CA1306429C - Monoclonal antibodies specific for equine chorionic gonadotrophin - Google Patents

Abstract

ABSTRACT

A monoclonal antibody against equine chorionic gonadotrophin produced from a continuous cell line which produces a monoclonal antibody against equine chorionic gonadogrophin, including a hybridoma formed by fusing a B
cell capable of producing antibodies against equine chorionic gonadotrophin with a myeloma cell and its use in purifying equine chorionic gonadotrophin and modifying the biological activity of equine chorionic gonadotrophin.

Description

The present invention relates to monoclonal antibodies, a method for the preparation thereof and use thereof.
Equine chorionic gonadotrophin or pregnant mare serum gonadotrophin was first described by Cole and Hart [Anat. Rec., 56, 275-293 (1930)] as anterior hypophyseal sex maturing hormone due to its potent stimulatory effect on the sexual maturity of the immature rat. Its presence was noted in the sera of pregnant mares from appro~imately day 37 till day 100 of pregnancy.
More recently PMSG has been referred to as equine chorionic gonadotrophin (eCG), a name chosen to reflect the source of the hormone and its close homology with human chorionic gonadotrophin (hCG).
eCG has a potent, dual LH-FSH bioactivity capable of inducing follicular growth and ovulation when injected into a wide range of domestic animals. The hormone has a particularly long half life, 21.2 hours in sheep, 50-120 hours in cattle and 26 hours in rats which is in part responsible or the high biological potency of the hormone in these species. This potent dual LH FSH bioactivity of eCG in non-equine species has led to descriptions of the hormone such as "very nearly the complete gonadotrophin". Single administrations of the hormone are capable of inducing follicular growth, oestrogen production, ovulation, luteinization and subsequent progesterone synthesis in sheep, cattle, goats, pigs, dogs, cats, rats and mice.
eCG is e2tensively used in veterinary practice for superovulation in embryo transfer programs, induction of ovulation and oestrus in anoestrus animals and for mild ovulation increases to give greater fecundity. However, variation in the response of treated animals is a major problem and may be attributed to batch variation and contaminants present in the commercial eC~ preparations that are currently available. Seasonal effects including follicular/ovarian status of the treated animals also has an effect. A further problem e~ists with superovulatory doses where the contaminating substances and the higher dose of eCG
causes continued stimulation of the ovaries and the production of persistent secondary follicles. These follicles often become large and cystic, causing excessiv~
oestrogen production and eventually contractile scar tissue in the ovaries. Excessive oestrogen can have deleterious effects on egg, sperm and embryo transfer within the reproductive tract of rabbits, rats, mice, guinea pigs, and domestic farm animals.
Particular difficulties are further encountered with respect to the prior art relating to methods of production and purification of eCG.
Two major extraction methods are used for the production of crude extracts of eCG from pregnant mare sera; (i) acid and ethanol precipita-tion, and (ii) acetone precipitation. Depending on the method used, unwanted serum proteins are precipitated by the addition of ethanol or acetone to 50% ~v/v). The further addition of ethanol or acetone to 70-75% then precipitates the active component. The techni~ue of Gospodarowicz and Papkoff, [Endocrinology, 80, 699-702 (1967)] produces crude eCG extracts of approximately 600iu/mg of protein while the Cartland and Nelson [J.Biol. Chem. 119, 59-67(1937)] method results in a crude extract containing 50-lOOiu/mg of protein.
Further purification steps using the crude serum extracts include gel filtration and sulphoethyl-Sephadex ion exchange chromatography (Gospodarowicz and Papkoff, 1967), and gel filtration and hydroxylapatite chromatography Moore and Ward, [J.Biol.
Chem., 255, 6923-6929, 1980]. Others have used wheat germ lectin Sepharose chromatography. The purified products obtained range in potency from 9,000-15,800iu/mg, however estimations of specific activity are often unreliable as the measurements are based on standards of variable quality. Commercial preparations range from 325 iu/mg to 2500 iu/mg or approximately 5-20%
purity.
According to an aspect of the invention, a process for producing a monoclonal antibody against ~3~

equine chorionic gonadotrophin (ECG) which method comprises providing a B c~ll capable of producing antibodies against equine chorionic gonadotrophin, and a myeloma cell, fusing the B cell with the myeloma cell, detecting antibody producing hybridomas, which have specificity for biologically active eCG wherein said antibody is substantially unreactive with tissue isolated from human or equine pituitaries and in active subunits and fragments of eCG, utilising an immunohistology assay on Western immunoblotting technique, subsequently cloning the fused cell hybrid selected into individual antibody producing cell lines, propagating each cell line in-vivo or _-vitro and harvesting an antibody produced by said hybridomaO
According to another aspect of the invention, a hybridoma cell line which produces a monoclonal antibody against equine chorionic gonadotrophin, comprises a hybridoma for.~ed by fusing a myeloma cell with a B cell capable of producing antibodies against e~uine chorionic gonadotrophin wherein said antibodies are specific for biologically active eCG and are substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG.
According to another aspect of the invention, a hybridoma cell line, comprises hybridoma cell line F51-99 having an accession number given by the European Collection of Animal Cell Cultures, an international depository authority, of 87100204.
According to another aspect of the invention, a monoclonal antibody against equine chorionic gonadotrophin produced from a hybridoma cell line, said monoclonal antibody comprises an antibody which is specific for biologically active eCG and substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG, ~, .. .

and wherein the cell line is a hybridoma formed by fusing a myeloma cell with a B cell capable of producing antibodies against biologically active equine chorionic gonadot~ophin which are substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG.
According to another aspect of the invention, a method of purifying equine chorionic yonadotrophin (ecg) wherein said method comprises:
providing a source of impure eCG and a monoclonal antibody against equine chorionic gonadotrophin formed by a hybridoma cell line, wherein said antibody is attached to a solid support to form an antibody matrix, wherein said monoclonal antibody i5 specific for biologically active eCG and substantially unreactive with tissues isolated from human or equine pituitaries and inactive subunits and fragments of eCG;
passing the impure eCG through the antibody matrix to capture the eCG; and eluting the captured eCG from the antibody matrix.
The B cell capable of producing antibodies against eCG may be selected from spleen cells and lymph node cells of a test animal. The test animal may be a mouse, rat or the like. The B cells may be obtained from an animal immunized with eCG or an immunogenic fragmen-t thereoE. A BALB/c mouse is the preferred animal for immunization. Once-primed or hyper-immunized animals may be used as the source of antibody producing lymphocytes or splenocytes. Mouse and rat splenocytes give a higher percentage of stable fusions with a mouse myeloma line described below. However, the use o~ rat, human and other mammalian cells as well as frog cells is also contemplated. In a particularly preferred form, hyperimmunised mouse spleen cells are used to make the fused cell hybrids.

4b The myeloma cell may be of any suitable type.
Specialized myeloma cell lines which have been developed from lymphocyt~ tumours for use in hybridoma-producing fusion procedures are known in the prior art. The cell lines have been developed for at least three reasons.
The first is to facilitate the selection of fused hybridomas among unfused and similarly indefinitely self-propagating myeloma cells. Usually, this is accomplished by using myelomas with enzyme deficiencies that render them incapable of growing in certain selective media that support the growth of hybridomas.
The second reason arises from the inherent ability of lymphocyte tumour cells to produce their own antibodies.
The purpose of using monoclonal techniques is to obtain immortal fused hybrid cell lines that produce the desired single specific antibody genetically directed by the somatic cell component of the hybridoma. To eliminate the production of tumour cell antibodies by the hybridomas, myeloma cell lines incapable o~ producing light or heavy immunoglobulin chains or those efficient in antibody secretion mechanisms are used. A third reason for selection of cell lines is for their suitability and efficiency for fusion.
Se~eral m~eloma cell lines may be used for the production of fused cell hybrids, including P3/X63-Ag 8, NS-I
type myeloma cells, e.g. P3/NSI/l-Ag 4-1, Sp2/0-Agl4 and S194/5.XXO.BU.l. The P3/X63-Ag ~ and P3/NSI/l-Ag 4-1 cells lines have been described by Kohler amd Milstein [Europ. J.
Immunol. 6: 511-519 (1976)]. Shulman et al. [Nature 276:
269-270 (1978)~ developed the Sp2/0-Agl4 rnyeloma line. The S194/5.XXO.BU.1 myeloma line was reported in an article by Trowbridge [J. Exp. Med. 148: 313 (1979)~. In the example of the present invention, a NS-l-type myeloma cell (derived from BALB/c mice) is the preferred cell line.
The fusion step according to this aspect of the present invention for generating hybrids of antibody-producing spleen or lymph node cells and myeloma cells includes mi~ing somatic cells with myeloma cells. The proportion may vary from approximately 20:1 to 1:1, preferably 10:1, respectively. The fusion step may be carried out in the presence of an agent or agents that promote the fusion of cell membranes. It is preferred that the same species of animal serve as the source of the somatic and myeloma cells used in the fusion procedure. The fusion-promoting agents used may include Sendai virus or polyethylene glycol (PEG).
Because fusion procedures produce viable hybrids at very low frequency (e.g., when spleens are used as a source of somatic cells, only one hybrid is obtained for roughly every 2 ~ 105 spleen cells), it is preferred to have a means of selecting the fused cell hybrids from the remaining unfused cells, particularly the unfused myeloma cells~ A
means of detecting the desired antibody-producing hybridomas among other resulting fused cell hybrids is also necessary.
Generally, the selection of fused cell hybrids is accomplished by culturing the cells in media that support the growth of hybridomas but prevent the growth of the myeloma cells which normally would go on dividing indefinitely.
These cells are selected against in hypoxanthine/aminopterin/
thymidine ~HAT) medium, a medium in which the fused cell ,~ , ;~ - 5 -~96~

hybrids survive due to the HPRT-positive genotype of the spleen cells. The use of myeloma cells with different genetic deficiencies (e.g., other enzyme deficiencies, drug sensitivities, etc.) that can be selected against in media supporting the growth of genotypically comp~tent hybrids is also possible.
Several weeks may be required to selectively culture the fused cell hybrids. Early in this time period, it is preferred to identify those hybrids which produce the desired antibody so that they may be subsequently cloned and propagated. Generally, around 10% of the hybrids obtained produce the desired antibody, although a range of from l to 30% is not uncommon.
The use of highly discriminatory screening assays durinq the cloning and selection of hybridoma cell lines assists in selecting for monoclonal antibodies having the desired specificity.
The detection of antibody-producing hybrids can be achieved by any one of several standard assay methods, including enzyme-linked immunoassay and radioimmunoassay techniques. Immunohistology can also be used for antibody detection and selection. The detection method preferred i5 an enzyme-linked immunoassay employing a peroxidase-conjugated anti-mouse immunoglobulin.
Once the desired fused cell hybrids have been selected and cloned into individual antibody-producing cell lines, each cell line may bP propagated in either of two standard ways. A sample of the hybridoma can be injected into a histocompatible animal of the type that was used to provide the somatic and myeloma cells for the original fusion. The injected animal develops tumours secreting the specific monoclonal antibody produced by the fused cell hybrid. The body fluids of the animal, such as serum or ascites fluid, can be tapped to provide monoclonal antibodies in high concentration. Alternatively, the individual cell lines may be propagated in vitro in laboratory culture vessels; the culture medium, also containing high concentrations of a single specific monoclonal antibody, can be harvested by decantation, filtration or centrifugation.
Accordingly, in a further aspect of the present ~3~

invention there is provided continuous cell lines which produce monoclonal antibodies against equine chorionic gonadotrophin including a h~bridoma formed by fusing a B cell capable of producing antibodies against equine chorionic gonadotrophin with a myeloma cell.
Preferably, the hybridoma cell line is formed by fusing an NS-l myeloma cell with a spleen B cell obtained from a BALB/c mouse immunised with equine chorionic gonadotrophin.
Hybridomas of interest may be dilution cloned to ensure the selection of a single cell line. Hybridomas may be dilution cloned at least four times. Particular examples of hybridoma cell lines are those designated F50-37, F50-75 and F51-99 as hereinafter described samples of which are maintained in the Bunge (Australia) Pty. Ltd. Cell Collection at 89 Flemington Road, North Melbourne, Australia. The inv~ntion described herein is not limited in scope by the cell lines exemplified since these embodiments are intended as illustrations of one aspect of the invention and any equivalent cell lines which produce a functionally equivalent monoclonal antibody are within the scope of the invention.
Tha hybridomas described above are a result of a fusion between a mouse splenocyte and a myeloma cell line classified as NS-l. The lymphocyte was from a spleen obtained from an inbred mouse representative of the BALB~c strain. The mouse had been immunised with eCG of high purity combined with an adjuvant. The hybridoma cell line will grow in tissue culture using a nutrient solution. The cell line when analysed for purity after repeated tissue culture passage by dilution cloning will exhibit antibody producing eCG specific clones at a frequency of 100~.
In a still further aspect of the present invention there is provided monoclonal antibodies against e~uine chorionic gonadotrophin.
The monoclonal antibodies may be produced from a hybridoma cell line as described above. Specific examples of monoclonal antibodies include those formed by hybridoma cell lines F50-37, F50-75 and F51-99, samples of which are maintained in the Cell Collection facility of Bunge (Australia) Pty. Ltd. at B9 Flemington Road, North Melbourne, ~3~

Australia.
Samples o the following hybridoma cell lines have been deposited with the European Collection of Animal Cell Cultures (ECACC).
Culture eposit No.

F51-99 8710020~
Some Physicochemical and Bioloaical Properties of three sPecific Monoclonal Antibodies to eCG and the Cell Lines which produce them (i~ RadioimmunoPreciRitation of eCG
The three putative monoclonal antibodies to eC~ gave positive binding reactions with 125I eCG (Figure 1).
Monoclonal antibodies F50-37 and F51-99 showed considerably stronger binding than F50-75 at lower dilutions. Monoclonal antibody F50-37 gave a classic antibody dilution curve tapering to background levels in eight doubling dilutions, while F51-99 gave a dilution curve which failed to return to baseline levels over the dilution range. Both monoclonal antibodies F50-37 and F51-99 had similar ma~imum binding capacities at a dilution of 1:4, and no detectable binding when neat. This prozone effect when neat was notably absent with monoclonal antibody F50-75.
(ii) Isotypinq All three monoclonal antibodies are of the IgG
class, having kappa light chains. Monoclonal antibody F50 37 is IgG2a while F50-75 and F51-99 are both IgGl. Media containing no mouse immunoglobulin and normal mouse serum gave reliable results in the assay used.
(iii) Scatchard Plot analy~Lis The antibody dilution curves showed monoclonal antibodies F50-37 and F51-99 to be more potent binders of 5I eCG than monoclonal antibody F50-75 at lower dilutions (Figure 2). Monoclonal antibodies F50-37 and F51-99 also had similar maximal capacities and showed a prozone effect similar to that seen in the earlier radio-immunoprecipitation experiment. The prozone effect was due to saturation of the second antibody. The antibody dilution curve for monoclonal antibody F51-99 over the e~panded dilution range returned to background levels. Monoclonal antibody F50 75 showed no prozone effect and r~quired higher concentrations of supernatant to bind the equivalent ma2imal counts of the other two supernatants. The control supernatant was negative throughout.
By convention, th~ selection of the antibody concentration from antibody dilution curves for use in the development of standard curves is that concentration capable of binding 50% of maximal counts, or, the concentration associated with the steepest part of the curve. These two criteria may coincide and are selected to give range and ma2imal sensitivity to the assay developed.
The following monoclonal antibody supernatant concentrations were chosen from the antibody dilution curves for standard curve construction.
F50-37 0.078 x conc.
F50-75 2.185 ~ conc.
F51-99 0.078 x conc.
These supernatant concentrations gave similar ma~imum binding capacities and reflected suitably steep regions of each antibody dilution curve.
E~amination of the standard curves for the three eCG
monoclonal antibodies (Figures 3, 4 and 5~ reveal obvious differences in shape and position. Each point shown on the standard curves is the mean of a duplicate set of tubes for one e~periment and the line plotted between them is the mean result of 2 similar e~periments.
For affinity comparison, this data is presented as a Scatchard plot (Figure 6). The slope of each line is the measure of its a f f inity. Monoclonal ankibody F50-37 has the highest affinity, while F50-75 has the lowest affinity.
Computsr analysis of the data found the affinities for each monoclonal antibody to be, F50-37 1.92 ~ 10 8 mol/l F50-75 1.84 x 10 mol/l F51-99 4.27 ~ 10 -8mol/1 The x intercept of the line in the Scatchard plot gives the binder concentration, or, the concentration of antibody binding sites. These are by computer analysis, _ 9 _ F50-37 7.5 nmol~l F50-75 52.0 nmol~l F51-99 14.9 nmol/l As antibody supernatant concentrations were salected to give similar 125I eCG ma~imal binding capacities for all three antibodies, it would be e~pected that higher affinity antibodies would require less total antibody to bind a given number of 125I eCG counts when the reactions are a~
equilibrium. That is, affinity and antibody binding site concentration should be inversely related for each antibody.
Comparing affinities and binding site concentrations for the three monoclonal antibodies shows this to be true.
Monoclonal antibody F50-37 with the highest affinity has the lowest concentration of binding sites and monoclonal antibody F50-75 with the lowest affinity has the highest concentration of binding sites.
By extrapolatin of binding site concentrations determined through Scatchard analysis, supernatant antibody concentrations can be determined.
Monoclonal antibody F50-37 diluted to 3.078~ and having 7.49 nmol/l at this dilution would therefore have an original supernatant binding site, or antibody concentration, of 7.49 0.078 nmol/l or 14.4 ug/ml By Scatchard analysis and extrapolation of the results, supernatant antibody concentrations of 3.6ug/ml and 28.7ug/ml were obtained for monoclonal antibodies F50-75 and F51-99 respectively.
(iv) Aqqlutination of eCG coated sheep red blood cells The three eCG monoclonal antibody supernatants, and one negative control supernatant, were titrated in 96 well round bottom trays (Flow Labs, USA) by doubling dilution of 50ul volumes from neat to 1:16 in PBS/Az. lOul of eCG coated sheep red blood cells ~SRBC) as supplied in The MIP Test Kit (Carter and Wallace, USA) were added to each well and thoroughly mixed prior to settling. After 2-3 hours settling ~3oe~

at room temperature, the agglutinating ability of each monoclonal antibody was assessed according to cell settling patterns. An even mat of cells represented strong agglutination, whereas a discrete dark point of cells represented a negative reaction or no agglutination.
Supernatant of the monoclonal antibody F50-75 demonstrated considerably more agglutinating ability from neat to a 1:4 dilution than did supernatants of the monoclonal antibodies F50-37 and F51-99 which both showed only weak agglutination over the same dilution range. The negative control supernatant showed no agglutination.
(v~ Immunohistological characterization of three eCG
monoclonal antibodY specificities The three eCG monoclonal antibodies were unreactive with human and ovine placental sections, but were strongly reactive with horse endometrial cup sections. Figure 7 shows positive staining of the pregnant mare endometrial cup sections by each eCG monoclonal antibody. The heavy staining in the large binuclear cells of the endometrial cup is probably localisPd to the golgi apparatus. This could be confirmed usin~ electron microscopy and immunogold staining.
Monoclonal antibody F50-37 showed considerable staining reactivity with scattered cells of the equine pituitary and no speciic binding on pituitary sections of the other species e~amined. Monoclonal antibody F50-75 was reactive with scattered cells of equine, ovine and human pituitaries~ Monoclonal antibody F51-99 was unreactive with all pituitary sections. Immunohistological staining cf equine and human pituitary sections with the three eCG
monoclonal antibodies is shown in Figure 8. Figure 9 shows the staining patterns found for sections of ovine pituitary.
Additional Hybridoma Cell Lines Producinq Mono_lonal Antibodies to eCG .
Six further monoclonal antibodies have been established for examination and these areo F65-6, F65-53, F65-88, F65-98, F65-130 and F65-191.
The three most promising, as assessed in agglutination and cross-reactivity studies using uncloned supernatants, have been cloned. These cell lines are F65-88, F65-98 and F65-130. F65-88 and 98 are of the IgM class of :~3~

immunoglobulin, while F65-130 is an IgGl class immunoglobulin.
Samples of each of the above mentioned hybridoma cell lines are maintained in the Bunge (Australia) Pty. Ltd.
Cell Collection at 89 Flemington Road, North Melbourne, Australia.
Established uses for the Described HYbridoma Monoclonal Antibodies (i) Monoclonal Antibody Affinity ChromatoaraPhy Purification of eCG
In a still further aspect of the present invention there is provided a method of purifying equine chorionic gonadotrophin (eCG~ which method includes providing a source of impure eCG and a monoclonal antibody bound to a matrix;
and passing the impure eCG through the monoclonal antibody matri~ to capture eCG. The method of purification may further include eluting the captured eCG from the monoclonal antibody matrix.
The monoclonal antibodies bound to the matri~ may be any of the monoclonal antibodies described above. Monoclonal antibodies F50-37 and F51-99 are preferred for use in affinity chromatography. F50-37 is preferred because of its high eCG affinity as demonstrated by Scatchard analysis as described below. F51-99 is preferred for its e~quisite specificity for eCG.
A simple method for the purification of aquine chorionic gonadotrophin is of importance since those methods known in the prior art are both costly and time consuming, and, result in a product of low purity. A purified eCG is of importance as it provides a greatly improved biological response in-vivo. High purity eCG is also extremely desirable for application in human medicine.
The development of technology enabling the purification of eCG through monoclonal antibody affinity chromatography has provided a process whereby the supply of commercial ~uantities of a highly purified eCG product is possible.
Highly purified eCG has some distinct advantages over the impure eCG products currently available. Batch variation of commercial eCG is extensive and the quality of ~3~

the in-vivo response is poor by comparison. eCG purified from impure batches of eCG or pregnant mare serum using the presently described technology has little if any batch variation and provides a greatly improved response in-vivo.
The improved response is evident in all species tested~
A major advantage of the highly purified product is the reduced incidence of undesirable large follicles which may form in the ovaries adjacent to the Corpora Lutea. The formation of such follicles is implicated in a number of fertilisation complications including restriction of the cervix, blocking of the ovary ducts, e~cessive oedema and the like. See Table A. By comparing in-vivo responses in the ewe for identical doses of high purity and commercially available eCG, it has been shown that the incidence of unwanted large follicles is greatly reduced in those ewes treated with purified eCG. There was no associated loss or decrease in ovulation rate.
Accordingly, in a still further aspect of the present invention there is provided a purified equine chorionic gonadotrophin (eCG). The purified equine chorionic gonadotrophin is characterised by reduced incidence of large follicles in the ovaries of the animal treated.
TABLE A
In vivo response of the ewe to varying doses of high purity eCG and crude eCG preparations.

Dose eCG(iu) Mean No.of CorPora Lutea Mean No.of Follicles Crude eCG Purified eCG Crude eCG Purified eCG
~00 2.43 2.25 1.63 0.25 800 3.38 3.14 2.83 0.29 1200 6.38 6.75 4.50 1.25 1600 7.38 7.00 6.14 1.~8 Negative control animals (animals receiving no eCG~ had a mean of 1.43 corpora lutea and 0 follicles N = 200 The matri~ used may be of any suitable type. A
chromatographic gel is preferred. An affinity chromatographic gel rnay be used. The gel sold under the trade designation AFFIPREP 10 and available from Bio-Rad Laboratories has been found to be suitable. AFFIPREP 15 is 3L3~

also suitable.
For small scale work and large scale affinity purification of partially purified material, the monoclonal antibody matrix may be contacted with the eCG solution by passing the solution through the antibody matri~ supported in a chromatoqraphy column. However, it has been found for large scale work with crude material that the monoclonal antibody matrix and the eCG solution may be contacted in large scale containers and vats. The larqe scale containers or vats may be rotated, agitated or stirred to ensure complete contact between the monoclonal antibody matrix and the eCG solution.
The monoclonal antibody matrix may be provided in any suitable form. The monoclonal antibody matrix may be provided in a solution. A buffer solution may be used. A
phosphate buffered saline solution (PBS) may be used.
The source of impure equine chorionic gonadotrophin may be of any suitable type. The eCG source may be pregnant mare plasma or sera. The eCG source may be a crude e~tract thereof. The eCG source may be in a freeze dried form. The freeze dried sample may be dissolved in a suitable solution.
A buffer solution may be used. A phosphate buffer saline solution may be used. It has been found that the pH of the buffer solution may be important in optimising eCG binding.
A pH in the range o appro~imately 6.0 to 7.0 is preferred;
more preferably 6.5.
The step of passing the impure eCG through the monoclonal antibody matri~ may be undertaken in any suitable container. An affinity column may be used for relatively small scale work. For large scale work, a large container or vat may be used. The large container or vat may be rotated, agitated or stirred to ensure complete contact between the impure eCG and the monoclonal antibody matrix. A rotor bott]e may be used as a suitable container.
The captured eCG maybe eluted from the monoclonal antibody matrix utilising any suitable elution solution. An organic acid solution may be used. A pH buffered solution may be used. A sodium chloride-glycine buffer with a pH of approximately 2.0 - 3.0 has been found to provide maximum elution of purified eCG.

