Skip Navigation

Click on any project title for a more detailed description of the project. For more information about any of these awards (e.g., PI contact information or associated publications), please use the corresponding project number to search for information at the NIH Reporter website.

Current Year | 2012 | 2011 | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999

Show/Hide All

Award
Type
Project # Year
of Award
PI Name(s)
All
Institution Title
Abstract Text (Official)
R21 CA099089-01 2003 FERRARI, MAURO OHIO STATE UNIVERSITY Nanomechanical Method for Molecular Analysis of Cancer
This project aims at the development of a novel method for the quantitative, objective, and rapid analysis of the expression of the molecule Her-2/neu on breast tissue biopsies. Her-2/neu is a prognostic and predictive biomarker for breast cancer, and its level of expression is a basis for the determination of the optimal treatment modality in contemporary clinical practice. The technology platform proposed herein employs characterization-mode ultrasound, in conjunction with immunotargeted micro- and nanoparticles, for the amplification of the molecular signature. The centerpiece of the approach is a novel theoretical framework, which enables the identification of previously inaccessible information on the physical properties of the probed tissue, at different length scales. Preliminary studies have yielded encouraging results on both native and particle-modified tissue samples. The hypothesis underlying the proposal is that molecular information can be translated into mechanical properties at specified length scales by way of nanoparticle immunotargeting, and that information based on these properties can be detected and analyzed using the ultrasonic characterization system and software proposed herein. In the R21 phase, we propose to establish the feasibility of this approach, starting with the development and refinement of robust protocols for the derivatization of nanoparticles with antibody conjugates; the development of sample preparation methods that are most suited for quantitative interrogation by ultrasonic characterization; the completion of an ultrasonic characterization system for detecting the nanoparticle signals; and the finalization of the data analysis tool under the novel, multi-scale mechanical field theory. In keeping with the proof-of-principle nature of this phase, the milestones proposed for the transition into the R33 phase are the attainment of specificity and sensitivity figures that are no less than those that pertain to prevalent current pathological investigation techniques, i.e., in this context, IHC and FISH. Even at identical sensitivity and specificity, the proposed system is believed to be potentially advantageous over IHC and FISH in that it is very rapid, automated, and objective, has high potential for interobserver and interlaboratory agreement, and requires significantly reduced level of investment of a specialist's time. In the R33 phase, the objectives of the proposed program would be to optimize the system's performance and create an actual prototype for clinical testing in the context of breast malignancies.
 
R21 CA103071-01 2003 MAJUMDAR, ARUNAVA UNIVERSITY OF CALIFORNIA BERKELEY Nanocapillary Electrophoresis for Profiling Cancer
The goal of this project is to develop nanocapillary electrophoresis array technology (NEAT) as a label-free, quantitative, and high-throughput assay of proteins and nucleic acids that are critical in early detection, monitoring, diagnosis, and prognostic evaluation of cancer. The NEAT chips will contain an array of glass nanocapillaries whose inner surfaces are functionalized with receptors to capture specific ligands. We have developed a process for synthesizing glass nanocapillaries with lengths in the 1-10 micron range and diameters in the 5-20 nm range, which is on the order of biomolecular size. Hence, when analyte biomolecules are electrophoretically transported through functionalized nanocapillaries in response to an applied voltage (approx. 1 V), specific ligand-receptor binding will lead to reduction in the ionic current flow due to partial blockage of the nanocapillary. Recent experiments have shown that indeed when a transmembrane ion channel protein is functionalized with a single strand of DNA, binding of its complementary strand can be detected through modulation of the ionic current with single-molecule sensitivity, and specificity of single base pair mismatches. By using the same principles, the goal of the R21 phase is to demonstrate that ionic current modulation could be used as a label-free assay for PSA protein and p53 exonic sequence with specificity and sensitivity sufficient for cancer diagnostics and monitoring. The goal of the R33 phase is to build upon the R21 knowledge and develop the NEAT chip that will consist of an array of microfluidic cells, with each cell containing a single nanocapillary functionalized with a distinct probe/receptor. An electronic system will allow multiplexed addressing of individual nanocapillaries such that their respective ionic currents can be simultaneously measured. Such an integrated system will enable multiplexed assays of literally hundreds of molecules, such as transcribed mRNAs or proteins. For protein expression, in particular, this technology would represent a new paradigm in the evaluation of multiple proteins from a single tissue or from serum and could represent a cost-effective way to assess multiple cancer antigens in cancer screening and monitoring programs. Such routine screening is now only done for very few cancer antigens (primarily prostate specific antigen, PSA) due to the expense such large-scale screening would incur. This technology would directly address this important public health issue.
 
