Production of interferon gamma by lymphocytes exposed to antibody-coated spermatozoa: a mechanism for sperm antibody production in females*

Transcription

1 FERTILITY AND STERILITY Copyright c 1988 The American Fertility Society Vol. 5, No.3, September 1988 Printed in U.S.A. Production of interferon gamma by lymphocytes exposed to antibody-coated spermatozoa: a mechanism for sperm antibody production in females* StevenS. Witkin, Ph.D.t Department of Obstetrics and Gynecology, Cornell University Medical College, New York, New York The ability of sperm to induce lymphocytes from female donors to produce interferon gamma (IFN)' was measured. Of nine samples where immunoglobulin was bound to ejaculated motile sperm, five (55.6%) induced IFN production. In contrast, only 1 of35 (2.9%) sperm samples lacking bound immunoglobulins induced IFN (P <.5). Incubation of donor sperm with sperm antibody containing sera from nine male patients led, in seven cases (77.8%), to acquisition by sperm of the ability to induce IFN. Lymphocytes from eight of eight females tested all responded to sperm with bound antibodies. IFN induction of Ia antigen expression on macrophage may be required for recognition of processed sperm antigens by the female's T helper cells. Fertil Steril5:498, 1988 Conflicting data exist on the possible occurrence of cell-mediated immune responses to sperm. There are studies claiming to document lymphocyte activation by allogeneic and autologous sperm, 1 autologous but not allogeneic sperm, 2-4 and allogeneic but not autologous sperm. 5 The inability of sperm to activate lymphocytes has also been reported.s-8 To add to the confusion, lymphocyte activation was observed only in response to living sperm, 5 or only in the presence of dead sperm, 9 or sometimes with living and sometimes with dead sperm. 1 Because nonsperm cells present in the ejaculate, but not sperm, contain major histocompatibility complex class I and class II molecules on their surface, 1 and seminal fluid-derived somatic cells, but not motile sperm, stimulate in vitro lymphocyte responses, 6-8 I would suggest that purified, living motile sperm are not lymphocyte activators. Possible artefacts leading to apparent sperm-induced lymphocyte activation include the failure to Received January 8, 1988; revised and accepted May 31, *Supported by the National Institutes of Health grant HD t Reprint requests: Steven S. Witkin, Ph.D., Department of Obstetrics and Gynecology, Cornell University Medical College, 515 East 71st Street, New York, New York 121. completely remove somatic cells from semen prior to analysis, the use of nonviable or defective sperm that leak mitogenic acrosomal enzymes, 11 the use of nonhuman sera whose components can adhere to sperm and induce a lymphocytic response, and the nonspecific binding of IgG to dead sperm. 12 In this communication, it is reported that motile sperm are not lymphocyte activators as evidenced by a lack ofproduction of interferon gamma (IFN) after sperm-lymphocyte coincubations. However, lymphocyte activation is induced when antisperm antibodies are present on the surface of the sperm. Based on these data, a mechanism for the induction of an immune response to sperm in females is proposed. Subjects MATERIALS AND METHODS Forty-four male partners of infertile marriages referred to our laboratory for sperm antibody evaluation comprised our study population. All subjects had a normal sperm analysis (>2 X 1 6 sperm/ml, 75% progressive motility) and were culture negative for bacteria in their semen. Semen 498 Witkin Sperm-induced interferon production

2 from one donor male of proven fertility and lymphocytes from nine donor females with no history of fertility problems and negative for antisperm antibodies were utilized for all experiments. Sperm Antibody Assay The presence of antibodies on the surface of patients' and control motile sperm, purified from ejaculates by a swim-up technique, and in patients' sera was determined by enzyme-linked immunosorbent assay (ELISA). The use of motile sperm and their incubation with sera in suspension before binding to microtiter plates yield results comparable to the Immunobead binding assayy Antisperm antibody-containing donor sperm were prepared by incubating purified motile donor sperm (1 X 1 7 /ml) with 1:4 dilutions of antisperm antibody positive or negative sera from male patients, in a final volume of.2 ml. After 6 minutes at 37oC and 6 minutes at 4 oc, motile sperm were reisolated by the swim-up technique, washed three times in phosphate-buffered saline, and sperm concentration determined with a hemocytometer. The transfer of antibodies to sperm was confirmed by ELISA, and their location on the sperm visualized by the Immunobead binding assay. 