Antibody binding patterns in infertile males and females as detected by immunobead test, gel-agglutination test, and sperm immobilization test

Transcription

1 FERTILITY AND STERILITY Copyright The American Fertility Society Printed in U.S.A. Antibody binding patterns in infertile males and females as detected by immunobead test, gel-agglutination test, and sperm immobilization test Sandra Ann Carson, M.D.* Jill Reiher, B.S.t Antonio Scommegna, M.D. t Gail S. Prins, Ph.D. t University of Tennessee, Memphis, Memphis, Tennessee, and Michael Reese Hospital and Medical Center, University of Chicago, Chicago, Illinois Sera from 214 infertile patients were assayed for antisperm antibodies using the IBT, GAT, and SIT. The new IBT methodology was compared with the more classical tests. Although SIT and GAT did not correlate to any particular antibody class, both were negative if IgA was present alone. The immunoglobulin class presented a preferential sperm region binding site: IgG to the head and tail, IgM to tail tip only, IgA to head and tail. Furthermore, these immunoglobulins from the serum of male patients bound differently to sperm than immunoglobulins from female serum. Finally, we were able to calculate the relative sensitivity, specificity, and predictive values for these tests. Fertil Steril 49:487, 1988 The presence of antisperm antibodies has been associated with a reduction in the fertility potential of a large number of individuals. 1 However, because of the complexity of the human immune system and the relative nature of immunoinfertility, proper diagnosis of this factor can be difficult. Antibodies directed against spermatozoa are a heterogeneous population, varying with respect to immunoglobulin class, antigenic specificity, region of binding, and dependency upon the complement system. The two classical methods used to detect these immunoglobulins in serum are based upon the ability of sperm-specific antibodies to induce Received September 9, 1987; revised and accepted October 27, * Reprint requests: Sandra Ann Carson, M.D., Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, University of Tennessee, Memphis, 956 Court Avenue, Room D324, Memphis, Tennessee t Department of Obstetrics and Gynecology, University of Chicago. agglutination or complement-dependent immobilization reactions in donor spermatozoa. Because these two bioassays differ in their mechanism of interaction with sperm, it is not surprising that they frequently produce discordant results. To maximize detection in the clinical laboratory, we routinely screen patient sera using both the gel-agglutination test (GAT) 2 and the sperm immobilization test (SIT). 3 Recently, a new methodology, called the immunobead test (IBT), was introduced. 4 6 This assay employs micron-sized polyacrylamide spheres covalently attached to rabbit anti-human immunoglobulins. When these immunobeads are mixed with antibody-bound sperm, they adhere to the antibody and this attachment can be visualized microscopically. In addition to identifying the specific immunoglobulin present (IgG, IgA, and/or IgM), the IBT demarcates the region of the sperm cell to which antibodies are directed. The latter may aid prognostically or with choice of appropriate therapy. Carson et al. Antisperm antibodies 487

2 In order for the IBT to become a widely accepted screening tool for antisperm antibodies, it is important that diagnostic correlations are made between this test and several other well-established methodologies. We have now employed the IBT using an indirect method in conjunction with the SIT and GAT on over 200 sera from patients who presented to our fertility clinic with a possible immune cause of infertility. In the present report, we provide the correlation of the indirect IBT with these two classical assays in terms of antibody sensitivity, incidence for each immunoglobulin class, and region of sperm binding. In addition, we analyzed the results for possible sex differences and have observed dissimilar antibody binding patterns in male and female sera. MATERIALS AND METHODS Sera was obtained from 214 infertile patients visiting a fertility clinic who were suspected of an immunologic factor due to one of the following indications: poor postcoital test, spontaneous agglutination of Sl)erm in seminal plasma, vasectomy reversal, or chronic genital tract infections. This population consisted of 95 women and 119 men. The sera was heat-inactivated at 56 C for 30 minutes, aliquoted, frozen, and stored at -35 C until assay. Figure 1 the IBT_ IMMUNOBEAD TEST: PASSIVE ANTIBODY TRANSFER TECHNIQUE SEMEN FROM ANTIBODY FREE FERTILE DONOR WASH SPERM 2x IN