- Email: [email protected]
Production and characterization of monoclonal antibodies to cross-reactive antigens of human and porcine zonae pellucidae
Journal of Reproductive Immunology, 7 (1985) 187-198
187
Elsevier JRI 00341
Production and characterization of monoclonal antibodies to cross-reactive antigens of human and porcine zonae pellucidae K. Koyama, A. Hasegawa, Y. Tsuji and S. Isojima Department of Obstetrics and Gynecology, Hyogo Medical College, Nishinomiya, Japan (Accepted for publication 7 September 1984)
Three monoclonal antibodies (Mabs), 3A4-2G1, 1D5-2B7 and 1F2-1B8 were produced against heat-solubilized porcine zona pellucida (ZP). Each Mab stained intact ZP but no other pig tissues using immunofluorescence staining. All three Mabs stained selectively zonae pellucidae (ZPe) from pigs and humans but not from hamsters, rats or mice, and showed no inhibitory effect on sperm binding to human oocytes. When goat antiserum to mouse -/-globulin was added to human oocytes pre-treated with 3A4-2G1 or 1D5-2B7, sperm binding to oocytes was completely blocked with formation of immune precipitates around them. SDS-PAGE analysis of the immune precipitates of 12SI-labeled porcine zona proteins and Mab showed that the antigen binding 3A4-2G1 was mainly composed of components with approximate molecular weights of 92,000, 65,000 and 23,000 and the antigen binding 1D5-2B7 contained two components with approximate molecular weights of 57,000 and 49,000, respectively. The epitope of ZP antigen, corresponding to 3A4-2G1, was found to be present in the molecule of 92,000 daltons as demonstrated by enzyme immunostaining of the proteins after blotting to nitrocellulose membrane from SDS-PAGE gels. Key words: zona pellucida, monoclonal antibodies, cross-reactions, inhibitory effects
Introduction
Since the presence of cross-reactive antigen between porcine and human ZP has been demonstrated by Sacco (1977) and the antibody cross-reacting to porcine ZP was detected in the sera of some sterile women by immunofluorescence staining (Shivers and Dunbar, 1977), many studies have been carried out attempting to isolate the zona specific cross-reacting antigen of porcine ZP for use in a human contraceptive vaccine (Dunbar et al., 1980, 1981; Dunbar and Raynor, 1980). Sacco et al. (1981) demonstrated that a rabbit antiserum to one component of porcine ZP inhibited homologous sperm binding to oocytes from human and monkey. Inhibition of fertility was also shown by active immunization of rabbits with porcine ZP (Wood et al., 1981; Freye and Mettler, 1982). It has been assumed that the inhibitory effect on fertility by active or passive immunization against ZP was due to the blocking of sperm binding to oocytes. However, it is not known whether the blocking effect depends upon antibodies to 0165-0378/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
188
the sperm receptors of ZP or to steric hindrance of the receptors by precipitates due to antibodies to the other antigenic components of ZP. Previous studies seem to support the latter mechanism (Ahuja and Tzartos, 1981; Tsunoda et al., 1981; Aitken et al., 1982). In order to study the distribution of various antigenic components of porcine ZP among other species and their biological function in fertilization, the production of monoclonal antibodies (Mabs) to ZP was essential. Recently, we succeeded in establishing hybridomas which produced Mabs to porcine ZP and their characteristic properties have been reported in our previous papers (Isojima et al., 1981, 1984). In the present study, Mabs to cross-reacting antigens of human and porcine ZP were produced and their inhibitory effects on in vitro fertilization of human oocytes were examined. The molecular nature of the antigens corresponding to the Mabs was also studied and the results are presented in this paper.
Materials and Methods
Isolation and solubilization of porcine zona pellucida Porcine oocytes were collected by washing minced ovaries through nylon screens with meshes of different pore size as described previously (Isojima et al., 1984). After gentle homogenization of the collected oocytes in a Dounce tissue grinder, the detached ZPe were isolated from other cellular debris by repeated washings of the homogenates through a nylon screen with 40 /~m meshes. Solubilization of the isolated ZP was carried out by heating in 0.05 M sodium carbonate-bicarbonate buffer (pH 9.6) at 60°C for 60 rain, and the solubilized materials of ZP were obtained by centrifugation at 21,000 x g for 20 rain.
Establishment of hybridomas Procedures for immunization and cell fusion were essentially the same as described earlier (Isojima et al., 1984), with a slight modification. Briefly, a male BALB/c mouse was injected subcutaneously with an emulsion of 250 porcine ZPe in complete Freund's adjuvant. The second injection was given in incomplete Freund's adjuvant and then followed by two intraperitoneal (i.p.) booster injections of 250 porcine ZPe in saline. Three days before removal of the spleen, the mouse was given an i.p. injection of the heat-solubilized porcine zona solution equivalent to 250 ZPe. The mouse spleen cells (1 x l08) were fused with mouse myeloma cells (P3/x63Ag8Ul:P3U1) (1.8 x 107) by polyethylene glycol (PEG 1000) and the fused cells were distributed into 480 wells, and then cultured in HAT (hypoxanthine-aminopterin-thymidine) selection medium. Forty seven cell cultures among 480 wells showed positive antibody production by ELISA assay as described in detail later. These 47 antibody positive supernatants were further examined by indirect immunofluorescence staining and it was found that 5 of them stained ZPe from both porcine and human oocytes. Three cell lines (3A4, 1D5, 1F2) with high antibody titers were cloned by the limiting dilution method and the one clone from each culture which showed positive staining of porcine and human oocytes by
189 immunofluorescence test was selected for establishment of the hybridoma cell line. The monoclonal antibodies were obtained either from the hybridoma culture media or from the ascites fluid which was produced in BALB/c mice by giving an i.p. injection of 1 × 107 hybridoma cells one week after i.p. administration of 0.5 ml of pristane. The ascites fluid was collected 7-10 days after the cell injections.
Enzyme-linked immunosorbent assay (ELISA) Antibody production to the soluble antigens of porcine ZP was examined by ELISA combined with a biotin-avidin reaction system. Heat-solubilized ZP antigens were coated on Titertek Micro-ELISA plate U-20001 (Dynatech Lab. Inc.) by adding 100 /~l/well of the solution containing 100 /~g ZP proteins/ml in 0.05 M sodium carbonate-bicarbonate buffer, pH 9.6, following incubating at 4°C overnight. After blocking residual protein binding sites on the wells by incubating with 1% bovine serum albumin (BSA) solution for 2 h at room temperature (RT), they were washed with 2 mM phosphate buffered saline (PBS), pH 7.4, and then 100 #1 of the supernatants of culture medium were added to the wells coated with antigen. The plates were incubated for 60 min at RT and then the wells were washed 5 times with PBS containing 0.05% Tween 20. One hundred/~1 of a biotinylated IgG solution (10 /~g/ml) of horse antiserum to mouse y-globulin (Vector Laboratories, Inc.) were added in the wells and then incubated for 60 min at RT. After washing with PBS, 100 #1 of horseradish peroxidase conjugated avidin (5/~g/ml of PBS: Vector Lab. Inc.) were added and further incubated for 30 rain at RT. The peroxidase activity was measured by adding 100 t~l of an O-phenylenediamine solution (1 mg/ml of 0.15 M citrate phosphate buffer, pH 5.0) containing 0.05% H202 after extensive washing of the wells. The wells were kept for 10 min at RT. After stopping the reaction by adding 50 t~l of HzSO4 solution to the wells the optical density was measured in a micro-plate photometer (Corona Electric, Japan).
