Current & Past Odyssey Fellows

Hussein Abbas

Odyssey Fellow (2014-2017)
Department of Molecular & Cellular Oncology

Bioinformatic analyses to understand the interplay of the p53 pathway in tumorigenesis

p53 is mutated in more than 50% of human cancers and plays a critical role in tumor suppression.  p63 and p73 are part of the p53 family and have multiple isoforms with different functions. The isoforms can be placed into two groups: the TA isoforms, which structurally resemble p53 and are thought to act as tumor suppressors, and the ∆N isoforms, which bind to p53, TAp63, and TAp73 and inhibit their function, thus acting as oncogenes. One alternative strategy to overcome p53 loss is to manipulate the p53 family members, p63 and p73, which interact with p53. It was found that deleting the ∆N isoforms of p63 and p73 could compensate for p53 loss. Further, p63 regulates Dicer and DGCR8, which are prominent effectors of microRNAs. Hence, it is hypothesized that non-coding RNAs regulated by p63 and p73 can be used to therapeutically target the p53 pathway.. Further, the expression and role of p63 and p73 isoforms in human tumors is not fully deciphered. In addition to identifying downstream non-coding RNA targets of p63 and p73, TCGA database to characterize p63 and p73 in human cancers will also utilized and identify potential therapeutic nodes in the network regulated by these transcription factors.

Wantong Yao

Odyssey Fellow (2014-2017)
Department of Genomic Medicine

Targeting Syndecan-1 in Pancreatic Cancer

Pancreatic ductal adenocarcinoma (PDAC) represents the forth leading cause of cancer death, with an estimated 43,920 new cases and 37,390 deaths in 2012 in the US alone. Median survival averages six months and only 5% of all patients survive beyond 5 years. PDAC arises from one of three precursor lesions, but it most frequently derives from pancreatic intraepithelial neoplasia (PanIN). The progression of PanIN lesions from minimally to severely dysplastic and ultimately to PDAC is accompanied by an accumulation of gene alterations that involves activation of oncogenes and dysfunction of tumor suppressors.This study sought to combine basic and clinical research to understand more deeply pancreatic cancer and further elucidate the roles and the underlying mechanism of SDC1 in PDAC initiation, progression, and maintenance, and to conduct preclinical studies to determine whether monoclonal antibody (mAb) targeting of SDC1 may exhibit anti-tumor effects..

Tony Gutschner, Ph.D.

Odyssey Fellow (2014-2017)
Department of Genomic Medicine

RNA Epigenetics — From Discovery to Therapeutic Application

The discovery of catalytic RNAs, so called ribozymes, in the 1980s and more recently the discovery of a phethora of functional non-coding RNAs have completely changed our view of RNA. Besides its "passive" role as mediator molecule to assist decoding of the genomic information to generate the "active" proteome, RNA itself controls gene expression and protein function by multiple mechanisms. However, in analogy to post-translational protein modifications like phosphorylation or methllation, cellular RNAs (ribosoma (rRNAs), transfer (tRNAs), messenger RNAs (mRNAs), small nuclear RNAs (snRNAs) and others contain more than a hundred structurally distinct post-transcriptional modifications. Although many of these modifications have been known for decades, the enzymes that post-transcriptional RNA modifications are commonly thought to have adaptive roles that can fine-tune structures and functions of mature transcripts to influence gene expression. However, recent studies show that some post-transcriptional RNA modifications can be dynamic and might have regulatory roles. Therefore, understanding the scope and mechanisms of post-transcriptional RNA modifications, which might alter genomic information and thus could be called 'RNA epigeneics', representa a new area in RNA biology research.

Jessica Galloway-Peña, Ph.D.

Odyssey Fellow (2014-2017)
Department of Infectious Diseases

Use of Microbiome Analyses To Predict and Treat Bacteremia Associated Sepsis in Patients with Acute Myelogenous Leukemia/Myelodysplastic Syndrome (AML/MDS)

Cancer patients have a higher likelihood of developing septicemia as a result of immunosuppression or neutropenia. Septicemia causes mortality directly or indirectly by restricting treatment options. Little progress has been made in the treatment of infections in cancer patients and multi-drug resistance often renders prophylactic and empirical antibiotic therapy ineffective. An increasing recognition of the key role the microbiome plays in infectious diseases and cancer has steered research to the premise that certain flora prevent colonization with pathogenic bacteria. In the advent of the next generation sequencing, it is now able to feasibility perform rapid and economic large scale DNA sequencing and whole genome analyses of patient microbiomes. This proposed research will test the hypothesis that the changes in patient microbiomes contribute to infection susceptibility and mortality seen in AML/MDS patients. Moreover, the objective is to accurately identify patients at high-risk for infections, predict impending infections, and offer alternatives to antimicrobial therapy.

Narkhyun Bae, Ph.D.

