**2. What happens during cryopreservation?**

Independently of the cell or tissue above mentioned being considered, the current methods for their cryopreservation fall into one of the two following categories: (i) slow equilibrium freezing or, (ii) rapid non-equilibrium vitrification, and variations within. In either case, the

epidydimal spermatozoa (retrieved by biopsy of the cauda) or by tissue sampling through testicular biopsies (Keros et al 2005, Curaba et al 2011). However, these approaches are not relevant for porcine breeding. Cauda epididymal spermatozoa are easier to slow freeze than ejaculated spermatozoa (Rath & Niemann 1997) and testicular biopsies are not advisable in

Preservation of female genetics can be done either by freezing of germplam (e.g. oocytes or embryos) or of ovarian tissue (slices or whole ovary), from which oocytes can thereafter be harvested for ART. Germplasm freezing in pigs has also followed a tortuous road, with deceiving results for decades, particularly related to the high sensitivity of pig oocytes and early embryos to chilling, similarly to other species containing large deposits of intracellular lipids (Zhou & Li 2009), in contrast to blastocysts where the lipid amounts were lower. Delipation (or side-dislocation of lipid depots by ultracentrifugation) was soon shown to increase the survival of oocytes/embryos subjected to freezing (Nagashima et al 1995), survival that could be enhanced if the cytoskeleton could be preserved from damage using exogenous chemicals (Shi et al 2006). Use of alternative methods such as vitrification instead of slow cooling led to better survival (see Massip 2001) including the birth of offspring (Berthelot et al 2000). However, large variation was seen among methods, sources and laboratories (Holm et al 1999, Cuello et al 2007, Somfai et al 2008, Ogawa et al 2010), including the method used for intrauterine deposition (Rodriguez-Martinez 2007b, Roca et

Cryopreservation of ovarian tissue (or even of whole ovaries) has been tested in several species including human (Isachenko et al 2009) pertaining the recovery of follicular oocytes for ART or ultimate autographs (Kim et al 2010). Procedures for porcine ovarian samples have followed methods tested in other species (Imhof et al 2004, Borges et al 2009) with promising results, albeit yet at an academic level, pertaining the advancement of xenografting (Moniruzzaman et al 2009). As well, experimental models using the porcine species have been developed for cryopreservation of genital tissues, particularly the uterus (Dittrich et al 2004, 2006) paving the way for human transplantation procedures (Diaz-

Thus, interest has been large to attempt routine cryopreservation of porcine gametes, embryos and genital tissues, yet with various degrees of success. Therefore, the present review aims to summarize the state-of-the-art regarding established and emerging methods for the cryopreservation of porcine gametes and embryos as germplasm, intending a critical revision of the underlying problems that still constrain their application for establishing repositories, their use in reproductive biotechnologies and, ultimately, for breeding. As well, it intends to describe our level of knowledge when attempting cryogenics of gonads and other genital tissues for comparative research, particularly on human regenerative medicine. The review is not exhaustive and focus on

Independently of the cell or tissue above mentioned being considered, the current methods for their cryopreservation fall into one of the two following categories: (i) slow equilibrium freezing or, (ii) rapid non-equilibrium vitrification, and variations within. In either case, the

boars owing to their highly vascularized testicular capsule (Ohanian et al 1979).

al 2011).

Garcia et al 2011).

methodological aspects of procedures.

**2. What happens during cryopreservation?** 

entire process basically concerns the way we handle the presence of water in and around the cells and whether its freezing is allowed (conventional cryopreservation, slow equilibrium freezing) or totally prevented (vitrification).

In the first method, which is the one traditionally used in biomedicine, particularly for sperm preservation, cells are subjected to slow cooling to temperatures below zero, with freezing rates of 0.5-100 ◦C/min). The method allows ice to form and solute to concentrate alongside the change in water phase. Both ice and high solute concentrations can cause direct (either initial or eutectic, Han & Bischof 2004a), respectively secondary damage, jeopardizing cell survival or handicapping vital cell functions post-thaw. At some moment during the process, water freezes to form ice, primarily extracellular, but even intracellular. Ice grows and becomes over time surrounded by an increasing amount of solutes which move to the areas where water did not yet changed phase. Cells balance ion concentrations at either side of the plasma membrane thus keeping proper osmotic pressure. Depending on the relative amounts of free and bound water, such a change of phase (either formation or dissolution of ice and de/rehydration phenomena) implies changes in ionic concentration caused by directional movement of water across the membrane, disturbing the homeostatic osmotic pressure of the cell/s (Pegg 2007). Cells respond by allowing water to leave the intra-cellular compartment, to compensate the increasing hyper-osmotic extra-cellular compartment caused by the progressive formation of ice. Those water movements lead both to cell dehydration and to a toxic hyper-concentration of solutes intracellular which, ultimately, affects cells viability (Watson & Fuller 2001). Freezing injury can then be related to high electrolyte concentration effects (solute effects), presence of intracellularly ice (formed direct or eutectic) and also the pressure of large extracellular crystals on the veins of concentrated (i.e. vitrified) extender and cells (Saragusty et al 2009). See **Figures 1and 2** for an illustration of these events.

