Accurate Method for Generating a Phase Diagram of a Polymer

20250347697 ยท 2025-11-13

    Inventors

    Cpc classification

    International classification

    Abstract

    The present invention refers to a method for generating a binodal curve of a polymer in a system by determining accurate values of both concentration of dilute and condensed phases of a polymer, in particular, protein or polynucleic acid, under different condition, such as at different temperatures and salt concentrations. Furthermore, the present invention refers to an assay method for identify bioactive compound(s), comprising the inventive method for generating a binodal curve of a polymer in a system.

    Claims

    1. A method for determining a concentration (c.sub.con) of a polymer in a condensed phase and a concentration (c.sub.dil) of the polymer in a dilute phase in a system comprising: A) determining a total volume (V.sub.tot) of the system and a total concentration (c.sub.tot) of the polymer; B) triggering phase separation of the polymer into a condensed phase and a dilute phase in said system; C) determining a volume fraction (V.sub.con/V.sub.tot) of the condensed phase of the polymer in said system; D) determining a concentration (c.sub.con) of the condensed phase of the polymer and a concentration (c.sub.dil) of the dilute phase of the polymer in said system by using a following linear form equation: V c o n V t o t = 1 c con - c dil .Math. c tot - c dil c con - c dil , E) changing at least one condition of said system and repeating the steps A) to D).

    2. The method according to claim 1, wherein the polymer is a protein, or a polynucleic acid, preferably RNA, DNA, a mixture of a protein and RNA or a mixture of a protein and DNA and/or wherein the system is an in vitro system and is selected from a solution, an emulsion, or cells.

    3. The method according to claim 1, wherein in step B), the phase separation of the polymer in said system is triggered by changing: the concentration of a component in the system selected from a salt, a crowding agent, or a buffer; the pH value; the pressure; or the temperature of the system.

    4. The method according to claim 1, wherein when the polymer is not labelled with a fluorophore or a fluorescent protein, in step A) the total volume (V.sub.tot) of said system and the total concentration (c.sub.tot) of the polymer are determined by means of bright-field, dark-field, phase-contrast, holographic, polarization, or differential interference correlation (DIC) microscopy, or light-scattering based approaches.

    5. The method according to claim 1, wherein when the polymer is labelled with a fluorophore or a fluorescent protein, in the step A) the total volume (V.sub.tot) of said system and the total concentration (c.sub.tot) of the polymer are determined by means of fluorescence microscopy.

    6. The method according to claim 1, wherein step C) comprises: C1a) encapsulating a portion of the system in an emulsion system; C2a) determining a volume (V.sub.em) of one emulsion droplet of the emulsion system, and a condensed volume (V.sub.con) of the condensed phase of the polymer in said emulsion system; C3a) determining a volume fraction (V.sub.con/V.sub.em) of the condensed volume (V.sub.con) of the condensed phase of the polymer in said emulsion system and the volume (V.sub.em) of one emulsion droplet of said emulsion system, wherein the measured volume fraction (V.sub.con/V.sub.em) is identical to the volume fraction (V.sub.con/V.sub.tot) of the condensed volume (V.sub.con) of the condensed phase of the polymer in the system and the total volume (V.sub.tot) of the system.

    7. The method according to claim 6, wherein in the step C2a) a volume (V.sub.em) of the emulsion system, and a condensed volume (V.sub.con) of the condensed phase of the polymer in said emulsion system are determined by means of fluorescence microscopy.

    8. The method according to claim 6, further comprising after the step C2a): C2b) adjusting the measured fluorescent image with a predetermined partition factor.

    9. The method according to claim 1, further comprising after step B): B) centrifuging the system.

    10. The method according to claim 1, wherein in step E) the at least one condition of said system is selected from a concentration of a component in the system wherein the component is selected from a salt, or a buffer; the pH value; the pressure; a crowding agent; or the temperature of the system, or a combination of the aforementioned conditions.

    11. An assay method for identifying bioactive compound(s), comprising: A) determining a total volume (V.sub.tot) of the system and a total concentration (c.sub.tot) of a polymer in the system; A) adding a predetermined amount of test compounds into said system; B) triggering phase separation of the polymer into a condensed phase and a dilute phase; C) determining a volume fraction (V.sub.con/V.sub.tot) of the condensed phase of the polymer in said system; D) determining a concentration (c.sub.con) of the condensed phase of the polymer and a concentration (c.sub.dil) of the dilute phase of the polymer in said system by using a following linear form equation: V c o n V t o t = 1 c con - c dil .Math. c tot - c dil c con - c dil , D) comparing the measured condensed volume (V.sub.con), the measured condensed concentration (c.sub.con), and/or the measured volume fraction (V.sub.con/V.sub.tot) of said polymer in the system with a condensed volume (V.sub.ref), a condensed concentration (c.sub.ref), and/or a volume fraction (V.sub.ref/V.sub.tot) of the polymer in the presence of a reference molecule in a control system; F) identifying the bioactive compounds from the test compounds; and a device comprises a plurality of systems.

    12. The assay method according to claim 11, wherein the condensed volume (V.sub.ref), the condensed concentration (c.sub.ref), and/or the volume fraction (V.sub.ref/V.sub.tot) of the polymer in the presence of a reference molecule is (are) predetermined or measured at the same time.

