Journal oJ Reprodu~ ttve Immunology, 20 ( 1991 ) 2 7 4 1
27
Elsevier Scientific Pubhshers Ireland Ltd.
JRI 00714
Sperm immobilizing and fertilization-blocking monoclonal antibody 2C6 to human seminal plasma antigen and characterization of the antigen epitope corresponding to the monoclonal antibody Kinu Kameda, Yoshitsugu Takada, Akiko Hasegawa, Y o s h i y u k i Tsuji, K o j i K o y a m a a n d S h i n z o I s o j i m a Department of Obstetrtcs and Gynecology, Hyogo Medtcal College, 1-1 Mukogawa-cho, N1shmomo'a 663 (Japan) (Accepted for pubhcatlon 28 November 1990)
Summary A monoclonal antibody (Mab 2C6) with strong sperm immobdizing and agglutinating activities was generated by cell fusion between spleen cells from a mouse immunized with human seminal plasma (HSP) and mouse myeloma cells. It also showed a strong inhibitory effect on human sperm-egg interaction. The corresponding antigen was present on the whole surface of ejaculated spermatozoa. In male genital organs, lmmunostaining with Mab 2C6 was observed in epididymis and seminal vesicle but not an testis. By Western blotting, lmmunostainlng with Mab 2C6 was detected around the 15--25 kDa region under both reducing and non-reducing conditions. The antigen corresponding to Mab 2C6 was susceptible to treatment with perlodate or trifluoromethanesulfonlc acid. The antigenac activities were shghtly increased by treatment with neuraminadase but reduced by further treatment with glycosldases. Enzymatic digestions with pronase and papain also reduced the antigenic activities. The antigen molecules exhibited a strong binding affimty to RCA lectln. These results indacated that Mab 2C6 recognized one of the components which might be secreted from epldldymas or seminal vesicle and bmd to ejaculated Correspondence to Shmzo Isojtma, M D , Department of Obstetrics and Gynecology, Hyogo Medical College, 1-1 Mukogawa-cho, Nlshmomlya 663, Japan 0165-0378/91/$03.50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
28
spermatozoa as a sperm coating antigen. The corresponding antigen seems to be a glycoprotem and its carbohydrate moiety has an important role m the conformation of the antigen epitope.
Key words: monoclonal antibody; human sperm immobilization; fertilizatlon blocking; carbohydrate.
Introduction
The correlation between sperm immobilizing antibodies (SI-Abs) and sterility in humans has been well established (Isojima et al., 1968, 1972; Ansbacher et al., 1973; Jones et al., 1973). For clarifying the mechanism of immunological sterility and possible application of SI-Abs to the development of a contraceptive vaccine, analyses of the sperm antigens corresponding to SI-Abs in patients' sera were essential. Previous studies revealed that SI-Abs in patients' sera could be absorbed with human seminal plasma (HSP) as well as ejaculated human spermatozoa in most cases (Isojima et al., 1972). Antigen analysis of HSP using the conventional antiserum to HSP on agar immunoelectrophoresis was performed and at least four HSP specific components (Nos. 4, 6, 9 and 10) and two other antigenic components (Nos. 3 and 7) cross-reactive to human milk (HM) were recognized (Isojima et al., 1974). In order to identify the antigen epitope relevant to SI-Ab, a rat-mouse heterohybridoma (1C4) secreting SI-Mab was successfully established in our laboratory (Shigeta et al., 1980). The antigenic component corresponding to the Mab IC4 was purified 196 times by using immunoaffimty chromatography on bound Mab (Isojlma et al., 1982). However, this antigen was found to be one of the cross-reactive antigens with HM, such as HSP No. 7 antigen Ferrisplan (Koyama et al., 1983), and may not be suitable as a candidate for a contraceptive vaccine. In the present study, we tried to establish hybridomas producing SI-Mab to HSP specific component and succeeded in obtaining hybridoma 2C6. The characteristic properties of the antigen corresponding to Mab 2C6 are also reported. Materials and Methods
Production of monoclonal antibody Spleen cells from a Balb/c mouse immunized with HSP were fused with mouse myeloma cells (P3U1) as described previously (Shigeta et al., 1980). HSP from azoospermic semen (2.5 mg/mouse) was injected with complete
29
and then incomplete Freund's adjuvant at an interval of 2 weeks. For antibody screening, a sperm immobilization test (SIT) was used as described below. To select the clones producing SI-Mabs to HSP specific antigen, culture supernatants were mixed with HSP or HM (1 mg/ml). After incubation at 4°C overnight, the supernatants were examined for the remaining SI activity. Among the hybridomas producing SI-Mabs which could be absorbed with HSP but not with HM, one clone producing the highest antibody titer was selected for further cloning and a hybridoma (2C6) was established. The hybridoma 2C6 or myeloma cells (P3U1) were innoculated intraperitoneally into Balb/c mice pretreated with pristane to obtain ascites, as described previously (Koyama et al., 1985). A 3,-globulin fraction of the ascites was prepared by precipitation with 50% saturated ammonium sulfate.
