- Email: [email protected]
Sucrose-induced shrinkage of in vitro produced bovine morulae: Effect on viability, morphology and ease of evaluation
ELSEVIER
SUCROSE-INDUCED SHRINKAGE OF IN VITRO PRODIJCED BOVINE MORULAE: EFFECI ON VIABILITY, MORPHOLOGY AND EASE OF EVALUATION
A. Van Soom,’ M.-T. Ysebaert,z A. Vanhoucke-De Medts,’ A. Van de Velde,’ S. Metton, A. DelvalP A. Van Langendonckt,S I. Donnay, G. Vanroose,t P.E.J. Bolst and A. de Kruift ‘Department of Reproduction, Obstetrics and Herd Health and ;?Deparmrent of Physiology, Biochemistry and Biostatistics, University of Gent, B-9000 Gent, Belgium, 3Holland Genetics, IVP Laboratory, 5437 BG Beers, The Netherlands 41RSIA Research Unit, Sat-t-Tilman, B-4000 Liege, Belgium, SCatholic University of Louvain, Veterinary Unit, B- 1348 Louvain-la-Neuve, Belgium Received for publication : February 8, 1996 Accepted : May 10, 1996 ABSTRACT Sucrose (0.3 M) was used to cause artificial compaction of the embryonic cell mass of in vitro produced bovine embryos to facilitate morphological evaluation. Embryos were produced using routine in vitro maturation (IVM) and fertilization (IVF) techniques. The time necessary to induce shrinkage in 0.3 M sucrose to 75% of the original volume of Day 5 morulae was found to be less than 1 min, and 95% of the volume was regained in PBS after 2.5 min. No detrimental effect was observed after a 5 to IO-min sucrose treatment on subsequent blastocyst formation at Days 6 and 7 (P> 0.05). Furthermore, no significant differences were observed in the total number of cells, or in the mitotic and pycnotic cell index of blastocysts in different treatment groups. Agreement among 7 evaluators grading 40 Day 6 embryos was examined using the kappa coefficient of agreement (tc). Overall agreement among evaluators for classification of quality grade was poor (48.2 %,
K =
0.31) for embryos evaluated in PBS, but
the rate improved when the same embryos were scored in sucrose (62.5 %, K = 0.49). Evaluating less compact in vitro produced bovine morulae in sucrose increases agreement among evaluators, since embryos in sucrose mimick the appearance of in vivo produced embryos. Thus, we conclude that scoring in vitro produced embryos in sucrose improves agreement among evaluators. Key words: bovine morula compaction, sucrose, embryo morphology
Acknowledgments The authors thank R. Verbeke for computing embryo volumes and Dr. W. Vanneste for performing the Kappa statistics. We also thank A.M. van Wagtendonk-De Leeuw for critical discussions on statistical methods. This study was subsidized by an IWONLCSVH (section 1, nr 5597A) grant.
Theriogenology 46:1131-1147. 1996 0 1996 by Elsevier Science inc.
0093-691 X/96/$1 5.00 PII SOO93-691X(96)00285-9
1132
Theriogenology INTRODUCTION
In recent years, it has become possible to produce bovine embryos both by in vivo and in vitro methods. With growing interest in embryo production, it has become clear that embryos generated in vitro are different from flushed embryos, both with respect to their morphology at the light microscopical level (24) and at the ultrastructuml level (21). Although in vitro produced bovine embryos can yield satisfactory ptegnancy rates after transfer, fetal loss during pregnancy is higher than with in vivo embryos (4,20,25). This pregnancy loss is costly, especially when the number of available recipients is limited, and a reliable method for assessing embryo viability prior to transfer would be of great value. Although a number of different viability tests have been described for use in bovine embryo transfer (2), to date no extended studies have been published to relate the outcome of methods like measurement of metabolic activity, use of fluorescent probes or determination of intracellular pH with the actual number of pregnancies which can be achieved after transferring such tested embryos (16). Thus we used morphological parameters, which have been widely used as a noninvasive method, to score the quality of flushedbovine embryos (10,22). The advantages of morphological scoring tests are that they are quick, noninvasive and require readily available equipment. The disadvantages are that they are a subjective measure and they require highly trained personnel (2). Morphological criteria for assessment of embryo quality of in vivo produced embryos have been defined by Lindner and Wright (10). Moreover, these criteria have been shown to be related to pregnancy rates after transfer (8,lO). Morulae flushed out of cows 6 d post insemination am characterized by a very obvious compaction, which is defined as the coalescing of individual blastomeres. The resulting embryonic mass occupies only 60 to 70% of the space inside the zona pellucida (10). Therefore the compact morula stage is particularly suited for evaluation of the presence of extruded blastomeres and degenerated cells or fragments, since they are clearly visible in the perivitelline space. The space between zona and cells has an average width of 9.1 inn at that time, compared with 6.9 km for the 16cell embryo and only 2.6 pm for the blastocyst (15). In vitro produced morulae do not have a distinct increase in the width of the perivitelline space at the time of compaction (24). Compaction of in vitro produced embryos can only be assessed by the flattening of cells and cell outlines that become invisible, as defied for the mouse embryo by Pratt (18). Consequently, morphological evaluation of in vitro produced morulae is more difficult due to this lack of compaction (7). This has been confirmed in a recent study by Farin et al. (3), in which they compared agreement among evaluators of bovine embryos produced in vivo and in vitro. They showed that agreement for developmental stage was very good with embryos produced in vivo and only fair for embryos produced in vitro. For classification of quality grade, however, agreement was only fair to poor, regardless of whether in vivo or in vitro produced embryos were used.
1133
Theriogenology
In the present study, we undertook to facilitate quality assessment of in vitro produced bovine morulae by incubating the embryo in 0.3 M sucrose. This method offers the possibility for making embryo morphology scoring easier. The sucrose-induced shrinkage technique was first reported by Yang et al. (31,32). It is a method for enlarging the perivitelliie space of rabbit oocytes, which has made sperm microinjection under the zona more feasible. Sucrose-tolerance of rabbit zygotes and embryos at different stages of development was also reported with the purpose of performing micromanipulations such as microinjection and nuclear transplantation (33-35). The possible use of sucrose for morphological evaluation of in vitro produced bovine embryos has not been reported. The objectives of this study were 1) to define the time necessary for morula shrinkage and recovery and to measure the change in embryo volume after incubation in 0.3 M sucrose; 2) to evaluate possible negative effects of the sucrose-treatment on embryo viability; 3) to define morphological criteria for the sucrose-treated morula and to compare the efficacy of scoring the embryo in sucrose or in PBS, in terms of agreement among evaluators. MATERIALS AND METHODS In Vitro Embryo Production Cattle embryos were produced by routine in vitro maturation and fertilization techniques (11). Compact cumulus-occyte complexes (CGC) were selected after aspirating antral follicles (2 to 8 mm) on the surface of bovine ovaries derived from the abattoir. They were matured in groups of f 100 oocytes per well for 22 to 26 hours in 500 pl of maturation medium. The medium was TCM 199 bicarbonate buffered medium (Cat. No. 31150-014, Gibco BRL, Merelbeke, Belgium) supplemented with 20 % heat-inactivated estrous cow serum (v/v), 0.2 mM sodium pyruvate (Cat. No. P-3662, Sigma, Bomem, Belgium), 0.4 mM glutamine (Cat No. G-6392, Sigma) and 50 pg/l gentamycin sulphate (Cat No. 157 lo-03 1, Gibco BRL). The culture environment was 5% C@ in air, at 39°C and maximum humidity. Motile spermatozoa were separated from frozen-thawed semen by centrifugation at 100 g for 30 min at 25 “c on a discontinuous Percoll@ gradient (Cat. No. 17-0891-01, Pharmacia, Uppsala, Sweden) composed of 2 ml each of 90 and 45% Percoll solutions in a Hepes buffered salt solution. Gocytes were inseminated in 500 fl of Tyrode’s solution (TALP), with 6 m&nl BSA Fatty Acid Free (Cat. No. A-6003, Sigma), 20 pg./ml D-penicillamine (Cat. No. P-4875, Sigma), 10 pM hypotaurine (Cat. No. H-1384, Sigma), 1 pM adrenaline (Cat. No. E-4250, Sigma) and 25 pg/ml heparin (Cat. No. H-3149, Sigma). Gocyte sperm contact lasted for 20 to 24 h; the moment of first contact was designated 0 h post insemination (hpi). Maturation and fertilization media were not overlaid with oil. Presumed zygotes were subjected to vortexing for 1.5 mitt to remove adherent cumulus cells and spermatozoa. They were then washed in Hepes buffered TALP and transferred to 50 p.l droplets of oviduct co-culture, overlaid with paraffin oil (Cat. No. 1.07160/1000, MetckBelgolabo, Gverijse, Belgium). The medium used for embryo culture was Men&o-B2 (Cat.
1134
Theriogenology
No. ZA 146, INRA, Paris, France) supplemented with 10% estrous cow serum, 2.5 p&l fungizone (Cat. No. 15290-018, Gibco) and a suspension of bovine oviduct epithelial cells (BOEC; 11). Thme days post insemination, 50 pl of fresh medium were added to the culture droplets. Preparation of Sucrose Solution and PBS Sucrose (0.3 M) was dissolved in Embryo Transfer Freezing medium (Cat. No. 11 lOO021, Gibco), which is modified PBS supplemented with 0.33 mM sodium pyruvate, 5.6 mM glucose, 0.4 4%BSA and 0.025 mg/ml kanamycin sulphate and sterile filtered. Thus when PBS is mentioned in the text, it refers to the Embryo Transfer Freezing medium. The 0.3 M sucrose solution was frozen in tubes of 3 ml and kept at -2O’C until use. Selection of Morulae and Assessment of their Survival Embryos were observed at 120 to 138 hpi with a Wild Ml0 stereomicroscope at a magnification of x 60. Depending on the experiment, only excellent compact morulae were selected (Experiments 1 and 2) or excellent morulae together with substantially fragmented morulae (Experiment 3). Survival of embryos was assessed by blastocyst formation. A blastocyst was defined as an embryo with a visible blastococl lined by flattened cell(s). Mom discrete signs of possible delayed embryonic development were assessed by Hoechst staining of blastocysts. The blastocysts were fixed in 2% formaldehyde and 2% glutaraldehyde in PBS and subsequently stained with 10 @ml Hoechst 33342 in 2.3% sodium citrate supplemented with 25% ethanol, after counter staining with 0.1% trypan blue in 2.3% sodium citrate (19). The embryos were mounted in 100% glycerol (UCB, Leuven, Belgium) and the cell number was evaluated. Cells were counted using a Leitz Dialux 20 fluorescence microscope (with DAPI-filter). Experimental Design Exueriment 1. The effect of sucrose on morula volume was determined. Six selected morulae of excellent quality were put into isotonic PBS and micrographed as a control for initial embryo volume, by means of a camera connected to an inverted microscope (Leitz Fluovert) at a magnification of x 100. Then they were transferred into 20-pl droplets of 0.3 M sucrose under paraffii oil and micrographed after 15 set (if possible), 30 set, 45 set, 1 min, 1 min and 15 set, 1 ruin and 30 set, 1 min and 45 sec. 2 min, 2 min and 30 set, 3 min, 4 min and 5 min. The embryos were then placed into a 20-pl drop of PBS to recover and they were micrographed after the same time intervals. Changes in morula volume were measured by scanning the micrographs by means of a computer. Scanned images were analyzed with Claris CAD 2.0, and volumes of morulae were calculated based on the assumption that embryos were spherical. Volumes in sucmse were expressed as a percentage of the initial volume.
