- Email: [email protected]
Immunoglobulin E levels and antisperm antibody titers in infertile couples
Immunoglobulin E levels and antisperm antibody titers in infertile couples S. MATHUR, H.
0.
WILLIAMSON,
E. R. BAKER, H.
H.
Charleston,
PH.D. M.D.
M.D.
FUDENBERG, South
M.D.
Carolina
lmmunoglobulin E (IgE) levels and antisperm antibody titers were determined in samples of serum and seminal plasma from 25 fertile men (Group A), 18 infertile men without measurable immunity to sperm (Group B), and 42 infertile men autoimmune to sperm (Group C), and in samples of serum and cervical and vaginal secretions from 25 fertile women (Group D), 28 infertile women without measurable immunity to sperm (Group E), and 32 infertile women isoimmune to sperm (Group F). Among the men, IgE levels in the serum, measured in international units per milliliter, were elevated in Group C (230 2 41, mean f SEM) as compared with Groups B (94 f 33, P < 0.05) and A (55 k 8, P < 0.001). In contrast, IgE levels in the seminal plasma in Group C (180 f 44) were not significantly different from those in Group B (48 + 21), but were higher than those in Group A (8 f 2, P < 0.05). In the women, serum IgE levels were higher (P < 0.001) in Group F, with isoimmunity to sperm (472 k 55) than in Groups E (219 + 32) and D (78 f 18). Wives of autoimmune husbands had somewhat, though not significantly, higher serum IgE levels (406 k 55) than had wives of nonautoimmune husbands (247 f 97). These results provide evidence for an elevated IgE response in subjects with significantly elevated antisperm antibody titers. (AM. J. OBSTET.
GYNECOL.
140:923,1981.)
A L LE R G Y to human seminal plasma in female subjects has been associated with urticaria, pharyngeal edema, and cardiovascular symptoms immediately after coitus.’ In such patients, skin tests with seminal plasma, but not with extracts of sperm, revealed immediate type hypersensitivity. The causative agent has been reported to be a protein of prostatic origin.* Since immunoglobulin E Publication No. 439 frm the Department of Basic and Clinical Immunology and Microbiology and the Section of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Medical University of South Carolina. Supported in part by United States Public Health Service grants HD-09938, CA-25746, and HDIAI-14365, and by United States Public Health Service National Research Service Award AI-07063. The Foundation Prize Basic Science Award presented the Forty-third .4nnual Meeting of the South Atlantic Association of Obstetricians and Gynecologists, Hot Springs, Virginia, January 25-28, 1981.
at
Reprint requests: Dr. Subbi Mathur, Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina, I71 Ashley Ave., Charleston, South Carolina 29425. 0002-9378/81/160923+08$00.80/0
0 1981 The C. V. Mosby
Co.
(IgE) is considered to be responsible for the anaphylactic reactions of antigen-exposed sensitized mast cells, leading to histamine production,3 information on the possible association of antisperm immunity and IgE levels in these subjects is of interest. Friberg4 has reported a lack of correlation between antisperm antibody titers and IgE levels in the sera and seminal plasma of infertile men, although increased levels of other serum immunoglobulins in men with antisperm antibodies have been reported.’ To date, no systematic study of the role of reaginic antibodies in immunologic infertility caused by antisperm immunity in male and female subjects has been undertaken, primarily because of a lack of effective screening methods for autoimmunity and isoimmunity to sperm. A new method, based on passive hemagglutination and Coombs test, has been established in our laboratory for the determination of antisperm antibody titers on sperm and in sera, seminal plasma, and cervical and vaginal secretions from couples with unexplained infertility.6 In the present study, we measured IgE levels and antisperm antibody titers in samples from fertile and infertile subjects. 923
924
Mathur
et al.
Table I. Serum IgE levels and antisperm antibody titers in men-fertile antisperm immunity (Group B), and infertile with antisperm immunity
(Group A), infertile (Group C)
without
Student’,i Group
No. studied IgE levels (IU/ml) Serum SP Antisperm antibody titers (ieciprocal) Sperm Polyvalent ‘gc,
@A w
Serum SP
A
Group
B
Group
/ thl
C
25
18
42
55 2 8* 822
94 + 33 48 iz 21
230 2 41
180 ? 44
NST P < 0.01
P < 0.05 NS
P < 0.001 P < 0.05
722 Sk2 15 ” 2 2&l 221 7+1
23 2 3
325 434 224 55 45
P = 0.1 NS NS NS NS
P< P< P< P< P< P<
P< P< P< P< P< P<
10 t 2 26 t 9 321 3kl
11 ir 1
+ 5 k 2 5
40 94 84 39 6
79 k 21
NS
0.001 0.001 0.001 0.001 0.05 0.025
0.001 0.001 0.001 0.001 0.05 O.ofll
*Mean ? SEM. tNS = Not significant.
Material and methods Patients. Studied were infertile couples who had unsuccessfully attempted conception for 2 to 10 years. The women were normal and ovulatory, with normal endocrine profiles, no tubal occlusion, and no signs of genital infections in the cervicovaginal area. The men were free of infections and had normal sperm counts and motility. Sixty couples were studied; 25 fertile men and 25 fertile women served as controls. When one or both partners had elevated antibody titers, they were re-evaluated at periodic intervals. All participants were informed of the nature and purpose of the study. Samples. Serum samples were obtained by venipuncture. Semen was obtained by masturbation after 2 to -I days of sexual abstinence. The e.jaculate was collected in a clean, wide-mouthed, glass jar and allowed to liquefy at room temperature for about 30 minutes before the assays. Seminal plasma was obtained by centrifuging the semen at 1,000 X g for 5 minutes and removing the supernatant. \‘iscous samples were centrifuged for 10 minutes. Vaginal fluid was collected b) swabbing away excessive leukorrhea with a clean, cotton-tipped applicator and then flooding the posterior fornix with 2 ml of sterile saline solution from a sterile syringe via a plastic cannula. The fluid was Hushed back and forth a total of four times before the final aspiration. Cervical mucus was obtained by inserting the tip of an endocervical aspirator cannula into the endocervix and withdrawing all the mucus that could be obtained. Viscous secretions were treated with 0.1% bromelin (Sigma Chemicals, St. Louis, Missouri) in order to liquefy them. The secretions were extracted with a known volume of saline solution at room temperature and centrifuged at 1,000 x g for 10 minutes to eliminate cells and debris. The clear supernatant that con-
tained the highly soluble immunoglobulins was used for study. Dilutions of the washings were calculated on the basis of the ratio of the original volume of mucus to the total volume after the addition of saline solution. The starting dilution was usually 1 : 20. Antisera. Rabbit antisera to human immunoglobulins and sperm were obtained from Calbiochem, Somerville, New Jersey, and were preabsorbed with human ABO red blood cells, platelets, and granulocytes. both at room temperature and at 34” C. Passive hemagglutination and Coombs test. As described previously,” human type 0, Rh ’ red blood cells were coated with sonicated sperm extra< t (SE) or seminal plasma (SP) from each male patient. Control cells were coated with SE and SP from a normal man without immunity to sperm. The red cells coated with the patient’s SE and SP were titrated against doubling serial dilutions of antisera to human immunoglobulins (polyvalent and class-specific) or of the patient’s serum and secretions in microtiter plates (Dynatech Corporation, Alexandria, Virginia). The samples were also titrated against SE and SP from a known normal man, to determine whether antibody reactions were specifically against a given patient’s sperm antigens. Occasionally, women who were highlv reactive against their husbands’ sperm were negative against sperm from normal donors. IgE determination. Quantitative determination of IgE levels in the various samples was carried out by use of the Phadebas IgE radioimmunosorbent test kit (Pharmacia, Uppsala, Sweden).‘, ’ Anti-IgE antibodies were covalently bound to Sephadex particles. The IgE in an unknown sample was allowed to compete with a fixed amount of “’ I-labeled IgE for the binding sites on the anti-IgE. The concentration of IgE in an un-
Volume Number
IgE and antisperm antibody titers in infertility
140 8
Table II. Serum IgE levels and antisperm antibody titers in women-fertile (Group antisperm immunity (Group E), and infertile with antisperm immunity (Group F)
D), infertile Student’s
Group
No. studied Serum IgE levels (IU/ml) Antispe& antibody titers (reciprocal) Serum Cervical secretion Vaginal secretion
D
25 76 2 16* 2k2 622
12 ? 2
Group
E
28
219 2 32 321 14 + 8 20 + 8
Group
F
32 472 f 55 22 420 203
-r- 12 f 12 k 24
925
without t test
D vs. E
E vs. F
D vs. F
P < 0.05
P < 0.025
P < 0.001
NSt NS NS
P < 0.001 P < 0.001 P < 0.001
P < 0.001 P < 0.001 P < 0.001
*Mean -C SEM. TNS = Not significant.
known sample was then determined by comparing its competitive capacity to that of IgE standards of known concentration. For this assay, the test samples (sera diluted tenfold), “‘I-labeled IgE solution, and Sephadexanti-IgE complex were mixed and incubated at 4” C. Bound and free IgE were separated by centrifugation and washing. The radioactivity of the solid phase (the Sephadex complex) was then measured. The amount of bound radioactivity was inversely proportional to the amount of IgE present in the sample. The counts (B) for each of the standards and unknown samples were measured as percentage of the mean counts of the zero standard (B,), where o/o Activity
bound
= (B/Be) x 100.