Sperm immobilization and agglutination tests Sperm immobilization and agglutination tests were carried out by a slightly modified method of the sperm immobilization test (SIT) (Isojima and Koyama, 1979) and the tray agglutination test (Friberg, 1974), as described previously (Isojima et al., 1987). The number of motile spermatozoa were counted in test specimens (T%) and control P3U1 culture supernatant (C%) and the relative sperm motility (T/C%) was calculated by the formula of T/C × 100. The sperm immobilization titer (SIs0 unit) was determined as described earlier (Koyama et al., 1988). Antibody absorption test The specificity of antibody in the culture supernatant was examined by an antibody absorption test (Isojima et al., 1987) with various human materials. Ejaculated spermatozoa were washed in saline three times. HSP, HM and normal human serum (NHS) were prepared by saturation with ammonium sulfate followed by dialysis with water and lyophilization. Tissue extracts were prepared as follows: human hver (HL) and kidney (HK) tissues were homogenized in saline by a Waring blender (Omega electric) for 5 rain and sonicated (Kubota, 100 W) for 5 min. After centrifugatlon at 9500 x g for 30 min, the supernatants were dialysed m distilled water and lyophilized. For a study of interspecies cross-reaction, Mab 2C6 was absorbed with lyophilized specimens of porcine and bovine seminal plasmas. Sperm-ELISA inhibition assay Washed human spermatozoa were coated on plates (Falcon) (Isojima et al., 1987). After blocking with 1% bovine serum albumin (BSA) in PBS, 50 ~1 of Mab 2C6 and samples were added to each well and incubated at 4°C overnight. After washing, wells were incubated with peroxidase conjugated goat anti-mouse IgM (Cappel) diluted to 1:500 for 1 h. The reaction was
30
developed by 0.02°/,, o-phenylenediamine/O.05% H202 in 0.1 M citratephosphate buffer (pH 5.0) and stopped by adding 10% H2SO4. Percent binding inhibition was calculated by the formula of (ODin - ODt/ODm - ODe) X 100, in which O D m w a s the maximum absorbance in the wells with no inhibitory antigens, O O t w a s the absorbance in the wells with test samples and ODe was the absorbance in the wells with P3U1 culture supernatant instead of Mab.
Immunohistochemical and immunofluorescent studies Formalin-fixed and paraffin-embedded tissue sections of male genital and somatic organs were stained using Vectastain ABC kit for mouse IgM (Vecter Labs). Briefly, tissue sections were incubated with 0.3% H202 in methanol for 30 min to block endogeneous peroxidase activity and then blocked with 2°/,, goat serum in PBS. The sections were reacted with culture supernatant of Mab, bxotinylated goat anti-mouse IgM and then ABC reagent according to the manufacturer's instructions. The slides were developed with 0.5% diaminobenzMme/0.02'/,, H~O~ in 50 mM Trls--HC1 buffer (pH 7.2) and counter-stained with Mayer's hematoxyhn. Ejaculated human spermatozoa were stained by indirect immunofluorescence. The smears of washed sperm were made on the shde, a~r dried and then fixed in methanol for 5 mln. The slide was incubated with Mab for 1 h and then with a 1:50 dilution of FITC conjugated anti-mouse IgM (Cappel) for 30 min. The slides were washed three times with PBS at each step. Fluorescence was observed under an epifluorescent microscope• •
'
0
(
-
Gel filtration and lectm affinity chromatograph)' HSP (10 mg/ml; 1.5 ml) was applied to Sephacryl S-300 column (1.5 × 90 cm) and eluted in PBS at 10 ml/h. In another experiment, HSP was treated with trypsin (1 mg/ml) (Type I, Sigma) at 37°C for 1 h before application and eluted in the same way. Each fraction (2 ml/tube) was assayed for absorbance at 280 nm and the antigen activity by Sperm-ELISA inhibition assay The void fraction (1.5 mg) obtained from Sephacryl S-300 chromatography of trypsin-treated HSP, which was designated as partially purified HSP (p-HSP), was further fractionated wxth RCA 120-Agarose column (bed volume 2 ml; Biochemicals Co. Ltd.). The sample was eluted with 0.1 M phosphate buffer (pH 7.2)/0.15 M NaC1. After complete elution of RCA 120 unbound fraction, 0.1 M lactose in the above buffer was used to elute R C A 120 bound fraction.
