PLoS ONE 2014 9(1): e78494. doi:10.1371/journal.pone.0078494

Two Different Conformations in hepatitis C Virus p7 Protein Account for Proton Transport and Dye Release.

Gan SW, Surya W, Vararattanavech A, Torres J

School of Biological Sciences. Nanyang Technological University. 60, Nanyang Drive. Singapore 637551.

 

Abstract

The p7 protein from the hepatitis C virus (HCV) is a 63 amino acid long polypeptide that is essential for viral replication, and is involved in protein trafficking and proton transport. This small protein has been the subject of intensive research over the last 10-15 years, but it has been refractory to structural and functional analysis because of its hydrophobicity and tendency to aggregate. There is a consensus for its structure: an α-helical hairpin formed by two transmembrane domains. Six or seven monomers would form an ion or proton channel. Intense research has been directed towards the discovery of molecules that can inhibit this channel activity, as these inhibitors also inhibit viral replication in vitro. Therefore, p7 constitutes a possible target for antivirals. In the search for p7 channel inhibitors, a high-throughput functional assay was proposed based on carboxyfluorescein (CF) release from liposomes after p7 addition. However, we were puzzled by the dual ability of p7 to serve both as a proton channel in infected cells and to permeabilize membranes to large molecules like CF. In this paper, we have recreated both activities in vitro, and we have examined the conformation of p7 in these two assays using infrared spectroscopy. We observed that addition of p7 to preformed liposomes results in about 50% of the sample adopting β-structure. However, after repeatedly extruding the sample, a fully helical form could be obtained, which was able to permeabilize membranes to protons. A first indication that the two assays involve different conformations of p7 is that full length p7 protein was not required to release CF from liposomes, and that fragment p7(27-63) alone could do it. In contrast, when p7 was present as a α-helical form, CF was not released from liposomes. Successive tests indicated that the release of CF was not due to a minor fraction of α-helical form of p7, but to an incorrectly folded form, where only the C-terminal part of p7 is inserted in lipid bilayers and the N-terminal half adopts β-structure. The observed inhibitory effect of some small compounds on CF release and proton transport can be explained by binding of these compounds to the same C–terminal inserted fragment. This mechanism may be similar to the one proposed for the influenza A M2-rimantadine interaction. A proton transport assay should be used instead in high throughput anti-HCV p7 drug discovery efforts.

PMID: 24409277

 

Supplement:

Hepatitis C virus (HCV) has chronically infected about 170 million people worldwide and no prophylactic or therapeutic vaccine is available. The p7 protein, encoded by HCV, is a small transmembrane (TM) protein that is 63 residues long and is found in infected cells mainly at internal membranes. Although p7 is not necessary for RNA replication, it is essential for infectivity, assembly, and release of infectious virions. Despite its importance, little is know about this protein because of its hydrophobicity and tendency to aggregate. This has hampered structural studies for a long time, but an accepted model is that it has two α-helical TM domains forming a sort of hairpin. In this model, the N-terminal TM helix would face the lumen of a channel formed by six or seven monomers. These oligomers show ion channel activity and, recently, membrane permeabilization to protons was found to be crucial for the production of infectious viruses. In addition to this proton transport, p7 is involved in the intracellular assembly and localization of several viral proteins, possibly via channel-independent mechanisms.

It has been shown that addition of p7 to liposomes that encapsulate carboxyfluorescein (CF) leads to dye release. Dye release was found to be blocked by the same molecules that inhibit channel activity, and therefore this method has been used in the literature as a functional assay to discover new p7 channel inhibitors. However, we found strange that an oligomer that transports ions or protons is also a facilitator for large molecules like CF. Therefore, our hypothesis was that the conformation of p7 when performing these two very different functions is not the same, and we set to test this using infrared spectroscopy of p7 in membranes. Yet, the fact that both activities seem to be inhibited, although weakly, by similar inhibitors, suggests that the binding mode of these inhibitors to p7 must be the same.

We established first that adding p7 to preformed liposomes always results in an amide I band consistent with ~50% of β-structure, when the expected p7 conformation should be α-helical. However, when the sample was subsequently vortexed, freeze-thawed, sonicated and repeatedly extruded, its β-structure content was reduced until the sample became mostly α-helical. This helical form could transport protons in an in vitro assay, consistent with its role in infected mammalian cells, acting to prevent acidification of intracellular vesicles, but it was not able to release CF from CF-loaded liposomes. This was striking because now in theory all p7 should be correctly incorporated and therefore one should observe an even more efficient CF release. We concluded that the helical form of p7 is not able to elicit CF release. The alternative then was that CF release is caused by an incorrectly folded p7. We also observed that even addition of the more hydrophobic fragment p7(27-63) was as efficient as full length p7 in releasing CF. These results again indicate that correct incorporation of p7 as an α-helical hairpin is not required to elicit CF release. The conclusion of these experiments are summarized in Fig. 1, where added p7 to liposomes has only one TM domain inserted in the lipid bilayer, whereas loop and TM1 would form predominantly β-structure. We propose that this form is responsible for CF release. The α-helical form, in contrast, is not able to allow CF release, although it is able to transport protons.

fig1Figure 1. (A) Fully α-helical structure of p7 protein in membranes after correct incorporation; (B) this form transports protons, like influenza A M2, but it did not elicit CF release; (C) the model for this p7 proton channel; (D) abundant β-structure in p7 protein after addition to pre-formed liposomes; (E) release of CF from liposmes; (F) incorrect incorporation of the N-terminal half is responsible for membrane disruption.

Acknowledgements:  This paper was funded by National Research Foundation grant NRF-CRP4-2008-02 and AcRF Tier 1 grant RG 38/08.

 

Contact:

Jaume Torres, Ph.D.
Associate Professor
School of Biological Sciences, College of Science
60, Nanyang Drive
Nanyang Technological University
Singapore 737551
jtorres@ntu.edu.sg

 

Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann