High-throughput Flow-cytometry Measurement of Cellular Mechanotype Based on Rupture and Delivery of DNA Tension Probes into Cells

Summary

This method measures a cell's propensity to rupture ligand-presenting immobilized DNA-duplexes to report relative cellular traction forces. Ruptured fluorescent oligos are delivered into the cell and analyzed by flow cytometry, recording the mechanical history of the cell. This allows for high-throughput measurement of cellular force generation in the context of cellular mechanophenotyping.

Abstract

Mechanical forces guide many critical cellular processes, and cellular mechanical phenotype, or mechanotype, is disrupted in many diseases. The most widely used method to quantify cellular forces is traction force microscopy, which traditionally captures forces that cells exert on their surroundings by high-resolution microscopy of cells deforming elastic substrates containing fluorescent beads, deflecting microposts, or altering the conformation of immobilized molecular tension sensors. We present here a high-throughput method to measure relative cellular traction forces using established DNA tension probes and quantified via flow cytometry. There are five overall steps involved in this method described below. First, expression and purification of the high-affinity integrin ligand echistatin fused to a DNA-linking HUH-tag; second, preparation of the DNA duplex probes containing reporters and ligand; third, surface preparation and immobilization of probes; fourth, execution of the "Rupture and Deliver" tension-gauge-tether experiment and flow readout. Finally, we will describe data analysis and interpretation of experimental results.

Introduction

Mechanical forces guide many critical cellular processes, from migration¹ to stem cell differentiation² to the ability of a T cell to bind to and be activated by the correct antigen-presenting cell³. Mechanical variables are dysregulated in disease resulting in measurable differences in cellular mechanical phenotype, or mechanotype, of cells^4,5. Mechanotype is an umbrella term for the mechanical properties of cells and tissues and can include parameters such as contractility, migration, adhesion, and deformability. Cell or tissue stiffness is the most common aspect of mechanotype measured as it can be measured in high throughput. For example, cancer cells are often stiffer than normal cells while metastatic cells are often more compliant^6,7. Force generation by cells is an important aspect of mechanotype altered in disease⁸ because it probes specific cell-to-environment interactions. Traction force microscopy is the most common method to measure cellular force generation and involves plating cells on deformable elastic substrates embedded with fluorescent beads or on arrays of microposts and tracking changes in bead movement or post deflection to calculate forces exerted by cells. Molecular tension sensors have also been used to measure traction forces and offer the advantage of sensitivity to smaller magnitudes of force and measuring forces generated by specific ligand-receptor pairs^9,10. However, measuring cellular forces with traditional TFM and molecular tension sensors relies on high-resolution imaging, limiting the throughput of measurement.

We recently adapted DNA-duplex tension probes into a high-throughput readout of cellular force generation in the context of cellular mechanophenotyping¹¹ (Figure 1). Briefly, DNA duplexes adapted from the extensively used tension-gauge-tether (TGT) probes¹² are immobilized in the wells of 96-well plates via biotin-neutravidin chemistry of the "anchor" strand¹³. The requisite rupture force of the TGT duplexes can be tuned by varying the position of the biotin on the anchor strand, which alters the geometry of the duplex resulting in in either a low rupture force unzipping conformation (Figure 2A) or a high rupture force shearing conformation (Figure 2B). The anchor strand contains biotin and a fluorescence quencher, while the "ligand" strand of DNA contains a fluorescent probe and a ligand recognizing a cell surface receptor involved in mechanosensing such as an integrin. The ligand is covalently linked to the DNA (Figure 3) via the use of a DNA-linking fusion tag called an HUH-tag fused to the ligand^14,15. HUH-tags are small HUH endonuclease domains involved in viral replication and other processes that recognize a 9 bp sequence/structure of single-stranded DNA, nick it, and form a covalent phosphotyrosine adduct to the 5' end of the nicked DNA.

