Phosphinic pseudo-peptides that may be used as matrix zinc metalloprotease inhibitors

In this example where R 3 represents CH 2 R′ 3 , the acrylate is synthesised according to the following synthesis process:

In this example, this synthesis process is used to prepare the acrylate 2 wherein R 3 represents the CH 3 (CH 2 ) 6 group, i.e. R′ 3 ═CH 3 —(CH 2 ) 5 . In an argon atmosphere, a hexyl bromide solution CH 3 (CH 2 ) 5 Br(1.65 g, 10 mmol) in 15 ml of pure diethyl ether is added drop by drop to a flask containing 0.22 g (11 mmol) of magnesium and iodine I 2 (catalytic) over a 90 minute period.

The reaction mixture is reflux boiled for 1 hour. After adding 0.18 mmol of CuI, the temperature of the mixture is decreased to −78° C.

A solution of 1.15 g (6.7 mmol) of the compound is then added slowly:

in 10 ml of Et 2 O: THF. The mixture is stirred for 30 minutes at ambient temperature. After treating the mixture with 0.5N HCl, 5% NaHCO 3 and water, the organic layer is dried on Na 2 SO 4 . The solvent is eliminated in a vacuum and the residue obtained is purified by column chromatography using a 40 to 60° C. petroleum ether/ether mixture (13:1) as the eluent. The compound 2f is obtained in this way with a 40% yield. The characteristics of compound 2f are given in table 1.

In this example, an acrylate 2 is prepared with R 3 ═R′ 3 —CH 2 — using the method described by Baldwin et al J. Chem. Soc. Chem. Comm. 1986, p 1339-1340, [12].

This corresponds to the following reaction diagram:

This method is used to prepare the acrylate 2 g wherein R 3 represents the group according to the formula:

A mixture of 1.9 g (10 mmol) of ethyl bromomethyl acrylate and 3.3 g (20 mmol) of sulphinic benzene acid sodium salt in 40 ml of methanol is reflux boiled for 12 hours. The solvent is eliminated and diethyl ether is added. The solution is washed with water, brine, dried on Na 2 SO 4 and concentrated to obtain 2 grams of sulphinic acrylate

in oil form, with an 80% yield.

To a mixture of 2 g (8 mmol) of sulphinic acrylate in 40 ml of dry benzene, 4.7 g (16.4 mmol) of stannic tri-n-butyl hydride and 0.15 g (0.96 mmol) of 2,2′-azobisisobutyronitrile (AIBN) are added. The mixture is reflux boiled for 1.5 hours, and water is then added and the organic layer extracted is dried on Na 2 SO 4 and concentrated. The unprocessed product is purified by column chromatography, using petroleum ether (40 to 60° C.)/ether (9:1) as an eluent. 2.9 g of alkylstannane is obtained

with a 90% yield.

1.47 g (3.6 mmol) of alkylstannane and 0.39 g (1.8 mmol) of 2-bromoethyl naphthalene are dissolved in 10 ml of dry benzene. After adding 0.064 g (0.39 mmol) of AIBN, the reaction mixture is reflux boiled for 2 hours. After adding water, the organic phase is separated, washed with water dried on Na 2 SO 4 and concentrated to dryness. In this way, 0.06 g of the acrylate 2 g given in table 1 is obtained, with a 13% yield, After purifying the residue on a chromatography column using petroleum ether (40 to 60° C.)/ether (9:1) as an eluent.

EXAMPLES 8 AND 9

In these examples, an acrylate 2 is prepared, wherein R 3 comprises a sulphur atom or an oxygen atom, according to the following synthesis diagram:

In this diagram, R 3 corresponds to RX—CH 2 where X═S or O.

The method is used to prepare the acrylates 2h and 2i wherein R 3 represents the paramethoxybenzyl-thiomethyl group and the benzylthiomethyl group, respectively.

A 10 mmol solution of suitable thiol R—SH in 15 ml of methanol is added drop by drop to a stirred solution cooled in an ice-bath of 9 mmol of sodium in 20 ml of methanol over a 30 minute period. After adding thiol, the mixture is concentrated to dryness and diethyl ether is added. The salt which precipitates is cooled in an ice-bath. The product is filtered, washed with cold Et 2 O and dried on P 2 O 5 to obtain the sodium salt with a 85 to 95% yield. To a mixture of 10 mmol of sodium salt in suspension in 40 ml of dry Et 2 O and cooled in an ice-bath, over a 45 minute period, a solution of ethyl bromomethacrylate (9 mmol) in 20 ml of diethyl ether is added drop by drop. The solution is stirred for 30 minutes at 0° C. and for 1 to 2 hours at ambient temperature. The reaction mixture is diluted with 20 ml of water and the organic layer is washed with water, and then dried on Na 2 SO 4 and evaporated. The compounds 2h and 2i obtained in this way are purified by column chromatography, using petroleum ether (40-60° C.)/Et 2 O (8:2) as an eluent.

