Regulation of LFA-1 Activity through Cytoskeleton Remodeling and Signaling Components Modulates the Efficiency of HIV Type-1 Entry in Activa
http://www.100md.com
免疫学杂志 2005年第14期
Abstract
Besides interactions between the viral envelope glycoproteins with cell surface receptors, interactions between cell-derived molecules incorporated onto virions and their ligand could also modulate HIV type-1 (HIV-1) entry inside CD4+ T lymphocytes. Although incorporation of host ICAM-1 within HIV-1 increases both virus attachment and fusion, the precise mechanism through which this phenomenon is occurring is still unclear. We demonstrate in this study that activation of primary human CD4+ T lymphocytes increases LFA-1 affinity and avidity states, two events promoting the early events of the HIV-1 replication cycle through interactions between virus-embedded host ICAM-1 and LFA-1 clusters. Confocal analyses suggest that HIV-1 is concentrated in microdomains rich in LFA-1 clusters that also contain CD4 and CXCR4 molecules. Experiments performed with specific inhibitors revealed that entry of HIV-1 in activated CD4+ T cells is regulated by LFA-1-dependent ZAP70, phospholipase C1, and calpain enzymatic activities. By using laboratory and clinical strains of HIV-1 produced in primary human cells, we demonstrate the importance of the LFA-1 activation state and cluster formation in the initial step of the virus life cycle. Overall, these data provide new insights into the complex molecular events involved in HIV-1 binding and entry.
Introduction
Entry of HIV type-1 (HIV-1)3 into susceptible target cells is a complex and dynamic process involving several microevents. Besides the normal association between the external envelope gp120 and the appropriate cell surface receptors, it has been proposed that interactions between host-derived molecules present on the virion and their cognate receptors on the target cell surface can also contribute to the formation of a fusion-competent environment. It is well-established that HIV-1 incorporates a vast array of host cell membrane molecules during budding, including the ICAM-1 (1, 2, 3). Interestingly, engagement of ICAM-1 with its natural counterligand LFA-1 increases viral infectivity by several folds. In CD4+ T lymphocytes that express functional LFA-1, this phenomenon is associated with an enhancement of both HIV-1 attachment and virus-cell fusion. It has been shown that the presence of host ICAM-1 onto virions favors fusion rather than endocytosis, the former being associated with a more productive HIV-1 infection process (4). Moreover, CD4 and CXCR4 molecules are essential to achieve an efficient infection with viruses lacking host ICAM-1 as well as with isogenic ICAM-1-bearing virions (4). The presence of virus-anchored host ICAM-1 confers also resistance to neutralization by Abs and to the fusion inhibitor T-20 (5, 6). This reduced susceptibility to the antiviral activity of T-20 illustrates the critical role played by ICAM-1/LFA-1 interaction in the initial steps of the virus replication cycle. The exact molecular process by which ICAM-1-bearing virions are more efficiently entering LFA-1-expressing cells remains to be more clearly defined.
