Báo cáo khoa học:Invited Lectures docx

Invited Lectures A1-L5 Mitotic chromosome condensation and segration C. D’Ambrosio 1 , Y. Katou 2 , K. Shirahige 2 , F. Uhlmann 1 1 Chromosome Segregation Laboratory, London, UNITED KINGDOM, 2 Division of Gene Research, Tokyo Institute of Technology, Yoko- hama, JAPAN. Mitotic chromosome structure depends on the chromosomal condensin complex. Without condensin, metaphase chromosomes remain undercondensed and lack structural stability. We have asked where along budding yeast chromosomes the condensin complex associates, and what we can learn from its binding pattern about the mechanism of chromosome condensation. Our results suggest that condensin, like its relative the cohesin complex, is loaded onto chromosomes by a loading factor, the Scc2/4 complex. Unlike cohesin, that moves away from its loading sites after the loading reaction, condensin remains at the loading sites. We discuss the implications of these pat- terns for interphase and mitotic chromosome structure. Furthermore, condensin is required during anaphase to promote sister chromatid resolution. In the absence of condensin, anaphase brid- ges and segregation defects are observed. How condensin promotes sister chromatid resolution is unknown. We have used the budding yeast rDNA as a model locus, whose segregation depends on condensin during anaphase. We show that anaphase bridges in a condensin mutant are resolved by ectopic expression of a foreign (Chlorella virus) but not endogenous yeast topoisomerase II (topo II). This sug- gests that catenation prevents sister rDNA segregation, and that yeast topo II is ineffective in decatenating the rDNA, and maybe other chromosomal regions, in the absence of condensin. B5-L1 The Meningococcal Transformation Machine T. Tønjum Centre for Molecular Biology and Neuroscience, University of Oslo, Oslo, NORWAY Neisseria meningitidis is a leading cause of bacterial meningitis and septicaemia worldwide. This bacterium is constitutively competent for transformation throughout its entire life-cycle. Transformation in neisserial species is particularly important for genetic exchange and diversity and is coupled to the expression of type IV pili. Meningococcal type IV pili are present on the cell surface as bundled filamen- tous appendages and are assembled, extruded and retracted by pilus biogenesis components. These proteins are homologous to type II secretion components in Gram-negative species. The binding and uptake of transforming DNA into the meningococcal cell can be divided into four stages: entry through an outer mem- brane pore, transit of the periplasm, transport across the inner membrane and genome integration. We propose that the early stage of meningococcal transformation is coupled to pilus retraction and that transforming DNA is non-specifically attached to retracting pili. Previously, we have shown that pili directly interact with the secretin PilQ in the outer membrane. We have evidence that DNA is intro- duced into the cell through the positively charged pore formed by the PilQ complex and that other DNA binding components process DNA inside the periplasm and inner membrane. To address the multi-step nature of DNA binding and uptake during transformation, we have identified DNA binding components in meningococcal cellular fractions. By using a combination of molecular and imaging approaches, DNA binding candidates are being assessed for their structure-function relationships to define how they act and interact, with each other and with DNA. The goal is to define how these DNA binding components provide dynamic multi-site targeting and entry of DNA. Invited Lectures Late Abstracts 341 E1-L2 Global Mapping of the Yeast Genetic Interaction Network C. Boone, III University of Toronto, Toronto, ON, CANADA. Synthetic Genetic Array (SGA) analysis automates yeast genetics, enabling a number of different large-scale/systematic studies. In partic- ular, we are attempting to generate the complete synthetic lethal genetic interaction map for yeast cells. This map can be used to define complexes and pathways in the cell, but perhaps more importantly, it adds functional information to the protein-protein interaction map, identifying complexes and pathways that buffer one another and somehow work together as backup systems. One of our major challenges has been to generate a quantitative model for scoring genetic interactions based upon plate images and the predicted fitness of the double mutant relative to each single mutant. Another challenge has been to develop robotic platforms for our own high-through- put analysis and individual users in other labs. I will present the results of our latest network compilation and describe our plans for tackling the yeast genome. In addition, I will describe how the yeast genetic interaction network can be used to interpret chemical-genetic interactions and link bioactive compounds to their target pathways. E2-L2 Multi-factorial disease and robustness: Where Systems Biology makes a difference H. V. Westerhoff Manchester Center for Integrative Systems Biology and Netherlands Institute for Systems Biology, Manchester, UNITED KINGDOM. As we now know what Systems Biology is, it may be worthwhile to examine the difference it will make. In particular it may be useful to examine where such a difference is needed and how it may be brought about by those who are interested. There are a number of diseases that can be related to a single faulty gene product and or a single invading microorganism. Quite a few of these ‘monofactorial’ diseases can be cured, or structural molecular biology has defined a strategy to identify or design the corres- ponding drugs. However, there is a more substantial number of diseases for which a vast amount of biomedical research has been moun- ted and that cannot be cured. The research has led to substantial scientific success and biochemical understanding, but to little if any progress in defining cures. The examples are on the long list of ‘multifactorial diseases’ and include type-2 diabetes, obesity, heart dis- ease, cancer and arthritis. The approach that has been successful for monofactorial diseases is not optimal vis-a ` -vis these diseases. Finding cures for these diseases may require a shift in strategy. These diseases are ‘network’ or ‘Systems Biology diseases’ and to manage them one needs to address faults in the network rather than in just a molecule. I shall discuss some of the ways the Manchester-Amsterdam Systems Biology axis tries to devise such network-based strategies. These include differential network based drug design against Trypanosoma brucei, modular kinetics of a connection between type-2 diabetes and obesity, and new paradigms for searching new anti-tumor drugs. Fragility and robustness are Leitmotive, but