1. Reagent setup
Benzophenone solution
To prepare the benzophenone solution used for hydrophilic surface treatment of the microfluidic chip, disperse benzophenone at 10 %wt in acetone. A gentle agitation for ten seconds is enough to obtain its complete solubilization. Due to its toxicity, proceed for the weighing and the solubilization under a fume hood and wear appropriate protective equipment. This solution has to be prepared before each surface treatment experiment. As acetone is highly volatile solvent, the transfer of this solution to the syringe used for its injection in the treated microfluidic chips must be realized as soon as the benzophenone is solubilized. 5 mL of the solution is enough to proceed to the treatment of about ten microfluidics chips.
Acrylic acid solution
The acrylic acid solution used for hydrophilic surface treatment of the microfluidic device is prepared by dilution of acrylic acid at 20 %wt in Milli-Q ultra-pure water. Acrylic acid is highly corrosive and flammable. Handle it under a fume hood and wear appropriate protective equipment. 5 mL is enough to proceed to the treatment of about ten microfluidics chips.
BSA/Biotinylated BSA solution
A mixture of BSA and biotinylated BSA is used to cover microfluidics chips of chemical anchor points for GUVs immobilization. To obtain it, solubilize by magnetic stirring for 1h, BSA and biotinylated BSA powders at respectively 90 %mol and 10 %mol in a 10 mM HEPES aqueous buffer, containing 95 mM of NaCl. Less than 500 µL of this solution is needed to proceed to the entire treatment of about ten microfluidic chips. This solution is stored at 4 °C for ~ 10 days.
Streptavidin solution
The streptavidin solution used for chemical immobilization of GUVs is prepared by solubilization of streptavidin powder at 25 µg/mL in a 10 mM HEPES aqueous buffer, containing 95 mM of NaCl. A magnetic stirring of 1h is sufficient to perform this solubilization. Less than 500 µL of this solution is needed to proceed to the entire treatment of about ten microfluidic chips. As recommended by the supplier, this solution is aliquoted and stored at -20 °C for several months or at 4°C for several weeks.
Inner Fluid phase (IF)
The inner fluid phase consists of Milli-Q ultra-purified water. IF have to be filtered with a 0.2 µm filter before injection to avoid dusts and bubbles, which can clog channels and led to changes in pressure and flow directions. Fluorescent probes such as calcein can be dispersed in IF for fluorescence microscopy visualizations. To do so, dispense the appropriate amount of calcein into water, so that the final concentration is 10 µM. To perform an experiment of GUVs formation with the microfluidic device, less than 200 µl of solution is needed. To avoid bacterial growth, it is recommended to store IF at 4°C.
Middle fluid phase (MF)
The middle fluid phase is composed of DOPC, DOPG, DSPE-PEG-biotin lipids, dispersed in oleic acid with the respectively molar ratio 98.9:1:0.1 and a total lipid concentration of 6.5 mM. The following stock solutions are prepared by solubilization of lipids in oleic acid:
- Dissolve DOPC (MW = 786.113 g/mol) to 10 mM.
- Dissolve DOPG (MW = 797.026 g/mol) to 1 mM.
- Dissolve DSPE-PEG-Biotin (MW = 3016.781 g/mol) to 0.1 mM.
Note: Use an ultrasonic bath at the maximum amplitude for 2 h at 30 °C to completely dissolve lipids within oleic acid.
Nile red (10 ppm) hydrophobic fluorescent probe can be added to the lipid mixture for fluorescence microscopy experiments to visualize the MF. In this case, the mixture is sonicated for 2 h at 30 °C in the bath sonicator after the probe addition.
These lipid stock solutions can be stored at -20°C for months. Then, mix the stock solutions to a molar ratio DOPC:DOPG:DSPE-PEG-biotin of 98.9:1:0.1. This final lipid dispersion can be used as MF for GUVs preparation, after sonication at 30°C for 2h. It can be stored and used to produce stable emulsion droplets, for several weeks at room temperature.
