Conformational features of CytR WT and mutants. (A) Molecular structure of polyP. (B) Amino acid sequence of CytR. Blue and red boxes indicate the positions of alanine and proline mutations in DM and P33A variants, respectively. (C) Disordered propensity of CytR predicted using IUPred334 with gray and black curves showing long and short disorder predictions, respectively. (D) Far-UV CD spectra of WT (green), DM (blue), and P33A (red) at 298 K in mean residue ellipticity (MRE) units of deg. cm2 dmol−1. (F) Thermal denaturation curves of CytR variants from far-UV CD experiments monitored at 222 nm and reported in MRE units. (F) Stokes Radius following the color code in panel E. (G) Electrostatic potential map35 of CytR in its folded conformation displaying a large positive electrostatic potential. (H) The hypothesis tested in the current work. WT (left), DM (middle), and P33A (right) ensembles could potentially form condensates or aggregates in the presence of polyP or DNA, and which could also display differential time-dependent properties. U, PF, and F represent unfolded, partially folded and folded conformations, respectively.

CytR undergoes phase separation with polyP in vitro. (A) Phase diagram illustrating the phase separation of WT at different protein and polyP concentrations. Empty circles represent no phase separation, and filled circles represent the extent of phase separation following the color bar which indicates the OD (turbidity) at 350 nm. (B, C) Representative DIC (B) and fluorescence (C) microscopy images of WT showing condensate formation in the presence of polyP. (D) Fluorescence microscopy images of NHS-rhodamine labeled WT at different protein concentrations in the presence of 22 µM PolyP. The scale bar is 10 µm. (E) Ionic strength dependence of turbidity at fixed WT (90 µM) and polyP (22 µM) concentrations. The error bar indicates the spread from experimental replicates. (F, G) DIC (left) and Fluorescence (right) microscopy images of NHS-rhodamine labeled DM (F) and P33A (G) in the presence of polyP. The scale bar is 10 µm. (H) Representative fluorescence microscopy images of condensates during FRAP on WT (left) and DM (right) immediately after polyP addition. (I) FRAP recovery curves at 0 h for DM (blue) and WT (green). The data represents an average of five experiments and the errors (shaded areas) are smaller than the size of the circles. (J) Time-lapse fluorescence images showing a dripping event for the WT (left column) and fission-fusion events for DM (right column).

Time-dependence of polyP-induced assemblies. (A) Turbidity (blue circles and left y-axis) and Thioflavin T fluorescence intensity (green circles and right y-axis) time dependence for 90 µM WT (left panel), DM (middle panel), and P33A (right panel) in the presence of 22 µM polyP. The average from experimental replicates (circles) and the spread in the respective data (shaded region) are shown. (B) Representative fluorescence images of NHS-rhodamine labeled WT (top panel) and DM (bottom panel) in the presence of polyP at different time points. The scale bar is 10 µm. (C-H) FRAP experiments for WT (panels C-E) and DM (panels F-H) in the presence of polyP at different time points - 0 min (pink), 30 min (blue), and 60 min (yellow) – and from five droplets (n = 5) and plotted as mean ± s.d. (C, F) The FRAP recovery curves of NHS-rhodamine labeled WT (panel C) and DM (panel F). The experimental errors (shaded areas) are also shown. (D, G) Recovery amplitudes or extents from FRAP experiments on the WT (panel D) and DM (panel G) at different time points. WT shows less recovery with time, indicating a liquid-to-solid transition, while the DM recovers fully. (E, H) FRAP recovery half-times for the WT (panel E) and DM (panel H) at the indicated time points. (I, J) Droplet size-distribution of WT (panel I) and DM (panel J) at different time points following the same color code in panels C-H. The numbers within the plot represent the mean droplet dimensions at the corresponding time points.

