"Life is short, and art is long": RNA degradation in cyanobacteria and model bacteria

Abstract

RNA turnover plays critical roles in the regulation of gene expression and allows cells to respond rapidly to environmental changes. In bacteria, the mechanisms of RNA turnover have been extensively studied in the models Escherichia coli and Bacillus subtilis, but not much is known in other bacteria. Cyanobacteria are a diverse group of photosynthetic organisms that have great potential for the sustainable production of valuable products using CO₂ and solar energy. A better understanding of the regulation of RNA decay is important for both basic and applied studies of cyanobacteria. Genomic analysis shows that cyanobacteria have more than 10 ribonucleases and related proteins in common with E. coli and B. subtilis, and only a limited number of them have been experimentally investigated. In this review, we summarize the current knowledge about these RNA-turnover-related proteins in cyanobacteria. Although many of them are biochemically similar to their counterparts in E. coli and B. subtilis, they appear to have distinct cellular functions, suggesting a different mechanism of RNA turnover regulation in cyanobacteria. The identification of new players involved in the regulation of RNA turnover and the elucidation of their biological functions are among the future challenges in this field.

Keywords: RNA maturation; RNA metabolism; RNA turnover; cyanobacteria; ribonucleases.

Figure 1

The principal pathway of messenger RNA (mRNA) degradation in bacteria. The major steps include the following: RNA phosphohydrolase converts the 5ʹ end of the primary transcripts from triphosphate to monophosphate; endoribonucleases internally cleave the transcripts into intermediate fragments; exoribonucleases degrade the intermediates into monoribonucleotides and generate the end products of 2‐ to 5‐nt oligoribonucleotides (nanoRNAs); and oligoribonucleases hydrolyze the oligoribonucleotides into monoribonucleotides, finishing the degradation process. Note that many transcripts in the triphosphorylated form can also be substrates of endoribonucleases and some transcripts may also be substrates of exoribonucleases before endoribonucleolytic cleavage. The major Escherichia coli and cyanobacterial enzymes involved in the degradation process are shown. The enzymes present in both E. coli and cyanobacteria (RNase E, RNase III, PNPase, and RNase II) are in black, those currently discovered in E. coli only (RppH, RNase R, and Orn) are in purple, and the one present in cyanobacteria but not in E. coli (RNase J) is in red. Note that RNase J acts as both an endoribonuclease and a 5ʹ−3ʹ exoribonuclease.

Figure 2

Schematic representation of the assembly of the Anabaena and the Escherichia coli RNA degradosome. (A) Anabaena RNA degradosome; (B) Escherichia coli RNA degradosome. RNase E forms tetramers; here, only one of them (in green) is shown for simplicity. The components of the degradosome are shown relative to the positions of the RNase E proteins. Four conserved subregions (C1, C2, C3 and C4) and three variable subregions (V1, V2 and V3) have been identified in the noncatalytic region of Anabaena RNase E. Note that the E. coli RNase E catalytic region is composed of a large domain that is equivalent to the catalytic region of Anabaena RNase E and a small domain that has no counterpart in Anabaena RNase E.

Figure 3

RNase E cleavage of messenger RNAs (mRNAs) encoding photosynthetic proteins in Synechocystis PCC 6803. (A) Upper panel: The psbA2 mRNA has a 49‐nt 5ʹ UTR that, in the dark, is sensitive to RNase E cleavage due to multiple RNase E cleavage sites within an AU‐rich sequence that extends into the likely ribosome binding site (RBS), , . Hence, rapidly interacting ribosomes under conditions such as high light, when this mRNA is strongly translated, can protect against degradation (lower panel). In addition, the antisense RNA (asRNA) PsbA2R is coregulated with the mRNA from a transcriptional start site leading to an overlap with the first 19 nucleotides of the psbA2 mRNA, hence protecting these sites against RNase E cleavage, as indicated by the red crosses. A short stem‐loop near the mRNA 5ʹ end might be relevant for recognition. Note that in vivo additional ribonucleases are likely involved. Note that the gene psbA3, which is almost identical to psbA2, is regulated in the same way, . (B) Upper panel: The psaL gene is preceded by a 55‐nt 5ʹ UTR that is not targeted by RNase E under most growth conditions. However, under high light, transcription of the gene for the sRNA PsrR1, which is located elsewhere in the genome, is stimulated. Except for two mismatches, PsrR1 interacts over 22 consecutive nucleotides with the psaL mRNA (lower panel). This interaction overlaps the 3ʹ end of the 5ʹ UTR, including the ribosome binding site, the start codon, and one additional nucleotide of the second codon. This interaction leads to conditional recruitment of RNase E, which then cleaves a single‐bond seven nucleotides into the coding sequence, effectively decapitating the psaL mRNA.

Figure 4

Grad‐seq analysis aids the analysis of ribonucleases, auxiliary proteins, and ribonucleoprotein complexes. A typical sedimentation profile obtained in the analysis of the cyanobacterium Synechocystis PCC 6803 is shown. The positions of several major macromolecular complexes as determined by mass spectrometry are given to the left, the respective fraction numbers and sucrose percentages are indicated along the gradient. The different colors result from the native pigmentation of protein–pigment complexes involved in photosynthesis. The distribution of distinct groups of RNAs is sketched by the colored lines to the right of the gradient. The positions of abundant RNA–protein complexes, such as two of the three CRISPR complexes and noncoding RNA–ribonucleoprotein complexes containing 6S RNA or transfer‐messenger RNA (tmRNA) are shown. Note that characterized regulatory small RNAs such as PsrR1 peaked in fraction 7 (F7) together with the bulk of mRNAs, but there were secondary peaks in mRNA abundance in other fractions (for details, see Riediger et al. ¹⁹⁰ ). Several proteins and RNA‐protein complexes involved in RNA metabolism, such as RNase D (D), RNase J (J), RNase E (E), RNase P (P), PNPase, enolase (Eno), and CrhR occur in the higher molecular fractions, indicating their association with larger complexes. Most RNAs were detected in overlapping fractions as well, indicating their likely direct association with such complexes. The striking overlap in the in‐gradient distribution of PNPase, enolase, RNase E and J, consistent with their possible colocalization into degradosomes is boxed. Two different gene products were detected for RNase II/R (II/R) and RNase III (III) in the lighter fractions, while Mini‐III was not detected at all. The strong correlation between RNase Z (Z) and the bulk of tRNAs is consistent with the role of this enzyme in tRNA maturation. Note that Hfq was found only in very light fractions, consistent with its non‐RNA binding character in cyanobacteria. Candidates for alternative RNA chaperones are the KhpA/B homologs Slr0287 and Slr1472. See Table 1 for the gene IDs of all other mentioned proteins. The entire data set is available at https://sunshine.biologie.uni-freiburg.de/GradSeqExplorer/. Reprinted in modified form with permission from Riediger et al..