- Email: [email protected]
Correlation between puberty and the development of autoimmunity to spermatozoa in men with cystic fibrosis
Vol. 58, No.6, December 1992
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright © 1992 The American Fertility Society
Correlation between puberty and the development of autoimmunity to spermatozoa in men with cystic fibrosis
Richard A. Bronson, M.D.*t Walter J. O'Connor, M.D.:j: Thomas A. Wilson, M.D.§
Susan K. Bronson, M.P.H.II Fred I. Chasalow, Ph.D.~ Kathleen Droesch, M.D.*
State University of New York (SUNY) at Stony Brook, Stony Brook, and Maimonides Medical Center, Brooklyn, New York
Objective: To test the hypothesis that puberty is a necessary factor in the pathogenesis of autoimmunity to sperm in men with cystic fibrosis (CF), we studied prepubertal and postpubertal males with CF versus an age-matched group of males with type 1 diabetes as controls. Design: Sera from CF and diabetic males treated at University Hospital, State University of New York, Stony Brook, were tested by indirect immunobead binding for antisperm antibodies and by radioimmunoassay for testosterone (T), luteinizing hormone, and follicle-stimulating hormone. The finding of autoantibodies to spermatozoa was correlated with chronological age, as well as with clinical and hormonal pubertal status. Results: Autoimmunity to sperm, as detected by humoral antisperm antibodies, was documented solely in postpubertal males, as judged by hormonal and clinical criteria. Eighty-three percent of sexually mature CF males and 6.3% (1 of 16) diabetic males exhibited autoantibodies to sperm. These antibodies were only detected when serum T levels were >8.7 nmol/L (250 ng/dL). Conclusions: These results suggest that puberty, and presumably, active spermatogenesis is a requirement for the development of autoimmunity to sperm in men with CF. Fertil SterilI992;58:1199-204 Key Words: Cystic fibrosis, congenital absence of vas deferens, antisperm antibodies
In 1959, Rumke and Hellinga (1) initially reported an association between acquired obstruction of a portion ofthe ductal system ofthe male reproductive tract and autoantibodies to spermatozoa. They speculated that extravasation of spermatozoa into the interstitial tissue might stimulate antibody pro-
Received March 6, 1992; revised and accepted August 13, 1992. * Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Health Sciences Center, SUNY. t Reprint requests: Richard A. Bronson, M.D., Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Health Sciences Center, T-9, 060 SUNY, Stony Brook, New York 11794-8091. :j: Cystic Fibrosis Program, Department of Pediatrics, Health Sciences Center, SUNY. § Pediatric Endocrinology, Department of Pediatrics, Health Sciences Center, SUNY. II Department of Pathology, Health Sciences Center, SUNY. 11 Department of Pediatrics, Maimonides Medical Center. Vol. 58, No.6, December 1992
duction. Subsequently, several investigators have documented an association between autoimmunity to spermatozoa and ductal obstruction, whether acquired secondary to infection or sterilization by vasectomy. Ansbacher (2) provided further evidence that vasectomy caused autoimmunity to spermatozoa, by documenting prospectively the development of agglutinating antisperm antibodies in sera of men after sterilization. An association of congenital bilateral absence of the vas deferens and autoimmunity to spermatozoa was initially reported by Amelar and associates in 1974 (3). In their practice, 101 of 5,112 patients who had undergone urologic evaluation were found to have congenital bilateral absence of the vasa deferentia. Sera from 29 of these men were submitted for sperm antibody testing, using a macroscopic gel agglutinization test. Significantly high titers, ranging from 1:32 to 1:4,096, were detected in 18 ofthese Bronson et al.
Autoimmunity to sperm in CF
1199
men (3). Other studies have confirmed this association between congenital obstruction of the vas deferens and autoimmunity to spermatozoa, although the frequency of those men demonstrating antisperm antibodies varied widely, ranging from 11 % to 62% (4). The vast majority of men with cystic fibrosis (CF) are sterile secondary to obstructive azoospermia because of abnormalities of the Wolfian duct derivatives (5). The head of the epididymis is usually spared, but absence and atrophy of the vas deferens, tail, and body ofthe epididymis and/or seminal vesicles are noted. Only rarely are totally normal ductal structures seen. In contrast, testicular volumes are normal, and on biopsy, Leydig cells appear normal in number and morphology. Active spermatogenesis is present within seminiferous tubules, and a full complement of spermatogonia has been noted to line the basement membrane. However, in some studies, abnormalities of spermatocytes and spermatids have been reported, and a diminished total number of mature sperm observed. D'Cruz et al. (6) have recently shown an association between autoimmunity to sperm and CF. However, we are unaware of any study that has attempted to answer the question of whether the development of autoimmunity to spermatozoa in these men is related to the onset of spermatogenesis at puberty. Indeed, only a single report has been published in which the incidence of autoimmunity to spermatozoa has been studied during childhood (7). In this study, sera were obtained from unselected males with CF and males with type 1 diabetes mellitus who served as controls. This latter group was chosen because there has been no prior report of an association of autoimmunity to spermatozoa and type 1 diabetes mellitus. MATERIALS AND METHODS
All of the patients with CF were under the care of one of the coauthors (W.J.O.), whereas the diabetic patients were managed by T.A.W. In addition to a general history and physical examination, pubertal status was judged using Tanner (8) score and the Shwachman-Kulczycki clinical scoring system, modified by Doershuk et al. (9) was used to assess CF. This clinical scoring system evaluates four parameters: [1] case history; [2] pulmonary physical findings and cough; [3] growth and nutrition; and [4] chest roentgenogram. One to 25 points are awarded to each item, and the maximum score attainable is 100. The higher the score, the better the 1200
