Human bone marrow and adipose tissue samples (N=5 per tissue) were obtained from healthy donors after written informed consent under protocols approved by the Institutional Review Boards of the Virgen de la Arrixaca University Hospital (Murcia, Spain). Bone marrow was collected by iliac crest aspiration in syringes containing 20U/mL sodium heparin (Laboratorios Rovi, Madrid, Spain) followed by density gradient centrifugation over Ficoll-Paque (Thermo Fisher Scientific, Waltham, MA, United States) at 540g for 20 min at room temperature (R/T). Adipose tissue lipoaspirates were first mechanically disaggregated using tweezers and a scalpel, enzymatically digested in DMEM medium containing 2 mg/mL collagenase type I (Roche Diagnostics, Basel, Switzerland) for 45 min at 37°C in a rotational stirrer, filtered through 40-μm nylon cell strainers (BD Biosciences, San Jose, CA, United States) and centrifuged at 540g for 10 min.

After centrifugation, mononuclear cells from bone marrow (BM) or adipose tissue (Ad) samples were collected, washed with Dulbecco’s phosphate-buffered saline (DPBS) (Sigma-Aldrich, St. Louis, MO, United States), plated in 175-cm² culture flasks (NUNC, Thermo Fisher Scientific) at 1.6x10⁵ cells/cm² in Alpha Minimum Essential Medium (MEM; Thermo Fisher Scientific) containing10% fetal bovine serum (FBS; Gibco, Waltham, MA, United States), 1% GlutaMAX (Thermo Fisher Scientific) and 1% penicillin/streptomycin (Lonza, Basel, Switzerland) and incubated at 37°C and 5% CO₂ ( Figure 1 , Steps I and II).

For culture passage, cells were lifted by treatment with TrypLE Express dissociation reagent (Gibco) for 5 min at 37°C. Both BMMSCs and AdMSCs were grown to a maximum of 70-80% confluence at each culture passage and expanded up to passages 3-4 for the subsequent experiments ( Figure 1 , Step III).

Human BMMSCs and AdMSCs were exofucosylated as previously reported (21). Briefly, cells were resuspended at 2x10⁷ cells/mL in the exofucosylation reaction buffer composed of Hanks´ Balanced Salt Solution (HBSS) (Thermo Fisher Scientific) containing 40 μg/mL fucosyltransferase VII (FTVII) (Bio-Techne, R&D Systems, Minneapolis, MN, United States), 10 mmol/L HEPES (Merck Millipore, Burlington, MA, United States), 0.1% human serum albumin (HSA; Albutein, Grifols, Barcelona, Spain) and 1 mmol/L guanosine 5’-diphosphate (GDP)-fucose (Sigma-Aldrich), and treated at 37°C for 1 hour with gentle shaking ( Figure 1 , Step IV). After, treated MSCs were collected by centrifugation and washed twice with DPBS. Cell viability after exofucosylation was assessed by trypan blue exclusion (normally more than 90% live cells), and exofucosylation efficacy was evaluated using the phycoerythrin-labeled anti-sLe^X monoclonal antibodyHECA-452 (BD Biosciences) by flow cytometry.

Five freezing solutions were evaluated for optimizing cryopreservation conditions of fucosylated human BMMSCs (FucBMMSCs) and AdMSCs (FucAdMSCs). These solutions were made as follows: 1) saline solution (0.9% NaCl; Braun, Frankfurt, Germany) containing 10% dimethyl sulfoxide (DMSO; CryoSure, WAK-Chemie Medical GmbH, Steinbach, Germany) and 2% HSA; 2) saline solution containing 5% DMSO, 5% polyethylene glycol (PEG; Sigma-Aldrich) and 2% HSA; 3) saline solution containing 7.5% propylene glycol (PG; Sigma-Aldrich), 2.5% PEG and 2% HSA; 4) NutriFreez^® D10 cryopreservation medium, a commercial serum-free cryopreservation solution containing methylcellulose, 10% DMSO and other undisclosed ingredients (Biological Industries, Cromwell, CT, United States); and 5) CryoStore^®CS10, an animal component-free cryopreservation medium containing 10% DMSO and other undisclosed ingredients (STEMCELL Technologies, Vancouver, Canada).

