Cryobiology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Cryobiology journal homepage: www.elsevier.com/locate/ycryo

Osmotic parameters of red blood cells from umbilical cord blood q Mariia Zhurova a,b, Locksley E. McGann a, Jason P. Acker a,b,⇑ a b

Department of Laboratory Medicine and Pathology, 8249-114 Street, Edmonton, AB T6G 2R8, Canada Research and Development, Canadian Blood Services, 8249-114 Street, Edmonton, AB T6G 2R8, Canada

a r t i c l e

i n f o

Article history: Received 8 March 2013 Accepted 1 April 2014 Available online xxxx Keywords: Erythrocyte Cryopreservation Permeability Modeling Biobanking Transfusion medicine Cryoprotectant Glycerol

a b s t r a c t The transfusion of red blood cells from umbilical cord blood (cord RBCs) is gathering significant interest for the treatment of fetal and neonatal anemia, due to its high content of fetal hemoglobin as well as numerous other potential benefits to fetuses and neonates. However, in order to establish a stable supply of cord RBCs for clinical use, a cryopreservation method must be developed. This, in turn, requires knowledge of the osmotic parameters of cord RBCs. Thus, the objective of this study was to characterize the osmotic parameters of cord RBCs: osmotically inactive fraction (b), hydraulic conductivity (Lp), permeability to cryoprotectant glycerol (Pglycerol), and corresponding Arrhenius activation energies (Ea). For Lp and Pglycerol determination, RBCs were analyzed using a stopped-flow system to monitor osmoticallyinduced RBC volume changes via intrinsic RBC hemoglobin fluorescence. Lp and Pglycerol were characterized at 4 °C, 20 °C, and 35 °C using Jacobs and Stewart equations with the Ea calculated from the Arrhenius plot. Results indicate that cord RBCs have a larger osmotically inactive fraction compared to adult RBCs. Hydraulic conductivity and osmotic permeability to glycerol of cord RBCs differed compared to those of adult RBCs with the differences dependent on experimental conditions, such as temperature and osmolality. Compared to adult RBCs, cord RBCs had a higher Ea for Lp and a lower Ea for Pglycerol. This information regarding osmotic parameters will be used in future work to develop a protocol for cryopreserving cord RBCs. Ó 2014 Elsevier Inc. All rights reserved.

Introduction The transfusion of RBCs derived from umbilical cord blood for the treatment of neonatal anemia has gained significant interest recently [4,5,12,13,19,25,28,30,62,63]. Fetal and neonatal anemias are among the most serious complications of pregnancy and postnatal development. The most commonly used treatment is transfusion of red blood cells (RBCs), either intrauterine or intravenous [24,36,64] to help replace the lost RBCs of the fetus or neonate. To date, RBCs used in intrauterine and neonatal (intravenous) Abbreviations: RBC, red blood cell; HbF, fetal hemoglobin; b, osmotically inactive fraction; Lp, hydraulic conductivity; Ps, permeability to solutes; Pglycerol, permeability to glycerol; Ea, Arrhenius activation energy; CPD, citrate–phosphate– dextrose; SAGM, Saline–adenine–glucose–mannitol; NaCl, sodium chloride; PBS, phosphate-buffered saline; V, Volt; V/V0, equilibrium relative RBC volume; F/F0, equilibrium relative RBC fluorescence intensity; p0/p, inverse relative osmolality; Pf, osmotic water permeability. q Statement of funding: Funding for this study was provided by the Canadian Institutes of Health Research. Graduate Fellowship to Mariia Zhurova was provided by the Canadian Blood Services. ⇑ Corresponding author at: Canadian Blood Services, 8249-114 Street, Edmonton, AB T6G 2R8, Canada. Fax: +1 (780) 702 8621. E-mail address: [email protected] (J.P. Acker).

transfusions are derived from adult donors [24,36,45,64]. Adult RBCs are different from those present in the blood of a fetus or neonate [11,27,39,48,50,51]. Perhaps not surprisingly, the practice of administering adult RBC transfusions to premature infants has been associated with a number of complications, such as retrolental fibroplasia [14,23,38] and bronchopulmonary dysplasia [15,17,32], usually caused by the delivery of unnecessarily high amounts of oxygen to tissues. Neonatal RBCs obtained from umbilical cord blood (cord RBCs) may offer a superior alternative for intrauterine and neonatal transfusions [6,22]. Cord RBCs are usually discarded during the isolation of stem cells from cord blood [8,52,61]. Due to the high concentration of fetal hemoglobin (HbF), which is practically absent in adult RBCs, cord RBCs have a potential to deliver a physiologically suitable amount of oxygen to fetal and neonatal tissues upon transfusion [37]. A number of studies have demonstrated that transfusions of autologous cord RBCs are both safe and effective for the treatment of anemic neonates [4,12,13,19,25,63]. However, unlike RBCs from adult blood, cord RBCs deteriorate quickly during traditional storage at 1–6 °C [19,29] and thus would benefit from low temperature preservation. Despite several reports describing some of the effects of cryopreservation on cord RBCs

http://dx.doi.org/10.1016/j.cryobiol.2014.04.002 0011-2240/Ó 2014 Elsevier Inc. All rights reserved.

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

2

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

[9,70], an effective cryopreservation method for cord RBCs still needs to be developed. To design effective cryopreservation procedures for cord RBCs, it is critical to know the osmotic parameters of these cells [18,31,35,55]. A cell undergoes a number of osmotic changes during cryopreservation. Addition and removal of cryoprotectant, as well as freezing and thawing, cause changes in solute concentration inside and outside the cell and induce flux of water and permeating solutes across the cell membrane. Such changes in intra- and extra-cellular osmolality lead to corresponding changes in cell volume. A number of osmotic parameters define the movement of water and solutes across the cell membrane. Hydraulic conductivity (Lp) and osmotic permeability to solutes (Ps) describe the rate at which these substances cross the cell membrane and are typically determined by measuring the rate at which cell volume changes when placed in an anisotonic environment. The osmotically inactive fraction (b) is the fraction of the cell volume that does not participate in the osmotic response of the cell. The Arrhenius activation energy (Ea) describes the temperature dependence of the membrane’s permeability to water and solutes [56]. These osmotic parameters can be used in mathematical models to predict an optimized cryopreservation protocol for the cells of interest [41,42,57]. It has been reported that, compared to adult RBCs, cord RBCs are less permeable to water [3,60], have a higher activation energy for osmotic water permeability [3], and are less permeable to some solutes, particularly the common cryoprotectant glycerol [46]. However, reports on cord RBC osmotic parameters are very limited at present. Moreover, with measurements made using different experimental techniques and under different experimental conditions, the absolute values of cord RBC osmotic parameters differ significantly between reports and, therefore, cannot be relied upon for use in mathematical modeling. The objective of this study is to measure and compare the osmotic parameters of adult and cord RBCs, such as osmotically inactive fraction, permeability to water and glycerol, and Arrhenius activation energies for these processes.

Materials and methods Source of red blood cells Two sources of RBCs were used in this study: RBCs from peripheral blood of adult donors (adult RBCs), and RBCs from umbilical cord blood (cord RBCs). The Canadian Blood Services Network Centre for Applied Development in Vancouver provided the adult RBCs. RBCs were leukocyte-reduced, stored in CPD (citrate–phosphate–dextrose) anticoagulant and SAGM (saline–adenine–glucose–mannitol) preservative at 1–6 °C, and were used in experiments within 15 days of collection. The hematocrit of adult RBCs was standardized to 60 ± 2%, if necessary, by the removal of supernatant or addition of saline. The Alberta Cord Blood Bank supplied cord RBCs, a waste product after stem cell isolation from umbilical cord blood. Cord blood collected from a placenta was stored at room temperature for up to 38 h prior to stem cell isolation (previously shown not to cause a decrease in RBC quality [68]). The leftover cord RBC product was washed 3 times with saline using centrifugation at 2200g at 4 °C for 5 min, to remove any residual pentastarch used in the stem cell isolation process [66]. The hematocrit of cord RBCs was then adjusted to 60 ± 2% by the addition of saline to the RBC pellet. Cord RBCs were stored at 1–6 °C and used in experiments within 24 h of isolation from cord blood. Ethics approval for the study was obtained from the University of Alberta Health Research Ethics Board (Biomedical Panel) and Canadian Blood Services Research Ethics Board.

