Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (562K) Freely available
Right arrow Supplementary Material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Liu, M.
Right arrow Articles by Pape, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, M.
Right arrow Articles by Pape, L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research, 2002, Vol. 30, No. 24 5347-5359
© 2002 Oxford University Press

Characterization of the fission yeast ribosomal DNA binding factor: components share homology with Upstream Activating Factor and with SWI/SNF subunits

Meilin Liu, Ailan Guo, Boris Boukhgalter, Kaitlin Van Den Heuvel1, Matthew Tripp1 and Louise Pape*,1

Department of Chemistry, New York University, New York, NY 10003, USA and 1 Department of Genetics and Biotechnology Center, 445 Henry Mall, University of Wisconsin-Madison, Madison, WI 53706, USA

*To whom correspondence should be addressed. Tel: +1 608 265 7935; Fax +1 608 262 2976; Email: lpape{at}facstaff.wisc.edu
Present addresses:
Meilin Liu, Imclone, New York, NY, USA
Ailan Guo, Consensus Pharmaceuticals, Inc., Medford, MA, USA
Boris Boukhgalter, Whitehead Institute, Cambridge, MA, USA
Matthew Tripp, Stanford University, Palo Alto, CA, USA

Received as resubmission October 17, 2002; Accepted October 25, 2002


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
A ribosomal DNA (rDNA) binding activity was previously characterized in fission yeast that recognized the upstream ribosomal RNA (rRNA) gene promoter in a sequence specific manner and which stimulated rRNA synthesis. It was found to share characteristics with Saccharomyces cerevisiae’s Upstream Activating Factor (UAF), an RNA polymerase I (pol I) specific transcription stimulatory factor. Putative fission yeast homologs of the S.cerevisiae UAF subunits, Rrn5p and Rrn10p, were identified. The Schizosaccharomyces pombe rDNA binding activity/transcriptional stimulatory activity was found to co-fractionate with both SpRrn5h and SpRrn10h. Analysis of polypeptides interacting with SpRrn10h uncovered a 27 kDa polypeptide (Spp27) homologous to a SWI/SNF component (now known to be homologous to Uaf30p). The contributions of the S.pombe and S.cerevisiae upstream rDNA promoter domains were assessed in cross-species transcriptional assays. Furthermore, comparative genomic analysis revealed putative Rrn5p, Rrn10p, Rrn9p and p27 homologs in multiple non-vertebrates. The S.pombe rDNA binding activity is proposed to be an RNA pol I specific SWI/SNF type factor.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
During active cellular growth, synthesis of the large ribosomal RNAs (rRNAs) accounts for nearly half of total RNA synthesis. This high level of rRNA synthesis is driven, in part, by the action of ribosomal DNA (rDNA) transcription factors that associate with the upstream region of rRNA gene promoters and/or intergenic regulatory sequences to mediate stimulation of this synthesis (1–3). In vertebrates, upstream binding factor (UBF), an HMG box protein, is such a factor that associates with rDNA promoters (4) and enhancers (5,6) in a sequence tolerant fashion (7) to increase levels of synthesis up to ~50-fold. UBF functions as a dimer (8); its activity is modified by phosphorylation (9,10), by acetylation (11,12) and by interaction with Retinoblastoma protein (13).

In baker’s yeast, an RNA polymerase I (pol I) transcriptional stimulatory factor, Upstream Activating Factor (UAF), was characterized both genetically and biochemically as being important for synthesis of the large rRNAs (14,15). It was found to stably associate with upstream rDNA promoter sequences in a sequence-specific manner and to consist of six subunits: Rrn5p, Rrn9p, Rrn10p, p30 (Uaf30p), and Histones H3 and H4 (14–16). Lesions in RRN5, RRN9 or RRN10 resulted in a severe decrease in growth rate and in synthesis of large rRNAs (15) and in an increase in switching of the catalytic enzyme for rRNA synthesis from RNA pol I to RNA polymerase II (pol II) (17).

It was unclear whether UBF was present in non-vertebrates or in lower eukaryotes and whether a UAF activity was present in higher eukaryotes. This study focuses on analysis of an rDNA binding factor in fission yeast that interacts specifically with the Schizosaccharomyces pombe upstream rDNA promoter and that increases levels of rRNA synthesis (18). Since it shares characteristics with Saccharomyces cerevisiae UAF, a search was conducted to identify potential S.pombe components of the rDNA binding factor based on homology to Rrn5p, Rrn9p and Rrn10p. Putative homologs were identified for Rrn5p and Rrn10p, and epitope-tagged versions were engineered for investigating molecular interactions and activities leading to species-specific activation of rRNA synthesis.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Identification of putative SpRrn5 and SpRrn10 homologs
Database searches of the S.pombe genome (Wellcome Trust Sanger Institute) uncovered putative homologs of S.cerevisiae Rrn5p and Rrn10p: SpRrn5h (S.pombe Rrn5p homolog, accession no. Q10360) and SpRrn10h (accession no. O14013, putative Rrn10p homolog). Searches for putative homologs of the S.pombe SpRrn5h and SpRrn10h (and later S.pombe p27) were also conducted in other available databases [those at NCBI; Neurospora crassa database at the Whitehead Institute; databases at DOE Joint Genome Institute including white rot fungus (Phanerochaete chrysosporium); Aspergillus fumigatus at TIGR; Candida albicans at the Stanford Genome Technology Center website, http://www-sequence.stanford.edu/group/candida; Cryptococcus neoformans at C.neoformans Genome Project, Stanford Genome Technology Center and The Institute for Genomic Research] and the Genolevures database (19) (http://cbi.labri.u- bordeaux.fr/Genolevures/advanced_blast.php3) using BLASTP, position-specific iterated BLAST (PSI-BLAST; 20) and TBLASTN. Putative homologs of SpRrn5h were identified in N.crassa [686 residues, contig 1.115, Neurospora Sequencing Project. Whitehead Institute/MIT Center for Genome Research (www-genome.wi.mit.edu; V. 3/12/01)], white rot fungus (504 residues, Scaffold 65, DOE Joint Genome Institute), A.fumigatus (preliminary sequence data was obtained from The Institute for Genomic Research website at http://www.tigr.org); C.neoformans (399 residues, CN11020013 AA, database: cneo_orfs011005), Kluyvero myces marxianus (accession no. AL424621) and Kluyveromyces lactis (accession no. AL426913). Putative Rrn10p homologs include: C.albicans Rrn10h (Contig 6-1536), N.crassa (contig 2.273), Aspergillus fumigatus (contig 690), Saccharomyces bayanus (AL402119), K.lactis (AL428395), K.marxianus (AL422400) and Pichia farinosa (AL417122). Multiple protein alignments were conducted using ClustalW (21) [http://searchlauncher.bcm.tmc.edu (Baylor College of Medicine)] and displayed using Boxshade (http://www.ch.embnet.org/).

Construction of pREP-Sp-MH-Rrn5h, pREP-Sp-FLAG-Rrn10h and pREP-Sp-MH-Rrn10h
The forward and reverse primers for PCR amplification of SpRrn5h were 5'-CCCATATGTCTTCTTCAATAAACGG and 5'-GAGATCTCTATTTTAAAGGGAGTGAC, and the template was an S.pombe pACT cDNA library (22; purchased from American Type Culture Collection). For clonings in general, PCR products were first inserted into pCR2.1-TopoTM (Invitrogen) before subcloning into the target vector. The SpRrn5h PCR product was isolated as an NdeI/BglII fragment and ligated into pREP42MH N (23) cleaved with NdeI and BamHI. Coding sequences of the putative S.pombe Rrn10 homolog were amplified from a S.pombe pACT cDNA library (since the coding sequences are interrupted by an intron) using the primers SpR10Fo2, CCCATGGCATCAAATCCACCAACTCG and SpR10Re2, CGGATCCTCATCTTTCTTCAAATTC. The PCR product was isolated as an NcoI-partial BamHI fragment and cloned into pREP-HA-FLAG cleaved with NcoI and BamHI. To construct pREP-HA-FLAG, coding sequences for the HA and the FLAG epitope tags were amplified from pGEM:HA-FLAG (kindly provided by Drs Alexander Hoffman and R. Roeder) with 5'-CGCGTCGACATCCATGGCCACAGATGCTC and 5'-GGAATTCCATATGGGATACCC. This fragment was cleaved with NdeI and SalI, purified, and ligated to NdeI/SalI cleaved pREP, creating pREP-HA-FLAG. pREP-Sp-MH-Rrn10h was made by inserting a PCR fragment (primers SpR10f9 GCCATATGTCAAATCCACCAACTCGGCC and SpR10r9 GCAGATCTTCATCTTTCTTCAAATTCTTCC) into NdeI/BamHI cleaved pREP42MH N.

