PT98970A - PREPARATION PROCESS OF IMODAZOLINE DERIVATIVES AND PHARMACEUTICAL COMPOSITIONS CONTAINING THEM - Google Patents

Description

73 058 DRH / TAB / AC / 35180/033 -2-

DESCRIPTIVE MEMORY

This invention relates to imidazoline derivatives, their salts, processes for their preparation and pharmaceutical compositions containing them.

In the description of our European patent Na. 0071368 we describe and claim dihydrobenzofuranylimidazolines of formula

Wherein R1 is hydrogen or C1 -C6 alkyl; R2 is hydrogen, methyl, chlorine, bromine or fluoro; R 3 is hydrogen, methyl, hydroxy, methoxy, fluoro, chloro or bromo; and their non-toxic salts. The compounds of Formula A contain an asymmetric carbon atom and the invention encompasses both the racemic mixtures and the optically active enantiomers. All exemplified compounds were in the form of racemates.

The compounds of Formula A possess Î ± 2 -adrenoceptor antagonist activity and, by virtue of this activity, have potential utility in the treatment of those conditions in which patients are implicated in the activities of the Î ± 2 -receptor antagonists such as endogenous depression, heart failure, diabetes, obesity, migraine, etc.

Our subsequent investigations of the compounds of Formula A led to the identification of 2- [2- (2-ethyl-2,3-dihydro-benzofuranyl)] - 2-imidazoline (in Formula A: R1 = ethyl, R2 = R 3 = hydrogen; INN efaroxane). Eryaroxane as the racemate has been shown to be a particularly potent and selective α 2 -adrenoceptor antagonist and, based on these properties, has been identified

As a potential anti-diabetic drug.

We have now increased the field of investigation of the compounds of Formula A and have found that efaroxane and the novel isopropyl analog compound (in Formula A: R1 = i-propyl, R2 = R3 = hydrogen) are pharmacologically different from analogous compounds where R1 is hydrogen , n-propyl or longer alkyl. In addition, we can now resolve the efaroxane and its isopropyl analogue to its optically active enantiomers and have found that the (-) - enantiomers have a completely unexpected pharmacological profile.

According to this invention, the (-) - enantiomer of a compound of Formula 1

where R is hydrogen or methyl, in substantially pure form, and the non-toxic salts thereof.

By (-) enantiomer of a compound of Formula 1, in substantially pure form, it is meant that the amount of (+) - enantiomer present does not exceed 5% and preferably does not exceed 2%. Examples of non-toxic salts are those obtained with inorganic acids such as hydrochloric, sulfuric or phosphoric acid; or with organic acids such as acetic, propionic, malonic, succinic, fumaric, tartaric or citric acids. A preferred salt is the hydrochloride. The invention also includes the preparation of pharmaceutical compositions comprising the (-) - enantiomer of a compound of Formula 1 in substantially pure form or a non-toxic salt thereof, together with a pharmaceutically acceptable diluent or carrier. The literature clearly indicates the relationship between α2-adrenoceptor antagonism and the stimulation of insulin secretion (Robertson RP and Porte D, Diabetes, 1973, 22,1; Efendic S, Cerasi E and Luft R, Acta Endocrinologica, 1973, 74 , 542-547, Linde J and Deckert T, Horm Metab Res 1973, 5, 391-395). In particular, the antidiabetic effects of Midaglizole (DG 5128) have been attributed to α2-antagonism (Kameda Kui-ya, Shin-etsu Ono, Isao Koyama and Yasushi Abiko, Acta Endocrinologica 1982, 99, 410-415).

Although secretion of insulin from pancreatic islets in response to glucose stimulation is the result of a complex sequence of events, changes in potassium permeability are of paramount importance. Thus, increased potassium permeability (potassium channel openings) reduces insulin secretion, and the opposite is true with potassium channel blockade. Several drugs affect the permeability of potassium channels and the antidiabetic function of sulfonylureas as potassium channel blockers. In contrast, vasodilator diazoxide, which can induce a diabetic state, increases the permeability of the potassium channels and thus reduces the secretion of insulin. The mechanism by which Î ± 2 -antagonists stimulate insulin secretion has been related to its effects on the K + channels of pancreatic islet cells. Thus, it has been shown that the reduction of permeability of the K + channels by efaroxane occurs over the same range of α 2 -adrenoceptor blocking concentrations (Sehlin J, Doxey JC and Lindstrom P, Diabetologia, 1987, 30, 7, 580A No 503). An opposite effect has been demonstrated by UK 14304, a selective a2-agonist.

