University of Southern California Dissertations and Theses

Study of a novel near-infrared conjugated MAOA inhibitor, NMI, against CNS cancer by NCI60 data analysis

(USC Thesis Other)

Study of a novel near-infrared conjugated MAOA inhibitor, NMI, against CNS cancer by NCI60 data analysis

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Study of a novel Near-infrared Conjugated MAOA inhibitor, NMI, against CNS Cancer by NCI60 Data Analysis by Qianhua Feng A Thesis Presented to the FACULTY OF THE USC SCHOOL OF PHARMACY UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (PHARMACEUTICAL SCIENCES) May 2021 Copyright 2021 Qianhua Feng ii ACKNOWLEDGMENTS I would like to express my sincere and special gratitude to my advisor Dr. Jean C. Shih, who has been patient and generous, and gave me encouragement and valuable guidance for my thesis study, thank you. I would like to thank my committee members, Dr. Ian Haworth and Dr. Curtis Okamoto, for their time, guidance, advice for my thesis, thank you. I would like to acknowledge the members in the Shih Lab for their help and advice. iii TABLE OF CONTENTS ACKNOWLEDGMENTS .............................................................................................................. ii LIST OF TABLES .......................................................................................................................... v LIST OF FIGURES ....................................................................................................................... vi ABBREVIATIONS ...................................................................................................................... vii ABSTRACT ................................................................................................................................... ix INTRODUCTION .......................................................................................................................... 1 1. Central nervous system (CNS) cancer .................................................................................... 1 2. Mechanism of gliomas progression ........................................................................................ 2 2.1. Genetic alterations: mutation and amplification ............................................................................................ 2 2.2. Microenvironment: Hypoxia, ROS, and EMT .............................................................................................. 4 3. Therapeutic challenges of CNS cancer ................................................................................... 5 4. FDA-approved anti-CNS cancer drugs ................................................................................... 6 5. Monoamine oxidases (MAOs) ................................................................................................ 7 MATERIALS AND METHODS .................................................................................................. 13 1. MAOA gene expression analysis .......................................................................................... 13 2. Screening methodology of NMI ........................................................................................... 13 3. Statistical analysis ................................................................................................................. 14 4. COMPARE analysis of NMI and other drugs ...................................................................... 15 RESULTS ..................................................................................................................................... 15 1. MAOA expression in normal tissues and tumor tissues ....................................................... 15 2. Activity of NMI in cancer cell lines ..................................................................................... 19 3. Correlation between MAOA expression and anti-cancer potency of NMI in CNS cancer .. 24 4. Potency comparison of NMI and four FDA-approved CNS cancer drugs in CNS cancer cell lines 25 5. Therapeutic index comparison .............................................................................................. 31 6. Mechanism prediction and comparison by using COMPARE Algorithm ............................ 32 DISCUSSION ............................................................................................................................... 34 1. Possible mechanism of NMI ................................................................................................. 34 1.1. Drugs with similar growth inhibition activities of NMI ...................................................................... 34 1.2. Correlation of other genes and NMI .................................................................................................... 40 2. The potency of NMI may be different for different stages of gliomas ................................. 42 CONCLUSION ............................................................................................................................. 45 REFERENCES ............................................................................................................................. 46 iv APPENDICES .............................................................................................................................. 52 v LIST OF TABLES Table 1. FDA approved drugs for brain tumors. ............................................................................ 6 Table 2. Median MAOA expression in normal brain tissues of different regions and median MAOA expression in tumor tissues. ............................................................................................. 17 Table 3. Raw data of GI50, TGI, LC50 and therapeutic index (TI) values of NMI and four FDA- approved drugs in CNS cancer cell lines. ..................................................................................... 31 Table 4. Compounds with high PCC with NMI. .......................................................................... 39 vi LIST OF FIGURES Figure 1. Chemical structure of NMI. ........................................................................................... 11 Figure 2.MAOA expression in non-glioma patients from CGGA. ............................................... 17 Figure 3. MAO A expression in normal cells versus in tumor cells of CNS and prostate cancer. 18 Figure 4. Five Dose-response curves for NMI in 59 cancer cell lines from NCI. ........................ 19 Figure 5. The waterfall plot comparing endpoints of NMI in prostate and CNS cancer cell lines. ....................................................................................................................................................... 21 Figure 6. The overall potency comparison of NMI in CNS and prostate cancer cell lines. ......... 22 Figure 7. Scatter plot of correlation between MAO A expression and NMI antiproliferative efficacy (GI50) in all CNS cancer cell lines. ................................................................................ 25 Figure 8. The endpoints analysis of NMI and FDA-approved CNS drugs from NCI60. ............. 27 Figure 9. Heatmap representing GI50, TGI, and LC50 of NMI and four CNS cancer drugs in CNS cancer cell lines. ................................................................................................................... 29 Figure 10. 3D scatter plot representing endpoints of all tested drugs in CNS cancer cell lines. .. 30 Figure 11. Heatmap comparing PCC of NMI and four FDA-approved CNS cancer drugs. ........ 33 Figure 12. Top 25 interacting genes of MAOA from UCSC Genomes Browser. ........................ 41 Figure 13. MAO A expression in different stages of brain cancers from CGGA and Gent2. ...... 43 Figure 14. MAO A expression in different IDH mutation, 1p/19q co-deletion, and progression status of brain cancers from CGGA. ............................................................................................. 45 vii ABBREVIATIONS 5 5-hydroxytryptamine 5-HT .............................................................. 7 A Alzheimer’s disease AD ................................................................ 7 B B-cell lymphoma 2 Bcl-2 ............................................................ 36 Bipolar disorder BPD ............................................................... 7 Blood brain barrier BBB ............................................................... 5 C Cadherin 1 CDH1........................................................... 40 Cancer RNA-Seq Nexus CRN ............................................................. 15 Catenin beta 1 CTNNB1 ..................................................... 40 Central nervous system CNS ............................................................... 1 Chinese Glioma Genome Atlas CGGA .......................................................... 15 Cyclin-dependent kinases CDK............................................................. 36 D Dimethyl sulfoxide DMSO ......................................................... 13 E Electron transport chain complex I CI 38 Epidermal growth factor EGF ............................................................... 3 Epidermal growth factor receptor EGFR ............................................................. 2 Epithelial cadherin E-cadherin ..................................................... 8 Epithelial–mesenchymal transition EMT............................................................... 4 F Fragments per kilobase of exon per million FPKM .......................................................... 16 G Genotype-Tissue Expression GTEx............................................................ 15 Glioblastoma GBM .............................................................. 1 Glioma stem cells GSCs .............................................................. 5 Glutathione disulfide GSSG ........................................................... 38 Glyceraldehyde 3-phosphate dehydrogenase GAPDH........................................................ 40 H Histone deacetylases HDAC .......................................................... 36 Hypoxia-inducible factor 1  HIF-1 ........................................................... 3 I Interleukin 6 IL-6 .............................................................. 37 Isocitrate dehydrogenases IDH ................................................................ 3 M Mammalian target of rapamycin mTOR ............................................................ 3 MAOA inhibitor MAOAI ........................................................ 10 Matrix metalloproteinase 9 MM9 ............................................................ 11 Mitochondrial electron transport chain mETC ........................................................... 38 Mitogen-activated protein kinases MAPK ............................................................ 3 Monoamine oxidases MAOs ............................................................ 7 Monoamine oxidases B MAOB ........................................................... 7 Myeloid-cell leukemia 1 Mcl-1............................................................ 37 viii N NADPH oxidase NOX ............................................................ 38 National Cancer Institute NCI .............................................................. 13 Near-infrared NIR .............................................................. 10 Neurofibromatosis type 1 NF1 .............................................................. 40 Neuropilin-1 NRP-1 ............................................................ 3 N-Myc downstream regulated 1 NDRG1 ........................................................ 40 Nuclear factor kappa B NF-kB .......................................................... 35 Nuclear receptor subfamily 3 group C member 1 NR3C1 ......................................................... 40 Nuclear respiratory factor 1 Nrf-1 ............................................................ 35 O Organic anion-transporting polypeptide OATPs ......................................................... 10 P Pairwise pearson correlation coefficient PCC ............................................................. 15 Parkinson’s disease PD .................................................................. 7 Phosphatase and tensin homolog PTEN ............................................................. 2 Phosphoinositide 3-kinase PI3K............................................................... 2 Prolyl hydroxylase domain enzymes PHD ............................................................... 4 Prostate-specific antigen PSA .............................................................. 11 Proteasome inhibitors PIs ................................................................ 35 Protein kinase B AKT ............................................................... 3 R Reactive oxygen species ROS ............................................................... 4 Receptor tyrosine kinase RTK ............................................................... 2 S Schizophrenia SCZ ................................................................ 7 Subependymal giant cell astrocytoma SEGA ............................................................. 6 Sulforhodamine B SRB .............................................................. 13 T Temozolomide TMZ ............................................................... 5 The European Bioinformatics Institute EMBL-EBI .................................................. 15 The United State Food and Drug Adiministration FDA ............................................................... 6 Transcription Factor 25 TCF25 .......................................................... 40 Transcripts per million TPM ............................................................. 15 Trichloroacetic acid TCA ............................................................. 13 Tumor necrosis factor alpha TNF-  .......................................................... 37 Tumor protein P53 TP53 ............................................................... 2 Twist family transcription factor 1 TWIST1 ......................................................... 5 V vascular endothelial growth factor VEGF ............................................................. 3 Verrucarin A VC-A............................................................ 36 W Wingless-related integration site WNT .............................................................. 5 World Health Organization WHO .............................................................. 1 Z Zinc finger E-box binding Homebox 1 ZEB1 .............................................................. 4 ix ABSTRACT Our previous studies have shown that higher monoamine oxidase A (MAOA) expression is in drug-resistant recurrent gliomas, MAOA inhibitor NMI (Near-infrared dye conjugate MAOA Inhibitor) shows antiproliferative activities and reduced gliomas growth. This study examined the potency of NMI on additional CNS cancer cell lines by NCI60 screening data analysis. Literature search and database analysis revealed that CNS cancer has increased MAOA expression in tumor tissues than in normal tissues, which is similar to prostate cancer. The potency, growth inhibition, and lethal doses of NMI in CNS cancer and prostate cancer were studied by using waterfall plots, 3D scatter plots, and heatmaps. The results of this study show that NMI is more sensitive to CNS (U251, SNB-19, SF-539, SNB-75, SF295) cancer cell lines than prostate cancer cell lines (PC3, DU145) base on NMI waterfall plots of GI50, TGI, and LC50. Two CNS cancer cell lines, SF-539 and U251, showed higher overall potency to NMI treatment than other CNS and prostate cancer cell lines. The linear regression between MAOA expression and GI50 of NMI shows a weak negative correlation (r = -0.35), which means the antitumor activity of NMI in the CNS cancer also correlates to other genes probably. The drug activities and therapeutic index of NMI and four FDA-approved CNS cancer drugs were compared in NCI60 CNS cancer cell lines, NMI exhibited the best antitumor activity among all tested drugs. The Pairwise Pearson Correlation Coefficient (PCC) showed that NMI has a unique mechanism compared to the traditional CNS drugs. This study shows that NMI may be a candidate drug for CNS cancer with high antitumor activity and multiple mechanisms that are different from existing CNS drugs. Key Words: NCI60, MAOA, NMI, CNS Cancer, glioma, GI50, TGI, LC50. 