This is an old, no longer maintained version of the website, for new one please visit: http://gsn.io.gliwice.pl

		Gliwickie Spotkania Naukowe 2005 Gliwice Scientific Meetings 2005 18-19 November DIAGNOSIS AND THERAPY OF BRAIN TUMORS Jan Barciszewski1, Stanislaw Nowak2 and Ryszard Zukiel2 1Institute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland and 2Department of Neurosurgery and Neurotraumatology, Karol Marcinkowski University of Medical Sciences, Poznan, Poland Neoplasms arising from glial cells make up the most common group of primary brain tumors. The clinical outcome, especially the survival rates of the patients with tumors depend on tumor grade expressing their malignancy grade. A prognosis for glioblastomas (WHO IV) is very poor, but for astrocytomas (WHO I and II) it is relatively favourable. There are evidence that oxidative stress and reactive oxygen species (ROS) are crucial in the etiology and progression of a number of human diseases, also neoplasms. An oxidative damage of DNA, lipids and proteins is caused mainly with hydroxyl radical (#183;OH), the most reactive ROS species and may be seriously deleterious. In the reaction with hydroxyl radical all DNA components can be modified. 5-methylcytosine (m5C) a rare but normal component of cellular DNA is relatively easily deaminated to thymine, which pairs with adenine and after a round of replication, CG to TA transition occurs. Because thymine is a normal DNA base, therefore the product of spontaneous deamination of m5C is not so easily detected by a cell's DNA repair system. Thus, 5-methylcytosine residue constitutes a mutational hotspot and DNA methylation pattern in patients might be useful as a primary diagnosis tool or as a marker for early detection of relapse of the disease. In recent years a new mechanism of posttranscriptional gene silencing called also RNA interference (RNAi) has been discovered. This phenomenon is based on specific mRNA degradation mediated by double-stranded RNA molecules, approximately 19-28 nucleotides in length, called short interfering or siRNAs. These molecules are produced from long dsRNAs by a dsRNA-specific endonuclease (DICER) and form 300 kD multi-enzyme complex (RISC) which by Watson-Crick base-pairing of noncoding strand with their mRNA-targets induce the specific cleavage. The high sequence-specificity of RNAi shows a new, promising tool in gene-function analysis as well as in potential therapeutics development. PROTEIN STRUCTURE PREDICTION BY COMBINATION OF FOLD-RECOGNITION WITH DE NOVO FOLDING Janusz M. Bujnicki1,x, Marcin Feder1, Michal J. Gajda1, Jan Kosinski1, Marcin Pawlowski1, Michal Boniecki1,2, Dominik Gront2, Andrzej Kolinski2 1Laboratory of Bioinformatics and Protein Engineering, International Institute of Molecular and Cell Biology in Warsaw, Poland, http://genesilico.pl. 2Laboratory of Theory of Biopolymers, Faculty of Chemistry, University of Warsaw, Poland xe-mail: iamb@genesilico.pl A new method for protein structure prediction was developed, which allows modeling regardless of the potential homology with any known protein structure. This method is a combination of the "FRankenstein's Monster" approach for comparative modeling by recombination of Fold-Recognition (FR) models [1], and a new implementation of a Replica Exchange Monte Carlo (REMC) method for protein folding de novo or with restraints [2,3]. The sequence of a modeled protein is submitted to the GeneSilico structure prediction meta server, which is a gateway to a variety of third-party methods for secondary structure prediction and FR analysis ( http://genesilico.pl/meta/ [4]). FR alignments are compared and ranked and up to 5 most frequently reported folds are selected for further analysis. For each fold, the target-template alignments are used as a starting point for modeling using the "FRankenstein's monster" approach [1]. Best models obtained (1-15 models for each fold) are evaluated and fragments with best scores are used to derive spatial restraints [5]. Additional restraints can be derived from methods for de novo structure prediction, such as ROSETTA. Secondary structure restraints are derived from the consensus of methods implemented in the GeneSilico meta server [4]. Seconday and tertiary restraints are used to guide the REMC folding simulation using a new high-resolution reduced lattice model CABS [2,3]. The conformations obtained in the course of CABS simulations are subject to the average linkage hierarchical clustering. For a represetnative structure from each cluster a full-atom representation is rebuilt. The performance of the new method will be discussed in the context of our successful predictions in the recent CASP-6 experiment. 