1 Introduction

Umbilical cord blood (CB) is increasingly used as a source of alternative hematopoietic stem cells for transplantation [1–3]. Most CB units in the world’s CB banks are cryopreserved in liquid nitrogen in individual bags for several years and then are quickly thawed just prior to hematopoietic cell transplantation (HCT). CB units, many of which have been cryopreserved for more than 10 years, are available for HCT in many worldwide CB banks. However, little information is available on the long-term survival of cryopreserved CB cells. Most reports focus on relatively short storage intervals [4–11]. Although a few reports are available on effects of long-term cryopreservation on CB cells [12–16], these CB units have used only for study not for HCT. The CB bank network was established in Japan in 1998, and CB units cryopreserved more than 10 years are considered for public cancellation. No reports have focused on the quality of long-term cryopreserved CB units for HCT, and it is not clear whether long-term cryopreservation affects the quality of CB units for HCT. Therefore, this study analyzed the effect of long-term cryopreservation on hematopoietic cells of CB units that were cryopreserved 10 years ago for HCT. The number of total nucleated cell (TNC) count, CD34+ cells count, colony-forming units-granulocyte/macrophage (CFU-GM) assay, viability of mononuclear cells were evaluated before cryopreserving and after thawing. The CD34+ CD38− phenotype identifies a UCB primitive subpopulation of hematopoietic stem cells that have a higher proliferative potential in response to cytokines and a higher cloning efficiency than the same immunophenotype in adult bone marrow [17–19]. The receptor for stromal cell-derived factor-1 (SDF-1) was subsequently identified on human lymphocytes as CXCR4 [20]. SDF-1 and CXCR4 are necessary for engraftment of human CD34+ cells in the NOD/SCID model [21]. Given this fact, the percentage of CD34 cells expressing CD38 and CXCR4 were also evaluated as stem cell homing and mobilization markers.

The aim of the study was to evaluate the effect of long-term cryopreservation on CB units for HCT.

2 Materials and methods

The study was performed at Showa University Fujigaoka Hospital in September 2008.

CB units were obtained from the Kanagawa Cord Blood Bank in Japan. All protocols used for the recruitment of subjects and the collection of umbilical cord blood were reviewed and approved by the ethics committee of Showa University School of Medicine. The study was approved by the ethical commission of the Showa University Fujigaoka Hospital. All volunteers were required to sign informed consent forms prior to admission into the study. The study group included 18 CB units that were collected between April 1998 and September 1998 and accepted for clinical use as HCT. These CB units were not suitable for HCT based on the cryopreserved period (more than 10 years) in Japan. The control group included 18 units collected between May 2008 and June 2008, and could not be used for HCT because of TNC. More than 8.0 × 108 TNC content was required to accept cord blood units for banking in Japan. These CB units were cryopreserved for 1 month. This study analyzed the number of TNC count, CD34 cells count, CFU-GM assay, viability, and the percentage of CD34 cells expressing CD38 and CXCR4 before cryopreserving and after thawing. The viability and the percentage of CD34 cells expressing CD38 and CXCR4 in the study group were not analyzed before cryopreserving. In addition, the relationship between CD34+ cells and the CFU-GM content of the CB units after thawing was analysed.

2.1 Collection of CB units

CB donors must provide informed consent before delivery. CB samples were obtained from normal full-term deliveries in the obstetric departments of collaborating Hospitals. Every CB unit was collected after delivery of the neonate and ligation of the cord, and prior to the expulsion of the placenta [12]. The cord was disinfected, the umbilical vein was punctured, and the CB was collected into a 200-ml collection bag (CBS-20A; Nipro Co, Osaka, Japan) containing 28 ml citrate–phosphate-dextrose as an anticoagulant.

2.2 Processing of CB units

Hydroxyethyl starch separation was performed according to Rubinstein et al. [22]. Aliquots were removed for routine testing and 6% (wt/vol) hydroxyethyl starch (HES40; Nipro Co) was added to the anticoagulated CB to a final concentration of 1.5% (1:3 volume ratio) to increase the low sedimentation rate of the red blood cells. A leukocyte-rich supernatant was separated by centrifuging the mixture in the collection bag (60g for 5 min at 10°C), transferring it into a separation bag (CBP-20D; Nipro Co), and centrifuging (400g for 10 min at 10°C) to sediment the CB cells. Surplus supernatant plasma was transferred into a second bag via the connecting tube. Finally, the sedimented leukocytes were resuspended in supernatant plasma to a total volume of 23 ml for use as the leukocyte concentrate.

