DOE Genome Contractor-Grantee Workshop IX
January 27-31, 2002 Oakland, CA
Low Dose Ionizing Radiation Abstracts
194. Damage Recognition, Protein Signaling, and Fidelity in Base Excision Repair
M. A. Kennedy1, M. K. Bowman1, G. W. Buchko1, P. D. Ellis1, J. H. Miller1, D. F. Lowry1, T. J. Straatsma1, Susan S. Wallace2, and David Wilson III3
1Battelle Pacific Northwest National Laboratory, Richland, WA 99352
Glycosylase studies: The base excision repair enzyme formamidopyrmidine-DNA glycosylas (FPG) from E. coli has been examined using solution-state NMR spectroscopy. Out of 252 possible amide proton-nitrogen correlations (269 residues - 16 prolines and the N-terminal amide), 209 amide cross peaks were observed (83%) and 180 of these were assigned (86%). Chemical shift perturbations have been observed for a subset of residues upon binding a 13 base pair DNA oligonucleotide containing a propane diol linker mimicking a nonhydrolyzable abasic site from which a chemical shift map of the DNA binding surface has been generated. Comparison of free-precession and CPMG HSQC data indicate the presence of slow millisecond to microsceond timescale motion for some residues in the absence of DNA that might be correlated with a hinge motion essential for DNA binding. Using similar analyses, attempts are being made to determine if DNA binding freezes the slow hinge motion observed in the absence of DNA.
APE1 studies: Histidine-aspartate (His-Asp) dyads are critical constituents in several key enzymatic reactions. The importance of these dyads is best exemplified in serine and trypsin-like proteases, where structural and biochemical NMR studies have revealed important pKa values and hydrogen-bonding patterns within the catalytic pocket. However, it has been debated whether the histidine in the His-Asp dyad can be partially charged, i.e. maintained at intermediate degrees of protonation. We used a new Oxford Instruments 21.1 Tesla superconducting magnet to measure the charge state of a critical histidine of apurinic endonuclease 1 (APE1), a human DNA repair enzyme that cleaves adjacent to abasic sites in DNA using an active site His-Asp pair. These observations imply that the role of the dyad in APE1 is not to directly bind a divalent cation as suggested. Rather, the dyad may activate a water molecule or bind DNA phosphate oxygen. EPR has also been used characterize the metal binding sites in APE1. Binding studies on APE-1 were investigated using oxovanadium, VO(II), as a surrogate probe for divalent metal binding sites. The CW-EPR spectra showed two different bound forms of VO(II) that could be explained either by coordination of three water and one hydroxyl as equatorial ligands to the oxovanadium or by three water and a histidine as ligands. Pulsed EPR spectra and 2D HYSCORE spectra showed water and histidine as direct ligands confirming the CW EPR analysis.
Pol b studies: Solid-state NMR is being used to characterize the metal environment in Pol b in relation to its catalytic function. We have refined our utilization of paramagnetic dopants to speed up the low temperature solid-state NMR experiment used to obtain magnesium spectra. We estimate that this has increased our data acquisition rates by a factor of 100. In addition, we have been examining the theory behind our experimental approach to in order to examine the potential of recovering Mg-Mg distance information (on the order of 3 to 5 Å). Such information will be critical to our understanding of the “relative rigidity” of the active site of Pol b during catalysis, i.e., does the active site remodel in the presence of either the correct or incorrect nucleotide triphosphate? Our initial results suggest we can obtain this information. Molecular dynamics simulations have been carried out for the human DNA Polymerase b with Gapped DNA, starting from the crystal structure reported by Sawaya et al. The reduced substrate ddCTP in the crystal structure has been replaced by dCTP, which is the correct nucleotide that is complementary to the template residue. This system was equilibrated for 400 ps, and a molecular dynamics simulation extending 1.5 ns has been completed. The molecular dynamics simulation was carried out using the massively parallel NWChem computational chemistry software, with the AMBER force field and with particle-mesh Ewald (PME) long range electrostatic contributions.
