Determination of Plasma Membrane Characteristics of Boar Spermatozoa and Their Relevance to Cryopreservation'

BOLOGY OF REPRODUCTON 58, 28-36 (1998) Determination of Plasma Membrane Characteristics of Boar Spermatozoa and Their Relevance to Cryopreservation' J.A. Gilmore,34 J. Liu, 3 A.T. Peter, 4 and J.K. Critser2,3,4 Cryobiology Research nstitute 3 and Department of Veterinary Clinical Sciences, 4 School of Veterinary Medicine, Purdue University, West Lafayette, ndiana 4797 ABSTRACT The osmotic tolerance limits for boar spermatozoa were determined at 22 C. These cells can swell to within 1.2 times and shrink to within.97 times their isosmotic volume and maintain > 7% motility. n the presence of an extender, cells can swell to within 1.1 times and shrink to within.97 times their isosmotic volume and maintain > 7% motility. Plasma membrane permeability coefficients were determined in the presence of 1 M dimethyl sulfoxide (DMSO), 1 M glycerol, and 2 M ethylene glycol (EG) at 22 C. Hydraulic conductivity (Lp) was estimated to be.12.16 (mean + SEM),.138 ±.6, and.24 +.21 lim/min/atm in the presence of DMSO, glycerol, and EG, respectively, at 22 C. Solute permeability (PcPA) was determined to be.93 ±.118,.481 +.45, and 1.98 ±.16 X 1-3 cm/min, for DMSO, glycerol, and EG, respectively. Subsequent experiments were performed at 8C and C. Activation energies were calculated for L in the presence of glycerol and EG to be 7.2 and 11.51 Kcal/mol, respectively. The activation energies for PCPA were 4.6 and 7.48 Kcal/mol for glycerol and EG permeability, respectively. These membrane characteristics were used to calculate volume flux during addition and removal of cryoprotectant agents as well as during cooling and warming. n addition, the potential for intracellular ice formation during cooling and warming was calculated. NTRODUCTON Cryopreservation of boar spermatozoa offers an effective means for long-term storage of important genetic material [1]. This method eliminates the difficulty associated with transporting animals or fresh semen over long distances or extended periods of time. Successfully cryopreserved boar semen could improve pork production and its associated economic value internationally. However, current methods for cryopreservation are inadequate. To date, the use of cryopreserved boar semen often involves cumbersome processing procedures, and it yields low artificial insemination doses per ejaculate due to low cell survival, resulting in both low fecundity rates and litter sizes [2]. The current work is based upon the hypothesis that through a more thorough investigation of the fundamental cryobiological and biophysical properties of boar spermatozoa and how these properties are influenced during exposure to cryoprotectant agents (CPAs) and cooling and warming conditions, more optimal methods for boar semen cryopreservation can be developed. Prior studies with boar spermatozoa have determined such fundamental properties Accepted August 19, 1997. Received April 29, 1997. 'This work was supported by a Career Development Award from the NH (K4-HD98 to J.K.C.), a grant from the USDA/NRCGP (93-3723-9272 to J.K.C. and A.T.P.), and a NATO Collaborative Research Grant (CRG 9217 to J.K.C.). 2Correspondence: John K. Critser, Cryobiology Research nstitute, Purdue University, Potter Building, Rm. 24, West Lafayette, N 4797. FAX: (317) 494-1193; e-mail: critser@iquest.net 28 as the osmotically inactive cell fraction of the cell [3, 4], the hydraulic conductivity (Lp) [3, 5], and the cells' ability to regulate volume during preservation [6]. By combining this information with knowledge of the cells' ability to tolerate osmotically driven volume changes and their response to permeating CPAs, one can better predict optimal conditions for cooling and warming. During the several steps involved in the cryopreservation process, cells undergo changes in volume as water and solute enter and leave the cells. This osmotic response can be potentially lethal to the cells if it causes cells to swell or shrink beyond their osmotic tolerance limits [7, 8]. These limits can be determined by experimentally exposing cells to isosmotic and anisosmotic conditions; however, these experiments have not been performed using boar spermatozoa. Once these parameters have been identified, methods for CPA addition and removal can be estimated that would minimize cell volume excursion [9]. n addition to determining the osmotic tolerance limits for boar spermatozoa, optimizing methods for cryopreservation also requires defining the response of boar spermatozoa to permeating CPAs. By determining the rate at which CPAs and water permeate the boar sperm plasma membrane, and their temperature dependence, one can predict the amount of water loss and the potential for intracellular ice formation that cells will experience during cooling and warming. Thus the most optimal CPA and methods for its addition and removal can be estimated from this information. The objectives of this study were 1) to determine the osmotic tolerance limits of boar spermatozoa; 2) to determine the permeability of boar spermatozoa to CPAs (glycerol, dimethyl sulfoxide [DMSO], and ethylene glycol [EG]) and to water, as well as their temperature dependence (Ea); 3) to predict an optimal CPA type and appropriate method for its addition and removal; and 4) to predict the amount of water loss experienced during cooling and warming and the associated probability of intracellular ice formation at various cooling and warming rates. MATERALS AND METHODS Animals Three mature Yorkshire boars (18 mo to 3 yr old, weighing 24-273 kg) were used in this study. The animals were fed, in an amount based on body weight, once a day with a mixed corn-soybean ration and were allowed free access to water. The boars could move at will inside a barn with straw on the floor or outside on a concrete slab. Media The media used were isosmotic Tyrode's lactate Hepes (TL Hepes)-buffered solution (285 5 mosmol/kg) [1], and isosmotic and anisosmotic PBS solutions. Hyposmotic

