Fibrillarin Genes Encode Both a Conserved Nucleolar Protein and a Novel Small Nucleolar RNA Involved in Ribosomal RNA Methylation inArabidopsis thaliana

Fibrillarin is a key nucleolar protein in eukaryotes which associates with box C/D small nucleolar RNAs (snoRNAs) directing 2′-O-ribose methylation of the rRNA. In this study we describe two genes in Arabidopsis thaliana,AtFib1 and AtFib2, encoding nearly identical proteins conserved with eukaryotic fibrillarins. We demonstrate that AtFib1 and AtFib2 proteins are functional homologs of the yeast Nop1p (fibrillarin) and can rescue a yeast NOP1-null mutant strain. Surprisingly, for the first time in plants, we identified two isoforms of a novel box C/D snoRNA, U60.1f and U60.2f, nested in the fifth intron of AtFib1 andAtFib2. Interestingly after gene duplication the host intronic sequences completely diverged, but the snoRNA was conserved, even in other crucifer fibrillarin genes. We show that the U60f snoRNAs accumulate in seedlings and that their targeted residue on the 25 S rRNA is methylated. Our data reveal that the three modes of expression of snoRNAs, single, polycistronic, and intronic, exist in plants and suggest that the mechanisms directing rRNA methylation, dependent on fibrillarin and box C/D snoRNAs, are evolutionarily conserved in plants. Fibrillarin is a key nucleolar protein in eukaryotes which associates with box C/D small nucleolar RNAs (snoRNAs) directing 2′-O-ribose methylation of the rRNA. In this study we describe two genes in Arabidopsis thaliana,AtFib1 and AtFib2, encoding nearly identical proteins conserved with eukaryotic fibrillarins. We demonstrate that AtFib1 and AtFib2 proteins are functional homologs of the yeast Nop1p (fibrillarin) and can rescue a yeast NOP1-null mutant strain. Surprisingly, for the first time in plants, we identified two isoforms of a novel box C/D snoRNA, U60.1f and U60.2f, nested in the fifth intron of AtFib1 andAtFib2. Interestingly after gene duplication the host intronic sequences completely diverged, but the snoRNA was conserved, even in other crucifer fibrillarin genes. We show that the U60f snoRNAs accumulate in seedlings and that their targeted residue on the 25 S rRNA is methylated. Our data reveal that the three modes of expression of snoRNAs, single, polycistronic, and intronic, exist in plants and suggest that the mechanisms directing rRNA methylation, dependent on fibrillarin and box C/D snoRNAs, are evolutionarily conserved in plants. S-adenosylmethionine small nucleolar RNA small ribonucleoprotein particles expressed sequence tag polymerase chain reaction kilobase pair(s) green fluorescent protein open reading frame base pair(s) 2′-O-ribose-methylated adenosine and guanosine, respectively In plants as in all eukaryotes, the 25 S, the 5.8 S, and the 18 S rRNAs are cotranscribed into a large pre-rRNA precursor that is processed to generate the rRNAs (1Eichler D.C. Craig N. Prog. Nucleic Acids Res. Mol. Biol. 1997; 49: 197-239Crossref Scopus (180) Google Scholar, 2Kressler D. Linder P. de La Cruz J. Mol. Cell. Biol. 1999; 19: 7897-7912Crossref PubMed Scopus (308) Google Scholar). Processing of pre-rRNA involves the cleavage of spacer sequences and the modification of specific rRNA nucleotides by either 2′-O-ribose methylation or base pseudouridylation. These modifications concern a wide number of rRNA nucleotides, nearly 200 in vertebrates and 100 in yeast, but their effect on ribosome function remains unclear (3Bachellerie J.