Hit-to-lead optimization and discovery of a potent, and orally bioavailable G protein coupled receptor kinase 2 (GRK2) inhibitor.

G-protein coupled receptor kinase 2 (GRK2), which is upregulated in the failing heart, appears to play a critical role in heart failure (HF) progression in part because enhanced GRK2 activity promotes dysfunction of ß-adrenergic signaling and myocyte death. An orally bioavailable GRK2 inhibitor could offer unique therapeutic outcomes that cannot be attained by current heart failure treatments that directly target GPCRs or angiotensin-converting enzyme. Herein, we describe the discovery of a potent, selective, and orally bioavailable GRK2 inhibitor, 8h, through high-throughput screening, hit-to-lead optimization, structure-based design, molecular modelling, synthesis, and biological evaluation. In the cellular target engagement assays, 8h enhances isoproterenol-mediated cyclic adenosine 3',5'-monophosphate (cAMP) production in HEK293 cells overexpressing GRK2. Compound 8h was further evaluated in a human stem cell-derived cardiomyocyte (HSC-CM) contractility assay and potentiated isoproterenol-induced beating rate in HSC-CMs.

Enhancement of kinase selectivity in a potent class of arylamide FMS inhibitors.

Structure-activity relationship (SAR) studies on a highly potent series of arylamide FMS inhibitors were carried out with the aim of improving FMS kinase selectivity, particularly over KIT. Potent compound 17r (FMS IC50 0.7 nM, FMS cell IC50 6.1 nM) was discovered that had good PK properties and a greater than fivefold improvement in selectivity for FMS over KIT kinase in a cellular assay relative to the previously reported clinical candidate 4. This improved selectivity was manifested in vivo by no observed decrease in circulating reticulocytes, a measure of bone safety, at the highest studied dose. Compound 17r was highly active in a mouse pharmacodynamic model and demonstrated disease-modifying effects in a dose-dependent manner in a strep cell wall-induced arthritis model of rheumatoid arthritis in rats.

Optimization of a potent class of arylamide colony-stimulating factor-1 receptor inhibitors leading to anti-inflammatory clinical candidate 4-cyano-N-[2-(1-cyclohexen-1-yl)-4-[1-[(dimethylamino)acetyl]-4-piperidinyl]phenyl]-1H-imidazole-2-carboxamide (JNJ-28312141).

A class of potent inhibitors of colony-stimulating factor-1 receptor (CSF-1R or FMS), as exemplified by 8 and 21, was optimized to improve pharmacokinetic and pharmacodynamic properties and potential toxicological liabilities. Early stage absorption, distribution, metabolism, and excretion assays were employed to ensure the incorporation of druglike properties resulting in the selection of several compounds with good activity in a pharmacodynamic screening assay in mice. Further investigation, utilizing the type II collagen-induced arthritis model in mice, culminated in the selection of anti-inflammatory development candidate JNJ-28312141 (23, FMS IC(50) = 0.69 nM, cell assay IC(50) = 2.6 nM). Compound 23 also demonstrated efficacy in rat adjuvant and streptococcal cell wall-induced models of arthritis and has entered phase I clinical trials.

Reducing ion channel activity in a series of 4-heterocyclic arylamide FMS inhibitors.

During efforts to improve the bioavailability of FMS kinase inhibitors 1 and 2, a series of saturated and aromatic 4-heterocycles of reduced basicity were prepared and evaluated in an attempt to also improve the cardiovascular safety profile over lead arylamide 1, which possessed ion channel activity. The resultant compounds retained excellent potency and exhibited diminished ion channel activity.

Pyrido[2,3-d]pyrimidin-5-ones: a novel class of antiinflammatory macrophage colony-stimulating factor-1 receptor inhibitors.

A series of pyrido[2,3-d]pyrimidin-5-ones has been synthesized and evaluated as inhibitors of the kinase domain of macrophage colony-stimulating factor-1 receptor (FMS). FMS inhibitors may be useful in treating rheumatoid arthritis and other chronic inflammatory diseases. Structure-based optimization of the lead amide analogue 10 led to hydroxamate analogue 37, which possessed excellent potency and an improved pharmacokinetic profile. During the chronic phase of streptococcal cell wall-induced arthritis in rats, compound 37 (10, 3, and 1 mg/kg) was highly effective at reversing established joint swelling. In an adjuvant-induced arthritis model in rats, 37 prevented joint swelling partially at 10 mg/kg. In this model, osteoclastogenesis and bone erosion were prevented by low doses (1 or 0.33 mg/kg) that had minimal impact on inflammation. These data underscore the potential of FMS inhibitors to prevent erosions and reduce symptoms in rheumatoid arthritis.

