Swiss Rna Workshop

s of oral presentations (in order of the program) 5 Analysis of pre-mRNA splicing in cell lines derived from patients with retinitis pigmentosa and dominant mutations in pre-mRNA splicing factor genes PRPF31, PRPF8 and PRPF3 Goranka Tanackovic, Adriana Ransijn, Jacques S. Beckmann, Eliot L. Berson and Carlo Rivolta Department of Medical Genetics, University of Lausanne, Switzerland The Berman-Gund Laboratory for the Study of Retinal Degenerations, Harvard Medical School, Boston, USA Pre-mRNA splicing is a process in which introns are removed from pre-mRNA and exons are joined together to produce mRNA. It is catalyzed by complex molecular machinery termed spliceosome, composed of five snRNPs and numerous non-snRNP proteins. In addition to providing mature mRNA molecules to the cell, it contributes to protein diversity in higher eukaryotes through the process of alternative splicing, which allows the production of different mRNA molecules, and thus proteins, from a single pre-mRNA. Dysfunctions of splicing often lead to diseases and one of these is Retinitis pigmentosa (RP). RP is the most common form of hereditary retinal degeneration, affecting 1/4000 individual worldwide. Clinically, it is characterized by progressive loss of vision and can lead to complete blindness because of the death of photoreceptors, the light-sensing cells of the retina. Genetically, this is a very heterogeneous disorder, caused by mutations in roughly 50 different genes. Most of these are retina-specific or have a well-defined role in the physiology of photoreceptors; others are expressed in many tissues or are ubiquitous and yet cause a phenotype that is retina-specific. Among the ubiquitous genes are PAP-1 (RP9), PRPF31 (RP11), PRPF8 (RP13), and PRPF3 (RP18), all encoding pre-mRNA splicing factors that are very conserved and essential for viability. All these genes encode proteins that are part of U4/U6-U5 tri-snRNP particle that represents, in addition to U1 and U2 snRNPs, a key component of the spliceosome. The aim of our work is to understand why mutations in the ubiquitously expressed genes: PRPF31, PRPF8 and PRPF3 affect only one tissue – the retina. Moreover, from the mechanistic point of view, the nature of the mutations in these RP genes is different. Most of the PRPF31 mutations lead to premature stop codons and degradation of the mutant mRNA by nonsensemediated decay, therefore in heterozygous patients PRPF31 is expressed at lower amounts and only from the wild-type allele, probably causing the disease through an haploinsufficiencydependent mechanism. The RP-causing mutations in PRPF8 and PRPF3 genes are missense mutations or indels leading to premature stops in the last exon and may therefore potentially act as dominant-negative alleles. We are interested to know how decreased levels of PRPF31 protein or mutations in PRPF3 and PRPF8 affect pre-mRNA splicing. We have studied in vitro spliceosome formation, as well as in vitro and in vivo splicing using lymphoblastoid cells derived from RP patients carrying various mutations in the abovementioned PRPF genes. Our results show that mutations in these genes slow down kinetics of spliceosome assembly; however they do not block splicing of the pre-mRNA tested. This could indirectly suggest that tissue-specific alternative splicing patterns could be affected in the case of RP11, RP13 and RP18 mutations and we are currently testing this hypothesis. Abstracts of oral presentations (in order of the program) 6 SR, hnRNP, PKC and EJC: Converging complexity to control the splicing of the apoptotic regulator Bcl-x Benoit Chabot Département de microbiologie et d’infectiologie, Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, Québec, Canada The Bcl-x pre-mRNA is alternatively spliced to produce the anti-apoptotic Bcl-xL and the pro-apoptotic Bcl-xS isoforms. We have identified a variety of exon elements that control the use of the 5’ splice sites of Bcl-x. One element is bound by SRp30c to activate the xL site while others are bound by hnRNP F/H et hnRNP K to activate and repress the xS site. Another element strongly represses the xS site and requires protein kinase C activity. This sequence element appears to represent a converging platform for many signaling routes that monitor stresses and translation efficacy. Further, our screening platform has identified two components of the exon junction complex (EJC) involved in the control of Bcl-x splicing. These proteins also coregulate