    13. The assay method according to claim 11, wherein the identified bioactive compound(s) increase(s) the measured condensed volume (V.sub.con), the measured condensed concentration (c.sub.con), the measured dilute concentration (c.sub.dil), and/or the measured volume fraction (V.sub.con/V.sub.tot) of the polymer obtained by the step C) and/or D) than the predetermined condensed volume (V.sub.ref), the measured condensed concentration (c.sub.ref), and/or the measured volume fraction (V.sub.ref/V.sub.tot) of the polymer of said polymer in the presence of a reference molecule in the control system.

    14. A device for determining a binodal curve of a polymer in a system comprising: i) means for measuring a total volume (V.sub.tot) of the system; ii) means for measuring a volume fraction (V.sub.con/V.sub.tot) of a condensed phase of the polymer in said system; iii) means for calculating a concentration (c.sub.con) of the condensed phase of the polymer and a concentration (c.sub.dil) of the dilute phase of the polymer in said system by using a following linear form equation: V c o n V t o t = 1 c con - c dil .Math. c tot - c dil c con - c dil .

    15. The device according to claim 14, wherein the means for calculating a concentration (c.sub.con) of the condensed phase of the polymer and a concentration (c.sub.dil) of the dilute phase of the polymer in said system is a computer which is connected to a controller and a detector, wherein the computer is configured to receive signals detected from the detector and wherein the computer is configured to calculate a concentration (c.sub.con) of the condensed phase of the polymer and a concentration (c.sub.dil) of the dilute phase of the polymer in said system by using a following linear form equation: V c o n V t o t = 1 c con - c dil .Math. c tot - c dil c con - c dil .

    16. The method according to claim 1, wherein the polymer is a mixture of polymers, and the mixture of polymers is selected from a group comprising or consisting of a mixture of proteins, a mixture of RNAs, a mixture of DNAs, a mixture of proteins and RNAs, a mixture of proteins and DNAs, and a mixture of proteins, RNAs and DNAs.

    17. The method according to claim 7, further comprising after the step C2a); c2b) adjusting the measurement fluorescent image with a predetermined partition factor.

    Description

    DESCRIPTION OF FIGURES

    [0306] FIG. 1: Concentration series in defined volumes allow for determination of c.sub.con and c.sub.dil. a, Schematic of the binodal of a phase diagram with dilute branch c.sub.out and condensed branch c.sub.con concentrations connected via a tie line. b, Schematic of the experimental protocol and setup design using a titration of c.sub.tot in water-in-oil emulsions mounted on a temperature-controlled device. c, The total concentration of protein c.sub.tot used in the volume V.sub.tot of an emulsion droplet separates into the dilute phase with c.sub.dil and V.sub.out and the condensed phase with c.sub.con and V.sub.in. d, Example images of PGL-3 condensates in emulsions mounted on the device in b. The brightfield channel is used to determine the volume of the emulsions droplets while the fluorescent channel is used to determine the condensed phase volume (scale bar 100 m). e, Zoom on segmentation of emulsion droplet and condensate of white square region in d (scale bar 10 m). f, Using mass and volume conservation, the volume fraction of the condensed phase (V.sub.con/V.sub.tot) describes a linear form with the slope (c.sub.conc.sub.dil).sup.1 and an offset of c.sub.dil/(c.sub.con-.sub.dil), allowing to derive both c.sub.dil and c.sub.con via a linear regression against c.sub.tot. g, equation 1 and linear regression of the ratio V.sub.in/V.sub.tot and the total polymer concentration, c.sub.tot. h, calibration of image analysis pipeline with fluorescent polystyrene beads of known size.

    [0307] FIG. 2: Salt and temperature dependence of phase separated FUS and PGL-3 protein. a, False colour experimental images of PGL-3+5% PGL-3::GFP for various temperatures and concentrations (binodal line is a guide to the eye). B, Temperature vs. protein concentration phase diagrams of dilute (c.sub.dil) and condensed (c.sub.con) branch of FUS::GFP for various salt concentration. Derived from linear regression to c.sub.tot concentration series for each salt concentration and temperature (error bars are SD). c, Temperature vs. protein concentration phase diagrams for PGL-3 spiked with 5% PGL-3::GFP. d, e, Data of b and c presented as salt concentration vs. protein concentration phase diagrams for FUS and PGL-3 respectively. f, Comparison of the dilute branch concentrations c.sub.dil of FUS and PGL-3 as surfaces in a salt concentration, temperature space.

    [0308] FIG. 3: Reversible temperature quenches in and out of the two-phase region for FUS::GFP. a, Dilute branch binodal (c.sub.dil) for FUS::GFP at different salt concentrations. Solid dots demark the condensation temperature T.sub.cond for the protein concentrations used at specific salt concentrations that allow to cross the binodal with a temperature quench. b, Timeseries of temperature quenches from 35 C. to 10 C. and back to 35 C. Upper panel shows experimental data for the volume fraction and measured temperature curves for four different salt and protein concentrations of FUS::GFP (10 s resolution). Lower panel shows example projections of 3D image stacks at specific time points (also indicated by the red dots on the temperature curve). Yellow circles indicate remaining condensates/aggregates for low salt concentrations after the temperature quench.

