Patterning of Microorganisms and Microparticles through Sequential Capillarity-assisted Assembly

This method allows producing user-defined patterns of microorganisms within a microfluidic channel. Once patterned, the microorganisms can be monitored to evaluate their long-term physiology and interaction with high throughput. The use of capillary forces enables a nonspecific route for patterning, which is applicable across different classes of materials.

For example, colloidal particles and bacteria. Moreover, it allows to produce large arrays with exquisite control of the spatial arrangement of the material of interest. To date, we tested and applied the method on colloidal particles and microbial cells.

As a result, it can find applications in material engineering and fields requiring quantitative single-cell analysis like drug screening. To begin, prepare a PDMS mixture by mixing the elastomer with its cross-linking agent. Then stir the mixture vigorously to blend the two components uniformly until air bubbles are formed and the PDMS mixture looks opaque.

Now, degas the mixture in a vacuum desiccator until all air bubbles are removed and the mixture looks transparent again. To obtain a 400 micrometer thick template floor of the microfluidic chip, pour three grams of the mixture on the silicon master and place the silicon master on a spin coater to spin coat at 21 times G for five seconds and 54 times G for 10 seconds. Degas again for removing the trapped air bubbles as described earlier.

Pour 20 grams of the PDMS mixture into the 3D printed mold to make the microchannel which will serve as the roof of the microfluidic chip and degas it for 30 minutes as described earlier. Bake the silicon wafer and the 3D printed mold at 70 degrees Celsius for at least two hours, then cut the PDMS along the outline of the 3D printed mold and peel it off. Cut the PDMS with a blade around the microchannels and punch the holes that will serve as inlet and outlet of the microfluidic channel.

Now cut the PDMS and peel it off the silicon master and cut the PDMS layer into smaller pieces with the same dimensions of the microfluidic channels that will be bonded on top of the templates. Gently rub the templates and microchannels using a 1%detergent solution for five minutes and then rinse with deionized water. Next, rinse the templates and microchannels with isopropanol, before rinsing them with deionized water.

Now dry the templates and microchannels at room temperature for one minute with compressed air at one bar. Place the templates and the microchannels in a plasma cleaner with the bonding surfaces facing up. After turning on the plasma cleaner, plasma treat the templates and microchannels for 40 seconds, then take them out from the plasma cleaner and immediately bond the microchannels on top of the templates.

Store the microfluidic chips in an oven at 70 degrees Celsius for five days to ensure PDMS hydrophobic recovery. On the day of the experiment, set the box incubator at 37 degrees Celsius several hours before the experiment, then set up the syringe pump and the heated glass plate on the microscope stage, setting the same temperature as that of the box incubator. 90 minutes before the experiment, put the microfluidic chip in a vessel filled with 100%ethanol and flush the channel with 100%ethanol for at least 10 minutes, then place the microfluidic chip in a vacuum desiccator and degas for at least 30 minutes.

Now exchange the ethanol with distilled water and vacuum treat the chip for at least 30 minutes. Put the microfluidic chip in the oven at 70 degrees Celsius for 10 minutes to remove any traces of liquid left in the channel. Pipette one milliliter of MOPS medium into a centrifuge vial and add 10 microliters of 0.132 molar potassium phosphate.

Aliquot 100 microliters of the overnight culture into the centrifuge vial and centrifuge the culture at 2, 300 times G for two minutes. Gently discard the supernatant to resuspend the pellet in one milliliter of fresh MOPS medium with 0.015%of Tween 20 and 0.01%of potassium phosphate and load the bacterial suspension in a one milliliter syringe. To secure the syringe and the tubing connection, directly insert a needle into the tubing.

Now mount the syringe on the syringe pump and inject the suspension into the microfluidic chip through the inlet located at the upstream part of the channel until the suspension covers the template region with traps. Set the syringe pump at a flow rate of 0.07 to 0.2 microliters per minute to withdraw the bacterial suspension and monitor the patterning process via microscope software. Once the template has been patterned with cells, increase the withdrawal flow rate to quickly empty the microfluidic channel and flush it with fresh LB that was previously degassed for at least 30 minutes and prewarmed at 30 degrees Celsius.

Now set the syringe pump at a flow rate of two microliters per minute to gently flush the channel. Once the channel has been filled, again increase the flow rate. Acquire images of growing bacteria at the desired magnification and time interval.

Pipette 900 microliters of a 0.015%Tween 20 aqueous solution into a centrifuge vial and then pipette 100 microliters of the original colloidal suspension into it. Centrifuge the suspension at 13, 500 times G for one minute and gently replace the supernatant with the aqueous Tween 20 solution. Load the colloidal suspension in a one milliliter syringe and connect the syringe to the chip through microfluidic tubing.

Inject the suspension into the microfluidic chip through the inlet located within the central section at the upstream part of the channel and gradually push the suspension until the template is covered. Withdraw the colloidal suspension at a flow rate of 0.07 to 0.2 microliters per minute and image the patterning process via microscope software. Increase the flow rate once the meniscus reaches the end of the template quickly.

The straight channel geometry was exploited to pattern stationary phased cells of a fluorescent Escherichia coli strain depositing 83%of the 5, 000 analyzed traps. Patterned bacteria resumed growth at different times within 1.5 hours from when the channel was filled with fresh LB.Once growth was resumed, single bacterial cells started forming individual colonies which expanded until a surface layer was formed and the single cell resolution was lost. Analysis conducted over 55, 000 traps showed that green-red dimers made by two and one micrometer diameter particles were formed in 93%and 89%of analyzed traps respectively.

And green-red-green trimmers were formed in 52%of the traps. The distance between patterned particles can be precisely controlled by performing two depositions in opposite directions, thus trapping such particles at the opposite ends of each trap. It is important to ensure uniform temperature across the template to prevent condensation during the patterning process.

To this end, we place a heated glass plate underneath the template and set it at the same temperature as the box incubator. This method can be used in a variety of biological studies involving quantitative single-cell analysis. A unique advantage is that its high throughput nature which allows patterning of thousands of cells providing large statistics within the same experimental arena.