High frequency of defective sperm±zona pellucida interaction in oligozoospermic infertile men

Transcription

1 Human Reproduction Vol.19, No.2 pp. 228±233, 2004 DOI: /humrep/deh067 High frequency of defective sperm±zona pellucida interaction in oligozoospermic infertile men De Yi Liu 1 and H.W.Gordon Baker University of Melbourne Department of Obstetrics and Gynaecology, Reproductive Services, Royal Women's Hospital and Melbourne IVF, Victoria 3035, Australia 1 To whom correspondence should be addressed. dyl@unimelb.edu.au BACKGROUND: The ability of sperm to interact with the zona pellucida (ZP) plays a critical role during the process of human fertilization. The aim of this study is to determine frequency of defective sperm±zp interaction in oligozoospermic infertile men. METHODS: Sperm±ZP binding assays and the ZP-induced acrosome reaction (AR) were performed in 72 infertile men with a sperm concentration < /ml. Oocytes that had previously failed to fertilize in a clinical IVF programme were used for the tests. Motile sperm ( /ml) selected by swim-up from each semen sample were incubated with four oocytes for 2 h. The number of sperm bound per ZP and the ZP-induced AR were assessed. Under these conditions, an average of <40 sperm bound/zp was de ned as low sperm±zp binding and a ZP-induced AR <16% was de ned as low ZP-induced AR. RESULTS: In the 72 oligozoospermic men, 28% (20/72) had low sperm±zp binding. Of those with normal sperm±zp binding, 69% (36/52) had low ZP-induced AR. Overall, 78% (56/72) had either low ZP-binding or normal ZP binding but low ZP-induced AR. This means that only 22% (16/72) had both normal sperm±zp binding and normal ZP-induced AR. CONCLUSION: Oligozoospermic men have a very high frequency of defective sperm±zp interaction, consistent with their low natural fertility or low fertilization rate in conventional IVF. Infertile couples with oligozoospermic semen should be treated by ICSI rather than by conventional IVF. Key words: male infertility/oligozoospermia/sperm±zona pellucida interaction Introduction Sperm concentration in semen is correlated with human pregnancy rates (Baker, 2001). Traditionally, oligozoospermia is de ned as a sperm concentration < /ml (World Health Organization, 1999). Although the low sperm number may be the main factor reducing the chance of natural conception, low motility and poor sperm morphology often accompany and accentuate the defect regardless of the cause of the oligozoospermia (Liu and Baker, 1992). Although many oligozoospermic men have enough motile sperm for them to be harvested by sperm preparation techniques for insemination of oocytes by standard IVF, previous experience of IVF treatment in couples with moderate-to-severe oligozoospermia (sperm count < /ml) was generally unsatisfactory. The fertilization rate was usually low, averaging ~35% although some pregnancies were achieved (Yovich et al., 1985; Baker et al., 1993). Natural pregnancies have also been reported in men with severe oligozoospermia (sperm concentration < /ml) (Bostofte et al., 1982; Burger and Baker, 1984; Sokol and Sparkes, 1987; World Health Organization Task Force, 1990). Therefore, other sperm defects rather than low sperm number alone may be critical for the severity of infertility in oligozoospermic men. The capacity of sperm to interact with the oocyte during the process of fertilization is crucial and involves the sequential steps: sperm±zp binding, the ZP-induced AR, sperm±zp penetration and sperm±oolemma fusion (Yanagimachi, 1994; Wassarman, 1999). While abnormalities of either sperm or oocytes could lead to failure of fertilization, in conventional IVF, failure of fertilization of all or most of the oocytes collected from the one woman is usually caused by defective sperm±zp binding and penetration associated mainly with abnormal sperm morphology and disordered ZP-induced AR (Liu and Baker, 2000; Liu et al., 