The percentage values obtained for the standards were plotted against the IgE concentrations on a semi-log paper, and a standard curve was constructed. The concentration of IgE in international units per milliliter, equivalent to 2 rig/ml in samples from the patients, was measured directly from the standard curve and multiplied by the dilution factor. The lower limit of sensitivity of this assay was approximately 1.5 IU/ml. Statistical analysis. Student’s t test and linear regression analysis were used for evaluation of the results.
Results IgE levels and antibody titers in fertile and infertile men. Of the 60 infertile men, 18 (Group B) had antibody titers in samples of sperm (23 ? 3 II-J/ml), serum (3 ? 1), and SP (11 & 1) similar to those in normal fertile men (Group A). The other 42 (Group C) had dramatically elevated antibody titers in sperm (325 ? 40), serum (45 + 6), and SP (79 ? 21), as compared with nonautoimmune Groups A and B. In the “sperm autoimmune”Group C, IgG (434 f 94) and IgA (224 ? 84) antibodies were present in sperm samples at titers higher than those for IgM (55 + 39). IgE levels in the serum (230 + 41) and SP (180 f 44) of these “sperm
Table III. Comparison of serum IgE levels in wives of husbands with autoimmunity to sperm and wives of nonautoimmune husbands Serum IgE levels (IUlml)
Wives with autoimmune husbands Wives with nonautoimmune husbands Student’s t test
406 247
f 55* k 97 NSt
*Mean 2 SEM. tNS = Not significant. autoimmune” men were significantly higher than those in Group A (P < 0.001 and P < 0.05, respectively). IgE levels in serum samples from Group C were higher than levels in Group B (P < 0.05); in contrast, the IgE levels in the SP for these two groups were not significantly different. The IgE levels in SP samples from nonautoimmune infertile men (Group B) were higher than the levels in the fertile men in Group A (P < 0.01) (Table I), but analysis of the data by linear regression disclosed no correlation between IgE levels and antibody titers in serum and seminal plasma in any of the men. IgE levels and antibody titers in fertile and infertile women. Of the 60 infertile women, 28 (Group E) had antibody titers in serum (3 & l), cervical secretions (14 + 8), and vaginal secretions (20 2 8) similar to those in fertile women (Group D). In 32 women (Group F), antibody titers were highly elevated in serum (22 + 13), cervical secretions (420 + 12), and vaginal secretions (203 ? 24), as compared with the titers in Groups D and E (P < 0.001). Serum IgE levels in Group F were significantly higher (472 * 55) than those in Groups E (219 2 32), P < 0.001) and D (76 5 16, P < 0.025). Serum IgE levels were higher in infertile nonisoimmune (Group E) than in fertile nonisoimmune (Group D) women (P < 0.05) (Table II). IgE levels in the cervicovaginal secretions were mostly below the limit of sensitivity. Linear regression analysis
926
Mathur et al.
0
It0 .--a o--o
Ml. C.B.
t56 \
‘1
o--o LB. o--o 1 C.B.
0
0
=P 000 2 4s
64 1
s
\
40 4
0
\
0
I 16-i \
0
-ii&--&
s~niarl pillal Mill Fig. 1. IgE levels in samples from infertile couples nisone therapy.
of serum and V. H. and C. B. before
Ion:le seminal plasma and after pred-
showed no correlation between antibody titers and IgE levels in the sera. Wives with auotoimmune husbands apparently had higher mean IgE levels in serum (406 * 55 IU/ml) than had those with nonautoimmune husbands (247 ? 97), but the difference was not significant (Table III). IgE levels in samples of serum and seminal plasma from infertile couples before and after treatment with prednisone. Two couples (initials of women being V. H. and C. B.) with high titers of antisperm antibodies were treated with immunosuppressive prednisone (15 mg/day for 3 months). Their antibody titers and IgE levels before and after treatment are shown in Figs. 1 and 2. The antibody titer in the serum of husband V. H. was unaltered after treatment, whereas his SP titer showed a reduction from 64 to 8. On the other hand, his serum IgE was reduced from 200 to 100 lU/ml and seminal plasma IgE, from 100 to 30 IUlml. The antibody titer in his wife’s serum was reduced from 32 to 4 and the IgE level, from 1,260 to 950 lU/ml. In couple C. B., antibody titer in the husband’s serum was reduced from 64 to 32 and IgE, from 150 to 100 lU/ml. In his SP, the antibody titer was reduced from 256 to 32 and IgE, from 40 to 30 II-J/ml. The antisperm antibody titer in the serum of his wife was reduced from 64 to 16 and IgE, from 930 to 420 IU/ml. Comment The exact role of antisperm antibodies in the etiology of human infertility is not clearly understood. This is due, at least partly, to a lack of efficient and reliable
Fig. 2. Antisperm
antibody titers (in reciprocal) in samples of
serum and seminal plasma from C. B. before and after prednisone
infertile therapy.
couples
V. H. and
methods for the determination of antisperm antibodies. We have described a method for measurement of antisperm antibody titers by passive hemagglutination and Coombs test,6 and we have employed this method in the evaluation of 208 couples with unexplained primary infertility, secondary infertility, prostatitis, oligospermia, and repeated abortions9 Clinical reports suggest that certain women, irrespective of their fertility status, may develop allergy to constituents of their husbands’ seminal plasma,‘, * and indeed, IgE-mediated allergy to human SP in female subjects has been described.” However, these studies provided no information as to the relationship between antisperm immunity and IgE levels in women or IgEmediated allergic reactions to secretions of the reproductive tract in men. Friberg,” using a tray agglutination method to detect antisperm antibodies, found no differences in IgE levels between men with high titers of sperm-agglutinating antibodies and those with low titers. However, agglutination of sperm is a nonspecific phenomenon which can be mediated by microbial and viral factors,” as well as detritus and sex steroids.i2 In the present study, men and women from infertile couples were classified according to their antisperm antibody titers. IgE levels in the serum and SP of men with autoimmunity to sperm were significantly higher than the levels in fertile and infertile men without antisperm immunity. IgE levels were similar in the sera of nonautoimmune men, irrespective of their fertility status, but somewhat higher in the SP of infertile compared to fertile men without antisperm immunity (Table I). IgE levels in the sera of normal fertile male and
Volume Number
140 8
female subjects were comparable to those reported in our previous studies.‘, * The plasma cells that produce IgE are located mainly in the mucosal tissues.13 It is possible that, in certain infertile men, plasma cells in the urogenital area may be triggered to produce IgE antibodies in response to seminal antigens long before other plasma cells are stimulated to produce IgG, IgA, or IgM. In a previous study,13 we noted that local immune responses to sperm antigens were usually greater than systemic immune responses. Significant titers of antisperm antibodies were detected in the seminal plasma before they could be detected in the serum, thus supporting the previous suggestion that a local immune system might be operative.13 Our studies suggest that a local reaginic antibody response may also be activated earlier than the agglutinating or cytotoxic antibody response in infertile men. In the present study, serum IgE levels were significantly higher in female subjects with sperm isoimmunity (Group F) than in nonisoimmune women, whether infertile (Group E) or fertile (Group D), and higher in Group E than in Group D. This suggests that female infertility in general may be associated with elevated levels of serum IgE. We are not aware of other reports that suggest a link between elevated IgE levels and infertility. Although IgE elevels in normal men and women were comparable,’ levels in infertile women were higher than those in infertile men (P < 0.01). This could be due to an exaggerated allergic response to SP antigens in infertile female subjects. We reported earlier I4 that autoimmunity to sperm in infertile men is associated with an unusually high incidence of isoimmunity to sperm in their wives. In the present study, women with autoimmune husbands had higher serum IgE levels than had those with nonautoimmune husbands (Table III). The spermatozoa in the semen of autoimmune men may undergo rapid deterioration and disintegration because of the presence of antisperm antibodies. Or the sperm may be defective and antigenically distinct and bear autoantibody prod-