1. Kosinski, J., Cymerman, I. A., Feder, M., Kurowski, M. A., Sasin, J. M., and Bujnicki, J. M. (2003) Proteins 53 S6, 369-379 2. Boniecki, M., Rotkiewicz, P., Skolnick, J., and Kolinski, A. (2003) J Comput Aided Mol Des 17, 725-738 3. Kolinski, A. (2004) Acta Biochim Pol 51, 349-371 4. Kurowski, M. A., and Bujnicki, J. M. (2003) Nucleic Acids Res 31, 3305-3307 5. Sasin, J. M., and Bujnicki, J. M. (2004) Nucleic Acids Res 32, W586-589 THE ROLE OF CROSSING-OVER IN THE COMPUTER SIMULATED POPULATIONS Stanislaw Cebrat Institute of Genetics and Microbiology, University of Wroclaw, ul. Przybyszewskiego 63/77, 51-148 Wroclaw, Poland e-mail: cebrat@microb.uni.wroc.pl We have used the Penna ageing model, based on the Monte Carlo method to study evolutionary significance of the linkage disequilibrium as well as the inside intron recombination.In our computer simulations we have implemented internal structure of genes by declaring the probability of recombination between exons. There are two trivial situations when organisms could profit from the inside intron recombination. The first one is generation of new genes by exons reshuffling. The second one is recovering functional genes from exons dispersed in the genetic pool of the population after a long period without selection for the function of this gene. Populations have to pass through the bottleneck, then. These events are rather rare and we have expected that there should be other phenomena giving profits from the inside intron recombination. In fact we have found that inside intron recombination is advantageous only in the case when after recombination, besides the recombinant forms, parental haplotypes are available and selection is set already on gametes. Nevertheless, parameters of simulations when inside introns recombination is advantageous are very rigorous. That is why we tried to use simpler models of population evolution to study the conditions when the inside introns recombination brings direct profits for evolving populations. EXAMPLES OF APPLICATION OF MASS SPECTROMETRY IN PROTEOMICS Michał Dadlez Department of Biophysics, Institute of Biochemistry and Biophysics, PAS, ul. Pawinskiego 5A, 02-106 Warsaw, Poland Proteomics became an integral part of systems biology, providing a necessary complement for genomic approach. The aim of proteomics is defined as enumeration of protein complement of the sample of interest along with their posttranslational modifications. Moreover protein-protein interaction nets can also be elucidated. The lecture will introduce to proteomics technology and mass spectrometry as a basic proteomic tool. Examples of application of proteomics technology will be given. Mass spectrometry (MS) has emerged in the recent years as the main proteomic tool due to its sensitivity. At present subfemtomole (10-15 M) of a protein are sufficient for identification. In addition MS analyses are fast and do not require homogeneity of the sample. On the contrary, in MS technology complex mixtures of proteins can be studied in a single experiment. Proteomic technology application will be illustrated by several results originating from our Laboratory. Successful examples of protein identification along with their posttranslational modifications will be presented. The application of differential proteomics for identification of a new protein involved in pathogenesis of cystic fibrosis and identification of partners of human helicase protein complex will shortly be outlined. Finally, the application of multidimensional liquid chromatography prefractionation coupled to MS (MudPIT) will be illustrated by the study of the contents of Arabidopsis nuclear proteome and the human blood plasma peptidome mapping. A SIMPLE MATHEMATICAL MODEL FOR CELL SIGNALING IN RADIATION EXPERIMENTS Philipp Getto Institute of Automatic Control, Silesian University of Technology, Gliwice, Poland As a reaction to irradiation cells send signaling molecules/ligands/chemicals, which influence the functioning of the other cells (bystander effect). In the ongoing work we are interested whether and how the sending and receiving of signals is dependent on the phase of the cell cycle. The mathematical model which will help to plan the experimental part of the work will be presented. MECHANISMS OF ACTIVATION AND SILENCING OF THE IMMUNOGLOBULIN KAPPA LOCUS THROUGH DNA LOOPING AND NUCLEAR REPOSITIONING TO HETEROCHROMATIN Zhe Liu, Piotr Widlak, Ying Zou, Misook Oh, Mee Young Chang, Jerry W. Shay and William T. Garrard Departments of Molecular Biology and Cell Biology, UT Southwestern Medical Center, Dallas, Texas 75390, USA We have investigated the higher-order chromatin structure of the mouse immunoglobulin (Ig) kappa locus as a function of activation and silencing during B cell