Glycogen synthesis and degradation. (A) The synthesis of glycogen involves primarily the action of GS, which catalyzes the addition of α-1,4-linked glucose to the nonreducing end of the growing glycogen chain, and the branching enzyme (not shown), which introduces α-1,6-linked ramifications. In fungi and animal cells, autoglucosylation of glycogenin (Gibbons et al, 2002) provides the initial glycogen primer for further elongation. (B) Glycogen breakdown involves the action of GP, which catalyzes the phosphorolysis of α-1,4 glucose linkages, and debranching enzyme (not shown), which hydrolyzes α-1,6 linkages. In fungi and animal cells, both GS and GP are regulated by allosteric ligand binding and phosphorylation.

GS structure. (A) Cartoon showing the general fold and secondary structure organization of AtGS. The color ramp from blue to red indicates N- to C-terminal direction. (B) Cartoon representing the topology of AtGS, color-coded as in (A). Secondary structural elements, as defined by PROCHECK (Laskowski et al, 1993), are shown as arrows for β-strands and cylinders for α-helices comporting more than four residues. (C) Molecular surface representation comparing the structure of the experimental ‘open' state of AtGS (solid) and the predicted ‘closed' conformation (transparent). The orientation is similar to (A). (D) Superposition of the C-terminal domains from the four crystallographically independent AtGS molecules (unliganded: blue and green; ADP-bound: red and yellow). The predicted closed conformation is also shown in pale blue.

ADP-binding pocket. (A) Molecular surface representation of AtGS C-terminal domain (monomer B) showing the ADP-binding pocket. The nucleotidic ligand was reproducibly bound with higher occupancy to monomer B in the asymmetric unit; further analyses in the text hence refer only to this monomer. (B) Schematic representation showing the interactions between AtGS residues and ADP. Different colors distinguish between residues actually observed in the AtGS–ADP crystal complex (in red) and those residues predicted to interact after the ‘closure' movement (in blue). Observed hydrogen bonding and ionic interactions are depicted with dashed lines indicating the distances in Å. Van der Waals interactions are marked with symbols. (C) Electron density map (contoured at 1σ) of the bound nucleotide.

Homology to GP. (A) Cartoon representing the secondary structure organizations of AtGS (shown in green) and E. coli MalP (in blue) in the same orientation. After structural alignment, superimposed equivalent residues are represented in solid material, while in transparent are shown the regions that are not aligned. (B) Histograms showing the number of equivalent residues after pairwise structural alignment between AtGS and several GT-B enzymes: maltodextrin phosphorylase (MalP, 1QM5), glycogen phosphorylase (GP, 1A8I), trehalose-6-phosphate synthase (OtsA, 1GZ5), UDP-N-acetylglucosamine 2-epimerase (epim., 1F6D), UDP-glucosyl transferase (Gtfb, 1IIR), UDP-N-acetylglucosamine-N-acetylmuramyl-(pentapeptide)pyrophosphoryl-undecaprenol N-acetylglucosamine transferase (MurG, 1NLM) and DNA-β-glucosyltransferase (Bgt, 1JG6). The N-terminal and C-terminal domains were aligned separately. At the top of each bar, the calculated r.m.s. (in Å) of aligned Cα positions is shown. (C) Comparative representation of the catalytic sites of GS in its closed conformation (left view) and a ternary complex of MalP with inorganic phosphate and GSG4, a nonhydrolyzable analog of maltopentaose (right view). The sites are shown in the same orientation after structural superposition of both structures. Atoms are colored red (O), blue (N), purple (P), gray (protein C) and yellow (ligand C). The glucose acceptor oligosaccharide (as seen in MalP) is shown in green in the AtGS structure. (D) Multiple alignment of selected regions among members of the GS (GT3 and GT5) and GP (GT35) families. Atu: A. tumefaciens; Ec: E. coli; Ath: Arabidopsis thaliana; Os: Oriza sativa; Pa: Pyrococcus abysii; Tm: Thermotoga maritima; Hsl: Homo sapiens (liver); Hsm: H. sapiens (muscle); Sc2: Saccharomyces cerevisiae (Gys2); Oc: Oryctolagus cuniculus. Nonaligned regions for GT3 enzymes are depicted with a gray box. The secondary structure of AtGS is shown as a reference at the bottom.

Structural alignment of AtGS, E. coli MalP and rabbit muscle GP. Structurally equivalent regions are boxed, conserved residues are shown in red. Secondary structural elements (α-helices and β-strands) are shown above the sequence for AtGS and below the sequence for GP. Every tenth residue in the AtGS and GP sequences is dotted.

Molecular surface representation of the GS core, showing the equivalent position of the arginine clusters in the mammalian/yeast (GT3) allosteric site (in red) with respect to the active center. Assuming an extended main-chain conformation, approximate distances are shown for two relevant phosphorylation sites, one in the N-terminal (2a) and the other in the C-terminal (3a) extensions of GT3 enzymes.