Demonstration of antispermatozoal antibodies in varicocelerelated infertility with an enzyme-linked immunosorbent assay

Transcription

1 FERTILITY AND STERILITY Copyright 1986 The American Fertility Society Printed in U.8A. Demonstration of antispermatozoal antibodies in varicocelerelated infertility with an enzyme-linked immunosorbent assay (ELISA) Jacob Golomb, M.D.*t Nurit Vardinon, Ph.D.* Zvi T. Homonnai, M.D. Zvi Braf, M.D. * Israel Yust, M.D.* Ichilov Hospital and Serlin Maternity Hospital, Tel Aviv Medical Center, Tel Aviv University Sackler School of Medicine, Tel Aviv, Israel To assess the existence of a possible immunologic factor in varicocele-associated infertility. we searched for antispermatozoal antibodies in serum, seminal plasma, and bound to spermatozoa in 32 infertile men with varicocele and 22 infertile patients without palpable varicocele, with the use of an enzyme-linked immunosorbent assay. In addition, we performed morphologic and microbiologic analyses of the semen and urethral smears for isolation of Chlamydia trachomatis. Twenty-nine men from the varicocele group (90.6%) demonstrated antispermatozoal antibodies, compared with only 9 men (40.9%) in the control group. The antibodies in both groups, when present, were mainly serum and seminal plasma immunoglobulins IgA and IgM. A significant quantitative difference between the varicocele and control groups was also observed for serum IgA, seminal plasma IgA and IgM, and sperm-bound IgG, IgA, and IgM. Oligozoospermia and asthenozoospermia were significantly more prevalent in the varicocele men. An asymptomatic genital tract infection with C. trachomatis, Ureaplasma urealyticum, and Escherichia coli was traced in 40.6% of the varicocele men and in 45.5% of the control group. No interaction could be demonstrated between the infection and antispermatozoal antibody formation. These data suggest that an immunologic factor may playa role in varicocele-associated infertility; however, its impact on reproduction has yet to be assessed. Fertil Steril45:397, 1986 Received April 23, 1985; revised and accepted November 12, *Department of Urology, Ichilov Hospital, Tel Aviv Medical Center, Tel Aviv University Sackler School of Medicine. treprint requests: Jacob Golomb, M.D., Department of Urology, Tel Aviv Medical Center, Ichilov Hospital, 6 Weizmann Street, Tel Aviv 64239, Israel. *The Henry Goldman Memorial Clinical Immunology Laboratory, Ichilov Hospital, Tel Aviv Medical Center, Tel Aviv University Sackler School of Medicine. Institute for the Study of Fertility, Serlin Maternity Hospital, Tel Aviv Medical Center, Tel Aviv University Sackler School of Medicine. Varicocele-related infertility is well established. The testicular and semen aberrations seen in subfertile men with varicocele have already been described. l, 2 Nevertheless, little consensus exists concerning the pathophysiology and causes of this syndrome. Venous stasis with impaired testicular metabolism, local hyperthermia, and reflux of noxious adrenal metabolites cannot explain the bilateral impairment of spermatogenesis. 3 It is also well-known that, for some obscure reason, the conventional treatment of varicocele Golomb et ai. Antisperm antibodies in varicocele 397

2 by occlusion of the varicose spermatic vein does not always restore fertility.4 An immunologic mechanism has not been previously explored. If the blood-testis barrier is somehow damaged by the venous stasis and hyperthermia, then antispermatozoal antibodies may be formed and exert an effect on fertility. This study is an evaluation of an immunologic factor in varicocele-associated infertility. Previous studies 5, 6 indicated a relationship between genital tract infection and antispermatozoal antibody formation. We therefore included evaluation of the presence of a coexisting genital tract infection. MEN Varicocele Group MATERIALS AND METHODS Thirty-two men with clinical varicocele from couples who were referred for infertility served as the varicocele group. The varicocele was assessed by the physician who applied the routine method of palpation in the upright position during Valsalva maneuver performed by the patient. Infertile Control Group Twenty-two men without palpable varicocele were selected at random from infertile couples who were referred for consultation. They served as the control group. Fertile Group Ten healthy, fertile men were randomly selected as a fertile group. They all had normal semen analysis and had fathered a child within the last 36 months. Antigen Group A pool of normal spermatozoa, obtained from a group of healthy fertile men; served as a pool of antigen. Semen was collected from all men by masturbation, after 3 days of abstinence. The semen was given into sterile containers and divided into aliquots for semen analysis and for immunologic and microbiologic tests. Urethral smears for isolation of Chlamydia trachomatis were performed in the varicocele and infertile control groups. 