Ribosome biogenesis factors—from names to functions

Abstract

The assembly of ribosomal subunits is a highly orchestrated process that involves a huge cohort of accessory factors. Most eukaryotic ribosome biogenesis factors were first identified by genetic screens and proteomic approaches of pre‐ribosomal particles in Saccharomyces cerevisiae. Later, research on human ribosome synthesis not only demonstrated that the requirement for many of these factors is conserved in evolution, but also revealed the involvement of additional players, reflecting a more complex assembly pathway in mammalian cells. Yet, it remained a challenge for the field to assign a function to many of the identified factors and to reveal their molecular mode of action. Over the past decade, structural, biochemical, and cellular studies have largely filled this gap in knowledge and led to a detailed understanding of the molecular role that many of the players have during the stepwise process of ribosome maturation. Such detailed knowledge of the function of ribosome biogenesis factors will be key to further understand and better treat diseases linked to disturbed ribosome assembly, including ribosomopathies, as well as different types of cancer.

As part of our 2023 RNA review series, Kutay and colleagues provide a comprehensive update on the diverse factors involved in ribosome biogenesis across different species.

Introduction

Ribosomes are at the heart of messenger (mRNA) translation, a key process of gene expression in all organisms. In eukaryotes, ribosomes consist of a small 40S and a large 60S subunit that together form the translation‐competent 80S ribosome. The 40S subunit comprises the 18S ribosomal RNA (rRNA) as well as 33 ribosomal proteins (RPs), whereas the 60S subunit contains three rRNAs, the 25S/28S, 5.8S, and 5S rRNAs, as well as 46 RPs in yeast and 47 RPs in humans (Table 1). During protein synthesis, mRNA is bound by the 40S subunit, which harbors the decoding center designed for pairing of an mRNA codon with a cognate tRNA. The peptidyl transferase center (PTC) in the 60S subunit catalyzes peptide bond formation in the emerging polypeptide chain, which leaves the ribosome through the peptide exit tunnel (PET) of the 60S subunit. As the catalytic PTC is formed by rRNA, the ribosome is classified as a ribozyme (Cech, 2000; Nissen et al, 2000).

Table 1.

Composition of ribosomes from different kingdoms.

	Bacteria	Archaea	Eukaryotes
	E. coli	P. furiosus	S. cerevisiae	H. sapiens
Molecular weight
Ribosome	2.3 MDa	2.6 MDa	3.3 MDa	4.3 MDa
LSU	1.45 MDa	1.7 MDa	2.1 MDa	3.1 MDa
SSU	0.8 MDa	0.9 MDa	1.2 MDa	1.2 MDa
Sedimentation coefficients
Ribosome	70S	70S	80S	80S
LSU	50S	50S	60S	60S
SSU	30S	30S	40S	40S
rRNAs
LSU	23S	23S	25S	28S
	5S	5S	5S	5S
			5.8S	5.8S
SSU	16S	16S	18S	18S
rRNA length (nucleotides)
Total	4,567	4,712	5,475	7,181
LSU (23/25/28S +5S (+5.8S))	2,904 + 121	3,096 + 121	3,396 + 158 + 121	5,034 + 156 + 121
SSU (18S)	1,542	1,495	1,800	1,870
Number of ribosomal proteins
Ribosome	54	69	79	80
LSU	33	42	46	47
SSU	21	27	33	33

Properties of ribosomes in bacteria (E. coli), archaea (P. furiosus), and lower (S. cerevisiae) and higher (H. sapiens) eukaryotes.

Due to their importance in mRNA translation, decoding center, PTC, PET, and translation factor binding sites are the evolutionarily most conserved regions of ribosomes (Klinge et al, 2012; Melnikov et al, 2012). In contrast to the conserved core structure, overall ribosomal composition, size, and complexity vary between the different kingdoms of life (Table 1). The number of rRNAs and their length increased significantly from bacteria and archaea to eukaryotes. Eukaryotic rRNAs display several large rRNA expansion segments with largely unexplored function. The accretion of these expansion segments bears the major contribution to the increase in rRNA length and ribosome size from yeast to vertebrates (Table 1) (Hariharan et al, 2022). Furthermore, eukaryotic ribosomes contain additional RPs, and most RPs of the conserved core harbor extensions and insertions (Spahn et al, 2001; Armache et al, 2010; Ben‐Shem et al, 2011; Klinge et al, 2011; Rabl et al, 2011; Melnikov et al, 2012; Khatter et al, 2015).

Assembly of both ribosomal subunits requires deposition of the numerous RPs on the pre‐rRNAs concomitant with rRNA transcription, modification, folding, and processing, which all occur in a hierarchical, highly orchestrated, and energetically expensive cellular process (Warner, 1999). Although bacterial ribosomal subunits can be reconstituted at elevated temperature in vitro by mixing mature rRNAs and ribosomal proteins, their assembly in vivo is supported by about 50 non‐ribosomal factors (Held et al, 1973; Nierhaus & Dohme, 1974; Kaczanowska & Rydén‐Aulin, 2007; Shajani et al, 2011; Gibbs & Fredrick, 2018). In contrast to prokaryotes, where the assembly process occurs in a single compartment, the cytoplasm, eukaryotic cells are highly compartmentalized and ribosomal subunit maturation starts in the nucleolus, continues in the nucleoplasm and is only finalized in the cytoplasm. The eukaryotic subunit assembly process involves several hundred non‐ribosomal factors, termed ribosome biogenesis factors (RBFs), which function as chaperones and as modification, processing, assembly, and remodeling factors. These RBFs transiently associate with pre‐ribosomal particles, but are not part of mature ribosomes. Along with the growing complexity of ribosome composition throughout evolution, the intricacy of the eukaryotic ribosome biogenesis pathway and the regulatory interplays with other processes also increased.

Eukaryotic ribosome assembly starts with the transcription of a polycistronic pre‐rRNA precursor by RNA polymerase I (RNAPI) in the nucleolus (Fig 1) (Turowski & Tollervey, 2015). Emerging rRNA stretches are quickly bound by some early assembling RPs as well as RBFs, giving rise to a large, 90S‐sized precursor particle in the nucleolus (Klinge & Woolford, 2019; vanden Broeck & Klinge, 2022). A critical endonucleolytic pre‐rRNA cleavage event then leads to the separation of the pre‐40S and pre‐60S particles (Baßler & Hurt, 2019; Bohnsack & Bohnsack, 2019). Both subunits subsequently undergo a number of nucle(ol)ar maturation steps, including the incorporation of the 5S rRNA, transcribed by RNA polymerase III (RNAPIII), into the pre‐60S subunit (Woolford & Baserga, 2013; de la Cruz et al, 2015; Kressler et al, 2017; Chaker‐Margot & Klinge, 2019; Frazier et al, 2021). After nuclear export, final assembly events take place in the cytoplasm, including the release of remaining RBFs and incorporation of missing RPs, giving rise to mature, translationally competent subunits (Kressler et al, 2017; Peña et al, 2017).

Figure 1. Overview of eukaryotic ribosome biogenesis.

In the nucleolus, a rRNA precursor is transcribed by RNAPI and co‐transcriptionally joined by RPs and RBFs, giving rise to a 90S pre‐ribosomal particle. After pre‐rRNA cleavage (site 2 in human, A₂ in yeast), the pre‐40S and pre‐60S particles further mature independently in the nucleolus and the nucleoplasm. After export through the nuclear pore complex, final maturation steps occur in the cytoplasm, yielding 40S and 60S subunits competent for mRNA translation. RNAPIII transcribes the 5S rRNA (nucleolar in human cells, nucleoplasmic in yeast), which joins pre‐60S particles in the nucleolus as part of the 5S RNP complex. RNAPII transcribes mRNAs of RP and RBF genes, which are translated by the 80S ribosome in the cytoplasm and then imported into the nucleus. Structural snapshots of maturing 90S (PDB ID: 6ZQA, 6ZQC), pre‐40S (PDB ID: 6G4W, 6G4S, 6ZQF), and pre‐60S particles (PDB ID: 6EM3, 6C0F, 6ELZ, 3JCT, 5JCS, 6LU8, 6LSR) as well as mature subunits (PDB ID: 6G5H, 3J7P) are shown. RBFs are displayed in orange, Rps in green, Rpl in blue, rRNA in gray. Structures solved in yeast are shown with reduced opacity, currently no structures of the corresponding human maturation stages are available.

Early work in the 1970s already indicated that yeast and human cells employ broadly similar principles for ribosome maturation, which differ substantially from prokaryotic ribosome synthesis (Darnell, 1968; Udem et al, 1971; Warner, 1971; Trapman et al, 1975). Many eukaryotic RBFs were first identified in genetic screens and proteomic approaches using budding yeast as a model organism, yet their function was often unraveled only later by biochemical studies (Hurt et al, 1999; Stage‐Zimmermann et al, 2000; Dragon et al, 2002; Nissan et al, 2002; Saveanu et al, 2003; Schäfer et al, 2003; Woolford & Baserga, 2013). Over time, an increasing number of structural snapshots of ribosomal pre‐particles from various stages of the assembly process has tremendously increased the understanding of RBF functionalities (Barrio‐Garcia et al, 2016; Kornprobst et al, 2016; Wu et al, 2016; Zhang et al, 2016b; Barandun et al, 2017; Cheng et al, 2017; Kater et al, 2017, 2020; Ma et al, 2017; Sun et al, 2017; Sanghai et al, 2018; Scaiola et al, 2018; Chaker‐Margot & Klinge, 2019; Kargas et al, 2019; Klinge & Woolford, 2019; Zhou et al, 2019a, 2019b). In the past 15 years, biochemical, cellular, and structural studies including large‐scale screening and proteomic approaches have revealed commonalities and differences in ribosome synthesis between lower and higher eukaryotes (Couté et al, 2008; Wild et al, 2010; Finkbeiner et al, 2011; Simabuco et al, 2012; Widmann et al, 2012; Tafforeau et al, 2013; Wyler et al, 2014; Zemp et al, 2014, 2009; Badertscher et al, 2015; Wandrey et al, 2015; Larburu et al, 2016; Raman et al, 2016; Fromm et al, 2017; Memet et al, 2017; Montellese et al, 2017, 2020; Ameismeier et al, 2018, 2020; Farley‐Barnes et al, 2018; Boneberg et al, 2019; Braun et al, 2020; Choudhury et al, 2020, 2019; Liang et al, 2020; Gerhardy et al, 2021; Ogawa et al, 2021; Pöll et al, 2021; Singh et al, 2021; Dörner et al, 2022; Sailer et al, 2022).

Underscoring the importance of ribosome maturation for cellular and organismal homeostasis, defects in ribosome biogenesis are associated with a variety of human diseases. Most RPs and RBFs are encoded by essential genes (Panić et al, 2006; Woolford & Baserga, 2013; Perucho et al, 2014). Genetic alterations in RPs and RBFs are frequently associated with haploinsufficiency and causative for a class of severe congenital diseases termed ribosomopathies (Bohnsack & Bohnsack, 2019; Farley‐Barnes et al, 2019; Kampen et al, 2020). A number of ribosomopathies are linked to increased cancer susceptibility and a growing body of evidence suggests that defects in ribosome biogenesis can drive tumorigenesis (Pelletier et al, 2018; Catez et al, 2019).

In this review, we present a current inventory of yeast and human ribosome biogenesis factors and their functions in eukaryotic ribosome maturation from rDNA organization and transcription, pre‐rRNA modification and processing, to subunit assembly and nuclear export.

rDNA organization in the nucleolus and pre‐rRNA transcription

Nucleoli are primarily dedicated to pre‐rRNA transcription and early steps of ribosomal subunit assembly. They are the most prominent nuclear, membrane‐less organelles. Nucleoli are built around rDNA loci which are organized in the form of nucleolar organizing regions (NORs) constituted by rDNA gene clusters (Ritossa & Spiegelman, 1965). While S. cerevisiae contains a single NOR on chromosome 12 with roughly 150 rDNA repeats, human cells possess NORs on the acrocentric human chromosomes 13, 14, 15, 21, and 22, harboring altogether roughly 400 rDNA repeats (Henderson et al, 1972; Stults et al, 2008; Lofgren et al, 2019). Not all NORs are active in a cell, and cell type‐specific differences in NOR activity exist (Roussel et al, 1996; Farley et al, 2015).

The organization of nucleoli differs between yeast and human. While yeast nucleoli consist of two subdomains, the fibrillar strands and granules, at least three subcompartments are found in human nucleoli, termed fibrillar center (FC), dense fibrillar center (DFC), and granular component (GC) (Fig 2A). rDNA loci are densely packed in the FC, whereas pre‐rRNA is transcribed at the boundary between FC and DFC (Koberna et al, 2002). Early processing steps including rRNA modification and cleavage also occur in the DFC, while pre‐ribosomal particles mature further in the surrounding GC. Nucleoli are considered to be multiphase liquid condensates (Lafontaine et al, 2021). However, their biogenesis and domain organization reflect the hierarchy in pre‐rRNA synthesis and ribosomal subunit assembly, highlighting the contribution of interaction specificity in shaping their molecular makeup and structure (Musacchio, 2022).

Figure 2. Organization of nucleoli, the rDNA locus and promoter architecture in yeast and human cells.

(A) Schematic representation of a yeast nucleolus composed of fibrillar strands (FS) and granules (G) and human nucleoli consisting of three subcompartments: fibrillar center (FC), dense fibrillar component (DFC), and granular component (GC). (B) Schematic representation of rDNA architecture in S. cerevisiae and human cells. (C) Comparison of 35S/47S rDNA promoter region with associated pre‐initiation complexes in yeast and human cells. Yeast promoters contain the upstream activation sequence (UAS) bound by the upstream activating factor (UAF) complex and the central element (CE) bound by the core factor (CF) complex. Human promoters also contain two elements; the upstream core element (UCE) bound by a UBF dimer and the central element (CE) bound by selectivity factor 1 (SL1) complex (Knutson & Hahn, 2013; Engel et al, 2018; Sadian et al, 2019; Pilsl & Engel, 2020; Baudin et al, 2022; Girbig et al, 2022). (D) Structural model of yeast RNAPI in complex with Rrn3, the CF (PDB ID: 7OBA), and UAF complexes, bound to Tbp and promoter DNA (PDB ID: 7Z0O). RNAPI subunits are in shadows of gray, factors are color‐coded, and DNA is shown in light blue.

Promoters of eukaryotic rDNA contain two regulatory elements: the core element (CE) and the upstream activating sequence or control element (UAS/UCE) (Fig 2C) (Knutson & Hahn, 2013). Although the general promoter architecture is conserved between yeast and mammals, regulation of RNAPI transcription likely differs between species, as there is little sequence similarity in the main promoter elements (Goodfellow & Zomerdijk, 2013). Several factors involved in yeast and human rDNA organization and transcription, including pre‐initiation complex (PIC) formation, elongation, and termination, have been identified (Table 2). In yeast, each rDNA repeat encodes two transcripts: the 5S rRNA, transcribed by RNAPIII, and the polycistronic 35S pre‐rRNA, produced by RNAPI. The 35S pre‐rRNA contains the 18S, 5.8S, and 25S rRNAs surrounded and separated by external and internal transcribed spacers (5′ ETS, ITS1, ITS2, 3′ ETS) (Fig 2B) (Woolford & Baserga, 2013). Recruitment of RNAPI to the rDNA promoter depends on the RNA polymerase I‐specific transcription factor Rrn3 and the heterotrimeric core factor (CF) complex, which recognizes the core promoter element upstream of the transcription start site (Fig 2C and D). The efficient initiation of pre‐rRNA transcription is further supported by the TATA‐binding protein (TBP) and the upstream activating factor (UAF) complex that binds an upstream activation sequence (UAS) (Russell & Zomerdijk, 2006; Girbig et al, 2022).