R21 CA102949-01 2003 MAKRIGIORGOS, G. MIKE DANA-FARBER CANCER INSTITUTE ERROR-FREE DNA AMPLIFICATION FOR MUTATION DETECTION
High selectivity mutation detection, which frequently relies on PCR, often falls short by 1-2 orders of magnitude of the selectivity required to identify cancer cells at an early stage, to investigate mechanisms of tumorigenesis, to detect mutations in single cells or to reliably detect minimal residual disease. A major cause of this deficiency is that PCR poses a selectivity limit, typically 1 mutant in 105- 106 wild type alleles, since all DNA polymerases invariably generate errors during DNA synthesis which can be misinterpreted as mutations (false positives). We have now developed hairpin-PCR, a novel method that allows elimination or major reduction (>100-fold) of PCR errors in a sequence of interest, and can supply existing mutation detection technologies with the necessary 'selectivity leap'. By converting a DNA sequence to a hairpin and performing PCR in a hairpin structure, true mutations can be separated from polymerase-generated misincorporations, thereby providing practically error-free DNA. This project will develop and optimize the technology to PCR amplify sequences from genomic DNA, eliminate error-containing sequences, and demonstrate its application for highly improved mutation detection using technologies previously limited by PCR errors. In the R21 phase we shall demonstrate amplification of hairpins up to 500 bp long directly from human genomic DNA, and we shall optimize isolation of hairpins containing genuine mutations (homoduplex hairpins) from hairpins containing polymerase errors (heteroduplex hairpins), using dHPLC separation. In the R33 phase the technology will be further developed to (a) reduce PCR errors by at least 1000 times, (b) demonstrate that hairpin PCR allows current mutation detection methods a major selectivity leap relative to their current limit, and (c) amplify large portions of the human genome in an error-free manner, for multi-gene mutation screening. By providing error-free amplified DNA for analysis, the present hairpin PCR will allow a major boost to almost every existing genotypic selection method and enable studies and diagnostic tests that were not possible with previous technology. The new technology will also improve the accuracy of microsatellite analysis and will have additional applications in molecular beacons and real time PCR, and in DNA cloning or protein functional analysis by in vitro translation.
 
R21/R33 CA101150-02 2003 ANGELETTI, RUTH HOGUE ALBERT EINSTEIN COLLEGE OF MEDICINE YESHIVA UNIVERSITY Proteomics of Hepatic Carcinogenesis
Hepatocellular carcinoma is the one of most frequent causes of death by cancer in the world. There is no reliable diagnosis prior to late stages of disease and no hope for cure except surgery. The overall goal of the proposal is to discover and identify distinctive alterations of protein expression in early precancerous lesions isolated from their physiological microenvironment. The data obtained will provide early molecular markers with diagnostic and therapeutic potential for early intervention in human liver cancer. We propose to apply innovative proteomics and laser capture microdissection microscopy (LCM) to study a well-characterized animal model of liver carcinogenesis (RH) that exhibits well-defined, synchronous stages of initiation and progression of liver cancer that are strikingly similar to those in liver cancer in humans. In the R21 phase, we will focus on demonstrating our ability to generate useful patterns/profiles with combined LCM and protein technologies using control tissue/serum and one preneoplastic stage of the biological model system. In the R33 phase, we will apply the pattern/profile generation technologies to the cells and sera of each of the early stages and control paradigms so that we can identify candidate markers and targets. We will use both global and targeted proteomics strategies to: 1) identify difference proteins in cells early in the development of liver cancer; 2) identify unique serum markers at early stages; 3) determine progressive changes in tubulin isotype composition, which is associated with development of drug resistance; and 4) identify changes in proteins associated with the cytoskeletal scaffold.
 
