A practical cryopreservation and staining protocol for immunophenotyping in population studies

Abstract

Large population-based cohort studies, through their prospective collection of a broad range of health information, represent an invaluable resource for novel insights into the pathogenesis of human diseases. Collection and cryopreservation of viable cells from blood samples is becoming increasingly common in large cohorts as these cells are a valuable resource for immunophenotyping and functional studies. We describe the cryopreservation of peripheral blood mononuclear cells (PBMCs), thawing and immunophenotyping protocols used to immunophenotype 9,938 participants in the Health and Retirement Study (HRS). We also outline the extensive quality control involved in a large scale immunophenotyping epidemiological study. We also summarize the existing literature on the effect of cryopreservation on various immune cell subsets including T, B, NK cells, monocytes and dendritic cells.

INTRODUCTION

Large population based cohort studies prospectively collect a broad range of health information along with social, behavioral, and psychological information on several thousand participants over extended time periods. They represent an invaluable resource for novel insights into the pathogenesis of human diseases and it is estimated that approximately 1.5 million volunteers in the United States of America (USA) have participated in various cohorts funded by the National Institutes of Health (https://www.genome.gov/pages/about/od/reportspublications/potentialuscohort.pdf). The majority of these studies have successfully collected and stored a variety of biospecimens using standardized protocols to enable measurement of biomarkers to further our understanding of disease etiology, early detection or prognosis of various diseases. Since appropriate cryopreservation techniques to process and store viable cells are both costly and labor intensive, viable peripheral blood mononuclear cells (PBMCs) are not commonly stored in large population studies. However, PBMCs can be used to study differences in cellular distribution and perform functional assays that help understand biological consequences of genetic variants associated with disease outcomes in large scale genomics studies, several population studies such as the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial (Landgren et al., 2009), Cancer Prevention Study-3 (CPS-3) (Stevens et al., 2007),Coronary Artery Risk Development in Young Adults (CARDIA) (Hughes et al., 1987, Friedman et al., 1988), large biobanks such as the UK Biobank (500,000 participants) (Elliott and Peakman, 2008) and the All Of Us Research Program (https://allofus.nih.gov/) (1,000,000 participants) have cryopreserved cells for use in future immunophenotyping and functional studies. These studies have used a variety of methods to cryopreserve PBMCs. A majority of studies including PLCO, CPS-3, UK Biobank and the All Of Us Research program are cryopreserving whole blood with dimethyl sulfoxide (DMSO), while studies such as CARDIA have opted to isolate PBMCs prior to cryopreservation. Several studies have reported that differences in cryopreservation protocols, time interval between sample collection and processing and procedures for thawing samples all influence the distribution of cells after cryopreservation (Kutscher et al., 2013; Lemieux J, Jobin C, Simard C and Néron S, 2017).

We recently collected and processed blood samples to obtain cryopreserved PBMCs in the Health and Retirement Study (HRS), a nationally representative longitudinal survey of more than 37,000 individuals over age 50 in 23,000 households in all 48 contiguous states in the USA. We performed multiparameter flow cytometry on cryopreserved PBMCs to estimate 33 immune cell subsets, including T cells, B cells, Natural Killer (NK) cells, dendritic cells (DC) and monocytes along with 3 B cell subsets, 19 T cell subsets, 2 NK subsets, 2 DC subsets and 2 monocyte subsets using two flow cytometers, an LSRII and a FORTESSA X20 (BD Biosciences, San Diego, CA). Venous blood samples were collected from 9,938 HRS participants in ~7300 households during 2016–2017. Blood samples were collected directly into a CPT™ vacutainer tube (BD Biosciences, San Diego, CA) at participant homes in all 48 contiguous states in the USA and shipped daily at room temperature to the Advanced Research and Diagnostics Laboratory (ARDL) at the University of Minnesota. More than 95% of the samples were processed within 48 hours of sample collection. PBMCs were isolated from venous using the CPT™ vacutainer and subsequently cryopreserved and stored for future studies. We describe the cryopreservation, thawing and immunophenotyping protocols used in HRS.

BASIC PROTOCOL 1

Cryopreservation of PBMCs using CPT™ vacutainer as a collection tube

Isolation of PBMCs from whole blood was performed prior to cryopreservation using CPT™ tubes that contain a citrate anticoagulant that lies above a polyester barrier in the tube, while a Ficoll Hypaque density gradient lies below it. This tube allowed both sterile blood collection and density-based cell separation in a single container. Centrifugation of the CPT™ tube yielded a layer of PBMCs that was transferred under sterile conditions in preparation for the cryopreservation procedure. The final cell suspension was aliquoted into cryovials that were placed in a −80°C freezer using controlled rate freezing to gradually lower the temperature of the suspension to −80°C in preparation for final storage conditions in liquid nitrogen vapors at −140°C to −180°C.

