- Email: [email protected]
Defining the valid hemizona assay: accounting for binding variability within zonae pellucidae and within semen samples from fertile males*
FERTILITY AND STERILITY
Vol. 56, No.6, Decemher 1991
Printed on acid-free paper in U.S.A.
Copyright" 1991 The American Fertility Society
Defining the valid hemizona assay: accounting for binding variability within zonae pellucidae and within semen samples from fertile males* Daniel R. Franken, Ph.D.t Charles C. Coddington, M.D.* Lani J. Burkman, Ph.D.*§ W. T. Oosthuizen, B.Sc., Honnt
Sergio C. Oehninger, M.D.* Thinus F. Kruger, M.D.t Gary D. Hodgen, Ph.D.*"
Tygerberg Hospital, Infertility Clinic, Parow, Republic of South Afr.ica, and The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia
Objective: To achieve a better understanding of the variability in sperm and oocyte binding capacities will optimize use of the hemizona assay (HZA) as a predictor of sperm function. Design: Limitations of the HZA were more clearly delineated by current studies: (1) variability of sperm binding capacity of men over a 90-day interval; (2) variability of sperm binding using different oocytes; and (3) lower limits of the number of sperm bound from the fertile control in two laboratories. Patients: Semen was obtained from proven fertile men and one subfertile individual. Main Outcome Measure: The number of sperm tightly bound to the hemizona were measured and compared. Results: In the initial study, 6 fertile control men exhibited a similar degree of variability in zona binding when studied over a 90-day interval. Average sperm binding for individuals ranged from 68 to 127. Second, 3 of the 15 simultaneous assays showed very low numbers of sperm bound, indicating that 20% of the zonae had poor binding. Third, from 18 men who had 0% fertilization in an in vitro fertilization system using mature oocytes, evaluation of their sperm by HZA was performed. The sperm bound poorly and the 95% confidence interval was 20 sperm bound. Thus, the fertile controls should bind >20 sperm to distinguish them from the infertile group in the HZA system resulting in a valid assay. Conclusions: With these guidelines, applications ofthe HZA may be made with greater reassurance of a valid bioassay of sperm fertilizing potential. Fertil SteriI1991;56:1156-61
Received April 16, 1990; revised and accepted August 9, 1991. * Supported in part by the Contraceptive Research and Development Program (CONRAD), Eastern Virginia Medical School, under a Cooperative Agreement (DPE-2044-A-00-606300) with the United States Agency for International Development (A.I.D.) Rosslyn, Virginia, and by the South African Medical Research Council, Republic of South Africa. The views expressed by the author do not necessarily reflect the views of A.I.D. t Department of Obstetrics and Gynecology, Tygerberg Hospital, Infertility Clinic. :j: Department of Obstetrics and Gynecology; The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School. § Present address: GIBCO Laboratories, Life Technologies, Inc., Grand Island, New York. II Reprint requests: Gary D. Hodgen, Ph.D., The Jones Institute for Reproductive Medicine, Principal Investigator, Contraceptive Research and Development (CONRAD) Program, Department of Obstetrics and Gynecology Eastern Virginia Medical School, 855 West Brambleton Avenue, Suite B Norfolk, Virginia 23510.
1156
Franken et aI.
HZA: defining the limitations
The hemizona assay (HZA) has been introduced as a new bioassay for diagnostic evaluation of sperm performance. Clinical testing has demonstrated a positive correlation between the number of sperm bound to the hemizona and success during in vitro fertilization (IVF).1-3 We have shown that the HZA can predict a fertilization failure under standard IVF conditions (50,000 motile sperm/mL at Norfolk, VA) while showing a false-positive rate that is <5% of the tests performed. Whether oocytes are harvested by ovarian follicular aspiration after a stimulation regimen or obtained as immature tissue, they vary in diameter, zona thickness, and sperm binding potential.4-6 The HZA uses two matching hemispheres of the zona pellucida cut from one human oocyte, thus permitting an internally controlled assay of tight binding Fertility and Sterility
capacity.7,s Moreover, among proven fertile men, we have observed that sperm binding capacity varies both between and within individuals. l Optimal utilization of the HZA will depend on a better understanding of the variables that influence data interpretation and clinical correlations. In this report, we have examined the variability of sperm binding capacity within and between fertile controls using repeated HZAs through a 90-day interval. We have also studied the acceptable lower limit for the number of sperm bound to the hemizona, using control specimens (known fertile men), thereby defining a valid assay. Here, our studies progressed in three parts: (1) sperm binding variability within each man over 90 days; (2) variability because of the unequal sperm binding capacities of different oocytes; and (3) lower limits of the number of sperm bound from the fertile control in two laboratories. MATERIALS AND METHODS Oocyte Handling
All sperm binding experiments used immature (germinal vesicle stage) human oocytes collected from surgically excised ovarian tissue after patient consent was obtained. As previously described,5 the chilled tissue was first minced, then passed through a series of sieves, thereby isolating the zona-intact oocytes. As detailed elsewhere,2,5 the oocytes were stored in a concentrated salt solution consisting of 1.5 M MgCl 2 (Merck Chemical, Darmstad, Germany), 0.1% polyvinylpyrrolidone (Merck Chemical) and 40 mM Hepes buffer. The pH ofthe storage solution was first adjusted to 7.2 using 1 N NaOH. The oocytes were stored at +4°C, then transferred to Ham's F-lO medium (GIBCO, Grand Island, NY) or Biggers, Whitten, and Wittingham medium, both with 0.3% bovine serum albumin. Each nonliving oocyte was cut into equal hemispheres using a microblade and a micromanipulator. The details ofthe cutting procedure have been reported. l,5 Sperm Preparation
After liquefaction, a basic analysis of semen quality was performed. The seminal volume, sperm concentration (by hemacytometer), average grade of the sperm velocity (scale of 0 to 4),9 and the percentage of motile cells (100 sperm scored) were assessed by manual methods. After staining the air-dried samples with Diff-Quick (American Scientific Products, McGraw Park, IL), the proportion of sperm cells Vol. 56, No.6, December 1991
with normal morphology was evaluated according to strict criteria.lo,n An aliquot (0.5 mL) of each semen sample was washed twice with 0.5 mL of Ham's F-lO culture medium that had been supplemented with 7.5% human fetal cord serum. The final sperm pellet was overlaid with 0.5 to 1.0 mL of Ham's medium, then incubated for 1 hour (37°C, 5% CO 2 in air) to allow motile sperm to swim-up. The supernatant sperm were diluted with Ham's medium to 500,000 motile sperm/mL for use in the HZA. Study 1: Variability in HZA Among F~ttile Men When Tested Serially (4 Times Ove~;90 Days)
The motile spermatozoa from each of six fertile men were tested at 0, 30, 60, and 90 days. The men in this fertile control group had been selected because they had fathered children in the last 24 months. On each of the four test dates, a fresh semen specimen was obtained after 2 days of sexual abstinence. After the preparation and swim-up described above, each motile sperm sample was diluted to 500,000 motile sperm/mL with Ham's F-lO medium. Because this study was not intended for the comparison of fertile versus patient samples, the two halves of one zona pellucida provided a duplicate assay. Therefore, the two matching hemizonae from the same oocyte were transferred to a single 50-#LL droplet of the diluted sperm. Coincubation lasted for 4 hours (37°C, 5% CO 2 in air) after which each hemizona was picked up with a fine pipette and rinsed five times to remove the loosely attached sperm. The number of sperm tightly bound on the outer surface were then counted. l The number bound was expressed as the average for the two matching hemizonae. Study 2: Oocyte-To-Oocyte Variability in Sperm Binding Capacity Among 15 Zonae
The data from study 1 suggested that a factor other than sperm variability contributed to the noticeable spread in the number of bound spermatozoa. Therefore, oocyte-to-oocyte differences in sperm binding potential were examined. Fifteen HZAs were performed simultaneously, using the same sperm suspension from one proven fertile man (control) and from a single subfertile man presenting with teratoasthenozoospermia (test subject). The 15 oocytes had been removed from the same ovary and were used for this experiment. During three previous IVF attempts, sperm from the subfertile man fertilized only 3 of 11 mature, metaphase II oocytes, using an insemination concentraFranken et aI.
