One important group of biologically active steroids is the oestrogens.

As is well known, the oestrogens are a class of hormones concerned principally, although not exclusively, with reproductive processes in the female and the development of secondary female characteristics.

Oestrogenic hormones are known to exert effects in a large number of different tissues but their main physiological action is generally considered to be stimulation of growth and development and maintenance of the female reproductive organs, in particular the uterus, vagina and mammary glands.

The biological properties generally regarded as being indicators of oestrogenic activity are the ability to produce oestrus and vaginal cornification in spayed rats and mice (see Textbook of Physiology and Biochemistry, 9th Edition, George H.Bell etal., Churchill Livingstone, 1976, p651).

Mammalian oestrogen hormones are steroid hormones characterised by the possession of an aromatic A-ring bearing a phenolic hydroxyl group as shown below.

The principal hormone in female humans is oestradiol (R, = OH, R2 = H), the other main oestrogen hormones naturally occurring in humans being oestrone (R1 = carbonyl oxygen, R2 = H) and oestriol (R1 = R2 = OH).

Oestrogenic activity is also displayed by numerous compounds having a non-steroidal structure, examples of which are diethylstilboestrol and hexoestrol. Many of the synthetic oestrogens share the common feature of an aryl ring bearing a phenolic hydroxyl group as found in oestradiol and related human oestrogens.

It is known (see Mechanisms of Hormone Action - C. R. Austin and R. V. Short, Cambridge University Press, 1979, ppl 57-184) thatoestrogenic activity is mediated by the binding of the oestrogen to oestrogen receptors, and compounds have been found which compete with active oestrogens for binding at the receptor sites and therefore inhibit or block oestrogenic activity at the receptors. Such compounds, which are only weakly active themselves, are generally referred to as anti-oestrogens. Examples of anti- oestrogens include clomiphene, nafoxidine and tamoxifen. Some anti- oestrogens, such as tamoxifen, have been used to treat oestrogen- dependent tumours, but a problem with such compounds is their toxicity.

An object of the present invention is to provide novel steroidal antagonists or agonists which are less toxic than existing steroidal antagonists or agonists. A particular object of the invention is to provide compounds which are anti-oestrogenic, but which are less toxic, but not less active, than known therapeutic anti-oestrogens.

Oestrogens such as oestradiol and oestrone are metabolised through the hydroxylation of the phenolic A-ring to give a catechol A-ring in which hydroxyl groups are present at both the carbon 2 and carbon 3, or the carbon 3 and carbon 4 positions. The hydroxylation reaction, which is catalysed by the 2-hydroxylase enzyme or the 4-hydroxylase enzyme, is followed by rapid methylation by the O-methyltransferase enzyme which terminates its biological function. The methylated catechol oestrogen is shed from the receptor site and is then excreted as a sulphate or glucuronate.

It has now been found that by complexing the catechol moiety of the catechol oestrogen with molybdenum, the ability of the catechol to bind at the receptor is unimpaired. Indeed, its biological half life on the receptor is lengthened, probably as a consequence of methylation being inhibited.

Moreover, it has been found that the molybdenum catechol oestrogen competes effectively with oestradiol for binding at the receptor. In other words, the molybdenum-catechol oestrogen complex is a potent anti- oestrogen.

On the basis of work carried out to date, it is contemplated that the substitution of the molybdenum with other complex forming elements or molecules such as transition metals will also result in the formation of anti- oestrogens. It is also contemplated that by substituting the catechol oestrogen by another ligand which is capable of binding to other steroid receptors, antagonists or agonists of such steroid receptors will be formed.

Molybdenum is known to exhibit a number of important biological properties and has been recognised as an important trace element. The biological importance of molybdenum arises at least in part from the fact that it forms part of a complex required as a co-factor for the enzymes xanthine oxidase, aldehyde oxidase and sulphite oxidase; see V.M. Sardesai, Nutr.

Clin. Pract., 8(6): 277-81, 1993, R. Hille and H. Sprecher, J. Biol. Chem.

1987 262(23): 10914-7; and C.A. Pritsos and D.L. Gustafson, Oncol. Res.

1994; 6(10-11): 477-81.

It has also been suggested that molybdenum, or molybdenum compounds, bind to and/or stabilise oestrogen receptors; see for example S.M. Hyder and J.L Witliff, J. Chromatogr. 1988, 444: 225-37; and R. Pasic etna!. Clin. Chem. 1989, 35(12): 2317-9.

