Viral infection

Abstract

Viruses have developed different survival strategies in host cells by crossing cell-membrane compartments, during different steps of their viral life cycle. In fact, the non-regenerative viral membrane of enveloped viruses needs to encounter the dynamic cell-host membrane, during early steps of the infection process, in which both membranes fuse, either at cell-surface or in an endocytic compartment, to promote viral entry and infection. Once inside the cell, many viruses accomplish their replication process through exploiting or modulating membrane traffic, and generating specialized compartments to assure viral replication, viral budding and spreading, which also serve to evade the immune responses against the pathogen. In this review, we have attempted to present some data that highlight the importance of membrane dynamics during viral entry and replicative processes, in order to understand how viruses use and move through different complex and dynamic cell-membrane structures and how they use them to persist.

Viruses are small structures lacking intravital metabolic pathways and mechanisms to assure their own survival and therefore, depend a fortiori on host-cell machinery to replicate their DNA or RNA genome, and to spread their infectious progeny. Hence, many viruses have developed different survival strategies to enter and infect cells, or to accomplish their replication process through exploiting or modulating membrane traffic and generating specialized compartments1–3 (Figs. 1–3), which also serve to evade the immune responses. In turn, increasing evidence points to that membrane dynamics, like the traffic of vesicles and their spatial reorganization, is key to cell defense against pathogen infections, as in the case of neutrophil-mediated phagocytosis, where orchestrated secretion of granules and vesicles allows cells to locate and destroy invading microorganisms in a controlled manner.4–7 Viruses use structural or non-structural proteins to exploit the major cellular trafficking pathways to navigate across their target cells by recruiting clathrin, coatomer protein complex (COPI) I and II or endosomal sorting complex required for transport (ESCRT) and their accessory proteins, in a non-structural or structural viral proteins-dependent manner (excellently reviewed in ref. 2, 3 and 8–10). In addition, many of the cellular proteins recruited by the virus to accomplish its viral cycle are small GTPases that are able to generate, move and/or fuse different endosomes or vesicular compartments.1–3,9

This work is not a comprehensive review of all membrane dynamics events reported to occur during the cycle of infection of different viruses. Instead, we have attempted to present some data that highlight the importance of the constant flux of membrane structures during early viral entry and replication processes, in order to understand how viruses move through the different complex and dynamic cell-membrane compartments to survive.

Viral Entry and Membrane Dynamics

As the initial barrier to viral entry, the plasma membrane together with membrane-trafficking machinery is also of fundamental importance in the first stages of the viral cycle.2,11–14 Although the mechanisms of viral entry for non-enveloped virus are poorly characterized, the molecular mechanism involved in viral fusion and entry for enveloped virus begins to be more clearly understood.2 It is thought that certain enveloped viruses such as human immunodeficiency virus type 1 (HIV-1), herpes simplex virus 1 (HSV-1), Sendai virus and many other retroviruses have pH-independent viral fusion proteins which allow the virus to penetrate into cells by fusing directly with the plasma membrane2,13,15–20 (Fig. 1; Viral fusion and entry and Viral endocytosis schemes).

Plasma membrane morphology and polarization appears to be regulated by cell cytoskeleton reorganization and direct association with the different components of cellular cortex,21–26 and by the insertion and uptake of membrane structures at cell-surface that regulates plasma membrane dynamics.27–31 Although cytoskeleton reorganization and dynamics have well-documented roles in HIV-1 fusion and entry events, the contribution of plasma membrane dynamics is less clear during these early viral infection steps. HIV-1 interacts with target cells through cell-surface CD4 and CXCR4 or CCR5 viral co-receptor, a process that is cooperative and requires cell signaling, actin polymerization and reorganization32–41 and stabilization of microtubules,22,42,43 in order to achieve pore fusion formation throughout which viruses reach the inner cell. In this concern, it has been reported that HIV-1 fusion and entry could occur in micropinosomes and endosomes,44,45 a process that has been proposed to be clathrindependent,46 pH-independent and dynamin-dependent.47,48 In fact, dynamin- and endosome-dependent HIV-1 entry and infection have been recently controverted.49–51

