In vitro production of porcine embryos: current status, future perspectives and alternative applications

Nuclear maturation

For the onset of nuclear maturation, a hormone trigger is required in vivo (Edwards 1965). Thus, the addition of various hormones and growth factors in IVM medium has been performed and this has shown positive effects in meiotic progression: luteinizing hormone (LH) and follicle-stimulating hormone (FSH) (Mattioli et al. 1991) transforming growth factor and androstenedione (Singh et al. 1993), pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) (Funahashi & Day 1993; Funahashi et al. 1994), insulin like growth factor-I (IGF-I) (Illera et al. 1998), and estradiol-17 (Funahashi & Day 1993; Bing et al. 2001). Besides hormones, pFF was also found to increase the rates of nuclear maturation, which may be due to an acidic substance found in the follicular fluid (Rath et al. 1995).

Cumulus cells also play an important role in the control of nuclear maturation by: (i) maintaining the meiotic block at the germinal vesicle (GV) stage (reviewed in Tanghe et al. 2002); and (ii) triggering the resumption of meiosis by the secretion of a meiosis-inducing substance (Xia et al. 2000). In addition, cumulus cells protect oocytes against oxidative stress-induced apoptosis (Tatemoto et al. 2000). Cumulus cells also help to reduce DNA fragmentation in oocytes (Wongsrikeao et al. 2005).

To synchronize the progress of maturation, and thus improve the developmental competence of IVM oocytes, cell cycle-dependent kinase inhibitors such as butyrolactone-1 (Wu et al. 2002) and roscovitine (Romar & Funahashi 2005), protein-synthesis inhibitors such as cycloheximide (Ye et al. 2005), and dibutyryl cyclic adenosine monophosphate (cAMP) (Funahashi et al. 1997; Somfai et al. 2003) have been added into IVM media. These reversible inhibitors of meiotic resumption were expected to prevent ageing caused by premature meiotic resumption and to accumulate developmental factors in oocytes through communication with cumulus cells before their meiotic resumption. However, only cycloheximide and dibutyryl cAMP were found to improve the development of the oocytes to the blastocyst stage (Funahashi et al. 1997; Somfai et al. 2003; Ye et al. 2005).

Approximately 10–30% of oocytes fail to reach the M-II stage at the end of IVM. A small portion of them remain at the GV stage, whereas most of the incompetent pig oocytes permanently arrest at an immature stage that is most often characterized by metaphase chromosomes and the lack of the first polar body (1PB) extrusion (reviewed in Kikuchi et al. 2009). Although such a state is often referred as ‘metaphase-I (M-I) arrest’ previous study revealed that a high portion of these oocytes are at a diploid M-II stage (Sosnowski et al. 2003). Incomplete progress to the M-II stage may contribute to the formation of embryos with abnormal ploidy since the ‘M-I arrest’ oocytes were able to be fertilized, resulting in triploid embryos (Somfai et al. 2005). Nevertheless, the successful extrusion of the 1PB is not a guarantee for normal chromosome numbers in the resultant embryos since numerous oocytes with 1PB extrusion show aneuploidy (Sosnowski et al. 2003).

Cytoplasmic maturation

The main issue of porcine IVM is poor cytoplasmic maturation which covers all the cytoplasmic changes occurring during the process of meiosis. The cytoplasmic changes are necessary for the oocytes to acquire their future ability to be fertilized, activated and to develop to normal embryos. Thus, poor cytoplasmic maturation is thought to be partly responsible for low developmental competence of post-fertilization such as low MPN formation and blastocyst formation rates (Nagai et al. 1993a). Despite their normal nuclear progression in vitro, matured porcine oocytes show a reduced developmental ability compared with their in vivo-matured counterparts, suggesting their inadequate cytoplasmic abilities to support development (Laurincik et al. 1994).

