Evaluation and Characterization of Human Bone Marrow Mesenchymal Stromal Cells Cryopreserved in Animal Component-Free, Chemically Defined, Serum-Free Conditions

Abstract

Mesenchymal stromal cells (MSCs) have the potential to treat various disease indications and are the future of cell therapy-based regenerative medicine. Typically, MSCs cryopreserved in serum-containing freezing formulation are supplied at the clinical site, which necessities that this formulation is removed before the administration. This is a cumbersome process, and there is an immediate need for identifying serum-free, xeno-free cryopreservation medium that can be readily used. Here, we analysed two commercially available serum-free, xeno-free, defined freezing media viz., CryoStor 5 (CS5) and CryoStor 10 (CS10) on their effect on human bone marrow MSCs at different freezing cell densities (5, 10, 12.5, 15 and 25 million cells per ml) over a period of 6 months and compared them to the in-house PlasmaLyte A (PLA)-based cryopreservation media. We found that the MSCs cryopreserved in CS5 and CS10 showed similar characteristics as compared with the in-house freezing media for the various parameters analysed including post-thaw recovery, viability, phenotypic marker expression, CFU-F ability and trilineage differentiation potential of the MSCs. Our results show that human MSC could be successfully cryopreserved using serum-free and xeno-free cryopreservation media and can be delivered to the bedside without any manipulations.

1. Introduction

The potential use of Bone marrow mesenchymal stromal cells (BM-MSCs) to treating various disease indications is in steady increase [1] and demands huge availability of clinical quality BM-MSCs to meet the growing demands. Typically, MSCs frozen in cryopreservation solution were supplied to the clinical site, and the method of freezing along with the composition of freezing media itself plays an important role in determining the characteristics of MSCs before infusion [2, 3]. One essential requirement of cGMP grade formulation reagents in cryopreservation is that they are free from animal serum proteins and toxic chemicals, as the xenogeneic compounds possess the risk of transmission of animal viral, prion and zoonose contamination [4].

Presently, the common practice of cryopreserving MSCs is in PlasmaLyte A supplemented with albumin and a cryoprotectant of DMSO and/or dextran [5, 6]. Conventionally, 10% DMSO is used because they readily penetrate cell membranes and thus confer protection to the intracellular components [7, 8]. But it is proven that freezing media containing a higher concentration of DMSO is toxic to patients [8, 9]. Efforts to reduce toxicity include the removal of DMSO prior to transfusion or decreasing the amounts used in the freezing process. But these post-thaw manipulations consume lots of time, decrease the viability of the cellular product and further require aseptic zone for processing and concentration of cells at site [8]. Several groups have investigated MSCs cryopreservation using low concentrations of DMSO and or many alterative to DMSO like polyvinylpyrrolidone (PVP), methylcellulose, polyethene glycol (PEG) trehalose and polymer mimics without or with DMSO from 2.5 to 7.5% [10, 11, 12, 13]. But it is necessary that these in-house cryopreservation formulations are prepared from USP grade or cGMP grade reagents to meet the safety requirements and may pose many regulatory challenges while filing an NDA and obtaining regulatory clearances. One of the possible alternatives is to screen the commercially available GMP compliance USP grade, serum-free cryopreservation media for MSCs. Several groups have evaluated the commercial MSC cryopreservation for a short period of time from 1 week to 1 month, but it is necessary to study the real-time stability testing for at least 6 months. In this study, we evaluated the commercially available cryopreservation solutions − CryoStor (CS) 5 and CS10 from BioLife Solutions, USA for their ability to cryopreserve BMMSCs and compared them to our in-house developed PlasmaLyte A-based cryopreservation formulation. We analysed post-thaw viability, total cell recovery, trilineage differentiation, phenotypic marker expression, CFU-F potential and apoptotic cell percentage at a testing interval of 1 week, 1, 2, 3 and 6 months time point and discussed the data here.