differentially so. Late Abstracts Invited Lectures 342 E5-L5 Insight into alternative-splicing mechanisms with the solution structures of several RRM-RNA complexes F. Allain Institute of Molecular Biology and Biophysics, Zu ¨ rich, SWITZERLAND. Alternative-splicing is probably the most effective post-transcriptional gene regulatory event, as more than 60% of the human genes are alternatively spliced. Defects in alternative-splicing are the cause of many genetic diseases. We recently determined the structure of sev- eral important human alternative-splicing factors in complex with RNA, namely the Poly-pyrimidine Tract Binding Protein (PTB1), Fox-12, SRp203 and of a potential splicing factor RBMY4. The structures of PTB and Fox-1 showed that these alternative-splicing fac- tors bind RNA sequence-specifically and might control the fate of an alternative-exon in remodeling the RNA. The structures of SRp20 and of RBMY in complex with RNA revealed unusual RNA recognition modes. SRp20 binds RNA in a semi-sequence specific manner 3 and the human RBMY binds a stem-loop RNA whereas its mouse homolog doesn’t 4 . Implications of these structural findings for under- standing the mechanism of action of these alternative-splicing factors will be discussed. References: 1. Oberstrass, F.C. et al. Structure of PTB bound to RNA: specific binding and implications for splicing regulation. Science 309, 2054–7 (2005). 2. Auweter, S.D. et al. Molecular basis of RNA recognition by the human alternative splicing factor Fox-1. Embo J 25, 163–73 (2006). 3. Hargous, Y. et al. Molecular basis of RNA recognition and TAP binding by the SR proteins SRp20 and 9G8. Embo J 25, 5126–37 (2006). 4. Skrisovska, L. et al. The testis-specific human protein RBMY recognizes RNA through a novel mode of interaction. EMBO Rep 8, 372–79 (2007). Invited Lectures Late Abstracts 343 Posters A1-13 Characterization of a subtelomeric satellite DNA in the mollusc Donax trunculus V. Petrovic 1 ,C.Pe ´ rez-Garcı ´ a 2 , J. J. Pasantes 2 , E. Prats 3 , M. Plohl 1 1 Rudjer Boskovic Institute, Zagreb, CROATIA, 2 University of Vigo, Vigo, SPAIN, 3 Institut de Biologia Molecular de Barcelona – CSIC, Barcelona, SPAIN. Satellite DNAs are non-coding, tandemly repeated DNA sequences that comprise long arrays in the genome and are usually located within heterochromatic regions of chromosomes. Among marine invertebrates, satellite DNAs have so far been studied in detail only in a couple of taxonomic groups. We detected a novel satellite DNA present in the genome of the bivalve mollusc Donax trunculus. The monomer repeat length of this satellite is 169 bp, while the sequence analysis reveals high sequence conservation maintained throughout the entire monomer length. In contrast to the other satellites detected previously in the genome of D. trunculus (Petrovic and Plohl, Gene 2005, 362:37, and references therein), this is a GC rich satellite that has also been shown to exhibit CpG site methylation. In addition, this is the most abundant (5 %) among all detected satellites. Fluorescence in situ hybridization revealed that this satellite is located in subtelomeric regions on more than half of D. trunculus chromosome pairs. Its location on a subset of chromosomes and homogeneity of randomly cloned variants indicate different homogenization mechanisms acting between chromosomes, and/or different evolutionary history of chromosomes in the D. trunculus genome. A1-14 The evolutionary role of the LCR16A element in Mendelian disorders O. Symmons, H. de Boussac, A. Va ´ radi, T. Aranyi Institute of Enzymology, Hungarian Academy of Sciences, Budapest, HUNGARY. With the completion of the human genome, a major challenge is to determine correlation between genome evolution and phenotypic changes. The best studied human phenotypes are Mendelian disor- ders. Two such examples are Polycystic Kidney Disease (PKD) and Pseudoxanthoma Elasticum (PXE) caused by PKD1 and ABCC6, respectively. Both genes were mapped to the short arm of Chr 16. However, their study is complicated by the presence of multiple pseudogenes of high degree of similarity and also located on Chr 16. Our aim was to characterize these pseudogenes and understand the mechanism and potential impact of their duplica- tion. We found that every pseudogene is situated in close proximity to an LCR16a element, which is a very proliferative duplicon on Chr 16. LCR16a was proposed to be the driving force for the for- mation of intrachromosomal duplication blocks. We analyzed the phylogenetic relationship of the human LCR16a elements, which form three distinct clusters. The duplicated copies of the ABCC6 and PKD1 genes are all associated with the same cluster. We dem- onstrated that some of the duplication events were hominin-specific and showed that this can lead to hominin-specific chimeric tran- scripts. We have also investigated the possible consequences of the presence of pseudogene sequences in the aetiology of PXE and PKD. Our results demonstrate that non-allelic homologous recom- bination has occurred and is manifested in these regions and gene conversion is still an on-going process. A2-27 The BAH domain of Sir3 is the primary nucleosome binding domain in the SIR silencing complex M. Onishi 1 , G. Liou 1,2 , D. Moazed 1 1 Harvard Medical School, Boston, MA, 2 Division of Molecular and Genomic Medicine, National Health Research Institutes, Miao-Li County, TAIWAN. In Saccharomyces cerevisiae, Silent information regulator (Sir) pro- teins are required for regional gene silencing at the silent mating type cassettes and telomeres. The initiation of transcriptional silen- cing at these domains is proposed to involve the recruitment of the Sir complex, composed of Sir2, Sir3, and Sir4, by DNA-binding proteins. This recruitment is followed by iterative cycles of NAD- dependent deacetylation, production of O-acetyl-ADP-ribose, and binding of Sir3 and Sir4 to the deacetylated nucleosomes, and leads to the creation of extended silent chromatin domains. These silenced regions have histones that are hypoacetylated and hypo- methylated, while disruption of enzymes that acetylate and methy- late histones leads to mislocalization of Sir3. Thus, histone modifications play a crucial role in the assembly of silent chroma- tin. However, while Sir3 has been shown to have histone and nu- cleosome binding properties in vitro, specific binding of Sir3 to nucleosomes, as it occurs in vivo, has yet to be observed. We show that the Bromo-adjacent-homology (BAH) domain of Sir3 is neces- sary for nucleosome binding and that this binding is regulated by histone acetylation and methylation. These results suggest that the BAH domain, found in many eukaryotic chromatin proteins, binds to specifically modified nucleosomes. A2-28 Post synthetic acetylation of HMGB1 protein