A Microbial Reductive Conversion of the Steffimycin Chromophore: Chemical and Mechanistic Studies on a Soluble, Nadhdependent, Nonflavin-Containing Reductase from Streptomyces Steffisburgensis
An understanding of a microbial reduction of the anthracycline antibiotic, steffimycin (1) was approached from two directions. First, the NADH-dependent enzyme responsible for the reduction was purified from the Streptomyces steffisburgensis cytosol and characterized. Second, information was obtained on the reduction and oxidation of 1.
The enzyme was a single polypeptide chain, acidic, and stable in solution. It had a molecular weight of 80,000 and no prosthetic groups.
The 5-step purification process resulted in greater than 1,000-fold purification and a 9% yield of the homogeneous enzyme from extracts of S. steffisburgensis cells.
The enzyme reduced 1 by a two-electron mechanism to a compound whose properties resembled those of other anthracycline hydroquinones, i.e., no visible chromophore, auto-oxidation to 1 in air, more polar than 1, absorbance at 254 nm 10-fold greater than that of 1, and no ESR spectrum.
Due to oxidation of this product in aerobic solutions the reduction reactions were observable using an oxygen electrode. Unfortunately we observed product formation only by sampling an anaerobic reaction mixture at intervals and analyzing these samples by reverse phase hplc. Spectroscopically monitoring the NADH concentration in aerobic reaction mixtures was the assay used for most studies reported.
At conditions used for enzymically reducing 1, therapeutic anthracyclines were not substrates for this enzyme. Reduced steffimycin, warfarin, Cibachron Blue dye, and p-(hydroxymercuri) benzoic acid, inhibited the enzyme.
In the second approach we investigated the reductive and oxidative behavior of 1 by electrochemical means, with inorganic reducing agents, and in enzymic systems. Upon reduction of the chromophore, steffimycin was significantly more stable than chromophore-reduced therapeutic anthracyclines. Whereas, reduction under any conditions of the 5,12 quinone groups of either daunomycin (Daunorubicin) or adriamycin (Doxorubicin) with one or two electrons produced a rapid and quantitative elimination of daunosamine to form their respective 7-deocyanthracycliones; similarly reduced 1 reacted neither as rapidly nor as quantitatively. This indicated stabilization of the reduced intermediate of 1.
Since the reduced intermediate of 1 is stabilized, it is a poor bioalkylating agent. This unique chemistry for an anthracycline lends weight to the bioalkylation theories of anthracycline toxicity proposed by Moore, Bachur and others and reviewed in this work.