Abstract

The cytosolic class aldehyde dehydrogenase (ALDH-3) present in human normal tissues/secretions is apparently much less able to catalyze the oxidation aldophosphamide to carboxyphosphamide than is the ALDH-3 present in human tumor cells/tissues, suggesting that the former may be less able to protect cells from the cytotoxic action of cyclophosphamide, mafosfamide, and other oxazaphosphorines. To test this notion, relatively large and approximately equal amounts of human normal stomach mucosa ALDH-3 and catechol-induced human breast adenocarcinoma MCF-7/0 ALDH-3 were first electroporated into cells (MCF-7/0) that constitutively express only very small amounts of the enzyme. The resultant preparations were then tested for sensitivity to mafosfamide. ALDH-3 activities (NADP-dependent catalysis of benzaldehyde oxidation) were 1.7, 212, and 183 mlU/10(7) cells in sham-electroporated MCF-7/0 cells, and MCF-7/0 cells electroporated with stomach mucosa ALDH-3 and catechol-induced MCF-7/0 ALDH-3, respectively. LC90 values (concentrations of mafosfamide required to effect a 90% cell kill) were 62, 417, and >1,000 microM, respectively. The three preparations were equisensitive to phosphoramide mustard (LC90 = approximately 850 microM). Inclusion of benzaldehyde in the drug exposure medium fully restored the sensitivity of MCF-7/0 cells electroporated with either enzyme to mafosfamide. These observations support the notions that 1) cellular sensitivity to the oxazaphosphorines decreases as the cellular content of ALDH-3 increases, 2) the foregoing is the consequence of ALDH-3-catalyzed oxidation (thus detoxification) of aldophosphamide, and 3) the ALDH-3 present in at least some tumor cells/tissues is a slight variant of the ALDH-3 present in normal tissues/secretions. Furthermore, they illustrate the utility of electroporation used as a tool to determine whether a given enzyme, or even more generally, protein or other macromolecule, is a determinant of cellular sensitivity to a given cytotoxic agent.