Nucleic Acids Research

Architecture of promoter regions ofribosomal protein genes

Molecular dissection of ribosomal protein (rp) gene promoter regions of Schizosaccharomyces pombe revealed a promoter type which does not contain a canonical TATA-box. Instead, these promoters display a TATA-analogue named the Homol D-box. We showed that the Homol D-box, represented by the sequence CAGTCACA or its reverse complementary sequence TGTGACTG, is involved in determining transcriptional start sites and is the target of protein factor(s) binding. The binding of this factor cannot be competed with TATA-box containing oligonucleotides (1). The Homol D-box has been compared with the TATA-box with respect to its potential to form local sequence-specific structures which may contribute to the binding specificity of trans-acting proteins (2; Table 1).

Table 1. Ribosomal protein genes of S.pombe and their proximal promoter Lit., rp genes isolated from S.pombe and their previous designations; NC rat, nomenclature for rat ribosomal proteins (8); NC S.c., new nomenclature for ribosomal proteins of the budding yeast S.cerevisiae (7); Homol E proximal activation sequence in promoters of rp genes, in bold: canonical sequence; D, distance in bp between Homol E and Homol D; Homol D is the TATA-box analogue, in bold: canonical sequence; ATG, distance in bp between Homol D-box and start codon ATG of the indicated rp gene; cosno, cosmid number under which the sequences are to be found in the S.pombe genome project (http://www.sanger.ac.uk/Projects/S_pombe/); the cosmid numbers can also be used to access the sequences in the databases DDBJ/EMBL/GenBank; *database accession numbers.

In some rp promoters we found the tandem repeat AGGGTAGGGT or its reverse complementary sequence ACCCTACCCT upstream of the Homol D-box. We named this sequence the Homol E-box and showed that it functions, in the proximal arrangement, with Homol D as an activation sequence. It is also the target of protein factors (3; Table 1).

The Sanger Centre (Cambridge, UK) together with 12 other European laboratories, is sequencing the S.pombe genome (http://www.sanger.ac.uk/Projects/S_pombe/). Finished and unfinished sequences are available and searchable. We searched these sequences, represented by cosmids of chromosomes I and II, and found 43 genes coding for ribosomal proteins, including several we and others had already sequenced (Table 1).

Strikingly, a comparison of the promoter region of the rp genes reveals a statistically significant pattern for Homol D- and Homol E-boxes. First, all promoters contain the Homol D-box in a position between 49 and 102 bp upstream of the start codon ATG (Table 1). Second, those promoters which unequivocally contain, in addition to Homol D, the Homol E-box, display Homol E as a proximal upstream element of Homol D. The distance between the two elements ranges from 0 to 14 bp with no obvious bias for a preferred distance (Table 1). It is also noticeable from Table 1 that both sequences, Homol D and Homol E, are highly conserved. Only one sequence of Homol D and one of Homol E show two changes in the canonical Homol-box (rpaP1-1, rp126, Table 1). Fewer than 8% show a change of one nucleotide. For Homol D we found CAGTCACG instead of CAGTCACA, and AGTGACTG instead of TGTGACTG. The tandem repeat AGGGTAGGGT of Homol E was found as AGGGTAGGGC or as AGGTTAGGGT, whereas the reverse complementary repeat ACCCTACCCT was always found as GCCCTACCCT or ACCCTAACCT. In the Homol E sequence, two nucleotides were changed as indicated in Table 1 (rp126; Table 1).

We reported previously that different ribosomal proteins in S.pombe are encoded by either one or two expressed genes. In one case we found a family of three genes (4). In both gene families, one promoter contained Homol E and Homol D and the other only Homol D (3). Whether this will continue to hold cannot be predicted. We found in the database two families with two genes (rp118a and rp132) containing Homol D only (Table 1). However, these might be families like rp18 consisting of three members.

We compared the promoter strength of two gene families (rp127a and rp18; Table 1) using lacZ as a reporter. One family consisted of three genes with two promoters displaying only Homol D and one containing the Homol E/Homol D arrangement; the other family had a promoter with a Homol D-box and a promoter with the Homol E/Homol D arrangement. These two families express equal amounts of [beta]-Gal activity (3). However, these data do not reveal whether Homol E plays a role in the regulation of rp genes. The data show that Homol E is a proximal transcription activation sequence which can activate through the Homol D-box (3). Moving Homol E more than 21 bp away from Homol D leads to a drastic decrease in [beta]-Gal activity, indicating that it is a proximal promoter element (I.Witt, personal communication).

The compilation of the rp gene promoters (Table 1) confirms our data and predictions published previously (1,3). We suggested that rp genes of fission yeast display a promoter type in which Homol D mediates basal transcription and Homol E is a proximal activation sequence. Here we propose that Homol E and Homol D, their relative position to each other and the position of Homol D relative to the start codon are used as signatures for this promoter type.

