Determination of optimal cryoprotectants and procedures for their addition and removal from human spermatozoa

hrep$$0118 Human Reproduction vol.12 no.1 pp.112 118, 1997 Determination of optimal cryoprotectants and procedures for their addition and removal from human spermatozoa J.A.Gilmore 1,2, J.Liu 1, D.Y.Gao 1 and by Polge et al. (1949). Since this discovery, the development J.K.Critser 1,2,3,4 of a more complete understanding of the mechanisms by which 1 CPA protect cells during cooling and warming has been Cryobiology Research Institute, Methodist Hospital of Indiana, Indianapolis, IN 46202, 2 Department of Veterinary Clinical emphasized. Sciences, School of Veterinary Medicine, Purdue University, A wide range of chemicals is used for cryopreservation, West Lafayette, IN 47907 and 3 Departments of Physiology/ including polyhydroxy alcohols, sugars, inorganic cations and Biophysics and Obstetrics/Gynecology, Indiana University School amino acids (Karow, 1969). CPA function either through of Medicine, Indianapolis, IN 46202, USA colligative properties which maintain intracellular and extra- 4 To whom correspondence should be addressed at: Cryobiology cellular solute concentrations at levels cells can tolerate Research Institute, Methodist Hospital of Indiana, 1701 N. Senate (permeating CPA), or by partially dehydrating the cell, stabiliz- Boulevard Wile Hall Room 611, Indianapolis, IN 46206 1367, ing cell membranes or both (non-permeating CPA; Critser USA et al., 1988). The present study focuses only on permeating The objective was to test the hypothesis that the optimal CPA. Glycerol, dimethyl sulphoxide (DMSO), propylene glycol cryoprotective agent for cryopreservation of human sper- and ethylene glycol were chosen as the CPA of interest. matozoa would be a solute for which cells have the highest Although the use of CPA is imperative for cell survival plasma membrane permeability, resulting in the least during cryopreservation, their presence places osmotic stress amount of volume excursion during its addition and on the cells due to solute and water flux. Upon exposure to removal. To test this hypothesis, theoretical simulations permeating solutes, cells will first shrink as water leaves were performed using membrane permeability coefficients through the plasma membrane, and then swell as water reto predict optimal procedures for the addition and removal enters, together with the CPA. During removal of solutes, cells of a cryoprotectant. Simulations were performed using will initially swell due to the influx of water, and then slowly data from four different cryoprotectants: (i) glycerol, return to iso-osmotic volume as CPA and water efflux. In (ii) dimethyl sulphoxide, (iii) propylene glycol and (iv) addition to the injury caused by freezing, this osmotic stress ethylene glycol. Thermodynamic formulations were applied due to the presence of CPA may also provoke cell injury to determine approaches for the addition and removal of (Mazur and Schneider, 1984; Leibo, 1986; Critser et al., 1988). 1 M and 2 M final concentrations of cryoprotectant, Therefore, procedures for the addition of CPA before cooling allowing the spermatozoa to maintain a cell volume within and removal of CPA after cooling and warming must be their osmotic tolerance limits. Based on these data, ethylene optimized according to specific cell characteristics to ensure glycol was predicted to be optimal for minimizing volume successful cryopreservation. Information is required regarding excursions among the solutes evaluated. These predictions the osmotic tolerance limits of cells, defined as the extent of were then experimentally tested using glycerol as the control volume excursion cells can withstand before irreversible loss cryoprotectant and ethylene glycol as the experimental of function occurs. Gao et al. (1995) determined that human cryoprotectant. The results indicate that there was a higher spermatozoa can swell to 1.1 times and shrink to 0.75 times (P 0.05) recovery of motile spermatozoa after cryo- their iso-osmotic cell volume and maintain functional integrity preservation when using 1 M ethylene glycol than with (motility). To determine whether cells will maintain a volume 1 M glycerol, supporting the hypothesis that use of the within their tolerance limits during CPA addition