More than 90% fertilization rates after intracytoplasmic sperm injection and artificial induction of oocyte activation with calcium ionophore*

Transcription

1 FERTILITY AND STERILITY Copyright c 1995 American Society for Reproductive Medicine Vol. 63, No.2, February 1995 Printed on acid-free paper in U. S. A. More than 90% fertilization rates after intracytoplasmic sperm injection and artificial induction of oocyte activation with calcium ionophore* Jan Tesarik, M.D., Ph.D.t Mario Sousa, M.D., Ph.D.:!: American Hospital of Paris, Neuilly sur Seine, France, and Institut National de la Sante et de la Recherche Medicale (INSERM), Unit 355, Clamart, France Objective: To examine whether fertilization rates after intracytoplasmic sperm injection can be increased by artificial oocyte activation. Design: Oocytes that failed to fertilize spontaneously by 24 hours after intracytoplasmic sperm injection were treated either with calcium ionophore to induce activation or with solvent only to serve as control. The ability of ionophore-treated and control oocytes to achieve delayed fertilization was compared. Setting: Private hospital and public research center. Patients: Infertile couples treated by intracytoplasmic sperm injection. Interventions: Intracytoplasmic sperm injection. Main Outcome Measures: Fertilization and cleavage rates. Results: The mean rate of spontaneous fertilization after intracytoplasmic sperm injection was 32%, but 88% of the oocytes that failed to fertilize spontaneously did so after subsequent exposure to calcium ionophore. Most of these oocytes underwent at least one apparently normal cleavage division. In contrast, delayed fertilization of oocytes not treated with ionophore was an exceptional finding. If only oocytes remaining intact after intracytoplasmic sperm injection are taken into account, the mean global fertilization rate of ionophore-enhanced intracytoplasmic sperm injection was 91%. Conclusions: These results show that the failure of oocyte activation is the main cause of fertilization failure after intracytoplasmic sperm injection. If an appropriate, clinically applicable treatment is found to overcome this problem, intracytoplasmic sperm injection can be expected to yield fertilization rates far exceeding those of standard IVF with normal spermatozoa. Fertil Steril 1995;63:343-9 Key Words: Intracytoplasmic sperm injection, oocyte activation, calcium, ionophore A23187 The current era of IVF is marked by a growing proportion of andrological indications. This is ap- Received December 6, 1993; revised and accepted September 12,1994. * Supported in part by a grant from the "Junta de Andalucia" reseach group 3137 (J.T.). t Reprint requests: Jan Tesarik, M.D., Ph.D., Center for Reproductive Biology and Medicine, American Hospital of Paris, 63 Boulevard Victor Hugo, Neuilly sur Seine, France (FAX: ). * Present address: Laboratory of Cell Biology, Institute ofbiomedical Sciences, University of Oporto, 4000 Porto, Portugal. parently because of accumulation of cases resistant to IVF treatment, those in which spermatozoa are unable to fertilize even under in vitro conditions. Micromanipulation-assisted fertilization has brought new hope to these patients. So subzonal insemination (SUZI) has proven successful in cases in which standard IVF failed (1-3). However, spermatozoa from some infertile men show a low fertilization performance after SUZI (4), possibly because of a defect of sperm-oocyte fusion (5). Direct intracytoplasmic sperm injection obviously can overcome this problem. As a matter of fact, pub- Vol. 63, No.2, February 1995 Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection 343

2 lished data about intracytoplasmic sperm injection results compare favorably with the efficacy of the other methods of micromanipulation -assisted fertilization. The rates of fertilized-intact injected 00- cytes presented in five successive papers by the Brussels Free University group were 66% (31/47) (6), 44% (240/548) (7), 51% (803/1,575) (8), 65% (830/1,280) (9), and 65% (4,546/6,994) (10). However, another group has reported a smaller series of intracytoplasmic sperm injection cases with fairly lower fertilization rates (47/251 [19%]) (11). It is known that for successful fertilization, the spermatozoon must activate the oocyte. The mechanism of this process in humans is still understood poorly. Moreover, there is some concern of how oocyte activation proceeds after intracytoplasmic sperm injection that bypasses the normal sequence of fertilization events. Edwards and Van Steirteghem (12) hypothesized that oocytes are activated by the injection of high-calcium medium during intracytoplasmic sperm