Structural approach to the hur function

Consejo Superior de Investigaciones Científicas Universidad de Sevilla Instituto de Bioquimica Vegetal y Fotosintesis cicCartuja Structural Approach to the HuR Function Aproximación Estructural a la Función de HuR Memoria presentada por el licenciado D. Rafael Manfred Scheiba para optar al grado de Doctor en Ciencias Biológicas Sevilla, 2013 Directora Irene Díaz - Moreno, PhD Profesora Titular de Universidad Acknowledgments I thank Dr. Irene Díaz-Moreno, Dr. Antonio Díaz-Quintana and Prof. Miguel Ángel de la Rosa for accepting me in their team, for their motivation and informative discussions, as well as for offering me the possibility to carry out this work at the cicCartuja, Seville. I am grateful to Dr. Francisco J. Blanco, Dr. María Luz Martínez-Chantar, Dr. Luis Alfonso Martínez-Cruz and Dr. Adriana Rojas Cardona for their pleasant coordinated guidance and helpful discussions at the CIC bioGUNE, Bilbao. I am indebted to Dr. Manuel Angulo (CITIUS, Sevilla) and Dr. Tammo Diercks (CIC bioGUNE, Bilbao) for their professional help and advice on NMR experiments. My acknowledgments also go to Dr. M. Gorospe (NIH, Baltimore, USA) and Dr. J. A. Steitz (Yale University, New Haven, USA) for providing us the HuR vectors. I thank the laboratory teams of the Structural Biology and Metabolomics Units at CIC bioGUNE for their help, especially Dr. Alain Ibáñez de Opakua for NMR data processing, Iker Oyenarte for crystallization experiments, Dr. Nieves Embade and Dr. David Fernández for cell culture. My gratitude goes to the working group members of laboratories 1 and 5 of the “Instituto de Bioquímica Vegetal y Fotosíntesis (IBVF)” at cicCartuja. Everyone contributed its part to the work - those who stayed and those who left. It was a great pleasure for me to have all them accompanying me along the way, so helping me to see the things from many different points of view. This work has been supported by: the Andalusian Government (P07-CVI-2896), the Spanish Research Council (JAEpre08-00375) and the European Social Fonds (ESF 2007-2013). The six months stay at the CIC bioGUNE was supported by a fellowship from the Spanish Scientific Council (2011ESTCSIC–6934). Dedicated to my parents. “No book can ever be finished. While working on it we learn just enough to find it immature the moment we turn away from it.” Karl R. Popper, 1950 (The Open Society and Its Enemies, Preface to the Second Edition) 2 Figure 3.10: Structural model of HuR RRM3 WT domain built using chemical-shifts of backbone atoms as restraints in the CS23D server ........ 81 Figure 3.11: Oligomerization of HuR RRM3 ................................................. 82 Figure 3.12: AU experiments using HuR RRM3 WT and RRM3 W261A species .......................................................................................................... 83 Figure 3.13: RNA binding of HuR RRM3 with the 5´-UUUUU-3´ oligonucleotide ............................................................................................. 86 Figure 3.14: RNA binding of HuR RRM3 with the 5´-AUUUA-3´ oligonucleotide. ............................................................................................ 87 Figure 3.15: RNA binding of HuR RRM3 with the 5´-UUUUU3´ (A,B) and 5´-AUUUA-3´ (C,D) oligonucleotides ............................................................ 88 Figure 3.16: HuR RRM3 S318D: A phosphomimetic mutant ........................ 89 Figure 3.17: HuR RRM3 S318D: A phosphomimetic mutant ........................ 90 Figure 3.18: Comparison of average chemical-shift differences (Δδ avg ) between free and RNA-bound HuR RRM3 WT (black) and HuR RRM3 S318D (red) in a ratio of 4:1 for 5´-UUUUU-3´ RNA oligo ........................................ 91 Tables Table 3.1 Percentage of secondary structure for the different N-terminal constructs of HuR RRM domains and mutant species ................................. 71 Table 3.2: T m values of HuR N-terminal RRM domains and their phosphomimetic mutants, as calculated by CD and DSF ............................ 74 Table 3.3: K D values of the HuR RRM12 construct and its phosphomimetic mutants, as calculated by CD titration experiments with c-fos 11-mer RNA (5´-AUUUUUAUUUU-3´) ............................................................................... 77 Table 3.4: Percentage of secondary structure for HuR RRM3 WT and mutant species ......................................................................................................... 90 Abbreviations 7 Abbreviations 40S and 60S small and large subunits of the ribosome ADAR2 Adenosine Deaminase Acting on RNA-2 APRIL Acidic Protein Rich In Leucine AREs Adenylate and Uridylate enRiched Elements Ars2 Arsenate resistance protein 2 ATP Adenosine Tri Phosphate AU Analytical Ultracentrifugation AUF1 AU-rich binding Factor 1 AU-rich Adenylate and Uridylate enriched CCR4 Cysteine-cysteinyl Chemokine Receptor 4 Cdk5 Cyclin-dependent kinase 5 c-fos cellular fos CGRP Calcitonin/calcitonin Gene-Related Peptide Chk2 Checkpoint kinase 2 c-myc cellular myc COX CycloOXigenase CR Coding Region CRM1 Chromosomal Region Maintenance Protein 1 CSD Cold Shock Domain CspA Cold shock protein A Abbreviations 8 CUG-BP2 CUG-Binding Protein 2 DAN Deoxyribonucleoside-Activated Nucleotidase Dcp1-3 Decapping enzyme 1-3 DcpS Scavenger mRNA Decapping enzyme DEAD/DEAH RNA helicase DEAD/ DEAH box DNA DesoxyriboNucleic Acid Dnd1 Dead end protein 1 DRBD Double-stranded RNA Binding Domain dsRBD double-stranded RNA Binding Domain Drosha ribonuclease type III, drosha ds double-stranded Edc1 Enhancer of decapping enzyme 1 ELAV Embryonic Lethal Abnormal Vision FBP FUSE Binding Protein FIR FBP-Interacting Repressor FL Full Length FMR1 Fragile X Mental Retardation transcript 1 FXR Fragile X Related protein, GAP-43 Growth-Associated Protein-43 HIV1 Human Immunodeficiency Virus type I hnRNA heterogeneous nuclear RNA hnRNP heterogeneous nuclear RiboNucleoProtein Abbreviations 9 HNS HuR Nucleocytoplasmic Shuttling sequence Hrp1 Heterogeneous nuclear ribonucleoprotein 1 Hu protein Human antigen protein K A Association constant K D Dissociation constant KH K-Homology domain KSRP KH-type Splicing Regulatory Protein LB Luria Bertani LSm Sm-Like proteins m 7 GDP 7 methylGuanosine DiPhosphate m 7 GTP 7 methylGuanosine TriPhosphate MALDI Matrix Assisted Laser Desorption/Ionization Me 2+ divalent Metal cation Mg 2+ Magnesium Ion miRNA microRNA mRNA messenger RNA NFκB Nuclear Factor kappa B NF1 NeuroFibromatosis type 1 PABP PolyAdenylate-Binding Protein PAGE Poly Acrylamide Gel Electrophoresis PAIP2 Poly(A)-binding protein-Interacting Protein 2 PAN Pab1p-dependent poly(A) Nuclease Abbreviations 10 PARN Poly(A)-specific RNase PAZ Piwi/Argonaute/Zwille domain PIWI P-element Induced WImpy testis PH Pleckstrin PKC Protein Kinase C PNPase PolyNucleotide Phosphorylase poly(A) multiple Adenosine monophosphates pp32 phosphoprotein 32 pre-mRNA RNA precursors qRRM quasi-RNA Recognition Motif RBD RNA Binding Domain RBP RNA Binding Protein REV REgulator of Virion expression RGG Arg-Gly-Gly box RISC RNA Induced Silencing Complex RNA RiboNucleic Acid RNAse RiboNucleAse RNP RiboNucleoProtein complex RppH RNA pyrophosphoHydrolase RRM RNA Recognition Motif SDS Sodium Dodecyl Sulfate SF1-5 Super Family 1-5 Abbreviations 11 siRNA small interfering RNA SIRT1 stress protein SIr2-Related protein Type 1 snRNP small nuclear RiboNucleoprotein Particles SR proteins Serine/aRginine-rich proteins ss single-stranded TAR TrAnsactivation response RNA sequences TAT Trans-AcTivator TEV Tobacco Etch Virus TIA-1 T-cell-restricted Intracellular Antigen 1 TIAR TIA-1 Related protein TFIIIA 5S gene-specific Transcription initiation Factor T m Melting point TNF Tumor Necrosis Factor TRN TRaNsportin TTP Tris-TetraProline U2B'' U2 snRNP-specific B'' protein UNR Upstream of N-Ras U-rich Uracile enriched UTR UnTranslated Regions UV UltraViolett WT Wild-Type Resumen 6 Además, la proteína silvestre RRM3 parece unirse al oligonucleótido 5´-UUUUU-3´ con una afinidad mayor que al 5´-AUUUA-3´. La presencia de un residuo cargado negativamente en la posición 318 dificulta el reconocimiento por el ARN. La región comprendida entre el Trp261 y la Thr271 de HuR RRM3 alberga residuos responsables de la dimerización, tanto en presencia como en ausencia de ARN. De hecho, la sustitución del Trp por Glu en el mutante W261E RRM3 desplaza el equilibrio monómero/dímero característico de la proteina HuR RRM3 silvestre hacia la forma monomérica. Los resultados de esta tesis han dado lugar a la publicación de un artículo en una revista de alto impacto y a otro enviado: 1. Scheiba RM, Aroca A, Díaz-Moreno I (2012) HuR thermal stability is dependent on domain binding and upon phosphorylation. Eur Biophys J 41:597-605 2. Scheiba RM, Ibañez Opakua A, Oyenarte I, Díaz- Quintana A, Martínez-Chantar ML, Martínez-Cruz LA, Blanco FJ and Díaz-Moreno I (2012) Shedding light on the most C- terminal RNA binding motif of HuR: its role in RNA recognition (enviado) Summary Summary The thesis reports a structural approach to the RNA binding protein HuR, so providing relevant information on its modular assembly and phosphorylation. Our experimental data demonstrate that the N-terminal domains HuR RRM1 and RRM2 work as a functional unit, as inferred from the thermal stability analysis of single- and two-domain constructs. The phosphorylation mimic mutants S88D, S100D and S158D designed on RRM12 di-domain show negligible changes in secondary structure with respect to the WT. Thus, the T m of the S100D construct does not change with respect to WT, while S88D is stabilized and S158D is slightly destabilized upon phosphorylation mimicking. With regard to RNA binding properties, RRM12 S88D and RRM12 S100D species showed similar K D s to that of WT, whereas RRM12 S158D has a 4-fold higher binding affinity to the c-fos RNA 5´-AUUUUUAUUUU-3´. The C-terminal HuR RRM3 tumbles independently from the RRM12 unit. A challenging detergent-based strategy to get HuR RRM3 solubilized has been developed for the first time. Structural models for RRM3 WT and its phosphomimetic mutant RRM3 S318D based on NMR restraints with the CS23D server show the canonical RRM topology (β 1 α 1 β 2 β 3 α 2 β 4 ). The RNA binding site of RRM3 comprises the two central β strands (β 1 and β 3 ), although it can be eventually expanded to the whole β-sheet. Additionally, WT protein seems to bind 5´-UUUUU-3´ with a higher affinity than 5´-AUUUA-3´ based on NMR chemical-shift perturbations. Negatively-charged residue at position 318 impairs the RNA recognition. The stretch from Trp261 to Thr271 of HuR RRM3 hosts residues responsible for protein dimerization even in absence of RNA. In fact, the Trp-by-Glu substitution shifts the monomer/dimer Summary 12 equilibrium characteristic of HuR RRM3 WT towards the monomeric form of HuR RRM3 W261E mutant. The thesis resulted in one publication in a high impact journal and another one submitted: 1. Scheiba RM, Aroca A, Díaz-Moreno I (2012) HuR thermal stability is dependent on domain binding and upon phosphorylation. Eur Biophys J 41:597-605 2. Scheiba RM, Ibañez Opakua A, Oyenarte I, Díaz- Quintana A, Martínez-Chantar ML, Martínez-Cruz LA, Blanco FJ and Díaz-Moreno I (2012) Shedding light on the most C- terminal RNA binding motif of HuR: its role in RNA recognition (submitted) Introduction -1.2 mRNA Stabilization and mRNA Decay 32 interact with the exosome (Chen et al., 2001), whereas distinct reactions for different RBPs were observed. Binding of AREs to KH-type Splicing Regulatory Protein (KSRP) and Tris-TetraProline (TTP) results in a rapid degradation of mRNA, whereas HuR and AU-rich binding Factor 1 (AUF1) have stabilizing effects and prevent degradation (Chen et al., 2001). These examples demonstrate the multiple ways in how far AREs and RBPs regulate stabilization and degradation of mRNAs and expression levels of these genes. There exist two publicly available web server databases for investigation of AREs under the directions: http://rna.tbi.univie.ac.at/AREsite. (Gruber et al., 2010) http://brp.kfshrc.edu.sa/ARED/ (Bakheet et al., 2001 and 2003) 1.2.2 RNA Degradation RNA degradation depends on trans-regulatory factors (proteins) and cis-regulatory elements (mRNA sequences) and occurs in three different ways: 3´ exonucleases degrade the transcripts beginning from 3´ site, 5´ exonucleases hydrolyze the RNAs beginning from the 5´ site, whereas endonucleases can cleave RNA molecules internally (Garneau et al., 2007; Houseley and Tollervey, 2009). Usually, the Pab1p-dependent Poly(A) Nuclease (PAN2-PAN3) complex performs a 3´ deadenylation of mRNA at the beginning of degradation. This shortens the polyadenylic tail to a length of about 80 nucleotides (Brown et al., 1996; Yamashita et al., 2005). Additionally, a deadenylation-independent mRNA decay occurs in the cell. Thus, the proteins Rps28 (Badis et al., 2004) and the Enhancer of decapping protein (Edc1; Muhlrad and Parker, 2005) decap and degrade RNAs without previous deadenylation. In the deadenylation-dependent pathway the PAN2-PAN3 complex and the Cysteine-cysteinyl Chemokine Receptor 4-NOT (CCR4-NOT; Tucker et al., 2001; Yamashita et al., 2005) are involved in decapping. Another important protein taking part in deadenylation is the Poly(A)-specific RNase (PARN, previously called Deoxyribonucleoside- Activated Nucleotidase, DAN), as shown by experiments with Xenopus Introduction - 1.2 mRNA Stabilization and mRNA Decay 33 laevis oocytes (Korner et al., 1998) and other Eukaryotes (Dehlin et al., 2000; Gao et al., 2000; Chiba et al., 2004; Milone et al., 2004; Reverdatto et al., 2004; Opyrchal et al., 2005). 