**3.2.3 Non-mammal vertebrates**

Whereas embryo cryopreservation in mammals shows some success, at least in those extensively studied species, situation lagging far behind in all other vertebrates (fishes, birds, reptiles and amphibians). It is true that considerably less efforts have been invested in embryo cryopreservation in most members of these groups, but the more important cause is the different structure embryos in these vertebrates have, difference that complicates their cryopreservation. From the little that has been done in these vertebrates, the vast majority of studies were done on fish (primarily the zebrafish; *Dino rerio*) and to a lesser extent also in amphibians – the two classes with the smaller oocytes among the non-mammalian vertebrates. The ensuing discussion will therefore be primarily on fishes as representatives for these classes. When sex chromosomes are the determination method, as is the case in most vertebrates, either the male or the female can be the heterogametic sex. In mammals the male carry both X- and Y-chromosomes while the female carries two copies of Xchromosome. In birds, on the other hand, it is the female that carry the Z- and Wchromosomes while the male carries two copies of the Z-chromosome. In fishes and amphibians both systems can be found. To have both chromosomes represented, one should aim to at least preserve enough gametes of the heterogametic sex. In many of the nonmammal species this means preserving the female's gametes, which, as will be discussed here, is problematic. Several attributes differentiate oocytes in these classes from those of mammals. To start with, they are considerably larger, resulting in lower surface area to volume ratio. For example, while the diameter of human oocyte is ~120 µm or that of the mouse is ~80 µm, oocyte of the zebrafish is ~750 µm (Selman et al., 1993) or that of the marsh frog (*Rana ridibunda*) is ~1,400 µm (Kyriakopoulou-Sklavounou & Loumbourdis, 1990), oocytes of the American alligator (*Alligator mississippiensis*) are ~4,000 µm (Uribe & Guillette, 2000), those of the pink salmon (*Oncorhynchus gorbuscha*) in the range of 5,150 to 6,340 µm, and the sizes go even higher in snakes such as kingsnakes (genus: *Lampropeltis*) with diameter of about 22,000 µm (Tryon & Murphy, 1982), and birds like the Japanese quail (*Coturnix coturnix japonica*) – 17,000 to 19,000 µm (Callebaut, 1973) or the domestic chicken (*Gallus gallus domesticus*) with a diameter of about 35,000 to 40,000 µm (Schneider, 1992). The consequence of this is relatively poor water and cryoprotectant movement across the cellular membrane during chilling, freezing and thawing. The difference in size also means considerably larger volume of water to vitrify, thus greatly increasing the risk for intracellular ice formation and cell death. Fish embryos contain a large yolk compartment, enclosed in the yolk syncytial layer (YSL). The behavior of the yolk during freezing defer

Genome Banking for Vertebrates Wildlife Conservation 327

blastodermal cells donor. The alternative, which seems to have higher potential from conservation point of view, is the preservation of primordial germ cells. These can later be allo- or xenotransplanted to produce viable offspring of the donor. As a demonstration of concept, primordial germ cells from pheasant (*phasianus colchicus*) were injected into the bloodstream of domestic chicken (*Gallus gallus domesticus*) embryos to produce pheasantchicken chimeras (Kang et al., 2008). Back-crossing chimera males with pheasant females produced 10 pheasant chicks with an efficiency of 17.5%. Chimera offspring were also generated in zebrafish by transplanting GPC from various sources including vitrified embryoid, an aggregate of cells derived form embryonic stem cells (Kawakami et al., 2010). The male chimeras were then mated with normal females through natural spawning to

In conclusion, cryopreservation of embryos in the few mammalian species in which it was attempted shows some, though very limited, success. The situation is much less advanced in all other vertebrates (fish, birds, reptiles and amphibians) where noticeably less efforts have been invested and the challenges are often considerably more complex. In comparison to mammals, embryos in all these classes are usually larger in volume, with large amount of yolk and multiple membranes showing varying permeability to water and cryoprotectants. All these make embryos in these classes highly susceptible to chilling injury and, with the currently available knowledge and techniques, make their cryopreservation extremely complicated and often practically impossible. The alternative approach, at least for now, would therefore be to preserve blastodermal cells and primordial germ cells, which can be

