Non-stop mRNA decay initiates at the ribosome

Abstract

The translation machinery deciphers genetic information encoded within mRNAs to synthesize proteins needed for various cellular functions. Defective mRNAs that lack in-frame stop codons trigger non-productive stalling of ribosomes. We investigated how cells deal with such defective mRNAs, and present evidence to demonstrate that RNase R, a processive 3'-to-5' exoribonuclease, is recruited to stalled ribosomes for the specific task of degrading defective mRNAs. The recruitment process is selective for non-stop mRNAs and is dependent on the activities of SmpB protein and tmRNA. Most intriguingly, our analysis reveals that a unique structural feature of RNase R, the C-terminal lysine-rich (K-rich) domain, is required both for productive ribosome engagement and targeted non-stop mRNA decay activities of the enzyme. These findings provide new insights into how a general RNase is recruited to the translation machinery and highlight a novel role for the ribosome as a platform for initiating non-stop mRNA decay.

Figure 1

[A] Schematic representation of the domain architecture of RNase R and RNase II. RNase R and RNase II share extensive similarity in the N-terminal cold-shock, central nuclease and C-terminal S1 domains. RNase R has two additional domains, a N-terminal putative helix-turn-helix (HTH) domain and a C-terminal lysine-rich (K-rich) domain. [B] RNase R truncation variants used in this study, missing either the N-terminal HTH domain or various lengths of the C-terminal K-rich domain. [C] Schematic representation of the λ-cI-N nonstop and λ-cI-N stop reporters. The λ-cI-N coding region is represented as a rectangle and the nucleotide sequence of the trpA terminator, located at the 3′-end of the transcript, is shown.

Figure 2

Decay rates for λ-cI-N nonstop reporter mRNA as estimated from rifampicin chase experiments. Aliquots of cells expressing the λ-cI-N nonstop and RNase R, or one of its C-terminal truncation variants, were taken at indicated time points post rifampicin addition and used for RNA purification. Total RNA isolated from equal number of cells was used for northern blots, which were hybridized with the λ-cI-N specific probe. Bands on the blot were quantified by scanning and subsequent analysis with Image-J (http://rsbweb.nih.gov/ij/). mRNA half-lives were calculated by linear regression analysis of data obtained from at least three independent experiments. pACYCDuet-1 was used as the vector control.

Figure 3

Decay rates for λ-cI-N stop reporter mRNA as estimated from rifampicin chase experiments. Aliquots of cells expressing RNase R or RNR⁷²³ and λ-cI-N stop were taken at indicated times post rifampicin addition and used for RNA purification. Total RNA isolated from equal number of cells was used for northern blots, which were hybridized with a λ-cI-N specific probe. Bands on the blot were quantified by scanning and subsequent analysis with Image-J (http://rsbweb.nih.gov/ij/). mRNA half-lives were calculated by linear regression analysis of data obtained from at least three independent experiments. pACYCDuet-1 was used as the vector control.

Figure 4

[A] Analysis of the ability of RNase II, RNase R, and RNase R truncation variants to degrade a single stranded RNA substrate (ss32) in vitro. The ³²P-labeled ss32 substrate was used to examine the degradation capacity of these RNases in vitro. Samples from each degradation reaction were taken at designated time points and resolved by electrophoresis on a denaturing polyacrylamide gel and visualized by autoradiography. [B] The ability of RNase II, RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-U overhang. The experiment was performed as described in Figure 4A. [C]. The ability of RNase II and RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-A overhang.

Figure 4

[A] Analysis of the ability of RNase II, RNase R, and RNase R truncation variants to degrade a single stranded RNA substrate (ss32) in vitro. The ³²P-labeled ss32 substrate was used to examine the degradation capacity of these RNases in vitro. Samples from each degradation reaction were taken at designated time points and resolved by electrophoresis on a denaturing polyacrylamide gel and visualized by autoradiography. [B] The ability of RNase II, RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-U overhang. The experiment was performed as described in Figure 4A. [C]. The ability of RNase II and RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-A overhang.

Figure 4

[A] Analysis of the ability of RNase II, RNase R, and RNase R truncation variants to degrade a single stranded RNA substrate (ss32) in vitro. The ³²P-labeled ss32 substrate was used to examine the degradation capacity of these RNases in vitro. Samples from each degradation reaction were taken at designated time points and resolved by electrophoresis on a denaturing polyacrylamide gel and visualized by autoradiography. [B] The ability of RNase II, RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-U overhang. The experiment was performed as described in Figure 4A. [C]. The ability of RNase II and RNase R, and RNase R truncation variants to degrade a RNA substrate with an internal stem-loop structure followed by a 3′ poly-A overhang.

Figure 5

RNase R is selectively enriched in ribosome fractions translating a λ-cI-N nonstop mRNA and the enrichment is dependent on the presence of the C-terminal K-rich domain of RNase R. Ribosomes translating reporter mRNAs, either the normal λ-cI-N stop or a λ-cI-N nonstop mRNA, were isolated from the total cellular ribosome pool, utilizing the N-terminal His₆ tag encoded on the reporter protein. Protein components of the isolated ribosomes were resolved by electrophoresis on a 10% SDS-polyacrylamide gel and the presence of RNase R, or the RNase R⁷²³ truncation variant, was detected by western blot analysis using RNase R specific antiserum. A representative western blot is shown in the top panel. Each experiment was repeated 3 times and the intensity of the bands corresponding to enriched RNase R or RNase R⁷²³ were quantified by scanning and subsequent analysis with Image-J (http://rsbweb.nih.gov/ij/). The average quantified data is presented in the lower panel.

Figure 6

The recruitment of RNase R to stalled ribosomes is dependent on the presence of tmRNA and SmpB protein. Reporter mRNAs, λ-cI-N stop or a λ-cI-N nonstop, were expressed in either ssrA⁻smpB⁻ strain or its isogenic parental strain. The isolation of ribosomes translating the reporter mRNAs and the detection of RNase R in these ribosomes were performed as described in the Fig. 5.