Brain slice glucose utilization.

Abstract

The metabolism of 2-deoxyglucose has been studied in 540 micron and 1,000 micron hypothalamic brain slices. Slice 2-deoxyglucose (2DG) and 2-deoxyglucose-6-phosphate (2DG6P) levels were measured after tissue homogenization and perchloric acid extraction. By analyzing the uptake and washout kinetics with nonlinear least-squares methods, we have determined the rate constants for three-, four-, or five-parameter kinetic models and obtained a value for the in vitro lumped constant (LC). The kinetic analysis reveals a small, slowly decaying, 2DG component that is not predicted by any of the models. If this component is treated as a separate, parallel compartment, then the four- and five-parameter models are essentially equivalent. To compare our data to prior in vivo data, we combined 2DG and 2DG6P to produce Ci*, the total slice radioactivity, and analyzed the first 45 min of uptake. These data were fit best by a three-parameter model and the slowly decaying pool was not identified. Calculation of glucose utilization from total tissue radioactivity, measured by whole slice homogenization and by image analysis of autoradiograms, showed excellent correlation between the two methods. Image analysis of radioactivity in the suprachiasmatic nucleus, which is present in these slices, revealed a spontaneous diurnal variation in in vitro glucose utilization in close quantitative agreement with prior in vivo measurements. The kinetic analysis of the 1,000 micron slice was qualitatively similar to that of the 540 micron slice but revealed an increase in the LC and a large decrease in k1 as well as the expected large increase in the hexokinase rate constant, k3. Overall, in vitro glucose utilization increased by about 60%. These results are consistent with our prior studies of the 1,000 micron slice and support our interpretation that the 1,000 micron slice is an excellent in vitro model for brain ischemia without infarction.