Field of the Invention

This invention relates to human trophoblast membrane expressed protein, in particular to the preparation of this protein in substantially pure form and to the use of this protein or an immunologically- active fragment thereof in an anti-fertility vaccine which will interrupt fertility in female animals, including women.

Introduction.

An immunological approach to interrupting fertility offers several potential advantages viewed in the light of appraisals of the long term safety and acceptability of current contraceptive methods. These advantages include:

(i) the use of a non-pharmacologically active agent; (ii) ease and convenience of administration, suitable for distribution by paramedical personnel; (iϋ) a long-lasting (at least 12 months) but potentially reversible effect; (iv) acceptability of the vaccine principle (of particular importance in developing countries); and (iv) the possible large scale synthesis and manufacture of vaccines at low cost.

SUBSTITUTE SHEET

Selection of a target antigen for immune attack for interrupting fertility must follow stringent criteria. The target antigen must be specific to the reproductive process, that is, it should not share antigenic determinants with normal tissue components in vaccine recipients. Otherwise, the immune response induced by vaccination could elicit damage to normal tissues and/or interfere with normal physiological mechanisms. If shared antigenic determinants are present in the target antigen, unique structures within the macromolecule will have to be identified and synthesised. The present invention meets the above criteria as it is concerned with an expressed target glycoprotein antigen of human trophoblast which is apparently restricted to human trophoblast.

There exist a number of early studies aimed at demonstrating the contraceptive efficacy of some placental antigen targets, including placental hormones, and non-hormonal placental antigens. Secreted glycoproteins produced by the placenta, including the hormones human placental lactogen and human chorionic gonadotrophin (hCG) and the non-hormones βl-pregnancy- specific glycoprotein from cyanomologous monkeys and placental protein 5 from humans, all showed initial promise, but, with the exception of hCG, have proved to be unreliable as target antigens on testing in sub-human primates' ⁸¹ .

Early attempts at demonstrating that placental membrane antigens exerted an anti-fertility effect employed crude placental extracts which acted as an abortifacient, as did administration of antisera to such extracts. These effects, however, were not due to specific immunisation to placental antigens since toxic effects and extra placental tissue damage were seen in the recipients. Several groups have prepared more

SUBSTITUTE SHEET

purified fractions of trophoblast membranes in an attempt to isolate unique protein antigens' ^9,10,11,12 '. The best characterised preparation, TA-1 ⁽⁹⁾ has been tested for anti-fertility effect in baboons and showed encouraging results. TA-1, however, shares antigenic determinants with lymphocytes.

An anti-fertility vaccine formulation based on the C-terminal peptide sequence of the placental hormone, hCG, has been developed ⁽¹³⁾ and tested for its immunological efficacy and safety ⁽¹⁴⁾ . This vaccine may exert its effect by preventing hCG stimulation of the corpus luteum to produce progesterone necessary to maintain a suitable uterine environment for the embryo, or possibly by a direct cytotoxic effect on the hCG- producing cells of the peri-implantation blastocyst.

One advantage of a vaccine based on an antigen of the trophoblast surface membrane is that it is expressed only at one anatomical site following fertilisation and is in intimate contact with maternal blood at a very early stage in pregnancy, thus allowing the maternal immune response direct access to the target antigen ⁽¹⁵⁾ . A second potential advantage of such a vaccine over a hormone-derived formulation is that there is no possible disruption to normal hormone production or function in the recipient. Since the vaccine of the present invention is based on a glycoprotein antigen expressed on the trophoblast membrane surface which is defined by a monoclonal antibody it is, therefore, detected by a single antigenic determinant. This offers a distinct advantage over extracts prepared by conventional purification techniques and defined by polyclonal antisera which contain antibodies reactive to a wide spectrum of proteins.

SUBSTITUTE SHEET

In a first aspect, this invention provides a vaccine capable of stimulating the production of antibodies which prevent or inhibit pregnancy in a female animal, including a woman, which comprises human trophoblast membrane expressed protein or an immunologically-active fragment thereof, as active immunogen, together with one or more pharmaceutically acceptable carriers or diluents. Preferably, the vaccine further comprises an adjuvant, for example an emulsion of muramyl dipeptide (as adjuvant) and squalene-Arlacel A (as vehicle). In addition, the vaccine may comprise a second immunogen, such as diphtheria toxin, in order to enhance the immunogenicity of the active immunogen.