Preparation and fractionation of extracts from fission yeast
Whole cell extracts (WCEs) were prepared from ~5 l of MP6-10B pREP-Sp-FLAG-Rrn10h pREP-Sp-MH-Rrn5h (‘Sp-MH-Rrn5h/Sp-FLAG-Rrn10h’) cells grown in selective Edinburgh minimal medium (24), as well as MP6-10B pREP-Sp-FLAG-p27 and pREP-Sp-MH-Rrn5h (‘Sp-FLAG-p27/Sp-MH-Rrn5h’) or pREP-Sp-MH-Rrn10h (‘Sp-FLAG-p27/Sp-MH-Rrn10h’). Control WCEs were prepared from 2.5 l of cells expressing just Sp-MH-Rrn10h, Sp-FLAG-Rrn10h or Sp-MH-Rrn5h. The extract proteins (94.1 mg, Sp-MH-Rrn5h/Sp-FLAG-Rrn10h; 89.1 mg, Sp-FLAG-p27/Sp-MH-Rrn5h; and 78 mg, Sp-FLAG-p27/Sp-MHRrn10h) were fractionated by ammonium sulfate precipitation, the pellet recovered and dialyzed, and proteins were then subjected to anion exchange chromatography using three 5 ml HiTrapQTM columns (Amersham Pharmacia Biotech) as described (18). The columns were developed using a KCl step gradient [0.1 M, 0.35 M, 0.7 M KCl–1x column buffer (CB: 25 mM HEPES pH 7.9, 0.2 mM EDTA, 5 mM MgCl2, 20% glycerol, 1 mM PMSF and 1 mM DTT)]. All buffers contain PMSF and DTT, added just prior to use. The RNA pol I transcription activity eluted with 0.35 M KCl–1x CB. Peak fractions were pooled (27 mg total for ‘SpRrn5/SpRrn10h’; 31 mg for ‘Spp27/SpRrn5h’; and 24 mg for ‘Spp27/SpRrn10h’), converted to 1x wash buffer (WB: 25 mM HEPES pH 7.9, 0.15 M KCl and 20% glycerol) and rotated with 1 ml anti-FLAGTM M2 affinity matrix (Kodak/IBI) or 1 ml Ni-NTA agarose matrix (Ni-nitrilotriacetic acid; Qiagen; 1x WB, 0.1% NP-40) for 2 h at 4°C, followed by transfer to a column. For the M2 affinity matrix: following five, 1 ml washes with 1x WB, polypeptides were eluted with 1x WB–125 µg/ml FLAG peptide (five x1 ml). For the nickel affinity matrix, the column was washed five times with 1 ml WB, 0.1% NP-40, 10 mM imidazole, and myc-his-tagged proteins were eluted with 1x WB, 0.1% NP-40, 0.3 M imidazole. Approximately 30 mg of the control extracts (MP6-10B expressing just Sp-MH-Rrn5h; Sp-FLAG-Rrn10h or Sp-MH-Rrn10h) were fractionated as above and in Figure 2D [Sp-FLAG-Rrn10h (Fig. 2C)] or in Figure 7G [Sp-MH-Rrn5h (Fig. 7C) or Sp-MH-Rrn10h (Fig. 7F)] but using one 5 ml HiTrapQ matrix. Concentration and buffer exchange was performed by ultrafiltration using Centricon 10 units.




View larger version (81K):
[in this window]
[in a new window]
  		Figure 2. Fractionation of WCE expressing Sp-FLAG-Rrn10h and Sp-MH-Rrn5h. WCEs prepared from fission yeast expressing both Sp-MH-Rrn5h and Sp-FLAG-Rrn10h were fractionated via ammonium sulfate precipitation followed by chromatographic separation on HiTrapQ columns and affinity purification using a nickel chelating matrix or anti-FLAG (M2) monoclonal antibody matrix. Polypeptides were separated on a 12% (A) or 16% (B) acrylamide gel, transferred to nitrocellulose membrane, and western analysis was performed with anti-myc (A) or anti-FLAG (B) monoclonal antibodies. Lane 1, ~120 µg S-100 protein (WCE); lanes 2–4, 15 µl of 0.1, 0.35 and 0.7 M KCl–1x CB eluates from HiTrap Q matrix; lanes 5 and 6, 15 µl of flowthrough (Ni-NTA FT) or 0.3 M imidazole wash of the nickel affinity matrix (Ni-NTA El). The immunoreactive bands corresponding to the ~75 kDa Sp-MH-Rrn5h are labeled ‘SpRrn5h’, and Sp-FLAG-Rrn10h, ‘SpRrn10h’. The fraction shown in (A) and (B), lane 6 was used for further analysis. (C) Western analysis of fractionation of WCE prepared from MP6-10B pREP-Sp-FLAG-Rrn10h. 120 µg of WCE protein and 15 µl of fractions were analyzed as in (B). The ~15.1 kDa immunoreactive band detected with M2 monoclonal antibody and corresponding to Sp-FLAG-Rrn10h is marked ‘SpRrn10h’. (D) Diagram of fractionation scheme.

 



View larger version (59K):
[in this window]
[in a new window]
  		Figure 7. Fractionation of Sp-FLAG-p27 and associated polypeptides. WCE was prepared from 5 l of S.pombe MP6-10B pREP-FLAG-Spp27 pREP-MH-SpRrn5h (A and B) or MP6-10B pREP-FLAG-Spp27 pREP-MH-SpRrn10h (D and E). (A and B) Following ammonium sulfate precipitation, polypeptides were fractionated via HiTrap Q chromatography (Q). Approximately 120 µg of WCE protein, 15 µl of peak fractions eluted with 0.1, 0.35 or 0.7 M KCl–1x CB (Q) and 15 µl of fractions eluted from the subsequent immunoaffinity matrix (anti-FLAG matrix) were resolved on a 16% (A) or 12% (B) acrylamide gel, transferred to membrane, and reacted with anti-FLAG M2 monoclonal antibodies (A) or anti-myc monoclonal antibodies (B). The immunoreactive bands corresponding to Sp-MH-Rrn5h and Sp-FLAG-p27 are marked. (C) Fractionation of extract prepared from MP6-10B pREP-MH-SpRrn5h. Approximately 120 µg of WCE (lane 1) and 15 µl of fractions were separated by 12% SDS–PAGE as in (B) and probed with anti-myc antibodies. (D and E) Fractionation of extract expressing FLAG-Spp27 and Sp-MH-Rrh10h. Approximately 120 µg of WCE (lane 1), 15 µl of HiTrap Q 0.1 M KCl, 0.35, 0.7–1x CB eluates (lanes 2–4), and 15 µl of M2 affinity flowthrough (lane 5) or M2 affinity eluate (lane 6) were resolved by 16 or 12% SDS–PAGE, and the immunoblots probed with anti-FLAG (D) or anti-myc (E) monoclonal antibodies. (F) Western analysis of WCE prepared from MP6-10B pREP-MH-SpRrn10h cells and fractionated as above. Approximately 120 µg of WCE (lane 1) and 15 µl of fractions (lanes 2–6) were separated by 16% SDS–PAGE and analyzed as in (E). (G) Diagram of fractionation scheme.

 
For western analysis, fractions were resolved on a 16 or 12% SDS–polyacrylamide gel and transferred to nitrocellulose membrane. Concentrations of fractions (in µg/µl) of Figure 2A and B (Sp-MH-Rrn5h/Sp-FLAG-Rrn10h) were Hi Trap Q (Q)—Q-0.1, 1.8 µg/µl; Q-0.35, 2.0; Q-0.7, 0.8; Ni-NTA FT, 1.3; Ni-NTA El, 0.1; Figure 2C (Sp-FLAG-Rrn10h): Q-0.1, 1.9 µg/µl; Q-0.35, 1.8; Q-0.7, 0.6; Ni-NTA FT, 1.3; Ni-NTA El, <0.1; Figure 7A and B (Sp-FLAG-p27/Sp-MH-Rrn5h): Q-0.1, 2.2 µg/µl; Q-0.35, 2.1; Q-0.7, 0.9; M2 aff.-FT, 1.5; M2 aff.-El, <0.1; Figure 7C (Sp-MH-Rrn5h), Q-0.1, 2.2 µg/µl; Q-0.35, 2.3; Q-0.7, 0.7; M2 aff.-FT, 1.6; M2 aff.-El, <0.1; Figure 7D and E (Sp-FLAG-p27/Sp-MH-Rrn10h), Q-0.1, 2.0 µg/µl; Q-0.35, 1.9; Q-0.7, 0.7; M2 aff.-FT, 1.5; M2 aff.-El, <0.1; Figure 7F (Sp-MH-Rrn10h): Q-0.1, 2.1 µg/µl; Q-0.35, 1.8; Q-0.7, 0.6; M2 aff.-FT, 1.6; M2 aff.-El, <0.1. Either anti-FLAG, M2 monoclonal antibody (Kodak/IBI; for Sp-FLAG-Rrn10h and Sp-FLAG-p27) or anti-myc monoclonal antibody [9E10 (BabCo); Sp-MH-Rrn5h or Sp-MH-Rrn10h] was used at 1:1500 dilution for immunodetection as described (25), with the secondary peroxidase linked anti-mouse antibody at 1:5000 dilution. Antibody/polypeptide complexes were detected with Enhanced Chemiluminescent reagents (Amersham Pharmacia Biotech). Polypeptide size markers were Kaleidoscope and broad range standards (Bio-Rad).