Deficiencies in molybdenum have been shown to result in the development of conditions such as cancers. For example, in Sardesai, idem, it is disclosed that studies suggest that molybdenum deficiency may be a factor in the higher incidence of oesophageal cancers in populations consuming food grown in molybdenum poor soil. There have also been reports that supplementing the diet with molybdenum can reduce the incidence of cancers, see for example, H.J. Wei metal. Chung-Hua-Chung-Liu- Tsa-Chih, 1987, May; 9(3): 204-207; and C.D. Seaborn and S.P. Yang, Biol.

Trace Elem. Res. 1993, Nov-Dec; 39(2-3): 245-56.

Antitumour activities have been reported for a number of molybdenum compounds and complexes. In F.T. Garzon metal., Cancer Lett. 1987,34(3): 325-30, the biological effects of the molybdenum complex dichlorobis (1- <BR> <BR> <BR> <BR> phenylbutane- 1 ,3-dionato)molybdenum (IV) ( IV) are described. H. Fujita et al., J. Exp. Med. 1992, 168(2): 421-6 describe the antitumour activity of inorganic polyoxomolybdates, and the antitumour activities of such compounds are also disclosed in EP-A-O 412 158. The use of transition metals such as molybdenum in the treatment of cancers is also disclosed in P. Kopf-Maier, Eur. J. Clin. Pharmacol. 1994; 47(1): 1-16.

However, so far as the applicants are aware, there are no previous disclosures relating to the ability of molybdenum and other transition metals and Lewis acids to form complexes with catechol oestrogens, and the anti- oestrogenic properties of such complexes.

Accordingly, in a first aspect, the invention provides a steroidal antagonist or agonist compound which is a complex formed from a transition metal, or a non-transition metal Lewis acid, and a receptor-binding ligand capable of binding to a steroid receptor.

The steroid receptor can be an oestrogen receptor.

The transition metal, or non-transition metal Lewis acid, is typically one which is pharmaceutically acceptable. It will usually be non-toxic, although a degree of toxicity may be acceptable when using the complexes to treat life threatening conditions such as tumours. The transition metal, or non-transition metal Lewis acid, is preferably one which forms a complex with cis diol or catechol groups.

It is presently preferred that the transition metal is molybdenum, but examples of alternative transition metals include vanadium and chromium.

An example of a non-transition metal Lewis acid is boron (III) which is known to form borate and boronate complexes with cis diols, and has been found by the present applicants to form a complex with 2-hydroxy oestrone.

The receptor-binding ligand can be a steroidal or non-steroidal structure provided that, when complexed with the transition metal, it has the ability to bind to steroid receptors. However, excluded from the definition of the receptor-binding ligand as used herein is the 1-phenylbutane-1,3- dionato ligand disclosed in F.T. Garzon idem, whether or not the said ligand has steroid receptor-binding activity.

Most preferably, the steroid receptor binding ligand is an organic ligand, and it can be, for example, a catechol-containing moiety.

The catechol-containing moiety can be, for example, a catechol oestrogen compound.

In one particular embodiment, the invention provides an anti- oestrogenic compound which is a complex formed from molybdenum and a catechol oestrogen compound.

The term "catechol oestrogen" as used herein means an analogue of an oestrogen compound wherein, in said analogue, an aromatic ring forming part of the skeleton of the oestrogen compound has an ortho-dihydroxy grouping. The catechol oestrogen itself may or may not have oestrogenic activity but it must have the ability to bind to oestrogen receptors.

The term oestrogen compound as used herein refers to both naturally occurring oestrogens and also synthetic oestrogens. Naturally occurring oestrogens include oestrogens having a steroidal structure such as oestradiol, oestrone and oestriol, and derivatives thereof, e.g. esters such as oestradiol 1 7P-esters. The term oestrogen compound as used herein also include synthetic oestrogens, of both steroidal and non-steroidal structure.

Examples of synthetic steroidal oestrogens include moxestrol (see US Patent No. 3579545), equilin (see US Patent No. 2221340) and ethynyl oestradiol (see US Patents Nos. 2243887 and 2251939). Examples of synthetic non- steroidal oestrogens include oestrogen compounds comprising a pair of aryl rings spaced apart by two planar carbon atoms, such as oestrogens based on the stilbene skeleton e.g. diethylstilboestrol.

