| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Department of Farm Animal Health,3Faculty of Veterinary Medicine, Utrecht University, 3584 CL Utrecht, The Netherlands
Department of Anatomy and Physiology,4 Royal Veterinary and Agricultural University, DK-1850 Frederiksberg C, Denmark
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
apoptosis, developmental biology, early development, female reproductive tract
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
A variety of morphological and biochemical parameters have been studied in relation to the developmental competence of the oocytes and to embryo quality during in vitro production. These studies showed that bovine embryos produced in vitro differ from their in vivo counterparts in many aspects, such as morphology, cell number of the total embryo and of the inner cell mass (ICM) [4], expression of specific genes [5, 6], and incidence of chromosomal abnormalities [7].
In the above-mentioned studies, in vivo-developed morulae and blastocysts were compared with embryos for which maturation, fertilization, and culture had been performed in vitro. Recently, some data have been reported concerning the effects of the respective, separate steps of the in vitro procedure on embryo quality. In vivo-occurring prematuration and maturation of the oocyte are crucial for enhancing the competence of the oocyte to develop in vitro into a blastocyst [8], but they have less influence on the quality of the embryo. Different modes of maturation, in vitro or in vivo, of prematured oocytes did not influence the level of expression of six developmentally important genes in blastocysts [6], but they did have a moderate effect on the incidence of chromosomal aberrations in cells of the blastocyst [9]. The postfertilization period appears to be the most important step in determining the quality of the embryo. Postfertilization culture in vitro reduced the postthaw survival rate of blastocysts following cryopreservation [10] and modified expression of different genes [11]. To unravel the effects of in vitro culture (IVC) versus in vivo development on the quality of the embryo, we investigated morulae and blastocysts after temporary development in vivo and subsequent culture in vitro. The blastulation rate is a criterion for evaluating the quality of in vitro-produced embryos [12]. In the cow, blastulation occurs at an earlier point of time after insemination in embryos derived in vitro than in embryos developed entirely in vivo [4]. In particular, the presence of serum during culture accelerates the formation of the blastocoele [13]. As a measure for cleavage rates, the number of cells of the total embryo and of the ICM has often been used to evaluate embryo quality. Small differences between in vitro- and in vivo-derived embryos already occur during development up to the 8-cell stage [7], but thereafter, until the blastocyst stage, the cell number is significantly smaller after IVC [7, 14, 15]. The cell number of the ICM is also decreased after IVC [4, 7]. The cell number of the total embryo and of the ICM appeared to be dependent on the culture medium used [4]. Therefore, the environment during early embryonic development clearly plays a role in determining the cell numbers of the embryo.
Although apoptosis can be considered to be a normal process in preimplantation embryos to eliminate deviating cells, a high incidence of apoptotic cells is correlated with abnormal morphology of the embryo [16]. In in vitro-produced mouse blastocysts, the percentage of apoptotic cells is significantly higher than in their in vivo-developed counterparts [17, 18]. Also, data from the cow indicate that in vitro-produced blastocysts possess a higher level of apoptosis than embryos developed completely in vivo [19]. Furthermore, the presence or absence of factors in the culture medium can affect the incidence of apoptosis in mouse [17, 18, 20] and bovine embryos [21]. It has been suggested that "survival factors" produced by the embryo itself and by the maternal reproductive tract regulate the incidence of apoptosis [17, 18]. Although the environment during early embryonic development evidently influences the level of apoptosis in mammalian blastocysts, it is not known at which stage during IVC increased levels of apoptosis are induced.
In the present study, embryos that had been collected in vivo before or after the major genome activation were cultured. The major genome activation presumably starts during the fourth cell cycle [22] that, in vivo, takes place between 45 and 100 h postovulation (po) [23]. Therefore, embryos from superovulated heifers were flushed from the genital tract at 45 and 100 h po and, subsequently, cultured until Day 7 po. For comparison, Day 7 embryos that were derived in vitro from slaughterhouse oocytes or after development entirely in vivo were analyzed. The quality of the embryos was assessed by determining the blastulation rate, the cell number of the total embryo and of the ICM, and the level of apoptosis in the embryo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
To obtain embryos at precisely defined times after ovulation, heifers were treated for superovulation using a procedure with a controlled LH surge as described by Vos et al. [24] with slight modifications. Clinically healthy, nonlactating Holstein-Friesian heifers were selected from the experimental herd of the Veterinary Faculty of Utrecht University on the basis of cyclicity, as established by measuring progesterone levels in peripheral blood samples taken three times a week for at least 4 wk before the experiment started. The heifers were fed silage and concentrate and were supplied with water ad libitum. The experiments were carried out as approved by the Ethical Committee of the Veterinary Faculty of Utrecht University in December 2000 (first session, n = 32 heifers) and October 2001 (second session, n = 27 heifers).