Outer Fluid phase (OF)
The outer fluid phase consists of a mixture of 10 %wt ethanol and 20 %wt glycerol. Glycerol is used to increase the viscosity of the OF, which increases the w/o/w emulsion droplets formation at the second flow focusing junction. Ethanol is used to solubilize the oleic acid of the double emulsions droplets that enhances the GUVs formation in the trapping chamber. To prepare it, mix ethanol and glycerol in the suitable proportions and proceed to an overnight gentle agitation of the mixture. The resulting solution can be used to produce stable emulsion droplets for one month by storing it at 4 °C. Other chemicals components can be added in the OF phase. For example, amphiphilic copolymers, such as Pluronics, can be used to enhance droplets stability (Chariot et al., 2014; Deng, Yelleswarapu, & Huck, 2016; Deshpande et al., 2016; Teh et al., 2011). Buffer molecules can also be added to this phase to control the vesicles pH.
α-Hemolysin (α-HL) solution
To prepare the α-hemolysin solution, used for pore formation within lipids bilayers, disperse it at 2 mM in ultrapure water, or in its desired aqueous buffer of use, by magnetic stirring for 1 h. The resulting solution has to be filtered with a 0.2 µm filter to remove α-HL aggregates before its injection in a microfluidics device. This solution is aliquoted and stored at -20 °C in dark conditions for several months. Each aliquot can be used in a day after defrosting to avoid α-HL aggregation or degradation.
2. Equipment setup
The setup for GUVs production, immobilization, and microscopic analysis is outlined in Figure 7. Briefly, the microfluidic chip, presented in Figure 2, is placed on a sample holder of an inverted microscope, connected to fluorescent and light sources and to a camera linked to a computer. GUVs are produced in this device by injections of IF, MF and OF phases contained in syringes placed on three different syringe pumps. These syringes are connected to the inputs of the microfluidic device using Tygon tubing, luer stubs, and stainless steel couplers. Microscopic images of droplets and channels are recorded by the use of an image acquisition software.
Note:An inverted microscope is more suitable than a straight microscope since it provides large free space on the top of the chips. This is suitable for the tubing inputs of the microfluidic device.
CAUTION! For better reproducibility in the droplets production, it is recommended to keep fixed the position of the syringes, the length of the tubing, and the position of the microscope from one experiment to another.
3. Procedure
For a sake of clarity, we first present the procedure concerning the production and use of "classical" PDMS/PDMS chips. The optimization of this procedure for CLSM and UV-autofluorescence microscopy (PDMS/glass, PDMS/quartz chips) is presented at the end of this part in IV.7.
3.1. Mask design • TIMING ~1-7 days
1) Use a Computer-Aided Design (CAD) drawing software such as Adobe Illustrator to design the patterns of the microfluidic device.
Note: The dimensions of the microfluidic circuit drawing are detailed in the supplementary information part in Figure S1.
2) Save the drawing in negative format as vector graphic (e.g. .ai file extension). The channels drew in black on white have to appear in white on black.
▲ CRITICAL STEP A vector graphic format of the drawing is essential to obtain a high-resolution mask in the next steps.
3) Order this drawing to a printing service to obtain an opaque black mask on which the channels appear transparent.
▲ CRITICAL STEP Several masks from different printing services were tested in this study. It was observed that a minimum printing resolution of 3000 d.p.i. is necessary for the fabrication of the small patterns of the microfluidic chip. Furthermore, the mask must be completely opaque at its black areas; otherwise, unwanted partial crosslinking of the SU-8 resin could take place during the UV-insolation. The opacity of the mask can be measured with a reflectometer. We used a mask printed at 4000 d.p.i. by the company Micro Lithography Services (U.K.). This mask presents a good resolution and a high level of opacity of the black areas.
3.2. Microlithography - master fabrication • TIMING ~1h30
▲ CRITICAL STEP The presence of dust in the working environment in steps 4-25 seriously affects the quality of the obtained microfluidic chips. If possible, carry out these steps in a clean room (ISO 5-8). Otherwise, perform these experiments after removing dusts from surfaces and utensils used by passing a pressurized airflow and by working under a laminar flow hood.
4) Clean a 100 mm diameter silicon wafer by the use of a Kimtech wipe soaked in isopropanol and of a stream of pressurized air. This is done to eliminate the major part of the impurities that may be present on the surface.