Cells plated on top of these probes bind to the ligand and rupture a percentage of the probes proportional to the forces they generate. The fluorescent ligand strand bound to its receptor is subsequently delivered into the cell. Thus, the mechanical history of the cells is recorded in the cells of interest via the magnitude of their fluorescence. The fluorescence can be quantified using flow cytometry. We call this method Rupture and Deliver-TGTS or RAD-TGTs. This method not only measures the fluorescence and, thus, the relative force generated by thousands of cells at once, but also the flow cytometry readout offers the potential of sorting the cells by their mechanotype, allowing downstream -omics readouts, as well as CRISPR-KO or drug screens that might identify genes or drugs that alter cellular mechanotype.

In the protocol below, we demonstrate how to prepare and execute a RAD-TGT experiment. We will highlight the expression and purification of a type of HUH known as WDV (from wheat dwarf virus) by itself to serve as a negative control and fused to echistatin, a ligand that tightly binds a family of adhesion receptors that transduce force between cell and environment known as RGD-binding integrins, to serve as an experimental condition. From this, we will demonstrate how to prepare RAD-TGTs for experimental use. We will explain how to proceed with experiments and how to analyze the resulting data. We will also highlight an application of RAD-TGTs by measuring how a ROCK inhibitor alters integrin-mediated forces in CHO-K1 cells (Figure 4). In summation, we hope this protocol serves as an entry to mechanobiology regardless of prior experience; the field is in its nascent stage and all backgrounds are essential to further understanding.

Protocol

1. HUH expression and purification

1. Transform Escherichia coli with WDV (± Echistatin) plasmid, approximately 1.5 µL of ~50 ng/µL plasmid per transformation, and grow on ampicillin agar plates.
2. Inoculate 5 mL of LB with a single colony and grow overnight at 37 °C.
3. Inoculate 1 L of LB with the overnight culture and grow until OD₆₀₀ reaches ~0.8.
4. Lower the temperature to 18 °C, induce with 500 µM isopropyl β-D-1-thiogalactopyranoside (IPTG), and incubate overnight.
5. Pellet the cells by centrifugation at 3,750 × g for 30 min. Resuspend the pellet with 40 mL of lysis buffer (300 mM NaCl, 50 mM Tris pH 7.5, 1 mM EDTA).
6. Lyse the cells with 3 x 90 s sonication on ice at 70% duty cycle, power 7. Spin down at 24,000 × g for 30 min at 4 °C and pour the supernatant into a new 50 mL tube.
7. Put Ni-NTA resin slurry into a 15 mL conical tube (1 mL of bead suspension/L of culture), spin down at 700 × g for 2 min, and remove the supernatant.
8. Add 6 mL of water to the resin and repeat the washing process in step 1.7.
9. Repeat the wash twice with wash buffer (300 mM NaCl, 50 mM Tris [pH 7.5], 1 mM EDTA, 20 mM Imidazole) instead of water. Resuspend the beads in 5 mL of wash buffer.
10. Add the bead slurry (step 1.9) to the lysed cells (step 1.6) and mix thoroughly. Rotate the lysed cells and beads for at least 1 h at 4 °C.
    NOTE: For the remainder of this protocol, never let the resin go dry and collect all samples that flow through the column.
11. Pour the mixture onto a 20 mL column and collect the flowthrough. Wash the column with 4 x 25 mL of wash buffer.
12. Elute with 1 x 10 mL of elution buffer (300 mM NaCl, 50 mM Tris [pH 7.5], 1 mM EDTA, 250 mM Imidazole) into a fresh 15 mL tube, collecting early and late elution separately (early elution is the first 1 mL, late elution is the final 1 mL).
13. Analyze all the collected fractions via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to ensure the HUH was collected during the elution (see Supplemental Figure S1). For best results, load a 4-20% gel and run it for 45 min at 35 W and 200 V.
14. Add the eluted protein into dialysis tubing and incubate overnight in dialysis buffer (same as lysis buffer) at 4 °C. To remove the HIS-tag, add ULP1 with no reducing agent directly to the eluted protein prior to putting the protein in the dialysis tubing.
15. The next morning, prepare Ni-NTA resin as described in steps 1.7-1.9 and incubate with the dialyzed protein. Pour the mixture of the protein and Ni-NTA resin into a column and collect the flowthrough. Wash the column with 10 mL of dialysis buffer and collect any flowthrough.
16. Assess the collected fraction on SDS-PAGE following the above settings to ensure the protein is present, and concentrate to the desired concentration using a 15 mL ultracentrifugal filter with a 10 kDa molecular weight cutoff (see Supplemental Figure S1).
    NOTE: The ideal final protein concentration is between 40 µM and 100 µM.