The heifers were presynchronized using an ear implant for 9 days (3 mg of norgestomet; Crestar; Intervet International BV, Boxmeer, The Netherlands) accompanied by treatment with 3 mg of norgestomet and 5 mg of estradiol-valerate i.m. Two days before the implant was removed, prostaglandin (15 mg; Prosolvin; Intervet International BV) was administered i.m. to ensure complete regression of the corpus luteum. On Day 8 of the synchronized cycle (estrus = Day 0), all follicles larger than 5 mm were ablated by transvaginal, ultrasound-guided puncturing to synchronize follicular development. At the time of puncturing, one of the animals was excluded from the experiment, because it showed cystic ovaries. On Day 9, the remaining heifers (n = 58) were administered another ear implant (Crestar) for 5 days, but without the additional administration of norgestomet and estradiol-valerate. From Day 10 onward, ovine FSH (Ovagen; ICP, Auckland, New Zealand) was administered i.m., twice daily, with decreasing doses during 4 days: on the first day, 2.0 ml; on the second day, 1.5 ml; on the third day, 1.0 ml; and on the fourth day, 0.5 ml (10 ml in total, equivalent to 176 IU of NIH-FSH-S1). Prostaglandin (22.5 mg of Prosolvin) was administered i.m. concomitant with the fifth dose of FSH, and 55 h later, the ear implants were removed and GnRH administered (1.0 mg of Fertagyl [Intervet International BV[ in 10 ml of saline i.m.) to induce an LH surge. Heparinized blood samples from the jugular vein were collected daily during the experimental cycle, every 3 h from 24 h after PG administration and every hour during the 7-h period after removal of the second implant. After immediate centrifugation (1800 g for 10 min) at 4°C, plasma was stored at -25°C. All animals were inseminated 12–14 h after GnRH administration with two straws of semen from a known-fertile bull (one straw/uterine horn). In both sessions, the heifers were assigned at random to the three in vivo groups. The time point of 24 h after the LH surge was taken as the starting time of multiple ovulations. The time between expected ovulation and slaughter to collect embryos was 45–48 h for the 45-h in vivo group, 100–103 h for the 100-h in vivo group, and 165–168 h for the in vivo group (Fig. 1).
|
RIAs for Progesterone and LH
Concentrations of progesterone in plasma were estimated by a solid-phase, 125I RIA method (Coat-A-Count TKPG; Diagnostic Products Corporation, Los Angeles, CA) according to the manufacturer's instructions as validated previously [25]. The sensitivity was 0.15 nmol/L, and the intra- and interassay coefficients of variation were 8% and less than 11%, respectively.
Concentrations of LH in plasma were estimated by a validated, homologous RIA method as described previously [26]. The sensitivity was 0.4 µg/L NIH-LH-B4. The intra- and interassay coefficients of variation were less than 9%.
Embryo Collection for the In Vivo Groups
After slaughter, genital tracts were placed in saline (37°C) and transported immediately to the laboratory in a thermocontainer. Per genital tract, the number of corpora lutea on the ovaries was counted. The time period between slaughter and flushing ranged from 45 min to 2 h. Oviducts and uterine horns were flushed with PBS (PBS-ET; Bio Whitaker Europe, Verviers, Belgium) at 37°C using a blunt needle and from the infundibulum toward the uterine horn under gentle massage. For the 45-h in vivo group only, the oviducts with the top of the uterine horn were flushed. The PBS was collected into an embryo-recovery filter (Embryo Concentrator; Immuno Systems, Inc., Spring Valley, WI), and the embryos were collected by rinsing the filter with saline supplemented with 0.005% (w/v) BSA (A 6003; Sigma, St Louis, MO).