Cytoplasmic maturation of pig oocytes is improved by reduction of oxidative stress caused by reactive oxygen species (ROS) production from stressed cumulus-oocyte complexes due to improper in vitro culture environment (Tatemoto et al. 2000, 2001). Metabolic pathways mediated by antioxidants such as GSH control cellular levels of ROS and protect the oocyte against the damaging effects of oxidative stress. GSH content of the oocyte can be increased by supplementation of thiol compounds such as cysteine (Yoshida et al. 1993), cysteamine, glutamine and β-mercaptoethanol and/or follicular fluid to the IVM medium (Jeong & Yang 2001). Besides, GSH also increases amino acid transport, and stimulates DNA and protein synthesis (Lafleur et al. 1994).

Various factors also improve cytoplasmic maturation when added into the maturation medium by increased MPN rate and/or blastocyst formation rate: LH (Mattioli et al. 1991); epidermal growth factor (EGF) (Illera et al. 1998; Abeydeera et al. 2000); follicular shells and cells (Ding & Foxcroft 1992; Abeydeera et al. 1998); and pFF (Naito et al. 1988; Rath et al. 1995; Vatzias & Hagen 1999). Moreover, supplementation of IVM medium with pFF was also found to support normal distribution of mitochondria during maturation, and this process was associated with improved developmental competence after parthenogenetic activation, suggesting the positive effect of pFF on the cytoplasmic microtubule network that controls mitochondrial reorganization (Brevini et al. 2005).

The importance of cumulus cells for oocyte cytoplasmic maturation has also been extensively studied and confirmed in farm animals (reviewed in Nagai 2001). To date it has been clarified that cumulus cells are indispensable for the oocytes to acquire normal cytoplasmic maturation and developmental competence during IVM since they synthesize and transport of GSH into oocytes (Maedomari et al. 2007). Recent findings have indicated that FSH induces GSH synthesis in cumulus cells and oocytes, although porcine oocytes are able to synthesize some GSH without gap junction-mediated support from cumulus cells, at least in the second half of the maturation culture (Ozawa et al. 2010). Another study showed that the condition of cumulus cells during maturation alters meiotic spindle morphology, which is considered to be one of the indices of cytoplasmic maturation and affects the normal pronuclear formation after IVF (Ueno et al. 2005).

Fertilization conditions

In traditional porcine IVF systems, although sperm penetration takes place around 3 h after insemination and is completed by 6 h, the incidence of polyspermy increases as the co-culture duration is extended (Funahashi et al. 2000). Therefore, numerous laboratories incubate gametes for about 6 h (Abeydeera & Day 1997; Funahashi et al. 1997; Yoshioka et al. 2003). However, Kikuchi et al. (2006) found that the incubation with sperm for 6 h doubles the average number of sperm per oocyte in comparison with 3 h, which leads to a higher polyspermy, while MPN formation is not significantly improved. Furthermore, when shorter co-culture periods were applied, the penetration and polyspermy rates were not affected, and the overall efficiency of IVP of normal embryos was not improved (Gil et al. 2004). Since then, 3 h gamete incubation has been preferable. Nevertheless, the optimal duration and sperm concentration for successful penetration with low incidences of polyspermy may differ among samples from different sources (ejaculate or epididymal), preservation (fresh, stored or frozen/thawed), boar or even the lot (i.e. ejaculate) even when taken from the same boar. Therefore, it is recommended to set up the optimal concentration and duration of sperm co-incubation with oocytes specifically for each frozen lot when large amounts of frozen/stored sperm are available for IVF.

Another pending question is the necessity and importance of cumulus cells for the successful fertilization of porcine oocytes in vitro. In cattle, it has been clarified that removal of cumulus cells from oocytes greatly reduces the rate of fertilized oocytes in IVF systems, therefore it has been concluded that cumulus cells play an important role during fertilization (Cox et al. 1993). Similarly, early research has indicated the importance of follicle cells on fertilization and MPN formation in in vitro-matured porcine oocytes (Kikuchi et al. 1993). On the other hand, our recent results showed that the removal of cumulus cells before IVF does not reduce the penetration rate when using certain lots of frozen sperm, and in some cases, it even results in increased penetration associated with very high rates of polyspermy (data not shown). Performing IVF with denuded oocytes has an advantage for the selection of M-II oocytes based on the presence of 1PB (those with normal haploid chromosome numbers), which enables elevated efficacy of normal embryo production. However, the optimal sperm concentration and IVF duration when using denuded oocytes might be different from those for success using cumulus-oocyte complexes for the same sperm lot.

