Sperm nuclear chromatin normality: relationship with sperm morphology, sperm-zona pellucida binding, and fertilization rates in vitro*

Transcription

1 FERTILITY AND STERILITY Copyright Q 1992 The American Fertility Society Vol. 58, No.6, December 1992 Printed on acid-free paper in U.S.A. Sperm nuclear chromatin normality: relationship with sperm morphology, sperm-zona pellucida binding, and fertilization rates in vitro* De Yi Liu, Ph.D.t H. W. Gordon Baker, M.D., Ph.D. University of Melbourne Department of Obstetrics and Gynaecology and Reproductive Biology Unit, Royal Women's Hospital, Melbourne, Victoria, Australia Objective: To study whether the results of tests of sperm chromatin and deoxyribonucleic acid (DNA) normality are related to fertilization rates in vitro. Design: Normal morphology, nuclear maturity determined by acidic aniline blue stain, and DNA normality determined by acridine orange fluorescence of sperm in insemination medium and the number of sperm bound to the zona pellucida (ZP) of the oocytes that had failed to fertilize in vitro were determined. The relationship between sperm test results and fertilization rates were analyzed by logistic regression. Setting: Samples were obtained from patients undergoing in vitro fertilization (IVF) treatment. Results: The number of sperm bound to the ZP, the percentage of sperm with normal morphology, and the percentage of sperm with normal DNA were the most significant factors related to fertilization rates in vitro. In patients with normal morphology z 15% or with >10 sperm bound per ZP, the percentage of sperm with normal DNA, the number of sperm bound to the ZP, and motility grade were significantly related to IVF rates. Conclusion: In patients with normal morphology z 15%, failure of fertilization may be because of defects of sperm-zp binding or abnormal DNA. Assessment of DNA normality of motile sperm in the insemination medium may aid prediction of fertilization rates in addition to normal morphology and sperm-zp binding. Fertil Steril1992;58: Key Words: Sperm morphology, nuclear chromatin, zona binding, in vitro fertilization Use of the results of in vitro fertilization (IVF) to evaluate clinical tests of human sperm function is a powerful approach because sperm used for insemination of oocytes can be tested and the results of sperm tests directly related to IVF rates. Logistic regression analysis is useful to determine which groups of sperm characteristics are most independently significantly related to fertilization rates (1). Received May 15, 1992; revised and accepted August 24, * Presented at the 10th Annual Scientific Meeting of Fertility Society of Australia, Lome, Victoria, Australia, November 18 to 22, t Reprint requests: De Yi Liu, Ph.D., Department of Obstetrics and Gynaecology, University of Melbourne, Royal Women's Hospital, Carlton, Victoria 3053, Australia. A number of such studies, including our reports, have shown that the percentage of normal morphology assessed with an improved method is one of the most significant sperm factors related to fertilization rates (1-3). However, the accuracy of prediction of IVF rates using sperm morphology alone is currently still poor. Therefore, other new sperm function tests such as sperm-zona pellucida (ZP) binding and assessment of sperm with normal intact acrosomes have been developed and used in combination with normal morphology to improve prediction of IVF rates (4-6). Using these sperm function tests, the majority of patients with failure of fertilization are found to have defects of sperm morphology, acrosomes, and ZP binding. However, the causes of failure of IVF for the other 25% of patients is unclear (Liu DY, 1178 Liu and Baker Sperm nuclear chromatin normality and IVF Fertility and Sterility

2 Baker HWG, unpublished observations). Therefore, it is necessary to study other defects of sperm or oocytes as causes of poor IVF results, particularly in those patients without severe defects of sperm morphology and sperm-zp binding. Jeulin et al. (7) reported that assessment of sperm nuclear maturity with acidic aniline blue stain could provide additional information for predicting fertility in vitro. However, our previous study in 106 IVF patients showed the percentage of sperm with immature sperm stained with acidic aniline blue was not significantly related to IVF rates (1). Others reported that assessment of sperm deoxyribonucleic acid (DNA) normality by staining with acridine orange (AO) may help in the prediction of fertility (8, 9). Although these tests are useful for studying some aspects of human sperm physiology, the relationship between the results and fertility has not been studied in detail. In the present study, we have determined if the percentages of sperm with immaturity of nuclear chromatin determined