THE JOURNAL OF EXPERIMENTAL ZOOLOGY 255114-119 (1990)
Improved Developmental Potential of Rabbit Oocytes Fertilized by Sperm Microinjection Into the Perivitelline Space Enlarged by Hypertonic Media XIANGZHONG YANG, JINGBO CHEN, YU &I CHEN, AND ROBERT H. FOOTE Department of Animal Science, Cornell University, Ithaca, N e w York 14853-4801 ABSTRACT
The objectives of the present study were: 1)to develop a simple and more efficient technique for sperm microinjection than is currently available, using the rabbit as a model, and 2) to evaluate the development of rabbit oocytes fertilized by single or multiple sperm microinjection. Hyperosmotic sucrose in phosphate-buffered saline (SPBS) was employed to dehydrate oocytes to increase the perivitelline space for sperm microinjection and prevent possible injury to the vitellus. In the first experiment, 58% (n = 29) oocytes treated with 0.5 M SPBS developed to morulae following multiple sperm microinjection compared, respectively, to 47% (n = 34) and 60% (n = 15) for control IVF with or without sucrose exposure (P > 0.05). Blastocyst development from microinjected oocytes, however, was much lower (P < 0.05) than that of controls (14% vs. 42% and 40%, respectively). Sham operation by puncturing the zona pellucida of the sucrose-treated oocytes with the microinjection pipette did not increase parthenogenesis (P > 0.05). In Experiment 2 a smallersize injection pipette and shorter sucrose exposure time after sperm microinjection resulted in 41% (n = 42) of the oocytes developing into blastocysts for the microinjection group, whereas only 21% (n = 24) developed to blastocysts in the control IVF group (P < 0.05). When relatively older oocytes (17 h r post ovulation injection) were used to test if microinjection could reduce the time to fertilization and cleavage (Expt. 3), a n average of 27% (n = 63) blastocysts resulted from microinjection vs. 0% In = 28) for the control IVF group.
Microinjection of sperm into oocytes has been conducted in a number of species, including mice (Mann, '88), hamsters (Uehara and Y anagimachi, '76), rats (Thadani, '79), rabbits (Hosoi et al., '88; Keefer, '891, and humans (Laws-King et al., '87; Sathananthan et al., '89), but the fertilization rate was low (Gordon and Laufer, '88). One reason for failure may be the mechanical damage during micromanipulation, particularly sperm microinjection into the ooplasm (Sathananthan et al., '89). Experiments reported here were designed to minimize this damage and to test fertility of oocytes of different ages following sperm microinjection or microinjection of single or multiple sperm with or without preincubation. A hyperosmotic solution of sucrose in phosphate-buffered saline (SPBS) was employed t o dehydrate oocytes so as to increase the perivitelline space for sperm microinjection. When 0.5 M or 1.0 M SPBS was used in preliminary studies the oocytes shrank t o about 40% or 37%, respectively, of their initial 0 1990 WILEY-LISS, INC.
volume (Yang et al., '88a). The enlarged perivitelline space reported in preliminary studies greatly facilitated sperm microinjection.
MATERIALS AND METHODS Animals and oocyte collection Mature Dutch-Belted females in our colony were superovulated (Kennelly and Foote, '65). Oocytes were collected from the oviducts 13 to 14 hr (Experiments 1 and 2) or 17 hr (Experiment 3) after giving luteinizing hormone. Sperm preparation Semen was collected from mature male rabbits and centrifuged twice at 500g for 5 min in Brackett's (Brackett and Oliphant, '75) defined medium (DM) before incubation in 5% COz and 95% air for 0 or 4.5 hr or overnight (20 hr) in different experiments. Before sperm microinjecReceived October 3, 1989; revision accepted January 12, 1990.
MICROINJECTED FRESH AND CAPACITATED SPERM
tion, semen was diluted with 10% polyvinyl pyrolidone (PVP) in PBS t o about 2 x 104/ml.Sperm were the only vector used for activating the oocytes compared with sham-injected controls. Also, microinjected sperm were placed in the expanded perivitelline space in results reported here because of the damage we encountered and reported by others when the vitelline membrane was pierced to place sperm within the oocyte.
Experiment 1 The viability of oocytes following hyperosmotic treatment (sucrose) and microfertilization by microinjected sperm was evaluated. Multiple sperm microinjection with a pipette 25 pm in outside diameter was performed, as more than one sperm in the perivitelline space is the normal pattern in rabbits. Cumulus cells were partially removed with 0.2% hyaluronidase followed by pipetting. Sperm were incubated in 5% C 0 2 and air at 39°C. Oocytes were then assigned randomly to several treatments. I n (A) several (two to five) spermatozoa were microinjected into the perivitelline space of oocytes pretreated with 0.5 M SPBS (1020 min) by using sperm capacitated in defined medium (DM) for 4.5 hr, or (B) in DM for 4.0 h r plus 0.5 h r with 60 pg/ml of added lysophosphatidylcholine. In rabbits several sperm frequently penetrate naturally into the perivitelline space. Oocytes pretreated with sucrose (C) or without sucrose (D) and mixed with sperm capacitated as in (A) served as IVF controls. Treatment (E) was sham-operated oocytes (punctured zona and injected with PBS only), and (F) oocytes without sperm co-incubation or other manipulation served as parthenogenetic controls. For multiple sperm microinjection, diluted semen in 10% polyvinyl pyrolidone (PVP) in PBS was pre-filled into the injection pipette. Two to five sperm were injected into the enlarged perivitelline space of the oocyte. Eggs (treatments A through F) were cultured at 39°C in 5% C 0 2 and 95% humidified air for 4 days. The medium used was BSM I1 (Kane and Foote, '701, which is a modified Ham's F10 medium.
115
and the poor oocyte development to blastocysts following sperm microinjection in Experiment 1, Experiment 2 was carried out with these modifications: 1)A smaller-size injection pipette, 8 pm in diameter, was used to pick up individual sperm from a 10%PVP sperm suspension; 2) after exposure of oocytes to 0.5 M SPBS for 10-20 min, individual oocytes were transferred to manipulation medium (PBS) and sperm were microinjected within 30 sec after oocyte transfer to PBS because sperm cannot survive well in sucrose PBS; 3) sperm were capacitated i n DM overnight at 39°C in a COZ incubator. In this experiment, only microinjection and control IVF were compared.
