The structure of the catalytic domain of the ER a-mannosidase I from S. cerevisiaehas recently been determined by X-ray crystallography by a collaboration of the Howell and Herscovics labs (5). The overall structure of the enzyme catalytic domain is an (aa)7 barrel composed of 14 consecutive helices concentrically alternating from the inside to the outside of the barrel structure.

The figure shown at the left displays the structure of yeast ER mannosidase I as determined by Vallee et al (6) in a ribbon form with the N-linked oligosaccharides in wireframe in white (PDB code 1DL2). The barrel has an approximate internal seven-fold symmetry (see figure at the right); however, the two ends of the barrel are not equivalent. On one end of the barrel the inner and outer helices are connected by short loops of up to 4 residues. The opposite end consists of a complex array of b-strands that creates a broad and flattened surface perpendicular to the central axis of the barrel (SEE THIS LINK FOR A CHIME VIEW OF THE STRUCTURE). A COOH-terminal b-hairpin extends into the short loop side of the barrel, creating a plug in the end of the barrel and forming a ~15 Å cavity extending between the inner helices of the barrel to the face of the barrel containing the b-sheet structures. Note that the protein molecule contains an N-linked oligosaccharide extending downward in the structure and that a large portion of the N-linked oligosaccharide is visible (and therefore relatively rigid) in the structure. This results from the fact that the oligosaccharide covalently attached to the protein at this site extends downward in the crystal lattice to the next symmetry-related protein molecule. The interaction between the oligosaccharide and the adjoining protein molecule is largely through hydrogen bonding interactions at the top of the cavity in what is presumed to be the aglycone substrate binding site. The figure below (AND THE CHIME VIEW HERE) shows two protein units in the crystal lattice with the N-linked oligosaccharide (in green) bridging between the two molecules.

Figure 1: Association of two monomers of the yeast ER mannosidase I in the crystal lattice.
(click the figure to enlarge)

Since the structure of the N-linked oligosaccharide at this site is identical to the product of the enzymatic reaction, it has been proposed that this structure, with the N-linked oligosaccharide bound to the face of the barrel, reflects an enzyme-product complex. Thus, many of the points of interaction between the enzyme and the substrate have been identified as a result of this structure and hypotheses have been developed to explain the substrate specificity of the enzyme. In a testing of these hypotheses, a single amino acid (Arg 273 in yeast a1,2-mannosidase) was found to interact with three mannose residues and one N-acetylglucosamine residue of the N-glycan (7). This interaction was hypothesized to be responsible, at least in part, for the specificity of the yeast ER mannosidase I. In mammalian Golgi a1,2-mannosidases that trim Man₉GlcNAc₂ to Man₅GlcNAc₂ this arginine residue is typically a leucine. Replacement of Arg 273 with leucine in yeast ER class 1 a1,2-mannosidase resulted in an enzyme that is capable of cleaving all four a1,2-linked mannose residues rather than just the single terminal residue of the middle arm of Man₉GlcNAc₂ (7).

Although the yeast enzyme structure provided many insights into the docking of the substrate in the active site, there are several aspects of the enzymatic mechanism which remained unclear from the structure. The yeast structure demonstrated the mode of interaction of the aglycone with the enzyme, but provided little detail into the mode of interaction of the glycone with the enzyme nor the residues that were involved in catalysis.
 
Part 2: Mechanisms of inverting glycosidases