TYRODES-BSA "SWIM UP" MOTILE SPERM 50 x 106 MOTILE SPERM/ml fdh u \ ~ HEAT INACTIVATE SERA 30 min at 56DC -PATIENT SERUM - ANTJSPERM ANTIBODY POSITIVE AND NEGATIVE CONTROLS DILUTE I :4 TYRODES-BSA / u f0'\ \ /"" ' l]}/\cj INCUBATE 30min at 37'C -- >, ~~ WASH SPERM FREE OF SERUM l AN;p/1-lgG B~ADS ~ANTHgM BEADS... tomln ANTI-IgA BEADS -SCORE 100 MOTILE SPERM (DUPLICATE) 8NO BINDING,@ 2:1 BEAD/SPERM -NOTE LOCATION OF(±)BINDING -.!:.20" BOUND IS A POSITIVE IBT Schematic diagram depicting the principles used in Indirect Immunobead Test Antisperm antibodies present in serum were absorbed onto antibody-free donor sperm using a passive antibody transfer technique, as shown in Figure 1. Fertile sperm donors from our donor insemination program were used as a routine source for antibody-free spermatozoa. Care was taken to maintain sterile conditions for the spermatozoa during the entire assay. Following liquefaction, semen samples were diluted in 3 volumes of Tyrode's buffer containing 0.3% bovine serum albumin (0.3% T-BSA), and washed twice by centrifugation at 325 X g for 10 minutes. The resulting pellet was gently loosened and 0.5 ml 0.3% T-BSA was carefully layered onto the top. The tube was placed upright in a 37 C incubator for 30 minutes to allow the highly motile sperm fraction to swim up into the medium. After that time, the top 0.4 ml was transferred to a new tube and the sperm count and motility were determined and adjusted to 50 X 10 6 motile sperm/mi. Heat-inactivated serum was thawed and diluted 1:4 in Tyrode's buffer containing 3% BSA (3% T-BSA) at 37 C. Known antibody-positive and -negative sera were run as controls in every assay. Sperm suspensions were mixed with 400 f.lg diluted serum (4.5 X 10 6 motile sperm/ ml final concentration) and allowed to incubate at 37 C for 30 minutes, at which time the sperm were washed free of serum by centrifugation for 5 minutes at 325 X g, resuspended in 0.5 ml3% T-BSA, and centrifugation was repeated. The final pellet was resuspended in 100 f.ll 3% T-BSA (10 X 10 6 motile sperm/ml) and these sperm were used in the immunobead assay. Rabbit antihuman IgG, IgA, and IgM immunobeads (50 mg, Bio Rad, Richmond, CA) were resuspended in 10 ml Tyrode's buffer under aseptic conditions and stored at 4 oc for up to 3 months. The day of assay, 0.2 ml (1 mg) was aseptically removed, washed once in 3% T-BSA, and resuspended to 2 mg/ml for use in the assay. Ten microliters of the sperm suspension and 20 f.ll of the bead suspension were mixed on a glass slide, covered with a coverslip, and allowed to react for 10 minutes at room temperature. Using a phase-contrast microscope (400X), 100 motile spermatozoa were scored for the presence and location of bead binding. Binding was considered positive when one or more beads adhered to the surface of a motile sperm cell. Location was demarcated as head only (H), tail only, head and tail (HIT), and tail tip (tt). Duplicate counts were made 488 Carson et al. Antisperm antibodies Fertility and Sterility

3 for each of the three separate bead suspensions for each serum tested. A specimen was considered IBT-positive when ~20% of the motile sperm showed positive bead binding for IgG, IgA, and/ or IgM. Sperm Immobilization Test The presence of cytotoxic antisperm antibodies in serum was determined using a complement-dependent SIT developed by Isojima et al. in Briefly, complement-inactivated sera from patients and known positive and negative controls were diluted 1:4 in saline. Fresh sperm from an antibody-negative fertile donor were washed twice and resuspended in saline to 60 X 10 6 sperm/ml. Guinea-pig sera served as a source of complement for the reaction. Patient or control serum, donor sperm, and guinea-pig serum were incubated at room temperature for 1 hour, after which the percentage of motile sperm in each sample was calculated. Evaluation was made by determining the ratio of the percentage of motile cells in control serum to the percentage in the patient serum. A value of ~2.0 was considered positive. Gelatin Agglutination Test The presence of agglutinating antisperm antibodies was determined using a macroscopic technique described by Kibrick et al. 2 Briefly, 60 X 10 6 motile sperm from a fertile donor were mixed with serial dilutions of patient and positive and negative control serum and added to small tubes containing 5% gelatin medium with Baker's buffer as a diluent. Following a 30-minute incubation at 37 C, the tubes were examined for the presence of agglutination within the gelatin. A specimen was considered positive if flocculation was present at a serum dilution of 1:4 or greater. Statistical Analysis Data