Expression of His Cys Box Homing Endonucleases from Nuclear rDNA Transcripts

The position of the ORFs of the His-Cys box homing endonucleases in nuclear rDNA genes poses several questions regarding the expression of these proteins. In the nuclei of eukaryotes, RNAs destined to be translated into protein are transcribed by RNA polymerase II (Pol II) and subjected to extensive processing (5' capping, splicing and polyadenylation). Transcription and processing appear to exhibit cross talk that can reciprocally increase the efficiency of each event (Maniatis and Reed 2002; Proudfoot et al. 2002; Le Hir et al. 2003). Furthermore, RNA processing events greatly affect the stability, export, localization, and translation of the messenger RNA. Conversely, rDNA is transcribed by RNA Pol I in the nucleolus, and rRNA transcripts undergo a very different processing (cleavage and modification) to ready them for assembly into the ribosomal subunits.

For the His-Cys box homing endonucleases found in these rRNA transcripts questions of which polymerase generates the translated message and what pre-mRNA processing is necessary for expression have been raised. This situation is in contrast to the other families of homing endonucleases that so far have been found exclusively in organelle genomes. In organelles, transcription and translation take place in the same compartment, and pre-mR-NA processing appears to be minimal. Also, it is believed that a single RNA polymerase is responsible for all organelle transcription. In addition to transcription and processing of the His-Cys box endonuclease RNAs, there is also the matter of which portion of the transcript is exported to the cytoplasm and eventually translated (Fig. 1). It has been shown that group I introns are primarily spliced co-transcriptionally from nascent rRNA transcripts (Brehm

Fig. 1. Schematic of transcription and processing of RNA containing the His-Cys box homing endonuclease ORF. IPS refers to the internal processing site of nucleolytic cleavage by the (sometimes second) ribozyme

et al. 1983). In addition to the splicing reaction, the ribozyme can alternately catalyze the circularization of the full intron (leaving the flanking exons unjoined). Although this reaction is not beneficial to the host and appears at a low frequency in vivo, it may generate an RNA species resistant to degradation that plays a role in either endonuclease expression or intron mobility (Zaug et al. 1983; Nielsen et al. 2003). Another ribozyme-mediated processing event described for group I introns is hydrolysis at internal sites, resulting in a separation of the ribozyme from the ORF sequence. Internal processing appears to be carried out by the same ribozyme responsible for self-splicing in most cases. However, the Didymium iridis S956 intron encoding the I-Dirl endonuclease and the Naegleria introns inserted at S516 both contain "twin" ribozymes: one that catalyzes intron excision and exon ligation and one that catalyzes hydrolysis at two internal processing sites to release itself along with the endonuclease ORF (Johansen and Vogt 1994; Decatur et al. 1995; Einvik et al. 1997).

Studies of I-Ppol expression from the PpLSU3 intron integrated into the S. cerevisiae genomic rDNA showed that endonuclease activity could not be detected when RNA Pol I was inactivated (Lin and Vogt 1998). This result indicates that I-Ppol is expressed from an RNA Pol I transcript. Additionally, expression was not dependent on internal processing of the intron. In fact, higher enzyme activity was observed when processing sites were mutated. This result and data showing that the full-length intron was found in the cytoplasm was taken as evidence that I-Ppol is translated from the full-length excised intron. Further studies in this system showed that a heterologous ORF (ß-ga-lactosidase) replacing the I-Ppol ORF could also be expressed from the RNA

Pol I transcript, although at lower efficiency (-3%) than a similar RNA Pol II transcript (Lin and Vogt 2000). The lower expression does not appear to be due to transcript levels, but is likely due to lower efficiency of nuclear export and translation, which are known to be stimulated by the 5' cap and poly-A tail found on RNA Pol II transcripts and are presumably lacking from the excised intron.

It appears that the stories for other His-Cys box endonuclease expression might be more complicated. For example, activity of the second ribozyme responsible for internal processing in the twin introns of Didymium and Naegleria are necessary for expression of their homing endonucleases (Vad-er et al. 1999; Decatur et al. 2000). Intriguingly, the Didymium I-Dirl ORF also contains what appear to be a small spliceosomal intron and a polyadenyla-tion site, hallmarks of RNA Pol II transcripts (Vader et al. 1999). When the Didymium transcripts were examined, the majority species in the cytoplasm consisted of the ORF released by internal processing with the spliceosomal intron sequence absent and the 3' end polyadenylated. This species also specifically associated with polysomes, indicating that it is the messenger RNA. Although production of this RNA from a cryptic RNA Pol II promoter has not been rigorously disproven, all current data point to this being a unique example of these selfish sequences somehow incorporating host organism pre-mRNA processing in a RNA Pol I transcript to improve expression.

0 0

Post a comment