Study of Pol5 protein from Saccharomyces cerevisiae and its human ortholog MYBBP1A in ribosome biogenesis

Departamento de Genética Universidad de Sevilla La Doctoranda V.° B.° El director de la Tesis Jesús de la Cruz Díaz Ana Ramos Sáenz Memoria presentada para aspirar al grado de Doctora en Biología Molecular, Biomedicina e Investigación Clínica Study of Pol5 protein from Saccharomyces cerevisiae and its human ortholog MYBBP1A in ribosome biogenesis This work was funded by a FPU fellowship from the Spanish Ministry of Universi�es (FPU17/05779), and project grants from the Andalusian Regional Government (P20_00581 and US-1380394) and the Spanish Government (PID2019-103850GB-I00). A mis padres, por creer en mí y apoyarme siempre “There's nothing in this universe that can't be explained. Eventually.” Gregory House- House, M.D. INDEX 1.- ABSTRACT .................................................................................................................... 1 2.- INTRODUCTION ........................................................................................................... 7 2.1.- The ribosome: structure and func�on ................................................................... 9 2.1.1.- Ribosomal proteins ......................................................................................... 11 2.2.- Ribosome biogenesis in Saccharomyces cerevisiae ............................................. 13 2.2.1.- Trans-ac�ng or ribosome assembly factors .................................................... 16 2.2.2.- Synthesis and processing of pre-ribosomal RNAs........................................... 17 2.2.3.- Assembly of ribosomal subunits ..................................................................... 20 2.2.3.1.- Assembly and matura�on of the SSU processome ................................... 20 2.2.3.2.- Nucleocytoplasmic matura�on of pre-40S ribosomal par�cles ................ 21 2.2.3.3.- Nucleolar matura�on of pre-60S ribosomal par�cles ............................... 23 2.2.3.4.- Nucleoplasmic matura�on and export of pre-60S ribosomal par�cles .... 26 2.2.3.5.- Cytoplasmic matura�on of pre-60S ribosomal par�cles ........................... 28 2.3.- The yeast ribosome assembly factor Pol5 ............................................................ 30 2.4.- The human MYBBP1A protein .............................................................................. 31 2.5.- Ribosome altera�ons: cancer and ribosomopathies ........................................... 34 3.- OBJECTIVES ................................................................................................................ 37 4.- MATERIALS AND METHODS ...................................................................................... 41 4.1.- Strains and microbiological methods ................................................................... 43 4.2.- Plasmids and oligonucleo�des ............................................................................. 43 4.3.- Mul�copy suppression analysis............................................................................ 44 4.4.- Polysome profiles, ribosomal subunit analysis and sucrose gradient frac�ona�on ...................................................................................................................................... 45 4.5.- RNA extrac�on and analysis of steady-state RNA levels ...................................... 46 4.6.- Protein extrac�on, western blot analysis and an�bodies .................................... 47 4.7.- Analysis of fluorescent proteins by microscopy ................................................... 47 4.8.- Affinity purifica�on of GFP-tagged proteins......................................................... 49 5.- RESULTS ..................................................................................................................... 53 5.1.- Func�onal study of the yeast Pol5 protein .......................................................... 55 5.1.1.- Pol5 is an essen�al protein ............................................................................. 55 5.1.2.- Pol5 is nucleolar .............................................................................................. 57 5.1.3.- Pol5 is associated to high-molecular-mass par�cles ...................................... 59 5.1.4.- Pol5 associates with pre-60S ribosomal par�cles........................................... 60 5.1.5.- Pol5 is required for 60S ribosomal subunit accumula�on .............................. 62 5.1.6.- Pol5 is required for pre-rRNA processing ....................................................... 65 5.1.7.- Pol5 deple�on impairs nucleo-cytoplasmic transport of pre-60S ribosomal par�cles ...................................................................................................................... 67 5.1.8.- The overexpression of the uL23 and eL27A ribosomal proteins suppresses the thermosensi�ve phenotype of the pol5-1 and pol5-3 mutants ................................ 70 5.1.9.- The apparent DNA polymerase ac�vity of Pol5 does not play a role in ribosome biogenesis .................................................................................................................. 73 5.2.- Study of the protein domains of Pol5 .................................................................. 75 5.2.1.- The whole sequence of Pol5 except the last α-helix is required for the synthesis of the 60S ribosomal subunit ..................................................................................... 78 5.2.2.- The N-terminal domain of Pol5 is required for efficient nucleolar localiza�on ................................................................................................................................... 82 5.2.3.- Pol5 associates with pre-ribosomal par�cles through its N-terminal domain ................................................................................................................................... 84 5.2.4.- Trans-complementa�on of Pol5 restores 60S ribosomal subunit synthesis... 87 5.2.5.- Trans-complementa�on does not totally restored the nucleolar localiza�on of the N-terminal domain mutant ................................................................................. 90 5.2.6.- The interac�on of N- and C-terminal truncated constructs may not be stable enough to be detected by co-purifica�on ................................................................. 91 5.3.- Study of the effects of the heterologous expression of human MYBBP1A in yeast ...................................................................................................................................... 93 5.3.1.- The expression of MYBBP1A has a nega�ve effect on growth ....................... 93 5.3.2.- Full-length MYBBP1A is nucleolar in yeast ..................................................... 97 5.3.3.- The nega�ve effect of MYBBP1A in yeast is related to the synthesis of 60S ribosomal subunits..................................................................................................... 97 5.3.4.- MYBBP1A associates with pre-60S ribosomal par�cles ............................... 100 5.3.5.- HA-MYBBP1A protein has a similar sedimenta�on distribu�on in sucrose gradients as HA-Pol5 protein ................................................................................... 101 5.3.6.- MYBBP1A has an effect on pre-rRNA processing ......................................... 103 5.3.7.- MYBBP1A expression causes the accumula�on of uL23 ribosomal protein in the nucleolus ............................................................................................................ 104 5.3.8.- Expression of MYBBP1A does not affect the subcellular localiza�on of Pol5 protein ...................................................................................................................... 105 5.3.9.- Pol5 associa�on with pre-ribosomal par�cles is not altered by expression of MYBBP1A ................................................................................................................. 107 5.3.10.- Overexpression of the eL27A r-protein does not suppress the nega�ve effect of MYBBP1A on growth ........................................................................................... 109 5.3.11.- Dele�on of BAS1 does not suppress the nega�ve effect of MYBBP1A on growth ...................................................................................................................... 110 6.- DISCUSSION ............................................................................................................. 113 7.- CONCLUSIONS ......................................................................................................... 129 8.- BIBLIOGRAPHY ........................................................................................................ 133 9.- PUBLICATIONS AND PRESENTATIONS ..................................................................... 157 6 ABSTRACT desprovistas de Pol5 muestran una acumulación de este precursor y un déficit de subunidades ribosómicas 60S. Hemos iden�ficado las proteínas ribosómicas uL23 y eL27A como supresores mul�copia de dis�ntas mutaciones de pérdida de función de Pol5, lo que sugiere que la función de Pol5 está relacionada con el correcto ensamblaje de estas proteínas ribosómicas en el dominio III del rRNA 25S. Por otro lado, también se ha descrito que Pol5 par�cipa en la síntesis de las subunidades ribosómicas 40S. En relación con esta afirmación, en esta Tesis hemos estudiado la idea de que Pol5 sea una proteína modular en la que cada una de sus funciones podría asignarse a un dominio específico de la proteína, tal y como se ha descrito para otros factores de ensamblaje ribosómico bifuncionales. Además, nuestros resultados muestran que la transcomplementación de Pol5 es posible cuando la proteína completa se recons�tuye a par�r de dos fragmentos truncados dis�ntos. MYBBP1A (Myb binding protein 1A) es el supuesto ortólogo mamífero de Pol5, y a pesar de toda la información sobre esta proteína como supresor de tumores y su relación con diferentes procesos celulares como el ciclo celular, la mitosis, el metabolismo o la senescencia, su papel en la biogénesis de ribosomas ha sido poco caracterizado. En este trabajo, hemos estudiado el papel de MYBBP1A en la biogénesis de ribosomas u�lizando la levadura S. cerevisiae como modelo. Nuestros resultados muestran que MYBBP1A se asocia con par�culas pre-ribosómicas 60S cuando se expresa de manera heteróloga en levadura. Sin embargo, MYBBP1A es incapaz de complementar la ausencia de su ortólogo esencial Pol5 en la levadura y, además, su expresión heteróloga provoca un efecto dominante nega�vo en el crecimiento de la levadura asociado a un defecto específico en la síntesis de la subunidad ribosómica 60S. 2.- INTRODUCTION 9 INTRODUCTION 2.1.- The ribosome: structure and func�on Ribosomes are the organelles that perform the transla�on process, which consists of decoding the informa�on contained in the messenger RNAs (mRNAs) to synthesize proteins [1]. Ribosomes are ribonucleoproteins composed of ribosomal proteins (r-proteins) and ribosomal RNAs (rRNAs), which are distributed in two ribosomal subunits (rsubunits): the small (SSU) and the large (LSU) one. They are also called 30S and 50S (prokaryotes) or 40S and 60S (eukaryotes) according to their sedimenta�on coefficient, and together they form the mature ribosomes, which are called 70S (prokaryotes) or 80S (eukaryotes) [2, 3]. Mitochondria and chloroplasts also have their own ribosomes [4, 5]. The SSU decodes the informa�on of the mRNAs and base-pairs their codons with the corresponding transfer RNAs (tRNAs), while the LSU catalyzes the forma�on of the pep�de bond. The nascent chain exits from the ribosome through the polypep�de exit tunnel (PET) located in the LSU [6, 7]. In the recent years, techniques such as X-ray crystallography and cryogenic electron microscopy (cryo-EM) have provided useful informa�on about the structure of r-subunits [8-11]. In the SSU, four structural domains are present: the head, the beak, the pla�orm, and the body. In the case of the LSU, the domains present are the L1 stalk, the P stalk, and the central protuberance (CP), the later being where the pep�dyltransferase center (PTC) is located (Figure 1) [11]. The rRNAs are mainly located in the interface between the two subunits, where the func�onal domains are situated, while the r-proteins are mainly located on the surface of the ribosome [12]. The rRNAs are the responsible for the cataly�c ac�vi�es, and for this reason the ribosome is considered a ribozyme [13]. Ribosomes are highly conserved throughout evolu�on, but they have gained complexity from prokaryotes to eukaryotes. A common core, where the func�onal domains are located, is very conserved, but in eukaryotes there is an increase in the number of r-proteins and their complexity, as a result of the presence of specific extensions that seem to be related to the associa�on of eukaryo�c specific factors. The eukaryo�c rRNAs also present expansion segments (ES) that are oriented to the surface, 10 INTRODUCTION whose func�on appears to be related to the associa�on of r-proteins and ribosome assembly factors. The presence of these eukaryo�c specific elements provides a possible explana�on for the differences observed in the transla�on ini�a�on process between eukaryotes and prokaryotes, since some of them interact with eukaryo�c ini�a�on factors [14]. The length of the ES varies even within the eukaryo�c kingdom, with human ribosomes having longer ES than yeast ribosomes, a scenario that also occurs for the number and length of r-proteins extensions (Figure 2) [15, 16]. Figure 1. Crystal structure of yeast 60S and 40S ribosomal subunits. (A) Solvent exposed view (le�) and interface view (right) of the 60S r-subunit. (B) Solvent exposed view (le�) and interface view (right) of the 40S r-subunit. CP, central protuberance. The r-proteins are colored in red and the rRNAs are in gray. Adapted from [17]. In this Thesis work, we have used the budding yeast Saccharomyces cerevisiae, a model organism that has been widely used to study the synthesis of ribosomes due to the conserva�on of the process. Working with this organism offers many advantages due to its rapid growth and the availability of a wide range of gene�c, biochemical and molecular tools, as well as the possibility of extrapola�ng the obtained results to other organisms. In S. cerevisiae, the LSU or 60S r-subunit is composed of 3 rRNAs (25S, 5.8S and 5S) and 46 r-proteins, and the SSU or 40S r-subunit is composed of 1 rRNA (18S) and 33 r-proteins [11]. head platform beak body L1 stalk CP P stalk solvent interface subunit interface A B 11 INTRODUCTION Figure 2. Composi�on and comparison of bacterial and eukaryo�c ribosomes. The common core shared by bacteria and eukaryotes consist of rRNA (light blue) and r-proteins (light red). Expansion segments of rRNAs are colored in blue. Domain-specific r-proteins, as well as extensions and inser�ons in these proteins, are colored in red. Dashed lines present in the common core indicate the posi�ons of flexible elements not solved in the X-ray structures. Taken from [14]. 2.1.1.- Ribosomal proteins The r-proteins are generally characterized by their small size, intrinsically disorder domains, and basic nature. They can interact with the rRNA (which is acidic) and neutralize its charges. As men�oned above, they present extensions that seem to op�mize the associa�on with the rRNAs and provide sites for the associa�on of specific factors [16]. Their basic nature, together with their disordered domains and flexible tails, predisposes r-proteins to form toxic aggregates when they stay orphans. Cells have different mechanisms to avoid the forma�on of these aggregates when the r-proteins are not properly assembled into the r-subunits, such as their co-transla�onal associa�on with general chaperones or with the so-called dedicated chaperones [18]. This last family of factors also par�cipates in the import into the nucleus and the correct assembly of their client r-proteins into nucle(ol)ar pre-ribosomal par�cles (r-par�cles) [19, 20]. Recently, it has been described that the absence of certain dedicated chaperones causes 12 INTRODUCTION the destabiliza�on of the mRNAs that encode their clients, sugges�ng some kind of auto regulatory mechanism for the expression of dis�nct r-proteins [21, 22]. The excess of rproteins, which are not subject to assembly, are efficiently eliminated by the ubiqui�n– proteasome system, in which the ubiqui�n ligase Tom1 ubiqui�nates residues in the rproteins that are only accessible when they are not assembled into the r-par�cles. This system is called “excess r-protein quality control” (ERISQ) [23, 24]. In S. cerevisiae, the majority of r-proteins are encoded by two paralogs genes, whose sequences are very similar or iden�cal in some cases [25]. The dele�on of one of them does not have the same effect that the dele�on of its counterpart and they do not express at the same levels. The existence of these paralogs genes opens the possibility of the forma�on of specialized ribosomes, that, for example, have a preference for the transla�on of a certain group of mRNAs [26-30]. The r-proteins also have extra-ribosomal func�ons, and it is even possible that the paralogs could play different roles in these func�ons (Figure 3) [31-34]. Figure 3. Representa�on of extra-ribosomal func�ons of ribosomal proteins in oncogenic pathways. Illustra�on of the rela�on of different human r-proteins with factors of pathways related to tumorigenesis. KD, knockdown. Taken from [35]. 13 INTRODUCTION In S. cerevisiae, there are 46 r-proteins in the 60S r-subunit and 33 r-proteins in the 40S r-subunit. The names of the yeast r-proteins are given in Table 1, which also includes the names of the bacterial and human orthologs and the new nomenclature established in the field [36], which we have used in this work. Table 1. Nomenclature of ribosomal proteins. Current and former names of r-proteins from the small (le�) and large (right) r-subunits. The classical names for bacterial, yeast and human rproteins are provided. Adapted from [36]. 2.2.- Ribosome biogenesis in Saccharomyces cerevisiae The process of ribosome synthesis requires a large amount of cellular resources. Under exponen�al growth condi�ons, each yeast cell produces around 2000 ribosomes per minute [37]. This process is �ghtly regulated and happens in a hierarchical manner. 14 INTRODUCTION To produce mature ribosomes, it is necessary to synthesize the rRNAs and the rproteins, but this process also requires the func�on of among 80 small nucleolar RNAs (snoRNAs) and more than 300 trans-ac�ng proteins or ribosome assembly factors (RAFs). Taking all this into account, most of the transcrip�onal ac�vity of the cell is dedicated to ribosome biogenesis [17, 37, 38]. This process starts in the nucleolus, where a precursor of the rRNAs (pre-rRNA) is transcribed and the first r-proteins and factors co-transcrip�onally assemble to the nascent pre-rRNA to form the earliest pre-ribosomal par�cles. Matura�on con�nues in the nucleoplasm and finalizes in the cytoplasm, where the pre-ribosomal par�cles undergo the final steps of matura�on with the incorpora�on of the last r-proteins and the dissocia�on of the last RAFs. These factors associate with and dissociate from the pre-ribosomal par�cles in a hierarchical manner, and they are not present in the mature ribosomes (Figure 4) [38]. Figure 4. Schema�c representa�on of yeast ribosome assembly. Ribosome assembly begins with the forma�on of the 35S pre-rRNA by RNA polymerase I and the co-transcrip�onal associa�on of r-proteins and RAFs, whose genes are transcribed by RNA polymerase II. All together, they form the earliest 90S pre-ribosomal par�cle, which undergoes different matura�on steps that generate the pre-40S and pre-60S r-par�cles, which con�nue their matura�on through the nucleolus and nucleoplasm with the assembly of r-proteins and the transient associa�on of RAFs. The 5S rRNA is synthesized by the RNA polymerase III. Matura�on of both r-subunits ends in the cytoplasm. Note that mature ribosomes lack RAFs. Taken from [8]. 15 INTRODUCTION In yeast, ribosomal DNA (rDNA) is located on the chromosome XII and each cell has approximately 150 tandem repeats of rDNA. Under exponen�al growth condi�ons, only about half of the repeats are ac�ve simultaneously. Altera�ons in these condi�ons can change the number of ac�ve repeats, and it has been proposed that this organiza�on provide a mechanism of protec�on against muta�ons, and that they maintain genome integrity [39, 40]. Each ac�ve repeat produces a 35S pre-rRNA (which, a�er processing, produces the mature 18S, 5.8S and 25S rRNAs) and a pre-5S rRNA. The three RNA polymerases are involved in the synthesis of ribosomes: RNA polymerase I transcribes the 35S pre-rDNA, RNA polymerase III synthesizes a pre-5S rRNA, and RNA polymerase II transcribes the genes for r-proteins and RAFs [41]. Ribosome produc�on depends on the growth condi�ons (nutrients availability, stress condi�ons…), and the TORC1 and protein kinase A (PKA) signaling pathways regulate the adapta�on of this process. These pathways regulate the synthesis of ribosomes by controlling the levels of pre-rRNAs and r-proteins in a co- and posttranscrip�onal manner [42, 43] (Figure 5). Figure 5. Regula�on of ribosome synthesis. The TORC1 and PKA signaling pathways regulate ribosome biogenesis. They control the transcrip�onal levels of r-proteins (RP), rDNA, and the ribosomal biogenesis regulon (RiBi). At the post-transcrip�onal level, these pathways regulate pre-rRNA processing, ribosome assembly, transla�on, and degrada�on of r-subunits. A feedback system regulates the genes of the r-proteins and ribosome biogenesis regulon (RBB). Taken from [42]. 