2001). Oocyte immaturity affects normal fertilization but does not affect sperm±zp binding and penetration in vitro (Overstreet et al., 1976; Lopata and Leung, 1988). Overstreet et al. (1976, 1980) originally reported that testing sperm±zp interaction using immature oocytes recovered from surgically removed human ovaries provides an important indicator for defective sperm fertilizing ability that may not be revealed by routine semen analysis. More recent studies clearly show that tests for sperm±zp 228 Human Reproduction vol. 19 no. 2 ã European Society of Human Reproduction and Embryology 2004; all rights reserved

2 Sperm±ZP interaction in oligozoospermic men binding and the ZP-induced AR using unfertilized human oocytes obtained from clinical IVF provide useful markers for sperm fertilizing ability in vitro (Burkman et al., 1988; Liu et al., 1988; Oehninger et al., 1989; Liu et al., 2001). Aitken et al. (1989) evaluated sperm function in oligozoospermic men using the calcium ionophore A23187-enhanced human sperm ZP-free hamster oocyte penetration assay and found that 88% of oligozoospermic men (n = 74) had <25% penetration and 58% had no penetration. This suggests that a high proportion of oligozoospermic men may have low calcium ionophore A23187-induced AR or other sperm defects preventing sperm±oolemma fusion, oocyte penetration and sperm head decondensation. Using the human ZP-induced AR test, we showed a high frequency of defective sperm±zp interactions in severely teratozoospermic infertile men (strict normal sperm morphology <5%, sperm count > /ml and motility >30%) (Liu and Baker, 2003b). In this study we have investigated the frequency of defective sperm±zp binding and the ZP-induced AR in 72 oligozoospermic (< /ml) infertile men who had suf cient motile sperm harvested by swim-up for insemination of oocytes in vitro. Materials and methods Subjects Semen samples were obtained from 72 oligozoospermic infertile men (sperm count < /ml) who attended infertility clinics in both The Royal Women's Hospital and Melbourne IVF between February 1996 to November The oligozoospermia was caused by untreatable primary seminiferous tubule defects. Human oocytes Oocytes which showed no evidence of two pronuclei or cleavage at 48±60 h after insemination in the clinical IVF programme were used for the sperm±zp interaction tests. If the oocyte had sperm bound to the ZP from the IVF insemination, these were removed by aspiration using a ne glass pipette with an inner diameter (120 mm) slightly smaller than the oocyte diameter (Liu and Baker, 1996a). Most of the oocytes were obtained from patients with partial failure of fertilization in standard IVF and >50% of these unfertilized oocytes had one or a few sperm penetrating the ZP from the IVF insemination. We have shown previously that oocytes with <10 sperm penetrating the ZP have an ability for subsequent sperm±zp binding and ZP-induced AR similar to those without sperm penetration (Liu and Baker, 1996a). Degenerate, activated or morphologically abnormal oocytes, as well as oocytes with >10 sperm penetrating the ZP were not used for the test. Oocytes were pooled from several patients, kept in the 5% CO 2 incubator and used within 3 days. All patients signed consent forms permitting use of their unfertilized oocytes or sperm samples for research. The Royal Women's Hospital Research and Ethics Committees approved the project. Semen analysis Semen samples were obtained by masturbation after 2±5 days abstinence. All sperm tests were performed after liquefaction of the semen within 1 h. Sperm concentration and motility in semen were determined using standard methods (World Health Organization, 1999). Semen samples with sperm counts < /ml or total motility <20% were not used in this study since they provided insuf cient motile sperm for performing the sperm±zp interaction test. The leukocyte concentration in samples used was not