polyamide 1 Down-regulated 2-fold 449 391 187 113 7426 1811 1575 Up-regulated 2- fold 148 217 142 290 3492 1383 709 Total 597 608 329 403 10918 3194 2284 P-value (ANOVA) <0.005 <0.01 <0.01 <0.01 Reference Ibid. (243) (242) (281) II.2.5. In Vivo Assessment of the Efficacy of OHM in a Mouse Tumor Xenograft Model. We next sought to investigate the potential of oxopiperazine 1 to reduce tumor growth rate in a mouse xenograft model. We began by determining the maximum tolerated dose of OHM 1 in BALB/c mice. Escalating doses of 2–100 mg/kg OHM 1 were injected intraperitoneally every other day. No signs of toxicity were found, as assessed by daily weight measurements and visual inspections of the appearance and the behavior of treated mice. The animals were then injected intravenously with 10 mg/kg and then 100 mg/kg of OHM 1, again without measurable weight loss or changes in appearance or behavior (Figure II.11). 93 Figure II.11. Effect of OHM 1 treatment on the weight of BALB/c mice. Weight measurements (left axis) of control (–O–) and OHM 1 treated (–■–) mice throughout the course of the study. Dose escalation curve (−•−, right axis) indicates the dose of OHM 1 and the day of injection. Error bars are ± SEM of the weight measurements of the mice within each experimental group. II.2.6. MDA-MB-231 Mouse Tumor Xenografts. To assess the efficacy of oligooxopiperazine OHM 1 against hypoxia-induced transcription in vivo, two types of xenografts were obtained: MDA-MB-231, where the activity of the hypoxia-inducible genes is primarily driven by the HIF1 heterodimer, and 786-O renal cell carcinoma of the clear cell type, where the hypoxia-induced gene expression is also mediated by both HIF1- and HIF2-dependent mechanisms. Since both factors recruit p300/CBP in their CH1 regions, we reasoned that direct targeting of the p300 CH1 region may result in a therapeutic benefit for both types of tumors. Mice bearing MDA-MB-231 xenografts were randomly assigned to treated and control groups when the tumor volumes reached 200 mm 3 . The treated groups received intraperitoneal injections of 15 mg/kg OHM 1 in PBS, whereas control groups received 94 injections of PBS (100 μL per animal). Tumor sizes were monitored daily for both groups of mice. In the treated group, the median tumor volume was smaller (103 %) as compared to the control group (186 %), indicating 45% median tumor volume reduction at the conclusion of the experiment (Figure II.12a). Treated and control mice maintained their weights at 100 ± 15% before euthanasia (Figure II.12b). At the endpoint of the experiment, mice were injected with near-infrared (NIR) tumor-targeting contrast agent IR-783 and imaged using the Xenogen IVIS 200 system. The intensity of the NIR signal in the OHM 1 treated mice was consistently lower than vehicle-treated mice (Figure II.12c). The tumors from control and treated mice were excised, dissected and used for histopathology evaluations. We used hematoxylin and eosin (H&E) stain in order to evaluate cell morphology (Figure II.13). Cells in the treated tumors appear more differentiated with a greater cytoplasm to nucleus ratio. Cell proliferation marker Ki-67 was also used to determine the pattern of proliferating cells. Tumors in the untreated mice exhibit three fold higher levels of cell proliferation (Figure II.13b) as assessed by quantification of the Ki-67 stained images (Figure II.13c).(305) 95 Figure II.12. Effect of OHM 1 treatment on tumor growth rate in MDA-MB-231 xenografts. (A) Box-and-Whisker plots of the percentages of tumor volumes measured throughout the duration of the experiment: boxes represent the upper and lower quartiles and the median, while the error bars show maximum and minimum tumor volumes. (B) Weight measurements of control (–O–) and OHM 1 treated (–■–) mice engrafted with MDA-MB-231 tumors through the course of the study. Error bars are ± SEM of the weight measurements of the mice within each experimental group. (C) Localization of the near-infrared contrast agent IR-783 in the tumors of the control and treated mice. The fluorescence output was processed with Living Image software with one representative sample for each group presented above. Mice from the OHM 1 treated group show lower intensity of the signal originating from the tumor-accumulated contrast agent as compared to the control group. 96 Figure II.13. Histopathology data. (a) Hematoxylin and eosin (H&E)-stained sections of MDA-MB-231 xenograft (purple: nuclei, pink: cytoplasm) treated with vehicle or OHM 1. (b) Anti-Ki-67 stained MDA-MB-231 xenografts (brown stain), treated with vehicle or OHM 1. Scale bars = 50 μm. (c) Quantification of Ki-67 stained images using ImageJ with the ImmunoRatio plugin.(305) *** P < 0.001, t-test. II.2.7. 786-O Mouse Tumor Xenografts. The 786-O tumor growth rate was highly irregular. Equal numbers of mice engrafted with 786-O cells with large (>500 mm 3 ), medium (300-500 mm 3 ), and small tumors (150-300 mm 3 ) were otherwise randomly assigned to each group. The treatment started for all tumor sizes at the same time. The treated groups received injections of OHM 1 in PBS at 20 mg/kg intratumorally, whereas control groups received injections of PBS (100 µl per animal). Animal weights and tumor sizes were monitored every 2-4 days. In the treated group, the median relative tumor volume was smaller (101%) as compared to the control group (287%), indicating 65% median tumor volume reduction (P<0.001), calculated throughout all days of experiment at its conclusion (Figure II.14A). Treated and control mice maintained their weights at 95 ± 8% before euthanasia (Figure II.14B). 97 Figure II.14. Effect of OHM 1 treatment on tumor growth rate in 786-O xenografts. (A) Box-and-Whisker plots of the percentages of tumor volumes measured throughout the duration of the experiment: boxes represent the upper and lower quartiles and the median, while the error bars show maximum and minimum tumor volumes. (B) Relative tumor volume and (C) weight measurements of control (–O–) and OHM 1 treated (–■–) mice engrafted with 786-O tumors through the course of the study (in % compared to Day 0). Error bars are ± SEM of the corresponding measurements of the mice within each experimental group. *** P < 0.001, t-test. 98 II.3. Discussion. In this study we describe rational design of peptidomimetics that target protein-protein interactions involved in transcriptional activation of oncogenic pathways in biochemical, cell culture, and in vivo assays. Oxopiperazine analogs, designed to mimic side chain orientations of protein helices, are shown to inhibit the interaction of hypoxia inducible factor 1α with the coactivator p300/CBP. The successful candidate, OHM 1, is a small molecule with molecular weight of less than 500 Daltons. Small molecules are often considered better candidates as inhibitors of enzyme function rather than protein-protein interactions; however, analysis of PPIs that have been inhibited by small molecules suggests that they contain hot spot residues clustered within relatively small volumes – a feature that is reminiscent of enzyme active sites.(265, 267, 268, 306) The HIF1α–p300/CBP interface features a high concentration of hot spot residues, making it an attractive candidate for inhibition by small molecules. Although other inhibitors of this transcription factor–coactivator interaction have been described,(282) the best-known example, chetomin,(227) inhibits the interaction by chelating essential zinc metals and denaturing the coactivator. We aimed to discover small molecule orthosteric inhibitors that target the interaction with high specificity. The designed compounds bind the target protein in a predictable manner as the single residue mutant OHM 3 shows an expected weaker affinity for p300 as compared to OHM 1. The p300 binding site for OHM 1 was confirmed by HSQC titration experiments. OHM 1 significantly reduces HIF promoter activity and down-regulates the expression of hypoxia inducible genes responsible for promoting angiogenesis, invasion, and glycolysis. 99 Despite its low molecular weight and a limited number of contacts with the target protein, OHM 1 shows high specificity on genome-wide scale, especially when compared to other reported inhibitors of HIF1α mediated transcription. Comparative analysis of the genome- wide effects of the OHM 1, OHM 2 and OHM 4 provide important insights into the ability of related compounds to disrupt transcriptional activity of hypoxia-inducible genes. The microarray gene expression heat map observed with the control tetraalanine OHM 4 is similar to that of the vehicle under normoxic conditions, underscoring the levels of specificity offered by the display of the correct side chains on the scaffold. To assess the in vivo potential of OHM 1, murine tumor xenografts derived from the triple- negative breast cancer cell line MDA-MB-231 were treated with the compound. Injections of OHM 1. In order to address the discrepancies and the similarities between HIF1α and HIF2α regulation, we employed another animal model, murine tumor xenografts derived from the renal cell carcinoma cell line 786-O, as hypoxic response in RCC is primarily governed by HIF2α. The treatment with OHM 1 reduced the median tumor volume by 65% as compared to the untreated group. This encouraging result indicates that the binding interfaces between p300 CH1 domain and CTAD domains of HIF1α/HIF2α are identical. Slightly better efficacy of OHM 1 in the RCC mouse model may be due to its better vascularization. Importantly, the OHM 1 treatment did not cause measurable changes in animal body weight or other signs of toxicity in tumor-bearing animals, nor increased the metastasis rate. The efficacy and specificity of the OHM derivatives, in genome-wide analysis, supports our hypothesis that protein-protein interactions that offer clefts can be targeted by judiciously designed small molecules. 100 II.4. Conclusion. Protein-protein interactions are attractive targets for drug design due to their fundamental role in biological function. However, small molecules that selectively target the intended interactions have been difficult to access using traditional drug discovery approaches. We described the design, synthesis, biochemical and in vivo evaluation of a small molecule oligooxopiperazine scaffold that captures the topography of α-helices. The interaction between the helices of the CTAD domain of the HIF1α and p300 regulates the transcription of key genes, whose expression contributes to angiogenesis, metastasis, and altered energy metabolism in cancer. We designed mimics of a key α-helical domain at the interface of HIF1α and p300 to develop inhibitors of hypoxia inducible signaling, which contributes to angiogenesis, metastasis, and altered energy metabolism in cancer. The designed compounds target the desired protein with high affinity and in a predetermined manner with the optimal ligand providing effective down-regulation of hypoxia-dependent genes, and reduction of the tumor burden in two experimental mouse tumor xenograft models. 101 II.5. Experimental Section for Chapter II. General. Commercial grade solvents and reagents were used without further purification. Fmoc amino acids and peptide synthesis reagents were purchased from Novabiochem. The Hoveyda-Grubbs (second-generation) catalyst, and molecular biology grade salts and buffers were obtained from Sigma. Synthesis of oxopiperazine dimers. An Fmoc amino acid linked to Knorr Rink Amide resin was extended to a dipeptide using standard Fmoc solid phase peptide synthesis methods in a solid phase reaction vessel.(307) The resultant dipeptide was deprotected with 20% piperidine/dimethylformamide (DMF) and resin was washed sequentially with DMF, dichloromethane (DCM), methanol (MeOH), and diethyl ether and dried under vacuum. o-Nitrobenzenesulfonyl chloride (Ns-Cl, 10 eq.) and collidine (10 eq.) were dissolved in dry DCM and added to the reaction vessel. The mixture was shaken for 2 h at 23°C to obtain II. The resin was washed sequentially with DMF, DCM, MeOH, and diethyl ether and dried for 12 h under vacuum. The resin was transferred to a glass microwave tube (CEM). Triphenylphosphine (PPh3, 10 eq.) was added and the tube was flushed with nitrogen gas for 30 minutes. Tetrahydrofuran (THF), diisopropyl azodicarboxylate (DIAD, 10 eq.), and 2-bromoethanol (10 eq.) were added, and the reaction mixture was subjected to microwave irradiation (200 watts, 250 psi) for 10 minutes at 100°C. Resin was washed sequentially with THF, DMF, and DCM. The resin was transferred to a solid phase vessel and treated with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in THF for 2 h. Resin was washed with THF, DMF, DCM, and diethyl ether, allowed to dry for 30 minutes, and then treated with DBU and 2-mercaptoethanol in DMF for 2 h. Compound III was then washed with DMF, DCM, MeOH, and diethyl either and 102 dried. The desired pre-activated Fmoc-amino acid was added to the resin, and the mixture was shaken at 23°C for 12 h affording III. Nosyl protection and the ring formation steps were repeated to obtain oxopiperazine dimers V after cleavage from the resin with 95% trifluoroacetic acid (TFA), 2.5% water, and 2.5% triisopropylsilane (TIPS). OHM 1 - LLAQ 1 H-NMR (600 MHz, d6-DMSO, 100°C) δ 6.87 (br, 3H), 6.61 (br, 3H), 5.36 (t, J = 7.47, 1H), 4.90-4.85 (m, 1H), 4.67 (q, J = 6.88, 1H), 3.96 (br, 2H), 3.63-3.24 (m, 8H), 1.95-1.82 (m, 3H), 1.72-1.58 (m, 3H), 1.57-1.49 (m, 1H), 1.38 (s, 3H), 1.06-0.75 (m, 12H). HRMS (ESI) C24H42N6O5 [M+H] + calc’d= 494.3217; found= 495.3502 OHM 2 - LLAA 1 H-NMR (600 MHz, d6-DMSO, 100°C) δ 6.88 (br, 2H), 5.37 (t, J = 7.20, 1H), 4.92 (q, J = 7.25, 0.8H), 4.87 (q, J = 7.25, 0.2H), 4.65 (q, J = 6.91, 1H), 4.00 (br, 1H), 3.88 (br, 1H), 3.57-3.34 (m, 7H), 3.30-3.23 (m, 1H), 1.92-1.80 (m, 2H), 1.69-1.57 (m, 3H), 1.56-1.48 (m, 1H), 1.35 (d, J = 6.88, 3H), 1.29 (d, J = 7.27, 3H), 0.95 (d, J = 6.80, 4H), 0.93 (d, J = 6.34, 5H), 0.90 (d, 6.34, 3H). HRMS (ESI) C22H39N5O4 [M+H] + calc’d= 438.3002; found= 438.3118 OHM 3 - LAAQ 1 H-NMR (400 MHz, d6-DMSO, 100°C) δ 6.77 (br, 2H), 6.52 (br, 2H), 5.34 (q, J = 6.91, 1H), 4.90-4.84 (m, 1H), 4.67 (q, J = 6.91, 1H), 4.01-3.91 (m, 1H), 3.76-3.69 (m, 1H), 3.57- 3.27 (m, 7H), 3.24-3.14 (m, 1H), 2.19-2.01 (m, 3H), 2.00-1.80 (m, 3H), 1.64-1.55 (m, 1H), 103 1.38 (d, J = 6.92, 3H), 1.27 (d, J = 6.92, 3H), 0.94 (t, J = 6.02, 6H). HRMS (ESI) C21H36N6O5 [M+H] + calc’d= 453.2747; found= 453.2863 OHM 4 - AAAA 1 H-NMR (400 MHz, d6-DMSO, 100°C) δ 6.71 (br, 2H), 5.34 (q, J = 6.87, 1H), 4.91 (q, J = 7.13, 1H), 4.65 (q, J = 6.94, 13.68, 1H), 4.05-3.91 (m, 1H), 3.82-3.79 (m, 1H), 3.53-3.38 (m, 4H), 3.37-3.33 (m, 2H), 3.31-3.25 (m, 1H), 3.20-3.12 (m, 1H), 1.38 (d, J = 6.97, 3H), 1.35 (d, J = 6.97, 3H), 1.29 (d, J = 7.20, 3H), 1.26 (d, J = 6.97, 3H). HRMS (ESI) C16H27N5O4 [M+H] + calc’d= 354.2063; found= 354.2139 Plasmids. The DNA sequence of human p300 CH1 domain (amino acid residues 323-423) was subcloned into a pUC57 plasmid by Genscript, Inc. After transformation of JM109 bacteria (Promega) with the plasmid, it was amplified and purified. Then the gene of interest was subcloned between BamHI and EcoRI restriction sites of pGEX-4T-2 expression vector (Amersham). Cloning and Expression of 15 N p300–CH1. The pGEX-4T-2-p300 fusion vector was transformed into BL21(DE3) competent E.coli (Novagen) in M9 minimal media with 15 NH4Cl as the primary nitrogen source. Protein production was induced with 1 mM IPTG at OD600 of 1 for 16 h at 15°C. Production of the desired p300-CH1-GST fusion product was verified by SDS-PAGE. Bacteria were harvested and resuspended in the lysis buffer with 20 mM Phosphate buffer (Research Products International, Corp.), 100 µM DTT (Fisher), 100 µM ZnSO4 (Sigma), 0.5% TritonX 100 (Sigma), 1 mg/mL Pepstatin A (Research Products International, Corp.), 10 mg/mL Leupeptin A (Research Products International, Corp.), 500 µM PMSF (Sigma), and 0.5% glycerol at pH 8.0. Pellets were 104 lysed by sonication and centrifuged at 4 o C and 20,000 rpm for 20 min. Fusion protein was collected from the bacterial supernatant and purified by affinity chromatography using glutathione Sepharose 4B beads (Amersham) prepared according to the manufacturer’s directions. GST-tag was cleaved by thrombin, and protein was eluted from the resin. Collected fractions were assayed by SDS-PAGE gel; pooled fractions were treated with protease inhibitor cocktail (Sigma) and dialyzed against a buffer containing 10 mM Tris, 50 mM NaCl, 2 mM DTT (Fisher) and 3 equivalents ZnSO4 at pH 8.0 to ensure proper folding (vide supra). Tryptophan Fluorescence Binding Assay. Spectra were recorded on a QuantaMaster 40 spectrofluorometer (Photon Technology International) in a 10 mm quartz fluorometer cell at 25°C with 4 nm excitation and 4 nm emission slit widths from 200 to 400 nm at intervals of 1 nm/s. Samples were excited at 295 nm and fluorescence emission was measured from 200-400 nm and recorded at 335 nm. OHM stock solutions were prepared in DMSO. Aliquots containing 1 μL DMSO stocks were added to 400 μL of 1 μM p300-CH1 in 50 mM Tris and 100 mM NaCl (pH 8.0). After each addition, the sample was allowed to equilibrate for 5 minutes before UV analysis. Background absorbance and sample dilution effects were subtracted by titrating DMSO into p300-CH1 in an analogous manner. Final fluorescence intensity is reported as the absolute value of [(F1-F0)/F1]*100, where F1 is the final fluorescence upon titration, and F0 is the fluorescence of the blank DMSO titration. EC50 values for each compound were determined by fitting the experimental data to a sigmoidal dose-response nonlinear regression model in GraphPad Prism 5.0, and the dissociation constants, KD, were obtained from equation (1). 105 = ( ×1 − + × ) − () where, P = Total concentration of protein F = Fraction of bound peptide = 0.5 1 H- 15 N HSQC NMR Spectroscopy. Protein samples were prepared as described above. Uniformly 15 N-labelled p300-CH1 was concentrated to 69 μM in NMR buffer (10 mM Tris, pH 8, 50 mM NaCl, 2 mM DTT, and 207 μM ZnSO4) using 3 kDa MWCO Amicon Ultra centrifugal filter (Millipore) and supplemented with 5% D2O. For HSQC titration experiments, data was collected on a 600 MHz Bruker four-channel NMR system at 25 ºC and analyzed with the TopSpin software (Bruker). For the HSQC titration experiments, 0, 1.2 and 7 molar equivalents of OHM 1 in DMSO were added to 15 N-labelled p300-CH1, and the data was collected as described above. Mean chemical shift difference (∆δNH) observed for 1 H and 15 N nuclei of various resonances were calculated as described,(308) where α is the range of H ppm shifts divided by the range of NH ppm shifts (α = 1/8). = 1 2 ∙ [ + ∙ ] Cell lines. Human alveolar basal epithelial adenocarcinoma (A549), human breast adenocarcinoma (MCF7), and human renal cell adenocarcinoma (786-O) cell lines were obtained from ATCC. Human breast epithelial adenocarcinoma cells stably transfected 106 with a hypoxia response element (HRE) luciferase construct (MDA-MB-231-HRE-Luc) was a gift of Dr. Robert Gillies. Cell culture. MCF7 cells were grown in RPMI 1640 (Invitrogen) supplemented with 10% of fetal bovine serum (FBS, Irvine Scientific), 50 units/mL penicillin and 50 µg/mL streptomycin (Pen-Strep, Invitrogen). A549 cells were grown in RPMI 1640 supplemented with 2% or 0.2% of FBS, 50 units/mL penicillin and 50 µg/mL streptomycin. 786-O cells were grown in RPMI 1640 supplemented with 10% of FBS, 50 units/mL penicillin and 50 µg/mL streptomycin. MDA-MB-231-HRE-Luc cells were grown in high glucose Dulbecco's Modified Eagle's Medium (DMEM, Invitrogen) supplemented with 10% FBS and 0.2 g/L geneticin (RPI). All cells were incubated at 37 o C in a humidified atmosphere with 5% CO2. Hypoxia was induced by GasPak EZ Anaerobe Pouch System (BD Biosciences), unless mentioned otherwise. Cell growth and morphology were monitored by bright field microscopy. Cells were detached using trypsin in PBS (0.05%, Invitrogen). Preparation of OHM compounds for in vitro assays. Stock solutions of OHM compounds (1 mM) were prepared in a buffer, containing 150 mM NaCl, 50 mM Tris, 1 mM DTT, 100 µM ZnCl2, 0.05-0.15% w/v NonidetP40, and 10% v/v glycerol for the preliminary biophysical studies and used without further changes for all in vitro assays. Cell viability assay. MCF7 cells were seeded in a 96-well plate at a density of 1.4×10 4 cells per well in 200 µl of medium and allowed to form a monolayer for 48 h. After that, the medium was replaced with 150 µL of fresh medium containing OHM 1 or OHM 2 at a concentration range from 0.5 µM to 60 µM and 0.1 % dimethyl sulfoxide (DMSO). After 48 h of incubation with compounds, a solution of 3-(4,5-dimethylthiazol-2-yl)-2,5- 107 diphenyltetrazolium bromide (MTT, Sigma) was added to each well (17 µL of 5 mg/mL, in PBS) and incubated at 37°C and 5 % CO2 for an additional 3 h. After that, the medium was carefully removed and purple formazan crystals were dissolved in DMSO (100 µl per well). The absorbance was measured at 570 nm with a correction at 690 nm in order to quantify the amount of formazan. All experiments were performed in quadruplicate. A549 cells were seeded in a 96-well plate at a density of 1.5×10 4 cells per well in 200 µl of medium with 2% FBS and allowed to form a monolayer for 24 h. After that, the medium was replaced with 150 µL of fresh medium containing 0.2% FBS, OHM 1, OHM 2, OHM 3, and OHM 4 (concentrations ranging from 0.01 µM to 100 µM) and 0.1% DMSO. After 48 h of incubation with compounds, a solution of 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT, Sigma) was added to each well (17 µL of 5 mg/mL, in PBS) and incubated at 37°C and 5 % CO2 for an additional 3 h. After that, the medium was carefully removed and purple formazan crystals were dissolved in DMSO (100 µL per well). The absorbance was measured at 570 nm with a correction at 690 nm in order to quantify the amount of formazan. All experiments were performed in quadruplicate. Bicinchoninic acid (BCA) assay. Cell lysate (10 µL) was added to the BCA reagent (200 µL), prepared as per the manufacturer’s protocol (Thermo Scientific). BSA standard solutions were prepared at a concentration range of 25 µg/mL to 2000 µg/mL. Absorbance was measured at 562 nm using a Synergy 2 microplate reader (BioTek). Sample concentrations were determined from a calibration curve. All experiments were performed in triplicate. 108 Analysis of the hypoxia-inducible promoter activity with luciferase assays. MDA- MB-231-HRE-Luc cells were seeded in 24-well plates (BD Falcon) at a density of 6.5×10 4 cells per well in 1 mL of medium. Cells were allowed to adhere and form monolayer for 48 h before the addition of the compounds (~80% confluence). After that, the cells were treated with 1 mL of the fresh medium containing 10% FBS, OHM 1, OHM 2, or OHM 4 at a range of concentrations from 0.1 µM to 20 µM, 0.1% DMSO; vehicle samples were treated with cell culture medium containing 0.1% DMSO. Cells were incubated for 6 h at 37°C and 5% CO2 and then hypoxia was induced for another 18 h. The lysates were isolated by using cell culture lysis reagent (Promega), supplemented with Halt protease inhibitor cocktail (Thermo Scientific). Cell lysate from each well was collected into a separate ice-cold polypropylene tube (USA Scientific) and centrifuged at 13,000 rpm for 10 min at 4°C. Supernatant was then transferred into another set of cold polypropylene tubes and the pellet was discarded. Luciferase assay reagent (100 µL, Promega) was added to cell lysate (20 µL) and the luminescence intensity was measured by Turner TD-20e luminometer. The results were normalized to the total protein concentration determined by the BCA assay. All experiments were performed in triplicate. Isolation of mRNA. A549 cells were seeded in 6-well dishes (Greiner) at a density of 2×10 5 cells per well in 2 mL of medium with 2% FBS and allowed to form a monolayer (~90% confluent) for 72 h. After attachment, cells were treated with 1.5 mL of fresh medium containing 0.2% FBS, OHM 1, OHM 2, OHM 3, and OHM 4 at concentrations of 5 µM and 10 µM, and 0.1% DMSO; vehicle samples were treated with cell culture medium containing 0.1% DMSO. After 6 h of incubation, hypoxia was induced and cells were incubated for another 42 h. Cells were washed twice with ice-cold PBS and total mRNA 109 was isolated with an RNeasy kit (Qiagen) according to the manufacturer’s instructions. The mRNA was further treated with DNase I (Invitrogen, Turbo DNA-free kit) to remove any remaining genomic DNA. Then the mRNA was quantified by UV absorbance at 260 nm. Reverse transcription was performed with Superscript III Reverse Transcriptase (Invitrogen) as recommended by the manufacturer. All experiments were performed in quadruplicate. Analysis of gene expression. Real-time qRT-PCR was used to determine the effect of OHM 1, OHM 2, OHM 3, and OHM 4 on VEGF, LOX, and SLC2A1 (GLUT1) genes under normoxia and hypoxia. 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. For GLUT1, the following primers were used: forward 5´- AGT ATG TGG AGC AAC TGT GTG G-3´ and 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 product of 73 bp. The mRNA levels were normalized to the expression levels of a housekeeping gene, β-glucuronidase. For β- glucuronidase the following primers was designed and used: forward 5´-CTC ATT TGG AAT TTT GCC GAT T-3´ and reverse 5´- CCG AGT GAA GAT CCC CTT TTT A-3´. Reactions were performed with Fast SYBR Green Master Mix (Applied Biosystems). Temperature cycling and detection of the SYBR green emission were performed with an ABI 7900HT Fast Real-Time PCR System. Analysis of the data was performed with Applied Biosystems Sequence Detection System, version 2.3. All experiments were performed in quadruplicate. 110 Western blot analysis of HIF1α levels. A549 cells were seeded in a 60 mm dish (VWR) and allowed to reach 80% confluence. Cells were treated with vehicle or OHM 1 at 10 µM concentration in the cell culture medium containing 2% FBS and 0.1% DMSO. Cells were incubated for 6 h and hypoxia was induced with 300 µM of DFO or GasPak EZ Anaerobe Pouch System. After incubation for an additional 18 h, the cells were washed twice with ice-cold PBS and then lysed with the cell culture lysis reagent (Promega). To ensure equal loading, protein concentration was determined by the BCA assay. A 1.5 mm 10% acrylamide denaturing gel was cast and an aliquot of each sample containing 30 µg of protein was loaded into the gel. The SDS-PAGE was carried out and then the gel was electroblotted onto the PVDF membrane (BioRad). After the transfer, the membrane was rinsed with tris-buffered saline with tween-20 (TBST) buffer and incubated with 5 % milk in TBST for 1 h. The membrane was then probed for HIF1α with a monoclonal mouse anti-human HIF1α antibody (BD Biosciences) or for β-actin (loading control) with a polyclonal rabbit anti-human β-actin antibody (Cell Signaling) overnight at 4°C and gentle rocking. Membrane was further washed three times with TBST for 10 min and incubated with horseradish peroxidase (HRP) conjugated secondary anti-mouse or anti-rabbit antibody (Santa Cruz Biotechnology), respectively. Signals were detected by treating the membrane with Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare) for 3 min followed up by the exposure of the CL-Xposure film (Thermo Scientific) to the membrane. Gene expression profiling. Experiments were carried out with A549 cells. The medium, time course, hypoxia induction, small molecule treatment, and mRNA isolation were conducted as described above in the “mRNA isolation” section. Cultured cells contained 111 vehicle, OHM 1, OHM 2, or OHM 4 at a concentration of 10 µM. Sample preparation and microarray analysis was performed at the Genome Technology Center, New York University School of Medicine. Labeled mRNA was hybridized to Affymetrix Genechip Human Gene 1.0 ST microarrays. Four data sets were collected: normoxic cells with vehicle, hypoxic cells with vehicle, hypoxic cells with OHM 1 and hypoxic cells with OHM 2, respectively. Gene expression was analyzed using GeneSpring GX 12.5 software (Agilent). Probe level data have been converted to expression values using a robust multi- array average (RMA) preprocessing procedure on the core probe sets and baseline transformation to median of all samples. A low-level filter removed the lowest 20 th percentile of all the intensity values and generated a profile plot of filtered entities. Significance analysis was performed by one-way ANOVA test with Benjamini-Hochberg correction and asymptotic P-value computation. Fold change analysis was applied to identify genes with expression ratios above 2-fold between treatments and control set (P < 0.05). Hierarchical agglomerative clustering was performed using Pearson's centered correlation coefficient and average-linkage as distance and linkage methods. Microarray Data. Gene expression profiling data reported in this manuscript have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/GEO (accession no. GSE48134). In vivo toxicity dose-finding study. To determine the maximum tolerated dose of OHM 1, 4 male BALB/c mice were injected intraperitoneally every other day with the escalating doses of OHM 1 (in mg/kg): 2, 4, 6.6, 10, 14, 24, 32, and 100. The animals were then injected intravenously with 10 mg/kg and then 100 mg/kg of OHM 1. Toxicity after each 112 injection of OHM 1 was assessed by daily weight measurements and visual inspection of treated mice. In vivo efficacy testing of OHM 1 in mouse xenograft tumor models. The T-cell deficient mice CrTac:NCr-Foxn1 nu (Taconic, Inc.) were used. The mice were housed in an A.L.A.C.C. approved barrier facility under the direct supervision of a professional veterinarian. 786-O Xenografts: Mice (n=20) were inoculated into the right flank with 786-O cells (2 × 10 6 ), which were allowed to grow into tumors for 17 days. After that, equal number of animals with large (>500 mm 3 ), medium (300-500 mm 3 ), and small tumors (150-300 mm 3 ) were otherwise randomly assigned to each treatment group. The treatment started for all tumor sizes at the same time. The primary endpoints of efficacy (the tumor volume and tumor volume rates of increase as compared to control) were evaluated when mice (n=4) were treated with OHM 1 at 20 mg/kg dissolved in sterile PBS given parenterally on Days 0, 4, 8, 12, 16, 20, 25, 28, 33, a total of 9 injections. In parallel, a control group (n=4) received injections of PBS (100 µl per animal). Tumor sizes were measured on Days 0, 2, 3, 4, 6, 8, 11, 13, 16, 20, 25, 28, and 33 using Vernier calipers. To address the question of whether tumor growth is affected by the treatment with OHM 1, a comparison of the tumor volumes of the control group and group treated with OHM 1 was made. MDA-MB-231 Xenografts: Mice (n=31) were inoculated with MDA-MB-231 cells (5 × 10 6 ) into the right flank and allowed to grow into tumors. After the tumors reached 200 mm 3 , animals were randomly assigned to the treatment groups. The primary endpoint of efficacy (the tumor volume rate of increase as compared to control) were evaluated when 113 mice (n=4) were treated with OHM 1 at 15 mg/kg dissolved in sterile PBS given parenterally on Days 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, a total of 15 injections. In parallel, a control group (n=4) received injections of PBS (100 µl per animal). Tumor sizes were measured daily using Vernier calipers. To address the question of whether tumor growth is affected by the treatment with OHM 1, a comparison of the tumor volumes of the control group and group treated with OHM 1 was made. Imaging. At the experimental endpoint of the in vivo efficacy study, mice were injected intraperitoneally with the tumor-targeting near-infrared dye IR-783 and imaged using Xenogen IVIS 200 small animal imager. Euthanasia was performed as recommended by the American Veterinary Panel (AVMA 202229-249, 1993). Tumors and organs (liver, kidneys, heart, and lungs) were collected. Tumors were examined in a histopathology study (vide infra). Immunohistochemistry. Tumor tissues were excised and fixed with 10% formalin, embedded in paraffin, and sectioned using a standard histological procedure. For overall morphological observations, the tissue sections were stained with hematoxylin and eosin (H&E). For Ki-67 staining, paraffin sections were deparaffinized in xylene and hydrated in decreasing concentrations of aqueous ethanol. The slides were immersed in 3% hydrogen peroxide (Sigma) for 20 min to block endogenous peroxidase activity and then washed in PBS. For antigen retrieval, the slides were placed in preheated working solution of Retrievagen A (BD Pharmingen, San Jose, CA) and heated in a steamer for 70 min. After cooling for 20 min at room temperature, slides were rinsed with PBS, treated with 10% normal FBS for 30 min, and then incubated with anti-Ki-67 antibodies for 2 h. Slides were then washed with PBS and incubated with the HRP-labeled goat anti-rabbit IgG 114 antibodies for 1 h at room temperature. After washing with PBS, a streptavidin-HRP (BD Pharmingen) was added and incubated for 30 min. Slides were then stained with 3,3´-diaminobenzidine (Vector Laboratories, Burlingame, CA) for 3 min. Counterstaining was performed with hematoxylin (Vector Laboratories). After washing with distilled water, the slides were dehydrated in increasing grades of ethanol, cleared with xylene, and mounted using permanent mounting medium (Vector Laboratories). The proliferation index was determined by measuring the percentage of Ki-67 positive cells. A total of 24 randomly selected fields at 20× objective magnification from the tumors of each treatment group were examined. The pictures were quantified with ImmunoRatio plugin for ImageJ.(305) The data were plotted as a mean ± s.e.m. and analyzed for significance with the unpaired two-tailed t-test. 115 OHM 1 1 H-NMR (600 MHz, d6-DMSO, 100°C) 116 OHM 2 1 H-NMR (600 MHz, d6-DMSO, 100°C) 117 OHM 3 1 H-NMR (400 MHz, d6-DMSO, 100°C) 118 OHM 4 1 H-NMR (400 MHz, d6-DMSO, 100°C) 119 Chapter III. In vivo Anti-Cancer Activity of Rhomboidal Pt(II) Metallacyclic Assemblies. 120 III.1. Introduction. Ordered molecular assemblies of nanoscale size with well-defined geometry are rapidly becoming a new paradigm in drug delivery. An essential feature of such particles is their ability for site-specific delivery, not only to cells of the desired organ, but also to a particular subcellular compartment. To date, liposomes, which are self-assembled lipid nanoparticles held together by weak interactions are among the most widely studied and clinically successful nanoparticle-based carriers for drugs. Their use largely allows drugs to achieve sustained plasma levels while encapsulated, with the size preventing the rapid clearance by the kidneys, which often occurs to the free drug. However, the application of lipid-based drug carriers for medical purposes has been hindered by their poor loading capacity (<5 wt %) and issues related to passing through biological barriers.(309, 310) Inorganic and hybrid porous materials, such as molecular organic frameworks (MOFs), have also shown promise due to their higher loading capacities (>25 wt %).(311-313) Historically, MOFs have been viewed as unstable in either aqueous or biological milieu, a reputation garnered from the poor hydrolytic stability of MOF-5.(314, 315) Recent studies on MIL-100(Cr) and MIL-100(Fe) (MIL stands for material from Institute Lavoisier), however, suggest that MOFs can persist in biologically relevant environments and act as effective nanocarriers of some anticancer and antiviral therapeutic agents.(316, 317) As such, investigations into the biomedical applications of supramolecular coordination complexes (SCCs) that preserve the attractive properties of MOFs, such as facile building block modularity, yet afford increased solubility and lend themselves to small-molecule characterization techniques due to their well-defined structure, have seen intense growth in the past 5 years.(318-330) 121 Although development of SCCs for biomedical applications is in its infancy, some SCCs, such as trigonal prisms, self-assembled from p-cymene ruthenium-based metal fragments with pyridyl donors, have demonstrated the ability to act as effective nanocarriers of potential therapeutic agents.(331-333) Moreover, a library of p-cymene ruthenium-based polygons and cages that displays an array of cytotoxic profiles was synthesized by Stang, Chi and coworkers.(318) However, for medical applications, it would be advantageous to have the information about the cellular uptake, delivery of a guest, and metabolism of the drug delivery vehicle. Currently, the fate of SCCs in biological environments is not understood well and further investigations are warranted. In a rare report, only Therrien and coworkers systematically investigated the structural stability of a water-soluble, hexacationic ruthenium-based trigonal prism, and it was determined that the ruthenium- based trigonal prisms decompose in the presence of histidine, lysine and arginine.(334) As such, in order to investigate the biocompatibility of a SCC, a method for the real-time monitoring of the structural integrity of SCCs in vitro is urgently needed. Despite the well-known cytotoxic properties of mono- and multi-nuclear platinum complexes,(335, 336) reports on the cytotoxicity of platinum-based SCCs are rare.(324, 337) Moreover, reports have demonstrated that platinum-based SCCs can act as effective hosts for guests and display interesting photo-physical properties.(338-343) In particular, we have developed highly emissive rhomboids based on aniline-containing donors and Pt- based metal acceptors that display significantly different photophysical properties than its constituent subunits.(341) As such, this system was determined to be an attractive target to probe the cytotoxicity while spectroscopically interrogating the structural integrity of platinum-based SCCs in vitro. Herein, for the first time, the structural stability of SCCs in 122 vitro is determined in real time by using laser-scanning confocal microscopy, evidence that these structures are internalized and remain intact is shown, and their in vivo anticancer efficacy is demonstrated. 123 III.2. Results. III.2.1. Preparation and evaluation of Pt-based SCCs. Scheme III.1. Synthesis of SCCs 4 and 5. The synthesis of D2h endohedral amine-functionalized Pt-based rhomboid (Scheme III.1; 4) was accomplished by stirring 2,6-bis(pyrid-4-ylethynyl) aniline (Scheme III.1; 1) and 2,9-bis[trans-Pt(PEt3)2NO3] phenanthrene (Scheme III.1; 3) in methanol at 50°C for 24 h.10c Rhomboid 4 was determined to have a low energy absorption and emission band maxima in dichloromethane (Table III.1) at 430 and 522 nm, respectively.(341) The low energy absorption and emission band maxima of ligand 1 were determined to be blue- shifted when compared to those of 4 by 57 nm (3554 cm -1 ) and 100 nm (4540 cm -1 ), respectively. 