    [0309] FIG. 4: Effect of temperature and salt on the partition factor and comparison to quantitative fluorescence. a, b, Partition factor PF.sub.c (c.sub.con/c.sub.dil) derived via inventive method (inPhase) for various temperatures and salt concentrations for FUS::GFP and PGL-3+5% PGL-3::GFP respectively (y-axis logarithmic). c, Partition factor as a function of the amount of labelled PGL-3 protein fraction. While PF.sub.c is derived via inPhase PF.sub.I is the partition factor calculated via fluorescence intensities in the dilute and condensed phase of the same sample. d, Partition factor PF.sub.I of individual condensates as a function of their volume for FUS::GFP and GFP and PGL-3+5% PGL-3::GFP respectively. Upper and lower panels contain the same data with the lower panels depicted with a logarithmic x-axis (dashed lines indicate PF.sub.C as derived via inPhase method for the same data-set). e, Relative partition factor of inPhase PF.sub.c divided by quantitative intensity-based PF.sub.I for various salt concentrations and temperatures for FUS and PGL-3.

    [0310] FIG. 5: Comparison of variations of the inventive method (inPhase) with standard bulk measurements for PGL-3. a, Overview of c.sub.tot titration series of projections of 3D image stacks for PGL-3+5% PGL-3::GFP in 386 well-plates. b, Comparison of dilute branch concentrations c.sub.dil for inPhase using fluorescence (fluo) and brightfield (BF) in emulsions and the well-plate approach versus bulk measurements via Pierson BCA (BCA) and 280 nm UV absorption measurements in a nanoDrop (nDrop). c, Comparison of condensed branch concentrations c.sub.con for inPhase and bulk approaches. Dashed line indicates the value for the emulsion fluorescence-based evaluation as a standard.

    [0311] FIG. 6: shows an exemplary setup for the determination of the height of a condensed phase layer within a container (no need for fluorescent labels). This representation uses the reflection of light at the interface to determine the thickness of a layer.

    [0312] FIG. 7: shows the time-dependent changes of condensed and solute concentration for two exemplary proteins.

    [0313] c.sub.dil and c.sub.con do not change considerably during the experimental window. a, b, Time dependent evaluation using the inventive method of dilute and condensed branch concentrations for PGL-3 (a) and FUS (b) (Errors from error propagation of linear regression confidence intervals). Shaded areas depict the time-window used for the temperature and salt dependent phase diagrams.

    [0314] FIG. 8: Protein concentration heterogeneity in C. elegans embryos allows to test the inventive method in vivo. a, Data of condensed volume fraction against total PGL-1::GFP concentration at 20 C. (dots, N=46). Linear regression to the data (dashed line) allows to use equation 1 and derive c.sub.dil and effective c.sub.cond concentrations. b, Data of condensed volume fraction against total PGL-3::GFP concentration at 20 C. (dots, N=29). Linear regression to the data (dashed line) allows to use equation 1 and to derive c.sub.dil and effective c.sub.cond concentrations. c, scheme of a one-cell C. elegans embryo. d, Temperature dependent phase diagram of PGL-1::GFP in C. elegans embryos. Adapted from Fritsch et al. PNAS, 2021, 118(37), e2102772118.

    [0315] FIG. 9: shows comparison of the effect of untreated control (DMSO) and Lipoamide on the phase diagram (top panel) and partition factor of FUS (bottom panel).

    [0316] FIG. 10: shows a schematic drawing of the device according to the invention to determine V.sub.frac using the difference in refractive index.

    [0317] FIG. 11: shows a diagram of the intensity of reflected light in dependency of the sample depth.

    [0318] FIG. 12: RNA to protein ratio dependence of phase separated FUS. a, molar RNA to protein ratio vs. total concentration (RNA+protein) phase diagrams of dilute (c.sub.dil) and condensed (c.sub.con) branch of FUS::GFP for various RNA to protein ratios. Derived from linear regression to c.sub.tot concentration series for each RNA to protein ratio; b, volume fraction vs. total concentration (RNA+protein) at a constant molar RNA to protein ratio of 0.01. Experiments were carried out as described in Example 5 with Poly(A) RNA as RNA and FUS-GFP as protein.

    [0319] FIG. 13: Linear regressions of V.sub.frac(c.sub.tot) data and phase diagrams for FUS. a, False colour experimental images of PGL-3+5% PGL-3::GFP for various temperatures and concentrations (binodal line is a guide to the eye). b-g, Linear regressions of V.sub.frac(c.sub.tot) of samples prepared at 100 (b), 150 (c), 175 (d), 200 (e), 225 (f) and 240 (g) mM KCl and measured at different temperatures. Experimental data is pooled from all independent repeats and then fitted to Equation 1. Each individual experiment consists of data points from many emulsion droplets. Error bars depict standard deviations. h, Quality of linear regression (R-squared value) of the data presented in panel b-g. Shaded areas depict standard deviations derived from individual repeats. h, i, Temperature (i) and salt (j) phase diagrams presenting the derived data for c.sub.con and c.sub.dil from the linear regressions in panel b-g. Lines connecting the data points are a guide to the eye. Shaded areas depict the errors derived from propagation of confidence intervals of the linear regression in panel b-g.