Experimental solutions Sodium chloride (NaCl) solutions were prepared by diluting 12% (w/v) NaCl stock solution (Baxter, Deerfield, IL, USA) with distilled water to yield final concentrations of 0.68%, 0.9%, 1.6%, and 3.5% (w/v). Phosphate-buffered saline (PBS) solutions were prepared by diluting 10 PBS solution (Calbiochem, Gibbstown, New Jersey) with distilled water to 0.5, 1, 3, 5, and 7 PBS. Lastly, a 5% (w/v) glycerol in 1 PBS solution was prepared by diluting 50 g glycerol (99.5+%, Sigma Aldrich, Inc., St. Louis, MO, USA) and 100 mL 10 PBS with distilled water to 1 L. The osmolality of the experimental solutions was measured using a freezing-point depression osmometer Osmette (Precision Systems Inc., Natick, Massachusetts). Prior to each experimental run, the osmometer was verified through quality control checks using both a 290 mmol/kg Opti-Mole standard (Wescor, Inc., Logan, Utah) and a 1500 mOsm/kg standard (Precision Systems Inc., Natick, Massachusetts). Hemolysis of RBCs in experimental solutions RBC hemolysis (membrane damage) was measured in experimental PBS and glycerol solutions. 50 lL of RBCs were pipetted into 1 mL of the experimental solution and allowed to equilibrate at room temperature for approximately 5 min. RBC hemolysis was determined by spectrophotometric measurement of total and supernatant cyanmethemoglobin according to Drabkin’s method [1,71]. The hematocrit of the RBC sample was required for the calculation and, therefore, was determined, using a microhematocrit centrifuge (Hettich, Tuttlingen, Germany), as the ratio of the volume occupied by packed RBCs to the volume of a whole RBC sample. Controls for total hemoglobin were prepared from Stanbio Tri-Level Hemoglobin controls (Stanbio Laboratory, Boerne, TX, USA). Measurement of RBC volume kinetics on stopped-flow Changes of RBC volume with exposure to solutions of different osmolalities were determined indirectly by monitoring changes in intrinsic hemoglobin fluorescence intensity, as described previously [69]. Using a SX20 stopped-flow reaction analyser (Applied Photophysics, Ltd., Leatherhead, UK), the RBC suspension was rapidly mixed with an equal volume of anisotonic experimental solution to induce osmotically-driven changes of RBC volume. RBC fluorescence intensity, which is directly related to RBC volume, was then recorded as a function of time after mixing. The RBC suspension for stopped-flow experiments was prepared by adding 20 lL of RBCs to 1 mL of 1 PBS (final osmolality of 287 mOsm/kg). To determine hydraulic conductivity, RBCs were exposed to 0.75, 2, 3, and 4 PBS solution (final osmolalities of 214, 562, 832, and 1112 mOsm/kg, respectively) – achieved by mixing the RBC suspension (in 1 PBS) rapidly in a 1:1 ratio with 0.5, 3, 5, and 7 PBS, respectively. One thousand data points were collected during the 10 s period immediately following mixing. As a control, RBCs exposed to 1 PBS (osmotic equilibrium conditions) were assessed. To determine glycerol permeability, RBCs were exposed to 2.5% (w/v) glycerol (final osmolality of 578 mOsm/kg) – achieved by rapid 1:1 mixing with 5% (w/v) glycerol in 1 PBS solution. Two thousand data points were collected during the 120 s period immediately following mixing: the time required for the complete equilibration of glycerol across the RBC membrane. As a control, RBCs were mixed at a 1:1 ratio with 1 PBS (establishing osmotic equilibrium conditions) and fluorescence was measured as a function of time. The background fluorescence of buffer solutions (without RBCs) was also measured. All samples were measured

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

3

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

in triplicate. Prior to each experimental run, the stopped-flow circuit was flushed with a solution having the same osmolality as the one about to be tested. This was done to prevent cross-contamination between experimental solutions and, therefore, to assure accurate osmolality during each experimental run. SX20 instrument settings The sample cell was illuminated at 280 nm, and emission was measured at either 314 nm (adult RBCs) or 315 nm (cord RBCs). Excitation and emission slit widths were set to 3 mm (equivalent to a bandwidth of 13.95 nm). The total stopped-flow drive volume was set to approximately 120 lL. The 20 lL optical cell had a 10 mm pathlength and 1 ms dead time (during which mixing occurred). Fluorescence was expressed in Volts (V). Data were acquired using ProData SX software (Applied Photophysics, Ltd., Leatherhead, UK). The stopped-flow system was calibrated in fluorescence emission mode using a quality control test (chemical reaction of binding of 8-anilino-1-naphthalene sulphonate to the protein, bovine serum albumin) according to the manufacturer’s recommendations. To determine the Arrhenius activation energy, the kinetics of osmotically-induced changes of RBC volume was measured at 3 different temperatures (mean ± SEM): 3.8 ± 0.1 °C (3.5–4.0 °C range), 19.4 ± 0.1 °C (18.6–21.0 °C range), and 35.2 ± 0.1 °C (34.8– 35.6 °C range) for adult RBCs and 3.7 ± 0.1 °C (3.5–3.9 °C range), 20.4 ± 0.1 °C (19.7–21.2 °C range), and 35.3 ± 0.1 °C (35.1–35.7 °C range) for cord RBCs. The temperature of the stopped-flow system was controlled through an attached water-filled circulator (CH/P temperature control system, Forma Scientific, Marietta, Ohio). Syringes containing RBC samples and experimental solutions were equilibrated in the water bath, set to experimental temperature, for a minimum of 5 min before each run. Fluorescence to cell volume conversion is described in the Appendix. Determination of osmotic parameters Osmotically inactive fraction The osmotically inactive fraction of adult RBCs was taken from the literature (value of 0.51 ± 0.02) [53]. The osmotically inactive fraction of cord RBCs was determined from measured equilibrium volumes, using Boyle-van’t-Hoff equation:

V p0 ¼ ð1  bÞ þ b V0 p

ð1Þ

where V is the equilibrium cell volume at the experimental osmolality (p), V0 is the isotonic cell volume at isotonic osmolality (p0), and b is the osmotically inactive fraction of the cell volume. First, equilibrium volumes of cord RBCs were measured at different osmolalities (all measurements were taken at room temperature). Boyle-van’t-Hoff plot was then created, which displays the equilibrium relative RBC volume (VV0 ) as a function of inverse relative osmolality (pp0 ). The osmotically inactive fraction of cord RBC volume was determined from y-intercept of the linear regression line [49]. Equilibrium volumes of cord RBCs in NaCl solutions were determined using a Coulter Electronic Particle Counter. RBCs were diluted in 10 mL of 0.68%, 0.9%, 1.6%, and 3.5% (w/v) NaCl to approximately 20,000 cells/mL and were allowed to equilibrate at room temperature for at least 5 min. Using solutions at each NaCl concentration, 5 lm latex beads (Beckman Coulter, Inc., Fullerton, CA, USA) with known bead volume were used to determine a calibration factor to convert the instrument output to RBC volumes [56]. Electrical current pulses, proportional to cell volumes, were measured as RBCs passed through the Counter’s 50 lm aperture. Measurements were made in triplicate for each experimental solution.

