Clp1, a novel RNA-kinase for 3´ end formation

RNA interference (RNAi) allows the analysis of gene function by introducing synthetic, short interfering RNAs (siRNAs) into cells. In contrast to siRNA and micro-RNA (miRNA) duplexes generated endogenously by the RNaseIII endonuclease Dicer and thereby bearing a 5` phosphate group, synthetic siRNAs display a 5` hydroxyl group. However, in order to become incorporated into the RNA-Induced Silencing Complex (RISC) and mediate target RNA cleavage, the guide strand of a synthetic siRNA must display a phosphate group at the 5` end. The responsible kinase activity has so far remained elusive. Monitoring siRNA phosphorylation, we applied a classical chromatographic approach that resulted in the identification of the protein hClp1 (human Clp1), a known component of both tRNA splicing and mRNA 3` end formation machineries, as the siRNA-kinase. We have revealed hClp1 as the kinase activity that phosphorylates and licenses synthetic siRNAs to be assembled into RISC and to mediate target RNA cleavage. In addition, we have shown that hClp1 phosphorylates the 5` end of the 3` exon during human tRNA splicing, an event that is required for the ligation of both exons to generate a mature tRNA. Here, we propose to investigate the role of hClp1 in mRNA 3` end formation. We will therefore address the function of hClp1 in mRNA cleavage/polyadenylation, and in transcription termination, where RNA polymerase II is dismantled from its DNA template by a 5` phosphate-dependent exonuclease activity. A second goal of this proposal stems from studying the role of hClp1 in tRNA splicing, and consists in the identification of the long- sought, and still elusive human tRNA ligase. One possibility is that tRNA ligase and cyclic-phosphodiesterase (CPD) activities might be present in the same polypeptide, in resemblance to the multifunctional - kinase, CPD and ligase - yeast tRNA ligase. Using bioinformatics, we have identified putative candidates for a human CPD, which will be tested for CPD and tRNA ligase activities, as well as for their physiological roles in tRNA splicing.

The aim of the project was the analysis of the function of the RNA kinase Clp1 in cellular RNA metabolism, in particular during the processes of tRNA splicing, mRNA 3` end cleavage and transcription termination. We revealed a novel and unpredicted role that Clp1`s ATP binding motif exerts on the activity of the tRNA splicing endonuclease, TSEN. We found that HeLa cell extracts depleted of ATP as well as immunoprecipitates of the TSEN-Clp1 complex from ATP-depleted HeLa extracts show reduced pre-tRNA cleavage, which is restored by adding back ATP. In collaboration with Boehringer Ingelheim we developed a small molecule that inhibited Clp1`s RNA-kinase activity. A similar inhibition was achieved with AMP-PNP, a ß- non-hydrolyzable ATP analogue, in semi-purified TSEN-Clp1 complexes. Strikingly, affinity-purified TSEN associated with Clp1 harboring a mutation in the ATP-binding motif showed inefficient pre-tRNA cleavage. We conclude that the ATP-hydrolysis and/or RNA-kinase activity of Clp1 are crucial for efficient pre-tRNA cleavage by TSEN-complex in-vitro. A manuscript describing these results is in preparation (see page 6). In contrast to those results, the absence of kinase activity in Clp1 was found to affect neither mRNA 3` end cleavage nor RNA Polymerase II transcription termination in vitro (in collaboration with Steven West and Nick Proudfoot, University of Oxford). We have attempted to identify endogenous RNA substrates for Clp1 by setting up novel RNA purification and cloning technologies for high-throughput sequencing, as well as RNA substrate candidate approaches. Unfortunately, due to technical difficulties, we could not establish conditions for the isolation and/or verification of Clp1 RNA kinase substrates. We successfully optimized extract preparations and in vitro assay conditions to reveal deficiencies in pre-tRNA cleavage by mutations in the Clp1-associated TSEN54 subunit occurring in patients with pontocerebellar hypoplasia (PCH), a neurodegenerative disorder (in collaboration with Frank Baas, Academic Medical Center, Amsterdam). Mice with deletions in the Clp1 gene (knock-out) or encoding a mutated, RNA-kinase inactive version of Clp1 (knock-in) were developed together with Josef Penninger`s group at IMBA. Knock-in mice suffer a motorneuron defect. We have lately made progress in finding a molecular link between the mutation and the defect. A manuscript is planned (see page 6). Spinning off the original proposal, we discovered Nol9, a human nucleolar RNA-kinase with high homology to Clp1 (Heindl and Martinez, 2010; see page 6). Nol9 was found to be essential in ribosomal RNA processing. The role of Grc3, the homologous enzyme in S. cereviseae was studied in collaboration with Nick Proudfoot, and shown to be important in RNA Pol I termination (Braglia et al., 2010; see page 6).