124 Table III.1. Molar absorption coefficients and emission band maxima for compounds 1,2,4,5. Compound Absorption Bands λmax / (nm) [ε × 10 -3 /(cm -1 M -1 )] λex / (nm) λem / (nm) 1 a,d 373 [13.7] 356 422 2 b,c 390 [16.7] 390 458 4 a,d 317 [112], 430 [39.9] 430 522 5 b,d 318 [91.4], 432 [29.5] 430 538 a Data obtained from reference (340). b Data obtained from reference (341). c Data were obtained in DMSO. d Data were obtained in aerated methylene chloride. III.2.2. Uptake and toxicity of SCCs in cells. In order to evaluate the uptake and biodistribution of 4 and 5, it was necessary to first determine their cytotoxicity. As such, a standard MTT cell viability assay was employed for the HeLa (human cervical epithelial adenocarcinoma) and A549 (human alveolar basal epithelial adenocarcinoma) cells; ligand 2 served as a control. After 48 h of incubation, both cell lines did not see any appreciable decline in cellular metabolism (Figure III.1), suggesting that ligand 2 and SCCs 4 and 5 have low cytotoxicity. 125 0.001 0.01 0.1 1 10 0 50 100 3 4 5 Concentration, μM Normalized Abs,% (A) 0.001 0.01 0.1 1 10 0 50 100 3 4 5 Concentration, μM Normalized Abs,% (B) Figure III.1. Organoplatinum rhomboid SCCs have low cytotoxicity. MTT assay data: (A) HeLa and (B) A549 cells were treated with the ligand 2 and rhomboids 4 and 5 in the range of concentrations from 1 nM and 5 µM for 48 h. The cellular uptake and localization of ligand 2 and rhomboids 4 and 5 in live cells were then probed via laser-scanning confocal microscopy (LSCM), taking advantage of the fluorescent nature of the compounds. For this assay, two cell lines were chosen: HeLa and A549 (vide supra). Cells were seeded in glass bottom microscopy dishes, and after attachment, the cells were incubated with rhomboids 4 and 5 or ligand 2 (negative control) for 4 h. Upon irradiation at an excitation wavelength (λ) of 488 nm, the emission of 4 and 5 was monitored at 514 nm and 525 nm, respectively, while the background emission of the ligand 2 (λex = 390 nm, λem= 458 nm) was utilized as a control. The confocal images of the HeLa cells revealed that both SCCs rapidly localize within cells, forming a punctate pattern due to lysosomal localization (SSC 4: Figure III.3 (A-C); SSC 5: Figure III.2 (B- D)). The signal of 4 and 5 was also readily detectable in A549 cells (SSC 4: Figure III.3 (D-F); SSC 5: Figure III.2 (F-H)). Furthermore, it was determined that 4 and 5 were photochemically stable inside the cells with no appreciable photobleaching detected. The emission of ligand 2 within the range of λem = 514 nm – 525 nm, as expected, was below 126 the detection limit of the confocal instrument upon excitation at λex = 488 nm (Figure III.2 (A, E)) since it does not absorb appreciably at the excitation wavelength. Figure III.2. LSCM images of SCCs localized in live cells. A. Overlay of confocal and differential interference contrast (DIC) images of HeLa cells treated with ligand 2. B-D. Images of HeLa cells treated with SCC 5 (B. Confocal image; D. DIC image; C. Overlay of the signals from the fields B and D). E. Overlay of confocal DIC images of A549 cells treated with ligand 2. F-H. A549 cells treated with SCC 5 (F. Confocal image; H. DIC image; G. Overlay of the signals from the fields F and H). Scale bars: 10 µm. Figure III.3. LSCM images of SCCs localized in live cells. A-C. Images of HeLa cells treated with SCC 4 (A. Confocal image; C. DIC image; B. Overlay of the signals from the fields A and C). D-F. A549 cells treated with SCC 4 (D. Confocal image; F. DIC image; E. Overlay of the signals from the fields D and F). Scale bars: 10 µm. 127 III.2.3. In Vivo Assessment of the Efficacy of SCCs in a Mouse Tumor Xenograft Model. We next sought to investigate the potential of SCC 4 to reduce tumor growth rate in a mouse xenograft model. Due to the lack of systemic toxicity in cell culture, we opted for not conducting the maximum tolerated dose study. Xenograft models of MDA-MB-231 cells were utilized for the in vivo efficacy study. Mice bearing xenografts were randomly assigned to treated and control groups when the tumor volumes reached 200 mm 3 . The solubility of SCC 4 is limited at 0.6 mg/ml in PBS:DMSO (1:1 v/v), which translates into the maximum of 300 µl per average animal, weighing 30 g. As such, the treated groups received intraperitoneal injections of 6 mg/kg SCC 4 in PBS:DMSO (1:1 v/v), whereas control groups received injections of PBS:DMSO (1:1 v/v) of 300 μL per animal. Tumor sizes were monitored daily for both groups of mice. No signs of toxicity were observed, as assessed by daily weight measurements and visual inspections of the appearance and the behavior of treated mice. Some mice in the control group developed metastases in the vicinity of the primary tumor; however only primary tumor volumes are reported for the fairness of comparison. In the treated group, the median primary tumor volume was smaller (213 mm 3 ) as compared to the control group (401 mm 3 ), indicating 88% median tumor volume reduction (P<0.001), calculated throughout all days of experiment at its conclusion (Figure III.4A). The tumor growth inhibition (T/C %) is defined as the ratio of the median tumor volume for the treated vs. control group at a particular day. Thus smaller value of T/C % ratio reflects the better tumor growth inhibition, and the effective criteria for the T/C % ratio according to the National Cancer Institute standard is 128 <42%.(344) Notably, we achieved T/C % value of 36% at the last day of the experiment, and thus satisfied the NCI effective criteria. Treated and control mice maintained their weights at 105±3% before euthanasia (Figure III.4C). At the endpoint of the experiment, mice were injected with near-infrared (NIR) tumor-targeting contrast agent IR-783 and imaged using the Xenogen IVIS 200 system. The intensity of the NIR signal in the SCC 4 treated mice was consistently lower than vehicle-treated mice (Figure III.4E). The tumors from control and treated mice were excised and their pictures were taken (Figure III.4D). Control tumor in the picture is clearly bipartite, as it turned out that the secondary tumor (vide supra) merged with the primary one. The tumors were dissected and used for histopathology evaluations. We used hematoxylin and eosin (H&E) stain in order to evaluate cell morphology (Figure III.5A). Cells in the treated tumors appear considerably more differentiated with a greater cytoplasm to nucleus ratio. Cell proliferation marker Ki-67 was also used to determine the pattern of proliferating cells. Tumors in the untreated mice exhibit 2.4-fold higher levels of cell proliferation (Figure III.5B) as assessed by the quantification of the Ki-67 stained images (Figure III.5C).(305) 129 Figure III.4. Effect of SCC 4 treatment on tumor growth rate in MDA-MB-231 xenografts. (A) Box-and-Whisker plots of the tumor volumes measured throughout the duration of the experiment: boxes represent the upper and lower quartiles and the median, while the error bars show maximum and minimum primary tumor volumes. *** P < 0.001, t-test. (B) Tumor volume and (C) weight measurements of control (–O–) and SCC 4 treated (–■–) mice engrafted with MDA-MB-231 tumors through the course of the study. Error bars are ± SEM of the corresponding measurements of the mice within each experimental group. (D) Representative pictures of tumors, excised from control and SCC 4 treated mice. (E) Localization of the near-infrared contrast agent IR-783 in the tumors of the control and treated mice. The fluorescence output was processed with Living Image software with one representative sample for each group presented above. Mice from the SCC 4 treated group show lower intensity of the signal originating from the tumor-accumulated contrast agent as compared to the control group. 130 Figure III.5. Histopathology data. (A) Hematoxylin and eosin (H&E)-stained sections of MDA-MB-231 xenografts (purple: nuclei, pink: cytoplasm) from the control and treated with SCC 4 mice. (B) Anti-Ki-67 stained MDA-MB-231 xenografts (brown stain), from the control and treated with SCC 4 mice. Scale bars = 50 μm. (C) Quantification of Ki-67 stained images using ImageJ with the ImmunoRatio plugin.(305) *** P < 0.001, t-test. 131 III.3. Conclusion. In summary, for the first time, the cellular uptake and intracellular localization of fluorescent, rhomboidal Pt(II)-based SCCs were explored. The SCCs were determined to be water-soluble and nontoxic to cells. Rhomboids 4 and 5 also remained intact upon cellular internalization and did not photobleach or otherwise degrade. The well-defined geometry, presence of an internal cavity, and ability to emit within the visible spectrum makes endohedral amine-functionalized SCCs attractive candidates for further development as vehicles for drug delivery. Importantly, SCC 4 by itself caused a substantial 64% reduction of the tumor burden in a mouse tumor xenograft model on the last day of experiment, accompanied by a 2.4-fold reduction of the proliferative marker Ki67 and substantial improvement of cell morphology. Additionally, the presence of multiple spin-active Pt nuclei opens an intriguing possibility for future development of such assemblies for image-guided drug delivery (IGDD).(345, 346) 132 III.4. Experimental Section for Chapter III. Cell lines. Human cervical epithelioid carcinoma (HeLa) and human alveolar basal epithelial adenocarcinoma (A549) cell lines were obtained from ATCC. Human breast epithelial adenocarcinoma cells stably transfected with a hypoxia response element (HRE) luciferase construct and neomycin resistant gene (MDA-MB-231) was a gift of Dr. Robert Gillies. Cell culture. HeLa cells were grown in high glucose Dulbecco's Modified Eagle's Medium (DMEM, Invitrogen) supplemented with 10% of fetal bovine serum (FBS, Irvine Scientific), 50 units/mL penicillin and 50 µg/mL streptomycin (Pen-Strep, Invitrogen). A549 cells were grown in RPMI 1640 (Invitrogen) supplemented with 10% or 1% of FBS, 50 units/mL penicillin and 50 µg/mL streptomycin. MDA-MB-231 cells were grown in DMEM supplemented with 10% of FBS, and 0.4 g/L geneticin. All cells were incubated at 37°C in a humidified atmosphere with 5% CO2. Cell growth and morphology were monitored by bright field microscopy. Cells were detached using trypsin in PBS (0.05%, Invitrogen). Cell viability assays. HeLa cells were seeded in a 96-well plate at a density of 5,000 cells/well in 200 µl of medium per well with 10% FBS and allowed to form a monolayer for 72 h. Next, the medium was replaced with 150 µL of fresh medium containing 10% FBS, 2, 4, or 5 at a concentration range from 0.001 µM to 5 µM (here and below the concentration of 4 and 5 is represented as concentration of Pt in these compounds) and 0.1 % dimethyl sulfoxide (DMSO). After 48 h of incubation with compounds, 17 µL of 5 mg/mL in PBS solution of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide 133 (MTT, Sigma) was added to each well and the plates were incubated at 37°C and 5% CO2 for additional 3 h. After that, the medium was carefully removed and purple formazan crystals were dissolved in DMSO (100 µL per well). The absorbance was measured at 570 nm with a correction at 690 nm in order to quantify the amount of formazan. All experiments were performed in quadruplicate. A549 cells were added to a 96-well plate at a density of 10,000 cells/well in 200 µl per well of medium with 10% FBS and allowed to form a monolayer for 48 h. After that, the medium was replaced with 150 µL of fresh medium containing 1% FBS, 2, 4, or 5 at a concentration range from 0.001 µM to 5 µM and 0.1% DMSO. After 48 h of incubation with compounds, 17 µL of 5 mg/mL in PBS solution of 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT, Sigma) was added to each well and the plates were incubated at 37°C and 5% CO2 for additional 3 h. After that, the medium was carefully removed and purple formazan crystals were dissolved in DMSO (100 µL per well). The absorbance was measured at 570 nm with a correction at 690 nm in order to quantify the amount of formazan. All experiments were performed in quadruplicate. Confocal microscopy. HeLa and A549 cells were seeded in glass bottom microscopy dishes (MatTek) at a density of 30,000 cells/dish in 400 µl of medium supplemented with 10% FBS per dish and allowed to form a monolayer for 20 h. The medium was replaced with 400 µL of fresh medium containing 10% FBS, 2, 4, or 5 at a concentration of 5 µM (in Pt) and 0.1% DMSO and the dishes were further incubated at 37°C and 5% CO2 for additional 4 h. Imaging was performed on a Zeiss LSM 510 inverted laser-scanning confocal microscope equipped with an ×63 oil-immersion objective lens at excitation λmax = 488 nm and emission λmax = 514 nm and 525 nm, respectively. 134 In vivo efficacy testing of SCC 4 in mouse xenograft tumor models. The T-cell deficient mice CrTac:NCr-Foxn1 nu (Taconic, Inc.) were used. The mice were housed in an A.L.A.C.C. approved barrier facility under the direct supervision of a professional veterinarian. Mice (n=31) were inoculated with MDA-MB-231 cells (5 × 10 6 ) into the right flank and allowed to grow into tumors. After the tumors reached 200 mm 3 , animals were randomly assigned to the treatment groups. The primary endpoint of efficacy (the tumor volume rate of increase as compared to control) were evaluated when mice (n=6) were treated with SCC 4 at 6 mg/kg dissolved in a sterile DMSO:PBS (1:1 v/v) at 0.6 mg/mL mixture given parenterally on Days 0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, a total of 11 injections. Slight ±1 day variations were allowed for some animals. In parallel, a control group (n=5) received injections of a sterile DMSO:PBS (1:1 v/v) mixture (300 µL per animal). Length, width, and height of the primary tumorswere measured daily using Vernier calipers. To address the question of whether tumor growth is affected by the treatment with SCC 4, a comparison of the tumor volumes of the control group and group treated with SCC 4 was made. Tumor growth inhibition (T/C, %), defined as the ratio of the median tumor volume for the treated vs. control group and calculated as T/C % = [(median tumor volume of treated group at last day)/(median tumor volume of control group at last day)] × 100.(344) Imaging. At the experimental endpoint of the in vivo efficacy study, mice were injected intraperitoneally with the tumor-targeting near-infrared dye IR-783 and imaged using Xenogen IVIS 200 small animal imager. Euthanasia was performed as recommended by the American Veterinary Panel (AVMA 202229-249, 1993). Tumors and organs (liver, 135 kidneys, heart, and lungs) were collected and stored in zinc formalin fixative (Sigma). Tumors were examined in a histopathology study (vide infra). Immunohistochemistry. Tumor tissues were excised and fixed with 10% formalin, embedded in paraffin, and sectioned using a standard histological procedure. For overall morphological observations, the tissue sections were stained with hematoxylin and eosin (H&E). For Ki-67 staining, paraffin sections were deparaffinized in xylene and hydrated in decreasing concentrations of aqueous ethanol. The slides were immersed in 3% hydrogen peroxide (Sigma) for 20 min to block endogenous peroxidase activity and then washed in PBS. For antigen retrieval, the slides were placed in preheated working solution of Retrievagen A (BD Pharmingen, San Jose, CA) and heated in a steamer for 70 min. After cooling for 20 min at room temperature, slides were rinsed with PBS, treated with 10% normal FBS for 30 min, and then incubated with anti-Ki-67 antibodies for 2 h. Slides were then washed with PBS and incubated with the HRP-labeled goat anti-rabbit IgG antibodies for 1 h at room temperature. After washing with PBS, a streptavidin-HRP (BD Pharmingen) was added and incubated for 30 min. Slides were then stained with 3,3´-diaminobenzidine (Vector Laboratories, Burlingame, CA) for 3 min. Counterstaining was performed with hematoxylin (Vector Laboratories). After washing with distilled water, the slides were dehydrated in increasing grades of ethanol, cleared with xylene, and mounted using permanent mounting medium (Vector Laboratories). The proliferation index was determined by measuring the percentage of Ki-67 positive cells. A total of 16 randomly selected fields at 20× objective magnification from the tumors of each treatment group were examined. The pictures were quantified with ImmunoRatio plugin for 136 ImageJ.(305) The data were plotted as a mean ± s.e.m. and analyzed for significance with the unpaired two-tailed t-test. 137 Chapter IV. Synergistic Effect of Simultaneous Inhibition of MAOA and Hypoxia Inducible Transcription in Prostate Cancer. 138 IV.1. Introduction. Androgens play the key role in the development of male mammalian physiology, and in particular, prostate. Reduction of testosterone by 5α-reductase yields dihydrotestosterone (DHT), which is both a more potent ligand for the androgen receptors and more abundant in prostate by an order of magnitude.(347) As early as 1940, Huggins and Hodges demonstrated the dependence of prostate cancer on androgens,(348) and as it has been recently shown, besides its normal regulatory function, DHT is also implicated in benign prostate hyperplasia (BPH),(349) as men with substantially lower levels of DHT due to 5α- reductase mutation do not experience BPH or prostate cancer. As such, androgen deprivation therapy (ADT) is widely used for the treatment of prostate cancer,(350) which is the leading type of cancer in men in the United States, with 28,600 deaths in 2008.(351) However, it has been recently shown that low blood levels of testosterone significantly correlate with high risk and/or high grade prostate cancer,(352) while other studies declare no such relationship.(353) This discrepancy is directly related to the lack of understanding of the mechanistic details of the androgen dependence of prostate cancer and possibly due to adrenal steroids being subject to intra-prostate conversion, contributing to certain ADT- independent constant minimum levels of testosterone and DHT.(354) Ca. 80% of patients with advanced prostate cancer respond to the ADT, however the remission lasts only for 24 months, on average, until the cancer becomes androgen- insensitive.(355) Interestingly, intermittent ADT has shown promise, compared to the continuous therapy.(356) Adverse effects of ADT pose a serious issue, as they include mood alterations, cognitive dysfunction, erectile dysfunction, loss of bone mineral density, anemia, fatigue, metabolic syndrome, cardiovascular morbidity and mortality.(356) 139 All that evidence warrants the development of a novel approach to the treatment of prostate cancer that would employ a molecular target within a well-established cancer treatment paradigm. In particular, the most common hallmark of solid tumors is hypoxia.(54) In this work, we have uncovered a potential connection between hypoxia and a prominent marker of high-grade prostate cancer, monoamine oxidase A (MAOA). MAO is an outer mitochondrial membrane bound protein, responsible for the oxidative deamination of monoamines. Two isoforms of MAO are known: MAOA and MAOB, first distinguished on the basis of their sensitivity to a particular inhibitor – clorgyline or deprenyl, respectively.(357) Until recently, MAOA was of interest only due to its effect on neurotransmitters – serotonin, melatonin, noradrenaline, and adrenaline, in particular.(358) MAOA catalyzes the oxidation of serotonin, while MAOB is responsible for oxidation of 2-phenylethylamine and benzylamine. Other neurotransmitters are substrates for both enzymes.(358) Well-differentiated Gleason grade 3 prostate cancer is curable by surgery in 95 % cases, while as the higher grade (4 and 5) cancer component increases, the tumors respond to the treatment much worse,(359) and exact molecular mechanisms of such behavior are not well understood. MAOA was found to be one of the most highly overexpressed genes in high- grade prostate cancer (Gleason grade 4 and 5) compared to grade 3,(360) as well as bone metastatic lesions (unpublished data). Additionally, a known MAOA inhibitor, clorgyline, was found to up-regulate androgen receptor in prostate cancer, thus promoting the differentiation of cancer cells and thus suppressing the tumor growth.(361) 140 IV.2. Results. Three spontaneously established cell lines PC3, DU145, and LNCaP, are by far the most commonly used cell culture models of prostate cancer,(362) and C42B is a sub-line of LNCaP.(362) PC3, DU145, and C42B cell lines are androgen insensitive and metastatic, and MAOA activity increases in these cell lines in this order from barely detectable to abundant (Figure IV.1). PC 3 DU1 4 5 C42 B 0 2 4 6 8 10 100 110 120 130 MAOA activity (mmol/ 20 min/mg) Figure IV.1. Baseline MAOA activity in normoxia. MAOA activity as determined by the serotonin catalytic enzyme activity assay. Error bars are ± SEM of experiments performed in triplicate. Hypoxia is a well-known hallmark of solid tumors, and several transcriptional regulators of hypoxia-dependent pathways have recently been reported.(243) As such, we elected PC3, DU145, and C42B to investigate the possible interaction of MAOA and hypoxia in this study. Deferoxamine (DFO) or GasPak EZ Anaerobe Pouch System (BD) were used for hypoxia induction where appropriate to achieve the highest levels of induction of a gene of interest. 141 We have established that MAOA is induced by hypoxia in prostate cancer (Figure IV.2). Levels of MAOA mRNA increased 2.6 fold in PC3 cells under hypoxia (Figure IV.2a), which correlated with 5.7 fold increase in the MAOA protein levels (Figure IV.2b) and 1.5 fold increase in MAOA enzymatic activity (Figure IV.2c), as determined by the serotonin catalytic enzyme activity assay. We have also established the similar relationship between the increase of MAOA levels (2.3 fold) and its enzymatic activity (3.7 fold) in DU145 cell line as well (Figure IV.3). The up-regulation of MAOA correlates with the increase in hypoxia-inducible factor 1α (HIF1α) protein levels in both cell lines (Figure IV.2b, Figure IV.3a). In C42B cells, HIF1α is induced by hypoxia as well (Figure IV.4), however, it is unexpectedly quite abundant under normoxia at 63% of its hypoxic amount. As to the MAOA expression in C42B cells, its mRNA levels are not increased appreciably and reliably under hypoxia (vide infra), and we are inclined to correlate these phenomena. Figure IV.2. MAOA is induced by hypoxia in PC3 cells. (a) Hypoxia induction leads to the increase in MAOA transcription levels as determined by qRT-PCR; (b) hypoxia induction leads to the increase HIF1α and MAOA protein levels, as determined by immunoblot; (c) induction of hypoxia leads to the increase in MAOA activity. Error bars are ± SEM of experiments performed in duplicates. ** P < 0.01, * P < 0.05, t-test. 142 Figure IV.3. MAOA and HIF1α are induced by hypoxia in DU145 cells. (a) hypoxia induction by DFO leads to the increase HIF1α and MAOA protein levels, as determined by immunoblot; (b) Hypoxia induction leads to the increase in MAOA activity. Error bars are ± SEM of experiments performed in triplicate. ** P < 0.01, t-test. Figure IV.4. In C42B cells, HIF1α is abundant in normoxia, and further induced by hypoxia. Immunoblot analysis of HIF1α: (a) representative picture of the bands on the film; (b) HIF1α levels, normalized to levels of β-actin. * P < 0.05, t-test. 143 One of the recently published designed transcriptional antagonists of hypoxia inducible genes, LS72, exhibits cytotoxicity towards all three studied human prostate cancer cell lines (Figure IV.5). The cells were supplemented with fetal bovine serum (FBS) at either 1% or 10% where appropriate in order to achieve optimum cell viability where higher concentrations of LS72 were warranted. Serum-starved cells (1% FBS) exhibit substantial resistance towards LS72, tolerating two to six times the concentration, compared to the cells supplemented with 10% FBS (Figure IV.5, Table IV.1). PC3 0.001 0.01 0.1 1 10 0 50 100 1% FBS 10% FBS Concentration of LS72, μM Normalized Abs, % PC3MAOA 0.001 0.01 0.1 1 10 0 50 100 1% FBS 10% FBS Concentration of LS72, μM Normalized Abs, % DU145 0.01 0.1 1 10 0 50 100 1% FBS 10% FBS Concentration of LS72, μM Normalized Abs, % C42B 0.001 0.01 0.1 1 10 0 50 100 1% FBS 10% FBS Concentration of LS72, μM Normalized Abs, % Figure IV.5. Prostate cancer is sensitive to transcriptional antagonist of hypoxia-inducible genes LS72. Cell viability data (MTT assay) for PC3, PC3MAOA, DU145, and C42B cells treated with LS72 at 1% and 10% FBS. 144 Table IV.1. GI50 values of LS72 in prostate cancer cell lines. Cell line FBS concentration, % 1 10 PC3 490 nM 120 nM PC3MAOA 440 nM 200 nM DU145 260 nM 55 nM C42B 200 nM 80 nM To study the effect of LS72 on transcriptional activity of HIF-dependent genes in prostate cancer, the mRNA levels of three prominent HIF-dependent genes (VEGFA, LOX, and GLUT1), responsible for angiogenesis, cell motility, and metabolism switch, were measured in PC3 and C42B cell lines (Figure IV.6), as they represent the opposite sides of the MAOA activity spectrum. The PC3 cells were treated with LS72 at 200 and 400 nM; the treatment resulted in a dose-dependent down-regulation of the transcription levels of all three genes under hypoxia, by as much as 98% for VEGFA, 88% for LOX, and 94% for GLUT1. Similarly, the levels of the three HIF-dependent genes were measured in C42B cell line. The cells were treated with LS72 at 30, 60, and 200 nM. Again, the expression levels of all three hypoxia markers went down in presence of LS72 under hypoxia, by as much as 38% for VEGFA, 41% for LOX, and 35% for GLUT1. We have also established that MAOA is significantly down-regulated by LS72 in C42B cells by as much as 37% (Figure IV.7). This effect is observed under both normoxia and hypoxia, consistent with the high basal levels of HIF1α and MAOA under normoxia. Thus in C42B cells HIF1α is abundant under normoxia, MAOA transcription is hypoxia- independent, and MAOA levels decline when the cells are treated with the known inhibitor 145 of HIF1α transcriptional activity. These facts strongly suggests the dependence of MAOA transcription on HIF1α. Figure IV.6. LS72 treatment decreases hypoxia-induced transcription levels of HIF- dependent genes (VEGFA, LOX, GLUT1) in PC3 and C42B cells. Expression levels of HIF-dependent genes in PC3 (a-c) and C42B (d-f) cells treated with LS72 at different concentrations. The values were normalized to corresponding mean values of vehicle hypoxia samples. Error bars are ± SEM. *** P < 0.001, ** P < 0.01, * P < 0.05, t-test. It is noteworthy, that in PC3 cells LS72 causes considerable relative elevation of MAOA levels (Figure IV.8). This phenomenon is likely to be rooted in the fact that the absolute MAOA expression levels in PC3 cells (average cycle threshold Ct is 27) under hypoxia are much lower than in C42B cells (Ct 21), consistent with the dramatic difference in activity (vide supra). Upon treatment with LS72, levels of MAOA in PC3 rise (Ct 24); nevertheless, they are still much lower than the down-regulated levels in C42B cells (Ct 22). LS72 has been reported to exert general non-specific cell toxicity,(243) resulting, in particular, in up- 146 regulation of HIF-dependent genes under normoxia, when their basal levels are very low. Lastly, inconsistent dose-independent MAOA up-regulation supports this hypothesis. Figure IV.7. MAOA is down-regulated by LS72 in a dose dependent manner. Levels of MAOA in C42B cells treated with LS72 at different concentrations were determined by qRT-PCR. Error bars are ± SEM. ** P < 0.01, * P < 0.05, t-test. Figure IV.8. MAOA is up-regulated by LS72 in PC3 cells. Levels of MAOA in PC3 cells treated with LS72 at different concentrations were determined by qRT-PCR. Error bars are ± SEM. *** P < 0.001, ** P < 0.01, t-test. 147 In C42B cells, MAOA inhibitor clorgyline does not significantly affect MAOA levels, however it does decrease the transcription levels of HIF-dependent gene LOX (Figure IV.9) rather dramatically (by 82%). It prompted us to investigate the effect of LS72 and clorgyline on the MAOA and LOX transcription levels in C42B cells. We have established that these compounds, when applied in combination, down-regulate both MAOA and LOX in a synergistic manner, resulting in a profound 69% decrease in MAOA mRNA levels (compared to just 22% with LS72 alone, Figure IV.9a), and 87% decrease in transcription levels of LOX (compared to 66% with LS72 alone, Figure IV.9b). It is noteworthy, that the combination treatment yields significantly lower levels of both MAOA and LOX than LS72 (by 61 and 71%, respectively) or clorgyline (by 78 and 27%, respectively) treatment alone. Figure IV.9. In C42B cells, MAOA inhibitor clorgyline and LS72 down-regulate transcription levels of (a) MAOA and (b) HIF-dependent gene LOX, as determined by qRT- PCR. The cells were treated with clorgyline at 30 μM and LS72 at 200 nM for 24 h in the medium with 10% FBS. Hypoxia was induced with DFO. Error bars are ± SEM of experiments performed in quadruplicate. *** P < 0.001, ** P < 0.01, * P < 0.05, t-test, unless specified, compared to Vehicle Hypoxia sample. 148 We have established that the treatment conditions that are anticipated to mimic the tumor microenvironment better are yielding more reliable results. In particular, we varied the levels of FBS and employed two methods of hypoxia induction. It is known that solid tumors appear to be deprived of oxygen and nutrients, and as such, serum starvation and oxygen deprivation appear to be the most adequate conditions for cancer cell lines.(363) As mentioned above, the serum starvation leads to the increased resistance of the cancer cells to the LS72 treatment, however this phenomenon allows for the attainment of higher concentration of LS72 in the medium (Table IV.1, vide supra). Lower systemic toxicity is beneficial, as the cellular response at elevated concentrations of LS72 (200 nM) appears to be more pronounced (Figure IV.7). We mimicked hypoxia with either deferoxamine (300 µM) or by oxygen deprivation in GasPak EZ pouch (BD). Deferoxamine chelates and inactivates prolyl hydroxylases that leads to the accumulation of HIF1α, however the off-target effects must be rather abundant, as it affects all iron-containing proteins. We observed a consistently higher induction of all genes we tested that are known to be hypoxia-dependent when the GasPak EZ pouch was used. In particular, using the GasPak EZ pouch to induce hypoxia in C42B cells, we were able to reproduce the results we obtained with DFO (Figure IV.10A-B), where clorgyline does not significantly affect MAOA levels, however it does decrease the transcription levels of HIF-dependent gene LOX by 60%, and when applied in combination with LS72, down-regulates both MAOA and LOX in a synergistic manner, resulting in a profound 70% decrease in MAOA mRNA levels (compared to 50% with LS72 alone, Figure IV.10A), and 70% decrease in transcription levels of LOX (compared to 50% with LS72 alone, Figure V.10B). In this case we were also able to demonstrate the synergistic down-regulation of another notorious 149 HIF-dependent gene GLUT1 by 50%. It is noteworthy, that the combination treatment again yields significantly lower levels of MAOA, LOX, and GLUT1 than LS72 or clorgyline treatment alone. Figure IV.10. In C42B cells, MAOA inhibitor clorgyline and LS72 down-regulate transcription levels of (a) MAOA, HIF-dependent genes (b) LOX and (c) GLUT1, as determined by qRT-PCR. The cells were treated with clorgyline at 30 μM and LS72 at 200 nM for 24 h in the medium with 1% FBS. Hypoxia was induced with hypoxic pouch. Error bars are ± SEM of experiments performed in quadruplicate. *** P < 0.001, ** P < 0.01, * P < 0.05, t-test, unless specified, compared to Vehicle Hypoxia sample. 150 IV.3. Conclusion. To summarize, hormone-independent metastatic prostate cancer cell lines C42B and PC3 exhibit elevated levels of MAOA under hypoxia. In C42B cells, MAOA and HIF1α are abundant under normoxia as well. We show that LS72, a designed transcriptional antagonist of hypoxia-inducible genes, decreases hypoxia-induced transcriptional levels of HIF-dependent genes in PC3 and C42B cells. We found that LS72 also down-regulates MAOA transcription in a dose-dependent manner under hypoxia. Clorgyline, an inhibitor of MAOA enzymatic activity and a known prostate cancer antagonist, and LS72 exhibit synergistic suppressive effect on the transcription of both MAOA and two HIF-dependent genes – LOX and GLUT1. As such, we hypothesize that MAOA is a hypoxia-inducible gene in certain prostate cancer cell lines (Figure IV.11), regulated by HIF, and the combination of the transcriptional antagonist of hypoxia-inducible genes and the inhibitor of MAOA enzymatic activity is efficacious for the treatment of prostate cancer. Figure IV.11. Schematic representation of existing and proposed interactions between prostate cancer, MAOA, and HIF1α. 151 IV.4. Experimental Section for Chapter IV. Cell lines. Human prostate adenocarcinoma cell line (PC3) and bone metastatic androgen-insensitive subline of human prostate adenocarcinoma LNCaP cell line (C42B) were kindly provided by Dr. Leland Chung at Cedars-Sinai Medical Center. Human prostate cancer PC3 cell line overexpressing MAOA (PC3-MAOA) was generated and kindly provided by Jason Wu. Cell culture. PC3 cells were grown in RPMI 1640 (Invitrogen) supplemented with 10% or 1% of fetal bovine serum (FBS, Irvine Scientific), 50 units/mL penicillin and 50 µg/mL streptomycin (Pen-Strep, Invitrogen). PC3-MAOA cells were grown in T- medium (Invitrogen) supplemented with 10% or 1% of FBS, and 0.4 g/L geneticin (RPI). C42B cells were grown in T-medium supplemented with 10% or 1% of fetal bovine serum (FBS, Irvine Scientific), 50 units/mL penicillin and 50 µg/mL streptomycin. DU145 cells were grown in EMEM (ATCC) supplemented with 10% or 1% of fetal bovine serum (FBS, Irvine Scientific), 50 units/mL penicillin and 50 µg/mL streptomycin. All cells were incubated at 37°C in a humidified atmosphere with 5% CO2. Hypoxia was induced by deferoxamine (DFO, 300 µM) or GasPak EZ Anaerobe Pouch System (BD Biosciences). Cell growth and morphology were monitored by bright field microscopy. Cells were detached using trypsin in PBS (0.05%, Invitrogen). Cell viability MTT assay. PC3, PC3-MAOA, DU145, and C42B cells were seeded in a 96-well plate at a density of 7,000 cells per well (PC3 and PC3-MAOA), 6,000 cells (DU145), and 14,000 cells (C42B) in 200 µl per well of RPMI (PC3), T-medium (PC3- MAOA and C42B), or EMEM (DU145) supplemented with 10% FBS, and allowed to 152 form a monolayer for 24 h (DU145 and C42B) or 48 h (PC3 and PC3MAOA). After that, the medium was replaced with 150 µL of fresh medium, supplemented with 10% FBS, 0.1 % dimethyl sulfoxide (DMSO), and LS72 at a concentration ranging from 0.001 µM to 10 µM. After 48 h of incubation with compounds, a solution of 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma) was added to each well (17 µL of 5 mg/mL, in PBS) and incubated at 37°C and 5 % CO2 for an additional 3 h. After that, the medium was carefully removed and purple formazan crystals were dissolved in DMSO (100 µl per well). The absorbance was measured at 570 nm with a correction at 690 nm in order to quantify the amount of formazan. Experiments were performed in a quadruplicate or quintuplicate (C42B). Cell viability MTS assay. PC3, PC3-MAOA, DU145, and C42B cells were seeded in a 96-well plate at a density of 7,000 cells per well (PC3 and PC3-MAOA), 9,000 cells (DU145), and 15,000 cells (C42B) in 200 µl per well of RPMI (PC3), T-medium (PC3- MAOA and C42B), or EMEM (DU145) supplemented with 10% FBS, and allowed to form a monolayer for 24 h (DU145 and C42B) or 48 h (PC3 and PC3MAOA). After that, the medium was replaced with 150 µL of fresh medium, supplemented with 1% FBS, 0.1 % dimethyl sulfoxide (DMSO), and LS72 at a concentration ranging from 0.001 µM to 10 µM. After 48 h of incubation with compounds, a solution of 3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS, CellTiter 96® Aqueous Non-Radioactive Cell Proliferation kit, Promega) was added to each well (30 µL) and the plates were incubated at 37°C and 5 % CO2 for an additional 3 h. After that, the absorbance was measured at 490 nm with a correction at 690 nm. All experiments were performed in quadruplicate. 153 Bicinchoninic acid (BCA) assay. Cell lysate (10 µL) was added to the BCA reagent (200 µL), prepared as per the manufacturer’s protocol (Thermo Scientific). BSA standard solutions were prepared at a concentration range of 25 µg/mL to 2000 µg/mL. Absorbance was measured at 562 nm using a Synergy 2 microplate reader (BioTek). Sample concentrations were determined from a calibration curve. All experiments were performed in triplicate. Isolation of mRNA. PC3, PC3-MAOA, and C42B cells were seeded in 6-well dishes (Greiner) at a density of 4×10 5 cells per well in 2 mL of medium with 10% FBS and allowed to form a monolayer (~70-90% confluent) for 24 h (PC3 and PC3MAOA) or 48 h (C42B). After attachment, cells were treated with 1.5 mL of fresh medium containing 1% (PC3 and PC3MAOA) or 10% FBS (C42B), 0.1% DMSO, and LS72 at concentrations 30 nM, 60 nM, 100 nM and 200 nM; vehicle samples were treated with cell culture medium containing 0.1% DMSO. After 6 h of incubation, hypoxia was induced with DFO (PC3 and PC3MAOA) or by placing the plates into a GasPak EZ Anaerobe Pouch, and cells were incubated for another 18 h. Cells were washed once (C42B, detach easily) or twice with ice-cold PBS, and total mRNA was isolated with an RNeasy kit (Qiagen) according to the manufacturer’s instructions. The mRNA was further treated with DNase I (Invitrogen, Turbo DNA-free kit) to remove any remaining genomic DNA. Then the mRNA was quantified by UV absorbance at 260 nm. Reverse transcription was performed with Superscript III Reverse Transcriptase (Invitrogen) as recommended by the manufacturer. All experiments were performed in quadruplicate. Analysis of gene expression. Real-time qRT-PCR was used to determine the effect of LS72 and/or clorgyline on VEGF, LOX, SLC2A1 (GLUT1), and MAOA genes under 154 normoxia and hypoxia. 