Crystallization of Proteins on Chip by Microdialysis for In Situ X-ray Diffraction Studies

This paper describes the fabrication protocol of microfluidic chips developed for on-chip protein crystallization with the dialysis method and in situ X-ray diffraction experiments at room temperature. One of the main advantages of these microchips is the choice of fabrication materials, rendering the chips compatible for in situ X-ray diffraction data collection at room temperature. The on-chip crystallization experiments require small volumes of the protein sample, reducing the overall cost of working with these high value macromolecules.

On-chip crystallization also eliminates the manual harvesting of the fragile protein crystals, frequently applied in combination with cryo cooling during conventional crystallography experiments. Moreover, our microchips use microdialysis for protein crystallization, which is a diffusion-based method aiming to equilibrate the precipitant concentration through a semi-permeable membrane, in order to approach the cromulent concentration for protein crystallization, and enables precise and reversible control over the crystallization conditions. Microdialysis combined with temperature control can be used to decouple nucleation from crystal growth for investigating phase diagrams, by changing the precipitant concentration while using the same protein sample.

First prepare 50 grams of PDMS silicone base and it's curing agent in a 10 to one mass ratio. Mix the two ingredients in a beaker with a spatula. After mixing, place the mixture in a vacuum chamber in order to remove all the air bubbles.

Put 25 grams of the premixed PDMS into the first SU8 to master containing the patterns of the microfluidic channel and the pillars, up to a height of approximately five millimeters. Then pour the remaining 25 grams of the PDMS into the second SU8 master, containing only the patterns of the pillars. Cure the two PDMS layers in an oven at 338 kelvin for one hour.

Cut the cured PDMS layers around the patterns of the SU8 masters with a scalpel, and gently peel off the PDMS molds from the silicone masters. Place the PDMS mold, featuring both the channels and the pillars on a rigid microscope glass slide with the patterns facing upwards. Cut a small piece of regenerated cellulose dialysis membrane in order to cover the central pillar of the PDMS mold, which is designed to be the protein chamber.

The molecular weight cutoff of the dialysis membrane is chosen accordingly to the molecular weight of the protein sample, and the precipitants of the crystallization solution. Then separate carefully the dry piece of the regenerated cellulose dialysis membrane. Cut a small piece of the separated dialysis membrane.

The size of this piece of membrane depends on the design of the chip. And specifically, it depends on the dimensions of the central pillar. Deposit the piece of the dialysis membrane on the central pillar of the PDMS mold, which is supported on the glass slide, with its pillar and channels facing upwards.

Then place the second PDMS mold featuring only the pillars facing downwards, on top of the PDMS mold that is already supported on the glass slide. Align all the pillars of the two PDMS molds. The piece of the regenerated cellulose dialysis membrane is sandwiched between the two central pillars of the PDMS molds.

This step of the fabrication procedure can be conducted under an optical microscope or simply by visually inspecting carefully the alignment of all the pillars. Place the assembly in a vacuum chamber and desiccate for 30 minutes in order to remove the trapped air bubbles within the PDMS molds, and to enhance the insertion of the photocurable NOA resin, that will be described in the next step of this protocol. Remove the assembly from the vacuum chamber and fill the empty space between the two PDMS molds with the photocurable thiolin-based resin, NOA 81, by capillary imbibition.

Apply the resin on three of the four sides of the assembly. Place the assembly in a cross-linker and cure the NOA resin by exposing to UV light for eight seconds, using a collimated UV lamp of 35 milliwatt per square centimeter. Cut a 175 micrometers thick PMMA piece in the standard dimensions of a microscope glass slide, and peel off the plastic protection from each side of the PMMA piece.

Once the first UV exposure is completed, remove the upper PDMS mold with a partially cross-linked NOA resin stuck on it from the bottom PDMS mold and the glass slide. Press gently the assembly of the upper PDMS mold and the partially cured NOA resin on the PMMA piece. Place the new assembly in the cross-linker and cure again the NOA resin by exposure to UV light for 60 seconds.

Remove carefully the upper PDMS mold. The dialysis chip made of the fully cross-link NOA resin embedding a regenerated cellulose dialysis membrane and being supported on a PMMA piece is ready. Bond commercially available connectors on the inlet and outlet points of the microfluidic channel with fast epoxy glue.

For the fluid handling, choose the diameter of the PTFE tubes based on the size of the connectors. The tubes are used for the introduction of the crystallization solution within the fluidic channel of the dialysis chip. Pipette manually a droplet of the protein sample inside the protein reservoir located right upon the dialysis membrane.

The volume of the protein droplet depends on the design of the microchip that is being used. And specifically, it depends on the volume of the central pillar, and it can be 0.1 or 0.3 microliters. Carefully apply a thin layer of high vacuum silicone grease all around the protein reservoir.

Cut a small piece of PMMA and gently place it above the thin layer of the silicone grease. Cut a piece of Kapton tape, large enough to cover the PMMA piece placed above the protein reservoir and to stick on the NOA chip around all edges. The protein sample is encapsulated within the reservoir and the chip can be used for crystallization experiments.

First prepare approximately 500 microliters of the crystallization solution by mixing appropriate volumes of the buffer and the precipitant solutions. Inject the crystallization solution in the point of the dialysis chip through the fluidic connectors, either manually with the syringe or with an automated pressure driven system. Once the fluidic channel is filled with the crystallization solution and no air is trapped within, seal the inlet and outlet ports of the chip with Parafilm tape.

Pipette the appropriate volume of the protein solution within the protein reservoir and encapsulate the protein sample as has already been described. The protein crystal growth can be visualized and recorded with a digital camera. For the fabrication of the dialysis microchips, we have chosen optically transparent and biologically inert materials, demonstrating high compatibility for in situ X-ray diffraction experiments.

The background noise generated by the materials comprising the protein compartment of the chip, which is in the direct path of the X-ray beam, has been evaluated. The protein compartment consists of the regenerated cellulose dialysis membrane, the Kapton tape, and two PMMA layers, one used as a substrate of the microchips, and one used for the encapsulation of the protein sample. In this picture, is illustrated the background noise generated by the Kapton tape, the regenerated cellulose dialysis membrane, the PMMA layer, and their assembly.

Even though the photocurable resin NOA comprises the main body of the microchips, is not included in this measurements as it is not part of the protein chamber. Diffuse rings attributed to the interactions of X-rays with the materials can be seen for the Kapton tape at the resolution lower than four angstroms, the PMMA between four to eight angstroms, and the dialysis membrane between four to five angstroms. The total background noise generated by the chip is mainly observed at the resolution lower than six angstrom, indicating that the treatment of lysozyme high resolution diffraction data is not affected.

The microchips were mounted in front of the x-ray beam with a 3D printed support that we designed for in situ X-ray diffraction experiments, and it can hold up to three chips simultaneously. Experiments were conducted in order to evaluate the efficiency of the microchips for on-chip crystallization of model-soluble proteins with the microdialysis method. lysozyme crystals were grown at 293 kelvin in two different conditions.

The first crystallization condition contained 1.5 molar sodium chloride, and 0.1 molar sodium acetate. The second crystallization condition contained one molar sodium chloride, 0.1 molar sodium acetate, and 30%polyethylene glycol 400, and was chosen to show the compatibility of the microchips and the microdialysis method, with precipitants of high molecular weight and viscous solutions. The crystallization experiments were carried out under static conditions.

Once lysozyme crystals were grown on the microchip, they were used for in situ X-ray diffraction experiments at room temperature. The microchips were mounted at the beam line with the aid of the 3D printed support and complete X-ray diffraction data sets were collected from two lysozyme crystals. After treatment of the X-ray diffraction data, the electron density map of a single lysozyme crystal was obtained at 1.95 angstrom resolution, as shown in figure A.Figure B shows the electron density map obtained at 1.85 angstroms resolution.

After merging two data sets obtained from two different lysozyme crystals, both electro density maps show detailed structural information that can be obtained by in situ X-ray diffraction experiments conducted directly on the dialysis microchips at room temperature, from single or multiple protein crystals. We demonstrated the fabrication procedure for microfluidic devices designed for on-chip protein crystallization with the microdialysis method, and in situ X-ray diffraction experiments at room temperature. We used fabrication materials with high X-ray transparency, compatible for in situ protein crystallography.

The diffraction data collection is automated with the use of a 3D printed support for the chips, that can be mounted directly in macromolecular crystallography beam lines. The versatility of this microchip stems from using microdialysis to reversibly control crystallization conditions and map phase diagrams using small protein volume. The prototyping of the device is straightforward, enabling the fabrication of up to 30 microchips in a single day in a clean room, with relatively inexpensive materials.

We expect that these features of the chips can be used for serial X-ray crystallography studies of more challenging protein targets.