SGM Journals
Research Article

During entry of alphaviruses, the E1 glycoprotein molecules probably form two separate populations that generate either a fusion pore or ion-permeable pores

Wengler, Gerd, Koschinski, Andreas, Wengler, Gisela, Repp, Holger

,†, Gisela Wengler1 and Holger Repp2

1 Institut für Virologie der Veterinärmedizin, Justus-Liebig-Universität, D-35392 Giessen, Germany
2 Rudolf-Buchheim-Institut für Pharmakologie, Justus-Liebig-Universität, D-35392 Giessen, Germany

Correspondence
Gerd Wengler
gerd.wengler{at}gmx.de

Journal of General Virology 2004; 85(6):1695 · https://doi.org/10.1099/vir.0.79845-0

View at publisher PubMed

Abstract

Studies using the alphavirus Semliki Forest virus have indicated that the viral E1 fusion protein forms two types of pore: fusion pores and ion-permeable pores. The formation of ion-permeable pores has not been generally accepted, partly because it was not evident how the protein might form these different pores. Here it is proposed that the choice of the target membrane determines whether a fusion pore or ion-permeable pores are formed. The fusion protein is activated in the endosome and for steric reasons only a fraction of the activated molecules can interact with the endosomal membrane. This target membrane reaction forms the fusion pore. It is proposed that the rest of the activated molecules interact with the membrane in which the protein is anchored and that this self-membrane reaction leads to formation of ion-permeable pores, which can be detected in the target membrane after fusion of the viral membrane into the target membrane.
This study was supported by grant RE 1046/1-2 to H. R. and by grant WE 518/3-3 to G. W. from the Deutsche Forschungsgemeinschaft.

Footnotes

,†, Andreas Koschinski2 †These authors made equal contributions to this work.



Entry of enveloped viruses involves fusion of the viral membrane into a cellular target membrane. A fusion pore is generated during this process by a viral fusion protein and the internal components of the virus gain access to the cell through this pore (reviewed by Flint et al., 2000; Young, 2001). In the case of alphaviruses, the E1 protein represents the viral fusion protein (reviewed by Marsh & Helenius, 1989; Garoff et al., 1994; Kielian, 1995). Virus-induced changes in the ion permeability of membranes of infected cells have been found early and late in infection in many systems (Carrasco, 1981; reviewed by Carrasco, 1995). The alphavirus E1 protein has been implicated in the formation of ion-permeable pores in two situations: (i) E1 protein synthesized in an infected cell is present in the plasma membrane and the membrane is exposed to low pH (Kempf et al., 1987; Lanzrein et al., 1993; Käsermann & Kempf, 1996; Dick et al., 1996; Nyfeler et al., 2001); (ii) E1 protein present on the virus surface is transferred into the target membrane during virus entry at low pH (Wengler et al., 2003; Koschinski et al., 2003). The ability of the E1 protein to form ion-permeable pores has not been widely accepted, in part possibly because it was not easily seen how a single protein might form pores of such different properties. In this manuscript we propose a simple model to show how the E1 protein might be able to form both types of pore.

During infection via the endosome, a low pH-induced reorganization of the alphavirus surface exposes the fusion peptide of the E1 protein. The ensuing interaction of the fusion peptide with the target membrane leads to the formation of a fusion pore by the E1 protein (reviewed by Garoff et al., 1994; Kielian, 1995). Evidently, because of steric reasons only a small fraction of the fusion protein molecules present on the surface of an individual virus particle can participate in this reaction. The generation of ion-permeable pores could be explained if it is assumed that the fusion proteins that have not reacted with the target membrane fold back and react with the viral membrane in which they are anchored and thereby generate ion-permeable pores. These two reactions might be called a target membrane reaction and a self-membrane reaction, respectively. We therefore propose that during virus entry the E1 protein molecules, present on the surface of a single virus particle, form two separate populations of molecules, which generate either a fusion pore by a target membrane reaction or ion-permeable pores by a self-membrane reaction. A schematic presentation of this concept is shown in Fig. 1. Together these processes allow the delivery of the viral core into the cytoplasm and a flow of ions through the target membrane. A possible role for this ion flow in the disassembly of alphavirus cores has been described (Wengler & Wengler, 2002).