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. For GLUT1, the following primers were used: forward 5´- AGT ATG TGG AGC AAC TGT GTG G-3´ and 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 product of 73 bp. For MAOA, we utilized the following primer pair: forward 5´-ATC ATG GGC TTC ATT CTT GC 3´ and reverse 5´-GAT CCC AGC ACT TTG GCA TA-3´ – to amplify a product of 104 bp. The mRNA levels were normalized to the expression levels of a housekeeping gene, β-glucuronidase. For β-glucuronidase the following primers was designed and used: forward 5´-CTC ATT TGG AAT TTT GCC GAT T-3´ and reverse 5´- CCG AGT GAA GAT CCC CTT TTT A-3´. Reactions were performed with Fast SYBR Green Master Mix (Applied Biosystems). Temperature cycling and detection of the SYBR green emission were performed with an ABI 7900HT Fast Real-Time PCR System. Analysis of the data was performed with Applied Biosystems Sequence Detection System, version 2.3. All experiments were performed in quadruplicate. Western blot analysis of HIF1α levels. PC3, PC3-MAOA, DU145, and C42B cells were seeded in a 60 mm dish (VWR) at a density of 8×10 5 cells (PC3, PC3-MAOA, and DU145), and 1×10 6 cells per dish (C42B) in 3 mL of corresponding medium (vide supra), and allowed to reach 80% confluence. Cells were treated with vehicle or LS72 at 200 µM in the cell culture medium containing 0.1% DMSO, 10% FBS (C42B) or 1% FBS (PC3, PC3-MAOA, and DU145). Cells were incubated for 6 h and hypoxia was 155 induced with DFO (300 µM, PC3, PC3-MAOA, and DU145) or GasPak EZ Anaerobe Pouch System (C42B). After incubation for an additional 18 h, the cells were washed twice with ice-cold PBS and then lysed with the cell culture lysis reagent (Promega). To ensure equal loading, protein concentration was determined by the BCA assay. A 1.5 mm 10% acrylamide denaturing gel was cast and an aliquot of each sample containing 30 µg of protein was loaded into the gel. The SDS-PAGE was carried out and then the gel was electroblotted onto the PVDF membrane (BioRad). After the transfer, the membrane was rinsed with tris-buffered saline with tween-20 (TBST) buffer and incubated with 5 % milk in TBST for 1 h. The membrane was then probed for HIF1α with a monoclonal mouse anti-human HIF1α antibody (BD Biosciences), MAOA with a polyclonal rabbit anti-human MAOA antibody (Santa Cruz Biotechnology), or for β-actin (loading control) with a polyclonal rabbit anti-human β-actin antibody (Cell Signaling) overnight at 4°C and gentle rocking. Membrane was further washed three times with TBST for 10 min and incubated with horseradish peroxidase (HRP) conjugated secondary anti-mouse or anti-rabbit antibody (Santa Cruz Biotechnology), respectively. Signals were detected by treating the membrane with Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare) for 3 min followed up by the exposure of the CL-Xposure film (Thermo Scientific) to the membrane. 156 Chapter V. Synthesis of Near-Infrared Dye MHI148. 157 V.1. Introduction. Near infrared (NIR) dyes are important as photoconductive materials in laser printing, photosensitive elements in infrared photography, pleochroic dyes in thermal writing liquid crystal displays, IR-absorbing laser filters, cosmetic ingredients, and dyes for polymers.(364) In 1942 the hematoporphyrin, a dye with one of the absorption maxima at 605 nm,(365) was found to be preferentially accumulated in solid tumors, and especially in their inner necrotic compartments.(366) A particular subclass of indocyanine family of NIR fluorescent dyes (Scheme V.1), heptamethine cyanine dyes, have promising application as fluorescent labelling agents for proteins,(367-369) fluorescent tags in DNA sequencing,(370) flow cytometry,(371) and enzyme kinetics.(372) Low tissue autofluorescence and deep tissue penetration allow NIR fluorescence dyes to be excellent probes for tumor imaging. Various mouse tumor xenografts and chemically-induced lung tumors in C57BL/6 mice were found to retain one of such NIR dyes, IR780, for at least 20 days with excellent signal to noise ratio.(373) Scheme V.1. General structure of indocyanine polymethine dyes. Novel heptamethine dye IR808 (later known as MHI148) was first reported in 1995 by Lee et al.,(374) and there have been just over a dozen different reports on the synthesis and application of this dye since then until 2014. From a synthetic standpoint, carboxyl groups 158 of MHI148 are a very convenient feature, facilitating the labeling, as it may be conducted in the form of an amide coupling. The ability of MHI148 to accumulate in human cancer cells in vitro and in vivo, including spontaneous tumors in mice was first investigated in 2010 by Yang et al.(375) The subcellular localization, mechanisms of uptake, and duration of retention of the dyes in human cancer cells were evaluated using organelle-specific tracking dyes and bromosulfophthalein to differentiate cancer cells from normal cells, and MHI148 was found to accumulate in the mitochondria and lysosomes of predominantly cancer cells. The mechanism was speculated to involve anion transporting peptides. The void in the research dedicated to MHI148 has recently started to be filled, as the interest to this dye spiked just recently, when it was successfully used as an optical imaging agent for the detection of human kidney cancer;(376) a label for tumor targeting and delivery of an MRI contrasting mesoporous silica-coated Gd silicate nanoparticles;(377) and a cancer-targeting moiety in a dual-modality 99 Tc-containing imaging probe, which has been shown to accumulate in breast adenocarcinoma MCF7 cells in vitro and in vivo.(378) 159 Heptamethine cyanine NIR dyes are prepared via condensation between unsaturated bis- aldehyde (often masked as a Schiff base), in general, in any suitable protic solvent with a base as a catalyst. In particular, most frequently reported conditions include the mixture of glacial acetic acid and acetic anhydride, optionally supplemented with sodium acetate,(379) or ethanol supplemented with sodium acetate(380), triethylamine, or pyridine.(381) Recently a synthesis of several symmetric and asymmetric NIR dyes using reflux in butanol/benzene (7/3 v/v) was reported,(382) and shown to be advantageous over more traditional methods when equipped with a Dean-Stark trap for the continuous removal of water from the reaction as azeotrope.(383) Several syntheses of MHI148 were reported in the literature, including quite a few very recently either during or right after we accomplished our synthetic effort in early 2013.(377, 380, 382, 384) The yields were either not reported, or ranged between 50 and 84%, required additional column purification, and albeit the syntheses achieved their goals of providing enough material for further studies, the product was never meticulously and systematically analyzed by different methods. Besides the lack of a commercial source for MHI148, this lack of systematic approach in the literature might explain why different syntheses are still published while the dye has been known for almost 20 years. For instance, Zhang et al. in early 2014 claim to have developed the synthesis of MHI148, without either citing any prior research or providing the synthetic procedure.(385) Thus we deemed that the development of an alternative synthetic approach was reasonably warranted, and devised the synthesis of MHI148 as a combination of known steps and known conditions applied to our reactants (Scheme V.2). We based our decisions on the perceived simplicity, safety, and cost of the synthesis and expected yield. 160 V.3. Discussion of Synthetic Steps. The synthesis of the first reactant, indolenium bromide V.4, can be conducted in nitromethane,(386) 1,2-dichlorobenzene,(387, 388) or in solvent-free conditions.(374) Similar quaternary salts have been obtained by heating the reactants in 1,2- dichlorobenzene,(367, 383) acetonitrile,(383) or in solvent-free conditions.(367, 383) We elected to conduct the alkylation of indolenine V.2 with 6-bromohexanoic acid V.3 to afford the indolenium bromide V.4 under reflux in acetonitrile, which resulted in a pure compound in a high yield. Next, Vilsmeier-Haack formylation of cyclohexanone is conducted with formaldehyde and POCl3 to obtain V.5 according to a known procedure.(389) However, the yield of V.5 was consistently half of what was stated in that procedure, and we also found out that the purification by recrystallization from acetone was not necessary. At the same time, the yield was two times higher than in one of the earlier reports,(374) where the bis-aldehyde was masked as enamine with aniline. Finally, the coupling of V.4 and V.5 was performed in ethanol using pyridine a base, without subsequent column purification. All reactions were repeated at least two times on a gram scale, and have been proven to be reproducible, in terms of both yields and purity. 161 Scheme V.2. Synthesis of MHI148 V.1. 162 V.3. Analysis. We have used ¹H and ¹³C NMR to establish identity of the intermediates V.4 and V.5. Proton NMR of the product of the alkylation reaction to obtain V.4 has unassigned singlet at 1.33 ppm, but otherwise the structure of the compound can be considered confirmed to be V.4, according to both ¹H and ¹³C NMR. Vilsmeier-Haack reaction to obtain V.5 succeeded according to the ¹H NMR, with the peak at 2.08 ppm corresponding to residual acetone. The ¹³C NMR spectrum, however, was more confusing. Unassigned peak at 20.63 ppm apparently corresponds to acetone, but otherwise we observed fewer peaks than expected theoretically. Peaks at 146.77, 31.36, and 24.37 ppm apparently correspond to the carbons 1, 4, and 3/5 (both) on the cyclohexene ring, while the other carbons did not show up on the spectrum. It is not unusual, however, for the quaternary carbons not to show up on the ¹³C spectra, and one can argue that is what happened in this case. Another noteworthy observation is that the compound V.5 is symmetrical, as both methylene groups at positions 3 and 5 of the ring show up as one peak in ¹H and ¹³C NMR. Such symmetry is probably due to rapid tautomerization. The final product (V.1) was analyzed with ¹H and ¹³C NMR and LC-MS, while the purity was additionally confirmed with HPLC. The ¹H NMR contains all the expected peaks (Figure V.1). The three unassigned triplets at 3.55 (J 6.5), 3.40 ppm (J 7.0), and 1.35 (J 7.0), and a singlet at 1.32 in ¹H NMR have very low intensity (< 0.3H), and we were not able to identify the source of contamination (see Experimental section). The ¹³C NMR has all the peaks of interest (Figure V.2), however there are also some unassigned peaks in the high field at 44.64, 34.15, 26.54, and 24.24 ppm, which all are located in the immediate 163 vicinity of some of the assigned peaks (see Experimental section), characterizing the aliphatic region of the molecule. Both from a theoretical standpoint, and judging by the ¹H NMR and the low field aromatic portion of the ¹³C NMR, the product ought to be symmetric due to resonance. As such, we do not have a good explanation for the phenomenon observed in the high field portion of ¹³C spectrum. Figure V.1. ¹H peak assignment for V.1. Figure V.2. ¹³C peak assignment for V.1. 164 We found that the product is >95% pure according to HPLC (see Experimental section), and has substantial absorption in the near-IR portion of the visible spectrum (Figure V.3), from 550 nm all the way to the detection limit (650 nm), as anticipated for a NIR dye. Figure V.3. 3D graph of the HPLC trace of V.1. The LC-MS analysis (ESI) returned the m/z value of 683.3763, which coincides with the theoretical calculated value of 683.3610. The total ion count (TIC) chromatogram shows three peaks (see Experimental section), which correspond to various products of V.1 fragmentation and methylation in the mass spectrometer (Figure V.4). Therefore, LC-MS analysis confirms conclusions made based on the NMR and HPLC data about the identity and > 95% purity MHI148. 165 Figure V.4. Products of V.1 fragmentation/methylation in mass spectrometer. The dye we obtained was shown to cross the blood brain barrier and specifically target brain tumors implanted in mice (Figure V.5, courtesy of Dr. Jason Wu, Cedars-Sinai Medical Center). 166 Figure V.5. MHI148 obtained via the modified procedure crosses blood-brain barrier (data courtesy of Dr. Jason Wu, Cedars-Sinai Medical Center). 167 V.4. Conclusion. We developed a procedure to synthesize the NIR dye MHI148 in high yield (88%), on a gram scale, confirmed its purity and conducted a thorough analysis, including detailed ¹H and ¹³C NMR peak assignment and assignment of all ion peaks in the LC-MS. The synthesis did not require column purification at any stage. The dye successfully crossed the blood-brain barrier, and targeted the brain tumor. 168 V.5. Experimental Section for Chapter V. 1-(5-carboxypentyl)-2,3,3-trimethylindolenium bromide (V.4). A single-neck 250 mL round-bottom flask is charged with indolenine V.2 (10 g, 62.8 mmol), acid V.3 (35 g, 179.4 mmol), and acetonitrile (80 mL). The contents of the flask are vigorously stirred to afford a dark-red solution. The flask is then outfitted with a Graham condenser, and the reaction mixture is vigorously stirred and brought to reflux. The mixture is heated at the reflux point for 22 h. ACN is removed under reduced pressure, and the viscous red liquid is then dissolved in DCM (ca. 30 mL) and transferred into a 500 mL beaker. Ether (80 mL) is added dropwise under continuous stirring to afford a pink precipitate covered in red oil. More ether (100 mL) is added and the precipitate and the oil are triturated by thorough dispersion in the ether-DCM mixture with a spatula. The pink precipitate with red oily chunks is filtered and washed with ether (25 mL, three times) on the filter, then transferred into a beaker, triturated with ether (50 mL), filtered, washed with ether (50 mL) on the filter and then collected and dried under high vacuum overnight. Yield 15.3614 g (69%) in the form of pink powder (Note 7). 2-chloro-3-(hydroxymethylene)-cyclohex-1-ene-1-carboxaldehyde (V.5). A single-neck 250 mL round bottom flask is charged with DMF (40 mL) and DCM (40 mL). POCl3 (38 mL) is dissolved in DCM (35 mL) in a 200 mL beaker. Both solutions are cooled to 0°C in an ice bath. POCl3 in DCM is then added into the reaction mixture, dropwise using an addition funnel, under stirring to yield a light pink solution. Cyclohexanone (10 g, 0.102 mol) is added into the reaction mixture, and the ice bath is removed. The flask is then outfitted with a Graham condenser, and the bright orange reaction mixture is brought to a reflux point under vigorous stirring, and heated at a reflux point for 4 h. Then the reaction 169 mixture is cooled down for 10 min at rt, then carefully and slowly poured on ice (200 g). Organic pale-orange and aqueous bright-orange phases are separated and the aqueous phase is placed into the freezer at -20°C overnight. Yellow crystals of V.5 are then filtered, and dried under high vacuum overnight (Note 7). Yield 6.61 g (38%) in the form of yellow needle-like crystals (Note 9). 1-(2-(1-(5-carboxypentyl)-3,3-dimethylindolenium-2-yl)-vinyl)-2-cloro-3-((2-(1-(5- carboxypentyl)-3,3-dimethylindol-2-ylene)methyl)methylene-cyclohex-1-ene bromide (MHI148, V.1). A two-neck 250 mL round-bottom flask is charged with compounds V.4 (2.01 g, 5.67 mmol), V.5 (0.466 g, 2.7 mmol) and EtOH (40 mL). The flask is then outfitted with a Graham condenser, and the reaction mixture is stirred vigorously. Then pyridine (0.542 mL) is added. Dark-red reaction mixture is brought to reflux under continuous stirring, and heated at the reflux point for 16 h. Then the dark-green reaction mixture is cooled to rt, and the contents of the flask is transferred into a separation funnel with DCM (80 mL). The organic phase is washed with HCl (0.1 N, twice), then brine (once). Aqueous wash is washed with additional DCM (30 mL, twice), then the DCM fractions are merged and dried with MgSO4. DCM is removed under reduced pressure to afford the glittery purple residue. The residue is redissolved in DCM (10 mL), then triturated with hexanes (80 mL) at -20°C overnight. Yellow supernatant is carefully decanted off the purple-red oil and discarded. The oil is redissolved in DCM (ca. 20 mL) and transferred into a glass vial. DCM is removed under reduced pressure. Glittery metallic magenta crystals of V.1 are dried under high vacuum overnight. Yield of V.1 1.8203 g (88.2%) in the form of magenta metallic crystals (Note 10). 170 V.6. Notes. 1. All reagents and solvents were obtained from commercial sources and were used as received, unless otherwise stated. 2. All reactions were conducted under argon atmosphere using anhydrous solvents and in flame-dried glassware, unless indicated otherwise. 3. Solid materials were introduced into reaction vessels directly through the vessel openings. 4. Hygroscopic liquids were transferred via a syringe and were introduced into reaction vessels through the rubber septa. 5. Nuclear magnetic resonance (NMR) spectra were collected on Varian 400 MHz instrument in the indicated solvents. The peak positions are reported with chemical shifts (δ) in parts per million (ppm) downfield from the signal for tetramethylsilane (0 ppm) and referenced to the signal resulting from the incomplete deuteration of a solvent used in the experiment (CDCl3: 7.26 ppm, DMSO-d6: 2.50 ppm). Carbon-13 chemical shifts are reported as δ values in ppm and referenced to the cabon-13 signal of a solvent used in the experiment (CDCl3: 77.2 ppm, DMSO-d6: 39.51 ppm). The coupling constants (J) are reported in Hertz (Hz). The following abbreviations are used: singlet (s), doublet (d), triplet (t), doublet of doublets (dd), doublet of doublets of doublets (ddd), pentet (p), multiplet (m). 6. High resolution electrospray ionization mass spectra were obtained on an Agilent 6210 time-of-flight LC-MS. 7. 1-(5-carboxypentyl)-2,3,3-trimethylindolenium bromide (V .4) has the following physical and spectroscopic properties: 1 H NMR (DMSO-d6, 400 MHz) δ: 7.95 (m, 1H), 171 7.84 (m, 1H), 7.61 (m, 2H), 4.43 (t, 2H, J 7.7), 2.82 (s, 3H), 2.48 (p, 2H, J 1.8), 2.21 (t, 2H, J 7.2), 1.82 (m, 2H), 1.52 (s, 6H), 1.41 (m, 2H), 1.33 (s, 2H). ¹³C NMR (DMSO- d6, 400 MHz) δ: 174.35, 141.88, 141.07, 129.43, 128.97, 123.53, 115.50, 54.17, 47.41, 33.37, 26.98, 25.44, 24.04, 22.02, 13.95. 8. The crystals may turn brown-black overtime under high vacuum without loss of purity. 9. 2-chloro-3-(hydroxymethylene)-cyclohex-1-ene-1-carboxaldehyde (V .5) has the following physical and spectroscopic properties: 1 H NMR (DMSO-d6, 400 MHz) δ: 8.80 (s, 2H), 2.35 (t, 4H, J 6.1), 1.57 (p, 2H, J 6.1). ¹³C NMR (DMSO-d6, 400 MHz) δ: 146.77, 31.36, 24.37. 10. 1-(2-(1-(5-carboxypentyl)-3,3-dimethylindolenium-2-yl)-vinyl)-2-cloro-3-((2-(1-(5- carboxypentyl)-3,3-dimethylindol-2-ylene)methyl)methylene-cyclohex-1-ene bromide (MHI148, V .1) has the following physical and spectroscopic properties: 1 H NMR (CDCl3, 400 MHz) δ: 10.2 (s, 1H), 8.34 (d, 1H, J 14), 7.40 (ddd, 1H, J 7.7, J 1.0), 7.38 (dd, 1H, J 7.7, J 0.9), 7.24 (ddd, 1H, J 7.7, J 7.6, J 0.7), 7.18 (dd, 1H, J 7.6, J n/a (broad)), 6.17 (d, ¹H, J 14), 4.12 (t, 2H, J 7.3), 2.71 (t, 2H, J 5.9), 2.51 (t, 2H, J 7.3), 2.00 (p, 2H, J 6.0), 1.87 (p, 2H, J 7.3), 1.77 (p, 2H, J 7.5), 1.71 (s, 6H), 1.56 (p, 2H, J 7.4). ¹³C NMR (CDCl3, 400 MHz) δ: 176.96, 172.44, 151.01, 144.67, 142.23, 141.14, 129.09, 127.91, 125.52, 122.42, 111.08, 101.33, 49.49, 45.01, 44.59, 34.46, 34.09, 32.40, 28.33, 27.12, 26.73, 26.38, 26.41, 24.62, 24.18. 172 173 174 175 176 177 178 179 180 181 182 183 Chapter VI. A Scalable Procedure for the Synthesis of 3-(4-methoxyphenyl)-6,8- dimethyl-2,4-dithia-6,8- diazabicyclo[3.2.2]nonane-7,9-dione. 184 VI.1. Introduction. A scalable procedure for the multi-gram scale synthesis of 3-(4-methoxyphenyl)-6,8- dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane-7,9-dione, VI.1 was developed. The title compound VI.1 was first reported by Fukuyama and Kishi as an intermediate in the total syntheses of gliotoxin,(390, 391) dehydrogliotoxin, and hyalodendrin.(391) It is a key intermediate in the synthesis of xylylene-linked, dimeric epidithiodiketopiperazines (ETPs),(242) which are a novel class of designed protein ligands that target the key protein- protein interactions responsible for mediating the growth and metastatic spread of solid tumors. Both naturally occurring and synthetic ETPs, containing two ETP cores, disrupt hypoxia inducible transcription factor 1 (HIF1) activity by targeting the interactions between HIF1α subunit and p300 coactivator, thereby blocking activation of the hypoxia- inducible gene expression(227, 242) that leads to angiogenesis and increased metastasis, resulting in a poor prognosis for cancer patients. In addition, 3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane- 7,9-dione is a key intermediate in the total synthesis of aspirochlorine,(392) a secondary metabolite from Aspergillus terreus. Aspirochlorine exhibits antifungal and antibacterial 185 activity and is effective in inhibiting the pathogenic fungus C. albicans by blocking the synthesis of fungal proteins.(393) VI.2. Results and Discussion. In order to evaluate the efficacy of the synthetic dimeric ETPs as antitumor agents and aspirochlorine as antifungal/antimicrobial agent in the preclinical setting, such compounds must be prepared on a 0.25 – 0.5 g scale, which requires synthesis of the title compound VI.1 on a scale of 5 – 10 g. The first step of the synthesis (Scheme VI.1), conversion of glycine anhydride VI.2 to sarcosine anhydride VI.3, requires little effort. However, we and others found that the next step leading to VI.4, originally reported by Trown,(394) is unreliable and difficult to scale up. With a rare and unpredictable exception, the reaction most frequently yields black tar that is impossible to dissolve in any solvent. To our knowledge, there is no better current alternative for the preparation of VI.4. At one point, we had to employ a custom synthesis company to make VI.1, and that cost us ca. $1,000 per gram. Even though they were supplied with the detailed procedure, it took them 6 months to make 20 grams of VI.1, and they claimed that the synthesis of VI.4 was the culprit. This made us look closer at the issue. We have conducted a careful investigation of the reaction conditions, analyzed the nature of the reactants, reagents, and product, and possible side reactions. We have proposed that the source of the problems is the slow rate of the reaction. Excessive temperature and quantity of bromine recommended by the original procedure certainly help to overcome the activation barrier, however we speculated, they also may lead to the decomposition of the starting material and possibly 186 radical ring-opening polymerization. We have attempted to implement a different source of bromine, namely N-bromosuccinimide, but the reaction did not succeed. We have thus sought to facilitate the radical bromination by conditioning a more efficient radical formation. This arguably small change resulted in a spectacular outcome: it allowed us to lower the temperature and bring the excess of bromine to almost zero percent, while increasing the yield and making the synthesis reproducible on a 10-gram scale. Scheme VI.1. Synthesis of the dithioacetal VI.1. As such, the proposed procedure utilizes irradiation by a 500W regular halogen sunlamp (Home Depot) and careful control of the temperature in order to obtain VI.4 from sarcosine anhydride VI.3 in high yield (89%, 7.5 g of VI.4), thus circumventing the key culprit of the synthesis. 187 Albeit the precursor to VI.4, sarcosine anhydride VI.3, is commercially available, it is ~10 times more expensive than glycine anhydride VI.2. * Since sarcosine anhydride VI.3 is required at the second step of the synthesis, its availability is paramount to the cost- effectiveness of the synthesis, thus warranting the synthetic effort, especially considering that the transformation of VI.2 into VI.3 is facile and straightforward (72%, 36 g of VI.3). Formation of bis-thioacetate VI.5 from dibromide VI.4 was effortlessly optimized for the reported scale (9.0 g of VI.5, 93% yield). Finally, to obtain VI.1 from VI.5 as developed previously, bis-thioacetate VI.5 is hydrolyzed to give bis-thiol VI.6, which is then constrained into the dithioacetal VI.1 (5.73 g of VI.1, 55% yield for two steps). However, we have looked carefully at these two steps, and deduced that they can be carried out under the same conditions. Indeed, since both hydrolysis and formation of dithioacetal can be conducted in a protic solvent like methanol, hydrochloric acid can be used to catalyze both reactions. Additionally, the pure final product precipitates out of methanol, facilitating its isolation (2.5 g of VI.1, 55% yield from VI.5, one pot). Importantly, all products are isolated and purified only by crystallization and filtration, without the need for column chromatography. The identity and purity of each compound have been confirmed by ¹H and ¹³C NMR and by high-resolution MS. * Glycine anhydride – $49.60 for 25 g (Sigma), sarcosine anhydride – $19.14 for 1 g (Alpha Aesar). 188 VI.3. Experimental Section for Chapter VI. 1,4-dimethyl-2,5-diketopiperazine (VI.3). DMF is dried by distillation over 4A molecular sieves at a reduced pressure. A 2-neck, 250 mL flask filled with Ar is charged with sodium hydride (30.8 g of 60% dispersion in oil, 0.77 mol of NaH), which is then dispersed in DMF (400 mL) at 4°C. Suspension of glycine anhydride 2 (40 g, 0.3 mol) in DMF (100 mL) is added to the slurry of NaH in DMF at 4°C, via addition funnel, dropwise, under vigorous stirring, and careful control of the temperature. After that, solution of dimethyl sulfate (88.3 g, 0.7 mol, 64 mL) was added to the reaction mixture at 4°C, via addition funnel, dropwise, under vigorous stirring, and careful control of the temperature. After that, the reaction mixture was brought up to rt, and left overnight under vigorous stirring. Methanol (150 mL) is added to the reaction mixture. The mixture is transferred into a 1 L flask and the solvent is carefully removed under reduced pressure in a water bath at 60°C until the residue stops changing viscosity visually. The precipitate was resuspended in DCM (600 mL), the suspension was brought to boil and filtered. The filtrate was collected, and the procedure was repeated two more times with the precipitate. The solvent in each fraction was evaporated under reduced pressure until the precipitation was observed. Diethyl ether (600 mL per fraction) was added to each fraction, and abundant precipitation was observed. White precipitate was isolated by filtration and washed with diethyl ether (200 mL per fraction) on the filter. Yield 36 g (74%) of VI.3 in the form of white needle- like crystals (Note 10). 1,4-dimethyl-2,5-diketopiperazine-3,6-dibromide (VI.4). A 2-neck, 250 mL flask is charged with sarcosine anhydride VI.3 (4.995 g, 35.2 mmol, Alfa Aesar), which is dispersed in o-dichlorobenzene (70 mL, J.T. Baker). A Graham condenser with medium 189 rate water flow at ambient temperature is attached to the flask, while the other neck is plugged with a rubber septum, then the flask is immersed into an oil bath at 150°C and the reaction mixture is stirred vigorously for 10 min. Upon dissolution of most of 1, bromine (12.48 g, 4 mL, 78.0 mmol, Sigma-Aldrich) is added to the reaction mixture under continuous vigorous stirring and UV irradiation. Addition of bromine is performed dropwise, in small portions (10-15 drops) and the atmosphere in the flask is allowed to clear before more bromine is added (20-60 seconds between the portions). The temperature of the oil bath is manually maintained at 145 ± 10 °C throughout the addition using the hotplate stirrer (Corning). Abundant precipitation is observed upon addition of bromine. The color of the precipitate changes from white to orange throughout the addition process. Upon completion of the bromine addition (takes ca. 40 min), the precipitate begins dissolving rapidly and the reaction is stirred at 145 ± 10 °C for additional 30 min. The hot dark-orange homogenous mixture is poured into a 1 L beaker and allowed to cool down to ambient temperature until precipitation starts, then 300 mL of hexanes is slowly added into the beaker. The beaker with beige suspension is put in the freezer at -20°C for 1 h. After that, the walls and the bottom of the beaker are thoroughly scraped clean with a spatula and the precipitate is homogenized. The beaker is returned into the freezer for additional 15 h (or more). The pale orange precipitate that forms is isolated by filtration, then dried on filter at ambient temperature for 3-5 min and then quickly transferred into appropriate glass container and dried under high vacuum to constant mass. Yield 9.20 g (87%) of VI.4 in the form of fine glittery pale orange crystals (Note 11). 1,4-dimethyl-2,5-diketopiperazine-3,6-bis-thioacetate (VI.5). 1,4-Dimethyl-2,5- diketopiperazine-3,6-dibromide VI.4 (10.0 g, 333 mmol) is added to 300 mL CH2Cl2 in a 190 1 L single neck round-bottom flask. Potassium thioacetate (10.0 g, 87.6 mmol, Sigma- Aldrich, crushed in the mortar into fine powder) is added in ~0.5 g portions over the course of 3-5 min. The reaction mixture is stirred vigorously for 3 h at ambient temperature. Next, the reaction mixture is washed 3 times with water, and then the organic phase is dried with anhydrous MgSO4 (EMD) and filtered. The solvent is removed under reduced pressure to afford beige crystals. The crystals are dried in vacuo overnight. Yield 9.0 g (93%) of VI.5 (Note 12). 1,4-dimethyl-2,5-diketopiperazine-3,6-bis-thiol (VI.6). A 500 mL single-neck round bottom flask is charged with 1,4-dimethyl-2,5-diketopiperazine-3,6-bis-thioacetate VI.5 (4.10 g, 14.1 mmol). Methanol (100 mL) is added next, followed by HCl in anhydrous ether (100 mL, 1 M solution). A Graham condenser is then attached to the flask and the reaction mixture is refluxed for 3 h. The solvent is removed under reduced pressure. Yield 4.27 g of VI.6 in the form of yellowish slurry. The product is used in the next step without further purification. 3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane-7,9-dione (VI.1). Version A: A 500 mL single-neck round bottom flask is charged with a solution of VI.6 (4.98 g, 24.1 mmol) in CH2Cl2 (300 mL). Next, p-anisaldehyde (20.5 g, 18.3 mL, 151 mmol, Sigma-Aldrich) is added and the reaction mixture is stirred for 16 h at ambient temperature. Next, the reaction mixture is washed with saturated aqueous solution of NaHCO3 (30 mL), water (100 mL), and brine (100 mL). The organic phase is dried with MgSO4, filtered and the solvent is removed under reduced pressure. The crude product is 191 triturated with diethyl ether and the resulting white precipitate is isolated by filtration and dried under high vacuum overnight. Yield 5.72 g (73%) of VI.1 in the form of beige crystals (Note 13). Version B: A 500 mL two-neck round bottom flask equipped with a Graham condenser is charged with bis-thioacetate VI.5 (4.09 g, 14.1 mmol). Next, methanol (70 mL) is quickly added, followed by HCl in methanol (113 mL, 1.25 M, 141.0 mmol, Fluka). The reaction mixture is then refluxed under continuous N2 flow and stirring for 5 h. After that, the reaction is cooled to ambient temperature with continuous stirring. Next, p- anisaldehyde (5.76 g, 5.138 mL, 42.4 mmol, Sigma-Aldrich) is added and the reaction mixture is stirred for another 18 h. After that, white precipitate is isolated by filtration and dried on the filter for 30 minutes at ambient temperature and then under high vacuum overnight. Yield 2.50 g (55%) of VI.1 in the form of beige crystals (Note 13). 192 VI.4. Notes. 1. All reagents and solvents were obtained from commercial sources and were used as received, unless otherwise stated. 2. All reactions were conducted under argon atmosphere using anhydrous solvents and in flame-dried glassware, unless indicated otherwise. 3. Solid materials were introduced into reaction vessels directly through the vessel openings. 4. Hygroscopic liquids were transferred via a syringe and were introduced into reaction vessels through the rubber septa. 5. A halogen sunlamp (500 W, Home Depot) was used as a light source. 6. Reaction product solutions were concentrated using a rotary evaporator at 30-150 mm Hg. 7. Nuclear magnetic resonance (NMR) spectra were collected on Bruker 400 MHz and Varian 400 MHz instruments in the indicated solvents. The peak positions are reported with chemical shifts (δ) in parts per million (ppm) downfield from the signal for tetramethylsilane (0 ppm) and referenced to the signal resulting from the incomplete deuteration of a solvent used in the experiment (CDCl3: 7.26 ppm). Carbon-13 chemical shifts are reported as δ values in ppm and referenced to the cabon-13 signal of a solvent used in the experiment (CDCl3: 77.2 ppm). The coupling constants (J) are reported in Hertz (Hz). The following abbreviations are used: singlet (s), doublet (d), triplet (t), doublet of doublets (dd), multiplet (m). 8. High resolution electrospray ionization mass spectra were obtained on an Agilent 6210 time-of-flight LC-MS. 193 9. Bromination to obtain VI.4 allows for considerable variations in the temperature during the addition of bromine, however one should not allow temperatures of less than 120°C and more than 160°C for longer than 10 minutes. The UV lamp heats up the oil bath, thus constant temperature monitoring and adjustment is required. 10. 1,4-dimethyl-2,5-diketopiperazine (VI.3) has the following physical and spectroscopic properties: 1 H NMR (CDCl3, 400 MHz) δ: 3.95 (s, 2H), 2.94 (s, 3H). ¹³C NMR (CDCl3, 400 MHz) δ: 163.29, 51.83, 33.46. 11. 1,4-dimethyl-2,5-diketopiperazine-3,6-dibromide (VI.4) has the following physical and spectroscopic properties: 1 H NMR (CDCl3, 400 MHz) δ: 6.01 (s, 1H), 3.10 (s, 3H). ¹³C NMR (CDCl3, 400 MHz) δ: 161.77, 60.14, 33.57. 12. 1,4-dimethyl-2,5-diketopiperazine-3,6-bis-thioacetate (VI.5) has the following physical and spectroscopic properties: 1 H NMR (CDCl3, 400 MHz) δ: 5.76 (s, 3H), 5.62 (s, 1H), 2.97 (s, 3H), 2.93 (s, 9H), 2.47 (s, 9H), 2.41 (s, 3H). ¹³C NMR (CDCl3, 400 MHz) δ: 193.50, 192.06, 163.85, 162.69, 64.42, 62.58, 33.10, 32.32, 30.72. 13. 3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane-7,9- dione (VI.1) has the following physical and spectroscopic properties: 1 H NMR (CDCl3, 400 MHz) δ: 7.84-7.82 (d, 2H, J 8.6), 7.36-7.34 (d, 2H, J 8.6), 5.14 (s, 1H), 5.03 (s, 1H), 4.88 (s, 1H), 3.79 (s, 3H), 3.19 (s, 3H), 3.05, (s, 3H). ¹³C NMR (CDCl3, 400 MHz) δ: 165.24, 163.45, 160.72, 130.71, 126.70, 114.52, 66.07, 63.94, 55.51, 49.44, 32.67, 32.00. 194 195 196 197 198 199 200 201 202 203 204 205 Chapter VII. Automatic Cell Counting with ImageJ. 206 VII.1. Introduction. Cell counting is an important routine task. However, to date there is no comprehensive, easy to use, and importantly cheap solution for routine cell counting, and most researchers have to succumb to manual counting, a lengthy and tiresome procedure. Here we report a complete solution that everyone can implement in their laboratory with minimal effort to count the cells automatically. We have worked out a robust approach to equip a conventional light microscope with a web camera to obtain good images. Based on ImageJ toolbox, we have devised two algorithms to automatically count mammalian cells in the pictures of the cell suspension in the hemocytometer assembly. We have vetted the approach, and established that it takes up to 10 times less time and yields more reliable and consistent results than manual counting. Counting mammalian cells for seeding them into the plates or dishes is an important mundane task performed by cell biologists and medical professionals sometimes several times a day. Hemocytometer is most frequently used to perform this task. The cells have to be manually counted on eight 1×1 mm areas of the hemocytometer located on its two panels. The number of cells per panel most frequently averages 100-300, thus on the average, one has to count ca. 800-2,500 cells with a manual clicker, while looking into the eyepiece of the microscope all the time. It is numbing, tiresome, frustrating and, importantly, has to be carried out on a daily basis. The most obvious solution is a commercial cell counter; however, the primary issue is the cost of ~$4,000 and up, excluding the pricey consumables. Some also have a less than 207 favorable operating range of 10K-500K cells/ml, while the average concentration of a cell suspension one has to work with is 1-4M cells/ml. Some commercially available cell counters are listed in Table VII.1. Table VII.1. Commercially available cell counters. Manufacturer Product Name Link Price, USD Biorad TC20 http://www.bio-rad.com/en-us/product/tc20-automated-cell-counter 3,995 Invitrogen Countess http://www.lifetechnologies.com/us/en/home/brands/product-brand/countess- automated-cell-counter.html 3,750 Millipore Scepter http://www.millipore.com/life_sciences/flx4/scepter_automated_cell_counter N/A Logos Biosystems Luna http://logosbio.com/cell-counters/luna/features.asp N/A If the image of the cells spread in one layer on a known area is already available, there are several software solutions for cell counting. Manual standalone cell counting assistants, plugins and guides for ImageJ(395, 396) are numerous. They facilitate cell counting by replacing the manual clicker with multiple digital ones,(395, 397, 398) or placing a semi- transparent grid over the picture to help the researcher focus,(399) however every cell still has to be visually registered and manually accounted for. There are also a few powerful software suites, which allow for automatic counting of mammalian cells (Cell Profiler),(400) fluorescently labeled bacterial cells (CellC),(401) mammalian (CellCounter)(402) and bacterial (OpenCFU)(403) colonies, and other circular objects. The common drawbacks of some of the existing solutions include: 208 - lack of focus on routine mammalian cell counting for cell seeding; - complexity; - very narrow application range or extreme versatility (directly proportional to complexity); - manual data input and lack of batch processing; - manual adjustment of counting parameters for every image; - use of separate, standalone software solely for counting; - requirement for expensive hardware for cell imaging with the microscope. First, to facilitate counting, a microscope needs to have a camera, as it is much easier to count the cells while looking at a screen, rather than into the eyepiece of a microscope. However, such attachments from the manufacturer are prohibitively expensive and generally are available only for the high-end microscopes. Addition of a regular camera to a microscope has been reported,(404) albeit it has never been systematically investigated, documented or streamlined for daily lab use. Moreover, known DIY solutions for camera mounting require the presence of a dedicated camera port and/or the manufacture of a costly holder for the camera. The problem on the side of the software is the lack of proper guidance and a completely automated approach. The standalone software tools listed above are not designed to capture the number of cells in the picture of a hemocytometer grid and hence unfortunately are not capable to do so accurately. The online guides tend to sum up the otherwise well- documented features of the ImageJ software suite.(405) ImageJ is the software of choice for many applications related to image processing in natural sciences, as it is powerful, free, open-source, and cross-platform. 