tmRNA determinants required for facilitating nonstop mRNA decay

  1. Preeti Mehta1,3,
  2. Jamie Richards1,3, and
  3. A. Wali Karzai1,2
  1. 1Department of Biochemistry and Cell Biology, Stony Brook University, Stony Brook, New York 11794, USA
  2. 2Center for Infectious Diseases, Stony Brook University, Stony Brook, New York 11794, USA

  1. ↵3 These authors contributed equally to this work.

 
Next Section

Abstract

In bacteria, ribosomes stalled on nonstop mRNAs are rescued by tmRNA in a unique process called trans-translation. The two known tmRNA functions in trans-translation are (1) a tRNA-like function, which transfers the partially synthesized protein fragment to itself; and (2) an mRNA-like function, which enables ribosomes to resume and terminate translation on tmRNA as a surrogate template. We present evidence to demonstrate that tmRNA performs a third function, namely, facilitating the degradation of the causative defective mRNA. Our investigations have revealed the identity of key sequence determinants that promote the degradation of the nonstop mRNA. These sequence determinants are located in the distal part of the tmRNA open reading frame, encoding the ultimate, penultimate, and anti-penultimate amino acids of the peptide tag. We show that mutation of these tmRNA sequence elements has an adverse affect on the disposal of the nonstop mRNA, while leaving the tRNA and mRNA functions entirely unaffected. More significantly, specific mutations that change the nucleotide sequence of the peptide-reading frame without altering the nature or identity of the encoded amino acids still exhibit the characteristic defect in nonstop mRNA decay. In contrast, mutations in codons 3, 4, 5, and 6 of the tmRNA open reading frame do not have an adverse affect on degradation of defective mRNAs. Based on these results, we propose that tmRNA plays an important role in promoting the decay of nonstop mRNAs and that sequence elements in the distal segment of the peptide-reading frame make sequence-specific contributions that are crucial for this activity.

Keywords

Previous SectionNext Section

INTRODUCTION

Ribosomes read and decode genetic information contained within messenger RNAs to synthesize proteins. To accurately synthesize the intended polypeptide, ribosomes must recognize the correct initiation and termination signal encoded in each message. In bacteria, the signals required for initiation of translation are located near the 5′ end of the mRNA, and an in-frame termination codon is required to recruit protein factors needed for release of the nascent polypeptide chain and recycling of the ribosomal machinery. In theory, mRNAs that lack the requisite initiation signals should not engage the translation machinery or cause any untoward consequences for the cell. Conversely, mRNAs that lack in-frame stop codons are fully competent to engage the translation machinery, initiate normally, and continue until the 3′ end of the message is reached. At this point, neither continuation of translation nor normal peptide-chain release is a viable option, and the ribosome stalls, leaving the partially synthesized protein chain attached to the P-site tRNA. Additionally, since each message is translated by multiple ribosomes, obstruction by the leading ribosome would result in stalling of all ribosomes trailing on the same message. Thus, nonstop mRNAs present three challenges for the cell that, if not corrected, could adversely affect its growth potential. The first is the need for recycling of the sequestered ribosomes to permit translation of new mRNAs. The second is the potentially deleterious consequences of releasing partially synthesized protein fragments into the cell, as these fragments might have poor solubility or unregulated activities. The third is dealing with the defective mRNAs that, if not effectively removed, could engage ribosomes in further futile cycles of translation.

How do cells deal with nonstop messages (mRNAs that lack a stop codon)? Bacteria have evolved a unique and highly conserved translation rescue mechanism, known as trans-translation (Karzai et al. 2000; Withey and Friedman 2003), which effectively resolves all three problems associated with nonstop mRNAs. This unique mechanism enables the cell to recognize and rescue stalled ribosomes, mark the linked incomplete polypeptides for proteolysis, and promote degradation of the causative defective mRNAs. The principal components of trans-translation are small stable RNA A (SsrA) (Keiler et al. 1996), also known as tmRNA, and its associated protein cofactor small protein B (SmpB) (Karzai et al. 1999; Sundermeier et al. 2005; Dulebohn et al. 2006). tmRNA is the only known RNA molecule that has the combined structural and functional properties of a tRNA and an mRNA. The current model (Keiler et al. 1996; Karzai et al. 2000; Withey and Friedman 2003) for tmRNA function proposes that alanine-charged tmRNA recognizes stalled ribosomes, binds the A site, and acts as a tRNA to accept the nascent polypeptide. The mRNA-like segment of tmRNA, with its in-frame stop codon, is then engaged to provide an alternate reading frame for the translation machinery and promote proper termination and recycling of ribosomes. The tmRNA-encoded peptide is thus cotranslationally appended to the C terminus of the nascent polypeptide and serves as a recognition signal for rapid degradation by cellular proteases (Gottesman et al. 1998). tmRNA has been implicated in degradation of defective mRNAs (Yamamoto et al. 2003). However, which parts of tmRNA are important for facilitating nonstop mRNA decay or what RNase(s) participate in degradation of the defective message have not yet be elucidated. Intriguingly, a complex of SmpB protein and tmRNA contained several associated factors, one of which was the highly processive 3′-to-5′ exoribonuclease RNase R (Karzai and Sauer 2001).

This study was undertaken to characterize the role of tmRNA in degradation of defective mRNA and define tmRNA sequence elements that facilitate the decay process. Here, we provide evidence to show that the peptide-reading frame of tmRNA plays a crucial role in facilitating degradation of defective mRNAs.

Previous SectionNext Section

RESULTS

Nonstop mRNA stability

These investigations were prompted by our repeated observations that defective messages accumulate in an ssrA− background. Hence, we wished to examine whether tmRNA played a role in the degradation of defective mRNAs, and if so, which determinants of tmRNA facilitate the decay process. To this end, we made use of a well-characterized reporter gene lacking in-frame stop codons. This reporter gene (λ-cI-N-trpAt) consists of the N-terminal domain of λ repressor, a Flag-tag epitope, and a trpA terminator (Keiler et al. 1996). The nonstop mRNA transcript from this gene, hereafter referred to as λ-cI-N, has been shown to be a substrate for the trans-translation process (Keiler et al. 1996; Karzai et al. 1999; Sundermeier et al. 2005).