IVF medium

Fertilization medium of in vitro systems is designed to closely resemble in vivo systems. IVF is usually carried out in one of the following media: TCM199 (Nagai & Moor 1990; Yoshida et al. 1990; Funahashi et al. 2000); Brackett and Oliphant (BO) medium (Kikuchi et al. 1993; Wang et al. 1995); Krebs-Ringer bicarbonate medium (Naito et al. 1988); modified Tris-buffered medium (Abeydeera & Day 1997); pig fertilization medium (PigFM) (Suzuki et al. 2002); or porcine gamete medium (PGM) (Yoshioka et al. 2003). Fraser (1995) reported that mammalian spermatozoa require extracellular calcium for capacitation and maximal acrosomal exocytosis. In addition to calcium, bovine serum albumin (BSA) and caffeine are also important modulators of sperm penetration (Abeydeera & Day 1997). However, the authors reported that the mean number of sperm per oocyte increased significantly with the addition of BSA to the IVF medium. Caffeine has also been shown to promote the capacitation of boar spermatozoa and thus fertilization (Wang et al. 1991; Nagai et al. 1993b). Besides, incubating oocytes in a medium containing 10% or 30% oviductal fluid prior to fertilization increased the incidence of monospermy without decreasing sperm penetration rate (Kim et al. 1996). Additionally, a 2.5 h sperm incubation in oviduct culture prior to fertilization reduced the rate of polyspermy 40–50% (Nagai & Moor 1990). Co-culture of oocytes with oviductal epithelial cells also resulted in a higher percentage of monospermic oocytes (Kano et al. 1994).

Culture conditions

It is widely accepted that the optimal incubation atmosphere for IVC of porcine embryos is under 5% CO₂ and 5% O₂ (Berthelot & Terqui 1996; Kikuchi et al. 2002). According to Kikuchi et al. (2002), when porcine IVM-IVF oocytes were cultured in an IVC medium supplemented with pyruvate and lactate for the first 2 days and then in the medium containing glucose for a subsequent 4 days under 5% O₂, the proportion and quality of blastocysts were significantly improved. Recent results have clearly indicated that glucose in IVC medium for the first 2 days of culture is detrimental to the development of embryos by generating the formation of ROS which induces apoptosis (Karja et al. 2006). Detrimental effects of oxidative stress can be effectively reduced by the application of a low oxygen tension during embryo culture (Karja et al. 2004a) or by the applications of antioxidants in medium (Ozawa et al. 2006). Further, the timing of glucose administration seems also to be critical for blastocyst formation; change to a glucose-supplemented medium from a glucose-free medium at 58 h after fertilization significantly enhances the rate of blastocyst formation compared which the glucose addition at 48 h after IVF (Medvedev et al. 2004; Karja et al. 2004b).

Culture medium

Whereas the optimal IVC condition is widely set under the atmosphere of 5% CO₂ and 5% O₂, a variety of media have been used for IVC of porcine embryos. However, almost all of the media contain various salts at low concentrations (Beckmann & Day 1993), energy sources (Kikuchi et al. 2002) and macromolecules (Petters & Wells 1993). The medium most often used, NCSU-23, was first reported by Petters and Wells (1993) to sustain viable porcine embryos. BSA has been shown to support the development of one-cell porcine embryos (Menino & Wright 1982) to the blastocyst stage (Dobrinsky et al. 1996). A combination of NCSU-23 medium supplemented with BSA has been shown to yield significantly higher percentages of blastocysts than other media (Rath et al. 1995; Long et al. 1999). The co-culture of porcine embryos with oviductal epithelial cells during the early stage of IVC also has a significant effect on the cell numbers of the blastocysts (Kikuchi et al. 2002). Recently, Yoshioka et al. (2008) demonstrated that porcine oocytes can mature, be fertilized and develop to the blastocyst stage in chemically defined, protein-free media and that the IVP blastocysts are developmentally competent to full term after ET.

