Despite dramatic increases in glucose influx during the transition from fasting to fed states, plasma glucose concentration remains tightly controlled. This constancy is in large part due to the capacity of skeletal muscle to absorb excess glucose and store it as glycogen. The magnitude of this capacity is controlled by insulin by way of regulated insertion of glucose transporters into the muscle cell membrane. Here, we examine the mechanism by which muscle cells are able to tolerate large flux increases across their transporters without significantly changing their own metabolite pools. MCA was used to probe data sets that measured the effects of changing plasma glucose and͞or insulin concentrations on the rates of glycogen synthesis and the concentrations of metabolites, particularly glucose-6-phosphate. We find that homeostasis is achieved by insulin-dependent phosphorylation changes in GSase sensitivity to the upstream metabolite glucose-6-phosphate. The centrality of GSase to homeostasis resolves the paradox of its sensitivity to allosteric and covalent regulation despite its minimal role in flux control. The importance of this role for enzymatic phosphorylation to diabetes pathology is discussed, and its general applicability is suggested.M etabolic control analysis (MCA) has revolutionized our understanding of flux control through theoretical and experimental studies (1-4). One of its key precepts is the flux control coefficient, a quantitative characterization of the influence of any enzyme over the flux of a metabolic pathway. As MCA has matured, it has expanded into associated areas. One of particular interest has been the mechanism of metabolite control, in which the central question is how metabolite homeostasis, the constancy of metabolite concentrations, is maintained during changes in flux (5-9). One well known example of such metabolite homeostasis comes from the study of fluxes through the energetic pathway of glucose oxidation. There, the flux can change by more than an order of magnitude whereas metabolites remain constant within small experimental error (10, 11). Although theoretical analyses of such homeostatic phenomena have been available since the first description of MCA, physical exploration of this important physiological area has been scarce due to experimental limitations. The primary hurdle has been the difficulties in simultaneously obtaining in vivo values of metabolite concentrations and of fluxes through the pathway.These limitations have been somewhat eased by the development of in vivo NMR methods. Valuable information about energetics has been obtained by 31 P NMR of high-energy metabolites (e.g., ATP, PO 4 , and creatine phosphate) and about metabolic regulation from phosphorylated pathway intermediates such as glucose-6-phosphate (G6P) (12). In complementary experiments, the weak natural abundance (1.1%) of the 13 C isotope allows direct 13 C NMR of labeled 13 C substrates to measure flux, by following the time course of label flow into metabolic pools (13). Comprehensive a...