-- 1~ --Prior to the elution step, the method of purification may further include washing the monoclonal antibody matri~-bound eCG with a washing solution. The washing solution may be similar to ths binding bu~fer.
The elution step may be carried out in any standard affinity chromatography column. A large size column may be used for large scale work.
(ii) Modification of eCG Activity in vivo_ The dual FSH/LH bioactivity and long-half life of eCG in sheep and cattle provid~s eCG with the potent gonad stimulation effect unique to this hormone. These properties have led to its use in veterinary practice and research for induction of oestrus during anoestrus, increased fecundity and superovulation for embryo transfer.
The larger doses of eCG required for superovulation of livestock often have deleterious effects due to e~cessive and prolonged over-stimulation of ovarian follicles. The contaminants o~ known eCG preparations and the e~tremely long half-life of eCG in these animals results in the formation of large cystic follicles which can cause contractile scar tissue damage to the ovaries and oviducts, and~ e~cessive oestrogen production. These problems greatly reduce the number and quality of recoverable embryos or ova. As a result of the over-stimulation resulting from eCG usage, a means of modifying or controlling eCG bioactivity has been sought. Partial dialy~ation ha~ been attempted but not fully evaluated Papkoff, [Bio.Chem., 8io.Phys., Acta., 532, 185-194, (1978)~. To date most studies have used polyclonal eCG antisera to modify the half-life of eCG in cattle and sheep. Bindon and Piper [Proc. Embryo transfer in cattle and sheep, Canberra 1981, Published A.S.R.B. (1982)] found that the administration of 2ml sheep anti-eCG serum 24, ~8 or 72 hours after eCG injection greatly reduced the superovulatory response of the ~we to eCG. Even at 72 hours, the antiserum effectively halved the mean number of ovulations. The degree of ovarian over-stimulation in each group receiving anti-eCG
serum was reportedly reduced but the criteria on which this was established was not presented.
The in-vivo response to eCG has been greatly improved through the application of the following new technologies (i) High purity eCG, and (ii) Monoclonal antibodies to eCG.
In a further aspect of the present invention there is pro~ided a method of inducing oestrus and/or increasing fecundity in female animals which method includes administering to the animal an effective amount of a purified equine chorionic gonadotrophin. Anoestrus animals are thereby induced to cycle and ovulate. Animals may also show an increase in the frequency of the incidence of associated unwanted large follicles in the ovaries of the animal trPated.
The term animals as used herein in the description and claims includes domestic animals including sheep, goats, cattle, pigs, chickens and the like as well as other animals including humans.
For sheep and goats the equine chorionic gonadotrophin may be administered in amounts of from approximately 250 to 1000 iu. For cattle the equine chorionic gonadotrophin may be administered in amounts of from approximately 250 to 1000 iu.
In a urther preferred aspect of the present invention there is provided a method of substantially increasing the number of ova produced during a single oestrus cycle in female animals ~superovulation). This method includes administering to the animal an effective amount of a purified equine chorionic gonadotrophin. The number of ova produced may be increased to a level o at least about 5 preferably 7. The use of highly purified eCG instead of the current commercially available products greatly reduces the incidence of associated unwanted large follicles in the ovaries of the animal treated.
For sheep and goats the equine chorionic gonadotrophin may be administered in amounts of from appro~imately 500 to 2000 iu.
For cattle the equine chorionic gonadotrophin may be administered in amounts of from appro~imately 500 to 3000 iu.
It has now been discovered that monoclonal antibodies to equine chorionic gonadotrophin may block and/or enhance the biological activity of equine chorionic gonadotrophin in animals.

~L3~

Control of eCG bioactivity ln vivo using monoclonal antibodies is advantag~ous when using e~G for ti) superovulation, (ii) increasing fecundity, and ~ iii) manipulation of the oestrus cycle.
The advantages already described for superovulation are also applied to the low dose usage for eCG for increasing fecundity, i.e. twinning and manipulation of the oestrus cycles. By using these monoclonal antibodies with low doses of eCC. the incidence of twin births in a variety of species can be better controlled.
Twinning is therefore easier and more repeatable, and the numbers of unwanted triple and high multiple births are reduced.
~ ccordingly in a further aspect o~ the present invention, there is provided a method of increasing ovulation to provide twinning in female animals which method includes administering to the animal an effective amount of a purified equine chorionic gonadotrophin characterised in that in use there is a reduced incidence of large follicles in the ovaries of the animal treated.
Preferably the animal to be treated is a sheep, goat or cow and the amount of equine chorionic gonadotrophin administered is in the range of approximately 100 to 1000 iu.
More preferably the method o~ increasing ovulation to provide twinning further includes administering simultaneously with or after equine chorionic gonadotrophin administration to the animal approximately 0.25 to 1.0 mg of a purified monoclonal antibody against equine chorionic gonadotrophin formed by a hybridoma cell line.
In a preferred aspect of the present invention the method of inducing oestrus and/or increasing fecundity further includes administering to the animal to be treated an effective amount of a monoclonal antihody against equine chorionic gonadotrophin (formed by a hybridoma cell line) simultaneously with or after equine chorionic gonadotrophin administration.

., , ~

17a ~3~4 The monoclonal antibody may be any of the antibodies described above. PrefPrably the monoclonal antibody is monorlonal antibody F51-99. The monoclonal antibody may be administered in any suitable e~ective amount. Fo~ s~eep and goats the monoclonal antibody may be administered in amounts from approximately 0025 to 1.0 mg. For cattle the ~onoclonal antibody may be administered in amounts from approximately l.o to 3.0 mg. For pigs the monoclonal antibody may be administered in amounts from approximately 0.5 to 1.5 mg.
In the method of super ovulation, the monoclonal antibody may be administered for sheep and goats in amounts of from approximately 0.5 to 2.0 mg. For cattl~
the monoclonal antibody may be administered in amounts of from approximately 0.5 to 3.0 mg.
When administering eCG monoclonal antibodies for control of eCG in vivo, an accurate assessment of the required dose is essential. Excessive amounts o~ the monoclonal antibody can lead to an increase in the number of unruptured follicles possibly due to cross-reactive inter~erence with endogenous ho~mones or stripping of eCG

/

. ~

from biological receptorsO
The antibody dose suitable is dependent on several variables including:
~ i) the monoclonal antibody selected, ~ ii) eCG dose, (iii) time following eCG injection of antibody injection, (iv) species in which it is being used, (v) size of the animal, and (vi) number of times the animal has been previously treated .
An injection of GnRH (gonadotrophin releasing hormone) may also be incorporated into the system. This may be given before antibody administration and about 24 hours before the time of ovulation.
The use of monoclonal antibodies for controlling eCG
in vivo whether for use in superovulation, increasing fecundity i.e. twinning or for induction of oestrus as described for ewes and cows has similar application in goats, humans, dogs, cats, pigs and other mammals.
The amount of antibody necessary to be effective will vary with the particular antibody used, the species of animal, the level of eCG treatment and time of administration following eCG injection. For example, where from appro~imately 1500 to 2000 iu of eCG is injected into a ewe, we have found that appro2imately 0.25 to 5.00 milligrams of antibody may be used depending on the variables already described. The monoclonal antibody treatment may be undertaken contemporaneously with the eCG treatment or may be undertaken after eCG treatment.
Monoclonal antibodies administered simultaneously or after eCG administration can modify the biological activity of eCG. This modification may reduce or enhance the activity of the eCG. Both enhancement and reduction of eCG
bioactivity can be incorporated in the same regime.
Preincubation, simultaneous injection or independent administration of eCG and the monoclonal antibody can enhance the activity of the eCG via the formation of antibody-eCG
compleses, result;ng in greater biological activity.
Equally so the formation of a different antibody-eCG

~3~

comple~ can effectively reduce the bioactivity of the eCG or previously complexed eCG by nullifying the biological activity of the eCG.
A trial was undertaken in rats to e~amine the enhancement properties of 14 eCG monoclonal antibodies. A
low (non-effective) dose of eCG was incubated overnight at 4C with approximately an equal molar amount of each antibody. Pairs of 21-28 day old female PV~ rats wers then injected with the mixture. 48 hours later rats were e~amined and scored as for the standard eCG bioassay.
Results All rats receiving 0.5 iu eCG alone or saline were negative. Identical doses of eCG incubated as described above with monoclonal antibody F65-191 had significant biological activity. This activity was equivalent to the 2.5 iu eCG injected into positive control rats. Monoclonal antibody enhancement of the biological activity of eCG was restricted to monoclonal antibody F65-191, all others tested were negative in this trial. See Table B.
The modification of bioactivity, whether it be enhancement of reduction, of other pharmaceuticals by selected monoclonal antibodies is also possible.

~3~

TABLE B

Monoclonal Antibody Unit eCG Response Rat No. 1 Rat No. 2 Saline only0.5 iu - -Saline only2.5 iu ~ -~
F50-37 0.5 iu F50-75 0.5 iu - -F51-99 0.5 iu F65-6 0.5 iu - -/
F65-50 0.5 iu F65-53 0.5 iu F65-56 0.5 iu F65-88 0.5 iu F65-98 0.5 iu F65-130 0.5 iu F65-135 0.5 iu F65-146 0.5 iu F65-165 0.5 iu F65-191 0.5 iu +

(iii) _G Detection and Quantitive As~ys In a still further aspect of the present invention there is provided a method for the detection of equine chorionic gonadotrophin (eCG~ in an equine mammal. The detection assay may be an immunofluorescent assay, radioimmuno-assay, en~yme-linked immuno assay or agglutination assay. For the IgM immunoglobulin class monoclonal antibodies an agglutination assay is preferred. A
haemagglutination assay is preferred.
In a particularly preferred form the detection assay is a haemagglutination inhibition assay.
For the IgG immunoglobulin class monoclonal antibodies, the RIA or ELISA assays are preferred.
The sample to be tested may be of any suitable type. Sera, plasma, saliva or urine samples or extracts thereof may be used.
Thus, in a preferred aspect, there is provided a method for detecting the presence of equine chorionic 2~

gonadotrophin, which method includes providing a sample from the equine mammal to be tested, an IgM class eCG monoclonal antibody from those described above, or mixtures thereof and a source of equine chorionic gonadotrophin (eCG)-coated mammalian red blood cells. Contacting the monoclonal antibody/s with a portion of the sample to bind any equine chorionic gonadotrophin present; and contacting the mixture of antibody and sample with the eCG-coated mammalian red blood cells.
It would be understood that the presence of eCG may then be detected by visual means. Where eCG is present in the serum of interest, the eCG will bind to available binding sites on the monoclonal antibodies. When the eCG coated mammalian red blood cells are added there is effectively no antibody to bind to the antigsn on the cells and there is a resultant inhibition of agglutination. Non-agglutinated cells subsequently slide around the round bottom of the solid support well to form a dark dot of cells.
The solid support may be of any suitable type. A
microtitre plate or tray or test tube may be used. A
microtitre plate or tray of the round bottom type is preferred. The monoclonal antibodies may be introduced as a solution. A phosphate buffered saline solution may be used.
The equine chorionic gonadotrophin coated mammalian red blood cells may be eCG coated sheep red blood cells.
In a preferred form the detection methods may be used for detecting other pregnancy proteins in other mammals. The detection method is particularly applicable to eCG in equine animals, although hCG in humans and other pregnancy proteins in other mammals may be similarly tested.
Thus in an alternative form, a human chorionic gonadotrophin-specific monoclonal antibody may be incorporated in any suitable assay to detect human chorionic gonadotrophin.
Thus in a preferred alternative aspect, there is provided a method for detectiny the presence of human chorionic gonadotrophin, which method includes providing a sample from the human to be tested, an hCG - (human chorionic gonadotrophin) - specific monoclonal antibody, or mixtures thereof an a source of hCG - coated mammalian red blood cells, contacting the mo~oclonal antibody with a portion of the sample to bind any hCG present; and contacting the mixture of antibody and sample with the hCG - coated mammalian red blood cells.
The hCG-specific monoclonal antibodies may be elected utilizing similar immunohistological techniques to those described above for selecting eCG-specific monoclonal antibodies.
Accordingly, in a further aspect of the present invention there is provided a test kit including a source of monoclonal antibody in a suitable container;
a source of eCG-coated mammalian red blood cells in a suitable container;
and a solid support.
The monoclonal antibodies may be provided in a solution. A buffered solution may be used.
The eCG coated mammalian red blood cells may be eCG
coated sheep red blood cells.
The solid support may be a plastic support. The solid support may be a microtitre plate or tray or test tube. A round bottomed microtitre plate or tray may bP
used. The solid support may take the form of plastic balls, paddles, tubes or membranes. The solid support may be in the form of a glass. Glass beads may be used. The solid support may be in the form of a paper. A filter paper or paper membrane may be used.
The established eCG specificity of monoclonal antibody F51-99 is ideally suited for the development of an eCG specific assay. The mono and/or bivalency of this class of immunoglobulin (IgG) usually renders it unsuitable for agglutination type assays but suitable for ELISA and RIA
systems. The ELISA system is well suited for "in~field" eCG
detection. In such a system enzyme conjugated F51-99 is mixed in solution with the sample in question. If sufficient eCG is present the labelled antibody is "mopped up", if not some remains free in the mixture of solutions. This free antibody is then removed for detection. This is accomplished by exposing this solution (mix) with immobili~ed eCG.
Unblocked or free antibody remaining in the solution then binds. Residual unbound comple~es are rinsed away Erom the antibody bound to the immobilised eCG. Simple e~posure o the eCG matri~, with or without antibody attached, to a substrate solution determines the presence or absence of antibody through eg. a colour change reaction.
The reverse ELISA system to that describsd here is also applicable where the antibody is immobilised and the eCG
is labelled with an enzyme.
Enzymes suitable for couplîng to the monoclonal antibody and eCG include urease, horse radish perogidase and alkaline phosphatase. Each has its own relevant substrate system and colour change reaction as known in the art.
Accordingly, in a further aspect the present invention provides a veterinary or pharmaceutical composition for the modification of the biological activity of eCG in mammals including an effective amount of a monoclonal antibody selected from those described above or mixtures thereof in a unit dosage form.
The veterinary or pharmaceutical composition may further include a veterinarily acceptable or pharmaceutically acceptable diluent carrier or e~cipient therefor. The monoclonal antibody may be provided in solution. A buffered saline solution may be used. A phosphate buffered solution may be used.
In an alternative aspect, there is provided a method of inducing abortion in pregnant mammals, which method includes administering to the mammal an efective amount of a monoclonal antibody selected from those described above or mi~tures thereof.
The method of inducing abortion of unwanted pregnancies is particularly applicable to equine animals.
This is particularly applicable in relation to thoroughbred horses, where abortions are required for mares carrying twins or foetuses of unknown paternity. The treatment method is in general similar to that used for blocking of the biological activity of eCG. However, in general larger dosages may be required.
The present invention will now be more fully described with reference to the accompanying e~amples. It should be understood, however, that the description following ~3~2~