Experiment 3 Experiment 2 demonstrated that oocytes fertilized by microinjection of sperm may result in earlier cleavage and more oocytes developed to blastocysts than oocytes fertilized by the routine IVF procedure. To examine these events further, relatively older oocytes (17 h r after LH injection) were used. The microinjection procedure in this experiment was the same as outlined in Experiment 2, except that sperm were capacitated in DM for 4.5 h r because of the relatively poor results previously with control IVF for sperm incubated overnight, and based on results presented here on sperm survival and acrosome reaction after 0, 4.5, and 20 hr. Multiple sperm microinjection was again used for this experiment, considering the ease of microinjection and the natural pattern of fertilization in rabbits. St a t istica 1 ana l p s is Embryo development t o morula or blastocyst stages following culture for 4 days in vitro was used as criteria for all three experiments. Embryo data were analyzed by chi-square, and sperm motility and acrosome reaction were analyzed by analysis of variance (Snedecor and Cochran, '80).
RESULTS Experiment 1 Results are in Table 1. There were no differ-
Experiment 2 ences among treatments in the fertilization rate Previous observations suggested that rabbit (cleavage t o 2-4 cells) or development to morulae sperm may not tolerate hyperosmotic SPBS expo- (P> 0.05). Development beyond the morula stage sure although zygotes and embryos can (Yang et to blastocysts, however, was lower (P< 0.05) for al., '88a,b). A preliminary trial on the tolerance of the group of sperm microinjected (14%,n = 29, sperm to hyperosmotic SPBS was conducted by compared to those of control IVF with sucrose exposing sperm to 0.25 M and 0.5 M SPBS for 42%,n = 12 or without sucrose treatment 40%,n various times followed by recovery in isotonic = 15). There was no spontaneous cleavage (parPBS (290 mOsm). Based on the results of this trial thenogenesis), as indicated by the two controls
X. YANG ET AL.
116
T A B L E 1 . Experiment 1: development of oocytes fertilized by sperm microinjection and in vitro fertilization Oocyte incubation
Treatment combinations Sperm incub. Fertilization
A. 0.5MSPBS B. 0.5M SPBS C. 0.5M SPBS D. Normal PBS E. 0.5 M SPBS F. Normal PBS
DM DM + LPC DM DM No sperm No sperm
Micro-inj. Micro-inj. IVF IVF Puncture Control
%1
No. of oocvtes
Cleaved
Morulae
Blastocvsts
29 12 34 15 10 11
81a.b 67b 74b 100" 0 0
58" 50" 47" 60" 0 0
14" NA 42 b2 40b 0 0
'Values with different superscript letters (a,b) within the same column differ (P > 0.05). NA = not available due to contamination at the morula stage. Only one replicate of 12 oocytes was included to calculate blastocyst development. Other eggs during culture were contaminated a t the morula stage.
TABLE 2. Experiment 2: comparison of fertilization and development of oocytes collected 13 hr after LH injection fertilized by microinjected sperm and by control IVF
'
%
Treatments Microinjection Control IVF
No. of oocytes
Cleaved
Morulae
Blastocysts
42 24
50" 46"
45" 33b
41" 21b
stage (P < 0.05). One point noted in this experiment was that cleavage was first observed 12 h r after sperm microinjection, whereas cleavage in the IVF controls did not appear until 18 h r after treatment. This probably signifies more rapid fertilization following microinjection of sperm and may be the reason for higher development.
Experiment 3 To examine further the fact that microinjected sperm resulted in earlier cleavage and more development of eggs into blastocysts than for IVF without sperm added. Sperm microinjected into controls, older oocytes (17 hr) which would reoocytes without SPBS exposure caused lysis alquire rapid fertilization to be viable were used in most every time; development was thus not rethis experiment. Although cleavage rate was relcorded, nor was the procedure continued. atively high for the control IVF, no embryos developed into blastocysts during 4 days of culture. Experiment 2 This contrasts with 27% blastocyst development Rabbit sperm did not tolerate the hyperosmotic for the average of the three microinjection groups SPBS well. Fresh rabbit sperm had a n average (P < 0.05). Fresh (not incubated at 39"C, but held motility of 83%. Once the semen was exposed to at room temperature for 4.5 hr) single or multiple 0.5 M SPBS for 1,5, 10, and 25 min, sperm motil- microinjected control sperm served as one set of ity decreased to less than 5%, 0%, O%, and 0% controls. Surprisingly, the oocytes fertilized with respectively. When the sperm were returned to these sperm developed into blastocysts as well as approximately isotonic medium by dilution 10 with incubated sperm (P > 0.05, Table 3 ) . Sperm referred to in Table 3 as capacitated times with isotonic prewarmed PBS, the corresponding sperm motilities were 50%, 2076, lo%, were sperm incubated at 39°C in a C 0 2 incubator and 21%, respectively. When the sperm were ex- for 4.5 hr. A comparison of the percentage of posed to 0.25 M SPBS for 5 min and then returned motile and of acrosome-reacted sperm is shown in to isotonic PBS, sperm motilities were, respec- Figure 1. As the incubation time increased the tively, 20% and 75%. With 1min exposure to 1.0 M percentage of motile sperm decreased signifiSPBS progressive motility of sperm was largely cantly (P < 0.05), being 77.9 & 3.3%, 49.8 -+ regained in PBS. These data provided informa- 11.676, and 17.5 -+ 8.7%, respectively. Corresponding values for the rates of acrosome reaction tion for the sperm microinjection experiment. Results on development of oocytes following were 3.3 -+ 2.1%, 28.7 2 14.9%, and 42.8 +sperm microinjection and control IVF are in Table 25.5%, respectively (P < 0.05). A substantial percentage of possible parthe2. No difference was observed in cleaved embryos (P > 0.05), but with microinjected sperm a higher nogenetic cleavage or fragmentation was obpercentage of embryos developed t o the blastocyst served for the no sperm control groups, compared
'Values with different superscript letters (a,b) within the same column differ ( P < 0.05).