injection. Notwithstanding, recent data have shown that the calcium transient produced by the intracytoplasmic sperm injection procedure is not sufficient to activate the human oocyte and that activation is started after a considerable lag period, probably by a soluble factor released from the injected spermatozoon (13). Consequently, if the injected spermatozoon is deficient in this presumptive oocyte-activating factor or if the oocyte does not respond to it for various reasons, intracytoplasmic sperm injection will not work. Interestingly, relatively low frequencies of spontaneous fertilization after intracytoplasmic sperm injection were reported in bovine (14, 15) and rabbit (16), but fertilization results in both species were improved significantly by a short incubation of sperm-injected oocytes with calcium ionophore. The present study was undertaken to assess whether fertilization rates after intracytoplasmic sperm injection in humans also can be increased by artificial oocyte activation. MATERIALS AND METHODS Experimental Design If the fertilization failure after intracytoplasmic sperm injection is caused essentially by a failure of oocyte activation, at least some of the oocytes that do not fertilize spontaneously after intracytoplasmic sperm injection would be likely to do so after subsequent exposure to an artificial activating agent. If, on the contrary, the fertilization failure is caused mainly by other factors, the artificially activated oocytes would not fertilize and would develop merely a female pronucleus. To distinguish between these two possibilities, we used oocytes from 15 patients treated by intracytoplasmic sperm injection at our center in whom fertilization was achieved spontaneously in some but not all oocytes. Intracytoplasmic sperm injection was performed following a protocol granted by the National Committee for the Protection of Persons. Oocytes developing spontaneously two pronuclei (2PN) by 24 hours after intracytoplasmic sperm injection were cultured separately and transferred to the patient after an additional day of culture. Oocytes that do not fertilize spontaneously by 24 hours after intracytoplasmic sperm injection generally are considered as abnormal and are not transferred (8, 9). Such oocytes were treated with ionophore and examined for the pronuclear development during an additional 24 hours. Some of these oocytes were cultured further to assess their capacity to cleave. None of these oocytes was transferred. Unfertilized oocytes from a similar group of other 10 patients treated by intracytoplasmic sperm injection were exposed to dimethylsulphoxide (DMSO) (solvent for the ionophore) at the same concentration and under the same conditions as the oocytes from the experimental group; these oocytes served as controls. Patients The criteria for patient inclusion in our intracytoplasmic sperm injection program and the protocol of ovulation induction were essentially those described previously for SUZI (17). Briefly, intracytoplasmic sperm injection was indicated in patients in which at least two previous IVF attempts gave zero fertilization, irrespective of the results of sperm examination. Alternatively, intracytoplasmic sperm injection was performed as a first treatment option in cases in which the number of motile, morphologically normal spermatozoa in the ejaculate was =s;;100,000, which indicated that the success of standard IVF would be highly improbable. The presence of at least a few living spermatozoa in the sediment after centrifugation of the ejaculate (assessed by supravital eosin staining) was the only requirement for sperm quality. No restriction ofpatient age was applied. The mean age of the husband was 36 years (range, 29 to 54) and that of the wife 34 years (range, 28 to 42). Preparation of Micromanipulation Instruments Microinjection needles and holding pipettes were prepared from Narishige G-l glass capillary tubes 344 Tesarik and Sousa Ionoplwre-enhanced intracytoplasmic sperm injection Fertility and Sterility