In this regard, it has been reported that HIV-1 internalization and infection in polarized trophoblasts is a pH-dependent process,52,53 which is driven by a clathrin-, caveolae- and dynamin-independent endocytic pathway, and requires free membrane cholesterol.54 A recent work from our group suggested that the fluidity of plasma membrane, regulated by the phosphatidyl-inositol-4-phosphate-5-kinase Iα (PI4P5-K Iα) and subsequent phosphatidylinositol-4,5-biphosphate (PIP₂) production, is crucial for HIV-1 entry and the early steps of infection in permissive lymphocytes.39 PIP₂ and the PI4P5-K Iα are functionally linked to the small GTPase, ADP-ribosylation factor 6 (Arf6), which regulates membrane trafficking and regeneration of plasma membrane.55–58 Furthermore, efficient early HIV-1 fusion, entry and infection require both an Arf6-dependent dynamic and regenerative plasma membrane at the virus/cell-surface interacting regions,51 and a correct cellsurface localization of viral receptors (Fig. 1 and Viral fusion and entry scheme, and associated Fig. 2). Thus, movement of PIP₂-associated membrane structures, driven by the Arf6-GTP/GDP cycle activity on plasma membrane, assures the regeneration of cell-surface membrane by coordinating the turnover of these PIP₂-associated vesicles, which has synergy with the key first HIV-1/receptors interactions.37,40 Altogether, this process promotes pore fusion formation, between the non-regenerative HIV-1 viral membrane and the dynamic cell-surface, thereby favoring efficient cell-to-cell viral transmission, entry and infection51 (Fig. 2). The alteration of the Arf6-GTP/GDP cycle either by using GDP-bound or GTP-bound inactive mutants, which provokes the accumulation of Arf6/PIP₂-associated membrane structures on cell-surface, or by specific Arf6 silencing inhibits HIV-1-envelope-induced membrane fusion, entry and infection of T lymphocytes and permissive cells, regardless of viral tropism. Conversely, Arf6 silencing or its dominant mutants did not affect fusion, entry and infection of viruses pseudotyped with the envelope-G protein of the vesicular stomatitis virus (VSV-G), or ligand-induced CXCR4 or CCR5 endocytosis, both clathrin-dependent processes. These results confirmed that early HIV-1 infection of CD4⁺ T lymphocytes requires Arf6-coordinated plasma membrane dynamics in order to promotes viral fusion and entry. The non-endocytic route followed by HIV-1 during early infection seems to be decisive to establish viral latent infection,49 thus Arf6-regulated HIV-1 pathway of infection could be key for HIV-1 infection and pathogenesis.

In this concern, a new broad-spectrum antiviral agent, acting only against enveloped viruses (e.g., HIV-1, VSV, Ebola, Marburg, influenza, hepatitis C (HCV) and West Nile viruses), inhibited free virus-cell fusion and infection by perturbing non-regenerative viral membranes, without affecting non-enveloped viruses in vitro (e.g., adenovirus, coxsackievirus and reovirus). However, this molecule failed to block cell-to-cell fusion and infection due to its inactivity on dynamic, regenerative plasma membranes.59 Considering that Arf6 is key to coordinate plasma membrane dynamics, and its functional implication on cellular invasion by HIV-1 and several microorganisms,51,60–66 it is therefore plausible that different enveloped viruses may also benefit from Arf6-coordinated plasma membrane traffic to promote entry and infection processes. Arf6 is the only member of the Ras-related Arf-family of small GTPases that affects cell-surface dynamics, thereby regulating plasma membrane/endosome trafficking and cortical actin reorganization.58,67–73 Remarkably, plasma membrane morphology and dynamics is also regulated by the traffic of PIP₂-associated membranes from plasma membrane to a non-clathrin intracellular compartment,67,68,74–78 which in turn relies on the membrane transport activity of Arf6.21,56–58,67–69,79 Therefore, it is also conceivable that Arf6-coordinated membrane movements might control entry and infection processes of some non-enveloped viruses, such as coxsackievirus.80

