RNA Helicases in 60S Ribosomal Subunit Biogenesis in Saccharomyces cerevisiae

DOCTORAL THESIS 2020 Julia Mª Contreras Fernández Universidad de Sevilla RNA Helicases in 60S Ribosomal Subunit Biogenesis in Saccharomyces cerevisiae Departamento de Genética Universidad de Sevilla RNA Helicases in 60S Ribosomal Subunit Biogenesis in Saccharomyces cerevisiae Memoria presentada para aspirar al grado de Doctora en Biología V.o B.o: El director de la Tesis 1 V.o B.o: El director de la Tesis 2 Eduardo Villalobo Polo Jesús de la Cruz Díaz La doctoranda Julia Mª Contreras Fernández This work was funded by a contract associated to the grant BFU2013-42958-P and a fellowship from the University of Seville (PPI2015-II.2). A short-term fellowships from the University of Seville (VIPPIT-2017-II.2 and VIPPIT-2018-EBRV) is also acknowledges. Un científico en su laboratorio no es sólo un técnico: es también un niño colocado ante fenómenos naturales que le impresionan como un cuento de hadas. Marie Curie A mis padres, siempre. INDEX 1.- Summary ……………………………………………………………………………………… 1-9 2.- Introduction • Ribosomes in Saccharomyces cerevisiae …………………………………. 10-12 • Ribosome biogenesis in S. cerevisiae ▪ Pre-ribosomal RNA synthesis and processing ………………. 13-15 ▪ Pre-ribosomal RNA modification …………………………………. 15 ▪ Assembly of ribosomal particles - Assembly of the pre-40S r-particles ………………………… 17-19 - Assembly of the pre-60S r-particles ………………………… 20-22 • RNA helicases ▪ RNA helicases classification …………………………………………… 22-24 ▪ DEAD-box helicases ………………………………………………………. 24-28 ▪ 3D structure of DEAD-BOX RNA helicases ……………………… 28-29 ▪ Biochemical properties of DEAD-box RNA helicases ……… 29-30 ▪ Helicases in ribosome biogenesis ………………………………….. 31-32 ▪ Dbp6 and Dbp9 in ribosome biogenesis. Complex containing Dbp6, Dbp9, Nop8, Rsa3, Rsa1 …………………………………………………. 33 ▪ Dbp6 and Dbp9 in ribosome biogenesis. Complex containing Npa1, Npa2, Dbp6, Dbp9, Nop8, Rsa3 and Rsa1 ………………………. 34-37 ▪ Dbp7 in ribosome biogenesis …………………………………………. 37 3.- Objectives ………………………………………………………………………………………. 38 4.- Materials and methods …………………………………………………………………… 39-60 5.- Results • In silico analysis of Dbp7 ……………………………………………………………. 64-65 • Phenotypic analysis of dbp7 point mutants ……………………………….. 65-68 • Phenotypic analysis of truncated dbp7 alleles in the amino- and carboxylterminal extensions of the Dbp7 protein …………………………………… 68-76 • Phenotypic analysis of Dbp7 nuclear localization signal ……………. 76-83 • In vitro properties of Dbp7 ▪ Expression and purification of wild-type Dbp7, Dbp7[K197A], and Dbp7[R553A] proteins ……………………………………………………. 84-86 ▪ ATPase activity of Dbp7: the colorimetric assays ……………. 87-97 ▪ ATPase activity of Dbp7: the radioactive assay ………………. 97-99 ▪ Determination of the Michaelis-Menten kinetics parameters for recombinant wild-type and mutant Dbp7 protein …………. 99-105 ▪ ATPase activity of Dbp7[R553A] ……………………………………. 105-106 ▪ RNA binding activity of the recombinant wild-type and mutant Dbp7 in vitro …………………………………………………………………… 106-107 ▪ Helicase activity of the recombinant wild-type Dbp7 and Dbp7[K197A] in vitro ………………………………………………………. 107-112 • In silico analysis of Dbp9 ……………………………………………………………. 113 • In vitro properties of Dbp9 ▪ Expression and purification of recombinant wild-type and [K68A] mutant variant Dbp9 ………………………………………………………. 114-117 ▪ ATPase activity of Dbp9: colorimetric assay ……………………. 117-120 ▪ ATPase activity of wild-type Dbp9 and Dbp9[K68A] proteins: radioactive assays …………………………………………………………… 120-122 ▪ RNA binding and unwinding activities of the recombinant wildtype Dbp9 protein in vitro ………………………………………………… 123 • Role of Dbp7 in the proper interaction with pre-ribosomal particles of a subset of snoRNAs …………………………………………………………………… 123-128 6.- Discussion …………………………………………………………………………………….. 129-137 7.- Conclusions …………………………………………………………………………………… 138 7.- Bibliography ………………………………………………………………………………….. 139-147 8.- Acknowledgments …………………………………………………………………………. 148-152 6 ABSTRACT función de algunas proteínas ribosómicas y/o factores de ensamblaje. También, este proceso ha sido relacionado con patologías autoinmunes y de modo directo con numerosos tipos de cáncer e infecciones víricas. Las helicasas de RNA dependientes de ATP están entre los factores de ensamblaje que favorecen el proceso de biogénesis de ribosomas. En sentido bioquímico, estas enzimas llevan a cabo el desenrollamiento de dúplices de RNA utilizando para ello la energía liberada por la hidrólisis de ATP. Las helicasas de RNA se clasifican atendiendo a su secuencia y propiedades estructurales en superfamilias (SF1-SF5) y éstas a su vez en subfamilias. De especial interés para nuestro trabajo son las helicasas de RNA pertenecientes a la subfamilia denominada de las helicasas DEAD-box, llamadas así por la presencia de los aminoácidos aspártico, glutámico, alanina y aspártico, en ese orden, en un lugar específico de sus secuencias, los cuales se involucran en la unión e hidrólisis de ATP. Las helicasas DEAD-box están formadas por dos dominios estructurales, cada uno similar a la proteína RecA que, a su vez, contienen una serie de motivos de secuencia altamente conservados. Estos motivos se encuentran implicados en la unión e hidrólisis de ATP, en la unión y desenrollamiento de los ácidos nucleicos o en la coordinación entre ambas reacciones. A día de hoy, se han identificado hasta 19 helicasas de la subfamilia DEAD-box implicadas en la biogénesis de ribosomas citoplásmicos. Sin embargo, la función exacta que desempeña la mayoría de estas enzimas en el proceso no se conoce con claridad. Por ello, el objetivo de este trabajo ha consistido en la caracterización bioquímica y funcional de dos helicasas relacionadas funcionalmente con la biogénesis de subunidades ribosómicas 60S, las llamadas Dbp7 y Dbp9. Ambas proteínas son necesarias para la biogénesis de subunidades 60S debido a su papel durante el procesamiento de los pre-rRNAs tempranos 27SA y, por tanto, para la acumulación de los rRNAs maduros 25S y 5.8S. Sin embargo, su función concreta en este proceso es aún desconocida. Para cumplir estos objetivos hemos acometido una caracterización funcional de la proteína Dbp7 in vivo. Los motivos conservados de esta proteína, así como sus dos largas 7 ABSTRACT extensiones N- and C- terminales se identificaron en su secuencia aminoacídica. En la extensión N-terminal se identificó una posible señal de localización nuclear (NLS). Se realizaron distintos mutantes para el estudio. Los resultados de los análisis de crecimiento y perfiles de polisomas en mutantes puntuales en los motivos I y VI, implicados en unión e hidrólisis de ATP, demostraron que estos motivos cumplen un papel importante en la función de Dbp7, puesto que mostraron un retraso en el crecimiento debido, posiblemente, a una disminución en los niveles de subunidades 60S. Resultados similares fueron obtenidos en mutantes de deleción de regiones concretas de las extensiones N- and C-terminales de la proteína. Los defectos observados no se debieron a una deslocalización de Dbp7, puesto que en todos los mutantes con deleciones en los extremos N- and C- terminales, la proteína se encontraba asociada a complejos de alto peso molecular, probablemente partículas ribosómicas pre-60S. Por otra parte, mediante epifluorescencia hemos comprobado que la posible NLS localizada en el extremo N-terminal de Dbp7 es realmente una secuencia de localización nuclear verdadera; no obstante, el mutante carente de esta NLS crece de manera similar a como lo hace una estirpe silvestre, siendo la proteína capaz aún de alcanzar el núcleo; por tanto, Dbp7 debe presentar otras NLS redundantes no identificables en un análisis de secuencia o importarse al núcleo a través de la unión a otra proteína que la transfiera a esa localización. Se ha demostrado que Dbp7 está asociada funcionalmente a un complejo proteico denominado complejo Npa1 y que está formado por las proteínas Dbp6, Nop8, Rsa3, Npa1 y Npa2. Npa1 se une a posiciones específicas de rRNA en los dominios I y VI del rRNA 25S por lo que se ha propuesto que el complejo Npa1 participa en la restructuración de estos dominios durante la maduración temprana nucleolar de subunidades 60S, proceso que implica el ensamblaje de la proteína ribosómica uL3. Dbp7 se encuentra genéticamente asociada a este complejo. En estas reacciones de reestructuración participan algunos snoRNAs, entre ellos snR190, cuyas secuencias dianas de rRNA son vecinas a la región donde Npa1 se une, y al que Npa1 también se une directamente. Nuestro grupo ha identificado que mutaciones en el rRNA 25S que 8 ABSTRACT interfieren con la unión de snR190 suprimen parcialmente los defectos de crecimiento de un mutante dbp7 nulo. En este trabajo, hemos generado una estirpe doble mutante dbp7∆ snR190 que recapitula la supresión del defecto de crecimiento de la mutación dbp7∆. En consonancia con este resultado, snR190, entre otros snoRNAs, queda retenido en partículas pre-ribosómicas en ausencia de Dbp7, fenómeno que no ocurre en una estirpe silvestre isogénica. Así, nuestro trabajo sugiere que Dbp7, probablemente ejerciendo su función como helicasa de RNA, participa en la liberación de al menos snR190 de las partículas pre-ribosómicas 60S tempranas en progresión de su maduración. En una segunda parte de esta Tesis, hemos estudiado las características enzimáticas de Dbp7 in vitro, más concretamente de sus posibles actividades ATPasa, de unión a RNA y helicasa. Haciendo uso de ensayos enzimáticos colorimétricos y radiactivos, hemos demostrado que una proteína Dbp7 recombinante expresada y purificada a partir de extractos de Escherichia coli presenta actividad ATPasa dependiente de RNA y de oligonucleótidos específicos de DNA (ej. poly dT de 30 nucleótidos de largo). Simultáneamente, hemos demostrado que Dbp7 se une a RNA y que presenta capacidad helicasa in vitro, pudiendo desenrollar dúplices de RNA tanto en la dirección 5´-3´como en la dirección 3´-5´. Esta actividad es dependiente de la hidrólisis de ATP. Hemos estudiado igualmente las actividades anteriores en mutantes de Dbp7 puntuales en dominios implicados en la unión e hidrólisis de ATP (Dbp7[K197A], Dbp7[R553A]). Sin embargo, ambos mutantes sorprendentemente mantiene las actividades de la proteína silvestre. Estos resultados se discuten en el contexto de los dominios estructurales y funcionales de Dbp7. Finalmente, en este trabajo, hemos iniciado la caracterización preliminar de las actividades bioquímicas de la helicasa de RNA Dbp9. De manera similar al trabajo realizado con Dbp7, hemos expresado y purificado la proteína Dbp9 recombinante a partir de extractos de E. coli e iniciado su caracterización bioquímica. Nuestros primeros ensayos sugieren que la proteína recombinante tiene actividad ATPasa dependiente de 9 ABSTRACT un oligonucleótido de DNA, pero ni tiene la capacidad de unirse a RNA ni actividad helicasa en las condiciones ensayadas. 2.INTRODUCTION INTRODUCTION 10 2.- INTRODUCTION Ribosomes are the macromolecular machines, made out of proteins and RNAs, that translate the mRNAs into proteins in all living beings. Structurally, all ribosomes are composed by two subunits (r-subunits), separable from each other, the so-called large and small subunit (LSU and SSU, respectively); these names are given by the difference in sedimentation coefficient, in sucrose density gradient, of each subunit. The size of each r-subunit is higher in eukaryotes than in prokaryotes since, on the one hand, the number of ribosomal proteins (r-proteins) is greater in eukaryotes and, on the other hand, the length of each ribosomal RNA (rRNA) is bigger in eukaryotes. The synthesis and assembly (biogenesis) of ribosome subunits take place for each subunit during a complex and tightly regulated process. Once each subunit is fully assembled, they bind to each other to yield a fully competent-to-translate ribosome (1). In a functional ribosome, the LSU catalyses the formation of the peptide bond between the peptidyltRNA and the nascent aminoacyl chain, while the SSU mediates the mRNA codon and tRNA anticodon pairing (1, 25, 69). The study of the ribosome biogenesis pathway has been performed in several eukaryotic organisms but so far, most of the information concerning this process comes from the budding yeast Saccharomyces cerevisiae. S. cerevisiae is a low complexity eukaryotic system. Its growth and cells dispersion are fast, and mutant isolation is easy. Besides all this, the genome of S. cerevisiae has been completely sequenced, and its biology is in constant revision. For all these advantages, S. cerevisiae was the eukaryotic organism selected to perform this work. 2.1.-Ribosomes in Saccharomyces cerevisiae In S. cerevisiae, the LSU and the SSU have sedimentation coefficients of 60S and 40S, respectively. These r-subunits build up an 80S ribosome. The LSU possesses three different rRNA species (25S, 5.8S and 5S) and 46 different ribosomal proteins (rproteins). The SSU possesses a single rRNA species (18S) and 33 different r-proteins (1). In addition to rRNAs and r-proteins, accessory proteins or trans-acting factors (from now INTRODUCTION 11 on factors), and small nucleolar RNAs (snoRNAs) bind to the ribosomal subunits transiently. The function of these factors could be involved in stablish or dissociate snoRNA-pre-rRNA base pairings, prepare the rRNA for an efficient nuclease attack or participate in specific modulations of rRNA, rRNA-protein or protein-protein interactions (71). Most snoRNAs participates in rRNA modifications that generally stabilize the ribosome structure, while other serve as RNA chaperones for rRNA folding (82). The 18S rRNA of the SSU is 1869 nucleotides long and is folded in four conserved domains, which constitute the secondary structure of the rRNA. These conserved domains are the 5´, central, 3´ major, and 3´ minor domains (Figure 1A). At the same time, the domains fold and together with the r-proteins form the tertiary structures of the SSU: the head (containing for example: uS2, uS3, uS7, uS9, eS10, uS15, uS17 and eS19) and the body, which has the shoulder (eS30), the platform (for example: eS6, eS8, uS17, uS19, and uS13), the foot (uS10) the decoding centre cleft (for example: uS3, eS4, uS7,eS7 and eS12) and the beak (eS31, eS12 and eS10) (76, 90) (75) (Figure 1B). These tertiary structures function as scaffold to which accessory factors bind and release until giving rise the ribosomal 40S particles (44). In the case of the LSU, the 25S rRNA is 3396 nucleotides long and is folded giving rise to six conserved domains (I-VI), each beginning by a root helix (37) (Figure 1C). The 5.8S rRNA (156 nucleotides) is bound by base pairing with the 25S rRNA in the domain I. As occurring for SSU, these domains fold and with the binding of the r-proteins yield the remarkable functional sites of the LSU: the central protuberance (CP) (containing for example: uL18 and uL5), where is localized the 5S rRNA (121 nucleotides) (91), the P-stalk (P0, P1 and P2), involved in recruitment and activation of translation factors, the L1 stalk (uL1) implicated in tRNA movements during translocation through the ribosome (83, 86), the peptidyl transferase centre (PTC) (containing for example: uL2, uL3, eL6, eL14, uL16, eL40, eL41 and eL43), which is responsible for peptide bond formation, and the polypeptide exit tunnel (PET) (for example: uL4, eL22, uL14, eL29, eL31 and eL39) (6, 59), from where the nascent growing polypeptide chain leaves the ribosome (Figure 1D) (40, 89). INTRODUCTION 12 A) B) C) D) Figure 1. Secondary and tertiary structures of the SSU and LSU. (A) Secondary structure of 18S rRNA and (B) 25S and 5.8S rRNAs from S. cerevisiae. The four domains of 18S rRNA and the six domains of 25S rRNA are showed in different colours. The 5.8S rRNA is showed in black and is bound by base pairing with the domain I of 25S rRNA. The 5S rRNA is not represented. (C) Crystal structure of the SSU and (D) LSU of the ribosome from S. cerevisiae. In the SSU, the head, platform, body and beak are indicated. In the LSU, the central protuberance (CP) the L1 and P-stalks and are indicated. The peptidyl transferase centre (PTC) and the polypeptide exit tunnel (PET) are not highlighted (89). INTRODUCTION 13 2.2.-Ribosome biogenesis in S. cerevisiae The ribosome biogenesis pathway is an integrated process, in which we can highlight three principal steps: pre-rRNA synthesis and processing, pre-rRNA modifications, and rsubunits synthesis and assembly. The rRNA is transcribed in the nucleolus and concomitantly, it undergoes process of folding and nucleotide modification. The spacer sequences are removed, and r-proteins are bound to rRNA. Then, the rRNA travel to the nucleoplasm, where different factors and r-proteins bind to rRNA and the ribosome preparticles are assembled. The pre-ribosomal particles do not contain the same r-proteins and conformation as the mature ribosomal subunits. When the pre-ribosomal particles are competent for the export, they travel to the cytoplasm where the pre-ribosomal particles mature with the acquisition of its definitive conformation, the binding of the last r-proteins and the elimination of remaining factors (89). 