Fig 1.
Representative diagrams of plasmids used in our experiments.
Diagrams of plasmids (A) and (B) are shown. Initiation codon and termination codons are shown in capital letter and highlighted for clarity. AUG is the initiation codon for upstream open reading frame (u-ORF) and AUG is that for downstream open reading frame (d-ORF). The initiation codon for d-ORF is overlapped with the termination codon for u-ORF ((A) UAA and (B) UGA). u-ORF and d-ORF are boxed. Nucleotide sequence 5’-AGGA-3’ shown in italic represents a Shine-Dalgarno (SD) sequence of u-ORF. LacZ represents the reporter lacZ(Δ27) gene for β-galactosidase. n in the bracket represents a nucleotide and numbers next to the bracket represent the number of nucleotides in the bracket. The nucleotide sequence labelled (n)88 is referred to [39,43] for details.
Fig 2.
Downstream reading of translationally coupled ORFs is dependent on the presence of u-ORF and the initiation triplet of d-ORF.
E. coli LJ14 cells harboring plasmids were grown at 28°C to a cell density of ~0.13 as described in Experimental procedures and IPTG (final 2 mM) was added to the culture for induction of the reporter lacZ(Δ27) gene. The culture was incubated for another two hours. At the indicated times after the addition of IPTG, an aliquot was taken and the β-galactosidase activity was measured. The junction sequences and presence (+) or absence (−) of u-ORF are shown above each panel. The nucleotide sequences around the junction sequence are shown at the bottom. Rep represents the reporter lacZ(Δ27) gene. Dotted lines (……) in (b) through (e) represent the same nucleotide sequences with that of (a) in the corresponding regions. Panels (a) and (c) represent the β-galactosidase synthesis through translational coupling. (b) and (d) are identical to (a) and (c), respectively, with the exception that AUG of u-ORF was changed to auu. (e) is identical with (c) with the exception that (e) has aUAA instead of AUGA. The reporter lacZ gene is in frame with AUG of UAAUG, AUGA and aUA of aUAA.
Fig 3.
With heat inactivation of RRF, ribosomes remain on the mRNA and translate the downstream region in all three frames.
Numbers in square brackets shown above each panel represent the reading frame of d-ORF counting from AUG or Aua of UAAUG or UAAua, respectively, shown on the right side of panels. Panel (a) (a control plasmid) has no junction sequence and represents the u-ORF reading. The scale of the β-galactosidase activity in panel (a) is four times larger than that of (b) through (g). Panel (b) has the junction sequence UAAUG and represents the β-galactosidase synthesis through the coupling mechanism. The reporter lacZ(Δ27) gene of (b) is in zero frame (in frame with AUG of UAAUG). Panels (c) and (d) represent the cases in which the reporter gene is in +1 (c) and +2 (d) frame, respectively, counting from AUG of UAAUG. Panels (e), (f) and (g) have the sequence UAAua instead of UAAUG. Dotted lines (……) represent the same nucleotide sequences with that of (b) in the corresponding regions.
Table 1.
Heat inactivation of RRF makes more u-ribosomes read downstream.
Fig 4.
Both high culture temperature and loss of RRF function are required for thermal frameshifts to all frames.
LJ14 cells grown at 28°C were divided into 5 parts and further incubated at various temperatures. Numbers in square brackets shown on the right side of panels represent the reading frame of d-ORF counting from AUG of the junction sequence UAAUG. In panels (a)—(e), AUG of UAAUG is in frame with the reporter lacZ(Δ27) gene. In (f)—(j) and (k)—(o), AUG of UAAUG is out of frame with the reporter gene. The nucleotide sequences labelled (n)n are 5’-gguctcaaagcaaaacacaaggaaaacctcgcgaaauacgua-3’ (panels (a)–(e)), 5’-gguctcaaagcaaaacacaaggaaaacctcgcgauacgua-3’ ((f)–(j)), and 5’-gguctcaaagcaaaacacaaggaaaacctcgcgaauacgua-3’ ((k)–(o)). At 39°C, the temperature is high enough to inactivate RRF and cause thermal frameshift, which is shown by reading at all three frames (panels (e), (j) and (o)).
Fig 5.
The distance between AUG and UAA of the junction sequence influences downstream reading only in the presence of RRF.
Nucleotide sequence and the distance (D and d) between AUG and UAA at the junction regions are indicated above each panel. AUG is the initiation codon of d-ORF and in frame with the reporter lacZ(Δ27) gene. UAA is the termination codon of u-ORF. D represents distance in nucleotides when AUG is upstream of UAA while d represents distance when AUG is downstream of UAA. The minus sign (−) in panel (d) indicates that UAA and AUG are overlapped by one nucleotide (d = −1). Cases where AUG is located upstream of UAA with distance 3 and 0 nucleotides (D = 3 and 0) are not shown because AUG is in frame with downstream UAA.
Fig 6.
The smaller the size of u-ORF, the less the d-ORF reading.
