Binding of annexin V to plasma membranes of human spermatozoa: a rapid assay for detection of membrane changes after cryostorage

Transcription

1 Molecular Human Reproduction vol.5 no.2 pp , 1999 Binding of annexin V to plasma membranes of human spermatozoa: a rapid assay for detection of membrane changes after cryostorage H.-J.Glander 1,3 and J.Schaller 2 1 Department of Dermatology/Andrology Unit, University of Leipzig, Liebigstrasse 21, D Leipzig, and 2 Department of Dermatology, St. Barbara Hospital, Duisburg, Germany 3 To whom correspondence should be addressed When the cell membrane is disturbed, phospholipid phosphatidylserine (PS) is translocated from the inner to the outer leaflet of the plasma membrane. This is one of the earliest signs of apoptosis and can be monitored by the calcium-dependent binding of annexin V. Therefore, annexin V-binding, in conjunction with flow cytometry, was used to evaluate the integrity of the sperm plasma membrane after different cryostorage protocols: i.e. 10% (v/v) glycerol; sperm maintenance medium (MM); freezing medium TEST yolk buffer (TYB); or cryostorage without protection (cryoshock). Using a combination of two fluorescent dyes, annexin V and propidium iodide (PI), led to three groups of spermatozoa being identified: (i) viable spermatozoa (annexin V-negative and PI-negative); (ii) dead spermatozoa (annexin V-positive and PI-positive); and (iii) cells with impaired but integer plasma membrane (annexin V-positive and PI-negative). The percentage of vital annexin V-negative spermatozoa increased significantly (P < 0.05) from spermatozoa treated by cryoshock ( %) to spermatozoa cryopreserved by TYB ( %) via cryopreservation by 10% (v/v) glycerol ( %) and by MM ( %) and was associated with the percentage of motile spermatozoa ( % by glycerol; % by MM and % by TYB; P ). Of the spermatozoa, 12 22% were annexin V-positive even though they did not bind to PI, indicating viability before as well as after cryostorage. The percentage of vital annexin V-positive spermatozoa was significantly correlated with different sperm motility parameters (velocity straight linear, r 0.601, P 0.018; percentage of linearly motile spermatozoa: r 0.549, P 0.034). We, therefore, concluded that annexin V-binding is more sensitive in detecting a deterioration of membrane functions than PI staining, and that a considerable percentage of spermatozoa might have dysfunctional plasma membranes besides dead or moribund cells. Of the cryopreservation protocols tested, TYB yielded the most viable spermatozoa. Therefore, we advocate the use of the annexin V-binding assay for the evaluation of the quality and integrity of spermatozoa. Key words: annexin V-binding/apoptosis/cryopreservation/human spermatozoa/plasma membrane Introduction The sperm plasma membrane is one of the key structures affected by cryopreservation (de Baulny et al., 1997). However, the integrity and normal functionality of the sperm plasma membrane are important prerequisites for successful fertilization. The numerous functions of the membrane are related to the cell metabolism for maintaining sperm motility, capacitation, acrosome reaction (Cross and Hanks, 1991) and sperm egg interaction (Hammerstedt et al., 1990). The assessment of membrane integrity is based on the examination of sperm morphology and motility, hypo-osmotic swelling tests (Jeyendran et al., 1984), supravital stains and flow cytometric techniques (Rodriguez-Martinez et al., 1997). Staining with combinations of fluorescent dyes is useful for evaluating viability and functionality of spermatozoa. In the past, 123 rhodamine was used to assess mitochondrial membrane potential and ethidium bromide was used to determine membrane integrity (Evenson et al., 1982). Later, propidium iodide (PI) was combined with other stains, i.e. carboxyfluorescein diacetate (Garner et al., 1986) to evaluate sperm functions. These methods enabled discrimination between live and dead or moribund spermatozoa but did not detect early phases of disturbed membrane functions. During the early phases of disturbed membrane function, asymmetry of membrane phospholipids occurs