ª Embryo advanced one day compared to stage of recipient cycle (days post ovulation) § Embryo delayed one day compared to stage of recipient cycle (days post ovulation) From (Martin et al., 1988)

However, there was a trend toward increase in pregnancy rates as embryo/recipient synchrony approached zero. The use of steroid treated ovariectomised mares as embryo transfer recipients allows progesterone treatment to begin the day of or a few days after donor ovulation. In a study at CSU (McKinnon et al., 1988), day 15 pregnancy rates following surgical embryo transfer into either progesterone (300 mg in oil daily) or progestin (altrenogest - 0.044 mg/kg/day) treated ovariectomised recipients were 75% (15/20) and 70% (14/20) respectively. The increased flexibility of ovariectomised steroid treated recipients is expected to decrease the cost of embryo transfer by minimising number of recipients maintained per donor. Another possible use of progestin treated recipients (intact) is supplementation of mares pregnant when poor uterine tone or endometrial folds are recognised. In these cases the progestin may be withdrawn when secondary CL are present (ie > 35 days). It is also possible to repeatedly measure the mare’s endogenous progesterone production while treating with a progestin such as Regumate, as it does not cross react in any standard progesterone assay. Identification of mares with a progesterone of > 4 ng/ml allows confident withdrawal of the progestin (Shideler et al., 1982; McKinnon et al., 1988).

Few studies have been conducted that permit comparison of surgical versus non-surgical transfer. However, in studies where direct comparison was made, surgical transfer resulted in a higher pregnancy rate than non-surgical transfer (Steffenhagen et al., 1972; Castleberry et al., 1980; Imel 1981; Imel et al., 1981; Squires et al., 1982). Several reasons have been proposed for the lower pregnancy rates after non-surgical transfer. These include: 1) transient, local inflammatory response due to introduction of bacterial contamination; 2) expulsion of the embryo following cervical stimulation that induces prostaglandin mediated uterine contractions, 3) trauma induced by the transfer pipette; 4) site of deposition; and 5) damage to the embryo during transfer. An interesting occurrence was a difference (P < 0.001) in 50 day pregnancy rate from embryos transferred surgically (flank) to the ipsilateral uterine horn (82.5%, 66/80) compared to those transferred to the contralateral uterine horn (54.2%, 32/59) (McKinnon and Squires 1988). Although migration of the early equine conceptus has been demonstrated from its first detection with ultrasound on day 9 to 10 (Ginther 1983), perhaps similar to cattle a local effect of progesterone in the uterine horn ipsilateral to ovulation it is beneficial to the equine embryo immediately post transfer. At the GVEH, embryos are transferred predominantly surgically. This is difficult due to time restraints, however the improvement in pregnancy rates generally make the extra effort worthwhile. Many studies report good pregnancy rates from non surgical transfer, however when comparing pregnancy rates, readers should try to discriminate between studies that purport high pregnancy rates from low numbers of embryos transferred. Although % pregnancy rates of 8/10 is identical to 80/100, it is not the same statistically as in the former just one mare either pregnant or empty will dramatically change the result.

Similar to results with cattle (Curtis et al., 1981), skill of the technician may be the most important factor affecting pregnancy rates when embryos are transferred non-surgically.