3 (Narishige Co., Tokyo, Japan). For the preparation of micro injection needles, the capillaries were pulled in a Campden Model 773 computer-controlled programmable horizontal micropipette puller (Campden Instruments, London, United Kingdom) with the use of a five-stage program including three pulling stages and two cooling stages. A beveled tip was obtained by grinding the needles at an angle of 30 in a Narishige EG-4 micropipette grinder. Finally, an extra sharp spike was made at the extremity of the beveled tip with the use of a De Fonbrune microforge (Alcatel, Malakoff, France). The final part of the needle was then bent to an angle of 30 to facilitate the horizontal approach to the oocyte during puncture. The internal diameter of the needles was 6 ~m. Holding pipettes were pulled with the use of a one-stage program and fire-polished in the microforge. Their external and internal diameters were 100 and 20 ~m, respectively. More details of these techniques have been published previously (4, 13, 17). Micromanipulation Techniques Microinjection was performed 2 to 6 hours after oocyte recovery with the use of Narishige manipulators (MN-151 mechanicaljoystickmanipulatorcontrolling the holding pipette, and MO-202 hydraulic remote-control manipulator controlling the microinjection needle) mounted on an Olympus IMT- 2 inverted research microscope equipped with a heated stage and with Hoffman modulation contrast optics (Olympus Europe, Hamburg, Germany). The procedure was carried out at 37 C in microdrops (3 ~L) of culture medium placed in plastic 35 X 10-mm tissue culture dishes (Costar, Cambridge, MA). The medium used was Sperm Preparation Medium (Medi-Cult, Copenhagen, Denmark) equilibrated with 5% CO 2 in air, and the microdrops of medium were covered with embryotested light mineral oil (Sigma, La Verpilliere, France). The ph values of the equilibrated medium were 7.3 to 7.4, and they did not change during 20 minutes under oil outside ofthe CO 2 incubator, presumably because of the stabilizing effect of the HEPES-buffer system making part of the medium. This time was sufficient to carry out intracytoplasmic sperm injection with each group of oocytes injected at a time. Previously described techniques were used for sperm preparation and injection (7, 8, 13). To facilitate the micromanipulation, four microdrops were placed in series in the culture dish. The first microdrop contained a suspension of spermatozoa to be injected in a solution of 10% polyvinylpyrrolidone (PVP) (average molecular weight, 360,000; ICN Biochemicals, Cleveland, OH) (13). The second microdrop was the 10% PVP solution only, whereas two to three oocytes were located in each of the third and fourth microdrops containing Sperm Preparation Medium. Only metaphase II oocytes were allocated to intracytoplasmic sperm injection. An oocyte was considered to be at metaphase II when it possessed one intact or fragmented polar body. When sperm motility was very poor or absent, PVP was not included in the first microdrop. A spermatozoon lacking apparent morphological abnormalities was aspirated from the first microdrop into the microinjection needle, washed in the second microdrop, immobilized if necessary by touching the flagellum with the needle, and injected into one of the oocytes in the third or the fourth microdrop. This procedure was repeated until all oocytes in the dish were injected. The time required for the injection of one oocyte (including the selection and washing of the spermatozoon) usually did not exceed 3 minutes. When all oocytes had been injected, they were washed immediately in B2 medium and returned back to the CO 2 incubator. Evaluation of Fertilization Results and Ionophore Treatment Oocytes were inspected 16 hours after intracytoplasmic sperm injection for the presence and number of pronuclei and of polar bodies. If no pronuclei were seen at this time, the oocyte was assessed again at 24 hours after intracytoplasmic sperm injection. Oocytes that developed spontaneously 2PN by that time were cultured separately for an additional 24 to 28 hours and then transferred back to the patient at the two-cell or four-cell stage of preimplantation development. Those oocytes that did not develop any pronucleus spontaneously by 24 hours after intracytoplasmic sperm injection and that had not separated a second polar body were incubated for 15 minutes (37 C at 5% CO 2 in air) in B2 medium containing 10 ~M ionophore A23187 (Sigma). The ionophore was added to the medium from a 4-mM stock solution prepared by dissolving the ionophore in DMSO (Sigma). Ten microliters ofthis stock solution was then added to 1 ml B2 medium, and 0.1 ml of this intermediate solution was mixed with 0.3 ml of B2 medium to give the final ionophore concentration of 10 ~M. Accordingly, the final concentration of DMSO in the oocyte incubation medium was Vol. 63, No.2, February 1995 Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection 345