is illustrative only and should not be taken in any way as a restriction on the generality of the invention described above.
EXAMPLES
MATERIALS AND METHODS
Serum Collection and Crude Extract Preparation Sera were collected fresh at a local knackery.
Twelve ml of 10% potassium oxalate was added to each litre of whole blood as an anticoagulant. Supernatants were collected after settling and frozen at -20C until required for processing.
Extractions from sera using the technique of Gospadorowicz and Papkof~ (1967) gave poor eCG yields (5-10%). The technique was subsequently altered as follows.
0.25M metaphosphoric acid was freshly prepared at 4C and added to cold serum with stirring until a pH 3.5 was reached. After centrifugation ~15,000g for 30 minutes at 4C), the supernatant was poured off and adjusted to pH 4.5 with l.OM sodium hydro~ide. The supernatant was then made up to a 40% solution (v/v) with ethanol by slow addition of absolute ethanol (-20C) while stirring. The supernatant was then removed by centrifugation ~as above) and made up to a 75% solution (v~v~ with absolute ethanol (-20C) while stirring. The resulting precipitate was harvested by centrifugation, dialysed against double distilled water, freeze-dried and stored at -20C until required.
eCG:Haemaqqlutination Inhibition Assay All procedures were performed at room temperature and the haemagglutination inhibition assay was based on reagents supplied in a commercially available mare pregnancy diagnosis kit having the trade name of "M.I.P. (Mare Immunological Pregnancy) Test Kit" (Carter-Wallace, Inc., N.J., USA). The eCG standard for the assay was the W.H.OA
2nd International Standard for PMSG ~W.H.O., London).
eCG test samples and -the eCG standard were titrated by doubling dilution in 50ul volumes of phosphate buffered saline containing 0.1% sodium azide (PBS/Az) in 96 roun~ well titration tra~s ~Flow Laboratories). The following solutions were added per well as directed by the manufacturer, to eliminate possible non-specific agglutination during the assay. 5ul of M.I.P, Test Kit Inhibitor solution was added and mi~ed by rocking for 90 seconds. 5ul of M.I.P. Test Kit Neutraliser solution was then quickly added. 15ul of M.I.P.
Test Kit Antisera solution, (the optimal dilution being determined by titration to give sufficient agglutination of lOul of Test Kit eCG coated Sheep Red Blood Cells, [eCG SR8C
Solution]), was then added and mi~ed by occasional rocking for S minutes. lOul of M.I.P. Test ~it eCG SRBC Solution was then added per well and the tray rapidly rotated on a Microshaker~(Cat. No. AM69, Dynatech, England) for 10 seconds to ensure good cell distribution prior to settling. End points for the assay as determined by red blood cell settling patterns, ranged from 0.5iu/ml - ~iu/ml depending on the batch of the kit used. To standardise the assay, results were checked periodically by comparison with the values obtained using the rat ovarian weight bioassay as described by Cole and Erway, Endocrinology 29 (1941), _LISA assay_of eCG
ELISA (96 well3 plates were coated with 50 ul of 8 ug/ml monoclonal antibody F50-37, 6 ug/ml F50-75 in 0.1 M
carbonate buffer pH 9 for 60 minutes at 37C in a humidified chamber. Mi~tures of other monoclonal antibodies were also successfully used.
The coating solution was then aspirated and 250 ul of blotto (4~ skim milk in PB53 added to each well. This was again incubated as above.
eCG samples were titrated by doubling dilution in blotto in a separate 96 well plate. After aspiration of the blocking solution and 3 distilled water washed titrated samples were then transferred to the coated ELISA trays. The trays were incubated for 60 minutes as above.
Urease conjugated monoclonal antibody F51-99 diluted 1:100 in blotto was added to each well after thorough washing of the ELISA tray and incubated as above for a further 60 minutes.
After a final 3 water washes 75 ul of urease substrate solution was added per well and the reaction allowed to proceed until background reactions began to appear. Trays were then read on a Multiscan Titretek~ELISA
reader and eCG concentrations determined by e~trapolation ~ ~ - 25 -from a standard curve. Folligon 100 iu/vial material was incorporated as the standard.
Bioassay of eCG
Pairs of 21-26 day old virgin female PVG rats were lightly anaesthetized with chloroform and injected intraperitonealy (IP) with 0.2ml of each dilution of a standard or test sample of eCG in PBS. Autopsies were performed 48 hours after injection and ovarian/uterine development, including size and vascularization, assessed and scored on a 4 point scale as "++", "+", "+-" or "-". some variation in end points was found between assays, but typically an injection of 4iu or more gave a "+~" result, 2iu "+", liu "+-" and 0.5iu a "-" result. This assay was adapted frorn that described by Cole and Erway [Endocrinology, 29, 514-519 (1941)~.
Production of monoclonal antibodies to eCG
IMMUNIZATION: Four BALB/c mice were injected IP
with 1,000 iu eCG dissolved in 100ul PBS and emulsified in a further 150ul of Freund's Complete Adjuvant (FCA). The same batch and quantity of eCG in Freund's IncompletP Adjuvant (FIA) was injec~ed IP weekly for a further 3 weeks. One week after the final injection, mouse sera obtained by tail bleeding were egamined for eCG reactivity and specificity by immunohistology on a range of pregnant and non-pregnant mare tissues. Mice having the highest serum titres were boosted 4 days prior to fusion with l,000iu eCG in 0.1ml PBS by tail vein injection.
FUSION AND CLONING: The following reagents were required for the production of monoclonal antibodies and were prepared as follows:
(i) GKN Solution: 8.0g NaCl, 0.4g KCl, 1.42g Na2HP04 2H20, 2.0y D-Glucose and 0.01g phenol red were dissolved per litre of double distilled water, filter sterilized and stored at room temperature.
(ii) Tissue culture media: Dulbecco's Modified Eagles medium (DME, powdered and sodium bicarbonate free, Gibco USA) and sodium bicarbonate (0.2% w/v) were dissolved in double distilled water as directed, filter sterilized and stored at 4C. Prior to use, the DME solution was supplemented with 10%

(v/v~ foetal calf serum (FC~, Flow Laboratories~ and 2mM L-glutamine (200mM L glutamine, Flow Laboratories). HAT selec-tive media contained a further 13.6mg hypo~anthine, O.l9mg aminopterin and 3.87mg thymidine per litre of media. HT selective media was identical to HAT selective media e~cept it contained no aminopterin.
Fusion and cloning steps were carried out as follows: Briefly, a primed mouse was killed by cervical dislocation and the spleen aseptically removed. Splenocytes were obtained by gentle s~ueezing and teasing of the spleen in lOml GKN solution. The cell suspension was allowed a short settling period (30-60 seconds) to permit the removal of large clumps of tissue. The cells remaining in solution were washed twice by centrifugation (200 g for 5 minutes at room temperature) and resuspensed in lOml GK~. After the final centrifugation the cells were resuspended in a further lOml GKN.
NS-l myeloma cells, 2 ~ 107, (P3/NSI/l-Ag4-1) grown in tissue culture media were collected by centrifugation (200 g for 5 minutes at room temperature) and resuspended in lOml GKN solution.
The lOmls of splenocytes and NS-l myeloma cells were mi~ed in a 50ml centrifuge tube (Falcon) and centrifuged at 70g for 5 minutes at room temperature. The supernatant was then poured off to leave a pellet of cells which were resuspended to form a slurry, by gently tapping the tube, in a minimum of residual supernatant. 1.5ml of sterile 50~
polyethylene-glycol at 37C [PEG, Art. 9727 Polyethyleneglycol 4000, Merck] prepared by autoclaving l.Og PEG with lml double distilled water in a sealed container) was then added to the slurry over 90 seconds, with gentle mixing using the pipette as a stirring rod. After a further 30 seconds of mi2ing, GKN at 37C was carefully added with stirring. The first 2ml of GKN was added over 2 minutes, and then a further 8ml was added over 3 minutes. Another 30ml GKN was then gently added down the side of the centrifuge tube.
The cell suspension was immediately centrifuged at room temperature for 3 minutes at 70g with the brake off, and the supernatant poured off. The tube was tapped gently to form a slurry of c~lls before resuspension in 50ml of HAT
selective media. 0.5ml samoles of the cell suspension were then added to each well of four 24 well flat bottom Linbro~
plates (Flow Laboratories) each containing 1.5 ~ 106 rat thymocytes in 1.5ml HAT selective media. The thymocy~es were prepared on the day prior to the fusion and incubated overnight at 37~ in a humidified 5% CO~ atmosphere to ensure sterility.
Hybrid colonies were monitored using an inverted microscope (Olympus). 10-ll days after fusion supernatants were screened by ELISA and immunohistology. Selected wells were dilution cloned in 96 well flat bottom Linbro plates (Flow Laboratories) with thymocytes in HAT selective media.
Three trays containing appro~imately 1, 10 and 50 hybridoma cells per well were preparQd.
ELISA Screeninq Assay 96 well flat bottom Linbro plates ~Flow Laboratories) were coated with antigen by incubating overnight at 4C with 50ul/well of a solution containing 15-120ug/ml of protein in 0.5M ~a2HP04, pH 9 containing 0.1% NaN3. Trays were rinsed 3 times with PBS prior to use. Sera and supernatants were diluted 1:500 and 1:1 in "blotto" (4% skim milk powder [Diploma, Australia], in PBS
and made fresh daily) respectively for screening. 50ul volumes of diluted antibody were added per well and the trays then incubated for 2-3 hours at room temperature in a humidified bo~.
Following three rinses with PBS, 50ul of pero~idase-conjugated rabbit anti-mouse immunoglobulin (RA~-HRP, Dako~ Denmark) diluted 1:1,000 in blotto was added per well and the tray incubated for 1-2 hours as above.
After a further 3 rinses with PBS, lOOul of substrate solution was added per well. This solution was prepared fresh on the day of use by dissolving 40mg O-phenyline diamine (BDH) in lOOml O.OlM citrate buffer pH 5Ø
Immediately prior to use, 80ul of 30% hydrogen pero~ide was added with stirring.-The reaction wa~ stopped within 30 minutes by theaddition of 25ul of 25~ (v/v) sulphuric acid per well, ~3~