In the anti-fertility vaccine of this invention, the active immunogen may be the entire full length protein or an immunologically-active fragment thereof. Such fragments in particular may be rendered immunogenic either by hapten coupling or by conjugation to a carrier protein of non-human origin. Similarly, the protein or a fragment thereof may be rendered immunogenic by using recombinant DNA technology to synthesise the protein or fragment in the form of a fusion protein, with a heterologous protein or peptide moiety. Suitable moieties for use in such a fusion protein include, for example, β-galactosidase or glutathione-S-transferase ⁽¹⁶⁾ .

In another aspect, the present invention provides a method for the prevention or inhibition of pregnancy in a female animal, including a woman, which comprises administration to said female animal of an antibody- stimulating effective amount of the vaccine as broadly described above.

SUBSTITUTE SHEET

According to another aspect of the present invention, there is provided human trophoblast membrane expressed protein in substantially pure form. By "substantially pure" is meant a form which is at least 70% pure in relation to other proteins, preferably at least 80% pure, and most preferably at least 90% pure.

In yet another aspect, this invention provides antibodies, particularly a monoclonal antibody, to human trophoblast membrane expressed protein.

The invention further extends to a method for the preparation of human trophoblast membrane expressed protein in substantially pure form, which comprises the step of contacting human placental tissue with antibody to the human trophoblast membrane expressed protein, preferably monoclonal antibody thereof, and recovering the desired protein antigen from the antigen/antibody complex.

Preferably, the antibody used is adsorbed or covalently bound to a solid support, such as agarose beads, to assist in separating the antigen/antibody complex prior to recovery of the desired protein antigen.

Following identification of the corresponding gene in a human placental expression library, it is also possible to prepare the human trophoblast membrane expressed protein of this invention by well-known recombinant DNA technology using conventional expression vectors and host cells to achieve expression of the nucleotide sequence coding for this protein, as set out in Table 1. The protein may be produced by this means as the entire full length protein, optionally in the form of a fusion protein as described above. Alternatively, only a part of this nucleotide sequence encoding an immunologically-active fragment may be expressed, and

SUBSTITUTE SHEET

again this fragment may optionally be in the form of a fusion protein. Suitable expression vectors, host cells, and recombinant DNA techniques are disclosed by Sambrook et al. (1989)' ¹⁷ ', the disclosure of which is incorporated herein by reference.

In summary, therefore, the protein antigen of the present invention may be produced either by conventional protein purification means from human placentas, or by using recombinant DNA technology.

Human trophoblast membrane expressed protein in accordance with the present invention is characterised by the following: (i) it is present on the membrane of villous syncytiotrophoblast of human first trimester and term placentae but is not detected on villous cytotrophoblast cells of human placentae; (ii) on SDS-polyacrylamide gel electrophoresis under both reducing and non-reducing conditions it migrates principally as a species of molecular weight of 63-67kDa; and (iii) it is a glycoprotein which is estimated to contain 3-6 0-linked oligosaccharide chains and 5-10 complex N-linked oligosaccharide chains.

The predicted amino acid sequence of this protein, based on the sequence of the corresponding gene identified in a human placental expression library, is set out in Table 1.

Further details of the human trophoblast membrane expressed protein of this invention, and of its preparation in substantially pure form and its use in interrupting fertility are described in the following paragraphs, and in the Examples.

SUBSTITUTE SHEET

1. Description of the Detecting Monoclonal Antibody. The human trophoblast membrane expressed protein of this invention was detected by a murine monoclonal antibody, designated FD0338P, which recognises an epitope on that protein. This antibody is a mouse monoclonal antibody secreted by a hybridoma cell line which grows indefinitely in tissue culture and can be stored frozen in liquid nitrogen. Further details of this monoclonal antibody, and the production thereof, are given in Example 1.

Using this antibody (immunoglobulin heavy chain class Gl, light chain Kappa), it has been established that the human trophoblast membrane expressed protein of this invention resides on the membrane of villous syncytiotrophoblast of human first trimester and term placentae. The epitope was also present on invading, non-villous cytotrophoblast cells of human decidua but at much lower epitope density. It was not detected on villous cytotrophoblast cells of human placentae. The epitope of this membrane expressed protein recognised by the FD0338P antibody was not detected on a wide range of other human tissues and cells.

The cell and tissue distribution of this trophoblast membrane expressed protein antigen is distinct from other trophoblast antigens previously described. Thus the monoclonal antibodies described by Johnson et.al.' ¹ ' which react with syncytiotrophoblast also recognise decidual glands (H315) or lymphocytes (H316). The antibody ND0G1 ⁽²⁾ binds syncytiotrophoblast and cytotrophoblast and recognises a carbohydrate epitope. Trop.l and Trop.2, described by Lipinski

SUBSTITUTE SHEET

δ et.al.' ³ ' bind both syncytiotrophoblast and villous cytotrophoblast. The monoclonal antibody described by Loke and Day ^(<) binds villous cytotrophoblast and endometrial glands' ⁵ '. The pattern of reactivity of FD0338P antibody on human trophoblast and other human tissues was distinct from 45 monoclonal antibodies reacting with human trophoblast components submitted by 12 laboratories to a World Health Organisation- sponsored Workshop' ⁵ '. These monoclonal antibodies included two other monoclonal antibodies, and different from FD0338P, contributed from the laboratory of the present inventors. These two antibodies have been described, viz. FD046B' ⁶ ' which recognises a different antigen on syncytiotrophoblast, and FD0161' ⁷ ' which detects an antigen on both syncytiotrophoblast and non- villous cytotrophoblast.

2. Production of the Trophoblast Membrane Expressed Protein.

The antigen detected by the monoclonal antibody FD0338P has been prepared from human term placentae. The placental tissue is solubilised using a detergent (CHAPS) to release individual membrane proteins. Agarose (Sepharose) beads with FD0338P antibody covalently attached are then added to the mixture. After suitable incubation the beads are recovered and the antigen, which is bound to the monoclonal antibody, is then dissociated from the antibody using an acid solution.

Molecular Weight of the Trophoblast Membrane Expressed Protein.

SUBSTITUTESHEET

On SDS-polyacrylamide gel electrophoresis, the antigen prepared by affinity chromatography (as described above) migrates as a series of molecular weight entities ranging from 30kDa through to the largest, 67kDa. The same bands are seen under both reducing and non-reducing conditions, indicating that the protein does not possess either inter- or intrachain disulphide bonds. The two largest and most abundant species namely, the 63 and 67kDa bands have been isolated to apparent homogeneity by electro-elution from the SDS-polyacrylamide gel.

4. Glycosylation of the Trophoblast Membrane Expressed Protein.

The protein eluted from FD0338P gel was shown to contain carbohydrate residues. The carbohydrate was removed with trifluoromethanesulphonic acid, resulting in reduction of molecular weight of the protein to 30kDa. Assuming that each oligosaccharide chain contributes between 2-4kDa to the apparent molecular weight on SDS-PAGE gels, it is estimated that the FD0338P glycoprotein has between 8-16 oligosaccharide chains. The nature of the oligosaccharide linkage to the polypeptide chain was investigated using endoglycosidases F and H. Endoglycosidase F cleaves both complex and high mannose N-linked (i.e. to asparagine residues), but not O-linked (i.e. serine/threonine residues) oligosaccharides. Cleavage with this enzyme reduced the molecular weights of the two major bands from 63 and 67kDa to 43 and 48kDa respectively. Endoglycosidase H cleaves only high mannose-type N-linked oligosaccharides. No reduction in molecular weight was seen after treatment with this enzyme indicating an absence

SUBSTITUTESHEET

of high mannose chains. Thus the FD0338P glycoprotein appears to have between 3-6 O-linked oligosaccharide chains and 5-10 complex N-linked oligosaccharide chains.

