Animal Reproduction Science 98 (2007) 39 55

Transcription

1 Animal Reproduction Science 98 (2007) Developmental competence of equine oocytes and embryos obtained by in vitro procedures ranging from in vitro maturation and ICSI to embryo culture, cryopreservation and somatic cell nuclear transfer C. Galli a,b,, S. Colleoni a, R. Duchi a, I. Lagutina a, G. Lazzari a a Laboratorio di Tecnologie della Riproduzione, Istituto Sperimentale Italiano Lazzaro Spallanzani, CIZ s.r.l., Via Porcellasco 7f, Cremona, Italy b Dipartimento Clinico Veterinario, Università di Bologna, Via Tolara di Sopra 50, Ozzano Emilia, Italy Available online 17 October 2006 Abstract Development of assisted reproductive technologies in horses has been relatively slow compared to other domestic species, namely ruminants and pigs. The scarce availability of abattoir ovaries and the lack of interest from horse breeders and breed associations have been the main reasons for this delay. Progressively though, the technology of oocyte maturation in vitro has been established followed by the application of ICSI to achieve fertilization in vitro. Embryo culture was initially performed in vivo, in the mare oviduct or in the surrogate sheep oviduct, to achieve the highest embryo development, in the range of 18 36% of the fertilised oocytes. Subsequently, the parallel improvement of in vitro oocyte maturation conditions and embryo culture media has permitted high rates of embryo development from in vitro matured and in vitro cultured ICSI embryos, ranging from 5 to 10% in the early studies to up to 38% in the latest ones. From 2003, with the birth of the first cloned equids, the technology of somatic cell nuclear transfer has also become established due to improvement of the basic steps of embryo production in vitro, including cryopreservation. Pregnancy and foaling rates are still estimated based on a small number of in vitro produced equine embryos transferred to recipients. The largest set of data on non-surgical embryo transfer of in vitro produced embryos, from ICSI of both abattoir and in vitro-matured Ovum Pick Up (OPU) oocytes, and from somatic cell nuclear transfer, has been obtained in our laboratory. The data demonstrate that equine embryos produced by OPU and then cryopreserved can achieve up to 69% pregnancy rate with a foaling rate of 83%. These percentages This paper is part of the special issue entitled Developmental Competence of Oocytes and Embryos guest edited by Barry Bavister. Corresponding author at: Laboratorio di Tecnologie della Riproduzione, Via Porcellasco 7f, Cremona, Italy. Tel.: ; fax: address: cesaregalli@ltrciz.it (C. Galli) /$ see front matter 2006 Elsevier B.V. All rights reserved. doi: /j.anireprosci

2 40 C. Galli et al. / Animal Reproduction Science 98 (2007) are reduced to 11 and 23%, respectively, for cloned embryos. In conclusion, extensive evidence exists that in vitro matured equine oocytes can efficiently develop into viable embryos and offspring Elsevier B.V. All rights reserved. Keywords: Equine; Oocyte; Ovum pickup; ICSI; Embryo culture; Cloning 1. Introduction From an historical perspective the development of assisted reproductive technologies in the horse dates back to the late nineteenth century when Sir Walter Heape established the first equine pregnancies obtained by artificial insemination (Heape, 1898). Since that time, progress in assisted reproduction in the horse has been continuous although at an irregular pace compared with other domestic species. Besides artificial insemination, which is now well developed in the equine industry (Squires, 2005), other technologies based on in vivo and in vitro procedures of embryo production have emerged only in recent years. This situation is in sharp contrast with the rapid development and application of such technologies in several other species including cattle, sheep and pigs, to the point of successful commercial exploitation. Recent progress in assisted reproduction in the horse is the focus of this review with the main emphasis on the procedures of in vitro embryo production, OPU, ICSI and cloning. A description of the procedures and the most relevant results will be provided together with personal observations and comments on the role that such technologies can play in equine breeding. 2. Oocyte maturation in vitro and conventional in vitro fertilization of abattoir-collected oocytes The first successful report of in vitro maturation of horse oocytes was that of Fulka and Okolski (1981). The first embryo production from in vitro matured horse oocytes was reported in 1989 (Zhang et al., 1989) when oocytes collected at the abattoir were matured in vitro and, after transfer to the oviducts of inseminated mares and recovery by uterine flushing 7 days later, developed to the blastocyst stage. At that time, research on the technology of in vitro maturation of abattoir oocytes was at its peak in ruminants, driven not only by basic scientific questions but also by the declared interest of the cattle industry. This favorable context was further enhanced by the virtually unlimited availability of abattoir-collected bovine oocytes and the total freedom to conduct research with such material. As a consequence, there was a flourishing of scientific publications and in a relatively short time in vitro embryo production technology, including in vitro maturation, fertilization and embryo culture, became established in ruminants. By contrast, the main limitations in the case of the horse were the scarce availability of abattoir ovaries, the much poorer oocyte collection rate compared to other species and above all, the very limited interest from horse breeders and breed associations. In addition, the recovery of oocytes from horse ovaries presents some more technical problems: the collection of oocytes requires incision of follicles and scraping of the follicle wall with a curette and extensive flushing to detach the cumulus oocyte complexes (COCs). These difficulties are reflected in an increased length of time and number of personnel required for collecting enough oocytes to carry out even the simplest experiment. In our laboratory the difference between species is over 10-fold both for time and for personnel: for example, to collect 100 horse oocytes requires 4 technicians working

