Investigation and utilisation of the promiscuity of the sesquiterpene synthase amorpha-4,11-diene

In spite of the existence of several total syntheses of artemisinin including the use of engineered yeast for its production, synthetic or semi-synthetic artemisinin production has so far proven to be cost-ineffective to provide the target drug molecule on an industrial scale. Through a combination of synthetic biology and flow chemistry we will deliver a cost-effective semi-synthetic approach to the manufacturing of artemisinin with added value since our approach will also be adaptable to the generation of novel artemisinin derivatives that are not easily accessible by conventional synthetic chemistry.
At the heart of artemisinin production lies the enzyme amorphadiene synthase (ADS), a terpene cyclase that converts the water-soluble substrate farnesyl diphosphate (1) into a hydrocarbon product that is water-insoluble. Hence product inhibition is a major issue for utilizing such enzymes in a time and cost-effective manner as the product is slow to diffuse from the enzyme active site. We will employ technology provided by a leading producer of high-value sesquiterpenes (Isobionics BV, Netherlands) who have developed a fermentation procedure using a proprietary Rhodobacter sp. This generates large quantities of any desired sesquiterpene with a known cyclase. We will introduce both the ADS gene and an gene coding for a P450 cytochrome that is known to generate 12-hydroxyfarnesyl diphosphate from FDP 1 (CYP124A1) into the producing strain of Rhodobacter and use this to generate dihydroartemisinic aldehyde directly from fermentation. To mitigate risk we will also use yeast strains developed in-house by us or from Huvepharma to perform the same synthetic biology. Huverpharma are a new company formed in 2015 created to purchase the Sanofi factory in Garessio, Italy. The latter yeast strain has been demonstrated to generate large quantities of artemisinic acid. We will convert this strain with the CYP124A1 gene to make dihydroartemisinic aldehyde as above and can of course utilize it directly to generate artemisinic acid and use our flow-oxidation methodology (vide infra) to generate artemsinin directly. Production of dihydroartemsinic aldehyde by fermentation reduces the necessary chemical steps to artemisinin and so cuts the cost of its production. In added value, as a longer-term goal, this system will be flexible as other P450 cyctochromes and/or methyltranferases can be added to the system to generate completely new analogues of artemisinin that would not be readily accessible through conventional synthetic chemistry. These will modify the skeleton of amorpha-4,11-diene prior to the in-flow oxidations to the artemisinin skeleton and hence not require chemistry that will disrupt the delicate oxygen-dense core essential to the antimalarial activity. This will provide a viable approach to combat the growing ACT resistance.