Materials

Equipment and Supplies

1. Vacutainer CPT™ (Cell Preparation Tube with Sodium Citrate), Becton Dickinson and Company (362761).

2. Filter unit with nylon membrane, 250 mL, 0.45 micron.

3. Filter flasks receivers, Nalgene (4550250).

4. Tube, conical tip, 50 mL, orange cap, sterile.

5. Centrifuge equipped with buckets suitable for CPT™ vacutainer and 50 mL conical tubes.

6. Pipettes, sterile, disposable, 5 mL, 10 mL and 25 mL.

7. Hemocytometer, Bright Line, Improved Neubauer, 0.1 mm deep.

8. American Optical Zeiss microscope, Bi-pin (low pressure), 10× ocular.

9. Cryovials, sterile, screw cap, 2 mL.

10. Mr. Frosty freezing container (Nalgene).

11. HBSS-PS (See Reagents and Solutions)

12. Hemolytic Buffer (See Reagents and Solutions)

13. Supplemented RPMI (HEPES) buffer (500 mL) (See Reagents and Solutions)

14. 20% FCS (See Reagents and Solutions)

15. 20% DMSO (See Reagents and Solutions)

16. 1X Roswell Park Memorial Institute (RPMI) with 25mM HEPES buffer, with L-glutamine, 500 mL.

Protocol steps—Step annotations

Peripheral blood mononuclear cells purification:

Work in the biological safety cabinet. Use sterile technique throughout the following procedure:

Part A: PBMCs isolation

• 1
  Keep CPT™ tubes at room temperature and in upright position until they are ready to process for 30 minutes after unpacking.

  • During shipment, it is not always possible to control the tube temperature (high ambient temperature in summer or low ambient temperature in winter). It is crucial for the proper functionality of the CPT Ficoll layer that the processing be done at room temperature between 20°C and 25°C. Incubating the tubes at room temperature for 30 minutes allow the tube temperature to equilibrate prior to processing.
• 2
  Mix CPT™ tubes by inverting gently 8 to 10 times prior to centrifugation. Spin for 30 minutes at 1700 g at room temperature.

  • Centrifugation time is increased to 30 minutes compared to the 20 minutes recommended by the manufacturer protocol in order to reduce contamination of the PBMC preparation by red blood cells.
• 3Following centrifugation, check to make certain that proper separation of tube components by density gradient has occurred. From top to bottom separation should show a plasma layer, a PBMC layer, gel layer, Ficoll layer, and then red blood cells (figure 1).
• 4Using a sterile 10 mL pipet, aspirate the plasma on the gel by tilting the tube and having the pipette tip touch the side of the tube, not the gel barrier. Transfer all the plasma from the CPT™ tube into one labeled 50 mL conical tube.
• 5Using a new sterile 5 mL pipette, add 3 mL HBSS-PS to each CPT™ tube. Using the same pipette, dispense the HBSS-PS down to rinse the side of the CPT™ tube, followed by gently pipetting the volume up and down 2 times to insure capture of all cells. Using the same pipette (once again avoiding pipetting gel with the liquid) transfer all HBSS-PS/sample mixture to the 50 mL conical tube holding the original plasma.
• 6Using a new sterile 25 mL pipette, add additional HBSS-PS to bring the volume in the conical tube to 30–35 mL.
• 7
  Centrifuge the conical tubes for 10 minutes at 330 g at room temperature. After centrifugation, visually check for an intact pellet. Pour off the supernatant. Disperse cell pellet by gently tapping tube or gently ’racking’ (racking involves drawing each conical tube at a 45 degree angle across the top of a plastic tube rack).

  • Do not vortex samples at this step in order to minimize damage to cells.
• 8
  Using a sterile 5mL pipette, add 3 mL hemolytic buffer and mix gently. Incubate for 5 minutes at room temperature, then add 3mL HBSS-PS using another 5mL sterile pipette and invert or mix gently.

  • When the CPT™ tubes are used within the recommended time guidelines from the manufacturer, there is very little red blood cells contaminant within the PBMC layer however due to the time delay of 24 to 48 hours, contamination of the PBMC layer by RBC is more likely so a RBC lysis step was included.
• 1010. Centrifuge for 10 minutes at 330 g at room temperature. Visually check for an intact pellet to make certain that proper centrifugation has occurred. Pour off the supernatant. Disperse cell pellet by gently tapping tube or gently ’racking’ (do not vortex).
• 1111. Add 10mL of HBSS-PS to the cell pellet.

Figure 1. Separation of blood component using a CPT™ vacutainer.

The diagram shows the separation of the blood component after centrifugation of the CPT™ tube. The plasma (white) lays on top of the PBMCs layer (light grey). Below the PBMC is the polyester gel (black). At the bottom of the tube can be found the red blood cells (RBC, dark grey) within the density gradient layer.

Part B: PBMCs cryopreservation

• 12Pipette 10 μL of the resuspended PBMCs from step 11 into a hemacytometer.
• 13
  Count the cells located in the four outer squares of the hemacytometer and calculate cell count using Equation 1:

  $$\text{Total cell number}=\text{cell count}\times 16\times \text{resuspension volume}\times 2500$$ (Equation 1)
• 14
  Calculate the total volume of 20% DMSO needed for a final cell concentration of 4 million cells/mL using Equation 2. Use this volume to resuspend the PBMCs in step 16 and 18.

