Introduction
In the intervening five decades the ability to maintain the early mammalian embryo in culture has increased significantly, culminating in higher pregnancy rates post-transfer, but the reality of working in vitro is that embryos are exposed to several stresses they do not experience in vivo. A direct result of stress in vitro is compromised embryo physiology, gene expression and development (Gardner & Lane, Reference Gardner and Lane2005). Some researchers reported that the embryo development ex vivo is associated with developmental delay (Harlow & Quinn, Reference Harlow and Quinn1982; Li et al., Reference Li, Chandrakanthan, Chami and O'Neill2007) and that the culture system and the composition of the medium can affect embryo quality (Abe et al., Reference Abe, Yamashita, Satoh and Hoshi2002; Rizos et al., Reference Rizos, Ward, Duffy, Boland and Lonergan2002). In fact, while the innate quality of the oocyte is the major factor that determines the blastocyst yield, there is considerable evidence supporting the notion that the post-fertilization culture environment is critical in determining the quality of the blastocyst (Rizos et al., Reference Rizos, Ward, Duffy, Boland and Lonergan2002, Reference Rizos, Gutierrez-Adan, Perez-Garnelo, de la Fuente, Boland and Lonergan2003; Costa Pereira et al., Reference Costa Pereira, Nunes Dodeb and Rumpf2005; Lonergan, Reference Lonergan, Fair, Corcoran and Evans2006). Embryo development in vitro is influenced by a number of factors, such as co-culture with somatic cells (Donnay et al., Reference Donnay, Van Langendonckt, Auquier, Grisart, Vansteenbrugge and Massip1997), addition of anti-oxidants to the culture medium (Ali et al., Reference Ali, Bilodeau and Sirard2003), oxygen tension (Van Soom et al., Reference Van Soom, Yuan, Peelman, de Matos, Dewulf, Laevens and De Kruif2002) and the number of embryos present in the culture drop (Hendriksen et al., Reference Hendriksen, Bevers and Dieleman1999; Gardner & Lane, Reference Gardner, Lane, Trounson and Gardner2000; Vajta et al., Reference Vajta, Peura, Holm, Paldi, Greve, Trounson and Callessen2000; Costa Pereira et al., Reference Costa Pereira, Nunes Dodeb and Rumpf2005).
However, it should be noted that, with the appropriate media and the appropriate laboratory systems, it is possible in the mouse model to obtain cleavage rates similar to that in vivo (Gardner & Lane, Reference Gardner and Lane2003). Furthermore, in human medicine, it is actually possible to establish 80% pregnancy rates with a donor oocyte model, indicating that, across several mammalian species, culture conditions can now maintain high rates of embryo viability (Gardner, Reference Gardner2008).
In the horse, the development of assisted reproductive technologies is relatively slow compared with other domestic species. Recently, several studies have attempted to improve the in vitro culture of embryos (IVC) in horse but, unlike other mammalian species, as ruminants and pigs, its efficiency remains very low. A comparison between the published reports on in vivo culture of ICSI early cleavage stage embryos in the oviducts of mares (Choi et al., Reference Choi, Love, Varner and Hinrichs2006a) or temporary recipient sheep (Galli & Lazzari, Reference Galli, Lazzari, Stout and Wade2001; Galli et al., Reference Galli, Crotti, Duchi, Mari, Lazzari, Katila and Wade2000, Reference Galli, Crotti, Turini, Duchi, Mari, Zavaglia, Duchamp, Daels and Lazzari2002; Lazzari et al., Reference Lazzari, Crotti, Turini, Duchi, Mari, Zavaglia, Barbacini and Galli2002) and in vitro culture in various culture media, clearly demonstrated that the in vivo environment supports high blastocyst development, approximately 36% of injected oocytes in both the mare and sheep oviducts (Galli et al., Reference Galli, Colleoni, Duchi, Lagutina and Lazzari2007). In vitro, 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., Reference Choi, Chung, Seidel and Squires2001), DMEM-F12 and CZB (Choi et al., Reference Choi, Love, Varner and Hinrichs2006b) and modified SOF (Galli et al., Reference Galli, Crotti, Turini, Duchi, Mari, Zavaglia, Duchamp, Daels and Lazzari2002). Earlier work evaluated co-culture with somatic cells including Vero cells (Dell'Aquila et al., Reference Dell'Aquila, Cho, Minoia, Traina, Lacalandra and Maritato1997), oviduct epithelial cells (Battut et al., Reference Battut, Bezard and Palmer1991), cumulus cells (Li et al., Reference Li, Morris and Allen2001), granulosa cells (Rosati et al., Reference Rosati, Berlinguer, Bogliolo, Leoni, Ledda and Naitana2002) or culture in conditioned media (Choi et al., Reference Choi, Chung, Seidel and Squires2001) but in most of these systems the blastocyst rates remained low, ranging from 4 to 16% (Galli et al., Reference Galli, Colleoni, Duchi, Lagutina and Lazzari2007).