HZA: defining the limitations
1157
-----------------------------------------------
tion of 100,000 motile spermjmL after swim-up preparation. Fifteen small Petri dishes were prepared alike, each containing one 50-~L droplet of the diluted control spermatozoa and a second 50-~L droplet containing sperm from the sub fertile man. The droplets were covered by mineral oil. In each dish, the droplet containing fertile sperm received one hemizona, whereas the matching hemizona was transferred to the droplet containing spermatozoa from the subfertile man. Using a temporally staggered protocol, initiation of coincubation differed by only 30 minutes between set-up of the 1st and 15th dish. All dishes were incubated at 37°C (5% CO 2 in air) for 4.0 hours. 3 At the appropriate time, each hemizona was removed, rinsed five times in medium using a small-bore pipette to dislodge the loosely bound sperm. The number of tightly bound spermatozoa were then counted. The sperm binding data were pooled for the control versus test group (15 replicate HZAs per group), and the means ± SD were calculated. Study 3: Lower Limit Number of Sperm Bound for a Functional Assay
A collaborative effort between Bloemfontein and Norfolk encompassed HZAs from 18 couples who had 0% fertilization during an IVF cycle. These couples were selected because of their lack of fertilization of mature oocytes, suggesting the problem resided in the sperm. The women underwent a normal IVF stimulation with Lupron (TAP Pharmaceuticals, Deerfield, IL) and gonadotropins to maximize oocyte yield. 12,13 The husband's sperm were prepared in standard fashion as described previously.14 When failure of fertilization occurred, the HZA was performed on semen specimen after 48 hours abstinence. Semen parameters were calculated for count, motility, and morphology. The HZA was performed using a known fertile (control) individual and comparing his performance with that of the infertile individual (test). The number of sperm bound to the test portion of each hemizona were carefully counted. Once the data were available for the 18 couples studied, a 95% confidence interval (CI) was determined. RESULTS Study 1
Table 1 presents a summary of the zona-binding data for each of the six fertile men. When tested over 90 days, using four different oocytes, each of the men exhibited overlapping ranges of sperm:zona 1158
Franken et al.
HZA: defining the limitations
Table 1 Serial HZA Data From Six Control Men, With Semen Specimens Tested Four Times Over 90 Days
Control men (fertile) 1 2 3 4 5 6
Binding range (low to high) 28 to 25 to 90 to 41 to 63 to 50 to
129 92 195 200 144 166
No. bound sperm (all assays) 69.5 ± 68.1 ± 127.0 ± 101.0 ± 87.0 ± 87.0 ±
17 8 16 30 19 20
(8)" (8)" (8)" (7)" (7)" (7)"
"Values are means ± SEM with number of hemizonae in pa· rentheses.
binding. The binding ranges shown in Table 1 indicate a binding differential (high value minus the low value) of 101,67, 105, 159,81, and 116, respectively, for the six men. Therefore, based on four different oocytes (8 hemizonae), all of the proven fertile men showed a widely varying number of bound sperm. The mean number of sperm bound was high for each of the men, varying between 68.1 and 127 (Table 1). Over the 90 days, there was no trend toward an increasing or decreasing number of bound sperm for any of the men; rather, the variability seemed attributable to some factor other than changing sperm function. Except for two instances, the number of sperm bound to the two halves of a hemizona pair were similar. The value obtained at each time point was computed as an average of two hemizonae and did not differ from the difference induced by cutting the zonae. 7 However, when the difference in the number of sperm bound between the two halves was expressed as the absolute value, the percentage difference between the two paired numbers varied between 0% and 29%. The average deviation between the two halves was only 13.9% ± 1.9% (mean and SE; n = 7 pairs for 3 donors and n = 8 for each of the 3 other donors). Study 2
Sperm binding to the hemizonae in experiments 2, 8, and 15 was much lower than in the other 12 experiments (Table 2). For those three eggs, binding in the fertile droplets was limited to 8, 15, and 24 sperm, compared with 66 to 218 for the remaining 12 experiments. Likewise, the subfertile binding data were lower for these three oocytes as expected. For these 15 hemizonae incubated simultaneously with the sperm from the fertile sample, the average binding was 103 sperm after 4 hours (range: 8 to Fertility and Sterility
Table 2 Measuring the Variability in Sperm Binding Capacity Between Different Zonae Pellucidae in the HZA Number of sperm tightly bound to the matching hemizonae
Oocyte no. 1 2 3 4 5 6 7 8 9 10 11
12 13 14 15 Mean ± SD
Control sperm droplet (n = 15)
Subfertile sperm droplet (n = 12)
132 15 200 106 99 101 218 24 145 101 66 128 82 125 8 103.3 ± 58.4 b
61 6 79 72 63 54 18 42 31 42 58 9 44.6 ± 23.2 b•c
Data not obtained. Simultaneous HZAs were performed using 15 hemizona pairs allocated in tandem with the sperm from one fertile man and one infertile man (study 2). c Mean is significantly lower than the control mean, P < 0.05.