Odyssey Fellow (2014-2017)
Department of Molecular Carcinogenesis

Using a Protein Array Approach to Comprehensively Characterize Arginine Methylation

For cells to respond to environmental cues, they make use of a process called signal transduction. This process is mediated by a vast array of postranslational modifications (PTMs), which initiate at the cell membrane, travels through the cytophasm into the nucleus, and culminates in the deposition of chromatin modifications that epigenetically regulate cellular function. The characterization of the enzymes that catalyze these modifications and the proteins that recognize and transduce specific PTMs to biological outcomes remains largely underdeveloped, particularly in the field of protein methylation. This study will focus on one key aspect of the methylome - Protein Arginine Methylation. The arginine motifs that is recognized by the different Protein Arginine Methyltransferases (PRMT) are poorly characterized, and in some case not yet identified. Moreover, very few "readers" (effector molecules) for methylarginine motifs have been identified. To aid in the better characterization of PRMTs and the discovery of effectors, it is proposed to develop a new approach utilizing redundant protein peptide arrays that represent all possible linear motifs for potential recognition by different PRMTs or effectors.

Lindsay Kelderhouse, Ph.D.

Odyssey Fellow (2014-2017)
Department of Cancer Systems Imaging

Developing Novel Imaging Agents Against Specific in vivo Targets

Development of novel imaging agents against specific in vivo targets is has become vital for the success of many targeted therapeutic agents. Currently, lead compounds are optimized into targeted imaging agents through multiple rounds of screening and medicinal chemistry prior to in vivo validation. Unfortunately, for many promising lead compounds, poor in vivo efficacy often hinders further translation into preclinical and clinical settings.  To circumvent this problem, we propose to use in vivo directed evolution to select high affinity peptide ligands against extracellular targets that specifically accumulate in tumor and inflammatory microenvironments.  To increase the probability of obtaining high affinity compounds with the requisite stability for in vivo applications, we have chosen to use mRNA display, a powerful biological display platform that can generate over 1 quadrillion unique peptide sequences for directed evolution.  In vivo mRNA display selections will allow us to identify high affinity peptides that bind specifically to tumors and sites of inflammation within a biological context without the need for iterative in vitro design steps. Once established, this technique can potentially provide a method to rapidly generate high affinity imaging agents that have been pre-selected for enhanced biostability and specific tumor uptake in living organisms.

Tracy W. Liu, PhD

Odyssey Fellow (2013-2016)
Department of Cancer Systems

Imaging single cell and their signaling cascades initiated during metastasis in vivo using window chamber models

It is now possible to image single cells, cell-cell interactions and molecular changes within a microenvironment using intravital imaging. However, intravital imaging techniques are invasive with small fields of views and limited applicability in relevant bone marrow sites, such as the spine and long bones, due to their thickness. Recently, femur and spinal chambers have been developed which allow for longitudinal studies (several months) of the bone marrow without any motor deficits or functional loss. The combination of intravital single cell imaging and bone window chamber models provides a dynamic imaging strategy that will begin to disentangle the complexity of metastasis in heterogeneous in vivo systems; for example, we can begin to identify molecular and structural changes that occur to create a favorable niche for metastases, signaling molecules that home metastatic cells to the bone and cell-cell interactions that promote metastatic formation. This strategy in concert with single cell isolation and genetic profiling may further identify relevant metastatic processes. Once established, this strategy may provide a means to identify novel metastatic targets and evaluate the effects of therapy potentially driving clinical translation.

Amanda Say, Ph.D. 

Odyssey Fellow (2013-2014)
Department of Experimental Radiation Oncology

Knock-in of low-molecular weight cyclin E into somatic cells and mice

Cyclin E is a key regulatory protein controlling the G1 to S phase transition in mammalian cells. In breast cancer, cyclin E is overexpressed and drives cells into the S phase more rapidly, resulting in increased proliferation. Cyclin E is processed at its amino-terminus by the serine protease elastase to generate LMWE isoforms of cyclin E that appear specifically in cancer cells. The LMW-E forms of cyclin E have an increased interaction with CDK2 and are resistant to CDK inhibitors. This enables the tumorigenic cells to bypass the G1/S phase checkpoint, resulting in increased tumorigenesis in both human mammary epithelial cells and in transgenic mouse models. To date, studies involving LMW-E tumorigenesis have relied mainly on the ectopic overexpression of LMW-E in cells with functional endogenous full-length cyclin E expressed in the background. These studies neglect to identify the direct contribution LMW-E provides to tumor formation as overexpression of full-length cyclin E alone does not result in tumorigenesis and could be obscuring the complete tumorigenic role of LMW-E in vivo. Preliminary studies support this idea as an increase cytoplasmic cyclin E, identified as LMW-E due to the loss of the nuclear localization sequence in the amino-terminus of LMW-E isoforms, correlates with an increase of poor prognosis in patients. This indicates full-length cyclin E expression maintains a level of stability in patients. The proposed plan is to generate both cell-line and mouse knock-in models of the LMW-E forms to examine the direct result of LMW-E expression in these model systems.

Hunain Alam, Ph.D.

