Breeding or Assisted Reproduction? Relevance of the Horse Model Applied to the Conservation of Endangered Equids


Author’s address (for correspondence): A Van Soom, Department of Reproduction, Obstetrics and Herd Health, Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium. E-mail:


Many wild equids are at present endangered in the wild. Concurrently, increased mechanization has pushed back the numbers of some old native horse breeds to levels that are no longer compatible with survival of the breed. Strong concerns arose in the last decade to preserve animal biodiversity, including that of rare horse breeds. Genome Resource Banking refers to the cryostorage of genetic material and is an approach for ex situ conservation, which should be applied in combination with in situ conservation programmes. In this review, we propose that, owing to the great reproductive similarity among the different members of the genus Equus, the domestic horse can be used to optimize cryopreservation and embryo production protocols for future application in wild equids. We will give this hypothesis a scientific underpinning by listing successful applications of epididymal sperm freezing, embryo freezing, intracytoplasmic sperm injection, oocyte vitrification and somatic cell nuclear transfer in domestic horses. Some ART fertilization methods may be performed with semen of very low quality or with oocytes obtained after the death of the mare.


Horses have evolved from Hyracotherium or Eohippus (the dawn horse), a small leaf-eating animal living approximately 50 million years ago in the northern hemisphere, to the large members of the genus Equus, which are now represented on all continents except Antarctica. Recently, the analysis of mitochondrial DNA of fossilized bones has demonstrated that the evolutionary progression of the domestic horse is in fact more complex than initially believed (Vilà et al. 2001). During the Eocene, several equine species appeared in North America and in Eurasia and became extinct, resulting in complete disappearance of the genus Equus from the American continent approximately 8000–10 000 years ago (Vega-Pla et al. 2005). Only with their reintroduction by Columbus in 1492, were horses again present on the American continent.

The family of Equidae included, in addition to Equus, approximately 35 other genera that are all extinct today. This natural extinction happened during the course of evolution over millions of years. During recent centuries, several members of the surviving genus Equus have become endangered or are now extinct, and this occurred at a much higher rate than during the natural process of evolution. A range of factors, including predation, hunting, habitat destruction and, more recently, climate change-related drought, are the underlying causes of this increased rate of extinction of wild equids. At present, the genus Equus consists of four subgenera: Equus (wild and domestic horses), Asinus (donkeys and wild asses), Dolichohippus (Grevy’s zebra) and Hippotigris (Plain’s zebra). Most wild equids are currently endangered or threatened in the wild, as indicated on the Red List of endangered animal species of the International Union for the Conservation of Nature (IUCN) (Adams et al. 2009). A species is considered endangered when its survival in the wild is unlikely if causal factors of extinction continue to operate, or if its population is composed of fewer than 50 mature individuals raised in captivity (Comizzoli et al. 2000). Up to the present, the only measure taken to prevent extinction of wild equids is breeding them in captivity (Kaczensky et al. 2011). For example, the Przewalski’s horse, which became extinct in the wild in the 1960s, survived as a species owing to captive breeding (Souris et al. 2007). Preservation of habitat and education of people are equally essential actions to be undertaken, but in many cases, they are underused for economic or social reasons. In addition, breeding in captivity cannot always prevent extinction and should be accompanied by other measures, such as preservation of genetic material by cryopreservation.

Wild horses were first domesticated in the Eurasian steppe approximately 5500–3500 BCE (Outram et al. 2009; De Weerd and Oldenbroek 2010; Warmuth et al. 2011). This domestication gave rise to a variety of horse breeds used for travel, warfare, trade and agriculture. However, with increasing mechanization in the twentieth century, horse power became redundant, first for travel (coach horses), later for agriculture (draught horses). Some breeds found new functions in the leisure industry, but particular breeds such as the Hackney, with its extravagant action making it a popular carriage/coach horse in the nineteenth century, soon became endangered. Like the IUCN (which is focusing on the preservation of wild animals), several national societies have been founded with the purpose of preserving rare native farm animal breeds. Among these societies are the Rare Breeds Survival Trust (RSBT) in the UK (Alderson 2005), the Nordic Gene Bank for Farm Animals (Norsik Genbank Husdyr, NGH) for the Nordic and Baltic countries (Saastamoinen and Mäenpäa 2005), the Centre for Genetic Resources Netherlands (CGN) (De Weerd and Oldenbroek 2010) and the Equus Survival Trust in the USA. These organizations have carried out surveys to assess the level of endangerment and have listed breeds according to the numbers of available breeding mares (Alderson 2005) (Table 1). For domestic breeds, populations are considered at risk when fewer than 1500–3000 females remain. The actions that are at present undertaken to preserve an endangered horse breed are breed structure analysis, DNA profiling and the creation of a semen bank (Alderson 2005).