1 INTRODUCTION 1. Central nervous system (CNS) cancer Central nervous system (CNS) cancer is a disease that abnormal cells form and grow in the tissues which are within or surrounding the brain and spinal cord 1 . The risk factors of causing CNS tumor include genetic disorder, family histories of CNS cancer, immunodeficiency, stresses, ages, and many other unclear environmental factors 2 . Tumors of the central nervous system can be either benign or cancerous, they are categorized into primary tumors and metastatic tumors 3 . According to the World Health Organization (WHO), CNS tumors can be classified into four different grades, grade I to grade IV 4 . In adults, the most common types of brain tumors are gliomas, approximately 25.5% of primary CNS cancers are gliomas and about 80.8% of primary brain cancers are malignant tumors in the United States from 2012 to 2016 5 . Various types of gliomas are developed from glial cells which is the important supporter and protector of the central nervous system. The main types of glioma tumors are oligodendrogliomas, astrocytomas, and glioblastoma, they are grouped as diffuse gliomas by WHO based on their similarities of growth behaviors and genetic mutations. Oligodendrogliomas, WHO grade II CNS tumors, grow slowly. As with astrocytoma, more aggressive oligodendrogliomas, WHO grade III anaplastic oligodendrogliomas, can be developed over time. Diffused astrocytomas start as WHO grade II and it can become more aggressive WHO grade III anaplastic astrocytomas. Astrocytomas grow slowly with diffusion and disruption to normal brain tissues 3 . Glioblastoma (GBM), a malignant Grade IV brain tumor, is the most aggressive and fast-growing cancer occurred in the brain or spinal cord. GBM is the most common malignant brain tumor, representing 2 57.3% of malignant gliomas. The majority of GBM tumors grow in elderly patients 4 . The median survival of the patients with GBM tumors is only about 14.5 to 16.6 months 6 . 2. Mechanism of gliomas progression The mechanisms of glioma initiation and progression are complicated, the growth of glioma tumor associates with complex and unclear microenvironments and various genetic mutations or amplifications. Genetic alterations have been reported as one of the main events that often happens in gliomas, contributing to the prognosis estimation and targeted treatment 4 . 2.1. Genetic alterations: mutation and amplification The tumor suppressor genes are one of the important features of tumor initiation and growth since they are the key regulators to control cell cycle progression and signaling pathway. The mutations of tumor suppressor genes, tumor protein P53 (TP53), affect normal cell progression and proliferation of glioma cells, promoting the initiation and growth of glioma tumors 7 . Approximately 86% of genetic alteration in human gliomas are p53 inactivation 8 . Receptor tyrosine kinase (RTK)-encoding genes are significant elements to regulate the growth factor signaling pathways. Amplification and mutation of these genes have been demonstrated to cause tumorigenesis by dysregulating phosphoinositide 3-kinase (PI3K) pathway. Mutations of phosphatase and tensin homolog (PTEN), another glioma tumor suppressor, also disrupts the PI3K pathway so that helps tumor cell survival 7 . Moreover, approximately 90% of genetic variations of glioma correlate to abnormalities of RTK- encoding genes in RTK/PI3K signaling pathways. The induction of RTKs is caused by activation of epidermal growth factor receptor (EGFR) and amplification of vascular 3 endothelial growth factor (VEGF) subsequently 8, 9 . EGFR and VEGF are two important elements in glioma cell growth. Overexpression of EGFR induces epidermal growth factor (EGF), a ligand that stimulates the expression of VEGF. The formation of new blood vessels from existing vasculature is promoted by VEGFA under normal physiological conditions. It also plays a significant role in the progression and migration of tumors through increasing pathologic angiogenesis, providing oxygen and nutrients for tumor cells. The association between VEGFA and the process of angiogenesis have been demonstrated 10, 11 . VEGFA plays an important role in the intracellular signaling pathways. While VEGFA interacts with neuropilin-1 (NRP-1), a transmembrane glycoprotein, the migration and angiogenesis of vascular endothelial cells occur to cause degradation of extracellular matrix 11, 12 . VEGF also has been reported to have high affinity to RTKs so that deregulation of VEGF then may activate RTKs/PI3K/ protein kinase B (AKT) and RAS/mitogen-activated protein kinases (MAPK) pathways. Amplification of Ras always associates with high-grade glioma and it often accompanies abnormalities of other genes or signaling pathways 8 . Several reports suggested that isocitrate dehydrogenases (IDH) is an important prognostic factor for glioma tumors. Different grades of glioma tumors can be identified by the status of the isocitrate dehydrogenases 1 and 2 (IDH1/2) genes. The IDH mutations were found in lower-grade glioma tumors such as oligodendrogliomas and astrocytomas; in contrast, IDH wildtypes were only observed in the high-grade malignant GBM tumors. The IDH 1/2 mutations promote the stabilization of hypoxia-inducible factor 1  (HIF-1) and subsequently activate its target genes, VEGFA, in PI3K/AKT/ mammalian target of rapamycin (mTOR) signal transduction pathway 4, 7-9, 13 . 4 2.2. Microenvironment: Hypoxia, ROS, and EMT Recently, many studies started focusing on the impact of the microenvironment in gliomas and investigating new therapy via changing the functions of the tumor microenvironments. It has been mentioned above that microenvironments may also contribute to cell proliferation and invasion in glioma tumors. Hypoxia, a significant component of the tumor microenvironment, has been reported as an essential stimulator of glioma initiation and progression. Hypoxia response has been demonstrated to be associated with the HIF-1 signaling pathway and can be regulated by HIF-1. Studies reported that HIF-1 can be degraded by prolyl hydroxylase domain enzymes (PHD) which is a dioxygenase in a normoxic environment 14 . Under a hypoxic microenvironment, PHDs are inhibited and diminished due to the lack of oxygen, and HIF-1 becomes more stable subsequently. The stabilization of HIF-1 helps to induce the target genes of HIF-1. Hypoxia also has been reported as an essential regulator of VEGF, and increased VEGF mRNA expression was observed in the cells under the hypoxic microenvironment. 11, 14, 15 . Several studies revealed that the elevated production of reactive oxygen species (ROS) in mitochondria causes by a relative hypoxia condition. ROS is an important regulator of cellular function, controlling cell proliferation and differentiation. The induction and progression of brain cancer through different downstream signaling pathways and DNA mutations are the consequences of increased ROS 15, 16 . Recent studies proposed that hypoxia and ROS might induce epithelial–mesenchymal transition (EMT), a biological process of losing cell-cell contact and increasing mesenchymal phenotype in epithelial cells, to initiate the growth of cancer cells 17 . In vitro studies reported that a hypoxid condition leads to the activation of EMT with higher expressions of zinc finger E-box binding Homebox 1 (ZEB1), 5 a transcriptional repressor, in GBM cells via the HIF-1 signaling pathway. ZEB1 has been demonstrated as a contributor for tumor invasion and metastasis of CNS cancers. The characteristics of EMT are also related to the twist family transcription factor 1 (TWIST1). The results of in vitro studies showed that increased TWIST1 expression in the HIF-1  pathway associated with hypoxia and EMT induces the invasion of GBM. Literature data also revealed increased expression of regulators and activators of EMT, such as SNAIL, SLUG, wingless-related integration site (WNT)/ -catenin, NOTCH, CD44, in GBM tumors 17-20 . 3. Therapeutic challenges of CNS cancer Previous studies found several therapeutic difficulties and limitations for treating CNS cancers due to their biological characteristics and drug resistance. One of properties of diffuse glioma is to spread cancer cells to surrounding normal brain tissues. The infiltration property of gliomas is one of the Achilles heels for effective and precise targeted therapies 3, 21, 22 . Otherwise, the presence of the blood brain barrier (BBB) has been demonstrated as the most common challenge for CNS cancer treatments and therapies. Many brain tumor drugs are restricted to treat CNS cancer effectively because they cannot cross the BBB 23 . In vivo studies showed that temozolomide (TMZ), one of the common and effective treatments for glioma tumors, associates with drug resistance while tumors recur 24 . Therefore, developing a novel drug or therapy that can overcome TMZ-resistance and the BBB, and target tumors precisely is necessary. Glioma stem cells (GSCs) of glioma tumors are also another therapeutic challenge for investigating CNS treatment. GSCs are the tumor cells that associate with the formation of heterogeneous glial tumors. The growth rate of GSCs is slow 6 so that they can recover after receiving chemotherapy or radiotherapy and then cause tumor recurrence. The chemoresistance of GSCs is caused by heterogenetic formation and resistant adaptation 25 . It is necessary to investigate a new drug to overcome glioma tumor recurrences. 4. FDA-approved anti-CNS cancer drugs Five cancer drugs for CNS cancer were approved by FDA, temozolomide, lomustine, carmustine, everolimus, and bevacizumab 26 . Temozolomide is a newly chemotherapy approved for CNS cancer; it is often used to treat GBM tumors. TMZ is an alkylating drug to kill tumor cells via DNA damage 27 . Both everolimus and bevacizumab are targeted therapies for recurrent GBMs patients. Everolimus is a kinase inhibitor which is usually used in subependymal giant cell astrocytoma (SEGA) patients, and bevacizumab is an angiogenesis inhibitor 27-29 . Lomustine is a monofunctional alkylating drug to regulate the cell cycles and it is usually used to control recurrent GBM tumors 30 . Carmustine is a nitrosourea derivative, also known as an alkylating agent, it is used alone with implantation method or in combination with other agents to treat glioma tumors 31 . The summary of FDA-approved drugs for CNS with detailed information are shown in Table 1. Among six FDA-approved CNS drugs, only lomustine, temozolomide, carmustine, and everolimus have in vivo screening data from NCI60. Table 1. FDA approved drugs for brain tumors. Drug Approval year Mechanism of Action Target Temozolomide (NSC 362856) 1999 DNA and RNA alkylation 32 DNA 33 Lomustine (NSC 79037) 1976 DNA and RNA alkylation 34 DNA Stathimin-4 35 Carmustine (NSC 409962) 1977 DNA and RNA alkylation Enzymatic Carbamoylation 36 DNA RNA 7 Mitochondrial 36 Everolimus (NSC 733504) 2016 37 mTOR inhibitor HIF inhibitor 38 Serine/threonine kinase mTOR 38 Bevacizumab 2004 39 VEGF inhibitor 39 VEGFA 39 5. Monoamine oxidases (MAOs) 5.1. Background of MAOs Monoamine oxidases (MAOs) to locate the outer membrane of mitochondria, functioning on the catalyzation of oxidative deamination to degrade monoamine neurotransmitters and dietary monoamines. MAOs have two different isoenzymes, MAOA and monoamine oxidases B (MAOB), and both have 70% amino acid identity. The encoded genes and substrates of these two isoenzymes are different 40, 41 . Grimsby et al. showed that the human MAOA and MAOB are genes from chromosome X and span at least 60 kb, consist of 15 exons containing 527 amino acids, and exhibit identical exon-intron organization. The covalent FAD-binding site of MAOA and MAOB is at Exon 12 which is the most retained exon and it highly attaches to an enzyme cysteine 42-44 . The monoamine neurotransmitters serotonin (5-hydroxytryptamine, 5-HT), norepinephrine, and dopamine are the substrates of MAOA and phenylethylamine and benzylamine are the substrates of MAOB 40, 45 . During the degradation of serotonin, norepinephrine, and dopamine by the MAOA, H2O2 is produced at the same time. The abnormal H2O2 production generates ROS and oxidative stress that may cause DNA damage, genetic mutations and disrupt normal neuronal function and survival 46 . The roles of MAOA then were studied in neurological disorder diseases such as schizophrenia (SCZ), bipolar disorder (BPD), Parkinson’s disease (PD), and Alzheimer’s disease (AD) 44, 47, 48 . Later, many studies began to focus on studying the functions of MAOA 8 in cancer and found that high MAOA expression contributes to malignant cancer growth including breast cancer, prostate cancer, and brain cancer. 24, 49-51 . 5.2. MAOA in Brain Cancer MAOA level plays an important role in proliferation, migration, and invasion of tumor cells. Previous studies proved that increased MAOA expression was observed in glioma tissues and cell lines 24, 52 . Overexpression of MAOA enhances the production of H2O2 to activate the hypoxic microenvironment and subsequently promotes ROS and EMT, the hypoxic microenvironment associates with genetic alterations and downstream signaling pathways of glioma tumors. The elevated level of ROS in brain cancer cells may promote gene mutation and oncogenic production. Increased ROS contributes to the mutation of glioma tumor suppressor genes, TP53 and RAS. TP53 regulates the cell cycle process, and RAS controls the cell growth and proliferation, so mutations of these two genes dysregulate the normal cell cycle process and cell proliferation 53 . Otherwise, mutation of RAS also affects the downstream signaling pathway (PI3K) and causes abnormalities of RTKs, contributing to glioma cell formation and invasion. The increased level of ROS can inactivate the tumor suppressor genes, PTEN, to trigger the PI3K signaling pathways which associate with increased HIF-1 54 . EMT, which is promoted by MAOA overexpression, induces numerous growth factors and transcription factors, and it also influences the signaling pathways of glioma tumors. TWIST, SNAIL, and SLUG are the promoters of EMT since they suppress epithelial cadherin (E- cadherin) and modulate the epithelial phenotype, such as loss of E-cadherin. TWIST is a target gene of HIF-1α, and overexpression of TWIST promotes EMT; therefore, EMT 9 induces invasion and migration of tumor cells with high correlation with the TWIST and HIF-1 pathway 18, 19 . Indeed, TWIST, SNAIL, and SLUG can be promoted by WNT/ - catenin, indicating WNT/ -catenin pathway also relates to EMT. A previous study reported that WNT/ -catenin pathway contributes to GSCs formation and higher levels of WNT/ - catenin were observed in patients with malignant glioma tumors. So, EMT may cause the growth of GBMs via the WNT/ -catenin pathway 18, 55 . The NOTCH signaling pathway, which is another inducer of EMT, activates the PI3K downstream signaling pathway, and it could be proposed that EMT, which is activated by NOTCH, induces proliferation of glioma tumor cells via indirect activation of the PI3K signaling pathway 19 . Overall, the increased level of MAOA expression promotes tumor hypoxic environment, ROS production, and EMT, helping to trigger tumorigenesis, progression, and metastasis for CNS cancers 15, 52, 56 . According to the information that mention above, inhibition of overexpressed MAOA in glioma tumors might be a new therapeutic method for the treatment of CNS cancers. 5.3. MAOA Inhibitors (MAOAI) Clorgyline is a well-known irreversible and selective MAOA inhibitor, and it was used as a drug for antidepression and antianxiety 57, 58 . The structure of clorgyline is shown in the blue circle of Figure 1. Several studies have proved that higher MAOA expression in several types of cancer, and the anti-cancer effect of clorgyline was proposed. Emerging studies investigated the anticancer effect of clorgyline. The studies demonstrated that clorgyline promotes cell apoptosis, changes the status of cell’s EMT, and restores sensitivity of cancer cells to anticancer drugs via reducing MAOA levels in prostate cancer, breast cancer, colon cancer, and drug-resistant prostate cancer 24, 49-51 . 10 Subsequently, a novel MAOAI, NMI was synthesized and developed. The structure of NMI was composed of clorgyline and a near-infrared (NIR) dye MHI-148 which only can be detected in tumor cells (Figure 1). It was designed to target tumors specifically. NMI has a dual function for diagnosis and therapy for prostate and brain cancers. NIR heptamethine carbocyanine dye, MHI-148, was reported to have high affinity and selectivity to tumor cells and can be activated and regulated by organic anion-transporting polypeptide (OATPs) and hypoxia. Increased uptake of NHI-148 in tumor cells also was observed under a hypoxic microenvironment via the activation of HIF1α/OATPs signaling axis. Therefore, NIR dye MHI-148 contributes to the precise tumor-targeting effect of NMI. Accumulation of NMI was observed in the mitochondria of LNCaP cells by co-staining with the mitochondrial specific dye MitoTracker Green, indicating NMI inhibits MAOA activity in mitochondria 51 . The results of in vivo studies showed that NMI can reduce migration and invasion of prostate cancer cells 51, 59 . Compared to clorgyline, it was illustrated that NMI has better efficacy to inhibit cancer growth in prostate cancer because of its precise tumor-targeting property and high retention under hypoxia. 