Human Reproduction vol.6 no.2 pp.267-276, 1991
Relationship between local anti-sperm antibodies and sperm—mucus interaction in vitro and in vivo*
W.Eggert-Kruse1, A.HofsaB, E.Haury, W.Tilgen2, I.Gerhard and B.Runnebaum Division of Gynaecological Endocrinology, Women's Hospital and 2 Department of Dermatology, University of Heidelberg, 6900 Heidelberg, FRG 'To whom correspondence should be addressed
Sperm-mucus interaction under in-vitro or in-vivo conditions can be affected by local anti-sperm antibodies (ASA). In a prospective study, a mixed antiglobulin reaction (MAR) test, using immunoglobulin-coated red cells and spermatozoa, was used to detect the presence of IgG and/or IgA on the spermatozoal surface in 209 males of infertile partnerships. The results of direct MAR tests range from 0 to 100% and a significant correlation between MAR % IgG and MAR % IgA (r = 0.74) was found. MAR positive semen samples were significantly more frequent in the group of patients with an inadequate outcome of postcoital testing (PCT). The sperm-mucus interaction in vitro, evaluated by means of the in-vitro sperm-cervical mucus penetration test (SCMPT) with cervical mucus (CM) of patients' partners, was particularly related to spermatozoal surface antibodies of the IgA class: MAR IgA positive ejaculates were seen in 13.9% of males with inadequate SCMPT versus 4.4% when SCMPT was adequate. The significant correlation between MAR IgA and the sperm penetration ability in vitro could also be proven when donors' CM was taken for the SCMPT, but not when a non-human material was used for in-vitro penetration testing. Mkrobial colonization of semen specimens did not interfere with the outcome of MAR testing. The pregnancy rate after 12 months was 23% (48/209) and was significantly lower when > 30% of spermatozoa were covered with surface ASA of the IgG or IgA class. The results suggest that the mixed antiglobulin reaction (MAR test) can be recommended as a valuable adjunct to sperm-mucus interaction testing to detect immunological causes of impaired sperm function. Key words: anti-sperm antibodies/mixed antiglobulin reaction/ sperm-mucus interaction/microbial colonization
Introduction The incidence of immunological factors as the cause of a couple's infertility is still unclear. The immense number of studies undertaken to show that infertility could have an immune basis (reviewed by Haas etal., 1980; Bronson, 1984a; Alexander and •Presented in part at the XIII World Congress on Sterility and Fertility, Held in October 1989, Marrakesh, Morocco © Oxford University Press
Anderson, 1987; Hendry, 1989) has produced conflicting data rather than absolute proof that immunity to spermatozoa can impair human reproduction. Most investigations with regard to immunological infertility were conducted in serum samples to detect circulating anti-sperm antibodies (ASA) in males and/or females of infertile partnerships. We reported on this recently (Eggert-Kruse et al., 1989a). However, some studies suggest that spermatozoal surface antibodies are more important for local compatibility (Kremer and Jager, 1988). Therefore, in the present investigation 209 males from couples with long-standing infertility were screened for anti-sperm antibodies in semen samples by means of the mixed antiglobulin reaction (MAR) test for detection of sperm-attached IgG and/or IgA immunoglobulins (Jager et al., 1978) and parallel testing was performed in serum samples with a standard test for detection of circulating antibodies, the tray agglutination test (TAT) (Friberg, 1974b). In addition, couples were screened for a variety of female and male infertility factors including extensive hormonal analyses, evaluation of tuba! patency and uterine factors in females and standardized tests for sperm function in males. The influence of local antibodies on sperm—mucus interaction was studied in vivo by means of the postcoital test (PCT) and in vitro with the sperm-cervical mucus penetration test (SCMPT) performed with cervical mucus (CM) of the patients' female partners as well as CM of fertile donors. The influence of microbial colonization of semen on the mixed antiglobulin reaction was evaluated and the correlation of local and circulating antibodies was determined. Furthermore, the clinical significance of local anti-sperm antibodies of the IgG or IgA class for the subsequent pregnancy rate was analysed in a prospective study.