prior to a progressively disturbed integrity of the plasma membrane (Vermes et al., 1995). In vital cells with intact plasma membranes, the negatively charged membrane phospholipid phosphatidylserine (PS) is located on the inner leaflet of the plasma membrane (Hammerstedt et al., 1990). The disturbance of membrane function starts with the translocation of PS from the inner to the outer leaflet of the plasma membrane and results in an exposure of PS on the external surface (Vermes et al., 1995). This translocation of PS is one of the earliest features of cells undergoing apoptosis and is apparently triggered through an active mechanism (Martin et al., 1995). Depending on Ca 2, PS has a high affinity for annexin V. The latter is a kda phospholipid binding protein belonging to the annexin family (van Heerde et al., 1995) that binds very selectively to PS. Annexin V staining enables the identification of cells with deteriorated European Society of Human Reproduction and Embryology 109

2 H.-J.Glander and J.Schaller Table I. Standard settings in the computer assisted sperm analysis (CASA) of motion Parameter Value Parameter Value Fields 5 Linearity threshold (%) 90 Temperature ( C) 36 Maximum radius for circle (µm) 100 Immotile threshold (IMO) (µm/s) 5 Dilution no Locally motile threshold (LOC) (µm/s) 15 Depth of chamber 10 µm AREA MIN (pixel) 40 FR MAX maximum of frames 32 AREA MAX (pixel) 200 FR MIN minimum of frames 16 Grey threshold 210 membrane integrity at an earlier stage than staining with supravital stains (Vermes et al., 1995). Therefore, annexin V might be suitable for characterization of spermatozoa. Since PS is present on the impaired plasma membrane as well as on the external surface of dead cells, the vital dye propidium iodide (PI) should be used in combination with annexin V. Dead spermatozoa have lost their ability to resist the influx of a membrane-impermeable dye, PI, resulting in intracellular staining. This method enables the identification of three distinct types of spermatozoa: (i) viable spermatozoa (annexin V-negative and propidium iodide-negative, AN /PI ), dead spermatozoa (annexin V-positive and PI-positive, AN /PI ) and spermatozoa with impaired but integer plasma membranes (annexin V-positive and PI-negative, AN /PI ). The objective of this study was to assess changes in the plasma membrane integrity of human spermatozoa using an annexin V-binding assay after different cryostorage protocols. Materials and methods Selection criteria of the semen samples Semen samples (n 25) from 25 men were collected by masturbation into sterile plastic Petri dishes after sexual abstinence for 3 5 days. The group of semen donors consisted of 16 male partners of couples consulting the Department of Andrology, University of Leipzig, Germany, for infertility problems and nine fertile donors who had already fathered a child. The donors were asked to deliver the semen samples after informed consent. Classical semen parameters, including sperm concentration, motility and morphology were examined according to the World Health Organization (WHO) standard guidelines (WHO, 1992). The ejaculates were used for further experiments only if the following requirements were fulfilled: a sperm:leukocyte ratio of more than 100:1; sperm concentration /ml, 30% of the spermatozoa with a normal morphological shape and 50% appearing progressively motile. Highly viscous ejaculates and semen samples with a positive mixed antiglobulin reaction (MAR) test, i.e. 10% spermatozoa with adherent particles, were excluded. Computer-aided sperm analysis (CASA) The Stroemberg Mika cell motion analyser (Version 4.4, Mika Medical GmbH, Rosenheim, Germany) was used to assess sperm motility (Paasch and Glander, 1997). The specimens were observed by a high resolution-ccd-video camera and a microscope Optiphot-2 (Nikon Co, Tokyo, Japan). Aliquots of semen samples (5 µl) were placed into 10 µm deep disposable counting chambers belonging to the Stroemberg Mika system on a 36 C microscope stage warmer. A minimum of 100 spermatozoa from at least four different fields was analysed from each specimen. The following parameters