Convergent and enantioselective syntheses of cytosolic phospholipase A 2 α inhibiting N -(1-indazol-1-ylpropan-2-yl)carbamates †

Cytosolic phospholipase A 2 α (cPLA 2 α ) is an important enzyme of the in ﬂ ammation cascade. Therefore, inhibitors of cPLA 2 α are assumed to be promising drug candidates for the treatment of in ﬂ ammatory dis-orders. Recently we have found that indole-5-carboxylic acid with a 3-(4-octylphenoxy)-2-(phenoxycar-bonylamino)propyl substituent in position 1 is an inhibitor of cPLA 2 α . We have now synthesized a corresponding derivative with the indole heterocycle replaced by an indazole ( 4 ) employing an analogous reaction sequence as for the synthesis of the indole derivative. Besides, a more convergent synthesis for 4 was established using an aziridine as central intermediate. Furthermore, a chiral-pool based enantioselective synthesis was developed for the synthesis of ( R )- and ( S )- 4 . Starting compound for both enantiomers was the ( R )-serine derived oxazolidine ( R )- 25 . Compound 4 proved to be a moderate inhibitor of cPLA 2 α , with the S -enantiomer being twice as active as the R -enantiomer. The racemate 4 and the enantiomers ( R )- and ( S )- 4 showed a high in vitro metabolic stability in rat liver S9 fractions.


Introduction
Cytosolic phospholipase A 2 α (cPLA 2 α) is an esterase that selectively cleaves the sn-2 position of arachidonoyl-glycerophospholipids of biomembranes to generate free arachidonic acid and lysophospholipids. 1,2 Subsequent metabolism of these products leads to a variety of inflammatory mediators including prostaglandins, leukotrienes and platelet activating factor (PAF). Mice with cPLA 2 α deficiency display a reduced eicosanoid production and are resistant to disease in a variety of models of inflammation. 3 Therefore, cPLA 2 α is considered as a target for the treatment of inflammatory diseases, such as rheumatoid arthritis, atopic dermatitis and Alzheimer's disease. [4][5][6] Today several potent inhibitors of cPLA 2 α are known, 7,8 which show activity in diverse animal models of inflammation after systemic or local application. However, none of these agents is actually in clinical development.
We have found that certain 1-(indol-1-yl)propan-2-ones, such as compound 1 (Fig. 1), inhibit cPLA 2 α with high potency. 9 The latter compound shows structural similarities to the propan-2-one inhibitor ARC-70484XX developed by Astra-Zeneca. 10 An important part of the pharmacophore of these substances is the activated electrophilic ketone moiety present in the middle part of the molecules. This structural element is supposed to form reversible covalent binding interactions with a serine residue of the active site of cPLA 2 α.
Recent studies have shown that 1-(indol-1-yl)propan-2-ones are extensively metabolized in vitro as well as in vivo. 11,12 Especially, the activated ketone is reduced to an alcohol resulting in compounds, which do not inhibit cPLA 2 α any more. Therefore, we have replaced the ketone by metabolically more stable polar moieties such as acyloxy, acylamino, urea and carbamate. 13 These variations led to a more or less pronounced drop of inhibitory potency. One of the most active compounds of the investigated substances was the carbamate substituted indole-5-carboxylic acid 2, which possessed an IC 50 against cPLA 2 α in the micromolar range.
Because structure-activity relationship studies on the 1-(indol-1-yl)propan-2-ones have revealed that replacement of the indole scaffold of 1 by an indazole (3) led to an about fivefold increase of activity, 14 we wanted to synthesize and evaluate the corresponding carbamate-substituted indazole derivative 4.
In contrast to the ketones 1 and 3, the carbamates 2 and 4 possess a stereogenic centre resulting in R-and S-configurated products. A further aim of this study, therefore, was to develop enantioselective syntheses for the enantiomers of 4 and to determine the influence of the configuration of this compound on cPLA 2 α inhibition.

