Assessment of human sperm acrosome reaction by flow cytometry: validation and evaluation of the method by fluorescence-activated cell sorting*

Transcription

1 FERTILITY AND STERILITY Vol. 60. No.6. December 1993 Copyright 1993 The American Fertility Society Printed on acid-free paper in U. s. A. Assessment of human sperm acrosome reaction by flow cytometry: validation and evaluation of the method by fluorescence-activated cell sorting* Meike L. Uhler, M.D.t:j: Andrew Leungt II Steven Y. W. Chan, Ph.D.t Ingrid Schmid'\[ Christina Wang, M.D.tll Cedars-Sinai Medical Center, and University of California, Los Angeles School of Medicine, Los Angeles, California Objective: To determine the applicability of flow cytometry to assess human sperm acrosome reaction. Design: Prospective evaluation of semen samples incubated overnight for the development of spontaneous acrosome reaction or exposed to the calcium ionophore A23187 (5 tim) for 3 hours for induction of the acrosome reaction. Setting: University-affiliated tertiary care center. Patients: Normal healthy volunteers. Interventions: The spermatozoa were stained with fluorescein isothiocyanate (FITC)-labeled pea agglutinin. The labeled samples were assessed visually and also subjected to analytic flow cytometry and fluorescence-activated cell sorting. Main Outcome Measures: Acrosome reaction assessed visually and by flow cytometry. Results: Flow cytometric analysis showed a single peak of FITC fluorescence in the washed semen samples. A second peak of lower FITC fluorescence intensity was noted after overnight incubation or exposure to A23187, suggesting loss of fluorescence, which indicated the occurrence of the acrosome reaction. There was a statistically significant correlation between the assessment by the two methods (n = 41). However, although the mean difference between the methods was small (2.49%), the difference between the two methods for each individual sample can vary between -24% to +29%. When the sperm cells were subjected to cell sorting based on green fluorescence intensity, reanalysis and visual scoring verified that the low intensity peak contained a majority of acrosome-reacted spermatozoa (77.52% ± 2.39%). Conclusion: These results validated the flow cytometric method for assessment of acrosomereacted spermatozoa. Although flow cytometry is more objective and less time consuming when many samples are assessed at the same time, the visual method remains a useful and practical procedure in the clinical andrology laboratory. Fertil Steril 1993;60: Key Words: Capacitation, human spermatozoa Received March 18, 1993; revised and accepted August 24, * Presented in part at the 48th Annual Meeting of The American Fertility Society, New Orleans, Louisiana, November 2 to 5, t Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center. :j: Division of Reproductive Endocrinology and Infertility, De- partment of Obstetrics and Gynecology, University of California, Los Angeles School of Medicine. Reprint requests and present address: Meike L. Uhler, M.D., Department of Obstetrics and Gynecology, Loyola U niversity Medical Center, 2160 South First Avenue, Maywood, Illinois II Division of Endocrinology and Metabolism, Department of Medicine, Cedars-Sinai Medical Center. 11 JCCC Flow Cytometry Core Laboratory, University of California, Los Angeles School of Medicine Uhler et al. Flow cytometry assessment of acrosome reaction Fertility and Sterility

2 Capacitation and the acrosome reaction are prerequisites for fertilization by human spermatozoa. The presence of a normal acrosome before zona binding appears to be necessary for fertilization in vitro (1, 2). The assessment of the ability for spermatozoa to undergo spontaneous or induced acrosome reaction by calcium ionophore or follicular fluid has been reported to be of clinical value in predicting the fertilizing capacity of men (3-5). Several methods for detecting acrosome-reacted human spermatozoa (6, 7) have been developed including triple staining techniques (8), indirect immunofluorescence (IIF) using monoclonal (9) or polyclonal antibodies (10, 11), chlortetracycline fluorescence assays (12), and fluorescein-labeled lectins (13). More recently, the use of flow cytometry with fluoresceinated lectin labeling of spermatozoa has been described as a more accurate, faster, simpler, and objective method than fluorescence microscopy for assessment of acrosome-reacted sperm (14, 15). The first objective of this study was to compare the assessment of acrosome reaction by flow cytometry with the conventional visual method using fluorescence microscopy. The second objective was to separate the fluorescein-labeled acrosome intact from the non-fluorescein-iabeled acrosome-reacted sperm cells using a fluorescence-activated cell sorter. The sorted cells were then manually counted using fluorescence