209 Here we discuss a method to outfit a conventional microscope with a cheap camera, and offer a complete solution for counting cells automatically. We do not aim either to offer a universal cell- and colony-counting panacea or make another dedicated application. Rather, we aim at creating a complete solution on the basis of existing toolset for a specific mundane task that everyone can effortlessly implement in their laboratory. The only way to make a new method competitive and facilitate its adoption is to make the setup beyond affordable and software familiar and understandable at a glance, without disrupting the routine a researcher is used to. 210 VII.2. Hardware. The primary concerns for the imaging hardware are the price and the viewfield diameter or the field of view (FOV), which most often tends to be smaller than that of the eye. It depends on the size of the camera sensor. The FOV depends on the field number of the eyepiece (FN), objective magnification M(o), and the tube lens magnification M(l): FOV = ∗ Thus the viewfield diameter is 5 mm at 4x magnification (FN = 20 mm, M(l) = 1.00) when looking through the eyepiece with a naked eye. The sensor most frequently utilized in most web cameras is 1/3” in so-called “video camera tube notation”, and has the diagonal of 6 mm,(406) while the retina of a human eye has a diameter of 20-22 mm. Thus the formula for the FOV of a camera sensor is slightly different: FOV = ∗ where SD is sensor diagonal and M(a) is adapter magnification. As such, without an adapter at 5x magnification, the FOV shrinks to a mere 1.5 mm (diagonal) when a 1/3” sensor is used, and that is the major culprit of most DIY approaches. Effectively, a costly reduction adapter with M(a) ~ 0.8 or smaller is required to achieve any reasonable FOV. 211 In this study, the measurements were conducted using Olympus CKX41 brightfield microscope with 4x objective and 10x/20 WHB eyepiece. Sensor on the circuit board from Microsoft LifeCam VX-5000 was used as a camera. We have elected a web camera as opposed to a point-and-shoot camera or a DSLR for the following reasons: - image quality is more than adequate for all the routine lab tasks: cell monitoring and counting, scratch assay, colony formation assay, histopathology; - dramatically lower cost; - continuous feed on a computer monitor using a dedicated camera software. We have found out that the eyepiece can play a role of a reduction adapter, when a sensor on a circuit board extracted out of a web camera is positioned at a proper distance from the eyepiece. We have been able to achieve a FOV of 2 mm (diagonal, 33% larger than original, corresponding to M(a) = 0.75). This allowed for the manufacture of a simple holder for the circuit board out of a PVC pipe, easily removable from the eyepiece. There are two primary details one needs to pay attention to when manufacturing such an adapter: 1) aligning the center of the sensor with the center of the eyepiece and 2) ensuring that the sensor is 5-7 mm away from the eyepiece. We are currently trying to develop a design for the adapter that would accommodate circuit boards from most web cameras with minimal modification and fit 99% of all light microscopes, with the intention to distribute the blueprints for 3D printing. 212 VII.3. Software. We used ImageJ as the platform of choice for the development of our analysis algorithm in the form of an ImageJ macros. ImageJ has several tools pertaining to our goal, making it possible to approach the problem differently. First, the “Find Maxima…” method determines the local intensity maxima in the picture and counts them. Second, “Threshold…” command can separate background from the particles of interest, which can then be counted by the “Analyze Particles…” command. In order for the first algorithm to find maxima properly, the image needs to be blurred. This filter replaces each pixel with the average of its 3×3 neighborhood, and as such small imperfections of high intensity, contributing to false positives will be eliminated. Next, the appropriate noise threshold was empirically estimated and applied (see Code, Macros 1). The second approach employs conversion to binary. The RGB image is converted to grayscale (8- or 16-bit), and the threshold is found – the intensity value that allows to color everything below this value white, and everything above – black. After that, the ImageJ analysis algorithm can be applied to count all the black particles of a particular shape and size. We have set it up to cut off all the particles with the area smaller than 100 px (noise), but did not impose any restrictions on shape. The reason for the latter is the tendency of cells to coalesce at higher concentrations; in this case the software would thus at least detect the aggregate, albeit only as one particle (see Code, Macros 2). We have integrated the core functionality into the interactive envelope that prompts for the folder containing the pictures to be analyzed, parses it for jpeg images, applies either core analytical algorithm, shows and saves the results, and copies them into the system 213 clipboard. Considering that plugin execution in ImageJ can be assigned to a keyboard shortcut (Plugins>Shortcuts>Create Shortcut…), cell counting turns into literally just a press of a button. 214 VII.4. Results. U251 cells were prepared, and counted manually by three investigators with considerable experience in mammalian cell culture. The pictures of each of the eight areas on two panels of the hemocytometer were taken (Figure VII.1A), and the cells were counted automatically (Table VII.2). Since the original images are 2 times larger than the 1×1 mm area designated for manual counting, for the purpose of fair comparison we cropped the original images accordingly. The images were processed using Macros 1, employing blurring of the image (Figure VII.1C), and counting the intensity maxima (Figure VII.1D), and Macros 2, involving estimation of the intensity threshold and subsequently making the image binary (Figure VII.1B). There are some cells that were not accounted for by either algorithm due to the tendency of cells to coalesce at elevated concentrations (vide supra), however the rate of these events normally does not exceed 10%, and is somewhat offset by the inevitable inclusion of the few false positives as well. Table VII.2. Results of manual and automatic cell counting on the 1×1 mm area and total manual cell counting and picture taking time. Dilution, fold Average time, min Cell count Manual count Taking 8 pictures Manual Macros 1 Macros 2 Mean SEM Mean SEM Mean SEM 1 13.98 1.60 185 5 180 6 177 8 2 6.21 1.35 85 6 88 4 91 4 4 3.52 1.57 41 2 45 3 43 3 215 Both algorithms have shown remarkable consistency with each other and the manual cell count (Table VII.2, Figure VII.2A) throughout the cell concentration range, from very high to very low, despite the issues with the cells coalescing at higher concentrations. Importantly, it took from 3 to 14 minutes to count the cells manually depending on the dilution factor, i.e. cell concentration (Table VII.2, Figure VII.2B), and only 1.5 minutes to carefully take 8 pictures of the areas of interest. We did not include the time it took 1) to create a separate folder and copy the pictures there for analysis and 2) run either macros, as we estimated it totals under 10 seconds. 216 Figure VII.1. Application of the automatic cell counting methods. Cells in suspensions prepared by serial dilutions were deposited into a hemocytometer and the pictures of each of the eight areas on two panels were taken and cropped to 1×1 mm. Representative picture of one area at 1x dilution is shown on panel (A). The image was processed using “Threshold…” method (B). Alternatively, the image (A) was blurred to afford (C), then the maxima were counted and all cells accounted for were laid over the original image as a mask of yellow crosshairs to give (D). Yellow arrows: representative cells that were not counted. A B C D 217 We have not included the time it took to crop the pictures (vide supra) either, as this has been done only for the purpose of the fair comparison within the scope of this work, and is not pertaining to how the automatic counting procedure is designed to be implemented. 1 2 3 4 0 50 100 150 200 Manual Automatic 1 Automatic 2 Dilution fold Cell count 1 2 3 4 0 5 10 15 Manual Automatic Dilution fold Time, min Figure VII.2. Comparison of manual and automatic counting. (A) Cells in suspensions prepared by serial dilutions were counted using a hemocytometer and two different algorithms Macros 1 and Macros 2 in ImageJ on the 1×1 mm area. (B) Average total time spent on manual and automatic (picture taking) counting as a function of dilution fold of the sample. Error bars are ± SEM. There seems to be no information in the literature regarding which range of cell concentration is most suitable for the counting with hemocytometer. We have discovered, however, that when the manual cell count data is normalized via multiplication by the corresponding dilution factor, the values obtained were significantly different from one another (Figure VII.3), while they are obviously supposed to be the same. Thus there is a well-defined trend for the significant decrease of the measured cell count as the dilution of the sample increases. We hypothesized that such discrepancy may arise due to uneven distribution of cells in the hemocytometer assembly, and the probability of such phenomenon increases with the decrease in the cell concentration. (A) (B) 218 1x 2x 4x 0 50 100 150 200 *** *** ** Dilution, fold Cell count Figure VII.3. Normalized manual counting results. Cells in suspensions prepared by serial dilutions were counted using a hemocytometer. The means and standard deviations were normalized via multiplication by the dilution factor. Error bars are ± SEM. *** P < 0.001, ** P < 0.01, t-test. Fortunately, when the cells are counted automatically, one does not have to limit themselves to any particular area of hemocytometer, with one exception of densely gridded vertical and horizontal medians (intersections of the lines heavily contribute to false positives). Moreover, the area captured by the camera (2 mm 2 ) is two times larger than the one suggested for manual cell counting, thus encompassing larger cell population. As such, we carefully examined all the areas of interest, took pictures of the portions that seemed most representative and homogenous, and analyzed them using our algorithms (Figure VII.4). The average total cell count was surprisingly consistent between samples of different dilutions (for both methods), however it was significantly higher at 220±17 cells per 1×1 mm area (for both methods) than the highest manual count. This result demonstrates that the key advantage of the automatic cell counting is not even simplicity, 219 but rather consistency, and poses an important question as to whether the manual cell counting even at reasonably high cell concentration provides a reliable result. 1 2 3 4 0 100 200 300 Automatic 1 Automatic 2 Dilution fold Cell count Figure VII.4. Normalized automatic counting results for preferred fields. Cells in suspensions prepared by serial dilutions were counted automatically. The means and standard deviations were normalized via multiplication by the dilution factor. Error bars are ± SEM. Lastly, such approach to cell counting leaves “a digital trace”. Macros 1 and 2 save the files with the results, and Macros 2 also saves the processed images in a separate directory within the one being analyzed. Both algorithms copy the results into the clipboard at the end of the run, hence all that is left for the user is to paste the data into any spreadsheet template. 220 VII.5. Conclusion. We have worked out a robust approach to equip a conventional light microscope with a camera to obtain usable images with a reasonably wide FOV. We have devised two algorithms to automatically count mammalian cells in the pictures of the cell suspension in the hemocytometer assembly. Such approach takes 2-10 times less time, leaves a digital trace, and yields the same results as manual count on the same area. Importantly, it is not mandatory for the user to count the cells in the designated areas of the hemocytometer. Thus such approach allows for the selection of the most representative areas and does not impose any penalty on the number of such areas to be processed, therefore yielding consistent and arguably more reliable results, independent of the sample dilution. 221 VII.6. Experimental Section for Chapter VII. U251 cells (Figure VII.5) at a 50% confluency in a T75 flask (Greiner) were washed with PBS (10 ml), and detached using 1 ml of trypsin (Invitrogen, 0.05%, in PBS). The cells were resuspended in 9 ml of medium (DMEM, Invitrogen), supplemented with 10% FBS (Irvine Scientific) and antibiotic (penicillin 50 units/ml and streptomycin 50 µg/ml, Invitrogen), and centrifuged at 1250 rpm for 5 min. The supernatant was aspirated, and the pellet was resuspended in 1 ml of medium. The suspension was serially diluted to afford two and four times diluted samples, and an aliquot (100 µl) was taken for cell counting from each of these samples. Each aliquot was diluted with trypan blue (100 µl, 67% v/v solution in PBS), and deposited into the hemocytometer (Bright-Line, Hausser Scientific). Figure VII.5. U251 cells prior to lysis. 222 Next, the manual counting was conducted by three separate investigators using a standard hand tally counter (VWR) on a total of eight 1 mm 2 areas located on the two panels of the hemocytometer. The average number of cells per area per investigator was calculated, then the grand mean and the composite standard deviation were established,(407) and the outliers were removed based on Chauvenet’s criterion. The time the investigators spent counting was recorded (just the counting time, excluding all the preparatory work) using a timer. The time it took to take pictures of eight areas of interest was accurately determined by analyzing the automatic timestamps in the filenames. 223 VII.7. Instructions. 1) If you have ImageJ, skip to step 2). Otherwise, download and install ImageJ: http://rsbweb.nih.gov/ij/download.html 2) Download the plugins: https://dl.dropboxusercontent.com/u/23540218/Automatic%20cell%20counting/Automatic%20cell%2 0counting.zip 3) Unpack the archive. 4) Place the plugins into the ImageJ plugins directory: C:\Program Files\ImageJ\plugins 5) Launch ImageJ. 6) To launch either Macros 1 or Macros 2, go to Plugins > Macros 1 Plugins > Macros 2 7) The plugin will prompt you for a folder containing the images to be analyzed. 8) Output: • Macros 1: a. File “Cell counting results.txt” in the folder with the analyzed images. b. Copy of the results in the system clipboard. • Macros 2: a. Folder with the processed images in the folder with the analyzed images. b. File “Cell counting results.txt” in the folder with the analyzed images. c. Copy of the results in the system clipboard. 224 VII.8. Code. VII.8.1. Macros 1. macro "cell_count--Maxima" { //show prompt for selection of source directory dir = getDirectory("Choose source directory"); list = getFileList(dir); //get the file list setBatchMode(true); //hide all the details from user //process every file... for (i=0; i<list.length; i++) { //...that has .jpg extension if (endsWith(list[i], ".jpg") { open(dir+list[i]); //smooth the image 20 times for (j=0; j<20; j++) { run("Smooth"); } //find intensity maxima run("Find Maxima...", "noise=20 output=[Count] exclude light"); selectWindow(list[i]); close(); } } //summarize all data and copy the results run("Summarize"); String.copyResults(); setBatchMode(false); //create a text file with counting results saveAs("Results",dir+"Cell counting results.txt"); } 225 VII.8.2. Macros 2. macro "Cell_count--Threshold" { //show prompt for selection of source directory dir = getDirectory("Choose source directory"); list = getFileList(dir); //get the file list //make directory to store processed images File.makeDirectory(dir+"Processed_images"); setBatchMode(true); //hide all the details from user //process every file... for (i=0; i<list.length; i++) { //...that has .jpg extension if (endsWith(list[i], ".jpg")) { open(dir+list[i]); fileNoExt = split(list[i], "."); run("16-bit"); //convert image to 16-bit setAutoThreshold("Default"); //run("Threshold...") run("Convert to Mask"); //convert to mask and analyze run("Analyze Particles...", "size=100-1000 circularity=0.00-1.00 show=Outlines clear include summarize"); // save the file in a new directory under a new name and close all windows saveAs("Jpeg", dir+"Processed_images"+File.separator+"Analysis_of_"+fileNoExt[0]); close(); selectWindow(list[i]); close(); } } //copy all relevant contents from the Summary window selectWindow("Summary"); text = getInfo("window.contents"); lines = split(text, "\n"); 226 //create a text file with counting results, output only cell count //copy the results into clipboard String.resetBuffer; //reset string buffer for (i=0; i<lines.length; i++) { if (i==0){ File.saveString("",dir+"Cell counting results.txt"); f = File.open(dir+"Cell counting results.txt"); } labels = split(lines[i], "\t"); print(f,labels[1]); String.append(labels[1]+"\n"); //append the value to string } File.close(f); //open the summary file, close Summary window open(dir+"Cell counting results.txt"); selectWindow("Summary"); run("Close"); String.copy(String.buffer); //copy all values into clipboard setBatchMode(false); } 227 References 1. D. Engelkamp, V. vanHeyningen, Transcription factors in disease. Curr Opin Genet Dev 6, 334-342 (1996); published online EpubJun (Doi 10.1016/S0959-437x(96)80011-6). 2. D. S. Latchman, Transcription-factor mutations and disease. The New England journal of medicine 334, 28-33 (1996); published online EpubJan 4 (10.1056/NEJM199601043340108). 3. D. S. Latchman, Inhibitory transcription factors. The international journal of biochemistry & cell biology 28, 965-974 (1996); published online EpubSep ( 4. F. Crick, Central Dogma of Molecular Biology. Nature 227, 561-563 (1970); published online Epub08/08/print ( 5. A. Steitz Thomas, in Structures and Mechanisms. (American Chemical Society, 2002), vol. 827, chap. 15, pp. 231-248. 6. F. Jacob, J. Monod, [Genes of structure and genes of regulation in the biosynthesis of proteins]. Comptes rendus hebdomadaires des seances de l'Academie des sciences 249, 1282-1284 (1959); published online EpubOct 5 ( 7. F. Jacob, D. Perrin, C. Sanchez, J. Monod, [Operon: a group of genes with the expression coordinated by an operator]. Comptes rendus hebdomadaires des seances de l'Academie des sciences 250, 1727-1729 (1960); published online EpubFeb 29 ( 8. F. Jacob, J. Monod, Genetic regulatory mechanisms in the synthesis of proteins. Journal of molecular biology 3, 318-356 (1961); published online EpubJun ( 9. F. Jacob, J. Monod, [Biochemical and Genetic Mechanisms of Regulation in the Bacterial Cell]. Bulletin de la Societe de chimie biologique 46, 1499-1532 (1964). 10. F. Jacob, A. Ullman, J. Monod, [the Promotor, a Genetic Element Necessary to the Expression of an Operon]. Comptes rendus hebdomadaires des seances de l'Academie des sciences 258, 3125-3128 (1964); published online EpubMar 16 ( 11. J. Monod, F. Jacob, Teleonomic mechanisms in cellular metabolism, growth, and differentiation. Cold Spring Harbor symposia on quantitative biology 26, 389-401 (1961). 12. J. Monod, J. P. Changeux, F. Jacob, Allosteric proteins and cellular control systems. Journal of molecular biology 6, 306-329 (1963); published online EpubApr ( 13. R. Mostoslavsky, Y. Bergman, DNA methylation: regulation of gene expression and role in the immune system. Biochimica et Biophysica Acta (BBA) - Reviews on Cancer 1333, F29-F50 (1997); published online Epub8/8/ (http://dx.doi.org/10.1016/S0304- 419X(97)00010-3). 14. E. U. Selker, DNA methylation and chromatin structure: a view from below. Trends in Biochemical Sciences 15, 103-107 (1990); published online Epub3// (http://dx.doi.org/10.1016/0968-0004(90)90193-F). 15. A. Razin, C. Webb, M. Szyf, J. Yisraeli, A. Rosenthal, T. Naveh-Many, N. Sciaky- Gallili, H. Cedar, Variations in DNA methylation during mouse cell differentiation in vivo and in vitro. Proc Natl Acad Sci U S A 81, 2275-2279 (1984); published online EpubApr ( 228 16. G. R. Smith, DNA supercoiling: Another level for regulating gene expression. Cell 24, 599-600 (1981); published online Epub6// (http://dx.doi.org/10.1016/0092- 8674(81)90085-4). 17. C. Naughton, N. Avlonitis, S. Corless, J. G. Prendergast, I. K. Mati, P. P. Eijk, S. L. Cockroft, M. Bradley, B. Ylstra, N. Gilbert, Transcription forms and remodels supercoiling domains unfolding large-scale chromatin structures. Nat Struct Mol Biol 20, 387-395 (2013); published online Epub03//print (http://www.nature.com/nsmb/journal/v20/n3/abs/nsmb.2509.html#supplementary- information). 18. B. Schuster-Bockler, B. Lehner, Chromatin organization is a major influence on regional mutation rates in human cancer cells. Nature 488, 504-507 (2012); published online EpubAug 23 (Doi 10.1038/Nature11273). 19. K. Struhl, Fundamentally Different Logic of Gene Regulation in Eukaryotes and Prokaryotes. Cell 98, 1-4 (1999); published online Epub7/9/ (http://dx.doi.org/10.1016/S0092-8674(00)80599-1). 20. P. P. Pandolfi, Transcription therapy for cancer. Oncogene 20, 3116-3127 (2001); published online EpubMay 28 (DOI 10.1038/sj.onc.1204299). 21. I. V. Shirinsky, V. S. Shirinsky, Targeting Nuclear Hormone Receptors: PPARalpha Agonists as Potential Disease-Modifying Drugs for Rheumatoid Arthritis. International journal of rheumatology 2011, 937843 (2011)10.1155/2011/937843). 22. A. H. Brivanlou, J. E. Darnell, Signal Transduction and the Control of Gene Expression. Science 295, 813-818 (2002); published online EpubFebruary 1, 2002 (10.1126/science.1066355). 23. J. E. Darnell, Transcription factors as targets for cancer therapy. Nature Reviews Cancer 2, 740-749 (2002); published online EpubOct (Doi 10.1038/Nrc906). 24. M. D. Delgado, J. Leon, Gene expression regulation and cancer. Clinical & translational oncology : official publication of the Federation of Spanish Oncology Societies and of the National Cancer Institute of Mexico 8, 780-787 (2006); published online EpubNov ( 25. A. L. Jochim, P. S. Arora, Assessment of helical interfaces in protein-protein interactions. Molecular bioSystems 5, 924-926 (2009); published online EpubSep (10.1039/b903202a). 26. Y. Zhang, Q. Mu, H. Zhou, K. Vrijens, M. F. Roussel, G. Jiang, B. Yan, Binding of carbon nanotube to BMP receptor 2 enhances cell differentiation and inhibits apoptosis via regulating bHLH transcription factors. Cell death & disease 3, e308 (2012)10.1038/cddis.2012.48). 27. C. J. Marcus-Sekura, Techniques for using antisense oligodeoxyribonucleotides to study gene expression. Analytical Biochemistry 172, 289-295 (1988); published online Epub8/1/ (http://dx.doi.org/10.1016/0003-2697(88)90447-2). 28. J. G. Karras. (Google Patents, 2013). 29. K. W. Dobie, S. M. Freier, J. G. Karras. (Google Patents, 2003). 30. S. Bhanot, X. X. Yu. (Google Patents, 2013). 31. B. F. Baker, S. M. Freier. (Google Patents, 2003). 32. H. Liang, Z. Lou, V. M. Maher, J. J. Mccormick. (Google Patents, 2005). 229 33. P. Nielsen, M. Egholm, R. Berg, O. Buchardt, Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science 254, 1497-1500 (1991); published online EpubDecember 6, 1991 (10.1126/science.1962210). 34. K. Kaihatsu, B. A. Janowski, D. R. Corey, Recognition of Chromosomal DNA by PNAs. Chemistry & Biology 11, 749-758 (2004); published online Epub6// (http://dx.doi.org/10.1016/j.chembiol.2003.09.014). 35. M. Schlee, V. Hornung, G. Hartmann, siRNA and isRNA: Two Edges of One Sword. Mol Ther 14, 463-470 (2006); published online Epub10//print ( 36. F. Heil, H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, S. Bauer, Species-Specific Recognition of Single-Stranded RNA via Toll-like Receptor 7 and 8. Science 303, 1526-1529 (2004); published online EpubMarch 5, 2004 (10.1126/science.1093620). 37. J. Cho, M. E. Parks, P. B. Dervan, Cyclic polyamides for recognition in the minor groove of DNA. Proceedings of the National Academy of Sciences of the United States of America 92, 10389-10392 (1995); published online EpubOctober 24, 1995 ( 38. B. Z. Olenyuk, G.-J. Zhang, J. M. Klco, N. G. Nickols, W. G. Kaelin, P. B. Dervan, Inhibition of vascular endothelial growth factor with a sequence-specific hypoxia response element antagonist. Proceedings of the National Academy of Sciences of the United States of America 101, 16768-16773 (2004); published online EpubNovember 30, 2004 (10.1073/pnas.0407617101). 39. P. Simon, F. Cannata, L. Perrouault, L. Halby, J.-P. Concordet, A. Boutorine, V. Ryabinin, A. Sinyakov, C. Giovannangeli, Sequence-specific DNA cleavage mediated by bipyridine polyamide conjugates. Nucleic Acids Research 36, 3531-3538 (2008); published online EpubJune 1, 2008 (10.1093/nar/gkn231). 40. M. A. Bill, C. Nicholas, T. A. Mace, J. P. Etter, C. Li, E. B. Schwartz, J. R. Fuchs, G. S. Young, L. Lin, J. Lin, L. He, M. Phelps, P.-K. Li, G. B. Lesinski, Structurally Modified Curcumin Analogs Inhibit STAT3 Phosphorylation and Promote Apoptosis of Human Renal Cell Carcinoma and Melanoma Cell Lines. PLoS ONE 7, e40724 (2012)10.1371/journal.pone.0040724). 41. K. Siddiquee, S. Zhang, W. C. Guida, M. A. Blaskovich, B. Greedy, H. R. Lawrence, M. L. R. Yip, R. Jove, M. M. McLaughlin, N. J. Lawrence, S. M. Sebti, J. Turkson, Selective chemical probe inhibitor of Stat3, identified through structure-based virtual screening, induces antitumor activity. Proceedings of the National Academy of Sciences of the United States of America 104, 7391-7396 (2007); published online EpubMay 1, 2007 (10.1073/pnas.0609757104). 42. R. N. Eisenman, Deconstructing Myc. Genes & Development 15, 2023-2030 (2001); published online EpubAugust 15, 2001 (10.1101/gad928101). 43. T. Berg, S. B. Cohen, J. Desharnais, C. Sonderegger, D. J. Maslyar, J. Goldberg, D. L. Boger, P. K. Vogt, Small-molecule antagonists of Myc/Max dimerization inhibit Myc- induced transformation of chicken embryo fibroblasts. Proceedings of the National Academy of Sciences of the United States of America 99, 3830-3835 (2002); published online EpubMarch 19, 2002 (10.1073/pnas.062036999). 44. X. Lu, P. K. Vogt, D. L. Boger, J. Lunec, Disruption of the MYC transcriptional function by a small-molecule antagonist of MYC/MAX dimerization. Oncology reports 19, 825- 830 (2008); published online EpubMar ( 230 45. Z. Arany, L. E. Huang, R. Eckner, S. Bhattacharya, C. Jiang, M. A. Goldberg, H. F. Bunn, D. M. Livingston, An essential role for p300/CBP in the cellular response to hypoxia. Proceedings of the National Academy of Sciences of the United States of America 93, 12969-12973 (1996); published online EpubNovember 12, 1996 ( 46. L. H. Gray, A. D. Conger, M. Ebert, S. Hornsey, O. C. A. Scott, The Concentration of Oxygen Dissolved in Tissues at the Time of Irradiation as a Factor in Radiotherapy. The British Journal of Radiology 26, 638-648 (1953); published online Epub1953/12/01 (10.1259/0007-1285-26-312-638). 47. G. Powis, L. Kirkpatrick, Hypoxia inducible factor-1α as a cancer drug target. Molecular Cancer Therapeutics 3, 647-654 (2004); published online EpubMay 1, 2004 ( 48. J. Pouyssegur, F. Dayan, N. M. Mazure, Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441, 437-443 (2006); published online Epub05/25/print ( 49. J. A. Bertout, S. A. Patel, M. C. Simon, The impact of O2 availability on human cancer. Nat Rev Cancer 8, 967-975 (2008); published online Epub12//print ( 50. G. L. Semenza, Evaluation of HIF-1 inhibitors as anticancer agents. Drug Discovery Today 12, 853-859 (2007); published online Epub10// (http://dx.doi.org/10.1016/j.drudis.2007.08.006). 51. D. Hanahan, R. A. Weinberg, The hallmarks of cancer. Cell 100, 57-70 (2000); published online EpubJan 7 (Doi 10.1016/S0092-8674(00)81683-9). 52. R. G. Bristow, R. P. Hill, Hypoxia and metabolism: Hypoxia, DNA repair and genetic instability. Nat Rev Cancer 8, 180-192 (2008); published online Epub03//print (http://www.nature.com/nrc/journal/v8/n3/suppinfo/nrc2344_S1.html). 53. G. L. Semenza, G. L. Wang, A Nuclear Factor Induced by Hypoxia Via Denovo Protein- Synthesis Binds to the Human Erythropoietin Gene Enhancer at a Site Required for Transcriptional Activation. Mol Cell Biol 12, 5447-5454 (1992); published online EpubDec ( 54. K. Ruan, G. Song, G. Ouyang, Role of hypoxia in the hallmarks of human cancer. Journal of cellular biochemistry 107, 1053-1062 (2009); published online EpubAug 15 (10.1002/jcb.22214). 55. G. L. Semenza, M. K. Nejfelt, S. M. Chi, S. E. Antonarakis, Hypoxia-inducible nuclear factors bind to an enhancer element located 3' to the human erythropoietin gene. Proc Natl Acad Sci U S A 88, 5680-5684 (1991); published online EpubJul 1 ( 56. G. L. Semenza, G. L. Wang, A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 12, 5447-5454 (1992); published online EpubDecember 1, 1992 (10.1128/mcb.12.12.5447). 57. G. L. Wang, B. H. Jiang, E. A. Rue, G. L. Semenza, Hypoxia-Inducible Factor-1 Is a Basic-Helix-Loop-Helix-Pas Heterodimer Regulated by Cellular O-2 Tension. Proceedings of the National Academy of Sciences of the United States of America 92, 5510-5514 (1995); published online EpubJun 6 (DOI 10.1073/pnas.92.12.5510). 58. B.-H. Jiang, E. Rue, G. L. Wang, R. Roe, G. L. Semenza, Dimerization, DNA Binding, and Transactivation Properties of Hypoxia-inducible Factor 1. Journal of Biological Chemistry 271, 17771-17778 (1996); published online EpubJuly 26, 1996 ( 231 59. L. Semenza Gregg, F. Agani, G. Booth, J. Forsythe, N. Iyer, B.-H. Jiang, S. Leung, R. Roe, C. Wiener, A. Yu, Structural and functional analysis of hypoxia-inducible factor 1. Kidney Int 51, 553-555 (1997); published online Epub02//print. 60. P. Carrero, K. Okamoto, P. Coumailleau, S. O'Brien, H. Tanaka, L. Poellinger, Redox- regulated recruitment of the transcriptional coactivators CREB-binding protein and SRC- 1 to hypoxia-inducible factor 1alpha. Mol Cell Biol 20, 402-415 (2000); published online EpubJan. 61. L. E. Huang, Z. Arany, D. M. Livingston, H. F. Bunn, Activation of Hypoxia-inducible Transcription Factor Depends Primarily upon Redox-sensitive Stabilization of Its α Subunit. Journal of Biological Chemistry 271, 32253-32259 (1996); published online EpubDecember 13, 1996. 62. J. Gu, J. Milligan, L. E. Huang, Molecular Mechanism of Hypoxia-inducible Factor 1α- p300 Interaction: A LEUCINE-RICH INTERFACE REGULATED BY A SINGLE CYSTEINE. Journal of Biological Chemistry 276, 3550-3554 (2001); published online EpubFebruary 2, 2001 (10.1074/jbc.M009522200). 63. U. R. JEWELL, I. KVIETIKOVA, A. SCHEID, C. BAUER, R. H. WENGER, M. GASSMANN, Induction of HIF-1α in response to hypoxia is instantaneous. The FASEB Journal 15, 1312-1314 (2001); published online EpubMay 1, 2001 (10.1096/fj.00- 0732fje). 64. P. Jaakkola, D. R. Mole, Y.-M. Tian, M. I. Wilson, J. Gielbert, S. J. Gaskell, A. v. Kriegsheim, H. F. Hebestreit, M. Mukherji, C. J. Schofield, P. H. Maxwell, C. W. Pugh, P. J. Ratcliffe, Targeting of HIF-α to the von Hippel-Lindau Ubiquitylation Complex by O2-Regulated Prolyl Hydroxylation. Science 292, 468-472 (2001); published online EpubApril 20, 2001 (10.1126/science.1059796). 65. N. Masson, C. Willam, P. H. Maxwell, C. W. Pugh, P. J. Ratcliffe, Independent function of two destruction domains in hypoxia‐inducible factor‐α chains activated by prolyl hydroxylation. (2001), vol. 20, pp. 5197-5206. 66. A. C. R. Epstein, J. M. Gleadle, L. A. McNeill, K. S. Hewitson, J. O'Rourke, D. R. Mole, M. Mukherji, E. Metzen, M. I. Wilson, A. Dhanda, Y.-M. Tian, N. Masson, D. L. Hamilton, P. Jaakkola, R. Barstead, J. Hodgkin, P. H. Maxwell, C. W. Pugh, C. J. Schofield, P. J. Ratcliffe, C. elegans EGL-9 and Mammalian Homologs Define a Family of Dioxygenases that Regulate HIF by Prolyl Hydroxylation. Cell 107, 43-54 (2001); published online Epub10/5/ (http://dx.doi.org/10.1016/S0092-8674(01)00507-4). 67. E. Berra, E. Benizri, A. Ginouvès, V. Volmat, D. Roux, J. Pouysségur, HIF prolyl‐ hydroxylase 2 is the key oxygen sensor setting low steady‐state levels of HIF‐1α in normoxia. (2003), vol. 22, pp. 4082-4090. 68. C. E. Dann Iii, R. K. Bruick, Dioxygenases as O2-dependent regulators of the hypoxic response pathway. Biochem Bioph Res Co 338, 639-647 (2005); published online Epub12/9/ (http://dx.doi.org/10.1016/j.bbrc.2005.08.140). 69. P. H. Maxwell, M. S. Wiesener, G.-W. Chang, S. C. Clifford, E. C. Vaux, M. E. Cockman, C. C. Wykoff, C. W. Pugh, E. R. Maher, P. J. Ratcliffe, The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399, 271-275 (1999); published online Epub05/20/print (http://www.nature.com/nature/journal/v399/n6733/suppinfo/399271a0_S1.html). 232 70. T. Kamura, S. Sato, K. Iwai, M. Czyzyk-Krzeska, R. C. Conaway, J. W. Conaway, Activation of HIF1α ubiquitination by a reconstituted von Hippel-Lindau (VHL) tumor suppressor complex. Proceedings of the National Academy of Sciences of the United States of America 97, 10430-10435 (2000); published online EpubSeptember 12, 2000 (10.1073/pnas.190332597). 71. D. Lando, D. J. Peet, D. A. Whelan, J. J. Gorman, M. L. Whitelaw, Asparagine Hydroxylation of the HIF Transactivation Domain: A Hypoxic Switch. Science 295, 858- 861 (2002); published online EpubFebruary 1, 2002 (10.1126/science.1068592). 72. D. Lando, D. J. Peet, J. J. Gorman, D. A. Whelan, M. L. Whitelaw, R. K. Bruick, FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes & Development 16, 1466-1471 (2002); published online EpubJune 15, 2002 (10.1101/gad.991402). 73. K. Lisy, D. J. Peet, Turn me on: regulating HIF transcriptional activity. Cell Death Differ 15, 642-649 (2008); published online Epub01/18/online. 74. J.-W. Jeong, M.-K. Bae, M.-Y. Ahn, S.-H. Kim, T.-K. Sohn, M.-H. Bae, M.-A. Yoo, E. J. Song, K.-J. Lee, K.-W. Kim, Regulation and Destabilization of HIF-1α by ARD1- Mediated Acetylation. Cell 111, 709-720 (2002); published online Epub11/27/ (http://dx.doi.org/10.1016/S0092-8674(02)01085-1). 75. J.-w. Kim, I. Tchernyshyov, G. L. Semenza, C. V. Dang, HIF-1-mediated expression of pyruvate dehydrogenase kinase: A metabolic switch required for cellular adaptation to hypoxia. Cell Metabolism 3, 177-185 (2006); published online Epub3// (http://dx.doi.org/10.1016/j.cmet.2006.02.002). 76. S. N. Greer, J. L. Metcalf, Y. Wang, M. Ohh, The updated biology of hypoxia‐inducible factor. (2012), vol. 31, pp. 2448-2460. 77. G. L. Semenza, Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr Opin Genet Dev 8, 588-594 (1998); published online Epub10// (http://dx.doi.org/10.1016/S0959-437X(98)80016-6). 78. W. N. William, J. V. Heymach, E. S. Kim, S. M. Lippman, Molecular targets for cancer chemoprevention. Nat Rev Drug Discov 8, 213-225 (2009); published online Epub03//print. 79. M. J. Gray, J. Zhang, L. M. Ellis, G. L. Semenza, D. B. Evans, S. S. Watowich, G. E. Gallick, HIF-1[alpha], STAT3, CBP//p300 and Ref-1//APE are components of a transcriptional complex that regulates Src-dependent hypoxia-induced expression of VEGF in pancreatic and prostate carcinomas. 24, 3110-3120 (2005); published online Epub02/21/online. 80. Q. Ke, M. Costa, Hypoxia-Inducible Factor-1 (HIF-1). Molecular Pharmacology 70, 1469-1480 (2006); published online EpubNovember 1, 2006 (10.1124/mol.106.027029). 81. I. Mylonis, G. Chachami, M. Samiotaki, G. Panayotou, E. Paraskeva, A. Kalousi, E. Georgatsou, S. Bonanou, G. Simos, Identification of MAPK Phosphorylation Sites and Their Role in the Localization and Activity of Hypoxia-inducible Factor-1α. Journal of Biological Chemistry 281, 33095-33106 (2006); published online EpubNovember 3, 2006 (10.1074/jbc.M605058200). 82. D. E. Richard, E. Berra, E. Gothié, D. Roux, J. Pouysségur, p42/p44 Mitogen-activated Protein Kinases Phosphorylate Hypoxia-inducible Factor 1α (HIF-1α) and Enhance the 233 Transcriptional Activity of HIF-1. Journal of Biological Chemistry 274, 32631-32637 (1999); published online EpubNovember 12, 1999. 83. H. Suzuki, A. Tomida, T. Tsuruo, Dephosphorylated hypoxia-inducible factor 1 alpha as a mediator of p53-dependent apoptosis during hypoxia. Oncogene 20, 5779-5788 (2001); published online EpubSep 13 (DOI 10.1038/sj.onc.1204742). 84. D. Flügel, A. Görlach, C. Michiels, T. Kietzmann, Glycogen Synthase Kinase 3 Phosphorylates Hypoxia-Inducible Factor 1α and Mediates Its Destabilization in a VHL- Independent Manner. Mol Cell Biol 27, 3253-3265 (2007); published online EpubMay 1, 2007 (10.1128/mcb.00015-07). 85. D. Feldser, F. Agani, N. V. Iyer, B. Pak, G. Ferreira, G. L. Semenza, Reciprocal Positive Regulation of Hypoxia-inducible Factor 1α and Insulin-like Growth Factor 2. Cancer Res 59, 3915-3918 (1999); published online EpubAugust 15, 1999. 86. T. Hellwig-Bürgel, K. Rutkowski, E. Metzen, J. Fandrey, W. Jelkmann, Interleukin-1β and Tumor Necrosis Factor-α Stimulate DNA Binding of Hypoxia-Inducible Factor-1. (1999), vol. 94, pp. 1561-1567. 87. K. Salnikow, W. G. An, G. Melillo, M. V. Blagosklonny, M. Costa, Nickel-induced transformation shifts the balance between HIF-1 and p53 transcription factors. Carcinogenesis 20, 1819-1823 (1999); published online EpubSeptember 1, 1999 (10.1093/carcin/20.9.1819). 88. J. Li, G. Davidson, Y. Huang, B.-H. Jiang, X. Shi, M. Costa, C. Huang, Nickel Compounds Act through Phosphatidylinositol-3-kinase/Akt-Dependent, p70S6k- Independent Pathway to Induce Hypoxia Inducible Factor Transactivation and Cap43 Expression in Mouse Epidermal Cl41 Cells. Cancer Res 64, 94-101 (2004); published online EpubJanuary 1, 2004 (10.1158/0008-5472.can-03-0737). 89. D. Mottet, V. Dumont, Y. Deccache, C. Demazy, N. Ninane, M. Raes, C. Michiels, Regulation of Hypoxia-inducible Factor-1α Protein Level during Hypoxic Conditions by the Phosphatidylinositol 3-Kinase/Akt/Glycogen Synthase Kinase 3β Pathway in HepG2 Cells. Journal of Biological Chemistry 278, 31277-31285 (2003); published online EpubAugust 15, 2003 (10.1074/jbc.M300763200). 90. D. P. Stiehl, W. Jelkmann, R. H. Wenger, T. Hellwig-Bürgel, Normoxic induction of the hypoxia-inducible factor 1α by insulin and interleukin-1β involves the phosphatidylinositol 3-kinase pathway. FEBS Letters 512, 157-162 (2002); published online Epub2/13/ (http://dx.doi.org/10.1016/S0014-5793(02)02247-0). 91. A. M. Arsham, D. R. Plas, C. B. Thompson, M. C. Simon, PI3K/Akt signaling is neither required for hypoxic stabilization of HIF-1α nor sufficient for HIF-1-dependent target gene transcription. Journal of Biological Chemistry, (2002); published online EpubFebruary 21, 2002 (10.1074/jbc.M111162200). 92. M. Y. Koh, T. R. Spivak-Kroizman, G. Powis, HIF-1 regulation: not so easy come, easy go. Trends in Biochemical Sciences 33, 526-534 (2008); published online Epub11// (http://dx.doi.org/10.1016/j.tibs.2008.08.002). 93. Z. Zhou, F. Liu, Z.-S. Zhang, F. Shu, Y. Zheng, L. Fu, L.-Y. Li, Human Rhomboid Family-1 Suppresses Oxygen-Independent Degradation of Hypoxia-Inducible Factor-1α in Breast Cancer. Cancer Res 74, 2719-2730 (2014); published online EpubMay 15, 2014 (10.1158/0008-5472.can-13-1027). 234 94. R. Ravi, B. Mookerjee, Z. M. Bhujwalla, C. H. Sutter, D. Artemov, Q. Zeng, L. E. Dillehay, A. Madan, G. L. Semenza, A. Bedi, Regulation of tumor angiogenesis by p53- induced degradation of hypoxia-inducible factor 1α. Genes & Development 14, 34-44 (2000); published online EpubJanuary 1, 2000 (10.1101/gad.14.1.34). 