  $$\text{Volume of media}(\text{mL})=\text{total cell number}/4\times {10}^6$$ (Equation 2)
• 15Centrifuge the cell suspension for 10 minutes at 330 g at room temperature.
• 16Pour off the supernatant. Disperse cell pellet gently. Add the appropriate amount (half of the volume calculated in step 14) of 20% FCS in supplemented RPMI with HEPES.
• 17For the freezing procedure, prepare a melting ice bath in a shallow tray. Place the cryovial holder containing a labeled cryovial in this ice bath. Prepare a second ice bath. Into this bath, place the 50 mL conical tube of lymphocyte suspension. Pipet the estimated amount of 20 % DMSO in RPMI supplemented with HEPES (half of the volume calculated in step 14) into a labeled 50mL conical tube and place this in the ice bath with the specimen.
• 18Dilute the cell suspension with an equal volume of cold 20% DMSO (1:1 ratio of RPMI to DMSO). Using a sterile pipette, very slowly add the appropriate volume, calculated in step 14 dropwise with continuous mixing of the suspension tube to avoid concentration gradients of DMSO.
• 19Using the same pipette, transfer 1 mL of the suspension into the appropriate pre-chilled labeled cryovial. Screw caps tightly. Keep cryovials on ice water until ready to freeze. Do not allow the cells to sit in DMSO for more than 30 minutes prior to freezing.
• 20Transfer vials into a cardboard box, keeping in upright position. Place box in a styrofoam container (rate freezer or a Mr frosty container may be used if available in place of styrofoam). Place box in a −80°C freezer and transfer cryovials to liquid nitrogen after a minimum of 8 hours or maximum of 24 hours at −80°C for long term storage.

  We typically obtain 2–3 cryovials (each containing 4–6 million PBMCs) from a single CPT™ processed using the conditions described in this protocol.

BASIC PROTOCOL 2

Thawing cryopreserved PBMCs for immunophenotyping

The protocol below indicates how cryopreserved PBMCs were thawed and rested in preparation for immunophenotyping.

Materials

Equipment and Supplies

1. 15ml conical polypropylene tube.

2. Transfer pipettes.

3. Centrifuge with buckets suitable for 5ml round bottom tube and 15ml conical tubes.

4. CO₂ water jacketed incubator (37C, 5% CO₂, 95% humidity).

5. 5ml round bottom 12×75mm polystyrene tube, Corning (352058).

6. 1X PBS (2L) (See Reagents and Solutions)

7. Complete RPMI (See Reagents and Solutions)

8. DNase (100U/μL), Life Technology (18047019). Store until manufacturer expiration date at −20°C.

Protocol steps—Step annotations

1. Place a cryopreserved PBMC vial in a 37°C water bath. Submerge the cryovial halfway and thaw for 60 seconds.

   • PBMCs were frozen at 4 million cells in 1ml of freezing medium. We found that 60 seconds was the optimal time to thaw the frozen PBMCs. Larger volumes may require a longer duration. The thawing step can be done in batches of 10 samples. We have found that thawing larger batches of cryovials increases failure rate during immunophenotyping.

2. Add 1 mL of warmed (37°C) complete RPMI slowly dropwise to the thawed PBMCs against the side of the cryovial using a transfer pipette.

3. Pour the thawed PBMCs in 5 mL of complete RPMI in a 15ml conical tube that has been warmed to 37°C (no pipetting). Make sure to wipe down the tube so that no water from the water bath will be transferred along with the cells.

   • Excessive pipetting during the thawing process is disruptive to some immune cell subsets. At this step, pipetting can be avoided by pouring the cells rather than pipetting them.

4. Rinse the cryovials once with 2 mL of warm complete RPMI, pour the PBMCs into the conical tube with the cell mixture (no pipetting).

5. Incubate the PBMCs for 5 minutes in the 37°C water bath.

   • Since we were thawing multiple samples at once for the study this allowed all the samples to rest shortly prior to the spin to remove the DMSO.

6. Centrifuge the PBMCs for 10 minutes at 330 g at room temperature.

7. Pour off the supernatant. This will leave a small amount of complete RPMI behind that aids in pellet resuspension in Step 8.

8. Add 1 mL of pre-warmed complete RPMI with 50U/ml of DNase. Mix by flicking or racking the tube (avoid pipetting up and down).

9. Incubate the PBMCs for one hour at 37°C in a CO₂ water jacketed incubator (37°C, 5% CO₂, 95% humidity). Leave the cap of the conical tube slightly loose so that gas transfer can occur.

   • There is no clear consensus on the amount of time the cells should be rested after thawing and prior to immunophenotyping. We found that resting the PBMCs at 37°C for one hour was necessary to provide immunophenotyping results in cryopreserved PBMCs comparable to whole blood immunophenotyping in monocytes. More details on resting is provided in the commentary section.

BASIC PROTOCOL 3

Quality control and immunophenotyping of PBMCs

Since we are using two flow cytometers and the immunophenotyping was performed over 18 months, we instituted several quality control (QC) steps that will be described in part A to monitor the flow cytometer, the antibodies and reproducibility across technicians. Part B will describe the steps of the staining protocol.

Materials

Equipment and Supplies

1. LSRII, BD Biosciences, with a 488nm blue laser (4 colors), a 633nm red laser (2 colors) a 407nm violet laser (4 colors) and a 355nm UV laser (3 colors).

2. FORTESSA X20, BD Biosciences, with a 488nm blue laser (4 colors), a 633nm red laser (2 colors) a 407nm violet laser (4 colors), a 355nm UV laser (3 colors) and a nm yellow laser (2 colors).

3. Centrifuge with buckets suitable for 5ml round bottom tube and 15ml conical tubes.

4. 15ml conical polypropylene tube.

5. 5ml round bottom 12×75mm polystyrene tube, Corning (352058).