    [0320] FIG. 14: Linear regressions of V.sub.frac(c.sub.tot) data and phase diagrams for FUS. a-e, Linear regressions of V.sub.frac(c.sub.tot) of samples prepared at 75 (a), 100 (b), 150 (c), 200 (d) and 300 (e) mM KCl and measured at different temperatures. Experimental data is pooled from all independent repeats and then fitted to Eq. 1. Each individual experiment consists of datapoints from many emulsion droplets. Error bars depict standard deviations. f, Quality of linear regression (R-squared value) of data presented in panel a-e. Shaded areas depict standard deviations derived from individual repeats. g, h, Temperature (g) and salt (h) phase diagrams presenting the derived data for c.sub.con and c.sub.dil from the linear regressions in panel a-e. Lines connecting the data points are a guide to the eye. Shaded areas depict the errors derived from propagation of confidence intervals of the linear regression in panel a-e.

    [0321] FIG. 15: Effect of temperature and salt on the partition factor and comparison to quantitative fluorescence for FUS and PGL-3. a, Partition factor PF.sub.inPhase (c.sub.con/c.sub.dil) derived via inPhase method for various temperatures and salt concentrations for FUS::GFP (left panel) and PGL-3+5% PGL-3::GFP (right panel). b, PF.sub.intensity of individual condensates as a function of their volume for FUS::GFP (left panel) and PGL-3+5% PGL-3::GFP (right panel). Upper and lower panels contain the same data with the lower panels using a logarithmic x-axis (dashed lines indicate PF.sub.inPhase for the same data-set). c, Ratio PF.sub.inPhase/PF.sub.intensity for various salt concentrations and temperatures for FUS (left panel) and PGL-3 (right panel). d, Partition factor (PF) as a function of the amount of labelled PGL-3 protein fraction. The upper panel shows the c.sub.con (dark gray) and c.sub.dil (light gray) calculated using inPhase. The lower panel shows the ratio of PF.sub.inPhase/PF.sub.intensity at different GFP percentages. PF.sub.intensity underestimates the partition factor. Error bars show SD.

    [0322] FIG. 16: Increasing the RNA to protein ratio reduces c.sub.dil. a, Titration of total protein concentration at a constant ratio of poly(A) RNA to protein concentration. R-squared values are from linear regression to the individual curves (errors depict standard deviation). b-c, Derived values of c.sub.dil and effective c.sub.con using inPhase on the data presented in panel a.

    EXAMPLES

    General Method

    [0323] The invention relates to the study of protein solutions that undergo phase separation into a dense and a dilute phase. It allows to determine the protein concentrations of both the dense and the dilute phase. It relies on measuring the volume fraction of the condensed phase in the total reaction for a titration series of total protein concentration c.sub.tot.

    [00025] V c o n V t o t = 1 c con - c dil .Math. c tot - c dil c con - c dil linear form ( 1 c con - c dil and c dil c con - c dil are constants )

    Protocol 1 (Water-In-Oil Emulsions):

    [0324] 1. Measure the protein concentration of the stock solutions. The stock solutions can be unlabelled, a mix of labelled and unlabelled or fully labelled protein. Solutions containing only unlabelled protein will be detected using brightfield (or phase-contrast, DIC, etc.) while the mixes that contain labelled protein can be also detected with fluorescence means. [0325] 2. Make a titration series of the protein solution. [0326] 3. Trigger phase separation of the protein samples. Phase separation can be triggered for example by a change in salt, pH or temperature. [0327] 4. Immediately after triggering phase separation, encapsulate in a water-in-oil system (e.g. PicoSurf oil (Sphere Fluidics, UK)). [0328] 5. Centrifugation after encapsulation is optional but helps in the segmentation of the condensates. This is because the centrifugation coarsens the condensates. [0329] 6. Load the titration series into chambers for microscopy imaging. A custom-built temperature-controlled stage for mounting was used, thereby allowing performing temperature dependent sweep of the binodal. [0330] 7. Acquire images of the emulsion droplets and the encapsulated condensates. This allows to determine the volume of the emulsion droplet as well as the condensate. Now the ratio condensate over emulsion droplet volume which is the volume fraction can be determined. In case of fluorescently labelled proteins the determination of the condensate volume can be complemented by fluorescence microscopy. [0331] 8. Perform the image and segmentation analysis and fit the measured volume-fractions to a linear regression over the protein concentrations. [0332] 9. Determine the c.sub.con and c.sub.dil using the linear form derived from mass- and volume-conservation equations. [0333] 10. Repeat under different conditions (temperature, pH, salt, small molecule concentration) to generate a phase diagram.

    Protocol 2 (Well-Plate):

    [0334] Step 1.-3. are the same as in Protocol 1. [0335] 4. Immediately after triggering phase separation, pipette the samples in the well plate. [0336] 5. Centrifugation can be used to coarsen the condensates but in this case, it is preferred to let the condensates sediment for 1 hour. [0337] 6. Acquire images of the condensates in the well. The sample volume in a well is known and thus the volume fraction is the result of the segmented condensate volume divided by the sample volume.

    [0338] Continue with steps 8.-10. in Protocol 1.

