Selection of Shine-Dalgarno sequences in plastids

Abstract

Like bacterial genes, most plastid (chloroplast) genes are arranged in operons and transcribed as polycistronic mRNAs. Plastid protein biosynthesis occurs on bacterial-type 70S ribosomes and translation initiation of many (but not all) mRNAs is mediated by Shine-Dalgarno (SD) sequences. To study the mechanisms of SD sequence recognition, we have analyzed translation initiation from mRNAs containing multiple SD sequences. Comparing translational efficiencies of identical transgenic mRNAs in Escherichia coli and plastids, we find surprising differences between the two systems. Most importantly, while internal SD sequences are efficiently recognized in E. coli, plastids exhibit a bias toward utilizing predominantly the 5'-most SD sequence. We propose that inefficient recognition of internal SD sequences provides the raison d'être for most plastid polycistronic transcripts undergoing post-transcriptional cleavage into monocistronic mRNAs.

Figure 1.

Construction of vectors to analyze various combinations of plastid translation initiation signals. (A) Physical map of the targeting region in the plastid genome after integration of transformation constructs of the pOD series. The BglII restriction sites used for RFLP analysis are marked. The transgenes are targeted to the intergenic region between the trnfM and trnG genes (42). The GFP expression cassette consists of the ribosomal RNA operon promoter (Prrn) fused to a Shine-Dalgarno (SD) sequence element (see panel B) and the 3′-UTR from the plastid rps16 gene (Trps16). The expected sizes of gfp transcripts are indicated (cf. Figure 3B). The location of the RFLP probe is shown as black bar. The selectable marker gene aadA is driven by a chimeric ribosomal RNA operon promoter (Prrn) and fused to the 3′-UTR from the psbA gene (TpsbA; ref. 45) (B) Schematic maps of the different translation initiation signals tested in this study (pOD vector series). SD sequences are shown in orange, start codons (ATG) and mini-ORFs are indicated in green with the sequence given below the map. TEV: tobacco etch virus peptidase cleavage site; GFP: gene for the green fluorescent protein. The difference between the two basic vectors pOD1 and pOD19 lies in the mutational elimination of an in-frame stop codon upstream of the SD to facilitate translation of GFP from the first SD in constructs pOD20 and pOD21.

Figure 2.

Analysis of gfp expression in E. coli. All experiments were repeated at least three times and identical results were obtained. Representative blots are shown here. (A) Western blot analyses to determine GFP accumulation levels in E. coli strains harboring the different pOD plasmids. Loaded amounts of total protein are indicated below the blots. Dashes denote empty lanes. As a control for loading, the high-molecular weight region of the gel (which was not blotted) was stained with Coomassie and is shown below each blot. pRB95: empty vector control. (B) Analysis of gfp mRNA accumulation in pOD strains. As a loading control, the blot was stripped and re-hybridized to a 16S rRNA-specific probe.

Figure 3.

Analysis of gfp expression in tobacco plastids. (A) Western blot analyses to determine GFP accumulation levels in transplastomic tobacco plants generated with pOD vectors. Loaded amounts of total protein are indicated below the blots. Dashes denote empty lanes. As a control for loading, the high-molecular weight region of the gel (which was not blotted) was stained with Coomassie and is shown below each blot. WT, wild-type (1 µg protein loaded); S5, transplastomic control (carrying the aadA but no gfp; ref. ; 1 µg protein loaded); GFP, purified recombinant GFP used as standard for quantitation. (B) Analysis of gfp mRNA accumulation in pOD transplastomic plants. The lower band represents monocistronic gfp mRNA, the upper band is the result of read through transcription and has been observed before in studies using the same plastid transformation vector backbone (pRB95; 28). To control for equal loading, the blot was stripped and re-hybridized to a 16S rRNA-specific probe.

Figure 4.

Confirmation of different translation initiation efficiencies by ribosome loading experiments. Polysome profiles of a weakly GFP accumulating transplastomic tobacco line (pOD8) and a strongly GFP accumulating line (pOD18) are shown. The lines were chosen, because they show similar gfp mRNA accumulation levels, but strong differences in GFP accumulation. Compared to the pOD8 plant, gfp mRNA distribution is shifted toward the bottom of the sucrose gradient in the pOD18 plant, indicating denser coverage with ribosomes and thus, higher translation efficiency. The bar diagrams above each blot show the amount of gfp mRNA present in each fraction (given in percent of the total gfp mRNA in the sample). As an internal control RNA, the distribution of the rbcL mRNA in the gradient was analyzed and, as expected, is identical in the two transplastomic lines. The wedge indicates the increasing sucrose concentration from the top to the bottom of the polysome gradient. For identification of polysome-containing gradient fractions, a puromycin-treated sample is also shown.