Controlling Drug Partitioning in Individual Protein Condensates through Laser-Induced Microscale Phase Transitions

Gelation of protein condensates formed by liquid–liquid phase separation occurs in a wide range of biological contexts, from the assembly of biomaterials to the formation of fibrillar aggregates, and is therefore of interest for biomedical applications. Soluble-to-gel (sol–gel) transitions are controlled through macroscopic processes such as changes in temperature or buffer composition, resulting in bulk conversion of liquid droplets into microgels within minutes to hours. Using microscopy and mass spectrometry, we show that condensates of an engineered mini-spidroin (NT2repCTYF) undergo a spontaneous sol–gel transition resulting in the loss of exchange of proteins between the soluble and the condensed phase. This feature enables us to specifically trap a silk-domain-tagged target protein in the spidroin microgels. Surprisingly, laser pulses trigger near-instant gelation. By loading the condensates with fluorescent dyes or drugs, we can control the wavelength at which gelation is triggered. Fluorescence microscopy reveals that laser-induced gelation significantly further increases the partitioning of the fluorescent molecules into the condensates. In summary, our findings demonstrate direct control of phase transitions in individual condensates, opening new avenues for functional and structural characterization.


Chemicals
All reagents were purchased from Sigma-Aldrich if not stated otherwise.

Protein expression and purification
NT2RepCT, NT2RepCT YF , NT, NT-GFP and NT*-GFP were expressed and purified as previously described except that sonication (5 min, 2 sec on/ 2 sec off, 30% amplitude) instead of a high-pressure cell disrupter was used for protein solubilization. 1,22Rep was expressed as a fusion protein with the NT domain and CT as a fusion protein with thioredoxin.
After purification of the fusion proteins in the same way as the other constructs, 2Rep and CT were released via thrombin cleavage (1:1000 w/w) overnight at 4°C and the tag removed by affinity chromatography.Purified proteins were dialyzed against deionized water and stored at -80°C. 15N-labeled NT2RepCT YF was expressed in E. coli BL21 (DE3) cells using 2x M9 medium.20 ml start culture containing 2x M9 medium and 75 µg/ml kanamycin was inoculated with NT2RepCT YF from a previously prepared glycerol stock stored in Luria broth (LB) medium and incubated overnight at 37°C.Cells from the start culture (5% volume of the main culture) were transferred in fresh 2x M9 medium containing 75 µg/ml kanamycin and incubated at 31°C until the OD600 nm reached 0.9.For protein expression the incubation temperature was lowered to 20°C, Isopropyl b-D-1-thiogalactopyranoside (IPTG) was added to 0.5 mM, and cells were incubated overnight.Protein purification was carried out as for unlabeled NT2RepCT YF .hTau was expressed and purified as previously described and stored at -20°C in 20 mM sodium phosphate buffer pH 7.4 with 0.2 mM EDTA. 3 Met-Ab42 was expressed and purified as previously described. 4

Protein labeling
Prior to the labeling reaction protein stocks were buffer exchanged to 10 mM sodium phosphate pH 8.3 using slide-A-Lyzer mini dialysis tubes (Thermo Scientific, 88401).Freeze dried Atto655 NHS-ester dye (Sigma-Aldrich, 76245) was dissolved in a small volume of acetonitrile, mixed in a dye to protein ratio of 2:1 mole/mole and incubated at room temperature for 30 min.Unbound dye was removed using Micro Bio-Spin 6 columns (Bio-Rad) against deionized water.The absorbance at 280 nm and 663 nm were measured and used to calculate the degree of labeling (DOL).

Droplet formation and maturation
Protein droplets of NT2RepCT YF were formed by mixing 25 µM protein in 0.5 M potassium phosphate pH 8 in low binding tubes (Axygen).Samples have been dispensed into black 96-well low-binding polystyrene microplates with a transparent bottom (Corningand analyzed directly after 30 min sedimentation time and after maturation at 37°C for 24 h or at room temperature for at least 72 h if not stated otherwise.For maturation experiments plates have been sealed with cover film to avoid evaporation.Droplet formation of hTau was induced by buffer exchanging 122 µM hTau (stored in 20 mM sodium phosphate, pH 7.4 with 0.2 mM EDTA) to 25 mM ammonium acetate pH 8 and further dilution to 10 µM in deionized water and addition of 8% PEG.

