Single-Cell Sorting of Immunophenotyped Mesenchymal Stem Cells from Human Exfoliated Deciduous Teeth

Summary

This protocol describes the use of fluorescence-activated cell sorting of human mesenchymal stem cells using the single-cell sorting method. Specifically, the use of single-cell sorting can achieve 99% purity of the immunophenotyped cells from a heterogeneous population when combined with a multiparametric flow cytometry-based approach.

Abstract

The mesenchymal stem cells (MSCs) of an organism possess an extraordinary capacity to differentiate into multiple lineages of adult cells in the body and are known for their immunomodulatory and anti-inflammatory properties. The use of these stem cells is a boon to the field of regenerative biology, but at the same time, a bane to regenerative medicine and therapeutics owing to the multiple cellular ambiguities associated with them. These ambiguities may arise from the diversity in the source of these stem cells and from their in vitro growth conditions, both of which reflect upon their functional heterogeneity.

This warrants methodologies to provide purified, homogeneous populations of MSCs for therapeutic applications. Advances in the field of flow cytometry have enabled the detection of single-cell populations using a multiparametric approach. This protocol outlines a way to identify and purify stem cells from human exfoliated deciduous teeth (SHEDs) through fluorescence-assisted single-cell sorting. Simultaneous expression of surface markers, namely, CD90-fluorescein isothiocyanate (FITC), CD73-peridinin-chlorophyll-protein (PerCP-Cy5.5), CD105-allophycocyanin (APC), and CD44-V450, identified the "bright," positive-expressors of MSCs using multiparametric flow cytometry. However, a significant drop was observed in percentages of quadruple expressors of these positive markers from passage 7 onwards to the later passages.

The immunophenotyped subpopulations were sorted using the single-cell sort mode where only two positive and one negative marker constituted the inclusion criteria. This methodology ensured the cell viability of the sorted populations and maintained cell proliferation post sorting. The downstream application for such sorting can be used to evaluate lineage-specific differentiation for the gated subpopulations. This approach can be applied to other single-cell systems to improve isolation conditions and for acquiring multiple cell surface marker information.

Introduction

Mesenchymal stem cells (MSCs) can be regarded as a scalable source of cells suitable for cell-based therapies and may be considered a gold standard system in regenerative medicine. These cells can be isolated from a variety of sources in the body with different tissue origins¹. Depending on their source tissue, each type of MSC displays an ambiguous in vitro behavior². This is well observed in their morphological and functional properties³. Multiple studies have shown intra-clonal variation in dimensions, including adult tissue differentiation, genomic state, and metabolic and cellular architecture of MSCs^2,4.

Immunophenotyping of cells has been a common application of flow cytometry for the identification of stem cells and this was utilized by the International Society for Cell and Gene Therapy (ISCT) in 2006 to prescribe a list of minimal criteria to identify cells as MSCs. It stated that along with plastic adherence and the ability to differentiate into three lineages (osteogenic, chondrogenic, and adipogenic) in vitro, ≥95% of the cell population must express CD105, CD73, CD90, and these cells must lack the expression (≤2% positive) of CD34, CD45, CD11b, CD14, and HLA-DR, as measured by flow cytometry⁵. Although the MSCs were defined by a set of biomarkers under the minimal criteria of the ISCT, their immune properties could not be benchmarked with these biomarkers and there was a need for more beyond these criteria to make cross-study comparisons and clonal variations easier to quantify².

Despite the guidelines set by ISCT, extensive research on MSCs has shown that heterogeneity exists in this population, which could arise due to a multitude of factors, mainly due to the ubiquitous nature of heterogeneity that arises between MSC donors⁶, tissue sources⁷, individual cells within a clonal population⁸, and culture conditions^2,9,10. Characterization and purification of these primary cells from a variety of tissue sources to ensure quality and cell fate are key steps in their production. The need to understand the displayed variations amongst the population requires an efficient method to resolve it into subpopulations that can be divided and collected separately¹¹. Single-cell level analyses help overcome the challenges of cell-cell variation, reduce biological noise arising from a heterogeneous population, and offer the ability to investigate and characterize rare cells¹².

Based on the purpose and chosen parameters, several methods can be employed to sort and enrich the selected populations. Cell-sorting techniques can comprise both bulk sorting and single-cell sorting methods. While bulk sorting can enrich target populations through Magnetic-activated cell sorting (MACS)¹³, fractionation¹⁴, and elutriation¹⁵, single-cell sorting can enrich more homogeneous populations by means of fluorescence-activated cell sorting (FACS)¹¹. A comparative analyses of each of these methods with its own set of advantages and disadvantages is highlighted in Table 1.

Table 1: Comparative analyses of different techniques: MACS, Fractionation, Elutriation, and FACS highlighting the differences in their principle and the advantages and disadvantages of choosing a particular technique over another. Abbreviations: MACS = Magnetic-activated cell sorting; FACS = Fluorescence-activated cell sorting. Please click here to download this Table.

Since the advent of the technique, single-cell flow cytometry has played a major role in enumeration¹⁶, detection, and characterization of a specific cell population in a heterogeneous sample¹⁷. Hewitt et al. in 2006 laid the foundation of automated cell sorting methodology to enhance the isolation of homogenous pools of differentiated human embryonic stem cells (hESCs)¹⁸. Single-cell sorting enriched the population of GFP-transduced hESCs facilitating the isolation of genetically modified clones, which opened a new dimension in clinical research. To improve the sort efficiency, two approaches have generally been taken; either the collection media of the sorted populations are modified to sustain viability and proliferation of post-sorted cells¹⁹ or the cell-sorting algorithm/software is appropriately modified¹².