Odyssey Fellow (2013-2016)
Department of Molecular & Cellular Oncology

Defining the Oncogenic Role of the Epigenetic Modifier KDM2A in Non-Small Cell Lung Cancer

Lung cancer is the leading cause of cancer-related deaths in the United States, and as much as 85% of the newly diagnosed lung cancers are non-small-cell lung cancer (NSCLC). Despite great advance in the targeted therapy, mortality rate of lung cancer has not been significantly improved during the last three decades. Only 60% of affected individuals survive for five years mostly because of the neoplastic heterogeneity. Thus an identification of new druggable targets will pave an alternative way for the development of novel strategies to treat NSCLC. This project will study the tumor-promoting function of KDM2A in NSCLC. Based on preliminary results, the central hypothesis is that dysregulated KDM2A may repress tumor suppressor genes by altering H3K36 methylation profiles at their gene promoters and consequently promotes NSCLC. The proposed study will delineate the mechanism underlying the oncogenic role of KDM2A using genome-wide mapping approaches (e.g. chromatin immunoprecipitation sequencing) and genetic mouse models. The proposed study will also provide novel molecular insights into how dysregulation of an epigenetic modifier is coupled to NSCLC events and may prove useful in development of new therapeutic agents, such as small molecule inhibitors of KDM2A for NSCLC.

Yuhui Jiang, Ph.D.

Odyssey Fellow (2013-2016) 
Department of Neuro-Oncology

The mechanisms of PKM2-regulated cell cycle progression

Tumor-specific pyruvate kinase M2 (PKM2) is essential for the Warburg effect. Besides its well-established role in aerobic glycolysis, PKM2 directly regulates gene transcription and G1-S phase transition. However, whether PKM2 plalys a role in regulating other phases of cell cycle in unknown. Preliminary results show that PKM2 depletion blocked the association of Bub3/Bub1 SAC proteins to kinetochore and disrupted proper chromosome-to-spindle attachments leading to abnormal chromosome segregation in its kinase activity-dependent manner. This proposed study aims to elucidate the mechanisms underlying PKM2-regulated cell mitosis progression and examine the significance of this regulation in EGFR-promoted tumorigenesis. PKM2-phosphorylated Bub3 Tyr residues will be identified and the role of PKM2-dependent Bub3 Tyr phosphorylatin in regulation of Bub1 functions, the recruitment of Bub3/Bub1 and microtubules to kinetochore, and chromosome segregation will be examined. The effect of regulation of PKM2-dependent mitosis on EGFR-promoted tumor cell growth, cell cycle progression, transformation and glycolysis in cultured tumor cells and EGFR activation-promoted brain tumorigenesis in mice will also be examined. 

Junjie Li, Ph.D.

Odyssey Fellow (2013-2016)
Department of Cancer Systems Imaging

Dual Photothermal Ablation-Chemotherapy in Orthotopic and Subcutaneous Rat Models of Hepatocellular Carcinoma (HCC)

The incident of hepatocellular carcinoma (HCC) and death caused by HCC continue to rise in the United States. Though surgical resection and liver transplantation offers the hope for cure, few patients meet the criteria due to their existing intrinsic liver disease or a scarcity of donor liver. Conventional chemotherapy and radiotherapy have shown very little efficacy with inoperable HCC. New therapies that combine multimodality treatments are desperately needed to improve tumor response and prolong survival in patients with HCC. The focus of this research is on developing novel therapeutic strategies to enhance anti-tumor efficacy. The investigation of necrosis avid contrast agents (NACAs) constitutes potential strategies not only for detecting various diseases using MRI and Nuclear imaging, but also for developing therapeutic agents to target malignancies. Since most current cancer therapies may leave tumor residues left behind coupled with close-by necrosis, this study will sought to make necrosis as a single target common to all solid tumors, which can be easily achieved by using noninvasive vascular disrupting agent (VDA), minimally invasive radiofrequency ablation (RFA) or most other anticancer therapies.

Wanding Zhou, Ph.D.

Odyssey Fellow (2013-2016)
Department of Bioinformatics & Computational Biology

Quantitative analysis of tumor sequencing data

Tumorigenesis is typically accompanied by many waves of somatic mutations. Understanding these somatic mutations and their roles in cancer development requires not only the knowledge of what genes have mutated but also when and how much. This research tackles the problem of accurately measuring the fraction of mutated genes---the variant allele frequency---from Next Generation Sequencing data which is extracted from tumor samples. Automated, systematic and unbiased methods for such purpose are crucial to the downstream etiological analyses on which depends the promise of personalized cancer therapy. The work aims at both developing and implementing these methods to support scientific and clinical applications.

Xin-Jian Li, Ph.D.

Odyssey Fellow (2013-2015)
Department of Neuro-Oncology

Deciphering the role of phosphoglycerate kinase 1 in tumor development

Coordinating the regulation of glycolysis and tricarboxylic acid (TCA) cycle, a feature of tumor cells, plays a key role in cancer cell metabolism and proliferation. Understanding the precise cellular signaling regulating cell metabolism provides multiple potential therapeutic targets for cancer treatment. This study will elucidate the role of phosphoglycerate kinase 1 (PGK1), which catalyzes 1, 3-Bisphosphoglycerate to phosphoglycerate and generates ATP, in coordinating glycolysis and TCA cycle and tumorigenesis. Investigation of the regulations of PGK1 in response to growth factor and hypoxia stimulation and identify the effect of these regulations of PGK1 on glycolysis and TCA cycle, which in turn affect the tumor cell proliferation and tumor progression will be conducted. The proposed study will also identify novel functions of the key glycolytic enzyme PGK1 and may provide a molecular target for treatment of cancer.