Embryo Culture
In vivo groups The embryos were transferred to the culture medium within 2 h after slaughter. The developmental stage of the viable and nonviable embryos (i.e., of all embryonic structures) was assessed by stereomicroscopy. The 1-cell and degenerated embryos as well as, at 100 h po, the embryos at the 2- to 7-cell stages were characterized as nonviable and excluded from further investigation. Before culture, embryos were washed in culture medium consisting of synthetic oviduct fluid (SOF) with BSA as described by van Wagtendonk-de Leeuw et al. [3]. The embryos were grouped per heifer, and at most, 10 ex vivo-collected embryos were placed in 20-µl droplets of SOF under oil (Mineral Oil for IVF; Reproline Medical GmbH, Rheinback, Germany) [3]. In vitro culture was performed at 39°C in humidified air of 5% CO2, 7% O2, and 88% N2. Morulae and blastocysts were collected after 120 h of culture for the 45-h in vivo group and after 72 h for the 100-h in vivo group (Fig. 1).
In vitro group A control group of embryos was collected that was produced completely in vitro. Cumulus-oocyte complexes were aspirated from 3- to 8-mm follicles of ovaries that had been collected at a local abattoir, but only those with a multilayered, compact cumulus investment were used for the experiments. The cumulus-oocyte complexes were rinsed once with Hepes-buffered M199 (Gibco BRL, Paisley, U.K.) supplemented with 10% (v/v) fetal calf serum (Gibco BRL) and once with maturation medium M199 supplemented with 10% fetal calf serum, 0.01 IU/ml of recombinant human FSH (Organon, Oss, The Netherlands), 11.36 µg/ml of cysteamine (M-6500; Sigma), and 1% (v/v) penicillin-streptomycin (Gibco BRL). Groups of 35 oocytes were allocated at random to each well of a four-well culture plate (Nunc A/S, Roskilde, Denmark) containing 500 µl of maturation medium with the additions described above. After maturation for 22 h (39°C, 5% CO2 in humidified air), all oocytes were fertilized.
Procedures for in vitro fertilization were performed as described by Parrish et al. [27] with minor modifications [28] using semen from the same bull as used in the in vivo experiments for artificial insemination. After 21 h of incubation (39°C, 5% CO2 in humidified air), the presumptive zygotes were freed from cumulus cells by vortexing for 3 min, and a maximum of 10 zygotes was placed in 20-µl droplets of SOF and cultured as described above. All cleavage stages were transferred to fresh culture droplets at Day 5 after the start of fertilization. Morulae and blastocysts were collected 165 h after the start of fertilization.
Determination of Embryonic Stage and Fixation
The developmental stage (morula, early blastocyst, blastocyst, expanded blastocyst, and hatched blastocyst) of all embryonic structures at Day 7 was assessed and scored by stereomicroscopy. The embryonic structures that had not attained the morula or blastocyst stage at 165–172 h po (in vivo groups) or after fertilization (in vitro group) were excluded from further analysis. These embryos were considered to be nonviable.
Half of all collected morulae and blastocysts were used for TUNEL; the other half were stored for another study. For TUNEL, embryos were washed three times with PBS containing 1 mg/ml of polyvinylacohol (P-8136; Sigma) and then fixed in 4% (w/v) paraformaldehyde (Merck, Darmstadt, Germany) in PBS for 1–2 h at room temperature. After fixation, the embryos were stored in 1% paraformaldehyde in PBS and processed for TUNEL within 2 wk.