and removal, cryoprotectant for which the cell has the highest perme- other osmotic characteristics such as the osmotically inactive ability will result in higher cell survival. cell volume, the hydraulic conductivity and the solute perme- Key words: cell volume/cryoprotectant/human spermatozoa/ ability must be defined. Gilmore et al. (1995) have determined permeability these properties for human spermatozoa. With this collective information, it is possible theoretically to predict the optimal procedure for the addition and removal of CPA, as well as the optimal CPA for human spermatozoa cryopreservation. The Introduction objectives of this study were: (i) to determine theoretically the In order for cells to survive the process of cryopreservation, magnitude of sperm cell volume excursion during the addition cryoprotective agents (CPA) must be present during cooling and removal of 1 M and 2 M glycerol, DMSO, propylene and warming. CPA are defined as a class of compounds which glycol and ethylene glycol, through the use of computer specifically act to maintain the viability of frozen animal cells simulations; (ii) to predict the optimal CPA for human spermatozoa (Karow, 1969); their protective nature was first characterized cryopreservation through the information gained from 112 European Society for Human Reproduction and Embryology

Addition and removal of cryoprotectants from spermatozoa the computer simulations; and (iii) to test experimentally the predictions for these CPA. Table I. Characteristics of human spermatozoa at 22 C Surface area A 120 µm 2a Iso-osmotic volume V iso 28.2 µm 3b Materials and methods Osmotically inactive cell volume V b 50.0 µm 3b Solute permeability P Gly 0.21 10 2 cm/min b Samples P DMSO 0.08 10 2 cm/min b P PG 0.23 10 2 cm/min b Human semen samples were obtained, with informed consent, from P EG 0.79 10 2 cm/min b healthy donors by masturbation after at least 48 h of sexual abstinence. Water permeability L Gly p 0.77 µm/min/atm b The samples were allowed to liquefy in an incubator (5% CO 2, 95% L DMSO p 0.84 µm/min/atm b air, 37 C and high humidity) for 30 min. Samples were layered onto L PG p 1.23 µm/min/atm b a 47 and 90% discontinuous Percoll gradient to separate motile from L EG p 0.74 µm/min/atm b immotile cells. The resultant motile population was then washed with a Tyrode s lactate HEPES (TL-HEPES) buffered solution (285 a From Du et al. (1994). 5 mosmol/kg) (Bavister et al., 1983) supplemented with pyruvate b From Gilmore et al. (1995). (0.01 mg/ml) and resuspended to 2.5 ml final sample volume. A 5 µl volume of the sample was analysed using computer-assisted semen analysis (CASA; Cell Soft, Version 3.2/C, CryoResources, Ltd., where C i salt is the intracellular salt concentration, V iso is the cell Montgomery, NY, USA). A minimum concentration of 2 10 7 volume (µm 3 ) at isosmolality, V b is the osmotically inactive cell spermatozoa/ml, with a minimum of 40% motility, was required for volume (µm 3 ), C i CPA is the intracellular CPA concentration and V CPA the samples to be included in the study. is the partial molal volume of CPA (µm 3 ). These are computed from solution tables in the Handbook of Chemistry and Physics (Weast, Media 1970 1971) for glycerol, propylene glycol and ethylene glycol, The media used were iso-osmotic TL-HEPES solution (285 5 mosmol/kg) and derivatives of iso-osmotic TL-HEPES. Hyperyielding values of V glycerol 71 ml/mol, V PG 70 ml/mol and V EG 54 ml/mol respectively. For DMSO, V DMSO 69 ml/mol osmotic medium was made by diluting 1 M sucrose with iso-osmotic (Kiyohara et al., 1975). The sperm volume is defined as V (µm 3 )at TL-HEPES medium to an osmolality of 575 5 mosmol/kg. time t (s). Volume, surface area, V b, V CPA, water and CPA perme- Osmolality was measured using a freezing-point depression osmoin Table I. abilities have been determined for human spermatozoa and are shown meter (Model 3D2; Advanced Instruments, Needham Heights, MA, USA). The cryoprotectant treatment solutions were prepared by As water flows into a cell and CPA flows out of a cell, the volume mixing glycerol or ethylene glycol with TL-HEPES to yield CPA reaches a maximum point and volume flux equals zero. At this point: concentrations of 2 M glycerol, 2 M ethylene glycol and 4 M ethylene glycol. All media used in the