Theriogenology
1135
m The effect of sucrose on further embryo development was evaluated by transferring selected morulae (n=60) of excellent quality for 5 or 10 min in 0.3 M sucrose (sucrose-treatment) and allowing them to recover for an equal time period in PBS at room temperature. Then they were transferred again to the CR-incubator into Men&o BZdroplets supplemented with BOEC and cultured for another 36 to 48 h. To exclude possible negative effects of handling the embryos outside the incubator, a second group of morulae (n=60) were also put in PBS at room temperature for 10 or 20 min (sham-treatment), which was equal to the total time period outside the incubator of the sucrose-treated embryos. Then they were cultured again in the presence of 5% C@ and BOEC. A third group of morulae (n=30) was cultured undisturbed in M&&o-droplets supplemented with BOEC (control). Selected morulae of different replicates were randomly distributed among treatments. Evaluation of embryo development was performed both with respect to further blastocyst development and cell number at the blastocyst stage. Blastulation was recorded at 144 and 168 hpi. Newly formed blastocysts were removed from the culture medium and were fixed and stained with Hoechst 33342 for cell counts. The number of mitotic and pycnotic cells was also determined. In mitotic nuclei, the condensed chromosomes were clearly distinguishable. Dead or pycnotic nuclei were characterized by very dense chromatin or by dispersed nuclear fragments. These dead cells were not included in the total number of cells. Mitotic and pycnotic cell indices were calculated according to Hardy et al. (6) : Mitotic cell index = (No. of metaphases/total no. of cells ) x 100 Pycnotic cell index = ( No. of pycnotic cells/total no. of cells+no. of pycnotic cells)xlOO. The effect of prolonged sucrose treatment on cell lysis was also investigated : 70 morulae were incubated in groups of 10 for 0, 15, 30, 45, 60, 120 and 180 min in sucrose 0.3 M at room temperature and subsequently allowed to recover in PBS for 15 min. Then they were stained in Ml99 bicarbonate supplemented with propidium iodide 10 @nl for 30 min at 39’C. Propidium iodide is a DNA-specific red-pink fluorescent stain which only penetrates lysed cell membranes. Embryos were fixed for 5 min in 100% ethanol supplemented with Hoechst 33342, which stains the nuclei of both intact and lysed cells blue. Embryos were mounted in 100% glycerol, and pink and blue fluorescing nuclei were counted by means of a Dialux 20 fluorescence microscope. Pink nuclei were expressed as a percentage of the total number of nuclei for each embryo. Exoeriment 3, Criteria which were originally defined by Lindner and Wright (10) for flushed bovine embryos, were modified for use with sucrose-treated, in vitro produced bovine embryos. Only the percentage of fragmentation was taken into account. Embryos were assigned to 1 out of 4 categories, as shown in Figure 1: excellent: No fragmentation = an ideal embryo, spherical symmetrical (1A, 1B); good: Less than 25% fragmentation = extruded or degenerated cells occupy no more than one quarter of the embryo (1 large extruded blastomere or a few smaller ones; 2A, 2B); fair: Between 25 and 50% fragmentation = larger cell fragments may occupy up to one-half of the embryo, viable cell mass is still fairly large (3A, 3B): poor : More than 50% fragmentation = fragmentation exists in more than half of the space inside the zona, but a small viable appearing cell mass is present (4A, 4B).
1136
Figure 1. Morphology of embryos assigned to the 4 categories in PBS (A) or in sucrose (B) l= Excellent morula. 2= Good morula. 3= Fair morula. 4= Poor morula. Using the above criteria, 40 in vitro produced embryos were evaluated by 7 different evaluators at 138 to 140 hpi, first in PBS and then in sucrose. The number of embryos and evaluators which were selected was based upon the study of Farin et al. (3). Stage of the embryos was not included in the analysis, although a limited number of the embryos showed the earliest sign of cavitation. The selected embryos were evaluated for the percentage of agreement and for agreement beyond chance among evaluators (3). Momlae were numbered individually and scored in 25-N drops of PBS under paraffin oil by means of a stereo microscope Ml0 at x 80. To get a complete view, the embryo was rotated in the medium with a micropipette. Time allowed for evaluation was between 30 and 45 set per embryo. When the evaluators had graded all 40 embryos, the embryos were washed in 1 ml 0.3 M sucrose, coded at random, and transferred to alphabetically numbered droplets of 0.3 M sucrose to be reevaluated. Evaluators were not aware of the classification assigned to the embryos by the other evaluators, and they were not aware of their own previous classifications of embryos in PBS. After evaluation was complete, the embryos were washed in PBS and transferred individually to co-culture droplets. At 168 hpi, all the embryos were stained with Hoechst 33342 to evaluate cell number. Statistical Analysis Cell number data in Experiment 2 were analyzed using one-way ANOVA and the Least Significant Difference method for pairwise comparison of means with control treatment. Proportions were first submitted to arcsine transformation before performing one-way ANOVA (Stat&ix 4.1, Analytical Software, Tallahassee, FL). In Experiment 3, the percentage of
1137
Theriogenology
agreement [(observed number of agreement between all possible pairs of evaluators / number of possible pairs of evaluators) x 1oOJ and kappa coefficients among evaluators were calculated using the Generalized Kappa module (Crunch Software Corp., San Francisco, CA) (3)). Kappa was interpreted as poor (K < 0.40), fair to good (K = 0.40 to 0.75), or excellent ( K > 0.75) agreement beyond chance (5). Both unweighted and weighted kappa-coefficients were calculated. The weighted kappa coefftcient included the rate of disagreement among evaluators, e.g., for an embryo scored excellent (Grade 1) by evaluator x and scored poor (Grade 4) by evaluator y. a weighting factor of 4-l = 3 was used.
RESULTS Experiment 1 Six compact morulae which were put into 20-pl droplets of 0.3 M sucrose had reached the bottom of the petri dish within 50 to 60 sec. As shown in Figure 2, from 40 set of exposure to sucrose onwards, the mean volume of the morulae reached its minimal value. The mean volume of the morulae had shrunk to 75% of their original volume. When the shrunken morulae were placed into isotonic PBS, they did not regain their total volume even after 150 see: the embryonic cells occupied only 95% of the original volume at that time (mean values represent data from 4 of 6 morulae). Relative volume of embryos
150
Time (seconds) Figure 2. Mean volume changes of in vitro produced bovine morulae exposed in 0.3M sucrose (--) for 150 seconds and recovered in PBS (---) for 150 seconds (data presented for only 4 of 6 morulae).
1138
Theriogenology
Experiment 2 No effect of handling the embryos outside the incubator could be detected (shamtreatment: data not shown). Out of 90 compact morulae which were selected for evaluating the effect of short-term sucrose treatment, 82 (91%) developed to the blastocyst stage (Table 1). There was no significant difference in development to the blastocyst stage at Day 7 between embryos which were exposed for 0,5 or 10 min to sucrose 0.3 M; the rates ranged from 83 to 97% across the treatments. Table 1.
In vitro development of sucrose-treated and control embryos to the blastocyst stage. No. of Cumulative no. of blastocysts (8) compacted morulae at Day 6 at Day 7
Treatment
Sucrose 5 minutes 30 30 Sucrose 10 minutes Control 30 Total 90 Not significantly different (One-way ANOVA).
17 (57) 16 (53) 14 (47) 47 (52)
29 28 25 82
(97) (93) (83) (91)
The mean number of nuclei (USEM) for Day 6 blastocysts did not differ among treatments (Figure 3): the total mean cell number for Day 6 blastocysts was 72.M 3.6 with a range of 32 to 116 (F= 0.78, df = 2 and 44, P= 0.46), and for Day 7 blastocysts 100.5 + 5.7 with a range of 54 to 165 (F = 0.53, df = 2 and 32, P=O.59).
120 4
loo-
2 =I
80 -_
;
60-o
;-”
Figure 3.
40
-.
Day 6 Day 7 Mean total cell number (X + SEM) of sucrose-treated and control blastocysts. Not significantly different : One-way ANOVA.
Theriogenology
1139
The mean mitotic cell index and the pycnotic cell index were not significantly different between treatments at Day 6 (Figure 4). At Day 7, however, the number of pycnotic cells was significantly lower in embryos treated with sucrose for 10 mitt (F= 9.9, df = 2 and 32, P=O.OOOS).Moreover, when embryos were exposed for longer periods to sucrose 0.3 M (from 15 to 180 min), no immediate effect on cell lysis could be detected (data not shown). The highest number of lysed cells which could be detected in 1 embryo was at 120 mitt (5%, 3 cells lysed out of 57). However, at 120 and 180 min, a total of 5 of 21 embryos displayed nuclei with apparent signs of degeneration.
Figure 4.
Day 6 Day 7 Day 7 Day 6 Mean mitotic and pycnotic cell indexes (XzbSEM) of sucrose-treated and control blastocysts. Arcsine transformation, One-way ANOVA, LSD pairwise comparison of means a,b Values with different superscripts differ significantly (P=O.ooOS).
Experiment 3 When the modified criteria of Lindner and Wright were used to evaluate embryos in PBS or in 0.3 M sucrose, overall agreement among evaluators for classification of quality grade was poor for embryos evaluated in PBS (48.2%, K= 0.31) but improved when the same embryos were scored in 0.3 M sucrose (62.5%, t&.49; Table 2). Embryos with less then 25% fragmentation or between 25 and 50% fragmentation were more difficult to agree upon than embryos with no fragmentation or with more than 50% of the embryonic space occupied by fragments. In other words, the use of sucrose did not facilitate the distinction between good (Grade 2) and fair (Grade 3) embryos. When a weighted kappa-coefficient was calculated, which includes the level of disagreement, overall weighted percentages of agreement increased both for evaluation in PBS (77.9% , K =0&l) and in sucrose (85.6%, K =0.61). When the 7 evaluators were observed individually for the proportion of agreement with the other evaluators, 3 of them were found to have a one-tailed upper 95% confidence limit for index below 1. Two of these evaluators were not experienced with the morphological aspect of
1140
Thetiogenology
Table 2.
