Portal de Periódicos da CAPES

Chemistry in Cancer Research: A Vital Partnership

2007; American Association for Cancer Research; Volume: 67; Issue: 14 Linguagem: Inglês

10.1158/0008-5472.can-07-1619

1538-7445

Shana J. Sturla, John J. Irwin, Richard N. Loeppky, Mark J. Mulvihill, Mark Searcey,

Cancer Treatment and Pharmacology

The first Chemistry in Cancer Research Conference jointly presented by the American Chemical Society (ACS) and American Association for Cancer Research (AACR) was hosted by the Chemistry in Cancer Research Working Group (CICR) of the AACR, February 4 to 7, 2007, in San Diego, CA. The purpose of the meeting was to provide an interactive and cross-disciplinary forum for scientists engaged in cancer research with a unifying focus on chemical structure and reactivity. Research areas spanned fields, such as biomarker analysis, carcinogenesis, chemical biology, drug discovery and development, molecular modeling, proteomics, and structural biology. An outstanding group of 26 invited talks, including two keynote talks, plus 22 proffered talks from selected abstracts, including early career scientists and students, and 109 poster presentations comprised this stimulating meeting. There were over 240 registrants, including 122 ACS members, 31 AACR members, and 42 members of both organizations. Travel awards were provided for 13 scholars-in-training and one minority-serving institution faculty. This meeting report presents highlights from each oral presentation, organized by session. A complete listing of oral and poster session presentations can be found in appendices. Further details, including reaction schemes with chemical structures available for some presentations, are available as supplementary data.A tribute was shared in memory of Christopher J. Michejda, who directed the chemistry and structural biology faculty of National Cancer Institute (NCI)'s Center for Cancer Research and who died suddenly 1 month before the meeting. Chris was a founding CICR member and a strong proponent of the power of chemistry in cancer research. He is remembered for his diverse scientific contributions in areas including cancer therapy and nitrosamine carcinogenesis, his infectious personality, and a belief in the potential of science to serve humanity.The meeting was opened by Stephen Fesik (Abbott Laboratories, Abbott Park, IL), who stressed that the biggest premium for mistakes in drug development are involved in clinical trials and that any preclinical information with the potential to influence clinical trials is extremely important and valuable. The research centered on the Bcl 2 family of proteins and ABT 737, a candidate that potentiates chemotherapy and radiotherapy and displays single-agent activity in cell lines. Understanding the mechanistic basis of unresponsiveness to ABT 737 in certain cell lines can guide patient population selection and the design of improved therapies. For example, high Mcl 1 is associated with resistance to ABT 737 and current work is under way to design a drug to reduce Mcl 1 that may synergize with ABT 737. An orally active ABT 737 analogue is currently in phase I trials; and Abbott plans to move forward with single-agent trials, expecting that an understanding of mechanism will influence the success of trials. Paul Wender (Stanford University, Stanford, CA) discussed that new therapies from new chemistries, diagnostic methods, and prevention strategies are anticipated to revolutionize cancer research. Chemistry in cancer research has evolved from using available material to a rational approach to designed molecules that achieve a desired function and can be obtained in a practical way: function-oriented synthesis. Illustrative examples included the apoptolidins and bryostatins. Simpler natural product analogues drastically reduce the number of synthetic steps to active molecules. A final concept centered on breaching biological barriers, with examples of drug/probe-transporter conjugates to enhance bioavailability and change efficacy of existing agents. Wender shared thoughts regarding the value of people involved in research, training programs, methods to bridge academia and industry, and the problems of limited resources available to support the richness of researcher ideas. After the lecture, it was discussed that “rules” in medicinal chemistry are good guidelines but that it is important to be open to eclectic agents, ultimately seeking the “biggest bang” from the smallest amount of chemical information.The drug discovery session began with Dale Boger (Scripps Research Institute, La Jolla, CA) and the message that problems dictate the choice of chemistry and approaches used. For example, with scarce structural information and without known leads, combinatorial chemistry is a powerful