In some mammalian species, somatic cell co-cultures improved the in vitro development of embryos (White et al., Reference White, Hehnke, Rickords, Southern, Thompson and Wood1989; Rexroad & Powell, Reference Rexroad and Powell1993; Park et al., Reference Park, Han, Lee, Kim, Kim, Lee, Lee and Lee2000; Li et al., Reference Li, Morris and Allen2001) probably, by means of secreted embryo-trophic factors during co-culture. In this view, the co-culture with mouse embryonic fibroblasts improved the development of bovine and ovine embryos (Rexroad & Powell, Reference Rexroad and Powell1993; Park et al., Reference Park, Han, Lee, Kim, Kim, Lee, Lee and Lee2000). Embryonic fibroblast cells may secrete various factors that enhance embryonic development (Park et al., Reference Park, Han, Lee, Kim, Kim, Lee, Lee and Lee2000; Hatoya et al., Reference Hatoya, Sugiyama, Torii, Wijewardana, Kumagai, Sugiura, Kida, Kawate, Tamada, Sawada and Inaba2006).
The effect of co-culturing with amniotic epithelial stem cells (AE-SCs) or bone marrow mesenchymal stem cells (BM-MSCs), on IVC of horse embryos or other species, has not yet been studied. In the present study, we investigated the feasibility of using horse AE-SCs and BM-MSCs as feeder to support embryo development in vitro, employing the bovine embryo as a model to standardize a protocol for the culture of equine embryos.
Materials and methods
Chemicals, unless otherwise indicated, were purchased from Sigma Aldrich, Milano, Italy.
Isolation and culture of amniotic epithelial stem cells (AE-SCs)
From three samples of allantoamnion retrieved at delivery, each amniotic membrane was stripped from the overlying allantois and, for isolation of the epithelial cells, it was digested with 0.05% trypsin for 45 min at 37 °C. Finally, the cells were washed in high glucose Dulbecco's Modified Eagle's Medium (DMEM-HG, CELBIO Euroclone) plus 10% fetal calf serum (FCS), centrifuged at 300 g for 10 min and counted before plating.
Isolated cells were plated on 25 cm2 flask in DMEM-HG supplemented with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.25 μg/ml amphotericin B, 2 mM/l l-glutamine and 10 ng/ml of epidermal growth factor (EGF).
Amniotic stem cells were previously characterized as published in Lange Consiglio et al. (Reference Lange Consiglio, Corradetti, Bizzarro, Cassano and Cremonesi2010).
Isolation and culture of bone marrow mesenchymal stem cells (BM-MSCs)
Bone marrow samples were collected using sterile conditions from three horses under sedation (mean age of 7 years old).
Bone marrow aspirates were obtained from the sternum by using a Jamshidi biopsy needle (10 cm; 11G) into 12500 UI/ml heparin preloaded 60 ml syringes. The total bone marrow volume collected from each horse was about 30 ml. MSCs isolation from bone marrow consisted of layering over Histopaque™ 1.080, centrifuging for 20 min at 400 g at 4 °C, collecting the cell buffy coat, washing two times in phosphate-buffered saline (PBS) and plating mononucleated cells in DMEM-HG supplemented with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.25 μg/ml amphotericin B, 2 mM l-glutamine and 10 ng/ml of basic fibroblast growth factor (bFGF).
Bone marrow stem cells were previously characterized as published in Cremonesi et al. (Reference Cremonesi, Violini, Lange Consiglio, Ramelli, Ranzenigo and Mariani2008).
Isolation and culture of cumulus cells (control)
Cumulus cells remaining attached to the plastic culture dishes after in vitro maturation (IVM) of bovine cumulus–oocyte complexes were cultured at 38.5 °C, 5% CO2 in air until starting of the in vitro embryo culture, when the IVM medium was replaced with the embryo culture medium.
Preparation of feeder layers
The primary AE-SCs and BM-MSCs were cultured until confluent and thereafter, proliferated through two subsequent passages. These cells were then frozen at −80 °C using 10% of dimethylsulfoxide (DMSO). Frozen cells were thawed in water at 37 °C and re-seeded at a density of 1.0 × 104 cells per 100 μl droplet in IVC plates. The AE-SCs and BM-MSCs were plated 1 day before IVM and formed monolayer within 3 days of preparation, while the cumulus cells were ready 42 h after the beginning of the IVM.
