Ternary Complex Crystal Structures of Glycogen Phosphorylase with the Transition State Analogue Nojirimycin Tetrazole and Phosphate in the T and R States^,

Abstract

Catalysis by glycogen phosphorylase involves a mechanism in which binding of one substrate tightens the binding of the other substrate to produce a productive ternary enzyme-substrate complex. In this work the molecular basis for this synergism is probed in crystallographic studies on ternary complexes in which the glucosyl component is substituted by the putative transition state analogue nojirimycin tetrazole, a compound which has been established previously as a transition state analogue inhibitor for a number of glycosidases. Kinetic studies with glycogen phosphorylase showed that nojirimycin tetrazole is a competitive inhibitor with respect to glucose 1-phosphate and uncompetitive with respect to phosphate. Ki values for the phosphorylase-AMP-glycogen complex and the phosphorylase-AMP-glycogen-phosphate complexes are 700 microM and 53 microM, respectively, indicating that by itself norjirimycin tetrazole has poor affinity for glycogen phosphorylase but that phosphate substantially improves the binding of norjirimycin tetrazole. X-ray crystallographic binding studies to 2.4 A resolution with T state phosphorylase b crystals showed that nojirimycin tetrazole binds at the catalytic site and promotes the binding of phosphate through direct interactions. Phosphate binding is accompanied by conformational changes that bring a crucial arginine (Arg569) into the catalytic site. The positions of the phosphate oxygens were definitively established in X-ray crystallographic binding experiments at 100 K to 1.7 A resolution using synchrotron radiation. X-ray crystallographic binding studies at 2.5 A resolution with R state glycogen phosphorylase crystals showed that the protein atoms and water molecules in contact with the nojirimycin tetrazole and the phosphate are similar to those in the T state. In both T and R states the phosphate ion is within hydrogen-bonding distance of the cofactor pyridoxal 5'-phosphate group and in ionic contact with the N-1 atom of the tetrazole ring. The results are consistent with previous time-resolved structural studies on complexes with heptenitol and phosphate. The structural and kinetic results suggest that nojirimycin tetrazole in combination with phosphate exhibits properties consistent with a transition state analogue and demonstrate how one promotes the binding of the other.

Activator anion binding site in pyridoxal phosphorylase b: The binding of phosphite, phosphate, and fluorophosphate in the crystal

“…The interacting phosphates hypothesis was confirmed crystallographically in the analysis of the transition-state intermediate-like GPb-heptulose 2-P complex (McLaughlin et al, 1984;Hajdu et al, 1987;Johnson et al, 1990), which showed a close P-P distance (of phosphorus atoms of phosphate to cofactor 5"phosphate) of 4.8 A. These studies and recent structural studies on the T-and R-state GPb-NJT-phosphate complexes (Mitchell et al, 1996) have definitively established the location of the substrate phosphate at or near the transition state of the reaction and are consistent with the proposed role for the cofactor 5"phosphate as a general acid.…”

“…The close proximity of the product phosphate to the cofactor 5"phosphate provides support for a mechanism in which phosphorolysis of heptenitol is catalyzed by general acid attack of the substrate phosphate, promoted by the cofactor 5"phosphate. The recent X-ray work with NJT and phosphate (Mitchell et al, 1996) also show directly that the substrate phosphate binding site is in the correct position for the general acid attack. The proton is donated by the substrate phosphate in a concerted reaction in which it immediately gains a proton from the cofactor 5"phosphate.…”

“…The close approach of two phosphate groups in other enzymes such as the kinases is usually stabilized by magnesium ions . GP does not require magnesium ions although there is a bridging water between the two phosphates in the GPb-NJT-phosphate complex structure (Mitchell et al, 1996). Klein et al (1984) have suggested that a slightly different structure of the PL-GPb-anion enzyme compared with the native holoenzyme might bring in a proton donor group of the enzyme that normally is not in the reaction path.…”

“…Since it was proposed that this com- (Mitchell et al, 1996). NJT was found to be a competitive inhibitor pound might act as a transition-state analogue for the a-glycosidase with respect to Glc-1-P, with a Ki value of 0.70 mM, but it binds strongly to the enzyme-AMP-glycogen-phosphate complex with a showed that both the NJT molecule and the phosphate ion bind Ki value of 53 pM, a result indicating that the complex enzymetightly at the catalytic site.…”

“…This subroutine calculates the RMSD between two sets of atomic coordinates and the rotation matrix and translation vector that best superposes the position vector of two sets of atoms. Coordinate sets for comparison were: T-state GPb (Acharya et al, 1991) (PDB code IGPB), T-state GPb-Glc complex (Martin et al, 1990) (PDB code 2GPB), T-state GPb-1-GlcNAc-IMP complex (Oikonomakos et al, 1995) (PDB code lPRJ), and T-state GPb-phosphate-NJT complex (Mitchell et al, 1996) (PDB code 1NOJ).…”