Human Reproduction vol.6 no.3 pp.423-431, 1991
Experience with zona drilling and zona cutting to improve fertilization rates of human oocytes in vitro
D.Payne1, K.J.McLaughlin2, H.T.Depypere3, C.A.Kirby, G.M.Warnes and C.D.Matthews Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, The Queen Elizabeth Hospital, Woodville, South Australia 5011, department of Obstetrics and Gynaecology, The University of Adelaide, GPO Box 498, Adelaide, South Australia 5001 and 3Academisch Ziekenhuis De Pintelann 185, Gent, Belgium 'To whom correspondence should be addressed
Zona drilling (ZD) and zona cutting (ZC) were used in an IVF programme to assist fertilization in semen defect patients. Twenty-seven patients consented to ZD where acidified Tyrode's was used to create a hole in the zona pellucida. In 19 patients, ZD increased the fertilization rate to 29% compared with 8% (P < 0.001) in their routine IVF cycles, and in eight patients precluded from routine IVF, a fertilization rate of 14% was achieved. Twenty-two patients consented to ZC where a slit in the zona is made mechanically. In 12 patients ZC increased the fertilization rate to 31% compared with 14% (P < 0.01) from previous routine IVF cycles, and in 10 patients precluded from routine IVF, a fertilization rate of 34% was achieved. In 13 cycles, 68 uncut control oocytes were inseminated. In five cycles both control and ZC oocytes were fertilized (n.s.d). In eight cycles no control oocytes were fertilized compared with 27% of ZC oocytes. The polyspermy rate was 4.6%. Twenty-four per cent of ZD and 12% of ZC (P < 0.01) oocytes and embryos were degenerate after 42 h. Both ZD and ZC can increase the fertilization rate of sub-optimal semen, however, in our hands neither technique produced a pregnancy. Key words: zona drilling/zona cutting/fertilization/semen factors Introduction Routine in-vitro fertilization (IVF) procedures have been used with variable success to treat male infertility (Yates and de Kretser, 1987; Gordon et al., 1988). However, some male factor patients consistently fail to fertilize all or most oocytes or have inadequate numbers of spermatozoa to be offered IVF. Typically, these males have severely compromised seminal values including oligozoospermia ( < 2 0 x 106 sperm/ml) and/or athenozoospermia (50). Usually inadequate numbers of morphologically normal, motile spermatozoa are available for insemination purposes. For fertilization to occur, spermatozoa must first penetrate the © Oxford University Press
cumulus and then the zona pellucida before reaching the oocyte. While the cumulus of the oocyte may be removed without adverse effect on fertilization or subsequent embryonic development (Mahadevan and Trounson, 1985), the zona pellucida still remains a formidable barrier to low numbers of weakly motile, abnormally shaped spermatozoa. As removal of the entire zona pellucida may impair the development and survival of the preimplantation embryo, micromanipulative techniques designed to allow spermatozoa access to the oocyte while minimizing damage to the zona may offer some advance. The injection of a single spermatozoon into die oocyte cytoplasm of hamster (Uehara and Yanagimachi, 1976) and human oocytes (Lanzendorf et al., 1988) has resulted in pronuclear formation, and the injection of spermatozoa under die zona pellucida of mouse (Mann, 1988) and human oocytes (Ng et al., 1988; Fishel et al., 1990) has resulted in pregnancies. However, these procedures involve the risk of damage to the oocyte and currently result in low rates of fertilization. Additionally, the artificial selection of a single or a few spermatozoa for injection is a major ethical concern, and is a particular problem in die treatment of teratozoospermia in the human. Localized acid digestion of die zona as in zona drilling (ZD) (Gordon and Talansky, 1986), or physical breaching of the zona as in zona cutting (ZC) (Tsunoda et al., 1986), allow spermatozoa freer access to the oocyte plasma membrane without compromising oocyte viability. Additionally, diese techniques permit a degree of natural sperm selection, as only those spermatozoa undergoing capacitation and a physiologically normal acrosome reaction will be able to fuse with die oolemma. Gordon and Talansky (1986) showed that significandy more ZD mouse oocytes were fertilized at low sperm concentrations when compared with intact oocytes, and proposed zona drilled mouse oocytes as a model for the management of the oligozoospermic human. Depypere et al. (1988) compared the effectiveness of both ZD and ZC for improving fertilization rates in mouse oocytes at low sperm concentrations and concluded diat more zona drilled oocytes were fertilized at equivalent suboptimal sperm concentrations. Both techniques were associated with low oocyte mortality and normal in vitro and in vivo development. The application of assisted fertilization techniques to couples who had failed or poor fertilization in routine IVF cycles might be expected to increase fertilization rates. Our initial clinical trial of the ZD technique was carried out between November 1987 and August 1988 in 27 patients. While fertilization rates were significantly improved following ZD, abnormal oocyte and embryo morphology was frequently observed and subsequently a trial of the ZC technique was performed from September 1988 to April 1989 in 22 patients. This paper reports the results of these two trials. 423
D.Payne et al.
Materials and methods Selection of patients Zona drilling Couples were admitted to the trial if they had previously had up to three cycles of routine IVF in which fertilization rates were c P < 0.05
8 (25.8)b
4 (44.4)
49 (98.0)
a > b P < 0.05
1.9 (33)*
1.3 (IO)b
2.3(11)
8 (25.8)*
20 (64.5)b
5 (55.6)
Number fertilized (%)
52 (29.3)*
Number of patients receiving a transfer (%)
17 (54.8)*
Average number of embryos transferred (number) No. cycles with no fertilization
16 (8.3)
3 4 (168)
Nil
a > b P < 0.001
a > b P < 0.01
T w o polyspermic embryos.
425
D.Payne et al.
Nineteen patients in the first group had 31 ZD cycles in which 177 oocytes were collected and 52 were fertilized. In this group, fertilization increased from 8.3% in pre-ZD cycles to 29.3% (P < 0.001) after ZD, and the number of cycles in which no fertilization occurred dropped from 64.5% to 25.8% (P < 0.01). The increased fertilization rate was lower than that achieved in our standard IVF comparison group (89.8%) (P < 0.001). Of the 62 fertilized oocytes, two (3.2%) were polyspermic. The percentage of patients receiving a transfer increased after ZD (25.8% versus 54.8%; P < 0.05) but again was less than that seen in standard IVF (98.0%; P < 0.001). The number of embryos available for transfer more than tripled (P < 0.001) and more embryos per patient were transferred (Table II). Eight patients in the second group had nine ZD cycles in which 72 oocytes were collected and 10 were fertilized. When semen parameters from these males were compared with patients who had previous routine IVF cycles, sperm counts (12 x 106 versus 29 x 106, P < 0.05) and quality of motion ( 2 - versus 2, P < 0.05) were significantly lower (Table I)- Corresponding fertilization rates were also significantly lower after ZD (13.9% versus 29.3%, P < 0.05), although the number of patients receiving a transfer and number of embryos available for transfer was not different (Table H). No pregnancy was achieved in either group after a total of 21 transfers. Zona cutting Twenty-two patients had a total of 23 ZC cycles in which 215 oocytes were collected, 47 out of 147 ZC oocytes fertilized and 33 ZC embryos transferred in 15 transfers. For the purpose of statistical comparison, patients were