With the advancement of technology, commercial flow cytometers and cell sorters have been able to help address challenges that were met while aseptically sorting fragile and rare cell populations, especially stem cells of different origins. One of the major challenges of stem cell biologists has been the clonal isolation of human pluripotent stem cells following transfection protocols required in gene editing studies¹⁹. This was addressed by sorting single cells into 96-well plates that were coated with Mouse Embryonic Fibroblasts (MEFs) along with supplements and commercial small molecule ROCK inhibitors. However, cell isolation strategies could be largely refined with the use of index sorting, a feature of the sorting algorithm that identifies the immunophenotype of individual cells sorted¹². This refined modality in single-cell sorting helped not only in enhancing sort efficiency for stem cells, especially with regard to rare hematopoietic stem cell populations, but also efficiently linked single-cell clones to their downstream functional assays²⁰.

This paper focuses on single-cell sorting of immunophenotyped stem cells from human exfoliated deciduous teeth (SHEDs) for the enrichment of sub-populations to study their functional differentiation capacities. Using a combination of two MSC-positive markers, CD90 and CD73, and a negative hematopoietic marker CD45, the MSCs were immunophenotyped and the dim and null expressors were identified. Based on their immunophenotype the subpopulations were identified as pure MSCs, single positive and double negative populations. They were sorted using the single-cell sort mode to obtain pure and enriched subpopulations for further functional studies to identify whether the differential expression of markers was an artifact of in vitro culture conditions or whether it has any effect on the functional properties as well. Cells that were not homogeneous expressors of the "positive MSC markers" were sorted to study their functional properties.

Protocol

Ethics approval and consent to participate: Human exfoliated deciduous dental pulp samples were received after obtaining informed consent and full ethical approval by Sri Rajiv Gandhi Dental College and Hospital (SRGCDS) Oral and Maxillofacial Department, Bengaluru, in accordance with the standards established by the Hospital Ethical Clearance Committee, SRGCDS. Following which isolation, culture, maintenance, and application of SHEDs were approved by and in compliance with the guidelines recommended by the Institutional Committee for Stem Cell Research (IC-SCR) at Manipal Institute of Regenerative Medicine, MAHE - Bengaluru. See the Table of Materials for details about all materials and reagents used in this protocol.

1. Preparation of reagents and buffers

1. For culture maintenance
   1. Prepare MSC cell culture media (10%) using basal medium for undifferentiated hESCs, 10% fetal bovine serum (FBS), 1% L-glutamine, and 1% penicillin-streptomycin (Pen-strep) (Table 2).
   2. Prepare MSC cell culture media (20%) using basal medium for undifferentiated hESCs, 20% FBS, 1% L-glutamine, and 1% Pen-strep (Table 2).
   3. Prepare neutralizing media using basal medium for undifferentiated hESCs, 1% L-glutamine, and 1% antibiotic-antimycotic (Anti-Anti) (Table 2).
2. For flow cytometric analysis and sorting
   1. Prepare staining buffer using 2% FBS in phosphate-buffered saline (PBS).
   2. Prepare a stock solution of 4′,6-diamidino-2-phenylindole (DAPI) (1 µg/mL) by adding 1 µL in 1 mL of PBS
3. For lineage-specific differentiation of MSCs
   1. Prepare differentiation media for osteogenic, chondrogenic, and adipogenic differentiation according to the composition described in Table 3. Store the prepared aliquots at 4 °C for the duration of the experiment.
   2. Prepare serum-starved media (2% media) using basal medium for undifferentiated hESCs, 2% FBS, 1% L-glutamine, and 1% Pen-strep (Table 2).

TYPE OF MEDIA			PURPOSE OF MEDIA			COMPOSITION FOR 50 mL
						FBS	Pen-Strep	L-Glutamine	BASAL MEDIUM FOR UNDIFFERENTIATED hESCs
10% media			MSC culture and maintenance			5 mL	500 μL	500 μL	44 mL
20% media			CFU-F assay			10 mL	500 μL	500 μL	39 mL
Serum starved (2%) media			Media for Control wells in Differentiation protocol			1 mL	500 μL	500 μL	48 mL
Neutralizing media			Media for neutralizing the cell suspension after trypsinization			-	500 μL	500 μL	49 mL

Table 2: Cell culture media for culture maintenance and assays. Abbreviations: MSC = mesenchymal stem cell; CFU-F = colony-forming unit-fibroblast.

COMPONENTS	OSTEOGENIC MEDIA	CHONDROGENIC MEDIA	ADIPOGENIC MEDIA
Basal Media	90 mL	90 mL	90 mL
Induction media	10 mL	10 mL	10 mL
Total Volume	100 mL	100 mL	100 mL

Table 3: Differentiation media for trilineage differentiation of SHEDs.

2. Culture and maintenance of SHEDs

1. Maintain cells in 10% MSC culture media and perform media changes every 2 days or as required.
2. Trypsinize cells at 95% confluency using 0.25% trypsin-EDTA.
3. Neutralize cells after trypsinization using neutralizing media.
4. Centrifuge the tube at 300 × g for 6 min at room temperature to obtain a cell pellet.
5. Decant the supernatant and resuspend the cell pellet in 10% MSC culture media.
6. Seed cells into freshly prepared cell culture dishes containing 10% MSC culture media for further experiments or sub-culturing.
   NOTE: Optimal seeding density for SHEDs is 0.2 × 10⁶ cells in a 100 mm dish and 0.8 × 10⁶ cells in a T-75 flask, to obtain 1.5 × 10⁶ cells and 4 × 10⁶ cells, respectively, at 90-100% confluency.