experiments were obtained from Sigma J v 0 (C e salt C i salt) σ(c CPA e C i CPA) 0 [2] Chemical Co. (St. Louis, MO, USA). where J v total volume flux and σ is the reflection coefficient. It is assumed that there is no interaction between water and CPA during Theory of optimal CPA addition and dilution membrane transport (Gilmore et al., 1995) and therefore the value of σ 1 (P CPA V CPA )/(RTL p ) (Kedem and Katchalsky, 1958). Hence, Optimal addition and dilution are defined here as the processes that minimize the number of addition and dilution steps as well as the C iso salt(v iso V b )(1 C i CPAV CPA ) osmotic cell volume excursion. A computer software program V max (CellSim V 1.0, Cryobiology Research Institute, Indianapolis, IN, C salt e σ(c CPA e C CPA) i V b [3] USA) was designed to predict optimal CPA addition and removal Determining the dilution procedure, which will maintain cell procedures for human spermatozoa. Under the given experimental volume within the upper tolerance limit, of CPA concentration is conditions (initial intracellular concentration, upper volume limit and crucial. From the above equation, for a given C e CPA, there is a lower volume limit of the cell and temperature), the program optimizes corresponding maximum volume. If V max is not allowed to exceed the addition and dilution steps automatically and provides the appro- the upper limit of volume excursion, priate dilutant concentration. Previous studies have indicated that CPA addition results in less damage to the cell than CPA removal C iso salt(v iso V b )(1 C i CPAV CPA ) V uplimit V max (Gao et al., 1995), and therefore, only further development of optimal C salt e σ(c CPA e C CPA) i dilution is presented here. The theoretical considerations for this V b [4] procedure are described below. which requires: The Kedem Katchalsky membrane transport model (Kedem and C iso salt(v iso V b )(1 C i saltv CPA ) C salt e Katchalsky, 1958) was used to characterize theoretically the cell C CPA e volume change in response to aniso-osmotic conditions. The mathematical (V uplimit V b )σ σ C i CPA [5] formulation includes two coupled first-order non-linear ordin- The above equation suggests that the cell volume excursion will ary differential equations which describe the total transmembrane not exceed the upper limit if the cells are diluted with a solution volume flux and the transmembrane permeable solute flux respectively. whose concentration is in the range which is equal to or higher than Because human spermatozoa behave as ideal osmometers within the the value given by equation [5]. Minimizing the dilution steps means studied osmolality range (Du et al., 1993; Gilmore et al., 1995), that C CPA e should approach zero as quickly as possible. In other intracellular salt concentration can be expressed as: words, C e CPA should be selected as the smallest value in that range. C salt(t) i C iso The C CPA e can decrease further if C salt e is increased, since increasing salt(v iso V b )(1 CCPA(t)V CPA i ) [1] C e salt can decrease V max. However, this is limited by the lower limit V(t) V b of cell volume change: 113

J.A.Gilmore et al. Figure 1. Theoretical simulation for the one-step addition of 1 M and 2 M glycerol, dimethyl sulphoxide (DMSO), propylene glycol and ethylene glycol to human spermatozoa at 22 C. Figure 2. Theoretical simulation for the one-step removal of 1 M and 2 M glycerol, dimethyl sulphoxide (DMSO), propylene glycol and ethylene glycol from human spermatozoa at 22 C. V min C iso salt(v iso V b ) cell suspension through the use of a vortex and then assessing motility. V b [6] Each approach for the addition of a CPA was performed at 22 C. A C salt e maximum of nine donors was used for the control (1 M glycerol) For a given cell type, if the transport parameters (L p, P CPA and σ) and a minimum of four donors was used for the abrupt addition of and upper and lower osmotic tolerance limits are given, dilution 1 M and 2 M ethylene glycol. optimization can proceed as follows: (i) determine C e salt by equation Experiment 2: removal of CPA [6]; (ii) find C CPA e for the first step dilution; (iii) repeat (ii) for new Removal of 1 M glycerol and 1 M ethylene glycol was achieved by step concentration C e CPA. Figure 1 illustrates the relative cell volume adding dropwise over 30 s 2000 