Agreement among 7 evaluators on the quality grade of in vitro produced bovine embryos scored either in PBS or in sucrose at Day 6 pi.
Agreement (%) and kappa coefficient (10~
Grade
Evaluation in PBS % K
Evaluation in sucrose % K
Excellent Good Fair Poor
80.0 68.3 67.9 80.2
0.45 0.26 0.15 0.39
88.3 71.0 75.2 90.5
0.64 0.38 0.35 0.68
0ve.ra.u Unweighted tcb
48.2
0.31
62.5
0.49
Weighted I&
77.9
0.44
85.6
0.61
a Kappa measures agreement beyond chance; a value of 1.0 indicates complete agreement among evaluators b Does not include level of disagreement. c Includes level of disagreement, does not apply for different grades. morulae produced by oviduct co-culture, as their embryo production system had used predominantly SOF-culture. The third evaluator was more involved with the selection of blastocysts than with evaluation of morulae. Omission of these evaluators led to a higher percentage of agreement and agreement beyond chance for embryos scored in both PBS and sucrose (Table 3). Again, the good and fair embryos were the most difficult to agree upon, and evaluation in sucrose did not improve evaluator agreement for those 2 grades of embryo quality, while agreement for excellent and poor embryos reached nearly excellent levels for the more experienced evaluators. To be able to control the accuracy of the morphological evaluation, the test embryos were divided into 4 groups, according to their number of nuclei determined at 7 dpi. According to this arbitrarily fmed “ideal score”, poor embryos should display a number of distinct nuclei ranging from none to no more than 25 nuclei; fair embryos from 26 to 50 nuclei; good embryos horn 51 to 75 nuclei and excellent embryos from 76 to 100 nuclei. The number of nuclei of the tested embryos ranged from no distinct nuclei (in 3 embryos) to 99 nuclei. Although an ideal score based solely upon cell number is a rigid way to correlate with embryo quality, it was the only measure we could use for statistical purposes. The “ideal score” of the test embryos was
1141
Theriogenology Table 3,
Agreement among the 4 most experienced evaluators on the quality grade of in vitro produced bovine embryos scored either in PBS or in sucrose at Day 6 pi.
Agreement (%) and kappa coefficient
Grade
Evaluation in PBS % K
(K)a
Evaluation in sucrose %
K
87.1 0.66 90.4 0.71 GOOd 76.7 0.42 74.6 0.49 Fair 82.9 0.54 86.7 0.54 Poor 80.8 0.46 90.8 0.72 overallb 63.8 0.52 71.3 0.60 a Kappa measures agreement beyond chance; a value of 1.0 indicates complete agreement among the 4 evaluators. b Unweighted K.
Excellent
added to Table 4, which also shows the relationship between mean total cell number of the embryos and embryo quality score given by the 7 evaluators. After linear regression of the scores versus total mean cell number, the regression coefficients of the ideal slope and of the scored slope by each evaluator were compared with a two sample t-test. Each evaluator differed significantly from the ideal slope, and the use of sucrose did not affect the accuracy of grading the embryo according to cell number. Evaluator 5 approached the ideal slope most, as well in PBS as in sucrose. However, neither method was useful for eliminating embryos with low cell numbers from Grade 1 and 2 categories. In fact, there was 1 embryo with only 10 cells which was graded as excellent by all 7 evaluators in PBS, and by 6 out of 7 evaluators in sucrose. The overall number of excellent graded embryos was lower in sucrose than in PBS. On the other hand, to avoid the assignment of embryos with high cell numbers to Grade 4, sucrose could be used since 3 embryos of 45, 74 and 96 cells respectively, which were scored by at least 1 evaluator to be of poor quality in PBS,were scored 1 or 2 categories higher in sucrose. DISCUSSION Under normal conditions, mammalian cells are bathed in physiological fluids, which are isosmotic to a sodium chloride concentration of 0.9%. In spite of this, cells am able to survive a&osmotic conditions to a certain extent. A cell with intact cell membranes which is exposed to a nonpermeable solute like sucrose will respond by losing water. The water leaves the cell in order to maintain equality between the chemical potentials of the intracellular and the extracellular water (14). The water loss inevitably evokes shrinkage of the cell mass. According to this principle sucrose can act as an osmotic counterforce to minimize swelling damage when glycerol-frozen embryos are thawed (9). Sucrose has also been used as a tool in the splitting of bovine morulae (1). And hypertonic sucrose can aid sperm injection or blastomere insertion by enlarging the perivitelliie space of oocytes (12,13,31-35). In the present study, sucroseinduced enlargement of the perivitelline space was used to facilitate morphological evaluation of in vitro produced bovine embryos at the morula stage.
1142
Table 4.
Theriogenology Total mean cell number of embryos graded into 4 categories by 7 different evaluators compared with the ideal score
Score Ideal score
Mean Cell Number (k SD) of embryos graded as Excellent Fair GOOd Poor 84.7rt9.3
69.7f 7.5
38.2k8.5
10.6rt9.3
Evaluator 1
PBS Sucrose
50.3f29.3 51.Lz+25.1
41.w32.0 52.6k31.3
42.4k34.1 39.1k29.4
13.0 f 10.9 8.9f10.3
Evaluator 2
PBS Sucrose
54.1f26.0 56.2f26.0
52.6226.9 41.2k32.7
41.2k33.9 43.2+28.2
16.4-2.0 10.7*11.1
Evaluator 3
PBS Sucrose
5 1.6f26.0 6O.ti28.3
46.1k26.0 51.1f29.1
44.4k33.9 38.5f30.0
13.3fll.5 9.5k11.7
Evaluator 4
PBS Sucrose
54.3k22.8 59.ti24.4
39.1f33.4 41.5f30.8
36.6f29.9 37.3k28.1
26.5k35.9 12.6f12.6
Evaluator 5
PBS Sucrose
58.1f24.5 55.lk24.2
44.2f29.9 57.9f28.6
37.3f30.9’ 32.1f27.3
26.3k31.6 15.5f22.2
Evaluator 6
PBS Sucrose
39.2f31.9 55.OZk30.5
46.5rt26.9 55.1f24.7
44.OLh37.3 30.0f29.1
14.3f17.1 3.25f3.4
Evaluator 7
PBS Sucrose
36.3f33.2 47.9f29.0
43.6f32.0 49.9f33.3
43.9k30.4 36.3f32.5
27.8ti9.3 15.7f14.4
In Experiment 1, it was demonstrated that in vitro produced bovine morulae are very responsive to a hypertonic treatment of 0.3 M sucrose in PBS. Within 1 min they had reached the bottom of a 20-pl drop, and within 40 set they had shrunken to about 73% of their original volume. This fast shrinkage rendered them morphologically almost identical to flushed cattle embryos, which occupy between 60 and 70% of the perivitelline space at the compact morula stage (10). The shrinkage is particularly noticeable in cells which are connected to one another, leaving extruded cells and cellular fragments clearly visible in the perivitelline space, thus facilitating the evaluation of morphology. If our technique is to be used for routine embryo evaluation, possible toxic effects of sucrose have to be excluded. It has been reported that very hypertonic sucrose solutions can cause membrane leakage in 8-cell mouse embryos (14). The same was observed in rabbit embryos where incubation of the embryo in 1.5 M sucrose in PBS for 30 min led to shrinkage followed by expansion in size. This size shift was not observed when the embryos were
Theriogenology
1143
incubated for 30 mitt in 0.5 M sucrose in PBS (34). Therefore, in our study, we chose to limit the sucrose concentration to 0.3 M and to limit the exposure time to 5 min, which is more than sufficient for thorough embryo evaluation. At this concentration and duration, no effect whatsoever on further embryonic development is expected. This was confirmed in Experiment 2, where development to the blastocyst stage was not adversely affected by sucrose treatment for up to 10 min. Normal total cell numbers of about 72 cells were found in Day 6 blastocysts, which is comparable with cell numbers found in flushed blastocysts (30). At Day 7 of development, both sucrose-treated and control blastocysts had a higher average cell number (=lOO) than reported for flushed embryos (=75; 30). Cell division, which is expressed by the presence of mitotic cells, continued for at least 2 d after sucrose treatment, and was not significantly different from that of the controls (Figure 3). We can conclude that sucrose does not slow down embryo development. Moreover, it does not increase cell death, as a lower pycnotic index was observed after a 10 min treatment with sucrose 0.3 M. On the other hand, it can be expected that sucrose has a negative influence on the cellular membrarte by the osmotic changes that are induced. Prolonged treatment with sucrose did not influence the percentage of lysed cells, although some embryos displayed degenerative nuclei at 120 and 180 min. In Experiment 3, embryos were graded into 4 morphological groups. These groups are basically the same as in the classification system of Lindner and Wright (lo), but only a degree of fragmentation is taken into account. Fragmentation is a common feature in in vitro culture of mammalian embryos : in culture systems without somatic cells it seems to occur at earlier stages of development than in somatic cell co-culture systems. Human embryos, for instance, are scored for fragmentation at the 2- to 6-tell stage (17), while sheep embryos which have been cultured in SOF-medium can already display substantial fragmentation at the 4-cell stage (28). It has been shown that cytoplasmic fragmentation can be reduced by adding somatic cells to the culture system (28,29). With the co-culture methods used in the present study, fragmentation of the cytoplasm in viable bovine embryos was not obvious until the morula stage, but then it became very difficult to grade the amount of fragmentation due to the lack of compaction. It is generally recognized that bovine embryos produced in co-culture look less compact than, for instance, embryos produced in mSOF (23,24). After artificially shrinking the embryo in sucrose, the aspect of the in vitro produced morula very much resembles the flushed compact morula. Morula evaluation has become much more easy to perform and can be done in a more objective way and by someone who is not acquainted with the less compact morphology of the co-cultured embryos, such as those used in our study. The same should be applicable to in vitro produced blastocysts, in which perivitelline space is totally absent. By using sucrose, the amount of fragmentation in the perivitelline space can be visualized and quantified. Evaluators agreed more often when they evaluated the quality of in vitro produced embryos in sucrose than in PBS. It has been shown that agreement among evaluators is higher for in vivo produced embryos than for in vitro produced embryos, where the stage of development is concerned (3). In our study, no evaluation of development was performed, although some embryos already showed the slightest sign of blastocoele formation. Only quality grade as assessed by amount of fragmentation was scored. By using sucrose to improve
Theriogenology
1144
the visualization of cellular fragments in the perivitelline space, we were able to show that agreement for quality grade, which is known to be moderate for in vitro produced embryos from this study (48.21, K= 0.31) and also from Farins study (57.7 8, K= 0.42, 3 ), was increased by evaluating the embryo in sucrose (62.5%
K=
0.49). Agieement beyond chance
was good for excellent (K= 0.64) and for poor (K = 0.68) embryos and was inclined to be excellent for the mom experienced evaluators. However, in our study, poor embryos or Grade 4 embryos did not resemble completely degenerated embryos but rather embryos with a small viable cell mass and severe fragmentation. Such embryos are not completely void of developmental capacity (26). Farin et al. (3) regarded Grade 4 embryos as degenerated and Grade 2 and 3 as morphologically defected. They found a significant improvement (approximately 20%) in evaluator agreement (78.7%, K= 0.60) when they combined Grade 2 and 3 embryos into a single grade (=embryos with morphological defects). Moreover, they suggest that a less complicated 3-point grading scheme can increase the level of agreement among graders when maximum agreement between evaluators is important. In our study, we were able to increase the level of agreement for Grade 4 embryos by using sucrose. The Grade 4 embryos in our study may be comparable to Grade 3 embryos in the Fat-in study (3) since this grade also represents embryos with many extruded cells and with debris. Subjectivity in morphological evaluation of in viva-produced cattle embryos was also found by van Wagtendonk-De Leeuw (27). She tested 29 persons from 4 ET-teams for their ability to score 18 frozen-thawed, in vivo-produced bovine embryos according to Lindner and Wright (10). In the study, one team graded embryos, in average, half an embryo grade higher than the other teams, and 5 individuals deviated substantially from the others in embryo grading. Uniformity in embryo grading is important when embryo grading results are correlated with survival of freezing and thawing, pregnancy rates and other factors. In evaluating embryo grade, clear definitions of quality grade should be explained to all evaluators to avoid, for example, some evaluators focusing on inclusion of lipid droplets into the cytoplasm or the estimated cell number of the embryo, while others do not. The evaluators in our study were asked to exclusively take the amount of fragmentation into account when scoring the embryos. Different interpretations by each evaluator of specific defects (3) or lack of familiarity with the morphology of in vitro cocultured bovine embryos may be the cause of deviation in embryo grading. A more consistent embryo grading system may be provided by the 3-scale grading scheme, as proposed by Farin et al. (3), evaluating the embryos in sucrose (this study), or by a combination of both. The use of embryo morphology for assessment of embryo quality remains questionable. Although there was a tendency to allocate the embryos with higher cell numbers to the excellent and good categories, no evaluator was competent to grade the embryos according to an idealized cell number standard score, neither in PBS nor in sucrose. However, it was not the purpose of the morphological evaluation to estimate the number of cells of the embryo, rather the amount of fragmentation of each embryo. Cell number was used as an objective quality standard by lack of other criteria.