Bronson et al. Autoimmunity to sperm in CF
patient's condition. Testicular volume was quantitated using a Prader orchidometer (Dalzell USA Medical Systems, Park Forest, IL). The study was approved by the Committee on Research Involving Human Subjects, and written consent was obtained. Sperm Antibody Studies
Sera were obtained from unselected males with type 1 diabetes mellitus and stored at -70°C until tested in batches. A single donor with normal semen parameters (10), whose spermatozoa were judged to be free of autoantibodies by direct immunobead binding was used in an indirect immunobead binding assay, to document the presence of circulating antisperm antibodies in the test sera (11). Sera were initially screened at 1:10 dilution. A highly motile fraction of spermatozoa obtained from the ejaculate by swim-up were incubated in dilute sera (1:10) for 1 hour at 37°C, washed free of serum, and mixed with a suspension of immunobeads. The proportion of motile spermatozoa binding isotype-specific immunobeads was scored for immunoglobulins (Ig) of the IgG, IgA, and IgM classes. The location of immunobead rosetting over the sperm head and tail was also noted. When 100% of spermatozoa were found to be antibody-labeled after their exposure to test serum, these sera were serially diluted and retested to determine an antibody titre. Hormonal Testing
Luteinizing hormone (LH) and follicle-stimulating hormone (FSH) were measured with reagents purchased from Diagnostic Products Corporation (Los Angeles, CA) and used according to their recommended protocol. Testosterone (T) was measured by the method used for the assay of corticosterone (12) with the following modifications: [1] antibody was generated in rabbits with T -6-carboxymethyl oxime: bovine serum albumin (obtained from Research Plus (Bayonne, NJ) as the antigen and [2] the tracer was [1,2,6,7 H 3 ]T obtained from Dupont (Cambridge, MA). Each ofthe assay procedures had 14% interassay coefficient of (CV) variation and 12% intra-assay CV; each assay set included three external quality control samples (Lyphochek I, II, III, lot No. 7000; BioRad, Richmond, CA). Statistical Methods
Comparison of hormonal values between men with and without antisperm antibodies was by nonpaired Student's t-test. Statistical analysis of the presence or absence of antisperm antibodies in postpubertal Fertility and Sterility
Table 1
Pubertal Status and Detection of Antisperm Antibodies in 15 Males with CF Tanner score
Patient no.
Age
Antibody-positive group 1 2 3 4 5 6 Mean± SEM Antibody-negative group 1 2 3 4 5 6 7 8 9 Mean ± SEM
Genitalia
Pubic hair
Testicular volume
Serum T
mL
nmollL
Clinical CF
4.2 24.8 10.5
10.3 3.1 7.2
10.0 13.1 12.5 ± 3.4
7.0 4.8 6.5 ± 1.2
87 73 74 96 41 90
lUlL
33 29 25 25 20 18
5 5 5 5 5 5
5 5 5 5 5 5
20 20 20 20 20 20 20 ± 0
11.1 9.3 12.1 18.7 9.5 12.9 12.3 ± 1.4
19 15 14 14 12 10 10 9 9
5 3 3 3 2 1 1 1 2
5 3 2 2 2 2 1 1 2
20 10 6
3.4 16.4 3.1 7.6 0.35 1.6 1.1 0.7 2.1 4.0 ± 1.7
5 3 4 3 4 7±2
Serum LH
Serum FSH
6.4 7.4 1.0 9.2 3.5 4.5 1.8
3.8 4.9 2.4 2.7 2.1 2.4 1.7
3.3 4.6 ± 0.99
2.6 2.9 ± 0.35
74 76 78 95 80 87 95 78 95
* Mean serum T (P < 0.01), LH (P < 0.01), and FSH (P < 0.05, Student's t-test) were significantly higher in the antibody positive group compared with the antibody-negative group.
males with CF versus diabetes mellitus was by Fisher's exact test.
RESULTS Of the 15 male CF patients studied, 7 men had Tanner scores of 5 for genital characteristics and pubic hair, and testicular volume measured 20 mL bilaterally. Their ages ranged from 19 to 33 years. Six of the 7 had antisperm antibodies in their serum (Table 1). Five of these 6 postpubertal males with autoimmunity to sperm had high levels of circulating antisperm antibodies. All donor spermatozoa were found to be coated over their entire surface with Igs of the IgG class after their incubation in these sera. Antisperm antibodies of the IgA and IgM classes were also detected but at lower levels (Table 2). The sixth patient had lower levels of antisperm antibodies but was currently under active immunosuppression in association with a liver transplant. The one 19-year-old male with CF who was postpubertal by clinical examination, yet did not exhibit autoimmunity to spermatozoa, had a low serum T concentration (3.4 nmoljL). The presence of autoimmunity to spermatozoa did not correlate with the clinical status of CF, as reflected in the modified Shwachman-Kulczycki score (Table 1). In contrast to the males with CF, antisperm antibodies were detected in the serum from only 1 of Vol. 58, No.6, December 1992
16 postpubertal diabetic men. This individual has been found to have a normal sweat test for CF but has not undergone genetic testing for CF heterozygosity nor semen analysis. Among all subjects whose serum T concentrations were >8.7 nmoljL, a midpubertal value consistent with the onset of spermatogenesis (spermarche), the incidence of antisperm antibodies was significantly greater among patients with CF as compared with patients with type 1 diabetes (Fisher's exact test, P < 0.0045). None of the 16 prepubertal diabetic males nor 8 prepubertal males with CF was found to exhibit au-
Table 2 Immunoglobulin Class and Regional Binding on the Spermatozoan Surface of Antisperm Autoantibodies Detected in Six Postpubertal Men with CF* IgG Patient no. 1 2 3 4 5 6
IgM
IgA
Head
Tail
Head
Tail
+++
+++
++
++
++ +
+++ +++ +++ +++
++ +++ ++ +++
++ ++ ++ +
++ ++ +
Head
Tail
* As detected by indirect immunobead binding. + = >20% motile spermatozoa positive for surface Igs. ++ = >50% positive; +++ = >90% positive. Patient number 2 was under active immunosuppression in association with a liver transplant. Bronson et aI.