6. Cytometer and Setup Beads, BD Biosciences (655050). Store until manufacturer expiration date at 4°C away from light.

7. 1X PBS (See Reagents and Solutions)

8. Ultra Rainbow Beads (URB), Spherotech (URFP-30-2). Store until manufacturer expiration date at 4°C away from light.

9. Viability dye (See Reagents and Solutions)

Part A: Flow cytometry QC for BD instruments

1. Turn on flow cytometer and the lasers if necessary.

2. Refill the sheath tank, empty the waste tank and prime twice.

3. Set the flow speed to low and the fine adjust knob to the middle position (5 turns). Run water for 5 minutes.

4. After 20 minutes of warm up, add one drop of Cytometer Setup and Tracking beads (CST) to 350 μL of 1X PBS in a 5 mL round bottom polystyrene tube.

5. Open the Cytometer setup page in FACSDiva and run the beads. Make sure to use the appropriate lot number and follow the prompt from the software.

6. If the CST passes, start the next QC step. If the CST does not pass, call for service.

   • If the CST reports are monitored closely, it is possible to detect changes in lasers alignment before CST fails and to call for service. Slight increases either in bright beads CVs or in ΔPMTVs or changes in laser delay are indicative of potential shifts in lasers alignment. It is preferable to catch this drift early so a service visit can be scheduled at a convenient time rather than as an emergency service call.

7. Add one drop of Ultra Rainbow beads (URB) to 350 μL of 1X PBS.

8. Setup the voltages for each detector so that the beads fall within a gate of your choosing for each detector. Use these voltages as a basis to save the application settings in FACSDiva which will allow the software to adjust the voltages for each detector for minimal changes in PMTV.

9. Use the application settings for each subsequent URB that will be run on the flow cytometer. Establish a range for the Median fluorescence intensity (MFI) for each detector. This range will be used to check subsequent URB.

10. Run the URB on low with the fine adjust knob set at the middle (5 turns).

11. Record 30,000 events.

12. For the URB to pass QC, the MFI for all detectors of the blue and red lasers must be within 3% of the average calculated at the beginning of the study, and within 5% of the average for each detectors for the violet and UV lasers (Kalina et al., 2012).

13. Thaw a compensation PBMCs vial which will be processed as a sample (see part B).

    • We have set aside vials of cryopreserved PBMCs from a healthy volunteer to be used daily as a compensation sample. Daily compensation does improve data quality and will avoid adjustment of compensation post data acquisition.

14. After the one-hour rest, resuspend the compensation sample into 2 mL of 1X PBS.

15. Add 100 μL of cells to 20 pre-labeled 5 mL round bottom 12×75mm polystyrene tubes one per each individual antibody listed in tables 1 and 2 and an unstained sample.

    • Antibody panels: information about antibodies and viability dye (manufacturer and product number) are listed in tables 1 and 2. Store until manufacturer expiration date at 4°C away from light.

Table 1.

Antibody list and volumes per sample for panel 1 that characterize T and B cells. BD stand for BD Biosciences.

Panel 1 (T cells and B cells)
Marker	clone	fluorochrome	provider (cat #)	Volume per sample (μl)
brilliant stain buffer	NA	NA	BD (659611)	55
viability dye	NA	FVS 570 (PE)	BD (564995)	12
CD3	UCHT1	APC	BD (555335)	4.4
HLA-DR	G46-6	PE-CF594	BD (562331)	1.1
CD19	SJ25C1	PE-Cy7	BD (557835)	1.1
CD27	O323	FITC	Biolegend (302806)	2.75
CD8	RPA-T8	BUV395	BD (563796)	1.1
IgD	IA6-2	BUV737	BD (564687)	1.1
CCR7	G043H7	BV421	Biolegend (353208)	2.75
CD28	CD28.2	BV510	Biolegend (302936)	2.75
CD95	DX2	BV605	Biolegend (305628)	2.75
CD45RA	HI100	BV711	Biolegend (304138)	2.75
CD4	RPA-T4	APC-Cy7	BD (557871)	1.1

Table 2.

Antibody list and volumes per sample for panel 2 that characterize dendritic cells, NK and monocytes. BD stand for BD Biosciences.

Panel 2 (monocytes, DC, NK
Marker	clone	fluorochrome	provider (cat #)	Volume per sample (ul)
brilliant stain buffer	NA	NA	BD (659611)	55
viability dye	NA	FVS 570 (PE)	BD (564995)	12
CD3	UCHT1	APC	BD (555335)	4.4
HLA-DR	G46-6	PE-CF594	BD (562331)	1.1
CD19	SJ25C1	PE-Cy7	BD (557835)	1.1
CD11c	B-ly6	BB515	BD (564490)	1.1
CD20	2H7	BUV395	BD (563781)	1.1
CD16	3G8	BUV737	BD (564433)	1.1
CD56	NCAM16.2	BV421	BD (562751)	1.1
CD14	MOP9	BV510	BD (563079)	1.1
CD123	9F5	BV711	BD (563161)	1.1
CD45	2D1	APC-Cy7	BD (560178)	0.55

Development of control materials for training and experimental reproducibility.

In addition to the standard QC procedures to ensure reliability of flow cytometers, ten CPT™ tubes were drawn from 4 healthy volunteers before the start of the study. PBMCs were isolated and aliquots were frozen at a concentration of 2 million cells/cryovial following the procedure described in basic protocol 1. We established a control range (mean±2 standard deviations) before starting the study with 4 technicians processing and analyzing each of the 4 controls 3 times each (12 replicates/control). These controls were subsequently used to monitor our antibody stocks, our individual technique and to train new flow technicians.