depending on development of the background wells. Trays were read on a Microtitre Multiskan (MCC, Finland) at 492nm.
Immunohistoloqy Fresh tissues were collected at a local knackery and fixed in Bouin's solution for 2 hours, or in cold 95% ethanol overnight. After trimming, tissues were embPdded in paraffin by processing through four 1 hour changes of absolute ethanol at 4C , four changes of xylene at room temperature, and four changes of liquified paraffin at 56C, and then leEt to solidify overnight at 4C.
Tissue sections of 3-5um were cut using a microtome ~Leitz) and floated onto glass slides from a warm 1% gelatin ~w/v) water bath. Slides were then incubated overnight at 37C to adhere the sections. Prior to use, sections were rehydrated by 5 minute incubations in each of the following solutions; xylene, fresh (wax free) xylene, absolute ethanol, fresh (xylene free) absolute ethanol, 75% ethanol and running tap water. Thorough paraffin removal was found to be essential to minimise background staining due to non-specific binding of the second antihody. After two rinses in double distilled water, sections were kept moist and used within 2 hours.
Neat hybridoma supernatants,or mouse sera diluted 1:100 in blotto were placed directly onto the sections and the slides incubated in a humidified box for 2-3 hours at rcom temperature. After gentle rinsing with PBS, sections were covered with rabbit anti mouse IgG conjugated to horse radish peroxidase (RAM-HRP) diluted 1:60 in blotto and incubated for a further 1-2 hours as before. After thorough but gentle rinsing, the slides were submerged in a substrate solution containing 0.14g 3,3'-diaminobenzidine tetrachloride in 250ml O.OlM citrate bufer, pH 5, plus 250ul of 30%
hydrogen peroxide added immediately prior to use. Within 10 minutes the reaction was stopped by rlnsing in tap water.
Sections were counterstained with Harris' haemato~ylin and lithium carbonate, and then dehydrated by 5 minute rinses through 70% ethanol, absolute ethanol, ~ylene and fresh xylene for mounting with coverslips and Depex.
Chloramine T Radiolabellina of eCG
lmCi 125I ~lOOmCi/ml, Amersham) was mixed with ~3~
lOul 2mg/ml chloramine T in 0.3M sodium phosphate pH 7.4 (freshly prepared) and 10-20ul lmg/ml eCG in Dulbecco's A + B
buffer plus 0.1% sodium azide (DAB~z) for 10-~0 seconds at room temperature. The reaction was stopped by the addition of 5ul saturated L-tyrosine solution. lOOul 10% BSA in DAB/Az was then added to the misture prior to loading onto a 5ml G-50 Sephadex 3 Smedium grade) column packed in a disposable lOml pipette. Prior to separation the column was washed with 250ul 10~ BSA in DAB/Az and equilibrated with DA3/Az.
The loaded sample was eluted with DAB~Az and collected as three 1.5ml fractions. Fraction 2 contained the labelled protein while free 125I was eluted late in fraction 3, as determined by 10% trichloroacetic acid precipitation o~ the fractions.
Sodium Dodecyl Sulphate Polyacrylamide Gel Ele~troDhoresis tSDS-PAGE~
SDS-PAGE analysis was based on the method of Laemmli [~ature, 22Z, 680-685, ~1970~] and used the following reagents:
~A) acrylamide/bisacrylamide (Bio-Rad) 37.5:1. A stock solution of 30.8g/lOOml of double distilled water was stored at 4C in the dark.
(B) 0.75M Tris-HCl, pH 8.8 (C) l.OM Tris-HCl, pH 6.8 ~D) 10% SDS in double distilled water (E) 1% (w/v~ ammonium persulphate (F3 TEMED (Sigma) 10% acrylamide gels were prepared by mi~ing ~Oml of (A), 15ml of (B), 0.3ml of (D), 3.2ml of double distilled water, 1.5ml of (E) and 0.02ml of (F3. (E~ and (F) were added just prior to pouring of the gel. The stacker gel was prepared with 0.8ml o (A~, l.Oml of (C), O.G8ml o (D), 4.7ml of double distilled water, 0.4ml of (E) and 0.012ml of (F). Again (E) and (F) were added immediately before use.
Gels were run over 4-5 hours at 160V (constant voltage) using a Bio-Rad~Protean cell kit and running buffer consisting of 0.025M Tris, 0.192M glycine and 0.1% SDS.
Samples were prepared by incubating at room temperature for 30 minutes or by boiling for 5 minutes in an ~3~6AZ5~
equal volume of sample buffer. Sample buffer was prepared by mixing lOul 5% bromophenol blue, 200ul glycerol, 250ul 20%
SDS and 500ul l.OM Tris-HCl pH 6.8. Reducing sample buffer contained an additional 4% dithiothreitol.
Coomassie blue staining of gels was performed by overnight immersion in a 45.5~ methanol, 9% glacial acetic acid solution containing 0.125~ (w/v) Coomassie blue. Gels were destained by overnight rocking in a 25% methanol, 7.5%
glacial acetic acid destain solution.
Molecular weight markers used were (Cat. No. SDS-7, Sigma); Mr 14,200 ~ -lactalbumin, Mr 20,100 soyabean trypsin inhibitor, Mr 24,000 trypsinogen, Mr 29,000 carbonic anhydrase, Mr 36~000 glyceraldehyde-3-phosphate dehydrogenase, Mr 45,000 egg albumin and Mr 66,300 bovine albumin.
Western Immunoblottinq SDS-PAGE gels were soaked for 15 minutes in Western transfer running buffer (0.25M Tris, 0.192M glycine, 20%
methanol) and loaded into a Bio-Rad Transblot apparatus as recommended by the manufacturer. Transfer of proteins onto the nitro-cellulose paper (O.lum) was performed overnight at 4C using a current of 150mA. The nitrocellulose paper was removed and blocked by incubation in blotto for l hour at room temperature. After three rinses in PBS/Az the paper was cut and placed into resealable plastic bags containing 20-30ml of monoclonal antibody supernatant. The bags were sealed and rocked for 2-3 hours at room temperature on a Bellco rocking platform (Cat. No. 7740-20220 Bellco, USA) prior to thorough rinsing with PBS/Aæ and the addition of 20ml RAM-HRP second antibody diluted 1:500 in blotto. After 1-2 hours incubation at room temperature while rocking, and thorough rinsing in PBS (no azide), the colour reaction was developed as described for immunoperoxidase staining of tissue sections.
Isoty~inq Monoclonal antibodies supernatants were isotyped using a dot-blot technique. Reagents were purchased as an immunoglobulin class and subclass detection kit (Chemicon International, USA3 and used according to the manufacturer's instructions. An alternative second antibody to that provided in the kit, a p~ro$idase conjugated goat an~i-rabbit IgG (Lot ~o. LP002) wa~ purchased ~rom the same company and diluted 1:500 for use.
Ascites Production and Purification of Mouse IqG
(i) Production:
Cloned hybridoma cell lines were slowly taken off HAT selective media via HT media into ~AT free media. 1 ~
107 hybridoma cells were washed three times in PBS by centrifugation, and then resuspended in 0.5ml PBS for IP
injection. Recipient BAI,B/c mice were injected with 0.5ml pristane (2, 6~ 10, 14-tetramethyl pentadecane 96~3 3 weeks prior to hybridoma cell injection.
Following cell injection mice were regularly observed and killed when sufficient ascites was present. The ascites fluid was centrifuged (240g for 5 minutes at room temperature~ and the supernatant removed. Sodium azide was then added to a final concentration of 0.1%, and the ascites fluid stored at 0-4C for periods of less than 2 days, or at -20C for longer periods, or at -70C without azideO
(ii~ Purifica~ion:
Ammonium sulphate precipitation and DE-52 chromatography was used to P~tract and purify immunoglobulin from ascites fluid. E~ual volumes of ascites fluid and saturated ammonium sulphate solution were mi~ed by dropwise addition of the latter, with rapid stirring. The mi~ture was left to stand for 1 hour at room temperature and then centrifuged (12,000g for 20 minutes at 20C). The pellet was washed twice by resuspension and centrifugation in 50%
saturated am~onium sulphate. The precipitate was then redissolved in double distilled water and dialysed extensively against 25mM Tris, 25mM NaCl, pH 7.~.
A 50ml DE-52 Cellulose (Whatman,~ Kent England) column was packed in a 50ml ~isposable syringe ~Terumo, Aust.) and e~tensively equilibrated with the 25mM NaCl, 25mM
Tris pH 7.4 starting buffer. When the dialysed sample and column effluent conductivities, as measured on a conductivity meter (Model 101, Orion) were + 1-2mS, the sample was pumped onto the column at 40ml/hr, and lOml fractions collected on a LKB fraction collector with a LKB W spectrometer and chart recorded. All chromatographical procedures were performed at ~ - 32 -~ ) .

z~

4C.
Once the elution profile had returned to baseline levels, following thorough washiny with the starting buffer, the column was eluted sequentially with 50mM, lOOmM, 150mM
and 200mM NaCl. Each eluted peak was then e~amined for activity by ELISA assay and the relevant peak pooled and made up to 0.1% ~aN3 for storage at 0-4C, short term, or -20C for longer periods.
Production of Rabbit Antiserum to Mouse IqG
Rabbit anti-mouse IgG was prepared by 4 fortnightly intramuscular injections of lmg purified mouse IgG (~mg/ml PBS~ emulsified in lml Freund's complete adjuvant (primary immunization only) or Freund's incomplete adjuvant. Rabbits were bled one week after the final immunization.
Preparation of Affinity Chromatoqraph~ Columns Purified immunoglobulin from ascites fluid ~as coupled to AFFIGE~ 10 ~Biorad) following dialysis in 0.lM
NaHC03 pH 8.5, according to the manufacturers instructions. One volume of gel was coupled at 4C with 2-3 volumes of antibody solution (16 mg/ml) by rotational mi~ing overnight.
PhQtoqraPhy Agfapan 25 film (.~gfa-Gevaert~ was used for photography of Western immunoblots. The film was developed hy Rodinal (1/25~ for 5 minutes at 18-21~C, washed in tap water, and then fi~ed for 5 minutes in rapid fi~er A
(Kodak3~ The negatives were printed on grade 3 paper (Kodak) using Dekto~ developer ~Kodak).
Agrachrome 50L film (Agfa Gevaert) was used for colour photography of immunohistological slides. Colour film was printed using the Ektaflex~ process (Kodak).
Magnifications cited or colour illustrations have been corrected for magnification during photography and printing.
EXAMPLE l The Production and Characteristi~s of MonQclonal Antibodies to çCG
Screeninq antiaens:
Three serum batches were processed according to the adapted e~traction technique of Gospodarowicz, endocrinology 80 (1967). ~e-se we~e ~L3~6~

(i, 1.0 1 of serum rom a day 60 pregnant mare (PMS) (ii) 0.5 1 normal mare serum (NMS), and (iii? 0.5 1 normal stallion serum (NSS).
All crude preparations were dialysed against double distilled water, freeze dried and stored at -20C. Further purification of 73mg of the crude PMS preparation was undertaken according to the method described by Gospodarowicz (1967) using Sephade~ G-100 gel-filtration. This resulted in a 4 fold increase in speciic activity from 400iu/mg to 1600iu/mg.
Immuno~ens:
Three different eCG preparations were used for immunization, (i) 1,600iu/mg material prepared as above.
(ii) lO,OOOiu/mg pure eCG (NIH).
(iii) 400 iu/mg "Twinject PMSG" (Batch 8367, Immunoproducts Melbourne) Screeninq assays:
Two screening techniques were used; enzym~ linked immunosorbent assay (ELISA) and immunohistology. Immune sera and hybridoma supernatants were screened by ELISA assay as described above using PMS and NMS coated 96 well flat bottomed PVC trays. Supernatants which gave positive binding to PMS coated wells and no binding to NMS coated wells were further e~amined by immunohistology on 5um sections of pregnant (day 60) and non-pregnant mare uteri. Antibody preparations staining specifically in tissues known to synthesise eCG were selected for further study.
Cell fusions:
Two fusions were performed one week apart according to the method described above and were code named F50 and F~l respectively. Supernatants from the hybridoma cell lines were screened 10-12 days after fusion. Hybridomas of interest were dilution cloned at least 4 times to assure the selection of a single cell line.
Of the 12 eCG immunized mouse sera screened, 4 gave strong positive staining by immunohistology on endometrial cup sections, 6 gave weak staining and 2 gave no staining at all. The groups of mice immunized with the ~IH and 1600iu~mg eCG preparations each contained 2 mice with strong positive ~L3~

