Biological and Physicochemical Characterization of a Serum-and Xeno-Free Chemically Defined Cryopreservation Procedure for Adult Human Progenitor Cells

Isolation and Culture of Progenitor Cell Populations

UCB-ECs were cultured as previously described (10). Briefly, cord blood was collected following elective cesarean sections (gestational age: week 38–40) and with informed consent of the mothers, according to institutional guidelines of the Medical University of Vienna, Austria. The blood was diluted 1:2 with PBS, and mononuclear cells (MNCs) were separated from erythrocytes by density separation on a Ficoll cushion. The mononuclear cells were washed twice with PBS and then resuspended in StemSpan Serum-Free Expansion Medium (SFEM; StemCell Technologies) supplemented with 2 U/ml erythropoietin (Erypo, Jannsen-Cilag, 10,000 U/ml), 100 ng/ml Stem Cell Factor (SCF), 1 × 10^–6 M dexamethasone (Dex), 40 ng/ml insulin-like growth factor-1 (IGF-1), and 20 μg/ml of a cholesterol-rich lipid mix (all from Sigma) at a concentration of approximately 10 × 10⁶ cells/ml. Partial medium changes were performed daily, and adherent cells were removed by daily transfer to fresh culture dishes. Outgrowth of erythroid progenitor cells was monitored by staining of cytospin preparations according to May-Gruenwald-Giemsa and by analyzing the CASY cell counter (Schaerfe, Reutlingen) profiles. As soon as erythroid cells start to dominate the culture (day 6–10), the cell concentration was adjusted to 2 × 10⁶ cells/ml and maintained at this value by daily partial medium changes. At day 14, more than 95% show the characteristics of erythroid progenitor cells in hematological staining, cell diameter, and flow cytometry (>95% of the cells are positive for CD71 and CD36).

UCB-ECFCs were cultured as previously described (56). Briefly, at term UCB units were collected immediately after elective cesarean section from women without signs of infection (gestational age: week 38–40). Informed consent of the mothers was obtained, according to institutional guidelines of the University Hospital Zurich, Switzerland. The citrate-phosphate-dextrose-adenine anticoagulant-1 bag system (MacoPharma, Mouvaux, France) was used for blood collection and the units were processed within 1 h. For MNC isolation, a UCB unit was diluted 1:2 with phosphate-buffered saline (PBS) containing 2 mM ethylenediaminetetraacetate (EDTA; Sigma, Buchs, Switzerland), then processed by Ficoll-Hypaque solution (Biochrom AG, Berlin, Germany) gradient centrifugation at 500 × g for 30 min at room temperature. MNCs were removed from the interphase and washed once with cold PBS/EDTA, pelleted, then resuspended in endothelial cell growth medium (EGM-2, Lonza) and plated on fibronectin-coated (Biomedical Technologies Inc., Stoughton, MA) T75 tissue culture plates at a density of 5 × 10⁵/cm². After overnight incubation, nonadherent cells were removed and fresh endothelial medium was added. Medium was refreshed every 7 days. First cobblestone-like colonies appeared between days 12 and 20 after plating. The colonies were enumerated and then harvested by accutase (Sigma) treatment and replated at seeding densities of 2–4 × 10³ cells/cm² in T75 flasks, then cultured till confluence and serially passaged.

AT-MSCs were cultured as previously reported (57). Briefly, an adipose tissue aspirate (obtained from liposuction after plastic surgery at the plastic surgeon cabinet with informed consent of the donor according to Belgian law) was separated into two fractions: a floating adipose fraction [containing processed lipoaspirate cells (PLA)) and a denser fluid (liposuction aspirate fluid cells (LAF)]. Next, 10 ml of the PLA fraction was digested with a 0.15% collagenase solution and centrifuged for 10 min at 800 × g. Cells were then resuspended in CCM, consisting of low glucose Dulbecco's modified essential medium (DMEM) supplemented with 15% fetal bovine serum (FBS) + 100 U/ml penicillin + 100 μg/ml streptomycin (all from Invitrogen), and filtered through a 40-μm mesh filter, before final plating in T75 culture flasks in CCM. Culture medium was refreshed every 3–4 days until 80% confluence. For initial subcultivation, cells were detached by trypsin/EDTA treatment and replated in T75 flasks at a density of 3 × 10³ cells/cm². Following establishment, AT-MSC cultures were expanded in T75 culture flasks in CCM. For splitting, cells were detached by trypsin/EDTA treatment and replated in T75 flasks at a density of 1.6 × 10³ cells/cm² in 20 ml CCM.

Potential mycoplasma contamination was excluded by PCR using either a mycoplasma detection kit (Miner-vaBiolabsassay, Berlin, Germany) or performed as previously described (5).