TUNEL Assay
Nuclei with degraded DNA were detected by using a cell death-detection technique based on the TUNEL principle [29] using fluorescein-conjugated dUTP as described previously [17, 21] with minor modifications. The embryos were washed three times in PBS with polyvinylpyrrolidone (PBS/PVP; 1 mg/ml; Sigma), permeabilized for 1 h in PBS with 0.5% (v/v) Triton X-100 (Sigma), and washed twice in PBS/PVP. As positive controls, embryos were incubated in 50 U DNase/ml PBS (RQ1; Promega, Bie & Bernsten, Rødovre, Denmark) for 30 min at 37°C and then washed two times in PBS/PVP. Embryos were incubated in 10 µl of terminal deoxynucleotidyl transferase and 90 µl of fluorescein-conjugated dUTP (TUNEL, In Situ Cell Death Detection Kit; Roche, Hvidovre, Denmark) for 60 min at 37°C in the dark. For negative controls, embryos were not incubated with the terminal transferase enzyme. The embryos were washed twice in the Triton X-100 in PBS and once in PBS/PVP. Embryos were washed once in Tris-buffer (40 mM Tris, 10mM NaCl, and 6 mM MgCl2; pH 8.0) and then incubated with 0.1 mg/ml of RNase A (Sigma) in Tris-buffer for 1 h at 37°C in the dark. The nuclear material was counterstained with 10 µg/ml of propidium iodide (PI; Sigma) in the Tris-buffer for 45 min at 37°C in the dark. Embryos were then transferred through a gradient of Vecta-Shield (Vector Laboratories, Burlingame, CA) at 50%, 75%, and 100% (v/v) in PBS in the dark, with each step lasting for 30 min; in the 100% Vecta-Shield, 0.05 µg/ml of PI was added. Then, the embryos were mounted on a slide with ring enforcement in 4 µl of the 100% Vecta-Shield solution and covered with a coverslip. Slides were stored at 4°C for up to 7 days before confocal laser scanning microscopy was performed.
Microscopy and Image Processing
The embryos were subjected to confocal laser-scanning microscopy on a Leica TCS4D microscope (Leica Laser Technik GmbH, Heidelberg, Germany) fitted with 25/40x PL Fluotar/0.75 oil objectives. An argon/krypton laser was used for excitation at 488 and 568 nm, and two-channel scanning was performed with a double-dichroic DD488/568 beam splitter and a band-pass BP530 barrier and a long-pass OG590 barrier filter for detection of TUNEL reaction and PI, respectively. A complete Z series of 20–25 optical sections at 3- to 4-µm intervals was acquired from each embryo using Leica Scanware software (Leica Laser Technik), and image stacks were reconstructed on a Silicon Graphics computer equipped with an Imaris image-analysis software (Bitplane, Zurich, Switzerland) package.
Quantitative Analysis of TUNEL Labeling and Apoptosis
Digitally recombined, composite images were analyzed using the Imaris software. All 20–25 optical sections were divided using a standard grid over each layer to count all nuclei as a measure of the cell number of the total embryo. Nuclei were scored for TUNEL labeling, signs of fragmentation, and condensation. Cells were judged to be apoptotic when the nucleus displayed both the biochemical feature (TUNEL labeling) and the morphological feature (fragmentation and/or condensation) as described by Gjørret et al. (see Note Added in Proof). The apoptotic index of the embryos was calculated as the percentage of apoptotic cells relative to the total number of cells. Allocation of nuclei to ICM and trophectoderm (TE) was based on their position in the reconstructed images (see Fig. 2F). The nuclei belonging to the polar TE were counted together with the ICM nuclei, leading to an overestimation of the ICM. Differential staining of the ICM to provide a more accurate number of nuclei for the ICM [30] was not performed, because this technique cannot be combined with TUNEL labeling.
|
Validation of the counting of cells and TUNEL-positive (T+) cells was done on a random selection of 10% (20/194) of all embryos. After renewed allocation of the nuclei to the ICM and TE, counting was performed a second time. The proportional difference between the first and second countings was 0.9% for the cell number per total embryo and 1.5% for the level of T+ nuclei. For the number of nuclei in the ICM and TE, the differences were both 1.6% and, for the level of T+ in the ICM and TE, the differences were 2.3% and 4.3%, respectively.
Statistical Analysis
Correlation analysis between cell number and incidence of apoptosis was performed by the linear regression test. All tests were performed using SPSS 8.0 statistical software (SPSS, Inc., Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
No differences were observed between the two sessions with regard to the LH surge. On average, it occurred 2.5 h after the administration of GnRH, with a maximum level (mean ± SEM) of 22.4 ± 1.2 µg/L (n = 56). In two heifers, an increase of the LH concentration was observed some hours before the GnRH administration. The LH data of these two animals were not included.