Young mares in athletic work make excellent candidates for embryo transfer (Woods and Steiner 1986). Twelve pregnancies from 4 donor mares in competition at the time at transfer were obtained. Clear effects of age on embryo recovery and other fertility parameters were demonstrated by (Vogelsang and Vogelsang 1989). They showed a reduced recovery rate of embryos from subfertile mares of 81/282 (28.7%) compared to maiden mares of 40/66 (60.6%). In addition transferred pregnancy rates were lower (38/78--48.7% versus 34/49--69.4%) and embryonic loss was higher (13/38--34.2% versus 4/34--11.8%). They showed an effect of age with embryo recovery rate from mares aged 2-8 YO, 9-17 YO and 18-28 YO of 80/132--60.6%, 94/183--51.4% and 93/309--30.1%, respectively. Pregnancy rates from the three age groups were 59/84--70.2%, 50/97--51.6% and 50/90--55.6%, respectively (Vogelsang and Vogelsang 1989). A land mark study (Ball et al., 1986) demonstrated similar pregnancy rates on day two in subfertile versus fertile mares (11/14 versus 10/14), which was different from the observed pregnancy rates diagnosed by ultrasonography on day 14 in the same mares (3/15 versus 12/15). This led to the hypothesis that early embryonic death in the oviduct was much higher than previously thought. However recent work from Wisconsin (Carnevale and Ginther 1995), suggested age associated defects in the oocytes were more likely than the oviductal environment as the cause of reduced pregnancy rate and embryo recovery rate in older mares. They showed that oocyte collection and transfer into oviducts of young mares resulted in a pregnancy rate of 11/12--92% versus 8/26--31% from young versus old mares. It should also be mentioned that in older mares reduction in uterine defence mechanisms leads to an abnormal uterine environment which will not only kill sperm, but may harm the embryo after it leaves the confines of the oviduct (McKinnon and Voss 1993a). These workers also previously demonstrated that pregnancy rates on day 12 were different for young pony mares (5-7 YO - n=9) (100%) versus old mares (>15 YO n=22) (32%) and that EED was higher in the old mares (62% versus 11%) (Carnevale and Ginther 1992).

Cooled transported embryos: An important change in the equine embryo transfer industry was the first report of commercial transfer of cooled, transported embryos (Cook et al., 1989). This allows veterinarians to participate in ET without holding and managing recipients (traditionally this has been a major headache). Recipients are expensive, time consuming and probably the most difficult aspect of ET. Centralised stations such as the GVEH have expertise and experience in maintaining large numbers of quality mares as recipients. They are generally maintained after the breeding season finishes to be available early the next year (if not used already). In Australia we have noted an increasing trend in veterinarians who, although unwilling after initial attempts to continue to maintain recipients, were quite excited about sending embryos from multiple different states to the GVEH for transfer. The results have been very satisfactory and in the future we feel it will become the most important aspect of our program. Few studies have critically investigated the effects of different media on the viability of equine embryos. Equine embryos have been recovered and stored in saline/(2%) gelatin mixture or saline/mare serum solution, (Oguri and Tsutsumi 1972) TCM 199 (Allen and Rowson 1972; Oguri and Tsutsumi 1974; Allen and Rowson 1975; Vogelsang et al., 1979; Douglas 1982) and Dulbecco's phosphate buffered saline (Allen and Rowson 1972; Steffenhagen et al., 1972; Castleberry et al., 1980; Griffin et al., 1981; Imel 1981; Squires et al., 1982; Douglas 1982). Most studies demonstrated (Douglas 1982) that in vitro maintenance of embryos for greater than six hours was detrimental to pregnancy rates. However the work of (Pashen 1987; Sertich et