In addition to small hydrophobic viral proteins, there are other virus products that promote membrane permeabilization. This occurs with a number of virus glycoproteins (GP) that are known to increase cell membrane permeability, such as the human immunodeficiency virus gp41 (Chernomordik et al., 1994; Arroyo et al, 1995), the Ebola virus GP (Yang et al., 2000), the cytomegalovirus US9 protein (Maidji et al., 1996), the Vaccinia virus A38L protein (Sanderson et al., 1996), rotavirus VP7 and NS4 proteins (Charpilienne et al., 1997; Newton et al., 1997), the hepatitis C virus E1 protein (Ciccaglione et al., 1998), and the alphavirus E1 protein (Nyfeler et al., 2001; Wengler et al., 2003).
The architecture of some viral glycoproteins is such that upon oligomerization, the transmembrane (TM) domains may form a physical pore. In principle, two different regions of a viral fusion glycoprotein could form pores. One such region contains the fusion peptide that would create a pore in the cell membranes upon insertion (Skehel and Wiley, 1998), while the TM domain would form a pore in the virion membrane (Wild et al., 1994). Moreover, sequences adjacent to the TM region could have motifs designed to destabilize membrane structure (Suarez et al., 2000). Entry of enveloped animal viruses leads to early membrane permeabilization, which is mediated by the formation of the two pores (fusion and TM) formed by viral fusion glycoproteins. This early permeabilization induced during the entry of virions requires conformational changes of the fusion glycoproteins. By contrast, after virus replication, newly synthesized glycoproteins may affect membrane permeability when they reach the plasma membrane (Figure 6.3). This modification is achieved only by the TM domain, while the fusion peptide does not participate in this late modification. In viruses that lack the typical viroporin, its function could be replaced by these pore-forming glycoproteins, while for other viruses viroporin activity may be redundant (Bour and Strebel, 1996). In the latter case, pore formation may be generated by viral glycoproteins and viroporins (Figure 6.3). We would like to propose the possibility that pore-forming glycoproteins play a key role mainly during virus entry and, in some cases, also during virus budding, while viroporins come into action when viruses need to exit the cell.
Early membrane permeabilization is always carried out by a virion component. In the case of enveloped viruses, this early event is executed by a structural glycoprotein, which is coupled to the fusion process. An understanding of fusion at the molecular level also requires an explanation of the phenomenon of early membrane permeabilization. We have advanced the idea that viral glycoproteins involved in membrane fusion participate in the dissipation of the chemiosmotic gradient, thus providing the energy to push the nucleocapsid and neighboring macromolecules to the cell interior (Carrasco, 1994; Irurzun et al., 1997). Fusion glycoproteins do not simply serve to bridge the cellular and the viral membrane, but instead are designed to open pores in both membranes. This pore-opening activity may be necessary to lower membrane potential and to dissipate ionic gradients. Several chapters of this book are devoted to the detailed description of the structure and function of these glycoproteins, so we will focus our attention on viral glycoproteins that permeabilize membranes when individually expressed in cells. These membrane active proteins may exhibit this activity later on in the virus life cycle.
Rotavirus infection provokes a number of alterations in cellular membranes during infection (del Castillo et al., 1991). Amongst these alterations, there is an increase in the
Figure 6.3. Participation of pore formation by viral glycoproteins and viroporins in membrane permeability. Early membrane permeabilization is coupled to the fusion activity of the corresponding viral glycoprotein. This fusion glycoprotein may create two pores. One is located at the target cell membrane and the other is formed by the TM domain. Late membrane permeabilization may be carried out by viroporins or by the TM domains of viral glycoproteins.
concentration of cytoplasmic calcium (Michelangeli et al., 1991). Several rotavirus proteins exhibit membrane-destabilizing activity. The enterotoxin NSP4 induces alterations in membrane permeability (Tian et al, 1994). The individual expression of the non-structural glycoprotein NSP4 has the ability to increase the concentration of cytoplasmic calcium. This increase may be mediated by activation of phospholipase C activity (Dong et al, 1997).
Rotavirus particles induce the co-entry of protein toxins into cells (Cuadras et al., 1997). At least two structural components possess the ability to permeabilize cells, including VP5 protein and VP7 glycoprotein (Charpilienne et al., 1997; Irurzun et al., 1997).
Infection of lymphocytic human cells by HIV-1 enhances membrane permeability to ions and several compounds (Voss et al., 1996; Gatti et al., 1998). There are at least three different HIV-encoded proteins responsible for these alterations: Vpu protein, the retroviral protease, and the fusion glycoprotein gp41. Apart from the fusion peptide, there are two regions of gp41 that exhibit membrane permeability; one is located at the carboxy terminus (Arroyo et al., 1995; Comardelle et al., 1997) and another corresponds to the membrane-spanning domain (Arroyo et al., 1995). The C-terminus of gp41 includes two 20-30 residues, which may form cationic amphipathic a-helices, designated as lentivirus lytic peptides 1 and 2 (LLP-1 and LLP-2). Synthetic LLP-1 peptide forms pores in planar phospholipid bilayers
(Chernomordik et al., 1994), permeabilizes HIV-1 virions to deoxyribonucleoside triphosphates (Zhang et al., 1996), and induces alterations in ion permeability of Xenopus oocytes (Comardelle et al, 1997).
Inducible expression of the hepatitis C virus E1 glycoprotein increases membrane permeability in bacterial cells. The ability of E1 to modify membrane permeability has been mapped to the carboxy terminus of the protein (Ciccaglione et al., 1998, 2001). Similar permeabilization was found with Escherichia coli cells that synthesize Semliki forest virus E1 glycoprotein after exposure to low pH (Nyfeler et al., 2001). Finally, overexpression of Vaccinia virus A38L glycoprotein produces changes in the morphology, permeability, and adhesion of mammalian cells. The potential capacity of A38L protein to form pores at the plasma membrane promotes the entry of calcium ions and PI and the release of lactic dehydrogenase into the culture medium (Sanderson et al., 1996).
Was this article helpful?