5. Isolation of FD0338P Gene

Using polyclonal antisera raised against the affinity purified FD0338P glycoprotein in sheep, a commercial human placental 32 week λgtll cDNA expression library was probed for the expression of the corresponding antigen using conventional procedures as detailed in Sambrook et. al. (1989)' ¹⁷ '. Isolation of the correct clone was confirmed on Western blots using the expressed fusion protein (i.e. β-galactosidase-FD0338P antigen). The Western blot was probed with monoclonal antibody designated FD078P, raised against affinity-purified FD0338P glycoprotein from human term placentas (see Example 2). The clone was sequenced and the DNA sequence is shown on Table 1, together with the predicted amino acid sequence for the FD0338P protein. Using the sequenced cDNA, and a computer aided data base search (Genbank & EMBL) it was found that this gene had previously been cloned and sequenced by

Rooney et. al . (1988)' ¹⁸ '. Sequence analysis of the cDNA showed that the gene encoded a protein related to the Schwangerschaft protein-1 (SP-1) group of protein, which in turn is a subset of the carcinoembyonic antigens (CEA) and a further subset of the super family of immunoglobulins. An interesting feature of the amino acid sequence of the protein is the presence of an "RGD" (Arg-Gly- Asp) motif. This motif is an integrin receptor consensus sequence. Integrin is a cell receptor protein thought to be intimately involved with binding to extracellular matrices ⁽¹⁹⁾ .

SUBSTITUTE SHEET

Consequently, one possible function of the protein of the invention is the facilitation of blastocyst implantation.

The following Examples describe in detail the production of a monoclonal antibody to the human trophoblast membrane expressed protein of this invention, and the immunogenicity and anti-fertility testing of this protein.

EXAMPLE 1

Production of Monoclonal Antibody a. Syncytiotrophoblast preparation.

First trimester placentas were obtained from elective terminations of apparently healthy pregnancies performed by aspiration at 6-10 weeks' gestation. Clotted blood and any adherent decidua were carefully dissected from the placentas. Syncytiotrophoblast was isolated by gently teasing the placentas through a 250-mesh sieve. The sheets of syncytiotrophoblast, being significantly larger than contaminating cells, readily sediment at unit gravity in Earle's Balanced Salt Solution (Flow Laboratories, Sydney, Australia). After sedimentation for approximately 2 min. the supernatant was decanted and the cells resuspended in fresh solution. This washing procedure was performed three times, then the cells were either used for immunisation in mice or placed into culture. The success of trophoblast isolation was confirmed by the synthesis of human chorionic gonadotrophin in culture after three days incubation. Human chorionic gonadotrophin concentration was measured with a solid phase two- site immunoradiometric assay (Hybritech,

California, USA).

SUBSTITUTE SHEET

Generation of monoclonal antibodies.

Balb/c mice were immunised by peritoneal injection with a wheat germ lectin eluate from a solubilised placental membrane extract. The wheat germ lectin eluate was prepared using first trimester placenta. Placental membrane proteins were solubilised by the addition of an 8mM solution of 3-[(cholamidopropyl)dimethylammonio]1- propanesulphonate (CHAPS) in 20mM Tris-Hcl, pH 8.0 buffer with 0.02% (w/v) sodium azide, 5mM

Ethylenediaminetetra-acetic acid (EDTA) and ImM Phenylmethylsulphonylfluoride (PMSF) in the ratio of 1:10 (w/v). After stirring for 17 hours at 4°C the insoluble material was removed by centrifugation at 2,000g for 15 minutes. Wheat

Germ Agglutinin (WGA) covalently linked to agarose beads (approx. 1 ml of packed beads) was incubated batchwise, with stirring for 17 hours, with the solubilised placental preparation (approx.30 mis). The beads were then recovered by centrifugation at

200 g for five minutes and transferred to a column for washing with fresh homogenisation buffer. The immobilised proteins were eluted from the gel with three volumes of 200 mM N-acetylglucosamine in homogenisation buffer.

The immunogen was given at weekly intervals following a three week delay after the primary injection. After six successive weekly injections the spleen cells from an immunised mouse were fused with mouse myeloma P3x63Ag8.653 cells. The fusion cell mixture was dispensed initially into between two and six 24-well plates and allowed to grow in the presence of RPMI 1640 (Flow Laboratories) medium containing 10% heat- inactivated fetal calf serum (Flow Laboratories), lxl0" ⁷ M hypoxanthine, 4xl0 ^"7 M aminopterin, 1.5xl0" ⁵ M

SUBSTITUTESHEET

thymidine, 100 units/ml penicillin and 100 μg/ml streptomycin at 37°C in a humid atmosphere of 5% C0 ₂ in air for about 10 days. A few cells from each hybrid colony were then picked out and transferred to individual wells of other 24-well plates. To both minimise the probability of identifying carbohydrate epitopes and facilitate the possible use in an immuno-affinity gel, IgG secreting clones were identified using an antigen capture ELISA technique. Briefly, ELISA plates were coated with anti-mouse Ig and the culture supernatant from each hybrid colony added. After adequate incubation time and washing, an IgG class specific enzyme conjugated second antibody was added and developed with the appropriate enzyme substrate. IgM secreting clones were discarded and the culture supernatants containing IgG or A were screened for antibody binding activity on frozen sections of first trimester placenta using an immunoperoxidase technique (see below).