95. A.-L. Nieminen, S. Qanungo, E. A. Schneider, B.-H. Jiang, F. H. Agani, Mdm2 and HIF- 1α interaction in tumor cells during hypoxia. Journal of Cellular Physiology 204, 364- 369 (2005)10.1002/jcp.20406). 96. C. C. Hudson, M. Liu, G. G. Chiang, D. M. Otterness, D. C. Loomis, F. Kaper, A. J. Giaccia, R. T. Abraham, Regulation of Hypoxia-Inducible Factor 1α Expression and Function by the Mammalian Target of Rapamycin. Mol Cell Biol 22, 7004-7014 (2002); published online EpubOctober 15, 2002 (10.1128/mcb.22.20.7004-7014.2002). 97. N. S. Kenneth, S. Rocha, Regulation of gene expression by hypoxia. Biochemical Journal 414, 19-29 (2008). 98. M. Ema, S. Taya, N. Yokotani, K. Sogawa, Y. Matsuda, Y. Fujii-Kuriyama, A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1α regulates the VEGF expression and is potentially involved in lung and vascular development. Proceedings of the National Academy of Sciences of the United States of America 94, 4273-4278 (1997); published online EpubApril 29, 1997. 99. H. Tian, S. L. McKnight, D. W. Russell, Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes & Development 11, 72-82 (1997); published online EpubJanuary 1, 1997 (10.1101/gad.11.1.72). 100. M. Krieg, R. Haas, H. Brauch, T. Acker, I. Flamme, K. H. Plate, Up-regulation of hypoxia-inducible factors HIF-1alpha and HIF-2alpha under normoxic conditions in renal carcinoma cells by von Hippel-Lindau tumor suppressor gene loss of function. Oncogene 19, 5435-5443 (2000); published online EpubNov 16 (10.1038/sj.onc.1203938). 101. H.-X. Sun, Y. Xu, X.-R. Yang, W.-M. Wang, H. Bai, R.-Y. Shi, S. K. Nayar, R. P. Devbhandari, Y.-z. He, Q.-F. Zhu, Y.-F. Sun, B. Hu, M. Khan, R. A. Anders, J. Fan, Hypoxia inducible factor 2 alpha inhibits hepatocellular carcinoma growth through the transcription factor dimerization partner 3/ E2F transcription factor 1–dependent apoptotic pathway. Hepatology 57, 1088-1097 (2013)10.1002/hep.26188). 102. J. E. S. Shay, M. Celeste Simon, Hypoxia-inducible factors: Crosstalk between inflammation and metabolism. Seminars in Cell & Developmental Biology 23, 389-394 (2012); published online Epub6// (http://dx.doi.org/10.1016/j.semcdb.2012.04.004). 103. Y. Makino, R. Cao, K. Svensson, G. Bertilsson, M. Asman, H. Tanaka, Y. Cao, A. Berkenstam, L. Poellinger, Inhibitory PAS domain protein is a negative regulator of hypoxia-inducible gene expression. Nature 414, 550-554 (2001); published online Epub11/29/print. 104. E. Maltepe, B. Keith, A. M. Arsham, J. R. Brorson, M. C. Simon, The Role of ARNT2 in Tumor Angiogenesis and the Neural Response to Hypoxia. Biochem Bioph Res Co 273, 231-238 (2000); published online Epub6/24/ (http://dx.doi.org/10.1006/bbrc.2000.2928). 105. S. Takahata, K. Sogawa, A. Kobayashi, M. Ema, J. Mimura, N. Ozaki, Y. Fujii- Kuriyama, Transcriptionally Active Heterodimer Formation of an Arnt-like PAS Protein, Arnt3, with HIF-1a, HLF, and Clock. Biochem Bioph Res Co 248, 789-794 (1998); published online Epub7/30/ (http://dx.doi.org/10.1006/bbrc.1998.9012). 235 106. G. L. Semenza, HIF-1 and human disease: one highly involved factor. Genes & Development 14, 1983-1991 (2000); published online EpubAugust 15, 2000 (10.1101/gad.14.16.1983). 107. N. V. Iyer, L. E. Kotch, F. Agani, S. W. Leung, E. Laughner, R. H. Wenger, M. Gassmann, J. D. Gearhart, A. M. Lawler, A. Y. Yu, G. L. Semenza, Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1α. Genes & Development 12, 149-162 (1998); published online EpubJanuary 15, 1998 (10.1101/gad.12.2.149). 108. W. Zhou, M. Choi, D. Margineantu, L. Margaretha, J. Hesson, C. Cavanaugh, C. A. Blau, M. S. Horwitz, D. Hockenbery, C. Ware, H. Ruohola‐Baker, HIF1α induced switch from bivalent to exclusively glycolytic metabolism during ESC‐to‐EpiSC/hESC transition. (2012), vol. 31, pp. 2103-2116. 109. T. Simsek, F. Kocabas, J. Zheng, R. J. DeBerardinis, A. I. Mahmoud, E. N. Olson, J. W. Schneider, C. C. Zhang, H. A. Sadek, The Distinct Metabolic Profile of Hematopoietic Stem Cells Reflects Their Location in a Hypoxic Niche. Cell Stem Cell 7, 380-390 (2010); published online Epub9/3/ (http://dx.doi.org/10.1016/j.stem.2010.07.011). 110. S. H. Lee, P. L. Wolf, R. Escudero, R. Deutsch, S. W. Jamieson, P. A. Thistlethwaite, Early Expression of Angiogenesis Factors in Acute Myocardial Ischemia and Infarction. New England Journal of Medicine 342, 626-633 (2000)doi:10.1056/NEJM200003023420904). 111. M. Bergeron, A. Y. Yu, K. E. Solway, G. L. Semenza, F. R. Sharp, Induction of hypoxia- inducible factor-1 (HIF-1) and its target genes following focal ischaemia in rat brain. European Journal of Neuroscience 11, 4159-4170 (1999)10.1046/j.1460- 9568.1999.00845.x). 112. K. J. Banasiak, G. G. Haddad, Hypoxia-induced apoptosis: effect of hypoxic severity and role of p53 in neuronal cell death. Brain Research 797, 295-304 (1998); published online Epub6/29/ (http://dx.doi.org/10.1016/S0006-8993(98)00286-8). 113. V. Nizet, R. S. Johnson, Interdependence of hypoxic and innate immune responses. Nat Rev Immunol 9, 609-617 (2009); published online Epub09//print. 114. H.-J. Jeong, H.-S. Chung, B.-R. Lee, S.-J. Kim, S.-J. Yoo, S.-H. Hong, H.-M. Kim, Expression of proinflammatory cytokines via HIF-1α and NF-κB activation on desferrioxamine-stimulated HMC-1 cells. Biochem Bioph Res Co 306, 805-811 (2003); published online Epub7/11/ (http://dx.doi.org/10.1016/S0006-291X(03)01073-8). 115. K. Bartels, A. Grenz, H. K. Eltzschig, Hypoxia and inflammation are two sides of the same coin. Proceedings of the National Academy of Sciences of the United States of America 110, 18351-18352 (2013); published online EpubNovember 12, 2013 (10.1073/pnas.1318345110). 116. J. Rius, M. Guma, C. Schachtrup, K. Akassoglou, A. S. Zinkernagel, V. Nizet, R. S. Johnson, G. G. Haddad, M. Karin, NF-[kgr]B links innate immunity to the hypoxic response through transcriptional regulation of HIF-1[agr]. Nature 453, 807-811 (2008); published online Epub06/05/print (http://www.nature.com/nature/journal/v453/n7196/suppinfo/nature06905_S1.html). 117. S. Frede, C. Stockmann, P. Freitag, J. Fandrey, Bacterial lipopolysaccharide induces HIF- 1 activation in human monocytes via p44/42 MAPK and NF-κB. Biochem J 396, 517-527 (2006); published online EpubJune 15, 2006 (10.1042/bj20051839). 236 118. C. Peyssonnaux, A. T. Boutin, A. S. Zinkernagel, V. Datta, V. Nizet, R. S. Johnson, Critical Role of HIF-1[alpha] in Keratinocyte Defense against Bacterial Infection. J Invest Dermatol 128, 1964-1968 (2008); published online Epub03/06/online. 119. T. D. Eubank, C. B. Marsh, HIFs: a-cute answer to inflammation? , (2011), vol. 118, pp. 485-487. 120. G. T. Furuta, J. R. Turner, C. T. Taylor, R. M. Hershberg, K. Comerford, S. Narravula, D. K. Podolsky, S. P. Colgan, Hypoxia-Inducible Factor 1–Dependent Induction of Intestinal Trefoil Factor Protects Barrier Function during Hypoxia. The Journal of Experimental Medicine 193, 1027-1034 (2001); published online EpubMay 7, 2001 (10.1084/jem.193.9.1027). 121. D. Chen, M. Li, J. Luo, W. Gu, Direct Interactions between HIF-1α and Mdm2 Modulate p53 Function. Journal of Biological Chemistry 278, 13595-13598 (2003); published online EpubApril 18, 2003 (10.1074/jbc.C200694200). 122. M. P. Scott, Intimations of a creature. Cell 79, 1121-1124 (1994); published online Epub12/30/ (http://dx.doi.org/10.1016/0092-8674(94)90001-9). 123. R. K. Bruick, Expression of the gene encoding the proapoptotic Nip3 protein is induced by hypoxia. Proceedings of the National Academy of Sciences of the United States of America 97, 9082-9087 (2000); published online EpubAugust 1, 2000 (10.1073/pnas.97.16.9082). 124. N. Akakura, M. Kobayashi, I. Horiuchi, A. Suzuki, J. Wang, J. Chen, H. Niizeki, K.-i. Kawamura, M. Hosokawa, M. Asaka, Constitutive Expression of Hypoxia-inducible Factor-1α Renders Pancreatic Cancer Cells Resistant to Apoptosis Induced by Hypoxia and Nutrient Deprivation. Cancer Res 61, 6548-6554 (2001); published online EpubSeptember 1, 2001. 125. G. L. Semenza, Hypoxia, Clonal Selection, and the Role of HIF-1 in Tumor Progression. Critical Reviews in Biochemistry and Molecular Biology 35, 71-103 (2000)doi:10.1080/10409230091169186). 126. C. W. Pugh, P. J. Ratcliffe, Regulation of angiogenesis by hypoxia: role of the HIF system. Nat Med 9, 677-684 (2003); published online Epub06//print ( 127. W. G. Kaelin, Molecular basis of the VHL hereditary cancer syndrome. Nat Rev Cancer 2, 673-682 (2002); published online Epub09//print. 128. C. Shen, W. G. Kaelin Jr, The VHL/HIF axis in clear cell renal carcinoma. Seminars in Cancer Biology 23, 18-25 (2013); published online Epub2// (http://dx.doi.org/10.1016/j.semcancer.2012.06.001). 129. A. E. Greijer, E. van der Wall, The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis. Journal of Clinical Pathology 57, 1009-1014 (2004); published online EpubOctober 1, 2004 (10.1136/jcp.2003.015032). 130. G. Q. Chen, J. Zhang, Hypoxia, Hypoxia-Inducing Factor-1 and Leukemic Cell Differentiation. Scientia Sinica Vitae 44, 114-118 (2014); published online Epub2014-02- 18 (10.1360/052013-353). 131. M. Scortegagna, K. Ding, Y. Oktay, A. Gaur, F. Thurmond, L.-J. Yan, B. T. Marck, A. M. Matsumoto, J. M. Shelton, J. A. Richardson, M. J. Bennett, J. A. Garcia, Multiple organ pathology, metabolic abnormalities and impaired homeostasis of reactive oxygen species in Epas1-/- mice. Nat Genet 35, 331-340 (2003); published online Epub12//print (http://www.nature.com/ng/journal/v35/n4/suppinfo/ng1266_S1.html). 237 132. D. J. Manalo, A. Rowan, T. Lavoie, L. Natarajan, B. D. Kelly, S. Q. Ye, J. G. N. Garcia, G. L. Semenza, Transcriptional regulation of vascular endothelial cell responses to hypoxia by HIF-1. (2005), vol. 105, pp. 659-669. 133. G. L. Semenza, Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3, 721-732 (2003); published online Epub10//print. 134. J. D. Gordan, M. C. Simon, Hypoxia-inducible factors: central regulators of the tumor phenotype. Curr Opin Genet Dev 17, 71-77 (2007); published online Epub2// (http://dx.doi.org/10.1016/j.gde.2006.12.006). 135. V. Wang, D. A. Davis, M. Haque, L. E. Huang, R. Yarchoan, Differential Gene Up- Regulation by Hypoxia-Inducible Factor-1α and Hypoxia-Inducible Factor-2α in HEK293T Cells. Cancer Res 65, 3299-3306 (2005); published online EpubApril 15, 2005 (10.1158/0008-5472.can-04-4130). 136. E. M. Conway, D. Collen, P. Carmeliet, Molecular mechanisms of blood vessel growth. Cardiovascular Research 49, 507-521 (2001); published online EpubFebruary 16, 2001 (10.1016/s0008-6363(00)00281-9). 137. N. Weidner, P. R. Carroll, J. Flax, W. Blumenfeld, J. Folkman, Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol 143, 401-409 (1993); published online EpubAug. 138. R. K. Vartanian, N. Weidner, Correlation of intratumoral endothelial cell proliferation with microvessel density (tumor angiogenesis) and tumor cell proliferation in breast carcinoma. Am J Pathol 144, 1188-1194 (1994); published online EpubJun. 139. N. Ferrara, Vascular endothelial growth factor. Trends in cardiovascular medicine 3, 244-250 (1993); published online EpubNov-Dec (10.1016/1050-1738(93)90046-9). 140. N. Ferrara, T. Davis-Smyth, The Biology of Vascular Endothelial Growth Factor. Endocrine Reviews 18, 4-25 (1997)doi:10.1210/edrv.18.1.0287). 141. D. I. Holmes, I. Zachary, The vascular endothelial growth factor (VEGF) family: angiogenic factors in health and disease. Genome Biol 6, 209 (2005)10.1186/gb-2005-6- 2-209). 142. A.-K. Olsson, A. Dimberg, J. Kreuger, L. Claesson-Welsh, VEGF receptor signalling ? in control of vascular function. Nat Rev Mol Cell Biol 7, 359-371 (2006); published online Epub05//print (http://www.nature.com/nrm/journal/v7/n5/suppinfo/nrm1911_S1.html). 143. H. Takahashi, M. Shibuya, The vascular endothelial growth factor (VEGF)/VEGF receptor system and its role under physiological and pathological conditions. Clin. Sci. 109, 227-241 (2005); published online EpubSep , 2005 (10.1042/cs20040370). 144. R. Roskoski Jr, VEGF receptor protein–tyrosine kinases: Structure and regulation. Biochem Bioph Res Co 375, 287-291 (2008); published online Epub10/24/ (http://dx.doi.org/10.1016/j.bbrc.2008.07.121). 145. K. A. Houck, N. Ferrara, J. Winer, G. Cachianes, B. Li, D. W. Leung, The vascular endothelial growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA. Mol Endocrinol 5, 1806-1814 (1991); published online EpubDec (10.1210/mend-5-12-1806). 146. K. A. Houck, D. W. Leung, A. M. Rowland, J. Winer, N. Ferrara, Dual regulation of vascular endothelial growth factor bioavailability by genetic and proteolytic mechanisms. 238 Journal of Biological Chemistry 267, 26031-26037 (1992); published online EpubDecember 25, 1992. 147. Z. Gluzman-Poltorak, T. Cohen, Y. Herzog, G. Neufeld, Neuropilin-2 and Neuropilin-1 Are Receptors for the 165-Amino Acid Form of Vascular Endothelial Growth Factor (VEGF) and of Placenta Growth Factor-2, but Only Neuropilin-2 Functions as a Receptor for the 145-Amino Acid Form of VEGF. Journal of Biological Chemistry 275, 18040- 18045 (2000); published online EpubJune 16, 2000 (10.1074/jbc.M909259199). 148. N. S. Levy, S. Chung, H. Furneaux, A. P. Levy, Hypoxic Stabilization of Vascular Endothelial Growth Factor mRNA by the RNA-binding Protein HuR. Journal of Biological Chemistry 273, 6417-6423 (1998); published online EpubMarch 13, 1998 (10.1074/jbc.273.11.6417). 149. N. Ferrara, K. J. Hillan, H.-P. Gerber, W. Novotny, Discovery and development of bevacizumab, an anti-VEGF antibody for treating cancer. Nat Rev Drug Discov 3, 391- 400 (2004); published online Epub05//print. 150. M. Inoue, J. H. Hager, N. Ferrara, H.-P. Gerber, D. Hanahan, VEGF-A has a critical, nonredundant role in angiogenic switching and pancreatic β cell carcinogenesis. Cancer Cell 1, 193-202 (2002); published online Epub3// (http://dx.doi.org/10.1016/S1535- 6108(02)00031-4). 151. A. Compagni, P. Wilgenbus, M.-A. Impagnatiello, M. Cotten, G. Christofori, Fibroblast Growth Factors Are Required for Efficient Tumor Angiogenesis. Cancer Res 60, 7163- 7169 (2000); published online EpubDecember 15, 2000. 152. G. Bergers, K. Javaherian, K.-M. Lo, J. Folkman, D. Hanahan, Effects of Angiogenesis Inhibitors on Multistage Carcinogenesis in Mice. Science 284, 808-812 (1999); published online EpubApril 30, 1999 (10.1126/science.284.5415.808). 153. G. Bergers, R. Brekken, G. McMahon, T. H. Vu, T. Itoh, K. Tamaki, K. Tanzawa, P. Thorpe, S. Itohara, Z. Werb, D. Hanahan, Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat Cell Biol 2, 737-744 (2000); published online Epub10//print. 154. S. Weis, J. Cui, L. Barnes, D. Cheresh, Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. The Journal of Cell Biology 167, 223-229 (2004); published online EpubOctober 25, 2004 (10.1083/jcb.200408130). 155. M. Schmitt, A. Horbach, R. Kubitz, A. Frilling, D. Häussinger, Disruption of hepatocellular tight junctions by vascular endothelial growth factor (VEGF): a novel mechanism for tumor invasion. Journal of Hepatology 41, 274-283 (2004); published online Epub8// (http://dx.doi.org/10.1016/j.jhep.2004.04.035). 156. S. Hiratsuka, K. Nakamura, S. Iwai, M. Murakami, T. Itoh, H. Kijima, J. M. Shipley, R. M. Senior, M. Shibuya, MMP9 induction by vascular endothelial growth factor receptor- 1 is involved in lung-specific metastasis. Cancer Cell 2, 289-300 (2002); published online Epub10// (http://dx.doi.org/10.1016/S1535-6108(02)00153-8). 157. H. Hurwitz, L. Fehrenbacher, W. Novotny, T. Cartwright, J. Hainsworth, W. Heim, J. Berlin, A. Baron, S. Griffing, E. Holmgren, N. Ferrara, G. Fyfe, B. Rogers, R. Ross, F. Kabbinavar, Bevacizumab plus Irinotecan, Fluorouracil, and Leucovorin for Metastatic Colorectal Cancer. New England Journal of Medicine 350, 2335-2342 (2004)doi:10.1056/NEJMoa032691). 239 158. C. H. Blood, B. R. Zetter, Tumor interactions with the vasculature: angiogenesis and tumor metastasis. Biochimica et biophysica acta 1032, 89-118 (1990); published online EpubJun 1. 159. F. Bladt, D. Riethmacher, S. Isenmann, A. Aguzzi, C. Birchmeier, Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 376, 768-771 (1995); published online Epub08/31/print. 160. T. Nakamura, S. Mizuno, K. Matsumoto, Y. Sawa, H. Matsuda, T. Nakamura, Myocardial protection from ischemia/reperfusion injury by endogenous and exogenous HGF. The Journal of clinical investigation 106, 1511-1519 (2000)10.1172/JCI10226). 161. G. K. Michalopoulos, M. C. DeFrances, Liver Regeneration. Science 276, 60-66 (1997); published online EpubApril 4, 1997 (10.1126/science.276.5309.60). 162. C. Boccaccio, P. M. Comoglio, Invasive growth: a MET-driven genetic programme for cancer and stem cells. Nat Rev Cancer 6, 637-645 (2006); published online Epub08//print. 163. T. Otsuka, H. Takayama, R. Sharp, G. Celli, W. J. LaRochelle, D. P. Bottaro, N. Ellmore, W. Vieira, J. W. Owens, M. Anver, G. Merlino, c-Met Autocrine Activation Induces Development of Malignant Melanoma and Acquisition of the Metastatic Phenotype. Cancer Res 58, 5157-5167 (1998); published online EpubNovember 15, 1998. 164. C. Birchmeier, W. Birchmeier, E. Gherardi, G. F. Vande Woude, Met, metastasis, motility and more. Nat Rev Mol Cell Biol 4, 915-925 (2003); published online Epub12//print (http://www.nature.com/nrm/journal/v4/n12/suppinfo/nrm1261_S1.html). 165. P. M. Comoglio, S. Giordano, L. Trusolino, Drug development of MET inhibitors: targeting oncogene addiction and expedience. Nat Rev Drug Discov 7, 504-516 (2008); published online EpubJun (10.1038/nrd2530). 166. V. Orian-Rousseau, L. Chen, J. P. Sleeman, P. Herrlich, H. Ponta, CD44 is required for two consecutive steps in HGF/c-Met signaling. Genes & Development 16, 3074-3086 (2002); published online EpubDecember 1, 2002 (10.1101/gad.242602). 167. L. Trusolino, A. Bertotti, P. M. Comoglio, A Signaling Adapter Function for α6β4 Integrin in the Control of HGF-Dependent Invasive Growth. Cell 107, 643-654 (2001); published online Epub11/30/ (http://dx.doi.org/10.1016/S0092-8674(01)00567-0). 168. S. Giordano, S. Corso, P. Conrotto, S. Artigiani, G. Gilestro, D. Barberis, L. Tamagnone, P. M. Comoglio, The Semaphorin 4D receptor controls invasive growth by coupling with Met. Nat Cell Biol 4, 720-724 (2002); published online Epub09//print (http://www.nature.com/ncb/journal/v4/n9/suppinfo/ncb843_S1.html). 169. S. Pennacchietti, P. Michieli, M. Galluzzo, M. Mazzone, S. Giordano, P. M. Comoglio, Hypoxia promotes invasive growth by transcriptional activation of the met protooncogene. Cancer Cell 3, 347-361 (2003); published online Epub4// (http://dx.doi.org/10.1016/S1535-6108(03)00085-0). 170. H. T. Jun, J. Sun, K. Rex, R. Radinsky, R. Kendall, A. Coxon, T. L. Burgess, AMG 102, A Fully Human Anti-Hepatocyte Growth Factor/Scatter Factor Neutralizing Antibody, Enhances the Efficacy of Temozolomide or Docetaxel in U-87 MG Cells and Xenografts. Clinical Cancer Research 13, 6735-6742 (2007); published online EpubNovember 15, 2007 (10.1158/1078-0432.ccr-06-2969). 171. T. Martens, N.-O. Schmidt, C. Eckerich, R. Fillbrandt, M. Merchant, R. Schwall, M. Westphal, K. Lamszus, A Novel One-Armed Anti-c-Met Antibody Inhibits Glioblastoma 240 Growth In vivo. Clinical Cancer Research 12, 6144-6152 (2006); published online EpubOctober 15, 2006 (10.1158/1078-0432.ccr-05-1418). 172. M. Sattler, Y. B. Pride, P. Ma, J. L. Gramlich, S. C. Chu, L. A. Quinnan, S. Shirazian, C. Liang, K. Podar, J. G. Christensen, R. Salgia, A Novel Small Molecule Met Inhibitor Induces Apoptosis in Cells Transformed by the Oncogenic TPR-MET Tyrosine Kinase. Cancer Res 63, 5462-5469 (2003); published online EpubSeptember 1, 2003. 173. H. Y. Zou, Q. Li, J. H. Lee, M. E. Arango, S. R. McDonnell, S. Yamazaki, T. B. Koudriakova, G. Alton, J. J. Cui, P.-P. Kung, M. D. Nambu, G. Los, S. L. Bender, B. Mroczkowski, J. G. Christensen, An Orally Available Small-Molecule Inhibitor of c-Met, PF-2341066, Exhibits Cytoreductive Antitumor Efficacy through Antiproliferative and Antiangiogenic Mechanisms. Cancer Res 67, 4408-4417 (2007); published online EpubMay 1, 2007 (10.1158/0008-5472.can-06-4443). 174. S. A. Krawetz, The origin of lysyl oxidase. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 108, 117-119 (1994); published online Epub5// (http://dx.doi.org/10.1016/0305-0491(94)90171-6). 175. H. M. Kagan, W. Li, Lysyl oxidase: Properties, specificity, and biological roles inside and outside of the cell. Journal of cellular biochemistry 88, 660-672 (2003)10.1002/jcb.10413). 176. S. Gao, J. Zhou, Y. Zhao, P. Toselli, W. Li, Hypoxia-Response Element (HRE)–Directed Transcriptional Regulation of the Rat Lysyl Oxidase Gene in Response to Cobalt and Cadmium. Toxicological Sciences 132, 379-389 (2013); published online EpubApril 1, 2013 (10.1093/toxsci/kfs327). 177. D. F. Higgins, K. Kimura, M. Iwano, V. H. Haase, Hypoxia-inducible factor signaling in the development of tissue fibrosis. Cell Cycle 7, 1128-1132 (2008); published online Epub05/01. 178. A. N. Santhanam, A. R. Baker, G. Hegamyer, D. A. Kirschmann, N. H. Colburn, Pdcd4 repression of lysyl oxidase inhibits hypoxia-induced breast cancer cell invasion. Oncogene 29, 3921-3932 (2010); published online Epub07/08/print (http://www.nature.com/onc/journal/v29/n27/suppinfo/onc2010158s1.html). 179. M. S. Kim, S.-S. Kim, S. T. Jung, J.-Y. Park, H.-W. Yoo, J. Ko, K. Csiszar, S.-Y. Choi, Y. Kim, Expression and Purification of Enzymatically Active Forms of the Human Lysyl Oxidase-like Protein 4. Journal of Biological Chemistry 278, 52071-52074 (2003); published online EpubDecember 26, 2003 (10.1074/jbc.M308856200). 180. B. López, A. González, N. Hermida, F. Valencia, E. de Teresa, J. Díez, Role of lysyl oxidase in myocardial fibrosis: from basic science to clinical aspects. (2010), vol. 299, pp. H1-H9. 181. T. G. Voloshenyuk, E. S. Landesman, E. Khoutorova, A. D. Hart, J. D. Gardner, Induction of cardiac fibroblast lysyl oxidase by TGF-β1 requires PI3K/Akt, Smad3, and MAPK signaling. Cytokine 55, 90-97 (2011); published online Epub7// (http://dx.doi.org/10.1016/j.cyto.2011.03.024). 182. O. Adam, K. Theobald, D. Lavall, M. Grube, H. K. Kroemer, S. Ameling, H.-J. Schäfers, M. Böhm, U. Laufs, Increased lysyl oxidase expression and collagen cross-linking during atrial fibrillation. Journal of Molecular and Cellular Cardiology 50, 678-685 (2011); published online Epub4// (http://dx.doi.org/10.1016/j.yjmcc.2010.12.019). 241 183. A.-M. Baker, T. R. Cox, D. Bird, G. Lang, G. I. Murray, X.-F. Sun, S. M. Southall, J. R. Wilson, J. T. Erler, The Role of Lysyl Oxidase in SRC-Dependent Proliferation and Metastasis of Colorectal Cancer. J Natl Cancer I 103, 407-424 (2011); published online EpubMarch 2, 2011 (10.1093/jnci/djq569). 184. L.-C. Chen, S.-H. Tu, C.-S. Huang, C.-S. Chen, C.-T. Ho, H.-W. Lin, C.-H. Lee, H.-W. Chang, C.-H. Chang, C.-H. Wu, W.-S. Lee, Y.-S. Ho, Human breast cancer cell metastasis is attenuated by lysyl oxidase inhibitors through down-regulation of focal adhesion kinase and the paxillin-signaling pathway. Breast Cancer Res Treat 134, 989- 1004 (2012); published online Epub2012/08/01 (10.1007/s10549-012-1986-8). 185. Q.-T. Le, J. Harris, A. M. Magliocco, C. S. Kong, R. Diaz, B. Shin, H. Cao, A. Trotti, J. T. Erler, C. H. Chung, A. Dicker, T. F. Pajak, A. J. Giaccia, K. K. Ang, Validation of Lysyl Oxidase As a Prognostic Marker for Metastasis and Survival in Head and Neck Squamous Cell Carcinoma: Radiation Therapy Oncology Group Trial 90-03. Journal of Clinical Oncology 27, 4281-4286 (2009); published online EpubSeptember 10, 2009 (10.1200/jco.2008.20.6003). 186. K. R. Levental, H. Yu, L. Kass, J. N. Lakins, M. Egeblad, J. T. Erler, S. F. T. Fong, K. Csiszar, A. Giaccia, W. Weninger, M. Yamauchi, D. L. Gasser, V. M. Weaver, Matrix Crosslinking Forces Tumor Progression by Enhancing Integrin Signaling. Cell 139, 891- 906 (2009); published online Epub11/25/ (http://dx.doi.org/10.1016/j.cell.2009.10.027). 187. K. J. Isselbacher, Sugar and Amino Acid Transport by Cells in Culture — Differences between Normal and Malignant Cells. New England Journal of Medicine 286, 929-933 (1972)doi:10.1056/NEJM197204272861707). 188. A. Scheepers, H. Joost, A. Schurmann, The glucose transporter families SGLT and GLUT: molecular basis of normal and aberrant function. Journal of Parenteral and Enteral Nutrition 28, 364-371 (2004); published online EpubSeptember 1, 2004 (10.1177/0148607104028005364). 189. F. Q. Zhao, A. F. Keating, Functional properties and genomics of glucose transporters. Current genomics 8, 113-128 (2007); published online EpubApr. 190. R. C. Hresko, M. Kruse, M. Strube, M. Mueckler, Topology of the Glut 1 glucose transporter deduced from glycosylation scanning mutagenesis. Journal of Biological Chemistry 269, 20482-20488 (1994); published online EpubAugust 12, 1994. 191. M. Mueckler, C. Caruso, S. Baldwin, M. Panico, I. Blench, H. Morris, W. Allard, G. Lienhard, H. Lodish, Sequence and structure of a human glucose transporter. Science 229, 941-945 (1985); published online EpubSeptember 6, 1985 (10.1126/science.3839598). 192. K. Takata, T. Kasahara, M. Kasahara, O. Ezaki, H. Hirano, Erythrocyte/HEPG2-type glucose transporter is concentrated in cells of blood-tissue barriers. Biochem Bioph Res Co 173, 67-73 (1990); published online Epub11/30/ (http://dx.doi.org/10.1016/S0006- 291X(05)81022-8). 193. O. Warburg, On the origin of cancer cells. Science 123, 309-314 (1956); published online EpubFeb 24. 194. R. Diaz-Ruiz, M. Rigoulet, A. Devin, The Warburg and Crabtree effects: On the origin of cancer cell energy metabolism and of yeast glucose repression. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1807, 568-576 (2011); published online Epub6// (http://dx.doi.org/10.1016/j.bbabio.2010.08.010). 242 195. A. Godoy, K. Salazar, C. Figueroa, G. J. Smith, M. de los Angeles Garcia, F. J. Nualart, Nutritional channels in breast cancer. Journal of Cellular and Molecular Medicine 13, 3973-3984 (2009)10.1111/j.1582-4934.2008.00544.x). 196. M. Younes, R. W. Brown, M. Stephenson, M. Gondo, P. T. Cagle, Overexpression of glut1 and glut3 in stage I nonsmall cell lung carcinoma is Associated with poor survival. Cancer 80, 1046-1051 (1997)10.1002/(SICI)1097-0142(19970915)80:6<1046::AID- CNCR6>3.0.CO;2-7). 197. R. S. Haber, A. Rathan, K. R. Weiser, A. Pritsker, S. H. Itzkowitz, C. Bodian, G. Slater, A. Weiss, D. E. Burstein, GLUT1 glucose transporter expression in colorectal carcinoma. Cancer 83, 34-40 (1998)10.1002/(SICI)1097-0142(19980701)83:1<34::AID- CNCR5>3.0.CO;2-E). 198. M. Younes, L. V. Lechago, J. Lechago, Overexpression of the human erythrocyte glucose transporter occurs as a late event in human colorectal carcinogenesis and is associated with an increased incidence of lymph node metastases. Clinical Cancer Research 2, 1151-1154 (1996); published online EpubJuly 1, 1996. 199. F. Hernández, M. Navarro, J. L. Encinas, J. C. López Gutiérrez, M. López Santamaría, N. Leal, L. Martínez, M. Patrón, J. A. Tovar, The role of GLUT1 immunostaining in the diagnosis and classification of liver vascular tumors in children. Journal of pediatric surgery 40, 801-804 (2005); published online Epub5// (http://dx.doi.org/10.1016/j.jpedsurg.2005.01.046). 200. N. W. Merrall, R. Plevin, G. W. Gould, Growth factors, mitogens, oncogenes and the regulation of glucose transport. Cell Signal 5, 667-675 (1993); published online Epub11// (http://dx.doi.org/10.1016/0898-6568(93)90028-K). 201. A. Behrooz, F. Ismail-Beigi, Dual Control of glut1 Glucose Transporter Gene Expression by Hypoxia and by Inhibition of Oxidative Phosphorylation. Journal of Biological Chemistry 272, 5555-5562 (1997); published online EpubFebruary 28, 1997. 202. M. Hayashi, M. Sakata, T. Takeda, T. Yamamoto, Y. Okamoto, K. Sawada, A. Kimura, R. Minekawa, M. Tahara, K. Tasaka, Y. Murata, Induction of glucose transporter 1 expression through hypoxia-inducible factor 1α under hypoxic conditions in trophoblast- derived cells. Journal of Endocrinology 183, 145-154 (2004); published online EpubOctober 1, 2004 (10.1677/joe.1.05599). 203. J. S. Flier, M. M. Mueckler, P. Usher, H. F. Lodish, Elevated Levels of Glucose Transport and Transporter Messenger RNA are Induced by ras or src Oncogenes. Science 235, 1492-1495 (1987)10.2307/1699050). 204. C. Chen, N. Pore, A. Behrooz, F. Ismail-Beigi, A. Maity, Regulation of glut1 mRNA by Hypoxia-inducible Factor-1: INTERACTION BETWEEN H-ras AND HYPOXIA. Journal of Biological Chemistry 276, 9519-9525 (2001); published online EpubMarch 23, 2001. 205. S. Ito, T. Fukusato, T. Nemoto, H. Sekihara, Y. Seyama, S. Kubota, Coexpression of Glucose Transporter 1 and Matrix Metalloproteinase-2 in Human Cancers. J Natl Cancer I 94, 1080-1091 (2002); published online EpubJuly 17, 2002 (10.1093/jnci/94.14.1080). 206. A. R. Nelson, B. Fingleton, M. L. Rothenberg, L. M. Matrisian, Matrix Metalloproteinases: Biologic Activity and Clinical Implications. Journal of Clinical Oncology 18, 1135 (2000); published online EpubMarch 1, 2000. 243 207. M. Kucia, K. Jankowski, R. Reca, M. Wysoczynski, L. Bandura, D. Allendorf, J. Zhang, J. Ratajczak, M. Ratajczak, CXCR4–SDF-1 Signalling, Locomotion, Chemotaxis and Adhesion. Histochem J 35, 233-245 (2004); published online Epub2004/03/01 (10.1023/B:HIJO.0000032355.66152.b8). 208. A. F. Schier, Chemokine Signaling: Rules of Attraction. Current Biology 13, R192-R194 (2003); published online Epub3/4/ (http://dx.doi.org/10.1016/S0960-9822(03)00122-2). 209. J. Libura, J. Drukala, M. Majka, O. Tomescu, J. M. Navenot, M. Kucia, L. Marquez, S. C. Peiper, F. G. Barr, A. Janowska-Wieczorek, M. Z. Ratajczak, CXCR4–SDF-1 signaling is active in rhabdomyosarcoma cells and regulates locomotion, chemotaxis, and adhesion. (2002), vol. 100, pp. 2597-2606. 210. A. BeachyPhilip, S. KarhadkarSunil, M. BermanDavid, Tissue repair and stem cell renewal in carcinogenesis. Nature 432, 324-331 (2004); published online Epub11/18/print. 211. M. Z. Ratajczak, E. Zuba-Surma, M. Kucia, R. Reca, W. Wojakowski, J. Ratajczak, The pleiotropic effects of the SDF-1-CXCR4 axis in organogenesis, regeneration and tumorigenesis. Leukemia 20, 1915-1924 (2006); published online Epub08/10/online ( 212. B. A. Teicher, S. P. Fricker, CXCL12 (SDF-1)/CXCR4 Pathway in Cancer. Clinical Cancer Research 16, 2927-2931 (2010); published online EpubJune 1, 2010 (10.1158/1078-0432.ccr-09-2329). 213. T. Schioppa, B. Uranchimeg, A. Saccani, S. K. Biswas, A. Doni, A. Rapisarda, S. Bernasconi, S. Saccani, M. Nebuloni, L. Vago, A. Mantovani, G. Melillo, A. Sica, Regulation of the Chemokine Receptor CXCR4 by Hypoxia. The Journal of Experimental Medicine 198, 1391-1402 (2003); published online EpubNovember 3, 2003 (10.1084/jem.20030267). 214. D. J. Ceradini, A. R. Kulkarni, M. J. Callaghan, O. M. Tepper, N. Bastidas, M. E. Kleinman, J. M. Capla, R. D. Galiano, J. P. Levine, G. C. Gurtner, Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med 10, 858-864 (2004); published online Epub08//print (http://www.nature.com/nm/journal/v10/n8/suppinfo/nm1075_S1.html). 215. O. Trédan, C. M. Galmarini, K. Patel, I. F. Tannock, Drug Resistance and the Solid Tumor Microenvironment. J Natl Cancer I 99, 1441-1454 (2007); published online EpubOctober 3, 2007 (10.1093/jnci/djm135). 216. B. J. Murphy, K. R. Laderoute, R. J. Chin, R. M. Sutherland, Metallothionein IIA Is Up- Regulated by Hypoxia in Human A431 Squamous Carcinoma Cells. Cancer Res 54, 5808-5810 (1994); published online EpubNovember 15, 1994. 217. G. C. Rice, C. Hoy, R. T. Schimke, Transient hypoxia enhances the frequency of dihydrofolate reductase gene amplification in Chinese hamster ovary cells. Proceedings of the National Academy of Sciences of the United States of America 83, 5978-5982 (1986); published online EpubAugust 1, 1986. 218. I. Tannock, Cell kinetics and chemotherapy: a critical review. Cancer treatment reports 62, 1117-1133 (1978); published online EpubAug. 219. L. R.-t. M.S, E. J. Hall, The effect of bleomycin on aerated and hypoxic cells in vitro, in combination with irradiation. International Journal of Radiation Oncology*Biology*Physics 5, 1491-1494 (1979); published online Epub9// (http://dx.doi.org/10.1016/0360-3016(79)90756-9). 244 220. P. Vaupel, O. Thews, M. Hoeckel, Treatment resistance of solid tumors. Med Oncol 18, 243-259 (2001); published online Epub2001/12/01 (10.1385/MO:18:4:243). 221. M. K. Bae, S. H. Kim, J. W. Jeong, Y. M. Lee, H. S. Kim, S. R. Kim, I. Yun, S. K. Bae, K. W. Kim, Curcumin inhibits hypoxia-induced angiogenesis via down-regulation of HIF-1. Oncology reports 15, 1557-1562 (2006); published online EpubJun. 222. H. Choi, Y.-S. Chun, S.-W. Kim, M.-S. Kim, J.-W. Park, Curcumin Inhibits Hypoxia- Inducible Factor-1 by Degrading Aryl Hydrocarbon Receptor Nuclear Translocator: A Mechanism of Tumor Growth Inhibition. Molecular Pharmacology 70, 1664-1671 (2006); published online EpubNovember 1, 2006 (10.1124/mol.106.025817). 223. A. C. Kasper, E. J. Moon, X. Hu, Y. Park, C. M. Wooten, H. Kim, W. Yang, M. W. Dewhirst, J. Hong, Analysis of HIF-1 inhibition by manassantin A and analogues with modified tetrahydrofuran configurations. Bioorganic & Medicinal Chemistry Letters 19, 3783-3786 (2009); published online Epub7/15/ (http://dx.doi.org/10.1016/j.bmcl.2009.04.071). 224. C. F. Hossain, Y.-P. Kim, S. R. Baerson, L. Zhang, R. K. Bruick, K. A. Mohammed, A. K. Agarwal, D. G. Nagle, Y.-D. Zhou, Saururus cernuus lignans—Potent small molecule inhibitors of hypoxia-inducible factor-1. Biochem Bioph Res Co 333, 1026-1033 (2005); published online Epub8/5/ (http://dx.doi.org/10.1016/j.bbrc.2005.05.191). 225. N. T. Dat, X. Jin, K. Lee, Y.-S. Hong, Y. H. Kim, J. J. Lee, Hypoxia-Inducible Factor-1 Inhibitory Benzofurans and Chalcone-Derived Diels−Alder Adducts from Morus Species. Journal of Natural Products 72, 39-43 (2008); published online Epub2009/01/23 (10.1021/np800491u). 226. Y. Xia, Y. Jin, N. Kaur, Y. Choi, K. Lee, HIF-1α inhibitors: Synthesis and biological evaluation of novel moracin O and P analogues. European Journal of Medicinal Chemistry 46, 2386-2396 (2011); published online Epub6// (http://dx.doi.org/10.1016/j.ejmech.2011.03.022). 227. A. L. Kung, S. D. Zabludoff, D. S. France, S. J. Freedman, E. A. Tanner, A. Vieira, S. Cornell-Kennon, J. Lee, B. Wang, J. Wang, K. Memmert, H. U. Naegeli, F. Petersen, M. J. Eck, K. W. Bair, A. W. Wood, D. M. Livingston, Small molecule blockade of transcriptional coactivation of the hypoxia-inducible factor pathway. Cancer Cell 6, 33- 43 (2004); published online EpubJul. 228. Y. Mie Lee, S.-H. Kim, H.-S. Kim, M. Jin Son, H. Nakajima, H. Jeong Kwon, K.-W. Kim, Inhibition of hypoxia-induced angiogenesis by FK228, a specific histone deacetylase inhibitor, via suppression of HIF-1α activity. Biochem Bioph Res Co 300, 241-246 (2003); published online Epub1/3/ (http://dx.doi.org/10.1016/S0006- 291X(02)02787-0). 229. F. N. Ko, C. C. Wu, S. C. Kuo, F. Y. Lee, C. M. Teng, YC-1, a novel activator of platelet guanylate cyclase. (1994), vol. 84, pp. 4226-4233. 230. Y.-S. Chun, E.-J. Yeo, E. Choi, C.-M. Teng, J.-M. Bae, M.-S. Kim, J.-W. Park, Inhibitory effect of YC-1 on the hypoxic induction of erythropoietin and vascular endothelial growth factor in Hep3B cells. Biochemical Pharmacology 61, 947-954 (2001); published online Epub4/15/ (http://dx.doi.org/10.1016/S0006-2952(01)00564-0). 231. S. H. Li, D. H. Shin, Y.-S. Chun, M. K. Lee, M.-S. Kim, J.-W. Park, A novel mode of action of YC-1 in HIF inhibition: stimulation of FIH-dependent p300 dissociation from HIF-1α. Molecular Cancer Therapeutics 7, 3729-3738 (2008); published online EpubDecember 1, 2008 (10.1158/1535-7163.mct-08-0074). 245 232. B.-H. Jiang, G. Jiang, J. Z. Zheng, Z. Lu, T. Hunter, P. K. Vogt, Phosphatidylinositol 3- Kinase Signaling Controls Levels of Hypoxia-inducible Factor 1. Cell Growth Differ 12, 363-369 (2001); published online EpubJuly 1, 2001. 233. P. K. Majumder, P. G. Febbo, R. Bikoff, R. Berger, Q. Xue, L. M. McMahon, J. Manola, J. Brugarolas, T. J. McDonnell, T. R. Golub, M. Loda, H. A. Lane, W. R. Sellers, mTOR inhibition reverses Akt-dependent prostate intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways. Nat Med 10, 594-601 (2004); published online Epub06//print (http://www.nature.com/nm/journal/v10/n6/suppinfo/nm1052_S1.html). 234. X. Wan, N. Shen, A. Mendoza, C. Khanna, L. J. Helman, CCI-779 inhibits rhabdomyosarcoma xenograft growth by an antiangiogenic mechanism linked to the targeting of mTOR/Hif-1alpha/VEGF signaling. Neoplasia 8, 394-401 (2006); published online EpubMay (10.1593/neo.05820). 235. N. J. Mabjeesh, D. E. Post, M. T. Willard, B. Kaur, E. G. Van Meir, J. W. Simons, H. Zhong, Geldanamycin Induces Degradation of Hypoxia-inducible Factor 1α Protein via the Proteosome Pathway in Prostate Cancer Cells. Cancer Res 62, 2478-2482 (2002); published online EpubMay 1, 2002. 236. A. Rapisarda, B. Uranchimeg, O. Sordet, Y. Pommier, R. H. Shoemaker, G. Melillo, Topoisomerase I-Mediated Inhibition of Hypoxia-Inducible Factor 1: Mechanism and Therapeutic Implications. Cancer Res 64, 1475-1482 (2004); published online EpubFebruary 15, 2004 (10.1158/0008-5472.can-03-3139). 237. A. Ardizzoni, Topotecan in the Treatment of Recurrent Small Cell Lung Cancer: An Update. The Oncologist 9, 4-13 (2004); published online EpubDecember 1, 2004 (10.1634/theoncologist.9-90006-4). 238. D. K. Armstrong, Topotecan Dosing Guidelines in Ovarian Cancer: Reduction and Management of Hematologic Toxicity. The Oncologist 9, 33-42 (2004); published online EpubFebruary 1, 2004 (10.1634/theoncologist.9-1-33). 239. H. Minegishi, S. Fukashiro, H. S. Ban, H. Nakamura, Discovery of Indenopyrazoles as a New Class of Hypoxia Inducible Factor (HIF)-1 Inhibitors. ACS Medicinal Chemistry Letters 4, 297-301 (2013); published online Epub2013/02/14 (10.1021/ml3004632). 240. B. Z. Olenyuk, G. J. Zhang, J. M. Klco, N. G. Nickols, W. G. Kaelin, Jr., P. B. Dervan, Inhibition of vascular endothelial growth factor with a sequence-specific hypoxia response element antagonist. Proc. Natl. Acad. Sci. U. S. A. 101, 16768-16773 (2004); published online EpubNov 30. 241. A. L. Kung, S. Wang, J. M. Klco, W. G. Kaelin, D. M. Livingston, Suppression of tumor growth through disruption of hypoxia-inducible transcription. Nat Med 6, 1335-1340 (2000); published online Epub12//print. 242. K. M. Block, H. Wang, L. Z. Szabo, N. W. Polaske, L. K. Henchey, R. Dubey, S. Kushal, C. F. Laszlo, J. Makhoul, Z. Song, E. J. Meuillet, B. Z. Olenyuk, Direct Inhibition of Hypoxia-Inducible Transcription Factor Complex with Designed Dimeric Epidithiodiketopiperazine. J. Am. Chem. Soc. 131, 18078-18088 (2009)10.1021/ja807601b). 243. R. Dubey, M. D. Levin, L. Z. Szabo, C. F. Laszlo, S. Kushal, J. B. Singh, P. Oh, J. E. Schnitzer, B. Z. Olenyuk, Suppression of tumor growth by designed dimeric epidithiodiketopiperazine targeting hypoxia-inducible transcription factor complex. J. Am. Chem. Soc. 135, 4537-4549 (2013); published online Epub2013/03/20 (10.1021/ja400805b). 246 244. K. M. Cook, S. T. Hilton, J. Mecinović, W. B. Motherwell, W. D. Figg, C. J. Schofield, Epidithiodiketopiperazines Block the Interaction between Hypoxia-inducible Factor-1α (HIF-1α) and p300 by a Zinc Ejection Mechanism. Journal of Biological Chemistry 284, 26831-26838 (2009); published online EpubSeptember 25, 2009 (10.1074/jbc.M109.009498). 