R21/R33 CA092752-02 2003 BARRON, ANNELISE EMILY NORTHWESTERN UNIVERSITY Fast Mutation Detection by Tandem SSCP/HA on Microchips
It is proposed to optimize, evaluate, and pilot rapid, scalable, and low-cost microchip electrophoresis technologies for sensitive and specific molecular detection of cancer by tandem single-strand conformational polymorphism (SSCP)/heteroduplex analysis (HA), using the p53 gene as a model system. We request a 1-year R21 phase and a 3-year R33 phase. The proposed project involves collaboration between members of Northwestern's Lurie Comprehensive Cancer Center, including researchers in Chemical Engineering, the Medical School, and Evanston Hospital. Microchannel "tandem" SSCP/HA is a novel mutation detection method recently developed in our laboratory, which involves the simultaneous generation and analysis of homo/heteroduplex DNA and SSCP conformers. Studies of a significant number of samples (32) indicate that tandem SSCP/HA allows for much higher-sensitivity mutation detection (100%) than SSCP alone (93%) or HA alone (75%), for p53 samples. We have developed and published optimized sample preparation protocols, gel formulations, and analysis conditions for capillary array electrophoresis (CAE). During the R21 phase, we will translate these methods to microfluidic electrophoresis chips, which offer a large increase in throughput and drop in cost of DNA analysis compared to CAE. The p53 gene, known to be mutated in >50% of human cancers, and whose mutation status can be predictive of patient response to chemotherapy, is the important model system chosen. However, microchip-based genetic analysis technologies to be developed should be easily applied to ANY cancer-related gene. In the R21 phase, we will analyze approximately 60 different DNA samples derived from tumor cell lines, representing a range of mutations in different p53 exons, to determine the impact of DNA sample characteristics and electrophoresis protocols on the sensitivity and specificity of the method, in a blinded study designed by collaborating biostatisticians. When optimized tandem SSCP/HA protocols have been developed for microchips, they will be piloted by the analysis of >200 selected samples amplified from frozen, solid tumors banked at Evanston Hospital. Via this blinded study, sensitivity and specificity (both expected to be at or near 100%) will be determined and reported for the first time using banked tumor tissue, providing necessary validation for clinical application of this technique, and making rapid, low-cost cancer genotyping technology widely available to physicians.
 
R21/R33 CA099139-02 2003 GOODLETT, DAVID ROBINSON INSTITUTE FOR SYSTEMS BIOLOGY Parallel Peptide Tandem Mass Spectrometry (MS)
It is the specific aim of the R21 phase of this proposal to develop the software tools necessary to interpret tandem mass spectra produced by collision induced dissociation (CID) of peptides in parallel rather than in series, as is commonly practiced in shotgun proteomics, and to prove its advantages over serial CID on samples of increasing complexity from peptide standards to proteins extracted from a medulloblastoma primary cell line lysate. Our proposed technology is referred to as shotgun CID and shotgun tandem mass spectrometry (MS/MS) to distinguish it from serial tandem MS and multiplex MS in a Fourier transform-ion cyclotron resonance-mass spectrometer (FT-ICRMS). It is the general aim of this combined R21-R33 proposal to provide a cost effective competitor to the advantages of multiplex MS in FT-ICR-MS. Our approach will use lower resolution and mass accuracy time-of-flight (TOF) mass analyzers and a continuously alternating data acquisition scheme of parent ion measurement followed fragment ion measurement throughout the chromatographic introduction of sample. Proteins will be identified by a combination of mass mapping of parent peptide ions across the entire chromatographic time, unique chromatographic constraints and their combined fragment ions. Primary brain tumors are the leading cause of cancer-related death in children. Medulloblastoma is a primitive neuroectodermal tumor that typically arises from the cerebellar vermis and shows variable degrees of arrested neural differentiation. The cerebellum requires endogenous retinoids for proper control of neuronal apoptosis and differentiation during development. In embryonal carcinoma and neuroblastoma cells, retinoids induce neural differentiation and cell-cycle arrest. Retinoids have recently been shown to induce extensive apoptosis and neuronal differentiation in medulloblastoma cell lines and freshly resected medulloblastoma cells. Together with Dr. Jim Olson of the Fred Hutchinson Cancer Research Institute, who is organizing a Phase III clinical trial on the effects of 13-cis retinoic acid in medulloblastoma therapy, we will develop shotgun CID within a model that seeks to: 1) facilitate identification of markers associated with retinoid-responsiveness in medulloblastoma cell lines and 2) define key components of the retinoic acid pathway modulated by treatment in retinoid-responsive medulloblastoma cells. This work will be done with medulloblastoma cell lines and primary cells lines derived from retinoid-sensitive and -resistant tumors.
 