was analyzed with chi-square analysis using the Yates correction, where appropriate. RESULTS Correlation of Immunobead Test with Agglutination and Immobilization Tests A total of 214 sera were tested for the presence of antisperm antibodies using the GAT, SIT, and IBT methods. In this select group, 83 (40%) were negative and 39 (18%) were positive for all three assays. In the remaining samples, there was a divergence of results, i.e., being positive in at least one test and negative in another. Table 1 shows the correlation of GAT and SIT results with the IBT. Overall, 53 sera were positive for SIT and, of these, 4 7 sera (89%) were IBT-positive. Similarly, 75 specimens were GAT-positive and 65 of these (87%) were IBT-positive. If both GAT and SIT were positive, concordant results were obtained by IBT 98% of the time, including cases in which GAT titers were 1:4. However, when the SIT and GAT were both negative, the incidence of positive results by IBT was 34%. This finding may be due to false-negative SIT and GAT results, or to questionable significance of some positive IBT binding. To further address the latter issue, we redefined our IBT-positive sera as those producing >50% sperm bound with immunobeads of at least one class. These results are presented in Table 2. With this new criteria, only 10% of SIT and GAT negative sera exhibited positive IBT binding. Thus, in terms of correlations with other clinical assays, perhaps 50% IBT binding should be considered significant. When the incidence of IBT-positive binding was broken down for the three immunoglobulin classes and separated according to GAT and SIT results, some interesting observations were noted (Table 1). First, if SIT and GAT were both positive, IgG was always present; however, IgG binding alone occurred with only 5% frequency, since the overall IgA incidence also was high. Second, while IgA occurred with a frequency of 62%, it never occurred alone when either GAT or SIT were positive. In contrast, IgA binding alone accounted for approxi- Table 1 Correlation of GAT and SIT with Immunobead Test (IBT)a and Class of Immunoglobulin with IBT Positive as >20% Binding GAT SIT n IBT(-) 1 (2%) 5 (38%) 9 (26%) 83 (66%) a GAT and SIT results versus IBT, x 2 = 56.3; P < b Parentheses in immunoglobulin columns denote number of IBT() 39 (98%) 8 (62%) 26 (74%) 43 (34%) 39 (2) 7 (2) 20 (3) 24 (11) 35 (0) 4 (0) 18 (O) 15 (4) 13 (O) 1 (1) 12 (3) 24 (15) sera that possessed positive immunobead binding for only that particular class. Carson et al. Antisperm antibodies 489

4 Table 2 Correlation of GAT and SIT with Immunobead Test (IBT)a and Class of Immunoglobulin with IBT Positive as >50% Binding GAT SIT n IBT(-) IBT() IgN (8%) 6 (46%) 16 (46%) 113 (90%) 37 (92%) 7 (54%) 19 (54%) 13 (10%) 37 (5) 7 (3) 16 (3) 11 (7) 30 (0) 4 (O) 11 (O) 4 (1) 7 (O) 2 (0) 6 (3) 3 (0) a GAT and SIT results versus IBT, X 2 = 95.9; P < b Parentheses in immunoglobulin columns denote number of sera that possessed positive immunobead binding for only that particular class. mately 10% of the sera that were SIT- and GATnegative, yet IBT-positive. This is not surprising since SIT is not sensitive to lga antibodies and GAT shows only low positive titers when lga is present alone. 6 Third, in specimens with SIT- and GAT-negative and IBT-positive results, the incidence of detecting only one immunoglobulin class was 70%. Fifteen of 43 (35%) ofthese samples were positive for lgm alone. Regional Binding Distribution in Agglutinationand Immobilization-Positive Sera The location of antibody binding to donor spermatozoa could be distinguished as head, tail, head and tail, or tail tip with IBT. Attachment to the tail tip occurred either alone or in addition to binding in other regions. This distribution was analyzed for the three immunoglobulin classes in GAT -positive and in SIT-positive sera to determine first, which region of binding was responsible for agglutination and immobilization reactions and second, if different antibody class and distribution patterns existed for these two assays. In GAT-positive specimens (Table 3), the incidence of head only or tail only binding was very low. For lgg, the majority of antibody binding was located on the entire sperm (head and tail), with a minor amount bound to the tail tip. For lga, approximately half of the binding was directed to the head and tail, while a high incidence of tail tip binding was noted ( 73% total). The majority of lgm binding was directed to the tail tip alone. Unexpectedly, a similar binding pattern emerged for SIT-positive sera. lgg was primarily observed on the entire sperm (head and tail), with a low incidence