cyte function, it was important to investigate the lower number of sperm bound to the hermizona that would distinguish the sperm function of fertile and infertile men. Eighteen men, who had no fertilization of mature oocytes, were studied via the HZA after 48 hours' abstinence. The concentration of their sperm was 64.6 X 106 ± 43 sperm/mL (mean ± SD), percent motility was 28.7% ± 17% and morphology by the strict criteria was 2.9 ± 2.6. The control sperm bound in a range of 10 up to 90 compared with the infertile group, which was 0 up to 30. Once the number of sperm were counted for each hemizona, a 95% CI was determined. The value of which 95% of the values fell below was 20 sperm for the infertile group. Thus, when the hemizona bound >20 sperm, the binding capability of the oocyte was confirmed, and any decreased binding to the paired hemizona was because of poor sperm function. DISCUSSION
a
b
218). Likewise, the 15 matching hemizonae had been coincubated with sperm from the one subfertile specimen; here, the mean number of bound sperm was significantly lower (44.6 ± 23.2; P < 0.05) than that seen for the fertile sample, with a range of 6 to 79 bound sperm. This comparatively poor performance was consistent with the partial fertilization failure seen previously with this patient and the borderline semen characteristics for the sample used in study 2 (sperm concentration of 30 X 106/mL, motility of 40%, and 6% normal forms). The data from this study would indicate that, by a conservative estimate, approximately 20% (3/15) of the zonae pellucida from germinal vesicle stage oocytes have demonstrably impaired sperm binding potential. Consequently, the lower 20% ofthe values given in Table 2 for control sperm are erroneous, resulting from an inferior zona pellucida. The decreased binding in these two cases cannot be attributed to poor sperm capacity. When interpreting the zona-binding data from a set of HZAs using spermatozoa from proven fertile men, study 2 shows that as many as 20% of the assay may need to be repeated because poor binding does not allow the distinction to be made between infertile and poor zona pellucida binding. Study 3
Having demonstrated that as many as 20% ofthe HZA might need to be repeated because of poor 00Vol. 56, No.6, December 1991
During fertility testing, the variability in egg quality as well as donor semen quality frequently has confounded in vitro assay interpretation. Using a homologous, rodent fertilization model l5.16 in which eggs are easily obtained, inter-egg variation is the principal reason for placing multiple ova (5 to 20) in each coincubation droplet. Similarly, the zonafree hamster egg sperm penetration assay used 10 to 20 ova per dish. 17 Cross et al. 15 have calculated that oocyte-to-oocyte differences represent one of the two most significant sources of experimental variability in a zona-binding study. Variation in the diameter of the immature human oocyte (140 to 210 JLm),7 in the appearance of the zona surface,6 and in zona thickness may be linked to differences in zona function, including tight sperm binding. IS We have frequently observed that the hemizonae from a particular oocyte will bind surprisingly few sperm from a proven fertile semen preparation. Because of this source of assay error, we have specifically addressed the issue of defining a "bad zona" and the invalid experiment that may need to be repeated. In addition to the inconsistency of egg quality, the sperm from individual semen donors exhibit different capacitative or egg-interactive potential. The time course and peak levels of hamster egg penetrations,19 as well as the time course for hyperactivated motility,20 show donor-dependent variation. In our laboratory, we likewise have observed differences within proven fertile men concerning the time of maximal sperm binding to the hemizona pellucida (3.5 to 5.5 hours) and the peak number of sperm Franken et al.
HZA: defining the limitations
1159
bound at that time (42 to 215,1 and 27 to 1423 ). Across the 90 days of evaluation in study 1, sperm from the fertile men did not exhibit a time-dependent change in zona-binding potential. Judging by their individual means for sperm binding (67 to 127), all of the men demonstrated good binding capacities; however, there was a range of about 100 between their low and high counts when comparing different hemizonae. This fact suggested that some factor other than semen quality was the major contributor to binding variability within a fertile control. In study 1, the use oftwo matching hemizonae in the same sperm droplet (duplicate assay) provided a new demonstration of binding equivalency for the matching zona halves. The current results (14% deviation within a hemizona pair) are consistent with the original publication. 7 The simultaneous assays in study 2, using the very same sperm suspension throughout, demonstrated great variability that was clearly because of zona quality alone. The staggered protocol here should have contributed little to variations in the number bound because the duration of all hemizonae coincubations was practically 4 hours. Three hemizonae pairs (3 eggs) had low binding potential and were labeled as invalid; however, the one value of 24 in study 2 is acceptable using guidelines of study 3. This would suggest that only 13.3% of the assays might need to be repeated (2/15 eggs = 13.3% with poor zonae). Future studies should examine whether approximately 20% of metaphase I or metaphase II oocytes demonstrate deficient binding function. U sing the individuals who had failed fertilization during in vitro procedures, samples of truly infertile individuals could be tested in the HZA. Stimulation of the female patients was performed according to previously published guidelines. 12 The grading of the morphology of the oocytes during the in vitro attempt were all excellent; thus, the quality of the 00cytes was maximizedY Although the sperm concentration was in the normal range of this group, the percent motility and morphology were reduced. These facts did not affect the ability to perform the standard assay because 500,000 motile sperm/mL were able to be obtained, and the sperm for each patient survived the 4-hour incubation with the HZ. Once the number of bound sperm were determined, the 95% CI was established at 20. This threshold of 20 bound sperm provides a practical means of identifying a poor HZA. At Norfolk, based on the current threshold value of 20 bound sperm, any HZA that yields <20 for the control hemizona is invalid and must be repeated. A binding value of <20 indicates that either the zona pellucida used was inferior in 1160
Franken et al.