In the case of the steroidal oestrogens, the catechol oestrogen typically is a compound in which the A-ring of the steroidal skeleton has either a 2,3-dihydroxy or a 3,4-dihydroxy substitution pattern. In the case of a catechol oestrogen derived from a non-steroidal oestrogen compound, the catechol moiety may be an ortho-dihydroxy aryl ring which is sterically equivalent to the A-ring of a steroidal oestrogen. For example, in the case of a catechol oestrogen derived from stilboestrol or dienoestrol, and like diphenyl oestrogens, one of the hydroxyphenyl groups may be replaced by an ortho-dihydroxypheny group, e.g. a 3,4-dihydroxyphenyl group.

In a preferred aspect, the invention provides an anti-oestrogenic compound which is a complex formed from a transition metal such as molybdenum and a catechol oestrogen wherein the catechol oestrogen is derived from a steroidal oestrogen and has a 2,3-dihydroxy or 3,4-dihydroxy substitution pattern in the steroidal A-ring.

The steroidal oestrogen from which the catechol oestrogen is derived is preferably selected from oestradiol, oestrone and oestriol.

The transition-metal/catechol-containing moiety complex (e.g. a molybdenum-catechol oestrogen complex) can contain one catechol- containing moiety per atom of transition metal (e.g. molybdenum) or it can contain more than one. Preferably the complex contains two catechol- containing moieties per transition metal (e.g. molybdenum) atom.

In addition to the catechol oestrogen(s), the complex may contain other ligands bound to the transition metal. Such ligands can be any pharmaceutically acceptable ligand capable of binding to the transition metal.

In addition, the invention provides compounds wherein any or all of the ligands are non-pharmaceutically acceptable, for use as intermediates in the preparation of compounds wherein the ligands are pharmaceutically acceptable. The ligands are preferably oxygen, nitrogen or sulphur ligands.

Adjacent ligands may be linked, ie they may form part of bidentate or polydentate ligands. For example, where two or more adjacent ligands are oxygen ligands, they may form part of a 1,2-diol, 1,2-dicarboxylate or 1- hydroxy-2-carboxylate unit. Alternatively, an adjacent pair of oxyligands may form part of a further transition metal oxide unit, e.g. another molybdate group.

When the transition metal atom in the complexes of the present invention is molybdenum, typically it is present in the Mo(VI) oxidation state.

One preferred compound of the present invention is represented by the formula M2[E2MoO2] or M'[E2MoO2] wherein M is a monovalent cation, M' is a divalent cation and E is a group having the formula: wherein R3 is a carbonyi oxygen or a hydroxy group, and R4 is hydrogen, or R3 and R4 are both hydroxy.

In a further aspect, the invention provides a steroid antagonist or agonist including, for example, an anti-oestrogenic compound as hereinbefore defined, for use in medicine, for example in the prevention or treatment of steroid dependent disorders including, for example, oestrogen- dependent tumours.

In another aspect, the invention provides the use of an anti- oestrogenic compound as hereinbefore defined for the manufacture of a medicament for the prevention and treatment of oestrogen-dependent tumours.

In a still further aspect, the invention provides a pharmaceutical composition comprising a steroidal antagonist or agonist compound as hereinbefore defined, and a pharmaceutically acceptable carrier.

The invention also provides a method of inhibiting or blocking a biological effect induced or stimulated by a steroid in a mammal, which method comprises administering an effective steroid receptor blocking amount of a compound as hereinbefore defined.

The invention also provides a method of blocking a steroid receptor (e.g. an oestrogen receptor) in a mammal, which comprises administering to the mammal a therapeutic steroid (e.g. oestrogen) receptor blocking amount of a compound as hereinbefore defined.

In a still further aspect, the invention provides a method of treating an oestrogen-dependent tumour in a mammal, which method comprises administering to the mammal a therapeutically effective non-toxic amount of an anti-oestrogenic compound as hereinbefore defined.

The steroidal antagonist or antagonist compounds of the present invention are preferably administered preformed to the patient. However, they can be generated in vivo in the patient, e.g. by co-administering the catechol oestrogen and the molybdenum, separately or together.

Alternatively, molybdenum may be administered in a form effective to generate molybdenum-catechol oestrogen complex from endogenous catechol oestrogens.

The compounds of the present invention can be prepared from an appropriate catechol-group-containing moiety (e.g. a catechol oestrogen and a suitable transition metal (e.g. molybdenum) compound (e.g. an alkali metal molybdate such as sodium molybdate) in accordance with standard methods. Catechol oestrogens such as 2-hydroxy oestrone are commercially available or can readily be synthesised. Where it is desired to form a complex in which a single molybdenum atom is complexed by two catechol oestrogen moieties, the ratio of the reactants will be adjusted accordingly.