Structural changes in polyP-induced assemblies. (A-C) Far-UV CD spectra of 55 µM WT (A), DM (B), and P33A (C) at different time points - 0 hour (pink), 1 hour (blue), and 4 hours (yellow) - after the addition of 22 µM polyP at 298 K and displayed in mean residue ellipticity (MRE) units of deg. cm2 dmol−1. (D) Time dependence of the signal at 222 nm for the variants studied. Note the clear secondary-structure acquisition in the DM with time. (E) Basis spectra from a global singular value decomposition (SVD) of time-wavelength far-UV CD data. The first and second components are shown in black and gray, respectively. (F, G) Amplitudes of first (F) and second (G) spectra as a function of time for CytR variants following the color code in panel D. (H-J) FTIR spectra for WT (H), DM (I), and P33A (J) showing normalized absorbance recorded at the wavenumber range of 1700 – 1600 cm-1. Black curves represent the FTIR spectra of protein in the absence of polyP. The blue and green curves are the FTIR spectra of protein after the addition of polyP at 0 min and 60 min, respectively. Red arrows indicate the change in peak intensity after the addition of polyP, showing beta-sheet-like conformation in WT and P33A between 1610-1630 cm-1. The vertical black line is at 1652 cm-1, the amide I frequency indicative of helical structure.

DNA induces metastable condensates that solubilize with time. (A) Changes in turbidity in solutions containing increasing concentrations of the WT and a fixed 1 µM concentration 45-bp specific DNA. The error bar indicates the spread from experimental replicates. (B) Fluorescence microscopy images of NHS-rhodamine labeled WT at different concentrations with 1 µM DNA. The scale bar is 10 µm. (C) Representative time-lapse images of WT showing a fusion event. (D) FRAP recovery curves of NHS-rhodamine-labeled CytR WT (green), DM (blue), and P33A (red). The data represents average from n=4 experiments. The experimental errors are shown as shaded areas. (E) Recovery half-times for the CytR variants at the earliest time points (0 minutes). P33A shows the fastest recovery, followed by WT and DM. (F, G) Fluorescence microscopy images of 25 µM DM (F) and P33A (G) in the presence of 1 µM 45 bp DNA. The scale bar is 10 µm. (H) Turbidity of 25 µM WT with 1 µM DNA as a function of time. Data (circles) are from replicates and the error bar indicates the spread. (I) Fluorescence microscopy images of NHS-rhodamine labeled WT with DNA at different time points. The scale bar is 10 µm. (J) Droplet size distribution curve for WT in the presence of DNA at different time points - 0 min (pink), 30 min (blue), and 60 min (yellow). Numbers within the plot represent the mean droplet dimensions at the corresponding time points. (K) Recovery half-times from FRAP experiment on WT with DNA at different time points following the color code for panel J. (L-N) Far-UV CD spectra of 25 µM WT (left), DM (middle), and P33A (right) at different time points - 0 hour (pink), 1 hour (blue), and 4 hours (yellow) - following the addition of 1 µM 45 bp DNA at 298 K. Black curve is the protein spectra in the absence of DNA. The data is reported in mean residue ellipticity (MRE) units of deg. cm2 dmol−1. (O) Time-dependent changes in far-UV CD signal at 222 nm for the CytR WT (green), DM (blue), and P33A (red). Dashed lines show MRE signal at 222 nm for the corresponding proteins in absence of DNA.

PolyP is able to discriminate between folded and molten-globular variants of FruR DBD while DNA does not. (A) 3D structure of the DNA binding domain of FruR highlighting an aromatic stacking interaction between Y19 and Y28. (B, C) Turbidity (blue circles and left y-axis) and ThT fluorescence intensity (green circles and right y-axis) curve for 90 µM FruR WT (B) and Y19A (C) in the presence of 22 µM polyP. The mean data from replicate measurements and the corresponding spread are shown as circles and shaded area, respectively. (D) Representative fluorescence microscopy images of NHS-rhodamine labeled FruR WT (top panel) and Y19A (bottom panel) in the presence of polyP at different time points (0, 0.5 and 1 hour). The scale bar is 10 µm. (E and F) Turbidity of 25 µM WT (E) and Y19A (F) with 1 µM DNA as a function of time. No changes in turbidity are observed with time.