These observations were confirmed in the mouse deferent duet by the work of Zimanyi I, Folly G and Vizi E S (J. Neu-roscience Res, 1988, 20, 102-108) who concluded that the stimulus

The α 2 -adrenoreceptors (by Î ± 2-agonists) leads to an increase in K + permeability.

These results clearly indicate that there is a direct correlation between α2-adrenoceptors and changes in permeability of K + channels in both islet tissues and in different tissues.

Chan SLF and Morgan NG (Eur J Pharmac, 1990, 176, 97-101) have recently concluded that the ability of efaroxane to stimulate insulin secretion from pancreatic islets isolated from rats could not be attributed to the interaction of efaroxane with α2- adrenoreceptors since the effect was not reproduced by the corresponding α2-antagonist idazoxane at the same concentration. However, very high concentrations of efaroxane have been required to achieve this effect and, since idazoxane is 3 to 5 times less potent than efaroxane, an active concentration level for idazoxane may not have been attained. Chan and Morgan also showed that efaroxan had a greater effect than idazoxane in reversing the diazoxide inhibitory effects on glucose-induced insulin release, although, again, very high concentrations of efaroxane were required.

It has now been shown that the Î ± 2 -adrenoceptor blocking effects of the Î ± 2 -antagonist racemics, involving interactions with a receptor for which the natural ligand is a pure enantiomer (noradrenaline), are stereospecific with the activity residing in an enantiomer. In the case of efaroxane is the (+) - enantiomer. Table 1 shows that in several systems the selectivity of the (+) enantiomer over (-), for the α2 -receptors, is about 600 to 5,000.

73 058 DRH / TAB / AC / 35180/033 -6-

Table 1

Separation of enantiomers in non-islet α2-adrenoreceptors "

Racemic (+) (-) Ratio (+) / (-)

In vitro rat deferent duet pA2 8.9 8.9 5.2 511 *

In vivo deferent rat duct DR2 Mltioles / kg IV 0.05 0.02 12.7 635 **

Binding of radio-ligand rat cerebral cortex

Kl nM 1.3 0.88 3.871 4 398 ** * Determined by taking the anti-log of pA2 (+) - pA2 (-) ie 3.7 ** Power ratio obtained by dividing the reciprocal of (+) by the reciprocal of (-) ie _1_ + 1 = 12.7 = 635 0.02 12.7 0.02

All three methods indicate good separation of activity. The details of the tests are as follows: 1. The potencies in vitro. (pA2 values) were determined in the prostatic section of the rat vas deferral) using methodology already described (Doxey et al, Br J Pharm, 1984, 83, 713). The α 2 -adrenoceptor agonist used in these studies was p-aminoclonidine. 2. In vivo, pre-binding potencies of Î ± 2 -adrenoceptor antagonists were determined in the deferred duet of dead rats by sectioning the spinal cord using a method already described (Welbourn et al., J Med Chem, 1986, 29, 7-DRH / TAB / AC / 35180/033 2000). The potencies of the antagonists were determined as the dose (μmol / kg, iv) required to produce a two-fold (DR2) shift in the UK-14304 dose response curve in response to contraction of the vas deferens. 3. Radio-ligand (Kl / nM) binding affinities were determined from their ability to displace the [3 H] -didazoxane saturable binding of the α 2 -adrenoceptor sites prepared from rat cerebral cortical membranes (Welbourn et al, J Med Chem, 1986, 29, 2000).

In addition to its relationship with Î ± 2 -adrenoceptor antagonism, the suggestion that inhibition of K + channel permeability was a stereospecific phenomenon had never been suggested. In fact, the very high concentration of efaroxane required to inhibit the effect of diazoxide on pancreatic islet cells (Chan and Morgan 1990) suggests a non-specific effect.