In the case of the VSV virus, which has had an important role in our increasing understanding of both innate and acquired immunity, as well as virology in general,81 the viral-envelope glycoprotein G binds to phosphatidylserine, a near-universal component of cell-surface membranes, enabling VSV to infect virtually all animal cells. Although phosphatidylserine had been proposed to serve as VSV receptor,82 this conviction has been recently challenged.83 Following attachment to the cell surface, the virus enters by endocytosis and, after a subsequent drop in endosomal pH, the glycoprotein catalyzes the fusion of viral and cellular membranes, releasing the viral ribonucleoprotein (RNP) into the cytoplasm. In fact, VSV fusion occurs in multivesicular transport intermediates, formed between early and late endosome, and not in late endosomes,84,85 where finally viral nucleocapsid is delivered to the cytoplasm (Fig. 1 and Viral endocytosis scheme). Of note, the extensive tissue tropism of VSV enables this virus to be therefore used as an anti-cancer agent in all types of tumors.81

Other viruses could enter cells by more than one mechanism that implies the passage from cell-surface through different membrane structures (Fig. 1 and Viral endocytosis scheme). The enveloped influenza retrovirus mainly enters cells by clathrin-mediated endocytosis, but it can also follow a clathrin-independent pathway,86 which appears to be regulated by Epsin 1.87 Moreover, simian virus 40 (SV40) is alternatively internalized either via caveolae88 or a CLICs/GEECs (clathrin-independent carriers/glycosylphosphatidylinositol-anchored proteins and enriched early endosomal compartments)-driven pathway.3,89,90 The capsid of the non-enveloped DNA SV40 virus is composed of VP1 homopentamers, which resemble cholera toxin B pentamers, and like the cholera toxin B subunit, SV40 binds gangliosides on the cell surface and enters via a raft/caveolar pathway.2,81,90,91 Echovirus 1, a picornavirus that binds to integrins, follows a caveolar/raft uptake process that involves protein kinase C, and penetration seems to occur in caveosomes without the involvement of the endoplasmic reticulum (ER).2,92,93 For some viruses, such as for Ebola virus, SARS coronavirus (that causes severe acute respiratory syndrome [SARS]), and the non-enveloped mammalian reoviruses, an acidic pH alone is not sufficient to induce virus-endosome fusion and entry, thus participation of cathepsins L and B acid-dependent endosomal proteases are required to acquire the penetration-competent state of viral proteins.94–96

It has been described for certain viruses, such as HIV-1, VSV, murine leukemia virus, human papillomavirus type 16 and the DNA-enveloped vaccinia virus, that before entry into their target cells these viruses interact with and surf upon cell-surface filopodia toward the base of the filopodia,97–99 where certain virus enter cells by macropinocytosis, as echovirus 1 and adenovirus serotype 3,100–102 or by triggering the formation of transient membrane blebs99 (Fig. 3 and Surfing on filopodia in Viral fusion and entry scheme, and surfing in viral synapse).

Altogether, the viral endocytic route could represent an easy way to overcome the barrier that cortical cytoskeleton imposes to the virus during early entry and infection processes, being also of importance to evade the immune responses against viruses.2,3,81