Linda Odenthal-Hesse, Ph.D.

Odyssey Fellow (2013-2013)
Department of Molecular Carcinogenesis

Mechanism of mutagenesis and genome instability during the meiotic cell cycle

Changes in the genetic makeup of an individual, in the form of mutations and genomic rearrangements, are a common fingerprint of cancer cells as well as inherited diseases. This project will study the processes causing genome instability to increase the general understanding of cancer etiology. Intriguingly, genomes experience substantial instability during each meiotic cell cycle – much of which is induced by the repair of programmed DNA double-strand breaks by homologous recombination. Further, a significant proportion of an individual’s mutation burden can be attributed to de novo germ line point mutations. To date it is not known which proportion of these mutations arose as a consequence of recombination during meiosis, and which proportion arose prior to meiosis during stem cell proliferation. Using mouse meiosis as a model system, new molecular assays to definitively determine at high resolution the frequency and distribution of homologous recombination and point mutations during formation of sperm cells are being established. Several mouse mutants will be used to test current models to gain fundamental insights into the mechanisms of mutagenesis and genome instability. 

Youngbok Lee, Ph.D.

Odyssey Fellow (2013-2014)
Department of Cancer Systems Imaging

Real Time In Vivo Metabolic Profiling of Cancer Using Hyperpolarized Imaging Agents 

The goal of this research is to develop a new cancer diagnostic platform based on real-time in vivo metabolic imaging by Dynamic Nuclear Polarization (DNP), which capitalizes on the significant gains in sensitivity and selectivity. Magnetic Resonance Imaging (MRI) of hyperpolarized nuclear spins affords a signal enhancement of several orders of magnitude when compared to conventional MRI technique. Thus, the hyperpolarization techniques, in particular DNP, make a contribution to overcoming sensitivity and selectivity limitations of the typical MRI method through signal enhancements in their measurements, allowing the interrogation of  non-equilibrium processes including metabolisms in real time. Metabolism is fundamental to the cell and in cancer is significantly altered for cell proliferation and cell survival. Hyperpolarized metabolic imaging allows for anatomical and biological data to be determined concurrently. Using hyperpolarized pyruvate, succinate and glycine, which are biologically important metabolites in glycolysis, TCA cycle, and glycine metabolism, a new tool will be developed  for the early diagnosis of cancer as well as method in determining the aggressive profile of the disease. The focus will be on simultaneous monitoring of metabolic profiles and fluxes in the three metabolic cycles in cancer. A different in vivo metabolic profile should be seen in the more aggressive animal model. It is expected that the metabolic profiles and metabolic flux rates of hyperpolarized pyruvate, succinate, and glycine in vivo to substantially increase current understanding of metabolism in cancer. Ultimately, valuable predictors and biomarkers of neoplastic behavior may emerge.

Hao Hu, Ph.D.

Odyssey Fellow (2012-2015)
Department of Epidemiology

Uncovering the genetic bases for human diseases with sequencing data
Personal genome sequencing presents new challenges and opportunities for clinical interpretation of novel variants in human genomes. In response, new approaches for variant prioritization and disease-gene identification are emerging. Among the most promising of these are so called ‘burden tests’. In contrast to standard genome-wide association study (GWAS) approaches that evaluate the statistical significance of frequency differences for each variant in cases vs. controls, burden tests aggregate variants into discrete features—usually genes—to obtain greater statistical power. My major research interest is developing new burden test algorithms for finding disease-causing genes in humans, namely, the Variant Annotation, Analysis, and Search Tool (VAAST).  VAAST is aprobabilistic search tool that finds damaged genes and causal variants using whole-genome or exome sequencing data. Since the initial publication, we have repeatedly shown the utility of VAAST in both rare Mendelian diseases (e.g.Miller Syndrome and Ogden syndrome) and common, complex genetic diseases (e.g. breast cancer, Crohn diseases and hypertrilyceridemia), proving it is a robust software package that is highly accurate for both Mendelian and complex genetic diseases. Currently, we are incorporating support for family-based association study into the VAAST package (pVAAST), which makes uses of both case-control study design and pedigree sequencing data, achieving higher statistical power compared to transitional linkage analysis.

Nicholas Whiting, Ph.D.

Odyssey Fellow (2012-2015)
Department of Experimental Diagnostic Imaging

Improved Colorectal Diagnostic Imaging using Hyperpolarized, Functionalized Silicon Nanoparticles
There is an ever-growing need for accurate and effective colonoscopies to detect inflammation and abnormal growths in the lower gastrointestinal tract, such as polyps, diverticulosis, and cancer. The main objective of the proposed research is to develop a real-time molecular imaging nanoplatform to employ hyperpolarized silicon nanoparticles as viable molecular imaging agents for the early diagnosis of colorectal abnormalities and diseases, including cancer. The ability to non-invasively detect colorectal cancer is a huge step from traditional colonoscopies, and the increased detection sensitivity (and the absence of ionizing radiation) of hyperpolarized silicon nanoparticles is a huge improvement from typical MRI or CT scans. The increased resolution of this method will allow the detection of smaller polyps and lesions, as well as promote rapid feedback for decision making regarding patient selection, disease staging, and treatment monitoring; patients will benefit from a reduced incidence of invasive procedures, improved accuracy, and an increased ability to identify recurrences following treatment. The goal of this research is to form a new imaging method for detecting colorectal cancer at the earliest stages of development, thus dramatically increasing the patient's chances of successful treatment. The proposed research will systematically optimize the many components of this emerging field of colorectal cancer detection (including improvements to sensitivity, chemical specificity, and in vivo imaging), thus allowing it to quickly and efficiently translate from basic research to clinical imaging.