Freezing injury during slow freezing can be minimized. Intracellular freezing is generally lethal but can be avoided by sufficiently slowing the rate of cooling. Solute-caused damage, which is the dominating feature under conventional slow freezing especially in cells in liquid suspension, can be minimized by the addition of CPA. Most CPA´s (as glycerol, dimethyl sulfoxide (DMSO), ethylenglycol (EG), propyleneglycol (PG)) are highly soluble, permeating compounds of low-to-medium toxicity, whose primary role is to reduce the amount of ice formed at any given sub-zero temperature, by simply increasing the total concentration of all solutes in the system, thus defining the concept of slow equilibrium freezing (Pegg 2002 2007). Introduction of sufficient CPA would eventually avoid freezing and a glassy of vitreous state could be produced instead. Such concept is the theoretical rationale for the second method listed above: rapid nonequilibrium vitrification. Vitrification is the physical process by which a highly concentrated cryoprotective solution supercools to very low temperatures (often to -120 to -130 ◦C) to finally solidify into a metastable glass, without undergoing crystallization at a practical, high speed cooling rate (i.e. dipping onto LN2). Use of ultra-high speed voids the need of penetrating CPA, open for using non-penetrating CPA (such as sucrose, fructose, glucose), but demands the use of small (5-50µL) suspension droplets. The glassy state is defined by its viscosity reaching 10-13 poises, sufficient for the aqueous material to behave as a solid, without any water crystallization. Once again, this waives the above listed sources for cell injury: ice crystals and increased/ill distributed solute

Cryopreservation of Porcine Gametes, Embryos and Genital Tissues: State of the Art 235

Fig. 2. Cryo-scanning electron microscopy (cryo-SEM) micrograph at higher magnification showing the contents of a maxi-straw frozen at the speed of 1,200ºC/min (by direct plunging into LN2 after initial cooling to +5ºC). The ice lakes (\*) are small, and surrounded by prominent veins. This fast cooling caused freezing of water both extra- and intracellularly, with clear evidence of sub-cellular distortion caused by the presence of intracellular ice crystals in the peri-nuclear and peri-axonemal areas, owing to a lack of sperm dehydration during the process. Note the fractured sperm head (large arrow) with marks of lethal intracellular ice, and the tail entrapped by extracellular ice (small arrow)

However, we should bear in mind that the physical phenomenon of vitrification (e.g. the process by which a liquid begins to behave as a solid during cooling without substantial changes in molecular arrangement or thermodynamic state variables) is as relevant to conventional freezing, where the cells survive in this glassy medium between ice crystals (see **Figure 1a-c**) as to vitrification *per se*, where the entire sample is vitrified (Wowk 2010). Therefore, seeding is quite relevant for supercooled vitrification solutions in conventional freezing, while it does not play any role during pure vitrification, provided that cooling rates are high. For instance, use of LN2-slush (e.g. lowering the temperature to near the freezing point of LN2, -205 to -210 ◦C by applying negative pressure to the LN2, Yavin et al 2009) increases the cooling rate 2 to 7-fold compared to simple plunging in LN2. Viscosity also plays a major role and must increase during cooling, until the glass transition (i.e. the change from liquid to solid) is reached. This concept opens for the freezing of highly

When thawing or re-warming occurs, the events above described basically reverse. Slow rewarming allows water to reflux to the areas where solutes are concentrated in cells treated

concentrated semen samples, provided the size of the sample is small enough.

with dislocation of the axoneme (Courtesy of Dr Hans Ekwall).

concentrations. The CPA used to vitrify cells include those used during conventional freesing but at very high concentrations (10-fold higher compared to slow freezing), near the maximum tolerated by the cells, thus becoming potentially harmful (Pegg 2005). Penetrating CPA-free vitrification was attempted already by the early 1940´s using rabbit spermatozoa plunged into LN2 (Hoagland & Pincus 1942). Use of non-penetrating "CPA" (CPA-free concept) such as sucrose has proven feasible for the spermatozoa of some species, including human (Isachenko et al 2004, 2005, 2008, Hossain & Osuamkpe 2007), primates (Dong et al 2009), or canine (Sanchez et al 2011), where sperm suspensions were vitrified (either drop-wise, Isachenko et al (2004, 2005) or contained in 50µL-plastic capillaries (Isachenko et al 2011)) by plunging in LN2, with a cooling rate of ̴ 10,000 ◦C/min. Basically, vitrification is therefore always determined by a relation between cooling rate, medium viscosity and sample size.