After exofucosylation, FucBMMSCs and FucAdMSCs were resuspended in ice-cold freezing solutions at 2x10⁶ cells/mL or 5x10⁶ cells/mL, transferred to 2 mL cryovials, stored at -80°C overnight in a freezing container (Corning CoolCell^®; Corning Inc., Corning, NY, United States) to gradually decrease the temperature (1°C/min cooling rate) and then stored in liquid nitrogen for one month ( Figure 2 , Steps IV and V). Before being used in the different experiments, FucBMMSCs and FucAdMSCs cryovials were immediately placed in a water bath at 37°C and the cell suspensions were then rapidly washed with 10 volumes of a thawing solution composed of saline solution containing 2.5% HSA and 5% anticoagulant citrate-dextrose solution, solution A (ACD-A) (Grifols), centrifuged at 400g for 5 min and then gently resuspended in growth culture medium for complete cryopreservant removal ( Figure 2 , Step VI), being available for immediate allogeneic patient treatment ( Figure 2 , Step VII).

Cell viability of just thawed cryopreserved MSCs (without being cultured) was assessed by Annexin-V and 7-AAD staining (Immunostep, Salamanca, Spain) following the manufacturer´s instructions. Briefly, 1x10⁵ cells were resuspended in 100 μL 1X Annexin-V buffer followed by incubation with 5 μL FITC-labeled Annexin-V and 5 μL 7-AAD for 15 min in the dark at R/T. To evaluate MSC immunophenotype, just thawed FucBMMSCs and FucAdMSCs were stained with fluorescence-conjugated monoclonal antibodies for the cell markers CD73, CD90, CD105, CD14, CD20, CD34 and CD45 (Human MSC Phenotyping Kit; Miltenyi Biotec, Bergisch Gladbach, Germany) following the manufacturer´s instructions. Finally, cells were analyzed in an LSR Fortessa X-20 flow cytometer (Becton Dickinson, Franklin Lakes, NJ, United States) within 1h of staining. Non-specific fluorescence was measured using specific isotype monoclonal antibodies. All experimental conditions were evaluated in five separate experiments and analyzed using FlowJo software (FlowJo, LLC, Ashland, OR, United States).

The cell proliferation kinetics of cryopreserved and thawed FucBMMSCs and FucAdMSCs was monitored by calculating the population doubling (PD) level up to 7 days in culture at 37°C using the formula: PD = ( 1 log 10 2 ) × log 10 ( N t N o ) , where No is the number of viable cells at seeding and Nt the number of viable cells at harvest, as previously reported (20). Cumulative PD was calculated as the sum of individual PD obtained at the previous time points.

Cryopreserved and thawed FucBMMSCs and FucAdMSCs in the different experimental conditions were analyzed for their osteogenic and adipogenic differentiation potential using the StemMACS™ OsteoDiff and StemMACS™ AdipoDiff Media for human MSCs (Miltenyi Biotec), respectively, as per the manufacturer’s instructions. Briefly, MSCs were seeded in 24-well plates and cultured at 37°C until reaching 80-100% confluence. Afterward, the osteogenic or adipogenic differentiation media was added and changed every 2-3 days. After 21 days of culture, osteogenic differentiation was assessed by analyzing alkaline phosphatase (ALP) activity and the formation of calcified matrix culture deposits. For ALP activity, cells were fixed with 4% paraformaldehyde in PBS for 10 min at R/T and stained with SigmaFAST™ BCIP-NBT (Sigma-Aldrich) for 10 min at R/T, whereas calcified matrix culture deposits were detected after incubation with Alizarin Red staining solution (Sigma-Aldrich) for 30 min at R/T. For adipogenic differentiation, cells were cultured for 14 days and assessed by detection of intracellular lipid droplets with Oil Red O staining. Briefly, cells were fixed in 10% formalin for 10 min at R/T and stained with 0.3% Oil Red O (Sigma-Aldrich) in 2-propanol for 15 min at R/T. Finally, bright-field images of representative fields were acquired with a phase-contrast microscope (Nikon Eclipse Ti2, Nikon, Tokyo, Japan).

Peripheral blood mononuclear cells from healthy donors were subjected to density gradient centrifugation over Ficoll-Paque (Thermo Fisher Scientific, Waltham, MA, United States) at 540g for 20 min at R/T. Then, T cells were isolated by magnetic separation using the Human Pan T Cell Isolation Kit (Miltenyi Biotec) following the manufacturer´s instructions. Isolated T cells were resuspended in RPMI 1640 medium (Gibco) containing 20% FBS, plated at 1x10⁵ cells/well in 96-well plates, stimulated to proliferate with 10 μg/mL phytohemagglutinin (PHA; Sigma-Aldrich) and co-cultured with decreasing MSC: T cell ratios 1:100, 1:50, 1:25 and 1:10. To prevent the MSC proliferation in the co-cultures but preserving the cell viability, MSCs were previously irradiated at 25 Gy. Finally, T cell proliferation in the co-cultures was measured 4 days later using an ELISA BrdU Colorimetric Kit (Roche Diagnostics), according to the manufacturer´s instructions.