3. Characterization of MSCs

1. Short-term cell growth assay
   1. Seed 4 × 10⁴ cells/well in triplicate into 6-well plates in 10% MSC culture media.
   2. Incubate the plates for 7 days at 37 °C, and perform media changes every 2 days.
   3. Harvest the cells on days 2, 4, and 8 using 0.25% trypsin treatment and wash with culture media.
   4. Centrifuge the cells at 300 g for 6 min at room temperature and resuspend the pellet in 1 mL of media.
   5. Count the cells using a hemocytometer and determine their viability by the trypan blue exclusion method.
   6. Calculate the proliferation rate using equation (1):
          (1)
2. Colony-forming unit-fibroblast (CFU-F) assay
   1. Seed 10,000 cells in a 100 mm dish and culture them in 20% MSC culture media.
   2. Incubate the plates for 14 days at 37 °C, and perform media changes every 3 days.
   3. After 14 days, rinse the colonies with PBS, fix them with crystal violet dye in methanol, and rinse again with PBS to remove the residual stain.
   4. Count and image the colonies.
      NOTE: Count colonies with >50 cells. Colony-forming efficiency is calculated as colony-forming unit numbers.
3. Immunofluorescence assay
   1. Seed the MSCs on 35 mm dishes and let them grow till 80-90% confluency.
   2. Remove the media and rinse the dishes with PBS once.
   3. Fix the cells with 1 mL of 4% paraformaldehyde (PFA) by either incubating for 1 h at room temperature or overnight at 4 °C.
   4. After fixation, wash the wells with PBST for 3 x 5 min on the rocker.
   5. Add 0.3% Triton X-100 in PBST (0.05% Tween 20 solution in PBS) for permeabilizing the cells. Keep it on the rocker for 15 min at room temperature.
   6. Wash the wells with PBST for 3 x 5 min on the rocker.
   7. Add 3% bovine serum albumin (BSA) for blocking and keep it on the rocker for 1 h at room temperature.
   8. Wash the wells with PBST for 3 x 5 min on the rocker.
   9. Add 800 µL of 1:500 dilution of mouse anti-vimentin, keep it on the rocker for 1 h at room temperature, and transfer the plate to 4 °C for overnight incubation.
   10. Next day, remove the primary antibody and wash the wells with PBST for 3 x 5 min on the rocker.
   11. Add 800 µL of 1:1,000 dilution of goat anti-mouse IgG (H+L) cross-adsorbed secondary antibody, Alexa Fluor 555, and keep it for 3 h at room temperature on the rocker.
   12. Wash the wells with PBST for 3 x 5 min on the rocker.
   13. Add Alexa fluor 488 Phalloidin Probes 240 µL in 1,000 µL of PBS and incubate at room temperature for 60 min on the rocker.
   14. Wash the wells with PBST for 3 x 5 min on the rocker.
   15. Add 700 µL of DAPI mountant and observe the cells under a microscope.
4. Lineage-specific differentiation
   1. Differentiation following 2D culture
      1. Take two 48-well plates and label them as osteogenic and adipogenic lineages, respectively.
      2. Seed 15,000 cells/well in four wells of each plate and culture them in 10% MSC culture media.
      3. Once the monolayer of cells has reached 90% confluency, label the first two wells as 'Control' and replace the existing media with serum-starved media (2% media). In the last two wells labeled as 'Test,' add differentiation media of either of the two lineages, adipogenic or osteogenic. Mark this as Day 0.
      4. Gently replace media every third day to circumvent peeling and be careful to avoid contamination.
      5. Maintain these conditions till the twenty-first day; then process for the staining experiment.
   2. Differentiation following 3D culture
      1. Take two 15 mL tubes for performing chondrogenic differentiation using 3D pellet culture.
      2. Transfer 1 × 10⁶ cells to each tube and centrifuge at 300 × g for 6 min to form a pellet. Label one tube as 'Control' and add 10% MSC culture media to it; label the other tube as 'Test' and add chondrogenic differentiation media. Place the tubes carefully in the incubator with their caps loosely screwed. Mark this as Day 0.
      3. Change the media every third day carefully, so as not to dislodge/disintegrate the pellet during media changes.
      4. Maintain these conditions till the twenty-first day and then, process the cells for further experiments.
5. Lineage-specific cytochemical staining
   1. For 2D cultures, use the differentiated (test) and undifferentiated MSCs (control) plates (from step 3.4.1.5.) for staining, and first, remove the media and wash twice with PBS. Fix the cells using 4% PFA for 30 min at room temperature, remove the supernatant, and wash once with PBS. Perform staining for each lineage as follows.
      1. For adipocyte lineage, add permeabilization solution from the kit and incubate the plate for 5 min at room temperature. Prepare and add 1 mL of Oil red O working solution and keep it for 10 min. Remove the stain and give five washes with distilled water.
      2. For osteocyte lineage, add 5% of freshly prepared silver nitrate (in distilled water) to each well and keep the plate under UV for 1 h. Remove the solution and add 2.5% of sodium thiosulphate to remove the unreacted silver; keep it for 5 min. Remove the stain, wash twice with distilled water, and observe the stained cells under a microscope.
   2. For 3D cultures, collect the pellet after the end of the differentiation period from step 3.4.2.3 and obtain cryosections of the differentiated tissue in the form of the pellet. Allow the slides to air dry and be at room temperature before proceeding to the staining.
      1. For chondrocyte lineage, follow the directions of the staining kit to add a sufficient volume of washing solution, remove it, add the fixing solution, and incubate for 30 min. Wash with distilled water, add the staining solution, and incubate for 30 min. Wash thrice with 0.1 N hydrochloric acid; add distilled water to neutralize the acidity. Observe the stained cells under a brightfield microscope.

4. Cell surface staining for immunophenotyping

NOTE: Recommended cell culture plates for getting an optimal number of cells in steps 4.2-4.5 are 100 mm dishes or T75 flasks.