the alternative splicing of other apoptotic genes. Because EJC components are important for the decay of aberrant mRNAs with premature termination codons, a reduction in the level of these proteins may trigger apoptosis by encouraging the production of pro-apoptotic splice forms. Abstracts of oral presentations (in order of the program) 7s of oral presentations (in order of the program) 7 In vivo splicing modulation rescues a severe mouse model for Spinal Muscular Atrophy Kathrin Meyer, Rachel Nlend Nlend, Judith Trüb, Julien Marquis and Daniel Schümperli, Institute of Cell Biology, University of Bern, Baltzerstrasse 4, CH-3012 Bern Spinal muscular atrophy (SMA), the most common autosomal recessive disorder leading to infant death, is caused by homozygous deletion/inactivation of the SMN1 gene (survival of motoneurons) and leads to the progressive degeneration of α-motoneurons in the spinal cord. SMN1 codes for an ubiquitously expressed, essential protein with a major role in small nuclear ribonucleoprotein (snRNP) assembly. Survival of affected individuals into early childhood is due to a second gene (SMN2) that emerged from gene duplication. However, SMN2 is no complete substitute, because a silent C-T transition impairs the inclusion of exon 7 in its mRNA. Most of the produced protein is truncated and unstable (SMN∆7). As all patients carry at least one SMN2 copy, boosting its exon 7 inclusion is hypothesised to be an effective SMA therapy. So far, however, this has not been achieved in vivo, as it is difficult to specifically alter the splicing of a single mRNA with conventional drugs. Our laboratory has pioneered methods to modulate the splicing of specific mRNAs with modified U7 snRNAs. In the case of SMN2, we found that a bifunctional U7 snRNA carrying an antisense sequence targeting exon 7 combined with a splicing enhancer sequence can efficiently stimulate exon 7 inclusion and SMN protein levels in SMA patient fibroblasts. Tests with an animal model lacking the mouse smn gene and carrying 2 copies of human SMN2 show that the disease phenotype usually leading to death of these mice within 1 week can be overcome if they additionally contain the bifunctional U7 snRNA gene. While the oldest of these SMA/U7 mice are now nearly one year old, the median survival time depends on the integrated copy number of our therapeutic RNA and lies around 123 days. Besides being an important milestone towards the development of a somatic gene therapy for SMA, this result effectively proves the therapeutic value of an SMN2 exon 7 inclusion strategy. As it is still unclear if SMA is rather a developmental disorder, the time point when a therapy would have to be started is also unknown. To address this question we have generated a Doxocyclin inducible U7 that will be used for future transgenesis. Like that we will be able to start the splicing correction at different time points and will be able to evaluate if a treatment starting after birth will show equal benefits to the SMA mice. Abstracts of oral presentations (in order of the program) 8 NMR structure of the tra21 splicing factor in complex with RNA revealed an unexpected extensive interaction with the C-terminal extremity of the RRM A. Cléry, C. Dominguez and F. Allain Institute for Molecular Biology and Biophysics, Swiss Federal Institute of Technology (ETH) Zürich, Switzerland Human transformer2-beta1 (tra21) protein is an SR-like protein that promotes in a concentration-dependent manner the inclusion of a large number of alternative exons [1, 2]. Despite its autoregulation, tra21 expression is unphysiologically high in breast and ovarian cancer [3, 4], hypoxia [5, 6], silicosis [7] and arteriosclerosis [8]. This protein contains two RS domains mainly involved in protein-protein interactions and a central RNA recognition motif (RRM) responsible for its specific interaction with RNA [1, 2]. Although the function of tra21 depends on its binding on pre-mRNA target sequences the molecular basis of this interaction is not well characterized. Using NMR spectroscopy, we identified the AAGAAC sequence as the best candidate to study the structure of the tra21 RRM in complex with RNA. We solved the solution structure of this protein in complex with the 5’-AAGAAC-3’ RNA. We obtained a precise structure due to 81 intermolecular NOEs. Interestingly, in addition to the -sheet surface, the C-terminal extremity of the RRM is extensively used to bind the target RNA. In addition, our data clearly indicate that the GAA nucleotides are specifically recognized. The determination of this RNA-protein complex brings essential information on the specificity of RNA recognition by tra21 and then on its function and enriches the current knowledge about the versatile RRM-RNA interaction. 