For IFNγ preincubation, 20 ng/mL recombinant human IFNγ (Bio-Techne, R&D Systems) was added in the MSC culture medium for 24 hours and washed three times extensively with PBS before T cell addition.

Cryopreserved and thawed FucBMMSCs and FucAdMSCs at passage 10, obtained after nine sequential freezing, thawing, and in vitro culture cycles, were seeded in 24-well plates at 1×10⁵ cells/well and allowed to attach O/N. Senescence was determined using the Senescence Cells Histochemical Staining Kit (Sigma-Aldrich) according to the manufacturer´s instructions. Briefly, 1X fixation buffer and 1X DPBS were made fresh from stocks and ultrapure distilled water on the day of the experiment. The staining mixture was made immediately before use by mixing 1 mL of the 10X staining solution, 125 μL of reagent B, 125 μL of reagent C, 250 μL of β-gal solution, 8.5 mL ultrapure distilled water (final volume: 10 mL), and then filtered through a 0.2-μm filter. The culture media were carefully aspirated off and cells were washed twice with DPBS. After, 1.5 mL of 1X fixation buffer was added to each well, incubated at R/T for 6 min, and washed twice with 1X DPBS. Finally, 1 mL of the prepared staining mixture was added to each well and each plate was sealed with parafilm. Dishes were then incubated at 37°C without CO₂ O/N. The next day dishes were imaged under a bright-field inverted microscope (Nikon). The total number of SA-β-gal positive blue-stained FucMSCs and their relative number (percentage) respect to the total number of cells was measured in 20 non-overlapping and randomly chosen high power fields (HPF) per experimental condition.

Statistical analysis was performed using GraphPad Prism v8 software (GraphPad, La Jolla, CA, United States). All data are presented as mean ± standard deviation (SD). One-way ANOVA with Tukey’s post-hoc multiple comparison tests was used to determine statistical differences between groups. A P value < 0.05 was considered statistically significant.

After isolation and culture, human BMMSCs and AdMSCs were subjected to cell surface glycoengineering by glycosyltransferase-programmed stereosubstitution (i.e., exofucosylation) using the α(1,3)-fucosyltransferase VII (FTVII) and its donor substrate GDP-fucose to modify CD44 to HCELL, a fucosylated glycoform of CD44 with strong binding affinity for E-selectin, as previously described (29, 30). As is shown in Figure 3A , flow cytometry analysis confirmed that both unmodified human BMMSCs (left) and AdMSCs (right) did not natively express the tetrasaccharide sLe^X as demonstrated by the absence of reactivity with the monoclonal antibody HECA-452. By contrast, after modification of MSCs by exofucosylation (i.e., FucBMMSCs and FucAdMSCs), more than 95% of the cells displayed a substantial upregulation of surface sLe^X expression, as evidenced by intense positive HECA-452 staining ( Figure 3A ), as previously reported for human and mouse BMMSCs and AdMSCs (20, 22, 29). Importantly, the exofucosylation protocol did not alter the surface expression of typical MSC markers, as freshly exofucosylated FucBMMSCs and FucAdMSCs, and similar to unmodified counterparts, were uniformly positive for CD73, CD90, and CD105 ( Figures 4A, B , upper histograms) and negative for CD14, CD20, CD34, and CD45 (not shown).

To evaluate the impact of cryopreservation on the persistence of enforced sLe^X expression after exofucosylation, cryopreserved FucBMMSCs and FucAdMSCs in the different freezing solutions and conditions were thawed and re-cultured at 37°C up to 72h. Fresh FucBMMSCs and FucAdMSCs that were not cryopreserved (i.e., “Fresh FucMSCs”) displayed the maximum level of sLe^X expression immediately after exofucosylation, decreased to ~30% and ~71% after 24h of culture, respectively, and returned to basal absence of sLe^X displayed after 72h ( Figure 3B ). Importantly, we also observed that sLe^X levels were significantly retained in cryopreserved and just thawed FucBMMSCs and FucAdMSCs. However, some differences were found between the different freezing solutions in terms of the persistence of sLe^X expression detection over the culture time. FucBMMSCs and FucAdMSCs cryopreserved in the DMSO and PG/PEG solutions displayed significant decreases of sLe^X expression compared to levels shown by freshly fucosylated MSCs at different time points (*p<0.05, **p<0.01, ***p<0.001). However, FucMSCs cryopreserved in DMSO/PEG, and mainly in both NutriFreez and CryoStor, displayed sLe^X levels significantly higher than those observed with the DMSO solution (Δp<0.05, ΔΔp<0.01, ΔΔΔp<0.001). Again, these differences were more pronounced when FucBMMSCs were cryopreserved at a cell density of 2x10⁶ cells/mL compared to 5x10⁶ cells/mL, and FucAdMSCs at 5x10⁶ cells/mL compared to 2x10⁶ cells/mL, respectively ( Figure 3B ).