al., 1988) were encouraging in the use of longer term storage. The ability to collect embryos on one premise and transport them to another location has many advantages. This would permit movement of embryos within and between countries and eliminate the need for recipient and donor mares to be on the same premise. Recipient mares could be held at a centralised facility in order to maximise use of personnel, expertise and equipment. Transport of equine embryos requires a system of maintaining embryo viability for 12 to 24 hours. Experiments at CSU compared various media for storage of equine embryos. Embryo quality for both whole and bisected embryos stored in Ham's F10 plus 10% foetal calf serum in 5% CO₂, 5% O₂ and 90% N₂ was better at 24 hours than for embryos stored in minimal essential medium with Hank's balanced salt plus 10% foetal calf serum in air (Clarke 1986) or Dulbecco's phosphate buffered saline plus 10% foetal calf serum (Slade et al., 1984). Embryos cultured in Ham's F10 at 5 or 24⁰C did not develop or grow during the initial 12 hour culture period compared to embryos stored at 37⁰C; however these embryos did advance in development when held at 37⁰C for an additional 12 h. In a later trial, embryos stored in Ham's F10 at 24⁰C for 12 h provided similar pregnancy rates compared to those transferred in less than one hour after collection from the donor mare (Clark et al., 1987). Preparation of media for cooling or transport of embryos under field conditions would be more practical with the use of a non gas medium such as Ham's F10 with Hepes buffer. However, at CSU in 1986 (Carnevale et al., 1987), day 14 pregnancy rates for embryos transferred surgically following storage in Ham's F10 plus 10% foetal calf serum plus 5% CO₂, 5% O₂, and 90% N₂ for 24 hours at 5⁰C were better (P < 0.05) (14/20, 70%) than embryos maintained identically but with Hepes buffering of the Ham's F10 media (4/20, 20%). Results from this study suggested commercial application of cooling and transporting embryos was possible. In 1987, preliminary data from commercial embryo transfer experiences with transported cooled embryos (in Ham's F10 plus 10% foetal calf serum gassed with 5% CO₂, 5% O₂ and 90% nitrogen immediately prior to shipment) resulted in a 14 day pregnancy rate of (71%, 12/17) compared to (65%, 11/17) for commercial embryo transfer performed during the same period with freshly harvested embryos. Further studies with much larger numbers have confirmed the initial results (Carney et al., 1991; Squires et al., 1992). In the study of Carnevale et al (Carnevale et al., 1987), 4 of the 18 (22.2%) pregnancies were lost between day 14 and day 35 and in the 1987 commercial application of this procedure, 25% (4/16) of mares lost their pregnancies by day 35. In addition, when data were combined from 1986 and 1987, only 3 of 8 (37.5%) embryos less than 300 um resulted in pregnancies at day 35 compared to 15 of 28 (54%) for larger embryos. The data from (Squires et al., 1992) demonstrated that synchrony of between +2 to -3 had no difference on fertility, however as expected, embryo quality did. The result of overall pregnancy rates of 58% for embryos less than 250 microns versus 73% for those > 500 microns was confirmation of the work above. It is interesting to speculate on why smaller and developmentally more immature embryos (morula and early blastocysts) appear to survive freezing better than cooling and storage at 5⁰C when compared to blastocysts. Veterinarians wishing to send embryos in to the GVEH or any other ET facility need to contact the facility to organise media. The embryos are generally loaded into 0.25 ml straws or are placed into sterile 5 ml plastic tubes and sealed, prior to being placed into an equitainer and transported. Embryo viability is excellent up to 24 hours using this system. It is possible to send embryos between almost any two destinations in Australia within a 24 hour period.

