Patent Application
20060270634
The present invention relates to the use of lysophosphatidic acid (LPA) analogs to modulate LPA receptor (LPA-R) activity, peroxisome proliferator-activated receptor gamma (PPARγ) activity, and to inhibit lysophospholipase D (autotaxin, ATX) activity.
Lysophosphatidic acid (LPA; 1-acyl-3-phosphoglycerol) interacts with both intracellular and extracellular targets. Included among the known targets of LPA are the cell-surface G-protein coupled receptors (GPCR), the nuclear peroxisome-proliferator-activated receptor γ (PPARγ), and the secreted cancer cell motility factor autotaxin.
Not surprisingly, given the number of targets with which LPA has thus far been found to interact, it has also been demonstrated to play a role in a variety of physiological pathways that have also been associated with cellular and tissue development, as well as pathways associated with disease states such as atherosclerosis, cancer, diabetes, and even acne. There are, therefore, numerous efforts being made in laboratories around the world to determine how best to modulate LPA in a more target-specific manner, as well as to develop agents that modulate LPA and/or the targets of LPA for therapeutic use. One such agent is the PPAR-γ agonist rosiglitazone maleate (5-((4-(2-(methyl-2-pyridinylamino)ethoxy)phenyl)methyl)-2,4-thiazolidinedione, which is currently produced by Glaxo Smith Kline under the brand name Avandia®. Rosiglitazone is an oral drug used for treating patients with type 2 diabetes.
Given the importance of LPA and its associated targets in metabolism, there is a need in the art for new agents that may selectively modulate the effects of these molecules and be of therapeutic value in the treatment and prevention of disease.
The present invention relates to LPA analogs that act as agonists or antagonists of LPA1, LPA1, LPA3, PPARγ, and/or autotaxin, the LPA analogs comprising acetal phosphatidic acids. Acetal phosphatidic acid LPA analogs as described herein may be described by the formula
where R1 is O or S, and R2 is a straight or branched chain saturated or unsaturated, linear or cyclic hydrocarbon, substituted or unsubstituted, preferably C6 to C24, having an agonist or antagonist effect on LPA1, LPA2, LPA3, PPARγ, and/or autotaxin. In some embodiments, the analogs may comprise water soluble salts of an acetal phosphatidic acid.
The invention also provides a therapeutic method of modulating an LPA-, PPARγ-, and/or autotaxin-mediated disease in a human or animal subject, the method comprising administering to the subject a therapeutically effective amount of at least one acetal phosphatidic acid having agonist or antagonist activity on LPA1, LPA1, LPA3, PPARγ, autotaxin, or a combination thereof.
Also provided are fatty acid alcohol derivatives having LPA receptor agonist/antagonist activity, PPARγ agonist/antagonist activity, and autotaxin modulating activity, as well as a method of use of such fatty acid alcohol LPA analogs to treat or prevent LPA-, PPARγ, and/or autotaxin-mediated disease.
In 1949 Vogt reported isolation of an acidic phospholipid from the horse intestine that was capable of effecting smooth muscle contraction. This substance named Darmstoff at that time by Vogt has been shown to be a mixture of acetal phosphatidic acids. The inventors have demonstrated that Darmstoff analogs constitute a new class of subtype-selective LPA agonists and antagonists and developed a general and facile method for the synthesis of water-soluble salts of these analogs. They have demonstrated that Darmstoff analogs provide subtype-specific LPA G-protein coupled receptor (GPCR) ligands, as well as activators of nuclear transcription factor PPARγ and inhibitors of lysophospholipase D (autotaxin). The structures of LPA and the three major acetal phosphatidic acids which generally comprise Darmstoff are shown in
Pyridinium chloride (PCC) mediated oxidation of fatty alcohols produced the corresponding aldehydes, which were condensed with glycerol in the presence of PTSA under conditions previously reported in the literature to give dioxolanes (
The synthesis of compounds 21 and 22 containing a phenyl ring in the lipid chain is shown in
To investigate the effect of stereochemistry on biological activity, the inventors synthesized all stereoisomers of Darmstoff analogs 13 and 14. 2,4-disubstituted-1,3-dioxolanes of this type exist as a mixture of four stereoisomers. The inventors' synthetic approach for the preparation of the four possible stereoisomers of 13 and 14 is outlined in
The biological effects of all synthesized compounds were testing using three high-throughput assays. Representative compounds are shown in Tables 1 and 2. First, intracellular calcium transients in rat hepatoma (RH777, an LPA receptor null cell) cell lines individually expressing either LPA1, LPA2, and LPA3 receptors were analyzed to evaluate compounds as agonists or antagonists. Wild type RH7777 cells did not respond to any of the Darmstoff analogs. Second, PPARγ activation was examined in CV1 cells (an African green monkey kidney cell line), transfected with an acyl-coenzyme A oxidase-luciferase (PPRE-Acox-Rluc) reporter gene construct according to a protocol similar to that described by McIntyre, et al. (Proc. Natl. Acad. Sci. USA. 100:131-136 (2003)). The PPRE-Acox-Rluc construct contains a renilla luciferase coding region, an acyl-CoA oxidase coding region, and a PPAR response element coding region. Third, inhibition of the lysophospholipase D autotoxin was determined using a previously described procedure.
