Cryopreservation of mammalian embryos: Derivation of a method.

In 1972, a procedure was derived to cryopreserve mouse embryos. Over the past four decades, this procedure has been adapted to freeze embryos of more than twenty-five mammalian species. Cryopreservation of embryos has become a routine procedure in both veterinary and human medicine, having been used to freeze millions of embryos of mice and cattle, and many hundreds of thousands of human embryos. After transfer into appropriate foster mothers, cryopreserved embryos have developed into innumerable live offspring. This article describes the background that led to the derivation of the procedure and the events that transpired during its development. The first successful embryo cryopreservation procedure was developed by collaboration of three investigators, each bringing a special expertise and perspective to the project.

Birth of a domestic cat kitten produced by vitrification of lipid polarized in vitro matured oocytes.

The ability to cryopreserve oocytes is an effective method to retain valuable genetic material of mammals, including that of endangered animals. Embryos of domestic cats are amenable to cryopreservation, whereas their oocytes are much less cryo-tolerant. The capability of oocytes to survive cryopreservation is affected by several factors, one of which has been hypothesized to be the high concentration of intracellular lipids. To test this hypothesis, in this study we polarized lipids of cat oocytes and tested their cooling and freezing sensitivity. We found that the sensitivity of oocytes to cooling and cryopreservation does appear to be related to their high intracellular lipid content, as indicated by higher cryosurvival and development into blastocysts when intracellular lipids of in vitro matured oocytes were polarized before vitrification. However, polarization of all intracellular lipids was detrimental to development of embryos. Cell numbers in blastocysts derived from fully polarized/vitrified oocytes were significantly lower than those of partially polarized/vitrified or non-vitrified/fresh oocytes. Although embryos derived from fully polarized/vitrified oocytes developed to the blastocyst stage at higher rates than those of partially polarized/vitrified or non-centrifuged/vitrified oocytes, their in vivo developmental competence was compromised. When embryos derived from fully polarized/vitrified oocytes were transferred, although two recipients became pregnant, all implanted embryos were reabsorbed. In contrast, when embryos derived from oocytes that were only partially lipid polarized before vitrification and then were transferred, one recipient did become pregnant and produced a live healthy kitten. The present results suggest that other approaches to altering intra-cellular lipid levels in cat oocytes should be evaluated to improve their functional survival after cryopreservation.

Different patterns of metabolic cryo-damage in domestic cat (Felis catus) and cheetah (Acinonyx jubatus) spermatozoa.

Felid spermatozoa are sensitive to cryopreservation-induced damage, but functional losses can be mitigated by post-thaw swim-up or density gradient processing methods that selectively recover motile or structurally-normal spermatozoa, respectively. Despite the importance of sperm energy production to achieving fertilization, there is little knowledge about the influence of cryopreservation or post-thaw processing on felid sperm metabolism. We conducted a comparative study of domestic cat and cheetah sperm metabolism after cryopreservation and post-thaw processing. We hypothesized that freezing/thawing impairs sperm metabolism and that swim-up, but not density gradient centrifugation, recovers metabolically-normal spermatozoa. Ejaculates were cryopreserved, thawed, and processed by swim-up, Accudenz gradient centrifugation, or conventional washing (representing the 'control'). Sperm glucose and pyruvate uptake, lactate production, motility, and acrosomal integrity were assessed. Mitochondrial membrane potential (MMP) was measured in cat spermatozoa. In both species, lactate production, motility, and acrosomal integrity were reduced in post-thaw, washed samples compared to freshly-collected ejaculates. Glucose uptake was minimal pre- and post-cryopreservation, whereas pyruvate uptake was similar between treatments due to high coefficients of variation. In the cat, swim-up, but not Accudenz processing, recovered spermatozoa with increased lactate production, pyruvate uptake, and motility compared to controls. Although confounded by differences in non-specific fluorescence among processing methods, MMP values within treatments were positively correlated to sperm motility and acrosomal integrity. Cheetah spermatozoa isolated by either selection method exhibited improved motility and/or acrosomal integrity, but remained metabolically compromised. Collectively, findings revealed a metabolically-robust subpopulation of cryopreserved cat, but not cheetah, spermatozoa, recovered by selecting for motility rather than morphology.

Integrating microsatellite and pedigree analyses to facilitate the captive management of the endangered Mississippi sandhill crane (Grus canadensis pulla).