22 INTRODUCTION This nucleoplasmic par�cle contains RAFs such as Rrp12, Pno1 (Dim2), Enp1 or Bud23, that are already bound in the SSU processome, and incorporates some other novel factors such as Dim1, Ltv1, Tsr1, Rio2 or Sxl9. Many of these factors are located protec�ng func�onal sites of the ribosome and some remain bound to the pre-40S rpar�cles un�l the cytoplasm. This avoids the premature forma�on of these domains un�l the pre-40S r-par�cle is exported. The Nob1 endonuclease, which performs the cleavage at site D, is also incorporated, but does not yet have access to this site [79, 85, 86]. The head is formed in the nucleus, but has a rotated posi�on regarding to its final conforma�on, which is achieved in the cytoplasm, where the beak is also completely formed [86, 87]. The pre-40S r-par�cles are exported to the cytoplasm through the nuclear pore complex (NPC). Pno1, Rio2 and Ltv1 present nuclear export sequences (NES) that allow the interac�on with the RanGTP-binding expor�n Crm1 [88-91]. Sxl9 has been proposed to mediate the associa�on of Rio2 with Crm1 [92]. The Mex67/Mtr2 mRNA export factor and Rrp12 also seem to be related with the pre-40S r-par�cles export [93, 94]. As men�oned above, Pno1 associates with the SSU processome, but its NES is not exposed un�l some conforma�onal changes take place, which avoid premature export of the rpar�cles [79]. In the cytoplasm, the pre-40S r-par�cle undergoes conforma�onal changes that lead to the forma�on of the beak, the rota�on of the head and the final matura�on of the func�onal domains. The RAFs Ltv1, Enp1, Tsr1 and Rio2 are released and Rio1 associates [86, 87, 95]. This induces the forma�on of the 80S-like ribosomes, in which a pre-40S r-par�cle and a mature 60S r-subunit associate and a transla�on-like cycle takes place, mediated by Fun12 (eIF5B). Pno1 is released and this allows the cleavage at site D by Nob1 to obtain the mature 18S rRNA (Figure 8). Forma�on of 80S-like ribosomes acts as a proofreading mechanism that ensures that only correctly matured 40S r-subunits acquire transla�onal competence [61, 62]. Finally, the last RAFs are released, and the last r-proteins are assembled, which allow the mature 40S r-subunit to enter in the transla�ng pool. 23 INTRODUCTION Figure 8. Model of the cytoplasmic matura�on steps of yeast pre-40S ribosomal par�cles. The final matura�on steps of pre-40S r-par�cles involve the associa�on and dissocia�on of several factors and the forma�on of the 80S-like ribosomes. The cleavage of 20S pre-rRNA at site D by the Nob1 endonuclease can take place in the context of these 80S-like ribosomes or a�er the dissocia�on of the pre-40S r-par�cle and the 60S r-subunit. Factors stably bound to pre-40S rpar�cles are shown in yellow, transient factors in magenta, mRNA in blue, r-proteins in white, pre-40S r-par�cles in light gray and 60S r-subunits in dark gray. Adapted from [61]. 2.2.3.3.- Nucleolar matura�on of pre-60S ribosomal par�cles The matura�on pathway of pre-60S r-par�cles is more complex than that of pre- 40S r-par�cles. At this point, a�er the cleavage at site A2, the pre-60S r-par�cles contain the 27SA2 pre-rRNA, which includes the sequences of the mature 5.8S and 25S rRNAs. 25S rRNA has six domains (I-VI), which fold following a hierarchical process. The domains I and II fold first, but the rest of them do not follow the transcrip�onal order for assembly (Figure 9). The 5.8S rRNA and the domains I and II form the solvent-exposed face of the pre-60S r-par�cles, the domains IV and V form the interface, and the domains III and VI interconnect all regions [96, 97]. The earliest par�cle that emerges a�er the cleavage at site A2 is very flexible, making it very difficult to obtain cryo-EM structures. The pre-rRNA suffers modifica�ons carried out by snoRNPs, and the different r-proteins and RAFs begin to associate with the 24 INTRODUCTION pre-rRNA in a co-transcrip�onal and hierarchical manner. Some of the first associated factors (Nop4, Noc2, Mak21 or Rrp5) assist the correct compac�on and folding of the pre-rRNA [96]. Rrp5 is par�cularly relevant because its C-terminal domain associates near site A2 while its N-terminal domain does it near site A3, coordina�ng the two cleavages [98]. Figure 9. Assembly sequence of the 25S pre-rRNA domains. The r-proteins and RAFs cotranscrip�onally associate with the nascent pre-rRNA. The first domains to fold are I and II, and then domain VI binds to them and to the 5.8S por�on of the pre-rRNA. The forma�on of the PET starts at this point (represented with a black circle) and ends when the domain V folds. Stable folding is represented by full colors while domains in the process of restructura�on are represented with some transparency. Taken from [99]. Other factors such as those that form the Npa1 module (Nop8, Rsa3, Dbp6, Npa1 and Npa2) and some RNA helicases (Mak5, Prp43, Dbp2, Dbp3, Dbp7 and Dbp9) are also involved in the correct 27SA pre-rRNA compac�on and folding, as well as in the release of some snoRNPs [100-104]. The Nsa1 module (Nsa1, Mak16, Rrp1 and Rpf1) also associates with the solvent-exposed face of the par�cles and stabilizes them [97]. To process ITS1 by the cleavage at site A3 and generate the 27SA3 pre-rRNA, the so-called A3-factors (Nop7, Erb1, Ytm1, Drs1, Nop15, Rlp7, Cic1, Pwp1, Nop12, Has1, Brx1 and Ebp2) must associate. Some of these factors do not bind near ITS1, sugges�ng that their func�ons are more related to achieving the correct structure to perform the cleavage [38, 105-109]. The A3 cleavage is coupled with the termina�on of the pre-rRNA transcrip�on and the processing of the 3’ETS, but how this happens is not yet clear [58]. 25 INTRODUCTION Cleavage at site C2 in ITS2 to generate the 7S and 25.5S pre-rRNAs requires the associa�on of the so-called B-factors (Nip7, Nop2, Rrs1, Rpf2, Spb4, Mak11, Dbp10, Nug1, Rlp24, Tif6, Nog1, Nsa2, Rsa4 and Noc3) [38, 110]. This family of factors contributes to the forma�on of the PET and the PTC, whose forma�on is coupled to the ITS2 processing [109, 111]. The AAA-ATPase Rix7 removes the Nsa1 module, allowing the forma�on of the ves�bule (outer part) of the PET [99, 112, 113]. The assembly of the uL23, eL31 and eL19 r-proteins forms the rim around the exit tunnel [114]. The L1 stalk begins to form with the interac�on of some factors with domains of the pre-rRNA, but it is located in the region that will become the CP, blocking the access of the 5S RNP [99, 115]. The AAA-ATPase Rea1 removes Erb1-Ytm1 [116-118]. Erb1 interacts with many other factors and its removal leads to most of the structural and composi�onal changes that happen when the par�cle is migra�ng from the nucleolus to the nucleoplasm (Figure 10) [99, 114]. Figure 10. Erb1 interac�ons. (A) Erb1 domain architecture. Interac�ons with rRNA are colored in gray. (B) Erb1 localiza�on in a par�cle and interac�ons with RAFs. (C) Schema�c 2D representa�on of Erb1 interac�ons with some r-proteins and RAFs. Taken from [99]. 26 INTRODUCTION 5S rDNA is transcribed by the RNA polymerase III and the 5S rRNA must associate with the uL18 and uL5 r-proteins to form the 5S RNP. The Syo1 chaperone co-imports these two r-proteins into the nucleus and assists in the 5S RNP assembly [20]. It associates with the domain V in the nucleolus with the assistance of the Rpf2 and Rrs1 factors [111, 119, 120]. The direc�onality of all these events is achieved by the presence of RAFs that are energy-dependent (e.g., ATP-dependent RNA helicases, AAA-ATPases and GTPases). The early pre-60S r-par�cles are formed in the nucleolus and, as they progress along the assembly pathway, they migrate to the nucleoplasm. This is due to the fact that, in the early steps, the pre-rRNA is not very well folded or compacted, which allows interac�ons with nucleolar RNA chaperones that retain it there. As matura�on progresses, the prerRNA becomes more compacted and has more r-proteins and factors associated, which make these interac�ons with nucleolar RNA chaperones more difficult and the migra�on to the nucleoplasm occurs [38, 96]. 2.2.3.4.- Nucleoplasmic matura�on and export of pre-60S ribosomal par�cles The release of the Spb1 factor allows the matura�on of the L1 stalk, which moves to its final posi�on; then, the uL2 and eL43 r-proteins stabilize it. This allows the displacement of the 5S RNP to the CP, but in a 180° rotated posi�on with respect to its mature conforma�on [111, 115]. Some factors, such as Arx1, Cgr1 and Rsa4, associate with the interface, and Nog2 also associates with the PTC region. The C-terminal domain of Nog1 enters the PET and promotes its matura�on [99, 111, 114, 115, 121]. The Sda1 factor and the L1 stalk allow the incorpora�on of the Rix1 complex, which recruits the ATPase Rea1 again for another round of ac�on. Now, Rea1 releases the Rpf2, Rrs1 and Rsa4 factors, allowing the 180° rota�on of the 5S RNP and the CP matura�on with the assembly of the eL29 and eL42 r-proteins (Figure 11) [111, 115]. At this point, with the PET and the PTC formed, ITS2 is processed at site C2 by the Las1 endonuclease [67, 68]. The Nop53 factor associates in the same region previously occupied by Erb1 and binds to Mtr4, which allows the recruitment of the nuclear 27 INTRODUCTION exosome. Then, the exosome performs the subsequent processing of the ITS2 region present on the 7S pre-rRNAs [97, 114, 122, 123]. With the removal of this 3’ fragment from the 7S pre-rRNAs, the so-called foot structure present in the pre-60S r-par�cles disappears [97, 115]. Figure 11. Key nuclear matura�on steps of yeast pre-60S ribosomal par�cles. In state E, the L1 stalk is displaced from its mature posi�on. The release of Spb1 allows its displacement, which can be observed in the Arx1/Nog2 par�cle, as well as the incorpora�on of 5S RNP. The release and binding of other RAFs and r-proteins are also indicated. Rix1-Rea1 allows the rota�on of the 5S RNP and the forma�on of the central protuberance (CP), which can be observed in the LN- par�cle. The L1 stalk is shown in red, the 5S rRNA in orange and the pre-rRNA in white. Each RAF is shown in a different color. Taken from [38]. As men�oned before, Rea1 in its second round of ac�on releases Rsa4, which causes the separa�on of Nog2. This allows the associa�on of the export adaptor Nmd3, whose binding serves as control of the correct forma�on of the P and E sites and of the L1 stalk [117, 118, 124-126]. Nmd3 permits the export to the cytoplasm through the RanGTP-binding expor�n Crm1 (Figure 12) [88, 127]. 28 INTRODUCTION Figure 12. Model of Nmd3 associa�on and pre-60S ribosomal par�cles export. Rea1 releases Rsa4 and this causes the subs�tu�on of Nog2 by Nmd3, which allows the export to the cytoplasm of the pre-60S r-par�cles by the RanGTP-binding expor�n Crm1. Adapted from [124]. Factors such as Rrp12, Ecm1, Bud20 or Npl3, interact with nucleoporins [94, 128- 131]. The Mex67/Mtr2 mRNA export factor, which is involved in the pre-40S r-par�cles export, also exports the pre-60S r-par�cles [93, 131, 132]. The associa�on of this factor is coupled to the correct forma�on of the P stalk, since apparently only when the Mrt4 factor is released, Mex67/Mtr2 can associate with the par�cles. Mrt4 is bound to the uL10 r-protein site in the P stalk, and as matura�on progresses, this r-protein assembles and replaces Mrt4, which is released [133]. The presence of different factors that allow the export of the par�cles is a reflect of the need to ensure the rapid access to new 60S r-subunits. 2.2.3.5.- Cytoplasmic matura�on of pre-60S ribosomal par�cles In the cytoplasm, the pre-rRNAs undergo their last processing steps, the last RAFs dissociate, the last r-proteins assemble, and the func�onal sites complete their matura�on and are tested. Several GTPases and ATPases are involved in the cytoplasmic matura�on of pre-60S r-par�cles. At this point, the Rlp24, Nog1, Nmd3, Arx1 and Tif6 factors are released from the par�cles [134]. The AAA-ATPase Drg1 dissociates Rlp24, allowing the assembly of the eL24 r-protein. This ATPase also releases other factors, such as Nog1, that are close to Rlp24 [134-136]. Nog1 is dissociated from the cytoplasmic pre-60S r-par�cles in two phases: in the first one, its own GTPase ac�vity releases its N-terminal and GTPase domains; and in the second one, its C-terminal domain is released from inside the PET by Drg1 [38]. Nog2 29 INTRODUCTION eL24 recruits Rei1, whose C-terminal domain, like that of Nog1, enters the PET and allows its matura�on [111, 121, 137]. Arx1 is located at the exit of the PET and prevents the entry of factors that would ini�ate a premature transla�on. Rei1, Jjj1 and the ATPase Ssa1/Ssa2 release the dimer Arx1-Alb1 and themselves [138-142]. Reh1 associates and its C-terminal domain takes the place of Rei1’s inside the PET [121, 125]. The P stalk completes its matura�on, and Nmd3, Reh1 and Tif6 are released. The eL40 and uL16 r-proteins now assemble and the GTPase Lsg1 releases Nmd3. By this way, the PTC completes its matura�on [134, 143, 144]. On the other hand, Tif6 avoids the premature interac�on of pre-60S r-par�cles with mature 40S r-subunits. Sdo1 and the GTPase Efl1 release Tif6, allowing the now mature 60S r-subunits to enter the transla�ng pool (Figure 13) [145-147]. Figure 13. Cytoplasmic matura�on events of the 60S ribosomal subunit in yeast. Drg1 dissociates Rlp24 and other factors such as Nog1. eL24 recruits Rei1, which is later replaced by Reh1. The eL40 and uL16 r-proteins assemble and Lsg1 releases Nmd3. Reh1 dissociates and Efl1 and Sdo1 release Tif6 to form the mature 60S r-subunit. The L1 stalk is shown in red, the 5S rRNA in orange and the pre-rRNA in white. Each RAF is shown in a different color. Taken from [38]. 30 INTRODUCTION 2.3.- The yeast ribosome assembly factor Pol5 Pol5 is an essen�al yeast protein whose name comes as it was iden�fied as the puta�ve fi�h essen�al DNA polymerase. It has typical sequences of a B-type polymerase, but it has been clearly demonstrated that this ac�vity is not essen�al for yeast growth [148]. Despite of this, the name has been maintained. Pol5 is a nucleolar protein. This localiza�on suggests a func�on related to rRNA metabolism. Indeed, Pol5 has been proposed to regulate the rDNA transcrip�on and rDNA repeat copy number [148, 149]. The Rrp14 protein seems to be associated with the transloca�on of Pol5 to the nucleolus in Schizosaccharomyces pombe [150]. Pol5 was iden�fied as a member of the UTP-A complex [151], but other studies determined that this associa�on is present at sub-stoichiometric levels in the SSU processome or in very early ribosomal complexes [152]. Subsequently, Pol5 was described as an accessory factor of the UTP-A complex rather than a component, which associates when the transcrip�on of the pre-rRNA begins and dissociates before the SSU processome is assembled [153]. In contrast to the rest of the UTP-A components, the major binding region of Pol5 is located within the 3’ region of 25S rRNA (Figure 14) [154, 155]. More recently, it has been proposed that the role of Pol5 in the SSU processome is related to the recycling of the t-Utps factors [155]. Figure 14. Pol5 crosslinks to the 5ʹ ETS, ITS2 and 25S rRNA sequences. Representa�on of the normalized number of reads in a cross-linking and analysis of cDNA (CRAC) experiment of an HTP-tagged Pol5 protein (red) and an untagged protein (black) as control over a 35S pre-rRNA scheme. Note that the major binding region of Pol5 is located in 25S rRNA. Adapted from [155]. 31 INTRODUCTION 2.4.- The human MYBBP1A protein MYBBP1A (Myb binding protein 1A), also called p160, has been iden�fied as the human ortholog of Pol5 [149]. It is involved in human ribosome synthesis, but its exact role in this process has not yet been iden�fied [156-158]. The MYBBP1A gene is located on human chromosome 17, in a region where loss of heterozygosity is very common in different types of cancer, which suggest the presence of a tumor suppressor, being MYBBP1A the main candidate [159]. It is a nucleolar protein, although it could translocate to the nucleoplasm under certain stress condi�ons [160]. MYBBP1A was first iden�fied by its ability to bind to the nega�ve regulatory domain (NRD) of the proto-oncogen c-Myb and by this way repressed the func�on of the later [160]. c-Myb is a transcrip�on factor that regulates the stem cell phenotype and prolifera�ve capacity of cells. c-Myb has been described to be essen�al in bone marrow (stem cells and progenitors), colon crypts and neurogenic regions. In cancer, it o�en loses its NRD and becomes hyperac�vated [161-164]. Bas1 has been described as the yeast ortholog of c-Myb [165]. MYBBP1A has also been implicated in other cellular processes such as metabolism [166, 167], cell cycle and mitosis [168, 169] or senescence [168, 170]. It is also essen�al for early embryonic development [168, 171]. Regarding ribosome biogenesis, MYBBP1A was iden�fied in a screening searching for RAFs involved in human pre-rRNA processing [156]. It has been described that MYBBP1A represses the transcrip�on by RNA polymerase I and is involved in pre-rRNA processing in humans cells, but the exact mechanism of how it par�cipates is s�ll unknown [157]. MYBBP1A also appears to repress rDNA transcrip�on in an epigene�c manner [158]. As men�oned above, although MYBBP1A is nucleolar, under certain stress condi�ons, it translocates to the nucleoplasm. Under normal condi�ons, MYBBP1A is located in the nucleolus and plays a role in regula�ng the synthesis and the processing of pre-rRNA. However, under stress condi�ons, usually nucleolar stress condi�ons that inhibit ribosome biogenesis, MYBBP1A undergoes a proteoly�c process to generate the 39 OBJECTIVES Ribosome biogenesis is a very complex process involving numerous RAFs and rproteins. Among these RAFs, some are s�ll unknown, and their study could further elucidate the process of ribosome synthesis. In this Thesis work, we have focused on studying the role of Pol5 in this process. In addi�on, we have inves�gated whether the yeast S. cerevisiae is a good model to study whether the human ortholog of Pol5, MYBBP1A, plays the same role in ribosome synthesis as Pol5. For these purposes, we have worked in the following objec�ves: - The func�onal study of Pol5 to determine its role during ribosome biogenesis, analyzing its associa�on with pre-ribosomal par�cles and the effects of its loss- of-func�on in pre-rRNA processing and r-subunits matura�on. - The study of the N- and C-terminal protein domains of Pol5 and their possible different func�ons. More specifically, we aim to study the possibility that Pol5 could be a modular protein with different func�ons assigned to each of its domains, as it has been described for other RAFs. - The analysis of the effects of the heterologous expression of MYBBP1A on ribosome biogenesis in yeast to determine whether the human protein has similar func�ons to Pol5 in ribosome biogenesis. 4.- MATERIALS AND METHODS 43 MATERIALS AND METHODS 4.1.- Strains and microbiological methods All the yeast strains used in this study are listed in Table 2. Yeast cells were grown at the indicated temperatures in rich YPAD medium (1% yeast extract, 2% peptone, 0.2% adenine, 2% glucose) or in SD medium (synthe�c dextrose: 0.15% yeast nitrogen base, 0.5% ammonium sulfate, 2% glucose) supplemented with the necessary amino acids and bases. When indicated, glucose was replaced by 2% galactose (YPAGal or SGal) or 2% raffinose (SRaf) as the carbon source. For prepara�on of plates, 2% agar was added to the media prior to steriliza�on. The used strains are derived from the BY4742 and W303 strains. To obtain the DGY115 strain, the Y20296 strain (Euroscarf), which is an heterozygous POL5/pol5::kanMX4 diploid, was transformed with the YCplac33-POL5 plasmid. This diploid was sporulated and tetrads were dissected on a Singer MSM 400 micromanipulator. Spore clones were analyzed by colony PCR and DNA sequencing. DGY115 was a representa�ve pol5∆ spore clone. The growth and handling of yeast and the prepara�on of the different yeast media were performed according to established procedures [195]. When required, yeast cells were transformed with the indicated vectors using the lithium acetate method [196]. 4.2.- Plasmids and oligonucleo�des The plasmids used in this study are listed in Table 3. The DH5α strain of Escherichia coli was used to generate and/or amplify these plasmids. The YCplac111- and YCplac33-POL5 plasmids were generated by PCR amplifica�on of the POL5 gene using yeast genomic DNA as template and the oligonucleo�des XbaIpol5up and BamHIpol5dw, which were placed ±1 kbp upstream and downstream from the start-stop codon of the POL5 ORF, respec�vely (Table 4). The PCR product was digested with the XbaI and BamHI enzymes and cloned into the empty YCplac111 and YCplac33 plasmids, that were also digested with the same enzymes. 44 MATERIALS AND METHODS The same procedure was carried out to generate the YEplac181-RPL25, YEplac181-RPL27A, YEplac195-RPL27A and pAS24-HA-MYBBP1A plasmids but using the corresponding oligonucleo�des pairs listed in Table 4 and the appropriate enzymes. In the case of pAS24-HA-MYBBP1A, we used the pACT-Flag-MYBBP1A (a gi� from the laboratory of Dr. A. Carnero) as the template for the PCR. To obtain the YCplac111-pol5dn and the pAS24-HA-MYBBP1A-p140 plasmids, the Q5 site-directed mutagenesis kit (NEB) was used with the plasmids YCplac111-POL5 and pAS24-HA-MYBBP1A as templates, respec�vely, and the oligonucleo�des indicated in Table 4. Details of cloning for the rest of the plasmids have been reported in the respec�ve studies where they were originally described, which are also indicated in Table 3. 4.3.- Mul�copy suppression analysis We used a YEplac181-based yeast genomic library to perform a screening for mul�copy suppressors of pol5 mutants [197]. We transformed the Y10143 strain (pol5- 1 mutant) with this library, and transformants were spread on SD-Leu plates and incubated at 37 °C (non-permissive temperature) for 3-5 days. Plasmids from the thermotolerant candidates were isolated, rescued in E. coli, and then used to backtransform the pol5-1 mutant. Out of about 35,000 transformants, only four clones were able to suppress the thermosensi�ve phenotype of the pol5-1 mutant. The inserts present in the rescued plasmids were sequenced. One candidate contained DNA from the RPL25-TRM13 region of chromosome XV, another one contained the POL5 gene, and the last two ones contained fragments from chromosome VIII, where only the DIA4 and RPL27A genes were shared. Further cloning of PCR-amplified fragments of the RPL27A and RPL25 genes inserted into the YEplac181 or YEplac195 plasmids, defined these genes as the truly mul�copy suppressors of the pol5-1 mutant. 45 MATERIALS AND METHODS 4.4.- Polysome profiles, ribosomal subunit analysis and sucrose gradient frac�ona�on Cell extracts for polysome and r-subunit analysis were prepared as previously described [198]. For polysome profile analysis, yeast cultures of 200 ml were grown un�l mid-log phase and then harvested at an OD600 of 0.8 approximately or maintained under the specific condi�ons of each experiment for the �me indicated, before harves�ng. Transla�on was blocked by the addi�on of cycloheximide (100 µg/ml) and then the cells were collected by centrifuga�on. Cells were resuspended in lysis buffer (10 mM Tris-HCl, pH 7.5, 100 mM NaCl, 30 mM MgCl2, 100 µg/ml cycloheximide, 200 µg/ml heparin and 0.2 µl/ml DEPC (diethyl pyrocarbonate)) and then an equal volume of glass beads was added. Cells were lysed by vortexing at maximum speed. The extracts were clarified by centrifuga�on and glycerol was added to a 10% of final concentra�on. RNA in the samples was quan�fied by measuring the A260. Samples were stored at -80 °C un�l used. For r-subunit analysis, the same protocol was followed but 0.006% NaN3 was used instead of cycloheximide to produce a polysome run-off in a low Mg2+ lysis buffer (50 mM Tris-HCl, pH 7.5, 50 mM NaCl and 1 mM DTT) to dissociate vacant ribosomes into r-subunits. Ten A260 units of each extract were loaded onto 7-50% sucrose gradients and centrifuged in a Beckman Coulter rotor SW41Ti at 39,000 rpm for 2 h 45 min (polysomes) or 4 h 30 min (r-subunits). An ISCO UA-6 system was then used to con�nuously record the A254 to obtain the corresponding profiles. Quan�fica�on of the 60S/40S ra�o of r-subunits was performed using the OriginPro so�ware to measure the area under each peak; then, the mean of the ra�os obtained for different biological replicates was calculated and sta�s�cal analysis was performed using the GraphPad Prism so�ware. When necessary, 0.5 ml frac�ons were collected; then, proteins were precipitated from each frac�on. To do so, the trichloroace�c acid and acetone protocol was used as previously described [199]. Precipitates from each frac�on were resuspended in 2X Laemmli loading buffer (160 mM Tris-HCl pH 6.8, 2% SDS, 20% glycerol, 10% β-mercaptoethanol, 0.004% bromophenol blue). Equal volumes of each 46 MATERIALS AND METHODS precipitated sample were loaded onto SDS-polyacrylamide gels and analyzed by western blo�ng with the appropriate an�bodies. 4.5.- RNA extrac�on and analysis of steady-state RNA levels Total RNA was extracted from samples corresponding to 10 OD600 units of exponen�ally grown cells by the acid-phenol method as previously described [200]. Yeast cultures were grown un�l mid-log phase and then harvested or maintained under the specific condi�ons of each experiment for the �me indicated before harves�ng. Cells were resuspended in TES buffer (10 mM Tris-HCl, pH 7.5, 10 mM EDTA, pH 8.0, 0.5% SDS), and an equal volume of acid phenol was added. A�er mixing, the samples were incubated at 65 °C for 60 min, with constant shaking. The samples were then centrifuged and washed once with a volume of acid phenol and once with a volume of chloroform. RNA was precipitated overnight at -20 °C by the addi�on of 2 volumes of 96% (v/v) ethanol and 1/10 volume of 3 M sodium acetate. The next day, the samples were washed with 2 volumes of 70% (v/v) ethanol and dried. Finally, the RNA pellets were resuspended in dis�lled water and quan�fied using a NanoDropTM spectrophotometer (Thermo Fisher Scien�fic). Samples were stored at -80 °C un�l used. Northern blot and primer extension analysis were performed as described in [201]. For northern blot analysis, equal amounts of total RNA (normally 5 µg) were loaded and resolved on a 1.2% agarose-6% formaldehyde gel or a 7% polyacrylamide-8 M urea gel to analyze high or low molecular-weight RNAs, respec�vely. RNAs were then transferred to a nylon membrane (Hybond™-N, GE Healthcare) and crosslinked using a Spectrolinker XL-1000 (Spectronics Corpora�on). Specific an�sense oligonucleo�des, indicated in Table 4, were used to detect the different pre- and mature rRNAs. Before use, they were labeled with [γ-32P] ATP (6,000 Ci/mmol; Perkin Elmer) at their 5’ ends. For primer extension analysis, equal amounts of total RNA (normally 5 µg) were reverse-transcribed using the probe g as primer (Table 4), previously labeled with [γ-32P] ATP at its 5’ end. The DNA obtained was separated by electrophoresis in a 7% polyacrylamide-8 M urea sequencing gel. Once resolved, the gel was dried on a Whatman filter paper under vacuum. 47 MATERIALS AND METHODS Phosphorimager analysis was performed by exposing the Whatman papers and the nylon membranes to Phosphorimager screens (Fuji, BAS-IP MS; GE Healthcare), which were revealed using a Typhoon™ FLA-9000 imaging system (GE Healthcare). When required, signal intensity was quan�fied using the GelQuant.NET so�ware. 4.6.- Protein extrac�on, western blot analysis and an�bodies Total yeast protein extracts were prepared and analyzed by western blo�ng as described in [202]. Proteins were first separated by SDS-PAGE (denaturing electrophoresis) and then transferred to a nitrocellulose membrane. The primary an�bodies used are listed in Table 5. Secondary goat an�-mouse or an�-rabbit horseradish peroxidase-conjugated an�bodies (Bio-Rad) were used. Proteins were detected using an enhanced chemiluminescence detec�on kit (Super-Signal West Pico, Pierce) and a ChemiDocTM imaging system (Bio-Rad). 4.7.- Analysis of fluorescent proteins by microscopy To study the subcellular localiza�on of Pol5, the YMP1 strain (POL5-GFP strain) was transformed with the pRS314-DsRed-NOP1 plasmid, which expresses the Nop1 nucleolar marker. Transformants were grown in SD-Trp medium at 30 °C un�l mid-log phase, then cells were washed in PBS buffer (8 mM Na2HPO4, 1.5 mM KH2PO4, 140 mM NaCl, 2.75 mM KCl, pH 7.3) and examined using an Olimpus BX61 fluorescence microscope equipped with a digital camera. The images were acquired using the Olympus cellSens so�ware and processed using the Adobe Photoshop so�ware (Adobe Systems, Inc.). The same strain was also used to test whether the Pol5 localiza�on changes with the expression of the MYBBP1A protein. In this case, the strain was also transformed with a pAS24 empty plasmid or a pAS24 plasmid containing the MYBBP1A cDNA under the control of a GAL promoter. Transformants were grown in SRaf-Leu-Trp medium at 30 55 RESULTS Part of the results presented in this Thesis work have been published as: Ramos- Sáenz A., González-Álvarez D., Rodríguez-Galán O., Rodríguez-Gil A., Gaspar S.G., Villalobo E., Dosil M. & de la Cruz J. (2019). Pol5 is an essen�al ribosome biogenesis factor required for 60S ribosomal subunit matura�on in Saccharomyces cerevisiae. RNA. 25: 1561-1575. Some of the figures and results from this publica�on have been adapted and completed with new informa�on. 5.1.- Func�onal study of the yeast Pol5 protein 5.1.1.- Pol5 is an essen�al protein Since Pol5 has been described as an essen�al protein [148], we used two different condi�onal systems to study its possible role in ribosome biogenesis. First, we used the YJP38 strain (herea�er named GAL::HA-POL5 strain), in which the promoter region of the POL5 genomic locus was replaced by the inser�on of a kanMX6-GAL1/10::HA cassete. In this strain, an N-terminally HA-tagged Pol5 fusion protein is expressed in media containing galactose as carbon source, but is repressed in media containing glucose. The growth of the GAL::HA-POL5 strain was tested along with an isogenic wild-type strain by serial dilu�ons onto YPAGal or YPAD plates. We observed no differences in growth between the two strains in the case of the YPAGal plates, meaning that the constructed allele is fully func�onal. However, we observed no growth of the GAL::HA-POL5 strain on YPAD plates, which confirmed that the protein is essen�al (Figure 18A). When grown in liquid YPAGal medium, the duplica�on �me of the GAL::HAPOL5 strain was similar to that of an isogenic wild-type strain in the same medium (approx. 1.8 h). When the GAL::HA-POL5 strain was transferred from YPAGal to YPAD medium, the duplica�on �me was similar to that of the wild-type strain for the first 6 h, but gradually increased to a value of >10 h a�er 12 h in YPAD medium. We also analyzed the levels of the HA-Pol5 protein by western blo�ng. The result showed that the protein was only barely present a�er a shi� of 3 h of the GAL::HA-POL5 strain from YPAGal to YPAD medium (Figure 18B). As a second condi�onal system for phenotypic analysis, we used the previously described pol5-1 (Y10143 strain, herea�er named pol5-1 strain) and pol5-3 (Y10144 56 RESULTS strain, herea�er named pol5-3 strain) mutants [148, 207]. These strains carry point muta�ons that lead to specific amino acid subs�tu�ons in the Pol5 sequence (V404M, K491E and Q671P for Pol5-1; W292R for Pol5-3). We tested the growth of these strains along with an isogenic wild-type strain by serial dilu�ons onto YPAD plates at different temperatures. In the case of the pol5-1 mutant, we observed no differences in growth with the wild-type strain at 22, 25 and 30 °C, but we could observe a marked slow growth at 37 °C (Figure 19). On the other hand, the pol5-3 mutant showed no differences in growth with the wild-type strain at 22 and 25 °C, but a slight slow growth at 30 °C and a severe growth defect at 37 °C (Figure 19). Figure 18. The deple�on of Pol5 inhibits cell growth. (A) Growth comparison of the wild-type (W303-1B) and GAL::HA-POL5 strains. Cells were grown in liquid YPAGal medium, diluted to an OD600 of 0.05 and then spoted in serial dilu�on steps onto YPAGal or YPAD plates, which were incubated at 30 °C for 4 days. (B) Deple�on of the Pol5 protein. Total extracts were prepared from cells of each strain harvested at the indicated �mes a�er a shi� from YPAGal to YPAD medium and then analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. HA-Pol5 protein was detected using an an�-HA an�body. Pgk1 was used as loading control, which was detected using an an�-Pgk1 an�body. As it will be demonstrated later (see Figure 31), both pol5-1 and pol5-3 muta�ons are recessive, since when the mutants are transformed with a centromeric plasmid harboring a wild-type POL5 allele, they show a wild-type growth at any temperature. Altogether, these results confirm that Pol5 is an essen�al protein. Glc Gal POL5 GAL::HA-POL5 A B HA-Pol5 Pgk1 0 24 h Glc 24 GAL::HA-POL5POL5 03 9 6 12 57 RESULTS Figure 19. The pol5-1 and pol5-3 muta�ons confer a thermosensi�ve phenotype. Growth analysis of the wild-type (BY4742) and the pol5-1 and pol5-3 strains. Cells were grown in liquid YPAD medium at 25 °C, diluted to an OD600 of 0.05 and then spoted in serial dilu�on steps onto YPAD plates, which were incubated at 22, 25, 30 and 37 °C for 4 days. 5.1.2.- Pol5 is nucleolar To study the localiza�on of Pol5, we used a strain expressing a C-terminally GFP- tagged Pol5 from its genomic locus as the only source of Pol5 (YMP1 strain, herea�er named POL5-GFP strain). This construct was obtained by the inser�on of a PCR-derived cassete into the genomic locus of POL5. We tested the growth of the POL5-GFP strain along with an isogenic wild-type strain by serial dilu�ons onto YPAD plates at different temperatures. We could not observe any growth impairment at any temperature, which indicates that the fusion protein is fully func�onal (Figure 20A). We also could detect the fusion protein by western blot analysis using an an�-GFP an�body (Figure 20B). Finally, polysome profile analysis revealed no defects in ribosome biogenesis in the POL5-GFP strain, including normal levels of each r-subunit (Figure 20C). We then analyzed the subcellular localiza�on of the Pol5-GFP protein by fluorescence microscopy. As a result, we observed that the Pol5-GFP protein was mainly located in the nucleolus, which is the main site of ribosome biogenesis. It co-localized with the DsRed-Nop1 reporter used as a nucleolus marker (Figure 21). This result agrees with previous ones that indicated that a 9xMyc-tagged Pol5 construct was located in the nucleolus [148]. The predominant localiza�on of Pol5-GFP in the nucleolus suggests a role for Pol5 in ribosome biogenesis. 22 °C 25 °C 30 °C 37 °C POL5 pol5-1 pol5-3 58 RESULTS Figure 20. Func�onal analysis of the POL5-GFP strain. (A) Growth analysis of the wild-type (W303-1A) and POL5-GFP strains. Cells were grown in liquid YPAD medium at 30 °C, diluted to an OD600 of 0.05 and then spoted in serial dilu�on steps onto YPAD plates, which were incubated at 22, 25, 30 and 37 °C for 4 days. (B) Expression of Pol5-GFP. The wild-type and POL5- GFP strains were grown in liquid YPAD medium at 30 °C un�l mid-log phase. Total extracts were prepared and analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. Pol5-GFP protein was detected using an an�-GFP an�body. Pgk1 was used as loading control, which was detected using an an�-Pgk1 an�body. (C) Polysome profile of POL5-G F P. T h e POL5- GFP strain was grown in YPAD medium at 30 °C un�l mid-log phase. Then, a whole cell extract was prepared and 10 A260 units were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. The peaks of free 40S and 60S rsubunits, 80S free couples/monosomes and polysomes are indicated. Figure 21. Pol5 is predominantly nucleolar. POL5-GFP cells were transformed with a plasmid expressing the nucleolar marker protein DsRed-Nop1 from its cognate promoter. Pol5-GFP and DsRed-Nop1 signals were revealed by fluorescence microscopy. Cells were also examined by light microscopy. Arrows point to nucleolar localiza�on. Note that Pol5-GFP and DsRed-Nop1 signals co-localized. 22 °C 25 °C 30 °C 37 °C POL5 POL5-GFP POL5-GFP 60S 40S 80S polysomes Pol5-GFP Pgk1 A BC A 254 Sedimentation Pol5-GFP DsRed-Nop1 merge visible 59 RESULTS 5.1.3.- Pol5 is associated to high-molecular-mass par�cles To study whether Pol5 associates with r-par�cles, we analyzed the sedimenta�on behavior of the HA-tagged version of Pol5 in sucrose gradients. For this purpose, we used the GAL::HA-POL5 strain that was grown in YPAGal medium un�l mid-log phase. Whole cell extracts were then prepared and subjected to low Mg2+ run-off sucrose gradient ultracentrifuga�on. Frac�ons were collected and analyzed by western blot to detect the HA-Pol5 protein. As shown in Figure 22, a large propor�on of the signal from the HA-Pol5 protein was found to be associated with frac�ons of high-molecular-mass par�cles. These frac�ons correspond to r-par�cles, as indicated by the sedimenta�on of eS8 and eL14 rproteins. Some HA-Pol5 protein was detected at the posi�ons expected for the free proteins, probably due to the overexpression of POL5 in this strain, since no signal of the Pol5-GFP protein could be detected at these posi�ons when the POL5-GFP strain, which expressed the protein from the genomic locus, was used to perform a similar experiment (see later Figure 58). Figure 22. Pol5 is present in high-molecular-mass frac�ons. GAL::HA-POL5 strain was grown in YPAGal medium at 30 °C un�l mid-log phase, then cells were harvested and whole cell extracts were prepared in a low Mg2+ buffer lacking cycloheximide to dissociate the r-subunits. 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. Frac�ons were collected from the gradients, the proteins were extracted from each one and equal volumes were analyzed by western blo�ng. The posi�ons on the profile correspond to those of the frac�ons on the blots. The HA-Pol5 protein was detected with an an�-HA an�body. Specific an�bodies were used to detect the eS8 and eL14 r-proteins. Peaks of total 40S and 60S r-subunits are indicated. HA-Pol5 eS8 eL14 1 3 5 6 7 8 9 11T 2 4 10 12 GAL::HA-POL5 Gal 60S 40S 60 RESULTS We conclude that the majority of Pol5 is stably associated with large par�cles, which likely correspond to r-par�cles. 5.1.4.- Pol5 associates with pre-60S ribosomal par�cles To study the associa�on of Pol5 with pre-ribosomal par�cles, we affinity purified Pol5-GFP-containing complexes using the GFP-Trap procedure (see Materials and Methods) and determined which pre-rRNAs intermediates were co-purified with Pol5- GFP by northern blot hybridiza�on. For this purpose, we used cell extracts of the POL5-GFP strain, which was grown in YPAD medium un�l mid-log phase. As a nega�ve control, we used an isogenic POL5 untagged strain grown in the same condi�ons. Figure 23. Pol5 associates with pre-60S ribosomal par�cles. The associa�on of pre- and mature rRNAs with Pol5 was studied using the GFP-Trap immunoprecipita�on procedure with cell extracts of the POL5-GFP strain and a wild-type strain (W303-1A) as untagged control. Cells were grown in YPAD at 30 °C un�l mid-log phase and then harvested. Total RNA was extracted from the whole cell extracts (T) and from the immunoprecipita�on (IP), and pre-rRNAs and mature rRNAs were analyzed by northern blo�ng. (A) Northern blot analysis of high-molecular- mass pre- and mature rRNAs. (B) Northern blot analysis of low-molecular-mass pre- and mature rRNAs. The probes used are indicated between parentheses and described in the Table 4 from the Materials and Methods sec�on. Signal intensi�es were measured by phosphorimager scanning and the values indicate the percentage of each RNA species detected in the purifica�on rela�ve to the total extract input. (c) (d) (g) (b) (c) T TIP IP Pol5 Pol5-GFP 35S 32S 27SA2 27SA 27SB 25S 18S 20S (h) 0.4 0.3 (35S) <0.1 3.0 0.1 2.5 0.2 0.3 <0.1 0.1 0.30.2 A 7S 5.8S L 5.8S S 5S T TIP IP Pol5 Pol5-GFP (f) (e) (5S) B 0.1 0.1 0.4 0.3 0.3 0.5 (32S) 0.6 0.8 61 RESULTS As shown in Figure 23, in the Pol5-GFP purifica�on, the 27SA2 and 27SB prerRNAs were clearly enriched compared to the control purifica�on of the untagged Pol5. Litle, although not above background levels, of the 35S and 32S pre-rRNAs also appear to co-purify with Pol5-GFP. No enrichment could be detected for the 7S and 20S prerRNAs, nor for the mature rRNAs, in the case of the Pol5-GFP purifica�on when compared to the control strain purifica�on (Figure 23). We also examined the presence of selected RAFs and r-proteins within the complexes purified with the GFP-tagged Pol5. In this case, we used the POL5-GFP strain transformed with an empty pAS24 plasmid and, as a control, an isogenic POL5 untagged strain transformed with the same plasmid (see later Results 5.3.9). The strains were grown in SRaf-Leu medium un�l mid-log phase and then shi�ed to SGal-Leu medium for 6 h. Our results showed that the RAFs Nop7 and Has1 and the eL14 r-protein were also co-purified with Pol5-GFP (Figure 24). Figure 24. Pol5 associates with ribosome assembly factors and ribosomal proteins from pre- 60S ribosomal par�cles. Associa�on of different proteins related to the matura�on of the 60S r-subunit with Pol5 was studied using the GFP-Trap immunoprecipita�on procedure with cell extracts from the POL5-GFP strain transformed with an empty pAS24 plasmid and a wild-type strain (W303-1A) with untagged Pol5 transformed with the same plasmid as a control. Cells were grown in SRaf-Leu at 30 °C un�l mid-log phase, then shi�ed to SGal-Leu medium for 6 h and harvested. Proteins were extracted from the whole cell extracts (T) and from the immunoprecipita�on (IP), and then analyzed by western blo�ng. We loaded 30% volume of each extract. The Pol5-GFP protein was detected using an an�-GFP an�body. Specific an�bodies were used to detect the RAFs Nop7 and Has1 and the eL14 r-protein. Note that the Pol5-GFP signal is lost in the total extracts but present in the purifica�on, probably due to amount of extract loaded. Pol5-GFP Nop7 Has1 IPT eL14 62 RESULTS These results indicate that Pol5 is a component of early/medium nucleolar pre-60S r-par�cles. Pol5 appears to associate with these par�cles at an early stage during 27SA2 pre-rRNA forma�on and dissociates concomitantly to 27SB pre-rRNA cleavage at site C2 in the nucle(ol)us, but before the export of the corresponding par�cles to the cytoplasm. 