Autoimmunity to sperm in CF
1201
toimmunity to spermatozoa. The 15-year-old CF male, whose serum Twas >8.7 nmoljL (16.4 nmoljL) but whose antisperm antibody status was negative, had not completed puberty, as judged by clinical criteria (Tanner stages 3/3 for genitalia and pubic hair, and testicular volume 10 mL). The testicular volume in this individual was approximately half that of a mature male, suggesting that spermatogenesis was not yet active. Because delayed sexual maturation is a common observation in CF, the findings of autoimmunity to sperm were also correlated with hormonal endocrine status, as well as chronological age. Mean (±SEM) serum T in the antibody-positive group was 12.3 ± 1.4 nmoljL versus 4.0 ± 1.7 nmoljL in the antibody-negative group (P < 0.01, Student's t-test). Mean gonadotropin levels were higher in the group with autoimmunity to sperm (FSH/LH 12.5 ± 3.4/6.5 ± 1.2 IU/L) than the antibody-negative group (FSH/LH 4.6 ± 0.99/2.9 ± 0.35 IU/L) of men with CF, consistent with their more advanced stage of puberty. Serum T levels for CF males with autoimmunity to sperm did not differ significantly from those of the postpubertal diabetic males (15.2 ± 1.2 nmol/L versus 12.3 ± 1.4 nmoljL, Student's t-test). Antisperm antibodies detected in all men with CF exhibited regional specificities for the sperm surface, as reflected in immunobead binding patterns that were similar to those seen in men with other etiologies of autoimmunity to sperm (13). Immunoglobulins of the IgG class directed against antigens on the sperm head and tail were most prominent (Table 2). DISCUSSION
The onset of sperm production (spermarche), as judged by the initial appearance of spermatozoa in the first voided morning urine, occurs relatively early in the developmental process at midpuberty. Hence, Nielsen et al. (14) have observed that the smallest testicular volumes associated with the presence of spermaturia were 4 to 6 mL with a median of 11.5 mL, consistent with mean serum T concentrations ~ 8.7 nmoljL (15). The correlation of autoimmunity to spermatozoa with later pubertal signs in men with CF suggests that active spermatogenesis is a requirement for the development of antisperm antibodies. Studies using monoclonal antibodies in subhuman primates have shown that new antigens are expressed on developing spermatocytes and spermatids after the initiation of spermatogenesis (16). These 1202
Bronson et al.
Autoimmunity to sperm in CF
antigens, to which the immune system may not be tolerant, could playa role in the genesis of autoimmunity to sperm. It has been postulated that the development of autoimmunity to spermatozoa may be prevented by sequestration of auto antigens on germ cells by the presence ofthe blood-testis barrier. That such a barrier exists has been documented by the presence of tight junctions between Sertoli cells and the inability of macromolecules to enter the luminal compartment of the seminiferous tubule after their intravenous administration (17). There is also recent evidence in support of active local immunoregulatory mechanisms that may be operative (18). Because unique autoantigens are expressed on cells ofthe basal compartment of the seminiferous tubule, which are accessible to antigen processing cells (19), most investigators have speculated that autoimmunity to sperm after vasectomy broaches both the sequestration of antigens as well as these local immunoregulatory mechanisms (20). The immunobead binding technique used in this study allows one to judge whether Igs are present on the surface of living spermatozoa. In a prior study, Tung and associates (7) documented the presence of antisperm antibodies in sera of men and women of all ages when tested against methanol fixed, permeabilized spermatozoa. Evidence was provided that these naturally occurring antibodies were directed against subsurface antigens of spermatozoa that cross-reacted with epitopes present on bacteria, suggesting the possibility that these antibodies are not directed primarily against sperm. The present investigation, in contrast, provides evidence that antibodies directed against epitopes expressed on the sperm surface are not commonly seen in children, suggesting that pubertal development may play a role in the evolution of these antibodies. In men with congenital obstructive azoospermia secondary to CF, the immune system may become exposed to developmental antigens expressed on spermatocytes and spermatids to which it is not tolerant after activation ofthe pituitary-testicular axis and the initiation of spermatogenesis (16). Alternatively, T-dependent, epididymally derived antigens might also be secreted at puberty and be involved in the etiology of autoimmunity to sperm. The corpus and cauda of the epididymis secrete substances that modify the surface of spermatozoa during their extratesticular passage through the epididymis (21). In this instance, the location of the congenital block to sperm egress and the presence or absence of the corpus of the epididymis might be Fertility and Sterility
a determinant in the etiology of autoimmunity to sperm. Recent work presented by Patrizio and associates (22) lends support to the hypothesis that the expression of developmental antigens on spermatozoa is a critical factor in the etiology of autoimmunity to sperm. Antibody-negative spermatozoa were aspirated from the efferent ducts of the testis and proximal head of the epididymis in men with congenital obstructive azoospermia in preparation for in vitro fertilization and shown to be free of antisperm autoantibodies by direct immunobead binding. These spermatozoa that had not yet transited the epididymis were then exposed to a panel of sera obtained from infertile men with normal semen parameters, yet who had been shown to exhibit autoantibodies to ejaculate spermatozoa. The isotypes and regional binding patterns of Igs detected on the surface of the spermatozoa obtained from men with congenital obstructive azoospermia were no different than those seen when ejaculated sperm were exposed to the same sera concurrently. These results suggest that at least one if not several immunodominant epitopes are expressed on spermatozoa during their intratesticular development before their transit through the epididymis. An unanswered question is whether men with CF who manifest no evidence of autoimmunity to sperm possess spermatozoa in their ejaculates. A very small minority of men with CF have been found to be fertile. Taussig et al. (23) presented case reports of two such men with normal semen analysis, one of whom had fathered children (23). Although the majority of men after vasectomy develop antisperm antibodies, approximately one third do not, indicating that an acquired block to sperm egress does not necessarily lead to autoimmunity to sperm. Whether this lack of immunological response is on a genetic basis, related to specific immune response genes, or to local immunoregulatory factors within the male genital tract is yet to be determined. Nor is it known whether these data can be extrapolated to men with congenital absence of the vas deferens. Recently, Rijot and associates (24) screened 19 men with obstructive azoospermia secondary to aplasia of the epididymis and vas deferens for the delta F508 mutation of the CF gene. Seven of these men were heterozygous for that gene deletion. This raised the question of whether a large proportion of men with obstructive azoospermia might indeed exhibit subclinical CF or be heterozygotes for a CF gene. In a more recent report, Anguiano et al. (25) have extended these observations, demonstrating a Vol. 58, No.6, December 1992
high incidence of heterozygosity for gene mutations associated with CF in men with congenital absence of the vas deferens in the absence of clinically manifest disease. Given these findings, an important correlate of this study is that men with obstructive azoospermia should undergo testing for autoimmunity to sperm. Although causes of obstructive azoospermia other than congenital absence of the vas deferens exist, the failure to detect spermatozoa repeatedly in ejaculates of a postpubertal male, in association with the finding of normal serum FSH and T concentrations, should raise the possibility of CF. In this specific subgroup of men, the detection of autoimmunity to spermatozoa provides circumstantial evidence for the presence of active spermatogenesis and suggests that azoospermia was not because of an absence of spermatogonia within the seminiferous tubules (Sertoli cell only syndrome) nor a defect in sperm production. In these men, it might be reasonable to evaluate the possibility of the CF carrier state, by performing sweat testing and genetic analysis.