R21/R33 CA103235-02 2003 KRON, STEPHEN J. UNIVERSITY OF CHICAGO BCR-ABL Kinase Assays for STI571 Sensitivity of Response
This phased innovation award proposal is focused on developing a robust approach to quantitative assay of specific protein tyrosine kinase activities from cancer cells. Our model is the oncogenic BCR-ABL fusion protein, the gene product of the t(9;22) Philadelphia chromosome (Ph1) translocation observed in the vast majority of Chronic Myelogenous Leukemia (CML) and in up to 30% of adults with Acute Lymphoblastic Leukemia and in other hematological neoplastic diseases. The activation of Abl kinase by fusion to BCR that is inferred to underly the malignant transformation of Phl positive CML is effectively opposed by the orally administered tyrosine kinase inhibitor (TKI) Imatinib Mesylate (IM, STI-571, Gleevec). The activity of IM as an Abl kinase inhibitor in vitro is thought to be the critical determinant of its efficacy in vivo. Nonetheless, a clinically useful assay for IM inhibition of BCR-ABL kinase activity in circulating CML leukemia cells is lacking. We propose to develop a protein/peptide chip-based assay for BCR-ABL that can detect the degree of inhibition by IM to evaluate dosing and drug resistance. Insofar as other activated tyrosine kinases may be critical mediators of malignancy in both leukemias and solid tumors, developing such an assay would be a powerful tool in evaluating other TKI drug candidates targeting these kinases for their efficacy in vivo. Thus, this project is directed at two major discovery objectives and three development objectives. First, in the initial project year, we intend to use our established methods for anti-phosphotyrosine antibody-based detection of purified Abl kinase activity on a peptide chip to 1) Recapitulate our Abl kinase assay with undiluted whole cell extracts from cell lines expressing BCR-ABL and 2) Use this assay to measure the inhibition of BCR-ABL by IM both in extracts and intact cells. During the development phase, we intend to use both BCR-ABL expressing cancer cell lines and circulating leukemic cells from treated patients as samples to 1) Optimize the BCR-ABL substrate and reaction conditions to enhance sensitivity and specificity of phosphorylation, 2) Examine alternative detection methods for BCR-ABL activity based on phosphospecific antibodies and thiophosphate targeted chemistry, and 3) Evaluate different chemistries for immobilizing BCR-ABL substrates on a surface and geometries for detection of phosphorylation. By these aims we intend to develop a highly versatile kinase assay system which can be applied to monitoring of patient response to IM and as a tool for discovery and testing of new TKI cancer drugs.
 
R21/R33 CA099191-02 2003 LABAER, JOSHUA HARVARD UNIVERSITY Functional Proteomics of Breast Cancer
In the past decade, biological research has witnessed a paradigm shift from focused reductionist approaches to a greater dependence on large "industrial-sized" projects. High-throughput (HT) biology began in earnest with the Human Genome Project, and increasingly these HT tools and approaches are being exploited for protein research. Given the importance of proteins in disease etiology and treatment, a major challenge facing biology is the elucidation of the physiological role of all proteins. In this light, the field of functional proteomics, a new approach to the HT study of proteins, will enable the expression and subsequent assay of proteins and their various properties such as subcellular location, interacting partners, biochemical activity or regulated modification at a scale of thousands at a time. A prerequisite for this approach is the need for large collections of cDNAs in a format conducive to HT protein expression. We and others (Walhout 2000, Brizuela 2001) have begun to create such collections of cDNAs using the novel technology of recombinational cloning that allows rapid transfer of DNA fragments from one vector to another, in frame and without mutation. However, high-quality collections of human clones are not yet available. Here we propose to build a collection of 1000 expression-ready human cDNA clones representing genes of significance to breast cancer (BC 1000). We are nearing completion of 100 such clones assembled in a pilot study and have already found them to be invaluable in the development of HT methods for both in vitro and in vivo studies. However, the relatively modest size of the collection prevents its application in any meaningful screening experiments. Thus it is important to expand this collection to a size that will enable more comprehensive screens and the development of experimental technologies that will truly exploit the HT setting. In order to exploit this resource, we have developed a novel method for creating protein microarrays that enables the HT functional study of proteins. This approach, called Nucleic Acid-Programmable Protein Array (NAPPA), replaces the complex process of spotting purified proteins with the simple process of spotting DNA. By exploiting the recombinational format of the BC1000, genes are then simultaneously transcribed/translated in a cell-free system and the resulting proteins are immobilized in situ, minimizing direct manipulation of the proteins and making this approach well suited to HT applications. Advantages of this approach include: the ability to express and interact proteins in a mammalian milieu, no requirement for HT expression-purification-storage of proteins, and real time collection of data, minimizing concerns about protein stability. We propose to adapt this method to a glass matrix and miniaturize it to allow for the screening of thousands of proteins simultaneously. We will demonstrate the effectiveness of this approach by executing a 1000 x 1000 interaction matrix with the BC1000.
 