Table 1.   Endangered domestic horse breeds
BreedsCountry of originGlobal status
  1. Critical/nearly extinct: <100 active breeding mares.

  2. Critical: 100–300 active breeding mares.

  3. Threatened: 300–500 active breeding mares.

  4. Vulnerable: 500–1500 active breeding mares.

  5. At risk: 1500–3000 active breeding mares.

  6. Watch: 3000–5000 active breeding mares (Level of endangerment varies with source – Equus National Trust; De Weerd and Oldenbroek (2010); Saastamoinen and Mäenpäa (2005); Alderson (2005).

Dartmoor ponyEnglandVulnerable
Exmoor ponyEnglandVulnerable
Kerry Bog ponyIrelandCritical
Gotland-Russ ponySwedenAt risk
Dales ponyEnglandVulnerable
Fell ponyEnglandVulnerable
Faer Island ponyFaer islands/DenmarkCritical/nearly extinct
Highland ponyScotlandVulnerable
Newfoundland ponyCanadaCritical
American cream draftUSACritical
ClydesdaleScotlandAt risk
English ShireEnglandVulnerable
Estonian draught horseEstoniaCritical/nearly extinct
Lithuanian heavy draught horseLithuaniaThreatened
American Mammoth Jack donkeyUSAThreatened
Poitou donkeyFranceCritical
Caspian horseIranCritical
Appalachian purebred gaitedUSACritical
Asiatic barb horseCentral AsiaCritical/Study
Canadian horseCanadaThreatened
Cleveland bayEnglandCritical
Colonial Spanish horseUSACritical
Groningen horseThe NetherlandsThreatened
Gelderland horseThe NetherlandsThreatened
Hackney horseEnglandCritical
Irish draughtIrelandAt risk
Lippit morganUSACritical

From the above, it is clear that most wild equids and several domestic horse breeds are at risk of extinction. Therefore, the purpose of this review is to discuss the application of Genome Resource Banking in equids. A Genome Resource Bank (GRB) is a frozen repository of biological materials, including sperm and embryos, tissue, blood products and DNA. When in situ conservation of a species or breed becomes impossible, a last resort is ex situ conservation, which refers to the establishment of a viable population (in captivity for eventual reintroduction) and the start up of a GRB. First, we will discuss why the domestic horse is a good model for the establishment of a GRB in wild equids. Next, we will describe which strategies are at present being used for GRB in horses. Finally, we describe why techniques such as intracytoplasmic sperm injection (ICSI) and cloning are ready to be implemented, and what further developments are needed to increase success rates in the future.

The Domestic Horse as a Model for Wild Equids?

Domestic horses are very similar to wild equids; in contrast to Bovidae, which include very diverse species such as bisons, antelopes, cattle, goats and sheep, the Equidae are homomorphic and monogeneric (Linklater 2000). It is a common knowledge that equids are able to interbreed freely with all other species in their genus and produce viable (although in most cases infertile) offspring, with the mule and hinny as the best known examples (Allen 2005). Although interbreeding is not a strategy for species conservation, it demonstrates the similarity among horses, donkeys, onagers and zebras. Another remarkable characteristic of equids is that they are able to carry to term true xenogeneic extraspecies pregnancies created by embryo transfer (Allen 2005), although some pairings (e.g. horse embryo in donkey) are more successful than others (e.g. donkey embryo in horse). This means that domestic horses, donkeys or mules (some of which can support pregnancy after embryo transfer; review, Allen et al. 2011) can be used as a recipient for exotic equine embryos such as those of Przewalski’s horse and Grant’s zebra (Kydd et al. 1985; Summers et al. 1987). This may be a useful strategy when combining assisted reproduction with GRB.

Working with threatened wild equids becomes especially challenging because routine techniques, which can be applied with ease in domestic horses, can meet with important practical problems in wild animals. For example, collecting semen from a wild stallion involves electro-ejaculation under general anaesthesia. However, progress is being made in this area as demonstrated by the development of new techniques for ultrasound-guided monitoring of the oestrous cycle without general anaesthesia in wild mares (Adams et al., 2009). An important concern when working with wild animals, including equids, is that there is insufficient background information on their reproductive biology (Holt and Lloyd 2009), which often complicates the application of even basic techniques such as artificial insemination and embryo transfer, let alone the more advanced reproductive technologies, despite the recent progress achieved in assisted reproduction in horses (Hinrichs 2010).