3 C. Galli et al. / Animal Reproduction Science 98 (2007) for 3 4 h while for the same number of bovine oocytes only 2 technicians working for min are sufficient. It is obvious that all these difficulties represent a major limitation for basic research on equine oocytes and for all the downstream technologies from in vitro fertilization to ICSI, embryo culture and cloning. Another interesting aspect peculiar to the horse oocyte is the frequent collection of oocytes with expanded cumulus that in our laboratory accounts for about a third of the recovered COCs. In other species such as ruminants and pigs, the presence of expanded cumulus is linked to the collection of oocytes from atretic follicles and these oocytes are generally discarded immediately because of their extremely low developmental capacity (de Loos et al., 1989). In horses, however, oocytes with expanded cumulus mature normally and have normal developmental competence. Table 1 presents the data from a large study conducted in our laboratory that gives an exact measure of the efficiency of oocyte collection from abattoir ovaries and the expected maturation rate from compact and expanded COCs. The data indicate that each ovary contains on average 5.3 follicles with a diameter above 5 mm that are suitable for oocyte collection. If these follicles are scraped and washed, the recovery is 3.8 oocytes corresponding to about 70% collection efficiency. The need to scrape the follicles derives from the tight connections between the cumulus and the membrana granulosa and between the latter and the follicle wall (Hawley et al., 1995). In horses, it is not unusual to recover either oocytes surrounded only by the corona cells or, conversely, oocytes surrounded by large sheets of membrana granulosa cells. This indicates in the first instance most probably a COC detached from the membrana granulosa by the extensive washing that follows the scraping of the follicle, and in the second instance the effect of the scraping itself. After the oocytes are collected and selected on the basis of cumulus morphology, they are transferred to maturation medium and allowed to mature for 24 h. In our culture conditions, as shown in Table 1, the maturation rate ranges from 51.1 to 60% for compact and expanded COCs, respectively. This difference is statistically significant and in agreement with other studies (Hinrichs and Williams, 1997; Hinrichs and Schmidt, 2000; Hinrichs et al., 2005a) in which oocytes with expanded cumulus were found more capable of complete maturation than oocytes with compact cumulus. In addition, the ability of oocytes with expanded cumulus to develop to the blastocyst stage is not statistically different from oocytes with compact cumulus (see Table 2), unlike the situation in ruminants or pigs. In calculating the maturation rate, we included the large number of degenerating oocytes identified after IVM that are another peculiarity of the equine species when oocytes are collected from abattoir ovaries. At the time of collection, the oocytes that will degenerate in culture cannot be identified but at the end of maturation and after removal of the surrounding somatic cells they are easily detected. If oocytes that degenerate in culture are not included in the calculation, the percentage of mature oocytes increases considerably into the range of 70 85% and is comparable to that obtained from Ovum Pick Up oocytes (see Ovum Pick Up ). We should mention at this point that it is not always clear from the literature how the maturation rate of equine oocytes is calculated and whether degenerated oocytes are taken into account. It is interesting to note that in Table 1 the distribution of the degenerated oocytes is equal between the compact and expanded COC groups. This is somewhat surprising because, in other species, the presence of a compact and healthy cumulus cell layer is considered an indicator for good oocyte morphology. Because no lysis is observed with OPU oocytes, this finding implies that the postmortem modifications occurring in the large equine ovaries are responsible for the degeneration of abattoir oocytes. Indeed, when equine oocytes were placed into maturation medium immediately after slaughter, the rate of maturation was greater and the rate of degeneration lower than for oocytes collected after transport of ovaries to the laboratory (Hinrichs et al., 2005b). In this

4 Table 1 Maturation competence of horse oocytes derived from expanded or compact COCs No. of ovaries No. of follicles (per ovary) No. of COCs expanded (per ovary) No. of COCs compact (per ovary) No. of COCs matured after IVM No. of COCs degenerated after IVM Expanded (%) Compact (%) Expanded (%) Compact (%) (60.0%) a (51.1%) b (30.0%) (33.4%) χ 2 -test: values with different letters (a and b) differ (p < 0.05). 42 C. Galli et al. / Animal Reproduction Science 98 (2007) 39 55

5 C. Galli et al. / Animal Reproduction Science 98 (2007) Table 2 Effect of cumulus morphology on the developmental competence of equine oocytes matured in vitro, fertilised by ICSI and cultured in vivo in the sheep oviduct Cumulus morphology No. injected (MII) No. cleaved Cleavage rate No. of blastocysts % Blastocyst/MII injected Compacted % a % a Expanded % b % a χ 2 -test: values with different letters (a and b) differ (p < 0.05). previous study, greater rates of degeneration in culture were observed only in oocytes having compact cumulus; evaluation of oocyte chromatin configurations led to the conclusion that these oocytes were juvenile and that their chromatin was labile to damage during post-mortem storage within the ovary. Interestingly, in the horse, in spite of the early success of in vitro maturation (Zhang et al., 1989), no more relevant research on the issue was reported for some time. Many efforts instead concentrated on conventional in vitro fertilization (IVF) and the failure to obtain consistent results forced the research to focus back on the use of in vivo matured oocytes. Unfortunately, the long series of successes with IVF in several other species never occurred with horses and reports of successful IVF in this species are very few. Only two foals were reported as born from IVF and both were derived from in vivo matured oocytes collected by OPU from gonadotropin-stimulated donors (Palmer et al., 1991; Bezard, 1992). There is no published report of equine pregnancies derived from both in vitro maturation and conventional IVF. Failure of IVF remains a mystery but is probably related to inefficient sperm capacitation (Alm et al., 2001), changes in the zona pellucida (Dell Aquila et al., 1999; Hinrichs et al., 2002) or to incomplete in vitro maturation (Li et al., 2001) although this latter aspect has been more recently addressed and resolved by using more suitable culture methods. A variety of oocyte maturation conditions has been evaluated using different culture media comprising TCM199 (Willis et al., 1991; Dell Aquila et al., 1997; Hinrichs and Schmidt, 2000; Galli et al., 2001; Galli et al., 2002; Lagutina et al., 2005), B2 (Willis et al., 1991) and Ham s F10 (Shabpareh et al., 1993), supplemented with different concentration of serum, hormones or follicular fluid. These conditions have resulted in maturation rates varying from 20 to 85% but none of these has increased the efficiency of conventional IVF. 3. Intracytoplasmic sperm injection (ICSI) and embryo culture The application of ICSI to the horse has overcome the formidable barrier of inefficient IVF, finally resulting in the first pregnancy derived from an in vitro matured oocyte (Squires et al., 1996), which was carried successfully to term. This success was followed by a period of variable results until the development of ICSI using the piezo drill, which removed most of the inconsistency of the technique. The advantage of ICSI as compared to IVF is the ability to widen the choice of the stallion to be used, including those with poor sperm motility and reproductive performance in vivo. In an original study, Lazzari et al. (2002) compared the developmental capacity of in vitro matured oocytes fertilized by ICSI with frozen-thawed stallion semen of different motility and/or different fertility in the field. There was no difference in either cleavage or advanced embryo development rates among oocytes injected with spermatozoa from stallions of good, poor and no fertility in the field, as long as a motile spermatozoon was selected for ICSI. In contrast, in semen