again divided into two groups on the basis of previous treatment regimes. The first group had previous routine IVF cycles (Table W) and a second group had previous ZD cycles (Table IV). Six patients are common to both groups. In the first group, 12 patients had 17 previous IVF cycles and 13 ZC cycles. Fertilization rates increased from 13.5% in pre-ZC IVF cycles to 30.7% after ZC (P < 0.01), and the number of patients receiving a transfer increased (35.2 versus 83.3%, P < 0.01) as did the number of embryos available for transfer (P < 0.01). The number of cycles in which fertilization did not occur was lower after ZC (P < 0.01). Ten patients who had no previous routine IVF cycles are included in the data of Table El. There was no difference between the two ZC groups and no difference in semen parameters was evident between the three groups (Table ITJ). In the second group, eight patients had 15 previous ZD cycles and eight ZC cycles. The percentage of ZC oocytes fertilized was less although it did not reach significance (32.6 versus 19.0%, P = 0.06). There was no significant difference in the number of patients receiving a transfer (86.7 versus 50%) or the number of cycles with no fertilization (13.3 versus 25%). The number of embryos transferred was lower after ZC (7 versus 25, P < 0.05). Fourteen patients who had no previous ZD cycles had 15 ZC cycles. They had a significantly higher fertilization rate (41.7 versus 19.0%, P < 0.01) and number of embryos transferred (2.4 versus 1.8%, P < 0.01) than ZC patients who had previous ZD. There was no difference in semen parameters between the three groups (Table IV). 426
Table i n . A comparison of ZC cycles and routine IVF cycles in 12 patients Routine IVF ZC cycles with routine IVF
ZC cycles Statistical with no significance routine IVF
Number of patients
12
12
10
Number of cycles
17
13
10
Number of oocytes
111
91
56
Number fertilized (%) Number of patients
15 (13.5)* 28 (3O.7)b*
19 (33.9^
a < b,c P < 0.01
6 (35 2) 1
10 (83.3)b*»
5 (50.0)
a < b P < 0 01
1.5 (9)*
2 0 (20)b
2.6 (13)c
a b P < 0.01
receiving
transfer (%) Average no of embryos transferred (number) Number of cycles with no fertilization (%)
0.01
Three fertilized ZC oocytes were polyspermic. •*One patient had two frozen ZC embryos
Table IV. A comparison of ZC cycles and ZD cycles in eight patients ZD cycles
Number of patients
ZC cycles ZC cycles Statistical with previous with no significance ZD previous ZD
8
8
14
Number of 15 cycles
8
15
Number of 95 oocytes
63
84
Number 31 (32 6)' fertilized
12 (19.0)b
35 (41.7)c
a > b P = 0.06 b < c P < 0.01
11(73.3)
NS
Number of 13(86 7) patients receiving transfer
4(50)
Average 1.9 (25)" no. of embryos transferred (number)
1.8 (7)b
2.4 (26)c
a > b P < 0.05 b < c P < 0.01
Number of cycles with no fertilization (%)
2(25)
3(20)
NS
2(13 3)
Zona manipulation in human IVF
Table V. A comparison of ZC cycles with fertilizaUon and with no fertilization of control oocytes Patients with fertilization in control oocytes Number of cycles Control oocytes: Number Fertilized (%)
Patients with no fertilization in control oocytes
Patients with no control oocytes
8
10
26 19 (73.1)
42 0
Nil Nil
25 13 (52)»
45 12 (26.7)b
77 22 (28.6)*
5
Statistical significance
ZC oocytes: Number Fertilized (%)
a > b P < 0.05
Three polyspermic embryos.
Additionally, in 13 of the 23 ZC cycles, approximately half the oocytes were retained as control oocytes and not ZC (Table V). Of these, five cycles showed no difference in the unexpectedly high fertilization rate of ZC and control oocytes (73.1 versus 52.0%). In eight cycles there was no fertilization in control oocytes, while 26.7% of ZC oocytes were fertilized. This was less than the fertilization rate in ZC oocytes where there was also fertilization in control oocytes (26.7 versus 52.0%, P < 0.05), but the same as the fertilization rate in ZC oocytes where all oocytes collected were ZC (26.7 versus 28.6%). There was no difference in semen parameters between the three groups (Table V). Overall, of the 47 fertilized ZC oocytes, three (6.4%) were polyspermic. No pregnancies were achieved after a total of 15 transfers. Survival of ZD and ZC oocytes and embryos in culture Oocytes or embryos were observed at 0, 17 and 42 h after insemination. Immediately after ZD and ZC all oocytes appeared to have normal morphology, but at 17 and 24 h increasing numbers of both unfertilized oocytes and embryos appeared degenerate with a dark shrunken cytoplasm, a phenomenon rarely encountered in standard IVF. The results of these observations are summarized and compared with observations from non-semen factor patients undergoing standard IVF in Table VI. The number of degenerate unfertilized oocytes/embryos was increased after 17 and 42 h in culture (7.9 and 23.8% respectively after ZD compared with 0.8 and 1.0% in undrilled oocytes; P < 0.001). Similarly, the number of degenerate ZC oocytes and embryos was increased after 17 and 42 h in culture (8.2 and 12.2% compared with unmanipulated oocytes; P < 0.001); however, after 42 h in culture, degeneration of ZC oocytes was less than that seen in ZD oocytes and embryos (12.2 versus 23.8%, P < 0.01). At transfer, all embryos were graded according to quality on a scale of 1 to 3 (Table VII) and the grades of ZD and ZC embryos at transfer were compared with embryo quality in standard IVF. Significantly fewer (P < 0.001) grade one embryos were transferred in ZD and ZC cycles (12.1% ZD, 6.1% ZC) compared with 45.7% standard IVF. To assess the ultrastructural effects of the ZD and ZC procedure, morphologically normal unfertilized and some fertilized but uncleaved ZD, ZC and unmanipulated oocytes were
Table VI. Numbers (%) of degenerate oocytes or embryos from ZD and ZC cycles at 0, 17 and 42 h after insemination compared with embryos and oocytes from routine IVF cycles. Eleven ZD oocytes which were transferred as pronuclear embryos are not included
Number of oocytes
Zona drilled
Zona cut
Routine IVF
240
147
383
0
0
0
Oh
Statistical significance
17 h (%)
19 (7 9)'
12 (8.2)b
3 (0.8)c
a,b > c P < 0.01
42 h (%)
57 (23.8)1
18 (12.2)b
4(1.0) c
a,b > c P < 0.001 a > b P < 0.01
Table VD. The quality of ZD and ZC' embryos at transfer compared with embryo quality from routine IVF cycles (11 ZD and 28 routine IVF PN embryos are not included) Zona drilled
Zona cut
Routine IVF
Statistical significance
Embryos transferred (%) Grade 1 Grade 2 Grade 3
4 (12.If 26 (78.8)1 3(9.1)'
2 (6.1)b 30 (90.9)b 1 (3.0)b
Total
33
33
64 (45.7)c 74 (52.8)c 2 (1.4)c
a,b < c P < 0.001 a,b > c P < 0.01 a > c P < 0.05
140
Grade 1: all blastomeres intact, no fragmentation. Grade 2. slight fragmentation, unequal sized blastomeres. Grade 3: gross fragmentation.
fixed and processed for transmission electron microscopy. At fixation, many spermatozoa were observed bound to the zona pellucida and often a large mass of spermatozoa was seen in the vicinity of the hole or the cut. These spermatozoa were removed during subsequent steps of EM preparation. While most ZD and ZC oocytes had a similar ultrastructural appearance to intact oocytes in that cortical granules were numerous and closely applied to the plasma membrane and cell organelles were evenly distributed throughout the cytoplasm (Figure 1), other ZD and ZC oocytes demonstrated a cortical zone largely devoid of cell oganelles and cortical granules (Figure 2). This phenomenon was not observed in unfertilized control oocytes. One fertilized but 427
D.Payne el al.
1
zp,*-
•;
*>:-•• ?
.
Fig. 1. Electron micrograph of a normal 48-h-old unfertilized ZD human oocyte. The hole in the zona pellucida (ZP) is indicated by the asterisk. Note abundant cortical granules (CG). Magnification X5200.