R21/R33 CA099246-02 2003 SOPER, STEVEN ALLAN LOUISIANA STATE UNIVERSITY A&M COLLEGE BATON ROUGE Microsampling Unit for Capturing Low Abundant Cells
Our group will be developing modular microsystems that can be assembled in a variety of different configurations to carry out complex biomedical assays appropriate for monitoring both genetic and protein markers in clinical settings. Our focus in this R21/R33 application will be screening diagnostic markers associated with breast cancer. The focus of our project is to build upon our existing expertise in LIGA to fabricate BioMEMs devices possessing high-aspect ratio microstructures (HARMs) fabricated out of a variety of polymer materials. Specifically, we will be fabricating the following modular devices: (1) Microsampling unit with target preconcentration capabilities of rare cancer cells in circulating blood. Our microsampling unit will consist of capture beds prepared using surface imprinted polymers to semi-selectively capture pre-selected targets (cells expressing EpCAM). (2) Cellular lysis system. Captured cells will be electrodynamically lysed using microfabricated electrodes to induce cell rupture via electroporation. (3) Multiplexed hybridization association unit. To signal the presence of either a mutant gene or protein that is under- or over-expressed, we will build arrays of elements consisting of tethered nucleic acid probes or antibodies (Ab) to associate selectively with our targets. Since both genetic and protein markers can be effective in diagnosing a particular disease state, we will build hybrid systems (hybrid-biosensors) that will monitor simultaneously both genetic and protein markers to minimize false positives and negatives associated with the diagnosis. (4) Micro-optical bench with high sensitivity laser-induced fluorescence components. The transduction of the hybridization of target DNAs to our tethered probes or association with tethered Ab will be determined using near-IR fluorescence with excitation provided by a surface mounted vertical cavity surface emitting laser (VCSEL) and the sensing elements situated on a planar polymer waveguide. (5) Ultrasensitive probes for fluorescence transduction. Our fluorescence detectors will be configured to read luminescence generated in the near-IR by developing labeling probes that possess spectral properties in the near-IR (excitation/emission maxima > 700 nm).
 
R21/R33 CA099835-02 2003 WOODS, VIRGIL L UNIVERSITY OF CALIFORNIA AT SAN DIEGO Enhanced Crystallography of Cancer-Implicated Proteins
We aim to enhance the ability to design protein crystallographic constructs and, thereby, substantially speed protein structure determination with the use of information provided by innovative peptide amide hydrogen exchange techniques. In this resubmission, we present extensive preliminary studies, performed since review, that directly address all reservations regarding our prior submission. Determination of high-resolution structure at an increasingly high throughput (HT) pace is required for a fundamental understanding of how modifications of cancer- implicated proteins can promote oncogenesis and metastasis. Unfortunately, HT crystallographic efforts have a single, dominating roadblock: they produce suitable crystals for a small minority of target proteins. Floppy, unstructured regions of failed proteins play a major role in this problem. The exchange rates of the many peptide amide hydrogens within a protein are determined by the protein's stability at the individual amino acid scale. We have developed an enhanced form of amide hydrogen/deuterium exchange-mass spectrometry (DXMS) that can rapidly and precisely measure such rates. We propose that DXMS data can be used to identify and localize such unstructured regions within a protein and thereby guide the design of modified protein in which such regions are selectively removed. Furthermore, many proteins require tertiary-quaternary contacts, provided by binding partners, to induce structure in such regions. For these proteins, DXMS can be used to rapidly select binding partners that provide the needed stabilizing contacts, allowing focused protein-binding partner co-crystallization efforts. Importantly, repeat DXMS study of the modified protein(s) can rapidly determine how well they have retained the structured elements of the original protein. In our R21 year, we will demonstrate that DXMS can guide the re-design of protein constructs sufficiently to produce a 50% increase in overall crystallization success rates for target proteins, and do this at a high throughput pace. This result will establish the ability of DXMS to speed throughput of present HT crystallographic efforts, and likely similarly enhance construct definition for conventional, specific-protein focused crystallography, with obvious benefits for the structural study of cancer related proteins. In our R33 years we will establish a crystallography-dedicated DXMS facility and further refine our ability to guide construct design by analysis of the protein targets studied by our collaborators at the Joint Center for Structural Genomics, with an emphasis on those with cancer-relevance. This construct-refinement resource will then be broadly extended as a community service to NCI-funded investigators for application to both conventional and HT crystallographic efforts.
 