195. Low Dose Ionizing Radiation-Induced Effects in Irradiated and Unirradiated Cells: Pathways Analysis in Support of Risk Assessment
Bruce E. Lehnert, Robert Cary, Donna Gadbois, and Goutam Gupta
Bioscience Division, Los Alamos National Laboratory, Los Alamos, New Mexico USA
We are investigating how human cells respond to low dose ionizing radiation (LDIR), i.e., <1-10 cGy, with emphasis on comparing and contrasting effects in directly irradiated cells and “bystander” cells. Since advanced risk assessment modeling and prediction ultimately require an in-depth understanding of the complexity of cellular responses to LDIR as they pertain to untoward or benign effects, we are undertaking a “systems biology” approach to obtain such information. The specific aims of the project are: 1) to assess the temporal changes in gene expression in directly irradiated and bystander cells as functions of LDIR dose, cell types, cell culture conditions, and subsequent LDIR exposure conditions, 2) relate gene expression profiles in directly irradiated and bystander cells with the corresponding temporal and spatial changes in expressed proteins, and 3) determine how LDIR-induced extracellular and intracellular oxidative/reductive landscapes, signal transduction pathways, and gene and protein expression profiles regulate cellular responses. Our central hypothesis is that differences in the gene expression profiles and temporal and spatial patterns of key proteins expressed in directly irradiated and bystander cells critically determine how the cells ultimately respond to LDIR. Consistent with DOE’s new Genomes to Life initiative, knowledge gained from the project will contribute to the genesis of future hypothesis-driven investigations of candidate genes and proteins as they pertain to individual human resistance and susceptibility to the effects of LDIR, while additionally leading to the development of mechanistic models for predicting cancer risk.
196. The Application of Genome Data to the Important Problem of Risk from Low Dose Radiation
Antone L. Brooks
Washington State University, 2710 University Drive, Richland, WA 99352
Efforts to sequence the human genome and to make this information available to the scientific community are already paying great dividends. New genomics data and technology make it possible to address important societal issues in biology, medicine, and even in health risk. One application has been to apply these techniques to determine the cellular and molecular responses induced by low doses of ionizing radiation. Before the genome project, it was not possible to determine biological responses to very low levels of ionizing radiation (below about 0.10 Gy). The Low Dose Radiation Research Program funded by the DOE Office of Biological and Environmental Research was made possible by the merging of new technological developments with the genome research. The overall goal of this program is to provide a sound scientific basis for radiation protection standards. The program has been in place for just over three years and is currently funding 54 projects. There have already been several major breakthroughs resulting in a re-evaluation of basic radiation paradigms on which current radiation risk standards were set. These breakthroughs are a direct result of the gene chip and sequencing technology generated by the genome program. There is now evidence that cells do not require a direct “hit” to exhibit changes in gene expression, gene mutation and chromosome damage, but may also respond if a neighbor cell is irradiated, a phenomenon called the “bystander” effects. Such observations make it necessary for us to re-evaluate the effective biological target size for radiation and the significance of the long held “hit theory” of radiation biology. It has also been demonstrated that exposure of the matrix on which cells grow can change both the pattern of gene expression and the cells phenotype to result in cell transformation without direct induction of mutations. Therefore, the relative role of mutations and gene expression in cancer induction must be redefined. This may result in potential impacts on the basic linear-no-threshold hypothesis that is used in standard setting. Finally, low dose studies have demonstrated that the pattern and type of genes expressed after low doses of radiation are different from those observed after higher doses. Research has also shown that these patterns of gene expression influence many important genes involved in repair of DNA damage, as well as in programmed cell death (apoptosis). Results of recent studies suggest that low doses of radiation may decrease the level of spontaneous cell transformation resulting in another expression of the “adaptive response”. Without the advances in genomics most of these observations would not have been possible. Their impact on radiation risk and standards remains to be determined. However, the research from the Low Dose Program will provide a sound scientific basis for radiation risks. Continued application of new equipment, methods and techniques will be important in addressing many important scientific and societal needs.
