Summary:Diabetes is associated with extensive microvascular pathology and decreased expression of the glucose transporter (GLUT-1) in retina, but not brain. To explore the basis of these differences, the authors simultaneously measured glucose influx (mol · g −1 · min) and blood flow (mL · g −1 · min −1 ) in retina and brain cortex of nondiabetic control rats (normoglycemic and acute-hyperglycemic) and in rats with streptozotocin-induced diabetes (with or without aminoguanidine (AMG) treatment) using a single-pass, dual-label indicator method. In addition, tissue glucose and adenosine triphosphate (nmol/mg dry weight) levels were measured. Glucose influx in retina exceeded that of cortex by about threefold for both the nondiabetic and diabetic groups. In contrast, blood flow in retina was significantly lower than in cortex by about threefold for each group. Retinal and cortical glucose influx in the diabetic rats was lower than in the nondiabetic acute-hyperglycemic group, but not in the normoglycemic group. Blood flow in these tissues remained relatively unchanged with glycemic conditions. The glucose levels in the diabetic retina (aminoguanidine untreated and aminoguanidine treated) were fourfold to sixfold greater than the nondiabetic retina. The cortical glucose levels remained unchanged in all groups. These data suggest that the accumulation of glucose in the diabetic retina cannot be explained by increased endothelialglucose uptake or increased vascular membrane permeability.

It is recognized that brain oxygen deprivation results in increased glycolysis and lactate accumulation. Moreover, glucose metabolism is altered during starvation or diet, resulting in increased plasma ketones (acetoacetate + beta-hydroxybutyrate; BHB). We investigated glucose and lactate adaptation to hypoxia in concurrence with diet-induced ketosis. Male Wistar rats were fed standard (STD), ketogenic (high fat; KG), or carbohydrate-rich (low fat; CHO) diets for 3 wks and then exposed to hypobaric (0.5 ATM) or normobaric atmosphere for 3 wks while on their diets. Lactate, ketones, and glucose concentrations were measured in plasma (mM) and brain tissue (mmol/g). Plasma and tissue ketone levels were elevated up to 12-fold in the KG fed groups compared with other groups (STD and CHO), with the hypoxic KG group reaching the highest levels (2.6 +/- 1.3 mM and 0.3 +/- 0.1 mmol/g; mean +/- SD). Tissue lactate levels in the hypoxic ketotic rats (4.7 +/- 1.3 mM) were comparable with normoxic STD (5.0 +/- 0.7 mM) and significantly lower (ANOVA P < .05) than the hypoxic STD rats (6.1 +/- 1.0 mM). These data indicate that adaptation to hypoxia did not interfere with ketosis, and that ketosis during hypoxia may lower lactate levels in brain, suggesting decreased glycolysis or increased glucose disposal.

The benefits of a mouse model are efficiency and availability of transgenics/ knockouts. Quantitation of cerebral blood in small animals is difficult because the cannulation procedure may introduce errors. The [14C]-iodoantipyrine autoradiography (IAP) method requires both the tissue concentration and the time course of arterial concentration of the [14C] radioactive tracer. A single point-analysis technique was evaluated for measuring blood flow in mice (30 g +/- 0.3 g; n = 11) by using computational models of sensitivity analysis, which quantitates relationships between the predictions of a model and its parameters. Using [14C]-IAP in conjunction with mathematical algorithms and assumed arterial concentration-versus-time profiles, cortical blood flow was deduced from single-point measurements of the arterial tracer concentration. The data showed the arterial concentration profile that produced the most realistic blood flows (1.6 +/- 0.4; mean +/- SD, ml/g/min) was a profile with a ramp time of 30 sec followed by a constant value over the remaining time period of 30 sec. Sensitivity analysis showed that the total experimental time period was a more important parameter than the lag period and the ramp period. Thus, it appears that the accuracy of the assumption of linearly increasing arterial concentration depends on the experimental time period and the final arterial [14C]-iodoantipyrine concentration.