Recruitment of protein constructs into droplets
25 μM NT2RepCT YF were mixed with 0.25 µM Atto655-labeled protein constructs and droplet formation was induced by addition of potassium phosphate to a final concentration of 0.5 M. Samples were prepared in black 96-well polystyrene microplates with a transparent bottom (Corning).Fluorescence microscopy images were acquired using a Nikon Eclipse Ti series inverted microscope (Nikon) equipped with a Crest X-light V2 series confocal unit (Nikon), a Plan Apo 40x objective (Nikon) and a Zyla VSC camera (Andor).A 640 nm laser was used, and laser power and exposure were kept constant throughout each experiment.Images were processed using ImageJ.Protein recruitment was analyzed by acquiring four separate images and plotting the fluorescence intensity of around 150 individual droplets against the circular droplet area.For GFP labeled proteins, differences in GFP fluorescence have been compensated by adjusting the added protein concentration according to the fluorescence emission intensity of GFP only.To test if GFP-tagged proteins are trapped in droplets, droplets have been matured for 72h at room temperature and subsequently washed three times with MQ.

Fluorescence emission spectroscopy and ThT kinetics
Thioflavin T (ThT) and pentameric formyl thiophene acetic acid (pFTAA) fluorescence emission was recorded using a SPARK 20M plate reader (Tecan).ThT kinetics were monitored using an excitation wavelength of 448 nm, an emission wavelength of 485 nm, 5 nm bandwidths and a gain of 150.Samples containing 25 µM NT2RepCT YF were prepared with 10 µM ThT in either deionized water or 0.5 M potassium phosphate pH 8.0 buffer.Five replicates, each containing 40 µl of reactant volume, were dispensed into black 386-well low-binding polystyrene microplates with a transparent bottom (Corning) and sealed with transparent cover film to avoid evaporation during incubation at 37°C.To measure pFTAA fluorescence protein samples were supplemented with 0.4 µM pFTAA prior to maturation in either deionized water or 0.5 M potassium phosphate pH 8.0.Emission spectra between 470 nm and 670 nm were recorded in 1 nm increments at 25 °C using an excitation wavelength of 455 nm, 7.5 nm bandwidths and a gain of 170.As amyloid control, amyloid- fibrils were generated from a 3 µM monomeric amyloid- solution in 20 mM sodium phosphate pH 8 with 0.2 mM EDTA.

Fluorescence microscopy of small molecules and FRAP
For fluorescence microscopy and FRAP experiments samples have been supplemented with either 10 µM ThT, 25 µM DroProbe, 25 µM Mitoxantrone, 25 µM Myricetin or 25 µM Riboflavin prior to droplet formation.Bright-field and fluorescence microscopy images in Figure 1 were acquired using a Nikon Eclipse Ti series inverted microscope (Nikon) equipped with a Crest X-light V2 series confocal unit (Nikon) using a Plan Apo 40x objective (Nikon) and a Zyla sCMOS camera (Andor).To compare different time points, laser power and exposure were kept constant throughout each experiment.
FRAP experiments were performed on a LSM980-Airy microscope (Zeiss) equipped with an Airy detector 2 using a C-Apochromat 40X water objective.FRAP excitation and emission settings are shown in Table 1.For all experiments the pinhole was set to 49 µm.For photobleaching a circular region of 1 µm diameter has been selected in which the the laser power was increased for 50 iterations and the scan speed lowered from 7 to 4 during bleaching.Fluorescence recovery was recorded using the same settings as before bleaching.Mitoxantrone partitioning inside droplets was visualized on the LSM980-Airy microscope performing a stack with 0.69 µm slices through the droplet.

