antibodies raised against human placenta reacted positively, ... reactivity on frozen sections of human early placenta. Only ... Human placental lactogen.
Human Reproduction vol.13 no.5 pp.1169–1174, 1998
Reactivity of human trophoblast monoclonal antibodies with marmoset monkey trophoblast cultures
Catherine S.Hawes1,3, Angela Petropoulos1, A.Lopata2,B. Kalionis1 and W.R.Jones1 1Department
of Obstetrics and Gynaecology, Flinders University of South Australia, Flinders Medical Centre, Bedford Park, South Australia 5042 and 2Department of Obstetrics and Gynaecology, University of Melbourne and Reproductive Biology Unit, Royal Women’s Hospital, Carlton, Victoria 3053
3To
whom correspondence should be sent at: School of Nursing, Flinders University of South, Australia, GPO Box 2100, Adelaide, South Australia 5001
The aim of this study was to determine whether cultured trophoblast tissues, derived from the trophectoderm of marmoset monkey blastocysts, contain homologues of human trophoblast antigens. This is an essential prerequisite to determine whether the marmoset may be a suitable model for preclinical testing of a human antitrophoblast antigen for fertility regulation. Previously evaluated monoclonal antibodies from the Flinders University laboratory, which reacted with human trophoblast with a high degree of specificity, were tested for immunohistochemical reactivity using an immunoperoxidase detection method on both frozen and paraformaldehyde-fixed sections of the cultured marmoset monkey trophoblast. All monoclonal antibodies raised against human placenta reacted positively, when compared to controls, suggesting that human and marmoset trophoblast cells share common epitopes. The specificity of the monoclonal antibodies was investigated by determining whether there was cross-reactivity with other marmoset monkey tissues, including adrenal, spleen, kidney, liver, muscle, ovary and testis. The specificities of the monoclonal antibodies on these marmoset tissues were similar to those previously found on the corresponding human tissues. We have concluded that marmoset monkey trophoblast exhibits homologues of human trophoblast antigens. The findings also suggest that marmoset monkeys should be evaluated further as a primate model to test suitable target antigens for antitrophoblast vaccines that may be useful contragestation agents in humans. Key words: marmoset/monoclonal antibodies/trophoblast
they are expressed at only one anatomical site following fertilization, and are accessible to the immune system (Jones, 1982; Ada et al., 1985; Ada and Griffin, 1991; Alexander and Bialy, 1994). The Flinders University laboratory has focused on human trophoblast antigens as a possible basis for a fertility-regulating vaccine. An important consideration in the development of such a vaccine is the selection of an appropriate animal model for preclinical safety and efficacy studies. Basic criteria for such a model include its availability, the feasibility of conducting antifertility and immunological studies, and its relevance to human reproduction (Griffin and Hendrickx, 1991). Amongst the infrahuman primates, the marmoset fulfils these criteria and has previously been used in evaluating fertility control by immunizing with the placental hormone, human chorionic gonadotrophin (Hearn, 1976, 1979; Thanavala et al., 1979). To identify specific candidate antigens expressed on the surface of human trophoblast, .8000 mouse monoclonal antibodies (Mab) were generated in the Flinders laboratory using early placental tissue and extracts as immunogens. All Mab generated were screened in the first instance for immunohistochemical reactivity on frozen sections of human early placenta. Only those that reacted with syncytiotrophoblast and/or non-villous cytotrophoblast were selected for further screening. After production of stable clones, the Mab were screened on frozen human somatic tissues. Only those that demonstrated a lack of reactivity to a wide variety of other human tissues were selected as promising candidates. Confirmation of the immunohistochemical screening was performed in a number of laboratories participating in World Health Organization-sponsored workshops. The results of the first and second workshops were published (Anderson et al., 1987; Hsi and Johnson, 1992). The third workshop (held in 1992) data are unpublished. All selected Mab reacted with human placental syncytiotrophoblast, but not with villous cytotrophoblast or mesenchyme. The aim of this study was to determine whether marmoset monkey trophoblast and human trophoblast share common epitopes, thereby allowing us to use the marmoset as a model for preclinical testing of a human antitrophoblast vaccine for fertility regulation.
Introduction
Materials and methods
An immunological approach to fertility control requires the establishment of an immune response to reproductive tract antigen(s). Target antigens must be specific to the reproductive process and, preferably, not be present continuously in vaccine recipients. Antigens on the surface of the early placental trophoblast are attractive targets for attack in the female since
Generation and characterization of trophoblast-reactive monoclonal antibodies The monoclonal antibodies (Mab) used were generated and characterized in the Flinders University laboratory (Table I). In previous publications, the antibodies were assigned a specific number and letter as well as the prefix ‘FDO’ that was common to all the Mab
© European Society for Human Reproduction and Embryology
1169
C.S.Hawes et al.
Table I. Selected monoclonal antibodies (Mab) produced at Flinders University evaluated for their reactivity with marmoset tissues in the present study Immunogen used to generate hybridomas
Mab
Corresponding antigen
Reference
Sieved cell preparations from early human placentas
46B (FDO46B) 2D (FDO2D) 161G (FDO161G)
Unknown Unknown 3β hydroxy-5-ene steroid dehydrogenase Unknown Unknown Human placental lactogen Unknown
Mueller et al. (1986)
Pregnancy specific β-glycoprotein
Unpublished
Solubilized early placental homogenate Wheat germ lectin extract of early placental homogenate
4L (FDO4L) 168L (FDO168L) 202N (FDO202N) 101X (FDO101X) 338P (FDO338P)
Mueller et al. (1987), Nickson et al. (1991) Latham et al. (1996)
Table II. Immunoperoxidase detection of reactivity of monoclonal antibodies on early human trophoblast Monoclonal antibody
Syncytiotrophoblast
Villous cytotrophoblast
Non-villous cytrophoblast
Mesenchyme
46B 202N 338P 161G 101X FDO2D 4L 168L P3X63.Ag8 (negative control)
11 11 111 111 11 11 11 11 –
– – – – – – – – –
– – – 11 – – – – –
– – – – – – – – –
Reactivity grading: 111 5 very strong; 11 5 strongly positive; – 5 negative.
being studied. For simplicity, the common tag ‘FDO’ is omitted from the nomenclature in the present paper (Table I). All of these Mab reacted strongly with syncytiotrophoblast (Table II) of early human placenta in frozen sections (Mueller et al., 1986, 1987; Hsi and Johnson, 1992). None reacted with villous cytotrophoblast or mesenchyme. All were IgG1 class, κ isotype. Preparation of early placental homogenate Early placental tissue was incubated overnight with stirring in 20 mM Tris–HCl buffer, pH 8.0, containing 8 mM 3[(cholamidopropyl)dimethylammonio]1-propane sulphonate (CHAPS; Boehringer Mannheim GmbH, Australia), 0.02% sodium azide, 5 mM EDTA and 1 mM phenylmethylsulphonylfluoride (PMSF) in a ratio of 1:10 (w/v). Insoluble material was removed by centrifugation at 2000 g for 15 min. To generate Mab 101X, Balb/c mice were immunized by i.p. injection of three 0.5 ml volumes over 2 weeks prior to fusion. Preparation of wheat germ lectin extract Early placental tissue (7 g) was homogenized in 100 ml of 20 mm Tris–HCl buffer, pH 8.0, containing 150 mM NaCl, 0.05% sodium azide, 1 mM PMSF and 0.5% NP40. The homogenate was incubated for 2 h at room temperature with 1 ml wheat germ lectin–Sepharose (Pharmacia Biotech Australia). After washing with buffer, the bound material was eluted with 200 mM N-acetyl-D-glucosamine in the same buffer. To generate 338P, the primary immunogen, an emulsion of 0.5 ml of extract in Freund’s complete adjuvant was administered
1170
