Streamlined approach to high-quality purification and identification of compound series using high-resolution MS and NMR.

Automated medicinal chemistry (parallel chemistry) has become an integral part of the drug-discovery process in almost every large pharmaceutical company. Parallel array synthesis of individual organic compounds has been used extensively to generate diverse structural libraries to support different phases of the drug-discovery process, such as hit-to-lead, lead finding, or lead optimization. In order to guarantee effective project support, efficiency in the production of compound libraries has been maximized. As a consequence, also throughput in chromatographic purification and analysis has been adapted. As a recent trend, more laboratories are preparing smaller, yet more focused libraries with even increasing demands towards quality, i.e. optimal purity and unambiguous confirmation of identity. This paper presents an automated approach how to combine effective purification and structural conformation of a lead optimization library created by microwave-assisted organic synthesis. The results of complementary analytical techniques such as UHPLC-HRMS and NMR are not only regarded but even merged for fast and easy decision making, providing optimal quality of compound stock. In comparison with the previous procedures, throughput times are at least four times faster, while compound consumption could be decreased more than threefold.

Biotechnological production of 20-alpha-dihydrodydrogesterone at pilot scale.

The human sex hormone progesterone plays an essential and complex role in a number of physiological processes. Progesterone deficiency is associated with menstrual disorders and infertility as well as premature birth and abortion. For progesterone replacement therapy, the synthetic progestogen dydrogesterone is commonly used. In the body, this drug is metabolized to 20α-dihydrodydrogesterone (20α-DHD), which also shows extensive pharmacological effects and hence could act as a therapeutic agent itself. In this study, we describe an efficient biotechnological production procedure for 20α-DHD that employs the stereo- and regioselective reduction of dydrogesterone in a whole-cell biotransformation process based on recombinant fission yeast cells expressing the human enzyme AKR1C1 (20α-hydroxysteroid dehydrogenase, 20α-HSD). In a fed-batch fermentation at pilot scale (70 L) with a genetically improved production strain and under optimized reaction conditions, an average 20α-DHD production rate of 190 µM day(-1) was determined for a total biotransformation time of 136 h. Combined with an effective and reliable downstream processing, a continuous production rate of 12.3 ± 1.4 g 20α-DHD per week and fermenter was achieved. We thus established an AKR-dependent whole-cell biotransformation process that can also be used for the production of other AKR1C1 substrates (as exemplarily shown by the production of 20α-dihydroprogesterone in gram scale) and is in principle suited for the production of further human AKR metabolites at industrial scale.