R33 CA103056-01 2003 ARAP, WADIH UNIVERSITY OF TEXAS MD ANDERSON CANCER CENTER STEM CELL-BRAIN TUMOR INTERPLAY & IN VIVO PHAGE DISPLAY
This proposal aims to develop innovative methods for brain cancer diagnosis and therapy that will combine the strengths of neural stem cell (NSC) biology and in vivo phage display technology. The proposal is based on our prior work that demonstrated a remarkable, apparently "magnetic" attraction of NSCs to glioblastoma brain tumor cells. When NSCs were injected into one cerebral hemisphere, and rat or human glioblastoma tumors into the other, the NSCs migrated across the midline and headed directly to the tumor masses. When the NSCs were injected intravenously, they entered the brain and selectively targeted on the tumor. NSCs attached even to single tumor cells which were in the process of invading normal brain tissue. When NSCs were engineered to deliver toxic molecules, tumor cells were killed. New experiments will build on these results. 1. Short- and long-term effects of NSCs will be analyzed on genetically-induced natural tumors, not only on grafted tumors. 2. Optimal cell numbers and optimal route of injection into mice will be explored with mouse and human NSCs, including determination of whether a carotid intra-arterial route might be more effective than intracerebral or intravenous routes. 3. As a step toward development of diagnostic procedures of higher sensitivity, for future use in humans, NSCs will be modified to carry molecules allowing radiological visualization, so that the NSCs will serve to delineate the positions, sizes, and number of tumor masses in the brain. 4. As model "proof-of principle" experiments, NSCs will be engineered genetically to synthesize and release agents that kill dividing cancer cells and/or other agents that may induce cancer cells to differentiate into stable, quiescent glial cells that no longer endanger life. 5. To uncover the basic molecular and cell biological mechanisms controlling the "cross-talk" between NSCs and tumor cells, the powerful phage display technology, which allows identification of ligands and their receptors without preexisting data about their natures, will be used in tissue culture and in intact mice to define host and tumor ligands that react with NSC receptors and attract NSCs to the tumor, as well as the reverse - - specific receptors on brain tumor cells and on their specialized blood vessels that bind peptide ligands released by NSCs.
 
R33 CA103068-01 2003 COLLINS, COLIN C UNIVERSITY OF CALIFORNIA SAN FRANCISCO Development of ESP: Structural & Functional Oncogenomics
The long-term objective of this proposal is to gain an enhanced understanding of the structural genomics of solid tumors through development of a novel, sequence-based method capable of identifying all types of structural rearrangements that occur in tumor genomes. Genome rearrangements can promote cancer development, progression and/or resistance to therapy by altering gene regulation and/or function, and the involved genes are potential therapeutic targets. This is well established in leukemia and lymphoma, but less so in solid tumors, in part because of the difficulty of identifying the genes involved in complex structural rearrangements. We describe here a powerful and high resolution, sequence-based analytical approach called End Sequence Profiling (ESP). ESP maps copy number aberrations and directly identifies and clones en masse genome breakpoints associated with genome rearrangements such as inversions, translocations, deletions and amplifications. ESP is accomplished by constructing a BAC library of a tumor genome, end sequencing a larger number of BAC clones, and mapping the BAC end sequences (BES) onto the normal genome sequence. Paired BES that map to different parts of the normal genome span structural rearrangements. Sequencing these clones will reveal exact breakpoints and involved genes. In Specific Aim 1 we will: Implement ESP as a cost effective sequence-based technology for determining the structural organization of tumor genomes and clone rearrangement breakpoints en masse. Determine the minimum sequencing depth needed to yield the maximum structural information. Determine if ESP can reproducibly identify recurrent rearrangements between tumors, and if so, whether specific sequence elements are associated with these rearrangements. In Specific Aim 2 we will: Develop robust computational methods for the analysis, visual representation, and integration of ESP data with the human reference sequence, making possible comparison of ESP data from independent tumors. Knowledge of how genome rearrangements such as inversions and translocations impact local gene expression is critical. Thus, we will integrate ESP-based structure data with expression microarray data and co-localize aberrantly expressed genes with genome rearrangement breakpoints. In Specific Aim 3: We will biologically and clinically validate key ESP findings. We believe ESP provides a rational framework for sequencing tumor genomes. In fact, ( 100 tumor genomes can be analyzed at ( 10 kb resolution for less than sequencing a single 3000 Mb genome yielding hundreds of novel biomarkers and therapeutic targets associated with translocations, inversions, and complex rearrangements. This is important because, just as a comprehensive systems-based knowledge of human biology is predicated on the structural organization and sequence of the human genome, a structure-based view of tumor genomes is essential for a comprehensive understanding of tumor biology.
 
