PARTICLE SORTING IN A MICROFLUIDIC SYSTEM

Abstract

The invention relates to a method for ordering, sorting and/or focusing particles in a first microfluidic channel system, the method comprising the steps of i) providing for a first microfluidic channel comprising at least a first and a second inlet and a first outlet, ii) injecting a first fluid into the channel through said first inlet, iii) injecting a second fluid into the channel through said second inlet, wherein the viscosity of the first fluid is higher than the viscosity of the second fluid, such that the two fluids flow in a laminar fashion unmixed side by side, and one of the two fluids comprises the particles to be ordered, sorted and/or focused. The invention also relates to a microfluidic channel system for sorting different particles into one droplet.

Claims

1. A method for ordering, sorting and/or focusing particles in a first microfluidic channel system, the method comprising the steps of i. providing for a first microfluidic channel comprising at least a first and a second inlet and a first outlet, ii. injecting a first fluid into the channel through said first inlet, iii. injecting a second fluid into the channel through said second inlet, wherein the viscosity of the first fluid is higher than the viscosity of the second fluid, such that the two fluids flow in a laminar fashion unmixed side by side, and one of the two fluids comprises the particles to be ordered, sorted and/or focused.

2. A method according to claim 1, wherein the particles are comprised preferably in the second fluid and the viscosity of the first fluid is selected such, that the particles in the second fluid are confined by the first fluid to the space occupied by the second fluid.

3. A method according to claim 1 or 2, wherein the height of the first microfluidic channel is selected from the group of between 2 m and 60 m, 5 m and 50 m, 10 m and 45 m, 15 m and 40 m, 25 m and 35 m.

4. A method according to any of the claims 1 to 4, wherein the angle of the first inlet to the second inlet is below or equal to 180.

5. A method according to any of the claims 1 to 5, wherein first microfluidic channel, the ordering channel is substantially straight and has a length selected from the group of between 1 mm and 40 mm, 2 mm and 35 mm, 5 mm and 25 mm, 8 mm and 20 mm, 10 mm and 20 mm, and 12 mm and 18 mm.

6. The method of any of the claims 1 to 6, wherein the first and the second fluid are aqueous fluids and the viscosity of the first fluid is between 100 cP and 2000 cP, and the viscosity of the second fluid is between 0.001 cP and 100 cP or wherein the first and the second fluid are aqueous fluids and the viscosity of the first and second fluid is between 0.001 cP and 100 cP.

7. The method of any of the claims 1 to 7, wherein the aqueous first fluid comprises a substance selected from the group comprising an organic polymer, a natural polymer, cellulose, glucose, fructose or any other sugar, DNA, RNA, poly(ethylene glycol), poly(ethylene glycol) diacrylate, bis-acrylamide, polyacrylamide, streptavidin-acrylamide, poly-N-acrylamide, poly N-isopropylpolyacrylamide, agarose, alginic acid or mixtures thereof.

8. The method of any of the claims 1 to 8, wherein the first outlet is an emulsion outlet and the at least one particle is encapsulated in a droplet when it leaves the first outlet.

9. The method of any of the claims 1 to 9, wherein the microfluidic system comprises a third inlet.

10. The method of any of the claims 1 to 10, wherein a third fluid is injected in the third inlet and the viscosity of the three fluids is selected such that first fluid separates the second fluid from the third fluid and all three fluids flow in a laminar manner substantially unmixed for at least the length of the ordering channel.

11. The method of any of the claims 1 to 11, wherein particle may be selected from the group comprising, a single cell or aggregate of cells, a eukaryotic cell, a prokaryotic cell, a bead, an antibody or a fragment thereof, an antigen, and a hydrogel bead.

12. A method for ordering, sorting and/or focusing particles, wherein a first microfluidic channel system according to any of the claims 1 to 12 is combined with a further microfluidic channel system according to any of the claims 1 to 12 and the outlets of both systems are joined in one common channel.

13. A microfluidic channel system comprising i. a first microfluidic channel comprising at least a first and a second inlet and a first outlet, ii. a first fluid, iii. a second fluid, wherein the viscosity of the first fluid is higher than the viscosity of the second fluid, such that the two fluids can flow in a laminar fashion unmixed side by side, and one of the two fluids comprises particles to be ordered, sorted and/or focused.

14. Use of a method according to any of the claims 1 to 13, or a system according to claim 14 for sorting different particles into one droplet.

15. System comprising two or more microfluidic channels coming together at one junction, wherein at least two of the two or more channels are loaded with two fluids flowing in parallel, one of which having a higher viscosity than the other.