Immunophenotyping

Immunophenotyping of UCB-ECs (10), UCB-ECFCs, and AT-MSCs (56) was performed as previously described. Briefly, cells were detached and antibody (Ab) staining was performed on live cells. Primary Abs were directly labeled with fluorescein thiocyanate (FITC), phycoerythrin (PE), or tetramethylrhodamine isothiocyanate (TRITC); in the case of unlabeled primary Abs, dye-labeled goat or rat anti-mouse IgG1, IgG2, or IgG3 (all from Becton Dickinson, Franklin Lakes, NJ, USA) were used as secondary Abs. For staining of intracellular von Willebrand factor (vWf), cells were permeabilized with 0.1% Triton X-100 for 3 min.

The following Abs were used for UCB-ECs: anti-CD3 (clone UCHT1), anti-CD34 (clone 581), anti-CD29 (clone K20), anti-CD81 (clone JS64) from Beckman-Coulter; anti-CD14 (clone MϕP9), anti-CD36 (clone CB38), anti-CD44 (clone G44-26), anti-CD45 (clone HI30), anti-CD71 (clone M-A712), anti-CD117 (clone YB5.B8), anti-CD235A (clone GA-R2/HIR2) all from BD Biosciences; and anti-CD133 (clone 293C3) from Miltenyi Biotec.

The following Abs were used for UCB-ECFCs: anti-CD45 (clone 5B1), anti-CD133 (clone AC133) from Miltenyi Biotec; anti-CD105 (clone MEM-226) from Abcam; anti-CD31 (Part No. 2003788), anti-CD54/ICAM-1 (Part No. 2003785), anti-von Willebrand factor (vWF, Part No. 2003787), anti-CD146 (Part No. 2003789) from Chemicon International; anti-KDR (89106) from R&D Systems; and anti-CD14 (clone M5E2), anti-CD62E/E-selectin (clone 68-5H11), anti-CD117 (clone YB5.B8), anti-CD144 (clone 55-7H1) from Becton Dickinson.

The following Abs were used for AT-MSCs: anti-CD11b (clone ICRF44), anti-CD79a (clone HM47), from BioLegend; anti-CD34 (clone AC136), anti-CD45 (clone 5B1) from Miltenyi Biotec; anti-CD14 (clone M5E2), anti-CD44 (clone G44-26), anti-CD73 (clone AD2), anti-CD90 (clone 5E10), anti-CD166 (clone GHI/61), anti-MHC-I (clone G46–2.6), anti-MHC-II (clone G46-6) from Becton Dickinson; and anti-CD105 (clone MEM-226) from Abcam.

Unspecific isotype Abs were used as controls. Cells were fixed in 4% buffered formalin and analyzed with a flow cytometer (FACSCalibur, Becton Dickinson). A minimum of 10⁴ gated events were acquired for each sample.

Functional Characterization of Fresh or Cryopreserved UCB-ECs

Terminal erythroid differentiation was induced by changing the cytokine composition in the cultivation medium as described (26). Cells were washed once in PBS and seeded in StemSpan SFEM supplemented with the following factors: 10 U/ml erythropoietin, insulin 4 × 10^–4 IE/ml (Actrapid, NovoNordisk 100 IE/ml), 10^–6 M T3 thyroid hormone, 3 × 10^–6 M mifepristone (RU486), 3% human serum (type AB, male) and 1 mg/ml transferrin (all from Sigma). The factors were replenished by daily partial medium changes. Differentiation was followed by monitoring the decrease in cell size with the CASY counter, by benzidine staining of cytospin preparations, and by a colorimetric hemoglobin assay as described in Dolznig et al. (10).

Functional Characterization of Fresh or Cryopreserved UCB-ECFCs

In order to visualize capillary formation, UCB-FCs were seeded at 5 × 10³ to 2 × 10⁴ cells/well of a 96-well tissue culture plate precoated with 50 μl Matrigel (Becton Dickinson). Capillary formation was monitored over the course of 8 h with an inverted phase microscope (DMIL microscope; Leica, Glattbrugg, Switzerland) equipped with a DML digital camera and LeicaQ Win Image Analysis software (Leica Imaging System Ltd, Cambridge England, UK). HUVEC cells were used as positive control in this assay. In order to monitor endothelial cell activation, UCB-ECFCs were incubated with 10 ng/ml of recombinant human tumor necrosis factor-α (TNF-α; Life Technologies) for 4 h at 37°C. Next, upregulation of CD62E and CD54 was determined by flow cytometry using PE-labeled anti-CD62E and anti-CD54 antibodies (Becton Dickinson).