BOAR SPERM PLASMA MEMBRANE CHARACTERSTCS 29 TABLE 1. Composition of Modena extender.* TABLE 2. Definitions of major symbols used in equations. Concentration ngredient Grams/liter (M) Glucose monohydrate anhydros 27.5.139 Sodium citrate 6.9.28 Sodium bicarbonate 1..12 Ethylene dinitro tetraacetic acid 2.35.7 (EDTA), disodium Tris, buffer 5.65.47 Citric acid 2.9.15 *See [111. PBS solutions (< 29 mosmol/kg) were prepared by diluting isosmotic PBS with reagent water, and hyperosmotic PBS solutions (> 29 mosmol/kg) were prepared by diluting 1-strength PBS with reagent water. Osmolality was determined using a freezing-point depression osmometer (Model 3D2; Advanced nstruments, Needham Heights, MA). The cryoprotectant treatment solutions were prepared by mixing glycerol, DMSO, or EG to yield final CPA concentrations of 1 M (glycerol and DMSO) and 2 M (EG). All media used in the experiments were obtained from Sigma Chemical Company (St. Louis, MO). Modena extender was used in experiments 1 and 3. The extender was prepared as described by Weitze [11] using the components shown in Table 1. Samples Semen samples were collected once a week by manual manipulation using a dummy and gloved hand. Filtered semen from the sperm-rich portion was diluted 1:3 in TL Hepes media in 15-ml conical tubes (Sarstedt, Newton, NC) and transported in a 37 C water bath during the 1-h transport time. Five microliters of the sample was analyzed using computer-assisted semen analysis (CASA; Cell Soft, Version 3.2/C; CRYOResources Ltd., Montgomery, NY) to determine percentage of motility and concentration. Motility was analyzed using this approach throughout experiments 1 and 3. Electronic Particle Counter A Coulter Counter (ZM model; Coulter Electronics, nc., Hialeah, FL) with a standard 5-im aperture tube was used for all volume measurements. Sperm cell volume was calibrated using spherical styrene beads (Duke Scientific Corporation, Palo Alto, CA) with a diameter of 3.98 +.3 pm (mean + SEM) and a volume of 33.1 ixm 3. Data Analysis Membrane permeability coefficients. The Kedem Katchalsky mass transport model [12] was applied to determine the permeability of boar spermatozoa to CPAs. The following equations describe the total transmembrane volume flux (J) and permeable solute flux () for a ternary solution composed of a permeable solute (glycerol, DMSO, or EG, subscript "s") and an impermeable solute (NaCl, subscript 1 dv Adt = Jv = Lp[(Pe - Pi) - RT(M e - Mi) dn A dt - ort(m - M)] (1) i 1 =-- Js = (1 - r)/tjv + P,(a e - a), (2) Symbol Description Units Value A surface area of cell Lm 2 parameter E, activation energy Kcal/mol parameter e,i Superscripts (e = external, - i = internal to cell) Lp hydraulic conductivity am/min/atm parameter m molality moles/kg H 2 O variable M% osmolality of external per- Osmol/kg H 2.29 meating salts Mi' osmolality of internal per- Osmol/kg H 2 O variable meating salts PCPA CPA permeability cm/min parameter R universal gas constant Kcal/mol/K 1.987 x 1-3 s,n Subscripts (s = solute, n - nonpermeating salts) t time seconds variable T temperature K variable V(t) volume of cell jim 3 variable Vb osmotically inactive cell pim 3 parameter volume V, partial molar volume of liter/mol solute DMSO.69 Glycerol.71 EG.54 r reflection coefficient - parameter where V is cell volume and n is the number of moles of solute (glycerol, DMSO, or EG) inside the cell, P is the hydrostatic pressure, M is osmolality and a is activity. Table 2 details the description for units, values, and symbols. n the term,hs = (me- m)/(lnme - nm), m is the concentration in molal. The superscripts "i" and "e" refer to the intra- and extracellular cell compartments, respectively. The terms Lp, PCPA, and (r are parameters for the hydraulic conductivity of water, the permeability coefficient of the solute, and the reflection coefficient, respectively. The temperature and universal gas constant are given by T and R, respectively. The relation between n and mi is given by: n. = (V- Vb- VO)m i, (3) where (V -Vb- Vs) denotes the volume of intracellular water. By combining equations 1, 2, and 3 and assuming that the hydrostatic pressure difference across the membrane is equal to zero, the following coupled nonlinear equations describing the cell volume and amount of solute in the cell as functions of time can be applied: and dv d = LpART[(Mi - e ) M + (r(mi - me)] V m r(1 - ) - m i ]dv dt V- Vb (1 + Vms)] dt + APs[(m -m)]. (4) (5) Activation energies for parameters L and PCPA. The Arrhenius relationship was used to determine the temperature dependence of the parameters Lp and PCPA [13]. The value of any parameter P, (Lp or PcPA) at any temperature T can be obtained by the following formula:

3 GLMORE ET AL. W E U Q ~5 C4 2B. 11 1-9 - 8 7 6-5 - A~ 4{l MU '4 1 2 3 4 Time (seconds) FG. 1. Kinetic volume changes of boar sperm upon dilution of permeant solute. The sample was exposed to EG and allowed to equilibrate at room temperature. The sample was then abruptly diluted in an isosmotic solution without CPA, and the volume was measured over time. or in another form: P,(T) = P.exp (T T] E= ln[pa(t)] - + Constant, RT (6a) (6b) where Ea is the activation energy for the process, expressed here in Kcal/mol, R is the universal gas constant, and T is the absolute temperature. The subscript "o" represents the values at a reference temperature T,. From equation 6b, data collected at different temperatures are plotted as ln[pa(t)] vs. /T, a linear plot, (Arrhenius plot), and the slope is defined as: slope = - from which Ea can be determined. Data Acquisition :4 *1 *A * :* E,/R, The electronic particle counter was interfaced to a microcomputer using a CSA-1 interface (The Great Canadian Computer