Anthracycline secondary alcohol metabolite formation in human or rabbit heart: biochemical aspects and pharmacologic...

Clinical use of the anticancer anthracyclines doxorubicin (DOX) and daunorubicin (DNR) is limited by development of cardiotoxicity upon chronic administration. Secondary alcohol metabolites, formed after two-equivalent reduction of a carbonyl group in the side chain of DOX or DNR, have been implicated as potential mediators of chronic cardiotoxicity. In the present study we characterized how human heart converted DOX or DNR to their alcohol metabolites DOXol or DNRol. Experiments were carried out using post-mortem myocardial samples obtained by ethically-acceptable procedures, and results showed that DOXol and DNRol were formed by flavin-independent cytoplasmic reductases which shared common features like pH-dependence and requirement for NADPH, but not NADH, as a source of reducing equivalents. However, studies performed with inhibitors exhibiting absolute or mixed specificity toward best known cytoplasmic reductases revealed that DOX and DNR were metabolized to DOXol or DNRol through the action of distinct enzymes. Whereas DOX was converted to DOXol by aldehyde-type reductase(s) belonging to the superfamily of aldo-keto reductases, DNR was converted to DNRol by carbonyl reductase(s) belonging to the superfamily of short-chain dehydrogenase/reductases. This pattern changed in cardiac cytosol derived from rabbit, a laboratory animal often exploited to reproduce cardiotoxicity induced by anthracyclines and to develop protectants for use in cancer patients. In fact, only carbonyl reductases were involved in metabolizing DOX and DNR in rabbit cardiac cytosol, although with different K(m) and V(max). Collectively, these results demonstrate that human myocardium convert DOX and DNR to DOXol or DNRol by virtue of different reductases, an information which may be of value to prevent alcohol metabolite formation during the course of anthracycline-based anticancer therapy. These results also raise caution on the preclinical value of animal models of anthracycline cardiotoxicity, as they demonstrate that the metabolic routes leading to DOXol in a laboratory animal may not be the same as those occurring in patients.