tmRNA

From Wikipedia, the free encyclopedia

tmRNA (also known as 10Sa RNA) stands for transfer-messenger-RNA. The gene encoding the tmRNA is ssrA. It is found in all bacterial genomes that have been sequenced, and is an important part of translation regulation. To remain stable, tmRNA associates with Small Protein B (SmpB). More at wikipedia..

Visualizaion tmRNA on the ribosome

Chemical probing of tmRNA structure in ribosomal complexes blocked at different stages of trans-translation allowed us to follow the structural changes starting from early steps of tmRNA passage, with the TLD in tmRNA at the ribosomal E-site, up to the termination step of trans-translation. Our data shows that neither pseudoknots nor stable helices (besides helix 5) are resolved in the process of trans-translation. Helix 5 is somehow stabilized in the complex with tmRNA-2 and 4. It is partly unwound in complex with tmRNA-5 and is completely resolved at the trans-translation termination stage in the complex with tmRNA-11.

The nucleotides in TLD with altered reactivity are the same for all steps of tmRNA passage through the ribosome. These data are in agreement with structural probing data (Barends et al. 2001; Ivanova et al. 2007) and the X-ray structure for SmpB interacting with TLD in solution (Gutmann et al. 2003; Bessho et al. 2007). Earlier we have found that one SmpB molecule is present in all studied complexes (Bugaeva et al. 2008). We propose that SmpB remains bound to TLD of tmRNA starting from the second codon at the ribosomal A-site, when TLD is at the ribosomal E-site, up to trans-translation termination.

In all the complexes we can observe an enhanced reactivity for the G base in the UGA stop-codon located at the ribosomal A-site. For tmRNA-4, 5 and 11 (elongation and termination of trans-translation) we find a conserved enhancement of the reactivity of the base located 6 nucleotide residues upstream of the stop-codon, indicating that the conformation of this mRNA part is similar for all later stages of trans-translation. Corresponding kinks in mRNA were found in an X-ray study (Yusupova et al. 2006). For all complexes an A-rich single-stranded region (79-86) is protected. The reactivity of G90 (first base in the resume codon) is increased for tmRNA-4 and 5.

We suggest that the region 79-86 represents a fixed binding site on the ribosome and the rest of the message can occupy any open ribosomal space. Such a fixed binding site would represent the interaction with the corresponding mRNA binding pocket on the 30S ribosomal subunit. This pocket is revealed by X-ray analysis of the region upstream of the P-site in an mRNA lacking a Shine-Dalgarno sequence (Yusupova et al. 2006).

The complex with tmRNA-2 is different from the others as judged by the protection pattern. This complex represents the initial step of trans-translation when TLD occupies the ribosomal E-site and the resume codon is at the ribosomal P-site. Since the single-stranded region A79-A86 also is protected in this complex we suggest that these protections reflect MLD entering the mRNA binding pocket.

A number of the bases whose reactivity was changed by complex formation are located in different single-stranded regions of tmRNA in all complexes, indicating conformational changes in tmRNA upon complex formation. In the case of helix 2 and pK1 some bases appear to be more strongly protected in the complex than in solution, suggesting that these structural elements interact with ribosomal components.

To visualize tmRNA on the ribosome at the elongation steps of trans-translation we have applied cryo-electron tomography. It should be mentioned that we made many attempts to use single particle reconstitution cryo-EM to study these complexes but were unable to detect clear tmRNA density suggesting tmRNA arch flexibility. Although cryo-EM tomography does not allow getting a high resolution structure it allows to visualize a single particle (Zhao et al. 2004a; Zhao et al. 2004b) and to follow particular tmRNA conformations in different ribosomal particles. Indeed we have found that the tmRNA arch can move from the shoulder to the top of the head of the 30S ribosomal subunit (Fig. 1).

Figure 1. Cry-electron tomography of tmRNA on the ribosome.

