Systematic Analysis of In Vitro Cell Rolling Using a Multi-well Plate Microfluidic System

Summary

This study used a multi-well plate microfluidic system, significantly increasing throughput of cell rolling studies under physiologically relevant shear flow. Given the importance of cell rolling in the multi-step cell homing cascade and the importance of cell homing following systemic delivery of exogenous populations of cells in patients, this system offers potential as a screening platform to improve cell-based therapy.

Abstract

A major challenge for cell-based therapy is the inability to systemically target a large quantity of viable cells with high efficiency to tissues of interest following intravenous or intraarterial infusion. Consequently, increasing cell homing is currently studied as a strategy to improve cell therapy. Cell rolling on the vascular endothelium is an important step in the process of cell homing and can be probed in-vitro using a parallel plate flow chamber (PPFC). However, this is an extremely tedious, low throughput assay, with poorly controlled flow conditions. Instead, we used a multi-well plate microfluidic system that enables study of cellular rolling properties in a higher throughput under precisely controlled, physiologically relevant shear flow^1,2. In this paper, we show how the rolling properties of HL-60 (human promyelocytic leukemia) cells on P- and E-selectin-coated surfaces as well as on cell monolayer-coated surfaces can be readily examined. To better simulate inflammatory conditions, the microfluidic channel surface was coated with endothelial cells (ECs), which were then activated with tumor necrosis factor-α (TNF-α), significantly increasing interactions with HL-60 cells under dynamic conditions. The enhanced throughput and integrated multi-parameter software analysis platform, that permits rapid analysis of parameters such as rolling velocities and rolling path, are important advantages for assessing cell rolling properties in-vitro. Allowing rapid and accurate analysis of engineering approaches designed to impact cell rolling and homing, this platform may help advance exogenous cell-based therapy.

Introduction

One of the major challenges in the successful clinical translation of cell-based therapy is the inefficient delivery or targeting of systemically infused cells to desired sites^3,4. Consequently, there is a constant search for approaches to improve cell homing, and specifically cell rolling, as a strategy to improve cell therapy. Cell rolling on blood vessels is a key step in the cell homing cascade, classically defined for leukocytes that are recruited to disease sites⁵. This step is governed by specific interactions between endothelial selectins, i.e. P-and E-selectin (P-and E-sel), and their counter ligands on the surface of leukocytes^5,6. Better understanding and improved efficiency of cell homing, and specifically the rolling step, are of great importance in the quest for new platforms to improve cell-based therapy. To date this has been achieved by using parallel plate flow chambers (PPFCs), comprising two flat plates with a gasket between them, with an inflow and outflow port located on the upper plate, through which a cell suspension is perfused by using a syringe pump^{7,8 ,9}. The surface of the bottom plate can be coated with a relevant cell monolayer/substrates and the interaction between perfused cells and the surface under shear flow is then explored⁷. However, PPFC is a low throughput, reagent-consuming, and fairly tedious method, with bubble formation, leakage, and poorly controlled flow presenting major drawbacks.

An alternative technique to the traditional PPFC is a multi-well plate microfluidic system, permitting higher throughput performance of cellular assays (up to 10 times higher than PPFCs) under accurate, computer-controlled shear flow, with low reagent consumption^1,10. Cell rolling experiments are performed inside the microfluidic channels, which can be coated with cell monolayers or engineered substrates and imaged using a microscope, with rolling properties readily analyzed using a suitable software. In this study, we demonstrate the capabilities of this multi-well plate microfluidic system by studying the rolling properties of human promyelocytic leukemia (HL-60) cells on different surfaces. HL-60 rolling on substrates like P-and E-sel, as well as on cell monolayers expressing different rolling receptors, was analyzed. In addition, antibody (Ab) blocking was used to demonstrate direct involvement of specific selectins in mediating the rolling movement of HL-60 on those surfaces. Rolling experiments were performed with increased throughput, under stable shear flow, with minimal reagent/cell consumption, allowing efficient analysis of key rolling parameters such as rolling velocity, number of rolling cells, and rolling path properties.