(99K):

 		Fig. 1. Schematic representation of the formation of the fusion pore and ion-permeable pores by the fusion protein during entry of alphaviruses. (1) A virus particle containing a core, a viral membrane and activated fusion protein is present in the endosome. The fusion protein molecules, which are shown as arrows, are anchored in the viral membrane. The exposed active fusion peptide is indicated by the arrowhead. (2) Fusion protein molecules located near the endosomal membrane interact with the endosomal target membrane. (3) Complete fusion of the viral and the endosomal membrane has occurred. The six unaltered arrows represent those fusion protein molecules that have not reacted with the target membrane. They perform the self-membrane reaction which leads to the structure shown in step 4. (4) The fusion pore has enlarged, the core has entered the cytoplasm and the self-membrane reaction has occurred as indicated by the six arrows that fold back on themselves. The formation of ion-permeable pores by this self-membrane reaction is indicated by large arrows. (5) The viral membrane has fused completely into the endosomal membrane and the ion-permeable pores are closed. The figure is schematic and solely shown to introduce the concept that the fusion proteins are segregated into two functionally different groups that perform a target membrane reaction, which generates a fusion pore, or a self-membrane reaction, which generates ion-permeable pores, and that these processes together generate ion-permeable pores in the target membrane after fusion. For instance, the self-membrane reaction that is indicated in step 4 might have occurred already at one of the earlier steps shown in this figure.

The low pH-induced self-membrane reaction apparently also occurs for the E1 protein molecules that accumulate in the plasma membrane during virus multiplication (see Nyfeler et al., 2001, and references therein). The experiment presented in Fig. 2 analyses this situation. During replication of alphaviruses, a first phase in which 42S genome RNA, which is translated into non-structural proteins, is synthesized is followed by a second phase in which a 26S subgenomic RNA is made. The E1 fusion protein is translated from the 26S RNA, incorporated into the plasma membrane and the assembly of virus starts at this time (reviewed by Strauss & Strauss, 1994; Schlesinger & Schlesinger, 2001). The appearance of the E1 protein in the plasma membrane at the beginning of the second phase should generate cells exhibiting ion-permeable pores at low pH. These pores will allow the flow of Ca2+ ions into the cytoplasm, which can be used to detect the pores (Koschinski et al., 2003, and data not shown) by ratiometric fluorescence measurements (Silver, 1998) and Ca2+ imaging as described previously (Koschinski et al., 2003). An experimental analysis of this system, using Semliki Forest (SF) virus-infected HEK-293 cells is shown in Fig. 2. Cells were loaded with Fura-2 for 20 min at 37 °C. The virus growth curve (Fig. 2A) shows that assembly of progeny virus and the beginning of the second phase occurred at 2·5 h post-infection (p.i.). It could be seen that the cells did not form ion-permeable pores at low pH at 2 h p.i. and were almost quantitatively converted into cells that generated such pores at the beginning of the second phase between 2 h 15 min p.i. and 3 h p.i. (Fig. 2B, B', C, C', D and D'). Similar data are observed in very sparse cell cultures indicating that cellcell interactions do not play a role in the formation of ion-permeable pores (data not shown, but see Fig. 3). The experiments reported in Fig. 2 were performed in Petri dishes containing two separated closely adjacent cell layers immersed in a single layer of growth medium. To generate these cultures, two holes of 8 mm diameter were punched into the bottom of a 35 mm diameter Petri dish, which was sealed on the outside using a glass coverslip. One layer was infected and the adjacent layer served as a mock-infected control. The data presented in Fig. 2(E) and (E') showed that in the mock-infected cells no ion-permeable pores were generated. Virus particles released into the growth medium, which was the same for both layers, therefore were not responsible for the formation of the ion-permeable pores identified in this experiment. The data reported in Fig. 2 support the work of Kempf et al. (1987). Obviously, the low pH-induced formation of pores in the membrane of infected cells analysed in this experiment represents an artificial situation. The corresponding physiological situation probably is the low pH-induced self-membrane reaction of E1 protein molecules that do not interact with the target membrane during virus entry, as shown in Fig. 1. The data presented in Fig. 2(C) and (C') indicated that individual infected cells could be identified in the presence of a large number of uninfected cells. This fact allowed us rapidly to determine the number of infectious virus particles in a preparation of alphaviruses in a focus assay (data not shown).