Materials and methods The study population consisted of 209 randomly chosen couples with a median duration of infertility of 4 years (range 1-19 years) who were submitted to a comprehensive infertility investigation. The median age of the women was 29.5 (range 21 -40) years, the median age of their male partners was 32 (range 23-57) years. A detailed medical history was obtained and physical examinations performed on both partners. The tubal factor was evaluated by hysterosalpingography (HSG) and/or laparoscopy with tubal dye transit. Tubal patency was good (both sides free) in 63%, slightly impaired in 32% and severely impaired in 5% of women, but patients were only included in the study when they had at least one free patent tube. Extensive hormonal analyses had been carried out, including 267
W.Eggert-Kruse et al.
tests for ovarian, pituitary, thyroid and adrenal function in the early follicular phase, multiple determinations of oestradiol and progesterone in the luteal phase and premenstrual endometrial biopsies. By means of this screening procedure, some hormonal disorders were found in 52 % of women and a severely impaired cycle function in 23% of female patients. Specific treatment was administered according to the results of hormonal evaluations whenever necessary. Males represented a wide range of andrological findings, but patients with azoospermia and without motile spermatozoa were excluded. Andrological medication was given according to the results of sperm analyses and hormonal evaluations, but none of the males received corticosteroids or antibiotics. All patients in this study were without signs or symptoms of infection of the lower genital tract. Semen analysis and microbial examination of ejaculates
air bubbles within the capillary tube and to leave one drop of CM protruding at the other end. Sperm penetration was observed microscopically Qow power field, (x 100), LPF) after 30 min, after 2 and 6 h incubation at 37°C in a moist atmosphere. Penetration parameters were recorded without knowledge of the other tests. The following criteria were examined: the penetration density, the migration distance on a scale from 0 to 60 mm and the quality and duration of motility. The results of each variable were graded from 0 to 3 and summarized in a cumulative SCMPT score. Results of SCMPT are reported as inadequate when the cumulative score after 6 h incubation time was < 6 , otherwise, testing was classified as adequate. For further analyses, patients were subdivided according to SCMPT results into four groups: SCMPT negative, poor, good and excellent. SCMPT was performed with semen and CM of the infertile couple and as a cross-matched penetration test with CM of fertile donors.
Evaluation of sperm—mucus interaction In order to obtain semen and cervical mucus for performance of the in-vitro sperm—cervical mucus penetration test, couples were seen at the Infertility Unit between the 9th and the 14th day of the women's menstrual cycle. SCMPT, testing for local and circulating anti-sperm antibodies and microbial examinations were performed simultaneously in the majority of patients. In order to standardize hormonal influences on the mucus quality, 80 fig of ethinyloestradiol (EE) per day were administered orally for at least 7 days before SCMPT. Treatment with oestrogens before SCMPT was compulsory for patients' partners as well as for female donors of CM. Any medication with a potentially negative effect on the mucus quality (e.g. clomiphene citrate) was stopped in the previous cycle. In female patients, the cervix was exposed without speculum lubricant and the cervical index according to Insler (Insler et al., 1972) was assigned. The exocervix was first cleaned of excess debris with a large sterile cotton swab. Cervical mucus was obtained from the endocervix by gentle aspiration with a sterile device (AspiglaireR, IMV, L'Aigle, France).
Sperm penetration in vivo. Postcoital testing (PCT) was evaluated on the day of presumed ovulation (by basal body temperature (BBT) chart, examination of CM and ultrasound monitoring) after a 5-day period of sexual abstinence as a screening procedure for sperm—mucus interaction in vivo. Endocervical mucus was gently aspirated as described above and mucus samples were examined microscopically using first the LPF and then the HPF magnification. The number of spermatozoa with forward progression in CM was counted, the mean of 20 fields was taken. When there were no spermatozoa found in preovulatory CM, a vaginal pool sample was additionally examined to ensure that semen had actually been deposited in the vagina, otherwise, the testing was repeated in the next cycle. The grading of the PCT outcome, based on a WHO classification (WHO, 1976), has been described in detail elsewhere (Eggert-Kruse et al., 1987). Briefly, PCT was regarded as adequate when at least two spermatozoa of highly progressive motility/HPF were counted in preovulatory CM 8—12 h after intercourse, otherwise it was classified as inadequate. For further evaluation, the PCT results were subdivided into four groups: PCT negative (no spermatozoa found in CM/LPF, but in vaginal secretions), PCT poor ( < 2 motile sperm/HPF), PCT good ( 2 - 6 motile sperm/HPF), and PCT excellent ( ^ 7 motile sperm/HPF). A poor or negative result was only accepted as valid if the mucus was in good condition and after the test had been repeated in at least one other cycle, an initial negative result was ignored if later positive. In patients with anovulatory cycles, very late ovulation or marked luteal insufficiency, PCT was initially performed after oral treatment with ethinyloestradiol as described above. In the majority of patients, PCT was controlled under the same standardized conditions to ensure a comparable mucus quality. As for the SCMPT, in the test cycle no other medication than EE, which could possibly interfere with the rheological properties of CM, was administered.
Sperm penetration in vitro. An aliquot of each semen sample was used to perform the in-vitro SCMPT. The penetration of spermatozoa within each sample of cervical mucus was assessed as described in detail previously (Eggert-Kruse et al., 1989b). Briefly, the testing consisted of transferring fresh CM into a capillary tube and inserting one end into a small reservoir of semen, used directly after liquefaction. Care was taken to avoid
Assessment of the sperm-penetration ability with a non-human material. Additionally, for evaluation of sperm penetration in vitro, a non-human medium was used. A special fraction of fresh hen's egg white (HEW) was used to fill the capillaries of a Kremer sperm penetration meter (Kremer, 1965). The procedure has been described in detail elsewhere (Eggert-Kruse et al., 1990a). The HEW assay was read after 30 min incuba-
Semen was obtained in the hospital by masturbation into sterile glass jars after 5 days of sexual abstinence. For sperm analysis, standard criteria were used (WHO, 1987). The number of round cells (undifferentiated) was determined in wet preparations. A Papanicolaou (Pap) stain was prepared for differential morphology. Directly after liquefaction, aliquots of each semen sample (10 /tl) were transferred to the transport media (Port-ACul-Universal, Becton Dickinson, Heidelberg) for detection of potentially pathogenic aerobic and anaerobic bacteria. For culture and identification of microorganisms, standard methods were applied (Department of Microbiology, University of Heidelberg). No urethral swabs were taken.
268
Influence of local antibodies on sperm — mucus interaction
tion at 37°C in a moist Petri dish in a test setting parallel to the SCMPT. Performance of the mixed qntiglobulin reaction (MAR) test The MAR test was performed as described in detail by Jager et al. (1978). Briefly, either IgG or IgA-coated erythrocytes were used to detect IgG and/or IgA antibodies on the sperm surface. For MAR IgG testing, the erythrocytes, sensitized with anti-D serum, were kindly prepared by the Blood Grouping Laboratory of the Institute of Serology of the University of Heidelberg. For preparation of IgA erythrocytes, four major steps were necessary to coat the cells with IgA: the gamma fraction was isolated from colostrum of the first or second day post partum, obtained from women having delivered at the Women's University Hospital. After some washing procedures and adjustment of the pH, the erythrocytes were fixed with 3% formaldehyde solution and tanned with 0.005 % tannic acid before coating. The IgA-coated erythrocytes were stored at — 20°C after freezing with liquid nitrogen in thermostable containers, until thawed shortly before MAR testing. All the indicator cells were prepared in the same way and by the same person. For MAR IgG testing, a drop of freshly ejaculated semen, a drop of undiluted monospecific antiserum (Behring Werke, Marburg) and a drop of the sensitized erythrocyte suspension were set separately but close to each other on a microscopic slide. The drops were thoroughly mixed and the mixture covered by a cover slip before the reaction was then studied at once under a light microscope. In positive reactions, the erythrocytes adhered to the motile spermatozoa directly after mixing of the three drops, forming mixed aggregates. The results were noted as the percentage of the motile spermatozoa incorporated into mixed agglutinates, this percentage is described as MAR %. No interpretation of the test was given unless agglutination of the red blood cells and the presence of motile spermatozoa were observed. The outcome of the MAR test was classified as positive (pos), if >30% of the motile spermatozoa were involved in the mixed agglutinates, otherwise it was classified as negative (neg). Results were considered strongly positive, if ^60% of the motile spermatozoa adhered to the coated erythrocytes. For MAR IgG, the slide method was used, for MAR IgA the reaction was observed in capillary tubes directly after aspiration and after 30 min incubation in a moist atmosphere at 37°C according to Jager et al. (1980). Both MAR IgG and IgA testing were performed in parallel with aliquots of the same semen sample by the same observer without knowledge of the other test results. The reproducibility of results was tested using erythrocytes after different times of storage (variability 3%), and by performing the MAR test by several observers and by checking the same ejaculate with the MAR test several times by the same observer with an overall correlation of results of r = 0.91 (P < 0.0001) (analysis of variance). Peformance of the tray-agglutination test (TA T) Serum samples were examined for the presence of circulating ASA with a standard micro-agglutination technique (TAT) according to Friberg (1974a), which included dilutions of 1:4
to 1:1024 of complement-inactivated serum. Blood was taken at the day of SCMPT. In case of positive TAT results, the outcome was controlled in a second test setting with an ejaculate of another donor and by the independent examination of a second observer. WHO reference sera (Aarhus, Denmark) were included in all of the test series. With regard to the suggestions of Friberg (1974b) and Friberg and Tilly-Friberg (1977), titres of ASA are reported as positive when > 1:8. In addition to the reference sera, serum samples of 32 virgins who consulted the outpatient clinic for endocrine disorders, mostly cycle irregularities or amenorrhoea and from whom blood was taken for endocrine function tests, were checked with the TAT as an internal control. All of these 32 TAT results were completely negative. Statistical analysis For statistical evaluation, the Wilcoxon rank sum, Kruskal — Wallis, Chi square, Fisher's two-tailed exact tests and Spearman rank correlation were used. Data were processed using the statistical analysis system (SAS). The level of significance was set at P < 0.05. Results Results of MAR testing Semen samples of 209 males were examined for IgG and/or IgA antibodies on the sperm surface and the results of MAR testing are presented in Figure 1. The percentage of spermatozoa covered with surface antibodies of the IgG class ranged from 0 to 100% with a median of 10%. The percentage of spermatozoa with IgA immunoglobulins on the surface also ranged from 0 to 100% with a median of 5 %. 13 % of semen samples (28/209) offered a MAR % IgG of > 30% (MAR IgG positive) and 9% (19/209) of ejaculates offered a MAR % IgA of > 30% (MAR IgA positive); 4% of semen samples showed a strong positive reaction with ^ 6 0 % of spermatozoa with IgG and/or IgA antibodies on the surface. When MAR % IgG and MAR % IgA were compared, a close positive correlation was noted. The Spearman rank correlation coefficient (r) was 0.74 (P < 0.0001). Correlation of antibody testing and results of sperm analysis Table I compares the parameters of sperm analysis in patients with positive and negative outcome of MAR IgG and MAR IgA testing (Wilcoxon rank sum tests). Significant differences were found particularly for sperm progressive motility and morphology, but not for sperm count, volume, fructose concentration, pH and number of round cells. Furthermore, after the selection of patients with regard to the results of sperm analysis in groups with 40% of spermatozoa with progressive motility, MAR IgG positive semen samples were significantly more frequent in ejaculates with reduced progressive motility (Chi square analysis). The incidence of IgG positive samples was 21% (15/73) versus 10% (13/136) (P < 0.05), and the incidence of MAR % IgA positive samples was 14 versus 6%, respectively. In ejaculates with marked asthenozoospermia (
