Effect of salt concentration and unfrozen water fraction on the viability of slowly frozen ram spermatozoa

CRYOBIOLOGY

25, 131-142 (1988)

Effect of Salt Concentration Viability of Slowly Department

and Unfrozen Water Fraction Frozen Ram Spermatozoa

on the

P. F. WATSON AND ANNE E. DUNCAN of Physiology, The Royal Veterinary College, Royal College Street, London NW1 OTU, United Kingdom

Ram spermatozoa were subjected to a slow rate of freezing (l”C/min) in various glycerolNaCl-water solutions of known composition such that the molal concentration of NaCl (m,) and the unfrozen fraction of water (U) could be calculated at subzero temperatures from the relevant phase diagram. Sperm motility was reduced as m, increased and U correspondingly decreased with temperature. However, by freezing spermatozoa in solutions of differing initial tonicities, but with a constant weight ratio of glycerol:salt, to various subzero temperatures, the effects of m, could be separated from those of U. Motility was found to decrease dramatically at values of U 40.07 regardless of ms but, at higher values of U, maximum motility was dependent on the final salt concentration in that fraction, being reduced as the osmolality increased. Sperm cell concentration had no apparent effect on the influence of m, or U on viability in the range studied (3-12 x 10’ spermatozoa/ml). In order to account for these observations, the effects of osmotic stress on spermatozoa were investigated. When subjected to sudden changes in osmolality of the suspending medium by increasing NaCl or sucrose concentration at room temperature, spermatozoa showed a decreased motility with increasing osmolality. Since no improvement in motility was found on returning the cells to isosmolar conditions cell damage appeared to be irreversible. Furthermore, when placed in solutions of increasing hypotonicity the number of swollen spermatozoa with looped tails increased with increasing hypotonicity. Since the drop in motility seen at low values of U corresponded to those spermatozoa exposed to a hypotonic starting solution, it is suggested that a hypotonic stress followed by a hypertonic stress during freezing and thawing may account for the profound loss of motility in these samples, while a hypertonic stress may account for the strong effect of m, seen at higher values of U. B 1988 Academic Press, IX.

Mazur (10) proposed that during freezing two factors with an opposite dependence on cooling rate are operative to explain the observation that for any given cell type there is an optimum cooling rate above and below which cell survival declines. At rates higher than the optimum, cell death is caused by the formation of intracellular ice; while at less than optimum rates the damaging effects of the rising solute concentration, the changing pH as buffer salts reach their solubilities, the increasing cellular dehydration, the diminishing fluid volume, and the rising cryoprotective concentration may all contribute to loss of viability. Mazur (10) subsumed all these changing effects Received April 20, 1987; accepted October 19, 1987.

at slow cooling rates under the term “solution effects.” There is little dispute about the lethal effects of more than small amounts of intracellular ice, but the importance of particular “solution effects” has not been ascertained. Lovelock (8) showed that red blood cells were hemolyzed at salt concentrations equivalent to those experienced during slow freezing and maintained that exposure to excessive salt concentration was the cause of injury. He proposed that glycerol acted to limit the rise in electrolyte concentration. Spermatozoa were shown to have a similar but more pronounced sensitivity to hypertonic solutions of electrolytes (9). An alternative hypothesis for the cell injury resulting from solute concentration centers on the osmotic forces created across cell 131 OOII-2240188 $3.00 Copyright All rights

0 1988 by Academic Press, Inc. of reproduction in any form reserved.

132

WATSON

AND

membranes. Meryman (14) proposed that cells have a minimum tolerable cell volume, below which during freezing an excessive hydrostatic pressure develops across the cell membrane because of an inability to maintain osmotic equilibrium by further loss of water and consequent reduction in volume. He envisaged that “compression” of the cell contents would lead to a resistance to volume reduction. The debate has recently been continued by Mazur and his colleagues, who have attempted to separate the effect of the fraction of solution remaining unfrozen from that of the rising salt concentration during slow cooling. By appropriate manipulation of the initial glycerol and salt concentration and the temperature to which cells are frozen, the unfrozen fraction may be varied independently of the salt concentration (12, 13). They found that when the unfrozen fraction of extracellular water decreased below 15%, the degree of hemolysis of red blood cells was dependent upon the size of the unfrozen fraction irrespective of the salt concentration. Only at high hematocrits (40-60%) did the electrolyte concentration become more important (11). Otherwise, an effect of salt concentration was only evident when less than 85% of the water was frozen (13). In view of the earlier observations on the sensitivity of spermatozoa to salt concentration (9) and the uniquely specialized structure of spermatozoa, we decided to investigate the effects of similar manipulations on sperm cells. MATERIALS