Protocol

1. Cell Culture

1. Human promyelocytic leukemia (HL-60) cells
   1. Culture HL-60 cells in 75 cm² flasks with 15 ml of Iscove's Modified Dulbecco's Medium (IMDM), supplemented with 20% (v/v) fetal bovine serum (FBS), 1% (v/v) L-Glutamine and 1% (v/v) Penicillin-Streptomycin.
   2. Change media every 3 days by aspirating half of the cell suspension volume and replacing it with complete IMDM media.
   3. For carboxyfluorescein diacetate, succinimidyl ester (CFSE) staining, centrifuge HL-60 cell suspension (400 x g, 5 min), resuspend in a 1 μM CFSE solution (prepared in prewarmed PBS) and incubate for 15 min at 37 °C. Then centrifuge cells, aspirate supernatant and resuspend cells in fresh prewarmed medium for 30 min. Wash cells in PBS and then use for rolling experiments (see Figure 1B for representative image of CFSE-stained HL-60 cells on P-sel-coated surface).

Note: CFSE staining is optional, and is presented here to demonstrate the rolling phenomenon in the microfluidic channel. Analysis of rolling parameters presented in this manuscript was performed on unstained cells using standard brightfield imaging.

2. Lung microvascular endothelial cells (LMVECs)
   1. Coat 100 mm Petri dishes with 0.1% gelatin solution (v/v in PBS) and incubate at 37 °C for at least 30 min.
   2. Culture LMVECs on gelatin-coated 100 mm Petri dishes in complete endothelial growth medium (endothelial basal medium-2 (EBM-2)), supplemented with a specific growth supplement kit, see REAGENTS). Change media every other day and sub-culture cells upon reaching 80-90% confluence.
   3. For sub-culture, wash cells with PBS and then detach cells with 4 ml of 1x Trypsin-EDTA for 3 min at 37 °C and neutralize in an equal volume of complete EBM-2 media. Transfer the cell suspension to a 15 ml tube and centrifuge (400 x g, 5 min). Following centrifugation, resuspend the pellet in 1 ml of complete endothelial media and count cells with a hemocytometer. Do not over-passage the cells, as this affects their morphology and function- use only cells under passage 7 for all experiments.

3. Chinese Hamster Ovary-P-selectin (CHO-P) Cells
   1. CHO-P cells, which are CHO cells stably transfected to express human P-sel, were provided by collaborators (Beth Israel Deaconess Medical Center, Harvard Medical School)^11,12.
   2. Culture CHO-P cells in T175 cm² flasks in 25 ml of F-12 media.
   3. For passaging, wash cells with 10 ml of PBS for 4-5 sec and then trypsinize in 10 ml of 1x Trypsin-EDTA for 3 min at 37 °C, followed by neutralization in full media.
   4. Centrifuge the cell suspension (400 x g, 5 min), carefully aspirate the supernatant, resuspend the cell pellet in 1 ml of full media and count the cells with a hemocytometer.

2. Operation of the Integrated Multi-well Plate Microfluidic System

1. Make sure all the equipment is properly connected and turn on the different modules: computer, controller, inverted microscope, and CCD camera.
2. Open the imaging software; make sure the multi-well plate module and the imaging module are properly presented on the screen.
3. Connect the tubes to the vapor trap (connected to the controller) and also connect them to the Pressure Interface.
4. Place the multi-well plate in the plate heater/adaptor. Add reagents to wells (described below) and attach interface on top of the plate. Place plate for imaging on automated stage.
5. The interface attaches to the top of the plate and applies a pneumatic pressure from the controller to the top of the wells, driving the fluid through the microfluidic channels at the defined flow rate, easily controlled using the multi-well plate module screen under Manual mode.
6. Reagents in the channel flow across an observation area, located between the wells. Microfluidic channel dimensions are 350 μm wide x 70 μm tall. The length of the linear channel is 1 mm and the bottom of the channels comprises a 180 μm coverslip glass, which is compatible with brightfield, phase, fluorescence and confocal microscopy.
7. Acquire videos using a CCD camera (stream acquisition, 11 frames/sec) and analyze via compatible software.

