Direct Phosphorylation of Psilocin Enables Optimized cGMP Kilogram-Scale Manufacture of Psilocybin

This paper (2020) describes and makes available the method for producing psilocybin on a large (1kg) scale.

Abstract

“A second-generation kilogram-scale synthesis of the psychedelic tryptamine psilocybin has been developed. The synthesis was designed to address several challenges first encountered with the scale-up of previously described literature procedures, which were not optimized for providing consistent yield and purity of products, atom economy, or being run in pilot plant-scale reactors. These challenges were addressed and circumvented with the design of the second-generation route, which featured an optimized cGMP large-scale Speeter–Anthony tryptamine synthesis to the intermediate psilocin with improved in-process control and impurity removal over the three steps. Psilocin was subsequently phosphorylated directly with phosphorous oxychloride for the first time, avoiding a tedious and poor atom economy benzyl-protecting group strategy common to all previously described methods for producing psilocybin. In this report, the challenges encountered in a 100 g scale first-generation literature-based synthesis are highlighted, followed by a detailed description of the newly developed second-generation synthesis to provide over one kilogram of high-purity psilocybin under cGMP.“

Authors: Robert B. Kargbo, Alexander Sherwood, Andrew Walker, Nicholas V. Cozzi, Raymond E. Dagger, Jessica Sable, Kelsey O’Hern, Kristi Kaylo, Tura Patterson, Gary Tarpley & Poncho Meisenheimer

Summary

Introduction

In the last two decades, there has been renewed interest in the clinical study of psychedelics for the potential treatment of a range of mental illnesses. Psilocybin was selected as the most suitable candidate to usher in the modern era of clinical research into psychedelics because of several reasons.

Results and Discssion

The first-generation synthesis of psilocybin was based on several known methods and provided 100 g of high-purity psilocybin, which enabled the initiation of clinical trials. However, the first-generation synthesis presented several challenges that were not well predicted by bench-scale development reactions.

From commercially available 4-acetoxyindole and oxalyl chloride, acyl chloride 2 was not isolated and telescoped directly into the formation of ketoamide 3. To accommodate variable levels of tetramethyloxamide impurity, a large excess of lithium aluminum hydride was found to be optimal at stage 3, and a lengthy washing process was required to maximize yield. The phosphorylation of psilocin in stages 4a and 4b is a complex reaction, and the parameters that influence the benzyl migration remain little understood.

The kinetics of the rearrangement of 5 to 5z on the multigram scale was found to be sensitive to control of temperature. This resulted in an ultrafine particle that impeded the final filtration. The zwitterion product was subjected to exhaustive catalytic hydrogenolysis at stage 5 to remove two benzyl groups and smoothly reveal zwitterionic psilocybin (6). This process provided just 173 g of crude-isolated 6 with 99.6% high-performance liquid chromatography purity.

In order to scale up the first-generation synthesis of psilocin under cGMP conditions, it was imperative to address several questions, including how to minimize or eliminate the tetramethyloxamide byproduct at stage 2, how to quickly purify psilocin, and how to avoid boiling water in the purification and final isolation of psilocybin. A second campaign was designed to improve the first campaign’s yields by reworking the first three steps and direct phosphorylation of psilocin to provide purified psilocybin.

The first task was to develop a scalable process for the synthesis of acyl chloride 2 in good yield and purity. The process involved treating 4-acetoxyindole (1) with 1.2 equiv of oxalyl chloride at 10 °C in methyl tert-butyl ether (MTBE), and washing the collected solid with 1:3 MTBE/heptane. After obtaining 2 in a relatively stable, pure, and high yielding form, the amidation step was straightforward. The ketoamide 3 was obtained in 83% yield and >98% area purity by HPLC.

Psilocin (4) was synthesized at stage 3 of a first-generation process. The stability of the product depended on the processing time in solution. The initial approach to make -hydroxy psilocin involved exothermic addition of LAH to a solution of 3 in THF at 0 °C followed by refluxing. The optimized process involved adding 3 as a slurry in 2-MeTHF to a solution of LAH at >60 °C and then heated to reflux.

The approach to quenching the reaction mixture was also found to be important for a successful process. A quench with THF/H2O (100:27) was used, to which silica gel and MTBE were added, and this led to a white to off-white solid that turned dark green rapidly.

Anhydrous sodium sulfate was diluted with DCM/ MeOH and filtered through a small pad of silica gel to give a colorless solution. The crude product was isolated by solvent swap to heptanes and purified by reslurry in heptanes and diisopropyl ether.

Direct phosphorylation of psilocin to psilocybin has been used in medicinal chemistry efforts and natural product synthesis. However, the initial attempts at direct phosphorylation failed to deliver psilocybin in isolable quantity.