of head only, tail only, and tail tip binding. Half of the lga binding was scattered over the entire surface, with the rest being distributed on the tail tip or head alone. Again, lgm was oriented to the tail tip. In addition to the distribution similarity, the class frequency (lgg, lga, or lgm) was not different between GAT- or SIT-positive sera. Regional Binding Distribution in Male and Female Sera Sera positive for antisperm antibodies by the IBT were divided according to sex and analyzed for class frequency and regional distribution patterns, and these results are shown in Table 4. In male sera, the incidence of head only or tail only binding was very low for the three classes of immunoglobulins. The majority of IgG binding was directed to the entire sperm surface, with a minor incidence to the tail tip. For lga, 48% of the binding was on the Table 3 Regional Binding Distribution in GAT and SIT Positive Sera 70 GAT positive and IBT positive sera 52 SIT positive and IBT positive sera IgGa IgAa n = 64 n =59 Head only 4 (6) 6 (10) Tail only 2 (3) 1 (2) Head and tail 51 (80) 27 (46) Tail tip only 7 (11) 21 (36) Tail tip totalb 16 (25) 43 (73) a Parentheses denote percentage of that class. b Tail tip attachment of immunobeads occurred either alone or in addition to binding on the head and tail. The tail tip total IgMa IgGa IgAa IgMa n = 21 n = 49 n = 45 n = 25 3 (14) 6 (12) 9 (20) 1 (4) 2 (10) (12) 0 41 (84) 23 (51) 0 16 (76) 2 (4) 11 (24) 21 (84) 16 (76) 6 (12) 27 (60) 21 (84) represents the total incidence of tail tip binding, whereas the tail tip row denotes the incidence of binding to the tail tip alone. 490 Carson et al. Antisperm antibodies Fertility and Sterility

5 Table 4 Regional Binding Distribution in IBT Positive Sera from Males and Females Male sera (n = 119) IgG lga" n = 67 n =59 Head only 4 (6) 3 (5) Tail only 3 (5) 2 (3) Head and tail 50 (75) 28 (48) Tail tip only 10 (14) 26 (44) Tail tip totalb 14 (21) 47 (80) Parentheses denote percentage of that class. b Tail tip attachment of immunobeads occurred either alone or in addition to binding on the head and tail. The tail tip total Female sera (n = 95) lgm lgg lga lgm n = 33 n = 26 n = 20 n = (65) 15 (75) 4 (17) 2 (6) (23) 2 (10) 0 31 (94) 3 (12) 3 (15) 19 (83) 31 (94) 3 (12) 3 (15) 19 (83) represents the total incidence of tail tip binding, whereas the tail tip row denotes the incidence of binding to the tail tip alone. entire sperm, 44% was at the tail tip exclusively, and the overall tail tip incidence was 80%. IgM binding was almost always observed on the tail tip. A significantly different distribution pattern was found for antisperm antibodies in females. In their sera, the majority of IgG and IgA binding was observed on the head exclusively: 65% for IgG and 75% for IgA. There was no incidence of tail-only binding in female serum. Similar to what was noted in male sera, female serum IgM was primarily located on the tail tip. DISCUSSION The IBT was introduced by Bronson et al. 4 as a specific test for antisperm antibodies. These investigators identified and validated the test in the serum of infertile couples. In addition, they found that antibodies bound to the sperm altered the sperm's ability to penetrate a zona-free hamster ovum. 7 Clarke et al. 5 and Shulman et al. 6 evaluated the nature of the different classes of immunoglobulin covering the sperm cells. These investigators found IgG and IgA more commonly than IgM. The latter article 6 also showed that IgG bound predominantly to the head or to both head and tail, IgA predominantly to the tail, and IgM was not described. Our present findings show somewhat similar results for these immunoglobulins. This study further adds to the descriptive information about immunobead testing. We investigated the comparability of IBT with both GAT and SIT and found that, when both GAT and SIT were positive, the IBT was positive in 98% of samples. If both tests were negative, IBT was negative 66% of the time. It is difficult, however, to determine whether it is the GAT and SIT that are producing the false-negative results, or the IBT that is binding to areas other than antibodies and producing a false-positive result in the 34% of cases that are negative in the GAT and SIT. Assuming, indeed, that GAT and SIT were the "gold standard" for immunoglobulin testing, the IBT would then have a sensitivity of 97.5% and a specificity of 66%, with a positive predictive value of 4 7.5%, according to our results. Conversely, if the situation is reversed and IBT is actually the true test, then GAT and SIT have a sensitivity of 47.5% and a specificity of 99%, with a positive predictive value