Fig. 2. Electron micrograph of an abnormal 48-h-old unfertilized ZD human oocyte. There are very few cortical granules (CG) and the cortex of the oocyte is largely and atypically free of cell organelles (arrows). Magnification X520O.
uncleaved ZC oocyte showed atypical retention of cortical granules (Figure 3).
cycles. Additionally, we achieved a 13.9% fertilization rate in the second group of ZD patients whose poor semen parameters precluded them from routine IVF. Similarly, in 12 of the ZC patients, who had a total of 17 previous routine IVF cycles, ZC improved the number of oocytes fertilized (30.7 versus 13.5%). WHO (1987) recommends that semen with >50% abnormal forms be classed as teratozoospermic. Both groups of ZD patients had median normal forms well below that of WHO guidelines (21 and 14%), and significantly different fertilization rates after ZD (29.3 versus 13.9%). The difference in fertilization rates, and also the inability of ZD to increase fertilization rates to that of routine IVF in non-semen factor patients (89.8%) may be related to sperm morphology, since per cent normal morphology correlates strongly with fertilizing ability (Shalgi et al., 1985: Marsh et al., 1987). ZC like ZD was able to increase the fertilization rate compared with that seen in previous routine IVF cycles for the same patients. However, in patients who had both ZD and ZC cycles. ZC oocytes appeared to be less likely to be fertilized compared with ZD oocytes (19.0 versus 32.6%). ZD produces a hole of - 12 fim in diameter, compared with ZC which produces a cut — 40 /tm long but only — 2 /*m wide. The oocyte area available to spermatozoa after ZD is — 230 unr compared with an estimated 80 /tm: after ZC. This may account for the difference in fertilization rates between the two techniques. In this study, a total of 22 patients had 23 cycles of ZC. There were 13 cycles in which approximately half the oocytes collected were ZC and half were retained intact as a control group. Mahadevan and Trounson (1985) found that there was no difference in fertilization rates between intact and cumulus free oocytes. so control oocytes were not denuded. In five cycles there
Discussion The definitive test of sperm function is its ability to fertilize homologous oocytes. In our initial ZD trial, we had a population of infertile patients comprised largely of oligozoospermic and/or asthenozoospermic men, in whom IVF was precluded or the frequency of fertilization was low (8.3%). Approximately 30% of these patients had previously undergone up to three routine IVF cycles in which no fertilization was achieved. Of the 22 couples participating in the ZC trial, 12 had previously had up to three cycles of routine IVF with a fertilization rate of 15.3% and 10 patients had semen parameters which precluded routine IVF. Previous experience had suggested that sperm function in these couples was abnormal, and they were unlikely to achieve a pregnancy using standard IVF procedures. Hence, the development and use of micromanipulative techniques to increase fertilization rates in vitro was warranted. Although we achieved a significant improvement in fertilization rates after ZD. compared with routine IVF (29.3 versus 8.3%). it was less than that reported by Cohen etal. (1988) (30-50%). However, their group of male factor patients had fertilization rates of 24% in control (non-drilled) oocytes, substantially higher than that seen in our group of patients in their previous routine IVF cycles. It is possible therefore that the group of male factor patients admitted to our ZD trial had a more severe form of sperm dysfunction. Our results are similar to the results of Gordon et al. (1988). who achieved a fertilization rate of 32% with ZD oocytes from 10 patients who had no fertilization in previous 428
Zona manipulation in human IVF
ZP CG
Fig. 3. Electron micrograph of a fertilized uncleaved 48-h-old ZC human oocyte. There is atypical retention of cortical granules (CG). The cut in the zona pellucida (ZP) is indicated by arrows. Magnification X32OO.
was no difference in the fertilization rate of control oocytes and ZC oocytes (73.1 versus 52.0%). However, in eight cycles there was no fertilization in control oocytes, while a 26.7% fertilization rate was achieved in ZC oocytes. In a further 10 cycles, all oocytes were ZC; as based on semen parameters complete failure in fertilization of intact oocytes was anticipated. In these cycles 28.6% ZC oocytes were fertilized of which three were polyspermic. Thus ZC appears to have little effect on fertilization rates if fertilization was going to occur, yet enhanced fertilization rates where fertilization of intact oocytes did not occur, or was not anticipated. Of a total of 109 oocytes fertilized after either ZD or ZC, only five were polyspermic (4.6%). Gordon et al. (1988) reported a 50% incidence of polyspermy in ZD oocytes and proposed that the rapid block to polyspermy exists at the zona pellucida in the human. Polyspermy rates of 12 and 18% have been reported following partial zona dissection (PZD) compared with polyspermy rates in control oocytes of 20 and 29%, respectively (Maker and Cohen, 1989; Cohen et al., 1989). Sathananthan et al. (1989) and Fishel et al. (1990) obtained monospermic fertilization after multiple spermatozoa were transferred into the perivitelline space of human oocytes. We have observed motile spermatozoa swimming in the perivitelline space of normally fertilized ZC oocytes—a phenomenon similar to that seen in fertilized rabbit oocytes where there exists a very strong vitelline block to polyspermy (Yanagimachi, 1988) We have also observed normozoospermic fertilization in all six ZC and six intact oocytes from one patient, indicating that sperm function was normal and a vitelline block to polyspermy was operating. Malter et al. (1989) showed that the polyspermy rate in
normozoospermic patients was 29% for oocytes younger than 25 h post-oocyte recovery, but that it increased to 65% in aged oocytes. This suggests the existence of a vitelline block to polyspermy which diminishes as the oocyte ages. Increased rates of polyspermy are also observed in immature human oocytes (Trounson et al., 1982; Van der Ven et al., 1985), presumably due to the existence of a relatively 'soft' zona pellucida (Tesarik et al., 1988) and lack of an efficient vitelline block to polyspermy in immature oocytes. We postulate therefore that there is a transient rapid block to polyspermy at the level of the oocyte plasma membrane and the increased rates of polyspermy which have been observed in zona dissected oocytes is due to oocyte immaturity or post-maturity at the time of insemination, rather than zona manipulation bypassing a rapid block to polyspermy resident solely in the zona. Of the oocytes that were ZD, 24% exhibited morphological evidence of degeneration after 42 h in culture. Of the embryos that cleaved and were transferred, less were of grade one quality compared with those obtained from standard IVF cycles (12 versus 46%). Cohen et al. (1988) also found that the survival of embryos from ZD cycles was compromised, and suggested that this was a result of damage to the oocyte cytoplasm by the acidified Tyrodes used to dissolve the hole through the zona. In addition. Gordon et al. (1988) reported morphological abnormalities in some human ZD embryos after 40 h in culture. Whilst it is possible that fertilization by genotypically abnormal spermatozoa contributes to this phenomenon, early cleavage patterns at least to the two-cell stage appear to be determined by the (xxryte cytoplasm (Magnuson and Epstein, 1987; Pedersen. 1988). Further, decreased rates of implantation of ZD fertilized 429
D.Payne el al.
oocytes compared to control oocytes in the rat have been reported even though fertile males were used for insemination (Van der Hyden et al., 1989) and Ng et al. (1989) described a high proportion of human oocytes exhibiting meiotic arrest following ZD. This evidence, along with our own evidence of ultrastructural damage to some ZD human oocytes, and lack of a pregnancy after 21 transfers of ZD embryos strongly suggests that ZD may be detrimental to human oocyte viability. It was for this reason that we discontinued ZD in favour of ZC. Using ZC, we were able to decrease the number of degenerate oocytes and embryos to 12%, after 42 h in culture; however, there was no change in the low percentage of grade one embryos that were transferred. ZC is a more difficult technique, partly due to the small perivitelline space of human oocytes, and distortion of the oocyte during cutting may contribute to degeneration or poor cleavage patterns. Shrinking the oocyte in a hypertonic solution (Cohen et al., 1988) may reduce distortion during cutting. We have transferred 33 embryos which have resulted from insemination of ZC oocytes in 15 inter-uterine transfers and have not achieved a pregnancy. In our programme we would expect a pregnancy rate of 15—20% after the transfer of two embryos and hence at least three pregnancies should have been established from 15 transfers of ZC embryos. The lack of a pregnancy after ZC may be due to the high proportion of grade 2 embryos arising after ZC. However, in our present programme, grade 2 embryos regularly give rise to pregnancies. It has been suggested that the cells of zona-manipulated rabbit embryos which have been transferred to the oviduct are vulnerable to leukocytic attack (Moore et al., 1968). Alternatively, Nichols and Gardener (1989) have proposed that embryos with slit zonae may extrude their contents during contractions of the reproductive tract and hence are not able to produce pregnancies. Concurrent with transferring human PZD embryos to the uterus, Cohen et al. (1990) administered short-term immunosuppression with antibiotic cover and reported a pregnancy rate of 28%. Additionally, Cohen et al. (1990) have reported a clinical pregnancy rate of 44% after transferring embryos with slits made in the zonae subsequent to cleavage to assist in blastocyst hatching. This suggests that there may be some immunological involvement in the survival of zonamanipulated embryos in utero and may explain our inability to produce a pregnancy. We have now changed our protocol to include the regimen of immunosuppression used by Cohen et al. (1989). This clinical study of the zona drilling and zona cutting techniques has shown that it is possible to increase the rate at which spermatozoa from infertile males will fertilize human oocytes. There appears to be an inherent difficulty with the ZD technique which was not foreseen when used with mouse oocytes. The use of acid Tyrodes to dissolve a hole in the zona pellucida causes damage to at least one-quarter of the oocytes, and maybe more if abnormalities in cleavage are taken into account. Hence the clinical usefulness of ZD appears to be limited in the human. In contrast, zona cutting does not appear to be accompanied by so many morphologically abnormal oocytes and embryos, yet leads to a similar overall fertilization rate. Reduction in distortion to the oocyte during the cutting procedure may lead to improved embryo quality. 430
Acknowledgements The authors wish to thank Mr M.Makinson for photographic assistance, and Helen Holmes and Carol Burford for typing the manuscript. Dr S.P.Flaherty is thanked for a critical appraisal of the manuscript.