LFA-1-mediated adhesion is playing a key role in immune surveillance and the mounting of a potent immune response. More specifically, LFA-1 is involved in the arrest of rolling lymphocytes along blood vessels facilitating extravasation and migration of the activated T cells to infection sites. The 2 integrin LFA-1 participates to the formation of the immunological synapse between T cells and APCs. These events require a rapid modulation of adhesion/deadhesion and the adhesion needs to be stable for hours to sustain the immunological synapse. Hence, activation of LFA-1 must be tightly regulated. Two mechanisms have been proposed for the regulation of LFA-1-mediated adhesion that include modulation of affinity and avidity (7, 8, 9). Affinity regulation refers to changes in individual L2 heterodimer that are linked to modifications in LFA-1 conformation. Recently, it has been discovered that LFA-1 can take three binding affinities, i.e., low, intermediate and high (10). In the low affinity conformation, the integrin adopts a bent state for which the head domain points toward cell membrane and cannot interact with its ligand. The acquisition of the intermediate and high affinity conformational states involves the unbending of both subunits to form the extended state. The intermediate binding affinity is induced by inside-out signaling triggered by an immunological synapse, chemokine, or phorbol ester treatment (11), whereas the induction of LFA-1 under a high affinity state is achieved upon ligand binding and involves separation of the and subunits at their cytoplasmic, transmembrane, and leg domains (12, 13). The high affinity state can be also generated in vitro by exposing cells to divalent cations such as manganese or magnesium (8). The avidity state of LFA-1 refers to distribution of the integrins in plasma membrane. Reorganization of LFA-1 into micro- and macroclusters on the cell surface follows ligand binding and requires the release of the integrin tail to the actin cytoskeleton. Engagement of ICAM-1 induces outside-in signaling that triggers lateral diffusion of heterodimers in plasma membrane, a process regulating the adhesion strengthening (14, 15, 16, 17, 18). In naive T cells, LFA-1 is maintained under a low affinity/avidity state (19, 20). Upon T cell activation, the inside-out signaling generates the intermediate affinity state and activates the cysteine protease calpain that releases LFA-1 from its cytoskeleton constraint and favors lateral diffusion of LFA-1 (12, 18, 21, 22). These phenomena trigger homotypic adhesion which induces in turn the formation of micro- and macroclusters of LFA-1 at cell-cell contact interfaces (11). While affinity and avidity are differently controlled, the two mechanisms act in a concerted manner to achieve and maintain stable cell-to-cell adhesion (23, 24, 25). Furthermore, ligand-mediated signaling events have been shown to be important for the creation of new linkages with the actin cytoskeleton that reinforce and maintain stable adhesion (18).
Considering that LFA-1 is tightly regulated and its activation state is crucial to permit a firm and stable ICAM-1 binding, we addressed whether the LFA-1 activation state, intermediate lateral mobility and clustering can lead to a more efficient virus attachment and entry. Our results show that capture and internalization of ICAM-1-bearing virions are more efficient in cells expressing activated LFA-1 molecules. Moreover, LFA-1-mediated cytoskeleton remodeling and signaling events were found as key events to achieve an increase in HIV-1 entry in primary CD4+ T lymphocytes. We provide evidence that the noticed augmentation in virus entry is associated with an enrichment of CD4 and CXCR4 molecules in LFA-1 clusters, a process that increases the likelihood of virus-cell fusion upon stable HIV-1 adhesion.
Materials and Methods
Abs and reagents
The anti-LFA-1 mAb MEM25 (anti-CD11a) was provided by Dr. V. Horejsi (Institute of Molecular Genetics, Prague, Czech Republic). The anti-LFA-1 Ab mAb24, which is specific for an activation epitope, was obtained from Dr. N. Hogg (Cancer Research U.K. London Research Institute, London, U.K.). Recombinant human SDF-1 (rhSDF-1) was purchased from PeproTech Canada. The anti-CXCR4 12G5 and the hybridoma cell line that produces the anti-CD4 SIM.2 mAb were provided by the AIDS Repository Reagent Program (Germantown, MD) while polyclonal anti-CD4 Abs were produced in rabbits following immunization with recombinant soluble CD4 (kindly provided by Dr. R. Sweet, SmithKline Beecham, King of Prussia, PA). The hybridoma cell lines producing the anti-CD3 OKT3 and anti-ICAM-1 R6.5 mAbs were obtained from the American Type Culture Collection (ATCC). Purified anti-CD28 Ab (clone 9.3) was kindly provided by Dr. J. A. Ledbetter (Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ). The goat anti-mouse was purchased from Jackson ImmunoResearch Laboratories. The hybridomas that produce 183-H12-5C and 31-90-25, two Abs recognizing different epitopes of the HIV-1 major viral core protein p24, were supplied by the AIDS Repository Reagent Program and ATCC, respectively. The recombinant-PE-conjugated goat anti-mouse IgG was purchased from Jackson ImmunoResearch Laboratories. Alexa 488-, Alexa 546-, and Cy5- conjugated Abs were obtained from Molecular Probes. Jasplakinolide, calpeptin, U73122, piceatannol, and PP2 were obtained from Calbiochem and cytochalasin D from Sigma-Aldrich. The fusion inhibitor T-20 was provided by Roche Bioscience.