Compound 12 (Table 1) containing a C13 alkyl chain and no double bond inhibited Ca2+ mobilization in cells expressing all three LPA GPCRs, thereby providing a pantagonist of LPA1-3. An increase in chain length to C18 and introduction of the C9═C10 double bond resulted in analog 13, which produced LPA1/3 antagonist activity. Oleoyl-LPA is an agonist of LPA1-3 while Darmstoff analog 13 containing an oleoyl chain at the C-2 position of the 1,3-dioxolane inhibited LPA1-3 receptors, indicating that the acetal moiety plays a significant role in ligand recognition. To examine the effect of this modification with the Darmstoff series, compound 14 was synthesized. This analog was an agonist at all three LPA receptor subtypes and was most potent at LPA3 (EC50 of 639 nM). The phosphate analog 15 with conjugated double bonds at C9═C10, C12═C13, and C15═C16 positions was an agonist for all three LPA receptors. Though analog 15 was less potent than 14, these compounds were identified as two LPA GPCR pan-agonists.
A multitude of aldehydes are produced via oxidative cleavage of unsaturated fatty acids and their phospholipid derivatives. The cis-olefinic bond of analogs 13-15 is susceptible to oxidative cleavage. In order to avoid this problem and to examine the effect of structural rigidity on biological activity, the inventors replaced the double bond with an aromatic ring and screened against LPA GPCR, PPARγ, and autotoxin. Incorporation of an aromatic ring in the alkyl chain gave compounds 21 and 22. Analog 21 was an antagonist of LPA1/3 receptors but had no effect against LPA2. The thiophosphate analog, compound 22 was a weak LPA, antagonist, without any effect on LPA2 but stimulated LPA3 with an EC50 of 692 nM (Emax=87%).
To examine the importance of stereochemistry on biological activity, the inventors analyzed pure stereoisomers (Table 3) with respect to LPA GPCR activation. Results indicated that, regardless of stereochemistry at C-2 and C-4, Darmstoff analogs 30, 34, 37, and 39 with phosphate head groups were LPA3 antagonists, whereas analogs with thiophosphate groups 31, 35, 38 and 40 were pan-agonists. Among the phosphate stereoisomers, analog 34 was identified as the most potent LPA3 antagonist with an IC50 of 136 nM (Ki=83 nM). Interestingly, compound 30 weakly activated LPA1/3 and was a partial LPA2 agonist with an EC50 of 1.17 μM (Emax=39%). Stereoisomers with a thiophosphate head group were found to be more potent than parent compound 14. In this series, all other stereoisomers (31, 35 and 38) with the exception of 40 were full agonists of LPA3 receptor, with the most potent being 31 (EC50 of 127 nM, Emax=127%).
PPARγ is a lipid-activated transcription factor that belongs to the nuclear hormone superfamily. The inventors examined the activity of all synthesized Darmstoff analogs as PPARγ activators in vitro in CV1 cells using a PPRE-Acox-Rluc reporter gene assay. Rosiglitazone, a known PPARγ agonist, was used as a positive control for comparison. The results shown in Table 3 indicate that all tested Darmstoff analogs, regardless of whether they had been found to be antagonists or agonists of LPA GPCR, activated the PPARγ reporter construct.