to Balb/c mice at several sites s.c. Boosters of 0.5 ml extract were given i.p. 3 and 5 weeks later. The fusion was performed at week 6. Marmoset monkey trophoblast culture Uterine embryos were collected from marmoset monkeys according to the surgical procedure described by Summers et al. (1987). Embryos were flushed at 4–5 days gestation and placed into minimum essential medium (MEM) supplemented with 10% fetal calf serum, insulin (25 µg/ml) and transferrin (25 µg/ml) obtained from Sigma, St Louis, MO, USA. Culture conditions were as described previously (Lopata et al., 1995). The primary trophoblast monolayer, which developed from blastocysts attached to a plastic surface consisted of two distinct cell types: a ring of large, bi- and multinucleate peripheral cells surrounding a region of smaller, more abundant, mononucleated cells which formed trophoblastic vesicles of various sizes up to 5 mm diameter (Summers et al., 1987). The mononuclear cells, which resemble cytotrophoblast, formed the wall of the fluid-filled trophoblast vesicles (Lopata et al., 1995). These were cut into 0.5–1.0 mm fragments to re-establish new cultures of monolayer and trophoblastic vesicle complexes. Such cultures were maintained through five or more passages and continued to secrete marmoset chorionic gonadotrophin into the culture medium (Summers et al., 1987; A.Lopata, unpublished results). Immunoperoxidase staining of frozen tissue sections Frozen blocks of marmoset trophoblast (syncytial monolayer and cytotrophoblast vesicles) were prepared by washing the vesicles in
Trophoblast monoclonal antibodies
MEM and immediately embedding into a mould containing O.C.T. Compound (Miles Inc., Elkhart, IN, USA) before freezing at –20°C. In addition, fresh marmoset tissues obtained from Division of Human Nutrition, CSIRO, Adelaide, Australia were snap-frozen in liquid nitrogen within 15 min of collection and stored at –80°C for up to 2 months. Tissues processed in this manner included adrenal, kidney, liver, muscle, spleen and testis. Blocks of frozen tissue or cultured marmoset trophoblast were cryostat-sectioned (American Optical, 5 µm sections) and air-dried on chrome-alum gelatin-coated slides. Tissue sections were covered with cell culture supernatants containing secreted Mab and incubated for either 2 h at room temperature or overnight at 4°C. Bound antibody was detected by using biotinylated sheep anti-mouse IgG (Jackson Immunoresearch Laboratories, Inc., PA, USA), at 25 000fold dilution for 1 h at room temperature followed by streptavidin– horseradish peroxidase conjugate (Silenus, Victoria, Australia, at 400fold dilution for 20 min at room temperature). The second antibody and conjugate were diluted in 20 mM imidazole, 150 mM NaCl buffer, pH 7.0 containing 0.1% bovine serum albumin. The enzyme substrate was diaminobenzidine (1 mg/ml, Sigma) with 0.01% (v/v) hydrogen peroxide in 20 mM imidazole buffer, pH 7.0. Sections were counterstained with Mayer’s haemalum. Endogenous peroxidase activity needed to be blocked in some tissues such as adrenal, kidney, spleen and liver. Tissue sections were blocked with 1% sodium azide 10% hydrogen peroxide in the imidazole buffer used throughout. Sections were incubated for 15 min at room temperature, then washed prior to adding the Mab. The IgG1 MAb P3X63.Ag8, which is secreted by a myeloma cell line (obtained from the Department of Clinical Immunology, Flinders Medical Centre, Australia), served as a negative control. In our extensive experience using this Mab, no reactivity has previously been detected on any human tissues (data not shown). The Mab 114G, produced in the Flinders laboratory, which reacts with human collagen (Hawes et al., 1994), was employed as a positive control. Immunoperoxidase staining of paraformaldehyde-fixed tissue sections Marmoset trophoblast vesicles were rinsed in MEM and immersed in 4% (w/v) paraformaldehyde in phosphate-buffered saline (PBS), pH 7.4, for 1.5 h at room temperature. Freshly obtained pieces of marmoset tissue were immersed in the fixative overnight. The fixed vesicles and tissues were washed in PBS, pH 7.4, and dehydrated through graded alcohol followed by histosol before embedding in wax. Slides bearing sections (5 µm) were dewaxed in xylene and rehydrated in graded alcohols and finally placed in water. Culture supernatants containing Mab were added to the sections and processed as described above.