SDS-PAGE and western blottmg The samples were applied to 12% polyacrylamide gels (14 × 14 × 0.1 cm) according to Laemmli (1970) and transferred to nitrocellulose membrane by
3r
the method of Towbin et al. (1979). The membrane was blocked in 50 mM Tris--HC1/200 mM NaCI (pH 7.4) (TBS) containing 3% BSA and 5% normal horse serum and incubated with Mab at 4°C overnight. After washing, the membrane was reacted with a 1:500 dilution of peroxidase conjugated antimouse IgM (Cappel). The reaction was developed with 0.05% 4-chloro-l-naphthol in TBS containing 0.02% H~_O2. Chemical and enzymatic treatments o f human seminal plasma For heat treatment, p-HSP (0.5 mg/ml) in PBS was incubated at 56°C, 70°C, 80°C and 100°C for 30 min. Periodate treatment was performed as follows: p-HSP (0.5 mg/ml) was incubated in 50 mM sodium acetate buffer (pH 4.5) containing 1--10 mM of sodium metaperiodate (NaIO4) at 4°C for 1--30 h and then sucrose was added to a final concentration of 5% to stop the reaction and the treated samples were dialysed against PBS. The treatment with trifluoromethanesulfonic acid (TFMS) was performed according to Edge et al. (1981). Briefly, lyophilized p-HSP (10 mg) was treated with 1 ml o f a TFMS-anisol mixture (2:1, v/v) at 4°C for 3 h. After addition of 50% aqueous pyridine and extraction with diethyl ether 4 times, the aqueous layer was dialysed against 2 mM pyridine-acetate buffer (pH 5.5) and distilled water and lyophilized. For treatments with pronase (Kaken), trypsin and o~-chymotrypsin (Sigma), p-HSP (0.5 mg/ml) was incubated with 10 or 50 t~g/ml of each enzyme in 0.1 M Tris--HC1 buffer (pH 7.8)/10 mM CaC12. The treatments with papain and pepsin (Sigma) were performed in 0.1 M phosphate buffer/1 mM EDTA (pH 7.2) and 0.1 M glycine-HC1 buffer/0.25 M NaC1 (pH 2.5), respectively. After incubation, BSA was added to the reaction mixtures to a concentration of 10 mg/ml and the remaining antigen activity was assayed by ELISA. For treatments with glycosidases, p-HSP (0.5 mg/ml) was incubated with 3-galactosidase from C. lampas (1.25 U/nl) or endo-3-N-acetylglucosaminidase from T. cornutus (0.25 U/ml) (Biochemicals Co. Ltd.) in 0.1 M citratephosphate buffer/0.5 M NaC1 (pH 4.0) at 37°C for 1--30 h. The treatment with neuraminidase from Clostridium (Boehringer) (0.18 U/ml) was performed in 0.1 M sodium acetate buffer (pH 4.5) at 37°C for 16 h. After neuraminidase treatment, the samples were further treated with 3galactosidase or mixed glycosidases from T. cornutus (Biochemicals Co. Ltd.) at a final concentration of 0.1 U/ml and 1 mg/ml, respectively, for 5--25 h. After incubation, the treated samples were boiled for 3 rain to inactivate enzymatic activity and assayed for the remaining antigen activity by ELISA. In vitro fertilization Human oocytes surrounded with cumulus cells were aspirated from large
32
growing follicles of ovaries which were obtained from the surgical specimens with agreement of the patients, and incubated in Earle's medium containing 7.5% human serum for 4 0 ~ 4 8 h at 37°C in a 5% CO2 incubator for in vitro maturation as described previously (Koyama et al., 1985). Ejaculated human spermatozoa from a healthy donor were washed three times and preincubated in the same medium for 3 h. The sperm suspension and the oocytes were transferred to a droplet containing -y-globulin fractions from 2C6-hybridoma ascites or P3Ul-ascites and incubated for 15 h. The number of spermatozoa bound to or penetrating into the zonae pellucidae of the oocytes was counted under a phase-contrast microscope. Zona-free hamster eggs were prepared by treating superovulated hamster eggs with 0.1% hyaluronidase and 0.1% trypsin according to the method reported by Yanagimachi et al. (1976). They were added to a droplet of preincubated human sperm suspension (1 × 106/ml) in Biggers, Whitten and Whittingham medium (BWW) containing 0.3% BSA and 7-globulin from 2C6-hybridoma or P3U1 ascites. The mixtures were incubated for 18 h and the number of the eggs penetrated by spermatozoa was counted after staining with 0.25% acetolacmoid solution. Results
Fused cells were seeded in 457 wells. Cell growth was observed in all wells seeded and SI activity was detected in the culture supernatant of 23 wells on the 10th day after cell fusion. Seven clones were found to produce SI-Mab which could be absorbed with HSP but not with HM. One of the seven hybridomas with the highest antibody titer was selected for further cloning and established as a stable hybridoma clone (2C6). Culture supernatant of hybridoma 2C6 possessed strong SI (SIs0 titer: 200 units) and SA (1:40 dilutions) activities. The immunoglobulin class of Mab 2C6 was found to be IgM (K) by immunodiffusion test. Results of antibody absorption experiments are summarized in Table 1. Although the absorption with ejaculated human spermatozoa was not complete under the experimental conditions, HSP from azoospermic semen could completely remove the SI and SA activities in the culture supernatant of hybridoma 2C6. HM, NHS and tissue extracts of HK and HL did not remove any antibody activity by absorption. Figure 1 shows the results of the quantitative absorption test of Mab 2C6 with various amounts of human spermatozoa, HSP from normal and vasectomized men and also with HM. The decrease of antibody activity by absorption with HSP from vasectomized men was less than that of normal men. A similar absorption experiment was performed using seminal plasma from bull and boar. The SI activity of Mab 2C6 was completely absorbed out with bull seminal plasma but not with boar seminal plasma at a concentration of 9 mg dry weight/ml.