-P. Cavaillé J. Grosjean H. Benne R. Modification and Editing of RNA. American Society for Microbiology, Washington D. C.1998: 255-272Google Scholar). All of these early pre-rRNA processings occur in the dense fibrillar component of the nucleolus characterized by fibrillarin (4Brown J.W. Shaw P.J. Plant Cell. 1998; 10: 649-657Crossref PubMed Scopus (56) Google Scholar). Fibrillarin is a key nucleolar protein for pre-rRNA processing conserved from vertebrates to archaebacteria (5Schimmang T. Tollervey D. Kern H. Frank R. Hurt E.C. EMBO J. 1989; 8: 4015-4024Crossref PubMed Scopus (236) Google Scholar, 6Lapeyre B. Mariottini P. Mathieu C. Ferrer P. Amaldi F. Amalric F. Caizergues-Ferrer M. Mol. Cell. Biol. 1990; 10: 430-434Crossref PubMed Scopus (106) Google Scholar, 7Aris J.P. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 931-935Crossref PubMed Scopus (156) Google Scholar, 8David E. McNeil B.M. Basile V. Pearlman R.E. Mol. Biol. Cell. 1997; 8: 1051-1061Crossref PubMed Scopus (21) Google Scholar, 9Wang H. Boisvert D. Kim K.K. Kim R. Kim S.-H. EMBO J. 2000; 19: 317-323Crossref PubMed Scopus (143) Google Scholar, 10Henriquez R. Blobel G. Aris J.P. J. Biol. Chem. 1990; 265: 2209-2215Abstract Full Text PDF PubMed Google Scholar). Nop1p, the yeast fibrillarin, is an essential protein that can be partially complemented by the vertebrate fibrillarins showing the functional conservation of these proteins (11Jansen R.P. Hurt E.C. Kern H. Lehtonen H. Carmo-Fonseca M. Lapeyre B. Tollervey D. J. Cell Biol. 1991; 113: 715-729Crossref PubMed Scopus (132) Google Scholar). Analysis of yeast Nop1p temperature-sensitive mutants revealed that it is required for pre-rRNA methylation, early cleavages, and ribosome assembly (12Tollervey D. Lehtonen H. Jansen R. Kern H. Hurt W.C. Cell. 1993; 72: 443-457Abstract Full Text PDF PubMed Scopus (407) Google Scholar). The recent crystallization of an archaebacterial fibrillarin-homolog identified anS-adenosylmethionine (AdoMet)1-dependent methyltransferase-like domain (9Wang H. Boisvert D. Kim K.K. Kim R. Kim S.-H. EMBO J. 2000; 19: 317-323Crossref PubMed Scopus (143) Google Scholar), which is conserved in the yeast Nop1p (9Wang H. Boisvert D. Kim K.K. Kim R. Kim S.-H. EMBO J. 2000; 19: 317-323Crossref PubMed Scopus (143) Google Scholar, 13Niewmierzycka A. Clarke S. J. Biol. Chem. 1999; 274: 814-824Abstract Full Text Full Text PDF PubMed Scopus (192) Google Scholar). Thus the genetic and structural data strongly suggest that fibrillarin is the rRNA methylase, but this activity could never be detected in vitro (9Wang H. Boisvert D. Kim K.K. Kim R. Kim S.-H. EMBO J. 2000; 19: 317-323Crossref PubMed Scopus (143) Google Scholar). Recent studies in vertebrates and yeast have shown that rRNA nucleotide modifications are targeted by two classes of small nucleolar RNAs (snoRNAs): the box C/D snoRNAs directing 2′-O-ribose methylation and the H/ACA box snoRNAs directing base pseudouridylation (14Lafontaine D.L. Tollervey D. Trends Biochem. Sci. 1998; 23: 383-388Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar). The box C/D snoRNAs have two phylogenetically conserved C (RUGAUGA) and D (CUGA) motifs flanked by short inverted repeats at the 5′ and 3′ termini of the snoRNA, respectively. These elements induce folding of a stem-loop RNA structure which is essential for snoRNA stability and nucleolar accumulation. Adjacent to the terminal box D or to an additional box D′ (CUGA or related motifs) there is an antisense sequence of 10 to 21 bases spanning the 2′-O-ribose methylation site on rRNA. This guide sequence forms a snoRNA·rRNA duplex targeting the modified nucleotide. Thus a specific antisense box C/D snoRNA is associated with each methylated residue in the rRNA. In addition, a few members of this class (including the most abundant U3, U8, and U14) are involved in pre-rRNA spacer cleavages (for recent reviews compiling these data, see Refs 3Bachellerie J.