HZA: defining the limitations
its ability to bind sperm or the control semen was subnormal that day. On occasion, an assay may show that the control hemizona bound <20 sperm, but the matching (test) hemizona bound >20 patients' spermatozoa. In this particular instance, it is clear that the zona pellucida is adequate, but the control sperm were inferior in that instance. Again, one should repeat this assay. Additional data from Norfolk, which included 61 hemizona tests, appear to strengthen the cutoff at 20. There were 61 assays from Norfolk (n = 5 donors) for which there was an average of 51.1 ± 31 (mean ± SEM) sperm tightly bound from these fertile men. Individual hemizona varied between 7 and 155 bound sperm. The five men differed in their individual averages, one having a mean of 20 and another a mean of 66 bound sperm. Based on the findings from study 2, the lower 20% of the Norfolk values was noted. There were 20 sperm bound to the hemizona at the point of the cutoff for 20% bad assays because of poor egg binding. For this group of fertile control men, the threshold value for a valid hemizona assay was ~20 sperm tightly bound to the hemizona control droplet. Thus, at a value of 20 bound sperm, one should have minimized the impact of a poorly binding hemizona and also be at the upper range of the sperm that would function poorly like sperm that had failed IVF. So during performance of an HZA, if >20 sperm bound to the control half of the zona, the data obtained should give a valid representation of sperm function. The current findings provide guidelines for the appropriate interpretation of the hemizona assay. Within the limits stated in this paper, the HZA is able to control for the intrinsic variability in zona pellucida binding capacity because of its use of the matching halves of a single egg. In contrast to assays using an intact zona pellucida and fluorescent sperm labels,21,22 the two hemizona halves can specifically accommodate study designs involving two different incubation protocols or zona pretreatments. For these latter purposes, the HZA remains uniquely appropriate.
Acknowledgement. We gratefully acknowledge the expert technical assistance of Deborah Johnson, M.A., and Ms. Dara WillettLeary for her editorial contribution and manuscript preparation. Both are from the Jones Institute of Reproductive Medicine, Norfolk, Virginia.
REFERENCES 1. Franken DR, Oehninger SC, Burkman LJ, Coddington CC,
Kruger TF, Rosenwaks Z, Acosta AA, Hodgen GD: The
Fertility and Sterility
2.
3.
4.
5.
6.
7.
8.
9.
10.
hemizona assay (HZA): a predictor of human sperm fertilizing potential in in vitro fertilization (IVF) treatment. J In Vitro Fert Embryo Transfer 6:44, 1989 Oehninger SC, Coddington CC, Scott R, Franken DR, Burkman LJ, Acosta AA, Hodgen GD: Hemizona assay: assessment of sperm dysfunction and prediction of in vitro fertilization outcome. Fertil Steril51:665, 1989 Burkman LJ, Coddington CC, Franken DR, Oehninger SC, Hodgen GD: Assessment of fertilizing potential by means of human sperm binding to the human zona pellucida. In Handbook of the Laboratory Diagnosis and Treatment of Infertility, Edited by BA Keel, BW Webster. Boca Raton, CRC Press, Inc., 1990, p 213 Mahadevan MD, Trounson AO, Wood C, Leeton JF: Effect of oocyte quality and sperm characteristics on the number of spermatozoa bound to the zona pellucida of human oocytes inseminated in vitro. J In Vitro Fert Embryo Transfer 4:223, 1987 Overstreet JW, Hembree WC: Penetration of the zona pellucida of nonliving human oocytes by human spermatozoa in vitro. Fertil Steril 27:815, 1976 Familiari G, Nottola SA, Micara G, Aragona C, Motta PM: Is the sperm binding capability of the zona pellucida linked to its surface structure: a scanning electron microscopic study of human in vitro fertilization. J In Vitro Fert Embryo Transfer 3:134, 1988 Burkman LJ, Coddington CC, Franken DR, Kruger TF, Rosenwaks Z, Hodgen GD: 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, 1988 Franken DR, Burkman LJ, Oehninger SC, Coddington CC, Veeck L, Kruger TF, Rosenwaks Z, Hodgen GD: Hemizona assay using salt-stored human oocytes: evaluation of zona pellucida capacity for binding human spermatozoa. Gamete Res 22:15, 1989 World Health Organization: Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction, Edited by MA Belsey, R Eliasson, AJ Gallegos, KS Moghissi, CA Paulson, MRN Prasad. Singapore, Press Concern, 1980, p 10 Kruger TF, Menkveld R, Stander FSH, Lombard CJ, Van der Merwe JP, van Zyl JA, Smith K: Sperm morphologic
Vol. 56, No.6, December 1991
11.
12.
13.
14.
15.
16. 17.
18.
19. 20.
features as a prognostic factor in in vitro fertilization. Fertil Steril46:1118, 1986 Kruger TF, Acosta AA, Simmons KF, Swanson RJ, Matta JF, Oehninger S: Predictive value of abnormal sperm morphology in in vitro fertilization. Fertil Steril 49:112, 1988 Brzyski RF, Jones GS, Oehninger S, Acosta AA, Kruithoff C, Muasher S: Impact of leuprolide acetate on follicular response in in vitro fertilization patients with normal basal gonadotropin levels. J In Vitro Fert Embryo Transfer 6:290, 1989 Veeck LL: Human oocytes at the time of follicular harvest. In Atlas of the Human Oocytes and Early Conceptus. Baltimore, Williams and Wilkins, 1986, p 6 McDowell JS: Preparation of spermatozoa for insemination in vitro. In In Vitro Fertilization-Norfolk, Edited by HW Jones, Jr, GS Jones, GD Hodgen. Baltimore, Williams and Wilkins, 1986, p 162 Cross NL, Lambert H, Samuels S: Sperm binding activity of the zona pellucida of immature mouse oocytes. Cell Bioi Rep 10:545, 1986 Bavister DB: A consistently successful procedure for in vitro fertilization of golden hamster eggs. Gamete Res 23:139, 1989 Yanagimachi R, Yanagimachi H, Rogers BJ: The use of zonafree animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. Bioi Reprod 15: 471, 1986 Chen C, Sathananthan AH: Early penetration of human sperm through the vestments of human eggs in vitro. Arch Androl 16:183, 1986 Perreault S, Rogers BJ: Capacitation pattern of human spermatozoa. Fertil Steril 38:258, 1982 Burkman LJ: Experimental approaches to the evaluation and enhancement of sperm function. In In Vitro FertilizationNorfolk, Edited by HW Jones, Jr, GS Jones, GD Hodgen. Baltimore, Williams and Wilkins, 1986, p 201
21. Blazak WF, Overstreet JW, Katz DF, Hanson FW: A competitive in vitro assay of human sperm fertilizing ability utilizing contrasting fluorescent sperm markers. J AndroI3:165, 1982 22. Liu DY, Clarke GN, Lopata A, Johnston WIH, Baker HWG: A sperm-zona pellucida binding test and in vitro fertilization. Fertil Steril 52:281, 1989
Franken et al.