33 TABLE 1 Absorption of SI-Mab 2C6 with various human materials Each 1 ml of culture supernatant of hybridoma 2C6 was mixed with various human materials and incubated at 4°C overnight. After centrifugation (9500 × g, 30 min) the supernatants were subjected to the sperm immobilization test and the sperm motility was counted; C(+), in the presence of guinea pig serum as a source of complement; C(-), in the presence of guinea pig serum inactivated at 56°C for 30 min; HSP, human seminal plasma from azoospermic semen; HM, human milk; NHS, normal human serum; HK, human kidney; HL, human liver. Materials used for absorption
Concentration of absorbent
None Spermatozoa HSP HM NHS Extract of HK Extract of HL
Sperm motility after absorption
-4 × 107/ml 9 mg/ml a 9 mg/ml a 9 mg/ml a 50 mg/ml a 50 mg/ml a
C(+)
c(-)
0% 42% 73% 0% 0% 0% 0%
74% 76% 73% 72% 74% 70% 74%
amg of dry weight of lyophilized material/ml.
100
o E E £z. 50 Q~
> "-G
#J
E I
I
I
Sperm
"
105
186
187 cells/m~
HSP
"
8.1
1.8
10
mg/m~
Fig. 1. Quantitative absorption of SI-Mab 2C6. Culture supernatants of hybridoma 2C6 were diluted 1:40 and incubated with various amounts of human spermatozoa ( • - - • ), HSP from normal ( • - - • ) and vasectomized ( o - - o ) men and HM ( [ ] - - [] ) at 4°C overnight. After centrifugation, the supernatants were assayed for SI activity by the sperm immobilization test and relative sperm motility was calculated as described in Materials and Methods.
34
The effect of Mab 2C6 on sperm-egg interaction was tested by human sperm binding and penetration of human oocytes and zona-free hamster eggs. As shown in Table 2, 3,-globulin (1 mg/ml) from mouse myeloma P3U1 ascites had no adverse effects on sperm binding nor penetration to both eggs, whereas 3,-globulin (0.12 or 0.04/~g/ml) of hybridoma 2C6 ascites showed a strong inhibitory effect on the sperm-egg interaction. During these fertilization assays, there were no significant sperm agglutinations observed in the presence of "y-globulin from 2C6 ascites at the concentrations used for the assays. When ejaculated spermatozoa were subjected to the indirect immunofluorescent staining, most cells were strongly stained all over the surface. In the immunohistological studies of formalin-fixed and paraffin-embedded tissue sections of male genital organs (Fig. 2), the epithelial cells of the epididymis was stained. No staining was observed in the sections of testis and prostate. Weak and discontinuous staining was observed in the epithelial cells of seminal vesicles. Among the sections of human somatic organs tested, proximal and distal tubules of kidney were weakly stained, but mammary gland and liver did not show staining. For characterization of the antigen corresponding to the Mab, either HSP or trypsin-treated HSP was fractionated by gel filtration through Sephacryl S-300 column and each fraction was assayed for antigenic activity by sperm-
TABLE 2 Blocking effect of SI-Mab 2C6 on human sperm binding and penetration of human oocytes a n d zona-free hamster eggs Human oocytes and zona-free hamster eggs were transferred to droplets of preincubated human sperm suspension in medium containing 3,-globulin from control ascites of mouse myeloma (P3UI) or ascites of hybridoma 2C6 and incubated for 15--18 h at 37°C in 5% CO 2 incubator. After incubation, the number of spermatozoa bound or penetrating into zonae pellucidae or the number of eggs penetrated by spermatozoa were counted as described in Materials and Methods. The concentrations of 7-globulin used were 1 mg/ml for control and 0.12 or 0.04/zg/ml for Mab 2C6. N.T., not tested. Experiments
Human oocytes Control 2C6 (0.12 t~g/ml) Zona-free hamster eggs Control 2C6 (0.12 /zg/ml) 2C6 (0.04 ~g/ml)
No. of eggs used
3 8
15 16 17
No. of sperm bound to zona
No. of sperm penetrated into zona
No. of eggs penetrated by sperm
56.0 4- 16.5 2.8 ± 4.1
4.7 ± 2.9 0.0 ± 0.0
3 0
N.T. N.T. N.T.
N.T. N.T. N.T.
15 2 6
Fig. 2. lmmunohistochemical staining of male genital organs in human with Mab 2C6. A, testis. B, epididymis, x 150. Arrows indicate positive staining with Mab 2C6.
36
ELISA inhibition assay. As shown in Fig. 3 for the latter sample, most antigenic activity was eluted in the void fraction. It indicated that the molecular weight (MW) of the corresponding antigen was greater than 670 000. However, when the same sample was applied to SDS-PAGE and transblotted to nitrocellulose membranes, the immunostaining by Mab 2C6 was detected as a broad band around 15--25 kDa under both non-reducing and reducing conditions (Fig. 4). The void fraction of HSP from Sephacryl S-300 chromatography, which was designated as p-HSP in this paper, was subjected to heat treatment (56--100°C) for 30 rain and its reactivity with Mab 2C6 was compared with that of non-treated material by competitive binding inhibition assay of Mab 2C6 to human spermatozoa coated on ELISA plates: The reactivity was completely retained even after boiling the sample for 30 min. When the same material was treated with sodium metaperiodate, the antigen activity was almost completely destroyed (Fig. 5). After deglycosylation by treatment with TFMS, p-HSP completely lost reactivity with Mab 2C6 as tested by either ELISA or antibody absorption test (Fig. 6).