Frozen Embryos: Cryopreservation of bovine embryos has become an integral part of the multi-million dollar bovine embryo transfer industry. Advantages of freezing embryos include: 1) embryos can be stored indefinitely, thus preserving important genetic lines; 2) embryos can be collected from donors and transferred into recipients at a later time, thus minimising the number of recipients; and 3) embryos can be transported within and exported from Australia at our convenience without regard to matching of recipient cycles. Cryopreservation of equine embryos (Slade 1984; Takeda et al., 1984; Slade et al., 1985) resulted in pregnancy rates equal to or better than results reported for frozen thawed bovine embryos, although the number of transfers was quite small (Table 3). Results could be improved further if only embryos at the morula or early blastocyst stage of development are frozen. An interesting development that nicely demonstrates the flexibility of ET is the freezing of equine embryos in the breeding season and successful establishment of pregnancies after transfer in the non breeding season into ovariectomised mares treated for six days with Regumate prior to transfer (Squires et al., 1989). In these cases recipients need to be kept on Regumate treatment until at least 100 days of pregnancy (Shideler et al., 1981).

Table 3. Effect of stage of development on pregnancy rate from frozen embryos.

Adapted from (Slade 1984).

Techniques involve addition of glycerol in one or two steps to a final concentration of 10%. Other cryoprotectants such as 1, 2 propanediol and ethylene glycol have been examined (Seidel, Jr. et al., 1989; Pfaff 1994) with the conclusion that ethylene glycol was toxic to embryos and that embryos of smaller diameter (<200 um) survived the freeze thaw process well in 1, 2 propanediol. However, none of the 1, 2 propanediol protected embryos were transferred (Seidel, Jr. et al., 1989). Numerous studies have been published with the following general conclusions quite consistently reported (Czlonkowska et al., 1985; Seidel, Jr. et al., 1989; Squires et al., 1989; Pfaff 1994). Success from cryopreservation of larger embryos is not commercially viable. The protocol first reported by (Slade et al., 1985) is currently the most efficacious for embryos < 200 um and can be modified to remove glycerol in 3 steps (inclusive of 10% sucrose). Expected pregnancy rates from smaller embryos (late morula’s and early blastocysts) after freeze/thaw is around 50%. The exact reason for failure of large embryos to survive the freeze thaw process is not currently known, although probably relates to differences in permeability at different embryo ages (Pfaff 1994) and the recognition that the inner cell mass cells are more susceptible to damage during cryopreservation than the trophectoderm (Wilson et al., 1987).

One of the major costs and inconveniences of ET is maintenance of recipients. They are expensive, time consuming and sometimes very difficult to synchronise withe donor, especially in small programs. The first demonstration of success with ovariectomised recipients (Hinrichs et al., 1986) resulted in an immediate interest and attendant related publications (Hinrichs et al., 1987; Hinrichs and Kenney 1987b; McKinnon et al., 1988; Hinrichs et al., 1989). These mares were subsequently proven to lactate and provide a normal environment for the foal to develop in. The attraction of using progestin treated ovariectomised mares is the reduced number needed per donor. Generally only one mare is necessary per donor compared to a minimum of 2 for intact cycling recipients (McKinnon and Squires 1988). This is particularly important during the transition into the breeding season, in anoestrus or in the case of inadequate facilities to handle an adequate number of recipient mares. The draw back in their use is the need to provide daily progesterone or progestins (Shideler et al., 1981; McKinnon et al., 1988). Reported pregnancy rates from those studies large enough to be meaningful are good (McKinnon et al., 1988). Control (ovarian intact) mares had a pregnancy rate of 16/20--80% versus 14/20-70% for mares treated with either Regumate, oestradiol primed plus progesterone or oestradiol primed plus Regumate (70% for each group). Removal of the recipients ovaries is not technically difficult via colpotomy (McKinnon et al., 1991).

The incorporation of superovulation methods in commercial embryo transfer programs is somewhat questionable at this time. Some of the advantages would be: 1) possibility for collection of multiple embryos, and 2) higher pregnancy rates per embryo recovery attempt. Some of the disadvantages are: 1) inconsistency in number of mares responding to superovulation treatments and number of embryos collected from multiple ovulating mares; 2) no availability of commercial superovulatory drugs; 3) need for several recipients per cycle; 4) limited information on the effect of successive superovulation treatments; 5) considerable cost of the superovulation treatment; 6) possibility of altering the oestrous cycle of the mare; and 7) possible lowered viability of multiple embryos. The primary disadvantage is the lack of reliable effect and the documented inability to maintain the CL in anoestrus ovulation induction. In cattle, sheep and a variety of non domestic species treatment with equine chorionic gonadotropin (eCG - initially known as pregnant mare serum gonadotropin-PMSG) or FSH reliably induce multiple ovulation and have improved the efficiency of ET. The horse is extremely refractory to stimulation with eCG. Some treatments attempted to induce superovulation in horses and their results are listed below.

GnRH: Gonadotropin releasing hormone (GnRH) has been reported to induce ovulation in seasonally anoestrus mares (Johnson 1986; Johnson 1987; Ginther and Bergfelt 1990), however it has been demonstrated to be ineffective in cycling mares during the regular breeding season for induction of multiple ovulation. (Harrison et al., 1990; Harrison et al., 1991).