1. Stoilov, P., et al., Hum Mol Genet, 2004. 13(5): p. 509-24. 2. Tacke, R., et al., Cell, 1998. 93(1): p. 139-48. 3. Fischer, D.C., et al., Oncol Rep, 2004. 11(5): p. 1085-90. 4. Watermann, D.O., et al., Cancer Res, 2006. 66(9): p. 4774-80. 5. Daoud, R., et al., J Neurosci, 2002. 22(14): p. 5889-99. 6. Matsuo, N., et al., J Biol Chem, 1995. 270(47): p. 28216-22. 7. Segade, F., et al., J Immunol, 1995. 154(5): p. 2384-92. 8. Tsukamoto, Y., et al., Am J Pathol, 2001. 158(5): p. 1685-94. Abstracts of oral presentations (in order of the program) 9s of oral presentations (in order of the program) 9 An assembly chaperone collaborates with the SMNcomplex to generate spliceosomal snRNPs Ashwin Chari, Monika M. Golas, Michael Klingenhäger, Nils Neuenkirchen, Björn Sander, Clemens Englbrecht, Holger Stark, and Utz Fischer Department of Biochemistry, University of Würzburg, Germany; Research Group of 3D Electron Cryomicroscopy, MPI für Biophysikalische Chemie, Göttingen, Germany A structural hallmark of splicing U snRNPs is a ring shaped core that forms upon binding of Sm proteins onto U snRNA. PRMT5and SMN-complexes are required, in trans, to mediate core assembly in vivo but their mode of function is unknown. We will report on the biochemical and structural dissection of core formation, which has yielded mechanistic insight into key steps of the assembly process. In an early phase of assembly, a ring-shaped subunit is formed on the PRMT5 complex composed of pICln and a distinct subset of Sm proteins. This stable pICln-Sm complex with a sedimentation coefficient of 6S is likely to be identical to an RNA-free assembly intermediate described by Blobel and co-workers decades ago, but whose function and composition has remained elusive. Sm proteins are incapable of assembling onto RNA out of this state due to a steric inhibition of the RNA binding site by pICln. The resulting kinetic trap is resolved by the SMN-complex, which contacts the 6S ring on the outer surface and ejects pICln. As a consequence, the formerly closed ring opens so that the Sm proteins can now contact the U snRNA. The SMN-complex then seals the ring around snRNA in a catalytic manner and releases the assembled U snRNP. Our data thus identify pICln as an assembly chaperone and the SMN-complex as a catalyst of the transition from pICln-Sm complexes to U snRNPs. Although disparate to DNA clamp loaders, this combined chaperone/enzyme system acts structurally similar. Abstracts of oral presentations (in order of the program) 10 Trans-acting antisense RNAs mediate transcriptional gene cosuppression in S. cerevisiae Nissrine BEYROUTHY, F. Stutz lab, Department of Cell Biology, University of Geneva Homology-dependent gene silencing, a phenomenon initially described as cosuppression in plants, depends on small interfering RNAs. We found evidence that in S. cerevisiae, which is missing the RNAi machinery, protein coding gene cosuppression exists. Indeed, transformation of a centromeric PHO84 plasmid or integration of an additional PHO84 gene within the genome result in the cosilencing of both the transgene and the endogenous gene. This repression is transcriptional, position independent and requires trans-acting antisense RNAs. We show that antisense RNAs induce transcriptional gene silencing both in cis and in trans and that the two pathways differ by the implication of the Hda1/2/3 HADC complex. Finally, we report that trans silencing is influenced by the Set1 histone methyltransferase, which regulates antisense RNA production. All together our data represent the first example of protein coding gene cosuppression in S. cerevisiae and highlight the role of antisense RNAs in mediating RNAi independent transcriptional gene silencing. Abstracts of oral presentations (in order of the program) 11s of oral presentations (in order of the program) 11 Yeast telomere repeat containing RNA (TERRA) and telomere length regulation Brian Luke, Andrea Panza, Nahid Iglesias, Sophie Redon and Joachim Lingner ISREC, chemin des boveresses 155, Epalinges 1066, Switzerland Recent advances have demonstrated that surprisingly telomeres are actively transcribed into a non-coding telomere repeat containing RNA (TERRA), which remains