Artemisinin Goes With the Flow | January 23, 2012 Issue - Vol. 90 Issue 4 | Chemical & Engineering News
Volume 90 Issue 4 | p. 4 | News of The Week
Issue Date: January 23, 2012

Artemisinin Goes With the Flow

Pharmaceuticals: Continuous-flow synthesis produces antimalarial at low cost
Department: Science & Technology
News Channels: Biological SCENE
Keywords: artemisinin, artemisinic acid, continuous flow reactor, malaria
Three chemical transformations are completed in a single continuous-flow process without the need to isolate and purify intermediates.
A reaction scheme showing the three oxidation reactions en route between artemisinic acid and artemisinin that can be acomplished with Seeburg's new process.
Three chemical transformations are completed in a single continuous-flow process without the need to isolate and purify intermediates.
Lévesque working at the continuous-flow reactor.
Credit: Ulrich Kleiner
Postdoc Francois Levesque working at the continuous-flow reactor
Lévesque working at the continuous-flow reactor.
Credit: Ulrich Kleiner

Gaining access to the antimalarial compound artemisinin could become easier and less expensive, thanks to a continuous-flow synthetic procedure developed by researchers in Germany (Angew. Chem. Int. Ed., DOI: 10.1002/anie.201107446). The new procedure improves upon current synthetic approaches, according to its inventors, and offers a more reliable route to the compound than extraction from its natural source, the plant Artemisia annua.

Artemisinin is currently the most effective treatment in the fight against multi-drug-resistant forms of malaria, and the World Health Organization recommends artemisinin-based combination therapies as first-line drugs. Although artemisinin can be extracted from A. annua, the plant is seasonal, making the supply unreliable and the cost volatile.

Scientists previously devised a semisynthetic route to artemisinin based on the intermediate artemisinic acid—a compound that can be produced by specially engineered yeast. That technology was invented by a group led by Jay D. Keasling, a professor at the University of California, Berkeley, and was developed commercially by the biotech firm Amyris. Amyris has licensed the technology to Sanofi with the hope of bringing semisynthetic-artemisinin-based therapies to the market by 2013.

But converting artemisinic acid into artemisinin is no simple task. It involves installing an endoperoxide group and generating three of artemisinin’s rings. Doing this complex chemistry on a scale sufficient to meet the demand of 225 million malaria patients in developing countries presents a formidable challenge.

That’s where the work of Peter H. Seeberger and François Lévesque comes in. The researchers, from Germany’s Max Planck Institute for Colloids & Interfaces, have now figured out how to combine three key reactions—photochemically induced oxidation with singlet oxygen, acid-mediated cleavage of an oxygen–oxygen bond, and oxidation with triplet oxygen—into a single continuous-flow process en route to converting artemisinic acid into artemisinin.

One crucial step in the transformation requires photochemical generation of singlet oxygen, which is highly reactive. This reaction is difficult to do in traditional large-scale reactors. Seeberger and Lévesque discovered they could generate singlet oxygen on a relatively large scale in a continuous-flow reactor by simply wrapping the tubing that contains the reactant flow around a cooled mercury lamp (Org. Lett., DOI: 10.1021/ol2017643).

Chemists often avoid singlet oxygen reactions “because of the hazardous batch operating conditions and the potentially explosive hydroperoxides formed,” notes Kevin Booker-Milburn, a continuous-flow chemistry expert at the University of Bristol, in England. “I believe that chemists will now be much more likely to use this reaction under the Seeberger flow conditions since the potentially dangerous peroxides can be dealt with continuously, as they are not ‘stockpiled’ as they would be in a conventional batch process,” he says.

“This eye-catching, technically enabled synthesis will very likely stimulate process chemists to consider new oxidation protocols that they would have traditionally shunned,” Booker-Milburn adds.

At present, Seeberger tells C&EN, the continuous-flow reactor can produce 800 g of artemisinin per day, and he estimates it would take 400 such reactors to make the world’s supply of the drug. “This is just the tip of the iceberg,” Seeberger says. “I believe there are many other drugs that could be made in this way.”

Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society
Prof C. Devakumar (January 23, 2012 3:05 PM)
Great innovation. A technological breakthrough. Now go and crack at some other giant natural products like azadirachtins, avermectins for instance. Hybridization of science and technology is the need of the hour for a continuous flow of such marvels for mitigation of the suffering, the world over.
Richard K Haynes (February 21, 2012 1:18 AM)
Whilst per se the technical aspects of improving the photochemical step are clear, the fact that such on their own are published in Angewandte Chemie International Edition is something of a surprise, as these constitute a process engineering exercise, though clearly with some merit given teh overall importance now of artemisinin. We were the authors who together with Acton and Roth (ref. 7 in the Seeberger at al. paper) conducted the singlet oxygenation-triplet oxygenation sequences on dihydroartemisinic acid to generate artemisinin. We were the first to identify the nature of the Hock cleavage as applied to artemisinic (qinghao acid), and actually isolated the hydroperoxides corresponding to 3 and enol 4, and described the nature of the conversion into hydroperoxide related to 5 (authors' Schemes 1 and 2)(see Haynes, Vonwiller et al. J. Am. Chem. Soc. 1995, 117, 11098-11105; Acc. Chem. Res., 1997, 30, 73-79 and all previous references). Whilst our conversion overall may be deemed 'technically difficult', it is not, and its application is obvious from our papers. Our Accounts review where we summarize application to the preparation of artemisinin derivatives from the appropriately functionalized artemisinic acid precursors is not even referenced. Seeberger et al. infer (ref. 7) that our conversions were 'partially successful'; our conversions indeed were 'completely successful'. They also make a play of using the least 'non-toxic' halogenated solvent dichloromethane for the ultimate oxygenation step; we found that copper(II) triflate in dichloromethane worked well for the oxygenation step as well, that is in contrast to the authors’ findings, who clearly had to labor with acid catalyzed formation of the lactone 12. The origin of the authors’ use of trifluoroacetic acid is also not referenced – this reagent was developed by Acton (ref. 7). We also lodged a patent on this process when we first worked on this in 1988 – the relevant references are in our JACS and ACR papers. There is no reference to our work as it pertains to descriptions in the text relating to any of the intermediates related to those described in Schemes 1 and 2 (compare our Scheme 1 in our ACR paper that clarifies all); we explained the stereochemical outcome of the reaction that is a consequence of the stereochemistry of the groups flanking the enol that lead to the alpha-disposed hydroperoxide. The citation of reference 16 referring to a series of ‘condensation reactions’ is in fact our own work that was used by Brown et al., again without citation of our work.
To smmarize, all aspects of the science inherent in the above conversion (with the exception of formation of lactone 12) were previously established by Vonwiller, Haynes et al., and, in the case of the ultimate catalyst used, trifluoroacetic acid, by Acton and Roth. The Vonwiller-Haynes results apply in particular to the nature of the Hock cleavage of the hydroperoxide leading to the formation of the ring expanded product 10, and the nature of its collapse to the enol. We were able to isolate and characterize the closely related enol from artemisinic acid by 1H NMR spectroscopy as highlighted in our JACS paper. Is it so that Seeberger et al. by not citing the sources of their chemistry mislead by omission? - this leaves rather a poor impression.

Leave A Comment

*Required to comment