High-throughput microfluidic cell sorting platform (MICS)

Genome-scale functional genetic screens can be used to interrogate determinants of protein expression modulation of a target of interest. Such phenotypic screening approaches typically require sorting of large numbers of cells (>10 8 ). In conventional cell sorting techniques (i.e. fluorescence-activated cell sorting), sorting time, associated with high instrument and operating costs and loss of cell viability, are limiting to the scalability and throughput of these screens. We recently established a rapid and scalable high-throughput microfluidic cell sorting platform (MICS) using immunomagnetic nanoparticles to sort cells in parallel capable of sorting more than 10 8 HAP1 cells in under one hour while maintaining high levels of cell viability (Ref. 1). This protocol outlines how to set-up MICS for large-scale phenotypic screens in mammalian cells. We anticipate this platform being used for genome-wide functional genetic screens as well as other applications requiring the sorting of large numbers of cells based on protein expression.


Introduction
Microfluidics offers precise spatio-temporal control over fluids and cells due to the laminar flow generated by their small geometric size. Using immunomagnetic nanobeads, i.e. antibodies coupled to magnetic beads, which bind specifically to a target protein found in or on a cell, these devices can be used to sort heterogenous cell populations based on the expression of the target protein. Although the low manufacturing and operating costs and high speed at which these devices operate have made them a competitive alternative to cell sorting by FACS, the limited throughput of these small devices has consistently remained a barrier to adoption for large-scale applications.
a. Buffer can be degassed by putting in Erlenmeyer flask under vacuum for 30min.

Chip Set Up
The day before sorting, chips must be degassed by filling with 1% m/v Pluronic® F-108 in DI water and leaving overnight. This process removes any air from the device and lubricates the SU-8 surface.
It is suggested that chips be prepared while cells are being labelled to save time.
1. Remove plunger from two 10ml syringes per chip, place in stand and add luer-lock fitting. Label one syringe as "sample" and the other as "sheath".

Possible Issue
Outlet syringes slipping on the pump

Possible Reason/Effect
If using double sided tape to hold chips to magnet, chips can be difficult to remove/ adjust, resulting in chips breaking. Besides rendering the chips unusable, the glass in the chips can easily cut a user.

Solution
When adjusting/removing chips, place fingers on sides of chip and twist away from the magnet. Do not pull up to remove the chip.

Possible Reason/Effect
Leaks in the chips can result from errors during fabrication. Chips with leaks will have non-ideal laminar flow patterns and result in erroneous sorting.

Solution
Ensure chips are degassed. If a chip has a leak, it will often become evident during degassing. Placing a paper towel under chips while they degas will also help to notice leakage (even if leaked fluid dries, the paper towel will look used). If chips leak, do not try to plug. Discard leaking chip and replace with a fresh one.

Possible Issue
Tubing connected to wrong syringe

Possible Reason/Effect
With many tubes, mix-ups can happen, connecting a tube to the wrong outlet/inlet syringe. This can result in the sample entering the chip where sheath fluid should be, or a pressure differential from the pump being applied to the wrong region of the chip. This will create wrong laminar flow patterns.

Solution
Be very careful to ensure tubes are correctly connected to syringes, label if necessary.

Possible Issue
Inlet columns draining at different rates.

Possible Reason/Effect
This is likely due to tubing being connected to the wrong syringe. The total time require for this protocol can range from 2 to 8hr, depending on the labelling method, flow rate chosen, number of cells, number of samples and operator experience.

Anticipated Results
If protein expression has been detected on the cells of interest, users should observe cell populations in the medium and high outlets of the chip. The amount depends on the expression levels of the protein target. Users may also asses the relative volume of fluid that ends up in each outlet of each chip, since this metric is independent of the target protein or flow rate. The percentage of total volume should be 56% (low), 28% (medium) and 11% (high). Deviations in excess of 5% indicate that the flow patterns within the chip were not ideal, likely due to air bubbles within the chip or erroneous