Stoichiometry optimization indicated that the reaction of 4 with 1.5 equiv POCl3 was smooth and complete, and that celite facilitated the formation of a stirrable mixture. The reaction was monitored at various stages, and the phosphorodichloridate intermediate 7 was formed relatively quickly.

The hydrolysis step of the psilocybin synthesis was completed by quenching the crude mixture containing primarily intermediate 7 into a premixed solution of 30% aqueous THF containing 6 equiv of triethylamine at 0 °C. The resulting slurry was filtered and the remaining 6 adhered to the celite was removed by washing with water.

Early patent data indicated that psilocybin could be recrystallized in boiling water or methanol, but subsequent thermal stability studies showed that psilocybin in water at 90 °C hydrolyzed to psilocin after 1 hour. The final purification was accomplished by employing sequential reslurry in methanol followed by reslurry in warm water.

The level of one earlier eluting impurity observed at RRT 0.58 was typically reduced during the final water reslurry step and drying process, but during kilo-scale manufacturing the level of this impurity remained above the ICH Q3A guidance for unidentified impurities at 0.11%. The impurity was identified as pyrophosphate structure 10 with high confidence.

Conclusions

A second-generation synthetic approach to psilocybin was developed, which enabled the manufacture of 1.21 kg of API in 17% overall yield with 99.7% HPLC assay purity. The new process featured a novel direct phosphorylation reaction, which avoided an unacceptable protecting group strategy.

Experimental Section

3-(2-Chloro-2-oxoacetyl)-1H-indol-4-yl Acetate (2) was prepared from oxalyl chloride and 4-acetoxyindole by adding them to MTBE and stirring. The reaction was maintained at 010 °C for 24 h, and the product was dried for at least 30 min under N2 atmosphere on the filter.

Tetrahydrofuran was charged to the vessel containing 3-(2-chloro-2-oxoacetyl)-1H-indol-4-yl acetate and stirred until complete dissolution. A solution of 2 M dimethylamine in THF was added over 1 h at 010 °C via a dip pipe. Triethylamine was diluted in tetrahydrofuran and warmed to 1520 °C. Heptane was added and the mixture was cooled to 05 °C and maintained within this temperature range for at least 1 h. The mixture was heated to 8085 °C, cooled to 1822 °C over at least 2 h, filtered, washed with 2-propanol at 010 °C, pulled dry under vacuum for at least 15 min, and dried for at least 16 h in vacuo at 3540 °C to afford the desired compound.

A yellow-colored slurry was observed after adding 3-(2-(dimethylamino)-2-oxoacetyl)-1H-indol-4-yl acetate and 2-Me-THF to a clean, dry reactor under a nitrogen atmosphere.

The reaction mixture was cooled to 020 °C and quenched with a solution of THF/H2O (100:27) while stirring for at least 1530 min at 0 20 °C. The temperature of the reaction mixture was adjusted to 2025 °C and a solution of DCM/MeOH was charged. A reaction mixture was poured onto a pad of celite, and the filtrate was drummed up to avoid unnecessary air exposure. A pad of silica was added, and the combined filtrates were concentrated to 5 vol, maintaining the internal temperature 50 °C.

To make 3-(2-(Dimethylamino)ethyl)-1H-indol-4-yl Dihydrogen Phosphate, charge THF and phosphorus oxychloride to a clean, dry reactor under a nitrogen atmosphere, and cool the reactor contents to 0 to 15 °C. Stir the mixture for 2 to 15 °C. During this time, a quench solution was prepared by adding THF/H2O and Et3N to a clean reactor, and the crude psilocin reaction mixture was slowly added into the quench solution, maintaining the internal temperature at 20 to 0 °C.

The quenched mixture was stirred at 0 to 20 °C for at least 60 min, and then filtered. The biphasic filtrate was transferred back to the reactor, and the lower aqueous phase was separated, and the organic phase was removed. Upon reaching the aqueous distillation, purified water was charged at 1825 °C and the solution was stirred for at least 24 h. Psilocybin normally precipitates at this time. The crude psilocybin was charged to a clean, dry reactor under the nitrogen atmosphere at 2025 °C, stirred for at least 12 h, filtered under nitrogen, washed in turn with purified water at 2025 °C, and pulled dry under the nitrogen atmosphere for at least 2 h. A white solid was afforded by drying the trihydrate form initially isolated to the desired anhydrate form A by XRPD.

Supporting Information

The authors would like to thank Dr. Paul Daley for performing GC/MS analyses and for recommending POCl as a direct phosphorylation reagent.

Notes

This paper is included in our ‘Top 10 Articles on Psychedelics in the Year 2020‘

This open-access paper describes how one could make psilocybin on a large scale. This work is made available by the non-profit Usona Institute, and (partially) prevents for-profit companies from taking out patents on a/this method for making psilocybin at scale.