with acidic aniline blue and abnormality of DNA determined with AO fluorescence of sperm in the insemination medium are related to failure of IVF, particularly in those patients without severe defects of morphology and sperm -ZP binding. Patients MATERIALS AND METHODS Investigations were performed in 91 IVF patients. In these couples, the wife's oocytes (~3) were inseminated with husband's sperm, and some or all of the oocytes failed to fertilize. All the patients (n = 42) with zero fertilization from January to December 1991 were included in the present study. Other patients (n = 49) were selected because they had IVF on the same day as those with zero fertilization, and a proportion of their oocytes failed to fertilize. The clinical and laboratory aspects of the IVF procedures are described elsewhere (1). The swim-up technique was used for selection of motile sperm for insemination of the oocytes. Diagnoses were male infertility in 22, unexplained infertility in 28, tubal occlusion in 16, endometriosis in 6, and mixed male and female (tubal occlusion or/and endometriosis) factors in 19. Sperm Samples Sperm in the insemination medium were obtained after embryos or unfertilized oocytes had been removed at 48 to 60 hours after insemination. The insemination medium from all the culture dishes was pooled for the same patient. The sperm were washed with 10 ml of 0.9% NaCI (sterile intravenous saline), and the sperm pellet was resuspended in 0.1 ml of 0.9% NaCl. Washed sperm were smeared on a glass slide, and three separate slides were prepared for staining of morphology, nuclear maturity, and DNA normality. Sperm in semen and swim-up preparation were not examined in this study because many of the patients had poor semen, and all the sample was used for sperm preparation for in semi - nation of oocytes. Assessment of Sperm Concentration, Motility, and Morphology Sperm concentration in the insemination suspension was determined using a hemocytometer. Sperm motility and motility grade in the insemination suspension were determined using phasecontrast microscope according to the methods of World Health Organization (WHO) (10). The percentage motility was calculated by counting 100 sperm. Four grades were used for classifying sperm motility as follows: 1 = immotile; 2 = nonprogressive motility; 3 = progressive but not rapid linear movement; 4 = rapid and linear progression. Morphology smears were prepared after adjusting the sperm concentration to approximately 80 X 10 6 /ml. The slides were stained with the Shorr method after the smear was fixed in 90% ethanol for 30 minutes (1, 7). Normal sperm morphology was assessed following WHO's criteria for the silhouette plus internal staining characteristics with the acrosomal region being clearly seen, regular in shape, and occupying at least half of the sperm head (1, 10, 11). The percentage of sperm with normal morphology was determined by assessing 200 sperm from more than 10 individual fields under oil immersion with magnification of Xl,OOO and brightfield illumination. Assessment of Sperm Nuclear Maturity and DNA Normality Maturity of nuclear chromatin was examined with acidic aniline blue (BDH Chemical, Poole, England) stain method described elsewhere (12). Briefly, after air drying, the sperm smear was fixed for 30 minutes in 3% glutaraldehyde in phosphate-buffered saline. The smear was stained for 15 minutes in 5% aqueous aniline blue in 4% acetic acid (ph 3.5). Sperm heads with immature nuclear chromatin stain dark blue Vol. 58, No.6, December 1992 Liu and Baker Sperm nuclear chromatin normality and IVF 1179

3 with acidic aniline blue and those with a mature nucleus do not stain (10). The percentage of sperm stained with aniline blue was determined by counting 200 sperm under oil immersion with X1,000 magnification and bright-field illumination. The normality of DNA was assessed by the AO fluorescence method described by Tejada et al. (8). After air drying, the sperm smear was fixed in Carnoy's solution (3 parts of methanol and 1 part glacial acetic acid) for at least 3 hours or overnight. The Carnoy's solution was prepared daily. The slides were then washed in distilled water and allowed to air dry for a few minutes before staining. The AO (Sigma Chemical Co., St Louis, MO) staining solution was prepared as follows: 10 ml of 1 % AO in distilled water was added to a mixture of 40 ml of 0.1 M citric acid and 2.5 ml of 0.3 M Na2HP047H20, ph 2.5. The AO staining solution was prepared daily. The 1 % AO stock solution was stored in the dark at 4 C for 4 weeks. Sperm smear was stained for 5 minutes. Then the slide was gently rinsed and mounted with distilled water. The percentage of sperm with normal DNA was determined by counting 200 sperm under a fluorescence microscope (Dialux 20; Leitz, Wetzlar, Germany) with