2.2.1.-Pre-ribosomal RNA synthesis and processing The rDNA containing the information to yield the different rRNA species is comprised in the chromosome XII and it is composed by approximately 150 rDNA repetitions in tandem localized in the nucleolus (Figure 2) (60). A single-copy rDNA is a fragment of 9.1 kb that codes for the 35S and the 5S pre-rRNA. The 35S pre-rRNA is the largest detectable rRNA precursor and contains the 18S, 25S, and 5.8S rRNAs. The 35S precursor also contains two internal transcribed spacers (ITS1 and ITS2) separating the three rRNA species, and two external transcribed spacers (5´ETS and 3´ETS). Within the spacer regions are located different cleavage sites, which are the target of distinct endo and exonucleases to generate the rRNA intermediates. Two non-transcribed regions, called NTS1 and NTS2 separate the 35S and 5S transcriptional units. Several elements acting in cis, such as the origin of DNA replication and the replication fork block site, are present in the NTS regions. The 35S and the 5S are transcribed by the RNA polymerase I (RNAP I) and RNA polymerase III (RNAP III), respectively. In fast growing S. cerevisiae cells, the nascent rRNA transcripts are processed co-transcriptionally (43). However, a less common post- INTRODUCTION 14 transcriptional pathway also exists (43, 64). With the purpose of simplify the explanation, only the post-transcriptionally pathway will be explained here. Figure 2. Structure of the rDNA region in S. cerevisiae. A 9.1 kb repetition unit is expanded to show the disposition of the 35S rRNA and the 5S rRNA coding regions and the non-transcribed spacers, NTS1 and NTS2. The transcription and processing of the 35S pre-rRNA give rise to the mature 18S, 5.8S and 5S rRNAs. The 35S rRNA precursor is flanked by two external spacer sequences (5’ETS and 3’ ETS) and contains two internal spacer sequences (ITS1 and ITS2). The 35S rRNA and 5S rRNA pre-RNAs are separated by the nontranscribed spacers, NTS1 and NTS2. The NTS region is amplified and positions of terminators 1 and 2 (T1 and T2), RFB (replication fork blocking) and ARS (autonomous replication sequence) are indicated. P and T indicate the promoter and terminator of the transcript. Modified from 60. During pre-rRNA transcription, the endonucleolytic cleavage by Rnt1 of the nascent prerRNA in the 3′ETS at site B0 generates the 35S pre-rRNA (24). The nascent 35S pre-rRNA is cleaved at the A0 and A1 sites by the Utp24 nuclease (64, 88), giving rise to the 33S and to 32S pre-rRNA intermediaries, respectively. The 32S intermediary is further cleaved in the A2 site by the endonuclease Rcl1 (41). This latter cleavage generates the 20S and the 27SA2 intermediaries (85). The 20S pre-rRNA intermediary is processed in the D site of the 3’ end by the Nob1 endonuclease (23, 47). In this step, the mature 18S rRNA is directly generated. During the ulterior processing of the 27SA2 intermediary, two different Cr.XII INTRODUCTION 21 binding of uL16 is a rate-limiting step for cytoplasmic pre-60S ribosomal particles maturation (Figure 7) (53). Figure 6. Nuclear pre-60S r-subunit maturation model including the AFs. The particles containing Nog2 and Rix1 are intermediate pre-60S particles. In the Nog2 particle the ITS2 spacer is present still and the 5S rRNA, that will form the CP, is in the premature conformation. In the Rix particle the ITS2 spacer is removed. The exit of Rpf2 and Rrs1 and the entrance of Rea1 promote the 5S rRNA rotation into the mature conformation. During this step the binding of the export factor Nmd3 takes place. Nmd3 recruitment makes the pre-60S particle competent for the export to the cytoplasm. Once in the cytoplasm Nmd3 and the remaining AFs are removed (64). Figure 7. Cytoplasmic pre-60S r-subunit maturation model. The last quality control steps take place in the cytoplasm. The remaining AFs in the pre-60S rparticles are removed in a modular way. The exit of these AFs gives rise to the mature LSU (64). At this point, it has been clear the importance of the AFs in the ribosome biogenesis process. Into the AFs group, RNA helicases are a group of important proteins involved in ribosome biogenesis. To date, it is known that 19 RNA helicases participate in yeast INTRODUCTION 22 ribosome biogenesis, most of them harbouring counterparts in higher eukaryotes (71). However, many of them remain poorly characterized. 2.3.-RNA helicases In the literature, the RNA helicases are defined as enzymes that catalyse ATP-dependent unwinding of RNA duplexes (dsRNA). Today, it is known that besides to open dsRNA, the RNA helicases are involved in others biochemical process, such as RNA clamping, dsRNA destabilization, protein displacement from RNA and strand annealing (50, 71). For this reason, the term RNA helicases could be incomplete to define this category of enzymes. 2.3.1.-RNA helicases classification RNA helicases are ubiquitous enzymes present in all forms of cellular life. Even many viruses can codify for these enzymes (22). RNA helicases can be classified in different superfamilies (SFs), according to the pioneering sequence analysis of Gorbalenya and Kooning (1993) (29), and to more recent structural and functional analysis of Jankowsky and co-workers (22). So far, five superfamilies (SFs) of RNA helicases have been proposed (from SF1 to SF5) (Figure 8). RNA helicases belonging to SF1 and SF2 seems to be active as monomers or dimers. Conversely, SF3, SF4 and SF5 are active as oligomers (mostly hexamers) enzymes. The SF3, SF4 and SF5 are encoded principally in genomes of viruses and bacteria (49). Helicases from the SF1 and SF2 superfamilies all contain a central core composed by two RecA-like identical domains (49). These domains resemble the structure of the bacterial recombination protein recombinase A (Rec A). The RecA-like domains are characterized by the presence of a few conserved motifs. SF1 and SF2 superfamilies are very similar with respect to the conserved motifs (15). The SF2 is formed by five different sub-families: DEAD-box helicases, DEAH-RHA helicases, RIG-I like, Ski2-like and NS3/NPH-II proteins. INTRODUCTION 23 Figure 8. Structural classification of RNA helicases families. In each group, a structure of a representative protein from Protein Data Bank (PDB) is showed. SF1 and SF2 are formed by monomeric proteins. SF2 is very varied in its composition, and the only common characteristics in all the sub-families is the presence of a central core formed by two RecA-like domains. SF3, SF4 and SF5 are composed by hexameric proteins, basically of viral origin (49). However, according to their action mechanism and catalytic properties, the SF2 can be further classified as processive and non-processive RNA helicases. Processive RNA helicases can perform multiple rounds of catalysis before substrate dissociation. Processive RNA helicases are able to unwind RNA duplexes by binding to a singlestranded region of the RNA and translocating along the dsRNA molecule using the ATP hydrolysis energy to open the structure (Figure 9A). After substrate release and ATP hydrolysis, the enzyme adopts an open conformation and the RecA2 domain form new contacts with the RNA substrate further downstream of the direction of translocation. The DEAH-box family is mostly composed by processive helicases. However, nonprocessive RNA helicases only use one substrate molecule by catalysis round (Figure 9B). DEAD-box proteins are non-processive enzymes that join to the RNA and open the duplexes by local strand separation. Thus, in the case of DEAH-box RNA helicases the INTRODUCTION 24 ATP hydrolysis drives the unwinding; however, in the DEAD-box helicases conversion of ATP to ADP generates the substrate release but not strand separation (8, 51). A) B) Figure 9. Schematic representation of the possible action mechanisms of RNA helicases from the SF2 superfamily. (A) Processive helicase, which can bind to a single-stranded region of the RNA and unwind the double RNA strand by translocation through it. In this case, the ATP hydrolysis is used to open the structure. (B) Non-processive helicase, which can bind to the double strand RNA molecule and unwind it by local strand separation. In this case the ATP hydrolysis is used to recycle the inorganic phosphate and to return to the original conformation. ATP is showed as a red circle and ADP is showed as a yellow circle. (49 and modified from 81) 2.3.2.-DEAD-box helicases The DEAD-box helicases have the Rec A-like domains connected by a flexible short linker. In eukaryotes, the RecA-like domains are composed by a set of conserved motifs with invariable consensus sequences. These conserved motifs are involved in ATP-binding and hydrolysis, nucleic acid binding, and coordination between both activities. In most of the cases, the RecA-like domains are flanked by both a N- and a C- terminal INTRODUCTION 25 extensions. It is generally thought that these extensions interact with other proteins or with RNA targets (Figure 10) (15, 49, 50, 71). Figure 10. General structural organization of DEAD-box RNA helicases. The central core of the DEAD-box helicases is formed by the RecA-like 1 and RecA like 2 domains. These domains are connected by a flexible short linker. The central core harbours a set of conserved motifs. The motifs are showed in different colours depending of its biochemical function: In red; involved in ATP binding and hydrolysis, in green; involved in nucleic acid binding and in blue involved in coordination between ATP and nucleic acid binding. The central core can be flanked by N- and C- terminal extensions of variable length (71). The conserved motifs are distributed in the two RecA-like domains forming an RNA substrate tunnel and a catalytic binding pocket for ATP. The ATP binding pocket contains also a divalent ion (e.g. Mg2+) and a water molecule (84). The N-terminal RecA-like domain or domain 1 contains the ATP binding motifs F, Q, I, and II, the ATP hydrolysis motif III, and the RNA binding motifs Ia and Ib. The C-terminal RecA-like domain or domain 2 bears the RNA binding motifs IV and V and the motif VI that is able to coordinate ATPase and unwinding activities (15). The motifs are mentioned from the N- to the C-terminal end of a general RNA helicase (7, 15, 80). - Motif F: It is a highly conserved phenylalanine. It is present in the 88% of the DEAD- box RNA helicases. This phenylalanine is localized generally 16 amino acids upstream of the Q motif. It seems to function as a molecular switch to regulate ATP binding and hydrolysis. - Motif Q: GaccPoh1Q (a, aromatic residue, c, charged amino acid, h, hydrophobic amino acid, o, alcohol and 1 aliphatic residue). The motif contains an invariant INTRODUCTION 26 glutamine residue that is present in the 99% of the DEAD-box RNA helicases. It is localized generally 16 amino acids upstream of motif I. This glutamine establishes direct interaction with the bound nucleotide and it is necessary for adenine recognition and binding. - Motif I (Walker A): AxTGoGKT (x, any amino acid). It is involved in ATP binding and hydrolysis by forming hydrogen bonds with the pyrophosphates of the nucleotide and by interacting with the coordinated Mg2+ ion. - Motif Ia: PTRELA. It participates in the structural rearrangements that occur upon ATP binding and hydrolysis. It is mainly involved in RNA binding. - Motif Ib: TPGRI. It is involved in RNA binding. - GG doublet: GG. It is proposed to participate in protein-protein interactions. - Motif II (Walker B): DEAD. It gives the name to the family. This motif is involved in ATP binding and hydrolysis through the coordination of the ion Mg2+ by the first aspartate and the water molecule by the glutamate. - Motif III. SAT. It establishes a communication between ATP-and RNA-binding sites via interaction with the -phosphate of the ATP. - Motif IV: 1IFhxT+cx. (+, positively charged amino acid). It contributes to substrate binding through interactions with the RNA. - Motif V: TDVuARGID. (u: A, G). It regulates the hydrolysis of ATP. This motif is also involved in RNA binding. - Motif Va: QxxR. It establishes a communication between ATP-and RNA-binding sites. - Motif VI. HRIGRTGR. It is involved in ATP binding and hydrolysis. Contains two important arginine residues (first and second) that could bind the -phosphate of the ATP. The DEAD-box helicases function in all processes that involve RNA: transcription, premRNA splicing, RNA editing, RNA interference, RNA nucleo-cytoplasmic export, translation initiation and termination, RNA degradation, RNA quality control and in ribosome biogenesis, among them (50, 71, 78). INTRODUCTION 27 The conformational ATP-dependent changes of DEAD-box RNA helicases during its catalytic cycle is possible due to the linker that bind both RecA-like domains. This flexible short linker allows the change of orientation of one domain respect to the other. In this way, the RecA-like domains can adopt two different conformations. One of them is the closed active conformation, in which both motifs are in contact forming a cleft in the middle of motifs. To reach this conformation is necessary that both substrates join to the RecA-like domains 1 and 2. This conformation allows the formation of the ATP binding pocket and the RNA binding site. The ATP and RNA binding sites are arranged in perpendicular position in the cleft formed by the two RecA-like motifs (68). However, in the opened inactive conformation both RecA-like domains are separated, and this fact impairs the ATP and RNA sites assembly. The DEAD-box RNA helicases interaction with the RNA is always sequence independent as the protein bind to the phosphate backbone of the RNA molecule (19). The catalytic activity responds to a model in which the double strand RNA separation is completed before ATP hydrolysis and depending of the length and stability of the RNA duplex (35). The unwinding process is limited to short RNA duplexes (10-15 bp) and it is accomplished in a single cycle of ATP-dependent conformational changes (19, 28, 36). The binding of ATP and dsRNA causes the closure step and the formation of the functional ATP and RNA binding sites. Then, the dsRNA is unwound; one of the strands remains bound to the enzyme while the other is released from the complex. After that, the ATP is hydrolysed to ADP and inorganic phosphate (Pi). This fact promotes the recycling of the enzyme and the release of the second bound RNA strand (Figure 11) (3, 28, 49). INTRODUCTION 28 Figure 11. Structural changes in DEAD-box helicases during dsRNA unwinding. In the inactive or open conformation both domains are separated. The binding of ATP to the domain 1 and the dsRNA to the domain 2 promote the joining of both domains and induce the active or closed conformation. One of the dsRNA strands is released from the complex while the other remains bound to the enzyme. The ATP hydrolysis promotes the release of the second RNA strand and the return of the enzyme to the open conformation (54). 2.3.3.-3D structure of DEAD-BOX RNA helicases The detailed understanding of how the RNA helicases convert chemical energy into RNA remodelling need the knowledge about the structure of both domains in active or closed conformation and in inactive or opened conformation. Today, crystal structures of tandem domains are available for several DEAD-box helicases (eIF4AI, eIF4AII, Dbp4, Rok1, etc…). A comparative analysis of these structures showed not only the specific features of the isoforms but also nucleotide specific positioning of flexible elements that are common to several proteins. Considering the DEAD-box RNA helicase model eIF4A, the tertiary structure of these proteins can be defined by a ‘‘dumbbell’’ molecule with two globular domains connected by an extended (and in all probability, flexible) linker (Figure 12) (9). The superposition of the crystal structures revealed the location of others flexible regions. In general, regions of high sequence conservation (the conserved motifs) contribute to the binding sites for nucleotide and for RNA, and these sites coincide with the highest structural similarity. Concerning to the surface charge distribution of the DEAD-domain structure, there is a conserved patch that forms the nucleotide binding site and part of the RNA binding site. This conserved patch is negatively charged and formed a channel that extends from the α-helices 8-10 to the Mg2+ binding site. However, the RNA binding cleft is positively charged in all DEAD domains. Referring to ATP binding site, the P-loop (motif I) and motif II coordinate the nucleotide phosphates. Superposition of the DEAD-domains showed that the conformation of the P-loop (motif I) is determined by the nucleotide phosphates and longer phosphate tails result in a more open loop. Motifs Ia, Ib and II seem unaffected by the state of ATP hydrolysis, and their conformations remain unchanged even in the crystal structures in which the nucleotide binding site is not occupied. INTRODUCTION 29 The helicase domain is composed fundamentally, by the domains V and VI. Apparently, four side chains are of importance: The aspartate of motif V coordinates the O3´of the ribose. The second arginine side chain of motif VI interacts with the -phosphate. The third arginine, which is also the putative arginine finger during ATP hydrolysis, coordinates all three ATP phosphates. It is seeming that the unblocking of the RNA binding site is related with α helix 8. When the two RecA-like domains are separated, α helix 8 would block the RNA binding site. It is proposed that the ATP facilitates the RNA binding since it pushes α helix 8 out of the RNA site (78). RecA-like 1 RecA-like 2 Figure 12. Structure of eIF4AIII in complex with single RNA (yellow) and ADPNP (green). The DEAD-box RNA helicases are formed by two RecA-like domains connected by a flexible short linker. Both domains bear conserved sequence motifs. In red are indicated motifs involved in ATP binding and hydrolysis, in blue are indicated motifs involved in RNA binding and in purple are indicated motifs involved in coupling ATP hydrolysis and duplex unwinding (2). 