Translational accuracy of a tethered ribosome

Abstract

Ribosomes are evolutionary conserved ribonucleoprotein complexes that function as two separate subunits in all kingdoms. During translation initiation, the two subunits assemble to form the mature ribosome, which is responsible for translating the messenger RNA. When the ribosome reaches a stop codon, release factors promote translation termination and peptide release, and recycling factors then dissociate the two subunits, ready for use in a new round of translation. A tethered ribosome, called Ribo-T, in which the two subunits are covalently linked to form a single entity, was recently described in Escherichia coli. A hybrid ribosomal RNA (rRNA) consisting of both the small and large subunit rRNA sequences was engineered. The ribosome with inseparable subunits generated in this way was shown to be functional and to sustain cell growth. Here, we investigated the translational properties of Ribo-T. We analyzed its behavior during amino acid misincorporation, -1 or +1 frameshifting, stop codon readthrough, and internal translation initiation. Our data indicate that covalent attachment of the two subunits modifies the properties of the ribosome, altering its ability to initiate and terminate translation correctly.

Figure 1.

Translation of lacZ with either the wt SD or the oSD sequences by the wild-type or oRiboT ribosomes. The constructs analyzed are shown on the left (A). Agarose gel electrophoresis of total RNA preparation from cells expressing either the wild-type ribosomes (1 (BL21 cells)) or the wild-type and oRiboT ribosomes (2 (BL21 + poRibo-T2 vector cells)). Results obtained with the RNAPlus kit are shown with, below, the quantification of the respective rRNA species (B). The β-galactosidase activities with arbitrary units are shown for the wt SD lacZ (SD, black) or the oSD lacZ (oSD, light gray) translated by wild-type (WT), or oRibo-T (oSD, hatched light gray) ribosomes. Translation of the oSD lacZ results (light gray and hatched light gray) are shown on a smaller scale on the right panel indicated by an arrow (C). Results are from at least six independent experiments, with the median value indicated

Figure 2.

Translational misreading errors. The ratio of luciferase activity to β-galactosidase activity is shown as arbitrary units for each construct carrying the indicated sequence of the codon 529 of the luciferase gene (wild-type AAA, or mutants TTT, CAA, AGA and AAT). The positive control (wild-type AAA) and the negative control (mutant TTT with no luciferase activity) are shown in (B), the CAA, AGA and AAT mutants in (C). The constructs analyzed are shown in (A), with the type of ribosomes present in the cells. Translations of the lacZ-luc fusion gene by wild-type (WT) (dark gray), or oRibo-T (oRibo-T) (light gray) ribosomes. Results are from at least six independent experiments.

Figure 3.

The -1 frameshifting sites. For each sequence the slippery sequence and the SD are underlined. Two frameshifting sites come from IS elements, and one from the cellular dnaX gene. IS911, dnaX and IS3 frameshifting sites use a single or double stem loop or a pseudoknot respectively as stimulatory elements.

Figure 4.

Translational efficiencies during -1 frameshifting. The constructs analyzed are shown on the left, with the type of ribosomes present in the cells, and in-frame controls in (C). The percent frameshifting is shown for the IS911, IS3 and dnaX sequences translated by wild-type (WT) (dark gray) and oRibo-T (light gray) ribosomes (A); and for the dnaX (dark gray) and dnaXoSD (light gray) sequences translated by oRibo-T or wild-type ribosomes (B). Results are from at least six independent experiments.

Figure 5.

prfB +1 frameshifting. Sequence of the E. coli frameshifting site (A). Translational efficiencies for prfB frameshifting, for wild-type ribosomes (dark gray) and oRibo-T ribosomes (light gray) (B); or for the prfB (dark gray) and prfBoSD (light gray) sequences translated by oRibo-T or wild-type ribosomes (C). The constructs analyzed are shown on the left, with the type of ribosomes present in the cells, and in-frame controls in (D). Results are from at least six independent experiments.

Figure 6.

Internal initiation of wild-type ribosomes. The constructs analyzed are shown on the left, with the type of ribosomes present in the cells. The ratio of luciferase activity to β-galactosidase activity measured for each construct is shown in the right box plot, as arbitrary units. Translations of the lacZ gene by wild-type ribosomes (WT), and of the luc gene with either a wild-type SD (wt SD) (dark gray) or no SD (black) by wild-type ribosomes (A). Translations of lacZ by wild-type ribosomes, and of luc with either no SD (black) or a oSD (light gray) by wild-type ribosomes (1), or a oSD (hatched) by wild-type and oRibo-T ribosomes (2) (B). Translations of lacZ by oRibo-T ribosomes and of luc with either no SD (black) or a oSD (light gray) by wild-type and oRibo-T ribosomes (C), or by oRibo-T ribosomes with (light gray) or without (black) a stem-loop structure upstream from luc (D). The SmaI site specific of the sequence without a stem−loop structure is indicated. Results are from at least six independent experiments.

Figure 7.

Translational readthrough efficiencies. The percent readthrough is shown for the three stop codons encountered by the wild-type (WT) (dark gray) and oRibo-T (light gray) ribosomes. The constructs analyzed are shown on the left, with the type of ribosome present in the cells. Results are from at least six independent experiments.