-P. Cavaillé J. Grosjean H. Benne R. Modification and Editing of RNA. American Society for Microbiology, Washington D. C.1998: 255-272Google Scholarand 14Lafontaine D.L. Tollervey D. Trends Biochem. Sci. 1998; 23: 383-388Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar). In vivo box C/D snoRNAs form small ribonucleoprotein particles (snoRNPs) that are immunoprecipitated from cell extracts with anti-fibrillarin antibodies (15Maxwell E.S. Fournier M.J. Annu. Rev. Biochem. 1995; 35: 897-934Crossref Scopus (543) Google Scholar). In addition to fibrillarin, two other proteins common to all box C/D snoRNPs have been identified by genetic and biochemical approaches in yeast and human cells corresponding to Nop56p and Nop58p/Nop5p (16Gautier T. Bergès D. Tollervey D. Hurt E. Mol. Cell. Biol. 1997; 17: 7088-7098Crossref PubMed Scopus (232) Google Scholar, 17Wu P. Brockenbrough J.S. Metcalfe A.C. Chen S. Aris J.P. J. Biol. Chem. 1998; 273: 16453-16463Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, 18Lafontaine D.L. Tollervey D. RNA. 1999; 5: 455-467Crossref PubMed Scopus (139) Google Scholar, 19Lyman S.K. Gerace L. Baserga S. RNA. 1999; 5: 1597-1604Crossref PubMed Scopus (58) Google Scholar). Altogether these evidences strongly suggest that fibrillarin is directly implicated in rRNA methylation targeted by the box C/D snoRNAs within the snoRNPs (2Kressler D. Linder P. de La Cruz J. Mol. Cell. Biol. 1999; 19: 7897-7912Crossref PubMed Scopus (308) Google Scholar). Another particular feature of snoRNAs in eukaryotes concerns their mode of expression. In vertebrates most snoRNAs are nested within introns of protein-coding genes (15Maxwell E.S. Fournier M.J. Annu. Rev. Biochem. 1995; 35: 897-934Crossref Scopus (543) Google Scholar). Splicing of the pre-mRNA releases the host intron that is then trimmed by 5′- and 3′-exonucleases to the mature snoRNA ends (20Kiss T. Filipowicz W. Genes Dev. 1995; 9: 1411-1424Crossref PubMed Scopus (138) Google Scholar, 21Cavaillé J. Bachellerie J.P. Biochimie (Paris). 1996; 78: 443-456Crossref PubMed Scopus (80) Google Scholar). Usually snoRNA host genes encode abundant proteins related to ribosome biogenesis, such as ribosomal proteins or nucleolin (15Maxwell E.S. Fournier M.J. Annu. Rev. Biochem. 1995; 35: 897-934Crossref Scopus (543) Google Scholar). But host genes are not always related to ribosomal biogenesis, and there are even examples of host genes with no coding information except for the intronic snoRNA (22Tycowski K.T. Shu M.-D. Steitz J.A. Nature. 1996; 379: 464-466Crossref PubMed Scopus (258) Google Scholar, 23Pelczar P. Filipowicz W. Mol. Cell. Biol. 1998; 18: 4509-4518Crossref PubMed Scopus (93) Google Scholar). Most yeast snoRNAs are either single genes with their own promoter or are clustered genes driven by a single promoter-producing polycistronic snoRNAs (24 and references therein). In contrast to other eukaryotic systems, little is known about pre-RNA processing and the nucleolar factors directing rRNA modifications in plants. So far the plant fibrillarin has not been described, although it has been immunodetected in onion and pea nuclei using human autoimmune sera (25Cerdido A. Medina F.J. Chromosoma (Berl.). 1995; 103: 625-634Crossref PubMed Scopus (40) Google Scholar, 26Beven A.F. Lee R. Razaz M. Leader D.J. Brown J.W. Shaw P.J. J. Cell Sci. 1996; 109: 1241-1251Crossref PubMed Google Scholar, 27De Cárcer G. Medina F.J. J. Struct. Biol. 1999; 128: 139-151Crossref PubMed Scopus (43) Google Scholar). Also, only a few plant snoRNAs have been identified and shown to be expressed (for review, see Ref. 4Brown J.W. Shaw P.J. Plant Cell. 1998; 10: 649-657Crossref PubMed Scopus (56) Google Scholar). These include U3, U14, and seven other box C/D and one box H/ACA snoRNAs described in Arabidopsis or maize (28Leader D.J. Clark G.P. Watters J. Beven A.F. Shaw P.J. Brown J.W. EMBO J. 1997; 16: 5742-5751Crossref PubMed Scopus (91) Google Scholar). Many other snoRNAs should exist in plants because early biochemical studies indicate the presence of numerous modified nucleotides in plant rRNAs (29Cecchini J.