R33 CA101136-01 2003 FURGE, KYLE A VAN ANDEL RESEARCH INSTITUTE Comparative Genomic Analysis of Microarray Data
Aneuploidy is a common feature of cancer and several lines of evidence suggest that cytogenetic aberrations can significantly influence cancer diagnosis, prognosis, and treatment. While molecular genetic based methods, such as comparative genomic hybridization (CGH), have traditionally been used to determine cell karyotypes, recent transcriptional profiling studies have suggested that it is possible to predict cytogenetic changes from microarray gene expression data. A technique we term comparative genomic microarray analysis (CGMA) is based on the observation that gene expression values show expression biases, either increased or decreased, in regions of chromosomal gain and loss, respectively. In tumor samples, CGMA predictions are made by mapping gene expression values to the public human genome assembly and scanning for genomic regions that contain a statistically significant upwards or downwards gene expression bias. While the first-generation of algorithms that identify gene expression biases produce reasonably good cytogenetic predictions, it is likely that more sophisticated algorithms could produce better results. The R21 phase of this proposal focuses on implementing and testing a set of refined CGMA algorithms to make more accurate and higher resolution cytogenetic predictions. Cytogenetic and transcriptional profiling data obtained from a small set of colon tumors will be used for algorithm testing. The R21 milestone is to establish a CGMA prediction method that matches CGH determinations with high accuracy. The R33 phase will focus on developing algorithms to make CGMA predictions across multiple samples and will test if frequently changed regions identified by CGMA match those regions previously identified by CGH. For the R33 phase, hepatocellular carcinoma (HCC) and renal cell carcinoma (RCC) will serve as models because large sets of gene expression and cytogenetic profiling data are currently available. Historically, candidate genes have been identified by determining if a gene located within a region of frequent cytogenetic change is either mutated or misregulated. In this proposal, candidate genes will be identified from the HCC and RCC gene expression profiles by first using CGMA to locate frequently changed genomic regions and then by using traditional gene expression analysis to identify abnormally expressed genes located within these regions of frequent cytogenetic change.
 
R33 CA099136-01 2003 LAM, KIT S UNIVERSITY OF CALIFORNIA DAVIS Small molecule microarrays for intracellular proteins
This project involves the application of the "one-bead one-compound" encoded small molecule combinational library method and chemical microarray technique to study functional proteomics. Five enormous libraries of small molecule ligands (a total of over 1 million compounds) will be generated and screened against whole cell extracts derived from a B lymphoma cell line (Ramos). Billions of possible molecular interactions will be examined concurrently. Beads containing compounds that bind to cellular proteins or protein complexes will be isolated and the compound chemical structure determined by our novel decoding method. Selected small molecule ligands will be resynthesized on Sepharose beads and used as affinity matrix to capture the binding proteins or protein-complexes. The identity of the bound proteins will then be determined by protein separation and mass spectroscopy. Based on the chemical structure of these ligands, a small molecule microarray (approximately 1000 compounds) will be developed to probe the functional state of the whole cell extract. Our hypothesis is that with the above experimental scheme, we can systematically select a finite number of small molecule ligands and use them as capturing agents to probe the functional state of a B lymphoma cell. We further hypothesize that some of the ligands that bind to unique protein targets in lymphoma cell can be used as lead compounds for the development of anti-lymphoma agents. Once validated in a large number of lymphoid malignant cell lines, peripheral blood lymphocytes, and a limited number of primary malignant lymphoid tissues, this microarray technology can be applied to biopsy specimens obtained from a large number of patients with lymphoid malignancies. This technique, if successful, can readily be applied to other cancer types as well.
 
R33 CA099135-01A1 2003 LIZARDI, PAUL M. YALE UNIVERSITY Genetic Analysis of Amplified Genomic DNA Archives
Modern DNA analytical methods and microarray technologies are potentially enabling tools for the comprehensive genetic analysis of tumors as well as premalignant lesions in patients at risk for cancer. A significant obstacle to such analysis, however, is the need for relatively large DNA samples. A recently developed isothermal whole genome amplification method promises to eliminate these barriers to comprehensive genetic analysis. This research project aims to demonstrate the utility of simple and robust amplification procedures for generating archival copies of genomic DNA from small tissue samples. Head and neck cancer will be used as a model to optimize and validate procedures for DNA archiving of neoplastic and pre-neoplastic lesions. Samples will be collected prospectively, and cancer tissue, cancer tissue margins, as well as other lesions identified as candidates for pre-neoplasia will be collected. A subset of the samples will be isolated using laser capture microdissection. All collected tissue samples and microdissected samples will be amplified to generate a DNA archive of head and neck cancer and premalignant lesion specimens. A variety of methods, including microsatellite analysis, detection of human papillomavirus subtypes, and comparative genomic hybridization (CGH) on BAC microarrays, will be used to investigate and validate the utility of the archive of amplified DNA. CGH analysis on BAC microarrays enables the detection of genetic alterations at thousands of gene loci in the archived DNA samples, which are representative of different stages of cancer, preneoplasia, or benign dysplasia. Bioinformatics tools will be used to construct alternative classification schemes based on distance-based trees or clustering algorithms, utilizing the complete data set of microsatellite, array-CGH, and HPV infection status observations. This study is intended to serve as a demonstration and validation of the utility of DNA archives, generated by isothermal whole genome amplification, for comprehensive studies in cancer genetics. The novel capabilities established by this study should be extensible to any human cancer model and should greatly expand the utilization of genetic analysis at the level of the entire genome by eliminating many of the constraints related to limited availability of biological material.
 