Cells
293T cells were provided by Dr. W. C. Greene (The J. Gladstone Institutes, San Francisco, CA) and were cultured in DMEM supplemented with 10% FCS. PBMCs from healthy donors were isolated by Ficoll-Hypaque gradient centrifugation and CD4+ T cells were purified from freshly isolated PBMCs by immunomagnetic negative selection as indicated by the manufacturer (Stem Cell Technologies). The purity of CD4+ T cells was determined by cytofluorometry analysis and was always >97%. Purified CD4+ T cells were cultured for 2 days in RPMI 1640 medium supplemented with 10% FCS in the presence of PHA (1 μg/ml) and rhIL-2 (50 U/ml).
Plasmids and virus production
pNL4-3 is a full length infectious molecular clone of HIV-1. This vector was provided by the AIDS Repository Reagent Program. The pCD1.8 is a eukaryotic expression vector containing the entire human ICAM-1 (a generous gift from Dr. T. Springer, The Center for Blood Research, Boston, MA). Viruses differing only by the absence or the presence of host-encoded ICAM-1 proteins on their surface were produced by the calcium phosphate coprecipitation method in 293T cells as described previously (1). Production of laboratory (NL4-3) and clinical (92HT599) X4-tropic isolates of HIV-1 was achieved by infecting PBMCs for 7 days. Virus preparations were normalized for virion content by using an in-house enzymatic assay specific for the major viral p24 protein as described previously (3). In this test, 183-H12-5C and 31-90-25 are used in combination to quantify p24 levels.
Virus capture assay
The presence of host-encoded ICAM-1 on the surface of HIV-1 particles either produced by 293T cells or PBMCs was investigated using magnetic beads (BioMag, Fc specific; Perspective Diagnostics) coated with an anti-ICAM-1 (R6.5) or an isotype-matched irrelevant Ab (IgG2a). Briefly, 12.5 μl of beads were incubated with virions (3 ng of p24) in a final volume of 1 ml of binding buffer (PBS supplemented with 0.5% BSA) and incubated for 1 h at 4°C on a rotating plate. Beads were then washed three times with binding buffer with a magnetic separation unit and resuspended in 200 μl of binding buffer. The amount of captured viruses was evaluated by the p24 assay.
Virus entry assay
PHA/IL-2-treated CD4+ T lymphocytes (5 x 105) were resuspended in 0.1 ml of culture medium containing isogenic HIV-1 particles either lacking (called NL4-3) or bearing host-derived ICAM-1 (called NL4-3/ICAM-1) (2.5 ng of p24 per 105 cells) or laboratory and clinical viruses produced in PBMCs (20 ng of p24 per 105 cells) and were incubated at 37°C for 1 or 2 h. To monitor the role played by the cytoskeleton and signaling proteins in the process of HIV-1 entry, cells were pretreated for 30 min at 37°C with various inhibitors before adding viruses. As a control, cells were pretreated with DMSO, the dissolving agent for the tested compounds. In some experiments, PHA activated CD4+ T cells were pretreated with an anti-LFA-1 (MEM25), an anti-CD4 Ab (i.e., SIM.2 at 20 μg/ml), rhSDF-1 (500 ng/ml) or T-20 (10 μg/ml) for 30 min at 37°C to block either ICAM-1/LFA-1 interaction, the CD4 primary receptor, the CXCR4 coreceptor or virus fusion. In other conditions, resting CD4+ T cells were incubated with viruses and rhSDF-1 (1 ng/ml), Mn2+ (2 mM) or OKT3/9.3/goat anti-mouse combination (1:1:10 μg/ml) to activate LFA-1. After an incubation of 2 h with viruses at 37°C, cells were washed and trypsinized for 5 min to remove uninternalized viruses. Next, cells were first washed once with RPMI 1640 supplemented with 10% FCS and then twice with PBS before lysis in 200 μl of ice-cold lysis buffer (20 mM HEPES (pH 7.4), 150 mM NaCl, 0.5% Triton X-100). The amount of viruses entering cells was estimated by the p24 assay. This virus entry assay has been shown to allow measurements of viruses entering target cells through both cytosolic delivery (i.e., productive infection) and endocytosis (i.e., abortive infection) (4).