stimulates DNA end-joining I. Ugrinova 1 , E. Mitkova 1 , C. Moskalenko 2 , I. Pashev 1 , E. A. Pasheva 1 1 Institut Molecular biology,Bulgarian.Academy of Sciences, Sofia, BULGARIA, 2 Laboratoire Joliot-Curie, Ecole Normale Superieure, Lyon, FRANCE. The ability of HMGB1 protein to recognize bent DNA and to induce bending in linear duplex DNA defines HMGB1 as an archi- tectural factor. We already demonstrated that the binding affinity of the protein to various bent DNA structures is enhanced upon in vivo acetylation at Lys 2. Here we investigate how this modifica- tion of HMGB1 affects its ability to bend DNA. We report that the modified protein cannot bend short DNA fragments but, instead, it stimulates joining of the same fragments via their ends. The same properties demonstrates in vivo acetylated HMGB1 lack- ing its acidic tail. Further in vitro acetylation with acetyltransferase CBP of the truncated protein at Lys81 (possible upon tail removal only), restores the protein’s bending ability, while the stimulation of DNA end-joining is strongly reduced. We conclude, therefore, that the ability of HMGB1 to bend DNA or to stimulate end-join- ing is modulated in vitro by acetylation. In an attempt to explain the properties of in vivo acetylated HMGB1, its complexes with DNA have been analysed by both protein-DNA cross linking and AFM (atomic force microscopy). Unlike the parental protein, bound mainly within the internal sequences, the acetylated HMGB1 binds preferentially to DNA ends. We propose that the loading of acetylated protein on DNA ends accounts for both the failure to bend DNA and the stimulation of DNA end-joining. Late Abstracts Posters 344 A3-11 Analysis of restriction fragments distribution in genomic fungal DNAs. V. N. Tomilov 1 , D. A. Gonchar 1 , M. A. Abdurashitov 1 , A. Schueffler 2 , H. Anke 2 , S. K. Degtyarev 1 1 SibEnzyme Ltd., Novosibirsk, RUSSIAN FEDERATION, 2 IBWF, Kaiserslautern, GERMANY. Theoretical diagrams of fungal chromosomal DNA (Magnaporthe grisea, Mycosphaerella graminicola and Schizosaccharomyces po- mbe) cleavage at more than 125 different 4 - 8 nucleotides sequences have been simulated based on recently determined DNA primary structures. All chosen DNA sequences are the recognition sites of restriction enzymes. In all diagrams obtained we have selec- ted the fragments presented in peak’s quantities. Each peak has been characterized by multiplication of the number of fragments by their length. In the analysis we have considered peaks with a value of 0.15% or higher compared to the length of the corres- ponding chromosomes. Calculated Schizosaccharomyces pombe DNA digestions did not contain such peaks. Theoretical cleavage of Magnaporthe grisea and Mycosphaerella graminicola DNAs at the appropriate recognition sites of restriction enzymes resulted in more than 10 such peaks. A more detailed study has shown that the presence of peaks in DNA digestions depends on genome com- plexity. Hydrolysis of the fungal DNA with several restriction enzymes has confirmed the presence of the predicted fragments. A3-12 Nitric oxide levels and eNOS gene polymorphism in patients with coronary artery disease documented by angiography G. Yilmaz 1 , I. Mehmetoglu 1 , S. Kurban 1 , H. Aacar 2 1 Meram Faculty of Medicine, Department of Biochemistry, Univer- sity of Selcuk, Konya, TURKEY, 2 Meram Faculty of Medicine, Department of Medical Genetics, University of Selcuk, Konya, TURKEY. The aim of this study was to investigate 4a/4b VNTR polymorph- ism in intron 4, the G10-T polymorphism in intron 23 and plasma NO levels in patients with coronary artery disease (CAD) docu- mented by angiorgaphy and healty controls. The study was performed on 106 patients (30 F, 76 M) aged 40–70 years and 89 healty controls (40 males, 49 females) aged 41–73 years. All patients had more than 50% stenosis in at least one cor- onary artery documented by angiography. Blood samples were obtained from all subjects after fasting. NO was measured as NOx by Griess reaction on EDTA samples and eNOS gene polymorphism was investigated by PCR-RFLP tech- nique. NO levels of the patients and the controls were 47,78 ± 27 lmol/ L and 41,44 ± 25,9 lmol/L respectively the difference of which was significant. There was no significant difference between allelic frequencies of eNOS gene intron 4 a/b VNTR and intron 23 polymorphisms of the groups. Therefore, our results show that eNOS gene intron 4 a/b VNTR and intron 23 polymorphisms are not independent risk faktors for CAD in a central area (Konya) of TURKEY. A4-11 The role of miRNAs in DNA replication and damage repair E. Zlotorynski, J. van Duijse, R. Agami The Netherlands Cancer Institute, Amsterdam, THE NETHER- LANDS. The miRNA genes encode short RNAs, which suppress the expres- sion of mRNAs bearing complementary target sequences. Each mi- RNA is predicted to target up to hundreds of genes, but the function of only a handful miRNAs has been determined, under- ling the need for a systematic screening approach to identify their roles in development and disease. Primary human cells grown with low levels of DNA replication inhibitors undergo irreversible growth arrest due to the accumulation of DNA damage. In order to identify miRNAs that can modulate DNA replication and DNA damage repair, we performed a screen for miRNAs that can affect replication inhibition-induced growth arrest. Human primary cells were transduced with a human miRNA expression library, and grown for 3 weeks with low levels of the replication inhibitors Aphidicolin, Hydroxyurea, Etoposide or Doxorubicin. The relative abundance of miRNA expressing cells was then measured by hybridization of the DNA of the different cell populations on a miRNA micro-array. We subsequently validated the screen results for each miRNA individually. Several miRNAs were found to override the replication inhibition-induced arrest, whereas others were found to enhance the effect of replication inhibition. We are currently studying the mechanism of action of these miRNAs, in order to understand their role in DNA replication and DNA dam- age repair. A4-12 Identification of novel regulators of microRNA function M. Kedde, R. Agami Netherlands Cancer Institute, Amsterdam, THE NETHERLANDS. The 3¢-untranslated regions (UTRs) of many mRNAs are subject to posttrancriptional regulation by microRNAs (miRNAs), which function as repressors of their target mRNAs. Little is known about the regulation of miRNA targeting and the reversibility of the process. Here I will present evidence that certain RNA binding proteins can repress miRNA activity both in vitro and in vivo. Posters Late Abstracts 345 A4-13 Role of miRNAs in replicative senescence of endothelial cells K. Fortschegger, M. Wieser, H. Katinger, R. Voglauer, J. Grillari University of Natural Resources and