Synthetic lethality in the tobacco plastid ribosome and its rescue at elevated growth temperatures

Abstract

Consistent with their origin from cyanobacteria, plastids (chloroplasts) perform protein biosynthesis on bacterial-type 70S ribosomes. The plastid genomes of seed plants contain a conserved set of ribosomal protein genes. Three of these have proven to be nonessential for translation and, thus, for cellular viability: rps15, rpl33, and rpl36. To help define the minimum ribosome, here, we examined whether more than one of these nonessential plastid ribosomal proteins can be removed from the 70S ribosome. To that end, we constructed all possible double knockouts for the S15, L33, and L36 ribosomal proteins by stable transformation of the tobacco (Nicotiana tabacum) plastid genome. We find that, although S15 and L33 function in different ribosomal particles (30S and 50S, respectively), their combined deletion from the plastid genome results in synthetic lethality under autotrophic conditions. Interestingly, the lethality can be overcome by growth under elevated temperatures due to an improved efficiency of plastid ribosome biogenesis. Our results reveal functional interactions between protein and RNA components of the 70S ribosome and uncover the interdependence of the biogenesis of the two ribosomal subunits. In addition, our findings suggest that defining a minimal set of plastid genes may prove more complex than generally believed.

Figure 1.

Construction of Transplastomic Double Knockout Plants for the Nonessential Ribosomal Protein Genes rps15, rpl33, and rpl36. (A) Physical map of the region in the tobacco plastid genome (Shinozaki et al., 1986) containing the rpl33 gene and map of the transformed plastid genome (transplastome) in Δrpl33 mutant plants (Rogalski et al., 2008). Genes above the line are transcribed from the left to the right, and genes below the line are transcribed in the opposite direction. The AccI restriction sites employed for RFLP analyses are indicated, and the resulting fragment sizes are given for the wild type and the rpl33 knockout. The hybridization probe used for DNA gel blot analysis (derived from the psaJ coding region) is also indicated. (B) Map of the genomic region containing rps15 and map of the transformed plastid genome in Δrps15 mutants (Fleischmann et al., 2011). In the Δrps15 single mutant and the Δrps15/Δrpl36 double knockout, the spectinomycin resistance marker aadA (Svab and Maliga, 1993) disrupts the rps15 gene, whereas in the Δrps15/Δrpl33 plants, rps15 is inactivated with the kanamycin resistance gene aphA-6 (Huang et al., 2002). (C) Map of the genomic region containing rpl36 and map of the transformed plastid genome in Δrpl36 mutants (Fleischmann et al., 2011). In the Δrpl36 single mutant, the aadA marker disrupts the rpl36 gene, whereas in the Δrps15/Δrpl36 and the Δrpl33/Δrpl36 double knockout plants, rpl36 is inactivated with the aphA-6 marker. (D) RFLP analysis of plastid transformants. The wild type, the three single mutants, and the three double knockouts (two independently generated transplastomic lines each) were analyzed by DNA gel blotting using the restriction enzyme AccI and specific radiolabeled probes (pr.) for each of the three knockout alleles (cf. panels [A] to [C]). Note that all transplastomic lines included in this blot are homoplasmic and show exclusively the bands diagnostic of the transgenic plastid genomes.

Figure 2.

Phenotypes of Wild-Type Plants, Single Knockout Plants, and Double Knockout Plants of the Three Nonessential Plastid Ribosomal Protein Genes rps15, rpl33, and rpl36. The plants were grown from stem cuttings on synthetic medium supplemented with Suc as carbon source. (A) Comparison of the wild type with the single mutants ∆rps15, ∆rpl33, and ∆rpl36. While the ∆rps15 and ∆rpl33 plants display a wild-type-like phenotype, the ∆rpl36 knockout plants are pale and show strongly delayed greening (Rogalski et al., 2008; Fleischmann et al., 2011). (B) Phenotypes of the ∆rps15/∆rpl33, ∆rps15/∆rpl36, and ∆rpl33/∆rpl36 double knockout plants. Note the strong pigment deficiency (particularly pronounced in the older leaves) in all three double mutants.

Figure 3.

Accumulation and Processing of rRNAs in Ribosomal Protein Mutants Grown Heterotrophically on Synthetic Medium. Wild-type plants, single mutants of rps15, rpl33, and rpl36, and the three double mutants were analyzed. (A) Accumulation of the 16S rRNA as a proxy for the accumulation of the 30S ribosomal subunit. The values give the ratio of the plastid 16S rRNA to the cytosolic 18S rRNA and represent the means of three biological replicates. The error bars indicate the sd. Statistically significant differences (determined by one-way ANOVA and Fisher’s LSD test, P < 0.05) are indicated by the letters above the bars. (B) Accumulation of the 23S rRNA as a proxy for the abundance of 50S ribosomal subunits. (C) Physical map and transcript processing pattern of the plastid rRNA operon. The 7.3-kb primary transcript, the different processing intermediates, and the mature forms of the 16S and 23S rRNAs are shown. The 23S rRNA is cut into three pieces, a phenomenon known as “hidden break” processing (Delp and Kössel, 1991). The positions of the hybridization probes for the 16S and 23S rRNAs are also indicated. (D) Accumulation and processing of the plastid 16S rRNA determined by RNA gel blot hybridization. As loading control, the ethidium bromide–stained gel region containing the cytosolic 18S rRNA is shown. (E) Analysis of the accumulation and processing pattern of the 23S rRNA. Note quantitative differences in the efficiency of hidden break processing, which is known to be influenced by developmental cues (Rosner et al., 1974).

