Synthesis of 5-acyl-4-methylene-1,2,3,4-tetrahydropyridines

Several approaches to monocyclic N -protected 4-methylene-1,2,3,4-tetrahydropyridines are reported. N -Boc- 4-methylenepiperidin-3-one was found to be unstable and its enol triflate was not easily accessible. However, N -protected 2,3-dihydropyridin-4-ones are readily available by oxidation of the corresponding piperidin-4-ones and can be converted into 5-iodo-2,3-dihydropyridin-4-ones. Methylenation then provided N -Boc- and N - nosyl-5-iodo-4-methylene-1,2,3,4-tetrahydropyridines. N -Boc-5-iodo-2,3-dihydropyridin-4-one and N -Boc-5- iodo-4-methylene-1,2,3,4-tetrahydropyridine were converted into the corresponding 5-acetyl-2,3-dihydropyridin-4-one and 5-acetyl-4-methylene-1,2,3,4-tetrahydropyridine by Stille reactions. Methylenation of N -Boc-5-ethoxycarbonyl-2,3-dihydropyridin-4-one using the Petasis reagent gave the corresponding 5-ethoxycarbonyl-4-methylene-1,2,3,4-tetrahydropyridine with the 5-acetyl-4-methylenetetrahydropyridine being a side product at higher temperatures.


Scheme 1. Synthesis of the 3-hydroxymethyl-4-methylenepiperidine 16.
During the course of this work, nosyl protection for the piperidine was investigated and the 4hydroxymethyl-1,2,5,6-tetrahydropyridine 20 was prepared from the 4-methylenepiperidine 18. 20 However attempts to O-alkylate this using tributyl(iodomethyl)tin gave the 2-nitrophenyl ether 21 formed by denosylation of the starting material by the sodium salt of another molecule of starting material, see Scheme 2. Although further work may have improved the synthesis of the 3-hydroxmethyl-4-methylenepiperidine 16, it was decided instead to check out alternative approaches to the target 4-methylene-1,2,3,4tetrahydropyridines 1 and 2. The enol triflate 24 derived from the 4-methylenepiperidin-3-one 23 was identified as a possible intermediate for the synthesis of the target compounds 1.However, the attempted oxidation of the known 3-hydroxy-4-methylenepiperidine 22, 21 see Scheme 3, using several oxidizing agents (Dess-Martin periodinane, 22 PCC, PDC, TPAP, MnO2) gave complex mixtures of products.The crude reaction mixture obtained from the Dess-Martin oxidation appeared to contain a single, less polar, product, but this gave a mixture on attempted isolation suggesting that the ketone 23 may have been obtained but that it was very unstable.Indeed, the isomeric 3-methylenepiperidin-4-one 25 is known to be unstable and had to be prepared in a large excess and used immediately. 23s a third approach to 5-substituted 4-methylene-1,2,3,4-tetrahydropyridines it was decided to see whether methylenation of the corresponding 2,3-dihydropyridin-4-ones, see Figure 2, could be carried out.Several procedures are known for the oxidation of N-protected piperidin-4-ones into 2,3-dihydropyridin-4ones. 24In our hands, Shvo's palladium(II) acetate catalysed procedure 25 with allyl diethyl phosphate as the oxidant was found to be useful with the oxidation of the N-nosyl and N-Boc-piperidinones 26 and 27 giving 68% and 84% yields of the 2,3-dihydropyridinones 28 and 29, 24 respectively.The regioselective iodination of the 2,3-dihydropyridin-4-ones following the precedent set by Comins 26 was then investigated in order to provide access to 5-substituted 4-methylene-1,2,3,4-tetrahydropyridines.In the event, iodination of the Nnosyldihydropyridinone 28 using iodine monochloride gave a good yield of the 5-iodo-2,3-dihydropyridin-4one 30 but better yields of the corresponding N-Boc-dihydropyridinone 31 were obtained using iodine, 27 see Scheme 4.