Applied Life Sciences, Vienna, AUSTRIA. In vitro replicative senescence of human umbilical vein endothelial cells (HUVECs) is widely used as a model system for aging. Indeed, senescent cells can be detected in atherosclerotic plaques and might contribute to aging per se and the development of car- diovascular disease. We aim to elucidate to what extent the expres- sion profile of micro-RNAs (miRNAs) is changed throughout in vitro senescence and whether these changes are responsible for observed differences in protein expression patterns. Early passage HUVECs of five donors are compared with their senescent counterparts regarding miRNA expression profiles. LNA-microarray analysis reveals consistent differential expression of at least 12 miRNAs, most of them upregulated in senescent cells. Differential regulation is confirmed by Northern blot analysis or quantitative realtime PCR. Selected miRNAs will be investi- gated regarding their potential targets and their influence on cul- tured cells when overexpressed or blocked. Furthermore cloning and sequencing of the miRNA-pool will help to identify HUVEC- specific small regulatory RNA species. Some of the miRNAs which we found upregulated in senescent cells have growth factor receptor mRNAs as their putative targets. The suppressed expression of such proteins would perfectly agree with the observed growth arrest of senescent cells. Future experi- ments will predominantly address this topic. Acknowledgement: Supported by the GEN-AU II Pilot Program A5-44 Efficient initiation of k DNA replication requires transcription and interaction between RNA polymerase and the kO replication protein A. Szambowska, G. Wegrzyn, M. Glinkowska University of Gdansk, Gdansk, POLAND. Initiation of k DNA replication in vivo and in crude in vitro system is strongly dependent on transcription initiated at the k p R promo- ter. This transcription event is indispensable for transcriptional activation of orik and expression of the kO and kP genes. Forma- tion of the k replication initiation complex requires both kO and kP initiators and many E. coli proteins; including DnaB, primase, SSB, DNA polymerase III holoenzyme as well as DnaK, DnaJ and GrpE chaperones. In this study, we investigated in more detail the role of transcrip- tion and RNA polymerase in initiation of replication at orik. Using gel mobility shift assay, we detected stimulation of kO bind- ing to iterones and its self-assembly into a nucleosome-like struc- ture by RNA polymerase and transcription. In addition, we observed enhanced dimmer formation by kO protein in the pres- ence of both RNA polymerase and transcription. Stimulation of O-some formation by trancription was also confirmed by gel filtra- tion. These experiments revealed, however, that transcription caused no significant disassembly of nucleoprotein structure formed by kO. Glutaraldehyde - crosslinking experiments revealed also a possible direct interaction between kO protein and b -sub- unit of the E. coli RNA polymerase. We obtained additional evidence supporting this hypotesis using in vitro replication experi- ments, in which a plasmid containing phi10 promoter of bacterio- phage T7, instead of the k p R promoter, was used as a template. Interestingly, such a plasmid did not replicate in vitro in the presence of T7 RNA polymerase. Using in vitro permanganate footprinting method we detected also that RNA polymerase and trancription process enhance orik unwinding. A5-45 External cell control PCR: a new strategy for qPCR normalization A. Bors 1 , P. Ribiczey 1 ,G.Ko ¨ blo ¨ s 2 , Z. Ujfaludi 3 ,A.Va ´ radi 2 , T. Kova ´ cs 1 , T. Aranyi 2 1 National Medical Centre, Institute of Haematology and Immuno- logy, Budapest, HUNGARY, 2 Institute of Enzymology, Hungarian Academy of Sciences, Budapest, HUNGARY, 3 University of Szeged, Faculty of Sciences, Dept of Biochemistry and Molecular Biology, Szeged, HUNGARY. Quantitative RT-PCR (qRT-PCR) is a widely used method to determine relative gene expression levels. Quantification of the observed expression levels becomes reliable after normalization to the expression of an internal standard gene. However, the expres- sion of commonly used internal standard genes is often unstable, which may compromise quantification. To overcome the drawback of unstable internal standards, we developed a new method, called external cell control PCR (eccPCR). This method is based on the addition of control cells to the studied cells before RNA extraction and qRT-PCR. Only the control cells express the reference gene, while only the studied cells express the gene of interest. Here we present the validation of the method in various model systems including both adherent and non-adherent cells and either mam- malian or Drosophila external control cells. A5-46 Role of Arabidopsis SR protein atRSp31 in plant growth and development. M. Maronova 1 , M. Kalyna 1 , C. G. Simpson 2 , J. W. Brown 2 , A. Barta 1 1 Max F. Perutz Laboratories, Vienna, AUSTRIA, 2 Scottish Crop Research Institute, Dundee, UNITED KINGDOM. SR proteins are a family of evolutionary conserved splicing factors that play crucial role both in constitutive and alternative splicing and also in post-splicing processes. The genome of Arabidopsis thaliana encodes eighteen SR proteins and their exact function is unclear. Our aim was to characterize the function of plant specific SR pro- tein atRSp31 in plant growth and development using plants over- expressing atRSp31 and atRSp31 mutant plants. A phenotypic study revealed that increased levels of atRSp31 negatively affected plants upon oxidative stress, salt, abscisic acid and sugar treat- ments. Loss of function mutation showed an improved response to these stress conditions. Interestingly, plants overexpressing atRSp31 showed late flowering phenotype and early senescence. Corroborating these data, microarray analysis of overall expression change in atRSp31 overexpressing plants pointed out genes involved in response to diverse biotic and abiotic stimuli, genes participating in ROS metabolism, carbohydrate metabolism, aging and senescence. Moreover, several other SR proteins appeared to be affected. Further, to monitor the influence of atRSp31 on the known alter- native splicing events, we used a high throughput RT-PCR screen. The most prominent changes were observed in the splicing patterns of the SR protein atSRp30, the floral repressor MAF2 (AGL31), the two DNA repair proteins RAD1 and RAD23, and several unknown proteins. Our results suggest that atRSp31 is a negative regulator of stress response. It participates in response to wounding, oxidative stress, abscisic acid and salt stress. It is involved in flowering control and aging process. Late Abstracts Posters 346 A5-47 Properties of the Escherichia coli DnaA protein required for transcriptional