Glucose is the main fuel for energy metabolism in the normal human brain. It is generally a ed that glucose transport into the brain is not rate-limiting for metabolism. Since brain glucose concentrations cannot be determined directly by radiotracer techniques, we used 13C NMR spectroscopy after infusing enriched D-[1-_3C]glucose to measure brain glucose concentrations at euglycemia and at hyperglycemia (range, 4.5-12.1 mM) in six healthy children (13-16 years old). Brain glucose concentrations averaged 1.0 ± 0.1 ,umol/ml at euglycemia (4.7 ± 0.3 mM plasma) and 1. 8-2.7 jimol/ml at hyperglycemia (7.3-12.1 mM plasma). MichaelisMenten parameters of transport were calculated to be Kt = 6.2 ± 1.7 mM and T.. = 1.2 ± 0.1 zmol/gimin from the relationship between plasma and brain glucose concentrations. The brain glucose concentrations and transport constants are consistent with transport not being rate-limiting for resting brain metabolism at plasma levels >3 mM.Under normal conditions, brain cells rely almost entirely on the availability of D-glucose for generation of energy. The lack of significant carbohydrate stores in the brain requires that a steady glucose supply from the blood be maintained. However, D-glucose cannot diffuse freely across the bloodbrain barrier but is transported by facilitated diffusion (1-4) mediated by specific transporter molecules (5-7). The importance of glucose transport to metabolism and brain function has spurred extensive research in animals (for reviews, see refs. 8 and 9), as well as in humans (10,11). Although the rate of glucose transport is considered to be rapid enough to meet the cellular energy demand under basal conditions, it may become rate-limiting during hypoglycemia (11,12). Alterations in brain glucose transport may also be involved in the central nervous system morbidity associated with diseases such as diabetes mellitus, seizures, or hypoxicischemic encephalopathy (13,14).Glucose transport into the brain can be evaluated by measuring brain glucose concentrations as a function of plasma levels (15, 16). Direct measurement of glucose concentration is possible in animal brain with rapid-freezing techniques (17,18). However, these in vitro methods cannot be applied to humans, and direct quantitation of human brain glucose concentration has thus far not been achieved.NMR spectroscopy has been shown to be a powerful noninvasive technique that can be used to determine the concentration of many metabolites in living tissue (19)(20)(21)(22)(23).Recent studies have shown that the a-and ,B-anomers of glucose can be measured in animal brain by 13C NMR, when D-[1-'3C] glucose is infused (16,24). We report here the quantitation of the glucose concentration in the human brain using localized 13C NMR spectroscopy and the determination of the kinetic parameters of D-glucose transport from the relationship between brain and plasma glucose concentration. METHODSSix nonobese healthy children (13-16 years old) were studied after written informed consent was obtained from the su...

No abstract

Cerebral metabolism of D[1-"C]glucose was studied with localized '3C NMR spectroscopy during in-

The actions of recombinant human insulin-like growth factor-I (rhIGF-I) and insulin were compared in 21 healthy young (24±1 yr) and 14 healthy middle-aged (48±2 yr) subjects during 3-h paired euglycemic clamp studies using one of three doses (rhIGF-I 0.2, 0.4, and 0.8 ag/kg -min and insulin 0.2, 0.4, and 0.8 mU/kg. min, doses chosen to produce equivalent increases in glucose uptake). In younger subjects, rhIGF-I infusions suppressed insulin by 19-33%, C-peptide by 47-59% and glucagon by 33-47% (all, P < 0.02). The suppression of C-peptide was less pronounced with insulin than with rhIGF-I (P < 0.007). The metabolic responses to rhIGF-I and insulin were remarkably similar: not only did both hormones increase glucose uptake and oxidation in a nearly identical fashion, but they also produced similar suppression of glucose production, free fatty acid levels, and fat oxidation rates. In contrast, rhIGF-I had a more pronounced amino acid-lowering effect than did insulin (P < 0.004).In middle-aged subjects, basal IGF-I levels were 44% lower (P < 0.0001) whereas basal insulin and C-peptide were 20-25% higher than in younger subjects. Age did not alter the response to rhIGF-I. However, insulin-induced stimulation of glucose uptake was blunted in older subjects (P = 0.05). Our data suggest that absolute IGF-I and relative insulin deficiency contribute to adverse metabolic changes seen in middle age. (J.