2.3.4.- Biochemical properties of DEAD-box RNA helicases So far, several DEAD-box RNA helicases involved in ribosome biogenesis have showed RNA dependent ATPase activity and ATP dependent helicase activity. Study of these properties, construction of mutants in conserved motifs and analysis of its phenotypes have allowed determine the function of each conserved motif in the DEAD-box RNA helicases subfamily. Into the category of DEAD-box helicases biochemical data have INTRODUCTION 30 been found for Ded1, Has1, Rok1 and Dbp9. Ded1 are involved in translation initiation and could has a role in ribosome binding in S. cerevisiae. The ATPase activity of Ded1 is stimulated by natural RNAs, such as rRNA and polyadenylated mRNA in vitro. Concerning to the helicase activity, a high concentration of protein respects the template is necessary to open the dsRNA. The unwinding of RNA only is possible in presence of ATP; thus, ATP hydrolysis is necessary for driving the unwinding reaction. The mutant DAAD in the motif II of Ded1 is able to bind RNA as the wild-type, however, the ATPase activity is 60 times lower than in the wild-type and is deficient for the unwinding activity. Another important characteristic of Ded1 is that ATP binding and hydrolysis increase the relative affinities of Ded1 for ssRNA versus dsRNA. All the data obtained about Ded1 make it a good candidate to be involved in the unwinding of the cap-proximal secondary structures (35). Has1 are involved in SSU and LSU biogenesis. This protein has RNA dependent-ATPase activity in vitro and it can unwind DNA/RNA heteroduplexes in an ATP-dependent manner. Has1 is able to unwind heteroduplexes in 5´-3´or 3´- 5´directions, showing preferences by the 5´-3´direction. Mutation in the motif I or motif VI decreases the ATPase activity of Has1. However, mutations in the motif III affect the unwinding activity of the protein (70). Rok1 is, as Has1, involved in ribosome biogenesis. The ATPase activity of Rok1 is independent of ssRNA. Mutations in the motif I had different phenotype depending of the changed residue. K172L and G166D showed a drastic defect in ATP hydrolysis, however, the mutation K172R hardly affects the ATPase activity of Rok1. Referring to the helicase activity Rok1 did not present the capacity to open dsRNA duplexes (61). Finally, Dbp9 is an RNA helicase involved in LSU biogenesis. In contrast to other members the ATPase activity of Dbp9 seems to be stimulated by DNA and inhibited by RNA in vitro. Concerning to the helicase activity, Dbp9 exhibits DNA-RNA and DNA-DNA helicase activity dependent of ATP, but also unwinds RNA in absence of an NTP. Mutants in the motif VI of Dbp9 did not present ATPase activity neither or helicase activity (39). When it was possible, the kinetics parameters of these RNA helicases were measured. The KM values oscillate from 300-400 µM in all cases. INTRODUCTION 37 The reads that mapped to snR190 are the most abundant by far, but other reads less abundant mapped to snR10, snR42, snR5 and snR61. Altogether, these data are concordant with the hypothesis that the Npa1 complex is involved in pre-rRNA folding events within the earliest pre-60S particle modulating interactions between snoRNAs and 25S rRNA into pre-60S particles (37). 2.3.8.-Dbp7 in ribosome biogenesis Dbp7 (YKR024C) belong to the DEAD-box subfamily of RNA helicases and is involved in ribosome biogenesis; however, its precise function is still unknown (18). Dbp7 is a nucleolar protein of 742 amino acids (pI: 9.33 and predicted mass 84 KDa). The gene that encodes for Dbp7 is located in the chromosome XI. All conserved amino acid motifs of the DEAD-box protein family are present in Dbp7 (Figure 16). Besides this, the RecA-like domains of Dbp7 are flanked by two longs N- and C- terminal extensions of approximately 200 amino acids each (18). Figure 16. Representation of the RecA-like domains of Dbp7. The conserved motifs are shown. Each motif is indicated by a different color. In blue, motifs involved in ATP binding and hydrolysis, in red, motifs involved in nucleic acid binding and in green, motifs involved in the coordination between both functions. The sequence of each motif is showed inside rectangles. The name of each domain is showed over the rectangles. The N- and C-terminal extensions are indicated by a line with the letter N or C, respectively. Dbp7 is a non-essential protein in S. cerevisiae. However, the deletion of DBP7 generated a deficit of 60S ribosomal subunit observed due to a delay in the processing of 27SA and the derived precursors that lead the mature 25S and 5.8S rRNA. The involvement of Dbp7 in the 60S ribosomal subunit metabolism makes it necessary for optimal vegetative growth of the cells. As shown in this Thesis book, the understanding of the function of Dbp7 in ribosome biogenesis is the main objective of this work. 3.OBJECTIVES 38 OBJECTIVES 3.- OBJECTIVES Ribosome biogenesis is a very important cellular process in cells. In eukaryotic organisms, many proteins and factors are involved in this proccess. Among these factors, RNA helicases belonging to DEAD-box family are of special relevance. These enzymes catalyse ATP-dependent unwinding of RNA duplexes (dsRNA). Some members of the DEAD family have been characterized, but their specific role in the process of ribosome biogenesis is barely known. In this work, we have studied RNA helicases involved in 60S ribosomal subunit biogenesis in S. cerevisiae . For this purpose, we raised the following objectives: - To study in vitro the catalytic properties of Dbp7 and Dbp9. - To study the in vivo function of Dbp7, including the role of N- and C-terminal extensions. - To dissect the functional environment of Dbp7 within pre-60S r-particles, including its relationship with selected snoRNPs. 4. MATERIALS AND METHODS 39 MATERIALS AND METHODS 4.-MATERIALS AND METHODS 4.1.-Microorganisms and growth conditions S. cerevisiae and E. coli strains used in this work are listed in the Table 1. Table 1. Yeast and bacterial strains used in this study. The relevant characteristics of the strain genotypes and the references that describe them are indicated. Strain Relevant genotype SOURCE W303-1A MATa/α ade 2-1/ade 2-1 his 3-11,15/his 3-11,15 leu 2-3, 112/ leu 2-3, 112 trp 1-1/trp 1-1 ura 3-1/ura 3-1 Thomas and Rothstein, 1989 YKL500 As W303-1B but NOP58-yEmCherry::natNT2 ade3::kanMX4 Pillet et al., 2015 JDY306 MATa/α dbp7:: HIS3MX6 /DBP7 Our lab collection snR190 cl 2 MATa/α ade 2-1/ade 2-1 his 3-11,15/his 3-11,15 leu 2-3, 112/ leu 2-3, 112 trp 1-1/trp 1-1 ura 3-1/ura 3-1, snR190 Henras et al, 2018 E. coli XL-1-Blue endA1 gyrA96 (nalR) thi-1 recA1 relA1 lac glnV44 F'[:: Tn10 proAB+lacIq Δ(lacZ)M15] hsdR17 (rK-mK+) Stratagene E. coli BL21 Codon Plus (DE3) -RIPL F– ompT hsdS (rB – mB –) dcm+ Tetr gal λ(DE3) endA Hte [argU proL Camr] [argU ileY leuW Step/Specr] Agilent The dbp7∆[YCplac22] strain was constructed by transformation of the dbp7∆[YCplac33- DBP7] strain with the plasmid YCplac22 (CEN, TRP1); transformants were counterselected on 5-FOA (5-fluoroorotic acid) containing plates (plasmid shuffling) to allow the growth of those clones that have lost the YCplac33-DBP7 plasmid. Some of the colonies 40 MATERIALS AND METHODS growing on this media were further transferred to SD-Trp plates. We also constructed a dbp7∆[YCplac22-HADBP7] strain (from now on HA-DBP7), using the same strategy. The Dbp7[K197A] or Dbp7[R553A] protein variants were performed by PCR of the ORF of DBP7 (with the correspondent point mutation) using Ycplac22-HA-DBP7 as a template. The ORF of DBP7 was expressed under the control of the cognate DBP7 promoter followed by a HA epitope. The N-terminally HA-tagged constructs were transformed in the strain dbp7∆[YCplac33-DBP7]. The transformants were selected on SD-Trp and then, the URA3-marked plasmid counter-selected on 5-FOA plates (plasmid shuffling). Some of the colonies growing on this media were further transferred to SD- Trp plates. The dbp7 truncated mutants were performed by PCR of the ORF of DBP7 (lacking the correspondent codons in the N- or C- terminal extension) using Ycplac22-HA-DBP7 as a template. The ORF of DBP7 was expressed under the control of the cognate DBP7 promoter followed by a HA epitope. The N-terminally HA-tagged constructs were transformed in the strain dbp7∆[YCplac33-DBP7]. The transformants were selected on SD-Trp and then, the URA3-marked plasmid counter-selected on 5-FOA plates (plasmid shuffling). Some of the colonies growing on this media were further transferred to SD- Trp plates. The Dbp7-N-terminal-EGFP strain was constructed by cloning the first 162 amino acids of Dbp7 (N-terminal extension), which contained the putative NLS in phase with three tandem GFP proteins, and expressed from the strong ADH1 promoter on a LEU2-marked pADH111 centromeric plasmid. In addition, we cloned in the same plasmid a variant of the N-terminal of Dbp7 lacking the NLS to construct the dbp7ΔNLS-N-terminal-EGFP strain. As a positive control we used the NLS of the large T antigen of the SV40 virus, which was also cloned in pADH111 plasmid. Each plasmid was transformed into the 41 MATERIALS AND METHODS YKL500 strain, which harboured a genomic functional copy of the gene coding the nucleolar reporter Nop58 tagged with the red fluorescent mCherry protein. For the HA-dbp7∆NLS mutant, a PCR of the ORF of DBP7 lacking the codons corresponding to the NLS was performed using Ycplac22-HA-DBP7 as a template. The ORF of DBP7 was expressed under the control of the cognate DBP7 promoter followed by a HA epitope. To use as a control, a full-length HA-Dbp7 protein was cloned into YCplac22. Upon transformation of these constructs into the dpb7∆[YCplac33-DBP7] strain, cells were subjected to plasmid shuffling on 5-FOA plates. Some of the colonies growing on this media were further transferred to SD-Trp plates. The proteins wild-type Dbp7-(GA)5-3xyEGFP, Dbp7∆NLS-(GA)5-3xyEGFP and Dbp7∆N162-(GA)5-3xyEGFP were constructed by fusion of the Dbp7 protein in the C- terminus to three variant GFPs (3xyEGFP) in tandem. These constructs were expressed from the cognate DBP7 promoter and cloned onto a YCplac111 plasmid. The plasmids, were transformed independently into the YKL500 strain. The snR190 strain was performed by the CRISPR-Cas9 system in the collaborative laboratory of Prof. Henry. This system included a couple of relevant nucleotides in the C box of snR190 that compromise its binding to rRNA. Thus, the sequence GGCCCUGAUGA corresponding to the first 11 nucleotides of snR190 were substituted by the sequence GGAACUAUUGA. To obtain a double mutant containing the non-functional version of the snR190 and lacking Dbp7, we crossed the strain snR190 with an isogenic dbp7∆[YCplc33-DBP7]. Diploids were selected and sporulated. After tetrad dissection, we confirmed that, as expected, the His3-marker that deleted-disrupted the DBP7 gene segregated 2+:2- in 42 MATERIALS AND METHODS complete tetrads. Then, we counter-selected the Ura3-marked YCplac33-DBP7 plasmid from different tetrads in plates containing 5-FOA. Finally, to accurately establish the genotypes of the spore-clones, we obtained total DNA for several complete tetrads and sequenced the genomic region comprising the coding sequence of the snR190. Unless otherwise indicated, all strains of S. cerevisiae were grown at 30 oC in rich YPD medium (1% yeast extract, 2% peptone) containing 2% glucose as carbon source or in synthetic minimal medium (SD) (0.15% yeast nitrogen base, 0.5% ammonium sulphate) supplemented with the appropriate nutritional requirements and containing 2% glucose as carbon source with or without 2% agar. E. coli was grown at 37 oC on LB broth medium (10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl) supplemented with 2% of agar and antibiotics when was required. The following antibiotics were used alone or in combination: ampicillin (100 g/mL), tetracycline (20 g/mL) and chloramphenicol (50 g/mL). For serial dilutions in solid medium, yeast cultures were grown until reaching an OD600 of 0.8. Subsequently, these cultures were diluted to an OD600 of 0.05, 1:5 serial dilutions were performed, and five microliters of each dilution was spotted onto the corresponding plates. The plates were incubated at the appropriate temperature. Growth curves were obtained by measuring the OD600 of cellular cultures at different incubation times and representing this OD600 versus time. To do so, cultures were previously grown to reach an OD600 between 0.5 and 0.8. Subsequently, these cultures were diluted to an OD600 of 0.05 and incubated at the interest temperature and 180 rpm until they reach an OD600 of 1.0, approximately. The measures of the OD600 were 43 MATERIALS AND METHODS performed in a spectrophotometer (Genesys 20, Thermo Scientific) along the day; samples were taken every 90 minutes. 4.2-Yeast transformation Cells were grown to an OD600 between 0.5 and 1. Then, they were harvested by centrifugation; the cellular pellet was resuspended in a mix containing 100 mM lithium acetate, 10 mM Tris-HCl, pH 8.0, and 0.5 mM EDTA (ethylenediaminetetraacetic acid) and mixed with 5 µL boiled single-stranded DNA from salmon sperm (10 mg/mL). After that, 0.5-2 µg of plasmid DNA was added into the mix. A solution containing 100 mM lithium acetate, 10 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, and 40% PEG 3350 (polyethylene glycol-3350) was also added. This final mix was incubated in a microcentrifuge tube at 30 oC and 180 rpm for 30 minutes. Next, cells were incubated in a water bath at 42 oC for 15 minutes, and then for 10 minutes on ice. Pellets were obtained by centrifugation at 6500 rpm in a microcentrifuge and resuspended in 50 µL of water. Finally, for selecting yeast transformants, cells were streaked onto selective media plates, which were incubated for 2-4 days in a 30 oC incubator. 4.3.-Preparation of bacterial competent cells E. coli was grown at 22 oC and 180 rpm in 50 mL of SOB medium (2% tryptone, 0.5% yeast extract, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, and 10 mM MgSO4) until an OD600 of 0.4. Cells were centrifuged at 4000 rpm for 10 minutes at 4 oC in a Sorvall ST 16R centrifuge (Thermo Scientific, TX-200 swinging bucket rotor). Cellular pellets were resuspended in 15 mL of TB buffer (10 mM KOH-Pipes (piperazine-N, N′-bis (2- ethanesulfonic acid)) pH 6.7, 15 mM CaCl2, 250 mM KCl, and 50 Mm MgCl2 and incubated on ice for 10 minutes. After centrifugation at 4000 rpm during 10 minutes at 4 oC, cells were frozen in a mix containing 5 mL of TB buffer and 98 µL of 8% DMSO (dimethylsulfoxyde) in liquid nitrogen. Before freezing, cells were distributed into 100 L aliquots and store at -80 oC. Alternatively, competent cells were also prepared by the so-called TSS method. A culture of 5 mL of E. coli was grown at 37 oC and 180 rpm to an OD600 of 0.35. 1 mL of cells were 44 MATERIALS AND METHODS centrifuged at 10000 rpm for 2 minutes in a microcentrifuge. The supernatant was discarded and the cells were resuspended in 75 µL cold LB broth. Next, cells were incubated on ice for 5 minutes and 75 µL of cold TSS medium (LB broth containing 10% PEG, 5% DMSO, and 50 mM Mg2+, pH 6.5) was added to the mix and incubated for 5 minutes on ice. 4.4.