Synthesis of Cocaine analogs from Arecoline Written by Beagle, edited by Rhodium
Introduction
This is one of the most promising area of research on cocaine analogs that I’ve come across. In the article [1], the authors were attempting to come up with an equivalent of methadone for cocaine addicts. That is, a substance that would have some of the effects of cocaine without all the euphoria (Coke-Lite®, I guess it will be called). The research was based on some work done in 1973 (by Clarke [2]) in which a series of similar compounds was made and found not to cause stimulation in mice (i.e. no fun). However, in this study, the compounds were found to be up to 30 times the potency of cocaine in blocking dopamine transport or binding of cocaine to its receptor. So the results are rather confusing. Possibly the researchers have actually found a lead in their search for coke-adone, or maybe they have found a series of ridiculously easy to synthesize highly potent coke analogs. Either way, interesting work. Currently they are shooting up monkeys to see if they like it or not, but it may be years before they publish their work. Maybe someone out there could help them out. I always say, never send a monkey to do a man’s work.
In Clarke’s 1973 paper [2], a series of structurally reduced cocaine analogs are synthesized in which 2 carbons and an ester group have been removed from the tropane skeleton, giving 4-phenyl-piperidines with an ester group at the 3-position. The most active compound in their series is about 30 times the potency of cocaine (in blocking dopamine uptake). The synthesis is much easier than any other active cocaine analog, and starts from non-exotic reagents.
The main starting material used by the authors is Arecoline (3,4-unsaturated N-methyl-piperidine-3-carboxylic acid methyl ester), the active principle of the popular red betel nut. Treatment of this alpha,beta-unsaturated ester with the Grignard reagent 4-Chlorophenylmagnesium Bromide gave rise to four products, easily grouped into two racemic forms of each cis– and trans– isomer. The racemic cis-isomer is twice as potent as cocaine, while the racemic trans-isomer is only half as strong.
The authors start with arecoline (the active compound in betel nut: N-methyl-piperidine- 3-carboxylic acid methyl ester with a double bond between the 3 and 4 positions). Reaction of this with a 4-chlorophenylmagnesium bromide grignard reagent gives a mixture of a total of four discrete entities: A cis– and trans-mixture in a 3:1 ratio, each constisting of a of two isomers, each isomer also being racemic, all which the which the authors resolve through fractional crystallization as their dibenzoyl tartaric acids, but this is not strictly necessary since each of the isomers are active. Arecoline Hydrobromide is a fairly cheap commercial starting material, $80/50g from Aldrich. Also, I think that it is used in veterinary medicine.
Synthetic Outline
The Arecoline freebase was prepared by treating the Arecoline Hydrobromide of commerce with concentrated aqueous sodium bicarbonate, followed by extraction with methylene chloride, thorough drying of the extracts over anhydrous Magnesium Sulfate, followed by suction filtration and evaporation of the solvent in vacuo. The freebase was then used as is in the next step.
To a solution of 166ml of 1M 4-chlorophenylmagnesium bromide in 700ml ether was added 12.9g of arecoline freebase in 300ml ether at -10°C. The mix was stirred at -10°C for 30 min, poured onto ice and treated with 200ml of 10% HCl. The aqueous layer was separated, washed with 200ml ether, cooled in an ice bath, and 100ml of saturated sodium bicarb. solution added. The solution was extracted with 2x100ml of ether, washed with brine, dried, and concentrated in vacuo. The crude mixture was crystallized from EtOAc/hexane to give racemic cis-isomer as a white solid (5g, 22%). Additional cis-isomer (Total: 12.4g, 56%, mp 98-99°C), as well as 2.0g (18%) of the trans-isomer was obtained by flash chromatography of the mother liquor.
Alternatively, it looks like that 4-piperidone that you fentanyl-heads have been dreaming about could also be used to synthesize these compounds. I picture reaction of 4-piperidone with e.g. methyl chloroformate to give piperidone 3-methylcarboxylate (hey drone: enolate chemistry!). This could then be reduced to the alcohol, and dehydrated to give arecoline. Or the piperidone 3-methylcarboxylate could be reacted with phenylmagnesium bromide, dehydrated, and reduced to give the final compound. Note the use of one precursor to make both highly potent opiate and cocaine analogs. Talk about speedballs!
Or what about starting from nicotinic acid (pyridine-3-carboxylate; a B-vitamin)? That should be cheap as dirt. Seems to me that reaction with phenyllithium would give 4-phenyl-3-carboxyl-tetrahydropyridine, which would only need esterification and reduction to give the final compound. Or maybe there would be a better way to start from nicotinic acid?
In the last two proposed syntheses, care would need to be taken to avoid formation of a potentially nerotoxic MPTP analog!
Another pertinent reference is US Pat 6472422, dealting with various cocaine analogs.
References
[1] Journal of Medicinal Chemistry, Vol 41, No 11, pp 1962-9 (1998)
[2] Journal of Medicinal Chemistry, Vol 16, No 11, pp 1260-7 (1973)