(76K):

 		Fig. 2. Appearance of cells forming ion-permeable pores at low pH during viral growth. HEK-293 cell monolayers were infected with SF virus at an m.o.i. of 10. The virus growth curve was determined by plaque assay on BHK cells. For the analyses of the intracellular Ca2+ concentration, Petri dishes containing two adjacent monolayers immersed in a single layer of growth medium were infected or mock-infected. At the appropriate time, both layers were incubated in extracellular solution (140 mM NaCl, 3 mM KCl, 2 mM MgCl2, 2 mM CaCl2, 10 mM glucose, 10 mM HEPES, pH 7·4) containing Fura-2 (5 µg ml1) for 20 min at 37 °C, washed with extracellular solution and mounted under a fluorescence microscope. Treatment at low pH was performed by the addition of extracellular solution containing MES buffered at pH 5·4. The following data are shown: (A) the viral growth curve; (B, B') the intracellular Ca2+ concentration of infected cell layers determined at 2 h p.i. at neutral pH (B) and at low pH (B'); (C, C') the intracellular Ca2+ concentration of infected cell layers determined at 2 h 15 min p.i. at neutral pH (C) and at low pH (C'); (D, D') the intracellular Ca2+ concentration of infected cell layers determined at 3 h p.i. at neutral pH (D) and at low pH (D'); (E, E') the intracellular Ca2+ concentration of a mock-infected control layer determined at 3 h p.i. at neutral pH (E) and at low pH (E'). Intracellular Ca2+ concentrations were monitored by taking two photographs using emission wavelengths of more than 520 nm every 5 s at excitation wavelengths of 340 nm and 380 nm, respectively. The intracellular Ca2+ concentration was determined ratiometrically from these photographs by the Metafluor software, as described previously (Koschinski et al., 2003). The pictures are the resulting pseudo-colour representations of the maximal Ca2+ concentrations, which were reached about 3040 s after incubation at low pH. The colour scale used is shown at the top of the figure.


(23K):

 		Fig. 3. Analysis of ion-permeable pores formed at low pH in SF virus-infected cells. HEK-293 cells grown on two adjacent monolayer cultures were either mock-infected or infected with SF virus as described in the legend to Fig. 2. At 3 h p.i., single cells were subjected to the whole-cell configuration of the patch-clamp technique and treated with extracellular solution at pH 5·4. The time of addition of low pH is indicated by arrows. (A) Current recorded from the plasma membrane of an infected cell, using a holding potential of 40 mV (lower trace) or +40 mV (upper trace). (B) Current recorded from the plasma membrane of an uninfected HEK-293 cell during low pH-induced entry of SF virus via the plasma membrane at a holding potential of 40 mV (lower trace) or +40 mV (upper trace). (C) Currentvoltage relationships. The data shown on the left and right graph were obtained from a virus-infected cell or from an uninfected cell during low pH-induced virus entry, respectively. In both graphs the currentvoltage relationships prior to low pH treatment and 40 s after low pH treatment are shown. Membrane current amplitudes were measured at the end of voltage pulses from 120 to +80 mV in 20 mV steps each lasting for 200 ms. Between every step the potential was switched back for 100 ms to the holding potential of 40 mV. (D) The membrane currents indicated by shaded rectangles in (A) (lower trace, derived from a virus-infected cell at low pH) and in (B) (lower trace, derived from virus entry at low pH) are shown at an enlarged time scale in the left and right graphs, respectively.