Prior to running study samples, new team members processed and analyzed each control at least 3 times and they were certified to analyze study samples only after all values were within the pre-established range. As experimental reproducibility controls, each experimenter processed one control per week along the HRS samples and rotated through the 4 controls in one month. The controls were immediately analyzed to ensure that all cell subsets were within the established ranges. If two cell populations in a particular control were outside established ranges we still analyzed study samples on that day but the same technician would repeat the control on the following day. If the cell populations continued to be out of range, then no study samples were analyzed and data from study samples analyzed the previous day were discarded. If three populations were out of range, we analyzed study samples only after controls were repeated and they passed established QC parameters on the next day. If controls did not pass on two consecutive days, the flow cytometer would be serviced prior to study samples being analyzed.

Part B: Staining protocol for immunophenotyping

1. Prepare the antibody cocktail for panels 1 and 2 as described in tables 1 and 2 respectively.

   • The antibody cocktail (antibody clone choices and fluorochrome choices) was optimized using a combination of the classic fluorescence minus one (FMO) experiments and a series of titration experiments. The cocktail can be prepared once a day and stored at 4°C in the dark. If this cocktail is going to be stored for longer than 4 hours, further validation is required.

     1. A pre-mix with antibodies and brilliant stain buffer (required when two or more brilliant fluorochrome conjugated antibodies are used) is prepared for the day samples, a 10% overage is added to each antibody and to the brilliant stain buffer (See tables 1 and table 2 for specific volumes).
     2. The viability dye is added to the tubes of cells after the panel mix is added. It is used slightly below the manufacturer recommended amount.

2. At the end of the one hour PBMCs rest (step 9, basic protocol 2), add 10 mL of 1X PBS (room temperature) to wash the cells prior to staining.

3. Centrifuge cells for 10 minutes at 330 g at room temperature.

4. Pour off the supernatant and pipet the remaining supernatant completely.

5. Resuspend the cells in an appropriate volume of 1X PBS to achieve a final volume of 200 μL. Mix by flicking the tube.

   • Depending on the level of experience of the technician, the volume of 1X PBS added can vary slightly. The point is to minimize the volume of wasted cells and be as close to the final volume of 200 μL. The volume of 1X PBS required is generally between 160 to 180 μL.

6. Add 100 μL of cells to two pre-labeled 5 mL round bottom polystyrene tubes (panel 1 and panel 2).

7. Add 71.5 μL of panel 1 premix or 62.5 μL of panel 2 premix to the appropriate tubes. Mix the tube gently by agitating the tube rack.

   • It is really crucial to ensure that all the liquid (cells and antibody panel mix) is at the bottom of the tube. Any cells potentially stuck to the side of the tube will remain unlabeled and will be erroneously counted in the negatively stained fraction.

8. Add the 12 μL of viability dye to each tube.

9. Incubate for exactly 20 minutes at room temperature protected from light.

10. Add 800 μL of 1X PBS.

11. Centrifuge cells for 10 minutes at 330 g at room temperature.

12. Pipet off the supernatant. Resuspend stained PBMCs in 500 μL of 1X PBS and store on ice in the dark until flow time. Run samples within 4 hours of staining.

REAGENTS AND SOLUTIONS

1. HBSS-PS: 5 mL PS + 495 mL stock HBSS. Store for 30 days at 4°C.

   1. 1X calcium and magnesium-free Hanks Balanced Salt Solution (HBSS), Gibco (14170-112). Store until manufacturer expiration date at 4°C.
   2. Penicillin-Streptomycin (PS), 10,000 U/mL penicillin-10,000 μg/mL streptomycin, Gibco (15140-122). Store until manufacturer expiration date at −20°C.
2. Hemolytic Buffer: 8.3 g NH₄Cl + 1.0 g NaHCO₃ + 0.04 g disodium EDTA. Dilute to 1000 mL with sterile water. Sterilize by filtration in filter flask (0.45 micron). Store for 6 months at 4°C.

   1. NH₄Cl, 500 g. bottle, dry powder. Store until manufacturer expiration date at room temperature.
   2. NaHCO₃, 500 g. bottle, dry powder. Store until manufacturer expiration date at room temperature.
   3. Disodium EDTA, Fisher (S311-100). Store until manufacturer expiration date at room temperature.
3. Supplemented RPMI (HEPES) buffer (500 mL): 5 mL heparin sodium + 5 mL PS + and 5 mL thawed 100X L-glutamine (mix the latter well prior to addition) + 480 mL 1X RPMI 1640. Store for 7 days at 4°C.

   1. 100X L-Glutamine (200mM) (L-GLUT), Gibco (25030-081). Store until manufacturer expiration date at −20°C.
   2. RPMI Medium 1640 (RPMI with HEPES), Gibco (22400-89). Store until manufacturer expiration date at 4°C.
   3. Penicillin-Streptomycin (PS), 10,000 U/mL penicillin-10,000 μg/mL streptomycin, Gibco (15140-122). Store until manufacturer expiration date at −20°C.
   4. Heparin Sodium injection, USP. Sagent (NDC 25021-400-10). Store until manufacturer expiration date at room temp.
4. 20% FCS (500 mL): To the top of a 500 mL sterile filter flask, add 400 mL supplemented RPMI with HEPES, then 100 mL FCS. Filter. Store for 7 days at 4°C.