development using the techniques of chromosome conformation capture (3C) and 3D 3-color FISH. We find that in the fully transcriptionally active state that each of the three enhancers form complexes with themselves and rearranged gene promoters, while the resulting intervening DNA is looped out. These results fit the looping model for the mechanism of enhancer action. Using chromatin immunoprecipitation in combination with 3C (e.g., ChIP-3C), we demonstrate that the transcription factor E47 is present in such looped complexes. Two of these enhancers also form complexes with a 3' boundary sequence that has CTCF sites. In addition, the 5' region of the active transcription unit exhibits a continuum of interactions with downstream chromatin segments. All of these interactions are B cell specific (Liu and Garrard, Mol. Cell. Biol. 25, 3220 [2005]). Previous studies have shown that allelic exclusion of the mouse Igkappa locus occurs by the combination of monoallelic silencing and a low level of monoallelic activation for rearrangement combined with a negative feedback loop blocking additional functional rearrangements. Using yeast artificial chromosome-based single-copy isotransgenic mice, we have identified a cis-acting element that negatively regulates rearrangement in this locus, specifically in B cells. The element resides in the V-J intervening sequence, and is termed Sis (silencer in the intervening sequence). Sis specifies the targeting of germline Igkappa transgenes in B cells to centromeric heterochromatin and their association with Ikaros, a repressor protein that also co-localizes with centromeric heterochromatin. Significantly, these are hallmarks of silenced endogenous germline Igkappa genes in B cells. These results provide new insights into the molecular mechanisms of allelic exclusion. Research supported by NIH and the Robert A. Welch Foundation. GENOMIC INSTABILITY OF THE HEREDITARY PROSTATE CANCER 1 (HPC1) REGION IN SOMATIC BREAST CANCER Lise Lotte Hansen Danish Centre for Molecular Gerontology, Institute of Human Genetics, The Bartholin building, University of Aarhus, DK-8000 Aarhus C, Denmark e-mail: Lotte@humgen.au.dk In a prospective study with 10 years of follow-up, we have established a significant correlation between allelic imbalance (AI) affecting 16q23.2 -24.2.3 and favourable prognosis by univariate and multivariate analysis of 36 highly polymorphic microsatellite markers along chromosome 16q. No association was found to other clinical parameters such as menopausal status, tumour size, lymph node status, histopathology, and lymph node capsule invasion. This makes allelic loss of 16q23.2-24.2 an independent marker of favorable prognosis for primary breast cancer. A number of cytogenetic studies published over the past two decades have shown that a complex rearrangement resulting in the gain of 1q in combination with allelic deletion of 16q is an early, perhaps the primary, event in breast cancer development. In the search for a possible link between the gain of 1q, loss of 16q and prognosis for breast cancer patients, we concentrated on the HPC1 region at 1q25.3. Our initial hypothesis was that activation of ribonuclease L (RNase L) at 1q25.3 may lead tumour cells, with a certain level of genomic instability, into apoptosis. Mutations in the gene have recently been found to segregate in prostate cancer families, and homo- and heterozygosity for the mutation R462Q is a strong predictive marker for inherited prostate cancer. RNase L is part of the interferon-induced pathway against viral infection, it is involved in stress-induced apoptosis in metastatic prostate cancer cell lines and in termination of the protein synthesis. We found no significant correlation between AI or mutations in the gene RNASEL and any prognostic parameter for breast cancer. We are currently analyzing 300 breast tumours for the expression level of RNase L, since overexpression due to gain of 1q can be related to increased apoptosis, depending on the degree of genomic instability of the tumour cell. We focused on the region flanking RNASEL, the HPC1 region, and analyzed a 740 Kb spanning from RGSL2 to LAMC2 at chromosome 1q25.3 for allelic imbalance with new inter- and intragenic markers, already identified and characterized. We found a high rate of AI across the entire region and numerous chromosomal breakpoints. Breaks affecting three genes have been mapped in detail, and the resulting microdeletions of various sizes are positioned in the coding region of RGS8, RGS16, and RGSL2. This leads to the hypothesis that genomic instability affecting one or more genes at chromosome 1q25.3 is blocking metastasis formation in breast cancer patients. Additional genomic lesions may promote