In vitro production of bovine embryos
Cumulus–oocyte complexes (COCs) were collected from ovaries obtained from an abattoir by aspirating follicles 2–8 mm in diameter and washing them twice in preincubated (38.5 °C, 5% CO2 in air) TCM 199-HEPES buffered culture medium supplemented with 10% FCS.
IVM was performed for 24 h in TCM199 Earle's Salt medium supplemented with 10% FCS, 5 μg/ml luteinizing hormone (LH) (Lutropin, Bioniche), 0.5 μg/ml follicle stimulating hormone (FSH) (Folltropin, Bioniche), 0.2 mM sodium pyruvate, 10 μg/ml gentamycin and 1 mg/ml estradiol 17β. Cultures were in 100 μl droplets (up to 20 oocytes/droplet) of the medium under oil, at 38.5 °C in 5% CO2.
In vitro fertilization (IVF) was performed in TALP medium containing 2 mM penicillamine, 1 mM hypotaurine, 250 mM epinephrine and 20 μg/ml heparin. Frozen–thawed semen (106 spermatozoa/ml) was prepared by Percoll gradient (Amersham Pharmacia Biotec). In a 15 ml conic tube, 1 ml Percoll 90% was added followed by 1 ml Percoll 45%. Semen was thawed at 37 °C for 30 s, placed on the top of the Percoll gradient and centrifuged for 10–12 min at 650 g. After removal of the supernatant, 4 ml Tyrode's-albumin-lactate-pyruvate (TALP) medium were added and the sample centrifuged again for 2 min at 200 g to remove excess Percoll.
Semen was coincubated with matured oocytes for 18 h at 38.5 °C in 5% CO2. At the end of gametes co-culture, the cumulus cells were completely removed and cumulus-free presumptive zygotes were randomly transferred into one of three co-culture systems in which they were cultured up to day 7: (1) co-culture with cumulus cells (control); (2) co-culture with BM-MSCs; and (3) co-culture with AE-SCs. For all the feeders under experiments, the embryo culture medium was TCM199 Earle's Salt medium plus 10% FCS, 0.2 mM pyruvate and 10 μg/ml gentamycin at 38.5 °C in 5% CO2. Medium was changed every 48 h throughout the culture period.
Statistical analysis
The data were analyzed by ANOVA using GraphPad Instat 3.00 for Windows (GraphPad Software, Inc.).
Results and Discussion
Isolated cells readily attached to plastic culture dishes. AE-SCs displayed typical cuboidal morphology (Fig. 1a) while BM-SCs were fibroblast-like (Fig. 1b). A monolayer of cumulus cells developed progressively in the culture drop.
Figure 1 Monolayer of amniotic epithelial stem cells (a) and of bone marrow stem cells (b). Magnification ×10.
Seven hundred and five cumulus–oocyte complexes with a homogeneous cytoplasm and two or more layers of cumulus cells were used.
Data of three replicates are presented in Tables 1 and 2.
Table 1 Rate of different stage of development of blastocysts during culture on different feeders cells in three replicates
Blast. = Blastocysts; Exp = Expanded Blastocysts; Hatc = Hatched Blastocycts, Rep. = Replicates.
Table 2 Rate of cleavage and blastocysts on different feeders
a–cValues with different superscripts within same column are significantly (p < 0.001) different.
Rates of cleavage were similar but blastocysts developmental rates were significantly different (p < 0.001) between control, AE-SCs and BM-MSCs. The co-culture with AE-SCs gave a significantly higher percentage of blastocysts (p < 0.001).
Co-culture of oocytes and embryos with various types of cells has been widely used to improve development of bovine and ovine in vitro derived embryos (Rexroad & Powell, Reference Rexroad and Powell1993; Duszewska et al., Reference Duszewska, Reklewski, Pienkowski, Karasiewicz and Modlinski2000). It has been established that somatic cells, as cumulus cells, mainly have a beneficial effect on the development of bovine embryos because of the secretion of embryo-trophic factors (Donnay et al., Reference Donnay, Van Langendonckt, Auquier, Grisart, Vansteenbrugge and Massip1997; Geshi et al., Reference Geshi, Yonai, Sakaguchi and Nagai1999).