Fig. 1. Micrographs of frozen boar semen illustrated with (a) transmission electron microscopy or (b,c) Cryo-scanning electron microscopy. Spermatozoa were extended and conventionally frozen in maxi-straws (a, 5 mL) or FlatPack™ (b,c, 5 mL) and subjected to freeze-substitution (a) or partial sublimation (b,c) to depict extracellular ice lakes (\* in a, marked with legend in b,c) and the veins of concentrated extender (e in a, legend in b,c). Note the presence of intracellular ice marks (arrows in a) and the dislocation of axoneme structures in the tails. Such marks are not seen in the FlatPack™ material (Photo: Dr Hans Ekwall, Uppsala, Sweden).

concentrations. The CPA used to vitrify cells include those used during conventional freesing but at very high concentrations (10-fold higher compared to slow freezing), near the maximum tolerated by the cells, thus becoming potentially harmful (Pegg 2005). Penetrating CPA-free vitrification was attempted already by the early 1940´s using rabbit spermatozoa plunged into LN2 (Hoagland & Pincus 1942). Use of non-penetrating "CPA" (CPA-free concept) such as sucrose has proven feasible for the spermatozoa of some species, including human (Isachenko et al 2004, 2005, 2008, Hossain & Osuamkpe 2007), primates (Dong et al 2009), or canine (Sanchez et al 2011), where sperm suspensions were vitrified (either drop-wise, Isachenko et al (2004, 2005) or contained in 50µL-plastic capillaries (Isachenko et al 2011)) by plunging in LN2, with a cooling rate of ̴ 10,000 ◦C/min. Basically, vitrification is therefore always determined by a relation between

Fig. 1. Micrographs of frozen boar semen illustrated with (a) transmission electron microscopy or (b,c) Cryo-scanning electron microscopy. Spermatozoa were extended and conventionally frozen in maxi-straws (a, 5 mL) or FlatPack™ (b,c, 5 mL) and subjected to freeze-substitution (a) or partial sublimation (b,c) to depict extracellular ice lakes (\* in a, marked with legend in b,c) and the veins of concentrated extender (e in a, legend in b,c). Note the presence of intracellular ice marks (arrows in a) and the dislocation of axoneme structures in the tails. Such marks are not seen in the FlatPack™ material (Photo: Dr Hans

cooling rate, medium viscosity and sample size.

Ekwall, Uppsala, Sweden).

Fig. 2. Cryo-scanning electron microscopy (cryo-SEM) micrograph at higher magnification showing the contents of a maxi-straw frozen at the speed of 1,200ºC/min (by direct plunging into LN2 after initial cooling to +5ºC). The ice lakes (\*) are small, and surrounded by prominent veins. This fast cooling caused freezing of water both extra- and intracellularly, with clear evidence of sub-cellular distortion caused by the presence of intracellular ice crystals in the peri-nuclear and peri-axonemal areas, owing to a lack of sperm dehydration during the process. Note the fractured sperm head (large arrow) with marks of lethal intracellular ice, and the tail entrapped by extracellular ice (small arrow) with dislocation of the axoneme (Courtesy of Dr Hans Ekwall).

However, we should bear in mind that the physical phenomenon of vitrification (e.g. the process by which a liquid begins to behave as a solid during cooling without substantial changes in molecular arrangement or thermodynamic state variables) is as relevant to conventional freezing, where the cells survive in this glassy medium between ice crystals (see **Figure 1a-c**) as to vitrification *per se*, where the entire sample is vitrified (Wowk 2010). Therefore, seeding is quite relevant for supercooled vitrification solutions in conventional freezing, while it does not play any role during pure vitrification, provided that cooling rates are high. For instance, use of LN2-slush (e.g. lowering the temperature to near the freezing point of LN2, -205 to -210 ◦C by applying negative pressure to the LN2, Yavin et al 2009) increases the cooling rate 2 to 7-fold compared to simple plunging in LN2. Viscosity also plays a major role and must increase during cooling, until the glass transition (i.e. the change from liquid to solid) is reached. This concept opens for the freezing of highly concentrated semen samples, provided the size of the sample is small enough.