5) Protect the inside of the spin coater against resin splashes with a protection liner. Then, center the wafer into the spin coater, activate the vacuum to fix it and proceed to the spin coating of approximatively 10 g of SU-8 2100 resin, at 2000 rpm for 35s with a ramp of 500 rpm.s-1. This leads to a resin thickness of ~ 140 µm as determined by profilometry experiments (data not shown). Until development of the master (step 10), the wafer with the resin must be protected as much as possible from any light source.
Note: Different resin thicknesses can be obtained by changing the spin coating duration. The longer the spin coating, the thinner the resin. Another more or less viscous resin can also be chosen to obtain a different thickness. For the same spin coating conditions (duration, speed), the more viscous the resin, the thicker it is.
▲ CRITICAL STEP A minimum resin thickness of 100 µm is needed to obtain well-defined traps geometries on the master in the later steps (resin development).
? TROUBLESHOOTING
6) Immediately, proceed to the prebaking of the spin-coated wafer by placing it on a 65°C ceramic hotplate for 5 min. Then, place it on a 95°C ceramic hot plate for 20 min.
7) Cool down the spin-coated wafer to room temperature and place it on the wafer support of a photolithography UV insolator (e.g. UV KUB 2).
8) Place the mask obtained in step 3 in contact with the top of the spin-coated wafer and UV expose the resin through the mask at a power per surface of 35mW/cm² and a wavelength of 365 nm for 15 s.
▲ CRITICAL STEP The insolation duration and the position of the mask are critical for the resolution of the pattern. Optimal conditions were observed for exposition duration below 20 s and hard contact conditions (masking distance = 0 um). Longer exposition duration induces resin crosslink in unexposed areas.
? TROUBLESHOOTING
9) Immediately, proceed to the postbaking of the insolated spin-coated wafer by placing it on a 65°C ceramic hotplate for 5min. Then, place it on a 95°C ceramic hot plate for 10 min.
Note: The reticulated resin representing the design must be visible in transparency.
? TROUBLESHOOTING
10) Cool down the spin-coated wafer to room temperature and place it in a bath of SU-8 developer (~ 200 mL) under manual gentle agitation for ~ 35-45 min. Renew the bath after 15 min of development. Wash the well-developed wafer with isopropanol and pass a pressurized air stream to dry it and to obtain the desired master.
CAUTION ! SU-8 Developer is highly flammable, volatile and toxic. Manipulate it under a fume hood and wear appropriate protective equipments.
▲ CRITICAL STEP The increase in development duration induces better solubilization of the non-photocrosslinked resin. However, for durations > 45min, the crosslinked resin may crack. It is necessary to check the development of the resin when designing very small patterns, by successive observations with a stereoscopic microscope.
? TROUBLESHOOTING
☐ PAUSE POINT The resulting master can be stored indefinitely at room temperature in a plastic Petri dish. Dust must be removed with a stream of pressurized air before the following steps.
3.3. PDMS molding – microfluidic chip fabrication • TIMING ~8h
11) Place the master obtained in step 10 on a piece of aluminium foil in a plastic Petri dish. Arrange the foil to create borders around the master; this will be used to contain the poured PDMS in the next steps.
12) Prepare two PDMS/crosslinker mixtures in disposable beakers, a first 30 g batch with 10 % wt of crosslinker and another of 20 g with 5 % wt of crosslinker. Each batch is vigorously mixed with a rigid utensil such as a pipette transfer to obtain a homogenous mixture. The resulting mixture is turbid due to the presence of incorporated air bubbles which will be removed under vacuum in step 14.
CAUTION ! A surface covered with uncrosslinked PDMS is very difficult to clean. It is therefore recommended to protect the surfaces and to wear appropriate protective equipments.
▲ CRITICAL STEP It is recommended to vigorously mix this mixture for at least 2-3 minutes. The insufficient mixing gives inhomogeneous crosslinked PDMS after the subsequent heating step. When insufficient crosslinked, PDMS can clog the channels during the assembly of the chip pieces. Contrarily, if it is excessively crosslinked, it cannot be stuck to the PDMS bottom (step 20) leading to leaks problems. The resolution of the pattern was found to be sensitive to PDMS storage conditions. We recommend to aliquot it in a few hundred grams. Store it at 4°C and use each aliquot rapidly after opening (~ 1-4 weeks).