considered in our experiments were determined on motile spermatozoa: average 110 path velocity (VAP; 15 µm/s); curvilinear velocity (VCL; µm/s); and straight line velocity (VSL; µm/s). The standard settings are summarized in Table I. Preparation of spermatozoa Spermatozoa were separated from seminal plasma by density gradient centrifugation. The isotonic gradient medium was composed of Hank s balanced salt solution (HBSS) and Percoll 40% (v/v) to avoid selection of spermatozoa and to remove the gelatinous masses from the semen samples before flow cytometry. The liquefied semen samples were diluted with an equal volume of HBSS and layered onto a 2 ml gradient. After centrifugation (20 min at 400 g) the pellets were washed (300 g for 10 min) with 10 ml HBSS containing 5 mg/ml bovine serum albumin (BSA). The separated sperm pellets were resuspended in HBSS/BSA to obtain a final concentration of ~ /ml before further investigation. Freezing technique The sperm suspensions were divided into four aliquots for four different treatments. The first aliquot was plunged into liquid nitrogen without cryoprotection. The remaining three aliquots were cryoprotected using different protocols: (a) dilution with 10% (v/v) glycerol; (b) dropwise dilution with sperm maintenance medium (MM) containing 28%( v/v) glycerol until a 3:1 sample to medium ratio is reached (Irvine Scientific; catalogue no ); and (c) dropwise dilution with equivalent volume of freezing medium [TES (N-trishydroxymethyl methyl-2-aminoethanesulphonic acid) and Tris] TEST yolk buffer (TYB) containing 12%( v/v) glycerol (Irvine Scientific). The diluted semen samples were placed into 1.8 ml Nunc Cryo Tubes (InterMed) and frozen with the system Nicool LM 10 (Compagnie Francaise de Produits Oxygenes) according to a standard protocol. A slow cooling rate at level 2 that decreases the temperature from 24 C to 5 C within 15 min was followed by an incubation for 15 min at level 10 for a decrease to 70 C (McLaughlin et al., 1990). Finally, the tubes were plunged into liquid nitrogen for storage at 196 C. Thawing was carried out by incubation in a 37 C water bath for 5 min. Determination of FITC-conjugated annexin V-binding After pre-treatment by cryostorage the spermatozoa were adjusted to a concentration of spermatozoa/ml HBSS/BSA and 200 µl were centrifuged at 500 g for 5 min. The pellets were resuspended in 190 µl ice-cold diluted binding buffer (140 mm NaCl, 2.5 mm CaCl 2, 10 mm HEPES/NaOH, ph 7.4) filtered through 0.2 µm pore filter according to the recommendations of the manufacturer of the kit (Zymed Laboratories Inc, San Francisco, CA, USA, distributed by LD Heyden, Germany) and incubated with 10 µl of ready-touse annexin V fluorescein isothiocyanate (FITC) solution, which contained 50 mm Tris, 100 mm NaCl, 1% BSA, 0.02 sodium azide, ph 7.4. The tubes were kept on ice and incubated for 10 min in the

3 Annexin V-binding of cryopreserved spermatozoa Figure 1. Example of a histogram for flow cytometry using the annexin V-binding assay. LFL1 green fluorescence of annexin V; LFL2 red fluorescence of propidium iodide; LFS forward scatter; LSS side scatter. dark. Afterwards, the spermatozoa were washed once, resuspended in 190 µl binding buffer and mixed with 10 µl of dissolved propidium iodide (PI) (20 µg/ml binding buffer). The FITC-labelled spermatozoa were analysed in a flow cytometer Epics Profile II (Coulter Electronics, Krefeld, Germany). A minimum of spermatozoa were examined for each assay at a flow rate of 100 cells/s. The sperm population was gated using 90 and forward-angle light scatter to exclude debris and aggregates. The excitation wavelength was 488 nm supplied by an argon laser at 250 mw. Green (FITC-derived) fluorescence was measured using a nm bandpass filter and the red fluorescence (PI) by a 610 nm filter. An example is shown in Figure 1. The percentage of positive cells and the mean fluorescence was calculated by the Epics software on a 1023 channel scale. As a positive control the binding of FITC-conjugated Pisum sativum agglutinin to the acrosome was used according to method of Miyazaki et al. (1990) for setting the