2. RAD-TGT preparation

1. Assemble duplexes in a 1.1:1 molar ratio of anchor strand (with quencher) to ligand strand (with fluorophore) at a final concentration of 40 µM ligand strand in a solution of 10 mM Tris (pH 7.5), 50 mM NaCl, and 1 mM EDTA.
2. Anneal by incubating at 98 °C for 5 min followed by cooling at room temperature for 1 h; protect from light.
   NOTE: Low-volume samples easily evaporate and can be difficult to recover completely. Prepare greater volume than necessary.
3. React the annealed duplexes with HUH (using the concentration of the least abundant strand when annealing; in this case, the ligand strand) in a 1:2 molar ratio at a final concentration of 10 µM duplex and 20 µM HUH in a solution containing 1 mM MnCl₂, 50 mM HEPES (pH 8), 50 mM NaCl.
4. Incubate the reaction for 30 min at 37°C in a thermal cycler or light-protected heat block. Following incubation, remove from heat and use a bench top minifuge to spin down the now assembled TGTs. Either plate the TGTs immediately or store at 4 °C in the dark.

3. Surface preparation

1. Immediately upon opening a 96-well glass bottom plate, cover all the wells with adhesive PCR plate foil to preserve unused wells for later use. Use a razor blade to cut out the foil over the wells of interest (preferably non-edge wells) and add 100 µL of cold PBS to the desired wells.
   NOTE: Once liquid is added to wells, do not let the wells go dry. If drying occurs, non-specific adsorption of biomolecules to the surface will negatively impact data. Keep 100 µL of excess volume continually in the well to prevent any drying.
2. Add 200 µL of cold PBS to wells, gently pipette mix 2-5x, and remove 200 µL of PBS. Repeat for a total of three washes.
   NOTE: When adding and removing liquid from the wells, keep the pipette tip on the edge of the well at a 45° angle to avoid disturbing the surface. Do this for all handling of the plate except when plating cells.
3. For experimental wells, prepare a solution of 0.1 mg/mL Biotinylated-BSA in PBS and add 80 µL of the solution to the desired wells. For control wells, prepare the same solution with the addition of 18.75 µg/mL fibronectin and add 80 µL to respective wells. Gently pipette mix all the wells used, place the lid on the plate, and cover with foil. Incubate at room temperature for 1 h.
4. Remove 80 µL of volume from the wells and wash the wells by repeating step 3.2. Add 80 µL of 0.1 mg/mL neutravidin in PBS solution in the same manner as step 3.3.
5. After incubation, remove 80 µL of volume from the wells and repeat step 3.2. Dilute the assembled TGTs (RAD-TGT Preparation step 2) to 0.1 µM in PBS and 80 µL solution to the wells in a similar manner as step 3.3. Either cover the wells with adhesive foil and store in the dark at 4 °C or protect from light and incubate at room temperature for 1 h before proceeding with experiments.
   NOTE: We recommend treating step 3.5 as a stopping point and wait until the next day to proceed with experiments.