REFERENCES
1. Halpern, B. N., Ky, T., and Robert, B.: Clinical and immunological study of an exceptional case of reaginic type sensitization to human seminal fluid, Immunology 12:247, 1967. 2. Levine, B. B., Siraganian, R. P., and Schenkein, I.: Allergy to human seminal plasma, N. Engl. J. Med. 288:894, 1973. 3. Svdbom, A.. and Karlsson, T.: Relationship between serum IgE levels and anaphylactic histamine release from isolated rat mast cells, Acta Physiol. Stand. 107:313, 1979. 4. Friberg, J.: Immunoglobulin concentration in serum and
IgE and antisperm antibody titers in infertility
927
ucts. This could also increase the “antigenic load” in their semen. Thus, wives of such men would be exposed to a greater concentration of antigen than would the wives of nonautoimmune husbands, which could result in the development of isoimmunity to sperm. The lack of correlation between antibody titers (primarily IgG, followed by IgA) and IgE levels is not surprising. Studies on IgE and mechanisms of reaginic hypersensitivity reactions over the past decade have revealed that IgE has a high affinity for mast cells and basophilic granulocytes, and that reactions of cellbound IgE antibody with allergen induce the release of chemical mediators from these cells.15 Other immunoglobulins, such as IgG and IgA, either do not bind to these cells or bind with much Iower affinity and, thus, are incapable of mediating reaginic hypersensitivity. We previously reported a reduction in total T-cell percentages in infertile subjects with elevated antisperm antibody titers,16 and it has been suggested that elevated IgE levels may be secondary to a reduction in T-cell regulatory function.” Treatment of infertile couples with antisperm immunity with prednisone has had varying degrees of success.1R Of the 25 couples whom we have treated, nine have achieved conception after prednisone therapy.s In two couples with elevated antisperm antibody titers and IgE levels before treatment, both were reduced (Figs. 1 and 2). Although we are not aware of any other studies on the effects of prednisone on IgE levels in infertile couples, Settipane and associateslg have reported significant decreases in IgE levels in patients with atopic asthma after prednisone therapy. The findings suggest that an immune mechanism involving reaginic IgE antibody, distinct from agglutinating and cytotoxic antisperm antibodies, may be operative in infertile couples with antisperm immunity. We are grateful for the excellent editorial of Charles L. Smith and the expert technical of J. T. Melchers and Melissa Flynn.
assistance assistance
seminal fluid from men with and without sperm-agglutinating antibodies, AM. J. OBSTET. GYNECOL. 136:671, 1980.
5. Fjallbrant, B.: Studies on sera from men with sperm antibodies,
Acta
Obstet.
Gynecol.
Stand.
48~131,
1968.
6. Mathur, S., Williamson,‘H. O., Landgrebe, S. C., Smith, C. L.. and Fudenbere. H. H.: Application of passive hemagglutination for evaluation ofantisperm a&tibodies and a modified Coombs test for detecting male autoimmunity to sperm antigens, J. Immunol. Methods 30:381, 1979. 7. Mathur, S., Mathur, R. S., Goust, J. M., Williamson, v
H. O., and Fudenberg,
H. H.: Sex steroid
hormones
and
928
Mathur et al.
circulating
IgE levels, Clin. Exp. Immunol. 30~403, 1977. S.. Goust, J. M., Horger, E. O., and Fudenberg, H. H.: IgE anti-Can&da antibodies and candidiasis, Infect. Immun. l&257, 1977. 9. Mathur, S., Baker, E. R., Williamson, H. O., Teague, K. J., and Fudenberg, H. H.: Clinical significance of sperm antibodies, Fertil. Steril. In press. F., Janne, J., and 10. Reunala, T., Koskimies, A. I., Bjorksten, Lassus, A.: lmmunoglobulin E-mediated severe allergy to human seminal plasma, Fertil. Steril. 28:832, 1977. Il. Taylor-Robinson, D., and Manchee, R. J.: Spermadsorption and spermagglutination by mycoplasmas, Nature 215:484, 1967. B., and Kay, D. J.: Agglutination of sper12. Boettcher. matozoa by human sera with added steroids, Andrologie 5:265, 1973. T., and Ishizaka, K.: Biology of immunoglobu13. Ishizaka. -’ lin E, Prog. Allergy 19:60, 1975. 14. Mathur. S.. Williamson, H. O., Derrick, F. C., Madyastha, I’. R., Melchers. J. T., Goltz, G. L., Baker, E. R., Smith, C. I... and Fudenberg, H. H.: A new microassay for sper-
8. Mathur,
Discussion DR. LUTHER M. TALBERT, lina. The authors have shown IgE in patients with presumed
Chapel
Hill, North
Caro-
increased serum levels immunologic infertility.
of
Since the reaginic antibodies belong to IgE and represent a distinct immunoglobulin class, and since plasma cells producing IgE are located mainly in mucous membranes, this approach may eventually lead to clinically important interpretation. Beginning at puberty, the internal environment of the testis is separated from the blood by the blood-testis barrier. Although small molecules and electrolytes do diffuse into and out of the rete testis and seminiferous tubules, larger molecules and spermatozoa in the normal man, for practical purposes, do not cross this barrier. Injection of the man with homogenates of one of his own testes has been shown to result in antibody formation and the destruction of the seminiferous tubules of the remaining testicle, which suggests that the immune system recognizes certain testicular components as foreign substances.’ Obstruction of the vas def‘erens or damage to the genital tract by trauma or infection is associated with a high incidence of’ the formation tibodies* and there is good evidence
of antisperm that the coating
anof
spermatozoa with antibody can result in poor cervical mucus penetration.3 Since an intact blood-testis barrier probably noglobulin
prevents passage of large into the testis, it is likely
amounts of immuthat sperm acquire
an antibody coating from the prostatic secretions. That the female human being is able to produce antibodies to spermatozoa and that her immunologic factors can play a role in infertility are suggested by many lines
of evidence.
The first clinical suggestion lem
in men
or women
sperm penetration,
may
of an immunologic be the
in a Huhner
recognition
probof poor
test, in the presence of
mocytotoxic antibody: Comparison with passive benlagglutination assay for antisperm antibodies, J. Immunol 126:905, 1981. 15. Ishizaka, T., Ishizaka, K.. Conrad, D. H., and Froesc, :I.,4 new concept of triggering mechanisms of IgE-mediated histamine release, J. Allergy Clin. Immunol. 61:320, 1978. 16. Mathur, S., Goust, J. M., Williamson, H. O., and Fuderlberg, H. H.: Antigenic cross-reactivity of sperm and ‘1‘ lymphocytes, Fertil. Steril. 34469. 1980. 17. Mathur, S.,G oust, J. M., Hol-ger, E. 0.. and Fudenberg. H. H.: Cell-mediated immune deficiency and heightened humoral immune response in chronic \,aginal candidiasis. J. Clin. Lab. Immunot. 1:129, 1978. 18. Hendry, W. F., Stedronska, J., Hughes, L.. Cameron. K. M., and Pugh, R. C. B.: Steroid treatment of mate subfertility caused by antisperm antibodies, Lancer 1:498. 1979. G. A., Pudupakkam, R. K., and McGowan, 19. Settipane, J. H.: Corticosteroid effect on immunoglobulins. J. .AIlergy Clin. Immunol. 62: 162, 1978.
normal-appearing mucus and the peculiar in loco-onl) motility of sperm in the spermatozoa-cervical mucus contact test.’ Various immunologic tests, which may be roughly separated into those detecting sperm agglutination and those
detecting
sperm
been described. clinical
immobilizing
antibodies,
The latter seem to correlate
infertility.
The
authors
have
have
better with
used
a mod-
ification of the Coombs test, original with them. and have also shown good correlation with infertility. As with many other attempts to evaluate specific facets of infertility, the problem of C(Jntl.01 groups has not been entirely resolved in this study, and the efficac! of immunosuppressive therapy remains speculative. However, other investigators also have reported a reduction in antibody titers in some men treated with glucocorticoids ner.’ Condom
termination therapy
for
and
in conception
therapy, of antibody
the
immunized
woman,
effective in some couples. From a clinical viewpoint, authors
to elaborate
are the indications ity?