3. Coating of Microfluidic Channels with a Protein Substrate or a Cell Monolayer

1. Coating microfluidic channel with fibronectin or P-/E-selectin
   1. Prepare 1 ml of 20 μg/ml fibronectin solution in PBS. Alter volume based on the number of channels to be coated (use 25-50 μl of fibronectin per channel).
   2. Add 25-50 μl of fibronectin solution to each inlet well. Apply shear force of 2 dyn/cm² for 5 min to perfuse the channel. Please note the bead of liquid appearing in the outlet well. Incubate for 30-45 min at R.T.
   3. Aspirate the solution from wells (do not aspirate directly from the middle circle that feeds the channel)^1,13. Add 200-500 μl of PBS into outlet well and wash channel with PBS by applying shear flow of 2 dyn/cm² for 5 min. The channel is now properly coated with fibronectin and ready to be used.
   4. To coat with P- or E-sel, prepare a 5 μg/ml solution of the desired human recombinant protein in PBS, and coat the channels as described above, with 1 hr incubation at 37 °C to allow surface coating.
2. Creation of CHO-P or LMVEC monolayer inside the microfluidic channel
   1. Gently trypsinize cells from culture dishes for 3 min, quench using a 2-fold volume of full media and centrifuge (5 min at 400 x g). Resuspend cells with 10 ml of full media and centrifuge (5 min at 400 x g) again.
   2. Count the cells to determine cell concentration in the suspension. To ensure the formation of a confluent LMVEC monolayer inside the channel, bring cell concentration to 15-20 million cells/ml. For a confluent CHO-P cell monolayer, use 50-60 million cells/ml. Use 25-50 μl of cell suspension for each channel - determine initial cell number used for the experiment accordingly.
   3. Add 25-50 μl of cell suspension in the appropriate concentration to the inlet well. Place the plate on the microscope stage and introduce cells into the channel (2 dyn/cm²) until cells are observed on the screen filling the entire channels, and then stop the flow.
   4. Fill both outlet and inlet with 200 μl of either full LMVEC or CHO media. Let the cells settle and adhere for 3 hr in the incubator (37 °C, 5% CO₂).
   5. Following the 3 hr incubation, wash the channel with full media (2 dyn/cm², 10-15 min) to remove unattached cells. Cells should now appear completely confluent and the channel is now ready for use. Depending on initial cell seeding density, additional 2-3 hr of settling time may be required to ensure complete coverage of the surface with the cells.

4. LMVEC Pro-inflammatory Activation and Antibody Blocking of P-/E-selectin

1. Prepare a TNF-α solution (10 ng/ml) in LMVEC basal media.
2. To induce inflammatory activation of LMVEC in the channels, add 100 μl of the TNF-α solution to the inlet well and introduce the solution into the channel by applying shear flow of 2 dyn/cm² for 5 min. For control channels (nonactivated ECs), add 100 μl of LMVEC basal media to the inlet well and introduce into the channel (2 dyn/cm² for 5 min). Channel is now ready for a rolling assay.
3. To block P-sel and E-sel on LMVECs and CHO-P cells, introduce neutralizing P-sel (clone AK4, 5 μg/ml in basal media) or E-sel (clone P2H3, 5 μg/ml in basal media) antibodies into the channel and incubate for 1 hr at 37 °C. Next, wash channels with basal media (2 dyn/cm² for 5 min). Channels are now ready for a rolling assay.

5. HL-60 Rolling Assay on Substrate/Cell Monolayer-Coated Microfluidic Channels

1. Carefully examine the channels under the microscope to confirm that channels are properly coated (in the case of coating with cells, a fully confluent cell monolayer should be observed).
2. To prepare HL-60 cell suspension for the rolling experiments, centrifuge HL-60 cell suspension (5 min at 400 x g) and wash once with basal media. Count the cells and resuspend in IMDM (basal media, containing Ca²⁺ and Mg²⁺) to create a HL-60 cell suspension with 5 million cells/ml. Use 25-50 μl of cell suspension for each channel to perform the rolling assay.
3. Add 25-50 μl of the cell suspension to outlet well, place plate inside the temperature-controlled plate holder (37 °C) and place on the microscope stage. Next, introduce cells into the channel by applying shear force of 2 dyn/cm² (cells should be observed within 10-15 sec flowing from outlet to inlet).
4. To examine the rolling response as a function of shear stress, reduce shear to 0.25 dyn/cm² and acquire 20-30 sec videos (using "stream acquisition" function) in each desired shear (increase shear gradually from 0.25 up to 5 dyn/cm². It is also possible to use higher shears).
5. Acquire videos using a CCD camera (stream acquisition, 11 frames/sec) and analyze rolling paths and rolling velocities via compatible software.

6. Flow Cytometry to Detect Expression Of Surface Molecules

1. Following trypsinization, prepare a cell suspension (using 1-2 x 10⁵ cells/sample) of desired cell type (HL-60, CHO-P or LMVECs) in PBS (-/-), supplemented with 2% FBS. Wash cells twice and bring sample volume to 50 μl (using the same buffer).
2. Incubate each sample with the desired fluorophore-conjugated Ab (see attached table for detailed information) at 4 °C for 20 min (cover with aluminum foil).
3. Wash the cells twice (same buffer) and bring final volume of stained cell suspension to 200 μl. Analyze samples using a flow cytometer to detect expression of surface molecules.