1145
Theriogenology
In conclusion, it can be stated that in vitro produced bovine morulae shrink very rapidly in 0.3 M sucrose, yielding improved visualization of the cell fragments. Limited exposure (up to 10 min) to 0.3 M sucrose does not impair embryo viability in terms of subsequent blastocyst formation and cell number. The use of sucrose in embryo evahration improves agreement among evaluators and can be of value under circumstances which require a standardized scoring system by different individuals, such as embryo transfer experiments under field conditions. REFERENCES 1. Breclbacka P. Factors affecting cell viability during bisection of bovine embryos. Theriogenology 1995;44: 159- 166. 2. Butler JE, Biggers ID. Assessing the viability of preimplantation embryos in vitro. Thetiogenology 1989;31: 115126. 3. Farin PW, Britt JI-I, Shaw DW, Slenning BD. Agreement among evaluators of bovine embryos produced in vivo or in vitro. Theriogenology 1995;44:339-350. 4. Farin PW, Farin CE. Bovine embryos produced in vivo or in vitro: embryo survival and fetal development. Biol Reprod 1995;52:676-682. 5. Fleiss JL. Statistical methods for rates and proportions. New York: John Wiley 8c Sons, Inc 1981;212-236. 6. Hardy K, Handyside AH, Winstons RML. The human blastocyst: cell number, death and allocation during late preimplantation development in vitro. Development 1989; 107:597604. 7. Hasler JF, Henderson WB, Hurtgen PJ, Jin ZQ, McCauley AD, Mower SA, Neely B, Shuey LS, Stokes JE, Trimmer SA. Production, freezing and transfer of bovine IVFembryos and subsequent calving results. Theriogenology 1995;43:141-152. 8. Hasler IF, McCauley AD, Lathrop WF, Foote RH. Effect of donor-embryo recipient interactions on pregnancy rate in a large scale bovine embryo transfer program. Theriogenology 1987;27: 139- 167. 9. Leibo SP. Cryobiology: preservation of mammalian embryos. In: JW Evans, Hollaender A (eds) Genetic Engineering of Animals. New York and London: Plenum Press, 1986; 251272. 10. Lindner GM, Wright RW Jr. Bovine embryo morphology and evaluation. Theriogenology 1983;20:407-416. 11. Mahmoudzadeh AR, Van Soom A, Bols P, Ysebaert MT, de Kruif A. Optimization of a simple vitrification procedure for bovine embryos produced in vitro : effect of developmental stage, twostep addition of cryoprotectant and sucrose dilution on embryonic survival. J Reprod Fertil 1995;103:33-30. 12. Malter HE, Cohen J. Blastocyst formation and hatching in vitro following zona drilling of mouse and human embryos. Gam Res 1989,24:67-80. 13. Malter HE, Cohen J. Partial ulna dissection of the human oocyte : a nontraumatic method using micromanipulation to assist zona pellucida penetration. Fertil Steril 1989,51:139148.
1146 14.
Theriogenology
Mazur P, Schneider U. Osmotic responses of preimplantation mouse and bovine embryos and their cryobiological implications. Cell Biophys 1986;8:259-285. 15. Nagakawa A, Suzuki T, Suga T, Inaue T, Takahashi Y, Kanagewa H. Morphology and size of embryo recovered from superovulated cows. J Vet Med Sci 1991;53:287-290. 16. Overstrom EW. In vitro assessment of embryo viability. Theriogenology 1996; 45:3-16. 17. Pellestor F, Girardet A, And& B, Amal F, Humeau C. Relationship between morphology and chromosomal constitution in human preimplantation embryo. Mol Reprod Dev 1994;39:141-146. 18. Pratt HPM, Ziomek CA, Reeve WJD, Johnson MH. Compaction of the mouse embryo: an analysis of its components. J Embryo1 Exp Morph01 1982;70: 113- 132. 19. Purse1 VG, Wall RJ, Rexroad CE, Hammer RE, Brinster RL. A rapid whole-mount staining procedure for nuclei of mammalian embryos. Theriogenology 1985;24:687-700. 20. Reichenbach HD, Liebrich J, Berg U, Brem G. Pregnancy rates and births after unilateral or bilateral transfer of bovine embryos produced in vitro. J Reprod Fertil 1992;95:363370. 21. Shamsuddin M, Larsson B, Gustaffson H, Gustari S, Bartolome J, Rodriguez-Martinez H. Comparative morphological evaluation of in vivo and in vitro produced bovine embryos. Proc 12th Int Congr Anim Reprod 1992;3:1333-1335. 22. Shea BF. Evaluating the bovine embryo. Theriogenology 1981;15:31-43. 23. Van Langendonckt A, Lebrun N, Grisart B, Massip A, Dessy F. Cinematographic analysis of early bovine embryo development in synthetic oviduct fluid medium. Pnx 4th Int Symp Reprod Dom Rumin, 1994, Townsville, Australia. 24. Van Soom A, de Kruif A. A comparative study of in vivo and in vitro derived bovine embryos. Proc 12th Int Congr Anim Reprod 1992; 3:1365-1365. 25. Van Soom A, Mijten P, Van Vlaenderen I, Van den Branden J, Mahmoudzadeh AR, de Kruif A. Birth of double muscled Belgian blue calves after transfer of in vitro produced embryos in to dairy cattle. Theriogenology 1994;41:855- 867. 26. Van Soom A, Vanroose G, Bols PET, de Kruif A. Evaluation of morphology of in vitro produced bovine embryos after sucrose treatment : preliminary results. 1le Reunion A.E.T.E.,1995:252 abstr. 27. van Wagtendonk-de Leeuw AM. Evaluation of uniformity among persons in embryo grading from video recordings. Theriogenology 1996;45:230 abstr. 28. Walker SK, Heard TM, Seamark RF. In vitro culture of sheep embryos without co-culture: successes and perspectives. Theriogenology 1992;37: 11 l- 126. 29. Wiemer KE, Hoffman DI, Maxson WS, Eager S, Muhlberger B, Fiore I, Cuetvo M. Embryonic morphology and rate of implantation of human embryos following co-culture on bovine oviductal epithelial cells. Hum Reprod 1993;8:97-101. 30. Wthth YA, van der Zee-Kotting WI, Dieleman SJ, Bevers MM, Kruip TAM. Presence of mitotic cells : a parameter of embryo quality . Anim Repmd Sci 1994;35: 173-182. 31. Yang X, Chen J, Chen Y, Foote RH. Survival of rabbit eggs shrunken to aid in sperm microinjection. Theriogenology 1988;29:336 abstr. 32. Yang X, Chen Y, Chen J, Foote RH. Blastocyst development from rabbit ova fertilized by injected sperm. J Reprod Fertil 1988; 1: 13 abstr.
Theriogenology 33.
34.
35.
1147
Yang X, Chen Y, Chen J, Foote RI-I. Improved developmental potential of rabbit ova fertilized by sperm injection into the perivitelline space enlarged by hypertonic media. J Exp Zoo1 1990;255:114-119. Yang X, Chen Y, Chen J, Foote RI-I. Potential of hypertonic medium treatment for embryo micromanipulation : I. Survival of rabbit embryos in vitro and in vivo following sucrose treatment. Mol Reprod Dev 1990,27:110-l 17. Yang X, Zhang L, Fovacs A, Tobback C, Foote RH. Potential of hypertonic medium treatment for embryo micromanipulation : II. Assessment of nuclear transplantation methodology, isolation, subzona insertion and electrofusion of blastomeres to intact or functionally enucleated oocytes in rabbits. Mol Reprod Dev 1990;27: 118- 129.