6 44 C. Galli et al. / Animal Reproduction Science 98 (2007) Table 3 Cleavage and development rate following ICSI with semen of different motility and fertility post-thawing (Lazzari et al., 2002) Group of stallions No. of metaphase II injected No. cleaved Cleavage (%) No. of comp. morulae and blastocysts % Comp. morulae and blastocysts/injected A a a B a a C a a D b b χ 2 -test: numbers within columns with different letters (a and b) differ (p < 0.05). having very poor motility post-thawing and no fertility, no motile spermatozoa were observed after dilution of semen in PVP to perform ICSI, injected oocytes had a significantly lesser cleavage rate and the few cleaved embryos were not capable of developing to the compact morula or blastocyst stages (see Table 3). To demonstrate the viability of the embryos obtained, four grade 1 fresh embryos produced by ICSI performed with semen not fertile in the field after freezing and thawing were transferred non-surgically in pairs into two recipients, giving rise to two twin pregnancies. These results indicate that ICSI can allow the use of semen with poor motility or no in vivo fertility, provided that a motile sperm cell is selected for injection. In another study (Choi et al., 2006a), oocytes injected with immotile spermatozoa isolated from semen subjected to two freeze-thaw cycles were capable of blastocyst development. A few groups have reported the occasional birth of foals following ICSI of in vitro matured oocytes, after either surgical transfer of early embryos to the oviduct (Cochran et al., 1998)orin vitro embryo culture to the blastocyst stage and transcervical transfer to the uterus (Li et al., 2001; Hinrichs et al., 2005a). The progress in IVM and ICSI technologies has increased efforts to design suitable culture systems for early cleavage stage embryos. Many different culture conditions have been reported for preimplantation development of ICSI fertilized horse oocytes, including defined media such as G1.2 (Choi et al., 2002), DMEM-F12 and CZB (Choi et al., 2004b) and modified SOF (Galli et al., 2002). Earlier work evaluated co-culture with somatic cells including Vero cells (Dell Aquila et al., 1997), oviduct epithelial cells (Battut et al., 1991), cumulus cells (Li et al., 2001), granulosa cells (Rosati et al., 2002) or culture in conditioned media (Choi et al., 2001). In most of these systems, however, the blastocyst rates remained low, ranging from 4 to 16%. In contrast, the culture of presumptive zygotes following ICSI, in vivo either in the mare oviduct or in the surrogate sheep oviduct, allowed much greater development. A comparison between the published reports on in vivo culture of ICSI early cleavage stage embryos in the oviducts of mares (Choi et al., 2004a) or temporary recipient sheep (Galli and Lazzari, 2001; Lazzari et al., 2002; Galli et al., 2002, 2000) and in vitro culture in various culture media (see above) clearly demonstrated that the in vivo environment supports high blastocyst development, approximately 36% of injected oocytes in both the mare and sheep oviducts. Recently, an in vitro culture system has been developed using DMEM/F-12 medium under a mixed gas atmosphere that provides blastocyst development rates similar to those seen in vivo (27 38%; Hinrichs et al., 2005a; Choi et al., 2006a,b). When cell number counts were compared (Tremoleda et al., 2003) among in vivo produced embryos and those produced by in vitro culture in a modified SOF medium, both on day 7 of development, the in vitro produced embryos had significantly fewer cell numbers, resembling a day 5 rather than a day 7 embryo. This difference