-Bacterial transformation Plasmid was added to 100 µL of thawed competent cells, the mix was incubated during 30 minutes on ice, then in a water bath at 42 oC for 90 seconds, and finally for 2 minutes on ice. Subsequently, 200 µL of LB broth was added into cells, which were incubated at 37 oC and 180 rpm for 60 minutes. Finally, cells were plated onto LB agar plates supplemented with those antibiotics suitable for plasmid selection and incubated overnight in a 37 oC incubator. 4.5.-NUCLEIC ACIDS MANIPULATIONS 4.5.1.-Plasmid DNA extraction from bacteria - Alkaline lysis. E. coli cellular pellets obtained from 3 mL of cells cultures grown overnight were resuspended in 100 µL of 50 mM glucose, 25 mM Tris-HCl, pH 8.0, and 10 mM EDTA. Cells were lysed on ice by addition of 200 µL of 0.2 N NaOH, and 1% SDS (Sodium dodecyl sulfate). Subsequently, 150 µL of 2 M potassium acetate, pH 6.7 was added to the cellular lysate. After centrifugation at 13000 rpm during 15 minutes at room temperature in a microcentrifuge, the supernatant was transferred to a fresh 1.5 mL tube. Then, plasmid DNA was precipitated with 1 mL of cold absolute ethanol. The precipitated DNA was washed twice with 500 µL of 70 % ethanol, centrifuged again at 13000 rpm for 2 minutes, the supernatant was discarded and the pellet air-dried. Finally, the plasmid DNA was resuspended in 25 µL of TER buffer (10 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, and 0.1 mg/mL RNase A). 51 MATERIALS AND METHODS 4.6.2.-Affinity purification A Ni-NTA agarose resin (HIS select Nickel Affinity Gel, Sigma) was equilibrated by adding 750 µL of 50 mM Hepes, pH 7.5, and 500 mM NaCl to 150 µL of resin. The supernatant from small cultures (10 mL) was added to the resin and the mix was incubated for 1 hour and 30 minutes at 4 oC in a rotating mixer of 26 cm of diameter. The mix was centrifuged at 5000 rpm for 10 minutes at 4 oC in a Sorvall ST 16R centrifuge (Thermo Scientific, TX- 200 swinging bucket rotor), in order to remove the unbound proteins. The supernatant was discarded and 1.5 mL of 50 mM Hepes, pH 7.5, 500 mM NaCl, 10% glycerol, 1 mM DTT, and 0.1% Triton X100 was added to the resin in order to eliminate undesired contaminant proteins. The mix was incubated for 5 min at 4 oC in the rotating mixer, and centrifuged as above-mentioned. This step was repeated three times. To elute bound proteins, 150 µL of 50 mM Hepes, pH 7.5, 500 mM NaCl, 10% glycerol, 1 mM DTT, 0.1% Triton X100, and 20 mM imidazole was added to the resin. After that, the resin was incubated during 10 min at 4 oC in the rotating mixer, and centrifuged as abovementioned. The elution step was repeated five times. In case of large volume cultures, a FPLC system AKTA Purifier (GE Healthcare) was used. The supernatant obtained after cell lysis (50 mL) was loaded on a 5ml HisTrap FF (GE Healthcare) column pre-equilibrated with 5 column volumes of 50 mM Tris-HCl, pH 7.4, 500 mM NaCl, 10% glycerol, 1 mM DTT, 0.5 mM PMSF, 0.1% SDS, and 10 mM imidazole. Then, the column was washed with the same buffer until the UV lecture was of zero units. Next, proteins were eluted with 20 column volumes of 50 mM Tris-HCl, pH 7.4, 500 mM NaCl, 10% glycerol, 1 mM DTT, 0.5 mM PMSF, 0.1% SDS, and 500 mM of imidazole. The obtained eluates were ready for further purification by gel filtration. 4.6.3.-Gel filtration Proteins (2 mL of eluate in 50 mM Tris-HCl, pH 7.4, 500 mM NaCl, 10% glycerol, 1 mM DTT, 0.5 mM PMSF, 0.1% SDS, and 500 mM of imidazole) were loaded on a Superdex 200 HiLoad 16/60 column (GE Healthcare) pre-equilibrated with 20 mM Tris pH 8.0, 100 mM KCl, 10% glycerol, and 1mM DTT. Protein elution was performed using the same 52 MATERIALS AND METHODS buffer. The elution was followed by measuring the A280. Fractions, which elution volumes corresponded with the size of the protein of the interest were saved. When it was necessary, the proteins were concentrated using an Amicon Ultra-0.5 mL centrifugal filters units (30K). 4.7.-Probe labeling To label different RNA probes used in this study, 10 picomole of oligonucleotide (see Table 3) were 5’ end labeled with 10 mCi/mL [ 32P]-ATP (Perkin Elmer, 6000 Ci/mmol) using T4 polynucleotide kinase (NEB) in 70 mM Tris-HCl, pH 7.6, 10 mM MgCl2, and 5 mM DTT. The mix was incubated at 37 oC for 60 minutes and at 65 oC for 25 minutes. The reaction was stopped by adding 0.5 M ammonium acetate and water to complete 50 µL. Labeled probe was purified with a microspin G50 column (GE Healthcare) by centrifugation at 3000 rpm for 2 minutes at room temperature. 4.8.-ATPase assays - By colorimetry assay: The reactions took place in a final volume of 60 µL containing 1.5 µg of protein and the nucleic acid substrate (see Table 3). ATP was used at 0.5 mM in 20 mM Hepes pH 6.0, 50 mM potassium acetate, 2 mM MgCl2, and 2 mM DTT. The reactions were incubated at 30 oC for 10 minutes. The ATPase activity was stopped by adding 10 % EDTA. Each reaction was sampled into a well of a microtiter plate. Then, 50 µL of reaction was added to a 200 µL green malachite mix (molybdate (5.72 % in 6N HCl)- malachite (0.08%)-water mix 1:1:1) and incubated 10 minutes at 30 oC. The Abs620 was measured in a Sinergy H1 microplate reader (Biotek). To calculate the phosphate released in each assay was used the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4. The conditions mentioned in this method are the standards but can vary depending of the experiment. The differences can be observed in the Results section. 53 MATERIALS AND METHODS - By radioactivity assay: Reactions were carried out in a final volume of 2.1 µL, containing 0.10 µg protein, 100 µM non-radioactive ATP, 3 µM of a single stranded RNA substrate (see Table 3), and 0.2 µM [α-32P] ATP ( 3000 Ci/mmol, Perkin Elmer) in buffer containing 25 mM Tris acetate, pH 8, 10 mM magnesium acetate, 0.2 mM DTT, and 100 µg/µl BSA. The mixture was incubated at 30 oC for 1 hour. One microliter of each reaction was spotted on a chromatography plate (Machery Nagel CEL300 PEI/UV254, Merck). The plate was introduced in 0.75M KH2PO4 until the buffer reached the top of the plate. Next, the plate was dried and exposed to an intensifying screen in a Fuji cassette and revealed in Typhoon Trio imaging system (GE Healthcare). The results were quantified by the MultiGauge V3.0 software (Fujifilm). 4.9.-Helicase assays Reactions were performed in 10 µL of final volume containing 0.6 µg of protein, 1 mM ATP, 1 nm [-32P] double stranded RNA substrate (see Table 3), 10 nM non-radioactive RNA oligonucleotide (ssDNA complementary to the displaced strand) in the same buffer as in the radioactive ATPase assays. Reactions were incubated at 30 oC for 3 minutes. Reactions were quenched by adding 2 µL of stop solution (25% glycerol, 1% SDS, 10 mM Tris-HCl pH 8.0, 1 mg/mL proteinase K, 0.3 µM stop oligo (ssDNA complementary to the displaced strand), 0.05% bromophenol blue, and 0.05% xylene cyanole), and kept it on ice. Reactions were run in an acrylamide gel under non-denaturing conditions (see below). The gel was dried and exposed to an intensifying screen in a Fuji cassette and revealed in a Typhoon Trio imaging system (GE Healthcare). The results were quantified by the MultiGauge V3.0 software (Fujifilm). The annealing reaction was carried out in 100 µL of final volume. Each reaction contained 10 nM of labeled oligonucleotide, 10 nM of non-labeled oligonucleotide in 50 mM Tris-HCl, pH 7.5, 50 mM NaCl, and 1 mM EDTA. The reaction was incubated at 80 oC for 2 minutes and later it was cooled slowly at room temperature. Finally, the annealed oligonucleotide was stored at 4 oC. 54 MATERIALS AND METHODS 4.10.-RNA binding assays The reactions were performed in 10 µL of final volume containing 1nM [-32P] ssRNA (see Table 3) in 25 mM Tris acetate pH 8.0, 10 mM magnesium acetate, 0.2 mM DTT, 100 µg/µL BSA, and 1.2 U RNasin Ribonuclease Inhibitor, and several protein amounts. Reactions were performed at 30 oC for 15 minutes and were stopped by adding 2 µL of loading buffer (0,01% xylene cyanole, 50% glycerol, and 1 mg/mL BSA). Next, they were run in an acrylamide gel under non-denaturing conditions. The gel was dried, exposed in an intensifying screen in a Fuji cassette and revealed in a Typhoon Trio imaging system (GE Healthcare). Table 3. Oligonucleotides used in the enzymatic assays. The name and the size of each oligonucleotide is indicated. Name 5’-3’ sequence Size (nt) Poly dT16 TTTTTTTTTTTTTTTT 16 Poly dT17 TTTTTTTTTTTTTTTTT 17 Poly dT30 TTTTTTTTTTTTTTTTTTTTTTTTTTTTTT 30 Poly dA18 AAAAAAAAAAAAAAAAAA 18 Poly dC18 CCCCCCCCCCCCCCCCCC 18 Poly GC18 GCGCGCGCGCGCGCGCGC 18 ssRNA1 UCGACGAUCGAUUUGCACGUCAGAACCGCUACACUGAAAUGAUGAAUUCCGGUCUCCC 58 ssRNA3 AACCGCUAC 9 ssRNA4 GUAGCGGUUCUGACGUGCAAAUCGAUCGUC 30 ssRNA5 GGGAGACCGGAAUUGAUCAUUUCAGUGUAGCGGUUCUG 38 ssDNA comp CAGAACCGCTACAC 15 Oligo stop AATGATCAATTCCGGTCTCCC 21 5.8S TTTCGCTGCGTTCTTCATC 19 5S GGTCACCCACTACACTACTCGG 22 snR190 CGAGGAAAGAAGAGACACCATTATC 25 snR5 GTCTACTTCCAGCCATTTGC 20 55 MATERIALS AND METHODS snR10 CATGGGTCAAGAACGCCCCGGAGGGG 26 snR42 TCAAACAATAGGCTCCCTAAAGCATCACAA 30 Table 4. Characteristics of snoRNAs studied. In each case is indicated the identification number obtained from the Saccharomyces genome database, the length of the snoRNA and the binding domain in the 25S rRNA (SnoRNABase, LBME, version 3). snoRNA SGD ID Length (nts) 25S domain snR190 S000006509 190 V snR5 S000007291 197 II snR10 S000007499 245 V snR42 S000006502 351 V 4.11.-Protein extractions Total proteins were extracted from yeast cells grown to an OD600 of about 1. Cellular pellets were obtained by centrifugation at 5000 rpm for 10 minutes at room temperature in a Sorvall ST 16R centrifuge (Thermo Scientific, TX-200 swinging bucket rotor). Cells were lysed with 0.2 N NaOH during 5 minutes at room temperature. Cellular lysates were recovered by centrifugation and resuspended in 160 mM Tris-HCl pH 6.8, 20 % glycerol, and bromophenol blue. The supernatants were recovered while the pellet was discarded. Proteins were also extracted from the fractions of sucrose gradients. To do this, proteins were precipitated in 20 % TCA (trichloroacetic acid), vortexed and incubated on ice for 10 minutes. After centrifugation at 13000 rpm for 10 minutes at 4 oC, 1 mL of cold acetone was added to the pellet of each fraction. Proteins were centrifuged and the pellets were washed again in 1 mL of cold acetone. A final centrifugation was performed, the pellets were dried at room temperature and resuspended in 160 mM Tris-HCl pH 6.8, 20 % glycerol and, bromophenol blue. 56 MATERIALS AND METHODS 4.12.-Polyacrylamide gel electrophoresis for proteins Protein electrophoresis was performed in denaturing acrylamide-bisacrylamide (29:1) gels. The stacking gel was composed by 0.125 M Tris-HCl, pH 6.8, 0.1 % SDS, 0.1 % APS (ammonium persulfate), 0.2% TEMED, and 4 % acrylamide-bisacrylamide (29:1). The running gel was composed by 0.375 M Tris-HCl, pH 8.8, 0.1 % of SDS, 0.1 %APS, 0.2 % TEMED, and 10% acrylamide-bisacrylamide (29:1). Gels were run at 160 V for 50 minutes in a buffer composed by 25 mM Tris, 192 mM glycine, and 10 % SDS. Proteins were stained using a Coomassie blue solution (50% ethanol, 10% acetic acid, and 0.5 g/L Coomassie Brilliant Blue R-250). Non-denaturing 8% acrylamide-bisacrylamide (19:1) or non-denaturing 6 % acrylamidebisacrylamide (37.5:1) gels were also used depending of the assay. Proteins gels for helicase assays were composed by 100 mM Tris-base, 90 mM boric acid, 1 mM EDTA, pH 8.3, 0.16 % APS ,0.05% TEMED, and 8 % of acrylamide-bisacrylamide (19:1). Gels were run at 80 V for 40 minutes at 4 oC in a buffer consisting of 100 mM Tris-base, 90 mM boric acid, and 1 mM EDTA, pH 8.3. Proteins gels for RNA binding assays were composed by 100 mM Tris-base, 90 mM boric acid, 1 mM EDTA, pH 8.3, 0.16 % APS, 0.05 % TEMED, and 6 % acrylamide-bisacrylamide (37.5:1). Gels were run at 120 V for 35 minutes at 4 oC in 100 mM Tris-base, 90 mM boric acid, and 1 mM EDTA, pH 8.3. 4.13.-RNA gel electrophoresis RNA gel electrophoresis was performed in denaturing 7 % polyacrylamide (19:1)-8 M urea gels. The RNA samples were heated at 65 oC in a solution containing 10 M urea, 2 mM EDTA, 0.025 % xylene cyanol, and 0.025 % bromophenol blue before loading. The gels were run at 350 V during 2 hours and 30 minutes in 100 mM Tris-base, 90 mM boric acid, and 1 mM EDTA, pH 8.3. 57 MATERIALS AND METHODS 4.14.-Gel to membrane transfer Proteins were transferred from gels to nitrocellulose membrane in sandwiches at 100 V for 30 minutes in 25 mM Tris, 192 mM glycine, 10 % SDS, and 20% ethanol. After transfer, proteins were visualised using a solution containing 0.1% red Ponceau and 5 % acetic acid. RNA was transferred from the gel onto a HybondTM-N nylon membrane overnight at 15 V and 4 oC in 50 mM Tris-base, 45 mM boric acid, and 0.5 mM EDTA. RNAs were fixed at 1200 X100 µJ/cm2 on the membranes using an UV crosslinker (Spectrolinker XL-1000). After transfer, RNAs were visualized using a solution containing 0.02% methylene blue and 0.3 M sodium acetate at pH 5.5. 4.15.-Inmunodetection Membranes were blocked in a solution containing 5% skimmed milk, 15 mM Tris-HCl pH 7.5, and 200 mM NaCl at room temperature for 30 minutes. Membranes were incubated with a primary antibody in 5 % skimmed milk or 3 % BSA (Bovin serum albumin), 15 mM Tris-HCl pH 7.5 and 200 mM NaCl. The secondary antibodies used were horseradish peroxidase-coupled antibodies. The incubations of the membrane with the secondary antibody were carried out in 15 mM Tris-HCl pH 7.5, and 200 mM NaCl. Washes of the membrane between the primary and the secondary antibodies and after the secondary antibody were performed in 15 mM Tris-HCl, pH 7.5, 200 mM NaCl, and 0.1% Tween 20. Chemiluminescence was produced by addition of luminol and hydrogen peroxide using a detection kit (Super Signal West Pico Plus, Thermo Scientific). Chemiluminescence was detected in an Odissey FC Dual Mode Imaging System (Li-Cor Biosciences). 4.16.-RNA-DNA hybridisation Prehybridisation of the membranes were done with a solution containing 0.3 M Na2HPO4, 0.2 M NaHPO4, 7 % SDS, 0.5 M EDTA, pH 8, and 10 mg/mL BSA at 42 oC for 30 minutes. The hybridisation of the membranes was done in the buffer aforementioned but adding 10 picomoles of the labelled probe at 42 oC, which were incubated overnight. Three washes of 5 minutes were performed with 2X SSC (0.3 M NaCl, and 0.03 M sodium citrate) and 0.5 % SDS, and three washes with 0.1X SSC (0.015 M NaCl, and 0.0015 M 58 MATERIALS AND METHODS sodium citrate) and 0.5% SDS at 38 oC. Membranes were exposed to intensifying screen in Fuji cassettes and revealed in a Typhoon FLA 9000 imaging system (GE Healthcare). 4.17.