The reaction between the transition metal compound (e.g. molybdate) and the catechol-containing moiety (e.g. the oestrogen catechol compound) is typically carried out in a suitable polar solvent such as water or aqueous ethanol. In the case of molybdenum, formation of the catechol oestrogen molybdenum complex is characterised by the appearance of a brown colour in the solution. Following formation of the complex, the solvent may be removed by known methods to give the complex in solid form; for example where the solvent is water, the solvent may be removed by freeze-drying.

The compounds of the present invention can be formulated in accordance with known techniques, and may be presented in solid, semi- solid or liquid form. Thus for example they can be presented in the form of a tablet, lozenge, pill, capsule, powder, or other solid dosage form.

Alternatively, they can be presented as a solution for injection, or as a dry powder for reconstituting into solution form for injection. Alternatively, the compound may be presented for delivery transdermally or transmucosally, for example by administration to the nasal, bronchial, or rectal mucosa. It is preferred that the compounds are formulated in such a way as to avoid contact with the gastric juices. Thus, where the compounds are administered orally, they may be encased or encapsulated in a release delaying coating which releases the compound atan appropriate location in the intestinal tract. Release delaying agents suitable for such purpose are well known and need not be described in detail here.

Certain compounds of the invention have been found to be sensitive to light and/or oxygen, storage of the compound in the presence of these factors leading to decomposition and loss of activity over a period.

Accordingly, the compounds of the invention can be formulated and/or packaged in a manner which protects them from light or contact with oxygen. For example, the compounds can be presented in tablet form in hermetically sealed blister packs. Alternatively, if it is preferred to present the compounds in liquid or semi-solid form, for example, they can be formulated in capsules, the capsules being rendered opaque to prevent light induced degradation. The compositions can also contain pharmaceutically acceptable anti-oxidant compounds such as alkylated hydroxytoluenes (e.g.BHT) to prevent or hinder oxidation of the compounds of the invention.

The compounds of the invention will be administered in amounts which are effective to provide the desired therapeutic effect. Such amounts will vary according to the nature and severity of the disease or condition being treated and ultimately will be at the discretion of the physician. For life threatening diseases, the compounds may be administered where required at concentrations which may cause some toxic effects. However, it is generally preferred that the compounds of the invention are administered at concentrations which are non-toxic. It is envisaged that an advantage of the compounds of the present invention is their reduced toxicity compared to known anti-oestrogens.

The invention will now be illustrated by reference to the foilowing non-limiting examples.

EXAMPLE I Preparation of Molvbdate 2-Hvdroxv-oestrone Complex ("Humphrin") Two solutions were prepared; Solution A consisting of sodium molybdate (84.4mg Na2MoO4.2H2O) dissolved in water (100ml) and Solution B consisting of 2-hydroxy-oestrone (5mg) dissolved in ethanol (2.5ml). 1 ml of Solution A was mixed with 1 ml of Solution B giving a molar ratio of Na2MoO4.2H2O:2-hydroxy-oestrone of 1:2. A brown colour indicative of the formation of a molybdate catechol oestrogen complex developed almost immediately. The reaction was followed spectrophotometrically, and a characteristic absorption plateau at 400-460 nm was observed indicating complex formation. The complex, kept in the dark, is stable for a minimum of 28 h (the maximum time that stability was tested for). The stability constant for the molybdenum-catechol oestrogen complex is the same as for other molybdenum pyrocatechol complexes, i.e.

they have Log Dissociation constants of 5.4.

EXAMPLE 2 Anti-Oestroqenic Activitv of the 2-Hvdroxv-oestrone-Molvbdate Complex Rat Uterus Studies The rat uterus was used as the target organ for determining the oestrogenic effect of the complex. The first effect of oestrogen on the uterus is water imbibition, so that the fresh weight of the uterus is a direct measure of the oestrogenic effect of any oestrogen injected into the rat (see Wotiz et al. (1978) Cancer Research, 38. 4012-4020). A single injection of oestradiol (see Clark et al. (1977) Endocrinology, 100 [No. 1], 91-96), or repeated injections (Wotiz et al.) as used in this example, will produce a maximum uterine response (which will differ according to the protocol used) in terms of water uptake, ie fresh weight of the uterus is a measure of oestrogenic response. The 2-hydroxy-oestrone; as Wotiz used, will not produce any oestrogenic effect if one injection is given but, if it is injected every 3 hours for 21 hours, will produce almost the same response, at 24 hours, as the maximum response given by oestradiol.