X400 (oil lens) magnification with excitation of 450 to 490 nm. Sperm with normal (double-stranded) DNA fluoresce green, and those with denatured or singlestranded DNA fluoresce red or yellow. Assessment of Number of Sperm Bound to the ZP The number of sperm bound to the ZP of all the unfertilized oocytes were counted with an inverted phase-contrast microscope with magnification of X250 after removing the cumulus or corona cells if present (13). The oocytes were pipetted gently several times to dislodge any sperm not tightly bound to the zona. When there were >100 sperm bound to one ZP, it is difficult to count the number of sperm accurately. In this situation the number was recorded as 100. To determine whether the ZP selectively bound sperm with a mature nucleus and normal DNA, 00- cytes with >20 sperm bound to the ZP from the same patient were pooled together and smeared on a glass slide. These slides were stained to assess nuclear maturity and DNA normality of the sperm bound to the ZP. Statistical Analysis The significance of differences between mean results of sperm tests for patients with zero (IVF rate = 0) or some fertilization (IVF rates> 0) was performed by t-test. Paired t-tests were used for differences between the percentages of sperm with normal DNA and mature nuclei bound to the ZP and in the insemination medium. Weighted mean binding rates were also calculated (14). Correlations between sperm test results and between sperm test results and IVF rates were performed by nonparametric (Spearman) test. Relationships between sperm test results and IVF rates were examined by logistic regression analysis using the statistical package SPIDA (Macquarie University, Sydney, Australia). The proportion of the deviance explained by each model was calculated from the difference between the deviance of the fitted model and the deviance of a null model containing only the constant. The resulting r values with logistic regression are lower than with linear regression because of the binary response. RESULTS Sperm Test and IVF Results The simple comparison of mean sperm test results between patients with no oocytes and some oocytes fertilized is shown in Table 1. There was a wide range for all the sperm test results and also the number of oocytes inseminated. There were significant differences in the percentages of sperm with mature nuclei, normal DNA, normal morphology, and the number of sperm bound to the ZP between patients Table 1 In Vitro Fertilization and Sperm Test Results for Patients With 0 and >0 Fertilization Rates Fertilization rates 0 >0 No. of subjects No. of oocytes inseminated 10 (3 to 22) 11 (3 to 27) No. of oocytes fertilized 0 6 (1 to 19) Fertilization rates (%) 0 50 (4 to 89) No. of sperm inseminated (10 3 /ml) 220 (30 to 900) 144 (60 to 500)' Motility (insemination medium, %) to 100) 78 (5 to 100) No. of sperm bound/zp 9 (0 to 81) 30 (1 to 100)t No. of patients with ~1O sperm bound/zp Normal morphology (insemination medium, %) 10 (0 to 39) 25 (3 to 53)t Normal DNA (insemination medium, %) 40 (1 to 91) 56 (6 to 95)' Mature nucleus (insemination medium, %) 74 0 to 98) 82 (15 to 97)' 'P < t P < Liu and Baker Sperm nuclear chromatin normality and IVF Fertility and Sterility

4 with IVF rates = 0 and IVF rates > O. There was an average of 10% of sperm with normal morphology in patients with zero fertilization compared with 25% in those with some fertilization. Similarly, patients with zero fertilization had on average 9 sperm bound per ZP compared with 30 in those with some oocytes fertilized. There was no significant difference in the number of oocytes inseminated between the two groups. Because it is current practice to increase the number of sperm in the insemination medium in patients with high proportions of sperm with abnormal morphology or previous failure of fertilization, there was a significantly higher number of sperm inseminated in patients with IVF rates = 0 than in those with IVF rates> O. Patients with diagnoses of male infertility, including those with mixed male and female factors infertility (23 of 41,56%), had a significantly higher frequency of zero fertilization (x 2 = 4.8, P < 0.05) than did patients with only female infertility (6 of 22, 27%). Of patients with unexplained infertility, 47% (13 of 28) had zero fertilization: not significantly different from male factor patients. Correlations Between Sperm Test Results There were significant correlations between most of the sperm test results, such as between motility, motility grade, and morphology; between morphology, motility, and motility grade and number of sperm bound to the ZP; and between morphology, normal DNA, and mature nuclei (Table 2). Overall, sperm morphology was most strongly correlated with the number of sperm bound to the ZP and