µl of iso-osmotic TL-HEPES to the of human spermatozoa as a function of time when 1 M or 2 M 200 µl of the cell solute mixture and then assessing motility; 2 M glycerol, DMSO, propylene glycol or ethylene glycol is added. Figure ethylene glycol was removed by adding dropwise over 30 s 2000 µl 2 illustrates the relative cell volume of human spermatozoa as a of 575 mosmol/kg sucrose. The suspension was centrifuged (400 g function of time when 1 or 2 M glycerol, DMSO, propylene glycol for 7 min), and the supernatant was aspirated. The sample was then or ethylene glycol is removed. resuspended in 500 µl of iso-osmotic TL-HEPES. Motility was measured after addition of 575 mosmol/kg sucrose and after the Experimental design sample was returned to iso-osmotic conditions. Removal of a given Experiment 1: addition of CPA CPA was also performed by abruptly adding 2000 µl of iso-osmotic Two approaches were used for the addition of 2 M glycerol, 2 M TL-HEPES to cell suspensions with 1 M glycerol or 1 M ethylene ethylene glycol and 4 M ethylene glycol to the sperm suspension, glycol, or by abruptly adding 2000 µl of 575 mosmol/kg to the 2 M yielding final concentrations of 1 M glycerol, 1 M ethylene glycol ethylene glycol cell mixture and, again, returning the sample to and 2 M ethylene glycol. In the first approach, 100 µl of a given iso-osmotic conditions and then measuring motility. Cryoprotectant CPA was added dropwise over 30 s to 100 µl of sperm suspension. The percentage motility was then assessed. The second approach was performed by abruptly adding 100 µl of a given CPA to 100 µl of 114 removal was performed at 22 C. A maximum of nine donors was used for the control (1 M glycerol) and a minimum of three donors was used for the removal of 1 M and 2 M ethylene glycol.

Addition and removal of cryoprotectants from spermatozoa Figure 3. Experimental results: motility recovery (mean SEM) Figure 4. Experimental results: motility recovery (mean SEM) of human spermatozoa after experimental treatments involving slow of human spermatozoa after experimental treatments involving addition or removal of cryoprotective agents. Different letters abrupt addition with or without removal of cryoprotective agents. indicate a statistically significant difference between treatments Different letters indicate a statistically significant difference (P 0.05). between treatments (P 0.05). Experiment 3: addition of CPA plus cooling and warming The two approaches used for CPA addition were those described in 2 M glycerol, 2 M ethylene glycol, or 4 M ethylene glycol at Experiment 1; however, before assessing motility, the samples were 22 C (Figure 3). The data indicate that there was not a cooled and rewarmed. The samples containing the given CPA were statistically significant difference in sperm motility (P 0.05) placed in 250 µl straws and cooled in a programmable cooling freezer between the slow addition of 1 M ethylene glycol (88.3 (Planar Products Ltd., Perkasie, PA, USA) using the following 2.6%, n 6) and 1 M glycerol (82.0 2.3%, n 9); protocol: 3 C/min from 25 to 5 C; hold at 5 C for 10 min; after 3 min at 5 C, samples were seeded by touching the straws for 1 s however, the motility after slow addition of 2 M ethylene with forceps cooled in liquid nitrogen; 10 C/min from 5 down to glycol (73.2 %, n 9) was significantly lower (P 0.05) 80 C; lastly, the straws were plunged into liquid nitrogen ( 196 C). than after slow addition of 1 M glycerol or 1 M ethylene The straws were warmed at ~400 C/min by placing them directly on glycol. Abrupt addition resulted in similar findings (Figure 4), the benchtop (22 C). Motility was then assessed. A maximum of and the results for 1 M glycerol and 1 M ethylene glycol were nine donors was used for the control (1 M glycerol) and a minimum of not significantly different (P 0.05) (80.5 2.2%, n 4 and four donors was used for the addition of 1 M and 2 M ethylene glycol. 