Effects of 2-deoxyglucose on drug-sensitive and drug-resistant human breast cancer cells: toxicity and magnetic resonance spectroscopy studies of metabolism.

The glycolytic inhibitor 2-deoxyglucose (2-DG) was tested as a potential chemotherapeutic agent for drug-resistant cancer cells. Previously it was found that Adriamycin-resistant human MCF-7 breast cancer cells (ADR) exhibit an enhanced rate of glycolysis compared to their parent wild-type (WT) cell line (R. C. Lyon et al., Cancer Res., 48: 870-877, 1987). We now describe a specific toxic effect of 2-DG on the ADR cells, which is more than 15-fold greater than for WT cells. Using 31P magnetic resonance spectroscopy of perfused MCF7 cells we continuously monitored the accumulation of 2-deoxyglucose 6-phosphate together with concomitant changes in other phosphate-containing metabolites. Kinetic measurements demonstrated that ADR cells accumulated 2-deoxyglucose 6-phosphate faster and to a greater extent than WT cells, while their depletion of high energy compounds (ATP, phosphocreatine) was more pronounced and became irreversible earlier. The phosphorylation of 2-DG could be followed more effectively by the use of 13C magnetic resonance spectroscopy of 2-DG enriched with 13C at C-6, since the signals of 2-DG and 2-deoxyglucose 6-phosphate are clearly resolved and, unlike 31P magnetic resonance spectroscopy, there are no other interfering signals. With the use of this technique with ADR and WT cells the rate of phosphorylation of 2-DG was found to be 11.2 x 10(-4) and 6.5 x 10(-4) mmol/min/mg protein, respectively. The results of these studies indicate that differences in the biochemistry of energy metabolism of resistant cells may make them targets for energy antimetabolites.

31P-NMR spectroscopy of human cancer cells proliferating in a basement membrane gel.

We describe a system in which proliferating human breast cancer cells are monitored by NMR spectroscopy for at least 6 days in basement membrane gel (BMG)1 threads. The cells are perfused under standard sterile cell culture conditions. 31P-NMR spectra obtained continuously for up to 64 h showed an increase in the signals owing to an increasing number of cells. Cell division in the BMG is easily observed by microscope or by the human eye as the gel opacifies. Spectra of cells in the BMG threads at 20% confluency show a more rapid signal increase than at 60% confluency. Cells grown in vivo in nude mice show a spectrum markedly similar to in vitro spectra in BMG threads, whereas the same cells in agarose threads lack peaks owing to Pi, glycerophosphocholine, and glycerophosphoethanolamine. With the high resolution obtained from this system we distinguished intracellular from extracellular Pi in vitro, and found that the intracellular pH is equal to that observed in the same cell line in vivo. This cell-BMG system is in effect a model tumor, but it is composed of a homogeneous cell population that can be observed indefinitely as the cells reproduce. The material needed is inexpensive, the technique is simple and efficient, and no adaptation of the spectrometer is required. This model will be useful for studying intracellular metabolism and the interaction of cells with the basement membrane.