Relationship between local anti-sperm antibodies and sperm—mucus interaction in vitro and in vivo*
W.Eggert-Kruse1, A.HofsaB, E.Haury, W.Tilgen2, I.Gerhard and B.Runnebaum Division of Gynaecological Endocrinology, Women's Hospital and 2 Department of Dermatology, University of Heidelberg, 6900 Heidelberg, FRG 'To whom correspondence should be addressed
Sperm-mucus interaction under in-vitro or in-vivo conditions can be affected by local anti-sperm antibodies (ASA). In a prospective study, a mixed antiglobulin reaction (MAR) test, using immunoglobulin-coated red cells and spermatozoa, was used to detect the presence of IgG and/or IgA on the spermatozoal surface in 209 males of infertile partnerships. The results of direct MAR tests range from 0 to 100% and a significant correlation between MAR % IgG and MAR % IgA (r = 0.74) was found. MAR positive semen samples were significantly more frequent in the group of patients with an inadequate outcome of postcoital testing (PCT). The sperm-mucus interaction in vitro, evaluated by means of the in-vitro sperm-cervical mucus penetration test (SCMPT) with cervical mucus (CM) of patients' partners, was particularly related to spermatozoal surface antibodies of the IgA class: MAR IgA positive ejaculates were seen in 13.9% of males with inadequate SCMPT versus 4.4% when SCMPT was adequate. The significant correlation between MAR IgA and the sperm penetration ability in vitro could also be proven when donors' CM was taken for the SCMPT, but not when a non-human material was used for in-vitro penetration testing. Mkrobial colonization of semen specimens did not interfere with the outcome of MAR testing. The pregnancy rate after 12 months was 23% (48/209) and was significantly lower when > 30% of spermatozoa were covered with surface ASA of the IgG or IgA class. The results suggest that the mixed antiglobulin reaction (MAR test) can be recommended as a valuable adjunct to sperm-mucus interaction testing to detect immunological causes of impaired sperm function. Key words: anti-sperm antibodies/mixed antiglobulin reaction/ sperm-mucus interaction/microbial colonization
Introduction The incidence of immunological factors as the cause of a couple's infertility is still unclear. The immense number of studies undertaken to show that infertility could have an immune basis (reviewed by Haas etal., 1980; Bronson, 1984a; Alexander and •Presented in part at the XIII World Congress on Sterility and Fertility, Held in October 1989, Marrakesh, Morocco © Oxford University Press
Anderson, 1987; Hendry, 1989) has produced conflicting data rather than absolute proof that immunity to spermatozoa can impair human reproduction. Most investigations with regard to immunological infertility were conducted in serum samples to detect circulating anti-sperm antibodies (ASA) in males and/or females of infertile partnerships. We reported on this recently (Eggert-Kruse et al., 1989a). However, some studies suggest that spermatozoal surface antibodies are more important for local compatibility (Kremer and Jager, 1988). Therefore, in the present investigation 209 males from couples with long-standing infertility were screened for anti-sperm antibodies in semen samples by means of the mixed antiglobulin reaction (MAR) test for detection of sperm-attached IgG and/or IgA immunoglobulins (Jager et al., 1978) and parallel testing was performed in serum samples with a standard test for detection of circulating antibodies, the tray agglutination test (TAT) (Friberg, 1974b). In addition, couples were screened for a variety of female and male infertility factors including extensive hormonal analyses, evaluation of tuba! patency and uterine factors in females and standardized tests for sperm function in males. The influence of local antibodies on sperm—mucus interaction was studied in vivo by means of the postcoital test (PCT) and in vitro with the sperm-cervical mucus penetration test (SCMPT) performed with cervical mucus (CM) of the patients' female partners as well as CM of fertile donors. The influence of microbial colonization of semen on the mixed antiglobulin reaction was evaluated and the correlation of local and circulating antibodies was determined. Furthermore, the clinical significance of local anti-sperm antibodies of the IgG or IgA class for the subsequent pregnancy rate was analysed in a prospective study.