1.0
t
./
H*
L/ Inhibition assay by ELI SA
x
0.8
27
Elsevier Scientific Pubhshers Ireland Ltd.
JRI 00714
Sperm immobilizing and fertilization-blocking monoclonal antibody 2C6 to human seminal plasma antigen and characterization of the antigen epitope corresponding to the monoclonal antibody Kinu Kameda, Yoshitsugu Takada, Akiko Hasegawa, Y o s h i y u k i Tsuji, K o j i K o y a m a a n d S h i n z o I s o j i m a Department of Obstetrtcs and Gynecology, Hyogo Medtcal College, 1-1 Mukogawa-cho, N1shmomo'a 663 (Japan) (Accepted for pubhcatlon 28 November 1990)
Summary A monoclonal antibody (Mab 2C6) with strong sperm immobdizing and agglutinating activities was generated by cell fusion between spleen cells from a mouse immunized with human seminal plasma (HSP) and mouse myeloma cells. It also showed a strong inhibitory effect on human sperm-egg interaction. The corresponding antigen was present on the whole surface of ejaculated spermatozoa. In male genital organs, lmmunostaining with Mab 2C6 was observed in epididymis and seminal vesicle but not an testis. By Western blotting, lmmunostainlng with Mab 2C6 was detected around the 15--25 kDa region under both reducing and non-reducing conditions. The antigen corresponding to Mab 2C6 was susceptible to treatment with perlodate or trifluoromethanesulfonlc acid. The antigenac activities were shghtly increased by treatment with neuraminadase but reduced by further treatment with glycosldases. Enzymatic digestions with pronase and papain also reduced the antigenic activities. The antigen molecules exhibited a strong binding affimty to RCA lectln. These results indacated that Mab 2C6 recognized one of the components which might be secreted from epldldymas or seminal vesicle and bmd to ejaculated Correspondence to Shmzo Isojtma, M D , Department of Obstetrics and Gynecology, Hyogo Medical College, 1-1 Mukogawa-cho, Nlshmomlya 663, Japan 0165-0378/91/$03.50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
28
spermatozoa as a sperm coating antigen. The corresponding antigen seems to be a glycoprotem and its carbohydrate moiety has an important role m the conformation of the antigen epitope.
Key words: monoclonal antibody; human sperm immobilization; fertilizatlon blocking; carbohydrate.
Introduction
The correlation between sperm immobilizing antibodies (SI-Abs) and sterility in humans has been well established (Isojima et al., 1968, 1972; Ansbacher et al., 1973; Jones et al., 1973). For clarifying the mechanism of immunological sterility and possible application of SI-Abs to the development of a contraceptive vaccine, analyses of the sperm antigens corresponding to SI-Abs in patients' sera were essential. Previous studies revealed that SI-Abs in patients' sera could be absorbed with human seminal plasma (HSP) as well as ejaculated human spermatozoa in most cases (Isojima et al., 1972). Antigen analysis of HSP using the conventional antiserum to HSP on agar immunoelectrophoresis was performed and at least four HSP specific components (Nos. 4, 6, 9 and 10) and two other antigenic components (Nos. 3 and 7) cross-reactive to human milk (HM) were recognized (Isojima et al., 1974). In order to identify the antigen epitope relevant to SI-Ab, a rat-mouse heterohybridoma (1C4) secreting SI-Mab was successfully established in our laboratory (Shigeta et al., 1980). The antigenic component corresponding to the Mab IC4 was purified 196 times by using immunoaffimty chromatography on bound Mab (Isojlma et al., 1982). However, this antigen was found to be one of the cross-reactive antigens with HM, such as HSP No. 7 antigen Ferrisplan (Koyama et al., 1983), and may not be suitable as a candidate for a contraceptive vaccine. In the present study, we tried to establish hybridomas producing SI-Mab to HSP specific component and succeeded in obtaining hybridoma 2C6. The characteristic properties of the antigen corresponding to Mab 2C6 are also reported. Materials and Methods
Production of monoclonal antibody Spleen cells from a Balb/c mouse immunized with HSP were fused with mouse myeloma cells (P3U1) as described previously (Shigeta et al., 1980). HSP from azoospermic semen (2.5 mg/mouse) was injected with complete
29
and then incomplete Freund's adjuvant at an interval of 2 weeks. For antibody screening, a sperm immobilization test (SIT) was used as described below. To select the clones producing SI-Mabs to HSP specific antigen, culture supernatants were mixed with HSP or HM (1 mg/ml). After incubation at 4°C overnight, the supernatants were examined for the remaining SI activity. Among the hybridomas producing SI-Mabs which could be absorbed with HSP but not with HM, one clone producing the highest antibody titer was selected for further cloning and a hybridoma (2C6) was established. The hybridoma 2C6 or myeloma cells (P3U1) were innoculated intraperitoneally into Balb/c mice pretreated with pristane to obtain ascites, as described previously (Koyama et al., 1985). A 3,-globulin fraction of the ascites was prepared by precipitation with 50% saturated ammonium sulfate.