117
MICROINJECTED FRESH AND CAPACITATED SPERM
TABLE 3. Experiment 3: fertilization a n d development of "aging" oocytes (collected 17 h r after LH injection) resulting from sperm microinjection or IVF No. of oocytes
Treatments Microinjection Fresh sperm, > 1 One fresh sperm Capacitated sperm IVF control No sperm + SPBS Nosperm + PBS
20 19 24 28 36 31
Embryo development (%) Cleavage Morulae Blastocysts 8 (40Ib 8 (42Ib
12 (50)"~~ 20 (71)" 7 (19) 5 (16)
8 (40)" 7 (37)" 9 (38)" 9 (32)" 0 0
7 (35)" 5 (26)" 5 (21)" 0 (OIb 0 0
'Values with different superscript letters (a,b) within the same column differ, P < 0.05.
f
E- ' 0 ° 1 & 80
-3
Motility
Acrosome Reaction
'0
m
60
W
5 e 9 L
0
-W .-
40 20
L
I
0 0
4.5
20
Time (hours) of Sperm Incubation
Fig. 1. The percentage of motile and the percentage of acrosome-reacted sperm following incubation in defined medium (DM) in a n atmosphere of 5% C02:95%air at 39°C for 0, 4.5, and 20 hr.
to Experiment 1. Possibly this was caused by using relatively old oocytes.
DISCUSSION Microinjection of sperm offers the possibility of investigating many aspects of sperm capacitation, fertilization, fertilizing potential of individual sperm with different morphology, use of sperm as a vector for introducing foreign genes (Lavitrano et al., '89), or treatment of oligospermia problems in human IVF (Sathananthan et al., '89). The results from microfertilization remain low (Hosoi et al., '88; Keffer, '89) and only a few live young have been reported in rabbits (Hosoi et al., '88) and other species (Laws-King et al., '87; Mann, '88). We reported in preliminary form (Yang et al., '88a) and in more detail here a novel approach for microfertilization which resulted in a higher rate of development than previously reported (see review by Perreault et al., '891, particularly in rab-
bits (Hosoi e t al., '88; Keefer, '89). This improvement is accomplished by a simple procedure of sucrose addition t o shrink the oocyte for easy and rapid sperm microinjection. It is likely that the improved fertility resulting from the "microfertilized" oocytes is due to minimizing mechanical damage. In recent reports, we demonstrated that zygotes (Yang et al., '88a) and embryos (Yang et al., '88b) could tolerate hypertonic sucrose (0.5 or 1.0 M) in PBS exposed for up to 1 h r without reducing their viability in vitro. In addition, when 192 rabbit embryos were exposed for 0 (control), 30, and 60 min t o 0.5 M sucrose in PBS and transferred to recipients, 39, 42, and 31% young were born, respectively (Yang et al., '90). These studies clearly indicated the value of using hypertonic sucrose t o aid sperm in microinjection of eggs. In Experiment 1, a high rate of development to morulae was obtained, but few embryos developed t o the blastocyst stage. This may have been due t o the use of a relatively large injection pipette (25 pm outside diameter) causing some mechanical damage or due to the 20-30 min of exposure of sperm to SPBS after microinjection. As demonstrated in this study the tolerance of sperm to hypertonic sucrose is much less than that of oocytes or embryos (Yang et al., '88a, '90). After changing both procedures, a significant improvement was demonstrated in successive experiments, with 41% of the microinjected oocytes developing into blastocysts (Table 2). Since these experiments were initially described (Yang et al., '88a,b) and reported here in detail, Maker and Cohen ('89) were able to obtain high rates of fertilization and development following zona-opening-assisted IVF in mice with a lower sucrose concentration. Our sperm SPBS tolerance trial demonstrated that 0.25 M SPBS might be a "safer" concentration than 0.5 M SPBS. However we did not use 0.25 M SPBS-
118
X. YANG ET AL.
treated oocytes for sperm microinjection in this paper. As shown in Tables 1and 2, both rates of cleavage (46%, n = 24 vs. 82%, pooled n = 49) and blastocyst development (21%, n = 24 vs. 41%, pooled n = 27) were lower (P < 0.05) for control IVF in Experiment 2 than in Experiment 1. Since oocytes in Experiments 1 and 2 were collected at the same age (13-14 hr), the only major difference was in sperm incubation time (4.5 hr vs. overnight, 20 hr), which might be responsible for the different results. Our data on sperm motility and rate of acrosome reaction at different incubation times demonstrated that overnight incubation (Fig. 1) decreased sperm motility and increased the percentage of acrosome-reacted sperm. Therefore in the next experiment, 4.5 hr incubation time was adopted. It is noteworthy that the first cleavage was observed as early as 12 hr after microinjection of sperm, whereas we did not see cleavage for the control IVF groups until 18 hr. Microinjection, by circumventing the first barrier for fertilization, the zona pellucida, made the vitelline membrane accessible t o sperm quickly and thus could have reduced time before sperm and oocyte membranes fused. Therefore, it is not surprising that a higher development and earlier cleavage occurred with microinjected sperm. To confirm this, Experiment 3 was conducted using relatively old or “aging” oocytes (17 hr after LH injection). It has been reported that these oocytes can be fertilized, but the development to blastocysts (Maurer et al., ’69) and t o term (Chen et al., ’89) is low. As expected, the fertilization rate and the rate of development t o morulae between the microinjection and the IVF groups were similar (P > 0.051, but no embryos developed into blastocysts in the IVF group during 4 days of incubation, whereas an average of 27% developed t o blastocysts in the microinjection groups. In Experiment 3, 35% and 26% of the oocytes developed to blastocysts for multiple sperm and single sperm injection, respectively, but the difference was not significant (P > 0.05). This result contrasts with the report by Lasalle et al. (’87), where lower fertilization resulted when fewer than five sperm were inserted in the perivitelline space of the hamster and human oocytes. In addition, our preliminary experiment indicated that microinjection of sperm not treated to induce the acrosome reaction resulted in similar fertilization and development rates (P > 0.05) compared to incubated sperm. This appears t o contrast with reports which demonstrated that sperm must be
acrosome-reacted t o fuse with the egg membrane in microinjected human oocytes (Laws-King et al., ’87; Yamada et al., ’88) or in zone-denuded hamster eggs (Yanagimachi, ’881, although more research is needed to confirm this. It is not clear if this difference is due t o a species difference or t o the fact that following microinjection of sperm and returning the oocyte t o normal PBS, the expanding oocyte rapidly filled the zona pellucida. This rapid contact with sperm may have facilitated sperm acrosome reaction or even sperm-egg fusion, as spermatozoa were observed to complete the acrosome reaction in the perivitelline space of human oocytes following sperm microinjection (Sathananthan et al., ’89). In conclusion, oocytes, but not sperm, can tolerate considerable exposure to hypertonic solutions containing sucrose. It is essential to minimize the exposure time of sperm in hypertonic sucrose solution. This approach led t o a high and more rapid development to the blastocyst stage than when sperm were allowed to penetrate the zona pellucida unassisted in vitro. These results indicate that sperm incubation by procedures that cause the acrosome reaction and presumably capacitation required for normal zona pellucida penetration are not required for rapid fertilization and cleavage of eggs following sperm microinjection. Further research is needed to investigate the acrosome reaction, sperm-egg fusion, time of pronuclear formation, and development of embryos following microinjection of sperm into oocytes.
