Requirements for ribosomal protein S1 for translation initiation of mRNAs with and without a 5' leader sequence

It has previously been proposed that Escherichia coli ribosomal protein S1 is required for the translation of highly structured mRNAs. In this study, we have examined the influence of structural features at or near the start codon of different mRNAs. The requirement for ribosomal protein S1 for translation initiation was determined when (i) the ribosome-binding site (RBS) was either preceded by a 5' non-translated leader sequence; (ii) the RBS was located 5' proximal to a mRNA start codon; and (iii) the start codon was the 5' terminal codon as exemplified by leaderless mRNAs. In vitro translation studies revealed that the leaderless lambda cl mRNA is translated with Bacillus stearothermophilusribosomes, naturally lacking a ribosomal protein S1 homologue, whereas ompA mRNA containing a 5' leader is not. These studies have been verified by toeprinting with E. coli ribosomes depleted for S1. We have shown that S1 is required for ternary complex formation on ompA mRNA but not for leaderless mRNAs or for mRNAs in which the RBS is close to the 5' end.     

Codon replacement in the PGK1 gene of Saccharomyces cerevisiae: experimental approach to study the role of biased codon usage in gene expression.

The coding sequences of genes in the yeast Saccharomyces cerevisiae show a preference for 25 of the 61 possible coding triplets. The degree of this biased codon usage in each gene is positively correlated to its expression level. Highly expressed genes use these 25 major codons almost exclusively. As an experimental approach to studying biased codon usage and its possible role in modulating gene expression, systematic codon replacements were carried out in the highly expressed PGK1 gene. The expression of phosphoglycerate kinase (PGK) was studied both on a high-copy-number plasmid and as a single copy gene integrated into the chromosome. Replacing an increasing number (up to 39% of all codons) of major codons with synonymous minor ones at the 5' end of the coding sequence caused a dramatic decline of the expression level. The PGK protein levels dropped 10-fold. The steady-state mRNA levels also declined, but to a lesser extent (threefold). Our data indicate that this reduction in mRNA levels was due to destabilization caused by impaired translation elongation at the minor codons. By preventing translation of the PGK mRNAs by the introduction of a stop codon 3' and adjacent to the start codon, the steady-state mRNA levels decreased dramatically. We conclude that efficient mRNA translation is required for maintaining mRNA stability in S. cerevisiae. These findings have important implications for the study of the expression of heterologous genes in yeast cells.     

Translation of aberrant mRNAs lacking a termination codon or with a shortened 3'-UTR is repressed after initiation in yeast

A novel mRNA surveillance for mRNA lacking a termination codon (nonstop mRNA) has been proposed in which Ski7p is thought to recognize stalled ribosomes at the 3' end of mRNA. Here we report our analysis of translation and decay of nonstop mRNAs in Saccharomyces cerevisiae. Although the reduction of nonstop mRNAs was only 4.5-fold, a level that is sufficient for residual protein synthesis, translation products of nonstop mRNAs were hardly detectable. We show that nonstop mRNAs were associated with polysomes, but not with Pab1p. We also show that ribosomes translating nonstop mRNA formed stable and heavy polysome complexes with mRNA. These data suggest that ribosome stalling at the 3' end of nonstop mRNA may block further rounds of translation, hence repressing protein synthesis. Furthermore, it was found that the 5' --> 3' decay pathway was accelerated for nonstop mRNA decay in the absence of Ski7p. We also found that translation of aberrant mRNAs with a shortened 3'-UTR was repressed, suggesting that an improper spatial distance between the termination codon and the 3' end of mRNA results in translation repression.     

Nonsense mutations in close proximity to the initiation codon fail to trigger full nonsense-mediated mRNA decay

