Ribosome Biogenesis: Ribosomal Dispatch RNA Synthesis As a Package Deal

Ribosome Biogenesis: Ribosomal Dispatch RNA Synthesis as a Package Deal

Brenda A. Peculis Nop5/58, a core-binding protein present on all members of the C/D class of snoRNAs, including U3 [2,3]. They additionally put a different tag on Mpp10, a Ribosome biogenesis encompasses a complicated U3-specific protein previously identified in Baserga’s series of events involving hundreds of transiently lab [11]. Plasmids encoding the two tagged proteins interacting components. Insight into a mechanism were introduced into yeast and cell extracts were run for coordinating some of these events may come over a Protein A column to bind the TAP tag and from characterization of a functional processing enrich for all the C/D box snoRNAs. The bound com- complex. plexes were eluted and run over an antibody column to enrich for the Mpp10-containing complexes. When the material eluted from the second column Ribosome biogenesis is a universal cellular process. was examined, Baserga’s lab found this fraction was In eukaryotes, polymerase I transcription of the greatly enriched in U3 snoRNA. Thus encouraged, ribosomal (r)DNA locus yields a single, large pre-rRNA their collaborators in the Hunt lab examined the transcript (Figure 1A) which, through a complex series protein components that co-purified with the U3 RNA of processing, modification and folding steps, via nanoflow high-performance liquid chromatogra- ultimately gives rise to mature 18S, 5.8S and 25/28S phy/electrospray ionization mass spectrometry. They rRNAs assembled in functional ribosomal subunits. identified 28 proteins: ten were known U3-binding This maturation and assembly process involves over proteins, one known to affect 18S maturation and sev- 100 accessory proteins, about as many small nucleo- enteen were not previously identified or known to lar ribonucleoprotein particles (snoRNPs) and 60–80 affect pre-rRNA processing. Baserga and colleagues ribosomal proteins [1–5]. While previous genetic and [6] named these novel components U three proteins biochemical studies had identified many processing (Utp1–17). Database searches showed that eleven of components, including snoRNPs, helicases, endonu- these seventeen U3-associated yeast proteins have cleases, exonucleases, chaperones and so on, it putative human homologues, and two of the six remained unclear how such a complex collection of remaining ones have invertebrate homologues. The processing events and components might be coordi- authors interpret this high degree of evolutionary con- nated. The characterization of a ‘small subunit proces- servation as indicative of a highly conserved process- some’ by Dragon et al. [6] is a first step towards ing machinery. understanding how the many different steps in ribosome biogenesis are coordinated. A 5′ETS ITS1 ITS2 3′ETS The rRNA processing pathway branches early with 18S 25/28S many of the events for 18S rRNA maturation indepen- 5.8S dent of those for 5.8S–25/28S. The work of Dragon et * al. [6] focused on early pathway events and on the 18S branch. One of the prime components in this regard is Terminal ball the U3 snoRNP, a ubiquitous particle that is associ- B ated with a defined set of proteins and is required for maturation of 18S rRNA in every species in which it Pre-rRNA transcript has been tested [1–3]. Potential binding sites for U3 ′ snoRNA were identified within both the 5 ‘external Chromatin transcribed spacer’ (ETS) and 18S rRNA [7–9] (Figure 1A); mutagenesis experiments showed that some of these sites are indeed essential for 18S rRNA maturation [1,2]. Current Biology To increase our understanding of how U3 functions, Dragon et al. [6] examined the proteins constituting Figure 1. Ribosomal RNA transcription and processing. the functional U3 particle, with the goal of identifying, (A) A schematic of pre-RNA, indicating the mature 18S (red), not just the core U3 proteins but, more importantly, 5.8S (green) and 25/28S (blue) rRNAs