Cryopreservation of ovarian tissue has several advantages over oocyte or embryo cryopreservation, but it also comes with its unique complications. As was discussed earlier, in the section on testicular tissue cryopreservation, tissue is a complex structure and thus presenting many difficulties with respect to cryopreservation. Ovarian tissue is available at any time, season, stage in cycle, and age – from fetus to old to deceased. It contains large number of oocytes and, to overcome the problems associated with *in vitro* development and maturation, it can be implanted so that this can take place *in vivo* (Candy et al., 1995) or after partial development *in vivo*, oocytes can be retrieved and matured *in vitro* (Liu et al., 2001). Ovarian tissue also contains premeiotic germ cells, even in aged animals whose ovaries are otherwise devoid of follicles (Niikura et al., 2009). By transplanting such ovaries into recipient young adult animals can help generate new follicles. Attempts to cryopreserve ovarian tissue were reported already in 1951 (Smith & Parkes), only two years after the same group discovered the protective effect of glycerol during freezing (Polge et al., 1949). The first live birth following ovarian tissue freezing and transplantation was reported in mice, in which the tissue was frozen to -79ºC (Parrott, 1960). Grafts can be transplanted to the owner of the tissue (autotransplantation), to another member of the species (allotransplantation) or to a member of a different species (xenotransplantation). All three possibilities were successfully used to support follicular development in grafted tissue. When it comes to wildlife conservation, ovarian tissue will not be used in a similar manner to the way it is used in human medicine, namely retransplanted into its donor. Rather, these cryopreserved tissues will be used to collect oocytes by isolation and maturation *in vitro* or by transplanting them to immune deficient host animals (usually mice or rats) that will support oocyte

produce offspring.

transplanted into host embryos to produce offspring.

**3.3 Ovarian tissue cryopreservation** 

from the behavior of other embryonic compartments, making freezing very complex. These embryos have at least three membrane structures (YSL, plasma membrane of the developing embryo and the chorionic membrane which surrounds the periviteline space) (Kalicharan et al., 1998; Rawson et al., 2000). Each of these membranes has a different permeability coefficient for water and cryoprotectants, resulting, for example, in water permeability in the range of one order of magnitude lower in fish embryos compared to other animals - 0.022 to 0.1 µm × min-1 × atm-1 in zebrafish (Hagedorn et al., 1997a) compared to 0.722 in drosophila (Lin et al., 1989) or 0.43 in mice (Leibo, 1980). To complicate things even further, the different embryonic compartments have different water content and different osmotically inactive water content (Hagedorn et al., 1997b). Since the chorionic membrane can be removed enzymatically (by pronase) and its removal does not hinder embryonic development (Hagedorn et al., 1997c), Hagedorn et al. (1997a) suggested that the YSL was the primary barrier to crtyoprotectants resulting in the yolk sac reaching lower levels of cryoprotection compared to other embryonic compartments. Using magnetic resonance microscopy, they have shown that while no cryoprotectant injected into the yolk was able to leave, some cryoprotectant was able to enter the blastoderm (Hagedorn et al., 1996). Attempts to solve this permeability issue by adding aquaporin 3 water channels to the zebrafish embryonic membranes (Hagedorn et al., 2002) or inserting cryoprotectants into the yolk by microinjection (Janik et al., 2000) were unsuccessful. Efforts to test various permeating and non-permeating cryoprotectants including methanol, Me2SO, glycerol, 1,2 propanediol, PG, EG, trehalose, and sucrose also took place. Embryos were shown to be very sensitive to glycerol and EG at a concentration of 1.5M, but less so to methanol, Me2SO or PG (Hagedorn et al., 1997c). Studies also showed that later-stage embryos were less chilling sensitive than early-stage ones and thus probably more suitable for cryopreservation (Zhang & Rawson, 1995). However, attempts to cryopreserve fish embryos by slow freezing or vitrification generally met with lack of success. (reviewed in Robles et al., 2009). For instance, when intact embryos were cryopreserved by slow freezing, only about 2% of the cells in them survived the process (Harvey, 1983). Attempts were also carried out to cryopreserve amphibian (the frog *Xenopus*) oocytes with similar lack of success (Guenther et al., 2006; Kleinhans et al., 2006).

So, if oocytes and embryos are not an option at the moment, the alternatives are blastodermal cells and primordial germ cells. These cells can be cryopreserved by slow freezing (Naito et al., 1992; Naito et al., 1994) or vitrification (Kohara et al., 2008; Higaki et al., 2010) with good over all post-thaw/warming viability. Goose blastodermal cells, cryopreserved by slow freezing resulted in relatively low survival rate of 25% or less, depending on the cryovial used (Patakine Varkonyi et al., 2007). In another study, quail blastodermal cells were isolated, cryopreserved and the thawed viable cells were used to create quail-chicken chimeras (Naito et al., 1992). Chicken primordial germ cells had survival rate of 85.8 ± 1.2% and 91.2 ± 2.8% for vitrified-warmed and frozen-thawed cells, respectively with no significant difference between treatments and the control (Kohara et al., 2008). Blastodermal cells can be used to create chimeras, which are organisms made out of cells from two or more donors with different genetic background. Using this system, duck blastodermal cells were injected into the subgerminal cavity of same stage gammairradiated chicken embryo to produce duck-chicken chimeras (Li et al., 2002). These chimeras were mated with ducks to produce six duck hatchlings (out of 622 eggs collected) indicating that, albeit at low efficiency, this system can produce offspring of the

