Interaction of IF2 with the ribosomal GTPase-associated center during 70S initiation complex formation

Abstract

Addition of an Escherichia coli 50S subunit (50S(Cy5)) containing a Cy5-labeled L11 N-terminal domain (L11-NTD) within the GTPase-associated center (GAC) to an E. coli 30S initiation complex (30SIC(Cy3)) containing Cy3-labeled initiation factor 2 complexed with GTP leads to rapid development of a FRET signal during formation of the 70S initiation complex (70SIC). Initiation factor 2 (IF2) and elongation factor G (EF-G) induce similar changes in ribosome structure. Here we show that such similarities are maintained on a dynamic level as well. Thus, movement of IF2 toward L11-NTD after initial 70S ribosome formation follows GTP hydrolysis and precedes P(i) release, paralleling movement of EF-G following its binding to the ribosome [Seo, H., et al. (2006) Biochemistry 45, 2504-2514], and in both cases, the rate of such movement is slowed if GTP hydrolysis is prevented. The 30SIC(Cy3):50S(Cy5) FRET signal also provides a sensitive probe of the ability of initiation factor 3 to discriminate between a canonical and a noncanonical initiation codon during 70SIC formation. We employ Bacillus stearothermophilus IF2 as a substitute for E. coli IF2 to take advantage of the higher stability of the complexes it forms with E. coli ribosomes. While Bst-IF2 is fully functional in formation of E. coli 70SIC, relative reactivities toward dipeptide formation of 70SICs formed with the two IF2s suggest that the Bst-IF2.GDP complex is more difficult to displace from the GAC than the E. coli IF2.GDP complex.

Figure 1. Distance between L11 residue 38 and IF2 residue 378 in a 70S.IF2

GMPPCP fMet-tRNA^fMet.mRNA complex. According to a cryoelectron microscopy structure (EMD-1172, 11).

Figure 2. Measures of 70SIC formation and reactivity using different 30SICs

(A) Light scattering. (B) GTP hydrolysis; (C) Pi release; (D) fMet-tRNA^fMet(prf) fluorescence; (E) fMetPhe-tRNA^Phe formation. Experiments in (A), (C), and (D) were carried out by rapid mixing of the various 30SICs with wt-50S subunits in a stopped-flow spectrofluorometer. In all experiments except those in (E) with pre-formed 70SIC, 30S and 50S subunits were present in final concentrations of 0.3 M and 0.5 M, respectively. Preformed 70SIC in (E) was present at 0.3 M. Other final concentrations were: IF2: 0.15 M (A), (E); 0.45 M (B) – (D). GTP: 100 M (A), (C), (D); 36 M (B); 200 M (E). TC: 1.0 M (E).

Figure 3. Measures of 70SIC formation and reactivity using different 50S subunits

(A) GTP hydrolysis; (B) fMetPhe-tRNA^Phe and fMet-puromycin formation; (C) Poly (U)-dependent polyPhe synthesis. fMetPhe-tRNA^Phe formation was carried out by rapid mixing of 30SIC^Cy3 with 50S subunits and Phe-tRNA^PheEF-Tu.GTP ternary complex. Final concentrations were: 50S, 0.5 M; 30S, 0.3 M; IF2, 0.15 M. Other final concentrations were: GTP: 36 M (A); 200 M (B); puromycin, 2.5 mM (B). (C) see Materials and Methods. Results with wt-50S and 50S^−L11 parallel those reported earlier (40).

Figure 4. FRET between L11^Cy5 and Bst-IF2^Cy3 in the 70SIC complex

(A) After 70SIC formation. (B) During 70SIC formation. Excitation was at 540 nm. DA samples contain 30SIC^Cy3 and 50S^Cy5; D samples contain 30SIC^Cy3 and 50S^L11; A samples contain 30SIC^Bst and 50S^Cy5. In (A) 30SIC and 50S subunits were incubated at 37 °C for 15’ prior to the taking of fluorescence spectra. Final concentrations were: 30S, 0.3 M (hatched lines) or 0.6 M (smooth lines); Bst-IF2 or Bst-IF2^Cy3, 0.15 (hatched lines) or 0.3 M (smooth lines); 50S^L11 or 50S^Cy5, 0.14 M; GTP, 100 M. In (B) 30SICs were rapidly mixed with 50S subunits. The D and A samples were monitored at 567 nm and 670 nm, respectively. The DA samples were monitored at both wavelengths, as indicated. Final concentrations were: 30S, 0.30 M; IF2, 0.25 M; 50S, 0.18 M; GTP, 100 M.