Francois Therriault-Proulx, Ph.D.

Odyssey Fellow (2012-2015)
Department of Radiation Physics

An in vivo plastic scintillation detector for ultrasound-guided seed implantation during low-dose-rate prostate brachytherapy
Plastic scintillation detectors (PSDs) have been shown to possess advantageous characteristics for in vivo dosimetry. Their sub-millimetric size, fast response (~ns), water-equivalence and independence with pressure and temperature are among the qualities of PSDs that other existing detectors do not have. In addition, the independence of their response to energies greater than 100 keV makes PSDs the ideal choice for the development of in vivo dosimeters for high-energy external beam radiation therapy. For irradiation modalities of lower energies, the energy-dependence of PSDs has limited their application. This is the case of LDR brachytherapy where radioactive sources of Iodine-125 (I-125) and Palladium-103 (Pd-103) are used, which possess polyenergetic emission spectra with average energies of 28 keV and 21 keV, respectively. The overall objective of this research project is to account for that energy-dependence and develop a clinically reliable PSD system that allows for accurate in vivo dosimetry during and after the implant of low-dose-rate (LDR) brachytherapy seeds.

Weiwei Yang, Ph.D.

Odyssey Fellow (2012-2013)
Department of Neuro-Oncology

The Mechanisms of PKM2-regulated Gene Expression
Tumor-specific pyruvate kinase M2 (PKM2) is essential for the Warburg effect. Besides its well-established role in aerobic glycolysis, PKM2 directly regulates gene transcription. However, the mechanism underlying this non-metabolic function of PKM2 remains elusive. This proposal aims to elucidate the mechanisms of PKM2-dependent histone H3 modification and examine the significance of this regulation in gene transcription and EGFR-promoted tumorigenesis. Overexpression or mutation of EGFR occurs in up to 60% of human glioblastoma. However, the efficacy of clinical treatment of some human cancers with EGFR inhibitors has been less significant than expected because of intrinsic and acquired resistance. Thus, the identification of novel key regulators for EGFR-regulated tumorigenesis may provide an alternative approach or combined approaches for treating EGFR-related glioblastoma.

Marc Ramirez, Ph.D.

Odyssey Fellow (2012-2014)
Department of Imaging Physics

Magnetic Resonance Imaging of Hyperpolarized 13C-Labeled Tracers and Their Metabolic Products: High-Throughput Technology for Preclinical and Translational Imaging Research
Developing safe and noninvasive imaging methods to accurately diagnose cancer and to rapidly evaluate response to therapy is challenging.  With its excellent soft-tissue contrast and high sensitivity to the abundant 1H nuclei in the body, magnetic resonance imaging (MRI) is well suited for high-resolution detection of morphological and functional changes in tumors.  However, metabolic imaging with traditional MRI methods is impractical. Recent advances in commercial technology to dramatically improve the MR signal of 13C-labeled tracers permit in vivo imaging of cancer metabolism. Preliminary results from preclinical and clinical trials are promising, yet engineering and economic challenges remain.  The hyperpolarized signal is short lived, and novel imaging methods to spatially, temporally, and spectrally separate the tracers from their metabolic products must be developed. Additionally, the cost per data point is remarkably high, requiring the consumption of a $15,000 per gram radical and requiring nearly an hour to prepare the tracer for a single animal acquisition.  The goal of this project is to promote the use of hyperpolarized MRI by developing high-throughput 13C imaging technology to reduce cost, improve statistical power, and lower experimental variance of preclinical cancer research, where imaging biomarkers can be established and ultimately be used to help guide patient-specific cancer therapies.

Hui Ling, Ph.D.

Odyssey Fellow (2012-2015)
Department of Experimental Therapeutics

The role of LONG1, a novel non-coding RNA located in 8q24 colon cancer risk locus, in Wnt signaling and chromosomal instability
In the 8q24 cancer risk region, it has been identified and cloned a novel long non-coding RNA transcript (LONG1, Long Oncogenic NcRNA Gene 1) that is highly over-expressed in microsatellite-stable (MSS) colorectal cancers (CRC), which have a poor prognosis. Functional studies showed that high LONG1 expression promoted tumor growth and metastasis, and exhibited resistance to chemotherapeutics drugs. In addition, exogenous LONG1 expression induced chromosomal instability (CIN) in HCT116, a near-diploid microsatellite-instable (MSI) CRC cell line with CIN-negative phenotype. Meanwhile, a consistent correlation between LONG1 and MYC expression both in cell line models and CRC patient samples were observed. However, LONG1-induced invasion and chemo-resistance cannot be rescued by reducing MYC expression, suggesting MYC is not the sole LONG1 target. Since MYC is one of the Wnt target genes, and higher β-catenin expression and higher Wnt activity in LONG1-overexpressing clones were observed, it is hypothesize that LONG1 may have a general activating effect on the Wnt signaling, and this enhanced Wnt signaling causes and maintains CIN  phenotype in the MSS type (the majority) of CRC.