Table 2.

Factors involved in rDNA transcription.

Yeast	35S	Human	47S	Function in rRNA transcription	Citation
RNA Polymerase I subunits
Rpb5	x	POLR2E	x	RNAPI subunits, shared with RNAPII and RNAPIII between polymerases	Russell and Zomerdijk (2006)
Rpb6	x	POLR2F	x
Rpb8	x	POLR2H	x
Rpb10	x	POLR2L	x
Rpb12	x	POLR2K	x
Rpa40	x	POLR1C	x	RNAPI subunits, shared with RNAPIII
Rpa19	x	POLR1D	x
Rpa190	x	POLR1A	x	RNAPI‐specific subunits
Rpa135	x	POLR1B	x
Rpa43	x	POLR1F	x
Rpa14	x
Rpa12	x	POLR1H	x
Rpa49	x	POLR1E	x
Rpa34	x	POLR1G	x
PIC formation/promoter escape
Tbp	x	TBP	x	TATA box binding protein	Comai et al (1994)
		UBF	x	Binds rDNA promoter as a dimer, has a role in promoter escape and regulation of elongation	Bell et al (1988)
Rrn3	x	RRN3	x	RNAPI‐specific initiation factor, stimulates RNAPI recruitment	Milkereit and Tschochner (1998)
Rrn6	x	TAF1C	x	Involved in PIC formation. In yeast: part of CF Complex; in humans: part of selectivity factor SL1	Comai et al (1994)
Rrn7	x	TAF1B	x
Rrn11	x	TAF1A	x
		TAF1D	x	Involved in PIC formation: part of human Selectivity factor SL1	Denissov et al (2007) and Gorski et al (2007)
		TAF12	x
Rrn5	x			Involved in PIC formation: part of yeast UAF complex	Keys et al (1996), Keener et al (1997) and Siddiqi et al (2001)
Rrn9	x
Rrn10	x
Uaf30	x
Hht1/H3	x	H3C1/H3
Hhf1/H4	x	H4C1/H4
Top1	x	TOP1	x	Facilitate PIC formation by removing super coils at rDNA promoters and promote elongation	Brill et al (1987) and Ray et al (2013)
Top2	x	TOP2A	x
Cka1		CK2A1	x	Involved in PIC formation; part of tetrameric CK2 complex, regulate interaction between UBF and SL1	Panova et al (2006)
Cka2		CK2A2	x
Ckb1		CK2N	x
Elongation
Spt4	x	SPT4H	x	DSIF complex, influences RNAPI activity; Regulates binding of UBF to rDNA	Schneider et al (2006)
Spt5	x	SPT5H	x
Paf1	x	PAF1	x	Paf1C complex, influences RNAPI activity	Zhang et al (2009)
Ctr9	x	CTR9	x
Cdc73	x	WDR61	x
Rtf1	x	RTF1	x
Leo1	x	LEO1	x
Spt16	x	SUPT16H	x	FACT complex, supports RNAPI transcription through nucleosomes	Birch et al (2009)
Pob3	x	SSRP1	x
Fcp1	x			Dephosphorylates RNAPI for efficient RNA synthesis	Fath et al (2004)
Termination
Rnt1	x			Endonucleolytic cleavage at end of 25S transcript	el Hage et al (2008)
Rat1	x			Release of RNAPI transcribing 3′ part of the transcript	el Hage et al (2008)
Reb1	x	TTF1	x	RNAPI transcription termination factors, bind to termination sequence T1, lead to RNAP I pausing	Reiter et al (2012)
Nsi1	x				Reiter et al (2012)
Fob1	x			Binds to replication fork barrier sequence, inhibits clashes with DNA replication machinery	el Hage et al (2008)
rDNA organization
Hmo1	x			Associates with active rDNA repeats, related to UBF	Gadal et al (2002)
		CTCF	x	Regional organization of rDNA	van de Nobelen et al (2010)
		MYC	x	Attachment of rDNA to nucleolar matrix; local histone acetylation	Grandori et al (2005)
		MAX	x	Interacts with MYC	Nair and Burley (2003)
		TCOF1	x	Facilitates rDNA transcription, interacts with UBF and RNAPI, and is involved in rRNA methylation	Werner et al (2015)
Srp40		NOLC1	x	Interacts with TCOF1	Valdez et al (2004)
Bdf2		BRD2	x	Regulation of RNAPI activity by histone acetylation, recruited by LYAR	Izumikawa et al (2019)
Bdf1		BRD4	x
		KAT7	x
Nto1		JADE3	x
YCR087C		LYAR	x	Binds to UBF, recruits BRD2/4‐KAT7	Izumikawa et al (2019)
Factors acting both in rRNA transcription and early processing steps
		SIRT7	x	Required for the activation of Pol I transcription at the exit from mitosis	Ford et al (2006) and Iyer‐Bierhoff et al (2018)
Nop1		FBL	x	Methylates rRNA and histone H2AQ104	Tessarz et al (2014)
Utp4		UTP4	x	t‐UTPs, function in both rRNA transcription and the SSU processome	Gallagher et al (2004) and Prieto and McStay (2007)
Utp5		WDR43/UTP5	x
Utp10		HEATR1/UTP10	x
Utp15		UTP15	x
Utp17		UTP17/WDR75	x

Human rDNA repeats have by‐and‐large a similar architecture, although the 5S rRNA is transcribed from a distinct repeat region comprising roughly 100 loci on chromosome 1, located in the nucleoplasm in nucleolar proximity (Little & Braaten, 1989; Haeusler & Engelke, 2006; Stults et al, 2008). Also in human cells, several transcription factors and transcription factor complexes mediate RNAPI pre‐initiation complex assembly (Russell & Zomerdijk, 2006; Grummt, 2010). Initial binding of upstream binding factor (UBF) to the rDNA promoter allows for recruitment of the SL1 complex (containing TBP, factors analogous to CF in yeast (Fig 2C)), as well as metazoan‐specific factors (Table 2), before RNAPI is recruited via RRN3. Since pre‐rRNA transcription presents an initial rate‐limiting step of ribosome assembly, it is not surprising that many cellular signaling pathways target UBF, the SL1 complex, and RRN3 to regulate the production of ribosomes in human cells (Grummt, 2010; Bywater et al, 2013).

Pre‐rRNA modification

rRNA is heavily modified, in particular at functionally important regions such as the decoding center, the PTC, and the subunit interface (Decatur & Fournier, 2002; Polikanov et al, 2015; Sloan et al, 2017; Bailey et al, 2022). Many modifications are carried out co‐transcriptionally, and aid folding and compaction of the pre‐rRNA during assembly, but also support translation efficiency and accuracy (Liang et al, 2009; Sloan et al, 2017; Ojha et al, 2020; Khoshnevis et al, 2022). 2′‐O‐methylation of the ribose group in the rRNA backbone and pseudouridylation by isomerization of uridines are by far the most frequent modifications, introduced in a site‐specific manner by so‐called box C/D and box H/ACA small nucleolar ribonucleoproteins (snoRNPs) (see Table 3), respectively. Both types of RNPs contain four structural proteins and a dedicated modification enzyme, that is the methyltransferase fibrillarin (Nop1 in yeast) or the pseudouridine synthase dyskerin (Cbf5 in yeast). Each snoRNP also contains a 60–170 nt long snoRNA (with a few longer exceptions) (Marz et al, 2011; Jorjani et al, 2016), which guides the respective complex to its target site on the pre‐rRNA. Importantly, ribosome and snoRNA biogenesis are tightly interwoven in metazoans, since many snoRNAs are encoded within the introns of RP or RBF genes (Hirose & Steitz, 2001).

Table 3.

Factors involved in rRNA modification.

Yeast	40S	60S	Human	40S	60S	Function in rRNA modification	Citation
snoRNP components
Nop1	x	x	FBL	x	x	Box C/D snoRNP components, 2′O‐methylation of the ribose group in the RNA backbone	Grandi et al (2002)
Nop56	x	x	NOP56	x	x
Nop58	x	x	NOP58	x	x
Snu13	x	x	SNU13	x	x
Rrp9	x	x	U3‐55K	x	x
Cbf5	x	x	DKC	x	x	Box H/ACA snoRNP components, pseudouridylation of rRNA	Kiss‐László et al (1996)
Nhp2	x	x	NHP2	x	x
Nop10	x	x	NOP10	x	x
Gar1	x	x	GAR1	x	x
Stand‐alone enzymes
Tsr3	x		TSR3	x		Tsr3: aminocarboxypropyl transferase; Emg1: methyltransferase, 18S‐m1acp3Ψ1240 (y:1191)	Wurm et al (2010) and Meyer et al (2016)
Emg1	x		EMG1	x
Kre33	x		NAT10	x		Acetyltransferase 18S‐ac4C1337,1842 (y:1280,1773)	Ito et al (2014) and Sharma et al (2017)
Bud23	x		WBSCR22	x		Methyltransferase 18S‐m7G1639 (y:1575)	White et al (2008)
Dim1	x		DIMT1	x		Methyltransferase 18S‐m2 6A1850/1 (y:1781/2)	Lafontaine et al (1998)
Rrp8		x	NML/RRP8	x	x	Methyltransferase 28S‐m1A1332 (y:645)	Peifer et al (2013)
Bmt2		x				Methyltransferase 25S‐m1A2142	Sharma et al (2013)
Rcm1		x	NSUN5		x	Methyltransferase 28S‐m5C3761 (y:2278)	Schosserer et al (2015)
Bmt5		x				Methyltransferase 25S‐m3U2634	Sharma et al (2014)
Bmt6		x				Methyltransferase 25S‐m3U2843	Sharma et al (2014)
Nop2		x	NOP2		x	Methyltransferase 28S‐m5C4414 (y:2870)	Sharma et al (2014)
Spb1		x			x	Methyltransferase 28S‐Gm4469 (y:2922)	Lapeyre and Purushothaman (2004)
			ZCCHC4		x	Methyltransferase 28S‐m6A4220	Ma et al (2019) and Pinto et al (2020)
			METTL5	x		Methyltransferase 18S‐m6A1832	van Tran et al (2019)
Trm112			TRMT112	x		Activator of methyltransferases (Bud23/WBSCR22, METTL5)	Zorbas et al (2015) and van Tran et al (2019)

Both the binding of snoRNAs to the pre‐rRNA and the resulting nucleotide modifications play important roles in pre‐rRNA folding (see recent reviews Mitterer & Pertschy, 2022; Oborská‐Oplová et al, 2022). Many snoRNA binding sites are found in pre‐rRNA regions that fold late during nucleolar assembly steps. By base‐pairing with these still unfolded pre‐rRNA elements, snoRNAs prevent premature or non‐productive formation of RNA helices. Furthermore, multivalent snoRNAs, exemplified by the U3 snoRNA that is involved in early pre‐rRNA folding and processing steps (see below), can force distant pre‐rRNA regions into a defined configuration while simultaneously inhibiting untimely RNA annealing events. Nucleotide modifications, in turn, affect base‐pairing preferences and can enhance the conformational rigidity of the RNA backbone (Sumita et al, 2005; Helm, 2006; Abou Assi et al, 2020).

In yeast, ~75 snoRNAs have been described, catalyzing the modification of 112 sites, while in humans 228 sites are targeted by more than 200 snoRNAs (Natchiar et al, 2017; Taoka et al, 2018, 2016). The lack of individual modifications is generally well tolerated in yeast cells, whereas their cumulative loss can cause defects in subunit assembly, translation, and cell growth (Liang et al, 2007, 2009). Notably, both in yeast and mammalian cells, there is some heterogeneity in the use of modification sites (Jaafar et al, 2021b), but evidence for a functional relevance of this variability remains scarce (Metge et al, 2021). More than 130 rRNA modifications have been visualized by single particle cryo‐EM in the human ribosome, of which 11 were found at universally conserved sites (Natchiar et al, 2017). As expected, the vast majority of modifications reside in the interior of the ribosome close to the functional centers. Interestingly, also a large number of novel sites were discovered, many of which contain base‐modified nucleotides.

Some rRNA modifications do not depend on snoRNPs, but are introduced by stand‐alone enzymes, many of which function in base methylation, but also support other modifications such as acetylation (listed in Table 3). For the majority of these enzymes only a single target nucleotide has been described, but the functions of these modifications remain largely enigmatic. The most complex ribosomal modification, a 1‐methyl‐3‐(3‐amino‐3‐carboxypropyl)‐pseudouridine (m1acp3Ψ) in helix 31 of the 18S rRNA (1248U), is established in a stepwise manner, starting with the formation of a pseudouridine catalyzed by the snR35 H/ACA snoRNP complex (Samarsky et al, 1995). After subsequent methylation by Emg1 (Wurm et al, 2010), the final modification is formed by the aminocarboxypropyl transferase Tsr3 during final cytoplasmic maturation (Meyer et al, 2016; Huang et al, 2022). Importantly, m1acp3Ψ, which is solvent exposed at the ribosomal P site, is frequently lost or hypomodified in cancer and has been suggested to lead to increased translation of RP mRNAs (Babaian et al, 2020). Recently, it was discovered that human rRNA also contains m⁶A modifications in the 18S and 28S rRNAs, mediated by the methyltransferases METTL5 and ZCCHC4 respectively, yet the molecular function of the two identified modification sites remains to be further explored (Ma et al, 2019; van Tran et al, 2019; Pinto et al, 2020).

Pre‐rRNA processing

A series of concerted pre‐rRNA cleavage and trimming reactions in the nucleus and cytoplasm leads to the excision of the mature 18S, 5.8S, and 25/28S rRNAs from the polycistronic rRNA precursor (Fig 3). The general hierarchical principle of sequential elimination of transcribed spacers and the function of several endo‐ and exonucleases (Table 4) in this process is largely conserved from yeast to human (for more details see Tomecki et al, 2017; Aubert et al, 2018; Bohnsack & Bohnsack, 2019). In both organisms, pre‐rRNA processing is tightly interconnected with other steps of ribosome biogenesis, such as rRNA modification, folding, and binding of RPs or RBFs. A critical step in the maturation pathway is the first cleavage event within the ITS1 (at site A₂ in yeast, site 2 in humans), which separates the emerging pre‐40S and pre‐60S particles. In yeast, ~70% of nascent transcripts are cleaved co‐transcriptionally within ITS1 (Koš & Tollervey, 2010). Human pre‐rRNA processing also starts co‐transcriptionally, at least in the 5′ ETS region (Delannoy & Sollner‐Webb, 1997; Osheim et al, 2004). However, it remains to be defined whether the endonucleolytic cleavage within the ITS1 occurs only post‐transcriptionally or also co‐transcriptionally in mammalian cells, and how this critical event is coordinated with subunit assembly. The prominent presence of the long 47S/45S precursors in pulse‐labeling experiments and Northern blots indicates that at least a substantial fraction of ITS1 cleavage events may only occur after synthesis of the precursor has been completed (Bowman et al, 1981; Strezoska et al, 2000). Notably, in both yeast and mammals, alternative pre‐rRNA processing pathways exist, and their relative contribution to rRNA production is suggested to be governed by the respective kinetics of the processing reactions (Axt et al, 2014; Henras et al, 2015).