First, we examined the stability of λ-cI-N transcript in a wild-type (ssrA +) strain and a tmRNA-deficient (ssrA−) strain. Previous work has shown that although expression of the λ-cI-N construct in this wild-type background results in the production of a nonstop mRNA, the presence of tmRNA enables the cell to rescue stalled ribosomes and tag the encoded protein products for proteolysis (Keiler et al. 1996; Karzai et al. 1999). The tagged proteins are recognized and degraded by C-terminal-specific proteases (Gottesman et al. 1998; Herman et al. 1998). Using Northern blot analysis, with a probe specific to λ-cI-N, we found that in wild-type cells, the λ-cI-N transcript was expressed and rapidly degraded (Fig. 1A, upper panel), with a half-life of ∼3 min. In contrast, in ssrA− cells, the λ-cI-N nonstop mRNA was substantially stabilized and decayed at a reduced rate, with a half-life of ∼14 min (Fig. 1A, lower panel). Similarly, in ssrA− cells complemented with a plasmid harboring wild-type tmRNA (ptmRNAWT) we found that the steady-state level of λ-cI-N transcript was substantially reduced (Fig. 1B, lane 2). In ssrA − cells complemented with a control plasmid (pKW1) there was a marked increase in the steady-state level of the λ-cI-N reporter transcript (Fig. 1B, lane 1), suggesting that tmRNA was required to facilitate the rapid degradation of nonstop mRNAs.

FIGURE 1.
View larger version:
FIGURE 1.

An absence of tmRNA slows nonstop mRNA decay, and tmRNA ORF variants are similarly deficient in nonstop mRNA degradation. (A) Northern blot analysis of λ-cI-N reporter mRNA from wild-type and ssrA − cells over time. The decay of nonstop mRNA in ssrA − cells takes substantially longer than in wild-type E. coli. (B) Northern blot analysis of steady-state λ-cI-N reporter mRNA from ssrA − cells complemented with the pKW1 control plasmid, or plasmids harboring tmRNAWT, tmRNAHis6, or tmRNADD. (C) Bar graphs, representing the mean and standard deviation of three independent Northern blot assays, display the steady-state level of λ-cI-N reporter mRNA in the presence of tmRNAWT, tmRNAHis6, or tmRNADD, as compared to cells lacking tmRNA function. (D) The mRNA-like sequences and helix 5 regions of tmRNAWT, tmRNAHis6, and tmRNADD are shown to indicate (in bold letters) nucleotide sequence changes and compensating mutations.


Mutations in the nucleotide sequence of the tmRNA ORF alter nonstop mRNA degradation

In an attempt to further evaluate the fate of both the defective mRNA and its encoded protein product, we examined the effect of two widely used tmRNA open reading frame (ORF) variants (tmRNADD and tmRNAHis6) on the stability of the λ-cI-N nonstop reporter mRNA. These tmRNA mutants encode modified tmRNA tag sequence variants from the native ANDENYALAA to AANDENYALDD in the case of tmRNADD, and to ANDEHHHHHH in the case of tmRNAHis6. It should be noted here that expression of the nonstop transcripts in the presence of the tmRNADD and tmRNAHis6 variants has been shown to result in production of tagged proteins that are resistant to proteolytic degradation (Keiler et al. 1996; Gottesman et al. 1998; Karzai et al. 1999; Roche and Sauer 1999; Sundermeier et al. 2005). In each experiment, a probe specific to 16S rRNA was used to confirm that similar quantities of total RNA were loaded onto each lane. Surprisingly, we found that in cells where tmRNADD and tmRNAHis6 variants were expressed in place of wild-type tmRNA, there was substantial stabilization of the λ-cI-N nonstop mRNA, close to the reporter transcript levels in cells lacking tmRNA function (Fig. 1B,C, cf. lane 2 and lanes 1,3,4). Nucleotides in the distal segment of the tmRNA peptide-reading frame are predicted to be involved in formation of helix 5 (Fig. 1D; Williams and Bartel 1996); changes to these bases could impinge on the stability of this helix, and thus affect the activity of tmRNA. To ensure that the hydrogen-bonding integrity of the stem–loop structure was not affected, we had introduced compensating mutations in the helix 5 region of both tmRNA variants (tmRNAHis6 and tmRNADD) used in our experiments (Fig. 1D, insets). Despite the presence of the compensating mutations, we still observed substantial accumulation of the λ-cI-N reporter transcript with these two tmRNA variants. These results suggested that the mRNA-like segment of tmRNA might be involved in targeting nonstop mRNAs for degradation.

Nonstop mRNA decay defects are not due to stabilizing effects of the tag peptide

In the experiments described above, the deletion of the ssrA gene or certain modifications of the mRNA portion of the tmRNA by as little as 3 nucleotides (nt) in the case of tmRNADD, resulted in severe reduction in degradation of the λ-cI-N transcript. This prompted us to systematically evaluate the role of the mRNA-like segment of tmRNA in the degradation of the λ-cI-N nonstop mRNA. As noted above, several laboratories, including ours, have used the tmRNADD and tmRNAHis6 variants extensively to identify natural substrates of the tmRNA quality-control system and gain insights into the mechanistic details of the trans-translation process (Keiler et al. 1996; Karzai et al. 1999; Roche and Sauer 1999, 2001; Wiegert and Schumann 2001; Fujihara et al. 2002; Hayes et al. 2002a, b; Farrell et al. 2005; Sundermeier et al. 2005; Wower et al. 2005). One consequence of using these variants, and hence, the rationale for their use, is that the tmRNADD and tmRNAHis6 tag sequences impart a much greater degree of proteolytic stability to target proteins (Keiler et al. 1996; Roche and Sauer 1999, 2001; Okan et al. 2006). Therefore, one could postulate that the increased stability of the nonstop mRNA in our experiments was due to an indirect effect caused by stabilization of and/or interference by downstream target proteins.

To scrutinize this possibility, we made mutations in nucleotide sequences encoding the last two amino acids of the tmRNA tag sequence. These changes were designed to alter the nucleotide sequence while retaining the propensity of the tag sequence to promote proteolytic degradation of the tagged proteins (Table 1). For the initial set of mutants, we altered the last two amino acids of the degradation tag from AA to VV, VA, or AV. We deemed these residues ideal as proteases that degrade tmRNA-tagged proteins (ClpX/P, ClpA/P) are known to preferentially degrade substrates carrying small hydrophobic amino acids (V, A, and L) at their C termini (Parsell et al. 1990; Silber et al. 1992; Milla et al. 1993; Gottesman et al. 1998, and references therein). To evaluate the consequences of these mutations, we expressed the λ-cI-N gene in the presence of the tmRNAVV, tmRNAVA, and tmRNAAV variants and examined the stability of the nonstop transcript by Northern blot analysis (Fig. 2; Table 1). This evaluation showed the tmRNAVV, tmRNAVA, and tmRNAAV variants to have reduced nonstop mRNA degradation phenotypes (Fig. 2; Table 1). Additionally, the nonstop mRNA decay defects exhibited by these tmRNA variants were not restored by the introduction of compensating mutations in helix 5 (Table 1), suggesting that the nonstop mRNA degradation defects exhibited by these tmRNA variants were not due to changes in helix 5 stability. Furthermore, these data suggested that the nonstop mRNA decay defects were not due to the potential stabilization of target proteins, as proteins tagged with the tmRNAVV-, tmRNAVA-, and tmRNAAV-encoded peptides should be rapidly degraded by C-terminal specific proteases (see the section below). What is more, these findings suggested that the nonstop mRNA decay defects of the tmRNADD and tmRNAHis6 variants were also unlikely to be caused by stabilization of target proteins.