In addition, some of the available assisted reproduction technologies will need to be further optimized in the domestic horse before they can be applied to endangered wild equids. Such optimization will be not only of potential long-term benefit for the preservation of wild equids, but may also find immediate applications in the preservation of endangered domestic horse breeds. The primary aim of GRB is protection and preservation of biodiversity. In threatened domestic horse breeds, this can easily be achieved and has already been started in several countries. Such a preservation programme, for example, for the Groningen horse, provides the ideal model for future applications in endangered wild equids. Being domesticated, such horses can be handled with ease, represent target populations for which funding is available from governmental or international organizations (such as the Food and Agricultural Organization, FAO) and are of interest to horse breeders who are willing to collaborate with scientists by offering their mares and stallions as donor animals for GRB.

Current Strategies for Genome Resource Banking in Horses

Cryopreservation of semen and of embryos have both been proposed for operational GRB in farm animals (Gandini et al. 2007). As yet, cryopreservation of oocytes is not practised in GRB of farm animals (Woelders et al. 2003) but could become a crucial step forward (Prentice and Anzar 2011). This would facilitate the conservation of female genetics and allow more flexibility in the application of assisted reproductive techniques in endangered horse breeds and in endangered, or even extinct, wild equids. Equine oocytes can be fertilized by ICSI: blastocyst rates between 25% and 35% are now routinely obtained after ICSI in the horse (Fig. 1), with pregnancy rates after transfer equal those obtained with in vivo-derived embryos (Hinrichs 2010) and pave the way for clinical and GRB applications. Another possibility is to transfer equine oocytes into the oviduct of an inseminated recipient mare (‘oocyte transfer’, Carnevale 2007). In addition, the preservation of somatic cells, with the aim of using them later for somatic cell nuclear transfer (cloning), could offer a simplified method for the preservation of genetic material from a large pool of animals, eliminating the need for semen collection, embryo production or oocyte retrieval, all of which are difficult in wild animals (Woelders et al. 2004). For each of these techniques, we will briefly discuss methods, associated problems and potential for achieving a pregnancy.

Figure 1.

 Horse oocytes and embryos produced in vitro; a. Immature expanded oocyte; b. Mature denuded oocyte with polar body I (arrow); c. Sperm injection of mature oocyte with polar body at 6 o’clock (arrow); d. Two-cell embryo; e. Eight-cell embryo; f. Day 9 blastocyst; g. Horse blastocysts produced in vitro in DMEM-F12; h. Horse blastocysts produced in vitro in DMEM-F12 supplemented with recombinant uterocalin 1 mg/ml (Smits et al. 2012b)

Collection and storage of semen

Semen can be routinely collected from stallions using an artificial vagina, and then stored frozen. Banking of cryopreserved semen is the technique of choice to conserve valuable genetics of farm animals at risk of extinction. The disadvantage of semen cryopreservation is that the whole genome cannot be recovered (Woelders et al. 2003). For the recovery of a lost breed, it will take at least six generations of backcrossing to restore the original genotype (Olivier et al. 1995, cited by Danchin-Burge et al. 2002). In a recent study, the costs of creating a gene bank for an endangered breed were hypothetically calculated for five species (horse, cattle, sheep, swine and rabbits) using various strategies: preservation of semen only, of embryos in combination with semen and of embryos only (Gandini et al. 2007). The costs of reconstructing an extinct breed based on frozen semen, using breeding with closely related females, were the least expensive method for cattle and pigs, but it also took more time and did not allow for a 100% reconstruction of the genome. The horse was the most expensive farm animal in which to reconstruct an extinct breed: over one million € with the use of semen only (Gandini et al. 2007). This could be attributed to the much lower yield of semen per collection in horses as compared to cattle and pigs. To obtain sufficient frozen semen to restore a breed, a minimum of 100 doses of semen from 25 males should be stored – in horses, this would mean 10 ejaculates of 25 stallions.