LPA is liberated as the product of lysophosphatidylcholine hyrolysis by the lysophospholipase D, autotaxin (ATX). The inventors screened Darmstoff analogs for ATX inhibition. The IC50 values and percentage of inhibition for Darmstoff analogs are listed in Table 3. As indicated by the data, all tested analogs are capable of ATX inhibition independently of their ligand properties at LPA GPCR and PPARγ. Of the tested compounds, 31, an LPA1-3 pan-agonist with preference for LPA3 was the most effective ATX inhibitor, with an IC50 of 252 nM.
Synthesized compounds were tested for induction and inhibition of LPA-induced calcium transients in rat hepatoma (RH7777) cell lines that stably express individual LPA1, LPA2, and LPA3 receptors as described in the literature using a FlexStation II automated fluorimeter (Molecular Devices, Sunnyvale, Calif.). The results are shown in Table 1. Compound 4 inhibited LPA response with an IC50 of 1.1 μM and 2.87 μM for LPA1 and LPA3 respectively. However, the same compound was a weak agonist of LPA2 with an EC50 of 1.18 μM (Table 1). Interestingly, compound 5 was a full LPA2 agonist, and LPA1/3 antagonist. Modifications of the head group and double bond in compound 5 to analogs 6 and 7 respectively, provided two new pan-agonists of LPA1/2/3. Replacement of the cis-olefinic bond in 6 (pan-agonist of LPA1/2/3) with an aromatic ring provided a sub-type selective LPA3 antagonist (15) which has no effect on LPA1/2.
NE = No Effect, ND = Not Determined
NEb
aEmax = maximal efficacy of drug/maximal efficacy of LPA 18:1, expressed as a percentage.
bNE = no effect observed at the highest concentration (30 μM) tested.
cWA = weak antagonist.
The invention therefore provides LPA receptor agonists and antagonists, as PPARγ agonists having the structure:
where R1 is O or S, and R2 is a straight or branched chain saturated or unsaturated, linear or cyclic hydrocarbon, substituted or unsubstituted, preferably C6 to C24, having an agonist or antagonist effect on LPA1, LPA2, LPA3, PPARγ, and/or autotaxin. Compounds described by the present invention may also include, for example, salts, preferably water-soluble salts, such as ammonium, diammonium, potassium salts of the acetal phosphatidic acids.
The invention also provides methods for treating LPA- and/or PPARγ-mediated diseases by administering to a patient a therapeutically effective amount of an acetal phosphatidic acid analog of lysophosphatidic acid. A compound of the present invention may be provided, for example, as a therapeutic agent for the treatment and/or prevention of atherosclerosis, diabetes, cancer, and other LPA- and/or PPARγ-mediated diseases.
Modified alkyl-phenyl-alkyl phosphoric acid esters and straight chain di-halo phosphonates were synthesized by methods described in U.S. patent application Ser. No. 10/963,085 (Publication No. 2006/0009507A1) and tested for activity against the LPA receptors. A preferred scheme for the synthesis of the difluoro-alkyl phosphonates and reagents (a) (i) LDA, −78° C., THF; (ii) C14H29Br, 40% (b) (i) TMSBr, CH2Cl2, 6 h., rt; (b) MeOH/H2O, 78% is shown below and effects of synthesized compounds on LPA receptor subtypes are shown in Table 4.
Based on the observation that the shorter chain LPAs exert little or no activity on LPA receptors, the inventors identified short chain phosphatidic acid derivatives dioctanoyl glycerol pyrophosphate (DGPP 8:0, 1) and phosphatidic acid 8:0 (PA 8:0, 2) as subtype-selective LPA1 and LPA3 receptor antagonists.
The inventors had discovered that the replacement of phosphate headgroup by thiophosphate, in fatty alcohol phosphates (FAP) series, had a positive effect by improving the agonist as well as antagonist activities at LPA GPCR. To develop improved agents for LPA-R binding, they synthesized stereoisomers of PA 8:0 analogs evaluated their interaction with LPA GPCR and PPARγ. Their data indicated that LPA receptors stereoselectively interact with the glycerol backbone modified ligands. With dioctyl PA 8:0 compounds, they observed stereospecific responses, in which (R)-isomers found to be agonists whereas the (S)-isomers were antagonists of LPA GPCR. From this series, they identified compound 13b as a potent LPA3 receptor subtype-selective agonist (EC50=3 nM), and 8b as a potent and selective LPA3 receptor agonist (Ki=5 nM). Serinde diamide phosphate 19b was identified as an LPA3 receptor specific antagonist with no effect on LPA1, LPA2 and PPARγ.