-Fluorescence microscopy Cell cultures of 5 mL were grown to reach an OD600 of about 1, centrifuged and resuspended in 50 µL of medium. Samples were examined with a Leica DMR fluorescence microscope equipped with a DC350F digital camera and its software. The excitation channel used for fluorescent eGFP and mCherry proteins were BP450-490 and BP 515-560 respectively. Adobe Photoshop CS2 (Adobe Systems Inc.) was used to process the images. Table 5. Plasmids used in this study. Name Relevant information Source pGEMT-Easy Ampr Promega pET-21 b (+) Ampr,6x His tag Novagen YCplac33 CEN, URA3 Gietz and Sugino, 1988 YCplac22 CEN, TRP1 Gietz and Sugino, 1988 YCplac111 CEN, LEU2 Gietz and Sugino, 1988 pADH111(GA5)-3X-yEGFP C-terminal yeGFP tag, CEN, LEU2 Koch, 2012 pRS415 CEN, LEU2 Sikorski and Hieter, 1989 YCplac111-yGFP7/TCYC1 C-terminal yGFP tag, TCYC1, LEU2 Koch, 2012 pADH111-SV40NLS-(GA5)3XyEGFP CEN, LEU2 Koch, 2015 59 MATERIALS AND METHODS Table 6. Plasmids constructed in this work. The primers used to obtain each plasmid are indicated in the table. The template from which the construction was obtained is showed also shown. Name Primers used Template pET-21 b-DBP7-His pET21aDbp7SacF pET21aDbp7NotR Genomic DNA S. cerevisiae pET-21 b-dbp7-His [K197A] Dbp7KAF Dbp7KAF pET21aDbp7SacF pET21aDbp7NotR pET-21 b-DBP7-His pET-21 b-dbp7-His [R553A] Dbp7RAF Dbp7RAR pET21aDbp7SacF pET21aDbp7NotR pET-21 b-DBP7-His pET-21 b-DBP9-His pET21aDbp9BamF pET21aDbp9SacR Genomic DNA S. cerevisiae pET-21 b-dbp9-His [K68A] Dbp9KAF Dbp9KAR pET21aDbp9BamF pET21aDbp9SacR pET-21 b-DBP9-His YCplac22-HA-DBP7 M13 REVERSE NarIR YCplac33-HA-DBP7 YCplac22-HA-dbp7[K197A] K197A_FW K197A_FW YCplac22-HA-DBP7 YCplac22-HA-dbp7[R553A] R553A_FW R553A_RV YCplac22-HA-DBP7 YCplac22-HA-dbp7∆N10 ∆N1F ∆NR YCplac22-HA-DBP7 YCplac22-HA-dbp7∆N162 ∆N3F ∆NR YCplac22-HA-DBP7 YCplac22-HA-dbp7∆N693 ∆CF ∆C1R YCplac22-HA-DBP7 YCplac22-HA-dbp7∆N635 ∆CF ∆C2R YCplac22-HA-DBP7 YCplac22-HA-dbp7∆NLS NONLSF YCplac22-HA-DBP7 60 MATERIALS AND METHODS NONLSR pADH111-DBP7 N-terminal-(GA5)3X- yEGFP N-terminal BamF N-terminal BamR YCplac22-HA-DBP7 pADH111-dbp7∆NLS-N-terminal-(GA5)3X- yEGFP N-terminal BamF N-terminal BamR YCplac22-HA-dbp7∆NLS YCplac111-yGFP7/TCYC1-DBP7-(GA5)3X- yEGFP Dbp7GFPXmaI Dbp7GFPSphI Genomic DNA S. cerevisiae YCplac111-yGFP7/TCYC1- dbp7∆NLS - (GA5)3XyEGFP DBP7-DelNLS-FW DBP7-DelNLS-RV YCplac111-yGFP7/TCYC1-DBP7- (GA5)3XyEGFP YCplac111-yGFP7/TCYC1- dbp7∆N162 - (GA5)3XyEGFP DBP7-DelNter-FW DBP7-DelNter-RV YCplac111-yGFP7/TCYC1-DBP7- (GA5)3XyEGFP 66 RESULTS S The constructs were transformed in the strain dbp7∆[YCplac33-DBP7]. The transformants were selected on SD-Trp, and then the URA3-marked plasmid counterselected on 5-FOA plates (plasmid shuffling). Some of the colonies growing on this media were transferred to SD-Trp plates for further use. To assay protein expression, the steady-state levels of the respective HA-Dbp7 in the mutant strains and its wild-type counterpart were determined by western blot analyses. The protein expression levels of the wild-type HA-Dbp7 and mutant variants HA- dbp7[K197A], and HA-dbp7[R553A] were compared to the levels of a loading control protein, which corresponded to phosphoglycerate enzyme, Pgk1. As shown in Figure 21, all the Dpb7 variants were detected by western blot analysis using an anti-HA antibody. This identification was specific as no signal was found when the protein extract of a dpb7∆ strain was assayed (Figure 21). Figure 21. Immunodetection of HA-Dbp7 variant proteins. The different HA-Dbp7 variant proteins were detected with a monoclonal anti-HA antibody. Whole cell extracts were prepared from the indicated strains, grown in SD-Trp to exponential phase at 30 oC. Equivalent amounts of extracts were analysed by western blotting. As a loading control, phosphoglycerate kinase (Pgk1) protein was used. A negative dbp7∆ strain control was also used. Growth was also evaluated. Firstly, this phenotype was measured by serial dilutions in solid medium. As shown in Figure 22A, the HA-dbp7[K197A] strain was slightly affected on growth, while the HA-dbp7[R553A] was significantly affected on this feature. Secondly, growth was also quantified in order to compare with that of isogenic wild-type HA-DBP7, and dpb7∆ strains (Figure 22B). The doubling times calculated for the different 67 RESULTS S strains in SD-Trp at 30 oC were of about 140, 180 and 270 minutes for the HA-DBP7, HA- dbp7[K197A] and HA-dbp7[R553A] strains, respectively. Figure 22. Growth features of the point mutant strains. The growth phenotype was measured by serial dilutions in solid medium SD-Trp medium. (C) Growth curves of the HA-DBP7, HA-dbp7[K197A] and HA- dbp7[R553A] strains. Cultures were initiated in SD-Trp liquid medium and incubated at 30 oC. Three biological replicas were done for each strain. The curves of a representative experiment are shown. Note that only exponential growth phase is plotted. To determine if these mutations affects ribosome biogenesis, we assayed steady-state levels of ribosomes by polysome profile analyses (Figure 23). Compared to wild-type HADBP7 cells, the HA-dbp7[K197A] and HA-dbp7[R553A] mutant cells showed a decrease in the levels of free LSUs, an increase in the levels of free SSUs, and an overall decrease in the 80S peak and in polysomes. These defects were indicative of an LSU shortage, which in the case of the HA-dbp7[K197A] mutant also led to the appearance of half-mers polysomes. 68 RESULTS S Figure 23. Polysome profile of (A) the HA-DBP7, (B) the HA-dbp7∆[K197A] and HA-dbp7[R553A] strains. These strains were grown to exponential phase in liquid SD-Trp medium at 30 oC. Cell extracts were prepared, and 10 A260 units resolved on 7-50% sucrose gradients. The A254 was continuously measured. Sedimentation comes from left to right. The absorbance peaks of free 40S, 60S r-subunits, 80S vacant ribosomes/monosomes and polysomes are indicated, and the appearance of half-mers polysomes, pointed out by arrows. 5.3.-Phenotypic analysis of truncated dbp7 alleles in the amino- and carboxylterminal extensions of the Dbp7 protein As above stated, Dbp7 harbours N- and C- terminal extension of approximately 200 amino acids. To get insight into the function of these extensions, different serial truncated dbp7 alleles were constructed from Ycplac22-HA-DBP7 and expressed under the control of the cognate DBP7 promoter followed by a HA epitope. Upon transformation of these constructs into the dpb7∆[YCplac33-DBP7] strain, cells were subjected to plasmid shuffling on 5-FOA plates. Some of the colonies growing on this media were transferred to SD-Trp plates for further use. The following N- and C- terminal truncations were done: (1) HA-dbp7∆N10, an allele harbouring a deletion of the codons corresponding to the first 10 amino acids of the Dbp7; (2) HA-dbp7∆N162, an allele harbouring a deletion of the codons corresponding 69 RESULTS S to the first 162 amino acids. (3) HA-dbp7N693, an allele harbouring a deletion of the codons corresponding to the last 49 amino acids, thus leading to a protein of 693 amino acids instead of 742 amino acids. (4) HA-dbp7N635, an allele harbouring a deletion of the codons corresponding to the last 107 amino acids of Dbp7. To test whether mutant proteins were expressed, we first determined the steady-state level of the corresponding HA-Dbp7 proteins by western blot analyses, and compared them to that of the isogenic wild-type counterpart. As shown in Figure 24, the mutant as well as the wild-type full-length HA-Dbp7 proteins were detected in the western blotting using a specific antibody against the HA epitope. As expected, no signal was found when extracts of a dpb7∆ strain were assayed. The level of expression varied for the different HA-Dbp7 proteins. The levels of expression were very low for the HA- dbp7N693 strain. Figure 24. Immunodetection of HA-Dbp7 truncated proteins. The different HA-Dbp7 variant proteins were detected with a monoclonal anti-HA antibody. Whole cell extracts were prepared from the indicated strains, which were grown in SD-Trp to exponential phase at 30 oC. Equivalent amounts of extracts were analysed by western blotting. As a loading control, phosphoglycerate kinase (Pgk1) protein was used. A negative dbp7∆ strain control was also used. Note that the levels of expression for the HA-dbp7N693 protein were lower than for the rest of proteins. The growth phenotype of these strains also was evaluated. First, serial dilutions were plated onto solid medium. As shown in Figure 25, the growth of the different mutants respect to the isogenic wild-type HA-DBP7 strain was as follow: the HA-dbp7∆N162, HA- dbp7N693, and HA-dbp7N635 mutants grew significantly worse than the isogenic HADBP7 strain. In fact, the growth of these mutants was very similar to that of the dbp7Δ 70 RESULTS S strain. Only the strain HA-dbp7∆N10 grew apparently similar to the HA-DBP7 strain (Figure 25A). When growth curves were done, the calculated doubling time for each strain were of about 115 minutes for HA-dbp7∆N10, 345 minutes for HA-dbp7∆N162, 440 minutes for HA-dbp7N693 and 400 minutes for HA-dbp7N635. The doubling time of the strain HA-dbp7∆N10 was practically identical to that of the HA-DBP7 strain, which was of about 90 minutes. However, the doubling time of the rest of the strains were closer to the doubling time of dbp7∆ strain, which we calculated to be 345 minutes (Figure 25B). Figure 25. Phenotypic analysis of truncated mutants at the N- and C- terminal extensions of Dbp7. (A) The growth phenotype was measured by serial dilutions in solid medium SD-Trp medium. (B) Growth curves of HA-DBP7, dbp7Δ and truncated mutant strains. Cultures were initiated in SD-Trp liquid medium and 71 RESULTS S incubated at 30 oC. Three biological replicas were done for each strain. The curves of a representative experiment are shown. Note that only exponential growth phase is plotted. We also assayed the effect of these mutations on the profiles of their ribosomes. As shown in Figure 26, the HA-dbp7∆N162, HA-dbp7N693 and HA-dbp7N635 strains presented polysome profiles similar to those of a dbp7Δ strain. Thus, these mutants showed a decrease in the free LSU peak, and a concomitant increase in the free SSU peak together with the appearance of half-mers polysomes. However, the polysome profile of the HA-dbp7∆N10 strain was very similar to that of the isogenic wild-type HADBP7 control (Figure 26). Figure 26. Polysome profiles of the truncated mutants in the N- and C- terminal extensions of the Dbp7 protein. These strains were grown to exponential phase in liquid SD-Trp medium at 30 oC. Cell extracts were prepared, and 10 A260 units resolved on 7-50% sucrose gradients. The A254 was continuously measured. Sedimentation comes from left to right. The absorbance peaks of free 40S, 60S r-subunits, 80S vacant ribosomes/monosomes and polysomes are indicated, and the appearance of half-mers polysomes, pointed out by arrows. 72 RESULTS S The profiles obtained for most mutants suggest that the respective mutations lead to an overall deficit of LSUs. To corroborate this, we calculated for each mutant strain the 60S/40S ratio, obtained from analysis of total r-subunits by low Mg2+ sucrose gradients. As a result, in all the truncated mutants, the 60S/40S ratios were lower than that obtained for the isogenic HA-DBP7 strain, except for mutant HA-dbp7∆N10, whose calculated ratio was similar to that of HA-DBP7 control strain (Figure 27). Figure 27. Quantification of the relative 60S/40S ratios of the truncated mutants in the N- and C- terminal extensions of the Dbp7 protein. The indicated strains were grown to an OD600 of 0.8 in liquid SD-Trp medium at 30 oC. Cell extracts were prepared for ribosome isolation, in a buffer lacking MgCl2 to dissociate ribosomes in their respective subunits. Ten A260 units of each extract was resolved in 7-50% sucrose gradients prepared in the same buffer. The areas under the peaks were determined, and the ratio between 60S and 40S areas was calculated. At least three independent experiments were done; means and standard deviations are shown. A T-student test (independent samples) was performed using p ˂ 0.05. Next, we examined whether the respective truncated Dbp7 proteins were still being associated with pre-60S r-particles. To do so, we analysed the sedimentation behaviour or of wild-type and mutant Dbp7 proteins in sucrose gradients. For this purpose, total extracts were prepared from the strains grown in liquid SD-Trp at 30 oC and subjected to low Mg2+ sucrose gradient ultracentrifugation. Then, fractions were collected and 73 RESULTS S analysed by western blotting. As shown in Figure 28, a peak of wild-type HA-Dbp7 was associated with high-molecular mass complexes; this indicates that Dbp7 associates relatively stable to pre-ribosomal particles, as also does Erb1 (89). As additional controls, we used the proteins uL3 and eS8, which assembles in LSUs and SSUs, respectively. Figure 28. Immunodetection of proteins present in fractions from ribosome-enriched particle fractionation in HA-DBP7 strain. (A) Profile obtained from ribosome-enriched particle fractionation in low Mg2+ sucrose gradients. Peaks from left to right corresponds to free fractions (low molecular weight complex and free proteins), total 40S, and total 60S subunits, respectively. Numbers at the bottom of the profile indicate the fraction. (B) Western blot analysis performed using the fractions collected from the above-mentioned gradient. T indicates the total extraction before particle fractionation, which is used as a control. The peaks corresponding to 40S and 60S subunits are indicated. Numbers at the bottom indicates the corresponding fractions. A representative profile and western blot analysis are shown from three replicates, which showed the same results. 74 RESULTS S In the case of truncated mutant strains, we could observe the same distribution as in HA-DBP7 strain. Thus, the truncated Dbp7 variant proteins were also stably associated to pre-ribosomal particles. The only exception was that of mutant HA-dbp7N693, where the protein could not properly detect due to its reduced levels (Figure 29). 75 RESULTS S 82 RESULTS S both, the nucleus and the cytoplasm. Interestingly, the Dbp7∆N162-(GA)5-3xyEGFP was detected exclusively in the cytoplasm (Figure 36). Figure 35. Immunodetection of the different GFP-tagged Dbp7 proteins. Whole cell extracts were prepared from YLK500 cells expressing the indicated constructs, which were grown to exponential phase in SD-Leu at 30 oC. Equivalent amounts of extracts were analysed by western blotting using a specific anti-GFP antibody. As a loading control, yeast phosphoglycerate kinase (Pgk1) protein was used. 83 RESULTS S Figure 36. Subcellular localization of Dbp7. Strain YKL500, was transformed with different plasmids; a (GA)5-3xyEGFP reporter alone expressed from the ADH1 promoter (Vector), a full-length Dbp7, a Dbp7 lacking the 31 amino acids described as an NLS and a Dbp7 lacking the 162 amino acids from its N- terminal tail. All the proteins were fused to a C-terminal (GA)5-3xyEGFP reporter. After transformation, cells were grown in SD-Leu medium at 30 oC, and the localization of the GFP fusions obtained by fluorescence microscopy. Nucleolus were revealed by the nucleolar marker protein Nop58-yEmCherry. Cells were identified under bright field illumination (Visible). About 200 cells were inspected for each reporter, and practically all cells gave the same results as the selected ones shown in the pictures. 84 RESULTS S 5.5.- In vitro properties of Dbp7 As a second aim, in addition to clarify the in vivo function of the N- and C- terminal extensions of Dbp7, we explored the enzymatic activities of recombinant Dbp7, making use of wild-type Dbp7,and the variants Dpb7[K197A] and Dbp7[R553A], to which in vivo point mutants have been discussed in 5.2 section from Results. 5.5.1.- Expression and purification of wild-type Dbp7, Dbp7[K197A], and Dbp7[R553A] proteins A C-terminal His6-tagged Dbp7 fusion protein was constructed by introduction of the open reading frame of DBP7 in the plasmid pET21b, under the control of an inducible T7 promoter. We also constructed Dpb7 mutant variants, which harbour the mutation K197A or R553A in motifs I and VI, respectively. To do so, the pET21b-His6-DPB7 plasmid was mutagenized in vitro. Each plasmid was transformed into the E. coli BL21 Codon Plus strain. Transformed cells were grown, the proteins overexpressed after IPTG induction, and, recombinant proteins purified by affinity chromatography using a nickel (Ni-NTA) agarose resin, in cases of small volume cultures, or a 5 mL HisTrap FF column (GE Healthcare) coupled to a FPLC system in cases of large volume cultures. The elution fractions from the affinity chromatography were run in a denaturing polyacrylamide gel; a prominent protein band above 75 kDa was observed (Figure 37A and B). This molecular mass coincides with that estimated for Dbp7 by sequence analysis (83 kDa). In the case of the purification by the FPLC system, the A260 profile indicated the elution profile (Figure 37C). The peak observed in the chromatography is coincident with the elution peak in the gel. The identity of the protein was verified by western blot analysis using anti-polyhistidine antibodies (data not shown). 85 RESULTS S Figure 37. Purification of Dbp7 wild-type by affinity chromatography. (A) and (B) 10% SDS-PAGE gel showing the fractions obtained by affinity purification with Ni-NTA agarose resin or by a HisTrap FF column in a FPLC system, respectively. W1, W2, and W3 indicate wash volumes 1, 2 and 3. E1, E2, E3, E4, and E5 indicate elution volumes 1, 2, 3, 4 and 5, respectively. I, indicates input, Std, protein standards and U, unbound proteins. (C) Chromatogram of recombinant wild-type Dbp7 obtained by affinity purification in a FPLC system. Dashed lines indicate the interval of fractions in which Dbp7 protein elutes. This interval corresponds to fractions 44 to 52. Fractions 45, 46, 47, 48 and 49 (marked in red in the gel B) were pooled and injected in a Superdex 200 HiLoad 16/60 column. The flow through and elution peaks are indicated. 86 RESULTS S In order to detect multisubunit complexes, selected fractions from the affinity column (from 45 to 49 in the example of Figure 37) were pooled and subjected to molecular exclusion chromatography in a Superdex 200 HiLoad 16/60 column. After this purification step, it could be observed that Dbp7 was purified at a high electrophoretic homogeneity (Figure 38A). Surprisingly, when the molecular mass of Dbp7 wild-type was calculated by gel filtration, the protein eluted at a volume (70,59 mL) which corresponds to about 146 kDa (Figure 38B). Thus, these results suggest that most wildtype Dbp7 protein eluted as a dimer. A shoulder was also observed in the main peak at an elution volume of approximately 80 mL, which likely corresponded to the size of the monomer (83 kDa). 