(2)
What
tests
in the
female
part-
preferably monitored by detiters, has been suggested as
on the
I would following
for immunologic should
be done?
and
appears
to be
like to ask the points:
(1) What
testing in infertilWill
your
mod-
ification of the Coombs test replace both the Isojima and Kibrick tests? (3) Do you beliel;e that 15 mg of prednisone daily, a relatively small dose, will be effective
in
controlled
studies,
or
wilf
larger
doses
he
required?
REFERENCES 1. Setchell. B.P.: The Mammalian Testis, Ithaca, New York, 1978, Cornell University Press. 2. Rumke, P. H., and Hellinga, G.: Autoantibodies against spermatozoa in sterile men, Am. .1. Clin. Pathol. 32~357.
1050.
3. Fjallbrant,
B.: Cervical
mucus
penetration
by human
sper-
Volume Number
140 8
matozoa treated with anti-spermatozoa1 antibodies from rabbit and man, Acta Obstet. Gynecol. Stand. 48:71, 1969. 4. Kremer, J., and Jager, S.: The sperm-cervical mucus contact test: A oreliminarv renort, Fertil. Steril. 27:335, 1976. 5. Hendry, W: F., Stedr&ia, J., Hughes, L., et al.: Steroid treatment of male subfertility caused by antisperm antibodies, Lancet 1:498, 1979. DR. JAMES D. KITCHIN, III, Charlottesville, Virginia. In a recent review, Jones’ discussed the role of immunologic factors in human infertility. He emphasized the widespread skepticism in regard to such an association and attributed this to confusion, contradictions, unjustifiable claims, and untidy methodology in the literature. However, there is no question but that isoimmunization in the woman and autoimmunization in the man against sperm and seminal plasma aritigens may be demonstrated by a variety of techniques. Dr. Mathur and co-authors have studied 60 couples with unexplained infertility-a large group if one agrees with the generally reported incidence of 5% to IO%, unexplained infertility among infertile couples. The patients have been placed in various groups according to the results of a passive hemagglutination and Coombs test method for determination of antisperm antibody. Quantitative determinations of IgE were performed and the various groups were compared with regard to the levels of this immunoglobulin detected. The authors emphasized that there is a lack of correlation between antibody titers (which reflect the presence of other immunoglobulins) and IgE levels. However, they showed an elevated IgE response in subjects with significantly elevated antisperm antibody titers. This is of considerable interest since it is not generally thought that IgE allergic hypersensitivity is an effective immune mechanism that could affect fertility. I am unaware of any evidence for IgE secretory antibody in cervical mucus, and if IgE is important in infertility, it would seem that it must be present in this location. Friberg’ reported a lack of correlation between antisperm antibody titers and IgE levels in the sera and seminal plasma of infertile men. The radioimmunosorbent test, which was used to quantitate IgE in this study, is subject to error in that low values tend to be overestimated because of nonspecific interference. I am told that a newer method, paper radioimmunosorbent test, is available and is more accurate. Further work is required before the data presented today can be interpreted. We need to determine whether elevated IgE levels really indicate clinically significant alterations of the immune mechanism which are of importance in fertility or whether such alterations are nonspecific and irrelevant. I would like to ask the authors about the patients who were treated with corticosteroids. Steroids may be responsible for various changes in the immune response, including decreased production of antibody, as well as
IgE and antisperm antibody titers in infertility
929
altered interaction of antibody with bound antigen. I ask the following with regard to the nine patients who conceived after prednisone therapy: (1) Did antibody titers and IgE levels decrease in all patients and did they differ from those in the nine patients who did not conceive? (2) Did the patients conceive during or after therapy? (3) If conception occurred after therapy, were antibody and IgE levels elevated in any patient? I would like to thank the authors for the opportunity to review their manuscript. In situations such as unexplained infertility, which by definition is poorly defined, it is difficult to know what sort of observations will ultimately prove to be useful in evaluation. Even though IgE has not been established as an important factor in infertility, further work, such as that reported, may ultimately show it to be so. REFERENCES
1. Jones, W. R.: Immunologic infertility-Fact or fiction? Fertil. Steril. 33:577, 1980. 2. Friberg, J.: Immunoglobulin concentration in serum and seminal fluid from men with and without sperm-agglutinating antibodies, AM. J. OBSTET. GYNECOL. 136:671, 1980. DR. WILLIAMSON AND DR. MATHUR (Closing). Indications for immunologic testing in infertility include a persistently poor postcoital test, such as those that show vibratory (nonprogressive) or no motility, no sperm, and inflamnatory cells unresponsive to appropriate therapy or with negative cultures. Also, lowered motility or otherwise poor semen quality, coexistence of other immunologic disorders and negative routine infertility assessments (normal seminal analysis, postcoital tests, patent tubes and ovulatory cycles) may be included. It is suggested that this be done prior to operative procedures, such as laparoscopy, unless other indications exist that suggest that it not be delayed. One cannot suggest that our tests should replace currently accepted methods until they are confirmed in other laboratories and a larger experience is gained to validate them. However, we have found that they are reproducible, and clinical experience is supporting our findings. The small dose of prednisone was adopted from the work of Shulman and associates (Shulman, S., Harlin, B., Davis, P., and Reyniak, J. V.: Immune infertility and new approaches to treatment, Fertil. Steril. 29:309, 1978) which has been used by others.’ We have tended to avoid the high dose but briefer and interrupted schedule of Shulman and associates2 because of mental aberrations and gastric distress reported in a few patients when it was used. The pregnancy rates of our series and of others appear to be comparable but are still low, which suggests that further manipulation of dosage may be desirable. Doubts have been expressed about the relevance of sperm antibodies to human infertility,3 mainly because of the unreliability of the existing methods. We devel-
930
Mathur et al.
oped two new methods for determination of sperm antibodies in infertile couples. The first, a modified Coombs test and passive hemagglutination, aids in determination of class specificity of antibodies bound to sperm and titers of agglutinating sperm antibodies present in sera and secretions. The second, a microspermocytotoxicity assay,” determines cytotoxic sperm antibody titers. Simultaneous use of both of these methods enables one to obtain information on the presence of sperm antibodies in local as well as systemic circulation. The results have been reproducible. Dr. Kitchin commented on the unusually high incidence of unexplained infertility in this series. Our patients are seen by referral and generally after the referring gynecologist has exhausted customary investigations and therapy. Of our patients subjected to immunologic testing, approximately 60% to 80% without other findings have positive tests, whereas outside referrals for testing run only about 20% positive. The lower limit of sensitivity of the radioimmunosorbent test is approximately 1.5 IU/ml. We have used this assay routinely in our laboratory and established the range of IgE levels in the sera from normal men and women. IgE levels in the sera of normal fertile men and women in the present study were comparable to those in our previous studies. Keview of the case histories of most of the infertile patients with elevated serum IgE levels failed to show evidence of other nonspecific allergic problems, such as hay fever or atopic dermatitis. Therefore, we believe that the elevated IgE levelsareassociated with infertility
i\ugust Am. J. Obstet.
15, 19x1 Qmcol.
in these subjects. However, further work is warranted before we can conclude that the IgE antibodies are specifically directed against sperm or seminal plasma antigens. We have outlined some changes in the antibody titers and IgE levels in two couples treated with prednisone. Studies are underway in order to determine whether prednisone treatment alters the IgE levels, even though it is known to do so in certain clinical conditions.6
REFERENCES
1. Hendry, W. F., Stedronska, J., Hughes, L., Cameron, K. M., and Pugh, R. C. B.: Steroid treatment of male subfertility caused by antisperm antibodies, Lancet 1:498, 1979. 2. Shulman, S., Harlin B., Davis, P., and Reyniak, J. V.: Immune infertility and new approaches to treatment, Fertil. Steril. 29:309, 1978. 3. Jones, W. R.: Immunologic infertility-Fact or tiction? Fertil. Steril. 33:577, 1980. 4. Mathur S., Williamson, H. O., Landgrebe, S. C., Smith, C. L., and Fudenberg, H. H.: Application of passive hemagglutination for evaluation of antisperm antibodies and a modified Coombs test for detecting male autoimmunity to sperm antigens, J. Immunol. Methods 30:381, 1979. 5. Mathur, S., Williamson, H. O., Derrick, F. C., Madyastha, P. R., Melchers, J. T., Holtz, G. L., Baker, E. R., Smith, C. L., and Fudenberg, H.: A new microassay for spermocytotoxic antibody: Comparison with passive hemagglutination assay for antisperm antibodies, J. Immunol. 126:905, 1981. 6. Settipane. G. A., Pudupakkam, R. K., and McGowan, J. H.: Corticosteroid effect on immunoglobulins, J. Allergy Clin. Immunol. 62:162, 1978.