4 0.25%. The incubation of the oocytes with ionophore was finished before the 25th hour after intracytoplasmic sperm injection. Oocytes were then washed thoroughly in fresh B2 medium and incubated under 5% CO 2 in air for an additional 1 to 2 days. During this period they were examined several times for signs of fertilization and cleavage. None of these oocytes was transferred to the patient. Statistical Analysis Percentages of fertilized oocytes in the ionophore-treated and control group were transformed by arc-sin. They were then compared by Student's t-test. RESULTS Spontaneous fertilization results in the cytes forming the experimental group are summarized in Table 1. On average, 32% of the injected oocytes fertilized spontaneously by 24 hours after intracytoplasmic sperm injection. However, 88% of the oocytes that did not fertilize spontaneously developed 2PN after the exposure to calcium ionophore (Fig. la). Fertilization results after ionophore treatment obtained with oocytes from individual patients are shown in Table 2. Together, the spontaneous fertilizations and those achieved after ionophore treatment accounted for an overall mean fertilization rate of 83%. Ten oocytes (9%) were damaged during the intracytoplasmic sperm injection procedure and degenerated by the time of the first inspection for the presence of pronuclei. If these oocytes not treated with ionophore are not taken into account, the overall fertilization rate calculated for the oocytes remaining intact after intracytoplasmic sperm injection was as high as 91 %. It has to be noted that the spontaneous development of one-pronucleated oocytes after intracytoplasmic sperm injection was observed rarely (1 %), but this condition was much more frequent after ionophore treatment (9%). In the control group of 76 oocytes, the rate of spontaneous fertilization by 24 hours after intracytoplasmic sperm injection was 33% (25 of76). None of the 51 oocytes that failed to fertilize spontaneously in this group showed any pronucleus, and all these oocytes were treated subsequently with 0.25 % DMSO (solvent for ionophore) under the same conditions as the oocyte from the experimental group. Fifty of these oocytes did not show any pronucleus during 24 hours after the exposure to DMSO; 2PN developed in one oocyte of this group (significant difference from the experimental group; P < 0.001). Of the total of 63 oocytes that fertilized spontaneously in the experimental (15 cycles) and control (10 cycles) group, 57 underwent at least one cleavage division and were transferred to the patient's uterus (25 ET procedures). Six of these patients showed positive serum hcg levels indicating the beginning of implantation, but 1 patient suffered an early abortion. Fetal heart was detected with all the remaining five clinical pregnancies. Further development of the oocytes fertilized after intracytoplasmic sperm injection and subse- Table 1 Spontaneous Fertilization Results Achieved by 24 Hours After Intracytoplasmic Sperm Injection No. of metaphase II oocytes No. of oocytes developing different nos. of PN by 24 hours after intracytoplasmic sperm injection Intact after intracytoplasmic Patient Injected sperm injection OPN IPN 2PN 3PN Totals Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection Fertility and Sterility

5 quent ionophore treatment was not followed consistently because some of the zygotes were processed for electron microscopy and others for cytogenetic analysis. However, 34 of the 59 zygotes were maintained in culture to evaluate their developmental potential. After an additional 24 hours, 5 (15%) ofthese zygotes remained at the pronuclear stage and 1 (3%) degenerated. The remaining 28 zygotes (82%) underwent cleavage to at least the two-cell stage (Fig. IB), and some achieved the four-cell stage (Fig. Ie). The development of these embryos beyond the four-cell stage was not attempted in this study. DISCUSSION We have shown recently that the calcium transient occurring in human oocytes during the intracytoplasmic sperm injection procedure is not a sufficient stimulus for oocyte activation, that activation fails in many oocytes having undergone intracytoplasmic sperm injection and when it occurs, it is initiated no sooner than several hours after the injection (13). Because most human oocytes are activated within 30 minutes after in vitro insemination when previously freed from the zona pellucida (Tesarik J, Sousa M, unpublished observation), it appears that in normal fertilization, oocyte activation begins at the time of sperm-oocyte fusion or soon thereafter. The observed delay or failure of oocyte activation after intracytoplasmic sperm injection may thus be caused either by the lack of some signaling event associated with the interaction between the sperm and oocyte plasma membranes or by perturbation by the micro injection procedure of the normal organization of cytoplasmic structures in the oocyte cortex, resulting in the oocyte inability to transduce properly the signal delivered by the injected spermatozoon to intracellular calcium stores. These findings have raised a question of whether