2. Materials and methods

2.1 Isolation and culture of human bone marrow-derived MSCs

The BM was collected from healthy donors of age group between 19 and 40 years, after obtaining approval from the Institutional Ethics Committee of the Kasturba Hospital, Manipal and signed informed consent. The mononuclear cells (MNC) were isolated following the density gradient centrifugation method using Lymphoprep (Axis-Shield PoC) as described previously [9]. The isolated MNCs were seeded at a density of 50 million cells per T-75 flask in Dulbecco’s modified Eagles medium-Knock Out (DMEM-KO) supplemented with 10% FBS (Hyclone, Waltham, MA), 2 mM L-glutaMAX (Invitrogen, Carlsbad, CA) and 1X penstrep (Invitrogen, Carlsbad, CA). The cells were passaged when they reached 80−90% confluency using trypsin (0.25%)/ Ethylenediamine tetra acetic acid (EDTA; 1 mM) (Gibco, USA). MSCs from three donors were pooled at passage 2 (P2) in equal proportion and cultured at a seeding density of 1000 cells per cm² in bFGF (2 ng/ml) enriched KO-FBS complete medium till P5.

2.2 Cryopreservation

Cryopreservation of BM-MSCs involves freezing of cells in 1 ml of formulation medium containing 10% (v/v) Dimethyl sulfoxide (DMSO) as a cryoprotectant with 5% (v/v) human serum albumin and quantified (QS) with PlasmaLyte A (Baxter Inc) in 5 ml Daikyo Crystal Zenith (CZ) vials from West Pharma. The commercial formulation of Cryostor CS10 and CS5 (BioLife Solutions Inc) were used for cryopreservation of BM-MSCs in CZ vials for comparison. MSCs after trypsinisation were centrifuged to remove the trypsin and the pellet was dislodged by gentle tapping to the pellet, 2−8°C refrigerated cryopreservation media was slowly added. The cells were resuspended at five different concentrations (5, 10, 12.5, 15 and 25 million cells per ml) in these three formulations. These cells were aliquoted @ 1 ml per CZ vials and cryopreserved by slow freezing @ 1°C/min till 80°C by keeping them in Cryomed controlled rate freezer (CRF). During this time an aliquot of the cells was taken to measure the pre-freeze 7AAD/viability. The frozen vials were then shifted to canister racks and were stored in the vapour phase of liquid nitrogen (LN₂).

2.3 Thawing

Each sample was thawed according to the internal Stempeutics SOP. Briefly, after a minimum of 1-week storage in liquid nitrogen, the vials were retrieved and thawed immediately by placing them in a water batch at 37°C till the ice crystals were just disappearing. The thawed cells were transferred into 15 ml tube and the sample was diluted at 1:9 with complete media. The cells were centrifuged at 1200 rpm for 10 minutes and the pellet was then resuspended in KO-FBS complete media.

2.4 Testing frequency, cell count and manual viability assessments

The samples from three different cryopreservation mediums at five different freezing concentrations were analysed at time points of 1 week, 1 month, 2 months, 3 months and 6 months. The total cell recovery (TCR) and viability analysis by manual method and by flow cytometry method were carried out immediately after the thawing procedure. The total cell count and manual viability were determined by the trypan blue (Fluka) exclusion method. The number of viable (non-stained) and non-viable (stained) cells were enumerated microscopically. TCR is calculated by adding the total number of stained and unstained cells. The viability percentage was calculated by dividing the total number of viable cells by TCR and multiplied by 100. For estimation of viability by flow cytometry, the cells were stained with 7AAD and were analysed following the protocol as described previously [14].

2.5 Immunophenotyping

We analysed a set of two positive (CD90-PE, CD73-PE) and two negative cell surface markers (CD14-FITC and CD19-FITC) by flow cytometry. All the antibodies used for these studies were purchased from BD Pharmingen, San Diego. The cryopreserved cells were thawed and resuspended in wash buffer containing phosphate buffer saline (PBS) supplemented with 1% (v/v) FBS and 1% (w/v) sodium azide for analysis. The cells were incubated with saturating concentrations of fluorescein iso-thiocyanate (FITC) or phycoerythrin-(PE) conjugated antibodies at 4°C for 30 minutes in dark. After that, the cells were washed with wash buffer three times and re-suspended in 0.5 ml of wash buffer. The labelled cells were analysed in EasyCyte (Guava Technology) flow cytometer after setting the instrument parameters with respective isotype-matched controls. For every sample, 10,000 events were captured and the data was analysed using Guava Express Pro software (Guava Technologies). Fluorescence intensity of 25% or above its isotype control is considered an antigenic event and was used for calculation.

