- Email: [email protected]
Inhibition of complement-mediated cytotoxicity of antisera by fluid secreted by the seminal vesicle of the house mouse
Journal of Reproductive Immunology, 3 (1981) 109-116
109
Elsevier/North-Holland Biomedical Press
INHIBITION OF COMPLEMENT-MEDIATED CYTOTOXICITY OF ANTISERA BY FLUID SECRETED BY THE SEMINAL VESICLE OF THE HOUSE MOUSE
B. PEITZ * and D. BENNETT Laboratory of Developmental Genetics, Sloan-Kettering Institute for Cancer Research, 425 East 68th Street, New York, N Y 10021, U.S.A.
(Received 20 October 1980; accepted 19 January 1981)
The fluid from the seminal vesicles of the house mouse inhibited complement-mediated cytotoxicity of antisera against both sperm and lymphocytes. This inhibition was not reduced by heating or by absorption with sperm. Fractionation of the seminal vesicle fluid on Sephadex G-100 columns revealed three peaks of inhibitory activity, one of which appeared in the void volume of the columns. This inhibitory action of the seminal vesicle fluid may protec t sperm from immunological attack in the female reproductive tract. It could explain the observation that immunization of female house mice with sperm does not prevent pregnancy. The relationship between this activity in the seminal vesicles of house mice, which is directed against the cytotoxicity of antisera, and the inhibition of cellmediated responses to sperm reported for human and bull semen have not been investigated.
INTRODUCTION The seminal plasma produced by the accessory glands of the male reproductive tract does not seem to be essential for fertilization since epididymal sperm have been used successfully for artificial insemination (Leckie et al., 1973). This has led to a series o f questions on the role o f seminal plasma in fertility (Rodger, 1975). Recently it has been reported (Pang et al., 1979) that surgical removal of the seminal vesicles, but not the prostate gland, of mice reduced fertility. This indicates that fertilization following normal mating requires some components of seminal plasma. The actual nature of the seminal components required and their role in fertilization is still speculative. Several reports have indicated that seminal plasma contains a factor which actively suppresses lymphocyte responses to antigens (Lord et al., 1977; Marcus et al., 1977, 1979; Pitout and Jordan, 1976; Prakash et al., 1976; Stites and Erickson, 1975). This immunosuppressive activity has been isolated from both human and bull semen, and molecular weight estimates using gel f'dtration on Sephadex range from 3000 to 200 000. The lymphocyte responses inhibited include DNA synthesis stimulated by a variety of mitogens and antigens (concancavalin A, phytohemagglutinin, poke weed mitogen, tetanus toxoid, Candida albicans extract and allogenic cells). The immunosuppressive fractions of seminal plasma also inhibited IgM production, lymphocyte cytotoxicity * Present address: Department of Biology, California State University, Los Angeles, CA 90032, U.S.A. 0 165-0378181/0000-0000/$02.50 © Elsevier/North-Holland Biomedical Press
110 (Lord et al., 1977) and E-rosette formation (Marcus et al., 1979). The presence of immunosuppressive substances in semen may explain why females do not normally develop antibodies to sperm after repeated matings, even though sperm are antigenic and the female reproductive tract is capable of an immunologic response (for reviews, see Smith and Beer, 1976; Beer and Billingham, 1976). It has been postulated that infertility in women with antibodies to sperm in the vaginal and uterine fluids may be caused by the failure of immunosuppressive action (Lord et al., 1977). This report describes the inhibition of complement-mediated antibody cytotoxicity by a soluble fraction of the secretion of the mouse seminal vesicles. This inhibitory activity appeared to be due to a number of proteins with different molecular weights, which showed no binding to sperm. These data extend our knowledge of the immunosuppressive effects of seminal plasma proteins to another species, the house mouse, and another immunological response, complement-mediated antibody cytotoxicity. MATERIALSAND METHODS
Animals The study was done primarily with 8 - 1 0 week old mald mice of the BTBRTF/Nev stock. However, seminal vesicle secretions from one outbred strain, ICR, were also tested for activity. All of the fractionations on Sephadex G-100 were done with material from BRTBRTF/Nev males. Since ejaculated semen is difficult to obtain in rodents, material for this study was collected from dissected accessory glands of the male reproductive tract. The seminal vesicles were used because they are the largest glands of the accessory gland complex in the mouse. It has also been estimated that they are the major source of proteins in the ejaculate in the rat (Geiger et al., 1974)and they are necessary for fertility in mice (Pang et al., 1979). In addition, the seminal vesicle fluid can readily be removed from these glands and recovered in quantity. Mice were anesthetized with ether, to reduce the chances of spontaneous ejaculation and loss of material, and then killed by cervical dislocation. The seminal vesicles were teased away from the coagulating glands and removed as close to the base as possible. They were weighed, and the secreted fluid striped into 1 ml of cold phosphate-buffered saline, pH 7.4, (PBS) for direct tests. The resulting suspension was mixed thoroughly and allowed to precipitate in the cold from 1 hour to overnight. The length of time in the cold (up to 24 h) did not affect the activity but longer precipitation times guaranteed a clear supernatant. After precipitation the material was centrifuged for 25 min at 29 000 rev./min at 4°C and the supernatant was used in the tests. The amount of protein in the solution was calculated from measurements of the optical density at 280 nm, using bovine serum albumin as a standard. Since only 10-17 mg of soluble protein was recovered from the seminal vesicles of a single mouse, the secretions from 6 - 1 0 males were pooled for fractionation on Sephadex G-100 columns. The pooled samples were treated in the same way as those from single males and the supernatant was applied to the columns. Inhibition of complement-mediated cytotoxicity Two test systems were used to measure the inhibitory activity: the complement-