Figure 3. Pre‐rRNA maturation in yeast and human cells.

Simplified processing pathway of the 35S pre‐rRNA in yeast and 47S pre‐rRNA in human cells, indicating processing sites of endo‐ and exonucleases in orange and blue, respectively. Several alternative processing pathways exist, which are reviewed elsewhere (Tomecki et al, 2017; Aubert et al, 2018).

Table 4.

Factors involved in pre‐rRNA processing.

Yeast	40S	60S	Human	40S	60S	Function in pre‐rRNA processing	Citation
Utp23			UTP23	x		Endoribonuclease, cleavage at site A0 (likely inactive)	Wells et al (2017)
Utp24/Fcf1	x		UTP24	x		Endoribonuclease, cleavage at site 1 (A1)	Bleichert et al (2006)
Nob1	x		NOB1	x		Endoribonuclease, cleavage at site 3 (D)	Fatica et al (2004)
RMRP		x	RMRP		x	Part of MRP complex, endoribonuclease with non‐canonical snoRNA, cleavage at site 2 (A3)	Perederina et al (2020), van Hoof et al (2000) and Goldfarb and Cech (2017)
Pop1		x	POP1		x
Pop3		x	RPP38		x
Pop4		x	RPP29		x
Pop5		x	POP5		x
Pop6		x	RPP25		x
Pop7		x	RPP20		x
Pop8		x	RPP14		x
Rpp1		x	RPP30		x
Rpr2		x	RPP21		x
Snm1		x
Rmp1		x
			RPP40		x
Las1		x	LAS1L			Endoribonuclease, maybe cleavage at site 4 (C2)	Schillewaert et al (2012)
Rnt1		x				Endoribonuclease (B0)	Kufel et al (1999)
Rat1		x	XRN2	x	x	Exoribonuclease, pre‐rRNA trimming	el Hage et al (2008)
Rrp17		x	NOL12			Exoribonuclease, pre‐rRNA trimming	Oeffinger et al (2009)
Rrp44		x	DIS3		x	Nuclear exosome, exoribonuclease, pre‐rRNA Trimming 5.8S rRNA maturation	Briggs et al (1998) and Sloan et al (2012)
Rrp6		x	EXOSC10		x
Csl4		x	EXOSC1		x
Rrp4		x	EXOSC2		x
Rrp40		x	EXOSC3		x
Rrp41		x	EXOSC4		x
Rrp46		x	EXOSC5		x
Mtr3		x	EXOSC6		x
Rrp42		x	EXOSC7		x
Rrp43		x	EXOSC8		x
Rrp45		x	EXOSC9		x
Mtr4		x	MTR4		x	Cofactor of nuclear exosome	Briggs et al (1998) and Sloan et al (2012)
Trf4		x	PAPD5		x
Trf5		x	PAPD7		x
Air1		x	ZCCHC7		x
Air2		x	ZCCHC8		x
Mpp6		x	MPP6		x
Lrp1/Rrp47		x	C1D		x
		x	RBM7		x
Rex1		x	REXO5			Exoribonuclease, pre‐rRNA trimming	van Hoof et al (2000)
Rex2		x	REXO2			Exoribonuclease, pre‐rRNA trimming	van Hoof et al (2000)
Rex3		x	REXO1			Exoribonuclease, pre‐rRNA trimming	van Hoof et al (2000)
Ngl2		x	Ccr4			Exoribonuclease, pre‐rRNA trimming, 5.8S rRNA maturation	Faber et al (2002)
			PARN	x		Exoribonuclease, pre‐rRNA trimming, 18S rRNA maturation	Montellese et al (2017)
			TUT4	x		Uridylyltransferase, pre‐rRNA trimming, 18S rRNA maturation	Montellese et al (2017)
			TUT7	x		Uridylyltransferase, pre‐rRNA trimming, 18S rRNA maturation	Montellese et al (2017)
			ISG20L2		x	Exoribonuclease, pre‐rRNA trimming 5.8S rRNA maturation	Couté et al (2008)
			ERI1		x	Exoribonuclease, pre‐rRNA trimming, 5.8S rRNA	Ansel et al (2008)

The fourth rRNA, the 5S rRNA, is transcribed by RNAPIII. The immature 5S rRNA is initially bound by the La protein and TFIIIA in higher eukaryotes, protecting it from degradation (Ciganda & Williams, 2011; Layat et al, 2013). It needs to be trimmed at its 3′ end before getting incorporated into the maturing pre‐60S particle. This trimming is performed by the exonucleases Rex1, Rex2, and Rex3 in yeast, which act redundantly (Table 4) (van Hoof et al, 2000). Factors involved in human 5S rRNA maturation remain elusive. The Rex1 homolog REXO5 was shown to be functionally conserved in flies (Gerstberger et al, 2017), although mouse REXO5 is not essential for survival (Silva et al, 2017), pointing to potential redundancy. The processed 5S RNA associates with two newly synthesized RPs, RPL5/uL18, and RPL11/uL5, forming the 5S RNP which is then incorporated into nascent 60S subunits (see below).

Chaperones of ribosomal proteins

Ribosomal proteins are synthesized in the cytoplasm, but most are incorporated into pre‐ribosomal particles in the nucleolus, posing a logistic challenge to the cell. As RPs are enriched in basic amino acids and contain flexible tails as well as intrinsically disordered regions (Klinge et al, 2011; Rabl et al, 2011), they are prone to aggregation. Therefore, newly synthesized RPs need to be kept away from undesired interactions until they are incorporated into pre‐ribosomal particles. Several mechanisms contribute to avoiding adverse effects of unincorporated ribosomal proteins: dedicated chaperones for specific RPs (see below), the general chaperone network (Gong et al, 2009; Albanèse et al, 2010; Koplin et al, 2010; Leidig et al, 2013; Pillet et al, 2017) and association with nuclear transport receptors en route into the nucleus (Jäkel et al, 2002). In addition, excess, unincorporated RPs are degraded (Warner, 1977; Lam et al, 2007; Sung et al, 2016).

About a dozen dedicated chaperones for RPs have been described (Table 5). Some of them capture nascent RPs co‐translationally (e.g., Yar1 for Rps3/uS3, Rrb1 for Rpl3/uL3, Syo1 for Rpl5/uL18, Sqt1 for Rpl10/uL16, and Acl4 for Rpl4/uL4), whereas others bind their clients later (Pausch et al, 2015; Pillet et al, 2017). Interestingly, it has recently been revealed that the absence of the RP chaperones Acl4 or Rrb1 leads to a destabilization of the mRNAs encoding for their respective clients Rpl4 and Rpl3, revealing a novel regulatory mechanism of RP homeostasis (Pillet et al, 2022). RP chaperones not only shield their clients but can also aid their delivery into the nucleus. For those RPs that get incorporated into nascent subunits in the nuclear compartment, the nuclear localization signal can either be provided by the RP itself (e.g., Rps3/uS3, Rpl3/uL3, Rpl4/uL4), or by the dedicated chaperone (e.g., Syo1; Kressler et al, 2012; Bange et al, 2013; Calviño et al, 2015).

Table 5.

Chaperones of ribosomal proteins.

Yeast	40S	60S	Human	40S	60S	Chaperoned RP	Citation
Rrb1		x	GRWD1			Rpl3/uL3	Iouk et al (2001)
Acl4		x				Rpl4/uL4	Stelter et al (2015)
Syo1		x	HEATR3		x	Rpl5/uL18 and Rpl11/uL5	Kressler et al (2012), Hannan et al (2022) and O'Donohue et al (2022)
Sqt1		x	AAMP		x	Rpl10/uL16	Eisinger et al (1997)
Bcp1		x	BCCIP		x	Rpl23/uL14	Ting et al (2017)
Loc1		x				Rpl43/eL43	Liang et al (2019)
Puf6		x				Rpl43/eL43	Liang et al (2019)
Tsr4	x		PDCD2			Rps2/uS5	Black et al (2019) and Rössler et al (2019)
Yar1	x					Rps3/uS3	Lindström and Zhang (2008)
Nap1	x					Rps6/eS6	Rössler et al (2019)
			NPM	x		Rps9/uS4	Lindström and Zhang (2008)
Fap7	x		AK/CINAP	x		Rps14/uS11, maybe also in complex with Rps26/eS26	Hellmich et al (2013) and Peña et al (2016)
			AROS	x		Rps19/eS19	Singh et al (2021)
Tsr2	x		TSR2	x		Rps26/eS26	Schütz et al (2014)

One peculiar case is yeast Tsr2, which serves as dedicated chaperone for Rps26/eS26. Tsr2 has been proposed to pick up newly synthesized Rps26/eS26 only after nuclear import by stimulating Rps26/eS26 release from importins in the nucleus, to then allow its nuclear association with pre‐40S subunits (Schütz et al, 2018). In the mammalian system, RPS26/eS26 is incorporated into the pre‐40S particle during its final cytoplasmic maturation (Ameismeier et al, 2018, 2020; Plassart et al, 2021). Here, TSR2 could help in preventing association of RPS26/eS26 with nuclear import receptors in the cytosol, but more work is needed to unravel the place of action of mammalian TSR2. Notably, also in yeast Tsr2 may harbor a cytosolic function, as it has recently been suggested to facilitate Rps26/eS26 release from and reincorporation into mature ribosomes in response to salt or pH stress (Yang & Karbstein, 2022).

The function of RP chaperones does not seem to be limited to factors acting in trans. In most eukaryotes, Rps31/RPS27A/eS31 and Rpl40/eL40 are initially synthesized as linear fusions with an N‐terminal ubiquitin, which facilitates their folding and enhances solubility of the respective RPs (Finley et al, 1989; Lacombe et al, 2009; Martín‐Villanueva et al, 2019, 2020, 2021). After synthesis, the ubiquitin moiety is rapidly released from these fusion proteins (Grou et al, 2015) and enters the cellular ubiquitin pool. Interestingly, in humans and other holozoan organisms, a second RP of the small subunit, RPS30/eS30, is synthesized as a fusion with a ubiquitin‐like protein called FUBI. Release of FUBI from the FUBI‐eS30 fusion protein is required for 40S subunit maturation, likely linked to its nuclear incorporation into pre‐40S subunits, and promoted by the deubiquitinase USP36 (van den Heuvel et al, 2021).

Formation and maturation of the SSU processome

When the pre‐rRNA emerges from the transcribing polymerase, it is soon bound by a subset of RPs of the small subunit, 40S RBFs as well as by small nucleolar ribonucleoproteins (snoRNPs), eventually giving rise to a large ribosomal pre‐particle that is referred to as the small subunit (SSU) processome or 90S pre‐ribosome (Fig 4) (Dragon et al, 2002; Grandi et al, 2002). Early pre‐rRNA processing events are executed during maturation of this complex, leading to the removal of the external transcribed spacer (5′ ETS) and cleavage within ITS1. Initially, high‐resolution structures of different fungal SSU processome particles were obtained, giving invaluable mechanistic insights into the function of a multitude of RBFs during early assembly, pre‐rRNA folding, and processing events (Kornprobst et al, 2016; Barandun et al, 2017; Chaker‐Margot et al, 2017; Cheng et al, 2017; Sun et al, 2017; Du et al, 2020; Lau et al, 2021). Then, in 2021, the first structures of the human SSU processome were published, providing visual evidence for the existence of SSU processome particles in higher eukaryotic cells and illustrating the vast conservation of its global architecture (Singh et al, 2021) (Fig 4, Table 6).

Figure 4. Structures of the yeast and human SSU processome.

Cryo‐EM structures of the yeast (PDB ID: 6ZQB, 7AJT, and 7AJU) and human (PDB ID: 7MQ8, 7MQ9, 7MQA) SSU processome particles in the pre‐A1, pre‐A1* and post‐A1 cleavage states, showing a comparison of their overall architecture. The conserved RNA components and subcomplexes are color‐coded. The pre‐rRNA (white) and individual sub‐complexes such as UTP‐A (pink), UTP‐B (light blue), UTP‐C (violet), Emg1 (green), ANN (light green) complex, and additional RBFs, are shown as surfaces. KRR1/Krr1, C1orf131/Faf1, and the NAT10/Kre33 module are indicated as pre‐A1‐specific RBFs (sand). They are released after A1 cleavage. The exosome complex (yellow) is bound to the pre‐A1* yeast particle and it is associated only with the human post‐A1 structure. In the post‐A1 structures, DHX37/Dhr1 and DIM1/Dim1 are displayed in orange.

Table 6.

Small subunit processome subcomplexes and early assembly factors.