Viral Replication and Membrane Dynamics

Intracellular membrane dynamics is not only essential during the first stages of viral fusion and subsequent entry but also to accomplish the viral replicative process. Replication of many viruses is associated with specific intracellular compartments so-called viral factories (VF; see VF schemes in Fig. 1). These are thought to provide a physical scaffold to concentrate viral components for genome replication and morphogenesis.8,103,104 The formation of VF often results in rearrangement of cellular membranes, reorganization of the cytoskeleton and recruitment of mitochondria.105,106 One of the early events in factory formation is the assembly of replication complexes (RCs) that associates with membranes derived from the ER, as observed in the case of flaviviruses, hepaciviruses, coronaviruses, arteriviruses and picornaviruses (Fig. 1 and VF schemes). On the other hand, RC of Togaviruses associate with membranes of endocytic origin, whereas RC of nodaviruses associate with mitochondrial membranes (reviewed in ref. 8). In fact, RNA viruses modify specific membranes of the factory to concentrate viral replicases and necessary cofactors for a more efficient replication of the viral genome.107–113 In this regard, rubella virus (RUBV), an important human teratogenic virus and the only member of the genus Rubivirus in the Togaviridae family,114 and Semliki Forest virus (SFV),115 anchors its RNA synthesis in membranes of a cell organelle known as the cytopathic vacuole (CPV),116–118 which derives from modified endosomes and lysosomes.109,117,119 In fact, ER cisternae, mitochondria and Golgi stacks are recruited around CPVs to build RUBV factories106 (Fig. 1 and VF schemes), which help the virus to evade host cell defense responses as well as connecting viral replication with assembly and maturation of new viral particles in recruited Golgi membranes.120 The Golgi apparatus is a highly dynamic organelle with functional and sustained membrane and protein flow121 and serves as a morphogenic mold for rubiviruses and other viruses, such as coronaviruses, arteriviruses and Bunyaviruses.8,105,106,120,122,123 The surface of CPVs consists of small vesicular invaginations or spherules (the sites of viral RNA replication) that line the vacuole membrane at regular intervals.109,117–119,124 These CPV structures also exhibit a variety of complex contacts with the endocytic pathway through its internal membranes that are interconnected with different transport vesicles.106 A remarkably case of VF occurs with the poxvirus vaccinia virus, an example of DNA-virus that does not replicate inside the nucleus of host cells. Thus, discrete cisternae derived from the rough ER enclose the cytoplasmic site of viral-DNA replication and it is thought to eventually resemble a cytoplasmic mini-nucleus for viral replication.125–127

Other RNA⁺ viruses that belong to the Flaviviridae family and the Nidovirales order induce the formation of spherical, double-membrane vesicles (DMVs) to replicate128–130 (Fig. 1 and double-membrane vesicles scheme), while polioviruses-RNA polymerase molecules can also assemble bidimensional arrays.110,131 In fact, one of the best-studied viruses that induce membrane rearrangements is the human pathogen poliovirus, a member of the Picornaviridae family that causes poliomyelitis. Thus the trafficking and alteration of intracellular membrane structures regulate poliovirus infection, where all cellular organelles except mitochondria are virtually converted into virus replication vesicles.132–136 Indeed, endosomes and lysosomes (Togaviruses), peroxisomes and chloroplasts (members of the genus Tombusvirus) and mitochondria (nodaviruses) represents protective environments used as sites for RNA replication, while all plus-strand RNA viruses replicate in association with cytoplasmic membranes of infected cells.107 Some data about the antiviral effect of brefeldin A (BFA) suggest the functional involvement of membrane trafficking proteins on virus replication of Enteroviruses, such as for the picornaviruses poliovirus and coxsackievirus.137 BFA inhibits Arf-GTP exchange proteins (Arf-GEFS), reduces the Arf1-GTP needed to generate COPI coats in the Golgi and blocks the recruitment of membranes into replication compartments.133 Thus sequestration of Arf-1 into the replication complex would also explain the block in secretion seen in infected cells.138 However, BFA does not prevent formation of densely packed vesicles by poliovirus, so vesicle formation does not require activated Arf-GEF proteins.136 Several picornaviruses are resistant to BFA, and their 3A proteins do not slow ER to Golgi transport as occurred with foot-and-mouth disease virus.139–141 For these viruses COPII coated vesicles may provide membranes for replication,142–145 while other studies implicate a role for autophagosomes.146

The induction of coated-pit formation has also been observed for reovirus and SFV.2,147 Internalized SFV early recruits the intermediate-endosome Rab7-small GTPase148 to later induce the formation of CPVs, important for viral RNA synthesis,118 which are derived from late endocytic compartments. In the case of HCV virus, it seems that recruitment of membrane trafficking proteins to ER-derived membrane scaffolds is key for viral replication. Hence, the NS5A protein is anchored to the cytoplasmic face of this membrane web to recruit the RNA-dependent RNA polymerase, NS5B. Both NS5A and NS5B bind VAMP associated proteins (VAPs) 149,150 and recruit Rab1, Rab5 and Rab7 small GTPases that could direct the transport of vesicles to fuse with or enlarge this viral replicative membrane-web compartment.9,151–153 It is conceivable that other RNA⁺ viruses such as Norwalk virus take advantage of the formation of ER-derived membrane scaffolds to replicate, as observed with the membrane-bound nsp48 protein that also binds VAP-A.9,154