-P. Miassod R. Eur. J. Biochem. 1979; 98: 203-214Crossref PubMed Scopus (24) Google Scholar), but these have never been mapped. In addition, manyArabidopsis genomic sequences displaying box C/D or H/ACA snoRNA features can be found in the data banks. Nevertheless in the absence of any evidence that their putative targeted nucleotides are modified in the rRNAs, the role of the plant snoRNAs directing rRNA methylation or pseudouridylation remains speculative. Interestingly, the plant snoRNA genes that have been described show a distinct mode of organization and expression. In Arabidopsis thaliana U3 is encoded by three different genes transcribed by RNA polymerase III, whereas in animals the U3 genes are transcribed by RNA polymerase II (30Kiss T. Marshallay C. Filipowicz W. Cell. 1991; 65: 517-526Abstract Full Text PDF PubMed Scopus (87) Google Scholar). A different mode of expression is used by plant U14. In maize U14 is encoded by multiple genes tightly clustered with other snoRNA genes. These clustered genes are driven by a single promoter producing a polycistronic pre-snoRNA. The pre-snoRNA is then processed by endonucleolytic digestion between the snoRNAs and exonucleolytic trimming to release the mature snoRNAs (28Leader D.J. Clark G.P. Watters J. Beven A.F. Shaw P.J. Brown J.W. EMBO J. 1997; 16: 5742-5751Crossref PubMed Scopus (91) Google Scholar). Here we report the identification of two genes, AtFib1 andAtFib2, encoding both fibrillarin proteins and two isoforms of an intronic box C/D snoRNA in A. thaliana. We first demonstrate that AtFib1 and AtFib2 proteins are the functional homologs of the yeast Nop1p and partially complement a yeastNOP1-null mutant strain. Then, and for the first time in plants, we describe novel box C/D snoRNAs nested in the fifth intron of both AtFib1 and AtFib2 genes and show that they accumulate after splicing of the fibrillarin pre-mRNAs. Moreover, we show that the target nucleotide of these novel box C/D snoRNAs on the 25 S rRNA is methylated in vivo. We discuss these results and their implications concerning the mechanism of rRNA methylation in plants related to other eukaryotes. If not specified, 15-day-old A. thaliana Columbia 0 plants cultivated in growth chamber with a 16-h day were used. Radish (Raphanus sativus National Rond Rose a bout rond) were cultivated in a greenhouse under daylight conditions. We identified several ESTs showing strong similarity to known fibrillarin sequences in data bases. One of these ESTs (accession no. N38537) was used to screen an A. thaliana Col. 0 cDNA library of immature siliques. This cDNA library was constructed in λ-ZapII phage and kindly provided by J. Giraudat (Gif-sur-Yvette, France). Hybridization was carried out at 65 °C in 5 × SSC (0.75m NaCl, 0.075 m sodium citrate), 5 × Denhardt's solution (0.1% bovine serum albumin, 0.1% Ficoll, 0.1% polyvinylpyrrolidone), 0.5% SDS, and 25 μg/ml calf thymus DNA. Nitrocellulose membranes (BA85, Schleicher & Schuell) were washed twice for 10 min in 2 × SSC, 0.1% SDS at room temperature and once for 15 min in 1 × SSC, 0.1% SDS at 65 °C. Phages corresponding to positive clones following screening were plaque purified three times before characterization and sequence analysis. A. thaliana Col. 0 and radish genomic DNA were extracted according to Doyle and Doyle (31Doyle J.J. Doyle J.L. Focus. 