111
mediated cytotoxicity test on sperm (Yanagisawa et al., 1974; Artzt and Bennet, 1977) and a standard H-2 typing test on lymphocytes. In both systems a volume of solution containing seminal vesicle proteins equal to the volume of serum used was added to the test suspension before the addition of complement. This increased the final volume of the test suspension by 33% and diluted both the serum and complement. The density of sperm in the initial suspension was adjusted so that the final density of cells in the test suspension was about twice that originally described (about 4 X lo6 sperm/ml incubated with serum and complement). These changes in cell density were made so that the test would be easier to count. For lymphocyte tests the final cell number was adjusted to 2 X 106/ml. The changes in serum and complement concentration were taken into account by running cell controls, complement controls and serum with complement controls with every test. These three controls were made by adding an appropriate volume of medium instead of the solution of seminal vesicle proteins. Rabbit complement for sperm tests was prepared as described by Artzt and Bennett (1977) and unabsorbed nontoxic rabbit serum was used as a complement source for lymphocyte tests. Antisperm sera were prepared by injecting sperm into BTBRTF/Nev females as described by Yanagisawa et al. (1974). The serum used in this study was a highly cytotoxic non-specific antisperm serum. A standard H-2 typing serum (A anti-EL4, a gift of Dr. E.A. Boyse) was used for lymphocyte tests. Anti-sperm sera was generally used at dilutions of l/8 and l/16 so that inhibition of the activity at two levels of antisera could be assessed. The average of the two values was used to calculated specific inhibition of the tested sample. Doubling dilutions of the supernatant solution prepared from the seminal vesicle were tested for inhibitory activity with anti-sperm sera. The anti-EL4 serum was tested at doubling dilutions of l/100 to l/800 so that inhibition at the maximum activity of the serum could be determined. The inhibitory activity of the seminal vesicle fluid was calculated as a percentage of the activity of the normal serum by the following formula: Inhibition = cytotoxic index of normal serum with complement - cytotoxic index of serum with seminal vesicle fluid cytotoxic index of normal serum with complement
Heat inactivation Samples of the solution of soluble seminal vesicle proteins were incubated at temperatures of 25,37,56 and 85’C for 30 min before being used in the cytotoxic tests. Sperm absorptions Samples of the solution of soluble seminal vesicle proteins were absorbed with sperm by the quantitative absorption method described by Artzt and Bennett (1977). The numbers of sperm used ranged from 15 X lo6 to 80 X lo6 for each 0.05 ml of solution. The absorbed solutions were then tested in the cytotoxic test systems to determine whether sperm removed the active components. Fractionation Sephadex G-100 (Pharmacia) was suspended in PBS, washed three times and poured into columns 2 cm in diameter. The height of the columns, flow rate of buffer through
112 the colunm, and amount of protein applied were all varied in attempts to achieve the best separation possible. The columns were run at 4°C. The optical density at 280 nm of each 2 ml fraction was determined and a protein prof'fle was graphed for each column. Fractions for columns with good separation between peaks were pooled, concentrated by pressure dialysis on an Amicon Ftitration system and tested for inhibitory activity. RESULTS
Inhibitory activity in solutions of soluble seminal vesicle proteins Only 10-17 mg of soluble protein was recovered when the seminal vesicle secretions of a single male were mixed with PBS. Since fluid recovery was sometimes difficult and incomplete when precipitation of the insoluble material occurred rapidly some of the soluble proteins were undoubtedly lost. Although some attempts were made to reduce precipitation losses none were successful and the amount of protein lost was not determined. Because of this loss of protein, the measured inhibitory activity was probably lower than that present in the fluids in vivo. Solutions prepared from seminal vesicles of individual males inhibited serum cytotoxicity by 40-80%. These inhibitory values were approximately the same with both the anti-sperm serum and anti-H-2 serum. The seminal vesicle preparations themselves were not cytotoxic to either sperm of lymphocytes. Animals of an outbred stock (ICR) had similar levels of activity as inbred mice (BTBRTF/Nev). The inhibitory activity of solutions prepared from the glands of a single male were tested following serial dilutions from 1/2 to 1/16. Activity decreased very rapidly with dilution and no activity was detectable in dilutions greater than 1/8. Tests of the more concentrated solutions, prepared for fractionation on Sephadex G-100, showed inhibition of 90-100% of serum cytotoxicity.