Research funded by US DOE Grant DEFG0399ER62787 to Washington State University
197. Genome-Scale Modeling of Low-Dose Irradiation Responses Using Microarray Based Gene Networks
Matthew Coleman1, Terence Critchlow1, Mike Colvin1, Tom Slezak1, David Nelson1, and Leif Peterson2
1Molecular and Structural Biology Division, L-448, Biology and Biotechnology
Research Program, Lawrence Livermore National Laboratory, Livermore, California
Cells and tissues with similar radiation response phenotypes are predicted to have common ionizing radiation (IR)-induced gene expression profiles that are controlled by shared groups of regulatory elements. Our overall objective is to utilize genome-scale expression microarray data in conjunction with DNA sequence/pattern databases to build a computer-based gene-network model for identifying, grouping and predicting regulatory elements that control differential aspects of the early cellular responses to IR. This research project will develop a prototype model by: (a) Grouping genes identified by microarray experiments into IR-responsive clusters based on their relative-transcript abundance and their differential IR radiation responses at low (10cGy) and high doses and (b) Identifying regulatory elements (and their locations relative to the open reading frame) that distinguish among separate IR responsive gene clusters. To provide initial evaluation of our model, we will determine whether the identified regulatory elements are conserved across species and valid for microarray IR-responsive data sets from other laboratories. This model will be expanded and refined by including additional microarray expression data (low versus high dose responses, early versus late responses, adaptive response, and tissue/cell differences) and additional sequence data as they become available. A validated model will allow us to identify new human genes likely to be IR modulated and identify genes/pathways that are associated with different radiation response phenotypes (e.g., low dose sensitivity, adaptive response, sensitivity to chromosome damage, etc.) The identification and characterization of regulatory element profiles of IR-responsive genes will provide valuable understanding of the genetic mechanisms of IR-response and should provide powerful biological indicators of genetic susceptibilities for tissue and genetic damage.
198. Molecular Mechanisms and Cellular Consequences of Low-Dose Exposure to DNA-Damaging Agents
Andrew J. Wyrobek1, Matthew Coleman1, Eric Yin1, Francesco Marchetti1, Sandra McCutchen-Maloney1, Allen Christian1, David Nelson1, Irene Jones1, Larry Thompson1, Leif Peterson2, and Jian-Jian Li3
1Molecular and Structural Biology Division, L-448, Biology and Biotechnology
Research Program, Lawrence Livermore National Laboratory, Livermore, California
It is well established that high-dose exposures to ionizing radiation (IR) leads to diverse pathologies of skin and other tissues and to serious late-onset diseases including cancers. The long-term objectives of this research are to investigate the early cellular effects of low-dose exposures (1-10cGy) in human and mouse cells and to determine whether the early changes in gene expression (mRNA or protein) of specific gene/pathways are associated with subsequent risk for cytogenetic damage. In this ongoing project, we utilized both LLNL-manufactured cDNA arrays and commercial oligonucleotide arrays to identify hundreds of low-dose (10cGy) responsive genes in irradiated human lymphoblastoid (HLB) cell lines and mouse brain. We also identified a group of human genes that show transcriptional changes associated with adaptive responses (AR) and identified several human protein peaks by SELDI mass-spectrometry. In parallel experiments in the mouse, we established baseline tissue variations in the expressions of DNA repair and stress response genes, identified hundreds of genes whose expression was modulated after 10cGy brain irradiation, and studied Rad54 KO mice to suggest that double strand break (DSB) repair is associated with altered expression response. The renewal project will: (a) use genome-scale microarrays to survey the human and mouse genome for low dose (1-10cGy) radiation-responsive genes (b) investigate the nature of dose- and time-response profiles across mouse tissues, human lymphoblastoid (HLB) cells, and fresh human lymphocytes; (c) identify and characterize genes whose changes in early expression are associated with adaptive response, beginning with candidate genes already identified by their mRNA and the protein levels; and (d) continue our studies in mice of tissue variations and the role of DSB repair in radiation response. This project will provide new knowledge of the early cellular responses to low-dose IR to reduce the uncertainty of assessing risk at low-dose levels. It is also expected to identify genes whose expression is associated with low dose IR exposure and susceptibility for adaptive response.