7 C. Galli et al. / Animal Reproduction Science 98 (2007) is then taken into account when embryos are transferred to synchronized recipients (see section on embryo transfer). The advantage of having embryos at earlier developmental stages, however, is to improve their cryopreservation. In fact, it is notoriously difficult to freeze in vivo recovered equine embryos because of their large size. To recover equine embryos of an appropriate size for freezing, accurate timing of ovulation and embryo recovery must be performed, e.g., performing uterine flush 8 days after administration of hcg (6.5 days after ovulation; Eldridge-Panuska et al., 2005). With in vitro-produced embryos the advantage is that it is possible to freeze the embryos at the best stage, i.e., at the late morula or early blastocyst stage, to maximize survival post-thawing. 4. Ovum Pick Up The application of in vitro reproductive techniques in the horse is limited to animals with fertility problems or those engaged in sporting activities. This means that, contrary to other farm animal species, there is no interest to produce embryos from abattoir-collected ovaries if not for research purposes. As a consequence, a number of techniques, including laparotomy, colpotomy or blind aspiration through the paralumbar fossa, have been attempted to recover oocytes from live donors without compromising their reproductive and sporting abilities. At present, the method in use for recovery of immature oocytes from live horses is ultrasound guided transvaginal oocyte recovery or Ovum Pick Up (OPU). Similarly to other species, the use of this technology started about 15 years ago and has been subsequently refined (Cochran et al., 1998; McKinnon et al., 1998). This technology has been proven safe and repeatable in cattle and has the considerable advantage of not requiring any hormonal stimulation of the donor. This aspect is of considerable importance in the horse because superovulation is still unsatisfactory and inconsistent. Oocyte recovery rates from immature follicles in the horse are minimal unless vigorous flushing and scraping of the follicle is performed, because of the close attachment of the COC to the follicle wall. The data reported in the literature are abundant as far as the technique of immature oocyte recovery is concerned (Bogh et al., 2003; Vanderwall et al., 2006) but very limited with reference to the assessment of the developmental potential of such oocytes for producing horse embryos from live donors on a routine basis. The most practical use of OPU or trans-abdominal follicle aspiration is to recover in vivo matured oocytes (Carnevale, 2004) for oocyte transfer. This technique involves the recovery of the oocyte from the preovulatory follicle after gonadotropin stimulation and its transfer to an inseminated recipient mare at the time that it would have been ovulated if left in the donor mare. These studies have shown that the developmental capacity, as expected, is enhanced compared with previously used technologies resulting in highly satisfactory pregnancy rates except for in very old donors whose oocytes are intrinsically compromised (Carnevale and Ginther, 1995; Carnevale et al., 2005). In our laboratory, we have conducted a large study on the developmental capacity of immature oocytes recovered by OPU from both experimental and commercial donors within a commercial OPU service exactly as it is performed in cattle. The experimental OPU programme was conducted on five donors, four Haflingers and one Trotter, for a total of 54 consecutive OPU sessions at days interval. The results, previously published in part (Galli et al., 2002), indicate an average number of 6.6 follicles aspirated and 3.2 oocytes recovered with a recovery rate of 48%. The total number of oocytes collected was 141 of which 101 reached metaphase II following in vitro maturation (71.6%). The matured oocytes were fertilized by ICSI giving rise to 74 (52%) cleaved embryos that were either transferred into the sheep oviduct after cleavage and recovered on day 7 (day 0 = fertilization day) or cultured in vitro. We obtained 32 blastocysts (0.71 per OPU)

8 46 C. Galli et al. / Animal Reproduction Science 98 (2007) corresponding to a 31.7 or 43.2% development rate calculated on the total number of oocytes or on the cleaved embryos, respectively. The data reported in Table 4 refer to repeated OPU collections from donors of various breeds within a commercial OPU programme. The average numbers of follicles aspirated and oocytes collected were considerably greater than the values above but the average recovery, maturation and cleavage rates were similar to what was reported for the experimental donors. Development to blastocyst, conducted completely in vitro in this case, was very variable with a total of 12 embryos produced from only 5 of 13 donors. There are, however, important differences between these two data sets that create some difficulties in making a direct comparison. Firstly, there were only five donors in the experimental group which were all young females, and four of which were Haflingers, a breed well known for its fertility. The commercial group was much more heterogeneous with a large range of ages and breeds, and most of them were referred to us because of fertility problems. In addition, both in vivo and in vitro culture systems were used for the experimental group of donors while only in vitro culture was used for the commercial mares, and finally the conditions for in vitro maturation were also different. These changes in the laboratory procedures have occurred over several years and the two data sets were generated in 2001 and in 2005, respectively. Over that period, the in vitro maturation conditions and subsequent embryo development in vitro were improved by changing the maturation medium from TCM199 to DMEM/F12, the first used in the 2001 study and the second in In vitro maturation conditions dramatically influence the development of fertilized oocytes to the blastocyst stage and different media are used for maturation of equine oocytes, although the most common one remains medium TCM199, probably because of its widespread and wellknown use for bovine oocyte maturation. In the last 2 years, we have tested other media including DMEM-F12 supplemented with 15% serum replacement (Invitrogen) instead of TCM 199 supplemented with 10% foetal calf serum. Hormones and growth factors were added to both media as previously reported (Lagutina et al., 2005). As shown in Table 5, the experiment included a large number of oocytes and was done in six replicates. We found a clear advantage in the use of a DMEM-F12 based medium not only on cleavage rate but also on embryo development to the blastocyst stage. The same stallion was used throughout the study, therefore, a male effect on cleavage rate can be ruled out. The greater cleavage rate is most likely due to more efficient oocyte maturation. This positive effect is evident also on blastocyst development, which was significantly higher in the DMEM-F12 group than in the TCM199 group. In our previous experiments comparing in vivo embryo culture in the sheep oviduct with in vitro culture (see Table 6) we have shown that the in vivo environment supports almost three times higher blastocyst development of in vitro matured, ICSI injected equine oocytes than does in vitro culture. This large difference in embryo development justified the use of in vivo culture for getting the best results from OPU donors. However, the sheep oviduct culture method is technically demanding, time consuming and expensive and therefore less suitable for routine use. With the new oocyte maturation medium, based on DMEM-F12, we found that the embryo development rate in vitro is more similar to the in vivo culture results, being in the range of 26% compared to 27 50% for the sheep oviduct system (Tables 3 and 6). This remarkable improvement increases the efficiency of equine embryo production and makes it possible to use a totally in vitro system following OPU in the horse. This again emphasises the importance of the developmental competence acquired by the oocytes during in vitro maturation. However, the remaining difference in embryo development between the in vitro and in vivo culture environments indicates the need for further improvement of the embryo culture system.