which must be removed from the precursor. An ordered set of cleavages remove the 5′ any interacting factors that might be critical for U3 and 3′ external transcribed spacers (ETS) and the internal tran- function in the early steps of pre-rRNA processing. A scribed spacers, ITS1 and ITS2. The bracket and asterisk indi- protein-based protein affinity purification method cate the region required for terminal ball formation and where effective for examining large protein complexes [10] U3 binds the pre-rRNA. (B) A schematic of the ‘christmas tree’ was used to pursue this goal. Specifically, a ‘tandem appearance of actively transcribed rDNA loci visualized by affinity purification’ (TAP) tag was placed on yeast electron microscopy: the chromatin is the ‘trunk’, the closely packed rRNA transcripts constitute the ‘branches’ and at the tip of each transcript is a ‘terminal ball’, all of which are indi- NIH/NIDDK/GBB, Building 8 Room 106, Bethesda, Maryland cated. The elongating transcript encodes 18S, 5.8S and 25/28S 20892-0824, USA. E-mail: [email protected] rRNAs color-coded as in panel A. Dispatch R624

Baserga’s lab [6] cloned the corresponding genes Of the 28 proteins identified as components of the and epitope tagged the seventeen novel Utp proteins, SSU processome, ten were previously identified as and found each could be co-immunoprecipitated with U3-associated proteins. This validates the methodol- both Mpp10 and the U3 snoRNA, suggesting that they ogy used by Dragon et al. [6] and confirms earlier are all part of a large U3-containing complex. Two biochemical and genetic studies. More importantly, approaches were used to examine whether this these new studies have identified seventeen novel U3–Utp complex acts as a functional unit in vivo. First, proteins, including some encoding helicase and cat- seventeen conditional Utp expression strains were alytically active domains. This work vastly expands generated to allow analysis of the effects of genetic the toolbox available to dissect the in vivo roles of U3 depletion of each Utp protein on 18S maturation and and the processome. U3 snoRNA stability. Upon depletion of any single Utp Dragon et al. [6] have characterized a very large func- protein, 18S accumulation was severely inhibited tional pre-rRNA processing complex which may be while U3 stability was unaffected. The requirement for present in all eukaryotic cells. While U3-binding proteins Utp proteins in pre-rRNA processing thus appears to have been reported previously, this is the first descrip- be direct, rather than an indirect effect attributable to tion of proteins comprising a functional pre-rRNA pro- U3 RNA stability. Second, the distribution of several of cessing machine. This small subunit processome may the proteins across a glycerol gradient was examined: coordinate critical events, including transcription, mod- the Utps and U3 cosedimented in the 80S region of ification, pre-rRNA folding, protein assembly and pre- the gradient. Taken together, these results indicate rRNA processing, which culminate in a basic cellular that these 28 proteins constitute a large nucleolar process generically called ‘ribosome biogenesis’. complex which functions in processing of the 18S rRNA. Dragon et al. [6] call this machinery the small References 1. Maxwell, E.S. and Fournier, M.J. (1995). The small nucleolar RNAs. subunit (SSU) processome. Annu. Rev. Biochem. 64, 897–934. Other ‘large nucleolar complexes’ have been de- 2. Venema, J. and Tollervey, D. (1999). Ribosome Synthesis in Sac- scribed. The process of rRNA transcription from chro- charomyces cerevisiae. Annu. Rev. Genet. 33, 261–311. matin was visualized by electron microscopy over 30 3. Kressler, D., Linder, P. and de La Cruz, J. (1999). Protein trans- acting factors involved in ribosome biogenesis in Saccharomyces years ago [12]. Micrographs of rRNA gene loci from a cerevisiae. Mol. Cell. Biol. 19, 7897–7912. wide variety of organisms all reveal a characteristic 4. Peculis, B.A. and Mount, S.M. (1996). Ribosomal RNA: small nucle- ‘christmas tree’ appearance with a ‘terminal ball’ at olar RNAs make their mark. Curr. Biol. 6, 1413–1415. the tip of each elongating transcript (Figure 1B) [12]. 