2.3 Cryopreservation and thawing of CB units

Dimethyl sulfoxide (DMSO; Cryoserv; Edwards Life-sciences, Irvine, CA, USA) was used at a final concentration of 10% (vol/vol). The required volume of sterile, chilled 50% DMSO in 5% (wt/vol) dextran 40 (Terumo, Tokyo, Japan) was slowly added to the leukocyte concentrates, and were chilled with wet ice throughout the addition. The final volume of leukocyte concentrates with DMSO/dextran was adjusted to 25 ml. The cell suspensions were mixed and transferred to a 25-ml double-compartment freezing bag consisting of a 20-ml large chamber and 5-ml small chamber (F-025; Nipro Co) and maintained at 4°C in preparation for cryopreservation. The freezing bag was enclosed in a soft plastic bag (ThermoGenesis, Rancho Cordova, CA, USA) to provide a barrier against infectious pathogens. A protective aluminum canister housing the freezing bag was wrapped with heat-insulating materials and deposited horizontally on the level surface of a styrene-foam plate inside a −85°C electric freezer (Sanyo Electric, Tokyo, Japan). The temperature of the freezing bag was monitored until it was below −80°C, and the frozen CB units were cryopreserved in the vapor phase of liquid nitrogen at −195°C using programmed freezer.

These CB units were thawed. CB units were removed from the liquid nitrogen tank, exposed to room temperature for 5 min, and thawed by rapid immersion in a 37°C water bath. A 0.5 ml sample of each concentrate was transferred to a 1.5-ml polypropylene tube (BM Bio, Tokyo, Japan), diluted with an isotonic salt solution containing 5% (wt/vol) human serum albumin (Wellfide, Osaka, Japan) and 10% (wt/vol) dextran 40 with continuous mixing, and placed on ice for 5 min. The cells were counted with an automated blood cell counter (Sysmex 2100; Toa, Kobe, Japan).

2.4 TNC count

TNC count was obtained with an automated blood cell counter (Sysmex 2100; Toa, Kobe, Japan). Postthawing recovery of TNC was calculated as follows:

Percent recovery = TNC after thawing/TNC before freezing × 100.

2.5 CD34+ cell analysis

CD34+ cells were quantified by flow cytometry. Briefly, 10 μL of a combination of phycoerythrin (PE)-conjugated anti-CD34 and fluorescein isothiocyanate (FITC)-conjugated anti-CD45 monoclonal antibodies (BD Biosciences, San Jose, CA, USA) and 10 μL of Via-Probe (BD Biosciences) were added to Trucount tubes (BD Biosciences) containing FlowCount fluorescent beads (Beckman Coulter, Brea, CA, USA). Via-Probe is a ready-to-use solution of the nucleic acid dye 7-aminoactinomycin D (7-ADD). Next, 50 μL of each CB samples was poured into each tube, vortexed, and mixed well. The sample was incubated for 15 min and 1 ml of Tris-ammonium chloride lysing reagent was added to each tube. The samples were analyzed by flow cytometry after 5 min (FACSCalibur; BD Pharmingen, San Diego, CA, USA) with CellQuest software (BD Biosciences). Sequential gating of the CD45+ cell population was used to identify the CD34+ subpopulation according to the single-platform guidelines of the International Society of Hematotherapy and Graft Engineering (ISHAGE) [23]. Adding a known number of FlowCount fluorescent beads enabled the determination of an absolute CD34+ cells count for each sample. At least 100,000 events were analyzed for each sample. The absolute number of CD34+ cells per microliter of sample was calculated according to the following formula:

CD34+ cells/μL = CD34 cell count × beads per test tube × dilution factor/intake beads count × test volume (50 μL).

Absolute CD34 cell counts were calculated by multiplying the number of CD34+ cells per μL by the unit volume.