Acknowledgment. We thank Ms. Kathy O'Brien for her excellent work in preparation of the manuscript.
REFERENCES 1. Rumke P, Hellinga G. Auto-antibodies against spermatozoa in sterile men. Am J Clin PathoI1959;32:357-63. 2. Ansbacher R. Vasectomy: sperm antibodies. Fertil Steril 1973;24:788-92. 3. Amelar RD, Dubin L, Schoenfeld C. Circulating sperm agglutinating antibodies in azoospermic men with congenital bilateral absence of the vasa deferentia. Fertil Steril1975;26: 228-31. 4. Patrizio P, Moretti-Rojas I, Ord T, Balmaceda J, Silber S, Asch RH. Low incidence of antisperm antibodies in men with congenital absence of the vas deferens. Fertil Steril 1989;52: 1018-21. 5. Reiter EO, Stern RC, Root AW. The reproductive endocrine system in cystic fibrosis. Am J Dis Child 1981;135:422-6. 6. D'Cruz OJ, Haas GG Jr, de La Rocha R, Lambert H. Occurrence of serum antisperm antibodies in patients with cystic fibrosis. Fertil SteriI1991;56:519-27. 7. Tung KSK, Cooke WD Jr, McCarthy TA, Robitaille P. Human sperm antigens and antisperm antibodies II. Age-related incidence of antisperm antibodies. Clin Exp ImmunoI1976;25: 73-9. 8. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity and stages of puberty. Arch Dis Child 1976;51:170-6. 9. Doershuk C, Matthews L, Tucker AS. A 5 year clinical evaluation of a therapeutic program for patients with cystic fibrosis. J Pediatr 1964;65:677-93. 10. World Health Organization. Laboratory manual for the examination of human semen and semen cervical mucus inter-
Bronson et at.
Autoimmunity to sperm in CF
1203
11.
12. 13. 14.
15.
16.
17.
actions. 2nd ed. Cambridge: The Press Syndicate ofthe University of Cambridge, 1987:27. 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 1982;2:222-4. Chasalow FI, Blethen SL. Modulation of glucocorticoid secretion by growth hormone. Pediatr Res 1985;19:823-7. Bronson RA, Cooper GW, Rosenfeld DL. Sperm antibodies: their role in infertility. Fertil Steril 1984;42:176-83. Nielsen CT, Skakkebaek NE, Richardson DW, Darling JAB, Hunter WM, Jorgensen M, et al. Onset of the release of spermatazoa (spermarche) in boys in relation to age, testicular growth, pubic hair and height. J Clin Endocrinol Metab 1986;62:532-6. Kulin HE, Frontera MA, Demers LM, Bartholomew MS, Lloyd T A. The onset of sperm production in pubertal boys: relationship to gonadotropin excretion. Am J Dis Child 1989;143:190-3. Isahakia MA. Characterization of baboon testicular antigens using monoclonal anti-sperm antibodies. BioI Reprod 1988;39: 889-99. Dym M, Caviacchia JC. Further observations on the bloodtestis barrier in monkeys. BioI Reprod 1977;17:390-403.
1204
Bronson et al.
Autoimmunity to sperm in CF
18. Ritchie A WS. Intraepitheliallymphocytes in the normal epididymis: a mechanism for tolerance to sperm auto-antigens. Br J Urol 1984;56:79-83. 19. Mahi-Brown CA, Yule TD, Tung KSK. Evidence for acute immunological regulation in prevention of testicular autoimmune disease independent of the blood-testis barrier. Am J Reprod ImmunoI1988;16:165-70. 20. Alexander NJ, Anderson DJ. Immunology of semen. Fertil SteriI1987;47:192-205. 21. Olson GE, Oregbin-Crist MC. Sperm surface changes during epididymal maturation. Ann NY Acad Sci 1982;83:372-9. 22. Patrizio P, Bronson R, Silber SJ, Ord T, Asch RH. Testicular origin of immunobead reacting antigens on human sperm. Fertil SteriI1992;57:183-6. 23. Taussig LM, Lobeck CC, di Sant' Agnese P A, Ackerman DR, Kauttwinkel J. Fertility in males with cystic fibrosis. N Engl J Med 1972;287:586-92. 24. Rigot JM, Lafitte JJ, Dumur V, Gervais R, Manouvrier S, Biserte J, et al. Cystic fibrosis and congenital absence of the vas deferens. N Engl J Med 1991;325:64-5. 25. Anguiano A, Oates RD, Amos JA, Dean M, Gerrard B, Stewart C, et al. Congenital bilateral absence of the vas deferens: a primary genital form of cystic fibrosis. J Am Med Assoc 1992;267:1994-7.
