Effect of cumulus cell coculture and oxygen tension on the in vitro developmental competence of bovine zygotes cultured singly

Available online at www.sciencedirect.com

Theriogenology 71 (2009) 729–738 www.theriojournal.com

Effect of cumulus cell coculture and oxygen tension on the in vitro developmental competence of bovine zygotes cultured singly I.G.F. Goovaerts a,*, J.L.M.R. Leroy a, A. Van Soom b, J.B.P. De Clercq a, S. Andries a, P.E.J. Bols a a

University of Antwerp, Faculty of Pharmaceutical, Biomedical and Veterinary Sciences, Laboratory of Veterinary Physiology, Universiteitsplein 1, Gebouw U, B-2610 Wilrijk, Belgium b Ghent University, Faculty of Veterinary Medicine, Department of Reproduction, Obstetrics, and Herd Health, Salisburylaan 133, B-9820, Belgium Received 24 April 2008; received in revised form 10 September 2008; accepted 11 September 2008

Abstract The customary practice in bovine in vitro embryo production (IVP) is to handle oocytes and embryos in groups; although there are several reasons for establishing an IVP system for individual embryos that allows for following a single oocyte from retrieval through development to the blastocyst stage. To date, reports of individual IVP are inconsistent, and in most cases, resulted in unsatisfactory blastocyst rates. The objective of this study was to develop an efficient system for routine in vitro culture of individual bovine embryos. Single culture of zygotes in 2 different culture volumes (20 and 500 mL) yielded less than 3% blastocysts in experiment 1. In an attempt to improve these results, cumulus cells were added to the culture medium in experiment 2, after which blastocyst rates increased from 2.9 to 21.8% (P < 0.05). The third experiment revealed that an atmospheric oxygen tension, which is commonly used with somatic cell coculture, was not beneficial during individual embryo-cumulus cell coculture, because it resulted in lower blastocyst rates (Odds ratio 0.57, P < 0.001) and in lower blastocyst cell numbers (P < 0.05), when compared to culture in 5% oxygen. Grouped vs. single culture and reduced oxygen tension did not have a significant effect on cleavage and hatching rates. In experiment 4, three different cumulus cell coculture conditions during individual culture were tested and compared with the cleavage, blastocyst and hatching rates, and cell number of group culture (73.2%, 36.4%, 66.7% and, 155.1  7.26, respectively). The outcome variables after individual embryo culture on a 5-day-old cumulus cell monolayer (74.1%, 38.2%, 71.9% and 133.4  9.16, respectively), and single culture in the presence of added cumulus cells (69.9%, 31.9%, 66.7% and 137.3  8.01, respectively) were not significantly different from those obtained after group culture (P < 0.05). Though, individual culture in a cumulus cell conditioned medium significantly reduced both the cleavage (59.0%) and blastocyst rates (6.3%). These results demonstrate that single culture of bovine zygotes can be fully sustained by coculture with cumulus cells in a low oxygen environment; implementation of these findings in our IVP system produced blastocysts comparable in quantity and quality to those obtained by group culture. These results were consistently achieved after acquiring experience and expertise in the handling of single zygotes. # 2009 Elsevier Inc. All rights reserved. Keywords: Bovine; Coculture; Cumulus cells; Individual in vitro embryo production; Oxygen tension

1. Introduction * Corresponding author. Tel.: +32 3 820 23 95; fax: +32 3 820 24 33. E-mail address: [email protected] (I.G.F. Goovaerts). 0093-691X/$ – see front matter # 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.theriogenology.2008.09.038

The production of in vitro bovine embryos (IVP) was established more than 20 years ago [1]. Because oocytes

730

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

are routinely cultured in groups, no standard protocols exist for individual IVP. In fact, most researchers report higher blastocyst rates and improved embryo quality following group, compared to single culture, because oocytes and embryos stimulate reciprocally during their in vitro development [2–7]. Several growth hormones are identified as possible embryotrophic factors, namely: insulin like growth factor I and II (IGF-I, IGFII), transforming growth factor a and b (TGF-a, TGFb), interferon t (IFN-t), epidermal growth factor (EGF), platelet-activated factor (PAF), and platelet derived growth factor (PDGF) [8–11]. Moreover, producing in vitro embryos in groups can reduce cost, as smaller amounts of media are required, and group manipulation of oocytes or embryos is less laborintensive. Despite these considerations, an in vitro production system in which a single oocyte can be followed from the moment of retrieval through development to the blastocyst stage, can be extremely valuable and serve many scientific purposes, such as: investigating the processes of folliculo- and oogenesis, supporting studies on non-invasive oocyte quality parameters, and studying oocyte/embryo metabolism and gene-expression [2,12]. Individual embryo production also provides a means of avoiding the conglomeration of oocytes when the zona pellucida is removed [13], and of culturing oocytes and embryos resulting from labor-intensive manipulative methods like nuclear transfer [14]. From a practical point of view, an individual oocyte culture system would be very useful when working with endangered species or other animals of high genetic value [2,4], and/or when OPU-sessions, or other retrieval methods yield only one or a few cumulus oocyte complexes [15,16]. Finally, an individual bovine in vitro embryo production system is a promising model for studying human infertility [17], particularly because in human protocols oocytes and embryos are generally cultured singly. To obtain the maximum number of good quality embryos in vitro, mimicking the in vivo environment is ideal. In utero, the zygote is surrounded by less than 1 mL of fluid [18], while in vitro, group culture is performed both in large medium volumes and in small droplets under oil. The current volumes used for individual embryo culture vary between 1 and 500 mL, and no consensus for the ideal volume has been reached [2,19,20]. Not only the microenvironment is critical in vivo, but also the contact with somatic cells, and mimicking this condition was necessary for the successful production of the first in vitro blastocysts [21]. Although, nowadays somatic cells are generally no longer used for bovine