Yeast	40S	60S	Human	40S	60S	Function in early nucleolar ribosome biogenesis	Citation
Utp4	x		UTP4/CIRH1A	x		UTP‐A complex, interacts with the start of 5′ ETS, required for recruitment of UTP‐B and U3 snoRNP, essential for formation of 5′ ETS particle	Krogan et al (2004), Prieto and McStay (2007), Freed et al (2012), Chaker‐Margot et al (2017) and Singh et al (2021)
Utp5	x		UTP5/WDR43	x
Utp8	x		NOL11	x
Utp9	x
Utp10	x		UTP10/HEATR1	x
Utp15	x		UTP15	x
Utp17	x		UTP17/WDR75	x
Pol5	x	x	MYBBP1A	x
Utp1/Pwp2	x		PWP2	x		UTP‐B complex, binds 5′ ETS and U3 snoRNA, supports structural pre‐rRNA remodeling, essential for formation of 5′ ETS particle, Utp18 contains exosome interaction motif	Krogan et al (2004), Sloan et al (2015), Barandun et al (2017), Barandun et al (2018) and Singh et al (2021)
Utp6	x		UTP6	x
Utp12/Dip2	x		UTP12/WDR3	x
Utp13	x		TBL3	x
Utp18	x		UTP18	x
Utp21	x		WDR36	x
			DDX21	x
			NOP2	x
Nop1	x		FBL	x		U3 snoRNP, U3 snoRNA base pairs with 5′ ETS, chaperones pre‐rRNA folding steps, essential for formation of 5′ ETS particle	Kiss‐László et al (1996), Grandi et al (2002), Barandun et al (2017) and Singh et al (2021)
Nop56	x		NOP56	x
Nop58	x		NOP58	x
Snu13	x		SNU13/15.5 K	x
Rrp9	x		U3‐55K/U3IP2	x
Mpp10	x		MPHOSPH10	x		Mpp10‐Imp3‐Imp4 complex, interacts with U3 snoRNA, supports formation of 5′ ETS particle	Lee and Baserga (1999) and Granneman et al (2003)
Imp3	x		IMP3	x
Imp4	x		IMP4	x
Rrp7	x		RRP7A	x		UTP‐C complex, chaperones 5′ domain of 18S rRNA	Baudin‐Baillieu et al (1997), Krogan et al (2004), Rudra et al (2007), Barandun et al (2017) and Singh et al (2021)
Utp22	x		NOL6	x
Cka1	x		CK2A1	x
Cka2	x		CK2A2	x
Ckb1	x		CK2N	x
Ckb2	x		CK2N	x
Rrp36	x		RRP36	x
Utp7	x		WDR46	x		Sof1‐Utp7 complex, aids organization of A1 cleavage site	Barandun et al (2017)
Utp14	x		UTP14A	x
Sof1	x		WDSOF1	x
Rcl1	x		RCL1	x		Rcl1‐Bms1‐complex, GTPase activity (Bms1), mediates cleavage at A2 site, also required for A0 and A1 cleavage, interacts with U3 snoRNP	Karbstein and Doudna (2006) and Horn et al (2011)
Bms1	x		BMS1	x
Enp1	x		ENP1/BYSL	x		In yeast: Nop14‐Noc4‐Enp1 complex, in humans: NOP14‐NOC4L‐UTP14A‐EMG1, structural role: binds to 3′ domain of 18S pre‐RNA, Emg1: methyltransferase	Liu and Thiele (2001), Kühn et al (2009), Warda et al (2016) and Barandun et al (2017)
Utp2/Nop14	x		NOP14	x
Noc4	x		NOC4L	x
Emg1	x		EMG1	x
Bfr2	x		AATF	x		in humans: ANN complex, required for cleavage at A0, 1 and in the ITS1, in yeast: Bfr2‐Enp2 recruit Dpb4	Soltanieh et al (2014) and Bammert et al (2016)
Lcp5	x		NGDN	x
Enp2	x		NOL10	x
			XRN2	x		XND complex, required for A′ cleavage, recruits XRN2 for degradation of excised spacer fragment	Dragon et al (2002) and Memet et al (2017)
			NKRF		x
Prp43			DHX15		x
Utp30	x		RSL1D1	x		Utp30‐Rrt14 complex, binds 5′ ETS and pre‐18S rRNA, Rrt14 is non‐essential	Barandun et al (2017)
Rrt14	x
Utp11	x		UTP11	x
Bud21/Utp16	x		NOL7	x		Required for 18S pre‐rRNA processing	Dragon et al (2002) and Singh et al (2021)
Fcf2	x		TDIF2	x		Binds U3 snoRNP, TDIF2 contains AIM putative motif	Rempola et al (2006), Barandun et al (2017) and Singh et al (2021)
Sas10	x		UTP3	x		Stabilizes and chaperones Mpp10 complex to nucleolus, blocks Emg1 active site	Zhao et al (2019)
Utp24	x		UTP24	x		Endoribonuclease, couples pre‐rRNA cleavages in yeast at sites A1 and A2, in humans: site 1 and E	Bleichert et al (2006) and Wells et al (2016)
Esf1	x		ESF1			Involved in early pre‐rRNA processing	Peng et al (2004)
Esf2	x		ABT1			Stimulates Dbp8	Granneman et al (2006b)
Dbp8	x		DDX49	x		DEAD box RNA helicase	Granneman et al (2006b) and Awasthi et al (2018)
Dbp4	x		DDX10	x		DEAD box RNA helicase	Granneman et al (2006a) and Turner et al (2009)
Rrp3	x		DDX47	x		DEAD box RNA helicase	Granneman et al (2006a)
Bud22	x		SRFBP1				Dakshinamurthy et al (2010)
Efg1	x					Required for 18S pre‐rRNA processing (A1, A2), initiates degradation of aberrant 23S pre‐rRNA	Choque et al (2018)
Rrp5	x	x	PDCD11	x	x	Structural SSU component supporting pre‐rRNA, compaction, important for 18S maturation (site A0‐A2) and 5.8S processing (site A3)	Venema and Tollervey (1996) and Lebaron et al (2013)
Krr1	x		KRR1	x		Interacts with Faf1, important for 40S platform, assembly as it is replaced by Dim2	Zheng et al (2014) and Sturm et al (2017)
Rok1	x		DDX52	x		DEAD box RNA helicase, releases Rrp5, releases snR30	Khoshnevis et al (2016)
Utp25	x		DEF/C1orf107	x			Charette and Baserga (2010) and Tao et al (2017)
Kri1	x		KRI1			Interacts with Krr1	Sasaki et al (2000)
Utp23	x		UTP23	x		Endoribonuclease, in yeast: likely inactive, in humans: cleavage at site A0	Wells et al (2017)
Fyv7	x						Peng et al (2003)
Mrd1	x		RBM19	x		Aids formation of central pseudoknot, required for dynamic U3 snoRNA‐rRNA interaction (release)	Segerstolpe et al (2013) and Lackmann et al (2018)
Fal1	x		DDX48/EIF4A3	x		DEAD box RNA helicase	Kressler et al (1997) and Davila Gallesio et al (2020)
Sgd1	x		NOM1	x		Fal1 cofactor	Davila Gallesio et al (2020)
Cms1	x		CMS1	x			Grandi et al (2002)
Nop9	x		NOP9			Impedes Nob1 cleavage	García‐Gómez et al (2011) and Zhang et al (2016a)
Nop6	x					Not essential for ribosome biogenesis	García‐Gómez et al (2011)
Utp20	x		UTP20	x		Associates with faulty intermediates, might play a role in rRNA quality control	Dez et al (2007)
Kre33	x		NAT10	x		ATP‐dependent RNA acetyltransferase	Ito et al (2014) and Sharma et al (2017, 2015)
Faf1	x		C1orf131			Interacts with Krr1/KRR1	Zheng et al (2014) and Singh et al (2021)
Dim2/Pno1	x		PNO1/DIM2	x		Interacts with Nob1	Chaker‐Margot et al (2017)
Rrp12	x		RRP12	x		Part of the 3′ minor domain	Ameismeier et al (2018)
Nob1	x		NOB1	x		Endoribonuclease, associates with ITS1, catalyzes removal of final part of ITS1 (site D/3), might only associate in nucleoplasm	Fatica et al (2004)
Dhr1	x		DHX37	x		DEAH box RNA helicase, releases U3 snoRNA	Boneberg et al (2019)
Dhr2	x					DEAH box RNA helicase, interacts with Utp25 and Nop19	Granneman et al (2006a)
Nop19	x					Important for Utp25 incorporation	Choque et al (2011)
Rrp8	x		NML/RRP8	x		Methyltransferase	Peifer et al (2013)
Slx9	x		FAM207A/C21orf70			Non‐essential, supports efficient ITS1 processing	Bax et al (2006)
Pol5	x	x	MYBBP1A	x		Turnover of 5′ ETS fragment, required for recycling of SSU RBFs	Braun et al (2020)
Bud23	x		WBSCR22	x		Methyltransferase, supports disassembly of SSU particle	Black et al (2020)
Nip7		x	NIP7	x		In contrast to yeast, human NIP7 and FTSJ3 were shown to function in 40S biogenesis	Morello et al (2011a)
Sbp1		x	FTSJ3	x

Small subunit processome assembly is initiated when the 5′ ETS region of the nascent pre‐rRNA transcript is bound by the so‐called UTP‐A complex, a seven‐membered protein complex that provides binding interfaces for later joining SSU subcomplexes (Pérez‐Fernández et al, 2007; Zhang et al, 2016b; Barandun et al, 2017; Cheng et al, 2017; Hunziker et al, 2019). The UTP‐A complex is also needed for pre‐rRNA synthesis, thereby linking pre‐rRNA transcription and ribosome assembly (Gallagher et al, 2004). Following the UTP‐A complex, the six‐membered UTP‐B complex and the U3 snoRNP are co‐transcriptionally recruited to the 5′ ETS (Chaker‐Margot et al, 2015; Hunziker et al, 2016; Kornprobst et al, 2016). The UTP‐B complex acts as a chaperone for both the 5′ ETS and the U3 snoRNA, and upon incorporation of additional factors, including the Mpp10 complex (Granneman et al, 2003), a large 5′ ETS‐associated particle comprising more than 25 RBFs is formed (Chaker‐Margot et al, 2015; Zhang et al, 2016b; Barandun et al, 2017). Within the 5′ ETS‐associated particle, the U3 snoRNP serves as a key organizer orchestrating SSU processome formation. In both human and yeast cells, the U3 RNA base pairs with two regions each of the 5′ ETS and the 18S rRNA (Granneman et al, 2009; Dutca et al, 2011; Barandun et al, 2017; Singh et al, 2021), thereby functioning as a critical structural constraint for pre‐rRNA folding during SSU processome assembly (Fig 4). Importantly, this prevents the premature formation of the central pseudoknot, a universally conserved element of rRNA tertiary structure that is part of the small ribosomal subunit decoding center (Sardana et al, 2015). Notably, many RBFs within the SSU processome bind to more than one site of the pre‐rRNA, thereby reducing the conformational freedom of the rRNA during folding. In particular, SSU RBFs such as Faf1, Utp11, Mpp10, Sas10, and Nop14 exhibit long extensions that bridge distant rRNA regions within these precursors, thereby contributing to correct pre‐rRNA organization (Chaker‐Margot et al, 2017; Cheng et al, 2017).

A striking difference between yeast and mammalian cells is the length of the 5′ ETS that comprises roughly 700 nucleotides in yeast, but is extended to 3,600 nucleotides in humans. Only relatively short, structured parts of the 5′ ETS could be visualized in the structural models of the human SSU processome (Singh et al, 2021). Surprisingly, a minimal human 5′ ETS comprising these RNA segments (in total only ~25% of the entire 5′ ETS region) is sufficient for correct 40S and 60S assembly (Singh et al, 2021). In light of this data, the function of the other 5′ ETS regions is enigmatic. It has been proposed that they might contribute to nucleolar organization by supporting the formation of biomolecular condensates and thereby nucleolar phase separation (Yao et al, 2019). Along the same line of thought, it is conceivable that the increase in 5′ ETS length contributes to the observed differences in subnucleolar organization between mammals and yeast.

The 5′ ETS particle has originally been suggested to serve as a binding platform for the hierarchical recruitment of further protein complexes to sequentially promote the folding of spatially distant rRNA domains (Kornprobst et al, 2016; Cheng et al, 2017; Sun et al, 2017). However, since individual 18S rRNA domains can in principle recruit their respective biogenesis factors independently, it has also been proposed that the 5′ ETS and downstream 18S rRNA domains may function as independent units in the recruitment of their respective assembly factor complexes which may then support SSU processome assembly based on their mutual dependence (Hunziker et al, 2019). In either case, further assembly factors are recruited to the nascent rRNA concomitant with ongoing transcription in an rRNA‐subdomain dependent fashion (Pérez‐Fernández et al, 2007), among them the Bms1‐Rcl1 complex, the Nat10/Kre33 module (Kre33–Brf2–Lcp5–Enp2 in yeast/NAT10‐(AATF‐NGDN‐NOL10) (ANN) complex in humans) and the UTP‐C complex (Bammert et al, 2016; Barandun et al, 2018). Notably, also other snoRNAs including U14, snR30/U17, and snR10 transiently associate with the 5′ and central domains of the 18S rRNA (Zhang et al, 2016b) and aid 18S rRNA processing and folding. In the course of SSU processome assembly, cleavage at site A0 in the 5′ ETS occurs, eventually giving rise to a large, stable intermediate referred to as the “pre‐A1” particle in which site A1 is still uncut (Fig 4) (vanden Broeck & Klinge, 2022). The human pre‐A1 particle is 3.3 MDa in size and contains about 21 RPs and 50 RBFs (Singh et al, 2021).

To initiate cleavage at site A1, which generates the mature 5′ end of the 18S rRNA, the PIN domain endonuclease Utp24/UTP24 must gain access to the processing site (Tomecki et al, 2017; Barandun et al, 2018, 2017; Singh et al, 2021; vanden Broeck & Klinge, 2022). This involves a series of events, commencing with the release of Lcp5/NGDN (state pre‐A1*; Fig 4), followed by structural remodeling of the SSU particle that affects the positioning of the U3 snoRNA. Particle remodeling is associated with the release of the Utp24/UTP24 inhibitory factor Faf1/C1ORF131 and the residuals of the Kre33/NAT10 module (state post A1; Fig 4) (Cheng et al, 2020; Du et al, 2020; Singh et al, 2021). What drives these events and how the enzymatic activity of Kre33/NAT10 is coordinated with SSU processome maturation, potentially aided by the juxtaposed ANN complex or other factors, remains to be unraveled. Importantly, also the nuclear exosome may play a crucial role in the remodeling of the SSU processome (Du et al, 2020) by driving the 3′–5′ unwinding of the 5′ ETS after cleavage at site A0 (Du et al, 2020; Lau et al, 2021), contributing to structural remodeling of the SSU‐bound 5′ ETS particle and the SSU “head domain”. Recent structural and biochemical data obtained from yeast indeed revealed that the nuclear exosome already binds 90S particles prior to cleavage at site A1 as visualized in pre‐A1* particles (Fig 4) (Du et al, 2020; Lau et al, 2021). Yet, in humans, the exosome was only found to be associated with the post‐A1 structure and its binding sites are blocked in the described human pre‐A1 particles (Singh et al, 2021).

The disassembly of the SSU processome involves the activity of the conserved RNA helicase Dhr1/DHX37 (Sardana et al, 2015; Boneberg et al, 2019; Choudhury et al, 2019), which drives the release of the U3 snoRNP to allow the formation of the conserved pseudoknot in the 18S rRNA. Dhr1/DHX37 is recruited to the SSU processome already upon release of the ANN complex, and initially kept in an inhibited state. Activation is tightly linked to the choreography of SSU processome disassembly, when the RNA helicase gains spatial proximity to its activator UTP14 (Boneberg et al, 2019; Choudhury et al, 2019; Singh et al, 2021). Recent in vitro work reconstituted the Dhr1‐dependent release of the U3 snoRNA, allowing the visualization of these ATP‐dependent remodeling steps and details of central pseudoknot maturation (Cheng et al, 2022).

Cleavage in the ITS1 leads to the separation of pre‐40S and pre‐60S particles, which further mature independently in nucleoli and nucleoplasm. In both yeast and human, the ITS1 is cleaved at two sites (site A2 and A3, E and 2, respectively) by the conserved enzymes UTP24 (site A2, E) and RMRP (site A3, 2) (Udem et al, 1971; Lygerou et al, 1996; Rouquette et al, 2005; Tomecki et al, 2017; Aubert et al, 2018). The order of cleavages however differs: in yeast cleavage at site A2 by Utp24 is responsible for splitting the pre‐ribosomal particles, while in human cells the majority of pre‐rRNAs is first processed at site 2 by RMRP resulting in pre‐40S and pre‐60S separation (Allmang et al, 2000; Preti et al, 2013; Sloan et al, 2013b).