Autofluorescence microscopy
Autofluorescence of fresh and matured NT2RepCT YF droplets, and hTau has been performed using the same microscope and objective that has been used for FRAP experiments.Droplets were excited using 405 nm (30% laser power), 561 nm (50% laser power) or 639 nm (100% laser power) wavelengths and the emission range was set to 431 nm -560 nm, 572 nm -736 nm and 644 nm -739 nm, respectively.The pinhole was set to 49 µm, the master gain to 850 V and the digital gain to 1.To compare different samples all settings were kept constant throughout each experiment, only the laser power has been adjusted dependent on the laser line.Images were processed using ImageJ.
Live cell imaging E.coli cell cultures suspended in growth media were placed on a microscope coverslip coated with poly-D-Lysine (A3890401 Thermo Fisher Scientific).Images were acquired using a Nikon Ti-E inverted microscope equipped with a 60×/1.4objective lens, and Crest OpticsX-lightV3 spinning disk confocal head (operated in the widefield mode).The fluorescence signal of eGFP-NT2RepCT was excited by continuous illumination on samples using 470 nm light from an LDI Laser Diode Illuminator (89 North) at 1% power level with an exposure time of 50 ms, while emission light between 485 nm and 535 nm was collected.A Gataca Systems iLas 2 unit coupled to a 100 mW OBIS LX 405 nm laser was used to generate a circular spot with a diameter of approximately 1 μm for photobleaching.The photobleaching on samples was carried out at a fixed 4% laser power level and at a controlled exposure time of 0.1 s.After photobleaching, the fluorescence images of samples were acquired using the above setting but with an exposure time of 50 ms to minimize imaging-induced photobleaching.The same setting of FRAP was used for E.coli cells grown at 18°C and 37°C.Processing of the images and quantification of the fluorescence recovery were carried out using the Fiji (2.3.0)software with a plugin developed by Jay Unruh at the Stowers Institute for Medical Research (Kansas City, MO). 5

Sample preparation for native mass spectrometry
To study the exchange of molecules between the soluble and dense phase of NT2RepCT YF by native MS, the following samples have been prepared.A fully soluble protein sample was prepared by mixing 100 µM NT2RepCT YF and 10 µM 15 N-labeled NT2RepCT YF in 100 mM ammonium acetate pH 8.This sample was directly analyzed by nMS.Phase separation of 100 µM NT2RepCT YF was induced in 0.35 M potassium phosphate buffer pH 8 in low binding tubes (Axygen) and 50 µl dispensed in a low-bind 96-well plate.Subsequently, 15 N -labeled NT2RepCT YF was added to a final concentration of 10 µM directly to the supernatant of phase separated NT2RepCT YF (fresh) or after NT2RepCT YF condensates have been aged for 72 h at RT (gelated).Immediately after addition of 15 N-labeled NT2RepCT YF samples have been collected in low-bind tubes and centrifuged at 15000xg for 2 min at 25°C.The supernatants of fresh and gelated samples were collected, and buffer exchanged to 100 mM ammonium acetate pH 8 using Zeba Spin desalting columns, 7k MWCO (Thermo Fisher Scientific) and diluted 1:1 with 100 mM ammonium acetate pH 8 prior to MS analysis.The pellet from freshly formed condensates has been resuspended in 100 mM ammonium acetate and 5% acetonitrile added prior to MS analysis.The pellet sample of matured condensates was not amendable to MS analysis.

Native Mass Spectrometry
Native mass spectra were acquired on a Waters Synapt G1 travelling wave ion mobility mass spectrometer (Waters, UK) equipped with an offline nanospray source.The capillary voltage was set to 1.5 kV, the source pressure was maintained at 7.5 mbar, the sampling cone voltage set to 20 V, the source temperature was 30 °C, and the collision energy in the ion trap was 30 V. Spectra were visualized using MassLynx 4.1 (Waters, UK).

Supporting Figures
Figure S1.(a) and (b) Solubility of NT2RepCT YF .The protein was incubated in 0.35 M KPO 4 , pH 8 at concentrations of 100 M or 10 M.The dense (D) and soluble (S) fractions were separated by centrifugation and analyzed by SDS-PAGE.(c) Quantification of the band intensities (shown in rectangles in a and b) from the SDS-PAGE.At 100 M, approximately 90% of the protein was in the dense phase, while at 10 M, > 95% remained in the soluble phase.(d) Mass spectra of NT2RepCT YF supernatant spiked with 15 N-labeled NT2RepCT YF for non-droplet conditions, for fresh droplets, and for gelated droplets (top to bottom).