AND

METHODS

Semen Collection

Ram semen was collected from Finnish Landrace rams by artificial vagina. Only ejaculates with good initial motility were used in the experiments. Sperm density was ascertained from light absorbance measurements on a previously calibrated colorimeter (7). Except where cell concentra-

DUNCAN

tion was being investigated, semen was diluted to give a sperm cell concentration of 3 x lO*/ml in a washing diluent consisting of 150 mA4 NaCl, reconcentrated to its original volume by centrifugation at 700g for 10 min, and resuspended in the appropriate test solution. Preparation of Glycerol Solutions Containing Isotonic NaCl

An appropriate volume of washed cells was mixed with isotonic NaCl containing 0, 0.5, 1.0, 1.5, or 2.0 M glycerol to give a cell concentration of 3 X 10’ spermatozoa/ml. The exact compositions of solutions are given in Table 1. Preparation of Solutions Designed to Separate the Effect of NaCl Concentration from that of the Unfrozen Water Fraction

An appropriate volume of washed cells was mixed with one of eight different glycerol-NaCl-water solutions, containing 0.6, 0.75, 1, and 2x the isotonic concentration (percentage by weight) of NaCl, and a corresponding increasing concentration of glycerol so the weight ratio of glycerol: NaCl (R) remained at 5.42 in one set of four solutions and 11.26 in the other set of four solutions. The composition of these solutions was identical to that given in Table 1 of Mazur et al. (13). Slow Cooling and Freezing

After mixing with the appropriate test solution, the diluted semen was packaged in TABLE 1 Composition of Glycerol-Saline-Water Used in Experiment 1

Solutions

Percentage by weight

Glycerol concentration (M)

Glycerol (g)

NaCl (g)

0 0.5 1.0 1.5 2.0

4.604 9.209 13.814 18.417

0.878 0.846 0.814 0.782 0.75

CAUSES

OF CRYOINJURY

IN

0.25-ml plastic straws (paillettes, IMV, L’Aigle, France) modified to accept only 0.1 ml diluted semen. The straws were sealed with polyvinyl acetate powder and cooled at a linear rate of 0.125”CYmin from room temperature to O”C, using a computer-controlled cooling bath. The straws were then transferred to a rack at 0°C in a computer-controlled freezing device over the surface of liquid nitrogen and cooled at a rate of l”C/min to -5°C. Ice nucleation was induced at -5°C by touching the straws with a precooled copper rod; seeding in this way standardizes the temperature of ice nucleation and prevents wide temperature fluctuations which may be detrimental. Seeded samples were held for a further 5 min at - 5°C to allow equilibration with respect to crystallization, before freezing at a linear rate of l”C/min. At temperatures determined by the protocol, straws were removed and thawed rapidly in water at 37°C. Preparation of Solutions Designed to Test the Effect of Osmolality

Washed spermatozoa were mixed either with various NaCl solutions with molarity ranging from 0.15-1.0 M or with solutions of corresponding osmolality derived from a range of O-1.7 M sucrose each containing 0.15 M NaCl in order to maintain a constant ionic strength. Actual osmolalities were measured with an osmometer (Fiske Associates, Inc., Bethel, CT). Spermatozoa were left in each solution for 10 min at 20°C before determining motility immediately, or after the addition of an appropriate volume of distilled water slowly during mixing to return the cells to isosmotic conditions. Determination

of Motility

Survival was determined by an estimate of both the percentage of motile spermatozoa and quality of motility after freezing and thawing. Thawed straws were opened into a test tube at 37°C and samples were

SLOWLY

FROZEN

RAM

SPERM

133

assessed in random order by microscopic examination of drops of diluted semen on a warm stage at 37°C. The percentage of motile spermatozoa was estimated to the nearest 10% and motility was scored on a scale of o-4 (4). All experiments were replicated over ejaculates from four individuals, and results were analyzed by a computerized analysis of variance after prior transformation of percentage data to angles. The pooled ejaculate terms were used as error to test the significance of the treatment mean squares. Since motility scores were found to follow the same trend as the percentage motile, only data for the latter are presented here. Exposure to Hypotonic

Conditions

When spermatozoa are exposed to hypotonic conditions the cells swell and the tails coil up within the plasma membrane, giving a characteristic appearance (3). This observation was used to investigate the effect of varying degrees of hypotonic stress on ram spermatozoa at 20°C. Semen was diluted lo-fold in mixtures of isotonic phosphatebuffered saline and distilled water designed to give steps of 50 mOsm. Actual osmolarities were measured. After 30 min, unfixed smears were prepared on polylysine-coated slides and counts were made of the percentage of spermatozoa showing swollen and looped tails. RESULTS

Survival of Spermatozoa in Partly Frozen Glycerol-NaCl-Water Solutions

Phase diagrams of solutions of known composition NaCl and glycerol can be used to calculate the molal concentration of NaCl in the unfrozen portion at various temperatures; similarly, the unfrozen fraction of the solution can be obtained from these phase diagrams (13). Spermatozoa were frozen to various subzero temperatures in solutions containing O-2 M glycerol in isotonic NaCl before rewarming, and

134

WATSON AND DUNCAN

The temperatures, the corresponding valpost-thaw motility was plotted against the corresponding NaCl concentration or un- ues of m, and U, and the resulting mean frozen water fractions to which the sperma- motilities are given in Table 2 for R = 5.42 tozoa had been exposed. The results indi- and R = 11.26, respectively. It is apparent cate that motility was gradually reduced as from Table 2 that in both sets of solutions the NaCl concentration increased (Fig. l), (R = 5.42 and R = 11.26) post-thaw motilbut was largely independent of the decreas- ity decreased with decreasing temperature, ing unfrozen fraction of water until very but the decline in motility occurred at low values were reached (Fig. 2). However, slightly lower temperatures in the solutions in this experiment salt concentration and containing higher glycerol levels (R = unfrozen fraction were not independent and 11.26). However, this effect was not signifvaried directly with one another. The R val- icant. A significant difference between inues of the solutions used ranged from O- dividual ejaculates was detected, a com24.56 and thus the temperatures required to mon observation with spermatozoa, indiachieve a particular salt concentration in cating differing sensitivities of spermatozoa the solutions varied widely, but the curves from different individuals to treatment resulting from the different solutions were stresses. In Figure 3 mean post-thaw motility was closely grouped. plotted for decreasing values of U at conRelative Contributions of m, and U to the stant values of m,. Motility decreased at Survival of Slowly Frozen Spermatozoa low values of U (~0.07) regardless of salt concentration, but at higher values, maxiSemen diluted in each of eight different mum motility was dependent on the salt glycerol-NaCl-water solutions was frozen concentration being reduced after exposure at l”C/min to 10 different temperatures to to the higher values. When mean post-thaw provide a series of U values for constant motility was plotted against m, at constant values of m, and then thawed rapidly in a values of U (Fig. 4) salt concentration was water bath at 37°C. found to have little effect on motility at lower values of U (0.03-0.06), but at the higher values of U, motility was clearly adversely affected by the concentration of salt 40 1 present. \ \ ;c*: Effect of Unfrozen Fraction at Different .qy + 30 -. '\ Cell Concentrations 41, .o .:...., '1 \ z$,.. C. \..::.. Concentrations of 3-12 x lo8 spermatoY!i 1 SM F 20l'OM'@p...*o,sM zoa/ml were diluted into experimental glyc0 .. I '. '. erol-NaCl-water solutions as described P '. '. .. previously, containing 0.6, 0.75, 1, and 2x '. IO'.., the isotonic concentration of NaCl with in.. .. '. creasing concentrations of glycerol such .. '. that the weight ratio of glycerol to NaCl '.b OM I I I O20 1.0 3.0 4.0 was 5.42. Figure 5 shows the resulting postMOLALITY NaCl (r&l thaw motilities with varying temperatures FIG. 1. Effect of NaCl molality (m,) on post-thaw and cell concentrations for the four solumotility of ram spermatozoa after freezing at l”C/min to various subzero temperatures while suspended in tions used. At the lowest tonicity (0.6x, solutions of the indicated molarity of glycerol in iso- Fig. 5a), motility appeared independent of cell concentration and decreased with temtonic NaCI.