Results Reactivity of Mab with cultured marmoset trophoblast All Mab were tested on three frozen blocks of cultured trophoblast comprising a spherical vesicle of mononuclear cytotrophoblast cells with some attached multinuclear syncytiotrophoblast (by which means the vesicle adheres to the surface of the culture dish) and a block of a paraformaldehyde-fixed preparation of a trophoblast vesicle. The reactivity is summarized in Table III and representative examples are illustrated in Figure 1. Figure 1A–F shows the cultured trophoblast cells which formed the vesicles, although their spherical shapes have been distorted by the embedding
Table III. Reactivity of monoclonal antibodies with cultured marmoset trophoblast detected by the immunoperoxidase method on frozen and fixed sections Monoclonal antibody
Frozen sections
Paraformaldehyde-fixed
46B 202N 338P 161G 101X 2D 4L 168L P3X63.Ag8 (negative control)
11 11 1 1 1 – – –
1 111 1 – – 11 1 1
–
–
Reactivity grading: 111 5 very strong; 11 5 strongly positive; – 5 negative.
and sectioning process. Sections of trophoblast fixed in paraformaldehyde are shown in Figure 1A–C, while those in Figure 1D–F are sections of frozen preparations. The layers of cells that formed the trophoblastic vesicle can be clearly seen. Positive brown staining of peroxidase reaction product is present in trophoblast in Figure 1B (Mab 2D), C (202N), E (101X) and F (46B) when compared with the negative controls in Figure 1A and D. In the sections of vesicles examined, the reactivity of mono- and multinucleated trophoblast could not be clearly distinguished. It should be noted, however, that the cultured vesicles consisted of an outer layer of trophoblast cells that reacted with the Mab and an inner layer of endoderm cells (Borg and Lopata, 1997) that did not react. As presented in Table III, the epitopes recognized by Mab 161G (3β-hydroxysteroid dehydrogenase) and 101X (Figure 1E), were only reactive on the frozen sections of trophoblast, suggesting that they did not survive the fixation process. The Mab 46B, 202N, and 338P were reactive on both fixed and frozen sections. The remaining Mab, 2D (Figure 1B), 4L and 168L, were only reactive on the fixed preparations (Table III). Reactivity of Mab 161G with marmoset placenta fixed in paraformaldehyde Placentation in the marmoset monkey, as in the human, is haemomonochorial in which only trophoblast and the endothelium of fetal capillaries separate the maternal and fetal blood circulations. In early pregnancy, in both species, syncytiotrophoblast forms the outer layer and cytotrophoblast the inner layer of the placental villi. In the marmoset, establishment of tertiary placental villi occurs around day 40 of pregnancy compared with day 16 or 17 in the human (Hearn, 1986). In both species the relationship between the fetomaternal blood flow is multivillous, although the branching and morphology of the villous tree has a ‘trabecular’ structure in the marmoset and a ‘villous’ structure in the human (Kaufmann, 1996). Despite some of these differences in placentation, Mab 161G reacted strongly with both human (Table II) and marmoset placental trophoblast, as illustrated in Figure 1G and H. Reactivity of Mab on frozen sections of other marmoset tissues All the Mab were tested for reactivity on the sections cut from frozen tissue preparations (spleen, liver, kidney, adrenal and 1171
C.S.Hawes et al.