LITERATURE CITED Brackett, G., and G. Oliphant (1975) Capacitation of rabbit spermatozoa in vitro. Biol. Reprod., 12:260-274. Chen, Y., X. Yang, and R.H. Foote (1989) Timed breeding of rabbits with fresh and frozen-thawed semen and evidence of acrosome alteration following freezing and thawing. Anim. Reprod. Sci., 18:35-41. Gordon, J.W., and N. Laufer (1988) Applications of micromanipulation to human in vitro fertilization. J. In Vitro Fert. Embryo Transfer, 5:57-60. Hosoi, Y., M. Miyake, K. Utsumi, and H. Iritani (1988) Development of rabbit oocytes after microinjection of spermatozoa. In: Proc. 11th Int. Congr. Anim. Reprod. & AI, Vol. 3, pp. 331a-331c. Dublin, Ireland, 1988. Kane, M.T., and R.H. Foote (1970) Fractionated serum dialysate and synthetic media for culturing 2- and 4-cell embryos. Biol. Reprod., 2:356-362. Keefer, C.L. (1989) Fertilization by sperm injection in the rabbit. Gamete Res., 22.59-69. Kennelly, J.J., and R.H. Foote (1965) Superovulatory response of pre- and post-pubertal rabbits to commercially available gonadotropins. J. Reprod. Fertil., 9:179-188. Lasalle, B., A.M. Courtot, and J. Testart (1987) In vitro fertilization of hamster and human oocytes by microinjection of human sperm. Gamete Res., 16:69-78.
MICROINJECTED FRESH AND CAPACITATED SPERM Lavitrano, M., A. Camaioni, V.M. Frazio, S. Dobei, M.G. Farace, and C. Spadafora (1989) Sperm cells as factors for introducing foreign DNA into eggs: Genetic transformation of mice. Cell, 57:717-723. Laws-King, A., A. Trounson, A.H. Sathananthan, and I. Kola (1987) Fertilization of human oocytes by microinjection of a single spermatozoon under the zona pellucida. Fertil. Steril., 48:637-642. Mann, J.R. (1988) Full term development of mouse eggs fertilized by a spermatozoon microinjected under the zona pellucida. Biol. Reprod., 38:1077-1083. Malter, H.E., and J. Cohen (1989) Blastocyst formation and hatching in vitro following zona drilling of mouse and human embryos. Gamete Res., 24:67-80. Maurer, R.R., R.H. Whitener, and R.H. Foote (1969) Relationship of in vivo gamete aging and exogenous hormones to early embryo development in rabbits. Proc. SOC. Exp. Biol. Med., 131:882-885. Perreault, S.D., S.J. Naish, and B.R. Zirkin (1989) Fertilization by sperm microinjection and zona drilling: Applications in the basic and clinical sciences. In: Hands-on IVF, Cryopreservation and Micromanipulation. N. First and A.H. DeCherney, eds. University of Wisconsin Press, Madison, WI, in press. Sathananthan, A.H., S.C. Ng, A. Trounson, A. Bangso, A. Laws-King, and S.S. Ratnam (1989) Human microinsemination by injection of single or multiple sperm: Ultrastructure. Hum. Reprod., 4,574-583.
119
Snedecor, G.W., and W.G. Cochran (1980) Statistical Methods. Ames, Iowa, Iowa State University Press. Thadani, V.M. (1979) Injection of sperm heads into immature rabbit oocytes. J. Exp. Zool., 210:161-168. Uehara, T., and R. Yanagimachi (1976) Microsurgical injection of spermatozoa into hamster eggs with subsequent transformation of sperm nuclei into male pronuclei. Biol. Reprod., 15:467-470. Yamada, K., A.F.G. Stevenson, and L. Mettler (1988) Fertilization through spermatozoa1 microinjection: Significance of acrosome reaction. Hum. Reprod., 3:657-661. Yanagimachi, R. (1988) Mammalian fertilization. In: The Physiology of Reproduction. E. Knobil, J. Neill, L. Ewing, G.S. Greenwald, C.L. Markert, and D.W. Pfaff, eds. Raven Press, New York, pp. 135-185. Yang, X., J. Chen, Y. Chen, and R.H. Foote (1988a) Survival of rabbit eggs shrunken to aid in sperm microinjection. Theriogenology, 29:336 (Abstract). Yang, X., Y. Chen, J. Chen, and R.H. Foote (1988b) Embryo survival following hyperosmotic treatment with special reference to sucrose. Proc. 11th Int. Congr. Anim. Reprod. & A.I., Vol. 4, pp. 485a-485c. Dublin, Ireland, 1988. Yang, X., Y. Chen, J. Chen, and R.H. Foote (1990) Potential of hypertonic medium treatment for embryo micromanipulation. I. Survival of rabbit embryos in vitro and in vivo following sucrose treatment. Molecular Reprod. Devel. (submitted).