2.6 Differentiation potential

The differentiation potential of frozen MSCs was analysed by their ability to differentiate into osteogenic, adipogenic and chondrogenic lineages. Osteogenic differentiation was induced by culturing P5 BM-MSCs in the KO-FBS supplemented with 10⁻⁸ M dexamethasone, 30 μg/ml ascorbic acid and 10 mM β-glycerophosphate (all Sigma-Aldrich). For adipogenic differentiation, cells were cultured in the KO-FBS supplemented with 1 μM dexamethasone, 0.5 mM isobutylmethylxanthine (IBMX), 1 μg/ml insulin and 100 μM indomethacin (all Sigma-Aldrich). The chondrogenic differentiation was induced using STEMPRO (Invitrogen) chondrogenesis differentiation medium. After 21 days of differentiation, the cells were fixed and stained with Von Kossa, Oil Red O and Safranine O, respectively, for osteo, adipo and chondro differentiation cultures. The images were captured using Nikon Eclipse 90i microscope (Nikon Corporation, Japan, www.nikon.com) and Image-Pro Express software (Media Cybernetics, Inc., Silver Spring, MD, www.mediacy.com).

2.7 CFU-F assay

For CFU-F assay, 100 MSCs from 5 different freezing concentrations of three cryopreservation media at six-month time points were plated in KO-FBS (n = 2 of each condition) on a 100 mm² cell culture dish. After 14 days in culture, the plates were stained with crystal violet and the number of colonies was counted.

2.8 Apoptosis analysis

Apoptosis analysis was carried out using Tali apoptosis kit following the manufacturer’s instructions (Tali Apoptosis kit – Annexin V Alexa Fluor 488 and propidium iodide, Cat # A10788. Life technologies). Briefly, 1 million cells/ml were resuspended in annexin binding buffer (ABB) and 5 ul of Annexin V were added per 100 μl of sample. The samples were incubated for 20 minutes in dark, centrifuged and resuspended in 100 μl of ABB. After adding 1 μl of PI and incubation for 5 min, the samples were read in Tali® Image-Based Cytometer.

2.9 Statistical analysis

All values are expressed as mean ± SEM (standard error of mean). Data were analysed by using Graphpad Prism (version 5, Graphpad Software Inc., La Jolla, CA, USA). Two-way ANOVA (Analysis of variances) was performed in order to compare means between groups and Bonferroni post-tests were carried out to find out the significance of the variables tested. P value <0.05 was considered significant.

3. Results

3.1 Post-thaw viability – 7 Amino-actinomycin D (7AAD)

BM-MSCs were scaled-up to P5 in multiple numbers of ten cell stacks and the cultures were frozen in various cell concentrations ranging from 5 x 10⁶ to 25 x 10⁶ per ml in 5 ml CZ vials using 3 different cryopreservation media (PLA, CS10 and CS5). Pre-freeze viability of BM-MSCs in PLA, CS10 and CS5 formulation media was >98.6 ± 1.2% (mean ± standard deviation). Upon, thawing, BM-MSCs viability of PLA, CS10 and CS5 of 5 different cell concentrations (5, 10, 12.5, 15 and 25 million cells per ml) were measured by flow cytometry using 7AAD at different time points viz., 1 week, 1, 2, 3 and 6 months. All the samples have shown >92% (Figure 1) viability by 7AAD in all the time points. No significant difference in percentage of viable cells was observed after six-month storage when compared to one-week storage. Two-way analysis of variance (ANOVA) was performed to assay differences over different time points and among different cryopreservation media. The results indicate that CS5 is equally good to that of CS10 and PLA and there are no significant differences in comparison with either time points or different cryopreservation media.

Figure 1.

Post-thaw viability by 7AAD of BM-MSCs following cryopreservation with PlasmaLyte a based cryopreservation solution (control), CryoStor (CS) -10, and CS5 variants. Cell viability assessed at 5 different time points (1st week, 1st month, 2nd month, 3rd month and 6th month) at 5 different freezing concentration (5, 10, 12.5, 15 and 25 million cells per ml) shows no significant differences among three cryopreservation media.

