Evolution and Single-Droplet Analysis of Fuel-Driven Compartments by Droplet-Based Microfluidics

Abstract

Active droplets are a great model for membraneless organelles. However, the analysis of these systems remains challenging and is often limited due to the short timescales of their kinetics. We used droplet-based microfluidics to encapsulate a fuel-driven cycle that drives phase separation into coacervate-based droplets to overcome this challenge. This approach enables the analysis of every coacervate-based droplet in the reaction container throughout its lifetime. We discovered that the fuel concentration dictates the formation of the coacervate-based droplets and their properties. We observed that coacervate-based droplets grow through fusion, decay simultaneously independent of their volume, and shrinkage rate scales with their initial volume. This method helps to further understand the regulation of membraneless organelles, and we believe the analysis of individual coacervate-based droplets enables future selection- or evolution-based studies.

Keywords: Artificial Organelles; Droplet-Based Microfluidics; Nonequilibrium Processes; Phase Transitions.

Figure 1

Schematic representation of the reaction cycle and the microfluidic setup. a) Reaction cycle of the precursor Ac‐F(RG)₃D‐OH with EDC. EDC converts the +1 charged precursor to the +3 charged anhydride product. The product can then hydrolyze back to the precursor. This cycle continues until EDC is depleted. b) Formation and dissolution of coacervates in a confined volume depending on the reaction cycle. The product (red) forms coacervates with polystyrene sulfonate (pSS). Once the product is hydrolyzed back to the precursor (blue), the coacervates dissolve. c) Schematic representation of the microfluidic chip that is used for microfluidic droplet formation and trapping of these droplets in dropspot chambers. EDC and precursor solutions are supplied from two different channels and are mixed right after the formation of the microfluidic droplets. Upon stopping the flow, the microfluidic droplets are trapped in the dropspot chambers. In these chambers, the microfluidic droplets are imaged via confocal time‐lapse imaging by excitation of sulforhodamine B for coacervate‐based droplets with pSS or Cy3‐A15 for coacervate‐based droplets with pU at 552 nm. The pseudocolor‐coded confocal image represents a maximum z‐projection of a z‐stack throughout one microfluidic droplet. The grey value scale from 0 to 255 is given next to the image. The grey line represents the periphery of the microfluidic droplet. The scale bar represents 20 μm.

Figure 2

Comparison of the behavior of coacervate‐based droplets in the bulk and the microfluidic setup. Conditions are 23 mM Ac‐F(RG)₃D‐OH, 4.1 mM pU, 0.1 μM Cy3‐A15 and 200 mM MES at pH 5.3 with 25 mM EDC. a) Representative images over one cycle of coacervate assembly and disassembly in a 33 pl water droplet entrapped in a microfluidic dropspot chamber. Images are recorded by excitation of Cy3‐A15 at 552 nm. The pseudocolor‐coded confocal image represents a maximum z‐projection of a z‐stack throughout one microfluidic droplet. The grey value scale from 0 to 255 is given next to the image. The scale bar represents 20 μm. b, c) Comparison of the average and the total volume of pU droplets between the microfluidic and the bulk setup. The total volume is given as volume percent, defined as the total volume of coacervate‐based droplets divided by the volume of the microfluidic droplet or the imaged volume for the bulk setup. Error bars represent the standard deviation of 5 independent experiments.

Figure 3

Analysis of coacervate‐based droplet formation. All experiments are performed with conditions of 8 mM Ac‐F(RG)₃D‐OH, 5 mM pSS, 0.1 μM sulforhodamine B, and 200 mM MES at pH 5.3 with varying fuel concentrations. a) Representative images of coacervates’ initial formation and growth when 20 mM of EDC is added. Images are recorded by excitation of sulforhodamine B at 552 nm. The images show one z‐plane in the middle of a microfluidic droplet. The scale bar represents 20 μm. b) Number of coacervate‐based droplets depending on the amount of EDC added. Coacervate‐based droplets are counted in one z‐plane in the middle of a microfluidic droplet over the first 3 min after the start of the reaction cycle. Error bars represent the standard deviation of at least 9 measurements from 3 independent experiments. c) The time the first coacervate‐based droplet could be detected (t _nuc) is shown as a function of the amount of EDC added. Error bars represent the standard deviation of at least 9 measurements from 3 independent experiments. The orange dotted line represents the nucleation times calculated by the kinetic model depending on the EDC concentration. The red line represents the EDC concentration below which no fuel‐driven nucleation of coacervate‐based droplets is possible according to the kinetic model. d) Anhydride concentration for different EDC concentrations at the time of nucleation. The kinetic model calculates anhydride concentrations. The orange dotted line represents the average anhydride concentration needed to nucleate coacervate‐based droplets. Error bars are calculated from the standard deviation of the nucleation times.

Figure 4

Analysis of the total volume of coacervate‐based droplets and their viscosity. All conditions are 8 mM Ac‐F(RG)₃D‐OH, 5 mM pSS, and 200 mM MES at pH 5.3 with varying fuel concentrations. a) Images of the coacervation cycle showing the maximum amount of coacervate volume at different EDC concentrations. Images are recorded by excitation of sulforhodamine B at 552 nm. The pseudocolor‐coded confocal images represent a maximum z‐projection of a z‐stack throughout one microfluidic droplet. The grey value scale is given next to the images. The scale bar represents 20 μm. b) Analysis of the total volume of coacervate‐based droplets over the entire reaction cycle. The total volume is given as volume percent, defined as the total volume of coacervate‐based droplets divided by the total volume of the microfluidic droplet. Error bars represent the standard deviation of 3 experiments. c) Maximum total volume of coacervate‐based droplets as a function of the EDC concentration. Error bars represent the standard deviation of 3 experiments. d) The diffusivity of NBD‐labeled product inside of coacervate‐based droplets. Error bars represent the standard deviation of 9 experiments. e) Time series of droplet fusion. Fusion is drastically slower at increased EDC concentrations. The depicted time represents the time of the fusion event and not the actual time in the reaction cycle. Images are recorded by excitation of sulforhodamine B at 552 nm. The scale bar represents 2 μm.

Figure 5

Tracking of coacervate‐based droplets in a microfluidic droplet. a) A z‐projection image time series. All droplets that fuse are assigned to the same population and marked with the same color. A solid grey circle marks the periphery of the microfluidic droplet. The dotted circles mark the coacervate‐based droplet that is tracked under (d)–(f). b) The combined pathways of every droplet of the different populations. c) Volumes of every coacervate‐based droplet of the different populations detected in the microfluidic droplet. d)–f) The volume of one coacervate‐based droplet the population picked at 2.8 min and followed until its dissolution. A fusion event is highlighted. g) The same experiment as above, but with a blue‐fluorescent pS bead. The bead incorporates into a coacervate‐based droplet and remains incorporated until its dissolution. The grey value scale is shown next to the images. All scale bars represent 20 μm. All conditions are 10 mM Ac‐F(RG)₃D‐OH, 5 mM pSS and 0.1 μM sulforhodamine B in 200 mM MES at pH 5.3.