Figure S1.
Device characterization and phase separation conditions. (A–C) Characterization of the diffusion profile inside the microfluidic device. In such laminar flow mixers (MI and MII; Fig. 1), mixing occurs only by diffusion (Gambin et al., 2011; Lemke et al., 2009). To illustrate how such a concentration profile looks, A shows a finite element simulation performed in COMSOL Multiphysics for the second mixer MII. A diffusion coefficient of 10−7 cm2/s (≈ expected diffusion coefficient for molecules of the size of Importin β) was used for the biomolecules perfused in the cargo inlet channels. We also characterized the diffusion profile experimentally: to empirically calculate the concentration of cargo in the center of the main channel, we flowed fluorescently labeled Importin β in the cargo inlet and measured its concentration at different points in the center of the channel. B shows an image of the microfluidic device when only fluorescently labeled Importin β was loaded into the cargo inlet (scale bar is 50 µm), while from mixer MI, only PBS was flowing in the MII region. The red lines indicate positions at which we calculated the ratio of fluorescent Importin β in the center of the channel versus the periphery. The resulting ratio, plotted into C, revealed a slight increase from ≈15% to 20% over time/distance traveled. This concentration profile was consistent with the results from the COMSOL simulation. (D) Quantification of the FG-Nup49 concentration inside the droplets. Five different dye concentration solutions were first introduced inside the device to obtain an internal calibration curve that was used to estimate the concentration of the droplets based on their fluorescent intensity. For the quantification, 50 individual droplets were analyzed. As the droplets can differ in size, we compared the protein concentration in relation to droplet area (as segmented in ImageJ) and calculated the corresponding percentage polypeptide according to Hinz (2012). (E) Experimental phase separation conditions on a coverslip assay. Fluorescence images of phase-separated FG-Nup49 particles formed in presence of different concentration of TMAO or crowder (PEG 8000; scale bar is 10 µm). Lyophilized FG-Nup49 was dissolved in a small volume of stock buffer (4 M GdmHCl and 1× PBS, pH 7.0) and rapidly diluted to a final concentration of 10 µM into 1× PBS containing different concentrations of TMAO or PEG, which yielded similar droplets that were allowed to settle on a coverslip. TMAO was chosen for droplet experiments on the microfluidic device, as it has faster diffusive properties. As in the device experiments, here also FG-Nup49 labeled with Cy5 was added to unlabeled FG-Nup at a very small concentration (≈20 nM) to visualize the self-assembled particles. Images were taken on a confocal microscope, Olympus FV3000, using a 63× oil-immersion objective.

Device characterization and phase separation conditions. (A – C) Characterization of the diffusion profile inside the microfluidic device. In such laminar flow mixers (MI and MII; Fig. 1), mixing occurs only by diffusion (Gambin et al., 2011; Lemke et al., 2009). To illustrate how such a concentration profile looks, A shows a finite element simulation performed in COMSOL Multiphysics for the second mixer MII. A diffusion coefficient of 10−7 cm2/s (≈ expected diffusion coefficient for molecules of the size of Importin β) was used for the biomolecules perfused in the cargo inlet channels. We also characterized the diffusion profile experimentally: to empirically calculate the concentration of cargo in the center of the main channel, we flowed fluorescently labeled Importin β in the cargo inlet and measured its concentration at different points in the center of the channel. B shows an image of the microfluidic device when only fluorescently labeled Importin β was loaded into the cargo inlet (scale bar is 50 µm), while from mixer MI, only PBS was flowing in the MII region. The red lines indicate positions at which we calculated the ratio of fluorescent Importin β in the center of the channel versus the periphery. The resulting ratio, plotted into C, revealed a slight increase from ≈15% to 20% over time/distance traveled. This concentration profile was consistent with the results from the COMSOL simulation. (D) Quantification of the FG-Nup49 concentration inside the droplets. Five different dye concentration solutions were first introduced inside the device to obtain an internal calibration curve that was used to estimate the concentration of the droplets based on their fluorescent intensity. For the quantification, 50 individual droplets were analyzed. As the droplets can differ in size, we compared the protein concentration in relation to droplet area (as segmented in ImageJ) and calculated the corresponding percentage polypeptide according to Hinz (2012). (E) Experimental phase separation conditions on a coverslip assay. Fluorescence images of phase-separated FG-Nup49 particles formed in presence of different concentration of TMAO or crowder (PEG 8000; scale bar is 10 µm). Lyophilized FG-Nup49 was dissolved in a small volume of stock buffer (4 M GdmHCl and 1× PBS, pH 7.0) and rapidly diluted to a final concentration of 10 µM into 1× PBS containing different concentrations of TMAO or PEG, which yielded similar droplets that were allowed to settle on a coverslip. TMAO was chosen for droplet experiments on the microfluidic device, as it has faster diffusive properties. As in the device experiments, here also FG-Nup49 labeled with Cy5 was added to unlabeled FG-Nup at a very small concentration (≈20 nM) to visualize the self-assembled particles. Images were taken on a confocal microscope, Olympus FV3000, using a 63× oil-immersion objective.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal