|Home | About | Journals | Submit | Contact Us | Français|
The present study evaluated the interactions among pre-cooling, cryoprotectant, cooling, and thawing for rhesus monkey sperm using a four-way factorial design. Specifically, pre-cooling and thawing were evaluated for two conditions: slow vs. fast. Cooling was evaluated at four rates of 5, 29, 200, and 400 °C/min. The types of cryoprotectant involved combinations of egg yolk and glycerol, egg yolk and ethylene glycol, and egg yolk alone without permeable cryoprotectants or buffer alone with glycerol but without egg yolk. Our findings showed strong interactions among cryoprotectants, cooling, and thawing rates, but not pre-cooling rate, on post-thaw motility and forward progression. The optimal combination of cooling and thawing for maximum post-thaw survival depended on the types of cryoprotectant. When glycerol was used as a permeable cryoprotectant in the presence of egg yolk, slow thawing yielded similar success as fast thawing in some males. However, when glycerol was replaced with ethylene glycol for the same treatment, post-thaw motility was significantly lower in samples that were thawed slowly than those that were thawed rapidly. In the absence of permeable cryoprotectant but the presence of egg yolk, fast cooling was always favorable. On the contrary, in the absence of egg yolk but the presence of permeable cryoprotectant (glycerol), post-thaw motility was significantly reduced especially when samples were thawed slowly. Generally, fast thawing was superior to slow thawing regardless of the types of cryoprotectant or cooling rates, and glycerol in the presence of egg yolk yielded the highest post-thaw motility in all treatment groups.
Sperm cryopreservation is an effective method for the preservation of germplasm from rhesus monkeys that will allow important genotypes to be recovered as live, breeding animals . Although live births using cryopreserved semen have been reported in rhesus macaques through intrauterine insemination (IUI)  or intracytoplasmic sperm injection (ICSI) , the use of cryopreserved sperm for standard, intravaginal artificial insemination (AI) has not been successful. Assisted reproduction techniques such as IUI or ICSI are not readily available for most primate colonies and where they are available; the cost is prohibitive for routine use. It is estimated that a vaginal AI program would cost only a few hundred dollars for each offspring produced. However, depending on the complexity of techniques used, assisted reproduction alternatives to vaginal AI can cost upwards of $7,500– 9,000 per live offspring when including costs of animal per diems, superovulation of oocyte donors with human hormones, in vitro fertilization/ICSI and embryo culture, surgical embryo transfer procedures are included . However, the true cost may approach $50,000 to $100,000 when accounting for the effort of highly trained personnel, poor efficiency of embryo transfer and expensive specialized equipment required to support these techniques. In addition, there are concerns that sperm selected for ICSI are not likely to be the sperm that would be selected for fertilization by the trial of passage through the female reproductive tract. In humans, chromosome aneuploidy has been reported with higher frequency in offspring of ICSI than controls [3,8]. Therefore, a program of conventional (vaginal) AI with cryopreserved semen would provide the most cost-effective system of genetic exchange.
Because thawed sperm can fertilize eggs when used in IUI or ICSI, the failure of standard vaginal insemination with current sperm cryopreservation methods may well indicate that thawed sperm are incapable of swimming up through the cervical canal to meet the egg. This may be due to barriers created by the interaction of sperm with cervical mucus or the length and complexity of the cervical canal that sperm must traverse . Although current sperm cryopreservation techniques are adequate for assisted reproductive technologies, improvement of sperm cryopreservation protocols is necessary to maximize post-thaw sperm function so that vaginal AI is feasible.
Sperm cryopreservation involves a series of sequential steps, and it is the interactions among many variables that determine the final outcome. The choice of cryoprotectant, pre-cooling, cooling, and thawing methods have been the four main foci for essentially all empirical studies of sperm cryopreservation. However, few studies examine all four factors simultaneously despite the fact that most researchers acknowledge there are interactions among them. Experimental designs involving multiple factors would allow simultaneous study of several variables with different treatment levels, and may help to quickly find specific successful freezing protocols [9,35]. A multiple factor analysis could also provide an efficient tool to overcome male-to-male variation in rhesus sperm cryopreservation and identify effective protocols for individual males . As a continuation of our previous efforts on systematic optimization of sperm cryopreservation in rhesus monkeys , this study determined the effects of various cryoprotectants, pre-cooling methods, cooling and thawing rates on motility of rhesus monkey sperm using a four-way factorial design.