Groups of rats were injected either with vehicle (50% ethanol:50% water,50,u1) as a control, with oestradiol (0.5 pig oestradiol in 50pal vehicle) every three hours for 21 hours, or with 2-hydroxy-oestrone (50 pug 2- hydroxy-oestrone in 50,us vehicle), or with molybdate-2-hydroxy-oestrone complex (50 pg 2-hydroxy-oestrone with 8.4 pg Mo in 50,us vehicle) at the same time intervals. The results are shown in Figure 1. As expected, oestradiol gave a huge response in fresh weight of the uterus. The 2- hydroxy-oestrone injected every three hours for 21 hours gave a very similar increase, slightly less but not statistically different from the oestradiol effect.

This confirms earlier work. However, when the molybdate-2-hydroxy- oestrone complex was injected at the same concentration of oestrogen as the 2-hydroxy-oestrone, there was no statistically significant increase at all: the fresh weight of the uterus was the same as the control. This could however simply mean that the complex was not binding to the oestrogen receptor, and therefore a further experiment was undertaken to determine its receptor-binding properties.

It is known from Lan & Katzenellenbogen (1976; Endocrinology, [No.1], 220-227) that oestrogen binds to the cytosolic receptors with maximum binding half an hour after injection. It is then transferred to the nucleus where it exerts its effect. Accordingly, four groups of rats were set up. A control group was given vehicle (50% ethanol:50% water, 50 pal) only, followed half an hour later by vehicle again. A second group was given vehicle (50,us) followed half an hour later by a dose of oestradiol (5,ug in 50,us vehicle) which would produce a maximum uterine response in terms of increased fresh weight in the 6 hour duration of this experiment. A third group was given 2-hydroxy-oestrone (50 pg 2-hydroxy-oestrone in 50,us vehicle) followed half an hour later by the same dose of oestradiol as the second group. The fourth group was given molybdate-2-hydroxy-oestrone complex (50 pg 2-hydroxy-oestrone with 8.4 pg Mo in 50 pal vehicle) followed half an hour later by the same dose of oestradiol as the second and third groups. Uterine weights were taken at +6 hours, the results are shown in Figure 2. Oestradiol alone gave the expected maximum uterine response. The 2-hydroxy-oestrone was effective in blocking the subsequent oestradiol effect in this case because of the precise timing of injections and, again, the molybdate-2-hydroxy-oestrone was equally effective in blocking the oestradiol effect. This confirms that the molybdate-catechol oestrogen complex is capable of binding to the oestrogen receptor and blocking subsequent oestradiol effects. Thus, the conclusion that can be drawn from the above experiments is that the molybdate complex is capable of binding to the oestrogen receptor, yet has no oestrogenic effect; i.e. it is an anti- oestrogen.

EXAMPLE 3 Effect of Molvbdate Alone on Uterine Fresh Weight Using the same general methodology described in Example 2 above, the effect of administering molybdate alone was investigated. Four groups of six rats were prepared. A control group received only an injection of vehicle (50% ethanol:50% water, 50,us); a second group received an injection of oestradiol (Spg in 50,u1 vehicle), an amount effective to produce maximum uterine response at +6 hours; a third group received a single injection of molybdate (8.4 pig in 50,up vehicle) followed by the same dose of oestradiol half an hour later; and a fourth group received two molybdate injections, one (8.4pg in 50 u1 vehicle) 16 hours before administration of the oestradiol, and the second identical dose an hour before the oestradiol injection. When a single injection of molybdate was given before the oestradiol, no blocking effect of oestradiol was observed, but when two molybdate injections were given, (i.e. 16 hours and 30 minutes prior to oestradiol injection), there was a measurable reduction on the oestradiol effect on fresh weight of the uterus. The results may indicate to that a sustained molybdate level is required to form a complex with endogenous catechol oestrogens, and thereby exert its anti-oestrogenic effect in inhibiting the subsequent oestradiol injection on uterine fresh weight.

EXAMPLE 4 Effect of Molvbdate 2-Hvdroxv-oestrone Complex (Humphrin) on Breast Carcinoma 3366 xenoaranhs in Nude Mice Studies The anti-tumouringenic activity of Humphrin on a breast tumour was investigated using nude mice as a model host.

24 female nude mice were subcutaneously innoculated with fragments of the breast carcinoma 3366 tumour on day 0 of the experiment. All animals were supplemented with 0.5mg/kg estradiol intramuscularly once every week to induce tumour growth.