sera. The positive sera of the 2 mice injected with the ~IH
preparation had a higher titre, and stained more specifically in endometrial cup tissues. These mice were chosen for the production of hybridomas.
Normal and immune mouse sera staining of day 60 endometrial cup sections from a 60 day pregnant mare uterus are compared in Figure 10. The large cells evident in the centre of the negative control section lO~a), which are not stained, are clearly immuno-reactive with the immu~e mouse serum in 10(b) as shown by the strong brown reaction product. Some cells associated with the overlying foetal chorion also showed positive staining. If staininq is specific for eCG, this may be due to local eCG production or the adherence of eCG secreted by the large uterine embedded decidual-like cells of the cup tissue through the glandular openings into the lumen. The tissue seen in the bottom left corner of Figure 10(b) shows very little if any staining, and is representative of staining by this antiserum on non-pregnant uterine tissue sections.
Fusion 50 produced 12 wells with PMS-reactive supernatants by ELISA assay, out of the 192 wells containing hybridomas. Of these 12 supernatants 10 were also NMS-reactive. The specificity of each supernatant was further checked using immunohistology and showed that the same 2 wells specific for PMS by ELISA assay, wells 37 and 75, were specific for endometrial cup tissue. Other supernatants stained non-specifically or not at all. Wells 37 and 75 were dilution cloned at least 4 times prior to coding the cell lines F50-37 and F50-75 respectively.
Fusion 51 produced only one PMS-positive, NMS-negative well, by EhISA and immunohistology, well 99.
This was dilution cloned 4 times and coded F51-99.
Large quantities of high titre supernatant were produced for all three cell lines. Uncloned wells from the initial 24 well trays were pooled; e~panded and frozen for storage in liquid nitrogen. Cell lines derived at each dilution cloning were also exp~nded and frozen in liquid nitrogen.
The ELISA screening assay was designed to select for monoclonal antibodies to antigens present in crude e~tracts 2~

of PMS and absent in NMS extracts. This assay did not exclude the possibility of selecting or antibodies directed against other pregnancy related antigens in the PMS crude extract. Further screenin~ using immunohistology restricted the specificity of the selected antibodies to those molecules found in endometrial cups of pregnant mares. By screening on a variety of endometrial cup sections originating from several different mares, the possibility of selecting for antibodies to paternally derived antigens expressed on the surface of the endometrial cup cells was eliminated.
Further studies were undertaken to demonstrate the reactivity of the three monoclonal antibodies with purified eCG.
Although the screening procedures used in the identification and subsequent cloning of hybridomas were selective for pregnancy related antigens localised to the endometrial cups of pregnant mares, conclusive eCG reactivity of each monoclonal antibody had not been demonstrated. To establish this, radio-immunoprecipitation experiments were performed to show binding of 125I labelled eCG by each monoclonal antibody. Once the eCG reactivity of each monoclonal antibody had been established, further studies to enable Scatchard plot analysis [6catchard, Ann. N.Y. Acad.
Sci.. 51, 660-572, (1949)] and comparison of the affinities for the three eCG monoclonal antibodies were undertaken.
Prior to commencing ths Scatchard plot analysis, a preliminary examination of the valency of each monoclonal antibody was made to provide an important insight into the binding capacities of each antibody. This included an examination of monoclonal antibody isotypes and their ability to aggulutinate eCG coated SRBC.
Once the eCG reactivity of each monoc]onal antibody had been established, further experiments to classify the specificity of each were undertaken. All previously reported eCG antisera when tested have shown considerable cross-reactivity with the equine pituitary gonadotrophins LH
and FSH.
To e~amine the specificities of the three eCG
monoclonal antibodies, supernatants were examined by immunohistology ~or cross-reactive binding on sections of ~3~

placenta and pituitary from horses, sheep and man. Sections of human placental tissue previously shown to contain large quantities of hCG by immunohistology with a polyclonal hCG
antiserum, were used as representative sections of human placenta.
Materials and Methods Radio-Immunoprecipitation ofl25I eC5: 5ug of pur~
eCG ~UCB Bioproducts, Belgium) was reacted with 0.5mCi 125I
(lOOmg/ml 125I, Amersham) using the chloramine T method.
0.5ml samples of each eCG monoclonal supernatant and one non-associated negative control supernatant were titrated by doubling dilution from neat to 1:128 in PBS~Az. 30,000 cpm 125I eCG in lOOul PBS/Az containing 1.25% normal BALB/c mouse serum, was added to each ~ube and incubated for 2.5 hours at room temperature (20-25C). lOul of RAM was added and incubated for a further 1 hour at room temperature.
Supernatants were removed by centrifugation (3000g for 30 minutes at 4C) and the immunoprecipitates counted using a Packard~gamma counter (Model 500C Packard Illinois, USA).
S~andard curve construction: 30ug of pure eCG ~UCB, Belgium) was labelled with lmCi 125I by enzymatic (lactopero~idase~ iodination using Enzymobeads~Bio-Rad) as desc~ibed in Bio-Rad Technical Bulletin No~ 1071. Iodinated ~CG was separated from reagent beads and free 125I by G-50 Sephade~ gel-filtration as described for the chloramine T
iodination m~thod. Antibody supernatants were concentrated (10~) using an Amicon~ Diaflo filtration system (Model 202 and Amicon PM30 filter (Amicon, Massachusetts, USA).
Several separation schemes incorporating Pro-tein A
Sepharose~ (Pharmacia), R~M and polyethylene glycol were investigated and the scheme giving the best results and repeatability used for the construction of the antihody dilution curves.
0.5ml aliquots of 10~ concentrated supernatants were titrated by doubling dilution from 5x conc. to O.OOOl9x conc. in PBS/Az containing 2% foetal cal~ serum ~FCS). The FCS was added to reduce background in tubes containing low concentrations of antibody due to non-specific adherenc~ of 125I eCG to the walls of the 3DT tubes. 55,000cpm 125I
eCG in lOOul P~S/Az, 2% FCS, was added to each 3DT tube and incubated for 18 hours at 4C with constant shaking on a Micro-shaker ~Cat. No. AM69, Dynatech, England). 50ul of RAM
was then added, mixed and incubated for a further hour at room temperature. 5ul of normal mouse serum was added to increase the volume of precipitate 1 minute prior to precipitate collection by centrifugation (3500g for 30 minutes at 4C). The volume of normal mouse serum added had previously been established by tikration as sufficient to increase the precipitate bulk, and insufficient to saturate the quantity of second antibody used. Supernatants were removed by aspiration and the precipitate washed by resuspension in 2ml PBS/Az and centrifuged as above. The precipitates were then counted using a Packard gamma counter.
The concentration of each monoclonal antibody capable of binding appro~imately 66% of the ma~imal precipitable counts was considered suitable for use in the construction of standard curves. The maximum percentage of the total counts precipitated ranged from 32% to 38% of counts added. Higher percentages were not precipitated and reasons for this include damage to the trace during iodination and/or storage, and, heterogeneity of the l25I
labelled eCG.
Duplicate samples of standard assay blanks and negative con-trols were run for each monoclonal antibody. The eCG standard concentrations ranged from 4.6iu/ml to lO,OOOiu/ml, and were prepared by dissolving eCG supplied as "Folligon Serum Gonadotrophin" (batch 376551 Intervet, Holland~, in PBS/Az containing 2% FCS. lOOul aliquots of each standard were dispensed into 3DT tubes and then 55,000 cpm l25I eCG in lOOul of PBS/Az containing 2% FCS was added per tube. 0.5ml volumes of each suitably diluted monoclonal antibody were added and incubated with shaking at 4C
overnight. Immune comple~es were immunoprecipitated using RAM and washed as described for the construction of the antibody dilution curves.
EX~MPLE 2 The extraction of purified, biologically active eCG
from crude e~tracts of pregnant mare serum by affinity chromatography confirms the specificity of the antibody and allows accurate assessment of the biological properties of - 3~ -~3~

purified eCG.
(i) MATERIALS AND METHODS
Preparation of affinitY columns:
Approximately 200mg of each monoclonal antibody was purified from ascites fluid and coupled to Affigel 10 at 16mg per ml of gel. It was established that 89% of monoclonal antibody F50-37, and 94% of F51-99 coupled with the gel during the two reactions. Three columns were prepared for each affinity gel preparation, one of llml packed gel volume in a reverse flow column (K 16/20 Pharmacia, NSW) and two of 0.5ml packed gel volumes in lml disposable syringes (Terumo, Melbourne). The two small columns were used in establishing optimal conditions for the binding and elution of antigen.
The procedures used for determining elution and binding conditions for each affinity column were the same and consequently the method reported here is only for the monoclonal antibody F50-37 affinity column. The optimal conditions established for the F51-99 affinity column are presented in the results sectionO
Establishment of Elution and Loading Conditions for the F50-37 affinity column:
Iodinated eCG was loaded onto a 0.5ml affinity column in PBS/Az. The column was then washed with the same buffer to remove unbound material from the column. When wash effluents had radioactivity counts of background levels, stepwise elutions of the column with 0.lM citric acid solutions ranging in pH from 6.75 to 2.0 werQ collected as 2ml fractions.
The procedure was repeated with the second 0.5ml column onto which 100iu eCG (Folligon Serum Gonadotrophin batch 376551, Intervet, Holland) had been loaded in the same buffer. Fractions collected were diluted 1:10 and measured for eCG activity using the MIP assay.
OPtimisation of eCG bindinq to the F50-37 affinit~ column:
Various buffers including phosphate, Tris and citrate-phosphate were examined at 0.5M and 0.lM
concentrations over a wide range of pH. Buffer type and molarity had minimal effect on the binding of eCG and the citrate-phosphate buffer was chosen for further investigation. A pH range of 3.5 - 8.0, was obtaîned by mixing various proportions of 0~lM citric acid and 0.2M
Na2HPO~.
0.5ml samples of each buffer containing approximately 100,000cpm I eCG were prepared in 3DT
tubes and 50ul of resuspended F50-37 affinity gel was added to each tube. After overnight incubation at 4C with shaking, supernatants were removed by aspiration. The gel was then washed twice with 3ml of the appropriate binding buffer, aspirated and counted. Gel samples were then washed with 0.1M citric acid pH 3.0 eluting buffer, and the supernatant collected for counting.
Once optimal binding and elution conditions had been established, a large scale trial was undertaken using the llml reverse flow column. A crude eCG extract binding solution containing 280,000iu eCG was prepared by dissolving the crude freeze-dried eCG sample in 150ml pH 6.5 citrate-phosphate loading buffer and then dialyzed against the same buffer overnight at 4C.
The column was equilibrated with loading buffer and the sample loaded at 5ml/hr at 4C. Unbound protein was washed rom the column with 50ml of loading buffer. The purified product was eluted by reverse flow elution with 0.lM
citric acid pH 3.0 at 5ml/hr (see Figure 13).
MIP and rat bioassay analyses of PotencY:
Samples of each affinity purified product and the crude extract starting material were e~amined for eCG potency by the MIP assay and rat bioassay.
SDS-PAGE analysis of the affinitY Purified eCG:
The eluted peak was pooled, dialysed against double distilled water and freeze dried. A lmg/ml solution of the final product was prepared for SDS-PAGE, MIP assay and rat bioassay analysis o purity and potency. SDS-PAGE analysis of the product included Coomassie blue staining of gels, and, amino black and immunostaining of Western transferred proteins, run as non-reduced/non-boiled, non-reduced/boiled and reduced/boiled samples.
tii) RESULTS:
OPtimization of eCG l ading and elution of the F50-37 affinity column:
The results for the elution of 125I eCG and MIP

~3~

assay positive eCG are shown in Table 1. Results for both experiments coincide in that ~H 3.0 0.lM citric acid elutes the bulk of eCG bound whether it is iodinated or non-iodinated from F50-37 affinity column gel.
Table 1 Effect of varying pH on the elution of iodinated eCG and eCG
detected using the MIP assay from 0.5ml F50-37 affinity columns using a 0.lM citric acid pH gradient as presented.

Elution BufferCPM 125I eCG in MIP Assay Sequential Fractions Detectable eCG
PBS Wash 2,940 PBS Wash 2,380 0.lM Citric Acid - pH 6.75 2,040 - pH 6.0 3,930 - pH 5.5 8,260 - pH 5.0 11,710 - pH 4.5 25,050 - pH 4.0 53,620 - pH 3.5 70,870 +
- pH 3.0 59,45~ +
- pH 2.5 11,220 - pH 2.0 4,510 PBS Wash 5,770 Optimal comple~ing of eCG to the F50-37 affinity column was obtained at pH 6.0 and above as shown in Table 2.

