Genes for human U3 small nucleolar RNA contain highly conserved flanking sequences

Role of pre-rRNA base pairing and 80S complex formation in subnucleolar localization of the U3 snoRNP

In the nucleolus the U3 snoRNA is recruited to the 80S pre-rRNA processing complex in the dense fibrillar component (DFC). The U3 snoRNA is found throughout the nucleolus and has been proposed to move with the preribosomes to the granular component (GC). In contrast, the localization of other RNAs, such as the U8 snoRNA, is restricted to the DFC. Here we show that the incorporation of the U3 snoRNA into the 80S processing complex is not dependent on pre-rRNA base pairing sequences but requires the B/C motif, a U3-specific protein-binding element. We also show that the binding of Mpp10 to the 80S U3 complex is dependent on sequences within the U3 snoRNA that base pair with the pre-rRNA adjacent to the initial cleavage site. Furthermore, mutations that inhibit 80S complex formation and/or the association of Mpp10 result in retention of the U3 snoRNA in the DFC. From this we propose that the GC localization of the U3 snoRNA is a direct result of its active involvement in the initial steps of ribosome biogenesis.

Nucleolar factors direct the 2'-O-ribose methylation and pseudouridylation of U6 spliceosomal RNA

The nucleolus has long been known as a functionally highly specialized subnuclear compartment where synthesis, posttranscriptional modification, and processing of cytoplasmic rRNAs take place. In this study, we demonstrate that the nucleolus contains all the trans-acting factors that are responsible for the accurate and efficient synthesis of the eight 2'-O-methylated nucleotides and three pseudouridine residues carried by the mammalian U6 spliceosomal small nuclear RNA. Factors mediating the formation of pseudouridine residues in the U3 small nucleolar RNA are also present and functionally active in the nucleolus. For selection of the correct target nucleotides in the U6 and U3 RNAs, the nucleolar 2'-O-methylation and pseudouridylation factors rely on short sequences located around the target nucleotide to be modified. This observation further underscores a recently proposed role for small nucleolar guide RNAs in the 2'-O-methylation of the U6 spliceosomal RNA (K. T. Tycowski, Z.-H. You, P. J. Graham, and J. A. Steitz, Mol. Cell 2:629-638, 1998). We demonstrate that a novel 2'-O-methylated nucleotide can be generated in the yeast U6 RNA by use of an artificial 2'-O-methylation small nucleolar guide RNA. We also show that a short fragment of the 5.8S rRNA, when expressed as part of the human U6 RNA, is faithfully 2'-O-methylated and pseudouridylated. These results are most consistent with a trafficking pathway in which the U6 spliceosomal RNA cycles through the nucleolus to undergo nucleolar RNA-directed modifications.

Human genes encoding U3 snRNA associate with coiled bodies in interphase cells and are clustered on chromosome 17p11.2 in a complex inverted repeat structure

Coiled bodies (CBs) are nuclear organelles whose morphological structure and molecular composition have been conserved from plants to animals. Furthermore, CBs are often found to co-localize with specific DNA loci in both mammalian somatic nuclei and amphibian oocytes. Much as rDNA sequences are called nucleolus organizers, we term these coiled body-associated sequences 'coiled body organizers' (CBORs). The only sequences that have been shown to be CBORs in human cells are the U1, U2 and histone gene loci. We wanted to determine whether other snRNA genes might also act as CBORs. In this paper we show that human U3 genes (the RNU3 locus) preferentially associate with CBs in interphase cells. In addition, we have analyzed the genomic organization of the RNU3 locus by constructing a BAC and P1 clone contig. We found that, unlike the RNU1 and RNU2 loci, U3 genes are not tandemly repeated. Rather, U3 genes are clustered on human chromosome 17p11.2, with evidence for large inverted duplications within the cluster. Thus all of the CBORs identified to date are composed of either tandemly repeated or tightly clustered genes. The evolutionary and cell biological consequences of this type of organization are discussed.