Cultures of interest were then cloned twice by the limiting dilution method and further expanded in culture.

c. Tissue specificity of monoclonal antibodies by immunoperoxidase staining of frozen tissue sections.

Sections of frozen tissue were cut at -20°C using an American Optical Cryostat. 5μm thick sections were air-dried on chrome-alum gelatin- coated slides for 2h. The tissue sections were incubated with cell culture supernatants containing antibodies for eight l-2h at room temperature or overnight at 4°C. Bound antibody was revealed by using biotin-labelled horse antibodies to mouse IgG followed by avidin- biotinylated peroxidase complex (Vector Laboratories, Burlingame, CA;. Diaminobenzidine

SUBSTITUTE SHEET

(DAB) (0.5 mg/ml) with 0.01% hydrogen peroxide in 20mM Imidazole buffer, pH 7.0, was the enzyme substrate. Sections were counterstained with Mayer's Haemalum. Cover slips were then mounted on air dried sections with DePeX.

For tissues containing substantial amounts of endogenous biotin such as liver and kidney, an indirect immunoperoxidase technique was used. An appropriate dilution of peroxidase-labelled goat anti-mouse IgG reduced the high background staining found with the avidin-peroxidase complex technique.

Human tissues were obtained as follows: first trimester placenta and decidua were obtained from elective pregnancy terminations, term placentas within lhr or delivery, and endometrium, myometrium, ovary, cervix, stomach, colon and rectum were obtained at surgery. The remainder were obtained within 6h postmortem. Tissues were frozen in liquid nitrogen, then stored at -70"C until required.

A monoclonal antibody (FD0114G) produced in the laboratory, which reacts with Type IV collagen, acted as a positive control on all tissues.

d. Selection of monoclonal antibody to human trophoblast membrane expressed protein.

The FD0338P monoclonal antibody was selected on the basis of reactivity with trophoblast cells of human first trimester placenta and the absence of reactivity on other normal (non-malignant) tissues.

SUBSTITUTE SHEET

EXAMPLE 2 Immunogenicity and Anti-fertility Testing a. Identification of the FD0338P protein on baboon trophoblast. The murine monoclonal antibody, FD0338P, described in Example 1, which detects the human trophoblast membrane expressed protein did not react with baboon trophoblast. However, a sheep polyclonal antiserum raised to purified FD0338P antigen extracted from human placenta, did react with baboon trophoblast. The baboon analogue had a similar molecular weight to the human protein.

The polyclonal antiserum was raised by immunising sheep with an emulsion composed of

Freund's adjuvant and I6μg of purified FD0338P antigen (purified as described above - Production of the Trophoblast Membrane Expressed Protein). Briefly, for the primary immunisation an emulsion formed between 2 volumes of Freund's Complete

Adjuvant and 1 volume of 3M potassium chloride solution, was prepared by mixing with two syringes interconnected via a narrow tube. Then 1 volume of purified FD0338P protein was blended into the pre-formed emulsion. Sheep were immunised by injecting a total volume of 1ml of the emulsion over 4 sites, subcutaneously. A second and third booster injection were given 21 and 42 days after the primary immunisation. These booster injections were prepared essentially as described above, except that Freund's Incomplete Adjuvant was substituted for the complete adjuvant. Antiserum was collected 21 days after the third injection.

Confirmation of the similarity between the FD0338P protein isolated from human placenta and

SUBSTITUTE SHEET

the corresponding baboon analogue was achieved by immunising mice with purified FD0338P antigen extracted from human placenta as described above. The mouse spleen cells were then fused with myeloma cells and two distinct cell lines were produced with secreted monoclonal antibodies reactive to the antigen. These monoclonal antibodies also reacted with both human and baboon trophoblast, and have been designated FD078P and FD093P. The presence of the described trophoblast membrane expressed protein in baboon placentae meant that trial testing of immunogenicity and efficacy could be initiated in baboons.