1.2.2.1 RNA Degradation in 5´ to 3´ Direction mRNA degradation by exonucleases acting in 5´ to 3´ direction considered the major mRNA degradation pathway (Chen et al., 2001). This type of degradation depends on the 7 methylguanosine triphosphate (m 7 GTP)-cap at the 5´ ends of mRNAs. The cap protects the nucleic acids from 5´ to 3´ exonucleases. Additionally, the Decapping enzyme (Dcp1 and Dcp2) complex is involved in this process (Beelmann et al., 1996; Lykke- Andersen, 2002; Steiger et al., 2003; van Dijk et al., 2002; Wang et al., 2002) and other factors are needed, as Sm-like proteins (LSm1-7 complex; Tharun et al., 2000 and 2001), LSm14 or RAP55 (Yang et al., 2006) and the enhancer of decapping enzyme 3 - Edc3p or Lsm16 (Kshirsagar and Parker, 2006). When decapped, the RNA is fragmented into two RNA stretches: one ending with a 5´ monophosphate and the other with the m 7 GDP rest. The monophosphate does not protect the RNA from exonuclease degradation anymore, so the transcripts are continuously degraded by the exoribonuclease Xrn1 (Stevens and Maupin, 1986). In bacteria, the process is similar but the 5´ end is modified by a triphosphate rest instead of the 5´ cap. After catalysis to a monophosphate by the RNA pyrophosphoHydrolase (RppH; Deana et al., 2008), the degradation can take place. 1.2.2.2 RNA Degradation in 3´ to 5´ Direction – the Exosome The alternative mRNA degradation pathway in 3´ to 5´ direction is carried out by the exosome, consisting of 9 to 11 proteins acting as a 3´ to 5´ exonuclease. The complex varies in numbers of subunits, depending on the organism and localization of the complex (Liu et al., 2006). The human exosome is highly similar to the yeast exosome (Chen et al., 2001), but differs in enzymatic activites of some complexes (Liu et al., 2006) and has also homolog proteins in bacteria, which have Pleckstrin Homology (PH) and PolyNucleotide Phosphorylase (PNPase) domains (Symmons et al., 2002; Lorentzen et al., 2005; Liu et al., 2006). In bacteria, the complex is Introduction -1.2 mRNA Stabilization and mRNA Decay 34 called degradosome, but the functions are the same: degradation of mRNA in 3´ to 5´ direction. Before degradation occurs by the exosome, the Trf4-Air1/Air2-Mtr4 polyadenylation complex (TRAMP; Lacava et al., 2005) marks the RNA targets with a poly(A) tail. When the RNAs are degraded, the remaining methylated cap is hydrolyzed by the scavenger mRNA decapping enzyme (DcpS; Liu et al., 2002). 1.2.3 RNA Stabilization Although the most frequently described consequence of RBP binding to AREs is the degradation of the mRNA targets, some RBPs have a stabilizing function upon binding to AREs. Besides the extensively studied mRNA stabilizing Hu proteins (see chapter 1.3 - Hu proteins), only few more proteins are known to stabilize mRNAs, such as CUG-Binding Protein 2 (CUG-BP2), Nucleolin, TINO, Poly(A)-binding protein-Interacting Protein 2 (PAIP2) and AUF1, which has both stabilizing and destabilizing functions (Barreau et al., 2005). The well studied Hu family member HuR binds to more than 50 RNA targets related to disease (Srikantan and Gorospe, 2012). The mayority of these RNAs are stabilized. Thereby, degradation processes (see chapter 1.2.2 – RNA degradation) as well as interaction of mRNA with miRNAs or other RBPs are prevented. 1.3 RNA Binding Proteins From the beginning of transcription, RNAs are always accompanied by RNA Binding Proteins (RBPs). RBPs are constituted by different types of small (30 to 100 amino acids in length) RNA Binding Domains (RBDs), which bind specifically to nucleic acid sequences and have diverse functions in the processing of the RNAs. In this context, the nucleic acids are described as the cis-regulatory elements and proteins as trans-regulatory factors. Dreyfuss and coworkers (Dreyfuss et al., 2002) reviewed the various functions in which RBPs are involved, such as transcriptional regulation (Tomonaga et al., 1995; Michelotti et al., 1996; Du et al., 1998; Miau et al., 1998), telomere length maintenance (Ishikawa et al., 1993; La Branche et al, 1998; Eversole et al., 2000; Fiset et al., 2001), immunoglobulin gene recombination (Dempsey et al., 1999), splicing (van der Houven van Oordt Introduction – 1.3 RNA Binding Proteins 35 et al., 2000; Mayeda et al., 1992; Caceres et al., 1994, Yang et al., 1994; Min et al., 1995; Chan et al., 1997; Chabot et al., 1997; Ashiya et al., 1997; Chou et al., 1999; Del Gatto-Konczak et al., 1999; Chen et al., 1999; Mourelatos et al., 2001) pre-ribosomal RNA processing (Russel et al., 1992) and 3’ end processing (Kessler et al., 1997; Moreira et al., 1998; Minvielle-Sebastia et al., 1998, Bagga et al., 1998). In the same review (Dreyfuss et al., 2002) is also reported that, besides these nuclear actions, RBPs have also functions in the cytosol and are involved in mRNA transport between nucleus and cytoplasm (Lee et al., 1996; Liu et al., 1995; Izzauralde et al, 1997; Gallouzi et al., 2001a). Further, RBPs play roles in mRNA localization (Hoek et al., 1998; Matunis et al., 1994; Cote et al., 1999; Mouland et al., 2001; Carson et al., 2001) translation (Habelhah et al., 2001; Ostareck et al., 1997; Collier et al., 1998; Silvera et al., 1999; Ostareck et al., 2001) and mRNA stability (Kiledjian et al., 1995 and 1997, Rajagopalan et al., 1998; Chkhheidze et al., 1998; Shih et al., 1999; Loflin et al., 1999; Xu et al., 2001). Different RBPs are able to interact specifically with either single-stranded RNA (ssRNA, as for example the RRM) or double-stranded RNA (dsRNA, as the doublestranded RBD), as also with small interfering RNA (siRNA, as PIWI and PAZ domains) and even bind to DNA (as zinc finger domains). The modular assembly and auxiliary domains in combination with linkers of different lengths determine the binding mechanisms and functions of the protein (for reviews see Lunde et al., 2007; Glisovic et al., 2008). Combination of RBDs results in a big variety of RNPs with different RNA sequence specificities (Lunde et al., 2007). Furthermore, the length of a linker between RBDs is important for the protein-RNA interaction. A long linker (about 50-60 residues; Lunde et al., 2007) allows the RBDs to interact with RNA targets, which are separated on the same RNA strand. Also, it allows binding to various differently structured RNA molecules, as it is the case for the Adenosine Deaminase Acting on RNA-2 protein (ADAR2) that can bind dsRNA on differently structured transcripts (Stefl et al., 2006). A short linker between RBDs, on the other hand, yields a high specificity to distinct RNA sequences, since the RBDs interact with RNA stretches with close proximity to each other. This is the case for the HuD protein (see chapter 1.4 – Hu proteins), whose N-terminal RRMs - RRM1 and RRM2 - work as a tandem domain, with a 3 10 -helix linker (Wang and Tanaka Hall, 2001). Introduction -1.3 RNA Binding Proteins 36 Figure 1. 2: RBPs and their composition in RBDs . The size of the RBPs can be estimated by the scale of amino acids at the bottom. Abbreviations: hnRNP: heterogeneous nuclear RiboNucleoProtein, PABP: PolyAdenylate-Binding Protein, FMR1: Fragile X Mental Retardation transcript 1, FXR: Fragile X Related protein, Ars2: Arsenate resistance protein 2, ADAR2: Adenosine Deaminase Acting on RNA-2, RRM: RNA Recognition Motif, RGG: Arg- Gly-Gly box, KH: K-Homology domain, ZnF: Zinc Finger, dsRBD: double stranded RNA-Binding Domain, DEAD/DEAH: RNA helicase DEAD/ DEAH box, PAZ: Piwi/Argonaute/Zwille domain. The figure is adapted from Glisovic et al., 2008. The length of linkers differ from four amino acids (Bj6 protein) to 122 amino acids (snRNP U1A protein) and can have a strong influence on the RNA binding characteristics, affecting sequences and affinities (Shamoo et al., 1995; Finger et al., 2004; Lunde et al., 2007). When binding to RNA, a flexible disordered linker can change to an ordered structure, as reported for the Heterogeneous nuclear ribonucleoprotein 1 (Hrp1; Perez-Cañadillas, 2006). A short linker increases the affinity of RNA binding by additional RNA contacts with the linker. On the other hand, long flexible linkers are predicted to be less involved in RNA binding (Lunde et al., 2007). For example, the 82 residue linker of nucleolin does not contact RNA. Thus, RNA binding is restricted to the RBDs (Stefl et al., 2006). In addition, protein-protein-interactions in RBPs occur frequently. On one hand, RNA binding proteins can dimerize, such as reported for HuR Introduction -1.3 RNA Binding Proteins 37 (Benoit et al., 2010) and the p19 protein (Vargason et al., 2003; Ye et al., 2003). This dimer formation modulates the protein-RNA interface, resulting in different binding characteristics of the protein. On the other hand, protein-protein interactions are involved in protein shuttling between compartments. For instance, HuR protein has a 60 amino acid HuR Nucleocytoplasmic Shuttling sequence (HNS) (Fan and Steitz, 1998a) between the two C-terminal RRMs. This region is responsible for the shuttling of the protein from the nucleus to the cytoplasm and vice versa. This mechanism is coupled to protein binding of transportins (TRN) 1 and 2 (Gallouzi et al., 2001a; Guttinger et al., 2004; Rebane et al., 2004) or in association with pp32 and the Acidic Protein Rich In Leucine (APRIL) by the Chromosomal Region Maintenance Protein 1 (CRM1; Gallouzi et al., 2001b; Brennan et al., 2000). Other proteins known to interact with HuR are SETalpha, SETbeta and Importin α (Wang et al., 2004). Further examples for binding of RBPs to different kind of proteins are well reported by Lunde et al., 2007 with important outcomes in mRNA fate. RBPs have been extensively explored at the structural level with the pioneer techniques x-ray crystallography and Nuclear Magnetic Resonance (NMR), although slightly more crystallographic structures (26 in total; Daubner et al., 2013) have been published in comparison with NMR solution structures (23 in total). Despite the fact that the atomic resolution of crystallographic structures gives more structural details about RNA bound protein, NMR is able to give information about dynamics and binding affinities that makes it an advantageous tool for analysis of RBPs (Cukier and Ramos, 2011). On the following, the most important RBDs are presented (for reviews see Dreyfuss et al., 2002; Lunde et al., 2007; Glisovic et al., 2008, Elliott and Ladomery, 2011) 1.3.1 The RRM Domain The RNA Recognition Motif (RRM, also known as RBD or RNP; for reviews see Maris et al., 2005, Clery et al., 2008; Muto et al., 2012) is the most common RBD and the best characterized one. It is found in viruses, bacteria and eukaryotes (Maris et al., 2005). About 1-2% of human genes Introduction -1.3 RNA