tool. Natural products studies allow the definition of the biological target, the nature of the interaction with the target, and the origin (if any) of tumor selectivity. Finally, structure-based drug design can generate novel design concepts. Dr. Boger enlarged on studies in each area, beginning with the structure-based design of glycinamide ribonucleotide transformylase inhibitors. Dr. Boger described solution-phase combinatorial chemistry producing a library of 80,000 compounds, at about the limit to which one could expect an academic laboratory to screen. Finally, DNA-binding duocarmycins illustrated that compounds need to be sufficiently stable to reach their target but sufficiently reactive to alkylate the target when they arrive. The next presentation reiterated the importance of target-structure driving chemisty. Gustavé Bergnes (Cytokinetics, San Francisco, CA) presented the development of motor kinesin spindle protein (KSP) inhibitors with a philosophy of “design, synthesis, assay, and learn.” Ispinesib is >70,000-fold selective for KSP versus other kinesin proteins, and its progression into clinical trials is a significant development for a small to medium company. Dr. Bergnes described further goals to anticipate potential risks and pitfalls that can block progress to the clinic. Thomas Penning (Abbott Laboratories) described an inhibitor of the DNA repair enzyme poly(ADP-ribose) polymerase that crosses the blood-brain barrier, potentiates temozolamide, and is currently in the first ever phase 0 clinical trial. Mark Mulvihill (OSI Pharmaceuticals, Inc., Farmingdale, NY) described the development of inhibitors of insulin-like growth factor I receptor (IGF-IR), a transmembrane tyrosine kinase. Using a cocrystal structure of an imidazopyrimidine bound to the protein, “northern” and “southern” domains that had potential for variation were identified. The northern domain contained a hydrophobic pocket that was amenable to locking the structure into a rigid quinoline ring system, whereas the southern domain was altered to a cyclobutane structure. The resulting compound has an excellent pharmacokinetic profile in mice and 90% inhibition of IGF-IR phosphorylation with significant tumor growth delay and no effect on blood glucose; OSI-906 is currently in clinical development. Louis Lombardo (Bristol-Myers Squibb, Princeton, NJ) described the Bristol-Myers Squibb approach to the development of vascular endothelial growth factor receptor inhibitors, followed by dasatinib, a marketed kinase inhibitor (Sprycel) for the treatment of chronic myelogenous leukemia (CML). Since the advent of imatinib mesylate (Gleevec or Glivec), research on kinase inhibitors has exploded and is a major part of the new drug pipeline. Compounds that bind the ATP pocket, however, have potential selectivity problems. A key to the success of dasatinib lies in the lack of cross-resistance with imatinib. Patrick Harran (UT Southwestern Medical Center, Dallas, TX) described how serendipity and a prepared mind could lead to unexpected discoveries of new antitumor agents. Serendipitous dimers of a second mitochondria-derived inhibitor of caspases (Smac) analogue, with nonpeptidic replacements for proline and phenylalanine, exhibited enhanced efficacy in cell-based functional assays. Work of this nature is reestablishing the balance between replication and death, the ultimate aim of all anticancer programs. Eileen Kennedy (Harvard University, Cambridge, MA) described hydrocarbon-stapled peptides to target inhibition of the androgen-receptor ligand binding domain. Frank Gu (Massachusetts Institute of Technology, Cambridge, MA) described nanoparticles for targeting prostate cancer. This series of presentations showed the range of approaches that chemists in industry and academia are applying to the discovery and potentiation of new molecules with therapeutic potential.Proteomics, the study of the whole of the protein content of the cell, has the potential to revolutionize the way cancer diagnosis and treatment is approached. This session showed the potential of proteomics to make a significant contribution to the identification of new targets and the effects of new drugs in both tumor cells and tissues. However, it also clearly highlighted the ingenuity of the various groups working in the area in solving problems associated with the analysis of information derived from studies of the global expression of proteins. Catherine Fenselau (University of Maryland, College Park, MD) began the session with the statement that “chemistry enables proteomics” and described protein changes in MCF-7 tumor cells that are resistant to various antitumor agents. The strategy focuses on specific subcellular organelles and incorporation of 18O, using labeled water for