Sodium bicarbonate, ammonium chloride, sodium ethylenediaminetetracetic acid (EDTA), 2,3-(4-benzoyl) benzoyl adenosine-5-triphosphate (BzATP) and adenosine-5-triphosphate were purchased from Sigma (St. Louis, MO). Trypan blue, phosphate buffered saline (PBS) without calcium and magnesium, fetal bovine serum (FBS) and Lymphocyte Separation Medium (density = 1.077 g/mL) were purchased from Fisher Scientific (Atlanta, GA). MethoCult® Classic was purchased from Stem Cell Technologies (Seattle, WA). Hydroxyethyl starch (Viastarch, HES) was purchased from Fresenius (Linz, Austria). The HES mean molecular weight, water content, and NaCl content were 259,700 Da, 4.64% and 0.11%, respectively[67]. Bovine serum albumin fraction-V 30% (BSA) was purchased from Gemini BioProducts (West Sacramento, CA). Low endotoxin glucose and trehalose were purchased from Pfanstiehl Laboratories (Waukegan, Illinois). Fluorescein-5-isothiocyanate isomer I (FITC) and propidium iodide (PI) was purchased from Molecular Probes (Eugene, OR).

8.3 g ammonium chloride, 1.0 g sodium bicarbonate, and 0.04 g sodium EDTA were dissolved in 1 L ultra pure sterile water. The solution was filter through 0.22 µM filter and stored at 4°C. Red blood cell (RBC) lysis buffer was warmed to 37°C before use.

All protocols were approved by the University of Colorado Health Sciences Center Internal Review Board (number 01-809). After informed consent, human cord bloods were obtained using sterile tubes with heparin from newborn umbilical cords following normal vaginal deliveries. Patient consent was written. A translator was available if required or a copy in the native language of the patient was made available as requested. Only UCB that tested negative for HIV were used in this study. Samples were processed within 12 hr of collection.

Cord blood specimens were diluted with an equal volume of RBC lysis buffer and transferred to a 50 mL conical tube. Each specimen tube was then rinsed with 2 mL of RBC lysis buffer and the rinse was added to specimens. Specimens were briefly vortexed and then incubated for 3 minutes at room temperature. Cells were pelleted by centrifugation at 1500 rpm for 10 minutes. The supernatant was decanted and cells were resuspended in PBS/1%BSA equivalent the original specimen volume.

Light-density mononuclear cells (MNC) were isolated by layering cell suspension carefully onto Lymphocyte Separation Medium (density = 1.077 g/mL, 2 volumes of diluted blood to 1 volume of separation medium) and centrifuging at 1500 rpm for 30 minutes at room temperature. MNC (monocytes and lymphocytes) have a lower buoyant density than the erythrocytes and the polymorphonuclear (PMN) leukocytes (granulocytes). Since the majority of MNC have densities below 1.077 g/mL, they can be isolated at the sample/medium interface. The interphase layer containing MNC was collected and washed twice with PBS/1%BSA. After 2 cycles of washing and centrifugation, cell pellets were resuspended in 5 mL of PBS/1%BSA. Cells were counted using a Coulter Counter, and cell viability was assessed by Trypan blue exclusion. No further cell identification or purification was performed.

MNC cells were cultured with MethoCult® Classic. Cytokines in this medium only support cells that give rise to CFU-E, BFU-E, CFU-GM, and CFU-GEMM. All references to these cells will be identified as HPC in the text. (See CFU assay for details)

Expression of P2Z receptor was measured by permeabilizing HPC with BD Cytofix/Cytoperm kit (BD Biosciences Pharmingen, San Diego, CA) using manufacturer's instructions, staining with anti-P2Z/P2X7 receptor (Calbiochem, La Jolla, CA) polyclonal antibody from rabbit and anti-Rabbit-PE secondary antibody developed in goat (Sigma, St. Louis, MO). Briefly, 1×10⁶ cells were washed twice with 1 mL of staining buffer (PBS/1% FBS/5% (w/v) sodium azide). To fix cells, HPC were pelleted by centrifugation (1100 rpm for 8 minutes) and then resuspended in 250 µL of BD Cytofix/Cytoperm solution for 30 minutes at 4°C. HPC were washed twice in 1 ml of 1X BD Perm/Wash solution. Cells were then resuspended in 50 µL of BD Perm/Wash solution containing anti-P2Z/P2X7 (1∶100) for 30 minutes at 4°C. Cells were washed twice in 1 mL of 1X BD Perm/Wash solution. Cells were then resuspended in 50 µL of 1X BD Perm/Wash solution containing 20 µL of anti-Rabbit-PE for 30 minutes at 4°C in the dark. Cells were washed twice in 1 mL staining buffer. Cells were then resuspended in 300 µL of staining buffer and stored at 4°C in the dark until flow cytometric analysis. Flow cytometric analysis was performed a Coulter Epics XL equipped with Summit v3.1 and v4 software.