Once implanted tumours reached 4-5mm in diameter, animals were divided into three groups of eight animals and treated either with vehicle (25% ethanol, diluted with saline) subcutaneously as a control every day except weekends for 4 weeks, or with 20mg/kg Humphrin subcutaneously at the same time intervals, or with 50mg/kg Tamoxifen intramuscularly twice a week for 4 weeks.

Body weight and tumour volume were calculated twice every week.

At the end of the experiment the weight of the tumour and of the uterus was determined. Tumours were conserved in formalin for histopathologic examination.

EXAMPLE 5 Cell-Based Studies of Anti-Oestroaenic Activitv of the 2-Hydroxy-oestrone- Molvbdate Complex Two cell-based assays were used to screen the 2-hydroxy-oestrone- molybdate complex (2-HOMC) for anti-oestrogenic activity, namely a proliferation assay and a tritiated (3H) oestrogen competitive binding assay.

An MCF-7 cell line was used in both forms of assay. MCF-7 cells are oestrogen-dependent human breast cancer cells widely used as a model system in studies on mechanisms of oestrogen-dependent cancers. In addition to the MCF-7 cells, an oestrogen receptor negative HeLa cell line was used in proliferation assays as a control to determine whether any effects seen with MCF-7 cells were due to the presence of oestrogen receptors.

(i) PROLIFERATION ASSAYS The purpose of the proliferation assays was to demonstrate anti- proliferative activity of the molybdate complex in vitro and compare the effective dose with the effective dose required for the known anti- oestrogens Tamoxifen and the Zeneca compound ICI-182780.

Methods Cell Culture conditions MCF-7 cells were grown to confluence in an atmosphere of 5% CO2 in air in vented 75cm2 flasks with RPMI medium (25ml) containing extra pyruvate and 10% foetal calf serum. Flasks with confluent growth (containing about 8 x 106 cells) were split 1:4.

Preparation of Solutions of Oestradiol and the three oestrogen antagonists humphrin, tamoxifen, ICI-182780 Solutions of oestradiol, tamoxifen and lCl-1 82780 were prepared by dissolving accurately measured weights of the powdered compounds in water. The compound of the invention, molybdate 2-hydroxy oestrone complex, hereinafter referred to for convenience as "humphrin") was synthesized by mixing solutions of sodium molybdate and 2,3-dihydroxy oestradiol generally as set out in Example 1 above except that the ratio of water: ethanol in the final solution was 3:1 water:ethanol.

Proliferation assays MCF-7 cells were set up in 24-well plates in RPMI containing phenol red and 1 % dextran coated charcoal stripped serum. The cells were allowed to settle and grow for 2-3 days when oestradiol, with or without antagonists, were added. Cell proliferation was estimated by assaying totai cell protein after 2 or 3 days growth. This corresponds to the period at when maximal antagonist effect has been reported for ICI-182780.

RESULTS Proliferation assays a) MCF-7 cells. A first batch of humphrin (2.5 x 10-5M in ethanol) that had been stored at 40C for two months inhibited cell proliferation, but at a relatively high dose (1 0-5M cf tamoxifen 10.7 and ICI-182780 10 9M; these doses for tamoxifen and lCI-182780 agree with those reported in the literature). The results of the assays are shown in Figures 3a to 3d.

b) MCF-7 cells. A second batch of humphrin (5 x 1 0-3M in 25% ethanol) that had been stored at 40C for fourteen days was slightly more effective; its threshold dose was greater than 1 0-7M and less than or equal to 106M. The results obtained for this batch are shown in Figure 4.

c) MCF-7 cells. A humphrin batch tested after 26 days of storage neither inhibited cellular proliferation nor antagonised the upregulation by oestradiol that had been successfully obtained. It had lost even more of its ability to inhibit oestradiol binding.

d) HeLa cells. None of the oestrogen antagonists humphrin, tamoxifen or ICI-182780 inhibited cell proliferation of the oestrogen receptor negative HeLa cell. However tamoxifen was cytotoxic at 106M. The results for this assay are shown in Figures 5a to 5c.

In most proliferation experiments MCF-7 cells were not upregulated by oestradiol but all three antagonists did reduce cell proliferation. The mode of action of oestrogen antagonism is thought to be through direct competition with the oestrogen receptors as the effect is abrogated in the presence of high concentrations of oestradiol. This indicates that simple mass action overcame the antagonism. The failure of humphrin, or the other oestrogen antagonists, to effect HeLa cell proliferation is additional evidence that the action of humphrin is mediated via the oestrogen receptor.

ii) COMPETITIVE BINDING ASSAYS According to conventional wisdom, a pre-requisite for anti-oestrogenic activity is that a drug should prevent oestrogen binding to its receptor. An effective oestrogen antagonist would compete for the receptor at low doses, should have a high affinity for, and long residence time on, the receptor.