FACS analysis
To monitor LFA-1 affinity change in PHA/IL-2-treated CD4+ T lymphocytes, cells were incubated at 37° C for 15 min with mAb24 or an isotype-matched irrelevant Ab in the presence or absence of Mn2+. Cells were next washed with PBS and then incubated with a secondary Ab conjugated to recombinant-PE for 30 min at 4° C. After two washes, cells were fixed in 2% paraformaldehyde and analyzed by cell sorting (Epics ELITE ESP; Coulter Electronics).
Confocal microscopy
To evaluate the role of the cytoskeleton and of some signaling proteins in LFA-1 distribution in PHA/IL-2-treated CD4+ T lymphocytes, cells were either left untreated or treated with various drugs for 30 min at 37° C. Cells were then washed with binding buffer before incubation with an anti-LFA-1 (MEM25) (5 μg/ml) for 30 min on ice followed by a FITC-conjugated secondary Ab. For colocalization of LFA-1 and CD4, cells were also incubated with a polyclonal anti-CD4 (5 μg/ml) followed by an Alexa 633-conjugated secondary Ab. Next, cells were washed three times with binding buffer, fixed in 2% paraformaldehyde for 20 min at 4°C and slides were mounted in 90% glycerol in PBS. To localize bound and internalized HIV-1 particles, cells (1 x 106 cells) were incubated for 90 min at 37°C with isogenic NL4-3 particles either lacking or bearing host ICAM-1 (100 ng of p24). Cells were fixed, permeabilized with 0.05% saponin, and incubated with rhodamine-conjugated phalloidin (Molecular Probes) and pooled human serum from HIV-1-positive patients for 45 min on ice to stain the actin cytoskeleton and HIV-1, respectively. Cells were next washed and incubated for 30 min with goat anti-human IgG secondary Ab conjugated to Alexa 488. Slides were mounted as described above. To colocalize LFA-1 and HIV-1, CD4 and HIV-1, or CXCR4 and HIV-1, infected cells were incubated with anti-LFA-1 (mouse), anti-CD4 (rabbit), or anti-CXCR4 (mouse) Abs for 30 min on ice followed by the appropriate secondary Ab (i.e., Cy5-, FITC- or Alexa 633-conjugated). Cells were washed three times with binding buffer, fixed in 2% paraformaldehyde for 20 min and permeabilized for 5 min at 37°C with 0.025% saponin in binding buffer. HIV-1 particles were stained as described above except for the secondary Ab which was conjugated with Alexa 546 instead of Alexa 488. LFA-1, CD4, CXCR4, and bound/internalized viruses were visualized by confocal laser scanning microscopy (Olympus Fluoview FV300) and digital images were processed with Adobe Photoshop. All the images were taken under similar experimental conditions (i.e., exposure time, magnification, and intensification) and the processing was also the same for all the images shown. Colocalization was analyzed with the Metamorph Offline software version 6.1 (Universal Imaging). The size of the slices used to acquire the confocal images was 1 μM.
Statistical analysis
Results presented are expressed as percentage of inhibition compared with untreated cells from three to five experiments. Statistical significance between groups was computed by Wilcoxon analysis. Calculations were made with Instat3 software. Values of p <0.05 were considered statistically significant.
Results
Entry of laboratory and clinical strains of HIV-1 produced in primary human cells within activated CD4+ T cells is promoted by cell surface LFA-1
We initially monitored the presence of host-derived ICAM-1 in progeny virus produced in human embryonic kidney 293T cells, expressing or not expressing ICAM-1, and in laboratory (NL4-3) and clinical (92HT599) X4-tropic variants of HIV-1 produced in PBMCs through the use of a virus capture assay. As expected, ICAM-1 is efficiently acquired by viruses produced in ICAM-1-expressing 293 T cells but not in ICAM-1-negative parental 293T cells (Fig. 1A). Moreover, both virus strains that were amplified in PBMCs were found to incorporate host-derived ICAM-1 (Fig. 1B). Next, to address whether the interaction between virus-anchored ICAM-1 and cell surface LFA-1 is contributing to the process of virus entry, a neutralizing anti-LFA-1 Ab (i.e., MEM25) was used to abolish this interaction. It should be noted that purified CD4+ T cells were used as targets to parallel in vivo situations. Because it has been shown that infection of quiescent peripheral CD4+ T cells by HIV-1 is nonproductive and results in incomplete, labile, reverse transcripts, those target cells were treated with the lectin PHA, a powerful mitogenic agent used to mimic the activation from the immunological synapse. This treatment induces the LFA-1 intermediate affinity state and its release from the cytoskeleton, two events known to be required for an optimal interaction with ICAM-1. Results depicted in Fig. 1C indicate that entry of virions carrying host-derived ICAM-1 into PHA-stimulated primary CD4+ T cells is higher compared with virions lacking this adhesion molecule. Furthermore, internalization of viruses produced in ICAM-1-expressing 293T cells is significantly diminished upon treatment with an Ab that blocks the ICAM-1/LFA-1 interaction (i.e., MEM25), in contrast to viruses produced in ICAM-1-negative 293T cells. Interestingly, entry of virions amplified in primary human cells was also reduced upon treatment with the blocking anti-LFA-1 Ab (Fig. 1D). We evaluated also the possible role played by CD4 and CXCR4 in entry of the studied virus preparations. Results demonstrate that entry of virions lacking ICAM-1 in activated CD4+ T cells are not affected by agents that block gp120/CD4 (i.e., SIM.2) and gp120/CXCR4 interactions (i.e., SDF-1) as well as by a fusion inhibitor (i.e., T-20) (Fig. 2). These data suggest that, at least under the tested experimental conditions (i.e., virus entry monitored 2 h following infection), virions lacking host ICAM-1 are mainly internalized by endocytosis and not through fusion in the studied target cells. In contrast, isogenic ICAM-1-bearing viruses were found to be sensitive to all blocking agents studied including MEM25. These results are in agreement with our previous work demonstrating that the presence of host ICAM-1 favors entry of HIV-1 particles through fusion, a process known to result in productive infection (4). Overall, our data indicate that the association between virus-anchored host ICAM-1 and cell surface LFA-1 plays a key role in the initial steps of HIV-1 life cycle in activated CD4+ T cells by increasing virus attachment and fusion as described previously (4, 26).
Cysteine protease calpain is important for HIV-1 internalization in CD4+ T cells
Cluster formation requires lateral mobility of LFA-1 molecules, a process that increases the frequency of interactions with dispersed ICAM-1. The cysteine protease calpain is crucial for promoting rapid LFA-1-mediated adhesion by allowing lateral movement of this integrin on the plasma membrane when ICAM-1 is present at a low density (33). It was reported that treatment of T cells with a calpain inhibitor reduces both PMA- and TCR-stimulated cell-to-cell adhesion (21). Furthermore, it was shown that calpain is essential in regulating LFA-1 affinity by cleavage of talin, a protein that links the integrin with the actin cytoskeleton and induces LFA-1 activation (12, 22). To assess the implication of calpain-mediated LFA-1 activation and lateral mobility in the observed enhancement of virus entry, target cells were treated with the calpain inhibitor calpeptin after PHA treatment. Calpeptin does not affect the distribution of LFA-1 in clusters on the surface of PHA-activated CD4+ T lymphocytes (Fig. 7A). Interestingly, entry of ICAM-1-bearing NL4-3 particles is reduced by 24% while internalization of isogenic viruses lacking ICAM-1 is not affected (Fig. 7B). Similarly, entry of NL4-3 and 92HT599 that were produced in PBMCs was diminished by 43 and 32%, respectively (Fig. 7C). This confirms that entry of ICAM-1-bearing virions in primary CD4+ T cells requires a lateral mobility of the integrin LFA-1.