Nonsense-mediated mRNA decay (NMD) is a surveillance mechanism that degrades mRNAs containing premature translation termination codons. In mammalian cells, a termination codon is ordinarily recognized as "premature" if it is located greater than 50-54 nucleotides 5' to the final exon-exon junction. We have described a set of naturally occurring human beta-globin gene mutations that apparently contradict this rule. The corresponding beta-thalassemia genes contain nonsense mutations within exon 1, and yet their encoded mRNAs accumulate to levels approaching wild-type beta-globin (beta(WT)) mRNA. In the present report we demonstrate that the stabilities of these mRNAs with nonsense mutations in exon 1 are intermediate between beta(WT) mRNA and beta-globin mRNA carrying a prototype NMD-sensitive mutation in exon 2 (codon 39 nonsense; beta 39). Functional analyses of these mRNAs with 5'-proximal nonsense mutations demonstrate that their relative resistance to NMD does not reflect abnormal RNA splicing or translation re-initiation and is independent of promoter identity and erythroid specificity. Instead, the proximity of the nonsense codon to the translation initiation AUG constitutes a major determinant of NMD. Positioning a termination mutation at the 5' terminus of the coding region blunts mRNA destabilization, and this effect is dominant to the "50-54 nt boundary rule." These observations impact on current models of NMD.     

Influenza B virus genome: sequences and structural organization of RNA segment 8 and the mRNAs coding for the NS1 and NS2 proteins.

Double-stranded DNA derived from influenza B virus genome RNA segment 8, which codes for the NS1 and NS2 proteins, was constructed by hybridization of full-length cDNA copies of RNA segment 8 and of the NS1 mRNA. This DNA was cloned in plasmid pBR322 and sequenced. The NS1 mRNA (approximately 1,080 viral nucleotides) contains nonviral nucleotides at its 5' end and is capable of coding for a protein of 281 amino acids. Sequencing of the NS2 mRNA has shown that it contains an interrupted sequence of 655 nucleotides and is most likely synthesized by a splicing mechanism. The first approximately 75 virus-specific nucleotides at the 5' end of the NS2 mRNA are the same as are found at the 5' -end of the NS1 mRNA. This region contains the initiation codon for protein synthesis and coding information for 10 amino acids common to the two proteins. The approximately 350-nucleotide body region of the NS2 mRNA can be translated in the +1 reading frame, and the sequence indicates that the NS1 and NS2 protein-coding regions overlap by 52 amino acids translated from different reading frames. Thus, between the influenza A and B viruses, the organization of the NS1 and NS2 mRNAs and the sizes of the NS2 mRNA and protein are conserved despite the larger size of the influenza B virus RNA segment, NS1 mRNA, and NS1 protein.     

Alteration of amino-terminal codons of human granulocyte-colony-stimulating factor increases expression levels and allows efficient processing by methionine aminopeptidase in Escherichia coli

We have improved the expression of recombinant human granulocyte-colony-stimulating factor (G-CSF), produced by either pL or trpP expression vectors in Escherichia coli, by altering the sequence at the 5' end of the G-CSF-coding region. Initial attempts to express G-CSF resulted in neither detectable G-CSF mRNA nor protein in the trpP system, and only G-CSF mRNA was detectable in the pL system. We modified both expression vectors to decrease the G + C content of the 5' end of the coding region without altering the predicted amino acid sequence. This resulted in expression of detectable G-CSF mRNA and protein in both systems. Expression reached 17% and 6.5% of the total soluble cellular protein in the pL and trpP expression systems, respectively. The N-terminal sequence of the recombinant G-CSF from the pL system was Met-Thr-Pro-Leu-Gly-Pro-. G-CSF isolated from several human cell lines (including the LD-1 cell line reported here), does not have an N-terminal methionyl residue. Deletion of the threonine codon at the beginning of the coding region for the mature G-CSF resulted in efficient removal of the N-terminal methionine residue during expression in E. coli.     

gamma and gamma' chains of human fibrinogen are produced by alternative mRNA processing

cDNAs and the genomic DNA coding for the gamma and gamma' chains of human fibrinogen have been isolated and characterized by sequence analysis. The cDNAs coding for the gamma and gamma' chains share a common nucleotide sequence coding for the first 407 amino acid residues in each polypeptide chain. The predominant gamma chain contains an additional four amino acids on its carboxyl-terminal end (residues 408-411). These four amino acids, together with the 3' noncoding sequences, are encoded by the tenth exon. Removal of the ninth intervening sequence following the processing and polyadenylation reactions yields a mature mRNA coding for the predominant gamma chain. The less prevalent gamma' chain contains 20 amino acids at its carboxyl-terminal end (residues 408-417). These 20 amino acids are encoded by the immediate 5' end of the ninth intervening sequence. This results from an occasional processing and polyadenylation reaction that occurs within the region normally constituting the ninth intervening sequence. Accordingly, the gene for the gamma chain of human fibrinogen gives rise to two mRNAs that differ in sequence on their 3' ends. These mRNAs code for polypeptide chains with different carboxyl-terminal sequences. Both of these polypeptides are incorporated into the fibrinogen molecule present in plasma.     