Nuclear maturation and export of the pre‐40S subunit

Recent structural analyses provided insights into the molecular organization of a series of nucleoplasmic 40S precursors from yeast and human cells (Cheng et al, 2022). Early nucleoplasmic 40S assembly intermediates already possess many features of mature 40S subunits, except that the head region remains largely delocalized and helix 18 of the 18S rRNA is kept in an immature conformation (Cheng et al, 2022). This is achieved by Bud23/WBRSC22, the binding of which has been implicated in the remodeling of the SSU processome into immature 40S particles (Black et al, 2020; Cheng et al, 2022). Bud23/WBRSC22 together with its adaptor Trm112/TRMT112 also acts as an RNA methyltransferase factor and is required for the methylation of a conserved guanosine in the P‐site of the 40S subunit (White et al, 2008; Figaro et al, 2012; Létoquart et al, 2014; Zorbas et al, 2015). Besides Bud23/WBRSC22, Tsr1/TSR1 and Slx9/C21orf70/FAM207A are also already bound to these early nucleoplasmic precursors (Table 7).

Table 7.

Factors involved in nucleoplasmic steps of pre‐40S maturation.

Yeast	40S	60S	Human	40S	60S	Function in nucleoplasmic steps of pre‐40S maturation	Citation
Tsr1	x		TSR1	x		Associates	Gelperin et al (2001)
Nob1	x		NOB1	x		Endoribonuclease, associates with ITS1, catalyzes removal of final part of ITS1 (site D/3), might associate earlier	Fatica et al (2004), Zemp et al (2009) and Ameismeier et al (2018)
Hrr25	x		CKI‐δ/CSNK1D	x		Associates	Schäfer et al (2006) and Zemp et al (2014)
Hrr25	x		CKI‐ε/CSNK1E	x		Associates	Schäfer et al (2006) and Zemp et al (2014)
Rio1	x		RIOK1/RIO1	x		Associates	Rouquette et al (2005) and Widmann et al (2012)
Rio2	x		RIOK2/RIO2	x		Associates	Vanrobays et al (2003) and Zemp et al (2009)
Ltv1	x		LTV1	x		Associates	Seiser et al (2006) and Zemp et al (2014)
			PARN	x		Processes 3′ end of the 18S pre‐rRNA	Montellese et al (2017)
Slx9			FAM207A/C21orf70	x			Wyler et al (2011)

The addition of further factors, including Hrr25/CKIδ/ε and Ltv1/LTV1, into the pre‐40S subunits then results in partial stabilization of the 40S head and neck region (Cheng et al, 2022). The large RBF RRP12, which embraces the immature head region, is central in further steps of head maturation. These entail the sequential joining of a set of ribosomal proteins, RPSA/uS2, RPS2/uS5, and RPS21/eS21, recruitment of Nob1/NOB1, and conformational changes in several rRNA helices (Ameismeier et al, 2018; Cheng et al, 2022). Incorporation of these RPs is followed by the release of Bud23, which occurs in coordination with the recruitment of Rio2 that associates with the decoding center between the platform and the head region, prior to nuclear export (Heuer et al, 2017; Ameismeier et al, 2018; Black & Johnson, 2022; Cheng et al, 2022). Stabilization of the head region thus precedes nuclear export of pre‐40S subunits, while beak formation is completed later in the cytoplasm.

Of the RBFs present in these 40S precursors, several are early‐associating RBFs that remain bound throughout maturation in the nucleoplasm and even accompany 40S precursors to the cytoplasm, namely Enp1/BYSL, Dim2/PNO1, and RRP12 (Table 6). One notable early‐binding RBF is the methyltransferase Dim1/DIM1, which remains particle‐bound until it methylates two adenosines in the 3′ region of the 18S rRNA (Lafontaine et al, 1994). While this occurs during cytoplasmic maturation steps in yeast, human DIM1 performs the corresponding modification earlier, in the nucleus, where it dissociates from maturing particles before they are eventually exported to the cytoplasm (Wyler et al, 2011; Zorbas et al, 2015).

Compared to yeast, the processing of the ITS1 is more complex in higher eukaryotes and involves an additional endonucleolytic cleavage step, generating the 21S pre‐rRNA (Fig 3). Moreover, 21S pre‐rRNA maturation also includes exonucleases, namely the exosome with its catalytic subunits DIS3 and EXOSC10, and the poly(A)‐specific ribonuclease PARN (Preti et al, 2013; Tafforeau et al, 2013; Sloan et al, 2013b; Montellese et al, 2017).

The association of several conserved RBFs, including Nob1/NOB1, Rio2/RIOK2, Ltv1/LTV1, Slx9/C21orf70/FAM207A, and Tsr1/TSR1 (Table 7) with 40S precursors during nuclear 40S maturation paves the way for nuclear export (Ferreira‐Cerca et al, 2005, 2007; Schäfer et al, 2006; Carron et al, 2011; Wyler et al, 2011; Zemp et al, 2014; Heuer et al, 2017; Ameismeier et al, 2018; Cheng et al, 2022). Pre‐ribosomal particles are among the largest transport cargos that pass through nuclear pore complexes (NPCs) and they need to be bound by multiple nuclear export receptors for NPC passage (Table 8). Binding of export factors to ribosomal pre‐particles is thought to function as a quality control step of subunit maturation (Johnson et al, 2002; Woolford & Baserga, 2013). In both yeast and mammalian cells, export of pre‐40S subunits depends on the RanGTP‐binding exportin Crm1/XPO1 (Hurt et al, 1999; Moy & Silver, 1999; Thomas & Kutay, 2003; Wild et al, 2010). The yeast 40S RBFs Dim2, Rio2, and Ltv1 contain nuclear export sequences (NES) and have been suggested to serve as redundant adaptors for Crm1 binding to the small subunit (Schäfer et al, 2003; Vanrobays et al, 2003, 2008; Seiser et al, 2006; Merwin et al, 2014). Further factors linked to pre‐40S export in yeast are the mRNA export factor Mex67/Mtr2 (Faza et al, 2012), Rrp12 (Oeffinger et al, 2004), and Slx9, the latter with a proposed role in mediating Crm1 binding to Rio2 (Fischer et al, 2015). Its human homolog SLX9/C21orf70/FAM207A is also associated with nucleoplasmic pre‐40S particles (Wyler et al, 2011), but is not known to accompany pre‐40S subunits into the cytoplasm. Like in yeast, pre‐40S export in human cells also exploits the atypical protein kinase RIOK2 as an adaptor for XPO1 (Zemp et al, 2009). Other NES‐containing RBFs are expected to contribute to the recruitment of XPO1 (Zemp et al, 2009), among them PDCD2L, which was suggested to support pre‐40S export in mammals (Landry‐Voyer et al, 2016).

Table 8.

Factors involved in nuclear export of ribosomal pre‐particles.

Yeast	40S	60S	Human	40S	60S	Function in nuclear export	Citation
Crm1/Xpo1	x	x	XPO1/CRM1	x	x	Facilitates export in RanGTP‐dependent manner	Hurt et al (1999) and Thomas and Kutay (2003)
Dim2/Pno1	x		PNO1/DIM2	x		Possible adaptor for XPO1‐mediated export	Vanrobays et al (2008)
Rio2	x		RIOK2	x		Possible adaptor for XPO1‐mediated export	Vanrobays et al (2003) and Zemp et al (2014)
Ltv1	x		LTV1	x		Possible adaptor for XPO1‐mediated export	Seiser et al (2006)
			PDCD2L	x		Possible adaptor for XPO1‐mediated export	Landry‐Voyer et al (2016)
Yrb2	x		RANBP3	x			Stage‐Zimmermann et al (2000) and Badertscher et al (2015)
Slx9	x		FAM207A/C21orf70				Fischer et al (2015)
Rrp12	x	x	RRP12			Potential function in export, interaction with FG repeat nucleoporins in vitro	Oeffinger et al (2004)
Mex67	x	x	NXF1			Mex67‐Mtr2 complex Export receptor	Yao et al (2008, 2007)
Mtr2	x	x	NXT1
Nmd3		x	NMD3		x	Adaptor for XPO1‐mediated export	Ho et al (2000) and Thomas and Kutay (2003)
Arx1		x	PA2G4/EBP1			Induces structural changes allowing export	Bradatsch et al (2007)
Bud20		x	ZNF593			Binds FG repeats	Altvater et al (2012)
Ecm1		x				Binds FG repeats	Yao et al (2010)
Npl3/Nop3		x	SRSF1			Binds FG repeats	Hackmann et al (2011)
Gle2/Rae1		x	RAE1	x	x	Binds to Nup116 and recruits pre‐60S via second binding site	Wild et al (2010) and Occhipinti et al (2013)
Msn5			XPO5		x	Facilitates export in RanGTP‐dependent manner	Wild et al (2010)

Cytoplasmic steps of 40S subunit biogenesis

In the cytoplasm, several final maturation steps occur on both ribosomal pre‐particles, including incorporation of late‐assembling RPs, structural rearrangements, final pre‐rRNA processing steps, and the release of late‐acting RBFs (Zemp & Kutay, 2007; Nerurkar et al, 2015). Several RBFs accompany these particles from the nucleus to the cytoplasm and are thought to prevent premature 40S‐60S joining and translation initiation by keeping the particle in an inactive conformation and shielding functional sites on the subunit interface (Greber, 2016).

In yeast, exported pre‐40S particles are bound by the RBFs Dim1, Dim2, Enp1, Nob1, Hrr25, Rio2, Rrp12, and Tsr1 (Schäfer et al, 2006). The function of these factors is widely conserved in mammalian cells, although DIM1 acts earlier and is not part of human late 40S precursors (Table 9). Recently published cryo‐EM structures of cytoplasmic 40S pre‐particles from yeast and human cells have highlighted that cytoplasmic 40S maturation mostly involves structural changes in the head and beak region, formation of the decoding center by rearrangement of helix 44 (h44) and final processing of the 18S pre‐rRNA (Fatica et al, 2003; Lamanna & Karbstein, 2011; Larburu et al, 2016; Heuer et al, 2017; Scaiola et al, 2018; Ameismeier et al, 2020, 2018). These steps are coordinated by the conserved kinases Hrr25/CKIδ/ε, Rio2/RIOK2, and Rio1/RIOK1 (Vanrobays et al, 2003, 2001; Rouquette et al, 2005; Widmann et al, 2012; Ferreira‐Cerca et al, 2014, 2012; Zemp et al, 2014, 2009; Mitterer et al, 2019; Plassart et al, 2021), with an additional kinase, RIOK3, supporting 18S pre‐rRNA processing in mammalian cells (Baumas et al, 2012; Widmann et al, 2012). While Hrr25/CKIδ/ε is thought to phosphorylate and thereby trigger the release of Enp1/BYSL and Ltv1/LTV1 (Schäfer et al, 2006; Zemp et al, 2014), no substrates are known for the RIO kinases. Both Rio1 and Rio2 have been suggested to act as ATPases rather than kinases (Ferreira‐Cerca et al, 2012, 2014), with respective conformational changes regulating their association with 40S precursors. It is still unclear whether the kinase activity of Rio kinases is indeed exploited for structural remodeling of pre‐40S particles as originally suggested (Ferreira‐Cerca et al, 2012).

Table 9.

Factors involved in cytoplasmic pre‐40S maturation.

Yeast	40S	60S	Human	40S	60S	Function in cytoplasmic pre‐40S maturation	Citation
Enp1	x		ENP1/BYSL	x		Released by Hrr25/CKI phosphorylation, Release allows stable incorporation of Rps3/uS3, Rps10/eS10 and Rps20/uS10, Formation of the beak structure	Schäfer et al (2006), Zemp et al (2014) and Ameismeier et al (2018)
Ltv1	x		LTV1	x
Tsr1	x		TSR1	x		Released, occludes binding sites for mRNA and translation initiation factors	Larburu et al (2016) and Scaiola et al (2018)
Dim1	x		DIMT1	x		Methyltransferase, yeast: modifies subsequent adenines near 3′ end of 18S rRNA; human: function occurs already in the nucleus	Strunk et al (2011) and Zorbas et al (2015)
Rrp12	x		RRP12	x		Released, which allows incorporation of Asc1 (RACK1 in humans)	Wyler et al (2011) and Larburu et al (2016)
Dim2/Pno1	x		PNO1/DIM2	x		Inhibits binding of eIF3; release contributes to Major structural rearrangements, allowing Nob1 activity	Ameismeier et al (2018) and Scaiola et al (2018)
Nob1	x		NOB1	x		Endonucleolytic cleavage of 20S pre‐rRNA (18S‐E pre‐rRNA in humans) to yield mature 18S rRNA	Fatica et al (2003) and Lamanna and Karbstein (2011)
Hrr25	x		CKI‐δ/CSNK1D	x		Phosphorylation of Enp1‐Ltv1‐Rps3/uS3 complex, which triggers release of Ltv1	Schäfer et al (2006) and Zemp et al (2014)
			CKI‐ε/CSNK1E	x
Rio1	x		RIOK1/RIO1	x		Orchestrates structural changes, pre‐rRNA maturation and trans‐acting factor release	Vanrobays et al (2003) and Widmann et al (2012)
Rio2	x		RIOK2/RIO2	x		Kinase contributes to pre‐rRNA maturation and trans‐acting factor release	Geerlings et al (2003) and Zemp et al (2009)
			RIOK3	x		Promotes 18S‐E processing	Geerlings et al (2003)
Prp43	x		DHX15			Conformational pre‐rRNA switch, which allows Nob1 activity	Pertschy et al (2009)
Pfa1/Sqs1	x		SON	x		Cofactor for Prp43	Sloan et al (2013b)
Pxr1/Gno1	x		PINX1	x		Cofactor for Prp43	Guglielmi and Werner (2002) and Chen et al (2014)
Hbs1	x		HBS1			In yeast: facilitates formation of 80S like particle	Lebaron et al (2012), Strunk et al (2012) and Hector et al (2014)
Dom34	x		PELO
Rli1	x		ABCE1
Fun12	x		EIF5B
			USP16	x		Deubiquitylates lysine 113 in RPS27A/eS31	Montellese et al (2020)
			LRRC47	x		Bound to late cytoplasmic pre‐40S particle	Ameismeier et al (2020)
			EIF1AD	x		Supports trans‐acting factor release and pre‐rRNA maturation	Ameismeier et al (2020)

Intriguingly, the endonuclease Nob1/NOB1, which mediates the final cleavage of 20S pre‐rRNA (18S‐E in humans) to mature 18S rRNA, is already associated with nuclear pre‐40S particles. However, its access to the cleavage site at the 3′ end of 18S rRNA is restricted by Dim2/PNO1, thereby preventing premature removal of the remaining ITS1 fragment (Fig 5) (Turowski et al, 2014; Scaiola et al, 2018; Ameismeier et al, 2020). It has been suggested that in yeast, the formation of mature 18S rRNA is supported by the interaction of pre‐40S particles with mature 60S particles forming an 80S‐like complex, stimulated by the translation initiation factor Fun12/eIF5B (Lebaron et al, 2012; Strunk et al, 2012). The formation of an 80S‐like particle has not been described for human cells. However, two additional, human‐specific cytoplasmic RBFs were observed on late particles, namely EIF1AD and LRRC47 (Ameismeier et al, 2020; Montellese et al, 2020; Plassart et al, 2021). While LRRC47 associates with the subunit interface and might prevent premature 60S joining, binding of EIF1AD leads to a series of events, including repositioning of RIOK1 and the central helix h44, triggering PNO1 release and final pre‐rRNA processing by NOB1. These final maturation steps, entailing incorporation of RPS26/eS26, ATP hydrolysis on RIOK1, and dissociation of the few remaining RBFs, renders 40S subunit competent for 60S joining and mRNA translation (Plassart et al, 2021). The deubiquitinase USP16, which deubiquitylates RPS27A/eS31 in a translation‐dependent manner, supports these last maturation events in a not yet fully understood fashion, potentially linking surveillance of subunit maturation to translation initiation (Montellese et al, 2020).