the oocytes that failed to fertilize spontaneously after intracytoplasmic sperm injection would do so if activation did occur. Alternatively, it was possible that other factors, such as the failure ofthe injected Figure 1 Development of an oocyte that failed to fertilize spontaneously by 24 hours after intracytoplasmic sperm injection and was then activated artificially by calcium ionophore. When inspected 16 hours after ionophore treatment, the oocyte showed 2PN (A). When examined after an additional 12 hours and 24 hours, it had developed to the two- (B) and four-cell (C) stage, respectively. Nuclei are pointed with arrows where visible. Vol. 63, No.2, February 1995 Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection 347

6 Table 2 Delayed Fertilization Results Achieved by 24 Hours After Ionophore Treatment of Oocytes That Failed to Fertilize Spontaneously by 24 Hours After Intracytoplasmic Sperm Injection* Patient No. of OPN oocytes treated with ionophore 24 hours after intracytoplasmic sperm injection No. of oocytes with different nos. of PN by 24 hours after ionophore treatment OPN IPN 2PN 3PN Totals See the OPN column of Table 1. spermatozoon to shed its membranes or the failure of sperm chromatin to undergo changes leading to the formation of the male pronucleus, also contributed substantially to fertilization failure after intracytoplasmic sperm injection so that artificial oocyte activation would not be sufficient to rescue such oocytes. The results of this study have shown that such a complex disorder of the fertilization mechanism is not a common finding after intracytoplasmic sperm injection and that the absence of oocyte activation is the only cause of the absence of fertilization in most cases when intracytoplasmic sperm injection fails. Our results also have pointed out that most of the oocytes that fail to fertilize spontaneously after intracytoplasmic sperm injection can do so when activated artificially by calcium ionophore. It must be remembered, however, that the ionophore treatment was applied between the 24th and 25th hour after intracytoplasmic sperm injection, and it is not known whether the same efficacity could be achieved if the oocytes were treated earlier after the injection. In particular, it is not known how much time after intracytoplasmic sperm injection is needed for sperm chromatin to become accessible to oocyte cytoplasmic factors. If ionophore is added before this time, there may be a risk that only the oocyte chromatin will respond, which could lead to asynchrony between the development of both pronuclei. Even though the developmental consequences of such an asynchrony are not known, the increased incidence of one-pronucleated oocytes observed in this study after ionophore treatment is a clear warning in this sense. Another question arises as to the ability of the ionophore-treated oocytes to reproduce completely the biochemical events that are set normally in motion by the fertilizing spermatozoon. In fact, the calcium transient occurring during normal fertilization has the form of repeated sharp increases of intracellular free calcium in all mammalian species studied so far including the human (18). To the best of our knowledge, ionophores have never been reported to produce a similar pattern of calcium oscillations. In spite of this, bovine oocytes activated with calcium ionophore after intracytoplasmic sperm injection have been shown to develop into viable embryos that were transferred into recipient cows and resulted in the birth of normal calves (15). This may mean that either calcium oscillations at fertilization are not an absolutely necessary prerequisite for normal development in all mammalian species or that some kind of calcium oscillations can in fact be produced when oocytes are exposed to ionophore after intracytoplasmic sperm injection. It is to be recalled in this context that the spermatozoon is already present in the oocyte cytoplasm at the time of ionophore treatment so that factors released previously from the spermatozoon to the ooplasm may have conditioned the oocyte to respond to ionophore in a different way than a nonconditioned cell. This point certainly deserves further focused study. In theory, one also could imagine that calcium 348 Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection Fertility and Sterility