1. Cell preparation for flow cytometric experiments
   1. Trypsinize and collect the cells from a confluent dish/ flask and centrifuge at 300 × g for 6 min to obtain the cell pellet.
   2. Resuspend the pellet in 1 mL of media and determine the viable cell count using a hemocytometer following the trypan blue exclusion method.
   3. Centrifuge the cell suspension again after counting and give two more washes to the pellet with 1 mL of staining buffer.
   4. Discard the supernatant and finally resuspend the pellet in an appropriate volume of the staining buffer depending on the protocol (see steps 4.2 to 4.5).
2. Preparation of compensation controls
   1. Take seven FACS tubes and label them as Unstained, DAPI, V450, FITC, PE, PerCP-Cy 5.5, and APC.
   2. Resuspend the final pellet in the staining buffer (see step 4.1.) keeping a cell density of 0.5 × 10⁶ cells per 50 µL per tube for the Unstained and DAPI tubes.
   3. Prepare the single-stained tubes for compensation as described in Table 4.
   4. Gently vortex each tube after preparation and incubate in the dark for 30 min.
   5. After incubation, give two washes by adding 1 mL of staining buffer to each tube, followed by brief vortexing and centrifugation at 200 g for 10 min at room temperature.
   6. Discard the supernatant, resuspend the pellet in 500 µL of staining buffer, and set it aside until the acquisition.
   7. For the DAPI tube, perform heat shock treatment by incubating it in a 60 °C water bath for 5 min followed by 15 min on ice. Add 5 µL of DAPI to the suspension and keep it in the dark until the run acquisition.
3. Preparation of fluorescence minus one (FMO) controls
   1. Resuspend the final pellet in staining buffer (see step 4.1) keeping a cell density of 0.5 × 10⁶ cells per 50 µL for each tube.
   2. Take five FACS tubes, and label them as CD44-V450 isotype, CD90-FITC, CD45-PE, CD73-PerCP Cy5.5 isotype, and CD105-APC isotype.
   3. Prepare the cell and antibody suspension according to Table 5.
   4. Vortex the tubes gently and incubate them for 30 min at room temperature in the dark.
   5. After incubation, give two washes with 1 mL of staining buffer to each tube followed by brief vortexing and centrifugation at 200 g for 10 min at room temperature.
   6. Discard the supernatant, resuspend the pellet in 500 µL of staining buffer, and set it aside until acquisition.
4. Preparation of samples for analysis
   1. Resuspend the final pellet in the staining buffer (see step 4.1) keeping a cell density of 0.5 × 10⁶ cells per 50 µL for each tube.
   2. Take two FACS tubes and label them as Mixed tubes 1 and 2.
   3. Prepare the cell and antibody suspension according to Table 6.
   4. Vortex the tubes gently and incubate them for 30 min at room temperature in the dark.
   5. After incubation, give two washes with 1 mL of staining buffer to each tube followed by brief vortexing and centrifugation at 200 g for 10 min at room temperature.
   6. Discard the supernatant, resuspend the pellet in 500 µL of staining buffer, and set aside until acquisition.
5. Preparation of samples for single-cell sorting
   1. Take two FACS tubes, add 1mL of FBS, and roll the tube around till an even layer of FBS is formed on the inside of each tube. Incubate this for 1-2 h while processing the sample for staining. Label these tubes as Mixed tubes 1 and 2.
   2. Resuspend the final pellet in the staining buffer (see step 4.1) keeping a cell density of 2-3 × 10⁶ cells per 50 µL for each tube.
   3. Prepare the cell and antibody suspension in Mixed tubes 1 and 2 according to Table 7.
   4. Vortex the tubes gently and incubate them for 30 min at room temperature in the dark.
   5. After incubation, give two washes with 1 mL of staining buffer to each tube followed by brief vortexing and centrifugation at 200 g for 10 min at room temperature.
   6. Discard the supernatant, resuspend the pellet in 500 µL of staining buffer, and set aside. Add 5 µL of DAPI 15 min prior to sorting.
      NOTE: Note that for sorting experiments higher cell density is recommended per tube.

Tube type	Positive Comp Beads*	Negative Comp Beads*	Cells	Antibody added
FITC tube	1 drop	1 drop	–	Anti-human CD90-FITC (2 µL)
V450 tube	1 drop	1 drop	–	Anti-human CD44-V450 (2 µL)
PerCP-Cy 5.5 tube	1 drop	1 drop	–	Anti-human CD73-PerCP Cy 5.5 (2 µL)
PE tube	1 drop	1 drop	–	Anti-human CD45-PE (2 µL)
APC tube	1 drop	1 drop	–	Anti-human CD105-APC (2 µL)
DAPI tube	–	–	50 µL	–
Unstained tube	–	–	50 µL	–
*1 drop= 60 µL of bead suspension

Table 4: Compensation control samples. Abbreviations: Comp = compensation; DAPI = 4',6-diamidino-2-phenylindole; FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein.

TUBE TYPE	ANTIBODY AGAINST POSITIVE MARKER (2 µL)			ANTIBODY AGAINST NEGATIVE MARKER  (2 µL)	ISOTYPE ANTIBODIES ADDED  (2 µL)	TOTAL VOLUME OF ANTIBODIES	VOLUME OF CELL SUSPENSION ADDED	VOLUME OF STAINING BUFFER ADDED
CD90-FITC FMO tube	- Anti-human CD44-V450			Anti-human CD45-PE	FITC IgG1 isotype	10 µL	50 µL	40 µL
	- Anti-human CD73 PerCP Cy 5.5
	- Anti-human CD105-APC
CD73-PerCP Cy5.5 FMO tube	- Anti-human CD44-V450			Anti-human CD45-PE	PerCP Cy 5.5 IgG1 isotype	10 µL	50 µL	40 µL
	- Anti-human CD90-FITC
	- Anti-human CD105-APC
CD44-V450 FMO tube	- Anti-human CD90-FITC			Anti-human CD45-PE	V450 IgG1 isotype	10 µL	50 µL	40 µL
	- Anti-human CD73-PerCP Cy 5.5
	- Anti-human CD105-APC
CD105-APC FMO tube	- Anti-human CD44-V450			Anti-human CD45-PE	APC IgG1 isotype	10 µL	50 µL	40 µL
	- Anti-human CD90-FITC
	- Anti-human CD73-PerCP Cy 5.5
CD45-PE FMO tube	- Anti-human CD44-V450			-	PE IgG1 isotype	10 µL	50 µL	40 µL
	- Anti-human CD90-FITC
	- Anti-human CD73-PerCP Cy 5.5
	- Anti-human CD105-APC

Table 5: FMO control samples. Abbreviations: FMO = fluorescence minus one; FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein.

