Worm Breeder's Gazette 12(1): 43 (September 1, 1991)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Why do some pre-mRNAs receive exclusively SL1 ,while others receive only SL2 ?It now seems unlikely that some sequence on the pre-mRNA is the signal, because comparison of the outrons (the name for the intron at the beginning of trans-spliced pre-mRNAs) of SL2 -acceptingpre-mRNAs revealed no sequences in common. At the last worm meeting we proposed that SL2 trans-splicing is specific for downstream genes in "polycistronic mRNAs." However, the presence of an AAUAAA signal for cleavage and polyadenylation in the intercistronic region would make it unlikely that a full length polycistronic mRNA would ever exist. Instead it seems probable that, if two genes are co-transcribed, the transcript would be cleaved and the upstream portion poly-adenylated before the downstream gene is transcribed. The downstream portion might become an uncapped RNA and this uncapped pre-mRNA could be the signal for SL2 trans-splicing. The hypothesis implies that the intercistronic region should contain neither a transcription terminator for the upstream gene nor a promoter for the downstream gene. This hypothesis was engendered by the observations that in two examples of SL2 splicing, the SL2 -acceptinggene is just downstream of another gene oriented in the same direction, with AAUAAA signals between the genes. For instance, the SL2 -acceptinggene, gpd-3 ,is 245 bp downstream of the SL1 -acceptinggene, gpd-2 ,and the two genes are temporally and spatially co-expressed.
To begin to test this idea, we have inserted the entire intercistronic region of gpd-2 / gpd-3 into the 5' untranslated region of a vit-2 /6fusion gene, between the vit-2 promoter and the coding portion of the fusion gene. This gene should be transcribed under vit control and direct SL2 -specifictrans-splicing. Adult transgenic worms carrying this construction (called pIC+) in rol-6 co-arrays express the fusion protein, so there is no functional transcription termination signal in the gpd-2 / gpd-3 intercistronic space. Furthermore, larvae do not express the transgene, suggesting that the intercistronic region lacks gpd-3 promoter activity. These preliminary findings are consistent with the idea that gpd-2 and gpd-3 could be co-transcribed.
The transcripts were analyzed by cDNA-PCR amplification of the fusion mRNAs using oligos complementary to vit-2 , SL1 ,and SL2 .Recall that when a vit-5 intron lacking its 5' splice site was inserted in this same position it was trans-spliced to SL1 ,suggesting that SL1 splicing is the default mode. We observed a different result with the pIC+ construction: some transcripts received SL1 while others received SL2 .All trans-splicing occurred at the normal gpd-3 trans-splice site. The SL1 and SL2 oligos do not cross-hybridize under the amplification conditions used, and so we feel confident that both SL1 and SL2 were trans-spliced to the fusion pre-mRNAs. All products detected by primer extension were the size expected for the trans-spliced product, indicating that processing of the pIC+ transcripts was efficient. One interpretation of these observations is that the fusion mRNA transcripts undergo alternative fates, depending on whether they have been cleaved. Perhaps the cleavage reaction is relatively inefficient because the cleavage site is so near the 5' end of the transcript. In the absence of cleavage, SL1 trans-splicing may occur, while following cleavage, SL2 trans-splicing may occur. We are currently eliminating the AAUAAA to see if SL1 trans-splicing becomes the exclusive fate of the pIC+ transcript.