Immunofluorescence staining The indirect immunofluorescence method was used for staining of cumulus-free oocytes from pigs and humans. The details of the staining procedure by fluorescent isothiocyanate (FITC) conjugated F(ab')z fragment of IgG from rabbit anti-mouse y-globulin serum (Cappel Laboratories, Inc.) was previously described (Isojima et al., 1984). Porcine and human oocytes were obtained by puncture of large follicles of each fresh ovary and cumulus cells surrounding the oocytes were removed mechanically.
In vitro fertilization (IVF) of human oocytes Oocytes surrounded with cumulus cells were obtained by puncture of large follicles of human ovaries which were partly removed by surgery in the case of polycystic ovaries. The collected oocytes were incubated in modified Krebs-Ringer bicarbonate medium (mKRB), pH 7.4, containing 7% inactivated human serum for 40-48 h at 37°C in a 5% CO2 incubator and then incubated with antibodies to porcine ZP for 30 min at 37°C. After washing with mKRB medium, these oocytes were transferred to a droplet of sperm suspension (1 × 106/ml) under liquid
190 paraffin. The sperm suspension was prepared by washing ejaculated human spermatozoa twice with mKRB medium containing 0.1% BSA and preincubating in the same medium for 3 h at a concentration of 1-5 × 107/ml. The concentration was adjusted to 1 × 106/ml with mKRB with 7% human serum before use. The inseminated oocytes were incubated for 15 h at 37°C in a 5% CO 2 incubator and the number of spermatozoa bound to or penetrated through the ZP of the oocytes was then counted under a phase-contrast microscope, after washing by repeated pipetting. Human oocytes were treated with conventional antiserum to porcine ZP or ascites from mice which were inoculated i.p. with established hybridomas, after heat inactivation of complement (56°C, 30 min) and extensive dialysis against mKRB medium.
Iodination of solubilized zona proteins Heat-solubilized porcine ZP (10 /~g) was labeled with Na125I (1 mCi) by the chloramine-T method and the 125I-labeled proteins were isolated by Sephadex G-25 column (0.8 × 15 cm). Immunoaffinity chromatography of 125I-labeled zona proteins In order to isolate the antigens corresponding to 3A4-2G1 and 1D5-2B7 Mabs, the 125I-labeled zona proteins (ca. 100 x 106 cpm) were passed through a column (0.5 x 1 cm) of Sepharose 4B gel linked with a Mab (v-globulin of culture supernatant, 2.0 mg/ml of gel) and then the antibody bound fraction was eluted with 0.2 M glycine HC1 buffer (pH 2.5) as described in our previous report (Isojima et al., 1982). The antibody bound fraction was precipitated by the corresponding Mab and goat antiserum to mouse v-globulin as second antibody. The immune precipitates were analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to the method described previously (Isojima et al., 1982). Transfer of proteins from SDS-PA GE to nitrocellulose membrane The heat-solubilized porcine zona proteins (220 /~g/lane) were applied to 10% polyacrylamide slab gels (0.5 × 13 × 0.2 cm) and separated by electrophoresis with constant current of 4 mA for 20 h in 0.025 M Tris-glycine buffer (pH 8.3) containing 0.1% SDS. The proteins separated by SDS-PAGE were transferred to nitrocellulose membrane sheets according to the method described by Towbine et al. (1979) at 220 mA for 3 h in SDS-PAGE buffer containing 20% methanol. After trans-blotting, the antigens specific to Mab were stained on the nitrocellulose membrane by the ELISA method. For staining, the membrane was incubated with Mab (1 : 40 diluted ascites) at 4°C overnight and followed by treatment with peroxidase-labeled goat anti-mouse IgG (peroxidase activity 1072 units/ml by pyrogallol method, IgG/peroxidase = 0.87, 1:500 dilution, Miles) at RT for 2 h. 0.0025% O-dianisidine in 10 mM Tris-HCl (pH 7.4) containing 0.01% H/O 2 was added to the nitrocellulose membrane after washing in order to visualize the reaction products by peroxidase.
191 TABLE 1 Characteristics of three hybridoma-produced monoclonal antibodies to porcine zona pellucida Hybridoma clones
Immunoglobulin classes a
Precipitinline to sol. zona b
Immunofluorescent staining of porcine tissues
3A4-2G1 1D5-2B7 1F2-1B8
IgG1 IgG1 IgM
+ -
+ + +
Oocytes
Testes L i v e r Kidney
Brain
m
a Determined by radial immunodiffusion test using class-specific goat anti-mouse IgM and IgG1 serum. b Heat-solubilized porcine zona proteins (5 mg/ml) were reacted to each monoclonal antibody by Ouchterlony's double immunodiffusion test.
Results Three hybridomas (3A4-2G1, 1D5-2B7, 1F2-1B8) which p r o d u c e d Mabs to the soluble antigens of porcine Z P were established. The characteristics of each M a b are summarized in Table 1. The classes of immunoglobulin secreted by the three hybridomas were IgG1, IgG1 and IgM, respectively. Immunofluorescence study showed that all three Mabs stained only oocytes and no other tissues from pigs. The M a b produced by 1D5-2B7 h y b r i d o m a formed an immunoprecipitin line against heat-solublized porcine zona proteins in Ouchterlony's agar immunodiffusion test. As shown in Fig. 1, a conventional mouse antiserum to porcine Z P formed at least two precipitin lines against the solubilized porcine zona antigens and the precipitin line of 1D5-2B7 M a b was fused with one of those two precipitin lines. Figure 1 also shows that the M a b to porcine Z P from B l l C 8 hybridoma, which was characterized in our previous report (Isojima et al., 1984), formed a precipitin line against the porcine zona antigen, but this precipitin line fused with another one of the two precipitin lines formed by the conventional antiserum. The precipitin lines formed by 1D5-2B7 and B l l C 8 Mabs completely crossed each other. Neither M a b 3A4-2G1 nor 1F2-1B8 m a d e an immunoprecipitin line against the heat-solublized porcine zona proteins.
!
m
Fig. 1. Immunodiffusion of heat-solubilized porcine zona proteins with conventional and monoclonal antibodies to porcine zona antigens. Center well: heat-solubilized porcine zona proteins (Prot. 5 mg/ml). Con. Ab: mouse antiserum to porcine zona antigens (undiluted antiserum). 1D5-2B7 and BllC8: monoclonal antibodies to porcine zona antigens ('t-globulin of each hybridoma ascites: 10 mg/ml).
192 As illustrated by immunofluorescence study (Fig. 2), all three Mabs stained ZPe of porcine and human oocytes but did not stain any oocytes of hamsters, rats and mice. The staining patterns of ZPe in both pigs and humans by the three Mabs were quite similar but the staining intensity by Mab 1F2-1B8 was rather weak compared with the other two, even though the same protein concentration from the culture medium was used for staining. The difference of specificity between the two Mabs 3A4-2G1 and 1D5-2b7 was proved by the analysis of SDS-PAGE patterns of immune precipitates which were formed by each Mab with 125I-labeled porcine zona proteins, even though the staining patterns and staining intensity of ZPe by both Mabs were quite similar. When human oocytes were treated with conventional antiserum to porcine ZP, the binding of human spermatozoa to ZP was completely blocked (Table 2). However, numerous spermatozoa could bind to the ZP of human oocytes which were treated with ascites from mice inoculated with either mouse myeloma cells (P3U1) or hybridomas (3A4-2G1, 1D5-2B7 and 1F2-1B8). Moreover, some spermatozoa could penetrate through the ZP of the treated oocytes. No inhibition of sperm binding and penetration was observed even when the two kinds of ascites (3A4-2G1 and 1D5-2B7) were mixed and used for the treatment of human oocytes. When human oocytes were first treated with ascites (3A4-2G1 or 1D5-2B7) and then treated with goat antiserum to mouse v-globulin as second antibody, the sperm binding and penetration were completely blocked with the light scattering precipitin layer over the ZP of the treated oocytes. The immune precipitin layer was also observed following treatment of human oocytes with conventional antiserum to porcine ZP,
pig
3A4-2GI
human
hamster
rat
mouse
mmn
1D5-2B7
1F2-1B8
I/
m
Fig. 2. FITC staining patterns of oocytes from pigs, humans, hamsters, rats and mice by monoclonal antibodies to porcine zona antigens. Note that the three monoclonalantibodies used stained only porcine and human oocytes.