Binding Proteins 38 contain sequences coding for RRMs, often in multiple copies in the same gene (Venter et al., 2001; Maris et al., 2005). The domain is built up by about 90 amino acids, forming the secondary topology of β 1 α 1 β 2 β 3 α 2 β 4 (Nagai et al., 1995). In most RRMs, RNA binding takes place at the highly conserved RiboNucleoProtein1 (RNP1; Adam, 1986; Swanson et al., 1987; Oubridge et al., 1994) and RNP2 motifs (Dreyfuss et al., 1988) at the inner β 1 and β 3 strands of the protein, which are constituted by 8 amino acids for the RNP1 motif (Lys/Arg-Gly-Phe/Tyr-Gly/Ala-Phe/Tyr-Val/Ile/Leu-X-Phe/Tyr) and 6 amino acids for the RNP2 (Ile/Val/Leu-Phe/Tyr-Ile/Val/Leu-X-Asn-Leu) motif. Four residues of these RNPs interact with RNA at distinct positions (Birney et al., 1993; Maris et al., 2005). Position 1 of RNP1 contacts the RNA phosphodiester backbone by a salt bridge, while stacking interactions between aromatic residues and RNA bases are found at position 3 and 5 of RNP1 and position 2 of RNP2. However, not always all the four contacts contribute together to RNA binding (Ding et al., 1999; Maris et al., 2005). An isolated RRM is able to recognize RNA stretches ranging from two (RRM2 of nucleolin; Allain et al., 2000; Johansson et al., 2004) to eight RNA bases (RRM1 of U2 snRNP-specific B protein, U2B''; Price et al., 1998). Interestingly, some exceptional cases of RNA binding by RRMs have been described (Muto and Yokoyama, 2012). Instead of binding to RNA at the inner β 1 and β 3 strands, some RRMs bind RNA in a non-classical manner, such as the quasi-RNA recognition motif (qRRMs; Dominguez et al., 2010) of hnRNP F, xRRM of p65 (Singh et al., 2013) or the bacterial RNA helicase YxiN (Wang et al., 2006). For these proteins also α-helices and loops are implicated in the RNA binding. The RRM domain is involved in many posttranscriptional processes, as pre-mRNA processing, splicing, mRNA stability, RNA editing, RNA export and pre-mRNA complex formation, translation and degradation. In addition to RNA association, RRM domains are also known to interact with other proteins (Gallouzi et al., 2001a; Maris et al., 2005; Cho et al., 2010). Wellstudied examples of RRM binding proteins constituted by RRM domains are the proteins hnRNP A1 (Ding et al., 1999), Hu proteins (Hinman and Lou, 2008) or Sex-lethal protein (Handa et al., 1999). Introduction -1.3 RNA Binding Proteins 39 1.3.2 The K-homology Domain The K-Homology (KH) domain binds to both single-stranded RNA (ssRNA) and ssDNA (Backe et al., 2005; Beuth et al., 2005; Braddock et al., 2002a and b; Du et al., 2005; Lewis et al., 2000; Liu et al., 2001). It is ubiquitous in Eukaryotes, Eubacteria and Archaea (Siomi et al., 1993). The typical KH domain consists of 70 amino acids and forms two different types of topologies: Type 1 has a β 1 α 1 α 2 β 2 β 3 α 3 constitution, whereas Type 2 is arranged according to the α 1 β 1 β 2 α 2 α 3 β 3 topology. This domain is able to bind four nucleotides in a cleft formed by the GXXG loop joining the neighboring helices, the strands β 2 (Type 1) or β 3 (Type 2) and the variable loop between β 2 β 3 (Type 1) and β 2 α 2 (Type 2). The interaction between nucleic acids and protein is performed by hydrogen bonds and electrostatic interactions. 1.3.3 The Double-Stranded RBD Domain The double-stranded RNA Binding Domain (dsRBD or DRBD) is constituted by 70 to 90 amino acids and has the topology α 1 β 1 β 2 β 3 α 2 . In contrast to other RBDs, this motif binds to double-stranded RNA (dsRNA). The interaction is not sequence dependent and involves two successive minor RNA grooves and the intervening major groove of the dsRNA helix. Thus, binding of dsRBDs to dsRNAs depends on the structure of the nucleic acids. Binding is performed by hydrogen bonds between the 2´-OH groups of the RNA phosphates and the protein backbone (Ryter and Schultz, 1998). 1.3.4 Zinc Finger Domains The Zinc Finger (ZnF) domains are known as classical DNA binding motifs, but they are also able to in interact with RNAs (Carballo et al., 1998; Picard et al., 1979). These domains consist of ca. 30 residues with a β 1 β 2 α 1 topology. The different kinds of metal coordination result in three types of ZnFs: Cys Cys His His (CCHH), CCCH and CCHC arrangements. ZnF proteins often comprise several ZnF domains, as shown for the 5 S gene-specific Transcription initiation Factor (TFIIIA; Clemens et al., 1993; Searles et al., 2000), which is constituted by nine ZnF domains with different RNA or DNA binding preferences. The binding mechanism depends on the structure of Introduction -1.3 RNA Binding Proteins 40 the nucleic acid. For instance, TFIIIA ZnFs 4 and 6 are associated with two RNA loops by the α-helix of the protein (Clemens et al., 1993). Zif268 is a typical example for a protein with CCHH domains and its DNA binding is known from the crystal structure (Pavletich et al., 1991; Wolfe et al., 2000). Here the major groove of DNA is involved in binding to an α-helix by hydrogen bonds. TIS11d is an example of a CCCH-type ZnF protein (Hudson et al., 2004) which binds to Adenine and Uridine Rich Elements (AREs) in 3´ UnTranslated Regions (UTRs) of RNA molecules. The interaction between RNA and TIS11d is based on hydrogen bonds. An example for the CCHC-type of ZnF domains is the Human Immunodeficiency Virus type I (HIV-1) nucleocapsid protein (De Guzman et al., 1998; D´Souza et al., 2004), which follows the same mechanism for RNA binding as the CCCH-type ZnF proteins. 1.3.5 Cold Shock Domains Cold Shock Domains (CSD) were primarily identified in bacteria, for example the Cold shock protein A (CspA) from E. coli, which is expressed in upon temperature changes from 37 ˚C to 10 ˚C (Jones et al., 1987). From the structural point of view, this small domain is built up by 70 amino acids arranged exclusively in five β strands forming two β-sheets. The proteinnucleic acid binding mechanism is similar to that of RRM domains: aromatic residues form stacking contacts with RNA bases and positively charged amino acids maintain contacts with phosphates from RNA and DNA. CSDs are also found in eukaryotes, such as the “Upstream of N-Ras” (UNR) protein (Chang et al., 2004) and the Y-box protein (Sommerville and Ladomery, 1996a and b; Matsumoto and Wolffe, 1998). UNR has a repetition of five CSDs, the Y-box protein contains one CSD and is accompanied by additional sequences. These domains have additional functions related to RNA binding, multimerization and phosphorylation, making the eukaryotic proteins behave as multifunctional proteins. On the contrary, bacterial CSDs are less specialized with a main function as chaperones, synchronizing transcription and translation events. Introduction -1.3 RNA Binding Proteins 41 1.3.6 PIWI Domain P-element Induced Wimpy testis (PIWI) domains are involved in the processing of mRNAs and small interfering RNAs (siRNAs). PIWI topology consists of a 5-stranded β-sheet with α−helices at the periphery. The PIWI domain appears in the Argonaute protein and has an endonuclease cleavage function through binding to the 5’ end of RNA (Ma et al., 2005; Yuan et al., 2005; Song et al., 2004; Parker et al., 2004 and 2005). Argonaute protein (Song et al., 2004), along with Dicer (Macrae et al., 2006), is involved in RNA degradation and RNA silencing pathways by forming an RNA Induced Silencing Complex (RISC; Hammond et al., 2000; Bartel et al., 2004). 1.3.7 PAZ Domain PIWI Argonaute and Zwille (PAZ) domains bind non-coding micro RNAs (miRNAs) and small interfering RNAs (siRNAs). The PAZ domain comprises 110 amino acids with an α 1 β 1 arrangement and siRNAs bind to a hydrophobic pocket by 2 nucleotide overhangs. Stacking interactions and hydrogen bonds stabilize the complex (Ma et al., 2004). Examples for proteins containing PAZ domains are Argonaute (Song et al., 2004) and Dicer proteins (Macrae et al., 2006) which are involved in RNA silencing pathways like the PIWI domains by forming RISC that degrades RNAs by interaction with siRNAs (Hammond et al., 2000; Bartel et al., 2004). 1.3.8 Arginine Rich Domains Arginine Rich Domains are known as gene expression regulators. They were first discovered in bacteria and viruses, such as the TAT (Trans- AcTivator) domain in Human Immunodeficiency Virus type I (HIV-1; Garcia et al., 1988; Frankel et al., 1989; Subramanian et al., 1991) and the REgulator of Virion expression (REV) domain from the same virus (Heaphy et al., 1990). TAT has a conserved 14 amino acid sequence, rich in arginine and lysine residues, which is involved in binding to the TrAnsactivation Response RNA sequences (TAR) and regulates the gene expression (Dingwall et al., 1989; Weeks et al., 1990). Actually, TAT binds specifically to the stem-loop structured TAR RNA by a single arginine (Calnan et al., 1991). Positively charged arginine residues also appear in Arg-Gly-Gly (RGG) boxes Introduction – 1.5 HuR 48 1.5.3 HuR Structure HuR belongs to the family of Hu proteins (HuR, HuB, HuC and HuD) and shows the highest sequence identity in relation to the homolog ELAV protein (Okano et al., 1997). All Hu proteins are constituted by three RRMs. RRM1 and RRM2 are separated by a 3 10 helix, whereas the third RRM domain is spaced out by a 60 residue spanning linker, containing the HuR Nucleocytoplasmic Shuttling sequence (HNS; Fan & Steitz, 1998a). The reader is referred to chapter 1.3 – Hu proteins and chapter 1.3.1 – The RRM Domain. HuR and its homologs have the ability to multimerize (Toba and White, 2008, Benoit et al., 2010, David et al., 2007) and additionally to form complexes with other RBPs (David et al., 2007). On one hand, HuR builds homodimers by disulfide bonds at Cys13 residue (Benoit et al., 2010), on the other hand, HuR RRM3 is responsible for protein multimerization (Meisner et al., 2007; Toba and White, 2008) and protein-protein interactions. Additionally, HuR RRM3 has an Adenosyl Transferase activity along with binding sites for Me 2+ and ATP (Meisner et al., 2009). Crystallographic structures are available for HuC (RRM1: 1D8Z.pdb, RRM2: 1D9A.pdb, RRM12: 1FNX.pdb; Inoue et al., 2000) and HuD (RRM12: 1FXL.pdb; Wang and Tanaka Hall, 2001). Recently the structures of HuR (RRM1: 3HI9.pdb; Benoit et al., 2010; RRM12: 4ED5.pdb, 4EGL.pdb; Wang et al., 2013) have been published, demonstrating the binding of RNA at the molecular level of the N–terminal RRM domains RRM1 and RRM2. Figure 1.5 shows the Richardson presentation of such structures. HuR is a well-characterized protein in vivo, but little information is available at the structural level, except for the crystal structures mentioned above. Although the structures of individual and tandem RRM domains have been resolved, there is no Full Length (FL) structure reported for a protein from the Hu family. A previously crystallization report of HuR FL was published (Iyaguchy et al., 2009) without further results. In particular, the structure and assembly of RRM3 in relation to RRM12 remain to be elucidated. Due to the biophysical characteristics and the low solubility of RRM3 domain, it explains the relevance of this thesis work. Introduction – 1.5 HuR 49 Figure 1.5 : Crystallographic structures of Hu proteins. A ) HuR RRM1 (3HI9 .pdb), left monomer, right tetramer B) HuR RRM12, bound to RNA (left, 4ED5.pdb), and free (right, 4EGL.pdb) C) HuC RRM12, bound to RNA (1FXN.pdb) D) HuD RRM12 bound to RNA (1FXL.pdb). Objectives 53 Objectives This thesis work reports a biophysical characterization of the RBP HuR so as to explore its main structural features. HuR is well-characterized in vivo, but working mechanisms at the molecular level are still unknown. As the protein is involved in regulatory mechanisms of human pathologies, such as inflammation, cardiovascular diseases and cancer, studying in depth its biophysical and structural properties will help to understand and treat these diseases in a long term view. The first objective is to characterize biophysically the three RRMs of HuR with respect to its modular assembly. It is proposed that each RRM has certain physical and functional properties, such as specific thermal or chemical stabilities. In this context, it is of interest how differently HuR domains behave and whether they