quantitation. Thus, protein-labeling in normal cells (forward labeling) is followed by protein-labeling in resistant tumor cell lines (reverse labeling). Ben Cravett (Scripps Institute, San Diego, CA) described activity-based protein profiling, which, when combined with mass spectrometry (MS), can aid in the identification of proteins with varying activities between cell types, such as metastatic versus nonmetastatic. Probes that carry a reporter group bind and label enzymes in the proteome with a dependence on activity. The laboratory aims to develop an integrated enzyme and metabolite profile for tumor cells. James Veal (Serenex, Durham, NC) gave insight into an approach that uses chemoproteomics and proteome mining and showed an application to drug discovery involving an HSP90 inhibitor that was refined and optimized to generate ultimately a clinical candidate. Colin Barry (University of Pennsylvania, Philadelphia, PA) discussed proteomic analysis of the folate-homocysteine pathway using a modified stable isotope labeling of amino acids by cell culture approach. William Griffiths (University of London, London, England) described cytochrome P450 (CYP) expression in tumor cells and potential implications for individualized drug and prodrug administration as the basis of personalized medicine in cancer, a pervasive concept addressed throughout the conference. A comparison of the CYP expression profile of the microsomal fraction of primary colorectal cancer tissue, liver metastases, and surrounding liver tissue shows differences in individual CYPs. Forest White (Massachusetts Institute of Technology, Cambridge, MA) addressed the general question of how phosphorylation regulates biological response. He discussed phosphoproteomics or global analysis of protein phosphorylation by looking at the epidermal growth factor receptor signaling pathway. Virginia Espina (George Mason University, Manassas, VA) discussed the importance of preserving tissue samples in clinical settings, for example, in the development of personalized cancer therapy. Finally, Yuebiao Yao (Morehouse School of Medicine, Atlanta, GA) discussed the use of two-dimensional gels in colorectal cancer to analyze primary versus metastatic tumors.Chemists have played and continue to play an important role in cancer prevention through research on carcinogenesis. Traditionally, this has involved various fields of chemistry: development of sensitive instrumentation and methods to gauge exposure, organic and inorganic chemical model studies on mechanisms of carcinogen formation and bioactivation studies, the determination of putative carcinogen metabolic paths, the elucidation of endogenous pathways which may have carcinogenic potential, and the generation and utilization of relevant synthetic procedures. Whereas accidental exposures and subsequent clear-cut epidemiology resulted in the positive identification of a small number of human carcinogens, many more substances are animal carcinogens. How can the true human carcinogenic potential of a substance or an endogenous chemical or biochemical process be assessed? The presentations given at the Chemical Biology of Carcinogenesis, as well as several in the Biomarkers sessions of this conference, show current research in this field. All emphasized the elucidation of DNA/protein modification and the consequences and significance of this chemistry. Tamoxifen, a selective estrogen receptor modulator (SERM), is used in treatment and prevention of breast cancer and osteoporosis but has been associated with an increased risk of uterine cancer. Dr. Judy Bolton discussed microsomal bioactivation of benzothiophene SERMs being examined as safe replacements for tamoxifen. Cellular protein targets were found by the invention and utilization of COATAG methodology: the SERM is covalently linked to a biotin tag, permitting immunoisolation and identification. Dr. Bolton emphasized that covalent SERM-protein binding may not be strictly deleterious but could contribute to chemoprevention through Keap1 alkylation and induction of detoxification enzymes. Guanine bases in DNA are readily oxidized to their 8-oxo derivatives by biologically relevant reactive oxygen species (ROS) but the carcinogenic consequences of this chemistry are not clear. Cynthia Burrows (University of Utah, Salt Lake City, UT) reported on her group's work directed at understanding how the products and mechanisms of guanine oxidation may provide a basis for mutagenesis and the possible role of DNA-protein cross-linking therein. Yelena Margolin (Massachusetts Institute of Technology) presented another aspect of base oxidation in DNA: the chemical basis of site selectivity for oxidation by ROS and other oxidants. Sensitive assays developed by Lawrence Marnett's group have shown surprisingly