The competitive 3H-oestrogen binding assay screens for competitive binding by measuring reduced binding of 3H-oestrogen in the presence of competing ligands. The humphrin of the invention was compared with various known anti-oestrogens (tamoxifen, ICI-182780) with regard to its ability to block the binding of oestrogen to MCF-7 cells.

Materials 3H-oestrogen was obtained from DuPont (DuPont NEN 2,4,6,7-3H- oestradiol, specific activity 84.1CI mmol1, 1 mCi ml~' in 100% ethanol).

Methods MCF-7 cells were set up in 24 well plates at 2-4x 104 cells per well in 1 ml RPMI medium containing 10% fetal calf serum. They were grown for 2-3 days until almost confluent and then washed in cold DPBS. To pre- cooled (vessels standing in ice) cells 3H-oestradiol (10nM, 0.42pCi) was added in 0.5ml 0.5% ethanol/DPBS (cold) with or without competing ligand; the final ethanol concentration was 0.6%. Plates were left to incubate for 20 hours at 40C. DPBS was aspirated from plates and cells were washed three times in 1 ml of ice-cold saline (0.9%). Cells were dissolved in 300pal 0.1M sodium hydroxide for 30 minutes at room temperature. Plates were checked under the microscope to ensure that cells were completely dissolved. 1 00pal was removed for liquid scintillation counting, the remaining 200,u1 was assayed for proteins to normalise counts.

In some assays the procedure was modified so that competing ligand was added thirty minutes before the 3H-oestradiol and left to bind at 370C.

Results I. Comparison of oestradiol, 2-hvdroxv oestradiol and sodium molybdate Sodium molybdate exhibited no competitive properties. Both oestradiol and 2-hydroxy oestradiol competed with 3H-oestradiol for receptors; oestradiol being effective between 108 and 10-6 M and 2- hydroxy oestradiol between 10-5 and 10-25M. The results are shown in Figures 6a and 6b.

2. Competitive binding properties of humphrin A series of assays with two separate batches of humphrin, after various storage times, was carried out and the results are shown in Table 1.

TABLE 1 Competitive binding properties of two batches of humphrin, after different periods of storage post synthesis Mean binding (%) of 3H-oestradiol in the absence of humphrin humphrin batch number and days storage2 post synthesis humphrin 97/011 97/011 97/012 97/012 97/012 97/012 concentration 0 33 2 6 7 8 (M)1 1 X 10'7 72 84 69 60 3x 10-7 43 64 5x10-7 54 1 x 10-6 53 36 41 30 32 36 3x10-6 26 5 x 106 30 34 1 X 10-5 32 26 35 19 22 16 concentration calculated on the amount of 2-hydroxy oestradiol used in the synthesis 2 storage conditions: -200C, under N2, 50% ethanol, in the dark.

3. Comparison of the competitive binding properties of oestradiol and 2- hydroxy oestradiol A series of experiments in which oestradiol and 2-hydroxy oestradiol were directly compared were carried out and the results are shown in Table 2.

TABLE 2 Comparison of the competitive binding properties of oestradiol and 2- hydroxy oestradiol Concentration Mean binding (%) of 3H-oestradiol in the absence of competing (M)1 ligand 20H-E E 20H-E E 20H-E E 20H-E E 1 x 10 149 87 97 72 3X10-7 82 56 5 x 1o-7 145 51 1 x 10-6 159 60 88 40 104 49 40 47 3X10-5 29 44 5 x 10.6 52 41 1 x 10- 127 53 56 35 23 24 20H-E = 2-hydroxy oestradiol E = oestradiol The third set of comparative data in Table 2 (a composite of four experiments) is shown graphically in Figure 8b.

4. Comparison of the competitive binding properties of oestradiol and 2- hydroxy oestradiol and humphrin A direct comparison of all three ligands is given in Figures 7a and 7b and 9a and 9b. A further experiment comparing humphrin to oestradiol alone was completed and the results shown in Figure 8a 5. Comparison of the comDetitive binding properties of oestradiol, humnhrin, Tamoxifen and ICI-182780 A humphrin preparation (9 days post synthesis, stored at 40C in 25% ethanol in the dark) was compared to the other three competing ligands and the results are shown in Figure 10. The order of potency of the four ligands was: Oestradiol > humphrin > tamoxifen > ICI-182780.