87 RESULTS S Figure 38. Purification of recombinant Dbp7 wild-type by molecular exclusion in a Superdex 200 HiLoad 16/60 column. (A) A 10% SDS-PAGE gel showing the fractions obtained after the chromatography. Number in red marks the fraction eluting at about 70.6 mL (dimer), and number in green fraction eluting in the shoulder of the peak (monomer). I indicates input and Std protein standards. (B) Chromatogram corresponding to a typical molecular exclusion chromatography of wild-type Dbp7. The shoulder of the elution peak is indicated by a red arrow. To analyze the importance of the motifs I and VI on the enzymatic properties of the recombinant Dbp7 protein, we also overexpressed in E. coli and purified the Dbp7[K197A] and Dbp7[R553A] proteins. The affinity purification profile was similar to wild-type Dbp7. Thus, an apparent molecular mass above 75 kDa was estimated for Dbp7[K197A] by SDS-PAGE, which is similar to that obtained for the wild-type protein. As above, two elution peaks, one of about 110 kDa and another of about 73 kDa, were obtained after tandem affinity and molecular exclusion chromatography, suggesting again the dimeric nature of the putative RNA helicase, although none of these calculated masses exactly corresponded to the molecular masses calculated for the wild-type protein. We also purified recombinant Dbp7[R553A]. In this case, its gel filtration profile and behaviour in an SDS-PAGE gel were identical to the equivalent features of the recombinant wild-type Dbp7 protein. 5.5.2.- ATPase activity of Dbp7: the colorimetric assays As aforementioned, Dbp7 contains motifs involved in ATP binding and hydrolysis. However, to our knowledge, whether Dbp7 hydrolyse ATP in vitro has not been determined so far. In order to determine whether Dbp7 has ATPase activity, reaction conditions for an enzymatic assay were set-up. First of all, the ATPase activity of Dbp7 was measured at different ATP concentrations. Afterwards, parameters such as the relationship between protein concentration and the reaction time respect the ATP hydrolysis were measured. The dependence of the cation used as a cofactor was also assessed, including assays in the presence of EDTA. In addition, we studied the relationship between the ATPase activity and the temperature or the pH in the reaction mixtures. Finally, the ATPase activity of Dbp7 was measured in presence of RNase A to 88 RESULTS S determine the RNA dependence, and using different RNA or DNA as substrates. All these assays were monitored using a standard colorimetric method, which evaluated the amount of inorganic phosphate (Pi) released during the course of the reaction. For this method, we used the semi-pure protein sample contained in the elution 1 from an experiment of affinity chromatography using the Ni-NTA agarose resin (see Figure 37A). A) ATP concentration Firstly, we evaluated the activity of Dbp7 with different concentrations of this nucleotide, ranging from 0 to 1 mM. No exogenous RNA was used in the assay. We thought that the preparation must contain co-purified RNA. As shown in the Figure 39, the level of hydrolyzed ATP augmented as the ATP concentration did. However, at 0.5 mM ATP, the ATP hydrolysis did not increase further, indicating the saturation of the assay. (Figure 39). Consequently, a concentration of 0.5 mM of ATP was thereafter used, unless otherwise stated. 89 RESULTS S Figure 39. Plot of ATPase activity of Dbp7 wild-type with respect to ATP concentration. (A) A colorimetric assay was used, which measures the amount of Pi released (y-axis) at different ATP concentration (x-axis) during the assay. In all assays, 1.5 µg protein was used in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped after 10 min of incubation. (B) Linear regression obtained upon measuring the values given by different concentrations of NaH2PO4. The regression was used to calculate the amount of free Pi in the colorimetric assay. B) Protein concentration Secondly, we evaluated the activity of Dbp7 with different protein concentrations, ranging from 0 to 2.5 µg. As shown in Figure 40, we found a linear positive correlation between the protein concentration and the Pi released. We decided to use 1.5 µg of protein for further assays. 90 RESULTS S Figure 40. Plot of ATPase activity of Dbp7 wild-type with respect to protein concentration. A colorimetric assay was used, which measures the amount of Pi released (y axis) at different protein concentration (xaxis). The reaction contained 0.5 mM ATP in a buffer with 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped after 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. C) Reaction time We next assayed the ATPase activity of Dbp7 at different times, ranging from 0 to 50 minutes. As shown in Figure 41, and as expected, we obtained a saturation curve, which reached the saturation point about the 25 minutes after the reaction was started. Figure 41. Plot of ATPase activity of Dbp7 wild-type along the time. A colorimetric assay was used, which measures the amount of Pi released (y-axis) at different times (x-axis). In the assays, 1.5 µg protein, and 0.5 mM ATP were used in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped at different times of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. D) Mg2+ dependence Usually, ATP-dependent RNA helicases use Mg2+ as a cofactor to hydrolyze ATP. To study the dependence of this cation, we determined the ATPase activity of wild-type Dbp7 at different Mg2+ concentrations, ranging from 0.5 to 2 mM. As shown in Figure 42, ATP hydrolysis increased when the Mg2+ concentration also did it; however, at concentrations about 0.5 mM Mg2+, ATP hydrolysis did not increase further. 91 RESULTS S Figure 42 Plot of ATPase activity of Dbp7 wild-type with respect to Mg+2 concentration. A colorimetric assay was used, which measures the amount of Pi released (y-axis) at different Mg+2 concentration (xaxis). The reaction contained 1.5 µg of protein, and 0.5 mM ATP. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped after 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values by different concentrations of NaH2PO4 given in the colorimetric assay. E) EDTA concentration A different approach to evaluate the dependence of Mg2+ on the ATPase activity of wildtype Dbp7 was performing assays at a fixed concentration of Mg2+ while increasing concentrations of EDTA, which is a Mg2+ chelating agent, are used. The EDTA concentration ranging from 15 to 200 µM. As shown in Figure 43, the ATPase activity of Dbp7 decreased drastically when EDTA was added at concentrations over 50 µM (Figure 43). 98 RESULTS S As shown in Figure 51A, all fractions harboured ATPase activity in these conditions, being the fraction 11 the one displaying the highest activity, (Figure 51B). Thus, we decided to use this fraction for further assays. Figure 51. ATPase activity assay for Dbp7 wild-type protein. (A) Thin layer chromatography of the reaction products of the assay. The assay contained 0.1 µg protein, 0.1 mM non-radiactive ATP, 0.2 µM [α-32P] ATP and, 3 µM ssRNA 1 in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. No RNase A was added during the protein purification. The upper spots corresponded to the amount of ADP obtained from the ATP hydrolysis and the lower spots the remained ATP, which was not hydrolyse along the duration of the assay. Different fractions of a gel exclusion column from the purification of wildtype Dbp7 (see Figure 38) were used. A control without protein was also loaded. (B) Quantification of the ATPase activities detected in A. For this, the percentage of ADP released respect the ATP remained was calculated. 99 RESULTS S To verify that the ATPase activity of Dbp7 is dependent of RNA, we performed new assays in the absence of the ssRNA 1. As shown in Figure 52, there was a strong decrease in the ATPase activity when the RNA molecule was excluded from the assay. Figure 52. ATPase activity of Dbp7 wild-type in the presence or the absence of ssRNA. The figure represented the percentage of ADP released in the assay versus time. The assay contained 0.1 µg protein, 0.1 mM non-radiactive ATP, 0.2 µM [α-32P] ATP and, 3 µM ssRNA 1 in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. No RNase A was added during the protein purification. 5.5.4.- Determination of the Michaelis-Menten kinetics parameters for recombinant Dbp7 wild-type and Dbp7[K197A] protein We tried to calculate the Michalis-Menten kinetics parameters KM and Vmax for wild-type Dbp7 with the colorimetric assay, but we were unable to properly calculate them, because, at a determined ATP concentration, there was an inhibition of the reaction and the curve headed down (Figure 53). 100 RESULTS S Figure 53. ATP hydrolysis velocity by Dbp7 wild-type at different concentration of ATP. The velocity increased when the ATP concentration also did it. Note that at certain concentration of ATP, the velocity decreased. Three replicas were done for this assay. Only a representative one is shown. The assay was performed with 1.5 µg protein, and 0.5 µM poly dT30 in a buffer containing 2 mM Mg2+, pH 6.0. Reactions were performed at 30 oC for 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. To try to determine the kinetics parameter of the variant Dbp7[K197A], we proceeded as we did for the wild-type Dbp7 protein. We performed colorimetric ATPase assay at different ATP concentrations, and measured the Pi released. Unfortunately, as for the wild-type Dbp7, we observed inhibition of the ATPase activity as the ATP concentration increased (Figure 54). 101 RESULTS S Figure 54. ATP hydrolysis velocity by Dbp7[K197A] at different concentration of ATP. The velocity increased when the ATP concentration also did it. However, note that at certain concentration of ATP, the velocity decreased. Three replicas were done for this assay. Only a representative one is shown. The assay was performed with 1.5 µg protein, and 0.5 µM poly dT30 in a buffer containing 2 mM Mg2+, pH 6.0. Reactions were performed at 30 oC for 10 min. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. In view of the above, we did an attempt to calculate these parameters by the radioactive assay. First, we assayed kinetically the ATPase activity at different ATP concentration and different times; then, we obtained the respective initial velocities (V0) for each ATP concentration. One represented, the slope of each line provided the different V0. If V0 is represented versus the ATP concentration (1 µM to 300 µM), Michaelis-Menten kinetic model scenario is obtained. As shown in Figure 55, and strikingly, when the ATP concentration increased, the calculated V0 of the reactions again decreased. 102 RESULTS S Figure 55. Representation of the ADP% along the time at different range of concentrations of ATP. Note the inhibition of the ATPase activity by increasing the ATP concentration. (A, B, C). The concentration of ATP used in the assays ranging from 1 µM to 300 µM. The concentration of [α-32P] ATP was 0.2 µM. The concentration of protein used was 0.27 µg, and the concentration ssRNA 1 was 3 µM in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. 103 RESULTS S In this circumstance, it was impossible to calculate kinetic parameters by a standard Lineweaver-Burk representation, neither by the colorimetric nor the radioactive assay. For this reason, in order to obtain at least an approximation of the kinetic parameters for the recombinant wild-type and mutant Dbp7 in vitro, we represented directly the Michaelis-Menten graphs using the data obtained after using the colorimetric assay represented in Figures 53 and 54. To obtain an approximation of the kinetics parameters, we eliminated from the curve the points that make the curve to head down. In this way, the curves roughly approached the model of enzymatic kinetic of Michaelis- Menten. Accordingly, the Vmax is approximated by the maximum rate achieved by the system under substrate saturation condition. The KM or Michaelis-Menten constant is the substrate concentration at which the reaction rate is half of Vmax. Following this approximation, we obtained a coarse KM of about 0.30 mM for ATP and a Vmax of about 0.075 mM/min for wild-type Dbp7 and a coarse KM of about 0.20 mM and a Vmax of about 0,07 mM/min for the mutant Dbp7 (Figure 56 and 57). Figure 56. ATPase activity of Dbp7 wild-type with respect to ATP concentration. Only 8 values have been used to calculate the kinetic parameters. The points in which an inhibition was clearly observed (Figure 53) were eliminated from the graph. These points were the corresponding to concentrations of 2.5, 3, 3.5, 4, 4.5, 5, and 5.5 mM of ATP. Each point represented the mean and the standard deviation of at least three replicates. The coarse KM and Vmax were calculated using the Graph Pad Prism software. 104 RESULTS S Figure 57. ATPase activity of Dbp7[K197A] with respect to ATP concentration. Only 8 values have been used to calculate the kinetic parameters. The points in which an inhibition was clearly observed (Figure 53) were eliminated from the graph. These points were the corresponding to concentrations of 2.5, 3, 3.5, 4, 4.5, 5, and 5.5 mM of ATP. Each point represented the mean and the standard deviation of at least three replicates. The coarse KM and Vmax were calculated using the Graph Pad Prism software. These data suggest an inhibition of ATPase activity by substrate (ATP) or by one of the products (ADP or Pi). To get insight of a possible inhibition by product, we performed ATPase assays in the presence of ADP and ATP or in presence of ATP. Two different assays were done: in the first one, we used on one hand 50 µM of ATP and on the other hand 50 µM of ATP and 50 µM of ADP, while in the second one we used on one hand 300 µM of ATP and on the other hand 100 µM of ADP and 300 µM of ATP. In both cases, the radioactive method was used. As shown in Figure 58, the ATPase activity of wildtype Dbp7 was very similar in the presence or the absence of ADP independently of the ATP concentration (Figure 58). 105 RESULTS S Figure 58. ATPase activity of Dbp7 wild-type protein in the presence of ADP. The assay contained 0.27 µg protein, 0.2 µM [α-32P] ATP, and 3 µM ssRNA 1 in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. The amount of ATP and ADP was 50 µM in the first assay and 100 µM of ADP and 300 µM of ATP in the second assay. 5.5.5.- ATPase activity of Dbp7[R553A] For the mutant Dbp7[R553A] variant protein, we also performed a radioactive ATPase assay to verify whether the mutant was able to hydrolyze ATP. To do the assay, we used the protein obtained after a gel filtration purification. The fraction of the mutant selected was the corresponding to a fraction equivalent to fraction number 11 in Dbp7 wild-type (see Figure 38). Surprisingly, this mutant protein had, as previously we also observed for the Dbp7[K197A] variant protein, ATPase activity (Figure 59A). When compared to the activities of the wild-type protein, the mutant Dbp7[R553A] protein had even a slightly higher ATPase activity that the wild-type Dbp7 protein (Figure 59B). 106 RESULTS S Figure 58. Radioactive ATPase activity assay for the mutant Dbp7[R553A] protein. (A) Thin layer chromatography of the reaction products of the assay. (B) Quantification of the percentage of ADP released in the assay versus time. The concentration of protein used in the assay was 0.1 µg, and the concentration of ssRNA 1 was 3 µM. The assay was performed with 0.3 mM of ATP as substrate and 0.2 µM [α-32P] ATP in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. No RNase A was added during the protein purification. The % of ATP hydrolyzed at each time is the average between two independent experiments. 5.5.6.- RNA binding activity of the recombinant Dbp7 wild-type and Dbp7[K197A] in vitro To investigate whether wild-type Dbp7 is able to bind RNA, we performed in vitro gelshift assays in which the delay in the electrophoretic migration of a labelled ssRNA is interpreted as the binding capability of the protein. The substrate used for this analysis was, as for the previous radioactive enzymatic assays, the one named ssRNA1. To set-up the assay conditions, a fix amount of RNA was incubated with increasing amounts of Dbp7 protein. In the all the cases, the gel shift assay indicated that protein- RNA complexes were formed (Figure 60). Two different complexes were formed. We also evaluated whether the mutant Dbp7[K197A] was able to bind RNA. Thus, similar assays as the above described were performed with the three different concentrations of Dbp7[K197A] protein. As before, two different RNA-protein complexes could be identified (Figure 60). 107 RESULTS S Figure 59. Gel shift assays of wild-type Dbp7 and Dbp7[K197A] protein. A ssRNA 1 at 1 nM was incubated during 15 min with increasing concentration of recombinant Dbp7 wild-type or Dbp7[K197A] protein (0.4, 0.6 and 0.8 µg). As a control, we used the ssRNA 1 substrate, which was incubated with no protein during the assay. Positions of free RNA and protein-bound to RNA forming complexes are shown at the bottom of the gel and at the top of the gel, respectively. 5.5.7.