Sequence-specific DNA binding of the B-cell-specific coactivator OCA-B

B-cell-specific transcription of immunoglobulin genes is mediated by the interaction of a POU domain containing transcription factor Oct-1 or Oct-2, with the B-cell-specific coactivator OCA-B (Bob-1, OBF-1) and a prototype octamer element. We find that OCA-B binds DNA directly in the major groove between the two subdomains of the POU domain, requiring both an A at the fifth position of the octamer element and contact with the POU domain. An amino-terminal fragment of OCA-B binds the octamer site in the absence of a POU domain with the same sequence specificity. Coactivator OCA-B may undergo a POU-dependent conformational change that exposes its amino terminus, allowing it to recognize specific DNA sequences in the major groove within the binding site for Oct-1 or Oct-2. The recognition of both the POU domain and the octamer sequence by OCA-B provides a mechanism for differential regulation of octamer sites containing genes by the ubiquitous factor Oct-1.

Small nucleolar RNA

A growing list of small nucleolar RNAs (snoRNAs) has been characterized in eukaryotes. They are transcribed by RNA polymerase II or III; some snoRNAs are encoded in the introns of other genes. The nonintronic polymerase II transcribed snoRNAs receive a trimethylguanosine cap, probably in the nucleus, and move to the nucleolus. snoRNAs are complexed with proteins, sometimes including fibrillarin. Localization and maintenance in the nucleolus of some snoRNAs requires the presence of initial precursor rRNA (pre-rRNA). Many snoRNAs have conserved sequence boxes C and D and a 3' terminal stem; the role of these features are discussed. Functional assays done for a few snoRNAs indicate their roles in rRNA processing for cleavage of the external and internal transcribed spacers (ETS and ITS). U3 is the most abundant snoRNA and is needed for cleavage of ETS1 and ITS1; experimental results on U3 binding sites in pre-rRNA are reviewed. 18S rRNA production also needs U14, U22, and snR30 snoRNAs, whereas U8 snoRNA is needed for 5.8S and 28S rRNA production. Other snoRNAs that are complementary to 18S or 28S rRNA might act as chaperones to mediate RNA folding. Whether snoRNAs join together in a large rRNA processing complex (the "processome") is not yet clear. It has been hypothesized that such complexes could anchor the ends of loops in pre-rRNA containing 18S or 28S rRNA, thereby replacing base-paired stems found in pre-rRNA of prokaryotes.

Nucleologenesis: U3 snRNA-containing prenucleolar bodies move to sites of active pre-rRNA transcription after mitosis

We have investigated the distribution of U3 snRNA and rRNA in HeLa cells and normal rat kidney cells during interphase and mitosis. U3 snRNA, known to be involved in pre-rRNA processing, was detected in nucleoli and coiled bodies during interphase, whereas rRNA was distributed in the nucleoli and throughout the cytoplasm. By comparison, ribosomal protein S6 was detected in nucleoli, coiled bodies, and in the cytoplasm. During nucleologenesis, pre-rRNA was observed in newly forming nucleoli during late telophase but not in prenucleolar bodies (PNBs), whereas U3 snRNA was detected in forming nucleoli and PNBs. Similar findings to those reported here for the localization of U3 snRNA have been reported previously for the U3 small nuclear ribonucleoprotein fibrillarin. These results suggest that components involved in pre-rRNA processing localize to discrete PNBs at the end of mitosis. The nucleolus is formed at specific telophase domains (nucleolar organizing regions) and the PNBs, containing factors essential for pre-rRNA processing, are recruited to these sites of rRNA transcription and processing.