5. Peculis, B. (1997). RNA processing: pocket guides to ribosomal RNA. Curr. Biol. 7, R480–482. While the nature and composition of the terminal balls 6. Dragon, F., Gallagher, J.E., Compagnone-Post, P.A., Mitchell, B.M., were not known, they were indirectly implicated in Porwancher, K.A., Wehner, K.A., Wormsley, S., Settlage, R.E., Sha- processing, as the 5′ terminal region of the pre-rRNA banowitz, J., Osheim, Y. et al. (2002). A large nucleolar U3 ribonucleoprotein required for 18S ribosomal RNA biogenesis. Nature 417, transcript where they are found was known to be crit- 967–970. ical for the early maturation events [13]. 7. Beltrame, M. and Tollervey, D. (1995). Base pairing between U3 and Dragon et al. [6] directly addressed the question of the pre-ribosomal RNA is required for 18S rRNA synthesis. EMBO whether the SSU processomes they identified bio- J. 14, 4350–4356. 8. Hughes, J.M. (1996). Functional base-pairing interaction between chemically are the cytologically visible ‘terminal balls’. highly conserved elements of U3 small nucleolar RNA and the small In novel work done in the Beyer lab, chromosomal ribosomal subunit RNA. J. Mol. Biol. 259, 645–654. spreads of wild-type yeast were compared to spreads 9. Borovjagin, A.V. and Gerbi, S.A. (2001). Xenopus U3 snoRNA GAC- Box A′ and Box A sequences play distinct functional roles in rRNA of yeast conditionally depleted for either U3 snoRNA, processing. Mol. Cell. Biol. 21, 6210–6221. the Utp7 protein, Imp3p or Imp4p (two previously 10. Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M. and described U3-specific proteins [14]). All of the deple- Seraphin, B. (1999). A generic protein purification method for protein complex characterization and proteome exploration. Nat. tion strains showed loss of the characteristic termi Biotechnol. 17, 1030–1032. -nal ball structure compared to wild-type chromatin 11. Lee, S.J. and Baserga, S.J. (1997). Functional separation of pre- spreads. These results imply that either the terminal rRNA processing steps revealed by truncation of the U3 small nucleolar ribonucleoprotein component, Mpp10. Proc. Natl. Acad. balls are the direct visualization of the functional SSU Sci. U.S.A. 94, 13536–13541. processome characterized by Dragon et al. [6], or 12. Miller, O.L., Jr. and Beatty, B.R. (1969). Visualization of nucleolar that some components of the SSU processome are genes. Science 164, 955–957. required for assembly of the terminal ball. 13. Mougey, E.B., O’Reilly, M., Osheim, Y., Miller, O.L. Jr., Beyer, A. and Sollner-Webb, B. (1993). The terminal balls characteristic of eukary- While direct evidence that the terminal ball struc- otic rRNA transcription units in chromatin spreads are rRNA pro- tures are the processomes is lacking, there is addi- cessing complexes. Genes Dev. 7, 1609–1619. tional indirect evidence consistent with this possibility. 14. Lee, S.J. and Baserga, S.J. (1999). Imp3p and Imp4p, two specific ′ components of the U3 small nucleolar ribonucleoprotein that are The presence of an intact 5 ETS is required for in vitro essential for pre-18S rRNA processing. Mol. Cell. Biol. 19, processing and in vivo terminal ball formation [13]; 5441–5452. rDNA plasmids that lack the 5′ ETS do not form termi- 15. Wehner, K.A. and Baserga, S.J. (2002). The sigma(70)-like motif: a eukaryotic RNA binding domain unique to a superfamily of proteins nal balls [13]. The Baserga lab [15] previously showed required for ribosome biogenesis. Mol. Cell 9, 329–339. that the only pre-rRNAs that co-immunoprecipitate with the Imp4 protein are the 35S and 23S precursors; later pathway intermediates lacking the 5′ ETS are not precipitated. Together with the above biochemical and genetic data [6] these provide substantial indirect evidence that the processome is the terminal ball.