other characteristics such as motility and mature nuclei and less significantly correlated with normal DNA. Correlation Between Sperm Tests Results and IVF Rates When all the data from 91 patients were analyzed by Spearman's correlation test, there were highly significant correlations between percentage of sperm with normal morphology (r = 0.607, P < 0.001), the number of sperm bound to the ZP (r = 0.530, P < 0.001), and the percentage of sperm with normal DNA (r = 0.302, P < 0.001), and IVF rates. Sperm motility (r = 0.296, P < 0.05), motility grade (r = 0.274, P < 0.05), and mature nuclei (r = 0.252, P < 0.05) were less significant. Logistic Regression Analysis To determine which groups of sperm characteristics were independently related to IVF rates, all of the data were examined by logistic regression analysis. Only number of sperm bound to the ZP, the percentages of sperm with normal morphology and normal DNA, and motility grade were strongly significantly related to IVF rates. All the other variables were not significant (Table 3). Because there was a significantly lower IVF rate (mean ± SD, 10 ± 22 versus 42 ± 31; P < 0.001) for patients with normal morphology < 15% (n = 39) than for patients with normal morphology ~ 15% (n = 52), the data from patients with normal morphology ~ 15% were selected and analyzed by logistic regression. In these, only number of sperm bound to the ZP, the percentages of sperm with normal DNA and mature nucleus, and motility grade were significantly related to IVF rates. Normal morphology and other test results were not significant (Table 3). Because patients with no oocytes fertilized usually have on average <10 sperm bound per ZP, the data from 43 patients with ~ 10 sperm bound per ZP were selected to determine which other sperm characteristics were related to fertilization in vitro. Only motility grade, percentage of sperm with normal DNA, and number of sperm bound to the ZP were related to IVF rates (Table 3). Table 2 Spearman Correlation Coefficients Between Test Results of Sperm in Insemination Medium Motility Motility grade Normal morphology No. of sperm bound/zp Normal DNA Motility grade Normal morphology (%) No. of sperm bound/zp Normal DNA (%) Mature nuclei (%) % * t t 0.397* 0.332t % 0.464* * % 0.339* * P < t P < P < Vol. 58, No.6, December 1992 Liu and Baker Sperm nuclear chromatin normality and IVF ll81

5 Table 3 Significant Variables Related to IVF Rates in Logistic Regression Analysis Regression Variables coefficient SE* zt Probability All the patients studied (n = 91),2:\: = Normal morphology (%) <0.001 No. of sperm bound/zp <0.001 Normal DNA (%) <0.001 Motility grade <0.001 Patients with normal morphology ;" 15% (n = 52),2 = No. of sperm bound/zp <0.001 Normal DNA (%) <0.001 Motility grade <0.01 Mature nucleus (%) <0.05 Patients with;" 10 sperm bound/zp (n = 43),2 = Motility grade <0.001 Normal DNA (%) <0.005 No. of sperm bound/zp <0.01 * SE, standard error. t z, normal deviate (regression coefficient/se). :\:,2, proportion of deviance explained by the model. Nuclear Maturity and DNA Normality of Sperm Bound to the ZP There were significantly higher proportions of sperm with mature nuclei (18 subjects; mean and range, 93 and 71 to 100 on the ZP versus 83 and 69 to 94 in insemination medium, P < 0.01) and normal DNA (19 subjects; 87 and 35 to 100 on the ZP versus 57 and 7 to 95 in the insemination medium; P < 0.001) bound to the ZP than these in the insemination medium (Fig. 1). However, some sperm with abnormal DNA or an immature nucleus were still able to bind to the ZP, but the mean rates of binding of these sperm were lower than for sperm with normal DNA or mature nuclei (normal DNA = 28.2, abnormal DNA = 8.4; mature nucleus = 21.1, immature nucleus = 12.0/ZP /10 5 per ml inseminated). DISCUSSION The present study shows that the percentage of sperm in the insemination medium with normal morphology and normal DNA, the number of sperm bound per ZP, and motility grade are the most significant sperm factors independently related to IVF rates. This result further confirms our previous reports that percentage of normal morphology and the ability of sperm to bind to the ZP are the most important sperm characteristics related to IVF (1, 4, 5, 13). Therefore, defects of sperm morphology and sperm-zp binding seem to be the most common cause offailure offertilization in vitro (11). However, the use of sperm morphology and sperm-zp binding results alone for predicting fertility is still inaccurate, particularly in the patients without severe defects of morphology and sperm-zp binding. It is possible that other sperm defects or oocyte abnormalities may be involved in failure of fertilization in some of these patients. In the present study, we found that sperm DNA normality assessed with AO stain was a