FSH :Early studies on the use of FSH (Irvine 1981), demonstrated a mean ovulation rate for FSH-P (porcine FSH) treated mares of 1.7 ± 0.6 versus 1.0 for controls. Another study found that FSH-P was not as efficient as equine pituitary extract (EPE), with ovulation rates of 2.2 for EPE, 1.6 for FSH and 1.0 for controls (Squires et al., 1986). Dosages of FSH were between 150 and 200 mg as often as twice daily in mid to late dioestrus. A study from Cornell University (Fortune and Kimmich 1993)showed an ovulation rate of 1.80, 1.54, and 1.50 for mares treated with 8,16 or 32 mg purified porcine FSH (Folltropin) compared to 1.2 for the controls. Injections were given twice daily from day 6 (PGF_2a on day 6) until follicles ³ 40 mm were detected. Unfortunately the next season the ovulation rates for the lowest effective dose when studied in more detail were not different from controls (Fortune and Kimmich 1993). In conclusion, although the higher doses of purified FSH have not been adequately studied, it appears that FSH is only marginally successful for increasing ovulation rates in the horse and that the doses necessary are likely cost prohibitive.

Equine pituitary extract (EPE:):

It has been clearly demonstrated that multiple ovulation can be induced in both seasonally anovulatory and cycling mares with EPE (Douglas et al., 1974; Lapin and Ginther 1977; Douglas 1979; Woods and Ginther 1982a; Woods et al., 1982b). However, the induction of ovulation in the seasonally anovulatory mare is impractical since a large number of those mares fail to continue to cycle after the induced ovulation and if they become pregnant fail to maintain the CL. In addition, effective doses of pituitary extract for induction of multiple ovulation is higher in anovulatory mares. Pituitary extract administered during dioestrus for a period of approximately one week appears to be an effective means of inducing multiple ovulation in cycling mares. However, ovulation rates are quite low in comparison to those obtained in cattle, usually ranging from 2 to 4 ovulations per mare. Unfortunately, at present no equine pituitary extract is commercially available. The questions of 1) are multiple ovulations are associated with a reduced number of embryos per ovulation ? and 2) are they reduced in viability ?, have not adequately been resolved. It is most likely that the reduced number of embryos (embryos/ovulation) is related to those mares that have multiple ovulations on the same ovary. In relation to viability, although it was suggested that reduction in viability of multiple embryos occurred by one group (Woods and Ginther 1982a; Woods et al., 1982b), much of this was recorded before ultrasonography. More recent experiments (Squires et al., 1987) tend to refute earlier work and encourage the use of embryos collected from superovulatory programs. A further study from CSU (Dippert et al., 1994) was able to show fertilisation rates and embryonic development from oviductal embryos collected from either EPE treated mares or controls were identical.

Inhibin vaccination: Inhibin is produced by the granulosa cells of the dominant follicle. It serves to suppress FSH production. It is generally inversely related to FSH at all stages of the cycle. Increasing FSH through inhibin immunisation has increased ovulation rates in cattle and sheep vaccinated against either the ovine or bovine alpha subunit of inhibin respectively. The GVEH were first to report on using inhibin immunisation to increase ovulation rates in mares (Hintz and Schryver 1972). Mares were immunised on day 0 and 35. Ovulation rates per immunisation and in the control groups were 1.2 per cycle. Between the two vaccinations ovulations were 1.86 and then rose to 2.29 in the cycle after immunisation. All mares either double or triple ovulated (McKinnon et al., 1992). This observation was confirmed by workers in California (McCue et al., 1992) who vaccinated normally cycling mares at three week intervals, five times. Similar results occurred and in addition embryos were recovered at the rate of 1.6 versus 0.7 per cycle in immunised and control mares respectively. In a later study passive immunisation against inhibin did increase ovulation rate, however it was not as high as active immunisation and had other attendant side effects such as hypersensitivity and on one occasion death (McCue et al., 1993). Although immunisation against inhibin has been successful it was not able to produce large numbers of ovulations. It is not to our knowledge being actively researched.

