FERTILITY AND STERILITY
Vol. 58, No.3, September 1992
Printed on acid-free paper in U.S.A.
Copyright IV 1992 The American Fertility Society
Identification of human sperm antigens reacting with antisperm antibodies from sera and genital tract secretions*
Shafrira Shai, D.Sc. Yehudith Naot, D.Sc.t Department of Immunology, Technion-Israel Institute of Technology Haifa, Israel
Objective: To identify sperm antigens reacting with antisperm antibodies relevant in human infertility. Design: The reactions of separated sperm antigens with antibodies present in sera and genital tract secretions from infertile and fertile females and males were examined by immunoblotting techniques. Setting: The patients were followed in an outpatient setting of a hospital clinic. Patients: One hundred consecutive infertile males and females, referred for determinations of antisperm antibodies, comprised the study group. Fifty hospital and faculty employees with proven fertility served as a control group. Results: A high proportion of sera from fertile and infertile humans contained antibodies reacting with at least one sperm antigen. However, two discrete bands of antigenic proteins with molecular weights of 44 and 72 kd reacted significantly more frequently with serum antibodies from infertile females than from fertile females. No apparent correlation could be demonstrated between any particular antigen and serum antibodies from infertile males. Nevertheless, antigenic proteins of 62 kd were identified as the major sperm antigens reacting with antibodies present in seminal plasmas from infertile males. Conclusions: The major sperm antigens reacting with systemic antibodies differ from the antigens recognized by local antisperm antibodies. Sperm antigens exhibiting relative molecular weights of 62 kd are major antigens reactive with local antisperm antibodies from infertile humans. Fertil Steril 1992;58:593-8 Key Words: Antisperm antibodies, sperm antigens, immune infertility
Identification and characterization of human sperm antigens capable of eliciting the production of antisperm antibodies is important for understanding the mechanisms involved in antibody-mediated impairment of reproduction. This knowledge is also necessary in attempts to provide new methods to regulate fertility as well as for development of reliable, clinically significant assays for determinations of antisperm antibodies. Received November 19, 1991; revised and accepted May 13, 1992. * Supported by a grant from the Technion 2000 Group, Los Angeles. California. t Reprint requests: Yehudith Naot, D.Sc., Department of Immunology, Faculty of Medicine, Technion, Bat-Galim, Post Office Box 9649, Haifa, Israel. Vol. 58, No.3, September 1992
Several studies were reported on identification of detergent-solubilized sperm antigens recognized by autoserum and isoserum antibodies from infertile individuals (1-16). Sperm antigens separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis technique (SDS-PAGE) were transferred to nitrocellulose membranes and exposed to reactions with sera containing antisperm antibodies. The reports by different groups of investigators revealed variable results on the molecular weights (MWs) of spermatozoal antigens that were recognized by serum antibodies. These differences in patterns and MW s of the identified sperm antigens that reacted with serum antibodies may have resulted from differences in the procedures used for extraction of sperm antigens and from differences in the serum Shai and Naot
Identification of human sperm antigens
593
samples used in each study. Furthermore, because in most of the reported studies the reactions of sperm antigens with control sera from fertile individuals were not reported, the clinical relevance ofthe identified sperm antigens in infertility remained unclear. A unique sperm specific fertilization antigen-1 (FA-I) was recently characterized by Naz (17) and Naz et al. (18). This antigen present on human and murine male germ cells reacted strongly with sera from infertile males and females and reacted only weakly with sera from control fertile individuals. It has been postulated by Naz (17) that because antibodies to the F A-I do not agglutinate or immobilize sperm cells, these antibodies block fertilization by affecting the interactions between sperm and zona. In a recent report, we have shown that although antibodies in sera and genital secretions from fertile individuals react with soluble sperm antigens, statistically significant differences in levels of antisperm antibodies exist between infertile and fertile males and females when determined by a reverse (antibody capture) enzyme-linked immunosorbent assay (ELISA) (19). Furthermore, it has been demonstrated that the presence of antibodies in sera was not necessarily reflected in individual's genital tract secretion and vice versa. It was therefore suggested that for correct evaluation of sperm immunity antisperm antibodies should be determined not only in sera but also in genital tract secretions. In the present study, we have attempted to identify those sperm specific antigens that are recognized by antibodies present in sera and in genital tract secretions from infertile individuals in comparison with the antigens reacting with antibodies present m samples from fertile males and females. MATERIALS AND METHODS Patients and Samples
Fifty serum, 49 seminal plasma, and 42 cervical mucus (CM) samples were obtained from 100 consecutive infertile males and females, defined as suffering from unexplained infertility, and referred for determinations of antisperm antibodies. The patients, followed in an outpatient setting of a hospital clinic, had a history of infertility of at least 2 years' duration with no indication for hormonal or physical causes for their infertility. Male patients showed no significant abnormalities in semen analysis, exhibiting at least 20 X 106 sperm cell/mL, 2 mL of volume, 40% progressive motility, 50% normal morphology, and were found negative for the presence 594