Rhesus monkeys were housed at the California National Primate Research Center (CNPRC) and maintained according to Institutional Animal Care and Use Committee protocols at the University of California. Ejaculates were collected from eight adult males that were individually caged at the CNPRC with lights on from 06:00 to 18:00 h at 25–27 °C. The males were trained to chair restraint and semen was collected by direct penile stimulation with a Grass 6 stimulator equipped with EKG pad electrodes (30–50 V, 20 msec duration, 18 pulses/sec) . Collections were done in the morning and each male was only scheduled for collection once every week. A total of 16 ejaculates were used in this study (1–3 ejaculates per male) and the mean volume of ejaculates ranged from 20 – 550 µL. Samples were allowed to liquefy for 30 min before processing.
A dilution of 1:20 (v/v) of semen to modified Tyrode’s medium supplemented with bovine serum albumin (TL-BSA, 290 mOsm/kg)  was used for sperm motility estimation of fresh semen, and a second dilution of 1:20 (v/v) of sperm to distilled water was used for hemacytometer counts (Hausser Scientific, Horsham, PA). The mean sperm density ranged from 1.5 – 36.7 × 108/mL. For comparisons among various concentrations of ethylene glycol, sperm suspensions were washed twice with TL-BSA at 300 × g for 10 min. For the 4-way factorial arrangement of treatments in the experiment, sperm were used directly without washing to ensure sufficient sperm numbers. Sperm were initially suspended to 1 × 108 cells/mL of total motile sperm (sperm density × initial motility) with TEST solution (43.25 g TES, 10.265 g Tris, 10g glucose in 1L distilled water, pH 7.4, 350 mOsm/kg, modified from ) with or without 20% egg yolk. Sperm suspensions were mixed with equal volume of double strength cryoprotectant (suspended in TEST or TEST-yolk) or TEST-yolk to obtain a final freezing density of 5 × 107 cells/mL. All chemicals used for preparation of solutions were of reagent grade (Sigma Chemical Corporation, St. Louis, Missouri).
Fifty µl aliquants of sperm suspensions were drawn into 0.25-mL French straws (IMV International, Minneapolis) manually with a 1cc syringe and were heat-sealed (MP-4 Impulse Sealer, Midwest Pacific). Unless otherwise specified, for pre-cooling, straws were placed into a 600-mL glass beaker containing 500 mL of room temperature distilled water, and equilibrated at 4 °C in a refrigerator for 2 h before initiation of the freezing process. Samples were cooled at 5 °C/min from 4°C to −80°C before plunging into liquid nitrogen with a programmable freezer (CryoMed 1010, CryoMed Instruments, Mt. Clemens, MI) or at higher rates in a liquid nitrogen vapor as described previously . In brief, a Styrofoam box (inside dimensions: 20.5×15.5×21.3 cm) was filled to a depth of 4 cm with liquid nitrogen and a 5 cm, 1 cm or 0.4 cm thick Styrofoam ‘boat’ was floated on top of it for 10 min, then straws were placed on top of the ‘boat’, and equilibrated for 10 min before being plunged into liquid nitrogen. The cooling rate of straws while on the ‘boat’ was measured using a data logger thermometer (Type T thermocouple, Omega, Stamford, CT) with the wire inserted into a 0.25-mL straw filled with 50 uL solutions. The actual cooling rates inside the straws depended somewhat on the medium composition with generally slower cooling rate observed at higher cryoprotectant concentration (e.g., 10% ethylene glycol) but similar cooling rates among the other conditions used in this study. Typical curves are shown in Fig. 1. We have used straws filled with 50 uL 3% glycerol-TEST-20% yolk (n = 5) as the standard to measure cooling rates as a function of boat thickness, in order to translate 'boat thickness' to 'cooling rate'. The average cooling rate between −10°C (after the 'plateau' of dissipation of heat of fusion) and −70°C were 29, 220, and 400 °C/min for the 5, 1, and 0.4 cm boats, respectively.
