Nucleoside analog phosphorylation and mitochondrial enzymes : Studies on molecular targets of the anti-leukemic compound 9-beta-D-arabinofuranosylguanine

Abstract: Nucleoside analogs are commonly used in treatment of cancer and viral infections. A way to improve the therapies would be to minimize the acquirement of resistance and side effects such as delayed cytotoxicity. Nucleoside analogs are phosphorylated and thereby activated by cellular kinases and to understand more about their phosphorylation by mitochondrial enzymes we have studied the molecular targets of the guanosine analog 9-beta-D-arabinofuranosylguanine (araG). This analog is a substrate of both the mitochondrial deoxyguanosine kinase (dGK) and the cytosolic deoxycytidine kinase (dCK). The prodrug of the biologically active araG, nelarabine, has proven highly efficient in particular in patients with T-cell acute lymphoblastic leukemia. Although the mechanism of action of araG is not fully understood, the accumulation of araG tripbosphates (araGTP) has been correlated to cytotoxicity both in vitro and in vivo. AraGTP acts as a structural analog of deoxyguanosine triphosphate (dGTP) and is thereby incorporated into DNA. The accumulation of araGTP is independent of the cell cycle, which is not surprising since both dCK and dGK are expressed throughout the cell cycle. Incorporation of araGMP into nuclear DNA has been suggested as a critical event for cytotoxicity. A recent study has suggested a role of mitochondria in the cell specific toxicity of dGTP with intra-mitochondrial accumulation of dGTP and inhibition of mtDNA repair. The doselimiting toxicity in the clinical trials with nelarabine has been neurotoxicity, but less pronounced adverse effects include other symptoms similar of drugs causing mitochondrial toxicity. We have shown that araG can be incorporated into mtDNA but the mtDNA incorporation does not, however, cause the acute cytotoxicity of araG and we do presently not know to what extent it contributes to the cytotoxic action of the analog. It cannot be excluded that long-term exposure to araG may cause mtDNA alterations with subsequent delayed mitochondrial toxicity. Several studies on mechanisms of resistance to araG have been performed. These studies have shown partly conflicting results as to the molecular mechanism of resistance. In our studies we found that araG resistance can occur by two separate molecular mechanisms that can occur sequentially. The first mechanism is associated with a decrease of araG incorporation into mtDNA and the second event is associated with loss of dCK activity, whereas the dGK activity remained at the same level as in the control cells. We do not yet know how the decreased incorporation of araG into mtDNA contributes to the resistant phenotype, but we know that araG does not cause mtDNA depletion or altered translation of mtDNAencoded genes. To study differences in gene expression in the araG resistant cells we have initiated microarray analysis. In the search for enzymes that could contribute to the activation of nucleoside analogs in the mitochondria we found a UMP-CMP kinase from Drosophila melanogaster that localized to the mitochondria. The recombinant enzyme accepted pyrimidine nucleoside monophosphates as substrates. The enzyme contained an N-terminal signal targeting the enzyme to the mitochondria. The identification of a functional mitochondrial import signal in the Dm.UMP-CMP kinase suggests that this enzyme and its homologues in other species may be involved in the mitochondrial phosphorylation of pyrimidine nucleoside monophosphates. However, the mitochondrial homologue in human cells remains to be identified. It was recently shown that mutations in the genes coding for dGK and the mitochondrial thymidine kinase 2 (TK2) are associated with mtDNA depletion in patients. However, for the majority of patients with mtDNA depletion syndromes (MDS) the genetic defect causing the syndrome remains to be identified. It is known that the yeast mitochondrial pyrophosphatase is necessary for maintained mtDNA content in the yeast cells, and the human mitochondrial pyrophosphatase would be a candidate gene for MDS. Biochemical properties of mammalian mitochondrial pyrophosphatases have been studied on enzyme purified from tissues, but cloning of the gene encoding the enzyme has not been reported previously. Based on sequence similarity to other pyrophosphatases we identified the cDNA encoding the human mitochondrial pyrophosphatase. We cloned the enzyme and showed that it encoded a functional N-terminal mitochondrial targeting signal. The recombinant enzyme was active and ubiquitously expressed with highest levels in tissues rich in mitochondria such as muscle, liver and kidney. The ubiquitous expression suggests that the mitochondrial pyrophosphatase, like the cytosolic pyrophosphatase, is involved in "house-keeping" hydrolyzing of pyrophosphate which is generated by different metabolic processes in the cells. To test if the human mitochondrial pyrophosphatase is required for normal mtDNA copy number, material from more than 50 patients in the USA with unknown cause of MDS is presently being screened for alterations in the gene encoding the mitochondrial pyrophosphate.