COMPOUNDS HAVING ACTIVITY AT IMIDAZOLINE RECEPTORS
This invention relates to the use of a class of compounds as selective ligands for imidazoline receptors. In particular, the invention relates to the use of β-carbolines and their derivatives for treating diseases or disorders as a result of their activity at imidazoline receptors.
The existence of imidazoline (I) receptors in mammalian brain has been recognised for over fifteen years, with a subdivision into I,- and l2-subtypes based on their pharmacology, location and putative functions. 1,-sites are those located in the brainstem proposed to mediate the hypotensive actions of a new generation of antihypertensive drugs including rilmenidine and moxonidine. I2 sites are thought to be part of the enzyme monoamine oxidase and therefore to effect the metabolism of several brain monoamines.
Historically l sites mediate the hypotensive effects of clonidine in the brainstem, notably in the lateral reticular nucleus (LRN). The existence of this functional site is supported by radioligand binding studies utilising radiolabelled clonidine and para-aminoclonidine in bovine brainstem membranes. These studies also showed this site to have an affinity for cimetidine, which supports the facts that [3H]- cimetidine also binds to non-histaminergic binding sites and clonidine potently inhibits this binding.
There are few selective l site ligands, most having a degree of affinity for α2-adrenoceptors; however, their I,- over l2-site selectivity is generally good. The oxazoline compound rilmenidine, which is chemically related to clonidine, is an antihypertensive agent which lacks the sedative effects of clonidine and shows selectivity for the I,- site.
In addition to their brainstem location where they modulate blood pressure, l sites have also been identified in bovine frontal cortex and in rat locus coeruleus where their activation increases spontaneous neuronal firing. The existence of l sites has also been reported in rat kidney, where they function to increase water and sodium excretion, in human platelets, where they are a marker of depression and dysphoric premenstrual syndrome and in rat pancreatic β cells where they mediate insulin secretion in response to efaroxan.
Most of the early evidence for l2-sites derives from radioligand binding studies utilising [3H]-idazoxan in membrane preparations from several tissues of several species. As mentioned above, clonidine, rilmenidine, moxonidine, efaroxan and the catecholamines adrenaline and noradrenaline are recognised with low affinity by l2-sites, whereas imidazolines such as naphazoline and cirazoline demonstrate high nanomolar affinity. Some compounds containing a guanidino moiety, such as guanoxan and guanabenz, also bind with high affinity. Interestingly the structurally related diuretic amiloride has been found to be of high affinity in only certain studies, notably in tissue homogenates from rabbit. This has led to the subdivision l2A-amiloride sensitive, and l2B-amiloride insensitive sites. l2-sites are widely distributed throughout mammalian brain and periphery. In rat brain l2-sites are localised to distinct brain nuclei such as interpeduncular nucleus, arcuate and pineal gland, whereas they are more widespread in rabbit and human brain. At the subcellular level they are associated with the mitochondrial fraction of membranes prepared from brain, liver, kidney, heart and striated muscle.
The function of l2-sites is still not clear. This is due in part to the lack of selective ligands since most have affinity for α2- adrenoceptors or some other family of receptors. The following are a
few examples of proposed functions. In rabbit kidney l2-site activation inhibits Na+ uptake into renal tubule cells. In cultured rat cerebral cortical astrocytes l2-site activation leads to an increase in levels of mRNA for glial fibrillary acidic protein (GFAP) . It has also been noted that l2-sites regulate levels of GFAP and that chronic treatment of rats with an l2-site selective compound, (2-(- benzofuranyD-imidazole) (2BFI), increased GFAP immunoreactivity in cerebral cortex. This association with GFAP is of great interest since the brain density of l2-sites in man increases with age and idazoxan's neuroprotective effects following brain ischaemia are proposed to be mediated via l2-sites. An l2-site selective compound has been noted to increase food consumption in rats, which may indicate a role for l2- sites in appetite.