For post-thaw motility estimation, except for the slow thawing method described below, three straws per treatment were thawed in a 37 °C water bath for 30 s (ISOTEMP 102, Fisher Scientific, Pittsburg, PA).
A 10-µl drop of pre-freeze or post-thaw semen, covered with a 22 mm square coverglass, was visualized with 20 × positive-phase objective and a condenser setting of 100 (pseudo-dark field) on an Olympus BH-series phase-contrast microscope (Scientific Instrument Co, Sunnyvale, CA) with a heated stage at 37 °C. The initial motility was in the range of 80% – 99%. Post-thaw motility was estimated without any dilution or washing immediately after thawing or after incubation at 37 °C in 5% CO2 in air for 0.5 h, 1h or 4h, depending on each experimental design. Forward progression was estimated with an adjusted motility index (AMI) as described previously . In brief, the percent forward progression was subjectively estimated with a five point scale (0–4), and this was integrated with the percent motility into one number with the formula as follows: AMI = (Scale value/4) × percent motility. All samples were scored by the same individual and were presented in random order and without identifiers to ensure that motility assessments were not biased.
Our preliminary data showed that 3% and 6% ethylene glycol offered similar protection to monkey sperm as that of glycerol. Here we selected an optimal concentration of ethylene glycol for the subsequent trials with multiple factor experiments, thus we further compared ethylene glycol at 2%, 4%, 6%, 8%, and 10% (equivalent to 0.36–1.79 M) in TEST-20% yolk in using slow pre-cooling in combination with two cooling methods (1 cm vs. 0.4 cm boat) on semen from six males. Post-thaw motility and AMI were evaluated immediately after thawing (0h) or after incubation at 37 °C in 5% CO2 in air for 4h.
We then used a four-way factorial design to evaluate the interactions among four main factors involving in the cryopreservation process. Our preliminary testing of glycerol at different concentrations within its optimal range (3% vs. 6%) did not show any significant difference or interactions among all variables between these two concentrations, thus we chose 3% glycerol along with three other cryoprotectant combinations in this trial. The four factors in this design were: Factor 1-cryoprotectant (4 combinations): 3% glycerol (0.4 M) in TEST-20% yolk, 2% ethylene glycol (0.36 M) in TEST-20% yolk, TEST-20% yolk without any permeable cryoprotectant, and 3% glycerol in TEST without egg yolk; Factor 2-precooling rate (2 levels): fast (placing straws on a metal plate prechilled on crushed ice for 10 min) vs. slow (see above); Factor 3-cooling rate (4 levels): 5, 29, 220, and 400°C/min; Factor 4-thawing rate (2 levels): fast (37°C water bath for 30 s) vs. slow (straws were left in goblets full of liquid nitrogen, and then let liquid nitrogen bubbled off from goblets at RT for 10 min). Thus, this design yields a final combination of 64 conditions (4 × 2 × 4 × 2). For each ejaculate, three straws were used for subsamples at each of the 64 conditions (a total of 192 straws per male). To allow for sufficient sperm volume, samples were loaded at 50 µL per straw at a final freezing density of 5 × 107/ml. This experiment was performed with five males, with a single ejaculate from each male, and each male on a different day. Post-thaw motility and AMI were evaluated after incubation at 37 °C in 5% CO2 in air for 0.5h.
Data were analyzed using repeated measures ANOVA (SAS 9.1) and two or four-way ANOVA. When a significant difference (P < 0.05) was observed among treatments, Tukey’s studentized range test was used for post-test comparisons. Percent motility was arcsine-square root transformed and means of 3 straws per treatment per male (each data point was a mean of 3 straws or subsamples) were used for analysis. Values presented are means ± SD.
Motility of sperm samples at 0h after thawing for 2% ethylene glycol were higher than those at 6% or higher (P < 0.001), and sperm motility at 4h after thawing was higher for the 2–6% ethylene glycol treatment than for 8% or 10% (Table 1). Similar trends were observed for the forward progression motility (AMI). There were no significant differences between the two cooling methods (P = 0.224) for motility and AMI, and interaction between ethylene glycol concentration and cooling rate was not significant (P = 0.538). Samples incubated at 37 °C for 4h after thawing showed a significant loss of motility and AMI (P < 0.001). Thus, ethylene glycol at 2% was chosen for the subsequent experiment.
