During ribosomal RNA (rRNA) maturation, cleavages at defined sites separate the mature rRNAs from spacer regions, but the identities of several enzymes required for 18S rRNA release remain unknown. PilT N-terminus (PIN) domain proteins are frequently endonucleases and the PIN domain protein Utp24 is essential for early cleavages at three pre-rRNA sites in yeast (A0, A1 and A2) and humans (A0, 1 and 2a). In yeast, A1 is cleaved prior to A2 and both cleavages require base-pairing by the U3 snoRNA to the central pseudoknot elements of the 18S rRNA. We found that yeast Utp24 UV-crosslinked in vivo to U3 and the pseudoknot, placing Utp24 close to cleavage at site A1. Yeast and human Utp24 proteins exhibited in vitro endonuclease activity on an RNA substrate containing yeast site A2. Moreover, an intact PIN domain in human UTP24 was required for accurate cleavages at sites 1 and 2a in vivo, whereas mutation of another potential site 2a endonuclease, RCL1, did not affect 18S production. We propose that Utp24 cleaves sites A1/1 and A2/2a in yeast and human cells.
Two proteins with PIN endonuclease domains, yUtp24(Fcf1)/hUTP24 and yUtp23/hUTP23 are essential for early pre-ribosomal (r)RNA cleavages at sites A0, A1/1 and A2/2a in yeast and humans. The yUtp24/hUTP24 PIN endonuclease is proposed to cleave at sites A1/1 and A2/2a, but the enzyme cleaving at site A0 is not known. Yeast yUtp23 contains a degenerate, non-essential PIN domain and functions together with the snR30 snoRNA, while human hUTP23 is associated with U17, the human snR30 counterpart. Using in vivo RNA–protein crosslinking and gel shift experiments, we reveal that yUtp23/hUTP23 makes direct contacts with expansion sequence 6 (ES6) in the 18S rRNA sequence and that yUtp23 interacts with the 3΄ half of the snR30 snoRNA. Protein–protein interaction studies further demonstrated that yeast yUtp23 and human hUTP23 directly interact with the H/ACA snoRNP protein yNhp2/hNHP2, the RNA helicase yRok1/hROK1(DDX52), the ribosome biogenesis factor yRrp7/hRRP7 and yUtp24/hUTP24. yUtp23/hUTP23 could therefore be central to the coordinated integration and release of ES6 binding factors and likely plays a pivotal role in remodeling this pre-rRNA region in both yeast and humans. Finally, studies using RNAi-rescue systems in human cells revealed that intact PIN domain and Zinc finger motifs in human hUTP23 are essential for 18S rRNA maturation.
The RNA exosome is a ubiquitously expressed complex of nine core proteins (EXOSC1-9) and associated nucleases responsible for RNA processing and degradation. Mutations inEXOSC3,EXOSC8,EXOSC9, and the exosome cofactorRBM7cause pontocerebellar hypoplasia and motor neuronopathy. We investigated the consequences of exosome mutations on RNA metabolism and cellular survival in zebrafish and human cell models. We observed that levels of mRNAs encoding p53 and ribosome biogenesis factors are increased in zebrafish lines with homozygous mutations ofexosc8orexosc9, respectively. Consistent with higher p53 levels, mutant zebrafish have a reduced head size, smaller brain, and cerebellum caused by an increased number of apoptotic cells during development. Down-regulation ofEXOSC8andEXOSC9in human cells leads to p53 protein stabilisation and G2/M cell cycle arrest. Increased p53 transcript levels were also observed in muscle samples from patients withEXOSC9mutations. Our work provides explanation for the pathogenesis of exosome-related disorders and highlights the link between exosome function, ribosome biogenesis, and p53-dependent signalling. We suggest that exosome-related disorders could be classified as ribosomopathies.
Ribosome production is an essential cellular process involving a plethora of trans-acting factors, such as nucleases, methyltransferases, RNA helicases and kinases that catalyse key maturation steps. Precise temporal and spatial regulation of such enzymes is essential to ensure accurate and efficient subunit assembly. Here, we focus on the maturation of the 3' end of the 18S rRNA in human cells. We reveal that human RIO2 is an active kinase that phosphorylates both itself and the rRNA methyltransferase DIM1 in vitro. In contrast to yeast, our data confirm that human DIM1 predominantly acts in the nucleus and we further demonstrate that the 21S pre-rRNA is the main target for DIM1-catalysed methylation. We show that the PIN domain of the endonuclease NOB1 is required for site 3 cleavage, while the zinc ribbon domain is essential for pre-40S recruitment. Furthermore, we also demonstrate that NOB1, PNO1 and DIM1 bind to a region of the pre-rRNA encompassing the 3' end of 18S and the start of ITS1, in vitro. Interestingly, NOB1 is present in the cell at higher levels than other pre-40S factors. We provide evidence that NOB1 is multimeric within the cell and show that NOB1 multimerisation is lost when ribosome biogenesis is blocked. Taken together, our data indicate a dynamic interplay of key factors associated with the 3' end of the 18S rRNA during human pre-40S biogenesis and highlight potential mechanisms by which this process can be regulated.
Two of the four human ubiquitin-encoding genes express ubiquitin as an N-terminal fusion precursor polypeptide, with either ribosomal protein (RP) RPS27a or RPL40 at the C-terminus. RPS27a and RPL40 have been proposed to be important for the induction of the tumour suppressor p53 in response to defects in ribosome biogenesis, suggesting that they may play a role in the coordination of ribosome production, ubiquitin levels and p53 signalling. Here, we report that RPS27a is cleaved from the ubiquitin-RP precursor in a process that appears independent of ribosome biogenesis. In contrast to other RPs, the knockdown of either RPS27a or RPL40 did not stabilise the tumour suppressor p53 in U2OS cells. Knockdown of neither protein blocked p53 stabilisation following inhibition of ribosome biogenesis by actinomycin D, indicating that they are not needed for p53 signalling in these cells. However, the knockdown of both RPS27a and RPL40 in MCF7 and LNCaP cells robustly induced p53, consistent with observations made with the majority of other RPs. Importantly, RPS27a and RPL40 are needed for rRNA production in all cell lines tested. Our data suggest that the role of RPS27a and RPL40 in p53 signalling, but not their importance in ribosome biogenesis, differs between cell types.