Furthermore, we investigated whether cryopreservation and subsequent thawing could affect the expression of positive MSC surface markers on FucBMMSCs and FucAdMSCs ( Figures 4A, B , middle histograms, and bottom graphs). In general, immediately thawed FucBMMSCs and FucAdMSCs showed a slight but significant downregulation of CD73, CD90, and CD105 compared to the levels shown by Fresh FucMSCs (represented by a dashed within the histograms). However, there were again some significant differences when comparing the different freezing solutions. Compared to cells that were cryopreserved in DMSO, FucMSCs cryopreserved in the DMSO/PEG, NutriFreez and CryoStor solutions showed significant higher maintenance of CD73, CD90, and CD105 expression levels (*p<0.05, ***p<0.001). This finding was observed in both FucBMMSCs and FucAdMSCs that had been cryopreserved at either 2x10⁶ cells/mL or 5x10⁶ cells/mL cell density ( Figures 4A, B , middle histograms and bottom graphs).

The effects of different conditions of cryopreservation on the viability of immediately thawed FucMSCs were also investigated through flow cytometry by using Annexin-V and 7-AAD staining ( Figure 5 ). As previously reported for native human MSCs (27, 31, 32), the viability of FucBMMSCs and FucAdMSCs cryopreserved in DMSO (i.e, the most widely used freezing solution for human MSCs) dropped to ~50-60% ( Figure 5 ). However, the percentage of live cells previously cryopreserved in DMSO/PEG, NutriFreez, and CryoStor solutions were significantly higher (at 57.2%-68.9% for FucBMMSCs, and 61.6%-78.6% for FucAdMSCs, respectively) compared to those observed in DMSO (at 52.9%-58.3% for FucBMMSCs, and 55.2%-63.1% for FucAdMSCs, respectively) (*p<0.05, **p<0.01, ***p<0.001), while in PG/PEG were significantly lower (at 29.6%-33.2% for FucBMMSCs, and 32.3%-43.6% for FucAdMSCs, respectively) (ΔΔΔp<0.001). In general, slightly higher viability percentages were obtained when FucBMMSCs were cryopreserved at a cell density of 2x10⁶ cells/mL compared to 5x10⁶ cells/mL, and FucAdMSCs at 5x10⁶ cells/mL compared to 2x10⁶ cells/mL, respectively ( Figure 5 ).

Altogether, these data suggest that FucBMMSCs and FucAdMSCs cryopreserved in DMSO/PEG, NutriFreez, and CryoStor, at 2x10⁶ cells/mL and 5x10⁶ cells/mL, respectively, showed better cell viability, persistence of expression of sLe^X and MSC markers after thawing than DMSO and PG/PEG solutions.

To analyze whether cryopreservation could affect FucMSC morphology directly after thawing, cryopreserved cells in the different solutions and conditions were seeded directly after thawing and cultured for up to 72h. As is shown in Figures 6A, B , Fresh FucBMMSCs and FucAdMSCs (i.e., not cryopreserved) displayed a rapid plastic adherence, cell spreading, and the typical fibroblast-like morphology of MSCs only after 24h of culture. However, as expected, cryopreserved and thawed FucMSCs displayed a delayed attachment compared to Fresh FucMSCs. Among all the freezing solutions tested, CryoStor, NutriFreez, and DMSO/PEG were those showing better and faster cell attachment and spreading, reaching 70-80% confluence after 72h of culture, mainly when FucBMMSCs and FucAdMSCs were cryopreserved at 2x10⁶ cells/mL and 5x10⁶ cells/mL, respectively. On the contrary, DMSO, and especially PG/PEG, showed a remarkable impairment of the reactivation of MSC growing in culture.

In addition, proliferation kinetics of the different cryopreserved and thawed FucMSCs were investigated using the cumulative population doubling (PD) calculation up to 7 days of culture. FucBMMSCs and FucAdMSCs cryopreserved in CryoStor solution at 2x10⁶ and 5x10⁶ cells/mL, respectively, exhibited significantly higher cumulative PD as compared to Fresh FucMSCs (i.e., not cryopreserved) (*p<0.05) and to the DMSO solution (**p<0.01, ***p<0.001), while NutriFreez displayed comparable cumulative PD levels to Fresh FucMSCs ( Figures 7A, B ).