HPC were incubated in poration buffer containing 100 µM BzATP or 5 mM ATP and 200 mM trehalose at a concentration of 1×10⁶ cells in 1 mL. Poration buffer is composed of 5 mM glucose, 1X essential amino acids, 1X nonessential amino acids and 1X Vitamin Stock in 10 mM K₂HPO₄/KH₂PO⁴ titrated to pH 7.45±0.03 (modified from Menze[68]. Cells were incubated in the reaction mixture for 0–90 min to accommodate trehalose loading. Pores were closed with the addition of 1 mM MgSO₄ and 10-fold dilution of BzATP or ATP using RPMI 1640.

Poration efficiency for HPC was assessed by FITC/PI staining using flow cytometry as previously described[48], [69], [70], [71], [72]. A stock solution of FITC for permeabilization studies was prepared by dissolving 1 mg FITC in 1 mL of DMSO, with further dilution to 0.1 mg/mL using PBS without calcium and magnesium. Briefly, porated cells were loaded with trehalose containing 100 µL of 0.1 mg/mL FITC. Pores were closed and excess dye was removed by washing twice with PBS/1%FBS. Approximately 1×10⁵ porated cells were removed and suspended in 500 µL of PBS/1%FBS. Cells were treated with 10 µL of 1.0 mg/mL PI, a membrane-impermeant dye, for 4 min prior to analysis. PI labels the nucleic acids of membrane-compromised cells. Cells which stained positive for FITC and negative for PI were gated and considered to be both viable and porated, whereas cells stained with PI were gated and considered to be dead. All reported values were normalized to total cell count. Flow cytometric analysis was performed on a Coulter Epics XL equipped with System II software.

After trehalose loading, HPC were lyophilized in a formulation containing 6.8% trehalose/2% HES/5% HSA (w/v). Approximately 1 mL of sample solution was transferred to 5-mL flat bottom borosilicate lyophilization vials (West Pharmaceutical, Lititz, PA), and the vials were placed on the shelf of an FTS DuraStop MP/DuraDry MP freeze-dryer (Stone Ridge, NY). The shelf was cooled from room temperature to −45°C at a cooling rate of 2.5°C/min until product temperature reached −45°C. Samples were maintained −45°C for 180 minutes. Primary drying was initiated by raising the shelf temperature from −45°C to −35°C at 2.5°C/ minute, reducing the chamber pressure to 80 mTorr and maintaining the shelf temperature at −35°C for 2160 minutes. To determine completion of primary drying, the step was monitored using the chamber pressure rise test and equilibration of the product temperature to the shelf temperature, as described previously[69], [70]. For secondary drying, samples were warmed by increasing the shelf temperature at a rate of 0.2°C/minute to the designated secondary drying temperature to be tested and maintaining the shelf at this temperature for 360 minutes. To determine where clonogenic and differentiation potential is lost during lyophilization, vials were removed at the end of primary drying (−35°C) and during secondary drying (at 5°, 10°, 15°, and 20°C) from the freeze-drying chamber into an antechamber using a retractable arm. The extracted vials were stoppered in the antechamber under the same pressure as the chamber (∼80 mTorr). At the end of the cycle, the trays were raised until the rubber stoppers were pushed down to seal the vials under vacuum before removing them from the unit. All samples were stored at −80°C until analysis.

Lyophilized samples were rapidly reconstituted using high purity water. An aliquot of ∼950 µL water was drawn into a 1 mL serological pipette, and the water was released down the sides of lyophilization vials. To ensure homogeneous resuspension of the cells, the solution containing the cells was then gently drawn into a 1 mL serological pipette and released down the sides of the vials 5-times. Approximately 2 mL of complete medium was added to each vial, and cells were transferred to 10 cm² flat bottom tube flasks (MidWest Scientific, St. Louis, MO), and incubated for 2 hr at 37°C in humidified air containing 5% CO₂ before plating cells in methylcellulose.

Progenitor cell proliferation of control, frozen and thawed or freeze dried, and rehydrated cells was quantified by an in vitro method using MethoCult® Classic, a semi-solid methylcellulose system containing stem cell factor (SCF), GM-CSF, interleukin 3 (IL-3), and erythropoietin (StemCell Technologies, Vancouver, BC, Canada) and by measuring clonogenic activity via CFU. The assay was performed with light-density MNC as described above. Cytokines in this medium only support cells that give rise to CFU-E, BFU-E, CFU-GM, and CFU-GEMM. The culture plates used in these studies do not support the growth of anchorage-dependent cells. All references to these cells will be identified as HPC in this text.