CAUSES

OF CRYOINJURY

IN

SLOWLY

FROZEN

RAM

135

SPERM

lo-

0’

I 01 UNFROZEN

I WATER

02 FRACTION

, 0.3 (U)

FIG. 2. Effect of the unfrozen water fraction (U) on post-thaw motility of ram spermatozoa after freezing at a rate of l”C/min to various subzero temperatures while suspended in solutions of the indicated molarity of glycerol in isotonic NaCI.

perature. At higher tonicities, motility also decreased with temperature, but was consistently highest at the sperm concentration of 6 x lO’/ml (P < 0.01). There was a greater tolerance of freezing at all cell concentrations in both iso- and hypertonic solutions compared with that in the hypotonic solutions. The effects of salt concentration and unfrozen water fraction were separated out as before (Fig. 6) and showed a similar trend to the previous results, i.e., motility decreased dramatically only at low values of U (~0.1) regardless of salt concentration, and m, only exerted an effect on motility at higher values of U. Analysis of the data showed that cell concentration had no apparent effect on the response of sperm cell motility to m, or U, within the range of concentrations studied. Effect of Osmotic Stress Figure 7 shows the resulting motility after exposure of the cells to increasing osmolality. There was no apparent difference between cells exposed to high concentrations of electrolyte and those exposed to increased concentrations of sucrose, and

no improvement in motility was found after returning the cells to isosmotic conditions (Fig. 8). At 2O”C, the cells showed a considerably more marked sensitivity to hyperosmotic conditions than when they were being frozen (Figs. 1 and 3). When ram spermatozoa were exposed to hypotonic conditions of 295, 250, 198, and 150 mOsm, the proportions of cells with looped tails were 3, 9, 40 and 65%, respectively. At lower tonicities still, many of the “looped” cells suddenly straightened as the plasma membrane was ruptured. DISCUSSION

Ram spermatozoa are very sensitive to cooling, and for cryopreservation, a protectant of membrane stability, such as egg yolk, would normally be included. Such an additive was considered to be undesirable for these experiments since it may have made interpretation of the results more difficult. As a consequence, however, the mean survival under optimal conditions in these experiments did not exceed 32.5%. Ram spermatozoa displayed an apparent sensitivity to low values of U regardless of salt concentration (Figs. 2, 3, and 6). How-

136

WATSON AND DUNCAN

TABLE 2 Mean Post-thaw Motilities of Suermatozoa Suspended in Various Glycerol-NaCl-Water l”C/min to Various Subzero Temperatures and Thawed Rapidly Relative NaCl tonicity

Temperature P-3

Molal NaCl

Solutions Frozen at

Fraction unfrozen

b-d

Motility m

R = 5.42

0.6 0.6 0.6 0.6 0.6 0.75 0.75 0.75 0.75 0.75 1.0 1.0 1.0 1.0 1.0 2.0 2.0 2.0 2.0 2.0

-11 - 17 -21 -26 -30 -11 -17 -21 -26 -30 -11 -17 -21 -26 -30 -11 -17 -21 -26 -30

0.6 0.6 0.6 0.6 0.6 0.75 0.75 0.75 0.75 0.75 1.0 1.0 1.0 1.0 1.0 2.0 2.0 2.0 2.0 2.0

- 18 -28 -34 -41 -48 -18 -28 -34 -41 -48 -18 -28 -34 -41 -48 -18 -28 -34 -41 -48

1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8

0.092 0.058 0.046 0.038 0.033 0.109 0.068 0.054 0.045 0.039 0.15 0.092 0.073 0.061 0.052 0.304 0.19 0.16 0.13 0.11