13 Peripheral Blood Mononuclear Cell, Sperm and Seminal Fluid Incubations Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood of nine female donors by Ficoll-Hypaque gradient centrifugation. Incubations in wells of a microtiter plate contained, in a final value of.2 ml, 1 X 1 6 sperm or a 1:1 dilution of seminal fluid, 8.5 X 1 5 PB MCs, and 5% heat-inactivated serum obtained from the PBMC donor in RPM1164 medium with 25 J.LM Hepes buffer and L-glutamine supplemented with penicillin (1 u/ml), streptomycin (1 J.Lgf ml), and kanamycin (1 J.Lg/ml). After a 24-hour incubation at 37 C,.15 ml of culture supernatant was removed from each well and stored at -2oC. before analysis. Each sperm sample was analyzed in quadruplicate. Blank samples consisting of cultures without sperm and positive controls consisting of cultures incubated with the mitogen concanavalin A (Con A) instead of sperm were incubated in parallel with the test samples. Interferon Gamma Assay IFN in culture supernatants was quantitated by ELISA. Wells of a microtiter plate were coated with mouse monoclonal antibody to human interferon gamma (Interferon Sciences, New Brunswick, NJ) by overnight incubation in carbonate buffer, ph 9.8. The wells were washed three times with phosphate-buffered saline (PBS) containing.5% Tween 2, and.1 ml of culture supernatant was added. The plate was incubated in a 3TC waterbath for 9 minutes, the wells washed three times with PBS-Tween 2, and.1 ml of a 1:2 dilution in PBS-Tween 2 of rabbit antibody to human interferon gamma (Chemicon, El Segundo, CA) was added. After an additional 9 minutes at 37 C, the wells were again washed three times with PBS-Tween 2, and a 1:2 dilution in PBS Tween 2 of alkaline phosphatase-conjugated goat antibody to rabbit IgG (ICN, Lisle, IL) was added. After 6 minutes at 37 C, the wells were washed and bound alkaline phosphatase-conjugated antibody was quantitated, as previously described. 6 The mean value for each sperm sample was used to calculate the level of IFN production. Blank cultures were always negative, and Con A cultures were always positive for IFN. ELISA values were converted to units of IFN/ml (U /ml) by the use of a standard curve. Purified human interferon gamma (Cellular Products, Buffalo, NY) was diluted in 5% serum-rpmi medium over a concentration range of 1 to 1 U/ml and assayed in parallel to the test samples. Detection of IFN by ELISA has been shown to be a specific and sensitive assay, comparable to a bioassay. 14 Statistics Differences in IFN production by antibody-positive and antibody-negative sperm were evaluated by chi -square analysis with the Yates correction factor. RESULTS Measurement of IFN secretion by T lymphocytes is a more sensitive and specific assay for lymphocyte activation than is a lymphocyte proliferation assay. 15 Therefore, we examined the ability of sperm to stimulate IFN production during a 24- hour co-culture with PBMNs from a female donor. The results are shown in Table 1. With ejaculated sperm lacking surface-bound antibodies, only 1 in 35 samples (2.9%) induced IFN production. The one positive sample was from a man with IgA antisperm antibodies in his serum. Sperm samples from none of the 11 other men with circulating Vol. 5, No.3, September 1988 Witkin Sperm-induced interferon production 499

3 Table 1 Lymphocyte Activation By Serpmatozoa" Interferon gamma productionb Sperm antibody No. positive/ status total no. Range % U/ml Antibodies on sperm 5/9 (55.6) Antibodies in serum 1/12 (8.3) 11 No antibodies /23 () Spermatozoa were incubated with peripheral blood mononuclear cells from a female donor. b Interferon in culture supernatants was measured by ELISA. antisperm antibodies induced IFN. This confirmed that purified motile sperm do not activate T lymphocytes. In marked contrast, however, of the nine sperm samples that contained bound antibodies, five (55.6%) induced IFN production. This difference from antibody negative sperm was highly significant (P <.5). Three of the positive samples contained only IgA on the sperm surface, the fourth contained only IgG, while the fifth was positive for IgG, IgA, and IgM. To examine whether lymphocyte activation might have been due to nonsperm components of the ejaculates that adhered to sperm antibody-positive sperm, the ability of the sperm-free seminal fluids to induce IFN was also tested (Table 2). Only 1 of the 35 seminal fluids examined induced detectable levels of IFN. Sperm from this same patient was also an IFN inducer, suggesting that in this case, but not in the other men with antisperm antibodies, nonsperm components may have contributed to the observed effect. No IFN was detected in this man's seminal fluid (data not given). To further analyze the relationship between antibodies on sperm and their ability to activate lymphocytes, I tested the ability of antisperm antibody-containing sera to convert donor sperm to an IFN inducer. The results are shown in Table 3. When donor sperm were incubated in five sera that were negative for antisperm antibodies they di.d not Table 2 Sperm antibody status Lymphocyte Activation by Cell-Free Seminal Fluid Interferon gamma production No. positive/ total no. U/ml Table 3 Lymphocyte Activation After Transfer of Antibodies to Donor Spermatozoa" Sperm antibody status No. Positive 