Structure and mechanism of a bacterial β-glucosaminidase having O-GlcNAcase activity

Abstract

O-GlcNAc is an abundant post-translational modification of serine and threonine residues of nucleocytoplasmic proteins. This modification, found only within higher eukaryotes, is a dynamic modification that is often reciprocal to phosphorylation. In a manner analogous to phosphatases, a glycoside hydrolase termed O-GlcNAcase cleaves O-GlcNAc from modified proteins. Enzymes with high sequence similarity to human O-GlcNAcase are also found in human pathogens and symbionts. We report the three-dimensional structure of O-GlcNAcase from the human gut symbiont Bacteroides thetaiotaomicron both in its native form and in complex with a mimic of the reaction intermediate. Mutagenesis and kinetics studies show that the bacterial enzyme, very similarly to its human counterpart, operates via an unusual 'substrate-assisted' catalytic mechanism, which will inform the rational design of enzyme inhibitors.

Insight into a strategy for attenuating AmpC‐mediated β‐lactam resistance: Structural basis for selective inhibition of the glycoside hydrolase NagZ

“…Rearrangement of this loop also moves the general acid/base residue, Asp243, a remarkable 7.5 Å away from its normal position, making it unable to interact with the inhibitor. In the Xray structure of BtGH84 bound to PUGNAc, Asp243 sits in the same position as the unliganded enzyme, 22,31 and forms a hydrogen-bond to the carbamate nitrogen of the N-phenylcarbamate moiety of PUGNAc. Although this interaction is lost in the N-butyryl-PUGNAc complex, it is interesting to note that the N-phenylcarbamate moiety of N-butyryl-PUGNAc is flipped 180 so that its carbonyl oxygen can instead accept a hydrogen bond from the imidazole of His443.…”

“…Although GH3 enzymes, including NagZ, are functionally related to the GH20 human b-hexosaminidase isoenzymes and GH84 O-GlcNAcase, recent kinetic [17][18][19] and structural studies 16,[20][21][22] have revealed that GH3 enzymes use a catalytic mechanism that differs from that used by GH20 and GH84 enzymes. This distinction should therefore offer a tractable route to generating selective inhibitors of NagZ, and clear comparisons of these enzymes at a structural level would greatly accelerate these efforts.…”

“…Structural studies of N-acetyl-b-glucosaminidases from families GH3, 16 GH20, 20,21,24,25 and GH84 22,26 have revealed significant differences in the active site structures of enzymes carrying out substrate assisted catalysis versus those proceeding via a glycosyl-enzyme intermediate. For GH3 enzymes, it has been found that the 2-acetamido group of the substrate is not essential for cleavage of the glycosidic bond, consistent with these enzymes using an enzymic nucleophile.…”

“…19,27 The X-ray structure of NagZ from Vibrio cholerae (VcNagZ) in complex with its GlcNAc product (PDB entry: 1Y65) corroborates this finding, clearly showing that the 2-acetamido group of bound GlcNAc is not positioned to participate directly in CAO bond cleavage. Crystallographic studies of GH20 human HexA 25 and HexB, 24 and GH84 bacterial homologues of human O-GlcNAcase from Bacteroides thetaiotaomicron (BtGH84) 22 and Clostridium perfringens 26 have revealed that these families of enzymes bind the 2-acetamido group in a strikingly different manner. To position the carbonyl oxygen of the 2-acetamido group as a nucleophile and favor formation of the bicyclic oxazoline intermediate, these enzymes rotate the amide moiety into a deep hydrophobic pocket that tightly envelops this substituent, positioning it within striking distance of the anomeric center.…”

“…deeper in O-GlcNAcase than it is in the GH20 b-hexosaminidases. 22 Thus, not only do large structural differences exist between the NagZ active site and the active sites of GH20 b-hexosaminidases and GH84 OGlcNAcase, there also exist less prevalent yet distinct differences in the active site structures of the latter two enzyme families.…”