Percent recovery = CD34 cell counts after thawing/CD34 cell counts before freezing × 100.

2.6 CFU-GM assay

The CFU-GM assay was performed using a commercially prepared complete methylcellulose medium (MethoCult GFH4434V; Stem Cell Technologies, Vancouver, BC, Canada). CFU-GMs were calculated as the sum of the two kinds of colonies. The cultures were plated at 5 × 104 cells/plate in two duplicate 35 mm diameter Petri dishes and incubated for 14 days at 37°C with 5% CO2 in a 100% humidified atmosphere. Colonies defined as aggregates of more than 40 cells were counted under an inverted microscope. The mean number of colonies in the two wells was used to calculate the number of colonies per 5 × 104 cells plated.

Percent recovery = CFU-GMs after thawing/CFU-MBs before freezing × 100.

2.7 Viability assay

The trypan blue dye-exclusion test was used to distinguish live and dead cells. A minimum of 500 unstained viable cells and stained dead cells were counted with an improved Neubauer hemocytometer (Heinz Herenz Medizinalbedarf, Hamburg, Germany), and cell viability was expressed as a percentage of the total cells counted. The 7-AAD stain was used in flow cytometry-based viability testing of CD45+ and CD34+ cells. CD45+ cell- and CD34+ cell- specific viabilities were obtained by evaluating the proportion of 7-AAD-negative cells in the CD45+ and CD34+ cell population.

2.8 Flow cytometry

The percentage of CD34 cells expressing CXCR4 and CD38 was assessed by two-color flow cytometry. Whole samples were incubated for 15 min at room temperature protected from light with the following antibodies: phycoerythrin (PE)-conjugated anti-CD34 (8G12, Becton–Dickinson, San Jose, CA, USA), fluorescein isothiocyanate (FITC)-conjugated anti-CD38 (T16, Becton–Dickinson) and anti-CXCR4 (12G5, Becton–Dickinson, San Jose, CA, USA). These incubated samples were then lysed with FACS lysing solution for 10 min and then washed twice with phosphate-buffered saline (PBS). A control sample was incubated with PE anti-CD34, FITC IgG2a (Becton–Dickinson). A total of 1.0 × 105 total events were acquired. CD34+ cells were isolated using ISHAGE gating strategy, and CD38 and CXCR4 were measured directly on this CD34+ cell population. Flow cytometry was performed using a FACS-caliber instrument (Beckton Dickinson), and the Cell Quest software was used for analysis.

2.9 Statistical analysis

Results are expressed as the mean ± SD. The correlation between CD34 cell counts and CFU-GM counts of CB units after thawing was evaluated by a linear regression analysis. Statistically significant differences between the study group and the control group were detected using Student’s t test. Statistical significant was set at p values of not greater than 0.05. All analyses were performed using the StatView software package (version5.0; SAS institute, Cary, NC).

3 Results

The, TNC, CD34+ cells and CFU-GMs content per bag before cryopreserving and after thawing for both groups are shown in Table 1. The number of TNC, CD34+ cells and CFU-GMs decreased significantly after thawing for both the groups. The recoveries of TNC, CD34+ cells and CFU-GMs are shown in Table 1. Recoveries of TNC, CD34+ cells and CFU-GMs were 88.72 ± 16.40, 68.39 ± 18.37 and 42.28 ± 38.16% in the study group and 80.17 ± 14.46, 72.67 ± 20.38 and 49.61 ± 36.39% in the control group (p = 0.106, p = 0.513 and p = 0.559, respectively). There were no significant differences for the recovery of TNC, CD34+ cells and CFU-GMs between the study group and the control group.

Table 1 Before freezing and after thawing data for cord blood units (n = 18)
Full size table

The percentage of viable cells, CD34+ CD38− cells, and CD34+ CXCR4+ cells are shown in Table 2. The mean basal percentage of viable cells, CD34+ CD38− cells, and CD34+ CXCR4+ cells after thawing were 83.69 ± 9.45, 9.11 ± 4.13 and 81.65 ± 10.82% in the study group and 86.73 ± 6.35, 19.94 ± 8.70 and 65.92 ± 11.95% in the control group (p = 0.266, p < 0.001 and p < 0.001, respectively). The differences between the groups were statistically significant, except for cell viability (p = 0.266).