Materials and methods The study population consisted of 209 randomly chosen couples with a median duration of infertility of 4 years (range 1-19 years) who were submitted to a comprehensive infertility investigation. The median age of the women was 29.5 (range 21 -40) years, the median age of their male partners was 32 (range 23-57) years. A detailed medical history was obtained and physical examinations performed on both partners. The tubal factor was evaluated by hysterosalpingography (HSG) and/or laparoscopy with tubal dye transit. Tubal patency was good (both sides free) in 63%, slightly impaired in 32% and severely impaired in 5% of women, but patients were only included in the study when they had at least one free patent tube. Extensive hormonal analyses had been carried out, including 267
W.Eggert-Kruse et al.
tests for ovarian, pituitary, thyroid and adrenal function in the early follicular phase, multiple determinations of oestradiol and progesterone in the luteal phase and premenstrual endometrial biopsies. By means of this screening procedure, some hormonal disorders were found in 52 % of women and a severely impaired cycle function in 23% of female patients. Specific treatment was administered according to the results of hormonal evaluations whenever necessary. Males represented a wide range of andrological findings, but patients with azoospermia and without motile spermatozoa were excluded. Andrological medication was given according to the results of sperm analyses and hormonal evaluations, but none of the males received corticosteroids or antibiotics. All patients in this study were without signs or symptoms of infection of the lower genital tract. Semen analysis and microbial examination of ejaculates
air bubbles within the capillary tube and to leave one drop of CM protruding at the other end. Sperm penetration was observed microscopically Qow power field, (x 100), LPF) after 30 min, after 2 and 6 h incubation at 37°C in a moist atmosphere. Penetration parameters were recorded without knowledge of the other tests. The following criteria were examined: the penetration density, the migration distance on a scale from 0 to 60 mm and the quality and duration of motility. The results of each variable were graded from 0 to 3 and summarized in a cumulative SCMPT score. Results of SCMPT are reported as inadequate when the cumulative score after 6 h incubation time was < 6 , otherwise, testing was classified as adequate. For further analyses, patients were subdivided according to SCMPT results into four groups: SCMPT negative, poor, good and excellent. SCMPT was performed with semen and CM of the infertile couple and as a cross-matched penetration test with CM of fertile donors.
Evaluation of sperm—mucus interaction In order to obtain semen and cervical mucus for performance of the in-vitro sperm—cervical mucus penetration test, couples were seen at the Infertility Unit between the 9th and the 14th day of the women's menstrual cycle. SCMPT, testing for local and circulating anti-sperm antibodies and microbial examinations were performed simultaneously in the majority of patients. In order to standardize hormonal influences on the mucus quality, 80 fig of ethinyloestradiol (EE) per day were administered orally for at least 7 days before SCMPT. Treatment with oestrogens before SCMPT was compulsory for patients' partners as well as for female donors of CM. Any medication with a potentially negative effect on the mucus quality (e.g. clomiphene citrate) was stopped in the previous cycle. In female patients, the cervix was exposed without speculum lubricant and the cervical index according to Insler (Insler et al., 1972) was assigned. The exocervix was first cleaned of excess debris with a large sterile cotton swab. Cervical mucus was obtained from the endocervix by gentle aspiration with a sterile device (AspiglaireR, IMV, L'Aigle, France).