Fertility and Sterility
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright © 1992 The American Fertility Society
Correlation between puberty and the development of autoimmunity to spermatozoa in men with cystic fibrosis
Richard A. Bronson, M.D.*t Walter J. O'Connor, M.D.:j: Thomas A. Wilson, M.D.§
Susan K. Bronson, M.P.H.II Fred I. Chasalow, Ph.D.~ Kathleen Droesch, M.D.*
State University of New York (SUNY) at Stony Brook, Stony Brook, and Maimonides Medical Center, Brooklyn, New York
Objective: To test the hypothesis that puberty is a necessary factor in the pathogenesis of autoimmunity to sperm in men with cystic fibrosis (CF), we studied prepubertal and postpubertal males with CF versus an age-matched group of males with type 1 diabetes as controls. Design: Sera from CF and diabetic males treated at University Hospital, State University of New York, Stony Brook, were tested by indirect immunobead binding for antisperm antibodies and by radioimmunoassay for testosterone (T), luteinizing hormone, and follicle-stimulating hormone. The finding of autoantibodies to spermatozoa was correlated with chronological age, as well as with clinical and hormonal pubertal status. Results: Autoimmunity to sperm, as detected by humoral antisperm antibodies, was documented solely in postpubertal males, as judged by hormonal and clinical criteria. Eighty-three percent of sexually mature CF males and 6.3% (1 of 16) diabetic males exhibited autoantibodies to sperm. These antibodies were only detected when serum T levels were >8.7 nmol/L (250 ng/dL). Conclusions: These results suggest that puberty, and presumably, active spermatogenesis is a requirement for the development of autoimmunity to sperm in men with CF. Fertil SterilI992;58:1199-204 Key Words: Cystic fibrosis, congenital absence of vas deferens, antisperm antibodies
In 1959, Rumke and Hellinga (1) initially reported an association between acquired obstruction of a portion ofthe ductal system ofthe male reproductive tract and autoantibodies to spermatozoa. They speculated that extravasation of spermatozoa into the interstitial tissue might stimulate antibody pro-
Received March 6, 1992; revised and accepted August 13, 1992. * Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Health Sciences Center, SUNY. t Reprint requests: Richard A. Bronson, M.D., Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Health Sciences Center, T-9, 060 SUNY, Stony Brook, New York 11794-8091. :j: Cystic Fibrosis Program, Department of Pediatrics, Health Sciences Center, SUNY. § Pediatric Endocrinology, Department of Pediatrics, Health Sciences Center, SUNY. II Department of Pathology, Health Sciences Center, SUNY. 11 Department of Pediatrics, Maimonides Medical Center. Vol. 58, No.6, December 1992
duction. Subsequently, several investigators have documented an association between autoimmunity to spermatozoa and ductal obstruction, whether acquired secondary to infection or sterilization by vasectomy. Ansbacher (2) provided further evidence that vasectomy caused autoimmunity to spermatozoa, by documenting prospectively the development of agglutinating antisperm antibodies in sera of men after sterilization. An association of congenital bilateral absence of the vas deferens and autoimmunity to spermatozoa was initially reported by Amelar and associates in 1974 (3). In their practice, 101 of 5,112 patients who had undergone urologic evaluation were found to have congenital bilateral absence of the vasa deferentia. Sera from 29 of these men were submitted for sperm antibody testing, using a macroscopic gel agglutinization test. Significantly high titers, ranging from 1:32 to 1:4,096, were detected in 18 ofthese Bronson et al.
Autoimmunity to sperm in CF
1199
men (3). Other studies have confirmed this association between congenital obstruction of the vas deferens and autoimmunity to spermatozoa, although the frequency of those men demonstrating antisperm antibodies varied widely, ranging from 11 % to 62% (4). The vast majority of men with cystic fibrosis (CF) are sterile secondary to obstructive azoospermia because of abnormalities of the Wolfian duct derivatives (5). The head of the epididymis is usually spared, but absence and atrophy of the vas deferens, tail, and body ofthe epididymis and/or seminal vesicles are noted. Only rarely are totally normal ductal structures seen. In contrast, testicular volumes are normal, and on biopsy, Leydig cells appear normal in number and morphology. Active spermatogenesis is present within seminiferous tubules, and a full complement of spermatogonia has been noted to line the basement membrane. However, in some studies, abnormalities of spermatocytes and spermatids have been reported, and a diminished total number of mature sperm observed. D'Cruz et al. (6) have recently shown an association between autoimmunity to sperm and CF. However, we are unaware of any study that has attempted to answer the question of whether the development of autoimmunity to spermatozoa in these men is related to the onset of spermatogenesis at puberty. Indeed, only a single report has been published in which the incidence of autoimmunity to spermatozoa has been studied during childhood (7). In this study, sera were obtained from unselected males with CF and males with type 1 diabetes mellitus who served as controls. This latter group was chosen because there has been no prior report of an association of autoimmunity to spermatozoa and type 1 diabetes mellitus. MATERIALS AND METHODS
All of the patients with CF were under the care of one of the coauthors (W.J.O.), whereas the diabetic patients were managed by T.A.W. In addition to a general history and physical examination, pubertal status was judged using Tanner (8) score and the Shwachman-Kulczycki clinical scoring system, modified by Doershuk et al. (9) was used to assess CF. This clinical scoring system evaluates four parameters: [1] case history; [2] pulmonary physical findings and cough; [3] growth and nutrition; and [4] chest roentgenogram. One to 25 points are awarded to each item, and the maximum score attainable is 100. The higher the score, the better the 1200