Isolation of U3 snoRNP from CHO cells: a novel 55 kDa protein binds to the central part of U3 snoRNA

U3 snoRNP, the most abundant of the small nucleolar ribonucleoprotein particles (snoRNPs), has previously been demonstrated to participate in pre-rRNA maturation. Here we report the purification of U3 snoRNP from CHO cells using anti-m3G-immunoaffinity and mono Q anion-exchange chromatography. Isolated U3 snoRNPs contain three novel proteins, of 15, 50 and 55 kDa respectively. These proteins may represent core U3 snoRNP proteins whose binding mediates the association of other proteins, such as fibrillarin, that are lost during purification. Using a rabbit antiserum raised against the 55 kDa protein, and an in vitro reconstitution assay, we have localised the 55 kDa protein binding site on the U3 snoRNA. Stable binding of the 55 kDa protein requires sequences located between nucleotides 97 and 204 of the human U3 snoRNA, including the evolutionarily conserved B and C sequence motifs.

Transcription of human 5S rRNA genes is influenced by an upstream DNA sequence

Six human 5S rRNA genes and gene variants and one pseudogene have been sequenced. The six genes/variants were transcribed in a HeLa cell extract with about equal efficiency. Three genes contain the Sp1 binding sequence GGGCGG in position -43 to -38 and three genes contain the Sp1 like sequence GGGCCG in this position. The six genes contain furthermore one Sp1 binding site in a position about -245 and one ATF recognition site in a position about -202. A 12 bp sequence (GGCTCTTGGGGC) found in position -32 to -21 strongly influenced the transcriptional efficiency in vitro. This 12-mer, designated the D box, has also been found upstream a 5S rRNA gene from hamster and mouse. Removal of the Sp1 binding sites had no effect on the transcription in vitro whereas the transcriptional efficiency decreased to 10% if the D box was removed from the human 5S rRNA gene.

Genes for Xenopus laevis U3 small nuclear RNA

Genomic Southern blots showed there are only 14 to 20 copies of U3 snRNA genes per somatic genome in Xenopus laevis, unlike the highly repetitive, tandem arrangement of other snRNA genes in this organism. Sequencing of two U3 snRNA genes from lambda clones of a genomic library revealed striking similarity upstream, but much more divergence downstream. Consensus motifs common to other U snRNA genes were also found: a distal sequence element (DSE, octamer motif at -222 to -215), a proximal sequence element (PSE, at -62 to -52) and a 3' Box (15 or 16 bp downstream of the U3 genes). The DSE of mammals also has an inverted CCAAT motif specific for U3 snRNA genes, and we find this is conserved in the amphibian U3 snRNA genes. The Xenopus inverted CCAAT motif is exactly one helical turn further upstream of the octamer motif than its mammalian counterpart, suggesting interaction of putative transcription factors bound to these motifs. Mutation of the inverted CCAAT motif and part of an adjacent Sp1 site greatly depresses transcription of the mutant U3 snRNA gene in Xenopus oocytes, implying a role in transcriptional efficiency. Electrophoretic mobility shift assays implicate transcription factor binding to this region.

Structure of the differentially expressed mouse U3A gene

Two markedly different forms of U3 RNA are present in mouse, the relative abundance of which largely depends upon the tissues. In all cases studied so far, the more abundant form is U3B, encoded by four previously characterized genes. We report here the isolation and analysis of the unique gene encoding the U3A variant, which completes the characterization of the mouse U3 multigene family. Comparisons with rat U3 genes indicate that the diversification of the A and B forms has predated the mouse/rat separation. The two forms of U3 RNA are submitted to similar, but not identical, primary and secondary structure constraints. As for the sequences flanking the RNA coding region, similar observations emerge for both types of genes: for each type, the 5' flanks are strongly conserved between mouse and rat, over at least the proximal 500 bp, whereas only about 30 bp of proximal 3' flanks are preserved, which include a signal for the formation of vertebrate U small nRNA 3' end. By contrast the 5' flanks of the two types of genes diverge extensively from each other, either in mouse or in rat, and could be involved in the differential expression of the two forms. Even over the few conserved motifs thought to be involved in the basic transcriptional control of vertebrate U small nRNA genes, the A and B forms of U3 genes exhibit specific differences maintained in the two rodent species.