The minimization of kinship in captive populations is usually achieved through the use of pedigree information. However, pedigree knowledge alone is not sufficient if pedigree information is missing, questionable, or when the founders of the captive population are related to one another. If this is the case, higher levels of inbreeding and lower levels of genetic diversity may be present in a captive population than those calculated by pedigree analyses alone. In this study, the genetic status of the critically endangered Mississippi sandhill crane (MSC) (Grus canadensis pulla) was analyzed using studbook data from the U.S. Fish and Wildlife Service managed captive breeding program as well as microsatellite DNA data. These analyses provided information on shared founder genotypes, allowing for refined analysis of genetic variation in the population, and the development of a new DNA-based studbook pedigree that will assist in the genetic management of the MSC population.

Oxidative phosphorylation is essential for felid sperm function, but is substantially lower in cheetah (Acinonyx jubatus) compared to domestic cat (Felis catus) ejaculate.

Compared with the normospermic domestic cat, sperm metabolic function is compromised in the teratospermic cat and cheetah, but the pathway(s) involved in this deficiency are unknown. Glycolysis is essential for sperm motility, yet it appears to function normally in spermatozoa of either species regardless of structural morphology. We conducted a comparative study to further understand the mechanisms of energy production in felid spermatozoa, with the hypothesis that oxidative phosphorylation is required for normal sperm function and is impaired in teratospermic ejaculates. Electroejaculates from both species were stained with MitoTracker to quantify mitochondrial membrane potential (MMP) or were incubated to assess changes in sperm function (motility, acrosomal integrity, and lactate production) after mitochondrial inhibition with myxothiazol. Sperm midpiece dimensions also were quantified. Sperm mitochondrial fluorescence (directly proportional to MMP) was ~95% lower in the cheetah compared with the normospermic and teratospermic cat, despite the cheetah having a 10% longer midpiece. In both species, MMP was increased 5-fold in spermatozoa with retained cytoplasm compared with structurally normal cells. Inhibition of oxidative phosphorylation impaired sperm function in both species, but a 100-fold higher inhibitor concentration was required in the cat compared with the cheetah. Collectively, findings revealed that oxidative phosphorylation was required for sperm function in the domestic cat and cheetah. This pathway of energy production appeared markedly less active in the cheetah, indicating a species-specific vulnerability to mitochondrial dysfunction. The unexpected, cross-species linkage between retained cytoplasmic droplets and elevated MMP may reflect increased concentrations of metabolic enzymes or substrates in these structures.

Glycolytic enzyme activity is essential for domestic cat (Felis catus) and cheetah (Acinonyx jubatus) sperm motility and viability in a sugar-free medium.

We have previously reported a lack of glucose uptake in domestic cat and cheetah spermatozoa, despite observing that these cells produce lactate at rates that correlate positively with sperm function. To elucidate the role of glycolysis in felid sperm energy production, we conducted a comparative study in the domestic cat and cheetah, with the hypothesis that sperm motility and viability are maintained in both species in the absence of glycolytic metabolism and are fueled by endogenous substrates. Washed ejaculates were incubated in chemically defined medium in the presence/absence of glucose and pyruvate. A second set of ejaculates was exposed to a chemical inhibitor of either lactate dehydrogenase (sodium oxamate) or glyceraldehyde-3-phosphate dehydrogenase (alpha-chlorohydrin). Sperm function (motility and acrosomal integrity) and lactate production were assessed, and a subset of spermatozoa was assayed for intracellular glycogen. In both the cat and cheetah, sperm function was maintained without exogenous substrates and following lactate dehydrogenase inhibition. Lactate production occurred in the absence of exogenous hexoses, but only if pyruvate was present. Intracellular glycogen was not detected in spermatozoa from either species. Unexpectedly, glycolytic inhibition by alpha-chlorohydrin resulted in an immediate decline in sperm motility, particularly in the domestic cat. Collectively, our findings reveal an essential role of the glycolytic pathway in felid spermatozoa that is unrelated to hexose metabolism or lactate formation. Instead, glycolytic enzyme activity could be required for the metabolism of endogenous lipid-derived glycerol, with fatty acid oxidation providing the primary energy source in felid spermatozoa.

Production of channel catfish with sperm cryopreserved by rapid non-equilibrium cooling.

