An Improved Synthesis of 5-(2,6-Dichlorophenyl)-2-(phenylthio)-6H-pyrimido[1,6-b]pyridazin-6-one (a VX-745 analog) †

An improved procedure for the synthesis an analog of the p38 inhibitor compound VX- 745 is reported.


Introduction
Mitogen-activated protein (MAP) kinases are key enzymes involved in signal transduction and the amplification of cellular responses to stumuli.P38 MAP kinase is a specific member of the MAP kinase family associated with the onset and progression of inflammation, and several groups are working to develop p38 MAP kinase inhibitors as potential treatments for inflammatory and neurological disease.Vertex discovered the 3D structure of p38MAP kinase in 1996 and computer modeling and testing suggested the design of VX-475 [1].
For comparison purposes we needed a sample of a VX-745 analog and followed the synthesis as described in a general procedure for the preparation of p38 inhibitor compounds by Bemis and coworkers [2,3].As we had difficulties in duplicating some of the experiments and only limited physical data were mentioned by the authors, we herein report our experience with and improvements of the synthesis of this VX-745 analog.

Results and Discussion
The use of NaH, instead of NaNH 2 [2,3], for the deprotonation of dichlorophenylacetonitrile 1 and aqueous workup using satd.NH 4 Cl (instead of satd.Na 2 CO 3 ) afforded 2 as a shiny red crystalline product without chromatographic purification.The yield could be improved to 43% on a 9 g scale, whereas Bemis [2,3] reported a yield of 28% without comments on the physical state.HPLC and NMR analysis (no benzylic protons could be detected) revealed the formation of a mixture of tautomers of 2. Since 2 obtained from the modified procedure was quite insoluble in the solvents given by Bemis, DMF was used as solvent for this step.A modified aqueous workup and direct recrystallization of the crude product from EtOH allowed us to avoid column chromatography and gave the desired product 3 as orange crystals in a yield of 72%.The absence of signals of the benzylic proton and the complex peak pattern proved again the formation of at least two tautomeric species.Bemis provides no information on the physical state of 4; while in the general procedure the compounds obtained are described as oils.

Cl
Partial hydrolysis of the nitrile moity of 3 was performed using shorter reaction times than described in the literature and a modified workup, (no neutralization of the H 2 SO 4 ) to afford 4 as bright orange crystals.Bemis and coworkers again report no details for the final conversion of 4 to 5. When pure 4 was reacted with dimethylformamide dimethyl acetal the VX-745 analog 5 was formed as a yellow p recipitate, which crystallized from AcOH/water with a yield of 41% (see Scheme 1).

Conclusions
We have prepared the VX-745 analog 5 in a 4-step sequence and increased the overall yield from 3.5% to 16%.Spectral data, HPLC-and CHN-analysis proved identity as well as high purity (> 99%).

General
Melting points were measured on a Kofler hot stage apparatus. 1H-NMR-spectra were recorded on a Bruker AC-200 (200 MHz) pulse Fourier-transform NMR spectrometer in CDCl 3 , DMSO-d 6 or TFA-d 1 using tetramethylsilane as an internal standard.Thin layer chromatography (TLC) was performed on precoated plates (Merck TLC aluminum sheets silica 60 F 254 ) with detection by UV light or with phosphomolybdic acid in aqueous EtOH by heating.All reactions were magnetically stirred under an argon atmosphere.