cooperate with each other. Second aim of this work is to perform biophysical characterization of isolated HuR RRM3 in solution by means of improving its solubility to further explore its structural features.The HuR C-terminal RRM3 domain has additional functions in comparison to the N-terminal RRM12 module, since it is known to form multimers and actively contributes to RNA modification by its adenosyl transferase activity. This domain has a low solubility due to its particular biophysical properties. Thus, a challenging protocol is necessary to increase RRM3 solubility and, further, to improve the purification of this domain. Ongoing analysis of RNA binding, interactions with the N-terminal RRM12 and multimerization events are of interest to elucidate molecular working mechanisms of HuR. The third aim of this thesis is to investigate the effects of phosphorylation at the molecular level and its functional consequence by the design of phosphorylation mimic mutants. HuR is tightly regulated by diverse posttranslational modifications at several residues. The effects of phosphorylation of residues Ser88 (RRM1), Ser100 (in between RRM1 and RRM2), Ser158 (RRM2) and Ser318 (RRM3) are known at the cellular level. These phosphorylations alter mRNA binding and localization of the protein Objectives 54 in vivo. In this work, biophysical properties of RRM12 and RRM3 species have been explored through a comparison between the phospho-mimic species and the wild-type proteins. In vitro experiments get new insights into how Ser phosphorylation modulates the RNA binding properties of HuR. Materials and Methods Materials and Methods - 2.7 Nuclear Magnetic Resonance 64 further automatically assigned by MARS server (Jung and Zweckstetter, 2004). Structural models of RRM3 WT and RRM3 S318D were created using the backbone assignment already deposited in BMRB database (entry codes are going to be published), along with the amino acid sequences, as input files for the CS23D server (Wishart et al., 2008). RNA binding of HuR RRM3 WT and RRM3 S318D was monitored by acquiring 15 N-HSQC spectra at a 500 MHz Bruker spectrometer provided with cryo-probe, along titrations of the 5´-AUUUA-3´ and 5´-UUUUU-3´ RNA oligonucleotides (IDT, Integrated DNA Technologies) into a sample of 50 μM 15 N-labeled RRM3 in 10 mM sodium phosphate buffer (pH 7.3), 3 mM DTT at RNA:protein ratios of 2:1 and 4:1. The pH value of the sample was verified after each titration step. Weighted average values of 15 N and 1 H chemical-shift perturbations (∆δ avg ) of each resonance was calculated as follows: ∆δ avg = (([∆δ H ] 2 + [∆δ N /5] 2 ) / 2) 1/2 , being ∆δ H and ∆δ N chemical-shift increments of 1 H and 15 N, respectively. Results 69 Figure. 3.1: Schematical view o f HuR. Results 3.1 HuR RRM12 Domain Adopts a Canonical Topology with Negligible Changes in its Secondary Structure upon Phosphorylation The crystallographic structure of HuR RRM1 – recently published by Benoit et al. (2010) – shows that the RRM folding adopts the canonical β 1 α 1 β 2 β 3 α 2 β 4 topology. As shown in Figure 3.1, HuR protein consists of three RRMs. A homology model of HuR RRM12 construct was obtained using the crystallographic structure HuD RRM12 as a template (PDB entry 1FXL; Wang and Tanaka Hall, 2001). Sequence identity to the target was 75.4%, and the model represented in Figure 3.2 was built with the SWISS-MODEL server (Arnold et al., 2006; Kiefer et al., 2009; Peitsch, 1995) and graphically represented using Chimera software (Pettersen et al., 2004). Figure 3.2 shows the superposition of both HuR structures: the homology model of RRM12 and the crystallographic structure of RRM1. The homology based model is in good agreement with the secondary structure contents for HuR constructs. Figure 3.3 shows the normalized far-UV CD spectra of isolated RRM1 and RRM2 domains, the tandem RRM12, and the HuR-FL protein. Notably, all HuR species show similar global secondary structures with minor differences, as summarized in Table 3.1. Whereas all constructs share similar β strand and turn contents, RRM2 differs from RRM1 and RRM12 in its higher α-helix content. Results - 3.1 HuR RRM12 Domain Adopts a Canonical Topology with Negligible Changes in its Secondary Structure upon Phosphorylation 70 Figure 3.3 : Far - UV (190 – 250 nm) CD spectra of different HuR N - terminal domain constructs . RRM domains are represented as follows: RRM1 by solid line, RRM2 by dashed line, RRM12 by dotted line and HuR-FL protein by dash-dotted line. Figure. 3.2: HuR RRM12. Superposition of the crystal structure of HuR RRM1 (PDB entry 3HI9; Benoit et al., 2010) and the homology model of HuR RRM12 built as described in ‘‘Materials and Methods.’’ The RMSD for backbone atoms of HuR RRM1 domain in both structures is 0.583 Å. Side-chains of serine residues to be phosphorylated are included. Quality information: The QMEAN Z-Score is 1.33. Results - 3.1 HuR RRM12 Domain Adopts a Canonical Topology with Negligible Changes in its Secondary Structure upon Phosphorylation 71 RRM12 mutants, in which Ser88, Ser100 and Ser158 have been substituted by aspartic acid residues to mimic phosphorylation events, exhibit a secondary structure similar to that of RRM12 wild-type (RRM12 WT). In addition, Ser-by-Ala control mutations show similar CD spectra (Figure 3.4; Table 3.1). Constructs α-helix (%) β-strand (%) Turn (%) Unstructured (%)* RRM1 6.01 ± 0.57 36.21 ± 1.25 19.49 ± 2.05 37.74 ± 3.95 RRM2 10.84 ± 0.22 34.11 ± 0.65 19.13 ± 0.90 30.81 ± 1.61 RRM12 WT 5.72 ± 0.77 39.67 ± 4.39 21.22 ± 1.80 33.03 ± 3.10 RRM12 S88D 11.03 ± 0.56 33.26 ± 2.05 19.64 ± 2.06 35.76 ± 4.75 RRM12 S88A 5.87 ± 0.44 39.00 ± 1.68 20.42 ± 1.41 34.25 ± 2.40 RRM12 S100D 5.34 ± 1.25 42.71 ± 4.30 19.25 ± 2.01 32.16 ± 3.81 RRM12 S100A 3.18 ± 0.39 40.88 ± 1.34 21.23 ± 1.15 34.78 ± 2.90 RRM12 S158D 5.14 ± 0.50 40.28 ± 1.58 20.51 ± 1.12 33.81 ± 3.17 RRM12 S158A 5.16 ± 0.34 39.79 ± 1.40 21.04 ± 1.65 33.75 ± 3.29 HuR FL 9.18 ± 1.79 34.95 ± 0.83 20.01 ± 2.39 35.23 ± 4.74 * This section refers to both disordered and flexible and ordered but nonregular structured parts of the protein. For comparison of thermal stability data all proteins have to be in a monomeric state. Cys13 promotes a homodimer formation for the RRM12 construct (Meisner et al., 2007; Benoit et al., 2010). For further investigation of the thermal stability of RRM12 WT and its mutants, the impact of Cys13 on the homodimer formation needs to be evaluated. Figure 3.5 shows a Sodium Dodecyl Sulfate (SDS) polyacrylamide gel Table 3.1 Percentage of secondary structure for the different N - terminal constructs of HuR RRM domains and mutant species. Results - 3.1 HuR RRM12 Domain Adopts a Canonical Topology with Negligible Changes in its Secondary Structure upon Phosphorylation 72 Figure 3.4: Far-UV (190–250 nm) CD spectra of RRM12 WT and its phosphomimetic mutants. RRM12 WT is shown in blue solid line, RRM12 S88A in green solid line, RRM12 S88D in green dashed line, RRM12 S100A in black solid line, RRM12 S100D in black dashed line, RRM12 S158A in red solid line and RRM12 S158D in red dashed line. electrophoresis (PAGE) of RRM12 WT in absence and presence of dithiothreitol (DTT) at 0.5 and 5 mM, as reducing agent. RRM12 WT is clearly a monomer upon DTT addition, although a monomer–dimer equilibrium appears in samples devoid of DTT. These data were recently confirmed by AU experiments on RRM12 WT samples containing 0.5 mM DTT (data not shown). Thus, RRM12 WT construct, which includes Cys13, behaves as a monomer, at least in the experimental conditions used in this work. Results - 3.1 HuR RRM12 Domain Adopts a Canonical Topology with Negligible Changes in its Secondary Structure upon Phosphorylation 73 3.2 RRM12 Works as a Functional Unit Recently, it has been demonstrated that the thermal stability of RBDs reveals interactions between neighboring modules (Aroca et al., 2011; Díaz-Moreno et al., 2010). Thermal unfolding studies on the single N-terminal RRM segments and the two-domain construct from HuR were performed to confirm the assembly between RRM1 and RRM2, as inferred from the homology model of HuR RRM12 and the crystal structure of HuD RRM12 (Wang and Tanaka Hall, 2001). CD spectroscopy shows that the T m for isolated RRM1 (335 ± 3 K) is slightly lower than that for RRM2 (341 ± 2 K; Table 3.2). Interestingly, RRM12 is as stable as RRM1 (335 ± 2 K), suggesting that interdomain interactions are taking place. Such interaction lowers the T m of RRM2 by ca. 6 K, as previously reported for other RBPs (Aroca et al., 2011; Díaz-Moreno et al., 2010). Figure 3.5 : PAGE electrophoresis of HuR RRM12 WT. Line 1 HuR RRM12 WT sample devoid of DTT; lines 2 and 3 protein samples previously incubated with 5 and 0.5 mM DTT, respectively, for 90 min before loading into the gel. In each line, 4 µg HuR RRM12 WT was loaded onto an 18% SDS-PAGE gel. M: Pro-stain protein molecular weight marker (Intron Technologies Inc.). Results - 3.5 Shedding Light on the HuR C-terminal RRM3 Domain 80 3.5.2 NMR Restraint-Based Model of RRM3 We have established a protocol for expression and purification of HuR RRM3 WT and RRM3 S318D. For this purpose we have tested a set of buffer conditions with different amounts of detergents and salts (data not shown). Best results were achieved using the anionic detergent sarkosyl to increase HuR RRM3 solubility. After removing detergent and salt from the protein preparations, concentrations of no more than 200 µM were reached. Due to this low solubility, we could record 3D NMR spectra based on the 13 C resonance to assign only backbone atoms, but not those experiments with which obtain Nuclear Overhauser Effect (NOE) signals and distance information to be further implemented in structure calculations. However, automatic assignment worked well using MARS server and most of resonances appearing at the 15 N HSQC spectrum of RRM3 WT were successfully assigned (Figure 3.9). Actually, non-labeled resonances mainly correspond to signals from the TEV cleavage site located between the 6×His-tag and the protein sequence. Using NMR chemical-shifts and protein sequences as restraints, both structural models of RRM3 WT and RRM3 S318D were built up with the CS23D server (see Figure 3.10; Wishart et al., 2008). Figure 3 .9: HuR RRM3 . 15 N - HSQC spectrum of HuR - RRM3 WT along with the assignment . Labels stand for the amino acids of RRM3 following the numbers in the HuR -FL protein. Results - 3.5 Shedding Light on the HuR C-terminal RRM3 Domain 81 Both models have the conserved structure of two α-helices packed against four anti-parallel β strands with the canonical β 1 α 1 β 2 β 3 α 2 β 4 topology characteristic of RRM motifs. Comparison between the RRM3 WT model with that of the phosphomimetic mutant (RRM3 S318D) reveals minor differences mainly found at flexible loop regions. In fact, the RMSD for backbone atoms between both models is ca. 1.14 Å. Figure 3.10 shows how the side-chain of Ser318 at the β 4 strand is well-exposed to solvent in order to be accessible to Protein Kinase C δ (PKCδ) and to become phosphorylated. Intriguingly, the residue stretch ranging between Trp261 and Thr271 was not possible to assign in either RRM3 WT or RRM3 S318D, as indicated in Figure 3.10. This suggests that the helix/coil element is in a conformational intermediate exchange on the NMR time scale. Previously, it has been described the importance of RRM3 for ELAV – a homologue to HuR in Drosophila – multimerization (Toba and White, 2008). Indeed, authors concluded that the non-detectable Trp261 by NMR – Trp419 in ELAV sequence - was essential for protein-protein interactions. Thus, we suggest that the Trp261-Thr271 stretch form a dimerization epitope that involves the C-end of helix α 1 . Such a case was reported for the interactions between HuR RRM1 motifs through helices α 1 and α 2 (Benoit et al., 2010). Figure 3 .10: Structural mode l of HuR RRM3 WT domain built using che mical - shifts of backbone atoms as restraints in the CS23D server. The side-chain of Ser318, which becomes phosphorylated, is highlighted in yellow. Those residues of RRM3 WT between Trp261 and Thr271, which are in intermediate exchange on the NMR timescale for chemical-shift perturbations are marked in green. Results - 3.5 Shedding Light on the HuR C-terminal RRM3 Domain 82 The Hε of Trp261 from RRM3 WT in the 15 N HSQC spectrum was easily assigned by comparison with that of RRM3 W261E mutant. It is worth to mention that the resonance of Hε (Trp261) was broader than Hε (Trp244) – (32.9 Hz in 15 N / 55.7 Hz in 1 H and 27.7 Hz in 15 N / 40.8 Hz in 1 H, respectively), in agreement with the monomer-dimer exchange proposed for this helix/coil element (Figure 3.10). 