11
Procedures now had to be evaluated for the conversion of 2,3-dihydropyridin-4-ones into 4-methylenetetrahydropyridines.Wittig reactions were first investigated.The conversion of the N-nosyldihydropyridinone 28 into the 4-methylenetetrahydropyridine 32 using a Wittig reaction proceeded in an acceptable yield of 68%.However, only a 30% yield of the N-Boc-4-methylenetetrahydropyridinone 17 was obtained using this procedure and methylenation of the 5-iododihydropyridinones 30 and 31 using Wittig reactions gave only low yields of the 5-iodo-4-methylene-1,2,3,4-tetrahydropyridines 33 and 34, see Scheme 4. Alternative procedures were therefore investigated for the conversions of the 5-iododihydropyridinones 30 and 31 into the 5-iodo-4-methylenetetrahydropyridines 33 and 34.Using the Petasis reagent (dimethyltitanocene) in toluene at 110 o C, [28][29][30][31] a 42% yield of the N-Boc-5-iodo-4-methylenetetrahydropyridine 34 was obtained, see Scheme 5. Microwave conditions using the Petasis reagent 32 gave a similar yield but lower yields were obtained using the Tebbe reagent [33][34][35] and the use of the Nysted reagent 36 gave complex mixtures of products.Only very low yields of the N-nosyl-5-iodo-4-methylenetetrahydropyridine 33 were obtained using these organometallic reagents.Elaboration of the vinylic iodides 31 and 34 was now investigated.A Stille reaction of the 5-iodo-2,3dihydropyridin-4-one 31 37 with the ethoxyvinylstannane 35 38 using tris(dibenzylideneacetone)dipalladium(0) as the catalyst gave a good yield of the diketone 36.However, the Stille reaction of the 5-iodo-4methylenetetrahydropyridine 34 with the ethoxyvinylstannane 35 gave only a 27% yield of the 5-acetyl-4methylenetetrahydropyridine 37 under these conditions.When bis(benzonitrile)palladium(II) chloride 39 was used as the catalyst, the yield of the acetylated tetrahydropyridine 37 increased to 40%, see Scheme 6, although the yield of the acetylated dihydropyridinone 36 dropped to 56% with this catalyst.Conversion of the iodotetrahydropyridine 34 into the corresponding Grignard reagent 40,41 and acylation with the acid chloride 38 gave the 5-acyl-4-methylenetetrahydropyridine 39 but in only a low yield, see Scheme 6.The 5-iodo-4methylenetetrahydropyridine 34 showed signs of decomposition after a month at −25 o C in benzene and this instability may account for the lowish yields of the ketones 37 and 39.Scheme 6. Preparation of the 5-acyl-4-methylene-1,2,3,4-tetrahydropyridines 37 and 39.
Finally, the keto-ester 42 was prepared via the selenide 41 from the ester 40 42,43 and its reaction with the Petasis reagent [28][29][30][31] investigated.At 110 o C, a mixture of the 5-acetyl-4-methylenetetrahydropyridine 37 and the expected 5-ethoxycarbonyl-4-methylenetetrahydropyridine 43 was obtained whereas at 65 o C only the ester 43 was isolated (40%), see Scheme 7. Perhaps the ketone 37 had been formed by hydrolysis on work-up of the enol ether formed by reaction of the Petasis reagent with the ethoxycarbonyl group.
Although this work led to the synthesis of the target compound 37 it also pointed the way for further work.For example, it should be possible to improve the synthesis of the 5-acetyl-4methylenetetrahydropyridine 37 directly by bis-methylenation of the readily available keto-ester 42.Alternatively, the previously reported elimination of HCN from the 2-cyanotetrahydropyridine 5 and the formation of the 4-methylenetetrahydropyridine 17 from the ether 14 suggest that dehydration of 4hydroxymethyltetrahydropyridines, e.g.alcohol 13, may lead to improved access to 4methylenetetrahydropyridines albeit that further work will be necessary to access more complex examples.

Experimental Section
General.Low resolution mass spectra were recorded on a Micromass Trio 200 spectrometer.High resolution mass spectra were recorded on a Kratos Concept IS spectrometer.Modes of ionisation were electron impact (EI), chemical ionisation using ammonia (CI + ), electrospray (ES) or atmospheric pressure chemical ionization (APCI).Infrared spectra were recorded on a Genesis FTIR as evaporated films on sodium chloride plates.
Proton NMR spectra ( 1 H NMR) were recorded on a Bruker Ultrashield 500 (500 MHz) or a Varian INOVA Unity 300 (300 MHz) spectrometer.Residual non-deuterated solvent was used as the internal standard.Chemical shifts (H) are quoted in parts per million (ppm) downfield from tetramethylsilane (TMS).Signal splitting patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), quintet (qu), sextet (sext) or multiplet (m).Coupling constants are quoted in Hertz (Hz).Carbon NMR spectra ( 13 C) were recorded on a Varian INOVA 300 at 75 MHz, or Bruker Ultrashield 500 at 125 MHz using deuterated solvent as the internal standard.Chemical shifts (C) are quoted in ppm downfield from TMS. NMR peak broadening due to rotamers is indicated (br) for some compounds with a Boc-protecting group.

Figure 3 .
Figure 3.An outline of a possible approach to the target compounds 1.