Figure 4.

Seedling Lethality of the ∆rps15/∆rpl33 Mutants and Rescue by Elevated Growth Temperatures. Wild-type, ∆rps15, ∆rpl33, and ∆rps15/∆rpl33 plants were germinated either on soil ([A] and [C]) or on synthetic Suc-containing medium ([B] and [D]) and grown either at 22°C ([A] and [B]) or at 35°C ([C] and [D]). (A) Plants germinated and grow on soil at 22°C. The photos were taken 29 d after sowing. (B) Plants germinated and grown on synthetic medium at 22°C in the absence of antibiotics (w/o sel.) or in the presence of 500 µg/mL spectinomycin (Spec) or 50 µg/mL kanamycin (Kan). The photos were taken 22 d after sowing. (C) Plants germinated and grown on soil at 35°C. (D) Plants grown as in (B) but at 35°C. The plants grown without selection and the plants grown under spectinomycin selection were photographed 15 d after sowing, and the plants grown under kanamycin selection were photographed 41 d after sowing. Due to the low stability of kanamycin at elevated temperatures, the kanamycin-containing medium was replaced every 14 d.

Figure 5.

Growth at Elevated Temperatures Alleviates the Molecular Phenotype of ∆rps15/∆rpl33 Mutants. (A) and (B) Accumulation of rRNAs as a proxy for the corresponding ribosomal subunits in seedlings of the wild type and the ribosomal protein mutants ∆rps15, ∆rpl33, and ∆rps15/∆rpl33. The plants were germinated and grown for 29 d on synthetic medium at either 22 or 35°C. The mean of three biological replicates is shown. The error bars indicate the sd. Statistically significant differences (two-way ANOVA and Scheffé test; P < 0.05) are indicated by the letters above the bars. (A) Ratio of the plastid 16S rRNA to the cytosolic 18S rRNA at 22°C (dark-gray bars) and 35°C (light-gray bars). (B) Ratio of the plastid 23S rRNA to the cytosolic 18S rRNA at 22 and 35°C. (C) and (D) Accumulation and processing patterns of plastid rRNAs at 22 and 35°C. The ethidium bromide–stained cytosolic 18S rRNA band of the agarose gel prior to blotting is shown below each blot as a loading control. For a schematic representation of rRNA processing patterns, see Figure 3C. (C) Comparison of 16S rRNA accumulation and processing at 22°C versus 35°C. (D) Comparison of 23S rRNA accumulation and processing at 22°C versus 35°C.

Figure 6.

Comparison of Translational Activities at 22 and 35°C by Polysome Analysis in Wild-Type and ∆rps15/∆rpl33 Mutant Plants. Polysome profiles of the polycistronic psaA/psaB/rps14 transcript are shown. Polysomal complexes were separated in Suc gradients and each gradient was fractionated into six fractions (numbered from the top to the bottom, as indicated above each panel). The extracted RNAs were separated by gel electrophoresis, blotted, and hybridized to a radiolabeled psaA probe. The gradient fractions containing the bulk of the transcripts are indicated by brackets. As a control, polysomes were isolated in the presence of the polysome-dissociating antibiotic puromycin. The ethidium bromide–stained gel region containing the cytosolic 26S and 18S rRNAs is shown below each blot. (A) Polysome loading of the psaA/psaB/rps14 transcript in wild-type plants at 35°C and 24 h after shifting the growth temperature to 22°C. (B) Polysome loading in ∆rps15/∆rpl33 mutant plants at 35°C and 24 h after shifting the growth temperature to 22°C. (C) Dissociation of polysomal complexes by puromycin treatment in wild-type and mutant plants.

Figure 7.

Position of the Ribosomal Proteins S15, L33, and L36 in the 70S Ribosome. Shown are the 16S rRNA (pink ribbon) and the S15 protein (dark gray and yellow) of the small ribosomal subunit (30S) as well as the 23S rRNA (gray ribbon), the L33 (light blue and yellow), and the L36 (brown) proteins of the large ribosomal subunit (50S). (A) Localization of S15, L33, and L36 in relation to the 16S and 23S rRNAs. (B) Close-up view revealing the interaction of S15 with the 23S rRNA via an RNA loop that protrudes into the 30S subunit.