determined using speci c staining, but samples with high numbers (> /ml) of round cells were excluded from the study. Morphology of sperm was assessed on smears from both insemination medium (swim-up motile sperm) and semen, prepared by washing of sperm with 10 ml 0.9% sodium chloride. Morphology slides were stained with the Shorr method after xing in 90% ethanol for 30 min (Jeulin et al., 1986; Liu and Baker, 1992; World Health Organization, 1999). For each slide, 200 sperm were scored from >10 individual elds using oil immersion with magni cation of under bright- eld illumination. The percentage of normal sperm morphology was assessed according to strict criteria (Kruger et al., 1988; World Health Organization, 1999). Sperm preparation Motile sperm were selected by a swim-up technique. The sperm pellet obtained by centrifugation of semen was carefully added to the bottom of a test tube (12375 mm) containing 0.7 ml human tubal uid (HTF; Irvine Scienti c, USA) supplemented with 10% heat-inactivated human serum (ICN Biomedicals, USA). Care was taken to avoid disturbing the interface between the semen and the medium. After incubation for 1 h, 0.5 ml of the top layer of the medium containing motile sperm was aspirated. The motile sperm suspension was then centrifuged at 1000 g for 5 min, the supernatant removed and the sperm pellet washed again with 1 ml fresh HTF by centrifugation at 1000 g for 5 min. The washed sperm pellet was resuspended with serum-supplemented HTF to a sperm concentration of /ml for subsequent experiments. Sperm±ZP binding For each sperm±zp interaction test, motile sperm ( ) in 1 ml of medium were incubated with four oocytes for 2 h at 37 C in5%co 2 in air. After incubation, the oocytes were transferred to phosphatebuffered saline (PBS), ph 7.4, containing 2 mg/ml bovine serum albumin (BSA) and washed by repeated aspiration with a glass pipette (inside diameter ~250 mm) to dislodge sperm loosely adhering to the surface of the ZP. Because a high concentration of sperm was used in the insemination medium (20-fold higher than IVF insemination), the number of sperm from fertile men bound tightly to the ZP was >100/ oocyte (Liu and Baker, 1994; Liu et al., 2001). This experimental design allows enough ZP-bound sperm to be recovered for assessment of the AR. Assessment of the human ZP-induced AR All sperm bound to surface of the four ZP were then removed by repeated vigorous aspiration with a narrow gauge pipette with an inner diameter (~120 mm) slightly smaller than the oocyte (Liu and Baker, 1994, 1996a). This was performed on a glass slide with ~5 ml PBS containing 0.2% BSA and the removed ZP-bound sperm were smeared in a limited area (~16 mm 2 ), which was marked on the back of the slide with a glass pen to help nd the sperm under the microscope. This pipetting procedure for removing sperm from the surface of ZP does not affect sperm motility, morphology or acrosome status (Liu and Baker, 1996a). The acrosome status of sperm removed from the ZP was determined with uorescein-labelled Pisum Sativum agglutinin (PSA; Sigma Co., USA) using a modi cation of the method of Cross et al. (1986). Sperm smears were xed in 95% ethanol for 30 min after drying in air and then stained in 25 mg/ml PSA in PBS for >2 h at 4 C. The slide was washed and mounted with distilled water and the percentage of AR was determined by scoring 200 of the sperm removed from all four ZP 229