metastases formation. LOGICAL ANALYSIS OF PROTEOMIC DATA Krzysztof Puszynski Silesian University of Technology, Gliwice An parallel implementation of proteomic ovarian cancer diagnosis system based on logical analysis of data is shown. The implementation is based on computational cluster elaborated in System Engineering Group at Silesian University of Technology. For verification of algorithm and software Ovarian Dataset 8-7-02 was used. This mass spectrometry data contains intensity levels of 15 154 peptides defined by their mass/charge ratios (m/z) in serum of 162 ovarian cancer and 91 control cases. An OpenMosix with MPI (Message Passing Interface) cluster technology was used to construct in LAD a fully reproducible models (1) using full range and (2) using only 700-12000 of m/z values of peptides and proved in multiple cross-validation leave-one-out tests to guarantee sensitivities and specificities of up to 100 %. CHROMATIN REMODELING AND LINKER HISTONES IN PLANT DEVELOPMENT T. Sarnowski1, G. Ríos2,3, J. Jásik2, S. Świeżewski1, S. Kaczanowski1, A. Kwiatkowska4, K. Pawlikowska1, M. Koźbiał4, P. Koźbiał4, C. Koncz2, Marta Prymkaowska-Bosak4, A. Wierzbicki4 and A. Jerzmanowski1,4x 1Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Pawinskiego 5A, 02-106 Warsaw, Poland 2Max-Planck Institut für Züchtungsforschung, Carl-von-Linné-Weg 10, D-50829 Köln, Germany 3Present address: Departamento Bioquimica y Biologia Molecular, Universidad de Valencia, Dr Moliner 50, 46100 Burjassot, Spain 4Warsaw University, Laboratory of Plant Molecular Biology, 02-106 Warsaw, Poland xe-mail: andyj@ibb.waw.pl Eukaryotic genomes are organised into chromatin, a structure that enables packaging of very long DNA molecules into a microscopic-size nucleus, while ensuring efficient use of the information contained in the DNA sequence. In flowering plants and mammals the regulatory mechanisms at the level of chromatin show a high degree of evolutionary conservation. In mammals, the mutations in genes encoding proteins involved in post-translational modifications of core histones, methylation of DNA, ATP-dependent chromatin remodeling and in stabilizing heterochromatic state often result in embryo lethality or in serious developmental defects. Plants are more tolerant to such mutations and allow the analyses of their effects in successive generations. Chromatin remodeling requires specialized ATPases that utilize the energy of ATP hydrolysis to alter chromatin state. All such enzymes characterized so far appear to function within multi-subunit complexes, with the number of subunits ranging from two to more than ten. The prototype yeast SWI/SNF (SWItch/Sucrose-Non-Fermenting) complex is built around the founding member of the remodeling ATPase family, the SWI2/SNF2. All SWI/SNF-type complexes studied so far contain a minimal structural and functional core composed of four evolutionarily conserved subunits, homologues of yeast proteins SWI2/SNF2, SNF5, SWI3 and SWP73. We have recently identified the Arabidopsis lines with T-DNA insertions in each of the four genes encoding members of the AtSWI3 family. The analysis of these lines documents an astonishing functional diversity of plant SWI3 proteins. An important part of our research concerns identification of the possible molecular mechanisms underlying the phenotypic changes in Arabidopsis thaliana observed upon RNAi-mediated decrease of the chromatin content of linker histones. We have recently shown that plants with a >90% reduction in H1 expression exhibit pleiotropic phenotypic defects which segregate independently of the anti-H1 dsRNA transgene. The phenotypic defects are correlated with minor but statistically significant changes in the methylation patterns of repetitive and single-copy sequences APOPTIN - A NOVEL, CANCER-SELECTIVE KILLER Subbareddy Maddika, Evan P. Booy, Ted Paranjothy, and Marek Los Manitoba Institute of Cell Biology, CancerCare Manitoba, University of Manitoba, Winnipeg,MB R3E OV9, Canada Apoptin, a chicken anemia virus derived protein, selectively induces apoptosis in cancer cells but not in primary, non-transformed cells, thus making it a promising candidate as a novel anti-cancer therapeutic. The mechanism of apoptin induced apoptosis is largely unknown. Apoptin likely kills cancer cells by a novel mechanism that has not been previously explored by known anticancer drugs. In transformed cells apoptin is known to mainly localized in the nucleus and kills the cells via apoptosis. Nuclear localization of apoptin protein is important for its selective toxicity that is executed by the activation of the intrinsic (mitochondria - dependent) apoptotic pathway. However the role of nuclear localization on apoptin activity and its