Shai and Naot
Identification of human sperm antigens
of Mycoplasma hominis and Ureaplasma urealyticum. Female patients were tested for hormonal concentrations, basal body temperature, and underwent hysterosalpinogography, laparoscopy, and ultrasound monitoring of follicular development. Antisperm antibodies were tested in sera and genital tract secretions using the reverse ELISA as previously described (19). The results showed similar distribution of positive and negative samples as our previously reported data. Of 50 sera tested, 8 (32%) females and 7 (28%) males exhibited serum antisperm antibodies. Fourteen (28.5%) CM specimens and 11 (26.2%) seminal plasma samples were positive for antisperm antibodies as determined by the reverse ELISA. Control samples, 50 sera, 20 seminal plasma, and 9 CM samples were obtained from fifty hospital and faculty employees with recent proven fertility of the same age group (mean ± SD, 29.1 ± 2.8; range, 22 to 41 years) as the infertile individuals. Cervical mucus samples were collected 48 to 72 hours before ovulation after at least 48 hours of sexual abstinence. Treatment and solubilization of CM samples with bromelin (B-2252; Sigma, St. Louis, MO) were performed as previously described (19). Seminal plasma samples were separated from sperm cells by centrifugation after a liquefaction period of 30 minutes at room temperature. Handling of samples and treatment with bromelin were performed as previously (19). Preparation of Sperm Antigens
Specimens were obtained from fertile volunteers with normal semen quality. The samples used for isolation of sperm antigens had at least 20 X 106 sperm cell/mL, 2 mL of volume, 40% progressive motility, 50% normal morphology, and were found negative for the presence of M. hominis and U. urealyticum. After liquefaction, semen samples were centrifuged at 1,000 X g for 10 minutes and seminal plasma specimens separated. Sperm cells were resuspended in Dulbecco modified phosphate-buffered saline (D-5773; Sigma) containing 3 mM phenylmethyl sulphonyl fluoride (Sigma) as a protease inhibitor. Sperm cells were washed two times by centrifugation at 1,000 X g for 10 minutes in the above buffer. Washed cells from at least 20 individuals were pooled, suspended at a concentration of about 200 X 106 cells/mL, and subjected to at least 20 cycles of freezing and thawing in liquid nitrogen. The suspension was sonicated for 3 minutes at 4°C at 12 JLm (150-watt ultrasonic disintegrator; MSE Crawley, United Kingdom). Unbroken sperm cells were Fertility and Sterility
Table 1
Results of Immunoblotting Reactions Between Sperm Antigens and Serum Samples From Infertile and Fertile Individuals No. of individuals with antibodies reacting with the specified sperm antigens:j:
P*/Tt Sera obtained from Infertile females Fertile females Infertile males Fertile males
21/25 19/25 16/25 18/25
(84) § (76) (64) (72)
~116
o 4
o o
110 95
90
87
82
72
67
62
54
50
45
44
33
26
24
19
15
14
~12
12 8 10 13
1 3
5 5 6 5
4
5 0 5 1
3 2 2 1
1 0 0 0
2 3 1 0
6 6 2 0
5 4 1 1
8 2 3 1
2 0 1 0
2 3 1 0
2 1 2 2
4 3 3 1
3 6 1 3
0 7 3 6
15 7 11 13
3 2 1 3
o 3
1 3 2
* Number of individuals with antibodies reacting with at least one sperm antigen. t Total number of individuals tested in a given group.
removed by centrifugation at 1,000 X g for 10 minutes. The supernatants collected that contained approximately 2.5 mg protein/mL were divided into 0.5 ml aliquots and stored at -70°C. Gel Electrophoresis and Immunoblotting
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed according to the method of Laemmli (20). Stacking gels composed of 4% acrylamide and resolving gels containing 10% acrylamide were run in a vertical slab gel unit (Hoefer Scientific Instruments, San Francisco, CA). Sperm antigens were suspended in TRIS-HCl (62.5 mM) buffer (pH 6.8) containing 2% SDS, 10% glycerol, 5% iJ-mercapto-ethanol, and 0.1% bromophenol blue and immersed for 4 minutes in boiling water. Gels were loaded with 100 jlg/well of sperm extracts and were run for 18 hours at a constant voltage of 20 volts. Molecular weight standards at the range of 14 to 205 kd MW (SDS-7 and SDS6H; Sigma) were run in parallel. After electrophoresis, Western blotting techniques as described by Towbin et al. (21) were performed. Briefly, sperm antigens were transferred onto nitrocellulose sheets (45 jlM pore size; Tamar Ltd., Jerusalem, Israel) by electrophoretic blotting for 1 hour at 1 amp and additional hour at 1.5 amp. The transfer buffer contained 10 mM Na2C03 pH 9.9 (22). For enzymelinked immunodetection of sperm antigens on nitrocellulose' membranes were first incubated in blotting buffer containing 10 mM TRIS-HCl, 0.15 M NaCl, 1 mM ethylenediaminetetraacetic acid, 0.1 % Tween-20, 0.04% NaN 3, and 3% (wt/vol) nonfat dry milk at pH 7.6, to block unsaturated sites, overnight at 4°C. Sera, CM, and seminal plasma samples were diluted in blotting buffer to dilutions of 1:100; 1:40, and 1:8, respectively, and incubated with antigens on strips for 18 hours at room temperature with Vol. 58, No.3, September 1992
:j: Relative MWs in kd. § Values in parentheses are percents.
constant agitation. Strips were washed four times for 10 minutes each wash, with blotting buffer. Alkaline phosphatase-conjugated, affinity-purified, rabbit immunoglobulins (Ig) to human IgM, IgG, and IgA (Dako Patts A190, Copenhagen, Denmark) diluted in blotting buffer were added at a concentration of 2.5 jlg/mL; and incubation was carried out for 1 hour at 37°C. Strips were washed again four times and subjected to substrate for alkaline phosphatase (5-bromo-4-chloro 3 indolyl phosphate, B-6149; Sigma) for 2 hours at 37°C. Statistical analysis of data was performed using Yate's corrected X2 test for independence.
