N-linked glycosylation in Archaea: a structural, functional, and genetic analysis

Abstract

N-glycosylation of proteins is one of the most prevalent posttranslational modifications in nature. Accordingly, a pathway with shared commonalities is found in all three domains of life. While excellent model systems have been developed for studying N-glycosylation in both Eukarya and Bacteria, an understanding of this process in Archaea was hampered until recently by a lack of effective molecular tools. However, within the last decade, impressive advances in the study of the archaeal version of this important pathway have been made for halophiles, methanogens, and thermoacidophiles, combining glycan structural information obtained by mass spectrometry with bioinformatic, genetic, biochemical, and enzymatic data. These studies reveal both features shared with the eukaryal and bacterial domains and novel archaeon-specific aspects. Unique features of N-glycosylation in Archaea include the presence of unusual dolichol lipid carriers, the use of a variety of linking sugars that connect the glycan to proteins, the presence of novel sugars as glycan constituents, the presence of two very different N-linked glycans attached to the same protein, and the ability to vary the N-glycan composition under different growth conditions. These advances are the focus of this review, with an emphasis on N-glycosylation pathways in Haloferax, Methanococcus, and Sulfolobus.

FIG 1

Schematic overview of the different types of protein glycosylation. Acceptor residues (S, serine; T, threonine; Y, tyrosine; HyP, hydroxyproline; HyL, hydroxylysine; N, asparagine; W, tryptophan) or conserved motifs (N-X-S/T and WXXW) within a newly synthesized peptide (blue band) are indicated in boldface type. The sugar donors, shown as either nucleotide-activated or lipid-bound intermediates, are shown next to the glycosylation site. The catalyzing enzyme or enzyme complex is indicated in boldface type. GPI, glycosylphosphatidylinositol.

FIG 2

The N-glycosylation pathways in Saccharomyces cerevisiae and Campylobacter jejuni.

FIG 3

Working model of the N-glycosylation pathway in Haloferax volcanii. Shown are the steps at which various known agl gene products act in the biosynthesis and assembly of the N-glycan and the steps for which Agl protein involvement remains to be identified. The actual arrangement, size, and shape of the various GTs and AglB and the mechanism by which AglB acts are not known.

FIG 4

Working model of the N-glycosylation pathway in Methanococcus maripaludis. Shown are the steps at which various known agl gene products act in the biosynthesis and assembly of the N-glycan and the steps for which Agl protein involvement remains to be identified. The actual arrangement, size, and shape of the various GTs and AglB and the mechanism by which AglB acts are not known.

FIG 5

Working model of the N-glycosylation pathway in Sulfolobus acidocaldarius. Shown are the steps at which various known agl gene products act in the biosynthesis and assembly of the N-glycan and the steps for which Agl protein involvement remains to be identified. The actual arrangement, size, and shape of the various GTs and AglB and the mechanism by which AglB acts are not known.

FIG 6

Schematic depiction of known agl genes in model Archaea. Genes are colored to reflect demonstrated functions, as shown in the inset.

FIG 7

Structure of the repeating-unit N-linked glycan of Halobacterium salinarum (171, 172).

FIG 8

Structure of the N-linked sulfated oligosaccharide glycan of Halobacterium salinarum (84).

FIG 9

Structure of the N-linked glycan of Methanococcus voltae (111).

FIG 10

Structure of the archaellin N-linked glycan of Methanococcus maripaludis (120).

FIG 11

Structure of the N-linked glycan of Sulfolobus acidocaldarius (137, 138).

FIG 12

Structure of the large high-mannose-type N-linked glycan of Thermoplasma acidophilum (184).

FIG 13

Structure of the N-linked glycan of haloarchaeal pleomorphic virus 1 (HRPV-1) VP4 (187).

FIG 14

Diversity of N-linked glycans in the domain Archaea. The phylogenetic tree is based on the alignment of the full-length 16S rRNA sequence. Cell wall structures are indicated. The number of copies of the aglB gene is shown. n.d., not determined. (Adapted from reference .)

FIG 15

Schematic model of the catalytic site of Pyrococcus furiosus AglB and the presumed functions of the conserved motifs. Four interactions are assumed to occur, at least transiently, during the oligosaccharyltransferase reaction. The Ser/Thr residue in the +2 position of the acceptor sequon interacts with the Ser/Thr-binding pocket in the C-terminal globular domain (green). The pocket consists of the WWD part of the highly conserved WWDYG motif and the lysine residue of the DK or DKi (DK with insertion) motif (the isoleucine residue of the MI motif in other AglBs). A divalent metal ion (M²⁺[Mg²⁺ or Mn²⁺]) is coordinated by three acidic residues: D55 contained in the DXD motif located in EL1, D167 located in EL2, and the glutamic acid (E348) of the TIXE motif located in EL5 (white). The asparagine residue in the sequon is recognized by the two acidic residues in the DXD and TIXE motifs (red). The pyrophosphate or monophosphate group of the lipid-linked oligosaccharide interacts with the divalent metal ion via electrostatic interactions (blue). (Courtesy of D. Kohda; reprinted with permission.)