Effect of changing tumor oxygenation on glycolytic metabolism in a murine C3H mammary carcinoma assessed by in vivo nuclear magnetic resonance spectroscopy.

The rate of conversion of D-[1-(13)C]glucose into [3-(13)C]lactate (apparent glycolytic rate) has been determined in C3H murine mammary carcinomas in vivo using tumor-selective (13)C nuclear magnetic resonance spectroscopy with (1)H-(13)C cross-polarization. Under conditions of acute hypoxia induced by breathing carbon monoxide at 660 ppm, the apparent glycolytic rate was 0.0239 +/- 0.0019 min(-1). The proportion of (13)C label incorporated into [4-(13)C]glutamate (measured in tumor extracts) was 25-fold lower than that incorporated into [3-(13)C]lactate, reflecting a very limited oxidative metabolism during this hypoxic episode. For animals breathing air or carbogen (95% O(2) + 5% CO(2)), the calculated glycolytic rates were correspondingly lower (0.0160 +/- 0.0021 min(-1) and 0.0050 +/- 0.0011 min(-1), respectively). Although (13)C labeling of glutamate at C4 was still an order of magnitude lower than that for lactate at C3 (11-fold for air and 9-fold for carbogen), these ratios did show a greater degree of oxidative metabolism than that seen in animals breathing carbon monoxide at 660 ppm. The marked difference in apparent glycolytic rate for this tumor model between well-oxygenated and hypoxic conditions demonstrates a substantial Pasteur effect (inhibition of glycolysis by oxygen). Dynamic (13)C nuclear magnetic resonance spectroscopy provides a noninvasive estimate of tumor glycolysis that can be used to evaluate the relationship between oxygenation and energy metabolism, and this has potential consequences for the sensitivity of hypoxic cells to treatment and their ability to promote angiogenesis.