5.1.5.- Pol5 is required for 60S ribosomal subunit accumula�on To inves�gate the possible role of Pol5 in ribosome biogenesis, we analyzed the polysome profile of the GAL::HA-POL5 strain grown in YPAGal medium or a�er a shi� of 3 h in YPAD medium at 30 °C. In YPAGal, the strain showed a normal profile, iden�cal to those obtained for an isogenic wild-type strain. However, when shi�ed to YPAD for 3 h, we observed an aberrant polysome profile in which we could detect a decrease in the levels of the free 60S r-subunits peak rela�ve to the levels of the free 40S r-subunits peak, a decrease in the 80S ribosome peak and the presence of half-mer polysomes (Figure 25). Half-mer polysomes correspond to mRNAs associated with a precise number of ribosomes and an addi�onal single 48S pre-ini�a�on complex stuck at the ini�a�on codon. Half-mer polysomes appear when there is a transla�on ini�a�on defect. This defect could be due to a 60S r-subunit accumula�on defect (ribosome biogenesis failure) or to a subunit joining defect (transla�on failure) [214]. This effect was exacerbated when the shi� to YPAD was of 8 h (Figure 25). In our case, we have a deficit in 60S rsubunit accumula�on. Considering the western blot results showing that we could not detect any HA- Pol5 protein a�er 6 h of deple�on (Figure 18B) and the presence of an aberrant polysome profile only a�er 3 h of deple�on (Figure 25), we decided to use a deple�on �me of 6 h for the majority of our further experiments. We also analyzed the polysome profiles of the pol5-1 and pol5-3 mutants and of an isogenic wild-type strain. Strains were grown in YPAD medium at 25 °C and subjected to a shi� to 37 °C for 6 h. As shown in Figure 26, we observed a normal polysome profile for the pol5-1 mutant at 25 °C, which was indis�nguishable from that obtained for the wild-type strain. However, a�er the temperature shi�, the pol5-1 mutant led to an 63 RESULTS aberrant profile, similar to that obtained upon Pol5 deple�on (Figure 25). In the case of the pol5-3 mutant, an aberrant polysome profile was obtained even at the permissive temperature of 25 °C. However, this phenotype was exacerbated when shi�ed to the restric�ve temperature of 37 °C (Figure 26). In all cases, the aberrant polysome profiles obtained are compa�ble with a deficit in the accumula�on of the 60S r-subunit. Figure 25. Pol5 deple�on causes a 60S ribosomal subunit shortage. The GAL::HA-POL5 strain was grown in YPAGal medium at 30 °C un�l mid-log phase and then shi�ed to YPAD medium for 3 or 8 h. Cells were harvested, whole cell extracts were prepared and 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. The peaks of free 40S and 60S r-subunits, 80S free couples/monosomes and polysomes are indicated. Half-mers are labeled with arrows. To confirm that half-mer polysomes appeared due to a deficit in 60S r-subunit accumula�on upon deple�on or muta�on of Pol5, we performed a quan�fica�on of total r-subunits using low Mg2+ run-off sucrose gradients. Magnesium is required for the stable interac�on between the 60S and 40S r-subunits in the course of transla�on, thus, by using this type of gradients, a complete dissocia�on of mono- and polysomes into 40S and 60S r-subunits can be achieved. Quan�fica�on of total r-subunits using this method showed about a 20% reduc�on of total 60S rela�ve to 40S r-subunits in the GAL::HA-POL5 strain following Pol5 deple�on by a shi� for 6 h to YPAD medium compared to permissive condi�ons of GAL::HA-POL5 Gal 60S 40S 80S polysomes GAL::HA-POL5 Glc 3 h 60S 40S 80S polysomes GAL::HA-POL5 Glc 8 h 60S 40S 80S polysomes A254 Sedimentation 70 RESULTS Taken together, these results indicate that deple�on of Pol5 impaired the intranuclear and nucleocytoplasmic movement of pre-60S r-par�cles. 5.1.8.- The overexpression of the uL23 and eL27A ribosomal proteins suppresses the thermosensi�ve phenotype of the pol5-1 and pol5-3 mutants For a beter understanding of the func�on of Pol5, we performed a screen for mul�copy suppressors of the thermosensi�ve phenotype of the pol5-1 mutant. In order to do this, we transformed this mutant with a YEplac181-based mul�copy library [197], and then incubated the transformants at 37 °C (non-permissive temperature) for 5 days. Only a few thermo-tolerant candidates were obtained. Their plasmids were recovered and back-transformed into the pol5-1 mutant. Four plasmids were able to suppress the original growth defect of this mutant. We found that one of them contained the wildtype POL5 gene, one of the genes expected from the screening. The others contained the genes for the eL27A and uL23 r-proteins. We introduced the PCR-amplified RPL27A and RPL25 genes encoding eL27A and uL23, respec�vely, into the YEplac181 and YEplac195 plasmids, and then transformed the pol5-1 and pol5-3 mutants with these plasmids alone or in combina�on to study whether the thermosensi�ve phenotype of these strains was suppressed when we increased the dosage of these r-proteins. In the case of pol5-1, we observed that overexpression of RPL25 or RPL27A alone improved the growth at 37 °C, but not to wild-type levels, whereas simultaneous overexpression of both r-proteins led to a beter suppression of the thermosensi�ve phenotype at 37 °C, but s�ll not to the wild-type extent. In the pol5-3 mutant, we observed that overexpression of RPL25 had no apparent effect on the growth phenotype, but overexpression of RPL27A alone or in combina�on with RPL25 suppressed the thermosensi�ve phenotype at all temperatures, prac�cally to the wildtype extent (Figure 31). 71 RESULTS Figure 31. Suppression of the growth defects of pol5-1 and pol5-3 by high-dosage of uL23 and eL27A. The pol5-1 and pol5-3 strains were transformed with this combina�on of plasmids: YCplac111-POL5 and YCplac33 (+POL5), mul�copy plasmids YEplac181 and YEplac195 (+vectors), YEplac181-RPL25 and YEplac195 (+RPL25), YEplac181 and YEplac195-RPL27A (+RPL27A), and YEplac181-RPL25 and YEplac195-RPL27A (+RPL25/+RPL27A). Cells were grown in liquid SD-Leu-Ura medium and diluted to an OD600 of 0.05, and then they were spoted in serial dilu�on steps onto SD-Leu-Ura plates which were incubated at 25, 30 and 37 °C for 3 days. Note that the co-overexpression of uL23 and eL27A suppresses the thermosensi�ve phenotype of pol5-1 beter than single overexpression of uL23 or eL27A. These results suggest that the role of Pol5 in ribosome biogenesis could be related to these two r-proteins of the 60S r-subunit. We then decided to analyze whether the overexpression of these proteins improved the 60S r-subunit synthesis in the pol5-1 mutant by performing polysome profile analysis. Thus, the pol5-1 strain transformed with the plasmids harboring RPL25, RPL27A or both genes was grown at 30 °C in SD-Leu- Ura medium and then shi�ed to 37 °C for 9 h. We observed that a�er the shi� to 37 °C, neither the increase of RPL25 nor that of RPL27A improved 60S r-subunit synthesis, and that only the overexpression of both r-proteins at the same �me slightly ameliorated the polysome forma�on (Figure 32). We also quan�fied the total r-subunit in the pol5-1 mutant using low Mg2+ run-off sucrose gradients. As a result, we observed that only the simultaneously overexpression of RPL27A and RPL25 slightly improved the 60S/40S ra�o (Figure 33). +POL5 +vectors 25 °C +RPL25 +RPL27A +RPL25/+RPL27A 30 °C 37 °C pol5-1 +POL5 +vectors 25 °C +RPL25 +RPL27A +RPL25/+RPL27A 30 °C 37 °C pol5-3 72 RESULTS Figure 32. Polysome profiles of the pol5-1 mutant transformed with mul�copy suppressor plasmids. The pol5-1 strain was transformed with this combina�on of plasmids: mul�copy plasmids YEplac181 and YEplac195 (+vectors), YCplac111-POL5 and YCplac33 (+POL5), YEplac181-RPL25 and YEplac195 (+RPL25), YEplac181 and YEplac195-RPL27A (+RPL27A) and YEplac181-RPL25 and YEplac195-RPL27A (+RPL25 + RPL27A). Cells were grown in liquid SD-Leu- Ura medium at 30 °C un�l mid-log phase and then shi� to 37 °C for 9 h. Cells were harvested, whole cell extracts were prepared and 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. The peaks of free 40S and 60S r-subunits, 80S free couples/monosomes and polysomes are indicated. Half-mers are labeled with arrows. All these data suggest that higher levels of uL23 and eL27A may qualita�vely improve the ribosome produc�on or func�on in the pol5-1 mutant, as we did not observe a pronounced increase in the total 60S r-subunit levels in the suppressed strains compared to the original pol5-1 mutant. 60S 40S 80S polysomes 60S 40S 80S polysomes 60S 40S 80S polysomes 60S 40S 80S polysomes 60S 40S 80S polysomes pol5-1 +POL5 pol5-1 +vectors pol5-1 +RPL25 pol5-1 +RPL27A pol5-1 +RPL25 + RPL27A A254 Sedimentation 73 RESULTS Figure 33. Ribosomal subunit quan�fica�on in the pol5-1 mutant transformed with mul�copy suppressor plasmids. The pol5-1 strain was transformed with this combina�on of plasmids: mul�copy plasmids YEplac181 and YEplac195 (+vectors), YCplac111-POL5 and YCplac33 (+POL5), YEplac181-RPL25 and YEplac195 (+RPL25), YEplac181 and YEplac195-RPL27A (+RPL27A) and YEplac181-RPL25 and YEplac195-RPL27A (+RPL25 + RPL27A). Cells were grown in liquid SD-Leu-Ura medium at 30 °C un�l mid-log phase and then shi� to 37 °C for 9 h. Cells were harvested and whole cell extracts were prepared in a low Mg2+ buffer lacking cycloheximide to dissociate the r-subunits. 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. Peaks of total 40S and 60S r-subunits are indicated. The area under each peak was measured and the 60S/40S ra�os were calculated. Values are the mean of different biological replicates. The standard devia�on does not exceed 7% of the mean values in all cases. Note than only the co-overexpression of uL23 and eL27A improves the 60S/40S ra�o, but not to the wild-type extent. 5.1.9.- The apparent DNA polymerase ac�vity of Pol5 does not play a role in ribosome biogenesis It has been described that Pol5 has DNA polymerase ac�vity in vitro [148]. To study the role of this ac�vity in ribosome biogenesis, we generated the mutant pol5dn, which encodes a Pol5 protein variant that harbors muta�ons that cause the protein to lack DNA polymerase ac�vity in vitro (D623N, D625N) [148]. 60S 40S 60S 40S 60S/40S=1.3 60S/40S=1.8 60S 40S 60S/40S=1.3 60S 40S 60S/40S=1.3 60S 40S 60S/40S=1.5 pol5-1 +vectors pol5-1 +POL5 pol5-1 +RPL25 pol5-1 +RPL27A pol5-1 +RPL25 + RPL27A A254 Sedimentation 74 RESULTS We tested the growth of this mutant compared to that of an isogenic wild-type strain and we observed no differences between the wild-type and the mutant version of Pol5 (Figure 34A). We also analyzed polysome profiles from extracts of pol5dn mutant cells grown at 30 °C. Again, we were not able to observe any defect in this mutant compared to the isogenic wild-type strain (Figure 34B). Figure 34. The apparent DNA polymerase ac�vity of Pol5 is not required for cell growth and ribosome biogenesis. The pol5 null strain was transformed with a plasmid expressing either a Pol5 protein without DNA polymerase ac�vity in vitro (YCplac111-pol5dn) or a wild-type Pol5 protein (YCplac111-POL5). (A) Growth analysis of the wild type and pol5dn strains. Cells were grown in liquid YPAD medium at 25 °C and diluted to an OD600 of 0.05 and then they were spoted in serial dilu�on steps onto YPAD plates which were incubated at 25, 30 and 37 °C for 4 days. (B) Polysome profile of the wild-type and pol5dn strains. Cells were grown in liquid YPAD medium at 30 °C un�l mid-log phase. Cells were harvested, whole cell extracts were prepared and 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. The peaks of free 40S and 60S rsubunits, 80S free couples/monosomes and polysomes are indicated. Half-mers are labeled with arrows. With these results, we conclude that the DNA polymerase ac�vity of Pol5 is apparently fully dispensable for its func�on in ribosome biogenesis. pol5dn POL5 80S 80S 40S 40S 60S 60S polysomes polysomes A B 30 °C 25 °C 37 °C POL5 pol5dn A 254 Sedimentation 75 RESULTS 5.2.- Study of the protein domains of Pol5 The laboratory of Dr. Pérez-Fernández has studied in parallel to this work the func�on of Pol5 in ribosome biogenesis. This group has described that Pol5 has a dual role in (i) 60S r-subunit assembly, reaching conclusions similar to those we report in the Results sec�on 5.1., and (ii) the recycling of pre-40S RAFs, thus, contribu�ng to regulate the balance between 40S and 60S r-subunit produc�on [155]. To further inves�gate this later statement, in collabora�on with this group, we have performed different trunca�on mutants and analyzed whether we could func�onally separate the roles of Pol5 in 40S and 60S r-subunit synthesis, since this group suspects that each func�on could be carried out by different domains of the Pol5 protein. Since there is no crystal or cryo-EM structure for Pol5, we first developed a model with the predicted ter�ary structure of Pol5 in collabora�on with Dr. Antonio J. Díaz- Quintana using the Roseta so�ware [218]. The resul�ng model showed three different globular domains for the Pol5 protein: the first one composed of eleven α-helices separated by 7-12 amino acids; the second one was divided into two all-α domains formed by thirteen and five α-helices each one; and the last domain was composed of eight α-helices (Figure 35A). Recently, AlphaFold, an ar�ficial intelligence driven system, has computa�onally predicted the structure of Pol5 with the AF-P39985 code [219, 220]. Visualiza�on of the Pol5 structure predicted by AlphaFold showed that the protein is formed by a succession of α-horseshoe helices with different loops which correspond to disordered domains (Figure 35B). Using the informa�on provided by these models, we arbitrarily divided Pol5 into two domains (N-terminal and C-terminal), which are separated by a disordered region that we included in the C-terminal domain. We performed trunca�ons at each end to generate different mutants. The Pol5∆N37, Pol5∆N658 and Pol5∆N677 mutants cons�tuted the truncated versions of Pol5 lacking different numbers of amino acids in the N-terminal domain of the Pol5 protein. The Pol5∆N37 mutant started at the amino acid P38, lacking the first 37 amino acids of the full-length Pol5, the Pol5∆N658 mutant started at K659, therefore lacking the first 658 amino acids, and the Pol5∆N677 mutant started at G678, thus lacking the first 677 amino acids. 76 RESULTS Figure 35. Predicted structure of Pol5. (A) Predicted ter�ary structure of Pol5 using Roseta so�ware in collabora�on with Dr. Antonio J. Díaz-Quintana. (B) Structure of Pol5 predicted by the AlphaFold system. The N-terminal domain is shown in purple, and the C-terminal domain is shown in red. Representa�ons elaborated using the UCSF Chimera So�ware. The first and last amino acid are indicated in each case. On the other end, we generated the C-terminal trunca�ons Pol5∆C994-1022, Pol5∆C953-1022 and Pol5∆C680-1022, which lacked different numbers of amino acids in the C-terminal domain. These constructs harbored dele�ons of the last 29, 70 and 343 amino acids, respec�vely. Figure 36A outlines the primary sequence of Pol5 and highlights the different N- and C-terminal trunca�ons, and Figure 36B shows a schema�c representa�on of the fulllength Pol5 protein and the different trunca�ons. Figure 36C shows the predicted 3D structure of the different trunca�on variants and of the full-length Pol5 protein. All trunca�on mutants were expressed from plasmids under the control of a Tetoff promoter and tagged with the Flag epitope in their N-terminal end to allow the detec�on of the different truncated proteins. As a posi�ve control, we used a full-length wild-type Pol5 also Flag-tagged and expressed from the same Tet-off promoter; as a nega�ve control, we used an empty plasmid. A B 77 RESULTS Figure 36. Pol5 trunca�on mutants. (A) Amino acid sequence of Pol5. The different trunca�on mutants are indicated in the sequence. The first (red) or the last (purple) amino acid present in each construct is indicated in bold font. (B) Schema�c representa�on of the Pol5 trunca�on variants. Pol5 has been arbitrarily divided into an N-terminal (NTD) and a C-terminal (CTD) domain. The N-terminal domain is colored in purple and the C-terminal domain in red. For mutants lacking regions within the N-terminal domain, the number in the names indicates the last truncated amino acid; for mutants lacking regions within the C-terminal domain, the numbers in the names indicate the region of truncated amino acids. (C) Predicted ter�ary structure of the different trunca�on mutants of Pol5 elaborated using the UCSF Chimera So�ware. The N-terminal domain is shown in purple, and the C-terminal domain is shown in red. The first and last amino acid are indicated in each case. Pol5 Full Length Pol5∆N37 Pol5∆N658 Pol5∆N677 Pol5∆C994-1022 Pol5∆C953-1022 Pol5∆C680-1022 NTD CTD Pol5 ∆C680-1022 ∆C994-1022 ∆C953-1022 ∆N658 ∆N677 ∆N37 E1022M1 M1 M1 M1 E1022 E1022 E1022 P38 K659 G678 W993 A952 L679 1 MTGKVNRDLF FKLASDLREE RLHAAVALIK DLSALDLPDD AEEWSYVLNR LIKGLSSDRN 61 SARLGFSLCL TEVINLAVNM PPGQRPKGLE STNEFLSTLS TILNVNVNEG TKKSMKGKDE 121 RGILFGKLFG LKSLLNEPLF SEIFVKDLEK GNTEFFIRFT EQLIDLALKK NWIKEPCFFT 181 LFQTMKMLLP FMDESSAEKI LLIYDKYDLT LTNEGLSTYL LLKYEGDESL IPSVLDLKNP 241 GWKDNDPLAR GNLPLLTKVL RNSSVIPDAN GGLKETKKQK NTNWNPRLHF VWSVLLPLFG 301 NGKLENTSHI SKKRKKTNNK KVQNSIQFPE FWKMAVDESF FNEKASSERK YLGFLIIDAA 361 FKAVPGSYIG FCFSQNVMRT LINQSIDSQR VLNKISQLTL DSIVKACEED SANRLVPCLN 421 AMLFGPHGSI NFDKLTKSGT VSKLIAIKEL PSTVLAQLLD VFFLQLQDKK GVLSHTLFAL 481 DSILHIVRAH KVEINDMDIM KPVLRPIVYM AFFKHTSDDL KLEQLHELAK ERLYSILGEL 541 TINKEIRCKD PEINSWQYLT LKLILDIENS HVGDLINPLD ENLENIKNEA ISCLSKVCRS 601 RTAQSWGLST LLSMCLVQLY AGDTDSISVI EELCEFSKHE NNSMVGITEI LLSLLAQKKA 661 LLRKLSLIIW QQFIEEVGLE ELQILLDILK ARENKQGFAQ LFEGEEEFEE IKEENDASED 721 ESKTGSESES ESESDSDDAD EKDEEDEANE DILNIDKEAT SALVKALNLP DNIVNDKGEV 781 DLDQLEGLSD DGGDDEDEES MDDEKMMELD DQLSEIFKRR KEALSSISTG NQRKFEVKQS 841 RENVISFKHR VVDMLAVYVK YCEKLTLANK SEHSNNLGGS LSKLVYFIIP MLKCVNETLD 901 RPLADKISKL LKGKIFKIKV TAFKDMNKDI ELMDLLKKTH KLMLTSKPGQ HAAVFYSMCS 961 TSSLFLSKLY VEIGGNDKLD ELIDLYTATT KEWMQKGKCG PNIFIDFINW LSSKKQTVMD 1021 KE* Pol5∆N658 Pol5∆N677 Pol5∆N37 Pol5∆C680-1022 Pol5∆C994-1022 Pol5∆C953-1022 A B C 78 RESULTS 5.2.1.- The whole sequence of Pol5 except the last α-helix is required for the synthesis of the 60S ribosomal subunit First, to address the func�onality of the mutants, we transformed the GAL::HAPOL5 strain with the different mutants and checked the growth phenotype of the truncated proteins by serial dilu�ons onto selec�ve SGal or SD plates. In SGal plates, we observed no difference in growth between the different mutants, the full-length Pol5 and the nega�ve control. This indicated that the trunca�ons had no nega�ve effect on growth (Figure 37A). However, in SD plates, upon deple�on of the endogenous wild-type HA-Pol5 protein, only the Pol5∆C994-1022 mutant, which expresses a Pol5 protein lacking only the last 29 amino acids, was able to support cell growth, apparently to the extent of the wild-type strain (Figure 37A). According to the predicted ter�ary structure of Pol5, the Pol5∆C994-1022 mutant lacks the theore�cal last α-helix, which does not appear to be essen�al. We also checked the expression of the different trunca�ons a�er a shi� to selec�ve SD medium for 6 h by western blot. As shown in Figure 37B, all truncated proteins were detected by western blo�ng, although they accumulated to a different extent than the control wild-type Flag-Pol5, especially the Pol5∆C953-1022 and Pol5∆C680-1022. Thus, it is likely that the C-terminal domain of the Pol5 protein may be required for the protein stability. We then studied the consequences of Pol5 trunca�on on ribosome synthesis. To do this, we analyzed the polysome profiles of the GAL::HA-POL5 strain transformed with the trunca�on mutants a�er a shi� from selec�ve SGal to SD medium for 6 h. As shown in Figure 38, we observed that only the full-length Pol5 protein and the Pol5∆C994-1022 mutant variant had a wild-type profile, while the rest of the mutants showed a typical profile of a 60S r-subunit deficit, with a decrease in the 60S and 80S peaks and the presence of half-mer polysomes. Overall, the profiles of all mutants, except that of the Pol5∆C994-1022 mutant, were similar to that obtained for the isogenic GAL::HA-POL5 strain transformed with an empty plasmid, although less pronounced, sugges�ng that the impairment caused by the serial dele�on of either the N- or C-terminal extension was less severe than that caused by the complete deple�on of the full-length Pol5 protein. 