9 Table 4 Results of a commercial equine OPU programme at LTR Donor Age No. of OPUs No. of follicles No. of oocytes Recovery rate No. of MII injected % MII No. of cleaved No. of blastocysts % Blast/injected , Total Per OPU % C. Galli et al. / Animal Reproduction Science 98 (2007)

10 48 C. Galli et al. / Animal Reproduction Science 98 (2007) Table 5 Effect of maturation media on maturation, cleavage rate and ICSI embryos development Maturation medium No. of oocytes No. of degenerated No. of metaphase II (%) No. injected No. of cleaved (% of injected) No. of blastocysts (% of injected) TCM (47.2%) a (58.1%) a 23 (12.0%) a DMEM/F (45.6%) a (77.1%) b 37 (26.4%) b Replicates = 6, χ 2 -test: numbers within columns with different letters (a and b) differ (p < 0.05). 5. Cloning The first equids cloned by somatic cell nuclear transfer, three mules and one horse, were reported in Three mules were produced from nuclear transfer of somatic cells obtained from a fetus at 45 days gestation into in vivo matured oocytes recovered by OPU from gonadotropinstimulated preovulatory follicles (Woods et al., 2003). The recombined oocytes were surgically transferred to the oviducts of recipient mares immediately after activation. One horse filly was obtained by nuclear transfer of adult somatic cells, using in vitro matured oocytes as recipients of the donor nuclei and in vitro culture to the blastocyst stage before transcervical transfer to recipient mares (Galli et al., 2003). Remarkably, the latter was born to the same mare that donated the cells used in the cloning procedure, representing an exclusive example of autologous pregnancy successfully gone to term. More recently, two other cloned foals were born (Lagutina et al., 2005) from an endurance champion by SCNT, demonstrating the applicability of this technology to the rescue of gelding genotypes with exceptional sporting performances but obviously unable to produce offspring. Additionally, two cloned foals were born in 2005 in another laboratory confirming that the cloning technology is now established in equids (Hinrichs et al., 2006) and it can be done by using in vitro matured oocytes. The success of somatic cell nuclear transfer is the result of the optimisation of many steps described above. We have also tested the effect of changing maturation medium from TCM199 to DMEM-F12 on nuclear transfer (NT) embryo development (Table 7) and we have confirmed the positive results obtained with ICSI embryos (Table 5). Another step, essential for nuclear transfer success, is the activation of the reconstructed embryo. This step was performed in our laboratory by using a combination of the two most common chemicals used in other species for induction of parthenogenetic development: 6-dimethylaminopurine (6- DMAP) and cycloheximide. We found that either of the two used alone, after ionomycin exposure, was not providing satisfactory results as shown in Table 8. However, by using a combination of the two the activation rate improved to over 90%. This method was used to obtain the three cloned horses born in our laboratory. Other workers (Hinrichs et al., 2006) have used a combination of injection of sperm extract and culture in 6-DMAP to produce embryos resulting in successful foaling of cloned offspring. The group that reported the birth of the three mules from fetal cells in 2003 used in vivo matured oocytes for nuclear transfer and transferred them immediately after embryo reconstruction into the oviducts of synchronised recipient mares via standing flank laparotomy. This method requires the use of large number of experimental animals as donors of in vivo matured oocytes (only one oocyte potentially available from each mare per oestrus cycle), it is extremely expensive and finds little acceptance on animal welfare grounds. No other reports of this kind in fact exist for cloning equids. The other two groups that produced cloned offspring relied on the use of in vitro matured oocytes, which is the only realistic option for cloning large animals. Development rate of SCNT embryos is greatly influenced by the cell line (Lagutina et al., 2005), which is well known in

11 Table 6 Oocyte recovery rate by OPU and effect of in vitro or in vivo sheep oviduct culture on embryos development (Galli and Lazzari, 2001) No. of OPUs No. of oocytes (no. per OPU) No. metaphase II (% of oocytes) (no. per OPU) No. cleaved (% of injected) (no. per OPU) No. of comp. morulae/blastocysts (% of injected) (no. per OPU) Sheep oviduct culture (3.0) 46 (76.7%) (2.3) 41 (89.1%) a (2.1) 23 (50.0%) a (1.2) In vitro culture (4.1) 36 (73.5%) (3.0) 25 (69.4%) a (2.1) 5 (13.9%) b (0.4) Student s t-test: numbers within columns with different letters (a and b) are significantly different (p < 0.05). C. Galli et al. / Animal Reproduction Science 98 (2007)

12 50 C. Galli et al. / Animal Reproduction Science 98 (2007) Table 7 Effect of maturation media on maturation, cleavage rate and NT embryos development Maturation medium No. of oocytes No. lysated No. metaphase II (%) No. of NT embryos No. cleaved (% of NT) No. of blastocysts (% of NT) TCM (39.0%) a (92.7%) a 4 (9.77%) a DMEM/F (40.4%) a (97.9%) a 13 (27.7%) b χ 2 -test: numbers within columns with different letters (a and b) differ (p < 0.05). other species and varies from 0 to 17%. In a large data set obtained in our laboratory, the average blastocyst development rate was 4.4% (83 blastocysts from 1866 cleaved embryos out of 2099 reconstructed embryos; Galli et al., 2003; Lagutina et al., 2005). 6. Embryo transfer, pregnancies and offspring The only true measure of oocyte developmental competence is the ability of the oocyte, after fertilization and embryo culture, to generate viable offspring following the transfer of the resulting embryos to synchronised recipient mares even though a percentage of failures can also be attributed to the recipient itself. Therefore, one of our research priorities, given also its practical application, was to show that the embryos produced with the techniques described above had normal developmental competence and, therefore, validated the steps involved including in vitro maturation, ICSI, embryo culture and conventional cryopreservation in 10% glycerol. Non-surgical embryo transfer to recipients was performed over five breeding seasons ( ) in Cremona, Italy, northern hemisphere, from February to October. The frozen embryos were mainly produced outside the breeding season. Haflinger mares were examined two to three times a week by ultrasound to determine the day of ovulation. Three to seven (mostly 5) days after ovulation, each mare received 1 3 (up to four for nuclear transfer embryos) day 7 or day 8 blastocysts (fresh or after cryopreservation and thawing) by non-surgical transfer. On day 17 after ovulation the animals were examined (5 MHz linear probe, Sonovet 600, Medison, Korea) for pregnancy diagnosis and thereafter were scanned weekly throughout the first trimester of pregnancy and later at monthly intervals until foaling. Where required, all the experiments that involved the use of recipient animals were performed under veterinary supervision in accordance to Decreto Legislativo 116/92 that regulates the use of animal experimentation in Italy. Some of the experimental pregnancies were induced to abort at day 30 due to shortage of recipients. Overall, from 34 embryos produced by ICSI transferred into 21 recipients, we obtained 13 pregnancies at day 17 (Table 9), indicating a survival rate of 38%. The best result involved oocytes collected Table 8 Effect of cycloheximide and 6-DMAP alone or in combination on the activation of horse oocytes following ionomycin treatment 10 g/ml CHX 2mM DMAP 2.5 g/ml CHX mm DMAP 5 g/ml CHX+1mM DMAP 10 g/ml CHX+2mM DMAP No of oocytes treated No of oocytes activated % Activation 30.6% a 60.0% b 61.5% b 94.2% c 93.1% c χ 2 -test: values with different letters (a and b) differ (p < 0.05).