Figure S2 .
Figure S2.(a) ThT fluorescence traces for individual fresh droplets (25 µM) after photobleaching at a wavelength of 405 nm with laser intensities set to 25, 50, and 100%.Error bars indicate the standard deviation of three independent repeats.The bleaching regime is indicated by a grey bar.(b) Photobleaching of htau droplets stained with Atto655 -labeled htau (top row), or ThT (bottom row) using the same maximum intensity laser pulse settings as for gelation of NT2RepCT YF .FRAP data show complete bleaching and slow fluorescence recovery, with no intensity overshoot.Scale bars are 2.5 m.(c) Droplet formation by a truncated spidroin.25 µM NT2Rep (without CTD) forms droplets in 0.7 M KPO 4 , pH8.NT2Rep droplets display the same overshoot in ThT fluorescence after photobleaching as NT2RepCT YF , indicating that gelation is driven by thermal denaturation of the NTD.Scale bar is 2.5 m.Fluorescence traces (right) are shown for a bleached droplet (blue trace) and a neighboring, unbleached droplet (gray trace).(d) NT2RepCT YF condensates are liquid prior to photobleaching, as judged by droplet fusion observed by brightfield microscopy.Scale bar is 2.5 m.

Figure S3 .
Figure S3.(a) Confocal microscopy of fresh (top row) or gelated NT2RepCT YF droplets (middle row) shows red-shifted fluorescence upon excitation at  405 and 561 nm, (emission  431 nm -560 nm and  572 nm -736 nm, respectively) which is significantly less pronounced upon excitation at  639 nm (emission  644 nm -739 nm).Droplets of htau (bottom row) exhibit prominent fluorescence at  405, but not at  561 nm or  639 nm.(b) FRAP analysis shows photobleaching at 405 nm and subsequent fluorescence overshoot, suggesting an increase in auto-fluorescent amyloid-like structures.Scale bars are 5 µm.All images are shown with the same color intensity scale below.

Figure S4 .
Figure S4.(a) Photobleaching of wild-type NT2RepCT in the presence of Mitoxantrone shows post-bleach fluorescence overshoot as detected for NT2RepCT YF .Scale bar is 5 m.(b) Photobleaching a droplet for a second time (left) does not elicit another fluorescence intensity overshoot, but shows normal fluorescence recovery (fluorescence intensity plot, right).Scale bars are 2.5 m.(c) Photobleaching of an individual droplet (arrow) induces resistance to 1,6-hexanediol.Brightfield and fluorescence microscopy before (left) and after (right) addition of 10% 1,6-hexanediol are shown.(d) Photobleaching of an individual droplet (arrow) induces resistance to 1,6-hexanediol and formic acid.Brightfield microscopy before (left) and after (right) addition of 10% 1,6-hexanediol and 10% formic acid are shown.Scale bars are 5 µm.(e) Z-stacks of gelated NT2RepCT YF droplets show uniform loading with Mitoxantrone.(f) Native MS analysis of 10 M NT in the presence of 100 M Mitoxantrone shows a minor population of protein-drug complexes, indicated by an arrow in the zoomed insert showing the 7+ charge state.

Figure S5 .
Figure S5.(a) Fluorescence microscopy images show the effects of increased laser power on the integrity of E. coli, indicating significant heating from high-energy FRAP.Top row: E. coli with intracellular mitoxantrone (white) before, during, and 2 s after a 100 ms laser pulse with 10% power shows no effect on cell integrity.Bottom row: Increasing the laser power to 20% results in cell lysis.Scale bars are 2 M.(b) Intracellular eGFP-NT2RepCT condensates formed at 37°C show no recovery after photobleaching (left), whereas those formed at 18°C do, indicating liquid-like properties.The standard deviations of each condition are plotted in the form of a shaded area.At least four independent repeats were performed for both expression temperatures.(c) Images of the eGFP-NT2RepCT condensates used to quantify Mitoxantrone colocalization in Figure 5 in the main manuscript.(d) Overlay of the Mitoxantrone (red) and NT2RepCT-eGFP (green) fluorescence images in Figure 5 b with phase contrast images showing the outlines of individual E. coli cells.