References CohenJ., Malter.H., FehUly.C, Wright.G., Elsner.C, Kort,H. and MasseyJ. (1988) Implantation of embryos after partial opening of oocyte zona pellucida to facilitate sperm penetration. Lancet, ii, 162. Cohen^I., Malter.H., Wright.G., Kort.H., Massey.J. and Mitchell,D. (1989) Partial zona dissection of human oocytes when failure of zona pellucida penetration is anticipated. Hum. Reprod., 4, 435—442. Cohen^I., Elsner.C, Kort,H., Malter,H., Massey.J., Mayer,M.P. and Weimer.K. (1990) Impairment of the hatching process following IVF in the human and improvement of implantation by assisting hatching using micromanipulation. Hum. Reprod., 5, 7—13. Depypere.H.T., McLaughlin.K.J., Seamark,R.F., Warnes.G.M. and Matthews,CD. (1988) Comparison of zona cutting and zona drilling as techniques for assisted fertilization in the mouse. J. Reprod. Fertil., 84, 205-211. Fishel,S., Antinori,S., Jackson,P., Johnson,!., Lisi,F., Chiariello,F. and Versaci,C. (1990) Twin birth after subzonal insemination. Lancet, i, 722-723. Forster.M.S., Smith.W.D., Lee,W.I., Berger,R.E., Karp,L.E. and Stenchever.M.A. (1983) Selection of human spermatozoa according to their relative motility and their interaction with zona-free hamster eggs. Fertil. Steril., 40, 655-660. Gordon,J.W. and Talansky.B.E. (1986) Assisted fertilization by zona drilling: a mouse model for correction of oligospermia. J. Exp. Tool., 239, 347-354. Gordon.J.W., Grunfeld,L., Garrisi,G.L., Talanski,B.E., Richards.C. and Laufer.N. (1988) Fertilization of human oocytes by sperm from infertile males after zona pellucida drilling. Fertil. Steril., 50, 68—73. Kerin^.F.P., Peek,J., Warnes.G.M., Kirby,C, Jeffrey.R., Matthews, C D . and Cox,L.W. (1984) Improved conception rate after intrauterine insemination of washed spermatozoa from men with poor quality semen. Lancet, i, 533-534. Lanzendorf.S.E., Maloney,M.K., Veeck,L.L., SlusserJ., Hodgen.G.D. and Rosenwaks.Z. (1988) A preclinical evaluation of pronuclear formation by microinjection of human spermatozoa into human oocytes. Fertil. Steril., 49, 835-842. Magnuson,T. and Epstein,CJ. (1987) Genetic expression during early mouse development. In Bavister.B.D. (ed.), The Mammalian Preimplantation Embryo. Regulation of Growth and Differentiation in Vitro. Plenum Press, New York, pp. 133-150. Mahadevan.M.M. and Trounson,A.O. (1985) Removal of the cumulus oophorus from the human oocyte for in vitro fertilization. Fertil. Steril., 43, 263-267. Malter.H.E. and Cohen.J. (1989) Partial zona dissection of the human oocyte: a non traumatic method using micromanipulation to assist zona pellucida penetration. Fertil. Steril., 51, 139-148. Malter,H., Talansky,B., GordonJ. and Cohen,J. (1989) Monospermy and polyspermy after partial zona dissection of reinseminated human oocytes. Gamete Res., 23, 377-386. Mann,J.R. (1988) Full term development of mouse eggs fertilized by a spermatozoon mkroinjected under the zona pellucida. Biol. Reprod., 38, 1077-1083. Marsh,S.K., Bolton.V.N. and Braude.P.R. (1987) The effect of morphology on the ability of human spermatozoa to penetrate zona free hamster oocytes. Hum. Reprod., 2, 499-503. Moore.N.W., Adams.CE. and Rowson.L.E.A. (1968) Developmental potential of single blastomeres of the rabbit egg. J. Reprod. Fertil., 17, 527-534.
Zona manipulation in human IVF Ng,S.C, Bongso.A., Ratnam.S.S., Sathananthan.H., Chan.C.L.K., Wong,P.C, Hagglund,L., Anandakumar.C, Wong,Y.C. and Goh.V.H.H. (1988) Pregnancy after transfer of sperm under zona. Lancet, il, 790. Ng,S.C, Bongso.T.A., Chang,S.I., Sathananthan.A.H. and Ratnam.S.S. (1989) Transfer of human sperm into the perivitelline space of human oocytes after zona drilling or zona puncture. Fertil. Steril., 52, 73-78. Nichols,J. and Gardner.R.L. (1989) Effect of damage to the zona pellucida on development of preimplantation embryos in the mouse. Hum. Reprod., 4, 180-187. Pedersen,R.A. (1988) Early mammalian embryogenesis. In Knobil,E., Neill.J.D., Ewing.L.L., Greenwald,G.S., Markert.C.L. and Pfaff.D.W. (eds), The Physiology of Reproduction. Raven Press, New York, p. 187. Quinn,P., Kerin,J.K. and Warnes.G.M. (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil. Steril., 44, 493-498. Sathananthan,A.H., Ng.S.C, Trounson.A., Bongso,A., Laws-King,A. and Ratnam,S.S. (1989) Human micro-insemination by injection of single and multiple sperm: ultrastructure. Hum. Reprod., 4, 574-583. Shalgi,R., Dor,J., Rudak,E., Lusky.A., Goldman,B., Mashiach,S. and Nebel.L. (1985) Penetration of sperm from teratospermic men into zona-free hamster eggs. Int. J. Androi, 8, 285—294. Tesarik,J., Pilka,L. and Travnik.P. (1988) Zona pellucida resistance to sperm penetration before the completion of human oocyte maturation. J. Reprod. Fertil., 83, 487-495. Trounson,A.O., Mohr.L.R., Wood.C. and Leeton,J.F. (1982) Effect of delayed insemination on in-vitro fertilization, culture and transfer of human oocytes. J. Reprod. Fertil, 64, 285-294. Tsunoda.Y., Yasui,T., Nakamura.K., Uchida.T. and Sugie.T. (1986) Effect of cutting the zona pellucida in the pronuclear transplantation in the mouse. J. Exp. Tool, 240, 119-125. Uehara.T. and Yanagimachi.R. (1976) Microsurgical injection of spermatozoa into hamster eggs with subsequent transformation of sperm nuclei into male pronuclei. Biol. Reprod., 15, 467—470. Van der Hyden,B., McLaughlin.K.J., Rutledge.J.M. and Armstrong, D.T. (1989) Zona drilling increases the penetrability of rat oocytes matured in vivo. Biol. Reprod., 40, 953-960. Van der Ven.H.H., Al-Hasani.S., Diedrich.K., Hamerich,U., Lehmann,F. and Krebs.D. (1985) Polyspermy in in vitro fertilization of human oocytes: frequency and possible causes. Ann. N. Y. Acad. Sci., 442, 88-95. World Health Organization (1987) WHO Laboratory Manual for the Examination of Human Semen and Semen —Cervical Mucus Interaction. Cambridge University Press, Cambridge, p. 27. Yanagimachi.R. (1988) Mammalian fertilization. In Knobil.E., Neill.J.D., Ewing.L.L., Greenwald,G.S., Markert,C.L. and Pfaff,D.W. (eds), The Physiology of Reproduction. Raven Press, New York, pp. 135-185. Yates.C.A. and De Kretser,D.M. (1987) Male-factor infertility and in vitro fertilization. J. In Vitro Fertil. Embryo Transfer, 4, 141 — 147. Received on May 11, 1990; accepted on October 5, 1990
431
Experience with zona drilling and zona cutting to improve fertilization rates of human oocytes in vitro
D.Payne1, K.J.McLaughlin2, H.T.Depypere3, C.A.Kirby, G.M.Warnes and C.D.Matthews Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, The Queen Elizabeth Hospital, Woodville, South Australia 5011, department of Obstetrics and Gynaecology, The University of Adelaide, GPO Box 498, Adelaide, South Australia 5001 and 3Academisch Ziekenhuis De Pintelann 185, Gent, Belgium 'To whom correspondence should be addressed