TUBE TYPE	ANTIBODY AGAINST POSITIVE MARKER (2 µL)		ANTIBODY AGAINST NEGATIVE MARKER (2 µL)	TOTAL VOLUME OF ANTIBODIES	VOLUME OF CELL SUSPENSION ADDED	VOLUME OF STAINING BUFFER ADDED
Mixed tube 1	- Anti-human CD44-V450		Anti-human CD45-PE	10 µL	50 µL	40 µL
	- Anti-human CD90-FITC
	- Anti-human CD73-PerCP Cy5.5
	- Anti-human CD105-APC
Mixed tube 2	- Anti-human CD44-V450		Anti-human CD45-PE	10 µL	50 µL	40 µL
	- Anti-human CD90-FITC
	- Anti-human CD73-PerCP Cy5.5
	- Anti-human CD105-APC

Table 6: Sample tubes for multicolor immunophenotyping of SHEDs. Abbreviations: SHEDs = stem cells from human exfoliated deciduous teeth; PE = phycoerythrin.

TUBE TYPE	ANTIBODY AGAINST POSITIVE MARKER (3 µL)		ANTIBODY AGAINST NEGATIVE MARKER (3 µL)	TOTAL VOLUME OF ANTIBODIES	VOLUME OF CELL SUSPENSION ADDED	VOLUME OF STAIN BUFFER ADDED
Mixed tube 1	- Anti-human CD90-FITC		Anti-human CD45-PE	9 µL	50 µL	41 µL
	- Anti-human CD73-PerCP Cy5.5
Mixed tube 2	- Anti-human CD90-FITC		Anti-human CD45-PE	9 µL	50 µL	41 µL
	- Anti-human CD73-PerCP Cy5.5

Table 7: Single-cell sorting reaction tubes. Abbreviations: FITC = fluorescein isothiocyanate; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein.

5. Single-cell sorting

1. Preparation of cell sorter
   1. Install a 100 µm nozzle in the sorter.
      NOTE: The appropriate nozzle is at least five times the diameter of the particle to be sorted. The sheath fluid to be used for sorting needs to be decided based on the sample type and the sensitivity of the experiment; for this experiment, proprietary sheath fluid has been used.
   2. Perform the daily instrument quality check (QC) and set up the sorter for the experiment. Refer to the instrument manual for a detailed guide to instrument setup.
2. Setting up the compensation matrix
   1. Set the compensation matrix using single-stain compensation tubes from step 4.2.
   2. In the proprietary software, select Experiment from the Toolbar and click on Compensation setup. Open Create compensation controls.
   3. Check the markers and confirm. A new specimen is added named "Compensation" under which new tubes named the marker controls are added automatically by the software.
   4. Select the Unstained tube and run it, to record 5,000 events. Drag the gate to the population of cells and apply it to all compensation controls. This is to set the voltages and negative gate for each fluorescent parameter.
   5. Similarly, load the single-stain compensation tubes separately, and record and save the data. Select the gate demarcating the population of interest and apply to all compensation controls. This is to set the positive gates for each fluorescent parameter.
   6. Select Experiment from the Toolbar and click on Calculate compensation values | link and save.
      NOTE: Once the auto-comp matrix is generated using the software, the voltage parameters of the fluorochromes cannot be changed for any of the channels in the Mixed tubes.
3. Data acquisition
   1. Record 10,000 cells in each tube from steps 4.3. and 4.4 to collect data for analysis of the immunophenotype of the cells.
4. Preparation of the collection devices
   1. Depending on the purpose of the sorted cell populations, choose between 6-well, 24-well, 48-well, or 96-well plates.
   2. Coat the wells with 200-500 µL of FBS and keep the plates undisturbed for 2 h.
   3. After 2 h, remove the residual FBS and add 200-500 µL of 10% MSC culture media.
5. Cell sorting in single-cell sort mode
   1. Run Mixed tube 1 (from step 4.5), and record 10,000 events to set the gates on the population of interest to be sorted, using the appropriate gating strategy.
   2. Load a collection plate and set target cell numbers between 2,500 and 5,000 cells/well and select the single-cell sort purity mask.
   3. Collect the sorted populations in the collection device and keep them on ice until the end of the sorting experiment.
   4. Once done, transfer the plates to the 5% CO₂ incubator to maintain the cultures at 37 °C.
      NOTE: Post acquisition, raw data files were exported as .fcs file format (v.3.0. onwards). The sort reports generated after every experiment recorded the number of events/cells sorted per assigned well and indicated the number of conflicts that were aborted.

Representative Results

The SHEDs were characterized with standard immunofluorescence assays showing the expression of vimentin (red, type III intermediate filaments), actin filaments (Alexa fluor 488 Phalloidin Probes), and nuclei stained with DAPI (Figure 1A). To estimate their proliferative and colony-forming capacities, standard short-term cell growth assays were performed. A 14.3-fold increase in proliferation rate from day 2 to day 8 has been shown in Figure 1B. The clonogenic properties of the SHEDs were determined from the CFU-F Assays shown in Figure 1C-E. As per the ISCT criteria, the multipotent MSCs were able to differentiate into the three mesodermal-derived lineages. Cytochemical staining with Oil Red O was used to detect the oil droplets in the differentiated adipocytes (Figure 1F); von Kossa staining was used to stain the calcium deposits found in the matrix of the differentiated osteoblasts (Figure 1G); and the aggrecans in the 3D culture were stained for Alcian Blue in the differentiated chondroblasts (Figure 1H).

To characterize the immunophenotype of the SHEDs, multiparametric flow cytometry assays were set. The experiments were performed along with the use of four types of experimental controls, namely compensation controls (Table 4), FMO and isotype controls (Table 5), and unstained controls. The isotypes of all the MSC and HSC antibodies used were added along with the FMO tubes to discriminate positive versus negative signals and high versus dim antigen expression. Figure 2A shows the distribution of the FMO and isotype controls used per experiment. The representative density plots showed no expression of the FITC isotype of CD90 FITC antibodies. Similarly, the histogram overlays (Figure 2B) compared the expression levels of the stained samples along with their controls in the respective channels of FITC, PerCP Cy5.5, and PE. Positive expression of the markers CD90 and CD73 (red histograms) and negative expression of CD45 are comparable with that of their unstained and FMO controls (blue and black histograms, respectively).

