. al., 1989; Drobnis et al., 1993; Holt and North, 1984; 1986; Parks and Lynch, 1992
typically within the temperature range 17–368C. Their occurrence shows species depen- dence, which could go some way towards explaining the variations in cryopreservation
sensitivity seen in spermatozoa from different species. It is also likely that during a typical freeze–thaw cycle, the sperm membranes must undergo phase transitions during
both cooling and rewarming.
Ž .
Evidence that cold shock i.e. damage due to rapid cooling above 08C is caused by Ž
. Ž
. lipid phase transition effects was presented by Drobnis et al. 1993 . Holt et al. 1992
obtained some evidence that phase transitions might be involved in the manifestation of cryoinjury during the rewarming of cells after thawing. Ram spermatozoa were stained
Ž .
with fluorescein diacetate FDA , a fluorescent probe of cell membrane integrity, cooled Ž
. to a series of minimum temperatures 58C, y108C and y208C and then rewarmed to
308C. Plasma membrane integrity was retained throughout cooling, but fluorescein leakage, indicating membrane disruption, occurred during the rewarming process.
Conversely, performing similar experiments in the presence of external adenosine
Ž .
triphosphate ATP and a sperm reactivation medium, showed that spermatozoa rendered immotile by cooling could be restored to motility by an influx of ATP when the plasma
membrane was breached. The threshold temperatures causing loss of membrane integrity were correlated with the minimum temperature reached during cooling. One interpreta-
tion of this data is that as the post-thaw temperature increases, the plasma membrane is subjected to structural rearrangements involving lipids and proteins, the extent and
nature of which are governed by interactions of temperature and solute effects during the freezing process.
Besides causing physical disruption of the plasma membrane by the induction of lipid packing faults, lipid phase transition effects cause non-linear kinetic responses in some
enzymes, including some of the membrane ATPases whose activity depends upon the Ž
. physical state of annular lipids Kimelberg, 1977 . It is likely that such effects are partly
responsible for the poor control of intracellular calcium concentration which is evident Ž
. at temperatures below about 178C Bailey et al., 1994; Robertson and Watson, 1986 .
Ž .
This is probably the rationale for including ethylenediaminetetra-acetic acid EDTA and citrate in some semen diluents; these would chelate calcium and diminish the concentra-
tion gradient across the sperm plasma membrane. Intracellular calcium concentrations Ž
. ; 0.1 mM are four orders of magnitude lower than those in the external milieux.
EDTA chelates other metallic ions, however, and might also act by inhibiting lipid peroxidation.
3. Cryoprotectants and additives
Ž Many compounds have been tested for their efficacy as sperm cryoprotectants see
. for example, Jeyendran and Graham, 1980; Molinia et al., 1994 , but most semen
preservation protocols still favour glycerol in the cryoprotective media, following the Ž
. example set by Polge et al., 1949 . In certain instances other cryoprotectants are
Ž .
possibly better; for example, dimethyl sulphoxide DMSO was preferred for elephant
Ž .
spermatozoa Jones, 1973 . The choice of cryoprotectant seems to have been a matter of trial and error in nearly all investigations; this is partly because a complete and
satisfactory explanation for the action of cryoprotectants does not exist. Cryoprotectant compounds can be roughly classified into groups, with differing
modes of action. Glycerol, together with substances such as methanol, ethylene glycol, 1,2-propanediol, butanediol, acetamide and DMSO, belong to a group which permeate
Ž .
into the cellular cytoplasm. Lovelock and Polge 1954 proposed that the protective effects of glycerol were mediated by its colligative properties; depression of freezing
point and the consequent lowering of electrolyte concentrations in the unfrozen fraction at any given temperature would help to counter the harmful ‘‘solution effects’’ imposed
during the freezing process. Although this hypothesis has found support through a
Ž .
number of investigations see Mazur, 1984 , it is becoming increasingly apparent that this is not the only way in which glycerol might protect cells during freezing. It is also
evident that glycerol is somewhat cytotoxic to spermatozoa. Species differences in ability to withstand glycerol exposure therefore interact in a complex fashion with the
freezing rates used, and the degree of cryoprotection conferred.
Ž .