245. B. P. Orner, J. T. Ernst, A. D. Hamilton, Toward proteomimetics: Terphenyl derivatives as structural and functional mimics of extended regions of an alpha-helix. Journal of the American Chemical Society 123, 5382-5383 (2001); published online EpubJun 6 (Doi 10.1021/Ja0025548). 246. T. Clackson, J. Wells, A hot spot of binding energy in a hormone-receptor interface. Science 267, 383-386 (1995); published online EpubJanuary 20, 1995 (10.1126/science.7529940). 247. A. A. Bogan, K. S. Thorn, Anatomy of hot spots in protein interfaces. Journal of molecular biology 280, 1-9 (1998); published online Epub7/3/ (http://dx.doi.org/10.1006/jmbi.1998.1843). 248. S. Kushal, B. B. Lao, L. K. Henchey, R. Dubey, H. Mesallati, N. J. Traaseth, B. Z. Olenyuk, P. S. Arora, Protein domain mimetics as in vivo modulators of hypoxia- inducible factor signaling. Proceedings of the National Academy of Sciences of the United States of America 110, 15602-15607 (2013); published online EpubSeptember 9, 2013 (10.1073/pnas.1312473110). 249. R. K. Jain, Determinants of Tumor Blood Flow: A Review. Cancer Res 48, 2641-2658 (1988); published online EpubMay 15, 1988. 250. P. Carmeliet, R. K. Jain, Angiogenesis in cancer and other diseases. Nature 407, 249-257 (2000); published online Epub09/14/print. 251. H. Hashizume, P. Baluk, S. Morikawa, J. W. McLean, G. Thurston, S. Roberge, R. K. Jain, D. M. McDonald, Openings between Defective Endothelial Cells Explain Tumor Vessel Leakiness. The American Journal of Pathology 156, 1363-1380 (2000); published online Epub4// (http://dx.doi.org/10.1016/S0002-9440(10)65006-7). 252. H. W. Chalkley, Method for the quantitative morphologic analysis of tissues. J Natl Cancer I 4, 47-53 (1943); published online EpubAug-Jun. 253. S. Hansen, D. A. Grabau, F. B. Sorensen, M. Bak, W. Vach, C. Rose, The prognostic value of angiogenesis by Chalkley counting in a confirmatory study design on 836 breast cancer patients. Clinical cancer research : an official journal of the American Association for Cancer Research 6, 139-146 (2000); published online EpubJan. 254. Z. Y. Li, S. H. Ni, X. L. Yang, N. Kiviat, A. Lieber, Xenograft models for liver metastasis: Relationship between tumor morphology and adenovirus vector transduction. Mol Ther 9, 650-657 (2004); published online EpubMay (DOI 10.1016/j.ymthe.2004.01.021). 255. D. Visigalli, D. Palmieri, A. Strangio, S. Astigiano, O. Barbieri, G. Casartelli, A. Zicca, P. Manduca, The carboxyl terminal trimer of procollagen I induces pro-metastatic changes and vascularization in breast cancer cells xenografts. Bmc Cancer 9, (2009); published online EpubFeb 18 (Artn 59 Doi 10.1186/1471-2407-9-59). 256. D. Tilkorn, A. Daigeler, I. Stricker, A. Schaffran, I. Schmitz, L. Steinstraesser, J. Hauser, A. Ring, H. U. Steinau, S. Al-Benna, Establishing efficient xenograft models with 247 intrinsic vascularisation for growing primary human low-grade sarcomas. Anticancer research 31, 4061-4066 (2011); published online EpubDec. 257. D. B. C Heneweer, D Emme, C Graeff, M Both, CC Glueer, H Kalthoff, M Heller, Comparison of tumour vascularisation by small animal ultrasound in a subcutaneous and an orthotopic xenotransplantation model in mice. Fortschr Röntgenstr 180 - A1_1, (2008)10.1055/s-0028-1085907). 258. T. Cascone, M. H. Herynk, L. Xu, Z. Du, H. Kadara, M. B. Nilsson, C. J. Oborn, Y. Y. Park, B. Erez, J. J. Jacoby, J. S. Lee, H. Y. Lin, F. Ciardiello, R. S. Herbst, R. R. Langley, J. V. Heymach, Up-regulated stromal EGFR and vascular remodeling in mouse xenograft models of angiogenesis inhibitor-resistant human lung adenocarcinoma. The Journal of clinical investigation 121, 1313-1328 (2011); published online EpubApr (10.1172/JCI42405). 259. J. L. Zhang, Z. Wang, W. Hu, S. S. Chen, X. E. Lou, H. J. Zhou, DHA regulates angiogenesis and improves the efficiency of CDDP for the treatment of lung carcinoma. Microvasc Res 87, 14-24 (2013); published online EpubMay (DOI 10.1016/j.mvr.2013.02.006). 260. X. Y. Sun, J. R. Kanwar, E. Leung, K. Lehnert, D. M. Wang, G. W. Krissansen, Angiostatin enhances B7.1-mediated cancer immunotherapy independently of effects on vascular endothelial growth factor expression. Cancer Gene Ther 8, 719-727 (2001); published online EpubOct (DOI 10.1038/sj.cgt.7700370). 261. C. A. Kristensen, L. M. Hamberg, G. J. Hunter, S. Roberge, D. Kierstead, G. L. Wolf, R. K. Jain, Changes in vascularization of human breast cancer xenografts responding to antiestrogen therapy. Neoplasia 1, 518-525 (1999); published online EpubDec. 262. L. D. Nagaprashantha, R. Vatsyayan, J. Singhal, P. Lelsani, L. Prokai, S. Awasthi, S. S. Singhal, 2'-Hydroxyflavanone inhibits proliferation, tumor vascularization and promotes normal differentiation in VHL-mutant renal cell carcinoma. Carcinogenesis 32, 568-575 (2011); published online EpubApr (DOI 10.1093/carcin/bgr021). 263. M. P. Steinkamp, K. K. Winner, S. Davies, C. Muller, Y. Zhang, R. M. Hoffman, A. Shirinifard, M. Moses, Y. Jiang, B. S. Wilson, Ovarian tumor attachment, invasion, and vascularization reflect unique microenvironments in the peritoneum: insights from xenograft and mathematical models. Frontiers in oncology 3, 97 (2013)10.3389/fonc.2013.00097). 264. K. W. McCrudden, B. Hopkins, J. Frischer, A. Novikov, J. Huang, A. Kadenhe, T. New, A. Yokoi, D. J. Yamashiro, J. J. Kandel, W. Middlesworth, Anti-VEGF antibody in experimental hepatoblastoma: suppression of tumor growth and altered angiogenesis. Journal of pediatric surgery 38, 308-314; discussion 308-314 (2003); published online EpubMar (10.1053/jpsu.2003.50099). 265. J. A. Wells, C. L. McClendon, Reaching for high-hanging fruit in drug discovery at protein-protein interfaces. Nature 450, 1001-1009 (2007); published online EpubDEC 13. 266. A. P. Higueruelo, A. Schreyer, G. R. Bickerton, W. R. Pitt, C. R. Groom, T. L. Blundell, Atomic interactions and profile of small molecules disrupting protein-protein interfaces: the TIMBAL database. Chem. Biol. Drug. Des. 74, 457-467 (2009); published online EpubNov (JPP889 [pii] 10.1111/j.1747-0285.2009.00889.x). 267. B. N. Bullock, A. L. Jochim, P. S. Arora, Assessing helical protein interfaces for inhibitor design. J. Am. Chem. Soc. 133, 14220-14223 (2011); published online EpubSep 14 (10.1021/ja206074j). 248 268. A. L. Jochim, P. S. Arora, Systematic analysis of helical protein interfaces reveals targets for synthetic inhibitors. ACS Chem. Biol. 5, 919-923 (2010); published online EpubOct (Doi 10.1021/Cb1001747). 269. A. D. Thompson, A. Dugan, J. E. Gestwicki, A. K. Mapp, Fine-Tuning Multiprotein Complexes Using Small Molecules. ACS Chemical Biology 7, 1311-1320 (2012); published online Epub2012/08/17 (10.1021/cb300255p). 270. A. A. Bogan, K. S. Thorn, Anatomy of hot spots in protein interfaces. J. Mol. Biol. 280, 1-9 (1998); published online EpubJul 3. 271. T. Clackson, J. A. Wells, A Hot-Spot of Binding-Energy in a Hormone-Receptor Interface. Science 267, 383-386 (1995); published online EpubJAN 20. 272. V. Azzarito, K. Long, N. S. Murphy, A. J. Wilson, Inhibition of alpha-helix-mediated protein-protein interactions using designed molecules. Nat. Chem. 5, 161-173 (2013). 273. C. G. Cummings, A. D. Hamilton, Disrupting protein-protein interactions with non- peptidic, small molecule alpha-helix mimetics. Curr. Opin. Chem. Biol. 14, 341-346 (2010); published online EpubJun (S1367-5931(10)00053-0 [pii] 10.1016/j.cbpa.2010.04.001). 274. E. Ko, J. Liu, K. Burgess, Minimalist and universal peptidomimetics. Chemical Society Reviews 40, 4411-4421 (2011). 275. M. K. P. Jayatunga, S. Thompson, A. D. Hamilton, α-Helix mimetics: Outwards and upwards. Bioorganic & Medicinal Chemistry Letters 24, 717-724 (2014); published online Epub2/1/ (http://dx.doi.org/10.1016/j.bmcl.2013.12.003). 276. P. Fraisl, J. Aragones, P. Carmeliet, Inhibition of oxygen sensors as a therapeutic strategy for ischaemic and inflammatory disease. Nat Rev Drug Discov 8, 139-152 (2009); published online EpubFeb (Doi 10.1038/Nrd2761). 277. S. J. Freedman, Z. Y. Sun, F. Poy, A. L. Kung, D. M. Livingston, G. Wagner, M. J. Eck, Structural basis for recruitment of CBP/p300 by hypoxia-inducible factor-1 alpha. Proc. Natl. Acad. Sci. U. S. A. 99, 5367-5372 (2002); published online EpubApr 16. 278. S. A. Dames, M. Martinez-Yamout, R. N. De Guzman, H. J. Dyson, P. E. Wright, Structural basis for HIF-1alpha/CBP recognition in the cellular hypoxic response. Proc. Natl. Acad. Sci. U. S. A. 99, 5271-5276 (2002); published online EpubApr 16. 279. K. Hirota, G. L. Semenza, Regulation of angiogenesis by hypoxia-inducible factor 1. Crit. Rev. Oncol./Hematol. 59, 15-26 (2006); published online EpubJul. 280. G. L. Semenza, Hypoxia-inducible factors: mediators of cancer progression and targets for cancer therapy. Trends in Pharmacological Sciences 33, 207-214 (2012); published online Epub4// (http://dx.doi.org/10.1016/j.tips.2012.01.005). 281. N. G. Nickols, C. S. Jacobs, M. E. Farkas, P. B. Dervan, Modulating hypoxia-inducible transcription by disrupting the HIF-1-DNA interface. ACS Chem. Biol. 2, 561-571 (2007)10.1021/cb700110z). 282. Y. Xia, H.-K. Choi, K. Lee, Recent advances in hypoxia-inducible factor (HIF)-1 inhibitors. European Journal of Medicinal Chemistry 49, 24-40 (2012); published online Epub3// (http://dx.doi.org/10.1016/j.ejmech.2012.01.033). 283. L. K. Henchey, S. Kushal, R. Dubey, R. N. Chapman, B. Z. Olenyuk, P. S. Arora, Inhibition of hypoxia inducible factor 1–transcription coactivator interaction by a hydrogen bond surrogate alpha-helix. J. Am. Chem. Soc. 132, 941-943 (2010). 249 284. T. H. Scheuermann, Q. Li, H.-W. Ma, J. Key, L. Zhang, R. Chen, J. A. Garcia, J. Naidoo, J. Longgood, D. E. Frantz, U. K. Tambar, K. H. Gardner, R. K. Bruick, Allosteric inhibition of hypoxia inducible factor-2 with small molecules. Nat Chem Biol 9, 271-276 (2013); published online Epub04//print (http://www.nature.com/nchembio/journal/v9/n4/abs/nchembio.1185.html#supplementar y-information). 285. E. Miranda, I. K. Nordgren, A. L. Male, C. E. Lawrence, F. Hoakwie, F. Cuda, W. Court, K. R. Fox, P. A. Townsend, G. K. Packham, S. A. Eccles, A. Tavassoli, A Cyclic Peptide Inhibitor of HIF-1 Heterodimerization That Inhibits Hypoxia Signaling in Cancer Cells. Journal of the American Chemical Society 135, 10418-10425 (2013); published online Epub2013/07/17 (10.1021/ja402993u). 286. J. R. Perkins, I. Diboun, B. H. Dessailly, J. G. Lees, C. Orengo, Transient Protein-Protein Interactions: Structural, Functional, and Network Properties. Structure 18, 1233-1243 (2010). 287. L. W. Lee, A. K. Mapp, Transcriptional switches: chemical approaches to gene regulation. J. Biol. Chem. 285, 11033-11038 (2010); published online EpubApr 9 (R109.075044 [pii] 10.1074/jbc.R109.075044). 288. A. N. Koehler, A complex task? Direct modulation of transcription factors with small molecules. Curr. Opin. Chem. Biol. 14, 331-340 (2010); published online EpubJun. 289. P. Ravindranathan, T. K. Lee, L. Yang, M. M. Centenera, L. Butler, W. D. Tilley, J. T. Hsieh, J. M. Ahn, G. V. Raj, Peptidomimetic targeting of critical androgen receptor- coregulator interactions in prostate cancer. Nat Commun 4, 1923 (2013); published online EpubMay 28 (10.1038/ncomms2912 [pii]). 290. A. Shaginian, L. R. Whitby, S. Hong, I. Hwang, B. Farooqi, M. Searcey, J. Chen, P. K. Vogt, D. L. Boger, Design, Synthesis, and Evaluation of an alpha-Helix Mimetic Library Targeting Protein-Protein Interactions. J. Am. Chem. Soc. 131, 5564-5572 (2009); published online EpubApr 22 (10.1021/ja810025g). 291. P. Tosovska, P. S. Arora, Oligooxopiperazines as Nonpeptidic Alpha-Helix Mimetics. Org. Lett. 12, 1588-1591 (2010)10.1021/ol1003143). 292. A. A. Patchett, R. P. Nargund, Privileged structures - An update. Annu Rep Med Chem 35, 289-298 (2000). 293. T. Kortemme, D. E. Kim, D. Baker, Computational alanine scanning of protein-protein interfaces. Sci. STKE 2004, pl2 (2004); published online EpubFeb 10. 294. R. Dial, Z. Y. Sun, S. J. Freedman, Three conformational states of the p300 CH1 domain define its functional properties. Biochemistry 42, 9937-9945 (2003); published online EpubAug 26 (10.1021/bi034989o). 295. R. N. De Guzman, J. M. Wojciak, M. A. Martinez-Yamout, H. J. Dyson, P. E. Wright, CBP/p300 TAZ1 domain forms a structured scaffold for ligand binding. Biochemistry 44, 490-497 (2005); published online EpubJan 18 (10.1021/bi048161t). 296. T. Shibata, N. Akiyama, M. Noda, K. Sasai, M. Hiraoka, Enhancement of gene expression under hypoxic conditions using fragments of the human vascular endothelial growth factor and the erythropoietin genes. Int. J. Radiat. Oncol. Biol. Phys. 42, 913-916 (1998). 297. Y. Fang, R. Sullivan, C. H. Graham, Confluence-dependent resistance to doxorubicin in human MDA-MB-231 breast carcinoma cells requires hypoxia-inducible factor-1 250 activity. Experimental Cell Research 313, 867-877 (2007); published online EpubMar (10.1016/j.yexcr.2006.12.004). 298. H. E. Ryan, J. Lo, R. S. Johnson, HIF‐1α is required for solid tumor formation and embryonic vascularization. The EMBO Journal 17, 3005-3015 (1998); published online Epub1998-06-01 00:00:00 (10.1093/emboj/17.11.3005). 299. G. L. Semenza, Cancer-stromal cell interactions mediated by hypoxia-inducible factors promote angiogenesis, lymphangiogenesis, and metastasis. Oncogene 32, 4057-4063 (2013); published online EpubAug 29 (10.1038/onc.2012.578). 300. M. Tian, H. Zhang, Y. Nakasone, K. Mogi, K. Endo, Expression of Glut-1 and Glut-3 in untreated oral squamous cell carcinoma compared with FDG accumulation in a PET study. Eur J Nucl Med Mol Imaging 31, 5-12 (2004); published online Epub2004/01/01 (10.1007/s00259-003-1316-9). 301. J. H. Wang, J. C. Wang, Y. X. Sun, W. Y. Song, J. E. Nor, C. Y. Wang, R. S. Taichman, Diverse signaling pathways through the SDF-1/CXCR4 chemokine axis in prostate cancer cell lines leads to altered patterns of cytokine secretion and angiogenesis. Cell Signal 17, 1578-1592 (2005); published online EpubDec (DOI 10.1016/j.cellsig.2005.03.022). 302. N. Vo, R. H. Goodman, CREB-binding protein and p300 in transcriptional regulation. J. Biol. Chem. 276, 13505-13508 (2001); published online EpubApr 27. 303. D. Hanahan, R. A. Weinberg, Hallmarks of cancer: the next generation. Cell 144, 646- 674 (2011); published online EpubMar 4 (10.1016/j.cell.2011.02.013). 304. Y. Benita, H. Kikuchi, A. D. Smith, M. Q. Zhang, D. C. Chung, R. J. Xavier, An integrative genomics approach identifies Hypoxia Inducible Factor-1 (HIF-1)-target genes that form the core response to hypoxia. Nucleic Acids Research 37, 4587-4602 (2009); published online EpubAugust 1, 2009 (10.1093/nar/gkp425). 305. V. J. Tuominen, S. Ruotoistenmaki, A. Viitanen, M. Jumppanen, J. Isola, ImmunoRatio: a publicly available web application for quantitative image analysis of estrogen receptor (ER), progesterone receptor (PR), and Ki-67. Breast cancer research : BCR 12, R56 (2010)10.1186/bcr2615). 306. M. Raj, B. N. Bullock, P. S. Arora, Plucking the high hanging fruit: A systematic approach for targeting protein-protein interactions. Bioorganic & Medicinal Chemistry 21, 4051–4057 (2013) http://dx.doi.org/10.1016/j.bmc.2012.11.023). 307. I. Coin, M. Beyermann, M. Bienert, Solid-phase peptide synthesis: from standard procedures to the synthesis of difficult sequences. Nat. Protoc. 2, 3247-3256 (2007). 308. M. P. Williamson, Using chemical shift perturbation to characterise ligand binding. Progress in Nuclear Magnetic Resonance Spectroscopy 73, 1-16 (2013) http://dx.doi.org/10.1016/j.pnmrs.2013.02.001). 309. L. Qiu, Y. Bae, Polymer Architecture and Drug Delivery. Pharm Res 23, 1-30 (2006); published online Epub2006/01/01 (10.1007/s11095-005-9046-2). 310. A. Samad, Y. Sultana, M. Aqil, Liposomal drug delivery systems: an update review. Current Drug Delivery 4, 297-305 (2007). 311. P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R. E. Morris, C. Serre, Metal–Organic Frameworks in Biomedicine. Chemical Reviews 112, 1232-1268 (2011); published online Epub2012/02/08 (10.1021/cr200256v). 251 312. A. C. McKinlay, R. E. Morris, P. Horcajada, G. Férey, R. Gref, P. Couvreur, C. Serre, BioMOFs: Metal–Organic Frameworks for Biological and Medical Applications. Angewandte Chemie, International Edition 49, 6260-6266 (2010)10.1002/anie.201000048). 313. S. Keskin, S. Kızılel, Biomedical Applications of Metal Organic Frameworks. Industrial & Engineering Chemistry Research 50, 1799-1812 (2011); published online Epub2011/02/16 (10.1021/ie101312k). 314. H. Li, M. Eddaoudi, M. O'Keeffe, O. M. Yaghi, Design and synthesis of an exceptionally stable and highly porous metal-organic framework. Nature 402, 276-279 (1999) http://www.nature.com/nature/journal/v402/n6759/suppinfo/402276a0_S1.html). 315. J.-P. Zhang, X.-M. Chen, Crystal engineering of binary metal imidazolate and triazolate frameworks. Chemical Communications, 1689-1699 (2006)10.1039/b516367f). 316. P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J. F. Eubank, D. Heurtaux, P. Clayette, C. Kreuz, J.-S. Chang, Y. K. Hwang, V. Marsaud, P.-N. Bories, L. Cynober, S. Gil, G. Ferey, P. Couvreur, R. Gref, Porous metal-organic-framework nanoscale carriers as a potential platform for drug delivery and imaging. Nature Materials 9, 172-178 (2010) http://www.nature.com/nmat/journal/v9/n2/abs/nmat2608.html#supplementary- information). 317. C. Serre, F. Millange, C. Thouvenot, M. Noguès, G. Marsolier, D. Louër, G. Férey, Very Large Breathing Effect in the First Nanoporous Chromium(III)-Based Solids: MIL-53 or CrIII(OH)·{O2C−C6H4−CO2}·{HO2C−C6H4−CO2H}x·H2Oy. Journal of the American Chemical Society 124, 13519-13526 (2002); published online Epub2002/11/01 (10.1021/ja0276974). 318. T. R. Cook, V. Vajpayee, M. H. Lee, P. J. Stang, K.-W. Chi, Biomedical and Biochemical Applications of Self-Assembled Metallacycles and Metallacages. Accounts of Chemical Research, (2013)10.1021/ar400010v). 319. R. Chakrabarty, P. S. Mukherjee, P. J. Stang, Supramolecular Coordination: Self- Assembly of Finite Two- and Three-Dimensional Ensembles. Chemical Reviews 111, 6810-6918 (2011); published online Epub2011/11/09 (10.1021/cr200077m). 320. N. P. E. Barry, N. H. Abd Karim, R. Vilar, B. Therrien, Interactions of ruthenium coordination cubes with DNA. Dalton Transactions, 10717-10719 (2009)10.1039/b913642h). 321. O. Zava, J. Mattsson, B. Therrien, P. J. Dyson, Evidence for Drug Release from a Metalla-Cage Delivery Vector Following Cellular Internalisation. Chemistry – A European Journal 16, 1428-1431 (2010)10.1002/chem.200903216). 322. M. Pernot, T. Bastogne, N. P. E. Barry, B. Therrien, G. Koellensperger, S. Hann, V. Reshetov, M. Barberi-Heyob, Systems biology approach for in vivo photodynamic therapy optimization of ruthenium-porphyrin compounds. Journal of Photochemistry and Photobiology B: Biology 117, 80-89 (2012)http://dx.doi.org/10.1016/j.jphotobiol.2012.08.012). 323. M. J. Hannon, Supramolecular DNA recognition. Chemical Society Reviews 36, 280-295 (2007)10.1039/b606046n). 324. R. Kieltyka, P. Englebienne, J. Fakhoury, C. Autexier, N. Moitessier, H. F. Sleiman, A Platinum Supramolecular Square as an Effective G-Quadruplex Binder and Telomerase 252 Inhibitor. Journal of the American Chemical Society 130, 10040-10041 (2008); published online Epub2008/08/01 (10.1021/ja8014023). 325. V. Vajpayee, Y. J. Yang, S. C. Kang, H. Kim, I. S. Kim, M. Wang, P. J. Stang, K.-W. Chi, Hexanuclear self-assembled arene-ruthenium nano-prismatic cages: potential anticancer agents. Chemical Communications 47, 5184-5186 (2011)10.1039/c1cc10167f). 326. V. Vajpayee, Y. H. Song, Y. J. Jung, S. C. Kang, H. Kim, I. S. Kim, M. Wang, T. R. Cook, P. J. Stang, K.-W. Chi, Coordination-driven self-assembly of ruthenium-based molecular-rectangles: Synthesis, characterization, photo-physical and anticancer potency studies. Dalton Transactions 41, 3046-3052 (2012)10.1039/c2dt11811d). 327. V. Brabec, S. E. Howson, R. A. Kaner, R. M. Lord, J. Malina, R. M. Phillips, Q. M. A. Abdallah, P. C. McGowan, A. Rodger, P. Scott, Metallohelices with activity against cisplatin-resistant cancer cells; does the mechanism involve DNA binding? Chemical Science, (2013)10.1039/c3sc51731d). 328. H. Yu, M. Li, G. Liu, J. Geng, J. Wang, J. Ren, C. Zhao, X. Qu, Metallosupramolecular complex targeting an [small alpha]/[small beta] discordant stretch of amyloid [small beta] peptide. Chemical Science 3, 3145-3153 (2012)10.1039/c2sc20372c). 329. S. Leininger, B. Olenyuk, P. J. Stang, Self-assembly of discrete cyclic nanostructures mediated by transition metals. Chemical Reviews 100, 853-907 (2000); published online EpubMar (10.1021/cr9601324). 330. P. J. Stang, B. Olenyuk, Self-assembly, symmetry, and molecular architecture: Coordination as the motif in the rational design of supramolecular metallacyclic polygons and polyhedra. Accounts of Chemical Research 30, 502-518 (1997); published online EpubDec (10.1021/ar9602011). 331. M. A. Furrer, F. Schmitt, M. Wiederkehr, L. Juillerat-Jeanneret, B. Therrien, Cellular delivery of pyrenyl-arene ruthenium complexes by a water-soluble arene ruthenium metalla-cage. Dalton Transactions 41, 7201-7211 (2012)10.1039/c2dt30193h). 332. J. Mattsson, O. Zava, A. K. Renfrew, Y. Sei, K. Yamaguchi, P. J. Dyson, B. Therrien, Drug delivery of lipophilic pyrenyl derivatives by encapsulation in a water soluble metalla-cage. Dalton Transactions 39, 8248-8255 (2010)10.1039/c0dt00436g). 333. B. Therrien, G. Süss-Fink, P. Govindaswamy, A. K. Renfrew, P. J. Dyson, The “Complex-in-a-Complex” Cations [(acac)2M⊂Ru6(p-iPrC6H4Me)6(tpt)2(dhbq)3]6+: A Trojan Horse for Cancer Cells. Angewandte Chemie, International Edition 47, 3773-3776 (2008)10.1002/anie.200800186). 334. L. E. H. Paul, B. Therrien, J. Furrer, Investigation of the Reactivity between a Ruthenium Hexacationic Prism and Biological Ligands. Inorganic Chemistry 51, 1057-1067 (2011); published online Epub2012/01/16 (10.1021/ic2021935). 335. S. E. Sherman, S. J. Lippard, Structural aspects of platinum anticancer drug interactions with DNA. Chemical Reviews 87, 1153-1181 (1987); published online Epub1987/10/01 (10.1021/cr00081a013). 336. A. V. Klein, T. W. Hambley, Platinum Drug Distribution in Cancer Cells and Tumors. Chemical Reviews 109, 4911-4920 (2009); published online Epub2009/10/14 (10.1021/cr9001066). 337. A. Mishra, S. Ravikumar, S. H. Hong, H. Kim, V. Vajpayee, H. Lee, B. Ahn, M. Wang, P. J. Stang, K.-W. Chi, DNA Binding and Unwinding by Self-Assembled Supramolecular 253 Heterobimetallacycles. Organometallics 30, 6343-6346 (2011); published online Epub2011/12/12 (10.1021/om200802v). 338. Y.-R. Zheng, Z. Zhao, H. Kim, M. Wang, K. Ghosh, J. B. Pollock, K.-W. Chi, P. J. Stang, Coordination-Driven Self-Assembly of Truncated Tetrahedra Capable of Encapsulating 1,3,5-Triphenylbenzene. Inorganic Chemistry 49, 10238-10240 (2010); published online Epub2010/11/15 (10.1021/ic1018373). 339. M. Wang, V. Vajpayee, S. Shanmugaraju, Y.-R. Zheng, Z. Zhao, H. Kim, P. S. Mukherjee, K.-W. Chi, P. J. Stang, Coordination-Driven Self-Assembly of M3L2 Trigonal Cages from Preorganized Metalloligands Incorporating Octahedral Metal Centers and Fluorescent Detection of Nitroaromatics. Inorganic Chemistry 50, 1506- 1512 (2011); published online Epub2011/02/21 (10.1021/ic1020719). 340. J. B. Pollock, T. R. Cook, P. J. Stang, Photophysical and Computational Investigations of Bis(phosphine) Organoplatinum(II) Metallacycles. J. Am. Chem. Soc. 134, 10607-10620 (2012); published online Epub2012/06/27 (10.1021/ja3036515). 341. J. B. Pollock, G. L. Schneider, T. R. Cook, A. S. Davies, P. J. Stang, Tunable Visible Light Emission of Self-Assembled Rhomboidal Metallacycles. Journal of the American Chemical Society 135, 13676-13679 (2013); published online Epub2013/09/18 (10.1021/ja4079607). 342. J. B. Pollock, T. R. Cook, G. L. Schneider, D. A. Lutterman, A. S. Davies, P. J. Stang, Photophysical Properties of Endohedral Amine-Functionalized Bis(phosphine) Pt(II) Complexes as Models for Emissive Metallacycles. Inorganic Chemistry 52, 9254-9265 (2013); published online Epub2013/08/19 (10.1021/ic400491q). 343. J.-S. Chen, G.-J. Zhao, T. R. Cook, K.-L. Han, P. J. Stang, Photophysical Properties of Self-Assembled Multinuclear Platinum Metallacycles with Different Conformational Geometries. Journal of the American Chemical Society 135, 6694-6702 (2013); published online Epub2013/05/01 (10.1021/ja402421w). 344. M. C. Bissery, G. G. Chabot, [History and new development of screening and evaluation methods of anticancer drugs used in vivo and in vitro]. Bulletin du cancer 78, 587-602 (1991). 345. T. Lammers, F. Kiessling, W. E. Hennink, G. Storm, Nanotheranostics and Image- Guided Drug Delivery: Current Concepts and Future Directions. Mol. Pharm. 7, 1899- 1912 (2010); published online EpubNov-Dec (10.1021/mp100228v). 346. G. M. Lanza, C. Moonen, J. R. Baker, E. Chang, Z. Cheng, P. Grodzinski, K. Ferrara, K. Hynynen, G. Kelloff, Y. E. K. Lee, A. K. Patri, D. Sept, J. E. Schnitzer, B. J. Wood, M. Q. Zhang, G. Zheng, K. Farahani, Assessing the barriers to image-guided drug delivery. Wiley Interdiscip. Rev.-Nanomed. Nanobiotechnol. 6, 1-14 (2014); published online EpubJan (10.1002/wnan.1247). 347. E. D. Crawford, Understanding the epidemiology, natural history, and key pathways involved in prostate cancer. Urology 73, S4-10 (2009); published online EpubMay (10.1016/j.urology.2009.03.001). 348. C. Huggins, C. V. Hodges, Studies on prostatic cancer - I The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1, 293-297 (1941); published online EpubApr. 349. G. Andriole, N. Bruchovsky, L. W. Chung, A. M. Matsumoto, R. Rittmaster, C. Roehrborn, D. Russell, D. Tindall, Dihydrotestosterone and the prostate: the scientific 254 rationale for 5alpha-reductase inhibitors in the treatment of benign prostatic hyperplasia. The Journal of urology 172, 1399-1403 (2004); published online EpubOct. 350. M. R. Cooperberg, G. D. Grossfeld, D. P. Lubeck, P. R. Carroll, National practice patterns and time trends in androgen ablation for localized prostate cancer. J Natl Cancer Inst 95, 981-989 (2003); published online EpubJul 2. 351. S. M. Lippman, E. A. Klein, P. J. Goodman, M. S. Lucia, I. M. Thompson, L. G. Ford, H. L. Parnes, L. M. Minasian, J. M. Gaziano, J. A. Hartline, J. K. Parsons, J. D. Bearden, E. D. Crawford, G. E. Goodman, J. Claudio, E. Winquist, E. D. Cook, D. D. Karp, P. Walther, M. M. Lieber, A. R. Kristal, A. K. Darke, K. B. Arnold, P. A. Ganz, R. M. Santella, D. Albanes, P. R. Taylor, J. L. Probstfield, T. J. Jagpal, J. J. Crowley, F. L. Meyskens, L. H. Baker, C. A. Coltman, P. C. Walsh, Effect of Selenium and Vitamin E on Risk of Prostate Cancer and Other Cancers: The Selenium and Vitamin E Cancer Prevention Trial (SELECT). J Urology 181, 1686-1687 (2009); published online EpubApr (DOI 10.1016/j.juro.2008.12.066). 352. M. A. Hoffman, W. C. DeWolf, A. Morgentaler, Is low serum free testosterone a marker for high grade prostate cancer? J Urology 163, 824-827 (2000); published online EpubMar (Doi 10.1016/S0022-5347(05)67812-3). 353. J. Morote, C. Ramirez, E. Gomez, J. Planas, C. X. Raventos, I. M. de Torres, R. Catalan, The relationship between total and free serum testosterone and the risk of prostate cancer and tumour aggressiveness. Bju Int 104, 486-489 (2009); published online EpubAug (DOI 10.1111/j.1464-410X.2009.08378.x). 354. K. O. Rove, F. M. Debruyne, B. Djavan, L. G. Gomella, H. K. Koul, M. S. Lucia, D. P. Petrylak, N. D. Shore, N. N. Stone, E. D. Crawford, Role of Testosterone in Managing Advanced Prostate Cancer. Urology 80, 754-762 (2012); published online EpubOct (DOI 10.1016/j.urology.2012.05.006). 355. M. Pether, S. L. Goldenberg, Intermittent androgen suppression. Bju Int 93, 258-261 (2004); published online EpubFeb. 356. N. D. Shore, E. D. Crawford, Intermittent androgen deprivation therapy: redefining the standard of care? Reviews in urology 12, 1-11 (2010); published online EpubWinter. 357. J. C. Shih, K. Chen, Regulation of MAO-A and MAO-B gene expression. Current medicinal chemistry 11, 1995-2005 (2004); published online EpubAug. 358. A. S. Kalgutkar, D. K. Dalvie, N. Castagnoli, Jr., T. J. Taylor, Interactions of nitrogen- containing xenobiotics with monoamine oxidase (MAO) isozymes A and B: SAR studies on MAO substrates and inhibitors. Chemical research in toxicology 14, 1139-1162 (2001); published online EpubSep. 359. P. A. Humphrey, Gleason grading and prognostic factors in carcinoma of the prostate. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 17, 292-306 (2004); published online EpubMar (10.1038/modpathol.3800054). 360. L. True, I. Coleman, S. Hawley, C. Y. Huang, D. Gifford, R. Coleman, T. M. Beer, E. Gelmann, M. Datta, E. Mostaghel, B. Knudsen, P. Lange, R. Vessella, D. Lin, L. Hood, P. S. Nelson, A molecular correlate to the Gleason grading system for prostate adenocarcinoma. Proc Natl Acad Sci U S A 103, 10991-10996 (2006); published online EpubJul 18 (10.1073/pnas.0603678103). 255 361. H. Zhao, V. Flamand, D. M. Peehl, Anti-oncogenic and pro-differentiation effects of clorgyline, a monoamine oxidase A inhibitor, on high grade prostate cancer cells. BMC medical genomics 2, 55 (2009)10.1186/1755-8794-2-55). 362. R. E. Sobel, M. D. Sadar, Cell lines used in prostate cancer research: a compendium of old and new lines--part 2. The Journal of urology 173, 360-372 (2005); published online EpubFeb (10.1097/01.ju.0000149989.01263.dc). 363. K. Izuishi, K. Kato, T. Ogura, T. Kinoshita, H. Esumi, Remarkable tolerance of tumor cells to nutrient deprivation: possible new biochemical target for cancer therapy. Cancer Res 60, 6201-6207 (2000); published online EpubNov 1. 364. J. Fabian, H. Nakazumi, M. Matsuoka, Near-Infrared Absorbing Dyes. Chemical Reviews 92, 1197-1226 (1992); published online EpubSep-Oct (Doi 10.1021/Cr00014a003). 365. R. C. Srivastava, V. D. Anand, W. R. Carper, A Fluorescence Study of Hematoporphyrin. Appl. Spectrosc. 27, 444-449 (1973); published online Epub1973/11/01. 366. H. Auler, G. Banzer, Untersuchungen über die Rolle der Porphyrine bei geschwulstkranken Menschen und Tieren. Z Krebs-forsch 53, 65-68 (1942); published online Epub1942/03/01 (10.1007/BF01792783). 367. R. B. Mujumdar, L. A. Ernst, S. R. Mujumdar, C. J. Lewis, A. S. Waggoner, Cyanine Dye Labeling Reagents - Sulfoindocyanine Succinimidyl Esters. Bioconjugate Chem 4, 105-111 (1993); published online EpubMar-Apr (Doi 10.1021/Bc00020a001). 368. L. Strekowski, M. Lipowska, G. Patonay, Substitution-Reactions of a Nucleofugal Group in Heptamethine Cyanine Dyes - Synthesis of an Isothiocyanato Derivative for Labeling of Proteins with a near-Infrared Chromophore. J Org Chem 57, 4578-4580 (1992); published online EpubAug (Doi 10.1021/Jo00043a009). 369. R. J. Williams, M. Lipowska, G. Patonay, L. Strekowski, Comparison of Covalent and Noncovalent Labeling with near-Infrared Dyes for the High-Performance Liquid- Chromatographic Determination of Human Serum-Albumin. Anal Chem 65, 601-605 (1993); published online EpubMar 1 (Doi 10.1021/Ac00053a019). 370. L. R. Middendorf, J. C. Bruce, R. C. Bruce, R. D. Eckles, D. L. Grone, S. C. Roemer, G. D. Sloniker, D. L. Steffens, S. L. Sutter, J. A. Brumbaugh, G. Patonay, Continuous, on- line DNA sequencing using a versatile infrared laser scanner/electrophoresis apparatus. ELECTROPHORESIS 13, 487-494 (1992)10.1002/elps.11501301103). 371. S. R. Mujumdar, R. B. Mujumdar, C. M. Grant, A. S. Waggoner, Cyanine-Labeling Reagents: Sulfobenzindocyanine Succinimidyl Esters. Bioconjugate Chem 7, 356-362 (1996); published online Epub1996/01/01 (10.1021/bc960021b). 372. Z. Zhang, J. Fan, P. P. Cheney, M. Y. Berezin, W. B. Edwards, W. J. Akers, D. Shen, K. Liang, J. P. Culver, S. Achilefu, Activatable Molecular Systems Using Homologous Near-Infrared Fluorescent Probes for Monitoring Enzyme Activities in Vitro, in Cellulo, and in Vivo. Mol. Pharm. 6, 416-427 (2009); published online EpubMar-Apr (Doi 10.1021/Mp800264k). 373. C. Zhang, T. Liu, Y. Su, S. Luo, Y. Zhu, X. Tan, S. Fan, L. Zhang, Y. Zhou, T. Cheng, C. Shi, A near-infrared fluorescent heptamethine indocyanine dye with preferential tumor accumulation for in vivo imaging. Biomaterials 31, 6612-6617 (2010); published online EpubSep (10.1016/j.biomaterials.2010.05.007). 256 374. L. G. Lee, S. L. Woo, D. F. Head, R. S. Dubrow, T. M. Baer, Near-IR dyes in three-color volumetric capillary cytometry: cell analysis with 633- and 785-nm laser excitation. Cytometry 21, 120-128 (1995); published online EpubOct 1 (10.1002/cyto.990210203). 375. X. Yang, C. Shi, R. Tong, W. Qian, H. E. Zhau, R. Wang, G. Zhu, J. Cheng, V. W. Yang, T. Cheng, M. Henary, L. Strekowski, L. W. Chung, Near IR heptamethine cyanine dye- mediated cancer imaging. Clinical cancer research : an official journal of the American Association for Cancer Research 16, 2833-2844 (2010); published online EpubMay 15 (10.1158/1078-0432.CCR-10-0059). 376. X. Yang, C. Shao, R. Wang, C. Y. Chu, P. Hu, V. Master, A. O. Osunkoya, H. L. Kim, H. E. Zhau, L. W. Chung, Optical imaging of kidney cancer with novel near infrared heptamethine carbocyanine fluorescent dyes. The Journal of urology 189, 702-710 (2013); published online EpubFeb (10.1016/j.juro.2012.09.056). 377. C.-S. Yeh, C.-H. Su, W.-Y. Ho, C.-C. Huang, J.-C. Chang, Y.-H. Chien, S.-T. Hung, M.- C. Liau, H.-Y. Ho, Tumor targeting and MR imaging with lipophilic cyanine-mediated near-infrared responsive porous Gd silicate nanoparticles. Biomaterials 34, 5677-5688 (2013); published online Epub7// (http://dx.doi.org/10.1016/j.biomaterials.2013.04.020). 378. Y. Zhang, L. Xiao, K. Popovic, X. Xie, M. D. Chordia, L. W. Chung, M. B. Williams, W. Yue, D. Pan, Novel cancer-targeting SPECT/NIRF dual-modality imaging probe (99m)Tc-PC-1007: synthesis and biological evaluation. Bioorg Med Chem Lett 23, 6350- 6354 (2013); published online EpubDec 1 (10.1016/j.bmcl.2013.09.074). 379. Q. Li, J. Tan, B. X. Peng, Synthesis and characterization of heptamethine cyanine dyes. Molecules 2, 91-98 (1997)Doi 10.3390/20600091). 380. L. W. K. Chung, D. Pan, Compositions and methods for tumor imaging and targeting by a class of organic heptamethine cyanine dyes that possess dual nuclear and near-infrared properties. Intl. Patent WO2013052776 A1, (Apr 11, 2013); published online EpubApr 11, 2013. 381. P.-A. Bouit, G. Wetzel, G. Berginc, B. Loiseaux, L. Toupet, P. Feneyrou, Y. Bretonnière, K. Kamada, O. Maury, C. Andraud, Near IR Nonlinear Absorbing Chromophores with Optical Limiting Properties at Telecommunication Wavelengths. Chemistry of Materials 19, 5325-5335 (2007); published online Epub2007/10/01 (10.1021/cm071557h). 382. S. L. Luo, X. Tan, Q. R. Qi, Q. Y. Guo, X. Z. Ran, L. L. Zhang, E. L. Zhang, Y. F. Liang, L. L. Weng, H. Zheng, T. M. Cheng, Y. P. Su, C. M. Shi, A multifunctional heptamethine near-infrared dye for cancer theranosis. Biomaterials 34, 2244-2251 (2013); published online EpubMar (DOI 10.1016/j.biomaterials.2012.11.057). 383. N. Narayanan, G. Patonay, A New Method for the Synthesis of Heptamethine Cyanine Dyes - Synthesis of New near-Infrared Fluorescent Labels. J Org Chem 60, 2391-2395 (1995); published online EpubApr 21 (Doi 10.1021/Jo00113a018). 384. V. E. Shershov, M. A. Spitsyn, V. E. Kuznetsova, E. N. Timofeev, O. A. Ivashkina, I. S. Abramov, T. V. Nasedkina, A. S. Zasedatelev, A. V. Chudinov, Near-infrared heptamethine cyanine dyes. Synthesis, spectroscopic characterization, thermal properties and photostability. Dyes Pigments 97, 353-360 (2013); published online EpubMay (DOI 10.1016/j.dyepig.2012.12.031). 385. E. L. Zhang, S. L. Luo, X. Tan, C. M. Shi, Mechanistic study of IR-780 dye as a potential tumor targeting and drug delivery agent. Biomaterials 35, 771-778 (2014); published online EpubJan (DOI 10.1016/j.biomaterials.2013.10.033). 257 386. M. V. Kvach, A. V. Ustinov, I. A. Stepanova, A. D. Malakhov, M. V. Skorobogatyi, V. V. Shmanai, V. A. Korshun, A convenient synthesis of cyanine dyes: Reagents for the labeling of biomolecules. Eur J Org Chem, 2107-2117 (2008); published online EpubApr (DOI 10.1002/ejoc.200701190). 387. X. Y. Chen, X. J. Peng, A. J. Cui, B. S. Wang, L. Wang, R. Zhang, Photostabilities of novel heptamethine 3H-indolenine cyanine dyes with different N-substituents. J Photoch Photobio A 181, 79-85 (2006); published online EpubJul 5 (DOI 10.1016/j.jphotochem.2005.11.004). 388. X. Tan, S. L. Luo, D. C. Wang, Y. P. Su, T. M. Cheng, C. M. Shi, A NIR heptamethine dye with intrinsic cancer targeting, imaging and photosensitizing properties. Biomaterials 33, 2230-2239 (2012); published online EpubMar (DOI 10.1016/j.biomaterials.2011.11.081). 389. J. J. Gao, K. H. Xu, B. Tang, L. L. Yin, G. W. Yang, L. G. An, Selective detection of superoxide anion radicals generated from macrophages by using a novel fluorescent probe. Febs J 274, 1725-1733 (2007); published online EpubApr (Doi 10.1111/J.1742- 4658.2007.05720.X). 390. Y. Kishi, T. Fukuyama, S. Nakatsuka, New method for the synthesis of epidithiodiketopiperazines. J. Amer. Chem. Soc. 95, 6490-6492 (1973)10.1021/ja00800a077). 391. T. Fukuyama, S. Nakatsuka, Y. Kishi, Total synthesis of gliotoxin, dehydrogliotoxin and hyalodendrin. Tetrahedron 37, 2045-2078 (1981)10.1016/s0040-4020(01)97960-8). 392. Z. Wu, L. J. Williams, S. J. Danishefsky, A three-step entry to the aspirochlorine family of antifungal agents. Angew. Chem., Int. Ed. 39, 3866-3868 (2000). 393. F. Monti, F. Ripamonti, S. P. Hawser, K. Islam, Aspirochlorine: A highly selective and potent inhibitor of fungal protein synthesis. J Antibiot 52, 311-318 (1999); published online EpubMar. 394. P. W. Trown, Antiviral Activity of N,N'-Dimethyl-Epidithiapiperazinedione a Synthetic Compound Related to Gliotoxins Ll-S88alpha and Beta Chetomin and Sporidesmins. Biochem Bioph Res Co 33, 402-& (1968)Doi 10.1016/0006-291x(68)90585-8). 395. http://rsbweb.nih.gov/ij/plugins/cell-counter.html. 396. http://labs.fhcrc.org/fero/Protocols/cellcountsimagej.html. 397. http://freeware.ronhoebe.com/free/CellCounter.asp. 398. http://sourceforge.net/projects/cellcount/. 399. http://www.dnabaser.com/download/Cell-counter/grid-cell-counter.html. 400. J. T. Carpenter AE, Lamprecht MR, Clarke C, Kang IH, Friman O, Guertin DA, Chang JH, Lindquist RA, Moffat J, Golland P, Sabatini DM, CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biology 7, (2006)10.1186/gb-2006-7-10-r100). 401. J. S. Jyrki Selinummi, Olli Yli-Harja, and Jaakko A. Puhakka, Software for quantification of labeled bacteria from digital microscope images by automated image analysis. BioTechniques 39, 859-863 (2005). 402. http://nghiaho.com/?page_id=1011. 258 403. Q. Geissmann, OpenCFU, a New Free and Open-Source Software to Count Cell Colonies and Other Circular Objects. PLOS One, (2013)10.1371/journal.pone.0054072). 404. http://www.microbehunter.com/connecting-a-camera-to-a-microscope/. 405. http://fiehnlab.ucdavis.edu/staff/kind/Collector/Cell_Counting. 406. DPReview_Staff, Making (some) sense out of sensor sizes. Digital Photography Review, (2012). 407. http://www.burtonsys.com/climate/composite_standard_deviations.html.