When semen has to be collected from endangered horse breeds or from wild equids, acceptance of the artificial vagina may be a difficult issue. Alternatives are to train the stallion (which takes time), to use electro-ejaculation, to collect semen using chemically induced ejaculations or to castrate the animal and collect epididymal semen. Electro-ejaculation in horses results in poor quality semen, often grossly contaminated with urine, making it an ineffective way of semen collection (Cary et al. 2004). Chemically induced ejaculations in stallions can be obtained using intravenous xylazine alone or in combination with oral imipramine (McDonnell 2001). Because post-thaw quality obtained following chemical ejaculation exceeds that of semen collected with the artificial vagina (McDonnell and Oristaglio Turner 1994), this technique could be valuable in gamete preservation. Collection of epididymal sperm allows for a once-only harvest of approximately 15–20 billion spermatozoa under optimal circumstances (Bruemmer 2006). Because pregnancies can be obtained using lower doses of sperm with hysteroscopic insemination (Morris and Allen 2002), or by ICSI, offspring might be produced even when only limited numbers of sperm are available.

Post-thaw quality of cryopreserved ejaculated semen is highly variable among stallions and poor in some individuals (Loomis and Graham 2008). It is interesting to note that the first foal to result from the use of frozen sperm was obtained with epididymal semen from a castrated Belgian stallion (Barker and Gandier 1957). Frozen-thawed epididymal semen was at first considered to be less suitable for artificial insemination, typically yielding low pregnancy rates (Heise et al. 2010), although exposure to seminal plasma before freezing (Heise et al. 2011) seemed to improve fertility. Other recent data suggest that the use of an adapted diluent and freezing curve for epididymal sperm can result in pregnancy rates of 66% (Papa et al. 2008) to 92% (Monteiro et al. 2011).

In endangered breeds such as the Groningen horse, there is a policy to freeze epididymal sperm of stallions that are gelded, but are still valuable for the preservation of their genetics in the gene bank, to preserve old blood lines and as a back-up for possible future applications (De Vos 2006). It is the divergent quality of cryopreserved epididymal spermatozoa that remains an obstacle for this approach to GRB.

Although artificial insemination with frozen semen is routinely used in horse breeding, its successful application is influenced by factors such as semen quality, dose, timing, method of insemination and mare management (Miller 2008). This weakens the practical applications of frozen semen, particularly in endangered horse breeds in which one cannot select stallions based on post-thaw semen quality.

Collection and storage of in vivo-produced embryos

In general, livestock embryos are collected by flushing the uterus of a superovulated female, non-surgically in large farm animals. This is an interesting alternative to semen cryopreservation, because it allows the preservation of genetics from both male and female. It is also the fastest method to restore an original genotype when needed (Woelders et al. 2003). Conversely, embryo cryopreservation is less versatile, because the embryo represents the combination of one stallion with a particular mare, a combination that cannot be altered. When evaluating the feasibility of embryo cryopreservation and transfer as a strategy for genetic conservation, one needs to consider three separate aspects: embryo production, embryo cryopreservation and the transfer of embryos to recipients.

Embryo production in domestic equids is severely limited when compared to other domestic livestock such as cattle. Mares can be used for embryo collection on consecutive cycles, but only have fertile oestrous cycles for approximately 9 months per year. More importantly, routine induction of multiple ovulation as performed in cattle still appears to be labour intensive, ineffective, expensive and unpredictable in horses, and hence usually only a single embryo can be flushed during each oestrous cycle (Stout 2006). Assuming a success rate of embryo collection of 50% per attempt, one could generate approximately six to seven embryos per year from a fertile mare. However, this requires intensive management of the donor mares and precise timing of the embryo collection in relation to the time of ovulation. The oestrous cycle of the mare has to be monitored (by hormonal assays or by ultrasound), and the time of ovulation has to be accurately determined. Seven days following ovulation, the embryo must be collected from the uterus of the donor mare by a non-surgical, transvaginal technique requiring adequate restraint. Sedated Przewalski’s horses and Grant’s zebra mares have been subjected to embryo recovery with good results (61% and 56% recovery, respectively) (Allen et al. 2011) (Table 2). Because careful preparation, follow-up and timing are required, embryos cannot be collected from animals that have died unexpectedly, whereas oocytes can be collected on short notice (see next section).

