Mapping of a Regulatory Site of the Escherichia coli ADP-Glucose Pyrophosphorylase

Abstract

The enzyme ADP-glucose pyrophosphorylase (ADP-Glc PPase) controls the biosynthesis of glycogen in bacteria and starch in plants. It is regulated by various activators in different organisms according to their metabolic characteristics. In Escherichia coli, the major allosteric activator is fructose 1,6-bisphosphate (FBP). Other potent activator analogs include 1,6-hexanediol bisphosphate (HBP) and pyridoxal 5'-phosphate (PLP). Recently, a crystal structure with FBP bound was reported (PDB ID: 5L6S). However, it is possible that the FBP site found is not directly responsible for the activation of the enzyme. We hypothesized FBP activates by binding one of its phosphate groups to another site ("P1") in which a sulfate molecule was observed. In the E. coli enzyme, Arg40, Arg52, and Arg386 are part of this "P1" pocket and tightly complex this sulfate, which is also present in the crystal structures of ADP-Glc PPases from Agrobacterium tumefaciens and Solanum tuberosum. To test this hypothesis, we modeled alternative binding conformations of FBP, HBP, and PLP into "P1." In addition, we performed a scanning mutagenesis of Arg residues near potential phosphate binding sites ("P1," "P2," "P3"). We found that Arg40 and Arg52 are essential for FBP and PLP binding and activation. In addition, mutation of Arg386 to Ala decreased the apparent affinity for the activators more than 35-fold. We propose that the activator binds at this "P1" pocket, as well as "P2." Arg40 and Arg52 are highly conserved residues and they may be a common feature to complex the phosphate moiety of different sugar phosphate activators in the ADP-Glc PPase family.

Keywords: allosteric regulation; arginine scanning mutagenesis; polysaccharide biosynthesis; pyridoxal 5′-phosphate activation; sugar phosphate regulation.

Figure 1

The regulation and synthesis of glycogen in E. coli. ADP-Glc PPase regulates the production of glycogen in bacteria. ADP-Glc PPase catalyzes the reaction between Glc1P and ATP in the presence of Mg²⁺, forming ADP-Glc and PPi. This reaction takes place under physiological conditions; however, the reaction is reversible in vitro. ADP-Glc PPase is activated by FBP and inhibited by AMP. The production of ADP-Glc leads to the synthesis of a linear α-1,4 glucan molecule by glycogen synthase, which is then branched with an α-1,6 linkage by the branching enzyme.

Figure 2

FBP saturation curves for the WT E. coli ADP-Glc PPase in the presence of varying concentrations of sulfate. Activity was measured in the presence of 1.0 mM Glc1P, 1.0 mM ATP, 7 mM MgCl₂, and 50 mM HEPES pH 7.5. The effector and sodium sulfate concentration were varied as indicated in the figure. The inset graph represents the linear trend between the calculated A_0.5 of FBP from each curve and the concentration of sulfate.

Figure 3

Structure of E. coli ADP-Glc PPase and positioning of various ligands. (A) Ribbon representation of homotetrameric E. coli ADP-Glc PPase in the presence of FBP and sulfate (PDB ID: 5L6S). (B) Localization of the FBP-binding site. P1 shows the accessible surface area of the sulfate in 5L6S. P2 and P3 designate the accessible surface area of modeled phosphates equivalent to the phosphate groups of FBP from 5L6S. (C) HBP modeled into P1 and P2. (D) The linear form of FBP modeled into the putative FBP binding site comprising P1 and P2. The 1- and 6-phosphate groups were modeled into P2 and P1, respectively. (E) PLP modeled into the putative regulatory binding site, forming a Schiff base with Lys39 and the phosphoryl moiety at P1.

Figure 4

Effect of FBP on the apparent affinity for ATP in the WT and mutants of the E. coli ADP-Glc PPase. The effect of the FBP on the ATP apparent affinity was assessed by the ratio of two different S_0.5 for ATP. One was measured in absence (control) and the other in presence of 1 mM FBP. Assays were performed in the presence of 1 mM Glc1P, 10 mM MgCl₂, and 50 mM HEPES pH 8.0 as indicated in section Materials and Methods.

Figure 5

Effect of FBP on WT, R423A, and R52A E. coli ADP-Glc PPase activity. Activity was measured as described in section Materials and Methods in the presence of 1.0 mM Glc1P, 1.5 mM ATP, 10 mM MgCl₂, and 50 mM HEPES pH 8.0. The effector concentration was varied as indicated in the figure.

Figure 6

Effect of PLP on WT, R423A, and R52A E. coli ADP-Glc PPase activity. Activity was measured as described in Materials and Methods in the presence of 1.0 mM Glc1P, 1.5 mM ATP, 10 mM MgCl₂, and 50 mM HEPES pH 8.0. The effector concentration was varied as indicated in the figure.

Figure 7

Effect of FBP on the thermal shift assays for the WT and mutants of ADP-Glc PPase. Thermal shift assays were performed as described in Materials and Methods in presence and absence (control) of FBP as indicated.

Figure 8

Effect of mutagenesis on the allosterism of the E. coli ADP-Glc PPase. Arg residues depicted in the figure were mutated to Ala and characterized as described in the text. In red (Arg40, Arg52), are residues with negligible activation after mutagenesis. In orange (Arg386), is the residue that displayed a significantly lower apparent affinity for FBP and PLP (37- and 56-fold, respectively). In yellow (Arg423), is the residue that displayed a near WT maximum activation but had 12-fold lower apparent affinity for FBP and higher activity with PLP (only 2.6-lower apparent affinity). In cyan, are Arg residues that did not display dramatic effects on the allosteric activation when they were mutated to Ala. P1, P2, and P3 are surfaces depicting putative binding sites for phosphate groups.

Figure 9

Sequence alignment of ADP-Glc PPases from different species. Amino acid alignment of ADP-Glc PPases from selected species were processed as described in section Materials and Methods. The accession numbers were attained from the NCBI database. Eco, Escherichia coli (P0A6V1.2); Atu, Agrobacterium tumefaciens (P39669.1); Neu, Nitrosomonas europaea (Q82T88.1); Nos, Nostoc sp. PCC 7120 (P30521.1); Sco, Streptoyces coelicolor (CAA61885); Gst, Geobacillus stearothermophilus (O08326.1); Ral, Ruminococcus albus (WP_024858442.1); Stu (L), Solanum tuberosum (large subunit) (P55242.1); Stu (S), Solanum tuberosum (small subunit) (P23509.2). Highlighted regions display the conservation of amino acids amongst the selected organisms and the black box shows N-terminal residues involved in phosphate binding at the P1 site.