79 RESULTS Figure 37. Growth and expression study of the Pol5 trunca�on mutants. (A) Growth analysis of the indicated strains compared to an isogenic Flag-Pol5 control. The GAL::HA-POL5 strain was transformed with plasmids harboring the different Pol5 trunca�on proteins tagged with Flag in the N-terminal domain. Cells were spoted in serial dilu�on steps onto SGal-Leu (Gal) and SD- Leu (Glu) plates which were incubated at 30 °C for 4 days. Note that only the full-length Pol5 and the Pol5∆C994-1022 mutant allow the growth of the GAL::HA-POL5 strain in SD medium. (B) Expression analysis of the Pol5 trunca�on mutants. The indicated strains were grown in liquid SGal-Leu medium at 30 °C un�l mid-log phase and then shi�ed to SD-Leu medium for 6 h. Cells were harvested, total extracts were prepared from cells of each strain at the indicated �mes a�er the shi� and then analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. The HA-Pol5 protein was detected using an an�-HA an�body, the different trunca�ons tagged with Flag were detected using an an�-Flag an�body, and Pgk1 was used as loading control, which was detected using an�-Pgk1 an�body. The asterisk indicates the HA- Pol5 signal, that is at the top of the blot; the signal at the botom corresponds to the Flag signal. GAL::HA-POL5 +Flag-Pol5 +vector +Flag-Pol5∆C680-1022 +Flag-Pol5∆C994-1022 +Flag-Pol5∆C953-1022 +Flag-Pol5∆N658 +Flag-Pol5∆N677 +Flag-Pol5∆N37 Gal Glc Flag-Pol5 Flag-∆C994-1022 Flag-∆N37 GAL::HA-POL5 0 2 64 h Glc Flag Pgk1 *** * ****HA-Pol5 * 0 2 64 0 2 64 h Glc vector GAL::HA-POL5 0 2 6 4 Flag Pgk1 *HA-Pol5 0 2 64 Flag-∆C953-1022 HA-Pol5 Pgk1 h Glc Flag-∆C680-1022 Flag-∆N677 Flag-∆N658 GAL::HA-POL5 0 2 64 Flag Pgk1 HA-Pol5 0 2 64 0 2 64 A B 86 RESULTS We then analyzed whether selected RAFs (Has1 and Nop7) or 60S r-proteins (eL14) were present in the purified complexes by western blo�ng. As shown in Figure 42, we detected efficient precipita�on of all GFP-containing proteins by the GFP-Trap beads, and that both GFP-Pol5 and GFP-Pol5∆C680-1022 precipitates could co-enrich Nop7, Has1 and eL14, indica�ng that Pol5 and Pol5∆C680-1022 bind to pre-60S r-par�cles. However, the Pol5∆N677 variant was unable to significantly precipitate these proteins above the background levels (Figure 42). Figure 42. The N-terminal domain of Pol5 associates with ribosome assembly factors and ribosomal proteins from pre-60S ribosomal par�cles. The associa�on of the GFP-tagged trunca�on mutants with different proteins related to the matura�on of the 60S r-subunit was studied using the GFP-Trap immunoprecipita�on procedure. The GAL::HA-POL5 strain was transformed with different plasmids harboring an N-terminal and a C-terminal GFP-tagged Pol5 truncated variant, the full-length Pol5, and an empty plasmid expressing the GFP tag. Cells were grown in SGal-Ura-Trp at 30 °C un�l mid-log phase, then shi�ed to SD-Ura-Trp for 6 h and harvested. Proteins were extracted from the whole cell extracts (T) and from the immunoprecipita�on (IP), and then analyzed by western blo�ng. We loaded 30% volume of each extract. The different GFP-tagged proteins were detected using an an�-GFP an�body and specific an�bodies were used to detect the RAFs Nop7 and Has1 and the eL14 r-protein. These results suggest that the N-terminal domain of Pol5 is required for its stable associa�on with pre-60S r-par�cles. T IP Nop7 GAL::HA-POL5 GFP Has1 eL14 87 RESULTS 5.2.4.- Trans-complementa�on of Pol5 restores 60S ribosomal subunit synthesis As the predicted structure of Pol5 establishes interac�ons between the different regions of the protein, sugges�ng the possibility that it could be formed by different independent subdomains, we decided to study whether the N- and C-terminal truncated proteins could be func�onal in trans. For this purpose, we co-transformed the GAL::HAPOL5 strain with combina�ons of two plasmids, one containing a mutant lacking the C- terminal domain and the other one containing a mutant lacking the N-terminal domain. The following combina�ons were used: GFP-Pol5∆C680-1022+Flag-Pol5∆N677, GFP- Pol5∆N677+Flag-Pol5∆C680-1022, GFP-Pol5∆C680-1022+Flag-Pol5∆N658 and GFP- Pol5∆N658+FlagPol5∆C680-1022. The first two combina�ons recons�tuted the fulllength Pol5 protein with a small overlap of 2 amino acids, with the Flag or the GFP tag between the two construc�ons, respec�vely. The last two combina�ons presented an overlap of 21 amino acids, with the Flag or the GFP tag between the two construc�ons, respec�vely. We also co-transformed the same strain with the GFP-tagged full-length Pol5 and an empty vector as a control (Figure 43A). We first analyzed the growth phenotype of these combina�ons by serial dilu�ons onto selec�ve SGal or SD plates. None of the combina�ons caused a dominant nega�ve effect on growth in SGal plates. Interes�ngly, cells expressing either the GFP-Pol5∆C680- 1022+Flag-Pol5∆N677 or GFP-Pol5∆N677+Flag-Pol5∆C680-1022 combina�on were able to grow in SD plates, where the endogenous HA-Pol5 is depleted. The growth was apparently similar to that obtained with expression of the full-length GFP-Pol5 protein (Figure 43B). These combina�ons that allowed growth restored the protein without a large overlap, only 2 amino acids. However, the other combina�ons with an overlap of 21 amino acids did not allow growth in SD plates (Figure 43B). To discard the possibility that this complementa�on was the result of a recombina�on event taking place in vivo, we checked the expression of the different truncated proteins by western blot analysis a�er a shi� to selec�ve SD medium for 6 h, using the Flag and GFP tags to track them. As shown in Figure 43C, neither a full-length GFP-tagged nor a full-length Flag-tagged Pol5 protein could be detected, only the different truncated proteins, thus, discarding the above possibility. 88 RESULTS Figure 43. Trans-complementa�on analysis of Pol5. (A) Schema�c representa�on (le�) and predicted ter�ary structure (right) of different mutant combina�ons. The N-terminal domain is represented in purple and the C-terminal domain in red. The ter�ary structure was elaborated using the UCSF Chimera So�ware (B) Growth analysis of trunca�on mutant combina�ons. The GAL::HA-POL5 strain was co-transformed with plasmids harboring the indicated N- and C- terminal Pol5 trunca�ons, tagged with GFP or Flag in their N-terminal end. A strain transformed with a plasmid expressing a GFP-tagged full-length Pol5 protein and an empty plasmid were used as control. This strain was also transformed with a plasmid expressing the nucleolar marker DsRed-Nop1 from its cognate promoter. Cells were spoted in serial dilu�on steps onto SGal- Leu-Ura-Trp (Gal) and SD-Leu-Ura-Trp (Glu) plates which were incubated at 30 °C for 4 days. Note that growth is only recovered in SD medium when a fully func�onal Pol5 could be restored. (C) Expression analysis of trunca�on mutant combina�ons. The indicated strains were grown in liquid SGal-Leu-Ura-Trp medium at 30 °C un�l mid-log phase and then shi�ed to SD-Leu-Ura- Trp medium for 6 h. Cells were harvested and total extracts were prepared from cells of each strain at the indicated �mes a�er the shi� and then analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. The HA-Pol5 protein was detected using an an�- HA an�body, the different trunca�ons tagged with GFP and Flag were detected using an�-GFP or an�-Flag an�bodies, and Pgk1 was used as loading control, which was detected using an�- Pgk1 an�body. The asterisk indicates the residual HA-Pol5 signal. Gal Glc +GFP-Pol5+vector GAL::HA-POL5 +GFP-Pol5∆C680-1022 +Flag-Pol5∆N658 +GFP-Pol5∆C680-1022 +Flag-Pol5∆N677 +GFP-Pol5∆N658 +Flag-Pol5∆C680-1022 +GFP-Pol5∆N677 +Flag-Pol5∆C680-1022 NTD CTD Pol5+vector ∆C680-1022+∆N677 ∆C680-1022 +∆N658 E1022M1 E1022 E1022 M1 M1 L679 L679 G678 K659 h Glc GAL::HA-POL5 GFP Pgk1 Flag HA-Pol5 0 6 0 6 0 6 ** GAL::HA-POL5 GFP Pgk1 HA-Pol5 Flag h Glc 0 6 0 6 A B C Pol5∆C680-1022 + Pol5∆N658 Pol5∆C680-1022 + Pol5∆N677 89 RESULTS We analyzed the polysome profiles of the different combina�ons to assess whether, in the cases where growth was restored in non-permissive condi�ons, the produc�on of the 60S r-subunit was also recovered to the wild-type extent. We used the same combina�ons as above and studied the polysome profile a�er a shi� from selec�ve SGal to SD medium for 6 h. As a result, a wild-type polysome profile was obtained for those combina�ons that restored growth in SD plates, while polysome profiles typical of a 60S r-subunit synthesis defect were obtained for those combina�ons that could not restore growth in SD plates (Figure 44). Figure 44. Trans-complementa�on of Pol5 restores 60S r-subunit synthesis. The GAL::HA-POL5 strain transformed with plasmids containing the indicated combina�on of trunca�on Pol5 variants and a plasmid expressing the nucleolar marker DsRed-Nop1 from its cognate promoter was grown in SGal-Leu-Ura-Trp medium at 30 °C un�l mid-log phase and then shi�ed to SD-Leu- Ura-Trp medium for 6 h. Cells were harvested, whole cell extracts were prepared and 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. The peaks of free 40S and 60S r-subunits, 80S free couples/monosomes and polysomes are indicated. Half-mers are labeled with arrows. A 254 Sedimentation GAL::HA-POL5 +Flag-Pol5 +vector 60S 40S 80S polysomes 60S 40S polysomes GAL::HA-POL5 +GFP-∆C680-1022 +Flag-∆N658 80S GAL::HA-POL5 +GFP-∆C680-1022 +Flag-∆N677 60S 40S polysomes 80S GAL::HA-POL5 +GFP-∆N658 +Flag-∆C680-1022 60S 40S 80S polysomes GAL::HA-POL5 +GFP-∆N677 +Flag-∆C680-1022 60S 40S polysomes 80S 90 RESULTS These results indicate that Pol5 contains two independent domains that can be physically separated, but recons�tuted in trans a fully func�onal protein that carries out its role during 60S r-subunit biogenesis. The presence of a substan�al overlapping surface does not allow the proper recons�tu�on of the structure of the full-length protein, thus preven�ng the complementa�on of the growth and of the 60S r-subunit synthesis defect upon deple�on of Pol5. 5.2.5.- Trans-complementa�on does not totally restored the nucleolar localiza�on of the N-terminal domain mutant As we have already shown, the trunca�on mutants lacking the N-terminal domain had a not completely nucleolar localiza�on (Figure 41). We decided to test whether the combina�on of the construct lacking the N-terminal domain with that lacking the C-terminal domain, which allowed trans-complementa�on, also allowed normal nucleolar localiza�on of the restored Pol5 protein. To study this, we analyzed by fluorescence microscopy the subcellular localiza�on of the GFP-tagged Pol5∆C680-1022 and Pol5∆N677 truncated mutants when they were combined in a GAL::HA-POL5 strain, recons�tu�ng the full-length Pol5 protein. As shown in Figure 45, the localiza�on of the GFP-tagged trunca�ons was assessed in selec�ve SGal and a�er a shi� to SD medium for 6 h. In the case of the GFP- Pol5∆C680-1022 protein, the construct localized in the nucleolus when combined with Flag-Pol5∆N677 in both SGal and SD medium, as indicated by the co-localiza�on with the nucleolar marker DsRed-Nop1. Surprisingly, despite the trans-complementa�on, the GFP-Pol5∆N677 protein s�ll showed the same localiza�on when was combined with Flag-Pol5∆C680-1022 as when expressed alone, in both growth condi�ons. This trunca�on showed a cytoplasmic accumula�on, probably forming an aggregate, and a minor nucleolar co-localiza�on with the DsRed-Nop1 marker (Figure 45). 91 RESULTS Figure 45. Subcellular localiza�on of different combina�ons of the N- and C-terminal trunca�ons of Pol5. The GAL::HA-POL5 strain was co-transformed with plasmids harboring the GFP-Pol5∆C680-1022 and Flag-Pol5∆N677 trunca�ons (le�) and the GFP-Pol5∆N677 and Flag- Pol5∆C680-1022 trunca�ons (right), as well as a plasmid expressing the nucleolar marker DsRed-Nop1 from its cognate promoter. Cells were grown in selec�ve SGal medium at 30 °C un�l mid-log phase, then cultures were maintained in SGal medium or shi�ed to SD medium for 6 h. The GFP and DsRed-Nop1 signals were analyzed by fluorescence microscopy. Arrows point the accumula�ons of the GFP-tagged trunca�ons and the nucleolar localiza�on. Note that the Pol5∆N677 trunca�on form aggregates outside the nucleus even when combined with the Pol5∆C680-1022 construct. 5.2.6.- The interac�on of N- and C-terminal truncated constructs may not be stable enough to be detected by copurifica�on We analyzed whether the Pol5∆C680-1022 and Pol5∆N677 trunca�ons interacted with each other when both were present in the GAL::HA-POL5 strain. To do this, we purified GFP-containing complexes using the GFP-Trap procedure (see Materials and Methods) a�er a shi� from selec�ve SGal to SD medium for 6 h of the strain transformed with the GFP-Pol5∆C680-1022+Flag-Pol5∆N677 or the GFP- Pol5∆N677+Flag-Pol5∆C680-1022 combina�ons, that allowed trans-complementa�on. GAL::HA-POL5 Gal Glc 6 h GFP-Pol5∆C680-1022 DsRed-Nop1 merge Gal Glc 6 h GFP-Pol5∆N677 DsRed-Nop1 merge 92 RESULTS However, as shown in Figure 46, regardless of the trunca�on that we used to perform the purifica�on, there was no enrichment of the other corresponding Pol5 fragment, as indicated by the absence of signal from the Flag an�body in both copurifica�on experiments. In contrast, Nop7, Has1 and eL14 could be detected in the case of the GFP-Pol5∆C680-1022 purifica�on, but not in the case of the GFP-Pol5∆N677 purifica�on, except for a slight enrichment of the eL14 r-protein (Figure 46). Figure 46. Analysis of the protein-protein interac�ons of different combina�ons of N- and C- terminal trunca�ons of Pol5. A GFP-Trap immunoprecipita�on procedure was used to study the possible interac�on between the Pol5∆C680-1022 and Pol5∆N677 trunca�ons. The GAL::HAPOL5 strain was co-transformed with the plasmids harboring each construc�on, tagged with GFP or Flag, as indicated. Cells were grown in SGal-Leu-Ura-Trp at 30 °C un�l mid-log phase, then shi�ed to SD-Leu-Ura-Trp medium for 6 h and harvested. Proteins were extracted from the whole cell extracts (T) and from the immunoprecipita�on (IP), and then analyzed by western blo�ng. We loaded 30% volume of each extract. The GFP-tagged and Flag-tagged trunca�ons were detected using an�-GFP or an�-Flag an�bodies, respec�vely, and specific an�bodies were used to detect the RAFs Nop7 and Has1 and the eL14 r-protein. Taken together, these results indicate that, although the trans-complementa�on func�onal assay strongly suggests that a full-length Pol5 protein could be recons�tuted from its N- and C-terminal domains, the interac�on between these independent domains is likely so weak than the harsh physicochemical constraints imposed by the GFP-Trap purifica�on experiment are not able to retain it. Nop7 GAL::HA-POL5 T IP GFP Has1 eL14 Flag 93 RESULTS 5.3.- Study of the effects of the heterologous expression of human MYBBP1A in yeast 5.3.1.- The expression of MYBBP1A has a nega�ve effect on growth Yeast Pol5 and mammalian MYBBP1A are predicted to be orthologs [149]. Therefore, we decided to assess whether MYBBP1A is func�onally equivalent to Pol5 and, thus, could supply the absence of Pol5 in yeast. To study this, we used a POL5 shuffle strain, harboring a pol5∆ genomic dele�on in which the POL5 gene has been replaced by a kanMX4 cassete. This strain also expresses a wild-type POL5 allele from a YCplac33- POL5 plasmid as the sole source of Pol5. We transformed this strain with a pAS24 plasmid containing the cDNA derived from the expressed full isoform of the human protein MYBBP1A (p160) tagged with HA in its N-terminal domain and expressed under the control of a GAL promoter. As controls, we also transformed the strain with an empty pAS24 plasmid, and a YCplac111-POL5 plasmid, which allows expression of wild-type yeast Pol5 from its cognate promoter, thus, independently of the carbon source used. With these strains, we performed growth tests on SD, SGal or SGal containing 5-FOA plates. Cells containing a func�onal URA3 gene are sensi�ve to 5-FOA, therefore transformants unable to lose the YCplac33-POL5 plasmid (URA3) should not grow on SGal plates containing 5-FOA. As shown in Figure 47A, on SD plates, there were no differences in growth between the strains, as expected, because in the three cases a wild-type POL5 allele was expressed and no expression of MYBBP1A was expected. On SGal plates, the wild-type POL5 alleles were expressed, but at the same �me there was expression of the HAMYBBP1A protein from its plasmid, which had a nega�ve effect on the growth of the corresponding strain. On SGal plates containing 5-FOA, only the transformants that were able to lose the YCplac33-POL5 plasmid could grow, and there was expression of the HAMYBBP1A protein in the corresponding strain. Only the transformant that had another source of Pol5 (YCplac111-POL5) was able to lose the YCplac33-POL5 plasmid, and in the case of the transformants harboring the empty pAS24 or the pAS24-HA-MYBBP1A plasmids, no growth could be detected. Thus, we conclude that the expression of human HA-MYBBP1A protein was unable to complement the lethal phenotype associated with 94 RESULTS the POL5 dele�on and caused a nega�ve effect on growth (Figure 47A). Western blot analysis indicated that the lack of complementa�on was not due to deficient HAMYBBP1A protein expression (Figure 47B). Figure 47. The human protein MYBBP1A cannot func�onally replace yeast Pol5. (A) Analysis of the effects of the expression of an HA-tagged construct of human MYBBP1A on yeast growth. The POL5 shuffle strain was transformed with the YCplac111-POL5, empty pAS24 and pAS24- HA-MYBBP1A plasmids. Transformants were grown in liquid SD-Leu medium and diluted to an OD600 of 0.05 and then they were spoted in serial dilu�on steps onto SD-Leu-Ura (Glc), SGal- Leu-Ura (Gal) and SGal containing 5-FOA (5-FOA) plates, which were incubated at 30 °C for 3-4 days. Note that HA-MYBBP1A causes a nega�ve growth phenotype when expressed in a strain with an endogenous copy of the POL5 gene and is unable to complement the lack of yeast Pol5. (B) Expression of HA-MYBBP1A. The POL5 shuffle strain transformed with the pAS24-HA- MYBBP1A plasmid was grown in liquid SD-Leu-Ura medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu-Ura for 12 h. The GAL::HA-POL5 strain was used as a control. Cells were harvested and total extracts were prepared and analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. HA-Pol5 and HA-MYBBP1A were detected using an an�-HA an�body. Pgk1 was used as loading control, which was detected using an�-Pgk1 an�body. It has been described that under stress condi�ons, MYBBP1A undergoes a proteoly�c processing to produce the p140 and p67 isoforms, which translocate from the nucleolus to the nucleoplasm. The p140 isoform only loses the region responsible for the nucleolar localiza�on of the full-length MYBBP1A protein in humans [172]. We then tested whether this isoform had the same nega�ve effect as the full-length MYBBP1A protein, and whether it could complement the lack of Pol5 in yeast. To obtain the p140 isoform, we introduced a point muta�on that led to the subs�tu�on of the K1170 amino acid by a stop codon in the corresponding cDNA from the full-length HA-MYBBP1A HA-Pol5 Pgk1 Glc Gal 5-FOA +POL5 +vector +HA-MYBBP1A pol5∆ [Ycplac33-POL5] A B 95 RESULTS MYBBP1A protein. Thereby, we ar�ficially obtained the p140 isoform, which we also tagged with an HA epitope in its N-terminal end and expressed under the control of a GAL promoter in the pAS24 plasmid. Growth assays were then performed as previously described. When the p140 isoform of MYBBP1A was expressed in a POL5 shuffle strain (SGal plates), no nega�ve effect on growth was detected. However, there was neither complementa�on to the absence of Pol5 as no growth could be detected on SGal plates containing 5-FOA (Figure 48A). Western blot analysis showed that the p140 isoform of MYBBP1A was also expressed (Figure 48B). Figure 48. The p140 isoform of MYBBP1A cannot func�onally replace yeast Pol5. (A) Analysis of the effects of the expression of an HA-tagged construct of the p140 isoform of human MYBBP1A on yeast growth. The POL5 shuffle strain was transformed with the following plasmids: YCplac111-POL5, empty pAS24, pAS24-HA-MYBBP1A and pAS24-HA-MYBBP1A-p140. Transformants were spoted in serial dilu�on steps onto SD-Leu-Ura (Glc), SGal-Leu-Ura (Gal) and SGal containing 5-FOA (5-FOA) plates, which were incubated at 30 °C for 4 days. Note that only full-length HA-MYBBP1A causes a nega�ve growth phenotype when expressed in a strain with an endogenous copy of the POL5 gene and is unable to complement the lack of yeast Pol5. In addi�on, the p140 isoform has no nega�ve effect on growth, but is also unable to complement the lack of yeast Pol5. (B) Expression of full-length HA-MYBBP1A and p140 isoforms. The POL5 shuffle strain with pAS24-HA-MYBBP1A or pAS24-HA-MYBBP1A-p140 was grown in liquid SRaf-Leu-Ura medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu- Ura for 6 h. Cells were harvested and total extracts were prepared and analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. HA-MYBBP1A and HA-MYBBP1A p140 were detected using an an�-HA an�body. Pgk1 was used as loading control, which was detected using an�-Pgk1 an�body. Glc Gal 5-FOA pol5∆ [Ycplac33-POL5] +POL5 +vector +HA-MYBBP1A +HA-MYBBP1A p140 HA-MYBBP1A Pgk1 HA-MYBBP1A p140 A B 102 RESULTS BY4742 strain with the pAS24 plasmid harboring the HA-MYBBP1A protein under the control of a GAL promoter. Transformants were grown in SRaf-Leu medium and then shi�ed to SGal-Leu medium for 6 h to induce the expression of HA-MYBBP1A. Then, we studied the localiza�on of the HA-MYBBP1A protein by western blo�ng in the frac�ons of the total r-subunits obtained using low Mg2+ run-off sucrose gradients. As a result, we observed that HA-MYBBP1A was predominantly located in the frac�ons corresponding to the peaks of the 40S and 60S r-par�cles, as indicated by the sedimenta�on of eS8 and eL14 r-proteins, being more abundant in the frac�ons of the 60S r-par�cles (Figure 54). We compared this patern with that obtained for HA-Pol5, which is also expressed under the control of the same GAL promoter and should have the same level of expression. The distribu�on was similar in both cases (Figure 22). Figure 54. HA-MYBBP1A protein is present in ribosomal par�cles frac�ons. The BY4742 strain transformed with the pAS24-HA-MYBBP1A plasmid was grown in SRaf-Leu medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu medium for 6 h. Cells were harvested and whole cell extracts were prepared in a low Mg2+ buffer lacking cycloheximide to dissociate the r-subunits. 