87.3 1.7%, n 7 respectively). The abrupt addition of 2 Experiment 4: removal of CPA after cooling and warming M ethylene glycol resulted in the greatest loss of motility The same two approaches were used for the removal of CPA as (67.6 2.3%, n 7). described in Experiment 2, and the same protocol for freezing was used as described in Experiment 3. A maximum of nine donors was used for the control (1 M glycerol) and a minimum of three donors Experiment 2: removal of CPA was used for the removal of 1 M and 2 M ethylene glycol. Motility recovery was measured after 1 M glycerol, 1 M ethylene glycol, or 2 M ethylene glycol was slowly removed Statistical analysis from the cells (Figure 3). After CPA removal, 1 M ethylene The percentage of motile spermatozoa was measured after each glycol had the least effect on sperm motility (90.8 3.3%, treatment and was normalized to an untreated control (treatment n 6). Use of 1 M glycerol (70.4 2.8%, n 9) and 2 M motility/control motility 100). The data are presented as normalized ethylene glycol (50.0 5.3%, n 9, after addition of values. The mean, standard error and coefficient of variation were computed for each treatment and the data were analysed using 575 mosmol/kg sucrose, and 65.3 5.8%, n 3, after standard analysis of variance approaches with the Statistical Analysis returning to iso-osmotic conditions) both caused significantly Software (SAS ) program, a software system for data analysis (SAS lower recoveries of sperm motility (P 0.05). Motility was Institute Inc., Cary, NC, USA). also assessed when CPA was abruptly removed (Figure 4): motility after the use of 1 M glycerol (55.2 2.4%, n 7) Results and 2 M ethylene glycol (33.9 4.8%, n 7, sperm motility after addition of 575 mosmol/kg sucrose, and 44.2 1.4%, Experiment 1: addition of CPA n 3, after return to iso-osmotic conditions) was significantly The percentage of motile human spermatozoa was calculated lower (P 0.05) than for 1 M ethylene glycol (92.4 1.4%, after aliquots of cell samples were slowly exposed to either n 4). 115

J.A.Gilmore et al. Figure 5. Experimental results: motility recovery (mean SEM) of human spermatozoa after experimental treatments in which spermatozoa were frozen in conjunction with slow addition with or without removal of cryoprotective agents. Different letters indicate a statistically significant difference between treatments (P 0.05). Figure 6. Experimental results: motility recovery (mean SEM) of human spermatozoa after experimental treatments in which spermatozoa were frozen in conjunction with abrupt addition with or without removal of cryoprotective agents. Different letters indicate a statistically significant difference between treatments (P 0.05). Experiment 3: addition of CPA plus freezing 575 mosmol/kg sucrose, and 31.4 3.0%, n 3, after return Human sperm motility recovery was measured after 2 M to iso-osmotic conditions) was used. glycerol, 2 M ethylene glycol, or 4 M ethylene glycol was added and the cell CPA suspension was cooled to 80 C, Discussion plunged into liquid nitrogen, and warmed to room temperature. Figure 5 shows normalized percentage motility when CPA Most cells are lethally damaged during the cooling and addition was slow, with 1 M ethylene glycol maintaining the warming processes of cryopreservation in the absence of highest motility (62.4 4.1%, n 6) and 1 M glycerol protective agents (Ashwood-Smith and Farrant, 1980). There- (45.5 3.0%, n 9) and 2 M ethylene glycol (33.0 3.6%, fore, CPA are required for maintaining cell survival after n 9) causing a significantly greater loss of motility (P 0.05). cooling and warming. After Polge et al. (1949) published their Motility was also assessed when CPA addition was abrupt, findings regarding the protective characteristics of glycerol for using the same freezing protocol (Figure 6). The highest bovine spermatozoa and later with red blood cells (Polge, percentage of motility was maintained with 1 M ethylene 1980), others began to investigate similar low-molecular- glycol (36.9 2.5%, n 4) and a significantly lower recovery weight penetrating compounds and their ability to protect (P 0.05) was obtained with 1 M glycerol (22.7 1.1%, n cells during cryopreservation. In 1959, Lovelock and Bishop 7) and 2 M ethylene glycol (20.2 2.2%, n 7). introduced DMSO as a cryoprotectant for red blood cells. Both glycerol and DMSO have been studied in depth since then, and their uses have been applied to many different cell and Experiment 4: removal of CPA after freezing tissue types. Additionally, other permeating polyols, such as Percentage motility of human spermatozoa was measured ethylene glycol and propylene glycol, first studied by Lovelock after the cells were cooled, warmed, and 1 M glycerol, 1 M (1954) with human red blood cells, have come into wider use ethylene glycol, or 2 M ethylene glycol was removed. Figure and their cryopreservation effects have been explored. 