The Genetics of Ribosome Stalling and Rescue

The power of bacterial genetics

Chemists regularly synthesize and characterize one molecule at a time; biochemists make rational mutations to proteins and nucleic acids and study the effects of each single mutation. In contrast, genetic approaches create millions of solutions to a problem at once and sort through them in a high-throughput manner to identify molecules with the desired activity. Genetic methods can also identify key molecules involved in a process of interest when very little information is available about the system.

Several conditions must be satisfied to implement a genetic study. There must be a source of diversity-usually error prone PCR, chemical mutagenesis, or synthesis of degenerate DNA oligonucleotides to create mutant RNA or protein molecules. A second requirement is a method to identify molecules with the desired activity from among the many mutants. In vivo selections tie the life of the cell to the activity of an RNA or protein. Mutants with desired activity can be easily identified because they allow cells to survive. We use in vivo genetic selections in E. coli to study the genetics and biochemistry of ribosome function.

The trans-translation model

In bacteria, translation of mRNA into protein begins even before mRNA is fully transcribed from the DNA. There is no chance for quality control of mRNA as there is in eukaryotes. If an mRNA transcript is broken or incomplete, the ribosome stalls when it reaches the 3'-end; it cannot be freed without release factors that recognize stop codons. Damaged mRNAs that lack stop codons therefore trap ribosomes.

Ribosome stalling is also caused by certain peptide sequences that interact with the ribosome and prevent translocation or termination of translation. For example, the SecM sequence WIxxxxGIRAGP interacts with the ribosomal exit tunnel as the SecM protein is synthesized, causing stalling before the Pro residue (Nakatogawa and Ito, Cell 2002, 108, 629-36). The sequence Glu-Pro-Stop causes stalling during the termination process, perhaps by interfering with the function of release factors (Hayes et al., J. Biol. Chem. 2002, 277, 33825-32). When ribosomes stall in the middle of a messenger RNA, the mRNA is cleaved in the ribosome, resulting in the same situation described above for the broken or truncated mRNA case.

tmRNA enters the aminoacyl-tRNA (or A) site of stalled ribosomes as a tRNA mimic (see figure from Molecular Biology of the Cell). The ribosome leaves the broken transcript and resumes peptide synthesis from an open reading frame on tmRNA coding for the peptide sequence ANDENYALAA. In this way tmRNA serves as both a tRNA and a template coding for a protease-recognition tag. This accomplishes two important functions: the ribosome is released at a stop codon in the tmRNA and recycled and the aborted protein product is marked for destruction by proteases, preventing potentially harmful effects of mistranslated proteins. For a review on tmRNA and trans-translation see Karzai et al., Nat. Struct. Biol. 2000, 7, 449-55.

Outstanding questions and areas of current research

We have developed a powerful genetic selection that identifies E. coli cells in which tmRNA-mediated tagging of proteins is functional. This "positive" selection relies upon the complete synthesis of a necessary protein, the kanamycin resistance protein, upon stalling and rescue (see Tanner et al., J. Biol. Chem. 2006, 281, 10561-6). Only when stalled ribosomes are rescued by tmRNA, adding the last fifteen amino acids of the KanR sequence, is active KanR protein made, conveying cellular survival (see figure). Using this selection, we are addressing the following questions:

1. What sequences and structures in tmRNA allow the ribosome to restart translation in the correct place on the tmRNA template? What role does the protein SmpB play in positioning tmRNA correctly? We can rapidly identify active tmRNA and SmpB mutants using our selection, allowing us to ascertain the contribution of different substructures and sequences in these molecules.

2. What other molecular events cause ribosome stalling? Are there other peptide sequences (besides SecM) that induce stalling as they are formed inside the ribosome? Are there mRNA secondary structures that prevent the ribosome from moving along the mRNA? We are screening large collections of mRNA (or nascent peptide) mutants to identify new causes of stalling.

   We have also developed a genetic selection that identifies E. coli cells in which tmRNA tagging does not occur. This "negative" selection kills cells in which tmRNA tagging occurs, because stalling and rescue complete the synthesis of a toxic protein. Using this negative selection, we are addressing the following questions:

3. Why does the ribosome stall during termination at C-terminal proline residues? Termination of translation requires release factor proteins to bind to stop codons on the mRNA inside the ribosome. The release factors then enhance the cleavage of the ester bonding linking the nascent polypeptide and the attached tRNA. It is unclear how this catalysis occurs, if the ribosomal RNA or release factors contain this enzymatic activity. Our working hypothesis is that the release factors simply cause a conformational shift in the rRNA that then allows the rRNA to perform this reaction. Proline at the end of a polypeptide inhibits this reaction. We are identifying and characterizing mutants in the rRNA that do not stall during termination. These mutants may teach us why ribosome stalling occurs and more generally how termination normally works.

4. Is the ribosome an endonuclease? Does the ribosomal RNA catalyze cleavage of the mRNA when the ribosome stalls? tmRNA can only enter ribosomes in which the A-site is empty; the mRNA is cleaved in the A-site upon stalling. Deletion of known endonucleases has failed to identify the enzyme responsible for this activity. We are screening for mutations in the ribosomal RNA that disrupt this endonuclease activity and therefore prevent tmRNA function and tagging.

It is an exciting time to work on translation, as the high-resolution x-ray structure for the ribosome was solved in 2000. We now have the structure of the whole E. coli ribosome and can map our mutations that we isolate onto the structure (see Schuwirth et al., Science 2005, 310, 827-34 for the 3.5 angstrom structure of the 70S E. coli ribosome). This greatly aids us in figuring out how these mutations cause the observed phenotypes. The 50S subunit from the Schuwirth structure is shown.