The development of new diagnostic/therapeutic modalities for cancer requires a specific understanding of how tumors differ from normal tissues. Though the key components involved in the selective accumulation of 2-deoxy-D-glucose (2-DG) analogs in tumors are known, the relative importance of each is controversial. For this reason glucose transport protein (GLUT) density, hexokinase/glucose-6-phosphatase (GP) activity, and 2-DG biodistribution were measured together in four tumor models and normal murine tissues. Direct binding studies with 3H-cytochalasin B showed that GLUT density was elevated 20-fold in LX-1 tumors. Immunohistochemically in all tumors, the expression of GLUT-1 was highest in the necrotic/ perinecrotic foci and similar in cells not adjacent to necrotic foci. As the retention of 3H-2-DG was similar in all tumors, these data suggest that the GLUT-1 in perinecrotic tumor cells were not rate limiting for 3H-2-DG uptake. Kidney, liver, and lung had high GP activity and rapid clearance of 3H-2-DG. Sodium orthovanadate (5 mumol), a GP inhibitor, increased the concentration of 3H-2-DG in these tissues, suggesting that GP is a rate-limiting enzyme for 3H-2-DG clearance. All tumor homogenates had low GP activity, and hexokinase activity was not elevated compared to normal tissues. Thus, in the tumors studied, the selective accumulation of 3H-2-DG consistently occurred in the absence of significant GP activity without the marked overexpression of hexokinase or GLUT.