Sperm immobilization and agglutination tests Sperm immobilization and agglutination tests were carried out by a slightly modified method of the sperm immobilization test (SIT) (Isojima and Koyama, 1979) and the tray agglutination test (Friberg, 1974), as described previously (Isojima et al., 1987). The number of motile spermatozoa were counted in test specimens (T%) and control P3U1 culture supernatant (C%) and the relative sperm motility (T/C%) was calculated by the formula of T/C × 100. The sperm immobilization titer (SIs0 unit) was determined as described earlier (Koyama et al., 1988). Antibody absorption test The specificity of antibody in the culture supernatant was examined by an antibody absorption test (Isojima et al., 1987) with various human materials. Ejaculated spermatozoa were washed in saline three times. HSP, HM and normal human serum (NHS) were prepared by saturation with ammonium sulfate followed by dialysis with water and lyophilization. Tissue extracts were prepared as follows: human hver (HL) and kidney (HK) tissues were homogenized in saline by a Waring blender (Omega electric) for 5 rain and sonicated (Kubota, 100 W) for 5 min. After centrifugatlon at 9500 x g for 30 min, the supernatants were dialysed m distilled water and lyophilized. For a study of interspecies cross-reaction, Mab 2C6 was absorbed with lyophilized specimens of porcine and bovine seminal plasmas. Sperm-ELISA inhibition assay Washed human spermatozoa were coated on plates (Falcon) (Isojima et al., 1987). After blocking with 1% bovine serum albumin (BSA) in PBS, 50 ~1 of Mab 2C6 and samples were added to each well and incubated at 4°C overnight. After washing, wells were incubated with peroxidase conjugated goat anti-mouse IgM (Cappel) diluted to 1:500 for 1 h. The reaction was
30
developed by 0.02°/,, o-phenylenediamine/O.05% H202 in 0.1 M citratephosphate buffer (pH 5.0) and stopped by adding 10% H2SO4. Percent binding inhibition was calculated by the formula of (ODin - ODt/ODm - ODe) X 100, in which O D m w a s the maximum absorbance in the wells with no inhibitory antigens, O O t w a s the absorbance in the wells with test samples and ODe was the absorbance in the wells with P3U1 culture supernatant instead of Mab.
Immunohistochemical and immunofluorescent studies Formalin-fixed and paraffin-embedded tissue sections of male genital and somatic organs were stained using Vectastain ABC kit for mouse IgM (Vecter Labs). Briefly, tissue sections were incubated with 0.3% H202 in methanol for 30 min to block endogeneous peroxidase activity and then blocked with 2°/,, goat serum in PBS. The sections were reacted with culture supernatant of Mab, bxotinylated goat anti-mouse IgM and then ABC reagent according to the manufacturer's instructions. The slides were developed with 0.5% diaminobenzMme/0.02'/,, H~O~ in 50 mM Trls--HC1 buffer (pH 7.2) and counter-stained with Mayer's hematoxyhn. Ejaculated human spermatozoa were stained by indirect immunofluorescence. The smears of washed sperm were made on the shde, a~r dried and then fixed in methanol for 5 mln. The slide was incubated with Mab for 1 h and then with a 1:50 dilution of FITC conjugated anti-mouse IgM (Cappel) for 30 min. The slides were washed three times with PBS at each step. Fluorescence was observed under an epifluorescent microscope• •
'
0
(
-
Gel filtration and lectm affinity chromatograph)' HSP (10 mg/ml; 1.5 ml) was applied to Sephacryl S-300 column (1.5 × 90 cm) and eluted in PBS at 10 ml/h. In another experiment, HSP was treated with trypsin (1 mg/ml) (Type I, Sigma) at 37°C for 1 h before application and eluted in the same way. Each fraction (2 ml/tube) was assayed for absorbance at 280 nm and the antigen activity by Sperm-ELISA inhibition assay The void fraction (1.5 mg) obtained from Sephacryl S-300 chromatography of trypsin-treated HSP, which was designated as partially purified HSP (p-HSP), was further fractionated wxth RCA 120-Agarose column (bed volume 2 ml; Biochemicals Co. Ltd.). The sample was eluted with 0.1 M phosphate buffer (pH 7.2)/0.15 M NaC1. After complete elution of RCA 120 unbound fraction, 0.1 M lactose in the above buffer was used to elute R C A 120 bound fraction.