6. Stability of the competitive binding properties of humphrin Early work suggested that humphrin (5 x 10-3M, in 25% ethanol) stored at 40C, in air and in the dark) was losing activity. Analysis of the data showed that humphrin activity declined with a half-life of about five days (Figure 11). The solution was still orange and clear, indicating that at least part of the catechol oestrogen complex was still intact.

This decline in activity contrasts sharply with the data on humphrin preparations that were stored at -200C, in 50% ethanol, under N2, and excluding light. These preparations were essentially stable over a 33-day period.

Discussion of competitive binding assay results Humphrin preparations clearly exhibited anti-oestrogenic properties.

However, it was essential to establish (a) that it was a molecular species formed in the synthesis of the preparations that was responsible for this observation, and (b) the level of activity compared to other leading anti- oestrogens.

The two components used in synthesizing humphrin are sodium molybdate and 2-hydroxy oestradiol. The data clearly indicate that sodium molybdate was inert as an anti-oestrogen. The data on 2-hydroxy oestradiol are more complex. This compound exhibits anti-oestrogenic activity when present in relatively high concentrations, i.e. greater than 10-5M. Fresh humphrin or those preparations kept under ideal conditions, exhibited anti- oestrogenic properties at lower concentrations, i.e 107M. Therefore the conclusion is that humphrin does contain a novel anti-oestrogen.

Without wishing to be bound by any theory, it is believed that the active component of humphrin is the compound formed by the reaction between molybdate and two 2-hydroxy oestradiol catechol moieties ("2- wing" humphrin). It is also possible that "1-wing" humphrin (the catechol oestrogen complex formed by combining one 2-hydroxy oestradiol moiety with a molybdate ion) may contribute to the anti-oestrogen properties.

The data also clearly demonstrate that humphrin is at least as anti- oestrogenic as Tamoxifen and ICl-182780 and is in fact probably more effective.

Potentiation (increase over control value) of 3H-oestradiol binding was observed in some early experiments. This may have been an artifact of the protocol. Adding competitive ligand before 3H-oestradiol at 370C followed by addition of 3H-oestradiol at 40C may have disturbed the equilibrium between competing and labelled ligand to different extents in different wells, depending on the time for which cells were exposed to 40C before the labelled ligand could be added. Potentiation was not observed when competing ligand and 3H ligand were added simultaneously.

Although ICl-182780 is a potent antagonist in vitro on celiular proliferation, extensive specific binding to sites other than the oestrogen receptor in MCF-7 cells has been reported in the literature. Thus the 3H- binding assay on its own may not always be a reliable indicator of anti- proliferative activity.

EXAMPLE 5 Characterisation of Molybdate 2-Hvdroxv-oestrone Complex (Humphrin) Molybdenum 2-hydroxy-oestrone complex (humphrin) was prepared by reacting sodium molybdate (NaMoO4.2H2O and 2-hydroxy oestradiol to give a solution containing 5 pmols per ml of the complex in a mixture of 50:50 ethanol and water. This sample will be referred for convenience hereunder as sample H1.

In order to characterise the H1 solution, mass spectrometry was carried out using a Finegan MAT90O spectrometer with electrospray sample presentation. This technique has an accuracy of j0.1 mass units. Because of the conditions used, the ions generated are negatively charged.

Although the technique is not quantitative, the relative abundance of the various ionic species obtained under the conditions used can be approximated by the intensity of their signal peaks.

Sample H1 was subjected to mass spectrometry using the foregoing apparatus and typical mass spectra are shown in Figures 1 2, 1 3 and 1 4. In Figures 12 and 13, hydroxy oestradiol (287.3) is the most intense peak.

The proposed structure for humphrin requires a molybdenum atom with two associated oxygen moieties to be linked to the catechol groups of two molecules of 2-dihydroxy oestradiol. Such a compound would have a molecular mass of 705 (calculated using the atomic weight of the most abundant molybdenum isotope (98). Therefore, the cluster about the ion at 703.6 in Figures 12 and 13 is believed to be that from the humphrin ion.

The spectrum also shows a cluster about the ion mass of 433.2 which is considered to be the humphrin ion minus one of the steroid ligands.

In Figure 14, small clusters typical of the mono-steroid and di-steroid forms of humphrin can also be seen, along with large peaks at 287.3 and 575.6. It is considered that the peak at 575.6 could be the dehydrodimeric of 2-hydroxy oestradiol.

The foregoing examples are merely by way of illustration and are not intended to limit the scope of the invention in any way, the true scope of the invention being as defined in the following claims.