398 Golomb et al. Antisperm antibodies in varicocele A medical history was taken and a complete physical examination carried out on the first interview. BLOOD ANALYSIS Blood samples were obtained in all patients by venipuncture and allowed to clot for 60 minutes at room temperature, and the serum separated by centrifugation. The serum was stored in aliquots at - 20 C until tested. SEMEN ANALYSIS Sperm Characteristics Two semen specimens were obtained from each patient, with an interval of 2 weeks. The routine laboratory examination included semen volume, ph, sperm motility (1 hour after ejaculation), sperm vitality, and sperm concentration, which was estimated with the use of a hemocytometer. A spermatocytogram was performed on every specimen. 7 Microbiologic Tests All semen samples were cultured for aerobic bacteria and for Ureaplasma urealyticum. Immunologic Tests All semen specimens were centrifuged at 3000 rpm for 15 minutes.. The pelleted spermatozoa were washed three times with phosphate-buffered saline (PBS), resuspended to a final concentration of 6 x 10 6 sperm/ml, and, after addition of glycerol 21% (vol/vo}), were stored at -70 C until analyzed. The cell-free seminal fluid was dialyzed against PBS overnight and centrifuged at 3500 rpm for 20 minutes; the supernatant was stored in aliquots at - 20 C. Antispermatozoal antibodies were assessed with the use of an enzyme-linked immunosorbent assay (ELISA), which was carried out as previously reported. 8, 9 Both an indirect assay for unbound antibodies and a direct assay for spermbound antibodies were performed. Spermatozoa from the normal pool were pelleted by centrifugation at 3000 rpm for 15 minutes, washed three times with PBS, and resuspended to a final concentration of 6 x 10 6 sperm/ml. Aliquots of 0.1 ml were added to wells of a round-bottom, nonradiated polystyrene microtiter plate (Dynatech Laboratories, Alexan- Fertility and Sterility

3 dria, V A). The plates were incubated overnight to facilitate sedimentation of the spermatozoa; then the buffer was gently aspirated and 0.2 ml of 0.5% glutaraldehyde in PBS was added to the wells to fix the spermatozoa. After 10 minutes, the solution was removed and the wells washed three times with PBS containing 0.05% Tween 20.. In the indirect assay, the tested serum was diluted and the seminal fluid 1110 in Tween wash, and 0.2-ml aliquots were added to duplicate wells that contained fixed-pool spermatozoa and to two blank wells. Known positive and negative controls were always tested in parallel. After an incubation of 120 minutes, the liquid was removed from the wells and the wells were washed three times with Tween wash. The class of antibody bound to the fixed spermatozoa was determined with the use of 0.2 ml of heavy-chain specific alkaline phosphatase conjugated goat antihuman IgG, IgA, or IgM (Sigma Chemical Company, St. Louis, MO) diluted in the Tween wash. After repeated incubation for 120 minutes, the wells were washed three times with Tween wash. Then the alkaline phosphatase was dissolved to a final concentration of 1 mg/ml in bicarbonate buffer, ph 9.8, and 0.2 ml of the substrate was incubated in each well. The absorbance was determined at 405 nm with the use of an ELISA Dynatech plate reader. The incubation was complete at 1 hour, when a positive reference serum had an optical density of Serum and seminal plasma from the fertile group was assayed for the presence of antispermatozoal antibodies. The results were always expressed as a ratio to the positive reference serum. A mean and standard deviation (SD) for serum, seminal plasma, and sperm-bound IgG, IgA, and IgM antibodies to spermatozoa was calculated for the fertile group. For qualitative