20.0 13.3 10.25 8.0 7.75 22.5 30.0 27.5 15.0 12.5 32.5 26.7 22.5 10.25 15.0 20.0 20.3 22.5 15.2 10.25

0.092 0.058 0.046 0.038 0.033 0.107 0.067 0.054 0.045 0.038 0.15 0.092 0.073 0.061 0.052 0.33 0.21 0.16 0.14 0.12

25.0 17.5 15.0 10.0 10.25 27.5 27.5 17.5 15.0 15.0 27.5 25.0 17.5 12.5 15.0 17.5 12.75 15.0 10.25 10.0

R = 11.26

ever, they also showed a very marked sensitivity to increasing salt concentration when the unfrozen water fraction was above the lowest values (Figs. 1, 3, 4, and 6). It is interesting, nevertheless, that the range of salt concentrations producing this

1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8 1.0 1.6 2.0 2.4 2.8

effect during freezing (1-3 M) was far in excess of that causing loss of motility at 20°C (Fig. 7). Little is known of the effects of temperature on membrane responses but these data suggest that the cell structures may be less susceptible to osmotic damage

CAUSES

OF CRYOINJURY

IN

SLOWLY

FROZEN

RAM

SPERM

137

1.0

1.0

0’

0.1 UNFROZEN

0.2 WATER

0.3 FRACTION

0.4 tU)

FIG. 3. Post-thaw motility of ram spermatozoa as a function of unfrozen water fraction (v) after freezing cells at a rate of l”C/min in NaCl-glycerol-H,O. U was varied independently of NaCl molality (WI,)by appropriate manipulation of the initial tonicity and temperature to which cells were frozen. (A) Glycerol:NaCl weight ratio, R = 5.42; (B) R = 11.26.

at low temperatures. Alternatively, a grad- same values of m, and U as in those with ual increase in osmolality may be less lower R values. Although motility was constressful than a sudden increase. Unfortusistently slightly lower in the solution with nately, even above freezing the effects of R = 11.26, i.e., cooled to lower temperatemperature on membrane responses are tures, this was not significant (P > 0.25). difficult to test with ram spermatozoa be- However, it cannot be maintained that such cause of their peculiar temperature sensi- an approach isolates the temperature effect tivity. entirely satisfactorily, since although the R The use of constant molar ratio of glyc- values are constant, the cells are exposed erol:salt enabled the separation of m, and to increasing concentrations of glycerol. U, but at the same time introduced the con- Ram spermatozoa are known to be sensifounding of temperature with the other fac- tive to the adverse effects of glycerol (2, 6, tors. By repeating the observations at a dif- 17); any mild differences in motility beferent constant R value, the effect of tem- tween the R values may be due to the influperature can be isolated since the samples ence of glycerol, but the effect, if present in in the higher R value solution must be this case, was only slight. Moreover, alcooled to lower temperatures to achieve the though temperature and glycerol concen-

138

WATSON

ok/

I

II0

DUNCAN

1

2.0 MOLALITV

AND

3.0 NaCl

(m3)

FIG. 4. Post-thaw motility of ram spermatozoa as a function of NaCl molality (m,) in the unfrozen water fraction (U) of NaCl-glycerol-H,0 solution after freezing slowly at l”C/min. m, was varied independently of U by appropriate manipulation of the initial tonicity and glycerol concentration. (A) Glycerol: NaCl weight ratio, R = 5.42; (B) R = 11.26.

tration were confounded in the results in Fig. 1, it can be seen that for a given molal salt concentration, achieved at a wide range of temperatures in the solutions of different glycerol:salt ratios, survival was not greatly affected. This also confirms that neither temperature nor glycerol concentration was causing profoundly adverse effects. The reduced motility at the highest values of U apparent in some of the curves of constant m, (Fig. 3) may be attributable to an effect of starting solution. In preliminary experiments it was verified that spermatozoa would survive brief exposure to 2x isosmotic NaCl but survival was reduced as compared with that in the isosmotic solution, as illustrated in Fig. 7. It was in starting solutions of 2x isosmotic that survival was also reduced after freezing and thawing compared with that in starting solutions of isosmotic NaCl (Fig. 3). Mazur and Rigopoulos (12) noted a similar pattern of re-

-10

-30

-20 TEMPERATURE

(DEG.