Two-hybrid analysis
For cloning into the parental vectors pGAD and pGBDU (26), SpRrn5h sequences were amplified using 5'-GGAATTCATGTCTTCTTCAATAAACGGATTAAATG and 5'-TGA AGATCTTTTTAAAGGGAGTGACAAC. The PCR fragment was isolated as an EcoRI/BglII fragment and inserted into pGBDU-C1, creating pGBD-SpRrn5h, while SpRrn10h cDNA was amplified using CGAATTCATGTCAAATCCACCAACTCGGC and CGTCGACTCATCTTTCTTCAAATTCTTCC. It was inserted into EcoRI/SalI cut pGBDU-C1 and pGAD-C1, creating pGBDU-SpRrn10h and pGAD-SpRrn10h. Plasmids were transformed into PJ69-4A (MATa trp1-901 leu2-3,112 ura3-52 his3-200 gal4{Delta} gal80{Delta} GAL2-ADE2 LYS2::GAL1-HIS3 met2::GAL7-lacZ (26), and activation of the three reporters, ADE2, HIS3 and lacZ, was assessed. PJ69-4A pGBDU-SpRrn10h was transformed with an S.pombe MatchmakerTM cDNA library (in pGAD; Clontech), to yield 5.5 million Ura+ Leu+ transformants. Transformants that independently activated all three reporters (i.e., Ade+ His+ and activated beta galactosidase) were further analyzed (M.Tripp and L.Pape, manuscript in preparation). One of interest contained coding sequences for a predicted 27 kDa polypeptide (Sanger Centre, SPCC285.17).

Construction of pREP-Sp-FLAG-p27 and full-length pGBD-Spp27
The primers used to amplify full-length S.pombe p27 (Spp27) sequences from a S.pombe pACT cDNA library (to avoid amplifying the three introns) were 5'-CCCATGGCAGAAGAGTACGAAACAGAC and 5'-CGGATCCTCAAGCGGTGGATTCGGCGG. The resultant NcoI/BamHI fragment was cloned into pREP-HA-FLAG cut with the same enzymes. The primers 5'-CGGATCCATGGAAGAGTACGAAACAGAC and 5'-CAGATCTCAAGCGGTGGATTCGGCGGTAG were used to amplify Spp27 cDNA for insertion into BamHI and BglII cleaved pGBDU-C1 as a BamHI/BglII fragment. The resultant plasmid was pGBDU-Spp27.

Alignment of S.pombe p27/S.cerevisiae Uaf30p homologs
Putative S.pombe p27 (AL031545) homologs include: white rot fungus (P.chrysosporium), DOE JGI, Scaffold 27; N.crassa p27 homolog, contig 1.821, NCU02204.1, Neurospora Sequencing Project, Whitehead Institute/MIT Center for Genome Research (www-genome.wi.mit.edu; V. 3/12/01); Drosophila melanogaster p27 homolog, AAF47684.1; Arabidopsis thaliana AAD43149.1 (386 amino acid); AAL24145 (385 amino acid) BAB01706 (452 amino acid), and ~13 additional ones; Bombyx mori (AU006413.1); S.cerevisiae Uaf30p (YOR295W; 16), Caenorhabditis elegans p27 homolog (AAA50726.1); Oryza sativa (AAK63939.1); Apis mellifera (bee; BI516000); Hordeum vulgare (BF621939); Chlamydomonas reinhardtii (BI528153); Chlamydia muridarum (NP_297118). Additional putative p27 homologs include cotton, tomato, corn, moss, soybean, C.albicans, A.fumigatus, Debaryomyces hansenii, etc.

Electrophoretic mobility shift analysis with SpRrn5h/SpRrn10h fraction
A 32P-5'-end-labeled fragment (0.01 pmol) bearing the S.pombe rDNA promoter (–243 to +31; labeled at the BamHI and EcoRI sites in p3'{Delta}+31 and electrophoretically purified) was incubated with 7.5 µl of immunopurified Sp-FLAG-Rrn10h/Sp-MH-Rrn5h and associated polypeptides (~60 ng/µl) for 30 min at 25°C. Competitive binding reactions contained purified, unlabeled fragments bearing S.pombe rDNA promoter sequences (–150, –150 to +89; –84, –84 to +89; –57, –57 to +89 isolated from p5'{Delta}-150, p5'{Delta}-84, and p5'{Delta}-57, or a 190 bp pBS SK+ HpaII fragment) at a 10- or 30-fold molar excess to the labeled rDNA promoter fragment. Electrophoresis was as described (18) except that the gel was 4% acrylamide–0.068% bisacrylamide–1x gel shift buffer–5% glycerol.

Construction of S.cerevisiae and S.pombe rDNA templates
The S.pombe wild-type template p3'{Delta}+31, containing a full promoter from –243 to +31, was previously described (27,28). An analogous full-length S.cerevisiae rDNA template was constructed by PCR amplification of sequences between –243 and +44 with ScF-243 5'-CGGGATCCTTCCGCAGTAAAAAATAG-3' and ScR+44 5'-CCGGAATTCGTTTCCAAACTCTTTTCG-3'. Template DNA was YEPrR8 (kindly sent by Dr J. Warner; 29). The fragment was cleaved with BamHI and EcoRI, gel isolated, and ligated between the BamHI and EcoRI sites of pBS SK+ to yield Sc-243/+44. PCR amplifications were cycled at: 94°C, 1 min; 55°C, 2 min; 72°C, 2 min for 27 cycles. A BamHI–PstI fragment containing sequences from –243 to –56 of the S.cerevisiae rDNA promoter was amplified with ScF-243, and ScR-56, 5'-AGAACTGCAGTCCCATTACAAACTAAAATC. Schizosaccharomyces pombe rDNA sequences from –49 to +31 were amplified with SpF-49 5'-AGAACTGCAGACCACAAATGTTTCTATA and T7 primer and p3'{Delta}+31 as template. The PCR product was cleaved with PstI, isolated and ligated into PstI cleaved pSc-243/-56. The orientation of the S.pombe rDNA was assessed by dideoxy sequence analysis. The primers ScR+44 (see above) and ScF-50, 5'-AGAACTGCAGGTTTAGTCATGGAGTAC, were used to amplify the S.cerevisiae core rDNA promoter (-50 to +44); the fragment was cleaved with PstI and EcoRI and inserted into pBS SK+ linearized with these same enzymes, to create ‘Sc core’.

The S.pombe upstream promoter region from –243 to –56 was amplified with T3 and SpR-56 (5'-AGAACTGCAGAATTTCGTCCATTTCGG) primers, using p3'{Delta}+31 as template. Trimolecular ligation of the BamHI–PstI fragment Sp-243/-56, the PstI–EcoRI fragment Sc-50/+44 and vector pBS SK+ cleaved with BamHI and EcoRI yielded the hybrid template Sp/Sc. The core promoter region of S.pombe, from –57 to +89, is in p5'{Delta}-57 (28). Plasmid DNAs were purified and concentrations quantitated as described (18). To construct LS-106/-97, 3'-107, 5'-CGGATCCGTTTCTCTCCACCAAAAG-3' was used with T3 to amplify rDNA from 3'{Delta}+31. The fragment was isolated as a BamHI/XhoI fragment and inserted into BamHI/XhoI cleaved 5'{Delta}-97 (sequences from –106 to –97 were AACGGATCCG instead of the wild-type GGAGGGATAT). For LS-68/-58, 5'-CGGATCCGCGTCGGTGCCTATTTCC and T7 were used to prime synthesis from 3'{Delta}+31. The BamHI/XhoI fragment was inserted into BamHI/XhoI cleaved 5'{Delta}-57, replacing wild-type sequences between –68 and –58, AATGGACGAA, with CGCGGATCCG.

In vitro transcription reactions
S-100 extracts were prepared from S.pombe MP6-10B (27) and S.cerevisiae W303 (30). Reaction conditions were as described (25), with template concentrations as noted in figure legends. Reconstitution reactions (Fig. 3) contained 5 µl of S.pombe RNA pol I/Rrn3p complex [(134 ng/µl); purified using a tagged SpRpa43 homolog and shown to contain S.pombe Rrn3p; M.Liu and L.Pape, unpublished data], 5 µl of a fraction with transcription initiation factor activity [affinity purified based on association of activity with a tagged S.pombe Rrn7p/TAFI-68 homolog (28)], 5 µl of nickel affinity purified SpRrn5h/SpRrn10h or anti-FLAG M2 affinity matrix purified Spp27/SpRrn5h, and template at 0.17 µg/ml. The rNTPs were added after a 15 min pre-incubation, and 20 U RNasin (Promega) was present in each reaction. Transcripts were detected by S1 analysis as described (27). The single-stranded probe for detecting rDNA transcripts from Sc, Sc core, or Sp/Sc was prepared by digestion of the Sc plasmid with SalI, dephosphorylation with calf-alkaline phosphatase, 5' end-labeling (at +76 on the template strand), cleavage with BamHI, and isolation of the single-stranded fragment as described (27). The probe for detecting specific transcripts from p3'{Delta}+31 or Sc/Sp was made by cleaving p3'{Delta}+31 at the unique XhoI site (position +76) or, for Sp core, by cleaving 5'{Delta}-57 at EcoRI, position 107, or XhoI, position 140) and labeling as described. Markers were end-labeled pBR322 HpaII fragments. Quantitation was performed by scanning the gels using a GS-250 Molecular ImagerTM (Bio-Rad).