activation of bacteriophage k p R promoter M. Glinkowska 1 , A. Szambowska 1 , M. S. Thomas 2 ,G.We˛grzyn 1 1 University of Gdan ´ sk, Gdansk, POLAND, 2 University of Sheffield, Sheffield, UNITED KINGDOM. DnaA protein is a bacterial replication initiator. It recognizes oriC and governs subsequent assembly of the replication complex. DnaA was also demonstrated to act as a transcription factor. Apart from influencing mRNA synthesis of several bacterial genes, it also regulates transcription starting from phage lambda p R pro- moter, through a mechanism that requires DnaA binding to two non-canonical DnaA-boxes situated downstream of transcription start site. DnaA belongs to the AAA+ superfamily of ATP-ases. For its function as a replication initiator, it requires formation of higher- order nucleoprotein complexes and self-oligomerization. Domain I and III of DnaA contain specific sites necessary for interaction of DnaA monomers. In this study we investigated the role of ATP binding by DnaA and domain I-mediated oligomerization in the stimulation of tran- scription starting from the p R promoter. We present an evidence that DnaA devoid of its first domain is both able to bind to the p R region and stimulate activity of this promoter in vitro. We also demonstrate that DnaA mutants, incapable of binding nucleotides or hydrolyzing ATP, are still active in transcriptional activation of the p R promoter both in vivo. The role of distinct DnaA domains in stimulating transcription from p R will be discussed and a mechanism for transcription acti- vation by DnaA will be proposed. A5-48 NMR investigation on 15.5K associated complexes P. Li, D. Raghunathan, S. Liu, S. Nottrott, O. Dybkov, R. Lu ¨ hrmann, M. C. Wahl, T. Carlomagno Max-Planck-Institute for Biophysical Chemistry, Goettingen, GERMANY. 15.5K is one of the U4 specific proteins in the spliceosome. Its binding to the 5’ stem-loop of U4 snRNA (U4 5’-SL) nucleates bindings of other U4/U6 specific proteins, namely hPrp31, which contains Nop domain, and 20K/60K/90K protein complex. The length of stem II of U4 5’-SL is found to be crucial for the binding of hPrp31. Binding of 15.5K is also require prior to the association of other Nop domain containing core proteins of Box C/D snoR- NAs in the nucleoli (Lukowiak, 2000; Watkins, 2002). We use NMR spectroscopy methods to study the specificity required in these binding events. Unlabeled full-length hPrp31 was titrated to [15N,2D,1HN]15.5K bound to unlabeled U4 5’-SL resulting a ca. 80KDa tri-complex. Saturation transfer experiment on the tri-com- plex revealed direct protein-protein contacts on the a2 helix and part of the a3 helix of 15.5K. hPrp31(aa.78–333) construct yield the same binding surface, suggesting that Nop domain is sufficient for binding to the primary RNP and the C-terminus of hPrp31 is not involve in protein-protein interaction. From the binding sur- face, we proposed models of the tri-complex with HADDOCK program. The crystal structure of the hPrp31(aa.78–333)-15.5K-U4 5’-SL complex and the HADDOCK model is consistent with the each other. We have shown that the Nop domain is a genuine RNP binding module exhibiting RNA and protein binding surfa- ces. The discrimination between the RNAs rises from structural rather than sequence differences. From the structure as well as our model, it is seen that the elongation of stem II results steric hindrance, which abolishes hPrp31 binding. Currently the com- plexes of 15.5K with U3 and U14 Box C/D snoRNAs are also been studied. A5-49 Cellular localization of the proto-oncoprotein TAF15 M. Marko, A. Vlassis, A. Guialis, M. Leichter National Hellenic Research Foundation, Athens, GREECE. TAF15 is the least studied member of the TET protein family (TLS/FUS, EWS and TAF15). Chromosomal translocations fusing TET genes with genes of specific transcription factors can lead to the expression of oncogenic fusion proteins. While the oncogenic chimeric proteins are extensively studied, much less is known about their normal counterparts. To under- stand how TAF15 is regulated on the spatial level we performed an analysis of the TAF15 domain structure with respect to nuclear and subnuclear localization. Nuclear localization signals were iden- tified by expression of TAF15 and TAF15 deletion mutants fused to EGFP. The experiments revealed that the very C-terminus of TAF15, as for EWS, functions as a nuclear retention signal. The N-terminus of the protein localizes to the cytoplasm and nucleus. When transcription is inhibited TAF15 localizes to nucleolar caps. The N- and C-terminal parts share, independently of each other, this property with the full length protein. Biochemical and in situ fractionation experiments show that a sub- stantial fraction of TAF15 is associated with the nuclear matrix, a property which can be mainly ascribed to the TAF15 N-terminus. Immunoprecipitation assays identified two scaffold attachment fac- tors SAF-A/hnRNPU and SAF-B as binding partners. Both SAF- B and SAF-A can associate directly with the TAF15 N-terminus in vitro. Together our findings indicate that nuclear and subnuclear localization of TAF15 is mediated by multiple domains of the pro- tein and by different binding partners. A5-50 The specific transcription factor TAF15 can associate with a subset of the spliceosomal U1 snRNP complex M. Leichter 1 , M. Marko 1 , M. Patrinou-Georgoula 1 , L. Tora 2 , A. Guialis 1 1 National Hellenic Research Foundation, Athens, GREECE, 2 IGBMC/CNRS, Strasbourg, FRANCE. Nuclear proteins able to bind RNA/ssDNA comprise an emerging class of multi-functional proteins with an anticipated role in coup- ling transcriptional and post-transcriptional events. This is the case of the highly homologous specific transcription factors of the TET family: TLS/FUS, EWS and TAF15 (TAFII68). We have been interested in ascribing structural/functional properties to the least studied member, TAF15. Anti-TAF15 immunoprecipitation assays (IPs) on HeLa nuclear fractions, prepared at higher than physiolo- gical salt-conditions, revealed the association of TAF15 with a minor fraction of the spliceosomal U1 snRNP, as shown by the detection in the immune pellet of both U1 snRNA and U1-70k and Sm proteins. Moreover, the presence of TAF15 in IPs of anti- U1-70k (as well as anti-U1 RNP autoAbs) was also evident. Pull- down assays with recombinant TAF15 and U1 snRNP-specific proteins were applied to identify components directly involved in this association. This revealed interactions between the N-terminal domain of TAF15 and U1C (and to a lesser degree U1A) proteins, while U1snRNA was not able to directly interact with TAF15 in vitro. The ability of TAF15 to directly contact RNA Pol II tran- scripts of a high turn over rate was shown in in vivo UV cross- linking studies. In addition the association