group culture; however, the development of individual embryos, impaired by the absence of surrounding embryos, can benefit from coculture with somatic cells [22]. Coculture with somatic cells can be tested in different systems, such as: a several days old monolayer of feeder cells, somatic cells added on the first day of embryo culture, or a conditioned medium, in which somatic cells were cultured previously [22,23]. The presence of somatic cells often influences the choice of atmosphere under which to culture embryos. In the mammalian oviduct and uterus, the oxygen tension is maintained between 1.5 and 9% oxygen [24]. In cell free embryo culture systems in vitro, 5% oxygen was demonstrated to be optimal [25]. However, when somatic cells are used in coculture with oocytes or embryos, 20% oxygen is customarily used [26–28]. Thus far, it is not clear if the addition of cumulus cells can enhance the developmental competence of individually cultured bovine zygotes, or if their development is influenced by oxygen tension. Moreover, the optimal culture conditions for the cumulus cells used to enhance single bovine embryo culture have not been defined. The aims of the present study were to develop a practical, effective, standard single bovine oocyte culture system, therefore, our goals were: (1) to evaluate the effect of culture medium volume on the developmental capacity of bovine embryos cultured individually and in groups; (2) to study which effect adding cumulus cells to the culture medium has on the developmental capacity of embryos cultured individually and in groups; (3) to compare 5 and 20% oxygen atmosphere, in the presence of cumulus cells, on the developmental capacity and quality of bovine embryos cultured individually and in groups; and (4) to study the developmental capacity and embryo quality of individual zygotes cocultured with cumulus cells comparing three different coculture methods. 2. Materials and methods All chemicals were purchased from Sigma1 (Bornem, Belgium), unless otherwise stated. 2.1. Collection of cumulus oocyte complexes Ovaries were collected at nearby slaughterhouses as soon after slaughter as possible, and transported immediately to the laboratory; they were then washed 3 times in warm saline solution supplemented with 5% kanamycin. Subsequently, all follicles with a diameter between 2 and 6 mm were punctured with an 18G needle affixed to a 10-mL syringe. The collected follicular fluid

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

was brought in a tube filled with 2 mL of Hepes-TALP, a salt solution with 114 mM NaCl, 3.1 mM KCl, 0.3 mM Na2HPO4, 2.1 mM CaCl2
2H2O, 0.4 mM MgCl2
6H2O, 2 mM bicarbonate, 1 mM pyruvate, 36 mM lactate, 1 M Hepes, 2 mL/mL phenol red, 0.4 mg/mL BSA and 50 mg/ mL gentamycin. Following precipitation, the supernatant was removed, and the pellet was placed in a Petri dish and resuspended in Hepes-TALP. Cumulus oocyte complexes (COCs) were collected and washed 2 times in HepesTALP. Only unexpanded COCs with homogeneous cytoplasm surrounded by five or more cumulus cell layers (quality grade I) were selected for maturation. 2.2. In vitro embryo production Maturation medium consisted of TCM199 supplemented with 20% FCS, 0.2 mM glutamine, 0.2 mM sodium pyruvate, 100 mM cysteamine and 50 mg/mL gentamycin. The COCs were first washed in 500 mL maturation medium and then a standard maturation protocol was followed with groups of 100 COCs in 500 mL maturation medium in 4-well plates (Nunc1) for 22–24 h in humidified air with 5% CO2 at 38.5 8C. Plates with maturation media were prepared and were allowed to equilibration in the CO2 incubator for at least 2 h before adding COCs. For fertilization, semen of the same IVF tested bull was used throughout these experiments. The frozen sperm was thawed in a warm water bath for 1 min and then separated on a discontinuous Percoll gradient (90 and 45%, Amersham Biosciences, The Netherlands). Standard fertilization techniques were performed in the same oocyte groups as for maturation, in 500 mL fertilization medium, and at a concentration of 1  106 sperspermatozoa/mL for 18–22 h in humidified air with 5% CO2 at 38.5 8C. Fertilization medium consisted of 114 mM NaCl, 3.1 mM KCl, 0.3 mM Na2HPO4, 2.1 mM CaCl2
2H2O, 0.4 mM MgCl2
6H2O, 25 mM bicarbonate, 1 mM pyruvate, 36 mM lactate, 2 mL/mL phenol red, 6 mg/mL BSA, 50 mg/mL gentamycin and 10 mL/mL heparin. Following fertilization, cumulus cells were removed from the presumptive zygotes by vortexing (3 min) in 2 mL Hepes-TALP. For an in vitro culture (IVC) control, 25  4 presumptive zygotes were cultured together in 50 mL SOF with 5% FCS under mineral oil. The SOF medium consisted of 108 mM NaCl, 7.2 mM KCl, 1.2 mM KH2PO4, 0.8 mM MgSO4.7H2O, 0.6% v/ v sodium lactate, 25 mM NaHCO3, 0.0266 mM phenol red, 0.73 mM sodium pyruvate, 1.78 mM CaCl2.2H2O, 0.34 mM sodium tricitrate, 2.755 mM myoinositol, 3% v/v BME 50, 1% v/v MEM 100, 0.4 mM glutamine

731

and 50 mg/mL gentamycin. Routine IVC was performed in a modular incubator under an atmosphere of 90% N2, 5% O2 and 5% CO2 at 38.5 8C. 2.3. Experiment 1: The effect of droplet volume on individual and group culture In experiment 1, 863 presumptive zygotes divided into 3 replicates were randomly assigned to 4 culture treatments: (1) single zygote in 20 mL culture medium under oil, (2) single zygote in 500 mL without oil, (3) 25  4 zygotes in 50 mL under oil, and (4) 25  4 zygotes in 500 mL without oil, all without cumulus cells. When small droplets of medium were used, an oil overlay was necessary to avoid evaporation, therefore, two culture conditions were compared: large droplets where oil is not necessary, and small droplets under oil. This experiment was performed in 4-well plates (Nunc1), and the media were equilibrated in the CO2 incubator for at least 2 h before use. Cleavage and blastocysts rates were assessed 3 and 8 days after fertilization, respectively. 2.4. Experiment 2: The effect of cumulus cells addition on individual and group culture After routine maturation and fertilization, 4 different culture treatments were tested (4 replicates). In treatments 1 and 2, the presumptive zygotes were individually cultured in 20 mL of culture medium, while in treatments 3 and 4, group culture was performed (25  4 zygotes in 50 mL of culture medium). For this experiment, 24-well plates (Nunc1) were used. Before the 1028 presumptive zygotes were denuded by vortexing, they were partly stripped of cumulus cells by repeated pipetting in the fertilization medium, with a glass pipette. The collected cumulus cells were washed in culture medium and then added in small amounts (1 mL) to the culture medium droplets of treatments 2 and 4. Cleavage, blastocyst and hatching rates (on the total blastocyst number on day 10) were assessed 3, 8 and 10 days after fertilization, respectively. 2.5. Experiment 3: The effect of oxygen tension on individual coculture with cumulus cells, and group culture To test the effect of the oxygen tension in the gas phase, 1152 COCs were allocated to 4 treatments (6 replicates): (1) single zygote in 20 mL culture medium droplet under oil, supplemented with cumulus cells, cultured in 90% N2, 5% O2 and 5% CO2; (2) single

732

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

zygote in 20 mL culture medium droplet under oil, supplemented with cumulus cells, cultured in humidified air with 5% CO2; (3) 25  4 zygotes in 50 mL culture medium droplet under oil in 90% N2, 5% O2 and 5% CO2; and (4) 25  4 zygotes in 50 mL culture medium droplet under oil in humidified air with 5% CO2. Cleavage, blastocyst and hatching rates were assessed 2, 8 and 10 days after fertilization, respectively. Embryo cell number was assessed by Hoechst staining. Briefly, on average, 15 blastocysts (not hatched) per treatment were collected on day 8 after fertilization, and fixed in 4% formaldehyde for at least 24 h. Subsequently, they were stained with Hoechst 33342 stain at 10 mg/mL in PBS for 10 min. The stained embryos were fixed on a glass slide and evaluated under fluorescence microscopy. The nuclei of each blastocyst were counted at least twice and the average of the two counts was used for statistical analyses.

considered conditioned medium was incubated under oil. For the coculture treatment, cumulus cells were added to the culture medium as was done in experiments 2 and 3. For these 3 culture treatments and a control (90% N2, 5% O2 and 5% CO2 at 38.5 8C), 1033 presumptive zygotes were divided and randomly assigned to the 4 culture conditions (6 replicates). Cleavage, blastocyst and hatching rates were assessed 2, 8 and 10 days after fertilization, respectively. Blastocyst stages were also evaluated for each treatment at day 8, when hatched and expanded blastocysts were considered as fast blastocysts, and normal and early blastocysts as slow blastocysts. The cell numbers of day 8 blastocysts (hatched blastocysts included, on average 43 blastocysts per treatment) were determined with Hoechst staining as described above.

2.6. Experiment 4: The effect of cumulus cell coculture conditions on individual embryo development

All statistical procedures were carried out with SPSS 13.0 for Windows (Chicago, IL, USA). Values of P < 0.05 were considered statistically significant. Cleavage, blastocyst and hatching rates, as well as fast and slow blastocysts (experiment 4), were analyzed using a binary logistic regression model. In experiments 1 and 4, the effects of treatment, replicate, and the interaction of these two factors were assessed. When necessary (multiple dependent comparisons) P-values were corrected according to the Bonferroni test. Experiments 2 and 3 were a 2  2 design. In experiment 2, group vs. single treatment, presence vs. absence of cumulus cells, replicate, and the interaction terms were included in the model. In experiment 3, group vs. single treatment, oxygen tension, replicate, and the interaction terms were included in the model. When the interaction term was not significant, it was excluded from the final model. Blastocyst cell numbers were analyzed using a mixed model ANOVA. No data transformations were necessary for inequality of variance between groups or for achieving normality. Treatment was inserted as a fixed factor and replicate as random factor together with the interaction term. In the absence of a significant interaction term, the term was excluded from the final model. Cell numbers are expressed as means  stanstandard error of the mean (SEM).