from the behavior of other embryonic compartments, making freezing very complex. These embryos have at least three membrane structures (YSL, plasma membrane of the developing embryo and the chorionic membrane which surrounds the periviteline space) (Kalicharan et al., 1998; Rawson et al., 2000). Each of these membranes has a different permeability coefficient for water and cryoprotectants, resulting, for example, in water permeability in the range of one order of magnitude lower in fish embryos compared to other animals - 0.022 to 0.1 µm × min-1 × atm-1 in zebrafish (Hagedorn et al., 1997a) compared to 0.722 in drosophila (Lin et al., 1989) or 0.43 in mice (Leibo, 1980). To complicate things even further, the different embryonic compartments have different water content and different osmotically inactive water content (Hagedorn et al., 1997b). Since the chorionic membrane can be removed enzymatically (by pronase) and its removal does not hinder embryonic development (Hagedorn et al., 1997c), Hagedorn et al. (1997a) suggested that the YSL was the primary barrier to crtyoprotectants resulting in the yolk sac reaching lower levels of cryoprotection compared to other embryonic compartments. Using magnetic resonance microscopy, they have shown that while no cryoprotectant injected into the yolk was able to leave, some cryoprotectant was able to enter the blastoderm (Hagedorn et al., 1996). Attempts to solve this permeability issue by adding aquaporin 3 water channels to the zebrafish embryonic membranes (Hagedorn et al., 2002) or inserting cryoprotectants into the yolk by microinjection (Janik et al., 2000) were unsuccessful. Efforts to test various permeating and non-permeating cryoprotectants including methanol, Me2SO, glycerol, 1,2 propanediol, PG, EG, trehalose, and sucrose also took place. Embryos were shown to be very sensitive to glycerol and EG at a concentration of 1.5M, but less so to methanol, Me2SO or PG (Hagedorn et al., 1997c). Studies also showed that later-stage embryos were less chilling sensitive than early-stage ones and thus probably more suitable for cryopreservation (Zhang & Rawson, 1995). However, attempts to cryopreserve fish embryos by slow freezing or vitrification generally met with lack of success. (reviewed in Robles et al., 2009). For instance, when intact embryos were cryopreserved by slow freezing, only about 2% of the cells in them survived the process (Harvey, 1983). Attempts were also carried out to cryopreserve amphibian (the frog *Xenopus*) oocytes with similar lack of

So, if oocytes and embryos are not an option at the moment, the alternatives are blastodermal cells and primordial germ cells. These cells can be cryopreserved by slow freezing (Naito et al., 1992; Naito et al., 1994) or vitrification (Kohara et al., 2008; Higaki et al., 2010) with good over all post-thaw/warming viability. Goose blastodermal cells, cryopreserved by slow freezing resulted in relatively low survival rate of 25% or less, depending on the cryovial used (Patakine Varkonyi et al., 2007). In another study, quail blastodermal cells were isolated, cryopreserved and the thawed viable cells were used to create quail-chicken chimeras (Naito et al., 1992). Chicken primordial germ cells had survival rate of 85.8 ± 1.2% and 91.2 ± 2.8% for vitrified-warmed and frozen-thawed cells, respectively with no significant difference between treatments and the control (Kohara et al., 2008). Blastodermal cells can be used to create chimeras, which are organisms made out of cells from two or more donors with different genetic background. Using this system, duck blastodermal cells were injected into the subgerminal cavity of same stage gammairradiated chicken embryo to produce duck-chicken chimeras (Li et al., 2002). These chimeras were mated with ducks to produce six duck hatchlings (out of 622 eggs collected) indicating that, albeit at low efficiency, this system can produce offspring of the

success (Guenther et al., 2006; Kleinhans et al., 2006).

blastodermal cells donor. The alternative, which seems to have higher potential from conservation point of view, is the preservation of primordial germ cells. These can later be allo- or xenotransplanted to produce viable offspring of the donor. As a demonstration of concept, primordial germ cells from pheasant (*phasianus colchicus*) were injected into the bloodstream of domestic chicken (*Gallus gallus domesticus*) embryos to produce pheasantchicken chimeras (Kang et al., 2008). Back-crossing chimera males with pheasant females produced 10 pheasant chicks with an efficiency of 17.5%. Chimera offspring were also generated in zebrafish by transplanting GPC from various sources including vitrified embryoid, an aggregate of cells derived form embryonic stem cells (Kawakami et al., 2010). The male chimeras were then mated with normal females through natural spawning to produce offspring.