Figure 5. Measures of 70SIC formation on combining 30SIC^Cy3 and 50S^Cy5: FRET, light scattering, and Pi formation

(A) Direct comparison of FRET (green trace, acceptor fluorescence – excitation 540 nm; monitoring 670 nm) and light scattering (blue trace – irradiation at 436 nm; monitoring via a 455 nm cutoff filter) changes during 70SIC formation. Both traces were determined for an identical solution having the following final concentrations: 30S, 0.3 µM; Bst-IF2^Cy3, 0.45 µM; 50S^Cy5, 0.60 µM. For ease of comparison, the changes in each value were normalized to the total change seen at the plateau for each measurement (~ 10 s). The ratio of normalized FRET change to normalized light scattering change is plotted in the inset. When fMet-tRNA^fMet was omitted, virtually no changes were seen in either FRET (red trace) or light scattering (yellow trace). (B) FRET changes (green, red, and blue traces) and Pi formation (orange trace). Final concentrations employed: green and orange traces: 30S, 0.3 µM; Bst-IF2^Cy3, 0.45 µM; 50S^Cy5, 0.60 µM; blue trace: 30S, 0.6 µM; Bst-IF2^Cy3 0.50 M; 50S^Cy5, 0.18 µM; red trace: 30S, 0.3 µM; Bst-IF2^Cy3 0.25 M; 50S^Cy5, 0.50 µM. FRET changes are normalized for the total change seen at the plateau for the green trace, as in (A). The Pi release is normalized for the total change seen at the plateau, achieved at ~ 5s. Final GTP concentration in (A) and (B), 100 M. All solid black lines are fits of the data to Scheme 1, with R² = 0.997 (R² is the square of the sample correlation coefficient between the outcomes and their predicted values). Attempts to fit the results in (A) and (B) to a two-step model resulted in significantly lower R² values. (C) Scheme 1, the minimal scheme accounting quantitatively for 70SIC formation in the presence of GTP.

Figure 6. Measures of 70SIC formation when GDPNP replaces GTP

(A) Light scattering increase during 70SIC formation (blue trace, GTP; green trace, GDPNP). Inset: extension of results to 10 s. (B) FRET efficiency and light scattering increases during 70SIC formation measured in the presence of GDPNP. FRET efficiency, blue and red traces in the presence or absence of fMet-tRNA^fMet, respectively. Light scattering, green and yellow traces in the presence or absence of fMet-tRNA^fMet, respectively. (C) FRET efficiency increases during 70SIC formation. Blue trace: higher 30SIC concentration in the presence of GTP; green and orange traces, higher 30SIC concentration in the presence of GDPNP in the presence or absence of fMet-tRNA^fMet, respectively; red trace, lower 30SIC concentration in the presence of GDPNP. Inset: Expanded time scale. Final concentrations in (A) and (B) were 30S, 0.3 µM; Bst-IF2^Cy3, 0.45 µM; 50S^Cy5, 0.60 µM. Final concentrations in (C) were 30S, 0.6 µM; Bst-IF2^Cy3, 0.5 µM; 50S^Cy5, 0.18 µM (blue and green traces –higher 30SIC) or 30S, 0.3 µM; Bst-IF2^Cy3, 0.25 µM; 50S^Cy5, 0.18 µM (red trace – lower 30SIC). Final GTP or GDPNP concentration in (A) – (C), 100 M. Solid black lines are fits of the GDPNP results to either a two-phase (light scattering) or three-phase (FRET) reaction. Fitted parameter values are: light scattering - k_app1, 35 ± 5 s⁻¹, k_app2, 1.10 ± 0.03 s⁻¹; FRET, higher 30SIC - k_app1, 20 ± 1 s⁻¹, k_app2, 2.2 ± 0.1 s⁻¹, k_app3, 0.19 ± 0.01 s⁻¹: relative FRET efficiency amplitudes; phase 1, 0.46 ± 0.05: phase 2, 0.51 ± 0.05: phase 3, 1.00. FRET, lower 30SIC - k_app1, 16 ± 1 s⁻¹, k_app2, 1.9 ± 0.1 s⁻¹, k_app3, 0.19 ± 0.01 s⁻¹: relative FRET efficiency amplitudes; phase 1, 0.64 ± 0.08: phase 2, 0.47 ± 0.06: phase 3, 1.00.

Figure 7. FRET monitoring of the fidelity function of IF3

The effect of IF3 on the rate and extent of FRET efficiency increase when the AUG initiation codon is replaced by AUU. 022AUG-mRNA, +IF3, blue trace; 022AUG-mRNA, -IF3, orange trace; 022AUU-mRNA, +IF3, green trace; 022AUU-mRNA, -IF3, red trace. Final concentrations were: IF2^Cy3, 0.15 M; 30S, 0.3 M; 50S^Cy5, 0.14 M; GTP, 100 M.