This report describes the feasibility of using vitrification for fish sperm. Vitrification can be used to preserve samples in the field and offers an alternative to conventional cryopreservation, although it has not been systematically studied for sperm of aquatic species. The overall goal of the project was to develop streamlined protocols that could be integrated into a standardized approach for vitrification of aquatic species germplasm. The objectives of the present study in channel catfish (Ictalurus punctatus) were to: (1) evaluate the acute toxicity of 5%, 10%, 20% and 30% methanol, N,N-dimethyl acetamide, dimethyl sulfoxide, 1,2-propanediol, and methyl glycol; (2) evaluate a range of devices commonly used for cryopreservation and vitrification of mammalian sperm; (3) compare vitrification with and without cryoprotectants; (4) evaluate the post-thaw membrane integrity of sperm vitrified in different cryoprotectant solutions, and (5) evaluate the ability of vitrified sperm to fertilize eggs. Cryoprotectant concentrations of higher than 20% were found to be toxic to sperm. Methanol and methyl glycol were the least toxic at a concentration of 20% with an exposure time of less than 5 min. We evaluated a method reported for human sperm, using small volumes in loops (15 µl) or cut standard straws (20 µl) with and without cryoprotectants plunged into liquid nitrogen. Cryoprotectant-free vitrification using loops did not yield fertilization (assessed by neurulation), and the fertilization rates observed in two trials using the cut standard straws were low (~2%). In general, fertilization values for vitrification experiments were low and the use of low concentrations of cryoprotectants yielded lower fertilization (<10%) than the use of vitrification solutions containing high cryoprotectant concentrations (as high as 25%). The highest neurulation obtained was from a mixture of three cryoprotectants (20% methanol+10% methyl glycol+10% propanediol) with a single-step addition. This was reflected in the flow cytometry data from which the highest membrane integrity using loops was for 20% methanol+10% methyl glycol+10% propanediol (~50%). We report the first successful sperm vitrification in fish and production of offspring from vitrified sperm in channel catfish. Although the fertilization values were low, at present this technique could nevertheless be used to reconstitute lines (especially in small aquarium fishes), but it would require improvement and scaling up before being useful as a production method for large-bodied fishes such as catfish.

Survival of mouse embryos frozen to -196 degrees and -269 degrees C.

Mouse embryos survived freezing to -196 degrees C. Survival required slow cooling (0.3 degrees to 2 degrees C per minute) and slow warming (4 degrees to 25 degrees C per minute). Depending on the specific rates used, 50 to 70 percent of more than 2500 frozen and thawed early embryos developed into blastocysts in culture after storage at -196 degrees C for up to 8 days. When approximately 1000 of the survivors, including some frozen to -269 degrees C (4 degrees K), were transferred into foster mothers, 65 percent of the recipients became pregnant. More than 40 percent of the embryos in these pregnant mice gave rise to normal, living full-term fetuses or newborn mice.

Cryopreservation of canine ovarian and testicular fibroblasts.

To derive a practical procedure to store canine somatic cells, fibroblasts isolated from testicular or ovarian tissues were cryopreserved in 1.2 M ethylene glycol or in 1.2 M dimethylsulfoxide prepared in Dulbecco's Modified Eagle Medium as cryoprotectants, and were frozen either in plastic straws or vials. Thawed cells were cultured for 24 hr at 38.5 degree C in a humidified atmosphere of 5 percent CO2 95 percent air, and then their membrane integrity was assayed with a double fluorescent stain, Fertilight. In addition, frozen-thawed fibroblasts were cultured for 4 days, and then their functional survival was measured after staining small colonies with trypan blue. After freezing and thawing, membrane integrity of testicular fibroblasts was 55-70 percent and functional survival ranged from 20-40 percent. With frozen-thawed ovarian cells, the average membrane integrity was 55-75 percent and the average functional survival was 35-40 percent. When frozen in ethylene glycol, functional survival of ovarian fibroblasts was significantly higher than that of testicular cells (P less than 0.05). These methods should prove useful to preserve cells collected from canids in the wild.

Cryopreservation of oocytes and embryos: optimization by theoretical versus empirical analysis.

Embryos and oocytes were first successfully cryopreserved more than 30 years ago. This procedure has come to be an important, almost essential component in the practice of assisted reproduction in animals and humans. Literally millions of animals of more than 20 species and undoubtedly hundreds of thousands of children have been born from frozen embryos. Nevertheless, there still remain shortcomings with methods used to cryopreserve oocytes and embryos. A wide variety of approaches has been used to try to improve and optimize methods of cryopreservation. These approaches range from the very basic to the completely empirical. Some investigators make use of rigorous mathematical formulations to define and describe the behavior of oocytes and embryos at subzero temperatures. Others conduct "trial-and-error" studies to improve the results by dint of many replicate experiments in which they examine the effects of various protective compounds, macromolecular supplements, and compare different cooling and warming conditions. This review considers both extremes.