SUCROSE-INDUCED SHRINKAGE OF IN VITRO PRODIJCED BOVINE MORULAE: EFFECI ON VIABILITY, MORPHOLOGY AND EASE OF EVALUATION
A. Van Soom,’ M.-T. Ysebaert,z A. Vanhoucke-De Medts,’ A. Van de Velde,’ S. Metton, A. DelvalP A. Van Langendonckt,S I. Donnay, G. Vanroose,t P.E.J. Bolst and A. de Kruift ‘Department of Reproduction, Obstetrics and Herd Health and ;?Deparmrent of Physiology, Biochemistry and Biostatistics, University of Gent, B-9000 Gent, Belgium, 3Holland Genetics, IVP Laboratory, 5437 BG Beers, The Netherlands 41RSIA Research Unit, Sat-t-Tilman, B-4000 Liege, Belgium, SCatholic University of Louvain, Veterinary Unit, B- 1348 Louvain-la-Neuve, Belgium Received for publication : February 8, 1996 Accepted : May 10, 1996 ABSTRACT Sucrose (0.3 M) was used to cause artificial compaction of the embryonic cell mass of in vitro produced bovine embryos to facilitate morphological evaluation. Embryos were produced using routine in vitro maturation (IVM) and fertilization (IVF) techniques. The time necessary to induce shrinkage in 0.3 M sucrose to 75% of the original volume of Day 5 morulae was found to be less than 1 min, and 95% of the volume was regained in PBS after 2.5 min. No detrimental effect was observed after a 5 to IO-min sucrose treatment on subsequent blastocyst formation at Days 6 and 7 (P> 0.05). Furthermore, no significant differences were observed in the total number of cells, or in the mitotic and pycnotic cell index of blastocysts in different treatment groups. Agreement among 7 evaluators grading 40 Day 6 embryos was examined using the kappa coefficient of agreement (tc). Overall agreement among evaluators for classification of quality grade was poor (48.2 %,
K =
0.31) for embryos evaluated in PBS, but
the rate improved when the same embryos were scored in sucrose (62.5 %, K = 0.49). Evaluating less compact in vitro produced bovine morulae in sucrose increases agreement among evaluators, since embryos in sucrose mimick the appearance of in vivo produced embryos. Thus, we conclude that scoring in vitro produced embryos in sucrose improves agreement among evaluators. Key words: bovine morula compaction, sucrose, embryo morphology
Acknowledgments The authors thank R. Verbeke for computing embryo volumes and Dr. W. Vanneste for performing the Kappa statistics. We also thank A.M. van Wagtendonk-De Leeuw for critical discussions on statistical methods. This study was subsidized by an IWONLCSVH (section 1, nr 5597A) grant.
Theriogenology 46:1131-1147. 1996 0 1996 by Elsevier Science inc.
0093-691 X/96/$1 5.00 PII SOO93-691X(96)00285-9
1132
Theriogenology INTRODUCTION
In recent years, it has become possible to produce bovine embryos both by in vivo and in vitro methods. With growing interest in embryo production, it has become clear that embryos generated in vitro are different from flushed embryos, both with respect to their morphology at the light microscopical level (24) and at the ultrastructuml level (21). Although in vitro produced bovine embryos can yield satisfactory ptegnancy rates after transfer, fetal loss during pregnancy is higher than with in vivo embryos (4,20,25). This pregnancy loss is costly, especially when the number of available recipients is limited, and a reliable method for assessing embryo viability prior to transfer would be of great value. Although a number of different viability tests have been described for use in bovine embryo transfer (2), to date no extended studies have been published to relate the outcome of methods like measurement of metabolic activity, use of fluorescent probes or determination of intracellular pH with the actual number of pregnancies which can be achieved after transferring such tested embryos (16). Thus we used morphological parameters, which have been widely used as a noninvasive method, to score the quality of flushedbovine embryos (10,22). The advantages of morphological scoring tests are that they are quick, noninvasive and require readily available equipment. The disadvantages are that they are a subjective measure and they require highly trained personnel (2). Morphological criteria for assessment of embryo quality of in vivo produced embryos have been defined by Lindner and Wright (10). Moreover, these criteria have been shown to be related to pregnancy rates after transfer (8,lO). Morulae flushed out of cows 6 d post insemination am characterized by a very obvious compaction, which is defined as the coalescing of individual blastomeres. The resulting embryonic mass occupies only 60 to 70% of the space inside the zona pellucida (10). Therefore the compact morula stage is particularly suited for evaluation of the presence of extruded blastomeres and degenerated cells or fragments, since they are clearly visible in the perivitelline space. The space between zona and cells has an average width of 9.1 inn at that time, compared with 6.9 km for the 16cell embryo and only 2.6 pm for the blastocyst (15). In vitro produced morulae do not have a distinct increase in the width of the perivitelline space at the time of compaction (24). Compaction of in vitro produced embryos can only be assessed by the flattening of cells and cell outlines that become invisible, as defied for the mouse embryo by Pratt (18). Consequently, morphological evaluation of in vitro produced morulae is more difficult due to this lack of compaction (7). This has been confirmed in a recent study by Farin et al. (3), in which they compared agreement among evaluators of bovine embryos produced in vivo and in vitro. They showed that agreement for developmental stage was very good with embryos produced in vivo and only fair for embryos produced in vitro. For classification of quality grade, however, agreement was only fair to poor, regardless of whether in vivo or in vitro produced embryos were used.
1133
Theriogenology
In the present study, we undertook to facilitate quality assessment of in vitro produced bovine morulae by incubating the embryo in 0.3 M sucrose. This method offers the possibility for making embryo morphology scoring easier. The sucrose-induced shrinkage technique was first reported by Yang et al. (31,32). It is a method for enlarging the perivitelliie space of rabbit oocytes, which has made sperm microinjection under the zona more feasible. Sucrose-tolerance of rabbit zygotes and embryos at different stages of development was also reported with the purpose of performing micromanipulations such as microinjection and nuclear transplantation (33-35). The possible use of sucrose for morphological evaluation of in vitro produced bovine embryos has not been reported. The objectives of this study were 1) to define the time necessary for morula shrinkage and recovery and to measure the change in embryo volume after incubation in 0.3 M sucrose; 2) to evaluate possible negative effects of the sucrose-treatment on embryo viability; 3) to define morphological criteria for the sucrose-treated morula and to compare the efficacy of scoring the embryo in sucrose or in PBS, in terms of agreement among evaluators. MATERIALS AND METHODS In Vitro Embryo Production Cattle embryos were produced by routine in vitro maturation and fertilization techniques (11). Compact cumulus-occyte complexes (CGC) were selected after aspirating antral follicles (2 to 8 mm) on the surface of bovine ovaries derived from the abattoir. They were matured in groups of f 100 oocytes per well for 22 to 26 hours in 500 pl of maturation medium. The medium was TCM 199 bicarbonate buffered medium (Cat. No. 31150-014, Gibco BRL, Merelbeke, Belgium) supplemented with 20 % heat-inactivated estrous cow serum (v/v), 0.2 mM sodium pyruvate (Cat. No. P-3662, Sigma, Bomem, Belgium), 0.4 mM glutamine (Cat No. G-6392, Sigma) and 50 pg/l gentamycin sulphate (Cat No. 157 lo-03 1, Gibco BRL). The culture environment was 5% C@ in air, at 39°C and maximum humidity. Motile spermatozoa were separated from frozen-thawed semen by centrifugation at 100 g for 30 min at 25 “c on a discontinuous Percoll@ gradient (Cat. No. 17-0891-01, Pharmacia, Uppsala, Sweden) composed of 2 ml each of 90 and 45% Percoll solutions in a Hepes buffered salt solution. Gocytes were inseminated in 500 fl of Tyrode’s solution (TALP), with 6 m&nl BSA Fatty Acid Free (Cat. No. A-6003, Sigma), 20 pg./ml D-penicillamine (Cat. No. P-4875, Sigma), 10 pM hypotaurine (Cat. No. H-1384, Sigma), 1 pM adrenaline (Cat. No. E-4250, Sigma) and 25 pg/ml heparin (Cat. No. H-3149, Sigma). Gocyte sperm contact lasted for 20 to 24 h; the moment of first contact was designated 0 h post insemination (hpi). Maturation and fertilization media were not overlaid with oil. Presumed zygotes were subjected to vortexing for 1.5 mitt to remove adherent cumulus cells and spermatozoa. They were then washed in Hepes buffered TALP and transferred to 50 p.l droplets of oviduct co-culture, overlaid with paraffin oil (Cat. No. 1.07160/1000, MetckBelgolabo, Gverijse, Belgium). The medium used for embryo culture was Men&o-B2 (Cat.
1134
Theriogenology
No. ZA 146, INRA, Paris, France) supplemented with 10% estrous cow serum, 2.5 p&l fungizone (Cat. No. 15290-018, Gibco) and a suspension of bovine oviduct epithelial cells (BOEC; 11). Thme days post insemination, 50 pl of fresh medium were added to the culture droplets. Preparation of Sucrose Solution and PBS Sucrose (0.3 M) was dissolved in Embryo Transfer Freezing medium (Cat. No. 11 lOO021, Gibco), which is modified PBS supplemented with 0.33 mM sodium pyruvate, 5.6 mM glucose, 0.4 4%BSA and 0.025 mg/ml kanamycin sulphate and sterile filtered. Thus when PBS is mentioned in the text, it refers to the Embryo Transfer Freezing medium. The 0.3 M sucrose solution was frozen in tubes of 3 ml and kept at -2O’C until use. Selection of Morulae and Assessment of their Survival Embryos were observed at 120 to 138 hpi with a Wild Ml0 stereomicroscope at a magnification of x 60. Depending on the experiment, only excellent compact morulae were selected (Experiments 1 and 2) or excellent morulae together with substantially fragmented morulae (Experiment 3). Survival of embryos was assessed by blastocyst formation. A blastocyst was defined as an embryo with a visible blastococl lined by flattened cell(s). Mom discrete signs of possible delayed embryonic development were assessed by Hoechst staining of blastocysts. The blastocysts were fixed in 2% formaldehyde and 2% glutaraldehyde in PBS and subsequently stained with 10 @ml Hoechst 33342 in 2.3% sodium citrate supplemented with 25% ethanol, after counter staining with 0.1% trypan blue in 2.3% sodium citrate (19). The embryos were mounted in 100% glycerol (UCB, Leuven, Belgium) and the cell number was evaluated. Cells were counted using a Leitz Dialux 20 fluorescence microscope (with DAPI-filter). Experimental Design Exueriment 1. The effect of sucrose on morula volume was determined. Six selected morulae of excellent quality were put into isotonic PBS and micrographed as a control for initial embryo volume, by means of a camera connected to an inverted microscope (Leitz Fluovert) at a magnification of x 100. Then they were transferred into 20-pl droplets of 0.3 M sucrose under paraffii oil and micrographed after 15 set (if possible), 30 set, 45 set, 1 min, 1 min and 15 set, 1 ruin and 30 set, 1 min and 45 sec. 2 min, 2 min and 30 set, 3 min, 4 min and 5 min. The embryos were then placed into a 20-pl drop of PBS to recover and they were micrographed after the same time intervals. Changes in morula volume were measured by scanning the micrographs by means of a computer. Scanned images were analyzed with Claris CAD 2.0, and volumes of morulae were calculated based on the assumption that embryos were spherical. Volumes in sucmse were expressed as a percentage of the initial volume.