analysis, we considered a reading from an infertile patient's sample as positive when the absorbance value was at least 2 SD greater than the mean from the fertile group. For quantitative analysis, the OD405 of the sample, compared with the standard reference serum, was used as an absolute value. For direct assay of sperm-bound antibodies, the patient's spermatozoa, resuspended to a concentration of 6 x 10 6 sperm/ml, were fixed to the wells, then incubated with alkaline phosphatase conjugated goat antihuman IgG, IgA, and IgM and processed as described above. Wells without spermatozoa were always assayed to assess the nonspecific binding of the conjugate. STATISTICAL ANALYSIS For qualitative results-patients positive or negative for antibodies-fisher's exact test (twotail) was applied. The two-way analysis of variance, by group and by infection, was used to analyze the quantitative data of the absorbance values in both groups. SEMEN ANALYSIS RESULTS The results of semen analysis are summarized in Table 1. We defined semen characteristics according to Hotchkiss,IO who regarded a sperm count of < 20 x 10 6 /ml as oligozoospermia, motility at 1 hour of < 40% of the spermatozoa as asthenozoospermia, and a rate of pathologic forms > 60% as teratozoospermia. In the varicocele group, 21 men (65.6%) were oligozoospermic, compared with 8 (36.4%) in the control group (P = 0.007). In 17 varicocele patients (53.1 %), asthenozoospermia was found, compared with 5 (22.7%) of the control patients (P = 0.009); teratozoosper- Table 1. Distribution of Oligozoospermia, Asthenozoospermia, Teratozoospermia, and Normozoospermia Among the Varicocele, Compared with the Control, Groupa Group n Varicocele Control Oligozoospermia Asthenozoospermia Teratozoospermia «20 x 10 6 /ml) «40%) (> 60%) Normozoospermia No. % No. % No. % No. % pb asperm characteristics defined according to Hotchkiss. 10 bfisher's exact test (two-tail). ens, not significant Nse Golomb et al. Antisperm antibodies in varicocele 399

4 Table 2. Distribution ofigg, IgA, and IgM Antibodies in Serum, Seminal Plasma, and Bound to Sperm in 32 Varicocele, Compared Serum Group n IgG IgA No. % No. % Varicocele Control Seminal plasma Boundto sperm IgM IgG IgA IgM IgG IgA IgM No. % No. % No. % No. % No. % No. % No. % p b NSc afigures represent number of antibody-positive men. bfisher's exact test (two-tail). ens, not significant NS NS NS NS mia occurred in 22 (68.7%) and 10 (45.5%), respectively (not significant). Four (12.5%) of the varicocele group had normal sperm, compared with 12 (54.5%) of the control group (P = 0.001). Spontaneous agglutination of sperm was not observed in either group. MICROBIOLOGIC STUDIES In the varicocele group, 13 men (40.6%) had an asymptomatic genital tract infection, proven by positive culture. Six were infected by C. trachomatis, six by U. urealyticum, and one by Escherichia coli. In the control group, ten asymptomatic infections were recorded (45.5%), one with C. trachomatis, eight with U. urealyticum, and one with E. coli. IMMUNOLOGIC STUDIES Table 2 summarizes the distribution of IgG, IgA, and IgM antibodies in serum and seminal plasma and bound to sperm in the varicocele, compared with the control group. The figures represent the number of men with positive results for antibodies. The varicocele group differed from the control group in serum IgA (P = 0.006), serum IgM (P = 0.02), seminal plasma IgA (P = ), and seminal plasma IgM (P = 0.01). No correlation was found between the presence of antisperm antibodies of a particular isotype bound to spermatozoa and those detected in seminal plasma. Only five patients demonstrating seminal plasma IgA were also positive for spermbound IgA (25%). As shown in Figure 1, the majority of the varicocele patients (62.5%) had antibodies in more than one location, mainly in serum and seminal plasma, and only 14% of the control group manifested antibodies in combined locations. Three men (9.4%) of the varicocele group 400 Golomb et ai. Antisperm antibodies in varicocele were totally negative for antibodies, compared with 13 men (59.1%) in the control group. We compared the absorbance values obtained for each class of antibody in serum, seminal plasma, or bound to sperm in the varicocele and control groups. The varicocele patients showed significantly higher absorbance values for serum IgA (P = 0.005), seminal plasma IgA (P = 0.01), seminal plasma IgM (P = 0.05), and sperm-bound. IgG (P = 0.003), IgA (P = 0.04), and IgM (P = 0.02) (Fig. 2a-f). In both the varicocele and control groups, no interaction was found between infection and the quantitative level of antispermatozoal antibodies. There was no correlation between sperm abnormalities (sperm count, motility, and morphologic features) and the presence or absence of antibodies in plasma and bound to spermatozoa. t 50. Q. 40 "0 j30 z 20 v * VARICOCELE C"' CONTROL SE.