Unexpectedly, we have now found that the inhibitory effect of efaroxane on K + channel permeability (and hence its ability to stimulate insulin secretion) resides in a single (-) enantiomer, which is not the Î ± 2 -agonist enantiomer. The results set forth in Figs. 1 to 3 were obtained as follows:

Ficrura 1.

Groups of isolated rat islets were incubated with 20 mM glucose, 1 μM UK-14304 and with increasing concentrations of (+) and (-) enantiomers of efaroxane. After incubation for 60 minutes at 37øC, samples were taken from the medium and assayed for insulin. Data are mean ± SEM values for 10-18 observations.

Figure 2

Groups of isolated rat islets were incubated with 20 mM glucose, 250 .mu.mozoxide diazoxide and increasing concentrations of

(+) And (-) enantiomers of the efaroxane. After incubation for 60 minutes at 37øC, samples were taken from the medium and assayed for insulin. Results are mean ± SEM values for 10-18 observations.

Ficmra 3_

Groups of isolated rat islets were incubated with 8 mM glucose in the presence of increasing concentrations of the (+) and (-) isomers of efaroxane (thus similar procedure to that used for Figures 2 and 3).

In Figure 1, insulin secretion stimulated by 20 mM glucose is inhibited by the α2-agonist UK-14304 (1 μΜ). This inhibitory effect is inverted, dose dependent, by (+) but not by (-) efaroxane. This confirms that Î ± 2 -antagonistic activity resides in the (+) - enantiomer, in pancreatic islets as well as in different cells. Fig. 2 shows that insulin secretion stimulated by 20 mM glucose can also be inhibited by diazoxide, a known K + channel opening agent, and this effect is reversed by (-) but not (+) efaroxane. Therefore, the effect of efavirane on the K + channel in pancreatic islet cells is due to the (-) enantiomer. Fig. 3 shows that the (-) enantiomer potentiates insulin secretion at a threshold concentration of glucose, an effect that is not shared by the (+) enantiomer. This confirms that the dominant effect of insulin secretion from efaroxane is controlled by changes in permeability of the K + channel caused by the (-) enantiomer.

Data on Î ± 2 -adrenoceptor antagonists and data on the K + channels for various racemic compounds of Formula A wherein (R2 = R3 = hydrogen) are shown in Table 2. The power Î ± 2 is expressed in terms of the potency of compound in relation to idazoxane in the (in vitro) assay in the rat vas deferens. Data on the K + channel are presented in terms of the inverse percentage reduction in insulin secretion (stimulated by 20 mM glucose) by diazoxide (250 μΜ), by the test substance 100 μΜ.

Table 2

(With respect to idazoxane 1 K +% H 1,7 Ethyl 1/8 100 i-propyl oo σ-89 n-propyl 1.9 n-pentyl 0.44 2

The results in Table 2 further highlight the unexpected and unpredictable aspect of the results with the compounds prepared by this invention further emphasize the point that the effects of the K + channels are totally independent of any activity of α2 that the compounds possess (i.e., there is indeed a relationship between α2 -adrenoceptors and changes in permeability of the K + channel in the compounds). Compounds wherein R 1 = H, ethyl, n-propyl and n-pentyl (racemates) all have appreciable Î ± 2 -antagonistic activity; with the exception of the latter, they are equipotent, being approximately 2x more potent than idazoxane. Only efaroxane (R1 = ethyl), however, has K + channel inhibitory properties. In contrast, the compound wherein R 1 = i-propyl sharply reduced α 2 -antagonist activity although it is approximately as effective as efaroxane in inhibiting K + channels.

Subsequent studies with the (+) and (-) enantiomers of the isopropyl analog of efaroxane gave unexpected results which highlight the single action of the (-) isomer of efaroxane on the K + channels. Thus, in a series of experiments performed in a manner similar to that of the enantiomers of efaroxane, and as shown in Figs. 4 to 6, a similar but not unexpected difference in the potency of Î ± 2-adrenoreceptors of the two isopropyl enantiomers was determined (Fig. 4). The (+) enantiomer showed that it substantially reversed the inhibition of insulin secretion caused by the α2-agonist UK-14304, whereas the (-) enantiomer was relatively ineffective (Fig 4). This is in complete agreement with other studies which have confirmed that the Î ± 2 -agonist activity of the isopropyl analogs resides in the (+) enantiomer.

The isopropyl analogs reside in the (+) - enantiomer.

Surprisingly, however, and in contrast to the situation with the enantiomers of efaroxane, both (+) and (-) enantiomers of the isopropyl analogs were able to reverse the inhibition of insulin secretion caused by diazoxide, K + (Fig. 5). As mentioned above with respect to efaroxane, this activity was stereospecifically confined to the (-) enantiomer of efaroxane (Fig. 2). In addition, both isopropyl enantiomers of efaroxan were able to directly stimulate insulin secretion (Fig. 6) an activity that was greatly confined, in the case of efaroxane, to the (-) enantiomer (Fig. 3).

It has always been recognized that the ability of efaroxane to stimulate insulin secretion through an a2 - adrenoreceptor mechanism could be compromised by its actions on non - islet a2 - adrenoceptors. In particular, the effects on sympathetic neurons would tend to increase the circulating catecholamines and thus could counteract any direct effect on Î ± 2-adrenoreceptors in the islets. In addition, increased plasma catecholamines could also raise blood pressure in a group of patients suffering from hypertension diseases. These limitations are not apparent in the (-) - enantiomer of efaroxane or its isopropyl analogue which directly influence insulin secretion by an action on the potassium channels and which have a very low affinity for Î ± 2 -adrenoceptors. The invention also includes the use of the (-) enantiomer of a compound of Formula 1 or a nontoxic salt thereof as a K + channel blocking agent in the treatment of diabetes. The invention further includes a method of treating diabetes which comprises administering to the man an effective, potassium channel blocking, (-) enantiomer of a compound of Formula 1 or a nontoxic salt thereof. The invention further includes the use of the (-) enantiomer of a compound

The preparation of Formula 1 or a non-toxic salt thereof in the preparation of a pharmaceutical composition as a potassium blocking agent in the treatment of diabetes without producing any significant effect on an a2 -adrenoreceptor. The (-) enantiomer of the efaroxane can be prepared from the efaroxane (in the base form) by conventional methods of preparing optically active enantiomers from racemic mixtures. Thus, the efaroxane is treated in solution with an optically active (-) acid, such as (-) - dibenzoyl-L-tartaric acid, the resulting (-) salt is separated and recrystallized to obtain optical purity, the (-) enantiomer of the efaroxane after the addition of a base such as potassium carbonate. The (-) - enantiomer of the compound of Formula 1 wherein R 2 is methyl may be similarly prepared from the racemate prepared by the base catalyzed alkylation of dihydrobenzofuran-2-carboxylic acid with 2-iodopropane, the resulting 2-substituted acid being converted to 2-imidazoline, using the standard procedures. The example illustrates the preparation of the (-) and (+) enantiomers of the efaroxane and the (-) and (+) enantiomers of its isopropyl analog. Optical rotations were measured on a Perkin El-mer 141 polarimeter. EXAMPLE 1

(-) enantiomer of efaroxane

The free base of efaroxane (6.0 g) was dissolved in hot acetone (180 ml) and added to a solution of (-) - dibenzoyl-L-tartaric acid (10.44 g) in acetone (180 ml). The cloudy solution was allowed to cool to room temperature, the resulting white solid was collected by filtration, washed with diethyl ether and dried to afford the (-) - dibenzoyl tartarate of efaroxane, with 14.1 g yield, optical rotation [α] D = -82.07 ° (c = 1.00, methanol). The salt (13.8 g) was recrystallized from methanol / diethyl ether to give a white solid: yield 4.2 g, optical rotation [α] D = -125.2 ° (c = 1.03, methanol ); 4.0 g of the salt were recrystallized from

A second time in methanol / diethyl ether, to give a sample whose rotation is no longer increased in additional crystallizations; yield 3.2 g; optical rotation [α] D = -128.1 ° (c = 1.03, methanol).

0.3 g of the purified salt was stirred at room temperature with a solution of K 2 CO 3 (3 g in 10 ml H 2 O) and the resulting solution was extracted with dichloromethane. The organic layer was dried and evaporated, yielding the (-) enantiomer of the efaroxane, with a white solid; yield 0.09 g; optical rotation [< *] d = -53.8 ° (c = 1.0, methanol); mp. HCl salt 258-261 ° C. EXAMPLE 2

Enantiomer (+) of Efaroxane

The free base of efaroxane (5.0 g) was dissolved in hot acetone (180 ml) and added to a solution of (+) - dibenzoyl-D-tartaric acid (8.7 g) in hot acetone ( 180 ml). The cloudy solution was allowed to cool to room temperature, the resulting white solid was collected by filtration, washed with diethyl ether and dried, yielding the (+) - dibenzoyltartrate salt of efaroxane; yield 8.4 g, optical rotation [Î ±] D = + 90.04Â ° (c = 1.01, methanol). The salt (8.2 g) was recrystallized from methanol / diethyl ether to give a white-off solid; yield 4.3 g, optical rotation [Î ±] D = + 116.9Â ° (c = 1.05, methanol); 4.0 g of the salt were recrystallized a second time from methanol / diethyl ether to give a sample whose rotation stopped increasing on subsequent crystallizations; yield 1.1 g; optical rotation [α] D = + 118.7 ° (c = 1.03, methanol).

0.3 g of the purified salt was stirred at room temperature with a solution of K 2 CO 3 (3 g in 10 ml H 2 O) and the resulting solution was extracted with dichloromethane. The organic layer was dried and evaporated, yielding the (+) enantiomer of the efaroxane, as a white solid; yield 0.09 g, optical rotation [Î ±] D = + 52.8Â ° (c = 1.00, methanol).

EXAMPLE 3 (-) - 2- [2- (2-Isopropyl-2,3-dihydrobenzofuranyl)] - 2-imidazoline

The free base of 2- [2- (2-isopropyl-2,3-dihydro-benzofuranyl)] - 2-imidazoline (4 g) was dissolved in hot acetone (110 ml) and, with stirring , to a solution of (-) - dibenzoyl-L-tartaric acid (6.54 g) in hot acetone (85 ml). The stirred solution was allowed to cool, stirred for 2 hours at ambient temperature, and the resulting white solid was filtered off and dried, yielding (-) - 2- [2- (2-isopropyl-2,3- dihydrobenzofuranyl)] - 2-imidazoline, dibenzoyltartrate salt; yield 4.97 g; optical rotation [Î ±] D = -106.6Â ° (c = 1.00, methanol). The salt (4.8 g) was recrystallized from methanol / diethyl ether to give a white solid; yield 3.4 g; optical rotation [a] D = -116.1 '(c = 1, methanol). A second time in methanol / diethyl ether was recrystallized 3.25 g of the salt to give a sample whose rotation ceased to increase on subsequent crystallizations; yield 2.32 g; optical rotation [α] D = -122.2 ° (c = 1, methanol).

1.95 g of the purified salt was partitioned between dichloromethane and 10% aqueous sodium carbonate solution, and the organic layer was washed with further 10% sodium carbonate solution. The organic layer was washed with water, dried and evaporated to give (-) - 2- [2- (2-isopropyl-2,3-dihydro-benzofuranyl)] - 2-imidazoline as white solid; yield 0.75 g; optical rotation [Î ±] D = -29.4Â ° (c = 1, methanol), m.p. of HCl salt, 277Â ° C (dec.). EXAMPLE 4 (+) - 2- (2-Isopropyl-2,3-dihydrobenzofuranyl) -2,2-imidazoline

The parent liquors of the crystallizations of Example 3 (1 x acetone, 2 x methanol / diethyl ether) were combined and the solvents removed in vacuo. The residue was partitioned between dichloromethane and 5% aqueous sodium carbonate, and the aqueous layer was extracted with more dichloromethane. The organic phases were combined and washed successively with further solution

Sodium carbonate, water and saturated sodium bicarbonate solution. The solution was dried and evaporated, yielding a (+) - 2- [2- (2-isopropyl-2,3-dihydrobenzofuranyl)] - 2-imidazoline enriched residue (2.95 g).

A portion of this material (2.85 g) was dissolved in hot acetone (60 ml) and added to a stirred solution of (+) - di-benzoyl-D-tartaric acid (4.65 g) in hot acetone (60 ml). The stirred solution was allowed to cool, stirred for 3 hours at ambient temperature and the resulting white solid was collected by filtration and dried to give (+) - 2- [2- (2-isopropyl-2 , 3-dihydrobenzofuranyl)] - 2-imidazoline; product 4.82 g; optical rotation [Î ±] D = + 112.8Â ° (c = 1, methanol). The salt (4.6 g) was recrystallized from methanol / diethyl ether to give a white solid; yield 3.32 g; optical rotation [Î ±] D = + 120.8Â ° (c = 1, methanol). A second time, in methanol / diethyl ether, 3.1 g of the salt was recrystallized to give a sample whose rotation ceased to increase on subsequent crystallizations; yield 2.75 g; optical rotation [Î ±] D = + 121.4Â ° (c = 1, methanol).

1.5 g of the purified salt was partitioned between dichloromethane and 10% aqueous sodium carbonate solution, and the organic layer was washed with further sodium carbonate solution. The organic layer was washed with water, dried and evaporated to give (+) - 2- [2- (2-isopropyl-2,3-dihydrobenzopuranyl)] - 2-imidazoline as a white solid; yield 0.57 g; optical rotation [Î ±] D = + 29.7Â ° (c = 1, methanol), m.p. of HCl salt, 274Â ° C (dec.).

The pharmaceutical compositions may be in a form suitable for oral or parenteral administration. These oral compositions may be in the form of capsules, tablets, granules or liquid preparations such as elixirs, syrups or suspensions.

The tablets contain a compound of Formula 1 as described above.

Or a non-toxic salt thereof, in admixture with suitable excipients for the manufacture of tablets. These excipients may be inert diluents such as calcium phosphate, microcrystalline cellulose, lactose, sucrose or dextrose; granulating and disintegrating agents such as starch; binding agents such as starch, gelatin, polyvinyl pyrrolidone or acacia; and lubricating agents such as magnesium stearate, stearic acid or talc.

The capsule compositions may contain the compound or a non-toxic salt thereof, mixed with an inert solid diluent such as calcium phosphate, lactose or kaolin, in a hard gelatin capsule.

Compositions for parenteral administration may be in the form of sterile injectable preparations, e.g. ex. solutions or suspensions on p. ex. water, saline or 1,3-butane diol.

For purposes of ease and precision of dosage, the compositions are advantageously used in a unit dosage form. For oral administration, the unit dosage form contains from 1 to 500 mg, preferably from 1 to 250 mg of the compound of Formula 1 or a non-toxic salt thereof.

Claims (5)

A process for preparing the (-) - enantiomer of a compound of formula 1 where R is hydrogen or methyl in substantially pure form and the non-toxic salts thereof, characterized in that a compound of formula 1 in the form of the racemate is treated in solution with an optically active acid (-), the salt ( -), separated and recrystallized to optical purity and then the (-) - enantiomer of the compound of formula 1 obtained after the addition of a base. 2. A process according to claim 1, wherein the optically active acid (-) is (-) - dibenzoyl-L-tartaric acid. 3. A process according to claim 1, wherein the enantiomer prepared is (-) - 2- [2- (2-ethyl-2,3-dihydro-benzofuranyl)] - 2-imidazoline in the form substantially and its non-toxic salts. A process for the preparation of a pharmaceutical composition which comprises combining a compound according to claim 1 or 2 or a nontoxic salt thereof with a pharmaceutically acceptable diluent or carrier. A process for the preparation of a pharmaceutical composition for oral administration according to claim 4, characterized in that 1 to 500 mg of the compound or a non-toxic salt thereof is combined per unit dosage. 73 058 DRH / TAB / AC / 3 5180/033 -17- Lisbon, 13. 1991 By KECKITT & COLMAN PRODUCTS LIMITED = 0 OFFICIAL AGENT =