Sperm penetration in vivo. Postcoital testing (PCT) was evaluated on the day of presumed ovulation (by basal body temperature (BBT) chart, examination of CM and ultrasound monitoring) after a 5-day period of sexual abstinence as a screening procedure for sperm—mucus interaction in vivo. Endocervical mucus was gently aspirated as described above and mucus samples were examined microscopically using first the LPF and then the HPF magnification. The number of spermatozoa with forward progression in CM was counted, the mean of 20 fields was taken. When there were no spermatozoa found in preovulatory CM, a vaginal pool sample was additionally examined to ensure that semen had actually been deposited in the vagina, otherwise, the testing was repeated in the next cycle. The grading of the PCT outcome, based on a WHO classification (WHO, 1976), has been described in detail elsewhere (Eggert-Kruse et al., 1987). Briefly, PCT was regarded as adequate when at least two spermatozoa of highly progressive motility/HPF were counted in preovulatory CM 8—12 h after intercourse, otherwise it was classified as inadequate. For further evaluation, the PCT results were subdivided into four groups: PCT negative (no spermatozoa found in CM/LPF, but in vaginal secretions), PCT poor ( < 2 motile sperm/HPF), PCT good ( 2 - 6 motile sperm/HPF), and PCT excellent ( ^ 7 motile sperm/HPF). A poor or negative result was only accepted as valid if the mucus was in good condition and after the test had been repeated in at least one other cycle, an initial negative result was ignored if later positive. In patients with anovulatory cycles, very late ovulation or marked luteal insufficiency, PCT was initially performed after oral treatment with ethinyloestradiol as described above. In the majority of patients, PCT was controlled under the same standardized conditions to ensure a comparable mucus quality. As for the SCMPT, in the test cycle no other medication than EE, which could possibly interfere with the rheological properties of CM, was administered.
Sperm penetration in vitro. An aliquot of each semen sample was used to perform the in-vitro SCMPT. The penetration of spermatozoa within each sample of cervical mucus was assessed as described in detail previously (Eggert-Kruse et al., 1989b). Briefly, the testing consisted of transferring fresh CM into a capillary tube and inserting one end into a small reservoir of semen, used directly after liquefaction. Care was taken to avoid
Assessment of the sperm-penetration ability with a non-human material. Additionally, for evaluation of sperm penetration in vitro, a non-human medium was used. A special fraction of fresh hen's egg white (HEW) was used to fill the capillaries of a Kremer sperm penetration meter (Kremer, 1965). The procedure has been described in detail elsewhere (Eggert-Kruse et al., 1990a). The HEW assay was read after 30 min incuba-
Semen was obtained in the hospital by masturbation into sterile glass jars after 5 days of sexual abstinence. For sperm analysis, standard criteria were used (WHO, 1987). The number of round cells (undifferentiated) was determined in wet preparations. A Papanicolaou (Pap) stain was prepared for differential morphology. Directly after liquefaction, aliquots of each semen sample (10 /tl) were transferred to the transport media (Port-ACul-Universal, Becton Dickinson, Heidelberg) for detection of potentially pathogenic aerobic and anaerobic bacteria. For culture and identification of microorganisms, standard methods were applied (Department of Microbiology, University of Heidelberg). No urethral swabs were taken.
268
Influence of local antibodies on sperm — mucus interaction
tion at 37°C in a moist Petri dish in a test setting parallel to the SCMPT. Performance of the mixed qntiglobulin reaction (MAR) test The MAR test was performed as described in detail by Jager et al. (1978). Briefly, either IgG or IgA-coated erythrocytes were used to detect IgG and/or IgA antibodies on the sperm surface. For MAR IgG testing, the erythrocytes, sensitized with anti-D serum, were kindly prepared by the Blood Grouping Laboratory of the Institute of Serology of the University of Heidelberg. For preparation of IgA erythrocytes, four major steps were necessary to coat the cells with IgA: the gamma fraction was isolated from colostrum of the first or second day post partum, obtained from women having delivered at the Women's University Hospital. After some washing procedures and adjustment of the pH, the erythrocytes were fixed with 3% formaldehyde solution and tanned with 0.005 % tannic acid before coating. The IgA-coated erythrocytes were stored at — 20°C after freezing with liquid nitrogen in thermostable containers, until thawed shortly before MAR testing. All the indicator cells were prepared in the same way and by the same person. For MAR IgG testing, a