Folding and retention of rotavirus glycoproteins in the endoplasmic reticulum

Abstract

The endoplasmic reticulum (ER) is a membrane bound organelle composed of a network of cisternae stretching throughout the cytoplasm. Functionally, the ER performs many essential processes, such as glycosylation, and biosynthesis of lipids; it provides a redox potential condition for forming S-S linkages, and a quality control system, and functions as an intracellular calcium store. Rotavirus, a segmented dsRNA virus, is one of very few viruses that undergoes a unique maturation process in the ER and has therefore been an attractive model for the study of ER-translocation, retention and protein folding. In this thesis, I have studied how rotavirus and rotavirus proteins assemble in the eucaryotic cells. The rotavirus genome encodes 12 structural and nonstructural proteins; two of these proteins, the outer capsid VP7 protein and the nonstructural enterotoxin NSP4 protein, are associated with the ER and have been of particular interest in this thesis.

We found that brefeldin A, a compound that causes relocation of resident Golgi proteins to the ER, had a pronounced effect on rotavirus assembly and oligosaccharide processing (paper I). The results show that VP7, but not NSP4, was trimmed by [alpha]-mannosidase II, a medial Golgi enzyme, during brefeld in a treatment. Electron microscopy analysis revealed that brefeldin A interfered in the transition from the intermediate enveloped particles to the mature triple shelled particles. We also found that glycosylation inhibitors caused misfolding of VP7, which lead to stable interdisulfide aggregates (paper II). These observations suggest that the major function of the carbohydrates on VP7 is to facilitate correct disulfide bond formation and to stabilize the conformation. In paper III, we found that NSP4 associated with calnexin, a lectin-like chaperone, during oligosaccharide processing. This interaction was found to be glycan- and time-dependent. Novel experiments with castanospermine, a glucosidase 1 and II inhibitor, revealed that glucose trimming and calnexin NSP4-interaction were not essential for assembly of infectious virus. The role of ATP for activity of certain molecular chaperones has previously been discussed. In paper IV, we show that VP7 rapidly misfolds in an energy-depleted milieu. We also found that VP7 attained a stable minimum-energy-state immediately after translation in the ER. Most surprisingly, energy-misfolded VP7 could be recovered and obtain correct disulfide bonds and antigenicity following a shift to an ATP-rich milieu.

Our results show, for the first time, that rotavirus VP7 disulfide bond formation is an ATP-dependent process. The most reasonable explanation for the energy requirement of VP7 during folding is thus a close interaction with an ATP-dependent chaperone, such as Bip (grp 78), and possibly protein disulfide isomerase. VP7 and N5P4 mature and accumulate in the ER. While three amino acids in the N-terminus have been proposed to function as a retention signal of VP7, no information is yet available on how NSP4 remains associated with the ER. The last paper describes a region between amino acids 85-123 in the cytoplasmic tail of NSP4 that is involved in ER retention. This region covers also a part of a domain involved in membrane destabilization and oligomerization. Taken together our observations provide new information about protein folding and retention of ER-associated proteins in general, and for rotavirus VP7 and NSP4 in particular.