of 97.5%. Mathematical manipulations, however, will notreplace collection of clinical data comparing fertility rates with IBT results. It has already been shown that clinical evaluation of GAT and SIT reveals only a slight correlation with pregnancy and absence of antisperm antibodies detected by these methods. This correlation is so little, in fact, that the risks of treatment outweigh benefits. This may not be the case for a test that is better correlated with reproductive failure. These studies reveal that GAT and SIT usually need more than one antibody class present to be positive. Out of the 43 cases in which these tests were negative and IBT was positive, 30 occurred because of the presence of one single class of immunoglobulin. This may be a result of simply a higher total antibody titer, or may reflect a need of the bioassay to require some type of synergy between the antibody classes. Indeed, when IgA was present alone, both GAT and SIT were negative, further illustrating the need of complement fixation in the SIT. We looked at the immunoglobulin distribution along the sperm in those sera that were GAT-positive and those that were SIT-positive. We found no difference in binding within each class of immunoglobulin, whether they were detected by GAT, by SIT, or by both; nor did we find a difference between the distribution of immunoglobulin along Carson et al. Antisperm antibodies 491

6 the sperm detected by SIT or by GAT. However, there was a difference of regional binding between the immunoglobulin classes: IgG bound mostly to head and tail, lga to head and tail, and lgm to tail tip only. These regional binding variations may later prove important in clinically evaluating which patients need be treated. Preliminary results by Clarke et al. 5 reveal that immunoglobulins bound to the tail tip impede sperm migration through microcolumns of cervical mucus, but have little impact on fertilization at in vitro fertilization in the human. Conversely, lga bound heavily to the head and tail impairs the pregnancy rate at IVF according to these authors. 5 Our studies indicate that there are differences in the characteristics of the antibodies found in male versus female sera. While the three classes of immunoglobulins are distributed equally in females, males have a higher percentage of lgg and lga. In addition, the regional binding of those immunoglobulins differs. lgg and lga from male sera bind heavily to almost the entire sperm-head and tail. However, these antibodies from female sera bind mostly to the sperm head. lgm binds similarly in both male and female. Once again, these findings require clinical correlation, but the identification of different patterns in male than female may result in different treatment regimens and indications. In conclusion, this study has allowed us to calculate a sensitivity and specificity of the IBT as compared with well-established methodologies, GAT and SIT. It is now clear that, if only SIT or GAT is used for evaluation, a 12% false-positive and up to 40% false-negative result may occur, as defined by immunobead binding. Also, we have identified regional binding differences among the classes of immunoglobulins, including IgM, which seems to prefer the tail tip. Finally, male and female antisperm antibodies of the lgg and lga classes differ in their regional distribution on sperm: females possess a higher incidence of head-directed and males have a higher prevalence of antibodies directed at head, tail, and tail tip. REFERENCES 1. Haas GG, Jr: Immunologic infertility: which approaches are best? Contemp Obstet Gynecol 22:141, Kibrick S, Belding DL, Merrill B: Methods for the detection of antibodies against mammalian spermatozoa. II. A gelatin agglutination test. Fertil Steril 3:430, Isojima S, Li TS, Ashitaka Y: Immunologic analysis of sperm immobilizing factor found in sera of women with unexplained sterility. Am J Obstet Gynecol101:677, Bronson RA, Cooper GW, Rosenfeld DL: Correlation between regional specificity of antisperm antibodies to the spermatozoan surface and complement-mediated sperm immobilization. Am J Reprod Immunol 2:222, Clarke GN, Elliott PJ, Smaila C: Detection of sperm antibodies in semen using the immunobead test: a survey of 813 consecutive patients. Am J Reprod Immunol 7:118, Shulman S, Pretorius E, Keane T: Antibodies to spermatozoa. XI. The use of immunobeads for the detection of sperm antibodies in serum. Am J Reprod Immunol 9:62, Bronson R, Cooper G, Rosenfeld D: Ability of antibodybound human sperm to penetrate zona-free hamster ova in vitro. Fertil Steril 36:778, Carson et al. Antisperm antibodies Fertility and Sterility