In order to understand how tmRNA is arranged on the ribosome at different steps of trans-translation and to elaborate a mechanism of template switching, we did modeling of the tmRNA structure in the ribosome. For modeling we used the published ribosome X-ray structure with a defined mRNA path (Yusupova et al. 2006). We have assumed that tmRNA ORF should occupy mRNA binding region for canonical mRNA. For tmRNA-2 the stop codon is at the A-site and the preceding codon together with tRNA should occupy the P-site. The TLD of tmRNA at the ribosomal E-site region keeps the structure of the acceptor arm of tRNA. The position of the L1 stalk was taken from (Harms et al. 2001). Only the above mentioned tmRNA elements were fixed. The program allowed the formation of the tmRNA structure with known secondary structure elements in the empty space in the ribosome or at the ribosome surface. The structure was optimized by energy minimization and manual corrections. The details of the models are available at http://rnp.genebee.msu.ru/bin/view/Projects/TmrnaEnglish.

Figure 2. Computer model of tmRNA-2 in the ribosome (click to download coordinates of tmRNA-2).

The structural model for tmRNA-2 inside the ribosome that encompasses the footprinting data presented here as well as available data (see Fig. 2). The SmpB protein occupies the site on TLD as known from its X-ray structure in complex with tmRNA fragment (Gutmann et al. 2003). The TLD is located at the ribosomal E-site. The protected A79-A86 loop is located at the entrance of the mRNA binding pocket; pK1 is nearby at the E-site of tRNA exit channel. The L1 stalk is moved from its position in the original X-ray structure (Yusupova et al. 2006) as suggested by cryo-EM data for conformational changes in the ribosome caused by exiting deacylated tRNA (Valle et al. 2003b; Harms et al. 2001). Only in this case there is enough space between the subunits at the L1 side of the ribosome in order to accommodate the TLD with SmpB and pK1 simultaneously. Helices 2a, 2b and 2c support the TLD location at the E-site, helix 2d creates a link between pK1 and the arch consisting of pK4, pK2 and pK3. The arch is surrounding the head of the 30S subunit starting from the shoulder; helix 5 is located at the entrance to the mRNA channel and can be easily unwound during subsequent steps of ORF translation without influencing the arch structure. The resume codon is at the P-site of the ribosome and the 2nd ORF codon is at the A site. One can see that all protected bases in the single-stranded regions are indeed involved in the interactions with the ribosome and the position of the bases with enhanced reactivity corresponds to possible distortions in the RNA chain.

Figure 3. Computer model of tmRNA-4 in the ribosome (click to download coordinates of tmRNA-2).

The same approach was applied to create a model for tmRNA-4 in the ribosome (Fig. 3). Again, there is a good agreement between the chemical probing data, cryo-electron tomography data and the proposed model. At this step of trans-translation the TLD-SmpB complex is moved out from the ribosome and is located at the platform side of the 30S subunit. It is not tightly fixed on the ribosome and can occupy any position such that it does not interfere with the deacylated tRNA that leaves the ribosome through the exit site. Helices 2a, 2b and 2c link TLD to pK4 (part of the arch) on one side and pK1 and the A79-A86 loop on the other side supported by helix 2d. This helix could be partially unwound, but hidden in the ribosome. It is thus protected from chemical modification even more strongly than in tmRNA being in solution. The protected A79-A86 loop is tightly bound in the mRNA binding pocket (Yusupova et al. 2006) that causes the conformational change of the first nucleotide of resume codon that becomes more exposed for modification. The arch that can be visualized by cryo-electron tomography consists of three pseudoknots. This arch is surrounding the head of the 30S subunit starting from the shoulder, as in the case of tmRNA-2. The model allows movement of the arch around the head of the 30S subunit. Helix 5 is located at the entrance to the mRNA channel. The 3rd and 4th codons of the tmRNA ORF are located at the P- and A-site positions, as determined by X-ray analysis (Yusupova et al. 2006), in the decoding center on the 30S subunit. Again, all protected bases in the single-stranded regions are indeed involved in the interactions with the ribosome and the position of the bases with enhanced reactivity corresponds to possible distortions in the RNA chain.

Figure 4. Computer model of tmRNA-A in the ribosome (click to download coordinates of tmRNA-A).