    Example 1

    PGL-3 Protein Purification:

    [0339] PGL-3 was purified from insect cells according to Saha et al. (Cell 166, 1572-1584.e16 (2016)) from SF9-ESF cells were infected with baculovirus containing the PGL-3-GFP-6HIS protein under the polyhedrin promoter. Cells were harvested after 3 days of infection by centrifugation at 500g for 10 min and then resuspended in lysis buffer (25 mM HEPES 7.25, 300 mM KCl, 10 mM imidazole, 1 mM DTT, 1 protease inhibitor). Cells were lysed by passing the cells 2 times through the LM20 microfluidizer at 15 000 psi. The lysate was then centrifuged at 20 000 rpm for 45 min at 15 C. The lysate was loaded in a pre-equilibrated Ni-NTA column with lysis buffer at 3 mL/min. The Ni-NTA column was rinsed with 10 C.V of wash buffer (25 mM HEPES 7.25, 300 mM KCl, 20 mM imidazole, 1 mM DTT, 1) and the protein was eluted in 1.5 mL fractions with elution buffer (25 mM HEPES 7.25, 300 mM KCl, 250 mM imidazole, 1 mM DTT). After elution the GFP tagged was cleaved to produce untagged PGL-3. The cleavage was performed using a TEV protease overnight at 4 C. PGL-3 and PGL-3-GFP proteins (SEQ ID No: 2) were diluted with Dilution buffer (25 mM Tris pH 8.0, 1 mM DTT) to reach 50 mM KCl before loading the protein in an anion exchange HiTrapQ HP 5 mL column. The HiTrap column was previously equilibrated first with HiTrapQ elution buffer (25 mM Tris pH 8.0, 50 mM KCl, 1 mM DTT) and then with HiTrapQ binding buffer (25 mM Tris pH 8.0, 1 M KCl, 1 mM DTT). The column was mounted in a Akta Pure FPLC system. After the sample was loaded the column was washed with HiTrapQ binding buffer. The sample was finally eluted with a linear gradient from 0 to 55% of HiTrapQ elution buffer (25 mM Tris pH 8.0, 1 M KCl 1 mM DTT) for 25 C.V. Finally, a 100% HiTrap elution buffer step was performed for 5 C.V. The pooled fractions were then loaded in a HiLoad 16/60 Superdex 200 size exclusion chromatography column that was previously equilibrated with Superdex buffer (25 mM HEPES 7.25, 300 mM KCl, 1 mM DTT). After size exclusion, the final samples were collected.

    FUS Protein Purification:

    [0340] Unlabeled FUS purified from a baculovirus construct containing N-HIS-MBP-FUS-TEV-SNAP. SF9-ESF cells were harvested after three days of infection by centrifugation at 500g for 10 min. The cell pellet was resuspended using 50 mL of lysis buffer (50 mM Tris pH 7.4, 500 mM KCl, 5% glycerol, 10 mM imidazole, 1 mM PMSF, 1 protease inhibitor) for every 50 mL of cultured cells. The cells were lysed by passing them 2 times through the LM20 microfluidizer at 15 000 psi. The lysate was then centrifuged at 20 000 rpm for 45 min at 15 C. The supernatant was collected and loaded into a Ni-NTA column that was previously equilibrated with lysis buffer. After loading the sample, the column was washed for 10 C.V. with Ni-NTA was buffer (50 mM Tris pH 7.4, 500 mM KCl, 5% glycerol, 20 mM imidazole). The protein was then eluted with Ni-NTA elution buffer (50 mM Tris pH 7.4, 500 mM KCl, 5% glycerol, 300 mM imidazole). The collected fractions where then loaded into a MBPTrap HP column preequilibrated with Ni-NTA elution buffer. The MBP column was washed for 10 C.V with MBP wash buffer (50 mM Tris pH 7.4, 500 mM KCl, 5% glycerol). After washing, the sample was eluted with MBP elution buffer (50 mM Tris pH 7.4, 500 mM KCl, 5% glycerol, 500 mM arginine, 20 mM maltose). The protein was diluted to a concentration of less than 15 M using MBP elution buffer. 3C and TEV proteases were then added to cleave the MBP and SNAP tags from the FUS construct. The cleavage reactions were incubated overnight at 18 C. Finally, the protein was loaded in a SepFast GF-HS-L 26600 mm gel filtration to remove the cleaved MBP and SNAP tags and exchange the buffer. The SepFast column was previously equilibrated in storage buffer (50 mM HEPES pH 7.25, 750 mM KCl, 5% glycerol, 1 mM DTT). The sample was concentrated to a final concentration of 15 L using 30 kDa Amicon centrifuge filters. FUS-GFP (SEQ ID No: 1) was purified as previously described (Wang, J. et al. A Molecular Grammar Governing the Driving Forces for Phase Separation of Prion-like RNA Binding Proteins, Cell, 2018, 174, 688-699.e16).