View larger version (61K):
[in this window]
[in a new window]
  		Figure 3. (A) Transcriptional analysis of affinity purified SpRrn5h/SpRrn10h fractions. In vitro transcription reaction products supported by the rDNA template 3'{Delta}+31 (at 0.48 µg/ml final concentration) were assessed by S1 analysis. Reactions contained either purified RNA pol I/S.pombe Rrn3p alone (‘pol I’, lane 2) or pol I/Rrn3p together with the S.pombe pol I transcription initiation factor [‘init. factor’, lane 3 (28)]. Lane 4 shows transcription supported by pol I, initiation factor, and the affinity purified SpRrn5h fraction also containing SpRrn10h and associated polypeptides. The 76 nt S1 fragment representing correctly initiated rRNAs is indicated with a ‘+1’. Lane 1 contains end-labeled pBR322 HpaII fragments (‘M’). (B) Electrophoretic mobility shift analysis of rDNA binding activity of affinity purified SpRrn5h/SpRrn10h fraction. Lane 1 shows electrophoresis of a fragment with rDNA promoter sequences from –243 to +31, incubated in the absence of protein prior to electrophoresis (lane 1; see ‘free DNA’). The binding reactions of lanes 2 through 11 contained 7.5 µl of the affinity purified SpRrn5h/SpRrn10h, ‘SpR5/R10’ (~60 ng/µl). Binding reactions were conducted in the presence of competitor DNA fragments containing the rDNA promoter from –150 to +89 (lanes 3 and 4), from –84 to +89 (lanes 5 and 6), or from –57 to +89 (lanes 7 and 8), or pBS SK+ fragment (lanes 9 and 10). The (10) and (30) refer to molar excess of the competitor DNA fragments (‘comp. DNA’), and the shifted complex is marked ‘complex’. All reactions were incubated for 30 min at 25°C prior to electrophoresis. The gel was dried and exposed to Kodak XAR-5 film.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Synthesis of wild-type levels of large rRNAs in fission yeast is dependent on the presence of the upstream rDNA promoter domain (28) and on a sequence-specific rDNA binding activity that stimulates rDNA transcriptional initiation (18). Since several properties of the fission yeast rDNA binding activity were reminiscent of S.cerevisiae UAF (14,15), S.pombe polypeptides uncovered as potential homologs of S.cerevisiae UAF subunits Rrn5p and Rrn10p were molecularly analyzed.

Characterization and molecular tagging of SpRrn5h and SpRrn10h
Alignment of a putative fission yeast Rrn5p homolog with S.cerevisiae Rrn5p is shown in Figure 1A. The putative S.pombe Rrn5p homolog (SpRrn5h, 556 residues) has a calculated molecular weight of 64.0 kDa and a 147 amino acid C-terminus not present in S.cerevisiae Rrn5p. The S.cerevisiae Rrn5p and putative S.pombe homolog share 28% identical and 41% similar residues over 176 amino acids (E-value, 6 x 10–5 in a PSI-BLAST search of fungal sequences with S.cerevisiae Rrn5p as query). In addition, SpRrn5h was shown to localize in the nucleolus (31). SpRrn5h contains a domain of particular interest: a SANT domain, located between residues 71 and 112 (E-value, 1 x 10–4, RPS-BLAST). The SANT domain was originally identified from SWI3, ADA2, N-CoR and TFIIIB' and it overlaps a myb domain (32).




View larger version (63K):
[in this window]
[in a new window]
  		Figure 1. (A) Alignment of SpRrn5h and S.cerevisiae Rrn5p. The putative S.pombe Rrn5p homolog is shown from residue 66 to 409 (of 556 residues total), and S.cerevisiae 52 to 344 (of 363 total). The calculated molecular weight of SpRrn5h is 64.0 kDa and its pI is 5.2. (B) Alignment of the putative SpRrn10h (‘SpR10h’) with S.cerevisiae Rrn10p (‘ScRrn10’): top residues 1–36, SpRrn10h; 1–38, Sc Rrn10p; bottom, residues 37–96, SpRrn10h; 61–119, Sc Rrn10p. The putative SpRrn10h (97 residues) has a calculated molecular weight of 11.0 kDa and pI of 4.13.

 
When fungal protein coding sequences are queried with S.cerevisiae Rrn10p, a putative S.pombe homolog (SpRrn10h) is aligned, yielding an E-value of 1.4 [PSI-BLAST (20)]. Although this value is less than threshold, the putative SpRrn10h is the top match besides Rrn10p itself (see Fig. 1B). The putative SpRrn10h shares 29% identical and 49% similar residues with S.cerevisiae Rrn10p over 55 residues. Furthermore, when fungal sequences are searched with SpRrn10h as the query, S.cerevisiae Rrn10p is the top match after SpRrn10h (E-value, 0.61, PSI-BLAST).

Sp-FLAG-Rrn10h cofractionates with Sp-MH-Rrn5h
A fission yeast strain, MP6-10B pREP-Sp-MH-Rrn5h pREP-Sp-FLAG-Rrn10h, was engineered to express N-terminal myc-epitope and poly-histidine (MH)-tagged SpRrn5h fusion protein and FLAG-tagged SpRrn10h. Fractionation of WCE prepared from these cells was conducted to determine whether the putative SpRrn5h and SpRrn10h copurified with each other, with transcriptional stimulatory activity, and/or with rDNA binding activity (Fig. 2D). Western analysis of WCE proteins revealed an anti-myc antibody immunoreactive polypeptide with an apparent molecular weight of ~75 kDa (Fig. 2A, lane 1), corresponding to Sp-MH-Rrn5h (its calculated molecular weight is 68.3 kDa), while a 15.1 kDa polypeptide, corresponding to Sp-FLAG-Rrn10h (calculated molecular weight, 14.5 kDa), was detected with M2 (anti-FLAG) monoclonal antibodies (Fig. 2B, lane 1). Neither of these immunoreactive polypeptides was detected in WCE prepared from control S.pombe cells (data not shown).

WCE proteins were subjected to ammonium sulfate precipitation, followed by fractionation on the anion exchange matrix, HiTrapQTM, and the nickel affinity matrix as described (18). Polypeptides that eluted from the HiTrapQ matrices with 0.35 M KCl column buffer contained all factors required for specific rRNA synthesis (18,25), as well as the putative SpRrn5h and SpRrn10h (as seen in the immunoblot of Fig. 2A and B, lanes 3). The peak fractions of the 0.35 M KCl HiTrapQ eluate were pooled, applied to a nickel affinity matrix and, following extensive washing of the matrix, polypeptides were eluted with 0.3 M imidazole buffer. In a parallel fractionation, extracts were prepared and fractionated as above, but the affinity purification in this alternate fractionation was an M2 (anti-FLAG) affinity matrix for purification of Sp-FLAG-Rrn10h and associated polypeptides.

As seen in Figure 2A, Sp-MH-Rrn5h (‘SpRrn5h’) was detected in fractions eluted from nickel affinity matrix with 0.3 M imidazole, as expected (lane 6, ‘Ni-NTA El’; compare to lane 5, ‘Ni-NTA FT’). A parallel immunoblot of proteins in the nickel affinity purified fractions was challenged with anti-FLAG (M2) antibody and revealed Sp-FLAG-Rrn10h largely in the bound fraction (‘Ni-NTA El’, Fig. 2B, lane 6) with a portion in the flow-through fraction (‘Ni-NTA FT’, lane 5). Further evidence for interaction came from co-immunoprecipitation analysis of Sp-MH-Rrn5h with Sp-FLAG-Rrn10h (data not shown). To assess the extent of binding of the putative Sp-FLAG-Rrn10h to the Ni-NTA matrix in the absence of Sp-MH-Rrn5h, WCE was prepared from S.pombe expressing just Sp-FLAG-Rrn10h and fractionated as in Figure 2D. As seen in Figure 2C, lane 5, SpRrn10h does not bind the Ni-NTA matrix in the absence of Sp-MH-Rrn5h but is present in the flow-through fraction (compare lane 5 with 6).