Dioctanoyl PA analogs were synthesized as shown in
The dialkyl PA 8:0 (APA 8:0) analogs were synthesized as shown in
The serinediamide phosphate/thiophosphate (SDP/SDTP) analogs were synthesized as outlined in
Previously reported DGPP 8:0 (1) and PA 8:0 (2) subtype selective antagonists of LPA1 and LPA3 receptors with an order of magnitude preference for LPA3 receptor were derived from the natural sources and were available only in (R)-enantiomeric form. The activities of the (S)-enantiomers had not assessed at LPA GPCR. The inventors hypothesized that the PA 8:0 scaffold interacts with LPA receptors in a stereoselective manner, and synthesized and evaluated several PA 8:0 analogs. Keeping the hydrophobic chain length constant as in PA 8:0, they modified the phosphate headgroup to a thiophosphate, glycerol backbone to a serine, and varied the hydrophobic chain linkage, to provide analogs that were tested for the agonist and antagonist activities at LPA1, LPA2 and LPA3 receptors.
RH7777 cells, which lack LPA GPCR, were stably transfected with individual LPA1, LPA2 and LPA3 receptors and used for the in vitro screening. The ability of these compounds to activate intracellular LPA receptor PPARγ was also assessed in CV1 cells transfected with an acyl-coenzyme A oxidase-luciferase (PPRE-Acox-Rluc) reporter gene. The results obtained are shown in Table 5.
(2S)-PA 8:0 compound (7a), like its enantiomer (2R)-PA 8:0 (2), showed subtype selective antagonism at LPA1 and LPA3 receptors with no effect on LPA2 receptor. PA analogs enantioselectively antagonize both LPA1 and LPA3 receptors with a moderate preference for S-isomer at LPA1. The antagonistic selectivity is reversed at LPA3 receptor, which showed a preference for (2R)-PA 8:0 over (2S)-isomer. (2R)-TPA 8:0 (8b) was a more potent antagonist than the phosphate analog (2R)-PA 8:0 (2) at both the LPA1 and LPA3 receptors, and a partial agonist of LPA2. 8b was identified as the most potent and selective LPA3 receptor antagonist reported so far with a Ki value of 5 nM and 75-fold selectivity over LPA1. (2S)-TPA 8:0 (8a) lacked LPA1/3 antagonism, but was a partial agonist at LPA2 and LPA3. Results for these newly-synthesized compounds, in accordance with previously published reports, show that LPA receptors exhibit stereoselectivity in interacting with the sn-2 substituted glycerol analogs.
To increase the stability of the acyl-PA 8:0 analogs against chemical as well as phospholipase A (PLA) degradation, the inventors synthesized 12a-b and 13a-b, alkyl derivatives of PA 8:0, and evaluated their agonist and antagonist properties at LPA GPCR. In general, thiophosphates were more potent than corresponding phosphates regardless of agonist/antagonist activity. Enantiospecific activation of LPA1/3 receptors by (2R)-APA 8:0 (12b) and 2(R)-ATPA 8:0 (13b) was observed. Compound 13b, which has the identical (R)-configuration as the endogenous LPA, was the most potent and LPA3 subtype-selective receptor full agonist (EC50=3 nM, Emax=109%), and was ˜230 and ˜1900 fold selective for LPA3 over LPA1 and LPA2, respectively. At LPA3 receptor, dioctyl thiophosphate analog 13b was a more potent agonist than the corresponding phosphate (12b) and LPA 18:1 (Table 5). In contrast to the (2R)-alkyl analogs, the opposite (2S)-enantiomers were antagonists at LPA1/3 receptors. Although compounds (R)-VPC12204 and (S)-VPC12249 were the first to demonstrate enantiospecific agonist and antagonist responses, respectively, at LPA, receptor, both enantiomers were antagonists at LPA3 receptor. The enantiosepecific activation of LPA1/3 receptors by APA analogs may be due to the favorable orientation of the conformationally flexible alkyl side chains of (R)-isomers in ligand binding pocket of these receptors. The sidechains of acyl PA analogs are relatively constrained due to the ester linkage to glycerol, and may not be able to have these favorable interactions with the receptors. Except 2(R)-ATPA 8:0 (13b), which was a weak and partial agonist of LPA2, alkyl PA analogs had no effect on LPA2 receptor.