Results and discussion
Chemistry For the preparation of the indazole 4 the synthetic route developed for the synthesis of the corresponding indole 2 13 was used (Scheme 1). Starting material was indazole-5-carboxylic acid (5), which first was converted into its tert-butylester 8 employing a reaction sequence published for the synthesis of tert-butyl benzotriazole-5-carboxylate. 15 Treatment of 8 with epichlorohydrin in presence of KOH and tetrabutylammonium bromide led to 9, which was reacted with 4-octylphenol to obtain the secondary alcohol 10. The alcohol functionality of this compound was converted to a mesylate (11) by reaction with methanesulfonyl chloride. From the latter intermediate the azide 12 was obtained by treatment with trimethylsilyl azide and tetrabutylammonium fluoride. Catalytic hydrogenation of the azide group of 12 with Pd on charcoal led to the amine-substituted compound 13. Reaction of this with phenyl chloroformate in presence of an amine base followed by hydrolysis of the tert-butyl ester with trifluoroacetic acid gave the desired target compound 4.
In Scheme 2, an alternative approach for the synthesis of 4 is outlined, which is more convergent in respect of the variation of the octylphenoxy-part of 4 during the course of structure-activity studies. This synthetic route started from indazole-5-carboxylic acid, which was converted into its allyl ester (15) by reaction with allyl bromide in presence of K 2 CO 3 . Using Cs 2 CO 3 as base, the indazole ester 15 was alkylated with epichlorohydrin in position 1 yielding the oxiranylmethylindazole derivative 16. The oxiranyl ring of 16 was opened to a azidoalcohol by reaction with sodium azide. 16 Then the azido group of 17 was transformed to an amine (18) by treatment with triphenylphosphine, which in turn was protected with BOC using di-tert-butyl dicarbonate. The alcohol group of obtained compound 19 was reacted with tosyl chloride in presence of 4-dimethylaminopyridine to give the tosylate 20. Cyclization to the aziridine 21 was accomplished by treatment of 20 with powdered KOH and tetrabutylammonium hydrogensulfate in CH 2 Cl 2 . Ring opening of the BOC-protected aziridine 21 with 4-octylphenol was achieved in methylene chloride with catalytical amounts of BF 3 -etherate. 17 The BOC group of 22 was removed by trifluoroacetic acid. Reaction of the generated amine 23 with phenyl chloroformate led to the carbamate 24. Finally, the allyl ester of the indazole heterocycle of 24 was cleaved with tetrakis(triphenylphosphine)palladium(0) to afford the desired target compound 4. By this way, derivatives with varying substituents at the phenoxy-residue can be obtained from a common intermediate (here 21) in only four steps instead of six necessary in the synthesis described above (Scheme 1).
Due to the chiral centre present in 4, this compound exists as two enantiomers. For the determination of their eudismic ratio, additionally an enantioselective, chiral-pool based synthetic approach was established. The synthesis of the R-configurated enantiomer of 4 started from the commercially availably, (R)-serine derived oxazolidine (R)-25, 18 which was converted to the tosylate (R)-27 as previously described (Scheme 3). 19,20 This central intermediate was reacted with 4-octylphenol to yield the phenol ether (S)-29. The acetone protecting group was removed by reaction with p-toluenesulfonic acid and the alcohol moiety of (R)-30 was converted to a tosyl ester with p-toluenesulfonyl chloride. Treatment of obtained compound (R)-31 with allyl indazole-5-carboxylate in DMF in presence of NaH afforded the BOC protected 1-(2-aminopropyl)indazole derivative (R)-22. After deprotection of the BOC group, the amine moiety of resulting compound (R)-23 was reacted with phenyl chloroformate in presence of an amine base. Subsequent cleavage of the allyl ester group of resulting compound (R)-24 using Pd(0) catalysis led to the desired target compound (R)-4.
The synthesis of the S-enantiomer of 4 is outlined in Scheme 4. With (R)-27, the identical central chiral intermediate was used as for the synthesis of the R-enantiomer. To get the antipodal configuration in the target compound, here the indazole heterocycle and not the 4-octylphenoxy residue was introduced at first into the molecule. After removing the acetone protection group of obtained compound (S)-32, the free alcohol moiety was esterified with tosyl chloride and the tosylate group substituted by a 4-octylphenoxy residue to afford (S)-22. From this compound (S)-4 was synthesized employing the same procedure as for the synthesis of (R)-4 from (R)-22. Chiral HPLC on a Chiralpak® IA column revealed that (R)-4 and (S)-4 were formed with 93% ee and 100% ee, respectively (for chromatograms see ESI †).
As mentioned above, a drawback of the indazolylpropan-2-one 3 is its metabolic liability. In in vitro experiments it could be shown that the carbamate 4 is much more stable against metabolism by rat liver homogenate than 3. After incubation of 3 in presence of the co-factor NADPH only 8% of the parent compound could be recovered, under the same conditions still 80-90% of 4, (S)-4 and (R)-4 were present in the incubation mixture.

Conclusions
Taken together, we have developed new effective synthetic approaches for the synthesis of 4 as well as for its enantiomers (S)-4 and (R)-4. The biological evaluation of these compounds revealed that the increase of metabolic stability achieved by replacement of the ketone function of 3 by a carbamate moiety is accompanied by a drastic loss of cPLA 2 α inhibitory potency. The aim of further studies will be to improve cPLA 2 α inhibitory potency of 4 by structural variations.

Experimental section
Chemistry General. Column chromatography was performed on silica gel 60, particle size 0.040-0.063 mm, from Macherey & Nagel. Melting points were determined on a Büchi B-540 apparatus and are uncorrected. 1 H NMR spectra were recorded on a Varian Mercury Plus 400 spectrometer (400 MHz), a Varian Unity Plus 600 spectrometer (600 MHz) or an Agilent VNMRS-600 spectrometer (600 MHz). 13 C NMR spectra were measured on a Varian Mercury Plus 400 spectrometer (101 MHz) or an Agilent VNMRS-600 spectrometer (151 MHz). Electron ionization (EI) mass spectra were obtained on a Finnigan GCQ apparatus. The high resolution mass spectra (HRMS) were recorded on a Bruker micrOTOF-Q II spectrometer using electro spray chemical ionization (ESI) or atmospheric pressure chemical ionization (APCI). Preparative reversed phase HPLC was performed using a Knauer pump P2.1L and a Shimazu SPD-6A UV detector. Chromatograms were recorded with MacDAcq32 Control Software from Bischoff. As stationary phase a Knauer RP18 Eurospher II 5 µm column (20 mm (I.D.) × 250 mm) with a RP18 Eurospher II 5 µm guard column (20 mm (I.D.) × 30 mm) was used. The mobile phase consisted of acetonitrile-water-formic acid (80 : 20 : 0.1, v/v/v). The flow rate was 25 mL min −1 . The compounds were dissolved in DMSO and the injected sample volume was 0.5-1 mL. The detection wavelength was set to 254 nm. The substances were obtained as solids after distilling off the organic solvent and freeze-drying the remaining aqueous phase using a Christ alpha 1-2 LD plus apparatus. Chiral HPLC-analysis was performed on a Daicel Chiralpak® IA 5 µm column (4.6 mm (I.D.) × 250 mm) using isohexane-methanol-isopropanol-formic acid (90 : 8.75 : 1.25 : 0.1, v/v/v/v) as mobile phase at a flow rate of 1 mL min −1 . The detection wavelength was 254 nm.