RESULTS
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of solubilized spermatozoal antigens revealed approximately 20 major protein bands with relative MWs ranging from 12 to 116 kd. A reaction was considered positive in the enzyme-linked immunoblotting reactions, regardless of the intensity of the stained bands of antigen-antibody complexes on nitrocellulose membranes. Table 1 summarizes the distribution of positive reactions obtained by immunoblotting between separated sperm antigens of different MWs and antisperm antibodies present in sera from fertile and infertile individuals. It can be seen from Table 1 that most of the sera tested exhibited antibodies to at least one sperm antigen whether obtained from infertile or from fertile donors. Statistical analysis of the data using Yate's corrected X2 test for independence showed that antisperm antibodies present in sera from infertile females reacted with two bands of sperm antigens with relative MWs of 44 and 72 kd significantly more frequently than antisperm antibodies in sera from fertile control females (P < 0.1). Unlike these differences between infertile Shai and Naot
Identification of human sperm antigens
595
and fertile females, sera from fertile and infertile males revealed statistically similar distributions of reactions with separated sperm antigens. Based on these results and on our previously reported data (19) that showed that antisperm antibodies present in sera are not necessarily reflected in individual's genital tract secretions and vice versa, we were further motivated to examine the reactions of sperm antigens with antibodies present in genital tract secretions. Electrophoretically separated sperm antigens were, therefore, exposed to reactions with solubilized samples of CM and seminal plasma from infertile and fertile human subjects. The results of immunoblotting assays are summarized in Table 2. Antisperm antibodies present in CM samples from infertile females reacted preferentially with sperm antigens exhibiting relative MWs of 62 kd and 87 kd, as compared with fewer reactions with other sperm antigens. However, comparison of the data obtained with samples from infertile women with those obtained with fertile controls showed statistically insignificant differences probably because of the limited number of samples available for testing from the group of fertile females. Seminal plasma samples from infertile men contained antisperm antibodies that reacted significantly more frequently (P < 0.1) with sperm antigens exhibiting MW of 62 kd than reactions of antibodies present in seminal plasmas from fertile males. Interestingly, these 62 kd sperm antigens rarely reacted with antibodies present in sera from either infertile or fertile males and females (Table 1). Considered together, the immunoblotting test results revealed differences in the antigens that react with serum antibodies and those reacting with antibodies present in genital tract secretions.
DISCUSSION
This study was carried out in an attempt to identify spermatozoal antigens recognized by antisperm antibodies present in sera and in genital tract secretions from infertile men and women in comparison with the sperm antigens reacting with antibodies from fertile males and females. Most of the studies carried out in the past used immunoblotting techniques as an experimental tool (1-16) and revealed variable results on the MW of spermatozoal antigens recognized by serum antisperm antibodies (3-6,8). Furthermore, because in most studies sera from fertile controls were not tested, the clinical significance of the identified sperm polypeptides remained obscure. More extensive immunoblotting assays with different groups of sera from fertile and infertile individuals of both sexes, performed previously (8, 9, 11) and in this study, clearly demonstrated that a high proportion of sera from all clinical categories contained different levels of antibodies reacting with at least one spermatozoal antigen. This observation, also revealed in immunofluorescence and ELISAs, which are similar in their basic methodology (19, 23,24), led to the suggestion (9) that antisperm antibodies in fertile subjects may reflect a natural physiological phenomenon. To avoid these background physiological reactions, observed when sera were diluted to 1:100, Lehmann et al. (9) also tested sera at higher dilutions of 1:2,000 and have found that antibodies to a sperm polypeptide with 14 kd MW are detected in a significantly higher proportion among males with unexplained infertility. This 14 kd polypeptide could not be observed in our study because we have tested sera from males at 1:100 dilution.
Table 2 Results of Immunoblotting Reactions Between Sperm Antigens and eM and Seminal Plasma Samples From Infertile and Fertile Individuals No. of individuals with antibodies reacting with the specified sperm antigens:j:
P*/Tt Genital tract secretions obtained from Infertile females Fertile females Infertile males Fertile males
21/42 1/9 19/49 2/20
(50) § (11) (38) (10)
~116
o o o o
110
95
90
87
82
72
67
62
54
50
45
44
33
26
24
19
15
14
~12
2
o o
o
1
o o
3
1 2
1 1
12 1 2 2
2 0 1 0
0 0 0 0
7 1 11 0
0 0 0 0
3 0 6 0
3 1 0 0
1 0 3 0
0 0 0 0
1 0 2 0
3 0 0 0
1 0 0 0
3 0 0 0
2 0 0 0
0 0 0 0
o o
1 1
* Number of individuals with antibodies reacting with at least one sperm antigen. t Total number of individuals tested in a given group. 596
Shai and Naot
Identification of human sperm ontigens
:j: Relative MW in kd. § Values in parentheses are percents.
Fertility and Sterility
Autoantibodies against various sperm antigens were also shown by Hald et al. (11) in all sera obtained from a group of 16 males before and after vasectomy. These investigators reported that although two postoperative patients developed serum agglutinating antibodies, there was no change in the pattern of their immunoblotting reactions. On the other hand, another patient without sperm agglutinins showed increased serum activity with a 31.5 kdpolypeptide. Other investigators (10, 12) extended these observations and demonstrated that although antisperm antibodies bind to discrete sperm-associated antigens, there is no correlation between the pattern of reactions of sera with separated discrete sperm antigens and the activity of sera in producing different modes of agglutination, in complement dependent immobilization of sperm cells and in stimulation or suppression of sperm-oocyte fusion. This lack of correlation between these biological effects of antibodies on viable sperm cells functions and between the reactions of antibodies with detergentsolubilized sperm antigens may reflect a true phenomenon. However, it may also be a result of the denaturing effects of immunoblotting procedures that denature native antigens and possibly expose new