3 D.Y.Liu and H.W.G.Baker Table I. Results of sperm tests in oligozoospermic infertile men Test n Mean 6 SD Range Sperm count (310 6 /ml, semen) ±19 Total motility (%, semen) ±80 Progressive motility (%, semen) ±70 Live (%, semen) ±97 Normal sperm morphology (%, semen) ±36 Progressive motility (%, swim-up) ±99 Normal sperm morphology (%, swim-up) ±56 Normal intact acrosome (%, swim-up) ±98 No. of sperm bound/zp ±100 ZP-induced AR (%) a ±57 a Test only carried out in 52 men with a mean of >40 sperm bound/zp (20 had <40 sperm bound/zp). Figure 1. An example for Pisum sativum agglutinin± uorescein isothiocyanate staining of the acrosome of sperm removed from the surface of zona pellucida. There were ve acrosome intact (AI) and four acrosome-reacted (AR) sperm. The photograph was taken with 3400 magni cation. per test, using a uorescence microscope with oil immersion at a magni cation of When more than half the head of a sperm was brightly and uniformly uorescing, the acrosome was considered to be intact. Sperm with a uorescing band at the equatorial segment or no staining of the head (a rare pattern) were considered to be acrosomereacted (Figure 1). Thresholds for low sperm±zp binding and low ZP-induced AR A threshold of <40 sperm bound/zp was used to de ne low sperm±zp binding. This threshold for a sperm concentration of /ml in the insemination medium is equivalent to <2 sperm bound/zp with the standard IVF insemination sperm concentration of /ml (Liu and Baker, 2000). The ZP-induced AR <16% was used for classi cation of low ZP-induced AR according to our previous reports that men with <16% ZP-induced AR have very low (<30%) sperm±zp penetration and fertilization rates with standard IVF (Liu and Baker, 1996b, 2003b). Statistical analysis The signi cance of correlations between sperm±zp binding, ZP-induced AR and other sperm characteristics were examined by non-parametric Spearman's r and Spearman's correlation coef cient and its P value are reported. c 2 -Test was used for comparison of the proportions of men with low (<16%) ZP-induced AR and nonparametric Wilcoxon's rank sum test for comparison of mean results of ZP-induced AR between oligozoospermic, normozoospermic or teratozoospermic men from previous studies (Liu et al., 2001; Liu and Baker, 2003a). Figure 2. Comparison of the results of ZP-induced AR in 52 oligozoospermic (Oligo, mean ZP-induced AR 15%) infertile men with those for 186 normozoospermic infertile men (Normo, mean ZP-induced AR 33%) and 86 severely teratozoospermic infertile men with strict normal morphology <5% (Terato, mean ZP-induced AR 23%) as published previously (Liu et al., 2001; Liu and Baker, 2003a). (A) Plot of individual results for the three groups (Wilcoxon rank sum test, P < for Oligo versus Normo or Terato versus Normo, and P = for Oligo versus Terato). (B) Proportion of men with ZP-induced AR <16% (c 2 -test, P < 0.01 for Oligo or Terato versus Normo, and P = for Oligo versus Terato). Results Sperm tests results Table I shows mean and ranges for all sperm test and sperm± oocyte interaction test results. The average number of sperm bound was 70/ZP and 20 men had an average of <40 sperm bound/zp. The mean ZP-induced AR for the 52 men with normal sperm±zp binding was 15% and ranged from 1 to 57% (Figure 2). 230 Frequency of defective sperm±zp binding and the ZPinduced AR As described above, a threshold of <40 sperm bound/zp was used to de ne low sperm±zp binding and the ZP-induced AR <16% was used for classi cation of low ZP-induced AR. Of the 72 oligozoospermic men, 28% (20/72) had low sperm±zp binding. Of the 52 men with normal sperm±zp binding, 69%

4 Sperm±ZP interaction in oligozoospermic men (36/52) had low ZP-induced AR. This frequency of low ZPinduced AR with oligozoospermia is much higher than seen in previous studies of normozoospermic and severely teratozoospermic (strict normal morphology <5%) infertile men (Figure 2; Liu et al., 2001; Liu and Baker, 2003a). Overall, a total 78% (56/72) oligozoospermic men had either low sperm± ZP binding or normal sperm±zp binding with low ZP-induced AR. Correlation between sperm tests The percentage of sperm with progressive motility in semen was signi cantly correlated with number of sperm bound/zp (r = 0.274, P < 0.05). Sperm concentration and all other semen analysis results did not correlate with either sperm±zp binding or the ZP-induced AR. There was a signi cant correlation