significant factor related to fertilization rates in patients with either normal morphology :2: 15% in insemination medium or in patients with :2: 10 bound per ZP. This result suggests that patients without severe defects of sperm morphology and sperm-zp binding may have failure of fertilization because of abnormal nuclear chromatin. Therefore, assessment of sperm chromatin abnormalities with these simple staining methods could improve the ability to predict fertilization in vitro. In the present study, the results of most sperm tests were correlated with each other; particularly the percentage of normal sperm morphology was most significantly correlated with the number of sperm bound to the ZP, which further confirm our previous report (13). We have also shown that the human ZP is highly selective for binding of sperm with normal morphology (14). Although some abnormal sperm do bind to the ZP, the binding rate of morphologically normal sperm is much higher than for any abnormal sperm (14). The binding rate of morphological normal sperm is also strongly related to fertilization rates in vitro. This may explain n = 19 n = 18 ~ ol Insem ZP Normal DNA Insem ZP Mature Nuclei Figure 1 Comparison of the percentages of sperm with normal DNA and mature nuclei bound to the ZP and in the insemination medium Liu and Baker Sperm nuclear chromatin normality and IVF Fertility and Sterility

6 why normal morphology is found by us and others to be a most important sperm characteristic for fertilization in vitro (1, 3, 7, 10). Although the percentage of sperm with mature nuclei is strongly correlated with normal morphology, normal DNA was weakly correlated with normal morphology. Thus, some morphologically normal sperm may have defects of DNA. Hofmann and Hilscher (15) reported that 23% to 78% of morphologically normal sperm had immature nuclei stained with acidic aniline blue in infertile patients in whom treatment by either artificial insemination with husband's sperm or IVF was unsuccessful. It is known that selection of motile sperm by swim-up or Percoll gradient centrifugation can improve the quality of the sperm recovered with respect to morphology, acrosome integrity, motility, and velocity (16). Le Lannou and Blanchard (17) reported that the proportion of sperm with mature or normal nuclear chromatin assessed by both acidic aniline blue and AO staining methods was significantly increased by selection of motile sperm by swim-up or Percoll gradient centrifugation. Thus motile sperm generally have better morphology and nuclear maturity. Katz et al. (18) reported that morphologically normal sperm swim faster, straighter, and with higher tail beat frequencies than do morphologically abnormal sperm in semen. Furthermore, the present study shows that percentages of sperm with mature nuclei and normal DNA were significantly higher for sperm bound to the ZP than for sperm in the insemination medium. This suggests that the human ZP selectively binds sperm with a mature nuclei and normal DNA. In the present study, nearly half the subjects (42 of 91) had zero fertilization, and various sperm defects may be included in this large group of patients with failure of fertilization. However, sperm morphology is still one of the most important characteristics related to fertilization rates in vitro. Even if the number of sperm bound to the ZP is strongly correlated with IVF rates in this or our previous study, normal sperm morphology is the most significant correlate of the number of sperm bound to the ZP (1, 4, 13). Acrosome status of sperm in the insemination medium was not determined in this study because the sperm had been incubated for 48 to 60 hours. However, we previously showed that the proportion of sperm with a normal intact acrosome in insemination medium was significantly related to fertilization in vitro in patients with poor morphology (4, 19). Other evidence suggests that acrosome-reacted sperm have little or no ability to bind to the ZP (20). In patients with an average of ~10 sperm bound per ZP, the percentage of sperm with normal DNA and motility grade were significantly related to IVF rates. This suggests that in patients with good sperm-zp binding, failure of fertilization may be because of failure of sperm penetration of the ZP because of poor motility or subsequent defects of fertilization because of the abnormal chromatin. The significant factors in the logistic regression model for all 91 patients: percentage of normal morphology, average number of sperm bound per ZP, percentage of normal DNA, and motility grade, explained 38% of the deviance. Logistic regression models for patients with normal morphology ~ 15% or greater than 10 sperm bind per ZP while omitting the percentage of normal morphology still included normal DNA as a significant factor