4. RAD-TGT experimental protocol

1. After TGTs have had adequate time to adhere, wash the wells (see step 3.2) 3x with PBS, followed by two washes with serum-free media. Leave the media in the wells after the final wash. Add 200 µL of PBS to the peripheral wells to mitigate any edge effects and transfer the plate to a 37 °C incubator to prewarm the plate.
2. Retrieve the cells of interest at ~80% confluency, and aspirate the media followed by the addition and aspiration of PBS to wash the cells. Add an appropriate volume of trypsin and place the plate in a 37 °C cell culture incubator until the cells dissociate (~5-10 min).
3. Once the cells are dissociated, add the original volume of complete cell media used to maintain cells to the plate and resuspend with a serological pipette. Add the cell suspension to a 15 mL conical tube.
4. Centrifuge the cells at 300 × g for 4 min, aspirate the media, and resuspend the pellet in PBS. Repeat centrifugation and aspiration, then resuspend in 5 mL of serum-free media and repeat centrifugation and aspiration. Resuspend the cells in ~2 mL of serum-free media.
5. Determine the density of the resuspended cells and adjust the concentration as desired.
   NOTE: We aim to plate ~15,000 cells per well so we prepare the cells to be at a concentration of 150,000 cells/mL, which allows for 15,000 cells to be plated with 100 µL of resuspended cells.
   1. Optional: If the effect of drugs or other pharmacological agents is to be tested, add the appropriate concentration of drug to the cells and return the cells to the incubator. Allow for adequate preincubation time before proceeding to the next step.
6. Retrieve the plate from the incubator (step 4.1) and remove the volume of media left from the wash from the experimental wells in the prewarmed plate. Wash once with the desired media. Carefully pipette 100 µL of the resuspended cells to the desired wells. Transfer the plated cells to the incubator for 90 min.
   NOTE: Pipette the cells into center of the wells gently, trying to keep the pipette vertical.
7. Following incubation, remove ~125-170 µL of volume from the wells. If required, wash the cells gently with 200 µL of warmed PBS. Add 50 µL of trypsin to each well and return the plate to the incubator until the cells begin to dissociate (5-10 min).
   NOTE: Ensure the well never becomes dry and that overtrypsinization does not occur.
8. Add 160 µL of freshly prepared flow buffer (1 mM EDTA, 2% w/v bovine serum albumin, in cold PBS) to the wells. Gently pipette up and down to dissociate the cells, add 196 µL of this cell suspension to PCR tubes followed by 4 µL of propidium iodide, and gently mix. Prepare one control sample without propidium iodide to serve as a control for viability gating.
9. Place the tubes in an ice bucket with the lid blocking light and proceed to measure the fluorescence intensity with a flow cytometer.

5. Data collection and analysis

1. Analyze each sample with a flow cytometer. Aim to collect at least 2,000 live single cells per sample.
   NOTE: Minimize light exposure for samples that have not yet been analyzed and keep on ice.
2. Once all the data are collected, transfer the files to the software of choice and gate the cells.
   1. Gate for cells by forward scatter area (FSC-A) vs side scatter area (SSC-A).
   2. Gate for singlets within the cell population by FSC-A vs forward scatter height (FSC-H).
   3. Gate for live cells from the singlet population by setting a vertical gate on the histogram of propidium iodide area intensity at the highest value for the no propidium iodide control prepared in step 4.8. Any events that do not exceed the gate are live cells
   4. Set a vertical gate on a histogram of Cy5 area (Cy5-A) intensity of the singlets to remove any non-fluorescent event on the Cy5-A channel caused by cell debris. Ensure that this vertical gate is at a value less than the unlabeled cells analyzed but greater than 0 fluorescent intensity.
   5. To analyze cell populations, set a bisector gate at the 99^th percentile of each respective negative control.
   6. Apply a gate at the same intensity as the corresponding experimental conditions, and collect the percentage of cells exceeding the negative control gate for each condition.
      NOTE: Percentages can be directly compared between biological replicates via appropriate statistical analysis; see representative data.
3. Analyze the data using median Cy5 fluorescence, dividing the median intensity of ligand-WDV by the median of the WDV alone and compare these values between replicates.