- Helicase activity of the recombinant Dbp7 wild-type and Dbp7[K197A] in vitro Most of the DEAD-box RNA helicases are able to unwind RNA duplexes (dsRNA) in an ATP-dependent manner. Typically, RNA helicases need a single stranded RNA region (overhand), where the helicase binds, to open the dsRNA. Some RNA helicases have a polarity or directionality, that is defined as the direction (5´-3 or 3´-5´) of the helicase movement on the ssRNA along which it is moving. However, others are capable of bind to both ssRNA ends independently and move in both directions (5´-3 or 3´-5´) along the ssRNA. In the case of Dbp7, so far it is unknown if the enzyme is able to unwind RNA duplexes. To test the unwinding activity of Dbp7 in vitro, we used two potential substrates, a dsRNA with an 3´-overhand (dsRNA 3´) and dsRNA with an 5´-overhand (dsRNA 5´). Since we did not know the adequate concentration of protein to carry out the assay, we decided to first perform the assay with a range of protein concentration from 0 to 0.8 µg. The fraction of wild type protein used in the assay was the number 11 from molecular exclusion chromatography. The fraction of the mutant [K197A] selected was the fraction equivalent to fraction number 11 in Dbp7 wild-type. As shown in the Figure 60A and 60B, Dbp7 wild-type and Dbp7[K197A] mutant had ATP-dependent helicase activity. Concerning to the protein concentration for the time of assay established (3 minutes), in both cases, the amount of ssRNA increased as the protein concentration increased, but at certain concentration of protein (0.7 µg), the amount of ssRNA generated did not increase even if the concentration of protein continued increasing (Figure 60C). 114 RESULTS S 5.7.- In vitro properties of Dbp9 5.7.1.- Expression and purification of recombinant Dbp9 wild-type and [K68A] mutant variant Dbp9 A C-terminal His6-tagged Dbp9 fusion protein was constructed by introduction of the open reading frame of DBP9 in the plasmid pET21b, under the control of an inducible T7 promoter. We also constructed a Dpb9 mutant variant, which harbour the point mutation K68A in motif I. To do so, the pET21b-His6-DPB9 plasmid was mutagenized in vitro. Each plasmid was transformed into the E. coli BL21 Codon Plus. Transformed cells were grown, proteins overexpressed after IPTG induction and, recombinant proteins purified by affinity chromatography. To the affinity chromatography a nickel (Ni-NTA) agarose resin or a 5 mL HisTrap FF column (GE Healthcare) coupled to a FPLC system were used. The elution fractions from the affinity chromatography were run in a denaturing polyacrylamide gel, in which a prominent protein band of about 70 kDa was observed (Figure 66). This molecular mass is similar to that estimated by sequence analysis (about 68 kDa). The band expected to be Dbp9 in the gel was confirmed by western blot using specific anti-polyhistidine antibodies (data not shown). Several other bands were also observed in the polyacrylamide gel. In the case of the FPLC purification, the peak observed in the chromatography was coincident with the elution peak in the gel. 115 RESULTS S Figure 66. Purification of Dbp9 wild-type by affinity chromatography. (A) and (B) 10% SDS-PAGE gel showing the fractions obtained by affinity purification with Ni-NTA agarose resin or by a HisTrap FF column in a FPLC system, respectively. W1, W2, and W3 indicates wash volumes 1, 2 and 3. E1, E2, E3, E4, and E5 indicates elution volumes 1, 2, 3, 4 and 5 respectively. I, indicates input, std, protein standards and U, unbound proteins (C) Affinity purification chromatogram of recombinant Dbp9 wild-type obtained by affinity purification in a FPLC system. Dashed lines indicate the interval of fractions in which Dbp9 protein elutes. This interval corresponds to fractions 35 to 47. Fractions 37, 38, 39, 40 and 41 (marked in red in 116 RESULTS S the gel B) were pooled and injected in a Superdex 200 HiLoad 16/60 column. The flow through and elution peaks are indicated. In order to detect multisubunit complexes, selected fractions from the affinity column from 37 to 41 in the example of Figure 66, were pooled and subjected to molecular exclusion chromatography. When calculated the molecular mass of the main peak of Dbp9 obtained by exclusion chromatography (fractions 16 to 29), the value was of about 77 kDa (Figure 67B). Thus, these results suggested that the wild-type Dbp9 protein likely eluted as a monomer. In addition, it was observed that Dbp9 was purified to electrophoretic homogeneity (Figure 67A). The lower band observed in SDS-PAGE gels (see red arrow) was also a product of Dbp9, as it was also revealed by anti-polyhistidine antibodies (data not shown). 117 RESULTS S Figura 67. Purification of recombinant Dbp9 wild-type by molecular exclusion. (A) A 10% SDS-PAGE gel showing the fractions obtained after the chromatography. In red is marked the fraction eluting at about 77 mL. I indicates input and std protein standards (B) Chromatogram corresponding to a typical molecular exclusion chromatography of Dbp9 wild-type. We also overexpressed in E. coli and purified the Dbp9[K68A] mutant version. The affinity purification profile by FPLC was very similar to that of wild-type Dbp9. Thus, an apparent molecular mass of about 70 kDa was calculated by SDS-PAGE for this variant, which is identical to that obtained for the Dbp9 wild-type protein. As above, the molecular exclusion profile obtained for the Dbp9[K68A] protein, and the further behaviour of the fractions obtained by SDS-PAGE gel were identical to the equivalent ones obtained for wild-type protein (data not shown). 5.7.2.- ATPase activity of Dbp9: colorimetric assay In order to determine whether the DEAD-box Dbp9 protein is able to hydrolyze ATP, we performed ATPase assays following the colorimetric procedure previously used for recombinant Dbp7 protein (see Section 5.5.2 from Results), which evaluated the amount of Pi released during the course of the reaction. We also optimized some parameters of the assay, such as the reaction time or the temperature of the assay. For this section, was used the Dbp9 protein sample contained in the fraction named "elution 1" from an affinity chromatography by Ni-NTA agarose resin (see Figure 66A). A) ATP concentration We evaluated different concentrations of this nucleotide in the absence of exogenous RNA in the assay, as we suspected the presence of RNA in the preparations (see also the results obtained for recombinant Dbp7). As shown in the Figure 68, and as expected, the amount of ATP hydrolyzed positively correlated with the initial ATP concentration without saturation. To normalize respect the assay conditions used for recombinant Dbp7, we arbitrarily decided to use a concentration of 0.5 mM of ATP for further assays. 118 RESULTS S Figure 68. Plot of ATPase activity of Dbp9 wild-type with respect to ATP concentration. A colorimetric assay was used, which measures the amount of Pi released (y axis) at different ATP concentration (x-axis) during the assay. In the assays, 1.5 µg protein was used in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped after 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. B) Protein concentration As shown in the Figure 69, the protein concentration positive correlated with the Pi released. In other words, the higher concentration of protein, the higher ATP hydrolysis was also detected. For further analysis, 3.5 µg of protein were employed per assay. 119 RESULTS S Figure 69. Plot of ATPase activity of Dbp9 wild-type with respect to protein concentration. A colorimetric assay was used, which measures the amount of Pi released (y axis) at different protein concentration (xaxis). The indicated amounts of protein were added to the reaction mixture; 0.5 mM ATP was used in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped after 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. C) Reaction time We next assayed the relation between time and ATPase activity. As shown in Figure 70, and as expected, we obtained a positive correlation between the incubation time and the concentration of Pi released. Figure 70. Plot of ATPase activity of Dbp9 wild-type along the time. A colorimetric assay was used, which measures the amount of Pi released (y-axis) at different times (x-axis). In the assays, 3.5 µg protein, and 0.5 mM ATP were used in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were performed at 30 oC and stopped at the indicated times. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. D) Temperature We also performed ATPase assays at a range of different temperatures from 22 °C to 47 °C. As shown in Figure 71, all the temperatures assayed gave rise to the similar level of ATP hydrolysis 120 RESULTS S Figure 71. Bar diagram of ATPase activity of Dbp9 wild-type at different temperatures. A colorimetric assay was used, which measures the amount of Pi released (y-axis) at different temperatures (x-axis). Reactions contained 3.5 µg of protein, and 0.5 mM ATP in a buffer containing 2 mM Mg2+, pH 6.5. Note that no exogenous RNA was added. Reactions were incubated at the indicated temperatures and stopped after 10 min of incubation. The amount of free Pi was calculated using the linear regression obtained upon measuring the values given by different concentrations of NaH2PO4 in the colorimetric assay. 5.7.3.- ATPase activity of Dbp9 wild-type and Dbp9[K68A] proteins: radioactive assays To evaluate the dependence on RNA by the ATPase activity, we performed radioactive assays, which were previously used for the biochemical characterization of Dbp7 wildtype and the mutant variant Dbp7[K197A]. To this aim, we used fractions obtained after exclusion chromatography of Dbp9 and Dbp9[K68A] protein (i.e. from 16 to fraction 22). As RNA substrate, we used the single stranded RNA (ssRNA 1) of 58 nucleotides long that was previously used for the equivalent assays for Dbp7 (see Section 4.8 from Material and Methods, and Table 3). Figure 72 shows the results obtained. Dbp9 did not present ATPase activity in the presence of ssRNA 1 as substrate (Figure 72A). Similar results were obtained for the Dbp9[K68A] protein (Figure 72B). 121 RESULTS S Figure 72. ATPase activity assay for Dbp9 wild-type and Dbp9[K68A] proteins. (A) and (B) Thin layer chromatography of the reaction products of the assays for Dbp9 wild-type and Dbp9[K68A] proteins, respectively. The assay contained 0.1 µg protein, 0.1 mM non-radioactive ATP, 0.2 µM [α-32P] ATP and, 3 µM ssRNA 1 in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. No RNase A was added during the protein purification. The upper spots corresponded to the amount of ADP obtained from the ATP hydrolysis and the lower spots the remained ATP, which was not hydrolyze along the duration of the assay. As a discrete ATPase activity was measured, assays in the presence of a different substrate were performed, in particular we made use of the DNA oligonucleotide previously used in the Dbp7 characterization, the oligonucleotide poly dT30. With this substrate, a more robust ATPase activity was detected using the same fractions used above (Figure 73 A and B). As shown in the quantification panel, no differences at the level of activity could be detected among the wild-type and the mutant Dbp9 (Figure 73C). 122 RESULTS S Figure 73. ATPase activity assay for Dbp9 wild-type and Dbp9[K68A] proteins using DNA oligonucleotide. (A) and (B) Thin layer chromatography of the reaction products of the assay for Dbp9 wild-type and Dbp9[K68A], respectively. The assay contained 0.1 µg protein, 0.1 mM ATP, 0.2 µM [α-32P] ATP, and 0.5 µM poly dT30 in a buffer containing 10 mM Mg2+, pH 8.0. The reaction time was of 1 hour at 30 oC. No RNase A was added during the protein purification. The upper spots corresponded to the amount of ADP obtained from the ATP hydrolysis and the lower spots the remained ATP, which was not hydrolyse along the duration of the assay. (C) Quantification of the ATPase activities detected in A and B. For this, the percentage of ADP released respect the ATP remained was calculated. 123 RESULTS S 5.7.4.- RNA binding and unwinding activities of the recombinant Dbp9 wild-type protein in vitro We only performed preliminary experiments regarding these activities using the same parameters previously established for the characterization of Dbp7 wild-type. The capacity to bind to RNA of the fractions 17, 18 and 19 of Dbp9 wild-type was measured in vitro using the ssRNA1 as substrate. However, in contrast to Dbp7, Dbp9 did not form RNA-protein complexes in any of the fraction assayed (data not shown). Concerning to the helicase activity, there were previous evidences that Dbp9 was able to unwind duplexes of nucleic acids even in absence of NTP. Interestingly, it was also reported that recombinant Dbp9 protein presented unwinding activity dependent of ATP hydrolysis on duplexes of DNA or of DNA-RNA (39). To test in vitro the unwinding activity of our Dbp9 wild-type protein preparation, we used the fraction 18 of the wildtype Dbp9 protein sample. As a substrate, we used a dsRNA with an overhand in the 3´ end (dsRNA 3´). However, Dbp9 did not present unwinding activity in the assay conditions (data not shown). 5.8.- Role of Dbp7 in the proper interaction with pre-ribosomal particles of a subset of snoRNAs In addition to protein trans-acting factors, about 80 smalls nucleolar RNAs are required to facilitate the accurate progression of ribosome biogenesis. To perform their activity, snoRNAs perform imperfect base pairing with the pre-rRNAs. Indeed, snoRNAs can be detected in stable association with pre-ribosomal particles upon immunoprecipitation or on sucrose density gradients (72, 80) . Our group has uncovered via synthetic lethal analysis a defined functional network involving the DEAD-box proteins Dbp6, Dbp7 and Dbp9, as well as the ribosomal protein uL3 and protein trans-acting factors Rsa3, Nop8, Rsa1, Npa1 and Npa2, all of them involved in early steps during the maturation of pre-60S ribosomal particles (18, 20, 73, 129 DISCUSSION 6.-DISCUSSION The yeast Dbp7 protein is classified as a DEAD-box ATP-dependent RNA helicase according to its amino acid sequence. To date, the knowledge about Dbp7 is scarce; only a few papers have been published containing information about this protein (17, 18, 22, 71). In 1998, Daugeron et al., showed that Dbp7 was a nucleolar protein and that its absence resulted in slow-growing cells with reduced amounts of LSUs and accumulation of 80S and polysome half-mers. Thus, they, therefore, concluded that Dbp7 is involved in LSU biogenesis (17). In order to recapitulate some of the experiments by these researchers, we constructed a null haploid mutant strain lacking Dbp7 (dbp7Δ). This mutant strain was obtained by sporulation of the same heterozygous diploid strain (dbp7Δ/DBP7) used in above mentioned study (17). The phenotypes of dbp7Δ and dbp7-1 strains were fully comparable. Additionally, we have generated a Dbp7-(GA)5-3xyEGFP construct that confirms that Dbp7 localises in the nucleolus. In growth experiments, dbp7-1 and dbp7Δ showed, as expected, very similar phenotypical behaviour. Thus, both strains showed a slow growth in YPD medium, being this phenotype more pronounced at high (37 oC) or low (22 or 18 oC) temperatures than at 30 oC. The dbp7Δ strain showed aberrant polysome profiles, which displayed a decrease in the peak corresponding to the free LSU, and half-mers in the 80S peak and polysomes peaks. This aberrant profile was indistinguishable to that previously reported by Daugeron et al. with their dbp7-1 strain. Therefore, our results confirm the fact, drawn by Daugeron et al., that nucleolar Dbp7 is a quasi-essential protein involved in LSU synthesis. DEAD-box helicases usually contain the N- and C- terminal extensions that constitute protein or RNA binding platforms (19), providing the respective binders the specificity that the RNA helicases must harbour in vivo. In some cases, these extensions may also contain sequences to target the proteins to their functional location. Dbp7 contains two long N- and C- terminal extensions flanking the RecA-like domains. Although Dbp7 was 130 DISCUSSION previously shown to be localized in the nucleolus (17), the mechanism that allow Dbp7 to travel to this subcellular compartment is still unknown. In our work, we have demonstrated the presence of an NLS in the N-terminal extension of Dbp7; this sequence allowed Dbp7 to reach the nucleus but not the nucleolus. Thus, another determinant must be present in the primary sequence of the protein to target Dbp7 to the nucleolus. Moreover, working with constructions of full length Dbp7, we concluded that this NLS is only necessary but not essential for the optimal transport of Dbp7 to the nucleus. Indeed, a Dbp7 construct lacking this NLS was still detected in the nucleus, although was also enriched in the cytoplasm, indicating that Dbp7 may contain additional sequences able to optimally bring the protein into the nucleus. These additional sequences must also map in the N-terminal tail of the protein, as a Dbp7 construct lacking the complete N-terminal extension of the