Affinity purified SpRrn5h/SpRrn10h fraction harbors sequence-specific rDNA binding and transcriptional stimulatory activities
Subsequently, the affinity purified Sp-MH-Rrn5h/Sp-FLAG-Rrn10h fraction (‘SpR5/SpR10’) was tested for its ability to confer stimulation of transcription supported by a full-length S.pombe rRNA gene promoter in in vitro reactions. These reactions were conducted under stringent conditions that reflect requirements for the upstream rDNA promoter region and for stimulatory transcription factors in directing efficient initiation. Correctly initiated rRNAs are represented by 76 bp S1-resistant products. Virtually no accurately initiated rDNA transcripts are detected when reactions contain either the RNA pol I fraction (RNA pol I/S.pombe Rrn3p; see Fig. 3A, lane 2) or RNA pol I and the rDNA transcription initiation factor [‘init. factor’, lane 3 (28)]. However, when the affinity purified SpRrn5h/SpRrn10h fraction was also present in the reactions, the level of rDNA transcriptional initiation was significantly stimulated (see lane 4). The SpRrn5h/SpRrn10h fraction does not support correct initiation on its own (data not shown). Furthermore, Sp-MH-Rrn5h as well as the rDNA transcriptional stimulatory activity were found to immunopurify with Sp-FLAG-Rrn10h using an anti-FLAG affinity matrix (data not shown).

If SpRrn10h and SpRrn5h are components of the S.pombe sequence-specific rDNA binding activity, this activity should copurify with these tagged polypeptides. The presence of an rDNA binding activity was assessed by electrophoretic mobility shift analysis. When an end-labeled fragment bearing S.pombe rDNA promoter sequences from –243 to +31 was incubated in the presence of the affinity purified SpRrn5h– SpRrn10h fraction, the electrophoretic mobility of the labeled rDNA promoter fragment was retarded (see ‘complex’, compare Fig. 3B, lane 2, with lane 1). In order to assess the specificity of this association, binding reactions were conducted in the presence of 10- or 30-fold molar excess of unlabeled competitor fragments bearing rDNA sequences from –150 to +89 (lanes 3 and 4) and –84 to +89 (lanes 5 and 6). Both competitors served to largely abolish the retarded complex. In contrast, 10-fold molar excess of a fragment bearing rDNA from –57 to +89 (lane 7) or non-specific vector fragment (pBS; lanes 9) does not affect formation of the complex (while 30-fold molar excess of –57 to +89 does; lane 8). Thus, the 5' border of the S.pombe rDNA promoter that is critical for directing association of the SpRrn5h/SpRrn10h fraction with rDNA lies between –84 and –57. This correlates with the rDNA promoter region, between –80 and –56, shown to be stably bound by the S.pombe rDNA binding activity (18), and with the 5' border required for mediating stimulated levels of rRNA synthesis that was also found to lie between –84 and –57 (28).

An illustration of the importance of this rDNA promoter region is seen in transcriptional analysis of rDNA promoters mutated in this region. The rDNA promoter of the template LS-68/-58 is full-length (from –243 to +89) but bears non-specific bases between –68 and –58. Comparison of its promoter strength with that of a wild-type template (3'{Delta}+31) or a neutral mutant rDNA template, LS-106/-97, reveals that it is severely impaired in its ability to support rDNA synthesis (compare Fig. 4A, lane 2 with lanes 1 and 4). In contrast, LS-35/-28, bearing non-specific bases between –35 and –28 (28), is only somewhat impaired (compare lane 3 with its wild-type control, lane 4).





View larger version (111K):
[in this window]
[in a new window]
  		Figure 4. (A) In vitro transcriptional analysis in S.pombe WCE supported by LS-106/-97 (lane 1), LS-68/-58 (lane 2), LS-35/-28 [lane 3 (28)] and the wild-type 3'{Delta}+31 (lane 4) was assessed by S1 analysis. The S1-protected fragment representing correctly initiated rRNAs supported by 3'{Delta}+31 and LS-35/-28 is 76 nt (‘+1*’), and is 140 nt for LS-106/-97 and LS-68/-57 (‘+1’). Template concentration was 125 ng/ml. (B) Analysis of transcriptional strength of hybrid S.cerevisiae–S.pombe rDNA promoters. Transcription reactions were conducted with S.pombe S-100 extract and templates at 20 or 50 ng per reaction (0.5 or 1.25 µg/ml). Lanes 1 and 2, wild-type S.cerevisiae rDNA template (Sc); lanes 3 and 4, S.cerevisiae rDNA core promoter (Sc core); lanes 5 and 6, hybrid template with S.pombe upstream rDNA promoter and S.cerevisiae rDNA core promoter (Sp/Sc); lanes 7 and 8, S.cerevisiae upstream rDNA promoter and S.pombe core (Sc/Sp); and lane 10, wild-type S.pombe rDNA template (Sp). Both single-stranded probes had a comparable specific activity. (C) Transcription supported by Sc/Sp (lanes 1 and 2), full-length S.pombe rDNA promoter (lane 3), and Sp core (lane 4). The 20 or 50 refers to ng template present in the reaction (0.5 or 1.25 µg/ml template). The ‘+1’ marks the 76 nt S1 protected fragment representing correctly initiated rRNAs from templates in lanes 1–3, while ‘+1 core’ is a 140 nt S1-protected fragment representing correctly initiated rRNAs from Sp core template. (D) Diagram of the hybrid rDNA templates used in this study. (E) Sequence of the hybrid and wild-type rDNA promoters at the junction site. Schizosaccharomyces pombe rDNA sequences are underlined, PstI linker bases are in italics and S.cerevisiae rDNA sequences are plain.

 
Analysis of cross-species function of S.cerevisiae and S.pombe upstream rDNA promoter regions
Since the respective S.cerevisiae or S.pombe upstream rDNA promoter domains mediate the effects of S.cerevisiae UAF (15) or the S.pombe rDNA binding activity (18), the cross-species activities of these domains were tested in in vitro S.cerevisiae and S.pombe transcription assays on templates bearing hybrid S.pombe/S.cerevisiae rDNA promoters. The hybrid templates bore the S.pombe upstream rDNA promoter fused to an S.cerevisiae core rDNA promoter (Sp/Sc) or the S.cerevisiae upstream promoter fused to the S.pombe core promoter (Sc/Sp), while the control rDNA templates consisted of full-length S.pombe (Sp) or S.cerevisiae rDNA promoters (Sc) or the ‘Sc core’ and ‘Sp core’ rDNA templates that lacked the upstream promoter domains entirely (see Fig. 4D and E).

When the promoter strength of this series was assessed in S.pombe in vitro transcription reactions (see Fig. 4B), no correctly initiated rRNAs were detected supported by the S.cerevisiae core (Sc core; lanes 3 and 4) or full-length rDNA promoter (Sc; lanes 1 and 2). In contrast, the Sp/Sc template does direct accurate rRNA synthesis (see Fig. 4B, lanes 5 and 6), and its start site was fine-mapped to +1 (data not shown). This demonstrates that the S.pombe upstream rDNA promoter can mediate correct initiation of the S.cerevisiae core rDNA promoter with S.pombe RNA pol I transcription factors. In contrast, a cis-located S.cerevisiae upstream rDNA promoter (Sc/Sp) functions to repress basal level transcription of the S.pombe core rDNA promoter (Sp core) (see Fig. 4B, lanes 7 and 8, compare to Sp, lane 10, and Fig. 4C, compare lane 4 with lanes 1 and 2). The molecular basis underlying this repressive effect is unknown.

When transcription reactions were conducted with S.cerevisiae WCE components, the S.cerevisiae full-length rDNA promoter (‘Sc’), as expected, supported efficient transcriptional initiation (see Fig. 5, lanes 3 and 4), while levels supported by the S.cerevisiae core rDNA promoter (from –49 to +44) were reduced 6-fold or more (compare lane 2 with lane 4, and lane 1 with lane 3, Figure 5). A cis-located S.pombe upstream rDNA promoter mediated a modest increase in transcription supported by the S.cerevisiae rDNA core promoter [compare lanes 1 and 2 (Sc core) to lanes 5 and 6 (Sp/Sc)]. In the Sc/Sp template, the cis-located S.cerevisiae upstream rDNA promoter domain increases transcription supported by the S.pombe rDNA promoter 50% compared to the wild-type Sp template (compare lane 9 with lane 11) and several-fold more compared to Sp core (lanes 12 and 13). As seen previously, transcription of the S.pombe rDNA promoter initiates at an altered site [see Fig. 5, lanes 8 to 13 (25)] in S.cerevisiae in vitro transcription reactions. A minor transcript supported by Sc/Sp (lane 9) may represent initiation at the correct +1 site or may be due to non-specific initiation.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 5. Transcriptional analysis of hybrid rDNA promoters conducted in S.cerevisiae S-100 extracts. The rDNA templates bearing Sc core (lanes 1 and 2), Sc (lanes 3 and 4), Sp/Sc (lanes 5 and 6), Sc/Sp (lanes 8 and 9), Sp (lanes 10 and 11) and Sp core (lanes 12 and 13) rDNA promoters were transcribed in vitro in an S.cerevisiae transcription extract, with the final template concentration at 0.5 µg/ml (20 ng; lanes 1, 3, 5, 8, 10 and 12) or 1.25 µg/ml (50 ng; lanes 2, 4, 6, 9, 11 and 13). Correctly initiated S.cerevisiae rRNAs are represented by a 76 nt S1 protected fragment (lanes 1–6). Lanes 8–11, the 67 nt S1 protected fragment represents initiation from Sc/Sp and Sp at +10 (fine-mapped; data not shown). Lanes 12 and 13: The 130 nt S1-protected fragment represents rRNAs initiated at +10 from Sp core. Transcriptional efficiency is noted relative to Sc 50 (lane 4) for lanes 1–6, and relative to Sp 50 (lane 11) for lanes 8–11.

 
A 27 kDa S.pombe polypeptide interacts in vivo with SpRrn10h
To investigate interactions of the putative SpRrn10h and SpRrn5h as well as to search for specifically interacting polypeptides, their coding sequences were cloned in frame with the GAL4 DNA binding domain or activation domain in the vectors pGBDU and pGAD (26). In a large-scale two-hybrid screen, polypeptides that specifically associated with SpRrn10h were searched for in a library of S.pombe GAL4AD-cDNAs (Clontech). One of the GAL4AD-cDNAs that encoded a fusion protein capable of activating three independent reporters together with GAL4BD-SpRrn10h, indicating interaction, was found to encode a ~27 kDa polypeptide. This S.pombe 27 kDa polypeptide was homologous to S.cerevisiae YOR295W [now reported as a subunit of UAF, Uaf30p (16)] and YMR233W (16) and shared some homology with the mammalian ~60 kDa SWI/SNF subunit (33). As seen in Figure 6, the activation of the ADE2 and the HIS3 reporters in PJ69-4A bearing both pGBDU-SpRrn10h (SpRrn10h) and pGAD-Spp27 (Spp27) allowed growth on medium lacking adenine or histidine (the lacZ reporter was also activated; data not shown). In contrast, no activation was observed in strains expressing GAL4AD-Spp27 (Spp27) and GAL4BD-SpRrn5h (SpRrn5h); GAL4BD-Spp27 and S.cerevisiae GAL4AD-ScRrn9p (ScRrn9p); Spp27 and vector; GAL4DB-SpRrn10h (SpRrn10h) and ScRrn9p; or SpRrn10h and SpRrn5h. The reason for the negative two-hybrid result between SpRrn5h and SpRrn10 is unknown, since they co-fractionate with each other (perhaps mediated by another associated factor) and with the S.pombe rDNA binding activity. Furthermore, the documented interaction of S.cerevisiae Rrn9p with Rrn10p (34,35) appears to be species-specific in nature and is not seen between S.cerevisiae Rrn9p and S.pombe SpRrn10h.




View larger version (95K):
[in this window]
[in a new window]
  		Figure 6. Two-hybrid analysis of interactions with SpRrn10h. Trans formants expressing GAL4 DNA binding domain (GBD)-SpRrn10h and GAL4 activation domain (GAD)-SpRrn5h, Spp27 or ScRrn9p fusion proteins. The top panels show growth of transformants on medium lacking adenine, leucine and uracil (left) or lacking histidine, leucine and uracil, with 5 mM 3 amino triazole (right). Transformants are PJ69-4A pGBDU-SpRrn10h pGADSpp27 (SpRrn10h x Spp27), PJ69–4A pGBDU-Spp27 x pGADSpRrn5h (Spp27 x SpRrn5h), PJ69-4A pGBDU-Spp27 pGADScRrn9p (Spp27 x ScRrn9p), PJ69-4A pGBDU-Spp27 x pGAD-C1 (Spp27 x vector), PJ69-4A pGBDU-SpRrn10h pGADScRrn9p (SpRrn10h x ScRrn9p), and PJ69-4A pGBDU-SpRrn10h pGADSpRrn5h (SpRrn10h x SpRrn5h). A diagram in the bottom panel depicts positions of the transformants: the top polypeptide listed of each pair is expressed as a fusion protein with the GAL4 DNA binding domain; the bottom of each pair as a fusion with the GAL4 activation domain.

 
Sp-FLAG-p27 cofractionates with Sp-MH-Rrn5h and with Sp-MH-Rrn10h
To test whether Spp27 co-fractionated with SpRrn5h—as expected if Spp27 were a component of the rDNA binding complex—pREP-Sp-FLAG-p27 encoding N-terminal FLAG-tagged Spp27 was introduced into S.pombe MP6-10B together with pREP-Sp-MH-Rrn5h. WCE was prepared and fractionated as described above, except that the affinity purification step used anti-FLAG affinity matrix (see Fig. 7G). Duplicate aliquots of the fractions were resolved on 16 and 12% SDS–polyacrylamide gels, transferred to nitrocellulose membrane and challenged with anti-FLAG (Fig. 7A) or anti-myc (Fig. 7B) monoclonal antibodies. Both Sp-MH-Rrn5h and Sp-FLAG-p27 co-eluted from the HiTrap Q matrix (Fig. 7A and B, lane 3) together with the RNA pol I transcriptional activity (18). Peak fractions eluted with 0.35 M KCl–1x CB were applied to an anti-FLAG, M2 monoclonal antibody affinity matrix. Following extensive washing, bound polypeptides were eluted with FLAG peptide. As expected, Sp-FLAG-p27 is specifically bound to the M2 affinity matrix and dissociates upon addition of FLAG peptide (Fig. 7A, lane 6). The reason for the discrepancy between its apparent molecular weight of ~42 kDa and the calculated value of 30.1 kDa is unknown, as is the presumptive modification underlying the slower migrating immunoreactive band of ~44 kDa. Sp-MH-Rrn5h (~75 kDa apparent molecular weight) is detected in the fraction bound to the M2 affinity matrix, presumably via interactions with Sp-FLAG-p27 [‘SpRrn5h’, Fig. 7B, lane 6, ‘M2 aff.-El’; for Sp-FLAG-p27, see Fig. 7A, lane 6 (‘Spp27’)], although a portion is also present in the flow-through fraction (‘M2 aff.-FT’, Fig. 7B, lane 5). In addition, fractionation of an extract prepared from fission yeast expressing just the epitope-tagged Sp-MH-Rrn5h revealed that it does not specifically bind to the M2 affinity matrix in the absence of Sp-FLAG-p27 (Fig. 7C). Thus, Sp-MH-Rrn5h co-fractionates with Sp-FLAG-p27, suggesting that SpRrn5h, SpRrn10h and S.pombe p27 (Spp27) associate in a UAF-like complex.

WCEs were also prepared and fractionated from S.pombe expressing Sp-FLAG-p27 and Sp-MH-Rrn10h (see Fig. 7D and E, lane 1), as above. Sp-MH-Rrn10h co-elutes from the HiTrap Q matrix with Sp-FLAG-p27 and is largely present in the bound fraction of the M2 affinity matrix (see Fig. 7D and E, lanes 3 and 6), consistent with their in vivo interaction (Fig. 6). In contrast, when WCE prepared from S.pombe cells expressing just Sp-MH-Rrn10h was fractionated as in Figure 7G, Sp-MH-Rrn10h was found in the flowthrough fraction of the M2 affinity matrix (Fig. 7F, lane 5).

The Spp27/SpRrn5h fraction was assayed to determine whether it contained an activity leading to stimulated levels of rDNA transcriptional initiation. Addition of the affinity purified Spp27/SpRrn5h fraction to a reaction containing basal level pol I transcription initiation components and a full-length rDNA promoter resulted in a significant increase in rRNA synthesis (compare lane 5 with lane 3, Fig. 3A), while it supported no initiation itself (data not shown). This Spp27/SpRrn5h fraction also harbored sequence specific rDNA binding activity (data not shown).

Identification of p27, SpRrn5h and SpRrn10h homologs in other eukaryotes
Sequence alignment of the S.pombe Spp27 with other eukaryotic homologs is seen in Figure 8A (and Supplementary Material, Fig. 9A). Extensive database searches reveal that non-vertebrates have a clearly identifiable homolog of Spp27 as well as a homolog of the SWI/SNF 60 kDa subunit. The latter polypeptide, however, shares less sequence homology with Spp27 than the putative Spp27 homologs. A clearly identifiable homolog of Spp27 was not found in the proteomes of vertebrates, although the 60 kDa SWI/SNF subunit homolog was.