low levels of an endogenous oxidative damage–associated nucleoside adduct in human urine. Charles Knutson (Vanderbilt University, Nashville, TN) presented data from 13C-labeling studies showing the metabolism and excretion of the M1dG adduct, which is expected to affect biomarker studies of oxidative DNA damage by this pathway. John Essigmann (Massachusetts Institute of Technology) began the second session with a discussion of the value of chemically based mutagenesis studies. By site-specifically modifying DNA oligomers with modified bases and incorporating them into viral genomes, Dr. Essigmann and his group have developed a method producing a “fingerprint” for each type of adduct which evaluates its replication efficiency, whether it induces the error prone SOS response, types of mutations, and relative frequencies. Essigmann discussed chemical mechanisms of repair by the AlkB protein, which may protect bacteria against macrophage action (humans have homologues of AlkB). Fred Guengerich (Vanderbilt University School of Medicine) used his group's research to illustrate complex connections between the chemistry and biochemistry of reactive intermediates and mutagenesis. For example, aflotoxin, a potent human carcinogen, is activated by CYP-mediated conversion to two stereoisomeric epoxides. Although the more reactive isomer has a t1/2 of only ∼1 s, it is vastly more genotoxic and binds DNA efficiently. Dr. Guengerich presented his group's extensive research concerning interactions of DNA polymerases with adducted bases, examining the effects on insertion and extension of the strand being copied by a translesional DNA polymerase. The process of establishing a suspected carcinogen to be a human carcinogen is an arduous task involving varied approaches. IARC has declared the tobacco-specific carcinogens, NNN and NNK, to be human carcinogens due in significant measure to the extensive research of Stephen Hecht and his collaborators. However, given the complex and numerous protective mammalian defenses against carcinogenesis, biomarkers and the nature of DNA adducts need to be established. An illustration was provided by the presentation of Pramod Upadhyaya (University of Minnesota, Minneapolis, MN), concerning CYP-mediated metabolism. Recent concerns have been raised about the possible genotoxicity of acrylamide, which is found in some fried foods, because it is activated through the corresponding epoxide. Further, it is structurally akin to carcinogens described by Dr. Guengerich. Matthias Baum (University of Kaiserslautern, Kaiserslautern, Germany) described comparisons of the genotoxicity of glycidamide in several different assays and with N-nitroso compounds known to be potent alkylating agents.The session on Biomarkers and Analytic Chemistry involved a range of experimental approaches for the identification and analysis of chemical biomarkers important in cancer. Ian Blair (University of Pennsylvania) described reaction pathways related to the overproduction of ROS with a focus on understanding the effects of this process when various enzymes involved are up-regulated, such as in cells under oxidative stress mediated by the overproduction of the cyclooxygenase-2 enzyme. Reactions of specific lipid peroxidation products with DNA and reduced glutathione (GSH) produce adducts. Moving from DNA and GSH to proteins, Daniel Liebler (Vanderbilt University School of Medicine) continued the discussion of lipid peroxidation and addressed problems and solutions in identifying targets. Using conjugated lipid-biotin probes, adducts were surveyed using streptavidin binding and proteins were identified by MS, an approach analogous to the COATAG method discussed earlier in Dr. Bolton's talk. Helpful analytic strategies were suggested, such as the construction of model peptides to determine MS behavior, mild ammonia hydrolysis methods, and wide-band activation. Elizabeth Grimm (University of Texas, M.D. Anderson Cancer Center, Houston, TX) introduced iNOS, an enzyme that catalyses intracellular NO production from l-arginine, and has carried out studies to establish iNOS expression as a predictive marker that correlates inversely with survival in stage-3 melanoma patients. The presentation of Chenguo Xing (University of Minnesota) directly related to Dr. Fesik's talk, i.e., apoptotic regulation by the Bcl2 family of proteins. Dr. Xing has established that HA 14-1 can synergize various standard chemotherapies; and data indicate that HA14-1 induces cell death by two distinct pathways, possibly converting antiapoptotic Bcl2 to a proapoptotic form. Natalia Tretyakova (University of Minnesota) transitioned the focus to exogenous carcinogens. Dr. Tretyakova explained that the isomers of diepoxybutane influence resulting patterns of DNA alkylation. In