HZA: defining the limitations
1161
Vol. 56, No.6, Decemher 1991
Printed on acid-free paper in U.S.A.
Copyright" 1991 The American Fertility Society
Defining the valid hemizona assay: accounting for binding variability within zonae pellucidae and within semen samples from fertile males* Daniel R. Franken, Ph.D.t Charles C. Coddington, M.D.* Lani J. Burkman, Ph.D.*§ W. T. Oosthuizen, B.Sc., Honnt
Sergio C. Oehninger, M.D.* Thinus F. Kruger, M.D.t Gary D. Hodgen, Ph.D.*"
Tygerberg Hospital, Infertility Clinic, Parow, Republic of South Afr.ica, and The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia
Objective: To achieve a better understanding of the variability in sperm and oocyte binding capacities will optimize use of the hemizona assay (HZA) as a predictor of sperm function. Design: Limitations of the HZA were more clearly delineated by current studies: (1) variability of sperm binding capacity of men over a 90-day interval; (2) variability of sperm binding using different oocytes; and (3) lower limits of the number of sperm bound from the fertile control in two laboratories. Patients: Semen was obtained from proven fertile men and one subfertile individual. Main Outcome Measure: The number of sperm tightly bound to the hemizona were measured and compared. Results: In the initial study, 6 fertile control men exhibited a similar degree of variability in zona binding when studied over a 90-day interval. Average sperm binding for individuals ranged from 68 to 127. Second, 3 of the 15 simultaneous assays showed very low numbers of sperm bound, indicating that 20% of the zonae had poor binding. Third, from 18 men who had 0% fertilization in an in vitro fertilization system using mature oocytes, evaluation of their sperm by HZA was performed. The sperm bound poorly and the 95% confidence interval was 20 sperm bound. Thus, the fertile controls should bind >20 sperm to distinguish them from the infertile group in the HZA system resulting in a valid assay. Conclusions: With these guidelines, applications ofthe HZA may be made with greater reassurance of a valid bioassay of sperm fertilizing potential. Fertil SteriI1991;56:1156-61
Received April 16, 1990; revised and accepted August 9, 1991. * Supported in part by the Contraceptive Research and Development Program (CONRAD), Eastern Virginia Medical School, under a Cooperative Agreement (DPE-2044-A-00-606300) with the United States Agency for International Development (A.I.D.) Rosslyn, Virginia, and by the South African Medical Research Council, Republic of South Africa. The views expressed by the author do not necessarily reflect the views of A.I.D. t Department of Obstetrics and Gynecology, Tygerberg Hospital, Infertility Clinic. :j: Department of Obstetrics and Gynecology; The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School. § Present address: GIBCO Laboratories, Life Technologies, Inc., Grand Island, New York. II Reprint requests: Gary D. Hodgen, Ph.D., The Jones Institute for Reproductive Medicine, Principal Investigator, Contraceptive Research and Development (CONRAD) Program, Department of Obstetrics and Gynecology Eastern Virginia Medical School, 855 West Brambleton Avenue, Suite B Norfolk, Virginia 23510.
1156
Franken et aI.
HZA: defining the limitations
The hemizona assay (HZA) has been introduced as a new bioassay for diagnostic evaluation of sperm performance. Clinical testing has demonstrated a positive correlation between the number of sperm bound to the hemizona and success during in vitro fertilization (IVF).1-3 We have shown that the HZA can predict a fertilization failure under standard IVF conditions (50,000 motile sperm/mL at Norfolk, VA) while showing a false-positive rate that is <5% of the tests performed. Whether oocytes are harvested by ovarian follicular aspiration after a stimulation regimen or obtained as immature tissue, they vary in diameter, zona thickness, and sperm binding potential.4-6 The HZA uses two matching hemispheres of the zona pellucida cut from one human oocyte, thus permitting an internally controlled assay of tight binding Fertility and Sterility
capacity.7,s Moreover, among proven fertile men, we have observed that sperm binding capacity varies both between and within individuals. l Optimal utilization of the HZA will depend on a better understanding of the variables that influence data interpretation and clinical correlations. In this report, we have examined the variability of sperm binding capacity within and between fertile controls using repeated HZAs through a 90-day interval. We have also studied the acceptable lower limit for the number of sperm bound to the hemizona, using control specimens (known fertile men), thereby defining a valid assay. Here, our studies progressed in three parts: (1) sperm binding variability within each man over 90 days; (2) variability because of the unequal sperm binding capacities of different oocytes; and (3) lower limits of the number of sperm bound from the fertile control in two laboratories. MATERIALS AND METHODS Oocyte Handling
All sperm binding experiments used immature (germinal vesicle stage) human oocytes collected from surgically excised ovarian tissue after patient consent was obtained. As previously described,5 the chilled tissue was first minced, then passed through a series of sieves, thereby isolating the zona-intact oocytes. As detailed elsewhere,2,5 the oocytes were stored in a concentrated salt solution consisting of 1.5 M MgCl 2 (Merck Chemical, Darmstad, Germany), 0.1% polyvinylpyrrolidone (Merck Chemical) and 40 mM Hepes buffer. The pH ofthe storage solution was first adjusted to 7.2 using 1 N NaOH. The oocytes were stored at +4°C, then transferred to Ham's F-lO medium (GIBCO, Grand Island, NY) or Biggers, Whitten, and Wittingham medium, both with 0.3% bovine serum albumin. Each nonliving oocyte was cut into equal hemispheres using a microblade and a micromanipulator. The details ofthe cutting procedure have been reported. l,5 Sperm Preparation
After liquefaction, a basic analysis of semen quality was performed. The seminal volume, sperm concentration (by hemacytometer), average grade of the sperm velocity (scale of 0 to 4),9 and the percentage of motile cells (100 sperm scored) were assessed by manual methods. After staining the air-dried samples with Diff-Quick (American Scientific Products, McGraw Park, IL), the proportion of sperm cells Vol. 56, No.6, December 1991
with normal morphology was evaluated according to strict criteria.lo,n An aliquot (0.5 mL) of each semen sample was washed twice with 0.5 mL of Ham's F-lO culture medium that had been supplemented with 7.5% human fetal cord serum. The final sperm pellet was overlaid with 0.5 to 1.0 mL of Ham's medium, then incubated for 1 hour (37°C, 5% CO 2 in air) to allow motile sperm to swim-up. The supernatant sperm were diluted with Ham's medium to 500,000 motile sperm/mL for use in the HZA. Study 1: Variability in HZA Among F~ttile Men When Tested Serially (4 Times Ove~;90 Days)
The motile spermatozoa from each of six fertile men were tested at 0, 30, 60, and 90 days. The men in this fertile control group had been selected because they had fathered children in the last 24 months. On each of the four test dates, a fresh semen specimen was obtained after 2 days of sexual abstinence. After the preparation and swim-up described above, each motile sperm sample was diluted to 500,000 motile sperm/mL with Ham's F-lO medium. Because this study was not intended for the comparison of fertile versus patient samples, the two halves of one zona pellucida provided a duplicate assay. Therefore, the two matching hemizonae from the same oocyte were transferred to a single 50-#LL droplet of the diluted sperm. Coincubation lasted for 4 hours (37°C, 5% CO 2 in air) after which each hemizona was picked up with a fine pipette and rinsed five times to remove the loosely attached sperm. The number of sperm tightly bound on the outer surface were then counted. l The number bound was expressed as the average for the two matching hemizonae. Study 2: Oocyte-To-Oocyte Variability in Sperm Binding Capacity Among 15 Zonae
The data from study 1 suggested that a factor other than sperm variability contributed to the noticeable spread in the number of bound spermatozoa. Therefore, oocyte-to-oocyte differences in sperm binding potential were examined. Fifteen HZAs were performed simultaneously, using the same sperm suspension from one proven fertile man (control) and from a single subfertile man presenting with teratoasthenozoospermia (test subject). The 15 oocytes had been removed from the same ovary and were used for this experiment. During three previous IVF attempts, sperm from the subfertile man fertilized only 3 of 11 mature, metaphase II oocytes, using an insemination concentraFranken et aI.