Table 2.   Recovery and transfer of Przewalski’s horse (Equus przewalskii) and Grant’s zebra (Equus quagga boehmi) embryos to domestic horse (Equus caballus) and donkey recipients (Equus asinus) (adapted from Allen et al. 2011)
Type of transfer recipientsEmbryo recovery attemptsNo. of embryos recovered (%)No. of embryos transferred/no. pregnant (%)Outcome of pregnancies
Przewalski’s horse-to-horse1811 (61)11/7 (64)2 resorbed (days 20–40)
1 resorbed (days 85–101)
1 stillborn at term
3 live foals
Zebra to horse2514 (56)5/3 (60)1 resorbed (days 59–64)
1 stillborn (day 350)
1 live foal
Zebra to donkey  8/2 (25)1 resorbed (days 53–64)
1 aborted (day 292)

The availability of appropriate surrogate mothers is not a limitation to the use of embryo transfer in endangered horse breeds or even for many wild equids; the domestic horse has been proven to be a suitable recipient for either. Pregnancy rates for Przewalski’s horse-to-horse and Grant’s zebra-to-horse extraspecific pregnancies were 64 and 60, respectively, with three live Przewalski’s horse and one live zebra foal being born (27% and 20% from embryos transferred) (Table 2: Kydd et al. 1985; Summers et al. 1987; Allen et al. 2011). Donkey recipients seem to be less suitable that only two pregnancies occurred and neither was carried to term (Table 2). For wild asses, domestic horses are probably not the best recipients, although to date, there are no publications on transfer of wild ass embryos to corroborate this assumption. At least for domestic donkeys, it has been shown that donkey-in-horse pregnancies are prone to conceptus degeneration and abortion between days 80 and 100 (70%) and to late-stage intrauterine foetal death or growth retardation and birth of dysmature foals (15%), probably due to abnormal placentation and/or a maternal cell–mediated rejection response as a result of the mare’s recognition of foreign foetal antigens (Allen 1982).

Cryopreservation of embryos for GRB remains an interesting option for endangered horse breeds, because it has been shown that using storage of both embryos and semen in horses is theoretically 5–10 times cheaper than storage of semen only (Gandini et al. 2007). Cryopreservation of horse embryos has been limited by the finding that whereas small embryos cryopreserve well, larger embryos (diameter > 300 μm) do not. This necessitates flushing embryos from the mare during a small window of time between their descent into the uterus (day 6) and their expansion (day 7). Significant progress in the cryopreservation of expanded equine embryos can be expected in the near future; a recent study reported high survival rates using vitrification after induced collapse of the blastocoele (Choi et al. 2011b), whereas microinjection of the cryoprotectant into the blastocoele may also be a workable method (Scherzer et al. 2008). It is estimated that approximately 200 embryos are needed to restore a breed (Woelders et al. 2003), making embryo transfer a potentially expensive and tedious approach. To date, there are no reports of a frozen embryo repository for endangered horse breeds, in contrast to the case for sheep, goats and cattle (Danchin-Burge et al. 2002; Prentice and Anzar 2011).

Collection and storage of oocytes

Whereas semen and embryo collections are fairly non-invasive, oocytes have to be collected in a semi-invasive way that requires specialized skill and equipment (Brinsko et al. 2011). Techniques have been developed for transvaginal ultrasound-guided puncture of antral follicles in livestock, and these techniques have been applied successfully in horses, albeit with a lower success rate than in cattle. The main advantages of collecting oocytes rather than embryos are that collection can be performed without precise knowledge of the stage of the oestrous cycle and consequently also in dead mares.

There are two main approaches to the recovery of oocytes from live mares: (i) aspiration of the in vivo maturing oocyte from the gonadotropin-stimulated pre-ovulatory follicle (with high recovery rates of 65–80%) and (ii) puncture of all visible follicles on the ovary, to collect immature oocytes for the purpose of in vitro maturation (often with low recovery rates, that is, <30%) (Hinrichs 2010). In the domestic horse, this technique is primarily used to produce embryos from mares that are unable to produce embryos using conventional techniques (Carnevale 2007). In dead mares, ovaries can be removed and transported to the laboratory for the recovery of oocytes from the antral follicles (Smits 2010). Estimated numbers of immature and mature oocytes that can be collected from an ovary by various techniques are summarized in Table 3.

Table 3.   Estimates of number of oocytes/embryos/foals that can be obtained from living mares after aspiration of the gonadotropin-stimulated mature pre-ovulatory follicle, unstimulated follicles or from ovaries of dead mares
Condition of mareLiving – stimulatedLiving – unstimulatedDead
  1. Assuming a 60% maturation rate, a 20% blastocyst rate from matured oocytes subjected to intracytoplasmic sperm injection (ICSI), a 70% foaling rate after transfer (Carnevale et al. 2003; Hinrichs 2010).