of TAF15 with pre mRNA was demonstrated by in vitro splicing assays. All these findings suggest the existence of a functionally discrete U1 snRNP subset and provide support to the current view on an extensive involvement of U1 snRNP components in the early steps of coordinated gene expression and the involvement of TAF15 in coupling these processes. Posters Late Abstracts 347 A5-51 Investigation of the structure and zinc-binding properties of the RING finger domain from the human splicing-associated pro tein RBBP6 M. A. Kappo 1 , T. Mulaudzi 1 , R. A. Atkinson 2 ,E.AB 3 , R. Boelens 3 , D. J. G. Rees 1 , D. J. R. Pugh 1 1 University of the Western Cape, Bellville, SOUTH AFRICA, 2 IGBMC, Strasbourg, FRANCE, 3 University of Utrecht, Utrecht, THE NETHERLANDS. RBBP6 is a 250 kDa human protein known to interact with both p53 and pRb tumour-suppressor proteins as well as being involved in mRNA processing and transcriptional regulation. RBBP6 con- tains a RING finger domain and an N-terminal ubiquitin-like domain, suggesting a role in ubiquitylation. Recently RBBP6, in particular the RING finger domain, was shown to be required for the ubiquitylation of p53 by Hdm2, the human homologue of Mdm2, with knock-down of RBBP6 leading to accumulation of p53, concomitant wide-spread apoptosis and embryonic lethality, prompting speculation that RBBP6 may function as a scaffold for the interaction of p53 with Hdm2. As a first step towards investigating the interaction of RBBP6 with p53 and Hdm2, we have expressed the RING finger domain from human RBBP6 as a GST-fusion and determined its solution struc- ture using NMR. In addition to the conserved Cysteines typical of RING fingers, the domain also shares the conserved hydrophobic residues characteristic of U-box proteins, which are able to fold in the absence of bound zinc, raising the question of whether the domain requires zinc in order to fold. Using NMR we have shown that zinc is required for the domain to fold. Addition of Cd-EDTA leads to changes in the 15N-HSQC spectrum of the domain consis- tent with the replacement of zinc ions by cadmium ions. Further refinement is being undertaken in order to determine the identities of the residues involved in coordination of the zinc ions. Attempts to identify the coordinating residues directly using 113Cd-1H HSQC spectra are also ongoing. A5-52 NMR and SAXS studies of the ribonuclease RegB / ribosomal protein S1 system. P. Aliprandi 1 , P. Salah 1 , P. Giraud 1 , F. Mareuil 1 , J. Perez 2 , M. Uzan 3 , C. Sizun 1 , F. Bontems 1 1 Institut de Chimie des Substances Naturelles, Gif-sur-Yvette, FRANCE, 2 Synchrotron SOLEIL, Gif-sur-Yvette, FRANCE, 3 Insti- tut Jacques Monod, CNRS et Universite ´ Paris VI et VII, Paris, FRANCE. S1 is the largest protein (61 kDa) of the Escherichia coli ribosome. It is strictly required to the correct recognition of the translation initiation codon by the 30S subunit on the messenger RNA when the Shine-Dalgarno region is degenerated. S1 is also used by sev- eral bacteriophages at their own benefit. In particular, it increases the reaction rate of the T4 endoribonuclease RegB, which inacti- vates some of the phage early mRNAs, when their translation is no more required, by cleaving them in the middle of their Shine- Dalgarno sequence. We are interested in analyzing and comparing S1 mode of actions in both functions. We studied by NMR the structure of RegB and demonstrated that it belongs to a family of ribonucleases acting on translated mRNA in the ribosome. We also showed that the same region of S1 is responsible for the recognition of the initiation codon and for RegB activation. We analyzed by NMR and SAXS the structural and functional properties of this region. We now have undertaken to study the ternary complex formed by RegB, S1 and some mRNA in solution and in the ribosome. A5-53 Cytoplasmic recycling of 60S pre-ribosomal factors depends on the AAA-protein Drg1 L. Kappel 1 , G. Zisser 1 , B. Pertschy 2 , M. Tengg 1 , E. Liebminger 1 , B. Nobis 1 ,G.Ho ¨ genauer 1 , H. Bergler 1 1 Karl-Franzens-Universita ¨ t Graz Institut fu ¨ r Molekulare Biowissenschaften, Graz, AUSTRIA, 2 Biochemie-Zentrum Heidelberg, Heidelberg, GERMANY. Allelic forms of DRG1/AFG2 confer resistance to the drug dia- zaborine, an inhibitor of ribosome biogenesis in yeast. We show here that the AAA-ATPase Drg1 is essential for 60S maturation and associates with 60S precursor particles in the cytoplasm. Functional inactivation of Drg1 leads to an increased cytoplas- mic localization of shuttling pre-60S maturation factors like Rlp24, Arx1 and Tif6. Surprisingly, Nog1, a nuclear pre-60S fac- tor, was also relocalized to the cytoplasm under these condi- tions, suggesting that it is a previously unsuspected shuttling pre-ribosomal factor that is exported with the precursor particles and very rapidly reimported. Proteins that became cytoplasmic under drg1 mutant conditions were blocked on pre-60S particles at a step that precedes the association of Rei1, a later acting pre-ribosomal factor. We conclude that functional Drg1 is required for the release of shuttling proteins from the pre-60S particles at a maturation step that closely follows the nuclear export of particles, thus defining a novel step in eukaryotic ribo- some maturation. B1-46 Mitochondrial competence: The DN A import mechanism N. Ibrahim 1 , F. Weber-Lotfi 1 , M. Koulintchenko 1,2 , Y. Konstanti- nov 3 , H. Handa 4 , R. Lightowlers 2 , A. Dietrich 1 1 IBMP-CNRS, Strasbourg Cedex, FRANCE, 2 Newcastle Univer- sity, Newcastle Upon Tyne, UNITED KINGDOM, 3 SIFIBR-RAS, Irkutsk, RUSSIAN FEDERATION, 4 NIAS, Tsukuba, JAPAN. There are considerable gaps in the understanding of the mitoch- ondrial genetic systems and sterility- or disease-related mutations in the mitochondrial DNA cannot be complemented. This is mainly due to the fact that conventional transformation of mito- chondria has been unsuccessful for plants and mammals. How- ever, we have shown that isolated plant and mammalian mitochondria have a natural potential to incorporate, repair and express foreign DNA. To understand and optimize the process, we studied the import mechanism through biochemical and phy- siological approaches. The voltage-dependent anion channel (VDAC) was identified as the putative translocator through the outer membrane. In the case of plant mitochondria, DNA import seems to follow nucleotide transport pathways through the inner membrane and to be concomitant with phosphate uptake. To directly identify the import complex, we designed DNA substrates destined to get stuck in the membranes during translocation. Concerning the in vivo relevance of the process, we hypothesize that it is the basis for paternal transmission of an 11.6 kb mit- ochondrial plasmid in Brassica napus. The idea is supported by