bitmap on dot plot. The negative controls represented spermatozoa only treated with buffer without annexin V and PI. Statistical analysis Data analyses were performed by the non-parametric Mann Whitney U-test for evaluation of the differences as appropriate for data type and distribution. The correlation coefficient (r) between various parameters was determined applying Spearman s rank correlation test utilizing the statistical computer program STATISTICA for Windows from StatSoft Inc (Tulsa, OK, USA). Results are expressed as SEM. P 0.05 was considered to be statistically significant. Results The investigated semen samples showed a sperm concentration of /ml, % of spermatozoa had normal morphology and % appeared to be motile (mean SD). Fresh semen samples contained 71.7% PInegative spermatozoa in total, whereas the cryopreserved samples had between 33.4 and 48.0% PI-negative spermatozoa (Table II, sperm group 1). Depending on the various cryostorage protocols used annexin V-binding was affected significantly and sperm motility decreased compared with fresh semen (Table II). All methods of cryostorage led to a significant decrease of sperm motility (Table III) as well as of percentage of vital and annexin V-negative spermatozoa, AN /PI (P 0.05; Table II, sperm group 2). The percentage of vital annexin V-negative spermatozoa (AN /PI ) increased from spermatozoa treated by cryoshock to spermatozoa cryopreserved by TYB via cryopreservation by glycerol and by MM (Table II, sperm group 2). The differences between the protocol without cryoprotection and the protocol with 10% glycerol were significant on the P 0.05 level. The percentage of vital annexin V-binding spermatozoa (AN /PI ) of the total vital spermatozoa (Table II, sperm group 4) increased from 28.2% in fresh semen samples to 40% after cryostorage. Surprisingly, this group of spermatozoa (Table II, sperm group 4) did not significantly differ comparing the various methods of cryostorage. The increased annexin V-binding indicates that almost half of the spermatozoa classified as vital by PI staining are effected by deterioration of their plasma membranes after cryostorage. Subsequently, the smallest portion of AN /PI was seen after cryopreservation using TYB (Table II, sperm group 5). The three groups of spermatozoa characterized by differential labelling by PI and AN (Table II) did not reach a total of 100%, because a small group of avital spermatozoa ( 8%) 111

4 H.-J.Glander and J.Schaller Table II. Influence of different cryostorage protocols on the staining of human spermatozoa by annexin V-FITC (AN) and propidium iodide (PI). Values are given as means SEM with median shown in parentheses Sperm Sperm Fresh Without 10% (v/v) MM TYB group characteristics semen cryoprotection glycerol 1 PI in m m total (%) (74.7) (25.7) (34.9) 43.7) (47.4) 2 AN and a e e a PI (%) (54.7) (13.1) (18.6) (20.2) (23.5) 3 AN and b g b,c,e c,g e PI (%) (20.2) (8.5) (11.9) (21.3) (20.4) 4 AN (% of PI n n spermatozoa in total) (27.9) (39.0) (36.5) (51.8) (46.5) 5 AN and k f d,f,i,j d j,i,k PI (%) (25.3) (65.7) (58.3) (53.6) (47.0) MM maintenance medium with glycerol (Irvine Sci. Catalogue No 99176); TYB freezing medium TEST yolk buffer (Irvine Sci, Catalogue No 9971); AN /PI annexin-v-negative and propidium iodide-negative sperm; AN /PI annexin V-positive and propidium iodide-positive; AN /PI annexin V- positive and propidium iodide-negative spermatozoa (for details see text). Mann Whitney U-test, values with the same character are significantly different: a P ; b P n.s.; c P 0.048; d P 0.023; e P 0.024; f P 0.024; g P 0.048; h P 0.023; i P 0.013; j P 0.002; k P ; m P ; n P Table III. Influence of different cryostorage protocols on the sperm motility. Values are shown as mean SE with median shown in parentheses Sperm Fresh Without 10% (v/v) MM TYB characteristics semen cryoprotection glycerol Motile spermatozoa (%) a a (63.3) (20.0) (22.7) (25.1) Linearly motile spermatozoa (%) (33.1) (8.0) (13.0) (13.8) VSL (µm/s) (35.8) (14.2) (18.9) (20.2) VAP (µm/s) (43.5) (16.0) (19.3) (22.0) VCL (µm/s) (59.1) (20.1) (23.0) (24.3) MM maintenance medium with glycerol (Irvine