To model the initiation complex with the acceptor arm of TLD at the A-site and tRNA bound to the cellular mRNA at the A-site, we used the X-ray structure of the ribosome with tRNAs bound to A- and P-site. This is under the assumption that TLD should have the conformation of the acceptor arm of tRNA at the A-site and the mRNA part of tmRNA should be in the mRNA channel. After application of the structure formation approach described above we obtained the model presented in Fig. 4. One can see that there is no room for pK1 in the decoding area of the ribosome and pK1 is located at the intersubunit space on the L7/L12 side. Already in the pre-initiation complex the SmpB protein was shown to change its binding site on tmRNA (Kaur et al. 2006) known to be occupied in solution (Barends et al. 2001; Ivanova et al. 2007). In the initiation complex further movement of SmpB to the novel position should take place. The only space where it is possible to place SmpB is the loop formed by a single-stranded RNA region (A79-G87) and pK1; in our model we propose the interaction of SmpB with one of these elements (A79-G87). Recently, it was suggested that there is an alternative binding site for SmpB on the A79-A86 loop (Metzinger et al. 2008). The protection of some nucleotides in this region was also determined for the SmpB-tmRNA complex in solution (Konno et al. 2007). A number of mutations in this region either inactivate or affect trans-translation (Williams et al. 1999; Lee et al. 2001; Ivanov et al. 2002; Miller et al. 2008). Our model allows SmpB interactions with the ribosomal A-site as it was recently proposed (Nonin-Lecomte et al. 2009).

Figure 5. Computer model of tmRNA-P in the ribosome (click to download coordinates of tmRNA-P).

The model for the next step of trans-translation is presented in Fig. 5. At this stage EF-G dependent translocation has taken place and the TLD of tmRNA has moved to the P-site and the resume codon appears at the A-site. One can see that at this stage the only possibility for SmpB, pK1 and the A79-G87 loop to be in the ribosome is to occupy the space in the E-site area on the 30S subunit.

Thus we can speculate that the mechanism of the first EF-G dependent translocation should include the rotational movement of this element from one side of the intersubunit space to another. We propose that EF-G can catalyze such a movement with the help of its domain IV. This movement results in the precise positioning of the resume codon at the ribosomal A-site since the position of SmpB containing element is restricted by the size of the cavity in the ribosomal E-site. The A79-A86 loop has been shown to be essential for correct resume codon positioning (Williams et al. 1999; Lee et al. 2001). Moreover, a mutation at position 86 allows for peptide transfer, but not for further translocation (Konno et al. 2007). The model predicts that pK1 also should play an essential role in the formation of such element, being a structural constraint that determines the position of the SmpB containing element in the ribosome. According to literature, mutations in pK1 affect trans-translation (Tanner et al. 2006). Application of a selection procedure allowed the determination of a helical element that can substitute pK1. This structural element that can occupy the space of pK1 is a helix of a certain size and thermodynamic stability (Tanner et al. 2006). Studies of pK1 function by deletion of different elements also showed that pK1 is essential for supporting the stability of tmRNA structure (Wower et al. 2008). After translocation of the SmpB containing element the space at the L7/L12 side becomes free and EF-Tu can bring aminoacyl-tRNA to start translation of the tmRNA ORF. At the next step of trans-translation the TLD should move to the ribosomal E-site. This movement should be accompanied by the partial disruption of the SmpB containing element and stimulate the A79-A86 loop displacement towards the mRNA binding pocket, thus allowing SmpB switching to its position on TLD (model for tmRNA-2 in Fig. 2). At the further steps of trans-translation the A79-A86 loop occupies the mRNA binding pocket and this position is fixed by pK1. These elements remain bound to this site during the whole trans-translation process in agreement with our chemical probing data, thus stimulating the movement of the ORF to loop out. This makes it available for chemical modifications, as in the complex with tmRNA-11.

Our model considers one SmpB molecule at the stages when tmRNA has already entered the ribosome in agreement with (Shpanchenko et al. 2005, Bugaeva et al. 2008, Sundermeier et al. 2007). However two molecules of SmpB could bind the ribosome at the pre-initiation step of trans-translation (Kaur et al. 2006, Hallier et al. 2004) stimulating recognition of the stalled ribosome by tmRNA.

-- AndreyGolovin - 01 Dec 2008