Polycistronic gene expression in yeast versus cryptic promoter elements

Abstract

Saccharomyces cerevisiae is a much preferred host for biotechnological applications. However, the expression of entire heterologous pathways, required for some potential products, is technically challenging in yeast. A possible tool would be polycistronic gene expression. Recent studies demonstrated that short 5' untranslated regions (5'UTRs) found upstream of certain genes support cap-independent translation in vitro. In this study 5'UTRs were used as linkers between genes in polycistronic constructs. Expression levels of genes located in the first, second and third position after a promoter were studied by replacing the respective gene by a promoterless green fluorescence protein (GFP) gene. S. cerevisiae transformed with these constructs was grown on different carbon sources and GFP expression was assayed. Our results demonstrate that (i) ribosomal read-through does not suffice for polycistronic gene expression in vivo, (ii) 5'TFIID and 5'HAP4 but not 5'L-A significantly improve the expression of a reporter gene located second in a bicistron, (iii) 5'TFIID, 5'HAP4 and 5'YAP1 but not 5'L-A can drive expression of a promoterless reporter gene, and (iv) expression driven from 5'TFIID, 5'HAP4 and 5'YAP1 is induced in the presence of raffinose or galactose but not in the presence of glucose. This implies that these elements unlike typical internal ribosome entry site-like structures contain small, potentially useful promoters which support carbon source-regulated expression.

Testing of internal translation initiation via dicistronic constructs in yeast is complicated by production of extraneous transcripts

“…Therefore, in contrast to the results reported by Hecht et al (2002), our analysis indicated that some transcription initiated upstream the GAL1 promoter. However, the expression data suggests that the level of 3′ cistron expression from these RNAs was extremely low.…”

“…Furthermore, the cryptic promoter activity did not appear to depend on the presence of the 5′ reporter sequence. Low expression from the promoterless monocistronic reference construct suggested that no strong cryptic promoter activity resided within the vector-derived part of the 5′ UTR, reference sequence, or upstream the GAL1 promoter region, closely paralleling earlier studies with pYES vector that found no cryptic promoter activity in the vector backbone (Hecht et al, 2002). Thus, it appeared more likely that the cryptic promoters resided in the studied viral elements.…”

Cap-dependent and cap-independent translation in eukaryotic systems

“…In her first concern, there is more than enough evidence to justify caution (van Eyll and Michiels, 2002;Han and Zhang, 2002;Hecht et al, 2002;Dumas et al, 2003;Verge et al, 2004). However, it should be noted that, while both cryptic promoters or alternate splicing could be problems for in vivo experiments, this author is unfamiliar with any published example of in vivo mRNA breakage generating a unique mRNA species for translation.…”

Regulation of translation via mRNA structure in prokaryotes and eukaryotes

“…The most common test involves transposing a suspected IRES from its normal 5′ position to the midpoint of a synthetic dicistronic transcript and asking whether this allows translation of the 3′ cistron. The test is undermined in many cases, however, because the candidate IRES turns out to harbor a cryptic promoter or splice site, causing the dicistronic DNA vector to produce an unintended monocistronic mRNA (Dumas et al, 2003;Han and Zhang, 2002;Han et al, 2003a,b;Hecht et al, 2002;Liu et al, 2005;Sherrill et al, 2004;Van Eden et al, 2004;Vergé et al, 2004;Wang et al, 2005b). In many other cases, the RNA analyses required to rule out a cryptic promoter or splice site simply were not done.…”