Introns in rp genes

The architecture of the promoter of rps15a-1 appears to be an exception to the rule since the distance between the Homol D-box and the start codon ATG is 213 bp (Table 1). Interestingly, comparative analysis of this promoter region reveals 5[prime] and 3[prime] splice sites including a branch site establishing possible S.pombe introns of 147 or 87 bp, respectively, in this untranslated region (5). We found introns in 40% of the 51 rp genes. A random count of 200 ORFs in the S.pombe database counting intron-less versus intron containing ORFs revealed 45% intron containing genes. This is consistent with the results of a survey which we conducted a few years ago. We searched a random sample of 100 genes for introns and found that 40% of them contained introns (5). Thus rp genes contain introns with a frequency found for all the other genes. The distribution of the introns in rp genes seems to be random. The smallest exon1 found in an ORF consists of just the start codon ATG, found in rps17 and rp17a-1 (A.Marchfelder, personal communication). However, there seems to be a bias in the number of introns within one gene. Eighty-five percent of the intron containing rp genes display only one intron. In other genes both the number and distribution of introns appears to be random. There are genes with any number of introns, whereas 15 introns is currently the highest number reported (6). Finally, at this stage of the rp gene count another bias seems to exist: the size of several introns found is between 215 and 278 bp. These introns belong to the larger class of S.pombe introns and are less often found than introns between 40 and 100 bp in size. The architecture of S.pombe introns has been discussed by Prabhala et al. (5).

Suggestion for nomenclature of S.pombe rp genes and proteins

The sequence of the genome of the budding yeast S.cerevisiae has been completed and is available for comparative analysis. Based on such a comparative analysis a new nomenclature for ribosomal proteins of S.cerevisiae has been proposed (7). The new numbering system of ribosomal proteins of S.cerevisiae follows, when possible, the nomenclature for mammalian ribosomal proteins. Significant differences still exist. One major difference is that there is no S.cerevisiae counterpart for mammalian ribosomal protein L28 (7,8).

We compared the amino acid sequences of the S.pombe ORFs identified in the Sanger Centre (Table 1) with the available databanks using BLAST 2.0, a similarity search tool developed by Altschul et al. (9). With this approach we found a gene on cosmid C2E11 (Table 1) which aligns with sequences from rat, Caenorhabditis elegans and Trypanosoma cruzi. Notably, the sequence of the rat (Fig. 1A, rnL28) shares 37% identical amino acids with the S.pombe sequence (Fig. 1A, spL28), and 36 and 31% with C.elegans and Trypanosoma, respectively. The rat sequence is described in the database as ribosomal protein L28 (Fig. 1A, rnL28). The ribosomal proteins of rat are the base for the proposed mammalian numbering system since the sequence determination of all rat ribosomal proteins has been almost completed (8; Table 1). We did not find an ORF in the genomic sequence of S.cerevisiae bearing significant similarity to these sequences.

Figure 1. Alignment of primary sequences of ribosomal proteins of S.pombe (sp), S.cerevisiae (sc) and Rattus norvegicus (rn). (A) Ribosomal protein rnL28 shows 37% identical amino acids with the ORF found in the S.pombe database on cosmid C2E11 (Table 1). No similar sequence was found in the S.cerevisiae genome. (B) Alignment of ribosomal protein S23 of S.cerevisiae and rat with the ORF found on cosmid C23C11 (Table 1). (C) Alignment of ribosomal protein L25 of S.cerevisiae and ribosomal protein L23a of rat with the ORF found on cosmid C19G10 (Table 1). The positions of conserved residues are highlighted by black (identical) and grey (similar).

All the other ORFS of S.pombe resemble ribosomal protein sequences of S.cerevisiae and rat, whereas in general the sequences share between 60 and 85% identical amino acids. For example, the ORF on cosmid C23C11 (Table 1) aligns with ribosomal protein S23 of S.cerevisiae (scS23) and rat (rnS23), sharing 85 and 75% identical amino acids, respectively (Fig. 1B). According to the new S.cerevisiae nomenclature, this rp has recently been designated S23 (7). Furthermore, the ORF on cosmid C19G10 (Table 1) aligns with ribosomal protein L25 of S.cerevisiae (scL25) and L23a of rat (rnL23a), sharing 64 and 61% identifical amino acids, respectively (Fig. 1C). L25 is the designation of the S.cerevisiae rp following the new nomenclature (7, Table 1). These two examples reflect the range of similarity found between the species for all the sequences which we have tested, except for L28 (Fig. 1).

We also found rp genes in the S.pombe databank encoding the counterparts of acidic ribosomal proteins (10). Based on sequences taken from the literature (10), we identified two promoters of acidic rp genes as containing canonical TATA-boxes (1). One of these genes (rpa3), was found in our search of the S.pombe database for promoters containing Homol E-/Homol D-boxes (Table 1). The two promoter regions are completely different in sequence, whereas the sequences spanning the structural gene, the intron sequences and the 3[prime] trailer of the gene are identical. This indicates that we identified the authentic rpa3 gene. The TATA-box promoter in front of rpa3 must be a cloning accident (10).