Hydraulic conductivity and permeability to glycerol The following equation by Jacobs and Stewart describes the rate of cell volume change when the cell is exposed to anisotonic conditions and was used to calculate hydraulic conductivity [18,26,56]:

  dV ¼ Lp ART ðC is  C es Þ þ ðC ii  C ei Þ dt

ð2Þ

where V is the cell volume (lm3), t is the time (min), Lp is the hydraulic conductivity (lm/min/atm), A is the cell surface area, either constant at a specified value, or the area of a sphere having the volume of the cell (lm2), R is the gas constant (L-atm/mol/K), T is the absolute temperature (K), C is , C es are the intracellular, extracellular concentration of permeant solute (mol/kg), C ii , C ei are the intracellular, extracellular concentration of impermeant solute (mol/kg). Glycerol movement across the cell membrane, as a function of time, was calculated using the following equation [18,56]:

dS ¼ Ps AðC es  C is Þ dt

ð3Þ

where S is the amount of glycerol (moles), t is the time (min), Ps is the glycerol permeability (lm/min), A is the cell surface area, either constant at a specified value, or the area of a sphere having the volume of the cell (lm3), C is , C es are the intracellular, extracellular concentration of glycerol (mol/kg). Experimental data on the rate of RBC volume change in PBS and glycerol solutions were fit to Eqs. (2) and (3) using the least squared method in Excel Solver. The following assumptions were made: solutions are dilute, the density of water is 1 kg/m3, cell surface area is calculated from the volume of the cell, and the reflection coefficient is zero (meaning there is no interaction between water and glycerol transport). Arrhenius activation energies The Arrhenius activation energy (Ea) describes the temperature dependence of hydraulic conductivity and solute permeability. Ea was determined from the slope of the plot of the natural logarithm of Lp or Ps is an inverse function of temperature [18,56]:

Lp or Ps ¼ k  exp

  Ea RT

ð4Þ

where Lp is the hydraulic conductivity (lm/min/atm), Ps is the solute permeability (lm/min), k is the fitting constant, R is the gas constant (l-atm/mol/K), Ea is the activation energy for Lp or Ps (kcal/ mol), and T is the absolute temperature (k). Comparison between Lp (lm/min/atm) and Pf (cm/s) The majority of published papers report values for RBC osmotic water permeability as Pf (cm/s). Pf can be determined from the following equation [67]:

dV=V 0 ðtÞ ¼ Pf  SAV  MVW  dt



C inðt¼0Þ =

  V ðtÞ  C out V0

ð5Þ

where V/V0 (t) is relative volume of the cell as a function of time, Pf is osmotic water permeability (cm/s), SAV is the cell surface area to volume ratio, MVW is the molar volume of water (18 cm3/mol), Cin(t=0) is the initial intravesicular osmolality (osm/kg), and Cout is the extravesicular osmolality (osm/kg). Combining Eqs. (2) and (5), Pf (cm/s) can be converted to Lp (lm/min/atm) using the equation below:

Pf ¼ Lp  ð273:15 þ  CÞ  7:5926  106

ð6Þ

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

4

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

To evaluate whether the method of analysis (Lp vs. Pf) impacted the results, a small subset of the RBC volume kinetics data was analyzed using both Eqs. (2) and (5). Lp and Pf were determined for one adult RBC sample (run in triplicate) at three experimental temperatures (4.0, 19.0, and 35.2 °C). To determine Pf, experimental data on the kinetics of RBC volume change in PBS solutions was fit to Eq. (5) using the least squared method in Excel Solver. Lp was calculated using Eq. (2) and converted to Pf using Eq. (6).

Statistical analysis Differences between experimental groups were analyzed using the Mann–Whitney non-parametric test, Student’s T-test and oneway analysis of variance (ANOVA). The significance level was set to 0.05. The strength of the relationships between RBC hemolysis and extracellular osmolality, hydraulic conductivity and experimental temperature, hydraulic conductivity and extracellular osmolality, and glycerol permeability and experimental temperature were each investigated using correlation analysis (Pearson’s correlation coefficients were calculated). Statistical analysis was performed using IBM SPSS Statistics software (version 19.0, SPSS, Inc., IBM Corporation, Armonk, New York). The number of adult and cord samples varied for different tests and is reported in the legend for each Table or Figure. Samples were measured in triplicate for all tests, except hemolysis (measured once).

Results Integrity of red cells in experimental solutions Upon equilibration in 0.75–4 PBS solutions, the range in hemolysis was 0.53–0.96% for adult RBCs and 0.83–1.5% for cord RBCs, however, statistically, no difference was found between the level of hemolysis of cord and adult RBCs in PBS solutions. Hemolysis significantly increased with the osmolality of the solution for cord RBCs (r = 0.580, p = 0.023), however not for adult RBCs (r = 0.326, p = 0.255). In 2.5% (w/v) glycerol solution, adult RBCs showed lower hemolysis compared to cord RBCs (0.44 ± 0.02% vs. 0.93 ± 0.17%, p = 0.05) (Table 1).

Osmotic parameters of adult and cord RBCs Osmotically inactive fraction of cord RBCs The average isotonic cell volume was 93.8 ± 1.1 lm3 (mean ± SEM) for adult RBCs and 115.1 ± 1.5 lm3 (mean ± SEM) for cord RBCs as determined using the Coulter AcT Series Analyzer. The Boyle van’t Hoff plot, created using measured equilibrium volumes of cord RBCs at different osmolalities, is shown in Fig. 1. The osmotically inactive fraction of cord RBCs determined from this plot was 0.56 ± 0.04, while the slope of the linear regression line was 0.51 ± 0.05.

Fig. 1. Boyle-van’t-Hoff plot for cord RBCs. Data points are equilibrium relative cell volumes in various sodium chloride concentrations (0.68%, 0.9%, 1.6%, and 3.5% (w/ v)). Osmotically inactive fraction (b) was determined form y-intercept of the linear regression fit to the data. N = 3 cord RBC samples.

Hydraulic conductivity Fig. 2 shows a representative plot for the adult RBC volume kinetics upon exposure to 2 PBS at 3.8 °C. As can be seen from the plot, RBC volume reaches osmotic equilibrium within approximately 0.01 min (0.6 s). The hydraulic conductivity of adult and cord RBCs in this solution (562 mOsm/kg (2 PBS) at the three different experimental temperatures are given in Table 2. Hydraulic conductivity increased with temperature in both adult RBCs (r = 0.586, p < 0.001) and cord RBCs (r = 0.725, p < 0.001). Statistical analysis shows that the hydraulic conductivity of cord and adult RBCs was the same at 3.8 °C (p = 0.83) and 19.9 °C (p = 0.075), though cord RBCs were more permeable to water at 35.3 °C (p < 0.001). Hydraulic conductivity values of adult and cord RBCs measured over the range of experimental osmolality (214–1112 mOsm/kg) and at room temperature are included in Table 3. The hydraulic conductivity of cord RBCs was significantly lower than that of adult RBCs in 214 mOsm/kg (0.75 PBS, p < 0.001) and 1112 mOsm/kg (4 PBS, p = 0.009). However, hydraulic conductivities were equal in 562 mOsm/kg (2 PBS, p = 0.075) and 832 mOsm/kg (3 PBS, p = 0.401) solutions. Hydraulic conductivity also significantly increased with extracellular osmolality in both adult (r = 0.381, p < 0.001) and cord (r = 0.402, p < 0.001) RBCs. Increase of hydraulic conductivity was observed between 214 and 832 mOsm/kg (0.75–3 PBS) for adult RBCs and from 214 to 562 mOsm/kg (0.75–2 PBS) for cord RBCs. In 1112 mOsm/kg (4 PBS), hydraulic conductivity was slightly lower compared to its value in less concentrated test solutions for adult (p = 0.004) and cord RBCs (p < 0.001). Comparison between Pf values calculated using the method of Zeidel et al. [67] vs. those determined from Lp values is shown in

Table 1 Hemolysis (%) of adult and cord RBCs in experimental solutions. Source of RBCs

Adult Cord

Experimental solution 0.75 PBS (214 mOsm/kg)

1 PBS (287 mOsm/kg)

2 PBS (562 mOsm/kg)

3 PBS (832 mOsm/kg)

4 PBS (1112 mOsm/kg)

2.5% (w/v) Glycerol (578 mOsm/kg)

0.96 ± 0.15 1.1 ± 0.17

0.53 ± 0.18 0.83 ± 0.12

0.81 ± 0.04 1.1 ± 0.14

0.93 ± 0.05 1.4 ± 0.28

0.96 ± 0.06 1.5 ± 0.30

0.44 ± 0.02 0.93 ± 0.17a

Values are means ± SEM (n = 3). a p = 0.05 vs. adult.