=- SERUM S.P", SEMINAL PLASMA 5 "SPERMATOZOA Figure 1 Distribution of anti spermatozoal antibodies in the varicocele and control groups. The ordinate shows the percentage of patients in whom antisperm antibodies were detected, according to location, and of those who were antibody-negative. Total varicocele group: n = 32. Total control group: n = 22. " SP. Sf,., Fertility and Sterility

5 " o 240 " z.. ;: IE IE :: : O... O w w i z b. ;; t Figure 2.; QJ20 Comparison of the absor- bance values obtained for u noninfected varicocele (NV), infected varicocele (IV), non g infected control (NG), and in-!1l fected control (IG) in serum, o...!!.!!.!!.... _._.- 0 _ seminal plasma, and bound c. to spermatozoa for selected classes of antibodies: (a), see a.1mo w 0 rum IgA (P ' l' = 0.005); (b), seminal plasma IgA (P " = =o ); (c), seminal plasma IE IE IgM (P = 0.05); (d), sperm- :: 0.211)._ bound IgG (P = 0.003); (e),.. o ;;v- sperm-bound IgA (P = 0.04); -I.V- kc I.C HV I.V N.C I.C NUM IER Of PAT! ENTS NUMBER Of PAT! ENTS and (fl, sperm-bound IgM (P = 0.02). (Statistical analysis NV = Non-Infected Varicocele N C = Non- infected Control by twoway analysis of vari- I.V = Infected Varicocele I.e = Infecied Control ance.) Q. DISCUSSION The concept of immunoinfertility has been widely accepted since the antigenicity of spermatozoa was first demonstrated by Landsteiner, in Antispermatozoal antibody formation has been reported after vasectomy,12 in experimental testicular injury, 13 after testicular biopsy, 14 and after testicular torsion.15, 16 Disruption of the blood testis barrier is probably a prerequisite for antisperm antibody formation. The testis in varicocele undoubtedly suffers a deleterious tissue effect, which sometimes leads to complete atrophy; this may be attributed to the prolonged venous stasis and hyperthermia. This tissue injury may induce unshielding of the protected spermatozoa, an antisperm immunologic response, and antibody formation.. In this preliminary study, we tested two infertile populations for antispermatozoal antibodies and demonstrated a significant difference between the varicocele and control groups: 29 men with varicocele (90.6%) had antispermatozoal antibodies, compared with only 9 (40.9%) in the control group. The antibodies were mainly of the IgA and IgM classes in serum, seminal plasma, and bound to sperm. There is an apparent disparity in the incidence of the immunity rate in our control group, compared with that found in some studies of un- selected infertile couples (approximately 10%) This can be attributed to the fact that in those series, only one location of antibodies was searched (serum, seminal plasma, or bound to sperm), usually analyzing only one isotype of antibody; we carried out a combined study in two levels (location and isotypes of antibodies). Presumably, therefore, we had a higher incidence rate of immunity to spermatozoa. In addition, the high sensitivity of the ELISA may contribute to the disparity in rate. It is apparent from our re;ults (Table 2) that the indirect ELISA was more sensitive than the direct assay, because the incidence of spermbound antibodies in the varicocele group was lower than that detected in the seminal plasma of those patients. A possible explanation is that technically, because of freezing and-thawing the spermatozoa and repeated washings, the antibodies were deleted from the sperm surface, leading to an erroneous lower reading. The incidence of a positive culture for C. trachomatis, U. urealyticum, or E. coli in the two groups was very similar, and no statistical interaction with antibody formation could be demonstrated. Although it had been reported5,6 that such an interaction exists, mainly regarding Mycoplasma and Chlamydia, 5 our findings are in agreement with those of Upadhyaya et al.,17, 18 who found no such correlation. Golomb et al. Antisperm antibodies in varicocele 401