Improved Developmental Potential of Rabbit Oocytes Fertilized by Sperm Microinjection Into the Perivitelline Space Enlarged by Hypertonic Media XIANGZHONG YANG, JINGBO CHEN, YU &I CHEN, AND ROBERT H. FOOTE Department of Animal Science, Cornell University, Ithaca, N e w York 14853-4801 ABSTRACT
The objectives of the present study were: 1)to develop a simple and more efficient technique for sperm microinjection than is currently available, using the rabbit as a model, and 2) to evaluate the development of rabbit oocytes fertilized by single or multiple sperm microinjection. Hyperosmotic sucrose in phosphate-buffered saline (SPBS) was employed to dehydrate oocytes to increase the perivitelline space for sperm microinjection and prevent possible injury to the vitellus. In the first experiment, 58% (n = 29) oocytes treated with 0.5 M SPBS developed to morulae following multiple sperm microinjection compared, respectively, to 47% (n = 34) and 60% (n = 15) for control IVF with or without sucrose exposure (P > 0.05). Blastocyst development from microinjected oocytes, however, was much lower (P < 0.05) than that of controls (14% vs. 42% and 40%, respectively). Sham operation by puncturing the zona pellucida of the sucrose-treated oocytes with the microinjection pipette did not increase parthenogenesis (P > 0.05). In Experiment 2 a smallersize injection pipette and shorter sucrose exposure time after sperm microinjection resulted in 41% (n = 42) of the oocytes developing into blastocysts for the microinjection group, whereas only 21% (n = 24) developed to blastocysts in the control IVF group (P < 0.05). When relatively older oocytes (17 h r post ovulation injection) were used to test if microinjection could reduce the time to fertilization and cleavage (Expt. 3), a n average of 27% (n = 63) blastocysts resulted from microinjection vs. 0% In = 28) for the control IVF group.
Microinjection of sperm into oocytes has been conducted in a number of species, including mice (Mann, '88), hamsters (Uehara and Y anagimachi, '76), rats (Thadani, '79), rabbits (Hosoi et al., '88; Keefer, '891, and humans (Laws-King et al., '87; Sathananthan et al., '89), but the fertilization rate was low (Gordon and Laufer, '88). One reason for failure may be the mechanical damage during micromanipulation, particularly sperm microinjection into the ooplasm (Sathananthan et al., '89). Experiments reported here were designed to minimize this damage and to test fertility of oocytes of different ages following sperm microinjection or microinjection of single or multiple sperm with or without preincubation. A hyperosmotic solution of sucrose in phosphate-buffered saline (SPBS) was employed t o dehydrate oocytes so as to increase the perivitelline space for sperm microinjection. When 0.5 M or 1.0 M SPBS was used in preliminary studies the oocytes shrank t o about 40% or 37%, respectively, of their initial 0 1990 WILEY-LISS, INC.
volume (Yang et al., '88a). The enlarged perivitelline space reported in preliminary studies greatly facilitated sperm microinjection.
MATERIALS AND METHODS Animals and oocyte collection Mature Dutch-Belted females in our colony were superovulated (Kennelly and Foote, '65). Oocytes were collected from the oviducts 13 to 14 hr (Experiments 1 and 2) or 17 hr (Experiment 3) after giving luteinizing hormone. Sperm preparation Semen was collected from mature male rabbits and centrifuged twice at 500g for 5 min in Brackett's (Brackett and Oliphant, '75) defined medium (DM) before incubation in 5% COz and 95% air for 0 or 4.5 hr or overnight (20 hr) in different experiments. Before sperm microinjecReceived October 3, 1989; revision accepted January 12, 1990.
MICROINJECTED FRESH AND CAPACITATED SPERM
tion, semen was diluted with 10% polyvinyl pyrolidone (PVP) in PBS t o about 2 x 104/ml.Sperm were the only vector used for activating the oocytes compared with sham-injected controls. Also, microinjected sperm were placed in the expanded perivitelline space in results reported here because of the damage we encountered and reported by others when the vitelline membrane was pierced to place sperm within the oocyte.
Experiment 1 The viability of oocytes following hyperosmotic treatment (sucrose) and microfertilization by microinjected sperm was evaluated. Multiple sperm microinjection with a pipette 25 pm in outside diameter was performed, as more than one sperm in the perivitelline space is the normal pattern in rabbits. Cumulus cells were partially removed with 0.2% hyaluronidase followed by pipetting. Sperm were incubated in 5% C 0 2 and air at 39°C. Oocytes were then assigned randomly to several treatments. I n (A) several (two to five) spermatozoa were microinjected into the perivitelline space of oocytes pretreated with 0.5 M SPBS (1020 min) by using sperm capacitated in defined medium (DM) for 4.5 hr, or (B) in DM for 4.0 h r plus 0.5 h r with 60 pg/ml of added lysophosphatidylcholine. In rabbits several sperm frequently penetrate naturally into the perivitelline space. Oocytes pretreated with sucrose (C) or without sucrose (D) and mixed with sperm capacitated as in (A) served as IVF controls. Treatment (E) was sham-operated oocytes (punctured zona and injected with PBS only), and (F) oocytes without sperm co-incubation or other manipulation served as parthenogenetic controls. For multiple sperm microinjection, diluted semen in 10% polyvinyl pyrolidone (PVP) in PBS was pre-filled into the injection pipette. Two to five sperm were injected into the enlarged perivitelline space of the oocyte. Eggs (treatments A through F) were cultured at 39°C in 5% C 0 2 and 95% humidified air for 4 days. The medium used was BSM I1 (Kane and Foote, '701, which is a modified Ham's F10 medium.
115
and the poor oocyte development to blastocysts following sperm microinjection in Experiment 1, Experiment 2 was carried out with these modifications: 1)A smaller-size injection pipette, 8 pm in diameter, was used to pick up individual sperm from a 10%PVP sperm suspension; 2) after exposure of oocytes to 0.5 M SPBS for 10-20 min, individual oocytes were transferred to manipulation medium (PBS) and sperm were microinjected within 30 sec after oocyte transfer to PBS because sperm cannot survive well in sucrose PBS; 3) sperm were capacitated i n DM overnight at 39°C in a COZ incubator. In this experiment, only microinjection and control IVF were compared.