HZA: defining the limitations
1157
-----------------------------------------------
tion of 100,000 motile spermjmL after swim-up preparation. Fifteen small Petri dishes were prepared alike, each containing one 50-~L droplet of the diluted control spermatozoa and a second 50-~L droplet containing sperm from the sub fertile man. The droplets were covered by mineral oil. In each dish, the droplet containing fertile sperm received one hemizona, whereas the matching hemizona was transferred to the droplet containing spermatozoa from the subfertile man. Using a temporally staggered protocol, initiation of coincubation differed by only 30 minutes between set-up of the 1st and 15th dish. All dishes were incubated at 37°C (5% CO 2 in air) for 4.0 hours. 3 At the appropriate time, each hemizona was removed, rinsed five times in medium using a small-bore pipette to dislodge the loosely bound sperm. The number of tightly bound spermatozoa were then counted. The sperm binding data were pooled for the control versus test group (15 replicate HZAs per group), and the means ± SD were calculated. Study 3: Lower Limit Number of Sperm Bound for a Functional Assay
A collaborative effort between Bloemfontein and Norfolk encompassed HZAs from 18 couples who had 0% fertilization during an IVF cycle. These couples were selected because of their lack of fertilization of mature oocytes, suggesting the problem resided in the sperm. The women underwent a normal IVF stimulation with Lupron (TAP Pharmaceuticals, Deerfield, IL) and gonadotropins to maximize oocyte yield. 12,13 The husband's sperm were prepared in standard fashion as described previously.14 When failure of fertilization occurred, the HZA was performed on semen specimen after 48 hours abstinence. Semen parameters were calculated for count, motility, and morphology. The HZA was performed using a known fertile (control) individual and comparing his performance with that of the infertile individual (test). The number of sperm bound to the test portion of each hemizona were carefully counted. Once the data were available for the 18 couples studied, a 95% confidence interval (CI) was determined. RESULTS Study 1
Table 1 presents a summary of the zona-binding data for each of the six fertile men. When tested over 90 days, using four different oocytes, each of the men exhibited overlapping ranges of sperm:zona 1158
Franken et al.
HZA: defining the limitations
Table 1 Serial HZA Data From Six Control Men, With Semen Specimens Tested Four Times Over 90 Days
Control men (fertile) 1 2 3 4 5 6
Binding range (low to high) 28 to 25 to 90 to 41 to 63 to 50 to
129 92 195 200 144 166
No. bound sperm (all assays) 69.5 ± 68.1 ± 127.0 ± 101.0 ± 87.0 ± 87.0 ±
17 8 16 30 19 20
(8)" (8)" (8)" (7)" (7)" (7)"
"Values are means ± SEM with number of hemizonae in pa· rentheses.
binding. The binding ranges shown in Table 1 indicate a binding differential (high value minus the low value) of 101,67, 105, 159,81, and 116, respectively, for the six men. Therefore, based on four different oocytes (8 hemizonae), all of the proven fertile men showed a widely varying number of bound sperm. The mean number of sperm bound was high for each of the men, varying between 68.1 and 127 (Table 1). Over the 90 days, there was no trend toward an increasing or decreasing number of bound sperm for any of the men; rather, the variability seemed attributable to some factor other than changing sperm function. Except for two instances, the number of sperm bound to the two halves of a hemizona pair were similar. The value obtained at each time point was computed as an average of two hemizonae and did not differ from the difference induced by cutting the zonae. 7 However, when the difference in the number of sperm bound between the two halves was expressed as the absolute value, the percentage difference between the two paired numbers varied between 0% and 29%. The average deviation between the two halves was only 13.9% ± 1.9% (mean and SE; n = 7 pairs for 3 donors and n = 8 for each of the 3 other donors). Study 2
Sperm binding to the hemizonae in experiments 2, 8, and 15 was much lower than in the other 12 experiments (Table 2). For those three eggs, binding in the fertile droplets was limited to 8, 15, and 24 sperm, compared with 66 to 218 for the remaining 12 experiments. Likewise, the subfertile binding data were lower for these three oocytes as expected. For these 15 hemizonae incubated simultaneously with the sperm from the fertile sample, the average binding was 103 sperm after 4 hours (range: 8 to Fertility and Sterility
Table 2 Measuring the Variability in Sperm Binding Capacity Between Different Zonae Pellucidae in the HZA Number of sperm tightly bound to the matching hemizonae
Oocyte no. 1 2 3 4 5 6 7 8 9 10 11
12 13 14 15 Mean ± SD
Control sperm droplet (n = 15)
Subfertile sperm droplet (n = 12)
132 15 200 106 99 101 218 24 145 101 66 128 82 125 8 103.3 ± 58.4 b
61 6 79 72 63 54 18 42 31 42 58 9 44.6 ± 23.2 b•c
Data not obtained. Simultaneous HZAs were performed using 15 hemizona pairs allocated in tandem with the sperm from one fertile man and one infertile man (study 2). c Mean is significantly lower than the control mean, P < 0.05.