- ~1 -Table 2 The effect of varying p~ on the binding of 125I eCG to F50-37 affinity column gel.

pH of LoadingTotal Counts Counts Eluted with pH 3.0 Buffer Bound O.lM Citric Acid 3.5 20,935 5,910 .0 23,080 8,120 .5 29,515 15,190 5.0 32,030 19,095 5.5 3~,355 20,715 6.0 4~,915 2~,295 6.5 42,760 29,540 6.75 41,760 25,785 7.0 43,~00 24,725 7.5 45,445 28,390 8.0 47,060 24,460 The optimum loading and eluting buffers for the F50-37 monoclonal antibody affinity column were, O.lM citric acid - 0.2M ~a2HP04 pH 6.5, and O.lM citric acid pH 3.0 respectively.
The buffers suitable for the elution and binding of eCG for the F51-99 monoclonal antibody affinity column were, binding in PBS/Az and elution with O.lM NaCl, O.lM glycine adjusted to pH 2.0 with lM HCl. The elution of the purified eCG from this column required conditions considerably more harsh than those required for elution from the F50-37 monoclonal antibody affinity column. Consequently, the eCG
purified on the F50-37 column was chosen for use in Western transblotting e~periments. Sufficient monoclonal antibody F51-99 affinity purified eCG was prepared for SDS-PAGE
analysis and comparison with the eCG purified on the F50-37 affinity column (Figure 11), and, MIP and rat bioassay analyses of potency.
Monoclonal antibody F50-37 affinity chromatoqraPhr Purification of ~CG:
Of the 280,000 iu eCG loaded onto the columrl, 80,000iu failed to bind during the single passage through the column, and were subsequently detected by MIP assay in the ~ 3~

loading and wash fractions. The same procedure was repeated using the F51-99 affinity column.
MIP and rat bioassaY analysis o potency:
The speciic activities of (i) th0 crude e~tract, and (ii) both of the affinity column final purified products were 660iu/mg and 11,250 iu/mg respectively as measured by MIP assay. Bioassay results showed both of the final products had potencies ~ lO,OOOiu/mg.
Examination of purified eCG:
Affinity purified eCG preparations from the monoclonal antibody F50-37 and F51-99 affinity columns were indistinguishable by SDS-PAGE analysis (Figure 12) and, MIP, ELISA and rat bioassay analyses of potency. Coomassie blue staining of non-reduced/non-boiled samples run on SDS-PAGE
showed one diffuse band around Mr 66,000. This band was not present in the crude e~tracts run on SDS-PAGE (Figur~
13). The same band was immunoreactive on nitrocellulose following Western transfer with the three eGG monoclonal antibodies and showed no reactivity with the negative control monoclonal antibody (Figure 14).
Boiling in non-reducing bufer dissociated the Mr 66,000 band to give a very diffuse M~ 55,000 band and a diffuse Mr 22,000 band with three associated discret~ bands (Figure 13, Lane e). Immunostaining of Western transferred boiled/non-reduced samples differs for each monoclonal antibody (Figure 14). Monoclonal antibody F51-99 was unreactive with all subunits and fragments while F50-37 and ~50-75 stained heavily in the Mr 22,000 ragion. Monoclonal antibody F50-75 also gave strong staining in the Mr 36,000 region. Boiling samples for five minutes in reducing buffer gave similar Coomassie blue staining on SDS-PA OE gels to boiling in non-reducing buffer (Figure 13). All monoclonal antibodies were unreactive with reduced samples.
Monoclonal antibody affinity chromatography purification of immunologically and biologically active eCG
using the monoclonal antibodies F50-37 and F51-99 has confirmed their reactivity with eCG. Immunological reactivity of the purified eCG as shown by Western transfer and immunostaining with the remaining eCG monoclonal antibody, F50-75, also confirmed its reactivity with eCG.

; - - 43 -The high purity of the affinity purified eCG
products has been shown by, SDS-PAGE Coomassie blue staining of a single diffuse band of appro~imately Mr 66,000, and by the high biological activity of the purified eCG in both MIP
assay and rat bioassay. SDS-PAGE results, and MIP a~say and bioassay estimations of the specific activity for each affinity purified eCG preparation are consistent with those previously reported for purified eCG [Gospodarowicz, Endocrinology, 91, 101-106, (1971)]; Moore and Ward, (1980)]. The broadness of the bands appearing in the SDS-PAGE gels following Coomassie blue staining, is also consistent with previous findings [Moore and Ward, ~1980)].
Monoclonal Antibody F51-99 Affinity Purification of eCG from Plasma An initial experiment was undertaken to examine the effect of temperature on Affigel F51-99 gel extraction of eCG
directly from plasma. Three temperatures were compared:
4C, 20C and 37C.
750ml pregnant mare plasma ~0O1% azide and thiomersol was filtered using several Whatman Nol filter papers. 3 aliquots of 250ml plasma, each containing 3.5ml Affigel F51-99 gel (packed gel volume), were then rotated overnight at the different temperatures stated.
AftPr settling for 1 hour the majority of the plasma was syphoned off. The residual plasma was then swirled to resuspend the gel suspensions and a column then packed using each entire volume.
The three ~ffigel F51-99 columns were then washed e~tensively with PBS~Az to remove residual plasma proteins.
Each column was eluted using O.lM NaCl, O.lM Glycine pH 2.0 buffer and the eCG peak pooled. This solution was immediately buffered to pH 6.5 to avoid possible damage to the eCG due to prolonged exposure to this extreme pH.
The 3 separate eCG products were dialysed and assayed for comparison.
Results showed Affigel F51-99 gel absorption of eCG
from plasma serum to be as effective at 4C as at 37C
and/or 20C.
A large scale trial was then undertaken at 4C
incorporating 11 ml of Affigel F51-99 gel and 2.5 litre of ~3~

pregnant mare plasma containing 49 iu/ml eCGO After overnight rotational mi~ing, the gel was remo~ed, washed and eluted as above. The eCG product was assayed and showed a recovery of 120-140,000 iu total, effectively 100%. The plasma remaining after e~traction was negative in all eCG
assays. The eCG product was biologically active and of very high purity (8,000 iu~mg). This specific activity was increased to 12,500 iu/mg following a second Affigel F51-99 chromatography step which removed the contaminating fibrinogen.

Larqe scale affinity purification of eCG from serum_and çrude serum e~tracts ~sina AFFIPREP 10 immobilized monoclonal antibody F51-99 Affiprep310 is a more stable matri~ for immobilising proteins such as monoclonal a~tibodies or the production of affinity column gels. Appro~imately 200 mg of F51-99 was coupled to 20 mls of Af f iprep 10 as described by the manufacturer. This ~ffinity matris (Affiprep 10/51-99~ was then e~amined in 2 eCG purification trials.
Trial 1 A crude extract ~325 iu/mg) eCG preparation ~500,000 iu) was dissolved in 200 ml PBS~Az and mixed in a roller bottle overnight at 4C. The Affiprep 10~51-99 gel was collected into an affinity column, washed and eluted with O.lM NaCl, O.lM glycine pH 2Ø
The eluted peak was quickly buffered to pH 6.0, dialysed against distilled water and freeze dried.
ELISA and rat bioassay analyses of specific activity were consistent and gave activities of 11,600 and >10,000 iu 1 mg respectively. 420,000 iu was recovered in the purified product and 60,000 iu were detected in the residual binding solution.
Trial 2 The Affiprep 10/51-99 gel was reacted overnight with 2 litres of pregnant mare plasma containing 68 iu/ml as described in trial 1. 125-130,000 iu were recovered with a specific of appro~imately 5-6,000 iu/mg. This was increased to 10,000 iu/mg by a second column chromatography purification of the eCG using the same Affiprep 10/51-99 ~ - 45 -~6~

affinity gel.

Another trial was undertaken in sheep to compare the time of eCG administration, and the effect of antibody vs no antibody administration.
Three doses of affinity purified (Affiprep iO/51-99~
eCG were given at either (i) time of progesterone plug pull or (ii) 48 hours before plug pull (-48 hrs). Half of the sheep in each group received 1 mg of monoclonal antibody at 48 hours after plug pull while the other half received an injection of saline only.
Results Time of eCG Dose Antibody +~- No of C.L. No of Follicles "O" - 1.64 800 - 7.8 0.4 800 + 7.3 0.O
- 48 Hrs1200 - 10.6 107 1200 -~ 10.~ 3.0 1600 - 9.5 1.7 1600 -~ 10.0 1.0 800 - 3.3 1.0 800 + 4.~ 0 O Hrs1200 - 3.4 1.4 1200 + 6.0 0.75 1600 - 4.2 ~.2 _ 1600 + 6.8 1 6 The number of corpora lutea obtained at -48 hours was significantly higher for all groups than those receiving eCG at plug pull.
The numbers of unwanted large follicles in all groups were again low when compared with previous trials incorporating crude commercial eCG preparations. These were reduced even further by the administration of a single antibody dose at +48 hours.
The reduction in the number of large follicles in antibody treated ewes is also accompanied by a rise in the numbers of corpora lutea and therefor ovulations.
The use of the monoclonal antibody has increased the ~L3~
ovulation rate and reduced the number of undesirable large follicles.

Five dosPs of a commercially available eCG were compared with five equivalent doses of this material after purification on the Affiprep 10/Sl-99 affinity column.
Results Dose eCG Number of Corpora Lutea Number of Large Follicles Crude Pure Crude Pure "O" 1.~3 0.18 200 iu 1.75 1.55 0.25 0.75 400 iu 2.66 2.3~ 0.12 0.06 800 iu 4.38 2.95 1.00 0.27 1200 iu 6.76 6.25 3.37 1.75 1600 iu 6.61 7.75 8.45 2.69 Tha results show a greatly reduced number o~ large follicles in those ewes receiving the purified eCG in the higher dose range.
EX~MPLE 6 A trial was undertaken in pigs to examine whether purified eCG alone or purified eCG plus antibody F51-99 could improve litter si~e. Three groups of sows were compared.
Each group received 2 injections, one at time of weaning and one at the time of first mating.
Results No. of 1st 2nd Av. Litter % Still Av. Born Alive Litters Iniect. Inject. SizeBornLitter Size 72 Saline Saline 10.11 6.2~ 9.49 74 1500iu eCG Saline 10.917.1% 10.14 68 1500iu eCG 1 mg 51-99 10.87 5.8% 10.24 -Those Sow8 receiving both eCG and antibody on average gave birth to an e~tra 0.75 live piglets per litter.

~ This e~ample examines the ability of the monoclonal `'`"`' antibodies F50-37, F50-75 an~ F51-99 to block and/or modify the e$fect of 1600iu eCG in the ewe. Idfally an eCG
superovulation protocol incorporating the use of one or more of the eCG monoclonal antibodies would result in high numbers of ovulations with reduced numbers of post-ovulatory cystic follicles.
MATERIALS AND METHODS
Oestrus in Merino/Border Leicester cross ewes was synchronized by intravaginal insertion of progesterone sponges (Repromap 60mg MAP, Upjohn, Sydney) for 13 days.
Immediately prior to eCG injection of the ewes, the sponges were removed and the sheep given one of the following treatments.
Trial 1 Treatment (a) Intramuscular (IM~ injection of saline (b) IM injection of 1600iu eCG (Pregnicol~ Heriot Agencies, Melb.) in lml saline (c) IM eCG as in ~b) plus immediate IV
(intravenous) injection of (i) lmg antibody F50-37 or (ii~ lmg antibody F50-75 or (iii)lmg antibody F51-99, all in lml of saline.
Trial 2 Treatment (a~ IM injection of saline (b) IM in~ection of 1600iu eCG (as for Trial 1) (c) IM injection 1600iu eCG at plug pull (as in Trial 1) plus, 48 hours after eCG injection, an IM injection of (i) lmg antibody F50-75 or (ii) lmg antibody F51-99, each in lml of saline .
Ovarian response was assessed by mid-ventral laproscopy 7-8 days after oestrus. The number of corpora lutea and follicles were recorded.
In Trial 1, a significant block of eCG activity was found in the eCG treated sheep following the injection of monoclonal antibody F51-99 compared with sheep receiving eCG
only (Table 3)O The probability that the mean responses of eCG only, and eCG~F51-99 treated sheep were not different was Ø001 using the unpaired t-test (Statsworks program, - ~8 -Model 512KA computer, Macintosh Appl~. The mean number of o~ulations, represented by the mean number of corpa lutea (CL) was lower in sheep receiving both eCG and F51-99 than in sheep re~eiving saline only. No follicles were found with either treatment. In sheep receiving eCG only, eCG plus monoclonal antibody F50-37, and, eCG plus monoclonal antibody F50-75, typical superovulatory respons~s were observed.
Monoclonal antibody F50-75 may have partially inhibited the eCG superovulatory action but this did not appear significant.
Trial 2 results were consistent with Trial 1 (Table 3). Intramuscular injection of monoclonal antibody F50-75 48 hours after eCG injection failed to d~crease the superovulatory action of eCG. Sheep injected with mono~lonal antibody F51-99, 48 hours a~ter eCG injection, had considerably fewer follicles and a slightly reduced number of ovulations when compared to eCG only treated sheep. The difference in ovulation rates were not significant.
Monoclonal antibody F50-37 was e~cluded from the second trial due to the lack of activity shown in Trial 1 results.
Tabl~_~
Summary of the results for Trials 1 and 2 on the modification of eCG bioactivity in the ewe by monoclonal antibodies to eCG.