Results

To demonstrate the utility and ease of use of RAD-TGTs, we analyzed the effects of Rho Kinase (ROCK) inhibitors on the mechanotype of CHO-K1 cells. The ROCK pathway is involved in actin organization and myosin activation. Previous studies using TGTs have shown that ROCK inhibition decreases the rupture of shearing TGTs^16,17,18. Shearing or high-force TGTs report on contractile forces exerted through focal adhesions, while unzipp...

Discussion

We have provided protocols on how to set up and perform RAD-TGT experiments with the aim of encouraging scientists across life sciences to incorporate mechanotyping within their workflow. A key advantage of RAD-TGTs, particularly for those with protein production capabilities, is the use of HUH endonucleases to ligate ligands to DNA duplexes with high fidelity without any chemical crosslinking and/or purification. HUH endonucleases encompass a family of nucleases that bind to single-stranded DNA in a site-specific manner...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
15 mL Conical Tube	Sarstedt	62.554.205
96 well glass bottom plate	Cellvis	P96-1.5H-N
Adhesive PCR plate foil	Thermo Fisher Scientific	AB-0626
Ampicillin	Apex Bioresearch Products	25-530
Bio-Rad 20 mL column	Bio-Rad	7321010
Biotinylated BSA	Thermo Fisher Scientific	29130
BL21 (DE3) Competent E. coli	New England Biolabs	C2527H
Bovine Serum Albumin	Thermo Fisher Scientific	BP9703100
Cell Culture Incubator	Thermo Fisher Scientific	51030284
Centrifuge	Thermo Fisher Scientific	75009521
CHO-K1	ATCC	CCL-61
Dialysis Tubing	Thermo Fisher Scientific	PI88242
DMSO	Thermo Fisher Scientific	D12345
EDTA	Thermo Fisher Scientific	S311-100
Flow Cytometer	BD	Accuri C6 Plus
FlowJo	BD	FlowJo v10.10
HEPES	Thermo Fisher Scientific	BP310-1
Imidazole	Thermo Fisher Scientific	O3196-500
IPTG	IBI Scientific	IB02125
LB agar	Fisher Scientific	BP1425-500
Ligand Stand	Integrated DNA technologies	user designed oligo	GCT ATA AAC TCA CCG TAA TTT TTT GGC CCG CAG CGA CCA CCC TTT /3Cy5Sp/
Lysogeny Broth (LB)	Thermo Fisher Scientific	BP1426-500
MnCl2	Thermo Fisher Scientific	M87-100
NaCl	Thermo Fisher Scientific	BP358-212
NeutrAvidin Protein	Thermo Fisher Scientific	31000
Nickel-NTA resin	Thermo Fisher Scientific	88222
PBS	Corning	21-031-CV
Petri Dish	VWR	25384-342
Propidium Iodide Solution	Cayman Chemicals	10008351
SDS-PAGE gel	Bio-Rad	4561096
Serum Free Media	Thermo Fisher Scientific	A4124801	Gibco Optimem Used, alternative serum free medias are fine to use
Shearing Anchor Strand	Integrated DNA technologies	user designed oligo	/5IAbRQ//iBiodT/G GGT GGT CGC TGC GGG CC
Sonicator	Thermo Fisher Scientific	15-345-139
Thermal Cycler	Bio-Rad	1861096
Tris	Thermo Fisher Scientific	BP152-500
Trypin	Thermo Fisher Scientific	12605036
ultracentrifugal filter, 10 kDa cutoff	Amikon
Unzipping Anchor Strand	Integrated DNA technologies	user designed oligo	/5IAbRQ/GG GTG GTC GCT GCG GGC C/3Bio/
WDV Plasmid	AddGene	226983
WDV-Echistatin Plasmid	AddGene	226985
Y-27632	Selleck Chemicals	S1049	Reconstituted in DMSO