Flow cytometry-based characterization of the immunophenotypes has shown the marker distribution from one of the early passages (P6) SHEDs (Figure 3A-H). Both the ISCT-recommended markers and CD44 determined the parent populations as MSCs. The tetra-positive population demonstrated in the density plots showed ≥98% expression of CD90⁺CD73⁺CD105⁺CD44⁺ and ≤2% expression of CD45 markers. More than five independent experiments showed similar results. The late-passage SHEDs (P12), which showed differential expression of the positive markers, were chosen for single-cell sorting of high and dim expressors (Figure 4A-H). Following the gating strategy (Figure 4H), singlets→ Live cells→ Non-HSCs→ Double positive CD73⁺CD90⁺ and single positive CD73⁺CD90^- populations were sorted using the single-cell sort mode. The sorted populations were collected in 96-well/48-well/6-well plate layouts with different seeding counts per well (Supplementary File 1). Three consecutive attempts were made in a single phase of sorting experiments and the survivability of cells post sort (in the double-positive expressors: CD90⁺CD73⁺) per experiment was 100%. Three independent phases of experiments have been conducted. We matched the number of cells that grew per well post sort to the number of cells sorted per well.

Figure 5A shows the CD90⁺CD73⁺ post-sorted cells collected in the 96-well plate showed adherence and proliferation like their parent population. The sorted cells have been observed to reach 70-80% confluency on the sixth day post sorting. The adhered and proliferating post-sorted cells were differentiated in the presence of adipogenic growth factors and then stained post Day 15 with Oil Red O using appropriate post-sorted control (undifferentiated) cells as well (Figure 5B,C).

Figure 1: Characterization of stem cells from human exfoliated deciduous teeth. (A) Immunofluorescence staining confirms that the isolated MSCs express vimentin (red, type III intermediate filaments), actin filaments (Alexa fluor 488 Phalloidin Probes), and DAPI (blue, nuclei). Photomicrograph obtained at original magnification of 20x. Scale bar = 10 µm. (B) Short-term cell growth assay of SHEDs exhibits their high proliferative capacity, as observed by the significant increase in the number of cells in culture by day 2 and a 14.3-fold increase in proliferation rate at the end of day 8 (n = 3, p-value<0.01). Standard deviation error bars have been shown. (C-E) CFU-F assay and Crystal violet staining confirm the colony-forming ability of SHEDs; (C) multiple colonies are expected to form after 14 days in culture; (D) single clonogenic colony originating from a single cell; (E) Crystal violet-stained single colony of SHEDs displays the typical fibroblast-like morphology of the cells. (F-H) MSCs under appropriate in vitro conditions are expected to undergo differentiation into adipogenic, osteogenic, and chondrogenic lineages by day 21. Cytochemical staining shows (F) Oil Red O (inset shows 40x image of adipocyte showing oil droplets), (G) von Kossa, and (H) Alcian Blue staining for adipogenic, osteogenic, and chondrogenic lineages, respectively, at Day 22. Photomicrographs are obtained at an original magnification of 20x. Scale bar = 10 µm. Abbreviations: MSCs = mesenchymal stem cells; DAPI = 4',6-diamidino-2-phenylindole; SHEDs = stem cells from human exfoliated deciduous teeth; CFU-F = colony-forming unit-fibroblast. Please click here to view a larger version of this figure.

Figure 2: Negative controls used in multiparametric experiment. (A) Fluorescence Minus One controls have been displayed in a tabular form, where FMO controls have been replaced by their specific isotypes. The representative density plots (left to right) show the unstained control (left), FMO of CD90 FITC (middle) with all antibodies added except CD90 FITC along with the addition of CD90 FITC isotype, and fully stained sample with all antibodies of the panel added (right). (B) Histogram overlay represents the comparison of three types of samples, black represents FMO, blue represents unstained, and red represents fully stained samples in their respective channels. Abbreviations: FMO = Fluorescence Minus One; FITC = fluorescein isothiocyanate; PerCP = peridinin-chlorophyll-protein; PE = phycoerythrin. Please click here to view a larger version of this figure.

Figure 3: Multiparametric immunophenotyping of SHEDs (P6). (A) SSC-A versus FSC-A: Based on their side and forward scatter properties, cells of interest are gated and separated from the debris; (B) FSC-H versus FSC-A and (C) SSC-H versus SSC-A:The two plots are employed for doublet discrimination, which enables the selection of singlets while eliminating aggregates that may generate false positive signals; (D) SSC-A versus CD45 PE: Exclusion gating allows selection of the CD45^- population from the singlets; (E) CD73 PerCP-Cy5.5 versus CD90 FITC: From the CD45^- live singlets, CD90⁺CD73⁺ cells are gated; (F) CD44 V450 versus CD105 APC: The CD90⁺CD73⁺ population, is further gated to define the tetra-positive CD90⁺CD73⁺CD44⁺CD105⁺ cells; (G) SSC-A versus Time(s): The plot depicts the stable acquisition over time (seconds) and that no events caused pressure fluctuations while acquisition; (H) Gating hierarchy. Abbreviations: SHEDs = stem cells from human exfoliated deciduous teeth; FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein; SSC-A = side scatter-peak area; FSC-A = forward scatter-peak area; FSC-H = forward scatter-peak height. Please click here to view a larger version of this figure.