   1. Fetal Calf Serum (FCS) also called Fetal Bovine Serum (FBS), Invitrogen (10437-028). Store until manufacturer expiration date at −20°C.
5. 20% DMSO (250 mL): To the top of a filter flask add 200 mL of 20% FCS. While slowly swirling the pipet in the FCS, add 50mL DMSO to the FCS by using a disposable pipette (DMSO must be added slowly or protein will precipitate and the filter will clog). Store for 7 days at 4°C.

   1. Dimethyl Sulfoxide (DMSO), Sigma (D5879-1L). Store until manufacturer expiration date at room temperature.
6. 1X PBS (2L): 2L of 10X PBS + 18L of deionized water. Store until manufacturer expiration date at room temperature.

   1. 10X phosphate buffered saline (PBS) pH 7.4, Lonza (17515Q).
7. Complete RPMI (500 mL): 445 mL 1X RPMI + 50 mL FBS (10%)+ 5 mL PS (100 U/Ml). Store for 7 days at 4°C.

   1. 1X RPMI, Gibco (11875093). Store until manufacturer expiration date at room temperature. 1X RPMI, Gibco (11875093). Store until manufacturer expiration date at room temperature.
   2. Fetal Bovine Serum (FBS), Invitrogen (10437-028). Store until manufacturer expiration date at −20°C.
   3. Penicillin-Streptomycin (PS), 10,000 U/mL penicillin-10,000 μg/mL streptomycin, Gibco (15140-122). Store until manufacturer expiration date at −20°C.
8. Viability dye: 594 μL of 1X PBS + 6 μl viability dye stock. Do not store.

   1. Viability Dye-FVS 450, BD Biosciences (564995), prepared according to manufacturer recommendations. Store for 90 days date at −20°C.

COMMENTARY

Background information

Numerous publications have reported on the effect of cryopreservation on various immune subsets (Lemieux et al., 2017; Draxler DF, Madondo MT, Hanafi G, Plebanski M and Medcalf RL, 2017; Verschoor CP, Kohli V and Balion C, 2017). The two major types of cryopreservation evaluated across studies are cryopreserved whole blood and PBMCs isolated via a density gradient and then cryopreserved. Several studies compared the viability and functionality of cryopreserved whole blood and cryopreserved PBMCs. DNA repair capacity and mutagen sensitivity assays showed similar results in cryopreserved whole blood and cryopreserved PBMCs (Cheng L, Wang LE, Spitz MR and Wei Q, 2001). While it is possible to stain immune cells for surface markers for immunophenotyping, fix them and then freeze them at lower temperature (−20°C or −80°C), this method does not result in viable cells for future functional assays (Pinto et al., 2004; Madelaine Paredes R, Tadaki DK, Sooter A, Gamboni F and Sheppard CDRF, 2017). Studies have also evaluated different methods for thawing samples with some studies thawing samples and performing immunophenotyping without any rest period while other studies provided rest periods (incubating cells at 37°C) of 1 hour to 24 hours prior to immunophenotyping (Lemieux et al., 2017; Wang et al., 2016, Kutscher et al., 2013). In general, these studies have shown that a short rest period of 1 hour generally improves comparability of various immune subsets in cryopreserved cells and fresh cells (Lemieux et al., 2017) though a longer rest period of up to 24 hours results in higher proportion of activated cells in the cryopreserved sample (Sattui et al., 2012). An 18 hour rest period reduced the number of apoptotic cells in thawed PBMCs and positively affected lymphocytes functionality (Wang et al., 2016). Published effects of cryopreservation and thawing procedures on the enumeration and functions of individual immune subsets are reviewed below.

T cells

Since T cells and their subsets are useful in numerous applications such as monitoring HIV patients (Sattui et al., 2012; Zhang et al., 2016), functional studies (Cheng et al., 2001), activation studies (Lemieux et al., 2017), and vaccine testing (Kutscher et al., 2013; Wang et al., 2016), the effect of cryopreservation on T cells has been extensively investigated. Numerous studies showed no effect of cryopreservation in the enumeration of T cells, helper and cytotoxic T cells (Weinberg et al., 2010; Draxler et al., 2017; Vershoor et al., 2017; Sattui et al., 2012). Though cryopreservation did not affect the distribution of helper and cytotoxic T cells, some T cell subsets such as ‘naive helper’ and ‘central memory helper’ and ‘cytotoxic’ were significantly lower and cytotoxic effector T cells were higher in cryopreserved samples compared to fresh whole blood (Costantini et al., 2003). A significant decrease in T cells and helper T cells observed after cryopreservation was not rescued by 1 hour rest (Lemieux et al., 2017). Though there were significant differences in several T cell subsets when immunophenotyping was performed on cryopreserved PBMCs without resting the cells, the proportion of naïve, central memory, effector, effector memory, Th1 and Th2 as well as activated subsets within the helper and cytotoxic T cell populations did not show any significant change with cryopreservation after a rest period (Lemieux et al., 2017). In their study, a one hour rest period also showed greater concordance of helper and cytotoxic T cell subsets with the fresh sample than the sample that was rested for 24 hours or not rested at all (Lemieux et al., 2017). In contrast, a rest period of 18 hours after thawing improved lymphocyte functionality. Smaller T cell subsets such as regulatory helper T cells (Tregs) and naïve helper or naïve cytotoxic T cells measured levels are comparable between cryopreserved PBMCs and fresh blood (Wang et al., 2016). Tregs play a major role in immune homeostasis and downregulate both Th1 and Th2 responses (Zhang et al., 2014) and are sensitive to cryopreservation. Though helper T cell enumeration remained constant between fresh and frozen PBMCs, Treg count was reduced in cryopreserved PBMCs regardless of HIV status (Sattui et al., 2012). However in this study, the PBMCs were not rested after being thawed. Indeed, Lemieux (2017) showed that a rest of 1 hour was sufficient to restore T cells phenotype in thawed PBMCs so it is comparable to fresh samples. Interestingly, they also showed that using a more sensitive flow data analysis method (viSNE) compared to more conventional flow analysis method (FlowJo or FCS express) could improve the detection of poorly represented populations such as Tregs (Lemieux et al., 2017). Resting of thawed PBMCs for 4 hours has also been shown to restore CD120b expression on Tregs regardless of HIV status (Zhang et al., 2016) though number of Tregs did not seem to be affected by either cryopreservation or duration of the rest period after thawing. But other studies showed that Tregs were decreased, but highly correlated, in cryopreserved samples (Elkord, 2009) without any resting of thawed samples.