subsequent signaling to the mitochondria is not known. My presentation will focus on the signaling molecules and pathways, that are involved in apoptin's signaling from the nucleus to mitochondria. It has been recently shown that some signaling molecules translocate from nucleus to mitochondria upon various apoptotic stimuli. Nur77/TR3/NGFI-B, an immediate early response gene and an orphan member of the steroid/thyroid receptor superfamily, translocates from nucleus to the cytoplasm, and targets mitochondria to induce cytochrome c release and apoptosis in response to apoptotic stimuli. By using different methodologies like siRNA technology, co-imunorprecipitation, confocal microscopy studies, apoptotic assays etc, we are exposing the signaling pathway employed by apoptin to kill cancer cells. CLINICAL SIGNIFICANCE OF OXIDATIVE DNA DAMAGE Ryszard Olinski Department of Clinical Biochemistry, Nicolaus Copernicus University, Collegium Medicum in Bydgoszcz, Karlowicza 24, 85-092 Poland Oxidative DNA damage is an inevitable consequence of cellular metabolism, with a propensity for increased levels following toxic insult. Whilst over twenty base lesions have been identified, only a fraction of these have received any appreciable study, most notably 8-hydroxy-2'deoxyguanosine. This lesion has been the focus of intense research interest, and been ascribed much importance, largely to the detriment of other lesions. The present work reviews the basis for the biological significance of oxidative DNA damage, drawing attention to the multiplicity of proteins with repair activities, along with a number of poorly considered effects of damage. Given the plethora of, often contradictory, literature reports describing pathological conditions in which levels of oxidative DNA damage have been measured, this review critically addresses the extent to which the in vitro significance of such damage has relevance for the pathogenesis of disease. It is suggested that a number of short-comings associated with biomarkers, along with gaps in our knowledge may be responsible for the failure to produce consistent and definitive results when applied to understanding the role of DNA damage in disease, highlighting the need for further studies. GENE EXPRESSION SIGNATURES OF NORMAL SQUAMOUS EPITHELIUM CAN DISTINGUISH AMONG GASTROESOPHAGEAL REFLUX DISEASE VARIANTS Jerzy Ostrowski, Tymon Rubel, Lucjan S. Wyrwicz, Michal Mikula, Andrzej Bielasik, Eugeniusz Butruk, Jaroslaw Regula Department of Gastroenterology, Medical Center for Postgraduate Education and Maria Skłodowska-Curie Memorial Cancer Center and Institute of Oncology, Warsaw Background and Aims: It has been proposed recently that gastroesophageal reflux disease patients may be categorized into three distinct groups: nonerosive reflux disease, erosive reflux disease, and Barrett's esophagus. Thus, different molecular mechanisms may underlie the development of each clinical variant. Among other predisposing factors, sex influences clinical outcomes of gastroesophageal reflux. The aim of this study was to identify distinct molecular subclasses in different variants of gastroesophageal disease. Methods: The measurements were done using the oligonucleotide DNA Affymetrix U133A 2.0 microarray and RNA isolated from mucosal samples of normal squamous esophageal epithelium in 29, 32, and 29 patients with nonerosive reflux disease, erosive reflux disease, and Barrett's esophagus, respectively. Results: Gene expression patterns successfully distinguished mucosal samples from Barrett's esophagus and nonerosive reflux disease patients, but not from erosive reflux disease. In addition, females developing nonerosive, but not erosive, reflux disease showed molecular changes that differed from those of males with Barrett's esophagus. Thus, both clinically and molecularly, erosive is clearly localized between nonerosive reflux disease and Barrett's esophagus. Conclusions: For the first time we have demonstrated different molecular makeup among gastroesophageal reflux disease patients. Our study suggests that this disease is a monophyletic disease developing on the basis of gastroesophageal reflux and that its clinical form is a result of process within epithelium modulated by sexual dimorphism. Differentially expressed genes in esophageal normal epithelium may depend on genetic predispositions but also may reflect different molecular responses of esophageal epithelium to damaging components of gastroesophageal reflux. SOME ALGORITHMS OF MOLECULAR DOCKING Marcin Pacholczyk Silesian University of Technology Institute of Automatic Control, Gliwice, Poland The study of geometric aspects of molecular docking is performed. Molecular docking can be regarded as a potential method