Immunogenicity and anti-fertility testing of FD0338P protein in baboons.

In an initial study to demonstrate utility, three baboons were immunised intramuscularly with 16μg of human FD0338P protein. The SDS-denatured protein was mixed with an emulsion of muramyl dipeptide and squalene-Arlacel A. This emulsion has been approved for, and used in, a Phase 1 Trial of the hCG-based anti-fertility vaccine in humans. Injections were given on Day 0, 21, 42 and 140 (2 animals only). Baboons were mated after the second injection and in subsequent cycles until pregnancy resulted.

All three animals responded by raising antibodies (tested by ELISA assay to the immunising antigen), the titres increased after each injection, indicating that the antigen is immunogenic. One animal became pregnant in the first mating cycle, the second in the second cycle while the third was protected until the fourth cycle. The overall fertility rate was 33%, compared to the average colony fertility rate of 70%. This preliminary and limited trial indicates

SUBSTITUTE SHEET

the potential use of the FD0338P protein in an anti-fertility vaccine formulation.

It will be appreciated that several approaches can be taken to improve both the immunogenicity and thus the anti-fertility effect of the FD0338P protein as described above. For example, the antigen used in this initial trial was a denatured preparation which bears little conformational resemblance to the native protein expressed on the surface of the trophoblast membrane. Use of the protein in its native conformational state and/or linked to a second, immunogenic protein, for example diphtheria toxoid, before injection is expected to enhance its effectiveness.

SUBSTITUTE SHEET

TABLE 1

DNA SEQUENCE AND PREDICTED AMINO ACID SEQUENCE

GGG CGG GAC AAC TGG TCT GAG TAC TAT GGC TGA TTT TCG CTG TCT GGC 48

ATT GAG AAG CCA CAC GCC CCT TTT GCT TAG GAG GCC TCT CTG CTG GAG 96

GAT GAC GAT GGC ATG GTT TAT CTA AGG CCA CTG ACA AGT CAT CAA TAT 14

AGG ACA GCA CAG CTG AGA GCC ATG CTC AGG AAG TTT CTG GAT CCT AGG 192

Met Gly Pro 3 CTC AGC TCC ACA GAG GAG AAC ACG CAG GCA GCA GAG ACC ATG GGG CCC 2 0 Leu Ser Ala Pro Pro Cys Thr Gin Arg lie Thr Trp Lys Gly Leu Leu 19 CTC TCA GCC CCT CCC TGC ACA CAG CGC ATC ACC TGG AAG GGG CTC CTG 288 Leu Thr Ala Ser Leu Leu Asn Phe Trp Asn Pro Pro Thr Thr Ala Gin 35 CTC ACA GCA TCA CTT TTA AAC TTC TGG AAC CCG CCT ACC ACT GCC CAA 336 Val Thr He Glu Ala Glu Pro Thr Lys Val Ser Lys Gly Lys Asp Val 51 GTC ACG ATT GAA GCC GAG CCA ACC AAA GTT TCC AAG GGG AAG GAC GTT 384 Leu Leu Leu Val His Asn Leu Pro Gin Asn Leu Ala Gly Tyr lie Trp 67 CTT CTA CTT GTC CAC AAT TTG CCC CAG AAT CTT GCT GGC TAC ATC TGG 432 Tyr Lys Gly Gin Met Lys Asp Leu Tyr His Tyr He Thr Ser Tyr Val 83 TAC AAA GGG CAA ATG AAG GAC CTC TAC CAT TAC ATT ACA TCA TAC GTA 80 Val Asp Gly Gin He He He Tyr Gly Pro Ala Tyr Ser Gly Arg Glu 99 GTA GAT GGT CAA ATA ATT ATA TAT GGG CCT GCA TAC AGT GGA CGA GAA 528 Thr Val Tyr Ser Asn Ala Ser Leu Leu He Gin Asn Val Thr Arg Glu 115 ACA GTA TAT TCC AAT GCA TCC CTG CTG ATC CAG AAT GTC ACC CGG GAG 576 Asp Ala Gly Ser Tyr Thr Leu His He Val Lys Arg Gly Asp Gly Thr 131 GAC GCA GGA TCC TAC ACC TTA CAC ATC GTA AAG CGA GGT GAT GGG ACT 624 Arg Gly Glu Thr Gly His Phe Thr Phe Thr Leu Tyr Leu Glu Thr Pro 147 AGA GGA GAA ACT GGA CAT TTC ACC TTC ACC TTA TAC CTG GAG ACT CCC 672 Lys Pro Ser He Ser Ser Ser Asn Leu Tyr Pro Arg Glu Asp Met Glu 163 AAG CCC TCC ATC TCC AGC AGC AAC TTA TAC CCC AGG GAG GAC ATG GAG 720 Ala Val Ser Leu Thr Cys Asp Pro Glu Thr Pro Asp Ala Ser Tyr Leu 179 GCT GTG AGC TTA ACC TGT GAT CCT GAG ACT CCG GAC GCA AGC TAC CTG 768 Trp Trp Met Asn Gly Gin Ser Leu Pro Met Thr His Ser Lc-u Gin Leu 195 TGG TGG ATG AAT GGT CAG AGC CTC CCT ATG ACT CAC AGC TIG CAG TTG 816