As mentioned above, there appears to be a strong link between l2-sites and MAO-A and MAO-B, and several l2-site selective compounds have been found to inhibit amine oxidation in liver cells and adipocytes. An interaction with MAO is also supported by microdialysis studies where selective ligands can increase the availability of dopamine and noradrenaline in rat brain. Also the MAO-A inhibitor clorgyline potently inhibits l2-site binding and, furthermore, chronic dosing of rats with clorgyline downregulates l2- sites in brain.
The past two years have seen the publication of a number of highly selective l2-site ligands, which should help define a clear function for these sites. Of course without a clear function it is difficult to say whether these new chemical entities are agonists or antagonists. Indeed if the l2-site is located on MAO then one may consider the possibility of positive and negative modulators of MAO, although to date only inhibition of MAO has been reported.
Compounds such as cirazoline and naphazoline have high affinity for l2-sites and also for α adrenoceptors. Isothiocyanate- tolazoline has been reported to show moderate affinity and selectivity in guinea pig cerebral cortex and porcine renal cortex membranes. The 1 ,3-benzodioxan isomer of idazoxan is also reported to show some 100 fold selectivity for l2-sites over α2-adrenoceptors in binding studies.
The vast majority of the compounds which are known to act as ligands at imidazoline receptors contain imidazoline groups or related groups such as oxazolidine.
The present invention is based on the discovery of a class of compounds which have high affinity for imidazoline receptors, even though they do not contain an imidazoline (or related) group.
Accordingly the present invention provides the use of a compound of formula (1 ) in the manufacture of a medicament for the treatment or prevention of a disease or a disorder by selective action at an imidazoline receptor, wherein formula ( 1 ) is:
(1 ) and X is NR, O or S
R is hydrogen, ZΛ to C6 alkyl, C, to C7 acyl, C, to C6 alkyloxycarbonyl, C2 to C6 alkenyl, C2 to C6 alkenylcarbonyl or
C2 to C6 alkenyloxycarbonyl A is a ring forming a fused ring system with the ring containing X and is selected from
R\ R2, R3, R4, R5, R6 and R7 are independently selected from:
(i) hydrogen, C, to C6 alkyl, OH, NH2, C, to C6 alkylamino, di(C, to C6 alkyDamino, C, to C6 alkylcarbonyl, C, to C6 alkyloxycarbonyl, C, to C6 alkylcarbonyloxy, carboxyl, halo, halof^ to C6)alkyl, aminoJC, to C6)alkyl, hydroxy(C1 to C6)alkyl, (C, to C6 alkoxy) C, to C6 alkyl, NO2, C, to C6 alkylthio, SO3H, C2 to C6 alkenyl, C2 to C6 alkenyloxy, C2 to C6 alkenylamino, di(C2 to C6 alkenyl)amino, (C, to C6 alkyl) (C2 to C6 alkenyDamino, C2 to C6 alkenylcarbonyl, C2 to C6 alkenyloxycarbonyl, C2 to C6 alkylcarbonyloxy, halo(C2 to C6)alkenyl, amino(C2 to C6)alkenyl, hydroxy(C2 to C6)alkenyl, (C, to C6 alkoxy) C2 to C6 alkenyl, C2 to C6 alkenylthio, C2 to C6 alkynyl, C2 to C6 alkynyloxy, C2 to C6 alkynylamino, di(C2 to C6 alkynyl)amino, C2 to C6 alkynylcarbonyl, C2 to C6 alkynyloxycarbonyl, C2 to C6 alkynylcarbonyloxy, halo(C2 to C6)alkynyl, amino(C2 to C6)alkynyl, hydroxy(C2 to C6)alkynyl, (C, to C6 alkoxy)C2 to C6 alkynyl;
(ii) C, to C6 alkoxy; and
(iii) aryl and aralkyl, optionally substituted on the aromatic ring with from one to five of the groups mentioned under (i) and/or (ii) above; R' is hydrogen, C, to C6 alkyl, C, to C7 acyl, C, to C6 alkyloxycarbonyl, C2 to C6 alkenyl, C2 to C6 alkenylcarbonyl or
C2 to C6 alkenyloxycarbonyl; one or more of the pairs of groups R1 and R2, R2 and R3, R3 and R4, R5 and R6, R6 and R', R' and R7 being optionally linked to form a
fused carbocyclic or heterocyclic, aromatic or non-aromatic, ring system with the rings to which they are bonded and pharmaceutically acceptable salts and prodrugs thereof.