When data were analyzed with four-way ANOVA, there were significant differences in post-thaw motility and AMI among treatments for the factors of cryoprotectant, cooling rate, and thawing rate (Table 2). Significance of two-factor interactions was observed between cryoprotectant and cooling rate, cryoprotectant and thawing rate, and cooling rate and thawing rate. All three-factor or four-factor interactions were non-significant (Table 2). Therefore, we re-analyzed the data with two-way ANOVA to examine the optimal cooling rate for each combination of cryoprotectant and thawing rate (Table 3, Table 4). Because there were no significant differences between the two pre-cooling methods for all combinations of cryoprotectant and thawing rate (P > 0.144), and also because percent motility and AMI exhibited the exact same patterns in responding to different treatments, the percent motility from the slow pre-cooling group were charted to demonstrate the different responses among males (Fig 2 –Fig 5).
For samples suspended in 3% glycerol in TEST-20% yolk (Table 3, Table 4), when thawed rapidly, the highest post-thaw motility/AMI was observed in samples cooled at 220 °C/min, which was not significantly (P > 0.05) different from those cooled at 29 or 400 °C/min, but higher than those at 5 °C/min (P < 0.035). Three males showed an optimal rate of 220 °C/min, and the other two yielded the highest motility at 29 and 400 °C/min (Fig. 2). When thawed slowly, the highest post-thaw motility/AMI was observed in samples cooled at 5 and 29°C/min, which were not significantly different from each other (P > 0.05), but higher than those cooled at higher rates (220 or 400 °C/min) (P < 0.001). Except for male 4, two males showed an optimal rate of 5 °C/min and another two males showed an optimal rate of 29°C/min (Fig. 2).
For samples suspended in 2% ethylene glycol in TEST-20% yolk (Table 3, Table 4), when thawed rapidly, the highest post-thaw motility/AMI was observed in samples cooled at 400 °C/min, but which was not significantly (P > 0.05) different from others at 5, 29, or 220 °C/min. Except for male 3, all other four males showed the highest post-thaw motility at 400 °C/min (Fig. 3). When thawed slowly, the highest post-thaw motility/AMI was observed in samples cooled at 5 and 29°C/min, which were not significantly different from each other (P > 0.05), but higher than those cooled at higher rates (220 or 400 °C/min) (P < 0.001). However, post-thaw motility/AMI were much lower than samples frozen in 3% glycerol and thawed slowly (Fig. 2). Except for male 1, all other four males showed the highest post-thaw motility at 5 °C/min (Fig. 3).
For samples suspended in TEST-20% yolk without permeable cryoprotectant (Table 3, Table 4), when thawed rapidly, the highest post-thaw motility/AMI was observed in samples cooled at 400 °C/min, which was not significantly (P > 0.05) different from those cooled at 220 °C/min, but higher than those cooled at slower rates of 5 or 29 °C/min (P < 0.001). Except for male 2, all other males had the highest post-thaw motility at 400 °C/min (Fig. 4). When thawed slowly, virtually no motile sperm were observed for any samples regardless of cooling rates (Fig. 4).
For samples suspended in 3% glycol in TEST without egg yolk (Table 3, Table 4), when thawed rapidly, the highest post-thaw motility/AMI was observed in samples cooled at 400 °C/min, which was not significantly (P > 0.05) different from those cooled at 29, or 220 °C/min, but higher than those cooled at 5 °C/min (P < 0.014). Three males had the highest post-thaw motility at 220 °C/min, and two males had the highest post-thaw motility at 400 °C/min (Fig. 5). When thawed slowly, the highest post-thaw motility/AMI was observed in samples cooled at 29°C/min, which were not significantly different from those cooled at 5 °C/min (P > 0.05), but higher than those cooled at faster rates (220 or 400 °C/min) (P < 0.009). The best post-thaw motility and AMI of samples in this group were generally lower than other groups with the presence of egg yolk in the cryoprotectant solutions. Except for male 2, all other males showed the highest post-thaw motility at 5 °C/min, though motility for these samples was generally low (< 25%) (Fig. 5).