Figure 5. Overview of late maturation steps of the small ribosomal subunit.

Front and back views of human pre‐40S particles at different stages of cytoplasmic maturation as derived from cryo‐EM analyses (PDB ID: 6G4S, 6G18, 6ZXE, 6ZXF, 6ZXH). Factors involved in these steps are color‐coded and pre‐rRNA is shown in white. After RRP12 release from the state B particle, RACK1 occupies its place and the pre‐rRNA is rearranged for head formation. PNO1 directly interacts with NOB1 and keeps it in an inactive state (from state B to F2). Association of EIF1AD, concomitant with rearrangements of RIOK1, triggers PNO1 dissociation, RPS26/eS26 incorporation, and final pre‐rRNA processing. LRRC47 association prevents 60S joining until the mature decoding region is formed.

Nucleolar pre‐60S biogenesis

After the subunit separating cleavage in the ITS1, the 60S precursor matures independently of the 40S subunit. At this point, the pre‐60S subunit contains the 27S (yeast) or 32S pre‐rRNA (human) that comprises both the 5.8S and 25S/28S rRNA portions (Fig 3) (Kater et al, 2017). Based on conserved secondary and tertiary structures, the 25/28S rRNA is subdivided into six domains, named I to VI from 5′ to 3′, which fold in a hierarchical process (Gamalinda et al, 2014; Klinge & Woolford, 2019). While the two most 5′ domains of the 25S rRNA were observed to fold first (Zhou et al, 2019a), later steps of 60S domain formation are more complex and intertwined as they do not follow the order of domain transcription, as illustrated by structural snapshots of several pre‐60S particles (Wu et al, 2016; Kater et al, 2017, 2020; Sanghai et al, 2018; Kargas et al, 2019; Zhou et al, 2019a, 2019b). Cryo‐EM structures of the first nucleolar pre‐60S particles could not be obtained so far, presumably due to the high flexibility and heterogeneity of these complexes, in which RNA–RNA and RNA‐protein interactions are gradually established (Burlacu et al, 2017; Pöll et al, 2017). In a recent study, an early pre‐60S was visualized in a bipartite structure with a 90S particle, but the existence of this intermediate needs to be confirmed in wild‐type cells (Ismail et al, 2022). The earliest visualized yeast particles reveal that 60S biogenesis first yields the solvent exposed subunit surface, initiated by folding of the 25S domains I and II (Zhou et al, 2019a). This pre‐particle is stabilized by several RBFs, including the Nsa1 module at the solvent‐exposed side. In intermediate pre‐60S particles, folding and positioning of domain VI toward the forming core can be observed, as well as folding of the pre‐5.8S rRNA, while domains III‐V remain too flexible to be resolved (Kater et al, 2017; Sanghai et al, 2018; Zhou et al, 2019a). In later stages of nuclear pre‐60S maturation (Fig 6), domains III, IV, and V, progressively fold and are positioned in the context of the maturing particle, leading to the gradual formation of PET, PTC, and the 60S/40S subunit interface (Barrio‐Garcia et al, 2016; Wu et al, 2016; Kater et al, 2017; Ma et al, 2017; Malyutin et al, 2017; Zhou et al, 2019b).

Figure 6. Overview of key nuclear maturation events of the large ribosomal subunit in yeast.

Cryo‐EM structures of yeast pre‐60S particles at different stages of maturation (PDB ID: 6ELZ, 3JCT, 6YLG, 6N8J). RBFs involved in these steps are color‐coded, as well as the pre‐rRNA (white), L1 stalk (red), and 5S rRNA (orange). The nucleolar pre‐60S in state E shows a displacement of the L1 stalk from its position in the mature subunit. The successive nucleoplasmic Arx/Nog2 particle is a result of a stepwise release and binding of the indicated RBFs and the 5S RNP. The Rix‐Rea1 remodeling machinery initiates the formation of the central protuberance (CP) and rotation of the 5S RNP to its mature conformation, visible in the successive late nuclear (LN) particle.

Pre‐rRNA folding, compaction, and cleavage steps, as well as incorporation of RPs, are supported by a number of 60S‐specific RBFs (Table 10). Many of these RBFs join the particle already early in nucleoli and the number of associated factors steadily decreases as 60S subunits mature on their way to the cytoplasm (Nerurkar et al, 2015). Importantly, directionality of the 60S assembly process is ensured by the activity of energy‐dependent RBFs, for example, ATP‐dependent RNA helicases (e.g., Has1/DDX18 and Dpb10/DDX54), AAA‐ATPases (e.g., Rix7/NVL2 and Rea1/MDN1) and GTPases (including Nug1/GNL3 and Nog1/GTPBP4) (Table 10) (Nissan et al, 2002; Bernstein et al, 2006; Ulbrich et al, 2009; Baßler et al, 2010; Kressler et al, 2010, 2008; Wild et al, 2010; Kappel et al, 2012; Dembowski et al, 2013; Matsuo et al, 2014; Manikas et al, 2016; Zhang et al, 2016b; Hiraishi et al, 2018; Klinge & Woolford, 2019). Several other RBFs involved in 60S biogenesis contain multiple RNA binding motifs, which likely provide structural support and reduce the conformational freedom of rRNA during folding and compaction. Importantly, correct binding and positioning of RPs also critically contribute to correct pre‐rRNA folding in the maturing 60S particle (de la Cruz et al, 2015; Pöll et al, 2021, 2009). This is exemplified by the largest Rpl, Rpl3/uL3, which binds very early during pre‐60S biogenesis and spans several rRNA domains (Ben‐Shem et al, 2011; de la Cruz et al, 2015). Rpl3 stabilizes the interaction between the 5′ and 3′ end of the 25S rRNA and its binding is prerequisite for the incorporation of most other Rpls (Pöll et al, 2009; Ohmayer et al, 2013; Gamalinda et al, 2014; de la Cruz et al, 2015).

Table 10.

Factors involved in nucleolar steps of 60S maturation.

Yeast	40S	60S	Human	40S	60S	Function in nucleolar steps of pre‐60S maturation	Citation
Rrp5	x	x	PDCD11	x		Part of Rrp5‐Noc1‐Noc2 complex, supports early steps of pre‐rRNA Compaction, Rrp5 contains multiple RNA binding motifs	Hierlmeier et al (2013)
Noc1		x	CEBPZ		x
Noc2/Rix3		x	NOC2L/NIR		x
Npa1/Urb1		x	URB1			Npa1‐Npa2‐Nop8‐Rsa3‐Dbp6 complex, organization of early pre‐rRNA compaction steps, Dbp6/DDX51 is a DEAD box RNA helicase DDX51: 28S 3′ end maturation, release of U8 snoRNA (only in metazoans)	Rosado et al (2007) and Srivastava et al (2010)
Npa2/Urb2		x	URB2
Nop8		x
Rsa3		x
Dbp6		x	DDX51		x
Dbp2		x	DDX5		x	DEAD box RNA helicases, involved in early pre‐rRNA remodeling steps, including release of snoRNPs (Dbp3)	Bond et al (2001) and Saporita et al (2011)
Dbp3		x					Aquino et al (2021)
Dbp7		x	DDX31		x		Bernstein et al (2006)
Dbp9		x					Bernstein et al (2006)
Mak5		x	DDX24		x		Bernstein et al (2006)
Nsa1		x	WDR74		x	Nsa1 module, stabilizes solvent exposed side, bridges 25S domains I and II, Rpf1 protrudes into PET	Kater et al (2017) and Lo et al (2017)
Rpf1		x	RPF1
Mak16		x	MAK16		x
Rrp1		x	RRP1		x
Nop4		x	RBM28			Binds 5′ end of 5.8S rRNA	Granneman et al (2011)
Puf6		x	PUM3			Chaperone for Rpl43/eL43, aids 7S processing, interacts with H63 (25S rRNA), required for export of 60S at low temperature	Liang et al (2019)
Loc1		x				Chaperone for Rpl43/eL43	Liang et al (2019)
Rrp15		x	RRP15	x	x	Rrp15‐Ssf1 complex, Ssf1 and Ssf2 are 94% identical	Fatica et al (2002) and de Marchis et al (2005)
Ssf1		x	PPAN
Ssf2		x	PPAN
Rrp14		x	SURF6				Oeffinger et al (2007)
			METTL18		x	Methylation of RPL3/uL3 His245	Małecki et al (2021)
“A3”‐factors, involved in A3 site processing and ITS1 trimming, incorporation of Rpl17/uL22, Rpl26/uL24, Rpl35/uL29, Rpl37/eL37, recruitment of Rrp17
Nop7		x	PES1		x	Erb1‐Ytm1‐Nop7/PeBoW complex, Stabilizes early fold of 5.8S and domain I of 25S rRNA; important for structuring of PTC and PET, DDX27 also functions independently in 47S 3′ end formation, Drs1 is not associated with Erb1‐Ytm1‐Nop7 complex	Rohrmoser et al (2007), Kellner et al (2015) and Konikkat et al (2017)
Erb1		x	BOP1		x
Ytm1		x	WDR12		x
Drs1		x	DDX27		x
Nop15		x	MKI67IP		x	Nop15‐Rlp7‐Cic complex, Supports folding of ITS2, part of the foot structure of pre‐60S particles	Sahasranaman et al (2011) and Kater et al (2017)
Rlp7		x	RLP7
Cic1		x
Pwp1		x	PWP1			Important for 5.8S folding, ubiquitylated by CRL4VPRBP	Talkish et al (2014) and Han et al (2020)
Nop12		x	RBM34			Important for 5.8S folding	Talkish et al (2014)
Has1		x	DDX18			DEAD‐box RNA helicase, binding triggers trimming of 5′ end of 5.8S pre‐rRNA, important for 25S domain I folding, facilitates incorporation of Rpl17/uL22 and other PET forming proteins, also: linked to release of U14 snoRNA	Dembowski et al (2013)
Brx1		x	BRIX1		x	Brx1‐Ebp2 complex, prevents premature RNA–RNA interactions of domains I and V, important for PET formation	Sanghai et al (2018)
Ebp2		x	EBNA1BP2		x
“B” factors, required for endonucleolytic cleavage at site C2 by Las1
Nip7		x	NIP7	x		Nip7‐Nop2 complex, aids formation of PTC, recruits Rpf2‐Rrs1, Nop2 is a methyltransferase	Morello et al (2011b) and Kater et al (2017)
Nop2		x	NOP2
Rrs1		x	RRS1		x	Rpf2‐Rrs1 complex, facilitates incorporation of 5S RNP	Wu et al (2016)
Rpf2		x	BXDC1		x
Spb4		x	DDX55		x	Binds domain IV of 28S rRNA	Wu et al (2016)
Mak11		x	PAK1IP1		x		Manikas et al (2016)
Dbp10		x	DDX54			Binds h89, role in PTC formation	Bernstein et al (2006) and Manikas et al (2016)
Nug1		x	GNL3		x	GTPase, required for Dbp10 binding	Manikas et al (2016)
Rlp24		x	RLP24/RSL24D1			Placeholder for Rpl24/eL24, recruits and activates Drg1	Kappel et al (2012)
Tif6		x	EIF6		x		Basu et al (2001)
Nog1		x	GTPBP4		x	GTPase, proofreading/maturation of PET	Wu et al (2016) and Liang et al (2020)
Nsa2		x	NSA2		x		Talkish et al (2012)
Rsa4		x	NLE1				de la Cruz et al (2005)
Pol5	x	x	MYBBP1A	x		Binds to domain III of 25S rRNA, important for PET formation	Braun et al (2020)
Noc3		x	NOC3L				Milkereit et al (2001)
Other factors
Nog2		x	GNL2		x		Dembowski et al (2013)
Nop16		x	NOP16		x		Pratte et al (2013)
Nop53		x	GLTSCR2/NOP53		x	Binds to similar position as Erb1, after Erb1 is released, recruits nuclear exosome by Mtr4possibly structural role of Nop53 rearranging and stabilizing the foot interface	Falk et al (2017), Sanghai et al (2018) and Bagatelli et al (2021)
Spb1		x	FTSJ3	x		Methyltransferase, involved in PTC formation	Kater et al (2017)
Mrt4		x	MRT4/MRTO4		x	Structural placeholder for P stalk	Rodríguez‐Mateos et al (2009)
Mtq2		x	N6AMT1			Acts together with its cofactor Trm112	Lacoux et al (2020)
YBL028C		x	LLPH		x		Kater et al (2017) and Liang et al (2020)
Bud20		x	ZNF593			Nuclear export factor	Altvater et al (2012)
Arx1		x	PA2G4/EBP1		x	Binds at PET exit, suggested to proofread PET exit region, supports structural changes allowing nuclear export	Greber et al (2012)
Rix7		x	NVL2		x	AAA‐ATPase, removes Nsa1 (+associated factors), which together with release of Erb1‐Ytm1 allows Las1 cleavage and transition of the particle to the nucleoplasm, NVL2 was shown to be associated with Mtr4 (exosome) and WDR74	Kressler et al (2008) and Hiraishi et al (2018)
Rea1/Mdn1		x	MDN1		x	AAA‐ATPase, releases Erb1‐Ytm1 complex, binds again at a later stage of pre‐60S maturation	Baßler et al (2010) and Wild et al (2010)
Las1		x	LAS1L		x	Endoribonuclease C2 cleavage	Schillewaert et al (2012)
Grc3		x	NOL9			Polynucleotide kinase, phosphorylation of ITS2 cleavage product, forms constitutive complex with Las1 in yeast	Pillon et al (2017)

In general, DExD/H‐box ATPases facilitate ribosome biogenesis by unwinding snoRNA‐pre‐rRNA base pairs and remodeling of RNA–RNA and protein‐RNA interactions, thereby supporting major structural rearrangements in the forming subunits as well as pre‐rRNA folding (Dembowski et al, 2013; Rodríguez‐Galán et al, 2013; Martin et al, 2014; Khoshnevis et al, 2016; Brüning et al, 2018). Seven DExD/H‐box ATPases have been implicated in the first steps of 60S maturation (Table 10) (reviewed in Mitterer & Pertschy, 2022). Recently, it was shown that absence of yeast Dbp3 and Prp43 results in a drastic reduction of rRNA modifications (Aquino et al, 2021; Bailey et al, 2022), and several snoRNPs accumulate on early 60S pre‐particles (Leeds et al, 2006; Bohnsack et al, 2009; Aquino et al, 2021). Similarly, DDX51 facilitates release of U8 snoRNA (Srivastava et al, 2010), whereas Dbp7 was proposed to regulate the association of snR190, a snoRNA that structurally inhibits aberrant folding of 25S rRNA (Jaafar et al, 2021a).