? TROUBLESHOOTING
13) Pass a pressurized airflow on the surface of the previously prepared master to eliminate dusts.
14) Under a laminar flow hood, pour the 10 %wt PDMS mixture into the master and the 5 %wt PDMS mixture into a Petri dish before placing them in a vacuum desiccator for 2 h, at pressure around 30 mBar, until the bubbles disappear completely.
▲ CRITICAL STEP The presence of bubbles on the surface of the PDMS at the end of this step must be avoided. It leads to defects in the finals microfluidics devices, which can generate visual distortions of the patterns during microscopic observations.
? TROUBLESHOOTING
15) Heat the master and the Petri dish, which contains PDMS in a 70°C oven for 20 min. Then let them cool to room temperature for 5 min under a laminar flow hood.
▲ CRITICAL STEP The resulting PDMS does not have to be completely crosslinked. You must be able to bend it on itself without breaking or deforming it. As mentioned in the critical step of the point 12, the optimal crosslinking of PDMS is essential to obtain a usable microfluidic device.
? TROUBLESHOOTING
16) Under a laminar flow hood, cut with a scalpel and gently peel the PDMS slab from the master.
? TROUBLESHOOTING
17) Using tape, remove the small residual pieces of PDMS that remained stuck to the cut PDMS slab.
18) Using a biopsy puncher (ID = 0.50 mm), create holes in the inlets and outlets patterns of the PDMS slab placed flatly on a cutting mat.
▲ CRITICAL STEP Punching must be done vertically in one go and the punched part must absolutely be completely removed from the PDMS slab. A non-vertical punching of the inlets and outlets can lead to leaks during the use of the microfluidic chips. In addition, the presence of residual PDMS pieces can block the channels of the chips.
? TROUBLESHOOTING
19) Using tape, remove the small residual pieces of PDMS that remained stuck after punching.
20) Put the PDMS slab in contact with the crosslinked PDMS that remained in the Petri dish and place them in a 70°C oven for 1 hour.
▲ CRITICAL STEP Position the PDMS slab carefully on the other PDMS piece from one edge to another.
Note: The patterns must be visible highlighted in backlight.
? TROUBLESHOOTING
21) Let cool the resulting two-part PDMS microfluidic chip to room temperature for 5 min, under a laminar flow hood.
22) Insert stainless steel plugs (ID = 0.5 mm) in the punched inlets and outlets.
▲ CRITICAL STEP For the same reason of the critical step of the point 18, the insertion of the plugs must be done vertically in one go.
? TROUBLESHOOTING
23) Cut pieces of plastic transfer pipette (~ 5 mm of high) using a scalpel and place it around the plugs inserted in the inlets and outlets to create reinforcements.
▲ CRITICAL STEP Make straight cuts to get stable reinforcements once the pipette pieces will be filled with PDMS in the next step.
24) Cut a plastic transfer pipette at ~ 2 cm of its end to make it wider. Use it to pour the PDMS mixture containing 5% wt of crosslinker in the pieces of plastic placed around the plugs.
▲ CRITICAL STEP Do not touch the plugs while filling in the uncrosslinked PDMS.
? TROUBLESHOOTING
25) Immediately, place the microfluidic chip in a 70°C oven for 2 h to proceed to the crosslinking of the reinforcements.
? TROUBLESHOOTING
☐ PAUSE POINT The resulting microfluidic chip can be stored indefinitely at room temperature in a dust-free closed plastic box. A schematic and a photograph of one of these resulting circuits are available in Figure 6A.
3.4. Hydrophilic surface treatment • TIMING ~6 h
Note: For a sake of clarity, we describe here the hydrophilic surface treatment of a single microfluidic device. This treatment can be carried out in series by connecting the output of a first one microfluidic circuit to the input of a second one. From our experience, carrying out a series processing is efficient for a maximum number of eight circuits. Beyond this number, the chips at the end of the series are poorly treated and remain hydrophobic.