Sci. Catalogue No 99176); TYB freezing medium TEST yolk buffer (Irvine Sci, Catalogue No 9971). Mann Whitney U-test, a P did not show any annexin V-binding at all, AN /PI (data not shown). In general, the total number of PI negative spermatozoa correlated with the portion of annexin-negative spermatozoa (r 0.863; P ; Figure 2) but not with the percentage of annexin V-positive ones (P 0.05). The association between vital spermatozoa (PI ) and the three groups of spermatozoa discriminated by annexin V- and PI-staining indicated the impact of the cryostorage method used. There was no correlation in staining without cryoprotection but a correlation of total vital sperm numbers (PI ) with the vital annexin V-negative spermatozoa (AN /Pl ) in the assay utilizing glycerol (r 0.678, P 0.015), MM (r 0.553, P 0.049) or TYB (r 0.602, P 0.029). The motility differed significantly between assays with or without cryoprotection (P 0.05; Table III) but not between the three cryoprotection protocols used (P 0.05). Considering clinical implications the relation between annexin V-binding and sperm motility might be of valuable interest. The percentage of PI- and annexin V-negative cells 112 Figure 2. Spearman s correlation of vital annexin V-negative spermatozoa, AN /PI (%) with percentage of vital (propidium iodide-negative, PI ) spermatozoa in total. (AN /PI ) corresponded well to the percentage of motile spermatozoa (Tables II and III) but exceeded the level of linear motile sperm counts (P 0.05). The group of vital

5 Annexin V-binding of cryopreserved spermatozoa spermatozoa stained by annexin V (AN /PI ) correlated significantly with the percentage of linearly motile spermatozoa, VSL and VAP (Table IV). The AN /PI staining pattern showed a significant, inverse correlation with the motility parameters tested at the P 0.05 level. Storage without cryoprotection resulted in a loss of any sperm motility, however, 15% of the spermatozoa were not stained by either PI or annexin V. Discussion Binding of annexin V to human ejaculated spermatozoa was determined in fresh sperm suspensions, after three different cryoprotection protocols and after freezing without cryoprotection. The various protocols resulted in different PI- and annexin V-binding pattern of spermatozoa. Of the spermatozoa, 12 22% did not show PI binding, indicating viability but was annexin V-positive before as well as after cryostorage. This finding implicates that a considerable number of spermatozoa might exist with altered and possibly dysfunctional plasma membranes besides moribund or dead cells. The annexin V-binding ligand, PS, is located on the outer leaflet on the plasma membrane and identifies cells with deteriorated membrane, which is one of the earliest features of cells undergoing apoptosis (Vermes et al., 1995). The specificity of annexin V-binding to PS has been demonstrated by inhibition experiments. Binding of annexin V to PS on the cell surface was effectively inhibited by PS liposomes but not by liposomes containing other phospholipids (Martin et al., 1995). Functional assays of plasma membrane integrity can potentially characterize the quality of spermatozoa. Several tests have been reported for evaluating plasma membranes, e.g. supravital staining techniques (Evenson et al., 1982; Garner and Johnson, 1995) or the hypo-osmotic swelling test according to Jeyendran et al. (1984). These methods can discriminate viable from dead or damaged cells, but do not monitor early phases of membrane dysfunction or initial phases of apoptosis like the annexin V-binding assay. Annexin V-binding does not involve enzymatic activity and does not require previous fixation of the cells. Thus, this assay allows monitoring apoptosis of living spermatozoa which is not possible by other apoptosis assays reported. The annexin V-binding assay is more sensitive at detecting the early deterioration of membrane functions than PI staining, since we could demonstrate annexin V-positive but PI-negative spermatozoa. Therefore, an improved assessment of sperm quality for assisted fertilization programmes might be achieved through the application of this