microscopy to validate the faster and less subjective method of flow cytometry to assess human sperm acrosome reaction. MATERIALS AND METHODS Preparation and Incubation of Spermatozoa Semen samples were obtained from healthy, normal male donors with normal semen parameters after at least 48 hours of sexual abstinence. Samples were allowed to liquefy for 30 minutes at room temperature before processing. After liquefaction, semen was washed twice by centrifugation at 300 X g for 10 minutes in an N-[2- hydroxyethyl] piperazine - N' - [ 2 - ethane sulfonic acid] (HEPES) buffered Ham's F-I0 medium (GIBCO, Grand Island, NY) with 0.1 % (wt/vol) bovine serum albumin (Sigma Chemical Co., St. Louis, MO). The washed spermatozoa were then suspended at a concentration of 20 X 10 6 /ml in Ham's F-I0 with 10% heat-inactivated pooled human serum (capacitation medium) obtained from preovulatory women undergoing IVF at our lllstitution. Aliquots of the samples were taken for assessment of the acrosome reaction. To induce spontaneous acrosome reaction, spermatozoa were incubated overnight at 37 C in capacitation medium in a 5% CO 2 humidity incubator and the acrosome reaction was assessed. Alternatively, to induce the maximum acrosome reaction (16), spermatozoa were exposed to the calcium ionophore A23187 (final concentration 5 llm; Sigma Chemical Co.) for 3 hours and then the acrosome reaction was assessed. Assessment of the Acrosome Reaction Detection of acrosome-reacted human sperm by the IIF method using the fluorescein isothiocyanate (FITC)-labeled Pisum sativum agglutinin (PSA; Sigma Chemical Co.) was performed according to Cross et al. (13). Briefly, the following procedures were carried out in small sterile culture tubes. The supravital stain Hoechst (Sigma Chemical Co.) was used to distinguish dead from live sperm. Spermatozoa were incubated at 37 C in 5% CO 2 :95% air for 5 minutes in Dulbecco's phosphate-buffered saline (PBS; Sigma Chemical Co.), containing 0.67 llg/ml of Hoechst The spermatozoa were fixed in 2% (vol/vol) paraformaldehyde, washed in Dulbecco's PBS, permeabilized in absolute ethanol, and washed again with Dulbecco's PBS. Fluorescein isothiocyanate-iabeled PSA, 1 mg/ml was added to each tube and incubated at room temperature for 10 minutes. After final washing in Dulbecco's PBS, small aliquots (50 lll) were dried onto microscope slides, and the remainder of the sample was stored at 4 C in the dark for evaluation within 1 week by flow cytometry. Preliminary experiments have demonstrated that cold storage in the dark for up to 1 week had no effect on the results of the scoring of human sperm acrosome reaction. The acrosome reaction was scored using an Olympus BH-2 epifluorescent microscope (Olympus Optical Company, Tokyo, Japan). Dead sperm as evidenced by Hoechst staining were excluded from the scoring of acrosome-reacted sperm. Acrosomal status of the viable sperm was determined by assessing between 100 to 200 consecutive cells for FITC-PSA labeling. Acrosome intact sperm had intensely green fluorescent acrosomal regions and acrosome-reacted sperm had patchy fluorescence (undergoing acrosome reaction), fluorescent equatorial segment (almost completed acrosome reaction), or no fluorescence (acrosome- Vol. 60, No.6, December 1993 Uhler et al. Flow cytometry assessment of acrosome reaction 1077

3 Table 1 Correlation Coefficients of Acrosome-Reacted and Unreacted Spermatozoa by Visual Assessment Versus Flow Cytometry Visual assessment Unreacted Total reactedt Patchy Equatorial No fluorescence Unreacted Flow cytometry (n = 41) Reacted 0.63* -0.61* -0.63* 0.61* -0.63* 0.62* :1: 0.31 * P < O.OOL t Total reacted = Patchy + Equatorial + No fluorescence. :I: P < reacted) in the head region, as previously described (17). All slides were examined by the same observer (M.L.U.). Flow Cytometry and Fluorescence-Activated Cell Sorting Analytic flow cytometry was done on a F ACScan (Becton Dickinson Immunocytometry Systems, San Jose, CA) equipped with an air-cooled 488 nm argon laser. This clinical flow cytometer does not have an ultraviolet tunable argon laser; therefore, Hoechst fluorescence, which we used for exclusion of dead sperm on the microscope could not be used. However, dead cells show altered light scatter properties, usually an increase in the 90 scatter and a decrease in the forward scatter signal (18). Consequently, a gate was set on dot plot distributions of forward versus 90 scatter to exclude debris and clumps from the analysis of intact spermatozoa. In each sample, between 5,000 and 10,000 sperm cells labeled with FITC-PSA were analyzed. Green FITC fluorescence was detected through a 530/30 band pass filter and was displayed on a linear scale. Fluorescence stability of the flow cytometer was monitored daily using standard beads (Calibrite; Becton Dickinson). Equivalent instrument settings