Abstract (if available)

Abstract Protein‐protein interactions are attractive targets for drug design due to their fundamental role in biological function. Of a particular interest is the interaction between the helical CTAD domain of the hypoxia‐inducible factor 1α (HIF1α) and its cofactor p300/CBP, as it regulates the transcription of key genes, whose expression contributes to angiogenesis, metastasis, and altered energy metabolism in cancer. However, small molecules that selectively target the intended interactions have been difficult to access using traditional drug discovery approaches. In Chapter I, we describe the design, synthesis, biochemical and in vivo evaluation of a small molecule oligooxopiperazine scaffold that captures the topography of a key α‐helical domain at the interface of HIF1α and p300 to develop inhibitors of hypoxia‐inducible signaling. The designed compounds target the desired protein with high affinity and in a predetermined manner with the optimal ligand providing effective down‐regulation of hypoxia‐dependent genes, and reduction of the tumor burden in two experimental mouse tumor xenograft models. ❧ Denaturation of one of the binding partners is another approach implemented for the targeting of some protein‐protein interactions. Epidithiodiketopiperazines eject Zn ions from the CH1 domain of p300/CBP, disrupting its global fold and thereby allosterically inhibiting its interaction with HIF1α, as well as other proteins interacting with the CH1 domain. However, the synthesis of epidithiodiketopiperazines poses a substantial challenge. In Chapter VI, we describe a scalable procedure for the multi‐gram scale synthesis of 3-(4-methoxyphenyl)-6,8-dimethyl-2,4-dithia-6,8-diazabicyclo[3.2.2]nonane-7,9-dione, a common precursor for several designed epidithiodiketopiperazines. ❧ In Chapter IV, we investigated the hypoxia‐dependent signaling in androgen‐insensitive metastatic prostate cancer cell lines C42B and PC3. Both of these cell lines exhibit elevated levels of monoamine oxidase A (MAOA) under hypoxia, and in C42B cells, MAOA and HIF1α are abundant under normoxia as well. We show that epidithiodiketopiperazine LS72, a designed transcriptional antagonist of hypoxia‐inducible genes, decreases hypoxia‐induced transcriptional levels of HIF‐dependent genes in PC3 and C42B cells. We found that LS72 also down‐regulates MAOA transcription in a dose‐dependent manner under hypoxia. Clorgyline, an inhibitor of MAOA enzymatic activity and a known prostate cancer antagonist, and LS72 exhibit synergistic suppressive effect on the transcription of both MAOA and two HIF‐dependent genes—LOX and GLUT1. As such, we hypothesize that MAOA is a hypoxia‐inducible gene in certain prostate cancer cell lines, regulated by HIF, and the combination of the transcriptional antagonist of hypoxia‐inducible genes and the inhibitor of MAOA enzymatic activity is efficacious for the treatment of prostate cancer. Drug delivery presents yet another challenging area of pharmaceutical research. ❧ In Chapter III, we describe the self‐assembled rhomboidal Pt(II)‐based fluorescent supramolecular coordination complexes (SCCs). We demonstrated that such SCCs are water‐soluble and nontoxic to cells, and analyzed their cellular uptake and intracellular localization. Rhomboids remained intact upon cellular internalization and did not photobleach or otherwise degrade. Therefore, the well‐defined geometry, presence of an internal cavity, and ability to emit within the visible spectrum makes such endohedral amine‐functionalized SCCs attractive candidates for further development as vehicles for drug delivery. Importantly, one of these SCCs by itself caused a substantial 64% reduction of the tumor burden in a mouse tumor xenograft model on the last day of experiment, accompanied by a 2.4‐fold reduction of the proliferative marker Ki67 and substantial improvement of the cell morphology. ❧ Live tissue imaging is an indispensable tool for the analysis of many pathologies, and in particular, cancer. Low tissue autofluorescence and deep tissue penetration allow heptamethine cyanine near‐infrared (NIR) fluorescence dyes to be excellent probes for tumor imaging. In Chapter V, we describe a procedure to synthesize the NIR dye MHI148 in high yield (88%), on a gram scale. We confirmed its purity and conducted a thorough analysis, including detailed 1H and ¹³C NMR peak assignment and assignment of all ion peaks in the LC-MS. The synthesis did not require column purification at any stage. The dye successfully crossed the blood‐brain barrier, and targeted the brain tumor. ❧ Lastly, in Chapter VII we describe a robust approach to equip a conventional light microscope with a camera to obtain usable images of the cells and cell cultures, in order to automate a mundane task—mammalian cell counting. We have devised two algorithms to automatically count mammalian cells in the pictures of the cell suspension in the hemocytometer assembly. Such approach takes 2-10 times less time, leaves a digital trace, and yields the same results as manual count on the same area. Importantly, it is not mandatory for the user to count the cells in the designated areas of the hemocytometer. Thus, such approach allows for the selection of the most representative areas and does not impose any penalty on the number of such areas to be processed, therefore yielding consistent and arguably more reliable results, independent of the sample dilution.