Previously, a monolayer of cumulus cells, derived from matured COCs from a separate batch of ovaries, was prepared, both for subsequent culture of zygotes on this monolayer as well as to obtain conditioned medium. To do so, standardly matured cumulus oocyte complexes were washed in Hepes-TALP and stripped with a glass pipette. The naked oocytes were removed and pipetting up and down through a small diameter pipette further dispersed the cumulus cells. Two microlitre of this cumulus suspension was mixed for 2 min with 2 mL of trypan blue and subsequently the concentration and the percentage of viable cumulus cells were assessed in a Bu¨rker counting chamber. For monolayer culture the suspension was diluted to 100 viable cumulus cells per mL in culture medium, and then 5 mL of this diluted suspension was added to 15 mL droplets of culture medium under oil in 24-well plates. The remainder of the cumulus suspension was divided and added to 500 mL of culture medium in a 4-well plate. The plates were cultured in humidified air with 5% CO2. One day after fertilization (1 day p.i.) of another batch of matured COCs, on the start of embryo culture, all plates were checked for cumulus cell proliferation, and without exception, in all wells a monolayer had developed. On this day the monolayer was 5 days old. For use in the conditioned media treatment, 200 mL of supernatant was taken from each well of the 4-well plates, and droplets of 20 mL were placed in the wells of 24-well plates and this, now

2.7. Statistical analyses

3. Results In all experiments, no interaction between replicate and treatment was found, so this interaction term was excluded from the model.

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

733

3.1. Experiment 1: The effect of droplet volume on individual and group culture The cleavage and blastocyst rates of experiment 1 are summarized in Fig. 1. No statistically significant difference could be found in cleavage rates between the 2 single treatments (in 20 mL: 126/231 or 54.5%; in 500 mL: 108/170 or 63.5%) or between the 2 group treatments (in 50 mL: 162/232 or 69.8%; in 500 mL: 143/230 or 62.2%). Group culture in small droplets resulted in significantly higher cleavage rates compared to single culture in small droplets. Group culture after 7 days (8 days p.i.) in 50 mL achieved equally blastocyst rates (57/232 or 24.6%) as in 500 mL (37/230 or 16.1%); however, both group treatments resulted in significantly higher blastocyst rates compared to individual culture conditions in either 20 or 500 mL, between which no significant difference was found (5/231 or 2.2% and 2/170 or 1.2%, respectively). 3.2. Experiment 2: The effect of cumulus cells addition on individual and group culture Fig. 2 shows the blastocyst and hatching rates for experiment 2. In this experiment no interaction between group vs. individual culture, and presence vs. absence of cumulus cells was found to effect cleavage rates. Individual culture decreased cleavage rates (1 - CC: 221/ 345 or 64.1% and 1 + CC: 188/307 or 61.2% vs. 25 - CC: 144/189 or 76.2% and 25 + CC: 137/187 or 73.3%; Odds ratio 0.57, P < 0.001), while the addition of cumulus cells had no effect. A significant interaction was found for the blastocyst and hatching rates. The four culture treatments were thus statistically tested apart. Coculture with cumulus cells was beneficial for blastocyst rates in single embryo culture (67/307 or 21.8% vs. 10/345 or

Fig. 1. Effect of culture droplet volume on cleavage and blastocyst rates after single (s20 + oil: individual culture in 20 mL droplets under oil; s500: individual culture in 500 mL) and group culture (g50 + oil: 25 zygotes in 50 mL droplets under oil; g500: 25 zygotes in 500 mL) in 5% O2 without cumulus cells. a,b Different superscripts above the same bar type indicate a statistical difference between treatments (P < 0.05).

Fig. 2. Effect of adding cumulus cells (CC) to individual (1 - CC and 1 + CC) and group culture (25 - CC and 25 + CC) in 5% O2 on blastocyst and hatching rates. a,b Different superscripts above the same bar type indicate a statistical difference between treatments (P < 0.05).

2.9%; P < 0.001), but not in group culture (46/187 or 24.6% and 58/189 or 30.7%). More embryos hatched when cultured in a group without cumulus cells (50/62 or 80.6%) than when cultured singly in the presence of cumulus cells (35/77 or 45.5%). The hatching rate after individual culture without cumulus cells (1/12 or 8.3%) was not significantly different from the hatching rate of single culture with cumulus cells because of the low number of blastocysts. Without cumulus cells, 60.7% (34/56) of the group cultured blastocysts formed on day 10, were hatched. 3.3. Experiment 3: The effect of oxygen tension on individual coculture with cumulus cells, and group culture The results of testing the effect of oxygen tension on embryo development (experiment 3) are summarized in Tables 1a and 1b. The interaction between the effect of oxygen tension and the effect of group vs. individual culture was never significant. Blastocyst rates were significantly lower (Odds ratio 0.57, P < 0.001) when 20% oxygen was used, compared to 5%. Culturing individually also significantly decreased blastocyst rates compared to group culture (Odds ratio 0.56, P < 0.001). Group or individual culture had no significant effect on blastocyst cell number, but culturing in high oxygen tension significantly decreased cell numbers, compared to culture in 5% oxygen. Cleavage and hatching rates were not significantly different between treatments. 3.4. Experiment 4: The effect of cumulus cell coculture conditions on individual embryo development In experiment 4, three treatments for individually cultured embryos (monolayer, adding cumulus cells, and conditioned medium) were compared to the group