Heat inactivation Tests of heated aliquots of seminal vesicle protein showed that none of the temperatures used caused any decrease in inhibitory activity. Thus, this activity was heat stable to 85°C.
Sperm absorptions Absorptions were done on aliquots of soluble seminal vesicle protein solutions from 6 different males. The solution from each male was tested unabsorbed and absorbed with different numbers of sperm. The numbers of sperm used for absorption ranged from 15 × 106/0.05 ml of solution to 60 X 106/0.05 ml of solution. Inhibition (% of normal serum) for unabsorbed solutions and samples absorbed with various numbers of sperm were averaged, and appear in Table 1. These results clearly show that absorption with sperm did not reduce inhibitory activity. In fact, inhibitory activity seemed to be increased following absorption. It was possible that some other component of the mixture of soluble proteins that restricts the inhibition was removed by the absorption. But further studies of this interaction were not conducted. In any case, the inhibitor was not removed and therefore does not bind to sperm.
113 TABLE 1 Inhibition of complement-mediated antisera cytotoxicity by solutions of soluble seminal vesicle proteins after absorption with spermatozoa Inhibition (% decrease in normal serum activity) (m -+ S.E.) Control: unabsorbed solution o f soluble seminal vesicle proteins Solution absorbed with 1 5 - 1 0 X 106 spermatozoa 2 5 - 3 5 × 106 spermatozoa 4 0 - 5 0 × 106 spermatozoa 5 5 - 6 0 X 106 spermatozoa
6
45.8 -+ 16
4 5 4 5
7915 71.6 96.0 82.7
-+ 6 -+ 14 -+ 9 -+ 18
Fractionation
Initial fractionations indicated that flow rates of between 6 and 16 ml/h did not affect the protein pattern. Both 1- and 2-m columns were initially tested and longer columns gave better separations. Samples containing 10-130 mg of protein were separated on 2 m × 2 cm Sephadex G-100 columns. Best separations occurred with samples of 5 0 - 9 0 mg. A typical protein prof'lle is shown in Fig. 1. There are 5 protein peaks, each numbered for identification purposes. Peak 1 was in the void volume and was only detectable when more than 40 mg protein was applied to the column. When samples containing over 30 mg of protein were fractionated, peaks 3, 4 and 5 were not well separated from each other. When the tubes in each peak were pooled, concentrated, and tested for inhibition of 0.8 E c 0.6 O G) GI
I I I
g o.4 g ~ 0.2
t 0,4
o.2~"..5
h o
o
I
4to
66
~o
~6o
Fraction
~o
Peak 1 2 3 4 5 Fig. 1. Typical protein profile from fractionation o f seminal vesicle proteins on Sephadex G-100 showing 5 main protein peaks and'3 peaks o f inhibitory activity..
114 cytotoxic activity, peak 5 showed no inhibition. Peaks 3 and 4 were most active but 1 and 2 also showed some inhibition (summary of the results of 9 columns). Since the inhibitory activity could not be localized in one major protein peak, the fractions were assayed for activity by pooling every 5 tubes, starting with tube 40, concentrating the protein and testing it to produce a profile of inhibitory activity for the whole column effluent. A typical activity pattern is presented in Fig. 1. It can clearly be seen that there are three peaks of inihibitory activity. The major peak of activity (A in Fig. 1) coincides most closely with protein peak 1, which was in the void volume of the column and therefore had a molecular weight of over 150 000. Some of this activity may also by detected in protein peak 2, probably as a result of incomplete separation. The two other activity peaks coincide with protein peaks 3 and 4. DISCUSSION These data indicate that the secretions of the seminal vesicles of the house mouse contain a factor that can protect sperm from complement-mediated antibody cytotoxicity. The major peak of this immunosuppressive activity was found in a soluble fraction which appeared in the void volume after gel filtration on Sephadex G-100. This indicated that the material has a molecular weight of over 150000. Two other peaks of activity with lower molecular weight were also identified. There are several possible explanations for the appearance of three peaks of activity. The active material of high molecular weight may either become degraded and release the lower molecular weight fragments, or conversely the high molecular weight component could be an aggregate of the lower molecular weight fractions. Alternatively, there may be three unrelated proteins with immunosuppressive activity. Fractionations of this material following treatment with urea or high salt concentrations to prevent aggregation have not yet been performed. While the association of activity with fractions of high molecular weight absorbing light at 280 nm implies that the inhibitory materials are proteins, they could also be glycoproteins. Studies by a number of investigators indicate that human seminal plasma contains several distinct immunosuppressive factors. The inhibitor of cell-mediated immune responses isolated from human seminal plasma has been shown to be heat stable, rich in carbohydrates and with a high molecular weight (Marcus et al., 1979). Human seminal plasma also contains a heat labile factor with a