were used for all samples. For data analysis, the fluorescence distributions were divided into low fluorescence (relative mean fluorescence < 300 units) and higher fluorescence (relative mean fluorescence > 300 units) intensity areas. The proportion of cells for each area under the curve was calculated and expressed as the percentage of acrosome-reacted and unreacted cells. For confirmation that the two distinct FITCfluorescence distributions represented acrosomereacted spermatozoa (low fluorescence intensity) and acrosome-un reacted spermatozoa (high fluorescence intensity), the remainder ofthe FITC-labeled samples was subjected to fluorescence-activated cell sorting based on green fluorescence on a F AC Star plus flow cytometer (Becton Dickinson) following analytic flow cytometry. To improve the ability of the instrument to distinguish small sperm cells from electronic noise, 90 scatter was used for setting a signal threshold. Aliquots of both sorted subpopulations of cells were dried onto microscope slides, examined under fluorescent microscopy, and scored. Statistical Analyses Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS, Chicago, IL). Correlations were determined using Pearson's rank correlation test. Bland and Altman statistics and plots (19) were used to show the agreement between the visual and flow cytometric methods of assessments of acrosome reaction. RESULTS The typical distributions of acrosome fluorescence as detected by flow cytometry is shown in Figure 1. Green fluorescence representing unreacted spermatozoa is generally found as a single population of cells at linear fluorescence intensity of 300 to 500 (arbitrary units, Fig. 1, top). Overnight incubation in capacitation medium (Fig. 1, middle) or exposure to the calcium ionophore A23187 (Fig. 1, bottom), increased the number of acrosome-reacted spermatozoa. This was demonstrated by the appearance of a peak of lower fluorescence «300 units) distribution, indicating loss of fluorescein -labeled lectin from the sperm head. The mean ± SEM percent total acrosomereacted spermatozoa assessed visually (patchy + equatorial + no fluorescence) and by flow cytometry was ± 2.42 and ± 2.28, respectively, with a significant correlation between the two readings (r = 0.61, P < 0.001) (Table 1). Similarly, the mean percent total acrosome unreacted spermatozoa determined visually and by flow cytometry was ± 2.42 and ± 2.30, respectively. For each subcategory of reacted spermatozoa determined visually, the percent of acrosome-reacted sperm cells was ± 1.66 for those classified as patchy, 3.48 ± 0.72 for equatorial, and 5.78 ± 1.07 for no fluorescence. Correlations for unreacted spermatozoa as Uhler et al. Flow cytometry assessment of acrosome reaction Fertility and Sterility

4 RELATIVE CELL NUMBER RELATIVE CELL NUMBER , , o LINEAR GREEN FLUORESCENCE (arbhrary unhs) , , between the two methods was 2.49% with 2 SD being % and %. Examination of the plot showed a large difference between the two methods, and this difference was not dependent on the mean value. This indicated that although there seemed to be a good general quantitative agreement between the two methods, the difference could be large in the analysis of a single sample. Fluorescence-activated cell sorting (Table 2) demonstrated a good correlation of low fluorescence intensity with a high proportion of acrosomereacted spermatozoa. The percent of acrosomereacted spermatozoa assessed manually in the samples before sorting was ± After cell sorting, reanalysis and visual scoring verified that the cells in the low fluorescence region contained a majority of acrosome-reacted spermatozoa (77.52% ± 2.39%). The increase in acrosome-reacted spermatozoa was evenly distributed in the categories scored as patchy, equatorial, or no fluorescence by the manual method. o BOO 1000 LINEAR GREEN FLUORESCENCE (arbhrary unhs) 5000-r ~ DISCUSSION U sing flow cytometry, we showed that the population of sperm cells represented by high intensity 40 RELATIVE CELL NUMBER 30 o LINEAR GREEN FLUORESCENCE (arbhrary units) Figure 1 Distribution of acrosome fluorescence detected by flow cytometry in washed spermatozoa from a normozoospermic man. Top: After incubation for 3 hours. Middle: After incubation overnight. Bottom: After exposure to the calcium ionophore A23187 for 3 hours. DIFFERENCE (%) ) I I sessed visually and by flow cytometry, as well as for each subcategory of reacted spermatozoa, are presented in Table 1. To further examine whether there was good agreement between the visual and flow cytometric methods of assessments of the acrosome reaction, the Bland and Altman plot of the mean versus the difference of the values obtained from these two methods is shown in Figure 2. The mean difference -30 'I o I I I I I MEAN(%) Figure 2 Bland and Altman plot of the mean versus the difference of the values of acrosome-reacted spermatozoa detected by visual assessment compared with flow cytometry. Solid line represents the mean difference between the two methods; dotted lines denote mean ± 2 SD. Vol. 60, No.6, December 1993 Uhler et ai. Flow cytometry assessment of acrosome reaction 1079

