Fig. 4: Increased RET to LOV2 and sensitisation can be achieved by linker optimisation.

a Alignments of XFP-optoNES variants compared with the original mCherry-fused LEXY⁹, showing the end of the fluorescent protein βbarrel and intervening sequences. AsLOV2 starts at leucine 404 or trimmed to 408 to increase RET; b The trimmed constructs are typically constitutively cytoplasmic because disruption of A’α helix by deleting LOV2 residues 404–407 renders the constructs functionally active as expected (the dC11 construct is shown; scalebar 20 µm, n = 3 wells in ×10 magnification, 1 example imaged at ×40); c The C7-C11 deletion series of mTq2 fused to 408LOV2-NES21 were subjected to kinetic RET analysis, revealing characteristic dequench-relaxation cycles (mean, n = 3; SEM shown in Supplementary Fig. 11); d Analysis of the dequench-relaxation as in Fig. 3 reveals that deletion of linker can almost double maximal dequench (at 14 µmol m⁻²) though relaxation rate is also affected. The dC7 construct has least deletion but greatest dequench, suggesting dipole angles or other parameters reduce RET in other constructs though mTq2-LOV2 distance in primary sequence is reduced. Best-fit parameters—maximum dequench, apparent sensitivity, relaxation time and ED50, dequench limit and sensitivity are compared in Table 1A. e Activation sensitivity of the mTurquoise-optoNES constructs from Table 1A correlates with dequench limit (equivalent to effective FRET efficiency; statistical analysis in Supplementary Table 2, best-fit ± S.E., n = 3 wells). f Best-fit relaxation rates obtained from the in situ data (green bars; Table 1A) are shown together with the corresponding values derived from measurements performed on lysates of cells expressing the same constructs (dC7, dC10, dC11 and the parental optoNES) at 25 and 37 °C (blue and red bars). This shows that temperature has a strong influence on relaxation rates of these constructs but, when the sequences used are identical, there is no significant difference between relaxation in intact cells and the cell lysates in any case. Mean ± standard error is shown (n = 8 cell-free, n = 6 and 3 for in-cell wild-type and variants respectively). *, **, *** denote P < 0.05, 0.01, and 0.0001, respectively (25 °C compared to either of the 37 °C values) by Tukey post-hoc test after two-way ANOVA. n.s. not significant. Source data are provided as a Source Data file.