Experiment 3 Experiment 2 demonstrated that oocytes fertilized by microinjection of sperm may result in earlier cleavage and more oocytes developed to blastocysts than oocytes fertilized by the routine IVF procedure. To examine these events further, relatively older oocytes (17 h r after LH injection) were used. The microinjection procedure in this experiment was the same as outlined in Experiment 2, except that sperm were capacitated in DM for 4.5 h r because of the relatively poor results previously with control IVF for sperm incubated overnight, and based on results presented here on sperm survival and acrosome reaction after 0, 4.5, and 20 hr. Multiple sperm microinjection was again used for this experiment, considering the ease of microinjection and the natural pattern of fertilization in rabbits. St a t istica 1 ana l p s is Embryo development t o morula or blastocyst stages following culture for 4 days in vitro was used as criteria for all three experiments. Embryo data were analyzed by chi-square, and sperm motility and acrosome reaction were analyzed by analysis of variance (Snedecor and Cochran, '80).
RESULTS Experiment 1 Results are in Table 1. There were no differ-
Experiment 2 ences among treatments in the fertilization rate Previous observations suggested that rabbit (cleavage t o 2-4 cells) or development to morulae sperm may not tolerate hyperosmotic SPBS expo- (P> 0.05). Development beyond the morula stage sure although zygotes and embryos can (Yang et to blastocysts, however, was lower (P< 0.05) for al., '88a,b). A preliminary trial on the tolerance of the group of sperm microinjected (14%,n = 29, sperm to hyperosmotic SPBS was conducted by compared to those of control IVF with sucrose exposing sperm to 0.25 M and 0.5 M SPBS for 42%,n = 12 or without sucrose treatment 40%,n various times followed by recovery in isotonic = 15). There was no spontaneous cleavage (parPBS (290 mOsm). Based on the results of this trial thenogenesis), as indicated by the two controls
X. YANG ET AL.
116
T A B L E 1 . Experiment 1: development of oocytes fertilized by sperm microinjection and in vitro fertilization Oocyte incubation
Treatment combinations Sperm incub. Fertilization
A. 0.5MSPBS B. 0.5M SPBS C. 0.5M SPBS D. Normal PBS E. 0.5 M SPBS F. Normal PBS
DM DM + LPC DM DM No sperm No sperm
Micro-inj. Micro-inj. IVF IVF Puncture Control
%1
No. of oocvtes
Cleaved
Morulae
Blastocvsts
29 12 34 15 10 11
81a.b 67b 74b 100" 0 0
58" 50" 47" 60" 0 0
14" NA 42 b2 40b 0 0
'Values with different superscript letters (a,b) within the same column differ (P > 0.05). NA = not available due to contamination at the morula stage. Only one replicate of 12 oocytes was included to calculate blastocyst development. Other eggs during culture were contaminated a t the morula stage.
TABLE 2. Experiment 2: comparison of fertilization and development of oocytes collected 13 hr after LH injection fertilized by microinjected sperm and by control IVF
'
%
Treatments Microinjection Control IVF
No. of oocytes
Cleaved
Morulae
Blastocysts
42 24
50" 46"
45" 33b
41" 21b
stage (P < 0.05). One point noted in this experiment was that cleavage was first observed 12 h r after sperm microinjection, whereas cleavage in the IVF controls did not appear until 18 h r after treatment. This probably signifies more rapid fertilization following microinjection of sperm and may be the reason for higher development.
Experiment 3 To examine further the fact that microinjected sperm resulted in earlier cleavage and more development of eggs into blastocysts than for IVF without sperm added. Sperm microinjected into controls, older oocytes (17 hr) which would reoocytes without SPBS exposure caused lysis alquire rapid fertilization to be viable were used in most every time; development was thus not rethis experiment. Although cleavage rate was relcorded, nor was the procedure continued. atively high for the control IVF, no embryos developed into blastocysts during 4 days of culture. Experiment 2 This contrasts with 27% blastocyst development Rabbit sperm did not tolerate the hyperosmotic for the average of the three microinjection groups SPBS well. Fresh rabbit sperm had a n average (P < 0.05). Fresh (not incubated at 39"C, but held motility of 83%. Once the semen was exposed to at room temperature for 4.5 hr) single or multiple 0.5 M SPBS for 1,5, 10, and 25 min, sperm motil- microinjected control sperm served as one set of ity decreased to less than 5%, 0%, O%, and 0% controls. Surprisingly, the oocytes fertilized with respectively. When the sperm were returned to these sperm developed into blastocysts as well as approximately isotonic medium by dilution 10 with incubated sperm (P > 0.05, Table 3 ) . Sperm referred to in Table 3 as capacitated times with isotonic prewarmed PBS, the corresponding sperm motilities were 50%, 2076, lo%, were sperm incubated at 39°C in a C 0 2 incubator and 21%, respectively. When the sperm were ex- for 4.5 hr. A comparison of the percentage of posed to 0.25 M SPBS for 5 min and then returned motile and of acrosome-reacted sperm is shown in to isotonic PBS, sperm motilities were, respec- Figure 1. As the incubation time increased the tively, 20% and 75%. With 1min exposure to 1.0 M percentage of motile sperm decreased signifiSPBS progressive motility of sperm was largely cantly (P < 0.05), being 77.9 & 3.3%, 49.8 -+ regained in PBS. These data provided informa- 11.676, and 17.5 -+ 8.7%, respectively. Corresponding values for the rates of acrosome reaction tion for the sperm microinjection experiment. Results on development of oocytes following were 3.3 -+ 2.1%, 28.7 2 14.9%, and 42.8 +sperm microinjection and control IVF are in Table 25.5%, respectively (P < 0.05). A substantial percentage of possible parthe2. No difference was observed in cleaved embryos (P > 0.05), but with microinjected sperm a higher nogenetic cleavage or fragmentation was obpercentage of embryos developed t o the blastocyst served for the no sperm control groups, compared
'Values with different superscript letters (a,b) within the same column differ ( P < 0.05).