Bronson et al. Autoimmunity to sperm in CF
patient's condition. Testicular volume was quantitated using a Prader orchidometer (Dalzell USA Medical Systems, Park Forest, IL). The study was approved by the Committee on Research Involving Human Subjects, and written consent was obtained. Sperm Antibody Studies
Sera were obtained from unselected males with type 1 diabetes mellitus and stored at -70°C until tested in batches. A single donor with normal semen parameters (10), whose spermatozoa were judged to be free of autoantibodies by direct immunobead binding was used in an indirect immunobead binding assay, to document the presence of circulating antisperm antibodies in the test sera (11). Sera were initially screened at 1:10 dilution. A highly motile fraction of spermatozoa obtained from the ejaculate by swim-up were incubated in dilute sera (1:10) for 1 hour at 37°C, washed free of serum, and mixed with a suspension of immunobeads. The proportion of motile spermatozoa binding isotype-specific immunobeads was scored for immunoglobulins (Ig) of the IgG, IgA, and IgM classes. The location of immunobead rosetting over the sperm head and tail was also noted. When 100% of spermatozoa were found to be antibody-labeled after their exposure to test serum, these sera were serially diluted and retested to determine an antibody titre. Hormonal Testing
Luteinizing hormone (LH) and follicle-stimulating hormone (FSH) were measured with reagents purchased from Diagnostic Products Corporation (Los Angeles, CA) and used according to their recommended protocol. Testosterone (T) was measured by the method used for the assay of corticosterone (12) with the following modifications: [1] antibody was generated in rabbits with T -6-carboxymethyl oxime: bovine serum albumin (obtained from Research Plus (Bayonne, NJ) as the antigen and [2] the tracer was [1,2,6,7 H 3 ]T obtained from Dupont (Cambridge, MA). Each ofthe assay procedures had 14% interassay coefficient of (CV) variation and 12% intra-assay CV; each assay set included three external quality control samples (Lyphochek I, II, III, lot No. 7000; BioRad, Richmond, CA). Statistical Methods
Comparison of hormonal values between men with and without antisperm antibodies was by nonpaired Student's t-test. Statistical analysis of the presence or absence of antisperm antibodies in postpubertal Fertility and Sterility
Table 1
Pubertal Status and Detection of Antisperm Antibodies in 15 Males with CF Tanner score
Patient no.
Age
Antibody-positive group 1 2 3 4 5 6 Mean± SEM Antibody-negative group 1 2 3 4 5 6 7 8 9 Mean ± SEM
Genitalia
Pubic hair
Testicular volume
Serum T
mL
nmollL
Clinical CF
4.2 24.8 10.5
10.3 3.1 7.2
10.0 13.1 12.5 ± 3.4
7.0 4.8 6.5 ± 1.2
87 73 74 96 41 90
lUlL
33 29 25 25 20 18
5 5 5 5 5 5
5 5 5 5 5 5
20 20 20 20 20 20 20 ± 0
11.1 9.3 12.1 18.7 9.5 12.9 12.3 ± 1.4
19 15 14 14 12 10 10 9 9
5 3 3 3 2 1 1 1 2
5 3 2 2 2 2 1 1 2
20 10 6
3.4 16.4 3.1 7.6 0.35 1.6 1.1 0.7 2.1 4.0 ± 1.7
5 3 4 3 4 7±2
Serum LH
Serum FSH
6.4 7.4 1.0 9.2 3.5 4.5 1.8
3.8 4.9 2.4 2.7 2.1 2.4 1.7
3.3 4.6 ± 0.99
2.6 2.9 ± 0.35
74 76 78 95 80 87 95 78 95
* Mean serum T (P < 0.01), LH (P < 0.01), and FSH (P < 0.05, Student's t-test) were significantly higher in the antibody positive group compared with the antibody-negative group.
males with CF versus diabetes mellitus was by Fisher's exact test.
RESULTS Of the 15 male CF patients studied, 7 men had Tanner scores of 5 for genital characteristics and pubic hair, and testicular volume measured 20 mL bilaterally. Their ages ranged from 19 to 33 years. Six of the 7 had antisperm antibodies in their serum (Table 1). Five of these 6 postpubertal males with autoimmunity to sperm had high levels of circulating antisperm antibodies. All donor spermatozoa were found to be coated over their entire surface with Igs of the IgG class after their incubation in these sera. Antisperm antibodies of the IgA and IgM classes were also detected but at lower levels (Table 2). The sixth patient had lower levels of antisperm antibodies but was currently under active immunosuppression in association with a liver transplant. The one 19-year-old male with CF who was postpubertal by clinical examination, yet did not exhibit autoimmunity to spermatozoa, had a low serum T concentration (3.4 nmoljL). The presence of autoimmunity to spermatozoa did not correlate with the clinical status of CF, as reflected in the modified Shwachman-Kulczycki score (Table 1). In contrast to the males with CF, antisperm antibodies were detected in the serum from only 1 of Vol. 58, No.6, December 1992
16 postpubertal diabetic men. This individual has been found to have a normal sweat test for CF but has not undergone genetic testing for CF heterozygosity nor semen analysis. Among all subjects whose serum T concentrations were >8.7 nmoljL, a midpubertal value consistent with the onset of spermatogenesis (spermarche), the incidence of antisperm antibodies was significantly greater among patients with CF as compared with patients with type 1 diabetes (Fisher's exact test, P < 0.0045). None of the 16 prepubertal diabetic males nor 8 prepubertal males with CF was found to exhibit au-
Table 2 Immunoglobulin Class and Regional Binding on the Spermatozoan Surface of Antisperm Autoantibodies Detected in Six Postpubertal Men with CF* IgG Patient no. 1 2 3 4 5 6
IgM
IgA
Head
Tail
Head
Tail
+++
+++
++
++
++ +
+++ +++ +++ +++
++ +++ ++ +++
++ ++ ++ +
++ ++ +
Head
Tail
* As detected by indirect immunobead binding. + = >20% motile spermatozoa positive for surface Igs. ++ = >50% positive; +++ = >90% positive. Patient number 2 was under active immunosuppression in association with a liver transplant. Bronson et aI.