0.
WILLIAMSON,
E. R. BAKER, H.
H.
Charleston,
PH.D. M.D.
M.D.
FUDENBERG, South
M.D.
Carolina
lmmunoglobulin E (IgE) levels and antisperm antibody titers were determined in samples of serum and seminal plasma from 25 fertile men (Group A), 18 infertile men without measurable immunity to sperm (Group B), and 42 infertile men autoimmune to sperm (Group C), and in samples of serum and cervical and vaginal secretions from 25 fertile women (Group D), 28 infertile women without measurable immunity to sperm (Group E), and 32 infertile women isoimmune to sperm (Group F). Among the men, IgE levels in the serum, measured in international units per milliliter, were elevated in Group C (230 2 41, mean f SEM) as compared with Groups B (94 f 33, P < 0.05) and A (55 k 8, P < 0.001). In contrast, IgE levels in the seminal plasma in Group C (180 f 44) were not significantly different from those in Group B (48 + 21), but were higher than those in Group A (8 f 2, P < 0.05). In the women, serum IgE levels were higher (P < 0.001) in Group F, with isoimmunity to sperm (472 k 55) than in Groups E (219 + 32) and D (78 f 18). Wives of autoimmune husbands had somewhat, though not significantly, higher serum IgE levels (406 k 55) than had wives of nonautoimmune husbands (247 f 97). These results provide evidence for an elevated IgE response in subjects with significantly elevated antisperm antibody titers. (AM. J. OBSTET.
GYNECOL.
140:923,1981.)
A L LE R G Y to human seminal plasma in female subjects has been associated with urticaria, pharyngeal edema, and cardiovascular symptoms immediately after coitus.’ In such patients, skin tests with seminal plasma, but not with extracts of sperm, revealed immediate type hypersensitivity. The causative agent has been reported to be a protein of prostatic origin.* Since immunoglobulin E Publication No. 439 frm the Department of Basic and Clinical Immunology and Microbiology and the Section of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Medical University of South Carolina. Supported in part by United States Public Health Service grants HD-09938, CA-25746, and HDIAI-14365, and by United States Public Health Service National Research Service Award AI-07063. The Foundation Prize Basic Science Award presented the Forty-third .4nnual Meeting of the South Atlantic Association of Obstetricians and Gynecologists, Hot Springs, Virginia, January 25-28, 1981.
at
Reprint requests: Dr. Subbi Mathur, Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina, I71 Ashley Ave., Charleston, South Carolina 29425. 0002-9378/81/160923+08$00.80/0
0 1981 The C. V. Mosby
Co.
(IgE) is considered to be responsible for the anaphylactic reactions of antigen-exposed sensitized mast cells, leading to histamine production,3 information on the possible association of antisperm immunity and IgE levels in these subjects is of interest. Friberg4 has reported a lack of correlation between antisperm antibody titers and IgE levels in the sera and seminal plasma of infertile men, although increased levels of other serum immunoglobulins in men with antisperm antibodies have been reported.’ To date, no systematic study of the role of reaginic antibodies in immunologic infertility caused by antisperm immunity in male and female subjects has been undertaken, primarily because of a lack of effective screening methods for autoimmunity and isoimmunity to sperm. A new method, based on passive hemagglutination and Coombs test, has been established in our laboratory for the determination of antisperm antibody titers on sperm and in sera, seminal plasma, and cervical and vaginal secretions from couples with unexplained infertility.6 In the present study, we measured IgE levels and antisperm antibody titers in samples from fertile and infertile subjects. 923
924
Mathur
et al.
Table I. Serum IgE levels and antisperm antibody titers in men-fertile antisperm immunity (Group B), and infertile with antisperm immunity
(Group A), infertile (Group C)
without
Student’,i Group
No. studied IgE levels (IU/ml) Serum SP Antisperm antibody titers (ieciprocal) Sperm Polyvalent ‘gc,
@A w
Serum SP
A
Group
B
Group
/ thl
C
25
18
42
55 2 8* 822
94 + 33 48 iz 21
230 2 41
180 ? 44
NST P < 0.01
P < 0.05 NS
P < 0.001 P < 0.05
722 Sk2 15 ” 2 2&l 221 7+1
23 2 3
325 434 224 55 45
P = 0.1 NS NS NS NS
P< P< P< P< P< P<
P< P< P< P< P< P<
10 t 2 26 t 9 321 3kl
11 ir 1
+ 5 k 2 5
40 94 84 39 6
79 k 21
NS
0.001 0.001 0.001 0.001 0.05 0.025
0.001 0.001 0.001 0.001 0.05 O.ofll
*Mean ? SEM. tNS = Not significant.
Material and methods Patients. Studied were infertile couples who had unsuccessfully attempted conception for 2 to 10 years. The women were normal and ovulatory, with normal endocrine profiles, no tubal occlusion, and no signs of genital infections in the cervicovaginal area. The men were free of infections and had normal sperm counts and motility. Sixty couples were studied; 25 fertile men and 25 fertile women served as controls. When one or both partners had elevated antibody titers, they were re-evaluated at periodic intervals. All participants were informed of the nature and purpose of the study. Samples. Serum samples were obtained by venipuncture. Semen was obtained by masturbation after 2 to -I days of sexual abstinence. The e.jaculate was collected in a clean, wide-mouthed, glass jar and allowed to liquefy at room temperature for about 30 minutes before the assays. Seminal plasma was obtained by centrifuging the semen at 1,000 X g for 5 minutes and removing the supernatant. \‘iscous samples were centrifuged for 10 minutes. Vaginal fluid was collected b) swabbing away excessive leukorrhea with a clean, cotton-tipped applicator and then flooding the posterior fornix with 2 ml of sterile saline solution from a sterile syringe via a plastic cannula. The fluid was Hushed back and forth a total of four times before the final aspiration. Cervical mucus was obtained by inserting the tip of an endocervical aspirator cannula into the endocervix and withdrawing all the mucus that could be obtained. Viscous secretions were treated with 0.1% bromelin (Sigma Chemicals, St. Louis, Missouri) in order to liquefy them. The secretions were extracted with a known volume of saline solution at room temperature and centrifuged at 1,000 x g for 10 minutes to eliminate cells and debris. The clear supernatant that con-
tained the highly soluble immunoglobulins was used for study. Dilutions of the washings were calculated on the basis of the ratio of the original volume of mucus to the total volume after the addition of saline solution. The starting dilution was usually 1 : 20. Antisera. Rabbit antisera to human immunoglobulins and sperm were obtained from Calbiochem, Somerville, New Jersey, and were preabsorbed with human ABO red blood cells, platelets, and granulocytes. both at room temperature and at 34” C. Passive hemagglutination and Coombs test. As described previously,” human type 0, Rh ’ red blood cells were coated with sonicated sperm extra< t (SE) or seminal plasma (SP) from each male patient. Control cells were coated with SE and SP from a normal man without immunity to sperm. The red cells coated with the patient’s SE and SP were titrated against doubling serial dilutions of antisera to human immunoglobulins (polyvalent and class-specific) or of the patient’s serum and secretions in microtiter plates (Dynatech Corporation, Alexandria, Virginia). The samples were also titrated against SE and SP from a known normal man, to determine whether antibody reactions were specifically against a given patient’s sperm antigens. Occasionally, women who were highlv reactive against their husbands’ sperm were negative against sperm from normal donors. IgE determination. Quantitative determination of IgE levels in the various samples was carried out by use of the Phadebas IgE radioimmunosorbent test kit (Pharmacia, Uppsala, Sweden).‘, ’ Anti-IgE antibodies were covalently bound to Sephadex particles. The IgE in an unknown sample was allowed to compete with a fixed amount of “’ I-labeled IgE for the binding sites on the anti-IgE. The concentration of IgE in an un-
Volume Number
IgE and antisperm antibody titers in infertility
140 8
Table II. Serum IgE levels and antisperm antibody titers in women-fertile (Group antisperm immunity (Group E), and infertile with antisperm immunity (Group F)
D), infertile Student’s
Group
No. studied Serum IgE levels (IU/ml) Antispe& antibody titers (reciprocal) Serum Cervical secretion Vaginal secretion
D
25 76 2 16* 2k2 622
12 ? 2
Group
E
28
219 2 32 321 14 + 8 20 + 8
Group
F
32 472 f 55 22 420 203
-r- 12 f 12 k 24
925
without t test
D vs. E
E vs. F
D vs. F
P < 0.05
P < 0.025
P < 0.001
NSt NS NS
P < 0.001 P < 0.001 P < 0.001
P < 0.001 P < 0.001 P < 0.001
*Mean -C SEM. TNS = Not significant.
known sample was then determined by comparing its competitive capacity to that of IgE standards of known concentration. For this assay, the test samples (sera diluted tenfold), “‘I-labeled IgE solution, and Sephadexanti-IgE complex were mixed and incubated at 4” C. Bound and free IgE were separated by centrifugation and washing. The radioactivity of the solid phase (the Sephadex complex) was then measured. The amount of bound radioactivity was inversely proportional to the amount of IgE present in the sample. The counts (B) for each of the standards and unknown samples were measured as percentage of the mean counts of the zero standard (B,), where o/o Activity
bound
= (B/Be) x 100.