CELSIUS)

FIG. 5. Effect of cell concentration on post-thaw motility of ram spermatozoa frozen at l”C/min to various temperatures. (A, B, C, D) Cells suspended in 0.6x, 0.75x, 1x, and 2X isotonic saline, respectively. Glycerol:NaCl weight ratio, R = 5.42. (@.....a) 3 x 10’ spermatozoa/ml, (-) 6 x 10’ spermatozoa/ml, (O- - - - -0) 12 X lo* spermatozoa/ml.

sponse with red blood cells at 4~ isosmotic NaCl and attributed it to to the osmotic stress of sudden exposure to the associated high glycerol concentrations. In the case of ram spermatozoa, in view of the demonstrated sensitivity to osmotic effects of salt, we consider the salt component to be at least as likely as glycerol to account for this observation. However, when the results were converted to percentages of initial survival the pattern of effects of m, and U was unchanged. Cell concentration in the range studied,

CAUSES

3o

1; I

0

OF CRYOINJURY

IN

SLOWLY

FROZEN

RAM

SPERM

139

c

*..f.&~” **.p& . .a :j 9

,,,,,..tl...“.;“;

*;I s

I

1

0.1

0.2 UNFROZEN

WATER

I 0.3 FRACTION

(U)

6. Effect of cell concentration on post-thaw motility of ram spermatozoa frozen at l”C/min to various temperatures. The unfrozen water fraction (.!I’)was varied independently of NaCl molality (m,) by appropriate manipulation of the initial tonicity and glycerol concentration. GlycerokNaCl weight ratio, R = 5.42. (A, B, C) 3, 6, and 12 x 10s spermatozoa/ml, respectively. FIG.

3-12 x lO’/ml, did not appear to influence the effects of either m, or U on the survival of spermatozoa, although the results with 6 X lo8 spermatozoa/ml were significantly better than with the other concentrations. This is similar to the report that ram spermatozoa yielded the highest survivals when frozen in concentrations of 9 X lO’/ml (2). A comparison of these results with those for red blood cells (11, 12) reveals some striking similarities, but also some impor-

tant differences. The viability response to U was similar, but occurred at values of U (when >90-95% of extracellular water was frozen, Figs. 3 and 4) lower than those seen with red cells (X30-85% water frozen (13)). A salt concentration effect at higher U values was found for red cells (1 l), but the effect of salt concentration on spermatozoa appeared to be more pronounced. Similarly, in both spermatozoa and erythrocytes the effect of temperature or glycerol was

140

WATSON AND DUNCAN

soSO-

70w

SO-

F

50-

-1

Fl

8

40302010 0

0.2

0.4

0.6 0.6 OSMOLALITY(Osm)

10

1.2

1.4

FIG. 7. Motility of ram spermatozoa as a function of osmolality of the medium, for cells suspended in solutions of increasing NaCl concentration or increasing sucrose concentration.

not significant. A clear cell concentration effect was demonstrated on the contribution of m, to erythrocyte survival at high hematocrits but no similar interaction was evident with spermatozoa; however, the hematocrit at which an effect was seen (4060%) represents cell concentrations of the order of 6-10 X 109/ml, far in excess of those sperm densities investigated here. An alternative explanation was recently proposed for the apparent dependence of red blood cell survival on the fraction of unfrozen water, which takes into account the extent of contraction or expansion of the cell (15). While a single curve applies to

OSMOLALITY(Osm1

FIG. 8. Motility of ram spermatozoa on returning to isosmotic conditions after exposure to the osmolalities shown on the abscissa, in either NaCl or sucrose solutions of varying concentrations.