Company, Edmonton, AB, Canada). Changes in cell volume were measured over time, as shown in Figure 1, and a commercial software package, MLAB (Civilized Software, nc., Bethesda, MD), was used to solve equations 4 and 5 using the Gear method [14]. The Marquard-Levenberg curve-fitting method [14], as implemented in MLAB, was used to fit the experimental data and determine the values of Lp and PCPA. A fixed value for Vb [3] was used in the fitting calculation, and the reflection coefficient, ar, was assumed to be noninteracting and was calculated using the following equation: (a = (1 - PCPAVcPA)/(RTLp). (7) (8) Statistical Analysis Using standard analysis of variance approaches [15] with the SAS program (SAS nstitute nc., Cary, NC), data were analyzed, and the mean, standard error, and coefficient of variation were computed among donors. Experimental Design Experiment (a-d): Osmotic tolerance of boar spermatozoa. Experiment la was conducted in anisosmotic solutions of nonpermeating solutes. Processed cell suspensions were exposed to control isosmotic (285 + 5 mosmol/kg) PBS solutions and to anisosmotic PBS solutions with the following osmolalities: 35, 75, 15, 6, 12, and 24 mosmol/kg. Ten microliters of cell suspension was transferred into 15 pr1 of anisosmotic solution in one step. The suspension-solution mixture was gently mixed, and the samples were allowed to equilibrate for 1 min before motility was assessed. A sample size of 3 boars (n = 3) was used for this experiment, and all data were collected at 22 C. Experiment lb was designed for return to isosmolality after exposure to anisosmotic conditions. One hundred microliters of cell suspension was added to 2 dl of anisosmotic solutions, as described above. After 1 min, the cells were returned to near-isosmotic conditions by the addition of 15 p. 1 of isosmotic PBS. Motility was assessed after the return to isosmolality for 1 min and again at 5 min. Experiment lc exposed cells to a narrower osmotic range of anisosmotic conditions. Exposure to anisosmotic conditions was performed as described above using the following osmolalities of PBS: 14, 22, 26, 28, 29, 3, 32, 36, 44, and 6 mosmol/kg. After mixing, the suspension was allowed to equilibrate for 1 min before motility was assessed. A total of three boars (n = 3) were used and all data were collected at 22 C. Experiment d was designed for return to isosmolality after exposure to anisosmotic conditions. After exposure, cells were returned to near isosmolality by adding 15 p. 1 of isosmotic PBS. Motility was assessed after 1 min and after 5 min. Experiment 2. Activation energy for boar spermatozoa solute and water permeability. One hundred microliters of 2 M glycerol, 2 M DMSO, or 4 M EG was added drop by drop over 6 sec to 1 p.l of cell suspension, yielding final CPA concentrations of 1 M and 2 M, respectively. Cells were allowed to equilibrate for approximately 3 min before all 2 l were returned to isosmotic media. Changes in cell volume were measured over time. A total of 3 donors (n = 3) were used, and each experimental run was replicated three times for each donor. The experiments were performed at 22 C, 8 C, and C in the presence of glycerol and EG, and at 22 C for DMSO. Experiment 3. Osmotic tolerance limits of boar spermatozoa in the presence of extender. The same procedures were performed as described in experiment lc; however, the spermatozoa were suspended approximately 1:3 in Modena extender. Experiment 4. Activation energy for boar spermatozoa glycerol and water permeability in the presence of extender. One hundred microliters of 2 M glycerol were added drop by drop over 6 sec to 1 pl of cell suspension diluted in Modena extender, yielding a final CPA concentration of 1 M glycerol. Cells were allowed to equilibrate for approximately 3 min before the 2 1 was returned to isosmotic media. Changes in cell volume were measured

BOAR SPERM PLASMA MEMBRANE CHARACTERSTCS 31..N Sn Z 1 8 6-4 - 2 + n anisosmotic Returned to isosmotic after min Returned to isosmotic after 5 min TABLE 3. Percentage of boar spermatozoa motility is given as a function of the upper osmotic tolerance limit (UTL) and lower osmotic tolerance limit (LTL), given in percentage of isosmotic cell volume, in the presence and absence of Modena extender. Absence of extender Presence of extender % Motility UTL LTL UTL LTL n anisosmotic conditions 9 1.1.99 1.1.99 8 1.1.98 1.7.98 7 1.2.97 1.11.97 After return to isosmotic conditions for 1 min 9 1..99 1.1.99 8 1.1.98 1.5.98 7 1.2.97 1.1.97 After return to isosmotic - conditions for 5 min, 4 9 1..99 1.1.99 2 3 4 5 6 8 1.1.98 1.5.98 7 1.1.97 1.9.97 1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~1 1 - these assumptions are incorrect, the results could potential- 8 ly increase cell loss. 6 4 2-1 2 3 4 5 6 Osmolality FG. 2. Normalized percentage of motility (mean SEM, n = 3) of boar spermatozoa that were abruptly exposed to different anisosmotic conditions (circles) and returned to isosmotic conditions after 1 min (squares) and after 5 min (triangles) at room temperature (A) or in the presence of Modena extender at room temperature (B). over time. A total of 3 donors (n = 3) were used, and each experimental run was replicated three times for each donor. The experiments were performed at 22 C, 8C, and C in the presence of glycerol and EG, and at 22 C for DMSO. Experiment 5. Simulation of cryoprotectant addition and removal. Computer simulations of glycerol and EG addition and removal to and from boar spermatozoa were performed using methods described by Gilmore et al. [9]. Experiment 6. Simulation of intracellular water volume flux during cooling and warming. Computer simulations of intracellular water volume flux during the cooling and warming of boar spermatozoa were performed using methods described by Liu et al. [16]. The theoretical predictions of water loss during cooling and warming were based upon