Ensuring embryo quality

For the success of porcine IVP it is elemental to produce embryos with normality. The normality of porcine IVP embryos is highly impaired by either: (i) high incidence of abnormal chromosome numbers caused by polyspermic fertilization or the fertilization of oocytes arrested at a diploid stage (reviewed in Kikuchi et al. 2009); or (ii) the imperfection of the culture system to maintain or support developmental competence of embryos.

To eliminate embryos with abnormal chromosome numbers it is important to avoid the fertilization of oocytes arrested at a diploid stage to reduce the generation of polyploid embryos. This can be achieved by the selection of M-II stage oocytes for IVF, which on the other hand implies the removal of cumulus cells for the visualization of 1PB. Since cumulus removal may affect the fertilization results, the specific setting of the IVF system for denuded oocytes is necessary by optimizing sperm concentrations and duration of IVF to achieve the best penetration results. Another approach can be the improvement of the IVM system to reduce the frequency of oocytes arrested at the diploid stage and thus to ensure the normal haploid status of the matured oocytes. However, little is known about the mechanism that lies behind the meiotic arrest during IVM. It has been reported that, in pigs, this malfunction is neither related to oocyte diameter nor the composition of basic media or donor age but seems to be affected by genotype (Lechniak et al. 2005, 2007). Further research is needed to clarify the exact mechanism and to avoid the factors which lead to this abnormality. Since most of the embryos with chromosomal abnormalities resulted from polyspermic penetration (reviewed in Kikuchi et al. 2009), it is important to make sure that embryos result from normal monospermic fertilization. Since in vitro-cultured polyspermic porcine embryos are known to be able to develop to the expanded blastocyst stage despite their abnormal chromosome numbers (most of them being polyploid or mixoploid) (Han et al. 1999; Somfai et al. 2008), the ability of the embryo to form blastocysts cannot be considered to be a reliable selection marker to guaranteed embryo quality. Monospermic porcine zygotes can be effectively selected based on the number of pronuclei after a centrifugation treatment (Somfai et al. 2008) or based on the early cleavages (Dang-Nguyen et al. 2011). Nevertheless, low penetration or monospermy rates greatly limit the numbers of such selected embryos. Therefore, the development of IVF systems with high frequencies of monospermic fertilization seems to be a much desirable way to solve the problem. There have been several approaches to reduce polyspermic fertilization in pigs, including the use of alternative vessels for insemination such as microchannels (Clark et al. 2005) or the use of oviduct fluid (Kim et al. 1996) or oviductal glycoproteins (Kouba et al. 2000; McCauley et al. 2003).

Further, a novel strategy to reduce polyspermy in pigs is ‘zona hardening’ in which the zona pellucida (ZP) is hardened by pre-treating oocytes with an amine-reactive cross-linker (Coy et al. 2008). By the pre-treatment of oocytes with the cross-linker, the incidence of monospermy after IVF increases five-fold, and the number of penetrated sperm per oocyte was reduced six times, and a 45% improvement for porcine IVF efficiency has been reported. Amine-reactive cross-linker produces the most number of bonds in the ZP amino acid sequence. This strategy is based on the fact that the fusion between a fertilizing sperm and the oocyte results in exocytosis of cortical granules (Tsai et al. 2010). Their contents harden the ZP to prevent penetration of excess spermatozoa. However, the frequency of polyspermy in pigs still remains at relatively high levels. Therefore, further research on polyspermic fertilization in pigs is still needed. Ensuring embryo quality in IVP systems also covers the improvement of their viability and developmental competence, which may be possible by further efforts to reduce the detrimental effects of the culture systems, such as by reducing oxidative stress or the development of new media specifically designed for pigs.

Chemically defined systems

Another alternative strategy for promotion of IVP system is to establish a chemically defined medium. Since a chemically defined medium eliminates undefined factors present in biological materials such as serum or serum albumin, application of a chemically defined medium to IVP of embryos has various great advantages, especially for studying the mechanism and effect of chemicals on embryonic development. Research on this would push the progress of optimizing culture medium. Efforts have been made to mature, fertilize and culture porcine oocytes/embryos without protein supplementations with various results. It can be concluded that, in pigs, blastocyst development and viable offspring can be achieved using defined media (Abeydeera et al. 2000; Yoshioka et al. 2003). However, in certain media, porcine oocytes/embryos produced by defined systems are still inferior in developmental competence compared with those that were matured/cultured in the presence of pFF or BSA (Brevini et al. 2005; Yoshioka et al. 2008). Further development of defined media will be possible in the future by the identification and application of certain serum- or pFF-derived factors that are responsible for improved cytoplasmic maturation and embryo development.