Functional Characterization of Fresh or Cryopreserved AT-MSCs

AT-MSCs were tested in vitro for their differentiation potential into three mesodermal lineages. Differentiation was induced by culture in commercial differentiation media: AdipoDiff, ChondroDiff, and OsteoDiff Induction Media (Miltenyi Biotec GmbH, Germany). Control cultures were grown in NH expansion medium (Miltenyi Biotec GmbH, Germany). For adipogenic differentiation, AT-MSCs were cultured at 5 × 10⁴ cells/well in 12-well plates (TPP, Switzerland) in AdipoDiff medium for 3 weeks, with fresh media added every 48 h. Oil Red O staining was used to visualize lipid droplets. Cells were washed twice with PBS, fixed in precooled methanol for 5 min, washed, and then incubated with 0.5% Oil Red O (Sigma-Aldrich AG, Switzerland) in isopropanol for 20 min at room temperature. Finally, stained cells were washed and analyzed microscopically. Chondrogenic differentiation was performed under micromass condiin ChondroDiff medium. Briefly, 5 × 10⁶ AT-MSCs pelleted by 5-min centrifugation at 300 × g. The micromass was kept in ChondroDiff medium for 24 days, with fresh media added every 48 h. The pellets were fixed in 10% formalin, and then embedded in paraffin. Staining for glycosaminoglycans was performed on deparaffinized 5-μm sections alone or in combination with periodic acid-schiff (PAS), which stains glycogen. For detection of glycoconjugates, sections were immersed in a solution of 1% Alcian Blue dissolved in 3% aqueous acetic acid (pH 2.5) for 10 min. The slides were washed and immersed in Schiffs reagent for 10 min. Subsequently, the sections were washed for 10 min, counterstained with Weigerts hematoxylin, dehydrated in graded ethanol, cleared in xylene, and mounted in resinous medium. For osteoenic differentiation, 3 × 10⁴ AT-MSCs were seeded in 12-well plates (TPP) and maintained in OsteoDiff induction medium for 3 weeks and fresh medium was added every 48 h. After 3 weeks, cells were fixed in 70% ethanol. Alkaline phosphatase production of the cells was evaluated by BCIP-NBT assay (GIBCO). Briefly, the BCIP-NBT solution was prepared and the cells were stained as described in the manufacturer's protocol. Finally, stained cells were washed and analyzed microscopically.

Cryomedia

For the experiments the following cryomedia were used: standard cryomedium (FBS-10:90% FBS + 10% DMSO), serum-/xeno-free chemically defined cryomedium 10 (CM-10: IBMTSTEM + 10% DMSO), cryomedium 5 (CM-5: IBMTSTEM + 5% DMSO), cryomedium 2 (CM-2: IBMTSTEM + 2% DMSO), or cryomedium 0 (CM-0: IBMTSTEM without DMSO; IBMT-media, Fischer Procryotect, Zurich, Switzerland, licensed by Fraunhofer IBMT, St. Ingbert, Germany).

Cryopreservation of Progenitor Cell Populations

Following harvesting of UCB-ECs, UCB-ECFCs and AT-MSCs, cells were washed twice with PBS and resuspended at a concentration of 1 × 10⁶ cells/ml in either FBS-10, CM-10, or CM-5. Next, 1.5-ml aliquots were transferred to ice-cooled cryovials and a 0.2-ml sample was directly removed from each tube to perform cell count and viability analysis prefreezing. Then, cell samples were cooled in a standard freezing container (“Mr. Frosty,” Nalgene, USA) to −80°C at −1°C/min. After ≥ 2 h, samples were transferred to a 1 N₂ tank where they were kept for at least 24 h prior to thawing. For thawing of cryopreserved samples, cryovials were placed in a 37°C water bath for rapid thawing until almost no ice was detectable. Directly after thawing, a 0.2-ml sample was removed from each tube to perform cell count and viability analysis postfreezing (= cell recovery analysis). Next, the cell suspension was diluted in 15 ml of fresh culture medium and incubated for 10 min at 37°C. Following centrifugation, cells were resuspended in 6 ml prewarmed culture medium and subcultured in duplicate culture. At different time points postplating (24 and 48 h for UCB-ECs, 48 and 96 h for UCB-ECFCs and AT-MSCs), cells (including supernatant) were harvested to perform cell count and viability analysis (= cell repopulation analysis). Each cell sample was counted in triplicate using a hemocytometer or automatic cell counter and cell viability was determined by propidium iodide exclusion during flow cytometric analysis.

Scanning Electron Microscopy (SEM) of Cryopreserved UCB-ECs, UCB-ECFCs, and AT-MSCs Postthaw

After cryopreservation in CM-0, CM-2, CM-5, CM-10, or FBS-10 and thawing, cell samples were washed in PBS, fixed in 2% glutaraldehyde in sodium cacodylate buffer, and kept at 4°C for at least 1 day. Next, samples were washed in PBS and seeded for at least 2 h in PBS or culture medium without FBS on ethanol-treated and dried aluminum foils. Samples were prepared for SEM according to standard procedures: washed in water, dehydrated in increased serial ethanol, critical point dried, and coated with heavy metals. Samples were examined with secondary electron (SE) or backscattered electron (BSE) mode in FESEM XL30 (Phillips, USA) at 10 kV accelerating voltage and a 10-mm working distance.