3.5.3 Oligomerization of HuR RRM3 To get further insight on HuR oligomerization and the above mentioned low solubility of this protein, the formation of HuR RRM3 multimers was investigated by AU assays and MALDI measurements (Figure 3.11). By AU experiments (Figure 3.12), we could determine that the apparent molecular weight was ca. 19.2 kDa, which differs from the expected one for a monomer (14.2 kDa). Figure 3.11: Oligomerization of HuR RRM3. MALDI mass spectrometry reveals that RRM3 WT is a mixture of monomers (MW of 14047.1 Da) and dimers (28131.2 kDa) at 150 µM of protein Results - 3.5 Shedding Light on the HuR C-terminal RRM3 Domain 83 Figure 3.12: AU experiments using HuR RRM3 WT and RRM3 W261A species. A) AU at sedimentation equilibrium conditions. The apparent molecular weight (MW) is determined as 19033 ± 1425 Da, differing slightly from the estimated one for a monomer (14342.2 Da). B) The sedimentation coefficient of WT protein calculated by sedimentation velocity at 45.000 rpm results in 1.96 S and 19.2 kDa. A small population ( ~ 5%) corresponds to 5.08 S and 79.8 kDa C) The sedimentation coefficient of W261E mutant at the same velocity as the WT construct results in 1.85 S and 14.1 kDa. A small population (~ 5%) corresponds to 5.4 S and 70.6 kDa. Results -3.5 Shedding Light on the HuR C-terminal RRM3 Domain 84 The results fit in the best way to a fast/intermediate exchange model between monomeric and dimeric species, with a K A of about 1.8 x 10 4 M -1 . Then, about 70% of HuR RRM3 WT is already in the monomeric form at protein concentration used in these AU assays. The dimeric population may be substantially increased at NMR experiments which demand larger protein amounts. In addition, a small fraction of the protein (5%) tends to form aggregates with a molecular weight of ca. 79.8 kDa (Figure 3.12 B). MALDI experiments on HuR RRM3 WT confirmed the dimer formation and even trimers, which must be stable (Figure 3.11). In contrast to HuR RRM3 WT, the RRM3 W261E mutant is mainly monomeric in AU experiments. Under similar conditions in sedimentation velocity experiments, the molecular weight (MW) of this species was 14.1 kDa, which corresponds to the MW of the monomer state (14.2 kDa). Aggregation could be also observed as an additional peak corresponding to the molecular weight of 70.6 kDa, as happens for the WT protein (Figure 3.12 C). Thus, the Trp-261-to-Glu261 mutation impairs dimerization of the RRM3 domain. 3.6 RNA Binding of HuR RRM3 To test the RNA sequence specificity by HuR RRM3 WT domain, its interactions with two short 5-mer RNA oligonucleotides, namely 5´-AUUUA-3´ and 5´-UUUUU-3´, were studied by NMR titrations (Figure 3.13 and 3.14). Average chemical-shift perturbations (∆δ avg ) inferred from the titrations of 15 N-labeled RRM3 with both RNA oligonucleotides revealed that RRM3 binds to 5´-UUUUU-3´ with an apparent higher affinity than 5´-AUUUA-3´ (Figure 3.15, A and C), which indicates a preferential binding for pyrimidine bases than for purine derivates However, as the chemicalshift perturbations were too small, quantification of the binding was not feasible. The two RNA molecules lie on the RRM3 canonical platform, which comprises aromatic residues mainly localized at the central β strands (β 1 and β 3 ) from β-sheet (Figure 3.15, panels B and D). In fact, backbone amides for residues ranging from Phe247 and Leu251 at β 1 -RNP2 and from Phe287 and Met292 at β 3 -RNP1 were all perturbed in titrations with both RNAs, as Results -3.6 RNA Binding of HuR RRM3 85 previously reported (Query et al., 1989; Shamoo et al., 1997). Interestingly, the protein platform with which HuR RRM3 binds to 5´-UUUUU-3´ RNA is extended to involve the whole β-sheet, including most of β 2 and few signals at β 4 signals (Figure 3.15, panels B and D). In addition to ∆δ avg , some RRM3 signals at the protein-nucleic acid interface broadened beyond the detection limit independently of the RNA oligonucleotide used. Residues such as Ile248, Leu251, Val275 and Lys320 showed this behavior for both 5´-UUUUU-3´ and 5´-AUUUA-3´ RNAs. Results - 3.6 RNA Binding of HuR RRM3 86 Figure 3.13: RNA binding of HuR RRM3 with the 5´ - UUUUU - 3´ oligonucleotide. Average chemical - shift differences ( ∆δ avg ) between free and RNA - bound HuR RRM3 in a ratio of 4:1. Secondary structure elements of RRM3 are symbolized by blue arrows for β strands and red coil symbols for αhelices. Asterisk symbols (*) st and for residues broadened beyond the detection upon RNA binding. Results - 3.6 RNA Binding of HuR RRM3 87 Figure 3.14: RNA binding of HuR RRM3 with the 5´ - AUUUA - 3´ oligonucleotide. Average chemical - shift differences ( ∆δ avg ) between free and RNA - bound HuR RRM3 in a ratio of 4:1. Secondary structure elements of RRM3 are symbolized by blue arrows for β strands and red coil symbols for αhelices. Asterisk symbols (*) stand for residues broadened beyond the detection upon RNA binding. Results - 3.6 RNA Binding of HuR RRM3 88 Figure 3.15: RNA binding of HuR RRM3 with the 5´ - UUUUU - 3´ (A,B ) and 5´ - AUUUA - 3´ (C,D) oligonucleotides. (A,C) Superposition of 15 N-HSQC spectra of HuR RRM3 WT of free (black) and bound to RNA oligos in an RNA:RRM3 ratio of 2:1 (blue) and 4:1 (red). A subset of four representative resonances is labeled in black. (B,D) Map of HuR RRM3 interface upon binding to RNA. RRM3 surface is rotated 180˚ around the vertical axis in each view. Residues are colored, according to ∆δ avg (ppm): blue for values < 0.025, yellow for 0.025 ≤ Δδ avg ≤ 0.075 and orange for values > 0.075. Resonances broadened beyond the detection limit are colored in red. Prolines and unassigned resonances are indicated in gray. Results - 3.7 The Phosphomimetic HuR RRM3 S318D Mutant 89 Figure 3.16: HuR RRM3 S318D: A phosphomimetic mutant. A) Far-UV (190–250 nm) CD spectra of HuR RRM3 domain constructs. RRM3 WT is presented in blue, RRM3 S318A in red and RRM3 S318D in black. B) Chemical denaturation with guanidine hydrochloride (GuaHCl) of RRM3 WT (blue) and RRM3 S318D (black). Fitted unfolding curves are presented as solid lines, superimposed on experimental data in squares. The midpoint GuaHCl concentrations (C m ) of the transitions are marked by dashed lines. A B 3.7 The Phosphomimetic HuR RRM3 S318D Mutant Phosphorylation at Ser318 residue of HuR RRM3 domain was mimicked by Ser-to-Asp single mutation, which had negligible effects on both the secondary structure (Figure 3.16 A and Table 3.4) and the chemical stability (Figure 3.16 B) of the protein, as inferred from far-UV CD and fluorescence measurements, respectively. Discussion - RRM12 Works as a Functional Unit 96 4.2 RRM12 Works as a Unique Functional Unit Thermal stability studies of HuR species indicate the importance of the cooperation between the two N-terminal RRM domains of HuR, which work as a functional unit. The comparison of T m values for isolated RRM1 or RRM2 and the two-domain construct RRM12 reveals that RRM12 shows the same thermal stability as RRM1, while RRM2 is substantially more stable. In addition, the fact that the denaturation curve of HuR RRM12 is not the sum of those from the two individual RRM1 and RRM2 domains suggests cooperativity between both modules. It is tempting to speculate that the RRM12 modular interaction is essential for RNA recognition activity, similarly to what was previously observed for RRM1–RRM2 motifs of the homologous HuD protein upon cfos RNA binding (Wang and Tanaka Hall, 2001). Indeed, the preferred orientation between RNA binding domains helps to establish a high-affinity RNA-binding platform (Vitali et al., 2006; Li et al., 2010) and/or to stabilize a suitable conformation that can adapt to the changes in the direction of the RNA chain inside the highly structured 3´ UTRs, as previously suggested (Díaz-Moreno et al,. 2010). Altogether, the results indicate that HuR RRM12 works as a functional unit, independently of RRM3. 4.3 Phosphorylation of HuR RRM12 To study changes in structure and stability of HuR induced by serine phosphorylation, we designed three Ser-by-Asp mutations. Two of them are localized at the RRM cores, while the third one is in the interdomain linker between RRM1 and RRM2. No significant changes in secondary structure were observed for any of these phosphomimetic mutants, unlike what has been recently published for other RNA binding domains (Díaz-Moreno et al., 2009). Therefore, phosphorylation effects on HuR seem to be essentially related to RNA binding properties and/or intermolecular protein interactions rather than to changes of the HuR structure, as confirmed by our CD RNA binding titrations. CD experiments in the far-UV range Discussion – 4.3. Phosphorylation of HuR RRM12 97 confirmed that the secondary structure is not changed significantly by phosphorylation. 4.3.1 Phosphorylation of HuR RRM12 and Thermal Stability The thermal stability of HuR constructs is regulated by phosphorylation. The phosphomimetic mutant S88D slightly stabilizes RRM1 in the RRM12 context, which can be explained by the addition of a negative charge to the protein loop mainly governed by two positively charged residues (Benoit et al., 2010). Thus, Asp88 could minimize the electrostatic repulsion between Arg85 and Lys89, which would restrict the loop mobility. Phosphorylation at the level of the RRM12 linker region at Ser100 residue also has a negligible effect on HuR stability. A plausible explanation is that this solvent-exposed residue does not make many contacts with neighbors. Intriguingly, phosphorylation at Ser100 increases RNA binding in vivo (Abdelmohsen et al., 2007b), although the equivalent serine in the homologous HuD Ser126 is facing away from the RNA in the HuD/c-fos mRNA crystal structure (Wang and Tanaka Hall, 2001). Slightly destabilizing phosphorylation of Ser158 could be explained based on electrostatic repulsion with another nearby negative residue Glu162, although the negatively charged Asp158 is added at the N-end of αhelix 2 of HuR RRM2. 4.3.2 Phosphorylation of HuR RRM12 and RNA Binding In terms of RNA binding, it has been previously reported that in vivo HuR phosphorylation at Ser88 increases the docking of RNA targets to the RNA binding sites (Abdelmohsen et al., 2007b). Also it is proposed that the phosphoserine at position 88 exhibits a Mg 2+ -ion-mediated interaction with a phosphate group from RNA (Benoit et al., 2010). In this thesis, no substantial differences in binding affinities were observed between RRM12 WT and the phosphomimetic RRM12 S88D mutant by performing in vitro CD titrations using c-fos RNA. Discussion – 4.3. Phosphorylation of HuR RRM12 98 Posttranslational modification of Ser158 at RRM2 domain mimicked by the RRM12 S158D mutation tightly regulates the binding of HuR RRM12 with c-fos RNA in vitro. Actually, the RNA binding affinity of RRM12 S158D is four times larger than that of RRM12 WT, in agreement with previous data in vivo (Doller et al,. 