Origins of techniques in human and animal embryology.

Increasingly innovative and imaginative techniques are being developed to investigate the development of animal and human embryos. Among the types of techniques that have been developed are ones that deal with oocyte maturation and culture, the isolation and utilization of stem cells, cryopreservation of reproductive cells and tissues, and various procedures to manipulate early embryos. To appreciate the derivation of these sophisticated techniques, it seems appropriate to consider the very early origins of these current techniques.

Effect of cooling and exposure to ethylene glycol on in vitro maturation and embryo development of rhesus oocytes.

The meiotic spindle of metaphase II-stage oocytes is damaged when mature oocytes are cooled to temperatures close to 0 degrees C, as occurs during cryopreservation by equilibrium cooling. Since a spindle has not yet formed within a germinal vesicle-stage oocyte, it has been suggested that immature oocytes may be more resistant than metaphase II oocytes to cryopreservation by equilibrium cooling. To test this proposition, we examined the effects on rhesus macaque oocytes of chilling and exposure to ethylene glycol (EG) on their maturation and embryo development. A total of 202 cumulus-intact oocytes was collected from adult female rhesus monkeys that had been given follicle stimulating hormone for controlled ovarian hyperstimulation. Within two hours of their having been aspirated and prior to germinal vesicle breakdown, oocytes were either cooled to 0 degrees C for 10 minutes or were exposed for 15 minutes at 35 degrees C to 1.5 M EG to be tested as a possible cryoprotectant. After being exposed, oocytes were cultured in maturation medium, fertilized in vitro with rhesus spermatozoa, and cultured. The maturation rate and subsequent development into blastocysts of those oocytes that had been exposed to EG or cooled to 0 degrees C did not differ significantly from untreated control oocytes. Additional germinal vesicle oocytes were exposed to 1.5 M EG at 35 degrees C for 3 minutes and then supercooled to -7 degrees C or frozen at -7 degrees C or frozen at 0.5 degrees C to -35 degrees C. Rates of maturation and embryo development of oocytes cooled to or frozen at -7 degrees C were significantly lower than rates for control oocytes; none of those frozen to -35 degrees C even underwent maturation. These results suggest that germinal vesicle-stage oocytes may be less susceptible to injury resulting from chilling or exposure to ethylene glycol, but are still damaged by freezing.

Pregnancy resulting from cryopreserved human embryos using a one-step in situ dilution procedure.

In vitro fertilization and embryo transfer require the use of hormonal manipulation and surgery that may reduce the receptivity of the patient's uterus during the stimulated cycle. Cryopreservation of human embryos eliminates the need for immediate transfer, permitting them to be stored until they can be transferred during subsequent unstimulated cycles. Embryo cryopreservation is an established procedure in the breeding of laboratory and domestic animals, but has only recently been applied to humans. We report on a pregnancy using a simple cryopreservation procedure that permits embryos to be diluted out of the cryoprotectant solution without removing them from the plastic straw in which they were cryopreserved.

Effects of chilling to 0 degrees C on the morphology of meiotic spindles in human metaphase II oocytes.

OBJECTIVE: To determine the effects of chilling to 0 degrees C on the meiotic spindle of human metaphase II oocytes, as observed by optical sectioning microscopy. DESIGN: Laboratory study. SETTING: Academic research laboratory in a medical school. PATIENT(S): Seventy-two women undergoing infertility treatment donated a total of 108 oocytes. INTERVENTION(S): Metaphase II oocytes were stripped of their cumulus cells, cooled directly to 0 degrees C, and held for periods of 1 to 10 minutes. They were then fixed at 37 degrees C, stained for immunofluorescence, and examined microscopically. MAIN OUTCOME MEASURE(S): Morphology of the meiotic spindle in chilled and control oocytes. RESULT(S): Microscopic evaluations of 46 chilled oocytes revealed various time-dependent changes in microtubules compared to 9 control oocytes. After 1 minute at 0 degrees C, spindle damage was negligible, but in oocytes cooled for 2 or 3 minutes, there was obvious shortening of the spindle and loss of polarity. Cooling to 0 degrees C for 4 to 9 minutes resulted in increasingly more drastic changes; by 10 minutes the spindles had totally disappeared. Despite depolymerization of microtubular tubulin at 0 degrees C, the chromosomes did not become dispersed, but remained anchored even in the absence of spindles. CONCLUSION(S): Even brief exposure of human oocytes to temperatures near 0 degrees C causes profound alterations of the meiotic spindle.