The percentage values obtained for the standards were plotted against the IgE concentrations on a semi-log paper, and a standard curve was constructed. The concentration of IgE in international units per milliliter, equivalent to 2 rig/ml in samples from the patients, was measured directly from the standard curve and multiplied by the dilution factor. The lower limit of sensitivity of this assay was approximately 1.5 IU/ml. Statistical analysis. Student’s t test and linear regression analysis were used for evaluation of the results.
Results IgE levels and antibody titers in fertile and infertile men. Of the 60 infertile men, 18 (Group B) had antibody titers in samples of sperm (23 ? 3 II-J/ml), serum (3 ? 1), and SP (11 & 1) similar to those in normal fertile men (Group A). The other 42 (Group C) had dramatically elevated antibody titers in sperm (325 ? 40), serum (45 + 6), and SP (79 ? 21), as compared with nonautoimmune Groups A and B. In the “sperm autoimmune”Group C, IgG (434 f 94) and IgA (224 ? 84) antibodies were present in sperm samples at titers higher than those for IgM (55 + 39). IgE levels in the serum (230 + 41) and SP (180 f 44) of these “sperm
Table III. Comparison of serum IgE levels in wives of husbands with autoimmunity to sperm and wives of nonautoimmune husbands Serum IgE levels (IUlml)
Wives with autoimmune husbands Wives with nonautoimmune husbands Student’s t test
406 247
f 55* k 97 NSt
*Mean 2 SEM. tNS = Not significant. autoimmune” men were significantly higher than those in Group A (P < 0.001 and P < 0.05, respectively). IgE levels in serum samples from Group C were higher than levels in Group B (P < 0.05); in contrast, the IgE levels in the SP for these two groups were not significantly different. The IgE levels in SP samples from nonautoimmune infertile men (Group B) were higher than the levels in the fertile men in Group A (P < 0.01) (Table I), but analysis of the data by linear regression disclosed no correlation between IgE levels and antibody titers in serum and seminal plasma in any of the men. IgE levels and antibody titers in fertile and infertile women. Of the 60 infertile women, 28 (Group E) had antibody titers in serum (3 & l), cervical secretions (14 + 8), and vaginal secretions (20 2 8) similar to those in fertile women (Group D). In 32 women (Group F), antibody titers were highly elevated in serum (22 + 13), cervical secretions (420 + 12), and vaginal secretions (203 ? 24), as compared with the titers in Groups D and E (P < 0.001). Serum IgE levels in Group F were significantly higher (472 * 55) than those in Groups E (219 2 32), P < 0.001) and D (76 5 16, P < 0.025). Serum IgE levels were higher in infertile nonisoimmune (Group E) than in fertile nonisoimmune (Group D) women (P < 0.05) (Table II). IgE levels in the cervicovaginal secretions were mostly below the limit of sensitivity. Linear regression analysis
926
Mathur et al.
0
It0 .--a o--o
Ml. C.B.
t56 \
‘1
o--o LB. o--o 1 C.B.
0
0
=P 000 2 4s
64 1
s
\
40 4
0
\
0
I 16-i \
0
-ii&--&
s~niarl pillal Mill Fig. 1. IgE levels in samples from infertile couples nisone therapy.
of serum and V. H. and C. B. before
Ion:le seminal plasma and after pred-
showed no correlation between antibody titers and IgE levels in the sera. Wives with auotoimmune husbands apparently had higher mean IgE levels in serum (406 * 55 IU/ml) than had those with nonautoimmune husbands (247 ? 97), but the difference was not significant (Table III). IgE levels in samples of serum and seminal plasma from infertile couples before and after treatment with prednisone. Two couples (initials of women being V. H. and C. B.) with high titers of antisperm antibodies were treated with immunosuppressive prednisone (15 mg/day for 3 months). Their antibody titers and IgE levels before and after treatment are shown in Figs. 1 and 2. The antibody titer in the serum of husband V. H. was unaltered after treatment, whereas his SP titer showed a reduction from 64 to 8. On the other hand, his serum IgE was reduced from 200 to 100 lU/ml and seminal plasma IgE, from 100 to 30 IUlml. The antibody titer in his wife’s serum was reduced from 32 to 4 and the IgE level, from 1,260 to 950 lU/ml. In couple C. B., antibody titer in the husband’s serum was reduced from 64 to 32 and IgE, from 150 to 100 lU/ml. In his SP, the antibody titer was reduced from 256 to 32 and IgE, from 40 to 30 II-J/ml. The antisperm antibody titer in the serum of his wife was reduced from 64 to 16 and IgE, from 930 to 420 IU/ml. Comment The exact role of antisperm antibodies in the etiology of human infertility is not clearly understood. This is due, at least partly, to a lack of efficient and reliable
Fig. 2. Antisperm
antibody titers (in reciprocal) in samples of
serum and seminal plasma from C. B. before and after prednisone
infertile therapy.
couples
V. H. and
methods for the determination of antisperm antibodies. We have described a method for measurement of antisperm antibody titers by passive hemagglutination and Coombs test,6 and we have employed this method in the evaluation of 208 couples with unexplained primary infertility, secondary infertility, prostatitis, oligospermia, and repeated abortions9 Clinical reports suggest that certain women, irrespective of their fertility status, may develop allergy to constituents of their husbands’ seminal plasma,‘, * and indeed, IgE-mediated allergy to human SP in female subjects has been described.” However, these studies provided no information as to the relationship between antisperm immunity and IgE levels in women or IgEmediated allergic reactions to secretions of the reproductive tract in men. Friberg,” using a tray agglutination method to detect antisperm antibodies, found no differences in IgE levels between men with high titers of sperm-agglutinating antibodies and those with low titers. However, agglutination of sperm is a nonspecific phenomenon which can be mediated by microbial and viral factors,” as well as detritus and sex steroids.i2 In the present study, men and women from infertile couples were classified according to their antisperm antibody titers. IgE levels in the serum and SP of men with autoimmunity to sperm were significantly higher than the levels in fertile and infertile men without antisperm immunity. IgE levels were similar in the sera of nonautoimmune men, irrespective of their fertility status, but somewhat higher in the SP of infertile compared to fertile men without antisperm immunity (Table I). IgE levels in the sera of normal fertile male and
Volume Number
140 8
female subjects were comparable to those reported in our previous studies.‘, * The plasma cells that produce IgE are located mainly in the mucosal tissues.13 It is possible that, in certain infertile men, plasma cells in the urogenital area may be triggered to produce IgE antibodies in response to seminal antigens long before other plasma cells are stimulated to produce IgG, IgA, or IgM. In a previous study,13 we noted that local immune responses to sperm antigens were usually greater than systemic immune responses. Significant titers of antisperm antibodies were detected in the seminal plasma before they could be detected in the serum, thus supporting the previous suggestion that a local immune system might be operative.13 Our studies suggest that a local reaginic antibody response may also be activated earlier than the agglutinating or cytotoxic antibody response in infertile men. In the present study, serum IgE levels were significantly higher in female subjects with sperm isoimmunity (Group F) than in nonisoimmune women, whether infertile (Group E) or fertile (Group D), and higher in Group E than in Group D. This suggests that female infertility in general may be associated with elevated levels of serum IgE. We are not aware of other reports that suggest a link between elevated IgE levels and infertility. Although IgE elevels in normal men and women were comparable,’ levels in infertile women were higher than those in infertile men (P < 0.01). This could be due to an exaggerated allergic response to SP antigens in infertile female subjects. We reported earlier I4 that autoimmunity to sperm in infertile men is associated with an unusually high incidence of isoimmunity to sperm in their wives. In the present study, women with autoimmune husbands had higher serum IgE levels than had those with nonautoimmune husbands (Table III). The spermatozoa in the semen of autoimmune men may undergo rapid deterioration and disintegration because of the presence of antisperm antibodies. Or the sperm may be defective and antigenically distinct and bear autoantibody prod-
REFERENCES
1. Halpern, B. N., Ky, T., and Robert, B.: Clinical and immunological study of an exceptional case of reaginic type sensitization to human seminal fluid, Immunology 12:247, 1967. 2. Levine, B. B., Siraganian, R. P., and Schenkein, I.: Allergy to human seminal plasma, N. Engl. J. Med. 288:894, 1973. 3. Svdbom, A.. and Karlsson, T.: Relationship between serum IgE levels and anaphylactic histamine release from isolated rat mast cells, Acta Physiol. Stand. 107:313, 1979. 4. Friberg, J.: Immunoglobulin concentration in serum and
IgE and antisperm antibody titers in infertility
927
ucts. This could also increase the “antigenic load” in their semen. Thus, wives of such men would be exposed to a greater concentration of antigen than would the wives of nonautoimmune husbands, which could result in the development of isoimmunity to sperm. The lack of correlation between antibody titers (primarily IgG, followed by IgA) and IgE levels is not surprising. Studies on IgE and mechanisms of reaginic hypersensitivity reactions over the past decade have revealed that IgE has a high affinity for mast cells and basophilic granulocytes, and that reactions of cellbound IgE antibody with allergen induce the release of chemical mediators from these cells.15 Other immunoglobulins, such as IgG and IgA, either do not bind to these cells or bind with much Iower affinity and, thus, are incapable of mediating reaginic hypersensitivity. We previously reported a reduction in total T-cell percentages in infertile subjects with elevated antisperm antibody titers,16 and it has been suggested that elevated IgE levels may be secondary to a reduction in T-cell regulatory function.” Treatment of infertile couples with antisperm immunity with prednisone has had varying degrees of success.1R Of the 25 couples whom we have treated, nine have achieved conception after prednisone therapy.s In two couples with elevated antisperm antibody titers and IgE levels before treatment, both were reduced (Figs. 1 and 2). Although we are not aware of any other studies on the effects of prednisone on IgE levels in infertile couples, Settipane and associateslg have reported significant decreases in IgE levels in patients with atopic asthma after prednisone therapy. The findings suggest that an immune mechanism involving reaginic IgE antibody, distinct from agglutinating and cytotoxic antisperm antibodies, may be operative in infertile couples with antisperm immunity. We are grateful for the excellent editorial of Charles L. Smith and the expert technical of J. T. Melchers and Melissa Flynn.