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. Frac�ons were collected from the gradients, proteins were extracted from each one and equal volumes were analyzed by western blo�ng. The posi�ons on the profile correspond to those of the frac�ons on the blots. We detected the HA-MYBBP1A protein with an an�-HA an�body. Specific an�bodies were used to detect the eS8 and eL14 r-proteins. Peaks of total 40S and 60S r-subunits are indicated. These results, along with the results of the subcellular localiza�on of HAMYBBP1A and the GFP-Trap assay, strongly suggest that MYBBP1A associates with nucleolar par�cles on the road to mature 60S r-subunits. BY4742 +HA-MYBBP1A Gal 6 h HA-MYBBP1A eS8 eL14 1 3 5 6 7 8 9 11T 2 4 10 12 60S 40S 103 RESULTS 5.3.6.- MYBBP1A has an effect on pre-rRNA processing We decided to study whether the nega�ve effect of MYBBP1A on the synthesis of 60S r-subunits was related to defects in the pre-rRNA processing pathway. Therefore, we analyzed the effects of MYBBP1A expression on the steady-state levels of pre- and mature rRNAs by northern blo�ng. We used a BY4742 strain transformed with the empty pAS24 plasmid or harboring HA-MYBBP1A under the control of a GAL promoter, taking samples every 3 h a�er a shi� from SRaf to SGal medium for up to 12 h. We also followed the expression of the protein by western blot analysis (Figure 55C). Figure 55. Expression of HA-MYBBP1A impairs pre-rRNA processing in yeast. The BY4742 strain transformed with empty pAS24 (vector) or pAS24-HA-MYBBP1A plasmid was grown in SRaf-Leu medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu medium. Cells were harvested and total RNA was extracted from cells of each strain at the indicated �mes a�er the shi�. Pre-rRNAs and mature rRNAs were analyzed by northern blo�ng using the same amount of RNA (5µg). (A) Northern blot analysis of high-molecular-mass pre- and mature rRNAs. (B) Northern blot analysis of low-molecular-mass pre- and mature rRNAs. The probes used are indicated between parentheses and described in the Table 4 from the Materials and Methods sec�on. Signal intensi�es were measured by phosphorimager scanning and the values were normalized to those obtain in the control strain in SRaf medium, arbitrarily set at 1.0. (C) Expression of HA-MYBBP1A. Cells were harvested and total extracts were prepared from cells of each strain at the indicated �mes a�er the shi� and then analyzed by western blo�ng. Equal amounts of protein were loaded in each lane. HA-MYBBP1A protein was detected using an an�- HA an�body. Pgk1 was used as loading control, which was detected using an�-Pgk1 an�body. (D) Schema�c simplifica�on of the main steps of pre-rRNAs processing in yeast. 35S 20S 18S 27SA2 27SA3 27SB 7S 5.8S 25.5S 25S 40S r-subunit 60S r-subunit 7S 0 6 12 0 63 9 12 HA-MYBBP1Avector 5.8SL 5.8SS 5S 1.0 1.1 0.9 0.7 0.6 0.8 0.6 0.5 1.0 1.1 1.2 0.9 1.1 1.3 1.1 0.8 (5.8SS) h Gal 1.0 1.0 1.0 0.8 1.0 1.0 1.0 0.7 (f) (e) (5S) h Gal HA-MYBBP1A Pgk1 0 6 12 0 63 9 12 HA-MYBBP1A vector C A B D 35S 32S 0 6 12 0 63 9 12 vector HA-MYBBP1A 27SA2 1.0 27SBL/S 25S 20S 18S 1.0 0.8 1.4 1.0 1.4 1.4 1.0 1.0 1.1 0.9 1.1 1.1 1.6 1.7 1.1 (32S) (35S) 1.0 1.5 1.2 1.1 1.6 1.9 1.7 1.0 1.0 1.2 1.0 1.0 1.5 2.6 2.9 1.6 1.0 0.6 0.7 0.9 0.7 0.7 0.6 0.4 1.0 1.0 0.9 0.9 1.0 1.6 1.5 1.0 h Gal 1.0 0.9 0.9 0.8 0.9 1.0 1.0 0.8 27SA (d) (d) (g) (h) (c) (b) (27SB) 104 RESULTS As shown in Figure 55A, expression of MYBBP1A disrupted the matura�on of 25S rRNA, but not that of 18S rRNA, as indicated by the decrease in the levels of 25S but not of 18S rRNA upon MYBBP1A expression. The levels of 35S and 32S pre-rRNAs increased, as well as those of 27SA2, 27SB and 20S pre-rRNAs, although the accumula�on was more pronounced in the case of 27SB pre-rRNA (Figure 55A). Analysis of low-molecular-mass rRNAs revealed no changes in the levels of the different pre- and mature rRNAs, except for a slight decrease in the levels of 7S pre-rRNA upon the expression of MYBBP1A (Figure 55B). It has been described that the small mature rRNAs are more stable than the 25S and 18S rRNAs and are, therefore, depleted more slowly upon a failure of their synthesis [216, 217], which could explain why we did not observe any significant changes for 5.8S and 5S rRNAs. These results suggest that MYBBP1A interferes with the normal processing of the 27SB pre-rRNA, which is likely to be the cause of the shortage of 60S r-subunits and the nega�ve effect on growth observed upon its expression. 5.3.7.- MYBBP1A expression causes the accumula�on of uL23 ribosomal protein in the nucleolus We also inves�gated whether the expression of MYBBP1A affects the nuclear export of pre-ribosomal par�cles to the cytoplasm. To do so, we used the uL23-eGFP and uS3-eGFP r-proteins as reporters of the 60S and 40S r-subunits, respec�vely. We transformed the BY4742 strain with either an empty plasmid or the plasmid pAS24 harboring HA-MYBBP1A under the control of a GAL promoter. The transformants were then grown in SRaf-Leu-Ura medium and shi�ed to SGal-Leu-Ura medium for 6 h to allow the expression of MYBBP1A. We then checked the localiza�on of the reporters by fluorescence microscopy. In the control strain, we observed that both uS3-eGFP and uL23-eGFP proteins were located in the cytoplasm. In contrast, upon expression of MYBBP1A, we observed that the uS3-eGFP protein was in the cytoplasm, but a small accumula�on of the uL23-eGFP protein could be detected in the nucleolus, which was revealed by the co-localiza�on with the nucleolar marker DsRed-Nop1 (Figure 56). 105 RESULTS Figure 56. Expression of MYBBP1A causes a slight nucle(ol)ar reten�on of uL23-eGFP ribosomal protein. The BY4742 strain transformed with empty pAS24 (vector) or pAS24-HA- MYBBP1A plasmid was also transformed with plasmids expressing DsRed-Nop1 and either uS3- eGFP or uL23-eGFP from their cognate promoters. Cells were grown in selec�ve SRaf medium at 30 °C un�l mid-log phase and then shi�ed to selec�ve SGal medium for 6 h. GFP and DsRed- Nop1 signals were analyzed by fluorescence microscopy. Arrows point the mild accumula�on of uL23-eGFP with the nucleolar marker DsRed-Nop1. This result suggests that the expression of MYBBP1A slightly interferes with the nuclear export of pre-60S r-par�cles to the cytoplasm. However, since Pol5 interacts with uL23 [155], we cannot exclude that this accumula�on could also be caused by a specific failure in the assembly of this r-protein caused by the expression of MYBBP1A. However, given the interference that MYBBP1A causes in pre-rRNA processing and its associa�on with pre-60S r-par�cles, this later scenario is unlikely. 5.3.8.- Expression of MYBBP1A does not affect the subcellular localiza�on of Pol5 protein We analyzed whether the nega�ve effects of MYBBP1A on ribosome synthesis were related to altera�ons in the normal func�on of Pol5. uS3-eGFP DsRed-Nop1 merge uL23-eGFP DsRed-Nop1 merge +vector +HA-MYBBP1A uS3-eGFP DsRed-Nop1 merge uL23-eGFP DsRed-Nop1 merge 106 RESULTS To clarify this, we first studied whether the expression of MYBBP1A alters the subcellular localiza�on of Pol5. For this purpose, we transformed a POL5-GFP strain with the empty pAS24 plasmid or harboring HA-MYBBP1A under the control of a GAL promoter. Transformants were then grown in SRaf-Leu-Trp medium and shi�ed to SGal- Leu-Trp medium for 6 h to induce the expression of MYBBP1A. We then analyzed the localiza�on of the Pol5-GFP protein by fluorescence microscopy. As a result, we observed no differences in the localiza�on of Pol5 between the control strain and the strain expressing the MYBBP1A protein. In both cases, the Pol5- GFP protein was nucleolar, as shown by the co-localiza�on of its signal with the nucleolar marker DsRed-Nop1 (Figure 57). Figure 57. The expression of the human MYBBP1A does not affect the subcellular localiza�on of yeast Pol5. The POL5-GFP strain was transformed with empty pAS24 (vector) or pAS24-HA- MYBBP1A plasmid and also with a plasmid expressing DsRed-Nop1 from its cognate promoter. Cells were grown in selec�ve SRaf medium at 30 °C un�l mid-log phase and then shi�ed to selec�ve SGal medium for 6 h. Pol5-GFP and DsRed-Nop1 signals were analyzed by fluorescence microscopy. Arrows point the nucleolar localiza�on of DsRed-Nop1. Note the co-localiza�on of Pol5-GFP and Nop1 in both cases. This result strongly suggests that the expression of MYBBP1A does not significantly interfere with the subcellular localiza�on of Pol5, thus, the perturba�on that its expression causes on the 60S r-subunit synthesis cannot be explained by this scenario. +vector Pol5-GFP +HA-MYBBP1A DsRed-Nop1 merge POL5-GFP 107 RESULTS 5.3.9.- Pol5 associa�on with pre-ribosomal par�cles is not altered by expression of MYBBP1A We also analyzed whether the associa�on of Pol5 with pre-ribosomal par�cles was affected by the expression of MYBBP1A. To study this, we checked whether the sedimenta�on patern of Pol5 in sucrose gradients changed upon MYBBP1A expression. We used a POL5-GFP strain transformed with either an empty pAS24 or pAS24-HA- MYBBP1A plasmid. Transformants were grown in SRaf-Leu medium and then shi�ed to SGal-Leu medium for 6 h to allow the expression of HA-MYBBP1A. We then studied the localiza�on of the Pol5-GFP protein by western blo�ng in the frac�ons of the total rsubunits obtained using low Mg2+ run-off sucrose gradients. A large propor�on of the Pol5-GFP signal was found to be associated with frac�ons of high-molecular-mass par�cles. These frac�ons correspond to r-par�cles, as indicated by the sedimenta�on of the eS8 and eL14 r-proteins. No protein was detected at the posi�ons expected for the free proteins (Figure 58A). Our results showed that this localiza�on did not change a�er expression of MYBBP1A (Figure 58B). As a second experiment, we affinity purified Pol5-GFP-containing complexes using the GFP-Trap procedure (see Materials and Methods) a�er the shi� from SRaf-Leu to SGal-Leu medium for 6 h. As a nega�ve control, we used an isogenic POL5 untagged strain transformed with the pAS24 plasmid harboring HA-MYBBP1A, which was subjected to the same procedure. We then analyzed the proteins that co-purified with Pol5-GFP. As a result, we observed an enrichment in the levels of the eL14 r-protein and the RAFs Has1 and Nop7 in the immunoprecipitated samples compared to the nega�ve control. It is worth men�oning that we did not observe any enrichment of the HAMYBBP1A protein in the obtained complexes (Figure 59). Although these results suggest that the Pol5-GFP and HA-MYBBP1A proteins do not interact with each other and are not simultaneously present in the same preribosomal par�cles, we firmly believe that other scenarios are s�ll possible (see Discussion). In addi�on, the expression of MYBBP1A does not appear to significantly affect the ability of Pol5 to associate with pre-ribosomal par�cles. 108 RESULTS Figure 58. The MYBBP1A expression does not change the sedimenta�on of Pol5 in sucrose gradients. (A) The POL5-GFP strain transformed with empty pAS24 (vector) was grown in SRaf- Leu medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu medium for 6 h. Cells were harvested and whole cell extracts were prepared in a low Mg 2+ buffer lacking cycloheximide to dissociate the r-subunits. 10 A260 units of each extract were resolved in 7-50% sucrose gradients. The A254 was con�nuously measured. Sedimenta�on is from le� to right. Frac�ons were collected from the gradients, proteins were extracted from each one and equal volumes were analyzed by western blo�ng. (B) The POL5-GFP strain transformed with a pAS24- HA-MYBBP1A plasmid was grown in SRaf-Leu medium at 30 °C un�l mid-log phase and then shi�ed to SGal-Leu medium for 6 h. The frac�ons were then analyzed as described above. Extracts were obtained and frac�ons were analyzed as described above. The posi�ons on the profiles correspond to those of the frac�ons on the blots. The Pol5-GFP protein was detected with an an�-GFP an�body. Specific an�bodies were used to detect the eS8 and eL14 r-proteins. Peaks of total 40S and 60S r-subunits are indicated. APOL5-GFP +vector Gal 6 h Pol5-GFP eS8 eL14 1 3 5 6 7 8 9 11 T2 4 10 12 60S 40S POL5-GFP +HA-MYBBP1A Gal 6 h Pol5-GFP eS8 eL14 135 6 7 8 9 11T 2 410 12 60S 40S B 109 RESULTS Figure 59. Pol5 and MYBBP1A seem not to interact with each other. Associa�on of different proteins related to the matura�on of the 60S r-subunit with Pol5 was studied using the GFP- Trap immunoprecipita�on procedure with cell extracts from the POL5-GFP strain transformed with an empty pAS24 (vector) or pAS24-HA-MYBBP1A plasmid and a wild-type strain (W303- 1A) with untagged Pol5 transformed with the same plasmids as a control. Cells were grown in SRaf-Leu at 30 °C un�l mid-log phase, then shi�ed to SGal-Leu medium for 6 h and harvested. Proteins were extracted from the whole cell extracts (T) and from the immunoprecipita�on (IP), and then analyzed by western blo�ng. We loaded 30% volume of each extract. The Pol5-GFP and HA-MYBBP1A proteins were detected using an an�-GFP and an an�-HA an�bodies, respec�vely. Specific an�bodies were used to detect the RAFs Nop7 and Has1 and the eL14 rprotein. Note that this figure completes the informa�on of Figure 24 from the Results sec�on 5.1.4. 5.3.10.- Overexpression of the eL27A r-protein does not suppress the nega�ve effect of MYBBP1A on growth Since we have already shown that the overexpression of the 60S r-protein eL27A suppresses the thermosensi�ve phenotype of the pol5-1 and pol5-3 mutants (Results 5.1.8.), we decided to check whether this overexpression could also suppress or alleviate the nega�ve effect caused by the MYBBP1A expression on growth. To analyze this, we used a BY4742 strain co-transformed with either an empty pAS24 or a pAS24-HA- MYBBP1A plasmid and with either an empty YEplac195 or a YEplac195-RPL27A plasmid. We performed growth tests with the transformants on SD and SGal plates. HA-MYBBP1A Pol5-GFP Nop7 Has1 IPT eL14 110 RESULTS No nega�ve effect was observed in the case of the control strains without MYBBP1A. However, with the expression of MYBBP1A, we observed the typical nega�ve effect on growth, independently of the overexpression of eL27A (Figure 60). This result suggests that the overexpression of the eL27A r-protein is not enough to suppress the nega�ve effect on growth caused by the expression of MYBBP1A. Figure 60. High-dosage of eL27A does not suppress the nega�ve growth defect caused by MYBBP1A expression. The BY4742 strain transformed with empty pAS24 (vector) or pAS24-HA- MYBBP1A plasmid was also transformed with the mul�copy plasmids YEplac195 and YEplac195- RPL27A. Cells were spoted in serial dilu�on steps onto SD-Leu-Ura (Glc) and SGal-Leu-Ura (Gal) plates which were incubated at 30 °C for 4 days. Note that the expression of HA-MYBBP1A causes a growth defect regardless of the dose of the eL27A r-protein. 5.3.11.- Dele�on of BAS1 does not suppress the nega�ve effect of MYBBP1A on growth MYBBP1A binds and represses the proto-oncogene c-Myb in humans [160]. To discard that part of the nega�ve effect that we observed in growth upon MYBBP1A expression was due to a possible interac�on with Bas1, the puta�ve ortholog of human c-Myb [165], we tested whether the loss-of-func�on of Bas1 alleviated the nega�ve effect caused by the expression of MYBBP1A. To do so, we used a bas1 null strain transformed with either an empty pAS24 or a pAS24-HA-MYBBP1A plasmid. Then, we performed growth assays with the transformants on SD and SGal plates. Glc Gal +vector +HA-MYBBP1A +vector +HA-MYBBP1A BY4742 +YEplac195 BY4742 +YEplac195-RPL27A 111 RESULTS As a result, no effect on growth was observed in the control strain, but a nega�ve effect upon the expression of MYBBP1A could be s�ll detected in this background (Figure 61). This result suggests that the nega�ve effect observed upon the expression of MYBBP1A is not related to Bas1, as the nega�ve effect on growth is apparently similar in strains with or without Bas1. Figure 61. The absence of Bas1 does not suppress the nega�ve growth defect caused by MYBBP1A expression. The bas1 null mutant was transformed with the empty pAS24 (vector) or pAS24-HA-MYBBP1A plasmid. Cells were spoted in serial dilu�on steps onto SD-Leu-Ura (Glc) and SGal-Leu-Ura (Gal) plates which were incubated at 30 °C for 4 days. Note that the expression of HA-MYBBP1A also causes a growth defect in the bas1 null mutant. Glc Gal bas1∆ +vector +HA-MYBBP1A 118 DISCUSSION were simultaneously overexpressed, the synthesis of the 60S r-subunit was slightly recovered (Figures 32 and 33). This suggests that, upon suppression, the func�onal competence of the ribosomes produced is likely improved, rather than an improvement in the 60S r-subunit produc�on. There are other published cases where mul�copy suppressors can rescue the defects on growth caused by muta�ons in RAFs without improving the corresponding defect in ribosome biogenesis. Examples of this are the suppression of the growth phenotype of ebp2 mutants by overexpression of eL36A/B rproteins, which produces only a par�ally restaura�on of 60S r-subunits synthesis [245], or the suppression of the growth phenotype of a puf6 mutant by the overexpression of eL43 r-protein without an improvement in the 60S r-subunit produc�on [235]. In a study performed by the group of Dr. Pérez-Fernández, it was determined by CRAC (crosslinking and analysis of cDNA) that Pol5 interacts with the domain III of 25S rRNA [155], where the uL23 and eL27A r-proteins also associate [11, 99, 111, 114]. This suggests that Pol5 would be involved in the correct assembly of these r-proteins, allowing the compac�on of the domain III and the recruitment of other r-proteins and factors necessary for ITS2 processing at site C2 [121, 243, 246-249]. This hypothesis is supported by the fact that overexpression of uL23 and eL27A reduces the growth defect of the pol5-1 mutant, and by the fact that upon deple�on of Pol5, the r-proteins uL2, eL19, eL22, uL23, eL27, eL34 and eL43 are present at lower levels in pre-60S r-par�cles isolated via the RAFs Rlp7 and Noc2 and analyzed by iTRAQ® [155]. These r-proteins also associate with the domain III of 25S rRNA and are essen�als for the forma�on of the PET [11, 99, 111, 114], fact that is coupled to the ITS2 processing [109]. Moreover, it has been described that deple�on of uL23 causes a reduc�on in the incorpora�on of other rproteins into pre-60S r-par�cles [248], thus, the deple�on of Pol5 could affect the assembly of uL23 and consequently the subsequent assembly of the rest of the indicated r-proteins, which ul�mately results in the failure of processing at site C2. We favor the following model: Pol5 is a B-factor that associates with pre-60S rpar�cles when the 27SA pre-rRNA is formed, and then facilitates the assembly of the uL23 and eL27A r-proteins into domain III of 25S rRNA. This allows the recruitment of other r-proteins and factors and the structural rearrangements necessary for cleavage at site C2. At this point, when the 27SB pre-rRNA is processed to generate the 7S and 25.5S 119 DISCUSSION pre-rRNAs, Pol5 is released. Upon loss-of-func�on of Pol5, uL23 and eL27A cannot properly assemble and none of the subsequent events can take place, resul�ng in the accumula�on of pre-60S r-par�cles containing 27SB pre-rRNA, the degrada�on of these par�cles and therefore the shortage of mature 60S r-subunits (Figure 62). Figure 62. Model of associa�on of Pol5 with pre-ribosomal par�cles and the consequences of its absence. (A) Pol5 associa�on in wild-type condi�ons. Pol5 associates with pre-60S r-par�cles during 27SA pre-rRNA forma�on and remains associated un�l 27SB pre-rRNA is processed to 7S and 25.5S pre-rRNAs. (B) Accumula�on of pre-60S r-par�cles upon deple�on of or muta�on in Pol5. Upon loss-of-func�on of Pol5, the processing of 27SB pre-rRNAs is impaired and pre-60S r-par�cles containing these pre-rRNAs transiently accumulate prior to turnover, preven�ng the forma�on of downstream par�cles and resul�ng in a shortage of 60S r-subunits. In the second part of this work, we have tried to func�onally separate the puta�ve dual role of Pol5 in the synthesis of the 60S r-subunit and in the recycling of the t-Utps factors described by the group of Dr. Pérez-Fernández. The t-Utps factors associate with 5’ETS and are essen�al for the forma�on of the SSU processome [50, 76, 78, 79, 81, 152]. The results obtained by this group indicate that upon deple�on of Pol5, these factors remain associated to 5’ETS and are not recycled back to form new pre-ribosomal par�cles, which should have an impact on the 