associated with the telomere and may regulate telomere structure and maintenance (Azzalin et al, 2007; Schoeftner & Blasco, 2007). Interestingly, in human cells, the localization of TERRA to chromosome ends is under the control of the nonsense-mediated RNA decay (NMD) machinery. Our results demonstrate that in yeast TERRA levels are controlled by the conserved 5’ to 3’ exonuclease, Rat1, as inactivation of the temperature sensitive rat1-1 allele results in rapid accumulation of TERRA. Through genetic approaches we have determined that the accumulation of TERRA in yeast, results in telomere shortening through the inhibition of telomerase. Strikingly TERRA induced telomere shortening can be overcome through the overproduction of RNAseH within the cells. These data support the idea that the formation of an RNA/DNA hybrid inhibits telomerase action at the chromosome end. We have uncovered a link between the maintenance of telomere length homeostasis through Rap1 and the production of TERRA. Interestingly, cells expressing two different alleles of RAP1, rap1-1 and rap1-2, have increased levels of TERRA while at the same time harbor short telomeres. The short telomere phenotype is not enhanced when the rat1-1 mutation is introduced into either of these mutant backgrounds indicating an epistatic relationship. Taken together these data indicate that the achievement of telomere length homeostasis through the counting of Rap1 molecules may go through a non-coding RNA intermediate (TERRA) and that the number of Rap1 molecules may be influencing the rate of TERRA transcription. Abstracts of oral presentations (in order of the program) 12 Stress-dependent Coordination of Transcriptome and Translatome in Yeast Regula E. Halbeisen, André P. Gerber Institute of Pharmaceutical Sciences, ETH Zurich, CH-8093 Zurich, Switzerland. Cells rapidly alter gene expression in response to environmental stimuli such as nutrients, hormones and drugs. During the imposed ‘remodeling’ of gene expression, changes in the levels of particular mRNAs do not necessarily correlate with those of the encoded proteins, which could in part rely on the differential recruitment of mRNAs to translating ribosomes. To systematically address this issue, we have established a novel approach to rapidly access the translational status of each mRNA in the yeast Saccharomyces cerevisiae by affinitypurification of endogenously formed ribosomes and the analysis of associated mRNAs with DNA microarrays. Using this method, we compared changes in total mRNA expression (transcriptome) with ribosome associations (translatome) after the application of different conditions of cellular stress. Severe stresses, induced by amino acid depletion or osmotic shock, stimulated highly correlated responses affecting about 15% of both transcriptome and translatome. Many of the regulated messages code for functionally related proteins, thus reflecting logical responses to the particular stress. In contrast, mild stress provoked by addition of Calcofluor-white and menadione altered the translatome of approximately 1% of messages with only marginal effects on total mRNA, implying largely uncorrelated responses of transcriptome and translatome. Among these putative translationally regulated messages were most components of the mitochondrial ATPase. Increased polysome associations of corresponding messages and higher mitochondrial ATPase activities upon treatment confirmed the relevance for regulation of this macromolecular complex. Our results suggest the presence of highly sensitive translational regulatory networks that coordinate functionally related messages. These networks are preferentially activated for rapid adaptation of cells to minor environmental perturbations. Abstracts of oral presentations (in order of the program) 13s of oral presentations (in order of the program) 13 Translational control by miRNAs and RNA-binding proteins Matthias W. Hentze European Molecular Biology Laboratory, Meyerhofstrasse 1, D-69117 Heidelberg, Germany Protein synthesis is controlled via two interdependent processes: the assembly of translation factors and ribosomes on the mRNA template, and the formation of mRNPs that promote or impede the former process. Hence, regulation of the activity of translation factors and modulation of mRNPs by regulatory trans-acting factors [RNA-binding proteins, microRNAs (miRs)] have emerged as the major principles for the control of protein synthesis. Since the beginning, we have