193 TABLE 2 Effect of monoclonal antibodies to porcine zona pellucida on in vitro fertilization of human oocytes Antibodies to zona pellucida
No. of oocytes examined
No. of spermatozoa bound to zona/oocyte
Zona penetration by spermatozoa
Antiserum a Control b
5 9
0 > 50
+
3A4-2G1 1D5-2B7 1F2-1B8
2 2 2
> 50 > 50 > 50
+ + +
3A4-2G1 + 1D5-2B7 c
3
> 50
+
3A4-2G1 + anti-fg d 1D5-2B7 + anti-Ig 1F2-1B8 + anti-Ig
4 4 3
0 0 > 50
-+
a b c d
Conventional mouse antiserum to porcine zona pellucida. Ascites from a mouse inoculated with P3U1 mouse myeloma cells. Ascites from 3A4-2G1 and 1D5-2B7 hybridomas were mixed at 1 : 1 ratio. After treatment with ascites from 3A4-2G1, the oocytes were further treated with goat antiserum to mouse y-globulin.
but no precipitin layer was observed by treatment only with ascites produced by the hybridoma. Contrary to 3A4-2G1 and 1D5-2B7, the 1F2-1B8 Mab neither blocked sperm binding to human oocytes nor made an immune precipitin layer over the zona after treatment with the second antibody. 125I-labeled porcine zona proteins were fractionated by immunoaffinity chromatography on Sepharose 4B gels treated with Mab (3A4-2G1 or 1D5-2B7) and the binding activity of each fraction with the corresponding Mab was measured. As shown in Table 3, the binding activity of the antibody-bound fraction was much higher than that of the unfractionated original material and the antibody-unbound fraction. The immune precipitates of the antibody-bound fraction with the correTABLE 3 Binding assay of the fractions from the antibody affinity chromatography of 125I-labeled zona proteins 125l-zona proteins
Original material Mab unbound fraction Mab bound fraction
Binding to Mab (%) a 3A4-2G1
1D5-2B7
5.0 0.7 16.3
17.9 7.6 38.2
a For the binding assay, a sample of 100/~1 of 12SI-labeled antigen (ca. 20,000 c.p.m.) and 100/~1 of each Mab (hybridoma culture supernatant) were mixed and kept for 48 h at 4°C. As a second antibody, 100 /L1 of rabbit anti-mouse ~,-globulin (1 : 5 diluted) was added and the mixture was kept for another 24 h at 4°C. The precipitate was separated by centrifugation (1800×g, 10 rain) and counted for the radioactivity by an auto-gamma scintillation counter. The net binding (%) was calculated by substracting the control binding (1 : 2000 diluted normal mouse serum) from that due to Mab.
194
sponding Mab were further fractionated by SDS-PAGE. Figure 3 shows the comparison of SDS-PAGE patterns of the original ]25I-labeled porcine zona proteins and the immune precipitates with Mabs of 3A4-2G1 and 1D5-2B7. Each sample showed a quite different pattern on SDS-PAGE. The main peaks of radioactivity were present in the zone of much higher than 100,000 daltons and around 67,000 daltons in the original 125I-labelled zona proteins, and around 92,000, 65,000 and cpm
4000
9°;00° ,.ooo,;ooo ,o.ooo 2o.,oo
2000
I
I
I
I
I
!
1500
I000
500
i
I
s00 0
i
I
!
I
10
20
30
40
fraction
number
Fig. 3. Radioactivity applied to SDS-PAGE was ca. 100,000 c.p.m, in 125I-labeled zona proteins, ca. 15,000 c.p.m, in the immune precipitate with 3A4-2G1 and ca. 8000 c.p.m, in the i m m u n e precipitate with 1D5-2B7. A dotted Line in the second panel shows a pattern of radioactivities (ca. 2000 c.p.m.) of lESI-zona proteins non-specifically immunoprecipitated with 4C7 (mouse IgG1 Mab established to a cell Line of h u m a n ovarian cancer in our laboratory).
195
23,000 daltons in the immune precipitates with Mab 3A4-2G1. The immune precipitate with Mab 1D5-2B7 showed peaks around 57,000 and 49,000 daltons, though the original ]25I-zona material contained components of these molecular sizes only in a small amount. These results may possibly be due to the concentration of minor components containing immunoprecipitable antigens by two step purifications of the corresponding antigen. Fig. 4 shows the staining patterns of antigens trans-blotted to Staining patterns of antigens transblotted to a nitrocellulose membrane from SDS-PAGE of heat-solubilized porcine zona proteins A. 1 iiiiiiii~iiii~!il!
2
3
B.. 1
2
3
F-l -
I ] <1 94K
"
I <1 67K
i!~!ili!i~iliii~ il~ii!;!!iiiiii!~
~iiiiiiiiiiii~iiiiiil ii!iiiiiiiiiii~iiiiiiii
I <1 43K
i~!!!,!i!!!2!31 i!ii!~!!!ii!~!~iiill
iii :~iili ~ iii!
Fig. 4. A is the original picture and B a schematic picture of the staining patterns. Channel 1 was stained with a conventional antiserum to porcine ZP, channel 2 stained with 3A4-2G1 and channel 3 stained with lD5-2B7.
196 a nitrocellulose membrane from SDS-PAGE of heat-solubilized porcine zona proteins. The staining pattern with the conventional antiserum to porcine ZP was broad and homogeneous with a distribution from large to small molecules. 3A4-2G1 Mab stained antigen molecules restricted to around 92,000 daltons but 1D5-2B7 Mab was unable to stain any molecules trans-blotted to the nitrocellulose membrane.
Discussion
In our previous study (Isojima et al., 1984), we established five hybridomas which produced Mabs to porcine ZP and showed that the Mab BllC8 which was generated to the species specific zona antigen of pigs strongly blocked the binding of boar spermatozoa to porcine oocytes. However, other Mabs which cross-reacted with hamster ZP did not block the binding of hamster spermatozoa to hamster oocytes. When goat antiserum to mouse T-globulin was applied as the second antibody, one (G10G5) of the Mabs which cross-reacted with hamster ZP showed a blocking effect on sperm binding to hamster oocytes, with formation of a light-scattering precipitin layer around the oocytes. From these results, we assumed that the Mab which recognized the species-specific antigen of ZP would block sperm binding to oocytes in the same species by reacting possibly to a sperm receptor antigen on ZP, because it is generally believed that the sperm receptor is species specific (Peterson et al., 1980). On the other hand, Mabs which recognize cross-reactive zona antigens other than pigs might bind to portions different from the sperm receptor of ZP in other species, because they were unable to block sperm binding. However, when antigen epitopes are present at high density on ZP they may possibly be located in proximity to the sperm binding sites of the ZP. Therefore the corresponding Mab could block sperm binding to ZP by forming dense immunoprecipitates with the second antibody .over the surface of the ZP. Similarly, if the antigen epitopes are present close to the sperm receptors of ZP, the sperm binding could be blocked by steric hindrance of the receptors with immunoprecipitates formed by the corresponding Mab and the second antibody, even though their distribution may be sparse. As shown in the present experiments, though the three Mabs (3A4-2G1, 1D5-2B7 and 1F2-1B8) generated to heat-solubilized porcine ZP cross-reacted with porcine and human ZP, they did not show any inhibitory effect on human sperm binding to homologous oocytes. However, when goat antiserum to mouse -/-globulin was added as the second antibody in addition to Mab 3A4-2G1 or 1D5-2B7, sperm binding was completely inhibited with the formation of an immune precipitin layer around the oocytes. Thus, the present experiment not only confirmed our previous results in a hamster experiment but also indicated the presence of at least three antigens (epitopes) in human ZP which were common to porcine ZP. In attempting to make a contraceptive vaccine from porcine ZP, isolation of porcine ZP antigen which is common to human ZP is essential. The establishment of hybridomas which produced Mabs 3A4-2G1 and 1D5-2B7 was very useful because these two Mabs may be very convenient and specific tools for the development of
197
such a contraceptive vaccine. The antigens (epitopes) corresponding to these two Mabs will not be involved in the sperm receptor proteins on human ZP. However, if molecules of porcine ZP including antigens (epitopes) which correspond to these Mabs are isolated, fertilization could possibly be blocked with conventional antibodies produced to the isolated molecules by lattice formation of immune precipitates on human ZP. The established Mabs (3A4-2G1 and 1D5-2BT) were therefore used for isolating the corresponding antigens from the heat-solubilized porcine zona proteins by immunoaffinity chromatography. The antigenic molecule corresponding to 3A4-2G1 seems to be composed of three molecules of different size (92,000, 65,000 and 23,000 daltons) and the antigen (epitope) proved to be present on the molecule of 92,000 daltons by the western blotting procedure after SDS treatment. On the other hand, the antigenic molecule corresponding to 1D5-2B7 seems to be composed of two molecules of different size (51,000 and 49,000 daltons), but it was not clear in which molecules the antigen (epitope) was preserved by western blotting after SDS treatment (see Fig. 4). This might be due to damage of the antigen epitope corresponding to 1D5-2B7 by the process of SDS-PAGE or that the antigen after SDS-PAGE could not be transferred to the membrane under these experimental conditions. The same explanation could apply to the molecule of 65,000 and 23,000 daltons corresponding to Mab 3A4-2G1. All these experiments were carried out by using very small amounts of radiolabeled materials of porcine zona proteins, but we are now trying to isolate the corresponding antigens on a large scale using the same technique. The sperm binding receptor of ZP must be species specific; therefore the blocking of fertilization of Mab to zona antigen could be expected only in the same species from which the ZP was used for immunization. On this assumption, we cannot expect to prevent fertilization in humans by Mabs to porcine ZP. We then have to aim to isolate those antigenic molecules close to the sperm binding receptor as the second choice. It is true that several cross-reacting porcine ZP antigens are distributed over the human ZP; therefore there is a possibility of isolating some common antigenic components densely present on human ZP or present close to the sperm binding receptor of human ZP. There are biochemical and immunological ways of purifying these antigenic components. The immunological procedure was preferred because it can select and pick up antigens common only to porcine and human ZPe. This study could indicate a very promising molecule of 92,000 daltons in the porcine ZP. Recently, Yurewicz et al. (1983) and Sacco et al. (1984) reported a molecule of porcine ZP purified by biochemical procedures. They showed that the molecular weight of the purified component was 58,000 and that this molecule possessed the sperm receptor activity to porcine spermatozoa. However, from porcine ZP, it will be difficult to isolate the antigen (epitope) in the sperm binding receptor of human ZP, but the molecule of 92,000 daltons in the antigen corresponding to Mab 3A4-2G1 seems to be present at high density on human ZP or located close to the sperm binding receptor of human ZP. Thus, there is a possibility that fertilization could be impaired by conventional antibodies to this molecule of 92,000 daltons.
198
References Ahuja, K.K. and Tzartos, S.J. (1981) Investigation of sperm receptors in the hamster zona pellucida by using univalent (Fab) antibodies to hamster ovary. J. Reprod. Fert. 61,257-264. Aitken, R.J., Holme, E., Richardson, D.W. and Hulme, M. (1982) Properties of intact and univalent (Fab) antibodies raised against isolated, solubilized, mouse zonae pellucidae. J. Reprod. Fert. 66, 327-334. Dunbar, B.S. and Raynor, B.D. (1980) Characterization of porcine zona pellucida antigens. Biol. Reprod. 22, 941-954. Dunbar, B.S., Wardrip, N.J. and Hedrick, J.L. (1980) Isolation, physicochemical properties, and macromolecular composition of zona pellucida from porcine oocytes. Biochemistry 19, 356-365. Dunbar, B.S., Lin, C. and Sammonds, D.W. (1981) Identification of the three proteins of porcine and rabbit zonae pellucidae by high resolution two dimensional gel electrophoresis: comparison with serum, follicular fluid, and ovarian cell proteins. Biol. Reprod. 24, 1111-1124. Freye, J.D. and Mettler, L. (1982) Fertility inhibition using low-dose immunization with porcine zonae pellucidae. Am. J. Reprod. Immunol. 2, 153-156. Isojima, S., Koyama, K. and Hasegawa, A. (1981) Production of monoclonal antibody to zona pellucida from porcine oocytes. Acta Obstet. Gynaecol. Jpn. 33, 1995-1996. Isojima, S., Koyama, K. and Fujiwara, N. (1982) Purification of human seminal plasma No. 7 antigen by immunoaffinity chromatography on bound monoclonal antibody. Clin. Exp. ImmunoL 49, 449-456. Isojima, S., Koyama, K., Hasegawa, A., Tsunoda, Y. and Hanada, A. (1984) Monoclonal antibodies to porcine zona pellucida antigens and their inhibitory effects on fertilization. J. Reprod. Immunol. 6, 77-87. Peterson, R.N., Russel, L., Bundman, D. and Freund, M. (1980) Sperm-egg interaction: evidence for boar sperm plasma membrane receptors for porcine zona pellucida. Science 207, 73-74. Sacco, A.G. (1977) Antigenic cross-reactivity between human and pig zona pellucida. Biol. Reprod. 16, 164-174. Sacco, A.G., Yurewicz, E.C., Subramanian, M.G. and De Mayo, F.J. (1981) Zona pellucida composition: species cross-reactivity and contraceptive potential of antiserum to a purified pig zona antigen (PPZA). Biol. Reprod. 25, 439-450. Sacco, A.G., Subramanian, M.G. and Yurewicz, E.C. (1984) Association of sperm receptor activity with a purified pig zona antigen (PPZA). J. Reprod. Immunol. 6, 89-103. Shivers, C.A. and Dunbar, B.S. (1977) Autoantibodies to zona pellucida: A possible cause for infertility in women. Science 197, 1082-1084. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350-4354. Tsunoda, Y., Sugie, T., Mori, J., Isojima, S. and Koyama, K. (1981) Effect of purified zona antibody on fertilization in the mouse. J. Exp. Zool. 271, 103-108. Wood, D.M., Lia, C. and Dunbar, B.S. (1981) Effect of alloimmunization and heteroimmunization with zonae pellucidae on fertility in rabbits. Biol. Reprod. 25, 439-450. Yurewicz, E.C., Sacco, A.G. and Subramanian, M.G. (1983) Isolation and preliminary characterization of a purified pig zona antigen (PPZA) from porcine oocytes. Biol. Reprod. 29, 511-523.