The cell-host plasma membrane undergoes an immense rearrangement and a deformability process during viral budding, a key event of the life cycle of enveloped viruses, thereby determining viral morphology and infectiveness (reviewed in ref. 10) (Figs. 1 and 3, Budding schemes). The majority of studied enveloped viruses bud from cells by co-opting the host ESCRT machinery,155–159 critical for budding of vesicles in multivesicular bodies (MVBs) that are important intermediates in endolysosomal transport.10,160,161 The HIV-1 viral polyprotein Gag binds, through its C-terminal region, both to cellular ESCRT-I complex and to ALIX protein, which both recruit ESCRT-III complex to the budding site to catalyze the scission of nascent virions, an event that it is thought to be carried out by the same process as for cleavage of intralumenal vesicles in MVBs.10,162

Furthermore, the tumor susceptibility gene 101 (Tsg101) subunit of the ESCRT-I complex, which mediates receptor sorting into MVBs,163,164 appears to be dispensable for viral envelope fusion with endosomal membranes and viral RNA transport to late endosomes, but is necessary for infection and RNA release.85 These data indicate that Tsg101, in addition to its role in receptor sorting into MVBs, may well play a direct role in the release of nucleocapsids from within MVBs to the cytoplasm by controlling the back-fusion process. However, several recently reported data indicate that the interaction of Gag viral protein with the ESCRT machinery is dispensable for HIV-1 viral budding and infectiveness.165–167 In this regard, the matrix protein of the enveloped influenza virus lacks an ESCRT binding domain, thus it buds in an ESCRT-independent manner.168,169

In turn, the HIV-1 Gag localizes on PIP₂-enriched plasma membrane regions, where PIP₂ plays a critical role in HIV-1 particle assembly.170,171 Overexpression of polyphosphoinositide-5-phosphatase IV (5ptaseIV), which hydrolyzes the phosphate at the D5 position and reduces plasma membrane PIP₂ levels,172 drastically impairs HIV-1 release by relocalizing the viral polypeptide Gag from the plasma membrane to CD63-positive intracellular compartments.171 Indeed, disrupting PIP₂ levels and localization by expression of Arf6/Q67L, a GTP-bound mutant of Arf6, reduces virus release by retargeting Gag to newly formed PIP₂-enriched endosomal vesicles.171

Finally, other viruses bud from plasma membrane of host cells through their matrix viral protein. This is the case of Newcastle disease virus (a Paramyxovirus) and VSV virus, where bud formation and scission from membranes are both matrix-dependent processes.173–175

Viral Spread: Membrane Dynamics at the Virological Synapse

There is growing evidence that a number of different viruses can exploit pre-existing mechanisms of physiological communication between cells to facilitate direct cell-cell viral spread.38 Studies revealing the co-clustering of viral egress machinery and viral receptors at the interface between conjugates of infected and uninfected cells respectively, have led to the definition of this structure as virological synapse (VS) 176,177 (Fig. 3 and Viral synapse scheme). The first detailed description of VS was reported for the human T cell leukemia virus type 1 (HTLV-1),178 and was then translated to the HIV field.179 The cell-free form of HTLV-1 is very inefficient at infecting T cells and is spread between, and within individuals by strictly dependent cell-to-cell transmission mechanisms. Noteworthy, cell-to-cell viral spread presents numerous advantages for the virus as compared to cell-free virus infections, which could be summarized as follows:

• A more rapid replication kinetics is obtained when viruses are transmitted across VS due to the higher concentration of viral particles released at the point of contact between the cell partners thus obviating the rate-limiting step of fluid-phase diffusion of free viral particles.180