No. of immature oocytes39
No. of mature oocytes126
No. of blastocysts after ICSI0.20.41.2
No. of foals0.140.280.84
No. of attempts to obtain one foal73.61.2

In theory, oocytes can be cryopreserved, and thus held for extended intervals. Such oocyte banks could enlarge the gene pool of a particular breed, because a limited number of valuable females could be used to provide oocytes on a regular basis (10–15 days between OPU sessions) (Galli et al. 2007). With an estimated yield of three oocytes per ovum pick-up session, this approach could produce more than 70 oocytes per year from an individual mare (Galli et al. 2007). In human fertility clinics, vitrification of mature oocytes has become routine, with vitrified oocytes yielding pregnancy rates as high as those obtained from fresh oocytes (Cobo and Diaz 2011). At present, equine oocyte cryopreservation is still very difficult; only two foals have been born after vitrification of mature horse oocytes (Maclellan et al. 2002), but attempts to improve current results have been reviewed recently (Saragusty and Arav 2011). An alternative approach may be to freeze ovarian tissue, which, after thawing, can be transplanted or cultured with the aim of producing mature fertilizable oocytes (for review, see Santos et al. 2010).

Equine oocytes can be transferred to the oviduct of an inseminated domestic mare to be fertilized in vivo (Carnevale et al. 2003). This should also work for oocytes of wild equids. Matings of a jack (male donkey) or zebra stallion to a horse mare are of normal fertility and thus one can assume that transfer of donkey or zebra oocytes into horse recipient mares will also result in pregnancies (reviewed by Allen et al. 2011).

Another approach is to fertilize the oocyte by ICSI. Because conventional in vitro fertilization has been very unsuccessful in horses to date, with only two live foals born so far (Palmer et al. 1991), ICSI has been widely adopted by reproductive scientists as a technique to generate horse embryos in vitro (Smits et al. 2012a). Only a few papers are available on fertilization of vitrified-warmed horse oocytes, and fertilization results have been disappointing (Table 4). So far only two papers have reported embryo production after applying ICSI on vitrified-warmed oocytes (Tharasanit et al. 2006; Maclellan et al. 2010). However, embryos generated by ICSI of fresh horse oocytes can easily be frozen owing to their small size (Smits et al. 2009) and can yield pregnancies (Galli et al. 2002).

Table 4.   Results of equine oocyte vitrification using different protocols in terms of oocyte maturation and embryo production in vitro
Oocyte stageCumulus cellsVitrification solutionVitrification methodOutcome parametersReferences
  1. GV, germinal vesicle stage; MII, metaphase II; B, blastocyst; OPS, open pulled straw; EFS, ethylene glycol, ficoll, sucrose; EDS, ethylene glycol, dimethyl sulphoxide, sucrose; SSV, solid surface vitrification; SIB, synthetic ice blocker; IVM, in vitro maturation; ND, not determined.

  2. a4/6 blastocysts transferred developed embryonic vesicles with embryo proper and heartbeat.

GVYesControl60% MIIHochi et al. (1996)
One-step EFS0.25 ml straw2% MII
Two-step EFS0.25 ml straw16% MII
GVCompControl58% MII, 21% BTharasanit et al. (2006)
Exp68% MII, 26% B
CompTwo-step EDSOPS41% MII, 4% B
ExpOPS46% MII, 0% B
In vivo semi-mature plus 18 h IVMNDCommercialCryotop72% survival, 83% cleavage, 40% BaMaclellan et al. (2010)
GV Control51% MIIde Leon et al. (2012)

Collection and storage of somatic cells

Somatic cells, such as white blood cells, skin fibroblasts or granulosa cells, can be collected from a living or recently deceased animal with minimal invasion and are readily cryopreserved. Cryopreservation of equine somatic cells for cloning purposes is already being offered commercially by several companies. The first horse clone was born in 2003 (Galli et al. 2003), and the procedure is quite efficient in horses compared to other species. Recent publications report an overall efficiency of 0.4–1.2% (fusion to foal) and 13–27% of live foals born from transferred blastocysts (Hinrichs et al. 2007; Choi et al. 2009; Table 5), which is higher than the reported 6–15% offspring after transferring cloned cattle embryos (Chavatte-Palmer et al. 2012). Live births of cloned horse foals have been reported by several groups: eight foals by the group of Galli (C. Galli, Cremona, Italy, personal communication), fourteen foals by Texas A&M University (Johnson et al. 2010), an undisclosed number of cloned foals, stated to be ‘70+’ by ViaGen (as reported by the popular press), and at least two foals of Criollo horses in Argentina (D. Salamone, Buenos Aires, Argentina, personal communication). These reported successes suggest that this technique may provide the best choice for the reconstruction of a lost breed or species.