11 Figure 1. Chemical structure of NMI. Blue circle is the structure of MAOA inhibitor, clorgyline, red circle is the structure of NIR dye, MHI-148. Clorgyline was conjugated with MHI-148 via thioester linker. In vivo studies reported that NMI localizes in the tumor to reduce tumor growth in mice that were implanted with C4-2B cells. NMI was proved to reduce prostate-specific antigen (PSA) and MAOA activity, indicating that NMI can slow down the growth rate of prostate xenografts in nude mice 51 . A recent study also showed that MAOA inhibitors overcome drug-resistant problems in prostate cancer, proposing NMI could be used as a co-treatment for drug-resistant cancer to improve the efficacy of these drugs 60 . Moreover, NMI was studied to use alone or in combination with TMZ to treat recurrent gliomas. TMZ-resistance is a common phenomenon in glioblastoma patients because of overexpressed O 6 - methylguanine methyltransferase 32 . NMI alone or in combination with TMZ were reported to increase survival of patients with gliomas by inhibiting matrix metalloproteinase 9 (MM9) and triggering inflammatory cells. Therefore, NMI is able to cause elimination of glioma proliferation and angiogenesis, and induction of macrophage infiltration 24 . Clorgyline NIR dye 12 6. Bioinformatic Studies of the effects of NMI on CNS cancer Previous studies showed that NMI decreased the MAOA expression to treat prostate cancer and TMZ-resistant CNS cancer 24, 51 . It could be assumed that NMI also can be a potent drug for CNS cancer which also has higher MAOA expression in tumor tissues. In this study, the expression of MAOA level in glioma tissues and in normal brain tissues was analyzed in multiple databases including GTEx, EMBL-EBI, CGGA, Gent2, MERAV, Oncopression, and CRN. Three endpoints and the overall potency of NMI in CNS and prostate cancer cells lines from NCI60 were compared by using waterfall plots, heatmaps, and 3D scatter plots. The correlation between MAOA expression levels and antiproliferative activity (GI50) of NMI then was studied. Moreover, the drug activity including the tumor growth inhibition and killing of tumor cell lines (GI50, TGI, and LC50) of NMI was compared with that of FDA- approved drugs for CNS cancer which were screened by NCI60 to investigate the potency of NMI to be a novel CNS cancer drug. Four FDA-approved CNS cancer drugs (temozolomide, lomustine, carmustine, and everolimus) were chosen to be compared with NMI. The NCI60 data of NMI and the FDA-approved anti-cancer drugs were further applied to compare the response pattern to demonstrate the unique mechanism of NMI among these anti-cancer drugs. The possible mechanism of action for NMI was discussed in this paper. The significance of this study is to demonstrate the high antiproliferation of NMI as a therapeutic agent in CNS cancer cells through data analysis from different databases and provide discussion on the mechanism of NMI. 13 MATERIALS AND METHODS 1. MAOA gene expression analysis Raw data of gene expression in normal brain cell lines and in brain cancer cell lines are collected and downloaded from GETx, EMBI-EBI, CGGA, Gent2. The boxplots of MAOA expression comparison all were generated by ggplot2 in R package. 2. Screening methodology of NMI The detailed description of screening methodology is on the NCI website https://dtp.cancer.gov/discovery_development/nci-60/methodology.htm. One dose screening for compound was tested to verify a significant growth inhibition then compound was tested at five concentrations in 60 cell line panel from nine cancers. Cells lines for dose screening were grown in RPMI 1640 medium containing 5% fetal bovine serum and 2 mM L- glutamine and seeded in 96 well microtiter plates with 5,000 to 40,000 cells/well density based on the doubling time of individual cell lines. Microtiter plates were incubated for one day under 37° C, 5 % CO2, 95 % air and 100 % relative humidity environment. After one- day incubation, two microtiter plates were fixed by trichloroacetic acid (TCA) and measured as initial cell density at time zero (Tz) without addition of NMI. NMI was solubilized in dimethyl sulfoxide (DMSO) and diluted to 400-fold the desired final maximum test concentration and stored frozen before use. NMI was added into the rest plates of each cell line and plates were incubated for an additional two days at 37° C, 5 % CO2, 95 % air and 100 % relative humidity. Cells were fixed in situ by the gentle addition of 50 μl of cold 50 % (w/v) TCA and then washed by tap water five times, then air-dried at room temperature. Each cell well in plates was stained by Sulforhodamine B (SRB) solution (100 μl) at 0.4 % 14 (w/v) in 1 % acetic acid for 10 minutes at room temperature. Unbound stain was removed and bound dye was dissolved by 10 mM Trizma base. The absorbance was measured at a wavelength of 515 nm in an automated plate reader. The percentage of growth is calculated at different NMI concentrations, it allows the determination of GI50, TGI, and LC50. Three endpoints of NMI were calculated from its dose response curve. Growth inhibition of 50 % (GI50) is calculated from [(Ti-Tz)/(C-Tz)] x 100 = 50, The drug concentration resulting in total growth inhibition (TGI) is calculated from Ti = Tz. The LC50, concentration of drug resulting in a 50% reduction in the measured protein at the end of the drug treatment as compared to that at the beginning, is calculated from [(Ti-Tz)/Tz] x 100 = -50 61 . 3. Statistical analysis All the waterfall plots, 3D scatter plots and heatmap in this study were derived by using R studio (version 3.6.1). Three endpoints values within prostate and CNS cancer cell lines were normalized between -1 to 1. A Pearson’s correlation coefficient (r) was used to calculate the strength of the linear regression which associates the GI50 of MAOA and expression levels of MAOA in CNS cancer cell lines. A Kruskal-Wallis test and pairwise comparison were used to determine the statistical significance of endpoints values of tested drugs in CNS cancer cell lines. 15 4. COMPARE analysis of NMI and other drugs The detailed description of COMPARE Algorithm is on the NCI60 website (https://dtp.cancer.gov/databases_tools/docs/compare/compare_methodology.htm). The mean graph of NMI and other FDA-approved drugs were created from the set of their GI50 which were calculated from the dose response curve. The GI50 values of NMI and other compared drugs in each cell line were converted to their log10 GI50 values. The log10 GI50 values of each tested drug then were averaged, and the averaged value was subtracted by each log10 GI50 values to generate the mean graph. The COMPARE Algorithm then was applied to compare the similarity of growth pattern of selected drugs with a pairwise pearson correlation (PCC) coefficient analysis which provides an index of similarity. Analyses and the PCC were calculated by using a commercial statistical package procedure (SAS Institute Inc, Cary, NC). RESULTS 1. MAOA expression in normal tissues and tumor tissues Base on the data from Genotype-Tissue Expression (GTEx), the median range of MAOA levels in normal brain tissues is from 4.3 to 18.5 transcripts per million (TPM), but they are slightly different in different regions of the brain (Table 2). In another dataset, the Chinese Glioma Genome Atlas (CGGA), the mRNA sequencing for non-glioma patients also shows that the median MAO A level in normal brain samples is about 11 TPM, which is close to the results from GTEx (Figure 2). Since both GTEx and CGGA only show the MAOA expression in normal brain tissues; in order to know the MAOA expression in brain tumor tissues, the data from EMBL-EBI and CRN were further analyzed. The data from EMBL- 16 EBI exhibited that the average MAOA expression in glioblastoma multiforme and glioma are 18 and 31 TPM, respectively. Increased MAO A expression in glioma tumor tissues was observed compared to that in normal tissue. In GBM tumors, higher MAO A level was observed comparing to most normal tissues except in the hypothalamus and substantia nigra. The data collected from Gent2 (Figure 3a) also proved that the log2 values of MAO A expression in tumor brain tissues are slightly higher than that in normal brain tissues and a similar result also shows in the MERAV database (Figure 3b). According to the Oncopression database, increased MAOA expression in cancer tissues was only observed in brain and prostate cancer, indicating that MAOA activity in brain cancer is similar to that in prostate cancer (Figure 3c). Similarly, the CRN databases showed that MAO A levels in normal CNS tissue, 8.1 fragments per kilobase of exon per million (FPKM), are lower than that in CNS cancer tissues, 10.9 FPKM (Figure 3d). It could be demonstrated that CNS cancer and prostate cancer both have increased MAOA expression in their tumor tissues Elevated MAOA expression in gliomas compared to that in normal human astrocytes was proved by Kushal et al 24 . The initiation and progression of glioma is promoted by increased MAOA level 62 . Wu et al. also found that enhanced MAOA expression associates with prostate cancer cell growth 52 . In vivo studies showed that lower MAOA expression could inhibit prostate cancer growth and migration 51, 63 . The results from databases are consistent with the statements from literature that higher MAOA activity in prostate cancer and glioma tissues compare with MAOA expression in normal prostate cells and brain tissues. 17 Table 2. Median MAOA expression in normal brain tissues of different regions and median MAOA expression in tumor tissues. Expression values are shown in TPM (Transcripts per Million). The expression values in normal tissues are from GTEx (https://www.gtexportal.org/home/gene/MAOA) and data in tumor tissue are from EMBL-EBI (https://www.ebi.ac.uk). Figure 2.MAOA expression in non-glioma patients from CGGA. MAO A expression values are shown in TPM (transcripts per million). Raw data was downloaded from CGGA (http://cgga.org.cn/download.jsp#PartG) and the boxplot was generated by R studio. 18 Figure 3. MAO A expression in normal cells versus in tumor cells of CNS and prostate cancer. (a) Boxplot comparing MAOA expression in brain cancer and normal brain. Tissue types are color-coded: red = Brain cancer, blue = Brain-Normal. Log2 values of MAOA expression are shown in TPM. Values was downloaded from Gent2 (http://gent2.appex.kr/gent2/) and the boxplot was generated by R studio. (b) Boxplot comparing MAOA expression in normal brain tissue and primary brain tumor. Boxplot was downloaded from MERAV (http://merav.wi.mit.edu/SearchByGenes.html). Tissue types are color-coded: pink = Normal tissue light blue = Primary tumor. Log2 values of MAOA expression are shown in TPM. (c) Boxplot comparing MAOA expression in normal brain tissues versus brain cancer tissues (top) and in normal prostate tissues versus prostate cancer tissues (bottom). Boxplots were downloaded from Oncopression (http://www.oncopression.com). Statistical significance of comparison with normal were shown at the right of boxplots. (d) MAOA expression comparison of normal brain samples and glioma samples. The figure was downloaded from CRN (http://syslab4.nchu.edu.tw/CRN). Subset 1 is normal brain samples, subset 2 is stage contrast- enhancing glioma samples. Highest expression is represented by red and lowest expression is represented by blue. Expression values are shown in FPKM (fragments per kilobase of exon per million). 19 2. Activity of NMI in cancer cell lines NMI was submitted to NCI60 for dose response screening and the NCI measured the growth percent in 59 cell lines from nine different cancers at five concentrations, from 10 -4 M to 10 -8 M. The NCI60 five dose response curve of NMI is shown in below (Fig. 4). The potential growth inhibition of NMI in all cell lines can be observed in Fig. 1, which 48 cell lines had 100% growth inhibition and 9 cell lines had 50% growth inhibition at 10 μM. Figure 4. Five Dose-response curves for NMI in 59 cancer cell lines from NCI. NMI were tested at five concentration levels, 10 -4 M to 10 -8 M, in all cancer cell lines. X- axis represents the five concentration levels of NMI. Y-axis represents the percentage growth. The 100% growth represents the observed growth of cells without treatment. The 0% growth indicates no observed growth of cells, corresponding to the quantity of cells at start point. The -100% growth means all cells are killed by treatment. 20 Three endpoints, GI50, TGI, and LC50, were calculated from the NCI60 five dose response curve. GI50 (50% cell growth inhibition) corresponds to the concentration of the drug required for inhibiting 50% growth of the cells; TGI (total growth inhibition) represents the concentration required for a growth percent of 0; LC50 (50% lethal concentration) is the concentration yielding 50% cell death. The potency of drugs can be determined by the values of endpoints from the NCI60 five dose response curve. High potential drugs have lower concentrations of endpoints, which means a lower concentration of drugs were used for inhibiting cell growth inhibition and killing 50% of the cells. GI50, TGI and LC50 values for NMI were arranged and made into waterfall plots to investigate the potency of NMI further (Figure S1). The lowest to highest concentrations of endpoints in each cell lines are listed from top to bottom. The cell lines which NMI has more potency to treat are listed at the top of the plots, and the following decrease progressively. In leukemia cell lines, NMI shows great potency for inhibiting 50% cell growth, the concentrations of GI50 are lowest; but the concentrations of TGI and LC50 are high so that the potency for 100% growth inhibition and 50% cell lethal are weak. The potency of NMI is only high for GI50 but not for other two endpoints while as a treatment for leukemic cancers, so the overall potency of NMI to leukemic cancers is non-ideal. It was demonstrated previously that CNS cancer and prostate cancer have the same trend of MAOA expression in normal tissues and in tumor tissues. The endpoints of NMI only in cell lines of these two cancers were extracted and compared to illustrate the potency of NMI for CNS cancer (Figure 5). The characteristics of CNS cell lines and prostate cancers are listed in Table S1. 21 Figure 5. The waterfall plot comparing endpoints of NMI in prostate and CNS cancer cell lines. X-axis represents -log value of GI50 (a), TGI (b), and LC50 (c) and y-axis are the names of tested cell lines. The most sensitive cell lines for each endpoint of NMI are at the top and the most resistant cell lines are at the bottom. Cancer types are color-coded: blue bar= CNS cancer; pink bar= prostate cancer. For both GI50 and TGI, NMI has higher potency in all CNS cancer cell lines than in DU-145 prostate cancer cell line. Four out of six CNS cancer cell lines (U251, SNB-19, SF-539, SNB- 75) showed higher potencies to be treated by NMI compared to another prostate cancer cell line, PC3. The potency of NMI for LC50 is greater in three CNS cancer cell lines (U251, DF-539, SF-295) than in PC3. The endpoints of NMI have lower concentrations in the majority of CNS cancer cell lines compared to prostate cancer cell lines. U251 is at the first top of all waterfall plots, meaning NMI has highest potency in U251 than other CNS cancer cell lines and prostate cancer cell lines. SF-539 is at the third top in GI50 waterfall plot and second top in both TGI and LC50 waterfall plots. Thus, NMI exhibits better overall potency in CNS cancer cell lines, especially in U251 and SF-539. DU-145 SF-295 SF-268 PC-3 SNB-75 SF-539 SNB-19 U251 0 2 4 6 -Log10 GI50 (mM) cell panel CNS Prostate Waterfall plot of GI50 in PC and CNS cancer cell lines SF-268 DU-145 SNB-19 SNB-75 PC-3 SF-295 SF-539 U251 0 2 4 6 -Log10 LC50 (mM) cell panel CNS Prostate Waterfall plot of LC50 in PC and CNS cancer cell lines DU-145 SF-268 SF-295 PC-3 SNB-75 SNB-19 SF-539 U251 0 2 4 6 -Log10 TGI (mM) cell panel CNS Prostate Waterfall plot of TGI in PC and CNS cancer cell lines a. b. c. 22 Figure 6. The overall potency comparison of NMI in CNS and prostate cancer cell lines. (a) Heatmap representing normalized GI50, TGI, and LC50 of NMI in prostate and CNS cancer cell lines. The values of normalized endpoints are from -1 to 1. The heatmap used a three color (red-yellow-green) color spectrum to represent the value ranges of endpoints. The normalized values which are close to -1 tend to be red, the normalized values which are close to 1 tend to be green, whereas the normalized values which are close to 0 tend to be yellow. (b) The 3D scatter plot representing endpoints of NMI in prostate and CNS cancer cell lines. The x-axis is log(GI50), y-axis is the log(TGI), and the z-axis is the log(LC50). Each point in the plot represents the potency of NMI in each cell line. Cancer types are color-coded: blue dots = CNS cancer, pink triangles = prostate cancer. The log values of NMI endpoints in prostate and CNS cancer cell lines then were normalized on a scale from -1 to 1. The heatmap for normalized endpoints of NMI were generated (Figure 6a). The normalized values of endpoints were coded on a red-to-green color-spectrum in which the more reddish color (dark red) corresponds to the lowest normalized values and more greenish color (dark green) corresponds to the highest normalized values and the yellowish color corresponds to the middle range of the normalized values. Figure 6a shows that the normalized GI50 of NMI in U251 was in the dark orange color scale which is the darkest reddish color in all tested cell lines. The normalization of NMI in SF-539 was observed to fall 3D Scatter Plot for endpoints comparison of NMI in PC and CNS cancer cell lines -6.