for computer aided drug design and optimization. A solution based on LUDI interaction surfaces and computer vision algorithm pose clustering adopted to the specific needs of molecular docking is shown. As an illustration a reconstruction of the native pose of SO4 ligand in trypanasomal isomerase complex 5TIM is presented. The 3D structure of the complex was retrieved from ProteinDataBank (PDB) database. As a measure of binding affinity RMS deviation from the reference pose was used. ENGINEERING OF MEGANUCLEASES Alfred Pingoud Institute of Biochemistry, Justus-Liebig-University, Giessen, Germany Homologous gene targeting is potentially the ultimate tool for gene targeting /gene therapy, but its use is limited by low efficiency and reproducibility. Site-specific DNA double-strand breaks can induce efficient and selective gene targeting. Engineered, highly specific DNA endonucleases, which are programmable according to the desired specificity are the key to a wider use of this technology in gene therapy. We are in the process of developing programmable "meganucleases" designed to cleave only positions of interest in a genomic context with the aim - based on homologous recombination - to solve the main problem in reverse genetics and gene therapy: to replace genes, efficiently, selectively and reproducibly. Two types of meganucleases are being engineered by us: (i) programmed restriction endonucleases, and (ii) redesigned homing endonucleases. The first approach depends on the high specificity of restriction enzymes and of triple helix formation; the target site of a restriction enzyme - TFO conjugate is a composite of the recognition site of the restriction enzyme and the triple helix forming site. We have fused a triple helix forming oligonucleotide to single chain PvuII (Simonscits et al. 2001) using a bifunctionasl crosslinker. This scPvuII-TFO heteroconjugate cleaves only the composite site, when preincubated with DNA in the absence of Mg-ions to allow triple helix formation, prior to addition of Mg-ions to initiate cleavage. With such programmable restriction enzymes any gene can in principle be targeted given the combinatorial flexibility this fusion offers in addressing a short, yet precisely recognized restriction site next to a defined triple-helix forming site. A drawback of this approach is that triple-helix formation is slow, which means that after delivery to the cell, the restriction enzyme part of the conjugate would cleave DNA at multiple sites before triple-helix formation has targeted the enzyme to the site of interest. A solution to this problem developed by us is to use a "caged" version of the enzyme that requires photoactivation to become active. For the second approach of meganuclease generation, homing endonucleases are particularly suitable. Homing endonucleases are a large class of proteins (several hundreds of members) found in bacteria, archaea, fungi and algae. They are encoded by mobile genetic elements such as group I introns and inteins, which promote their proper dissemination through their endonuclease activity. They recognize DNA sequences of up to 35 base pairs in length and cleave large genomes only at few positions. Different from restriction endonucleases, they are tolerant to base substitutions. Homing endonucleases of the LAGLIDADG family have a modular architecture, which in principle allows generating homing endonucleases of new (desired) specificity by domain swapping approaches (Chevalier et al. 2002; Epinat et al. 2003). We have engineered a version of I-DmoI, composed of two copies of domain A of the monomeric wildtype I-DmoI. This homodimeric A2 variant recognizes and cleaves a palindromic sequence composed of two inverted repeats of the half-site 'A' of the I-DmoI recognition sequence; it does not cleave the recognition sequence of wild type I-DmoI ('A-B'). Given the large number of known LAGLIDADG enzymes it should be possible to generate a huge number of variants by domain swapping, which should allow addressing an even larger number of recognition sequences, considering the sequence tolerance of homing endonucleases. ASSEMBLY OF NUCLEAR MATRIX - BOUND PROTEIN COMPLEXES INVOLVED IN NON-HOMOLOGOUS END JOINING IS INDUCED BY INHIBITION OF DNA TOPOISOMERASE II. Omar L. Kantidze, Olga V. Iarovaia and Sergey V. Razin Laboratory of Structural and Functional Organization of Chromosomes, Institute of Gene Biology RAS, Vavilov Street 34/5, 119334, Moscow, Russia Topoisomerases maintain the DNA structure by relieving the torsional stress and alleviating other topological problems occurring in DNA during transcription and replication. Topoisomerase II appears to have a close association with the family of proteins involved in the