9 Negative 5 Interferon gamma production No. positive Range % U/ml 7 (77.8) 8-24 Donor spermatozoa were incubated for 12 minutes with sperm antibody-containing sera from males. The sperm were then washed, incubated with donor lymphocytes, and interferon gamma production was measured. acquire the ability to activate T lymphocytes. However, seven of the nine sera (77.8%) that contained antisperm antibodies successfully converted sperm from nonactivators to activators of T lymphocyte IFN secretion. Of the positive sperm samples, four contained bound IgG, two contained IgA and IgM, while the last was positive for IgG and IgM. Both head-directed and tail-directed antibodies on sperm were able to induce IFN synthesis. To assess whether IFN production is a general characteristic of lymphocytes after exposure to sperm with bound antibodies, the above experiment was repeated using seven different female lymphocyte donors. The results (Table 4) confirmed that donor sperm can be converted to IFN inducers by attachment of antibodies to the sperm surface. Table 4 Activation of Lymphocytes from Different Female Donors by Sperm with Bound Antibodies" Lymphocyte Sperm-bound Interferon gamma donor antibodies (U/ml) <l Antibodies on sperm Antibodies in serum No antibodies % 1/8 (12.5) /9 () /18 () 3. a Donor sperm were incubated with sera containing sperm antibodies or with sperm antibody-free sera. The sperm were then washed, incubated with lymphocytes, and interferon production was measured. 5 Witkin Sperm-induced interferon production

4 DISCUSSION The inability of human sperm to typically activate T lymphocytes from a female undoubtedly contributes to the lack of antisperm antibody production in most sexually active women. The association of sperm with surface-bound antibodies converted the gametes into inducers of T cell activation and, thereby, increases the likelihood of a sperm specific immune response in the female partner. Enhanced immunogenicity of antigens by virtue of their association with specific antibody has been demonstrated in other systems The increased responsiveness requires the presence of intact immunoglobulin Fe regions and macrophages and may be due to enhanced macrophage uptake and processing of antigen when it is combined with antibody18 or the immune-enhancing effect of Fe fragments produced by macrophage processing of antigen-antibody complexes.19 It is unlikely that the results reported here were due to a greater bacterial contamination of the semen from patients with antibodies on their sperm than from the other semen samples. The swim-up technique used to isolate motile sperm is effective in eliminating bacteria from sperm samples.2 Possible viral contributions to the IFN induction cannot be excluded, and this seems an attractive explanation for the positive result obtained with one cell-free seminal fluid. I am unaware of any data linking the occurrence of antibodies on sperm to the presence of viruses in semen. Viruses or other microorganisms that attach to sperm, however, would probably also convert sperm into lymphocyte activators. These observations lead me to propose a mechanism for antisperm antibody formation in females (Fig. 1). After coitus, sperm interaction with the complement system leads to the formation of activation products which are chemotactic for macrophages. 21 The congregated macrophages phagocytose spermatozoa, eliminating them from the va:. gina. Although the macrophages may degrade the spermatozoa and process and display sperm antigens on its surface, a sperm antigen-specific immune response is not initiated in most women since synthesis by the macrophages of a second class of molecules required for immune activation, Ia antigens, has not been induced. T lymphocytes are capable of recognizing and responding to antigens presented by macrophages only when the antigens are associated with Ia antigens, a gene prod- Vol. 5, No.3, September 1988 ~= """ ~ IFNy. Figure 1 Mechanism for the induction of sperm antibodies in the female genital tract. In most females, sperm do not activate T lymphocytes (T) and are phagocytosed by macrophages (Mt/>). The macrophages may degrade sperm and display processed sperm antigens on its surface (D, D), but in the absence of concomitant Ia antigen expression (~, ~) by the M<t>, T helper cells (T H) are unable to react with the sperm antigens and no antigen-specific immune response is initiated. If antibodies (Y) or microorganisms (, D) are present on sperm, however, the gametes activate T cells to produce interferon gamma (IFN-y). This induces Ia expression on macrophages, thereby allowing T helper cell recognition of antigen and subsequent stimulation of B lymphocytes (B) to produce specific antibodies. uct ofthe major histocompatibility class II locus.22 In a minority of women, however, due to the association of immune system activators with sperm, semen deposition leads both to the induction of IFN production by T cells and to the generation of macrophage chemotactic factor. Since IFN is a potent inducer of Ia antigen expression on macrophages, 23 the macrophages now display on their surfaces both sperm antigens and Ia antigens. T helper lymphocytes recognize and interact with the!a-sperm antigen complex and stimulate a sperm antigenspecific antibody response by B lymphocytes. A similar mechanism for antisperm antibody induction may also be operative in the male genital tract, where T cells, B cells, and macrophages are present. 