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 Log10 GI50 (mM) Log10 TGI (mM) Log10 LC50 (mM) CNS Prostate GI50 TGI LC50 SF-268 SF-295 SF-539 SNB-19 SNB-75 U251 PC3 DU-145 CNS Prostate -1 0 1 a. b. 23 in the light orange color scale. The normalized TGI and LC50 of NMI in U251 and SF-539 were in the darkest reddish color scale compared to all of the other tested cell lines. Among all prostate and CNS cancer cell lines, normalized GI50, TGI, and LC50 all are lowest in U251 and second lowest in SF-539. It is demonstrated that NMI has higher overall potency, as indicated by low normalized GI50, TGI, and LC50, in U251 (-0.63, -1.00, -1.00) and SF-539 (-0.39, -0.87, -0.87) than in PC3 (-0.22, -0.67, -0.57) and DU-145 (0.07, -0.15, 0.29). NMI might be a highly potent drug not only for prostate cancer but also for CNS cancers. The three-dimensional scatter plot is a better method, correlating three endpoints in one graph, for visualizing and comparing the overall potency of NMI in all cancer cell lines. The points at the bottom left are the most potent cell lines and at the top right are the least potent cell lines to be treated by NMI (Fig. S2). One of the CNS cancer cell lines, the blue point at the bottom left of the plot, shows the highest potency among all of the cell lines for nine cancers. Other CNS and prostate cancer cell lines, blue and pink points at the center of the plot, are less potent to be treated by NMI. To compare the overall potency of NMI only in prostate and CNS cancer cell lines, the 3D scatter plot for GI50, TGI, and LC50 of NMI in CNS and prostate cancer cell lines was made (Figure 6b). Blue dots represent the values of endpoints in CNS cancer cell lines and pink triangle points represent the values of endpoints in prostate cancer cell lines. Two blue dots (U251, SF-539) were observed at lower left position than the pink triangle points at the center of the plot (PC3 and DU-145). NMI has been demonstrated that it has better potency in two CNS cancer cell lines (SF-539, U251) with low GI50, TGI, and LC50, (10 -5.92 ,10 -5.57 ,10 - 5.22 ) and (10 -6.13 ,10 -5.68 ,10 -5.30 ) than in prostate cancer cell line (PC-3, DU-145), with (GI50, TGI, LC50) of (10 -5.77 ,10 -5.40 ,10 -5.02 ) and (10 -5.52 ,10 -4.96 ,10 -4.46 ). 24 3. Correlation between MAOA expression and anti-cancer potency of NMI in CNS cancer The correlation analysis of MAOA expression and efficacy of inhibiting cancer growth then was further investigated. The scatter plot which correlates the MAOA expression and log(GI50)of NMI in all tested CNS cancer cell lines was generated (Figure 7). Pearson’s correlation coefficient (r) was calculated to determine the relationship between those two variables. Both Pearson’s correlation coefficient and Spearman’s correlation coefficient are constrained between -1 to 1. An r of 1.0 identifies a perfect positive correlation, an r of -1.0 denotes a perfect negative correlation, and an r of 0 means there is no correlation between the two variables. Figure 7 shows that higher concentration of NMI with lower MAOA expression, meaning higher concentration of NMI can slightly inhibit the growth of MAOA and the anti-cancer activity of NMI possibly relates to other genes or pathways which can affect the expression of MAOA. It is a weak negative linear correlation (r = -0.35) between MAOA expression and efficacy of cancer growth inhibition by NMI. 25 Figure 7. Scatter plot of correlation between MAO A expression and NMI antiproliferative efficacy (GI50) in all CNS cancer cell lines. The regression linear has a Pearson’s correlation coefficient (r=-0.35). X-axis is Log2 values of MAO A expression, which are shown in FPKM (fragments per kilobase of exon per million); y-axis is log GI50 values of NMI. 4. Potency comparison of NMI and four FDA-approved CNS cancer drugs in CNS cancer cell lines The comparison of GI50, TGI, and LC50 values of NMI and four FDA-approved CNS cancer drugs in six CNS cancer cell lines were then analyzed by boxplots separately. The GI50 boxplot shows that the light blue color box is at the right-most position of the plot, the green color box is at the left position of the light blue color box and other color boxes are at Log 2 of MAOA expression (FPKM) Log(GI50) of NMI SF-268 SF-295 SF-539 SNB-19 SNB-75 U251 R ² = 0.1209 -6.2 -6.1 -6.0 -5.9 -5.8 -5.7 -5.6 -5.5 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 r = -0.35 26 the left position of the green box. This means that everolimus (light blue box) has the lowest concentration of GI50 and NMI has the second low concentration of GI50 among all five drugs in all CNS cancer cell lines (Figure 8a). The Kruskal-Wallis test was chosen to determine if there are significances among all tested drugs, and if the p-value is less than 0.05, which means at least two groups have a significant difference without knowing which two groups. The Kruskal-Wallis p-value of the GI50 boxplot is 3.5e-05, indicating at least two drugs have a significant difference among all five drugs. The GI50 of NMI also shows significant differences (**) compared to that of other drugs, which indicated that NMI needs a lower concentration to inhibit 50% tumor cell growth than the majority of FDA-approved CNS drugs in the CNS panel. The TGI of NMI has the greatest potency compared to all FDA-approved CNS cancer drugs in all cell lines (Figure 8b). The green color box is at the right-most position of the plot and other color boxes are at the left of the green color box. The Kruskal-Wallis p-value of the TGI boxplot is 9.8e-05 and the TGI of NMI has significant differences compared to the TGI of four CNS drugs. The boxplot of LC50 (Fig. 8c) also shows the same result that green color box is at the rightest position of the plot, meaning NMI has the lowest concentration of LC50 compared to that of four CNS drugs. The Kruskal-Wallis p-value of TGI is 1.1e-05, and it shows that there are significant differences between NMI and the three CNS drugs except for everolimus. These results illustrated that the GI50, TGI, and LC50 of NMI are better than the majority of FDA- approved drugs, indicating the high overall potency of NMI as an anticancer drug for CNS cancer. 27 Figure 8. The endpoints analysis of NMI and FDA-approved CNS drugs from NCI60. (a) Boxplot of GI50 comparison of NMI and CNS drugs. (b) Boxplot of TGI comparison of NMI and CNS drugs. (c) Boxplot of LC50 comparison of NMI and CNS drugs. Drugs are color-coded: pink = lomustine (NSC 79307); blue green = temozolomide (NSC 362856); orange = carmustine (NSC 409962); light blue = everolimus (NSC 733504); light green = NMI (NSC 791228). * p  0.05; ** p  0.01. The three endpoints’ values of NMI and four FDA-approved anti-CNS cancer drugs from NCI60 were changed to heatmap analysis for further comparison (Figure 9). The values of the endpoints were coded on a red to green color-spectrum in which the more reddish color (dark red) corresponds to the lowest concentration of endpoints and more greenish color (dark green) corresponds to the highest concentration of endpoints and the orange color correspond to the middle range of the concentration. The GI50 results of everolimus and NMI in all CNS cancer cell lines were observed in red and orange color range, which contrasted with the results of lomustine, temozolomide, and carmustine, in the yellow and green color range. Interestingly, only the TGI results of NMI were observed in the orange color range (orange, light orange) in all tested cell lines, and the TGI results of other drugs were contrasted, which were in the green and yellow color range. The same results were observed in LC50, in that only the results of NMI were in the orange and yellow color ranges in the majority of tested Kruskal-Wallis, p = 3.5e-05 ** ** ** ** 362856 409962 733504 79037 791228 -9 -8 -7 -6 -5 -4 Log10 GI50 (mM) nsc nsc 362856 409962 733504 79037 791228 GI50 Comparison of NMI vs. FDA-approved CNS Drugs Kruskal-Wallis, p = 9.8e-05 ** ** * ** 362856 409962 733504 79037 791228 -7 -6 -5 -4 Log10 TGI (mM) nsc nsc 362856 409962 733504 79037 791228 TGI Comparison of NMI vs. FDA-approved CNS Drugs Kruskal-Wallis, p = 1.1e-05 * ** ** ns 362856 409962 733504 79037 791228 -8 -7 -6 -5 -4 Log10 LC50 (mM) nsc nsc 362856 409962 733504 79037 791228 LC50 Comparison of NMI vs. FDA-approved CNS Drugs a. b. c. 28 cell lines except for SF-268. The results of the other drugs were in the yellow and green color range. NMI has a better potency in terms of GI50, TGI, and LC50 in CNS cancer cell lines than most of these four other CNS cancer drugs. The overall potency of NMI and four CNS cancer drugs were then visualized by 3D scatter plot (Figure 10). The points at the bottom left are the most potent drugs and at the top right are the least potent drugs. Each drug was represented by different colors, and each cell line was represented by different symbol dots. The 3D-scatter plot shows that most points of NMI are at the bottom left and some points of everolimus are at the left of the plot but higher than NMI. The overall potencies of other drugs in all tested cell lines are at the top right of the plot. The result illustrated that NMI has the highest overall potency, as indicated by the lowest concentration of GI50, TGI, and LC50, in U251 and SF-539. Thus, NMI was demonstrated to be a great potent drug to treat CNS cancer alone or in combination with other agents. 29 Figure 9. Heatmap representing GI50, TGI, and LC50 of NMI and four CNS cancer drugs in CNS cancer cell lines. The log10 values of endpoints are from -3.6 to -8.0. The heatmap used a three color (red-yellow- green) color spectrum to represent the value ranges of endpoints. The lowest concentration of endpoints tends to be red, while the highest concentration of endpoints tends to be green, whereas the yellow color represents the middle ranges of concentration. 30 Figure 10. 3D scatter plot representing endpoints of all tested drugs in CNS cancer cell lines. The x-axis is log(GI50), y-axis is the log(TGI), and the z-axis is the log(LC50). The drugs are color-coded: pink = lomustine (NSC 79307); blue green = temozolomide (NSC 362856); orange = carmustine (NSC 409962); blue = everolimus (NSC 733504); light green = NMI (NSC 791228). Cell lines are represented by different symbols (explanation below figure title). Overall potency of CNS drugs and NMI in CNS cell lines -8 -7 -6 -5 -4 -5.5 -5.0 -4.5 -4.0 -3.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 Log10 GI50 (mM) Log10 TGI (mM) Log10 LC50 (mM) 79037 362856 409962 733504 791228 SF-268 SF-295 SF-539 SNB-19 SNB-75 U251 31 5. Therapeutic index comparison Therapeutic index (TI) is a measurement, a ratio of GI50 and LC50, to know the relative safety of a drug. Higher TI means the ratio of efficacy and toxicity of the drug is larger so that the drug is safer. Table 3 shows that TI of lomustine, temozolomide, and carmustine in all tested cell lines are low (TI < 10), indicating that these three drugs are dangerous. Interestingly, the TI of everolimas in all tested cell lines are very high (TI > 100), which means that the efficacy of everolimas is non-ideal. NMI has a great therapeutic index (10<TI<100) in SF-268, SNB-19, and SNB-75 cell lines, which are contracted to SF-295, SF-539, and U251 with low TI (<10). Table 3. Raw data of GI50, TGI, LC50 and therapeutic index (TI) values of NMI and four FDA-approved drugs in CNS cancer cell lines. The unit of cellular inhibition values is M, and with no unit of TI. *The GI50 and TGI are higher than 100 M, the TI was calculated as both GI50 and TGI are at 100 M. Drugs Cell lines SF-268 SF-295 SF-539 SNB-19 SNB-75 U-251 NMI (NSC 791228) GI50 TGI LC50 TI 1.84 6.75 100.00 54.4 1.91 4.06 8.63 4.5 1.20 2.70 6.10 5.1 1.12 3.64 28.18 25.2 1.30 3.85 13.77 10.6 0.73 2.11 5.04 6.9 Lomustine (NSC 79037) GI50 TGI LC50 TI 21.04 57.94 100.00 4.8 91.83 100.00 100.00 1.1 16.41 34.04 70.47 4.3 100.00 100.00 100.00 1.0 23.33 64.57 100.00 4.3 15.78 51.29 100.00 6.3 Temozolomide (NSC 362856) GI50 TGI LC50 TI 100.00 100.00 100.00 1.0 100.00 100.00 100.00 1.0 100.00 100.00 100.00 1.0 100.00 100.00 100.00 1.0 >100.00 100.00 100.00 1.0 100.00 100.00 100.00 1.0 Carmustine (NSC 409962) GI50 TGI LC50 TI >100.00 98.4 >100.00 1.0 >100.00 >100.00 >100.00 1.0 >100.00 92.68 >100.00 1.0 >100.00 >100.00 >100.00 1.0 >100.00 >100.00 >100.00 1.0 >100.00 66.37 >100.00 1.0 Everolimus (NSC 733504) GI50 TGI LC50 0.27 21.09 53.95 0.01 17.26 62.95 0.01 17.34 41.88 0.25 19.50 45.19 0.01 14.09 38.90 0.01 18.11 42.66 32 TI 199.8 6295 4188 180.8 3890 4266 6. Mechanism prediction and comparison by using COMPARE Algorithm The COMPARE algorithm was provided by Developmental Therapeutics Program’s anticancer screening program to compare whether compounds have similar growth inhibition pattern (GI50) for treating cancer cell lines. The GI50 value from dose response curves for each drug were converted to the mean graph pattern which exhibits the sensitivity and resistance of compounds to the cancer cell lines and then compared the mean graph pattern of each compound to find out the similarity of their activity profiles. The Pairwise Pearson Correlation Coefficient (PCC) of each drug was calculated as a measurement of pattern similarity. A PCC of 1.0 identifies a perfect match, a PCC of -1.0 denotes a perfect mirror image, whereas a PCC of 0 means there is no correlation between the two patterns. High PCC (> 0.8) means two drugs have similar patterns, which indicates that they have similar mechanisms in treating these cancer cell lines 64 . Since NMI outperforms existing anti-cancer drugs, the mechanism of tumor growth inhibition of NMI is important to explore as an anti-CNS cancer drug. The possible mechanism of NMI was investigated by using the COMPARE algorithm and compared to the mean graph pattern of four FDA-approved CNS anti-cancer drugs with PCC. The results of PCC of NMI and FDA- approved anticancer drugs were converted and visualized in a heatmap (Figure 10). The green color range corresponds to low PCC values (<0.6), the yellow color range corresponds to the middle PCC values range (0.6<PCC<0.8), and the red color range corresponds to high PCC values (>0.8). Figure 10 shows that the PCC values of NMI and each of the four CNS anti- 33 cancer drugs are all in green (< 0.6) (shown in the last row of Fig. 10), even with lomustine having the highest PCC (0.48) to NMI, is still in the green color range. The PCC comparisons exhibit all low PCC values among NMI and FDA-approved CNS drugs, indicating that NMI has a unique mechanism to inhibit CNS cancer cell growth compared with traditional CNS drugs. Figure 11. Heatmap comparing PCC of NMI and four FDA-approved CNS cancer drugs. The range of PCC is from -1 to 1. High PCC (> 0.8) are shown in yellow to red color, low to middle range of PCC (< 0.6) are shown in green color and negative range of PCC are 34 highlighted in black. The PCC analysis of NMI and other tested drugs are in the last row of the heatmap. The name and NSC number of drugs are listed beside of the heatmap. DISCUSSION 1. Possible mechanism of NMI 1.1. Drugs with similar growth inhibition activities of NMI CellMiner, a web-based bioinformatics tool, can be applied to analyze data from NCI60. The pattern comparison under NCI60 Analysis Tools from this web-application allows to compare mean graph patterns of all drugs from NCI60 and find out the drugs have similar activity patterns 65 . Previous results showed that the mechanism of NMI is not similar to conventional CNS cancer drugs, it was assumed that the mechanism of NMI may be similar to other drugs. To figure out the possible mechanism of NMI, the web-tool CellMiner was applied to identify the compounds with high PCC (>0.72). Table 4 lists the information about the compounds with high PCC. The highest PCC (0.80) of NMI was reported for NSC 720622 which is one of the derivatives of Pyrazoloindole; however, few findings are about NSC 720622. The study of another Pyrazoloindole derivative, GS-2, was found. The only difference between the structure of NSC 720622 and GS-2 is an additional nitrogen dioxide in NSC 720622. The antitumor activities of GS-2 have been demonstrated by increasing intracellular ROS expression to cause strong topo genes (topo I/II) suppression, DNA damage, reduction of mitochondrial membrane potential, and caspase-3/9 activation 66 . Since NSC 720622 exhibits high PCC to NMI, NMI may use the similar mechanism of NSC 720622 to inhibit tumor growth. The PCC of ONX-0914 is the second highest, which is 0.77. Several studies 35 illustrated that ONX-0914, a tripeptide epoxyketone, is the immunoproteasome inhibitor for 5i and 2i. Upregulation of nuclear respiratory factor 1 (Nrf-1) and immunoproteasomes were observed in brain cancers and the 2i subunit of the proteasome impacts the expression of Nrf-1. The inactivation of Nrf1 occurs during the process of inhibiting 2i subunit by ONX-0914. Meanwhile, a previous study also exhibited that the ONX-0914 in combination with proteasome inhibitors (PIs) can regulate the ubiquitin proteasome pathway by decreasing the expression of NF-kB and inducing apoptosis 67, 68 . Both Nrf-1 and NF-kB also have been reported to relate to the prostate cancer growth. Nrf-1 mediates androgen receptor signaling which is induced by MAOA in prostate