187
Elsevier JRI 00341
Production and characterization of monoclonal antibodies to cross-reactive antigens of human and porcine zonae pellucidae K. Koyama, A. Hasegawa, Y. Tsuji and S. Isojima Department of Obstetrics and Gynecology, Hyogo Medical College, Nishinomiya, Japan (Accepted for publication 7 September 1984)
Three monoclonal antibodies (Mabs), 3A4-2G1, 1D5-2B7 and 1F2-1B8 were produced against heat-solubilized porcine zona pellucida (ZP). Each Mab stained intact ZP but no other pig tissues using immunofluorescence staining. All three Mabs stained selectively zonae pellucidae (ZPe) from pigs and humans but not from hamsters, rats or mice, and showed no inhibitory effect on sperm binding to human oocytes. When goat antiserum to mouse -/-globulin was added to human oocytes pre-treated with 3A4-2G1 or 1D5-2B7, sperm binding to oocytes was completely blocked with formation of immune precipitates around them. SDS-PAGE analysis of the immune precipitates of 12SI-labeled porcine zona proteins and Mab showed that the antigen binding 3A4-2G1 was mainly composed of components with approximate molecular weights of 92,000, 65,000 and 23,000 and the antigen binding 1D5-2B7 contained two components with approximate molecular weights of 57,000 and 49,000, respectively. The epitope of ZP antigen, corresponding to 3A4-2G1, was found to be present in the molecule of 92,000 daltons as demonstrated by enzyme immunostaining of the proteins after blotting to nitrocellulose membrane from SDS-PAGE gels. Key words: zona pellucida, monoclonal antibodies, cross-reactions, inhibitory effects
Introduction
Since the presence of cross-reactive antigen between porcine and human ZP has been demonstrated by Sacco (1977) and the antibody cross-reacting to porcine ZP was detected in the sera of some sterile women by immunofluorescence staining (Shivers and Dunbar, 1977), many studies have been carried out attempting to isolate the zona specific cross-reacting antigen of porcine ZP for use in a human contraceptive vaccine (Dunbar et al., 1980, 1981; Dunbar and Raynor, 1980). Sacco et al. (1981) demonstrated that a rabbit antiserum to one component of porcine ZP inhibited homologous sperm binding to oocytes from human and monkey. Inhibition of fertility was also shown by active immunization of rabbits with porcine ZP (Wood et al., 1981; Freye and Mettler, 1982). It has been assumed that the inhibitory effect on fertility by active or passive immunization against ZP was due to the blocking of sperm binding to oocytes. However, it is not known whether the blocking effect depends upon antibodies to 0165-0378/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
188
the sperm receptors of ZP or to steric hindrance of the receptors by precipitates due to antibodies to the other antigenic components of ZP. Previous studies seem to support the latter mechanism (Ahuja and Tzartos, 1981; Tsunoda et al., 1981; Aitken et al., 1982). In order to study the distribution of various antigenic components of porcine ZP among other species and their biological function in fertilization, the production of monoclonal antibodies (Mabs) to ZP was essential. Recently, we succeeded in establishing hybridomas which produced Mabs to porcine ZP and their characteristic properties have been reported in our previous papers (Isojima et al., 1981, 1984). In the present study, Mabs to cross-reacting antigens of human and porcine ZP were produced and their inhibitory effects on in vitro fertilization of human oocytes were examined. The molecular nature of the antigens corresponding to the Mabs was also studied and the results are presented in this paper.
Materials and Methods
Isolation and solubilization of porcine zona pellucida Porcine oocytes were collected by washing minced ovaries through nylon screens with meshes of different pore size as described previously (Isojima et al., 1984). After gentle homogenization of the collected oocytes in a Dounce tissue grinder, the detached ZPe were isolated from other cellular debris by repeated washings of the homogenates through a nylon screen with 40 /~m meshes. Solubilization of the isolated ZP was carried out by heating in 0.05 M sodium carbonate-bicarbonate buffer (pH 9.6) at 60°C for 60 rain, and the solubilized materials of ZP were obtained by centrifugation at 21,000 x g for 20 rain.
Establishment of hybridomas Procedures for immunization and cell fusion were essentially the same as described earlier (Isojima et al., 1984), with a slight modification. Briefly, a male BALB/c mouse was injected subcutaneously with an emulsion of 250 porcine ZPe in complete Freund's adjuvant. The second injection was given in incomplete Freund's adjuvant and then followed by two intraperitoneal (i.p.) booster injections of 250 porcine ZPe in saline. Three days before removal of the spleen, the mouse was given an i.p. injection of the heat-solubilized porcine zona solution equivalent to 250 ZPe. The mouse spleen cells (1 x l08) were fused with mouse myeloma cells (P3/x63Ag8Ul:P3U1) (1.8 x 107) by polyethylene glycol (PEG 1000) and the fused cells were distributed into 480 wells, and then cultured in HAT (hypoxanthine-aminopterin-thymidine) selection medium. Forty seven cell cultures among 480 wells showed positive antibody production by ELISA assay as described in detail later. These 47 antibody positive supernatants were further examined by indirect immunofluorescence staining and it was found that 5 of them stained ZPe from both porcine and human oocytes. Three cell lines (3A4, 1D5, 1F2) with high antibody titers were cloned by the limiting dilution method and the one clone from each culture which showed positive staining of porcine and human oocytes by
189 immunofluorescence test was selected for establishment of the hybridoma cell line. The monoclonal antibodies were obtained either from the hybridoma culture media or from the ascites fluid which was produced in BALB/c mice by giving an i.p. injection of 1 × 107 hybridoma cells one week after i.p. administration of 0.5 ml of pristane. The ascites fluid was collected 7-10 days after the cell injections.
Enzyme-linked immunosorbent assay (ELISA) Antibody production to the soluble antigens of porcine ZP was examined by ELISA combined with a biotin-avidin reaction system. Heat-solubilized ZP antigens were coated on Titertek Micro-ELISA plate U-20001 (Dynatech Lab. Inc.) by adding 100 /~l/well of the solution containing 100 /~g ZP proteins/ml in 0.05 M sodium carbonate-bicarbonate buffer, pH 9.6, following incubating at 4°C overnight. After blocking residual protein binding sites on the wells by incubating with 1% bovine serum albumin (BSA) solution for 2 h at room temperature (RT), they were washed with 2 mM phosphate buffered saline (PBS), pH 7.4, and then 100 #1 of the supernatants of culture medium were added to the wells coated with antigen. The plates were incubated for 60 min at RT and then the wells were washed 5 times with PBS containing 0.05% Tween 20. One hundred/~1 of a biotinylated IgG solution (10 /~g/ml) of horse antiserum to mouse y-globulin (Vector Laboratories, Inc.) were added in the wells and then incubated for 60 min at RT. After washing with PBS, 100 #1 of horseradish peroxidase conjugated avidin (5/~g/ml of PBS: Vector Lab. Inc.) were added and further incubated for 30 rain at RT. The peroxidase activity was measured by adding 100 t~l of an O-phenylenediamine solution (1 mg/ml of 0.15 M citrate phosphate buffer, pH 5.0) containing 0.05% H202 after extensive washing of the wells. The wells were kept for 10 min at RT. After stopping the reaction by adding 50 t~l of HzSO4 solution to the wells the optical density was measured in a micro-plate photometer (Corona Electric, Japan).