Gene banking

Since the artificial insemination of frozen semen has not been satisfactorily developed in pigs, IVP seems to be the only way to-date to generate live piglets using frozen sperm stored in gene banks. Successful IVF of porcine oocytes by frozen boar sperm was first achieved by Nagai et al. (1988). IVP technology involves some extension of embryo culture which – due to the imperfect media and stresses – cannot provide embryo development in the way it happens in vivo. Thus, to achieve successful pregnancies the early transfer of cleavage-stage embryos surgically in the oviduct have been favored (Kikuchi et al. 1999). Although live piglets are generated by the transfer of IVP embryos at the blastocyst stage (Marchal et al. 2001; Kikuchi et al. 2002; Yoshioka et al. 2003), satisfactory results were achieved only by the intrauterine transfer of numerous excellent quality blastocysts (Kikuchi et al. 2002).

Gene banking also includes the cryopreservation of IVP embryos. In general, vitrification of the blastocyst or morula-stage embryos is considered more favorable than traditional slow freezing for cryopreservation in pigs (Dobrinsky 2001). However, in most cases, transfer of cryopreserved IVP blastocysts failed to result in pregnancies. The first piglets from cryopreserved IVP embryos were generated by the removal of intracellular lipids by micromanipulation (Li et al. 2006; Nagashima et al. 2007). Recently, Somfai et al. (2009) reported the successful production of piglets from IVP embryos vitrified and transferred at the zygote stage without lipid removal, suggesting that cellular damages caused by cryopreservation could be recovered in IVP embryos under in vivo conditions. As an option in gene banking, cryopreservation of immature or mature oocytes before IVF is also required. Recently, Somfai et al. (2010) have succeeded in producing blastocysts in vitro using oocytes vitrified at the germinal vesicle stage, which may lead to successful piglet production in the future.

Production of embryonic stem cells

Since pigs are thought to be useful as biomedical models outside of agricultural areas, many efforts have been made to establish porcine embryonic stem (ES) cells from early embryos (Evans et al. 1990; Hall 2008; Telugu et al. 2010) mainly because of the biological similarities pigs share with human beings, making them suitable for therapeutic research. However, authentic porcine ES cells lines have not yet been established. Unlike mouse ES cells, both intrinsic and extrinsic factors likely contribute to the difficulties in domesticated ungulates (Telugu et al. 2010; our unpubl. data). Further, the inefficient IVP system in pigs is also thought to be partly responsible for these difficulties. In addition, the lack of information, such as morphology and surface markers also holds up the establishment of porcine ES cells. Recently, authentic ES cells have been captured from rat embryos by using a medium including small molecules that specifically inhibit : glycogen synthase kinase (GSK)3β and MEK (Buehr et al. 2008; Li et al. 2008). Although it is obscure whether the fundamental mechanism underlying ES cell physiology is conserved in mammals or is particular to rodents, these findings appear to advance stem cell research. From a different perspective, it has been reported that induced pluripotent stem (iPS) cells are generated from porcine somatic cells (Ezashi et al. 2009; Wu et al. 2009; West et al. 2010). As the porcine iPS cell properties are similar to those of ES cells, the data acquired from these studies would provide some important insights for derivation of porcine ES cells from embryos.

Sperm-mediated gene transfer

Recently, IVF and intra-cytoplasmic sperm injection (ICSI) technologies were successfully used for gene transfer in monkeys, cattle and pigs by binding exogenous DNA to spermatozoa prior to the fertilization/injection process (Chan et al. 2000; Shemesh et al. 2000; García-Vázquez et al. 2010). Nevertheless this technique is rather detrimental to sperm (Canovas et al. 2010) and therefore it is still mainly used with ICSI. Further improvement of this technique is needed for successful application with IVF.