Embryo Collection In Vivo
In the majority of the heifers, a variable proportion of the collected embryonic structures was characterized as nonviable. The embryos of the few animals (n = 7) with an exceptionally high proportion of nonviable embryonic structures (>85%) were not used for the experiment.
The superovulatory responses with regard to the number of corpora lutea and the recovery rate of the embryonic structures relative to the number of corpora lutea were similar between the two sessions. The recovery rate was not different for the 45-h in vivo, 100-h in vivo, and in vivo groups, being 75% (n = 18 heifers), 66% (n = 13), and 70% (n = 20), respectively. The developmental stages of the embryos flushed at 45 h, 100 h, and Day 7 po are shown in Tables 1 and 2.
|
|
Embryo Culture Results and Developmental Rate
A significantly higher (P < 0.05) proportion of the embryos selected for culture developed in vitro to the morula or blastocyst stage when culture was started after 100 h of development in vivo (90%) than after 45 h (68%) (Table 2). In both the 45- and 100-h in vivo groups, the proportion of embryos developing further in vitro was much higher than the proportion of oocytes developed completely in vitro to the morula and blastocyst stages (29% for the in vitro group). The proportion of embryos at the blastocyst stage was high in the 45- and 100-h in vivo groups (91% and 100%, respectively), similar to that in the in vitro group (91%), and the proportion of embryos at the morula stage was low. On the contrary, after development entirely in vivo, the proportions of embryos at the morula and blastocyst stages were more alike (43% and 57%, respectively) (Table 2).
Cell Count
In general, the cell number of the total embryo was significantly lower for embryos produced completely in vitro compared to that of embryos developed entirely in vivo (Table 3). The cell numbers of embryos in the 45-h in vivo group were not significantly different from those in the in vitro and the in vivo group at corresponding stages. However, in the 100-h in vivo group, the cell number of the embryos at the expanded blastocyst and hatched blastocyst stages was significantly higher than that in the in vitro group and similar to that in the in vivo group. At the blastocyst stage, the cell number of the total embryo tended to increase concurrently with extension of the period of development in vivo.
|
In 15% of the embryos that were characterized as blastocyst, expanded blastocyst, or hatched blastocyst by stereomicroscopy, the respective numbers of nuclei of the ICM and TE could not be assessed because of collapse of the blastocoele. As shown in Table 4, the average cell number of the ICM was significantly lower in embryos of the in vitro group over all stages with a blastocoele (blastocyst, expanded blastocyst, and hatched blastocyst) than in those of the in vivo group. Likewise, the cell number of the ICM in the 45-h in vivo group was not significantly different from that in the in vitro and in vivo group. As observed for the cell number of the total embryo, the embryos at the expanded blastocyst and hatched blastocyst stages in the 100-h in vivo group again showed a significantly higher number of cells, but now for the ICM, than in the in vitro group and a number similar to that in the in vivo group. For the TE, the number of cells was markedly lower than in the corresponding ICM, showing only minor variation between groups at specific developmental stages.
|
Apoptosis
From all nuclei with clear signs of TUNEL labeling, 98% were condensed, and 79% were condensed and fragmented (Fig. 2). The apoptotic index of the embryos is shown in Table 5. In general, it tended to be higher in the in vitro group at all embryonic stages than in the corresponding stages in the 45- and 100-h in vivo groups and the in vivo group. However, the difference was significant only between the expanded blastocyst and hatched blastocyst stages of the in vitro group and the 45-h in vivo group. The average apoptotic index over all embryos, regardless of the developmental stage, was rather similar between the 45- and 100-h in vivo groups and the in vivo group. The apoptotic index in blastocysts (blastocyst, expanded blastocyst, and hatched blastocyst) of all four groups was significantly higher in the ICM than in the TE (Table 6 and Fig. 2, E and F). In the 45-h in vivo group and the in vivo group, the apoptotic index in the ICM was significantly lower than that in the in vitro group.
|
|
The distribution of the embryos according to low (<5%), average (5%–10%), and high (>10%) levels of apoptosis in the total embryo was not different between the four groups (results not shown).