the determined cryobiological parameters of boar spermatozoa and made the following assumptions: 1) above-zero activation energies were extrapolated to subzero temperatures, 2) a generic phase diagram was used for both glycerol and EG, and 3) the predictions assumed that optimal methods for CPA addition and removal were used. f any of RESULTS Experiment 1: Osmotic Tolerance Limits for Boar Spermatozoa Figure 2 demonstrates the relative percentage of motile cells as a function of osmolality. Motility is evaluated in anisosmotic conditions, and 1 and 5 min after return to isosmolality. The results indicate that there was no significant effect of treatment (cells in anisosmotic conditions vs. returned to isosmotic conditions; p =.943) on motility. Table 3 summarizes the percentages of volume excursions cells can tolerate during swelling and shrinking while still maintaining 9%, 8%, and 7% motility, respectively. Experiment 2: Activation Energy for Boar Spermatozoa Solute and Water Permeability Changes in cell volume, measured over time, were fitted to compute PCPA and Lp at 22 C, 8C, and C; results can be seen in Table 4. Hydraulic conductivity (L ) was estimated to be.12 ±.16 (mean + SEM),.138 +.6, and.24 +.21 ipm/min/atm in the presence of DMSO, TABLE 4. Boar spermatozoa permeability characteristics (mean + SEM). L, PCPA Solute (tm/min/atm) (1 3 cm/min) o- (n)* Glycerol 22 C.138 ±.6'.481 ±.45'.896 _.11' 3 8 C.72 +. 5 b.316 +. 2 9 b.852 ±.28, 3 C.52 +.6c.279 +.28 c.812 ±.26, 3 Ethylene glycol 22 C.24 ±.21, 1.98 +.16'.763 ±.34, 3 8 b C.92 +. 5 b 1.8 ±. 6 8.719 1.17' 3 O C.41 ±.3 :.74 ±.33 c.561 ±.41 b 3 DMSO 22'C.12 ±.16.93 ±.118.724 +.77 3 * Number of animals used. -' Values with different superscripts are different (p <.5) within CPA and within columns.

1 32 GLMORE ET AL., -2-3 -4-5 -6-7. 3.335-1 -4-5 -6-7 -8-9 3.35 3.4 3.45 3.5 3.55 3.6 3.65 3.7( -1-2 -3-4 -5-6 Lp: Ea = 11 5 Kcal/mol R2=.98 PEG: Ea = 7 5 Kcal/mol R2=.99 3.4 3.55 3.6 3.65 3.4 3.45 3.5 3.55 3.6 3.65 3.7 LP: Ea= 7.8 Kcal/mol R 2 =.99 PGilCr_: E- 4.1 Kcal/mol R 2 =.97 _-P--~~~~ '---i------+~ Lp: E a = 6.9 Kcal/mol in presence of extender R2=.99 PGlyrl: R2=.95 - Ea - 4.1 Kcal/mol in presence of extender -7-8 X [-- - J [ 3.35 3.4 3.45 3.5 3.55 3.6 3.65 3.7 1/Temperature (K) FG. 3. An Arrhenius plot of the L (circles) and P,, (A) or Pgycero (squares) (B) of boar spermatozoa. C) Same as A but in the presence of Modena extender. Permeability values were determined at 22 C, 8 C, and C (n = 3 at each temperature). glycerol, and EG, respectively, at 22 C. Cryoprotectant permeability (PCPA) was determined to be.93.118,.481.45, and 1.98.16 x 1 3 cm/min for DMSO, glycerol, and EG, respectively, at 22 C. The reflection coefficients were estimated to be.724.77,.896.11, and.763 +.34 for DMSO and water, glycerol and water, and EG and water, respectively. Subsequent experiments were performed at 8C and C to determine the activation energy for glycerol, EG, water and glycerol, and water and EG permeability. Hydraulic conductivity (Lp) was estimated to be.72.5 and.52 ±.6 pxm/min/atm in the presence of glycerol at 8C and A B C C, respectively; and.92.5 and.41 +.3 Rpm/min/atm in the presence of EG at 8 C and C, respectively. Cryoprotectant permeability (PCPA) was determined to be.316 +.29, and.279.28 x 1 3 cm/min for glycerol, at 8C and C, respectively, and 1.8 +.68 and.74 +.33 x 1-3 cm/min for EG, at 8C and C, respectively. Reflection coefficients were determined to be.852 +.28, and.812 +.26, for glycerol and water at 8C and C, respectively. Reflection coefficients were determined to be.719.17 and.561.41 for EG and water at 8C and C, respectively. The activation energies (Ea) for Lp in the presence of glycerol and EG were estimated, using the Arrhenius relationship, to be 7.8 and 11.51 Kcal/mol, respectively. The estimated activation energies for PCPA were 4.6 and 7.48 Kcal/mol for glycerol and EG, respectively, and can be seen in Figure 3. Experiment 3: Osmotic Tolerance Limits of Boar Spermatozoa in the Presence of Extender Figure 2 shows the relative percentage of motile cells as a function of osmolality. Motility was evaluated in anisosmotic conditions, and 1 and 5 min after return to isosmolality. Results indicate that there was no significant effect of treatment (p =.849) on motility. However, the results did indicate that there is a significant interaction (p =.1) between extender and osmolality. Table 3 summarizes the percentages of volume excursions cells can tolerate during swelling and shrinking while still maintaining 9%, 8%, and 7% motility, respectively. Experiment 4. Activation Energy for Boar Spermatozoa Glycerol and Water Permeability in the Presence of Extender Glycerol permeability and its associated Lp was determined in the presence of Modena extender at 22 C, 8 C, and C; the results can be seen in Table 5. Hydraulic conductivity was.156.31,.86 +.13, and.64 +.9 prm/min/atm at 22 C, 8C, and C, respectively. Solute permeability was determined to be.681.711,.473 +.566, and.293.663 X 1-3 cm/min, at 22 C, 8 C, and C, respectively. The reflection coefficients were estimated to be.865.25,.825.46, and.84 +.5 at 22 C, 8C, and C, respectively. Activation energy for hydraulic conductivity was estimated to be 6.88 Kcal/mol and 5.95 Kcal/mol for glycerol permeability (Fig. 3). Experiment 5: Simulation of Cryoprotectant Addition and Removal Computer simulations of the addition and removal of glycerol (Fig. 4) and EG (Fig. 5) show the resulting relative cell volume change over time at room temperature. The results indicate that glycerol causes more cell volume ex- TABLE 5. Boar spermatozoa permeability characteristics in the presence of Modena extender (mean SEM). Temperature L, Pglycrao1 ( C) (LJ.m/min/atm) (1 3 cm/min) o- (n)a 22.156 ±.31.681.711.865 +.25 3 8.86.13.473.566.825 +.46 3.64 ±.9.293.663.84 +.5 3 Ea (Kcal/mol) 6.88 5.95 a Number of animals used.