117
MICROINJECTED FRESH AND CAPACITATED SPERM
TABLE 3. Experiment 3: fertilization a n d development of "aging" oocytes (collected 17 h r after LH injection) resulting from sperm microinjection or IVF No. of oocytes
Treatments Microinjection Fresh sperm, > 1 One fresh sperm Capacitated sperm IVF control No sperm + SPBS Nosperm + PBS
20 19 24 28 36 31
Embryo development (%) Cleavage Morulae Blastocysts 8 (40Ib 8 (42Ib
12 (50)"~~ 20 (71)" 7 (19) 5 (16)
8 (40)" 7 (37)" 9 (38)" 9 (32)" 0 0
7 (35)" 5 (26)" 5 (21)" 0 (OIb 0 0
'Values with different superscript letters (a,b) within the same column differ, P < 0.05.
f
E- ' 0 ° 1 & 80
-3
Motility
Acrosome Reaction
'0
m
60
W
5 e 9 L
0
-W .-
40 20
L
I
0 0
4.5
20
Time (hours) of Sperm Incubation
Fig. 1. The percentage of motile and the percentage of acrosome-reacted sperm following incubation in defined medium (DM) in a n atmosphere of 5% C02:95%air at 39°C for 0, 4.5, and 20 hr.
to Experiment 1. Possibly this was caused by using relatively old oocytes.
DISCUSSION Microinjection of sperm offers the possibility of investigating many aspects of sperm capacitation, fertilization, fertilizing potential of individual sperm with different morphology, use of sperm as a vector for introducing foreign genes (Lavitrano et al., '89), or treatment of oligospermia problems in human IVF (Sathananthan et al., '89). The results from microfertilization remain low (Hosoi et al., '88; Keffer, '89) and only a few live young have been reported in rabbits (Hosoi et al., '88) and other species (Laws-King et al., '87; Mann, '88). We reported in preliminary form (Yang et al., '88a) and in more detail here a novel approach for microfertilization which resulted in a higher rate of development than previously reported (see review by Perreault et al., '891, particularly in rab-
bits (Hosoi e t al., '88; Keefer, '89). This improvement is accomplished by a simple procedure of sucrose addition t o shrink the oocyte for easy and rapid sperm microinjection. It is likely that the improved fertility resulting from the "microfertilized" oocytes is due to minimizing mechanical damage. In recent reports, we demonstrated that zygotes (Yang et al., '88a) and embryos (Yang et al., '88b) could tolerate hypertonic sucrose (0.5 or 1.0 M) in PBS exposed for up to 1 h r without reducing their viability in vitro. In addition, when 192 rabbit embryos were exposed for 0 (control), 30, and 60 min t o 0.5 M sucrose in PBS and transferred to recipients, 39, 42, and 31% young were born, respectively (Yang et al., '90). These studies clearly indicated the value of using hypertonic sucrose t o aid sperm in microinjection of eggs. In Experiment 1, a high rate of development to morulae was obtained, but few embryos developed t o the blastocyst stage. This may have been due t o the use of a relatively large injection pipette (25 pm outside diameter) causing some mechanical damage or due to the 20-30 min of exposure of sperm to SPBS after microinjection. As demonstrated in this study the tolerance of sperm to hypertonic sucrose is much less than that of oocytes or embryos (Yang et al., '88a, '90). After changing both procedures, a significant improvement was demonstrated in successive experiments, with 41% of the microinjected oocytes developing into blastocysts (Table 2). Since these experiments were initially described (Yang et al., '88a,b) and reported here in detail, Maker and Cohen ('89) were able to obtain high rates of fertilization and development following zona-opening-assisted IVF in mice with a lower sucrose concentration. Our sperm SPBS tolerance trial demonstrated that 0.25 M SPBS might be a "safer" concentration than 0.5 M SPBS. However we did not use 0.25 M SPBS-
118
X. YANG ET AL.
treated oocytes for sperm microinjection in this paper. As shown in Tables 1and 2, both rates of cleavage (46%, n = 24 vs. 82%, pooled n = 49) and blastocyst development (21%, n = 24 vs. 41%, pooled n = 27) were lower (P < 0.05) for control IVF in Experiment 2 than in Experiment 1. Since oocytes in Experiments 1 and 2 were collected at the same age (13-14 hr), the only major difference was in sperm incubation time (4.5 hr vs. overnight, 20 hr), which might be responsible for the different results. Our data on sperm motility and rate of acrosome reaction at different incubation times demonstrated that overnight incubation (Fig. 1) decreased sperm motility and increased the percentage of acrosome-reacted sperm. Therefore in the next experiment, 4.5 hr incubation time was adopted. It is noteworthy that the first cleavage was observed as early as 12 hr after microinjection of sperm, whereas we did not see cleavage for the control IVF groups until 18 hr. Microinjection, by circumventing the first barrier for fertilization, the zona pellucida, made the vitelline membrane accessible t o sperm quickly and thus could have reduced time before sperm and oocyte membranes fused. Therefore, it is not surprising that a higher development and earlier cleavage occurred with microinjected sperm. To confirm this, Experiment 3 was conducted using relatively old or “aging” oocytes (17 hr after LH injection). It has been reported that these oocytes can be fertilized, but the development to blastocysts (Maurer et al., ’69) and t o term (Chen et al., ’89) is low. As expected, the fertilization rate and the rate of development t o morulae between the microinjection and the IVF groups were similar (P > 0.051, but no embryos developed into blastocysts in the IVF group during 4 days of incubation, whereas an average of 27% developed t o blastocysts in the microinjection groups. In Experiment 3, 35% and 26% of the oocytes developed to blastocysts for multiple sperm and single sperm injection, respectively, but the difference was not significant (P > 0.05). This result contrasts with the report by Lasalle et al. (’87), where lower fertilization resulted when fewer than five sperm were inserted in the perivitelline space of the hamster and human oocytes. In addition, our preliminary experiment indicated that microinjection of sperm not treated to induce the acrosome reaction resulted in similar fertilization and development rates (P > 0.05) compared to incubated sperm. This appears t o contrast with reports which demonstrated that sperm must be
acrosome-reacted t o fuse with the egg membrane in microinjected human oocytes (Laws-King et al., ’87; Yamada et al., ’88) or in zone-denuded hamster eggs (Yanagimachi, ’881, although more research is needed to confirm this. It is not clear if this difference is due t o a species difference or t o the fact that following microinjection of sperm and returning the oocyte t o normal PBS, the expanding oocyte rapidly filled the zona pellucida. This rapid contact with sperm may have facilitated sperm acrosome reaction or even sperm-egg fusion, as spermatozoa were observed to complete the acrosome reaction in the perivitelline space of human oocytes following sperm microinjection (Sathananthan et al., ’89). In conclusion, oocytes, but not sperm, can tolerate considerable exposure to hypertonic solutions containing sucrose. It is essential to minimize the exposure time of sperm in hypertonic sucrose solution. This approach led t o a high and more rapid development to the blastocyst stage than when sperm were allowed to penetrate the zona pellucida unassisted in vitro. These results indicate that sperm incubation by procedures that cause the acrosome reaction and presumably capacitation required for normal zona pellucida penetration are not required for rapid fertilization and cleavage of eggs following sperm microinjection. Further research is needed to investigate the acrosome reaction, sperm-egg fusion, time of pronuclear formation, and development of embryos following microinjection of sperm into oocytes.