Pre‐rRNA processing during 60S maturation commences with removal of the remaining ITS1 spacer at the 5′ end of 27S/32S pre‐rRNAs, before cleavage in the ITS2 (at site C2/4) separates the pre‐5.8S and pre‐25/28S rRNAs (Lygerou et al, 1996; Rouquette et al, 2005; Schillewaert et al, 2012; Gasse et al, 2015; Pillon et al, 2017). Removal of the ITS1 spacer is initiated by endonucleolytic cleavage at site A₃ in yeast, while it is removed solely exonucleolytically in human cells (Tomecki et al, 2017). Depletion experiments revealed that a set of 12 yeast RBFs (Ytm1, Erb1, Nop7, Rlp7, Cic1, Nop15, Has1, Drs1, Rpf1, Pwp1, Nop12, and Rrp1) is required for A₃ site processing, and they were consequently termed A₃‐factors (Table 10) (Merl et al, 2010; Granneman et al, 2011; Sahasranaman et al, 2011; Shimoji et al, 2012; Dembowski et al, 2013; Woolford & Baserga, 2013; Talkish et al, 2014). Structural data showed that the binding sites of most of these factors are not located in proximity of the ITS1 fragment, indicating that they are rather needed as structural components for correct folding and configuration of a pre‐60S assembly intermediate to render it amenable to A₃ cleavage (Sahasranaman et al, 2011; Woolford & Baserga, 2013; Kater et al, 2017; Konikkat & Woolford, 2017; Sanghai et al, 2018; Zhou et al, 2019a). Interestingly, A₃ cleavage was found to be coupled with termination of RNAPI transcription and 3′ ETS processing, suggesting a link between 60S assembly and transcription, yet the underlying molecular mechanism remains to be deciphered (Allmang & Tollervey, 1998; Lebaron et al, 2012; Gamalinda et al, 2014; Chen et al, 2017).

After ITS1 removal, pre‐rRNA processing is focused on the separation of the 27SB/32S precursor into the 7/12S and 26/28.5S pre‐rRNAs by cleavage in the ITS2. More than a dozen RBFs, originally termed “B‐factors,” are required to prepare the particle for this processing step and many of them are bound to the subunit interface (Talkish et al, 2012; Woolford & Baserga, 2013). Interestingly, some B‐factors, including Nip7 and Nop2, are already part of the 90S particle, before endonucleolytic separation of the two subunits occurs (Kater et al, 2017). B‐factors contribute to the construction of the PET (e.g., Nog1; Fuentes et al, 2007; Wu et al, 2016) and the assembly of the PTC (e.g., Nsa2, Dpb10, Nug1, and Rsa4; Baßler et al, 2014; Matsuo et al, 2014; Barrio‐Garcia et al, 2016; Wu et al, 2016).

Before ITS2 cleavage can take place, pre‐60S subunits undergo major structural rearrangements driven by the AAA‐ATPases Rix7/NVL2 and Rea1/MDN1. First, Rix7 releases Nsa1, which leaves the particle together with Rpf1, Rrp1, and Mak16 (Saveanu et al, 2003; Kressler et al, 2008; Lo et al, 2017, 2019). This allows the formation of the outer part of the PET (Kater et al, 2017; Sanghai et al, 2018; Zhou et al, 2019a). Rix7 activity is then followed by ATP hydrolysis by Rea1. This giant 550 kDa protein removes the Erb1‐Ytm1 subcomplex (Fig 6), which interacts with many RBFs covering the yet to form intersubunit surface (Baßler et al, 2010, 2014; Thoms et al, 2016; Kater et al, 2017; Chen et al, 2018; Ahmed et al, 2019). The MIDAS domain of Rea1 was biochemically shown to bind the UBL domain of Ytm1, yet a structure illustrating the binding of Rea1 to Ytm1 is still missing. Erb1 likely stabilizes the premature architecture at this stage with its N‐terminus deeply embedded in the particle (Kater et al, 2017; Prattes et al, 2019). Its active removal by Rea1 presumably contributes to the major structural rearrangements and compositional changes observed during particle transition from the nucleolus to the nucleoplasm. Release of Erb1 also allows the recruitment of the GTPases Nog2 just before ITS2 cleavage and transition to the nucleoplasm (Talkish et al, 2012; Fromm et al, 2017; Biedka et al, 2018). Finally, cleavage at site C₂/4 within ITS2 is mediated by the conserved endonuclease Las1/LAS1, which functions in a complex with the polynucleotide kinase Grc3, together called RNase PNK (Schillewaert et al, 2012; Castle et al, 2013; Gasse et al, 2015; Pillon et al, 2017; Frazier et al, 2021).

The release of pre‐particles from nucleoli could be governed by the state of pre‐rRNA compaction and processing. According to this model, pre‐ribosomal particles that still expose binding sites for general nucleolar RNA chaperones such as nucleophosmin (NPM) (Szebeni & Olson, 1999; Box et al, 2016) or nucleolin (NCL) (Mongelard & Bouvet, 2007) (Table 14) would remain partitioned in the nucleolar “phase” by low‐affinity interactions of exposed RNA segments with these multivalent RNA chaperones. Once the pre‐rRNAs in the emerging subunits have been sufficiently compacted, processed, and covered by RPs and RBFs, they would no longer be retained and released into the nucleoplasm. Along these lines, in vitro experiments showed a preferential partitioning of protein‐free bacterial rRNA but not mature ribosomal subunits into droplets formed by human NPM (Riback et al, 2020). Furthermore, a similar model has recently been proposed based on bioinformatic analyses of cryo‐EM structures, which revealed that earlier nucleolar ribosomal precursors contain more unstructured rRNA regions as well as RBFs with predicted intrinsically disordered regions compared to nucleoplasmic subunit assembly intermediates, and both these elements could contribute to nucleolar retention of subunits (LaPeruta et al, 2022).

Table 14.

Other factors involved in ribosome biogenesis.

Yeast	40S	60S	Human	40S	60S	Function in rRNA transcription	Citation
			NPM	x	x	Nuclear RNA‐binding protein and chaperone	Szebeni and Olson (1999)
Nsr1	x	x	NCL	x	x	Nuclear RNA‐binding protein and chaperone	Mongelard and Bouvet (2007)
			DDX1				Suzuki et al (2021)
			HECTD1			Ubiquitylates ZNF622	Lv et al (2021)
			USP36			Promotes FAU processing and snoRNP maturation	Ryu et al (2021) and van den Heuvel et al (2021)
			UBR5	x		Supports rRNA maturation by regulation of H/ACA RNPs	Saez et al (2020)
			VPRBP	x		Ubiquitylates PWP1	Han et al (2020)

Formation of the 5S RNP and its incorporation into pre‐60S subunits

During its maturation, the pre‐60S subunit must incorporate the mature 5S rRNA. The 5S rRNA first associates with its partner RPs Rpl5/uL18 and Rpl11/uL5, which are co‐imported into the nucleus with help of the RP chaperone Syo1/HEATR3 which also serves as a platform for 5S RNP assembly (Kressler et al, 2012; Calviño et al, 2015). The 5S RNP is then bound by pre‐60S particles in the nucleolus aided by the associated Rpf2‐Rrs1 complex (Table 11) (Wu et al, 2016). In mature 60S subunits, the 5S RNP forms the central protuberance, yet the 5S RNP is not immediately placed in its final position in the pre‐60S, but rotated by about 180°, as revealed by structures of Arx1/Nog2 pre‐60S particles from yeast (Leidig et al, 2014; Wu et al, 2016).

Table 11.

Factors involved in 5S RNP formation and incorporation.

Yeast	40S	60S	Human	40S	60S	Function in 5S RNP maturation	Citation
Rex1		x	REXO5			Exoribonuclease, 5S pre‐rRNA trimming	van Hoof et al (2000)
Rex2		x	REXO2			Exoribonuclease, 5S pre‐rRNA trimming	van Hoof et al (2000)
Rex3		x	REXO1			Exoribonuclease, 5S pre‐rRNA trimming	van Hoof et al (2000)
La		x				Binds immature 5S pre‐rRNA	Madru et al (2015)
TFIIIA		x	TFIIIA		x	Transcription factor for 5S DNA locus, but also binds 5S rRNA	Layat et al (2013) and Sloan et al (2013b)
Syo1		x	HEATR3		x	Chaperone for Rpl5/uL18 and Rpl11/uL5	Kressler et al (2012), Hannan et al (2022) and O'Donohue et al (2022)
Rrs1		x	RRS1		x	Rpf2‐Rrs1 complex, aids incorporation of 5S RNP into mature ribosomes, in yeast also important for nucleolar localization of 5S RNP	Wu et al (2016)
Rpf2		x	BXDC1		x

In mammalian cells, the incorporation of the 5S RNP into the nascent 60S subunit serves as an important checkpoint to relay defects in nucleolar ribosome synthesis into the p53 pathway, referred to as the “nucleolar stress response” (reviewed in Chakraborty et al, 2011; Bohnsack & Bohnsack, 2019). If nucleolar 60S maturation is perturbed, the unincorporated 5S RNP accumulates, to then bind and inhibit the p53 E3 ubiquitin ligase MDM2, leading to p53 stabilization and cell cycle arrest (Donati et al, 2013; Sloan et al, 2013a). This 5S RNP‐dependent mechanism has emerged as a key nuclear stress response pathway that reacts to a broad range of insults ranging from diverse DNA damaging insults, proteasome inactivation to nuclear export inhibition (Hannan et al, 2022), putting the 5S RNP and ribosome biogenesis defects into the center of nuclear stress sensing.

Nucleoplasmic assembly steps and export of pre‐60S subunits

The transition of pre‐60S particles from the nucleolus to the nucleoplasm is accompanied by the release and binding of a significant number of RBFs, resulting in substantial compositional and structural changes in the emerging subunits (Fig 6) (Kater et al, 2017, 2020; Sanghai et al, 2018). These include the repositioning of the L1 stalk, which is first delocalized and then accommodated into its mature position following the release of Spb1 (Kater et al, 2020). Furthermore, 2′O‐methylation of G2922 in the PTC A‐site loop by Spb1 was suggested to be prerequisite for stable binding of Nog2 to H92 (Kressler et al, 1999; Lapeyre & Purushothaman, 2004; preprint: Yelland et al, 2022). Finally, the association of Arx1, Cgr1, and Rsa4 to the forming subunit interface complete the well‐characterized Arx1/Nog2 particle (Fig 6) (Bradatsch et al, 2012; Leidig et al, 2014; Wu et al, 2016).

The L1 stalk together with the RBF Sda1 provides a binding platform for the Rix1 complex (Table 12) (Barrio‐Garcia et al, 2016; Wu et al, 2016; Kater et al, 2020). Association of Rix1‐Ipi3‐Ipi1 initiates the formation of the central protuberance as it recruits the ATPase Rea1 for a second round of action (Fig 6). Rea1 then catalyzes the release of the Rpf2‐Rrs1 complex powered by ATP hydrolysis (Baßler et al, 2010; Matsuo et al, 2014; Barrio‐Garcia et al, 2016). This leads to a 180° rotation of the 5S RNP and accommodation in its final position, coupled to maturation of PET and PTC (Micic et al, 2020). Moreover, the RBF Nop53 joins the pre‐60S subunits (Falk et al, 2017; Kater et al, 2020) and binds the RNA helicase Mtr4, thereby recruiting the nuclear exosome to trim the ITS2 part of the 7S pre‐rRNA (Michael et al, 2018). Elimination of this ITS2 fragment results in the removal of the so‐called foot region visible in cryo‐EM structures of Nog2 particles (Fromm et al, 2017; Zhou et al, 2019a). Coupled to its role in 5S RNP accommodation, Rea1 ATPase activity also triggers the release of Rsa4 (Baßler et al, 2014, 2010), and is prerequisite for GTP hydrolysis‐dependent release of the GTPase Nog2 (Matsuo et al, 2014). The precise molecular mechanism by which the mechano‐chemical force of ATP hydrolysis by Rea1 leads to these structural rearrangements needs to be further investigated, but the liberation of the Nog2 binding site on pre‐60S particles sets the stage for the recruitment of the export adaptor Nmd3. Nmd3 binding serves as a quality control checkpoint probing the correct assembly of the E‐ and P‐sites as well as of the L1 stalk (Sengupta et al, 2010; Matsuo et al, 2014; Malyutin et al, 2017).

Table 12.

Factors involved in nucleoplasmic steps of pre‐60S maturation.

Yeast	40S	60S	Human	40S	60S	Function in nucleoplasmic steps of pre‐60S maturation	Citation
Alb1		x				Arx1 binding partner	Greber et al (2012)
Cgr1		x	CCDC86				Thoms et al (2018)
Sda1		x	SDAD1			Sda1 suggested to initiate Rpf2‐Rrs1 release as it partially overlaps with Rpf2 binding site	Klinge and Woolford (2019)
Rix1		x	PELP1			Rix1/PELP1 complex, important for Rea1/MDN1 positioning PELP1 is SUMOylated, which is required for its interaction with MDN1	Finkbeiner et al (2011), Barrio‐Garcia et al (2016) and Gordon et al (2022)
Ipi3		x	TEX10
Ipi1		x	WDR18
Rea1/Mdn1		x	MDN1		x	AAA‐ATPase, rebinds to 60S particles in nucleoplasm, releases Rsa4, triggers GTPase activity and release of Nog2	Baßler et al (2010)
			SENP3		x	SUMO specific protease, activity results in disengagement of MDN1 and PELP1 deSUMOylates NPM	Finkbeiner et al (2011) and Raman et al (2016)
			NF45/ILF2		x	NF45‐NF90 complex	Wandrey et al (2015)
			NF90/ILF3		x
Nmd3		x	NMD3		x	Nuclear export adaptor	Ho et al (2000) and Thomas and Kutay (2003)

Notably, the various RNA expansion segments, located at the solvent‐exposed surface of mature ribosomes (Yusupova & Yusupov, 2014), also play an important role in pre‐rRNA processing and subunit assembly. Studies in yeast have demonstrated that individual deletion of the majority of expansion segments in 25S rRNA leads to 60S biogenesis defects (Jeeninga et al, 1997; Ramesh & Woolford, 2016). Interestingly, certain eukaryotic RPs and their extensions (Ramesh & Woolford, 2016) as well as some RBFs such as Arx1 (Bradatsch et al, 2012), Nop7 (Granneman et al, 2011), Rlp7 (Dembowski et al, 2013), and Rrp5 (Lebaron et al, 2013) make contact to RNA expansion segments (Granneman et al, 2011; Bradatsch et al, 2012; Babiano et al, 2013; Dembowski et al, 2013; Lebaron et al, 2013), guiding models of coevolution of expansion segments with RPs and RBFs (Ramesh & Woolford, 2016).