This is the first study to simultaneously compare 64 treatments involving 4 factors that are essential to the success of sperm cryopreservation in rhesus macaques. Our findings showed strong interactions among cryoprotectants, cooling, and thawing rates, but not pre-cooling rate on post-thaw motility and forward progression. The lack of significant differences between the two pre-cooling methods suggests that rhesus monkey sperm have a relatively high tolerance to cold shock. This study also showed that rhesus monkey sperm can tolerate a wide range of cooling rates (5–400 °C/min) in the presence of permeable cryoprotectants and egg yolk. It is important to note that the very high cooling rates (220 and 400 °C/min for the 1 and 0.4 cm boat, respectively) occurred during the second phase of freezing (below −10°C), whereas cooling rates were much lower right after the onset of ice formation until reaching −10°C. In this way, the physical properties during the freezing procedure with the Styrofoam box 'automatically' provides the slow rate when sperm need it (during removal of the bulk of intracellular water) and provides a higher rate after most water efflux is finished.
Fast thawing was superior to slow thawing regardless of the types of cryoprotectant or cooling rates. The type of cryoprotectant affected the optimal combination of cooling and thawing for maximal post-thaw survival. When glycerol was used as a permeable cryoprotectant in the presence of egg yolk, slow thawing and fast thawing yielded similar success in some males. However, when glycerol (0.41 M) was replaced with ethylene glycol (0.36 M) for the same treatment, post-thaw motility was significantly lower in samples thawed slowly than those thawed rapidly. In the absence of permeable cryoprotectant but the presence of egg yolk, fast cooling was always favorable. On the contrary, in the absence of egg yolk but the presence of permeable cryoprotectant (glycerol), post-thaw motility was significantly reduced, but especially when samples were thawed slowly.
Fast warming led to higher post-thaw motility suggesting that cryodamage associated with rapid freezing is most likely due to the re-growth of small ice crystals (recrystallization) during slow thawing  or reduced exposure to the concentrated extracellular solutes and cryoprotectant  during the freezing process. This detrimental effect of slow thawing on fast cooled samples was further manifested by freezing sperm in the egg yolk alone without a permeable cryoprotectant as post-thaw motility was approximately zero for all samples. However, a small number of sperm (6%–27% on average) did survive the slow thawing process in the presence of glycerol or ethylene glycol, supporting the hypothesis that permeable cryoprotectants minimize the formation of ice crystals during cooling and/or reduce the recrystallization damage during thawing for fast cooled and slow thawed samples.
The mechanisms of processes associated with slow freezing cryodamage are rather complicated and controversial (see  for review). Existing theories include the rise in solute concentration (especially electrolytes) [16,17], the reduction of the unfrozen fraction  or increase in volume excursion , and the direct result of dehydration of the cell membrane (e.g., ). Consequently, the protective effects of cryoprotectants with low-molecular weight have been ascribed to their nonspecific colligative ability to reduce the solute concentration, increase the unfrozen fraction, or reduce volume excursion, or alternatively, ascribed to solute-specific interactions with the phospholipid bilayers that lead to the stabilization of the plasma membrane . Previous studies have demonstrated that responses to thawing rate were variable with slowly cooled cells; in some reports there were no differences between slow and rapid thawing, while in other reports slow thawing is either better or worse than rapid thawing depending on conditions and/or species . In this study, except for male 5 in the glycerol and egg yolk treatment group, fast thawing in general yielded higher post-thaw motility than slow thawing for slowly cooled rhesus monkey sperm (5 °C/min). This pattern agrees with previous findings for other mammalian sperm such as ram , boar , human , mouse , and bovine . Possible explanations for this outcome could be any or all of the protective mechanisms indicated above, or due to the reduced ability of the cell membrane to withstand deformation at low temperature , or simply because of an increase in the total exposure time of cells at subzero temperatures for slow thawing . Another possibility is that devitrification of glassy water contributes to cryodamage during the slow thawing process. Even at low cooling rates, a portion of freezable water may undergo vitrification to form a metastable glass [23,24].