199. Molecular Mechanism of the 9-1-1 Checkpoint Response to DNA Damage Based on Protein Structure Prediction
Ceslovas Venclovas, Michael Colvin, and Michael P. Thelen
Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory
Deciphering the molecular mechanisms of DNA replication and repair is hindered by the lack of detailed structural information for the protein machinery involved. For structurally uncharacterized proteins, modeling by comparison can generate hypotheses for fundamental DNA repair mechanisms, serving as a powerful guide for experimental design. Here we describe such a three-dimensional structure prediction for two interacting complexes of nine proteins that are central to the DNA damage checkpoint in eukaryotic cells.
The checkpoint response to DNA damage requires Replication Factor C (RFC), a hetero-pentameric protein complex that is essential to eukaryotic replication. Current experimental evidence indicates that during normal DNA replication, RFC binds to primed DNA and uses ATP to drive the loading of PCNA, the sliding clamp that tethers polymerase to DNA for processive DNA synthesis. Upon DNA damage, Rad17 complexes with RFC and causes it to load a different sliding clamp protein, the Rad9-Rad1-Hus1 (9-1-1) heterotrimer. Based on our comparative modeling studies and more recent experimental observations, the 9-1-1 complex behaves similarly to PCNA, as a polymerase processivity/fidelity factor, but is specifically needed when the DNA template contains breaks or adducts that would halt normal replication.
Five distinct RFC subunits and Rad17 have detectable sequence similarity to each other and to functionally analogous proteins from bacteria and archaea. We used recently determined structures, including the RFC small subunit from Pyrococcus furiosus, to produce high confidence, all-atom models for each RFC subunit. The quaternary structure of the RFC complex was then assembled by analogy to the E. coli clamp loader, or gamma complex. Constraints derived from available structural, biochemical and genetic data were used to predict relative positions for the individual Rad17 and RFC subunits within the complex. The resulting architectural model of Rad17/RFC includes interactions with the 9-1-1 ring, and with DNA and ATP, leading to a first approximation of the eukaryotic mechanism for dealing with DNA damage during the cell division cycle. The model indicates specific protein-protein contact regions that could be tested by site-directed mutation.
Work performed under auspices of U. S. Department of Energy by the University of California, LLNL under Contract No. W-7405-Eng-48
200. Phylogenetic Analysis of Two Human Proteins that are Homologues of Proteins Involved in Base Excision Repair, Formamidopyrimidine DNA Glycosylase and Endonuclease VIII
Sirisha Sunkara, Susan S. Wallace, and Jeffrey P. Bond
Department of Microbiology and Molecular Genetics, Markey Center for Molecular Genetics, University of Vermont, Stafford Hall, Burlington, VT 05405-0068, USA
It is important to identify human proteins involved in DNA repair because understanding DNA repair is essential for understanding carcinogenesis. We report identification of two human proteins that are members of a family of proteins involved in base-excision repair, the formamidopyrimidine DNA glycosylase (Fpg)/endonuclease VIII (Nei) family. Phylogenetic analysis suggests that an ancestor of plants, fungi, and metazoa possessed Fpg and that a monophyletic group of Nei sequences exists. On the basis of phylogenetic analysis we classify one of the sequences as Fpg and the other as Nei. Analysis of alignments of the sequences of the human proteins with previously identified Fpg and Nei sequences in the context of the structure of a bacterial Fpg suggests that the human proteins have structures similar to that of the bacterial Fpg. In particular, sequence similarities between the human sequences and previously identified Fpg/Nei sequences include highly conserved regions associated with each of two Fpg structural domains and amino acids involved in catalysis. Eukaryotic Fpg sequences are missing the zinc finger motif seen in bacterial Fpg. Human Nei has a C-terminal extension that includes a second zinc finger motif and a region similar to portions of certain AP endonuclease and topoisomerase sequences.
The online presentation of this publication is a special feature of the Human Genome Project Information Web site.