5 Table 2 Unreacted Total reacted Patchy Equatorial No fluorescence Percent of Total Acrosome-Reacted Spermatozoa* Presort ± 2.57t ± ± ± ± 0.91 * n = 22 samples. t Values are means ± SEM. Low fluorescence ± ± ± ± ± 1.93 Postsort High fluorescence ± ± ± ± ± 0.40 green fluorescence was distributed above fluorescence intensity 300 (arbitrary units) before incubation in capacitation media. Spermatozoa incubated overnight or exposed to A23187 to induce the acrosome reaction produced a leftward shift of the fluorescence distribution frequently leading to the appearance of a second peak of lower intensity. This was consistent with the findings of Purvis et al. (15) who indicated that flow cytometry could also be used to monitor the dynamics of the acrosome reaction. However, detection of acrosome-reacted spermatozoa by flow cytometry using labeled FITC PSA is limited to those samples with good initial fluorescence intensity and low background fluorescence. The present study showed a statistically significant correlation (r = 0.61) between the method of flow cytometry to detect fluorescein-labeled acrosome-reacted spermatozoa compared with the visual, manual method that used fluorescence microscopy. This confirmed the findings of several investigators who suggested that flow cytometry offered a fast and quantitative assessment of sperm cell viability and acrosomal integrity (20-22). Additionally, our results were in agreement with that of Miyazaki et al. (14) who also demonstrated a high correlation between the percent acrosome-reacted sperm determined by flow cytometry and fluorescence microscopy. Engh et al. (23) extended their study to acrosomal integrity assessed by flow cytometry in men with intermediate and severely reduced sperm quality. They reported that peanut lectin binding to spermatozoa was diminished and became more variable as the degree of sperm pathology increased. Our investigation was limited to semen samples obtained from healthy, normal male donors with normal semen parameters. The flow cytometric analysis for abnormal semen samples remains to be validated in further studies. Although a significant correlation was demonstrated between flow cytometry and visual assessment of acrosome-reacted spermatozoa, further examination using a plot of the mean versus the difference of the values obtained from these two methods suggested a more cautious interpretation. The Bland and Altman plot showed that the mean of the two methods was close to zero, but the difference for each individual sample could be very large. Thus, when comparing flow cytometry with the visual method, for any single sample, the difference between the two methods can vary between -24% to +29%. Some of the variability might be caused by the different techniques for dead cell discrimination that were used visually and on the flow cytometer. Although both are based on membrane permeability, flow cytometry in addition is using altered 90 scatter signals that could change results in some samples. Furthermore, some ofthe variability may be due to a counting error in the visual method as only between 100 to 200 cells were assessed for the acrosome reaction. We attempted to minimize observer error by having all the slides scored by one observer. The use of flow cytometry to assess acrosomereacted spermatozoa was subsequently validated by subjecting the sperm cells to fluorescence-activated cell sorting. Visual determination of the population of cells sorted into low and high fluorescence intensity regions confirmed the prevalence of acrosomereacted spermatozoa in the low fluorescence area. These findings confirmed for the first time the validity of flow cytometry to assess spermatozoa on the basis of fluorescence intensity. Although the number of acrosome-reacted cells is high in the low fluorescence fraction, it did not reach 100% because of the difficulty in obtaining pure populations of cells as small as spermatozoa through cell sorting. The proportion of acrosome-reacted spermatozoa scored visually after sorting is, therefore, influenced by the purity of the sorted populations and could be considerably higher. Flow cytometry assessment of human sperm acrosome reaction is more objective and less time consuming than the visual method that requires scoring a minimum of 100 to 200 individual cells for the acrosome reaction. However, there are certain limitations to the flow cytometry technique that deserve mention. First, the method of flow cytometry is efficient only when large numbers of samples are assessed in a batch, particularly for research purposes. Second, detection of acrosome-reacted spermatozoa by flow cytometry is limited to those 1080 Uhler et al. Flow cytometry assessment of acrosome reaction Fertility and Sterility