5 illustrates sperm recovery when CPA removal was slow. Use The mechanism by which these permeating solutes are able of 1 M ethylene glycol maintained the highest recovery of to protect during cooling and warming has been closely motility (67.7 3.1%, n 6) and 1 M glycerol (39.1 examined. It is understood that permeating CPA function 2.6%, n 9) and 2 M ethylene glycol (28.6 1.8%, n 9, through colligative properties which help maintain solute after addition of 575 mosmol/kg sucrose, and 43.2 3.0%, concentrations inside and outside the cells at non-damaging n 3, after return to iso-osmotic conditions) led to a levels. It is also recognized that the cryoprotective ability of significantly lower motility (P 0.05). Motility was also any given compound varies widely across cell and tissue type assessed after CPA was removed abruptly (Figure 6). Sperm and, therefore, an optimal CPA for cryopreservation must be recovery was highest after removal of 1 M ethylene glycol studied for each cell type under investigation (Karow, 1969). (50.6 4.9%, n 4; Figure 6). Motility was significantly Although CPA enhance cell survival after cooling and less (P 0.05) when 1 M glycerol (21.4 1.3%, n 7) or warming, their presence induces potentially damaging volume 2 M ethylene glycol (11.6 2.7%, n 7, after addition of excursions. Lovelock (1953a,b) first described how this effect 116

Addition and removal of cryoprotectants from spermatozoa as compared to ethylene glycol, because glycerol will require more time to leave the cell and thereby allow more water to permeate the membrane. It should also be noted that the permeability rates used to predict theoretical optimal CPA addition and removal (Gilmore et al., 1995) are derived from the use of 1 M glycerol and 2 M ethylene glycol, and not 1 M ethylene glycol. It is possible that permeability coefficients are concentration dependent. If permeability is affected by concentration, permeability rates would differ from those published when using other solute concentrations. Prior studies (Papanek, 1978) with red blood cells have shown that water permeability is directly affected (decreased) by external solute conditions. Papanek (1978) suggested that, with higher solute concentration, water permeability in the red blood cell is substantially reduced and, therefore, would affect the rate at which the cells would be optimally cooled and warmed. Work is currently ongoing to determine concentration dependence of the effect of CPA on human spermatozoa. In addition to the greater membrane permeability to ethylene glycol, preliminary studies with human spermatozoa have also suggested that the temperature dependence of ethylene glycol and its associated water permeability are lower than those of glycerol. This lower activation energy for ethylene glycol indicates that membrane transport of ethylene glycol is less affected by changes in temperature compared to that of glycerol. Further investigation is necessary to determine if the mechanism for membrane transport across the plasma membrane differs for the two solutes. There are several possible explanations for the poor motility recovery when using 2 M ethylene glycol with human spermatozoa. It may be that this concentration is chemically toxic to human spermatozoa. Although ethylene glycol has been found to be relatively non-toxic to many cell and tissue types such as bovine embryos (Pollock et al., 1991) and mouse morulae (Ali and Shelton, 1993), it has been found to be toxic to other tissues, such as rabbit common carotid arteries (Wusteman et al., 1996). The length of time and temperature of exposure would be a relevant issue in determining toxicity effects, and more investigation may be needed to understand further the effects of 2 M ethylene glycol on human sper- matozoa. Another possible explanation for the decrease in cell survival when using 2 M ethylene glycol may be related to the initial cell volume. All media were prepared on a molar (moles of solute/litre of solution) basis. Therefore, 2 M ethylene glycol was prepared by volumetrically measuring 2 M ethylene glycol and adding it to