When thawing or re-warming occurs, the events above described basically reverse. Slow rewarming allows water to reflux to the areas where solutes are concentrated in cells treated

Cryopreservation of Porcine Gametes, Embryos and Genital Tissues: State of the Art 237

per ejaculate (5-8). For examples of current protocols see Eriksson & Rodriguez-Martinez (2000), Saravia et al (2005), Parrilla et al (2009) or Rath et al (2009) and methods cited therein. This general current protocol fits most boars but considering the large variation between ejaculates and –particularly- among boars for their capacity to sustain cryopreservation (Roca et al 2006a), the protocol has to be modified to accommodate those with sub-optimal sperm freezability (the so-called bad freezers), particularly regarding glycerol concentration and warming rates (Hernandez et al 2007a). Those changes usually allow for minimum acceptable cryosurvival (i.e. around 40%). However, it clearly shows that the methodology is still sub-optimal. Current semen cryopreservation techniques are technically demanding and expensive, both in terms of labour- and laboratory equipment costs, as well as timeconsuming (rev by Roca et al 2006b, 2011). Last but not least, there is a lack of reliable laboratory tests for the accurate assessment of semen quality in vitro, that limits our capacity to properly monitor the methods used to freeze-thaw boar semen and, particularly, its relationship to AI-fertility (Rodriguez-Martinez, 2007b). This is critical, since despite having acceptable post-thaw survival (even above 60%) this cryosurvival is not reflected in fertility after AI. Thus, boar spermatozoa are considered one of the most demanding cell types with respect to sustaining viability during freezing and thawing, with a large proportion of the spermatozoa not surviving these procedures (Penfold & Watson 2001). Moreover, those surviving spermatozoa are usually a mixture of cells, some of which survive well while others show modified motility and a shortened lifespan, factors which compromise their fertilising ability. Insemination with such spermatozoa leads, ultimately, to lowered pregnancy rates and fewer piglets born, compared with AI using liquid-stored semen (Knox 2011). In sum, although freezing methods are nowadays rather stable in many laboratories and yield above 50% of sperm survival post-thaw, fertility after AI is extremely variable (Parrilla et al 2009). The major constrain is not only the inherent difficulties to freeze spermatozoa from this species (Holt, 2000a,b), but -within the species- the sire-dependent cryosurvivability to the current procedures (Eriksson et al 2002, Holt et al 2005, Gil et al 2005, Waterhouse et al 2006, Hernandez et al 2006, 2007a, Roca et al., 2006a, Parrilla et al

This variation is usually compensated by the AI of excessive sperm numbers (at least 5x109 spermatozoa per AI-dose), i.e. double the numbers of total spermatozoa present in liquid semen doses. Fertility post-AI is nowadays substantially better, closer to AI with liquid semen (Eriksson et al 2002). See **Table 1** for an overview of fertility after conventional (cervical) AI with frozen-thawed boar semen. Fertility with lower sperm numbers is also becoming acceptable when deep intrauterine AI is practiced, although data are still restricted in numbers (Bathgate et al 2006, Roca et al 2006b, 2011). But, even with these huge sperm numbers, overall fertility (as farrowing rates) and prolificacy (as litter size) are still lower than for liquid semen (around 10-30 % lower farrowing rates, and 1-3 less piglets), indicating that other factors are limiting, such as the timing of insemination respective to spontaneous ovulation (Bolarín et al 2006, Wongtawan et al 2006). This implies that we are far from reaching the goals set up by the industry for the use of frozen-thawed semen: 85% of conception rates and a litter size of 11 piglets (Knox 2011). So, frozen-thawed boar semen is still basically limited to research, genetic banking or the export of semen for selected

2009, Roca et al 2011).

nuclei lines, constituting barely above 1% of all AIs.

by slow freezing, but the time elapsing is not short enough to avoid the toxicity that the solute concentration exerts on the cells, either leading to cell death or dysfunction. If the rewarming is too slow, ice (intracellular in particular) can damage organelles and the cytoskeleton. Rapid rewarming diminishes these risks since the toxic solutes or CPA are only momentarily present.

For either method listed, the CPA has to gain access to all areas of the cell/tissue/organ. Traditional cooling and re-warming rates affect the fluidity of the membranes of the cell and the organelles through the rearrangement of structural proteins and the dislocation of constituent lipids. If these changes affect diffusion and/or osmosis, they can jeopardize -by causing changes in the viscosity of fluids or inducing osmotic inbalance- the proper distribution of the CPA, its introduction and removal and ultimately, the freezing and the thawing process (Morris 2006, Morris et al 2007). Cooling can disrupt the integrity of the cytoskeleton and of the chromatin structure, including DNA damage (Watson & Fuller 2001, Fraser et al 2011). In cells in suspension, such as spermatozoa, both the form and volume of the sample to be cooled/re-warmed, and the concentration of the contained cells play major roles during the most damaging interval in the process, i.e. during the changes in phase of the extra-cellular water, when heat is either dissipated (during cooling) or incorporated (during re-warming) (Mazur & Cole 1989, Morris et al 1999). It is therefore obvious that samples (cells, tissues, organs) have to pass cooling and re-warming under conditions where cell injury can be minimized (Morris 2006, Morris et al 2007).