6 samples with good initial fluorescence intensity and low background fluorescence. Third, flow cytometry cannot provide as detailed information about the subcategories of acrosome-reacted spermatozoa as the visual method, and fourth, flow cytometric analysis of abnormal semen samples has not been validated. Thus, we conclude that the visual method is a reliable, adequate, useful routine laboratory procedure and is probably the current preferred method of assessment of the human acrosome reaction in the clinical setting. The value of flow cytometry for assessment of acrosome reaction could be for research investigations examining the induction of acrosome reaction by physiological or pharmacological agents. The significance of the acrosome reaction determined by flow cytometry needs to be further evaluated in clinical studies. Acknowledgments. We are grateful to Howard L. Judd, M.D., Division of Reproductive Endocrinology, University of California, Los Angeles School of Medicine, Los Angeles, California, for his support and encouragement of our research endeavors. We thank David Bockstoce, J CCC Flow Cytometry Core Laboratory, University of California, Los Angeles School of Medicine, Los Angeles, California, for his expert advice and technical assistance. REFERENCES 1. 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, 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 Fertil 1990;89: Calvo L, Vantman D, Banks SM, Tezon J, Koukoulis GN, Dennison L, et al. Follicular fluid-induced acrosome reaction distinguishes a subgroup of men with unexplained infertility not identified by semen analysis. Fertil Steril 1989;52: Cummins JM, Pember SM, Jequier AM, Yovich JL, Hartmann PE. A test of the human sperm acrosome reaction following ionophore challenge: relationship to fertility and other seminal parameters. J AndroI1991;12: Pilikian S, Guerin JF, Adeleine P, Ecochard R, Czyba JC. Spontaneous and ionophore induced acrosome reaction in asthenozoospermic infertile men. Hum Reprod 1992;7: Cross NL, Meizel S. Methods for evaluating the acrosome status of mammalian sperm. BioI Reprod 1989;41: Wolf DP. Acrosome status quantitation in human sperm. Am J Reprod ImmunoI1989;20: Talbot P, Chacon RS. A triple stain technique for evaluating normal acrosome reactions of human sperm. J Exp Zool 1981;215: Wolf DP, Boldt J, Byrd W, Bechtol KB. Acrosomal status evaluation in human ejaculated sperm with monoclonal antibodies. BioI Reprod 1985;32: Toepfer-Petersen E, Auerbeck J, Weiss A, Friess AE, Schill W-B. The sperm acrosome: immunological analysis using polyclonal and monoclonal antibodies directed against the outer acrosomal membrane of boar spermatozoa. Andrologia 1986;18: Sanchez R, Toepfer-Petersen E, Aitken RJ, Schill W-B. A new method for evaluation of the acrosome reaction in viable human spermatozoa. Andrologia 1991;23: Lee MA, Trucco GS, Bechtol KB, Wummer N, Kopf GS, Blasco L, et al. Capacitation and acrosome reactions in human spermatozoa monitored by a chlortetracycline fluorescence assay. Fertil Steril 1987;48: Cross NL, Morales P, Overstreet JW, Hanson FW. Two simple methods for detecting acrosome-reacted human sperm. Gamete Res 1986;15: Miyazaki R, Fukuda M, Takeuchi H, Itoh S, Takada M. Flow cytometry to evaluate acrosome-reacted sperm. Arch Androl 1990;25: Purvis K, Rui H, Scholberg A, Hesla S, Clausen OPF. Application of flow cytometry to studies on the human acrosome. J Androl 1990;11: Aitken RJ, Ross A, Hargreave T, Richardson D, Best T. Analysis of human sperm function following exposure to the ionophore A23187: comparison of normospermic and oligospermic men. J AndroI1984;5: Mortimer D, Curtis EF, Camenzind AR, Tanaka S. The spontaneous acrosome reaction of human spermatozoa incubated in vitro. Hum Reprod 1989;4: Loken MR, Herzenberg LA. Analysis of cell populations using FACS. Ann NY Acad Sci 1975;254: Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1: Garner DL, Pinkel D, Johnson LA, Pace MM. Assessment of spermatozoal function using dual fluorescent staining and flow cytometric analyses. Bioi Reprod 1986;34: Fenichel P, Hsi BL, Farahifar D, Donzeau M, Barrier-Delpech D, Yeh CJG. Evaluation of the human acrosome reaction using a monoclonal antibody, GB24, and fluorescenceactivated cell sorter. J Reprod FertiI1989;87: Graham JK, Kunze E, Hammerstedt RH. Analysis of sperm cell viability, acrosomal integrity, and mitochondrial function using flow cytometry. BioI Reprod 1990;43: Engh E, Clausen OPF, Purvis K. Acrosomal integrity by flow cytometry in men with variable sperm quality. Hum Reprod 1991;6: Vol. 60, No. 0, December 1993 Uhler et al. Flow cytometry assessment of acrosome reaction 1081