Cooperation between CCAAT and octamer motifs in the distal sequence element of the rat U3 small nucleolar RNA promoter

Mammalian U3 small nucleolar RNA promoters possess a highly conserved distal sequence element (DSE) consisting of CCAAT and octamer motifs separated by 11-12 base pairs. We show here that both motifs are required for transcription of a rat U3D gene in Xenopus oocytes. Deletion of the CCAAT motif leaves residual DSE activity, while removal of the octamer motif does not. Changing the conserved spacing between the two motifs generally inhibits transcription less than deletion of either motif, but increasing the spacing between the motifs by one helical turn of DNA preserves normal levels of transcription. We also show that the rat U3D DSE is functionally equivalent to the human U2 snRNA DSE, which consists of adjacent GC and octamer motifs, and that elements from the Herpes Simplex Virus thymidine kinase promoter can replace part or all of the U3D DSE. These data are apparently paradoxical; despite high evolutionary conservation, the U3 DSE is relatively insensitive to mutation, and other upstream motifs are also able to drive transcription from the U3 basal promoter. We suggest that the conserved structure of the U3 DSE may be required for regulation rather than efficiency of U3 transcription.

Organization of the gene family encoding mouse U3B RNA: role of gene conversions in its concerted evolution

In mouse, U3B small nuclear RNA is encoded by a small family of four functional genes. Three of these genes have recently been identified, with two of them being located 5 kb apart, in the same orientation [Mazan and Bachellerie, J. Biol. Chem. 263 (1988) 19461-19467]. By analyzing a cosmid library of mouse genomic DNA we have isolated the fourth gene, U3B.4, which is located 12 kb downstream from the previously reported pair of linked genes, U3B.1 and U3B.2, in the same orientation. Its sequence analysis demonstrates that four mouse U3B genes have undergone a very efficient concerted evolution, with a perfect sequence identity not only over the entire RNA coding region but also over 254 bp of 5' flanking DNA which contain essential transcription signals. Moreover, in the regions which immediately flank the domain of perfect sequence identity, a patchwork of sequence similarities among the four genes is apparent, with a series of striking reversals in relationships between adjacent localized areas. This suggests that gene conversions have played a major role in the concerted evolution of the U3B gene family. A superimposition of conversion events, which probably occurred at distinct stages of rodent evolutions can be inferred. This has involved not only the three clustered copies, but also the remaining gene, U3B.3, the location of which relative to the cluster remains unknown. B1 repetitive elements are found at the boundaries of conversion domains, suggesting their possible role in control of the sequence homogenization process.

Molecular analysis of a U3 RNA gene locus in tomato: transcription signals, the coding region, expression in transgenic tobacco plants and tandemly repeated pseudogenes

By screening a tomato genomic library with a tomato U3 RNA probe, we detected a U3 genomic locus whose coding region was determined by primer extension (5' end) and direct RNA sequencing of purified U3 RNA from tomato (3' end). Tomato U3 RNA is 216 nucleotides long, contains all the four evolutionarily highly conserved sequence blocks (Boxes A to D), has at its 5' end a cap not precipitable with anti-m3G antibodies and can be folded into a peculiar secondary structure with two stem-loops at its 5' end. A tagged derivative of the U3 gene was faithfully expressed in transgenic tobacco plants. In the 5' flanking region both plant-specific UsnRNA transcription signals [the TATA-like sequence and the upstream sequence element (USE)] were present, but were positioned closer to each other and also to the cap site in the U3 gene than in the genes for the plant spliceosomal UsnRNAs studied so far. The 3' flanking region of the tomato U3 gene lacked the consensus sequence of the putative termination signal established for the plant spliceosomal UsnRNA genes and contained a pyrimidine-rich tract (R1) followed by four tandemly repeated U3 pseudogenes (U3.1 ps to U3.4 ps) flanked by slightly altered forms (R2 to R5) of R1 and most probably generated by DNA-mediated events. Our results are in line with the conjecture that the enzyme transcribing the tomato U3 gene has different structural requirements for transcriptional activity than the enzyme transcribing plant U1, U2 and U5 genes.