Cecil Han, Ph.D.

Odyssey Fellow (2012-2015)
Department of Cancer Biology

Functions of a Novel RNA-binding Protein DBP-RB in microRNA processing
Defects in the DNA damage response (DDR) have been associated with various human diseases, including cancer. Emerging evidence suggests that microRNAs (miRNAs) play a key role in controlling the expression of genes involved in the initiation, activation and maintenance of the DDR. Recently, it has been shown that DNA damaging stress induces global changes in miRNAs expression. However, mechanism about how miRNA expression is regulated in response to DNA damage is largely unknown. Preliminary results showed that DBP-RB, one of RNA binding protein is interacted with both Drosha/DGCR8 and Dicer complexes, two primary miRNA processors. It has also been found that DBP-RB is colocalized with DNA damage foci and physically interacts with ATM (Ataxia Telangiectasia Mutated), a major signaling protein in the DDR. In particular, DBP-RB-depleted cells showed significant changes in the expression of a subset of miRNA and these DBP-RB-dependent miRNAs were also regulated in DNA damage-dependent manner. Based on the preliminary data, it is hypothesize that DBP-RB function as a mediator by transducing DNA damage signal to the miRNA processors to modulate miRNA processing by facilitating Drosha/DGCR8 and Dicer complexes. These proposed studies will be focused on understanding mechanisms by which DNA damage signaling is linked to miRNA biogenesis, providing valuable and mechanistic insights into the sensitivity and resistance of cancer cells to genotoxic drugs and a basis to establish novel therapeutic strategies.

Derrick Ong, Ph.D.

Odyssey Fellow (2012-2015)
Department of Genomic Medicine

Identification of factors that promote self-renewal of aged neural stem cells
Cytotoxic cancer therapy is associated with significant systemic toxicities across many organs, including the brain. As more cancer patients achieve long term survival benefit, these neurodegenerative sequelae are amplified and increase the susceptibility of patients to the neurodegenerative diseases including Alzheimer’s disease (AD). Advancing age itself is also a major driver of the neurodegenerative diseases and these conditions are fueled further by a greatly increased life expectancy worldwide. The impact of both cancer therapy and changing demographics, coupled with the lack of effective treatments for the neurodegenerative diseases, promises to extract a heavy social and economic toll worldwide, thus representing an urgent medical need. The aim of this project is to identify factors that promote self-renewal of aged neural stem cells. Interesting molecular targets will be identified and rigorously characterized with an eye on further drug development against these targets.  

Youting Sun, Ph.D.

Odyssey Fellow (2012-2013)
Department of Pathology

Systematic optimization of the next generation sequencing pipeline for cancer research
The goal of the project is to optimize the next generation sequencing (NGS) pipeline in order to facilitate biomarker discovery and key regulatory network identification in cancer biology. Currently, much of the complicated work flow of NGS remains as a black box to researchers. However by systematic modeling and simulation, key factors that affect sequence coverage, signal detection, and results accuracy will be revealed, which will shed light on NGS experiment design. In addition, better understanding of the workflow may lead to the development of more effective feature extraction methods, which in turn render better sensitivity and specificity in biomarker discovery. A second project is to detect signature patterns of genomic/epigenetic changes associated with tumor subtypes, recurrence, response to therapy, and survival. To this end, integrative multidimensional data analysis will be performed. Data sources include genomic, epigenetic, proteomic, and clinical data.

Sudha Krishnamurthy, Ph.D.

Odyssey Fellow (2012-2013)
Department of Head and Neck Surgery

Tumor Initiating Cells and Genomic alterations in Head and Neck Cancers
Head and neck squamous cell carcinoma (HNSCC) causes significant morbidity and mortality in the United States and worldwide, with over 250,000 diagnoses and 125,000 deaths annually. Despite the various advancements in treatment modalities, the 5 year survival rate is only 20-30% in advanced head and neck squamous cell carcinomas. This has been mainly attributed to the increased rates of recurrences and distant metastasis, the processes of which are poorly understood. Cancer stem cells (CSC), also known as tumor initiating cells, are believed to be the drivers of tumor recurrences and metastatic spread with its properties of chemo-radio resistance, and epithelial to mesenchymal transition.  The main goal of this project is to understand, and develop better CSC models to examine response to existing anti-cancer treatment, and to define new therapeutic targets. Specifically it hopes to understand how the genomic alterations and mutations identified in oral cancer affects the behavior and maintenance of oral cancer stem cells, including their capacity to avoid differentiation, apoptosis, and senescence.

Tyler Moss, Ph.D.