HIV-1 colocalizes with LFA-1 clusters in activated CD4+ T lymphocytes
Our previous observations suggest that the initial steps in the HIV-1 life cycle when put in contact with activated CD4+ T lymphocytes require a dynamic clustering of LFA-1 to achieve an efficient virus entry process. These results imply that virions are most likely patched on the plasma membrane and colocalized with LFA-1 clusters. This possibility was first tested by analyzing the distribution of viruses either lacking or bearing host ICAM-1 on the studied target cells. Confocal microscopic analyses revealed that a barely detectable amount of virions lacking ICAM-1 is found associated with activated CD4+ T cells, whereas large quantities of ICAM-1-bearing viruses are concentrated in some specific areas on target cells (Fig. 8). Localization of LFA-1, CD4, and HIV-1 was then monitored in an attempt to detect possible colocalization events through the use of the Metamorph Imaging System. This is a powerful system for performing operations such as quantitative measurement of colocalization events. This tool allows measurements for the area, averages gray scale intensity, and integrates gray scale intensity of regions for which different fluorescent probes are overlapped. Results were thus generated using the most recent bioimaging technology. Colocalization images represent the superposition of pixels from two and/or three colors calculated by the Metamorph algorithms. As depicted Fig. 9, very few colocalization events between CD4 and HIV-1 (Fig. 9F, magenta color), CXCR4 and HIV-1 (panel L, white color) CXCR4, CD4 and HIV-1 (panel N, pink color) are seen upon infection of activated CD4+ T lymphocytes with ICAM-1-bearing virions. However, a high proportion of HIV-1 particles carrying host ICAM-1 colocalizes with LFA-1 following incubation with PHA-activated CD4+ T lymphocytes (Fig. 9E, yellow color). Moreover, a large fraction of CD4 molecules is present in LFA-1 clusters (around 40%) (Fig. 9D, white color). However, colocalization of HIV-1 particles with CD4 relies primarily on the colocalization of virions with LFA-1. Indeed, colocalization of LFA-1, CD4, and ICAM-1-bearing viruses (Fig. 9G, pink color) is similar to colocalization of CD4 and virions ((Fig. 9F, magenta color). Given that attachment of virions lacking host ICAM-1 is barely detectable, colocalization events cannot be detected (data not shown). It should be stated that the very low amounts of ICAM-1-negative viruses were not concentrated in LFA-1 clusters but were randomly distributed around surface of CD4+ T lymphocytes (data not shown). Altogether, these data indicate that a high amount of ICAM-1-bearing viruses is captured by LFA-1 under an activated state and only a small fraction of progeny virus is also attached to CD4 and CXCR4.
Discussion
Intermediate affinity, lateral mobility, and clustering of LFA-1 are thought to play a dominant role in cell-to-cell adhesion because these two processes facilitate the interaction between LFA-1 and ICAM-1 as well as the transport of molecules to the cell-to-cell contact area. The data shown in the present work indicate that LFA-1-mediated adhesion is also important for HIV-1 to attach to and gain entry within activated CD4+ T lymphocytes, a cell type considered as a major cellular reservoir in infected individuals. It can be proposed that molecular events involved in attachment of HIV-1 to LFA-1-expressing cells might be different from events required for the more traditional ICAM-1/LFA-1-mediated cell-to-cell adhesion. Indeed, when one considers the small diameter of HIV-1 (100 nm compared with at least 10 μm for most cells of the immune system), viruses carrying host ICAM-1 on their exterior surface might behave as a multivalent soluble ligand. We provide evidence that activation of LFA-1 which results in induction of intermediate binding affinity state, lateral mobility and clustering of the integrins is the mechanism through which PHA-activated CD4+ T cells more efficiently capture ICAM-1-bearing viruses compared with isogenic virions lacking host-derived ICAM-1. The physiological relevance of such findings is high considering that the adhesion molecule ICAM-1 is efficiently acquired by all tested clinical strains of HIV-1 bearing different tropisms (R5 and X4) that were produced in primary human cells (3, 34, 35, 36, 37). Data presented in this work were confirmed with a X4-tropic variant of HIV-1 (92HT599) expanded in PBMCs that was used to inoculate primary CD4+ T lymphocytes. Thus, the overall importance of the interaction between virus-anchored host ICAM-1 and cell surface LFA-1 has been established with a primary isolate of HIV-1, which provides clinical relevance to these observations. An indirect proof of the relative contribution of the ICAM-1/LFA-1 association when natural cellular reservoirs of HIV-1 are infected with primary viral isolates has been obtained in previous works. For example, it was demonstrated that infection of human lymphoid tissue cultured ex vivo with primary R5- and X4-tropic isolates of HIV-1 that were expanded in PBMCs was severely reduced by a treatment with a blocking anti-ICAM-1 Ab (3). More recently, a reduction in virus replication was seen when R5- and X4-tropic field isolates of HIV-1 amplified in PBMCs were used to infect mitogen-stimulated PBMCs that were treated this time with a blocking anti-LFA-1 Ab (3).
Previous studies have indicated that changes in binding affinity, lateral mobility, and clustering of the 2 integrins such as LFA-1 is influenced by a number of molecular events including polymerization/depolymerization of actin, generation of intracellular second messengers (calcium and diacylglycerol), activation of PKC, PI3K, and the small GTPases Rap1 and RapL (38, 39, 40, 41). The formation and lateral movement of LFA-1 clusters on the plasma membrane are active processes that require the release from cytoskeleton constraints. Conversely, actin polymerization is crucial for stabilizing clusters and maintaining cell-to-cell adhesion (12, 18, 21, 22, 23, 24, 25). Although the involvement of cytoskeleton remodeling in LFA-1 activation is well-established, we wanted to investigate its role in the early steps of the HIV-1 life cycle. We found that alteration of polymerization/depolymerization of actin cytoskeleton disrupts LFA-1 patches and reduces the ability of HIV-1 to enter inside activated CD4+ T lymphocytes. The importance of ICAM-1/LFA-1 interaction in the observed phenomenon is depicted by the observation that entry of isogenic virions lacking ICAM-1 is not modulated by agents that affect the actin cytoskeleton. An intact cytoskeleton is thus essential for maintaining the dynamic structure of LFA-1 clusters, which are subjected to a rapid turnover. Changing the equilibrium between free and cytoskeleton-linked LFA-1 impairs the ability of the integrin to efficiently bind to its natural ligand ICAM-1. LFA-1-mediated signaling also requires a dynamic cytoskeleton to cluster together to enhance T cell adhesion upon ligand binding (30).
