The Ribosome Derives the Energy to Translocate and Unwind mRNA from EF-G Binding

1 Supplementary Information The Ribosome Derives the Energy to Translocate and Unwind mRNA from EF-G Binding Hossein Amiri1,2,3,*,✉, William J. Van Patten1,2,4,*, Gillian Rexroad5,6, Varsha P. Desai1,2, Benjamen A. Sterwerf1,2, Laura Lancaster5,6, Harry F. Noller5,6,✉, Carlos Bustamante1,2,4,7,8,9,10,✉ 1Institute for Quantitative Biosciences-QB3, University of California, Berkeley, CA 94720, USA. 2Jason L. Choy Laboratory of Single-Molecule Biophysics, University of California, Berkeley, CA 94720, USA. 3Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720, USA. 4Biophysics Graduate Group, University of California, Berkeley, CA 94720, USA. 5Center for Molecular Biology of RNA, University of California, Santa Cruz, CA 95064, USA. 6Department of Molecular, Cell and Developmental Biology, University of California, Santa Cruz, CA 95064, USA. 7Department of Chemistry, University of California, Berkeley, CA 94720, USA. 8Department of Physics, University of California, Berkeley, CA 94720, USA. 9Howard Hughes Medical Institute, University of California, Berkeley, CA 94720, USA. 10Kavli Energy Nanoscience Institute, University of California, Berkeley, CA 94720, USA. * These authors contributed equally ✉ correspondence: [email protected] (HA), [email protected]du (HFN), [email protected] (CB) This file contains: Supplementary Figures 1 – 10 Supplementary Tables 1 – 5 2 3 Supp. Fig. 1| Hairpin unwinding fleezers trajectories for ribosomal translocation with WT EF-G. a, A full representative fleezers trajectory for a single ribosome translocating in the presence of WT EF-G and 1 mM GTP, showing successive unwinding steps (top plot) accompanied by Cy3-labeled EF-G binding events (bottom plot). The raw and 10-point smoothed data are shown in pale and dark colors, respectively, in each plot. Changes in fluorescence baseline are due to mixing of the translation mix in the optical tweezers chamber and opening and closing of the shunt containing translation mix. b, Three examples of individual productive WT EF-G binding events. c, A fleezers trajectory for a ribosome that does not show activity for over 10 minutes, demonstrating the dependence of unwinding on ribosome activity. 4 Supp Fig. 2| Possible effects of hindered GTP hydrolysis or Pi release on hairpin unwinding in fleezers trajectories. Lengthening of either τunwinding or τrelease is expected if hairpin unwinding or EF-G unbinding, respectively, is affected by the slowed rates in each case. Measured τunwinding and τrelease values for WT EF-G are shown for reference. 5 Supp. Fig. 3| Occasional lengthening of τrelease but not τunwinding with WT EF-G and a mixture of GTP and GTP analogs. a,b, Examples of productive WT EF-G binding events with a mixture of GTP and GDPNP (a) or a mixture of GTP and GTPγS (b) are shown. In each case, an unwinding step that is associated with a long τrelease (blue arrowhead) is followed by steps (pink arrowhead) with normally short binding events. The time window highlighted in yellow is expanded on the right for each example. c, Cumulative distribution plot of τrelease for GTP-only condition (magenta) and three GDPNP:GTP ratios 1:9 (pale blue), 1:2 (medium blue), and 3:1 (dark blue), with bi-exponential fits shown as dashed lines. The time axis is broken between 30 and 150 seconds due to a lack of data points. 6 Supp. Fig. 4| Lengthening of τrelease but not τunwinding with ribosomal mutant V67D L7/12 (slow Pi release) or with EF-G mutant H92A (slow GTP hydrolysis and Pi release). a, Examples of productive WT EF-G binding events in the presence of ribosomes containing mutant V67D L7/L12. b, Examples of productive hydrolysis mutant (H92A) EF-G in the presence of WT ribosomes. c, Summary of τunwinding and τrelease measurements for V67D L7/L12 and H92A EF-G, represented as mean ± standard error. Note that the bar for τrelease in the case of H92A exceeds the width of the page at the linear scaling used. 7 Supp. Fig. 5| Lengthening of τrelease but not τunwinding with EF-G mutant H92Q (slow Pi release). Two examples of productive H92Q EF-G binding events are shown. The time window highlighted