Previous studies from other laboratories, using rabbit reticulocyte lysate filtered through Sephadex G-25 or G-50, have demonstrated that glucose 6-phosphate is required to maintain active rates of translation, but its mechanism of action is currently unsettled. We have tested whether glucose 6-phosphate is required to prevent activation of the hemin-controlled translational repressor and the phosphorylation of the smallest or alpha subunit of eukaryotic initiation factor 2 (eIF-2). We have found that antibody to the hemin-controlled translational repressor can completely restore protein synthesis in reticulocyte lysate, filtered through Sephadex G-25, that is incubated in the absence of hemin and presence of glucose 6-phosphate, but cannot restore protein synthesis in such lysate incubated in the presence of hemin and absence of glucose 6-phosphate. We have also found, using a modification of the method of Matts and London [1984) J. Biol. Chem. 259, 6708-6711) to measure the ability of gel-filtered lysate to dissociate and exchange GDP from eIF-2.GDP, that this endogenous eIF-2B activity is reduced to the same low level in the presence of hemin and absence of glucose 6-phosphate as it is in the absence of hemin and presence of glucose 6-phosphate. Although there is a low level of phosphorylation of eIF-2 alpha in gel-filtered lysate given hemin but no glucose 6-phosphate, it cannot account for the loss of eIF-2B activity, since this phosphorylation is removed by antibody to the hemin-controlled translational repressor or isocitrate, which do not restore protein synthesis or eIF-2B activity, and not by fructose 1,6-diphosphate, which does partially restore protein synthesis and eIF-2B activity. These findings suggest that sugar phosphates may exert a direct effect on eIF-2B and may be required for its proper function. Additional support for this conclusion is our finding that protein synthesis and eIF-2B activity in partially hemin-deficient lysate can be restored by high levels of glucose 6-phosphate or fructose 1,6-diphosphate without a reduction in the level of phosphorylated eIF-2 alpha, suggesting that such levels of sugar phosphate may permit restoration of normal function with a limiting amount of eIF-2B.