BOAR SPERM PLASMA MEMBRANE CHARACTERSTCS 33 1.4 1..96 -- 1 Step min -- 3 min - 5 min -- 7 min - 9 min 1.2 1. Step min 3 min - -5 min --- 7 min 9 min U..92.88 1.2 1 2 3 4 5 6 2 2 V.98.96 ) - a8 v,.94 UM 1.1 V) P4 1.8 1.16 1.6 1.12 1.8 1.4 1.4 1.2 1. 1 2 3 4 5 6 Time (seconds) FG. 4. Computer simulation for the addition (A) and dilution (B) of 1 M glycerol to boar spermatozoa at 22 C. Addition procedures considered a 1:1 dilution of cell suspension to CPA. Addition rates, resulting in different cell responses, varied between and 9 min. Dilution procedures considered a 9:1 dilution of isosmotic media to CPA/cell suspension. Dilution rates varied between and 9 min. The horizontal bar refers to the osmotic limits within which 8% sperm motility could be maintained. 1. 1 2 3 4 5 6 Time (seconds) FG. 5. Computer simulation for the addition (A) and dilution (B) of 1 M EG to boar spermatozoa at 22 C. Addition procedures considered a 1: 1 dilution of cell suspension to CPA. Addition rates, resulting in different cell responses, varied between and 9 min. Dilution procedures considered a 9:1 dilution of isosmotic media to CPA/cell suspension. Dilution rates varied between and 9 min. The horizontal bar refers to the osmotic limits within which 8% sperm motility could be maintained. cursion during its addition and removal than does EG, at room temperature. Experiment 6. Simulation of ntracellular Water Volume Flux during Cooling and Warming Computer simulations of intracellular water volume excursions experienced during the cryopreservation process can be seen in Figure 6. Data are shown as relative cell water volume as a function of temperature in degrees centigrade until C is reached, at which point it becomes a function of time in seconds. The results indicate that both glycerol and EG could be used effectively to cryopreserve boar spermatozoa when cooling rates of 1 C/min and 1 C/min, respectively, are used, with a warming rate of 12 C/min. However, EG is ineffective for boar spermatozoa cryopreservation due to the increased cell water volume excursion when cooling rates exceed 1 C/min with a similar warming rate. The potential for intracellular ice formation remained low (< 5%) for each CPA across all three cooling rates. DSCUSSON Osmotic Tolerance of Boar Spermatozoa During the cryopreservation process, cells are exposed to multiple steps in which they experience volume excursions. During CPA addition, cells transiently shrink as water flows out of the cell to a hyperosmotic environment, and subsequently, the cells swell as water and CPA enter. The reverse occurs during CPA removal. The cells initially swell as water enters, and subsequently shrink as water and CPA leave the cell. These volume excursions induced by the addition and removal of CPAs are separate and distinct from those experienced during the actual cooling and warming processes of cryopreservation. As a result, these two steps of cryopreservation can both be potentially osmotically damaging to cells [7]. By determining the osmotic limits cells can endure, one can better minimize the damage associated with volume excursion and therefore optimize cryopreservation. Little information is available regarding osmotic responses of boar spermatozoa. Current literature does indicate that boar spermatozoa behave as linear osmometers [3, 4] down to 185 mosmol/kg. The present study expanded this scope of understanding to include the extent of volume excursion these cells can tolerate (minimum and maximum volumes) while still maintaining functional integrity and motility. The results indicate that cells can tolerate shrinking and swelling only to.97 and 1.2 times their isosmotic cell volume and still maintain > 7% motility. These tolerance limits are broadened somewhat in the presence of an extender, which may be the result of plasma membrane modification or stabilization by the extender. Previous reports support the notion that extenders may interact with boar spermatozoa plasma membranes, allowing for extended cell viability. For example, it has been reported that certain extender components, such as BSA, can remove lipid breakdown products from membranes, thereby changing membrane permeability [11]. The ANDROHEP extender has been reported to allow boar spermatozoa to regulate

34 GLMORE ET AL. 1 1 8 8 6 4 2.N E Z 6 4 E 2 - -2-4 -6-8 -6-4 -2 1 2 3 4 5 U94 14 12 1 8 6 4 2 2 4 6 8 1 12 Osmolality FG. 7. The estimated normalized percentage motilities of human spermatozoa [7], mouse spermatozoa [8], and boar spermatozoa in the presence of extender (present paper) after exposure to anisosmotic conditions and return to isosmotic conditions. Figure 7 compares the osmotic tolerance limits for human sperm, mouse sperm, and boar sperm in the presence of an extender. - -2-4 -6-8 -6-4 -2 1 2 3 4 5 Temperature (C) Time (seconds) FG. 6. Cell water volume changes during cooling and warming of boar spermatozoa in the presence of glycerol (A) or EG (B) using cooling rates of 1 C/min (solid line), 1 C/min (dotted line), and 1 C/min (dashed/dotted line), and a warming rate of 12 C/min. cell volume over prolonged periods of time [6]. Extenders containing low-density lipoprotein fractions of egg yolk have been shown to modify protein components of the cells, thereby enhancing protection during storage [17]. Graham and Foote [18] revealed that media containing specific lipids, phosphatidylserine alone or in combination with phosphatidylcholine, can protect sperm during cooling. Further experiments should be conducted to determine the mechanism for extender protection of boar spermatozoa, and, more specifically, to determine what components of the extender offer this protection. n addition, a more thorough investigation