The yields are given in table 2.

In this example, the acrylate 2j wherein R 3 corresponds to ROCH 2 and represents the group according to the formula C 6 H 4 —CH 2 —O—CH 2 is prepared according to the reaction diagram illustrated above, with R OH.

In a completely dry flask, small pieces of sodium (10 mmol) are added to 20 ml of dry diethyl ether. To this reaction mixture, a 10 mmol solution of benzyl alcohol in 10 ml of diethyl ether is added over a 2 hour period, under a moderate reflux. The mixture is reflux boiled for 6 additional hours. The white precipitate obtained is cooled in an ice-bath, and then filtered, washed with dry and cold diethyl ether and dried on P 2 O 5 to obtain the sodium salt with an 85 to 95% yield.

The reaction of said salt with ethyl bromomethacrylate and the purification of the compound obtained are performed as described in example 8.

The acrylate 2j obtained is listed in table 2.

EXAMPLES 11 TO 21

In these examples, phosphinic blocks 3 are prepared with R 2 and R 3 given in table 3, using the procedure described by Yiotakis et al, J. Org. Chem. 61, 1996, 6601-6605 [8].

This procedure firstly comprises the Michael reaction described in FIG. 1 . The initial phosphinic acid 1 (FIG. 1) was prepared according to the method described by Baylis et al, J. Chem. Soc. Perkin Trans I, 1984, p. 2845-2853 [13].

The acrylates 2 were prepared in examples 1 to 10. To prepare the phosphinic blocks, proceed as follows.

A suspension of 1 mmol of N-benzoyloxy-carbonylphosphinic acid 1 and 5 mmol of hexamethyl-disilazane is heated to 110° C., for 1 hour, in a nitrogen atmosphere.

1.3 mmol of the suitable acrylate 2 is then added drop by drop over a 15 minute period. The reaction mixture is placed under stirring for 3 additional hours at 110° C. The mixture is allowed to cool to 70° C. and 3 ml of ethanol is added drop by drop. After cooling to ambient temperature, the reaction mixture is concentrated. The residue is purified by column chromatography using a chloroform/methanol/acetic acid (7:0.5:0.5) mixture as an eluent. In this way, the phosphinic blocks 3a to 3k are obtained with the yields given in table 3.

In this table, the R 3 column indicates which acrylate 2 was used for the synthesis of the phosphinic block. Table 3 also gives the Rf of the blocks obtained.

EXAMPLES 22 AND 23

In these examples, the phosphinic blocks 3l and 3m complying with the formulas given in table 3 are synthesised using the following procedure.

2.1 mmol of diisopropylamine and 2.1 mmol of trimethylsilyl chloride are added to a solution chilled in ice of 1 mmol of phosphinic acid 1 in 2 ml of chloroform.

The mixture is placed under stirring at ambient temperature for 3 hours. After cooling to 0° C., 1.4 mmol of suitable ethyl acrylate 2 (2e or 2g) is added, and the reaction mixture is stirred at ambient temperature for 16 hours. After adding ethanol drop by drop, the solvents are eliminated in a vacuum. The compounds 3l or 3m in table 3 are obtained after column chromatography as for examples 11 to 21.

The yields and Rf values of the blocks obtained are given in table 3.

In addition, the blocks 3b, 3d and 3e to 3k were characterised with proton NMR, carbon 13 and phosphorus. The results obtained are appended.

In this example, the compounds 4a to 4m in FIG. 1 are prepared from compounds 3a to 3m according to the following procedure.

1 mmol of compound 3a to 3m and 1.2 mmol of 1-adamantyl bromide are dissolved in 10 ml of chloroform. The reaction mixture is reflux boiled. 2 mmol of silver oxide is then added in five equal parts, over a 50 minute period. The solution is reflux boiled for 30 more minutes. After the solvents are eliminated, the residue is treated with diethyl ether and filtered on Celite. The filtrates are concentrated. The residue is purified by column chromatography, using the chloroform/isopropanol (9.8:0.2) mixture as an eluent. In this way, the compounds 4a to 4m are obtained with the yields given in table 4, the Rf values of said compounds are also given in table 4.