Table 2 Percentage of viability, CD34+/CD38−, CD34+/CXCR4+ populations before freezing and after thawing
Full size table

Figure 1 shows the relationship between CD34+ cells and CFU-GMs after thawing in the both groups. There was positive relationship between CD34+ cells and CFU-GMs in the study group, although the correlation is moderate (r = 0.430; p = 0.068; Fig. 1a). The control group showed a significant, correlation between CD34+ cells and CFU-GMs (r = 0.706; p < 0.001; Fig. 1b).

Fig. 1
figure 1

Relationship between CD34+ cells and CFU-GM content of cord blood units after thawing. a 10 years cryopreserved (n = 18; r = 0.430, p = 0.068). b 1 month cryopreserved (n = 18, r = 0.706, p < 0.001)

Full size image

4 Discussion

CB is the only hematopoietic progenitor cell (HPC) source that is cryopreserved for years before use. CB units, many of which have been stored for more than 10 years, are available for HCT in many worldwide CB banks. However, little information is available on long-term survival of cryopreserved CB cells.

The results of this study indicate that long-term cryopreservation does not affect the quality of the CB units for HCT.

The number of TNC that was measured before cryopreservation is utilized to determine the suitability of a CB unit prepared for HCT [24]. Although the TNC decreased significantly after thawing in both the groups in the present study, there was no significant difference in the recovery of TNC between the study group and the control group. The results are similar to previous reports of long-term cryopreserved CB units [12–14, 16].

In addition, the number of CD34+ cells and CFU-GM may be a good indicator of quality [1, 25]. The importance of CD34+ cell content is also apparent, because recent studies have shown a correlation between the number of CD34+ cells in the product and the time to complete engraftment [26, 27]. Although CD34+ cells decreased significantly after thawing in both the groups, there was no significant difference in the recovery of CD34+ cells between the study group and the control group. The recovery data for CD34+ cells of long-term cryopreserved CB units have not been reported previously. Therefore, the current results were compared with previously published reports that included short-term cryopreserved data.

Some report a good recovery of CD34+ cells after thawing [4, 7, 10, 11]. Laroche and co-workers [4] reported that the recovery of CD34+ cells was 97% after thawing. Although they think these results would be that CB processing has little impact on CD34+ cell content, they said one should be cautious before drawing such a conclusion. Because the number of CB units included in their study was relatively low, and because a large variability in CD34+ cells was collected, the standard deviation (SD) in their sampling was elevated.

The CB units in those reports were used only for the experimental study, not for HCT. This may explain why the recovery was good. There are a few reports that have so far evaluated the recovery of CD34+ cells in CB units for HCT [6, 8, 28]. Ogawa and co-workers [28] reported that the recovery of CB units which were used for HCT was 78% for CD34+ cells. These results are similar to the current study. These authors conclude that there may be variability in the measurement methods.