40S r-subunit synthesis [155]. The group of Dr. Pérez-Fernández has speculated that this may indicate that Pol5 is one of the few factors involved in the synthesis of both r-subunits. This par�cular set of factors could play a role in maintaining the needed stoichiometry between the mature r-subunits. Currently, the only factors described to perform a dual func�on during yeast ribosome 120 DISCUSSION synthesis are Rrp12 [94], Prp43 [250-252], Has1 [102, 253], and Rrp5 [254]. Rrp12 par�cipates in the export of both pre-ribosomal par�cles to the cytoplasm [94]. Prp43 is involved in remodeling events that allow the access of Nob1 to the site D in 20S prerRNA [60] and in the associa�on with and dissocia�on from pre-60S r-par�cles of some snoRNPs [255]. Has1 par�cipates in the release of some snoRNPs from pre-40S rpar�cles [256, 257] but it is also an A3-factor required for the processing of ITS1 within pre-60S r-par�cles [107]. In the case of Rrp5, this factor is essen�al for processing at sites A0, A1, A2 and A3; its deple�on causes a reduc�on in the produc�on of the mature 5.8S and 18S rRNAs [254]. An interes�ng feature of this factor is that it can be separated into two func�onal domains: the N-terminal domain, which par�cipates in the synthesis of the mature 5.8S rRNAs and associates near site A3 in pre-ribosomal par�cles, and the C- terminal domain, which is involved in the synthesis of the mature 18S rRNA and associates near site A2 in pre-ribosomal par�cles [98, 258]. The N-terminal domain of Rrp5 ac�vates the endonuclease RNase MRP, which performs the cleavage at site A3, while the C-terminal domain of Rrp5 seems to mediate the associa�on of some snoRNPs necessary for the cleavages at sites A0, A1 and A2 [98]. It has also been described that Rrp5 couples the cleavage at site A2 to the transcrip�on of 25S rDNA, establishing a checkpoint between matura�on of pre-40S and pre-60S r-par�cles [259]. Since each Rrp5 func�on is associated with a specific protein domain, we analyzed whether a similar scenario could be iden�fied for Pol5. Hence, we studied whether the role of Pol5 in the 60S r-subunit synthesis and its role in the t-Utps recycling are associated with different protein domains. To do so, we analyzed different trunca�on mutants lacking the N- or the C-terminal domains of Pol5. Except for the last α-helix (last 29 amino acids), the complete Pol5 protein is required for growth (Figure 37A) and for 60S r-subunit biogenesis, as shown by the polysome profiles of the different trunca�on mutants (Figure 38). By quan�fying total rsubunits, we assessed possible quan�ta�ve differences between mutants lacking the N- or the C-terminal domains. We expected that if the 40S r-subunit synthesis was less affected in some trunca�ons, we might detect differences in the 60S/40S ra�o compared to the nega�ve and posi�ve controls. However, regardless of which domain the mutant lacked, we found ra�os very similar to the nega�ve control in all cases (Figure 39). This 121 DISCUSSION suggests that Pol5, in contrast to Rrp5, is not a modular protein containing independent func�onal domains. Moreover, as we have shown that almost the en�re Pol5 protein is required for 60S r-subunit synthesis, its possible role in the recycling of t-Utps, and thus in 40S r-subunit synthesis, could be masked in all cases by the 60S r-subunit defect. Something similar has been observed upon deple�on of or muta�on in Prp43 and Has1, where polysome profiles showed a deficit in only one of the r-subunits, being difficult to detect the effects on the other r-subunit by this technique [252, 257]. The subcellular localiza�on of the analyzed truncated mutants indicates that the sequences cri�cal for an efficient nucleolar import of Pol5 are located in its N-terminal domain (Figure 41). Interes�ngly, MYBBP1A, the human ortholog of Pol5, harbors these signals in its C-terminal domain [173]. In the mutants lacking the N-terminal domain, we could s�ll detect some nucleolar localiza�on, sugges�ng that other import mechanisms may be involved in the nucleolar localiza�on of Pol5. In S. pombe, it has been described that Rrp14 facilitates the import of Pol5 to the nucleolus, which could be also happening in S. cerevisiae as the RAFs Rrp14 is apparently well conserved [150]. The N-terminal domain of Pol5 is also responsible for its associa�on with preribosomal par�cles, as indicated by the enrichment on the levels of the eL14 r-protein, and the RAFs Nop7 and Has1 in our affinity purifica�on experiments of the mutant harboring the N-terminal domain, while this is not happening in the mutant lacking this domain (Figure 42). It has been well demonstrated that the separate expression of the two func�onal domains of Rrp5 allows the recons�tu�on of a full protein and the recovery of the wildtype phenotype [258]. Similar results have been obtained for Pwp2, a member of the UTP-B complex [260]. In both cases, the expression of two separate domains allows the recons�tu�on in trans of a fully func�onal protein. Our results, despite Pol5 seems not to be a modular protein from the func�onal point of view, indicate that it is also possible to recons�tute a complete func�onal protein at the levels of growth and 60S r-subunit synthesis from two separated truncated parts of Pol5 (Figures 43B and 44). However, the presence of an overlap of 21 amino acids avoids this trans-complementa�on, which could be an indica�on of a physical interac�on between the two regions. Unfortunately, we could not detect this interac�on with the affinity purifica�on procedure used in this 122 DISCUSSION work, the GFP-Trap procedure (Figure 46). This could be due to the harsh physicochemical constraints imposed by the experiment, which cause the loss of the trunca�on that it is not GFP-tagged during the precipita�on as we speculate that the interac�on with the other trunca�on is not strong enough to resist the purifica�on procedure. Surprisingly, we did not observe an improvement in the nucleolar localiza�on of the mutant lacking the N-terminal domain a�er the trans-complementa�on (Figure 45). It is possible that the interac�on between the two Pol5 fragments takes places in the nucleolus, and that the small amount of the trunca�on mutant lacking the N-terminal domain which is imported into the nucleolus is sufficient to recover some amount of complete Pol5 protein, which is enough to restore the growth and the 60S r-subunits synthesis. Thus, we cannot exclude the possible role of Pol5 in 40S r-subunit synthesis, but with the approaches used in this work we could not detect it, nor to associate this role to a specific domain of Pol5. Furthermore, according to the interpreta�on of all our analyses, the N-terminal domain of Pol5 seems to be responsible for the nucleolar localiza�on of this protein and for its associa�on with pre-ribosomal par�cles. In the last part of this work, we have analyzed the effects of the heterologous expression in yeast of MYBBP1A, the puta�ve human ortholog of Pol5 [149]. MYBBP1A is well characterized as a tumor suppressor, but its role in ribosome biogenesis has been poorly characterized. MYBBP1A is nucleolar as Pol5 [160], and it has been described that knockdown of MYBBP1A causes effects on pre-rRNA processing similar to those caused by deple�on of Pol5 in yeast: accumula�on of the equivalent human precursors of 35S and 27SB pre-rRNA (47S and 32S pre-rRNAs, respec�vely) [156]. MYBBP1A interacts with several RAFs and r-proteins from both r-subunits, and among them have been found some of the r-proteins that we have described in yeast that Pol5 helps to assemble, highligh�ng the presence of uL23 and eL27A r-proteins [157]. Complementa�on between human and yeast orthologs genes has been described previously [261-265]. Some yeast cytoskeletal genes could be replaced by their human orthologs [261]; some genes related to different cellular processes such as 123 DISCUSSION DNA repair or replica�on, metabolism, RNA processing, and transla�on, among others, have also been shown to be complemented by their human orthologs [262]. When there are mul�ple human orthologs for a single yeast gene, there are cases where all of them can complement the yeast gene, but some�mes only some of them can do it [264]. Complementa�on between human and yeast orthologs has also been used to study disease-associated variants of human genes [263]. Given the similari�es between Pol5 and MYBBP1A, we expected that a complementa�on between them would be possible. Unfortunately, heterologous expression of the human protein does not complement the absence of Pol5 in our experimental condi�ons. Interes�ngly, its expression even has a dominant nega�ve effect on growth (Figures 47A, 48A and 49). Under certain stress condi�ons, MYBBP1A undergoes a proteoly�c process that generates the p67 and p140 isoforms, which translocate to the nucleoplasm a�er losing their nucleolar localiza�on signals [160, 172]. The p67 isoform is cytoplasmic when ar�ficially expressed because it has lost all the signals that allowed its nucle(ol)ar localiza�on [173]. When processed under biological condi�ons, it is possible that remains nuclear by interac�ng with different nuclear factors [172, 173]. This isoform seems to be more related to interac�ons with other factors such as c-Myb than to a role in ribosome biogenesis, since its overexpression has apparently no effect on this process, in contrast to the full-length isoform [157, 160]. In the case of the p140 isoform, the puta�ve proteoly�c site is assigned in a cluster of basic amino acids that are highly conserved and that appear to be responsible, at least in part, for the nucleolar and nuclear localiza�on of MYBBP1A [172, 173]. This isoform should be par�ally nuclear when ar�ficially expressed because it s�ll harbors one of the basic amino acid repeats considered responsible for this localiza�on [173]. Taking into considera�on these premises, we decided to ar�ficially generate a p140 isoform using the human sequence and check the effect of its expression on yeast. The nega�ve effect detected upon expression of full-length MYBBP1A could not be reproduced with this isoform, nor was any complementa�on detected (Figures 48A and 49). The full-length MYBBP1A is nucleolar in yeast, but in our hands the p140 isoform is cytoplasmic (Figure 50). This may explain the lack of a nega�ve effect and complementa�on with this isoform and may indicate that at least two of the basic amino 124 DISCUSSION acid repeats present in the C-terminal domain of MYBBP1A are necessary for its nucle(ol)ar localiza�on, since an ar�ficially expressed isoform harboring two of them is par�ally nuclear in human cell lines [173]. MYBBP1A has been described to associate with pre-ribosomal par�cles in human cell lines [157], and we could detect a similar situa�on in yeast. The affinity purifica�on experiments of Nop7-GFP containing complexes showed the associa�on of MYBBP1A with these complexes, and presumably with pre-60S r-par�cles (Figure 53). The localiza�on of MYBBP1A in the frac�ons corresponding to r-par�cles in our low Mg2+ run-off sucrose gradient ultracentrifuga�on experiments also supports this idea (Figure 54). Interes�ngly, the sedimenta�on distribu�on of MYBBP1A in sucrose gradients seems to be very similar to that of Pol5 (Figure 22). The nega�ve effect on growth observed with the expression of MYBBP1A is caused by a defect in the synthesis of 60S r-subunits (Figures 51 and 52), which in turn is due to altera�ons in pre-rRNA processing. Northern blot experiments showed that MYBBP1A expression par�cularly affected the processing of precursors for the 60S rsubunit (Figure 55), with a decrease in the levels of mature 25S rRNA and a pronounced increase in the levels of 27SB pre-rRNAs, and a more moderate increase in the levels of 27SA2, 35S and 32S pre-rRNAs. The levels of 7S pre-rRNAs were also nega�vely affected by MYBBP1A expression. Interes�ngly, 20S pre-rRNA levels were also increased, sugges�ng a possible effect on the synthesis of 18S rRNA, but the levels of this mature rRNA were not affected by MYBBP1A expression. In line with the defects observed in the 60S r-subunit synthesis upon MYBBP1A expression, a slight reten�on of pre-60S rpar�cles containing the uL23 r-protein was detected (Figure 56), which, as we men�oned above, could be associated with a quality control mechanism that avoid defec�ve pre- 60S par�cles from acquiring export competence [217]. In human cell lines, it has been described that overexpression of MYBBP1A leads to altera�ons in pre-rRNA processing, with an increase in 47S, 45S, 41S, 30S and 21S prerRNAs [157]. The 47S, 45S and 21S pre-rRNAs are the human equivalents of 35S, 32S and 20S pre-rRNAs, respec�vely, whose accumula�on is occurring in yeast upon MYBBP1A expression, as this Thesis work has revealed. However, no decrease in 28S rRNA (25S rRNA in yeast) has been described, and the other precursors of this rRNA have not been 125 DISCUSSION studied so far [157]. This raises the possibility that the nega�ve effect observed in yeast is due to the fact that MYBBP1A is overexpressed under the control of the GAL promoter, and the effects that we observed are the same that have been detected in human cell lines upon overexpression of MYBBP1A, although it is not clear how this occurs. However, the overexpression of Pol5 with the same GAL promoter does not have any nega�ve effect, sugges�ng that the effect observed upon the overexpression of MYBBP1A may have a different explana�on in this case. We have studied the possibility that the nega�ve effect caused by overexpression of MYBBP1A was directly related to a disrup�on in the normal func�on of Pol5. We have analyzed whether MYBBP1A replaces Pol5 in pre-ribosomal par�cles, preven�ng its associa�on and forming a non-produc�ve pre-60S r-par�cle that cannot mature further. Our experiments show that the expression of MYBBP1A does not change the subcellular localiza�on of Pol5, which is s�ll located in the nucleolus (Figure 57). In addi�on, Pol5 is s�ll able to associate with pre-ribosomal par�cles upon MYBBP1A expression, as shown by the posi�oning of the Pol5-GFP protein in the frac�ons corresponding to the rpar�cles upon MYBBP1A expression (Figure 58B), and by the increase in the levels of the RAFs Has1 and Nop7 and the uL14 r-protein in affinity purifica�on experiments, although at slight lower levels than the control without MYBBP1A expression (Figure 59). Interes�ngly, with this experiment, we were not able to detect interac�on between Pol5 and MYBBP1A, which suggests that the nega�ve effect of the later is probably not related to the kidnapping of Pol5 by MYBBP1A, and therefore, the effects that we have observed are not consequence of a Pol5 deple�on. Moreover, the overexpression of eL27A does not suppress or alleviate the nega�ve effect caused by MYBBP1A expression, even when the overexpression of this r-protein suppresses the growth defects of the mutants pol5-1 and pol5-3 (Figure 60). As MYBBP1A was first described as a nega�ve regulator of the proto-oncogen c- Myb, we have also discarded the possibility that the nega�ve effect caused by the overexpression of MYBBP1A was due to its interac�on with the yeast ortholog of c-Myb, Bas1 [165]. The dele�on of this gene does not improve the growth defect observed upon MYBBP1A expression, which suggests that the nega�ve effect observed it is not related to the interac�on between MYBBP1A and Bas1 (Figure 61). 126 DISCUSSION With these results, we cannot sa�sfactorily explain the causes of the nega�ve effect observed upon MYBBP1A expression in yeast. We have discarded the possibility that MYBBP1A kidnaps Pol5 or replaces it in pre-ribosomal par�cles, since the subcellular localiza�on of Pol5 does not change and it is s�ll able to associate with preribosomal par�cles upon MYBBP1A expression. Another possibility is that MYBBP1A associates with pre-ribosomal par�cles at the same �me as Pol5, but in a different localiza�on inside these par�cles, and somehow interferes with the normal processing at site C2, causing the accumula�on of the 27SB pre-rRNA and the shortage of 60S rsubunits. This hypothesis is consistent with our results showing that Pol5 s�ll associates with Nop7-containing pre-60S r-par�cles upon expression of MYBBP1A, and that the human protein is present in Nop7-GFP containing complexes. However, the absence of MYBBP1A enrichment in the Pol5-GFP affinity purifica�on experiment does not support this idea. Our third hypothesis is based on recent findings that indicate that Pol5 dimerizes, probably through its C-terminal domain (Dr. Pérez-Fernández, pers. comm.). MYBBP1A has also been described to dimerize through its C-terminal domain [172, 178]. In both cases, it is not clear whether these proteins carry out their func�ons as dimers or whether these dimers are the result of a mechanism to regulate their func�on. MYBBP1A and Pol5 could therefore form a heterodimer in yeast, probably through their C-terminal domains, which associates with pre-ribosomal par�cles, probably through the N-terminal domain of Pol5, since our results seem to indicate that the N-terminal domain is responsible for the associa�on of Pol5 with pre-60S r-par�cles. However, this dimer is not able to perform the normal func�on of Pol5, which causes the shortage in 60S r-subunit synthesis. This scenario is consistent with our results showing that Pol5 and MYBBP1A associate with pre-ribosomal par�cles, but again, we could not detect any interac�on between MYBBP1A and Pol5 with our affinity purifica�on assays. A possible explana�on for this is that the GFP tag in the C-terminal domain of the Pol5-GFP protein interferes with the associa�on with MYBBP1A, making this interac�on strong enough to cause the nega�ve effect in vivo, but not strong enough to resist the GFP-Trap procedure. This scenario is the most likely, since interac�on between Pol5 and MYBBP1A has been detected in analyses performed by our collaborator under different experimental condi�ons (Dr. Pérez-Fernández, pers. comm.). Repea�ng the affinity purifica�on assays with an N-terminally GFP-tagged version of Pol5 or altering the composi�on of our 127 DISCUSSION purifica�on buffers (e.g. changing the NaCl or MgCl2 concentra�on of the buffers) could perhaps resolve this issue, but, unfortunately, these experiments could not be performed in this work due to �me restric�ons. Further experiments to try to “humanize” the yeast should be performed in the future, such as introducing the cDNA of MYBBP1A into the locus of Pol5 by recombina�on to allow the expression of the human protein under the control of the POL5 promoter. This approach could resolve the ques�on of whether part of the observed nega�ve effects is due to overexpression of MYBBP1A. It has also been described cases where the expression of the human ortholog gene under the control of a cons�tu�ve yeast promoter from a plasmid does not allow complementa�on, but the expression of the same human gene from its yeast ortholog locus does it [261, 262], so with this approach we could also test whether the lack of complementa�on between Pol5 and MYBBP1A is due to the promoter or plasmid used. Unfortunately, these experiments could neither be performed due to �me restric�ons. In summary, in this Thesis work we have iden�fied Pol5 as a novel B-factor required for the synthesis of 60S r-subunits. We have shown that its N-terminal domain is primarily responsible for its nucleolar localiza�on and associa�on with pre-60S rpar�cles. We have also shown that trans-complementa�on of Pol5, as occurs with some other RAFs, is possible by co-expressing two fragments of the protein, allowing recons�tu�on of a full-length variant of the protein. Under the condi�ons used in this study, we cannot detect complementa�on between Pol5 and its human ortholog, MYBBP1A, which also causes a dominant-nega�ve phenotype in yeast associated with a deficit in 60S r-subunit produc�on. 135 BIBLIOGRAPHY 1. Schmeing T.M. & Ramakrishnan V. (2009). What recent ribosome structures have revealed about the mechanism of transla�on. Nature 461: 1234-1242. 2. Davis J.H. & Williamson J.R. (2017). Structure and dynamics of bacterial ribosome biogenesis. Philos Trans R Soc Lond B Biol Sci 372: 20160181. 3. Klinge S., Voigts-Hoffmann F., Leibundgut M. & Ban N. (2012). Atomic structures of the eukaryo�c ribosome. Trends Biochem Sci 37: 189-198. 4. Greber B.J. & Ban N. (2016). Structure and func�on of the mitochondrial ribosome. Annu Rev Biochem 85: 103-132. 5. Bieri P., Leibundgut M., Saurer M., Boehringer D. & Ban N. (2017). The complete structure of the chloroplast 70S ribosome in complex with transla�on factor pY. EMBO J 36: 475-486. 6. Steitz T.A. (2008). A structural understanding of the dynamic ribosome machine. Nat Rev Mol Cell Biol 9: 242-253. 7. Green R. & Noller H.F. (1997). Ribosomes and transla�on. Annu Rev Biochem 66: 679- 716. 8. Greber B.J. (2016). Mechanis�c insight into eukaryo�c 60S ribosomal subunit biogenesis by cryo-electron microscopy. RNA 22: 1643-1662. 9. Rabl J., Leibundgut M., Ataide S.F., Haag A. & Ban N. (2011). Crystal structure of the eukaryo�c 40S ribosomal subunit in complex with ini�a�on factor 1. Science 331: 730- 736. 10. Ben-Shem A., Jenner L., Yusupova G. & Yusupov M. (2010). Crystal structure of the eukaryo�c ribosome. Science 330: 1203-1209. 