molecular weight of between 20 000 and 60 000 which reduces the activity of complement components C 1 and C3 by more than 50% (Petersen et al., 1980). It is not clear whether these varied components interact in any way. Since the seminal fractions from the house mouse inhibit complement-mediated cytotoxicity of both anti-sperm and anti-H-2 sera the activity was probably directed against some general feature of antibodies or some component of the complement system. However, the specific component has not yet been identified. It was clear from the absorption studies with sperm that the active material did not have a specific binding site on sperm that was necessary for inhibition to occur. Further tests will be needed to determine why the absorptions with sperm increased the inhibitory activity. These data indicate that following natural matings sperm are protected against antibody attack. This would explain why immunizations against sperm have not totally pre-
115 vented pregnancy, even in female mice that developed a high titer of anti-sperm antibodies (McLaren, 1966; Bell and McLaren, 1970). Even though the inhibitory material does not bind to sperm it could still play a protective role during the passage of sperm through the uterus since insemination is intra-uterine in the house mouse. The absence of thins factor from mixtures of epididymal sperm used for artificial insemination may account for the relatively low success rate of artificial insemination as compared to natural matings in the mouse, as well as the decrease in fertility following surgical removal of the seminal vesicle reported by Pang et al. (1979). Several studies have indicated that mouse semen inhibits cell-mediated immune responses. Mouse sperm inhibit the mixed leukocyte reaction (Erickson and Stites, 1975) and mouse seminal plasma can prevent lymphocyte activation (Stites and Erickson, 1975). Further studies are needed to determine the relationship between the seminal factors that inhibit cell-mediated immunological responses and those which inhibit complement-mediated antisera cytotoxicity. In addition, a comparative study of the immunosuppressive fractions from the semen of a number of species would be most informative. ACKNOWLEDGMENT B.P. was supported by NIH Fellowship No. 5 F32 HD05033-02. REFERENCES Artzt, K. and Bennet, D. (1977) Serological analysis of sperm of antigenically cross reacting T/thaplotypes and their recombinants. Immunogenetics 5, 97-107. Beer, A.E. and Billingham, R.E. (1976) The Immunobiology of Mammalian Reproduction. PrenticeHall, Englewood Cliffs, N.J. Bell, E.B. and McLaren, A. (1970) Reduction of fertility in female mice isoimmunized with a subcellular sperm fraction. J. Reprod. Fertil. 22, 345-356. Erickson, R.P. and Stites, D.P. (1975) Effects of murine spermatozoa on the mixed leukocyte reaction of mice. Transplantation 20, 263-265. Geiger, B., Frensdorff, A. and Kraicer, P.F. (1974) The soluble proteins of rat seminal vesicle fluid; some physico-chemical and immunological properties. Immunology 27, 729-738. Leckie, P.A., Watson, J.G. and Chaykin, S. (1973) An improved method for the artificial insemination of the mouse (Mus musculus). Biol Reprod. 9, 420-425. Lord, E.M., Sensabaugh, G.F. and Stites, D.P. (1977) Immunosuppressive activity of human seminal plasma. I. Inhibition of in vitro lymphocyte activation. J. Immunol. 118, 1704-1711. Marcus, Z.H., Freisheim, J., Herman, J.H., and Hess, E.V. (1977) Inhibition of mitogen induced blast transformation by male genital components. In Immunological Influence on Human Fertility (Boettcher, B., ed.), pp. 333-339. Academic Press, New York. Marcus, Z.H., Hess, E.V., Herman, J.H., Troiano, P. and Freisheim, J. (1979) In vitro studies in reproductive immunology. 2. Demonstrations of the inhibitory effect of male genital tract constituents on PHA-stimulated mitogenesis and E-rosette formation of human lymphocytes. J. Reprod. Immunol. 1, 97-107. McLaren, A. (1966) Studies on the isoimmunization of mice with spermatozoa. Fertil. Steril. 17, 492-499. Pang, S.F., Chow, P.H. and Wong, T.M. (1979) The role of the seminal vesicles, coagulating glands and prostate glands on the fertility and fecundity of mice. J. Reprod. Fertil. 56, 129-132. Petersen, B.H., Lammel, C.J., Stites, D.P. and Brooks, G.F. (1980) Human seminal plasma inhibition of complement. J. Lab. Clin. Med. 96,582-591.
116 Pitout, M.J. and Jordan, J.H. (1976) Partial purification of an antimitogenic factor from human semen. Int. J. Biochem. 7, 149-151. Prakash, C., Coutinho,'A. and M611er, G. (1976) Inhibition of in vitro immune responses by a fraction from seminal plasma. Scand. J. Immunol. 5, 72-85. Rodger, J.C. (1975) Seminal plasma, an unnecessary evil? Theriogenology 3,237-247. Smith, W.G. and Beer, A.E. (1976) Current concepts of reproductive immunobiology. J. Perinat. Med. 4, 59-71. Stites, D.P., and Erickson, R.P. (!975) Suppressive effect of seminal plasma on lymphocyte activation. Nature (London) 253,727-729. Yanagisawa, K., Bennett, D., Boyse, E.A., Dunn, L.C. and Dimeo, A. (1974) Serological identification of sperm antigens specified by lethal t-alleles in the mouse. Immunogenetics 1, 57-67.