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

5

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

Table 4 Comparison between calculated and converted values for osmotic water permeability (Pf). Pf value

Calculated according to Zeidel et al. Converted from Lp

Experimental temperature (°C) 4.0

19.0

35.2

0.012 ± 0.001

0.018 ± 0.002

0.027 ± 0.005

0.018 ± 0.001

0.027 ± 0.004

0.040 ± 0.008

‘‘Calculated’’ Pf values were determined using Eqs. (4)–(6) from Zeidel et al. [67]. ‘‘Converted’’ Pf values were determined by converting corresponding Lp values into cm/s using Eq. (6). Data is presented for one adult RBC sample, run in triplicate. Values are mean ± SEM.

Fig. 2. Representative plot of the kinetics of adult RBC volume upon exposure to 2 PBS at 4 °C. Rhombuses are the relative cell volumes and solid line is the fitted curve to data.

Table 2 Hydraulic conductivity (Lp) of adult and cord RBCs at various temperatures. Source of RBCs

Experimental temperature (°C)

Adult Cord

3.8 ± 0.1

19.8 ± 0.1

35.3 ± 0.1

8.19 ± 0.38 10.96 ± 1.48

18.11 ± 1.51 23.16 ± 2.51

23.80 ± 1.48 33.13 ± 1.53a

Lp values were obtained on exposure of RBCs to 2 PBS (562 mOsm/kg) solution at 3 different experimental temperatures. a p < 0.05 vs. adult. Values are means ± SEM (n = 8 for adult RBCs, n = 7 for cord RBCs).

Table 3 Hydraulic conductivity (Lp) of adult and cord RBCs at defined osmolality. Source of RBCs

Adult Cord

Fig. 3. Arrhenius plot of the natural logarithm for hydraulic conductivity (Lp, lm/ min/atm) of adult (d) and cord (s) RBCs as a function of inverse absolute temperature (K1). Activation energies (Ea) were calculated from the slope of the linear regression fits to the data. Values are means ± SEM (n = 8 for adult RBCs, n = 7 for cord RBCs).

Experimental osmolality (mOsm/kg) at RT 214

562

832

1112

9.51 ± 0.35 6.66 ± 0.22a

18.11 ± 1.51 23.16 ± 2.51

23.76 ± 0.70 24.50 ± 0.54

16.94 ± 0.20 16.18 ± 0.18a

Lp values were obtained at room temperature (19.8 ± 0.1 °C) on exposure of RBCs to four different concentrations of PBS solution. a p < 0.05 vs. adult. Values are means ± SEM (n = 8 for adult RBCs, n = 7 for cord RBCs).

Table 4. Converting Lp from lm/min/atm to cm/s always resulted in a Pf value that was 50% larger than the one calculated using Zeidel’s approach [67]. The Arrhenius activation energy for hydraulic conductivity was 4.8 kcal/mol for adult RBCs and 5.4 kcal/mol for cord RBCs (Fig. 3). No statistically significant differences were found between activation energies for adult and cord RBCs (p = 0.275). Permeability to glycerol Fig. 4C shows a representative plot of adult RBC volume kinetics upon exposure to 2.5% (w/v) glycerol in 1 PBS solution at 20 °C. The first part of the curve depicts RBC shrinking (V/V0 decreases with time), as water leaves the cell. The second part of the curve represents RBC swelling (V/V0 increases with time), as glycerol enters the cell, followed by a plateau, as glycerol reaches equilibrium on both sides of the RBC membrane. As can be seen from the plot, water efflux from the RBC is complete within approximately 0.01 min (0.6 s), and glycerol influx is complete within approximately 0.6 min (36 s). Values of glycerol permeability (Pglycerol) for adult and cord RBCs are shown for three experimental

temperatures at Table 5. Permeability to glycerol increased with temperature for both adult (r = 0.837, p < 0.001) and cord RBCs (r = 0.801, p < 0.001). Pglycerol of cord RBCs was higher than that of adult RBCs at 3.8 °C (p < 0.001), though permeabilities of adult and cord RBCs to glycerol were equal at both 20.2 °C (p = 0.079) and 35.3 °C (p = 0.435). The Arrhenius activation energies for glycerol permeability were 11.5 kcal/mol and 9.7 kcal/mol for adult and cord RBCs, respectively (Fig. 5), and thus significantly lower for cord RBCs than for adult RBCs (p = 0.0003). Discussion In this study we determined the osmotic permeability of cord and adult RBCs to water and glycerol, as well as the Arrhenius activation energies for these processes. To do this, we used a newly developed method that records the rapid kinetics of osmoticallyinduced RBC volume changes via intrinsic RBC hemoglobin fluorescence [69]. Integrity of red cells in experimental solutions A RBC hemolysis assay demonstrated that 99% RBCs were intact in experimental PBS and glycerol solutions and were thus able to respond to changes in extracellular solution composition by changing their cell volume. A slight increase in hemolysis in 3 and 4 PBS may be due to the large degree of shrinkage of RBCs in these solutions, which results in seemingly smaller hematocrit than in isotonic solutions. The value for hematocrit is used in the

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

6

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx Table 5 Permeability of adult and cord RBCs to glycerol (Pglycerol) at defined temperatures. Source of RBCs

Adult Cord

Experimental temperature (°C) 3.8 ± 0.01

19.9 ± 0.06

35.3 ± 0.01

1.24 ± 0.19 2.30 ± 0.31a

8.49 ± 0.87 10.46 ± 0.63

19.96 ± 1.60 21.98 ± 0.03

Pglycerol values were obtained upon exposure of RBCs to 2.5% (w/v) glycerol in 1 PBS solution at 3 different experimental temperatures. a p < 0.05 vs. adult. Values are means ± SEM (n = 8 for adult RBCs, n = 7 for cord RBCs).

Fig. 5. Arrhenius plot of the natural logarithm for glycerol permeability (Pglycerol, lm/min) of adult (d) and cord (s) RBCs as a function of inverse absolute temperature (K1). Activation energies (Ea) were calculated from the slope of the linear regression fits to the data. Values are means ± SEM (n = 8 for adult RBCs, n = 7 for cord RBCs). ⁄p < 0.05 vs. adult.

previously unreported, was here determined to be 0.56 ± 0.04. Cord RBCs have a larger osmotically inactive fraction compared to that reported for adult RBCs (b = 0.41–0.51) [16,33,53]. Temperature and osmolality dependence of osmotic parameters Fig. 4. (A) Curves representing kinetics of RBC volume after mixing with 1 PBS (RBCs+1 PBS) and 2.5% (w/v) glycerol (RBCs+2.5% glycerol). (B) Curve generated after RBCs+1 PBS curve was subtracted from RBCs+2.5% glycerol curve. (C). Curve generated after isotonic RBC volume was added to each data point on the curve generated in (B). Rhombuses are the relative cell volumes and solid line is the fitted curve to data.

equation to calculate percent hemolysis in such a way that a lower hematocrit would result in a higher percent hemolysis. In the 2.5% (w/v) glycerol solution, hemolysis of cord RBCs was slightly greater than for adult RBCs. Since cord RBCs are known to be more mechanically fragile than adult RBCs [44], this increased hemolysis could have been the result of damage during centrifugation while processing the samples. It could also be due to cord RBC’s lower stability during hypothermic storage (1–6 °C) compared with adult RBCs, which results in higher levels of hemolysis [19,29], especially in the absence of any preservatives in the solution. Osmotically inactive fraction The average isotonic cell volumes obtained for adult and cord RBCs were in a good agreement with values reported in the literature [10]. The volumes of cord RBCs were larger than those of adult RBCs and the osmotically inactive fraction for cord RBCs, which was