R33 CA095996-01A2 2003 RONINSON, IGOR B ORDWAY RESEARCH INSTITUTE, INC. Function-based selection of target genes in tumor cells
This proposal uses a new methodology to identify human genes that are required for tumor cell growth. Such genes, which provide potential targets for cancer treatment, will be identified through expression selection of genetic suppressor elements (GSEs). GSEs are biologically active sense- or antisense-oriented cDNA fragments that inhibit the function of the gene from which they are derived. Genes that are essential for cell proliferation are expected to give rise to GSEs that inhibit cell growth. Such GSEs can be isolated by bromodeoxyuridine (BrdU) suicide selection from a normalized (reduced-redundance) library of human cDNA fragments in an inducible retroviral vector. In preliminary studies, selection for growth-inhibitory GSEs has been carried out in breast carcinoma cells, yielding growth-inhibitory GSEs from about 60 genes. Many of the genes identified by GSE selection are known oncogenes or positive regulators of cell growth, while other genes have no known function or had not been previously implicated in cell proliferation. This analysis will now be extended to several other types of tumor and normal cells. A normalized cDNA fragment library in an inducible retroviral vector will be generated from a mixture of RNA preparations from multiple human tumor cell lines. This library will be transduced into recipient cell lines derived from several major types of human cancer and into telomerase-immortalized lines of normal human cells. Prior to transduction, the recipient cell lines will be derivatized to provide for high efficiency of retroviral infection and for the ability to regulate gene expression from retroviral vectors. The transduced cells will be subjected to BrdU suicide selection for growth-inhibitory GSEs, and GSE-enriched population of cDNA fragments will be recovered from the selected cells. Genes enriched by GSE selection will be identified by sequencing, and representative GSEs from each gene will be tested by several functional assays. The role of GSE-cognate genes in cell proliferation will be confirmed via siRNA inhibition. Genes identified through GSE selection will be prioritized as potential targets by comparing the ability of their cognate GSEs to inhibit cell growth in different types of tumor and normal cells and by analyzing the ability of the GSEs to induce tumor cell death through mitotic catastrophe. This analysis will provide a database of potential new targets for the development of anticancer drugs.
 
R33 CA103455-01 2003 TAYLOR, CLIVE R UNIVERSITY OF SOUTHERN CALIFORNIA Retrieval of DNA, RNA, and Protein from Archival Tissue
The current phase I R21 study has accomplished three Specific Aims. The feasibility of retrieval/extraction of protein, RNA, and DNA from archival formalin-paraffin tissues by modified Antigen Retrieval (AR) methods was demonstrated. PCR based methods were adapted and applied successfully for the amplification and quantitation of RNA/DNA extracted by modified AR from formalin-paraffin tissues. Three model systems were developed to monitor the efficacy of the modified AR process, including novel simulated or 'faux tissues' and purified 'protein matrix pellets'. Continuation of these studies in this R33 proposal focuses upon the incorporation of these advances into an 'integrated systems approach' to the molecular analysis of cancer tissues that have been preserved as formalin-paraffin blocks. Currently, immunohistochemical (IHC), in situ hybridization (ISH) and PCR based assays are widely applied to formalin-paraffin cancer tissues in 'routine diagnosis' and in basic cancer research, but the reproducibility and the validity of the findings are open to serious question due to the lack of uniform methods, in particular the total lack of 'standard reference materials', which are considered essential to the use of analogous methods in the diagnostic clinical laboratory. This proposal addresses pre-analytical, analytical, and post-analytical aspects of the use of IHC, ISH and PCR based assays as applied to formalin-paraffin cancer tissues. Integral to this approach is the development of reference standards for key analytes that are, or may be, employed as cancer or prognostic markers, against which results obtained by the analysis of formalin-paraffin cancer tissues will be calibrated to yield strict quantitative findings for diagnostic and research purposes. To meet the goal of objective quantitation in tissue sections, spectral image analysis will be performed at the USC Image Core, for which we have shown the feasibility of quantitative analysis of up to four different analytes simultaneously in the same formalin-paraffin section. This proposal is multi-disciplinary, including collaborations with the USC Cellular Imaging Core at Childrens Hospital of Los Angeles, the Huntington Medical Research Institute in Pasadena, Roche Molecular Systems Inc., the National Institute of Standards and Technology (NIST), and the UNIVERSITY of California, San Francisco.