Description

DETAILED DESCRIPTION OF THE FIGURES

[0040] FIG. 1 shows a general mask design for the co-encapsulating particles in droplet using a device according to the present invention.

[0041] FIGS. 2A-C shows particular mask designs for the co-encapsulating particles in droplet device according to the present invention.

[0042] FIG. 3 shows different distribution of set of cells A and B in their respective channels prior to co-encapsulation. In both channels, cells are confined by a stream viscous media, as shown in FIG. 2.

[0043] FIG. 4 shows the laminar stream of the two fluids.

[0044] FIG. 5 shows the better enrichment of particles as compared to the Poisson distribution, in microfluidics device depicted in FIG. 1 or 2.

[0045] FIG. 6 shows the probability mass function of single particle/cell encapsulation following Poisson statistics or experimental data. It shows cell concentration equivalent to a lambda 0.56 for 300 s period of droplet production and cell concentration equivalent to a lambda 0.85 for 300 s period of droplet production.

[0046] FIG. 7 shows the particle/cell co-encapsulation probability of the method according to the present invention, wherein the cell concentration is equivalent to a lambda 0.56 for 300 s period of droplet production and to a lambda 0.85 for 300 s period of droplet production.

[0047] FIG. 8 shows distribution of the cells in the microfluidic channels according to the present invention.

[0048] FIG. 9 shows ordering of particles/cells in a microfluidics device, where the ordering viscous media consists in fluorinated oil.

[0049] FIG. 10 shows the method for analyzing the cell temporal positioning in a micro-channel.

[0050] FIG. 11 shows the analysis of the inter-cellular spacing flowing after a determined length of channel, based on the method described in FIG. 10.

[0051] FIG. 12 shows a particular mask design for the co-encapsulating particles/cell in droplet using a device according to the present invention achieving better co-encapsulation of particles in droplet compared to Poisson statistics described in FIGS. 6 and 7.

EXAMPLES

[0052] Microfluidic Chip Device Design, Fabrication & Preparation

[0053] All microfluidic devices were designed using a 2D computer aided design (CAD) software (AutoCAD 20, Autodesk). Prior to coating SU-8, 4 silicon wafers were dried at 200 C. for at least 5 min on a hotplate.

[0054] A SU-8 layer was applied on a wafer by spin coating SU-8 resin (SU-8-2025 or SU-2035, MicroChem) at 2000-4000 RPM as final speed for 30 sec followed by a pre-bake first at 65 C. for 1-3 min and then at 95 C. for 3-10 min and left to cool down to room temperature before UV exposure. SU-8 coated wafers were 365 nm UV exposed through a high resolution transparency mask (JD Photo Tools, UK) using a mask aligner (MJB4 contact mask aligner, SUSS MicroTec) to structure the microchannels. The mask aligner was used in WEC contact mode and exposing the resin for typically 10-40 sec with a dose of 10-16 mW/cm.sup.2. Exposed SU-8 coated wafers were post-baked first at 65 C. for 1-3 min and then at 95 C. for 1-10 min and left to cool down to room temperature before development. Post-baked SU-8 wafers were developed using PGMEA (Microchem, Y020100) for 1-10 min in a glass container on an orbital shaker at 100 RPM and wafers were subsequently dried with nitrogen and hard-baked at 200 C. for at least 5 min. Microchannel heights on the SU-8 mold were measured with a contact stylus profilometer (Dektak 6M, Veeco) or white light interferometry (NT9100, Veeco). Polydimethylsiloxane (PDMS) (Sylgard 184, DowCorning) was mixed in a ratio of 1:10 (curing agent:base agent) in a dish and poured over the SU-8 mold. PDMS was subsequently degassed in a vacuum desiccator for several minutes to remove air bubbles from the mold. PDMS was cured at 70 C. in an oven (VWR, France) for 2 hr. Cured PDMS was peeled off the mold and holes for inlet and outlet ports punched with a 0.75 mm biopsy punch (Harris, USA). Punched PDMS slabs were cleaned from particles and dust using Scotch tape followed by rinsing with isopropanol and de-ionized water. Cleaned PDMS slabs were dryed using pressurised nitrogen. The microfluidic channel network on the PDMS slab was bonded to a glass slide (50 mm75 mm, Dow Corning) by exposing the PDMS slab and the glass to an oxygen plasma (Pico, Diener Plasma) for 1 min and bringing them in contact after exposure. As last step, a fluorophilic coating was applied to the microfluidic channel walls by flushing the network with a 1% (v/v) silane (Alfa Aesar, L16584) in HFE-7500 (Novec, 3M) solution, followed by rinsing the channel with HFE-7500 and purging remaining HFE-7500 with pressurised nitrogen gas.