the cell volume at subzero temperatures irrespective of the solution in which the cells were suspended (12), the initial cell volume varies with solutions of different initial tonicities. The volume excursion is a function of both the initial solution and the final temperature. Pegg (15) suggested that, since the volume excursion of the cell varies as well as the unfrozen fraction and salt concentration, the cell damage may be related to shrinkage and expansion. When the results obtained with spermatozoa were examined from this standpoint, it was clear that there were some very distinct differences from red cells. Whereas the survival of red cells appeared to be directly related to volume changes after freezing and thawing, spermatozoa were additionally influenced by the initial salt concentration; postthaw motility was reduced in those samples which had initially been exposed to hypertonic media. In order to explore this further, spermatozoa were subjected to changes in salt concentration at 20°C. Survival decreased very rapidly when the salt concentration was increased in the range of 1.25 to 2.25~ isotonic saline (Fig. 8), comparable values to those involved in starting solutions of the freezing experiments. Lovelock and Polge (9) had previously demonstrated a similar response with bull spermatozoa and had interpreted it as an effect of high salt concentrations. However, when sucrose was substituted for the additional salt (maintaining an isotonic saline) the response was identical (Fig. 7), demonstrating that the effect was due to rising osmolality rather than salt concentration per se. The cell damage was irreversible since no improvement in motility occurred on redilution to isosmolar conditions (Fig. 8). Meryman (14) recognized that although the salt concentration in this experimental design externally remained isotonic, it would rise internally as water was withdrawn irrespective of whether the external osmolality was increased by electrolyte or nonelectrolyte. The mechanism of cell damage may still, thus, relate to in-

CAUSES

OF CRYOINJURY

IN

creased ionic strength. Bull spermatozoa, which are structurally very similar to ram spermatozoa, have been found to be capable of only a limited decrease in size in hyertonic solutions compared to the increase found in hypotonic solutions (1). Bredderman and Foote (1) found that the mean cell volume in isosmotic saline was 25.2 um3, rising to 32.4 urn3 in hypotonic saline (94 mOsm), and decreasing to 20.0 urn3 in hypertonic Tris buffer (712 mOsm). More importantly, killed cells, whose membranes were no longer osmotically active and which had presumably leaked cytoplasmic contents, had mean volumes of 20.420.8 um3. They comment that in isotonic media spermatozoa “appear to be closer to a fixed minimum volume than to their maximum possible volume” (1). Thus the damaging effects on spermatozoa appear to be attributable directly to exposure to hyperosmotic conditions rather than to excessive volume changes of the cells or to excessive salt concentration. This would account for the strong effect of salt concentration seen at higher values of U (Figs. 1, 3, 4, and 6). It is not surprising that spermatozoa and erythrocytes behave differently during slow freezing. Spermatozoa have a complex and rather rigid substructure comprising mostly condensed nucleus in the head region and mitochondria and flagellar motile components in the tail. There is little cytoplasm and the cell water content is relatively very low (see Ref. (16)). The substructure presumably strongly resists cell shrinkage and thus osmotic forces exerted across membranes would become of considerable importance during dehydration. It appears that the behavior of spermatozoa largely supports the hypothesis of a “minimum cell volume” below which cells cannot shrink during freezing and instead become progressively damaged (14). The importance of the unfrozen fraction is not adequately explained by this concept. Mazur and Cole (11) suggested that, as the unfrozen fraction reached low values, the advancing ice front might exert rheological

SLOWLY

FROZEN

RAM

SPERM

141

forces on the cells and, in addition, that cell-to-cell interactions became important at high salt concentration and high hematocrit. While such an explanation is not excluded by these results another possibility is suggested by Pegg’s approach (15). The profound losses of motility at the very low values of unfrozen fraction are all derived from treatments in which the spermatozoa had been exposed to an initially hypotonic medium. From the results of the final experiment reported, up to 50% of the spermatozoa would have responded to hypotonic stress by looping of the tail structures in the 0.6~ isosmotic NaCl. The additional hypotonic stress could well be sufficient to render these cells nonfunctional after cooling, freezing, and thawing. It is known that the post-thaw motility of ram spermatozoa is improved in hypertonic freezing diluents as compared to hypotonic diluents (5). Similarly we have reported here that spermatozoa tolerated freezing better in hyper- or isotonic solutions than in hypotonic solutions (Fig. 5). This suggests the view that spermatozoa withstand cooling and freezing stresses less well after exposure to hypo- rather than hypertonic conditions. We suggest that the hypotonic stress in some samples prior to cooling, combined with the rise and fall of hypertonicity during freezing and thawing, may be sufficient to account for both the marked salt concentration effect at higher U values and the profound loss of motility at the lowest U values. ACKNOWLEDGMENTS

We would like to thank Dr. D. E. Pegg for helpful discussions and for allowing us to see material prior to publication, and also Mr. W. J. Anderson for expert technical assistance. The work was supported by a grant from the Agricultural and Food Research Council.