Hammerstedt and Graham 1992 , addressing the issue of poultry sperm cryopreser- vation, reviewed the actions of glycerol on spermatozoa, which in this instance include
the abolition of fertilizing ability. Glycerol removal does, however, restore fertility. They pointed out that since glycerol reaches the interior of the cell it probably affects
cytoplasmic viscosity, thereby changing rates of all diffusion limited processes. Previous
Ž .
evidence that cytoplasmic viscosity differs between species Hammerstedt et al., 1978 suggests that glycerol could have species-specific effects on spermatozoa. This argument
can be applied more generally to other permeating cryoprotectants such as DMSO, ethylene glycol and methanol. Experimentally, it is also known that glycerol is able to
Ž .
insert into the membrane bilayer; Hammerstedt and Graham 1992 suggested that exposure of cells to 0.5M glycerol in cryoprotective media would yield an intramembra-
nous concentration of about 1 mM. This might contribute to the alteration of cell membrane properties by inducing changes in lipid packing structure and hence the
stability and water permeability of the cell membrane would be altered. Membrane fusogenicity and the responses of signal transduction pathways could also be affected by
such changes, thus contributing to the possibility that post-thaw sperm longevity is
Ž .
reduced through accelerated capacitation Watson, 1995 . Glycerol itself is known to be metabolized by ram, bull, boar and goat spermatozoa,
Ž .
see for example, Aalbers et al., 1961; Jones et al., 1992 ; the metabolic pathway differs from that operating in tissues such as the liver where glycerol is phosphorylated by a
Ž kinase, being recruited instead by an NADP-dependent dehydrogenase Jones et al.,
. Ž
. 1992 . Hammerstedt et al., 1990 argued in some detail that in its capacity as a
substrate, glycerol would alter the bioenergetic status of spermatozoa, perhaps interfer- ing with the balance between ATP synthesis and utilization. If an ATP deficit occurred,
for example during cooling, metabolic control over ion-dependent cellular processes might be compromised, leading to inappropriate activation of phospholipases and
proteases and irreversible cell damage. These authors did not believe that a mismatch between ATP synthesis and utilization occurred in bull or ram spermatozoa, but
considered that the concept may be valid for other species.
In this context, it is worth noting that interspecific variation in glycerol tolerance can be very marked. Recent studies of marsupial spermatozoa have indicated that they have
Ž an unusually high tolerance of glycerol Johnston et al., 1993; Molinia and Rodger,
. 1996; Rodger et al., 1991; Taggart et al., 1996 . Indeed, it seems that unless concentra-
tions exceed about 10, and approach 20, there is little recovery of motility although the fertility of these spermatozoa has yet to be investigated. In contrast to these
requirements for high glycerol concentrations, boar spermatozoa suffer loss of fertility if
Ž .
the concentration exceeds 3 Johnson, 1985 , an effect caused by increased acrosomal damage. Mouse spermatozoa, whose preservation has also been investigated recently,
Ž seem unable to withstand more than about 1.75 glycerol Penfold and Moore, 1993;
. Tada et al., 1990, 1993 . Some investigators have avoided glycerol completely for
mouse spermatozoa, preferring instead a diluent based upon the non-permeating cryopro- Ž
. tectant raffinose
Nakagata and Takeshima, 1993 . Despite the possible perturbing actions of glycerol discussed above, there is at present no plausible explanation for these
interspecific differences in cryoprotectant sensitivity. While comparative studies of semen cryopreservation are likely to be valuable in
providing explanations for the observed variations in cryosensitivity, there is some indication that between-strain studies of inbred mice might also be useful. Nakagata and
Ž .
Takeshima 1993 collected and cryopreserved spermatozoa from caudae epididymides of eight different strains of mice and then used them for in vitro fertilization assays.