cancer and NF-kB associates with the induction of EMT which contributes to prostate cancer growth 52, 69, 70 . It can be proposed that the mechanism of NMI may correlate to immunoproteasome inhibition to inhibit prostate and CNS cancer growth by reducing Nrf-1 and NF- kB. Olivomycin, a well-known DNA- binding antibiotic drug, has a high PCC (0.76) to NMI. An in vivo study revealed that the characteristics of olivomycin in anti-cancer activities include tumor cell apoptosis induction and suppression of p53-induced transcriptional genes and other cellular stressors which disrupt DNA damage by forming complexes with DNA 71, 72 . NMI for prostate and brain tumor inhibition may associate with increased tumor cell apoptosis and decreased transcriptional genes in the p53 pathway. The structural study of phyllanthoside illustrated that it acts like a tRNA able to inhibit eukaryotic ribosome and protein synthesis by targeting the catalytic sites in the ribosome. The ribosome, which locates at the surface of endoplasmic reticulum membrane, controls secretory and integral membrane protein synthesis in several diseases and cancers. Dysregulated ribosome and protein synthesis were demonstrated as a factor to trigger cancers 72-74 . Protein synthesis, a very general important function in cells, 36 contributes to the proliferation not only in the normal cells but also in the tumor cells. The deficiency of tumor suppressors like p53 and activation of inflammatory cytokines like IL6 and oncogenes like MYC may stimulate ribosome biogenesis and protein synthesis 75 . NMI may share a similar mechanism to phyllanthoside to kill cancer cells by regulating the proteins which involve ribosome biogenesis and protein synthesis. A handful of information and data about NSC 800374 and NSC 795502 compounds, which are substituted quinazolinone compounds, were found. Thakur, et. al. found out that a series of substituted quinazolinone-based derivatives are dual inhibitors of PI3K/ histone deacetylases (HDAC). The inhibition of the PI3K pathway and inactivation of HDACs are beneficial for cancer therapy through affecting downstream genes and regulating tumor suppressor and transcription factors. The study exhibited that some substituted quinazolinone have good potency in several cancer lines including multiple mutant and resistant cell lines due to necrosis induction 76 . Since there are limited studies about NSC 800374 and NSC 795502, it is difficult to understand their precise mechanisms of action. Roridin A and verrucarin A (VC-A), a macrocyclic toxin, is a type D trichothecenes. Several studies showed the high anticancer activity of macrocyclic trichothecenes against multiple cancers including prostate cancer. In the study from Liu, et. al, VC-A was proved to inhibit prostate cancer growth through reducing tumor proliferation and activating cell cycle arrest in the AKT/NF- kB/mTOR signaling pathway. Downregulations of Bcl-2, NF-kB, and cyclin-dependent kinases (CDK4 and CDK6) in prostate cell lines were observed after treating with VC-A 77, 78 . This result is consistent with the study of gene expression profiling of NMI-treated samples, indicating that NMI might inhibit proliferation and trigger apoptosis by affecting genes and proteins in the downstream AKT/NF-kB/mTOR signaling pathway 51 . Besides, 37 roridin A and VC-A show a response to macrophages due to TNF-  induction. These two compounds also activate inflammatory cytokines such as IL-6 and enhance the proinflammatory response, contributing to reduce proliferation and angiogenesis of cancer 79 . These findings of roridin A and VC-A are also similar to the results that increased macrophage infiltration in NMI-treated glioma from Kushal, et. al. Dinaciclib is a selective cyclin-dependent kinase inhibitor with a moderate PCC (0.72) to NMI. Parry, et. al. showed an excellent antiproliferative activity of dinaciclib in prostate and brain tumor cell lines without a selectivity for inhibiting specific tumors by reducing Mcl-1 80 . It has been reported that dinaciclib is a dose-dependent CDK inhibitor to inhibit cell proliferation without apoptosis in glioma tumors. Interestingly, increasing cell death and the activation of mitochondrial pathway were observed after dinaciclib combined with a Bcl-2 inhibitor in glioma cells. It could be hypothesized that NMI increases apoptosis and decreases cell proliferation by inhibiting Mcl-1 subsequently reducing Bcl-2 to inhibit prostate and CNS cancer growth, which is also consistent with the results of gene expression study from Wu et al 81 . Bouvardin has been reported as a protein synthesis inhibitor in eukaryotic cells. Bouvardin inhibits protein synthesis by increasing the effect of ionizing radiation and, it also exhibits significant inhibitory activity in a xenograft model 82 . The previous study also showed that the growth of a xenograft tumor was inhibited by NMI significantly in PCa xenograft mouse models, assuming that NMI might be a protein synthesis inhibitor that synergizes with the radiation effect 59 . Interestingly, some studies showed that higher ROS expression induces p53-dependent apoptosis; however, the in vivo study proved that ROS which is induced by MAOA can 38 cause apoptosis inhibition and autophagy activation in neuroendocrine prostate cancer cells 15, 83 . Paradoxically, NSC 720622, which has the highest PCC to NMI, inhibits tumor cells by increasing intracellular ROS expression to cause apoptosis. Indeed, intracellular ROS mediate both apoptosis and anti-apoptosis depending on their level, NADPH oxidase (NOX) and mitochondrial electron transport chain (mETC) are two main supporters for ROS production. Increasing ROS moderately by MAOA overexpression activates several genes in the downstream signaling pathways contributing to tumor cell survival; a mild increased level of ROS which is generated by NOX, also involves in multiple signaling pathways and transcription factors to induce tumor cell proliferation and inhibit apoptosis. However, high levels of ROS were produced by mETC to achieve tumor cell apoptosis. Lin, et. al. also showed that low levels of ROS which promoted by MAOA may cause accumulation of p53 in the mitochondrial matrix because of the reaction between glutathione disulfide (GSSG) and the thiol groups of p53 83 . Another study demonstrates that ROS derived from NOX correlates to arsenite-induced HIF-1  stabilization 84 . A hypothesis could be proposed that increased MAOA causes mild elevated ROS generation and subsequently induces glutathionylation of p53, GSSG, and stabilization of HIF-1, these processes may also correlate to increased NOX activity. Actually, NOX is a significant controller of complex I (CI) which is one of transmembrane proteins of the electron transport chain in mitochondrial. The role of CI is to transfer electrons from NADPH to ubiquinone. A previous study in human neuroblastoma cells showed that blocking CI can promote ROS production which mediates hypoxia; however, decreased MAOA level contributes to reduce ROS levels and activates complex I 85, 86 . It could be proposed that higher MAOA levels may disrupt the 39 normal roles of CI by interacting with NADPH. NMI may have significant effects on this route. Autophagy has also been proved to mediate cancer cell growth and death, which is similar to ROS 87 . It could be assumed that mildly increased ROS may relate to autophagy activation triggering tumor cell proliferation and highly increased ROS may cause apoptosis- independent cell death via autophagy. In addition, it could be hypothesized that NMI may be a regulator of NOX-induced ROS and it might have multiple mechanisms of action to inhibit cancer growth. One could be reducing ROS meanwhile increasing apoptosis and inhibiting autophagy. Another mechanism might decrease ROS which is generated by NADPH through inhibiting MAOA and transcription factors. Then it is possible to activate higher levels of intracellular ROS which is produced by mETC to promote apoptosis and necrosis. The levels of ROS in NMI-treated prostate cancer and glioma cancer could be further studied. Table 4. Compounds with high PCC with NMI. Compounds NSC number PCC Mechanism of Action 4-(4- Dimethylaminobenzyliden)-1- methyl-6-nitro-2-phenyl-4H- pyrazolo indolium trifluoromethanesulfonate 720622 0.80 Topoisomerase (topo) inhibitor ONX-0914 762152 0.77 Immunoproteasome inhibitor Olivomycin 76411 0.76 P53-induced transcription genes inhibitor Tumor cell apoptosis Phyllanthoside 342443 0.76 Protein synthesis inhibitor Eukaryotic ribosome inhibitor Transcription/ translation inhibitor (S)-4-(((2-(1-((2-Amion-6- Methylpyrimidin-4- yl)amino)propyl)-4-oxo-3- phenyl-3,4-Dihydroquinazolin- 800374 0.76 PI3K/HDAC dual inhibitor 40 5-yl)amino)methyl)-N- hydroxybenzamide Roridin A Verrucarin A 327993 0.76 AKT/NF-kB/mTOR inhibitor (S)-4-(((2-(1-((9H-Purin-6-yl) amino)propyl)-4-oxo-3-phenyl- 3,4-Dihydroquinazolin-5- yl)amino)methyl)-N- hydroxybenzamide 795502 0.73 PI3K/HDAC dual inhibitor Dinaciclib 800091 0.72 Cyclin-Dependent Kinase inhibitor Bouvardin 259968 0.72 Protein synthesis inhibitor Transcription/ translation inhibitor 1.2. Correlation of other genes and NMI In order to figure out the connection between MAOA and genes or proteins mentioned above, the “gene interaction” web-based tool from UCSC Genome Browser was used to provide a brief picture of genes connected to of MAOA. The top 25 related genes are in Figure 12. The black color-coded circles represent cancer genes. NDRG1 is the only cancer gene in prostate cancer which relates to MAOA. NDRG1 correlates multiple downstream genes and proteins of prostate and brain cancer, such as TP53, CDH1, MYCN, and NFKB1 (Figure S2). Other transcription factors and proteins like NR3C1, GAPDH, TCF25 also strongly correlate with MAOA and they have interactions with NF1, BCL 2, AKT1/2/3, HIF1A, TNF, CTNNB1, and MAPK family proteins which correlates with EGFR and VEGFA-associated prostate and glioma tumor invasion (Figure S3, S4, S5). 41 Figure 12. Top 25 interacting genes of MAOA from UCSC Genomes Browser. Figure was downloaded from the UCSC Genomes Browser (https://genome.ucsc.edu/index.html). Blue color-corded circles are proteins and black color- coded circles are cancer genes. The previous results showed that the correlation of MAOA level and growth inhibition of NMI is weak. It is assumed that the mechanism of NMI might correlate with other genes which have indirect connections with MAOA. Wu, Et. al. found that not only expression of MAOA was changed after treating with NMI but also other genes, including upregulation of TP53, CDKN1A, CDKN2A, and CDH1 as well as downregulation of proto-oncogenes (FOS, JUN, NFKB1, MYC), cell-cycle regulator genes (CCND1, CCNE1, CDK4, and CDK6), anti- apoptotic gene (BCL2), and genes which involve in downstream signaling (VIM, SNAI1, SNAI2, TWIST1, VEGFA, GLUT1, IL6, IL8, MMP2, MMP9, MET). A previous in vivo study proved that loss of E-cadherin is one of the characteristics of EMT caused by an elevation of MAOA expression in PC3 cell line. In addition, elimination of E-cadherin caused by decreased miR-203 expression subsequently to activate EMT in drug-resistant 42 GBM cells have been demonstrated. NMI might be a potent treatment for recurrent GBM tumors to inhibit EMT via inducing the gene of E-cadherin, CDH1 52, 88 . The CDKN1A and CDKN2A are cell cycle regulators. NMI promotes the expression of these two genes to inhibit the progression of cell cycle and eliminate tumor cell proliferation via DNA damage. These two genes strongly correlate with the p53 pathway, an important pathway for gliomas, since both CDKN1A and CDKN2A encoded p21 protein which is a target of p53. Increased TP53 also was observed in the NMI-treated sample compare to the control samples, indicating that the mechanism of NMI involves in the induction of apoptosis. Meanwhile, the BCL-2 gene, a p53 transcriptional gene, also showed elimination after treating with NMI, which also involved in the p53 pathway to reduce activities of anti-apoptosis 52, 89 . Otherwise, the elevated genes of downstream signaling were demonstrated to occur frequently in glioma tumors. NMI can regulate these genes to inhibit CNS tumor growth. MET contributes to tumor cell proliferation and angiogenesis so that downregulation of this gene can diminish tumor cell growth and invasion. The genes correlating to EMT and the HIF-1  pathway decreased after NMI treatment indicating that the mechanism of NMI is to inhibit cancer cell proliferation and increase apoptosis of tumor cells 51 . The antitumor activities of NMI match that of high PCC compounds which were mentioned in the previous section, suggesting that NMI might have similar mechanisms to those drugs. 2. The potency of NMI may be different for different stages of gliomas Various stages of gliomas associate with different levels of MAOA expression. The data from CGGA and Gent2 show that MAOA expression is higher in low-stage (II/III) gliomas than in the high-stage (IV) gliomas. Highest median MAOA expression are exhibited in grade II gliomas and lowest in grade IV gliomas in samples from CGGA (Figure 13a). These 43 data are consistent with the result in grade II gliomas, and the lowest median MAOA expression is exhibited in grade I and grade IV gliomas in samples from Gent2 (Figure 13b). In addition, other factors also may affect the MAOA expression such as IDH status, 1p/19 co-deletion status, and primary status. Figure 14 (from CGGA data) shows that increased median MAOA level in IDH mutant, 1p/19q codel, and primary gliomas compare with IDH wildtype (a), 1p/19q non-codel (b), and recurrent gliomas (c). Therefore, it could be proposed that the efficacy of NMI might be better in stage II gliomas, and the antiproliferative activity of NMI may be affected by the factors mentions above. Further, an in vivo study of NMI in primary brain cancer with different stages, IDH status, and 1p/19 co-deletion status of gliomas could be developed and extended. Figure 13. MAO A expression in different stages of brain cancers from CGGA and Gent2. (a) Boxplot of MAOA expression in various stages of gliomas from CGGA. MAOA expression are shown in log2 values of expression. Different stages are color-coded: pink = WHO II; green = WHO III; blue = WHO IV. (b) Boxplot of MAOA expression in various stages of gliomas from Gent2. Y-axis is the log2 value of MAOA expression level and x-axis is the stage of glioma tumors. 44 45 Figure 14. MAO A expression in different IDH mutation, 1p/19q co-deletion, and progression status of brain cancers from CGGA. (a) Boxplot of MAOA expression in wildtype and IDH mutant brain cancers. Different status are color-coded: pink = IDH mutant; green = Wildtype. (b) Boxplot of MAOA expression in 1p/19q codel and 1p/19q non-codel brain cancers. Different status are color-coded: pink = 1p/19q codel; green = 1p/19q non-codel. (c) Boxplot of MAOA expression in primary and recurrent brain cancers. Different status are color-coded: pink = Primary; green = Recerrent. MAOA expression are shown in log2 values of expression. CONCLUSION This study shows that the high potency of NMI against CNS cancer cell lines and strong anticancer activity of NMI outperforms and other FDA-approved CNS drugs. It was also found that NMI uses a unique mechanism of action to inhibit CNS cancer growth compared with conventional CNS drugs. The results of this study provide a strong basis for further development with NMI as a potential candidate as a CNS anti-cancer drug. 46 REFERENCES 1. Vasilev A, Sofi R, Tong L, Teschemacher A, Kasparov S. In Search of a Breakthrough Therapy for Glioblastoma Multiforme. Neuroglia. 2018;1(2):292-310. doi:10.3390/neuroglia1020020 2. Ostrom QT, Fahmideh MA, Cote DJ, et al. Risk factors for childhood and adult primary brain tumors. Neuro-Oncology. 2019;21(11):1357-1375. doi:10.1093/neuonc/noz123 3. Shan FY, Zhong D, Hu W, et al. Neoplasms of Central Nervous System: A Diagnostic Approach. InTech; 2018. 4. Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathologica. 2016;131(6):803-820. doi:10.1007/s00401-016-1545-1 5. Ostrom QT, Cioffi G, Gittleman H, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2012–2016. Neuro- Oncology. 2019;21(Supplement_5):v1-v100. doi:10.1093/neuonc/noz150 6. Wen PY, Reardon DA. Progress in glioma diagnosis, classification and treatment. Nature Reviews Neurology. 2016-02-01 2016;12(2):69-70. doi:10.1038/nrneurol.2015.242 7. Van Meir EG, Hadjipanayis CG, Norden AD, Shu HK, Wen PY, Olson JJ. Exciting New Advances in Neuro-Oncology: The Avenue to a Cure for Malignant Glioma. CA: A Cancer Journal for Clinicians. 2010;60(3):166-193. doi:10.3322/caac.20069 8. Harrison RA, De Groot JF. Cell Signaling Pathways in Brain Tumors. Topics in Magnetic Resonance Imaging. 2017;26(1):15-26. doi:10.1097/RMR.0000000000000112 9. Cheng F, Guo D. MET in glioma: Signaling pathways and targeted therapies. Journal of Experimental and Clinical Cancer Research. 2019;38(1):1-13. doi:10.1186/s13046-019-1269-x 10. Gordon MS. Vascular endothelial growth factor as a target for antiangiogenic therapy. Journal of Clinical Oncology. 2000;18(21 SUPPL.):1000-1017. 11. Tabernero J. The Role of VEGF and EGFR Inhibition: Implications for Combining Anti– VEGF and Anti–EGFR Agents. Molecular Cancer Research. 