FIGURE 2.
View larger version:
FIGURE 2.

Comparison of tmRNAWT with tmRNAAV and tmRNAVV, tmRNA variants that should not affect tagged protein stability. (A) Northern blot analysis of steady-state λ-cI-N reporter mRNA and 16S rRNA from ssrA − cells complemented with plasmids harboring tmRNAWT, tmRNAAV, or tmRNAVV. These tmRNA mutants do not affect the proteolytic stability of the target tagged protein, yet they affect nonstop mRNA stability (see text for details). (B) Bar graphs, representing the mean and standard deviation of three independent Northern blot assays normalized to the level of 16S rRNA, display the steady-state level of λ-cI-N reporter mRNA in the presence of tmRNAWT, tmRNAAV, or tmRNAVV, as compared to cells lacking tmRNA function.


View this table:
TABLE 1

tmRNA peptide-reading frame variants used in this work


Nucleotides within the distal part of the tmRNA ORF are important for nonstop mRNA decay

An alternative explanation for our findings could be that nucleotide sequence elements in the peptide-reading frame of tmRNA are important in facilitating targeted nonstop mRNA degradation. To test this possibility, we constructed site-specific tmRNA mutants carrying conservative substitutions in the third nucleotide position of codons encoding the tmRNA peptide tag sequence (Table 1). This strategy permitted us to change the nucleotide sequence without altering the amino acid sequence of the native tag. Interestingly, Northern blot analysis showed significant reduction in degradation of the λ-cI-N transcript with conservative mutations in the last three codons of the tmRNA ORF (Fig. 3; Table 1). Since we were making minimal changes in this region, we reasoned that the phenotypes associated with the mutants were not due to helix 5 stability. In fact, nucleotides in positions 28–37 are part of the loop in the stem–loop structure and are not predicted to participate in any local H-bonding interactions (Fig. 3C). Furthermore, introduction of compensating mutations for nucleotides that were part of helix 5 did not substantially alter the decay phenotype. For instance, in Figure 3, compare lane 3 and lane 4 (tmRNAA27C-U30C and tmRNAA27C-U30C-U39G), or lane 5 and lane 10 (tmRNAA27U-U30A and tmRNAA27U-U30A-U39A), and the quantified data in Table 1. Mutations in the ultimate, penultimate, and anti-penultimate codons, encoding the last three amino acid residues of the peptide tag sequence, exhibited the greatest level of mRNA stabilization, showing the same relatively low level of nonstop mRNA decay as the tmRNAVV or tmRNADD mutants. Moreover, changing the native alanine codons to alternate alanine codons, either alone or in combination, and with or without compensating mutations, demonstrated a substantially deficient mRNA degradation phenotype (see the mutant listed above and in Table 1). We also made mutations in the second stop codon (U34-A35-A36) of the tmRNA ORF, changing the nucleotide sequence to C34-A35-A36. This mutation had a similar effect on message decay, impeding nonstop mRNA decay to the same extent as mutations in the ultimate and penultimate codons of the tmRNA ORF (Fig. 3; Table 1). Taken together, these findings are consistent with the hypothesis that part of the nucleotide sequence of the stem–loop in the mRNA region plays an important role in tmRNA-facilitated degradation of defective mRNAs.

FIGURE 3.
View larger version:
FIGURE 3.

Conservative mutations in the last three codons of the tmRNA ORF have adverse effects on degradation of nonstop mRNA. Northern blot analysis of the steady-state level of λ-cI-N reporter mRNA from ssrA − cells complemented with plasmids that harbor tmRNAWT or one of a series of tmRNA ORF variants, all of which encode for the wild-type peptide tag. The pKW1 vector, tmRNAWT, and tmRNAVV were included as controls. (B) Bar graphs represent the mean and standard deviation of three independent Northern blot assays, normalized to the level of 16S rRNA, and display the steady-state level of λ-cI-N reporter mRNA in the presence of each indicated tmRNA ORF variant. (C) The nucleotide sequence of the mRNA-like segment of tmRNA, including the conserved upstream nucleotides, the resume codon (nucleotides 1–3), the helix 5 stem–loop structure, and the stop codon. Green circles highlight those nucleotide changes that resulted in wild-type-like nonstop mRNA degradation phenotypes (see Fig. 4). Red circles denote those nucleotide changes that resulted in reduced degradation of nonstop mRNA. The nucleotides encoding the degradation tag are shown in lowercase. The peptide tag sequence is highlighted in gray.


Mutations in the proximal segment of the tmRNA ORF do not affect nonstop mRNA decay

Thus far, our analysis focused on the distal part of the tmRNA peptide-reading frame. Next, we examined whether the resume-codon proximal segment of the tmRNA peptide-coding sequence, prior to the stem–loop structure, also played a role in facilitating degradation of nonstop mRNA. To this end, we made tmRNA mutants that altered the third nucleotide of codons for each of the four amino acids after the alanine resume codon. This region of the tmRNA peptide tag has been previously shown to constitute the binding site for SspB, a specificity-enhancing factor for ClpX/P-mediated proteolytic degradation of tmRNA-tagged proteins (Levchenko et al. 2000, 2003; Flynn et al. 2001). Therefore, we chose to change the coding sequence in this region without altering the encoded amino acids, so as not to affect polypeptide tagging and targeted proteolysis (Table 1). Analysis of these tmRNA variants revealed that the λ-cI-N nonstop transcript was degraded to a similar extent as in cells harboring wild-type tmRNA (Fig. 4), suggesting that changes in this region have no effect on the degradation of nonstop mRNA. These data lend further support to the conclusion that the nucleotide sequences in the distal segment of the tmRNA ORF, encoding the last three amino acids of the tmRNA tag, play an important role in determining whether or not defective mRNAs are targeted for degradation.

FIGURE 4.
View larger version:
FIGURE 4.