Table 5.   Pre- and post-implantation development of equine embryos cloned by somatic cell nucleus transfer (adapted from Galli et al. 2008)
AuthorsNo. of oocytes recombinedNo. of cleavedNo. of blastocystsNo. of transferredNo. of pregnantNo. of foalsOverall efficiency (foal/oocyte), %Transfer efficiency (foal/oocyte or blastocyst transferred), %
  1. aTransferred immediately after fusion to the oviduct.

  2. bMule embryos.

Woods et al. (2003)b307NANA305a21311
Galli et al. (2003)8417532117410.16
Vanderwall et al. (2004)72NANA62a70
Lagutina et al. (2005)15081339101101920.12
Hinrichs et al. (2006)5674571410420.314
Hinrichs et al. (2007)58948035261691.227
Choi et al. (2009)4243801514920.413

Apparently, MI oocytes can also give rise to healthy foals after nucleus transfer (Choi et al. 2009). If this finding can be confirmed, theoretically vitrified-warmed oocytes might be better used for cloning than for ICSI, since during oocyte vitrification, the maternal chromatin is often damaged (Tharasanit et al. 2006). This damage would be less important in cloning, because the chromatin is removed during the procedure and only the cytoplast remains.

Practical Examples of Successful Salvaging of Genetics

In the current literature, several examples illustrate the feasibility of the above-mentioned techniques in horses and other species. Below, we describe a few equine examples and how emerging techniques may find application in these programmes.

Salvaging mare genetics by oocyte collection

Following the untimely death of a mare, it is critical that the oocytes be maintained in a way that will maximize their viability. There are two approaches to this: (i) keeping the oocyte inside the ovary during transportation; or (ii) holding the oocytes in special media after their removal from follicles. Shipment of ovaries to the laboratory within 10 h post-mortem (Carnevale et al. 2003) yielded good results; remarkably, this delay in removal from the ovary actually increased developmental competence of equine oocytes (Hinrichs et al. 2005). Holding immature oocytes immediately after recovery from the follicle in a special holding medium at room temperature allowed for holding of the immature oocytes for 16–18 h, before being put into maturation conditions, with no effect on developmental competence (Choi et al. 2007). Both approaches can be used in the field when a valuable mare (e.g. of an endangered breed) dies unexpectedly or has to be slaughtered.

Post-mortem embryo production programmes are at present available at Texas A&M University (Hinrichs 2010) and Colorado State University (Carnevale and Maclellan 2006). Oocyte transfer has been applied in clinical practice to obtain offspring from 25 mares that died or were euthanized for medical reasons, with on average 11 oocytes per mare being collected. After in vitro maturation, a mean of 5 oocytes were transferred to the oviducts of inseminated recipients, with 15% of the oocytes forming embryonic vesicles and 24% of the donor mares (6 of 25) producing one or more foals as a result (Carnevale and Maclellan 2006). A similar procedure was used at CSU for salvaging oocytes of a fatally injured mare. Four blastocysts, produced after ICSI of 14 mature oocytes, were transferred and resulted in the birth of two foals (Moellenberg 2009). At Texas A&M University, 110 oocytes recovered from ovaries shipped after the death of valuable donor mares resulted in 21 blastocysts, and ten healthy foals (48%) (Choi and Hinrichs 2011). These observations suggest the feasibility of the technique for application to endangered species.

Salvaging stallion genetics by sperm collection

Intracytoplasmic sperm injection involves the injection of a single immobilized sperm cell into an oocyte and can be applied with sperm of very low quality (in terms of concentration and motility). Using such low quality semen may not be advisable for conventional breeding because it may lead to perpetuating genetics related to low fertility, but in case of endangered breeds or species, it has its value as a last resort. Blastocysts have been produced using sperm of subfertile stallions (Lazzari et al. 2002), sperm that has been frozen-thawed twice (Choi et al. 2006), and even lyophilized sperm (Choi et al. 2011a), in the latter case resulting in the production of a foal (Table 6). Recently, equine morulae have been generated after ICSI using air-dried sperm (Alonso et al. 2011). ICSI is, therefore, very suitable for sperm samples of very low quality or can be applied in circumstances where no liquid nitrogen is available for semen storage.