• Infected cells entering a new host could adhere to, and cross by transmigration, a mucosal epithelial barrier that would otherwise be impermeable to cell free virions.38

• The formation of stable junctions whereby viruses traffic, shields the virus from immune response or drugs both sterically and kinetically in terms of exposure time although it is still unclear whether this is also the case for HIV.181,182

At the molecular level, VS synchronize both viral egress and entry processes in the synaptic cleft. Most recent literature suggests that HIV-1 cell-to-cell transmission results from microtubules-mediated polarized and massive HIV budding and subsequent entry of viral particles into target cells182,183 (Fig. 3 and Viral synapse scheme). Therefore, most of the mechanisms that regulate cell-free virus entry by perturbing membrane dynamics also apply for cell-to-cell virus transmission. Indeed, it is known that membrane-cholesterol sequestering agents184 and inhibitors of cytoskeleton motility185 interfere with retroviral transmission between infected and uninfected cells. Similarly, the blockade of Arf6-coordinated plasma membrane dynamics by siRNA-Arf6 silencing or expression of dominant mutants has detrimental effects on cell-to-cell HIV-1 transmission51 (Fig. 3 and Arf6-dependent viral fusion and entry scheme).

Despite these similarities, some specific events of synaptic contacts that involve membrane dynamics should be also considered. It is well known that massive viral endocytosis accompanies viral entry during cell-to-cell HIV transmission186 (Fig. 3). In contrast, HTLV-1 appears to be transmitted from biofilm-like structures that accumulate viral particles on the surface of infected cells.187 Other viruses have developed alternative strategies, pseudo-rabies virus transferred from cellular projections, HSV are transported within long membrane protrusions towards adjacent cells, and murine leukaemia virus moves along the surface of filopodia before viral entry at the cell body of fibroblasts188 (Fig. 3 and surfing scheme at Viral synapse). Such filopodia that also participate in HIV transmission extend through receptor-mediated mechanisms from uninfected towards the infected cell.188 A similar retroviral surfing has been described over narrower membranous structures called nanotubes that can connect cells separated by up to 100 mm,189 which also transfer viral proteins at the inner inside. In this case, receptor specificity seems to play a minor role, as these actin-driven structures seems to extend from HIV infected cells in the absence of receptor-envelope interactions (Fig. 3 and Nanotubes scheme).

The formation of different bridging tubular structures within the synapses, usually involve huge membrane invaginations engulfing cellular fragments,183 which may lead to the exchange of membranes between counteracting cells, a synapse-specific event that is known as trogocytosis.190 Trogocytic exchange of membrane patches, initially described during cellular contacts among cells of the immune system, has been proposed as a possible mechanism to control the length and the stability of the synapse and therefore regulate its outcome.190 Recently, trogocytosis has been reported to occur at the HIV induced VS, a process dependent on gp120 binding to CD4. Interestingly, HIV particles as well as membrane components (such as CD4 molecules) were transferred in an unidirectional way from the uninfected towards the infected cells182 (Fig. 3 and Trogocytosis scheme). This mechanism could have a huge impact rendering cells permissive to HIV infection as suggested in a recent study.191

Besides the direct transfer from infected to uninfected cells, several pathogens take advantage of other mechanisms of immune communication between antigen presenting cells, mainly dendritic cells (DCs) and T cells. As DCs are professional pathogen hunters, several viruses, such as HIV or cytomegalovirus (CMV), have developed strategies to stably associate in an infectious state to these cells and infect lymphocytes during antigen presentation.192,193 HIV has been shown to reside in CD81 enriched salk-like structures inside DC that are released at the synaptic space during T cell scanning or antigen presentation.193 Interestingly, the journey of HIV in DCs appears to be reminiscent of the exosomal pathway of antigen presentation (Figs. 1 and 3, Exosome schemes), suggesting that HIV hijacks this cellular mechanism in its own benefit and highlighting the similarities between retroviruses and exosomes.194 Importantly, multiple pathogens, including some bacteria and other parasites, hijack the complex mechanisms of recognition and capture by DC to increase their spread efficiency in the infected hosts.195