In conclusion, cryopreservation of embryos in the few mammalian species in which it was attempted shows some, though very limited, success. The situation is much less advanced in all other vertebrates (fish, birds, reptiles and amphibians) where noticeably less efforts have been invested and the challenges are often considerably more complex. In comparison to mammals, embryos in all these classes are usually larger in volume, with large amount of yolk and multiple membranes showing varying permeability to water and cryoprotectants. All these make embryos in these classes highly susceptible to chilling injury and, with the currently available knowledge and techniques, make their cryopreservation extremely complicated and often practically impossible. The alternative approach, at least for now, would therefore be to preserve blastodermal cells and primordial germ cells, which can be transplanted into host embryos to produce offspring.

#### **3.3 Ovarian tissue cryopreservation**

Cryopreservation of ovarian tissue has several advantages over oocyte or embryo cryopreservation, but it also comes with its unique complications. As was discussed earlier, in the section on testicular tissue cryopreservation, tissue is a complex structure and thus presenting many difficulties with respect to cryopreservation. Ovarian tissue is available at any time, season, stage in cycle, and age – from fetus to old to deceased. It contains large number of oocytes and, to overcome the problems associated with *in vitro* development and maturation, it can be implanted so that this can take place *in vivo* (Candy et al., 1995) or after partial development *in vivo*, oocytes can be retrieved and matured *in vitro* (Liu et al., 2001). Ovarian tissue also contains premeiotic germ cells, even in aged animals whose ovaries are otherwise devoid of follicles (Niikura et al., 2009). By transplanting such ovaries into recipient young adult animals can help generate new follicles. Attempts to cryopreserve ovarian tissue were reported already in 1951 (Smith & Parkes), only two years after the same group discovered the protective effect of glycerol during freezing (Polge et al., 1949). The first live birth following ovarian tissue freezing and transplantation was reported in mice, in which the tissue was frozen to -79ºC (Parrott, 1960). Grafts can be transplanted to the owner of the tissue (autotransplantation), to another member of the species (allotransplantation) or to a member of a different species (xenotransplantation). All three possibilities were successfully used to support follicular development in grafted tissue. When it comes to wildlife conservation, ovarian tissue will not be used in a similar manner to the way it is used in human medicine, namely retransplanted into its donor. Rather, these cryopreserved tissues will be used to collect oocytes by isolation and maturation *in vitro* or by transplanting them to immune deficient host animals (usually mice or rats) that will support oocyte

Genome Banking for Vertebrates Wildlife Conservation 329

(*Sminthopsis crassicaudata*) (Shaw et al., 1996). The last two are of special interest as they demonstrate that even when xenografting between species so philogentically distant as marsupials and mice, the graft is still supported and oocytes can develop. Primordial oocytes in ovarian tissue are probably less prone to cooling and cryopreservation damages when compared to mature ones because they are smaller in size and they lack zona pellucida. Still, recovery rate is low. In cats, for example, only 10% of the follicles survived freezing, thawing and transplantation-associated ischemia (Bosch et al., 2004). To overcome

The alternative cryopreservation approach that has been applied to gametes and embryos, namely vitrification, has been applied to ovarian tissue as well. Naturally, to achieve good cryoprotectant penetration and proper heat transfer the sample should be thin enough, normally in the range of 1 mm or less. Several groups have experimented with this approach, cryopreserving tissue samples from humans (Isachenko et al., 2009), mice (Salehnia et al., 2002), sheep (Baudot et al., 2007), pig (Gandolfi et al., 2006), cow (Kagawa et al., 2009), goat (Santos et al., 2007), dog (Ishijima et al., 2006) and cynomolgus and rhesus macaques (Yeoman et al., 2005). The general trend in recent years is for similar outcome

Cryopreservation of large volumes, including whole organs, involves several aspects, which make any attempt at cryopreservation a challenge (Arav & Natan, 2009). These difficulties include: 1) the need for efficient heat transfer throughout the tissue. When a thick tissue or whole organs are involved, this is very difficult to accomplish, 2) the need for efficient cryoprotectant penetration to all cells in the tissue. This is challenging because of the tissue thickness and because different cell types in it have different permeability coefficients and different sensitivities. Excessive exposure time may be damaging to some cells in the tissue due to cryoprotectant toxicity while shorter time might not provide sufficient protection to others. Thus, the optimal time slot is to be identified, 3) supercooling (cooling below the solution's freezing point without crystallization) may take place in some parts of the tissue. This may lead to damages from uncontrolled intra- and extracellular ice formation once crystallization occurs, 4) attaining homogenous cooling rate while avoiding the excessive build-up of toxic concentrations of cryoprotectants, 5) during cryopreservation, latent heat is released from the solution. This released heat can induce recrystallization and extend the isothermal stage, resulting in the development of a large temperature difference between the tissue/organ and the surrounding. This may lead to faster-than-optimal cooling once all latent heat has been released, 6) recrystallization may also occur during thawing because of inhomogeneous warming of the sample. Still, if these issues can be overcome, whole ovary presents one very important advantage over ovarian tissue when it comes to cryopreservation. One of the major problems with cryopreserving ovarian cortical tissue is the ischemia the graft goes through when transplanted. This ischemia cause both graft loss and death of large portion of the follicles within surviving grafts. Cryopreserving whole ovary, including its vascular pedicle, can ensure blood supply as soon as the organ has been transplanted (Bromer & Patrizio, 2009). For the grafted ovary to become fully functional, both ovaries of the recipient should be removed (Liu et al., 2008). Grafting the ovary can be done to its natural position or to any other location in the body that may provide easy