The compounds of the invention have surprising affinity and/or selectivity for imidazoline I, and l2 receptors relative to other receptor types and, therefore, are useful in the treatment or prevention of all diseases or disorders in which agonist or antagonist activity at these sites is pharmacologically useful. Diseases or disorders which may benefit from treatment in this way include, for example: hypertension; Alzheimer's disease; Parkinson's disease; Huntington's chorea; other neurological disorders, such as depression, anxiety and eating disorders; opiate addiction; and diabetes. Also, the compounds may be used as neuroprotective agents, following brain ischaemia or in the treatment of other neuronal injuries, for increasing water and/or sodium excretion or as MAO inhibitors.
Those skilled in the art will appreciate that compounds having, agonist activity, as well as those having antagonist activity, are potentially useful as therapeutic agents.
The compounds of the invention may have a greater affinity or selectivity for one imidazoline subtype over another (e.g., greater for I T than l2 or for l2 than I,) or may have substantially equal affinity and selectivity for both l and l2-sites.
Preferably, X is NH and the compounds are based on a β- carboline ring system.
R3 is preferably selected from the groups mentioned under (i) and (iii) above.
R6 is preferably hydrogen.
It is also preferred that R1 , R2, R4, R5 and R6 are all hydrogen.
Preferred values for R3 are either hydrogen or hydroxyl (OH), most preferably hydrogen.
R7 is preferably hydrogen or C, to C6 alkyl e.g., methyl. R', when present, is preferably hydrogen.
Compounds which have been found to be particularly suitable as ligands for I, and l2 receptors are the following:
(A) (B)
(E) (F)
(G) (H)
Of these, compounds (A) and (B) are particularly preferred on account of their high affinity and selectivity for l2 sites; and compound (C) for its selectivity for both I, and l2 sites over other receptors.
The present invention also contemplates a method of treating or preventing a disease or disorder of an animal or human of the types mentioned above, comprising the administration to a patient of a
pharmacologically effective amount of a compound of formula ( 1 ), as defined above.
The term "alkyl" as used herein is intended to cover groups having straight and branched chains and, for C3 to C6 groups, optionally substituted alicyclic rings, such as cyclopropyl, cyclobutyl, cyclopentyl and cyclohexyl, for example. The terms "acyl", "alkenyl" and "alkynyl" are defined similarly, the difference being that they comprise a terminal carbonyl group, one or more carbon-carbon double bonds and one or more carbon-carbon triple bonds, respectively.
Halo means fluoro, chloro, bromo or iodo. Groups containing halo substitutents, such as halo(C1 to C6)alkyl, may contain more than one halogen atom e.g., trifluoromethyl.
The term "aryl" covers aromatic, carbocyclic and heterocyclic, monocyclic and polycyclic ring systems such as phenyl and naphthyl, for example. "Aralkyl" means C, to C6 alkyl substituted with aryl e.g., benzyl.
Where the compounds contain a chiral centre, they may be used as one enantiomer, a mixture enriched in one enantiomer or as a racemic mixture.
The compounds of the invention may be in the form of prodrugs, such as in vivo hydrolysable esters or amides, or pharmaceutically acceptable salts such as acid addition salts (e.g., hydrochlorides) and, where the compounds contain an acidic functional group, acid salts. Certain of the compounds, and their salts, are known to form solvates and these solvates may also be used in the invention. Preferred solvates are the hydrates.
The compounds may be formulated in conventional ways for use in the invention.
While it is possible to administer the compounds of the invention directly without any formulation, the compounds are preferably employed in the form of a pharmaceutical formulation comprising a pharmaceutically acceptable excipient and at least one compound of the invention. Such compositions contain from about 0.1 % by weight to about 90.0% by weight of the compound.
In formulating the compositions, the compound is usually mixed with an excipient which can be a carrier, or a diluent or be diluted by a carrier, or enclosed within a carrier which can be in the form of a capsule, sachet, paper or other container. When the carrier serves as a diluent, it can be a solid, semi-solid, or liquid material which acts as a vehicle, excipient, or medium for the compound. Thus, the composition can be in the form of tablets, pills, powders, lozenges, sachets, cachets, elixirs, emulsions, solutions, syrups, suspensions, aerosols (as a solid or in a liquid medium), and soft and hard gelatin capsules.
The compounds of the invention may be delivered transdermally, if desired. Transdermal permeation enhancers and delivery systems, including patches and the like, are well known to those skilled in the art.
Examples of suitable carriers, excipients, and diluents include lactose, dextrose, sucrose, sorbitol, mannitol, starches, gum acacia, calcium phosphate, alginates, calcium silicate, microcrystalline cellulose, polyvinylpγrrolidone, cellulose, tragacanth, gelatin, syrup, methyl cellulose, methyl- and propylhydroxy- benzoates, talc, magnesium stearate, water, and mineral oil. The formulations may also include wetting agents, sweetening agents or flavouring agents. The formulations of the invention may be formulated so as to provide quick, sustained, or delayed release of the active ingredient after
administration to the patient by employing procedures well known in the art.
For oral administration, a compound of this invention ideally can be admixed with carriers and diluents and moulded into tablets or enclosed in gelatin capsules.
The compounds of the invention exhibit selectivity for I, and l2 receptors. Some of the compounds show further selectivity for l2 receptors over \ receptors. The term "selectivity" means that the compounds show a higher affinity for the imidazoline receptors than for other receptors which are present in the human or animal body in the same region. For example, the compounds exhibit a higher affinity for I, and l2 sites than benzodiazepine sites in the mammalian brain.
The invention is illustrated by the following non-limiting examples.
EXAMPLE 1
Compounds (A) to (H), whose structural formulae are given on page 7, were obtained from Sigma.
Methods:Rat brain and kidney P2 membranes were prepared from male Wistar rats (250g) by homogenisation in ice-cold buffer (20 vol 50mM Tris-HCI, pH7.4), and washed twice by repeated centrification. For I, binding, aliquots of kidney membrane (0.5mg protein/ml) were incubated (45 min, 22°C) in triplicate with 3nM [3H]clonidine (in the presence of 1 0 mM rauwolscine to preclude α2- adrenoceptor binding) alone or with increasing concentrations of the test compounds. For l2 binding, aliquots of brain membrane (0.5mg protein/ml) were incubated (45min, 22°C) in triplicate with 1 nM [3H]2- (2-benzofuranyl)2-imidazoline ([3H12BFI) alone or with increasing
concentrations of the test compounds. Assays were terminated by ice-cold filtration and bound radioligand determined by scintillation counting. Inhibition constants (Ki & IC50 values) were determined using computer-assisted curve fitting software, capable of fitting data to a single or two-site curve.
In the l2 imidazoline binding assay, it was found that the unsubstituted β-carboline (norharmane) (A) was able to displace [3H]2BFI from rat brain membranes with high affinity and a calculated Ki value of 90nM, whilst tetrahydro β-carboline (B) had ten-fold higher affinity with a Ki of 9nM. These compounds, both found in brain, displaced [3H]2BFI in a biphasic manner, from which nonlinear curve fitting (Prism, GraphPad Inplot) could determine two components of binding, revealing a high affinity component representing around 50% of the sites labelled, and the remainder, representing a low affinity component. This is consistent with our previous findings where the selective high affinity l2 site ligands BU224 and 2BFI show a similar pattern of displacement, probably reflecting subtypes of l2 sites and their association with monoamine oxidase.
A range of β-carbolines (harmane (C), harmine (D) and harmaline (E)) were also found to have high affinity at l2 imidazoline receptors (Table 1 ). As shown in Table 1 , the classic benzodiazepine agonist diazepam was a relatively weak inhibitor of binding to imidazoline receptors, showing that the imidazoline binding site is pharmacologically distinct from the site located on the benzodiazepine receptor. This fact was further supported by the evidence that the highest affinity compounds at the l2 receptor were demonstrated by the tetrahydro β-carboline (B) and harmaline (E) which have been found to have very low affinity for the benzodiazepine receptor (Table 1 ) .
What is clear from our present study is that these β-carbolines all have several orders of magnitude higher affinity than agmatine which has previously been proposed as the endogenous ligand.
Given the above results it was important to establish whether this range of β-carbolines also had high affinity for the I, receptor, the site reported to be responsible for the antihypertensive action of drugs such as clonidine, rilmenidine and moxonidine. The results of the assay for compounds (A) to (H) are also shown in Table 1 . In this assay, harmane (C) demonstrated high affinity whilst norharmane (A) demonstrated moderate affinity.
Table 1 Ki values at high affinity component of I, binding vs I, site and benzodiazepine affinity
Compound IC50 (nM) 1, Ki (nM) l2 IC50 (nM) Benzo * *
Tryptamine* 26,700±4300
Tryptophan * > 1 00,000
5-Hydroxytryptamine * > 100,000
Agmatine* 41 6,700±1 1 8,800
Norharmane (A) (β-carboline) 655.7 89.6 8,000
Noreleagnine (B) 8830 9.0 920,000
(Tetrahydro-β-carboline)
Harmane (C) 34.01 36.3 7,000
Harmine (D) 61 60.0 1 8.0
Harmaline (E) 1 4470.0 1 1 .7 390,000
Pinoline (F) + 3526±279 358±1 1 8
Harmalol (G) 1 000 65.0
3-Methoxycarbonylamino-β- 2963±370 61422+2951 5 carboline (H) +
Diazepam * 100,000 3.6
* Comparative Examples
* * Benzodiazepine binding, data taken from EJP (1 981 ) 70,409-416
+ Mean of four experiments
EXAMPLE 2
Rabbit (New Zealand, either sex, 1 .5-3.5 kg) brains were homogenised ( 1 0 w/v 50mM Tris-HCI buffer, pH 7.4 containing 320mM sucrose) and membranes prepared by the methods of Lione et al ( 1 996) . Aliquots of thawed membrane were incubated (45min) with 1 nM [3H]2BFI to label l2 binding sites. Specific binding was defined by 1 0μM BU224 (Lione, L., et al., (1 996) . Eur.J. Pharmacol., 304, 221 -229) . The β-carbolines were examined for their ability to compete with [3HJ2BFI over the range of 0.01 nM-1 mM. Bound ligand was separated by rapid filtration and determined by liquid scintillation counting. Results were analysed by Prism (GraphPAD Software, 1 994) . Agmatine was of low affinity whereas the precursor of the β- carbolines, tryptamine, demonstrated reasonable affinity with a K, of 3μM. Several β-carbolines displaced [3H]2BFI in a biphasic manner yielding high affinity (K,H, approx. 50% of labelled sites) and low affinity (K,L) components (Table 2). Of these compounds harmane (C) showed the highest affinity for l2 sites with a K,H of 9.7 nM. In comparison the exogenous β-carbolines, harmine (D) and harmaline (E), also demonstrated very high affinity for around 50% of labelled sites (Table 2) . Ethyl β-carboline-3-carboxylate (β-CCE) displaced from an apparent single site with very low affinity (Table 2).
Table 2 Affinity of compounds for I, binding sites in rabbit whole brain membranes
Compound KjH (nM) n K;L (nM) n
Tryptamine 3027±21 9 3
Agmatine 856633±4068 3
Norharmane (A) 52.87±7.5 4 30490.0±8240.6 4
Noreleagnine (B) 36.6±3.7 3 3041 6.7±8424.4 3
Harmane (C) 9.7±2.4 4 3591 .8±585.6 4
Harmine (D) 1 5.415.2 3 1 048.61343.2 3
Harmaline (E) 1 2.5±2.9 3 6650.7+1 104.7 3 β-CCE 7661 7±2978 3
Data are mean K, values + s.e.mean derived from three to four experiments performed in triplicate.
These data demonstrate particular β-carbolines show higher affinity for rabbit l2 binding sites than agmatine, the proposed endogenous ligand. These results parallel those for rat brain.
Lione, L., et al., (1 996) . Eur.J. Pharmacol., 304, 221 -229