Embryo splitting: Embryo bisection to create identical twins is useful for research into diseases that have an inherited basis with an environmental predisposition such as OCD and other forms of DOD (developmental orthopaedic disease). The matched full siblings removes allot of the inherent genetic variation. Commonly one twin is used as a control and the other is assigned to a specific treatment. The reduction in numbers of experimental animals decreases costs and unnecessary animal experimentation. Another reason to bisect equine embryos is the production of almost perfectly matched twins as show or carriage animals. A further reason would be to improve the overall number of pregnancies from a given number of embryos. In cattle the technique results in ~ 50% more pregnancies than transfer of single non manipulated embryos with approximately 20% production of identical (monozygotic) twins (Williams et al., 1984). Early attempts to create identical twins were successful with embryos recovered from the oviduct but were quite time and labour intensive (Allen and Pashen 1984). Two pairs of identical twins were obtained from 13 demi and 17 quarter embryos (Allen and Pashen 1984). Workers at CSU were first to obtain identical twin pregnancies from non surgically recovered day 5.5 or 6 embryos (Slade et al., 1984). Another study at CSU (McKinnon et al., 1989) demonstrated that the zona pellucida was not necessary for the continuing development of bisected embryos in vivo and the complex, expensive and sophisticated micro-manipulation equipment were not necessary. These workers showed that pregnancy rates per original embryo were decreased (14/30) compared to the non manipulated embryos (12/15) and that 5 pairs of identical twins could be obtained from 30 embryos (1:6 ratio). Only morulae and early blastocyst were split in this experiment. In England 2 sets of identical twins were obtained from 14 embryos bisected (Skidmore et al., 1989). It was interesting to note that 0/16 demi embryos from blastocysts established a pregnancy compared to 8/12 demi embryos from morulae. To our knowledge the success reported above have not been improved on and this has probably been responsible for the limited usefulness of the technique.

Transfer of day 10 embryos: We have been experimenting on the use of ten day old embryos for transfer. When the donor is identified as pregnant the embryo is harvested and placed either surgically or non surgically in a recipient mare. The first foal from this technique was born at the GVEH in 1993, however the techniques needs refining as to date we have only utilised day 6 recipients and the pregnancy rates of surgical versus non surgical transfer need to be directly compared in the same breeding season. The technique if demonstrated efficacious offers distinct advantages for the veterinarian involved only casually in equine ET as the use of expensive microscopic equipment is avoided and only mares that are pregnant are actually flushed.

Oocyte retrieval: For a variety of reasons, gamete research in the horse has lagged behind species such as cattle. This has had the advantage that equine researchers have been able to benefit from the information explosion associated with other species, including human programmes. In cattle, the in vitro maturation of oocytes from immature follicles harvested from either slaughterhouse collected ovaries or by trans-vaginal ultrasonographic guided ovum pick up (OPU) has become a routine procedure. A recent report on OPU in cattle demonstrated that over a six month period it was possible for 16% of immature oocytes to develop to a transferable embryo stage and a pregnancy rate of 40% was obtained. This was associated with the production of about 50 calves in a six month period, which is a fourfold increase over the expected annual production with standard ET techniques.

Initial studies in the horse concentrated on mature oocyte collection and most were performed through the flank in the standing mare by aspiration through large bore needles of follicular fluid while the follicle was held in situ pre rectum (McKinnon et al., 1986; Vogelsang et al., 1986; Palmer et al., 1986; McKinnon et al., 1988) Oocyte recovery rates of 6/18 (33.3%) (McKinnon et al., 1988), 45/63 (71.4%) (McKinnon et al., 1986), 11/36 (30.5%) (Vogelsang et al., 1986) and 9/16 (56.3%) (Palmer et al., 1986) were reported respectively. Other techniques such as colpotomy were also described (Hinrichs and Kenney 1987a) but were not associated with an improved recovery (9/13 in hCG stimulated mares compared to 15/21 in non stimulated follicles) and were abandoned.

