BRAIN GLUCOSE-LEVELS IN PORTACAVAL-SHUNTED RATS WITH CHRONIC, MODERATE HYPERAMMONEMIA - IMPLICATIONS FOR DETERMINATION OF LOCAL CEREBRAL GLUCOSE-UTILIZATION

Rates of glucose utilization (lCMR(glc)) in many structures of the bra in of fed, portacaval-shunted rats, when assayed with the [C-14]deoxyg lucose (DG) method in our laboratory, were previously found to be unch anged (30 of 36 structures) or depressed (6 structures) during the fir st 4 weeks after shunting, but to rise progressively to higher than no rmal values in 25 of 36 structures from 4-12 weeks. In contrast, lCMR( glc), when assayed with the [C-14]glucose method in another laboratory , was depressed in most structures of brains of 4-8-week shunted rats that had relatively high brain ammonia levels. There was a possibility that the increases in lCMR(glc) obtained with the [C-14]DG method may have been artifactual, due, in part, to a change in brain glucose con tent which could alter the value of the lumped constant of the DG meth od. Brain glucose levels of shunted rats were, therefore, assayed by b oth direct chemical measurement in freeze-blown samples and by determi nation of steady-state brain:plasma distribution ratios for [C-14]meth ylglucose; the methylglucose distribution ratio varies as a function o f plasma and tissue glucose contents. Within a week after shunting, am monia levels in blood and brain rose to 0.25-0.30 mM and 0.35-0.70 mu mol/g, respectively, and mean plasma glucose levels fell from 9-10 mM to 7.4-8.5 mM, and then remained nearly constant. Brains of fed-shunte d rats had normal glycogen levels and stable but moderately reduced gl ucose contents between 1 and 12 weeks (i.e., 1.9-2.2 mu mol/g), [C-14] Methylglucose distribution ratios were essentially the same as those i n controls in 22 brain structures at 2 and 8 weeks after shunting. Bec ause brain glucose levels remained stable from 1 to 12 weeks after shu nting, there is no evidence to support the hypothesis that the value o f the lumped constant would have changed and caused an artifactual ris e in lCMR(glc).