The permeability of adult and cord RBCs to water and glycerol was strongly dependent on temperature. In particular, the hydraulic conductivity and glycerol permeability both increased significantly with temperature in both adult and cord RBCs (Tables 2 and 5). The results further suggest an osmolality-dependence on hydraulic conductivity. Data describing osmolality-dependence on hydraulic conductivity in the literature are controversial. Ross-Rodriguez showed that the hydraulic conductivity of TF-1 cells did not change significantly with extracellular osmolality [56]. On the other hand, Rich et al. reported that the hydraulic conductivity of human and dog erythrocytes decreased with an increase in osmolality of the medium [54]. In that study, the hydraulic conductivity of human erythrocytes decreased from 1.87 cm3/dyne s in 199 mOsm/kg solution to 0.76 cm3/dyne s in 516 mOsm/kg solution [54]. Liu et al. reported different results still, demonstrating the opposite trend in the relationship between hydraulic conductivity and extracellular osmolality in canine RBCs. Their data generally suggested that hydraulic conductivity increased when osmolality increased (r = 0.35). For example, increase in hydraulic conductivity of dog RBCs was observed with increase in osmolality in 220–360 mOsm/kg solutions [34]. However, in 445 mOsm/kg solution, a slight drop in hydraulic conductivity was observed. Our data are consistent with the observations of Liu et al. [34]: the hydraulic

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

7

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

conductivity of RBCs increased with increase in osmolality (r2 = 0.343 for adult and 0.526 for cord RBCs) over the range of most experimental osmolalities. However, in PBS solution of very high osmolality (1112 mOsm/kg or 4 PBS), hydraulic conductivity decreased compared to values in lower PBS concentrations. The threshold of osmolality where hydraulic conductivity starts decreasing differs between these two studies. This may be attributed to the difference in species (dog vs. human). There are two possible reasons for the decrease in RBC Lp at very high osmolalities. First, Farrant and Woolgar showed that at high extracellular concentrations of NaCl (1200 mmol/kg sodium) RBCs begin losing potassium and taking up sodium ions [21]. If this were true here, efflux of water from RBCs upon exposure to the 1112 mOsm/kg PBS solution would be counter-balanced by an influx of water into the RBCs in the direction of osmotic pressure gradient, caused by increasing concentration of intracellular sodium ions. As a result, cell shrinking will be slowed, and this will be reflected by a lower value for hydraulic conductivity. As a second possibility, 4 PBS has been reported to be a lower threshold for RBC osmotic tolerance [43]. In solutions of such high osmolality, osmotically-induced shrinking of RBCs may be limited (cell approaches its osmotically inactive volume) which, again, will be reflected by a lower value for hydraulic conductivity.

Hydraulic conductivity and its activation energy Differences in hydraulic conductivity between adult and cord RBCs varied depending on the experimental conditions (temperature and extracellular osmolality). Cord RBCs showed a lower Lp than adult RBCs when subjected to extreme swelling or shrinking (in 214 and 1112 mOsm/kg solutions, Table 3). At the same time, when subjected to more moderate cell volume changes (in 562 and 832 mOsm/kg solutions, Tables 2 and 3), the Lp of cord RBCs was equal or higher (depending on the experimental temperature that the measurements were taken at) than that of adult RBCs. The literature reports that, overall, the hydraulic conductivity of fetal RBCs is lower than that of adult RBCs. Using a stopped-flow

light scattering technique, Agre et al. measured the hydraulic conductivity of adult and fetal RBCs (from neonates 28–40 weeks gestational age) after doubling isotonic osmolality at 37 °C [3]. They reported that fetal RBCs were 1.4 times less permeable to water compared to adult RBCs. The reduced osmotic water permeability of fetal RBCs was correlated with a lower expression of aquaporin CHIP protein – the major water transporter – in their membranes. In a different study, Sjolin determined the hydraulic conductivities of adult and cord RBCs by measuring the light scattering of cells suspended in prehemolytic and hemolytic NaCl concentrations [60]. He similarly reported that RBCs from cord blood were approximately 1.8 (hemolytic NaCl concentrations) and 2.5 (prehemolytic NaCl concentrations) less permeable to water compared to adult RBCs. The Arrhenius activation energy for hydraulic conductivity was found to be equivalent for cord and adult RBCs (5.4 kcal/mol for cord and 4.8 kcal/mol for adult, p > 0.05). This finding differs from results reported by Agre et al. who measured the activation energy for osmotic water transport for adult and cord RBC ghosts at 8–39 °C using the stopped-flow light scattering technique [3]. They found that the activation energy for cord RBCs was 1.3 times higher than for adult RBCs (p < 0.01). Our values for activation energies for cord and adult RBCs are in agreement with those reported in the literature. In particular, activation energies for osmotic water permeability have been reported to be 3.6 ± 0.4 kcal/mol [67], 3.3 ± 0.4 kcal/mol [65], and 4.6 ± 0.8 kcal/mol [3] for adult RBCs, and 6.0 ± 0.5 kcal/mol for cord RBCs [3]. Low activation energies such as these (around 5 kcal/mol on average) are known to correlate with the presence of aquaporins (water transporting channels) in the cell membrane [2,20].

Comparison between hydraulic conductivity (Lp) and osmotic water permeability (Pf) The majority of published papers report values of adult RBC osmotic water permeability as Pf in cm/s; in order to compare our data to published data, we therefore converted our Lp values

Table 6 Summary of literature on osmotic parameters for adult and cord RBCs. Source

Method

Osmotic water permeability values for adult RBCs Agre, P. et al. J. Clin. Invest. 1994 Stopped-flow by light scattering Zeidel, M.L. et al. Biochemistry 1992 Stopped-flow fluorimetry with carboxyfluorescein diacetate Mlekoday, H.J. et al. J. Gen. Physiol. 1983 Stopped-flow by light scattering Sidel, V.W and Solomon, A.K. J. Gen. Physiol. Stopped-flow by light scattering 1957 Liu, L. et al. J. Comp. Physiol. B 2011 Stopped-flow by light scattering Sjölin, Acta paediatrica Supplement 1954 Light scattering: Hemolytic NaCl concentrations Prehemolytic NaCl concentrations Osmotic water permeability values for fetal RBCs Agre, P. et al. J. Clin. Invest. 1994 Stopped-flow fluorimetry by light scattering

Sjölin, Acta paediatrica Supplement 1954

Light scattering: Hemolytic NaCl concentrations Prehemolytic NaCl concentrations

Activation energy values for osmotic water permeability for adult RBCs Agre, P. et al. J. Clin. Invest. 1994 Stopped-flow by light scattering Zeidel, M.L. et al. Biochemistry 1992 Stopped-flow fluorimetry with carboxyfluorescein diacetate Vieira, F.L. et al. J. Gen. Physiol. 1970 Rapid reaction continuous flow system analysis for tritiated water (THO) radioactivity Activation energy values for osmotic water permeability for fetal RBCs Agre, P. et al. J. Clin. Invest. 1994 Stopped-flow by light scattering

Value 0.037 ± 0.007 (Pf ± SD) cm/s at 37 °C 0.040 ± 0.009 (Pf ± SEM) cm/s 0.2 cm/s 0.23 ± 0.03 cm4/osm s 0.029 ± 0.004 (Pf ± SEM) cm/s 0.0084 cm/s; 0.022 cm/s 0.016 cm/s

17–28 wk: 0.029 ± 0.009 (Pf ± SD) cm/s at 37 °C 28 wk-term: 0.026 ± 0.006 (Pf ± SD) cm/s at 37 °C 0.0047 cm/s; 0.012 cm/s 0.0066 cm/s

4.6 ± 0.8 kcal/mol at 8–39 °C 3.6 ± 0.4 kcal/mol 3.3 ± 0.4 kcal/mol at 7–37 °C

17–28 wk: 6.7 ± 0.4 kcal/mol at 8–39 °C 28 wk-term: 6.0 ± 0.5 kcal/mol at 8–39 °C