iso-osmotic TL-HEPES to achieve the final volume. As a result, the salt concentration of the TL-HEPES is diluted by the CPA and, therefore, the osmolality of the TL- HEPES that the spermatozoa are exposed to is 285 5 mosmol/kg. Specifically, the osmolality of the TL-HEPES would be ~237 mosmol/kg in a 2 M ethylene glycol solution. Theoretical simulations predict that during removal of 2 M ethylene glycol in hyposmotic TL-HEPES, human sperm cell volume would increase to 1.38 times its iso-osmotic cell volume, exceeding its upper osmotic tolerance limit. According to Gao et al. (1995), human spermatozoa are expected to lose ~60% motility when their volume exceeds 1.38 times their 117 of solutes during cryopreservation results in lethal cell damage and how, in addition, the damage related to the mechanics of freezing. The present study was performed to determine which CPA would minimize cell volume excursion during cooling and warming and, thus, would be optimal for cryopreservation of human spermatozoa. A series of experiments investigated four different cryoprotectants for human spermatozoa cryopreservation: (i) glycerol, (ii) DMSO, (iii) propylene glycol and (iv) ethylene glycol. First, theoretical predictions of how cells respond to the addition and removal of CPA were made using computer simulations. Data regarding permeating rates of the solute and water across the sperm plasma membrane (Gilmore et al., 1995) and information regarding the osmotic tolerance limits of the cells (Gao et al., 1995) were used for the simulations. The data indicated that the optimal CPA would be one that could permeate the cell in the shortest period of time causing the least amount of volume excursion during its addition and removal. The theoretical predictions were then experimentally tested. Based on the computer simulations regarding the maintenance of minimum and maximum cell volume, 1 M and 2 M ethylene glycol were chosen as the optimal CPA for human spermatozoa cryopreservation. The results of the experiments confirmed that the use of 1 M ethylene glycol during cooling and warming of human spermatozoa results in higher cell survival in comparison to 1 M glycerol. These data indicate that, regardless of slow or abrupt addition or removal, the percentage of motile cells is greater when using 1 M ethylene glycol. However, in contradiction to the theoretical predictions, 2 M ethylene glycol did not result in greater cell survival during cryopreservation. When compared to 1 M ethylene glycol and 1 M glycerol, 2 M ethylene glycol was shown to lead to significantly lower sperm motility (P 0.05) when added and removed abruptly after cryopreservation. However, when 2 M ethylene glycol was added and removed slowly after cryopreservation, it did not result in significantly different sperm motility as compared to 1 M glycerol. It should be noted that 2 M ethylene glycol resulted in a greater percentage of motile spermatozoa once the suspension was returned to iso-osmotic conditions. These data are consistent with those of Gao et al. (1995), which indicated that cells will maintain a lower percentage motility while suspended in hyper-osmotic media, but will recover some motility when returned to iso-osmotic conditions. This study illustrated through computer simulations and experimental results that CPA addition causes much less damage to the cell than CPA removal, and therefore, only further development of optimal dilution is presented here. Previous work (Gao et al., 1995) and the present study have both indicated that abruptly removing 1 M glycerol decreases cell motility by ~60%, which can be explained by examining the volume excursion during the removal procedure (Figure 2). Because human sperm membrane permeability to ethylene glycol is approximately four times greater than that to glycerol, while the associated water permeabilities are approximately the same (Gilmore et al., 1995), the maximum volume excursion during glycerol dilution is greater as compared to ethylene glycol dilution. In other words, human spermatozoa will sustain a greater increase in cell volume when glycerol is removed,

