Introduction Inducible 6-phosphofructo-2-kinase (iPFK-2; PFKFB3) has been found to be overexpressed by the neoplastic cells of human solid tumors, required for human leukemia growth in athymic mice and to be a transcriptional target of hypoxia inducible factor 1, a protein required for the survival and growth of tumors. Based on these observations, the substrate (fructose-6-phosphate) binding site of iPFK-2 is an attractive target for the identification of pharmacophore-based compounds as antineoplastic agents.
Methods/Results The x-ray structures of human liver and rat testes PFKFB isozymes served as input homologue sequences and structural templates for the determination of structurally conserved regions and the fructose-6-phosphate (F6P) binding site of iPFK-2. The Advanced Chemical Directory (v.2002-2) then was computationally screened for high fit molecules of the F6P binding site of iPFK-2 and 45 compounds were identified. We exposed Jurkat human T cell leukemia cells to 6 of the compounds (termed MT1-6) at concentrations of 100 nM-1 mM, and then enumerated live and dead cells by propidium iodide exclusion staining after 48-72 hours of exposure. Whereas 5 compounds exhibited cytotoxic effects at a concentration of 1 mM, compound MT3 was cytotoxic at a concentration of 10 μM and cytostatic at a concentration of 1 μM in vitro. Growth inhibitory properties of MT3 were also observed against A459 human lung adenocarcinoma cells, mouse Lewis lung carcinoma cells, and mouse B16 melanoma cells. Exposure of MT3 to Jurkat cells suppressed (i) the product of iPFK-2, F2,6BP; (ii) lactate secretion; and (iii) incorporation of glucose-derived carbons into pyrimidine but not purine riboses via the pentose shunt. Toxicity studies in C57BL/6 mice indicated that daily intraperitoneal administration of 0.5 mg of compound MT3 was well tolerated. We subcutaneously injected 8 × 105 Lewis lung carcinoma cells (in 100 μL PBS) into C57Bl/6 mice and, after tumors were established, administered 0.5 mg MT3 or vehicle i.p. every 24 hours. After 14 days we observed a reduction in tumor growth in mice that were administered daily MT3 (1042.8 ± 192.5 mm3) compared to vehicle alone (2034.78 ± 395.18 mm3; p value < .0001).
Conclusions We conclude that pharmacophore targeting of the substrate binding site of iPFK-2 may prove to be a useful strategy in the development of novel antineoplastic agents. The selective reduction of 13C-glucose incorporation into pyrimidine riboses suggests that MT3 may decrease the availability of aspartate, a product of the citric acid cycle (via oxaloacetate) and a precursor of the pyrimidine ring.