from slow freezing and vitrification (see recent review by Amorim et al., 2011)

this low harvesting rate, multiple grafts are required.

**3.4 Whole ovary cryopreservation** 

development *in vivo*. Although ovarian tissue grafting is usually done under the kidney's capsule where ample of blood vessels are found, other locations like subcutaneous grafting for easy access have also been reported (Cleary et al., 2003). Transplantation can be to either female or male recipient (Weissman et al., 1999; Snow et al., 2002) and, interestingly, in a study on human ovarian cortex transplantation to non-obese diabetic-severe combined immune deficiency (NOD-SCID) mice, more males (76.5%, 13/17) supported follicular development than females (30%, 6/20) (Weissman et al., 1999). In another study, while more xenografts were retrieved from females, the number of oocytes recovered from each xenograft was higher in those transplanted to males (Snow et al., 2002). Oocytes developed in males, however, showed reduced fertilizing ability and none of the transferred embryos resulted in implantation. The tissue, cut of its blood supply from harvesting till about 48h after transplantation, needs to rely on its surrounding for supply of oxygen and nutrients and removal of CO2 and other wastes. If not completely lost or rejected, ischemia can thus lead to the death of more than half of the follicles in the graft (Candy et al., 1997). The surviving follicles, though may grow and develop after transplantation, often contain oocytes of suboptimal quality (Kim et al., 2005). Transplanted ovarian tissue, like any transplanted tissue, carries the risk of transmitting diseases from donor to recipients, a risk that is greatly elevated by the need to use immune-deficient recipients to reduce the risk of graft rejection. The alternative to grafting is growing the follicles to maturation *in vitro*. This, however, has been demonstrated thus far only in mice where primordial follicles (Eppig & O'Brien, 1996) or primary follicles (Lenie et al., 2004) were cultured successfully *in vitro*.

The standard cryopreservation protocol, which seems to work for many different species, is cryopreservation of ovarian cortical tissue slices with a size of 1 to 2 mm3 in cryoprotective solution containing Me2SO, ethylene glycol or 1,2-propanediol. The tissue and the cryoprotective solution are equilibrated at 0ºC and then again at -5 to -7ºC. Seeding to initiate extracellular freezing is performed and the sample is then cooled at a slow and constant rate of 0.3ºC to 0.5ºC/min till somewhere between -30ºC and -80ºC, before being plunged into liquid nitrogen for storage (for review see Paris et al., 2004). An alternative technique proposed a few years ago does not require expensive equipment and is suitable for work under field conditions (Cleary et al., 2003). Following this technique, equilibration is performed on ice, and the sample is then placed in a passive freezing device that is placed on dry ice. Using this device, a cooling rate of about 1ºC/min can be achieved. This is faster than optimal cooling rate but still tolerable. When freezing wombat (*Vombatus ursinus*) ovarian cortical tissue slices this way, 134 ± 32 intact follicles per graft were found compared to 214 ± 55 for the controlled-rate freezing machine.

Cryopreserved ovarian tissue, which was later auto-, allo- or xenografted, has been done in a variety of species including humans (Weissman et al., 1999; Gook et al., 2001; Gook et al., 2003; Donnez et al., 2004), non-human primates - rhesus macaque (*Macaca mulatta*) (Lee et al., 2004), cynomolgus macaque (*Macaca fascicularis*) (Schnorr et al., 2002) and common marmoset (*Callytrix jacchus jacchus*) (von Schönfeldt et al., 2011), bovine (Herrera et al., 2002), sheep (Gosden et al., 1994), cats (Gosden et al., 1994; Jewgenow et al., 1997; Bosch et al., 2004; Jewgenow & Paris, 2006; Luvoni, 2006), mice (Parrott, 1960; Liu et al., 2000; Liu et al., 2001), rabbits (Almodin et al., 2004), common wombat (*Vombatus ursinus*) (Wolvekamp et al., 2001; Cleary et al., 2003), African elephant (*Loxodonta Africana*) (Gunasena et al., 1998), Amur leopard (*Panthera pardus orientalis*) and African lion (*Panthera leo*) (Jewgenow et al., 2011), tammar wallaby (*Macropus eugenii*) (Mattiske et al., 2002), and Fat-tailed dunnart