Conservative mutations in the tmRNA ORF proximal region do not have an effect on degradation of nonstop mRNA. Northern blot analysis of the steady-state level of λ-cI-N reporter mRNA from ssrA − cells complemented with plasmids that harbor tmRNAWT or tmRNA ORF variants with third codon changes in the resume-codon proximal region. The pKW1 vector and tmRNAWT were included as controls. (B) Bar graphs represent the mean and standard deviation of three independent Northern blot assays normalized to the level of 16S rRNA, and display the steady-state level of λ-cI-N reporter mRNA in the presence of each tmRNA ORF proximal variant.


tmRNA mutants have wild-type levels of tmRNA

One possible explanation for the observed reduction in tmRNA-dependent decay of defective mRNA could be that mutations in the distal part of the tmRNA peptide-reading frame might affect the expression or intracellular levels of tmRNA. To address this possibility, we evaluated the amount of tmRNA in whole cell lysates by Northern blot analysis. tmRNA mutants that were deficient in degradation of the λ-cI-N RNA transcript were analyzed in this manner. Along with these mutants, we also tested the tmRNAHis6 variant, which has previously been shown to have wild-type levels of tmRNA expression (Moore and Sauer 2005). All of the mutants tested showed a strong band corresponding to tmRNA. Northern blot analysis of all the mutants generated in this work showed each to have wild-type levels of tmRNA (data not shown; Fig. 5). These data demonstrate that the point mutations introduced into the ssrA gene do not alter the level or stability of tmRNA in the cell, thus supporting a sequence-specific role for the distal segment of the tmRNA peptide-reading frame in nonstop mRNA decay.

FIGURE 5.
View larger version:
FIGURE 5.

Steady-state levels of tmRNA are not affected by mutations in the distal region of the tmRNA peptide-reading frame. Northern blot analysis of total tmRNA from ssrA − cells harboring the indicated tmRNA ORF mutants. The pKW1 vector and tmRNAWT were included as controls. Although there is some variation in the level of tmRNA in some mutants, the variations are not uniform and vary from experiment to experiment. Overall, there is no substantial difference between tmRNAΔU42, the least affected mutant, and tmRNAVV, one of the most severely affected mutants (see also Fig. 3).


tmRNA mutants are fully proficient in tagging target proteins

As noted earlier, the tmRNAHis6 and tmRNADD variants have been extensively used and shown to be proficient in tagging substrate proteins. It could be argued that the nonstop mRNA degradation defect exhibited by the tmRNA variants described here might be due to an overall reduction in trans-translation proficiency. In theory, a reduction in the tagging efficiency of tmRNA could stabilize the target nonstop mRNA. To scrutinize this possibility, we examined two representative tmRNA mutants, tmRNAΔU42 and tmRNAA24G, which both encode for the wild-type peptide tag but have distinguishing mRNA degradation phenotypes. The tmRNAΔU42 mutant facilitates degradation of the nonstop message to an extent only slightly lower than tmRNAWT, while the tmRNAA24G mutant accumulates significant amounts of nonstop mRNA (Fig. 3). Whole cell lysates were examined by Western blot analysis to determine whether the λ-CI-N protein was tagged by these tmRNA variants. As controls, tmRNAWT, tmRNAHis6, and tmRNADD were also tested. The tmRNAHis6 and tmRNADD variants of tmRNA have been previously shown to result in tagged proteins that are much more resistant to proteolytic degradation (Keiler et al. 1996; Roche and Sauer 1999; Karzai and Sauer 2001). Western blots were probed with anti-Flag monoclonal antibody, which recognizes the Flag-tag epitope near the C-terminal end of the λ-CI-N-Flag protein and is present in both tagged and untagged species (Fig. 6A). This analysis showed accumulation of an untagged lower-molecular-weight product in the sample lacking tmRNA (Fig. 6B). In the presence of tmRNAWT, the λ-CI-N-Flag protein was successfully tagged and targeted for proteolysis, deduced from the distinct lack of both tagged and untagged bands in Western blots. In cells expressing the tmRNAΔU42 and tmRNAA24G mutants, there was a similar absence of untagged and tagged λ-CI-N protein, most likely due to efficient tagging and subsequent proteolytic degradation of the tagged proteins (Fig. 6B).

FIGURE 6.
View larger version:
FIGURE 6.

tmRNA peptide-reading frame variants are fully proficient in peptide-tagging activity. (A) Schematic of the fate of the tagged l-CI-N protein in the protease proficient and deficient backgrounds. Western blot analysis of whole cell lysates probed with anti-Flag epitope monoclonal antibody. Experiments were carried out in the E. coli ssrA - (B) or ssrA- clpX-/P-/lon- (C) backgrounds with the pPW500 plasmid, which encodes the l-CI-N protein. The ssrA- cells were complemented with plasmids harboring the pKW1 vector, tmRNAWT, or one of the indicated tmRNA ORF variants.


Although the lack of tagged protein products in the presence of wild-type tmRNA, or the tmRNAΔU42 and tmRNAA24G mutants, is consistent with efficient tagging and rapid degradation via targeted proteolysis, the lack of a band does not necessarily prove that efficient tagging and degradation have occurred. The ClpX/P, ClpA/P, and Lon ATP-dependent proteases are known to play a key role in the recognition and degradation of tmRNA-tagged proteins (Levchenko et al. 2000, 2003; Farrell et al. 2005; J. Choy and W. Karzai, unpubl.). Therefore, to directly assess the tagging proficiency of these tmRNA variants, we analyzed the stability of the λ-CI-N protein in a ΔclpX/P/lon triple protease-deficient strain. In this background, tmRNA-tagged proteins should not be subject to rapid proteolysis and accumulate within the cell. In the ΔclpX/P/lon deficient strain, and in the absence of functional tmRNA, we observed the expected accumulation of the untagged λ-CI-N protein product (Fig. 6B, lane 1). In the presence of wild-type tmRNA, tmRNAΔU42, or tmRNAA24G, we observed substantial accumulation of tmRNA-tagged λ-CI-N protein (Fig. 6B), demonstrating that the defective mRNA was translated and the resulting protein product efficiently tagged. This result demonstrated directly that although the tmRNAA24G variant had a severe defect in its ability to facilitate degradation of nonstop mRNA, it was fully competent in supporting all known functions of tmRNA in trans-translation. Additionally, this result suggests that the role of tmRNA in nonstop mRNA decay is separable from its role in ribosome rescue and peptide tagging for directed degradation. Taken as a whole, these results suggest that the resume-codon distal region of the tmRNA peptide-reading frame plays a crucial role in facilitating nonstop mRNA decay, in a process that is distinct from its tRNA- and mRNA-like functions in trans-translation.