Based on the data presented here, we suggest the S.pombe genome project and the scientific community use the mammalian nomenclature for the cytoplasmic ribosomal proteins (7,8). We also suggest indicating the fission yeast ribosomal protein genes as rps3-1, rps3-2, rp17a-1, rp130, etc. If there are two copies, they should be labeled -1, -2, etc. (Table 1). The proteins coded by the genes should be called Rps3.1, Rps3.2, Rp17a.1, Rp130, etc. (Table 1). The gene names rps and rp1 for ribosomal proteins of the small and large subunit, respectively, have been proposed by Kohli (11). The acidic ribosomal protein genes isolated by Beltrame and Bianchi (10) have been named rpa1, rpa2, rpa3 and rpa4. In the new S.cerevisiae numbering system and in the mammalian system (7), the acidic equivalents of Rpa1 and Rpa3 are designated P1; the equivalents of Rpa2 and Rpa4 are named P2. In addition, we found in the EMBL databank two cDNAs named rpa5 and rpa6 (AJ002733 and AJ002734). Rpa5 belongs to the P1 and Rpa6 belongs to the P2 family of acidic ribosomal proteins. Therefore, we suggest the designation RpaP1.1, RpaP1.3 and RpaP1.5, and RpaP2.2, RpaP2.4 and RpaP2.6 for these two families (Table 1).

Since many ribosomal protein genes are represented by two copies and 35 different ribosomal proteins have been identified in the S.pombe genome project, we would expect that at least another 50 ribosomal protein genes will be sequenced and named. If the mammalian nomenclature is used and the new nomenclature of the S.cerevisiae counterparts is always added in brackets (7), we are confident that the currently unbearable confusion in nomenclature of ribosomal proteins will be under control.

ACKNOWLEDGEMENT

We thank Drs Isabell Witt and Anita Marchfelder for communicating unpublished results.

REFERENCES

1. Witt,I., Straub,N., Käufer,N.F. and Gross,T. (1993) EMBO J., 12, 1201-1208. MEDLINE Abstract

2. Harrington,R.E. and Winicov,J. (1994) Prog. Nucleic Acid Res. Mol. Biol., 47, 195-270. MEDLINE Abstract

3. Witt,I., Kwart,M., Gross,T. and Käufer,N.F. (1995) Nucleic Acids Res., 23, 4296-4302. MEDLINE Abstract

4. Gatermann,K.B., Teletsky,C., Gross,T. and Käufer,N.F. (1989) Curr. Genet., 16, 361-367. MEDLINE Abstract

5. Prabhala,G., Rosenberg,G.H. and Käufer,N.F. (1992) Yeast, 8, 171-182. MEDLINE Abstract

6. Nakamura,T.M., Morin,G.B., Chapman,K.B., Weinrich,S.L. Andrews,W.H., Lingner,J., Harley,C.B. and Cech,T.R. (1997) Science, 277, 955-959. MEDLINE Abstract

7. Mager,W.H., Planta,R.J., Ballesta,J.-P.G., Lee,J.C., Mizuta,K., Suzuki,K., Warner,J.R. and Woolford,J. (1997) Nucleic Acids Res., 25, 4872-4875. MEDLINE Abstract

8. Wool,I.G., Chan,Y.L. and Gluck,A. (1995) Biochem. Cell Biol., 73, 933-947. MEDLINE Abstract

9. Altschul,S.F., Madden,T.L., Schaffer,A.A., Zhang,J., Zhang,Z., Miller,W. and Lipman,D.J. (1997) Nucleic Acids Res., 25, 3389-3402. MEDLINE Abstract

10. Beltrame,M. and Bianchi,E.M. (1990) Mol. Cell. Biol., 10, 2341-2348. MEDLINE Abstract

11. Kohli,J. (1987) Curr. Genet., 11, 575-589. MEDLINE Abstract

---

*To whom correspondence should be addressed. Tel: +49 531 391 5774; Fax: +49 531 391 5765; Email: n.kaeufer@tu-bs.de

---

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 2 Jul 1998
Copyright©Oxford University Press, 1998.
CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				B. Shor, J. Calaycay, J. Rushbrook, and M. McLeod Cpc2/RACK1 Is a Ribosome-associated Protein That Promotes Efficient Translation in Schizosaccharomyces pombe J. Biol. Chem., December 5, 2003; 278(49): 49119 - 49128. [Abstract] [Full Text] [PDF]

				O. Lecompte, R. Ripp, J.-C. Thierry, D. Moras, and O. Poch Comparative analysis of ribosomal proteins in complete genomes: an example of reductive evolution at the domain scale Nucleic Acids Res., December 15, 2002; 30(24): 5382 - 5390. [Abstract] [Full Text] [PDF]

This Article Abstract Print PDF (98K) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (8) Request Permissions Commercial Re-use Guidelines for Open Access NAR Content Google Scholar Articles by Gross, T. Articles by Kaufer, N. F. Search for Related Content PubMed PubMed Citation Articles by Gross, T. Articles by Kaufer, N. F. Social Bookmarking          What's this?