▲ CRITICAL STEP The inlets and outlets of the PDMS microfluidic devices are fragile. Do not disconnect tubing unnecessarily from step to step. Only disconnect the syringe when possible.
26) Remove the stainless-steel plugs from the OF phase inlet and the trapping chamber outlet (identified by areas (1) and (6) in Figure 2).
? TROUBLESHOOTING
27) Under a hood, fill a syringe with 5 mL of a solution of benzophenone dispersed at 10% wt in acetone and install it on a syringe pump.
CAUTION ! As indicated before, the benzophenone solution is toxic, flammable and volatile. Stay under the hood for the rest of the surface treatment experiment and wear appropriate protective equipments. Take immediately the solution into the syringe after its preparation to avoid concentration of benzophenone by acetone evaporation.
28) Cut two portions of PTFE tubing with a length of approximately 20 cm.
29) Connect one of these tubing to the OF inlet released in step 26 and to the syringe using a stainless-steel coupler and a 22ga luer stub.
30) Connect the other tubing to the outlet of the trapping chamber released in step 26 using a stainless-steel coupler and place its free end in an empty glass beaker.
31) Inject the benzophenone into the microfluidic chip at a flow rate of 200 µL/min for 10 min.
▲ CRITICAL STEP It is important that the treatment solutions injected into the device as shown in Figure 4A, do not flow to the area before the 2nd flow focusing junction.
? TROUBLESHOOTING
32) Disconnect the syringe from the inlet tubing of the microfluidic circuit by disconnecting the luer stub. Under a hood, connect this same luer stub to a nitrogen generator and pass a stream of nitrogen gas through the chip, to remove the benzophenone solution.
▲ CRITICAL STEP The benzophenone solution must be completely removed from the chips to obtain a good surface treatment. Visualise its progress towards the exit within the device during the injection of the nitrogen gas.
? TROUBLESHOOTING
33) Place the microfluidic circuit in a vacuum desiccator for 10 min to induce the diffusion of the residual benzophenone molecules within PDMS.
34) Under a hood, fill a syringe (Hamilton Gastight, U.S) with 5 mL of an aqueous solution of acrylic acid 20% wt.
CAUTION ! Acrylic acid is corrosive, so continue to wear appropriate protective equipments.
35) Connect the microfluidic circuit to the syringe filled with acrylic acid with the same tubing as used for the injection of benzophenone (connection to the OF inlet) and inject it into the device at 350 µL/min for 2 min, placing the free end of the outlet tubing in a beaker.
? TROUBLESHOOTING
36) Stop the injection of acrylic acid and immediately close the inlet and outlet of the device by removing the tubing and inserting plugs (ID = 0.5 mm) to keep this solution within the circuit.
▲ CRITICAL STEP It is important to close the circuit quickly to avoid acrylic acid loss. Insufficient acrylic acid in the device, during the next step of insolation, would lead to poorly treated chip.
? TROUBLESHOOTING
37) Place the microfluidic circuit within the UV KUB 2 insulator and proceed to its exposure at a power per surface of 35 mW/cm² and a wavelength of 365 nm for 5 min.
38) Remove the plugs from the processing inlet and outlet.
39) Fill a syringe with a 96% ethanol solution (~ 25 mL) and install it on a syringe pump.
40) Cut two portions of tygon tubing with a length of approximately 20 cm.
41) Connect one of these tubing to the OF inlet released in step 38 and to the syringe using a stainless-steel coupler and a 22ga luer stub.
42) Connect the other tubing to the outlet of the trapping chamber released in step 38 using a stainless-steel coupler and place its free end in an empty glass beaker.
43) Inject the ethanol solution into the microfluidic circuit at a flow rate of 350 µL/min for 1 hour.
? TROUBLESHOOTING
44) Similarly, inject an aqueous solution at pH 11, prepared using 4M NaOH, for 1 hour at a flow rate of 350 µL/min.
CAUTION ! Because NaOH is corrosive, continue to wear appropriate protective equipments.