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

8

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

from lm/min/atm to cm/s using Eq. (6). Summarized literature data on adult and cord RBC permeability are included in Table 6, our calculated Pf values for adult RBCs are shown in Table 4, and our calculated Pf values for cord RBCs are not tabulated, but were 0.023 ± 0.003, 0.052 ± 0.006, and 0.078 ± 0.004 cm/s at 4.0, 19.0, and 35.2 °C, respectively. If we convert our hydraulic conductivity data to osmotic water permeability and compare it to literature data, our values will be about 1.5 times higher than in the majority of published reports. As we have stated before, the process of converting Lp to Pf would result in a 50% increase in Pf value. Accounting for the impact of the conversion, it is reasonable to conclude that our values for osmotic permeability of RBCs are in good agreement with published data. Glycerol permeability and its activation energy There were also differences in glycerol permeability between adult and cord RBCs. Specifically, cord RBCs were more permeable to glycerol at 4 °C. Though, at other investigated temperatures (20 °C and 35 °C), the glycerol permeabilities of cord and adult RBCs were observed to be equivalent (Table 5). There are few published studies on the permeability of cord RBCs to glycerol. Both studies were done in 0.3 M glycerol, which is equivalent to 2.8% (w/v) glycerol and, therefore, comparable to the glycerol concentration we used in our study. Moore reported that the average 50% hemolysis time in 0.3 M glycerol was approximately twice as long for fetal RBCs than for adult RBCs. He concluded that adult RBCs are twice as permeable to glycerol compared to fetal RBCs, with the difference presumably caused by the higher percentage of lecithin in adult erythrocyte membranes [46]. In his study, Moore used a glycerol lysis test to estimate cell permeability to glycerol. In this test, erythrocytes are exposed to a hypotonic salt solution (about 0.3% NaCl) containing 0.3 M glycerol. Glycerol prevents rapid flux of water inside the cell in the direction of osmotic gradient, and, as both glycerol and water slowly enter the cell, hemolysis is monitored as a function of time. This method was originally developed as the test for osmotic fragility, specifically used as a screening test for hereditary spherocytosis – a condition in which an increased osmotic fragility of erythrocytes is caused by mutations in cytoskeletal proteins. Bautista et al. used the same test to again show that it takes fetal RBCs longer to lyse in 0.3 M glycerol than adult RBCs, and interpreted this as a measure of decreased osmotic fragility of fetal erythrocytes [7]. In the glycerol lysis test, it would make sense that the time to hemolysis is determined by membrane permeability not only to glycerol, but also to water (since water enters the cell from extracellular hypotonic solution), which would be a confounding factor. As the permeability of the cell membrane to water in hypotonic solutions is lower for cord than for adult RBCs (Table 2), it would take longer for water to enter cord RBC and cause lysis. This would result in a seemingly decreased permeability of the cell to glycerol. Previously it was thought that glycerol, as well as other small molecules, cross the cell membrane predominantly by diffusion through the lipid bilayer [59]. However, it is now recognized that glycerol transport across the human erythrocyte membrane occurs through specialized channels called aquaglyceroporins [58]. The differences in glycerol permeability between cord and adult RBCs therefore need to be interpreted in light of these new findings. Although expression of aquaglyceroporins has not been evaluated in fetal RBCs yet, based on our data it is reasonable to believe that differences in glycerol permeability would imply differences in the expression and/or function of these channels in fetal vs. adult erythrocytes. Several studies have reported values for Pglycerol in adult RBCs. Naccache and Sha’afi measured the glycerol permeability

coefficient of adult RBCs exposed to 0.3 M glycerol in hypotonic saline solution at 19–24 °C using a combination of hemolysis and a stopped-flow technique [47]. They reported Pglycerol as 0.58 ± 0.04  105 cm/s, which is equivalent to 3.48 lm/min. Mazur and Miller estimated the glycerol permeability coefficient for adult RBCs using 1) the time to 50% hemolysis in hypotonic saline solutions containing 1 and 2 M glycerol, and 2) the time it took RBCs in isotonic saline solution containing 1 or 2 M glycerol to undergo osmotic shock upon tenfold dilution with isotonic saline buffer [40]. They reported glycerol permeability coefficients as 2.5  104 cm/min at 20 °C and 0.9  104 cm/min at 0 °C, which is equivalent to 2.5 lm/min and 0.9 lm/min, respectively. The glycerol permeability of adult RBCs measured in our study was higher than that presented in these two reports (8.49 ± 0.87 lm/ min at 20 °C). This could have been caused by significant differences in the methods used to estimate Pglycerol. In particular, Naccache’s and Mazur’s methods were based on measuring the time to RBC lysis in hypotonic solutions in the presence of various amounts of glycerol. The limitations of this methodological approach have been discussed above. The Arrhenius activation energy for glycerol permeability was found to be lower for cord RBCs than for adult RBCs (9.7 kcal/mol for cord vs. 11.5 kcal/mol for adult, p < 0.05). Activation energy of 11.5 kcal/mol is slightly higher than the 7.2 kcal/mol reported by Mazur and Miller as an activation energy for glycerol permeability for adult RBCs [40].

Conclusion This study provides values for osmotic parameters of cord and adult RBCs, such as the osmotically inactive fraction, water and glycerol permeability, and the activation energies for these processes. Osmotic parameters were determined over a broad range of experimental temperatures and osmolalities, providing further understanding and insights to previous work in which osmotic parameters were determined under very specific and limited experimental conditions. Moreover, this study adds to the limited published data available for cord red cell osmotic permeability to water and glycerol. We conclude that cord RBCs have a larger osmotically inactive fraction than adult RBCs. The hydraulic conductivity and osmotic permeability to glycerol of cord RBCs differ from those of adult RBCs with the extent of these differences dependent on experimental conditions, such as temperature and osmolality. In the future, these osmotic parameters can help inform the use of mathematical modeling to determined optimized cryopreservation protocol for cord RBCs.

Conflict of interest The authors have no financial conflicts of interest to declare. Acknowledgments We would like to thank Dr. John Akabutu, Nanni Zhang and Sally Shahi from the Alberta Cord Blood Bank for providing cord RBC samples. We are also grateful to the Alberta Cord Blood Bank cord blood donors and the Canadian Blood Services adult peripheral blood donors for providing samples for this study. We would like to thank Dr. Lisa Ross-Rodriguez (University of Alberta) and Dr. James Benson (Northern Illinois University) for their insights and Dr. Andy Holt (Department of Pharmacology, University of Alberta) for granting us access to the stopped-flow system for this study.

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

Appendix. Conversion from fluorescence to cell volume Normalization of fluorescence kinetics curves (background subtraction) For analysis of hydraulic conductivity, fluorescence of PBS buffer was averaged over 10 s. This value was then subtracted from each data point on the curve representing fluorescence of RBCs mixed with this buffer. For analysis of glycerol permeability, fluorescence of 2.5% (w/v) glycerol and 1 PBS buffers was averaged over 10 s. These values were then subtracted from fluorescence of RBCs mixed with 2.5% (w/v) glycerol and 1 PBS (osmotic equilibrium conditions, control), respectively. Conversion from fluorescence to volume A linear regression was used to convert relative fluorescence     F to relative volume VV0 . F0