Immunofluorescence staining The indirect immunofluorescence method was used for staining of cumulus-free oocytes from pigs and humans. The details of the staining procedure by fluorescent isothiocyanate (FITC) conjugated F(ab')z fragment of IgG from rabbit anti-mouse y-globulin serum (Cappel Laboratories, Inc.) was previously described (Isojima et al., 1984). Porcine and human oocytes were obtained by puncture of large follicles of each fresh ovary and cumulus cells surrounding the oocytes were removed mechanically.
In vitro fertilization (IVF) of human oocytes Oocytes surrounded with cumulus cells were obtained by puncture of large follicles of human ovaries which were partly removed by surgery in the case of polycystic ovaries. The collected oocytes were incubated in modified Krebs-Ringer bicarbonate medium (mKRB), pH 7.4, containing 7% inactivated human serum for 40-48 h at 37°C in a 5% CO2 incubator and then incubated with antibodies to porcine ZP for 30 min at 37°C. After washing with mKRB medium, these oocytes were transferred to a droplet of sperm suspension (1 × 106/ml) under liquid
190 paraffin. The sperm suspension was prepared by washing ejaculated human spermatozoa twice with mKRB medium containing 0.1% BSA and preincubating in the same medium for 3 h at a concentration of 1-5 × 107/ml. The concentration was adjusted to 1 × 106/ml with mKRB with 7% human serum before use. The inseminated oocytes were incubated for 15 h at 37°C in a 5% CO 2 incubator and the number of spermatozoa bound to or penetrated through the ZP of the oocytes was then counted under a phase-contrast microscope, after washing by repeated pipetting. Human oocytes were treated with conventional antiserum to porcine ZP or ascites from mice which were inoculated i.p. with established hybridomas, after heat inactivation of complement (56°C, 30 min) and extensive dialysis against mKRB medium.
Iodination of solubilized zona proteins Heat-solubilized porcine ZP (10 /~g) was labeled with Na125I (1 mCi) by the chloramine-T method and the 125I-labeled proteins were isolated by Sephadex G-25 column (0.8 × 15 cm). Immunoaffinity chromatography of 125I-labeled zona proteins In order to isolate the antigens corresponding to 3A4-2G1 and 1D5-2B7 Mabs, the 125I-labeled zona proteins (ca. 100 x 106 cpm) were passed through a column (0.5 x 1 cm) of Sepharose 4B gel linked with a Mab (v-globulin of culture supernatant, 2.0 mg/ml of gel) and then the antibody bound fraction was eluted with 0.2 M glycine HC1 buffer (pH 2.5) as described in our previous report (Isojima et al., 1982). The antibody bound fraction was precipitated by the corresponding Mab and goat antiserum to mouse v-globulin as second antibody. The immune precipitates were analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to the method described previously (Isojima et al., 1982). Transfer of proteins from SDS-PA GE to nitrocellulose membrane The heat-solubilized porcine zona proteins (220 /~g/lane) were applied to 10% polyacrylamide slab gels (0.5 × 13 × 0.2 cm) and separated by electrophoresis with constant current of 4 mA for 20 h in 0.025 M Tris-glycine buffer (pH 8.3) containing 0.1% SDS. The proteins separated by SDS-PAGE were transferred to nitrocellulose membrane sheets according to the method described by Towbine et al. (1979) at 220 mA for 3 h in SDS-PAGE buffer containing 20% methanol. After trans-blotting, the antigens specific to Mab were stained on the nitrocellulose membrane by the ELISA method. For staining, the membrane was incubated with Mab (1 : 40 diluted ascites) at 4°C overnight and followed by treatment with peroxidase-labeled goat anti-mouse IgG (peroxidase activity 1072 units/ml by pyrogallol method, IgG/peroxidase = 0.87, 1:500 dilution, Miles) at RT for 2 h. 0.0025% O-dianisidine in 10 mM Tris-HCl (pH 7.4) containing 0.01% H/O 2 was added to the nitrocellulose membrane after washing in order to visualize the reaction products by peroxidase.
191 TABLE 1 Characteristics of three hybridoma-produced monoclonal antibodies to porcine zona pellucida Hybridoma clones
Immunoglobulin classes a
Precipitinline to sol. zona b
Immunofluorescent staining of porcine tissues
3A4-2G1 1D5-2B7 1F2-1B8
IgG1 IgG1 IgM
+ -
+ + +
Oocytes
Testes L i v e r Kidney
Brain
m
a Determined by radial immunodiffusion test using class-specific goat anti-mouse IgM and IgG1 serum. b Heat-solubilized porcine zona proteins (5 mg/ml) were reacted to each monoclonal antibody by Ouchterlony's double immunodiffusion test.
Results Three hybridomas (3A4-2G1, 1D5-2B7, 1F2-1B8) which p r o d u c e d Mabs to the soluble antigens of porcine Z P were established. The characteristics of each M a b are summarized in Table 1. The classes of immunoglobulin secreted by the three hybridomas were IgG1, IgG1 and IgM, respectively. Immunofluorescence study showed that all three Mabs stained only oocytes and no other tissues from pigs. The M a b produced by 1D5-2B7 h y b r i d o m a formed an immunoprecipitin line against heat-solublized porcine zona proteins in Ouchterlony's agar immunodiffusion test. As shown in Fig. 1, a conventional mouse antiserum to porcine Z P formed at least two precipitin lines against the solubilized porcine zona antigens and the precipitin line of 1D5-2B7 M a b was fused with one of those two precipitin lines. Figure 1 also shows that the M a b to porcine Z P from B l l C 8 hybridoma, which was characterized in our previous report (Isojima et al., 1984), formed a precipitin line against the porcine zona antigen, but this precipitin line fused with another one of the two precipitin lines formed by the conventional antiserum. The precipitin lines formed by 1D5-2B7 and B l l C 8 Mabs completely crossed each other. Neither M a b 3A4-2G1 nor 1F2-1B8 m a d e an immunoprecipitin line against the heat-solublized porcine zona proteins.
!
m
Fig. 1. Immunodiffusion of heat-solubilized porcine zona proteins with conventional and monoclonal antibodies to porcine zona antigens. Center well: heat-solubilized porcine zona proteins (Prot. 5 mg/ml). Con. Ab: mouse antiserum to porcine zona antigens (undiluted antiserum). 1D5-2B7 and BllC8: monoclonal antibodies to porcine zona antigens ('t-globulin of each hybridoma ascites: 10 mg/ml).
192 As illustrated by immunofluorescence study (Fig. 2), all three Mabs stained ZPe of porcine and human oocytes but did not stain any oocytes of hamsters, rats and mice. The staining patterns of ZPe in both pigs and humans by the three Mabs were quite similar but the staining intensity by Mab 1F2-1B8 was rather weak compared with the other two, even though the same protein concentration from the culture medium was used for staining. The difference of specificity between the two Mabs 3A4-2G1 and 1D5-2b7 was proved by the analysis of SDS-PAGE patterns of immune precipitates which were formed by each Mab with 125I-labeled porcine zona proteins, even though the staining patterns and staining intensity of ZPe by both Mabs were quite similar. When human oocytes were treated with conventional antiserum to porcine ZP, the binding of human spermatozoa to ZP was completely blocked (Table 2). However, numerous spermatozoa could bind to the ZP of human oocytes which were treated with ascites from mice inoculated with either mouse myeloma cells (P3U1) or hybridomas (3A4-2G1, 1D5-2B7 and 1F2-1B8). Moreover, some spermatozoa could penetrate through the ZP of the treated oocytes. No inhibition of sperm binding and penetration was observed even when the two kinds of ascites (3A4-2G1 and 1D5-2B7) were mixed and used for the treatment of human oocytes. When human oocytes were first treated with ascites (3A4-2G1 or 1D5-2B7) and then treated with goat antiserum to mouse v-globulin as second antibody, the sperm binding and penetration were completely blocked with the light scattering precipitin layer over the ZP of the treated oocytes. The immune precipitin layer was also observed following treatment of human oocytes with conventional antiserum to porcine ZP,
pig
3A4-2GI
human
hamster
rat
mouse
mmn
1D5-2B7
1F2-1B8
I/
m
Fig. 2. FITC staining patterns of oocytes from pigs, humans, hamsters, rats and mice by monoclonal antibodies to porcine zona antigens. Note that the three monoclonalantibodies used stained only porcine and human oocytes.
