(70a) Integration of a Green and Efficient Synthesis Step Enables Enhanced Production of the Antimalarial Precursor Artemisinin from Plant-Based Extraction

Artemisinin is the precursor to several important active pharmaceutical ingredients (API), which are essential in the treatment of malaria. Although an industrial process based on biotechnological and chemical conversion is already available, artemisinin is still mainly produced by extraction of the plant Artemisia annua, where it is formed as a secondary metabolite. The low content of the API in the dried matter (max. 1.4 wt‑%) and the long cultivation time of the plant caused high and fluctuating prices in recent years. Reducing the cost of the treatments and increasing their availability also for the population of developing countries, is a major issue in the fight against the disease.

Our approach to improve the production pathway is to use the plant not only as source for artemisinin but also for the co-extracted substance dihydroartemisinic acid (DHAA) – a natural precursor to artemisinin. A. annua contains up to equal amounts of DHAA as of artemisinin depending on the plant strain and the cultivation conditions. This former waste product of the extraction can be converted chemically to additional artemisinin via a photooxidation and subsequent acid-catalyzed reaction sequence [1]. The aim is to design the reaction step in manner so that it can be easily integrated to existing equipment without major adjustments to the upstream extraction and the downstream purification unit.

In previous studies [2] 9,10-dicyanoanthracene (DCA) and trifluoroacetic acid (TFA) were utilized as catalysts - both toxic and expensive substances, which require recycle after the synthesis. In addition, the reaction was performed at a low temperature of ‑20°C and necessitated highly concentrated conditions, the use of pure oxygen and pure solvents. In this study we demonstrate that these catalysts can be substituted by green alternatives. Co-extracted chlorophyll initiates the photooxidation making the addition of other photocatalysts unnecessary. Immobilized heterogeneous acids are used as catalysts for the second reaction to improve reusability and process safety (Figure 1). In addition, we will show that purified starting material and extensive cooling are not required to convert the DHAA contained in the extract.

The extract used for the reaction experiments was obtained by batch-extraction of dried, ground leaves of A. annua in toluene. To perform the partial synthesis, the obtained crude extract was fed into a mini-channel flow system combined with oxygen and irradiated by highly intensive visible light. The small dimensions of the reactor ensure high mass transfer rates between gas and liquid phase and strong illumination of the reactant solution.

The obtained extract is very diluted with respect to artemisinin and DHAA. To compare the catalyst performance with previous published studies performed in purified conditions [2], the partial synthesis was first investigated by adding additional DHAA to the extract (0.5 M DHAA) before testing the concept on the diluted conditions (0.0085 M DHAA).

The performance of co-extracted chlorophyll as photocatalyst was evaluated in comparison to 9,10-dicyanoanthracene at -20°C and 7 bar oxygen. In purified solutions, DHAA is completely converted in 2 min residence time and a yield of 90% of the desired hydroperoxide PO1 is obtained. Using the crude extract as reaction medium, gave the same conversion and yield after a residence time of 5 min, although chlorophyll is 4 times less concentrated than the photocatalyst in the reference case. When trifluoroacetic acid is added to the extract, artemisinin is gained with a yield of 67% equaling the results obtained in purified solutions [2]. The impurities present in the crude extract affected neither the yield nor the selectivity of artemisinin significantly. So prior purification of DHAA is not required before performing the partial synthesis. Additionally, the photooxidation benefits from the strong and broad absorption of chlorophyll enabling the use of different wavelengths in the visible range.

In the next step, trifluoroacetic acid was substituted by heterogenous acids. To identify suitable catalysts, various solid acids ranging from oxides to heteropoly acids were screened in batch experiments at ambient conditions. The best performing candidates, a β-zeolite (Si/Al = 360) and the ion-exchange resin Amberlyst 36, exceeded the performance of TFA in these low-pressure conditions with an artemisinin yield of 56% and 46%, respectively, in comparison to 33% with TFA. Both materials were applied as fixed-bed reactor directly coupled to the photoreactor. The reactor configuration was varied to identify flow conditions with sufficient mass transfer between gas and liquid, as well as liquid and solid phase. The best results were obtained with the β-zeolite (Si/Al = 360) packed in a single-particle reactor achieving a maximum artemisinin yield of 40%.

All results presented so far were performed with a starting concentration of DHAA exceeding the natural concentration in the extract 50-fold. Performing the partial synthesis in diluted solutions at otherwise optimized conditions (-20°C, 7 bar O₂, 0.25 M TFA) resulted in a drop of the final yield to 40%. Experiments with varied initial concentration of DHAA added to the extract showed, that this drop in selectivity to artemisinin was caused by the decreased excess of DHAA in comparison to other co-extracted metabolites.

Adaption of milder reaction conditions did not affect the performance of the partial synthesis at these diluted conditions. DHAA is converted still with 40% yield to artemisinin, when increasing the reaction temperature to ambient conditions, using pressurized air instead of oxygen and reducing the added amount of acid 4-fold to 0.06 M. The application of β -zeolite as acidic catalyst resulted in equal observed yields to artemisinin.

In conclusion, the conversion of the former waste product dihydroartemisinic acid to artemisinin in an additional synthetic step is an attractive extension of the conventional extraction process. Utilizing co-extracted chlorophyll as photocatalyst eliminates the addition and recycle of expensive artificial photosensitizers leading to a significant simplification of the process. The substitution of organic acids by heterogenous acidic catalyst increases the safety and sustainability of the process further. By applying these tools, 40% of DHAA extracted from A. annua can be converted to additional artemisinin by treating the extract only with visible light and air at ambient temperatures. Since no components are added for the reaction step, avoiding interference with the downstream purification units, the partial synthesis can be easily integrated to existing plants for extraction-based production of artemisinin.

References:

[1] Levesque, F. and P. H. Seeberger, Continuous-flow synthesis of the anti-malaria drug artemisinin. Angewandte Chemie - International Edition, 2012. 51(7): 1706-1709.

[2] Kopetzki, D., F. Levesque, and P. H. Seeberger, A continuous-flow process for the synthesis of artemisinin. Chemistry-a European Journal, 2013. 19(17): 5450-6.

[3] Triemer, S., K. Gilmore, G.T. Vu, P.H. Seeberger, and A. Seidel-Morgenstern, Literally Green Chemical Synthesis of Artemisinin from Plant Extracts. Angewandte Chemie - International Edition, 2018. 57(19): 5525-8.