SDS-PAGE and western blottmg The samples were applied to 12% polyacrylamide gels (14 × 14 × 0.1 cm) according to Laemmli (1970) and transferred to nitrocellulose membrane by
3r
the method of Towbin et al. (1979). The membrane was blocked in 50 mM Tris--HC1/200 mM NaCI (pH 7.4) (TBS) containing 3% BSA and 5% normal horse serum and incubated with Mab at 4°C overnight. After washing, the membrane was reacted with a 1:500 dilution of peroxidase conjugated antimouse IgM (Cappel). The reaction was developed with 0.05% 4-chloro-l-naphthol in TBS containing 0.02% H~_O2. Chemical and enzymatic treatments o f human seminal plasma For heat treatment, p-HSP (0.5 mg/ml) in PBS was incubated at 56°C, 70°C, 80°C and 100°C for 30 min. Periodate treatment was performed as follows: p-HSP (0.5 mg/ml) was incubated in 50 mM sodium acetate buffer (pH 4.5) containing 1--10 mM of sodium metaperiodate (NaIO4) at 4°C for 1--30 h and then sucrose was added to a final concentration of 5% to stop the reaction and the treated samples were dialysed against PBS. The treatment with trifluoromethanesulfonic acid (TFMS) was performed according to Edge et al. (1981). Briefly, lyophilized p-HSP (10 mg) was treated with 1 ml o f a TFMS-anisol mixture (2:1, v/v) at 4°C for 3 h. After addition of 50% aqueous pyridine and extraction with diethyl ether 4 times, the aqueous layer was dialysed against 2 mM pyridine-acetate buffer (pH 5.5) and distilled water and lyophilized. For treatments with pronase (Kaken), trypsin and o~-chymotrypsin (Sigma), p-HSP (0.5 mg/ml) was incubated with 10 or 50 t~g/ml of each enzyme in 0.1 M Tris--HC1 buffer (pH 7.8)/10 mM CaC12. The treatments with papain and pepsin (Sigma) were performed in 0.1 M phosphate buffer/1 mM EDTA (pH 7.2) and 0.1 M glycine-HC1 buffer/0.25 M NaC1 (pH 2.5), respectively. After incubation, BSA was added to the reaction mixtures to a concentration of 10 mg/ml and the remaining antigen activity was assayed by ELISA. For treatments with glycosidases, p-HSP (0.5 mg/ml) was incubated with 3-galactosidase from C. lampas (1.25 U/nl) or endo-3-N-acetylglucosaminidase from T. cornutus (0.25 U/ml) (Biochemicals Co. Ltd.) in 0.1 M citratephosphate buffer/0.5 M NaC1 (pH 4.0) at 37°C for 1--30 h. The treatment with neuraminidase from Clostridium (Boehringer) (0.18 U/ml) was performed in 0.1 M sodium acetate buffer (pH 4.5) at 37°C for 16 h. After neuraminidase treatment, the samples were further treated with 3galactosidase or mixed glycosidases from T. cornutus (Biochemicals Co. Ltd.) at a final concentration of 0.1 U/ml and 1 mg/ml, respectively, for 5--25 h. After incubation, the treated samples were boiled for 3 rain to inactivate enzymatic activity and assayed for the remaining antigen activity by ELISA. In vitro fertilization Human oocytes surrounded with cumulus cells were aspirated from large
32
growing follicles of ovaries which were obtained from the surgical specimens with agreement of the patients, and incubated in Earle's medium containing 7.5% human serum for 4 0 ~ 4 8 h at 37°C in a 5% CO2 incubator for in vitro maturation as described previously (Koyama et al., 1985). Ejaculated human spermatozoa from a healthy donor were washed three times and preincubated in the same medium for 3 h. The sperm suspension and the oocytes were transferred to a droplet containing -y-globulin fractions from 2C6-hybridoma ascites or P3Ul-ascites and incubated for 15 h. The number of spermatozoa bound to or penetrating into the zonae pellucidae of the oocytes was counted under a phase-contrast microscope. Zona-free hamster eggs were prepared by treating superovulated hamster eggs with 0.1% hyaluronidase and 0.1% trypsin according to the method reported by Yanagimachi et al. (1976). They were added to a droplet of preincubated human sperm suspension (1 × 106/ml) in Biggers, Whitten and Whittingham medium (BWW) containing 0.3% BSA and 7-globulin from 2C6-hybridoma or P3U1 ascites. The mixtures were incubated for 18 h and the number of the eggs penetrated by spermatozoa was counted after staining with 0.25% acetolacmoid solution. Results
Fused cells were seeded in 457 wells. Cell growth was observed in all wells seeded and SI activity was detected in the culture supernatant of 23 wells on the 10th day after cell fusion. Seven clones were found to produce SI-Mab which could be absorbed with HSP but not with HM. One of the seven hybridomas with the highest antibody titer was selected for further cloning and established as a stable hybridoma clone (2C6). Culture supernatant of hybridoma 2C6 possessed strong SI (SIs0 titer: 200 units) and SA (1:40 dilutions) activities. The immunoglobulin class of Mab 2C6 was found to be IgM (K) by immunodiffusion test. Results of antibody absorption experiments are summarized in Table 1. Although the absorption with ejaculated human spermatozoa was not complete under the experimental conditions, HSP from azoospermic semen could completely remove the SI and SA activities in the culture supernatant of hybridoma 2C6. HM, NHS and tissue extracts of HK and HL did not remove any antibody activity by absorption. Figure 1 shows the results of the quantitative absorption test of Mab 2C6 with various amounts of human spermatozoa, HSP from normal and vasectomized men and also with HM. The decrease of antibody activity by absorption with HSP from vasectomized men was less than that of normal men. A similar absorption experiment was performed using seminal plasma from bull and boar. The SI activity of Mab 2C6 was completely absorbed out with bull seminal plasma but not with boar seminal plasma at a concentration of 9 mg dry weight/ml.