assistance assistance
seminal fluid from men with and without sperm-agglutinating antibodies, AM. J. OBSTET. GYNECOL. 136:671, 1980.
5. Fjallbrant, B.: Studies on sera from men with sperm antibodies,
Acta
Obstet.
Gynecol.
Stand.
48~131,
1968.
6. Mathur, S., Williamson,‘H. O., Landgrebe, S. C., Smith, C. L.. and Fudenbere. H. H.: Application of passive hemagglutination for evaluation ofantisperm a&tibodies and a modified Coombs test for detecting male autoimmunity to sperm antigens, J. Immunol. Methods 30:381, 1979. 7. Mathur, S., Mathur, R. S., Goust, J. M., Williamson, v
H. O., and Fudenberg,
H. H.: Sex steroid
hormones
and
928
Mathur et al.
circulating
IgE levels, Clin. Exp. Immunol. 30~403, 1977. S.. Goust, J. M., Horger, E. O., and Fudenberg, H. H.: IgE anti-Can&da antibodies and candidiasis, Infect. Immun. l&257, 1977. 9. Mathur, S., Baker, E. R., Williamson, H. O., Teague, K. J., and Fudenberg, H. H.: Clinical significance of sperm antibodies, Fertil. Steril. In press. F., Janne, J., and 10. Reunala, T., Koskimies, A. I., Bjorksten, Lassus, A.: lmmunoglobulin E-mediated severe allergy to human seminal plasma, Fertil. Steril. 28:832, 1977. Il. Taylor-Robinson, D., and Manchee, R. J.: Spermadsorption and spermagglutination by mycoplasmas, Nature 215:484, 1967. B., and Kay, D. J.: Agglutination of sper12. Boettcher. matozoa by human sera with added steroids, Andrologie 5:265, 1973. T., and Ishizaka, K.: Biology of immunoglobu13. Ishizaka. -’ lin E, Prog. Allergy 19:60, 1975. 14. Mathur. S.. Williamson, H. O., Derrick, F. C., Madyastha, I’. R., Melchers. J. T., Goltz, G. L., Baker, E. R., Smith, C. I... and Fudenberg, H. H.: A new microassay for sper-
8. Mathur,
Discussion DR. LUTHER M. TALBERT, lina. The authors have shown IgE in patients with presumed
Chapel
Hill, North
Caro-
increased serum levels immunologic infertility.
of
Since the reaginic antibodies belong to IgE and represent a distinct immunoglobulin class, and since plasma cells producing IgE are located mainly in mucous membranes, this approach may eventually lead to clinically important interpretation. Beginning at puberty, the internal environment of the testis is separated from the blood by the blood-testis barrier. Although small molecules and electrolytes do diffuse into and out of the rete testis and seminiferous tubules, larger molecules and spermatozoa in the normal man, for practical purposes, do not cross this barrier. Injection of the man with homogenates of one of his own testes has been shown to result in antibody formation and the destruction of the seminiferous tubules of the remaining testicle, which suggests that the immune system recognizes certain testicular components as foreign substances.’ Obstruction of the vas def‘erens or damage to the genital tract by trauma or infection is associated with a high incidence of’ the formation tibodies* and there is good evidence
of antisperm that the coating
anof
spermatozoa with antibody can result in poor cervical mucus penetration.3 Since an intact blood-testis barrier probably noglobulin
prevents passage of large into the testis, it is likely
amounts of immuthat sperm acquire
an antibody coating from the prostatic secretions. That the female human being is able to produce antibodies to spermatozoa and that her immunologic factors can play a role in infertility are suggested by many lines
of evidence.
The first clinical suggestion lem
in men
or women
sperm penetration,
may
of an immunologic be the
in a Huhner
recognition
probof poor
test, in the presence of
mocytotoxic antibody: Comparison with passive benlagglutination assay for antisperm antibodies, J. Immunol 126:905, 1981. 15. Ishizaka, T., Ishizaka, K.. Conrad, D. H., and Froesc, :I.,4 new concept of triggering mechanisms of IgE-mediated histamine release, J. Allergy Clin. Immunol. 61:320, 1978. 16. Mathur, S., Goust, J. M., Williamson, H. O., and Fuderlberg, H. H.: Antigenic cross-reactivity of sperm and ‘1‘ lymphocytes, Fertil. Steril. 34469. 1980. 17. Mathur, S.,G oust, J. M., Hol-ger, E. 0.. and Fudenberg. H. H.: Cell-mediated immune deficiency and heightened humoral immune response in chronic \,aginal candidiasis. J. Clin. Lab. Immunot. 1:129, 1978. 18. Hendry, W. F., Stedronska, J., Hughes, L.. Cameron. K. M., and Pugh, R. C. B.: Steroid treatment of mate subfertility caused by antisperm antibodies, Lancer 1:498. 1979. G. A., Pudupakkam, R. K., and McGowan, 19. Settipane, J. H.: Corticosteroid effect on immunoglobulins. J. .AIlergy Clin. Immunol. 62: 162, 1978.
normal-appearing mucus and the peculiar in loco-onl) motility of sperm in the spermatozoa-cervical mucus contact test.’ Various immunologic tests, which may be roughly separated into those detecting sperm agglutination and those
detecting
sperm
been described. clinical
immobilizing
antibodies,
The latter seem to correlate
infertility.
The
authors
have
have
better with
used
a mod-
ification of the Coombs test, original with them. and have also shown good correlation with infertility. As with many other attempts to evaluate specific facets of infertility, the problem of C(Jntl.01 groups has not been entirely resolved in this study, and the efficac! of immunosuppressive therapy remains speculative. However, other investigators also have reported a reduction in antibody titers in some men treated with glucocorticoids ner.’ Condom
termination therapy
for
and
in conception
therapy, of antibody
the
immunized
woman,
effective in some couples. From a clinical viewpoint, authors
to elaborate
are the indications ity?