drop of freshly ejaculated semen, a drop of undiluted monospecific antiserum (Behring Werke, Marburg) and a drop of the sensitized erythrocyte suspension were set separately but close to each other on a microscopic slide. The drops were thoroughly mixed and the mixture covered by a cover slip before the reaction was then studied at once under a light microscope. In positive reactions, the erythrocytes adhered to the motile spermatozoa directly after mixing of the three drops, forming mixed aggregates. The results were noted as the percentage of the motile spermatozoa incorporated into mixed agglutinates, this percentage is described as MAR %. No interpretation of the test was given unless agglutination of the red blood cells and the presence of motile spermatozoa were observed. The outcome of the MAR test was classified as positive (pos), if >30% of the motile spermatozoa were involved in the mixed agglutinates, otherwise it was classified as negative (neg). Results were considered strongly positive, if ^60% of the motile spermatozoa adhered to the coated erythrocytes. For MAR IgG, the slide method was used, for MAR IgA the reaction was observed in capillary tubes directly after aspiration and after 30 min incubation in a moist atmosphere at 37°C according to Jager et al. (1980). Both MAR IgG and IgA testing were performed in parallel with aliquots of the same semen sample by the same observer without knowledge of the other test results. The reproducibility of results was tested using erythrocytes after different times of storage (variability 3%), and by performing the MAR test by several observers and by checking the same ejaculate with the MAR test several times by the same observer with an overall correlation of results of r = 0.91 (P < 0.0001) (analysis of variance). Peformance of the tray-agglutination test (TA T) Serum samples were examined for the presence of circulating ASA with a standard micro-agglutination technique (TAT) according to Friberg (1974a), which included dilutions of 1:4
to 1:1024 of complement-inactivated serum. Blood was taken at the day of SCMPT. In case of positive TAT results, the outcome was controlled in a second test setting with an ejaculate of another donor and by the independent examination of a second observer. WHO reference sera (Aarhus, Denmark) were included in all of the test series. With regard to the suggestions of Friberg (1974b) and Friberg and Tilly-Friberg (1977), titres of ASA are reported as positive when > 1:8. In addition to the reference sera, serum samples of 32 virgins who consulted the outpatient clinic for endocrine disorders, mostly cycle irregularities or amenorrhoea and from whom blood was taken for endocrine function tests, were checked with the TAT as an internal control. All of these 32 TAT results were completely negative. Statistical analysis For statistical evaluation, the Wilcoxon rank sum, Kruskal — Wallis, Chi square, Fisher's two-tailed exact tests and Spearman rank correlation were used. Data were processed using the statistical analysis system (SAS). The level of significance was set at P < 0.05. Results Results of MAR testing Semen samples of 209 males were examined for IgG and/or IgA antibodies on the sperm surface and the results of MAR testing are presented in Figure 1. The percentage of spermatozoa covered with surface antibodies of the IgG class ranged from 0 to 100% with a median of 10%. The percentage of spermatozoa with IgA immunoglobulins on the surface also ranged from 0 to 100% with a median of 5 %. 13 % of semen samples (28/209) offered a MAR % IgG of > 30% (MAR IgG positive) and 9% (19/209) of ejaculates offered a MAR % IgA of > 30% (MAR IgA positive); 4% of semen samples showed a strong positive reaction with ^ 6 0 % of spermatozoa with IgG and/or IgA antibodies on the surface. When MAR % IgG and MAR % IgA were compared, a close positive correlation was noted. The Spearman rank correlation coefficient (r) was 0.74 (P < 0.0001). Correlation of antibody testing and results of sperm analysis Table I compares the parameters of sperm analysis in patients with positive and negative outcome of MAR IgG and MAR IgA testing (Wilcoxon rank sum tests). Significant differences were found particularly for sperm progressive motility and morphology, but not for sperm count, volume, fructose concentration, pH and number of round cells. Furthermore, after the selection of patients with regard to the results of sperm analysis in groups with 40% of spermatozoa with progressive motility, MAR IgG positive semen samples were significantly more frequent in ejaculates with reduced progressive motility (Chi square analysis). The incidence of IgG positive samples was 21% (15/73) versus 10% (13/136) (P < 0.05), and the incidence of MAR % IgA positive samples was 14 versus 6%, respectively. In ejaculates with marked asthenozoospermia (