should be conducted regarding the interaction between extender components and osmotic tolerance limits. n comparison to spermatozoa of other mammalian species, boar spermatozoa are highly sensitive to osmotic changes. Human spermatozoa have been shown to tolerate swelling to 1.1 times their isosmotic volume and shrinking to.75 times their isosmotic volume and still maintain 9% motility [7]. Studies of mouse spermatozoa indicate that they are as sensitive as human spermatozoa but more tolerant than boar spermatozoa to osmotically driven changes. Mouse sperm can shrink and swell to.76 and 1.24 times their isosmotic cell volume, respectively, and still maintain 8% motility [8]. However, boar spermatozoa can tolerate swelling and shrinking only to 1.2 and.97 times their isosmotic cell volume and still maintain > 7% motility. Activation Energy for Boar Spermatozoa CPA and Water Permeability The CPAs described in this paper are thought to protect cells through their colligative properties, which maintain internal and external cellular solute concentrations at tolerable levels [19]. However, their presence can be damaging, and what may be protective for one cell type may not be appropriate for another. For example, glycerol successfully cryoprotects bull spermatozoa; however, the same procedures for the cryopreservation of boar spermatozoa have not been as successful [2]. t has been demonstrated in human spermatozoa that an optimal CPA is one that can permeate the cell rapidly, thereby minimizing cell volume excursion, with low temperature dependence (activation energy) and low toxicity to the cell [9]. The present study focused on DMSO, glycerol, and EG, and their ability to permeate boar spermatozoa. Cells were exposed to the CPAs, and changes in cell volume were measured over time. A two-parameter model was used to fit the permeability coefficients. The reflection coefficient was calculated, and a fixed value for Vb was chosen [3], versus fitting Vb as done in other studies [21]. DMSO was studied at room temperature only. The results indicated that DMSO permeates the boar sperm plasma membrane relatively slowly, and because of its infrequent use with boar sperm cryopreservation, no further investigation was performed. Because of glycerol's common use as a CPA with boar sperm and because of EG's relatively high permeability, these CPAs were chosen for further study. t was noted that although EG may permeate the cell at a higher rate, it also has a higher associated activation energy than glycerol. There appeared to be little statistical variation in solute and water permeabilities, with the exception of glycerol permeability at low temperatures, which were more variable. Nevertheless, the values obtained for glycerol and EG permeability, and their associated water

BOAR SPERM PLASMA MEMBRANE CHARACTERSTCS 35 permeabilities, are similar to those found in human spermatozoa [22], as well as in other cell types such as human red blood cells [23] and human platelets [24]. The presence of extender does not appear to change the permeability of the boar spermatozoa plasma membrane to glycerol or water (glycerol and water permeability values are similar in the absence or presence of Modena extender), and this finding is consistent across temperatures. t could be speculated that the modification or stabilization of the membrane by the extender may be similar to the modification present with CPAs. Simulation of Cryoprotectant Addition and Removal By combining the information formulated from the osmotic tolerance limits of boar spermatozoa along with membrane permeability characteristics, optimal procedures for CPA addition and removal can be determined that minimize osmotic volume excursion and maximize cell viability. Because EG permeates the boar spermatozoa plasma membrane more rapidly than does glycerol, there is less volume excursion during its addition and removal. However, because of the high degree of boar sperm sensitivity to osmotic changes, an abrupt single-step addition or removal of EG does exceed the estimated osmotic tolerance limits of boar spermatozoa. Computer simulations indicate that glycerol addition and removal is even more damaging than addition and removal of EG. Therefore, during CPA addition and removal, it will be necessary to add and remove the CPAs over time. Figures 4 and 5 illustrate how boar sperm cell volume excursion could be minimized by extending the time allotted for addition and removal procedures. Volume excursions are maintained well within the boar sperm's osmotic tolerance limits if glycerol is added drop by drop over a minimum of 5 min, and if EG is added drop by drop over a minimum of 3 min. Removal is similar, with glycerol requiring a 3-min dilution and EG requiring a 1-min dilution. Simulation of ntracellular Water Volume Flux during Cooling and Warming Fundamental cryobiology is an essential tool in determining optimal methods for boar spermatozoa cryopreservation. Recently there has been discussion in the literature questioning the applicability of fundamental cryobiology to sperm cryopreservation [25, 26]. This concern has focused on the apparent discrepancies between predicted theoretical approaches, based upon fundamental cryobiological parameters (e.g., membrane permeability characteristics and activation energy), and current empirical data. Although these previous discrepancies have been evident, they have not been due to the use of fundamental cryobiology but rather to a lack of available information. The human spermatozoan is an example of a cell type for which historically, theoretical data have not been viewed as empirically applicable. During some of the first investigations into the