193 TABLE 2 Effect of monoclonal antibodies to porcine zona pellucida on in vitro fertilization of human oocytes Antibodies to zona pellucida
No. of oocytes examined
No. of spermatozoa bound to zona/oocyte
Zona penetration by spermatozoa
Antiserum a Control b
5 9
0 > 50
+
3A4-2G1 1D5-2B7 1F2-1B8
2 2 2
> 50 > 50 > 50
+ + +
3A4-2G1 + 1D5-2B7 c
3
> 50
+
3A4-2G1 + anti-fg d 1D5-2B7 + anti-Ig 1F2-1B8 + anti-Ig
4 4 3
0 0 > 50
-+
a b c d
Conventional mouse antiserum to porcine zona pellucida. Ascites from a mouse inoculated with P3U1 mouse myeloma cells. Ascites from 3A4-2G1 and 1D5-2B7 hybridomas were mixed at 1 : 1 ratio. After treatment with ascites from 3A4-2G1, the oocytes were further treated with goat antiserum to mouse y-globulin.
but no precipitin layer was observed by treatment only with ascites produced by the hybridoma. Contrary to 3A4-2G1 and 1D5-2B7, the 1F2-1B8 Mab neither blocked sperm binding to human oocytes nor made an immune precipitin layer over the zona after treatment with the second antibody. 125I-labeled porcine zona proteins were fractionated by immunoaffinity chromatography on Sepharose 4B gels treated with Mab (3A4-2G1 or 1D5-2B7) and the binding activity of each fraction with the corresponding Mab was measured. As shown in Table 3, the binding activity of the antibody-bound fraction was much higher than that of the unfractionated original material and the antibody-unbound fraction. The immune precipitates of the antibody-bound fraction with the correTABLE 3 Binding assay of the fractions from the antibody affinity chromatography of 125I-labeled zona proteins 125l-zona proteins
Original material Mab unbound fraction Mab bound fraction
Binding to Mab (%) a 3A4-2G1
1D5-2B7
5.0 0.7 16.3
17.9 7.6 38.2
a For the binding assay, a sample of 100/~1 of 12SI-labeled antigen (ca. 20,000 c.p.m.) and 100/~1 of each Mab (hybridoma culture supernatant) were mixed and kept for 48 h at 4°C. As a second antibody, 100 /L1 of rabbit anti-mouse ~,-globulin (1 : 5 diluted) was added and the mixture was kept for another 24 h at 4°C. The precipitate was separated by centrifugation (1800×g, 10 rain) and counted for the radioactivity by an auto-gamma scintillation counter. The net binding (%) was calculated by substracting the control binding (1 : 2000 diluted normal mouse serum) from that due to Mab.
194
sponding Mab were further fractionated by SDS-PAGE. Figure 3 shows the comparison of SDS-PAGE patterns of the original ]25I-labeled porcine zona proteins and the immune precipitates with Mabs of 3A4-2G1 and 1D5-2B7. Each sample showed a quite different pattern on SDS-PAGE. The main peaks of radioactivity were present in the zone of much higher than 100,000 daltons and around 67,000 daltons in the original 125I-labelled zona proteins, and around 92,000, 65,000 and cpm
4000
9°;00° ,.ooo,;ooo ,o.ooo 2o.,oo
2000
I
I
I
I
I
!
1500
I000
500
i
I
s00 0
i
I
!
I
10
20
30
40
fraction
number
Fig. 3. Radioactivity applied to SDS-PAGE was ca. 100,000 c.p.m, in 125I-labeled zona proteins, ca. 15,000 c.p.m, in the immune precipitate with 3A4-2G1 and ca. 8000 c.p.m, in the i m m u n e precipitate with 1D5-2B7. A dotted Line in the second panel shows a pattern of radioactivities (ca. 2000 c.p.m.) of lESI-zona proteins non-specifically immunoprecipitated with 4C7 (mouse IgG1 Mab established to a cell Line of h u m a n ovarian cancer in our laboratory).
195
23,000 daltons in the immune precipitates with Mab 3A4-2G1. The immune precipitate with Mab 1D5-2B7 showed peaks around 57,000 and 49,000 daltons, though the original ]25I-zona material contained components of these molecular sizes only in a small amount. These results may possibly be due to the concentration of minor components containing immunoprecipitable antigens by two step purifications of the corresponding antigen. Fig. 4 shows the staining patterns of antigens trans-blotted to Staining patterns of antigens transblotted to a nitrocellulose membrane from SDS-PAGE of heat-solubilized porcine zona proteins A. 1 iiiiiiii~iiii~!il!
2
3
B.. 1
2
3
F-l -
I ] <1 94K
"
I <1 67K
i!~!ili!i~iliii~ il~ii!;!!iiiiii!~
~iiiiiiiiiiii~iiiiiil ii!iiiiiiiiiii~iiiiiiii
I <1 43K
i~!!!,!i!!!2!31 i!ii!~!!!ii!~!~iiill
iii :~iili ~ iii!
Fig. 4. A is the original picture and B a schematic picture of the staining patterns. Channel 1 was stained with a conventional antiserum to porcine ZP, channel 2 stained with 3A4-2G1 and channel 3 stained with lD5-2B7.
196 a nitrocellulose membrane from SDS-PAGE of heat-solubilized porcine zona proteins. The staining pattern with the conventional antiserum to porcine ZP was broad and homogeneous with a distribution from large to small molecules. 3A4-2G1 Mab stained antigen molecules restricted to around 92,000 daltons but 1D5-2B7 Mab was unable to stain any molecules trans-blotted to the nitrocellulose membrane.