Zona drilling (ZD) and zona cutting (ZC) were used in an IVF programme to assist fertilization in semen defect patients. Twenty-seven patients consented to ZD where acidified Tyrode's was used to create a hole in the zona pellucida. In 19 patients, ZD increased the fertilization rate to 29% compared with 8% (P < 0.001) in their routine IVF cycles, and in eight patients precluded from routine IVF, a fertilization rate of 14% was achieved. Twenty-two patients consented to ZC where a slit in the zona is made mechanically. In 12 patients ZC increased the fertilization rate to 31% compared with 14% (P < 0.01) from previous routine IVF cycles, and in 10 patients precluded from routine IVF, a fertilization rate of 34% was achieved. In 13 cycles, 68 uncut control oocytes were inseminated. In five cycles both control and ZC oocytes were fertilized (n.s.d). In eight cycles no control oocytes were fertilized compared with 27% of ZC oocytes. The polyspermy rate was 4.6%. Twenty-four per cent of ZD and 12% of ZC (P < 0.01) oocytes and embryos were degenerate after 42 h. Both ZD and ZC can increase the fertilization rate of sub-optimal semen, however, in our hands neither technique produced a pregnancy. Key words: zona drilling/zona cutting/fertilization/semen factors Introduction Routine in-vitro fertilization (IVF) procedures have been used with variable success to treat male infertility (Yates and de Kretser, 1987; Gordon et al., 1988). However, some male factor patients consistently fail to fertilize all or most oocytes or have inadequate numbers of spermatozoa to be offered IVF. Typically, these males have severely compromised seminal values including oligozoospermia ( < 2 0 x 106 sperm/ml) and/or athenozoospermia (50). Usually inadequate numbers of morphologically normal, motile spermatozoa are available for insemination purposes. For fertilization to occur, spermatozoa must first penetrate the © Oxford University Press
cumulus and then the zona pellucida before reaching the oocyte. While the cumulus of the oocyte may be removed without adverse effect on fertilization or subsequent embryonic development (Mahadevan and Trounson, 1985), the zona pellucida still remains a formidable barrier to low numbers of weakly motile, abnormally shaped spermatozoa. As removal of the entire zona pellucida may impair the development and survival of the preimplantation embryo, micromanipulative techniques designed to allow spermatozoa access to the oocyte while minimizing damage to the zona may offer some advance. The injection of a single spermatozoon into die oocyte cytoplasm of hamster (Uehara and Yanagimachi, 1976) and human oocytes (Lanzendorf et al., 1988) has resulted in pronuclear formation, and the injection of spermatozoa under die zona pellucida of mouse (Mann, 1988) and human oocytes (Ng et al., 1988; Fishel et al., 1990) has resulted in pregnancies. However, these procedures involve the risk of damage to the oocyte and currently result in low rates of fertilization. Additionally, the artificial selection of a single or a few spermatozoa for injection is a major ethical concern, and is a particular problem in die treatment of teratozoospermia in the human. Localized acid digestion of die zona as in zona drilling (ZD) (Gordon and Talansky, 1986), or physical breaching of the zona as in zona cutting (ZC) (Tsunoda et al., 1986), allow spermatozoa freer access to the oocyte plasma membrane without compromising oocyte viability. Additionally, diese techniques permit a degree of natural sperm selection, as only those spermatozoa undergoing capacitation and a physiologically normal acrosome reaction will be able to fuse with die oolemma. Gordon and Talansky (1986) showed that significandy more ZD mouse oocytes were fertilized at low sperm concentrations when compared with intact oocytes, and proposed zona drilled mouse oocytes as a model for the management of the oligozoospermic human. Depypere et al. (1988) compared the effectiveness of both ZD and ZC for improving fertilization rates in mouse oocytes at low sperm concentrations and concluded diat more zona drilled oocytes were fertilized at equivalent suboptimal sperm concentrations. Both techniques were associated with low oocyte mortality and normal in vitro and in vivo development. The application of assisted fertilization techniques to couples who had failed or poor fertilization in routine IVF cycles might be expected to increase fertilization rates. Our initial clinical trial of the ZD technique was carried out between November 1987 and August 1988 in 27 patients. While fertilization rates were significantly improved following ZD, abnormal oocyte and embryo morphology was frequently observed and subsequently a trial of the ZC technique was performed from September 1988 to April 1989 in 22 patients. This paper reports the results of these two trials. 423
D.Payne et al.
Materials and methods Selection of patients Zona drilling Couples were admitted to the trial if they had previously had up to three cycles of routine IVF in which fertilization rates were c P < 0.05
8 (25.8)b
4 (44.4)
49 (98.0)
a > b P < 0.05
1.9 (33)*
1.3 (IO)b
2.3(11)
8 (25.8)*
20 (64.5)b
5 (55.6)
Number fertilized (%)
52 (29.3)*
Number of patients receiving a transfer (%)
17 (54.8)*
Average number of embryos transferred (number) No. cycles with no fertilization
16 (8.3)
3 4 (168)
Nil
a > b P < 0.001
a > b P < 0.01
T w o polyspermic embryos.
425
D.Payne et al.
Nineteen patients in the first group had 31 ZD cycles in which 177 oocytes were collected and 52 were fertilized. In this group, fertilization increased from 8.3% in pre-ZD cycles to 29.3% (P < 0.001) after ZD, and the number of cycles in which no fertilization occurred dropped from 64.5% to 25.8% (P < 0.01). The increased fertilization rate was lower than that achieved in our standard IVF comparison group (89.8%) (P < 0.001). Of the 62 fertilized oocytes, two (3.2%) were polyspermic. The percentage of patients receiving a transfer increased after ZD (25.8% versus 54.8%; P < 0.05) but again was less than that seen in standard IVF (98.0%; P < 0.001). The number of embryos available for transfer more than tripled (P < 0.001) and more embryos per patient were transferred (Table II). Eight patients in the second group had nine ZD cycles in which 72 oocytes were collected and 10 were fertilized. When semen parameters from these males were compared with patients who had previous routine IVF cycles, sperm counts (12 x 106 versus 29 x 106, P < 0.05) and quality of motion ( 2 - versus 2, P < 0.05) were significantly lower (Table I)- Corresponding fertilization rates were also significantly lower after ZD (13.9% versus 29.3%, P < 0.05), although the number of patients receiving a transfer and number of embryos available for transfer was not different (Table H). No pregnancy was achieved in either group after a total of 21 transfers. Zona cutting Twenty-two patients had a total of 23 ZC cycles in which 215 oocytes were collected, 47 out of 147 ZC oocytes fertilized and 33 ZC embryos transferred in 15 transfers. For the purpose of statistical comparison, patients were again divided into two groups on the basis of previous treatment regimes. The first group had previous routine IVF cycles (Table W) and a second group had previous ZD cycles (Table IV). Six patients are common to both groups. In the first group, 12 patients had 17 previous IVF cycles and 13 ZC cycles. Fertilization rates increased from 13.5% in pre-ZC IVF cycles to 30.7% after ZC (P < 0.01), and the number of patients receiving a transfer increased (35.2 versus 83.3%, P < 0.01) as did the