B cells

B cells can be successfully isolated and immortalized in thawed whole blood that was cryopreserved even after a delay of 24 hours after blood collection (Peakman and Elliot, 2007). B cells are increased in cryopreserved blood as compared to a fresh sample (Verschoor et al., 2017). The increase in B cells in cryopreserved PBMCs observed by Lemieux (2017) remained unchanged by either a 1 or 24 hours rest period (Lemieux et al., 2017).

NK cells

NK cell frequencies are lower in cryopreserved PBMCs but still highly correlated in cryopreserved blood compared to fresh blood (Verschoor et al., 2017). A rest period after thawing does not rescue the NK cells enumeration in cryopreserved PBMCs (Lemieux et al., 2017). In contrast, another study that included no rest period after thawing found that NK cells enumeration was not affected by cryopreservation of PBMCs compared to fresh blood (Draxler et al., 2017). The reasons for the discrepancy between these studies is unclear.

Monocytes

Two studies showed that monocytes count was more elevated in cryopreserved blood or cryopreserved PBMCs compared to fresh blood (Draxler et al., 2017; Verschoor et al., 2017). However, neither one of these studies allowed the cells to rest prior to staining. Proportion of monocytes in cryopreserved PBMCs was similar to fresh blood after both 1 and 24 hours of rest after thawing (Lemieux et al., 2017).

Dendritic cells

Dendritic cells are some of the least represented cells in PBMCs and are sensitive to freeze thaw cycles. Makino and Baba (1997) showed that upon proper cryopreservation condition (25% FBS and 10% DMSO), dendritic cells can be frozen and properly thawed and cultured in the same manner as they would in freshly isolated PBMCs. But while the dendritic cells retain their function after cryopreservation, they appear to increase in numbers (Draxler et al., 2017; Verschoor et al., 2017). Dendritic cell subsets are affected by PBMC manipulation. A study showed that Ficoll separation regardless of cryopreservation status increases the ratio of plasmacytoid dendritic to myeloid dendritic cells (Gerrits et al., 2007). These cells also had a higher expression level of maturation markers.

Critical parameters and troubleshooting

Standardization of cryopreservation and thawing procedures

In order to obtain reproducible results during immunophenotyping, it is important to minimize pre-analytical variability by standardizing cryopreservation and thawing of PBMCs. These include:

1. The time between blood collection and sample processing should remain consistently within 48 hours as viability of cells after cryopreservation is dramatically reduced if the delay between sample collection and processing is more than 48 hours (data not shown).

2. Though CPT™ tubes are shipped in styrofoam containers with room temperature gel packs, it is critical to keep the tubes upright at room temperature for 30 minutes prior to processing so that the gel layer can equilibrate to room temperature.

3. Rate of cooling during sample freezing is an important step. The cooling rate must be maintained at its optimum rate as cell survival will decrease if more or less of the optimal cooling rate is used (Hubel and Skubitz, 2017). We stored the PBMCs cryovials in a styrofoam container at −80°C in order to control the cooling rate. One can also use a freezer specifically designed to cryopreserve cells or a Mr. Frosty (Nalgene) which is a container using the cooling property of isopropanol to maintain an optimum cooling rate of 1°C per minute.

4. Using cold solutions during the cryopreservation process has been shown to improve viability and functional recovery of PBMCs (Tree et al., 2004). When comparing the use of cold freezing media versus room temperature freezing medium, Tree (2004) noted that PBMCs frozen using cold media performed more similarly to fresh blood in a functional assay (T cells response to Tetanus). We used cold 20% DMSO (in 1X RPMI) solution and cold 20% FBS (in 1X RPMI) to process the PBMCs.

5. Storage temperature and more specifically changes in storage temperature can affect cell viability and subsets recovery (Cosentino et al., 2003). Storing the control PBMCs and study samples in the same liquid nitrogen freezer will allow evaluation of any differences in subset recovery or viability due to changes in storage temperature.

6. Fast thawing at 37°C is an important step to standardize across technicians. We instituted a thawing time of 60 seconds to maintain consistency between technicians.

7. Based on previously published data (Lemieux et al., 2016), we rested cells at 37°C for one hour after thawing to restore most immune cell subsets to their level in fresh blood. The cell subsets of interest are the major determinant of whether to include a rest period and decide on the duration of the rest period.