Figure 1. Reactivity of the human trophoblast monoclonal antibodies (Mab) on marmoset and human tissues. Detection was by the immunoperoxidase method, and sections were counterstained with Mayer’s haemalum. (A–C) Paraformaldehyde-fixed cultured marmoset trophoblast vesicle; A shows the lack of reactivity (blue counterstain only) of the negative control Mab, P3X63Ag8; B and C show the positive brown staining of Mab 2D and 202N respectively. Bars 5 25 µm. (D–F) Frozen preparations of cultured marmoset trophoblast vesicle; D shows the lack of reactivity of the negative control Mab, P3X63Ag8; the positive brown staining of Mab 101X and 46B is shown in E and F respectively. Bars 5 50 µm. (G–I) The strong reactivity of the Mab 161G can clearly be seen on early human placenta (G), marmoset placenta (H) and the interstitial cells of marmoset testis (I). Bars 5 100 µm.
Table IV. Immunoperoxidase detection of reactivity of monoclonal antibodies (Mab) on human and corresponding marmoset tissues Mab
46B 202N 338P 161G 101X 2D 4L 168L P3X63.Ag8a 114Gb
Spleen
Liver
Muscle
Kidney
Adrenal
Testis
H
M
H
M
H
M
H
M
H
M
H
M
– nd – – – – – nd – 1
– – –
– nd – – – – – – – 1
– – – 1 – – – – – 1
– nd – – – – – nd – 1
– – – – – – – – – 1
– – – – – – – – – 1
– – – – – 1 – 1s – 1
– – – 1 – – – – – 1
nd – – 11 – –/1 – – – 1
– – – 1 1/– – 1/– – – 1
– –/1 – 1 – –/1 – – – 11
– – – 1s – 11
H 5 human; M 5 marmoset. aNegative control; bpositive control. – 5 no reactivity; –/1 5 weak reactivity; 1s 5 reactivity on selected cells only; 1 5 positive reactivity; 11 5 strong reactivity; nd 5 not done.
testis) obtained from two animals (Table IV). These results, for most Mab tested, were largely similar to those previously found on comparable human tissues (Table IV). The positive 1172
control Mab 114G reacted with all tissues tested. The Mab 46B, 4L, 338P and 101X were negative on all the frozen tissues, including ovary that is not shown in Table IV. This
Trophoblast monoclonal antibodies
lack of reactivity of somatic marmoset tissues matched previous results obtained on corresponding human tissues. In keeping with the known reactivity of 161G with 3βhydroxysteroid dehydrogenase (Nickson et al., 1991), this Mab reacted strongly with marmoset adrenal cells and the interstitial cells of the testis (see Figure 1I) in frozen sections. In contrast to human tissues, however, reactivity with Mab 161G was also found in the marmoset liver (Table IV). The Mab 202N, which recognizes human placental lactogen, reacted only with marmoset trophoblast and human syncytiotrophoblast. Mab 2D and 168L, which recognize unidentified epitopes on trophoblast, reacted with marmoset kidney tubules and selected spleen cells respectively. These reactivities were not found on corresponding human tissues (Table IV). Discussion The Mab evaluated in this study were previously found to react predominantly with human syncytiotrophoblast, although the Mab 161G also reacted with non-villous cytotrophoblast. All of these antibodies were found to have either limited reactivity, or failed to react, with other human tissues (Mueller et al., 1986, 1987; Hsi and Johnson, 1992). All of the tested Mab also reacted with marmoset trophoblast, suggesting that human and marmoset trophoblast cells share common epitopes. The demonstration of their reactivity on marmoset trophoblast provided valuable confirmation of the identity of the cultured cells and raised prospects for future work. Moreover, the specificities of the Mab on several non-trophoblastic tissues in the marmoset were found to be similar to those previously detected on the corresponding human somatic tissues. Taking into account the reactivity of the Mab on both human and marmoset tissues, the most promising candidate antigens for antifertility vaccines are antibodies 46B, 338P, 101X and 4L. In screening the whole panel of Mab in both species it was found that this subgroup of four antibodies specifically recognized the embryonic tissues, but did not react with somatic tissues in the female adult. Further work needs to be done to identify and characterize the antigens recognized by these antibodies in trophoblast cells. Although 338P is known to recognize pregnancy specific β-glycoprotein, its identity needs further clarification (Flinders laboratory, unpublished) In