Previous SectionNext Section

DISCUSSION

In this study, we evaluated the role of nucleotide sequence within the tmRNA peptide-reading frame to identify elements that are involved in nonstop mRNA decay. Our results demonstrate that specific mutations in the distal part of the tmRNA ORF adversely affect only the disposal of nonstop mRNA, while leaving the tRNA- and mRNA-like functions entirely unaffected. Therefore, we conclude that sequence information contained within this segment of the tmRNA ORF (nucleotides 24–33 of the ORF) (Fig. 7A,B) is used to affect nonstop mRNA decay. tmRNA is thus able to facilitate a trans-translation-dependent disposal of defective mRNAs that cause ribosome stalling, complementing well the known ribosome-rescue and peptide-tagging functions of this versatile RNA.

FIGURE 7.
View larger version:
FIGURE 7.

Conservation of the tmRNA orf and its encoded peptide tag. Graphical sequence logo representations of nucleotide sequence conservation of (A) the tmRNA ORF and (B) the tag peptide amino acid sequence of the 69 available tmRNA sequences within the Enterobacteriaceae γ-protobacteria, generated using WebLogo (Crooks et al. 2004). The nucleotide sequence of the tmRNA peptide-reading frame and the encoded amino acid (above) and nucleotide numbers (below) are shown in panel A. Despite the observed variations in nucleotide sequence, the encoded peptide tag is highly conserved in the resume-codon distal region (B).


Although tmRNA has previously been implicated in degradation of the defective mRNAs (Yamamoto et al. 2003), specific sequence determinants required for this function had not been identified. In this study, we examined the role of the tmRNA peptide-reading frame and discovered that sequence determinants within this region of the RNA facilitate the degradation of defective mRNAs that promote ribosome stalling. We systematically examined and ruled out several possible explanations for the observed defect in nonstop mRNA decay instigated by ORF variants of tmRNA.

First, we examined and discounted the possibility that the increased stability of nonstop mRNA, with some tmRNA ORF variants, was an indirect effect of tagged-protein stabilization (Figs. 2–4; Table 1). Second, we examined and ruled out the possibility that the introduced mutations altered the overall expression or stability of tmRNA. We found that none of the mutants tested showed a significant difference in the quantity of tmRNA present in the cell (Fig. 5). Finally, we examined and ruled out a model in which mutations in the tmRNA ORF might result in a defect in trans-translation-mediated protein tagging and ribosome rescue, and thus impart a defect in nonstop mRNA decay. Evaluation of the tagging ability of variants that were defective in nonstop mRNA decay revealed that they were fully capable of tagging and targeting a reporter substrate for tagging and degradation (Fig. 6). Based on these findings, we propose that nucleotides in this segment of the ORF participate in intra- or intermolecular interactions with tmRNA elements, ribosomal components, or protein factors required for facilitating the removal of nonstop mRNAs.

What information is encoded within the amino acid sequence of the tmRNA-encoded peptide tag? The 10-amino-acid tmRNA-encoded peptide tag (A1N2D3E4N5Y6A7L8A9A10), which is cotranslationally appended to the C termini of target substrates, marks these proteins for degradation by the ClpX/P, ClpA/P, and FtsH ATP-dependent proteases (Gottesman et al. 1998; Herman et al. 1998). Tag sequence mutagenesis studies have shown that ClpX recognizes residues 8–10 (L8A9A10), whereas ClpA recognizes residues 7–9 (A7L8A9) of the tag peptide (Flynn et al. 2001). SspB, the ClpX specificity-enhancing factor, recognizes residues 1–3 and 7 (A1-N2-D3, and A7) of the peptide tag to heighten recognition and degradation by ClpX/P, and diminish binding and recognition by ClpAP, presumably through steric overlap with ClpA binding sites (Levchenko et al. 2000, 2003). Therefore, both the N and C termini of the tmRNA-encoded peptide tag contain sufficient information to facilitate recognition by several distinct processing and regulatory proteins.

What information is encoded within the nucleotide sequence of the tmRNA ORF? Engagement of the tmRNA reading frame by the rescued ribosome must a priori be different from engagement of all other mRNAs. For one, tmRNA function requires the ribosomal subunits to remain intact, in a 70S ribosome form with peptidyl-tRNA occupying the P site and still base-paired with the last codon of the defective mRNA. Additionally, tmRNA does not possess a Shine–Delgarno ribosome binding site or a standard AUG start codon. Instead, a highly conserved GCA codon is used as part of the resume signal required to reinitiate translation with tmRNA as its template (Fig. 7). How is the tmRNA peptide-reading frame established? The mechanistic details of this process have not yet been elucidated. What we do know is gleaned from mutagenesis studies of resume-codon proximal sequences (Williams et al. 1999). These studies show that the first 3 nt of the tmRNA peptide-reading frame (the GCA resume codon) and a conserved trinucleotide located 2 nt upstream of the resume codon are critical for establishing the reading frame. Mutations that alter these signals abolished both ribosome rescue and peptide-tagging activities of tmRNA. The tmRNA peptide-reading frame does contain a standard termination signal to permit normal recycling of ribosomes. This U-A-A/G termination signal is highly conserved throughout eubacteria. The tmRNA ORF thus possesses the necessary information content for both, resumption of translation and termination and recycling of ribosomes. It is thus reasonable to postulate that tmRNA possesses additional information content required for an all-encompassing trans-translation reaction, a reaction that solves all three problems caused by nonstop mRNAs. Our studies show that, indeed, the distal part of the tmRNA ORF contains the information required for facilitating degradation of nonstop mRNAs.