REFERENCES

1. Bredderman, P. J., and Foote, R. H. Volume of stressed bull spermatozoa and protoplasmic

142

WATSON AND DUNCAN

droplets and the relationship of cell size to motility and fertility. .Z. Anim. Sci. 28, 496-501 (1969). 2. Colas, G. Effect of initial freezing temperature, addition of glycerol, and dilution on the survival and fertilizing ability of deep-frozen ram semen. .I. Reprod. Fertil. 42, 277-285 (1975). 3. Drevius, L.-O., and Eriksson, H. Osmotic swelling of mammalian spermatozoa. Exp. Cell Res. 42, 136-156 (1966). 4. Emmens, C. W. The motility and viability of rabbit spermatozoa at different hydrogen-ion concentrations. 1. Physiol. 106, 471481 (1947). 5. Fiser, P. S., Ainsworth, L., and Langford, G. A. Effect of osmolality of skim-milk diluents and thawing rate on cryosurvival of ram spermatozoa. Cryobiology 18, 39-03 (1981). 6. Fiser, P. S., and Fairfull, R. W. The effect of glycerol concentration and cooling velocity on cryosurvival of ram spermatozoa frozen in straws. Cryobiology 21, 542-551 (1984). 7. Foote, R. H. Standards for sperm concentration: Polystyrene latex particles as an aid in quality control. In “Proceedings, 2nd Tech. Conf. on AI. and Reprod,” pp 95-97. Chicago, NAAB, 1968. 8. Lovelock, J. E. The haemolysis of human red blood cells by freezing and thawing. Biochim. Biophys. Acta. 10, 414-426 (1953). 9. Lovelock, J. E., and Polge, C. The immobilization of spermatozoa by freezing and thawing and the protective action of glycerol. Biochem. J. 55, 618-622 (1954). 10. Mazur, P. The role of cell membranes in the freez-

ing of yeast and other single cells. Ann. N. Y. Acad. Sci. 125, 658-676 (1965).

11. Mazur, P., and Cole, K. W. Influence of cell concentration on the contribution of unfrozen fraction and salt concentration to the survival of slowly frozen human erthrocytes. Cryobiology 22, 509-536 (1985). 12. Mazur, P., and Rigopoulos, N. Contributions of unfrozen fraction and of salt concentration to the survival of slowly frozen human erthrocytes: Influence of warming rate. Cryobiology 20, 274-289 (1983). 13. Mazur, P., Rall, W. F., and Rigopoulos, N. Relative contributions of the fraction of unfrozen water and of salt concentration to the survival of slowly frozen human erthrocytes. Biophys. J. 36, 653-675 (1981). 14. Meryman, H. T. The exceeding of a minimum tolerable cell volume. In “The Frozen Cell” (G. E. W. Wolstenholme and M. O’Connor, Eds.), pp 51-67. Churchill, London, 1970. 15. Pegg, D. E. Mechanisms of freezing damage. In “Temperature and Animal Cells” (K. Bowler and B.J. Fuller, Eds.) 41, 363-378. The Company of Biologists Ltd. Cambridge, 1987. SEB Symposium. 16. Watson, P. F. The preservation of semen in mammals. In “Oxford Reviews of Reproductive Biology.” (C. A. Finn, Ed.) Vol. 1, pp. 283351. Oxford Univ. Press, London/New York, 1979. 17. Watson, P. F., and Martin, I. C. A. Effects of egg yolk, glycerol and the freezing rate on the viability and acrosomal structures of frozen ram spermatozoa. Ausf. J. Biol. Sci. 28, 153-159 (1975).