CFU assays have also been shown to correlate with time to engraftment [25]. Although CFU-GMs decreased significantly after thawing in both the groups in the current study, there was no significant difference in the recovery of CFU-GMs between the study group and the control group. These results are not consistent with other reports [13, 15, 16]. However, Harris and co-workers [12] reported the CFU-GMs to statistically decrease from 164 ± 20.8 × 105 before freezing to 62 ± 17.1 × 105 after thawing. This is consistent with the current results. In addition, previous reports focused on the effect of short-term cryopreservation, there are some reports [7, 10] that the mean recovery of CFU-GMs was >90%, and others [4, 14] that CFU-GMs were statistically decreased from after thawing. Ogawa and co-workers [28] said the CFU-GM assay has much variability because there is significant laboratory-to-laboratory and technician-to-technician variability. In addition, all the reports showing a good recovery used CB units for their studies, and reports [6, 28] that used CB units for HCT, found the CFU-GMs to be 65 ± 31.1 and 76 ± 27%, respectively. Therefore, the difference could be explained by the type of CB units. The CD34+ cells and CFU-GMs decreased remarkably in some of the current CB units after thawing in the both groups. The relationship between CD34+ cells and CFU-GM content of CB units after thawing was analyzed because there might be some problems in the examination methods. There was a positive relationship between CD34+ cells and CFU-GMs. These results suggested that there are some CB units in which the quality can be reduced to a remarkable degree after cryopreservation. Kobylka and co-workers [14] reported that two of eight long-term (15 years) cryopreserved CB units have no CFU-GM colonies, and CFU-GMs decrease remarkably in some CB units even after short-term cryopreservation. A retrospective analysis of CB HCT [29] found two patients who experienced engraftment failure after receiving CB units that produced no CFU-GM post-thaw. They concluded that CFU-GM after thaw is one of the most useful parameters when selecting CB units for HCT. Recent data suggest that the viability dye 7-AAD be applied to discriminate between dead or early apoptotic cells and living cells [30, 31]. Although no prefreeze data were available in the study group in the present study, no significant difference was observed in the viability after thawing between the study group and the control group. No reports have so far analyzed the viability of long-term cryopreserved CB units. Previous reports [4–7] focused only on the effect of cryopreservation, cryopreserved CB units had high viability after thaw. Even long-term cryopreserved CB units had high viability in the current series. This result indicates that there the length of the cryopreservation period therefore appears to have little effect on the long-term viability.

The presence of CD38 identifies already committed CD34+ cells [32], while the CD34+ CD38− phenotype identifies a UCB primitive subpopulation of hematopoietic stem cells with a higher proliferative potential in response to cytokines and a higher cloning efficiency than the same immunophenotype in adult bone marrow [17–19]. CD34+ CD38− cells are a very immature population highly enriched in SCID repopulating cells [33]. The CD34+ CD38− population was shown to be 16% in a previous report [34].

A chemotactic factor was isolated and identified as stromal cell-derived factor-1 (SDF-1), a CXC chemokine previously cloned from mouse BM stromal cells [35]. The receptor for SDF-1 was subsequently identified on human lymphocytes as CXCR4 [20], a seven-transmembrane domain protein member of the α chemokine receptor family. SDF-1 and its receptor CXCR4 are necessary for the engraftment of human CD34+ cells in the NOD/SCID model [21]. Donor CD34+ cells that express higher levels of CXCR4 are statistically correlated with earlier reconstitution of hematopoiesis after HCT [36]. A recent study by Dabusti et al. [37] demonstrated the CXCR4 expression in CB CD34+ cells to be 94.8%.

The current results showed that the number of CD34+ CD38− cells was significantly lower in the study group than in the control group, and the number of CD34+ CXCR4+ cells was significantly higher in the study group than in the control group. There is no explanation for this observation. However, the results that the mean basal percentage of CD34+ CXCR4+ cells and CD34+ CD38− cells after thawing were 81.65 ± 10.82 and 9.11 ± 4.13, respectively, for the study group, thus indicating the quality of long-term cryopreserved CB units to therefore be maintained. Timeus and co-workers [5] reported that, CD34+ CD38− cells, CXCR4 expression on CD34+ CD38− cells did not change significantly after the freeze–thaw procedures. Although the percentage of cells expressing CD38 and CXCR4 was assessed by a three-color flow cytometry analysis in this report, we assessed their expression by two-color flow cytometry analysis. We believe that analysis of the number of CD34CD38− and CD34+ CXCR4+ cells (known simply as hematopoietic progenitors) is the most important indicator of the identity of the cells.

The present results also showed that the CD34+ CD38− cells and CD34+ CXCR4+ cells did not change significantly after short-term cryopreservation.

In conclusion, the current results suggest that long-term cryopreservation does not affect the quality of CB units for transplantation, although several types of HPCs were statistically reduced. However, some units have an acceptable cell number but very low or even no clonogenic potential after thaw and these units would not be selected for transplantation. These results suggest that long-term cryopreserved CB units can be used for HCT, but cord blood banks should prepare and supply small aliquots of cryopreserved samples from several candidate units at least several weeks before HCT to allow for the analysis of the CD34+ cell count, viability, and CFU-GM assay in order to select the best units possible.