A Convenient Route to 4-Carboxy-4-Anilidopiperidine Esters and Acids

The route selection and development of a convenient synthesis of 4-carboxy-4-anilidopiperidines is described. Previous routes were hampered by the low yield of the target esters as well as the inability to convert the esters to the required free acids. Considerations for large-scale production led to a modified synthesis that utilised a tert-butyl ester of 4-carboxy-4-anilidopiperidines which resulted in a dramatic increase in the overall yield of the target N-propionylated- 4-anilidopiperidine-4-carboxylic acids and their corresponding methyl esters. These compounds are now available for use as precursors and reference standards, of particular value for the production of 11C and 18F-labelled 4-carboxy-4-anilidopiperidine radiotracers.

The original synthesis of 6a is based on the preparation of -phenylamino nitrile (2a) from 1-(2-phenylethyl)-4-piperidone (1a), aniline and KCN in a Strecker-addition (route A: 1a2a3a4a5a6a, Scheme 2 and Scheme 3). Nitrile hydrolysis yields carboxamide 3a, which is finally reacted with KOH in 1,2-ethanediol at 190 °C to yield the free acid 4a. Conversion to the methyl ester 5a followed by acylation of 5a with propionic anhydride results in 6a. A limitation of this method is the very low overall yield (1.2% 3) of 6a, mainly caused by low conversion of the nitrile (2a) to the corresponding amide (3a, 14% 4, 3% 13). Moreover, we found the modified reaction pathway 4 for the synthesis of 6a not to provide any improvement over the original procedure (route B: 1b2b3b4b5b6b6c6a, 1.2% 15, Scheme 2 and Scheme 3) in contrast to the corresponding yield of 11% in the original report 4. Furthermore, the preparation of 6g according to procedures reported in the literature (route C: 1a2a3a4a5c6i6g, Scheme 1 and Scheme 2) 4,8,12,13 also resulted in only a very low overall yield (0.5%). Finally, we identified more recently developed methods for the synthesis of 6a, 6g and 6h [13,16] to be applicable only for reactions on the milligram scale.

Results and Discussion
The need for gram amounts of pure carfentanil acid for use as precursor in radiolabelling, as well as for the corresponding authentic reference compounds of the radiotracers in question, prompted us to address the development of an improved method of preparing desmethyl carfentanil free acid (6g), desmethyl carfentanil sodium salt (6h) and carfentanil (6a) itself. The literature procedures and the identified improved synthetic sequences are summarized in Scheme 2 and Scheme 3.

Scheme 2.
Preparation of 4-phenylamino-1-substituted-4-piperidine carboxylic acid derivatives. For the synthesis of 6a, 6g-h, compound 4b was used as a key intermediate. 4b was prepared 4,15 in a good overall yield (33%). An alternative method 17 for the preparation of 4b was based on the reaction of -aminonitrile 2b with CSI, followed by cyclization of the resulting amide (7) by treatment with 1 M HCl to yield a 1-phenyl-spirohydantoin (8) derivate. Alkaline hydrolysis 18 of the 2,4-imidazilidinedione derivative yielded -amino acid 4b in an overall yield of 39%.
For our new synthesis route (route D: 4b5d6d6e6f6g6h and 6g6a, Scheme 2 and Scheme 3) a tert-butyl group was chosen for protecting the carboxylic acid function of 4b, ultimately providing the new compounds 5d, 6d-f. This protecting group has several advantages: The introduction of tert-butyl goup in 4b is readily performed, and the final cleavage of the tert-butyl ester, subsequent to the required transformations, can be performed under mild conditions. Scheme 3. Synthesis of 4-carboxy-4-anilidopiperidine derivatives.  For the introduction of the tert-butyl group, 4b was reacted with N,N-dimethylformamide di-tert-butyl acetal 19 or, alternatively, with tert-butyl 2,2,2-trichloroacetamidate 20 to yield the tert-butyl ester 5d (71%/43%). N-propionlyation of 5d was initially attempted by refluxing the amine in neat propionic anhydride, but this procedure led to the removal of the tert-butyl group. In contrast propionyl chloride in the presence of Hünig-base yielded 6d in 60% yield. Hydrogenolysis followed by N-alkylation led to the new tert-butylester of carfentanil (6f). Deprotection of 6f with neat TFA at ambient temperature afforded the target compound 6g. Overall yields from 4b were: 6g (16.8%); 6h (13.8%); 6a (13.2%). Overall yields starting from 1b were as follows: 6g (5.54%); 6h (4.55%); 6a 4.35% which compare favourably to that of the literature procedures: 0.4-0.5%.

General
Starting materials and reagents were obtained from major commercial suppliers and were used without further purifications. Melting points were measured with a Büchi-535 instrument and the reported data are uncorrected. 1 H-NMR and 13 C-NMR spectra were obtained with a Bruker 500 spectrometer, and measurements were obtained at 20 C in CDCl 3 , CD 3 OD and DMSO-d 6

Conclusions
A simple and effective synthesis of 4-carboxy-4-anilidopiperidines has been developed based on coverting 4b to the corresponding t-Bu ester for use as a key intermediate. The improved method facilitates the production of 4-carboxy-4-APs in general, and more specifically, opens a route for preparation of carboxy-4-APs for use in PET imaging.