Translation of the leaderless Caulobacter dnaX mRNA.

The expression of the Caulobacter crescentus homolog of dnaX, which in Escherichia coli encodes both the gamma and tau subunits of the DNA polymerase III holoenzyme, is subject to cell cycle control. We present evidence that the first amino acid in the predicted DnaX protein corresponds to the first codon in the mRNA transcribed from the dnaX promoter; thus, the ribosome must recognize the mRNA at a site downstream of the start codon in an unusual but not unprecedented fashion. Inserting four bases in front of the AUG at the 5' end of dnaX mRNA abolishes translation in the correct frame. The sequence upstream of the translational start site shows little homology to the canonical Shine-Dalgarno ribosome recognition sequence, but the region downstream of the start codon is complementary to a region of 16S rRNA implicated in downstream box recognition. The region downstream of the dnaX AUG, which is important for efficient translation, exhibits homology with the corresponding region from the Caulobacter hemE gene adjacent to the replication origin. The hemE gene also appears to be translated from a leaderless mRNA. Additionally, as was found for hemE, an upstream untranslated mRNA also extends into the dnaX coding sequence. We propose that translation of leaderless mRNAs may provide a mechanism by which the ribosome can distinguish between productive and nonproductive templates.     

Structure of a cluster of mouse histone genes.

The four mouse histone genes (2 H3 genes, an H2b gene and an H2a gene) present in a cloned 12.9 kilobase fragment of DNA have been completely sequenced including both 5' and 3' flanking regions. These genes are expressed in cultured mouse cells and the 3' and 5' ends of the mRNA have been determined by S1 nuclease mapping. These genes code for a minor fraction of the histone mRNAs expressed in cultured mouse cells. They comprise at most 5-8% of the total histone mRNA of each type. The two H3 genes code for H3.2 and H3.1 histone proteins, while the H2b gene codes for an H2b.1 protein with a single amino acid change (val-leu) at position 18. Only the 3' portion of the H2a gene is contained in the clone and there is an amino acid change (alanine-proline) at position 126. Comparison of the 5' and 3' flanking sequences reveals a conserved sequence at the 3' end of the mRNA which forms a hairpin loop structure. The codon usage in the genes is non-random and there has been no discrimination against CG doublets in the coding region of the genes.     

The size heterogeneity of rat insulin-like growth factor-I mRNAs is due primarily to differences in the length of 3'-untranslated sequence

Previous studies have revealed multiple size classes of rat insulin-like growth factor-I (IGF-I) of estimated size 7.5-7.0, 1.9-1.5, and 1.2-0.9 kilobases (kb). Available sequence information accounts for only 2.1 kb of the 7.5-7.0 kb IGF-I mRNAs. We used oligomer directed ribonuclease H (RNase H) mapping to define the extent to which the unknown sequence in the large molecular weight mRNAs lies 5' or 3' to known sequence. Rat liver polyadenylated RNAs were incubated with oligomer probes complementary to internal rat IGF-I precursor (E domain) coding sequences. RNase H was used to hydrolyze IGF-I mRNAs at the point of annealment with the oligomers. Resultant 5' and 3'-IGF-I mRNA fragments were analyzed on Northern blots. A probe specific for type 1 (class C) 5'-sequences (the most predominant of multiple 5'-sequence types found on rat IGF-I mRNAs) identifies intact IGF-I mRNAs of 7.5-7.0, 1.9-1.5 and 1.2-0.9 kb but, after oligomer directed RNase cleavage of these mRNAs, identified only a single IGF-I mRNA 5'-fragment. Major differences in the length of sequence 5' to the IGF-I coding sequence therefore, do not account for the multiple size classes of type 1 (class C) IGF-I mRNAs. The size of the 5'-fragment suggests that the extent of sequence 5' to the IGF-I coding sequence is 0.4-0.7 kb in type 1 (class C) IGF-I mRNAs. Identification of multiple 3'-fragments of IGF-I mRNAs demonstrated heterogeneity in the 3'-ends of rat IGF-I mRNAs.(ABSTRACT TRUNCATED AT 250 WORDS)