Results

HL-60 cells roll on P- and E-selectin surfaces, but not on fibronectin

HL-60 cells are considered gold standard "rollers" as they express a variety of homing ligands, including the rolling ligands P-sel glycoprotein ligand-1 (PSGL-1) and Sialyl-Lewis X (SLeX)^5,14 (Figure 1A). The surface protein PSGL-1 acts as a scaffold for the tetra-saccharide SLeX, mediating specific interaction with P- and E-sel, which are up-regulated on the endothelium during inflammation...

Discussion

One of the major challenges in successful translation of exogenous cell-based therapy is the inability to efficiently deliver cells to sites of injury and inflammation with high engraftment efficiency³. Cell rolling represents a critical step in the process of cell homing, facilitating the deceleration of cells on the walls of blood vessels, eventually leading to their firm adhesion and transmigration through the endothelium into the tissue⁵. Better understanding of the rolling process for candidate...

Materials

Name	Company	Catalog Number	Comments
Human Lung Microvascular Endothelial Cells	Lonza	CC-2527
P-selectin-expressing Chinese Hamster Ovary Cells (CHO-P)	Kind gift by Dr. Barbara Furie11,12
HL-60 Cells	ATCC	CCL-240
Cell Culture Reagents
Endothelial Basal Medium	Lonza	CC-3156
EBM-2 Media	Lonza	CC-3156
Endothelial Basal Medium Supplements	Lonza	CC-4147
EGM-2 MV SingleQuots	Lonza	CC-4147
IMDM - Iscove's Modified Dulbecco's Medium 1x	Gibco	12440
F-12 (1x) Nutrient Mixture (Ham)	Gibco	11765-054
Penicillin Streptomycin (P/S)	Gibco	15140
L-Glutamine (L/G) 200 mM	Gibco	25030
Fetal Bovine Serum (FBS)	Atlanta Biologicals	Sa550
Petri Dishes	BD Falcon	BD-353003
100 mm Cell Culture Dish, Tissue-Culture Treated Polystyrene
Centrifuge Tubes (15 ml polypropylene conical tubes)	MedSupply Partners	TC1500
T75 Flasks	BD Falcon	353136
Gelatin Solution (2%)	Sigma	G1393
dPBS (without calcium chloride and magnesium chloride)	Sigma	D8537
Trypsin-EDTA Solution (10x)	Sigma	T4174
Antibodies
Anti-hE-Selectin/CD62E	R&D Systems	BBA21
FITC Conjugated Mouse IgG1	R&D Systems	BBA21
Anti-hP-Selectin	R&D Systems	BBA34
FITC Conjugated Mouse IgG1	R&D Systems	BBA34
FITC Mouse IgG­1 κ Isotype Control	BD Bioscience	555748
Anti-SLeX /CD15s Ab, Clone: 5F18	Santa Cruz	SC70545
FITC Conjugated	Santa Cruz	SC70545
Normal Mouse IgM-FITC Isotype Control	Santa Cruz	SC2859
PE Mouse Anti-Human CD162, Clone: KPL-1	BD Pharmingen	556055
PE Mouse IgG1 k Isotype Control	BD Pharmingen	550617
Anti-P-Selectin Ab (AK4)	Santa Cruz	SC19996
Anti-E-Selectin Ab, Clone P2H3	Millipore	MAB2150
Mouse IgG1 Isotype Control	Santa Cruz	SC3877
Other Reagents
Recombinant Human TNF-alpha	PeproTech	300-01A
Cell Trace CFSE Cell Proliferation Kit - For Flow Cytometry	Invitrogen	C34554
Human P-selectin-FC recombinant protein	R&D Systems	137-PS-050
Human E-selectin-FC recombinant protein	R&D Systems	724-ES-100
Fibronectin Human, Plasma	Invitrogen	33016-015
Equipment
Bioflux 1000	Fluxion Biosciences	Bioflux Montage was the software used to run the experiments and analyze the data
BioFlux 48-well plates	Fluxion Biosciences
BD Accuri C6 Flow Cytometer	BD Bioscience	CFlow Plus was the software used to run the experiments and analyze the data
Nikon Eclipse Ti-S	Nikon
CoolSnap HQ2 CCD camera	Photometrics