Bacterial hydrolysis of host glycoproteins - powerful protein modification and efficient nutrient acquisition

Abstract

Glycoproteins are ubiquitous in nature and fundamental to most biological processes, including the human immune system. The glycoprotein carbohydrate moieties, or glycans, are very diverse in their structure and composition, and have major effects on the chemical, physical and biological properties of these glycoproteins. The hydrolysis of glycoprotein glycans by bacterial glycosidases can have dramatic effects on glycoprotein function and, thereby, be beneficial for the bacteria in different ways. This review gives an introduction to the expanding field of extracellular glycosidases from bacterial pathogens with activity on host glycoproteins, describes some known and proposed consequences for the host and the bacteria and discusses some evolutionary and regulatory aspects of bacterial glycosidases.

Fig. 1

Schematic representation of glycan hydrolysis of N-linked and core-1 O-linked glycans. Arrows indicate cleavage sites of glycosidases (discussed in this review): with a solid line = endoglycosidases and with a dotted line = exoglycosidases, with activity only on terminal carbohydrates. Enzyme names in bold indicate cleavage sites (shown or proposed in the references). Other indicated cleavage sites are suggested cleavage sites due to similarities with other enzymes. Please note that EngPA and EngEF also have activity on core-3 O-linked glycans and that EngSP activity probably relies on the cleavage of the sialic acid by NanA. Gal = Galactose; GalNAc = N-acetylgalactosamine; GlcNAc = N-acetylglucosamine; Man = mannose; Sia = sialic acid. Symbols and coloring adhere to guidelines devised by the Consortium for Functional Glycomics (www.functionalglycomics.org/static/consortium/Nomenclature.shtml).

Fig. 2

Phylogenetic analysis of bacterial glycosidases. They were first aligned using the ClustalW algorithm of the MacVector^TM 11.0.2 sequence analysis software suite. Phylogenetic trees were generated using the neighborjoining method, gaps were distributed proportionally with uncorrected p values and were validated by bootstrap analysis (32,000 repetitions). The GenBank accession numbers of the enzymes are indicated in parentheses together with the organism name.