Conclusions and Perspectives

This article attempt a non-comprehensive review about the journey that different viruses accomplish to successfully infect their target cells, describing viral entry pathways and intracellular trafficking, from cell-surface and through complex and dynamic cell-membrane structures, which often are newly organized by the viruses in order to replicate. The different host intracellular compartments, such as ER, endosomes, lysosomes or mitochondria, serve as an adaptable membrane source to form viral replication factories and allow viruses to adopt their own functional morphology with the appropriated lipid composition. Indeed, it is conceivable that these membrane structures help to confine the process of RNA replication to a specific cytoplasmic location, just preventing the activation of certain host defense mechanisms that can be triggered by the viral genome during unwinding and replication.

Although several data have been reported about the functional role of membrane dynamics during viral infection, many related questions remain to be addressed. For example, how could different viral proteins, from divergent viruses and host cells, gain the ability to modify the same intracellular membrane compartments to assure viral replication, without critically affecting important cellular processes? In turn, why viruses choose different subcellular-membrane compartments for their replication? And, how viral proteins are targeted to and moving across those membranes and what host factors are recruited or involved in these events? The answer of these questions will surely contribute to our understanding of some of the basic mechanisms that control membrane dynamics, both in cellular functions and viral infection cycle. This may open the door to the design of new antiviral strategies aimed at effectively target the dynamic of viral-cell interactions, which hopefully would lead to new therapies for combat viral infections, such as HIV-1 infection and Acquired Immunodeficiency Disease Syndrome (AIDS). Moreover, it may also provide new rational designs by which non-viral gene delivery systems can be improved and therapeutically used in different gene-originated tumor and immune diseases.

Finally, the two- and three-dimensional study of viral fusion, entry and trafficking at the level of cells and virus particles will bring the technical development of new and more powerful microscopy devices, from fluorescent and high-end light microscopy and total internal reflection fluorescence microscopy (TIRFM) to transmission electron microscopy (TEM) with cellular electron tomography (ET),51,106,196–199 which allow to analyze these events with increased spatial and temporal resolution, in order to understand how viruses move through the different complex and dynamic cell-membrane structures to infect and survive.

Figures and Tables

Figure 1 Membrane dynamics processes in host cells that are involved in the life cycle of viral infection. The schematic representation of a cell and the different intracellular membrane-compartments, constitutive or putative sites formed by viruses to accomplish their infection processes, are shown. The non-regenerative viral membrane of enveloped viruses and the dynamics cell-host membranes play an important role during early infection process, since these two opposed membranes need to fuse, either at cell-surface or in an endocytic route (clathrin-, caveolae- or pinocytosis-dependent), to promote viral entry and infection. Endocytosis is initiated at the plasma membrane and progress through early and late endosomes, where some viruses replicate and are recycled back to the plasma membrane or transported to lysosomes to complete the life cycle. In the case of HIV-1, this enveloped virus requires Arf6-membrane dynamics to efficiently fuse with plasma membrane and promote entry and infection of CD4⁺ T lymphocytes (Asterisk scheme and Fig. 2). The non-endocytic route followed by HIV-1 during early infection is decisive to establish viral latent infection. Once inside the cell, many viruses accomplish their replication process through exploiting or modulating membrane traffic, and generating specialized compartments to assure viral survival, such as Viral Factories (VF), multivesicular bodies (MVB), double-membrane compartments, budding on plasma membrane and exosomes (it is conceivable that some viruses may actually be released as exosomes ). These membrane structures, cell-constitutive or arranged by the different viral proteins, are required for viral-gene replication, morphogenesis, export, viral maturation and release from cell-surface, and also serve to evade the immune responses against viral genomes. Viral proteins could enter the secretory pathway by co-translational translocation into the endoplasmic reticulum (ER; only a part of the perinuclear ER is shown), to be further transported from the ER to the Golgi complex in vesicles and in a coatomer protein complex (COP) II-dependent manner. Viral complexes formed inside MVBs, in communication with vesicles, mitochondria, Golgi cisternae and ER-membranes, could be transported through the Golgi network to the plasma membrane to be released as viral particles. Viral budding of enveloped viruses is mainly under the control of the activity of ESCRT-III complexes that are recruited to the site of viral release by ESCRT-I or Alix proteins that interacts with matrix viral proteins located on cell-surface.