Replacement of the glycerol backbone by serine is well tolerated at LPA GPCR. Surprisingly, SDP 8:0 analogs (19a-b) were identified as LPA3 receptor subtype-specific antagonists with no effect on LPA1 and LPA2 receptors. SDP 8:0 isomers also demonstrated enantioselectivty in LPA3 antagonism with a preference for (S)-isomer (19a) over R (19b). Thiophosphate head group modification in serinediamides (20a-b) not only improved the LPA3 antagonistic activity but also resulted in loss of LPA3 subtype-specificity by rendering the LPA, antagonistic ability.
Results of in vitro PPARγ activation assay of these compounds in CV1 cells, transfected with PPARγ and PPRE-Acox-Rluc reporter gene, are shown in
Compound 13b was identified as a potent LPA3 receptor subtype-selective agonist and compound 8b (Ki=5 nM) as the most potent subtype-selective LPA3 receptor antagonist so far. Finally, using serine as a backbone substitute, an LPA3 receptor specific antagonist 19b was discovered with no effect on LPA1, LPA2 and PPARγ.
aEmax = maximal efficacy of the drug/maximal efficacy of LPA 18:1, expressed as the percentage.
bNE = no effect.
The invention therefore provides compounds comprising LPA analogs
where R1 is O or S and R2 is a linear or branched chain, saturated or unsaturated, substituted or unsubstituted hydrocarbon having an agonist or antagonist effect on LPA receptor 1, 2, 3, or PPARγ. Compounds described by the present invention may also include, for example, salts, preferably water-soluble salts, such as ammonium, diammonium, potassium salts of the fatty alcohol phosphate.
As used herein, an “analog” is a chemical compound that is structurally or functionally similar to a known compound. Compounds of the present invention have demonstrated that they provide a benefit in modulating the effects of LPA through its extra-cellular receptors LPA1, LPA2, and/or LPA3, as well as through its intracellular receptor PPARγ and the enzyme autotaxin. Compounds as described herein, or pharmaceutically acceptable salts of the compounds, may be administered to a subject separately or together in any conventional dosage forms, including, oral, buccal, sublingual, ocular, topical (e.g., transdermal), parenteral (e.g., intravenous, intramuscular, or subcutaneous), rectal, intracisternal, intravaginal, intraperitoneal, intravesical, local (e.g., powder, ointment, or drop), nasal and/or inhalation dosage forms. For therapeutic use, compounds of the present invention may be provided to a patient in oral form by means of tablets, capsules, liquids, softgels, or other modes of delivery known to those of skill in the art. Patients may also receive treatment using one or more compounds of the present invention delivered intravenously, intraperitoneally, intranasally, via a device to dispense medication at a steady rate, at predetermined intervals, as determined by the patient, or as the need for the medication is detected. Therapeutic medications comprising compounds pharmaceutically acceptable salts thereof may be administered in the form of a pharmaceutical composition comprising a pharmaceutically acceptable carrier, vehicle, or diluent.
As indicated herein, compounds have been described that have been determined to have effect at nanomolar concentrations. Some compounds demonstrate certain effects at millimolar concentrations. Determination of appropriate dosages for therapeutic use, given the information regarding effective concentrations provided herein, is within the skill of those in the art of pharmaceutical design and production.