734

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

Table 1a Cleavage, blastocyst and hatching rates (on total blastocyst number on day 10) and cell numbers after culture in 5 and 20% oxygen during individual culture with cumulus cells (1 + CC) and in a group (25, without cumulus cells). Culture treatment

Number of oocytes

Cleavage (%)

Blastocysts (%)

Cell number D8  SEM

% Hatched

1 + CC in 5% O2 1 + CC in 20% O2 25 in 5% O2 25 in 20% O2

324 311 259 258

216 226 189 180

77 48 94 62

106.9  9.67 84.8  13.76 116.9  8.36 110.2  6.30

62.5 57.1 69.7 72.5

(66.7) (72.7) (73.0) (69.8)

(23.8) (15.4) (36.3) (24.0)

Table 1b The effect of culture in 5 and 20% oxygen during individual culture with cumulus cells (1 + CC) and in a group (25, without cumulus cells) on cleavage, blastocyst and hatching rates (on total blastocyst number on day 10) and cell numbers. Culture treatment

Cleavage (%)

Blastocysts (%)

Odds ratio

P-value

Odds ratio

P-value

Group/single Group Single

Ref 0.93

– 0.60

Ref 0.56

– <0.001

Oxygen tension 5% oxygen 20% oxygen

Ref 1.09

– 0.52

Ref 0.57

– <0.001

Cell number D8

% Hatched

F-value

P-value

Odds ratio

P-value

3.20

0.08 Ref 0.67

– 0.20

Ref 0.90

– 0.75

4.70

culture control (Table 2). Culturing individual zygotes in a cell free cumulus cell conditioned medium significantly reduced the cleavage rate (59.0%) and the blastocyst rate (6.3%) compared to group culture (cleavage rate: 73.2%; blastocyst rate: 36.4%). By contrast, no differences in either cleavage or blastocyst rates could be found between individual culture on a monolayer (cleavage rate: 74.1%; blastocyst rate: 38.2%) and the grouped controls, or between individually cultured embryos with added cumulus cells (cleavage rate: 69.9%; blastocyst rate: 31.9%) and the control. Individual culture in conditioned medium also significantly reduced the percentage of fast developing blastocysts (hatched and expanded, as a percentage of total blastocysts) on day 8 following fertilization (14.3%), compared to the group culture (71.4%). The percentage of fast developing blastocysts did not differ significantly between the embryos cultured in

0.04

the individual monolayer treatment (66.7%), the treatment where cumulus cells were added to individual culture (73.6%), and the control. Because not enough blastocysts could be obtained using the conditioned medium treatment, it was excluded from the statistical analyses of cell numbers (Table 2). Cell numbers of blastocysts from the monolayer treatment (133.4  9.16) and the treatment with addition of cumulus cells (137.3  8.01) were not significantly different from the cell numbers of blastocysts in the group treatment (155.1  7.26). Moreover, hatching rates at day 10 p.i. were also not affected by these culture conditions. 4. Discussion The overall aim of this study was to establish a standard method to routinely culture bovine in vitro

Table 2 Effect of group culture (group contr), individual culture on an existing monolayer (1 on monol), with added cumulus cells (1 + cocult), or in conditioned medium (1 in condit), on cleavage, blastocyst and hatching rates, percentage of fast developed blastocysts on day 8 (hatched and expanded blastocysts), and on blastocyst cell numbers, after culture in 5% O2. Culture treatment

Number of oocytes

Cleavage (%)

Group contr 1 on monol 1 + cocult 1 in condit

365 220 226 222

267 163 158 131

a,b

(73.2)a (74.1)a (69.9)a (59.0)b

Blastocysts (%) 133 84 72 14

(36.4)a (38.2)a (31.9)a (6.3)b

% Fast blastocysts D8

Cell number D8  SEM

% Hatched D10

71.4a (95) 66.7a (56) 73.6a (53) 14.3b (2)

155.1  7.26a 133.4  9.16a 137.3  8.01a –

66.7a 71.9a 66.7a 42.9a

Different superscripts in the same column indicate a statistical difference (P < 0.05).

(26/39) (23/32) (16/24) (3/7)

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

embryos singly, as the first step in generating an individual embryo production system. Our results showed that with individual culture of zygotes in a cumulus cell coculture system at low oxygen tension, the number of good quality blastocysts obtained can be equal to that from routine group culture. In the first experiment, individual culture did not yield acceptable blastocyst rates, and was independent of the culture drop volume. Embryo development data in single culture experiments are inconsistent among different studies, and only a few researchers report acceptable blastocyst rates [2,13,29,30]. This shows that single embryo culture is far from efficient, even more so because blastocyst quality is frequently not satisfactory, indicated by low cell numbers [5,31], low hatching rates [2,19] and reduced tolerance to cryopreservation [5]. For in vitro culture, a small embryo:medium ratio is optimal [13], but individual culture in very small volumes (<10 mL) can compromise embryo development as well, probably due to the accumulation of toxic metabolites like ammonium [2]. In contrast, when large volumes of medium are used, or when media are replaced during culture, autocrine embryotrophic factors are diluted and/or washed out of the culture droplet [6]. The results of our first experiment indicate that group culture leads to higher blastocyst rates, even though the same ‘‘embryo density’’ as in individual culture (1 embryo/20 mL in treatments 1 and 4) was used and all zygotes originated from a uniform maturation and fertilization procedure. An increase in the observed 8- to 16-cell block, probably caused by the rapid depletion of reserves due to inadequate culture conditions [32], further indicates that individual culture led to suboptimal results. A somatic cell coculture system can be one solution to overcome the impaired maternal-embryonic transition period, indicated by the 8- to 16-cell block [22]. The use of autologous cumulus cells to support in vitro development offers several opportunities. Firstly, it is practical, because no oviductal cells need to be prepared, avoiding labor-intensive steps, and (difficult) cell lines need not be cultured. Secondly, one of the disadvantages of somatic cell coculture is countered, namely the risk of contamination [13]. Although a few researchers used cumulus cells in coculture during individual embryo culture [5,13,30], until now, their beneficial effect was untested. Our second experiment showed that adding cumulus cells to individually cultured zygotes indeed led to significantly higher blastocyst rates. Coculture with somatic cells remains controversial, and a hot topic for scientific debate [22]. On the one hand, the amount of non-defined factors in

