WO1997000678A1

WO1997000678A1 - Methods related to the treatment of and isolation of compounds for treatment of ischaemic conditions

Info

Publication number: WO1997000678A1
Application number: PCT/AU1996/000381
Authority: WO
Inventors: Jennifer Ann Kennedy; John David Horowitz
Original assignee: The Queen Elizabeth Hospital
Priority date: 1995-06-23
Filing date: 1996-06-21
Publication date: 1997-01-09
Also published as: AUPN380695A0

Abstract

A method of isolating an anti-ischaemic drug useful for treatment of heart conditions or ischaemias including the step of screening candidate compounds for their ability to inhibit CPT-1. The anti-ischaemic drug may be structural analogues of amiodarone or perhexiline. These drugs may be useful in the treatment of other conditions including angina pectoris, evolving acute myocardial infarction, aortic stenosis, congestive cardiomyopathy, skeletal myopathy, intermittent claudication and renal or cerebral ischaemia. Also disclosed is a method of monitoring the treatment of ischaemia by assaying for the effect of perhexiline or amiodarone using a CPT-1 inhibition assay.

Description

METHODS RELATED TO THE TREATMENT OF AND ISOLATION OF COMPOUNDS FOR TREATMENT OF ISCHAEMIC CONDITIONS

FIELD OF THE INVENTION

This invention relates to ischaemia and related conditions, to the monitoring of existing treatments of ischaemic and related conditions, the treatment of ischaemic and related conditions, to a method of screening for useful compounds for the ischaemic and related conditions, and to compounds isolated for the treatment of ischaemic and related conditions.

BACKGROUND

The quality of life of individuals with a heart condition is often maintained by administration of certain drugs. A number of such drugs are presently prescribed including nitrates, β-adrenoceptor blockers and L-channel calcium antagonists.

A large group of patients with angina pectoris remains refractory to maximally tolerated conventional therapy of nitrates, β-adrenoceptor blockers and L-channel calcium antagonist These are often the patients who are not suitable for coronary bypass surgery. For these patients therapeutic options are limited, and such patients would potentially benefit from a novel, anti-ischaemic approach.

Perhexiline is a drug which has been in clinical use for over 25 years, always as perhexiline maleate. There is good evidence for its ability to relieve angina, with efficacy documented in patients refractory to the more conventional anti-anginal drugs listed above (Mir and Kafetzakis, 1978, White and Lowe, 1983, Horowitz et al, 1986, Cole et al, 1990). In a double-blind, randomised trial, perhexiline treatment improved exercise performance and quality of life, in addition to reducing anginal frequency and severity in patients already receiving optimal therapy with other anti-anginal agents (Cole et al, 1990). In addition, short-term perhexiline therapy for patients on the waiting Ust for elective coronary artery bypass surgery has been shown to be highly effective in the control of anginal symptoms (White and Lowe, 1983, Horowitz et al, 1986).

Although perhexiline is extremely effective, its use was previously associated with a high incidence of eventual development of hepatitis (Pessayre et al, 1979) and peripheral neuropathy (Laplane et al, 1978). However, recent understanding of its pharmacokinetics has helped to limit the occurrence of these side effects by monitoring plasma levels of the drug (Horowitz et al, 1986, Cole et al, 1990), especially in those patients who are slow metabolisers of the drug (Shah et al, 1982).

The basis of the development of perhexiline as an anti-anginal agent was the betief that it exerted a coronary vasodilator effect, a conclusion which has not stood the test of time. During the period from 1975 to 1980, initial enthusiasm at the considerable therapeutic efficacy of the drug was gradually tempered by increasing awareness of the insidious development in many patients of severe hepato-toxic and neuro-toxic adverse effects (Laplane etal, 1978, Horowitz and Mashford, 1979). This adverse experience led to a decline in the use of the drug, which coincided with the development of a new hypothesis for the basis of its action (blockade of L-type calcium channels) (Fleckenstein-Grun et al, 1978). As other calcium antagonists were less toxic, there was no reason to use perhexiline.

Clinical experience with perhexiline, on the other hand, tended to distinguish it from other anti-anginal agents on the basis of efficacy (Horowitz and Mashford, 1979 ,Mir and Kafetzakis, 1978, White and Lowe, 1983, Cole et al, 1990) and lack of inotropic effects (Silver, 1985). Thus perhexiline produced incremental anti-anginal effects when added to β-adrenoceptor antagonists, (Horowitz and Mashford, 1979,White and Lowe, 1983, Cole et al, 1990) calcium antagonists, (Horowitz and Mashford, 1979, Cole et al, 1990) or even a combination of these agents and prophylactic nitrates (Cole et al, 1990). Furthermore the calcium antagonistic actions are weak (Barry et al, 1985) and therapeutically there is Uttle support for these two mechanisms of action because perhexiline produces Uttle or no haemodynamic or negative inotropic effects at the doses used clinically (Pepine et al, 191 A, Pepine etal, 1973). Controlled studies have demonstrated both improved exercise performance (Mir and Kafetzakis, 1978,Cole et al, 1990) and quaUty of life (Cole et al, 1990) in such patients. Laboratory studies readily distinguished the insignificant negative inotropic effects ofthe perhexiline from those of conventional calcium antagonists (Barry et al, 1985), and there have been no reports of aggravation of cardiac failure with perhexiline in clinical use. There were also no more effects of the drug on cardiac electrophysiological parameters, although there have been occasional reports of development of torsade de pointes. Despite these favourable characteristics, because of the fear of long-term toxicity, perhexiline has been gradually relegated to a role in treating frail patients with otherwise intractable angina (Chamberlain, 1987). Prospects for preventing perhexiline toxicity have been predicated upon improved understanding of the pharmacokinetics of the drug, and the relationship between steady-state plasma perhexiline concentration, efficacy and toxicity. Perhexiline metabolism is readily saturable within usual clinical dose ranges and as a result dosage adjustments lead to disproportionate changes in plasma pe±exiline concentrations (Horowitz et al, 1981). For most patients, maximal daily clearance of perhexiline (Vmax) is approximately 600mg. A further element contributing to the marked inter individual variabiUty in the metaboUc clearance of pe±exiline relates to a genetic polymorphism ofthe major metaboUc pathway by cytochrome P4502D6 (Shah et al, 1982). Up to 10% of Caucasions are 'poor' metaboUsers with regard to this enzyme and as a result clear the drug at rates less than 10% of those who are fast metaboUsers. Our recent experience indicates that the ratio of steady state perhexiline concentration to daily drug dosage may vary approximately 100-fold within the Caucasian population (Button et al, 1993). The key to the safe long-term use of perhexiline is monitoring of the drug's plasma concentrations to guide adjustment of dose, optimisation of efficacy and minimisation of risks of toxicity.