More recently mature oocyte collection has been performed with a trans-vaginal ultrasound guided approach (Gardner et al., 1993; Farstad et al., 1993). In Utrecht five Dutch warmblood mares were used to aspirate 200 follicles (Gardner et al., 1993). They reported 34 oocytes retrieved in 24 sessions. Pre treatment with PGF_2µ or hCG did not improve oocyte recovery compared to non treated mares (8.2%, 15.3% and 26.9% respectively), however better recovery was obtained with double channel needles compared to single (24.4% versus 12.3%) (Gardner et al., 1993). At CSU 99 preovulatory follicles yielded a total of 62 oocytes (63%). The best results were obtained from use of a 12 gauge double lumen continuous flushing device (36/43-84%)(Farstad et al., 1993). During dioestrus recovery rates were low (71/323-22%) and were not affected by the type of needle (Farstad et al., 1993). The procedure has also been successful in recovering oocytes from mares that have been superovulated (Farstad et al., 1993) and early pregnant mares (Gutierrez et al., 1993). Immature oocytes are much more difficult to obtain due to their close association to the granulosa calls (Yurewicz et al., 1993; Macilwain 1993). They are best obtained from slaughter house ovaries by dissection of the follicle and gentle scraping of the surface (Funahashi et al., 1993; Yurewicz et al., 1993), however there are numerous reports of their recovery with OPU (Gardner et al., 1993; Farstad et al., 1993).

Transfer of in vivo matured oocytes into the oviduct of an already bred recipient mare was first reported at CSU (McKinnon et al., 1988). Fifteen oocytes were transferred and two days later three fertilised oocytes were identified from a total of ten recovered (30%). These were transferred and two established pregnancies, with one producing a live foal (McKinnon et al., 1988). The technique has numerous advantages for those mares that fail to provide embryos by standard methods. Because it involves more physiological techniques than IVF it is commercially applicable (McKinnon et al., 1988) and the equipment needed is much less sophisticated (McKinnon et al., 1986). The main drawback is the need to recover the recipients own oocyte that she would have ovulated . In cases where this is not possible, the identity of the foal will be in question until confirmed with blood or DNA typing. A variation was reported in England (Funahashi et al., 1993), with in vitro maturation of immature oocytes before transferring them into oviducts of bred recipients. More recently the ability of in-vivo matured oocytes to be fertilised in oviducts of young fertile mares was compared for young fertile mares and old infertile mares (Carnevale and Ginther 1995). They found that more (P<0.005) oocytes fertilised and developed from young mares (11/12-92%) compared to old mares (8/26-31%) and hypothesised that it was the oocyte of the older mare, not the oviduct that caused the problem of infertility (Carnevale and Ginther 1995). The very high pregnancy rate reported in this study was apparently due to a modification of the original technique (McKinnon et al., 1988) to include a culture period of 16-20 hours prior to transfer (Carnevale and Ginther 1995). Another study reported lower pregnancy rates (2/26-7.7%) (Ray et al., 1994). However it has been proposed that a commercial application for GIFT should be forthcoming (Hinrichs et al., 1998) based on 6 of 8 transferred oocytes establishing a pregnancy.

In vitro oocyte maturation: Immature oocytes for maturation studies have traditionally been recovered from slaughter houses or ovariectomy, until the advent of equine OPU (Shabpareh et al., 1993). It is possible the future of equine IVF in the horse will be related to in vitro maturation (IVM) studies because in vivo matured oocytes are expensive (one per cycle) and time consuming to obtain. Fertilisation but not pregnancies have been obtained from IVM oocytes (Zhang et al., 1989; Zhang et al., 1990; Zhang et al., 1991; Delcampo et al., 1995). The problem may be associated with the culture of the oocytes or capacitation of the spermatozoa (necessary for penetration and fertilisation) (Blue et al., 1989; Ellington et al., 1993). We expect that intra-cytoplasmic sperm injection (ICSI) will replace the current techniques involving difficult capacitation procedures.