735

the production system increases [13]. On the other, the somatic cells support embryo development and improve embryo quality because they secrete embryotrophic factors, like growth factors and cytokines, and may neutralize embryotoxic components [33–35]. Individual bovine embryo development has been facilitated by coculture with other somatic cells, such as bovine oviductal epithelial cells (BOEC) [36], and buffalo rat liver cells (BRL) [4]. Edwards et al. [37] showed that 48 h of incubation of these cells in TCM199 resulted in lower glucose, and higher lactate and pyruvate concentrations compared to the initial concentrations in TCM199, thereby creating a propitious environment for embryonic development. The beneficial effect of cumulus cell coculture might, at least partly, be attributed to a comparable beneficial alteration in the medium. It is generally accepted that low oxygen tension leads to increased blastocyst rates and improved embryo quality in the absence of somatic cells (for review: see [38]). However, generally 20% oxygen is maintained when somatic cells are utilized for coculture [26–28]. Our third experiment showed that the blastocyst rates at day 8 after fertilization were significantly lower in 20% oxygen, compared to 5%, in the presence of cumulus cells. Also, embryo quality was lower at high oxygen tension, as indicated by lower cell numbers. Lim et al. [39] found no effect of oxygen tension on blastocyst rates in a group culture study with cumulus cells. This study, as well as ours, shows a preferred oxygen tension that differs from other reports [26–28]. This is probably caused by the use of somatic cell types other than cumulus, and/or varying culture systems [23]. It was shown that BOEC and BRL cells, incubated for 48 h in TCM199, significantly reduce the oxygen tension in the medium [37]. Additionally, the presence of human OEC during mice zygote culture in high oxygen tension resulted in reduced superoxide anion levels, which are associated with oxidative stress, compared to culture in a cell free control medium [34]. It is not clear if cumulus cells act in the same way, they might have had a significant effect on oxygen tension in the medium, but our study showed that in the presence of bovine cumulus cells, a low oxygen atmosphere was preferable and improved individual developmental capacity. Thus far, for individual embryo culture with cumulus cells, use of 20% oxygen has been reported [3,5,30]. Despite the fact that acceptable blastocyst rates could be obtained by adding cumulus cells during individual culture (experiment 2), group culture still yielded improved rates. This is probably caused by the fact that cumulus cells obtained after fertilization might

736

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

be too mature, and in addition, spermatozoa were added simultaneously with these cumulus cells. Dead sperm cells in the culture medium might cause higher ROS concentrations through free amino oxidases, which catabolize spermine and spermidine to free radicals [40]. Therefore, we designed and carried out the fourth experiment, in which two other culture conditions were tested: (1) culture on a monolayer of matured cumulus cells, grown for 5 days, and (2) culture in a cumulus cell conditioned medium. Along with these 2 treatments, group and individual culture with added cumulus cells was tested. No statistical difference between individual culture in the presence of added cumulus cells and group culture could be found; this clearly indicates that individual culture can achieve results comparable to those of group culture. Moreover, culture on a monolayer resulted in as many blastocysts (38.2%) as did group culture (36.4%), and the embryo quality, as assessed by cell number, did not differ significantly. In contrast however, using a cumulus cell conditioned medium significantly reduced both the cleavage and blastocyst rates, and additionally, blastocoele formation was slower, compared to group culture. Comparable results were achieved with a monolayer of granulosa cells obtained from follicular fluid after removing the COCs; culture of single zygotes on this monolayer, in CR2 medium, increased blastocyst rates from 0.5 to 12.4% [4]. This tendency could also be observed in embryo groups cultured in a protein-free medium, e.g., embryo development was enhanced by a granulosa cell conditioned medium, although coculture resulted in even higher blastocyst rates [41]. Many studies have used a mouse model, and a recent report [35] showed that a murine OEC conditioned medium enhanced the development of mouse embryos, in contrast to a medium conditioned with caprine OEC. The authors concluded that when heterologous somatic cells are used, immediate communication between the embryo and the feeder cells is demanded, while this is apparently not the case with homologous cells. This is in agreement with Joo et al. [34], who demonstrated that human cell conditioned medium did not enhance murine embryo development. Yet, researchers disagree on the need for direct contact between the feeder cells and the embryos [34,35]. Our study showed that homologous bovine cumulus cell conditioned medium did not effectively enhance individual bovine embryo development. Expertise in handling single zygotes influences the blastocyst development. The same protocol for adding cumulus cells to the individual embryo culture medium was used during 3 experiments, while the subsequent