development *in vivo*. Although ovarian tissue grafting is usually done under the kidney's capsule where ample of blood vessels are found, other locations like subcutaneous grafting for easy access have also been reported (Cleary et al., 2003). Transplantation can be to either female or male recipient (Weissman et al., 1999; Snow et al., 2002) and, interestingly, in a study on human ovarian cortex transplantation to non-obese diabetic-severe combined immune deficiency (NOD-SCID) mice, more males (76.5%, 13/17) supported follicular development than females (30%, 6/20) (Weissman et al., 1999). In another study, while more xenografts were retrieved from females, the number of oocytes recovered from each xenograft was higher in those transplanted to males (Snow et al., 2002). Oocytes developed in males, however, showed reduced fertilizing ability and none of the transferred embryos resulted in implantation. The tissue, cut of its blood supply from harvesting till about 48h after transplantation, needs to rely on its surrounding for supply of oxygen and nutrients and removal of CO2 and other wastes. If not completely lost or rejected, ischemia can thus lead to the death of more than half of the follicles in the graft (Candy et al., 1997). The surviving follicles, though may grow and develop after transplantation, often contain oocytes of suboptimal quality (Kim et al., 2005). Transplanted ovarian tissue, like any transplanted tissue, carries the risk of transmitting diseases from donor to recipients, a risk that is greatly elevated by the need to use immune-deficient recipients to reduce the risk of graft rejection. The alternative to grafting is growing the follicles to maturation *in vitro*. This, however, has been demonstrated thus far only in mice where primordial follicles (Eppig & O'Brien, 1996) or primary follicles (Lenie et al., 2004) were cultured successfully *in vitro*.

The standard cryopreservation protocol, which seems to work for many different species, is cryopreservation of ovarian cortical tissue slices with a size of 1 to 2 mm3 in cryoprotective solution containing Me2SO, ethylene glycol or 1,2-propanediol. The tissue and the cryoprotective solution are equilibrated at 0ºC and then again at -5 to -7ºC. Seeding to initiate extracellular freezing is performed and the sample is then cooled at a slow and constant rate of 0.3ºC to 0.5ºC/min till somewhere between -30ºC and -80ºC, before being plunged into liquid nitrogen for storage (for review see Paris et al., 2004). An alternative technique proposed a few years ago does not require expensive equipment and is suitable for work under field conditions (Cleary et al., 2003). Following this technique, equilibration is performed on ice, and the sample is then placed in a passive freezing device that is placed on dry ice. Using this device, a cooling rate of about 1ºC/min can be achieved. This is faster than optimal cooling rate but still tolerable. When freezing wombat (*Vombatus ursinus*) ovarian cortical tissue slices this way, 134 ± 32 intact follicles per graft were found compared

Cryopreserved ovarian tissue, which was later auto-, allo- or xenografted, has been done in a variety of species including humans (Weissman et al., 1999; Gook et al., 2001; Gook et al., 2003; Donnez et al., 2004), non-human primates - rhesus macaque (*Macaca mulatta*) (Lee et al., 2004), cynomolgus macaque (*Macaca fascicularis*) (Schnorr et al., 2002) and common marmoset (*Callytrix jacchus jacchus*) (von Schönfeldt et al., 2011), bovine (Herrera et al., 2002), sheep (Gosden et al., 1994), cats (Gosden et al., 1994; Jewgenow et al., 1997; Bosch et al., 2004; Jewgenow & Paris, 2006; Luvoni, 2006), mice (Parrott, 1960; Liu et al., 2000; Liu et al., 2001), rabbits (Almodin et al., 2004), common wombat (*Vombatus ursinus*) (Wolvekamp et al., 2001; Cleary et al., 2003), African elephant (*Loxodonta Africana*) (Gunasena et al., 1998), Amur leopard (*Panthera pardus orientalis*) and African lion (*Panthera leo*) (Jewgenow et al., 2011), tammar wallaby (*Macropus eugenii*) (Mattiske et al., 2002), and Fat-tailed dunnart

to 214 ± 55 for the controlled-rate freezing machine.

(*Sminthopsis crassicaudata*) (Shaw et al., 1996). The last two are of special interest as they demonstrate that even when xenografting between species so philogentically distant as marsupials and mice, the graft is still supported and oocytes can develop. Primordial oocytes in ovarian tissue are probably less prone to cooling and cryopreservation damages when compared to mature ones because they are smaller in size and they lack zona pellucida. Still, recovery rate is low. In cats, for example, only 10% of the follicles survived freezing, thawing and transplantation-associated ischemia (Bosch et al., 2004). To overcome this low harvesting rate, multiple grafts are required.