Figure 4: Single-cell sorting of immunophenotyped SHEDs (P12). (A-F) A step-by-step exclusion-based gating strategy to sort pure MSCs in the following sequential order: Cells to singlets_1, to singlets_2, to DAPI^- live cells, to CD45^- non-Hematopoietic cells, and finally to CD90⁺CD73⁺ cells. The gated populations shown in (F) were sorted using single-cell sort mode. (G) The SSC-A versus Time(s) plot describes the uninterrupted sorting of the samples. (H) Gating hierarchy. (n=3, Sort efficiency = 100%). Abbreviations: SHEDs = stem cells from human exfoliated deciduous teeth; MSCs = mesenchymal stem cells; DAPI = 4',6-diamidino-2-phenylindole; FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein; SSC-A = side scatter-peak area. Please click here to view a larger version of this figure.

Figure 5: Morphological and functional characterization of post-sorted SHEDs. (A) Phase contrast images of post-sorted populations were collected in a 96-well plate on Day 6 at an original magnification of 10x, Scale bar = 10 µm (inset shows a 20x image depicting the sorted single cells changing morphology into fibroblast-like MSCs). Cytochemical staining for adipogenic differentiation using Oil Red O observed in (B) undifferentiated (control) and (C) differentiated post-sort population at an original magnification of 20x, Scale bar = 10 µm (inset shows a 40x image depicting the oil droplets observed). Abbreviations: SHEDs = stem cells from human exfoliated deciduous teeth. Please click here to view a larger version of this figure.

Supplementary File 1: Compiled sort reports generated per experiment. Please click here to download this File.

Supplemental Table S1: Optical Configuration of BD FACSAria Fusion. Abbreviations: LP = long pass; PMT = photomultiplier tube; DAPI = 4',6-diamidino-2-phenylindole; FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein. Please click here to download this File.

Discussion

In the field of tissue engineering and regenerative medicine, among the postnatal sources, oral tissue-derived MSCs have attracted profound interest because of their minimal ethical obligations and notable multilineage differentiation potential²¹. Dental pulp stem cells (DPSCs) from the impacted third molar and SHEDs have garnered the most attention among dental MSCs for their therapeutic potential in neurodegenerative and traumatic diseases²². The protocol described in this manuscript helps characterize SHEDs using a multiparametric approach and further sort populations of interest using the single-cell sort approach. However, the application of this protocol is not only limited to oral MSCs but also can be used for immunophenotyping and sorting of MSCs from other source tissues in the human body²³.

Several critical steps in this protocol need to be discussed here for its successful implementation for the biological problem addressed. The key to every sorting protocol is to check two important components: its biological controls and the experimental controls. To begin with, clean culture maintenance ensures that the sample used does not get compromised while being sorted. The cell viability in the single-cell suspension pre-sorting shall ensure that more than 70% viable cell populations are started with, followed by analysis and sorting of healthy subpopulations. The stain buffer used in the single-cell suspension should contain 2% FBS and experiments are best done at room temperature. Selection of the correct confluence of these primary cells (80-85% confluency) avoids compromising their cell viability and reduces the chances of large aggregates. Additionally, the use of a live/dead marker (DAPI in this protocol) is essential for every sorting experiment. In this study, as the samples used were MSCs with an approximate diameter of 20-35 μm, the nozzle size chosen was 100 μm to sort them at a pressure of 20 psi on the BD FACSAria Fusion. This allowed cell sorting without compromising the cell viability during the process at lower sheath pressure. For detailed sort setup conditions, it is recommended to refer to the software BD FACSDiva guide²⁴.

An essential prerequisite for multicolor experiments in flow cytometry is the need for appropriate controls; there are four major types of controls used here: autofluorescence, isotypes, FMO, and compensation controls. The auto-compensation matrix allowed accurate compensation, preventing spectral bleedthrough in the detectors, and used both compensation beads and cells for this purpose. The unstained sample set the threshold for autofluorescence of the MSCs. The mixed tubes with all four antibodies include the isotype and FMOs of each antibody used. DAPI compensation control was performed using MSCs, which were processed separately (with heat shock). The antibody panel (as indicated in Table 8) used here is specifically designed for hMSCs based on the criteria stated by the ISCT and the instrument configuration of the BD FACSAria Fusion cell sorter (refer to Supplemental Table S1). This is very crucial during the planning of multiparametric flow cytometry experiments. The characterization panel used for immunophenotyping consists of four positive MSC markers (CD90, CD105, CD73, and CD44) and one negative marker (CD45). In contrast, the panel for the sorting experiment contains two positive markers (CD90 and CD73), one negative marker (CD45), and a live/dead marker (DAPI) to ensure a high count of live MSCs.

Antibody/Marker	Fluorochrome	hMSC Surface Profile	Laser Wavelength (nm)	Excitation peak (nm)	Emission peak (nm)
CD44	V450	Should express; >95%	405	405	450
CD90	FITC		488	495	519
CD73	PerCP-Cy5.5		488	488	695
CD105	APC		640	650	660
CD45	PE	Should NOT express; <2%	561	566	574
DAPI	-	Dead cell marker	405	358	461

Table 8: Multicolor panel designed for characterization and sorting of SHEDs by flow cytometry. Abbreviations: FITC = fluorescein isothiocyanate; APC = allophycocyanin; PE = phycoerythrin; PerCP = peridinin-chlorophyll-protein.

Selection of the sort purity mask
Considering the unpredictability of cellular heterogeneity in MSCs, bulk sorting may not necessarily achieve the level of purity among the sorted populations. We optimized the single-cell sorting protocol in this study to identify the single-cell clones and track their functional properties post sort in terms of potency. Single-cell sorting was performed on a BD FACSAria Fusion, which had the flexibility of the following settings: choice of 100 µm nozzle, 20 psi sheath pressure, a flow rate that is adjustable (maintained at 20 μL/min), and threshold rate at 380 events/s. The chances of selecting two target cells instead of one are minimal in this method. Normally, the single-cell sort mode of the algorithm is selected that can resolve the sorting of pure populations of target cells and consider how many drops will be sorted into the collection device of choice. The Precision mode ultimately should be accurately chosen such that it can combine the masks available in the sorting algorithms and create a stringent sorting condition to sort the purest populations of interest. Single-cell sort mode does not compromise on a target cell if it is aggregated to a non-target cell while sorting and that achieves the level of purity we look forward to in this approach. This was the module selected in this protocol and the sorted cells were collected in 96-well, 48-well, 24-well, and 6-well plate devices.