2007;5(3):203-220. doi:10.1158/1541-7786.mcr-06-0404 12. Herzog B, Pellet-Many C, Britton G, Hartzoulakis B, Zachary IC. VEGF binding to NRP1 is essential for VEGF stimulation of endothelial cell migration, complex formation between NRP1 and VEGFR2, and signaling via FAK Tyr407 phosphorylation. Molecular Biology of the Cell. 2011;22(15):2766-2776. doi:10.1091/mbc.e09-12-1061 13. Cohen AL, Holmen SL, Colman H. IDH1 and IDH2 Mutations in Gliomas. Current Neurology and Neuroscience Reports. 2013;13(5)doi:10.1007/s11910-013-0345-4 14. Monteiro A, Hill R, Pilkington G, Madureira P. The Role of Hypoxia in Glioblastoma Invasion. Cells. 2017;6(4):45. doi:10.3390/cells6040045 15. Perillo B, Di Donato M, Pezone A, et al. ROS in cancer therapy: the bright side of the moon. Experimental and Molecular Medicine. 2020;52(2):192-203. doi:10.1038/s12276-020- 0384-2 16. Bell EL, Klimova TA, Eisenbart J, Schumacker PT, Chandel NS. Mitochondrial Reactive Oxygen Species Trigger Hypoxia-Inducible Factor-Dependent Extension of the Replicative Life Span during Hypoxia. Molecular and Cellular Biology. 2007;27(16):5737-5745. doi:10.1128/mcb.02265-06 47 17. Aggarwal V, Tuli H, Varol A, et al. Role of Reactive Oxygen Species in Cancer Progression: Molecular Mechanisms and Recent Advancements. Biomolecules. 2019;9(11):735. doi:10.3390/biom9110735 18. Colella B, Faienza F, Di Bartolomeo S. EMT Regulation by Autophagy: A New Perspective in Glioblastoma Biology. Cancers. 2019;11(3):312. doi:10.3390/cancers11030312 19. Iwadate Y. Epithelial-mesenchymal transition in glioblastoma progression. Oncology Letters. 2016;11(3):1615-1620. doi:10.3892/ol.2016.4113 20. Chen S-C, Liao T-T, Yang M-H. Emerging roles of epithelial-mesenchymal transition in hematological malignancies. Journal of Biomedical Science. 2018;25(1)doi:10.1186/s12929- 018-0440-6 21. Aldape K, Brindle KM, Chesler L, et al. Challenges to curing primary brain tumours. Nature Reviews Clinical Oncology. 2019;16(8):509-520. doi:10.1038/s41571-019-0177-5 22. Claes A, Idema AJ, Wesseling P. Diffuse glioma growth: a guerilla war. Acta Neuropathologica. 2007;114(5):443-458. doi:10.1007/s00401-007-0293-7 23. Bhowmik A, Khan R, Ghosh MK. Blood brain barrier: A challenge for effectual therapy of brain tumors. BioMed Research International. 2015;2015doi:10.1155/2015/320941 24. Kushal S, Wang W, Vaikari VP, et al. Monoamine oxidase A (MAO A) inhibitors decrease glioma progression. Oncotarget. 2016;7(12):13842-13853. doi:10.18632/oncotarget.7283 25. Osuka S, Van Meir EG. Overcoming therapeutic resistance in glioblastoma: the way forward. Journal of Clinical Investigation. 2017;127(2):415-426. doi:10.1172/jci89587 26. Drugs Approved for Brain Tumors. National Cancer Institute. Accessed February 3, 2021. https://www.cancer.gov/about-cancer/treatment/drugs/brain#1 27. Carter TC, Medina-Flores R, Lawler BE. Glioblastoma Treatment with Temozolomide and Bevacizumab and Overall Survival in a Rural Tertiary Healthcare Practice. BioMed research international. 2018;2018:1-10. doi:10.1155/2018/6204676 28. Current Treatments for Brain Tumors. 2017. Accessed February 16, 2021. https://braintumor.org/wp-content/assets/2017_NBTS_CurrentTreatmentOptions_083017.pdf 29. Hasskarl J. Everolimus. Springer Berlin Heidelberg; 2014:373-392. 30. Weller M, Le Rhun E. How did lomustine become standard of care in recurrent glioblastoma? Cancer Treatment Reviews. 2020;87:102029. doi:10.1016/j.ctrv.2020.102029 31. Xiao Z-Z, Wang Z-F, Lan T, et al. Carmustine as a Supplementary Therapeutic Option for Glioblastoma: A Systematic Review and Meta-Analysis. Frontiers in Neurology. 2020;11doi:10.3389/fneur.2020.01036 32. Lee SY. Temozolomide resistance in glioblastoma multiforme. Genes & Diseases. 2016;3(3):198-210. doi:10.1016/j.gendis.2016.04.007 33. Temozolomide. DrugBank. February 19, 2020. https://go.drugbank.com/drugs/DB00853 34. GLEOSTINE (lomustine) capsules, for oral use (2016). 35. Lomustine. DrugBank. Accessed February 19, 2020. https://go.drugbank.com/drugs/DB01206 36. Carmustine. DrugBank. Accessed February 19, 2020. https://go.drugbank.com/drugs/DB00262 37. Everolimus (Afinitor). U.S Food and Drug Administration. Accessed February 19, 2020. https://www.fda.gov/drugs/resources-information-approved-drugs/everolimus-afinitor 38. Everolimus. DrugBank. Accessed February 19, 2020. https://go.drugbank.com/drugs/DB01590#BE0002386 48 39. Bevacizumab. DrugBank. Accessed February 19, 2020. https://go.drugbank.com/drugs/DB00112 40. Shih JC, Chen K, Ridd MJ. MONOAMINE OXIDASE: From Genes to Behavior. Annual Review of Neuroscience. 1999;22(1):197-217. doi:10.1146/annurev.neuro.22.1.197 41. Bortolato M, Chen K, Shih JC. Monoamine oxidase inactivation: From pathophysiology to therapeutics ☆. Advanced Drug Delivery Reviews. 2008;60(13-14):1527-1533. doi:10.1016/j.addr.2008.06.002 42. Grimsby J, Chen K, Wang LJ, Lan NC, Shih JC. Human monoamine oxidase A and B genes exhibit identical exon-intron organization. Proceedings of the National Academy of Sciences of the United States of America. 1991;88(9):3637-3641. doi:10.1073/pnas.88.9.3637 43. Gaweska H, Fitzpatrick PF. Structures and mechanism of the monoamine oxidase family. BioMolecular Concepts. 2011;2(5):365-377. doi:10.1515/bmc.2011.030 44. V R, Husain Rs A. MAOA Gene Associated with Aggressive Behavior in Humans. Journal of Down Syndrome & Chromosome Abnormalities. 2017;03(01)doi:10.4172/2472- 1115.1000120 45. McCauley R, Racker E. Separation of two monoamine oxidases from bovine bran. Molecular and cellular biochemistry. 1973;1(1):73-81. doi:10.1007/bf01659940 46. Naoi M, Riederer P, Maruyama W. Modulation of monoamine oxidase (MAO) expression in neuropsychiatric disorders: genetic and environmental factors involved in type A MAO expression. Journal of Neural Transmission. 2016;123(2):91-106. doi:10.1007/s00702- 014-1362-4 47. Riederer P, Laux G. MAO-inhibitors in Parkinson's Disease. Experimental Neurobiology. 2011;20(1):1-17. doi:10.5607/en.2011.20.1.1 48. Takehashi M, Tanaka S, Masliah E, Ueda K. Association of monoamine oxidase A gene polymorphism with Alzheimer's disease and Lewy body variant. Neuroscience letters. 2002;327(2):79-82. doi:10.1016/s0304-3940(02)00258-6 49. Satram-Maharaj T, Nyarko JNK, Kuski K, et al. The monoamine oxidase-A inhibitor clorgyline promotes a mesenchymal-to-epithelial transition in the MDA-MB-231 breast cancer cell line. Cellular signalling. 2014;26(12):2621-2632. doi:10.1016/j.cellsig.2014.08.005 50. Zhao H, Flamand V, Peehl DM. Anti-oncogenic and pro-differentiation effects of clorgyline, a monoamine oxidase A inhibitor, on high grade prostate cancer cells. BMC Medical Genomics. 2009;2(1):55. doi:10.1186/1755-8794-2-55 51. Wu JB, Lin T-P, Gallagher JD, et al. Monoamine Oxidase A Inhibitor–Near-Infrared Dye Conjugate Reduces Prostate Tumor Growth. Journal of the American Chemical Society. 2015;137(6):2366-2374. doi:10.1021/ja512613j 52. Wu JB, Shao C, Li X, et al. Monoamine oxidase A mediates prostate tumorigenesis and cancer metastasis. Journal of Clinical Investigation. 2014;124(7):2891-2908. doi:10.1172/jci70982 53. Rinaldi M, Caffo M, Minutoli L, et al. ROS and Brain Gliomas: An Overview of Potential and Innovative Therapeutic Strategies. International Journal of Molecular Sciences. 2016;17(6):984. doi:10.3390/ijms17060984 54. Cully M, You H, Levine AJ, Mak TW. Beyond PTEN mutations: the PI3K pathway as an integrator of multiple inputs during tumorigenesis. Nature Reviews Cancer. 2006;6(3):184-192. doi:10.1038/nrc1819 55. Sandberg CJ, Altschuler G, Jeong J, et al. Comparison of glioma stem cells to neural stem cells from the adult human brain identifies dysregulated Wnt- signaling and a fingerprint 49 associated with clinical outcome. Experimental Cell Research. 2013;319(14):2230-2243. doi:10.1016/j.yexcr.2013.06.004 56. Koundouros N, Poulogiannis G. Phosphoinositide 3-Kinase/Akt Signaling and Redox Metabolism in Cancer. Frontiers in Oncology. 2018;8doi:10.3389/fonc.2018.00160 57. Finberg JPM, Rabey JM. Inhibitors of MAO-A and MAO-B in Psychiatry and Neurology. Frontiers in Pharmacology. 2016;7doi:10.3389/fphar.2016.00340 58. Lipper S, Murphy DL, Slater S, Buchsbaum MS. Comparative behavioral effects of clorgyline and pargyline in man: A preliminary evaluation. Psychopharmacology. 1979;62(2):123-128. doi:10.1007/bf00427124 59. Wu JB, Shao C, Li X, et al. Near-infrared fluorescence imaging of cancer mediated by tumor hypoxia and HIF1α/OATPs signaling axis. Biomaterials. 2014;35(28):8175-8185. doi:10.1016/j.biomaterials.2014.05.073 60. Wang K, Luo J, Yeh S, et al. The MAO inhibitors phenelzine and clorgyline revert enzalutamide resistance in castration resistant prostate cancer. Nature Communications. 2020;11(1)doi:10.1038/s41467-020-15396-5 61. NCI-60 Screening Methodology. Updated August 26, 2015. Accessed February 17, 2020. https://dtp.cancer.gov/discovery_development/nci-60/methodology.htm 62. Li P-C, Chen S-Y, Xiangfei D, Mao C, Wu C-H, Shih JC. PAMs inhibits monoamine oxidase a activity and reduces glioma tumor growth, a potential adjuvant treatment for glioma. BMC Complementary Medicine and Therapies. 2020;20(1)doi:10.1186/s12906-020-03041-z 63. Xu S, Adisetiyo H, Tamura S, et al. Dual inhibition of survivin and MAOA synergistically impairs growth of PTEN-negative prostate cancer. British Journal of Cancer. 2015;113(2):242-251. doi:10.1038/bjc.2015.228 64. Holbeck SL, Collins JM, Doroshow JH. Analysis of Food and Drug Administration– Approved Anticancer Agents in the NCI60 Panel of Human Tumor Cell Lines. Molecular Cancer Therapeutics. 2010;9(5):1451-1460. doi:10.1158/1535-7163.mct-10-0106 65. Reinhold WC, Sunshine M, Liu H, et al. CellMiner: A Web-Based Suite of Genomic and Pharmacologic Tools to Explore Transcript and Drug Patterns in the NCI-60 Cell Line Set. Cancer Research. 2012;72(14):3499-3511. doi:10.1158/0008-5472.can-12-1370 66. Ji YY, Zhu YM, Wang JW. GS-2, a pyrazolo[1,5-a]indole derivative with inhibitory activity of topoisomerases, exerts its potent cytotoxic activity by ROS generation. Environmental toxicology and pharmacology. 2013;36(3):1186-1196. doi:10.1016/j.etap.2013.09.019 67. Narayanan S, Cai C-Y, Assaraf YG, et al. Targeting the ubiquitin-proteasome pathway to overcome anti-cancer drug resistance. Drug resistance updates. 2020;48:100663. doi:10.1016/j.drup.2019.100663 68. Bhawe K, Felty Q, Yoo C, Roy D. Abstract 2441: Association between NRF1 regulatory gene networks and cancer and aging. Bioinformatics, Convergence Science, and Systems Biology. 2019:2441-2441. Proceedings: AACR Annual Meeting 2019; March 29-April 3, 2019; Atlanta, GA. doi:10.1158/1538-7445.AM2019-2441 69. Schultz MA, Hagan SS, Datta A, et al. Nrf1 and Nrf2 Transcription Factors Regulate Androgen Receptor Transactivation in Prostate Cancer Cells. PLoS ONE. 2014;9(1):e87204. doi:10.1371/journal.pone.0087204 70. Chua HL, Bhat-Nakshatri P, Clare SE, Morimiya A, Badve S, Nakshatri H. NF-κB represses E-cadherin expression and enhances epithelial to mesenchymal transition of mammary epithelial cells: potential involvement of ZEB-1 and ZEB-2. Oncogene. 2007;26(5):711-724. doi:10.1038/sj.onc.1209808 50 71. Simonova VS, Samusenko AV, Filippova NA, et al. Olivomycin Induces Tumor Cell Apoptosis and Suppresses p53-Induced Transcription. Bulletin of Experimental Biology and Medicine. 2005;139(4):455-459. doi:10.1007/s10517-005-0321-3 72. Needham RJ, Bridgewater HE, Romero-Canelón I, Habtemariam A, Clarkson GJ, Sadler PJ. Structure-activity relationships for osmium(II) arene phenylazopyridine anticancer complexes functionalised with alkoxy and glycolic substituents. Journal of Inorganic Biochemistry. 2020;210:111154. doi:10.1016/j.jinorgbio.2020.111154 73. Chan J. Eukaryotic protein synthesis inhibitors identified by comparison of cytotoxicity profiles. RNA. 2004;10(3):528-543. doi:10.1261/rna.5200204 74. Garreau De Loubresse N, Prokhorova I, Holtkamp W, Rodnina MV, Yusupova G, Yusupov M. Structural basis for the inhibition of the eukaryotic ribosome. Nature. 2014-09-01 2014;513(7519):517-522. doi:10.1038/nature13737 75. Gilles A, Frechin L, Natchiar K, et al. Targeting the Human 80S Ribosome in Cancer: From Structure to Function and Drug Design for Innovative Adjuvant Therapeutic Strategies. Cells. 2020;9(3):629. doi:10.3390/cells9030629 76. Thakur A, Tawa GJ, Henderson MJ, et al. Design, Synthesis, and Biological Evaluation of Quinazolin-4-one-Based Hydroxamic Acids as Dual PI3K/HDAC Inhibitors. Journal of Medicinal Chemistry. 2020;63(8):4256-4292. doi:10.1021/acs.jmedchem.0c00193 77. Liu Y, Gao X, Deeb D, et al. Mycotoxin verrucarin A inhibits proliferation and induces apoptosis in prostate cancer cells by inhibiting prosurvival Akt/NF-kB/mTOR signaling. Journal of experimental therapeutics and oncology. 2016;11(4):251-260. 78. Wu Q, Wang X, Nepovimova E, et al. Trichothecenes: immunomodulatory effects, mechanisms, and anti-cancer potential. Archives of Toxicology. 2017-12-01 2017;91(12):3737- 3785. doi:10.1007/s00204-017-2118-3 79. Kankkunen P, Rintahaka J, Aalto A, et al. Trichothecene Mycotoxins Activate Inflammatory Response in Human Macrophages. The Journal of Immunology. 2009;182(10):6418-6425. doi:10.4049/jimmunol.0803309 80. Parry D, Guzi T, Shanahan F, et al. Dinaciclib (SCH 727965), a Novel and Potent Cyclin- Dependent Kinase Inhibitor. Molecular Cancer Therapeutics. 2010;9(8):2344-2353. doi:10.1158/1535-7163.mct-10-0324 81. Jane EP, Premkumar DR, Cavaleri JM, Sutera PA, Rajasekar T, Pollack IF. Dinaciclib, a Cyclin-Dependent Kinase Inhibitor Promotes Proteasomal Degradation of Mcl-1 and Enhances ABT-737-Mediated Cell Death in Malignant Human Glioma Cell Lines. Journal of Pharmacology and Experimental Therapeutics. 2016;356(2):354-365. doi:10.1124/jpet.115.230052 82. Gladstone M, Frederick B, Zheng D, et al. A translation inhibitor identified in a Drosophila screen enhances the effect of ionizing radiation and taxol in mammalian models of cancer. Disease Models & Mechanisms. 2012;5(3):342-350. doi:10.1242/dmm.008722 83. Lin Y-C, Chang Y-T, Campbell M, et al. MAOA-a novel decision maker of apoptosis and autophagy in hormone refractory neuroendocrine prostate cancer cells. Scientific Reports. 2017;7(1):46338. doi:10.1038/srep46338 84. Li Y-N, Xi M-M, Guo Y, Hai C-X, Yang W-L, Qin X-J. NADPH oxidase-mitochondria axis-derived ROS mediate arsenite-induced HIF-1α stabilization by inhibiting prolyl hydroxylases activity. Toxicology letters. 2014;224(2):165-174. doi:10.1016/j.toxlet.2013.10.029 51 85. Zhao RZ, Jiang S, Zhang L, Yu ZB. Mitochondrial electron transport chain, ROS generation and uncoupling (Review). International journal of molecular medicine. 2019;doi:10.3892/ijmm.2019.4188 86. Fitzgerald JC, Ugun‐Klusek A, Allen G, et al. Monoamine oxidase‐A knockdown in human neuroblastoma cells reveals protection against mitochondrial toxins. The FASEB Journal. 2014;28(1):218-229. doi:10.1096/fj.13-235481 87. Azad MB, Chen Y, Gibson SB. Regulation of Autophagy by Reactive Oxygen Species (ROS): Implications for Cancer Progression and Treatment. Antioxidants & redox signaling. 2009;11(4):777-790. doi:10.1089/ars.2008.2270 88. Liao H, Bai Y, Qiu S, et al. MiR-203 downregulation is responsible for chemoresistance in human glioblastoma by promoting epithelial-mesenchymal transition via SNAI2. Oncotarget. 2015;6(11):8914-8928. doi:10.18632/oncotarget.3563 89. Yin S, Van Meir EG. p53 Pathway Alterations in Brain Tumors. Humana Press; 2009:283-314. 52 APPENDICES Table S1. Characteristic of CNS and prostate cancer cell lines from NCI60. 53 54 Figure S1. The waterfall plot comparing endpoints of NMI in cell lines from nine types cancers of NCI60. X-axis represents -log value of GI50 (a), TGI (b), and LC50 (c) and y-axis are the names of tested cell lines. The name of each cell line is listed in y-axis. The most sensitive cell lines for each endpoint of NMI are at the top and the most resistant cell lines are at the bottom. Cancer types are color-coded: red = Breast cancer; blue = CNS cancer; green = Colon cancer; purple = Leukemia; orange = Melanoma; yellow = Non-small cell lung cancer; brwon = Ovarian cancer; pink = prostate cancer; grey = Renal 55 Figure S2. Top 25 interacting genes of NDRG1. Blue color-corded circles are proteins and black color-coded circles are cancer genes. Figure S3. Top 25 interacting genes of NR3C1. Blue color-corded circles are proteins and black color-coded circles are cancer genes. RPN2 GSR NDRG1 PPAP2C ACTG1 ACTG2 ASS1 CKB DARS MYL6 VDR ARAF CPP PHYHIP NFKB1 RELA TP53 IGKC EGF EGR1 MTOR MYCN DDX39B HSPA5 CDH1 TBX21 CARM1 NR3C1 GTF2F1 GTF3C3 GTF3C1 GTF3C4 XRCC5 PRL EGR1 SP1 NF1 BDP1 HSD11B1 ANXA1 NTS GSK3B IFNG BGLAP GATA3 FOXA3 FOXA2 ARHGAP35 ACTB ARID1A 56 Figure S4. Top 25 interacting genes of GAPDH. Blue color-corded circles are proteins and black color-coded circles are cancer genes. Figure S5. Top 25 interacting genes of TCF25. Blue color-corded circles are proteins and black color-coded circles are cancer genes. ELANE GAPDH HSPA6 HSPA5 HSPA4 SUMO1 HSPA9 SUMO2 SUMO4 HIF1A INS RAB2A AKT2 AKT3 SRC PARP1 PPP1CB BCL2 PPA1 DBN1 IGKC SGPL1 SMG1 IGHG2 AKT1 MAPK14 TCF25 TNF MAPK15 MAPK10 MAPK11 MAPK12 MAPK13 WNT4 MAPK3 MAPK1 WNT3 WNT2 WNT1 WNT6 WNT16 WNT11 RELA EPHB2 IFNG IRF6 IL4 NFKB1 CTNNB1 INS

Abstract (if available)

Abstract Our previous studies have shown that higher monoamine oxidase A (MAOA) expression is in drug-resistant recurrent gliomas, MAOA inhibitor NMI (Near-infrared dye conjugate MAOA Inhibitor) shows antiproliferative activities and reduced gliomas growth. This study examined the potency of NMI on additional CNS cancer cell lines by NCI60 screening data analysis. ❧ Literature search and database analysis revealed that CNS cancer has increased MAOA expression in tumor tissues than in normal tissues, which is similar to prostate cancer. The potency, growth inhibition, and lethal doses of NMI in CNS cancer and prostate cancer were studied by using waterfall plots, 3D scatter plots, and heatmaps. The results of this study show that NMI is more sensitive to CNS (U251, SNB-19, SF-539, SNB-75, SF295) cancer cell lines than prostate cancer cell lines (PC3, DU145) base on NMI waterfall plots of GI50, TGI, and LC50. Two CNS cancer cell lines, SF-539 and U251, showed higher overall potency to NMI treatment than other CNS and prostate cancer cell lines. The linear regression between MAOA expression and GI50 of NMI shows a weak negative correlation (r = −0.35), which means the antitumor activity of NMI in the CNS cancer also correlates to other genes probably. The drug activities and therapeutic index of NMI and four FDA-approved CNS cancer drugs were compared in NCI60 CNS cancer cell lines, NMI exhibited the best antitumor activity among all tested drugs. The Pairwise Pearson Correlation Coefficient (PCC) showed that NMI has a unique mechanism compared to the traditional CNS drugs. This study shows that NMI may be a candidate drug for CNS cancer with high antitumor activity and multiple mechanisms that are different from existing CNS drugs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Bioinformatics analysis of the anti-cancer potency of NMI on non-small cell lung cancer and its potential mechanism