V F ¼m þc V0 F0

ðA1Þ

where V is the equilibrium RBC volume at an experimental osmolality, V0 is the isotonic RBC volume, F is the equilibrium RBC fluorescence at an experimental osmolality, and F0 is the isotonic RBC fluorescence, m is the slope, and c is the y intercept. Equilibrium RBC volumes in anisotonic NaCl solutions were measured on Coulter Electronic Particle Counter (ZB1, Coulter Electronics, Inc., Hialeah, FL, USA) equipped with a pulse-height analyzer (The Great Canadian Computer Company, Spruce Grove, AB, Canada) [45]. Isotonic volume of RBCs was measured on Coulter AcT Series Analyzer (Beckman Coulter, Inc., Brea, California). Equilibrium RBC autofluorescence in NaCl solutions of various tonicities was obtained using stopped-flow system. Data for adult and cord RBCs were fit to Eq. (7) to derive values for m and c. For adult RBCs, m = 0.8406 and c = 0.2227. For cord RBCs, m = 0.7278 and c = 0.2129. For analysis of hydraulic conductivity, each data point on the normalized curve representing fluorescence of RBCs mixed with PBS was converted to RBC volume using Eq. (7). Analysis of glycerol permeability was performed in three steps. First, each data point on the normalized curves representing fluorescence of RBCs mixed with 2.5% (w/v) glycerol and 1 PBS was converted to RBC volume using Eq. (7) (Fig. 4A). Second, each data point on the osmotic equilibrium curve (RBCs+1 PBS) was subtracted from each corresponding data point on the curves representing fluorescence of RBCs mixed with 2.5% (w/v) glycerol (RBCs+2.5% glycerol) (Fig. 4B). Third, isotonic RBC volume was added to each data point on the curve generated in step 2 (Fig. 4C). Subtraction of equilibrium curve in step 2 was done to correct for cell settling in the flow cell that resulted in the RBC fluorescence increase over time after about 10 s spent in a stopped-flow optical cell. References [1] J.P. Acker, I.M. Croteau, Q.L. Li, An analysis of the bias in red blood cell hemolysis measurement using several analytical approaches, Clin. Chim. Acta 413 (2012) 1746. [2] P. Agre, L.S. King, M. Yasui, W.B. Guggino, O.P. Ottersen, Y. Fujiyoshi, A. Engel, S. Nielsen, Aquaporin water channels – from atomic structure to clinical medicine, J. Physiol. 542 (2002) 3. [3] P. Agre, B.L. Smith, R. Baumgarten, G.M. Preston, E. Pressman, P. Wilson, N. Illum, D.J. Anstee, M.B. Lande, M.L. Zeidel, Human red cell aquaporin CHIP. II. Expression during normal fetal development and in a novel form of congenital dyserythropoietic anemia, J. Clin. Invest. 94 (1994) 1050. [4] M.V. Appalup, T.A. Fedorova, The effectiveness and safety of autologous umbilical blood derived red blood cells in a treatment of postoperative anaemia in newborns with a surgical pathology, Vox Sang. 99 (2010) 408.

9

[5] A. Ballin, E. Arbel, G. Kenet, M. Berar, D. Kohelet, A. Tanay, H. Zakut, D. Meytes, Autologous umbilical cord blood transfusion, Arch. Dis. Child. 73 (1995) 181. [6] F.C. Battaglia, W. Bowes, H.R. McGaughey, E.L. Makowski, G. Meschia, The effect of fetal exchange transfusions with adult blood upon fetal oxygenation, Pediatr. Res. 3 (1969) 60. [7] M.L. Bautista, W. Altaf, R. Lall, R.A. Wapnir, Cord blood red cell osmotic fragility: a comparison between preterm and full-term newborn infants, Early Human Dev. 72 (2003) 37. [8] F. Bertolini, M. Battaglia, C. Zibera, G. Baroni, V. Soro, C. Perotti, L. Salvaneschi, G. Robustelli della Cuna, A new method for placental/cord blood processing in the collection bag. I. Analysis of factors involved in red blood cell removal, Bone Marrow Transplant. 18 (1996) 783. [9] O.M. Bobrova, Influence of cryoprotectants and low temperatures on thermodynamic parameters and structural state of cord blood component human placenta, Institute for Problems of Cryobiology and Cryomedicine National Academy of Sciences of Ukraine Kharkiv, 2007. [10] C. Brugnara, Appendices. Reference values in infancy and childhood, in: D.G. Nathan, S.H. Orkin, D. Ginsburg, A.T. Look (Eds.), Hematology of Infancy and Childhood, Saunders, Philadelphia, Pennsylvania, 2003, p. 1835. [11] C. Brugnara, O.S. Platt, The neonatal erythrocyte and its disorders, in: D.G. Nathan, S.H. Orkin, D. Ginsburg, A.T. Look (Eds.), Hematology of Infancy and Childhood, Saunders, Philadelphia, Pennsylvania, 2003, p. 19. [12] T. Brune, H. Garritsen, R. Hentschel, F. Louwen, E. Harms, G. Jorch, Efficacy, recovery, and safety of RBCs from autologous placental blood: clinical experience in 52 newborns, Transfusion 43 (2003) 1210. [13] T. Brune, H. Garritsen, R. Witteler, A. Schlake, J. Wüllenweber, F. Louwen, G. Jorch, E. Harms, Autologous placental blood transfusion for the therapy of anaemic neonates, Biol. Neonate 81 (2002) 236. [14] C. Clark, J.A. Gibbs, R. Maniello, E.W. Outerbridge, J.V. Aranda, Blood transfusion: a possible risk factor in retrolental fibroplasia, Acta Paediatr. Scand. 70 (1981) 537. [15] K.J. Collard, S. Godeck, J.E. Holley, Blood transfusion and pulmonary lipid peroxidation in ventilated premature babies, Pediatr. Pulmonol. 39 (2005) 257. [16] J.S. Cook, Nonsolvent water in human erythrocytes, J. Gen. Physiol. 50 (1967) 1311. [17] R.W. Cooke, J.A. Drury, C.W. Yoxall, C. James, Blood transfusion and chronic lung disease in preterm infants, Eur. J. Pediatr. 156 (1997) 47. [18] S.L. Ebertz, L.E. McGann, Cryoprotectant permeability parameters for cells used in a bioengineered human corneal equivalent and applications for cryopreservation, Cryobiology 49 (2004) 169. [19] H. Eichler, T. Schaible, E. Richter, W. Zieger, K. Voller, A. Leveringhaus, S.F. Goldmann, Cord blood as a source of autologous RBCs for transfusion to preterm infants, Transfusion 40 (2000) 1111. [20] H.Y. Elmoazzen, J.A. Elliott, L.E. McGann, The effect of temperature on membrane hydraulic conductivity, Cryobiology 45 (2002) 68. [21] J. Farrant, A.E. Woolgar, Human red cells under hypertonic conditions; a model system for investigating freezing damage. I. Sodium chloride, Cryobiology 9 (1972) 9. [22] J.C. Fouron, H. Bard, J.C. Le Guennec, M.R. Van Amerigen, Effect of fetal or adult red cells on tissue oxygenation and myocardial function in normoxemic newborn lambs, Pediatr. Res. 15 (1981) 967. [23] W.R. Hepner, A.C. Krause, Retrolental fibroplasia: clinical observations, Pediatrics 10 (1952) 433. [24] H. Hume, Red blood cell transfusions for preterm infants: the role of evidencebased medicine, Semin. Perinatol. 21 (1997) 8. [25] K. Imura, H. Kawahara, Y. Kitayama, A. Yoneda, M. Yagi, N. Suehara, Usefulness of cord-blood harvesting for autologous transfusion in surgical newborns with antenatal diagnosis of congenital anomalies, J. Pediatr. Surg. 36 (2001) 851. [26] M.H. Jacobs, D.R. Stewart, A simple method for the quanttative measurement of cell permeability, J. Cell. Comp. Physiol. 1 (1931) 71–82. [27] S.K. Jain, The neonatal erythrocyte and its oxidative susceptibility, Semin. Hematol. 26 (1989) 286. [28] C.M. Khodabux, A. Brand, The use of cord blood for transfusion purposes: current status, Vox Sang. 97 (2009) 281. [29] C.M. Khodabux, J.M. van Beckhoven, J.G.M. Scharenberg, F. El Barjiji, M.C. Slot, A. Brand, Processing cord blood from premature infants into autologous redblood-cell products for transfusion, Vox Sang. 100 (2011) 367. [30] C.M. Khodabux, J.S. Von Lindern, J.A. Van Hilten, S. Scherjon, F.J. Walther, A. Brand, A clinical study on the feasibility of autologous cord blood transfusion for anemia of prematurity, Transfusion 48 (2008) 1634. [31] F.W. Kleinhans, Membrane permeability modeling: Kedem-Katchalsky vs a two-parameter formalism, Cryobiology 37 (1998) 271. [32] P. Korhonen, O. Tammela, A.M. Koivisto, P. Laippala, S. Ikonen, Frequency and risk factors in bronchopulmonary dysplasia in a cohort of very low birth weight infants, Early Human Dev. 54 (1999) 245. [33] S.W. Levin, R.L. Levin, A.K. Solomon, A. Pandiscio, D.H. Kirkwood, Improved stop-flow apparatus to measure permeability of human red cells and ghosts, J. Biochem. Biophys. Methods 3 (1980) 255. [34] J. Liu, J.A. Christian, J.K. Critser, Canine RBC osmotic tolerance and membrane permeability, Cryobiology 44 (2002) 258. [35] J. Liu, S. Mullen, Q. Meng, J. Critser, A. Dinnyes, Determination of oocyte membrane permeability coefficients and their application to cryopreservation in a rabbit model, Cryobiology 59 (2009) 127. [36] N.L. Luban, Neonatal red blood cell transfusions, Vox Sang. 87 (Suppl. 2) (2004) 184.