between sperm count and motility (r = 0.295, P < 0.05), and between normal sperm morphology and both total motility (r = 0.951, P < 0.01) and progressive motility (r = 0.973, P < 0.01). Eleven men had two ZP-induced AR tests performed on sperm from two separate ejaculates within 2±10 weeks, with similar results obtained (mean 6 SD %, range 5±30% versus %, range 4±25%, mean difference 6 SD %). Discussion The ZP-induced AR is highly correlated with sperm±zp penetration and fertilization rate in conventional IVF in patients with normal semen analysis (Liu and Baker, 1996b, 2003b; Esterhuizen et al., 2001; Bastiaan et al., 2003). Patients with unexplained infertility with ZP-induced AR <16% have average fertilization rates of <30% with conventional IVF (Liu and Baker, 2003b). Therefore, the ZP-induced AR is a useful clinical test for sperm fertilizing ability and for selection between conventional IVF and ICSI for treatment of patients. The present study shows that >28% of oligozoospermic men had low sperm±zp binding and 69% of the remainder with normal ZP binding had low ZP-induced AR. Overall, defective sperm±zp binding or ZP-induced AR was present in 78% of oligozoospermic men, indicating that a high frequency of defective sperm±zp interaction contributes to the infertility in oligozoospermia. Oligozoospermic men with defective sperm± ZP interaction should be treated by ICSI to avoid low or zero fertilization rates in conventional IVF. On the other hand, ~22% of oligozoospermic infertile men had normal sperm±zp binding and normal ZP-induced AR. These men may have better chances of producing fertilization either in vivo or with conventional IVF. The frequency of low ZP-induced AR (69%) in this group of oligozoospermic men who could be tested is much higher than in normozoospermic (25±29%) and severely teratozoospermic (48%) infertile men (Liu et al., 2001; Liu and Baker, 2003a,b). It is interesting that >64% of severely teratozoospermic men had low ZP-induced AR if they also had sperm counts between 20 and < /ml (Liu and Baker, 2003a). It is likely that these high frequencies of defective sperm±zp interaction are related to the degree of disordered spermatogenesis. It has been reported that abnormalities of sperm plasma membrane may reduce uidity and affect the AR in oligozoospermic men (Ladha, 1998). Further study is required to determine the cause for low ZP-induced AR in men with low sperm production. Others have reported that the ability of sperm to undergo the calcium ionophore A2387- induced AR may also predict sperm fertilizing ability in vitro (Cummins et al., 1991; Yovich et al., 1994). However, we found A23187-induced AR correlates with fertilization rate of IVF only in patients with teratozoopsermia (strict normal sperm morphology <15%; Liu and Baker 1998). Also A23187-induced AR does not correlate with the ZP-induced AR or sperm±zp penetration in normozoospermic men (Liu and Baker, 1996a,b). Therefore, the A induced AR may predict other sperm dysfunction but not defective ZP-induced AR. Similarly, the AR induced by progesterone or human follicular uid may also predict sperm fertilizing ability in men with unexplained infertility (Calvo et al., 1989; Krausz et al., 1995, 1996). However, these progesterone-induced AR tests usually involve long periods (18±20 h) of preincubation of sperm which are inconvenient for routine testing. Also sperm lose the ability to bind to the ZP after 18 h preincubation (Singer et al., 1985). We have been unable to study relationship between the ZP-induced AR and the AR induced by progesterone and human follicular uid since we have not been able to induce the AR with either agent. Human sperm binding to the ZP is highly species speci c and human sperm do not bind to the ZP of most other mammalian oocytes (Bedford, 1977; Liu et al., 1991). Human oocytes are essential for clinical testing of the sperm±zp interaction. In