the questions-and-answers portion of Dr. Tretyakova's talk, links between findings for diepoxybutane and nitrogen mustards and the CML drug sulfam were discussed. Following on the theme of DNA-small molecule adduct formation, Peter Farmer (University of Leicester, Leicester, England) presented an overview of current analytic methodologies. The focus was on MS as an approach for obtaining an accurate indication of dose that takes into account individual biological differences. Dr. Farmer established a detection limit benchmark of one damaged base in 108 nucleotides or less. Comparing adduct levels between MS methods and 32P postlabeling indicated that the methods correlate, but in some cases 32P assay may underestimate adduct levels. Finally, Dr. Farmer presented a new site selective mutation assay as a variant on the SupF assay for human cells. It was concluded that future studies should address the biological influences of low levels (one or two adducts per cell) of DNA damage. Roberto Diaz (Vanderbilt University) addressed the identification of peptides recovered from in vivo biopanning of enzyme libraries from animal tumors, and Surojit Sur (Johns Hopkins Medical Institutions, Baltimore, MD) presented an example of drug discovery in a multidisciplinary academic setting.Throughout the conference, computational strategies aided in the design and evaluation of cancer therapies and a session was focused on emerging methods and applications in this area. In the first talk, David Covell (NIH, Frederick, MD) presented bioinformatic and chemical informatics strategies for mining the screening data from the “NCI60” panel of 60 immortalized human-derived cell lines that have been accumulated at NCI over the past 19 years. Recently, cell screening has been augmented with high content DNA, protein microarray, and xenograft data resulting in a database with over two million datapoints. Interpretation of data linking small molecules, molecular targets, and mechanism of action remains an area of active research. John Irwin (University of California, San Francisco, CA) talked about the ZINC database, a collection of commercially available compounds in biologically relevant representations for virtual screening and chemical informatics applications. Dr. Irwin discussed the importance of the representation of molecules in their bioactive forms, particularly of protonation and tautomeric variants, as well as forms appropriate for metalloenzymes. He presented the Directory of Useful Decoys database for benchmarking docking programs, challenging them to expose weaknesses and as a basis for improving docking methods. James Wright (Carleton University, Ottawa, ON, Canada) talked about the problem of quinone formation via P450 metabolism of estradiol used in hormone replacement therapy. A goal is to develop new estrogen mimics with good relative bioavailability and that cannot form dangerous quinones. Lei Jiam (New York University, New York, NY) discussed molecular modeling and dynamics studies of oxidative lesions in hNeil1, part of the base excision repair pathway. William Jorgensen (Yale University, New Haven, CT) discussed computer-aided lead generation and optimization focusing on nonnucleoside inhibitors of HIV reverse transcriptase as a therapeutic context. Jorgensen described the Biochemical and Organic Model Building program, which generates conformers of a dynamically created combinatorial in silico library. His approach trains a scoring function against experimental data, and ligands are filtered for properties (i.e., solubility, cell permeablity) using Qikprop and Monte Carlo/free energy perturbation to refine predictions including water molecules and flexible ligand and receptor. Brian Shoichet (University of California) presented virtual screening for structure-based inhibitor discovery, describing simple, artificial protein binding sites to study the performance of docking programs. Docking works well for simple binding sites but poorly against real drug targets. Experiments to address why involved docking against AmpC β lactamase in a head-to-head comparison of docking and HTS in collaboration with the National Chemical Genomics Center. Of 70,500 compounds screened, 1,274 were primary hits; but after follow-up, none were single, noncovalent, reversible, competitive inhibitors. Docking the same library resulted in 16 compounds picked to test, of which two were found to be active, the best with a Ki of 37 μM. Lihua Wang (New York University) has used QM/MM calculations to investigate catalytic mechanisms of a DNA polymerase. Mark Klein (University of Minnesota) discussed the use of a nuclear magnetic resonance (NMR) and alanine scanning-derived three-dimensional pharmacophore to search the NCI database for compatible small