The influence of co-culture with fetal fibroblast cells was studied by Li et al. (Reference Li, Morris and Allen2001) on in vitro maturation of equine oocytes and by Hatoya et al. (Reference Hatoya, Sugiyama, Torii, Wijewardana, Kumagai, Sugiura, Kida, Kawate, Tamada, Sawada and Inaba2006) on IVM, IVF and IVC of canine oocytes. Signficantly higher proportions of blastocysts were produced from fetal fibroblast cells co-culture group than from the groups of control oocytes, demonstrating a beneficial influence of co-culture with fetal fibroblasts for nuclear and cytoplasmic maturation of oocytes in vitro.
These fibroblast cells secreted several cytokines, (e.g. leukemia inhibitory factor, steel factor and basic fibroblast growth factor (bFGF) (Dolci et al., Reference Dolci, Williams, Ernst, Resnick, Brannan, Lock, Lyman, Boswell and Donovan1991; Matsui et al., Reference Matsui, Toksoz, Nishikawa, Nishikawa, Williams, Zsebo and Hogan1991; Cheng et al., Reference Cheng, Gearing, White, Compton, Schooley and Donovan1994; Xie et al., Reference Xie, Lin, Luo, Luo and Lu2004), and growth factors that affected meiotic maturation of oocytes in vitro (Hatoya et al., Reference Hatoya, Sugiyama, Torii, Wijewardana, Kumagai, Sugiura, Kida, Kawate, Tamada, Sawada and Inaba2006).
In the present study, we examined the effects of co-culturing bovine embryos with horse AE-SCs and BM-MSCs on in vitro development. We have previously studied the characteristic of stemness of AE-SCs and BM-MSCs used in this work as feeders to culture embryos, and we have showed that these cells express mesenchymal stem cells mRNA markers (CD29, CD105, CD44), are negative for CD34, express embryonic marker as Oct-4, Sox 2 and SSEA-4 and are plastic because they are able to differentiate into osteogenic and adipogenic lines (Cremonesi et al., Reference Cremonesi, Violini, Lange Consiglio, Ramelli, Ranzenigo and Mariani2008, Reference Cremonesi, Maggio and Lange Consiglio2009; Lange Consiglio et al., Reference Lange Consiglio, Corradetti, Bizzarro, Cassano and Cremonesi2010). Moreover, AE-SCs are negative for the major histocompatibility complex (MHC) type II until fifth passages (data not shown).
Our results showed that the co-culture with AE-SCs gave a significantly higher percentage of blastocysts compared with control and BM-MSCs feeders. It's possible that AE-SCs monolayer create a more suitable microenvironment necessary for inducing local cell activation and proliferation of the growing embryos in comparison with BM-MSCs and cumulus cells used as control.
Koizumi et al. (Reference Koizumi, Inatomi and Sotozono2000) studied human amniotic epithelial cells and their gene expression and reported that they excrete or maintain on the membrane surface many proteins as growth factors [epidermal growth factor (EGF), keratinocyte growth factor, hepatocyte growth factor (HGF), basic fibroblast growth factor (bFGF), transforming growth factor-α (TGFα), TGFβ1, TGFβ2, TGFβ3] and two growth receptors (keratinocyte growth factor receptor and epatocyte growth factor receptor).
The mesenchymal stem cells by bone marrow or fat tissue express mRNA for interleukin (IL)-6, IL-11, leukemia inhibitory factor (LIF), macrophage colony-stimulating factor (M-CSF), and stem cell factor (SCF) (Majumdar, Reference Majumdar2000); other studies evidenced expression of vascular endothelial growth factor (VEGF), HGF, bFGF, granulocyte–macrophage colony stimulating factor (GM-CSF) and TGF-β (www.vet-stem.com/technologies/regenerativecells.php).
A growth factor that differentiates the AE-SCs from the BM-MSCs is EGF, that in vivo is produced by endometrial cells and for which the embryo possesses receptors. We cannot affirm that this factor is responsible for the different results of the in vitro culture of the bovine embryo, but certainly it is one of the greatest candidates to play a positive role of the monolayer of AE-SCs in embryonic development as it happens in vivo for the endometrium. Moreover, it can be suggested that these cells secrete biologically active substances, including signalling molecules and growth factors of epithelial nature, different to those of the BM cells of mesenchymal origin.
It is also important to underline that for embryonic development it is not necessary that these factors must be species-specific, considering the different affiliation of species of the equine amniotic epithelial stem cells in comparison with the bovine embryos, the object of this study.
In conclusion, this approach revealed that the complexity of the environment provided by this cellular feeder makes the AE-SCs particularly interesting for the in vitro co-culture of embryos and represents the initial step towards standardization of a protocol for the culture of equine embryos.
Acknowledgements
Regione Lombardia is acknowledged for ‘Dote Ricercatori’ fellowship to V.M.