in yellow is expanded on the right for each example to better resolve the short time between EF-G binding and the hairpin unwinding step. 8 Supp. Fig. 6| Insensitivity of unproductive EF-G binding events to GTP hydrolysis or Pi release. a, A fleezers trajectory showing multiple unproductive EF-G binding events (blue arrowheads) before a productive EF-G binding event accompanied by an unwinding step (pink arrowheads). b, A cumulative distribution plot for τunproductive under various conditions. The inset in the plot shows a magnified view for the shorter observed times. c, Summary of τunproductive measurements for WT EF-G, H92Q EF-G, the GDPNP mixture, and the GTPγS mixture, represented as mean ± standard error. The number of data points (N) is indicated for each condition. 9 16 Supp. Table 1| MLE fitting results for a double exponential distribution describing τunwinding. Errors are reported as two standard deviations (95%) confidence intervals of the fits. Fitting τunwinding to a single exponential distribution for any condition always yielded a poor fit. The last column shows sample mean and standard error of the mean (SEM). τunwinding 𝑓 1 𝑘1 𝑘2 N event N ribosome Mean ± SEM WT 39 ± 24% 6.7 ± 3.7 s-1 45 ± 25 s-1 67 24 70 ± 14 ms H92Q 24 ± 16% 3.3 ± 1.9 s-1 29 ± 11 s-1 97 29 100 ± 20 ms GDPNP+GTP (1.12+0.38 mM) 18 ± 16% 1.5 ± 1.2 s-1 30 ± 17 s-1 58 40 160 ± 60 ms GDPNP+GTP (0.5+1 mM) 64 ± 25% 14 ± 6 s-1 95 ± 75 s-1 63 27 50 ± 8 ms GDPNP+GTP (0.15+1.35 mM) 21 ± 20% 5 ± 4 s-1 44 ± 22 s-1 58 23 60 ± 15 ms GTPγS+GTP (0.5+1 mM) 22 ± 27% 11 ± 9 s-1 51 ± 24 s-1 69 20 35 ± 7 ms 17 Supp. Table 2| MLE fitting results for τrelease. Errors are reported as two standard deviations (95%) confidence intervals of the fits. A single-exponential fit was performed if sufficient. The last column shows sample mean and standard error of the mean (SEM). τrelease ksingle 𝑓 1 𝑘1 𝑘2 N event N ribosome Mean ± SEM WT - 45 ± 27% 1.4 ± 0.7 s-1 9 ± 6 s-1 67 24 374 ± 70 ms H92Q 0.5 ± 0.1 s-1 - - - 97 29 2 ± 0.2 s GDPNP+GTP (1.12+0.38 mM) - 45 ± 13% 0.06 ± 0.02 s-1 3.4 ± 1.5 s-1 58 40 8.2 ± 3.4 s GDPNP+GTP (0.5+1 mM) - 35 ± 15% 0.11 ± 0.05 s-1 2.8 ± 1.4 s-1 63 27 3.5 ± 0.9 s GDPNP+GTP (0.15+1.35 mM) - 11 ± 12% 0.4 ± 0.3 s-1 3.6 ± 1.1 s-1 58 23 504 ±100 ms GTPγS+GTP (0.5+1 mM) - 45 ± 16% 0.6 ± 0.2 s-1 10 ± 5 s-1 69 20 825 ± 160 ms 18 Supp. Table 3| MLE fitting results for τunproductive. Errors are reported as two standard deviations (95%) confidence intervals of the fits. A single-exponential fit was performed if sufficient. The last column shows sample mean and standard error of the mean (SEM). τunproductive ksingle 𝑓 1 𝑘1 𝑘2 N event N ribosome Mean ± SEM WT - 21 ± 18 1 ± 0.8 8.2 ± 3.4 63 14 300 ± 80 ms H92Q - 26 ± 56 1.0 ± 1.3 3.3 ± 2.3 29 11 500 ± 120 ms GDPNP+GTP (0.5+1 mM) 8 ± 3 - - - 28 11 125 ± 30 ms GTPγS+GTP (0.5+1 mM) - 24 ± 24 3.5 ± 3.4 36 ± 20 25 9 90 ± 30 ms 19 Supp. Table 4| Evaluation of statistical significance of differences (two-sided) in CDF distributions as compared to those of WT EF-G with GTP. p-values for two-sample Welch’s ttests, Kolmogorov-Smirnov tests (KS tests), and two-sample Mann-Whitney U Tests are shown in the table below. Note that for GTP analog conditions, these statistical tests were run on the entire distribution which includes a subset of normal GTP events. Welch p-value KS p-value MW p-value H92Q τunwinding 0.29 0.99 0.57 GTPγS+GTP τunwinding (0.5+1 mM) 0.02 0.10 0.04 GDPNP+GTP τunwinding (0.5+1 mM) 0.19 0.77 0.43 H92Q τrelease 1.1e-12 2.3e-15 1.3e-16 GTPγS+GTP τrelease (0.5+1 mM) 0.01 0.25 0.39 GDPNP+GTP τrelease (0.5+1 mM) 5.8e-4 1.9e-3 3.3e-5 H92Q τunproductive 0.17 0.01 0.001 GTPγS+GTP τunproductive (0.5+1 mM) 0.02 1.1e-3 1.5e-4 GDPNP+GTP τunproductive (0.5+1 mM) 0.06 0.09 0.2 20 Supp. Table 5| Counts of productive EF-G binding events and number of ribosomes for lowerthroughput single-molecule conditions. N event N ribosome EF-G H92A (hairpin assay) 5 4 L7 V67D (hairpin assay) 6 5 EF-G WT (assisting force assay) 7 6 EF-G + FA (assisting force assay) 8 6 EF-G H92Q (assisting force assay) 7 7 EF-G H92A (assisting force assay) 5 4 EF-G WT + GDPNP + GTP (assisting force assay) 16 13 EF-G WT + GTPγS + GTP (assisting force assay) 10 9