2007). Additionally, in vitro CD RNA-RRM12 titrations reveal no effect of the S100D mutation on RNA recognition with respect to RRM12 WT. Therefore, phosphorylation at this site would influence RRM2– interdomain linker interactions and the rearrangement between RRM domains, rather than directly repulsing RNA (Benoit et al., 2010). Perturbations in HuR stability upon posttranslational modifications such as phosphorylation may explain the HuR behavior in binding RNA molecules, as well as in determining their lifetime and translation rate. 4.4 HuR RRM3 To shed light on the most C-terminal HuR RRM3 domain, which has been the HuR unknown module up to date from the structural point of view, a novel expression and purification protocol to improve its solubility has been developed. HuR RRM3 is the domain of HuR that is of special interest due to its differences to the N-terminal RRM domains RRM1 and RRM2. This domain is separated from the N-terminal RRM1 and RRM2 by a sixty residue linker that is cleaved at residue Asp226 to separate the protein under apoptotic conditions (Mazroui et al., 2008). The biophysical properties of the RRM3 are distinct to the N-terminal ones, resulting in low aquaeous solubility. This thesis reports how HuR RRM3 domain tumbles independently in solution with respect to the RRM12 tandem. This might make the HNS linker accessible ― in parpcular its Ser226 ― for caspase-dependent cleavage, as previously reported by Mazroui et al., 2008. Thus, the finding that RRM3 behaves as an autonomous working module within HuR FL can provide the protein with additional functions like adenosyl transferase activity (Meisner et al., 2009) and HuR-protein molecular recognition processes, as was shown for the homolog Drosophila protein (Toba and Discussion - 4.4 HuR RRM3 99 White, 2008) that enables more alternatives for HuR in RNA binding and processing. In addition, the RRM3-containing cleavage product acquires new functions in triggering apoptosis because it selectively binds to and stabilizes caspase-9 mRNA in an ARE-dependent manner (von Roretz et al., 2012). This agrees with our NMR titrations that demonstrate a binding of cleaved RRM3 to AREs, although previous studies suggested that the HuR RRM3 role ― as part of HuR FL ― in binding to single-stranded ARE-bearing mRNAs was negligible (Chung et al., 1996; Yeap et al., 2002, Fialcowitz- White et al., 2007, Barker et al., 2012). Actually, HuR RRM3 associates with the poly(A) tails of mRNAs (Ma et al., 1997; Anderson et al., 2000), which is crucial for effective mRNA stabilization by HuR (Fan and Steitz, 1998b). 4.4.1 Multimerization of HuR RRM3 On an intermediate NMR time scale, isolated RRM3 is in a monomer/multimer exchange that makes the resonances belonging to the Trp261-Thr271 stretch at the C-end of helix α 1 non-detectable. Thus, we suggest that RRM3 dimerization involves its α 1 -helix placed at the surface opposite to the RNA-binding platform, as previously proposed for homologous proteins such as ELAV in Drosophila upon mutating Trp419 (Trp261 in HuR; Toba and White, 2008). Toba and colleagues reported that multimerization of the homolog ELAV protein from Drosophila is dependent on RRM3 and the linker part (Toba and White, 2008). In this work it could be demonstrated that also the isolated domain of the human homolog HuR is able to form such multimers. This is especially interesting due to the cleavage of RRM3 and that it appears isolated within the cell under apoptotic conditions (Mazroui et al., 2008). It could be the case that, despite multimerization, the protein transports distinct Uracile-rich RNA targets and accumulates them by multimerization in form of stress granules (Gallouzi et al., 2000). Furthermore we can speculate that HuR RRM3 may be involved in degradation pathways, as it is already reported that certain RBPs, such as KSRP or AUF1, interact with the exosome (Chen et al., 2001). Our AU- and MALDI analysis confirm the oligomerization of RRM3, and that RRM3 W261E mutant shifts the monomer/dimer equilibrium towards the monomeric RRM3 form. Therefore, RRM-RRM interactions Discussion - 4.4 HuR RRM3 100 often take place through their α-helices. For instance, RRM3 and RRM4 from Polypyrimidine Tract Binding protein (PTB) mainly contact through helix α 2 of RRM4 and helix α 1 and α 2 of RRM3, resulting in the perpendicular positioning of both RRM (Oberstrass et al., 2005). On the other hand, in FBP-Interacting Repressor (FIR) protein, the α-helix face of RRM1 packs onto the β-sheet face of RRM2, creating a stable interface (Crichlow et al., 2008; Cukier et al., 2010). Interestingly enough is the crystallographic structure of HuR RRM1 motif, which reveals a homodimer assembled through helices α 1 and α 2 (Benoit et al., 2010). Multimerization of Hu protein family is then dependent on RRM3 motif and more especially on its well-conserved Trp261 (Toba and White, 2008). In fact, the whole Trp261-Thr271 stretch is highly conserved among Hu proteins. Our structural studies on HuR RRM3 also demonstrate that the monomer/dimer exchange of HuR RRM3 takes place even in absence of RNA, in contrast to previous reports suggesting that RNA promotes HuR FL multimerization (David et al., 2007; Fialcowitz-White et al., 2007). Importantly, no additional resonances corresponding to the Trp261-Thr271 gap appear on the 15 N-HSQC NMR spectra upon binding to short singlestranded 5-mer RNA oligonucleotides, from which it follows that the oligomerization exchange remains unaltered. The fact that HuR FL forms stable oligomeric complexes in presence of long RNA fragments suggests that additional experiments are necessary to explain how far the oligomerization of isolated RRM3 would change by long and highlystructured ARE-bearing RNAs. On the other hand, the multimerization of HuR could be prevented by the W216E mutation, as already showed for the homolog Drosophila protein (Toba and White, 2008). For this mutant, the AU experiments demonstrate a unique monomeric protein population, also independent of RNA interactions. For future investigation, it would be of interest what consequences this mutation has at the cellular level. 4.4.2 RNA Binding of HuR RRM3 RRM-comprising RBPs are usually known to bind single-stranded nucleic acids by stacking with aromatic residues placed at the central βstrands (β 1 and β 3 ) of the β-sheet, although a high variety of RNA binding mechanisms have been described for RRM modules (for a review, see Clery Discussion - 4.4 HuR RRM3 101 et al., 2008). By NMR experiments we confirm the RNA binding to these secondary structure elements of RRM3 for both 5´-AUUUA-3´ and 5´- UUUUU-3´ oligonucleotides. In addition, we found out that the HuR RRM3 protein platform to interact with 5´-UUUUU-3´ RNA is extended and involves the whole β-sheet. This agrees with a preferential binding of RRM3 to pyrimidine bases rather than to purine derivates, based on chemical-shift perturbations data. Preferences by 5-mer U-rich elements over AU-rich motifs were already observed for the heterogeneous nuclear RiboNucleoprotein C protein (hnRNP1; Sokolowski et al., 2001) and HuR FL (Sokolowski et al., 1999; López de Silanes et al., 2004). More recently, it has been also demonstrated that HuR FL recognizes mRNAs containing U-richness and single-strandedness, where a substitution of U by A or C has a moderate effect on HuR binding affinity, whereas an exchange by G has a drastic effect (Barker et al., 2012). The interruption of U-rich sequences with Cs to result UC-rich motifs is also recognized by HuR FL, although its RRM3 module does not bind this RNA probe at all (Yeap et al., 2002). Meriting particular interest, the RRM3-containing cleavage product specifically binds ARE1 of caspase-9 mRNA, which lacks of 5´-AUUUA-3´ sequences but contains a U-rich region (von Roretz et al., 2012). Thus, ARE1 of caspase-9 mRNA is classified as a Class III ARE, according to the categories defined by Chen and coworkers (Chen et al., 1995). Ongoing experiments would be necessary to find out which sequence specificity and number of nucleotides are preferred for distinct RRM domains not only from HuR protein but also from other RBPs. This observation is a small contribution to the complex field of the interaction between proteins and RNA inside the cell, where we have a sum of many more factors, as competition with other RBPs by RNA sequences and cellular conditions. Indeed, the complex network of elements acting in cis- and trans-factors within the cell makes it difficult to predict those exact working mechanisms for controlling mRNA fate. 4.4.3 Phosphorylation of HuR RRM3 The RBPs which act as trans-factors are often upperregulated by posttranslational modifications, such as Discussion - 4.4 HuR RRM3 102 phosphorylation. In HuR, phosphorylation at RRM3 Ser318 residue implies an important mode in its regulation, with consequences in colon carcinoma cells due to HuR dysregulation (Doller et al., 2011). The Ser318 residue of HuR RRM3, which is phosphorylated by PKCδ (Doller et al., 2008a), was mutated by an aspartate to mimic phosphorylation events. Both RRM3 WT and RRM3 S318D models agree on the secondary structure and the chemical stability. The major difference was found at the level of RNA recognition, since RRM3 S318D binds U-rich stretches with an apparent lower affinity than RRM3 WT, although RNA docks on both RRM3 species using similar platforms. This result can be explained by an electrostatic repulsion effect between Asp318 and the negatively charged RNA (Barker et al., 2012). Indeed, the backbone amide of Asp318 is physically in contact with RNA, as inferred from chemical-shift data. In contrast to our findings, Doller and coworkers (Doller et al., 2010) reported a higher binding affinity of phosphorylated HuR FL with AU-bearing mRNA stretches (Doller et al., 2010). Plausible explanations arise from: i) differences in length, degree of structure and sequence of AU-containing long mRNAs and 5´-UUUUU-3´ short RNA molecules; ii) differences in number of RRMs making up HuR FL and RRM3 S318D species; iii) differences on how Ser318 residue is modified by PKCδ-dependent phosphorylation of HuR FL or by Ser318-by-Asp318 substitution on isolated HuR RRM3. Our study underlines the exceptional properties of HuR RRM3 as a multi-functional domain which may lead to HuR oligomerization and binding to U-rich RNA targets at once, using two opposite RRM surfaces. Conclusions 113 References Abdelmohsen K, Lal A, Kim HH, Gorospe M (2007a) Posttranscriptional orchestration of an anti-apoptotic program by HuR. Cell Cycle 6:1288-1292 Abdelmohsen K, Pullmann R, Jr., Lal A, Kim HH, Galban S, Yang X, Blethrow JD, Walker M, Shubert J, Gillespie DA, Furneaux H, Gorospe M (2007b) Phosphorylation of HuR by Chk2 regulates SIRT1 expression. Mol Cell 25:543-557 Abdelmohsen K, Srikantan S, Kuwano Y, Gorospe M (2008) miR-519 reduces cell proliferation by lowering RNA-binding protein HuR levels. Proc Natl Acad Sci U S A 105:20297-20302 Abdelmohsen K, Srikantan S, Yang X, Lal A, Kim HH, Kuwano Y, Galban S, Becker KG, Kamara D, de Cabo R, Gorospe M (2009) Ubiquitin-mediated proteolysis of HuR by heat shock. Embo J 28:1271-1282 Abdelmohsen K, Kim MM, Srikantan S, Mercken EM, Brennan SE, Wilson GM, Cabo R, Gorospe M (2010) miR-519 suppresses tumor growth by reducing HuR levels. Cell Cycle 9:1354-1359 Adam SA, Nakagawa T, Swanson MS, Woodruff TK, Dreyfuss G (1986) mRNA polyadenylate-binding protein: gene isolation and sequencing and identification of a ribonucleoprotein consensus sequence. Mol Cell Biol 6:2932-2943 Al-Ahmadi W, Al-Ghamdi M, Al-Haj L, Al-Saif M, Khabar KS (2009) Alternative polyadenylation variants of the RNA binding protein, HuR: abundance, role of AU-rich elements and auto- Regulation. Nucleic Acids Res 37:3612-3624 References 114 Allain FH, Bouvet P, Dieckmann T, Feigon J (2000) Molecular basis of sequence-specific recognition of pre-ribosomal RNA by nucleolin. Embo J 19:6870-6881 Anderson KD, Morin MA, Beckel-Mitchener A, Mobarak CD, Neve RL, Furneaux HM, Burry R, Perrone-Bizzozero NI (2000) Overexpression of HuD, but not of its truncated form HuD I+II, promotes GAP-43 gene expression and neurite outgrowth in PC12 cells in the absence of nerve growth factor. J Neurochem 75:1103-1114 Arnold K, Bordoli L, Kopp J, Schwede T (2006) The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22:195-201 Aroca A, Díaz-Quintana A, Díaz-Moreno I (2011) A structural insight into the C-terminal RNA recognition motifs of T-cell intracellular antigen-1 protein. FEBS Lett 585:2958-2964 Ashiya M, Grabowski PJ (1997) A neuron-specific splicing switch mediated by an array of pre-mRNA repressor sites: evidence of a regulatory role for the polypyrimidine tract binding protein and a brain-specific PTB counterpart. Rna 3:996-1015 Atasoy U, Watson J, Patel D, Keene JD (1998) ELAV protein HuA (HuR) can redistribute between nucleus and cytoplasm and is upregulated during serum stimulation and T cell activation. J Cell Sci 111 (Pt 21):3145-3156 