investigated the molecular mechanisms underlying translational control by trans-acting factors. Methodologically, the establishment of faithful cell-free systems has frequently served as a key step towards deciphering the underlying mechanisms. Specifically, I plan to discuss new insights into the regulation of Drosophila msl-2 mRNA translation by the RNA-binding protein Sex-lethal (SXL) and the co-repressor UNR, as well as the mechanism of translational control by Drosophila miR 2. Thermann and Hentze, Nature 447, 875-879 (2007) Till et al., Nature Struc. Mol. Biol. 14, 897-903 (2007) Duncan et al., Genes & Dev. 20, 368-379 (2006) Beckmann et al., Cell 122, 529-540 (2005) Abstracts of oral presentations (in order of the program) 14 Function and Regulation of the let-7 MicroRNA Xavier Ding, Ingo Büssing, Monika Fasler and Helge Grosshans, Friedrich Miescher Institute for Biomedical Research, CH-4002 Basel. [email protected] MicroRNAs (miRNAs) are a novel class of genes that account for >1% of genes in a typical animal genome. They constitute an important layer of gene regulation that affects diverse processes such as apoptosis, cell differentiation, and stem cell maintenance. Although it is generally agreed that miRNAs regulate their targets posttranscriptionally, through an antisense mechanism, the precise mechanism(s) of action remains controversial. Regulation at the level of mRNA stability, translation initiation, translation elongation, polyadenylation, and/or protein stability have all been invoked. Guided by results from a genetic screen, where we established interaction between C. elegans let-7 and the translation machinery, we used polysome profiling by sucrose density gradient centrifugation to demonstrate that let-7 blocks translation of its endogenous targets at the initiation level. This finding overturns a dogma that miRNA repression occurs downstream of translation initiation in vivo, and we demonstrate repression of translation initiation for multiple, endogenous targets of multiple miRNAs. In addition, and reconciling some controversies in the field, we find that miRNA target transcript degradation frequently occurred together with translational repression, but appears to constitute a separate process. Finally, we find that the GW182 proteins AIN-1 and AIN-2 are essential both for miRNA target degradation and repression of translation initiation – an unexpected result as GW182 proteins were previously thought to play an auxiliary role, largely restricted to aiding mRNA degradation. We have also begun to identify additional genes that are required for miRNA function, and we will discuss our progress on the characterization of these genes. Abstracts of oral presentations (in order of the program) 15s of oral presentations (in order of the program) 15 To be or not to be a miRNA target site Jean Hausser, Markus Landthaler, Dimos Gaidatzis, Thomas Tuschl, Mihaela Zavolan University of Basel, Biozentrum. 4056 Basel, Switzerland How miRNAs recognize their target sites is a puzzle that many experimental and computational studies aimed to solve. Several features such as perfect pairing of the miRNA seed, additional pairing in the 3' region of the miRNA, relative position in the 3' UTR and the A/U content of the environment of the putative site have been found relevant. Based on a large number of experiments comprising transcriptomics and proteomics datasets in addition to mRNA affinity profiles to the main effector of RNAi obtained through Ago2 immuno-precipitation, we have assessed the power that each of these features has in distinguishing between putative sites that do and those that do not appear to be functional. Our results indicate that the experimental datasets give widely different pictures of the importance of these features. They additionally suggest that miRNA target sites have been selected in evolution on their ability to have energetically favorable interactions with the miRNAs and to trigger mRNA degradation. Finally, using the features we identified as most relevant for miRNA targeting, we trained a model to predict miRNA targets with improved accuracy. We show that moving away from the common "seed-based" approach to the prediction of miRNA-mRNA interactions, our model can predict sites that only match 5 of the first 8 nucleotides of the miRNA and yet induce mRNA degradation. Abstracts of oral presentations (in order of the program) 16 MicroRNA feed back loop in cocaine induced synaptic plasticity and