number of embryos available for transfer (P < 0.01). The number of cycles in which fertilization did not occur was lower after ZC (P < 0.01). Ten patients who had no previous routine IVF cycles are included in the data of Table El. There was no difference between the two ZC groups and no difference in semen parameters was evident between the three groups (Table ITJ). In the second group, eight patients had 15 previous ZD cycles and eight ZC cycles. The percentage of ZC oocytes fertilized was less although it did not reach significance (32.6 versus 19.0%, P = 0.06). There was no significant difference in the number of patients receiving a transfer (86.7 versus 50%) or the number of cycles with no fertilization (13.3 versus 25%). The number of embryos transferred was lower after ZC (7 versus 25, P < 0.05). Fourteen patients who had no previous ZD cycles had 15 ZC cycles. They had a significantly higher fertilization rate (41.7 versus 19.0%, P < 0.01) and number of embryos transferred (2.4 versus 1.8%, P < 0.01) than ZC patients who had previous ZD. There was no difference in semen parameters between the three groups (Table IV). 426
Table i n . A comparison of ZC cycles and routine IVF cycles in 12 patients Routine IVF ZC cycles with routine IVF
ZC cycles Statistical with no significance routine IVF
Number of patients
12
12
10
Number of cycles
17
13
10
Number of oocytes
111
91
56
Number fertilized (%) Number of patients
15 (13.5)* 28 (3O.7)b*
19 (33.9^
a < b,c P < 0.01
6 (35 2) 1
10 (83.3)b*»
5 (50.0)
a < b P < 0 01
1.5 (9)*
2 0 (20)b
2.6 (13)c
a b P < 0.01
receiving
transfer (%) Average no of embryos transferred (number) Number of cycles with no fertilization (%)
0.01
Three fertilized ZC oocytes were polyspermic. •*One patient had two frozen ZC embryos
Table IV. A comparison of ZC cycles and ZD cycles in eight patients ZD cycles
Number of patients
ZC cycles ZC cycles Statistical with previous with no significance ZD previous ZD
8
8
14
Number of 15 cycles
8
15
Number of 95 oocytes
63
84
Number 31 (32 6)' fertilized
12 (19.0)b
35 (41.7)c
a > b P = 0.06 b < c P < 0.01
11(73.3)
NS
Number of 13(86 7) patients receiving transfer
4(50)
Average 1.9 (25)" no. of embryos transferred (number)
1.8 (7)b
2.4 (26)c
a > b P < 0.05 b < c P < 0.01
Number of cycles with no fertilization (%)
2(25)
3(20)
NS
2(13 3)
Zona manipulation in human IVF
Table V. A comparison of ZC cycles with fertilizaUon and with no fertilization of control oocytes Patients with fertilization in control oocytes Number of cycles Control oocytes: Number Fertilized (%)
Patients with no fertilization in control oocytes
Patients with no control oocytes
8
10
26 19 (73.1)
42 0
Nil Nil
25 13 (52)»
45 12 (26.7)b
77 22 (28.6)*
5
Statistical significance
ZC oocytes: Number Fertilized (%)
a > b P < 0.05
Three polyspermic embryos.
Additionally, in 13 of the 23 ZC cycles, approximately half the oocytes were retained as control oocytes and not ZC (Table V). Of these, five cycles showed no difference in the unexpectedly high fertilization rate of ZC and control oocytes (73.1 versus 52.0%). In eight cycles there was no fertilization in control oocytes, while 26.7% of ZC oocytes were fertilized. This was less than the fertilization rate in ZC oocytes where there was also fertilization in control oocytes (26.7 versus 52.0%, P < 0.05), but the same as the fertilization rate in ZC oocytes where all oocytes collected were ZC (26.7 versus 28.6%). There was no difference in semen parameters between the three groups (Table V). Overall, of the 47 fertilized ZC oocytes, three (6.4%) were polyspermic. No pregnancies were achieved after a total of 15 transfers. Survival of ZD and ZC oocytes and embryos in culture Oocytes or embryos were observed at 0, 17 and 42 h after insemination. Immediately after ZD and ZC all oocytes appeared to have normal morphology, but at 17 and 24 h increasing numbers of both unfertilized oocytes and embryos appeared degenerate with a dark shrunken cytoplasm, a phenomenon rarely encountered in standard IVF. The results of these observations are summarized and compared with observations from non-semen factor patients undergoing standard IVF in Table VI. The number of degenerate unfertilized oocytes/embryos was increased after 17 and 42 h in culture (7.9 and 23.8% respectively after ZD compared with 0.8 and 1.0% in undrilled oocytes; P < 0.001). Similarly, the number of degenerate ZC oocytes and embryos was increased after 17 and 42 h in culture (8.2 and 12.2% compared with unmanipulated oocytes; P < 0.001); however, after 42 h in culture, degeneration of ZC oocytes was less than that seen in ZD oocytes and embryos (12.2 versus 23.8%, P < 0.01). At transfer, all embryos were graded according to quality on a scale of 1 to 3 (Table VII) and the grades of ZD and ZC embryos at transfer were compared with embryo quality in standard IVF. Significantly fewer (P < 0.001) grade one embryos were transferred in ZD and ZC cycles (12.1% ZD, 6.1% ZC) compared with 45.7% standard IVF. To assess the ultrastructural effects of the ZD and ZC procedure, morphologically normal unfertilized and some fertilized but uncleaved ZD, ZC and unmanipulated oocytes were
Table VI. Numbers (%) of degenerate oocytes or embryos from ZD and ZC cycles at 0, 17 and 42 h after insemination compared with embryos and oocytes from routine IVF cycles. Eleven ZD oocytes which were transferred as pronuclear embryos are not included
Number of oocytes
Zona drilled
Zona cut
Routine IVF
240
147
383
0
0
0
Oh
Statistical significance
17 h (%)
19 (7 9)'
12 (8.2)b
3 (0.8)c
a,b > c P < 0.01
42 h (%)
57 (23.8)1
18 (12.2)b
4(1.0) c
a,b > c P < 0.001 a > b P < 0.01
Table VD. The quality of ZD and ZC' embryos at transfer compared with embryo quality from routine IVF cycles (11 ZD and 28 routine IVF PN embryos are not included) Zona drilled
Zona cut
Routine IVF
Statistical significance
Embryos transferred (%) Grade 1 Grade 2 Grade 3
4 (12.If 26 (78.8)1 3(9.1)'
2 (6.1)b 30 (90.9)b 1 (3.0)b
Total
33
33
64 (45.7)c 74 (52.8)c 2 (1.4)c
a,b < c P < 0.001 a,b > c P < 0.01 a > c P < 0.05
140
Grade 1: all blastomeres intact, no fragmentation. Grade 2. slight fragmentation, unequal sized blastomeres. Grade 3: gross fragmentation.
fixed and processed for transmission electron microscopy. At fixation, many spermatozoa were observed bound to the zona pellucida and often a large mass of spermatozoa was seen in the vicinity of the hole or the cut. These spermatozoa were removed during subsequent steps of EM preparation. While most ZD and ZC oocytes had a similar ultrastructural appearance to intact oocytes in that cortical granules were numerous and closely applied to the plasma membrane and cell organelles were evenly distributed throughout the cytoplasm (Figure 1), other ZD and ZC oocytes demonstrated a cortical zone largely devoid of cell oganelles and cortical granules (Figure 2). This phenomenon was not observed in unfertilized control oocytes. One fertilized but 427
D.Payne el al.
1
zp,*-
•;
*>:-•• ?
.
Fig. 1. Electron micrograph of a normal 48-h-old unfertilized ZD human oocyte. The hole in the zona pellucida (ZP) is indicated by the asterisk. Note abundant cortical granules (CG). Magnification X5200.