Atasoy U, Curry SL, Lopez de Silanes I, Shyu AB, Casolaro V, Gorospe M, Stellato C (2003) Regulation of eotaxin gene expression by TNF-alpha and IL-4 through mRNA stabilization: involvement of the RNA-binding protein HuR. J Immunol 171:4369-4378 References 115 Backe PH, Messias AC, Ravelli RB, Sattler M, Cusack S (2005) X-ray crystallographic and NMR studies of the third KH domain of hnRNP K in complex with single-stranded nucleic acids. Structure 13:1055-1067 Badis G, Saveanu C, Fromont-Racine M, Jacquier A (2004) Targeted mRNA degradation by deadenylation-independent decapping. Mol Cell 15:5-15 Bagga PS, Arhin GK, Wilusz J (1998) DSEF-1 is a member of the hnRNP H family of RNA-binding proteins and stimulates pre-mRNA cleavage and polyadenylation in vitro. Nucleic Acids Res 26:5343-5350 Bakheet T, Frevel M, Williams BR, Greer W, Khabar KS (2001) ARED: human AU-rich element-containing mRNA database reveals an unexpectedly diverse functional repertoire of encoded proteins. Nucleic Acids Res 29:246-254 Bakheet T, Williams BR, Khabar KS (2003) ARED 2.0: an update of AU- rich element mRNA database. Nucleic Acids Res 31:421-423 Barker A, Epis MR, Porter CJ, Hopkins BR, Wilce MC, Wilce JA, Giles KM, Leedman PJ (2012) Sequence requirements for RNA binding by HuR and AUF1. J Biochem 151:423-437 Barreau C, Paillard L, Osborne HB (2005) AU-rich elements and associated factors: are there unifying principles? Nucleic Acids Res 33:7138-7150 Bartel DP (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116:281-297 Bhattacharyya SN, Habermacher R, Martine U, Closs EI, Filipowicz W (2006) Relief of microRNA-mediated translational repression in human cells subjected to stress. Cell 125:1111-1124 References 116 Beelman CA, Stevens A, Caponigro G, LaGrandeur TE, Hatfield L, Fortner DM, Parker R (1996) An essential component of the decapping enzyme required for normal rates of mRNA turnover. Nature 382:642-646 Bellavia D, Mecarozzi M, Campese AF, Grazioli P, Talora C, Frati L, Gulino A, Screpanti I (2007) Notch3 and the Notch3- upregulated RNA-binding protein HuD regulate Ikaros alternative splicing. Embo J 26:1670-1680 Benoit RM, Meisner NC, Kallen J, Graff P, Hemmig R, Cebe R, Ostermeier C, Widmer H, Auer M (2010) The x-ray crystal structure of the first RNA recognition motif and site-directed mutagenesis suggest a possible HuR redox sensing mechanism. J Mol Biol 397:1231-1244 Beuth B, Pennell S, Arnvig KB, Martin SR, Taylor IA (2005) Structure of a Mycobacterium tuberculosis NusA-RNA complex. Embo J 24:3576-3587 Birney E, Kumar S, Krainer AR (1993) Analysis of the RNA-recognition motif and RS and RGG domains: conservation in metazoan pre-mRNA splicing factors. Nucleic Acids Res 21:5803-5816 Bleichert F, Baserga SJ (2007) The long unwinding road of RNA helicases. Mol Cell 27:339-352 Bolognani F, Tanner DC, Merhege M, Deschenes-Furry J, Jasmin B, Perrone-Bizzozero NI (2006) In vivo post-transcriptional regulation of GAP-43 mRNA by overexpression of the RNA- binding protein HuD. J Neurochem 96:790-801 Bouvet P, Diaz JJ, Kindbeiter K, Madjar JJ, Amalric F (1998) Nucleolin interacts with several ribosomal proteins through its RGG domain. J Biol Chem 273:19025-19029 References 117 Braddock DT, Baber JL, Levens D, Clore GM (2002a) Molecular basis of sequence-specific single-stranded DNA recognition by KH domains: solution structure of a complex between hnRNP K KH3 and single-stranded DNA. Embo J 21:3476-3485 Braddock DT, Louis JM, Baber JL, Levens D, Clore GM (2002b) Structure and dynamics of KH domains from FBP bound to single-stranded DNA. Nature 415:1051-1056 Brennan CM, Gallouzi IE, Steitz JA (2000) Protein ligands to HuR modulate its interaction with target mRNAs in vivo. J Cell Biol 151:1-14 Brennan CM, Steitz JA (2001) HuR and mRNA stability. Cell Mol Life Sci 58:266-277 Brown CE, Tarun SZ, Jr., Boeck R, Sachs AB (1996) PAN3 encodes a subunit of the Pab1p-dependent poly(A) nuclease in Saccharomyces cerevisiae. Mol Cell Biol 16:5744-5753 Bushati N, Cohen SM (2007) MicroRNA functions. Annu Rev Cell Dev Biol 23:175-205 Caceres JF, Stamm S, Helfman DM, Krainer AR (1994) Regulation of alternative splicing in vivo by overexpression of antagonistic splicing factors. Science 265:1706-9 Calnan B, Tidor B, Biancalana S, Hudson D, Frankel A (1991) Argininemediated RNA recognition: the arginine fork. Science 252:1167-1171 Carballo E, Lai WS, Blackshear PJ (1998) Feedback inhibition of macrophage tumor necrosis factor-alpha production by tristetraprolin. Science 281:1001-1005 References 118 Carson JH, Cui H, Barbarese E (2001) The balance of power in RNA trafficking. Curr Opin Neurobiol 11:558-563 Chabot B, Blanchette M, Lapierre I, La Branche H (1997) An intron element modulating 5' splice site selection in the hnRNP A1 pre-mRNA interacts with hnRNP A1. Mol Cell Biol 17:1776- 1786 Chan RC, Black DL (1997) The polypyrimidine tract binding protein binds upstream of neural cell-specific c-src exon N1 to repress the splicing of the intron downstream. Mol Cell Biol 17:4667-4676 Chang TC, Yamashita A, Chen CY, Yamashita Y, Zhu W, Durdan S, Kahvejian A, Sonenberg N, Shyu AB (2004) UNR, a new partner of poly(A)-binding protein, plays a key role in translationally coupled mRNA turnover mediated by the c-fos major coding-region determinant. Genes Dev 18:2010-2023 Chen CD, Kobayashi R, Helfman DM (1999) Binding of hnRNP H to an exonic splicing silencer is involved in the regulation of alternative splicing of the rat beta-tropomyosin gene. Genes Dev 13:593-606 Chen CY, Gherzi R, Ong SE, Chan EL, Raijmakers R, Pruijn GJ, Stoecklin G, Moroni C, Mann M, Karin M (2001) AU binding proteins recruit the exosome to degrade ARE-containing mRNAs. Cell 107:451-464 Chen CY, Shyu AB (1995) AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem Sci 20:465-470 References 119 Chen CY, Xu N, Shyu AB (2002) Highly selective actions of HuR in antagonizing AU-rich element-mediated mRNA destabilization. Mol Cell Biol 22:7268-7278 Cherry J, Karschner V, Jones H, Pekala PH (2006) HuR, an RNA-binding protein, involved in the control of cellular differentiation. In Vivo 20:17-23 Chiba Y, Johnson MA, Lidder P, Vogel JT, van Erp H, Green PJ (2004) AtPARN is an essential poly(A) ribonuclease in Arabidopsis. Gene 328:95-102 Chkheidze AN, Lyakhov DL, Makeyev AV, Morales J, Kong J, Liebhaber SA (1999) Assembly of the alpha-globin mRNA stability complex reflects binary interaction between the pyrimidinerich 3' untranslated region determinant and poly(C) binding protein alphaCP. Mol Cell Biol 19:4572-4581 Cho SJ, Zhang J, Chen X (2010) RNPC1 modulates the RNA-binding activity of, and cooperates with, HuR to regulate p21 mRNA stability. Nucleic Acids Res 38:2256-2267 Chou MY, Rooke N, Turck CW, Black DL (1999) hnRNP H is a component of a splicing enhancer complex that activates a csrc alternative exon in neuronal cells. Mol Cell Biol 19:69-77 Chung S, Jiang L, Cheng S, Furneaux H (1996) Purification and properties of HuD, a neuronal RNA-binding protein. J Biol Chem 271:11518-11524 Clemens KR, Wolf V, McBryant SJ, Zhang P, Liao X, Wright PE, Gottesfeld JM (1993) Molecular basis for specific recognition of both RNA and DNA by a zinc finger protein. Science 260:530-533 References 120 Clery A, Blatter M, Allain FH (2008) RNA recognition motifs: boring? Not quite. Curr Opin Struct Biol 18:290-298 Collier B, Goobar-Larsson L, Sokolowski M, Schwartz S (1998) Translational inhibition in vitro of human papillomavirus type 16 L2 mRNA mediated through interaction with heterogenous ribonucleoprotein K and poly(rC)-binding proteins 1 and 2. J Biol Chem 273:22648-22656 Cote CA, Gautreau D, Denegre JM, Kress TL, Terry NA, Mowry KL (1999) A Xenopus protein related to hnRNP I has a role in cytoplasmic RNA localization. Mol Cell 4:431-437 Crichlow GV, Zhou H, Hsiao HH, Frederick KB, Debrosse M, Yang Y, Folta-Stogniew EJ, Chung HJ, Fan C, De la Cruz EM, Levens D, Lolis E, Braddock D (2008) Dimerization of FIR upon FUSE DNA binding suggests a mechanism of c-myc inhibition. Embo J 27:277-289 Crick F (1970) Central dogma of molecular biology. Nature 227:561- 563 Crick FH (1958) On protein synthesis. Symp Soc Exp Biol 12:138-163 Cukier CD, Hollingworth D, Martin SR, Kelly G, Díaz-Moreno I, Ramos A (2010) Molecular basis of FIR-mediated c-myc transcriptional control. Nat Struct Mol Biol 17:1058-1064 Cukier CD, Ramos A (2011) Modular protein-RNA interactions regulating mRNA metabolism: a role for NMR. Eur Biophys J 40:1317-1325 Daubner GM, Clery A, Allain FH (2013) RRM-RNA recognition: NMR or crystallography...and new findings. Curr Opin Struct Biol 23:100-108 References 121 David PS, Tanveer R, Port JD (2007) FRET-detectable interactions between the ARE binding proteins, HuR and p37AUF1. Rna 13:1453-1468 Dean JL, Wait R, Mahtani KR, Sully G, Clark AR, Saklatvala J (2001) The 3' untranslated region of tumor necrosis factor alpha mRNA is a target of the mRNA-stabilizing factor HuR. Mol Cell Biol 21:721-730 Deana A, Celesnik H, Belasco JG (2008) The bacterial enzyme RppH triggers messenger RNA degradation by 5' pyrophosphate removal. Nature 451:355-358 De Guzman RN, Wu ZR, Stalling CC, Pappalardo L, Borer PN, Summers MF (1998) Structure of the HIV-1 nucleocapsid protein bound to the SL3 psi-RNA recognition element. Science 279:384-388 Dehlin E, Wormington M, Korner CG, Wahle E (2000) Cap-dependent deadenylation of mRNA. Embo J 19:1079-1086 Del Gatto-Konczak F, Olive M, Gesnel MC, Breathnach R (1999) hnRNP A1 recruited to an exon in vivo can function as an exon splicing silencer. Mol Cell Biol 19:251-260 Dempsey LA, Sun H, Hanakahi LA, Maizels N (1999) G4 DNA binding by LR1 and its subunits, nucleolin and hnRNP D, A role for G- G pairing in immunoglobulin switch recombination. J Biol Chem 274:1066-1071 Díaz-Moreno I, Hollingworth D, Frenkiel TA, Kelly G, Martin S, Howell S, Garcia-Mayoral M, Gherzi R, Briata P, Ramos A (2009) Phosphorylation-mediated unfolding of a KH domain regulates KSRP localization via 14-3-3 binding. Nat Struct Mol Biol 16:238-246 References 128 Hoek KS, Kidd GJ, Carson JH, Smith R (1998) hnRNP A2 selectively binds the cytoplasmic transport sequence of myelin basic protein mRNA. Biochemistry 37:7021-7029 Houseley J, Tollervey D (2009) The many pathways of RNA degradation. Cell 136:763-776 Hudson BP, Martinez-Yamout MA, Dyson HJ, Wright PE (2004) Recognition of the mRNA AU-rich element by the zinc finger domain of TIS11d. Nat Struct Mol Biol 11:257-264 Iyaguchi D, Yao M, Tanaka I, Toyota E (2009) Cloning, expression, purification and preliminary crystallographic studies of the adenylate/uridylate-rich element-binding protein HuR complexed with its target RNA. Acta Crystallogr Sect F Struct Biol Cryst Commun 65:285-287 Inoue M, Muto Y, Sakamoto H, Yokoyama S (2000) NMR studies on functional structures of the AU-rich element-binding domains of Hu antigen C. Nucleic Acids Res 28:1743-1750 Ishikawa F, Matunis MJ, Dreyfuss G, Cech TR (1993) Nuclear proteins that bind the pre-mRNA 3' splice site sequence r(UUAG/G) and the human telomeric DNA sequence d(TTAGGG)n. Mol Cell Biol 13:4301-4310 Izaurralde E, Jarmolowski A, Beisel C, Mattaj IW, Dreyfuss G, Fischer U (1997) A role for the M9 transport signal of hnRNP A1 in mRNA nuclear export. J Cell Biol 137:27-35 Jain RG, Andrews LG, McGowan KM, Pekala PH, Keene JD (1997) Ectopic expression of Hel-N1, an RNA-binding protein, increases glucose transporter (GLUT1) expression in 3T3-L1 adipocytes. Mol Cell Biol 17:954-962 References 129 Janga SC, Mittal N (2011) Construction, structure and dynamics of post-transcriptional regulatory network directed by RNA- binding proteins. Adv Exp Med Biol 722:103-117 Johansson C, Finger LD, Trantirek L, Mueller TD, Kim S, Laird-Offringa IA, Feigon J (2004) Solution structure of the complex formed by the two N-terminal RNA-binding domains of nucleolin and a pre-rRNA target. J Mol Biol 337:799-816 Jones PG, VanBogelen RA, Neidhardt FC (1987) Induction of proteins in response to low temperature in Escherichia coli. J Bacteriol 169:2092-2095 Jung YS, Zweckstetter M (2004) Mars - robust automatic backbone assignment of proteins. J Biomol NMR 30:11-23 Kang MJ, Ryu BK, Lee MG, Han J, Lee JH, Ha TK, Byun DS, Chae KS, Lee BH, Chun HS, Lee