Multiphoton Laser-Scanning Cryomicroscopy (MLSCM) of Cryopreserved UCB-ECFCs and AT-MSCs

The multiphoton laser-scanning microscope LSM510-Meta-NLO (Zeiss, Germany) was used to improve spatial resolution, penetration depth, and photo quality of cell observation during cryopreservation (9). Cells (1.5 × 10⁴–3.8 × 10⁴)/cm² were cultivated for at least 24 h on Thermanox coverslips (Nunc, Germany) with a diameter of 13 mm. The cytoplasm of cells was stained with 2 μg/ml calcein (Invitrogen, UK), the nucleus with 10 μg/ml Hoechst 33,342 (Sanofi-Aventis, Frankfurt, Germany) by incubating for 15 min at 37°C. After staining, cells were incubated for 5 min at 4°C in CM-0, CM-2, CM-5, CM-10, or FBS-10. Afterwards, cells were placed in a liquid nitrogen-cooled cryostage with computerized temperature feedback (temperature range +600°C down to −196°C, MDS 600, Linkam Scientific Instruments, UK). The cryostage is connected to a temperature controller (TMS 94), a liquid nitrogen pump (all Linkam), and a Dewar for liquid nitrogen supply. Stained samples were frozen at 1°C/min from 4°C to −80°C and thawed with 60°C/min to 20°C. For excitation a turnkey Ti: Sapphire femtosecond (fs) laser system (Mai Tai Broadband, Spectra-Physics, CA), which delivers 120 fs laser pulses at a repetition rate of 80 MHz, was used with a central wave length of 800 nm. Images were collected and analyzed using the Zeiss LSM image browser software, version 3,5,0,376.

Statistical Analysis

For in vitro cell recovery, repopulation, and differentiation studies, results are expressed as mean ± SD and comparisons were validated using Student's t-test with a value of p < 0.05 considered to be statistically significant using GraphPad Prism version 5.01 for Windows (GraphPad Software, San Diego, CA, USA). For recovery and repopulation data, results are presented by day-per-day comparisons between the different groups with a value of p < 0.006 (for 0.05 critical significance level according to Bonferroni correction) considered to be statistically significant. The individual effects of FBS-10, CM-10, and CM-5 with respect of UCB-EC volume reduction and HB content during differentiation postthaw in function of time were validated with values of p < 0.002 and p < 0.003, respectively (for 0.05 critical significance level according to Bonferroni correction) considered to be statistically significant.

Derivation and Characterization of Primary Cell Cultures

In order to perform comparative cryopreservation experiments, three distinct ex vivo cultured human adult primary progenitor cell (PC) populations were initiated and consequently tested for mycoplasma contamination, characterized by morphology and immunophenotype: i) UCB-ECs (n = 3 individual donors), ii) UCB-ECFCs (n = 3 individual donors), and iii) adipose tissue AT-MSC (n = 6 individual donors). All three PC populations were mycoplasma free, displayed uniform morphology (Fig. 1A, C, E) and characteristic immunophenotype (Fig. 1B, D, F), according to internationally well-established and accepted criteria (42,48,54). In addition, functional studies were performed to further characterize the established PC populations (Fig. 2). For UCB-ECs, in vitro differentiation studies into red blood cells indicated a uniform decrease in cell volume and increase in hemoglobin content as demonstrated by cytospin analysis (Fig. 2A), CASY-mediated cell volume measurement (Fig. 2B), and hemoglobin determination (Fig. 2C). Proliferating UCB-ECs (Fig. 2A, upper panel) consist of undifferentiated blasts with a cell size of 800 fl. Twenty-four hours after induction of differentiation, the cell size started to decrease to a final volume of 200 fl (Fig. 2B) with concurrent accumulation of hemoglobin detected by benzidin staining (Fig. 2A, middle and lower panel) and colorimetric assay (Fig. 2C). For UCB-ECFCs, capillary formation on Matrigel (Fig. 2D) and upregulation of ICAM-1 (Fig. 2E) and E-selectin (Fig. 2F) upon TNF-α stimulation demonstrated functional endothelial cell properties. For AT-MSCs, in vitro differentiation studies indicated their broad spectrum of differentiation potential towards cell types of mesodermal origin. Differentiation into adipocytes (Fig. 2G), osteocytes (Fig. 2H), and chondrocytes (Fig. 2I) could be demonstrated upon specific stimulation.

OPEN IN VIEWER

Figure 1.

Isolation and characterization of UCB-ECs, UCB-ECFCs, and AT-MSCs. Morphology (A, C, E) and immunophenotype (B, D, F) of umbilical cord blood (UCB)-derived erythroid cells (UCB-ECs, A, scale bar: 20 μm, and B), UCB-derived cobblestone-like endothelial colony forming cells (UCB-ECFCs, C, scale bar: 100 μm, and D), and adipose tissue (AT)-derived mesenchymal stromal cells (AT-MSCs, E, scale bar: 100 μm, and F). Immunophenotyping was performed as described in Materials and Methods.