routine clinical IVF, ~20±30% of oocytes inseminated fail to fertilize and these unfertilized oocytes provide valuable biological material for testing sperm function for other infertile patients before they commence assisted reproduction treatment. Most oocytes that fail to fertilize and immature (germinal vesicle) oocytes unsuitable for ICSI are useful for the test. However, degenerate, spontaneous activated and morphologically abnormal oocytes and those with large numbers of sperm (>10) penetrating the ZP are not used. It is important to use a group of oocytes (optimally four) rather than a single oocyte for each test because of the variability in the quality of individual oocytes. Previous tests showed that ~10±20% of unfertilized oocytes obtained from clinical IVF do not bind sperm from fertile men (Liu et al., 1988). However, it is rare for all oocytes recovered from one woman to fail to bind sperm (Liu and Baker, 2000). Solubilized human ZP can also be used for induction of the AR and a microassay has been reported (Franken et al., 2000; Bastiaan et al., 2003). Nevertheless, access to human oocytes for routine tests of sperm oocyte interaction is still limited. At the moment, no chemical or biological material can substitute for native human ZP in this test. Recombinant human ZP still remains unavailable despite some claims of the production of biologically active material (van Duin et al., 1994; Brewis et al., 1996; Whitmarsh et al., 1996; Dong et al., 2001). This study indicates that the infertility of oligozoospermic men occurs not only because of the low sperm numbers but also because of the high frequency of defective sperm±zp inter- 231

5 D.Y.Liu and H.W.G.Baker action. Importantly, sperm from most (78%) of the oligozoospermic men have either low sperm±zp binding or low ZP-induced AR, both of which are likely to cause failure of sperm±zp penetration and fertilization. However, 22% have near normal sperm±oocyte interaction and may be more likely to conceive naturally or with conventional IVF. For clinical management of patients, sperm±zp interaction tests are useful for distinguishing between these two groups so that effective treatments can be chosen. Practically, if sperm±zp interaction tests are not available, patients with oligozoospermia should be treated with ICSI since ~78% of them have either defective sperm±zp binding or low ZP-induced AR. This will minimize the risk of fertilization failure with conventional IVF. The underlying biochemical and molecular causes of defective sperm±zp interaction in oligozoospermic men require further study. Acknowledgements The authors thank Mingli Liu for technical assistance, Dr Claire Garrett for preparation of the manuscript and all scientists in both IVF and Andrology Laboratories for their help collecting the oocytes and semen samples for this study. References Aitken RJ, Clarkson JS, Hargreave TB, Irvine DS and Wu FC (1989) Analysis of the relationship between defective sperm function and the generation of reactive oxygen species in cases of oligozoospermia. J Androl 10,214±220. Baker HWG (2001) Male Infertility. In DeGroot LJ and Jameson JL (eds), Endocrinology. 4th edn, WB Saunders Co, Philadelphia, PA, pp 2308±2328. Baker HWG, Liu DY, Bourne H and Lopata A (1993) Diagnosis of sperm defects in selecting patients for assisted fertilization. Hum Reprod 8,1779± Bastiaan HS, Windt ML, Menkveld R, Kruger TF, Oehninger S and Franken DR (2003) Relationship between zona pellucida-induced acrosome reaction, sperm morphology, sperm±zona pellucida binding, and in vitro fertilization. Fertil Steril 79,49±55. Bedford JM (1977) Sperm±egg interaction: the speci city of human spermatozoa. Anat Rec 188,477±483. Bostofte E, Serup J and Rebbe H (1982) Relation between sperm count and semen volume, and pregnancies obtained during a twenty-year follow-up period. Int J Androl 5,267±275. Brewis IA, Clayton R, Barratt CLR, Hornby DP and Moore HDM (1996) Recombinant human zona pellucida glycoprotein 3 induces calcium in ux and acrosome reaction in human spermatozoa. Mol Hum Reprod 2,583±589. Burger HG and Baker HWG (1984) Therapeutic considerations and results of gonadotropin treatment in male hypogonadotropic hypogonadism. Ann NY Acad Sci 438,447±453. Burkman LJ, Coddington CC, Franken DR, Krugen TF, Rosenwaks Z and Hogen GD (1988) The hemizona assay (HZA): development of a diagnostic test for the binding of human spermatozoa to the human hemizona pellucida to predict fertilization potential. Fertil Steril 49,688±697. Calvo L, Vantman D, Banks SM, Tezon J, Koukoulis GN, Dennison L and Sherins R (1989) Follicular uid-induced acrosome reaction distinguishes a subgroup of men with unexplained infertility not identi ed by semen analysis. Fertil Steril 52,1048±1054. Cross NL, Morales P, Overstreet JW and Hanson FW (1986) Two simple methods for detecting acrosome-reacted human sperm. Gamete Res 15,213± 226. Cummins JM, Pember SM, Jequier AM, Yovich JL and Hartmann PE (1991) A test of the human sperm acrosome reaction following ionophore challenge (ARIC). Relationship to fertility and other seminal parameters. J Androl 12,98±103. Dong KW, Chi TF, Juan YW, Chen CW, Lin Z, Xiang XQ, Mahony M, Gibbons WE and Oehninger S (2001) Characterization of the biologic activities of a recombinant human zona pellucida protein 3 expressed in 232 human ovarian teratocarcinoma (PA-1) cells. Am J Obstet Gynecol 184,835±843. Esterhuizen AD, Franken DR, Lourens JDH and van Rooyen LH (2001) Clinical importance of zona pellucida induced acrosome reaction (ZIAR): predictive value of the ZIAR test for in vitro fertilization. Hum Reprod 16,138±144. Franken DR, Bastiaan HS and Oehninger SC (2000) Physiological induction of the acrosome reaction in human sperm: validation of a microassay using minimal volumes of solubilized, homologous zona pellucida. J Assist Reprod Genet 17,156±161. Jeulin C, Feneux D, Serres C, Jouannet P, Guillet-Rosso F, Belaosch-Allart J, Frydman R and Testart J (1986) Sperm factors related to failure of human in vitro fertilisation. J Reprod Fertil 76,735±744. Krausz C, Bonaccorsi L, Luconi M, Fuzzi B, Criscuoli L, Pellegrini S, Forti G and Baldi E (1995) Intracellular calcium increase and acrosome reaction in response to progesterone in human spermatozoa are correlated with in-vitro fertilization. Hum Reprod 10,120±124. Krausz C, Bonaccorsi L, Maggio P, Luconi M, Criscuoli L, Fuzzi B, Pellegrini S, Forti G and Baldi E (1996) Two functional assays of sperm responsiveness to progesterone and their predictive values in in-vitro fertilization. Hum Reprod 11,1661±1667. Kruger TF, Acosta AA, Simmons KF, Swanson RJ, Matta JF and Oehninger S (1988) Predictive value of abnormal sperm morphology in in vitro fertilisation. Fertil Steril 49,112±117. Ladha S (1998) Lipid heterogeneity and membrane uidity in a highly polarized cell, the mammalian spermatozoon. J Membr Biol 165,1±10. Liu DY and Baker HWG (1992) Tests of human sperm function and fertilisation in vitro. Fertil Steril 58,465±483. Liu DY and Baker HWG (1994) Disordered acrosome reaction of sperm bound to the ZP (ZP): a newly discovered sperm defect causing infertility with reduced sperm±zp penetration and reduced fertilisation in vitro. Hum Reprod 9,1694±1700. Liu DY and Baker HWG (1996a) A simple method for assessment of the human acrosome reaction of spermatozoa bound to the zona pellucida: lack of relationship with ionophore A23187-induced acrosome reaction. Hum Reprod 11,551±557. Liu DY and Baker HWG (1996b) Relationship between the zona pellucida (ZP) and ionophore A23187 induced acrosome reaction and the ability of sperm to penetrate the ZP in men with normal sperm±zp binding. Fertil Steril 66,312±315. Liu DY and Baker HWG (1998) Calcium ionophore-induced acrosome reaction correlates with fertilization rates in vitro in patients with