Hepatitis C virus (HCV) belongs to the Hepacivirus genus in the Flaviviridae family. Among the least known viral proteins in this family is the nonstructural protein NS4B, which has been suggested to be a part of the replication complex. Hydrophobicity plots indicate a common profile among the NS4B proteins from different members of the Flaviviridae family, suggesting a common function. In order to gain a deeper understanding of the nature of HCV NS4B, we have determined localization and topology of this protein by using recombinant HCV NS4B constructs. The protein localized to the endoplasmic reticulum (ER), but also induced a pattern of cytoplasmic foci positive for markers of the ER. Computer predictions of the membrane topology of NS4B suggested that it has four transmembrane segments. The N and C termini were anticipated to be localized in the cytoplasm, because they are processed by the cytoplasmic NS3 protein. By introducing glycosylation sites at various positions in HCV NS4B, we show that the C terminus is cytoplasmic and the loop around residue 161 is lumenal as predicted. Surprisingly, the N-terminal tail was translocated into the lumen in a considerable fraction of the NS4B molecules, most likely by a posttranslational process. Interestingly, NS4B proteins of the yellow fever and dengue viruses also have their N termini located in the ER lumen due to an N-terminal signal peptide not found in NS4B of HCV. A shared topology achieved in two different ways supports the notion of a common function for NS4B in Flaviviridae.

Among the least-known hepatitis C virus proteins is the non-structural protein 4B (NS4B). It localizes to the endoplasmic reticulum (ER) membrane and induces membrane changes, resulting in a membranous web that is reported to be the locale for virus replication. A model was presented previously for the topology of recombinant HCV NS4B of the 1a genotype based on in vitro data. In this model, the N-terminal tail of a considerable fraction of the NS4B molecules was translocated into the ER lumen via a post-translational process, giving the protein a dual transmembrane topology. It is now reported that translocation of the N terminus also occurs for processed NS4B expressed in cells in the context of the polyprotein. In the presence of NS5A, however, a lower degree of translocation was observed, which may indicate that NS5A influences the topology of NS4B. In vitro expression studies of NS4B from all major genotypes demonstrated that translocation of the N terminus to the ER lumen is conserved across genotypes. This clearly suggests an important function for this feature. Furthermore, when disrupting a previously reported amphipathic helix (AH) in the N terminus of NS4B, translocation was inhibited. As a disrupted AH also abolished the ability of NS4B to rearrange membranes, these data indicate for the first time an association between translocation of the N terminus and membrane rearrangement. Finally, the present experiments also confirm the predicted location of the first luminal loop to be around aa 112. INTRODUCTIONHepatitis C virus (HCV) is a positive-stranded RNA virus that is classified in the genus Hepacivirus, which, together with the genera Flavivirus and Pestivirus, makes up the family Flaviviridae. The genome is a positive-strand RNA that encodes a polyprotein of about 3000 aa, which is processed proteolytically by cellular and viral proteases. The structural proteins (core, E1 and E2) are located in the Nterminal part of the polyprotein and the non-structural proteins (NS2, NS3, NS4A, NS4B, NS5A and NS5B) in the latter part (Major et al., 2001). Between the two regions there is also a small protein, p7, but the question about to which region it belongs to is still disputed.Little is known about the details of HCV's replication process, but it is generally believed that most of the NS proteins are involved. This assumption is based on comparisons with other members of the family Flaviviridae and the fact that all NS proteins are localized together with newly synthesized viral RNA on the endoplasmic reticulum (ER) or membranes originating from it (El-Hage & Luo, 2003;Gosert et al., 2003;Mottola et al., 2002). Most NS proteins also interact with several other NS proteins, and some of them interact with all (Dimitrova et al., 2003).The NS2 protein is an autoprotease that, together with NS3, digests the junction between them (Hijikata et al., 1993). NS2 has been shown to interact with all other NS proteins, but is not required for replication in a subgenomic replicon (Blight et al., 2000;Dimitrova et al., ...

No abstract

4782 Ramos cells were labeled with Type I (Rituximab and Ocrelizumab) and Type II (B1/H299 and GA101) anti-CD20 antibodies and then fixed and stained with marker gold labeled 2ary antibodies. Protein complexes connected to marker gold were analyzed by Protein TomographyTM at SIDEC and the corresponding 3D-structures visualized (∼100 structures per antibody). Less than 5 % of the refined structures were dimeric. While most (60 – 90%) CD20 molecules were present in tetramers or even higher order defined complexes, a sizeable proportion was also engaged in large protein networks (11 -38%). For the multimeric complexes, we could clearly distinguish between extended (or open) and ring-like (or closed) conformations. Although cells had been labeled with an excess of antibody, as suggested by FACS binding curves, the antibodies bound CD20 monovalently in most refined structures. Bivalent binding was overall more prevalent with Rituximab and Ocrelizumab than with B1 and GA101 (1/3 vs 1/6 structures). The proportion of CD20 molecules present in ringlike complexes was higher for the Type II than for the Type I antibodies. In co-localization experiments, Rituximab-CD20 and GA101-CD20 complexes were also found to only partially colocalize, while a large fraction of the two antibodies were found in separate cell surface compartments. These findings suggest that the different antibodies favor different CD20 conformations that seem to be associated with different protein complexes and might form the basis for initiation of different downstream signaling processes. Disclosures: Niederfellner: Roche: Employment. Braennstroem:SIDEC: Employment. Lindstrom:SIDEC: Employment. Jansson:SIDEC: Employment. Lundin:SIDEC: Employment. Schaefer:Roche: Employment. Mundigl:Roche: Employment. Umana:Glycart: Employment, Equity Ownership, Patents & Royalties. Klein:Roche: Employment, Equity Ownership, Patents & Royalties.