FoUowing initial reports that perhexUine toxicity was associated with plasma perhexiline concentrations greater than lOOOmg/L (Singlas et al, 1978), two studies have estabUshed that toxicity can be prevented if plasma drug levels are kept between 150 and 600μg/L by regular monitoring; therapeutic efficacy is maintained. (Cole et al, 1990, Horowitz et al, 1986) Reports of toxicity associated with plasma concentrations in the range of 600-1200 are relatively rare (Singlas et al, 1978, Horowitz et al, 1986) and it is possible that in some cases incremental therapeutic response may be achieved associated with such higher plasma perhexUine concentrations.

The mechanism of perhexiline toxicity is moderately well understood. Histological investigations indicate the progressive development of drug-induced phosphoUpoidosis, with lameUated inclusions in hepatocytes, Schwann ceUs and other tissues (Albert and Lullman-Rauch, 1983), Similar changes have been documented with amiodarone (Simone et al, 1984).

The mechanism of beneficial effects of perhexiUne has, to date, remained less certain, despite numerous experiments. However, it is clear that the drug does not interact significantly with cell surface calcium channels at normal therapeutic doses, (Barry et al, 1985) It is tempting to postulate that the biochemical mechanisms of toxicity and anti-anginal effect are Unked. As long ago as 1980, it was hypothesised by Vaughan WilUams that perhexUine may act by causing a shift from fatty acid to glucose metabolism, thereby exerting a protective effect on the myocardium during periods of decreased oxygen availabiUty. The benefit of using less fatty acids Ues in the fact that carbohydrates require less oxygen than fats to generate the same amount of high energy phosphates (Leidtke, 1981). Such a shift in metabolism should theoreticaUy allow the heart to function more efficiently in other words it would be able to do the same amoimt of work utiUsing less oxygen, an advantage in the ischaemic heart where regional oxygen supply may be Umited. However me calculated benefit is relatively smaU (Cole et cd, 1990) and there has in the past been considerable doubt whether this could fuUy account for the considerable cUnical effect ofthe drug. This potential mechanism of perhexiline has received Uttle attention.

However very recent evidence has been obtained by Jeffrey et al (1995) to support the hypothesis that perhexiline shifts the metabolism of the rat heart away from fatty acids to carbohydrate, and that it does so in concentrations which are similar to therapeutic perhexUine concentrations in patient plasma. At the same time the rat heart was found to perform a greater amount of work for the same amount of oxygen consumed. However the authors of this study did not identify the mechanism by which perhexiline is able to alter cardiac metabolism and thereby efficiency.

Amiodarone is a drug originaUy developed as an anti-anginal agent, which has been used mainly to treat cardiac arrhythmias (reviewed by Freedman and Somberg, 1991) and has multiple actions including prolongation of action potential duration (Singh et al, 1989), L-channel calcium antagonism (Nattel et al 1987), non-competitive β- adrenoceptor blockade (Nokin et al, 1983) and sodium channel blockade (Mason et al, 1983). However, the recent GESICA trial using low dose amiodarone therapy has shown that amiodarone reduced the 2 year mortality and frequency of hospital admission in patients on optimal therapy for heart faUure and that this protective effect occurred irrespective of the presence or absence of complex ventricular arrhythmia (Dove et al, 1994). The mechanism of mis protective effect of amiodarone is not clear but could conceivably involve anti-ischaemic as distinct from anti-arrhythmic effects.

Although structuraUy dissimUar, there are a number of paraUels between perhexiline and amiodarone. They are both highly lipophiUc drags and produce similar toxic effects including hepatitis and peripheral neuropathy (Pessayre etal, 1979, Simon etal, 1984, Freedman and Somberg, 1991). The histopathology of both perhexUine and amiodarone-induced hepatitis and peripheral neuropathy is that of phosphoUpidosis (Pessayre etal, 1979, PouceU etal, 1984, Guigui etal, 1988). However, no direct studies to date have linked this phosphoUpid accumulation with the basic mechanism(s) of action of the drugs. To date, no reliable methods have been developed for predicting risk of development of amiodarone toxicity in individual patients.

Trimetazidine is a drug which is not structuraUy similar to either perhexiline or amiodarone and which has been shown to act as a prophylactic anti-anginal agent. The pecise mechamsm of action of trimetazidine is currently uncertain, but it exerts anti- ischaemic effects independent of haemodynamic charges and has been suggested to alter ceUular metaboUsm (Lavanchy et al, 1987).