Although glycerol has been the most widely used cryoprotectant for non-human primates, recent measurements of osmotic tolerance and membrane permeability characteristics of rhesus monkey sperm suggest that an alternative cryoprotectant such as ethylene glycol may be more appropriate as it has the highest permeability coefficient among five tested cryoprotectants including glycerol . The present study verified the effectiveness of ethylene glycol as a cryoprotectant for rhesus monkey sperm in samples cooled slowly or quickly and thawed rapidly, but not in samples subject to slow thawing. Agca et al.  suggested that ethylene glycol may be the most appropriate cryoprotectant for rhesus sperm cryopreservation, as ethylene glycol could lead to less osmotic stress than glycerol during addition and removal of cryoprotectant. Despite that we found no advantage of using ethylene glycol over glycerol. The use of 2% ethylene glycol in combination with egg yolk yielded post-thaw motility that is comparable to that of 3% glycerol when samples were thawed rapidly. Slow thawing, however, led to less than half of the yield in post-thaw motility for samples suspended in ethylene glycol (10.9%–20.2%) than those suspended in glycerol (41.5% –45.4%). Thus the protection from permeable cryoprotectants in slow freezing and slow thawing of rhesus monkey sperm seemed in part to be related to solute-specific interactions, in addition to the non-specific colligative mechanisms, since glycerol (0.41 M) and ethylene glycol (0.36 M) at approximately equal molar concentrations gave different levels of protection. Our findings also indicate that glycerol still is the most suitable cryoprotectant for rhesus monkey sperm, and further refinement of concentration from 1% to 5% showed no significant differences in retaining post-thaw motility or percent forward progression (data not shown).
The importance of membrane stabilization in the ability of sperm to overcome the damage associated with the cryopreservation process is further evidenced by findings in the present study where egg yolk alone without permeable cryoprotectant subjected to fast cool and rapid thaw yielded higher post-thaw motility than samples suspended in TEST-glycerol without egg yolk. Though the exact means by which egg yolk helps sperm survive the cryopreservation process is unknown, it has generally been assumed that egg yolk stabilizes the sperm plasma membrane either directly, by forming a protective film  or replacing lost phospholipids [11,12,30], or indirectly by reducing the deleterious effect of seminal plasma proteins on the sperm membrane . Yet, in the rhesus, a direct association of egg yolk with the sperm membrane seems to be the more likely explanation as egg yolk has been found to be equally necessary for protecting epididymal sperm as it is for ejaculated sperm .
It is worth noting that the optimal cooling rate for any particular ejaculate/male varied with the choice of cryoprotectant for either slow or fast thaw scenarios. For example, with rapid thawing, the optimal cooling rates for single ejaculate collected from males 1 and 3 could be 29, 220, or 400 °C/min depending on which cryoprotectant is used, while ejaculates from males 2, 4, and 5 exhibited a narrower optimal range of 220 to 400 °C/min for the same treatments. Although more replicates of ejaculates per male would be necessary to reveal whether there is male-to-male variation in response to different freezing parameter combinations, this phenomenon has been observed in our previous report . Despite the broad recognition of the male variation phenomenon in sperm cryopreservation of almost any species, few studies have examined the underlying causes. Future research efforts on this aspect of cryopreservation may also improve the understanding of mechanisms related to freezing and thawing damage. It follows that development of individually customized freezing protocols should be a high priority for future research due to the unlikelihood that a single freezing procedure will be equally effective for all males of the same species . In fact, the development of customized freezing protocols based on males’ susceptibility to cryopreservation is the ultimate approach for semen banking to reach its full potential because the genetics of many males with poor freezibility could also be represented. Otherwise, sperm cryopreservation programs will select for the genetics of only those males whose sperm can survive this process. Multiple factor designs similar to the one used in this study could serve as powerful tool to efficiently identify different combinations of optimal cryopreservation parameters for individual males.
We thank S. Rodenburg and T. Tollner for critical review.
This work was supported by NIH grants RR00169 and RR13439.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.