The data shown in Fig. 2 indicated that at 3 h p.i. almost all cells of infected cultures formed ion-permeable pores at low pH. This situation is well suited for analyses of these pores by patch-clamp measurements. Such analyses are shown in Fig. 3. Cells were grown at low density in Petri dishes containing two adjacent layers as described above. At 3 h p.i., individual cells were measured using the whole-cell recording configuration of the patch-clamp technique (Hamill et al., 1981) and treated with buffer at pH 5·4 as described previously (Wengler et al., 2003). The cells in the mock-infected control layers did not show ion-permeable pores (n=5, data not shown). Typical results obtained from SF virus-infected HEK-293 cells at membrane holding potentials of 40 mV and +40 mV are shown in Fig. 3(A). A rapid increase of membrane current generated by small steps that led to a maximum inward current of about 0·8 nA was observed after low pH treatment at 40 mV. A similar change in ion-permeability also occurred at +40 mV holding potential and generated an outward current of about +0·8 nA. For comparison, the ion-permeable pores generated during low pH-induced entry of virus particles at the plasma membrane of uninfected cells at 40 mV and at +40 mV holding potential are shown in Fig. 3(B). The pores generated at 40 mV have been analysed previously and the formation of individual pores, which are probably generated during entry of single virus particles and conduct currents of about 12 pA, have been identified (Wengler et al., 2003; Koschinski et al., 2003). The data also showed that at +40 mV membrane potential the virus particles fused into the plasma membrane. The linear currentvoltage relationships shown in Fig. 3(C) indicated that the pores were at least permeable for Na+ and K+. The formation of similar pores has been observed in patch-clamp analyses of SF virus-infected Aedes albopictus cells (Lanzrein et al., 1993). The shaded parts of the traces shown in Fig. 3(A) and (B) are shown at higher time resolution in Fig. 3(D). It can be seen that the permeability changes observed at the membrane of infected cells and during virus entry were qualitatively equal but differed in detail. During virus entry, a group of E1 homotrimers, derived from the 240 E1 molecules of a single virus particle, is inserted into the target membrane and the combined ion permeability of this group is determined as a single event in the corresponding patch-clamp measurements (Koschinski et al., 2003). The data shown in Fig. 3(D) indicated that at the membrane of infected cells such groups of ion channels were not formed and that many more smaller permeability changes were observed. These differences indicated that the permeability changes observed at the plasma membrane of infected cells corresponded to pores generated by the opening and closure of individual or small numbers of pores. It is important to note that the data presented in Fig. 3(B) did not allow us to conclude that the observed virus fusion at positive potential difference across the target membrane led to productive infection, since no attempt was made to detect a virus biochemical function in the cell after the observed fusion. Currently it is not known with certainty whether or not electrical forces in the target membrane play a role in the establishment of a productive infection by alphaviruses (see also Pérez & Carrasco, 1994; Samsonov et al., 2002).

The model of the functions of the E1 protein of alphaviruses during virus entry presented above allows us to combine a number of different experimental observations into a coherent picture: (i) the formation of a fusion pore by the E1 protein in the endosome; (ii) the formation of ion-permeable pores in the plasma membrane at low pH by the E1 protein accumulating in the membrane during virus multiplication; (iii) the co-entry of small molecules into the cytoplasm during virus entry; and (iv) the formation of ion-permeable pores in the plasma membrane during low pH-induced virus entry at the plasma membrane. The formation of ion-permeable pores by a viral protein during virus entry is found in a number of virus systems, including influenza virus (reviewed by Lamb & Krug, 2001) and poliovirus (Danthi et al., 2003). Current evidence indicates that these processes do not have any evolutionary relationship but have evolved independently. In the case of influenza virus, the M2 membrane protein allows a flow of protons from the endosome into the virus particle at low pH and is involved in virus fusion and uncoating. A similar function of the E1 protein has been proposed for alphaviruses (Schlegel et al., 1991; Spyr et al., 1995). This process is directly compatible with the model presented in Fig. 1. Further analyses of alphavirus entry will be required as a result of this model. It will be important to analyse further whether the formation of ion-permeable pores occurs in vivo before, during or after the opening of the fusion pore and whether the formation of the fusion pore might be a necessary prerequisite for the formation of the ion-permeable pores. From a functional point of view, the most important work will probably be direct studies of whether or not the formation of ion-permeable pores is necessary for the establishment of a productive infection, e.g. with a role in core disassembly. From a structural point of view, it remains to be seen whether the target membrane reaction and the self-membrane reaction lead to identical or different stable end-product complexes of the E1 protein.

References

Carrasco, L. (1981). Modification of membrane permeability induced by animal viruses early in infection. Virology 113, 623629.[CrossRef][Medline]

Carrasco, L. (1995). Modification of membrane permeability by animal viruses. Adv Virus Res 45, 61112.[Medline]

Danthi, P., Tosteson, M., Li, Q. & Chow, M. (2003). Genome delivery and ion channel properties are altered in VP4 mutants of poliovirus. J Virol 77, 52665274.[Abstract/Free Full Text]

Dick, M., Barth, B. U. & Kempf, Ch. (1996). The E1 protein is mandatory for pore formation by Semliki Forest virus spikes. Virology 220, 204207.[CrossRef][Medline]

Flint, S. J., Enquist, L. W., Krug, R. M., Racaniello, V. R. & Skalka, A. M. (2000). Virology. Molecular Biology, Pathogenesis, and Control. Washington, DC: ASM Press.