antigens. This limitation of SDS-PAGE procedures was overcome by Saji et al. (13) who first used isoantiserum to isolate a 15 kd glycopeptide from sperm cell membranes and then showed that this antigen reacted with sera from infertile females possessing sperm-immobilizing antibodies, whereas sera from fertile females and males did not bind to this sperm glycopeptide. Saji et al. (13) suggested that this spermatozoal antigen is one of the corresponding antigens to sperm-immobilizing antibodies. The isoantigen described by Saji et al. (13) differs from the FA-1 isolated by Naz et al. (17, 18) not only in their relative MWs but also in their biological significance, inasmuch as antibodies produced against the FA-1 were demonstrated to be nonreactive in sperm agglutination and immobilization tests (17, 18). In SDS-PAGE experiments, isolated FA-1 exhibited MW s of 47 to 50 kd, which may correspond to one of the two discrete sperm polypeptides with relative MW s of 44 kd and 72 kd demonstrated in this study to react significantly more often with serum antibodies from infertile females than with serum antibodies from fertile females or from both infertile and fertile males. Unlike previous reports, this study differentiated between sperm antigens reactive with serum antibodies and those antigens that are recognized by Vol. 58, No.3, September 1992
local antibodies. Our data demonstrated that although serum antibodies react with antigenic determinants on 44 and 72 kd sperm antigens, local antisperm antibodies from infertile females recognize epitopes present on the 62 and 87 kd isoantigens of normal sperm cells. The dichotomy between local and systemic humoral responses, reflected by the differences in specificities of systemic and local antisperm antibodies, was exhibited not only in infertile females but also in infertile males. A sperm autoantigen of 62 kd MW was identified as a major sperm antigen reacting with antisperm antibodies present in seminal plasma samples obtained from infertile males. This particular sperm antigen with 62 kd MW reacted only rarely with genital secretions from fertile subjects or with serum antibodies, whether obtained from females or from males and regardless of their reproductive potential. It should be mentioned that recently Mathur and associates (14) suggested that sperm cells from infertile males contain, in addition to common antigens present on normal cells, special antigens recognized by autoimmune sera. This novel proposal may explain the inability to demonstrate statistically significant differences in reactions of sera from infertile and fertile males with normal sperm antigen. Whether the proposed special antigens observed on sperm from autoimmune infertile men are also triggering antibodies in genital tract secretions of both sexes remains to be examined. REFERENCES 1. Young LG, Goodman SA. Characterization of human sperm cell surface components. BioI Reprod 1980;23:826-35. 2. Hjort T, Poulsen F. Analysis of auto-antigens in the human sperm membrane by a F(ab)2 blocking system. J Clin Lab ImmunoI1981;6:61-7. 3. Poulsen F, Hjort T. Identification of auto-antigens of the human sperm membrane. J Clin Lab ImmunoI1981;6:69-74. 4. Poulsen F. The nature of an iso-antigen of the human sperm membrane. J Reprod Immunol 1983;5:49-54. 5. Lee C-Y G, Huang Y oS, Hu P-C, Gomel V, Menge AC. Analysis of sperm antigens by sodium dodecyl sulfate gel protein blot radioimmunobinding method. Anal Biochem 1982;123: 14-22. 6. Lee Coy G, Lum V, Wong E, Menge AC, Huang Y-S. Identification of human sperm antigens to antisperm antibodies. Am J Reprod ImmunoI1983;3:183-7. 7. Lee Coy G, Wong E, Richter DE, Menge AC. Monoclonal antibodies to human sperm antigens-II. J Reprod Immunol 1984;6:227 -8. 8. Naaby-Hansen S, Bjerrum OJ. Auto and iso antigens of human spermatozoa detected by immunoblotting with human sera after SDS-PAGE. J Reprod ImmunoI1985;7:41-57. 9. Lehmann D, Temminck B, Rugna DDA, Leibundgut B, Muller H. Blot immunobinding test for the detection of anti sperm antibodies. J Reprod Immunol 1985;8:329-36.
Shai and Naot
Identification of human sperm antigens
597
10. Aitken RJ, Hulme MJ, Henderson CJ, Hargreare TB, Ross A. Analysis of the surface labelling characteristics of human spermatozoa and the interaction with antisperm antibodies. J Reprod FertiI1987;80:473-85. 11. Hald J, Naaby-Hansen S, Egense J, Hjort T, Bjerrum OJ. Autoantibodies against spermatozal antigens detected by immunoblotting and agglutination. A longitudinal study of vasectomized males. J Reprod Immunol 1987;10:15-26. 12. Parslow JMM, Poulton TA, Hay FC. Characterization of sperm antigens reacting with human antisperm antibodies. Clin Exp ImmunoI1987;69:179-87. 13. Saji F, Ohashi K, Kamiura S, Negoro T, Tanizawa O. Identification and characterization of a human sperm antigen corresponding to sperm immunobilizing antibodies. Am J Reprod Immunol Microbiol 1988;17:128-33. 14. Mathur S, Chao L, Goust JM, Milroy GT, Woodley-Miller C, Caldwell JZ. Special antigens on sperm from autoimmune infertile men. Am J Reprod Immunol 1988;17:5-13. 15. Howe SE, Grider SL, Lynch DM, Fink LM. Antisperm antibody binding to human acrosin: a study of patients with unexplained infertility. Fertil Steril 1991;55:1176-82. 16. Cunningham DS, Fulgham DL, Rayl DL, Hansen KA, Alexander NJ. Antisperm antibodies to sperm surface antigens in women with genital tract infection. Am J Obstet Gynecol 1991;164:791-6. 17. N az RK. The fertilization antigen (FA -1): applications in
598
Shai and Naot
Identification of human sperm antigens
18.
19.
20. 21.
22.
23.
24.
immunocontraception and infertility in humans. Am J Reprod Immunol MicrobioI1988;16:21-7. Naz RK, Phillips TM, Rosenblum BB. Characterization of the fertilization antigen 1 for the development of a contraceptive vaccine. Proc Natl Acad Sci USA 1986;83:5713-7. Shai S, Bar-Yoseph N, Peer E, Naot Y. A reverse (antibody capture) enzyme linked immunosorbent assay for detection of antisperm antibodies in sera and genital tract secretions. Fertil Steril 1990;54:894-901. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T 4 • Nature 1970;227:680-5. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350-4. Dunn SD. Effects of the modification of transfer buffer composition and the renaturation of proteins in gels on the recognition of proteins on western blots by monoclonal antibodies. Anal Biochem 1986;157:144-53. Hjort T, Poulsen F. Reactivity of F(ab')2 fragments of sperm antibodies in agglutination immobilization and immunofluorescent test. Arch Androl 1978;1:83-9. Mettler L, Czuppon AB, Alexander N, D'Alnedia M, Hass GG Jr, Hjort T, et al. Antibodies to spermatozoa and seminal plasma antigens detected by various enzyme-linked immunosorbent ELISA assays. J Reprod ImmunoI1985;8:301-12.