Odyssey Fellow (2012-2015)
Department of Systems Biology

MicroRNA Regulation of Signaling Networks Important in Cancer  
The focus of this project is on understanding the role of miRNA on signaling networks in cancer cells and to identify miRNA that can be utilized as therapy in combination with FDA approved chemotherapy. Screening of 879 miRNA coupled to phospho proteomics using 170 antibodies, and RPPA platform in three different breast and ovarian cancer cell lines have been completed. From these screening, it has been identified that 155 miRNA are significant in their modulation of signaling. Tyler will study a sub-set of these miRNA by using a systems approach combining computational modeling of miRNA effect on networks, bioinformatics mining of patient data from the TCGA and biochemistry and experimental biology. A second project that Tyler will work on is to integrate cellular automata models of cell-cell interaction with co-culture single cell analysis of cancer cells and fibroblasts to understand the role of signaling networks in modulating interactions with the microenvironment. These studies are based on clinical data with collaborators in Gynecologic Oncology and specific WNT signaling components has been identified as being upregulated in ovarian cancer cells in their interactions with fibroblasts.

Hoon Kim, Ph.D.

Odyssey Fellow (2011-2014)
Department of Bioinformatics & Computational Biology

Integrated Genomic Analysis of Regulatory Networks in Ovarian Cancer Subtypes 
The underlying vision of this research is to enhance the understanding of biological mechanisms governing cancer. Network analysis on high-throughput cancer data for this purpose will be performed. These studies may identify master regulators that can be therapeutically targeted. Barriers to achievement of network-level analysis of cancer can be the effect of tumor infiltrating normal cells and the lack of systematic network analysis integrating multiple types of high-throughput cancer data. To address these issues, the study will focus on high grade serious ovarian carcinomas (HGSC) while extension to other cancers is possible once appropriate data are available for analysis. Successful realization of the study will help to identify the gene regulatory networks responsible for molecular pathogenesis of tumors cell in HGSC, and will provide clues for biological mechanisms of non-tumor cell infiltration present in some tumors while not in others and will lead to potential applications in the prevention and treatment of ovarian cancer. 

Bo Zhong, Ph.D.

Odyssey Fellow (2011-2013)
Department of Immunology

Regulation of TLR- and TCR-Mediated Signaling by Ubiquitin-Specific Protease 25
This project investigates how TCR signaling and T cell differentiation are regulated by ubiquitination and deubiquitination process and how dysregulation of this process leads to autoimmune diseases and tumorigenesis. Although there are several reports indicating the link between ubiquitination defect and tumorigenesis, the exact mechanisms are poorly understood. USP25 as a negative regulator for both TLR- and TCR-mediated signaling has been identified. With autoimmune disease models and tumorigenesis models established, the plan is to elucidate the mechanisms by which USP25 regulates TLR and TCR signaling and investigate whether and how dysregulation of USP25-mediated deubiquitination regulates tumorigenesis. This study will further the understanding of how TLR- and TCR-mediated signaling is finely tuned by ubiquitination and deubiquitination and how deubiquitination-mediated dysregulation of TLR and TCR signaling leads to tumorigenesis.

Da Yang, Ph.D.

Odyssey Fellow (2011-2014)
Department of Pathology

Defects in microRNA biogenesis: a new hallmark of colorectal cancer

Colorectal cancer (CRC) remains one of the major causes of cancer deaths in the United States. Prevention and detection in its early stages as well as monitoring disease progression and predicting therapy outcome are hindered by suboptimal compliance with strategies. Identification of biomarkers for detection and prognosis and candidates for therapeutic intervention is a significant area of CRC investigation. MicroRNA (miRNA) is ∼22nt small non-coding RNA that plays a pivotal role in multiple cellular processes by inhibiting mRNA translation. MiRNAs are closely involved in tumorigenesis through either suppressing tumor by targeting oncogenes (e.g. miR-15a targeting BCL2) or promoting tumor by targeting tumor suppressors (e.g. miR-21 targeting PTEN and miR-17-92 cluster targeting E2F1). Recently, accumulating evidence has indicated that aberrant expression of miRNAs is associated with colorectal cancer development and progression. These discoveries are poised to lead to novel targets or tools for colorectal cancer therapeutics. Moreover, given miRNAs could be secreted by tumor into plasma, where they are stably protected from RNases, miRNAs that are dysregulated in tumor have a great potential as non-invasive cancer biomarkers. In this regards, the elucidation of the cause and consequence of miRNA dysregulation in CRC will help better understand the pathogenesis of colorectal cancer and discover novel molecular targets for the development of prognosis biomarkers and anticancer therapeutics.

Yun Zhang, Ph.D.

Odyssey Fellow (2010-2013)
Department of Genetics

A mouse model of sporadic breast tumors with a conditional p53 mutation
Missense mutations in the tumor suppressor gene p53 are present in more than 50% of human tumors. In particular, the arginine-to-histidine mutation at codon 175 (p53 ^R175H) has been found in more than 4% of human breast cancers. p53 mutations fall into two general categories: germline mutations that are associated with hereditary tumors, and somatic mutations that can cause sporadic turmors. Mouse models for cancers induced by p53 germline mutations have been established and characterized. However, hitherto there are no accurate animal models for sporadic tumors induced by p53 somatic missense mutation, although this mechanism of inactivating p53 occurs in a large fraction of human cancers. The project proposes to generate mouse models for sporadic breast cancer that can be induced by a conditional p53 ^R175H  mutation (corresponding to R175H in humans) specifically in mammary tissues. Experiments include generating engineered mice carrying a conditional p53 ^R175H  mutation; monitoring breast tumor formation following sporadic expression of p53R172H and investigating the molecular mechanisms of tumor development.