Numbers of codons in u-ORF are shown above each panel. SD represents the SD sequence (5’-AGGA-3’). AUG of the junction sequence UAAUG is in frame with the reporter lacZ(Δ27) gene. n in the bracket represents a nucleotide and numbers next to the bracket show the number of nucleotides in the bracket. The nucleotide sequences around the junction sequence are shown at the bottom. Dotted lines (……) in (b) through (e) represent the same nucleotide sequences with that of (a) in the corresponding regions. Panel (a) has no junction sequence and represents the u-ORF reading. The nucleotide sequences labelled (n)n are 5’-aca-3’, 5’-accauc-3’, 5’-accaugauu-3’, 5’-accaugauuacc-3’, and 5’-accaugauuacgaau-3’ for panel (b), (c), (d), (e) and (f), respectively. Dotted lines (……) represent the same nucleotide sequences with that of panel (a) in the corresponding regions.
Fig 7.
The relative positions of the initiation and termination codons in relation to the position of SD influences the d-ORF reading.
Numbers of codons in u-ORF are shown above each panel. SD represents the SD sequence (5’-AGGA-3’). UAA and UGA are the termination codons of u-ORF. AUG is the initiation codon of d-ORF and in frame with the reporter lacZ(Δ27) gene. Dotted lines (……) represent the same nucleotide sequences with that of the plasmid of panel (a) in the corresponding regions. Panel (a) has no junction sequence representing the u-ORF reading. In panel (b), AUG is downstream of UAA. In (c), AUG is upstream of UGA. In (d), termination codon UGA of u-ORF is 94 nucleotides far from AUG and exists within the reporter gene.
Fig 8.
027mRNA sequence and translated peptides.
Two ways of downstream reading that depend on the presence or absence of RRF are shown. u-ORF is in italic. SD and the junction sequence are in bold. Amino acid residues corresponding to the triplets of the reading frames are shown in single capital letters.
Fig 9.
u-ORF translation is stimulated by RRF.
u-ORF reading was measured by incorporation of [3H]-Phenylalanine coded for by UUU and UUC codons into hot TCA insoluble materials. Reaction mixtures were as described in Materials and Methods. The concentration of Mg2+ was 4 mM.
Fig 10.
RRF stimulates translational coupling.
A RRF stimulates downstream reading in frame with AUG. Downstream reading was measured by incorporation of [14C]-Leucine coded for by CUG and CUA codons in frame with AUG. Reaction mixture was identical to that of Fig 9 except for labeled amino acid (leucine). Mg2+ concentration was 4 mM. B At high Mg2+ concentration, downstream reading in frame with AUG is more dependent on RRF. The reaction mixture was identical to that of Fig 10A except that Mg2+ concentration was 7 mM to observe ribosome releasing activity of RRF more clearly. Note that requirement for RRF is clearer under the condition (7 mM Mg2+) where non-enzymatic release of ribosomes from mRNA is harder than that in 4 mM Mg2+.
Fig 11.
RRF prevents wrong downstream reading in frame with UAA; RRF releases ribosomes from UAA of UAAUG.
Downstream reading in frame with UAA of the junction sequence was measured by [14C]-Valine incorporation coded for by triplet GUC which was in frame with UAA. The components of the reaction mixture were identical to that of Fig 9 except for labeled amino acid (valine).
Fig 12.
Complete ribosome splitting is not needed for translational coupling.
A ribo-T is not split while wild-type 70S ribosomes are split into 30S and 50S subunits in the presence of RRF. Upper figures show wild-type ribosome sedimentation and lower figures show that of ribo-T as described in Experimental procedures. Mg2+ concentration was 4 mM. B Complete splitting of ribosome by RRF is unnecessary for translational coupling; ribo-T functions in a similar manner to wild-type ribosomes. Incorporation of [14C]-Leucine coded for by CUG and CUA codons in frame with AUG was measured as in Fig 10B. The reaction mixture contained 7 mM Mg2+ where RRF requirement was more distinctly observed.
Fig 13.
RRF’s function in translational coupling without SD.
Complex (a); Post Termination Complex (PoTC) at the junction sequence of translationally coupled ORFs. The junction sequence UAAUG is shown in bold. Complex (b); RRF is bound on the ribosome at A/P-site. Complex (c); Completed 70S initiation complex. Process from (b) to (c); The ribosome and deacylated tRNA are released from the mRNA by RRF and EF-G action. The ribosome may or may not split into subunits. Both the 70S ribosomes and the ribosome subunits that engaged in the reading of the u-ORF may participate in downstream reading. The released 70S ribosomes or the dissociated ribosome subunits bind to the adjacent AUG on the same mRNA with the help of fMet-tRNA.
Fig 14.
Possible mRNA secondary structures around the junction sequence at the non-permissive temperature of tsRRF.
Stem and loop structures were constructed according to Freier et al. [56]. The SD sequence (5’-AGGA-3’), initiation and termination codons are represented by capital letters. Termination codons of u-ORF are boxed for clarity. Bars between nucleotides represent Watson-Crick type base-pairing.