Fertility and Sterility
Vol. 58, No.3, September 1992
Printed on acid-free paper in U.S.A.
Copyright IV 1992 The American Fertility Society
Identification of human sperm antigens reacting with antisperm antibodies from sera and genital tract secretions*
Shafrira Shai, D.Sc. Yehudith Naot, D.Sc.t Department of Immunology, Technion-Israel Institute of Technology Haifa, Israel
Objective: To identify sperm antigens reacting with antisperm antibodies relevant in human infertility. Design: The reactions of separated sperm antigens with antibodies present in sera and genital tract secretions from infertile and fertile females and males were examined by immunoblotting techniques. Setting: The patients were followed in an outpatient setting of a hospital clinic. Patients: One hundred consecutive infertile males and females, referred for determinations of antisperm antibodies, comprised the study group. Fifty hospital and faculty employees with proven fertility served as a control group. Results: A high proportion of sera from fertile and infertile humans contained antibodies reacting with at least one sperm antigen. However, two discrete bands of antigenic proteins with molecular weights of 44 and 72 kd reacted significantly more frequently with serum antibodies from infertile females than from fertile females. No apparent correlation could be demonstrated between any particular antigen and serum antibodies from infertile males. Nevertheless, antigenic proteins of 62 kd were identified as the major sperm antigens reacting with antibodies present in seminal plasmas from infertile males. Conclusions: The major sperm antigens reacting with systemic antibodies differ from the antigens recognized by local antisperm antibodies. Sperm antigens exhibiting relative molecular weights of 62 kd are major antigens reactive with local antisperm antibodies from infertile humans. Fertil Steril 1992;58:593-8 Key Words: Antisperm antibodies, sperm antigens, immune infertility
Identification and characterization of human sperm antigens capable of eliciting the production of antisperm antibodies is important for understanding the mechanisms involved in antibody-mediated impairment of reproduction. This knowledge is also necessary in attempts to provide new methods to regulate fertility as well as for development of reliable, clinically significant assays for determinations of antisperm antibodies. Received November 19, 1991; revised and accepted May 13, 1992. * Supported by a grant from the Technion 2000 Group, Los Angeles. California. t Reprint requests: Yehudith Naot, D.Sc., Department of Immunology, Faculty of Medicine, Technion, Bat-Galim, Post Office Box 9649, Haifa, Israel. Vol. 58, No.3, September 1992
Several studies were reported on identification of detergent-solubilized sperm antigens recognized by autoserum and isoserum antibodies from infertile individuals (1-16). Sperm antigens separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis technique (SDS-PAGE) were transferred to nitrocellulose membranes and exposed to reactions with sera containing antisperm antibodies. The reports by different groups of investigators revealed variable results on the molecular weights (MWs) of spermatozoal antigens that were recognized by serum antibodies. These differences in patterns and MW s of the identified sperm antigens that reacted with serum antibodies may have resulted from differences in the procedures used for extraction of sperm antigens and from differences in the serum Shai and Naot
Identification of human sperm antigens
593
samples used in each study. Furthermore, because in most of the reported studies the reactions of sperm antigens with control sera from fertile individuals were not reported, the clinical relevance ofthe identified sperm antigens in infertility remained unclear. A unique sperm specific fertilization antigen-1 (FA-I) was recently characterized by Naz (17) and Naz et al. (18). This antigen present on human and murine male germ cells reacted strongly with sera from infertile males and females and reacted only weakly with sera from control fertile individuals. It has been postulated by Naz (17) that because antibodies to the F A-I do not agglutinate or immobilize sperm cells, these antibodies block fertilization by affecting the interactions between sperm and zona. In a recent report, we have shown that although antibodies in sera and genital secretions from fertile individuals react with soluble sperm antigens, statistically significant differences in levels of antisperm antibodies exist between infertile and fertile males and females when determined by a reverse (antibody capture) enzyme-linked immunosorbent assay (ELISA) (19). Furthermore, it has been demonstrated that the presence of antibodies in sera was not necessarily reflected in individual's genital tract secretion and vice versa. It was therefore suggested that for correct evaluation of sperm immunity antisperm antibodies should be determined not only in sera but also in genital tract secretions. In the present study, we have attempted to identify those sperm specific antigens that are recognized by antibodies present in sera and in genital tract secretions from infertile individuals in comparison with the antigens reacting with antibodies present m samples from fertile males and females. MATERIALS AND METHODS Patients and Samples
Fifty serum, 49 seminal plasma, and 42 cervical mucus (CM) samples were obtained from 100 consecutive infertile males and females, defined as suffering from unexplained infertility, and referred for determinations of antisperm antibodies. The patients, followed in an outpatient setting of a hospital clinic, had a history of infertility of at least 2 years' duration with no indication for hormonal or physical causes for their infertility. Male patients showed no significant abnormalities in semen analysis, exhibiting at least 20 X 106 sperm cell/mL, 2 mL of volume, 40% progressive motility, 50% normal morphology, and were found negative for the presence 594