Theriogenology
1135
m The effect of sucrose on further embryo development was evaluated by transferring selected morulae (n=60) of excellent quality for 5 or 10 min in 0.3 M sucrose (sucrose-treatment) and allowing them to recover for an equal time period in PBS at room temperature. Then they were transferred again to the CR-incubator into Men&o BZdroplets supplemented with BOEC and cultured for another 36 to 48 h. To exclude possible negative effects of handling the embryos outside the incubator, a second group of morulae (n=60) were also put in PBS at room temperature for 10 or 20 min (sham-treatment), which was equal to the total time period outside the incubator of the sucrose-treated embryos. Then they were cultured again in the presence of 5% C@ and BOEC. A third group of morulae (n=30) was cultured undisturbed in M&&o-droplets supplemented with BOEC (control). Selected morulae of different replicates were randomly distributed among treatments. Evaluation of embryo development was performed both with respect to further blastocyst development and cell number at the blastocyst stage. Blastulation was recorded at 144 and 168 hpi. Newly formed blastocysts were removed from the culture medium and were fixed and stained with Hoechst 33342 for cell counts. The number of mitotic and pycnotic cells was also determined. In mitotic nuclei, the condensed chromosomes were clearly distinguishable. Dead or pycnotic nuclei were characterized by very dense chromatin or by dispersed nuclear fragments. These dead cells were not included in the total number of cells. Mitotic and pycnotic cell indices were calculated according to Hardy et al. (6) : Mitotic cell index = (No. of metaphases/total no. of cells ) x 100 Pycnotic cell index = ( No. of pycnotic cells/total no. of cells+no. of pycnotic cells)xlOO. The effect of prolonged sucrose treatment on cell lysis was also investigated : 70 morulae were incubated in groups of 10 for 0, 15, 30, 45, 60, 120 and 180 min in sucrose 0.3 M at room temperature and subsequently allowed to recover in PBS for 15 min. Then they were stained in Ml99 bicarbonate supplemented with propidium iodide 10 @nl for 30 min at 39’C. Propidium iodide is a DNA-specific red-pink fluorescent stain which only penetrates lysed cell membranes. Embryos were fixed for 5 min in 100% ethanol supplemented with Hoechst 33342, which stains the nuclei of both intact and lysed cells blue. Embryos were mounted in 100% glycerol, and pink and blue fluorescing nuclei were counted by means of a Dialux 20 fluorescence microscope. Pink nuclei were expressed as a percentage of the total number of nuclei for each embryo. Exoeriment 3, Criteria which were originally defined by Lindner and Wright (10) for flushed bovine embryos, were modified for use with sucrose-treated, in vitro produced bovine embryos. Only the percentage of fragmentation was taken into account. Embryos were assigned to 1 out of 4 categories, as shown in Figure 1: excellent: No fragmentation = an ideal embryo, spherical symmetrical (1A, 1B); good: Less than 25% fragmentation = extruded or degenerated cells occupy no more than one quarter of the embryo (1 large extruded blastomere or a few smaller ones; 2A, 2B); fair: Between 25 and 50% fragmentation = larger cell fragments may occupy up to one-half of the embryo, viable cell mass is still fairly large (3A, 3B): poor : More than 50% fragmentation = fragmentation exists in more than half of the space inside the zona, but a small viable appearing cell mass is present (4A, 4B).
1136
Figure 1. Morphology of embryos assigned to the 4 categories in PBS (A) or in sucrose (B) l= Excellent morula. 2= Good morula. 3= Fair morula. 4= Poor morula. Using the above criteria, 40 in vitro produced embryos were evaluated by 7 different evaluators at 138 to 140 hpi, first in PBS and then in sucrose. The number of embryos and evaluators which were selected was based upon the study of Farin et al. (3). Stage of the embryos was not included in the analysis, although a limited number of the embryos showed the earliest sign of cavitation. The selected embryos were evaluated for the percentage of agreement and for agreement beyond chance among evaluators (3). Momlae were numbered individually and scored in 25-N drops of PBS under paraffin oil by means of a stereo microscope Ml0 at x 80. To get a complete view, the embryo was rotated in the medium with a micropipette. Time allowed for evaluation was between 30 and 45 set per embryo. When the evaluators had graded all 40 embryos, the embryos were washed in 1 ml 0.3 M sucrose, coded at random, and transferred to alphabetically numbered droplets of 0.3 M sucrose to be reevaluated. Evaluators were not aware of the classification assigned to the embryos by the other evaluators, and they were not aware of their own previous classifications of embryos in PBS. After evaluation was complete, the embryos were washed in PBS and transferred individually to co-culture droplets. At 168 hpi, all the embryos were stained with Hoechst 33342 to evaluate cell number. Statistical Analysis Cell number data in Experiment 2 were analyzed using one-way ANOVA and the Least Significant Difference method for pairwise comparison of means with control treatment. Proportions were first submitted to arcsine transformation before performing one-way ANOVA (Stat&ix 4.1, Analytical Software, Tallahassee, FL). In Experiment 3, the percentage of
1137
Theriogenology
agreement [(observed number of agreement between all possible pairs of evaluators / number of possible pairs of evaluators) x 1oOJ and kappa coefficients among evaluators were calculated using the Generalized Kappa module (Crunch Software Corp., San Francisco, CA) (3)). Kappa was interpreted as poor (K < 0.40), fair to good (K = 0.40 to 0.75), or excellent ( K > 0.75) agreement beyond chance (5). Both unweighted and weighted kappa-coefficients were calculated. The weighted kappa coefftcient included the rate of disagreement among evaluators, e.g., for an embryo scored excellent (Grade 1) by evaluator x and scored poor (Grade 4) by evaluator y. a weighting factor of 4-l = 3 was used.
RESULTS Experiment 1 Six compact morulae which were put into 20-pl droplets of 0.3 M sucrose had reached the bottom of the petri dish within 50 to 60 sec. As shown in Figure 2, from 40 set of exposure to sucrose onwards, the mean volume of the morulae reached its minimal value. The mean volume of the morulae had shrunk to 75% of their original volume. When the shrunken morulae were placed into isotonic PBS, they did not regain their total volume even after 150 see: the embryonic cells occupied only 95% of the original volume at that time (mean values represent data from 4 of 6 morulae). Relative volume of embryos
150
Time (seconds) Figure 2. Mean volume changes of in vitro produced bovine morulae exposed in 0.3M sucrose (--) for 150 seconds and recovered in PBS (---) for 150 seconds (data presented for only 4 of 6 morulae).
1138
Theriogenology
Experiment 2 No effect of handling the embryos outside the incubator could be detected (shamtreatment: data not shown). Out of 90 compact morulae which were selected for evaluating the effect of short-term sucrose treatment, 82 (91%) developed to the blastocyst stage (Table 1). There was no significant difference in development to the blastocyst stage at Day 7 between embryos which were exposed for 0,5 or 10 min to sucrose 0.3 M; the rates ranged from 83 to 97% across the treatments. Table 1.
In vitro development of sucrose-treated and control embryos to the blastocyst stage. No. of Cumulative no. of blastocysts (8) compacted morulae at Day 6 at Day 7
Treatment
Sucrose 5 minutes 30 30 Sucrose 10 minutes Control 30 Total 90 Not significantly different (One-way ANOVA).
17 (57) 16 (53) 14 (47) 47 (52)
29 28 25 82
(97) (93) (83) (91)
The mean number of nuclei (USEM) for Day 6 blastocysts did not differ among treatments (Figure 3): the total mean cell number for Day 6 blastocysts was 72.M 3.6 with a range of 32 to 116 (F= 0.78, df = 2 and 44, P= 0.46), and for Day 7 blastocysts 100.5 + 5.7 with a range of 54 to 165 (F = 0.53, df = 2 and 32, P=O.59).
120 4
loo-
2 =I
80 -_
;
60-o
;-”
Figure 3.
40
-.
Day 6 Day 7 Mean total cell number (X + SEM) of sucrose-treated and control blastocysts. Not significantly different : One-way ANOVA.
Theriogenology
1139
The mean mitotic cell index and the pycnotic cell index were not significantly different between treatments at Day 6 (Figure 4). At Day 7, however, the number of pycnotic cells was significantly lower in embryos treated with sucrose for 10 mitt (F= 9.9, df = 2 and 32, P=O.OOOS).Moreover, when embryos were exposed for longer periods to sucrose 0.3 M (from 15 to 180 min), no immediate effect on cell lysis could be detected (data not shown). The highest number of lysed cells which could be detected in 1 embryo was at 120 mitt (5%, 3 cells lysed out of 57). However, at 120 and 180 min, a total of 5 of 21 embryos displayed nuclei with apparent signs of degeneration.
Figure 4.
Day 6 Day 7 Day 7 Day 6 Mean mitotic and pycnotic cell indexes (XzbSEM) of sucrose-treated and control blastocysts. Arcsine transformation, One-way ANOVA, LSD pairwise comparison of means a,b Values with different superscripts differ significantly (P=O.ooOS).
Experiment 3 When the modified criteria of Lindner and Wright were used to evaluate embryos in PBS or in 0.3 M sucrose, overall agreement among evaluators for classification of quality grade was poor for embryos evaluated in PBS (48.2%, K= 0.31) but improved when the same embryos were scored in 0.3 M sucrose (62.5%, t&.49; Table 2). Embryos with less then 25% fragmentation or between 25 and 50% fragmentation were more difficult to agree upon than embryos with no fragmentation or with more than 50% of the embryonic space occupied by fragments. In other words, the use of sucrose did not facilitate the distinction between good (Grade 2) and fair (Grade 3) embryos. When a weighted kappa-coefficient was calculated, which includes the level of disagreement, overall weighted percentages of agreement increased both for evaluation in PBS (77.9% , K =0&l) and in sucrose (85.6%, K =0.61). When the 7 evaluators were observed individually for the proportion of agreement with the other evaluators, 3 of them were found to have a one-tailed upper 95% confidence limit for index below 1. Two of these evaluators were not experienced with the morphological aspect of
1140
Thetiogenology
Table 2.