fundamental cryobiology of human spermatozoa [27], parameters were derived (e.g., water permeability) that, when used to determine cooling and warming rates, were not supported by empirical data [25, 28]. However, as the fundamental cryobiological investigations continued with this cell type, the discrepancy between theoretical predictions and experimental results became minimal. Gilmore et al. [22] reported data regarding the water permeability of human spermatozoa in the presence of CPAs and the ways in which these values differ from those previously reported. These data were later applied to resolve previous apparent discrepancies between theoretical predictions and empirical outcomes. They were then used to develop more optimal cryopreservation protocols for human spermatozoa [9]. The same approach to fundamental cryobiology should be used as a paradigm for the optimization of boar spermatozoa cryopreservation. Theoretical computer simulations of intracellular water volume flux during cooling and warming indicate that boar spermatozoa respond similarly in the presence of glycerol and EG at relatively low cooling rates. The data suggest that boar spermatozoa should tolerate cooling rates of 1 C/ min and 1 C/min in combination with warming rates of 12 C/min with minimal damage. However, excessively higher cooling rates (e.g., 1 C/min) in the presence of EG do result in potentially damaging cell volume excursions. Similar experimental findings of damaging cell volume excursions upon cooling and warming have been previously reported in other cell types such Chinese hamster ovary cells [29]. n addition, the theoretical simulations indicate that cell loss during cooling and warming is not primarily a result of intracellular ice formation. The probability of intracellular ice formation in boar sperm is less than 5%, in the presence of either glycerol or EG, when using cooling rates of 1 C/min, 1 C/min, and 1 C/min. Although boar spermatozoa have a relatively low water permeability in the presence of CPAs (.12,xrm/min/atm, in the presence of DMSO), compared to other cell types such as murine oocytes (.8 ilm/min/atm in the presence of DMSO) [3] regarding intracellular ice formation, the cells also have a counterbalancing low intracellular water volume (31%) [3], and a very high surface area:volume ratio (surface area:volume = 156 txm 2 :26.3 im 3 ) [3]. Therefore, the low probability of intracellular ice formation even when relatively high cooling rates are used appears consistent with the complete biophysical characteristics of these cells. Therefore, these data suggest that the majority of cell loss predicted at high cooling rates is not associated with intracellular ice formation but is due to excessive cell swelling upon rapid warming. This predicted cell loss is greater when EG rather than glycerol is used as the CPA for sperm cryopreservation. This finding presents a complex dilemma. The CPA that has been found to have the highest permeability (EG), and therefore has been predicted to be most optimal for minimizing cell volume excursions upon its addition and removal, also has the highest associated temperature dependence, resulting in potentially damaging cell swelling when rapid cooling and warming rates are used. These data indicate a three-way interaction among CPA type, cooling rate, and warming rate. Therefore, in developing new cryopreservation protocols, it may be most effective to prioritize selection of CPA type first (using the highest PCPA as the criterion) and use as high a concentration of that CPA as the cell type of interest will tolerate upon addition and removal (using osmotic tolerance limits as criteria). Then, on the basis of selection of CPA type and concentration, the procedure should predict an optimal cooling and warming rate based upon: LpCPA, PCPA, o, and their activation energies. By establishing this order, one is able to work within the cells' osmotic tolerance limits (which are currently more difficult to modify) and subsequently adjust the cooling and warming rates (which are easily adjustable using current programmable cooling/ warming units). Application of fundamental cryobiology appears to be species-dependent, and cryopreservation of boar sperma-

36 GLMORE ET AL. tozoa is considerably more challenging and complicated than the cryopreservation of spermatozoa from other species. However, with the knowledge of the osmotic tolerance limits of the cells, their response to CPA addition and removal, and the characterization of the water loss of the cell experienced during cooling and warming, this challenge can become more manageable. Future efforts should focus on engineering methods for expanding the osmotic tolerance limits of boar spermatozoa. Other extenders and their components should be studied to determine their effects on boar sperm permeability and osmotic tolerance. n addition, further investigations should be conducted to determine optimal cooling and warming rates, as well as their interactions, on a CPA-dependent basis. When applied, this information on the fundamental cryobiology of boar spermatozoa would insure minimal volume excursion during cooling and warming, and CPA addition and removal, thereby maximizing post-treatment functional viability. ACKNOWLEDGMENTS The authors would like to thank Dr. Dayong Gao for helpful discussion of the manuscript and Katherine Vernon for manuscript preparation. REFERENCES. Reed H. Current use of frozen boar semen-future need of frozen boar semen. n: Johnson LA, Larsson K (eds.), Proceedings First nternational Conference on Deep Freezing of Boar Semen. Uppsala, Sweden: Swedish University of Agricultural Sciences, Uppsala; August 1985: 225-237. 