The alternative cryopreservation approach that has been applied to gametes and embryos, namely vitrification, has been applied to ovarian tissue as well. Naturally, to achieve good cryoprotectant penetration and proper heat transfer the sample should be thin enough, normally in the range of 1 mm or less. Several groups have experimented with this approach, cryopreserving tissue samples from humans (Isachenko et al., 2009), mice (Salehnia et al., 2002), sheep (Baudot et al., 2007), pig (Gandolfi et al., 2006), cow (Kagawa et al., 2009), goat (Santos et al., 2007), dog (Ishijima et al., 2006) and cynomolgus and rhesus macaques (Yeoman et al., 2005). The general trend in recent years is for similar outcome from slow freezing and vitrification (see recent review by Amorim et al., 2011)

#### **3.4 Whole ovary cryopreservation**

Cryopreservation of large volumes, including whole organs, involves several aspects, which make any attempt at cryopreservation a challenge (Arav & Natan, 2009). These difficulties include: 1) the need for efficient heat transfer throughout the tissue. When a thick tissue or whole organs are involved, this is very difficult to accomplish, 2) the need for efficient cryoprotectant penetration to all cells in the tissue. This is challenging because of the tissue thickness and because different cell types in it have different permeability coefficients and different sensitivities. Excessive exposure time may be damaging to some cells in the tissue due to cryoprotectant toxicity while shorter time might not provide sufficient protection to others. Thus, the optimal time slot is to be identified, 3) supercooling (cooling below the solution's freezing point without crystallization) may take place in some parts of the tissue. This may lead to damages from uncontrolled intra- and extracellular ice formation once crystallization occurs, 4) attaining homogenous cooling rate while avoiding the excessive build-up of toxic concentrations of cryoprotectants, 5) during cryopreservation, latent heat is released from the solution. This released heat can induce recrystallization and extend the isothermal stage, resulting in the development of a large temperature difference between the tissue/organ and the surrounding. This may lead to faster-than-optimal cooling once all latent heat has been released, 6) recrystallization may also occur during thawing because of inhomogeneous warming of the sample. Still, if these issues can be overcome, whole ovary presents one very important advantage over ovarian tissue when it comes to cryopreservation. One of the major problems with cryopreserving ovarian cortical tissue is the ischemia the graft goes through when transplanted. This ischemia cause both graft loss and death of large portion of the follicles within surviving grafts. Cryopreserving whole ovary, including its vascular pedicle, can ensure blood supply as soon as the organ has been transplanted (Bromer & Patrizio, 2009). For the grafted ovary to become fully functional, both ovaries of the recipient should be removed (Liu et al., 2008). Grafting the ovary can be done to its natural position or to any other location in the body that may provide easy

Genome Banking for Vertebrates Wildlife Conservation 331

(gametes, embryos, somatic cells, or anything else we can put your hands on) of the genetic diversity, and just sit and wait for some new technology to come by or for breakthrough in one of the still experimental technologies at hand, genetic diversity within species and possibly entire species will be lost for ever. We should therefore aim to create banks that will hold samples from each endangered species on earth and of as wide a diversity of genetic make up as possible in each. Cryopreservation is a more mature technology for this purpose but many other options are advancing and may one day play an important role in long-term banking for wildlife conservation. New and much better technologies may emerge with time but we cannot sit and watch species going extinct and take no action. Collections should be