protein is fully retained into the cytoplasm. The fluorescent microscopy data are complemented by the phenotype of a strain lacking the NLS sequence (HA-dbp7ΔNLS), which showed an apparent wild-type growth phenotype and no LSU shortage at the level of polysome profiles. We assumed that, therefore, the herein identified NLS must be complemented by a second NLS present in the N-terminal end of the Dbp7 sequence to target it to the nucleus. Alternatively, Dbp7 can be redundantly imported to the nucleus bound to a second protein, a phenomenon known as piggy-backing. The deletion of the complete N-terminal extension, or deletions of different size in the C-terminal extension of Dbp7 led to slow growth phenotype, decrease in free LSUs, and 80S and polysome half-mers. However, the truncated Dbp7 proteins showed the same sedimentation behaviour in low Mg2+ sucrose gradients than the full length Dbp7 protein. Therefore, the terminal extensions of Dbp7 seems not to be involved in the association of Dbp7 with pre-60S particles, but instead, provide other roles to the in vivo function of the protein, as indicated by the slow growth and ribosome biogenesis defects of the mutants in the N- and C- terminal extensions. These roles might consist on the 131 DISCUSSION recruitment of other factors to execute whatever restructuring reaction might exert Dbp7 in vivo or the optimisation of the enzymatic catalytic activities of the protein. We have also analysed the role of some of the conserved motifs within the two RecA- like domains of Dbp7. The amino acid residues that belong to the conserved motifs vary in importance. It is known that some of them carry out specific functions in ATP binding and hydrolysis, nucleic acid binding and unwinding or coordination between the ATP hydrolysis and the unwinding functions (7, 15, 80). For other RNA helicases, it has been reported that mutations in motif I, either in the first alanine residue, in the conserved lysine, or in the last threonine, reduce the affinity towards ATP and its hydrolysis (15). In motif VI, the first and second arginine residues could bind the -phosphate of the ATP, therefore, mutations in these residues affect ATP binding and hydrolysis (15). The functions that carry out the different residues (ATP binding and hydrolysis, and nucleic acid binding, and unwinding) are possible due to the interactions established between the different conserved motives (15). To analyse the importance of the lysine of the motif I and the second arginine of the motif VI, we constructed the mutant strains HA-dbp7[K197A] and HA-dbp7[R553A] by site-directed mutagenesis. Usually, the equivalent changes of amino acids in the motifs I and VI of others DEAD-box helicases seriously affect the functionality of the respective proteins in vivo, and the binding and hydrolysis of ATP activities in vitro. Thus, DEAD-box helicases mutated in equivalent residues in motif I such as K92A in Has1, K52A in Dbp8, and K52R in Dbp8 were unable to complement the corresponding null mutants (30, 70). In motif VI, the mutation R376A in Has1 also showed a lethal phenotype (70). In our series of experiments, the growth of the HA-dbp7[K197A] and HA-dbp7[R553A] strains was significantly slower than that of the wild-type strain at 30 oC, although not as drastically affected as a complete loss-of-function mutation does. In agreement, the HA- dbp7[K197A] and HA-dbp7[R553A] showed a decrease in the peak of the free LSU subunits, with the appearance of half-mers polysomes only in HA-dbp7[K197A] strain. We have purified recombinant Dbp7 protein to electrophoretical homogeneity. Our in vitro experiments showed that wild-type Dbp7 had ATPase activity dependent of ssRNA. 132 DISCUSSION Strikingly, Dbp7 showed ATPase activity when using a ssDNA substrate, a poly dT of 30 nucleotides. Dbp7 is not the only DEAD-box protein that has been described to use DNA as a substrate since a similar case was observed for yeast Dbp9, whose ATPase activity was stimulated by DNA (39). Another examples of protein that use DNA in ATP hydrolysis are Drosophila Vasa or HCV CAP-Rf proteins and S. cerevisiae Mer3 (39). We have also evaluated the ATPase activity of the Dbp7[K197A] mutant protein and surprisingly, it showed similar rates of ATPase as the wild-type protein. This fact could explain why the HA-dbp7[K197A] mutant strain has a very slight growth defect as its catalytically activities seems not to be impaired. The KM of Dbp7 wild-type measured using the deoxynucleotide poly dT of 30 nucleotides was about 300 µM. This KM is in the range of the constant calculated for others DEAD- box helicases purified from E. coli as Has1 (440 µM), Ded1 (300 µM), Prp5 (438 µM), or eIF4A (470 µM) (15, 34, 70). The KM of Dbp7[K197A] protein for the same DNA substrate was about200 µM, even lower than that of the wild-type protein. In contrast, mutations in the equivalent amino acid in others ATP-dependent RNA helicases generate an increase in the KM values, and a decrease in the ATPase activity. Such is the case of Has1[K92A], which showed 2-5% of the wild-type activity and almost 20 times higher KM for ATP, or Dbp8, whose mutants Dbp8[K52A] and, Dbp8[K52R] showed a decrease  10-fold in ATPase activity respect the wild-type protein (30, 70). The function of this lysine of the motif I is coordinating the β-phosphate of the ATP, stabilizing the binding of ATP-Mg2+ during the catalysis (80). It is known that not only motif I but other motifs as F, II, V and VI participate in ATP binding and hydrolysis (7, 15, 80). Thus, we speculate, as the most parsimonious explanation, Dbp7[K197A] protein did not display a decreased ATPase activity, because the change K by A does not distort the motif of the enzyme and therefore its activity. Another plausible explanation is that Dbp7[K197A] could have a redundant motif with motif I, which could be the one promoting ATPase activity. Finally, we could not discard that other residues can replace the function of K197; K184, A181 or T188 are alternative residues able to conform a functional non-orthodox motif I. It is known that the two last residues are also involved 133 DISCUSSION in ATP binding and hydrolysis. In fact, in some enzymes the last threonine of the motif I is found in the vicinity of the Mg2+ ion, being able to also stabilize the binding ATP-Mg2+ (15). Motif VI plays an important role in ATPase and helicase activities through the communication with the motif II. Mutants in the second arginine of the motif VI in Dbp7 (Dbp7[R553A]) showed slightly higher ATPase activity than the wild-type protein. However, in other DEAD-box helicases such as eIF4A, equivalent mutations of the conserved arginine residues of this motif (R359Q, R362Q, R365Q) impair ATP hydrolysis, RNA binding, and helicase activity, likely due to the loss of the coordination between motif VI and motif II (15, 63). Schneider and Schwer in 2001, showed that the mutation of the second arginine of motif VI in Prp22 drastically reduced the ATPase activity of the protein (77). Additionally, Dbp9 mutants in the second arginine of motif VI (R414I, R414T and R414K) did not have ATPase activity (39). Maybe the increase of the ATPase activity of the Dbp7[R553A] protein is caused by a communication breakdown between motifs VI and II. In this regard, it is worth to note that the amino acid sequence of the motif II of Dbp7 deviates (DEGD) respect with the canonical sequence (DEAD). The ATPase activity of Dbp7 had a particular behaviour, since the increase in the substrate concentration generates a decrease in the ATP hydrolysis velocity. It is possible that this behaviour relates to the non-processive helicases mechanism. The non-processive enzymes only unwind about 6-10 base pairs of RNA duplex in each ATP hydrolysis cycle (13, 19, 35). When the short duplex is opened the enzyme can inhibit the reaction if the unwinding activity is not required yet. Nonetheless, it is more probably that this behaviour, although unclear for us, could be due to product inhibition. Upon ATP hydrolysis, two products are generated: ADP and Pi. In this scenario, the ADP could bind to the ATP binding pocket of the enzyme and inhibit the reaction. This is apparently the case for the inhibition exerted by the recombinant Has1 protein (70). However, we were unable to detect inhibition by ADP at the concentration used. We cannot discard a possible inhibition by Pi. It is possible that the Pi could bind to a second site in the enzyme and act as an allosteric inhibitor. This mechanism could avoid the 134 DISCUSSION unnecessary use of the substrate and the products accumulation. An inhibition by substrate (ATP) is also possible, but in this case, it may exist a negative cooperative relationship between protein and substrate; when the substrate binds to the protein the affinity of the protein by the substrate decreases. In this regard, the DEAD-box helicase Has1 contains a consensus sequence of the motif I localised downstream of the motif VI (383GTKGKGKS390), which is probably also involved in ATP binding. It was demonstrated that mutation K389A in this second Walker A motif increased the ATP hydrolysis rate of the Has1 mutant enzyme. Interestingly, an in vivo K389A mutant had no effect on cell growth. This fact suggests that Has1 has two putative motifs for ATP binding; one of high and a second of low ATP affinity. According to this possibility, ATP can be accommodated in two ATP binding sites: the site in the motif I (high ATP affinity) or the site in the motif downstream of the motif VI (low ATP affinity). It has been suggested that when the lysine in the second Walker A motif is mutated (K389A), ATP could only bind in the high affinity site and the hydrolysis occur faster (70). Despite this possibility, we cannot find a possible second ATP binding motif with a sequence consensus of motif I in Dbp7. In contrast, Dbp9 contains downstream of motif VI a possible consensus sequence of the motif I (415ARGGKT420). Therefore, the presence of a putative second ATP binding motif could be a characteristic shared by different members of the DEAD-box family of proteins. According to our in vitro experiments, Dbp7 had also the ability to bind RNA. We detected two different bands corresponding to different Dbp7-RNA complexes in gel shifts assays. Similar results were obtained with the recombinant Dbp7[K197A] mutant protein. This result suggests that Dbp7 could bind the RNA substrate in form of a monomer or a multiprotein complex. In fact, our molecular exclusion chromatography data suggest that Dbp7 could have a tendency to form dimers. ATP dependent RNA helicases are able to open RNA duplexes following different mechanism. Usually, RNA helicases need a single stranded RNA region (overhand), where the helicase binds, to open the dsRNA but in other cases, these enzymes are able to open blunt ends RNA duplexes. Moreover, some RNA helicases can show polarity or 135 DISCUSSION directionality, that is defined as the direction (5´-3 or 3´-5´) of the helicase movement on the ssRNA along which it is moving. Our in vitro experiments also demonstrated that wild-type Dbp7 and Dbp7[K197A] proteins are able to unwind RNA duplexes in both 5´- 3´and 3´-5´directions. The RNA helicase activity of Dbp7 is comparable to that of others DEAD-box RNA helicases, such as Ded1, which is able to unwind blunt ends duplexes (34). Similarly, Has1 unwinds RNA/DNA duplexes in both directions, showing higher extend of unwinding with the 5´ RNA/DNA duplexes (70). In light of the in vitro results, Dbp7 has ATPase activity, it is able to bind RNA, and displays helicase activity, as several others DEAD-box RNA helicases do. However, our results show some Dbp7 features that make it special. Such features are related with the conserved residues that contain the DEAD-box proteins. Firstly, the residue K197 of motif I seems to be no essential for ATP binding and hydrolysis, while in others DEAD- box RNA helicases this motif is strictly essential for ATPase activity; secondly, the conserved motif II, which is DEAD in the most of DEAD-box helicases, is deviated to DEGD in Dbp7. To better understand why the mutation K197A in Dbp7 did not affect the ATPase activity of the protein in vitro, we decided to study the equivalent mutation in a Dbp7-related DEAD-box helicase. The DEAD-box helicase chosen was Dbp9 because i) the sequence of the motif II of Dbp9 (DEVD) is also deviated respect the canonical sequence (DEAD), as it occurs in Dbp7 (DEGD); ii) both proteins are involved in LSU biogenesis, most likely, in the same early step of this process; and iii) Dbp7 and Dbp9 genetically interact, as the null dbp7 allele synthetically enhances the slow growth phenotype of distinct dbp9 alleles (16, 17, 18). The enzymatic activities of Dbp9 were previously studied by Kikuma et al. in 2004. They determined that the ATPase activity of Dbp9 is stimulated by DNA and inhibited by RNA, Dbp9 presents RNA/DNA helicase activity dependent of ATP, and mutations in the second arginine residue in the motif VI abolish this helicase activity (39). In consonance, our results showed that Dbp9 does not display ATPase activity in presence of a ssRNA, unlike Dbp7. As it occurs for Dbp7, ATPase activity could be detected for recombinant Dbp9 when a DNA poly dT oligonucleotide is included in the 136 DISCUSSION reaction mix. Surprisingly, as it also occurs for Dbp7, no differences could be detected for the ATPase activity of Dbp9 wild-type and mutant Dbp9[K68A] protein, which harbours a mutation in motif I. Dbp7 and Dbp9 show similar features that could led us to understand why the mutants in the lysine of the motif I have ATPase activity at the same level as the wild-type proteins. However, it is necessary to perform more in vitro and in vivo assays to decipher the action mechanism of these enzymes. In the last part of this thesis, we have explored the precise in vivo role of Dbp7 in LSU biogenesis. Dbp7 is part of a functional interaction network that also includes Dbp6, Dpb9, Npa1, Npa2, Nop8, Rsa3, Rsa1, and the ribosomal protein uL3 (18). Among these factors, Npa1, Npa2, Rsa3, Nop8 and Dbp6 form the Npa1 complex. (74). Dbp7 is not part of this physical complex, but it could interact with it through the binding to Dbp6 or Dbp9, as a part of a helicase heterodimer. The function of the Npa1 complex has been related to the control of snoRNA/pre-RNA interactions; it has been shown to be important for the structural rearrangement of rRNA domains I and II during the earliest reaction of formation of LSU precursor particles (37). It is known that Npa1 interact with several snoRNAs involved in the decoding centre and the peptidyl transferase centre modifications in the environment of the binding sites of these snoRNAS on 25S rRNA (37). Thus, Dbp6, Dbp9, and Dbp7 would be the responsible for the establishment of the binding or dissociation of these snoRNAs from 25S rRNA (37). We determined that mutations in the 25S rRNA region, where snR190 binds, are able to suppress the severe slow growth of the dbp7∆ mutant (21). Apparently, these mutations will relax or impede the interaction of snR190 with the 25S rRNA, suggesting that the in vivo function of Dbp7 must be related to the dissociation of distinct snoRNAs from early pre-60S ribosomal particles. To explore this view, a mutant lacking Dbp7 and harbouring a snR190, apparently unable to base pair to its binding site in 25S rRNA, was constructed. When we analysed the growth of this double mutant, a clear suppression of the growth defect phenotype of the dbp7∆ mutation was observed. This result suggests that in the absence of Dbp7, snR190 exert a negative role in ribosome biogenesis, and therefore in growth. 137 DISCUSSION To test the association status of different snoRNAs with pre-ribosomal particles in the presence or absence of Dbp7, we obtained cell extracts of isogenic wild type and dbp7Δ, which were fractionated through low Mg+2 sucrose gradients. RNA was prepared from the fractions and analysed by northern blotting with specific probes against a set of snoRNAs included snR190. Interestingly, this analysis showed a clear and specific accumulation of snR190 in the fractions corresponding to pre-60S and 60S r-particles in the dbp7Δ strain, compared to the wild-type strain. These data strongly suggest that Dbp7 is required for snR190 dissociation from pre-60S ribosomal particles. Therefore, we envisage a scenario where the loss-of-function of Dbp7 impairs snR190 dissociation from early pre-60S ribosomal particles leading to formation of aberrant particles that are unable to progress in their maturation, and consequently being subjected to degradation. In this scenario, the simultaneous absence of a productive interaction of snR190 with these particles might partially alleviate the conformational constraints in their maturation. Further experiments are required to demonstrate this view. 7. CONCLUSIONS