13 C. Galli et al. / Animal Reproduction Science 98 (2007) Table 9 Pregnancy and foaling rates following non-surgical transfer of equine embryos produced in vitro at LTR Type of embryo No. of embryos transferred No. of recipients No. of pregnancies (%) No. of vesicles No. of foals Abbattoir derived in vivo culture fresh transfer Abbattoir derived in vivo culture after cryopreservation Ovum Pick Up in vitro culture after cryopreservation Somatic cloning in vitro culture both fresh and frozen a (66.6%) b (25%) (72.7%) 9 c 5 d (28.2%) 13 3 a Two pregnancies were terminated on day 30. b Terminated on day 30. c One vesicle of a twin pregnancy was ablated. d Two pregnancies are still ongoing (over the sixth month). by OPU, with embryos cultured in vitro and cryopreserved, in which 9 embryonic vesicles were observed on day 17 out of 13 embryos transferred into 11 recipients. The pregnancy rate after transfer was 69% and the subsequent foaling rate was 83% of established pregnancies (5/6, two pregnancies are still ongoing). All but one of the OPU pregnancies were from the commercial OPU program. Despite the great variability of embryo production observed in the commercial group Fig. 1. The first foal derived from in vitro produced embryos obtained from in vitro matured oocytes collected by OPU, fertilized by ICSI, cultured in vitro to day 7, frozen conventionally in 10% glycerol and transferred non-surgically after thawing. The recipient received two thawed embryos; both developed to day 17 when the pregnancy was diagnosed by ultrasound. One vesicle was aspirated with an OPU needle and the second developed into a healthy foal born in 2002.

14 52 C. Galli et al. / Animal Reproduction Science 98 (2007) Fig. 2. The first foal derived from a commercial equine OPU service. The embryo derived from one in vitro matured oocyte, fertilised by in ICSI, cultured in vitro and frozen conventionally on day 7 in 10% glycerol. The recipient received one frozen-thawed embryo by non-surgical transfer and the foal was born in of OPU donors (see Table 4), the pregnancy rate has been very favorable especially considering that the embryos were cryopreserved. In the somatic cloning group we have transferred 118 embryos, fresh or frozen, into 46 recipients obtaining 13 pregnancies with one embryonic vesicle each, indicating a survival rate of 11% and a foaling rate of 23% (3/13) (Galli et al., 2003; Lagutina et al., 2005). No difference was observed between fresh and frozen embryos. No particular difficulties were encountered at foaling and some of the foals were the first in the world to be born as a result of combining in vitro oocyte maturation, ICSI, embryo culture in vitro and cryopreservation (Figs. 1 and 2). 7. Conclusions In this paper, we have summarized most of the research and applied results concerning assisted reproduction in equids. Oocyte developmental competence is clearly the key requirement for all the technologies that have been described. In vivo matured oocytes represent the gold standard for oocyte competence but they are in extremely short supply because equids are mono-ovulating species in which superovulation is problematic. Oocyte transfer, briefly described here, relies on this gold standard and provides very high results and pregnancy rates. Also, the very first equids (mules) derived from somatic cloning were obtained by nuclear transfer into in vivo matured oocytes. However, the paucity of available in vivo matured oocytes and the requirement for large numbers of oocyte donors, together with the animal welfare and economic issues involved, have provided a strong rationale for development of in vitro procedures for oocyte maturation. Nevertheless, the true assessment of developmental capacity of in vitro matured oocytes still