J.A.Gilmore et al. iso-osmotic volume. One method of preventing this increase in cell volume during exposure to a CPA is to prepare the CPA medium on a molal (moles of solute/kg of solvent, where the solvent is water) basis. This method adjusts for the dilution of salts when a CPA is present, and therefore the desired osmolality is maintained (Pegg, 1984). Acknowledgements The authors would like to thank Dr Locksley McGann (University of Alberta, Alberta, Canada) for helpful discussion of the manuscript, Jason Bisel for technical support, and Katherine Vernon for manuscript preparation. This work was supported by Methodist Hospital of Indiana, Inc., a Career Development Award from the NIH (HD00980 to J.K.C.), a grant from USDA/NRICGP (93 37203 9272) and a NATO Collaborative Research Grant (CRG 920170). Polge, C., Smith, A.U. and Parkes, A.S. (1949) Revival of spermatozoa after vitrification and dehydration at low temperatures. Nature, 164, 666 676. Pollock, G.A., Hamlyn, I., Macguire, S.H. et al. (1991) Effects of four cryoprotectants in combination with two vehicle solutions on cultured vascular endothelial cells. Cryobiology, 28, 413 421. Weast, R.C. (ed.) (1971) Handbook of Chemistry and Physics, 51st edn. The Chemical Rubber Co., Cleveland, OH. Wusteman, M.C., Busza, A., Boylan, S. et al. (1996) Toxicity of ethylene glycol when used as a cryoprotectant for rabbit common carotid arteries. Cryobiology, 17, 85 92. Received on July 15, 1996; accepted on October 12, 1996 References Ali, J. and Shelton, J.N. (1993) Design of vitrification solutions for the cryopreservation of embryos. J. Reprod. Fertil., 99, 471 477. Ashwood-Smith, M.J. and Farrant, J. (eds) (1980) Low Temperature Preservation in Medicine and Biology. Pitmans Medical, Tunbridge Wells. Bavister, B.D., Leibfried, M.L. and Lieberman, G. (1983) Development of preimplantation embryos of the golden hamster in a defined culture medium. Biol. Reprod., 28, 235 247. Critser, J.K., Huse-Benda, A.R., Aaker, D.V. et al. (1988) Cryopreservation of human spermatozoa. III. The effect of cryoprotectants on motility. Fertil. Steril., 50, 314 320. Du, J., Kleinhans, F.W., Mazur, P. and Critser, J.K. (1993) Osmotic behavior of human spermatozoa studied by EPR. Cryo-Lett., 14, 285 294. Du, J., Kleinhans, F.W., Mazur, P. and Critser, J.K. (1994) Human spermatozoa glycerol permeability and activation energy determined by electron paramagnetic resonance. Biochim. Biophys. Acta, 1194, 1 11. Gao, D.Y., Liu, J., Liu, C. et al. (1995) Prevention of osmotic injury to human spermatozoa during addition and removal of glycerol. Hum. Reprod., 10, 1109 1122. Gilmore, J.A., McGann, L.E., Liu, J. et al. (1995) Effect of cryoprotectant solutes on water permeability of human spermatozoa. Biol. Reprod., 53, 985 995. Karow, A. (1969) Cryoprotectant a new class of drugs. J. Pharm. Pharmacol., 21, 209 223. Kedem, O. and Katchalsky, A. (1958) Thermodynamic analysis of the permeability of biological membranes to nonelectrolytes. Biochim. Biophys. Acta, 27, 229 246. Kiyohara, O., Perron, G. and Desnoyers, J.E. (1975) Volumes and heat capacities of dimethylsulfoxide, acetone, and acetamide in water and of some electrolysis in these mixed aqueous solvents. Can. J. Chem., 52, 3263 3268. Leibo, S.P. (1986) Cryobiology: preservation of mammalian embryos. Basic Life Sci., 37, 251 272. Lovelock, J.E. (1953a) The haemolysis of human red blood cells by freezing and thawing. Biochim. Biophys. Acta, 10, 414 426. Lovelock, J.E. (1953b) The mechanism of the protective action of glycerol against haemolysis by freezing and thawing. Biochim. Biophys. Acta, 11, 28 36. Lovelock, J.E. (1954) The protective action of neutral solutes against haemolysis by freezing and thawing. Biochem. J., 56, 265 270. Lovelock, J.E. and Bishop, M.W.H. (1959) Prevention of freezing damage to cells by dimethyl sulphoxide. Nature, 183, 1394 1395. Mazur, P. and Schneider, U. (1984) Osmotic consequences of cryoprotectant permeability and its relation to the survival of frozen thawed embryos. Theriogenology, 21, 68 79. Papanek, T. (1978) The Water Permeability of the Human Erythrocyte in the Temperature Range 25 C to 10 C. Doctor of Philosophy thesis. Massachusetts Institute of Technology, Cambridge, MA. Pegg, D.E. (1984) Red cell volume in glycerol/sodium chloride/water mixtures. Cryobiology, 21, 234 239. Polge, C. (1980) Freezing of spermatozoa. In Ashwood-Smith, M.S. and Farrant, J. (eds), Low Temperature Preservation in Medicine and Biology. Pitman, Bath, pp. 45 64. 118