Fatty acids are the 'preferred' fuel of the heart (Van der Vusse et al, 1992). They are metabolised to acetyl CoA by B-oxidation in the mitochondria of aU cells including the cardiac cells. For long chain fatty acids to enter the matrix of the mitochondrion where B-oxidation takes place, they must cross the inner mitochondrial membrane. To cross this membrane they must form long chain acyl carnitine (LC acyl carnitine), a reaction which is catalysed by the regulatory enzyme, carnitine palmitoyl transferase- 1 (CPT-1) (reviewed by McGarry et al, 1991). Once on the matrix surface of the inner mitochondrial membrane LC acylcarnitine is reconverted to LC acyl CoA by another distinct enzyme, CPT-2. Hence, CPT-1 controls the access of long chain fatty acids to the enzymes of B-oxidation. This is shown diagrammaticaUy in Figure 1.

There have been a number of drugs developed which have been shown to reduce fatty acid metabolism and increase carbohydrate metabolism by inhibiting CPT-1 and thereby preventing the long chain fatty acids entering the mitochondria of the ceU. These include 2-tetradecylglycidic acid (TDGA), sodium 2-[5(4-chlorophenyl) pentyl] oxirane-2-carboxylate (POCA), etomoxir and oxfenicine. AU of these drugs need to be converted to active metabolites in the body before they are able to inhibit CPT- 1. The first three need to be converted to CoA metaboUtes (McGarry et al, 1991) whUe oxfenicine needs to be transaminated to form 4-hydroxy phenyl glyoxylate (HPG) (Stephens et al, 1985). Only oxfenicine has been administered to cardiac patients, and this was in a small, uncontroUed study using a single dose (Bergman et al, 1980). Oxfenicine increased the time to angina during pacing and this effect was associated with reduced myocardial oxygen consumption. Simultaneously carbohydrate extraction was enhanced and fatty acid extraction was reduced by the myocardium during pacing. This patient study was conducted at a time when the mechanism of oxfenicine was never tested as a specific CPT-1 antagonist in patients. No controUed clinical trials were carried out to test whether oxfenicine had anti-anginal efficacy and interest in the drug has waned due to its toxicity in chronic animal studies (Bachmann and Weber, 1988). Continued interest in CPT-1 inhibitors is related primarily to the potential treatment of diabetes (Wolf and Engel, 1985, McGarry et al, 1991), not heart faUure or angina.

SUMMARY OF THE INVENTION

The inventors have provided the first evidence that the major biochemical basis of the anti-ischaemic effect of two structurally disparate drugs (perhexiline and amiodarone) is inhibition of the enzyme carnitine palmitoyl transferase- 1 (CPT-1) within the heart. Furthermoe theyhave shown that trimetazidine is a less potent inhibitor of CPT-1 within the heart.

The finding leads to a strategy for development of further anti-ischaemic drugs by screening structural analogues of perhexiline and/or amiodarone and/or trimetazidine for inhibition of CPT- 1, and for screening entirely unrelated compounds for inhibition of CPT-1.

Monitoring of the extent of CPT-1 inhibition in patients receiving pe±exiUne, amiodarone and or other CPT-1 inhibitors, structuraUy related or unrelated to perhexiline, amiodarone or trimetazidine, may faciUtate the therapeutic use of these agents.

Our discovery is that pe±exiUne inhibits CPT-1 in a dose-dependent manner in rat heart and Uver mitochondria. In this respect it is more potent than its major hydroxylated metaboUtes. This suggests that pe±exiUne is the active form and does not need to be converted by the body to a metaboUte to act as a CPT- 1 inhibitor. We have compared the activity of pe±exiUne to that of other known inhibitors of the enzyme and showed its potency to be of the same order of magnitude as that of the active metaboUte of oxfenicine, but one to two orders of magnitude less in potency than the main endogenous regulator of CPT-1, namely malonyl CoA. We have identified that amiodarone is also a CPT-1 inhibitor although approximately 2 to 3-fold less potent than pe±exiUne. We propose that inhibition of CPT-1 is the basis of perhexiline's anti¬ anginal action and that amiodarone may have similar actions. It should be possible to develop new, more convenient and/or more potent anti- ischaemic agents useful not only for angina, but for evolving myocardial infarction, or protection of other tissues from ischaemia. The basis of this approach would be the screening of analogues of pe±exiUne and/or amiodarone and/or trimetazidine or other compounds for inhibitory effects on CPT-1 in appropriate tissue preparations.

Accordingly in one form the invention could be said to reside in a method of isolating an anti-ischaemic drug useful for treatment of heart conditions or ischaemia in other tissues, induding the step of screening candidate compounds for Λeir abiUty to inhibit CPT-1.

Preferably the candidate compound would exhibit a dose dependent inhibition of CPT- 1, with affinity for the enzyme sufficient to achieve significant inhibition in the target tissue (e.g. myocardium skeletal muscle, platelet, kidney, brain) with normal oral or intravenous doses of the agent.

Preferably the candidates are selected from derivatives of known anti-ischaemic drugs so that the potential side-effects might be predicted or at least in some way anticipated.

In a second form the invention could be said to reside in a drug found or produced by the first form of the invention, or at least when the drug is used in treatment of an ischaemic or related condition.

In a third form the invention could be said to reside in a method of treating an ischaemic or related condition comprising the step of administering a drug identified in the first form of the invention in a pharmaceuticaUy acceptable carrier at a pharmaceuticaUy effective dose.

Reference has been made in this document to ischaemia and or related conditions that CPT-1 inhibitors might be used for and these are understood to include: angina pectoris, evolving acute myocardial infarction, aortic stenosis, congestive cardiomyopathy, skeletal myopathy, intermittent claudication or cerebral and renal ischaemia. Whilst the conditions may primarily be related to heart conditions they may also be conditions related to the protection of other tissues from ischaemia.

Alternatively the CPT-1 inhibitors might also be useful in Umiting ischaemic damage to heart or brain during cardiac or neuro-surgery, and in Umiting platelet aggregation in patients prone to thrombosis.

In a fourth form the invention could be said to reside in a method of monitoring the plasma level of a drag useful for treating an ischaemic or related condition which drag is effective by reason of its inhibition of the enzvme CPT-1, said step of monitoring including the method of monitoring the level of activity of CPT- 1 in individuals receiving treatment with the drug.

Preferably the condition is a heart condition.

It is proposed that Perhexiline and amiodarone therapy may be monitored more accurately on the basis of the extent of inhibition of CPT-1 activity (for example in circulating platelets). This may provide a particular advance with respect to amiodarone considering the lack of a reUable method for predicting its toxicity at present

In a fifth form the invention could be said to reside in a method of treating an ischaemic or related condition including administering a drug effective by reason of its inhibition of the enzyme CPT-1, the step of monitoring the level of said drug including the monitoring of the level of activity of CPT- 1 , and the step of adjusting the level of drag administered by taking into account the level of inhibition of CPT- 1 so monitored.

Preferably the condition is a heart condition.

BRIEF DESCRIPΗON OF THE DRAWINGS

For a better understanding the invention wiU now be described with reference to several examples and reference to drawings wherein:

Figure 1 is a diagrammatic representation of the transfer of fatty acids across the mitochondrial membranes, to iUustrate the role of Carnitine palmitoyl transferase I and is adapted from McGarry etal (1991),

Figure 2 is a graph of the inhibition of rat heart CPT- 1 by perhexiline maleate, amiodarone, trimetazidine, and malonyl CoA, the Y axis being the percentage inhibition of CPT- 1, and the X axis showing the concentration of inhibitor in μmol/L, Figure 3a is a Lineweaver Burk plot of the effect of palmitoyl CoA concentration on the inhibition of CPT-1 derived from rat hearts by perhexiline the range of perhexilne concentrations bemg 0, 100, 150, and 200 μmol/L; the X axis plotting 1/concentration of palmitoyl CoA concentration in 1/ μmol/L, the

Y axis plotting the inverse of the rate of CPT-1 action in nmol/mg protein /min,

Figure 3b is a Lineweaver Burk plot of the effect of carnitine concentration on the inhibition of CPT- 1 derived from rat hearts by pe±exiUne the range of perhexUine concentrations being 0, 100, and 200 μmol/L; the X axis plotting 1/concentration of carnitine concentration in 1/ μmol L, the Y axis plotting the inverse of the rate of CPT-1 action in nmol/mg protein /min., and

Figure 3c is a Lineweaver Burk plot of the effect of palmitoyl CoA concentration on the inhibition of CPT-1 derived from rat liver by perhexUine the range of perhexUine concentrations being 0, 50, 100, 150, and 200 μmol/L; the X axis plotting 1/concentration of palmitoyl CoA concentration in 1/ μmol/L, the Y axis plotting the inverse of the rate of CPT- 1 action in nmol/mg protein /min.

Figure 4a shows the effect of nagarase on inhibition of cardiac CPT- 1. Mitochondria were incubated with nagarase 5μm/ml (open symbols) or vehicle (close symbols) for 10 min at 37°C. Following treatment wrth 20% BSA w/v, and washing, the mitochondria were preincubated for 15 min with inhibitors and then incubated for 4 min with palmitoyl-CoA 50 μmol/L and carnitine 400 μm/L at 37°C. Malonyl-CoA ■, HPG • , Perhexiline A; and nagarse pre-treated, Malonyl-CoA D, HPG O, pe±exiUne Δ. Values are means ± SEM. * P<0.05, compared with no nagarse treatment, paired t-test, n=4. CPT-1 activity was 19.3+1.4 nmol/mg protein /min in the nagarse pre-treated mitochondria (n=4).

Figure 4b is the same plot as in Figure 4a except that the the inhibitors were Malonyl- CoA ■, Amiodarone •, Trimetazidine A; and nagarse pre-treated,

Malonyl-CoA D, Amiodarone O, Trimetazidine Δ. Figure 5a Yonetani-TheoreU plot of the inhibition of cardiac mitochondrial CPT-1 by malonyl-CoA, in the presence A; and in the absence H of perhexiline 60μmol/L. Enzvme preparations were preincubated with the inhibitors for 15 mins and then CPT-1 activity estimated by incubating with palmitoyl- CoA 50μmol/L and carnitine 400μmol/L for 4 min. Values are means ±

SEM. *P<0.05, compared with no pe±exiUne, paired t-test, n=5.

Figure 5b is the same plot as in Figure 5a except with HPG,

Figure 5c is the same plot as in Figures 5a and 5b except with CoA,

Figure 6 is a graph of the inhibition of ADP-induced platelet aggregation by perhexiline D and amiodarone , the Y axis being % inhibtion of aggregation and the X-axis the log of concentration of inhibition in mol/L, and

Figure 7 shows the chemical structure of pe±exiUne and examples of strucmral analogues of pe±exiUne that may be useful.

EXAMPLE 1

Inhibition of rat heart CPT-1 with varying levels of Amiodarone, perhexiline maleate.

Tissue Preparation Male Sprague-Dawley rates of 300g body weight were kiUed by exsanguination under Ught ether anaesthesia. The heart and liver were rapidly removed and placed on ice. Homogenates and mitochondria were prepared essentiaUy as described by Kiorpes et al (1984) based on the method of Johnson and Lardy (1967). Briefly, hearts were homogenised at 4°C in isolation buffer (lOmmol/L Tris HCl, ph7.4, containing 250mmol/L sucrose and lmmol/L EDTA) at 1 :5 (w:v), using a Polytron PT-3000 (Kinematica). The post-nuclear supernatant was obtained after centrifugation for 10 minutes at 600xg. In some experiments the supernatant was used as such, whUe in other experiments a mitochondrial peUet was obtained by centrifugation at 4°C for 10 minutes at 7,000xg. The peUet was washed by resuspension to the original volume in isolation buffer and centrifugation at 7,000xg for 10 minutes, followed by resuspension in isolation buffer to approximately 40mg/ml protein. FoUowing protein estimation by BioRad protein assay kit using BSA standard, the homogenate and mitochondrial preparations were dUuted to an appropriate protein concentration for the CPT-1 assay.

CPT-1 Assay. CPT-1 activity was measured by the formation of palmitoyl-^I^-carnitine from palmitoyl- CoA and (³H)-1 carnitine, essentiaUy as described by McGarry et al (1978), with the exception that a preincubation period was used prior to enzyme estimation as described by Kiorpes et al (1984). The incubate consisted of 0.4ml of incubation buffer (pH 7.4) containing 6.25 μmol Tris-HCl, 72 μmol KCl, 2.85 μmol reduced glutathione, 1.45 μmol KCN, 1 μmol MgCl2, 5 μl ethanol, 2.7mg defatted bovine serum albumin (BSA), and potential inhibitors as indicated below. Preincubation in a shaking water bath at 37°C was initiated by addition of 50μl of homogenate or mitochondrial preparation. For assays of the cardiac enzyme 50 μg of mitochondrial protein and 150 μg of homogenate protein were utilised per assay to ensure linearity with protein concentration. Corresponding protein concentrations were increased to lOOμg and 300 μg for assays of the hepatic enzyme. The reaction was initiated by addition of 50 μl of substrate mix containing palmitoyl -CoA and (?H)-l -carnitine. Unless otherwise indicated the final concentrations were 100 μmol/L palmitoyl-CoA and 400 μmol/L carnitine. The reaction was terminated by the addition of 50 μl of concentrated HCl. Samples were dUuted with 1.45ml of distiUed water and then extracted with 1 ml n-butanol as described by Kiorpes et al (1984). Blanks were identical in composition to reaction mbes except that concentrated HCl was added prior to addition of substrate mix. The incubation time was 4 min for dose response curves to inhibitors. However for kinetic smdies utiUsing lower concentrations of palmitoyl- CoA and 1 -carnitine, the incubation time was shortened to 1.5 min since preliminary smdies indicated that the rate of palmitoyl carnitine formation was linear at aU substrate concentrations up to 2 min.

Platelet aggregation studies Platelet aggregation is quantitated via whole blood aggregometry utliUzing graded concentrations of ADP as a pro-aggreganL Extent of inhibition of ADP-induced aggregation via putative anti-aggregatory agents is determined utiUzing blood to which the requisite concentration of CPT- 1 inhibiton had been added five (5) minutes prior to ADP. Results

Effects of perhexiline (Pex) and amiodarone (Ami) on carnitine palmitoyl transferase -1 (CPT-1) activity, as measured by the rate of formation of ^H-palmitoyl carnitine from palmitoyl-CoA and ^H-camitine, were compared with those of known inhibitors of CPT-1, malonyl-CoA (Mal-CoA) and 4-hydroxy phenyl glyoxylate (HPG), in rat cardiac mitochondria. The breakdown product of pe±exiUne hydroxylated PerhexUine (OH-Pex) was also tested for inhibitory action.

The concentrations of the inhibitors (μmol/L) that gave either 30% inhibition IC30 or 50% inhibition IC50 were determined as a measure of the strength of inhibition. These are set out below in Table 1.

TABLE 1

Agent Mal-CoA HPG Pex Ami OH-Pex

IC30 0.36 16 53 92 95 (μmol/L) (0.26-0.49) (7-34) (44-64) (78-108) (55-166)

IC50 1.23 42 77 >200 >200 (μmol/L) (0.76-1.97) (12-148) (63-96)

(mean [95% confidence limits]); n=4-7

Clearly PerhexUine inhibits CPT-1 but is not as effective as Malonyl CoA. Amiodarone and the hydroxylated Pe±exiUne also inhibit CPT-1 but less effectively than Perhexiline.

Kinetic analysis indicated that Pex is a competitive inhibitor ofthe enzyme with respect to palmitoyl-Co A, but non-competitive with respect to carnitine. In this regard it was simUar to malonyl-CoA.

Clearly Pex and Ami are CPT-1 inhibitors. This effect may account for the anti- ischaemic effects of Pex and contribute to those of Ami. Dose dependence inhibition of rat heart CPT-1

Inhibition of CPT-1 activity was measured as in example 1 but only for perhexiline, maleate, amiodarone, trimetazidine, and malonyl CoA. A variety of inhibitor concentrations were used and a plot of the inhibition is seen in Figure 1.

Rat heart CPT-1 is inhibited dose dependently by pe±exiUne and amiodarone and trimetazidine. The affinity of the enzyme for inhibitors is in the order malonyl-CoA> pe±exiUne > amiodarone > trimetazidine. In addition pe±exiUne and trimetazidine display a much steeper dose response curve for inhibition than either malonyl-CoA or amiodarone. This may be related to preliminary evidence that pe±exiUne shows positive cooperativity with respect to CPT-1 inhibition in rat heart.

EXAMPLE 2. Perhexiline is competitive with respect to palmitoyl CoA in rat heart and rat liver.

The competition of inhibition of activity of CPT-1 isolated from mitochondria of the rat heart and rat Uver and activity by pe±exiUne against palmitoyl CoA and carnitine, concentration was tested. Assays were conducted as described in Example -1.

The inhibition of CPT- 1 by a range of concentrations of perhexiline was measured in conjunction with a range of concentrations of palmitoyl CoA, and carnitine.

Lineweaver-Burk plots of inhibition of CPT- 1 by pe±exiUne demonstrate that perhexiline is competitive with respect to palmitoyl-CoA in rat heart and rat Uver as evidenced by the increase in Km for palmitoyl-CoA, with no change in Vmax, with increasing perhexiline concentrations. In contrast, perhexiline is non-competitive with respect to carnitine since increasing concentrations of perhexUine have no effect on the Km for carnitine but decrease the V max of CPT- 1. In this respect inhibition of CPT- 1 by pe±exiUne is simUar to that of malonyl-CoA. The source of the enzyme CPT-1 appears to have no significant effect on this characteristic, in as much as the CPT- 1 isolated from rat Uver did not behave differently to the CPT-1 isolated from mitochondria isolated from rat heart.

The data in Figure 4a indicate that pretreatment of cardiac mitochondria with the protease, nagarse, markedly reduced the degree of inhibition of cardiac CPT-1 by both malonyl-CoA and HGP, but had no significant effect on the inhibition caused by perhexUine. Nagarse treatment did not significantly affect the basal CPT-1 activity in these preparations. These data indicate a difference between the inhibitory site for pe±exiUne on the one hand, and that for malonyl-CoA and HPG on the other. Nevertheless, the abiUty of perhexiline to interfere with the binding of both malonyl- CoA and HPG to cardiac CPT-1 is demonstrated by the Yonetani-TheoreU plots of Figure 5a and 5b respectively. Perhexiline 60μmol/L competed with HPG for inhibition of cardiac CPT-1, as demonstrated by the paraUel plots of HPG in the presence, and absence of prehexiUne. However, pe±exiUne appeared to inhibit the effect of lower concentrations of malonyl CoA on CPT-1 and the effect did not appear to conform to either a competitive or non-competitive interaction. In contrast perhexiUne displayed a non-competitive interaction with Co-A, indicating that it acts at a separate site from CoA on cardiac CPT- 1 (Figure 5c).

Despite the finding that perhexiline displayed competitive inhibition with respect to paUtoyl-CoA, the current experiments do not permit deteimination of the site of binding of perhexiUne to CPT- 1. Previous data suggest that malonyl-CoA, the principal endogenous inhibitor of CPT-1, interacts with two sites. Although malonyl-CoA is structurally simUar to palmitoyl-Co A and inhibits CPT-1 competitively, most of its inhibitory effect is exerted remote from the active site, probably at a 'regulatory site" on the cytoplasmic side of the outer mitochondrial membrane (Murthy and Pande, 1987) . Moreover, drags such as HPG (the active metabolite of oxfenicine) appear to bind to the same cytoplasmic site on hepatic CPT-1 as malonyl-CoA (Cook, et al ,1994).

From the current experiments, perhexUine appears to act at a mitochondrial site which is protected from the action of protease treatment. In this regard it differs from both HPG and malonyl-CoA, whose inhibitory effects on cardiac CPT-1 (Murthy and Pande, 1987) and hepatic CPT-1 (Murthy and Pande, 1987, Kashfi etal, 1994) are reduced after protease treatment of the mitochondria. Nevertheless, the dual inhibitor studies indicated that perhexUine interacts in some way with the malonyl-CoA binding site in that its inhibition is less than additive with both HPG and malonyl-CoA on cardiac CPT- 1. The interaction of perhexiline and malonyl-CoA with cardiac CPT-1 was more compUcated than that of perhexiUne with HPG, in that coincubation with malonyl -CoA and perhexUine did not lead to greater inhibition than that due to perhexiUne alone. At present we have no explanation for the difference between perhexiUne's interaction with HPG and that with malonyl-CoA. Although the binding site for perhexiUne on CPT-1 is protected from protease treatment in a simUar way to that for active-site directed agents like Co-A (Kashfi et al, 1994), the non-competitive interaction of perhexiUne with Co-A indicates that they inhibit the enzyme at different sites. HPG and malonyl- CoA both show similar non-competitive interactions with Co-A on hepatic CPT-1 (Kashfi et al, 1994). Hence perhexUine appears to be unique amongst CPT-1 inhibitors so far described. Pe±exiUne appears to interact with the same site on CPT-1 as HPG and malonyl -CoA, but unlike the latter two agents it does not appear to require a protein facing the cytoplasmic aspect of the mitochondrion for inhibitory activity. Hence Λe inhibitory site for pe±exiUne on CPT-1 is not identical to that of any previously described endogenous or exogenous inhibitor.

Cardiac mitochondria were preincubated with a range of concentrations of trimetazidine and CPT-1 activity was determined. Trimetazidine produced a dose-dependent inhibtion of cardiac CPT-1 (Figure 2) with an IC₅0 of 1.31 (0.95-1.81) mmol/L (n=5) and a threshold at approximately 200 μmol/L. As such it is much less potent as a CPT- 1 inhibtor than either perhexiUne, amiodarone or the endogenous inhibitor malonyl -CoA. When mitochondria were pretreated with nagarse, 5μmol/L, there was no significant effect on the inhibitory effect of trimetazidine or amiodarone (Figure 4b) while the inhibitory effect of malonyl-CoA was significantly reduced. In this respect trimetazidine and amiodarone were simUar to perhexiline. In dual inhibitor smdies, amiodarone was non-competitive with respect to Co-A in a manner simUar to the non¬ competitive interaction of perhexiline with Co-A (Figure 5c). Hence, pe±exiUne, amiodarone and trimetazidine aU bind to a nagarse-insensitive biding site on cardiac CPT- 1. This site appears to differ from the binding site for Co-A on CPT-1.

Inhibition of platelet aggregation by perhexiUne has been reported previously (Ono & Kimura, 1981), but has not been perceived as a therepeutically relevant observation. Nevertheless, the phenomenon might have important clinical impUcations. Firstly, platelet hyperaggregabiUty is associated with ischaemic heart disease and is considered as a possible aetiological factor of cardiovascular abnormaUties. Secondly, suppression of increased platelet aggregabUity by a drag such as perhexiUne might be beneficial in Λe Λerapeutic management of angina pectoris, complementary to anti-platelet effects exerted by aspirin, or by nitroglycerine and related nitrates.

The effects of perhexiline and amiodarone on platelet aggregation (Figure 6) indicate Λat Λe relative potency of Λese two drags as inhibitors of platelet aggregation is commensurate wiΛ Λeir effects on CPT-1 activity in rat heart. It is Λerefore possible, but not certain, Λat Λis effect is mediated via CPT- 1. Inhibition of platelet aggregation is useful in Λe treatment of myocardial ischaemia, but also in Λe prevention of thrombosis in oΛer vascular beds. This finding Λerefore provides an example whereby Λe identification of a compound as a CPT-1 inhibitor may faciUtate its future Λerapeutic utility via manipulation of ceUular and/or organ fucntion in any tissue in which CPT-1 activity is a significant determinant of ceUular metaboUsm. Such tissues and ceUs include heart, skeletal muscle, kidney, lungs, brain, platelets and lymphocytes.

EXAMPLE 3

Method of screening and testing for new compounds

InitiaUy candidate compounds will be incubated with cardiac mitochondria in vitro, and effects on CPT-1 activity documented. Any potentially useful compound would have CPT-1 inhibiting effects at least comparable to Λose of amiodarone.

Examples of what are beUeved to be possible candidate compounds are shown in

Figure 7, and Λeir structures compared to perhexiUne as marked.

EXAMPLE 4

Method of monitoring level ofCPT-1 as a measure of monitoring perhexiline effect.

Monitoring of Λe level of CPT- 1 may be achieved by monitoring Λe level in blood ceUs, perhaps leucocytes or platelets, or alternatively from skin fibro blasts. These could be tested according to a number of different meΛods which are outiined in Λe foUowing scientific papers which are incoφorated herein by reference.

Scholte et al J Neurological Sciences, 1979: 40; 39-51, Trevisan et al Neurology 1984:34;353-356, Iida et al Molecular and Cellular Biochemistry 1991: 103 ;23-30, Pourfarzam et al Clin Chem 1994:40;2267-2275.

EXAMPLE 5

Method of treating ischaemia

MeΛod of treating patients wiΛ Λe drags wiU be determined empiricaUy in accordance wiΛ known meΛods. Thus for example for perhexiline Λe following is a guide.

Initial dose of perhexiUne maleate is 100 mg twice daily taken oraUy in encapsulated or tablet form. One week later plasma levels of Λe drag are determined and Λe dose adjusted to bring Λe plasma concentration within Λe range 150-600 mg L. Plasma levels are checked at least once every Λree monΛs. Dosing may range from 400 mg per day to 500mg per week. The dose and plasma level to be maintained wiU be determined for each of Λe compounds separately.

REFERENCES

Albert C, LuUman-Rauch R. Arzneim Forsch 33; 125-7 (1983)

Bachmann E, Weber E. Pharmacology 36: 238-248 (1988).

Barry WH, Horowitz JD, SmiΛ TW. Br J Pharmacol 85:51-60 (1985).

Bergman G, Atkinson L, Metcalfe J, Jackson N, Jewitt DE. European Heart Journal 1:247-253 (1980).

Button IK, Davies HR, Zeitz CJ Wuttke RD, Horowitz JD. Aust NZ J Med 25;599 (1993)

Chamberlain DA. Br Heart J 58:547-51 ( 1987)

Cole PL, Beamer AD, McGowan N, CantiUon CO, Benfell K, Kelly RA, et al. Circulation 81:1660-1270 (1990).

Cook, et al J Biol Chem 269:8803-8807 (1994)

Doval HC, Nul DR, Grancelli HO, Penrone SV, Bortman GR, Curiel R. Lancet 344:493-498 (1994).

Fleckenstein-Grun G, Fleckenstein A, Byon YK, Kem KW. In: PerhexUine maleate. Proceedings of a symposium. Amsterdam: Excerpta Media, 1978: 1-22.

Freedman MD, Somberg JC. J CUn Pharmacol 31:1061-1069 (1991).

Guigui B, Perrot S, Berry JP, Fleury-FeiΛ J, Martin N, Metreau JM, Dhumeaux D, et al. Hepatology 8:1063-1068 (1988).

Horowitz JD, Mashford ML. Med J. Aust 1: 485-8 (1979) Horowitz JD, Morris PM, Drummer OH, Goble AJ, Loius WJ. J Pharm Sci 70; 320- 1 (1981)

Horowitz JD, Sia STB, Macdonald PS, Goble AJ, Louis WJ. Int J Cardiol 13:219- 229 (1986).

Hudak WJ, Lewis RE, Kuhn WL. J Pharmacol Exper Ther 173:371-382 (1970).

Jeffrey FMH, Alvarez L, Diczku V, Sherry D, MaUoy CR. J Cardiovasc Pharmacol 25: 469-472 (1995).

Kashfi etal BBA 1212:245-252 (1994).

Lavanchy N, Martin J, Rossi A. Arch Int Pharmacol Ther. 286: 97-110 (1987).

Laplane D, Bousser MG, Bouche P, Touboul PJ. in : PerhexiUne Maleate. Proceedings of a Symposium Amsterdam, Excepta Medica: 89-96 (1978).

Leidtke AJ. Progress in Cardiovascular Diseases 23:321-335 (1981).

Mason JW, Hondeghem LM, Katzung BG. Pflugers Arch 396:79-81 (1983).

McGarry JD, Sen A, Esser V, Woeltje KF, Weis B, Foster DW. Biochimie 73:77-84 (1991).

Mir MA, Kafetzakis EM. Am Heart J 96:350-354 (1978).

Murthy and Pande, Proc Natl Acad Sci USA 84: 378-382 (1987)

Nattel S, Talajic M, Quantz M, DeRoode M. Circulation 76:442-449 (1987).

Nokin P, CUnet M, Schoenfeld P. Biochem Pharmacol 32:2473-2477 (1983).

Ono H, Kimura M. Arzneim. Forsch. 31:1131-1134 (1981).

Pepine CJ, Schang SJ, BemiUer CR. Post Grad Med J 49(Supp 3):43-46 (1973). Pepine CJ, Schang SJ, BemUler CR. Circulation 49:887-893 (1974).

Pessayre D, Bichara M, Feldmann G, Degott C, Potet F, Benhamou J-P. Gastroenterology 76:170-177 (1979).

PouceU S, Ireton J, Valencia-Mayoral P, Downar E, Larratt L, Patterson J et al. Gastroenterology 86:926-936 (1984).

Shah RR, Oates NS, Idle JR, SmiΛ RL, Lockhart JDF. Br Med J 284:295-299 (1982).

Silver PJ. Eur J Pharmacol 1985; 106: 217-8

Simone JB , Manley PN, Brien JF, Armstrong PW. N Engl J Med 311: 167-172 (1984).

Singh BN, Venkatesh N, Nademanee K, Josephson MS, Kannan R. Prog Cardiovasc Dis 31:249-237 (1989).

Singlas E, Goujer MA, Simon P. Eur J Clin Pharmacol 14: 195-201 (1978)

Stephens TW, Higgins AJ, Cook GA, Harris RA. Biochem J 227:651-660 (1985).

Van der Vusse GJ, Giatz JR£, Stam HCG, Reneman RS. Physiological Reviews 72:881-940 (1992).

Vaughan WiUiams EM. Anti-arrhythmic Action and Λe Puzzle of Perhexiline, London, Academic Press (1980).

White HD, Lowe JB. Int J Cardiol 3:145-155 (1983).

Wolf HOP, Engel DW. Eur J Biochem 146:359-363.

Claims

1 A meΛod of isolating an anti-ischaemic drug useful for treatment of a heart condition or extracardiac ischaemia including Λe step of screening candidate compounds for Λeir abiUty to inhibit CPT- 1.

2. The meΛod of claim 1 including Λe step of checking for dose dependent inhibition of CPT- 1.

3. The meΛod of claim 1 including Λe step of checking for reversal of Λe inhibition of CPT-1 by a protease.

4. The meΛod of claim 3 wherein Λe protease is nagarse.

5. The meΛod as in eiΛer claim 1 or 2 wherein Λe candidates are selected from structural analogues of anti-ischaemic drags.

6. The meΛod as in claim 1 wherein Λe candidates are selected from structural analogues of perhexiUne.

7. The meΛod as in claim 1 wherein Λe candidates are selected from structural analogues of amiodarone.

8. The meΛod as in claim 1 wherein Λe candidates are selected from structural analogues of trimetazidme.

9. An anti-ischaemic agent identified by screening candidate compounds for Λeir abiUty to inhibit CPT-1.

10. An anti-ischaemic agent as in claim 9 wherein Λe CPT- 1 inhibition is shown by a detectable reduction in CPT-1 activity wiΛ less Λan or equal to lmmol/L of Λe compound in at least one potential target tissue.

11. An anti-ischaemic agent as in claim 9 giving dose dependent inhibition of CPT- 1.

12. An anti-ischaemic agent as in claim 9 wherein Λe inhibition of CPT- 1 is not reversed by a protease.

13. An anti-ischaemic agent as in claim 12 wherein Λe protease is nagarse.

14. The agent of claim 9 wherein Λe agent is a structural analogue of perhexiline.

15. The agent of claim 9 wherein Λe agent is a structural analogue of amiodarone.

16. The agent of claim 9 wherein Λe agent is a structural analogue of trimetazidine.

17. A meΛod of treating an ischaemic or related condition comprising Λe step of administering an anti-ischaemic agent according to any one of claims 9 to 16 in a pharmaceuticaUy acceptable carrier at a pharmaceuticaUy effective dose.

18. A meΛod according to claim 17 wherein Λe condition is selected from Λe group comprising, angina pectoris, evolving acute myocardial infarction, aortic stenosis, congestive cardiomyopaΛy, skeletal myopaΛy, intermittent claudication and cerebral or renal ischaemia.

19. A meΛod according to claim 17 wherein Λe condition is an ischaemia.

20. A meΛod according to claim 19 wherein Λe ischaemia is a heart condition.

21. A meΛod as in claim 17 wherein Λe meΛod is used for Umiting ischaemic damage to heart or brain during cardiac or neuro-surgery.

22. A meΛod as in claim 17 wherein Λe meΛod is for limiώig platelet aggregation in patients prone to Λrombosis.

23. A meΛod of monitoring Λe effect of a drug useful for treating an ischaemic or related condition which drug is effective by reason of its inhibition of Λe enzyme CPT- 1, said step of monitoring including Λe meΛod of monitoring Λe level of activity of CPT-1 in individuals receiving treatment wiΛ Λe drag.

24. A meΛod as in claim 23 wherem Λe condition is a heart condition.

25. A meΛod as in eiΛer claim 23 or 24 wherem Λe drag is eiΛer perhexiUne or amiodarone.

26. A meΛod of treating an ischaemic or related condition including administering a drag effective by reason of its inhibition of Λe enzyme CPT- 1 , Λe step of monitoring

Λe effect of said drug including Λe monitoring of Λe level of activity of CPT- 1 , and Λe step of adjusting Λe level of drag administered by taking into account Λe level of inhibition of CPT- 1 so monitored.

27. A meΛod as in claim 26 wherein Λe condition is a heart condition.