cyte function, it was important to investigate the lower number of sperm bound to the hermizona that would distinguish the sperm function of fertile and infertile men. Eighteen men, who had no fertilization of mature oocytes, were studied via the HZA after 48 hours' abstinence. The concentration of their sperm was 64.6 X 106 ± 43 sperm/mL (mean ± SD), percent motility was 28.7% ± 17% and morphology by the strict criteria was 2.9 ± 2.6. The control sperm bound in a range of 10 up to 90 compared with the infertile group, which was 0 up to 30. Once the number of sperm were counted for each hemizona, a 95% CI was determined. The value of which 95% of the values fell below was 20 sperm for the infertile group. Thus, when the hemizona bound >20 sperm, the binding capability of the oocyte was confirmed, and any decreased binding to the paired hemizona was because of poor sperm function. DISCUSSION
a
b
218). Likewise, the 15 matching hemizonae had been coincubated with sperm from the one subfertile specimen; here, the mean number of bound sperm was significantly lower (44.6 ± 23.2; P < 0.05) than that seen for the fertile sample, with a range of 6 to 79 bound sperm. This comparatively poor performance was consistent with the partial fertilization failure seen previously with this patient and the borderline semen characteristics for the sample used in study 2 (sperm concentration of 30 X 106/mL, motility of 40%, and 6% normal forms). The data from this study would indicate that, by a conservative estimate, approximately 20% (3/15) of the zonae pellucida from germinal vesicle stage oocytes have demonstrably impaired sperm binding potential. Consequently, the lower 20% ofthe values given in Table 2 for control sperm are erroneous, resulting from an inferior zona pellucida. The decreased binding in these two cases cannot be attributed to poor sperm capacity. When interpreting the zona-binding data from a set of HZAs using spermatozoa from proven fertile men, study 2 shows that as many as 20% of the assay may need to be repeated because poor binding does not allow the distinction to be made between infertile and poor zona pellucida binding. Study 3
Having demonstrated that as many as 20% ofthe HZA might need to be repeated because of poor 00Vol. 56, No.6, December 1991
During fertility testing, the variability in egg quality as well as donor semen quality frequently has confounded in vitro assay interpretation. Using a homologous, rodent fertilization model l5.16 in which eggs are easily obtained, inter-egg variation is the principal reason for placing multiple ova (5 to 20) in each coincubation droplet. Similarly, the zonafree hamster egg sperm penetration assay used 10 to 20 ova per dish. 17 Cross et al. 15 have calculated that oocyte-to-oocyte differences represent one of the two most significant sources of experimental variability in a zona-binding study. Variation in the diameter of the immature human oocyte (140 to 210 JLm),7 in the appearance of the zona surface,6 and in zona thickness may be linked to differences in zona function, including tight sperm binding. IS We have frequently observed that the hemizonae from a particular oocyte will bind surprisingly few sperm from a proven fertile semen preparation. Because of this source of assay error, we have specifically addressed the issue of defining a "bad zona" and the invalid experiment that may need to be repeated. In addition to the inconsistency of egg quality, the sperm from individual semen donors exhibit different capacitative or egg-interactive potential. The time course and peak levels of hamster egg penetrations,19 as well as the time course for hyperactivated motility,20 show donor-dependent variation. In our laboratory, we likewise have observed differences within proven fertile men concerning the time of maximal sperm binding to the hemizona pellucida (3.5 to 5.5 hours) and the peak number of sperm Franken et al.
HZA: defining the limitations
1159
bound at that time (42 to 215,1 and 27 to 1423 ). Across the 90 days of evaluation in study 1, sperm from the fertile men did not exhibit a time-dependent change in zona-binding potential. Judging by their individual means for sperm binding (67 to 127), all of the men demonstrated good binding capacities; however, there was a range of about 100 between their low and high counts when comparing different hemizonae. This fact suggested that some factor other than semen quality was the major contributor to binding variability within a fertile control. In study 1, the use oftwo matching hemizonae in the same sperm droplet (duplicate assay) provided a new demonstration of binding equivalency for the matching zona halves. The current results (14% deviation within a hemizona pair) are consistent with the original publication. 7 The simultaneous assays in study 2, using the very same sperm suspension throughout, demonstrated great variability that was clearly because of zona quality alone. The staggered protocol here should have contributed little to variations in the number bound because the duration of all hemizonae coincubations was practically 4 hours. Three hemizonae pairs (3 eggs) had low binding potential and were labeled as invalid; however, the one value of 24 in study 2 is acceptable using guidelines of study 3. This would suggest that only 13.3% of the assays might need to be repeated (2/15 eggs = 13.3% with poor zonae). Future studies should examine whether approximately 20% of metaphase I or metaphase II oocytes demonstrate deficient binding function. U sing the individuals who had failed fertilization during in vitro procedures, samples of truly infertile individuals could be tested in the HZA. Stimulation of the female patients was performed according to previously published guidelines. 12 The grading of the morphology of the oocytes during the in vitro attempt were all excellent; thus, the quality of the 00cytes was maximizedY Although the sperm concentration was in the normal range of this group, the percent motility and morphology were reduced. These facts did not affect the ability to perform the standard assay because 500,000 motile sperm/mL were able to be obtained, and the sperm for each patient survived the 4-hour incubation with the HZ. Once the number of bound sperm were determined, the 95% CI was established at 20. This threshold of 20 bound sperm provides a practical means of identifying a poor HZA. At Norfolk, based on the current threshold value of 20 bound sperm, any HZA that yields <20 for the control hemizona is invalid and must be repeated. A binding value of <20 indicates that either the zona pellucida used was inferior in 1160
Franken et al.