☐ PAUSE POINT The microfluidic device is now treated. It can be stored for a month at 4 °C immersed in pH 11 water to maintain its surface properties.
? TROUBLESHOOTING
3.5. Surface treatment for chemical immobilization of vesicles • TIMING ~45min
Note: As for the hydrophilic surface treatment, for a sake of clarity, we describe here the surface treatment for chemical immobilization of vesicles of a single microfluidic device. This treatment can also be carried out in series with the same precautions.
45) Using stainless steel plugs, close the inlet of the IF, MF and OF phases of the microfluidic circuit (identified by areas (1) (2) and (3) in Figure 2).
46) Fill a syringe with a mixture of BSA/biotinylated BSA dispersed in a 10 mM HEPS buffer in the presence of 95 mM of NaCl.
47) Cut two portions of tygon tubing with a length of approximately 20 cm.
48) Connect one of these tubing to the first outlet identified by area (5) in Figure 2 and to the syringe using a stainless-steel coupler and a 22ga luer stub.
49) Connect the other tubing to the outlet of the trapping chamber identified by area (6) in Figure 2.
50) Inject the BSA mixture into the microfluidic circuit at a flow rate of 50 µL/min for 2 min.
? TROUBLESHOOTING
51) Remove the tubing from the two outlets and quickly close them with plugs to allow the BSA/biotinylated BSA mixture to incubate in the chip for 20 min.
? TROUBLESHOOTING
52) Fill a syringe with the HEPES buffer solution. Inject it into the microfluidic circuit at a flow rate of 200 µL/min for 2 min.
? TROUBLESHOOTING
53) In the same way as for steps 50 to 52, inject a solution of streptavidin dispersed in a 10 mM HEPES buffer in the presence of 95 mM of NaCl at a flow rate of 50 µL/min, let it incubate for 10 min. Wash the chip with Milli-Q ultrapure water injected at a flow rate of 200 μL/min for 2 min.
? TROUBLESHOOTING
☐ PAUSE POINT The resulting circuit is treated for chemical immobilization of vesicles. It can be stored for one day at room temperature when it is filled with ultra-pure water. Water evaporation will damage the surface treatment, so put the chip in a Petri dish and seal it with parafilm unless the prepared chip is used immediately.
3.6. Semi-permeable GUVs production and immobilization • TIMING ~ 30 min
54) Place the microfluidic device contained in a Petri dish on the support of an inverted microscope.
55) Inspect the microfluidic device to check that channels are free of PDMS residue, dust, or other dirt. Also, check if they are not clogged.
▲ CRITICAL STEP The presence of dirt or clogged channels in a microfluidic device can make it unusable.
? TROUBLESHOOTING
56) Fill three syringes with approximately 100 µL of the OF, MF, and IF phases.
Note: A few tens of microliters are sufficient for this experiment. Be careful to avoid the formation of air bubbles in the syringes during their filling. These bubbles can disturb the fluids' flows within the device.
57) Using Tygon tubing (ID = 0.5 mm), luer stub 22ga, and stainless-steel coupler 22ga, connect the syringes to their respective inlets (areas (1) (2) and (3) represented in Figure 2) and place them on 3 syringe pumps as shown in Figure 7.
58) Close the outlet of the trapping chamber (shown in area 6 in Figure 2) using a stainless-steel plug 22ga.
59) At the first outlet of the microfluidic device (represented in Figure 2 by area 5), connect a Tygon tubing using a coupler and place its other end in an empty beaker.
60) Prepare the inverted microscope for viewing the microfluidic device by switching on its illumination source (fluorescence or light) and adjusting the x4 objective.
61) Adjust the position of the microfluidic device so that you can observe the flow-focusing junction area (represented by area 4 in Figure 2).
62) Activate the OF phase syringe pump at a flow rate of 50 µ/min.
63) Check that this phase flows to the first outlet and not to the trapping chamber or to another inlet.
? TROUBLESHOOTING
64) Once the OF phase arrives at the flow-focusing area, activate the syringe pump of the MF phase at a flow rate of 10 µL/min. Then wait for the formation of an oil-in-water emulsion at the second flow-focusing junction.
? TROUBLESHOOTING
65) Activate the syringe pump of the IF phase at a flow rate of 5 µL/min to generate a double w/o/w emulsion.
Note: This induces the production of double emulsion droplets as represented in Figure 1A.
? TROUBLESHOOTING
66) Let the droplets go out of the microfluidic circuit for about 2 min.
Note: At the start of injections, the flow rates may be unstable. This is mainly due to the deformability of PDMS and the mechanical instability of the syringe pumps. It is therefore recommended to wait 2 min, for the flow rates stabilization.
67) Very quickly, open the outlet of the trapping chamber (represented by area 6 in Figure 2) by removing the inserted plug. Then, immediately remove the tubing from the outlet connected to the beaker (area 5 in Figure 2) and insert it into the outlet of the trapping chamber. Then, close the first outlet (area 5 in Figure 2) with a plug. This will direct the w/o/w emulsion droplets into the trapping chamber.
? TROUBLESHOOTING
68) Once droplets are trapped in the PDMS traps, as represented in Figure 1B, take off the IF and MF tubing and quickly close these inlets by using plugs.
? TROUBLESHOOTING
69) Inject the OF phase at 50 µL/min for 20 min into the trapping chamber filled with droplets of double emulsion.
Note: This induces the elimination of the MF phase of the double emulsion droplets through the traps, as illustrated in Figure 1B, and leads to the formation of GUVs.
? TROUBLESHOOTING
70) Fill a syringe with a 2 µM α-HL solution. 200µL of solution is sufficient for this experiment.
71) Open the first outlet of the microfluidic device (represented by area 5 in Figure 2) by removing the inserted plug.
72) Quickly connect the hemolysin syringe to the first outlet opened in the previous step.
▲ CRITICAL STEP The opening of the circuit, as well as the connection of the hemolysin syringe, must be operated as quickly as possible to avoid air bubbles and pressures modifications. These can lead to a leak of the liposomes from the traps by inducing a rearward movement.
Note: Chemical trapping of liposomes reduce the rearward movement.
? TROUBLESHOOTING
73) Using a syringe pump, initiate the injection of hemolysin at a rate of 50 µL/min.
? TROUBLESHOOTING
74) Quickly, stop the OF flow and disconnect its inlet tubing from the microfluidic circuit. Then close this inlet with a plug to direct the hemolysin solution in the trapping chamber.
? TROUBLESHOOTING
▲ CRITICAL STEP To avoid the release of the liposomes from the PDMS pads, it is essential to initiate the injection of hemolysin before stopping the OF phase flow. This allows a constant flow of fluid directed from the inlet of the storage chamber to its outlet. Thus, the liposomes are directed towards the traps.
75) The hemolysin is injected for 1 min 30s at 50 µL/min. This provides the formation of hemolysin pore on the liposomes membrane as represented in Figure 1C. The resulting liposome are semi-permeable.
? TROUBLESHOOTING
3.7. Optimized procedure for CLSM and UV-autofluorescence microscopy
I.7.1. Mask design and master preparation • TIMING ~1-7 days
1’-11’) Steps 1 to 11 of the previous procedure (part IV.1 and IV.2) related to the design of the masks and the fabrication of the SU-8 resin master remain unchanged for the production of the adapted chips.
I.7.2. Preparation of the optimized microfluidic devices • TIMING ~6-8 h
Most of the steps for the preparation of optimized chips are similar to those described in the previous procedure (part IV.3). The main difference is that the PDMS slab (10 % wt of crosslinker) is bonded to a glass or a quart coverslip by oxygen plasma technics.
12’ Steps 11-19 of the previous procedure (Part IV) related to the PDMS slab preparation are slightly modified as follows:
- Reinforcements of the chip cannot be created because the un-patterned side of the PDMS slab must be place flatly during the next plasma exposition step (12’). It must be thicker to be more resistant to tubing movements. So, 40 g of PDMS/crosslinker mixture is poured on the master (vs 30 g in step 12).
- PDMS slab must be completely crosslinked before the bonding step. The master is heated for two hours instead of 20 min in step 15.
13’) Place a quartz coverslip (ESCO Optics, U.S.) or a glass coverslip (24x60 mm Thermo Scientific, U.S.), and the PDMS slab, on the treatment support of an oxygen plasma cleaner. Place them flatly with the patterned PDMS surface directed to the top side.
14’) Expose the patterned side of the PDMS slab and the quartz (or glass) slide to oxygen plasma for 1 min (at the higher intensity of plasma and a pressure of 120 Pa) using a Plasma-cleaner.
▲ CRITICAL STEP For an effective bonding of PDMS on glass (or quartz) surface in the next step, plasma have to be formed properly. Check it by the observation of the formation of a kind of blue/purple cloud. Also, to check the correct treatment of the surfaces, place a drop of water on the surface of the treated PDMS slab. This drop must spread out ; PDMS becomes hydrophilic when its Si-CH3 bonds are modified into Si-OH bonds as explained in Figure S3.
? TROUBLESHOOTING
15’) Immediately after the plasma treatment put in contact the glass (or quartz) surface with the patterned side of the PDMS slab and gently press them for 1 min flatly.
▲ CRITICAL STEP If the pressure exerted is not sufficient, the PDMS traps patterns are not sufficiently bonded and the trapping of the droplets is difficult to perform. Excessive pressure can however lead to distortion of the patterns.
? TROUBLESHOOTING
16’) Place the microfluidic chip in an oven at 90 °C for 30 min to promote the formation of PDMS/quartz or PDMS/glass bonds.
☐ PAUSE POINT The resulting microfluidic chip can be stored indefinitely at room temperature in a dust-free closed plastic box.
3.7.3. Surface treatments of optimized microfluidic devices • TIMING ~6h
17’) Steps 26 to 53 of the previous procedure (part IV.4 and IV.5) related to the chips surface treatments remain unchanged for the production of the chips adapted to CLSM.
18’) Due to the design singularity of the chip made for UV-autofluorescence microscopy, the following few modifications to the steps 26 to 53 of the previous procedure have to be considered for its surface treatments procedure:
The injection of fluids for the surface treatments (benzophenone, acrylic acid, ethanol, water at pH 11, BSA, HEPES, streptavidin ...) must be done through the direct entry of the circuit highlighted by the area (1’) in Figure 6C. The outflow of its fluids must be through the outlet of the trapping chamber highlighted in this same figure by the area (4’). Also, the circuit inlets corresponding to areas (2’) and (3’) of this circuit must be closed with plugs.
3.7.4. Semi-permeable GUVs production and immobilization with optimized microfluidic devices • TIMING ~30min
19’) Steps 54 to 75 of the previous procedure related to the semi-permeable GUVs production and immobilization remain unchanged for the use of the chips adapted to CLSM.
20’) For the UV-autofluorescence microscopy chip, the following few modifications to the steps 54 to 75 of the previous procedure (part IV.6) have to be done for the semi-permeable GUVs production and immobilization.
21’) The double emulsion droplets are produced using a PDMS/PDMS chip by following steps 54-66.
22’) Once the droplets of double emulsions are formed using this first circuit and the flow rates have been stabilized, close the inlets 2' and 3' (Figure 6) of an optimized circuit (PDMS-quartz chip) using plugs.
23’) Using a stainless-steel coupler, connect a tygon tubbing to the outlet of the trapping chamber of this circuit (area 4’ in Figure 6).
24’) Connect the inlet (area 1’ in Figure 6) to the output of the chip of generation using tygon tubing and a coupler.
Note: At this stage, the droplets pass from one circuit to another and are immobilized in the optimized circuit.
? TROUBLESHOOTING
25’) Proceed as in steps 68-75 to generate semi-permeable GUVs.
▲ CRITICAL STEP The trapping chamber of the optimized circuit is size-reduced compared to the conventional circuit. The flow rates of the phases are therefore increased at the center of the channels. Consequently, it is more difficult to trap droplets while maintaining an external phase flow despite the surface treatment allowing the adhesion of biotinylated lipids. The connection/disconnection of the syringes at the inlets induces strong flow disturbances and significantly increase the rearward movement of the droplets. Hermetic valves may help in controlling the inlet flow in and out.