3.2 Total cell recovery and viability

The Post-thaw cell recovery and viability of five different concentrations at 5 different time points (1st (first) week, 1st month, 2nd month, 3rd month and 6th month) in 3 different cryopreservation media were assayed using trypan blue dye exclusion assay. The total viable and non-viable cell count was taken as total cell recovery and shown in Figure 2A. The results demonstrated that no cell loss was observed upon thawing the samples at all time points in 3 cryopreservation media. There are Figure 2B also no significant differences in total cell recovery percentage in the lowest and highest freezing densities (5 and 25 million cells/ml respectively) at 3rd and 6th month of recovery after freezing (). Simultaneously viability by dye exclusion method (DEM) was analysed in all of the samples and the results were shown in Figure 3. Viability by DEM was between 85 to 95% in all test samples (Figure 2C).

Figure 2.

Post-thaw Total cell recovery (TCR) of BM-MSCs cryopreserved in CS10, CS5 compared to control. (A) TCR in millions as assessed by summing total viable and non-viable cell counts after trypan blue staining. (B) TCR in percentage between the lowest and highest freezing densities at 3^rd and 6^th month. (C) Viability percentage as calculated by trypan blue exclusion method. There is no significant differences among the three cryopreservation media.

Figure 3.

Immuno-phenotype of BM-MSCs following cryopreservation with PlasmaLyte A-based cryopreservation solution (control) (A), CryoStor (CS) 10 (B), and CS5 (C) at six month time point. Two positive markers of CD90 and CD73 and two negative markers of CD19 and CD14 were analysed. No significant differences among three cryopreservation media at 5 different freezing concentrations except for CD 73 expression variation among CS10 group.

3.3 Immunophenotyping

The surface marker expression was evaluated and analysed by using flow cytometry for BM-MSCs cryopreserved in control, CS10 and CS5 of 5 different cell concentrations. The freeze–thaw BM-MSCs in all three cryopreservation media across all time points were positive for CD90 and CD73 and negative for CD19 and CD14. BM-MSCs showed similar expression of CD markers in CS5 compared with that of CS10 and control (Figure 3). There are no significant differences in positive and negative marker expression observed among different seeding densities in any of those three groups except for CD 73 expression variation in CS10 group. There is a significant difference (n = 5,*p < 0.05) between 12.5 and 25 million per ml and this significance is even stronger (n = 5, **p < 0.01) between 5 and 12.5 million per ml. Howsoever, these expression levels were more than 95% for positive markers and < 2.5% for negative markers in all three cryopreservation media at different cell densities.

3.4 Differentiation potential

We investigated the in-vitro functional tri-lineage differentiation potential of P5 BM-MSCs at the 6th month time point after cryopreservation in PLA-based cryopreservation medium (control), CS10 and CS5 at 5 different cell freezing concentrations. The differentiation towards adipocytes was evident by the formation of fatty vacuole deposits and was observed by Oil Red O staining (Figure 4A). The differentiation towards osteoblasts was observed by Von Kossa staining (Figure 4B) and that of chondrocytes was observed by safranine O staining (Figure 4C) after 21 days of differentiation induction. The cells cryopreserved at different cell densities in different cryopreservation media stained for all three lineages, showing that the trilineage differentiation ability is maintained in all these conditions. The results of CS5 and CS10 were comparable with that of the control cryopreservation medium for all 5 different cell freezing concentrations.

Figure 4.

Trilineage differentiation of BM-MSC in 3 different cryopreservation medium at 6th month time point. BM-MSC cryopreserved in PlasmaLyte a based cryopreservation solution (control), CryoStor (CS) 10 and CS5 were differentiated into (A) adipocytes (B) osteocytes and (C) chondrocytes. Adipocytes stained with oil red O, osteocytes with Von kossa staining and chondrocytes by safronin O stain. All pictures captured with magnification of 10X.

3.5 CFU-F assay

Clonogenic potential of cryopreserved MSCs at 5 different cell freezing concentrations (5, 10, 12.5, 15 and 25 million cells per ml) in each of the 3 different cryopreservation media (control, CS10 and CS5) at 6-month time points were assessed by counting the number of colonies with more than 50 cells. There is no significant change in the number of colonies formed by cells cryopreserved at different concentrations in all the three different cryopreservation media, except for two different cell concentrations in CS5 storage (Figure 5). The number of colonies formed by 5 and 10 million per ml freezing concentration (n = 2, P < 0.001) was significantly lower compared to other concentrations in CS5.