blastocyst rates increased from 21.8 and 23.8% in experiments 2 and 3, to 31.9% in experiment 4. Moreover, in contrast to experiments 2 and 3, blastocyst rates achieved in the fourth experiment, equalled those of the zygotes cultured in groups. This illustrates that reducing the exposure time to air and/or lower temperatures probably enhanced development, although warming plates were used at all times. As both, culture on an older cumulus cell monolayer, and adding cumulus cells on the first day of culture resulted in a high number of good quality embryos, the method of choice can depend on the aim of the study. In the monolayer treatment, each zygote is cultured under the same conditions, while, when adding cumulus cells, autologous cumulus cells can help sustain each zygote throughout all the separate steps of IVP. Thus, the developmental capacity of the complete cumulus oocyte complex will be tested for the entire duration of the IVP protocol. In conclusion, we found that the developmental competence, as well as embryo quality of individually cultured zygotes, benefits from the presence of cumulus cells. This demonstrable benefit can be accomplished either by adding cumulus cells on the first day of individual embryo culture, or by culturing single zygotes on a 5-day-old cumulus cell monolayer. Additionally, more blastocysts with a higher cell number can be obtained when zygotes are cultured in an oxygen tension lower than that of the atmosphere. Through this study we determined a standard protocol for culturing individual bovine zygotes in a manner which produces as many good quality embryos as obtained in routine group culture. Acknowledgement The authors thank M. Julian (JustMe Editing, Storrs, CT) for editing and critical reading of the manuscript. References [1] Gordon I. Laboratory production of cattle embryos, Second Edition, Oxon, UK–Cambridge, USA: CABI Publishing; 2003. [2] Carolan C, Lonergan P, Khatir H, Mermillod P. In vitro production of bovine embryos using individual oocytes. Mol Reprod Dev 1996;45:145–50. [3] Donnay I, Van Langendonckt A, Auquier P, Grisart B, Vansteenbrugge A, Massip A, et al. Effects of co-culture and embryo number on the in vitro development of bovine embryos. Theriogenology 1997;47:1549–61. [4] O’Doherty EM, Wade MG, Hill JL, Boland MP. Effects of culturing bovine oocytes either singly or in groups on development to blastocysts. Theriogenology 1997;48:161–9.

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738 [5] Pereira DC, Dode MAN, Rumpf R. Evaluation of different culture systems on the in vitro production of bovine embryos. Theriogenology 2005;63:1131–41. [6] Fujita T, Umeki H, Shimura H, Kugumiya K, Shiga K. Effect of group culture and embryo-culture conditioned medium on development of bovine embryos. J Reprod Dev 2006;52:137–42. [7] Feng WG, Sui HS, Han ZB, Chang ZL, Zhou P, Lui DJ, et al. Effects of follicular atresia and size on the developmental competence of bovine embryos: a study using the well-in-drop system. Theriogenology 2007;67:1339–50. [8] Paria BC, Dey SK. Preimplantation embryo development in vitro: cooperative interactions among embryos and role of growth factors. In: Proceedings of the National Academy of Sciences of the United States of America; 1990. p. 4756–60. vol. 87. [9] Thibodeaux JK, Myers MW, Hansel W. The beneficial effects of incubating bovine embryos in groups are due to platelet-derived growth factor. Theriogenology 1995;43:336. [10] Lim JM, Hansel W. Roles of growth factors in the development of bovine embryos fertilized in vitro and cultured singly in a defined medium. Reprod Fertil Dev 1996;8:1199–205. [11] O’Neill C. Evidence for the requirement of autocrine growth factors for development of mouse preimplantation embryos in vitro. Biol Reprod 1997;56:229–37. [12] Oyamada T, Fukui Y. Oxygen tension and medium supplements for in vitro maturation of bovine oocytes cultured individually in a chemically defined medium. J Reprod Dev 2004;50:107–17. [13] Vajta G, Peura TT, Holm P, Paldi A, Greve T, Trounson AO, et al. New method for culture of zona-included or zona-free embryos: the well of the well (WOW) system. Mol Reprod Dev 2000;55: 256–64. [14] Mizushima S, Fukui Y. Fertilizability and developmental capacity of bovine oocytes cultured individually in a chemically defined maturation medium. Theriogenology 2001;55:1431–45. [15] Ward FA, Lonergan P, Enright BP, Boland MP. Factors affecting recovery and quality of oocytes for bovine embryo production in vitro using ovum pick-up technology. Theriogenology 2000;54: 433–46. [16] Bols PEJ. Puncture of immature ovarian follicles in bovine assisted reproduction. Verhandelingen 2005;LXVII(3):177–82. [17] Campbell B, Souza C, Gong J, Webb R, Kendall N, Marsters P, et al. Domestic ruminants as models for the elucidation of the mechanisms controlling ovarian follicle development in humans. Reprod Suppl 2003;61:429–43. [18] Wheeler MB, Beebe DJ, Walters EM, Raty S. Microfluidic technology for in vitro embryo production. In: Proceedings of the 2nd annual international IEEE-EMBS special topic conference on microtechnologies on medicine & biotechnology; 2002. p. 104–8. [19] Hendriksen PJM, Bevers MM, Dieleman SJ. Single IVP using BRL cell co-culture and serum yields a lower blastocyst rate than a group culture. Theriogenology 1999;51:319. [20] Li R, Wen L, Wang S, Bou S. Development, freezability and amino acid consumption of bovine embryos cultured in synthetic oviduct fluid (SOF) medium containing amino acids at oviductal or uterine-fluid concentrations. Theriogenology 2006;66:404–14. [21] Goto K, Kajihara Y, Kosaka S, Koba M, Nakanishi Y, Ogawa K. Pregnancies after co-culture of cumulus cells with bovine embryos derived from in-vitro matured follicular oocytes. J Reprod Fertil 1988;83:753–8. [22] Orsi NM, Reischl JB. Mammalian embryo co-culture: trials and tribulations of a misunderstood method. Theriogenology 2007;67:441–58.