Table 6.   Blastocyst development after intracytoplasmic sperm injection (ICSI) with sperm of low quality
AuthorsSperm statusNo. of injectedPercentage of injected that cleavedPercentage of injected forming blastocysts
  1. 2F, sperm subjected to two freeze-thaw cycles; SE, injection of sperm extract at the time of ICSI; Lyo, sperm lyophilized from fresh semen; Frz-Lyo, sperm lyophilized from frozen-thawed semen; Frz-D-Lyo, sperm lyophilized from frozen-thawed semen passed through a density gradient before lyophilization.

  2. aFoals born.

Lazzari et al. (2002)Highly motile fertile1177548
Highly motile subfertile786343
Low motility fertile817946
Low motility no pregnancy4690
Choi et al. (2006)Motile control634727
Motile 2F653623
Non-motile 2F632413
Non-motile 2F sperm extract65312
Choi et al. (2011a,b)Control362531a
Control + SE343012a
Lyo + SE35356a
Frz-Lyo + SE36346a
Frz-D-Lyo + SE383211a

Salvaging any animal’s genetics by somatic cell collection

Jurassic Park may well become true in the near future. Preservation of somatic cells allows reproduction of the entire genome of the donor animal. One can preserve genetics from individuals that have died prematurely or unexpectedly; are experiencing reproductive dysfunction in their prime reproductive years; are in their post-reproductive years and have failed or lacked the opportunity to breed; or are prepuberal or castrated animals. As of now, most animal clones have been produced using freshly isolated cells or after cryopreservation of somatic cells in liquid nitrogen. However, a few exceptional studies have generated great promise for the resurrection of extinct or endangered species. In the mouse, two studies have independently shown that it is possible to use somatic cells, frozen without cryoprotectant, for cloning (Li and Mombaerts 2008; Wakayama et al. 2008). In the first study, mouse cells were kept for almost 1 year at −80°C; in the second, the donor cells were derived from a whole mouse that had been in the −20°C freezer for 16 years! Another breakthrough was made in Spain; the Pyrenean Ibex, a wild goat, became the first species ever to become ‘un-extinct’ when, for 7 min in January 2009, a cloned kid was born alive and survived until it died of lung failure. Frozen skin cells from the ear of the last female ibex were used as donor cells and transferred to recipient goat oocytes: 285 embryos were reconstructed, 54 were transferred into 12 mountain goats and mountain goat–domesticated goat hybrids. Two pregnancies resulted and one kid was born (Folch et al. 2009). Because cloning in horses is quite efficient, extraspecific pregnancies are possible. Therefore, it should only be a matter of time before similar results are achieved with donor cells from extinct equids, such as the quagga.


The two most obvious examples of recent extinction of wild equids are the tarpan, the Eurasian wild horse of which the last living individual died in captivity in 1909, and the quagga, a subspecies of the Plain’s zebra, of which the last mare died in the Amsterdam zoo in 1883. If appropriate breeding programmes had been organized at that time, these species might have been saved from extinction. Animal conservationists are now taking action to prevent the extinction of the presently living equids. Besides habitat conservation, breeding programmes have a role to play in that action. The Przewalski’s horse, with a last confirmed sighting of a wild stallion in 1969 (Boyd and King 2011), was in the 1970s considered to be extinct in the wild. During the twentieth century, a founder population of approximately 13 animals captured in the wild was used to breed a new population of captive horses, which, at the end of the century, were successfully reintroduced into Mongolia. Despite some losses caused by harsh winters, diseases and predation of foals by wolves, there are now approximately 306 free-ranging reintroduced and native-born Przewalski’s Horses in Mongolia (Boyd and King 2011). The rebreeding of extinct equine species, making use of genetically close animals, is also underway. For rebreeding the tarpan, the Polish Konik pony has been used; for the quagga, animals of the subspecies Equus quagga burchelli with brown hindquarters (Harley et al. 2009).

As we have reviewed here, techniques in assisted reproduction in horses have evolved to a level that already makes them applicable to the preservation of endangered breeds and, in the near future, possibly to the saving of endangered equine species and perhaps even the resurrection of some that have become extinct.

Conflicts of interest

None of the authors have any conflicts of interest to declare.

Author contributions

All authors took part equally in the development of the manuscript.