Does sequence conservation analysis provide any additional insights? Our mutagenesis studies were conducted with E. coli tmRNA. Therefore, we focused our sequence information content and conservation analysis on the tmRNA ORF nucleotide and tag peptide amino acid sequence of the 69 available tmRNA sequences within the Enterobacteriaceae γ-protobacteria. This analysis shows that nucleotides 23–33 of the ORF, encoding L8A9A10 and the termination codon, are very highly conserved (Fig. 7A). One might reason that this segment of the tmRNA ORF is conserved to ensure that the nature of encoded amino acid tag is unaltered for subsequent recognition by the ClpX/P protease. However, little or no variation is observed in the third nucleotide position of codons 8, 9, 10, and 11. Third nucleotide changes of codons in principle alter the nucleotide sequence while leaving the encoded amino acids unaltered. It is important to note that in this group of tmRNA peptide-reading frames, the ultimate and penultimate amino acids of the peptide tag, both alanines, are invariably encoded by the GCA-GCU codons. In contrast, codon 7 of the ORF, which also codes for an alanine, A7 of the tag, displays greater nucleotide sequence variation at the third nucleotide position, with GCU and GCA appearing at approximately equal distribution (Fig. 7A). This sequence variation is present despite the fact that the tag amino acid sequence is highly conserved in this region (Fig. 7B). If GCA and GCU both code for alanine, why is it that within this group of bacteria, codons 9 and 10 of the ORF are invariant while codon 7 has an equal distribution of GCA and GCU? Equally interesting is the fact that variations in codon 8 are confined to the first nucleotide, while the second and third nucleotides are highly conserved (Fig. 7B). The leucine encoded by codon 8 (L8) is invariant in this set of sequences (Fig. 7C), which agrees with the notion that the amino acid sequence of the peptide tag might be conserved to ensure efficient proteolysis of tagged proteins by the ClpX/P protease complex. We reason that the nucleotide sequence in this region, between nucleotides 23 and 33, is conserved for sequence-specific functional reasons. More explicitly, we propose that these nucleotides make sequence-specific contacts to facilitate degradation of defective mRNAs that cause unproductive ribosome stalling. Likewise, the sequence alignment analysis indicates that the first two codons of the ORF, the resume codon and the adjacent codon encoding N2 of the peptide tag, might also be conserved for functional reasons. These indications are in agreement with previous studies that established the significance of these codons for tmRNA ORF function, showing the resume codon to be essential for restarting translation on the tmRNA ORF (Williams et al. 1999), and asparagines 2 (N2) to be a key determinant for SspB binding to tagged protein (Levchenko et al. 2000).

Overall, although some general features of the tmRNA-encoded peptide tag are conserved throughout the bacterial kingdom, at the nucleotide and amino acid levels neither the ORF nucleotide sequence nor the tag-peptide amino acid sequence is absolutely conserved in all known tmRNA sequences. For instance, not all tmRNA-encoded tags are 10 amino acids in length, nor do they all start with alanine or end with alanine. If the resume GCA codon is essential for restarting translation and the second coded amino acid is essential for SspB binding (Levchenko et al. 2000), how can variation in this segment of the peptide-reading frame be explained? One possible rationale could be that the nucleotide sequence and its interacting partner might have coevolved, thus giving the appearance that the specific determinants are not important. This might likely be the case for the GCA resume codon. Alternatively, the interacting factor, for instance, SspB, might be quite distinct or not present in all bacterial species (Flynn et al. 2001). This latter possibility would permit the particular sequence to either acquire and tolerate the observed changes or interact with a new or related factor with distinct sequence specificity. Likewise, we reason that although the nucleotide sequence of the distal part of the tmRNA peptide-reading frame is conserved in large subclasses of bacteria (Fig. 7B), it may not be uniformly conserved throughout the bacterial kingdom. It is also not known whether the tmRNA-mediated degradation-of-nonstop-mRNAs function is present in all bacterial species. Therefore, the presence of this sequence variation might thus be due to coevolution with interacting partners, lack of interacting partners in some subclasses of bacteria, or acquisition of new interacting partners to serve the same ultimate function of promoting degradation of defective mRNAs.

What are the possible mechanisms for tmRNA-facilitated nonstop mRNA decay? The observed nonstop mRNA decay is SmpB dependent (Richards et al. 2006). This strict SmpB requirement suggests that the SmpB–tmRNA complex must be actively engaged with stalled ribosomes and in the early stages of the trans-translation process. Based on these facts, we propose the following possible mechanisms for tmRNA-mediated degradation of nonstop mRNAs: First, sequence elements within tmRNA, more specifically within the resume-codon distal part of the tmRNA ORF, could be involved in interactions with ribosomal elements that promote binding, rearrangement, or loading of an RNase(s) onto the soon-to-be-discarded defective mRNA. The RNase loading must occur prior to disengagement of the aberrant mRNA from the rescued ribosome or full engagement of the ribosome with the tmRNA reading frame. Second, this segment of tmRNA might be involved in interactions with other sequence elements in tmRNA or tmRNA-associated factors that then facilitate RNase binding or loading onto the defective mRNA. A third possibility could be that this segment of tmRNA might interact directly with an RNase and deliver it to the defective message. What RNase(s) might be involved in degradation of the nonstop mRNAs? In E. coli, the three primary 3′-to-5′ RNases, PNPase, RNase II, and RNase R, are likely candidates. RNase R, in particular, is an ideal candidate for this mRNA decay activity, as it was previously shown to be associated with the SmpB–tmRNA complex (Karzai and Sauer 2001). Indeed, in a related study we have demonstrated that of the three E. coli 3′-to-5′ exoribonucleases, RNase R has the unique distinction of being involved in tmRNA-dependent decay of aberrant mRNAs (Richards et al. 2006).

It is clear from these studies that tmRNA is required for the facilitated degradation of nonstop mRNAs, and that mutations in the distal part of the tmRNA ORF affect this facilitated mRNA decay. The mRNA decay phenotype also holds true for other well-studied and widely used tmRNA ORF variants, such as tmRNAHis6, which have been shown to be biologically active (Roche and Sauer 1999, 2001; Moore and Sauer 2005). In agreement with the findings reported herein, we observed that the tmRNAHis6-induced mRNA stabilization phenotype was present, in at least two previous studies (Hayes and Sauer 2003; Moore and Sauer 2005). These findings support our conclusion that sequence elements in the distal part of the tmRNA peptide-reading frame play a crucial role in facilitating nonstop mRNA decay.

Previous SectionNext Section

MATERIALS AND METHODS

Strains and plasmids

The Escherichia coli strains used in this work were grown in Luria–Bertani (LB) broth (10 g of tryptone, 5 g of yeast extract, and 5 g of NaCl per liter) or on LB plates (containing 1.5% agar). Ampicillin was used at 100 μg/mL, rifampicin at 150 μg/mL, and tetracycline at 24 μg/mL where required. All strains were constructed in the E. coli X90 ssrA∷cat background (hereafter referred to as ssrA−), which has been described previously (Keiler et al. 1996). The tmRNA expression plasmids, pKW11 encoding ssrA WT (Karzai et al. 1999), pKW23 encoding ssrA DD (Keiler et al. 1996), and pKW24 encoding ssrA His6 (Roche and Sauer 2001), have all been used in previous work. The pKW1control plasmid is derived from the pACYC184 vector, has a p15A origin of replication, and encodes a tetracycline-resistance marker. The pPW500 plasmid encoding the λ-cI-N-trpAt nonstop mRNA, under the control of an IPTG inducible promoter, has been described previously (Keiler et al. 1996).

Site-directed mutagenesis

Mutations in the ssrA coding sequence were introduced, using appropriate mutagenic primers and pKW11 as template by Quick-Change site-directed mutagenesis in accordance with the manufacturer's recommendations. Individual clones were analyzed by sequencing to confirm the presence of the intended mutations and verify the remaining ssrA sequence as unaffected by the procedure.

Induction λ-cI-N-trpAt nonstop mRNA

E. coli X90-ssrA− cells and a complementing plasmid-borne copy of ssrA were grown overnight at 30°C with ampicillin and tetracycline selection. Fresh cultures of ssrA− cells harboring pPW500 and a complementing plasmid-borne copy of ssrA were grown at 37°C in LB media to an OD600 of 0.5. The cultures were induced with IPTG at 1 mM final concentration. The control strain, lacking ssrA, contained the pPW500 and the pKW1 control plasmids. Cells were grown for 90 min post-induction, and 3 mL of cells were harvested by centrifugation. E. coli X90 cells are rifampicin resistant, therefore, experiments aimed at analyzing the decay rate of the reporter mRNA were performed using MG1655-ssrA− and the isogenic wild-type cells harboring pPW500, the λ-cI-N nonstop mRNA expression plasmid. In these experiments, the induction period with IPTG was 15 min. At this point, rifampicin was added to the culture, and 3 mL samples of the culture were taken at 0, 2, 4, 8, 16, and 32 min after the addition of rifampicin. Whole cell pellets were resuspended in equal volumes of sterile water, and OD600 was measured to enumerate cell numbers. Total RNA from equal numbers of cells (corresponding to equal OD600) was then extracted using Tri-Reagent (Molecular Research Center Inc). The quantity of recovered RNA was determined by absorbance at 260 nm.

Northern blot analysis

Equal amounts of RNA for each sample were resolved by electrophoresis on a denaturing 1.5% (v/v) formaldehyde agarose gel, and 15 μL RNA samples were mixed with 10 μL of denaturing solution (6.5 μL of formamide, 2 μL of formaldehyde, 1.5 μL of 10× MOPS buffer), heated at 65°C for 10 min, and 5 μL of sample dye was added (2% bromophenol blue, 10 mM EDTA, 50% glycerol) to each sample before loading on the gel. Resolved RNAs were transferred to Hybond N+ membranes (Amersham Biosciences) by overnight capillary transfer with 20× SSC buffer (175.3 g/L sodium chloride, 88.2 g/L sodium citrate). Membranes were cross-linked by UV and pre-hybridized at 45°C in hybridization solution (50% formamide, 5× SSC, 2.5× Dendhardt's solution, 1% SDS, 50 μg/mL salmon sperm DNA). Biotin-labeled probe was denatured for 10 min at 65°C and then added to the hybridization solution; hybridization was carried out overnight at 65°C. The biotin-labeled DNA probes were generated from purified PCR fragments 200–400 nt in length using EZ-Link Psoralen-Biotin reagent (Pierce). Unbound probe was washed from Northern blots with three low-stringency washes (0.2× SSC/0.1% SDS at room temperature) followed by three high-stringency washes (0.1× SSC/0.1% SDS at 65°C). Blots were developed using alkaline phosphatase-conjugated streptavidin and chemiluminescent substrate (Biotin Luminescent Detection Kit; Roche). Bands were quantified using ImageJ software (NIH, http://rsb.info.nih.gov/ij/).

Tagging and Western blot analysis

Cultures were induced as for the mRNA analysis experiments. After the induction period, 1 mL of culture was taken, and the cells were recovered by centrifugation. Induced whole cell pellets were resuspended in 2× SDS-sample buffer (20% glycerol, 4% SDS, 1.42 M β-mercaptoethanol, 125 mM Tris at pH 8.0, with bromophenol blue) according to the final OD600 of the culture (50 μL per 1.0 OD600). Samples were heated to ∼95°C for 10 min and vortexed. An equal volume of each sample was resolved by electrophoresis on a 15% Tris-tricine gel in duplicate. One gel was reserved for staining with Coomassie blue and the other blotted onto PVDF membrane (Millipore) using a semidry electro-blotter. Blots were probed with anti-Flag M2 mouse monoclonal antibody primary and rabbit anti-mouse peroxidase-conjugated secondary antibodies (Sigma). Detection was with ECL reagents (Amersham Biosciences). Bands were quantified using ImageJ software (NIH, http://rsb.info.nih.gov/ij/).

Previous SectionNext Section

ACKNOWLEDGMENTS

We thank Bob Haltiwanger and Tom Sundermeier for insightful comments on the manuscript. We also thank Latt Latt Aung for constructing the protease-deficient strains, and other members of the Karzai laboratory for helpful discussions and suggestions. We are grateful to Jorge Benach and members of The Center for Infectious Diseases for their continued support. This research was supported in part by grants (to A.W.K.) from the National Institutes of Health, and the Pew Scholars Program.

Previous SectionNext Section

Footnotes

  • Reprint requests to: A. Wali Karzai, Department of Biochemistry and Cell Biology, Center for Infectious Diseases, Stony Brook University, Stony Brook, NY 11794, USA; e-mail: akarzai{at}ms.cc.sunysb.edu; fax: (631) 632-8575.

  • Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.247706.

