M. OHNO

Because the nuclear export of mRNA occurs only after the splicing reaction is completed, intron-containing pre-mRNA does not normally appear in the cytoplasm. As a mechanism to secure this, intron-containing RNA is retained in the nucleus via formation of the spliceosome. Therefore, the process of releasing spliced mRNA from the spliceosome after completion of splicing is an essential step for triggering the nuclear export of the spliced mRNA. In budding yeast, RNA helicase-like protein Prp22 is implicated in this process. Here we demonstrate the function of HRH1, a human protein homologous to Prp22, in mammalian cells using dominant-negative HRH1++ mutants (dn-HRH1). dn-HRH1 protein stalls on the spliceosome and prevents release of the spliced RNA from the spliceosome in vitro. Expression of dn-HRH1 in mammalian cells leads to inhibition of splicing and to extensive nuclear export of unspliced pre-mRNA, probably because of the incapability of recycling spliceosome components that normally retain the pre-mRNA in the nucleus. The arginine/serine-rich domain (RS domain) of HRH1, which is missing in Prp22, confers a nuclear localization signal, and appears to facilitate the interaction of HRH1 with the spliceosome. This is the first report on a bona fide mammalian homolog of yeast Prp splicing factor, and also on a mammalian RNA helicase-like splicing factor.

The effect of the 5' cap structure on the splicing of precursor mRNAs was investigated after the RNAs were injected into Xenopus oocyte nuclei. The precursor mRNAs synthesized in vitro in a prokaryotic transcription system with a dinucleotide, ApppG, as a primer, were extremely stable when injected into the nuclei yet behaved like uncapped pre-mRNAs in the in vitro splicing reaction. The ApppG-primed precursor mRNAs served as a control (uncapped) in the injection experiments, and their splicing reactions were compared with those of their capped (m7GpppG-primed) counterparts. The capped precursors were spliced more efficiently than the uncapped precursors. Examination of splicing of the precursor mRNA that contained three exons and two introns with a single molecule has revealed that the cap structure exerts its effect primarily on the 5'-proximal intron. Thus, the cap structure not only stabilizes precursor mRNAs but also plays a positive role in the splicing of precursor mRNAs in cells.

In the budding yeast Saccharomyces cerevisiae, a number of PRP genes known to be involved in pre-mRNA processing have been genetically identified and cloned. Three PRP genes (PRP2, PRP16, and PRP22) were shown to encode putative RNA helicases of the family of proteins with DEAH boxes. However, any such splicing factor containing the helicase motifs in vertebrates has not been identified. To identify human homologs of this family, we designed PCR primers corresponding to the highly conserved region of the DEAH box protein family and successfully amplified five cDNA fragments, using HeLa poly(A)+ RNA as a substrate. One fragment, designated HRH1 (human RNA helicase 1), is highly homologous to Prp22, which was previously shown to be involved in the release of spliced mRNAs from the spliceosomes. Expression of HRH1 in a S. cerevisiae prp22 mutant can partially rescue its temperature-sensitive phenotype. These results strongly suggest that HRH1 is a functional human homolog of the yeast Prp22 protein. Interestingly, HRH1 but not Prp22 contains an arginine- and serine-rich domain (RS domain) which is characteristic of some splicing factors, such as members of the SR protein family. We could show that HRH1 can interact in vitro and in the yeast two-hybrid system with members of the SR protein family through its RS domain. We speculate that HRH1 might be targeted to the spliceosome through this interaction.

To apply accurate and uniform osmotic pressures to liposomes, they can be formed using the spontaneous transfer method in solutions with different osmolarities. The majority of liposomes unexpectedly opened large holes (several micrometers in diameter) in response to the osmotic pressure regardless of its strength, that is, the difference between the outside and inside solute (sucrose or KCl) concentrations. However, the lag time for any response, including the opening of a hole, after the formation of the liposome decreased with increasing osmotic pressure.