737

[23] Voelkel SA, Hu YX. Effect of gas atmosphere on development of one-cell bovine embryos in two culture systems. Theriogenology 1992;37:1117–31. [24] Fischer B, Bavister BD. Oxygen-tension in the oviduct and uterus of rhesus-monkeys, hamsters and rabbits. J Reprod Fertil 1993;99:673–9. [25] Yuan YQ, Van Soom A, Coopman FOJ, Mintiens K, Boerjan ML, Van Zeveren A, et al. Influence of oxygen tension on apoptosis and hatching in bovine embryos cultured in vitro. Theriogenology 2003;59:1585–96. [26] Khatir H, Anouassi A, Tibary A. In vitro and in vivo developmental competence of dromedary (Camelus dromedarius) embryos produced in vitro using two culture systems (mKSOMaa and oviductal cells). Reprod Dom Anim 2005;40: 245–9. [27] Locatelli Y, Cognie´ Y, Vallet JC, Baril G, Verdier M, Poulin N, et al. Successful use of oviduct epithelial cell coculture for in vitro production of viable red deer (Cervus elaphus) embryos. Theriogenology 2005;64:1729–39. [28] Correˆa GA, Rumpf R, Mundim TCD, Franco MM, Dode MAN. Oxygen tension during in vitro culture of bovine embryos: effect in production and expression of genes related to oxidative stress. Anim Reprod Sci 2008;104:132–42. [29] Hagemann LJ, Weilert LL, Beaumont SE, Tervit HR. Development of bovine embryos in single in vitro production (sIVP) systems. Mol Reprod Dev 1998;51:143–7. [30] Han ZB, Lan GC, Wu YG, Han D, Feng WG, Wang JZ, et al. Interactive effects of granulosa cell apoptosis, follicle size, cumulus-oocyte complex morphology, and cumulus expansion on the developmental competence of goat oocytes: a study using the well-in-drop culture system. Reproduction 2006;132: 749–58. [31] Ahern TJ, Gardner DK. Culturing bovine embryos in groups stimulates blastocyst development and cell allocation to the inner cell mass. Theriogenology 1998;49:194. [32] Lonergan P, Rizos D, Gutierrez-Adan A, Fair T, Boland MP. Effect of culture environment on embryo quality and geneexpression–experience of animal studies. Reprod BioMed Online 2003;7:657–63. [33] de Wit AAC, Kruip ThAM. Bovine cumulus-oocyte-complexquality is reflected in sensitivity for a-amanitine, oocyte-diameter and developmental capacity. Anim Reprod Sci 2001;65: 51–65. [34] Joo BS, Kim MK, Na YJ, Moon HS, Lee KS, Kim HD. The mechanism of action of coculture on embryo development in the mouse model: direct embryo-to-cell contact and the removal of deleterious components. Fertil Steril 2001;75:193–9. [35] Tan XW, Ma SF, Yu JN, Zhang X, Lan GC, Liu XY, et al. Effects of species and cellular activity of oviductal epithelial cells on their dialogue with co-cultured mouse embryos. Cell Tissue Res 2007;327:55–66. [36] Blondin P, Coenen K, Guilbault LA, Sirard MA. In vitro production of bovine embryos: developmental competence is acquired before maturation. Theriogenology 1997;47: 1061–75. [37] Edwards LJ, Batt PA, Gandolfi F, Gardner DK. Modifications made to culture medium by bovine oviduct epithelial cells: changes to carbohydrates stimulate bovine embryo development. Mol Reprod Dev 1997;46:146–54. [38] Harvey AJ. The role of oxygen in ruminant preimplantation embryo development and metabolism. Anim Reprod Sci 2007;98:113–28.

738

I.G.F. Goovaerts et al. / Theriogenology 71 (2009) 729–738

[39] Lim JM, Reggio BC, Godke RA, Hansel W. Development of invitro-derived bovine embryos cultured in 5% CO2 in air or in 5% O2, 5% CO2 and 90% N2. Hum Reprod 1999;14:458–64. [40] Agarwel A, Allamaneni SSR. Role of free radicals in female reproductive diseases and assisted reproduction. Reprod BioMed Online 2004;9:338–47.

[41] Maeda J, Kotsuji F, Negami A, Kamitani N, Tominaga T. In vitro development of bovine embryos in conditioned media from bovine granulosa cells and vero cells cultured in exogenous protein- and amino acid-free chemically defined human tubal fluid medium. Biol Reprod 1996;54:930–6.