The efficacy of cryopreserved hamster ova in the sperm penetration assay.

To solve the logistical problems of the sperm penetration assay (SPA) to provide just a sufficient number of hamster ova exactly when they are needed, a new method to cryopreserve the ova has been devised. The ova, suspended in a 1.5 M solution of propylene glycol as a cryoprotectant in an isotonic salt solution, were frozen in 1/4 mL plastic straws. Included in each straw was a sucrose solution, isosmotic to the propylene glycol solution, to serve as an osmotic buffer during dilution of the cryoprotectant out of the ova. This one-step method of dilution permitted the ova to be recovered and diluted out of the cryoprotectant within the straw in which they had been originally frozen. A total of 547 cryopreserved ova were thawed, 504 (92.1%) of which were morphologically normal after they had been incubated at 37 degrees C for 3 hours. After removal of the zonae, the frozen-thawed ova were compared with fresh, control ova in SPAs of donor and patient semen that had been capacitated in TEST-yolk buffer. The percent penetration and penetration index of fresh versus cryopreserved ova did not differ significantly for either donor or patient semen.

Cryopreservation of germ cells from bovine fetal ovaries.

Bovine fetuses at stages required for studies of female germ cells (primordial germ cells and oogonia) become available from the abattoir at unpredictable times. To alleviate this logistical problem, a procedure to cryopreserve these ovarian germ cells has been devised. Fetal ovarian cells were dispersed and suspended in 1.5 M dimethyl sulfoxide (DMSO) prepared in modified TCM 199 medium. The suspensions were aspirated into plastic semen straws, cooled, seeded to induce ice formation at -7 degrees C, and then cooled at 1 degree C min-1 to -70 degrees C before being plunged into liquid nitrogen at -196 degrees C for storage. The straws were thawed at a moderate rate of approximately 250 degrees C min-1, the DMSO was diluted 28-fold with culture medium, and then the cells were cultured for > 2 h before their viability was tested or they were used for nuclear transfer. No statistically significant difference in viability before and after cryopreservation was detected by vital staining with fluorescein diacetate (P > 0.05). When frozen-thawed germ cells were fused to cytoplasts, the cleavage rate of the resultant reconstructed embryos 44 h after fusion was 31%, although none developed into blastocysts. It is concluded that cryopreservation of bovine fetal ovarian germ cells is feasible and can play a major role in facilitating future experimentation.

Glycine betaine improves survival of fresh bovine spermatozoa.

Fresh spermatozoa from bulls established as 'good freezers' and 'poor freezers' (consistently > or = 50% or < 20% motile spermatozoa after cryopreservation, respectively) were incubated for 96 h in Tes/Tris-egg yolk or TALP-egg yolk media at 37 degrees, 20 degrees , 5 degrees or 0 degrees C. The TALP extender contained 0, 100 or 200 mM glycine betaine (GB) to test the hypothesis that GB would efficiently maintain spermatozoa function during long-term incubation. The percentage of motile spermatozoa declined over time in a temperature- and medium-dependent fashion. No spermatozoa were motile by 24 h incubation at 37 degrees C or by 72 h incubation at 0 degrees C, and there were no significant differences in the percentage of motile spermatozoa from either category of bull when spermatozoa were incubated in any media for less than 24 h. Spermatozoa from poor freezers were significantly more motile than spermatozoa from good freezers after 96 h at 20 degrees or 5 degrees C in TALP alone; however, GB at both 100 and 200 mM increased the percentage of motile spermatozoa in poor and good freezers and eliminated these differences. Overall, the presence of GB at either 100 or 200 mM significantly improved the percentage of motile spermatozoa at 20 degrees, 5 degrees and 0 degrees C, but not at 37 degrees C.

Relationship between the characteristics of frozen-thawed ram spermatozoa and in vitro embryo production.

