Messenger RNAs transcribed from yeast linear cytoplasmic plasmids possess unconventional 5’ and 3’ UTRs and suggest a novel mechanism of translation

Václav Vopálenský; Michal Sýkora; Tomáš Mašek; Martin Pospíšek

The unique 5' ends of the pGKL mRNAs immediately raised a question about the organization of their 3' ends. Previously reported indications suggested that pGKL mRNAs were polyadenylated at their 3' ends (18,87), which is reasonable, especially in the case of non-capped pGKL mRNAs, because 3' polyadenylation of eukaryotic mRNAs plays an important role in their stability (88,89) and the efficiency of their utilization by cellular translation apparatus (90-92). To answer this question, we performed a modified 3' RACE where a purified and DNase I-treated total yeast RNA was 3' tailed by an E. coli poly(A) polymerase using CTP instead of ATP. Following reverse transcription with an oligo(dG) anchored primer, PCR amplifications were performed with a universal anchor primer and appropriate gene-specific primers (see primer sequences in Table S2 ). At least twenty randomly selected 3' RACE fragments cloned in pCR4-TOPO plasmids were sequenced for each ORF. The 3' RACE analyses of all 15 pGKL ORFs are exemplified by the results from K1ORF1 ( Fig. 3B ; Table S1C), K2ORF5 and K2ORF10 (Table S1C) . Surprisingly, we found that pGKL mRNAs do not contain a 3' poly(A) tail. Transcripts produced from the pGKL plasmids contain heterogeneous 3' ends, suggesting the existence of more than one independent putative termination signal. Transcripts corresponding to only four ORFs are uniform at their 3' ends. This group is exemplified by K2ORF10 (Table S1C ). Four pGKL ORFs gave rise to transcripts that fell into two distinct groups with regard to their 3' UTR lengths, suggesting two independent transcription terminators. These transcripts are exemplified by those of K2ORF5 (Table S1C ). Transcripts of the remaining seven pGKL ORFs probably utilize three or more termination sites. Our initial bioinformatic analyses failed to determine the transcription termination signals. We detected only very weak RNA stem loops (∆G from -1.0 to -4.0) at the 3' UTR sequences that are probably not able to induce transcription termination.