109
Elsevier/North-Holland Biomedical Press
INHIBITION OF COMPLEMENT-MEDIATED CYTOTOXICITY OF ANTISERA BY FLUID SECRETED BY THE SEMINAL VESICLE OF THE HOUSE MOUSE
B. PEITZ * and D. BENNETT Laboratory of Developmental Genetics, Sloan-Kettering Institute for Cancer Research, 425 East 68th Street, New York, N Y 10021, U.S.A.
(Received 20 October 1980; accepted 19 January 1981)
The fluid from the seminal vesicles of the house mouse inhibited complement-mediated cytotoxicity of antisera against both sperm and lymphocytes. This inhibition was not reduced by heating or by absorption with sperm. Fractionation of the seminal vesicle fluid on Sephadex G-100 columns revealed three peaks of inhibitory activity, one of which appeared in the void volume of the columns. This inhibitory action of the seminal vesicle fluid may protec t sperm from immunological attack in the female reproductive tract. It could explain the observation that immunization of female house mice with sperm does not prevent pregnancy. The relationship between this activity in the seminal vesicles of house mice, which is directed against the cytotoxicity of antisera, and the inhibition of cellmediated responses to sperm reported for human and bull semen have not been investigated.
INTRODUCTION The seminal plasma produced by the accessory glands of the male reproductive tract does not seem to be essential for fertilization since epididymal sperm have been used successfully for artificial insemination (Leckie et al., 1973). This has led to a series o f questions on the role o f seminal plasma in fertility (Rodger, 1975). Recently it has been reported (Pang et al., 1979) that surgical removal of the seminal vesicles, but not the prostate gland, of mice reduced fertility. This indicates that fertilization following normal mating requires some components of seminal plasma. The actual nature of the seminal components required and their role in fertilization is still speculative. Several reports have indicated that seminal plasma contains a factor which actively suppresses lymphocyte responses to antigens (Lord et al., 1977; Marcus et al., 1977, 1979; Pitout and Jordan, 1976; Prakash et al., 1976; Stites and Erickson, 1975). This immunosuppressive activity has been isolated from both human and bull semen, and molecular weight estimates using gel f'dtration on Sephadex range from 3000 to 200 000. The lymphocyte responses inhibited include DNA synthesis stimulated by a variety of mitogens and antigens (concancavalin A, phytohemagglutinin, poke weed mitogen, tetanus toxoid, Candida albicans extract and allogenic cells). The immunosuppressive fractions of seminal plasma also inhibited IgM production, lymphocyte cytotoxicity * Present address: Department of Biology, California State University, Los Angeles, CA 90032, U.S.A. 0 165-0378181/0000-0000/$02.50 © Elsevier/North-Holland Biomedical Press
110 (Lord et al., 1977) and E-rosette formation (Marcus et al., 1979). The presence of immunosuppressive substances in semen may explain why females do not normally develop antibodies to sperm after repeated matings, even though sperm are antigenic and the female reproductive tract is capable of an immunologic response (for reviews, see Smith and Beer, 1976; Beer and Billingham, 1976). It has been postulated that infertility in women with antibodies to sperm in the vaginal and uterine fluids may be caused by the failure of immunosuppressive action (Lord et al., 1977). This report describes the inhibition of complement-mediated antibody cytotoxicity by a soluble fraction of the secretion of the mouse seminal vesicles. This inhibitory activity appeared to be due to a number of proteins with different molecular weights, which showed no binding to sperm. These data extend our knowledge of the immunosuppressive effects of seminal plasma proteins to another species, the house mouse, and another immunological response, complement-mediated antibody cytotoxicity. MATERIALSAND METHODS
Animals The study was done primarily with 8 - 1 0 week old mald mice of the BTBRTF/Nev stock. However, seminal vesicle secretions from one outbred strain, ICR, were also tested for activity. All of the fractionations on Sephadex G-100 were done with material from BRTBRTF/Nev males. Since ejaculated semen is difficult to obtain in rodents, material for this study was collected from dissected accessory glands of the male reproductive tract. The seminal vesicles were used because they are the largest glands of the accessory gland complex in the mouse. It has also been estimated that they are the major source of proteins in the ejaculate in the rat (Geiger et al., 1974)and they are necessary for fertility in mice (Pang et al., 1979). In addition, the seminal vesicle fluid can readily be removed from these glands and recovered in quantity. Mice were anesthetized with ether, to reduce the chances of spontaneous ejaculation and loss of material, and then killed by cervical dislocation. The seminal vesicles were teased away from the coagulating glands and removed as close to the base as possible. They were weighed, and the secreted fluid striped into 1 ml of cold phosphate-buffered saline, pH 7.4, (PBS) for direct tests. The resulting suspension was mixed thoroughly and allowed to precipitate in the cold from 1 hour to overnight. The length of time in the cold (up to 24 h) did not affect the activity but longer precipitation times guaranteed a clear supernatant. After precipitation the material was centrifuged for 25 min at 29 000 rev./min at 4°C and the supernatant was used in the tests. The amount of protein in the solution was calculated from measurements of the optical density at 280 nm, using bovine serum albumin as a standard. Since only 10-17 mg of soluble protein was recovered from the seminal vesicles of a single mouse, the secretions from 6 - 1 0 males were pooled for fractionation on Sephadex G-100 columns. The pooled samples were treated in the same way as those from single males and the supernatant was applied to the columns. Inhibition of complement-mediated cytotoxicity Two test systems were used to measure the inhibitory activity: the complement-