The invention may be further described by means of the following non-limiting examples:
Synthesis of Darmstoff Analogs
Bis-(2-cyano-ethyl)-2-heptadec-8-enyl-(1,3)dioxolan-4yl methyl phosphate (5). To a solution of alcohol 4 (0.245 g, 0.72 mmol) in dichloromethane (15 ml), 1H-tetrazole (0.2 g, 2.85 mmol) was added. After 10 minutes, to this solution biscyano ethyl diisopropyl phosphoramidite (0.39 g, 1.44 mmol) was added and stirred for 30 minutes. H2O2 (0.25 ml) was added to the reaction mixture and stirred for an additional 30 minutes. The reaction mixture was diluted with dichloromethane (100 ml) and the solution was washed sequentially with saturated aqueous Na2S2O5, saturated NaHCO3, water, brine and dried over Na2SO4. Solvent was removed in vacuo and the residue was purified by column chromatography (silica gel, acetone: hexanes) to give 0.3 g (80%) of 5. 1HNMR (CDCL3, 300 mHz) δ 0.89 (t, J=6.6 Hz, 3H), 1.29 (m. 24H), 2.03 (M, 4H), 2.82 (m, 4H), 3.57-3.68 (m, 1H), 3.84-3.96 (m, 1H), 4.1-4.2 (m, 2H), 4.25-4.45 (m, 5H), 4.57 (t, J=5.1 Hz, 0.5H), 4.88-5.0 (dt, J=4.8 Hz, 0.5H), 5.35 (m, 2H); ESIMS m/z 549.5 (M++23).
2-Heptadec-8-enyl-(1,3)dioxolan-4yl methyl dipotassium phosphate (6). A solution of 5 (0.5 g, 0.095 mmol) in methanolic KOH (1M, 2 ml) was stirred for 3 hours, concentrated in vacuo, and the residue was dissolved in water and passed through a sep-pak syringe cartridge (C18). Eluted with methanol, fractions containing product were pooled and solvent was removed in vacuo to give 6 (0.045 g, 85%) as amorphous powder. 1HNMR (CD3OD, 300 MHz) δ 0.86 (m, 3H), 1.27 (m, 24H), 2.0 (m, 4H), 3.77-4.0 (m, 3H), 4.14-4.21 (m, 1H), 4.26-4.35 (m, 1H), 4.5-4.66 (m, 1H), 4.89 (t, J=4.8 Hz, 0.5H), 5.0 (t, J=4.8 Hz, 0.5H), 5.35 (, 2H); ESIMS m/z 419.5 (M+−1).
Bis-(2-cyano-ethyl)-2-heptadec-8-enyl-(1,3)dioxolan-4yl methyl thiophosphate (7). To a solution of alcohol, 4 (0.274 g, 0.80 mmol) in dichloromethane (15 ml), 1H-tetrazole (0.17 g, 2.4 mmol) was added. After 10 minutes, to this solution biscyano ethyl diisopropyl phospharamidite (0.44 g, 1.66 mmol) was added and stirred for 30 minutes. Sulfur powder (0.076 g, 2.4 mmol) was added to the reaction mixture and refluxed for 2 h. The mixture was then cooled to RT, concentrated in vacuo and the residue was purified by column chromatography (silica gel, acetone:hexanes) to give 0.32 g (73%) of 7. 1HNMR (CDCl3, 300 MHz) δ 0.86 (t, J=6.9 Hz, 3H), 1.29 (brs, 24H), 2.03 (m, 4H), 2.69-2.86 (m, 4H), 3.6-4.0 (m, 4H), 4.1-4.4 (m, 5H), 4.56 (t, J=4.8 Hz, 0.35H), 4.88-5.0 (dt, J=4.5 Hz, 0.65H), 5.35 (m, 2H).
2-Heptadec-8-enyl-(1,3)dioxolan-4yl methyldipotassium thiophosphate (8). Compound 8 was prepared by the same procedure as that of 6 as amorphous powder (0.058 g, 88%). 1HNMR (D2O, 300 MHz), δ 0.78 (brs, 3H), 1.19 (brs, 24H), 1.93 (m, 4H), 3.76-3.85 (m, 3H), 4.05-4.25 (m, 2H), 4.4-4.47 (m, 0.4H), 4.75 (m, 0.75H) 4.97 (m, 0.3H), 5.25 (m, 2H).
This application claims the benefit of priority of earlier-filed U.S. provisional patent applications No. 60/678,498 filed May 6, 2005 and 60/705,556 filed Aug. 4, 2005.
The present invention was made, at least in part, with funding received from the National Institutes of Health under grants CA92160 and HL61469, and from the Department of Defense under grant DMAD 17-01-0830. The U.S. government may retain certain rights in this invention.
| Number | Date | Country | |
|---|---|---|---|
| 60678498 | May 2005 | US | |
| 60705556 | Aug 2005 | US |