Autoimmunity to sperm in CF
1201
toimmunity to spermatozoa. The 15-year-old CF male, whose serum Twas >8.7 nmoljL (16.4 nmoljL) but whose antisperm antibody status was negative, had not completed puberty, as judged by clinical criteria (Tanner stages 3/3 for genitalia and pubic hair, and testicular volume 10 mL). The testicular volume in this individual was approximately half that of a mature male, suggesting that spermatogenesis was not yet active. Because delayed sexual maturation is a common observation in CF, the findings of autoimmunity to sperm were also correlated with hormonal endocrine status, as well as chronological age. Mean (±SEM) serum T in the antibody-positive group was 12.3 ± 1.4 nmoljL versus 4.0 ± 1.7 nmoljL in the antibody-negative group (P < 0.01, Student's t-test). Mean gonadotropin levels were higher in the group with autoimmunity to sperm (FSH/LH 12.5 ± 3.4/6.5 ± 1.2 IU/L) than the antibody-negative group (FSH/LH 4.6 ± 0.99/2.9 ± 0.35 IU/L) of men with CF, consistent with their more advanced stage of puberty. Serum T levels for CF males with autoimmunity to sperm did not differ significantly from those of the postpubertal diabetic males (15.2 ± 1.2 nmol/L versus 12.3 ± 1.4 nmoljL, Student's t-test). Antisperm antibodies detected in all men with CF exhibited regional specificities for the sperm surface, as reflected in immunobead binding patterns that were similar to those seen in men with other etiologies of autoimmunity to sperm (13). Immunoglobulins of the IgG class directed against antigens on the sperm head and tail were most prominent (Table 2). DISCUSSION
The onset of sperm production (spermarche), as judged by the initial appearance of spermatozoa in the first voided morning urine, occurs relatively early in the developmental process at midpuberty. Hence, Nielsen et al. (14) have observed that the smallest testicular volumes associated with the presence of spermaturia were 4 to 6 mL with a median of 11.5 mL, consistent with mean serum T concentrations ~ 8.7 nmoljL (15). The correlation of autoimmunity to spermatozoa with later pubertal signs in men with CF suggests that active spermatogenesis is a requirement for the development of antisperm antibodies. Studies using monoclonal antibodies in subhuman primates have shown that new antigens are expressed on developing spermatocytes and spermatids after the initiation of spermatogenesis (16). These 1202
Bronson et al.
Autoimmunity to sperm in CF
antigens, to which the immune system may not be tolerant, could playa role in the genesis of autoimmunity to sperm. It has been postulated that the development of autoimmunity to spermatozoa may be prevented by sequestration of auto antigens on germ cells by the presence ofthe blood-testis barrier. That such a barrier exists has been documented by the presence of tight junctions between Sertoli cells and the inability of macromolecules to enter the luminal compartment of the seminiferous tubule after their intravenous administration (17). There is also recent evidence in support of active local immunoregulatory mechanisms that may be operative (18). Because unique autoantigens are expressed on cells ofthe basal compartment of the seminiferous tubule, which are accessible to antigen processing cells (19), most investigators have speculated that autoimmunity to sperm after vasectomy broaches both the sequestration of antigens as well as these local immunoregulatory mechanisms (20). The immunobead binding technique used in this study allows one to judge whether Igs are present on the surface of living spermatozoa. In a prior study, Tung and associates (7) documented the presence of antisperm antibodies in sera of men and women of all ages when tested against methanol fixed, permeabilized spermatozoa. Evidence was provided that these naturally occurring antibodies were directed against subsurface antigens of spermatozoa that cross-reacted with epitopes present on bacteria, suggesting the possibility that these antibodies are not directed primarily against sperm. The present investigation, in contrast, provides evidence that antibodies directed against epitopes expressed on the sperm surface are not commonly seen in children, suggesting that pubertal development may play a role in the evolution of these antibodies. In men with congenital obstructive azoospermia secondary to CF, the immune system may become exposed to developmental antigens expressed on spermatocytes and spermatids to which it is not tolerant after activation ofthe pituitary-testicular axis and the initiation of spermatogenesis (16). Alternatively, T-dependent, epididymally derived antigens might also be secreted at puberty and be involved in the etiology of autoimmunity to sperm. The corpus and cauda of the epididymis secrete substances that modify the surface of spermatozoa during their extratesticular passage through the epididymis (21). In this instance, the location of the congenital block to sperm egress and the presence or absence of the corpus of the epididymis might be Fertility and Sterility
a determinant in the etiology of autoimmunity to sperm. Recent work presented by Patrizio and associates (22) lends support to the hypothesis that the expression of developmental antigens on spermatozoa is a critical factor in the etiology of autoimmunity to sperm. Antibody-negative spermatozoa were aspirated from the efferent ducts of the testis and proximal head of the epididymis in men with congenital obstructive azoospermia in preparation for in vitro fertilization and shown to be free of antisperm autoantibodies by direct immunobead binding. These spermatozoa that had not yet transited the epididymis were then exposed to a panel of sera obtained from infertile men with normal semen parameters, yet who had been shown to exhibit autoantibodies to ejaculate spermatozoa. The isotypes and regional binding patterns of Igs detected on the surface of the spermatozoa obtained from men with congenital obstructive azoospermia were no different than those seen when ejaculated sperm were exposed to the same sera concurrently. These results suggest that at least one if not several immunodominant epitopes are expressed on spermatozoa during their intratesticular development before their transit through the epididymis. An unanswered question is whether men with CF who manifest no evidence of autoimmunity to sperm possess spermatozoa in their ejaculates. A very small minority of men with CF have been found to be fertile. Taussig et al. (23) presented case reports of two such men with normal semen analysis, one of whom had fathered children (23). Although the majority of men after vasectomy develop antisperm antibodies, approximately one third do not, indicating that an acquired block to sperm egress does not necessarily lead to autoimmunity to sperm. Whether this lack of immunological response is on a genetic basis, related to specific immune response genes, or to local immunoregulatory factors within the male genital tract is yet to be determined. Nor is it known whether these data can be extrapolated to men with congenital absence of the vas deferens. Recently, Rijot and associates (24) screened 19 men with obstructive azoospermia secondary to aplasia of the epididymis and vas deferens for the delta F508 mutation of the CF gene. Seven of these men were heterozygous for that gene deletion. This raised the question of whether a large proportion of men with obstructive azoospermia might indeed exhibit subclinical CF or be heterozygotes for a CF gene. In a more recent report, Anguiano et al. (25) have extended these observations, demonstrating a Vol. 58, No.6, December 1992
high incidence of heterozygosity for gene mutations associated with CF in men with congenital absence of the vas deferens in the absence of clinically manifest disease. Given these findings, an important correlate of this study is that men with obstructive azoospermia should undergo testing for autoimmunity to sperm. Although causes of obstructive azoospermia other than congenital absence of the vas deferens exist, the failure to detect spermatozoa repeatedly in ejaculates of a postpubertal male, in association with the finding of normal serum FSH and T concentrations, should raise the possibility of CF. In this specific subgroup of men, the detection of autoimmunity to spermatozoa provides circumstantial evidence for the presence of active spermatogenesis and suggests that azoospermia was not because of an absence of spermatogonia within the seminiferous tubules (Sertoli cell only syndrome) nor a defect in sperm production. In these men, it might be reasonable to evaluate the possibility of the CF carrier state, by performing sweat testing and genetic analysis.