While nucle(ol)ar pre‐60S maturation in yeast is relatively well understood, only a limited number of studies have addressed this process in human cells (Wild et al, 2010; Finkbeiner et al, 2011; Tafforeau et al, 2013; Wandrey et al, 2015; Dörner et al, 2022). Although the human homologs of many yeast RBFs have been identified (Wild et al, 2010; Tafforeau et al, 2013; Badertscher et al, 2015; Dörner et al, 2022), the functional conservation of most factors remains to be investigated (Table 12). Interestingly, SUMOylation of PELP1, a component of the PELP1‐TEX10‐WDR18 complex (Rix1 complex in yeast), was shown to be essential for recruitment of MDN1 (the human ortholog of yeast Rea1) to the pre‐60S particle (Finkbeiner et al, 2011; Raman et al, 2016). Recently, the structures of two late nuclear human pre‐60S particles already associated with the export factor NMD3 have been described (Liang et al, 2020). Overall, the structures showed similar architecture and composition as yeast particles at similar stages, indicating conservation of the function of bound RBFs. Yet, they also revealed mammalian‐specific features of 60S subunit maturation, for example, interaction of the N‐terminal domain of ZNF622 (Rei1 in yeast) with expansion segment ES27, which is much longer in human cells. Cryo‐EM structures of earlier nucle(ol)ar human pre‐60S particles remain to be solved and will provide additional insights into similarities and differences of the process between yeast and mammals.

Binding of the NES‐containing export adaptor Nmd3/NMD3 licenses pre‐60S subunits for Crm1/XPO1‐dependent nuclear export both in fungi and metazoan cells (Table 8) (Ho et al, 2000; Gadal et al, 2001; Thomas & Kutay, 2003; Trotta et al, 2003). In vertebrate cells, a second RanGTP‐binding exportin, XPO5, was shown to support pre‐60S export in addition (Moy & Silver, 1999; Wild et al, 2010). In yeast, a number of further auxiliary factors facilitate pre‐60S translocation through the NPC by directly interacting with FG‐repeats of nucleoporins, including Rrp12, Bud20, Ecm1, and Npl3 (Oeffinger et al, 2004; Yao et al, 2010; Hackmann et al, 2011; Altvater et al, 2012; Nerurkar et al, 2015) (Table 8).

In addition, the mRNA export receptor Mex67/Mtr2 aids export of both 40S and 60S pre‐particles in yeast, a function that is not conserved in human cells (Yao et al, 2008, 2007). Interestingly, recruitment of Mex67/Mtr2 has been linked to the assembly of the P stalk. Pre‐60S particles initially contain Mrt4, a structural homolog of the P‐stalk protein uL10/P0. As long as Mrt4 is bound, the recruitment of the nuclear export receptor Mex67/Mtr2 is inhibited (Sarkar et al, 2016). Yvh1/DUSP12 then dissociates Mrt4, thereby allowing P‐stalk formation by incorporation of the uL10/P0 (Kemmler et al, 2009; Rodríguez‐Mateos et al, 2009; Lo et al, 2010, 2009; Sarkar et al, 2016; Zhou et al, 2019b; Klingauf‐Nerurkar et al, 2020). It must be noted, however, that the exact timing of P‐stalk assembly is not fully resolved, since Mrt4 can be found on cytoplasmic particles upon expression of dominant‐negative Drg1, a cytoplasmic 60S‐RBF, indicating that exchange may occur later, in the cytoplasm (Klingauf‐Nerurkar et al, 2020).

Cytoplasmic 60S maturation steps

While cytoplasmic pre‐40S subunit maturation is primarily driven by kinases, cytoplasmic maturation of pre‐60S particles relies on the function of GTPases and AAA‐ATPases (Table 13) (Lo et al, 2010). These ensure the timely release of some remaining assembly factors (Fig 7), including the ribosomal‐like protein Rlp24/RSL24D1, the GTPase Nog1/GTPBP4, the export adaptor Nmd3/NMD3, Arx1/PA2G4, and Tif6/EIF6. At the same time, RBFs acting in the cytoplasm drive the incorporation of the last RPs and proofread the assembly state of the functional centers.

Table 13.

Factors involved in cytoplasmic pre‐60S maturation.

Yeast	40S	60S	Human	40S	60S	Function in cytoplasmic pre‐60S maturation	Citation
Drg1/Afg2		x	SPATA5		x	ATPase activity releases Rlp24 and Nog1	Pertschy et al (2007), Kappel et al (2012) and Ni et al (2022)
			SPATA5L		x
			CINP		x
			C1orf109		x
Rlp24		x	RLP24/RSL24D1		x	Released by Drg1, which allows incorporation of Rpl24/eL24 (placeholder)	Kappel et al (2012)
Bud20		x	ZNF593		x	Released	Srivastava et al (2010)
Nug1		x	GNL3		x	Released, timing of release not clear in humans	Altvater et al (2012) and Ma et al (2017)
YBL028C		x	LLPH		x	Released	Klingauf‐Nerurkar et al (2020) and Liang et al (2020)
TMA16		x	TMA16		x	Might be released before export	Liang et al (2020)
Nsa2		x	NSA2			Released	Altvater et al (2012) and Ma et al (2017)
Nog1		x	GTPBP4/NOG1		x	Inserts flexible helix into PET, which could function as proof reading or maturation step, released after/with Rlp24, coordinates bifurcation of pre‐60S maturation pathway	Pertschy et al (2007) and Kappel et al (2012)
Arx1		x	PA2G4/EBP1		x	Prevents binding of proteins and complexes typically engaging with the nascent peptide chain binding near the tunnel exit, released by Rei1 (with Ssa and Jjj), which frees peptide exit tunnel, function in human cells is less clear	Bradatsch et al (2012) and Greber et al (2012)
Alb1		x				Binding partner of Arx1, released	Greber et al (2012)
Rei1		x	ZNF622		x	Inserts flexible helix into PET, which could function as proof reading or maturation step; functions together with the ATPase Ssa and Jjj to release Arx1	Meyer et al (2010), Bradatsch et al (2012) and Greber et al (2012)
Jjj1		x	DNAJC21		x	Cofactor for release of Arx1 by Rei1/Ssa	Demoinet et al (2007) and Greber et al (2012)
Ssa1		x	HSPA1A			Binds together with Rei1 and Jjj1, ATPase Activity releases Arx1	Pertschy et al (2007) and Ma et al (2017)
Ssa2		x	HSPA1B
Reh1		x				Inserts C‐terminal helix into PET	Parnell and Bass (2009)
Mrt4		x	MRT4/MRTO4		x	Placeholder for P stalk, released, which allows incorporation of ribosome stalk, initiated by binding of RplP0/uL10	Kemmler et al (2009) and Lo et al (2009)
Yvh1		x	DUSP12		x	Phosphatase, releases Mrt4	Kemmler et al (2009) and Lo et al (2009)
Efl1/Ria1		x	EFL1/EFTUD1		x	GTPase activity releases Tif6/EIF6, works together with Sdo1/SBDS	Basu et al (2001) and Finch et al (2011)
Sdo1		x	SBDS		x	Works together with Efl1/EFL1	Basu et al (2001) and Finch et al (2011)
Tif6		x	EIF6		x	Anti‐association factor, preventing premature association with 40S subunits, released by Efl1/EFL1 (with Sdo1/SDAD)	Gartmann et al (2010)
Lsg1/Kre35		x	LSG1		x	GTPase activity releases Nmd3/NMD3	Kallstrom et al (2003) and Hedges et al (2005)
Sqt1		x	AAMP		x	Chaperone of Rpl10/uL16	Kallstrom et al (2003) and Hedges et al (2005)
Nmd3		x	NMD3		x	Released by Lsg1/LSG1, allows incorporation of Rpl10/uL16, before release: prevents joining of premature subunits	Kallstrom et al (2003) and Hedges et al (2005)

Figure 7. Cytoplasmic maturation of 60S subunit in yeast.

Cryo‐EM structures of pre‐60S particles illustrating snapshots of cytoplasmic maturation events (PDB ID: 6N8L, 7Z34, 6N8M, 6N8N, 6RZZ, 6QTZ). RBFs are color‐coded, as well as the pre‐rRNA (white), L1 stalk (red), and 5S rRNA (orange). After nuclear export, the ATPase Drg1 dissociates Rpl24 from early cytoplasmic pre‐60S particles. The release of multiple factors allows the association of the GTPase Lsg1, which in turn dissociates Nmd3 upon the incorporation of Rpl10/uL16. Following Reh1 dissociation, the GTPase Efl1 together with Sdo1 triggers the release of Tif6, resulting in mature 60S subunits.

In a first step, the AAA‐ATPase Drg1 releases Rlp24 (Fig 7), which acts as a placeholder for Rpl24/eL24 (Table 13) (Pertschy et al, 2007; Lo et al, 2010; Kappel et al, 2012). While Drg1 is the sole factor known to be responsible for Rlp24 dissociation in yeast, release of the human homolog of Rlp24, RLP24/RSL24D1, was recently reported to involve two Drg1‐related AAA‐ATPases, SPATA5, and SPATA5L1 (Ni et al, 2022). Interestingly, SPATA5 is localized in the cytosol, similar to yeast Drg1 (Puusepp et al, 2018), whereas SPATA5L1 is predominantly nuclear (Richard et al, 2021), suggesting that either RLP24 release can in principle occur in both compartments or that only one of these factors functions directly in RLP24 exchange. In addition to SPATA5 and SPATA5L, RLP24 release was suggested to depend on CINP and C1orf109 (Ni et al, 2022), two structurally related proteins. However, how these factors function together with SPATA5 and SPATA5L1 in RLP24/eL24 exchange remains to be mechanistically defined.

Drg1 not only acts on Rlp24, but also contributes to the dissociation of the GTPase Nog1 and additional factors that bind in close proximity to Rlp24. The GTPase Nog1 is deposited on the pre‐60S subunit already during nucleolar maturation steps and projects its long C‐terminal tail into the PET, almost reaching back to the PTC (Wu et al, 2016). Nog1 is liberated in a two‐step process. The first step exploits its own GTPase activity, driving the dissociation of both its N‐terminal and GTPase domains. In the second step, the C‐terminal part of Nog1 is dissociated by Drg1 (Pertschy et al, 2007; Kappel et al, 2012). After release of Nog1, the PET is further functionally probed and potentially matured by insertion of the C‐terminal domain of Rei1/ZNF622, which is recruited via Rpl24 (Greber et al, 2016, 2012; Kargas et al, 2019; Zhou et al, 2019b). Arx1 is bound close to the PET exit where it sterically impedes the premature loading of nascent chain binding factors. It is set free by Rei1, Jjj1 and the ATPase activity of Ssa1/Ssa2 (Hsp70) (Hung & Johnson, 2006; Lebreton et al, 2006; Meyer et al, 2010; Bradatsch et al, 2012; Greber et al, 2012). Then, the PET is again occluded by the C‐terminal α‐helix of Reh1, which binds inside the PET (Ma et al, 2017).

Cytoplasmic 60S maturation ends with the dissociation of Reh1, Nmd3, and Tif6, which are bound on the subunit interface. These factors are suggested to function as anti‐association factors, inhibiting premature interaction of cytoplasmic pre‐60S particles with mature 40S subunits (Gartmann et al, 2010; Weis et al, 2015; Ma et al, 2017). After incorporation of RPL40/eL40 and RPL10/uL16, Nmd3 is dissociated by the GTPase Lsg1/LSG1 (Fig 7) (Malyutin et al, 2017; Kargas et al, 2019; Zhou et al, 2019b). Finally, Tif6 removal is mediated by the GTPase Efl1/EFTUD1 and its guanine nucleotide exchange factor Sdo1/SBDS (Bécam et al, 2001; Senger et al, 2001; Menne et al, 2007), which have also been suggested to probe functionality of the P stalk, P‐site, and PTC (Ma et al, 2017; Zhou et al, 2019b).

Most factors involved in cytosolic pre‐60S maturation in yeast have human homologs, and recently solved structures of late human pre‐60S subunits revealed that the observed human homologs of yeast RBFs bind to similar positions and are likely functional homologs (Liang et al, 2020). One notable exception is PA2G4/EBP1, which is significantly smaller than its yeast homolog Arx1. PA2G4 has been implicated both in ribosome biogenesis and translation, as it was shown to bind to pre‐60S as well as mature 80S ribosomes in vivo (Liang et al, 2020; Bhaskar et al, 2021; Kraushar et al, 2021). Several structural features of yeast Arx1 that might play a role in 60S biogenesis are missing in PA2G4. Thus, further analysis is needed to address its functional conservation as an RBF.

Altogether, eukaryotic cells exploit several hundred RBFs along the intricate ribosome assembly line from the nucleolus to the cytoplasm (Tables 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14). These RBFs not only promote RP deposition, pre‐rRNA folding, processing, and maturation as well as remodeling of the emerging ribosomal particles, but also probe the correct configuration of the functional centers. Once all RBFs are released, both subunits are ready to fulfill their function in mRNA translation.

Concluding remarks

Over the past years, the field has made impressive progress in deciphering the molecular mechanisms of ribosomal subunit assembly and maturation. Today, we know for the vast majority of RBFs at which step of the complex pathway they function, how they associate with precursor particles, and which task they perform. What initially seemed an overwhelming list of accessory factors, first merely named as players in either 40S or 60S synthesis, has coalesced into an almost coherent molecular picture of the subunit assembly lines. Of course, some pieces of the puzzle are still missing. For instance, while we appreciate the impressive molecular snapshots of the SSU processome, we still do not fully understand how this gigantic RNP, similar in size to a mammalian ribosome, is assembled in the first place. How are structural and compositional remodeling steps of the SSU processome that are associated with pre‐rRNA folding and maturation driven? We also know little about whether and how early nucleolar steps of 40S and 60S biogenesis are coupled before RNA cleavage separates the precursors to both subunits. Likewise, we still miss a structural depiction of some key 60S subunit assembly intermediates.

But not only the assembly process itself still contains some uncharted territory, also other areas related to ribosome synthesis are expected to hide a number of exciting secrets. For instance, it is becoming increasingly clear that early errors in the assembly line affect nucleolar morphology and structure. But how is nucleolar organization governed in first place? And how are maturing particles expelled from nucleoli for further maturation into the nucleoplasm? Some first hints suggest, as discussed above, that pre‐rRNA length and compaction may govern this decisive step. Another almost unexplored area concerns quality control of ribosome synthesis. How are aberrant premature subunits recognized and eliminated? This question especially pertains to the fate of the proteinaceous parts of the precursor particles. Hence, which mechanisms counteract the potential proteotoxicity of unassembled RPs or aberrant precursors, especially in mammalian cells? Clearly, these and many other questions remain to be resolved. Thus, despite the impressive progress made, research on ribosome biogenesis will remain an active and rewarding field of biology in the future.

Author contributions

Kerstin Dörner: Conceptualization; data curation; formal analysis; writing – original draft; writing – review and editing. Chiara Ruggeri: Conceptualization; data curation; formal analysis; writing – review and editing. Ivo Zemp: Conceptualization; data curation; formal analysis; writing – original draft; writing – review and editing. Ulrike Kutay: Conceptualization; data curation; formal analysis; supervision; funding acquisition; writing – original draft; project administration; writing – review and editing.

Disclosure and competing interests statement

The authors declare that they have no conflict of interest.

The EMBO Journal (2023) 42: e112699