contrast, since the Mab 2D and 168L showed reactivity, albeit weak, with other adult female marmoset tissues besides trophoblast, the corresponding antigens may not be suitable candidates for antifertility studies in this species. However, such antibodies may be of great value in furthering our knowledge of trophoblast antigens. At this stage of the work we are not able to explain the presence of a weak reaction with marmoset somatic tissues but a lack of reaction with corresponding human tissues for Mab 2D and 168L. Although the trophoblast enzyme 3β-hydroxysteroid dehydrogenase (defined by 161G Mab) is not a candidate antigen for a fertility vaccine due to its presence in other tissues, its possible role in reproduction and related endocrine events could be further investigated in the marmoset model. The reactivity of 161G on cultured marmoset trophoblast found in this study complements earlier work in which the
reactivity of Mab 161G was tested on marmoset embryos collected 9–14 days post conception. Positive staining of both the embryos and the implantation site was found from day 12 onwards, coinciding with the time of implantation and formation of trophoblast (unpublished work in collaboration with P.M.Summers). The enzyme was also located in the marmoset adrenal and the interstitial cells of the testis, corresponding with the known activity of 3β-hydroxysteroid dehydrogenase in these human tissues (Hawes et al., 1994). The Mab 202N and 338P, which recognize human placental lactogen and human pregnancy-specific β-glycoprotein respectively, reacted with cultured marmoset trophoblast. These findings suggest that the marmoset trophoblast contained homologues of such pregnancy specific proteins. We believe that this has not previously been described in the marmoset. A further outcome of our studies, therefore, is the possibility for future investigation of the role of placental lactogen and pregnancy specific β-glycoprotein in the reproductive process by evaluating the effects of selected Mab, described in the current work, during early pregnancy. A considerable amount of work still needs to be done to identify the other unknown antigens recognized by the Mab used in the present studies. The reactivity of 2D, 4L and 168L with fixed, but not the frozen, preparations of marmoset trophoblast, indicates that they may be directed towards soluble antigens which were not retained in the preparation of frozen sections. In future studies the influence of the endocytotic uptake (Bajoria et al., 1997) of Mab, and their complexes with antigens, on the reactivity of trophoblast will also need to be evaluated. The present study has identified epitopes common to both human and cultured marmoset trophoblast. This suggests the feasibility of extending the studies to an examination of the reactivity of the human Mab with marmoset implantation sites, using the Mab that reacted with trophoblast but not with somatic tissues (46B, 338P, 101X and 4L). After the antigens and their corresponding Mab are characterized at various stages of placental development, their potential to interrupt early pregnancy could be assessed by appropriate immunization trials in the marmoset. Pregnancy in the marmoset can be readily diagnosed by hormone assays, thereby allowing an accurate assessment of the efficacy of suitable contragestational treatment. It should be noted that this model would also allow assessment of the effect of antitrophoblast antigen immunization on ovulation, fertilization, the development of preimplantation embryos as well as on implantation. Such studies in the marmoset may ultimately provide a useful primate model for the preclinical testing of the antifertility effects and safety of immunization with human trophoblast antigens. However, further questions regarding the safety, efficacy and acceptability of new vaccines can be answered only after the implementation of carefully designed clinical trials.
Acknowledgements We thank M.Mano, CSIRO Division of Human Nutrition, Adelaide, South Australia for the generous gift of marmoset tissues. This work was supported in part by the World Health Organization and by
1173
C.S.Hawes et al.
Flinders Technologies Pty Ltd. We also thank Angela Nelson for her devoted care of the marmoset monkey facility in Melbourne.