11. Ben-Shem A., Garreau de Loubresse N., Melnikov S., Jenner L., Yusupova G. & Yusupov M. (2011). The structure of the eukaryo�c ribosome at 3.0 Å resolu�on. Science 334: 1524-1529. 12. Wilson D.N. & Doudna Cate J.H. (2012). The structure and func�on of the eukaryo�c ribosome. Cold Spring Harb Perspect Biol 4: a011536. 13. Steitz T.A. & Moore P.B. (2003). RNA, the first macromolecular catalyst: the ribosome is a ribozyme. Trends Biochem Sci 28: 411-418. 14. Melnikov S., Ben-Shem A., Garreau de Loubresse N., Jenner L., Yusupova G. & Yusupov M. (2012). One core, two shells: bacterial and eukaryo�c ribosomes. Nat Struct Mol Biol 19: 560-567. 15. Armache J.P., Jarasch A., Anger A.M., Villa E., Becker T., Bhushan S., Jossinet F., Habeck M., Dindar G., Franckenberg S., et al. (2010). Cryo-EM structure and rRNA model of a transla�ng eukaryo�c 80S ribosome at 5.5-Å resolu�on. Proc Natl Acad Sci USA 107: 19748-19753. 136 BIBLIOGRAPHY 16. Melnikov S., Manakongtreecheep K. & Soll D. (2018). Revising the structural diversity of ribosomal proteins across the three domains of life. Mol Biol Evol 35: 1588-1598. 17. Woolford J.L., Jr. & Baserga S.J. (2013). Ribosome biogenesis in the yeast Saccharomyces cerevisiae. Genetics 195: 643-681. 18. Pillet B., Miterer V., Kressler D. & Pertschy B. (2017). Hold on to your friends: Dedicated chaperones of ribosomal proteins: dedicated chaperones mediate the safe transfer of ribosomal proteins to their site of pre-ribosome incorpora�on. Bioessays 39: 1-12. 19. Kressler D., Bange G., Ogawa Y., Stjepanovic G., Bradatsch B., Prate D., Amlacher S., Strauss D., Yoneda Y., Katahira J., et al. (2012). Synchronizing nuclear import of ribosomal proteins with ribosome assembly. Science 338: 666-671. 20. Calviño F.R., Kharde S., Ori A., Hendricks A., Wild K., Kressler D., Bange G., Hurt E., Beck M. & Sinning I. (2015). Sympor�n 1 chaperones 5S RNP assembly during ribosome biogenesis by occupying an essen�al rRNA-binding site. Nat Commun 6: 6510. 21. Pillet B., Méndez-Godoy A., Murat G., Favre S., Stumpe M., Falquet L. & Kressler D. (2022). Dedicated chaperones coordinate co-transla�onal regula�on of ribosomal protein produc�on with ribosome assembly to preserve proteostasis. eLife 11: e74255. 22. Yueh L.Y., Tseng Y.T., Chu C.Y. & Lo K.Y. (2022). The dedicated chaperones of eL43, Puf6 and Loc1 can also bind RPL43 mRNA and regulate the produc�on of this ribosomal protein. J Biochem 171: 85-96. 23. Sung M.K., Reitsma J.M., Sweredoski M.J., Hess S. & Deshaies R.J. (2016). Ribosomal proteins produced in excess are degraded by the ubiqui�n-proteasome system. Mol Biol Cell 27: 2642-2652. 24. Sung M.K., Porras-Yakushi T.R., Reitsma J.M., Huber F.M., Sweredoski M.J., Hoelz A., Hess S. & Deshaies R.J. (2016). A conserved quality-control pathway that mediates degrada�on of unassembled ribosomal proteins. eLife 5: e19105. 25. Kellis M., Birren B.W. & Lander E.S. (2004). Proof and evolu�onary analysis of ancient genome duplica�on in the yeast Saccharomyces cerevisiae. Nature 428: 617-624. 26. Ghulam M.M., Catala M. & Abou Elela S. (2020). Differen�al expression of duplicated ribosomal protein genes modifies ribosome composi�on in response to stress. Nucleic Acids Res 48: 1954-1968. 27. Simoff I., Moradi H. & Nygard O. (2009). Func�onal characteriza�on of ribosomal protein L15 from Saccharomyces cerevisiae. Curr Genet 55: 111-125. 28. Komili S., Farny N.G., Roth F.P. & Silver P.A. (2007). Func�onal specificity among ribosomal proteins regulates gene expression. Cell 131: 557-571. 137 BIBLIOGRAPHY 29. Bauer J.W., Brandl C., Haubenreisser O., Wimmer B., Weber M., Karl T., Klausegger A., Breitenbach M., Hintner H., von der Haar T., et al. (2013). Specialized yeast ribosomes: a customized tool for selec�ve mRNA transla�on. PLoS One 8: e67609. 30. Malik Ghulam M., Catala M., Reulet G., Scot M.S. & Abou Elela S. (2022). Duplicated ribosomal protein paralogs promote alterna�ve transla�on and drug resistance. Nat Commun 13: 4938. 31. Warner J.R. & McIntosh K.B. (2009). How common are extraribosomal func�ons of ribosomal proteins? Mol Cell 34: 3-11. 32. Lu H., Zhu Y.F., Xiong J., Wang R. & Jia Z. (2015). Poten�al extra-ribosomal func�ons of ribosomal proteins in Saccharomyces cerevisiae. Microbiol Res 177: 28-33. 33. Bursac S., Brdovcak M.C., Dona� G. & Volarevic S. (2014). Ac�va�on of the tumor suppressor p53 upon impairment of ribosome biogenesis. Biochim Biophys Acta 1842: 817-830. 34. Park S.Y., Seo D., Jeon E.H., Park J.Y., Jang B.C., Kim J.I., Im S.S., Lee J.H., Kim S., Cho C.H., et al. (2023). RPL27 contributes to colorectal cancer prolifera�on and stemness via PLK1 signaling. Int J Oncol 63: 93. 35. El Khoury W. & Nasr Z. (2021). Deregula�on of ribosomal proteins in human cancers. Biosci Rep 41: BSR20211577. 36. Ban N., Beckmann R., Cate J.H., Dinman J.D., Dragon F., Ellis S.R., Lafontaine D.L., Lindahl L., Liljas A., Lipton J.M., et al. (2014). A new system for naming ribosomal proteins. Curr Opin Struct Biol 24: 165-169. 37. Warner J.R. (1999). The economics of ribosome biosynthesis in yeast. Trends Biochem Sci 24: 437-440. 38. Dörner K., Ruggeri C., Zemp I. & Kutay U. (2023). Ribosome biogenesis factors-from names to func�ons. EMBO J 42: e112699. 39. Ide S., Miyazaki T., Maki H. & Kobayashi T. (2010). Abundance of ribosomal RNA gene copies maintains genome integrity. Science 327: 693-696. 40. Kobayashi T. (2011). Regula�on of ribosomal RNA gene copy number and its role in modula�ng genome integrity and evolu�onary adaptability in yeast. Cell Mol Life Sci 68: 1395-1403. 41. Nomura M., Nogi Y. & Oakes M. (2004) Transcrip�on of rDNA in the yeast Saccharomyces cerevisiae. Nucleolus,(Olson MOJ, ed.) pp. 128-153. Landes Bioscience/Eurekah.com, Georgetown. 42. de la Cruz J., Gómez-Herreros F., Rodríguez-Galán O., Begley V., Muñoz-Centeno M.C. & Chávez S. (2018). Feedback regula�on of ribosome assembly. Curr Genet 64: 393-404. 138 BIBLIOGRAPHY 43. Gu�errez-San�ago F. & Navarro F. (2023). Transcrip�on by the three RNA polymerases under the control of the TOR signaling pathway in Saccharomyces cerevisiae. Biomolecules 13: 642. 44. Kressler D., Linder P. & de la Cruz J. (1999). Protein trans-ac�ng factors involved in ribosome biogenesis in Saccharomyces cerevisiae. Mol Cell Biol 19: 7897-7912. 45. Kressler D., Hurt E. & Bassler J. (2010). Driving ribosome assembly. Biochim Biophys Acta 1803: 673-683. 46. Khreiss A., Bohnsack K.E. & Bohnsack M.T. (2023). Molecular func�ons of RNA helicases during ribosomal subunit assembly. Biol Chem 404: 781-789. 47. Watkins N.J. & Bohnsack M.T. (2012). The box C/D and H/ACA snoRNPs: key players in the modifica�on, processing and the dynamic folding of ribosomal RNA. Wiley Interdiscip Rev RNA 3: 397-414. 48. Henras A.K., Plisson-Chastang C., Humbert O., Romeo Y. & Henry Y. (2017). Synthesis, func�on, and heterogeneity of snoRNA-guided postranscrip�onal nucleoside modifica�ons in eukaryo�c ribosomal RNAs. Enzymes 41: 169-213. 49. Ojha S., Malla S. & Lyons S.M. (2020). snoRNPs: func�ons in ribosome biogenesis. Biomolecules 10: 783. 50. Pérez-Fernández J., Román A., de las Rivas J., Bustelo X.R. & Dosil M. (2007). The 90S preribosome is a mul�modular structure that is assembled through a hierarchical mechanism. Mol Cell Biol 27: 5414-5429. 51. Phipps K.R., Charete J.M. & Baserga S.J. (2011). The small subunit processome in ribosome biogenesis-progress and prospects. Wiley Interdiscip Rev RNA 2: 1-21. 52. Kressler D., Hurt E. & Bassler J. (2017). A puzzle of life: cra�ing ribosomal subunits. Trends Biochem Sci 42: 640-654. 53. Koš M. & Tollervey D. (2010). Yeast pre-rRNA processing and modifica�on occur cotranscrip�onally. Mol Cell 37: 809-820. 54. Wells G.R., Weichmann F., Colvin D., Sloan K.E., Kudla G., Tollervey D., Watkins N.J. & Schneider C. (2016). The PIN domain endonuclease Utp24 cleaves pre-ribosomal RNA at two coupled sites in yeast and humans. Nucleic Acids Res 44: 5399-5409. 55. Fernández-Pevida A., Kressler D. & de la Cruz J. (2015). Processing of preribosomal RNA in Saccharomyces cerevisiae. Wiley Interdiscip Rev RNA 6: 191-209. 56. Kufel J., Dichtl B. & Tollervey D. (1999). Yeast Rnt1p is required for cleavage of the pre-ribosomal RNA in the 3' ETS but not the 5' ETS. RNA 5: 909-917. 57. El Hage A., Koper M., Kufel J. & Tollervey D. (2008). Efficient termina�on of transcrip�on by RNA polymerase I requires the 5' exonuclease Rat1 in yeast. Genes Dev 22: 1069-1081. 139 BIBLIOGRAPHY 58. Allmang C. & Tollervey D. (1998). The role of the 3' external transcribed spacer in yeast pre-rRNA processing. J Mol Biol 278: 67-78. 59. Fa�ca A., Oeffinger M., Dlakic M. & Tollervey D. (2003). Nob1p is required for cleavage of the 3' end of 18S rRNA. Mol Cell Biol 23: 1798-1807. 60. Pertschy B., Schneider C., Gnadig M., Schafer T., Tollervey D. & Hurt E. (2009). RNA helicase Prp43 and its co-factor Pfa1 promote 20 to 18 S rRNA processing catalyzed by the endonuclease Nob1. J Biol Chem 284: 35079-35091. 61. Strunk B.S., Novak M.N., Young C.L. & Karbstein K. (2012). A transla�on-like cycle is a quality control checkpoint for maturing 40S ribosome subunits. Cell 150: 111-121. 62. Lebaron S., Schneider C., van Nues R.W., Swiatkowska A., Walsh D., Botcher B., Granneman S., Watkins N.J. & Tollervey D. (2012). Proofreading of pre-40S ribosome matura�on by a transla�on ini�a�on factor and 60S subunits. Nat Struct Mol Biol 19: 744-753. 63. Chu S., Archer R.H., Zengel J.M. & Lindahl L. (1994). The RNA of RNase MRP is required for normal processing of ribosomal RNA. Proc Natl Acad Sci USA 91: 659-663. 64. Axt K., French S.L., Beyer A.L. & Tollervey D. (2014). Kine�c analysis demonstrates a requirement for the Rat1 exonuclease in cotranscrip�onal pre-rRNA cleavage. PLoS One 9: e85703. 65. Oeffinger M., Zenklusen D., Ferguson A., Wei K.E., El Hage A., Tollervey D., Chait B.T., Singer R.H. & Rout M.P. (2009). Rrp17p is a eukaryo�c exonuclease required for 5' end processing of Pre-60S ribosomal RNA. Mol Cell 36: 768-781. 66. Johnson A.W. (1997). Rat1p and Xrn1p are func�onally interchangeable exoribonucleases that are restricted to and required in the nucleus and cytoplasm, respec�vely. Mol Cell Biol 17: 6122-6130. 67. Gasse L., Flemming D. & Hurt E. (2015). Coordinated ribosomal ITS2 RNA processing by the Las1 complex integra�ng endonuclease, polynucleo�de kinase, and exonuclease ac�vi�es. Mol Cell 60: 808-815. 68. Pillon M.C., Sobhany M., Borgnia M.J., Williams J.G. & Stanley R.E. (2017). Grc3 programs the essen�al endoribonuclease Las1 for specific RNA cleavage. Proc Natl Acad Sci U S A 114: E5530-E5538. 69. Faber A.W., Van Dijk M., Raué H.A. & Vos J.C. (2002). Ngl2p is a Ccr4p-like RNA nuclease essen�al for the final step in 3'- end processing of 5.8S rRNA in Saccharomyces cerevisiae. RNA 8: 1095-1101. 70. Thomson E. & Tollervey D. (2010). The final step in 5.8S rRNA processing is cytoplasmic in Saccharomyces cerevisiae. Mol Cell Biol 30: 976-984. 71. Geerlings T.H., Vos J.C. & Raué H.A. (2000). The final step in the forma�on of 25S rRNA in Saccharomyces cerevisiae is performed by 5'-3' exonucleases. RNA 6: 1698-1703. 140 BIBLIOGRAPHY 72. Ciganda M. & Williams N. (2011). Eukaryo�c 5S rRNA biogenesis. Wiley Interdiscip Rev RNA 2: 523-533. 73. Kos-Braun I.C., Jung I. & Koš M. (2017). Tor1 and CK2 kinases control a switch between alterna�ve ribosome biogenesis pathways in a growth-dependent manner. PLoS Biol 15: e2000245. 74. Grandi P., Rybin V., Bassler J., Pe�alski E., Strauss D., Marzioch M., Schäfer T., Kuster B., Tschochner H., Tollervey D., et al. (2002). 90S pre-ribosomes include the 35S prerRNA, the U3 snoRNP, and 40S subunit processing factors but predominantly lack 60S synthesis factors. Mol Cell 10: 105-115. 75. Barandun J., Hunziker M. & Klinge S. (2018). Assembly and structure of the SSU processome-a nucleolar precursor of the small ribosomal subunit. Curr Opin Struct Biol 49: 85-93. 76. Barandun J., Chaker-Margot M., Hunziker M., Molloy K.R., Chait B.T. & Klinge S. (2017). The complete structure of the small-subunit processome. Nat Struct Mol Biol 24: 944-953. 77. Hunziker M., Barandun J., Pe�alski E., Tan D., Delan-Forino C., Molloy K.R., Kim K.H., Dunn-Davies H., Shi Y., Chaker-Margot M., et al. (2016). UtpA and UtpB chaperone nascent pre-ribosomal RNA and U3 snoRNA to ini�ate eukaryo�c ribosome assembly. Nat Commun 7: 12090. 78. Chaker-Margot M., Barandun J., Hunziker M. & Klinge S. (2017). Architecture of the yeast small subunit processome. Science 355: eaal1880. 79. Chaker-Margot M. (2018). Assembly of the small ribosomal subunit in yeast: mechanism and regula�on. RNA 24: 881-891. 80. Chaker-Margot M., Hunziker M., Barandun J., Dill B.D. & Klinge S. (2015). Stagespecific assembly events of the 6-MDa small-subunit processome ini�ate eukaryo�c ribosome biogenesis. Nat Struct Mol Biol 22: 920-923. 81. Cheng J., Lau B., La Venuta G., Ameismeier M., Berninghausen O., Hurt E. & Beckmann R. (2020). 90S pre-ribosome transforma�on into the primordial 40S subunit. Science 369: 1470-1476. 82. Sardana R., Liu X., Granneman S., Zhu J., Gill M., Papoulas O., Marcote E.M., Tollervey D., Correll C.C. & Johnson A.W. (2015). The DEAH-box helicase Dhr1 dissociates U3 from the pre-rRNA to promote forma�on of the central pseudoknot. PLoS Biol 13: e1002083. 83. Lau B., Cheng J., Flemming D., La Venuta G., Berninghausen O., Beckmann R. & Hurt E. (2021). Structure of the maturing 90S pre-ribosome in associa�on with the RNA exosome. Mol Cell 81: 293-303 e294. 141 BIBLIOGRAPHY 84. Cheng J., Lau B., Thoms M., Ameismeier M., Berninghausen O., Hurt E. & Beckmann R. (2022). The nucleoplasmic phase of pre-40S forma�on prior to nuclear export. Nucleic Acids Res 50: 11924-11937. 85. Black J.J. & Jonhson A.W. (2022). Release of the ribosome biogenesis factor Bud23 from small subunit precursors in yeast. RNA 28: 371-389. 86. Heuer A., Thomson E., Schmidt C., Berninghausen O., Becker T., Hurt E. & Beckmann R. (2017). Cryo-EM structure of a late pre-40S ribosomal subunit from Saccharomyces cerevisiae. eLife 6: e30189. 87. Scaiola A., Pena C., Weisser M., Bohringer D., Leibundgut M., Klingauf-Nerurkar P., Gerhardy S., Panse V.G. & Ban N. (2018). Structure of a eukaryo�c cytoplasmic pre-40S ribosomal subunit. EMBO J 37: e98499. 88. Zemp I. & Kutay U. (2007). Nuclear export and cytoplasmic matura�on of ribosomal subunits. FEBS Lett 581: 2783-2793. 89. Vanrobays E., Leplus A., Osheim Y.N., Beyer A.L., Wacheul L. & Lafontaine D.L. (2008). TOR regulates the subcellular distribu�on of DIM2, a KH domain protein required for cotranscrip�onal ribosome assembly and pre-40S ribosome export. RNA 14: 2061-2073. 90. Vanrobays E., Gelugne J.P., Gleizes P.E. & Caizergues-Ferrer M. (2003). Late cytoplasmic matura�on of the small ribosomal subunit requires RIO proteins in Saccharomyces cerevisiae. Mol Cell Biol 23: 2083-2095. 91. Merwin J.R., Bogar L.B., Poggi S.B., Fitch R.M., Johnson A.W. & Lycan D.E. (2014). Gene�c analysis of the ribosome biogenesis factor Ltv1 of Saccharomyces cerevisiae. Genetics 198: 1071-1085. 92. Fischer U., Schauble N., Schütz S., Altvater M., Chang Y., Faza M.B. & Panse V.G. (2015). A non-canonical mechanism for Crm1-export cargo complex assembly. eLife 4: e05745. 93. Faza M.B., Chang Y., Occhipin� L., Kemmler S. & Panse V.G. (2012). Role of Mex67- Mtr2 in the nuclear export of 40S pre-ribosomes. PLoS Genet 8: e1002915. 94. Oeffinger M., Dlakic M. & Tollervey D. (2004). A pre-ribosome-associated HEAT- repeat protein is required for export of both ribosomal subunits. Genes Dev 18: 196-209. 95. Turowski T.W., Lebaron S., Zhang E., Peil L., Dudnakova T., Pe�alski E., Granneman S., Rappsilber J. & Tollervey D. (2014). Rio1 mediates ATP-dependent final matura�on of 40S ribosomal subunits. Nucleic Acids Res 42: 12189-12199. 96. Klinge S. & Woolford J.L., Jr. (2019). Ribosome assembly coming into focus. Nat Rev Mol Cell Biol 20: 116-131. 97. Zhou D., Zhu X., Zheng S., Tan D., Dong M.Q. & Ye K. (2019). Cryo-EM structure of an early precursor of large ribosomal subunit reveals a half-assembled intermediate. Protein Cell 10: 120-130. 142 BIBLIOGRAPHY 98. Lebaron S., Segerstolpe A., French S.L., Dudnakova T., de Lima Alves F., Granneman S., Rappsilber J., Beyer A.L., Wieslander L. & Tollervey D. (2013). Rrp5 binding at mul�ple sites coordinates pre-rRNA processing and assembly. Mol Cell 52: 707-719. 99. Kater L., Thoms M., Barrio-García C., Cheng J., Ismail S., Ahmed Y.L., Bange G., Kressler D., Berninghausen O., Sinning I., et al. (2017). Visualizing the assembly pathway of nucleolar pre-60S ribosomes. Cell 171: 1599-1610. 100. Joret C., Capeyrou R., Belhabich-Baumas K., Plisson-Chastang C., Ghandour R., Humbert O., Fribourg S., Leulliot N., Lebaron S., Henras A.K., et al. (2018). The Npa1p complex chaperones the assembly of the earliest eukaryo�c large ribosomal subunit precursor. PLoS Genet 14: e1007597. 101. Rodríguez-Galán O., García-Gómez J.J. & de la Cruz J. (2013). Yeast and human RNA helicases involved in ribosome biogenesis: current status and perspec�ves. Biochim Biophys Acta-Gene Regulatory Mechanisms 1829: 775-790. 102. Brüning L., Hackert P., Mar�n R., Davila Gallesio J., Aquino G.R.R., Urlaub H., Sloan K.E. & Bohnsack M.T. (2018). RNA helicases mediate structural transi�ons and composi�onal changes in pre-ribosomal complexes. Nat Commun 9: 5383. 103. Aquino G.R.R., Krogh N., Hackert P., Mar�n R., Gallesio J.D., van Nues R.W., Schneider C., Watkins N.J., Nielsen H., Bohnsack K.E., et al. (2021). RNA helicasemediated regula�on of snoRNP dynamics on pre-ribosomes and rRNA 2'-O-methyla�on. Nucleic Acids Res 49: 4066-4084. 104. Miterer V. & Pertschy B. (2022). RNA folding and func�ons of RNA helicases in ribosome biogenesis. RNA Biol 19: 781-810. 105. Sahasranaman A., Dembowski J., Strahler J., Andrews P., Maddock J. & Woolford J.L., Jr. (2011). Assembly of Saccharomyces cerevisiae 60S ribosomal subunits: role of factors required for 27S pre-rRNA processing. EMBO J 30: 4020-4032. 106. Shimoji K., Jakovljevic J., Tsuchihashi K., Umeki Y., Wan K., Kawasaki S., Talkish J., Woolford J.L., Jr. & Mizuta K. (2012). Ebp2 and Brx1 func�on coopera�vely in 60S ribosomal subunit assembly in Saccharomyces cerevisiae. Nucleic Acids Res 40: 4574- 4588. 107. Dembowski J.A., Kuo B. & Woolford J.L., Jr. (2013). Has1 regulates consecu�ve matura�on and processing steps for assembly of 60S ribosomal subunits. Nucleic Acids Res 41: 7889-7904. 108. Talkish J., Campbell I.W., Sahasranaman A., Jakovljevic J. & Woolford J.L., Jr. (2014). Ribosome assembly factors Pwp1 and Nop12 are important for folding of 5.8S rRNA during ribosome biogenesis in Saccharomyces cerevisiae. Mol Cell Biol 34: 1863-1877. 109. Konikkat S. & Woolford J.L., Jr. (2017). Principles of 60S ribosomal subunit assembly emerging from recent studies in yeast. Biochem J 474: 195-214. 143 BIBLIOGRAPHY 110. Talkish J., Zhang J., Jakovljevic J., Horsey E.W. & Woolford J.L., Jr. (2012). Hierarchical recruitment into nascent ribosomes of assembly factors required for 27SB pre-rRNA processing in Saccharomyces cerevisiae. Nucleic Acids Res 40: 8646-8661. 111. Wu S., Tutuncuoglu B., Yan K., Brown H., Zhang Y., Tan D., Gamalinda M., Yuan Y., Li Z., Jakovljevic J., et al. (2016). Diverse roles of assembly factors revealed by structures of late nuclear pre-60S ribosomes. Nature 534: 133-137. 112. Kressler D., Roser D., Pertschy B. & Hurt E. (2008). The AAA ATPase Rix7 powers progression of ribosome biogenesis by stripping Nsa1 from pre-60S par�cles. J Cell Biol 181: 935-944. 113. Lo Y.H., Romes E.M., Pillon M.C., Sobhany M. & Stanley R.E. (2017). Structural analysis reveals features of ribosome assembly factor Nsa1/WDR74 important for localiza�on and interac�on with Rix7/NVL2. Structure 25: 762-772. 114. Sanghai Z.A., Miller L., Molloy K.R., Barandun J., Hunziker M., Chaker-Margot M., Wang J., Chait B.T. & Klinge S. (2018). Modular assembly of the nucleolar pre-60S ribosomal subunit. Nature 556: 126-129. 115. Kater L., Miterer V., Thoms M., Cheng J., Berninghausen O., Beckmann R. & Hurt E. (2020). Construc�on of the central protuberance and L1 stalk during 60S subunit biogenesis. Mol Cell 79: 615-628. 116. Thoms M., Ahmed Y.L., Maddi K., Hurt E. & Sinning I. (2016). Concerted removal of the Erb1-Ytm1 complex in ribosome biogenesis relies on an elaborate interface. Nucleic Acids Res 44: 926-939. 117. Bassler J., Kallas M., Pertschy B., Ulbrich C., Thoms M. & Hurt E. (2010). The AAA- ATPase Rea1 drives removal of biogenesis factors during mul�ple stages of 60S ribosome assembly. Mol Cell 38: 712-721. 118. Ahmed Y.L., Thoms M., Miterer V., Sinning I. & Hurt E. (2019). Crystal structures of Rea1-MIDAS bound to its ribosome assembly factor ligands resembling integrin-ligand- type complexes. Nat Commun 10: 3050. 119. Zhang J., Harnpicharnchai P., Jakovljevic J., Tang L., Guo Y., Oeffinger M., Rout M.P., Hiley S.L., Hughes T. & Woolford J.L., Jr. (2007). Assembly factors Rpf2 and Rrs1 recruit 5S rRNA and ribosomal proteins rpL5 and rpL11 into nascent ribosomes. Genes Dev 21: 2580-2592. 120. Madru C., Lebaron S., Blaud M., Delbos L., Pipoli J., Pasmant E., Réty S. & Leulliot N. (2015). Chaperoning 5S RNA assembly. Genes Dev 29: 1432-1446. 121. Biedka S., Wu S., LaPeruta A.J., Gao N. & Woolford J.L., Jr. (2017). Insights into remodeling events during eukaryo�c large ribosomal subunit assembly provided by high resolu�on cryo-EM structures. RNA Biol 14: 1306-1313.