PDF

NMI (near-infrared dye conjugate MAO A inhibitor) outperformed FDA-approved prostate cancer drugs with a unique mechanism based on bioinformatic analysis of NCI60 screening data

PDF

NMI: a near infrared conjugated MAO-A inhibitor as a novel targeted therapy for colorectal and other cancers

PDF

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

PDF

Co-expression of monoamine oxidase A and prostate cancer stem cell markers in Pten knockout mice

PDF

Deregulation of CD36 expression in cancer presents a potential targeting therapeutic opportunity

PDF

Monoamine oxidase inhibitors regulate tumorigenesis and mitochondrial function in a prostate cancer mouse model

PDF

MAO a deficient mice exhibit an altered immune system in the brain and prostate

PDF

Monoamine oxidase and cancer

PDF

Comparing drug release kinetics of rapamycin bound FKBP-ELP fusion proteins

PDF

Potential therapeutic effect of monoamine oxidase (MAO) inhibitor on human neuroblastoma

PDF

Inhibition of MAO-A by Dual MAO-A/HDAC inhibitors: in silico approach for ligand binding and affinity prediction

PDF

Characterization of IL-1β secretion by fusing elastin-like polypeptides to pro-caspase-1

PDF

Targeting cellular redox modulations for pancreatic cancer treatment

PDF

The role of translocator protein (TSPO) in mediating the effects of mitotane in human adrenocortical carcinoma cells

PDF

Genome engineering of filamentous fungi for efficient novel molecule production

PDF

Discovery of novel small molecules for ovarian cancer treatment

PDF

Chimeric Antigen Receptor targeting Prostate Specific Membrane Antigen (PSMA)

PDF

Characterization of actin based motility in mammalian cells through LIM and SH3 domain protein 1 (LASP1) and elastin like polypeptide (ELP) fusion protein

PDF

Development of engineered antibodies as novel anti-cancer agents

Asset Metadata

Creator Feng, Qianhua (author)

Core Title Study of a novel near-infrared conjugated MAOA inhibitor, NMI, against CNS cancer by NCI60 data analysis

School School of Pharmacy

Degree Master of Science

Degree Program Pharmaceutical Sciences

Publication Date 04/25/2021

Defense Date 04/22/2021

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

Tag CNS cancer,GI50,glioma,LC50,MAOA,NCI60,NMI,OAI-PMH Harvest,TGI

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Haworth, Ian (committee member), Okamoto, Curtis (committee member), Shih, Jean (committee member)

Creator Email fancyyyfenggg@gmail.com,qianhuaf@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-449482

Unique identifier UC11668586

Identifier etd-FengQianhu-9527.pdf (filename),usctheses-c89-449482 (legacy record id)

Legacy Identifier etd-FengQianhu-9527.pdf

Dmrecord 449482

Document Type Thesis

Rights Feng, Qianhua

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