Agreement among 7 evaluators on the quality grade of in vitro produced bovine embryos scored either in PBS or in sucrose at Day 6 pi.
Agreement (%) and kappa coefficient (10~
Grade
Evaluation in PBS % K
Evaluation in sucrose % K
Excellent Good Fair Poor
80.0 68.3 67.9 80.2
0.45 0.26 0.15 0.39
88.3 71.0 75.2 90.5
0.64 0.38 0.35 0.68
0ve.ra.u Unweighted tcb
48.2
0.31
62.5
0.49
Weighted I&
77.9
0.44
85.6
0.61
a Kappa measures agreement beyond chance; a value of 1.0 indicates complete agreement among evaluators b Does not include level of disagreement. c Includes level of disagreement, does not apply for different grades. morulae produced by oviduct co-culture, as their embryo production system had used predominantly SOF-culture. The third evaluator was more involved with the selection of blastocysts than with evaluation of morulae. Omission of these evaluators led to a higher percentage of agreement and agreement beyond chance for embryos scored in both PBS and sucrose (Table 3). Again, the good and fair embryos were the most difficult to agree upon, and evaluation in sucrose did not improve evaluator agreement for those 2 grades of embryo quality, while agreement for excellent and poor embryos reached nearly excellent levels for the more experienced evaluators. To be able to control the accuracy of the morphological evaluation, the test embryos were divided into 4 groups, according to their number of nuclei determined at 7 dpi. According to this arbitrarily fmed “ideal score”, poor embryos should display a number of distinct nuclei ranging from none to no more than 25 nuclei; fair embryos from 26 to 50 nuclei; good embryos horn 51 to 75 nuclei and excellent embryos from 76 to 100 nuclei. The number of nuclei of the tested embryos ranged from no distinct nuclei (in 3 embryos) to 99 nuclei. Although an ideal score based solely upon cell number is a rigid way to correlate with embryo quality, it was the only measure we could use for statistical purposes. The “ideal score” of the test embryos was
1141
Theriogenology Table 3,
Agreement among the 4 most experienced evaluators on the quality grade of in vitro produced bovine embryos scored either in PBS or in sucrose at Day 6 pi.
Agreement (%) and kappa coefficient
Grade
Evaluation in PBS % K
(K)a
Evaluation in sucrose %
K
87.1 0.66 90.4 0.71 GOOd 76.7 0.42 74.6 0.49 Fair 82.9 0.54 86.7 0.54 Poor 80.8 0.46 90.8 0.72 overallb 63.8 0.52 71.3 0.60 a Kappa measures agreement beyond chance; a value of 1.0 indicates complete agreement among the 4 evaluators. b Unweighted K.
Excellent
added to Table 4, which also shows the relationship between mean total cell number of the embryos and embryo quality score given by the 7 evaluators. After linear regression of the scores versus total mean cell number, the regression coefficients of the ideal slope and of the scored slope by each evaluator were compared with a two sample t-test. Each evaluator differed significantly from the ideal slope, and the use of sucrose did not affect the accuracy of grading the embryo according to cell number. Evaluator 5 approached the ideal slope most, as well in PBS as in sucrose. However, neither method was useful for eliminating embryos with low cell numbers from Grade 1 and 2 categories. In fact, there was 1 embryo with only 10 cells which was graded as excellent by all 7 evaluators in PBS, and by 6 out of 7 evaluators in sucrose. The overall number of excellent graded embryos was lower in sucrose than in PBS. On the other hand, to avoid the assignment of embryos with high cell numbers to Grade 4, sucrose could be used since 3 embryos of 45, 74 and 96 cells respectively, which were scored by at least 1 evaluator to be of poor quality in PBS,were scored 1 or 2 categories higher in sucrose. DISCUSSION Under normal conditions, mammalian cells are bathed in physiological fluids, which are isosmotic to a sodium chloride concentration of 0.9%. In spite of this, cells am able to survive a&osmotic conditions to a certain extent. A cell with intact cell membranes which is exposed to a nonpermeable solute like sucrose will respond by losing water. The water leaves the cell in order to maintain equality between the chemical potentials of the intracellular and the extracellular water (14). The water loss inevitably evokes shrinkage of the cell mass. According to this principle sucrose can act as an osmotic counterforce to minimize swelling damage when glycerol-frozen embryos are thawed (9). Sucrose has also been used as a tool in the splitting of bovine morulae (1). And hypertonic sucrose can aid sperm injection or blastomere insertion by enlarging the perivitelliie space of oocytes (12,13,31-35). In the present study, sucroseinduced enlargement of the perivitelline space was used to facilitate morphological evaluation of in vitro produced bovine embryos at the morula stage.
1142
Table 4.
Theriogenology Total mean cell number of embryos graded into 4 categories by 7 different evaluators compared with the ideal score
Score Ideal score
Mean Cell Number (k SD) of embryos graded as Excellent Fair GOOd Poor 84.7rt9.3
69.7f 7.5
38.2k8.5
10.6rt9.3
Evaluator 1
PBS Sucrose
50.3f29.3 51.Lz+25.1
41.w32.0 52.6k31.3
42.4k34.1 39.1k29.4
13.0 f 10.9 8.9f10.3
Evaluator 2
PBS Sucrose
54.1f26.0 56.2f26.0
52.6226.9 41.2k32.7
41.2k33.9 43.2+28.2
16.4-2.0 10.7*11.1
Evaluator 3
PBS Sucrose
5 1.6f26.0 6O.ti28.3
46.1k26.0 51.1f29.1
44.4k33.9 38.5f30.0
13.3fll.5 9.5k11.7
Evaluator 4
PBS Sucrose
54.3k22.8 59.ti24.4
39.1f33.4 41.5f30.8
36.6f29.9 37.3k28.1
26.5k35.9 12.6f12.6
Evaluator 5
PBS Sucrose
58.1f24.5 55.lk24.2
44.2f29.9 57.9f28.6
37.3f30.9’ 32.1f27.3
26.3k31.6 15.5f22.2
Evaluator 6
PBS Sucrose
39.2f31.9 55.OZk30.5
46.5rt26.9 55.1f24.7
44.OLh37.3 30.0f29.1
14.3f17.1 3.25f3.4
Evaluator 7
PBS Sucrose
36.3f33.2 47.9f29.0
43.6f32.0 49.9f33.3
43.9k30.4 36.3f32.5
27.8ti9.3 15.7f14.4
In Experiment 1, it was demonstrated that in vitro produced bovine morulae are very responsive to a hypertonic treatment of 0.3 M sucrose in PBS. Within 1 min they had reached the bottom of a 20-pl drop, and within 40 set they had shrunken to about 73% of their original volume. This fast shrinkage rendered them morphologically almost identical to flushed cattle embryos, which occupy between 60 and 70% of the perivitelline space at the compact morula stage (10). The shrinkage is particularly noticeable in cells which are connected to one another, leaving extruded cells and cellular fragments clearly visible in the perivitelline space, thus facilitating the evaluation of morphology. If our technique is to be used for routine embryo evaluation, possible toxic effects of sucrose have to be excluded. It has been reported that very hypertonic sucrose solutions can cause membrane leakage in 8-cell mouse embryos (14). The same was observed in rabbit embryos where incubation of the embryo in 1.5 M sucrose in PBS for 30 min led to shrinkage followed by expansion in size. This size shift was not observed when the embryos were
Theriogenology
1143
incubated for 30 mitt in 0.5 M sucrose in PBS (34). Therefore, in our study, we chose to limit the sucrose concentration to 0.3 M and to limit the exposure time to 5 min, which is more than sufficient for thorough embryo evaluation. At this concentration and duration, no effect whatsoever on further embryonic development is expected. This was confirmed in Experiment 2, where development to the blastocyst stage was not adversely affected by sucrose treatment for up to 10 min. Normal total cell numbers of about 72 cells were found in Day 6 blastocysts, which is comparable with cell numbers found in flushed blastocysts (30). At Day 7 of development, both sucrose-treated and control blastocysts had a higher average cell number (=lOO) than reported for flushed embryos (=75; 30). Cell division, which is expressed by the presence of mitotic cells, continued for at least 2 d after sucrose treatment, and was not significantly different from that of the controls (Figure 3). We can conclude that sucrose does not slow down embryo development. Moreover, it does not increase cell death, as a lower pycnotic index was observed after a 10 min treatment with sucrose 0.3 M. On the other hand, it can be expected that sucrose has a negative influence on the cellular membrarte by the osmotic changes that are induced. Prolonged treatment with sucrose did not influence the percentage of lysed cells, although some embryos displayed degenerative nuclei at 120 and 180 min. In Experiment 3, embryos were graded into 4 morphological groups. These groups are basically the same as in the classification system of Lindner and Wright (lo), but only a degree of fragmentation is taken into account. Fragmentation is a common feature in in vitro culture of mammalian embryos : in culture systems without somatic cells it seems to occur at earlier stages of development than in somatic cell co-culture systems. Human embryos, for instance, are scored for fragmentation at the 2- to 6-tell stage (17), while sheep embryos which have been cultured in SOF-medium can already display substantial fragmentation at the 4-cell stage (28). It has been shown that cytoplasmic fragmentation can be reduced by adding somatic cells to the culture system (28,29). With the co-culture methods used in the present study, fragmentation of the cytoplasm in viable bovine embryos was not obvious until the morula stage, but then it became very difficult to grade the amount of fragmentation due to the lack of compaction. It is generally recognized that bovine embryos produced in co-culture look less compact than, for instance, embryos produced in mSOF (23,24). After artificially shrinking the embryo in sucrose, the aspect of the in vitro produced morula very much resembles the flushed compact morula. Morula evaluation has become much more easy to perform and can be done in a more objective way and by someone who is not acquainted with the less compact morphology of the co-cultured embryos, such as those used in our study. The same should be applicable to in vitro produced blastocysts, in which perivitelline space is totally absent. By using sucrose, the amount of fragmentation in the perivitelline space can be visualized and quantified. Evaluators agreed more often when they evaluated the quality of in vitro produced embryos in sucrose than in PBS. It has been shown that agreement among evaluators is higher for in vivo produced embryos than for in vitro produced embryos, where the stage of development is concerned (3). In our study, no evaluation of development was performed, although some embryos already showed the slightest sign of blastocoele formation. Only quality grade as assessed by amount of fragmentation was scored. By using sucrose to improve
Theriogenology
1144
the visualization of cellular fragments in the perivitelline space, we were able to show that agreement for quality grade, which is known to be moderate for in vitro produced embryos from this study (48.21, K= 0.31) and also from Farins study (57.7 8, K= 0.42, 3 ), was increased by evaluating the embryo in sucrose (62.5%
K=
0.49). Agieement beyond chance
was good for excellent (K= 0.64) and for poor (K = 0.68) embryos and was inclined to be excellent for the mom experienced evaluators. However, in our study, poor embryos or Grade 4 embryos did not resemble completely degenerated embryos but rather embryos with a small viable cell mass and severe fragmentation. Such embryos are not completely void of developmental capacity (26). Farin et al. (3) regarded Grade 4 embryos as degenerated and Grade 2 and 3 as morphologically defected. They found a significant improvement (approximately 20%) in evaluator agreement (78.7%, K= 0.60) when they combined Grade 2 and 3 embryos into a single grade (=embryos with morphological defects). Moreover, they suggest that a less complicated 3-point grading scheme can increase the level of agreement among graders when maximum agreement between evaluators is important. In our study, we were able to increase the level of agreement for Grade 4 embryos by using sucrose. The Grade 4 embryos in our study may be comparable to Grade 3 embryos in the Fat-in study (3) since this grade also represents embryos with many extruded cells and with debris. Subjectivity in morphological evaluation of in viva-produced cattle embryos was also found by van Wagtendonk-De Leeuw (27). She tested 29 persons from 4 ET-teams for their ability to score 18 frozen-thawed, in vivo-produced bovine embryos according to Lindner and Wright (10). In the study, one team graded embryos, in average, half an embryo grade higher than the other teams, and 5 individuals deviated substantially from the others in embryo grading. Uniformity in embryo grading is important when embryo grading results are correlated with survival of freezing and thawing, pregnancy rates and other factors. In evaluating embryo grade, clear definitions of quality grade should be explained to all evaluators to avoid, for example, some evaluators focusing on inclusion of lipid droplets into the cytoplasm or the estimated cell number of the embryo, while others do not. The evaluators in our study were asked to exclusively take the amount of fragmentation into account when scoring the embryos. Different interpretations by each evaluator of specific defects (3) or lack of familiarity with the morphology of in vitro cocultured bovine embryos may be the cause of deviation in embryo grading. A more consistent embryo grading system may be provided by the 3-scale grading scheme, as proposed by Farin et al. (3), evaluating the embryos in sucrose (this study), or by a combination of both. The use of embryo morphology for assessment of embryo quality remains questionable. Although there was a tendency to allocate the embryos with higher cell numbers to the excellent and good categories, no evaluator was competent to grade the embryos according to an idealized cell number standard score, neither in PBS nor in sucrose. However, it was not the purpose of the morphological evaluation to estimate the number of cells of the embryo, rather the amount of fragmentation of each embryo. Cell number was used as an objective quality standard by lack of other criteria.