OPEN IN VIEWER

Figure 2.

Functional properties of UCB-ECs, UCB-ECFCs, and AT-MSCs. Functional characterization of cultured UCB-ECs (A–C), UCB-ECFCs (D–F), and AT-MSCs (G–I). For UCB-ECs, in vitro differentiation studies into red blood cells indicated a uniform decrease in cell volume and increase in hemoglobin content as demonstrated by cytospin analysis (A, scale bar: 20 μm), CASY-mediated cell volume measurement (B, black circles indicate cells in the presence of maintenance and white circles in differentiation medium), and colorimetric hemoglobin determination (C, black circles indicate cells in the presence of maintenance and white circles in differentiation medium). For UCB-ECFCs, formation of capillary-like structures on Matrigel (D, scale bar: 100 μm) and upregulation of ICAM-1/CD54 (E) and E-selectin/CD62E (F) following TNF-α stimulation (isotyp-contol Abs in gray, unstimulated in bold line, and stimulated in dashed line) demonstrated functional endothelial cell properties. For AT-MSCs, in vitro differentiation studies indicated their tripotential differentiation capacity to produce adipocytes (G, Oil Red O staining, scale bar: 40 μm), osteocytes (H, BCIP-NBT assay, scale bar: 40 μm), and chondrocytes (I, PAS-staining, scale bar: 100 μm). Insets indicate unstimulated control samples.

Selection of a Serum- and Xeno-Free Chemically Defined Cryopreservation Medium for UCB-ECs, UCB-ECFCs, and AT-MSCs

First, a prospective analysis of commercial serum-free cryomedia was performed (Table 1, commercial cryomedia) and four media were selected for potential studies: IBMT^PLUS, IBMT^STEM, Cryostor CS-10, and Cryopan II. These media all contained 10% DMSO with the exception of IBMT^STEM, which only contains 5% DMSO. Osmolarities of the media containing 10% DMSO was in the range of 2,100–2,500 mOsmol, and freezing points varied between −26.8°C and −33.4°C. Only three media, IBMT^PLUS, IBMT^STEM, and Cryostor CS-10, fulfilled the criterion of being free of xenogenic compounds. The IBMT media utilize the compound Pluronic F-68 at a concentration of 1% in a chemically fully defined, although not disclosed, mixture (14). As initial tests with this cryomedium were very promising, we decided to use this solution for all further comparisons to serum-based media. Additionally, we used DMSO-free formulation of IBMT^STEM and customized the DMSO concentration for our tests. The physical properties of three different customized media (IBMT^STEM with 0%, 5%, and 10% DMSO) were analyzed (Table 1, CM-0, CM-5, and CM-10). This analysis clearly showed that the osmolarity strongly increases and the freezing point decreases with increasing DMSO concentrations. These customized media were used for all subsequent experiments. Furthermore, all comparisons were made versus cryopreservation in commonly used serum-containing cryomedium (Table 1, common cryomedia: 90% FBS + DMSO).

OPEN IN VIEWER

Table 1.

Comparison of Physical Properties and Ingredients of Different Cryomedia

					Standard Cryomedia
	Commercial Cryomedia				90% FBS + DMSO (FBS-10)	Culture Medium + DMSO	Tested Cryomedia
	IBMTPlus	IBMTStem	Cryostor CS-10	Cryopan II			CM-0	CM-5	CM-10
DMSO (%)	10	5	10	10	10	10	0	5	10
Endotoxin free	no	yes	yes	ND	ND	ND	yes	yes	yes
Xeno free/protein free	yes	yes	yes	ND	no	no	yes	yes	yes
Buffer system	MOPS	NA	HEPES	ND	—	—	NA	NA	NA
Pluronic F-68 (%)	1	1	0	0	0	0	1	1	1
Osmolarity (mOsmol)	2163	1053	2592	2246	2244	2420	284	1111	2104
Freezing point (°C)	–30.4	–28.6	–33.4	–26.8	–29.4	–30.0	–28.1	–30.1	–33.6
Phenol red indicators	yes	no	no	no	no	yes	no	no	no

Cryomedia investigated in this study are shown in bold. ND, not determined, NA, data not available.

Serum- and Xeno-Free, Chemically Defined Cryopreservation of UCB-ECs, UCB-ECFCs, and AT-MSCs