111
mediated cytotoxicity test on sperm (Yanagisawa et al., 1974; Artzt and Bennet, 1977) and a standard H-2 typing test on lymphocytes. In both systems a volume of solution containing seminal vesicle proteins equal to the volume of serum used was added to the test suspension before the addition of complement. This increased the final volume of the test suspension by 33% and diluted both the serum and complement. The density of sperm in the initial suspension was adjusted so that the final density of cells in the test suspension was about twice that originally described (about 4 X lo6 sperm/ml incubated with serum and complement). These changes in cell density were made so that the test would be easier to count. For lymphocyte tests the final cell number was adjusted to 2 X 106/ml. The changes in serum and complement concentration were taken into account by running cell controls, complement controls and serum with complement controls with every test. These three controls were made by adding an appropriate volume of medium instead of the solution of seminal vesicle proteins. Rabbit complement for sperm tests was prepared as described by Artzt and Bennett (1977) and unabsorbed nontoxic rabbit serum was used as a complement source for lymphocyte tests. Antisperm sera were prepared by injecting sperm into BTBRTF/Nev females as described by Yanagisawa et al. (1974). The serum used in this study was a highly cytotoxic non-specific antisperm serum. A standard H-2 typing serum (A anti-EL4, a gift of Dr. E.A. Boyse) was used for lymphocyte tests. Anti-sperm sera was generally used at dilutions of l/8 and l/16 so that inhibition of the activity at two levels of antisera could be assessed. The average of the two values was used to calculated specific inhibition of the tested sample. Doubling dilutions of the supernatant solution prepared from the seminal vesicle were tested for inhibitory activity with anti-sperm sera. The anti-EL4 serum was tested at doubling dilutions of l/100 to l/800 so that inhibition at the maximum activity of the serum could be determined. The inhibitory activity of the seminal vesicle fluid was calculated as a percentage of the activity of the normal serum by the following formula: Inhibition = cytotoxic index of normal serum with complement - cytotoxic index of serum with seminal vesicle fluid cytotoxic index of normal serum with complement
Heat inactivation Samples of the solution of soluble seminal vesicle proteins were incubated at temperatures of 25,37,56 and 85’C for 30 min before being used in the cytotoxic tests. Sperm absorptions Samples of the solution of soluble seminal vesicle proteins were absorbed with sperm by the quantitative absorption method described by Artzt and Bennett (1977). The numbers of sperm used ranged from 15 X lo6 to 80 X lo6 for each 0.05 ml of solution. The absorbed solutions were then tested in the cytotoxic test systems to determine whether sperm removed the active components. Fractionation Sephadex G-100 (Pharmacia) was suspended in PBS, washed three times and poured into columns 2 cm in diameter. The height of the columns, flow rate of buffer through
112 the colunm, and amount of protein applied were all varied in attempts to achieve the best separation possible. The columns were run at 4°C. The optical density at 280 nm of each 2 ml fraction was determined and a protein prof'fle was graphed for each column. Fractions for columns with good separation between peaks were pooled, concentrated by pressure dialysis on an Amicon Ftitration system and tested for inhibitory activity. RESULTS
Inhibitory activity in solutions of soluble seminal vesicle proteins Only 10-17 mg of soluble protein was recovered when the seminal vesicle secretions of a single male were mixed with PBS. Since fluid recovery was sometimes difficult and incomplete when precipitation of the insoluble material occurred rapidly some of the soluble proteins were undoubtedly lost. Although some attempts were made to reduce precipitation losses none were successful and the amount of protein lost was not determined. Because of this loss of protein, the measured inhibitory activity was probably lower than that present in the fluids in vivo. Solutions prepared from seminal vesicles of individual males inhibited serum cytotoxicity by 40-80%. These inhibitory values were approximately the same with both the anti-sperm serum and anti-H-2 serum. The seminal vesicle preparations themselves were not cytotoxic to either sperm of lymphocytes. Animals of an outbred stock (ICR) had similar levels of activity as inbred mice (BTBRTF/Nev). The inhibitory activity of solutions prepared from the glands of a single male were tested following serial dilutions from 1/2 to 1/16. Activity decreased very rapidly with dilution and no activity was detectable in dilutions greater than 1/8. Tests of the more concentrated solutions, prepared for fractionation on Sephadex G-100, showed inhibition of 90-100% of serum cytotoxicity.