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002

10

M. Zhurova et al. / Cryobiology xxx (2014) xxx–xxx

[37] B.F. Ludvigsen, Hemoglobin synthesis and function, in: E.A. Stiene-Martin, C.A. Lotspeich-Steininger, J.A. Koepke (Eds.), Clinical Hematology: Principles, Procedures, Correlations, Lippincott, Philadelphia, Pennsylvania, 1997, p. 73. [38] H. Mallek, P. Spohn, Retrolental fibroplasia, Can. Med. Assoc. J. 63 (1950) 586. [39] L.M. Matovcik, W.C. Mentzer, The membrane of the human neonatal red cell, Clin. Haematol. 14 (1985) 203. [40] P. Mazur, R.H. Miller, Permeability of the human erythrocyte to glycerol in 1 and 2 M solutions at 0 or 20 degrees C, Cryobiology 13 (1976) 507. [41] L.E. McGann, S.L. Ebertz, J.A.W. Elliott, Optimal cooling rates from osmotic simulation of cellular low temperature responses, 2002, pp. 255. [42] L.E. McGann, J.A.W. Elliott, Optimization of cryopreservation protocols using computer simulations, 2003, pp. 255. [43] H.T. Meryman, Cryopreservation of living cells: principles and practice, Transfusion 47 (2007) 935. [44] J. Meyburg, T. Bohler, O. Linderkamp, Decreased mechanical stability of neonatal red cell membrane quantified by measurement of the elastic area compressibility modulus, Clin. Hemorheol. Microcirc. 22 (2000) 67. [45] K.J. Moise Jr., Intrauterine transfusion with red cells and platelets, West. J. Med. 159 (1993) 318. [46] T.J. Moore, Glycerol permeability of human fetal and adult erythrocytes and of a model membrane, J. Lipid Res. 9 (1968) 642. [47] P. Naccache, R.I. Sha’afi, Patterns of nonelectrolyte permeability in human red blood cell membrane, J. Gen. Physiol. 62 (1973) 714. [48] R.L. Nagel, Hemoglobins: normal and abnormal, in: D.G. Nathan, S.H. Orkin, D. Ginsburg, A.T. Look (Eds.), Hematology of Infancy and Childhood, Saunders, Philadelphia, Pennsylvania, 2003, p. 745. [49] P.S. Nobel, The Boyle-Van’t hoff relation, J. Theor. Biol. 23 (1969) 379. [50] F.A. Oski, Fetal hemoglobin, the neonatal red cell, and 2,3-diphosphoglycerate, Pediatr. Clin. North Am. 19 (1972) 907. [51] F.A. Oski, The unique fetal red cell and its function. E. Mead Johnson award address, Pediatrics 51 (1973) 494. [52] P. Perutelli, S. Catellani, L. Scarso, P. Cornaglia-Ferraris, G. Dini, Processing of human cord blood by three different procedures for red blood cell depletion and mononuclear cell recovery, Vox Sang. 76 (1999) 237. [53] R.C. Prickett, J.A.W. Elliott, S. Hakda, L.E. McGann, A non-ideal replacement for the Boyle van’t Hoff equation, Cryobiology 57 (2008) 130. [54] G.T. Rich, I. Sha’afi, A. Romualdez, A.K. Solomon, Effect of osmolality on the hydraulic permeability coefficient of red cells, J. Gen. Physiol. 52 (1968) 941. [55] L.U. Ross-Rodriguez, Using simulations to design a cryopreservation procedure for hematopoietic stem cells without DMSO, Medical Sciences – Laboratory Medicine and Pathology (Edmonton University of Alberta), 2004, pp. 1. [56] L.U. Ross-Rodriguez, J.A. Elliott, L.E. McGann, Characterization of cryobiological responses in TF-1 cells using interrupted freezing procedures, Cryobiology 60 (2010) 106.

[57] L.U. Ross-Rodriguez, J.A.W. Elliott, L.E. McGann, Investigating cryoinjury using simulations and experiments. 1: TF-1 cells during two-step freezing (rapid cooling interrupted with a hold time), Cryobiology 61 (2010) 38. [58] N. Roudier, J.M. Verbavatz, C. Maurel, P. Ripoche, F. Tacnet, Evidence for the presence of aquaporin-3 in human red blood cells, J. Biol. Chem. 273 (1998) 8407. [59] R.I. Sha’afi, C.M. Gary-Bobo, Water and nonelectrolytes permeability in mammalian red cell membranes, Prog. Biophys. Mol. Biol. 26 (1973) 103. [60] S. Sjolin, The resistance of red cells in vitro; a study of the osmotic properties, the mechanical resistance and the storage behaviour of red cells of fetuses, children and adults, Acta Paediatr. Suppl. 43 (1954) 1. [61] T. Sousa, M.E. de Sousa, M.I. Godinho, C. Mendes, A. Carvalhais, I.L. Barbosa, Umbilical cord blood processing: volume reduction and recovery of CD34+ cells, Bone Marrow Transplant. 19 (1997) 311. [62] R.G. Strauss, Autologous transfusions for neonates using placental blood. A cautionary note, Am. J. Dis. Child. 146 (1992) 21. [63] D.V. Surbek, R. Glanzmann, H.P. Senn, I. Hoesli, W. Holzgreve, Can cord blood be used for autologous transfusion in preterm neonates?, Eur J. Pediatr. 159 (2000) 790. [64] W.H. Tooley, Neonatal anemia, in: W.H. Tooley (Ed.), Intensive Care Nursery House Staff manual, UCSF Children’s Hospital at UCSF Medical Center, 2004, p. 108. [65] F.L. Vieira, R.I. Sha’afi, A.K. Solomon, The state of water in human and dog red cell membranes, J. Gen. Physiol. 55 (1970) 451. [66] H. Yang, J.P. Acker, D. Abley, L.E. McGann, J. Akabutu, High-efficiency volume reduction of cord blood using pentastarch, Bone Marrow Transplant. 27 (2001) 457. [67] M.L. Zeidel, A. Albalak, E. Grossman, A. Carruthers, Role of glucose carrier in human erythrocyte water permeability, Biochemistry 31 (1992) 589. [68] M. Zhurova, J. Akabutu, J. Acker, Quality of red blood cells isolated from umbilical cord blood stored at room temperature, J. Blood Transfus. 2012 (2012) 102809, http://dx.doi.org/10.1155/2012/102809. [69] M. Zhurova, A. Olivieri, A. Holt, J.P. Acker, A method to measure permeability of red blood cell membrane to water and solutes using intrinsic fluorescence, Clin. Chim. Acta 431 (2014) 103. [70] P.M. Zubov, Modification of proteins of membrane-cytoskeleton complex and lipid asymmetry in erythrocytes under cooling and freezing in presence of cryoprotectant PEO-1500, Institute for Problems of Cryobiology and Cryomedicine National Academy of Sciences of Ukraine Kharkiv, 2008. [71] A. Zwart, O.W. van Assendelft, B.S. Bull, J.M. England, S.M. Lewis, W.G. Zijlstra, Recommendations for reference method for haemoglobinometry in human blood (ICSH standard 1995) and specifications for international haemiglobinocyanide standard (4th edition), J. Clin. Pathol. 49 (1996) 271.

Please cite this article in press as: M. Zhurova et al., Osmotic parameters of red blood cells from umbilical cord blood, Cryobiology (2014), http://dx.doi.org/ 10.1016/j.cryobiol.2014.04.002