molecules, filtered for “drug likeness.” The resultant collection was docked to the NMR model using Glide, from which top scoring ligands were selected as putative CDK4/6 inhibitors and experimental testing of the docking predictions is under way.The structural biology talks involved NMR and X-ray crystallography studies of diverse proteins with a focus on novel techniques, gaining an understanding of mechanism and the design of small-molecule inhibitors. These are highlighted here, and details from the studies presented are described in the supplementary data. Andrew Byrd (NCI, Frederick, MD) presented high-field (900 MHz) NMR work centered on the STAT proteins, a family of cytoplasmic proteins that mediate cellular responses to cytokines via membrane receptors coupled to Janus-activated kinase. Their large size (∼100 kDa) has limited solution NMR studies, and structures have been solved primarily by X-ray crystallography. On the basis of Dr. Byrd's recent solution structure of STAT4-NT, he concluded that both X-ray and NMR dimer arrangements can exist in solution with both being relevant. Stephen Burley (Structural GenomiX, San Diego, CA) provided an overview of the company and its capabilities, highlighting its fragment-based drug discovery approach, noting that many of the fragment hits are more selective than one would predict based upon their low molecular weight and lack of complexity. Wei Yang (NIH-National Institutes of Diabetes, Digestive and Kidney Diseases, Bethesda, MD) presented the application of X-ray crystallography, molecular biology, and various biochemical and biophysical approaches to elucidate the molecular mechanisms behind the Y-family of DNA polymerases. These catalyze template-dependent DNA synthesis with low fidelity and low processivity for normal DNA but synthesize past-damaged template bases. The structures reveal a conventional right hand–like catalytic core, with unusually small finger and thumb domains resulting in an open and capacious active site. This information is central to understanding biological effects of DNA-damaging agents as discussed in the earlier chemical biology of carcinogenesis session. Ann McDermott (Columbia University, New York, NY) presented solid state NMR studies of binding and function of membrane-associated proteins. NMR spectra of uniformly labeled (15N, 13C) solid state proteins can be well resolved and may provide a basis for structural and functional studies. Her group uses a magic-angle spinning technique (40 μL of solvent with a few milligrams of protein) and has studied many small proteins. Ian Hardcastle (Northern Institute for Cancer Research, Newcastle University, Newcastle upon Tyne, England) presented on his work regarding small molecule inhibitors of MDM2-p53 protein interactions. Inhibition is expected to reactivate normal p53 pathways in cells overexpressing MDM2, causing proapototic effects. HSQC NMR structural studies indicated a plausible binding mode for a compound in complex with MDM2. Ann Alcaraz (Emory University, Atlanta, GA) presented on the structure of colchicine and a model that indicates a new ligand binding site.A lunchtime session was held to discuss chemistry careers in cancer research. Information was presented by Dr. John Hunt (Bristol Meyers Squibb) and Dr. Lawrence Marnett (Vanderbilt University). Dr. Hunt discussed recent trends in the industry, including an increasing integration of science throughout the drug pipeline. Dr. Marnett overviewed traditional tenure track positions involving some combination of teaching and research, as well as research professorships and opportunities in core academic research facilities. He gave useful hints and advice for developing research ideas and seeking funding. Representatives from academia and industry participated in roundtable discussions with participants.A number of session- and discipline-spanning concepts emerged during the meeting, including predictive strategies for responsiveness of individual cancers to specific therapies, a focus on correlating chemical structure and target selectivity, and chemical pathways in drug toxicity and carcinogenesis. The meeting was considered highly successful by participants and a consensus called for developing a regularly scheduled event. At the AACR Annual Meeting in April 2007, plans for the meeting to become a biennial event were announced, with the next to be held in Spring 2009. Participants indicated that the ACS-AACR union represented by this conference marked a beginning of alliances that will greatly benefit cancer research.The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.We acknowledge the conference chair, Stephen S. Hecht, and the program committee (see supporting information). We thank the AACR staff for preparation of the appendices.