View larger version (119K):
[in this window]
[in a new window]
  		Figure 8. (A) Multiple alignment of S.pombe p27 (Sp.p27; 117–190 of 233 residues total) with putative homologs: white rot fungus (wrf.p27h; 264–337 of 363 total), N.crassa (Nc.p27h 186–259 of 265 total), A.thaliana (At.p27h; 251–324 of 386; note—there are ~16 putative A.thaliana p27 homologs), O.sativa (Os.p27h; 67–141 of 144 total), H.vulgare (Hv.p27h), C.muridarum (Cm.p27h; 9–83 of 86 total), S.cerevisiae Uaf30P (Sc.UAF30, 120–192 of 228 total), D.melanogaster (Dm.p27h; 168–241 of 244 total), bee (bee.p27h), B.mori (Bm.p27h), C.elegans (Ce.p27h; 264–337 of 415 total) and C.reinhardtii (Cr.p27h). (B) Multiple alignment of S.pombe SpRrn5h (residues 68–160) and S.cerevisiae ScRrn5p (56–139) with putative Rrn5p homologs: K.marxianus KmRrn5h, K.lactis KlRrn5h, white rot fungus wrfRrn5h (66–165), C.neoformans CnRrn5h (52–147), A.fumigans AfRrn5h (108–196) and N.crassa NcRrn5h (182–275). SpRrn5h, NcRrn5h, wrfRrn5h and AfRrn5h have clearly identifiable SANT domains: SpRrn5h, between residues 71 and 112 [E-value 1 x 10–4 RPS-BLAST (20)]; N.crassa Rrn5h, between 147 and 185 (E-value 1.3 x 10–2); white rot fungus Rrn5h, between 72 and 107 (E-value 6.4 x 10–2) and A.fumigatus from 114 to 158 (E-value 2.1 x 10–2). All have conserved residues in this domain. (C) Alignment of putative Rrn10p homologs: S.cerevisiae Rrn10p (ScRrn10p, 63–110 of 145 total residues), S.bayanus SbRrn10h (63–110 of 145 total), K.marxianus KmRrn10h, K.lactis KlRrn10h (49–96 of 125), P.farinosa PfRrn10h (93–143 of 183 total), C.albicans CaRrn10h (residues 102–155 of 195 total), A.fumigans AfRrn10h; N.crassa NcRrn10h [residues 89–157, with 7 residues removed ‘//’ (of 261 total)], and S.pombe SpRrn10h (residues 39–87 of 97 total).

 
The genomes of N.crassa, P.chrysosporium (white rot fungus), A.fumigatus, C.neoformans, K.marxianus and K.lactis encode putative Rrn5p homologs that align with S.pombe SpRrh5h and S.cerevisiae Rrn5p as seen in Figure 8B. (Full-length alignments are shown in Supplementary Material, Fig. 9B.) The N.crassa, white rot fungus and A.fumigatus Rrn5p homologs, like SpRrn5h, also harbor a clearly identifiable SWI3-like SANT domain and all share conserved residues in this domain. While no definitive Rrn5p homologs have been found in eukaryotes other than in fungi, this may be due to a divergence in primary sequence that masks their recognition.

Similarly, putative Rrn10p homologs identified recently in several fungal databases are shown in Figure 8C. A search of C.albicans genomic sequences homologous to Rrn10p revealed a putative homolog (CaRrn10h). When fungal sequences are queried with the putative C.albicans Rrn10h, S.cerevisiae Rrn10p aligns with an E-value of 1 x 10–6 (PSI-BLAST; 32% identities and 50% similarities over 98 residues), and SpRrn10h with an E-value of 2.2 x 10–2 (29% identical and 53% similar residues over 57 amino acids). A putative N.crassa Rrn10 homolog (NcRrn10h) was also identified based on homology with CaRrn10h that is 50% identical and 63% similar over 52 residues, from 51 to 102, of CaRrn10h. Alignment of other putative yeast Rrn10p homologs from the hemiascomycetous class of yeast (S.bayanus, K.marxianus, K.lactis and P.farinosa), identified in the Genolevures database (19) with S.cerevisiae Rrn10p as query, is also shown in Figure 8C. (See Supplementary Material, Fig. 9C, for full-length alignments.)


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Stimulation of rRNA synthesis in eukaryotes is mediated by cis-acting sequences in the upstream rDNA promoter as well as by the stimulatory factors, UBF in vertebrates and UAF (15) in S.cerevisiae (1–3 and references therein). These factors are vastly different in subunit structure: UBF is a homodimer composed of UBF1 or UBF2 molecules, while UAF is composed of six subunits: Rrn5p, Rrn9p, Rrn10p (15), histones H3 and H4 (14) and Uaf30p (16). UBF is abundant (up to 100 000 molecules per cell), while UAF is present at ~200 copies per cell (15). Both UAF and the S.pombe rDNA binding activity recognize their species’ upstream rDNA promoter in a sequence-specific manner (15,18), in contrast to UBF, which associates with rDNA in a sequence tolerant fashion (7,36). Both serve important roles in stimulating levels rRNA synthesis.

The presence of a UAF-like activity and UAF subunit homologs that co-purify with this activity in S.pombe demonstrate that UAF is not restricted to S.cerevisiae. As reported here, epitope-tagged versions of putative fission yeast SpRrn5h, SpRrn10h and Spp27 were found to co-fractionate with each other and with the S.pombe rDNA binding/rDNA transcriptional stimulatory activity. A fundamental question raised when UAF was first characterized (15) was whether other eukaryotes harbor UAF, and if they do, whether they regulate levels of rRNA synthesis via both UBF and UAF? While clear homologs of Uaf30p/S.pombe p27 have been found in invertebrates, including fungi, plants, D.melanogaster, silk worm, etc., putative Rrn5p homologs and Rrn10p homologs have only been identified in S.pombe and other fungi. These include representatives of basidiomycetes (white rot fungus) and ascomycetes: archaeascomycetes (S.pombe), hemiascomycetes (S.cerevisiae, K.lactis, C.albicans, etc.), and euascomycetes (N.crassa). The apparent lack of non-fungal, invertebrate homologs of Rrn5p and Rrn10p may be due to considerable sequence divergence at the primary structure level. If so, purification of UAF-like complexes via a tagged Spp27 or UAF30p homologous subunit could facilitate determination of the presence of such complexes as well as their subunit composition. Since disruption of YMR233W did not affect cell growth or rRNA transcription despite its shared homology with UAF30p (16), the Spp27 homologs have been called p27h until it is demonstrated that they play a role in rRNA synthesis. Chlamydia, a bacterial human pathogen, has the SWIB-Topoisomerase I fusion (16) and a ‘stand-alone SWIB’ protein (37) that are Spp27 homologs among other chromatin-associated genes phylogenetically derived from eukaryotes.

It is of interest that at least two of the components of the S.pombe rDNA binding factor share homology with components of the SWI/SNF complex. SpRrn5h bears a SANT domain [first noticed in the SWI/SNF subunit SWI3, Ada3, N-CoR, and a TFIIIB component (32)]. Homology to Swi3p, however, extends beyond this domain (see Supplementary Material, Fig. 10). One function of a SANT domain was recently reported: a SANT domain in SMRT and N-CoR was shown to be important for interaction with and activation of histone deacetylase (38).

A second component of the S.pombe rDNA binding activity also bears homology to a SWI/SNF subunit: the SpRrn10h interacting factor Spp27 to SWI/SNF 60 kDa (16; this paper). Taken together, these observations suggest that the S.pombe rDNA binding activity that co-fractionates with tagged SpRrn10h, SpRrn5h, and Spp27 (a UAF30p homolog) functions in chromatin remodeling and is targeted to the upstream rDNA promoter by its sequence-specific recognition of this region. If so, its ability to stimulate rDNA transcription may result, at least in part, from such chromatin remodeling activity (reviewed in 39–41).

While a homolog of the fourth RNA pol I specific subunit of UAF, Rrn9p, has yet to be characterized in S.pombe, putative Rrn9p homologs have been identified in multiple yeasts (see Supplementary Material, Fig. 11A and B). An intriguing observation obtained in a search of the S.pombe proteome with the putative Kluyveromyces thermotolerans Rrn9p homolog, KtRrn9h, as query was that S.pombe Snf2p was the top match (Supplementary Material, Fig. 11C); the putative A.fumigatus Rrn9h was shown to align with Snf2 as well (see Supplementary Material, Fig. 11D). However, since the homologies seen were limited, it remains to be tested whether fungal Rrn9ps share any activity of Snf2p.

The polypeptides with the highest degree of homology to Spp27 are clearly distinct from the mammalian 60 kDa SWI/SNF subunit class. While Spp27 shares some homology with this SWI/SNF subunit, non-vertebrates including S.pombe, S.cerevisiae [Uaf30p (16)], white rot fungus, N.crassa, D.melanogaster, C.elegans and A.thaliana, have Spp27 homologous coding sequences (see Fig. 8A). A search of eukaryotic coding sequences with these Spp27 homologs reveals that they align, in general, to yield E-values significantly higher with each other than with the SWI/SNF 60 kDa subunit homologs. With the exception of the Anopheles gambiae genome, the Spp27 homologs are found in eukaryotic genomes that do not have a clearly identifiable UBF.