novel approach. Annexin V-binding assays could supplement the combined Pisum sativum agglutinin FITC/Hoechst staining described by Hinsch et al. (1997). Binding of annexin V can be analysed by flow cytometry, which enables the analysis of several thousand cells in the liquid phase in a short time but does not provide information on the location of fluorescence on the sperm surface. Visual examination of air-dried smears after the cytospin technique may localize the annexin V-binding. However, cytospin techniques may cause damage to the spermatozoa and therefore, Table IV. Spearman s significant correlations (P 0.05) of the sperm motility characteristics with the groups of spermatozoa after annexin V and propidium iodide staining Characteristics Motility Correlation Significance (r ) (P) AN /PI Percentage linear motility VSL VAP AN /PI Percentage linear motility VSL VAP AN /PI annexin V-positive and propidium iodide-negative spermatozoa; AN /PI annexin V-positive and propidium iodide-positive; VCL curvilinear velocity (µm/s); VAP average path velocity (µm/s) and VSL straight line velocity (µm/s). this method is not suitable for assessing the exact localization of the annexin V-binding on the sperm surface. After cooling of biological membranes, a reordering of membrane components is likely (Hammerstedt et al., 1990). The cryopreservation usually causes sublethal cryodamage to spermatozoa, decreasing post-thaw cell viability (Alvarez and Storey, 1993). More precise information is needed on the exact proportion of only slightly disturbed spermatozoa and how this ratio might be affected by cryopreservation (de Baulny et al., 1997). The pattern of annexin V-binding demonstrated in this study indicates that the annexin V-binding assay might be a promising candidate providing the required information. However, the existence of a subgroup of AN /PI spermatozoa becoming immotile after cryoshock freezing in liquid nitrogen diminishes this optimism. As expected the portion of AN /PI spermatozoa decreased significantly after storage in liquid nitrogen without protection (P 0.05) indicating the damaging effects of cryostorage. The cryostorage with glycerol used as cryoprotectivum exclusively yielded the smallest percentage of AN /PI spermatozoa after cryoprotection (Table II, sperm group 2). This result is in agreement with the data of other authors (Alvarez and Storey, 1993; Jedrzejczak et al., 1996) who found significantly lower values with respect to progressive sperm motility and viability using glycerol alone. Our cryoprotection protocols were based on different final concentrations of glycerol, 10% (only glycerol protection), 7% (MM) and 6% (TYB). A glycerol concentration of 6 10% did not cause significantly different motility patterns determined by CASA (Paasch and Glander,1997) but different percentages of annexin V-positive spermatozoa (Table II). It should be considered that other ingredients of the cryoprotective media besides glycerol might influence the plasma membrane, e.g. TYB containing zwitterions which have been shown to improve cryotolerance of spermatozoa (Graham et al., 1972). Furthermore, TEST-yolk buffer (TYB) enhanced the capacitation, acrosome reaction (Ragni et al., 1993) and promoted the sperm binding to zona pellucida (Lanzendorf et al., 1992). The enhanced binding capacity of human spermatozoa following treatment with egg yolk-containing media may be a result of a decrease in cholesterol:phospholipid molar ratio in spermatozoa (Gamzu et al., 1997). Reduction of the cholesterol:phospholipid molar ratio accompanies the 113

