A common core structure for U3 small nucleolar RNAs

Abstract

U3 nucleolar small RNA (snRNA) is involved in early processing of the primary rRNA transcript. A secondary structure model for the unusually small Trypanosoma brucei U3 snRNA was deduced by comparative analysis of U3 snRNA sequences and by chemical modification and enzymatic cleavage of U3 snRNA in deproteinized and ribonucleoprotein (RNP) forms. Comprehensive alignment of U3 snRNAs from vertebrate, plant, fungal and protozoan species clearly delineated conserved and divergent features. The 5' domain of the T.brucei U3 snRNA appears to form one small, flexible 5' stem loop structure followed by a long single-stranded region; this model is a variation on 5' domain structures proposed for other U3 snRNAs which do not conform to a single model. The 3' domain of T.brucei U3 snRNA contains four single-stranded sequences conserved between U3 snRNAs. Of these, structural probing determined that the configurations of GAU region and box B and C sequences are altered by protein interactions in U3 snRNP. Conspicuously, the 3' domains of trypanosomal U3 snRNAs lack stem loops 11 and Ill, indicating that these structures are not required for conserved U3 snRNA functions.

An unexpected, conserved element of the U3 snoRNA is required for Mpp10p association

“…We did not expect to find that helix 1b9 (nt 47-63) was required for Mpp10p association+ Previous results from studies in several different organisms have implicated only the 39 domain (nt 73 the 39 end) of the U3 snoRNA in protein binding (Parker & Steitz, 1987;Jeppesen et al+, 1988;Segault et al+, 1992;Hartshorne & Agabian, 1994;Mereau et al+, 1997)+ Indeed, the 39 domain of the U3 snoRNA is sufficient for association of the common proteins, Nop56, Nop5/58, fibrillarin, and Snu13p as well as the specific protein, Rrp9p (Lubben et al+, 1993;Pluk et al+, 1998;Lukowiak et al+, 2000;Venema et al+, 2000;Watkins et al+, 2000)+ Because the stem on the distal side of helix 1b9 is protected during chemical and enzymatic mapping in vivo (Segault et al+, 1992;Mereau et al+, 1997), it has been proposed that helix 1b9 may form an additional basepairing interaction with the pre-rRNA+ One possible interpretation of our results is that this interaction is required for Mpp10p association and for overall U3 snoRNP function+ Alternatively, these nucleotides in the U3 snoRNP are protected because Mpp10p and perhaps other proteins are situated there+ Because Mpp10p is found in yeast and metazoans, if the sequence constituting helix 1b9 is truly important for Mpp10p association it would be conserved among different species+ Indeed, phylogenetic comparison of hinge region sequences indicates the capability to form a similar hairpin in the majority of organisms whose sequence is known+ In mammalian U3 snoRNAs, multiple hairpins are even possible+ Of the 20 sequenced U3 snoRNAs from different species, there are only four U3 snoRNAs where no hairpin can be drawn, those from Dictyostelium discoideum, Schizosaccharomyces pombe, Xenopus laevis, and Xenopus borealis+ In general, the sequences that comprise the stem of helix 1b9 are also highly conserved+ For example, two hairpins representing potential helix 1b9 structures can be drawn for the human U3 snoRNA, although the length of the loop differs from that of yeast+ However, although both stem-loop structures in the 59 domain and the hinge region are well conserved, it is thought that they do not form in vivo when the U3 snoRNA is base paired with the pre-rRNA (Mereau et al+, 1997;Antal et al+, 2000)+ Consistent with this is the high level of tolerance of Mpp10p association to mutation of the involved sequences, even to the point of eliminating the helix 1b9 structure+ Conservation, then, of the sequences required for Mpp10p association is likely due to a functional requirement other than maintenance of the structure of helix 1b9+…”

“…The 39 domain of the U3 snoRNA is highly folded and complexed with proteins (Parker & Steitz, 1987;Jeppesen et al+, 1988;Baserga et al+, 1991;Lubben et al+, 1993;Hartshorne & Agabian, 1994;Mereau et al+, 1997;Speckmann et al+, 1999)+ In all U3 molecules, boxes B/C and boxes C9/D are predicted to be in close proximity separated by phylogenetically conserved terminal and central stems+ The remaining nonconserved hairpins of the U3 snoRNA in yeast are dispensable and are not required for U3 function or stability (Samarsky & Fournier, 1998)+ Structural probing of boxes B/C and boxes C9/D in trypanosomes, yeast, and human U3 snoRNPs revealed these boxes are protected and most likely covered with proteins (Parker & Steitz, 1987;Baserga et al+, 1991;Hartshorne & Agabian, 1994;Mereau et al+, 1997)+ In yeast, substitutions of the sequences comprising box B disrupt U3 function (Samarsky & Fournier, 1998)+ In Xenopus, substitution of these same sequences has no affect on function; however, substitution of sequences comprising box C affect function and nucleolar localization of the U3 snoRNA (Lange et al+, 1998)+ Furthermore, mutations in box C affect fibrillarin association with U3 (Baserga et al+, 1991;Lange et al+, 1998;Samarsky & Fournier, 1998;Speckmann et al+, 1999)+ Similarly, fibrillarin has been shown to bind directly to the box C/D sequence of the U16 snoRNA in Xenopus oocytes and in ooycte extracts (Fatica et al+, 2000)+ Substitution of sequences in boxes C9 or D in yeast abolish U3 RNA levels, indicating that these sequences are required for U3 snoRNA accumulation (Samarsky & Fournier, 1998)+ Furthermore, mutations in box D and the 39 terminal stem block cap formation and affect nucleolar localization and nuclear import in Xenopus (Baserga et al+, 1992;Terns et al+, 1995;Lange et al+, 1998)…”

Interaction of the U3-55k protein with U3 snoRNA is mediated by the Box B/C motif of U3 and the WD repeats of U3-55k

“…Substitution of the other sequence elements did not prevent interaction (Fig. 3, U3∆A′, U3∆A, U3∆H, U3∆C′ and U3∆D, lanes [7][8][9][10][11][12][13][14][15][16][17][18][25][26][27]. The results indicate that Boxes B and C of U3 snoRNA are necessary for interaction with U3-55k in vivo.…”

“…Sequence analysis of U3 snoRNA reveals six evolutionarily conserved sequence elements, termed Boxes A, A′, B, C, C′ and D (17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27). Boxes A and A′ are located in the 5′ domain of U3 snoRNA and, along with the 'hinge' region, base pair to sites within the pre-rRNA to direct the multiple endonucleolytic cleavages necessary for the maturation of 18S rRNA (15,16,23,25,28).…”

“…Boxes A and A′ are located in the 5′ domain of U3 snoRNA and, along with the 'hinge' region, base pair to sites within the pre-rRNA to direct the multiple endonucleolytic cleavages necessary for the maturation of 18S rRNA (15,16,23,25,28). Boxes B, C, C′ and D are found in the 3′ domain of U3 and have been characterized as sequences required for protein association (17,19,23,25,29). The secondary structure of the 3′ domain positions Boxes B and C, as well as Boxes C′ and D into discrete loops flanked by base paired stems to form the Box B/C and Box C′/D motifs, respectively (23,27,30).…”