In order to perform a comparative survival analysis between PCs cryopreserved in a chemically defined versus FBS-containing cryomedium, we used three different parameters: “cell recovery,” “cell repopulation,” and “Δ recovery and repopulation” (Fig. 3). Cell recovery is defined as the percentage of viable cells recovered following cryopreservation in chemically defined freezing or standard FBS-containing freezing medium directly postthaw compared to the number of viable cells before freezing. Cell repopulation is defined as the percentage of viable cells recovered following cryopreservation in chemically defined or standard FBS-containing freezing medium at different time points postplating of cryopreserved cells (24 and 48 h for UCB-ECs, 48 and 96 h for UCB-ECFCs and AT-MSCs). Δ recovery and repopulation is defined as the mean cell repopulation of PC cryopreserved in chemical defined medium versus PCs cryopreserved in FBS-containing medium. During each cryopreservation experiment, three cryovials were prepared for every condition (i.e., cryopreservation in standard FBS-containing medium with 10% DMSO) (Fig. 3, FBS-10), chemically defined cryopreservation medium with 10% DMSO (Fig. 3, CM-10), and chemically defined cryopreservation medium with 5% DMSO (Fig. 3, CM-5). The absolute number of viable cells in each tube were directly counted before (Fig. 3A, C, and E; dashed line indicating 100%) and after (Fig. 3A, C, and E; white bars postthaw) cryopreservation using a hemocytometer/automatic cell counter combined with FACS analysis, as described in Materials and Methods. In addition, following cryopreservation, triplicate cultures were plated and the absolute number of viable cells in each culture was determined again after 24 h (Fig. 3A; gray bars) and 48 h (Fig. 3A; black bars) of culture for UCB-ECs and after 48 h (Fig. 3C, E; gray bars) and 96 h (Fig. 3C, E; black bars) of culture for UCB-ECFCs and AT-MSCs. Figure 3 provides a representative example for a comparative analysis of cell recovery and cell repopulation for UCB-ECs (Fig. 3A), UCB-ECFCs (Fig. 3C), and AT-MSCs (Fig. 3E). The graphs on the right (Fig. 3B, D, F) provide the Δ recovery and repopulation summarized for all experiments performed with UCB-ECs (n = 8), UCB-ECFCs (n = 9, passages 5–7), and AT-MSCs (n = 15, passages 2–8) and indicate no statistical difference for CM-10 versus CM-5.

OPEN IN VIEWER

Figure 3.

Cell recovery and repopulation of UCB-ECs, UCB-ECFCs, and AT-MSCs after serum- and xeno-free chemically defined cryopreservation. “Cell recovery,” “cell repopulation,” and “Δ recovery and repopulation” for UCB-ECs (A, B), UCB-ECFCs (C, D), and AT-MSCs (E, F) cryopreserved in FBS-10, CM-10, and CM-5. Representative example given of cell recovery provided as the percentage of viable cells recovered following cryopreservation directly postthawing compared to the number of viable cells before freezing (A, C, E; dashed line and white bars postthaw). Cell repopulation is provided as the percentage of viable cells recovered following cryopreservation at different time points postplating of cryopreserved cells [24 and 48 h for UCB-ECs (n = 3 individual donors), 48 and 96 h for UCB-ECFCs (n = 3 individual donors), and AT-MSCs (n = 6 individual donors; gray and black bars)]. Δ recovery and repopulation provided as the mean cell repopulation of PCs cryopreserved in CM-5 and CM-10 versus PCs cryopreserved in FBS-10 medium for all time points tested (B, D, F). Values are mean ± SD.

Multiphoton Laser-Scanning Cryomicroscopy of UCB-ECFCs and AT-MSCs Before, During, and After Cryopreservation

After identifying comparable cryopreservation efficiency of the chemically defined media even with reduced DMSO concentrations, we investigated the effects on membrane integrity during cryopreservation and thawing in real time. Moreover, the possibility of further DMSO reduction was analyzed. Therefore, AT-MSCs (Fig. 4A–O) and UCB-ECFCs (Fig. 4P–AD) cultured on coverslips were fluorescently labeled with calcein (green fluorescence) to stain the cytoplasm of viable cells and with Hoechst 33,342 (blue fluorescence) to counterstain the nuclei. Labeled UCB-ECFCs and AT-MSCs were covered with CM-0, CM-2, CM-5, CM-10, or FBS-10 cryomedium and placed on a temperature-controlled liquid N₂-cooled cryostage, as described in Materials and Methods. The processes of ice crystallization and membrane disruption before (4°C), during (–80°C), and after (4°C) cryopreservation were examined by multiphoton laser-scanning cryomicroscopy. Cryopreservation of AT-MSCs (Fig. 4A, B, C) and UCB-ECFCs (Fig. 4P, Q, R) in CM-0 resulted in significant intercellular (arrows) and intranuclear (asterisks) ice formation at −80°C. Cryopreservation in CM-2 resulted mainly in the formation of large intercellular ice crystals for both AT-MSCs (Fig. 4E, dashed arrows) and UCB-ECFCs (Fig. 4T, dashed arrows). Following thawing after cryopreservation in CM-0 and CM-2, cytoplasm was released due to membrane disruption as indicated by fading of cytoplasm staining and an increased green background fluorescence of AT-MSC (Fig. 4C, F) and UCB-ECFC samples (Fig. 4R, U). During cryopreservation in CM-5, only limited intercellular ice crystals were observed for AT-MSCs (Fig. 4H, dashed arrows), but not for UCB-ECFCs (Fig. 4W). No intercellular or intracellular ice crystal formation was observed when cells were cryopreserved in CM-10 or FBS-10 for both AT-MSCs (Fig. 4K, N) and UCB-ECFCs (Fig. 4Z, AC). Likewise, after thawing only limited cytoplasmic loss was observed for AT-MSCs cryopreserved in CM-5 (Fig. 4I), but not for AT-MSCs cryopreserved in CM-10 and FBS-10 (Fig. 4L, O) and for UCB-ECFCs cryopreserved in CM-5, CM-10, and FBS-10 (Fig. 4X, AA, AD). Based on these data, we conclude that lack of cell recovery following cryopreservation of UCB-ECFCs and AT-MSCs in CM-0 and CM-2 (data not shown) is due to intercellular and intracellular ice crystal formation, which is not observed following cryopreservation in CM-5, CM-10, and FBS-10.