KY, Kim HJ, Chi SG (2008) NF-kappaB activates transcription of the RNA-binding factor HuR, via PI3K-AKT signaling, to promote gastric tumorigenesis. Gastroenterology 135:2030-2042 Kannt A, Lancaster CR, Michel H (1998) The role of electrostatic interactions for cytochrome c oxidase function. J Bioenerg Biomembr 30:81-87 Katsanou V, Papadaki O, Milatos S, Blackshear PJ, Anderson P, Kollias G, Kontoyiannis DL (2005) HuR as a negative posttranscriptional modulator in inflammation. Mol Cell 19:777-789 Kawai T, Lal A, Yang X, Galban S, Mazan-Mamczarz K, Gorospe M (2006) Translational control of cytochrome c by RNA-binding proteins TIA-1 and HuR. Mol Cell Biol 26:3295-3307 References 130 Kedde M, Agami R (2008) Interplay between microRNAs and RNA- binding proteins determines developmental processes. Cell Cycle 7:899-903 Keene JD (1999) Why is Hu where? Shuttling of early-response-gene messenger RNA subsets. Proc Natl Acad Sci U S A 96:5-7 Kessler MM, Henry MF, Shen E, Zhao J, Gross S, Silver PA, Moore CL (1997) Hrp1, a sequence-specific RNA-binding protein that shuttles between the nucleus and the cytoplasm, is required for mRNA 3'-end formation in yeast. Genes Dev 11:2545- 2556 Kiefer F, Arnold K, Kunzli M, Bordoli L, Schwede T (2009) The SWISSMODEL Repository and associated resources. Nucleic Acids Res 37:D387-392 Kiledjian M, Dreyfuss G (1992) Primary structure and binding activity of the hnRNP U protein: binding RNA through RGG box. Embo J 11:2655-2664 Kiledjian M, Wang X, Liebhaber SA (1995) Identification of two KH domain proteins in the alpha-globin mRNP stability complex. Embo J 14:4357-4364 Kiledjian M, DeMaria CT, Brewer G, Novick K (1997) Identification of AUF1 (heterogeneous nuclear ribonucleoprotein D) as a component of the alpha-globin mRNA stability complex. Mol Cell Biol 17:4870-4876 Kim HH, Yang X, Kuwano Y, Gorospe M (2008) Modification at HuR (S242) alters HuR localization and proliferative influence. Cell Cycle 7:3371-3377 References 131 Kim HS, Wilce MC, Yoga YM, Pendini NR, Gunzburg MJ, Cowieson NP, Wilson GM, Williams BR, Gorospe M, Wilce JA (2011) Different modes of interaction by TIAR and HuR with target RNA and DNA. Nucleic Acids Res 39:1117-1130 Korner CG, Wormington M, Muckenthaler M, Schneider S, Dehlin E, Wahle E (1998) The deadenylating nuclease (DAN) is involved in poly(A) tail removal during the meiotic maturation of Xenopus oocytes. Embo J 17:5427-5437 Kshirsagar M, Parker R (2004) Identification of Edc3p as an enhancer of mRNA decapping in Saccharomyces cerevisiae. Genetics 166:729-739 Kundu P, Fabian MR, Sonenberg N, Bhattacharyya SN, Filipowicz W (2012) HuR protein attenuates miRNA-mediated repression by promoting miRISC dissociation from the target RNA. Nucleic Acids Res 40:5088-5100 LaBranche H, Dupuis S, Ben-David Y, Bani MR, Wellinger RJ, Chabot B (1998) Telomere elongation by hnRNP A1 and a derivative that interacts with telomeric repeats and telomerase. Nat Genet 19:199-202 LaCava J, Houseley J, Saveanu C, Petfalski E, Thompson E, Jacquier A, Tollervey D (2005) RNA degradation by the exosome is promoted by a nuclear polyadenylation complex. Cell 121:713-724 Lal A, Mazan-Mamczarz K, Kawai T, Yang X, Martindale JL, Gorospe M (2004) Concurrent versus individual binding of HuR and AUF1 to common labile target mRNAs. Embo J 23:3092-3102 Lal A, Kawai T, Yang X, Mazan-Mamczarz K, Gorospe M (2005) Antiapoptotic function of RNA-binding protein HuR effected through prothymosin alpha. Embo J 24:1852-1862 References 132 Laue T M, Shah B D, Ridgeway T M & Pelletier S L (1992) Computeraided interpretation of analytical sedimentation data for proteins in Analytical Ultracentrifugation in Biochemistry and Polymer Science (Eds Harding, S. E., Rowe, A.J. & Horton, J.C.) 90−125 (Royal Society of Chemistry, Cambridge, 1992). Lee MS, Henry M, Silver PA (1996) A protein that shuttles between the nucleus and the cytoplasm is an important mediator of RNA export. Genes Dev 10:1233-1246 Lee Y, Ahn C, Han J, Choi H, Kim J, Yim J, Lee J, Provost P, Radmark O, Kim S, Kim VN (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425:415-419 Levine TD, Gao F, King PH, Andrews LG, Keene JD (1993) Hel-N1: an autoimmune RNA-binding protein with specificity for 3' uridylate-rich untranslated regions of growth factor mRNAs. Mol Cell Biol 13:3494-3350 Lewis HA, Musunuru K, Jensen KB, Edo C, Chen H, Darnell RB, Burley SK (2000) Sequence-specific RNA binding by a Nova KH domain: implications for paraneoplastic disease and the fragile X syndrome. Cell 100:323-332 Li H, Park S, Kilburn B, Jelinek MA, Henschen-Edman A, Aswad DW, Stallcup MR, Laird-Offringa IA (2002) Lipopolysaccharideinduced methylation of HuR, an mRNA-stabilizing protein, by CARM1. Coactivator-associated arginine methyltransferase. J Biol Chem 277:44623-44630 Li H, Shi H, Wang H, Zhu Z, Li X, Gao Y, Cui Y, Niu L, Teng M (2010) Crystal structure of the two N-terminal RRM domains of Pub1 and the poly(U)-binding properties of Pub1. J Struct Biol 171:291-297 References 133 Liu X, Mertz JE (1995) HnRNP L binds a cis-acting RNA sequence element that enables intron-dependent gene expression. Genes Dev 9:1766-1780 Liu Z, Luyten I, Bottomley MJ, Messias AC, Houngninou-Molango S, Sprangers R, Zanier K, Kramer A, Sattler M (2001) Structural basis for recognition of the intron branch site RNA by splicing factor 1. Science 294:1098-1102 Liu H, Rodgers ND, Jiao X, Kiledjian M (2002) The scavenger mRNA decapping enzyme DcpS is a member of the HIT family of pyrophosphatases. Embo J 21:4699-4708 Liu Q, Greimann JC, Lima CD (2006) Reconstitution, activities, and structure of the eukaryotic RNA exosome. Cell 127:1223- 1237 Loflin P, Chen CY, Shyu AB (1999) Unraveling a cytoplasmic role for hnRNP D in the in vivo mRNA destabilization directed by the AU-rich element. Genes Dev 13:1884-1897 Lopez de Silanes I, Zhan M, Lal A, Yang X, Gorospe M (2004) Identification of a target RNA motif for RNA-binding protein HuR. Proc Natl Acad Sci U S A 101:2987-2992 Lorentzen E, Walter P, Fribourg S, Evguenieva-Hackenberg E, Klug G, Conti E (2005) The archaeal exosome core is a hexameric ring structure with three catalytic subunits. Nat Struct Mol Biol 12:575-581 Lund E, Guttinger S, Calado A, Dahlberg JE, Kutay U (2004) Nuclear export of microRNA precursors. Science 303:95-98 Lunde BM, Moore C, Varani G (2007) RNA-binding proteins: modular design for efficient function. Nat Rev Mol Cell Biol 8:479-490 References 134 Lykke-Andersen J (2002) Identification of a human decapping complex associated with hUpf proteins in nonsensemediated decay. Mol Cell Biol 22:8114-8121 Ma WJ, Cheng S, Campbell C, Wright A, Furneaux H (1996) Cloning and characterization of HuR, a ubiquitously expressed Elavlike protein. J Biol Chem 271:8144-8151 Ma WJ, Chung S, Furneaux H (1997) The Elav-like proteins bind to AU- rich elements and to the poly(A) tail of mRNA. Nucleic Acids Res 25:3564-3569 Ma JB, Ye K, Patel DJ (2004) Structural basis for overhang-specific small interfering RNA recognition by the PAZ domain. Nature 429:318-322 Ma JB, Yuan YR, Meister G, Pei Y, Tuschl T, Patel DJ (2005) Structural basis for 5'-end-specific recognition of guide RNA by the A. fulgidus Piwi protein. Nature 434:666-670 Macrae IJ, Zhou K, Li F, Repic A, Brooks AN, Cande WZ, Adams PD, Doudna JA (2006) Structural basis for double-stranded RNA processing by Dicer. Science 311:195-198 Marasa BS, Srikantan S, Martindale JL, Kim MM, Lee EK, Gorospe M, Abdelmohsen K (2010) MicroRNA profiling in human diploid fibroblasts uncovers miR-519 role in replicative senescence. Aging 2:333-343 Maris C, Dominguez C, Allain FH (2005) The RNA recognition motif, a plastic RNA-binding platform to regulate post-transcriptional gene expression. Febs J 272:2118-2131 Matsumoto K, Wolffe AP (1998) Gene regulation by Y-box proteins: coupling control of transcription and translation. Trends Cell Biol 8:318-823 References 135 Matunis EL, Kelley R, Dreyfuss G (1994) Essential role for a heterogeneous nuclear ribonucleoprotein (hnRNP) in oogenesis: hrp40 is absent from the germ line in the dorsoventral mutant squid. Proc Natl Acad Sci U S A 91:2781- 2784 Mayeda A, Krainer AR (1992) Regulation of alternative pre-mRNA splicing by hnRNP A1 and splicing factor SF2. Cell 68:365-375 Mazan-Mamczarz K, Galban S, Lopez de Silanes I, Martindale JL, Atasoy U, Keene JD, Gorospe M (2003) RNA-binding protein HuR enhances p53 translation in response to ultraviolet light irradiation. Proc Natl Acad Sci U S A 100:8354-8359 Mazroui R, Di Marco S, Clair E, von Roretz C, Tenenbaum SA, Keene JD, Saleh M, Gallouzi IE (2008) Caspase-mediated cleavage of HuR in the cytoplasm contributes to pp32/PHAP-I regulation of apoptosis. J Cell Biol 180:113-127 McMullen MR, Cocuzzi E, Hatzoglou M, Nagy LE (2003) Chronic ethanol exposure increases the binding of HuR to the TNFalpha 3'-untranslated region in macrophages. J Biol Chem 278:38333-38341 Meisner NC, Hintersteiner M, Mueller K, Bauer R, Seifert JM, Naegeli HU, Ottl J, Oberer L, Guenat C, Moss S, Harrer N, Woisetschlaeger M, Buehler C, Uhl V, Auer M (2007) Identification and mechanistic characterization of lowmolecular-weight inhibitors for HuR. Nat Chem Biol 3:508- 515 Meisner NC, Hintersteiner M, Seifert JM, Bauer R, Benoit RM, Widmer A, Schindler T, Uhl V, Lang M, Gstach H, Auer M (2009) Terminal adenosyl transferase activity of posttranscriptional regulator HuR revealed by confocal onbead screening. J Mol Biol 386:435-450 References 136 Miau LH, Chang CJ, Shen BJ, Tsai WH, Lee SC (1998) Identification of heterogeneous nuclear ribonucleoprotein K (hnRNP K) as a repressor of C/EBPbeta-mediated gene activation. J Biol Chem 273:10784-10791 Michelotti EF, Michelotti GA, Aronsohn AI, Levens D (1996) Heterogeneous nuclear ribonucleoprotein K is a transcription factor. Mol Cell Biol 16:2350-2360 Min H, Chan RC, Black DL (1995) The generally expressed hnRNP F is involved in a neural-specific pre-mRNA splicing event. Genes Dev 9:2659-2671 Michelotti EF, Michelotti GA, Aronsohn AI, Levens D (1996) Heterogeneous nuclear ribonucleoprotein K is a transcription factor. Mol Cell Biol 16:2350-2360 Milone J, Wilusz J, Bellofatto V (2004) Characterization of deadenylation in trypanosome extracts and its inhibition by poly(A)-binding protein Pab1p. Rna 10:448-457 Min H, Chan RC, Black DL (1995) The generally expressed hnRNP F is involved in a neural-specific pre-mRNA splicing event. Genes Dev 9:2659-2671 Minvielle-Sebastia L, Beyer K, Krecic AM, Hector RE, Swanson MS, Keller W (1998) Control of cleavage site selection during mRNA 3' end formation by a yeast hnRNP. Embo J 17:7454- 7468 Moore MJ, Proudfoot NJ (2009) Pre-mRNA processing reaches back to transcription and ahead to translation. Cell 136:688-700 References 137 Moreira A, Takagaki Y, Brackenridge S, Wollerton M, Manley JL, Proudfoot NJ (1998) The upstream sequence element of the C2 complement poly(A) signal activates mRNA 3' end formation by two distinct mechanisms. Genes Dev 12:2522- 2534 Mouland AJ, Xu H, Cui H, Krueger W, Munro TP, Prasol M, Mercier J, Rekosh D, Smith R, Barbarese E, Cohen EA, Carson JH (2001) RNA trafficking signals in human immunodeficiency virus type 1. Mol Cell Biol 21:2133-2143 Mourelatos Z, Abel L, Yong J, Kataoka N, Dreyfuss G (2001) SMN interacts with a novel family of hnRNP and spliceosomal proteins. Embo J 20:5443-5452 Muhlrad D, Parker R (2005) The yeast EDC1 mRNA undergoes deadenylation-independent decapping stimulated by Not2p, Not4p, and Not5p. Embo J 24:1033-1045 Muto Y, Yokoyama S (2012) Structural insight into RNA recognition motifs: versatile molecular Lego building blocks for biological systems. WIREs RNA 2012, 3:229–246. doi: 10.1002/wrna.1107 3:229-246 Myer VE, Fan XC, Steitz JA (1997) Identification of HuR as a protein implicated in AUUUA-mediated mRNA decay. Embo J 16:2130-2139 Nagai K, Oubridge C, Jessen TH, Li J, Evans PR (1990) Crystal structure of the RNA-binding domain of the U1 small nuclear ribonucleoprotein A. Nature 348:515-520 Oberstrass FC, Auweter SD, Erat M, Hargous Y, Henning A, Wenter P, Reymond L, Amir-Ahmady B, Pitsch S, Black DL, Allain FH (2005) Structure of PTB bound to RNA: specific binding and implications for splicing regulation. Science 309:2054-2057