Extended Data Fig. 5: Addition of substrates ATP and L-Phe transiently favours the A-conformation prior to stabilizing the T-conformation in a product-inhibited state.
From: Subdomain dynamics enable chemical chain reactions in non-ribosomal peptide synthetases
a, b, Positive correlations between aCC1 and aCC2 (a), and between hCC1 and hCC2 (b) in bulk-FRET steady-state distances under various substrate conditions showing a minimal population of the A-conformation (<10%) that has distinct CC1 and CC2 distances. Bulk-FRET distance errors were propagated from 1 s.d. of three measurements of bulk-FRET transfer efficiency, those from measured extinction coefficients, labelling efficiency and calculated Förster radius \({R}_{\circ }\). c, d, Steady-state bulk-FRET spectra of aCC2 (c) and hCC2 (d) in smFRET buffer, 1-mM ATP, 1-mM L-Phe and 1-mM ATP/L-Phe. Fluorescence spectra were normalized by total peak areas. Spectral data are presented as mean ± 1 s.d. from three repeats. e, Stopped-flow bulk-FRET measurements of aCC2 (150 nM, n = 4 for both substrates and n = 3 for other conditions) revealed a transient biasing of A-conformation after mixing with ATP/L-Phe. A single-exponential fit is shown in black. f, Adenylation-inactive aCC2-K517A (50 nM, n = 4 for both substrates and L-Phe-only and n = 3 for other conditions) shows no changes in the stopped-flow bulk-FRET measurements within 2 s in all four solutions. Data in e and f are shown as mean ± 1 s.d. from specified number of stopped-flow FRET traces. g, Close-up of the modelled active site of the GrsA A-___domain in A-conformation (PDB 1AMU with L-Phe). ATP and Mg2+ were taken from PDB 3C5E after aligning the proteins. h, smFRET distance histograms showing that aCC2-K517A is folded and capable of binding Phe-AMS. The solid lines are conformational probability density functions extracted by maximum-entropy deconvolution. The uncertainty was plotted in dash as 1 s.d. from 25 bootstrapped data sets. Inset: a native PAGE gel showing aCC2-K517A and hCC2-K517A binding to Phe-AMS. n = 2 for independently repeated native PAGE assays with similar mobility as shown in the (h) inset. Bottom: smFRET peak positions (uncertainty indicated by translucent vertical bars’ thickness), relative populations with vertical error bars, and normalized transition rates (horizontal arrows). Centres of peak positions and populations are shown as motional narrowing fits of nine time-resolved smFRET conformational distributions and errors are 1 s.d. of motional narrowing fits from 25 bootstrapped smFRET datasets.
