A nontracer amount (0.25 mmol/kg of body weight) of 2-deoxyglucose (DG) was intravenously injected into rats, which were frozen 2 and 4 min later in liquid nitrogen. Freeze-dried samples of CNS regions and cell bodies of spinal motor neurons were prepared, and the concentrations of glucose, glucose 6-phosphate, DG, and DG 6-phosphate (DG6P) in them were microassayed after 3,000-1,500,000-fold amplification using an enzymatic amplification reaction, NADP cycling. Based on the time course of glucose, DG, and DG6P concentrations in arterial plasma and the anterior horn of the spinal cord, the Sokoloff-type rate equations for DG and DG6P concentrations were mathematically solved, and the resultant DG and DG6P concentration functions were fitted to the data points using the nonlinear least-squares fitting SALS package program. This fitting provided four rate constants for the functions and supported the theoretical basis for our calculations of glucose utilization rate (GUR) when DG was administered in nontracer amounts. The GUR was highest in the spinal motor neurons and lowest in the white matter of the cerebellum. Neuron-rich structures, such as the cerebellar molecular and granular layers and the anterior horn of the spinal cord, had higher GUR values than the white matter of the cerebellum and spinal cord.