addiction. Vijay Chandrasekar and Jean-Luc Dreyer. Institute of Biochemistry, University of Fribourg. Rue du Musée 5, CH-1700 Fribourg, Switzerland. Cocaine orchestrates synaptic plasticity by a surfeit of gene regulatory mechanisms. MicroRNAs, a class of small non-coding regulatory molecules, play a major role in temporal development and neurogenesis of complex physiological pathways in the brain. The focus of the present study is the elucidation of miRNA pathway in behavioral sensitization to cocaine and implications of differentially regulated miRNAs in the cocaine-induction of synaptic plasticity. Cocaine is one of the most reinforcing psycho-stimulant and blocks the dopamine transporter, thus inhibiting dopamine re-uptake and producing increased synaptic dopamine levels. Many axon guidance molecules implicated in cocaine mediated neural plasticity are regulated by miRNAs. A screening for cocaine specific miRNAs revealed the role of specific miRNAs. Quantitative RT-PCR analysis of several precursor and mature miRNAs showed that these miRNAs were significantly regulated under chronic cocaine administration in the mesolimbic dopaminergic system compared to saline treated controls. We found that mature miRNA profile correlates as expected with the pattern from precursor miRNAs in specific brain areas. These results were confirmed by the miR-VANA method (Ambion), by Northern blots and by in-situ hybridization analysis using LNA probes. Furthermore, miRNAin vitro reporter assays using lentiviruses expressing miRNAs consolidated our observation of an inversely proportional concentration of targeting miRNAs and the targeted genes in cocaine addiction. Our results suggests, that miRNAs are involved in a complex signal transduction cascade activated by cocaine and may be involved in the regulation of many downstream genes upon drug treatment. The behavioral implication of expression of these miRNA genes in specific brain areas will be described. Abstracts of oral presentations (in order of the program) 17s of oral presentations (in order of the program) 17 SMG6 promotes endonucleolytic cleavage of nonsense mRNA in human cells Andrea B. Eberle, Søren Lykke-Andersen, Oliver Mühlemann & Torben Heick Jensen; presented by Pamela Nicholson Institute of Cell Biology, University of Bern, Switzerland. Centre for mRNP Biogenesis and Metabolism, Department of Molecular Biology, Aarhus University, Denmark. In eukaryotes, mRNAs harbouring premature translation-termination codons (PTCs) are recognized and eliminated by nonsense-mediated mRNA decay (NMD). In addition to its quality control function, NMD constitutes a translation-dependent post-transcriptional pathway to regulate the expression levels of alternatively spliced mRNAs. Moreover, with an estimated 30% of all known human disease-associated mutations generating a nonsense mRNA, NMD is an important modulator of genetic disease phenotypes. Although NMD is conserved from yeast to man, previous reports suggest significantly different RNA degradation mechanisms between species. In Drosophila, an unidentified endonuclease initiates NMD by cleaving nonsense mRNA near the PTC, which exposes the two new ends for subsequent rapid exonucleolysis. In contrast, the current view posits that in yeast and human cells, elimination of PTC-containing RNA occurs from one or both termini, initiated by deadenylation and/or decapping followed by exonucleolytic decay. Here we report that also in human cells, degradation of nonsense mRNA can be initiated by a PTC-proximal endonucleolytic cleavage. Furthermore, we identify the metazoan-specific NMD factor SMG6 as the responsible endonuclease by demonstrating that mutation of conserved residues in the catalytic center of its nuclease domain the C-terminal PIN motif – abolishes endonucleolysis in vivo and in vitro. Thus, our data reveal a new degradation pathway for nonsense mRNA in human cells, and suggest that endonucleolytic cleavage is a conserved feature in metazoan NMD. Abstracts of oral presentations (in order of the program) 18 Genome-wide analysis of the non-canonical poly(A) polymerases Trf4p and Trf5p: more than RNA quality control Salvatore San Paolo, Walter Keller Lab, Biocenter, University of Basel Trf4p and Trf5p are non-canonical yeast poly(A)polymerases that promote exosome-mediated degradation of aberrant and short lived RNA substrates. To assess the level of functional redundancy between Trf4p and Trf5p and to investigate the role of the Trf4-dependent