Translational Control by Ribosome Pausing in Bacteria: How a Non-uniform Pace of Translation Affects Protein Production and Folding

Abstract

Protein homeostasis of bacterial cells is maintained by coordinated processes of protein production, folding, and degradation. Translational efficiency of a given mRNA depends on how often the ribosomes initiate synthesis of a new polypeptide and how quickly they read the coding sequence to produce a full-length protein. The pace of ribosomes along the mRNA is not uniform: periods of rapid synthesis are separated by pauses. Here, we summarize recent evidence on how ribosome pausing affects translational efficiency and protein folding. We discuss the factors that slow down translation elongation and affect the quality of the newly synthesized protein. Ribosome pausing emerges as important factor contributing to the regulatory programs that ensure the quality of the proteome and integrate the cellular and environmental cues into regulatory circuits of the cell.

Keywords: cotranslational folding; nascent peptide; prokaryotes; ribosome pausing; tRNA; translation; translation efficiency.

Figure 1

Factors contributing to translational efficiency (TE) and protein folding in bacterial translation. The schematic follows the mRNA direction from the 5' (left) to the 3' (right) end. The TE of an mRNA is largely determined at the translation initiation step when the 30S ribosomal subunit is recruited to the start codon on the mRNA. In some cases, next 30S subunit can be recruited to a stand-by site upstream of the initiation site. A rare-codon ramp of 10–15 A/U-rich codons at the beginning of the coding region can increase TE by disfavoring mRNA secondary structures at the start codon. During translation, rare codons, low-abundance aa-tRNAs, lack of tRNA modifications and the interactions of the nascent chain with the polypeptide exit tunnel of the ribosome may cause ribosome pausing. mRNA secondary structures can regulate ribosome occupancy at the upstream sequences. Some mRNA contexts, such as particular bi-codons, poly(Pro) and poly(Lys) sequences, cause rearrangements in the peptidyl transferase center (PTC) and promote formation of unusual structures in the A-site, thereby promoting ribosome stalling. Interactions between ribosomes in a polysome and of a leading ribosome with the RNA polymerase (RNAP) may provide yet another source of pausing.

Figure 2

Regulation of the A-site accessibility by unconventional mRNA secondary structures formed upon ribosome pausing. (A) Positions of the P- (dark blue) and A-site (light blue) tRNAs on the mRNA (dark red) in Escherichia coli ribosome (PDB: 7K00; Watson et al., 2020). The density of the ribosome is omitted for clarity. The A-site is occupied by tRNA. In panels (B–D), the A-site tRNA is absent and the P-site tRNA is used for alignment. (B) An mRNA element in the A-site of yeast ribosome stalled on CGA-CCG inhibitory bicodon (PDD: 6T4Q; Tesina et al., 2020). Inset: a close up of the secondary structure element. (C) A hairpin formed in the A-site of E. coli ribosome stalled on a take-off site of gene 60 mRNA of bacteriophage T4 prior to bypassing (PDB: 5NP6; Agirrezabala et al., 2017). Inset: a close up of the secondary structure element. (D) Mammalian ribosome stalled on a poly(A) sequence (PDB: 6SGC; Chandrasekaran et al., 2019). Inset: a close-up of the single stranded helix in the A-site. Residues A1825 and C1698 of 18S rRNA (green) stabilize helix formation by stacking.

Figure 3

Examples of how rare codons can affect protein folding and function. (A) Gamma-B crystallin (Buhr et al., 2016). Left panel: Crystal structure of bovine gamma-B crystallin (PDB: 4GCR). Indicated are the protein positions used to introduce FRET labels. Middle panel: Cotranslational folding of un-optimized (U) and harmonized (H) gamma-B crystallin monitored in real time. The translation times, indicated by the delay in the time courses, are 35 s for the H and 50 s for the U construct; the delay coincides with the emergence of the N-terminal domain of gamma-B crystallin from the exit tunnel of the ribosome. The folding times, derived from exponential fitting, are 39 s for the H and 59 s for the U construct, demonstrating that folding kinetics is affected by synonymous codon replacements. Right panel: Expression of U and H proteins in E. coli. Shown are the total protein (Total) for the U and H variants as well as soluble (S) and pellet (P) fractions for each construct. (B) Stability of an E. coli protein Sufl (Zhang et al., 2009). Comparison of the proteinase K sensitivity of the wild-type wt Sufl and SufI-Δ25–28, in which two rare Leu codons at positons 244 and 252 were substituted by abundant codons. (C) Stability of the E. coli chloramphenicol acetyltransferase (CAT; Walsh et al., 2020). The CAT activity was measured in a ClpXP degradation assay of ssrA-tagged wt CAT and a ssrA-tagged Shuf1-CAT variant. In Shuf1-CAT, the pattern of synonymous codons is changed locally, but the global codon usage was largely unchanged. (D) Activity of the circadian clock protein FRQ (Zhou et al., 2013) tracked via Luciferase (Luc) reporter assay. Right panel: Effect of codon optimization of the circadian rhythm. m-frq, rare codons were optimized; f-frq, all codons were optimized. Left panel: Comparing periodicity and expression levels of wt and optimized FRQ variants. (E) Analysis of 500 E. coli proteins determining the fraction of conserved, slowly translated, rare codon-enriched regions that account for the predicted intermediates in a cotranslational folding model (orange) vs. changing threshold value of p (Jacobs and Shakhnovich, 2017). Randomized control sequences are shown in blue. **p < 0.01.