DNA import experiments with isolated Brassicacea mitochondria. From all these results, we expect to progress towards mitochond- rial transformation in vivo. Late Abstracts Posters 348 B1-47 VAMP2 affects the U-type inactivation of Kv2.1 through interaction with N-terminus of the channel D. Chikvashvili, A. Lvov, D. Greizer, I. Lotan Tel-Aviv University, Tel-Aviv, ISRAEL. We have previously demonstrated that physical and functional interactions of Kv2.1 with syntaxin 1A or the binary t-SNARE complex were mediated by the C-terminus of the channel. Recently, we focused on VAMP2, the vesicle associated partner of Syx and SNAP-25 in the ternary SNARE complex. Two-electrode voltage clamp analysis in oocytes showed that VAMP2 has an effect on the steady-state and kinetic parameters of the U-shaped inactivation of Kv2.1. Using co-immunoprecipitation from plasma membranes of Xenopus oocytes combined with in vitro binding assay with recombinant proteins, we demonstrated that VAMP2 interacts physically with Kv2.1. Notably, although VAMP2 can bind both the C- and N-termini of Kv2.1, as determinated by in vi- tro binding analysis, we present evidence that the interaction with the N-terminus is important for its functional effect on inactiva- tion. Firstly, effects of VAMP2 on Kv2.1 and Kv2.2 channels, which share high degree of their sequences homology at the N-ter- minus and the transmembrane region, are similar, reflecting VAMP2’s ability to bind these channels at their N-termini. Second- arily, deletions at the C-terminus of Kv2.1 do not abolish the effects of VAMP2 on the channel inactivation. Finally, injection of recombinant protein corresponding to the N-terminus reverses the effect of VAMP2 on the inactivation of Kv2.1. Taken together our results show that VAMP2 functions as a regulatory protein that modulates the U-shaped inactivation of Kv2.1 through interaction with the N-terminus of the channel. B1-48 Enzymatic properties and subcellular localization of Arabidopsis beta-N-acetylhexosa minidases E. Liebminger 1 , J. S. Bondili 2 , J. Schoberer 1 , F. Altmann 2 , J. Glo ¨ ssl 1 , H. Steinkellner 1 , L. Mach 1 , R. Strasser 1 1 Institute of Applied Genetics and Cell Biology, Vienna, AUSTRIA, 2 Departement of Chemistry, Vienna, AUSTRIA. Plant glycoproteins contain substantial amounts of N-glycans lack- ing terminal GlcNAc residues at their non-reducing ends. It has been proposed that this is due to the action of b-hexosaminidases during late stages of N-glycan processing or in the course of N-glycan turnover. We have now cloned the three putative b-hexosaminidase sequences from Arabidopsis thaliana. When heterologously expressed as soluble forms in insect cells, the enzymes (termed HEXO1-3) could all hydrolyze substrates like pNP-GlcNAc, pNP-GalNAc, chitotriose and chitobiose, albeit to a varying extent. With N-glycan substrates, HEXO1 displayed a more than 1000 times higher specific activity than HEXO2 and HEXO3. Subcellular localization studies with HEXO-fluorescent protein fusions transiently expressed in Nicotiana benthamiana plants showed that HEXO1 is a vacuolar protein. In contrast, HEXO2 and HEXO3 are mainly located at the plasma membrane. Analysis of HEXO expression in leaves revealed that HEXO1 is present in the soluble protein fraction, while HEXO3 is mainly membrane-bound and HEXO2 was not detected at all. These results indicate that HEXO1 participates in N-glycan trimming in the vacuole, whereas HEXO3 could be responsible for the process- ing of N-glycans present on secretory glycoproteins. B2-20 2D-PAGE versus shotgun proteomics: Secretome analysis of immature and differently stimulated human dendritic cells N. C. Gundacker, V. J. Haudek, A. Slany, H. Wimmer, E. Bayer, V. Bochkov, J. Stoeckl Medical University of Vienna, Vienna, AUSTRIA. Dendritic cells (DC), the most potent and specialised antigen-pre- senting cells, play a key role in the regulation of adaptive immu- nity. The secretome contains the most directly acting proteins affecting other cells. Immature human DCs were functionally acti- vated with lipopolysaccharide (LPS), lipid oxidation products derived from 1-palmitoyl-2-arachidoyl-sn-glycerol-3-phosphorylch- olin (OxPAPC) and human rhinovirus (HRV). During treatment, cells were metabolically labeled with 35 S-methionine/cysteine. In 2D-gels, secreted proteins were specifically detected by autoradiog- raphy and accessible to accurate quantification. Protein identifica- tion, however, was more successful by shotgun analysis using nano-flow HPLC peptide separation combined with a electrospray ion trap mass spectrometer. Here, a semi-quantitative assessment was performed with the spectral count method. By means of a home-made database, contaminants from residual fetal calf serum and cytoplasmic proteins were subtracted. As a result, 148 different secreted proteins were identified. Secretion was induced by LPS and HRV, but rather repressed by OxPAPC. 18 proteins were exclusively found secreted by LPS-treated, 6 by OxPAPC-stimula- ted and 16 by HRV-infected DCs. The present approach proved to be very powerful for the analysis of this meaningful protein frac- tion. B3-20 Characterising the multitude of chloroplast protein import receptors in Arabidopsis thaliana M. Gutensohn, B. Hust Martin-Luther-University Halle-Wittenberg, Halle/Saale, GERMANY. The majority of chloroplast proteins are encoded in the nucleus and have to be imported into the organelle after their synthesis in the cytosol as precursor proteins. The transport of precursors across the two chloroplast envelope membranes is mediated by the interaction with two import machineries, the Toc and Tic complex. The core of the Toc complex consists of two receptor proteins, Toc34 and Toc159, involved in initial binding of precursor proteins at the chloroplast surface and a translocation pore, Toc75. In Ara- bidopsis two homologs of Toc34 (atToc33, atToc34) and four homologs of Toc159 (atToc159, atToc132, atToc120, atToc90), however, only one ortholog of Toc75 (atToc75-III), have been identified. For the functional characterisation we have isolated knockout mutants for all six Arabidopsis Toc receptors as well as for the translocation pore. The atToc75-III knockout mutant has an embryo lethal phenotype with an extremely early arrest of development. In contrast, the Toc receptor mutants show remark- ably different phenotypes suggesting more specialized functions for these import receptors. Detailed characterisation of the Toc recep- tor mutants, including proteome and expression analyses, as well as in vitro studies demonstrated that each of these Toc receptors preferentially interacts with different groups of precursor proteins. The composition of Toc complexes in Arabidopsis has been ana- lysed by biochemical approaches using specific antibodies as