^S UBSTITUTE _S HEET

TABLE 1 (CONT. )

196 Ser Lys Asn Lys Arg Thr Leu Phe Leu Phe Gly Val Thr Lys Tyr Thr 211

817 TCC AAA AAC AAA AGG ACC CTC TTT CTA TTT GGT GTC ACA AAG TAC ACT 864

212 Ala Gly Pro Tyr Glu Cys Glu He Arg Asn Pro Val Ser Ala Ser Arg 227

865 GCA GGA CCC TAT GAA TGT GAA ATA CGG AAC CCA GTG AGT GCC AGC CGC 912

228 Ser Asp Pro Val Thr Leu Asn Leu Leu Pro Lys Leu Pro Lys Pro Tyr 243

913 AGT GAC CCA GTC ACC CTG AAT CTC CTC CCG AAG CTG CCC AAG CCC TAC 960

244 He Thr He Asn Asn Leu Asn Pro Arg Glu Asn Lys Asp Val Leu Ala 259

961 ATC ACC ATC AAC AAC TTA AAC CCC AGG GAG AAT AAG GAT GTC TTA GCC 1008

260 Phe Thr Cys Glu Pro Lys Ser Glu Asn Tyr Thr Tyr He Trp Trp Leu 275

1009 TTC ACC TGT GAA CCT AAG AGT GAG AAC TAC ACC TAC ATT TGG TGG CTA 1056

276 Asn Gly Gin Ser Leu Pro Val Ser Pro Arg Val Lys Arg Pro He Glu 291

1057 AAT GGT CAG AGC CTC CCG GTC AGT CCC AGG GTA AAG CGA CCC ATT GAA 1104

292 Asn Arg He Leu He Leu Pro Ser Val Thr Arg Asn Glu Thr Gly Pro 307

1105 AAC AGG ATC CTC ATT CTA CCC AGT GTC ACG AGA AAT GAA ACA GGA CCC 1152

308 Tyr Gin Cys Glu He Gin Asp Arg Tyr Gly Gly He Arg Ser Tyr Pro 323

1153 TAT CAA TGT GAA ATA CAG GAC CGA TAT GGT GGC ATC CGC AGT TAC CCA 1200

324 Val Thr Leu Asn Val Leu Tyr Gly Pro Asp Leu Pro Arg He Tyr Pro 339

1201 GTC ACC CTG AAT GTC CTC TAT GGT CCA GAC CTC CCC AGA ATT TAC CCT 12 8

340 Ser Phe Thr Tyr Tyr His Ser Gly Glu Asn Leu Tyr Leu Ser Cys Phe 355

1249 TCA TTC ACC TAT TAC CAT TCA GGA GAA AAC CTC TAC TTG TCC TGC TTC 1296

356 Ala Asp Ser Asn Pro Pro Ala Glu Tyr Ser Trp Thr He Asn Gly Lys 371

1297 GCG GAC TCT AAC CCA CCA GCA GAA TAT TCT TGG ACA ATT AAT GGG AAG 1344

372 Phe Gin Leu Ser Gly Gin Lys Leu Phe He Pro Gin He Thr Thr Lys 387

13 5 TTT CAG CTA TCA GGA CAA AAG CTC TTT ATC CCC CAG ATT ACT ACA AAG 1392

388 His Ser Gly Leu Tyr Ala Cys Ser Val Arg Asn Ser Ala Thr Gly Met 403

1393 CAT AGC GGG CTC TAT GCT TGC TCT GTT CGT AAC TCA GCC ACT GGC ATG 1440

404 Glu Ser Ser Lys Ser Met Thr Val Lys Val Ser Ala Pro Ser Gly Thr 419

1441 GAA AGC TCC AAA TCC ATG ACA GTC AAA GTC TCT GCT CCT TCA GGA ACA 1488

420 Gly His Leu Pro Gly Leu Asn Pro Leu ***