    Protocol In Vitro Protein Handling:

    [0341] Protein solutions were prepared from stocks that are frozen at 80 C. The protein stock solutions were cleared from potential aggregates using centrifugal filters (UFC30HV00, Merck, Germany). The protein concentrations were determined via extinction coefficient measurements at 280 nm (NanoDrop, Thermo Fisher Scientific, USA). Next, the desired label fraction, in the main experiments 5% GFP labeled protein, was prepared. In order to keep the protein from phase separating to early the titration of the final protein concentrations was done at salt concentrations of the protein stock (300 mM KCl for PGL-3 and 750 mM KCl for FUS; both 25 mM HEPES and 1 mM DTT at 7.4 pH). Only in the final step the salt concentration was dropped to the desired value, using a no salt buffer (25 mM HEPES and 1 mM DTT at 7.4 pH), and the solution was immediately encapsulated into the water-in-oil solution. Encapsulation was performed using twice the amount of PicoSurf oil (2% (w/w) in Novec 7500, Sphere Fluidics, UK) compared to the protein solution (usually 5 L). After adding the oil on top of the protein solution in standard 100 l microcentrifuge tubes the solution is agitated using a 10 l pipette until the desired size distribution of water-in-oil solutions is reached. This solution can be used directly for imaging. Alternatively, the coarsening of the phase separated condensates can be speed up by a mild centrifugation step of the water-in-oil emulsions (100-1000 g for 3 min). This will lead to predominantly one single condensate of condensed phase per emulsion droplet.

    Preparation of Emulsions:

    [0342] Protein solutions were prepared from stocks that are frozen at 80 C. The protein stock solutions were cleared from potential aggregates using small volume centrifugal filters (UFC30W25, Merck, Germany). The protein concentrations were determined via extinction coefficient measurements at 280 nm (NanoDrop, Thermo Fisher Scientific, USA). Next, the desired fraction of labelled protein was prepared (main experiments 5% GFP labelled protein for PGL-3, 100% for FUS). The protein concentrations were adjusted using the same salt concentration as the protein stock (300 mM KCl and 25 mM HEPES for PGL-3 and 500 mM KCl and 50 mM HEPES for FUS; both 1 mM DTT at pH 7.4). In the final step the salt concentration was reduced to the desired value, using a buffer containing no salt (HEPES and 1 mM DTT at pH 7.4) to trigger phase separation. The protein solution was then immediately encapsulated into a water-in-oil solution. Encapsulation was performed using twice the amount of PicoSurf oil (2% (w/w) in Novec 7500, Sphere Fluidics, UK) compared to the protein solution. Here, we add the oil to the protein solution in standard 100 l PCR tubes and the solution is agitated using a 20 l pipette and 200 l tips until the desired size distribution of water-in-oil solutions is reached. To coarsen the phase separated condensates within the emulsion drops we used a centrifugation step (200-400 g for 4 min) at the desired temperature of the first data point. This will lead to a single condensate per emulsion droplet in most emulsion droplets. A titration series of the total protein concentration was then mounted on a temperature controlled microscope stage.

    Sample Loading and Imaging Procedure for Emulsions:

    [0343] As a sample holder a custom-made temperature stage based on Peltier elements connected to a sapphire microscope slide (2575 mm) was used. This allows for fast and precise temperature changes both below and above room temperature. Between the sapphire slide and a standard cover slide (2222 mm, 170 m, Menzel, Germany) lanes of Paraflim (Bernis, USA) stripes were sandwiched. The sandwich is connected by heating the stage to at least 55 C. which will connect the Parafilm to the top and bottom by applying soft pressure on the cover slide. This allows to load multiple samples in a separated fashion in one measurement. Emulsion solutions are loaded into the individual lanes by pipetting the turbid upper layer in the microcentrifuge tubes. Each lane is sealed using addition curing silicone (Picodent twinsil speed, Picodent, Germany) to ensure no evaporation and movement of the water-in-oil emulsions upon temperature changes. Sample imaging was performed via CellSens software (Olympus, Japan) on an Olympus IX83 microscope connected to a Yokogawa W1 SoRa spinning-disc system (Yokogawa, Japan) and a Hamamatsu Orca Flash v3 sCMOS camera (Hamamatsu, Japan) using a 40 air objective (UPLXAPO, 0.95NA, Olympus, Japan). For each sample, parafilm lane, a multi-tile brightfield image and a corresponding multi-tile 3-dimensional fluorescence stack was recorded using a piezo stage (Mad city labs, USA). The main temperature experiments started at 10 C. and increased the temperature by 5 C. steps. Between consecutive temperatures we added 5 min of equilibration time before imaging.

    Image Analysis:

    [0344] Images were processed by a custom-written pipeline using MATLAB software (The Mathworks, USA). The brightfield images of each lane were used to detect the contours of the water-in-oil emulsion droplets. The location as well as the volume of each emulsion droplet was determined considering the height of approximately 100 m of the Parafilm chambers. Corresponding positions in the fluorescence image stack are further analyzed after removing emulsion droplets that are intersected by the margins of the imaged tiles as well as overlapping emulsions. For each emulsion droplet its condensed phase is determined by image registration and parameters like total condensed phase and intensity values are extracted. In brief, after basic image filtering, an intensity gradient approach was used to find a good estimate for the outline of the condensates.

    [0345] The inventors then assume sphericity of each condensate to calculate its volume. For large condensates the spherical assumption might introduce overestimation of the actual volume since they can flatten out due to gravity. To get valid estimates for the dilute phase intensities the inventors first dilate the masks of the condensed phase and remove them from the mask of the emulsion droplet. This helps to minimize the influence of intensity values close to the margins of the condensate phase. Furthermore, the inventors can also use the brightfield images to estimate the volume of the condensed phase. The image analysis pipeline was calibrated using monodisperse fluorescent beads of known size immersed in glycerol solution to mimic the refractive index difference found for condensates in buffer.

    Calculation of Dilute and Condensed Branch Concentrations:

    [0346] To get to the form of Equation 1 the inventors assume volume and mass conservation in the reaction container. While there could be processes that lead to a change of total volume in case of a phase transition in our case of protein phase separation the actual volumes of the condensed phase are orders of magnitude smaller than the total reaction volume and thus even in such a case the linear form will be valid for small volume fractions V.sub.frac=V.sub.con/V.sub.tot.

    [0347] The conservation reads as V.sub.tot=V.sub.dil+V.sub.con for the volume and as V.sub.tot.Math.c.sub.tot=V.sub.dil.Math.c.sub.dil+V.sub.con.Math.c.sub.con for the mass given that the mass equals the volume times the concentration (m=V.Math.c). Converting the mass conservation to

    [00026] V frac = c tot c con + V dil V tot .Math. c dil c con

    and substituting the volume conservation as

    [00027] V dil V tot = 1 + V con V tot

    leads to

    [00028] V frac = 1 c con - c dil .Math. c tot - c dil c con - c dil . ( Equation 1 )

    [0348] The volume fraction data for each total concentration was pooled from all repeats and used for linear regression of the concentration series in c.sub.tot. Using the linear form V.sub.frac=a.Math.c.sub.tot+b the inventors get

    [00029] c con = 1 - a b and c dil = - b a

    according to Equation 1. Via the confidence interval of the fit and error propagation, the inventors can determine an error for the dilute- and condensed-phase concentrations.

    Example 2. Time Dependent Behavior of the Condensates

    [0349] It has been observed, that the material properties of condensates change over time. The material properties of a condensate are relevant for their function. This is especially interesting for the ageing of condensates that are linked to neurodegenerative diseases (e.g. hardening of FUS condensates related to ALS). However, it remains challenging to measure material properties on such small scales. Here, it is shown that changes in the phase (diagram) can be tracked over time using the inventive approach (FIG. 7). [0350] 1. Prepare the sample according to protocol 1 or 2 (e.g., a total protein concentration titration at phase separating conditions, at the desired external conditions (temperature, salt concentration, pH, pressure, buffer composition, crowding agent)). [0351] 2. Perform time-dependent measurement of the volume fraction of the condensed phase (e.g. via fluorescence microscopy based imaging using a labeled protein, or brightfield imaging using unlabeled protein, or any of the other techniques outlined herein). [0352] 3. Use linear regression to fit the volume fraction versus total concentrations data for each time-point and apply our formula to get to a time-dependent c.sub.in(t) and c.sub.out(t). [0353] 4. A time-dependent plot of either c.sub.in or c.sub.out (or any derivative value of the two, e.g. the ratio) allows assessing the ageing of the protein condensed phase of interest.

    Example 3. Effect of Small Molecule Drugs

    [0354] To demonstrate the possibilities of drug screening, a small molecule that has been indicated to affect the phase separation of FUS, a protein related to ALS has been used. It has so far been difficult to accurately the extent of the effect because of a lack of quantitative measures (FIG. 9).

    [0355] Here the effect of Lipoamide on the phase behavior of FUS is shown, where a reduced amount of conformational changes at higher temperatures was observed.

    [0356] This is indicated by the earlier bent of the condensed phase branch of the binodal for the untreated control (DMSO). This can also be seen in the partition factor c.sub.in/c.sub.out. [0357] 1. Prepare the sample according to protocol 1 or 2 (e.g., a total protein concentration titration at phase separating conditions). [0358] 2. Add small molecule drug of interest to the final protein mix before encapsulation for the emulsion approach or before/after adding protein mix to the well-plates (depending on intention to measure the effect of the drug during formation of the condensates (before) or on the already formed condensates (after)). [0359] 3. Use linear regression to fit the volume fraction versus total concentrations data for each drug concentration and apply our formula to get to a dose-dependent c.sub.in and c.sub.out compared to a control (e.g., DMSO control). [0360] 4. This allows preparing an IC50 plot for the effect of the small molecule drug.

    Example 4. Phase Diagrams In Vivo

    [0361] In general, the method of the present invention can be used for any phase separating system. This is also true for in vivo settings of phase separating proteins. The natural or induced fluctuations in the total protein concentration of cells are equivalent to our protein concentration titration series. Thus, the inventive method can be applied to calculate c.sub.dil and c.sub.con. As previously discussed, the volume fraction of the phase and the total concentration of protein (e.g. via GFP labelled protein) is determined. This is shown exemplary for C. elegans embryos and the proteins PGL-1 and PGL-3 that are constitutive for P granules (FIG. 8). [0362] 1. Determine the volume fraction of the condensed phase of a protein inside a cell (e.g., using a fluorescently labelled protein (live imaging) or fixed cells with immunofluorescence). [0363] 2. Determine the total concentration of the protein of interest within the given cell. For example, by using the mean intensity of the protein in the cell and relating it to a known standard (e.g., fluorophore concentration titration, quantitative immunoblots see (Fritsch et al. PNAS, 2021, 118(37), e2102772118), mass spectrometry see (Saha et al Cell 166, 1572-1584.e16 (2016))). [0364] 3. Use the intrinsic variability of the total concentration between cells to plot the volume fraction over total protein concentration. [0365] 4.a. In addition, RNAi can be used to decrease the total concentration of a given protein in vivo in a controlled way. [0366] 4.b In addition, inducible gene expression (e.g. doxycycline based) can increase the total concentration of a given protein in vivo in a controlled way. The gene can be integrated in the genome or transfected in a plasmid. [0367] 5. Use linear regression to fit the volume fraction versus total concentrations data for and apply equation 1 to get c.sub.dil and an apparent c.sub.con.