Figure 5.

CFU-F assay of BM-MSC in 3 different cryopreservation medium at 6th month time point. The number of colonies formed in 5 different freezing concentration at 6th month time point show no statistically difference between control and CS10. There is a significant difference (***P < 0.001) observed in CS5 at lower freezing density of 5 and 10 million cells per mL compared to other 3 freezing concentration (12.5, 15 and 25 million cells per mL).Abbreviation:CFU — colony forming unit fibroblast.

3.6 Apoptosis analysis

The comparative post-thaw apoptosis assay was performed at 6th month time point of BM-MSCs cryopreserved in control, CS10 and CS5 of 5 different freezing concentrations. The percentage of apoptotic cells increased with the decrease in freezing concentration, with the lowest concentration of 5 million/ml having the highest value and vice versa in all the three freezing media (Figure 6). The differences are significant (n = 2,*P < 0.05, **P < 0.01, ***P < 0.001) except between 25 and 15 million cells per ml of CS10 and CS5. Overall, CS10 and CS5 have a lesser percentage of apoptotic cells compared to control in all concentrations tested.

Figure 6.

Apoptosis assay. Percentage of apoptotic cells in 3 cryopreservation media (control, CS5, CS10) in 5 different freezing concentration at 6th month time point reveals significant increase in percentage of apoptosis from 5 million cells per ml to 25 million cells per ml in all three cryopreservation media. (*P < 0.05,**p < 0.01***P < 0.001).

4. Discussion

The demand for MSCs in clinical application is growing and necessitates availability of good quality cryopreserved MSCs with minimal pre-transplantation manipulations at the clinical trial site or final usage of the product. Optimal cryopreservation protocols including freezing density were not yet tested or published globally and no consensus has reached for the standard freezing density of MSCs. Moreover, published efforts were guided by HSCs cryopreservation protocols [15, 16]. Cryopreservation technology for MSCs is still evolving, as MSCs seem to lose viability very rapidly and are mostly attributed due to the rapid development of apoptotic processes and cryopreservation-induced delayed-onset cell death (CIDOCD) [17]. Most commonly, MSCS were cryopreserved in formulations containing 10% DMSO that needs to be removed from the final cellular product before infusion for human use. The removal of DMSO has its own challenge and requires the sample to be handled in a GMP compliance area and any improper removal may cause several ailments like headache, nausea, vomiting, sedation, high blood pressure, bradycardia or anaphylactic shock to the patients [18].

The total substitute of DMSO is not advisable, as it affects the viability and decreases the shelf life of MSCs during cryopreservation. Though many in-house cryopreservation formulations have been reported [19, 20], it is important that all their constituents meet the bio-safety standards. The use of approved and commercially available freezing formulations will ease the complications and facilitates hassle-free filing to NDA (New Drug Application), auditing, screening and testing etc.

In the present study, we have critically evaluated 5% and 10% DMSO containing commercial cryopreservation media (CS5 & CS10, respectively) for post-thaw viability, CFU-F, phenotypic markers expression, differentiation potential and percentage apoptotic cells of cryopreserved BMMSCs and compared to in-house formulations (PlamsLyte A with 10% DMSO and 5% HSA) at different time points up to 6 months. PlamsLyte A is an isotonic solution with the physiochemical properties closely resembling human plasma and is widely used as perioperative fluid [21]. PlasmaLyte A is commonly used in cryopreservation of MSCs and in our earlier studies it is known to preserve the viability and efficiency of BM-MSCs for clinical trials [22].

We have demonstrated comparable results for all the characteristics analysed in all these formulations at different time points. Our results showed that the post-thaw viability analysed by two methods viz., trypan blue dye exclusion and 7AAD were > 85% in all conditions tested up to 6 months time period. Few studies reported a reduction in post-thaw viability in time with the use of DMSO-based cryoprotectant [10, 23] but they used different freezing concentrations and DMSO percentages. The viability of >85% of what we obtained should not be a problem as there are reports stating that even with 70% viability, the cells were demonstrated to have enhanced immunosuppression within 6 months of time period [24]. We also demonstrated that the total cell recovery of BM-MSCs is >85% in all these conditions and it probably seems that higher cell freezing densities yields lower cell recovery compared to lower freezing density, but the differences in not significant in the conditions tested. It may be probably that the lower concentration cells tend to thaw faster and dilute DMSO faster during post-thawing procedures maintaining cell integrity and lower osmotic cell shock to the cells.

