Partial purification and characterization of a new human membrane-bound carbonyl reductase playing a role in the deactivation of the anticancer drug oracin

Carbonyl reducing enzymes play important roles in the biotransformation and detoxification of endo- and xenobiotics. They are grouped into two protein superfamilies, the short-chain dehydrogenases (SDR) and aldo–keto reductases (AKR), and usually are present in the cytoplasm of a cell. So far, only one membraneous carbonyl reductase has been described, 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), which is located in the endoplasmic reticulum and which significantly contributes to the metabolism of a variety of carbonyl containing drugs and toxicants. Oracin is a new and prospective anticancer drug bearing a prochiral carbonyl moiety. The main metabolic pathway of oracin is carbonyl reduction to 11-dihydrooracin (DHO) which, however, eliminates the therapeutic potential of the drug, because the two DHO enantiomers formed have significantly less anti-tumor activities. Therefore, the oracin inactivating enzymes should urgently be identified to search for specific inhibitors and to enhance the chemotherapeutic efficacy. Interestingly, the calculation of enzyme specific activities and stereospecificities of (+)-DHO and (−)-DHO formation strongly suggested the existence of a second, hitherto unknown microsomal oracin carbonyl reductase in human liver. Therefore, the aim of the present study was to provide proof for the existence of this new enzyme and to develop a purification method for further characterization. First, we succeeded in establishing a gentle solubilization technique which provided a favourable detergent surrounding during the further purification procedure by stabilizing the native form of this fragile protein. Second, we could partially purify this new microsomal carbonyl reductase by a two step separation on Q-sepharose followed by Phenyl-sepharose. The enzyme turned out to be NADPH specific, displaying kinetic values for oracin carbonyl reduction of Km = 42 μM and Vmax = 813 nmol/(30 min × mg protein). Compared to the microsomal fraction, the enzyme specific activity towards oracin could be enhanced 73-fold, while the stereospecificity of (+)-DHO formation shifted from 40% to 86%. Considering these data for 11β-HSD1, as described in previous reports, it is clear that the microsomal carbonyl reductase investigated in the present study is new and has a great potential to significantly impair the chemotherapy with the new anticancer drug oracin.