Allosteric regulation of mammalian fructose-1,6-bisphosphatase

Fructose-1, 6-bisphosphate (D-fructose-1, 6-bisphosphate 1-phosphohydrolase; EC 3. 1. 3; FBPase) is an essential regulatory enzyme in gluconeogenesis and has long been considered as a drug target towards type II diabete. In mammalian, the activity of FBPase is regulated by AMP and fructose 2,6-bisphosphate (Fru-2,6-P2). AMP is an allosteric inhibitor that binds to FBPase with positive coopertivity, and Fru-2,6-P2 is an active site inhibitor which is up-regulated by hormone. Despite the 30 Å distance between their binding sites, both of AMP and Fru-2,6-P2 transform FBPase from active R-state to inactive T-state. Large conformational rearrangements are coupled to the R- to T-state transition: subunit pairs within tetrameric FBPase rotate over ten degree relative to each other and an essential catalytic loop (residue 50-72) is forced away from active site. Mutagenesis, kinetics, crystallography and molecular dynamics simulations are combined here to investigate structure-function relationship of FBPase. Tetramer is a functional unit of FBPase; disturbing the tetrameric packing of FBPase leads to loss of AMP cooperativity. A hydrophobic cavity at the center of FBPase tetramer is populated by well-defined clathrate-like waters. The cavity together with waters in it is shown to be thermodynamic determinant for quaternary states of FBPase. Kinetics and crystallographic studies indicate a negative correlation between subunit pair rotation and relative activity of FBPase. Filling the cavity by point mutations selectively hinges subunit pair rotation induced by Fru-2,6-P2 and largely reduce the synergism between AMP and Fru-2,6-P2. Mutation that stops subunit pair rotations causes complete loss of AMP inhibition but retains Fru-2,6-P2 inhibition; whereas mutation promotes the subunit pair rotation turn off cooperative binding of AMP as well as AMP/Fru-2,6-P2 synergism. MD simulation together with crystal structures of intermediate states of FBPase reveals correlation between subunit pair rotation and status of loop 50-72. Moreover, knowledge of allosteric regulation of porcine liver FBPase and FBPase from Escherichia coli was used predict the regulatory properties of all Type-I FBPases, for which sequence information is available. Subsequent expression of FBPase from a bacterial organism, predicted to have the regulatory properties of a eukaryotic FBPase, proved correct and established a basis for the evolution of regulatory properties for all Type-I FBPases