OPEN IN VIEWER

Figure 4.

Multiphoton laser-scanning cryomicroscopy (cryo-MPLSM) of UCB-ECFCs and AT-MSCs. Multiphoton laser-scanning cryomicroscopy images of AT-MSCs (A–O) and UCB-ECFCs [P-AD before (left column, 4°C), during (middle column, −80°C), and postthaw (right column, 4°C)] in CM-0, CM-2, CM-5, CM-10, and FBS-10. Representative pictures are provided for each condition. Green fluorescence indicates calcein staining of intracellular proteins. Blue fluorescence indicates Hoechst 33,342 staining of nuclei. Scale bars: 50 μm.

Scanning Electron Microscopy (SEM) of UCB-ECs, UCB-ECFCs, and AT-MSCs Postthaw

For proving these results, SEM was used to analyze membrane integrity and structural alterations of UCB-ECs (Fig. 5A–F), UCB-ECFCs (Fig. 5G–L), and AT-MSCs (Fig. 5M–R) after cells had been cryopreserved in chemically defined cryomedia containing various concentrations of DMSO. UCB-ECs are shown before (Fig. 5A) and after (Fig. 5B–F) cryopreservation. Most of the cells had a round shape (Fig. 5A, left, arrow), although some also were elongated (Fig, 5A, left, arrowhead). Further, UCB-ECs possessed heterogeneous surface features: few cells were smooth (Fig. 5A, left, arrow), whereas most of cells had a microvillous relief (Fig. 5A, left, arrowhead) or were covered with lots of vesicles (Fig. 5A, right, arrowhead) and with short microvilli (Fig. 5A, right, arrow). After cryopreservation with CM-0 or CM-2, swelling of cells was observed, as the cell diameter increased from 7–8 μm (Fig. 5A, left) to approximately 12 μm (Fig. 5B, C, left). Signs of cell damage (e.g., disturbed cell surfaces without microvilli and numerous holes) (Fig, 5B, right, arrows) were visible on the cell surface. Cryopreservation with CM-2 resulted in improved membrane integrity (Fig. 5C). But nevertheless, any feature of cell damage ranging from compact and smooth (Fig. 5C, inset) to disturbed surface with numerous ruffles and vesicles was seen (Fig. 5C, right, arrows). Cryopreservation of UCB-ECs with CM-5 and CM-10 showed the best results (Fig. 5D, E). Thus, the form, size, and cell surface hardly displayed any differences from control preparations without cryopreservation. Some cells had a wrinkled surface with single microvilli (Fig. 5E, right, arrow), others a microvillous relief (Fig. 5D, left, arrow). On the contrary, cells frozen with FBS-10 presented various responses to cryopreservation. Some cells were covered with numerous vesicles and only few microvilli (Fig. 5F, right, arrow), whereas others had a wrinkled surface with deep disruptions (Fig. 5F, left, arrow) of the plasma membrane.

OPEN IN VIEWER

Figure 5.

Scanning electron microscopy (SEM) of UCB-ECs, UCB-ECFCs, and AT-MSCs following serum- and xeno-free chemically defined cryopreservation. SEM images of UCB-ECs (A–F), UCB-ECFCs (G–L), and AT-MSCs (M–R) before (A, G, M) and after cryopreservation in CM-0 (B, H, N), CM-2 (C, I, O), CM-5 (D, J, P), CM-10 (E, K, Q), and FBS-10 (F, L, R). For each condition a series of two images is provided at different magnification as indicated by scale bars. Description is given in the Results section.