33 TABLE 1 Absorption of SI-Mab 2C6 with various human materials Each 1 ml of culture supernatant of hybridoma 2C6 was mixed with various human materials and incubated at 4°C overnight. After centrifugation (9500 × g, 30 min) the supernatants were subjected to the sperm immobilization test and the sperm motility was counted; C(+), in the presence of guinea pig serum as a source of complement; C(-), in the presence of guinea pig serum inactivated at 56°C for 30 min; HSP, human seminal plasma from azoospermic semen; HM, human milk; NHS, normal human serum; HK, human kidney; HL, human liver. Materials used for absorption
Concentration of absorbent
None Spermatozoa HSP HM NHS Extract of HK Extract of HL
Sperm motility after absorption
-4 × 107/ml 9 mg/ml a 9 mg/ml a 9 mg/ml a 50 mg/ml a 50 mg/ml a
C(+)
c(-)
0% 42% 73% 0% 0% 0% 0%
74% 76% 73% 72% 74% 70% 74%
amg of dry weight of lyophilized material/ml.
100
o E E £z. 50 Q~
> "-G
#J
E I
I
I
Sperm
"
105
186
187 cells/m~
HSP
"
8.1
1.8
10
mg/m~
Fig. 1. Quantitative absorption of SI-Mab 2C6. Culture supernatants of hybridoma 2C6 were diluted 1:40 and incubated with various amounts of human spermatozoa ( • - - • ), HSP from normal ( • - - • ) and vasectomized ( o - - o ) men and HM ( [ ] - - [] ) at 4°C overnight. After centrifugation, the supernatants were assayed for SI activity by the sperm immobilization test and relative sperm motility was calculated as described in Materials and Methods.
34
The effect of Mab 2C6 on sperm-egg interaction was tested by human sperm binding and penetration of human oocytes and zona-free hamster eggs. As shown in Table 2, 3,-globulin (1 mg/ml) from mouse myeloma P3U1 ascites had no adverse effects on sperm binding nor penetration to both eggs, whereas 3,-globulin (0.12 or 0.04/~g/ml) of hybridoma 2C6 ascites showed a strong inhibitory effect on the sperm-egg interaction. During these fertilization assays, there were no significant sperm agglutinations observed in the presence of "y-globulin from 2C6 ascites at the concentrations used for the assays. When ejaculated spermatozoa were subjected to the indirect immunofluorescent staining, most cells were strongly stained all over the surface. In the immunohistological studies of formalin-fixed and paraffin-embedded tissue sections of male genital organs (Fig. 2), the epithelial cells of the epididymis was stained. No staining was observed in the sections of testis and prostate. Weak and discontinuous staining was observed in the epithelial cells of seminal vesicles. Among the sections of human somatic organs tested, proximal and distal tubules of kidney were weakly stained, but mammary gland and liver did not show staining. For characterization of the antigen corresponding to the Mab, either HSP or trypsin-treated HSP was fractionated by gel filtration through Sephacryl S-300 column and each fraction was assayed for antigenic activity by sperm-
TABLE 2 Blocking effect of SI-Mab 2C6 on human sperm binding and penetration of human oocytes a n d zona-free hamster eggs Human oocytes and zona-free hamster eggs were transferred to droplets of preincubated human sperm suspension in medium containing 3,-globulin from control ascites of mouse myeloma (P3UI) or ascites of hybridoma 2C6 and incubated for 15--18 h at 37°C in 5% CO 2 incubator. After incubation, the number of spermatozoa bound or penetrating into zonae pellucidae or the number of eggs penetrated by spermatozoa were counted as described in Materials and Methods. The concentrations of 7-globulin used were 1 mg/ml for control and 0.12 or 0.04/zg/ml for Mab 2C6. N.T., not tested. Experiments
Human oocytes Control 2C6 (0.12 t~g/ml) Zona-free hamster eggs Control 2C6 (0.12 /zg/ml) 2C6 (0.04 ~g/ml)
No. of eggs used
3 8
15 16 17
No. of sperm bound to zona
No. of sperm penetrated into zona
No. of eggs penetrated by sperm
56.0 4- 16.5 2.8 ± 4.1
4.7 ± 2.9 0.0 ± 0.0
3 0
N.T. N.T. N.T.
N.T. N.T. N.T.
15 2 6
Fig. 2. lmmunohistochemical staining of male genital organs in human with Mab 2C6. A, testis. B, epididymis, x 150. Arrows indicate positive staining with Mab 2C6.
36
ELISA inhibition assay. As shown in Fig. 3 for the latter sample, most antigenic activity was eluted in the void fraction. It indicated that the molecular weight (MW) of the corresponding antigen was greater than 670 000. However, when the same sample was applied to SDS-PAGE and transblotted to nitrocellulose membranes, the immunostaining by Mab 2C6 was detected as a broad band around 15--25 kDa under both non-reducing and reducing conditions (Fig. 4). The void fraction of HSP from Sephacryl S-300 chromatography, which was designated as p-HSP in this paper, was subjected to heat treatment (56--100°C) for 30 rain and its reactivity with Mab 2C6 was compared with that of non-treated material by competitive binding inhibition assay of Mab 2C6 to human spermatozoa coated on ELISA plates: The reactivity was completely retained even after boiling the sample for 30 min. When the same material was treated with sodium metaperiodate, the antigen activity was almost completely destroyed (Fig. 5). After deglycosylation by treatment with TFMS, p-HSP completely lost reactivity with Mab 2C6 as tested by either ELISA or antibody absorption test (Fig. 6).
1.0
t
./
H*
L/ Inhibition assay by ELI SA
x
0.8