polyadenylation in vivo we used DNA microarrays to compare the transcription profiles of the wild type (WT) strain of S. cerevisiae with either that of trf4∆ or trf5∆ mutant strains or the trf4∆ mutant expressing the polyadenylation defective Trf4(DADA) protein. We found that most of the transcripts that were upregulated in the trf4∆ or the trf5∆ mutants fell into two distinct groups of trf4-dependent and trf5-dependent transcripts. Surprisingly most RNAs, whose expression was affected by the trf4 deletion, were restored to WT levels by the overexpression of TRF4(DADA) suggesting that the Trf4p-DADA is functional in vivo. Apart from previously reported RNAs, our analysis revealed that subsets of spliceosomal introns were also substrates for the Trf4 and Trf5 protein complexes (TRAMP4/TRAM5) and did not require the Trf4p-dependent polyadenylation for degradation. In vivo cross-linking and RNA immunoprecipitation-chip experiments (RIP-chip) provided further experimental support for a direct role of Trf4p in intron decay. In addition, although subtelomeric RNAs accumulated in the trf4∆, the trf5∆ and the rrp6∆ mutants, only disruption of trf4 caused a severe shortening of telomeres, suggesting that TRF4 functions in telomere maintenance. Finally, we show that TRF4, the exosome and TRF5 participate in antisense-RNA mediated regulatory pathways that modulate the expression of genes involved in phosphate metabolism. Abstracts of oral presentations (in order of the program) 19s of oral presentations (in order of the program) 19 In vivo analysis of selenocysteine synthesis shows a single pathway in eukaryotes Eric Aeby, Sotiria Palioura, Mascha Pusnik, Janine Marazzi, Allyson Lieberman, Elisabetta Ullu, Dieter Söll, and André Schneider 1 Department of Chemistry and Biochemistry, University of Bern, Freiestr. 3, CH-3012 Bern, Switzerland; 2 Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520-8114, USA; 3 Department of Internal Medicine, Yale University Medical School, New Haven, CT 06536-0812, USA. Selenocysteine (Sec) is the 21 amino acid that is cotranslationally inserted into proteins of many organisms in response to UGA codons. A comprehensive in vivo study on the mechanism of selenoprotein synthesis in eukaryotes is still lacking. Here we have analyzed the in vivo Sec-tRNA formation pathway in the parasitic protozoa Trypanosoma brucei: selenoprotein synthesis was completely abolished in null mutants of either phosphoseryltRNA kinase or Sep-tRNA:Sec-tRNA synthase, demonstrating that there is a single pathway for Sec-tRNA formation requiring the sequential action of both enzymes. Growth of the two knock out strains was not impaired indicating that in trypanosomes, unlike in mammals, selenoproteins are dispensable for life. Moreover, analysis of conditional RNAistrains indicates the essential roles of seryl-tRNA synthetase, selenophosphate synthase and Sec-specific elongation factor for selenoprotein synthesis. These results together with the recently in vitro reconstituted mammalian pathway strongly suggest that eukaryotes have a single highly conserved Sec-tRNA formation pathway. Abstracts of oral presentations (in order of the program) 20 Alternative RNA editing creates novel proteins in trypanosome mitochondria Torsten Ochsenreiter, Institute of Cell Biology, University of Bern The mitochondrial genome of trypanosomes is composed of thousands of topologically interlocked circular DNA molecules that form the kinetoplast DNA (kDNA). Most genes encoded by the kDNA require a posttranscriptional modification process called RNA editing to form functional mRNAs. Here we show that alternative editing of the mitochondrial cytochrome c oxidase III (COXIII) mRNA in Trypanosoma brucei produces a novel DNA binding protein, alternatively edited protein-1 (AEP-1). AEP-1 localizes to the region of the cell between the kDNA and the flagellum and purifies with the tripartide attachment complex (TAC), a structure believed to physically link the kDNA and flagellar basal bodies. Expression of the DNA binding domain of AEP-1, results in aberrant kDNA structure and reduced cell growth indicating that AEP-1 is involved in the maintenance of the kDNA. Perhaps most important, our studies show a gain of function through an alternatively edited mRNA and for the first time, provide a link between the unusual structure of the kDNA and RNA editing in trypanosome mitochondria. Abstracts of poster presentations (numbered) 21s of poster presentations (numbered) 21