Figure 2 Arf6-membrane dynamics regulates efficient HIV-1 infection. HIV-1 requires Arf6-coordinated membrane dynamics to efficiently fuse with plasma membrane and promote entry and infection of CD4⁺ T lymphocytes. In fact, movement of PIP2-associated membrane structures, driven by the Arf6-GTP/GDP cycle activity on plasma membrane from a sorting and recycling endosomal compartment, assures the regeneration of cell-surface membrane by coordinating the turnover of these PIP₂-associated vesicles. This membrane traffic has synergy with the key first HIV-1/receptors interactions to promote pore fusion formation, between the non-regenerative HIV-1 viral membrane and the dynamic cell-surface, thereby favouring efficient virus-cell fusion, entry and infection (scheme corresponding to early fusion and entry steps of the HIV-1 infection process, also indicated in Fig. 1 by an asterisk). The alteration of the Arf6-GTP/GDP cycle, by GDP-bound or GTP-bound mutants provokes an accumulation of Arf6/PIP₂-membrane structures on the plasma membrane. Specific Arf6 silencing also inhibits HIV-1-envelope-induced membrane fusion, entry and infection of T lymphocytes and permissive cells, regardless of viral tropism.

Figure 3 Membrane dynamics at the virological synapse. At the virological synapse (VS), some viruses either attach to the plasma membrane or surf along the filopodia and finally bind to specific receptors on the target cell. Viruses can also directly fuse with the plasma membrane, as in the case of HIV-1. Cell-to-cell transfer of HIV-1 takes place in an Arf6-membrane dynamics-dependent manner, which hijack endocytic pathways, including clathrin-dependent, caveolin-dependent or both independent pathways for viral internalization. The VS represents an efficient environment for viral budding, where the membrane of the infected cell is polarized towards the synaptic junction by the movement for of vesicles or MVBs coordinated by the translocation of the microtubule organizing centre (MTOC). This scaffolding allows for a subsequent viral infection and spread, that favours viral fusion and entry, viral endocytosis and viral protein/gene transfer from the infected to the close non-infected cell. Besides, long membrane nanotubes may also be formed between neighboring cells, which promote viral protein traffic and also HIV surfing and infection, from infected cell to non-infected cell. Other membrane dynamics events involved or occurred during viral infection and spreading are Trogocytosis and exosomal transport. Trogocytosis of cell-surface patches, containing CD4/HIV-1-bound molecules, occurs from non-infected to infected cells in a gp120/CD4-dependent manner. Exosomes are membrane vesicles, formed from MVB that could account for viral infection and spreading within membrane structures that are protected from immune responses.

Acknowledgments

This work and A.V.F. are supported by SAF2008-01729 (MICINN, Spain), European Regional Development Fund (ERDF), 24661/07 and 24-0740-09 (Fundacion Investigacion y Prevencion del SIDA en España), and ProID20100020 (Agencia Canaria de Investigación, Innovación y Soc. Información; Gobierno de Canarias) grants. The HIVACAT Program, the FIS project PI08/1306 and the Spanish AIDS network (RD06/0006). J.B.G., L.G.E. and L. de A.R. are supported by FIPSE-24-0740-09, SAF2008-01729 and ProID20100020 associated fellowships, respectively. J.D.M. is supported by the Ramón y Cajal program (R&C-2010-06256, MICINN). J.B. is supported by the ISCIII and Health Department (Generalitat de Catalunya). The authors declare that they have no conflicting financial interests. We specially thank to Prof. Manuel Feria for critical reading of the manuscript and for his continuous and generous support. We apologize for all research works and reviews that we have not reported or considered in this minireview, where we have tried to avoid omissions by our unpremeditated unknowledge.