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

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

PDF

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

PDF

Study of the mechanism by which hypoxia-inducible-factor-1 alpha (HIF-1α) regulates heat shock protein-90 alpha (Hsp90α) secretion in breast cancer

PDF

Modulation of Runx proteins by steroid hormone receptors

PDF

Discovery of novel small molecules for ovarian cancer treatment

PDF

Discovery of novel small molecules targeting cancer cell metabolism

PDF

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

PDF

Development of novel small molecules targeting mitochondrial and oxidative stress signaling pathways for pancreatic cancer therapy

PDF

Wnt/β-catenin/p300 induced transcription is critical for the differentiation and maintenance of Paneth cells

PDF

Identification of small-molecules targeting CXCR2 function and signaling

PDF

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

PDF

Design of novel anticancer agents targeting cellular stress response pathways

PDF

Flipping the switch on protein activity activity: elastin-like polypeptides assemble into cell switches and vesicles

PDF

HIF-1α gene polymorphisms and risk of severe-spectrum hypertensive disorders of pregnancy: a pilot triad-based case-control study

PDF

pRB structures and functions mediating suppression of retinoblastoma initiation

PDF

Multi-omic data mining to elucidate molecular adaptation mechanisms of filamentous fungi exposed to space environment

PDF

Genome engineering of filamentous fungi for efficient novel molecule production

PDF

Temperature triggered protein assembly enables signaling switching and peptide drug delivery

PDF

Mechanistic studies of novel small molecule anti-cancer agents using next generation sequencing

Asset Metadata

Creator Grishagin, Ivan V. (author)

Core Title Small molecule modulators of HIF1α signaling

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 08/07/2015

Defense Date 06/16/2014

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

Tag cancer,helix mimetics,HIF1α,hypoxia,OAI-PMH Harvest,p300/CBP,protein‐protein interaction,small molecules

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Olenyuk, Bogdan Z. (committee chair), Okamoto, Curtis Toshio (committee member), Wang, Clay C.C. (committee member)

Creator Email grishagin@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-456649

Unique identifier UC11286747

Identifier etd-GrishaginI-2790.pdf (filename),usctheses-c3-456649 (legacy record id)

Legacy Identifier etd-GrishaginI-2790.pdf

Dmrecord 456649

Document Type Dissertation

Format application/pdf (imt)

Rights Grishagin, Ivan V.

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