1989; 12: 13-15Google Scholar). Total RNA was extracted as described previously (32Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-159Crossref PubMed Scopus (63145) Google Scholar). RNA was first subjected to a DNase treatment for 30 min at 37 °C in 100-μl reactions with 5 units of RQ1-DNase (Promega), 20 units of RNasin in 40 mm Tris-HCl, 10 mm MgSO4, 1 mm CaCl2, pH 8.0, and then phenol-chloroform extracted before ethanol precipitation. If not specified, all techniques were performed according to standard protocols (33Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). For cloning of the fifth intron of AtFib2, the full intron was first amplified by PCR using 500 ng of A. thaliana Col. 0 genomic DNA and 10 pmol of oligonucleotides AtFib2In5-1 and AtFib2In5-R2 with the following standard cycling steps: 94 °C for 45 s, 50 °C for 45 s and 72 °C for 1 min. The PCR products were purified and inserted into the pGEM-T easy vector (Promega) by T-A cloning yielding the pAtFib2In5 clone. The same strategy and oligonucleotides were used for cloning the fifth intron of the radish fibrillarin gene. The constructs pG3RPAtFib1 and pG3RPAtFib2 used for RNase protection were obtained by cloning the PCR-amplified 5′-untranslated region of theArabidopsis fibrillarin genes with 10 ng of the corresponding cloned cDNAs and the couples of oligonucleotides AtFib1RP-1 and AtFib1RP-R2 for AtFib1 and AtFib2RP-1 and AtFib2RP-R2 for AtFib2. The amplified fragments were subcloned into the pGEM-T easy vector (Promega) and reinserted into theEcoRI restriction site of the pGEM-3Zf(+) vector (Promega). For translational fusions with green fluorescent protein (GFP) and expression in onion cells, AtFib2 ORF was cloned downstream from a duplicated 35S promoter in theNcoI restriction site of a ppk100 vector (derived from pEGFP1 CLONTECH and kindly provided by R. Blanvillain). NcoI restriction sites were incorporated in AtFib2 cDNA by PCR using oligonucleotides NcoAtfib2 and NcoAtfib2-R. For expression of AtFib1, AtFib2, and Nop1p in yeast, all ORFs were amplified by PCR incorporating NotI sites at their extremities with the following oligonucleotides: NotAtFib1-H and NotAtf-IR for AtFib1, NotAtFib2-H and NotAtFib2-IR for AtFib2, Notnop1-E3 and Notnop1-FR for Nop1. The PCR products were then inserted downstream from a yeast alcohol dehydrogenase promoter in theNotI restriction site of the shuttle vector pADNs that carry a Leu2 selectable marker (34Colicelli J. Birchmeier C. Michaeli T. O'Neill K. Riggs M. Wigler M. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 3599-3603Crossref PubMed Scopus (162) Google Scholar). All cloned fragments were sequenced using the Prism Ready Reaction dye deoxy terminator cycle sequencing Kit and an Applied Biosystems 377 automated DNA sequencing apparatus. Onion epidermal layers (Allium cepa) were transfected according to Saez-Vasquez and co-workers (35Sáez-Vásquez J. Gallois P. Delseny M. Plant Sci. 2000; 156: 35-46Crossref PubMed Scopus (35) Google Scholar) using 5 μg of a plasmid containing a AtFib2-EGFP1 translational fusion. Fluorescence was observed 24 h after transfection using an Axioplan microscope (Zeiss). The riboprobes were obtained by transcription of the desired sequences in antisense orientation with the SP6 or T7 RNA polymerases (Promega) and incorporating 5 μl of radiolabeled 5,000 Ci/mmol [α-32P]CTP. All probes were gel purified on a 7 m urea and 8% polyacrylamide gel. For detection of U60.2f, the pAtFib2In5 construct linearized byNcoI was used. For detection of AtFib1 and AtFib2 mRNAs, constructs ppG3RPAtFib1 and pG3RPAtFib2 were linearized by aBamHI digestion. RNase A/T1 protection was performed as described previously (36Goodall G.J. Wiebaur K. Filipowicz W. Methods Enzymol. 