Amifostine: a tonic or toxin to myeloid progenitors

Abstract

It should come as no surprise, that the aminothiol prodrug, Amifostine (EthyolTM), whose actions as a hematopoietic promoter were recently realized, should now be considered to exert opposing biologic effects in leukemic cells [1]. A product of the United States Defense Department research emanating from the cold war, the relative selective cytoprotection afforded by amifostine to normal cells compared to tumor cells seemed at first glance somewhat difficult to accept. Indeed, relative differences in alkaline phosphatase distribution and pH within the tumor milieu appeared plausible [2–4], but insufficient to explain the agent’s ability to potentiate antineoplastic cytotoxicity in selected malignancies while conferring no protection to leukemic cells and protecting normal tissues [5–8]. The answer to this apparent dichotomy lies in amifostine’s chemical derivation and structural resemblance to endogenous polyamines, abundant but highly regulated organic cations involved in the homeostatic balance of cell death and survival. The dephosphorylated form of amifostine, WR-1065, and its symmetrical disulfide oxidation product, WR33278, display striking homology to spermidine and spermine, respectively [9,10]. In the oxidized extracellular environment, WR-33278 is believed to predominate, gaining entry into cells as a substrate for the polyamine facilitative transporter [11,12]. The activity of these transporters are strictly regulated by the feedback inhibitor, antizyme, which serves as a gate-keeper to limit intracellular polyamine content [13,14]. Within the cell, WR-33278 can be reduced to WR-1065. These novel aminothiols can then modulate cell behavior in two ways. They can compete with the naturally occurring polyamines for binding to DNA [15] and as substrates for amine oxidases [16]. Like endogenous polyamines, they induce the activity of spermine/spermidine acetyl transferase (SSAT), the first enzyme in polyamine catabolism, which is transcriptionally activated by the erythroid-specific transcription factor GATA-1 as an early event in erythroid differentiation [16,17]. In addition, these molecules can activate unique patterns of gene expression via polyamine-independent mechanisms, such as the redox sensitive tumor suppressor gene, p53 [18]. Like the polyamines, the cationic charge of the aminothiols facilitates their avid DNA binding, where they exert their many biologic activities including activation of early response genes, thymidine kinase, NFk-B, and other genes involved in the maintenance of genomic integrity [19–21]. Polyamines and polyamine analogues, like classical thiol antioxidants, afford protection against nuclear oxidation by serving as scavengers of reactive oxygen intermediates [22–24]. How then can we expect divergent effects on cell survival in normal, myelodysplastic and leukemic progenitors? Proliferating tumor cells maintain a high cellular polyamine content via ornithine decarboxylase-controlled biosynthesis; thereby blocking cell import by the induction of antizyme. Significant changes in polyamine homeostasis in tumor cells can be catastrophic, resulting in growth arrest and/or programmed cell death. Although active import of WR33278 is inhibited [25], passive inward diffusion of WR-1065 accelerates polyamine degradation by induction of SSAT, resulting in lowered apoptotic threshold. Sensing exceedingly high intracellular polyamine content, antizyme binds to and destabilizes ornithine decarboxylase, triggering its rapid proteosome degradation [26,27]. In normal cells where polyamine generation is not upregulated, the disulfide is actively imported into the cell, affording added defense against oxidative stress and resistance to apoptosis induction. The latter is indeed the case for normal hematopoietic progenitors,