LITERATURE CITED Brackett, G., and G. Oliphant (1975) Capacitation of rabbit spermatozoa in vitro. Biol. Reprod., 12:260-274. Chen, Y., X. Yang, and R.H. Foote (1989) Timed breeding of rabbits with fresh and frozen-thawed semen and evidence of acrosome alteration following freezing and thawing. Anim. Reprod. Sci., 18:35-41. Gordon, J.W., and N. Laufer (1988) Applications of micromanipulation to human in vitro fertilization. J. In Vitro Fert. Embryo Transfer, 5:57-60. Hosoi, Y., M. Miyake, K. Utsumi, and H. Iritani (1988) Development of rabbit oocytes after microinjection of spermatozoa. In: Proc. 11th Int. Congr. Anim. Reprod. & AI, Vol. 3, pp. 331a-331c. Dublin, Ireland, 1988. Kane, M.T., and R.H. Foote (1970) Fractionated serum dialysate and synthetic media for culturing 2- and 4-cell embryos. Biol. Reprod., 2:356-362. Keefer, C.L. (1989) Fertilization by sperm injection in the rabbit. Gamete Res., 22.59-69. Kennelly, J.J., and R.H. Foote (1965) Superovulatory response of pre- and post-pubertal rabbits to commercially available gonadotropins. J. Reprod. Fertil., 9:179-188. Lasalle, B., A.M. Courtot, and J. Testart (1987) In vitro fertilization of hamster and human oocytes by microinjection of human sperm. Gamete Res., 16:69-78.
MICROINJECTED FRESH AND CAPACITATED SPERM Lavitrano, M., A. Camaioni, V.M. Frazio, S. Dobei, M.G. Farace, and C. Spadafora (1989) Sperm cells as factors for introducing foreign DNA into eggs: Genetic transformation of mice. Cell, 57:717-723. Laws-King, A., A. Trounson, A.H. Sathananthan, and I. Kola (1987) Fertilization of human oocytes by microinjection of a single spermatozoon under the zona pellucida. Fertil. Steril., 48:637-642. Mann, J.R. (1988) Full term development of mouse eggs fertilized by a spermatozoon microinjected under the zona pellucida. Biol. Reprod., 38:1077-1083. Malter, H.E., and J. Cohen (1989) Blastocyst formation and hatching in vitro following zona drilling of mouse and human embryos. Gamete Res., 24:67-80. Maurer, R.R., R.H. Whitener, and R.H. Foote (1969) Relationship of in vivo gamete aging and exogenous hormones to early embryo development in rabbits. Proc. SOC. Exp. Biol. Med., 131:882-885. Perreault, S.D., S.J. Naish, and B.R. Zirkin (1989) Fertilization by sperm microinjection and zona drilling: Applications in the basic and clinical sciences. In: Hands-on IVF, Cryopreservation and Micromanipulation. N. First and A.H. DeCherney, eds. University of Wisconsin Press, Madison, WI, in press. Sathananthan, A.H., S.C. Ng, A. Trounson, A. Bangso, A. Laws-King, and S.S. Ratnam (1989) Human microinsemination by injection of single or multiple sperm: Ultrastructure. Hum. Reprod., 4,574-583.
119
Snedecor, G.W., and W.G. Cochran (1980) Statistical Methods. Ames, Iowa, Iowa State University Press. Thadani, V.M. (1979) Injection of sperm heads into immature rabbit oocytes. J. Exp. Zool., 210:161-168. Uehara, T., and R. Yanagimachi (1976) Microsurgical injection of spermatozoa into hamster eggs with subsequent transformation of sperm nuclei into male pronuclei. Biol. Reprod., 15:467-470. Yamada, K., A.F.G. Stevenson, and L. Mettler (1988) Fertilization through spermatozoa1 microinjection: Significance of acrosome reaction. Hum. Reprod., 3:657-661. Yanagimachi, R. (1988) Mammalian fertilization. In: The Physiology of Reproduction. E. Knobil, J. Neill, L. Ewing, G.S. Greenwald, C.L. Markert, and D.W. Pfaff, eds. Raven Press, New York, pp. 135-185. Yang, X., J. Chen, Y. Chen, and R.H. Foote (1988a) Survival of rabbit eggs shrunken to aid in sperm microinjection. Theriogenology, 29:336 (Abstract). Yang, X., Y. Chen, J. Chen, and R.H. Foote (1988b) Embryo survival following hyperosmotic treatment with special reference to sucrose. Proc. 11th Int. Congr. Anim. Reprod. & A.I., Vol. 4, pp. 485a-485c. Dublin, Ireland, 1988. Yang, X., Y. Chen, J. Chen, and R.H. Foote (1990) Potential of hypertonic medium treatment for embryo micromanipulation. I. Survival of rabbit embryos in vitro and in vivo following sucrose treatment. Molecular Reprod. Devel. (submitted).