Acknowledgment. We thank Ms. Kathy O'Brien for her excellent work in preparation of the manuscript.
REFERENCES 1. Rumke P, Hellinga G. Auto-antibodies against spermatozoa in sterile men. Am J Clin PathoI1959;32:357-63. 2. Ansbacher R. Vasectomy: sperm antibodies. Fertil Steril 1973;24:788-92. 3. Amelar RD, Dubin L, Schoenfeld C. Circulating sperm agglutinating antibodies in azoospermic men with congenital bilateral absence of the vasa deferentia. Fertil Steril1975;26: 228-31. 4. Patrizio P, Moretti-Rojas I, Ord T, Balmaceda J, Silber S, Asch RH. Low incidence of antisperm antibodies in men with congenital absence of the vas deferens. Fertil Steril 1989;52: 1018-21. 5. Reiter EO, Stern RC, Root AW. The reproductive endocrine system in cystic fibrosis. Am J Dis Child 1981;135:422-6. 6. D'Cruz OJ, Haas GG Jr, de La Rocha R, Lambert H. Occurrence of serum antisperm antibodies in patients with cystic fibrosis. Fertil SteriI1991;56:519-27. 7. Tung KSK, Cooke WD Jr, McCarthy TA, Robitaille P. Human sperm antigens and antisperm antibodies II. Age-related incidence of antisperm antibodies. Clin Exp ImmunoI1976;25: 73-9. 8. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity and stages of puberty. Arch Dis Child 1976;51:170-6. 9. Doershuk C, Matthews L, Tucker AS. A 5 year clinical evaluation of a therapeutic program for patients with cystic fibrosis. J Pediatr 1964;65:677-93. 10. World Health Organization. Laboratory manual for the examination of human semen and semen cervical mucus inter-
Bronson et at.
Autoimmunity to sperm in CF
1203
11.
12. 13. 14.
15.
16.
17.
actions. 2nd ed. Cambridge: The Press Syndicate ofthe University of Cambridge, 1987:27. 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 1982;2:222-4. Chasalow FI, Blethen SL. Modulation of glucocorticoid secretion by growth hormone. Pediatr Res 1985;19:823-7. Bronson RA, Cooper GW, Rosenfeld DL. Sperm antibodies: their role in infertility. Fertil Steril 1984;42:176-83. Nielsen CT, Skakkebaek NE, Richardson DW, Darling JAB, Hunter WM, Jorgensen M, et al. Onset of the release of spermatazoa (spermarche) in boys in relation to age, testicular growth, pubic hair and height. J Clin Endocrinol Metab 1986;62:532-6. Kulin HE, Frontera MA, Demers LM, Bartholomew MS, Lloyd T A. The onset of sperm production in pubertal boys: relationship to gonadotropin excretion. Am J Dis Child 1989;143:190-3. Isahakia MA. Characterization of baboon testicular antigens using monoclonal anti-sperm antibodies. BioI Reprod 1988;39: 889-99. Dym M, Caviacchia JC. Further observations on the bloodtestis barrier in monkeys. BioI Reprod 1977;17:390-403.
1204
Bronson et al.
Autoimmunity to sperm in CF
18. Ritchie A WS. Intraepitheliallymphocytes in the normal epididymis: a mechanism for tolerance to sperm auto-antigens. Br J Urol 1984;56:79-83. 19. Mahi-Brown CA, Yule TD, Tung KSK. Evidence for acute immunological regulation in prevention of testicular autoimmune disease independent of the blood-testis barrier. Am J Reprod ImmunoI1988;16:165-70. 20. Alexander NJ, Anderson DJ. Immunology of semen. Fertil SteriI1987;47:192-205. 21. Olson GE, Oregbin-Crist MC. Sperm surface changes during epididymal maturation. Ann NY Acad Sci 1982;83:372-9. 22. Patrizio P, Bronson R, Silber SJ, Ord T, Asch RH. Testicular origin of immunobead reacting antigens on human sperm. Fertil SteriI1992;57:183-6. 23. Taussig LM, Lobeck CC, di Sant' Agnese P A, Ackerman DR, Kauttwinkel J. Fertility in males with cystic fibrosis. N Engl J Med 1972;287:586-92. 24. Rigot JM, Lafitte JJ, Dumur V, Gervais R, Manouvrier S, Biserte J, et al. Cystic fibrosis and congenital absence of the vas deferens. N Engl J Med 1991;325:64-5. 25. Anguiano A, Oates RD, Amos JA, Dean M, Gerrard B, Stewart C, et al. Congenital bilateral absence of the vas deferens: a primary genital form of cystic fibrosis. J Am Med Assoc 1992;267:1994-7.
Fertility and Sterility