6 According to McShane,19 oligozoospermia and poor sperm motility have been noted in some se" ries of men with sperm autoimmunity without evidence of autoimmune orchitis on biopsy study. We have demonstrated both decreased sperm quality and increased sperm autoimmunity in the varicocele group, compared with the control group. It remains to be assessed whether sperm immunity leads to oligozoospermia and poor motility or whether there is an underlying defect in spermatogenesis that contributes to' both autoimmunity and decreased sperm quality. Ideally, we should have examined fertile varicocele patients for sperm quality and antisperm antibodies and compared them with infertile varicocele patients, but this is technically almost impossible, because fertile men usually do not seek medical aid in an infertility clinic, and symptomatic varicocele is rare. The contribution of our observations to the enigma of varicocele infertility is yet unclear. As already reported,19, 20 antispermatozoal antibodies may affect reproduction by different mechanisms; for example, agglutination or immobilization of spermatozoa, sperm cytotoxicity, impairment of sperm penetration into cervical mucus, prevention of capacitation or penetration of ova membranes, and enhanced phagocytosis of sperm in the genital tract by macrophages. Therefore, any single assessment for the detection of the presence of anti spermatozoal antibodies usually does not correlate with subsequent fertility. Our ability to differentiate the class of antibody and to quantify the antibody level may allow for such correlations to be sought in subsequent investigations. Acknowledgments. We wish to thank Sophie Livio for her technical assistance and Ilana Gelernter for the statistical workup. The secretarial assistance of Chaya Fogel is greatly appreciated. REFERENCES, 1. MacLeod J: Seminal cytology in the presence of varicocele. Fertil Steril 16:735, McFadden MR, Mehan DJ: Testicular biopsies in 101 cases of varicocele. J Urol 119:372, Turner TT: Varicocele: still an enigma. J Urol 129:695, Getzoff PL: Surgical management of male infertility: results of a survey. Fertil Steril 24:553, Witkin SS, Toth A: Relationship between genital tract infections, sperm antibodies in seminal fluid, and infertility. Fertil Steril 40:805, Mathur S; Baker ER, Williamson HO, Derrick FC, Teague KJ, Fudenberg HH: Clinical significance of sperm antibodies in infertility. Fertil Steril 36:486, Homonnai ZT, paz GF, Weiss IN, David MP: Relation between semen quality and fate of pregnancy: retrospective study on 534 pregnancies. Int J Androl 3:574, Witkin SS: Enzyme linked immunosorbent assay (ELISA) for detection of antibodies to spermatozoa. Res Reprod 15:1, Witkin SS, Zelikovsky G, Good RA, Day NK: Demonstra' tion of 11S IgA antibody to spermatozoa in human seminal fluid. Clin Exp Immunol 44:368, Hotchkiss RS: Infertility in the male. Iri Urology, Vol I, Edited by MF Campbell, JH Harrison. Philadelphia, W. B. Saunders Company, 1970, p Landsteiner K: Zur Kenntnis der Spezifisch auf Blutkorperchen wirkenden Sera. Zentralbl Bakteriol 25:546, Sullivan MJ, Howe GE: Correlation of circulating antisperm antibodies to functional success in vasovasostomy. J Urol 117:189, Rappaport FT, Sampath A, Kano K, McCluskey RT, Milgrom F: Immunological effects of thermal injury. J Exp Med 130:1411, Hjort T, Husted S, Linnet-Jepsen P: The effect of testis biopsy on autosensitization against spermatozoal antigens. Clin Exp ImmunoI18:201, Nagler HM, White RDV: The effect of testicular torsion on the contralateral testis. J Urol 128:1343, Merimsky E, Orni-Wasserlauf R, Yust I: Assessment of immunological mechanism in infertility of the rat after experimental testicular torsion. Urol Res 12:179, Upadhyaya M, Hibbard BM, Walker SM: Antisperm antibodies and male infertility. Br J Urol 56:531, Upadhyaya M, Hibbard BM, Walker SM: The effect of Ureaplasma urealyticum onsemen characteristics. Fertil Steril 41:304, McShane PM: Immunological aspects of male infertility. Semin Urol 2:107, London SN, Haney AF, Weinberg JB: Diverse humoral and cell-mediated effects of antisperm antibodies on reproduction. Fertil Steril 41:907, Golomb et ai. Antisperm antibodies in varicocele Fertility and Sterility