To select rams suitable for ovine in vitro embryo production (IVP), the predictive values of the screening tests used to identify unsuitable rams need to be established. The present study examined some characteristics of frozen-thawed ram spermatozoa that might be evaluated routinely in a commercial breeding programme. These included sperm motility, plasma membrane integrity, morphology, and acrosome and capacitation status of the sperm population. Cryopreserved spermatozoa from four Dorset rams, which had previously satisfied the selection criteria for inclusion in a commercial breeding programme, were used for IVP. The overall contribution of the four rams and the ejaculates within each ram to the variability (R2) in the production of blastocysts was very small (2.1% and 2.5% respectively). The analysis of the sperm characteristics by logistic regression revealed a significant and positive association between total post-thaw sperm motility, viability and longevity with in vitro blastocyst production. However, there was no association between the other surface characteristics of the spermatozoa measured in this study with embryo production. Despite the absence of differences between the rams in the low incidence of polyspermic fertilization, the significant and detrimental effects of polyspermic fertilization on in vitro blastocyst production rates were quantified by logistic regression analysis. A large proportion of the variability within the IVP system was unaccounted for by the analysis of sperm and oocyte characteristics evaluated in this study. Thus, the identification of other factors contributing to the variability in the production of embryos in vitro warrants further investigation. No single sperm characteristic was sufficient to predict the ultimate outcome of blastocyst production. Rather, assessments of multiple characteristics within the IVP system are required to make accurate predictions.

A one-step method for direct nonsurgical transfer of frozen-thawed bovine embryos.

One impediment to the more widespread use of freezing as an adjunct to embryo transfer in cattle has been the method used to remove the protective compound from the thawed embryos. Recently, a method has been developed that permits bovine embryos to be diluted out of the protective solution within the plastic straw in which the embryos were originally rrozen and thawed. The method requires only about 10 minutes to perform and does not require a microscope or other laboratory equipment. Therefore, frozen bovine embryos can be thawed, diluted, and transferred nonsurgically into recipient cattle under conditions quite similar to those used for artificial insemination. A large number of field trials of this method have been performed during the past 2 1 2 years. A total of 327 pregnancies have been established by the nonsurgical transfer of 1259 embryos that had been frozen, thawed, and diluted by this method.

Recovery of motile, membrane-intact spermatozoa from canine epididymides stored for 8 days at 4 degrees C.

The purpose of this study was to determine how long canine spermatozoa remain motile and with intact membranes when maintained within epididymides stored at 4 degrees C, and to determine whether such stored spermatozoa are able to bind to canine zonae pellucidae. Testes with attached epididymides, obtained from 32 dogs (26 purebred; six mixed breeds) at orchiectomy, were refrigerated at 4 degrees C, and spermatozoa were collected from caudae epididymides at nine time intervals ranging from 5 to 192 h. The effects on spermatozoa that had been refrigerated within epididymides for various times were determined by assaying sperm motility, integrity of plasma membranes and of acrosomes, and measuring binding of membrane-intact spermatozoa to canine zonae pellucidae. Membrane integrity was assessed using a double fluorescent dye, and acrosome integrity by staining with Pisum sativum agglutinin. For the zona-binding assay at various refrigeration time points, duplicate sets of six oocytes each, isolated from ovaries retrieved at elective ovariohysterectomy, were placed into 100 microl droplets of sperm capacitation medium containing 5 x 10(6) spermatozoa/ml. One minute later, oocytes were rinsed vigorously by pipetting, and then incubated for 1 h at 38.5 degrees C in a humidified atmosphere of 5% CO2 in air; the number of membrane-intact spermatozoa bound to zonae were counted. There was no significant decrease in membrane integrity and acrosome integrity of spermatozoa recovered from epididymides stored at 4 degrees C within the first 48 h of refrigeration. In contrast, sperm motility decreased significantly within the first 5 h of refrigeration (P < 0.05), but then declined more gradually thereafter. Some spermatozoa recovered from epididymides that had been refrigerated for 192 h retained their capability to bind to zonae pellucidae, although the mean number of refrigerated spermatozoa (0.4) bound to zonae was less than that of fresh samples (9.0). Membrane integrity of spermatozoa recovered from epididymides refrigerated for various times was highly correlated (r = 0.88) with sperm motility. Even after storage for 192 h (8 days) at 4 degrees C, motile spermatozoa could be recovered from the epididymides, and such refrigerated spermatozoa were capable of binding to zonae. We interpreted these data to indicate that it might be possible to recover functional spermatozoa from postmortem specimens of domestic and nondomestic canids.