Post-thaw motilities ranged from 23 to 62, and fertility in vitro ranged from 25.5 to 88.9. The strain mostly used for transgenic work, C57BLr6N, consistently showed
the worst post-thaw motility and fertility rates. Although it is premature to attribute these differences to biochemical properties of the cell, rather than to simpler explanations such
as sperm head shape or flagellar efficiency, the intriguing possibility should be consid- ered that genetically determined differences in membrane properties are involved. The
gross differences between species are undoubtedly under genetic control, but the subtle use of a within-species experimental model, where membrane organization and function
is likely to be consistent, may permit more detailed investigations of, for example,
Ž .
membrane permeability and ion transport, to be performed. Willoughby et al. 1996 Ž
. included comparisons between two mouse strains outbred ICR and inbred B6C3F1 in
their determinations of mouse sperm membrane properties, and failed to observe statistically significant inter-strain differences in the recovery of sperm motility after
exposure to a range of anisosmotic conditions. Despite this formal result, their data seems to show that spermatozoa from the inbred strain had lower tolerance to extremes
Ž .
of osmolarity Willoughby et al., 1996 . A significant difference between strains was also observed in the maintenance and recovery of mitochondrial integrity after exposure
to the same range of anisosmotic conditions. It would be interesting to confirm these results, identify the source of such variability if it really exists, and then correlate such
differences across, rather than within, species.
Besides glycerol and the other penetrating cryoprotectants, sugars such as raffinose Ž
. and lactose, polymers such as polyvinyl pyrollidone PVP and the amphipathic com-
pounds glycine betaine, glutamine and proline have been identified as potentially cryoprotective. Raffinose has been used, with and without glycerol, for the preservation
Ž .
of mouse spermatozoa see above ; 11 lactose in combination with glycerol has been
found useful in combination with pellet freezing methods, where it has been used for Ž
. carnivore e.g. ferret, Howard et al., 1991; Giant panda, Moore et al., 1984 as well as
Ž .
for ram and boar spermatozoa Salamon and Lightfoot, 1969; Wilmut and Polge, 1977 . Sugars are thought to act by increasing the percentage of unfrozen water at any given
temperature or reducing the concentration of salts in the unfrozen aqueous solution. Glycine betaine, proline and trehalose are thought to interact directly with membrane
Ž lipids and proteins, altering their phase transition behaviour and hydration state Rudolph
. et al., 1986 . Experimentally, however, these amphiphatic substances have only proved
effective in the presence of glycerol and egg yolk when tested with ram and stallion Ž
. spermatozoa Koskinen et al., 1989; Sanchez-Partida et al., 1992 .
In addition to these various cryoprotective compounds, egg yolk is routinely included in cryopreservation protocols for semen from domestic animals and many exotic species.
Egg yolk is regarded as protecting against cold-shock, a lipid-phase transition effect Ž
. Drobnis et al., 1993 . Given current needs for disease control and therefore the
avoidance of biologically derived substances in cryoprotective media, there is a pressing requirement to find an egg yolk substitute. However, until the action of egg yolk in
conferring membrane cryoprotection is better understood little progress in the search for Ž
. alternatives can be made. Watson 1976 showed that the active component of egg yolk
is a low-density lipoprotein, but direct evidence for its mode of action has remained elusive. Evidence from cryomicroscopic studies of ram spermatozoa showed that egg
yolk protected against membrane damage and loss of motility induced below the
Ž . Ž
. extracellular freezing point y208C
Holt et al., 1992 . In that study, the onset of membrane damage during thawing was detected by the loss of intracellular fluorescein
where, in contrast to unprotected cells where fluorescein was lost soon after thawing, egg yolk inhibited fluorescein loss until the cells were rewarmed above q208C. In an
earlier cryomicroscopic study of cold-shock effects in ram spermatozoa, it was evident that egg yolk prevented sperm flagellae from bending into a rigid ‘‘bow-like’’ configu-
Ž .
ration during cooling Holt et al., 1988 . Furthermore, egg yolk abolished the tendency for these flagellae to undergo sudden, irreversible, midpiece bending through 1808 when
the temperature declined to about 12–148C. As the sudden bending phenomenon could not be induced by detergent-mediated membrane permeabilization, it was suggested that
a localized membrane lesion permitted influx of ions which activated the axonemal mechanism in a highly specific region.