Shai and Naot
Identification of human sperm antigens
of Mycoplasma hominis and Ureaplasma urealyticum. Female patients were tested for hormonal concentrations, basal body temperature, and underwent hysterosalpinogography, laparoscopy, and ultrasound monitoring of follicular development. Antisperm antibodies were tested in sera and genital tract secretions using the reverse ELISA as previously described (19). The results showed similar distribution of positive and negative samples as our previously reported data. Of 50 sera tested, 8 (32%) females and 7 (28%) males exhibited serum antisperm antibodies. Fourteen (28.5%) CM specimens and 11 (26.2%) seminal plasma samples were positive for antisperm antibodies as determined by the reverse ELISA. Control samples, 50 sera, 20 seminal plasma, and 9 CM samples were obtained from fifty hospital and faculty employees with recent proven fertility of the same age group (mean ± SD, 29.1 ± 2.8; range, 22 to 41 years) as the infertile individuals. Cervical mucus samples were collected 48 to 72 hours before ovulation after at least 48 hours of sexual abstinence. Treatment and solubilization of CM samples with bromelin (B-2252; Sigma, St. Louis, MO) were performed as previously described (19). Seminal plasma samples were separated from sperm cells by centrifugation after a liquefaction period of 30 minutes at room temperature. Handling of samples and treatment with bromelin were performed as previously (19). Preparation of Sperm Antigens
Specimens were obtained from fertile volunteers with normal semen quality. The samples used for isolation of sperm antigens had at least 20 X 106 sperm cell/mL, 2 mL of volume, 40% progressive motility, 50% normal morphology, and were found negative for the presence of M. hominis and U. urealyticum. After liquefaction, semen samples were centrifuged at 1,000 X g for 10 minutes and seminal plasma specimens separated. Sperm cells were resuspended in Dulbecco modified phosphate-buffered saline (D-5773; Sigma) containing 3 mM phenylmethyl sulphonyl fluoride (Sigma) as a protease inhibitor. Sperm cells were washed two times by centrifugation at 1,000 X g for 10 minutes in the above buffer. Washed cells from at least 20 individuals were pooled, suspended at a concentration of about 200 X 106 cells/mL, and subjected to at least 20 cycles of freezing and thawing in liquid nitrogen. The suspension was sonicated for 3 minutes at 4°C at 12 JLm (150-watt ultrasonic disintegrator; MSE Crawley, United Kingdom). Unbroken sperm cells were Fertility and Sterility
Table 1
Results of Immunoblotting Reactions Between Sperm Antigens and Serum Samples From Infertile and Fertile Individuals No. of individuals with antibodies reacting with the specified sperm antigens:j:
P*/Tt Sera obtained from Infertile females Fertile females Infertile males Fertile males
21/25 19/25 16/25 18/25
(84) § (76) (64) (72)
~116
o 4
o o
110 95
90
87
82
72
67
62
54
50
45
44
33
26
24
19
15
14
~12
12 8 10 13
1 3
5 5 6 5
4
5 0 5 1
3 2 2 1
1 0 0 0
2 3 1 0
6 6 2 0
5 4 1 1
8 2 3 1
2 0 1 0
2 3 1 0
2 1 2 2
4 3 3 1
3 6 1 3
0 7 3 6
15 7 11 13
3 2 1 3
o 3
1 3 2
* Number of individuals with antibodies reacting with at least one sperm antigen. t Total number of individuals tested in a given group.
removed by centrifugation at 1,000 X g for 10 minutes. The supernatants collected that contained approximately 2.5 mg protein/mL were divided into 0.5 ml aliquots and stored at -70°C. Gel Electrophoresis and Immunoblotting
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed according to the method of Laemmli (20). Stacking gels composed of 4% acrylamide and resolving gels containing 10% acrylamide were run in a vertical slab gel unit (Hoefer Scientific Instruments, San Francisco, CA). Sperm antigens were suspended in TRIS-HCl (62.5 mM) buffer (pH 6.8) containing 2% SDS, 10% glycerol, 5% iJ-mercapto-ethanol, and 0.1% bromophenol blue and immersed for 4 minutes in boiling water. Gels were loaded with 100 jlg/well of sperm extracts and were run for 18 hours at a constant voltage of 20 volts. Molecular weight standards at the range of 14 to 205 kd MW (SDS-7 and SDS6H; Sigma) were run in parallel. After electrophoresis, Western blotting techniques as described by Towbin et al. (21) were performed. Briefly, sperm antigens were transferred onto nitrocellulose sheets (45 jlM pore size; Tamar Ltd., Jerusalem, Israel) by electrophoretic blotting for 1 hour at 1 amp and additional hour at 1.5 amp. The transfer buffer contained 10 mM Na2C03 pH 9.9 (22). For enzymelinked immunodetection of sperm antigens on nitrocellulose' membranes were first incubated in blotting buffer containing 10 mM TRIS-HCl, 0.15 M NaCl, 1 mM ethylenediaminetetraacetic acid, 0.1 % Tween-20, 0.04% NaN 3, and 3% (wt/vol) nonfat dry milk at pH 7.6, to block unsaturated sites, overnight at 4°C. Sera, CM, and seminal plasma samples were diluted in blotting buffer to dilutions of 1:100; 1:40, and 1:8, respectively, and incubated with antigens on strips for 18 hours at room temperature with Vol. 58, No.3, September 1992
:j: Relative MWs in kd. § Values in parentheses are percents.
constant agitation. Strips were washed four times for 10 minutes each wash, with blotting buffer. Alkaline phosphatase-conjugated, affinity-purified, rabbit immunoglobulins (Ig) to human IgM, IgG, and IgA (Dako Patts A190, Copenhagen, Denmark) diluted in blotting buffer were added at a concentration of 2.5 jlg/mL; and incubation was carried out for 1 hour at 37°C. Strips were washed again four times and subjected to substrate for alkaline phosphatase (5-bromo-4-chloro 3 indolyl phosphate, B-6149; Sigma) for 2 hours at 37°C. Statistical analysis of data was performed using Yate's corrected X2 test for independence.
RESULTS
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of solubilized spermatozoal antigens revealed approximately 20 major protein bands with relative MWs ranging from 12 to 116 kd. A reaction was considered positive in the enzyme-linked immunoblotting reactions, regardless of the intensity of the stained bands of antigen-antibody complexes on nitrocellulose membranes. Table 1 summarizes the distribution of positive reactions obtained by immunoblotting between separated sperm antigens of different MWs and antisperm antibodies present in sera from fertile and infertile individuals. It can be seen from Table 1 that most of the sera tested exhibited antibodies to at least one sperm antigen whether obtained from infertile or from fertile donors. Statistical analysis of the data using Yate's corrected X2 test for independence showed that antisperm antibodies present in sera from infertile females reacted with two bands of sperm antigens with relative MWs of 44 and 72 kd significantly more frequently than antisperm antibodies in sera from fertile control females (P < 0.1). Unlike these differences between infertile Shai and Naot
Identification of human sperm antigens
595
and fertile females, sera from fertile and infertile males revealed statistically similar distributions of reactions with separated sperm antigens. Based on these results and on our previously reported data (19) that showed that antisperm antibodies present in sera are not necessarily reflected in individual's genital tract secretions and vice versa, we were further motivated to examine the reactions of sperm antigens with antibodies present in genital tract secretions. Electrophoretically separated sperm antigens were, therefore, exposed to reactions with solubilized samples of CM and seminal plasma from infertile and fertile human subjects. The results of immunoblotting assays are summarized in Table 2. Antisperm antibodies present in CM samples from infertile females reacted preferentially with sperm antigens exhibiting relative MWs of 62 kd and 87 kd, as compared with fewer reactions with other sperm antigens. However, comparison of the data obtained with samples from infertile women with those obtained with fertile controls showed statistically insignificant differences probably because of the limited number of samples available for testing from the group of fertile females. Seminal plasma samples from infertile men contained antisperm antibodies that reacted significantly more frequently (P < 0.1) with sperm antigens exhibiting MW of 62 kd than reactions of antibodies present in seminal plasmas from fertile males. Interestingly, these 62 kd sperm antigens rarely reacted with antibodies present in sera from either infertile or fertile males and females (Table 1). Considered together, the immunoblotting test results revealed differences in the antigens that react with serum antibodies and those reacting with antibodies present in genital tract secretions.
DISCUSSION
This study was carried out in an attempt to identify spermatozoal antigens recognized by antisperm antibodies present in sera and in genital tract secretions from infertile men and women in comparison with the sperm antigens reacting with antibodies from fertile males and females. Most of the studies carried out in the past used immunoblotting techniques as an experimental tool (1-16) and revealed variable results on the MW of spermatozoal antigens recognized by serum antisperm antibodies (3-6,8). Furthermore, because in most studies sera from fertile controls were not tested, the clinical significance of the identified sperm polypeptides remained obscure. More extensive immunoblotting assays with different groups of sera from fertile and infertile individuals of both sexes, performed previously (8, 9, 11) and in this study, clearly demonstrated that a high proportion of sera from all clinical categories contained different levels of antibodies reacting with at least one spermatozoal antigen. This observation, also revealed in immunofluorescence and ELISAs, which are similar in their basic methodology (19, 23,24), led to the suggestion (9) that antisperm antibodies in fertile subjects may reflect a natural physiological phenomenon. To avoid these background physiological reactions, observed when sera were diluted to 1:100, Lehmann et al. (9) also tested sera at higher dilutions of 1:2,000 and have found that antibodies to a sperm polypeptide with 14 kd MW are detected in a significantly higher proportion among males with unexplained infertility. This 14 kd polypeptide could not be observed in our study because we have tested sera from males at 1:100 dilution.
Table 2 Results of Immunoblotting Reactions Between Sperm Antigens and eM and Seminal Plasma Samples From Infertile and Fertile Individuals No. of individuals with antibodies reacting with the specified sperm antigens:j:
P*/Tt Genital tract secretions obtained from Infertile females Fertile females Infertile males Fertile males
21/42 1/9 19/49 2/20
(50) § (11) (38) (10)
~116
o o o o
110
95
90
87
82
72
67
62
54
50
45
44
33
26
24
19
15
14
~12
2
o o
o
1
o o
3
1 2
1 1
12 1 2 2
2 0 1 0
0 0 0 0
7 1 11 0
0 0 0 0
3 0 6 0
3 1 0 0
1 0 3 0
0 0 0 0
1 0 2 0
3 0 0 0
1 0 0 0
3 0 0 0
2 0 0 0
0 0 0 0
o o
1 1
* Number of individuals with antibodies reacting with at least one sperm antigen. t Total number of individuals tested in a given group. 596
Shai and Naot
Identification of human sperm ontigens
:j: Relative MW in kd. § Values in parentheses are percents.
Fertility and Sterility
Autoantibodies against various sperm antigens were also shown by Hald et al. (11) in all sera obtained from a group of 16 males before and after vasectomy. These investigators reported that although two postoperative patients developed serum agglutinating antibodies, there was no change in the pattern of their immunoblotting reactions. On the other hand, another patient without sperm agglutinins showed increased serum activity with a 31.5 kdpolypeptide. Other investigators (10, 12) extended these observations and demonstrated that although antisperm antibodies bind to discrete sperm-associated antigens, there is no correlation between the pattern of reactions of sera with separated discrete sperm antigens and the activity of sera in producing different modes of agglutination, in complement dependent immobilization of sperm cells and in stimulation or suppression of sperm-oocyte fusion. This lack of correlation between these biological effects of antibodies on viable sperm cells functions and between the reactions of antibodies with detergentsolubilized sperm antigens may reflect a true phenomenon. However, it may also be a result of the denaturing effects of immunoblotting procedures that denature native antigens and possibly expose new antigens. This