Unfrozen control cells were suspended at a concentration of 1×10⁶ cells/mL and 100 µL of cells were removed for culture in 1% methylcellulose. (∼1×10⁵ as determined by both a Coulter counter and trypan blue exclusion). The 100 µL of cells was added to 5 mL of 1% methylcellulose (∼20,000 cells/mL). The clonogenic activity of unfrozen control cells was standardized in this manner and considered to be 100% clonogenic potential. Colony identification was based on color and morphology. Only multi-lineage myeloid colonies were scored (i.e., CFU-GEMM, CFU-GM, or BFU-E).

For freeze-thawing experiments, cell samples were removed at the end of the freezing step of lyophilization and stored at −80°C until analysis. Frozen samples were rapidly thawed in a 37°C water bath. For experiments of cells that were freeze-dried, samples were stored at −80°C until analysis. Dried samples were rehydrated as described in the rehydration section above. Following thawing or rehydration, approximately 2 mL of complete medium was added to each sample. Cells were mixed by pipetting with a 5-mL serological pipette 5-times to ensure homogeneous suspension. Cells were then transferred to 10 cm² flat bottom tube flasks (MidWest Scientific, St. Louis, MO) and incubated for 2 hr at 37°C in humidified air containing 5% CO₂. Following incubation at 37°C, approximately 50 µL of cell suspension was transferred to a 15-mL conical tube with 6 mL of 1% methylcellulose containing growth factors.

Four individual culture plates were scored for each group. All samples were cultured at 37°C in humidified air containing 5% CO₂ for 14 days, after which colonies were enumerated on an inverted Nikon microscope (1 colony ≥20 cells).

All samples were analyzed on a Perkin-Elmer Diamond DSC equipped with an Intercooler II and the data were analyzed on Pyris 7.0 software. The DSC was calibrated using indium (for temperature and enthalpy of melting). Vials containing dried samples used for T_g measurements were placed in a nitrogen-purged dry glove box to avoid possible hydration resulting from exposure to environment and hermetically sealed in aluminum pans. Approximately 10 mg of solid was measured each measurement. Lyophilized samples were heated from 0 to 200°C at100°C/minute, maintained at 200°C for 5 minutes, then cooled to 0°C and reheated again to 250°C at the same rates to measure T_g in the second scan. Values are reported as the onset temperature of the glass transition.

Residual water content of lyophilized samples was determined by Karl Fischer titration using a Mettler DL37 coulometric moisture analyzer (Hightstown, NJ) containing pyridine free vessel solutions (Photovolt Instruments, Inc., St. Louis Park, MN). Samples were prepared in a nitrogen-purged dry glove box to avoid possible hydration resulting from exposure to environment. Lyophilized samples were dissolved in 1 mL of dimethyl formamide. The samples were sonicated for 15–25 minutes to ensure complete solubility prior to analysis. Using a gas-tight syringe, 250 µL were withdrawn through the rubber stopper and moisture analysis was performed as previously described⁷⁰. Values are reported as mass percent water (gH₂O/100 g solid).

After lyophilization, stoppered vials containing dehydrated samples were sealed with aluminum caps and stored in an incubator at 25°C (so-called “controlled room temperature”) in the dark for 4 weeks. At the end of the storage period, samples were rehydrated and analyzed as described above.

Statistical analyses were performed on CFU assay results using paired-student t test (SigmaPlot 2001) and ANOVA using Dennett's test, which controls overall error rate for comparing all other means with a designated control group's mean (JMP v6-8). Differences among groups were considered significant when p value was less than 0.05. Data reported as the mean ± the standard deviation of three independent experiments each with quadruplicate samples.

We investigated the expression of P2Z receptor on HPC using indirect staining and flow cytometric analysis (Figure 1). Staining with secondary antibody alone marked the background fluorescence level by non-specific binding of this antibody to dead or other cells (1.6%±0.3, Figure 1A). Fluorescence levels above the background were considered specific for P2Z and shown to be approximately 58%±4% (Figure 1B) compared to secondary antibody alone (Overlay of A and B as shown in Figure 1C).