Xindong Liu, Ph.D.

Odyssey Fellow (2010-2013)
Department of Immunology

The Molecular Control of T-cell Tolerance
T-cell tolerance represents a negative regulatory mechanism to maintain T cell hyporesponsive to self tissues which minimizes the risk of autoimmunity. Importantly, induction of T-cell tolerance is hijacked by tumors in dampening the T cell anti-tumor function, which leads to tumor evasion. In addition to T cell receptor (TCR), accessory costimulatory receptors on T cells, such as CD28 and ICOS, play important roles in productive T-cell activation. It was found that in the absence of positive costimulation (both CD28 and ICOS signaling), activation of T cells by TCR signal alone results in their tolerance, similar to in vivo-rendered T-cell tolerance. The induction of T-cell tolerance is accompanied by selective activation of genes (Grail and Cbl-b) that inhibits TCR signaling as well as repression of T-cell cytokine genes (such as IFN-γ and IL-4). This in vitro T-cell tolerance induction model is a powerful and physiological system that would enable large quantities of tolerant T-cell to be generated, characterize the molecular control of their gene expression and study the stability or plasticity of T-cell telerance program. The project proposes to elucidate the molecular mechanism that controls T-cell tolerance program. Through the analysis of gene expression profiles and epigenetic modifications in tolerant T cells together with immunological analysis of tolerant T cell behavior in vivo, the aim is to identify key factors that control T-cell tolerance induction and maintenance. This project has broad therapeutic implications in cancers and autoimmune diseases. 

Prabodh Kapoor, Ph.D.

Odyssey Fellow (2010-2013)
Department of Carcinogenesis

Mechanisms of the INO80 Chromatin Remodeling Complex
The structure of chromatin has profound influence on nuclear functions, such as transcription, DNA replication, recombination and repair. Disruptions of these functions can lead to defects in gene expression and DNA damage repair, which constitute major underlying causes of many human cancers. Recent findings indicate that eukaryotic cells utilize two major classes of enzyme complexes to modify chromatin structure. These multi-protein complexes alter nucleosome architecture, either by covalent modifications of the histones (such as the histone acetyltransferases, HATs), or by ATP-dependent perturbations of histone-DNA interactions (the SWI/SNF family of complexes). Recent studies have shown that the INO80 complex is involved in multiple fundamental biological processes, such as in DSB repair and checkpoint regulation. The involvement of a single complex in different functions necessitates distinct regulatory mechanisms. Despite progress in linking chromatin remodeling to multiple processes involved in genome maintenance, mechanistic understanding of these links is currently lacking. As such, it is critical to reveal the biochemical and structural mechanisms of the INO80 complex underlying its activities in distinct biological processes.

Sofie Claerhout, Ph.D.

Odyssey Fellow (2009-2012)
Department of Systems Biology

Tumor dormancy and autophagy: implications for breast cancer
Recurrent metastatic disease is a major clinical problem among breast cancer patients, resulting in high morbidity and mortality. Hormone receptor positive (estrogen receptor, ER+) breast cancers can enter a prolonged period of dormancy resulting in recurrence 10-20 years later. Indeed, the long term outcome for patients with ER positive breast cancers is almost as bad as for those with ER negative cancers. In the case of late recurrence, cells that escape the initial therapy survive in a dormant state and ‘hide’ for years or decades ultimately giving rise to incurable metastases. Recent collaborative studies have suggested that autophagy, known as a survival mechanism in many cancers including breast cancer, may be required for tumor dormancy in ovarian cancer. The molecular mechanisms of cancer dormancy that allow cells to enter a dormant phase, escape from therapy, ‘reawaken’ and become occult after many years of dormancy are poorly understood. Moreover, one of the biggest problems that have hampered this field is the lack of defined molecular markers. Therefore, this project aims to investigate the genes involved in breast cancer dormancy and to examine whether autophagy can render breast cancer cells dormant and indolent for a long time period, only to awake with lethal consequences.

Marites Melancon, Ph.D.

Odyssey Fellow (2009-2012)
Department of Imaging Physics

Targeted Nanoshell-Based Agents for MRI-Guided Thermal Ablation of Head and Neck Cancer
In patients with advanced head and neck cancer (HNC), local tumor progression is the main cause of cancer-induced morbidity and mortality. Among patients with HNC treated with aggressive local-regional therapy, the local-regional recurrence rate is approximately 50%. In patients with local-regional recurrence, further surgery or radiotherapy is often not feasible owing to limited normal tissue tolerance. Thus, there is a need for innovative noninvasive or minimally invasive techniques that target recurrent HNC while exerting minimal effects on normal tissue. Among the techniques being investigated, thermal ablation is one of the promising options. However, thermal ablation for HNC currently has a major limitation: heat is delivered not only to tumor but also to surrounding normal tissues, which means that thermal ablation for treatment of lesions proximal to vital blood vessels and neural structures is associated with a substantial potential for life-threatening complications. If this limitation is to be overcome, thermal ablation will need to be modified such that heat is preferentially delivered to tumor and normal tissues are spared.