15 C. Galli et al. / Animal Reproduction Science 98 (2007) required their transfer into inseminated recipients, until recently, when ICSI technology using the piezo drill really opened the way for in vitro embryo production in the horse, making it possible to assess the developmental competence of both oocytes and embryos. ICSI has allowed testing of the relationship between in vivo and in vitro fertility of stallions, demonstrating that even semen with poor fertility in the field can be successfully used for embryo production in vitro. A satisfactory estimate of oocyte competence is provided by the ability to develop into blastocysts and then by the ability of these blastocysts to establish a pregnancy capable of developing to term. Blastocyst production rate is influenced not only by culture environment, in vivo as compared to in vitro, but also by oocyte maturation conditions. The change of maturation medium from conventional TCM199 to DMEM-F12 has a pronounced effect on blastocyst rate in our laboratory, suggesting that most probably there is still room for improvement of in vitro maturation conditions of equine oocytes. The culture environment, in vivo and in vitro, has been extensively tested and although the in vivo methods, either the mare oviduct or the surrogate sheep oviduct, have proven superior, the in vitro culture systems have improved considerably allowing greater rates of blastocyst development. A common feature of all the in vitro culture systems used is that the embryos tend to be retarded with fewer cell numbers than expected. This finding has suggested that we should allow some time for the day 7 or 8 IVP embryos to catch up with normal development by transferring them into recipient mares 5 days after ovulation instead of 7 days. Interestingly, if retarded embryos are kept in culture up to days 8 or 9 they can still produce a reasonable number of cells, while this is not the case with bovine, sheep and pigs in which species the inner cell mass of the blastocyst quickly undergoes pycnosis and degenerates. Pregnancy rates, following non-surgical transfer of ICSI embryos, have varied between 25 and 73%, the latter obtained with OPU cryopreserved embryos. Similarly, cloned embryos produced from in vitro matured oocytes were able to establish pregnancies after in vitro culture to the blastocyst stage followed by non-surgical transfer. Finally, in this paper we have provided extensive evidence that in vitro matured horse oocytes can develop to the blastocyst stage and develop into offspring, the real test of developmental competence. All this work opens the way to the successful application of assisted reproduction technologies in the horse. Acknowledgements The authors are grateful to Gabriella Crotti and Paola Turini for excellent technical assistance and to Gaetano Mari, Sandro Barbacini, Vittorio Marchi and Giovanni Zavaglia for helpful discussions. Part of this work was funded by MIUR (FIRB grant no. RBNE01HPMX and FIRS grant TECLA). References Alm, H., Torner, H., Blottner, S., Nurnberg, G., Kanitz, W., Effect of sperm cryopreservation and treatment with calcium ionophore or heparin on in vitro fertilization of horse oocytes. Theriogenology 56, Battut, I., Bezard, J., Palmer, E., Establishment of equine oviduct cell monolayers for co-culture with early equine embryos. J. Reprod. Fertil. Suppl. 44, Bezard, J., In vitro fertilization in the mare. In: Proceedings of the International Scientific Conference on Biotechnics in Horse Reproduction. Agricultural University of Cracow, Poland, Abstract 12. Bogh, I.B., Brink, P., Jensen, H.E., Lehn-Jensen, H., Greve, T., Ovarian function and morphology in the mare after multiple follicular punctures. Equine Vet. J. 35, Carnevale, E.M., Oocyte transfer and gamete intrafallopian transfer in the mare. Anim. Reprod. Sci. 82/83, , Review.

16 54 C. Galli et al. / Animal Reproduction Science 98 (2007) Carnevale, E.M., Ginther, O.J., Defective oocytes as a cause of subfertility in old mares. Biol. Reprod. Monogr. 1, Carnevale, E.M., Coutinho da Silva, M.A., Panzani, D., Stokes, J.E., Squires, E.L., Factors affecting the success of oocyte transfer in a clinical program for subfertile mares. Theriogenology 64, Choi, Y.H., Chung, Y.G., Seidel Jr., G.E., Squires, E.L., Developmental capacity of equine oocytes matured and cultured in equine trophoblast-conditioned media. Theriogenology 56, Choi, Y.H., Love, C.C., Love, L.B., Varner, D.D., Brinsko, S., Hinrichs, K., Developmental competence in vivo and in vitro of in vitro-matured equine oocytes fertilized by intracytoplasmic sperm injection with fresh or frozen-thawed spermatozoa. Reproduction 123, Choi, Y.H., Roasa, L.M., Love, C.C., Varner, D.D., Brinsko, S.P., Hinrichs, K., 2004a. Blastocyst formation rates in vivo and in vitro of in vitro-matured equine oocytes fertilized by intracytoplasmic sperm injection. Biol. Reprod. 70, Choi, Y.H., Love, L.B., Varner, D.D., Hinrichs, K., 2004b. Factors affecting developmental competence of equine oocytes after intracytoplasmic sperm injection. Reproduction 127, Choi, Y.H., Love, C.C., Varner, D.D., Hinrichs, K., 2006a. Equine blastocyst development after intracytoplasmic injection of sperm subjected to two freeze-thaw cycles. Theriogenology 6, Choi, Y.H., Love, L.B., Varner, D.D., Hinrichs, K., 2006b. Holding immature equine oocytes in the absence of meiotic inhibitors: effect on germinal vesicle chromatin and blastocyst development after intracytoplasmic sperm injection. Theriogenology 66, Cochran, R., Meintjes, M., Reggio, B., Hylan, D., Carter, J., Pinto, C., Paccamonti, D., Godke, R.A., Live foals produced from sperm-injected oocytes derived from pregnant mares. J. Equine Vet. Sci. 18, de Loos, F., van Vliet, C., van Maurik, P., Kruip, T.A., Morphology of immature bovine oocytes. Gamete Res. 24, Dell Aquila, M.E., Cho, Y.S., Minoia, P., Traina, V., Lacalandra, G.M., Maritato, F., Effects of follicular fluid supplementation of in-vitro maturation medium on the fertilization and development of equine oocytes after in vitro fertilization or intracytoplasmic sperm injection. Hum. Reprod. 12, Dell Aquila, M.E., De Felici, M., Massari, S., Maritato, F., Minoia, P., Effects of fetuin on zona pellucida hardening and fertilizability of equine oocytes matured in vitro. Biol. Reprod. 61, Eldridge-Panuska, W.D., di Brienza, V.C., Seidel Jr., G.E., Squires, E.L., Carnevale, E.M., Establishment of pregnancies after serial dilution or direct transfer by vitrified equine embryos. Theriogenology 63, Fulka, J., Okolski, A., Culture of horse oocytes in vitro. J. Reprod. Fertil. 61, Galli, C., Lazzari, G., In vitro and in vivo culture in the sheep oviduct of equine embryos obtained by IVM and ICSI. In: Stout, T.A.E., Wade, J.F. (Eds.), Proceedings of the Second Meeting of the European Equine Gamete Group. Loosdrecht, The Netherlands Newmarket: R&W Publications Abstract, pp Galli, C., Crotti, G., Duchi, R., Mari, G., Lazzari, G., Embryonic development of equine oocytes fertilised by ICSI. Havemeyer Foundation Monograph Series no. 3, pp In: Katila, T., Wade, J.F. (Eds.), Proceedings of the Fifth International Symposium on Equine Embryo Transfer. Finland, Saari Unit of Helsinki University, July 6 9, Newmarket: R&W Publications. Galli, C., Crotti, G., Turini, P., Duchi, R., Mari, G., Zavaglia, G., Duchamp, G., Daels, P., Lazzari, G., Frozen-thawed embryos produced by ovum pickup of immature oocytes and ICSI are capable to establish pregnancies in the horse. Theriogenology 58, Galli, C., Lagutina, I., Crotti, G., Colleoni, S., Turini, P., Ponderato, N., Duchi, R., Lazzari, G., Pregnancy: a cloned horse born to its dam twin. Nature 424, 635. Hawley, L.R., Enders, A.C., Hinrichs, K., Comparison of equine and bovine oocyte-cumulus morphology within the ovarian follicle. Biol. Reprod. Monogr. 1, Heape, W., On the artificial insemination of mares. Veterinarian 71, Hinrichs, K., Schmidt, A.L., Meiotic competence in horse oocytes: interactions among chromatin configuration, follicle size, cumulus morphology, and season. Biol. Reprod. 62, Hinrichs, K., Williams, K.A., Relationships among oocyte-cumulus morphology, follicular atresia, initial chromatin configuration, and oocyte meiotic competence in the horse. Biol. Reprod. 57, Hinrichs, K., Love, C.C., Brinsko, S.P., Choi, Y.H., Varner, D.D., In vitro fertilization of in vitro-matured equine oocytes: effect of maturation medium, duration of maturation, and sperm calcium ionophore treatment, and comparison with rates of fertilization in vivo after oviductal transfer. Biol. Reprod. 67, Hinrichs, K., Choi, Y.H., Love, L.B., Varner, D.D., 2005a. Transfer of in vitro produced equine embryos. In: Proceedings of the Sixth International Equine Embryo Transfer Symposium, pp