We studied whether exofucosylation and subsequent cryopreservation could affect the osteogenic ( Figure 8 ) and adipogenic differentiation potential ( Figure 9 ) of human FucBMMSCs and FucAdMSCs. For this purpose, cryopreserved and thawed FucMSCs were cultured for 21 days or 14 days in an osteogenic or adipogenic differentiation media, respectively, and subsequently subjected to osteogenic- or adipogenic-specific qualitative staining. Except for the time required to reach cell confluence when using some cryoprotectants and initiating the differentiation cultures, cryopreserved FucBMMSCs and FucAdMSCs in all conditions and thawed showed abundant mineral matrix depositions around the cells, positive alkaline phosphatase activity and lipid droplet formation comparable to those observed in Fresh FucMSCs. In addition, the cell density used for cryopreservation did not significantly alter the osteogenic and adipogenic differentiation capacities of human FucMSCs ( Figures 8A, B , 9A, B ).

To assess whether cryopreservation could affect the immunomodulatory properties of FucMSCs, we analyzed the capacity of FucBMMSCs and FucAdMSCs to inhibit mitogen-stimulated T cell proliferation. Assays with human FucMSCs from different donors and cryopreserved at both 2x10⁶ and 5x10⁶ cells/mL showed that Fresh FucMSCs inhibited PHA-stimulated T cell proliferation in a dose-dependent manner (from a ratio of 1:100 to 1:10 MSC: T cell) more effectively than cryopreserved and thawed cells ( Figure 10 ). However, FucBMMSCs and FucAdMSCs cryopreserved in NutriFreez or CryoStor at 2x10⁶ and 5x10⁶ cells/mL, respectively, showed an immunosuppressive effect comparable to that of Fresh FucMSCs at all MSC: T cell ratios. Regarding the other cryoprotective solutions, FucMSCs frozen in DMSO/PEG also showed a significantly higher immunosuppressive capacity than those cryopreserved in DMSO ( Figure 10 ).

Since DMSO-cryopreserved and thawed MSCs were previously reported to be refractory to IFNγ-preincubation to stimulate their immunosuppressive properties (31), we next studied whether this detrimental effect extended to non-DMSO cryopreservatives. As is shown in Figure 11 , cryopreserved FucBMMSCs and FucAdMSCs in DMSO or PG/PEG solutions and subsequently stimulated with IFNγ were unable to significantly inhibit PHA-activated T cell proliferation compared to Fresh FucMSCs or cryopreserved FucMSCs in DMSO/PEG, NutriFreez and CryoStor counterparts. Strikingly, IFNγ-stimulated FucBMMSCs and FucAdMSCs cryopreserved in CryoStor at 2x10⁶ cell/mL and 5x10⁶ cells/mL, respectively, exhibited a significantly higher anti-mitogenic effect on T cell proliferation at a 1:10 MSC: T cell ratio compared to those observed with Fresh FucMSCs (^# p<0.05) ( Figure 11 ).

Taken together, these results suggest that there are significant differences in the immunomodulatory capacity of cryopreserved FucMSCs in the context of T cell proliferation assays after mitogen stimulation, depending on the cryoprotectant used. Importantly, NutriFreez and CryoStor solutions increase the immunoregulatory activity beyond that observed with Fresh FucMSCs.

To further investigate the relevance of cryopreservation on FucMSC senescence after in vitro culture, we analyzed senescence-associated (SA)-β-galactosidase (gal) expression in FucBMMSCs and FucAdMSCs at passage 10 after nine sequential freezing steps in DMSO, NutriFreez and CryoStor solutions, thawing and in vitro culture ( Figure 12 ). Although low numbers of SA-β-gal positive cells were detected in all cultures, the levels of senescent cryopreserved FucBMMSCs in CryoStor and FucAdMSCs in NutriFreez were significantly decreased and increased, respectively, compared to the DMSO conditions (ΔΔΔp<0.001, ***p<0.001) ( Figure 12 ). Likewise, the percentage of SA-β-gal positive cells in relation to the total number of FucMSCs also showed the same trend (FucBMMSCs: DMSO (14.00% ± 2.3%), NutriFreez (10.3% ± 6.1%), and CryoStor (6.6% ± 3.37%); FucAdMSCs: DMSO (17.6% ± 11.8%), NutriFreez (25.9% ± 10.0%), and CryoStor (17.7% ± 7.5%) (data not shown).