Structural insights into glycoprotein transport and viral escape

Abstract

The correct trafficking of proteins through the secretory pathway is essential for the maintenance of cellular functions. Increasing evidence suggests that at least for some glycoproteins, export from the endoplasmic reticulum (ER) is mediated by interaction with cargo receptors. The mannose selective lectin p58/ERGIC-53 is the first identified cargo receptor for glycoproteins. The structure determination of the carbohydrate recognition domain (CRD) of this protein revealed a striking structural similarity to the CRD of plant Ltype lectins, suggesting that this scaffold was present in a common lectin ancestral of plants and animals. However, interesting structural differences are found between p58/ERGIC-53 and plant L-type lectins.

An unpredicted calcium binding site is present in the CRD of this protein, which is 6 A away from the metal binding site conserved between these two lectin families. This site might be part of an extended saccharide binding site that has evolved in p58/ERGIC-53 to bind high mannose type glycans. This is consistent with the absence of a loop in p58/ERGIC-53 which interacts with sugars and restricts the size of the pocket where monosaccharides bind in the plant L-type lectins. The transition from the calcium-free to the calcium bound form of the CRD of p58/ERGIC-53 is associated with localized but large conformational changes.

Viral escape from cytotoxic T-lymphocytes (CTL) recognition is thought to play an important role in the establishment of persistent viral infections. Lymphocytic choriomeningitis virus (LCMV) is a small RNA virus that infects mice and is capable of establishing persistent infection in its host. Most of the host CTL response against the virus is directed against one immunodominant epitope (GP33) derived from the viral glycoprotein and restricted by both H-2Kb and H-2Db class I major histocompatibility complex (MHC) in H-2b mice. Structural analyses of the GP33 index peptide and three of its point mutants, derived from LCMV escape variants, in complex with both H-2Kb and H2Db revealed that very subtle structural changes can have dramatic effects in CTL recognition.

The viral escape mutations affect Tcell receptor (TCR) contacting residues from the MHC molecule, as well as potential TCR contacts of the peptide. Each mutation affects the two peptide/MHC complexes differently, highlighting how under CTL pressure the virus can develop efficient variants to escape immune recognition in the context of two different MHC class I.