8. Appropriate training of technicians is key to standardization. While CPT™ vacutainer are easier to use than a Ficoll gradient, cryopreserving PBMCs still requires a certain level of technical skill. The technicians processing the samples for cryopreservation were trained by a senior member of the staff on all aspects of the procedure (cell count, cooling of the samples, recording of reagents lot numbers and so on). Trainees were then observed closely by a senior staff member before being allowed to work on study samples. A similar training method was used in the flow laboratory. To monitor proper thawing technique, we used our HRS controls as a test for appropriate and timely technique.

Standardization of immunophenotyping procedures

Variability in immunophenotyping assays can be minimized using the following procedures:

1. Standardize flow cytometers configuration across multiple instruments. For example both flow cytometers used for HRS were equipped with similar lasers (Blue, UV, red and violet) and the same filters for each PMT.

2. Establish a thorough flow cytometer daily QC as described in basic protocol 3. We minimized downtime by carefully monitoring our CST and URB in order to catch laser drift prior to outright QC failure. We also used frozen PBMCs from a healthy volunteer to perform daily compensation setup and calculations.

3. During the experimental setup, perform all the steps required to check the antibodies used in the study. This includes testing all the antibodies in a dilution series and including all the FMO (fluorescence minus one) controls. If possible, purchase antibodies in bulk so that they are from the same lot. However, if that is not possible test every new lot of reagents using control samples and ensure similar values for all subsets before using new lots of reagents on study samples. We used our PBMC controls to check the consistency of our panels over the course of the study.

4. In the context of immunophenotyping, we used our controls to check both our reagents (antibodies and buffers) as well as our technical abilities and consistency during the study. We also trained new technicians during the study and the PBMC controls proved to be an invaluable tool to test the readiness of each new technician. As described in basic protocol 3, each new trainee would immunophenotype each control 3 times on different days. If all measured immune populations were within the established range, the new trainee would start running HRS samples. If immune populations were out of range, additional training was provided.

5. When we started running samples for HRS, we grouped ten samples per batch for processing from thawing to immunophenotyping. Once all the technicians became proficient, we slowly increased the number of samples per batch (up to 16) however we noticed an increase in the number of samples with low cell count (data not shown). This increase in number of samples per batch only affected total thawing time. We suspect that the increased time at 37°C before centrifuging and removing the DMSO affected cell viability.

6. During initial panel testing, we identified lower numbers of dendritic cells if the samples are kept on ice water rather than “regular ice”. Hence we processed our daily samples in five scaled batches of ten samples each to minimize sample time on ice (generally 2 hours maximum per batch), which minimized variability in dendritic cell measurement.

7. The fluidics in the flow cytometer must be cleaned daily and a full clean of the fluidics should be performed monthly.

Time Considerations

Cryopreservation

Cryopreservation of PBMCs is time consuming and requires an experienced technician. CPT™ were chosen in order to save time over a Ficoll separation. It has been shown that viability and T cells subsets enumeration is similar in cryopreserved PBMCs that have been isolated using either a CPT™ vacutainer or the Ficoll separation method (Ruitenberg et al., 2006). The protocol for cryopreservation of PBMCs (basic protocol 1) is straightforward and can be done in roughly three hours for one CPT™ tube. Since the central laboratory received more than one sample a day, the separation and cryopreservation of PBMCs was batched. This only marginally increased processing time to 4 hours per batch. If several trained technicians were assigned to cryopreservation and processing was staggered, upward of 80 samples could be cryopreserved in a day without the need for automation.

Flow cytometry setup

To limit sample loss due to instrument failure, no samples are thawed until both (or at least one) flow cytometers has passed our QC procedure which takes roughly 40 minutes. The lasers need to warm up for a minimum of 20 minutes prior to use. This allows enough time to refill both sheath tank, empty the waste tank, prime the fluidic systems of the flows. After the CST have been ran, setup the URB experiment and its associated application settings. The cryopreserved PBMCs used for compensation setup are also thawed during this time.

PBMCs preparation from thawing to immunophenotyping

The study samples were processed in batches of 10 samples from thawing to staining. During a an 8 hours workday two technicians prepared 5 sets of 10 samples without requiring the need for any automation and without compromising data quality. Immunophenotyping of PBMCs from removal of cryovials from liquid nitrogen to staining took approximately 2 hours with 35 to 40 minutes of hands on work at the bench. By staggering the batches appropriately, we have been able to minimize the time the samples spent on ice and process all 50 samples during an 8 hours workday.

Anticipated results

Using the cryopreservation procedures described in this manuscript, the central laboratory was able to process up to 80 samples daily and all samples were processed on the same day samples were received at the laboratory. We were also able to immunophenotype up to 50 samples daily following the above mentioned protocols. The total failure rate was ≤ 5% when we used the QC procedures outlined in this manuscript.

Summary

Previous studies indicate that cryopreserving whole blood or isolated PBMCs both yield similar results for immunophenotyping and functional studies. The effect of cryopreservation on various cell subsets has been described previously and use of a rest period after thawing can help restore immune cell distributions to those seen in fresh blood. Standardization of cryopreservation, thawing and immunophenotyping procedures and use of control samples, stored and analyzed in a manner identical to study samples, can significantly reduce variability of immunophenotyping measurements in large scale population studies.

Significance Statement.

Lack of standardization in cryopreservation and thawing techniques can increase variability in immunophenotyping. Adherence to standardized procedures along with continuous monitoring of quality control procedures is essential for long term stability of immunophenotyping assays in population studies.