Fig. 2. Electron micrograph of an abnormal 48-h-old unfertilized ZD human oocyte. There are very few cortical granules (CG) and the cortex of the oocyte is largely and atypically free of cell organelles (arrows). Magnification X520O.
uncleaved ZC oocyte showed atypical retention of cortical granules (Figure 3).
cycles. Additionally, we achieved a 13.9% fertilization rate in the second group of ZD patients whose poor semen parameters precluded them from routine IVF. Similarly, in 12 of the ZC patients, who had a total of 17 previous routine IVF cycles, ZC improved the number of oocytes fertilized (30.7 versus 13.5%). WHO (1987) recommends that semen with >50% abnormal forms be classed as teratozoospermic. Both groups of ZD patients had median normal forms well below that of WHO guidelines (21 and 14%), and significantly different fertilization rates after ZD (29.3 versus 13.9%). The difference in fertilization rates, and also the inability of ZD to increase fertilization rates to that of routine IVF in non-semen factor patients (89.8%) may be related to sperm morphology, since per cent normal morphology correlates strongly with fertilizing ability (Shalgi et al., 1985: Marsh et al., 1987). ZC like ZD was able to increase the fertilization rate compared with that seen in previous routine IVF cycles for the same patients. However, in patients who had both ZD and ZC cycles. ZC oocytes appeared to be less likely to be fertilized compared with ZD oocytes (19.0 versus 32.6%). ZD produces a hole of - 12 fim in diameter, compared with ZC which produces a cut — 40 /tm long but only — 2 /*m wide. The oocyte area available to spermatozoa after ZD is — 230 unr compared with an estimated 80 /tm: after ZC. This may account for the difference in fertilization rates between the two techniques. In this study, a total of 22 patients had 23 cycles of ZC. There were 13 cycles in which approximately half the oocytes collected were ZC and half were retained intact as a control group. Mahadevan and Trounson (1985) found that there was no difference in fertilization rates between intact and cumulus free oocytes. so control oocytes were not denuded. In five cycles there
Discussion The definitive test of sperm function is its ability to fertilize homologous oocytes. In our initial ZD trial, we had a population of infertile patients comprised largely of oligozoospermic and/or asthenozoospermic men, in whom IVF was precluded or the frequency of fertilization was low (8.3%). Approximately 30% of these patients had previously undergone up to three routine IVF cycles in which no fertilization was achieved. Of the 22 couples participating in the ZC trial, 12 had previously had up to three cycles of routine IVF with a fertilization rate of 15.3% and 10 patients had semen parameters which precluded routine IVF. Previous experience had suggested that sperm function in these couples was abnormal, and they were unlikely to achieve a pregnancy using standard IVF procedures. Hence, the development and use of micromanipulative techniques to increase fertilization rates in vitro was warranted. Although we achieved a significant improvement in fertilization rates after ZD. compared with routine IVF (29.3 versus 8.3%). it was less than that reported by Cohen etal. (1988) (30-50%). However, their group of male factor patients had fertilization rates of 24% in control (non-drilled) oocytes, substantially higher than that seen in our group of patients in their previous routine IVF cycles. It is possible therefore that the group of male factor patients admitted to our ZD trial had a more severe form of sperm dysfunction. Our results are similar to the results of Gordon et al. (1988). who achieved a fertilization rate of 32% with ZD oocytes from 10 patients who had no fertilization in previous 428
Zona manipulation in human IVF
ZP CG
Fig. 3. Electron micrograph of a fertilized uncleaved 48-h-old ZC human oocyte. There is atypical retention of cortical granules (CG). The cut in the zona pellucida (ZP) is indicated by arrows. Magnification X32OO.
was no difference in the fertilization rate of control oocytes and ZC oocytes (73.1 versus 52.0%). However, in eight cycles there was no fertilization in control oocytes, while a 26.7% fertilization rate was achieved in ZC oocytes. In a further 10 cycles, all oocytes were ZC; as based on semen parameters complete failure in fertilization of intact oocytes was anticipated. In these cycles 28.6% ZC oocytes were fertilized of which three were polyspermic. Thus ZC appears to have little effect on fertilization rates if fertilization was going to occur, yet enhanced fertilization rates where fertilization of intact oocytes did not occur, or was not anticipated. Of a total of 109 oocytes fertilized after either ZD or ZC, only five were polyspermic (4.6%). Gordon et al. (1988) reported a 50% incidence of polyspermy in ZD oocytes and proposed that the rapid block to polyspermy exists at the zona pellucida in the human. Polyspermy rates of 12 and 18% have been reported following partial zona dissection (PZD) compared with polyspermy rates in control oocytes of 20 and 29%, respectively (Maker and Cohen, 1989; Cohen et al., 1989). Sathananthan et al. (1989) and Fishel et al. (1990) obtained monospermic fertilization after multiple spermatozoa were transferred into the perivitelline space of human oocytes. We have observed motile spermatozoa swimming in the perivitelline space of normally fertilized ZC oocytes—a phenomenon similar to that seen in fertilized rabbit oocytes where there exists a very strong vitelline block to polyspermy (Yanagimachi, 1988) We have also observed normozoospermic fertilization in all six ZC and six intact oocytes from one patient, indicating that sperm function was normal and a vitelline block to polyspermy was operating. Malter et al. (1989) showed that the polyspermy rate in
normozoospermic patients was 29% for oocytes younger than 25 h post-oocyte recovery, but that it increased to 65% in aged oocytes. This suggests the existence of a vitelline block to polyspermy which diminishes as the oocyte ages. Increased rates of polyspermy are also observed in immature human oocytes (Trounson et al., 1982; Van der Ven et al., 1985), presumably due to the existence of a relatively 'soft' zona pellucida (Tesarik et al., 1988) and lack of an efficient vitelline block to polyspermy in immature oocytes. We postulate therefore that there is a transient rapid block to polyspermy at the level of the oocyte plasma membrane and the increased rates of polyspermy which have been observed in zona dissected oocytes is due to oocyte immaturity or post-maturity at the time of insemination, rather than zona manipulation bypassing a rapid block to polyspermy resident solely in the zona. Of the oocytes that were ZD, 24% exhibited morphological evidence of degeneration after 42 h in culture. Of the embryos that cleaved and were transferred, less were of grade one quality compared with those obtained from standard IVF cycles (12 versus 46%). Cohen et al. (1988) also found that the survival of embryos from ZD cycles was compromised, and suggested that this was a result of damage to the oocyte cytoplasm by the acidified Tyrodes used to dissolve the hole through the zona. In addition. Gordon et al. (1988) reported morphological abnormalities in some human ZD embryos after 40 h in culture. Whilst it is possible that fertilization by genotypically abnormal spermatozoa contributes to this phenomenon, early cleavage patterns at least to the two-cell stage appear to be determined by the (xxryte cytoplasm (Magnuson and Epstein, 1987; Pedersen. 1988). Further, decreased rates of implantation of ZD fertilized 429
D.Payne el al.
oocytes compared to control oocytes in the rat have been reported even though fertile males were used for insemination (Van der Hyden et al., 1989) and Ng et al. (1989) described a high proportion of human oocytes exhibiting meiotic arrest following ZD. This evidence, along with our own evidence of ultrastructural damage to some ZD human oocytes, and lack of a pregnancy after 21 transfers of ZD embryos strongly suggests that ZD may be detrimental to human oocyte viability. It was for this reason that we discontinued ZD in favour of ZC. Using ZC, we were able to decrease the number of degenerate oocytes and embryos to 12%, after 42 h in culture; however, there was no change in the low percentage of grade one embryos that were transferred. ZC is a more difficult technique, partly due to the small perivitelline space of human oocytes, and distortion of the oocyte during cutting may contribute to degeneration or poor cleavage patterns. Shrinking the oocyte in a hypertonic solution (Cohen et al., 1988) may reduce distortion during cutting. We have transferred 33 embryos which have resulted from insemination of ZC oocytes in 15 inter-uterine transfers and have not achieved a pregnancy. In our programme we would expect a pregnancy rate of 15—20% after the transfer of two embryos and hence at least three pregnancies should have been established from 15 transfers of ZC embryos. The lack of a pregnancy after ZC may be due to the high proportion of grade 2 embryos arising after ZC. However, in our present programme, grade 2 embryos regularly give rise to pregnancies. It has been suggested that the cells of zona-manipulated rabbit embryos which have been transferred to the oviduct are vulnerable to leukocytic attack (Moore et al., 1968). Alternatively, Nichols and Gardener (1989) have proposed that embryos with slit zonae may extrude their contents during contractions of the reproductive tract and hence are not able to produce pregnancies. Concurrent with transferring human PZD embryos to the uterus, Cohen et al. (1990) administered short-term immunosuppression with antibiotic cover and reported a pregnancy rate of 28%. Additionally, Cohen et al. (1990) have reported a clinical pregnancy rate of 44% after transferring embryos with slits made in the zonae subsequent to cleavage to assist in blastocyst hatching. This suggests that there may be some immunological involvement in the survival of zonamanipulated embryos in utero and may explain our inability to produce a pregnancy. We have now changed our protocol to include the regimen of immunosuppression used by Cohen et al. (1989). This clinical study of the zona drilling and zona cutting techniques has shown that it is possible to increase the rate at which spermatozoa from infertile males will fertilize human oocytes. There appears to be an inherent difficulty with the ZD technique which was not foreseen when used with mouse oocytes. The use of acid Tyrodes to dissolve a hole in the zona pellucida causes damage to at least one-quarter of the oocytes, and maybe more if abnormalities in cleavage are taken into account. Hence the clinical usefulness of ZD appears to be limited in the human. In contrast, zona cutting does not appear to be accompanied by so many morphologically abnormal oocytes and embryos, yet leads to a similar overall fertilization rate. Reduction in distortion to the oocyte during the cutting procedure may lead to improved embryo quality. 430
Acknowledgements The authors wish to thank Mr M.Makinson for photographic assistance, and Helen Holmes and Carol Burford for typing the manuscript. Dr S.P.Flaherty is thanked for a critical appraisal of the manuscript.