2. Almid T, Hofmo PO. A brief review of frozen semen application under Norwegian Al service conditions. n: Rath D, Johnson LA, Weitze KF (eds.), Boar Semen Preservation. Proceedings of the Third nternational Conference on Boar Semen Preservation held at Mariensee, Germany. Berlin: Blackwell Wissenschafts-Verlag GmbH Publishers; August 1995: 31:231-253. 3. Gilmore JA, Du Junying, Tao Jun, Peter AT, Critser JK. Osmotic properties of boar spermatozoa and their relevance to cryopreservation. J Reprod Fertil 1996; 17:87-95. 4. Du Junying, Tao Jun, Kleinhans FW, Peter AT, Critser JK. Determination of boar spermatozoa water volume and osmotic response. Theriogenology 1994; 42:1183-1191. 5. Curry MR, Watson PE Stopped-flow measurement of the membrane water permeability characteristics of boar spermatozoa. Cryobiology 1996; 33:634. 6. Petzoldt R, Wellmann R, Waberski D, Weitze KF. Volume regulation of boar spermatozoa under conditions of preservation. n: Rath D, Johnson LA, Weitze KF (eds.), Boar Semen Preservation. Proceedings of the Third nternational Conference on Boar Semen Preservation held at Mariensee, Germany. Berlin: Blackwell Wissenschafts-Verlag GmbH Publishers; August 1995: 31:265. 7. Gao DY, Liu J, Liu C, McGann LE, Watson PF, Kleinhans FW, Mazur P, Critser ES, Critser JK. Prevention of osmotic injury to human spermatozoa during addition and removal of glycerol. Hum Reprod 1995; 1:119-1122. 8. Willoughby CE, Mazur P, Peter AT, Critser JK. Osmotic tolerance limits and properties of murine spermatozoa. Biol Reprod 1996; 55: 715-727. 9. Gilmore JA, Liu J, Gao DY, Critser JK. Determination of optimal cryoprotectants and procedures for their addition and removal from human spermatozoa. Hum Reprod 1997; 12:112-118. 1. Bavister BD, Leibfried ML, Lieberman G. Development of preimplantation embryos of the golden hamster in a defined culture medium. Biol Reprod 1983; 28:235-247. 11. Weitze KE Long-term storage of extended boar semen. n: Johnson LA, Rath D (eds.), Boar Semen Preservation. Proceedings of the Second nternational Conference on Boar Semen Preservation held at Beltsville, Maryland. Berlin and Hamburg: Paul Parey Scientific Publishers; 199: 191:231-253. 12. Kedem O, Katchalsky A. Thermodynamic analysis of the permeability of biological membranes to nonelectrolytes. Biochim Biophys Acta 1958; 27:229-246. 13. Levin RL, Cravalho EG, Huggins CE. A membrane model describing the effect of temperature on water conductivity of erythrocyte membranes at subzero temperatures. Cryobiology 1976; 13:419-429. 14. Bunow B, Knott G. Mlab Reference Manual. Bethesda, MD: Civilized Software, nc.; 1995: 213. 15. Steel RGD, Torrie JH. Principles and Procedures of Statistics. New York: McGraw-Hill; 196: 194-231. 16. Liu J, Zieger M, Lakey J, Woods E, Critser JK. The determination of membrane permeability coefficients of canine pancreatic islet cells and their application to islet cryopreservation. Cryobiology 1997; 35:1-13. 17. Demianowicz W, Strzezek J. The effect of lipoprotein fraction from egg yolk on some of the biological properties of boar spermatozoa during storage of the semen in liquid state. n: Rath D, Johnson LA (eds.), Boar Semen Preservation. Proceedings of the Third nternational Conference on Boar Semen Preservation held at Mariensee, Germany. Berlin: Blackwell Wissenschafts-Verlag GmbH Publishers; August 1995: 31:279. 18. Graham JK, Foote RH. Effect of several lipids, fatty acyl chain length, and degree of unsaturation on the motility of bull spermatozoa after cold shock and freezing. Cryobiology 1987; 24:42-52. 19. Critser JK, Huse-Benda AR, Aaker DV. Cryopreservation of human spermatozoa.. The effect of cryoprotectants on motility. Fertil Steril 1988; 5:314-32. 2. Polge C. Artificial insemination in pigs. Vet Rec 1956; 68:62-76. 21. Benson CT, Liu C, Gao DY, Critser JK. Determination of the osmotic characteristics of pancreatic islets and isolated pancreatic islet cells. Cell Transplant 1993; 2:461-465. 22. Gilmore JA, McGann LE, Liu J, Gao DY, Peter AT, Kleinhans FW, Critser JK. Effect of cryoprotectant solutes on water permeability of human spermatozoa. Biol Reprod 1995; 53:985-995. 23. Mazur P, Miller RH. Survival of frozen-thawed human red blood cells as a function of cooling and warming velocities. Cryobiology 1976; 13:57-522. 24. Meyer MM, Verkman AS. Human platelet osmotic water and nonelectrolyte transport. Am J Physiol 1986; 251:c549-c557. 25. Curry MR, Millar JD, Watson PE Calculated optimal cooling rates for ram and human sperm cryopreservation fail to conform to empirical observations. Biol Reprod 1994; 51:114-121. 26. Penfold LM, Garner DL, Donoghue AM, Johnson LA. Comparative viability of bovine sperm frozen on a cryomicroscope or in straws. Theriogenology 1997; 47:521-53. 27. Noiles EE, Mazur P, Watson PF, Kleinhans FW, Critser JK. Determination of water permeability coefficient for human spermatozoa and its activation energy. Biol Reprod 1993; 48:99-19. 28. Henry MA, Noiles EE, Gao DY, Mazur P, Critser JK. Cryopreservation of human spermatozoa V: the effects of cooling rate and warming rate on the maintenance of motility, plasma membrane integrity, and mitochondrial function. Fertil Steril 1993; 6:911-918. 29. Griffiths JB, Cox CS, Beadle DJ, Hunt CJ, Reid DS. Changes in cell size during the cooling, warming, and pot-thawing periods of the freeze-thaw cycle. Cryobiology 1979; 16:141-151. 3. McGrath JJ, Gao DY, Tao J, Benson C, Crister ES, Critser JK. Coupled transport across the murine oocyte plasma membrane: water and cryoprotective agents. n: Toner M, Flik M, Webb DW, Vader DT, Arimilli RV, Sauer HJ, Georgiadis J, Prasad V (eds.), Topics in Heat Transfer. HTD Vol. 26-2. New York: American Society of Mechanical Engineers Press; 1992: 1-14.