The direct modulation of sperm plasma membrane lipid phase transition behaviour by interaction with egg yolk is an attractive idea, but one which has little evidence in its
favour. Freeze-fracture electron microscopy failed to demonstrate that egg yolk influ- enced the extent of intramembranous particle aggregation induced when ram, bull or
Ž boar spermatozoa were cooled and stored at 0–58C De Leeuw et al., 1990; Holt and
. North, 1984 . Taking particle aggregation as evidence of phase transitions having
occurred, no prevention of the effect by the presence of egg yolk was evident. If modulation of phase transition behaviour does not occur, the alternative is likely to be
egg yolk binding to the cell surface, modification of membrane permeability and
Ž .
activation of cellular adenylate cyclase Okamura et al., 1991 . Water and ionic
permeability may be modified by the close association of a lipoprotein with the glycocalyx or outer lipid bilayer and this, together with the activation of ionic pumps
through increased phosphorylation could affect the cells osmotic behaviour and response to permeating cryoprotectants such as glycerol.
Several investigators have concluded that the addition of surfactant to egg yolk diluents improves the post-thaw sperm motility, acrosomal integrity, survival and
Ž .
fertility for further details, see Bwanga, 1991 . Most diluents for boar semen contain Orvus ES paste, a detergent. The benefits of its use have also been documented for
Ž several large ruminant species e.g. zebra, elephant, scimitar-horned oryx, Eld’s deer,
. wildebeest and greater kudu . The consensus explanation for this beneficial effect is that
the detergent modifies egg yolk particles, thus facilitating a more efficient interaction Ž
. with the sperm plasma membrane Pontbriand et al., 1989 . If this were true, it is
presumably reasonable to expect equally improved performance with all species. How- ever, very few studies of semen preservation in primates, carnivores and other taxa have
Ž .
examined the value of including detergent. Penfold and Moore 1993 developed a cryopreservation diluent for mouse spermatozoa, and included 0.1 sodium lauryl
sulphate. These authors specifically commented that the detergent solubilized protective lipids in the egg yolk and did not act directly upon the sperm plasma membrane. In
Ž .
contrast, Kaplan and Mead 1992 evaluated semen diluents for the Western spotted skunk and noted that BF5 containing sodium dodecyl sulphate performed badly in the
comparison. In this case, glycerol also proved less effective than DMSO as a cryoprotec- tant.
A number of studies have implicated membrane lipid peroxidation as a cause of Ž
. defective sperm function, both in natural male infertility for review, see Aitken, 1995
Ž and after semen cryopreservation for a detailed discussion of this topic, see Salamon
. and Maxwell, 1995a . Attempts to overcome peroxidation during semen cryopreserva-
tion have included processing under anaerobic conditions, addition of antioxidants and the inclusion of chelating agents. Confirming the effectiveness of these strategies has
Ž .
Ž been somewhat problematic see Salamon and Maxwell, 1995a . Glutathione Slaweta
. Ž
. and Laskowska, 1987 and dithiothreitol Rao and David, 1984 have both been reported
as providing protection against peroxidation during cryopreservation, with improvements Ž
. in post-thaw motility and acrosomal integrity. Butylated hydroxytoluene BHT , a free
radical scavenger known to interact with biological membranes altering their fluidity and phase transition behaviour, has been shown to decrease the permeability of bovine sperm
Ž .
membranes Hammerstedt et al., 1976 . Little benefit on cold-shock resistance has subsequently been observed with ram spermatozoa; this is perhaps not surprising as
BHT reportedly enhances freeze–thaw damage and membrane fragility in mammalian Ž
. cells Law et al., 1986; Shertzer et al., 1991 .
In addition to the choice of cryoprotectant and various potential additives, semen diluents must be prepared in an aqueous medium. Some commonly used formulations,
especially those with high sugar content, do not contain a pH buffer even though components such as egg yolk can affect the solution pH. Many media include sodium
Ž Ž
. .
citrate, tris tris hydroxymethyl aminomethane or zwitterionic buffers such as TES Ž
Ž .
. N-tris hydroxymethyl methyl-2-aminoethane sulphonic acid . Tris titrated with TES
Ž .
TEST media has proved a particularly successful choice for wild species owing to its wide applicability, especially as background data on sperm responses to diluents is
frequently non-existent. BF5, a widely used freezing diluent for routine boar semen
Ž .
Ž freezing Pursel and Johnson, 1975 , contains the TEST combination Graham et al.,
. 1972 . The comparative merits of buffer systems were discussed in some detail by
Ž .
Ž .
Watson 1990 and Salamon and Maxwell 1995a .
4. Practical aspects of semen cryopreservation