(2)
What
tests
in the
female
part-
preferably monitored by detiters, has been suggested as
on the
I would following
for immunologic should
be done?
and
appears
to be
like to ask the points:
(1) What
testing in infertilWill
your
mod-
ification of the Coombs test replace both the Isojima and Kibrick tests? (3) Do you beliel;e that 15 mg of prednisone daily, a relatively small dose, will be effective
in
controlled
studies,
or
wilf
larger
doses
he
required?
REFERENCES 1. Setchell. B.P.: The Mammalian Testis, Ithaca, New York, 1978, Cornell University Press. 2. Rumke, P. H., and Hellinga, G.: Autoantibodies against spermatozoa in sterile men, Am. .1. Clin. Pathol. 32~357.
1050.
3. Fjallbrant,
B.: Cervical
mucus
penetration
by human
sper-
Volume Number
140 8
matozoa treated with anti-spermatozoa1 antibodies from rabbit and man, Acta Obstet. Gynecol. Stand. 48:71, 1969. 4. Kremer, J., and Jager, S.: The sperm-cervical mucus contact test: A oreliminarv renort, Fertil. Steril. 27:335, 1976. 5. Hendry, W: F., Stedr&ia, J., Hughes, L., et al.: Steroid treatment of male subfertility caused by antisperm antibodies, Lancet 1:498, 1979. DR. JAMES D. KITCHIN, III, Charlottesville, Virginia. In a recent review, Jones’ discussed the role of immunologic factors in human infertility. He emphasized the widespread skepticism in regard to such an association and attributed this to confusion, contradictions, unjustifiable claims, and untidy methodology in the literature. However, there is no question but that isoimmunization in the woman and autoimmunization in the man against sperm and seminal plasma aritigens may be demonstrated by a variety of techniques. Dr. Mathur and co-authors have studied 60 couples with unexplained infertility-a large group if one agrees with the generally reported incidence of 5% to IO%, unexplained infertility among infertile couples. The patients have been placed in various groups according to the results of a passive hemagglutination and Coombs test method for determination of antisperm antibody. Quantitative determinations of IgE were performed and the various groups were compared with regard to the levels of this immunoglobulin detected. The authors emphasized that there is a lack of correlation between antibody titers (which reflect the presence of other immunoglobulins) and IgE levels. However, they showed an elevated IgE response in subjects with significantly elevated antisperm antibody titers. This is of considerable interest since it is not generally thought that IgE allergic hypersensitivity is an effective immune mechanism that could affect fertility. I am unaware of any evidence for IgE secretory antibody in cervical mucus, and if IgE is important in infertility, it would seem that it must be present in this location. Friberg’ reported a lack of correlation between antisperm antibody titers and IgE levels in the sera and seminal plasma of infertile men. The radioimmunosorbent test, which was used to quantitate IgE in this study, is subject to error in that low values tend to be overestimated because of nonspecific interference. I am told that a newer method, paper radioimmunosorbent test, is available and is more accurate. Further work is required before the data presented today can be interpreted. We need to determine whether elevated IgE levels really indicate clinically significant alterations of the immune mechanism which are of importance in fertility or whether such alterations are nonspecific and irrelevant. I would like to ask the authors about the patients who were treated with corticosteroids. Steroids may be responsible for various changes in the immune response, including decreased production of antibody, as well as
IgE and antisperm antibody titers in infertility
929
altered interaction of antibody with bound antigen. I ask the following with regard to the nine patients who conceived after prednisone therapy: (1) Did antibody titers and IgE levels decrease in all patients and did they differ from those in the nine patients who did not conceive? (2) Did the patients conceive during or after therapy? (3) If conception occurred after therapy, were antibody and IgE levels elevated in any patient? I would like to thank the authors for the opportunity to review their manuscript. In situations such as unexplained infertility, which by definition is poorly defined, it is difficult to know what sort of observations will ultimately prove to be useful in evaluation. Even though IgE has not been established as an important factor in infertility, further work, such as that reported, may ultimately show it to be so. REFERENCES
1. Jones, W. R.: Immunologic infertility-Fact or fiction? Fertil. Steril. 33:577, 1980. 2. Friberg, J.: Immunoglobulin concentration in serum and seminal fluid from men with and without sperm-agglutinating antibodies, AM. J. OBSTET. GYNECOL. 136:671, 1980. DR. WILLIAMSON AND DR. MATHUR (Closing). Indications for immunologic testing in infertility include a persistently poor postcoital test, such as those that show vibratory (nonprogressive) or no motility, no sperm, and inflamnatory cells unresponsive to appropriate therapy or with negative cultures. Also, lowered motility or otherwise poor semen quality, coexistence of other immunologic disorders and negative routine infertility assessments (normal seminal analysis, postcoital tests, patent tubes and ovulatory cycles) may be included. It is suggested that this be done prior to operative procedures, such as laparoscopy, unless other indications exist that suggest that it not be delayed. One cannot suggest that our tests should replace currently accepted methods until they are confirmed in other laboratories and a larger experience is gained to validate them. However, we have found that they are reproducible, and clinical experience is supporting our findings. The small dose of prednisone was adopted from the work of Shulman and associates (Shulman, S., Harlin, B., Davis, P., and Reyniak, J. V.: Immune infertility and new approaches to treatment, Fertil. Steril. 29:309, 1978) which has been used by others.’ We have tended to avoid the high dose but briefer and interrupted schedule of Shulman and associates2 because of mental aberrations and gastric distress reported in a few patients when it was used. The pregnancy rates of our series and of others appear to be comparable but are still low, which suggests that further manipulation of dosage may be desirable. Doubts have been expressed about the relevance of sperm antibodies to human infertility,3 mainly because of the unreliability of the existing methods. We devel-
930
Mathur et al.
oped two new methods for determination of sperm antibodies in infertile couples. The first, a modified Coombs test and passive hemagglutination, aids in determination of class specificity of antibodies bound to sperm and titers of agglutinating sperm antibodies present in sera and secretions. The second, a microspermocytotoxicity assay,” determines cytotoxic sperm antibody titers. Simultaneous use of both of these methods enables one to obtain information on the presence of sperm antibodies in local as well as systemic circulation. The results have been reproducible. Dr. Kitchin commented on the unusually high incidence of unexplained infertility in this series. Our patients are seen by referral and generally after the referring gynecologist has exhausted customary investigations and therapy. Of our patients subjected to immunologic testing, approximately 60% to 80% without other findings have positive tests, whereas outside referrals for testing run only about 20% positive. The lower limit of sensitivity of the radioimmunosorbent test is approximately 1.5 IU/ml. We have used this assay routinely in our laboratory and established the range of IgE levels in the sera from normal men and women. IgE levels in the sera of normal fertile men and women in the present study were comparable to those in our previous studies. Keview of the case histories of most of the infertile patients with elevated serum IgE levels failed to show evidence of other nonspecific allergic problems, such as hay fever or atopic dermatitis. Therefore, we believe that the elevated IgE levelsareassociated with infertility
i\ugust Am. J. Obstet.
15, 19x1 Qmcol.
in these subjects. However, further work is warranted before we can conclude that the IgE antibodies are specifically directed against sperm or seminal plasma antigens. We have outlined some changes in the antibody titers and IgE levels in two couples treated with prednisone. Studies are underway in order to determine whether prednisone treatment alters the IgE levels, even though it is known to do so in certain clinical conditions.6
REFERENCES
1. Hendry, W. F., Stedronska, J., Hughes, L., Cameron, K. M., and Pugh, R. C. B.: Steroid treatment of male subfertility caused by antisperm antibodies, Lancet 1:498, 1979. 2. Shulman, S., Harlin B., Davis, P., and Reyniak, J. V.: Immune infertility and new approaches to treatment, Fertil. Steril. 29:309, 1978. 3. Jones, W. R.: Immunologic infertility-Fact or tiction? Fertil. Steril. 33:577, 1980. 4. Mathur S., Williamson, H. O., Landgrebe, S. C., Smith, C. L., and Fudenberg, H. H.: Application of passive hemagglutination for evaluation of antisperm antibodies and a modified Coombs test for detecting male autoimmunity to sperm antigens, J. Immunol. Methods 30:381, 1979. 5. Mathur, S., Williamson, H. O., Derrick, F. C., Madyastha, P. R., Melchers, J. T., Holtz, G. L., Baker, E. R., Smith, C. L., and Fudenberg, H.: A new microassay for spermocytotoxic antibody: Comparison with passive hemagglutination assay for antisperm antibodies, J. Immunol. 126:905, 1981. 6. Settipane. G. A., Pudupakkam, R. K., and McGowan, J. H.: Corticosteroid effect on immunoglobulins, J. Allergy Clin. Immunol. 62:162, 1978.