Although the correlation between cell number of the total embryo and apoptotic index was significant (R2 = 0.026, P < 0.05) (Fig. 3), the decrease of the apoptotic index with increasing cell number was only marginal.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
The rate of development into morula and/or blastocyst stages in the in vitro group was much lower than in the groups with temporary or total in vivo development, which is in accordance with earlier observations that prematuration and maturation are decisive factors for oocyte developmental competence [8, 9]. The rate of blastocyst development relative to the number of viable embryos in the in vivo group was lower than in the other three groups, concurrent with a higher rate of morula development. The present results indicate that in vivo initiation of blastulation occurs after 100 h po and may be mediated by factors from the uterine environment. The majority of the embryos used for further IVC were collected before (45-h in vivo group, 92% at the 5- to 8-cell stages) or after (100-h in vivo group, 79% at the 8- to 32-cell stages) completion of the fourth cell cycle. The rate of development into morula and/or blastocyst stages relative to the embryos selected for culture was higher in the 100-h in vivo group than in the 45-h in vivo group, which seems to indicate that passage of the fourth cell cycle may stimulate embryo development. However, this difference appears to result from selection of embryos for IVC. In the 100-h in vivo group, a considerably higher proportion of nonviable embryos (27%), which was similar to that in the in vivo group, had been discarded for IVC compared to the proportion in the 45-h in vivo group (5%). When the proportion of embryos developed to the morula and/or blastocyst stages is calculated over the total number of embryonic structures, the values are almost identical (45-h in vivo group, 107/167 = 64%; 100-h in vivo group, 107/163 = 66%). This is in agreement with earlier studies, in which the developmental competence of in vitro- or in vivo-derived zygotes did not change with further culture in vitro or in vivo in the ewe oviduct [10, 15]. The present data suggest that part of in vivo-derived, 5- to 8-cell stage embryos are not competent to develop beyond the fourth cell cycle, regardless of in vitro or in vivo conditions. The differences between the embryos of the four in vitro and in vivo groups were probably not caused by a low number of embryos per volume during culture. In an earlier study, the rate of blastocyst development decreased significantly below 72% when less than four 8-cell stage embryos were cultured in 50-µl SOF droplets under oil, and the number of ICM cells also decreased [31]. In the present study, a higher ratio of, on average, nine embryos per 20 µl was employed.
The number of cells of the total embryo, as a measure of the growth rate, at 7 days po was higher in the 100-h in vivo and the in vivo group than in the in vitro group. In the 45-h in vivo group, the number was intermediate. This pattern was particularly evident in the expanded blastocysts constituting the major proportion of the embryos at the respective blastocyst stages. It is interesting to note that the growth rate of embryos in the in vitro group was much slower than of that of in vivo embryos but not as much as observed earlier using Menezo B2-coculture [7, 32] instead of SOF (as used in the present study). Effects of different culture media on the growth rate have been shown by van Soom et al. [30]. The number of ICM cells of embryos in the four groups followed a pattern parallel to that observed for the total number of embryo cells, which is in accordance with the higher number of ICM cells reported for in vivo-developed embryos versus embryos after IVC [4]. It was suggested that contact with the maternal tract might be of importance to switch on certain genes that encode for developmentally important processes, such as tight junction formation in the case of inner cell allocation. The fact that the total number of cells and of the ICM in the 100-h in vivo group was similar to that in the in vivo group suggests that the growth rate is defined at 100 h po, when the fourth cell cycle is completed. It is speculated that in vivo, the embryonic cells are more stimulated to cleave because of interaction with the oviduct. However, a higher cell death, specifically at earlier stages of development, may have been the cause of the lower cell numbers in the in vitro group. Cell death appears to be correlated with cell number in mouse and human blastocysts [18], and in the present study, embryos with low cell numbers showed a higher apoptotic index than embryos with high cell numbers.