To produce embryos *in vivo* or *in vitro*, conspecific spermatozoa and good quality oocytes are required, both or either of which often prove very difficult to obtain. An alternative that can circumvent this, at least in part, is preservation of somatic cells, to be later used for somatic cell nuclear transfer (SCNT, Wilmut et al., 1997). In SCNT, also known as cloning, nucleus of a somatic cell is microinjected into enucleated oocyte, which is then grown *in vitro* and can be later transferred to recipient females for development to term, with or without a cryopreservation step in between. Somatic cells from a wide variety of sources can be used for this purpose. Such diverse sources include cells from tissues preserved without cryoprotectant at -80ºC for more than a decade, or cells from tissues kept at -20ºC for as long as 16 years (Hoshino et al., 2009), cells isolated from mummified animals (Kato et al., 2009), freeze-dried somatic cells (Loi et al., 2008a; Ono et al., 2008; see next section), semen-derived somatic cells (Nel-Themaat et al., 2008a; Nel-Themaat et al., 2008b; Liu et al., 2010), cells collected postmortem (Oh et al., 2008), cell line (Campbell et al., 1996), and of course both fetal and adult cells are suitable for this purpose (Wilmut et al., 1997). SCNT has indeed an obvious potential for the multiplication of rare genotypes (Corley-Smith & Brandhorst, 1999; Loi et al., 2008a; Loi et al., 2008b), but its wide application is prevented by the currently low efficiency in terms of offspring outcome. To date, successful cloning was reported in sheep (Campbell et al., 1996; Wilmut et al., 1997; Loi et al., 2008a; Loi et al., 2008b), cow (Cibelli et al., 1998), mice (Wakayama & Yanagimachi, 1998), goat (Baguisi et al., 1999), pigs (Polejaeva et al., 2000), cats (Shin et al., 2002), dogs (Jang et al., 2007), rabbits (Chesne et al., 2002), ferrets (Li et al., 2006), mule (Woods et al., 2003), horse (Galli et al., 2003), gaur (*Bos gaurus*) (Lanza et al., 2000), buffalo (*Bubalus bubalis*) (Lu et al., 2005; Shi et al., 2007), mouflon (*Ovis orientalis musimon*) (Loi et al., 2001), African wild cat (*Felis silvestris libica*) (Gómez et al., 2003), wolves (*Canis lupus*) (Kim et al., 2007), mountain bongo antelope (*Tragelaphus euryceros isaaci*) (Lee et al., 2003) and eland (*Taurotragus oryx*) (Nel-Themaat et al., 2008b). When dealing with already extinct species, we can anticipate survival of nucleus DNA but not for viable oocytes. The only hope is then to use oocytes from closely related species. Interspecies SCNT (ISCNT), performed by injecting the nucleus from one species into the oocyte of another has also been carried out in a variety of species (for a recent review see Loi et al., 2011a). These include ISCNT from the endangered mouflon to a domestic sheep (*Ovis aries*) (Loi et al., 2001), from red panda (*Ailurus fulgens*) to rabbit (Tao et al., 2009), from sand cat (*Felis margarita*) to domestic cat (Gómez et al., 2008), from Canada lynx (*Lynx canadensis*) to both domestic cat and caracal (*Caracal caracal*) (Gómez et al., 2009), from water buffalo (*Bubalus bubalis*) to cow (*Bos taurus*) (Srirattana et al., 2011), and most strikingly – from a

created with any and all possible technologies in mind.

**4.1 Somatic cells cryopreservation for SCNT** 

access. Of course ovary transplanted to another location can produce oocytes that should be harvested for use *in vitro*. First whole ovary cryopreservation reported was in sheep (Revel et al., 2001; Revel et al., 2004). This report used directional freezing technique, which is claimed to provide a solution to many of the issues involved in large volume cryopreservation mentioned above (Arav & Natan, 2009). Most other cryopreservation experiments used controlled-rate freezing equipment to achieve the desired very slow (~0.1ºC/min) cooling rate needed. This first report was followed by reports on cryopreserving ovaries of various other species such as rats (Wang et al., 2002; Qi et al., 2008), mice (Liu et al., 2008), bovine (Arav, 2003), pigs (Imhof et al., 2004), human (Bedaiwy et al., 2006) and another study on sheep (Onions et al., 2009) . In some of these studies, pregnancies were achieved and live young were produced. Interestingly, to date transplantation of cryopreserved whole human ovary has not been reported (Bromer & Patrizio, 2009) despite the fact that ovarian transplantation has been in practice for several years now and whole human ovary cryopreservation was attempted by several researchers.

Although vitrification is an attractive procedure for cryopreservation of whole ovaries, the current knowledge in cryobiology is insufficient to overcome the multiple problems involved in large volume vitrification (Fahy et al., 1990), primarily when tissue, rather than suspension, is involved. Keeping in mind the relationship between the three factors determining the probability of vitrification mentioned earlier (see section on semen vitrification and also Saragusty & Arav, 2011), to avoid cryoprotectant toxicity, very high cooling rates and very small sample volume are needed. Attempts at whole ovary vitrification did take place and in some cases, when the ovaries were sufficiently small, were even successful. An attempt to vitrify whole sheep ovary resulted in complete loss of all follicles (Courbiere et al., 2009). On the other hand, in studies on mice and rats, vitrification of whole ovary was successful (Migishima et al., 2003; Hoshina et al., 2009). One study showed acceptable post warming viability by *in vitro* evaluations of mice ovaries (Migishima et al., 2003). In another study follicular growth was demonstrated after autotransplantation under the kidney capsule of vitrified warmed rat ovaries (Sugimoto et al., 2000). In yet another study, live offspring were produced when the donor mice were transgenic so that their ovaries expressed anti-freeze protein type III as an additional mean of cryoprotection (Bagis et al., 2008).

With the big potential whole ovary cryopreservation holds for wildlife conservation, this procedure is yet to be reported in any animal other than laboratory or domestic species.