The phenotypic marker expression of MSCs cryopreserved in all the three different cryopreservation media was comparable with >95% expression for CD90 and CD73 and < 3% for CD19 and CD14 markers at all tested variables. There was no significant difference in expression of CD markers in all three freezing media with respect to different freezing densities at different time points except for CD 73 markers in different CS10 concentrations. It should not be a matter of concern as the expression in all of them is above 95% as stated by ISCT guidelines. Though contradictory studies showed the stable expression [25] or decreased expression [23] of phenotypic markers over different time points, we have not observed such difference in any of the time points with respect to freezing media or different freezing densities.

The trilineage differentiation ability of MSCs is also not compromised in any of these three formulations. There are not many studies, which compare the differentiation characteristics of MSCs at different freezing densities and various time points. Nevertheless, a study by Naaldijk et al. 2012 found a slight reduction in osteogenesis capacity of MSCs with higher DMSO concentrations [2]. In our study, we did not do any quantification but we observe that all of these conditions retain the trilineage differentiation capabilities.

Additionally, we evaluated the CFU-F ability and apoptotic cell percentage in cryopreserved BM MSC in different cryopreservation formulations at different cell densities at 6-month time point. These extended assays were done to comply with the stability testing of new drug substances and products of ICH guidelines 21 CFR 31.2.23(a) (7) (ii) which requires a minimum testing period of 6 months to confirm the functionality of the product. The CFU-F ability is one of the markers of stemness and proliferation capacity of MSCs. We observed a similar competence in the number of CFU-Fs formed at different cell concentrations in all three cryopreservation media, except for the two lowest cell densities in CS5 formulation. There are no studies that report the effect of freezing cell concentrations on CFU-F, however, a study reports no difference in number of CFU-Fs between 5% and 10% DMSO concentration in freezing media [26]. With regard to percentage of apoptotic cells at 6 months time point, we found that the lower freezing densities are prone to higher apoptotic rates compared to higher freezing densities in all the formulations. Higher level of intrinsic proteolytic activity may be higher in lower freezing concentration, as the DMSO availability per cell is high, compared to higher freezing cell density, where the DMSO availability per cell is low. Usually, post-thaw activation of caspase-3 demonstrated the proteases activity and subsequently increase intrinsic proteolytic activity following cryopreservation [20, 27]. Hence higher freezing density of BM-MSCs has a lesser apoptotic percentage compared to a lower freezing density.

5. Conclusion

In this study, we demonstrated the possibility of using reduced 5% DMSO containing cryopreservation media (CS5) for cryopreserving BM-MSCs without any impact on viability, phenotypic characteristics and functional properties of MSCs. This was the first study to provide the characterization and comparison of human BM-MSCs cryopreserved in different freezing densities ranging from as low as 5 M cells per ml to higher freezing densities as 25 M cells per ml in two commercially available variants of CS (CS10 and CS5) and comparing it to in-house formulation.

Based on our presented data, we can conclude that chemically defined reduced DMSO-based formulation of CS5 addresses challenges and minimizes the post-cryopreservation manipulation of MSCs for clinical use. However, these data needs to be backed up by safety and efficacy studies, with long-term stability program up to 1 year with more intrinsic molecular, proteomics analysis and immune-suppressive ability of cryopreserved MSCs before employing the CS5 for the cryopreservation of MSCs for therapeutic applications.

Acknowledgments

This work was fully funded by Stempeutics Research Pvt. Ltd., India.

Conflict of interest

The authors declare no conflict of interest.

Author contribution statement

SK: Design of studies, data analysis, interpretation and manuscript writing. SB: Design of studies, perform experiment, data collection and manuscript writing. PKG: Design of studies and manuscript correction. UK: Design of studies and data analysis. Correction and final approval of the manuscript.

Ethics statement

The authors confirm that the ethical policies of the journal, as noted on the journal’s author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. Obtaining bone marrow from consenting healthy donors was approved by the Institutional Ethics Committee (IEC) at the Manipal Hospital, Bangalore, India.