Identification of AglE, a second glycosyltransferase involved in N glycosylation of the Haloferax volcanii S-layer glycoprotein

Abstract

Archaea, like Eukarya and Bacteria, are able to N glycosylate select protein targets. However, in contrast to relatively advanced understanding of the eukaryal N glycosylation process and the information being amassed on the bacterial process, little is known of this posttranslational modification in Archaea. Toward remedying this situation, the present report continues ongoing efforts to identify components involved in the N glycosylation of the Haloferax volcanii S-layer glycoprotein. By combining gene deletion together with mass spectrometry, AglE, originally identified as a homologue of murine Dpm1, was shown to play a role in the addition of the 190-Da sugar subunit of the novel pentasaccharide decorating the S-layer glycoprotein. Topological analysis of an AglE-based chimeric reporter assigns AglE as an integral membrane protein, with its N terminus and putative active site facing the cytoplasm. These finding, therefore, contribute to the developing picture of the N glycosylation pathway in Archaea.

FIG. 1.

H. volcanii Dpm1-B (AglE) is a homologue of eukaryal Dpm1. The deduced amino acid sequence of H. volcanii Dpm1-B (AglE) was aligned with the sequence of the Mus musculus Dpm1 protein (NP_034202.1). Identical amino acids are in black boxes, while similar amino acids are shaded gray.

FIG. 2.

Deletion of the dpm1-B (aglE) gene does not affect cell viability. (A) Left panel: PCR amplification was performed by using a forward primer directed at the 5′ dpm1-B flanking region and a reverse primer directed at a sequence within the dpm1-B coding region (lanes 1a, 2a, and 3a) or a sequence within the trpA sequence (lanes 1b, 2b, and 3b), together with genomic DNA from cells of the WR536 background strain (bkgnd, lanes 1a and 1b), from plasmid-incorporating cells (pop-in, lanes 2a and 2b) or from cells that had replaced the dpm1-B gene (pop-out, lanes 3a and 3b), as a template. For the left panel, PCR amplification was performed with primers directed against the dpm1-B coding region together with genomic DNA from cells of the WR536 background strain (bkgnd) or the dpm1-B-deleted strain (deletion). (B) RT-PCR was performed with primers directed at dpm1-B (upper panel) or sec11b (lower panel) and genomic DNA from WR536 background cells or cDNA or RNA from cells with dpm1-B deleted as a template.

FIG. 3.

MALDI-TOF analysis of an Asn-13-containing, H. volcanii S-layer glycoprotein-derived glycopeptide. The MALDI-TOF spectra of the Asn-13-containing tryptic peptides derived from the S-layer glycoprotein of the WR536 background (upper panel) and dpm1-B (aglE)-deleted strains (lower panel) are shown. The components of the glycopeptide-associated sugar residues are shown in the inset box, while the glycan moieties decorating the peptide peaks are marked on the MALDI-TOF spectra, accordingly.

FIG. 4.

Expression and topology of the CBD-AglE construct. (A) The protein contents of cells of the WR536 background strain or the same cells expressing CBD-AglE were precipitated, separated by SDS-PAGE, and immunoblotted with antibodies directed against the cellulose-binding domain. The positions of molecular mass markers are shown on the right. (B) H. volcanii cells expressing CBD-AglE were broken by sonication, membrane and supernatant fractions were obtained by ultracentrifugation, and the different fractions were immunoblotted with antibodies to the S-layer glycoprotein (SLG), SRP54 or CBD. Membrane fractions treated with 6 M urea and ultracentrifuged through a 40% sucrose cushion are shown in the right lanes of each panel. The same treatments were performed with H. volcanii cells expressing the soluble CBD moiety alone (bottom panel). (C) H. volcanii cells (1 ml) expressing CBD-AglE were subjected to proteolysis with 1 mg of proteinase K/ml at 55°C. Aliquots (100 μl) were removed every 30 min, TCA (15%) precipitated, separated by SDS-12% PAGE, and either Coomassie-stained (S-layer glycoprotein, upper panel) or immunoblotted with antibodies to SRP54 (middle panel) or CBD (lower panel) at the concentrations given in Materials and Methods. Antibody binding was detected by enhanced chemiluminescence.