In vitro fertilisation: The first foal born from IVF was born in France on June 14, 1990 (Palmer et al., 1991). Unfortunately another has not been forthcoming in any laboratory anywhere in the world. This highlights the difficulty of obtaining IVF foals in the horse. The procedure is well documented in humans and cattle. The horse has a zona pellucida that is very resistant to sperm penetration or spermatozoa that are very difficult to capacitate. Most successful fertilisation under IVF conditions in the horse has revolved around the use of a toxic agent, calcium ionophore A23187 (Blue et al., 1989). This ionophore dramatically reduces the life of spermatozoa. Another difficulty that holds back progress in this area is the lack of a suitable number of in vivo matured oocytes.

Intracytoplasmic sperm injection (ICSI): Successful ICSI has been reported in the horse (Squires et al., 1996).In this experiment 4 injected eggs from slaughter house oocytes were transferred into the oviduct of recipient mares and one pregnancy established and resulted in the first ICSI foal. More recently we have obtained foals from ICSI with a different approach than originally reported (McKinnon et al., 1998). We obtained in vivo matured oocytes and injected them with a single frozen sperm. Two foals have been born from this procedure in 1998. The most important demonstration of the later technique was that the procedure has been shown to be repeatable and the eggs were collected from live cycling mares.

This is quite exciting as it bypasses capacitation procedures to help the sperm penetrate the zona pellucida. We expect the immediate future of IVF to involve ICSI of either in vitro or in vivo matured oocytes.

Xenogenous fertilisation: Apparently attempted fertilisation in the oviduct of another species of animal has only been reported once (McKinnon et al., 1988). This area although not reported successful in a small number of rabbits deserves further attention as other species may be easier to work with for the procedures of oocyte fertilisation and recovery.

Other gamete manipulations: Instead of transferring the oocytes to the oviduct as in GIFT (McKinnon et al., 1988) another technique is to transfer immature oocytes to the follicle of another horse. If this horse is bred then the possibility exist to collect multiple embryos from the recipient mare and then transfer them into subsequent recipients. The genetics of the foals can be sorted out at a later date. The first embryos produced by this technique were reported from Tufts University (Hinrichs and DiGiorgio 1991). They transferred 146 oocytes into 12 mares and recovered 18 embryos from 6 mares and no embryos from 6 mares. Embryos in excess of the number of ovulations were obtained in four mares (excess embryos of 1,2,3 and 6). It was interesting to note that most (127) oocytes were immature and that the only recipient to receive expanded oocytes (n=6) was the one that provided the 6 excess embryos. The relatively low success as a method to mature immature oocytes suggests that they may need more time in the follicle prior to ovulation (most mares ovulated around 40 hours after transfer) or that they had been collected from atretic follicles. In 1989 at the GVEH we transferred 40 oocytes that had been collected from ovariectomised mares and then stored in liquid nitrogen after vitrification. Eight mares were bred and a total of five single vesicles were identified. This indicated that the follicle went on to mature properly despite intra-follicular transfer of oocytes.

Equine ET is slowly being used more widely as more people understand it benefits. The costs and expertise necessary have probably precluded its more widespread use. The benefits of ET research are far more reaching than the improvement in fertility of ET programmes. All of us, no matter what reproductive discipline have benefited by knowing we can treat a mare with an intra-uterine infusion after ovulation and when the embryo is due to arrive in the uterus. Perhaps it would be best to end this paper with a direct quote from Keith Betteridge whilst he was summarising the papers from the Third Symposium on Equine Embryo Transfer in Argentina in 1993. "I think that it is fair to say that embryo transfer and related techniques have taken us a long way forward in our understanding of equine reproduction. It is equally certain that these techniques can take us very much further, as even the most cursory look at their application in other species will testify. We are limited, it is true, by an innate distrust of the concept of artificial breeding in important and influential segments of the equine industry and it is up to us to explain the advantages that are to be had from research that depends entirely on efficient (more efficient !) use of embryo related studies. Equally important to our future, in my opinion, is the necessity to explain to those who govern the financing of research the fundamental biological value of the horse in comparative studies of reproduction, medical as well as veterinary. Too often we apologise for the expense of keeping horses rather than mice; perhaps we need to stress how much germinal or embryonic material can be painlessly harvested from a single mare or stallion."(Betteridge 1993)

Reference List

Return to Angus McKinnon Title Page