7 ionophore may support fertilization after intracytoplasmic sperm injection by inducing the acrosome reaction (AR) in spermatozoa that have been injected into oocytes with still intact acrosomes. However, the induction of the human sperm AR with calcium ionophores at concentrations similar to that used in this study is known to require milimolar concentrations of external calcium (19). In contrast, free cytosolic calcium concentration in human oocytes, observed during fertilization-induced calcium spikes, achieves maximal levels in the micromolar range (18), and ionophore-induced calcium increases attain similar maxima (Tesarik J, unpublished data). With a micromolar concentration of ionophore in the oocyte-bathing medium (and probably much lower within the oocyte) and with a concentration of calcium approximately one thousand times lower that the optimum, the possibility of a direct action of calcium ionophore on the AR of injected spermatozoa appears unlikely. As a corollary, in spite of the absence of apparent toxicity of ionophore under the conditions of this study (100% survival ofthe treated oocytes), these results should by no means encourage the use of ionophore to boost freshly injected human oocytes until the mechanism of action of the ionophore and the potential risk associated with its use fully are elucidated. ADDENDUM After the acceptance of this paper, some of the points raised in the Discussion have been clarified. The treatment with a calcium ionophore has been shown to support the development of the physiological pattern of activation in sperm-injected, but not control, human oocytes (J. Tesarik and J. Testart, Biology of Reproduction 1994;51:385-91). Moreover, by using a modified injection technique producing a major calcium influx into oocytes, similar to that provoked by ionophore, we managed to increase the rate of spontaneous fertilization after intracytoplasmic sperm injection to 81 % (233 twopronucleated eggs out of 288 metaphase II oocytes injected over the last two months). REFERENCES 1. Fishel S, Jackson P, Antinori S, Johnson J, Grossi S, Versaci C. Subzonal insemination for the alleviation of infertility. Fertil Steril1990;54: Ng SC, Bongso A, Ratnam SS. Microinjection of human oocytes: a technique for severe oligoasthenoteratozoospermia. Fertil Steril 1991;56: Cohen J, Alikani M, Adler A, Berkeley A, Davis 0, Ferrara T A, et al. Microsurgical fertilization procedures: the absence of stringent criteria for patient selection. J Asst Reprod Genet 1992;9: Tesarik J. Subzonal sperm insertion with aged human 00- cytes from failed in-vitro fertilization attempts: fertilization results and some applications. Hum Reprod 1993;8: Tesarik J, Thebault A. Fertilization failure after subzonal sperm insertion associated with defective fusional capacity of acrosome-reacted spermatozoa. Fertil Steril1993;60: Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of a single spermatozoon into an oocyte. Lancet 1992;340: Palermo G, Joris H, Derde M-P, Camus M, Devroey P, Van Steirteghem A. Sperm characteristics and outcome of human assisted fertilization by subzonal insemination and intracytoplasmic sperm injection. Fertil Steril 1993;59: Van Steirteghem AC, Liu J, Joris H, Nagy Z, Janssenswillen C, Tournay H, et al. Higher success rate by intracytoplasmic sperm injection than by subzonal insemination. Report of a second series of 300 consecutive treatment cycles. Hum Reprod 1993;8: Van Steirteghem AC, Nagy Z, Joris H, Liu J, Staessen C, Smitz J, et al. High fertilization and implantation rates after intracytoplasmic sperm injection. Hum Reprod 1993;8: Van Steirteghem A, Liu J, Nagy Z, Joris H, Tournay H, Liebaers I, et al. Use of assisted fertilization. Hum Reprod 1993;8: Tucker M, Wiker S, Massey J. Rational approach to assisted fertilization. Hum Reprod 1993;8: Edwards RG, Van Steirteghem AC. Intracytoplasmic sperm injection (ICSI) and human fertilization: does calcium hold the key to success? Hum Reprod 1993;8: Tesarik J, Sousa M, Testart J. Human oocyte activation after intracytoplasmic sperm injection. Hum Reprod 1994;9: Keefer CL, Younis AI, Brackett BG. Cleavage development of bovine oocytes fertilized by sperm injection. Mol Reprod Dev 1990;25: Goto K. Bovine microfertilization and embryo transfer. Mol Reprod Dev 1993;36: Hosoi Y, Iritani A. Rabbit microfertilization. Mol Reprod Dev 1993;36: Tesarik J, Mendoza C. Most living acrosome-reacted spermatozoa do not fuse with the oocyte when inserted into the perivitelline space. Fertil Steril 1994;61: Taylor CT, Lawrence YM, Kingsland CR, Biljan MM, Cuthbertson K. Oscillations in intracellular free calcium induced by spermatozoa in human oocytes at fertilization. Hum Reprod 1993;8: Tesarik J. Comparison of acrosome reaction -inducing activities of human cumulus oophorus, follicular fluid and ionophore A23187 in human sperm populations of proven fertilizing ability in vitro. J Reprod FertiI1985;74: Vol. 63, No.2, February 1995 Tesarik and Sousa Ionophore-enhanced intracytoplasmic sperm injection 349