The upstream rDNA promoter domain is recognized in a species-specific manner: the S.cerevisiae upstream rDNA promoter domain does not mediate any increase in rRNA synthesis when 5' to either an S.cerevisiae or S.pombe rDNA core promoter domain in S.pombe in vitro transcription reactions. The possibility that different spacing between the upstream promoter and the core domains in the S.cerevisiae and S.pombe hybrid rDNA templates would increase their transcriptional efficiency, as was seen in the case of a Xenopus rDNA template in a heterologous murine transcription system (42), remains to be tested. While the S.pombe upstream rDNA promoter domain mediates a modest increase in rRNA synthesis in S.cerevisiae transcription reactions, it is critical for activating rRNA synthesis directed from the S.cerevisiae core rDNA promoter in in vitro S.pombe transcription reactions. Despite the homology observed between SpRrn5h and S.cerevisiae Rrn5p, and between SpRrn10h and S.cerevisiae Rrn10p, the S.pombe factors do not complement an S.cerevisiae rrn5– or rrn10– mutant strain, respectively (M.Tripp, K.van den Heuvel and L.Pape, data not shown). For SpRrn10h, this may be due to its inability to interact with S.cerevisiae Rrn9p (Fig. 6), in contrast to the interaction between S.cerevisiae Rrn10p and Rrn9p (34).

Recent studies point to nucleolar histone modification and chromatin remodeling activity playing an important role in regulation of eukaryotic rRNA synthesis. Histone modifications and DNA methylation were shown to play a crucial role in activation of rRNA genes in hybrid plants (43,44).A histone acetyltransferase activity was found to co-purify with Xenopus RNA pol I (45), and mammalian rDNA transcription was shown to increase in vivo and in vitro when histone deacetylase was inhibited (12). Furthermore, a nucleolar, chromatin remodeling activity that associates with the RNA pol I termination factor, TTF-I, was characterized in mammals (46), which aids in repressing rRNA synthesis by targeting histone deacetylase (47). The RSC nucleosome remodeling complex has been found to target RNA polymerase III transcribed promoters as well as its pol II targeted promoters (48). The hypothesis that fungal UAF complexes are RNA pol I specific chromatin remodeling factors, suggested based on the homology several subunits share with those of the SWI/SNF complex, can now be molecularly tested.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at NAR Online.


   ACKNOWLEDGEMENTS
 
This work was supported by NSF Grants MCB-9996390, MCB-9219220 and MCB-9604724 to L.P. NSF is also thanked for its support of the NYU Medical Center computing resources through grant BIR-9318128. Sequence data for N.crassa was obtained from the Neurospora Sequencing Project, Whitehead Institute/MIT Center for Genome Research (www-genome.wi.mit.edu). Sequence data for C.albicans was obtained from the Stanford Genome Technology Center website at http://www-sequence.stanford. edu/group/candida. Sequencing of C.albicans was accomplished with the support of the NIDR and the Burroughs Wellcome Fund. Sequence data for C.neoformans was obtained from C.neoformans Genome Project, Stanford Genome Technology Center, funded by the NIAID/NIH under cooperative agreement U01 AI47087, and The Institute for Genomic Research, funded by the NIAID/NIH under cooperative agreement U01 AI48594. Preliminary sequence data for A.fumigatus was obtained from The Institute for Genomic Research website at http://www.tigr.org. Sequencing of A.fumigatus was funded by the National Institute of Allergy and Infectious Disease U01 AI 48830 to David Denning and William Nierman.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Grummt,I. (1999) Regulation of mammalian ribosomal gene transcription by RNA polymerase I. Prog. Nucleic Acid Res. Mol. Biol., 62, 109–154.[ISI][Medline]

  2. Reeder,R.H. (1999) Regulation of RNA polymerase I transcription in yeast and vertebrates. Prog. Nucleic Acid Res. Mol. Biol., 62, 293–327.[ISI][Medline]

  3. Paule,M.R. and White,R.J. (2000) Transcription by RNA polymerases I and III. Nucleic Acids Res., 28, 1283–1298.[Abstract/Free Full Text]

  4. Bell,S.P., Pikaard,C.S., Reeder,R.H. and Tjian,R. (1989) Molecular mechanisms governing species-specific transcription of ribosomal RNA. Cell, 59, 489–497.[ISI][Medline]

  5. Pikaard,C.S., McStay,B., Schultz,M.C., Bell,S.P. and Reeder,R.H. (1989) The Xenopus ribosomal gene enhancers bind an essential polymerase I transcription factor, xUBF. Genes Dev., 3, 1779–1788.[Abstract/Free Full Text]

  6. Pikaard,C.S., Pape,L.K., Henderson,S.L., Ryan,K., Paalman,M.H., Lopata,M.A., Reeder,R.H. and Sollner-Webb,B. (1990) Enhancers for RNA polymerase I in mouse ribosomal DNA. Mol. Cell. Biol., 10, 4816–4825.[Abstract/Free Full Text]

  7. Copenhaver,G.P., Putnam,C.D., Denton,M.L. and Pikaard,C.S. (1994) The RNA polymerase I transcription factor UBF is a sequence-tolerant HMG-box protein that can recognize structured nucleic acids. Nucleic Acids Res., 22, 2651–2657.[Abstract/Free Full Text]

  8. McStay,B., Frazier,M.W. and Reeder,R.H. (1991) xUBF contains a novel dimerization domain essential for RNA polymerase I transcription. Genes Dev., 5, 1957–1968.[Abstract/Free Full Text]

  9. O’Mahony,D.J., Xie,W.Q., Smith,S.D., Singer,H.A. and Rothblum,L.I. (1992) Differential phosphorylation and localization of the transcription factor UBF in vivo in response to serum deprivation. In vitro dephosphorylation of UBF reduces its transactivation properties. J. Biol. Chem., 267, 35–38.[Abstract/Free Full Text]

  10. Stefanovsky,V.Y., Pelletier,G., Hannan,R., Gagnon-Kugler,T., Rothblum,L.I. and Moss,T. (2001) An immediate response of ribosomal transcription to growth factor stimulation in mammals is mediated by ERK phosphorylation of UBF. Mol. Cell, 8, 1063–1073.[ISI][Medline]

  11. Pelletier,G., Stefanovsky,V.Y., Faubladier,M., Hirschler-Laszkiewicz,I.I., Savard,J., Rothblum,L.I., Cote,J. and Moss,T. (2000) Competitive recruitment of CBP and Rb-HDAC regulates UBF acetylation and ribosomal transcription. Mol. Cell, 6, 1059–1066.[ISI][Medline]

  12. Hirschler-Laszkiewicz,I., Cavanaugh,A., Hu,Q., Catania,J., Avantaggiati,M.L. and Rothblum,L.I. (2001) The role of acetylation in rDNA transcription. Nucleic Acids Res., 29, 4114–4124.[Abstract/Free Full Text]

  13. Hannan,K.M., Hannan,R.D., Smith,S.D., Jefferson,L.S., Lun,M. and Rothblum,L.I. (2000) Rb and p130 regulate RNA polymerase I transcription: Rb disrupts the interaction between UBF and SL-1. Oncogene, 19, 4988–4999.[ISI][Medline]

  14. Keener,J., Dodd,J.A., Lalo,D. and Nomura,M. (1997) Histones H3 and H4 are components of upstream activation factor required for the high-level transcription of yeast rDNA by RNA polymerase I. Proc. Natl Acad. Sci. USA, 94, 13458–13462.[Abstract/Free Full Text]

  15. Keys,D.A., Lee,B.S., Dodd,J.A., Nguyen,T.T., Vu,L., Fantino,E., Burson,L.M., Nogi,Y. and Nomura,M. (1996) Multiprotein transcription factor UAF interacts with the upstream element of the yeast RNA polymerase I promoter and forms a stable preinitiation complex. Genes Dev., 10, 887–903.[Abstract/Free Full Text]

  16. Siddiqi,I.N., Dodd,J.A., Vu,L., Eliason,K., Oakes,M.L., Keener,J., Moore,R., Young,M.K. and Nomura,M. (2001) Transcription of chromosomal rRNA genes by both RNA polymerase I and II in yeast uaf30 mutants lacking the 30 kDa subunit of transcription factor UAF. EMBO J., 20, 4512–4521.[ISI][Medline]

  17. Vu,L., Siddiqi,I., Lee,B.S., Josaitis,C.A. and Nomura,M. (1999) RNA polymerase switch in transcription of yeast rDNA: role of transcription factor UAF (upstream activation factor) in silencing rDNA transcription by RNA polymerase II. Proc. Natl Acad. Sci. USA, 96, 4390–4395.[Abstract/Free Full Text]

  18. Guo,A., Chen,L., Zhao,A., Boukghalter,B. and Pape,L. (2000) Fission yeast contains an rDNA binding activity that interacts specifically with regulatory sequences for ribosomal RNA synthesis. Gene, 242, 183–192.[ISI][Medline]

  19. Souciet,J., Aigle,M., Artiguenave,F., Blandin,G., Bolotin-Fukuhara,M., Bon,E., Brottier,P., Casaregola,S., de Montigny,J., Dujon,B. et al. (2000) Genomic exploration of the hemiascomycetous yeasts: 1. A set of yeast species for molecular evolution studies. FEBS Lett., 487, 3–12.[ISI][Medline]

  20. Altschul,S.F., Madden,T.L., Schaffer,A.A., Zhang,J., Zhang,Z., Miller,W. and Lipman,D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res., 25, 3389–3402.[Abstract/Free Full Text]

  21. Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Aci