Structure-based optimization of a potent class of arylamide FMS inhibitors.

An anti-inflammatory 1,2,4-phenylenetriamine-containing series of FMS inhibitors with a potential to form reactive metabolites was transformed into a series with equivalent potency by incorporation of carbon-based replacement groups. Structure-based modeling provided the framework to efficiently effect this transformation and restore potencies to previous levels. This optimization removed a risk factor for potential idiosyncratic drug reactions.

Design and synthesis of a pyrido[2,3-d]pyrimidin-5-one class of anti-inflammatory FMS inhibitors.

A series of pyrimidinopyridones has been designed, synthesized and shown to be potent and selective inhibitors of the FMS tyrosine kinase. Introduction of an amide substituent at the 6-position of the pyridone core resulted in a significant potency increase. Compound 24 effectively inhibited in vivo LPS-induced TNF in mice greater than 80%.

Discovery of novel FMS kinase inhibitors as anti-inflammatory agents.

The optimization of the arylamide lead 2 resulted in identification of a highly potent series of 2,4-disubstituted arylamides. Compound 8 (FMS kinase IC(50)=0.0008 microM) served as a proof-of-concept candidate in a collagen-induced model of arthritis in mice.

Synthesis and evaluation of novel 3,4,6-substituted 2-quinolones as FMS kinase inhibitors.

A series of 3,4,6-substituted 2-quinolones has been synthesized and evaluated as inhibitors of the kinase domain of macrophage colony-stimulating factor-1 receptor (FMS). The fully optimized compound, 4-(4-ethyl-phenyl)-3-(2-methyl-3H-imidazol-4-yl)-2-quinolone-6-carbonitrile 21b, has an IC(50) of 2.5 nM in an in vitro assay and 5.0 nM in a bone marrow-derived macrophage cellular assay. Inhibition of FMS signaling in vivo was also demonstrated in a mouse pharmacodynamic model.

Cyclic nucleotide phosphodiesterase type 4 inhibitors: evaluation of pyrazolo[1,5-a]-1,3,5-triazine ring system as an adenine bioisostere.

A series of 8-substituted pyrazolo[1,5-a]-1,3,5-triazines were considered as a bioisosteric replacement for the 9-substituted adenine derivatives resulting in the discovery of 8-(2-methoxybenzyl)-4-(N-methylamino)-2-n-propylpyrazolo[1,5-a]-1,3,5-triazine (14d) and 2-trifluoromethyl-8-(2-methoxybenzyl)-4-(N-methylamino)pyrazolo[1,5-a]-1,3,5-triazine (14e) as a new structural class of potent phosphodiesterase type 4 inhibitors (IC(50)=13 nM and 11 nM, respectively) with high isoenzyme selectivity. An original tandem of reactions involving a palladium-mediated cross-coupling reaction (PMCCR) of the readily available 8-iodo-2-methyl-4-(N-methyl-N-phenylamino)pyrazolo[1,5-a]-1,3,5-triazine (11a) and arylboronic acids or alkynes followed by the displacement of the N-methyl-N-phenylamino group constitute the key steps in a novel synthetic approach developed herein. The treatment of 11a-c with n-BuLi and selected aldehydes represents an interesting alternative to the PMCCR for the synthesis of benzylic derivatives 14a-i. Preliminary biological testing has shown that compounds 14d and 14e strongly inhibit LPS-induced TNFalpha release from human mononuclear cells from healthy subjects. These two compounds were selected for further biological evaluation.

Identification of novel short chain 4-substituted indoles as potent alphavbeta3 antagonist using structure-based drug design.