limitation of SDS-PAGE procedures was overcome by Saji et al. (13) who first used isoantiserum to isolate a 15 kd glycopeptide from sperm cell membranes and then showed that this antigen reacted with sera from infertile females possessing sperm-immobilizing antibodies, whereas sera from fertile females and males did not bind to this sperm glycopeptide. Saji et al. (13) suggested that this spermatozoal antigen is one of the corresponding antigens to sperm-immobilizing antibodies. The isoantigen described by Saji et al. (13) differs from the FA-1 isolated by Naz et al. (17, 18) not only in their relative MWs but also in their biological significance, inasmuch as antibodies produced against the FA-1 were demonstrated to be nonreactive in sperm agglutination and immobilization tests (17, 18). In SDS-PAGE experiments, isolated FA-1 exhibited MW s of 47 to 50 kd, which may correspond to one of the two discrete sperm polypeptides with relative MW s of 44 kd and 72 kd demonstrated in this study to react significantly more often with serum antibodies from infertile females than with serum antibodies from fertile females or from both infertile and fertile males. Unlike previous reports, this study differentiated between sperm antigens reactive with serum antibodies and those antigens that are recognized by Vol. 58, No.3, September 1992
local antibodies. Our data demonstrated that although serum antibodies react with antigenic determinants on 44 and 72 kd sperm antigens, local antisperm antibodies from infertile females recognize epitopes present on the 62 and 87 kd isoantigens of normal sperm cells. The dichotomy between local and systemic humoral responses, reflected by the differences in specificities of systemic and local antisperm antibodies, was exhibited not only in infertile females but also in infertile males. A sperm autoantigen of 62 kd MW was identified as a major sperm antigen reacting with antisperm antibodies present in seminal plasma samples obtained from infertile males. This particular sperm antigen with 62 kd MW reacted only rarely with genital secretions from fertile subjects or with serum antibodies, whether obtained from females or from males and regardless of their reproductive potential. It should be mentioned that recently Mathur and associates (14) suggested that sperm cells from infertile males contain, in addition to common antigens present on normal cells, special antigens recognized by autoimmune sera. This novel proposal may explain the inability to demonstrate statistically significant differences in reactions of sera from infertile and fertile males with normal sperm antigen. Whether the proposed special antigens observed on sperm from autoimmune infertile men are also triggering antibodies in genital tract secretions of both sexes remains to be examined. REFERENCES 1. Young LG, Goodman SA. Characterization of human sperm cell surface components. BioI Reprod 1980;23:826-35. 2. Hjort T, Poulsen F. Analysis of auto-antigens in the human sperm membrane by a F(ab)2 blocking system. J Clin Lab ImmunoI1981;6:61-7. 3. Poulsen F, Hjort T. Identification of auto-antigens of the human sperm membrane. J Clin Lab ImmunoI1981;6:69-74. 4. Poulsen F. The nature of an iso-antigen of the human sperm membrane. J Reprod Immunol 1983;5:49-54. 5. Lee C-Y G, Huang Y oS, Hu P-C, Gomel V, Menge AC. Analysis of sperm antigens by sodium dodecyl sulfate gel protein blot radioimmunobinding method. Anal Biochem 1982;123: 14-22. 6. Lee Coy G, Lum V, Wong E, Menge AC, Huang Y-S. Identification of human sperm antigens to antisperm antibodies. Am J Reprod ImmunoI1983;3:183-7. 7. Lee Coy G, Wong E, Richter DE, Menge AC. Monoclonal antibodies to human sperm antigens-II. J Reprod Immunol 1984;6:227 -8. 8. Naaby-Hansen S, Bjerrum OJ. Auto and iso antigens of human spermatozoa detected by immunoblotting with human sera after SDS-PAGE. J Reprod ImmunoI1985;7:41-57. 9. Lehmann D, Temminck B, Rugna DDA, Leibundgut B, Muller H. Blot immunobinding test for the detection of anti sperm antibodies. J Reprod Immunol 1985;8:329-36.
Shai and Naot
Identification of human sperm antigens
597
10. Aitken RJ, Hulme MJ, Henderson CJ, Hargreare TB, Ross A. Analysis of the surface labelling characteristics of human spermatozoa and the interaction with antisperm antibodies. J Reprod FertiI1987;80:473-85. 11. Hald J, Naaby-Hansen S, Egense J, Hjort T, Bjerrum OJ. Autoantibodies against spermatozal antigens detected by immunoblotting and agglutination. A longitudinal study of vasectomized males. J Reprod Immunol 1987;10:15-26. 12. Parslow JMM, Poulton TA, Hay FC. Characterization of sperm antigens reacting with human antisperm antibodies. Clin Exp ImmunoI1987;69:179-87. 13. Saji F, Ohashi K, Kamiura S, Negoro T, Tanizawa O. Identification and characterization of a human sperm antigen corresponding to sperm immunobilizing antibodies. Am J Reprod Immunol Microbiol 1988;17:128-33. 14. Mathur S, Chao L, Goust JM, Milroy GT, Woodley-Miller C, Caldwell JZ. Special antigens on sperm from autoimmune infertile men. Am J Reprod Immunol 1988;17:5-13. 15. Howe SE, Grider SL, Lynch DM, Fink LM. Antisperm antibody binding to human acrosin: a study of patients with unexplained infertility. Fertil Steril 1991;55:1176-82. 16. Cunningham DS, Fulgham DL, Rayl DL, Hansen KA, Alexander NJ. Antisperm antibodies to sperm surface antigens in women with genital tract infection. Am J Obstet Gynecol 1991;164:791-6. 17. N az RK. The fertilization antigen (FA -1): applications in
598
Shai and Naot
Identification of human sperm antigens
18.
19.
20. 21.
22.
23.
24.
immunocontraception and infertility in humans. Am J Reprod Immunol MicrobioI1988;16:21-7. Naz RK, Phillips TM, Rosenblum BB. Characterization of the fertilization antigen 1 for the development of a contraceptive vaccine. Proc Natl Acad Sci USA 1986;83:5713-7. Shai S, Bar-Yoseph N, Peer E, Naot Y. A reverse (antibody capture) enzyme linked immunosorbent assay for detection of antisperm antibodies in sera and genital tract secretions. Fertil Steril 1990;54:894-901. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T 4 • Nature 1970;227:680-5. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350-4. Dunn SD. Effects of the modification of transfer buffer composition and the renaturation of proteins in gels on the recognition of proteins on western blots by monoclonal antibodies. Anal Biochem 1986;157:144-53. Hjort T, Poulsen F. Reactivity of F(ab')2 fragments of sperm antibodies in agglutination immobilization and immunofluorescent test. Arch Androl 1978;1:83-9. Mettler L, Czuppon AB, Alexander N, D'Alnedia M, Hass GG Jr, Hjort T, et al. Antibodies to spermatozoa and seminal plasma antigens detected by various enzyme-linked immunosorbent ELISA assays. J Reprod ImmunoI1985;8:301-12.
Fertility and Sterility