1990; 181: 148-161Crossref PubMed Scopus (149) Google Scholar). For the detection of the snoRNA mature 5′-ends, primer extension was essentially performed according to standard procedures (37Boorstein W.J. Craig E.A. Methods Enzymol. 1990; 180: 347-369Crossref Scopus (144) Google Scholar) with few modifications. Briefly, 5 μg of total DNase-treated RNA was hybridized to 0.5 pmol of specific 5′-end labeled oligonucleotides passed over a G-25 Microspin column (Amersham Pharmacia Biotech): oF1 for U60.1f and UF2 for U60.2f. Hybridization was carried out in 10 μl of aqueous hybridization buffer overnight at 45 °C and reverse transcription performed for 1 h at 42 °C in 20-μl reactions containing 1 mm dNTPs, 20 units of RNasin (Promega), 1 × avian myeloblastosis virus-RT buffer, and 10 units of AMV-RT (Promega). Products were analyzed on a 7 m urea and 8% polyacrylamide sequencing gel. For detection of ribose methylations, the protocol is similar except that three reactions were performed in parallel using 2 μg of RNA and various concentrations of dNTPs ranging from 1 to 0.004 mm. Oligonucleotides At25S-R1 and At25S-R5 were used for detection of ribose-methylated at positions 2114/2116 and 2781, respectively, on the 25 S rRNA. Ladders were obtained by performing sequencing reactions by the Sanger method with the fmol DNA cycle sequencing system (Promega) and 1.5 pmol of the corresponding 5′-end-labeled primer. Growth and handling ofSaccharomyces cerevisiae were performed as described by Johnston (38Johnston J.R. Rickwood D. Hames B.D. Molecular Genetics of Yeast. IRL Press at Oxford University Press, New York1994Google Scholar). The construction of the D255 strain carryingNOP1 gene under the control of an inducible GAL10 promoter (MATa, URA3-pGAL10::NOP1,leu2–3,112, ade1–100,his4–519) was described in (39Tollervey D. Lehtonen H. Carmo-Fonseca Hurt W.C. EMBO J. 1991; 10: 573-583Crossref PubMed Scopus (269) Google Scholar) and kindly provided by D. Tollervey (Edinburgh, Scotland). Transformation after lithium acetate treatment was carried out as described by Ito et al. (40Ito H. Fukuda Y. Murata K. Kimura A. J. Bacteriol. 1983; 153: 163-168Crossref PubMed Google Scholar). Transformants were first selected on minimal medium lacking leucine, and the presence of the different plasmids was confirmed by PCR. Transformants were subsequently transferred on glucose or galactose media at increasing dilutions. At25S-R1, CCGCAGGGACCATCGCAATGCTTTG; At25S-R5, GCAGGTGTCCTAAGATGAGCTCAACG; AtFib2In5-1, GCTTCTTTTGTTTAACTCACAATGAATGTC; AtFib2In5-R2, CACGGTTTTAAATGGCACAGAAAATTTCAT; oF1, CATCAGAAGCTTGGAGTTTGCAAC; oF2, ACATCAGAAACTTGGAGATTGCAT; AtFib1RP-1, CGTCTTTCGTTCTTCACTTTTAGACAAG; AtFib1RP-R2, GCCCACTACGGCCTCTGTCA; AtFib2RP-1, CTCTCTCATCAAAAGCTTTTCTCCTTG; AtFib2RP-R2, CCACCACGGTCGCTGAAACCT; NotAtFib1-H, CCGCGGCCGCATGAGACCCCCAGTTACAGG; NotAtFib1-IR, CCGCGGCCGCTTATGAGGCTGGGGTCTTTTG; NotAtFib2-H, CCGCGGCCGCATGAGACCTCCTCTAACTGG; NotAtFib2-IR, CCGCGGCCGCTTAAGCAGCAGTAGCAGCCTTTG; NotNop1-E3, GAGCGGCCGCTTAACTCAAATCAACTAAAACAG; NotNop1-FR, CCGCGGCCGCTTATTTCTTCAAACCGCTTC. All oligonucleotides were purchased at EUROGENTEC (Belgium). Restriction sites are underlined. To characterize the plant fibrillarin we identified an EST from A. thaliana whose conceptual translation product showed strong similarity to eukaryotic fibrillarins (41Cooke R. Raynal M. Laudié M. Grellet F. Delseny M. Morris P.C. Guerrier D. Giraudat J. Quigley F. Clabault G. Li Y.F. Mache R. Krivitzky M. Gy I.J. Kreis M. Lecharny A. Parmentier Y. Marbach J. Fleck J. Clement B. Philipps G. Herve C. Bardet C. Tremousaygue D. Lescure B. Lacomme C. Roby D. Jourjon M.F. Chabrier P. Charpenteau J.L. Desprez T. Amselem J. Chiapello H. Höfte H. Plant J. 1996; 9: 101-124Crossref