4-[4-(4-Fluorophenyl)-2-methyl-5-oxo-2,5-dihydroisoxazol-3-yl]-1-methylpyridinium iodide–4-[3-(4-fluorophenyl)-2-methyl-5-oxo-2,5-dihydroisoxazol-4-yl]-1-methylpyridinium iodide (0.6/0.4)

The crystal structure of the title compound, C16H16FN2O2 +·I−, was determined as part of a study of the biological activity of isoxazolone derivatives as p38 mitogen-activated protein kinase (MAPK) inhibitors. The X-ray crystal structure of 4-[4-(4-fluorophenyl)-2-methyl-5-oxo-2,5-dihydroisoxazol-3-yl]-1-methylpyridinium iodide showed the presence of the regioisomer 4-[3-(4-fluorophenyl)-2-methyl-5-oxo-2,5-dihydroisoxazol-4-yl]-1-methylpyridinium iodide. The synthesis of the former compound was achieved by reacting 4-(4-fluorophenyl)-3-(4-pyridyl)isoxazol-5(2H)-one after treatment with Et3N in dimethylformamide, with iodomethane. The unexpected formation of the regioisomer could be explained by a rearrangement occurring via aziridine of the isoxazolone compound. The regioisomers have site occupancies of 0.632 (4)/0.368 (4). The two six members rings make a dihedral angle of 66.8 (2)°.


Comment
Compound (II) (Scheme 1) was prepared in the course of our study on isoxazolones derivatives bearing the typical vicinal 4-pyridyl and 4-fluorophenyl pharmacophores of MAP Kinase inhibitors. Isoxazolones are described in the literature as inhibitors for p38 MAP Kinase (Laughlin et al., 2005;Clark et al., 2002).
The prototypical pyridinylimidazole SB 203580 is one of the best studied p38 inhibitors reported until now. Fig. 1 shows the most important interactions between the ATP binding sites of p38 kinase and the imidazole inhibitor SB203580 (Wang et al., 1998). The 4-fluorophenyl ring of SB203580 occupies a hydrophobic back pocket gaining selectivity. Vicinal to this interaction site, 4-pyridinyl moiety forms a hydrogen bond from the backbone NH group of Met 109 of p38 MAP Kinase ( Fig. 1).
However, certain liver toxicities, such as increased liver size and increased cytochrome P450 induction, have been reported (Foster et al., 2000;Adams et al., 1998). In light of this potential toxicity and the risks associated with developing human drugs, a continuing need exist for potent new small molecules inhibitors of cytokine production with improved pharmacokinetic and safety profiles.
Replacement of the core heterocycle representing a strategy to dissect inhibition of p38 from interferences with cytochrome P450 (CYP450).
Accordingly, and based on the research published by Laughlin and co-authors (Laughlin et al., 2005), we plane to prepare N-alkylated derivatives of compound (I) in order to get more accurate and comparable information about isoxazolones as p38 MAP Kinase inhibitors in terms of biological activity.
By testing compounds (I) and (II) in the in vitro p38-alpha MAPK assay (Laufer et al., 2005), only compound (I) was found to posses biological activity.
The loss of the biological activity of compound (II) can be attributed to the absence of hydrogen bond donor on the pyridine ring and, consequentely, impossibility of interaction with Met109.

Experimental
For the synthesis of 2-(4-Fluoro-phenyl)-3-oxo-3-pyridin-4-yl-propionic acid ethyl ester (see scheme 1), to a suspension of 3.3 g (26.8 mmol) of isonicotinic acid in 15 ml of DMF, 7.3 g (45 mmol) of CDI were added. The reaction mixture was stirred at 298 K for 1 h. The limpid solution was then cooled at 273 K and 5 g (27.4 mmol) of (4-Fluoro-phenyl)-acetic acid ethyl and 1.7 g (70.8 mmol) of NaH were added. The reaction mixture was stirred at 273 K for 15 min, then the supplementary materials sup-2 temperature was raised to 298 K and kept under vigorous stirring for 4 h. The reaction was then poured into water/ice, the pH adjusted to value 6 and extracted with ethylacetate. The combined organic layers were then collected, dried over Na 2 SO 4 and concentrated under vacuum affording an oil that was chromatographed over SiO 2 using acetone as eluent y ielding 75% of 2-(4-Fluoro-phenyl)-3-oxo-3-pyridin-4-yl-propionic acid ethyl ester. For the synthesis of (I), a suspension of 5.2 g (18.1 mmol) of 2-(4-Fluoro-phenyl)-3-oxo-3-pyridin −4-yl-propionic acid ethyl ester and 1.41 g (20.27 mmol) of hydroxylamine hydrochloride in 1.5 ml of H 2 O was warmed to 353 K. 8 ml of MeOH were added and the resulting solution allowed to reflux 4 h. The reaction mixture was then cooled to 298 K and stored at 277 K overnight whereupon a yellow solide precipitated, yielding 75% of (I). For the synthesis of (II) and (III), a suspension of 620 mg (2.5 mmol) of (I) in 1 ml of DMF was added of 0.620 ml (4.5 mmol) of Et3N and refluxed for 2 h. The reaction mixture was then cooled at 298 K, added of 0.231 ml (3.75 mmol) of iodomethane and stirred at 298 K for 2 h. The reaction mixture was then added of ethylacetate and the resulting precipitate separated by filtration and then crystalized from MeOH yielding 54% of (II) and (III).

Refinement
Hydrogen atoms attached to carbons were placed at calculated positions with C-H = 0.95 A% (aromatic) or 0.99-1.00 Å (sp 3 C-atom). All H atoms were refined with isotropic displacement parameters (set at 1.2-1.5 times of the U eq of the parent atom). The regioisomers (II) and (III) have s.o.f.s of 0.632 (4)/0.368 (4). The coordinates and a.d.p.'s of the disorderd C, N and F atoms were constrained to be equal to achieve a good convergence of the refinement procedure. Fig. 1. Schematic drawing of important interactions between the prototypical pyridin-4-yl imidazole inhibitor SB 203580 and the ATP binding site of p38. Fig. 2. Schematic drawings of 4-[4-(4-Fluoro-phenyl)-2-methyl-5-oxo-2,5-dihydro-isoxazol-3-yl]-1-methyl-pyridinium iodide, (II), and 4-[3-(4-Fluoro-phenyl)-2-methyl-5-oxo-2,5-dihydro-isoxazol-3-yl]-1-methyl-pyridinium iodide, (III).   Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.