Garoff, H., Wilschut, J., Liljeström, P. & 7 other authors (1994). Assembly and entry mechanisms of Semliki Forest virus. Arch Virol 9, 329338.

Hamill, O. P., Marty, A., Neher, E., Sakmann, B. & Sigworth, F. J. (1981). Improved patch-clamp techniques for high resolution current recording from cells and cell-free membrane patches. Pflügers Arch 391, 85100.[CrossRef][Medline]

Käsermann, F. & Kempf, C. (1996). Low pH-induced pore formation by spike proteins of enveloped viruses. J Gen Virol 77, 30253032.[Abstract/Free Full Text]

Kempf, C., Michel, M. R., Kohler, U. & Koblet, H. (1987). Can viral envelope proteins act as or induce proton channels? Biosci Rep 7, 761769.[CrossRef][Medline]

Kielian, M. (1995). Membrane fusion and the alphavirus life cycle. Adv Virus Res 45, 113151.[Medline]

Koschinski, A., Wengler, G., Wengler, G. & Repp, H. (2003). The membrane proteins of flaviviruses form ion-permeable pores in the target membrane after fusion: identification of the pores and analysis of their possible role in virus infection. J Gen Virol 84, 17111721.[Abstract/Free Full Text]

Lamb, R. A. & Krug, R. M. (2001). Orthomyxoviridae: the viruses and their replication. In Fields Virology, 4th edn, pp. 14871531. Edited by D. M. Knipe & P. M. Howley. Philadelphia: Lippincott Williams & Wilkins.

Lanzrein, M., Weingart, R. & Kempf, C. (1993). pH-dependent pore formation in Semliki Forest virus-infected Aedes albopictus cells. Virology 193, 296302.[CrossRef][Medline]

Marsh, M. & Helenius, A. (1989). Virus entry into animal cells. Adv Virus Res 36, 107151.[Medline]

Nyfeler, S., Senn, K. & Kempf, C. (2001). Expression of Semliki Forest virus E1 protein in Escherichia coli. J Biol Chem 276, 1545315457.[Abstract/Free Full Text]

Pérez, L. & Carrasco, L. (1994). Involvement of the vacuolar H+-ATPase in animal virus entry. J Gen Virol 75, 25952606.[Abstract/Free Full Text]

Samsonov, A. V., Chatterjee, P. K., Razinkov, V. I., Eng, C. H., Kielian, M. & Cohen, F. S. (2002). Effects of membrane potential and sphingolipid structures on fusion of Semliki Forest virus. J Virol 76, 1269112702.[Abstract/Free Full Text]

Schlegel, A., Omar, A., Jentsch, P., Morell, A. & Kempf, C. (1991). Semliki Forest virus envelope proteins function as proton channels. Biosci Rep 11, 243255.[CrossRef][Medline]

Schlesinger, S. & Schlesinger, M. J. (2001). Togaviridae: the viruses and their replication. In Fields Virology, 4th edn, pp. 895916. Edited by D. M. Knipe & P. M. Howley. Philadelphia: Lippincott Williams & Wilkins.

Silver, R. B. (1998). Ratio imaging: practical considerations for measuring intracellular calcium and pH in living tissue. Methods Cell Biol 56, 237251.[Medline]

Spyr, C. A., Kasermann, F. & Kempf, C. (1995). Identification of the pore forming element of Semliki Forest spikes. FEBS Lett 375, 134136.[CrossRef][Medline]

Strauss, J. H. & Strauss, E. G. (1994). The alphaviruses: gene expression, replication and evolution. Microbiol Rev 58, 491562.[Abstract/Free Full Text]

Wengler, G. & Wengler, G. (2002). In vitro analyses of factors involved in the disassembly of Sindbis virus cores by 60S ribosomal subunits identify a possible role of low pH in this process. J Gen Virol 83, 24172426.[Abstract/Free Full Text]

Wengler, G., Koschinski, A., Wengler, G. & Dreyer, F. (2003). Entry of alphaviruses at the plasma membrane converts the viral surface proteins into an ion-permeable pore that can be detected by electrophysiological analyses of whole-cell membrane currents. J Gen Virol 84, 173181.[Abstract/Free Full Text]

Young, J. A. T. (2001). Togaviridae: virus entry and uncoating. In Fields Virology, 4th edn, pp. 87103. Edited by D. M. Knipe & P. M. Howley. Philadelphia: Lippincott Williams & Wilkins.

Received 25 November 2003; accepted 26 January 2004.