HZA: defining the limitations
its ability to bind sperm or the control semen was subnormal that day. On occasion, an assay may show that the control hemizona bound <20 sperm, but the matching (test) hemizona bound >20 patients' spermatozoa. In this particular instance, it is clear that the zona pellucida is adequate, but the control sperm were inferior in that instance. Again, one should repeat this assay. Additional data from Norfolk, which included 61 hemizona tests, appear to strengthen the cutoff at 20. There were 61 assays from Norfolk (n = 5 donors) for which there was an average of 51.1 ± 31 (mean ± SEM) sperm tightly bound from these fertile men. Individual hemizona varied between 7 and 155 bound sperm. The five men differed in their individual averages, one having a mean of 20 and another a mean of 66 bound sperm. Based on the findings from study 2, the lower 20% of the Norfolk values was noted. There were 20 sperm bound to the hemizona at the point of the cutoff for 20% bad assays because of poor egg binding. For this group of fertile control men, the threshold value for a valid hemizona assay was ~20 sperm tightly bound to the hemizona control droplet. Thus, at a value of 20 bound sperm, one should have minimized the impact of a poorly binding hemizona and also be at the upper range of the sperm that would function poorly like sperm that had failed IVF. So during performance of an HZA, if >20 sperm bound to the control half of the zona, the data obtained should give a valid representation of sperm function. The current findings provide guidelines for the appropriate interpretation of the hemizona assay. Within the limits stated in this paper, the HZA is able to control for the intrinsic variability in zona pellucida binding capacity because of its use of the matching halves of a single egg. In contrast to assays using an intact zona pellucida and fluorescent sperm labels,21,22 the two hemizona halves can specifically accommodate study designs involving two different incubation protocols or zona pretreatments. For these latter purposes, the HZA remains uniquely appropriate.
Acknowledgement. We gratefully acknowledge the expert technical assistance of Deborah Johnson, M.A., and Ms. Dara WillettLeary for her editorial contribution and manuscript preparation. Both are from the Jones Institute of Reproductive Medicine, Norfolk, Virginia.
REFERENCES 1. Franken DR, Oehninger SC, Burkman LJ, Coddington CC,
Kruger TF, Rosenwaks Z, Acosta AA, Hodgen GD: The
Fertility and Sterility
2.
3.
4.
5.
6.
7.
8.
9.
10.
hemizona assay (HZA): a predictor of human sperm fertilizing potential in in vitro fertilization (IVF) treatment. J In Vitro Fert Embryo Transfer 6:44, 1989 Oehninger SC, Coddington CC, Scott R, Franken DR, Burkman LJ, Acosta AA, Hodgen GD: Hemizona assay: assessment of sperm dysfunction and prediction of in vitro fertilization outcome. Fertil Steril51:665, 1989 Burkman LJ, Coddington CC, Franken DR, Oehninger SC, Hodgen GD: Assessment of fertilizing potential by means of human sperm binding to the human zona pellucida. In Handbook of the Laboratory Diagnosis and Treatment of Infertility, Edited by BA Keel, BW Webster. Boca Raton, CRC Press, Inc., 1990, p 213 Mahadevan MD, Trounson AO, Wood C, Leeton JF: Effect of oocyte quality and sperm characteristics on the number of spermatozoa bound to the zona pellucida of human oocytes inseminated in vitro. J In Vitro Fert Embryo Transfer 4:223, 1987 Overstreet JW, Hembree WC: Penetration of the zona pellucida of nonliving human oocytes by human spermatozoa in vitro. Fertil Steril 27:815, 1976 Familiari G, Nottola SA, Micara G, Aragona C, Motta PM: Is the sperm binding capability of the zona pellucida linked to its surface structure: a scanning electron microscopic study of human in vitro fertilization. J In Vitro Fert Embryo Transfer 3:134, 1988 Burkman LJ, Coddington CC, Franken DR, Kruger TF, Rosenwaks Z, Hodgen GD: 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, 1988 Franken DR, Burkman LJ, Oehninger SC, Coddington CC, Veeck L, Kruger TF, Rosenwaks Z, Hodgen GD: Hemizona assay using salt-stored human oocytes: evaluation of zona pellucida capacity for binding human spermatozoa. Gamete Res 22:15, 1989 World Health Organization: Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction, Edited by MA Belsey, R Eliasson, AJ Gallegos, KS Moghissi, CA Paulson, MRN Prasad. Singapore, Press Concern, 1980, p 10 Kruger TF, Menkveld R, Stander FSH, Lombard CJ, Van der Merwe JP, van Zyl JA, Smith K: Sperm morphologic
Vol. 56, No.6, December 1991
11.
12.
13.
14.
15.
16. 17.
18.
19. 20.
features as a prognostic factor in in vitro fertilization. Fertil Steril46:1118, 1986 Kruger TF, Acosta AA, Simmons KF, Swanson RJ, Matta JF, Oehninger S: Predictive value of abnormal sperm morphology in in vitro fertilization. Fertil Steril 49:112, 1988 Brzyski RF, Jones GS, Oehninger S, Acosta AA, Kruithoff C, Muasher S: Impact of leuprolide acetate on follicular response in in vitro fertilization patients with normal basal gonadotropin levels. J In Vitro Fert Embryo Transfer 6:290, 1989 Veeck LL: Human oocytes at the time of follicular harvest. In Atlas of the Human Oocytes and Early Conceptus. Baltimore, Williams and Wilkins, 1986, p 6 McDowell JS: Preparation of spermatozoa for insemination in vitro. In In Vitro Fertilization-Norfolk, Edited by HW Jones, Jr, GS Jones, GD Hodgen. Baltimore, Williams and Wilkins, 1986, p 162 Cross NL, Lambert H, Samuels S: Sperm binding activity of the zona pellucida of immature mouse oocytes. Cell Bioi Rep 10:545, 1986 Bavister DB: A consistently successful procedure for in vitro fertilization of golden hamster eggs. Gamete Res 23:139, 1989 Yanagimachi R, Yanagimachi H, Rogers BJ: The use of zonafree animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. Bioi Reprod 15: 471, 1986 Chen C, Sathananthan AH: Early penetration of human sperm through the vestments of human eggs in vitro. Arch Androl 16:183, 1986 Perreault S, Rogers BJ: Capacitation pattern of human spermatozoa. Fertil Steril 38:258, 1982 Burkman LJ: Experimental approaches to the evaluation and enhancement of sperm function. In In Vitro FertilizationNorfolk, Edited by HW Jones, Jr, GS Jones, GD Hodgen. Baltimore, Williams and Wilkins, 1986, p 201
21. Blazak WF, Overstreet JW, Katz DF, Hanson FW: A competitive in vitro assay of human sperm fertilizing ability utilizing contrasting fluorescent sperm markers. J AndroI3:165, 1982 22. Liu DY, Clarke GN, Lopata A, Johnston WIH, Baker HWG: A sperm-zona pellucida binding test and in vitro fertilization. Fertil Steril 52:281, 1989
Franken et al.
HZA: defining the limitations
1161