PubMed Scopus (177) Google Scholar). We used this EST to screen a cDNA library prepared from Arabidopsis immature siliques (see "Experimental Procedures") and cloned two different cDNAs containing an ORF encoding similar proteins of 309 and 320 amino acids, respectively. Both proteins, AtFib1 and AtFib2, have strong similarity to fibrillarins and are 69% and 64% identical to human fibrillarin and yeast Nop1, respectively (Fig.1). Variability is restricted to the NH2-terminal glycine/arginine-rich domain. This domain is characteristic of many nucleolar proteins and is required for the nucleolar localization of these proteins 2F. Barnèche and M. Echeverrı́a, unpublished data. (42Creancier L. Prats H. Zanibellato C. Amalric F. Bugler B. Mol. Biol. Cell. 1993; 4: 1239-1250Crossref PubMed Scopus (76) Google Scholar, 43Sicard H. Faubladier M. Noaillac-Depeyre J. Léger-Silvestre I. Gas N. Caizergues-Ferrer M. Mol. Biol. Cell. 1998; 9: 2011-2023Crossref PubMed Scopus (23) Google Scholar).Arabidopsis fibrillarins also display the putative AdoMet-dependent methyltransferase motifs and domain recently identified in the yeast Nop1p (9Wang H. Boisvert D. Kim K.K. Kim R. Kim S.-H. EMBO J. 2000; 19: 317-323Crossref PubMed Scopus (143) Google Scholar) and in the crystal structure of an archaebacterial fibrillarin homolog (13Niewmierzycka A. Clarke S. J. Biol. Chem. 1999; 274: 814-824Abstract Full Text Full Text PDF PubMed Scopus (192) Google Scholar). This domain is formed by 7 β-sheet and 7 α-helix folds extending over a large conserved region of the protein (Fig. 1). In addition a Prosite search also predicts a AdoMet binding motif in both AtFib1 and AtFib2 between amino acids 203–214 and 214–235, respectively (not shown). Next we performed BlastN alignment (44Altschul S.F. Stephen F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Nucleic Acids Res. 1997; 25: 3389-3402Crossref PubMed Scopus (59748) Google Scholar) with the two cDNAs on theArabidopsis genomic data banks and identified the corresponding genes AtFib1 on chromosome 5 (BAC AB019226) and AtFib2 on chromosome 4 (cosmid ATL73G19). Alignment of the cDNAs with the genomic sequences revealed six introns interrupting the ORF coding region in each gene (Fig.2). In addition, TBlastN alignment of AtFib1 amino acid sequence with the Arabidopsis genomic sequences revealed a third gene, AtFib3, located 2.2 kbp downstream from AtFib1 on chromosome 5 (Fig. 2).AtFib3 has a predicted ORF with no stop codon encoding a protein of 292 amino acids with 67% identity to AtFib1 and AtFib 2 (not shown). The ORF coding region is interrupted by four introns with perfect 5′/GT and 3′AG/splice junctions (Fig. 2). No EST could be found in data banks for this gene, and we never detected expression of AtFib3 in Arabidopsis thaliana tissues by reverse transcriptase PCR, so AtFib3 is either expressed only under very particular physiological conditions or is a pseudogene. We also estimated the fibrillarin gene copy number in A. thaliana by Southern blot analysis. Hybridization of the AtFib1 cDNA probe to genomic DNA digested with different restriction enzymes detected two DNA fragments corresponding to theAtFib1 and AtFib2 genes (not shown). This probe did not recognize the AtFib3 gene in agreement with the divergence at the nucleotide level of this third gene. The identification of ESTs corresponding to both AtFib1 andAtFib2 indicates that both genes are expressed. We used RNase A/T1 protection analysis to quantitate their relative expression in different tissues from Arabidopsis. Specific riboprobes F1 and F2 were designed to hybridize to 164- and 209-base divergent regions of the 5′-untranslated regions of AtFib1 and AtFib2 mRNAs, respectively (see "Experimental Procedures"). RNase protection analysis on total seedling RNAs with F1 specifically yielded a fragment of 164 bases as expected from protection of AtFib1 mRNA (Fig.3, lane 2). Conversely, a single 209-base fragment was protected with F2 probe as expected for AtFib2 mRNA (lane 4). No protected RNA was detected using a tRNA sample with both probes simultaneously (lane 5). We detected expression of AtFib1 andAtFib2 in the different tissues with both F1 and F2 probes. In all tissues we detected the two fragments of 164 and 209 bases, indicating that both genes are constitutively expressed (lanes 6–10). AtFib1 has significantly higher expression thanAtFib2 (and probe F1 has lower specific activity than F2), but the ratio of expression between these genes is constant in all tissues. Strongest expression of both genes is found in roots (lane 8), a tissue with a high nucleolar activity, and lowest expression is detected in desiccating siliques (lane 7). Finally, we demonstrated that an AtFib1-GFP protein fusion expressed in transfected onion cells specifically accumulates in the nucleolus as expected (Fig. 4). These results indicate that the nucleolar protein fibrillarin is encoded by two different genes in A. thaliana. The sequence conservation of the Arabidopsisfibrillarins with yeast Nop1p suggests that they are functional homologs. To test this, we complemented a yeast NOP1 null mutant strain depleted in Nop1p with AtFib1 and AtFib2. NOP1is an essential gene in yeast so we complemented aGAL::nop1 strain carrying a NOP1gene under the control of a GAL10 promoter (39Tollervey D. Lehtonen H. Carmo-Fonseca Hurt W.C. EMBO J. 1991; 10: 573-583Crossref PubMed Scopus (269) Google Scholar). Growth of this strain depends on galactose and is inhibited in glucose medium, thus only yeast transformants expressing a functional fibrillarin/Nop1p should grow on glucose medium. The GAL::nop1 strain was transformed with the yeast vector pADNs directing constitutive expression of AtFib1, AtFib2 or Nop1p (see "Experimental Procedures"). We first confirmed that all transformants, including those with an empty pADNs vector, grew at similar rate on galactose (Fig. 5 A). On glucose medium, growth of transformants with the empty pADNs was arrested (Fig.5, rows 8 and 12) whereas transformants expressing Nop1p grew as well as in galactose media (compare rows 3 and 7). Strains transformed with a vector expressing AtFib1 and AtFib2 also grew at a significant rate on glucose media (rows 5, 6 and 9, 10) compared with the strain carrying the empty vector (rows 8and 12). Nevertheless they grew at a lower rate compared with the transformants expressing the Nop1p (rows 7 and11). These results demonstrate that AtFib1 and AtFib2 can rescue the NOP1 null-mutant strain and thus are functional homologs of the yeast Nop1p, but complementation is partial. The AtFib1 and AtFib2 genes are within a 447-kbp fragment found both in chromosomes 4 and 5 resulting from large ancestral duplications of the Arabidopsis genome (45Blanc, G., Barakat, A., Guyot, R., Cooke, R., and Delseny, M.Plant Cell, in pressGoogle Scholar). To assess the evolution of these genes after duplication we compared the nucleotide sequences of AtFib1 andAtFib2 using dot-matrix analysis (46Sonnhammer E.L.L. Durbin R. Gene (Amst.). 1995; 167: 1-10Crossref PubMed Scopus (662) Google Scholar). The dot-plot reveals that the exonic sequences coding the ORF are very well conserved (Fig.6 A). Only the sequences encoding the variable NH2-terminal glycine/arginine-rich domains in the second exon show some divergence (Fig. 6 A). By contrast all intronic sequences diverged completely with the remarkable exception of 70 bp in intron 5 (Fig. 6 A). Alignment of the c