The vitronectin receptor alpha(v)beta(3) has been identified as a promising potential target for the treatment of osteoporosis, diabetic retinopathy and cancer. We have recently reported 5-substituted indoles 3-[5-[2-(5,6,7,8-tetrahydro[1,8]naphthyridin-2-yl)ethoxy]indol-1-yl]-3-(3-pyridyl)propionic acid 3 and 3-[5-[2-(5,6,7,8-tetrahydro[1,8]naphthyridin-2-yl)ethoxy]indol-1-yl]-3-(3,4-methylenedioxyphenyl)propionic acid 4, as an original series of potent alpha(v)beta(3) antagonists with subnanomolar activity. Ligand-protein docking analyses have been performed to generate binding models of three different chemical classes of known alpha(v)beta(3) antagonists with alpha(v)beta(3). Results of this docking study suggested that indoles bearing the basic tetrahydronaphthyridine group at position 4 can easily adopt the correct binding conformation and should be as potent as our current 5-substituted indole leads 3 and 4. This hypothesis was nicely demonstrated by the synthesis of a series of 1,4-disubstituted indoles through a tandem of reactions involving: (i) the N-alkylation of indoles 15 and 22 with propargyl esters and cesium fluoride, and (ii) a Heck coupling reaction between 4-bromoindole and 7-vinyl-3,4-dihydro-2H-[1,8]naphthyridine-1-carboxylic acid tert-butyl ester 12, or (iii) a reductive amination involving the N-substituted-4-aminoindole 23 and the BOC-protected tetrahydro[1,8]naphthyridine aldehyde 13. Among the compounds assayed, 3-(3-pyridyl)-3-[4-[2-(5,6,7,8-tetrahydro[1,8]naphthyridin-2-yl)ethyl]indol-1-yl]propionic acid 21 showed the most promising activity on alpha(v)beta(3) (IC(50)=0.5 nM), and was found to have the same potency as our current leads 3 and 4, while maintaining selectivity over alpha(IIb)beta(IIIa). Moreover, based on the reasonable apparent permeability coefficient in an in vitro CACO-2 cell monolayer assay (P(app) apical/basolateral=2.2 x 10(-6)cm/s, P(app) basolateral/apical=2.5 x 10(-6)cm/s), compound 21 is expected to be absorbed through the intestine in human. Thus, 1,4-disubstituted indole 21 represents a new lead for this novel class of conformationally restricted alpha(v)beta(3) antagonists. Additionally, this study validates the pharmacophore model previously postulated and provides an improved basis for further structure-based drug design in the field of alpha(v)beta(3).

Effect of construct design on MAPKAP kinase-2 activity, thermodynamic stability and ligand-binding affinity.

MAPK-activated protein kinase-2 (MAPKAPK2) regulates the synthesis of tumor necrosis factor and other cytokines and is a potential drug target for inflammatory diseases. Five protein constructs were produced in 4-10mg quantities per liter of culture media using baculovirus-infected insect cells and characterized for kinase activity, thermal stability, and ligand-binding affinity. Compared to construct 1-370, removal of the C-terminal autoinhibitory peptide in 1-338 resulted in a destabilized but partially active nonphosphorylated enzyme; phosphorylation of 1-338 by p38alpha further increased activity 12-fold. A putative constitutively active mutant, 1-370/T222E/T334E, was 6.3-fold less active than phosphorylated 1-370. ThermoFluor, an equilibrium ligand-binding assay, was used to measure nucleotide analogue affinity for various constructs. Binding of phosphorylated nucleotides was Mg(2+)-dependent. Residues 1-40 were required for high-affinity binding of ADP, ATPgammaS, staurosporine, and K252a. A mutation M138A rendered 1-370 susceptible to p38-inhibitors SB-203580 and SB-202190 with IC50 values of 17.4 and 14.1 microM, respectively. Taken together, these studies provide information on the mechanism of ligand-binding to MAPKAPK2 that can be used in the search for selective small-molecule inhibitors.

Discovery, SAR, and X-ray structure of novel biaryl-based dipeptidyl peptidase IV inhibitors.

The discovery, SAR, and X-ray crystal structure of novel biarylaminoacyl-(S)-2-cyano-pyrrolidines and biarylaminoacylthiazolidines as potent inhibitors of dipeptidyl peptidase IV (DPP IV) are reported.

Comparison of automated docking programs as virtual screening tools.

The performance of several commercially available docking programs is compared in the context of virtual screening. Five different protein targets are used, each with several known ligands. The simulated screening deck comprised 1000 molecules from a cleansed version of the MDL drug data report and 49 known ligands. For many of the known ligands, crystal structures of the relevant protein-ligand complexes were available. We attempted to run experiments with each docking method that were as similar as possible. For a given docking method, hit rates were improved versus what would be expected for random selection for most protein targets. However, the ability to prioritize known ligands on the basis of docking poses that resemble known crystal structures is both method- and target-dependent.