well as by a genetic approach using a series of Toc receptor double mutants. Posters Late Abstracts 349 B3-21 The Pex3p-Pex19p interactions modulated by PMP cargo proteins C. P. Grou 1,2 , M. P. Pinto 1,2 , I. S. Alencastre 1 , M. E. Oliveira 1,2 , C. Sa ´ -Miranda 1 , M. Fransen 3 , J. E. Azevedo 1,2 1 IBMC, Porto, PORTUGAL, 2 ICBAS, Porto, PORTUGAL, 3 Katholieke Universiteit Leuven, Leuven, BELGIUM. Biogenesis of the mammalian peroxisomal membrane seems to require the action of only three peroxins: Pex3p, Pex16p, two intrinsic membrane proteins of the peroxisome, and Pex19p, a pro- tein that displays a dual subcellular distribution (cytosolic and per- oxisomal). Pex3p is believed to be the docking protein for Pex19p- PMP complexes at the peroxisomal membrane. We have analyzed the Pex3p-Pex19p interaction by Native-PAGE. Our data indicate that the Pex3p-binding affinity of Pex19p is highly increased when this peroxin is pre-loaded with a PMP. These findings suggest the existence of a cargo-induced peroxisomal targeting for Pex19p. Supported by: Fundac¸ a ˜ o para a Cieˆ ncia e Tecnologia Grant POCTI2010 and FEDER Funds, Portugal, European Union VI Framework program Grant LSHG-CT-2004-512018, Peroxisomes in Health and Disease, the Flemish Government (Geconcerteerde Onderzoeksacties Grant GOA/2004/08) and Fonds voor Wet- enschappelijk Onderzoek Vlaanderen (Onderzoeksproject Grant G.0237.04). B3-22 The role of Pex19p in the biogenesis of peroxisomal membrane proteins M. P. Pinto 1,2 , C. P. Grou 1,2 , I. S. Alencastre 1,2,2 , M. E. Oliveira 1,2 ,C.Sa ´ -Miranda 1 , M. Fransen 3 , J. E. Azevedo 1,2 1 IBMC, Porto, PORTUGAL, 2 ICBAS, Porto, PORTUGAL, 3 Katholieke Universiteit Leuven, Leuven, BELGIUM. Only three proteins, Pex3p, Pex16p and Pex19p, were found to be involved in the biogenesis of the mammalian peroxisomal mem- brane. Pex3p and Pex16p are two intrinsic membrane proteins of the peroxisome, while Pex19p localizes both to cytosol and to the organelle. The ability of Pex19p to interact with most peroxisomal membrane proteins (PMPs) led to the proposal that this peroxin functions as an import receptor or as a chaperone in the assembly/ disassembly of membrane protein complexes. Using an in vitro import system, we show that insertion of a reporter protein into the peroxisomal membrane is Pex19p-dependent and does not require ATP/GTP hydrolysis. By programming the system with recombinant versions of Pex19p, we demonstrate that Pex19p forms a complex with its cargo PMP before the peroxisomal dock- ing/insertion steps. Supported by: Fundac¸ a ˜ o para a Cieˆ ncia e Tecnologia Grant POCTI2010 and FEDER Funds, Portugal, European Union VI Framework program Grant LSHG-CT-2004-512018, Peroxisomes in Health and Disease, the Flemish Government (Geconcerteerde Onderzoeksacties Grant GOA/2004/08) and Fonds voor Wet- enschappelijk Onderzoek Vlaanderen (Onderzoeksproject Grant G.0237.04). B3-23 Mdm38p and its role in mitochondrial dynamics K. Nowikovsky 1 , R. J. Schweyen 1 , S. Reipert 2 , R. J. Devenish 3 1 MFPL, Department of Genetics, Wien, AUSTRIA, 2 MFPL, Department of Molecular Cell Biology, Wien, AUSTRIA, 3 Depart- ment of Biochemistry and Molecular Biology, and the ARC Centre of Excellence in Structural and Functional Microbial Genomics, Monash University, Victoria, AUSTRALIA. Yeast MDM38/YOL027c and its human homologue LETM1 encode an integral protein of the inner mitochondrial membrane. As previously reported, LETM1 is the candidate gene for seizures in the Wolf-Hirschorn-syndrom (Zollino et al., 2003). Loss of the MDM38 gene product in yeast mitochondria results in a variety of phenotypic effects including reduced content of respiratory chain complexes, altered mitochondrial morphology and loss of mitochondrial K+/H+ exchange activity resulting in osmotic swelling. By use of doxycycline-regulated shut-off of MDM38 gene expression we showed that loss of K+/H+ exchange activ- ity and mitochondrial swelling are early events, associated with a reduction in membrane potential and fragmentation of the mit- ochondrial reticulum. The use of a novel fluorescent biosensor directed to the mitochondrial matrix revealed that the loss of K+/H+ exchange activity was immediately followed by morpho- logical changes of mitochondria and vacuoles, the close associ- ation of these organelles and finally uptake of mitochondrial material by vacuoles. Nigericin, a K+/H+ ionophore, fully pre- vented these effects of Mdm38p depletion. We conclude that osmotic swelling of mitochondria triggers selective mitochondrial autophagy, or mitophagy. Furthermore, we provide evidence that fragmentation of the mitochondrial network is a prerequisite for mitochondrial degradation. B3-24 Biogenesis of b-barrel proteins in the outer membrane of human mitochondria V. Kozjak-Pavlovic, K. Ross, N. Benlasfer, S. Kimmig, A. Karlas, T. Rudel Max Planck Institute for Infection Biology, Berlin, GERMANY. Voltage-dependent anion-selective channel (VDAC), also known as mitochondrial porin, is one of the few proteins in the outer mit- ochondrial membrane predicted to have a ß-barrel topology. From studies in fungi, it is known that VDAC, similar to other ß-barrel proteins, requires the translocase of the outer mitochondrial mem- brane (TOM), as well as the sorting and assembly machinery (SAM) for its translocation and proper integration into the outer membrane of mitochondria. We studied the biogenesis of VDAC in human mitochondria by depleting the components of the mit- ochondrial import machinery by using RNA interference. We gen- erated stably transduced cell lines that showed inducible knockdown of Tom40, Sam50, Tom70 and metaxin 2. Using these cell lines, we were able to show that besides the TOM complex and Sam50 (Tom55/Omp85), so far the only known component of the human SAM complex, metaxins were also necessary for VDAC biogenesis. Metaxin 2-depleted mitochondria had reduced levels of both metaxin 1 and metaxin 2, and were deficient in the import and assembly of VDAC and Tom40, another ß-barrel precursor, but not in the import of three matrix-targeted precursors. We also observed a reduction in the amounts of metaxin 1 and metaxin 2 in Sam50-depleted mitochondria, implying a connection between these three proteins. However, we found metaxin 1 and metaxin 2 in a high-molecular-weight complex distinct from the Sam50-con- taining complex. We propose that in human mitochondria Sam50 and metaxins act sequentially to promote integration of ß-barrel proteins into the outer mitochondrial membrane. Late Abstracts Posters 350 . Invited Lectures A1-L5 Mitotic chromosome condensation and segration C. D’Ambrosio 1 ,. binding components provide dynamic multi-site targeting and entry of DNA. Invited Lectures Late Abstracts 341 E1-L2 Global Mapping of the Yeast Genetic Interaction