In this study, we sorted subpopulations based on their differential immunophenotype, namely CD90⁺CD73⁺CD45^-. In both characterization and sorting protocols, our gating strategy (as shown in Figure 3 and Figure 4) has been based on exclusion gating from the CD45^- population to give us the non-hematopoietic population (as these have been validated and characterized prior with the ISCT prescribed markers), after the identification of singlets. Eventually, the MSCs were identified from the CD45^- population using the positive ISCT markers. It is worth noting here that there is not much difference in percentages in first gating the positive expressors of ISCT markers followed by the negative expressors of CD45. However, as the purpose of this sorting has been to identify cells that show altered expression of the known MSC markers, we have concluded our analysis with the positive expression plots.

In the representative figures 97% of the CD45^- cells are MSCs, but during quantification of the antigen expressions, we have seen 75% of the cells are double-positive expressors of the two markers CD90 and CD73, while there are several populations that are single-positive expressors of the same markers (shown in Figure 3F) and few cells which do not express the markers at all.

During the characterization of these cells, we have identified the subpopulations as tetra-positive MSCs, triple-positive and double-negative populations. This differential expression of MSC markers expands with increasing passages and thus the need comes to study and correlate this difference with their functional properties. The ultimate purpose of sorting MSCs in our study is to create a comparative analysis between the identified subpopulations and determine whether the differential expressions of positive markers have functional implications.

Materials

Name	Company	Catalog Number	Comments
Alcian Blue Stain	HiMedia	CCK029-1KT
Antibiotic-Antimycotic (100x)	Gibco by ThermoFisher	15240062
BD CompBead Plus Anti-Mouse Ig, κ/Negative Control (BSA) Compensation Plus (7.5 µm) Particles Set	BD Biosciences	560497
BD FACS Accudrop Beads	BD Biosciences	345249	Used to set up the Laser delay when the sort module opens.
BD FACS Aria Fusion Flow cytometer	BD Biosciences	---
BD FACS Diva 9.4	BD Biosciences	---
BD FACS Sheath Fluid	BD Biosciences	342003	Used as sheath fluid for both analysis and sorting experiments in the BD FACSAria Fusion
BD FACSDiva CS&T Research Beads	BD Biosciences	655050	Used for Instrument configuration depending on the nozzle size.
BD Horizon V450 Mouse Anti-Human CD44	BD Biosciences	561292
BD Horizon V450 Mouse IgG2b, κ Isotype Control	BD Biosciences	560374	CD44-V450 isotype
BD Pharmingen APC Mouse Anti-Human CD105	BD Biosciences	562408
BD Pharmingen APC Mouse IgG1, κ Isotype Control	BD Biosciences	555751	CD105-APC isotype
BD Pharmingen DAPI Solution	BD Biosciences	564907	DAPI Stock solution of 1 mg/mL
BD Pharmingen FITC Mouse Anti-Human CD90	BD Biosciences	555595
BD Pharmingen FITC Mouse IgG1, κ Isotype Control	BD Biosciences	555748	CD90-FITC isotype
BD Pharmingen PE Mouse Anti-Human CD45	BD Biosciences	555483
BD Pharmingen PE Mouse IgG1, κ Isotype Control	BD Biosciences	555749	CD45-PE isotype
BD Pharmingen PerCP-Cy 5.5 Mouse Anti-Human CD73	BD Biosciences	561260
BD Pharmingen PerCP-Cy 5.5 Mouse IgG1, κ Isotype Control	BD Biosciences	550795	CD73-PerCP-Cy 5.5 isotype
BD Pharmingen Purified Mouse Anti-Vimentin	BD Biosciences	550513
Bovine serum albumin	Hi-Media	TC548-5G
Crystal violet	Nice chemical pvt ltd	C33809
Dulbecco's Phosphate Buffered Saline	Sigma-aldrich	D5652-50L	dPBS used for culture work and maintenance.
Ethanol	---	---	Used for general sterlization.
Fetal Bovine Serum	Gibco by ThermoFisher	10270-106
Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 555	ThermoFisher Scientific	A-21422
KO-DMEM	Gibco by ThermoFisher	10829018	Basal medium for undifferentiated hESCs, used in the preparation of culture media
L-Glutamine 200mM (100x)	Gibco by ThermoFisher	25030-081
Methanol, for Molecular Biology	Hi-Media	MB113
Oil red O	HiMedia	CCK013-1KT
Paraformaldehyde	loba chemie	30525-89-4
Penicillin Streptomycin (100x)	Gibco by ThermoFisher	15140- 122
Phalloidin (ActinGreen 488 ReadyProbes reagent)	Invitrogen	R37110
Silver Nitrate	HiMedia	MB156-25G
Sodium Thiosulphate pentahydrate	Chemport	10102-17-7
Sphero Rainbow Fluorescent Particles, 3.0 - 3.4 µm	BD Biosciences	556291
Staining buffer	Prepared in MIRM	----	It was prepared using 2% FBS in PBS
StemPro Adipogenesis Differentiation Basal Media	Gibco by ThermoFisher	A10410-01	Basal media for Adipogenic media
StemPro Adipogenesis Supplement	Gibco by ThermoFisher	A10065-01	Induction media for Adipogenic media
StemPro Chondrogenesis Supplement	Gibco by ThermoFisher	A10064-01	Induction media for Chondrogenic media
StemPro Osteogenesis Supplement	Gibco by ThermoFisher	A10066-01	Induction media for Osteoogenic media
StemPro Osteogenesis/Chondrogenesis Differentiation Basal Media	Gibco by ThermoFisher	A10069-01	Basal media for both Ostegenic and Chondrogenic media
Triton-X-100	Hi-Media	MB031
Trypan Blue	Gibco by life technologies	15250-061
Trypsin - EDTA Solution 1x	Hi-media	TCL049
Tween-20	MERCK	9005-64-5