organization of chromatin in a series of loops on the proteinaceous chromosomal matrix. Beyond its physiological functions, topoisomerase II is the target for some of the most active anticancer drugs. Inhibition of the topoisomerase II function can result in DNA double-strand breaks (DSBs) and, thus, lead to chromosomal translocations. The earliest event during DSB repair is phosphorylation of histone H2AX at S139 (so-called XH2AX) which is believed to serve as a focal point for the assembly of repair proteins at the DSB. In this work, we have demonstrated the formation of XH2AX foci in two human cell lines - K562 and HeLa - after suppression of topoisomerase II activity with etoposide. Furthermore, these foci remained visible at nuclear matrices and colocalized with the major components of non-homologous end joining system of DSBs repair. Thus, inhibition of topoisomerase II activity triggers assembly of NHEJ complexes at the nuclear matrix. SATELLITE DNA, MEIOTIC TELOMERES AND CHROMOSOME EVOLUTION Harry Scherthan1,2 1Institute of Radiobiology (Bundeswehr), Neuherberg Str.11, D-80937 Munich, Germany 2MPI for Molecular Genetics, D-14195 Berlin e-mail: scherth@web.de Exposure to ionizing radiation and genotoxic compounds leads to formation DNA double strand breaks (DSBs) that may trigger the formation of chromosome rearrangements, cancer or cell death. DSBs occur also in physiological context, e.g. during meiosis. Here DSBs are instrumental for generating crossing over, chromosome reduction, and genetic diversity in the offspring. During evolution, viable chromosome rearrangements propagated through the germ line have constantly reshaped karyotypes leading to the diverse karyomorphs encountered in extant species. Extreme karyotype variation and a particularly rapid karyotypic evolution characterize the genus Muntiacus comprising small Asian deer. Diploid chromosome numbers range from 2n=46 in the Chinese muntjac (M. reevesi) to as low as 2n=6/7 (female/male) in the Indian muntjac (M. muntjak vag.), with the latter being the lowest known chromosome number among extant mammals. The drastic chromosome number reduction in the Indian muntjac has involved numerous tandem and a few Robertsonian fusions. Molecular and FISH analysis suggests that repetitive DNA sequences found at centromeres and telomeres might have played a role in this fusion process. Genes involved in telomere metabolism appear to be functional in muntjac cells. Based on molecular and FISH analyses it is proposed that recombinogenic DSBs between telomere and muntjac satellite sequences during the bouquet stage of meiotic prophase, where chromosome ends cluster in a limited region of the nuclear periphery, may have sustained karyotypic orthoselection for tandem fusions that shaped the extant Indian muntjac karyotype. MOLECULAR MARKERS OF TUMOR PROGRESSION AND PROGNOSIS ESTABLISHED IN HEAD AND NECK CANCER K.Szyfter1,2, K.Szukała1, M.Jarmuż1, M.Giefing1, M.Rydzanicz1, M.Wierzbicka2, W.Szyfter2 1Institute of Human Genetics, Polish Academy of Sciences, Poznan, Poland 2Dept. of Otolaryngology, K.Marcinkowski Medical University, Poznan, Poland An efficiency of head and neck cancer treatment remains low. An occurrence of multiple primary tumors (MPT), local metastasis and tumor relapse are the main causes of treatment failure. An identification of genetic markers of all the listed symptoms is a task for molecular epidemiology and experimental oncology studies. The research efforts were focused on identification of a specific genotype predisposing MPT formation, differentiation between MPT and metastasis, discovering a marker of entering into micrometastasis, identification of genetic background of tumor tendency to relapse. The main results are as follows: 1. Cytogenetic (classical, FISH, CGH) findings indicate for the regions of oncogenes amplification and tumor suppressor genes deletion to be studied further for a loss of heterozygosity. 2. An amplification of 11q13 (CCND1 locus) and translocations involving 11q13 region determined together with 3q gain is a reliable marker of poor prognosis. Interpretation of CCND1 expression should take into account the cyclin D1 polymorphism. 3. Comparative genomic hybridization (CGH) and loss of heterozygosity (LOH) appear to be a proper attempt to determine cancer cells clonality and to differentiate between metastasis and MPT. 4. LOH analysis in tumor and tumor free-areas of larynx confronted with laryngeal cancer progression requires a thorough selection of microsatellite markes to get reach a predictive significance of tumor relapse. 5. Genotyping of the selected genes coding carcinogen metabolizing enzymes and DNA repair enzymes did not show a distinct genotype predisposing to MPT.