In this example, the procedure described in FIG. 1 is followed to convert compounds 4a to 4m into compounds 5a to 5m. For this purpose, the following procedure is used.

1 ml of 4N NaOH is added drop by drop to a stirred solution of 1 mmol of compound 4a to 4m in 5.5 ml of methanol. The reaction mixture is stirred for 18 hours, and the solvent is then eliminated. The residue is diluted with water, and then acidified with 0.5 N HCl in an ice bath, dried on Na 2 SO 4 and concentrated to obtain compounds 5a to 5m in white solid form, with the yields given in table 5. The Rf values of said compounds are also given in table 5.

In this example, phosphinic pseudo-peptides wherein the formulas are given in tables 5 to 7 are prepared from the phosphinic blocks in table 4, by means of Fmoc (fluorenylmethoxycarbonyl) chemistry solid phase synthesis, according to the procedure described by Yotakis et al, 1996, J. Org. Chem. 61, pages 6601-6605 [8] and Jiracek et al in J. Biol. Chem., 1995, 270, pages 21701-21706 [9] and in J. Biol. Chem., 1996, 271, pages 19606-19611 [10].

The coupling is carried out by means of the in situ strategy using 2-(1H-benzotriazol-1-yl-1,1,3,3-tetramethyluronium-hexafluorophosphate(HBTU)/diiso-propylethylamine. The coupling conditions are as follows: three equivalents of amino acid Fmoc derivative and 4 equivalents of diisopropylethylamine in dimethylformamide are added to the resin and left to react for 30 min. For the coupling of the phosphinic blocks:

1.5 block equivalents are used.

The splitting conditions of the Fmoc group are 50% piperidine in dimethylformamide for 30 minutes. The Fmoc group is (fluoremylmethoxy) carbonyl.

In this example, the pseudo-peptides given in table 8 comprising different types of R 1 group are synthesised. The R 1 group introduction conditions are given below, according to the type of group.

If R 1 is an acid comprising an indole or quinoline group (compounds 30, 31, 32, 33 in table 8), the R 1 group was coupled on the peptide of the generic formula

still coupled to the resin, under the following conditions: 3 equivalents of acid diluted in a small volume of N-methyl-pyrolidone, 3 equivalents of HBTU (0.4 M), 10 equivalents of diisopropylethylamine (1.2 M), with a coupling time of 45 min. The incorporation of the R 1 group is followed by a Kaiser test. When required, this operation was repeated several times, until the amine function was completely substituted.

The R 1 ═Ph—CH 2 —O—CH 2 —CO— (compound 26) group was incorporated using the corresponding chlorinated derivative under the following conditions: the chlorinated derivative (25 equivalents, 0.5 M/dichloromethane) and diisopropylethylamine (25 equivalents, 0.5 M/dichloromethane) are added to the resin. The acylation reaction requires 1/2 h.

For the synthesis of compounds 27, 28 and 29, the peptide

still coupled to the resin, was firstly acylated with the Br—CH 2 —CO—Br derivative under the following conditions: the brominated derivative (25 equivalents, 0.5 M/dichloromethane) and diisopropylethylamine (25 equivalents, 0.5 M/dichloromethane) are added to the resin, for a reaction time of 1/2 h. The second step corresponds to the alkylation of the corresponding aniline derivatives. For this purpose, the aniline derivative (50 equivalents/DMSO) is added to the resin, for a coupling time of 2.5 h.

The splitting of the peptides of the resin, and the hydrolysis of the protective groups were carried out using a trifluoroacetic acid solution containing 2.5% water, 2.5% thioanisoi, 1.25% thiophenol, 1.25% ethanedithiol and 1.25% triisopropylsilane.

All the peptides synthesised in examples 26 and 27 were purified by reverse phase HPLC using gradients produced with water, acetonitrile solutions containing 0.01% trifluoroacetic acid. In the majority of cases, four peaks are observed in the chromatograms, corresponding to the four forms of diastereoisomers generated by the synthesis protocol of said phosphinic compounds. All the phosphinic peptides purified in this way were inspected using mass spectroscopy.

In this example, the affinity constant Ki of the pseudo-peptides in tables 5 to 8 are examined with respect to the different matrix metalloproteases MMP.

The MMPs used were produced in recombinant form, in an E. coli expression system, and then purified using different types of chromatographies. Apart from stromelysin-3 (MMP-11), the MMPs produces correspond to human sequences. The Stromelysin-3 used in this study corresponds to the murine form.

The activity of each MMP was determined using two fluorogenic synthetic substrates. The cutoff of said substrates, which generates a fluorescent signal proportional to the quantity of split substrate, enables a precise determination of the kinetic parameters. The values of the constants Km of the substrates, taken into account to determine the Ki values, are given in the table below. The affinity constants Ki were calculated using the equation of Horowitz et al, Proc. Natl. Acad. Sci. USA, 1987, 84, p. 6654-6658 [14], accounting for the dependency of the percentage of inhibition measured experimentally according to the concentration of inhibitor, as a function of the enzyme and substrate concentration.

The experiments for conducted in a Tris/50 mM HCl buffer, pH 6.8, 10 mM CaCl 2 , 25° C.

The results obtained are given in tables 5 to 8.

The generic formula of the inhibitors synthesised in this case indicates that they are phosphinic pseudo-tripeptides. According to the assumed bonding mode of said compounds with the MMPs' active site, said inhibitors must interact with the sub-sites S1, S1′ and S2′ of said enzymes, respectively. The R1group comprised on most of the inhibitors is assumed to interact with the junction of sub-sites S3/S2.

In table 5, the influence of the type of R 3 residue on the efficacy of the pseudo-peptide as an MMP inhibitor.

The interaction of the inhibitors with the sub-site S1′, involving the R 3 group of the compounds, was the subject of an in-depth study, since this sub-site is thought to mainly control enzyme-inhibitor interaction selectivity.

The analysis of table 5 demonstrates that the size and nature of the R 3 residue plays a critical role in the affinity of compounds: in this way, the activity of the inhibitors appears to be multiplied by a factor of 30 in the case of MMP-8, when the R 3 group changes from a benzyl to phenylpropyl residue (compounds 7 and 9). It is noted that the gain in affinity corresponding to this substitution is not constant for the different MMPs, suggesting that the sub-site S1′ of each MMP may be sensitive to the nature of the R 3 residue. In this respect, it is interesting to observe that MMP-11 is the only MMP to prefer phenethyl residue, in relation to a phenylpropyl residue (compound 8 and 9).

The comparison of compounds 9, 10 and 11 which only differ by a single atom (C, O, S) on the R 3 residue, suggests the existence of a specific interaction between the sulphur atom in the γ position of R 3 of the inhibitors and the sub-site S1′ of the MMPs. Irrespective of the type of MMP, compound 11 is always the most powerful of these three inhibitors.

In table 6, the inhibitors are characterised by the presence of a phenyl residue in R 2 , instead of a methyl residue (table 5). The comparison of tables 5 and 6 reveals that the presence of a benzyl residue makes it possible to increase the overall affinity of the inhibitors. In the specific case of MMP-11, the methyl->benzyl substitution induces a much more significant increase for this MMP, than for the others. This result indicates the preference of the sub-site S1 of MMP-11 for a benzyl residue, in relation to a methyl residue.

Introducing a greater diversity in R 3 in this series made it possible to obtain a clearer idea of the influence of the residue in R 3 . This series demonstrates that, to obtain powerful phosphinic inhibitors with respect to certain MMPs, in the R 3 position, it is necessary to introduce a long arylalkyl chain. In this way, compound 18 is an example of a very powerful inhibitor of MMP-8, MMP-11, MMP-2 and MMP-9.

In relation to these MMPs, it is noted that the inhibitors reported in this study appears to be less powerful with respect to MMP-14.

In addition, as a general rule, these pseudo-peptides appear to have a low activity with respect to MMP-1 and MMP-7.

It should be noted that compound 14 comprising a phenethyl R 3 residue appears to be very powerful on MMP-11, but it is much less active on other MMPs.

Table 7 illustrates the importance of the tryptophan residue on the R 4 position, and the importance of the tryptophan residue configuration. Compound 24 indicates that the tryptophan in these inhibitors may be replaced by another aromatic residue Dpa (dinitrobenzyl) in the case of MMP-11 and MMP-8

Table 8 illustrates the effect of different modifications on the R 1 group (Table IV). The analysis of the results indicates that the nature of R 1 influences the affinity of compounds with respect to different MMPs. Compound 31 is an example of a very powerful inhibitor with respect to MMP-11, showing a certain selectivity for said enzyme. Compounds 34 and 35 represent examples of powerful inhibitors wherein R 1 corresponds to a natural amino acid.

In this example, the influence of the configuration of the two positions R 2 and R 3 on the efficacy of pseudo-peptides as an MMP inhibitor is studied.

The synthesis process used in the above examples to prepare the phosphinic derivatives according to this patent produces each inhibitor in the form of a mixture of four diastereoisomers, due to the presence to two asymmetric centres on the alpha carbon comprising the R 2 and R 3 residues. To evaluate the influence of the configuration of these two positions, compound 15 was resynthesised using optically pure phosphinic phenylalanine amino acid, of an R or S configuration. Each synthesis produces a mixture of two diastereoisomers:

Z-(S)PheY(PO 2 CH 2 ) (S)pPhe-Trp-NH 2 and Z-(S)PheY(PO 2 CH 2 ) (R)pPhe-Trp-NH 2 or

Z-(R)PheY(PO 2 CH 2 ) (S)pPhe-Trp-NH 2 and Z-(R)PheY(PO 2 CH 2 ) (R)pPhe-Trp-NH 2

that can easily be separated by reverse phase HPLC. Table 9 shows the activity of the four diastereoisomers of compound 15 with respect to the different MMPs. It is interesting to note that for this class of phosphinic compounds, at least three diastereoisomers inhibit the different MMPs in an almost equivalent manner. This property, apart from being a means to control inhibitor selectivity, could also prove to be very important in terms of metabolism and pharmacokinetics, two parameters which could be sensitive to the molecular stereochemistry.

The results of the characterisation by proton NMR, carbon 13 and phosphorus of the RI fraction of compound 15 are appended.

In this example, the anti-tumour efficacy of compound 15 (RI fraction) in table 9 (RXPO3) is measured.

The tumorigenesis model used to test the effect of said RXPO3 inhibitor in vivo consists of a subcutaneous graft of malignant murine C26 cells in syngenic mice (with a genetic background similar to that of the injected cells).

The C26 cells established from Balb c mouse colic cancer (Corbett et al., 1975, Cancer Res, 35:2434-2439 [15] are cultured until subconfluence. After trypsinisation, the cells are centrifuged at 1000 g for 10 minutes. The sediment is washed and resuspended in 1×PBS. A 200 ml volume, containing 5×10 4 C26 cells, is injected subcutaneously at 2 sites on the back of 8 to 9-week old mice. The inhibitor is solubilised in 1× PBS and 150 mg of inhibitor in a 150 ml volume is administered by the intraperitoneal route. The treatment starts on the day of the injection of the C26 cells and continues at a rate of one injection a day for 25 days. The tumoral volumes are measured every day.

Three identical and independent tumorigenesis experiments were conducted, according to the protocol given above. The results obtained were similar. One of them is reported below.

The experiment related to 12 animals, 6 receiving a treatment (150 μg of inhibitor/mouse/day) and 6 used as controlled animals (150 μl 1×PBS). FIG. 2 shows the median values of the tumoral volumes as a function of the time elapsed since the injection of the C26 cells (Days 10 to 25).

It is observed that the tumours start to appear on day D10, and that their mean and median volumes are always lower in the animals treated with the inhibitor RXPO3. This difference in tumour size is of the order of 50% on day D15 to day D20. Subsequently, the efficacy of the inhibitor appears to be lower. These results are in line with the recent observations made on another tumorigenesis model using wild or deficient animals for the expression of stromelysin-3 (MMP-11) and demonstrating that this matrix metalloprotease is involved in the initial implantation steps and favours the development of tumours. According to this model, the efficacy of the inhibitors must be maximal during the first days of tumour development.

References

[1]: Brown, Medical Oncology, 1997, 14, p. 1-10.

[2]: Beckett et al, 1998, Exp. Opin. Ther. Patents, 8, p. 259-289.

[3]: Goulet et al, Bioorg. Med. Chem. Lett. 4, 1994, p. 1221-1224.

[4]: Caldwell et al, Bioorg. Med. Chem. Letter 6, 1996, p.323-328.

[5]: FR-A-2 676 059.

[6]: FR-A-2 689 764.

[7]: EP-A-0 725 075

[8]: Yotakis et al, J. Org. Chem., 1996, 61, p. 6601-6605.

[9]: Jiracek et al, J. Biol. Chem., 1995, 270, p. 21701-21706.

[10]: J. Biol. Chem., 1996, 271, p. 19606-19611.

[11]: Eistetter et al, J. Med. Chem., 25, p 109-113, 1982.

[12]: Baldwin. et al, J. Chem. Soc. Chem. Comm. 1986, p 1339-1340.

[13]: Baylis et al, J. Chem. Soc. Perkin Trans I, 1984, p. 2845-2853.

[14]: Horowitz et al, Proc. Natl. Acad. Sci. USA, 1987, 84, p. 6654-6658.

[15]: Corbett et al., 1975, Cancer Res, 35:2434-2439.