Discussion
In our previous study (Isojima et al., 1984), we established five hybridomas which produced Mabs to porcine ZP and showed that the Mab BllC8 which was generated to the species specific zona antigen of pigs strongly blocked the binding of boar spermatozoa to porcine oocytes. However, other Mabs which cross-reacted with hamster ZP did not block the binding of hamster spermatozoa to hamster oocytes. When goat antiserum to mouse T-globulin was applied as the second antibody, one (G10G5) of the Mabs which cross-reacted with hamster ZP showed a blocking effect on sperm binding to hamster oocytes, with formation of a light-scattering precipitin layer around the oocytes. From these results, we assumed that the Mab which recognized the species-specific antigen of ZP would block sperm binding to oocytes in the same species by reacting possibly to a sperm receptor antigen on ZP, because it is generally believed that the sperm receptor is species specific (Peterson et al., 1980). On the other hand, Mabs which recognize cross-reactive zona antigens other than pigs might bind to portions different from the sperm receptor of ZP in other species, because they were unable to block sperm binding. However, when antigen epitopes are present at high density on ZP they may possibly be located in proximity to the sperm binding sites of the ZP. Therefore the corresponding Mab could block sperm binding to ZP by forming dense immunoprecipitates with the second antibody .over the surface of the ZP. Similarly, if the antigen epitopes are present close to the sperm receptors of ZP, the sperm binding could be blocked by steric hindrance of the receptors with immunoprecipitates formed by the corresponding Mab and the second antibody, even though their distribution may be sparse. As shown in the present experiments, though the three Mabs (3A4-2G1, 1D5-2B7 and 1F2-1B8) generated to heat-solubilized porcine ZP cross-reacted with porcine and human ZP, they did not show any inhibitory effect on human sperm binding to homologous oocytes. However, when goat antiserum to mouse -/-globulin was added as the second antibody in addition to Mab 3A4-2G1 or 1D5-2B7, sperm binding was completely inhibited with the formation of an immune precipitin layer around the oocytes. Thus, the present experiment not only confirmed our previous results in a hamster experiment but also indicated the presence of at least three antigens (epitopes) in human ZP which were common to porcine ZP. In attempting to make a contraceptive vaccine from porcine ZP, isolation of porcine ZP antigen which is common to human ZP is essential. The establishment of hybridomas which produced Mabs 3A4-2G1 and 1D5-2B7 was very useful because these two Mabs may be very convenient and specific tools for the development of
197
such a contraceptive vaccine. The antigens (epitopes) corresponding to these two Mabs will not be involved in the sperm receptor proteins on human ZP. However, if molecules of porcine ZP including antigens (epitopes) which correspond to these Mabs are isolated, fertilization could possibly be blocked with conventional antibodies produced to the isolated molecules by lattice formation of immune precipitates on human ZP. The established Mabs (3A4-2G1 and 1D5-2BT) were therefore used for isolating the corresponding antigens from the heat-solubilized porcine zona proteins by immunoaffinity chromatography. The antigenic molecule corresponding to 3A4-2G1 seems to be composed of three molecules of different size (92,000, 65,000 and 23,000 daltons) and the antigen (epitope) proved to be present on the molecule of 92,000 daltons by the western blotting procedure after SDS treatment. On the other hand, the antigenic molecule corresponding to 1D5-2B7 seems to be composed of two molecules of different size (51,000 and 49,000 daltons), but it was not clear in which molecules the antigen (epitope) was preserved by western blotting after SDS treatment (see Fig. 4). This might be due to damage of the antigen epitope corresponding to 1D5-2B7 by the process of SDS-PAGE or that the antigen after SDS-PAGE could not be transferred to the membrane under these experimental conditions. The same explanation could apply to the molecule of 65,000 and 23,000 daltons corresponding to Mab 3A4-2G1. All these experiments were carried out by using very small amounts of radiolabeled materials of porcine zona proteins, but we are now trying to isolate the corresponding antigens on a large scale using the same technique. The sperm binding receptor of ZP must be species specific; therefore the blocking of fertilization of Mab to zona antigen could be expected only in the same species from which the ZP was used for immunization. On this assumption, we cannot expect to prevent fertilization in humans by Mabs to porcine ZP. We then have to aim to isolate those antigenic molecules close to the sperm binding receptor as the second choice. It is true that several cross-reacting porcine ZP antigens are distributed over the human ZP; therefore there is a possibility of isolating some common antigenic components densely present on human ZP or present close to the sperm binding receptor of human ZP. There are biochemical and immunological ways of purifying these antigenic components. The immunological procedure was preferred because it can select and pick up antigens common only to porcine and human ZPe. This study could indicate a very promising molecule of 92,000 daltons in the porcine ZP. Recently, Yurewicz et al. (1983) and Sacco et al. (1984) reported a molecule of porcine ZP purified by biochemical procedures. They showed that the molecular weight of the purified component was 58,000 and that this molecule possessed the sperm receptor activity to porcine spermatozoa. However, from porcine ZP, it will be difficult to isolate the antigen (epitope) in the sperm binding receptor of human ZP, but the molecule of 92,000 daltons in the antigen corresponding to Mab 3A4-2G1 seems to be present at high density on human ZP or located close to the sperm binding receptor of human ZP. Thus, there is a possibility that fertilization could be impaired by conventional antibodies to this molecule of 92,000 daltons.
198
References Ahuja, K.K. and Tzartos, S.J. (1981) Investigation of sperm receptors in the hamster zona pellucida by using univalent (Fab) antibodies to hamster ovary. J. Reprod. Fert. 61,257-264. Aitken, R.J., Holme, E., Richardson, D.W. and Hulme, M. (1982) Properties of intact and univalent (Fab) antibodies raised against isolated, solubilized, mouse zonae pellucidae. J. Reprod. Fert. 66, 327-334. Dunbar, B.S. and Raynor, B.D. (1980) Characterization of porcine zona pellucida antigens. Biol. Reprod. 22, 941-954. Dunbar, B.S., Wardrip, N.J. and Hedrick, J.L. (1980) Isolation, physicochemical properties, and macromolecular composition of zona pellucida from porcine oocytes. Biochemistry 19, 356-365. Dunbar, B.S., Lin, C. and Sammonds, D.W. (1981) Identification of the three proteins of porcine and rabbit zonae pellucidae by high resolution two dimensional gel electrophoresis: comparison with serum, follicular fluid, and ovarian cell proteins. Biol. Reprod. 24, 1111-1124. Freye, J.D. and Mettler, L. (1982) Fertility inhibition using low-dose immunization with porcine zonae pellucidae. Am. J. Reprod. Immunol. 2, 153-156. Isojima, S., Koyama, K. and Hasegawa, A. (1981) Production of monoclonal antibody to zona pellucida from porcine oocytes. Acta Obstet. Gynaecol. Jpn. 33, 1995-1996. Isojima, S., Koyama, K. and Fujiwara, N. (1982) Purification of human seminal plasma No. 7 antigen by immunoaffinity chromatography on bound monoclonal antibody. Clin. Exp. ImmunoL 49, 449-456. Isojima, S., Koyama, K., Hasegawa, A., Tsunoda, Y. and Hanada, A. (1984) Monoclonal antibodies to porcine zona pellucida antigens and their inhibitory effects on fertilization. J. Reprod. Immunol. 6, 77-87. Peterson, R.N., Russel, L., Bundman, D. and Freund, M. (1980) Sperm-egg interaction: evidence for boar sperm plasma membrane receptors for porcine zona pellucida. Science 207, 73-74. Sacco, A.G. (1977) Antigenic cross-reactivity between human and pig zona pellucida. Biol. Reprod. 16, 164-174. Sacco, A.G., Yurewicz, E.C., Subramanian, M.G. and De Mayo, F.J. (1981) Zona pellucida composition: species cross-reactivity and contraceptive potential of antiserum to a purified pig zona antigen (PPZA). Biol. Reprod. 25, 439-450. Sacco, A.G., Subramanian, M.G. and Yurewicz, E.C. (1984) Association of sperm receptor activity with a purified pig zona antigen (PPZA). J. Reprod. Immunol. 6, 89-103. Shivers, C.A. and Dunbar, B.S. (1977) Autoantibodies to zona pellucida: A possible cause for infertility in women. Science 197, 1082-1084. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350-4354. Tsunoda, Y., Sugie, T., Mori, J., Isojima, S. and Koyama, K. (1981) Effect of purified zona antibody on fertilization in the mouse. J. Exp. Zool. 271, 103-108. Wood, D.M., Lia, C. and Dunbar, B.S. (1981) Effect of alloimmunization and heteroimmunization with zonae pellucidae on fertility in rabbits. Biol. Reprod. 25, 439-450. Yurewicz, E.C., Sacco, A.G. and Subramanian, M.G. (1983) Isolation and preliminary characterization of a purified pig zona antigen (PPZA) from porcine oocytes. Biol. Reprod. 29, 511-523.