1145
Theriogenology
In conclusion, it can be stated that in vitro produced bovine morulae shrink very rapidly in 0.3 M sucrose, yielding improved visualization of the cell fragments. Limited exposure (up to 10 min) to 0.3 M sucrose does not impair embryo viability in terms of subsequent blastocyst formation and cell number. The use of sucrose in embryo evahration improves agreement among evaluators and can be of value under circumstances which require a standardized scoring system by different individuals, such as embryo transfer experiments under field conditions. REFERENCES 1. Breclbacka P. Factors affecting cell viability during bisection of bovine embryos. Theriogenology 1995;44: 159- 166. 2. Butler JE, Biggers ID. Assessing the viability of preimplantation embryos in vitro. Thetiogenology 1989;31: 115126. 3. Farin PW, Britt JI-I, Shaw DW, Slenning BD. Agreement among evaluators of bovine embryos produced in vivo or in vitro. Theriogenology 1995;44:339-350. 4. Farin PW, Farin CE. Bovine embryos produced in vivo or in vitro: embryo survival and fetal development. Biol Reprod 1995;52:676-682. 5. Fleiss JL. Statistical methods for rates and proportions. New York: John Wiley 8c Sons, Inc 1981;212-236. 6. Hardy K, Handyside AH, Winstons RML. The human blastocyst: cell number, death and allocation during late preimplantation development in vitro. Development 1989; 107:597604. 7. Hasler JF, Henderson WB, Hurtgen PJ, Jin ZQ, McCauley AD, Mower SA, Neely B, Shuey LS, Stokes JE, Trimmer SA. Production, freezing and transfer of bovine IVFembryos and subsequent calving results. Theriogenology 1995;43:141-152. 8. Hasler IF, McCauley AD, Lathrop WF, Foote RH. Effect of donor-embryo recipient interactions on pregnancy rate in a large scale bovine embryo transfer program. Theriogenology 1987;27: 139- 167. 9. Leibo SP. Cryobiology: preservation of mammalian embryos. In: JW Evans, Hollaender A (eds) Genetic Engineering of Animals. New York and London: Plenum Press, 1986; 251272. 10. Lindner GM, Wright RW Jr. Bovine embryo morphology and evaluation. Theriogenology 1983;20:407-416. 11. Mahmoudzadeh AR, Van Soom A, Bols P, Ysebaert MT, de Kruif A. Optimization of a simple vitrification procedure for bovine embryos produced in vitro : effect of developmental stage, twostep addition of cryoprotectant and sucrose dilution on embryonic survival. J Reprod Fertil 1995;103:33-30. 12. Malter HE, Cohen J. Blastocyst formation and hatching in vitro following zona drilling of mouse and human embryos. Gam Res 1989,24:67-80. 13. Malter HE, Cohen J. Partial ulna dissection of the human oocyte : a nontraumatic method using micromanipulation to assist zona pellucida penetration. Fertil Steril 1989,51:139148.
1146 14.
Theriogenology
Mazur P, Schneider U. Osmotic responses of preimplantation mouse and bovine embryos and their cryobiological implications. Cell Biophys 1986;8:259-285. 15. Nagakawa A, Suzuki T, Suga T, Inaue T, Takahashi Y, Kanagewa H. Morphology and size of embryo recovered from superovulated cows. J Vet Med Sci 1991;53:287-290. 16. Overstrom EW. In vitro assessment of embryo viability. Theriogenology 1996; 45:3-16. 17. Pellestor F, Girardet A, And& B, Amal F, Humeau C. Relationship between morphology and chromosomal constitution in human preimplantation embryo. Mol Reprod Dev 1994;39:141-146. 18. Pratt HPM, Ziomek CA, Reeve WJD, Johnson MH. Compaction of the mouse embryo: an analysis of its components. J Embryo1 Exp Morph01 1982;70: 113- 132. 19. Purse1 VG, Wall RJ, Rexroad CE, Hammer RE, Brinster RL. A rapid whole-mount staining procedure for nuclei of mammalian embryos. Theriogenology 1985;24:687-700. 20. Reichenbach HD, Liebrich J, Berg U, Brem G. Pregnancy rates and births after unilateral or bilateral transfer of bovine embryos produced in vitro. J Reprod Fertil 1992;95:363370. 21. Shamsuddin M, Larsson B, Gustaffson H, Gustari S, Bartolome J, Rodriguez-Martinez H. Comparative morphological evaluation of in vivo and in vitro produced bovine embryos. Proc 12th Int Congr Anim Reprod 1992;3:1333-1335. 22. Shea BF. Evaluating the bovine embryo. Theriogenology 1981;15:31-43. 23. Van Langendonckt A, Lebrun N, Grisart B, Massip A, Dessy F. Cinematographic analysis of early bovine embryo development in synthetic oviduct fluid medium. Pnx 4th Int Symp Reprod Dom Rumin, 1994, Townsville, Australia. 24. Van Soom A, de Kruif A. A comparative study of in vivo and in vitro derived bovine embryos. Proc 12th Int Congr Anim Reprod 1992; 3:1365-1365. 25. Van Soom A, Mijten P, Van Vlaenderen I, Van den Branden J, Mahmoudzadeh AR, de Kruif A. Birth of double muscled Belgian blue calves after transfer of in vitro produced embryos in to dairy cattle. Theriogenology 1994;41:855- 867. 26. Van Soom A, Vanroose G, Bols PET, de Kruif A. Evaluation of morphology of in vitro produced bovine embryos after sucrose treatment : preliminary results. 1le Reunion A.E.T.E.,1995:252 abstr. 27. van Wagtendonk-de Leeuw AM. Evaluation of uniformity among persons in embryo grading from video recordings. Theriogenology 1996;45:230 abstr. 28. Walker SK, Heard TM, Seamark RF. In vitro culture of sheep embryos without co-culture: successes and perspectives. Theriogenology 1992;37: 11 l- 126. 29. Wiemer KE, Hoffman DI, Maxson WS, Eager S, Muhlberger B, Fiore I, Cuetvo M. Embryonic morphology and rate of implantation of human embryos following co-culture on bovine oviductal epithelial cells. Hum Reprod 1993;8:97-101. 30. Wthth YA, van der Zee-Kotting WI, Dieleman SJ, Bevers MM, Kruip TAM. Presence of mitotic cells : a parameter of embryo quality . Anim Repmd Sci 1994;35: 173-182. 31. Yang X, Chen J, Chen Y, Foote RH. Survival of rabbit eggs shrunken to aid in sperm microinjection. Theriogenology 1988;29:336 abstr. 32. Yang X, Chen Y, Chen J, Foote RH. Blastocyst development from rabbit ova fertilized by injected sperm. J Reprod Fertil 1988; 1: 13 abstr.
Theriogenology 33.
34.
35.
1147
Yang X, Chen Y, Chen J, Foote RI-I. Improved developmental potential of rabbit ova fertilized by sperm injection into the perivitelline space enlarged by hypertonic media. J Exp Zoo1 1990;255:114-119. Yang X, Chen Y, Chen J, Foote RI-I. Potential of hypertonic medium treatment for embryo micromanipulation : I. Survival of rabbit embryos in vitro and in vivo following sucrose treatment. Mol Reprod Dev 1990,27:110-l 17. Yang X, Zhang L, Fovacs A, Tobback C, Foote RH. Potential of hypertonic medium treatment for embryo micromanipulation : II. Assessment of nuclear transplantation methodology, isolation, subzona insertion and electrofusion of blastomeres to intact or functionally enucleated oocytes in rabbits. Mol Reprod Dev 1990;27: 118- 129.