[0055] Instrumentation Setup

[0056] For droplet production and reinjection, experiments were carried out on an inverted, epi-fluorescent microscope (TiE, Nikon, France), which has been modified and connected with a 488 nm, 561 nm and 638 nm laser (Omicron, Germany) and 4 photomultiplier tubes (PMT) (H10723, Hamamatsu) with corresponding band pass filters (PMT1: 440/40, PMT2: 525/40, PMT3: 593/46, PMT4: 708/75, PMT5: 809/81, Semrock). The signal from the PMTs was fed into an FPGA card (NI-USB7856R, National Instruments) and droplet data recorded with a proprietary software routine (uDrop 3.5-3.9, HiFiBiO).

[0057] A high-speed camera (Phantom Series, Vision Research) was attached to the left port of the microscope to monitor droplet production and cell encapsulation. A 5-channel syringe pump (Nemesys, Cetoni) was used to drive all fluids in the microfluidic chip.

[0058] For static droplet arrays, excitation light was provided by a LED source (SOLA light engine, Lumencor Inc.). Fluorescence signals for the specific channels were recorded using appropriate bandpass filters (GFP and TRITC filter sets, Nikon, and Cy5 filter set, Semrock) and camera settings (Orca R2, Hamamatsu) at room temperature (25 C.) and ambient oxygen concentrations. Images were acquired using a 10 objective (NA 0.45).

[0059] Cell Culture & Cell Labelling

[0060] The CHO-S (Freestyle, ThermoFisher) cell line was cultured in sterile Erlenmayer flasks on an orbital shaker at 125 rpm in a standard cell culture incubator (37 C., 5% CO.sub.2). The cells were cultured in CHO-S Freestyle medium supplemented with penicillin/streptomycin, 0.5% Pluronic F-68 (ThermoFisher), and L-Glutamine.

[0061] Cell suspensions were typically labelled with Calcein AM Green, CellTracker Orange, CellTracker Red or NucRed (all ThermoFisher) according to standard protocols provided by the manufacturer.

[0062] Preparation Cellulose

[0063] 2% (w/w) methyl-cellulose (Sigma-Aldrich) was dissolved in D.I. water under agitation using a magentic stirrer and stirrer bar (500 RPM-2000 RPM) in an Erlenmayer flask at room temperature.

[0064] Cell Encapsulation Procedure & Monitoring

[0065] Cells were harvested and filtered through a 15 um, 10 um filter. Counting by flow cytometer (Guava EasyCyte, Millipore) and cell concentration was adjusted to 10 m-80 m cells/mL. Cells were then aspirated into custom-made reservoirs which are compatible with the microfluidic devices. The continuous phase consisted of 2% (w/w) 008-FluoroSurfactant (RAN Biotechnologies) in Novec HFE7500 fluorinated oil. Aqueous phases were co-flowed on-chip. The flow rates (around 1000-6000 l/h for oil, and 100-800 l/h for each aqueous solution were adjusted to generate typically monodisperse droplets of 80 to 400 pl to measure the effect of loading. During droplet formation, the cell suspension was cooled to 5 C. using a homemade accessory to slow down antibody secretion and preserve cell viability. Data were collected using the laser/PMT system or using high speed imaging at rates of 10000 to 100000 frames per second. In case the laser/PMT system was used for measurements, data processing and analysis was carried out by uDrop. In case droplets were imaged using a static droplet array, the emulsion was either directly injected into the 2D chamber system, or collected in a 1.5 ml Eppendorf tube containing fluorinated oil with 0.1% (w/w) 008-FluoroSurfactant, either on ice (cell experiments) or at room temperature (cell-free experiments). If collected in a tube, the droplets were diluted 1:1 using fluorinated oil containing 0.1% (w/w) 008-FluoroSurfactant before introduction into the observation chamber using a custom-made PDMS valve. Droplets were re-injected into the chamber using a flow rate of 750 l/h. After chamber filling was complete, the chamber was gently closed and mounted onto a fluorescence microscope (Ti Eclipse, Nikon).

[0066] Image Analysis

[0067] High-speed images were processed using a proprietary ImageJ and Matlab routine. Using ImageJ, mean gray levels of an area of interest were extracted over time resulting in a gray level curve. In Matlab, further image processing was carried out by removal of the background (mean gray level of the entire curve) and inversion of the signal resulted in a curve with peaks representing cells.

[0068] The routine would determine the spacing between peaks using a built-in function and then the data was aggregated into a histogram.