1489 GGA CAT CTT CCT GGC CTT AAT CCA TTA TAG CAG CCG TGA TGT CAT TTC 1536

1537 TGT ATT TCA GGA AGA CTG GCA GAC AGT TGC TTT CAT TCT TCC TCA AAG 1584

1585 TAT TTA CCA TCA GCT ACA GTC CAA AAT TGC TTT TTG TTC AAG GAG ATT 1632

1633 TAT GAA AAG ACT CTG ACA AGG ACT CTT GAA TAC AAG TTC CTG ATA ACT 1680

SUBSTITUTE SHEET

TABLE 1 ( CONT . )

1681 TCA AGA TCA TAC ATG GAC TAA GAA CTT TCA AAA TTT TAA TGA ACA GGC 1728

1729 TGA TAC TTC ATG AAA TTC AAG ACA AAG AAA AAA ACC CAA TTT TAT TGG 1776

1777 ACT AAA TAG TCA AAA CAA TGT TTT CAT AAT TTT CTA TTT GAA AAT GTG 1824

1825 CTG ATT CTT TGA ATG TTT TAT TCT CCA GAT TTA TGC ACT TTT TTT CTT 1872

1873 CAG CAA TTG GTA AAG TAT ACT TTT GTA AAC AAA AAT TGA AAC ATT TGC 1920

1921 TTT TGC TCC CTA AGT GCC CCA GAA TTG GGA AAC TAT TCA GGA GTA TTC 1968

1969 ATA TGT TTA TGG TAA TAA AGT TAT CTG CAC AAA CCC

SUBST _IT UTE SHEET

REFERENCES:

1. Johnson, P.M. et.al. (1981). Am. J. Reprod. Immunol. 3. :246.

2. Sunderland, CA. et.al. (1981). Immunology 43 :541.

3. Lipinski, M. et.al. (1981). Proc. Natl. Acad. Sci. (USA). 78 : 5147.

4. Loke, Y.W. and Day, S. (1984). Am. J. Reprod. Immunol. 5. : 106.

5. Anderson, D.J. et.al. (1987). J.Reprod. Immunol. 10 :231.

6. Mueller, U.W. et.al. (1986. Immunology 59 :135.

7. Mueller, U.W. et.al. (1987). Histochem. J.. 19 : 288.

8. Jones, W.R. (1982). "Immunological Fertility Regulation", Blackwell, Oxford.

9. Faulk, W.P. et.al. (1987). Immunological Influence on Human Fertility (ed. B.Boettcher). p.153, Academic Press, Sydney.

10. Faulk, W.P. et.al. (1978). Proc. Natl. Acad. Sci. (USA). 75 : 1947.

11. Ogbimi, A.O. et.al. (1979). J. Reprod. Immunol. 1 : 127.

12. Whyte, A. and Loke, Y.W. (1979). Clin. Exp. Immunol. 37 : 359.

13. Stevens, V.C. (1986). Immunology Today 7 : 369.

14. Jones, W.R. et.al. (1988). Lancet 1 : 1295.

15. Ada, G.L. et.al. (1985). Nature 317 : 288.1

16. Smith and Johnson (1988). Gene. 67 : 31.

17. Sambrook et.al. (1989). "Molecular Cloning : A Laboratory Manual." Cold Spring Harbor, New York.

18. Rooney et.al. (1988). Gene. 71 : 439.

19. Ruoslahti and Pierschbacher (1987). Science. 238 : 491.

^S U ^BST ITUTESHEET