17 C. Galli et al. / Animal Reproduction Science 98 (2007) Hinrichs, K., Choi, Y.H., Love, L.B., Varner, D.D., Love, C.C., Walckenaer, B.E., 2005b. Chromatin configuration within the germinal vesicle of horse oocytes: changes post-mortem and relationship to meiotic and developmental competence. Biol. Reprod. 72, Hinrichs, K., Choi, Y.H., Love, C.C., Chung, Y.G., Varner, D.D., Production of horse foals via direct injection of roscovitine-treated donor cells and activation by injection of sperm extract. Reproduction 131, Lagutina, I., Lazzari, G., Duchi, R., Colleoni, S., Ponderato, N., Turini, P., Crotti, G., Galli, C., Somatic cell nuclear transfer in horses: effect of oocyte morphology, embryo reconstruction method and donor cell type. Reproduction 130, Lazzari, G., Crotti, G., Turini, P., Duchi, R., Mari, G., Zavaglia, G., Barbacini, S., Galli, C., Equine embryos at the compacted morula and blastocyst stage can be obtained by intracytoplasmic sperm injection (ICSI) of in vitro matured oocytes with frozen-thawed spermatozoa from semen of different fertilities. Theriogenology 58, Li, X., Morris, L.H., Allen, W.R., Influence of co-culture during maturation on the developmental potential of equine oocytes fertilized by intracytoplasmic sperm injection (ICSI). Reproduction 121, McKinnon, A.O., Lacham-Kaplan, O., Trounson, A.O., Pregnancies produced from fertile and infertile stallions by intracytoplamic sperm injection (ICSI) of single frozen/thawed spermatozoa into in vitro matured mare oocytes. In: Proceedings of the Seventh International Symposium on Equine Reproduction, July University of Pretoria, 137, abstract. Palmer, E., Bezard, J., Magistrini, M., Duchamp, G., In vitro fertilization in the horse. A retrospective study. J. Reprod. Fertil. Suppl. 44, Rosati, I., Berlinguer, F., Bogliolo, L., Leoni, G., Ledda, S., Naitana, S., The effect of co-culture on the development of in vitro matured equine oocytes after intracytoplastic sperm injection. Equine Vet. J. 34, Shabpareh, V., Squires, E.L., Seidel, G.E., Jasko, D.J., Methods for collecting and maturing equine oocytes in vitro. Theriogenology 40, Squires, E.L., Integration of future biotechnologies into the equine industry. Anim. Reprod. Sci. 89, , abstract. Squires, E.L., Wilson, J.M., Kato, H., Blaszczyk, A., A pregnancy after intracytoplasmic sperm injection into equine oocytes matured in vitro. Theriogenology 45, 306. Tremoleda, J.L., Stout, T.A., Lagutina, I., Lazzari, G., Bevers, M.M., Colenbrander, B., Galli, C., Effects of in vitro production on horse embryo morphology, cytoskeletal characteristics, and blastocyst capsule formation. Biol. Reprod. 69, Vanderwall, D.K., Hyde, K.J., Woods, G.L., Effect of repeated transvaginal ultrasound-guided follicle aspiration on fertility in mares. J. Am. Vet. Med. Assoc. 228, Willis, P., Caudle, A.B., Fayrer-Hosken, R.A., Equine oocyte in vitro maturation: influences of sera, time, and hormones. Mol. Reprod. Dev. 30, Woods, G.L., White, K.L., Vanderwall, D.K., Li, G.P., Aston, K.I., Bunch, T.D., Meerdo, L.N., Pate, B.J., A mule cloned from fetal cells by nuclear transfer. Science 301, Zhang, J.J., Boyle, M.S., Allen, W.R., Galli, C., Recent studies on in vivo fertilization of in vitro matured horse oocytes. Equine Vet. J. Suppl. 8,