    Example 5: Effect of RNA on Protein Phase Separation

    [0368] FUS-GFP containing samples were prepared as described in Example 1. Phase separation was triggered by simultaneously lowering the salt concentration of the sample and adding the Poly(A) RNA in an amount to maintain a fixed molar ratio of Poly(A):FUS-GFP. The volume fraction of the condensed phase was determined as described in Example 1 and c.sub.dil and c.sub.con, were determined, wherein c.sub.con represents an apparent concentration containing both Poly(A) (SEQ ID No: 3) and FUS-GFP concentration. The concentration of the single components of the sample can be resolved by the knowledge of the molar ratio of RNA to protein in the condensed phase (FIG. 16).

    Example 6: Multi-Well Plate Assay

    [0369] The inventors load equal amounts of 20 l sample volume for the total protein concentration titration in 384 well plates (PhenoPlate, PerkinElmer, USA). The well plates were chosen due to the superior non adhesive properties that allow to assume spherical shape of the condensates. Well plates are centrifuged at 200 g for 10 min with low acceleration and deceleration of the rotor to minimize coarsening into one condensate. The segmentation routine used for the emulsion-based approach was calibrated for the well plates and used to determine the volume fraction of the condensed phase for each total concentration. The inventors then use equation 1 to derive c.sub.dil and c.sub.con and the respective errors.

    Example 7: Bulk Concentration Measurements

    [0370] Centrifugation of a phase separated suspension provides us with the supernatant containing the dilute phase and a condensed phase at the bottom of a PCR tube that can both be used for the bulk measurements when using large amounts of sample. The inventors could use measured volume fractions of condensed phase for given external parameters to arrive at volumes of condensed phase suitable for pipetting. The inventor used the Pierce BCA protein assay kit (ThermoFischer, USA) together with a TECAN Spark 20M plate reader (TECAN, Swiss) as measurement a triplicate of an BSA standard curve was used to convert intensities to concentration. For each measurement a triplicate of a total concentration (c.sub.tot) series below and above the saturation concentration (c.sub.dil) were carried out. Below c.sub.dil we see a match of c.sub.tot and the measured concentrations (c.sub.BCA) and as expected above c.sub.dil a plateau-like region for c.sub.BCA. The intersect of linear fits to the plateau and the initial slope provides us with c.sub.dil. To determine the condensed phase concentration c.sub.con via fluorescence absorption measurement with a NanoDrop spectrometer (ThermoFisher, USA) the inventors used a negative displacement pipette (Eppendorf, Germany). This was necessary because of the high viscosity of the condensed phase and potential errors using air pipettes. A known volume of condensed phase was diluted into high salt buffer before absorption measurement to ensure dissolution of the phase. In addition, guanidinium chloride can be used to dissolve condensates for proteins that would not dissolve again in high salt buffer.

    Example 8: Temperature Shift Experiments

    [0371] Emulsions were prepared as described in the section Preparation of emulsions as described in Example 1 except without a final centrifugation step. This step was not necessary since the total protein concentrations were adjusted to be outside the binodal at the temperatures used for preparation. We used FUS::GFP at the salt concentrations of 100, 150, 200, and 300 mM KCl at protein concentrations of 2.4, 4.8, 7.9, and 15.9 M respectively (all pH 7.4). A low density of emulsion droplets was mounted in Parafilm chambers on the temperature-controlled stage to ensure stationary emulsions during temperature shift experiments. Images were recorded using the described spinning disc confocal system at 10 s temporal resolution and 1 m resolution in the z-axis using a 40 air objective.

    Example 9: One-Cell C. elegans Embryo Experiments

    [0372] One-cell embryos were imaged and analyzed according to a protocol described previously (Fritsch, A. W. et al. Local thermodynamics govern formation and dissolution of Caenorhabditis elegans P granule condensates. PNAS, 2021, 118.). This allowed to derive the volume fraction of condensed phase as well as the total concentration of labelled protein per embryo. In brief, known total protein concentrations (PGL-1::GFP, PGL-3::GFP respectively) in the different worm strains were related to fluorescence intensities to derive a scaling factor between florescence intensity and concentration. To derive the total condensate volume as well as the embryo volume, the inventors used a custom-written MATLAB (The MathWorks) segmentation routine. The worm lines used were TH586, pgl-1::mEGFP pgl-1(dd54[pgl-1::mEGFP]) and TH561, pgl-3::mEGFP (pgl-3(dd29[pgl-3::mEGFP]). Here, the proteins PGL-1 and PGL-3 were labelled at their endogenous genomic locus with monomeric enhanced GFP using the co-CRISPR method.