References CohenJ., Malter.H., FehUly.C, Wright.G., Elsner.C, Kort,H. and MasseyJ. (1988) Implantation of embryos after partial opening of oocyte zona pellucida to facilitate sperm penetration. Lancet, ii, 162. Cohen^I., Malter.H., Wright.G., Kort.H., Massey.J. and Mitchell,D. (1989) Partial zona dissection of human oocytes when failure of zona pellucida penetration is anticipated. Hum. Reprod., 4, 435—442. Cohen^I., Elsner.C, Kort,H., Malter,H., Massey.J., Mayer,M.P. and Weimer.K. (1990) Impairment of the hatching process following IVF in the human and improvement of implantation by assisting hatching using micromanipulation. Hum. Reprod., 5, 7—13. Depypere.H.T., McLaughlin.K.J., Seamark,R.F., Warnes.G.M. and Matthews,CD. (1988) Comparison of zona cutting and zona drilling as techniques for assisted fertilization in the mouse. J. Reprod. Fertil., 84, 205-211. Fishel,S., Antinori,S., Jackson,P., Johnson,!., Lisi,F., Chiariello,F. and Versaci,C. (1990) Twin birth after subzonal insemination. Lancet, i, 722-723. Forster.M.S., Smith.W.D., Lee,W.I., Berger,R.E., Karp,L.E. and Stenchever.M.A. (1983) Selection of human spermatozoa according to their relative motility and their interaction with zona-free hamster eggs. Fertil. Steril., 40, 655-660. Gordon,J.W. and Talansky.B.E. (1986) Assisted fertilization by zona drilling: a mouse model for correction of oligospermia. J. Exp. Tool., 239, 347-354. Gordon.J.W., Grunfeld,L., Garrisi,G.L., Talanski,B.E., Richards.C. and Laufer.N. (1988) Fertilization of human oocytes by sperm from infertile males after zona pellucida drilling. Fertil. Steril., 50, 68—73. Kerin^.F.P., Peek,J., Warnes.G.M., Kirby,C, Jeffrey.R., Matthews, C D . and Cox,L.W. (1984) Improved conception rate after intrauterine insemination of washed spermatozoa from men with poor quality semen. Lancet, i, 533-534. Lanzendorf.S.E., Maloney,M.K., Veeck,L.L., SlusserJ., Hodgen.G.D. and Rosenwaks.Z. (1988) A preclinical evaluation of pronuclear formation by microinjection of human spermatozoa into human oocytes. Fertil. Steril., 49, 835-842. Magnuson,T. and Epstein,CJ. (1987) Genetic expression during early mouse development. In Bavister.B.D. (ed.), The Mammalian Preimplantation Embryo. Regulation of Growth and Differentiation in Vitro. Plenum Press, New York, pp. 133-150. Mahadevan.M.M. and Trounson,A.O. (1985) Removal of the cumulus oophorus from the human oocyte for in vitro fertilization. Fertil. Steril., 43, 263-267. Malter.H.E. and Cohen.J. (1989) Partial zona dissection of the human oocyte: a non traumatic method using micromanipulation to assist zona pellucida penetration. Fertil. Steril., 51, 139-148. Malter,H., Talansky,B., GordonJ. and Cohen,J. (1989) Monospermy and polyspermy after partial zona dissection of reinseminated human oocytes. Gamete Res., 23, 377-386. Mann,J.R. (1988) Full term development of mouse eggs fertilized by a spermatozoon mkroinjected under the zona pellucida. Biol. Reprod., 38, 1077-1083. Marsh,S.K., Bolton.V.N. and Braude.P.R. (1987) The effect of morphology on the ability of human spermatozoa to penetrate zona free hamster oocytes. Hum. Reprod., 2, 499-503. Moore.N.W., Adams.CE. and Rowson.L.E.A. (1968) Developmental potential of single blastomeres of the rabbit egg. J. Reprod. Fertil., 17, 527-534.
Zona manipulation in human IVF Ng,S.C, Bongso.A., Ratnam.S.S., Sathananthan.H., Chan.C.L.K., Wong,P.C, Hagglund,L., Anandakumar.C, Wong,Y.C. and Goh.V.H.H. (1988) Pregnancy after transfer of sperm under zona. Lancet, il, 790. Ng,S.C, Bongso.T.A., Chang,S.I., Sathananthan.A.H. and Ratnam.S.S. (1989) Transfer of human sperm into the perivitelline space of human oocytes after zona drilling or zona puncture. Fertil. Steril., 52, 73-78. Nichols,J. and Gardner.R.L. (1989) Effect of damage to the zona pellucida on development of preimplantation embryos in the mouse. Hum. Reprod., 4, 180-187. Pedersen,R.A. (1988) Early mammalian embryogenesis. In Knobil,E., Neill.J.D., Ewing.L.L., Greenwald,G.S., Markert.C.L. and Pfaff.D.W. (eds), The Physiology of Reproduction. Raven Press, New York, p. 187. Quinn,P., Kerin,J.K. and Warnes.G.M. (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil. Steril., 44, 493-498. Sathananthan,A.H., Ng.S.C, Trounson.A., Bongso,A., Laws-King,A. and Ratnam,S.S. (1989) Human micro-insemination by injection of single and multiple sperm: ultrastructure. Hum. Reprod., 4, 574-583. Shalgi,R., Dor,J., Rudak,E., Lusky.A., Goldman,B., Mashiach,S. and Nebel.L. (1985) Penetration of sperm from teratospermic men into zona-free hamster eggs. Int. J. Androi, 8, 285—294. Tesarik,J., Pilka,L. and Travnik.P. (1988) Zona pellucida resistance to sperm penetration before the completion of human oocyte maturation. J. Reprod. Fertil., 83, 487-495. Trounson,A.O., Mohr.L.R., Wood.C. and Leeton,J.F. (1982) Effect of delayed insemination on in-vitro fertilization, culture and transfer of human oocytes. J. Reprod. Fertil, 64, 285-294. Tsunoda.Y., Yasui,T., Nakamura.K., Uchida.T. and Sugie.T. (1986) Effect of cutting the zona pellucida in the pronuclear transplantation in the mouse. J. Exp. Tool, 240, 119-125. Uehara.T. and Yanagimachi.R. (1976) Microsurgical injection of spermatozoa into hamster eggs with subsequent transformation of sperm nuclei into male pronuclei. Biol. Reprod., 15, 467—470. Van der Hyden,B., McLaughlin.K.J., Rutledge.J.M. and Armstrong, D.T. (1989) Zona drilling increases the penetrability of rat oocytes matured in vivo. Biol. Reprod., 40, 953-960. Van der Ven.H.H., Al-Hasani.S., Diedrich.K., Hamerich,U., Lehmann,F. and Krebs.D. (1985) Polyspermy in in vitro fertilization of human oocytes: frequency and possible causes. Ann. N. Y. Acad. Sci., 442, 88-95. World Health Organization (1987) WHO Laboratory Manual for the Examination of Human Semen and Semen —Cervical Mucus Interaction. Cambridge University Press, Cambridge, p. 27. Yanagimachi.R. (1988) Mammalian fertilization. In Knobil.E., Neill.J.D., Ewing.L.L., Greenwald,G.S., Markert,C.L. and Pfaff,D.W. (eds), The Physiology of Reproduction. Raven Press, New York, pp. 135-185. Yates.C.A. and De Kretser,D.M. (1987) Male-factor infertility and in vitro fertilization. J. In Vitro Fertil. Embryo Transfer, 4, 141 — 147. Received on May 11, 1990; accepted on October 5, 1990
431