6 H.-J.Glander and J.Schaller capacitation process and may also modulate the annexin V-binding. The significantly increased percentage of AN /PI spermatozoa in the fresh sperm suspension compared with cryostorage without cryoprotection ( % versus %; P 0.05) was an unexpected result (Table II, sperm group 3), and might reflect the differential distributions of the sperm groups and a relatively higher percentage of dead spermatozoa using the latter procedure. It was demonstrated that the number of annexin V-positive spermatozoa calculated as percentage of PI-negative cells in total were similar after cryostorage (Table II, sperm group 4). This result might lead to the conclusion that all four methods of storage in liquid nitrogen did not induce profound differential effects on plasma membrane asymmetry related to the number of vital human spermatozoa in total. Of the spermatozoa, ~2 8% were non-viable (PI-positive) but were not stained by annexin V FITC, AN /PI. This group may be characterized by a very high degree of membrane disorganization, which could prevent any binding of annexin V. The Stroemberg Mika cell motion analyser indicated lower sperm motility than other systems, e.g. Celltrak/S system (Kraemer et al., 1998). However, our results corresponded to those determined by de Wijchman et al. (1995) who also used the Stroemberg Mika system, i.e. the results are the consequence of the system applied (Paasch and Glander, 1997) and of the standardized settings (Davis and Katz, 1993). In fresh semen samples of undoubtedly fertile donors we found ~20% annexin V-positive spermatozoa in the vital sperm group. The underlying mechanism of this observation is unknown; however, a severe deterioration of fertility can be excluded since the donors had already fathered children. Prior to cryostorage we separated the spermatozoa from the seminal plasma to exclude possible interactions between seminal plasma and the plasma membrane of spermatozoa during cryostorage. In rabbit seminal plasma, a compound similar to human annexins has been found (Okabe et al., 1993). If this compound also exists in human seminal plasma, it does not affect our examinations because we determined differences between washed spermatozoa cryostored by several protocols. Although motility is not directly related to the fertilizing capacity it is generally accepted to be an important indicator of cryopreserved sperm quality (Cross and Hanks, 1991). As expected there was a negative correlation between sperm motility and the percentage of spermatozoa bound by both annexin V and PI. We found a positive correlation between the percentage of vital spermatozoa stained by annexin V and progressive motility parameters (Table III), i.e. early phases of loss of sperm membrane asymmetry are combined with an increase of sperm forward motility. It cannot be estimated whether the combination of these sperm functions is associated with the capacitation process. The portion of motile spermatozoa corresponded to the percentage of spermatozoa binding neither annexin V nor PI (Table II, sperm group 2) in all assays performed except the cryoshock procedure. This association could be a coincidence as well as the result of a relationship between membrane structure and motility. The latter hypothesis is supported by analogous results in our three 114 different cryoprotection protocols. The annexin V-binding assay seems to provide additional information about sperm deterioration besides conventional motility analysis and supravital staining. Furthermore, immotile spermatozoa were occasionally not stained in the annexin V-binding assay suggesting that not all of the immotile spermatozoa represent irreversibly dead cells. The aim of our study was to draw the attention to this assay which allows the examination of important plasma membrane binding groups on spermatozoa. Further investigations combined with additional sperm function tests and other markers of apoptosis are required in the future. Acknowledgements This study was made possible by a grant of the German Research Council, Deutsche Forschungsgemeinschaft, grant-nr. Gl 199/1 3. The authors gratefully acknowledge C.von Bormann for skilful technical assistance, J.Kleine-Tebbe and A.Jahreis for editorial assistance. References Alvarez, J.G. and Storey, B.T. (1993) Evidence that membrane stress contributes more than lipid peroxidation to sublethal cryodamage in cryopreserved human sperm: glycerol and other polyols as sole cryoprotectant. J. Androl., 14, de Baulny, B.O., Le Vern, Y., Kerboeuf, D. and Maisse, G. (1977) Flow cytometric evaluation of mitochondrial activity and membrane integrity in fresh and cryopreserved rainbow trout spermatozoa. Cryobiology, 