Heat inactivation Tests of heated aliquots of seminal vesicle protein showed that none of the temperatures used caused any decrease in inhibitory activity. Thus, this activity was heat stable to 85°C.
Sperm absorptions Absorptions were done on aliquots of soluble seminal vesicle protein solutions from 6 different males. The solution from each male was tested unabsorbed and absorbed with different numbers of sperm. The numbers of sperm used for absorption ranged from 15 × 106/0.05 ml of solution to 60 X 106/0.05 ml of solution. Inhibition (% of normal serum) for unabsorbed solutions and samples absorbed with various numbers of sperm were averaged, and appear in Table 1. These results clearly show that absorption with sperm did not reduce inhibitory activity. In fact, inhibitory activity seemed to be increased following absorption. It was possible that some other component of the mixture of soluble proteins that restricts the inhibition was removed by the absorption. But further studies of this interaction were not conducted. In any case, the inhibitor was not removed and therefore does not bind to sperm.
113 TABLE 1 Inhibition of complement-mediated antisera cytotoxicity by solutions of soluble seminal vesicle proteins after absorption with spermatozoa Inhibition (% decrease in normal serum activity) (m -+ S.E.) Control: unabsorbed solution o f soluble seminal vesicle proteins Solution absorbed with 1 5 - 1 0 X 106 spermatozoa 2 5 - 3 5 × 106 spermatozoa 4 0 - 5 0 × 106 spermatozoa 5 5 - 6 0 X 106 spermatozoa
6
45.8 -+ 16
4 5 4 5
7915 71.6 96.0 82.7
-+ 6 -+ 14 -+ 9 -+ 18
Fractionation
Initial fractionations indicated that flow rates of between 6 and 16 ml/h did not affect the protein pattern. Both 1- and 2-m columns were initially tested and longer columns gave better separations. Samples containing 10-130 mg of protein were separated on 2 m × 2 cm Sephadex G-100 columns. Best separations occurred with samples of 5 0 - 9 0 mg. A typical protein prof'lle is shown in Fig. 1. There are 5 protein peaks, each numbered for identification purposes. Peak 1 was in the void volume and was only detectable when more than 40 mg protein was applied to the column. When samples containing over 30 mg of protein were fractionated, peaks 3, 4 and 5 were not well separated from each other. When the tubes in each peak were pooled, concentrated, and tested for inhibition of 0.8 E c 0.6 O G) GI
I I I
g o.4 g ~ 0.2
t 0,4
o.2~"..5
h o
o
I
4to
66
~o
~6o
Fraction
~o
Peak 1 2 3 4 5 Fig. 1. Typical protein profile from fractionation o f seminal vesicle proteins on Sephadex G-100 showing 5 main protein peaks and'3 peaks o f inhibitory activity..
114 cytotoxic activity, peak 5 showed no inhibition. Peaks 3 and 4 were most active but 1 and 2 also showed some inhibition (summary of the results of 9 columns). Since the inhibitory activity could not be localized in one major protein peak, the fractions were assayed for activity by pooling every 5 tubes, starting with tube 40, concentrating the protein and testing it to produce a profile of inhibitory activity for the whole column effluent. A typical activity pattern is presented in Fig. 1. It can clearly be seen that there are three peaks of inihibitory activity. The major peak of activity (A in Fig. 1) coincides most closely with protein peak 1, which was in the void volume of the column and therefore had a molecular weight of over 150 000. Some of this activity may also by detected in protein peak 2, probably as a result of incomplete separation. The two other activity peaks coincide with protein peaks 3 and 4. DISCUSSION These data indicate that the secretions of the seminal vesicles of the house mouse contain a factor that can protect sperm from complement-mediated antibody cytotoxicity. The major peak of this immunosuppressive activity was found in a soluble fraction which appeared in the void volume after gel filtration on Sephadex G-100. This indicated that the material has a molecular weight of over 150000. Two other peaks of activity with lower molecular weight were also identified. There are several possible explanations for the appearance of three peaks of activity. The active material of high molecular weight may either become degraded and release the lower molecular weight fragments, or conversely the high molecular weight component could be an aggregate of the lower molecular weight fractions. Alternatively, there may be three unrelated proteins with immunosuppressive activity. Fractionations of this material following treatment with urea or high salt concentrations to prevent aggregation have not yet been performed. While the association of activity with fractions of high molecular weight absorbing light at 280 nm implies that the inhibitory materials are proteins, they could also be glycoproteins. Studies by a number of investigators indicate that human seminal plasma contains several distinct immunosuppressive factors. The inhibitor of cell-mediated immune responses isolated from human seminal plasma has been shown to be heat stable, rich in carbohydrates and with a high molecular weight (Marcus et al., 1979). Human seminal plasma also contains a heat labile factor with a