To assess poration efficiency, HPC were permeabilized by exposing them to either 5 mM ATP or 100 µM BzATP, which opens the P2Z cell surface receptor, in medium containing 200 mM trehalose and 0.1 mg/mL FITC. FITC was employed as an indicator of poration because it is normally impermeable to cellular membranes and has a similar molecular mass (MW 389) to that of trehalose (MW 342). In the absence of porating agent, less than 1% of cells were positive for the presence of dye. There was a gradual increase in the number of cells positive dye when permeabilized in the presence of ATP, reaching a maximum of ∼31% after 75 minutes (Figure 1D). Permeabilization in the presence of BzATP (a more potent agonist for the P2Z receptor than ATP)[57], [60], resulted in a steady increase in the number of cells positive for dye, with 58±4% of cells positive for dye after 90 minutes. The membrane integrity of cells was also monitored during poration using PI. There was a gradual loss in membrane integrity after 90 minutes of permeabilization using BzATP as the porating agent, with 68±3% of cells remaining intact whereas 90±4.5% of cells remained intact when the porating agent was ATP. These results indicate that HPC can be permeabilized via P2Z receptors for up to 60 minutes using BzATP and for up to 90 minutes using ATP without a significant decrease in cell viability.

Cells must first survive freezing before freeze-drying can be attempted. To demonstrate the use of trehalose as an appropriate intracellular cryoprotectant, we permeabilized cells in medium containing 100 µM BzATP and 200 mM trehalose for 75 minutes. Trehalose loaded cells were suspended in a formulation containing 6.8% trehalose/2% HES/5%HSA and then slowly cooled (1°C/minute) to −80°C followed by storage at −80°C for one week. Cryopreserved cells were rapidly thawed in a 37°C water bath and tested for clonogenic potential. Control groups for these experiments were freshly isolated cells and cells cryopreserved with 10% DMSO (1.4 M). As seen in Figure 2, cells cryopreserved trehalose produced comparable CFU to cells cryopreserved with DMSO. These data document that primary HPC loaded with trehalose can survive freeze-thawing when cryopreserved in medium containing 6.8% extracellular trehalose/2% HES/5% HSA. Furthermore, the frequency of CFU-GM, BFU-E, and CFU-GEMM is comparable to results obtained with fresh cells and those cryopreserved with DMSO.

A representative plot of the temperature profile during the freeze drying cycle shows when vials were removed during lyophilization without interrupting the cycle to assess the effect of each step on the differentiation and clonogenic potential of the cells (Figure 3). Trehalose loaded cells were lyophilized in a formulation containing 6.8% trehalose/2% HES/5% HSA (w/v). Samples were removed during processing using an extractable arm into an antechamber (See methods) and stored at −80°C until analysis. There were three main types of colonies that were enumerated: CFU-GM, BFU-E, and CFU-GEMM (Figure 4). There was no difference in the number of CFU produced by control groups and cells that were removed at the end of the freezing step and thawed (Figure 5A). There was a gradual decrease in the number of CFU-GM and BFU-E for samples removed at the end of primary drying and during secondary drying; however, there were no differences in the number of CFU-GEMM (∼82%). An ANOVA was performed on the results and the means were compared using Dennett's test, which controls overall error rate for comparing all other means with a designated control group's mean. CFU produced by thawed cells removed at the end of the freezing step and rehydrated cells removed at the end of primary drying were not statistically different from that for control cells. There was a gradual decrease in the number of CFU-GM and BFU-E produced by cells removed at increasing temperatures during secondary drying, which were statistically different from control groups. In marked contrast, cells produced approximately 82% CFU-GEMM at all secondary drying shelf temperatures compared to control cells. There were no statistical differences in the means between CFU-GEMM produced by rehydrated cells and control cells.

In addition to measuring the differentiation and clonogenic potential of lyophilized cells, we also quantified the Tg and residual water content of the dry cake (Figure 5B). Not surprisingly, the Tg of samples increased as water was removed from samples. The final Tg of samples processed at a secondary drying temperature of 20°C was 197±3°C.

At the end of the primary drying step, there was approximately 39.5±0.8 gH₂O/100 g solid. As expected, the residual water content decreased as secondary drying continued, which resulted in a concomitant decrease in cells' differentiation and clonogenic potential. This observation is consistent with results reported by Toner and colleagues who suggest that there may be an optimal residual water content required for stable dry cells[31].

To determine storage stability of lyophilized HPC, cells dried at a final secondary drying shelf temperature of 15°C were stored for 4 weeks at 25°C in the dark (Table 1). Cells reconstituted immediately after lyophilization produced approximately 40% CFU-GM, 40% BFU-E, and ∼82% CFU-GEMM relative to control. Cells reconstituted on after 28 days of storage at room temperature produced approximately 35% CFU-GM 26% BFU-E, and 82% CFU-GEMM compared to control. There was an initial decrease in the frequency of CFU-GM and BFU-E upon reconstitution immediately after lyophilization, but there was no further change in clonogenic output after 4 weeks of storage. There was not a significant difference in the frequency of CFU-GEMM by cells stored for 4 weeks at 25°C and untreated control cells. These results indicate that the initial loss observed during secondary drying is the main loss of cells during freeze-drying and storage in the lyophilized formulation.