Dose of eCG Time of antibody Number of Mean Response Treatm~nt Ewes CL Follicles Trial 1 0 - 22 1.95 0 1600iu - 6 6.66 1.16 1600iu Immediate IV F50-37 5 5.4 1.2 1600iu Immediate IV F50-75 5 4.4 O.6 1600iu Immediate IV F51-99 5 1.4 0 Trial 2 0 - 7 1.14 0 1600iu - 8 6.38 2.62 1600iu 48 hrs IM F50-75 7 6.57 3.43 1600iu 48 hrs IM F51-99 7 4.43 0.43 CL = Corpora Lutea The results of both trials showed F51-99 to be an efEective monoclonal antibody for controlling eCG bioactivity 49 ~

in the ewe. Trial 1 results showed that lmg monoclonal antibody F51-99 when injected IV immediately after eCG
administration, totally abolished the superovulatory activity of 1600iu eCG in the ewe. Ewes receiving both eCG and monoclonal antibody F51-99, had fewer ovulations than those receiving no eCG at all. Monoclonal antibodies F50-37 and F50-75 had no effect under the same conditions.
Differences in the results of the two trials shows that variation in time and route of monoclonal antibody administration can affect the results obtained. A
significant difference is seen between eCG/F51-99 treated ewes in Trials 1 and 2. The probability that the means of each group were not different was Pr <0.05. By delaying the administration of the antibody in Trial 2, a substantial proportion of the desired eCG superovulatory activity is obtained with only minimal ovarian overstimulation. The mean number of unruptured follicles in this group was 0.43, while those receiving no antibody or monoclonal antibody F50-75 had mean numbers of 2.62 and 3.43 follicles per ews respectively. This difference was not statistically significant due to the small numbers of sh0ep used, and the distribution of data within each group. The decreased number of ovulations associated with the decrease in unruptured follicle numbers in F51-99 treated sheep, was an undesirable effect but was not statistically significant. It is likely however, that with better timing and optimisation of the amount of the antibody administered, the desired reduction in numbers of unruptured follicles may be obtained without a decrease in the number of ovulations.
Coinciding with decreased numbers of unruptured follicles should be an expected decrease in oestrogen levels. A similar return to physiological oestrogen levels in sheep, may permit natural mating for embryo transfer programs. To date natural matings of superovulated sheep have produced poor yields of embryos, which is probably the result of decreased sperm mobility/penetration into the uterus [Killeen, Proc. Symp Embryo Transfer in Cattle, Sheep and Goats. Published ASRB (1982~]. As non-surgical artificial insemination is extremely difficult in sheep due to the constriction of the cervix, surgical or laproscopic :~3~

insemination of the uterine horns has been adopted with good results [Trounson and Moore, Aust. J. Biol. Sci., 27, 301-309, ~1974)]. Studies have commenced to axamine the possibility of natural mating following superovulation with eCG when controlled by the monoclonal antibody, F51-99.
Trial Three A further trial was conducted to analyse the effect of antibody dose rate on ovarian response. The results are set out in Table 4. Sheep received 1600 iu eCG
at sponge withdrawal and the various antibody doses were administered 48 hours later.

SheeP Trial n=12 (No. of ewes I mean CL (corpus I mean follicles per qroup I iutea)¦
Hiqh dose 3mq j 5.9 L 3.1 Mid dose 0.75mq I 6.2 I 1.0 Low dose 0.25ma I 5.8 I 2.6 No Dose "0" antibody I
CONTROL I 6.0 I 3.45 The results show the need for optimization of antibody dosage.

BOVINE DATA
Trials have been undertaken in cows superovulated with eCG. The use of the monoclonal antibody F51-99, 72 and 96 hours after eCG injection has demonstrated that the benefits resulting from the use of this monoclonal antibody usage in the ewe are also applicable in cattle.

Cattle Trial Results I eCG Qa~
Overall ¦ Ovulation rat~ Follicles means ¦ 14.25 3.0 ¦ n = 12 ¦ eCG & F51~99 (72 hrs) Overall ¦ Ovulation rate Follicles means 1 4.5 1.8 I n = 12 ¦ eCG & F51-99 (96hrs) Overall ¦ Ovulation rate Follicles means 1 10.1 1.7 n = ll TWI NM I NG DATA
Depending on the batch o eCG used, 200-800 iu eCG
injected at the time of sponge removal and a subsequent injection of 0.1 - 0.5 mg of antibody will induce a higher number of twin births than in non-eCG treated or eCG-non antibody treated sheep. eCG may also be administered prior to sponge removal. The important fact irrespective of the regime chosen to induce double ovulation is that low doses of PMSG and a subsequent dose of antibody can effectiv21y incr~ase the incidence of twin births in sheep and other live stock including cattle and goats.
EWE DATA
SPONGE 13 days. B00 iu at time of spongs removal.
48 hours later either (i) saline injection ~1 ml) (ii~lml 0.5 mg/ml F51-99 in Sal.

Ouads Triplet$ Twins Single None eCG only l 4 1 0 3 eCG & F51-99 0 1 ~ _ 4 .., ~,: .
~ ~ - 52 -SUPPLEMENT~RY DISCL~S~RB

Trials have been undertaken in cows superovulated with pure eCG purified by monoclonal antibody affinity chromatography using the monoclonal antibody F51-99.
The use o~ di~ferent doses demonstrated th~t pura eCG
can be used at much lower concentratlons to achieve superovulation comparQd to commercially available preparations. The use of low doses o~ eCG rasulted in no overstimulation o~ ovaries, a serious problem with the presently available commercial preparations of eCG
used for superovulation.
Hereford Cattle. 3-4 years o~ age, 2nd or 3rd calving.
Averaae Liveweiqht 334-371 Kq (ranqe) Treab~ No. O~'n R~e O.I.** C3~s on ~ted Ra~e(Mean) (Mean3 O~y ~p~L~Ve A 6000 iu 8 1.8 1-3 4.0 P~ - _ All ~igni~icantly B 3000 iu 8 6O0 1-16 3.7 ov~imLlated pure 5 - NonE~
C 1500 i.u. 8 7.~ 2-1~ 2.0 3- Signi~cantly ov~inula,tion 6 - Nonæ~
D 750 8 8.0 1-18 0.7 2 - Slight over-st~ on 6 - Sigmfi ~ tly over E 300 8 9.0 1-19 l.9 st~ation ~llinq~n* 2 - ~onE~
-6 - Non~
F ~ Lne 8 1.0 1-l 0.3 2 - Slight contr~1 ower-st~a~on *.~olligon - commercially available preparation (Intervek, Australia) ** O.I. re~ers to Overstimulation Index "0" normal, "5" ovaries as large as tennis balls.
Comments Use of pure eCG 750 iu gives same superovulation response as 3000 iu of khe commercially available preparation Folligon. No overstimulation of ovaries ~3~

observed with pure eCG at 750 iu. Use of 3000 iu of Folligon produced significant overstimulation.
It is therefore apparent that much lower amounts of eCG may ba used when purified according to the process o~ the present invention to achieve same number of ovulations with a near complate absence of overstimulation compared to commercially available productsO
Finally, it is to be understood that various other modifications and/or alterations may be made without departing from the spirit of the present invention as outlined herein.

~,~

Claims (23)

1. A process for producing a monoclonal antibody against equine chorionic gonadotrophin (eCG) which method comprises providing a B cell capable of producing antibodies against equine chorionic gonadotrophin, and a myeloma cell, fusing the B cell with the myeloma cell, detecting antibody producing hybridomas, which have specificity for biologically active eCG wherein said antibody is substantially unreactive with tissue isolated from human or equine pituitaries and inactive subunits and fragments of eCG, utilising an immunohistology assay on Western immunoblotting technique, subsequently cloning the fused cell hybrid selected into individual antibody producing cell lines, propagating each cell line in-vivo or in-vitro and harvesting an antibody produced by said hybridoma.

2. A process according to claim 2 wherein the B cell capable of producing antibodies is obtained from an animal immunized with equine chorionic gonadotrophin or an immunogenic fragment thereof; and the myeloma cell line is an NS-I type myeloma cell line.

3. A process according to claim 2 wherein the fusion step comprises mixing somatic cells with myeloma cells in the presence of a fusion promoting agent selected from the group consisting of Sendai virus and polyethylene glycol.

4. A hybridoma us cell line which produces a monoclonal antibody against equine chorionic gonadotrophin, comprising a hybridoma formed by fusing a myeloma cell with a B cell capable of producing antibodies against equine chorionic gonadotrophin wherein said antibodies are specific for biologically active eCC and are substantially unreactive with tissue isolated from human and eguine pituitarias and inactive subunits and fragments of eCG.

5. A hybridoma cell line according to claim 4 formed by fusing an NS-1 type myeloma cell with a spleen B cell obtained from a BALB/c mouse immunized with equine chorionic gonadotrophin.

6. A hybridoma cell line, comprising hybridoma cell line F51-99 having an accession number given by the European Collection of Animal Cell Cultures, an international depository authority, of 87100204.

7. A monoclonal antibody against equine chorionic gonadotrophin (eCG) formed by a hybridoma cell line, said monoclonal antibody comprising an antibody which is specific for biologically active eCG and substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG, and wherein the cell line is a hybridoma formed by fusing a myeloma cell with a B cell capable of producing antibodies against biologically active equine chorionic gonadotrophin which are substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG.

8. A monoclonal antibody according to claim 7 wherein the hybridoma cell line is hybridoma cell line F51-99.

9. A method of purifying equine chorionic gonadotrophin (eCG) wherein said method comprises:
providing a source of impure eCG and a monoclonal antibody against equine chorionic gonadotrophin formed by a hybridoma cell line, wherein said antibody is attached to a solid support to form an antibody matrix, wherein said monoclonal antibody is specific for biologically active eCG and substantially unreactive with tissues isolated from human or equine pituitaries and inactive subunits and fragments of eCG;
passing the impure eCG through the antibody matrix to capture the eCG; and eluting the captured eCG from the antibody matrix.

10. A method according to claim 9 wherein the matrix is an affinity chromatographic gel.

11. A method according to claim 10 wherein the antibody matrix support is in a chromatography column or in a large scale container or vat which may be rotated, agitated or stirred.

12. A method according to claim 9 wherein the monoclonal antibody is formed by the hybridoma cell line F51-99.

13. A method according to claim 12 wherein the impure equine chorionic gonadotrophin source is a buffered solution of pregnant mare plasma or sera at a pH of approximately 6.0 to 8Ø

14. A method according to claim 13 wherein the elution step is conducted utilising a sodium chloride-glycine solution buffered to a pH of approximately 2.0 to 3Ø

15. A veterinary composition including an effective amount of a monoclonal antibody against equine chorionic gonadotrophin formed by a hybridoma cell line, said monoclonal antibody comprising an antibody which is specific for biologically active eCG
and substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG, and wherein the cell line is the hybridoma formed by fusing a myeloma cell with a B cell capable of producing antibodies against biologically active equine chorionic gonadotrophin which are substantially unreactive with tissue isolated from human and equine pituitaries and inactive subunits and fragments of eCG; and a veterinarily acceptable diluent, carrier or excipient therefor.

16. A veterinary composition according to claim 15 further including an effective amount of a purified equine chorionic gonadotrophin produced by a method comprising purifying equine chorionic gonadotrophin (eCG) wherein said method comprises:
providing a source of impure eCG and a monoclonal antibody against equine chorionic gonadotrophin formed by a hybridoma cell line, wherein said antibody is attached to a solid support to form an antibody matrix, wherein said monoclonal antibody is specific for biologically active eCG and substantially unreactive with tissues isolated from human or equine pituitaries and inactive subunits and fragments of eCG;
passing the impure eCG through the antibody matrix to capture the eCG; and eluting the captured eCG from the antibody matrix.

17. The use of purified equine chorionic gonadotrophin made in accordance with a method of claim 9 in treatment of female animals to induce oestrus and/or to increase fecundity.

18. The use of monoclonal antibodies of claim 7 in a treatment of female animals for inducing oestrus and/or increasing fecundity.

19. The use of monoclonal antibodies of claim 7 simultaneaously with purified equine chorionic gonadotrophin made in accordance with a method of claim 9 in treatment of female animals to induce oestrus and/or increase fecundity.

20. The use of purified equine chorionic gonadotrophin in accordance with claim 17 to increase number of ova produced during a single oestrus cycle in female animals.

21. The use of the monoclonal antibodies of claim 7 in combination with or in conjunction with the use of purified equine chorionic gonadotrophin in accordance with claim 17.

22. The use of purified equine chorionic gonadotrophin in accordance with claim 17 to increase ovulation to provide twinning in female animals.

23. The use of monoclonal antibodies of claim 7 in combination with or in conjunction with the use of purified equine chorionic gonadotrophin in accordance with claim 17.