The ribosomal spacer in Xenopus laevis is transcribed as part of the primary ribosomal RNA

Volume 14 Number 15 1986 Nucleic Acids Research The ribosomal spacer in Xenopus laevis is transcribed as part of the primary ribosomal RNA Ronald F.J.De Winter and Tom Moss*"1" Biophysics Laboratories, Portsmouth Polytechnic, St Michael's Building, White Swan Road, Portsmouth PO1 2DT, UK Received 15 May 1986; Revised 14 July 1986; Accepted 21 July 1986 SI mapping of Xenopus laevis ribosomal RNA transcripts, both in oocyte microinjection experiments and in vivo, shows that all but 212 bp of the so-called "non-transcribed" spacer (NTS) of the ribosomal DNA repeat is transcribed as part of the primary ribosomal transcript. The 40S pre-ribosomal RNA (pre-rRNA) is therefore a processing intermediate. The primary ribosomal transcript co-terminates with the previously described spacer transcripts [Moss (10)], at a site 213 bp upstream of the 40S pre-rRNA initiation site. This mode of transcription suggests a simple mechanism for the recently proposed phenomenon of "readthroughenhancement", [Moss et al (9), Moss (10)], where readthrough transcription from an upstream gene may enhance transcription of a gene immediately downstream in the tandem ribosomal repeat. INTRODUCTION The gene coding for the 18S, 5.8S and 28S ribosomal RNAs (rRNAs) in eukaryotes are transcribed as part of a single oligocistronic transcript, (see refs 1, 2 for early reviews). For Xenopus laevis and other amphibia the "Miller" spread (3-5) elegantly supported earlier biochemical arguments that the tandemly repeated transcribed regions were separated from eath other by "non-transcribed" spacers (NTS). In agreement with these data the primary ribosomal transcript in X. laevis was identified as a 40S pre-rRNA molecule which terminated at the end of the 28S coding sequence (6,7). Although it was recently discovered that pre-rRNA transcription continued a little downstream of the 28S gene (8), this did not change the general view that a large untranscribed stretch of DNA separates consecutive transcription units. Even the earlier discovery of active RNA polymerase I promoters in the ribosomal DNA, (rDNA), NTS of X. laevis, Drosophila melanogaster andD. virilis, (see ref. 9 for review) did not affect this view. It has been shown that the X. laevis ribosomal (rDNA) spacer sequences lying between the first spacer promoter or Bam Island I, (SpPr 1 in fig. 1) and the 40S promoter proximal termination site at -213 bp, (T in fig. © IRL Press Limited, Oxford, England. 6041 ABSTRACT Nucleic Acids Research 1) are transcribed (10). Thus If the 40S pre-rRNA transcript is terminated, this must occur either upstream of the first spacer promoter or at the promoter proximal -213 bp site. We have found that in oocyte microinjection experiments >40% of the transcripts initiated from the 40S pre-rRNA promoter on pX£lO8c (and derivative clones), run through the vector region of these circular (plasmld) constructs and back into the rDNA insert, (Mitchelson and Moss, in preparation). These transcripts are designated pre-rRNA in fig. 1. Hence transcription on each of these rDNA plasmids resembles that on the tandemly repeated rDNA of the X. laevis chromosome. It has also been shown that under the conditions used for microinjection, no detectable level of This situation was therefore used to study the effects of deleting putative termination sequences on the transcription of downstream NTS sequences. The results show that all but 212 bp of the ribosomal NTS in X. laevis is transcribed as part of the primary ribosomal transcript. Mapping of RNA from X. laevis tissue culture cells and oocytes shows that this mode of transcription also occurs in vivo. MATERIALS AND METHODS Enzymes Restriction enzymes were obtained from Boehringer Mannheim, DNA T4 ligase from New England Biolabs, T4 DNA polymerase, T4 polynucleotide kinase and SI nuclease from P-L Biochemicals. Enzymes were generally used as recommended by the suppliers. Construction recombInant plasmids pX£ANS9c, pX£ANS10c and pX^ANSllc were all constructed by restricting pXdlO8c with BamHl totally, (for pX£ANS9c), or partially, (for pX£ANS10c and pXdANSllc), followed by religation and transformation into E. coli HB101. pX£ANS13f was constructed by totally restricting pX£lO8f with Hindlll and BamHl. pX£lO8f is a vector deletion mutant of pX£lO8c lacking the pBR322 Hindlll-PvuII fragment and contains only the two BamHl sites in the rDNA insert (11). The two Hindlll-BamHl fragments of pX£lO8f were separated on low melting agarose. The smaller fragment was totally digested with Hinfl, the Hindlll-Hinfl, Hinfl-BamHl and larger Hinfl-Hinfl fragments isolated, ligated into the larger HindiII-BamHl fragment from pX£lO8f and transformed into E. coli HB101. As the Hindlll-Hinfl and Hinfl-BamHl fragments could not self ligate due to the different sequences of the Hinfl sites, only the HinflHinfl fragment was included in the recombinant plasmid. 6042 Transformants were transcription is initiated at any site(s) other than the already identified polymerase I promoters (op. cit.). Nucleic Acids Research analysed by restriction enzyme analysis. The resultant recombinants are shown in figure 1. Probes Probes specific for 40S pre-rRNA transcripts and spacer promoter initiated transcripts have been described previously (10,12). In brief, the 40S pre-rRNA specific probe is the Pstl-TaqI fragment crossing the 40S initiation site. This probe was also used for detecting the 3' terminus of spacer transcripts by labelling the Pstl end as described previously (10), Fig. 1. The probe specific for spacer promoter initiated transcripts is the BamHl-Avall fragment from the first spacer promoter (12), (see also Fig. 1). from pBR322 as an extension. A second spacer transcript probe was constructed by cloning the Avail-Avail fragment from the first spacer promoter linked to the larger BamHl-PvuII fragment of pBR322. Fragment and vector were filled out with T4 DNA polymerase before ligation. In this way, at the PvuII-Avall junction the Avail site was restored, whereas at the BamHlAvall junction the Avail site was destroyed. By choosing the appropriate orientation of the Avail fragment, it was then possible to recover a Ddel to Avail fragment containing the original Avail-Avail rDNA fragment but extended by vector sequences. When 5' end labelled, this fragment could distinguish transcripts reading through the spacer promoter I from those reading through more downstream spacer promoters, see fig. 4. DNA 5' and 3' end labelling 32 P have previously been described (10,11,12). with Microinjection and transcript detection Equimolar amounts of plasmid were microinjected in X. borealis oocytes in the presence of ct-amanitin (BCL) and transcripts were detected by Sl-nuclease mapping as described previously (10,11,12). Endogenous tissue culture and oocyte transcripts were also detected by SI mapping as previously described (10,11,12). RNAase sensitivity of t (...truncated)