References Ada, G.J. and Griffin, P.D. (1991) The process of reproduction in humans: antigens for vaccine development. In Ada, G.L. and Griffin, P.D. (eds), Vaccines for Fertility Regulation. Cambridge University Press, Cambridge, p. 13. Ada, G.L., Basten, A. and Jones, W.R. (1985) Prospects for developing vaccines to control fertility. Nature, 317, 288–289. Alexander, N.J. and Bialy, G. (1994) Contraceptive vaccine development. Reprod. Fertil. Dev., 6, 273–280. Anderson, D.J., Johnson, P.M., Alexander, N.J. et al. (1987) Report of WHOsponsored workshops on monoclonal antibodies to human sperm and trophoblast antigens. J. Reprod. Immunol., 10, 231–257. Bajoria, R., Sooranna, S.R. and Contractor, S.F. (1997) Endocytotic uptake of small unilamellar liposomes by human trophoblast cells in culture. Hum. Reprod., 12, 1343–1348. Borg, J. and Lopata, A. (1997) Characteristics of trophoblastic vesicles derived from marmoset monkey blastocysts (Abstract O-58). In Proceedings of the XVI Annual Scientific Meeting, Fertility Society of Australia, Adelaide, December 2–4, 1997, p. 73. Griffin, P.D. and Hendrickx, A.G. (1991) Animal models for assessing the safety and efficacy of antifertility vaccines. In Ada, G.L. and Griffin, P.D. (eds), Vaccines for Fertility Regulation. Cambridge University Press, Cambridge, p. 27. Hawes, C.S., McBride, M.W., Petropoulos, A. et al. (1994) Epitopic heterogeneity of human 3β-hydroxysteroid dehydrogenase in villous and extra-villous trophoblast. J. Mol. Endocrinol., 12, 273–281. Hearn, J.P. (1976) Immunization against pregnancy. Proc. R. Soc. Lond., B195, 149–160. Hearn, J.P. (1979) Immunological interference with the maternal recognition of pregnancy in primates. In Maternal Recognition of Pregnancy (Ciba Foundation Symposium 62). Excerpta Medica, Amsterdam, p. 353. Hearn, J.P. (1986) The embryo–maternal dialogue during early pregnancy in primates. J. Reprod. Fertil., 76, 809–819. Hsi, B.L. and Johnson, P.M. (1992) Report of WHO Workshop on monoclonal antibodies to trophoblast. Trophoblast Res., 6, 125–134. Jones, W.R. (1982) Immunological Fertility Regulation. Blackwell, Oxford. Kaufmann, P. (1996) Placental types. In Benirschke, K. and Kaufmann, P. (eds), Pathology of the Human Placenta. Springer-Verlag, New York, pp. 1–13. Latham, S.E., Suskin, H.A., Petropoulos, A. et al. (1996) A monoclonal antibody to human placental lactogen hormone facilitates isolation of fetal cells from maternal blood in a model system. Prenat. Diagn., 16, 813–821. Lopata, A., Berka, J., Simula, A.P. et al. (1995) Differential distribution of mRNA for the alpha and beta subunits of chorionic gonadotrophin in the implantation stage blastocyst of the marmoset monkey. Placenta, 16, 335–346. Mueller, U.W., Hawes, C.S. and Jones, W.R. (1986) Cell surface antigens of human trophoblast: definition of an apparently unique system with a monoclonal antibody. Immunology, 59, 135–138. Mueller, U.W., Hawes, C.S. and Jones, W.R. (1987) Identification of extravillous trophoblast cells in human decidua using an apparently unique murine monoclonal antibody to trophoblast. Histochem. J., 19, 288–296. Nickson, D.A., McBride, M.W., Zeinali, S. et al. (1991) Molecular cloning and expression of trophoblast antigen FDO161G and its identification as 3β hydroxy-5-ene steroid dehydrogenase. J. Reprod. Fertil., 93, 149–156. Summers, P.M., Taylor, C.T. and Hearn, J.P. (1987) Characteristics of trophoblastic tissue derived from in vitro culture of preimplantation embryos of the common marmoset monkey. Placenta, 8, 411–422. Thanavala, Y.M., Hearn, J.P., Hay, F.C. and Hulme, M. (1979) Characterisation of the immunological response in marmoset monkeys immunised against hCG β-subunit and its relationship with their subsequent fertility. J. Reprod. Immunol., 1, 263–273. Received on July 21, 1997; accepted on February 14, 1998
1174