24 Increased macrophage phagocytosis of sperm after pretreatment of the sperm with sera containing antisperm antibodies has been reported. 25 This mechanism, should it be validated by further experimental analyses, would have significant practical applications. The deliberate induction of Ia antigen expression on macrophages following antigen deposition in the female genital tract may ultimately be of value for the development of an immunological contraceptive. Conversely, inhibi- Witkin Sperm-induced interferon production 51

5 tors of Ia antigen expression may reduce or eliminate sperm antigen-specific antibody production in the genital tracts of sensitized individuals. Acknowledgments. The outstanding technical assistance of Ms. Jan Jeremias and Ms. Debra Viti is gratefully acknowledged. REFERENCES 1. Thestrup-Pedersen K, Husted S, Hjort T: Lymphocytetransformation test with spermatozoal antigens in men from infertile couples. I. A methodical study of different spermatozoal preparations as antigens. Int J Fertil 21:218, Festenstein H, Halim K, Arnaiz-Villena A: Selection of haploid spermatozoa and its application to HLA-D typing. Scand J lmmunol6:511, Halim K, Festenstein H, Farrant J: HLA-D typing of human lymphocytes using frozen and thawed spermatozoa. J Immunol Meth 25:31, Mathur S, Pathak S, Kandel A, Ziegler J, Rust PF, Williamson HO: Leukocyte migration inhibitory factor (LIF) to sperm from autoimmune men in infertile couples. J lmmunol136:4444, Levis WR, Whalen JJ, Sherins RJ: Mixed cultures of sperm and leukocytes as a measure of histocompatibility in man. Science 191:32, Misko IS, Boettcher B, Roberts TK, Kay DJ: Spermatozoa cells in human semen do not stimulate allogeneic leukocytes in culture. Lancet 1:56, Misko IS, Boettcher B, Gruszynski R: Lymphocyte stimulation by allogeneic cells in sperm-free ejaculates. Lancet 2: 479, Rodriguez-Cordoba S, Arnaiz-Villena A: Human seminal cells other than spermatozoa stimulate lymphocyte cultures. Tissue Antigen 19:313, Dor J, Nebel LA, Soffer Y, Mashiach S, Serr DM: Cell mediated and local immunity to spermatozoa in infertility. Int J Fertil 24:94, Anderson DJ, Bach DL, Yunis EJ, DeWolfWC: Major histocompatibility antigens are not expressed on human epididymal sperm. J Immunol129:452, Kurpisz M, Witt M: The role of acrosomal enzymes in lymphocyte stimulation by spermatozoa. Am J Reprod Immunol5:129, 12. Richards JM, Witkin SS: Non-immune lgg binding to the surface of spermatozoa by disulphide rearrangement. Clin Exp lmmunol58:493, 13. Bronson R, Cooper G, Rosenfeld D, Witkin SS: Detection of spontaneously occurring sperm-directed antibodies in infertile couples by Immunobead binding and enzymelinked immunosorbant assay. Ann NY Acad Sci 438:54, 14. Chang TW, McKinney S, Liu V, Kuag PC, Vilcek J, Le J: Use of monoclonal antibodies as sensitive and specific probes for biologically active human interferon. Proc N atl Acad Sci USA 81:5219, 15. Hao XS, Le J, Vilcek J, Chang TW: Determination of human T cell activity in response to allogeneic cells and mitogens. An immuno-chemical assay for interferon is more sensitive and specific than a proliferation assay. J lmmunol Meth 92:59, Banks KL: The effect of antibody on antigen-induced lymphocyte transformation. J Immunol11:79, Collisson EW, Andersson B, Ronnholm M, Lamon EW: Potentiation of antibody responses by specific lgm: Specificity and thymus dependency. Cell Immunol 791:44, Oppenheim JJ: Modulation of in vitro lymphocyte transformation by antibodies: Enhancement by antigen-antibody complexes and inhibition by antibody excess. Cell Immunol3:341, Morgan EL, Weigle WO: Aggregated human-globulin-induced proliferation and polyclonal activation of murine B lymphocytes. J Immunol125:226, Kuzan FB, Hillier SL, Zarutskie PW: Comparison of three wash techniques for the removal of microorganisms from semen. Obstet Gynecol 7:836, Maroni ES, Wilkinson PC: Selective chemotaxis of macrophages towards human and guinea pig spermatozoa. J Reprod Fertil27:149, Unanue ER, Beller DI, LuCY, Allen PM: Antigen presentation: Comments on its regulation and mechanism. JImmunol132:1, 23. Beller DI: Functional significance of the regulation of macrophage Ia expression. Eur J Immunol14:138, 24. El-Demiry Ml, Hargreave TB, Busuttil A, Elton R, James K, Chisholm GD: Immunocompetent cells in human testis in health and disease. Fertil Steril48:47, London SN, Haney AF, Weinberg JB: Macrophages and infertility: enhancement of human macrophage-mediated sperm killing by antisperm antibodies. Fertil Steril43:274, Witkin Sperm-induced interferon production