UCB-ECFCs are shown before cryopreservation (Fig. 5G) and postthaw (Fig. 5H–L). Cells were mostly round or oval in shape within a size of up to 30 μm covered with numerous vesicles (Fig. 5G, arrowhead) and some short microvilli (Fig. 5G, arrow). Some cells in the same sample were round with only 10 μm in diameter (Fig. 5G, right). In most SEM micrographs of UCB-ECFCs small structures (below 1 μm) could be observed both on the cell surface and the substrate surface (data not shown). Our analysis showed that as origin of these structures a contamination can be excluded. Shrinkage of cells with deep holes (Fig. 5H, arrowheads) was seen immediately after cryopreservation using CM-0. Addition of 2% DMSO resulted in smooth cell surfaces with less signs of cell damages (Fig. 5I). Smoothened ruffled membrane of UCB-ECFCs, cryopreserved in CM-5 (Fig. 5J), changed to wrinkled surface with single microvilli by cryopreservation in CM-10 (Fig. 5K) or serum-containing medium (Fig. 5L).

AT-MSCs before and after cryopreservation are presented in Figure 5M and Figure 5N–R, respectively. Cells had a size of 20–30 μm, a round or oval shape, and a surface relief with microvilli and ruffles. Some-times also vesicles were seen on the surface (Fig. 5M, arrow). As expected, cryopreservation in CM-0 significantly changed cell morphology, leading to cells covered with numerous vesicles (Fig. 5N, arrows) and small holes were seen on the cell surface (Fig. 5N, arrowhead). Addition of 2% DMSO already resulted in a visible improvement (Fig. 5O). Cells were covered with numerous microvilli (Fig. 5O, left) without any signs of cell damage, and some cells presented dense vesicles (Fig. 5O, arrows). Only minor differences were seen between CM-5 (Fig. 5P), CM-10 (Fig. 5Q), and FBS-10 (Fig. 5R). In each preparation, cells were densely covered with larger microvilli; the appearances of fine long filaments can be also noted (Fig. 5P, Q, R, arrows). Nevertheless, after cryopreservation with CM-10 some shrunken cells with cavities on the cell surface could be observed (Fig. 5Q, inset).

Functional Characterization of UCB-ECs, UCB-ECFCs, and AT-MSCs Following Chemically Defined Cryopreservation

Following establishment of chemically defined cryopreservation conditions, additional functional studies were performed to demonstrate differentiation potential of UCB-ECs (Fig. 6A–C), UCB-ECFCs (Fig. 6D–F), and AT-MSCs (Fig. 6G–I) after cryopreservation in FBS-10, CM-10, or CM-5. For UCB-ECs, no difference was observed following in vitro differentiation into red blood cells, as indicated qualitatively by a uniform change of morphology (Fig. 6A) and quantitatively by a decrease in cell volume (Fig. 6B). However, after cryopreservation in FBS-10, CM-10, or CM-5, cells, which had been cryopreserved in CM-5, showed a significant lower hemoglobin content at day 7 versus FBS-10 (5.49 ± 0.12 vs. 8.03 ± 0.15, p = 0.0002) and no difference versus CM-10 (5.49 ± 0.12 vs. 8.35 ± 0.53, p = 0.006) (Fig. 6C). For UCB-ECFCs, neither quantitative differences in the upregulation of ICAM-1 (Fig. 6D) and E-selectin (Fig. 6E) following TNF-α stimulation, nor qualitative difference in capillary formation on Matrigel (Figure 6F), were observed after cryopreservation in FBS-10, CM-10, or CM-5. For AT-MSCs, in vitro differentiation studies indicated their broad spectrum of differentiation potential from cell types of mesodermal origin following cryopreservation in CM-10. These cells displayed adipogenic (Fig. 6G), osteogenic (Fig. 6H), and chondrogenic (Fig. 6I) differentiation following specific stimulation.

OPEN IN VIEWER

Figure 6.

Functional characterization of UCB-ECs, UCB-ECFCs, and AT-MSCs following serum- and xeno-free chemically defined cryopreservation. Functional characterization of cultured UCB-ECs (A–C), UCB-ECFCs (D–F), and AT-MSCs (G–I) post-thaw. For UCB-ECs, in vitro differentiation studies into red blood cells indicated a uniform decrease in cell volume and increase in hemoglobin content as demonstrated by cytospin analysis (A, scale bar: 20 μm), CASY-mediated cell volume measurement (B, circle FBS-10, triangle pointing up CM-10, and triangle pointing down CM-5), and colorimetric hemoglobin determination (C, circle FBS-10, triangle up CM-10, and triangle down CM-5). Values are mean ± SD, n = 3 individual donors. ^*p < 0.003. For UCB-ECFCs, upregulation of ICAM-1/CD54 (D) and E-selectin/CD62E (E) following TNF-α stimulation and formation of capillary-like structures on Matrigel (F, scale bar: 100 μm) demonstrated functional endothelial cell properties. Values are mean ± SD, n = 3 individual donors. For AT-MSCs, in vitro differentiation studies indicated their tripotential differentiation capacity after freeze-thaw in chemically defined cryopreservation medium to produce adipocytes (G, Oil Red O staining, scale bar: 40 μm), osteocytes (H, BCIP-NBT assay, scale bar: 40 μm), and chondrocytes (I, PAS and Alcian blue staining, scale bar: 100 μm). Insets show unstimulated control samples.