34, Cross, N.L. and Hanks, S.E. (1991) Effects of cryopreservation on human sperm acrosomes. Hum. Reprod., 6, Davis, R.O. and Katz, D.F. (1992) Standardization and comparability of CASA instruments. J. Androl., 13, de Wijchman, J.G., Wolf, B.T. and Jager, S. (1995) Evaluation of a computeraided semen analysis system with sperm tail detection. Hum. Reprod., 10, Evenson, D.P., Darzynkiewicz, Z. and Melamed, M.R. (1982) Simultaneous measurement by flow cytometry of sperm cell viability and mitochondrial membrane potential related to cell motility. J. Histochem. Cytochem., 30, Gamzu, R., Yogev, L., Paz, G. et al. (1997) Reduction of sperm cholesterol: phospholipid ratio is a possible mechanism for enhancement of human sperm binding to the zona pellucida following incubation with phosphatidylcholine liposomes. Biol. Reprod., 57, Garner, D.L. and Johnson, L.A. (1995) Viability assessment of mammalian sperm using SYBR-14 and propidium iodide. Biol. Reprod., 53, Garner, D.L., Pinkel, D., Johnson, L.A. and Pace, M.M. (1986) Assessment of spermatozoal function using dual staining and flow cytometric analyses. Biol. Reprod., 34, Graham, E.F., Crabo, B.G. and Brown, K.I. (1972) Effects of some zwitterion buffers on the freezing and storage of spermatozoa. J. Dairy Sci., 55, Hammerstedt, R.H., Graham, J.K. and Nolan, J.P. (1990) Cryopreservation of mammalian sperm: what we ask them to survive. J. Androl., 11, Hinsch, E., Ponce, A.A., Hägele, W. et al. (1997) A new combined in-vitro test model for the identification of substances affecting essential sperm functions. Hum. Reprod., 12, Jedrzejczak, P., Pisarski, T. and Kurpisz, M. (1996) Evaluation of sperm cryoprotective media in respect to fertilizing capacity tested in vitro. Andrologia, 28, Jeyendran, R.S., Van der Ven, H.H., Perez-Pelaez, M. et al. (1984) Development of an assay to assess the functional integrity of the human sperm membrane and its relationship to other semen characteristics. J. Reprod. Fertil., 70, Kraemer, M., Fillion, C., Martin-Pont, B. and Auger, J. (1998) Factors influencing human sperm kinematic measurements by the Celltrak computerassisted sperm analysis system. Hum. Reprod., 13,

7 Annexin V-binding of cryopreserved spermatozoa Lanzendorf, S.E., Holmgren, W.J. and Jeyendran, R.S. (1992) The effect of egg yolk medium on human sperm binding in the hemizona assay. Fertil. Steril., 58, Martin, S.J., Reutelingsperger, C.P.M., McGahon, A.J. et al. (1995) Earlier distribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exper. Med., 182, McLaughlin, E.A., Ford, W.C. and Hull, M.G. (1990) A comparison of the freezing of human semen in the uncirculated vapour above liquid nitrogen and in a commercial semi-programmable freezer. Hum. Reprod., 5, Miyazaki, R., Fukuda, M., Takeuchi, H. et al. (1990) Flow cytometry to evaluate acrosome-reacted sperm. Arch. Androl., 25, Okabe, M., Kishi, Y., Ying, X. et al. (1993) Characterization of capacitation inhibitory protein from rabbit seminal plasma: homology with human Annexins. Biol. Pharm. Bull., 16, Paasch, U. and Glander, H.-J. (1997) Optimisation of cryopreservation of subfertile sperm samples. A computer-assisted study. In Phillips, G.O., von Versen, R., Strong, M. and Nather, A. (eds), Advances in Tissue Banking. Vol.1. World Scientific, pp Ragni, G., Lombroso, G.C., Caccamo, A. et al. (1993) Enhanced quality of capacitated spermatozoa from oligoasthenozoospermic men after incubation in TEST yolk medium. Andrologia, 25, Rodriguez-Martinez, H., Larsson, B. and Pertoft, H. (1997) Evaluation of sperm damage and techniques for sperm clean-up. Reprod. Fertil. Dev., 9, Van Heerde, W.L., Degroot, P.G. and Reutelingsperger, C.P.M. (1995) The complexity of the phospholipid binding protein annexin V. Thromb. Haemost., 73, Vermes, I., Haanen, C. and Reutelingsperger, C.P.M. (1995) A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescence labelled annexin V. J. Immunol. Methods, 180, WHO (1992) Laboratory Manual for the Examination of Human Semen and Sperm Cervical Mucus Interaction. 3rd edn. Cambridge University Press, Cambridge, UK. Received on May 15, 1998; accepted on October 27,