and explained 25% and 19% of the deviance, respectively. Thus although the individual factors are highly significant in the statistical sense, there is a large amount of the variability in the fertilization rates unaccounted for. This may result from chance or undiscovered factors. This study further emphasizes that multiple sperm factors are often involved in failure of fertilization. Therefore groups of sperm function tests are required to predict fertility accurately. Logistic regression analysis is useful to determine which several sperm function tests are independently related to fertilization rates. Study of the causes of failure of fertilization in vitro also provides valuable information for understanding and improving treatment of male infertility. In summary, defects of sperm morphology and the ability of sperm to bind to the ZP are the most common causes for failure of fertilization in vitro. Assessment of maturity of nuclear chromatin and normality of DNA could improve prediction of fertilization rates in vitro in patients with either ~15% of sperm with normal morphology or ~10 sperm bound per ZP. Acknowledgments. We thank Ms. Mingli Liu for technical assistance and all scientists in the IVF laboratory for collecting the oocytes and sperm samples for this study. REFERENCES 1. Liu DY, Du Plessis YP, Nayudu PL, Johnston WIH, Baker HWG. The use of in vitro fertilization to evaluate putative tests of human sperm function. Fertil Steril1988;49: Vol. 58, No.6, December 1992 Liu and Baker Sperm nuclear chromatin normality and IVF 1183

7 2. Liu DY, Elton RA, Johnston WIH, Baker HWG. Spermatozoal nuclear chromatin decondensation in vitro: a test for sperm immaturity, comparison with results of human in vitro fertilisation. Clin Reprod FertiI1987;5:191-20l. 3. Kruger TF, Acosta AA, Simmons KF, Swanson RJ, Matta JF, Oehninger S. Predictive value of abnormal sperm morphology in in vitro fertilization. Fertil Steril1988;49: Liu DY, Baker HWG. The proportion of human sperm with poor morphology but normal intact acrosomes detected with pisum sativum agglutinin correlates with fertilization in vitro. Fertil Steril 1988;50: Liu DY, Lopata A, Johnston WIH, Baker HWG. A human sperm-zona pellucida binding test using oocytes that failed to fertilize in vitro. Fertil Steril 1988;50: Liu DY, Clarke GN, Lopata A, Johnston WIH, Baker HWG. A sperm-zona pellucida binding test and in vitro fertilization. Fertil Steril1989;52: Jeulin C, Feneux D, Serres C, Jouannet P, Guillet-Rosso F, Belaosch-Allart J, et al. Sperm factors related to failure of human in vitro fertilization. J Reprod FertiI1986;76: Tejada RI, Mitchell JC, Norman A, Marik JJ, Friedman S. A test for the practical evaluation of male infertility by acridine orange (AO) fluorescence. Fertil Steril 1984;42:87-9l. 9. Dadoune JP, Mayaux MJ, Guihard-Moscato ML. Correlation between defects in chromatin condensation of human spermatozoa stained by aniline blue and semen characteristics. Andrologia 1988;20: World Health Organization. WHO laboratory manual for the examination of human semen and semen-cervical mucus interaction. 2nd edition. Cambridge: The Press Syndicate of the University of Cambridge, 1987: Liu DY, Baker HWG. Tests of human sperm function and fertilization in vitro. Fertil Steril 1992;58: Terquem A, Dadoune JP. Aniline blue staining of human spermatozoa chromatin: evaluation of nuclear maturation. In: Andre J, editor. The sperm cell. London: Martinus Nijhoff Publishers, 1983: Liu DY, Lopata A, Johnston WIH, Baker HWG. Human sperm-zona binding, sperm characteristics and in-vitro fertilization. Hum Reprod 1989;4: Liu DY, Baker HWG. Morphology of spermatozoa bound to the zona pellucida of human oocytes that failed to fertilize in vitro. J Reprod Fertil 1992;94: Hofmann N, Hilscher B. Use of aniline blue to assess chromatin condensation in morphological normal spermatozoa in normal and infertile men. Hum Reprod 1991;6: Ng FLH, Liu DY, Baker HWG. Comparison of Percoll, Mini Percoll and swim-up methods for sperm preparation from abnormal semen samples. Hum Reprod 1992;7: Le Lannou D, Blanchard Y. Nuclear maturity and morphology of human spermatozoa selected by Percoll density gradient centrifugation or swim-up procedure. J Reprod FertiI1988;84: Katz DF, Diel L, Overstreet JW. Differences in the movement of morphologically normal and abnormal human seminal spermatozoa. BioI Reprod 1982;26: Liu DY, Baker HWG. Relationships between human sperm acrosin, acrosomes, morphology and in vitro fertilization. Hum Reprod 1990;5: Liu DY, Baker HWG. Inducing the human acrosome reaction with a calcium ionophore A23187 decreases sperm-zona pellucida binding with oocytes that failed to fertilize in vitro. J Reprod FertiI1990;89: Liu and Baker Sperm nuclear chromatin normality and IVF Fertility and Sterility