    • Received July 31, 2006.
    • Accepted September 18, 2006.
Previous Section
 

REFERENCES

  1. ↵
    1. Crooks, G.E.,
    2. Hon, G.,
    3. Chandonia, J.-M.,
    4. Brenner, S.E.
    (2004) WebLogo: A sequence logo generator. Genome Res. 14:1188–1190.
    Abstract/FREE Full Text
  2. ↵
    1. Dulebohn, D.P.,
    2. Cho, H.J.,
    3. Karzai, A.W.
    (2006) Role of conserved surface amino acids in binding of smpB protein to SsrA RNA. J. Biol. Chem. 281:28536–28545.
    Abstract/FREE Full Text
  3. ↵
    1. Farrell, C.M.,
    2. Grossman, A.D.,
    3. Sauer, R.T.
    (2005) Cytoplasmic degradation of ssrA-tagged proteins. Mol. Microbiol. 57:1750–1761.
    CrossRefMedlineISI
  4. ↵
    1. Flynn, J.M.,
    2. Levchenko, I.,
    3. Seidel, M.,
    4. Wickner, S.H.,
    5. Sauer, R.T.,
    6. Baker, T.A.
    (2001) Overlapping recognition determinants within the ssrA degradation tag allow modulation of proteolysis. Proc. Natl. Acad. Sci. 98:10584–10589.
    Abstract/FREE Full Text
  5. ↵
    1. Fujihara, A.,
    2. Tomatsu, H.,
    3. Inagaki, S.,
    4. Tadaki, T.,
    5. Ushida, C.,
    6. Himeno, H.,
    7. Muto, A.
    (2002) Detection of tmRNA-mediated trans-translation products in Bacillus subtilisGenes Cells 7:343–350.
    Abstract
  6. ↵
    1. Gottesman, S.,
    2. Roche, E.,
    3. Zhou, Y.,
    4. Sauer, R.T.
    (1998) The ClpXP and ClpAP proteases degrade proteins with carboxy-terminal peptide tails added by the SsrA-tagging system. Genes & Dev. 12:1338–1347.
    Abstract/FREE Full Text
  7. ↵
    1. Hayes, C.S.,
    2. Sauer, R.T.
    (2003) Cleavage of the A site mRNA codon during ribosome pausing provides a mechanism for translational quality control. Mol. Cell 12, 903.
  8. ↵
    1. Hayes, C.S.,
    2. Bose, B.,
    3. Sauer, R.T.
    (2002a) Proline residues at the C terminus of nascent chains induce SsrA tagging during translation termination. J. Biol. Chem. 277:33825–33832.
    Abstract/FREE Full Text
  9. ↵
    1. Hayes, C.S.,
    2. Bose, B.,
    3. Sauer, R.T.
    (2002b) Stop codons preceded by rare arginine codons are efficient determinants of SsrA tagging in Escherichia coliProc. Natl. Acad. Sci. 99:3440–3445.
    Abstract/FREE Full Text
  10. ↵
    1. Herman, C.,
    2. Thevenet, D.,
    3. Bouloc, P.,
    4. Walker, G.C.,
    5. D'Ari, R.
    (1998) Degradation of carboxy-terminal-tagged cytoplasmic proteins by the Escherichia coli protease HflB (FtsH) Genes & Dev. 12:1348–1355.
    Abstract/FREE Full Text
  11. ↵
    1. Karzai, A.W.,
    2. Sauer, R.T.
    (2001) Protein factors associated with the SsrA.SmpB tagging and ribosome rescue complex. Proc. Natl. Acad. Sci. 98:3040–3044.
    Abstract/FREE Full Text
  12. ↵
    1. Karzai, A.W.,
    2. Susskind, M.M.,
    3. Sauer, R.T.
    (1999) SmpB, a unique RNA-binding protein essential for the peptide-tagging activity of SsrA (tmRNA) EMBO J. 18:3793–3799.
    CrossRefMedlineISI
  13. ↵
    1. Karzai, A.W.,
    2. Roche, E.D.,
    3. Sauer, R.T.
    (2000) The SsrA-SmpB system for protein tagging, directed degradation and ribosome rescue. Nat. Struct. Biol. 7:449–455.
    CrossRefMedlineISI
  14. ↵
    1. Keiler, K.C.,
    2. Waller, P.R.,
    3. Sauer, R.T.
    (1996) Role of a peptide tagging system in degradation of proteins synthesized from damaged messenger RNA. Science 271:990–993.
    Abstract
  15. ↵
    1. Levchenko, I.,
    2. Seidel, M.,
    3. Sauer, R.T.,
    4. Baker, T.A.
    (2000) A specificity-enhancing factor for the ClpXP degradation machine. Science 289:2354–2356.
    Abstract/FREE Full Text
  16. ↵
    1. Levchenko, I.,
    2. Grant, R.A.,
    3. Wah, D.A.,
    4. Sauer, R.T.,
    5. Baker, T.A.
    (2003) Structure of a delivery protein for an AAA+ protease in complex with a peptide degradation tag. Mol. Cell 12:365–372.
    CrossRefMedline
  17. ↵
    1. Milla, M.E.,
    2. Brown, B.M.,
    3. Sauer, R.T.
    (1993) P22 Arc repressor: Enhanced expression of unstable mutants by addition of polar C-terminal sequences. Protein Sci. 2:2198–2205.
    Abstract
  18. ↵
    1. Moore, S.D.,
    2. Sauer, R.T.
    (2005) Ribosome rescue: tmRNA tagging activity and capacity in Escherichia coliMol. Microbiol. 58:456–466.
    CrossRefMedlineISI
  19. ↵
    1. Okan, N.A.,
    2. Bliska, J.B.,
    3. Karzai, A.W.
    (2006) A role for the SmpB–SsrA system in Yersinia pseudotuberculosis pathogenesis. PLoS Pathog. 2, e6.
  20. ↵
    1. Parsell, D.A.,
    2. Silber, K.R.,
    3. Sauer, R.T.
    (1990) Carboxy-terminal determinants of intracellular protein degradation. Genes & Dev. 4:277–286.
    Abstract/FREE Full Text
  21. ↵
    1. Richards, J.,
    2. Mehta, P.,
    3. Karzai, A.W.
    (2006) RNase R degrades non-stop mRNAs selectively in an SmpB-tmRNA-dependent manner. Mol. Microbiol. (in press).
  22. ↵
    1. Roche, E.D.,
    2. Sauer, R.T.
    (1999) SsrA-mediated peptide tagging caused by rare codons and tRNA scarcity. EMBO J. 18:4579–4589.
    CrossRefMedlineISI
  23. ↵
    1. Roche, E.D.,
    2. Sauer, R.T.
    (2001) Identification of endogenous SsrA-tagged proteins reveals tagging at positions corresponding to stop codons. J. Biol. Chem. 276:28509–28515.
    Abstract/FREE Full Text
  24. ↵
    1. Silber, K.R.,
    2. Keiler, K.C.,
    3. Sauer, R.T.
    (1992) Tsp: A tail-specific protease that selectively degrades proteins with nonpolar C termini. Proc. Natl. Acad. Sci. 89:295–299.
    Abstract/FREE Full Text
  25. ↵
    1. Sundermeier, T.R.,
    2. Dulebohn, D.P.,
    3. Cho, H.J.,
    4. Karzai, A.W.
    (2005) A previously uncharacterized role for small protein B (SmpB) in transfer messenger RNA-mediated trans-translation. Proc. Natl. Acad. Sci. 102:2316–2321.
    Abstract/FREE Full Text
  26. ↵
    1. Wiegert, T.,
    2. Schumann, W.
    (2001) SsrA-mediated tagging in Bacillus subtilisJ. Bacteriol. 183:3885–3889.
    Abstract/FREE Full Text
  27. ↵
    1. Williams, K.P.,
    2. Bartel, D.P.
    (1996) Phylogenetic analysis of tmRNA secondary structure. RNA 2:1306–1310.