teratozoospermic semen. Hum Reprod 13,905±910. Liu DY and Baker HWG (2000) Defective sperm±zona pellucida interaction: a major cause of failure of fertilization in clinic in vitro fertilization. Hum Reprod 15,702±708. Liu DY and Baker HWG (2003a) Frequency of defective sperm±zona pellucida interaction in severely teratozoospermic infertile men. Hum Reprod 18,802±807. Liu DY and Baker HWG (2003b) Disordered zona pellucida induced acrosome reaction and failure of in vitro fertilization in patients with unexplained infertility. Fertil Steril 79,74±80. Liu DY, Lopata A, Johnston WIH and Baker HWG (1988) A human sperm± zona pellucida binding test using oocytes that failed to fertilize in vitro. Fertil Steril 50,782±788. Liu DY, Lopata A, Pantke P and Baker HWG (1991) Horse and marmoset sperm bind to the zona pellucida of salt stored human oocytes. Fertil Steril 56,764±767. Liu DY, Clarke GN, Martic M, Garrett C and Baker HWG (2001) Frequency of disordered zona pellucida (ZP)-induced acrosome reaction in infertile men with normal semen analysis and normal semen analysis and normal spermatozoa-zp binding. Hum Reprod 16,1185±1190. Lopata A and Leung PC (1988) The fertilizability of human oocytes at different stages of meiotic maturation. In Jones HW Jr and Schrader C (eds) In Vitro Fertilization and Other Assisted Reproduction. New York Academy of Sciences, New York, pp 324±336. Oehninger S, Coddington CC, Scott R, Franken DA, Burkman LJ, Acosta AA and Hodgen GD (1989) Hemizona assay: assessment of sperm dysfunction and prediction of in vitro fertilization outcome. Fertil Steril 51,665±670. Overstreet JW and Hembree WC (1976) Penetration of the zona pellucida of nonliving human oocytes by human spermatozoa in vitro. Fertil Steril 27,815±831. Overstreet JW, Yanagimachi R, Katz F and Hanson FW (1980) Penetration of

6 Sperm±ZP interaction in oligozoospermic men human spermatozoa into the human zona pellucida and zona-free hamster egg: a study of fertile and infertile patients. Fertil Steril 33,534±542. Singer SL, Lambert H, Overstreet JW, Hanson FW and Yanagimachi R (1985) The kinetics of human sperm binding to the human zona pellucida and zonafree hamster oocyte in vitro. Gamete Res 12,29±39. Sokol RZ and Sparkes R (1987) Demonstrated paternity in spite of severe idiopathic oligospermia. Fertil Steril 47,356±358. van Duin M, Polman JEM, de Breet ITM, Bunschoten H, van Ginneken K, Grootenhuis A, Brindle J and Aitken RJ (1994) Recombinant human ZP protein ZP3 produced in Chinese hamster ovary cell induces the human sperm acrosome reaction and promotes sperm±egg fusion. Biol Reprod 50,607±617. Wassarman PM (1999) Mammalian fertilization molecular aspects of gametes adhesion, exocytosis, and fusion. Cell 96,175±183. Whitmarsh AJ, Woolnough MJ, Moore HDM, Hornby DP and Barratt CLR (1996) Biological activity of recombinant human ZP3 produced in vitro: potential for a sperm function test. Mol Hum Reprod 2,911±919. World Health Organization (1999) WHO Laboratory Manual for Examination of human Semen and Semen±Cervical Mucus Interaction. Cambridge University Press, Cambridge, UK, pp 1±20. World Health Organization Task Force on Methods for the Regulation of Male Fertility (1990) Contraceptive ef cacy of testosterone-induced azoospermia in normal men. Lancet 336,955±959. Yanagimachi R (1994) Mammalian fertilisation. In Knobil E and Neill J (eds), The Physiology of Reproduction. 2nd edn, Raven Press, New York, pp 189± 317. Yovich JL, Stanger JD and Yovich JM (1985) The management of oligospermic infertility by in vitro fertilization. Ann N Y Acad Sci 442,276±286. Yovich JM, Edirisinghe WR and Yovich JL (1994) Use of the acrosome reaction to ionophore challenge test in managing patients in an assisted reproduction program: a prospective, double-blind, randomized controlled study. Fertil Steril 61,902±910. Submitted on May 13, 2003; resubmitted on August 14, 2003; accepted on October 18,