The incidence of apoptosis in the in vitro group was higher than in the 45-h in vivo group and the in vivo group, particularly in expanded blastocysts and their ICM. In the 100-h in vivo group, the apoptotic index was intermediate, which may have been caused by the change from the in vivo to the in vitro environment. Stress has been reported to increase apoptosis. Recently, it was demonstrated that heat stress can induce TUNEL labeling at the late 8- to 16-cell stages, but not at earlier stages [33]. In the 45-h in vivo group, stress caused by change of environment possibly did not affect the incidence of apoptosis at the blastocyst stage, because in vivo, the apoptotic machinery is not apparent before the 21-cell stage as no DNA degradation was observed before that stage (see Note Added in Proof). Although in the bovine apoptosis is not evident before the 21-cell stage, various molecular components may already be present at early cleavage stages, as reported in mouse [34], human [35], and bovine embryos [33, 36]. Furthermore, in the 45-h in vivo group, the apoptotic cells that were eventually induced probably had already been phagocytosed during further culture until collection at Day 7 po. Phagocytosis has been observed in human and bovine blastocysts [37, 38]. The entire time span for apoptosis, from the beginning of cell rounding and blebbing to the final lysis of the cell, takes 12–24 h [39, 40], which implies that the apoptotic nuclei we observed form a "snapshot." Overall, the present data show small differences in the apoptotic index of the total embryo between groups, which indicates that SOF provides an environment in which apoptosis in bovine embryos occurs in a manner fairly similar to that in vivo. Kölle et al. [36] observed an even lower level of apoptosis in embryos cultured in SOF compared to that in our in vitro group. In other studies, in bovine embryos using Menezo B2-coculture [19] and in murine embryos [41], the difference between completely in vitro- and entirely in vivo-derived embryos was more pronounced.
The variable overestimation of the number of cells of the ICM did influence the ICM:TE ratio, but not the relation of the apoptotic indexes between the four groups. In all four groups, the apoptotic index of the ICM was significantly higher than that of the TE, which is in agreement with studies in bovine [19, 42, 43] and murine [18, 44, 45] embryos. It is assumed that apoptosis in the ICM may regulate this cell population, because the number of ICM cells reaches a plateau in later-stage blastocysts without decrease in mitotic division [46]. At the blastocyst stage, apoptosis likely acts to eliminate cells that are damaged, are in excess, are no longer required, are developmentally incompetent, or have acquired TE potential. This cellular "quality control" within the ICM is critical. This lineage forms the fetus and contains the germline, and furthermore, an aberrant TE:ICM ratio has been suggested to be related with the large offspring syndrome of in vitro-produced embryos [47].
In the cow, passage of the fourth cell cycle coincides in vivo with the period of major genome activation. In other species, this period appears to induce competence for apoptosis [48] and to activate expression of genes that serve to suppress cell death in developing embryos [41, 49, 50]. The present study shows that passing the period of major genome activation in vitro or in vivo does not affect the level of apoptosis in bovine Day 7 embryos.
In conclusion, development in vivo during the first 100 h po up to the fifth cell cycle is decisive for the cell number of the total embryo and the ICM in the expanded blastocyst at 7 days po. However, timing of blastulation appears to be affected by uterine factors after 100 h po. The level of apoptosis appears to be determined in vitro in SOF at an earlier stage of development, before 45 h po.
|
NOTE ADDED IN PROOF |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Hiemke M. Knijn, Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 7, 3584 CL Utrecht, The Netherlands. FAX: 31 30 2521887; h.knijn{at}vet.uu.nl
Received: 17 March 2003.
First decision: 8 April 2003.
Accepted: 4 June 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
. Biol Reprod 1997 56:1088-1096[Abstract]This article has been cited by other articles:
|
|
 |
|
  M. Takenaka, T. Horiuchi, and R. Yanagimachi Effects of light on development of mammalian zygotes PNAS, September 4, 2007; 104(36): 14289 - 14293. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pfister-Genskow, C. Myers, L. A. Childs, J. C. Lacson, T. Patterson, J. M. Betthauser, P. J. Goueleke, R. W. Koppang, G. Lange, P. Fisher, et al. Identification of Differentially Expressed Genes in Individual Bovine Preimplantation Embryos Produced by Nuclear Transfer: Improper Reprogramming of Genes Required for Development Biol Reprod, March 1, 2005; 72(3): 546 - 555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Beker-van Woudenberg, H. T.A. van Tol, B. A.J. Roelen, B. Colenbrander, and M. M. Bevers Estradiol and Its Membrane-Impermeable Conjugate (Estradiol-Bovine Serum Albumin) During In Vitro Maturation of Bovine Oocytes: Effects on Nuclear and Cytoplasmic Maturation, Cytoskeleton, and Embryo Quality Biol Reprod, May 1, 2004; 70(5): 1465 - 1474. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS |
