Therapeutic use of compounds

This invention relates to uses of compounds. In particular, it relates to the use of compounds in the treatment or prevention of diseases and biomedical conditions such as, seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea and schizophrenia, ischemia, cancer and fatal blood loss.

Epilepsy is a widespread, serious neurological condition presenting considerable personal, social and economic difficulty. It affects 0.5-1% of the population, of which 30% have epilepsy that is not adequately treated with present antiepileptic drugs (Bialer and White 2010). These patients have a high mortality and morbidity rate. The understanding of cellular and molecular aspects of seizures giving rise to epilepsy is unclear, although current research has centered on defining the molecular pathways necessary for seizure progression and the development of new treatments for seizure control.

Valproic acid (VPA; 2-propylpentanoic acid; Epilim.RTM.), a short chained branched fatty acid, is the most widely used anti-epileptic world-wide, but its mechanism of action in seizure control has remained relatively unclear for over 40 years (Lagace et al., 2005; Perucca, 2002). Having been accidentally found to be effective in seizure control (Carraz G., 1967), VPA is now also used for bipolar disorders and migraine treatment, in addition to a variety of potential new therapies including cancer and HIV treatment. VPA has previously been shown to have a chronic effect in controlling inositol depletion (Williams et al., 2006; Williams, 2005), and this long-term effect is likely to be related to its efficacy in bipolar disorders.

With regard to epilepsy treatment, the therapeutic effects of VPA have been proposed to occur via directly elevating gamma-amino butyric acid (GABA) signalling (Lagace et al., 2005) and inhibiting sodium channel activity (Costa et al., 2006). Of prime importance in VPA's mechanism of action in epilepsy treatment is that it blocks seizure activity acutely--within 30 minutes of administration--corresponding to the peak concentration of VPA in the brain following intravenous injection (Aly and bdel-Latif, 1980). Despite such rapid action suggesting a direct action on channels or a biochemically-based (rather than transcriptionally-based) epilepsy target, few acute effects of VPA have been identified (Lagace et al., 2005), making rapid VPA-catalysed effects of great potential therapeutic importance. The acute effect of VPA was recently analysed using the simple model Dictyostelium (Xu et al., 2007). It demonstrated that VPA induced an inhibition of phosphatidylinositol-(3,4,5)-trisphosphate (PIP.sub.3) production and a reduction in phosphatidylinositol monophosphate (PIP) and diphosphate (PIP.sub.2) phosphorylation.

Bipolar disorder post mortem brain samples also show altered levels of enzymes associated with fatty acid turnover (Kim et al., 2009) as well as altered fatty acids within cell membranes (Chiu et al., 2003). Numerous studies have also shown an increase in arachidonic acid (AA) release after seizure catalysed by increased PLA.sub.2 activity (Siesjo et al., 1982, Rintala et al., 1999, Bazan et al., 2002, Basselin et al., 2003), and attenuation of this process may thus provide some benefit in seizure control and epileptogenesis (Rapoport and Bosetti, 2002).

VPA has been shown to reduce AA turnover in the brain through an unknown mechanism (Chang et al., 2001). AA is an essential fatty acid and is the major polyunsaturated fatty acid in most membrane phospholipids (Svennerholm, 1968), and plays a central role in inflammatory signalling (Yedgar et al., 2006). It remains unclear if the effect of VPA on AA turnover is related to specific VPA-treatable conditions. For example, this effect may be related to bipolar disorder prophylaxis (Rapoport, 2008b) since a similar reduction in AA signalling has also been observed with other structurally independent bipolar disorder treatments such as lithium (Basselin et al., 2005) and carbamazepine (Bazinet et al., 2006a).

Although widely prescribed for multiple diseases, VPA has a number of unwanted side effects including teratogenicity and hepatotoxicity. Therefore, more potent antiepileptic drugs with reduced side effects are urgently needed.

International patent application WO 99/02485 discloses a family of VPA analogs for treating epilepsy, migraine, bipolar disorders and pain. The compounds specifically disclosed in WO 99/02485 are 2-propylheptylacetic acid, 2-propyldecanyl acetic acid and 1-O-stearoyl-2-propylheptylacetoyl-sn-glycero-3-phosphotidylcholine.

Using the biomedical model Dictyostelium, the inventors found that the effect of VPA is to cause a rapid attenuation of phosphoinositide turnover, and this effect is not based upon the direct inhibition of phosphatidylinositol-3-kinase (PI3K) activity, nor is it caused through regulation of inositol recycling. They also found that VPA induced both a reduced release and an increased uptake of radiolabelled AA and palmitic acid (a saturated long chain fatty acid). This VPA-catalysed effect is not caused by reducing fatty acid activation.

In addition, structure-activity relationship (SAR) studies showed a high degree of structural specificity for these mechanisms of action. This enabled the identification of a group of compounds showing therapeutic potential similar to VPA but with the potential for reduced side effects and/or increased therapeutic efficacy.

In accordance with a first aspect of the present invention, there is provided a compound having the Formula R1-COOH (I) wherein R1 is an alkyl or alkenyl group having a C.sub.7-11 backbone, optionally branched with a C.sub.1-6 alkyl group at any C position in the backbone, or a pharmaceutically acceptable salt, amide or ester thereof, wherein the backbone of the alkyl or alkenyl group, and/or the branched alkyl groups, are optionally interrupted by one or more heteroatoms, provided that when R1 is an alkyl group having a C.sub.7 backbone, the branching does not consist only of a hexyl group at the .alpha. carbon of R1, or only of a methyl group at the .gamma. carbon of R1, or of only single methyl groups at both the .beta. and .omega.-1 carbons of R1, and provided that when R1 is an alkyl group having a C.sub.8 or C.sub.11 backbone, the branching does not consist only of a propyl group at the .alpha. carbon of R1, for use in the treatment or prevention of a disease or a biomedical condition selected from a seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea, schizophrenia, ischemia, cancer and fatal blood loss, provided that, when the compound is 2-methyl-2-pentenoic acid, the disease or condition is not bipolar disorder or epilepsy.

The compounds described herein have been found to cause rapid attenuation of phosphoinositol turnover and/or attenuation of fatty acid turnover. Since attenuation of phosphoinositol and fatty acid turnover have been identified as mechanisms of action of VPA, these compounds may have the potential to be useful in the treatment or prevention of VPA-treatable conditions, such as seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea and schizophrenia, in particular, epilepsy, bipolar disorders and migraine.

In an embodiment, when R1 is an alkyl group having a C.sub.7 backbone, the branching does not consist only of a methyl group at the .omega.-1 carbon of R1, and preferably does not comprise a methyl group at the .omega.-1 carbon of R1.

Compounds that can be used for the purpose of the invention include, but are not limited to, nonanoic acid, decanoic acid, 4-ethyloctanoic acid, 2-propyloctanoic acid, 2-butyloctanoic acid, 4-methylnonanoic acid, 8-methylnonanoic acid, 3-methylnonanoic acid, and 3-Methylundecanoic acid.

The term `an alkyl group having a C.sub.x-y backbone` as used herein refers to a linear saturated hydrocarbon group containing from x to y carbon atoms. For example, an alkyl group having a C.sub.1-4 backbone refers to an unbranched saturated hydrocarbon group containing from 1 to 4 carbon atoms. Examples of an alkyl group having a C.sub.1-4 backbone include methyl, ethyl, propyl, and butyl.

The term `an alkenyl group having a C.sub.x-y backbone` as used herein refers to a linear unsaturated hydrocarbon group containing from x to y carbon atoms and at least one (e.g. 1, 2, 3 or 4) double bonds. For example, an alkenyl group having a C.sub.3-5 backbone refers to an unbranched unsaturated hydrocarbon group containing from 3 to 5 carbon atoms. Examples of an alkenyl group having a C.sub.3-5 backbone include propylene, butylene and pentylene.

The terms ".alpha. carbon of R1", ".beta. carbon of R1" and ".gamma. carbon of R1" refer to the first, second and third carbon atoms, respectively, in a chain of carbon atoms forming R1, counting from, but not including, the COOH group of Formula (I). The term ".omega.-1 carbon of R1" refers to the penultimate carbon atom of a chain of carbon atoms forming R1, again counting from the COOH group of Formula (I). In other words, ".omega.-1 carbon of R1" is the carbon next to the terminal methyl or methylene group in R1.

The term `C.sub.x-y alkyl` as used herein refers to a branched or unbranched saturated hydrocarbon group containing from x to y carbon atoms. For example, C.sub.1-4 alkyl refers to a branched or unbranched saturated hydrocarbon group containing from 1 to 4 carbon atoms. Examples of C.sub.1-4 alkyl groups include methyl, ethyl, n-propyl, isopropyl, n-butyl, isobutyl, sec-butyl and tert butyl.

`Pharmaceutically acceptable salts` of compounds of the present invention include salts with inorganic bases, salts with organic bases, salts with inorganic acids, salts with organic acids and salts with basic or acidic amino acids. Salts with bases may, in particular, be employed in some instances. The compound of the present invention may be in either hydrate or non-hydrate form.

`Pharmaceutically acceptable amides` of compounds of the present invention are derivatives in which the carboxyl (i.e. --C(O)OH) groups of the said compounds are modified by reaction with an amine --NHR1'R2' so as to yield --C(O)NR1'R2' groups, wherein R1' and R2' are optionally independently selected from H, C.sub.1-8 alkyl (eg C.sub.1-6 alkyl), aryl, heteroaryl and C.sub.3-8 cycloalkyl group.

`Pharmaceutically acceptable esters` of compounds of the present invention are derivatives in which the carboxyl (i.e. --C(O)OH) groups of the said compounds are modified by reaction with an alcoholic moiety W--OH so as to yield --C(O)OW groups, wherein W may be C.sub.1-18 alkyl (e.g. C.sub.1-6 alkyl), aryl, heteroaryl, or C.sub.3-8 cycloalkyl.

General methods for the preparation of salts, amides and esters are well known to the person skilled in the art. Pharmaceutical acceptability of salts, amides and esters will depend on a variety of factors, including formulation processing characteristics and in vivo behaviour, and the skilled person would readily be able to assess such factors having regard to the present disclosure.

Where compounds of the invention exist in different enantiomeric and/or diastereoisomeric forms (including geometric isomerism about a double bond), these compounds may be prepared as isomeric mixtures or racemates, although the invention relates to all such enantiomers or isomers, whether present in an optically pure form or as mixtures with other isomers. Individual enantiomers or isomers may be obtained by methods known in the art, such as optical resolution of products or intermediates (for example chiral chromatographic separation (e.g. chiral HPLC)), or an enantiomeric synthesis approach. Similarly, where compounds of the invention may exist as alternative tautomeric forms, the invention relates to the individual tautomers in isolation, and to mixtures of the tautomers in all proportions.

In certain embodiments of the invention, R1 is an alkyl group.

In certain embodiments of the invention, R1 is an alkyl group having a C.sub.7-10 backbone.

In certain embodiments of the invention, R1 is an alkyl group having at least one point of branching, for example one, two or three points of branching.

In some embodiments, R1 is a C.sub.7-10 backbone alkylene group comprising branching at any position in the backbone, preferably at the .alpha., .beta., .gamma. or .omega.-1 carbon of R1.

In certain embodiments of the invention, the branching consists of a C.sub.1-4 alkyl group, such as a methyl, ethyl, propyl or butyl group, preferably a methyl, ethyl or propyl group.

In particular embodiments, R1 is a C.sub.7 backbone alkyl group having a branched ethyl, propyl or butyl group.

In particular embodiments, R1 is a C.sub.8 backbone alkyl group having a branched methyl group.

In certain embodiments of the invention, R1 is an unbranched alkyl group.

In particular embodiments, R1 is a C.sub.8-9 unbranched alkyl group.

In some embodiments, the one or more heteroatoms in the alkyl or alkenyl groups is selected from the group consisting of oxygen, sulphur and nitrogen. Preferably, the one or more heteroatoms is oxygen.

In certain embodiments, the compound used for the present invention is given separately, simultaneously or sequentially in combination with another pharmaceutically active agent which is known to be useful for the treatment or prevention of a disease or a biomedical condition selected from seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea, schizophrenia, ischemia, cancer and fatal blood loss, or co-morbidities thereof.

In certain embodiments, two or more of the compounds used in accordance with the first aspect of the invention can be used separately, simultaneously or sequentially in combination.

In a second aspect, the invention also provides a method of treatment or prevention of a disease or a biomedical condition selected from seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea, schizophrenia, ischemia, cancer and fatal blood loss, in particular, epilepsy, bipolar disorders and migraine, the method comprising the administration, to a subject in need of such treatment or prevention, of a therapeutically effective amount of a compound used according to the first aspect of the invention.

The compound may be administered with one or more conventional non-toxic pharmaceutically acceptable carrier, adjuvant or vehicle. Pharmaceutically acceptable carriers, adjuvants and vehicles that may be used in accordance with this invention are those conventionally employed in the field of pharmaceutical formulation, and include, but are not limited to, sugars, sugar alcohols, starches, ion exchangers, alumina, aluminium stearate, lecithin, serum proteins, such as human serum albumin, buffer substances such as phosphates, glycerine, sorbic acid, potassium sorbate, partial glyceride mixtures of saturated vegetable fatty acids, water, salts or electrolytes, such as protamine sulphate, disodium hydrogen phosphate, potassium hydrogen phosphate, sodium chloride, zinc salts, colloidal silica, magnesium trisilicate, polyvinyl pyrrolidone, cellulose-based substances, polyethylene glycol, sodium carboxymethylcellulose, polyacrylates, waxes, polyethylene-polyoxypropylene-block polymers, polyethylene glycol and wool fat. The compound can be administered orally, parenterally, by inhalation spray, rectally, nasally, buccally, vaginally or via an implanted reservoir. The term parenteral as used herein includes subcutaneous, intracutaneous, intravenous, intramuscular, intra-articular, intrasynovial, intrasternal, intrathecal, intralesional and intracranial injection or infusion techniques.

The compound used in the present invention may be in administered in the form of a sterile injectable preparation, for example, as a sterile injectable aqueous or oleaginous suspension. This suspension may be formulated according to techniques known in the art using suitable dispersing or wetting agents (such as, for example, Tween 80) and suspending agents. The sterile injectable preparation may also be a sterile injectable solution or suspension in a non-toxic parenterally-acceptable diluent or solvent, for example, as a solution in 1,3-butanediol. Among the acceptable vehicles and solvents that may be employed are mannitol, water, Ringer's solution and isotonic sodium chloride solution. In addition, sterile, fixed oils are conventionally employed as a solvent or suspending medium. For this purpose, any bland fixed oil may be employed including synthetic mono- or diglycerides. Fatty acids, such as oleic acid and its glyceride derivatives are useful in the preparation of injectables, as are natural pharmaceutically-acceptable oils, such as olive oil or castor oil, especially in their polyoxyethylated versions. These oil solutions or suspensions may also contain a long-chain alcohol diluent or dispersant such as that described in Ph. Helv, or a similar alcohol.

The compound used for this invention may be orally administered in any orally acceptable dosage form including, but not limited to, capsules, tablets, powders, granules, and aqueous suspensions and solutions. These dosage forms are prepared according to techniques well-known in the art of pharmaceutical formulation. In the case of tablets for oral use, carriers which are commonly used include lactose and corn starch. Lubricating agents, such as magnesium stearate, are also typically added. For oral administration in a capsule form, useful diluents include lactose and dried corn starch. When aqueous suspensions are administered orally, the active ingredient is combined with emulsifying and suspending agents. If desired, certain sweetening and/or flavouring and/or colouring agents may be added.

The compound used for this invention may also be administered in the form of suppositories for rectal administration. For this purpose, the compound may be mixed with a suitable non-irritating excipient which is solid at room temperature but liquid at the rectal temperature and therefore will melt in the rectum to release the active components. Such materials include, but are not limited to, cocoa butter, beeswax and polyethylene glycols.

The compound used for this invention may be administered by nasal aerosol or inhalation. For this purpose, the compound is prepared according to techniques well-known in the art of pharmaceutical formulation and may be prepared as solutions in saline, employing benzyl alcohol or other suitable preservatives, absorption promoters to enhance bioavailability, fluorocarbons, and/or other solubilising or dispersing agents known in the art.

The compounds used for the present invention may be administered in a dose of around 1 to around 20,000 .mu.g/kg per dose, depending on the condition to be treated or prevented, and the characteristics of the subject being administered with the compound. In many instances, the dose may be around 1 to around 1500 .mu.g/kg per dose. The dosing regimen for a given compound could readily be determined by the skilled person having access to this disclosure.

In accordance with a third aspect, the present invention provides the use of a compound as defined according to the first aspect of the invention in the preparation of a medicament for the treatment or prevention of a disease or a biomedical condition selected from seizure-related disorders, bipolar disorders, mania, depression, migraine, attention deficit hyperactivity disorders, latent HIV infection, Alzheimer's disease, chorea, schizophrenia, ischemia, cancer and fatal blood loss.

The invention will now be described in more detail by way of example only with reference to the figures listed below.

FIG. 1 shows time- and dose-dependent effect of VPA in attenuation of phosphoinositide signalling in Dictyostelium. Phosphoinositide labeling was monitored by incorporation of a radio-labelled phosphate into newly-formed lipids, followed by extraction, TLC separation and quantification using a Typhoon phosphorimager (Pawolleck & Williams 2009). (a) Analysis of PIP and PIP.sub.2 turnover in cells treated with VPA (0.5 mM) for indicated times. (b) Analysis of PIP and PIP.sub.2 turnover in cells treated for 9 min with varying concentrations of VPA. Results are provided for triplicate experiments with duplicate samples.+-.SD where *P<0.05; **P<0.01; ***P<0.001 for PIP levels.

FIG. 2 shows phosphoinositide signalling and VPA sensitivity in wild type (wt) and knockout mutants lacking phospholipid kinase and inositol recycling enzymes with or without 9 min VPA treatment (0.5 mM). (a) Comparison of phosphoinositide levels in untreated isogenic mutant lines lacking indicated lipid kinase activities: five type 1 phosphatidylinositol 3-kinase (5.times.PI3K; DBS0252654) genes; the phosphatidylinositol-4-phosphate 5-kinase (rPKA; DDB0191443) gene; and the phosphatidylinositol-4-phosphate 5-kinase gene (PIPKinA; DDB0185056) gene. VPA and PI3K inhibitor sensitivity was monitored using 0.5 mM VPA or 50 .mu.M LY2946004 respectively for: (b) the 5.times.PI3K mutant; (c) the rPKA mutant and (d) the PIPKinA mutant. (e) Schematic of phosphoinositide signalling showing the role of phospholipase C (PLC), prolyl oligopeptidase (PO), inositol monophosphatase (IMPase) and myo-inositol synthase (INO1) in the generation and recycling of phosphoinositides. (f) Ablation of PLC and PO genes did not alter VPA-attenuated PIP and PIP.sub.2 signalling (g) Extended VPA treatment (60 min, 0.5 mM) further reduced phosphoinositide signalling, and this effect was not reversed following overexpression of IMPase. Results are provided for triplicate experiments with duplicate samples.+-.SD where *P<0.05; **P<0.01; ***P<0.001 for PIP levels.

FIG. 3 shows seizure control with VPA and phosphoinositide-attenuating compounds using in vitro acute seizure mode-pentylentetrazol (PTZ) model and low magnesium model. (a) A combined entorhinal cortex-hippocampal slice preparation was placed in a submerged recording chamber and perfused with artificial cerebrospinal fluid containing high [K.sup.+] and PTZ to induce epileptiform activity prior to addition of VPA or novel compounds at 1 mM. (b) Illustration of trace samples of burst discharges following application of VPA, 4-methyloctanoic acid (4-MO), 2-propyloctanoic (2-PO), and 2-butyloctanoic acid (2-BO)). (c) Summary of the frequency of burst discharges following application of drugs plotted against time. The drugs were applied from time 0 to 40 minutes. (d) Trace samples and (e) frequency of burst discharges for longer straight chain nonanoic acid (NA) and decanoic acid (DA). (f) Illustration of trace samples of low magnesium-induced burst discharges by application of VPA (1 mM), 4-methyloctanoic acid (4-MO, 1 mM). (g) Summary of the frequency of low magnesium-induced burst discharges following application of drugs (VPA 1 mM, n=5; 4-MO 1 mM, n=5). The frequency of epileptiform activity induced by low magnesium plotted against time. The drugs were applied from time 0 to 40 minutes. Application of VPA resulted in a significant decrease in discharge frequency (72.7.+-.3.6% of baseline, 30-40 minutes after application and the effect of suppression is reversible after wash out), whereas application of 4-MO abolished the epileptiform discharge (1.6.+-.3.1% of baseline, p<0.01 compared to control; P<0.01 compared to VPA). The epileptiform activities in both treatment recovered during drug washout (n=5 for each drug). (h) Comparison of the mean frequency of low Mg2+-induced burst discharges for the last 10 min during drug application with different treatments, demonstrating a significant effect of all compounds in attenuating seizure activity. * P<0.05, ** P<0.01 compared to control; +P<0.05, ++P<0.01, compared to VPA treated group. Data are presented as means.+-.SEM.

FIG. 4 shows seizure control with phosphoinositide-attenuating compounds in an in vivo seizure model. (a) Summary of the procedure--electrical induction of self-sustained status epilepticus (SSSE). Rats were electrically stimulated via the perforant pathway with 4-5 mA, 50 .mu.s monopolar pulses at 20 Hz for 2 hours to induce SSSE seven days after electrode implantation. Three hours after the induction of the SSSE, the rats were given diazepam (10 mg/kg) by intraperitoneal injection (i.p.) to terminate the seizure activity. (b) Illustrative trace samples of EEG from status epilepticus animal. Administration of VPA (400 mg/kg) or 4-methyloctanoic acid (400 mg/kg) resulted in attenuation of seizure activity, whereas DMSO had no effect. (c) Time course of the effects on spike amplitude following administration of DMSO (n=5), VPA (n=7) and 4-MO (n=7). (d) Time course of the effects on spontaneous spike frequency following administration of DMSO (n=5), VPA (n=7) or 4-MO (n=7).

FIG. 5 demonstrates that VPA induces changes in fatty acid uptake and release in a time and concentration dependent manner. Dictyostelium wild type (Ax2) cells were pre-incubated with .sup.3H arachidonic acid (A) or palmitic acid (C) and the release of .sup.3H into external buffer is shown in the presence/absence of VPA. Fatty acid uptake was measured by incubation of cells with or without VPA and .sup.3H AA (B) or palmitic acid (D) simultaneously. Uptake of .sup.3H into Dictyostelium cell pellet is shown. All results are expressed as control at 60 minutes. Insets show dose response curves. Statistics and dose response curves were calculated using Graphpad Prizm.TM. software. All data are replicates of at least 3 independent experiments and show mean.+-.SEM.

FIG. 6 demonstrates that VPA induces endocytosis independent lipid droplet accumulation of bodipy in Dictyostelium. (a) Images of bodipy fatty acid accumulation in Dictyostelium in the absence (i) or presence (ii) of 0.5 mM VPA. VPA significantly increased the droplet intensity and average diameter of lipid droplets compared to control (b) (t test, *** p<0.001, * p<0.05). The actin polymerising inhibitor latrunculin (10 .mu.M) did not completely inhibit VPA induced increase in .sup.3H arachidonic acid uptake. All data are replicates of at least 3 independent experiments and show mean.+-.SEM.

FIG. 7 demonstrates that knockout PlaA cells are protected from VPA-induced inhibition of development. (A) Alignment of PlaA protein sequence from Dictyostelium discoideum (XP_642421.1) and an iPLA.sub.2 protein sequence from Homo sapiens (AAD08847). Sequences show conserved homology at ATP binding and lipase sites. Alignment was carried out using BLAST software. (B) Radiolabel release from wild type and PlaA cells show a similar reduction of radiolabel release in the presence of VPA. (C) Development images of Dictyostelium wild type (Ax2) cells or PlaA-ve cells at 30 hours in the absence or presence of 1 mM VPA (as indicated). Scale bars represent 500 .mu.m.

FIG. 8 demonstrates that PLA.sub.2 inhibitors phenocopy VPA-induced .sup.3H release from fatty acid labelled cells. (A) Aggregation competent Dictyostelium wild type (Ax2) cells were pre-incubated with .sup.3H AA and the release of .sup.3H into external buffer is shown in the presence/absence of PLA.sub.2 inhibitors. Inhibitor x=80 .mu.M BEL, a Ca.sup.2+ PLA.sub.2 inhibitor, y=20 .mu.M BPB, a general PLA.sub.2 inhibitor and z=50 .mu.M MAFP, a Ca.sup.2+ dependent and Ca.sup.2+ independent cytosolic PLA.sub.2 inhibitor. Mix=combination of all inhibitors. (B) PLA.sub.2 inhibitors do not mimic VPA-dependent fatty acid uptake. Cells were incubated in the presence of VPA and PLA.sub.2 inhibitor mix (see methods). Uptake of .sup.3H into Dictyostelium cell pellet is shown. (C) VPA induced .sup.3H arachidonic acid uptake is independent of CoA activation. Quantification of CoA activated palmitic acid in wild type, wild type+0.5 mM VPA or fcsA -/- Dictyostelium (one way ANOVA, Dunnet's post hoc *** p<0.001). (D) Uptake of .sup.3H arachidonic acid in wild type fcsA -/- or fcsB --/- cells in the absence or presence of 0.5 mM VPA (one way ANOVA, Bonferroni post hoc, *** p<0.001).

FIG. 9 shows the effect of VPA analogues on .sup.3H arachidonic acid uptake and release. VPA induced parallel enhanced uptake and reduced release of radiolabel, effects which were enhanced by 2-isopropyl pentanoic acid (PIA) and reduced by 4-methyloctanoic acid. PLA.sub.2 inhibitor cocktail inhibited both the uptake and the release of radiolabel (one way ANOVA, Bonferroni post hoc, ns not significant, *** p<0.001).

1. Materials & Methods

1.1 Chemicals and Mutants

All chemicals were provided by Sigma UK Ltd. Valproic acid congeners were provided by Sigma Aldrich UK, Alfa Aesar/Avocado, ChemSampCo, ChemCo, The NCI/DTP Open Chemical Repository (http://dtp.cancer.gov) or TCI europe. Radiolabelled ATP was provided by Perkin Elmer Ltd. Dictyostelium mutants were provided by Dictybase.org, apart from the quintuple PI3K (and PTEN) knockout strain and the rPKA knockout kindly provided by R.Kay (Cambridge, UK) and A.Noegel (Koeln, Germany). Dictyostelium media (Axenic) was supplied by Formedium (Norfolk, UK). .sup.3H arachidonic acid purchased from Hartmann analytic (Germany) and .sup.3H palmitic acid from Perkin Elmer (Cambridge, UK).

1.2 Cells and Development

Dictyostelium cells were grown in Axenic medium or on Sussmans media plates in association with Raoultella planticola (Drancourt et al., 2001). All cell labelling used cells shaking (120 rpm) at 22.degree. C., and cells were artificially developed by pulsing with cAMP (25 nM final concentration) every 6 min for 4 hours at 2.5.times.10.sup.6 cells/ml in phosphate buffer (16.5 mM KH.sub.2PO.sub.4, 3.8 mM K.sub.2HPO.sub.4 pH 6.2) as described previously (Boeckeler et al., 2006).

1.3 Dictyostelium Phospholipid Labelling and Inositol Analysis

A saponin-based cell permeabilization protocol for Dictyostelium was adapted for these experiments (Pawolleck & Williams, 2009). Dictyostelium AX2 cells were developed for 5 hours as previously described (Boeckeler et al., 2006) (pulsed with cAMP to achieve final concentration of 25 nM), transferred to still dishes (2.5 cm), allowed to settle to give a confluent monolayer in KK2 (20 mM potassium phosphate buffer, pH 6.1) and pre-treated with compound (0.5 mM VPA or related compound or 50 .mu.M LY294002) for 3 min. At regular time intervals, buffer was replaced with labelling solution (139 mM sodium glutamate, 5 mM glucose, 5 mM EDTA, 20 mM PIPES pH 6.6, 1 mM MgSO4.2H2O, 0.25% (w/v) saponin, 1.times. phosphatase inhibitor cocktails 1 and 2 (Roche Ltd), and 1 .mu.Ci/ml .gamma.[32P]ATP) supplemented with compounds at defined concentrations.

Following a 6 min incubation, labelling solution was removed and cells were lysed in acidified methanol and phospholipids were separated as previously detailed (Williams et al., 1999). Phospholipid labelling was quantified using a Typhoon phosphor-imager. Even loading was determined using total lipid stain with copper sulphate. Inositol levels were measured from five hour developed cells (similar to phospholipid labelling), following lyophilisation, as previously described (Maslanski & Busa, 1990).

1.4 In Vitro Epilepsy Model

The rats (p21) were decapitated after killing by intraperitoneal injection with an overdose of pentobarbitone (500 mg/kg). The brain was removed and placed in ice-cold sucrose solution in mM: NaCl 87, KCl 2.5, MgCl.sub.2 7, CaCl2 0.5, NaH.sub.2PO.sub.4 1.25, sucrose 75, glucose 25, equilibrated with 95% O.sub.2/5% CO2 (pH 7.4). Horizontal combined entorhinal cortex-hippocampus slices (350 .mu.m) were prepared with a Leica vibratome (Leica VT1200S) and were then stored in an interface chamber that contained artificial cerebrospinal fluid solution (aCSF) containing in mM: NaCl 119, KCl 2.5, MgSO.sub.4 4, CaCl2 4, NaHCO.sub.3 26.2, NaH.sub.2PO.sub.4 1, glucose 11, and gassed with 95% O.sub.2/5% CO.sub.2. They were stored for over one hour before being transferred to a submersion recording chamber continually perfused with carbogenated aCSF for recording. Field potential recordings were made by placing glass microelectrodes (.about.1-2 M.OMEGA.) filled with aCSF solution in stratum radiatum of CA1. Bipolar stimulating electrodes were positioned in the Schaffer collateral/commissural fibre pathway in stratum radiatum to confirm slice viability. In the PTZ acute seizure model, PTZ (2 mM) was added to the perfusate and [K+] was increased to 6 mM in order to induce epileptiform activity (Armand et al., 1998). In the low Mg.sup.2+ acute seizure model, Mg.sup.2+ free aCSF was applied to generate rhythmic short recurrent discharges. Novel anticonvulsants were applied once the frequency and amplitude of the epileptiform discharges were stable over a period of 10 min. Anticonvulsant effects were evaluated by measuring the variation of frequency of the discharges every minute. The data acquired from the 30 to 40 minutes after application novel anticonvulsants were compared by ANOVA followed by post-hoc testing using Tukey test, using SPSS statistical analysis.

1.5 In Vivo Status Epilepticus

This method has been described in detail previously (Walker et al., 1999). In brief, male Sprague Dawley rats (300-400 mg) were anesthetized with 1-2% isoflurane in O.sub.2. An earth electrode was positioned subcutaneously, and a monopolar recording electrode was implanted stereotactically into the right hippocampus (coordinates, 2.5 mm lateral and 4 mm caudal from bregma). A bipolar stimulating electrode was implanted in the right hemisphere and advanced into the angular bundle (coordinates, 4.4 mm lateral and 8.1 mm caudal from bregma) to stimulate the perforant path. The depths of the electrodes were adjusted to maximize the slope of the dentate granule cell field potential (Guo et al., 1999). The electrodes were held in place with dental acrylic and skull screws. The animals were allowed to recover from anaesthesia. Seven days later, the perforant path was electrically stimulated with 4-5 mA 50 .mu.sec monopolar pulses at 20 Hz for 2 hr; this induced self-sustaining status epilepticus. After 10 min of self-sustaining status epilepticus, compounds or vehicle were administered and the behavioural seizures and EEG were monitored for 3 hours. At this point diazepam (10 mg/kg) was administered to all animals to stop the status epilepticus. Groups were compared by ANOVA followed by post-hoc testing using Tukey test, using SPSS.

1.6 Fatty Acid Uptake and Release

Dictyostelium cells were labelled with tritiated fatty acid in shaking liquid culture at 1.5.times.10.sup.6 cells/ml with 0.5 .mu.Ci of .sup.3H labelled fatty acid added in 0.5% BSA (fatty acid free BSA) per 2 treatments. Samples were taken at indicated times by removing 4.5.times.10.sup.6 cells, washing once in phosphate buffer and re-suspending in phosphate buffer prior to scintillation counting. For fatty acid release experiments, cells were pulsed (as above) for 4 hours, and cells were resuspended in phosphate buffer with fatty acid free BSA (0.5%) at 1.5.times.10.sup.6 cells/ml and time points were taken over one hour. Cells (4.5.times.10.sup.6 per time point) were washed to remove unincorporated radioactivity and at indicated times and the supernatant was analysed via scintillation counting. Modelling results employed using Graphpad Prism software. Bodipy labelling employed 4 hour pulsed cells, incubated with fluorescent fatty acid (Invitrogen) for 30 min in the presence or absence of VPA (0.5 mM) and images were recorded on an Olympus IX 71 inverted fluorescence microscope with Retiga FastA 1394 camera and analysed by ImagePro.TM. software.

1.7 Mutant Isolation and Recapitulation and Development

Screening of a REMI library was carried out as previously described (Kuspa & Loomis, 2006) using Ax2 background, with VPA resistant mutants selected for the ability to develop in the presence of 1 mM VPA on R. planticola. Identification of the ablated gene, enabled the identified PLAa (DDB_G0278525) to be recapitulated using by homologous recombination of a knockout cassette. Primers used for amplifying region within the open reading frame of the gene were (5' ATGGGAGATAATAAAAAAGAAAATATCAG and 3' TAAGAATTCATGGGAGATAA TA AAAAAGAAAATATCAG, cloned by pCR2.1 TOPO (Invitrogen Ltd)), cloned into pUC19 using EcoR1 digestion, and SmaI digested fragment from pBLPblp (Faix et al, 2004) was inserted into the EcoRV site of the insert. Genetic ablation was confirmed by PCR analysis. Developmental resistance to VPA was assessed by plating cells (1.times.10.sup.6) on 47 mm nitrocellulose filters (Millipore) soaked in phosphate buffer containing either 1 mM VPA or control, and development was recorded after 30 h unless otherwise stated. Development images were observed using a Leica CLS 150.times. microscope and images recorded using QICAM FAST 1394 camera. Fatty acid activation was determined by the method established by Wilson et al. (Wilson et al, 1982) with slight modifications. Briefly, extracts were prepared by sedimentation of 1.times.10.sup.7 axenically grown cells, washing them once in 10 ml precooled 1 M Tris-HCl (pH 7.5) and lysing them for 30 min in 100 .mu.l 1 M Tris-HCl (pH 7.5) containing 1% Triton and Protease Inhibitor Cocktail (P8340, Sigma-Aldrich, Germany) at 4.degree. C. Twenty mg of protein extract in a volume of 140 ml were diluted into 400 .mu.l of a buffer containing 250 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 3 mM ATP, 0.6 mM EDTA, 0.25% Triton, and 2.5 mM DTT. 40 .mu.l of unlabelled palmitic acid (P9767, Sigma-Aldrich, Germany) from a 100 .mu.M methanol stock and 5 .mu.l 3H-palmitic acid (20 .mu.M, 1 mCi/ml) served as substrates. The reaction was started by addition of 20 .mu.l 10 mM coenzymeA-solution and incubated at 35.degree. C. To stop the reaction 500 .mu.l Dole's medium (0.4 ml isopropanol, 0.1 ml n-heptane, 10 .mu.l H.sub.2SO.sub.4) was added after 10 min. Separation of the phases was achieved by centrifugation for 30 sec at 14,000 rpm in a tabletop centrifuge. The organic phase was discarded and the aqueous phase was washed six times with 300 .mu.l of n-heptane to remove non-activated fatty acids before the radioactivity of the acyl-CoA thioester remaining in the aqueous phase was determined in 2 ml of Lumasafe.TM. Plus fluid (Lumac LSC, Groningen, The Netherlands) in a scintillation counter.

2. Results

2.1 VPA Attenuates Phosphoinositide Signalling

Since the acute effect of VPA on phosphoinositide signalling has not been characterised, the inventors firstly examined the time- and concentration-dependence of drug action in Dictyostelium (FIG. 1). A rapid reduction in phosphoinositide phosphorylation was seen following 0.5 mM VPA treatment, with a 49% reduction in both radio-labelled PIP and PIP.sub.2 turnover during 6 min VPA treatment, increasing to 94% reduction in the turnover of each phosphoinositide compound following 60 min treatment (FIG. 1a). These values indicate a combination of phosphoinositide synthesis and degradation within the cell, thus reflect phosphoinositide turnover, although phosphatase activity is blocked using an inhibitor cocktail. The acute nature of this effect occurs with a similar speed to that of seizure control following intravenous VPA injection in a mouse seizure model (Honack & Loscher, 1992). The attenuation of phosphoinositide signalling was also concentration dependent, with a 25% and 42% inhibition of PIP and PIP.sub.2 turnover respectively at 0.25 mM VPA following 9 min treatment increasing to a 68% and 79% reduction at 0.5 mM VPA respectively (FIG. 1b)--these concentrations are found in the therapeutic use of VPA (0.4-0.7 mM in plasma) (DSM IV, 2000). Under these conditions, this inhibitory effect of VPA provides an EC50 of 154 .mu.M, and is independent of uptake (since cells are permeabilized with saponin). The acute, strong inhibition of phosphoinositide signalling caused by VPA made this effect of potential therapeutic interest.

A rapid reduction in phosphoinositide phosphorylation was initially thought to occur through inhibition of an unidentified lipid kinase activity. Analysis of lipid kinases traditionally employs pharmacological inhibition with enzyme class-specific compounds, but these studies are complex due to the large number of phosphatidylinositol kinase enzymes and overlapping effects between inhibitors. Since previous studies have suggested a role for VPA in attenuating the phosphatidylinositol 3-kinase (PI3K) signalling pathway (Xu et al., 2007), the inventors analysed the effect of ablating five different type 1 PI3K genes in a single cell line (Hoeller & Kay, 2007) on phosphoinositide signalling (FIG. 2a). These cells showed a 28% and 44% reduction in the formation of PIP and PIP.sub.2 production respectively compared to wild type cells, suggesting a major role of these enzymes in phosphoinositide signalling. To test for a PI3K-dependence of the VPA-catalysed phosphoinositide reduction, VPA (0.5 mM) was added to these cells. It was found that VPA reduced PIP and PIP.sub.2 production by 48% and 70%, respectively compared to untreated cells following 9 min treatment (FIG. 2b), indicating that these five ablated enzymes are not the target of VPA in attenuating phosphoinositide turnover.

To investigate a role of other lipid kinases in VPA-catalysed phosphoinositide attenuation, the inventors analysed two other non-related phosphatidylinositol kinases: the rPKA knockout mutant lacking an endosomal G-protein-coupled receptor protein containing a phosphatidylinositol 5 kinase (PIPSK) domain (Bakthavatsalam et al., 2006); and the PIPKinA mutant that lacks a nuclear phosphatidylinositol 4/5 kinase activity (Guo et al., 2001). Ablation of rPKA (FIG. 2a) showed a 30% and 54% reduction in PIP and PIP.sub.2 production, respectively compared to wild type cells, with VPA treatment causing an additional 72% and 33% reduction compared to untreated cells (FIG. 2c). Ablation of PIPKinA showed no significant change in PIP and PIP.sub.2 production (FIG. 2a), with VPA treatment causing a 54% and 68% reduction in PIP and PIP.sub.2 synthesis respectively compared to untreated cells (FIG. 2d). All three cell lines were still sensitive to pharmacological inhibition of PI3K activity (using 50 .mu.M LY294006--an inhibitor of PI3K activity), confirming these variations were related to attenuated phosphoinositide turnover (FIGS. 2b-d). The reduced sensitivity of all three lipid kinase mutants suggests a common mechanism of VPA action independent of specific phosphatidylinositol kinase action.

Since another mechanism for regulating phosphoinositide signalling is the recycling of phosphatidylinositol, via inositol phosphates (FIG. 2e), the inventors analysed phosphoinositide turnover in isogenic mutants with this recycling pathway blocked or activated. Cells lacking the single phospholipase C gene (Drayer et al., 1994) showed no significant reduction in PIP and PIP.sub.2 turnover compared to wild type cells (FIG. 2f), and showed a VPA-catalysed reduction in PIP and PIP.sub.2 signalling close to that for wild-type cells (73 and 75% for PIP and PIP.sub.2 respectively). Cells with approximately three-fold higher inositol trisphosphate (InsP.sub.3) caused by prolyl oligopeptidase ablation (PO; Williams et al., 1999, Williams et al., 2002) showed a slight decrease in PIP levels in untreated cells (and no significant change in PIP.sub.2 levels) and a VPA-catalysed reduction in PIP and PIP.sub.2 signalling by 66% and 68% respectively. Furthermore, the inventors have previously shown that inhibition of inositol monophosphatase (IMPase) activity by 10 mM lithium does not attenuate phosphoinositide signalling following acute (9 min) treatment (King et al., 2009), however extended lithium treatment (60 min) reduces PIP and PIP.sub.2 levels, and this effect is overcome by over-expression of IMPase (King et al., 2009). In comparison, over-expressing IMPase did not overcome extended VPA treatment (60 min; 0.5 mM) (FIG. 2g). These results suggest that elevating or reducing recycling of inositol through inositol phosphate signalling does not play a major regulatory role in acute phosphoinositide production and does not overcome VPA-catalysed acute reduction in phosphoinositide signalling in this model. These results thus provide the first strong evidence for a mechanism of action of VPA--independent of inositol depletion--in targeting phosphoinositide signalling.

2.2 Identifying Novel Compounds Showing Increased Phosphoinositide Attenuation

The identification of an acute effect of VPA in attenuating phosphoinositide signalling enabled the investigation of the structural requirements for this effect. The effect of compounds tested for the present invention on phosphoinositide attenuation are summarised in Table 1 below.

TABLE-US-00001 TABLE 1 PIP Chemical Chemical Chemical Level (% category (common name) (IUPAC nomenclature) control) SD valproic acid (VPA) 2-propylpentanoic acid 32.0 8.7 Shorter than 5 carbons backbone (the longest aliphatic side chain) acids Isovaleric Acid 3-methylbutanoic acid 37 9 3-methylbutanoic 99.1 11.4 GABA 4-aminobutanoic acid 60.8 3.8 TBA tert-butylacetic acid 21.4 4.1 PIA propylisopropylacetic acid 15.4 2.2 DIA diisopropylacetic acid 18.6 4.5 5 carbon backbone acids 4-methylpentanoic acid 60 11.8 2-methyl-2-pentenoic acid 14.8 3.8 4-methyl-2-pentenoic acid 54.8 14.1 2,4-dimethyl-2-pentenoic acid 38.4 6.8 trans-pent-2-enoic acid 50.4 9.1 2-methylpentanoic acid 59 7.3 3-methylpentanoic acid 64 13.6 4-methyl-2-pentenoic acid 121 25 2,4-dimethyl-2-pentenoic acid 94 14 3-methylvaleric acid 3-methylpentanoic acid 66 12 4-methylvaleric acid 4-methylpentanioc acid 102 18 3-methylpentanoic acid 67 16 2,2-dimethyl-4-pentenoic acid 60 7 3-methyl-4-pentenoic acid 84 6 6 Carbon backbone acids 4-methylhexanoic acid 68 3 2-methylhexanoic acid 68.1 14.8 5-methylhexanoic acid 7.2 0.7 2-ethylhexanoic acid 22.2 5.3 2,2-dimethylhexanoic 29.7 9.7 3,5,5-trimethylhexanoic acid 32 13 4-hexenoic acid, (cis + trans) 58 15 7-9 carbon backbone acids 2-methylheptanoic 16.1 6.4 4-methyloctanoic acid 12 1.3 4-ethyloctanoic acid 13.2 1.8 4-methylnonanoic acid 45 16 11 carbon backbone acids 3-methylundecanoic acid 50 0 Straight-chain acids valeric acid pentanoic acid 66.9 7.8 n-caproic acid hexanoic acid 49.2 10.1 enanthoic acid heptanoic acid 31.2 5 caprylic acid octanoic acid 16.7 3.4 pelagonic acid nonanoic acid 8 1.7 capric acid decanoic acid 12.9 1.4 Lauric acid dodecanoic acid 42 3 Margaric acid heptadecanoic acid 92.3 2.8 Other acids Diphenylacetic acid 2,2-diphenylacetic acid 39 11 TMCA tetramethylcyclopropane 33.9 7.7 carboxylic acid Derivatized carboxylic acids (amides) valpromide (VPD) 2-propylpentamide 69.3 13.3 valnocatmide (VCD) 2-ethyl-3-methyl valeramide 64 2.9 TMCD tetramethylcyclopropane- 72.5 7.2 carboxamide MTMCD N-methyl-tetramethyl- 50.8 6.9 -cyclopropane carboxamide PID propylisopropylacetamide 59 12.2 Tert-butyl amide 47.6 12 n-propyl 2-methylvalerate 121 13.6 Aldehydes methylvalerate 46 5.5 valeraldehyde pentanal 52 10.8 octanal 260 199 nonanal 99 13 Alcohols 2-propyl-1-pentanol 101 13 2-butyl-1-octanol 93 53 2-hexyl-1-decanol 191 38

Although the majority of compounds analysed showed some inhibitory effect on phosphoinositide turnover (Table 1), a number of structures showed greater phosphoinositide signalling inhibition than VPA. These highly active compounds fit into two structural groups: the first comprising branched fatty acids with a roughly similar structure to VPA; and a second novel group of compounds with or without short side chains in various positions on the backbone. Within this latter group, fatty acids show a strong dependence on length, whereby 8-10 carbon backbone acids are highly active (e.g. 4-methyloctanoic acid reduces PIP and PIP.sub.2 signalling by 88% and 93% respectively, and nonanoic acid by 92% and 93% respectively; Table 1) and increased or decreased backbone length reduces activity. All highly active compounds in this group are fatty acids, without predicted teratogenicity (Eickholt et al., 2005) and show a positive association with lipophilicity. This effect is also independent of acidic function, since variable activity is shown with straight carbon acids of equivalent acidity (pKa, Table 2 which shows a comparison of phosphoinositide attenuation and pKA values for VPA and straight chain acids in Dictyostelium). These structural distinctions provide the first characterization of VPA congeners for this effect of phosphoinositide attenuation. Interestingly, high structural specificity has previously been show for fatty acids in both anticonvulsant as well as antiallodynic (anti-neuropathic pain) activities (Kaufmann et al., 2009). Preliminary observation of behaviour in animal models for one related compound does not suggest a strong sedative effect.

TABLE-US-00002 TABLE 2 Total number of PIP level Acid Carbon atoms pKa % control SD VPA 8 4.6 32.0 8.7 Pentanoic acid 5 4.84 66.9 7.8 Hexanoic acid 6 4.85 49.2 10.1 Heptanoic acid 7 4.89 31.2 5 Octanoic acid 8 4.89 16.7 3.4 Nonanoic acid 9 4.95 8 1.7 Decanoic acid 10 4.90 12.9 1.4 dodecanoic acid: 11 4.85 42 3 heptadecanoic acid 17 4.78 92.3 2.8

VPA has been identified as an inhibitor of de novo inositol biosynthesis, indirectly blocking the production of inositol-1-phosphate from glucose-6-phosphate (Shaltiel et al., 2004, Shaltiel et al., 2007a, Vaden et al., 2001). A role for VPA-attenuation of inositol signalling has been widely shown in models ranging from yeast (Vaden et al., 2001) and Dictyostelium (Williams et al., 1999, Williams 2002) to Caenorhabditis elegans (Tokuoka et al., 2008), rats and humans (Shaltiel et al., 2007a, Shaltiel et al., 2007b). Measurement of inositol and inositol trisphosphate (InsP3) levels and the inositol-dependent spreading of mammalian growth cones have all been used to show inositol depletion. Since it is not clear if a VPA-induced reduction in inositol levels may cause the a cute reduction in phosphoinositide signalling shown here, the inventors analysed phosphoinositide turnover using VPA-related compounds shown to be active in inositol depletion. Compounds showing strong InsP3 depletion in Dictyostelium with concomitant inositol-depletion dependent enlargement in mammalian growth cones include 2-methyl-2-pentenoic acid (Eickholt et al., 2005) and this compound showed stronger phosphoinositide attenuation than VPA. Interestingly, substituting the carboxylic acid moiety of compounds showing high phosphoinositide attenuation (VPA) with a carboxamide group (yielding the corresponding amide (VPD) reduces the inhibitory effect on phosphoinositide turnover and reduces growth cone spreading, and VPD shows weak inhibition of human myo-inositol synthase proposed as the VPA-target in inositol depletion (MIP synthase (Shaltiel et al., 2004, Shaltiel et al., 2007a)). These inositol-depleting and phosphoinositide-attenuating compounds are found mainly within the first structural group of compounds (described above), and also contain a number of potent anticonvulsants second generation to VPA currently under investigation (Bialer & Yagen, 2007). None of the novel family of longer backbone compounds identified in this study have been analysed in inositol depletion studies.

Since reduction in the inositol levels may provide the mechanism of these compounds in phosphoinositide attenuation, the inventors analysed inositol levels in treated Dictyostelium cells using a range of compounds from both structural groups showing variable phosphoinositide attenuation). In these experiments, VPA gave no significant reduction in inositol levels in the time period shown to cause phosphoinositide attenuation (9 min), nor did any other compound tested, and thus no correlation was found between phosphoinositide attenuation and inositol depletion. This conclusion is in agreement with previous data, based in Dictyostelium, showing the acute inhibition of inositol monophosphatase (by lithium) does not give rise to phosphoinositide attenuation (King et al., 2009), and depletion of inositol trisphosphate in this model by VPA requires 6 hour treatment (Williams et al., 1999)--considerably longer than the time periods used here. These experiments therefore suggest phosphoinositide attenuation provides a novel effect of VPA in Dictyostelium and identifies a range of compounds showing increased efficacy for this effect. Since increased PIP and PIP.sub.2 levels have been observed during seizures in animal models (Van Rooijen et al., 1986), and the inventors have discovered a novel family of compounds causing this effect.

2.3 Novel Compounds Show Enhanced Efficacy in In Vitro Epileptiform Models

Since it is not possible to repeat these radio-labelling experiments in in vivo animal systems to replicate the inventors' mechanism-dependent findings in higher models, the inventors instead analysed the efficacy of the novel family of compounds in seizure control. For these experiments, they employed a VPA-sensitive pentelenetetrazol (PTZ) in vitro model of epileptiform activity (Armand et al., 1998) to analyse three compounds from the novel family (FIGS. 3a,b) with an eight carbon backbone with variable side chain position and length. VPA significantly decreased the frequency of epileptiform discharges (Armand et al., 1998; FIG. 3b; VPA: 75.1.+-.1.7.quadrature.). the application of equimolar concentrations of each eight carbon backbone compound also strongly reduced discharges with a significantly greater efficacy than VPA (FIGS. 3b,c). Application of all three novel compounds greatly reduced seizure discharge frequency (4-methyloctanoic acid, 49.1.+-.4.4%, 2-propyloctanoic acid is 5.3.+-.3.3 and 2-butyloctanoic acid 5.2.+-.5.0% all P=0.005 compared to VPA). The inventors also extended these compounds to show a similar efficacy for straight chain nine- and ten-carbon backbones (nonanoic acid, 20.9.+-.7.50, P=2.times.10-6 compared to VPA; decanoic acid 0.23.+-.0.23%, P=2.times.10-6 compared to VPA FIGS. 3d, e). This activity was not seen with shorter backbones (e.g. 5 carbon pentanoic acid--data not shown). These highly potent compounds have not previously been associated with seizure control, and would not be predicted to show teratogenic effects (Guo et al., 1999).

To show that the effect of these compounds was not seizure-model specific, the inventors further investigated the effect of one of these compounds in the in vitro low Mg.sup.2+ seizure model (FIGS. 3f, g, h). The inventors chose 4-methyloctanoic acid (hircinoic acid), since this is endogenous to animal systems (Johnson et al., 1977). VPA (1 mM) weakly reduced the frequency of recurrent short discharges in this model. Application of 4-methyloctanoic acid almost abolished the frequency of recurrent short discharges 30-40 minutes after application. These data therefore suggest that 4-methyloctanoic acid shows enhanced activity over VPA in multiple in vitro models of epileptiform activity.

2.4 Novel Compounds Show Enhanced Status Epilepticus Control

To demonstrate further efficacy in animal seizure models with these compounds, the inventors tested 4-methyloctanoic acid in an in vivo model of status epilepticus (FIG. 4). For this test, status epilepticus was induced by stimulation of the perforant path in awake, freely moving rats as has been previously described (Holtkamp et al., 2001, Walker et al., 1999; FIG. 4a). The inventors have previously found that VPA is effective in this model at high dose (600 mg/kg) but has only partial effectiveness at a lower dose (400 mg/kg). The inventors therefore compared the efficacy of 4-methyloctanoic acid (400 mg/kg) against VPA (400 mg/kg). 4-methyloctanoic acid has a marginally higher molecular weight (MW=158) than that of VPA (MW=144) and so this dose represents a slightly lower molar dose of 4-methyloctanoic acid. VPA strongly attenuated seizures in this model 2 hours after treatment, with reduced efficacy three hours post treatment (FIG. 4b), whereas 4-methyloctanoic acid protected against seizures over the test period. Both compounds reduce spike amplitude and frequency (FIG. 4): VPA reduced spike amplitude (75.2.+-.9.2% in the first hour; 61.1.+-.8.3% in the second hour; 55.1.+-.6.6% in the third hour) (FIGS. 4e,f). In comparison, 4-methyloctanoic showed significantly better control, reducing the mean spike frequency to 39.3.+-.11.3% in the first hour (significantly better than VPA p<0.05); 18.1.+-.10.3% in the second hour (p<0.05 compared to VPA) and 26.9.+-.12.3% in the third hour (FIGS. 4e,f). 4-methyloctanoic terminated status epilepticus (defined as a spike frequency of less than 1 Hz) in all status epilepticus animals. Furthermore, 4-methyloctanoic acid completely stopped the seizures in all 7 animals after 2 hours, whilst VPA decreased seizure severity but did not terminate the seizures in any (P=0.0003, Fisher's exact test), and this effect was maintained in five out of seven animals given 4-methyloctanoic acid by three hours (P=0.01, Fisher's exact test).

2.5 VPA Regulates Fatty Acid Uptake and Release

To analyse a role for VPA-mediated regulation of arachidonic acid release in Dictyostelium, the inventors developed an assay based upon the release of radiolabel from cells containing tritiated fatty acid over time. Using this assay, the inventors showed that following .sup.3H-AA labelling of cells, the release of radiolabel into media was linear over a 45 min period (FIG. 5A). The effect of VPA on radiolabel release from AA-labelled cells was acute and dose dependent, whereby VPA induced a decrease in the release with an IC50 of 89 .mu.M. This effect was not specific to AA, since release of tritiated palmitic acid was also inhibited in the presence of VPA with an IC50 of 163 .mu.M (FIG. 5C). The acute nature of this effect is seen with a significant inhibition following 30 min exposure (p<0.05).

Since reduced release of labelled fatty acid may be due to its re-incorporation into lipids, the inventors also measured fatty acid uptake by measuring radiolabel incorporation of fatty acids into cells. Like fatty acid release, incorporation of tritiated AA was linear over a 30 min period (FIG. 5B), however, VPA caused a dose-dependent increase in the uptake of AA, with an EC50 of 47 .mu.M. This effect was also seen using palmitic acid, with an EC50 value of 160 .mu.M (FIG. 5D) and was significant following 30 min drug treatment (p<0.05).

In order to test if the above effects occur through simple fatty acid membrane insertion, the inventors visualised fatty acid uptake using a compound containing a 12 carbon fatty acid chain linked to fluorescent head group (bodipy; FIG. 6A) (Worsfold et al., 2004). Upon incubation of cells and bodipy-labelled lipid, 0.5 mM VPA caused an increase in the intensity and diameter of fluorescent lipid droplets within cells compared to untreated cells (FIG. 6B). This results show that VPA also increased the uptake of this fatty acid, and that the drug increased fatty acid storage within lipid droplets.

2.6 VPA-induced Fatty Acid Uptake Occurs Independently of Actin Dynamics

Uptake (and release) of compounds in Dictyostelium is likely to be regulated by cellular mechanisms controlling macropinocytosis, thus changes in fatty acid incorporation may be due to simple regulation of this process. To examine this, and since macropinocytosis is dependent upon actin polymerisation, the inventors used latrunculin (10 .mu.M), an inhibitor of actin polymerisation (de Oliveira and Mantovani, 1988), to observe the effects on fatty acid uptake. Inhibition of actin polymerisation decreased uptake in control cells. However, latrunculin failed to attenuate VPA-induced AA uptake suggesting VPA-induced fatty acid regulation was independent of vesicle dynamics (FIG. 6C).

2.7 Genetic Ablation of PLA.sub.2 Activity does not Reverse Fatty Acid Perturbation

To identify specific genes controlling the effect of VPA in this model, the inventors carried out a restriction enzyme mediated integration (REMI) mutant screen to identify loci controlling the effect of VPA during development (Kuspa and Loomis, 2006). VPA inhibits the development of Dictyostelium at concentrations found in plasma of patients receiving VPA treatment (0.28-0.7 mM; (Silva et al., 2008)) (FIG. 7B). One mutant isolated in this screen contained an ablated PLA.sub.2 gene (van Haastert et al., 2007), with the encoded protein showing similarity to Ca.sup.2+-independent enzymes, and containing conserved ATPase and lipase motifs (FIG. 7A), and the mutant showed partial resistance to VPA during development (FIG. 7B). However, the knockout cell line did not attenuate radio-label release from cells (FIG. 7C) suggesting that although disruption of the PLA.sub.2 gene offered partial protection to VPA during development, it was not enough to prevent gross VPA-induced fatty acid release.

2.8 VPA Regulation of Fatty Acid Signalling is not Phenocopied by PLA.sub.2 Inhibition

Since VPA has been suggested to regulate phospholipase A2 (PLA.sub.2) related signalling (Rao et al., 2008), and since ablation of a single PLA.sub.2 gene provided only partial resistance to VPA during development and no effect on gross radiolabel release or fatty acid uptake (FIG. 7), the inventors assessed the role of pharmacological inhibition of PLA.sub.2 on AA regulation. Chemical inhibitors of different PLA.sub.2 class specificity (BEL [80 .mu.M], a Ca.sup.2+-independent PLA.sub.2 inhibitor (Ackermann et al., 1995); MAFP [50 .mu.M], a Ca.sup.2+-dependent and Ca.sup.2+-independent cytosolic PLA.sub.2 inhibitor (Balsinde and Dennis, 1996, Lio et al., 1996); and BPB [20 .mu.M], a phospholipase A2 inhibitor (Mitchell et al., 1976) all reduced radiolabel release from .sup.3H-AA cells in a similar manner to VPA treatment (FIG. 8). Differing specificity for these inhibitors was shown since a cocktail of all three inhibitors provided a cumulative inhibition of release. In contrast to the effect of VPA on release (causing an increase in fatty acid over time), chemical inhibition of PLA.sub.2 activity caused a reduced uptake of fatty acid (FIG. 8). This data suggests that PLA.sub.2 inhibition partially phenocopies the effect of VPA in modifying AA signalling, but that VPA has a more generalised effect on fatty acid signalling.

2.9 VPA-Induced Fatty Acid Uptake is not Dependent on Fatty Acid Activation

To test whether the incorporation of fatty acids was dependent on activation, the inventors firstly tested the ability of cell extracts to activate palmitic acid to form PaA-CoA. In these experiments, incubation of cell extracts with .sup.3H-PaA and coenzyme A enabled the activation of the fatty acid that was subsequently separated by differential solvent solubility and quantified (von Lohneysen et al., 2003). Inclusion of VPA (1.0 mM) either with cell extracts during the activation assay, or by pre-treatment of cells prior to preparation of extracts (10 min, 1 mM) had no effect on fatty acid activation, whereas ablation of the peroxisomal fatty acid CoA synthase A enzyme (FcsA)--the enzyme responsible for fatty acid CoA activation in endosome--showed a significant reduction in PaA-CoA synthesis (FIG. 8) compared to wild-type cells. Furthermore, cell lines lacking fcsA (DDB_G0269242; (von Lohneysen et al., 2003)) showed a VPA-induced increase in fatty acid uptake in a similar manner to wild type cells (FIG. 8). These results suggest that the effect of VPA in regulating fatty acid signalling is independent of fatty acid CoA activation as was previously suggested (Bazinet et al., 2006b).

2.10 Structural Specificity of Induced Fatty Acid Release

SAR studies identify the physical requirements for compounds to cause an effect, and these studies can help to distinguish between discrete effects of a compound. To examine the structural dependency of VPA on radiolabel release following .sup.3H-AA incorporation, the inventors employed a range of compounds related to VPA with varying carbon backbone and side chain lengths, head group, enantiomeric specificity and saturation and measured release following 30 min 0.5 mM treatment. The results obtained are summarised in Table 3 below.

TABLE-US-00003 TABLE 3 AA release [% Compound structure of control] Control 100 VPA (2-propylpentanoic acid) ##STR00001## 50 2-propyloctanoic acid ##STR00002## 45 2-propyldecanoic acid ##STR00003## 56 decanoic acid ##STR00004## 62 4-methylnonanoic acid ##STR00005## 63 nonanoic Acid ##STR00006## 66 4-methyloctanoic acid 66 octanoic acid ##STR00007## 72 2-ethyldecanoic ##STR00008## 73 4-ethyloctanoic acid ##STR00009## 87 dodecanoic acid ##STR00010## 89

The compounds tested showed a range of inhibitory activity, with high structural specificity, and the strength of activity was independent of acidity and lipophilicity (pKA and log P values respectively).

Since VPA gave a reduction in activity to 50% of control, the inventors defined compounds, such as 2-propyloctanoic acid, as highly active since they reduce activity to 45% or below. These highly active compounds were carboxylic acids (since valpromide gave virtually no inhibition, data not shown) branched at the second carbon, with the most active compounds containing an isopropyl group. Unlike teratogenicity (Eickholt et al 2006), a tertiary-substituted C2 still showed activity, and long- and medium-length straight chain fatty acids were still active. Branched compounds were stronger than corresponding straight chains, with a preference for longer side chains (propyl- giving stronger inhibition that methyl- side chains). Finally, unsaturated compounds showed a reduction in inhibitory activity. This SAR study represents a novel description of a VPA-catalysed effect.

To confirm that the dual effects of attenuated fatty acid release and increased fatty acid uptake occur at a single site, the inventors analysed two compounds showing either strong or weak inhibitory effects on radiolabel release (isopropyl-pentanoic acid and 4-methyloctanoic acid, respectively) for effects on .sup.3H-AA uptake (FIG. 9). From these experiments, enhanced inhibition of radiolabel release corresponded with an elevated uptake of fatty acid into cells as compared with VPA, and a reduced effect on release corresponded with a reduced effect on uptake. This data suggest a single, highly structurally-specific site of action for VPA and related compounds in the regulation of fatty acid signalling.

3. Discussion

VPA is used to treat a number of current medical conditions including epilepsy, Bipolar disorders, and migraine and its role is likely to expand widely to include cancer (Blaheta et al., 2006), HIV and Alzheimer's (Qing et al. 2008) treatment (reviewed in Lagace et al., 2005, Terbach & Williams, 2009, Bialer & Yagen, 2007), ischemia (Costa et al. 2006), and fatal blood loss (Alam et al. 2009). Understanding how these conditions are controlled by VPA has proved highly complex since it triggers a variety of cellular changes with unknown primary targets and these changes have not been related to specific clinical conditions, and few structure-function studies have been carried out (reviewed in Lagace et al., 2005, Terbach & Williams, 2009, Bialer & Yagen, 2007, Nalivaeva et al., 2009).

3.1 Effect on Phosphoinositol Signalling

The inventors have shown that VPA causes a dose-dependent reduction in PIP and PIP.sub.2 (FIGS. 1a&b) in the biomedical model, Dictyostelium, and this provides one of the few effects of VPA found to occur in the acute time period shown to protect against induced seizures (FIG. 1b; Honack & Loscher, 1992). The vast majority of research into VPA mechanisms has been complicated by long term treatment leading to changes in gene expression (likely to be mediated by teratogenic effects (Gurvich et al., 2004, Phiel et al., 2001)) or in time periods enabling regulation of multiple indirect targets. Therefore a rapid attenuation of phosphoinositide signalling provides a significant insight into the acute function of VPA. In light of increased phosphoinositide levels during seizures (Van Rooijen et al., 1986), the results shown here provide an exciting breakthrough in our understanding of seizure control.

It is very difficult to identify the target(s) of a drug which modulates phosphoinositide metabolism in vivo in mammalian systems, due to the large number of kinases' and phosphatases, and the promiscuous nature of lipid kinases substrate selection. An advantageous approach, facilitated in Dictyostelium, is to employ isogenic cell lines containing ablated kinase genes. Since it has previously been shown that VPA attenuates PIP3 production (Xu et al., 2007), the inventors analysed phosphoinositide turnover in cell lines lacking all type 1 PI3K activity (Hoeller & Kay, 2007), and in two unrelated lipid kinases null mutants (a receptor-linked phosphatidylinositol 5 kinase (rPKA) (Bakthavatsalam et al., 2006); and a nuclear phosphatidylinositol 4/5 kinase activity (PIPKinA) (Guo et al., 2001; FIG. 2). All three mutant cell lines showed large and significant reductions in PIP turnover following VPA treatment, albeit at reduced levels compared to wild type cells, indicating that these enzymes cannot be the direct target of VPA in phosphoinositide attenuation and that testing of further kinases would be of little benefit.

The inositol depletion theory (Berridge et al., 1989) provides a well-supported theory for action of VPA in bipolar disorder phrophylaxis (Williams et al., 2002, Williams, 2005), potentially through the indirect inhibition of the enzyme responsible for de novo inositol biosynthesis, myo-inositol-1-phosphate synthase (MIP). A strong inhibitory effect on MIP activity has also been shown for a limited number of VPA congeners (Shaltiel et al., 2004; Shaltiel et al., 2007), suggesting inositol regulation may be related to seizure control. However, the newly discovered compounds showing strongly improved seizure control (e.g. 4-MO and nonanoic acid)--do not acutely deplete inositol in the time frame for seizure control, arguing against a role for inositol signalling in this clinical treatment. Einat et al., 2008 also showed that pharmacological inhibition of MIP (through compounds structurally unrelated to VPA) does not control inositol sensitive pilocarpine-induced seizures model. Instead, the data shown here adds weight to the identification of phosphoinositide signalling as playing a key role of seizure occurrence (Backman et al., 2001) and suggests that VPA's effect on phosphoinositide turnover and seizure control is unrelated to that of inositol depletion.

The majority of research concerning VPA targets and clinical efficacy has centred around compounds with either a five carbon backbone (with a branch point on the second carbon) or cyclic derivatives (Bialer & Yagen, 2007). Thus the discovery of a novel family of compounds with longer carbon backbones showing inhibition of phosphoinositide turnover and seizure control provides a major advance in the development of new therapeutics. This novel chemical family includes compounds branched at the second and fourth carbon, suggesting efficacy in compounds with variable branching, and thus provide a new large family of compounds of potential clinical interest.

Here, the inventors tested the efficacy of five compounds within this novel family of fatty acid in an in vitro model of epileptiform activity that is a key model for screening potential antiepileptic drugs (Piredda et al., 1985). All compounds give rise to a greatly increased but reversible reduction in epileptiform activity in a PTZ model (see FIG. 3). This effect is not model specific, since one compound (4-methyloctanoic acid) shows efficacy in a low Mg.sup.2+ model. However, efficacy in in vitro models does not necessarily imply efficacy in vivo, because other factors such as drug metabolism, brain penetration and access to the drug target play an important role. The inventors therefore tested 4-methyloctanoic acid further in an in vivo model of status epilepticus (FIG. 4), where it proved more potent than VPA. This model has been previously shown to be resistant to phenyloin and to respond only to high doses of other anticonvulsants (Chang et al., 2009). These results therefore suggest that this family of compounds may provide a novel seizure control agent with increased efficacy, and since they are predicted to not show teratogenicity, they may also provide reduced side effects compared to VPA treatment.

These findings identify a mechanism of action of VPA in attenuating phosphoinositide turnover in the simple biomedical model, Dictyostelium, and have shown that this effect occurs independently of a range of phosphoinositide kinase enzymes, inositol recycling and depletion. The inventors used this system to identify a novel family of compounds showing increased phosphoinositide attenuation. They then translated this simple model-based research to several clinical models of seizure control and show a large increase in efficacy for five of these compounds over VPA in a hippocampal slice model of epileptiform activity and for one of these compounds in a whole-animal model of status epilepticus. These studies thus suggest seizure control efficacy for a novel family of compounds with longer backbone length and variable side chain length and position compared to VPA. Continued analysis of the biosynthetic pathways controlling phosphoinositide signalling may give rise to significant advances in understanding epilepsy and other VPA-treatable disorders such as bipolar disorder and migraine.

3.2 Effect on Fatty Acid Turnover

Previous studies have suggested a possible mechanism of action of VPA is the attenuation of arachidonic acid turnover (Chang et al., 2001). Here the inventors have demonstrated in the model Dictyostelium discoideum that VPA significantly reduces the release of radiolabel following .sup.3H-AA incorporation. Surprisingly, the inventors also observed that the uptake of .sup.3H-AA was enhanced in the presence of VPA. Since arachidonic acid is not an endogenous fatty acid in Dictyostelium the inventors also verified these results with palmitic acid, a fatty acid found in Dictyostelium (Weeks, 1976). The similar results observed for both poly-unsaturated and saturated fatty acids suggest a common mechanism for the modulation of fatty acid metabolism by VPA. The similarity of these results shown here and those seen in in vivo animal studies, such as decreased AA turnover (Chang et al., 2001) and increased lipid accumulation (Kesterson et al., 1984) suggest that Dictyostelium may be a useful model in the study of VPA induced fatty acid dynamics.

Few studies of fatty acid regulation by VPA in animal models have examined a role for VPA in simply elevating fatty acid transport into cells. Thus, to examine this, and since uptake of extra-cellular nutrients in Dictyostelium is regulated by vesicle dynamics (macropinocytosis), the inventors showed that blocking vesicle function by inhibition action polymerisation did not inhibit the VPA-catalysed increase in fatty acid uptake. These results are supported by their previous results, since they have shown that VPA treatment also inhibits vesicle dynamics in Dictyostelium (Xu et al., 2007), and these results therefore point towards an effect of VPA in regulating fatty acid signalling which is independent of uptake.

Polyunsaturated fatty acid turnover is primarily regulated by PLA.sub.2-catalysed cleavage from lipids to release free fatty acids, and VPA has previously been suggested to regulate PLA.sub.2-dependent signalling (Rapoport and Bosetti, 2002). A role for PLA.sub.2 in VPA-dependent signalling in our model was suggested when a screen for mutants resistant to the effect of VPA during development revealed that ablation of a single PLA.sub.2 gene gave partial resistance to VPA. This provided an exciting link to a potential clinical function of VPA, since elevated levels of PLA.sub.2 have been shown in bipolar disorder patients (Ross et al., 2006) and during seizures (Siesjo et al., 1982; Rintala et al., 1999a; Bazan et al., 2002; Basselin et al., 2003a). However, ablation of the single isoform of PLA.sub.2 identified in the genetic screen did not affect net change in VPA-induced radiolabel release (FIG. 8B). This may be due to the presence of 18 other PLA.sub.2-like genes present in the genome (Fey et al., 2009), since the activities of these gene products are likely to hide small changes caused by single gene ablation in whole-cell assays. However, the identified gene may play a critical, targeted role in Dictyostelium chemotaxis and development--as has been shown (Chen et al., 2007, Kortholt and van Haastert, 2008), and thus partially overcome the VPA-related development effects. This VPA-dependent inhibition of development may not be visible in the assays employed here, since the assay does not differentiate between cell-type specific function nor do it explore multiple time points over development.

To further examine PLA.sub.2 signalling in the observed effect of VPA, the inventors used pharmacological inhibitors of PLA.sub.2 activity, and showed these reduced radiolabel release from .sup.3H-AA labelled cells--thus confirming that VPA mimics the effect of PLA.sub.2 inhibition in this model. However, this effect of VPA is not through direct PLA.sub.2 inhibition, since the VPA-induced increase in fatty acid uptake was not reproduced by PLA.sub.2 inhibitors (FIG. 9). This data, suggests that a PLA.sub.2 inhibition-like effect of VPA may provide only one aspect of the drugs effect in regulating fatty acid turnover, and confirms that PLA.sub.2 is not the primary target of VPA in this effect.

If PLA.sub.2 is not the pharmacological target of VPA, the observed partial phenocopying of PLA.sub.2 inhibitors may point to an upstream disruption in lipid metabolism. For example, a reduced activity of fatty acid acyl CoA synthases may cause a reduced incorporation of radio-labelled fatty acid into phospholipids, resulting in a reduced release of the radio-labelled fatty acid (thus resembling a PLA.sub.2-inhibitory like effect). The inventors show that VPA does not inhibit CoA activation of fatty acids either directly or indirectly. Furthermore, genetic ablation of both of fatty acid CoA synthases (FcsA and FcsB) still showed VPA-dependent regulation of AA uptake, thus confirming that fatty acid activation is not the target of VPA in this effect, in contrast to that reported earlier at high VPA concentrations (Bazinet et al., 2006b).

Finally, the inventors have employed a SARs study to investigate VPA induced fatty acid regulation. This approach provides a highly important insight into VPA action, since VPA is used in the therapeutic treatment of a large range of conditions (Terbach and Williams, 2009), with numerous cellular effects remaining un-associated with each therapeutic role, and very few of these effects have an identified primary target (Lagace and Eisch, 2005, Terbach and Williams, 2009). SARs studies can be used to help differential these therapeutic treatments, mechanisms and targets. For example, range of VPA-related compound showing histone deacetylase inhibition as a cellular function have been shown to cause teratogenicity as a biomedical action, and the structural definition can now predict teratogenicity in novel compounds (Phiel et al., 2001). From the SARs study for fatty acid turnover, this effect is not related to teratogenicity since some compounds show strong fatty acid regulation but no predicted teratogenicity (Radatz et al., 1998). Another cellular effect of VPA in Dictyostelium and mammalian neurons is the inhibition of inositol based signalling (Williams et al., 2002, Eickholt et al., 2005, Shimshoni et al., 2007). Although some compounds (e.g. isopropyl-pentanoic acid) are strongly active in both inositol signalling attenuation and fatty acid regulation, these effects are not shared by all compounds, thus indicating these fatty acid- and inositol-based signalling effects are independent. Thus fatty acid regulation, HDAC inhibition and, inositol depletion provide three independent mechanisms of action for VPA. Employing VPA structures showing increased or decreased activity in fatty acid regulation may therefore give rise to increased therapeutic activity or a reduction in unwanted side effects in VPA-treatable conditions.

One clinical corollary of this work is shown in VPA-dependent lipid droplet formation. This effect has been shown in systems ranging from S. cerevisiae (Sun et al., 2007) to hippocampus and neocortex of the rat brain (Sobaniec-Lotowska, 2005), clearly indicating that the observed effects reported here are unlikely to be model specific. Furthermore, lipid droplet formation has also been associated with hepatotoxicity, although the mechanism remains unclear (Fujimura et al., 2009). The identification of a structural specificity for this effect in lipid regulation provides a potential mechanism for selection of novel therapeutics lacking this effect. The structural isolation of VPA-dependent increased lipid accumulation may also go some way to explaining the weight gain associated with patients undergoing VPA treatments (Wirrell, 2003, Masuccio et al., 2010, Verrotti et al., 2010).

In conclusion the inventors have described here a model for the study of VPA-induced fatty acid regulation. PLA.sub.2 inhibition phenocopies some but not all of these VPA-dependent effects, but PLA.sub.2 is unlikely to be the primary target of VPA. The role of this effect is likely to function in both bipolar disorder treatment and seizure control, since increased PLA.sub.2 activity is implicated in both conditions (Yegin et al., 2002, Rao et al., 2007). Identifying novel structures for this effect, comprising carboxylic acids, branched on the C2 position, with short (five carbon) to medium length (nine carbon) backbone and side chain (ethyl or propyl) provides potential new therapies for both conditions. The future definition of the primary site of action for this effect will significantly aid our understanding of VPA and related therapeutics.

All documents cited herein are hereby incorporated by reference in their entirety within this disclosure.

REFERENCES

Ackermann E J, Conde-Frieboes K, Dennis E A, Journal of Biological Chemistry 270, 445-450 (1995).

Alam et al. Surgery 146, 325-333 (2009)

Armand, V., Louvel, J., Pumain, R., & Heinemann, U. Epilepsy Res. 32, 345-355 (1998).

Backman, S. A. et al. Nat. Genet. 29, 396-403 (2001).

Bakthavatsalam, D., Meijer, H. J., Noegel, A. A., & Govers, F. Trends Microbiol. 14, 78-382 (2006).

Blaheta, Michaelis, Driever & Cinatl Med. Res. Rev. 25, 383-397 (2005). 3. Deubzer et al. Leuk. Res. 30, 1167-1175 (2006)

Balsinde J, Dennis E A, Journal of Biological Chemistry 271, 6758-6765 (1996).

Basselin M, Chang L, Bell J M, Rapoport S I, Neuropsychopharmacology 31, 1659-1674 (2005).

Basselin M, Chang L, Seemann R, Bell J M, Rapoport S I, J. Neurochem. 85, 1553-1562 (2003).

Bazan N G, Tu B, Rodriguez de Turco E B, Prog. Brain. Res. 135, 175-185 (2002).

Bazinet R P, Rao J S, Chang L, Rapoport S I, Lee H J, Biol. Psychiatry 59, 401-407 (2006a).

Bazinet R P, Weis M T, Rapoport S I, Rosenberger T A, Psychopharmacology (Berl) 184, 122-129 (2006b).

Berridge, M. J., Downes, C. P., & Hanley, M. R. Cell 59, 411-419 (1989).

Bialer, M. & White, H. S, Nat Rev Drug Discov. 9, 68-82 (2010).

Bialer, M. & Yagen, B. Neurotherapeutics. 4, 130-137 (2007).

Boeckeler K, Adley K, Xu X, Jenkins A, Jin T, Williams R S, Eur. J. Cell Biol. 85, 1047-1057 (2006).

Chang, P., Chandler, K. E., Williams, R. S., & Walker, M. C. Epilepsia (2009).

Chang M C, Contreras M A, Rosenberger T A, Rintala J J, Bell J M, Rapoport S I, J. Neurochem. 77, 796-803 (2001).

Chapman, A. G., Meldrum, B. S., & Mendes, E. Life Sci. 32, 2023-2031 (1983).

Chen C T, Green J T, Orr S K, Bazinet R P Prostaglandins Leukot Essent. Fatty Acids 79, 85-91 (2008).

Chen L, Iijima M, Tang M, Landree M A, Huang Y E, Xiong Y, Iglesias P A, Devreotes P N, Dev. Cell 12, 603-614 (2007).

Chiu C C, Huang S Y, Su K P, Lu M L, Huang M C, Chen C C, Shen W W, Eur. Neuropsychopharmacol 13, 99-103 (2003).

Costa et al. Stroke 37, 1319-1326 (2006)

de Oliveira C A, Mantovani B, Life Science 43, 1825-1830 (1988).

Deutsch, J., Rapoport, S. I., & Rosenberger, T. A. Neurochem. Res. 28, 861-866 (2003).

Drancourt M, Bollet C, Carta A, Rousselier P, Int. J. Syst. Evol. Microbiol. 51, 925-932 (2001).

Drayer, A. L., Van Der, K. J., Mayr, G. W., & Van Haastert, P. J. EMBO J. 13, 1601-1609 (1994).

DSMV IV American Psychiatric Association: Diagnostic and statistical manual of mental disorders (American Psychiatric Association, Washington D.C., 2000).

Eickholt B J, Towers G J, Ryves W J, Eikel D, Adley K, Ylinen L M, Chadborn N H,

Harwood A J, Nau H, Williams R S, Mol. Pharmacol. 67, 1426-1433 (2005).

Eikel D, Lampen A, Nau H, Chem. Res. Toxicol. 19, 272-278 (2006).

Einat, H., Tian, F., Belmaker, R. H., & Frost, J. W. J. Neural Transm. 115, 55-58 (2008).

Eyal S, Yagen B, Shimshoni J, Bialer M, Biochem. Pharmacol. 69, 1501-1508 (2005).

Faix J, Kreppel L, Shaulsky G, Schleicher M, Kimmel A R, Nucleic Acids Research 32, e143 (2004).

Fey P, Gaudet P, Curk T, Zupan B, Just E M, Basu S, Merchant S N, Bushmanova Y A, Shaulsky G, Kibbe W A, Chisholm R L, Nucleic Acids Research 37, D515-519 (2009).

Fujimura H, Murakami N, Kurabe M, Toriumi W, J. Appl. Toxicol. 29, 356-363 (2009). Guo, Q. et al. Nat. Med. 5, 101-106 (1999).

Gurvich, N., Tsygankova, O. M., Meinkoth, J. L., & Klein, P. S. Cancer Res. 64, 1079-1086 (2004).

Hoeller, O. & Kay, R. R. Curr. Biol. 17, 813-817 (2007).

Honack, D. & Loscher, W. Epilepsy Res. 13, 215-221 (1992).

Holtkamp, M., Tong, X., & Walker, M. C. Ann. Neurol. 49, 260-263 (2001).

Isoherranen, N., Yagen, B., & Bialer, M. Curr. Opin. Neurol. 16, 203-211 (2003).

Johnson C. B., Wong E., & Birch E. J. Lipids 12: 340-347 (1977).

Kaufmann, D., Bialer, M., Shimshoni, J. A., Devor, M., & Yagen, B. J. Med. Chem. 52, 7236-7248 (2009).

Keane, P. E., Simiand, J., Mendes, E., Santucci, V., & Morre, M. Neuropharmacology 22, 875-879 (1983).

Kesterson J W, Granneman G R, Machinist J M, Hepatology 4, 1143-1152 (1984).

Kim H W, Rapoport S I, Rao J S, Mol. Psychiatry. (2009).

King, J. S. et al. Dis. Model. Mech. 2, 306-312 (2009).

Kortholt A, van Haastert P J, Cell Signal 20, 1415-1422 (2008).

Kuspa A, Loomis W F, Methods Mol. Biol. 346, 15-30 (2006).

Lagace D C, Eisch A J, Psychiatr. Clin. North Am. 28, 399-414 (2005).

Lands W, Crawford C, New York: John Wiley & Sons (1976).

Lio Y C, Reynolds L J, Balsinde J, Dennis E A, Biochim. Biophys. Acta. 1302, 55-60 (1996).

Liu, M. J. & Pollack, G. M. P Epilepsia 35, 234-243 (1994).

Loscher, W., Fisher, J. E., Nau, H., & Honack, D. J. Pharmacol. Exp. Ther. 250, 1067-1078 (1989).

Loscher W, Nau H, Neuropharmacology 24, 427-435 (1985).

Maslanski, J. A. & Busa, W. B. Methods in Inositide Research (ed. Irvin, R. F.) 113-126 (Raven Press Ltd., New York, 1990).

Masuccio F, Verrotti A, Chiavaroli V, de Giorgis T, Giannini C, Chiarelli F, Mohn A, J. Child Neural. (2010).

Meunier H, Carraz G, Neunier Y, Eymard P, Aimard M, Therapie 18, 435-438 (1963).

Mitchell S M, Poyser N L, Wilson N H, Br. J. Pharmacol. 58, 295P (1976).

Mora, A., Gonzalez-Polo, R. A., Fuentes, J. M., Soler, G., & Centeno, F. Eur I Biochem. 266, 886-891 (1999).

Mora, A., Sabio, G., Alonso, J. C., Soler, G., & Centeno, F. Bipolar Disord. 4, 195-200 (2002).

Nalivaeva, N, N., Belyaev, N. D., & Turner, A. J. Trends Pharmacol. Sci. 30, 509-514 (2009).

Pawolleck, N. & Williams, R. S. Methods Mol. Biol. 571, 283-290 (2009).

Phiel, C. J. et al. J. Biol. Chem. 276, 36734-36741 (2001).

Piredda, S., Yonekawa, W., Whittingham, T. S., & Kupferberg, H. J. Epilepsia 26, 167-174 (1985).

Qing et al. J. Exp. Med. 205, 2781-2789 (2008)

Radatz M, Ehlers K, Yagen B, Bialer M, Nau H, Epilepsy Research 30, 41-48 (1998).

Rao J S, Ertley R N, Rapoport S I, Bazinet R P, Lee H J, J. Neurochem. 102, 1918-1927 (2007).

Rao J S, Lee H J, Rapoport S I, Bazinet R P, Mol. Psychiatry. 13, 585-596 (2008).

Rapoport S I, J. Nutr. 138, 2515-2520 (2008a).

Rapoport S I, Prostaglandins Leukot Essent. Fatty Acids 79, 153-156 (2008b),

Rapoport S I, Bosetti F, Arch. Gen. Psychiatry 59, 592-596 (2002).

Rintala J, Seemann R, Chandrasekaran K, Rosenberger T A, Chang L, Contreras M A, Rapoport S I, Chang M C, Neuroreport 10, 3887-3890 (1999).

Ross B M, Hughes B, Kish S J, Warsh J J, Bipolar Disord. 8, 265-270 (2006).

Shaltiel, G., Mark, S., Kofman, O., Belmaker, R. H., & Agam, G. Pharmacol. Rep. 59, 402-407 (2007a).

Shaltiel, G., Dalton, E. C., Belmaker, R. H., Harwood, A. J., & Agam, G. Bipolar. Disord. 9, 281-289 (2007b).

Shaltiel, G. et al. Valproate decreases inositol biosynthesis. Biol. Psychiatry 56, 868-874 (2004).

Shimshoni, J. A. et al. Mol. Pharmacol. 71, 884-892 (2007).

Siesjo B K, Ingvar M, Westerberg E, J. Neurochem. 39, 796-802 (1982).

Silva M F, Aires C C, Luis P B, Ruiter J P, Ijlst L, Duran M, Wanders R J, Tavares de Almeida I, J. Inherit. Metab. Dis. (2008).

Sobaniec-Lotowska M E, Int. J. Exp. Pathol. 86, 91-96 (2005).

Storey, N. M., O'Bryan, J. P., & Armstrong, D. L. Curr. Biol. 12, 27-33 (2002).

Sun Q, Bi L, Su X, Tsurugi K, Mitsui K, FEBS Lett. 581, 3991-3995 (2007).

Terbach N, Williams R S, Biochem. Soc. Trans. 37, 1126-1132 (2009).

Tokuoka, S. M., Saiardi, A., & Nurrish, S. J. Mol. Biol. Cell 19, 2241-2250 (2008).

Vaden, D. L., Ding, D., Peterson, B., & Greenberg, M. L. J. Biol. Chem. 276, 15466-15471 (2001).

van Haastert P J, Keizer-Gunnink I, Kortholt A, J. Cell Biol. 177, 809-816 (2007).

Van Rooijen, L. A., Vadnal, R., Dobard, P., & Bazan, N. G. Biochem. Biophys. Res. Commun. 136, 827-834 (1986).

Verrotti A, Manco R, Agostinelli S, Coppola G, Chiarelli F, Epilepsia 51, 268-273 (2010).

von Lohneysen K, Pawolleck N, Ruhling H, Maniak M, Eur. J. Cell Biol. 82, 505-514 (2003).

Walker, M. C. et al. Epilepsia 40, 359-364 (1999).

Weeks G, Biochim. Biophys. Acta. 450, 21-32 (1976).

Williams, R. S. B. Clinical Neuroscience Research 4, 233-242 (2005).

Williams R S, Cheng L, Mudge A W, Harwood A J, Nature 417, 292-295 (2002).

Williams R S, Eames M, Ryves W J, Viggars J, Harwood A J, EMBO J. 18, 2734-2745 (1999).

Wilson D B, Prescott S M, Majerus P W (1982) Discovery of an arachidonoyl coenzyme A synthetase in human platelets. J Biol Chem 257: 3510-3515

Wirrell E C, Pediatr, Neurol. 28, 126-129 (2003).

Worsfold O, Toma C, Nishiya T, Biosens. Bioelectron. 19, 1505-1511 (2004).

Xu X, Muller-Taubenberger A, Adley K E, Pawolleck N, Lee V W, Wiedemann C, Sihra T S, Maniak M, Jin T, Williams R S, Eukaryot. Cell 6, 899-906 (2007).

Yedgar S, Cohen Y, Shoseyov D, Biochim. Biophys. Acta. 1761, 1373-1382 (2006).

Yegin A, Akbas S H, Ozben T, Korgun D K, Acta. Neurol. Scand. 106, 258-262 (2002).