molecular weight of between 20 000 and 60 000 which reduces the activity of complement components C 1 and C3 by more than 50% (Petersen et al., 1980). It is not clear whether these varied components interact in any way. Since the seminal fractions from the house mouse inhibit complement-mediated cytotoxicity of both anti-sperm and anti-H-2 sera the activity was probably directed against some general feature of antibodies or some component of the complement system. However, the specific component has not yet been identified. It was clear from the absorption studies with sperm that the active material did not have a specific binding site on sperm that was necessary for inhibition to occur. Further tests will be needed to determine why the absorptions with sperm increased the inhibitory activity. These data indicate that following natural matings sperm are protected against antibody attack. This would explain why immunizations against sperm have not totally pre-
115 vented pregnancy, even in female mice that developed a high titer of anti-sperm antibodies (McLaren, 1966; Bell and McLaren, 1970). Even though the inhibitory material does not bind to sperm it could still play a protective role during the passage of sperm through the uterus since insemination is intra-uterine in the house mouse. The absence of thins factor from mixtures of epididymal sperm used for artificial insemination may account for the relatively low success rate of artificial insemination as compared to natural matings in the mouse, as well as the decrease in fertility following surgical removal of the seminal vesicle reported by Pang et al. (1979). Several studies have indicated that mouse semen inhibits cell-mediated immune responses. Mouse sperm inhibit the mixed leukocyte reaction (Erickson and Stites, 1975) and mouse seminal plasma can prevent lymphocyte activation (Stites and Erickson, 1975). Further studies are needed to determine the relationship between the seminal factors that inhibit cell-mediated immunological responses and those which inhibit complement-mediated antisera cytotoxicity. In addition, a comparative study of the immunosuppressive fractions from the semen of a number of species would be most informative. ACKNOWLEDGMENT B.P. was supported by NIH Fellowship No. 5 F32 HD05033-02. REFERENCES Artzt, K. and Bennet, D. (1977) Serological analysis of sperm of antigenically cross reacting T/thaplotypes and their recombinants. Immunogenetics 5, 97-107. Beer, A.E. and Billingham, R.E. (1976) The Immunobiology of Mammalian Reproduction. PrenticeHall, Englewood Cliffs, N.J. Bell, E.B. and McLaren, A. (1970) Reduction of fertility in female mice isoimmunized with a subcellular sperm fraction. J. Reprod. Fertil. 22, 345-356. Erickson, R.P. and Stites, D.P. (1975) Effects of murine spermatozoa on the mixed leukocyte reaction of mice. Transplantation 20, 263-265. Geiger, B., Frensdorff, A. and Kraicer, P.F. (1974) The soluble proteins of rat seminal vesicle fluid; some physico-chemical and immunological properties. Immunology 27, 729-738. Leckie, P.A., Watson, J.G. and Chaykin, S. (1973) An improved method for the artificial insemination of the mouse (Mus musculus). Biol Reprod. 9, 420-425. Lord, E.M., Sensabaugh, G.F. and Stites, D.P. (1977) Immunosuppressive activity of human seminal plasma. I. Inhibition of in vitro lymphocyte activation. J. Immunol. 118, 1704-1711. Marcus, Z.H., Freisheim, J., Herman, J.H., and Hess, E.V. (1977) Inhibition of mitogen induced blast transformation by male genital components. In Immunological Influence on Human Fertility (Boettcher, B., ed.), pp. 333-339. Academic Press, New York. Marcus, Z.H., Hess, E.V., Herman, J.H., Troiano, P. and Freisheim, J. (1979) In vitro studies in reproductive immunology. 2. Demonstrations of the inhibitory effect of male genital tract constituents on PHA-stimulated mitogenesis and E-rosette formation of human lymphocytes. J. Reprod. Immunol. 1, 97-107. McLaren, A. (1966) Studies on the isoimmunization of mice with spermatozoa. Fertil. Steril. 17, 492-499. Pang, S.F., Chow, P.H. and Wong, T.M. (1979) The role of the seminal vesicles, coagulating glands and prostate glands on the fertility and fecundity of mice. J. Reprod. Fertil. 56, 129-132. Petersen, B.H., Lammel, C.J., Stites, D.P. and Brooks, G.F. (1980) Human seminal plasma inhibition of complement. J. Lab. Clin. Med. 96,582-591.
116 Pitout, M.J. and Jordan, J.H. (1976) Partial purification of an antimitogenic factor from human semen. Int. J. Biochem. 7, 149-151. Prakash, C., Coutinho,'A. and M611er, G. (1976) Inhibition of in vitro immune responses by a fraction from seminal plasma. Scand. J. Immunol. 5, 72-85. Rodger, J.C. (1975) Seminal plasma, an unnecessary evil? Theriogenology 3,237-247. Smith, W.G. and Beer, A.E. (1976) Current concepts of reproductive immunobiology. J. Perinat. Med. 4, 59-71. Stites, D.P., and Erickson, R.P. (!975) Suppressive effect of seminal plasma on lymphocyte activation. Nature (London) 253,727-729. Yanagisawa, K., Bennett, D., Boyse, E.A., Dunn, L.C. and Dimeo, A. (1974) Serological identification of sperm antigens specified by lethal t-alleles in the mouse. Immunogenetics 1, 57-67.