It is well-established that engagement of LFA-1 by ICAM-1 triggers a signaling cascade leading to activation of c-Src kinases and ZAP70, tyrosine phosphorylation of PLC1, phospholipid hydrolysis, activation of different isoforms of PKC, and mobilization of intracellular calcium (42, 43, 44, 45, 46, 47). Those signaling events are important for the establishment of new linkages with the actin cytoskeleton that reinforce and maintain stable cellular adhesion through clustering. Our results demonstrate for the first time the contribution of ZAP70 but not c-Src kinases in entry of HIV-1 particles in primary human CD4+ T cells. Therefore, LFA-1-mediated activation of ZAP70 that is mediated upon HIV-1 attachment seems to be important for the entry process. Even though inhibition of ZAP70 has no effect on LFA-1 distribution, it can perturb some more downstream events. PLC1, a known effector of receptor tyrosine kinase signaling, is also recognized as a signal transducer that is acting downstream from ZAP70. This protein is essential for spatial organization of LFA-1 in clusters and the induction of intermediate and high affinity states. Inhibition of PLC1 leads to disruption of LFA-1 clusters and results in a reduced entry of HIV-1 particles harboring ICAM-1 in their envelope. By hydrolyzing phosphatidylinositol-4',5'-bisphosphate (PIP2), PLC1 generates two important second messengers, calcium and diacylglycerol, that are crucial for LFA-1 activation through Rap1 (32). In contrast, PIP2 is an important regulator of the actin cytoskeleton. Several lines of evidence indicate that accumulation of PIP2, following inactivation of PLC1, increases actin polymerization and cytoskeleton constraints of LFA-1 are enhanced through the activation of the actin-linker talin (48). The cleavage of talin by the protease calpain is known to lead to LFA-1 activation (20, 21). Calpain is a cysteine protease activated by calcium, a second messenger produced by the activation of PLC1. Inhibition of calpain has no effect on preformed LFA-1 clusters, but might considerably affect subsequent cluster formation and strengthen adhesion because its inhibition diminishes entry of ICAM-1-bearing virions. Together, our results indicate that clustering of LFA-1 is a key event to allow an efficient HIV-1 attachment to and entry in CD4-expressing T cells. Given that we provide evidence that most ICAM-1-bearing HIV-1 particles colocalize with LFA-1 clusters, it can be postulated that attachment of such virions to cell surface LFA-1 clusters might occur even before engagement of CD4 molecules. Binding of viruses to LFA-1 might in turn trigger outside-in signaling that might favor a firmer docking of the viral entity onto the target cell surface, a process that might facilitate the subsequent encounter with an appropriate number of CD4 and coreceptors.
The formation of an immunological synapse is always accompanied by clustering of LFA-1. It was recently reported that HIV-1 dissemination through a direct cell-to-cell transfer is efficiently achieved upon the creation of a virological synapse (49). This structure relies on rapid recruitment to the cellular interface of CD4, CXCR4, talin, and LFA-1 on the target cell, and of virus-produced Env and Gag proteins on the effector cell. We hypothesize here that the presence of ICAM-1 onto HIV-1 particles and of LFA-1 clusters on target cells can have a similar positive effect on virus-cell interactions as it does for cell-to-cell transfer of the virus. By using LFA-1 clusters to firmly attach to the surface of target cells, HIV-1 particles may not have to reorganize CD4 and CXCR4 distribution on the cellular membrane and fusion can thus be facilitated. Interestingly, data from confocal analyses aimed at localizing LFA-1 and CD4 molecules on PHA-activated CD4+ T cells revealed that CD4 is primarily located in LFA-1 clusters, independently of viral infection. Therefore, microdomains that contain LFA-1 clusters represent ideal environments on the cell surface membrane to gain entry into target cells. This hypothesis is confirmed by the observation that ICAM-1-bearing virions colocalize with LFA-1 clusters and also partly with CD4. Among viral particles colocalizing with the CD4 glycoprotein, a small proportion also colocalizes with the chemokine coreceptor CXCR4. Once strongly attached to LFA-1, HIV-1 may bind CD4 leading to a conformational change in the viral envelope permitting its engagement with the appropriate coreceptor. Because clustering of LFA-1 might lead to enrichment of CD4 and CXCR4 molecules in close vicinity, the process of fusion of viral and cellular membranes might be facilitated.
In summary, this study confirms the key role played by interactions between virus-associated host ICAM-1 and cell surface LFA-1 in HIV-1 biology. These results provide additional clues for the understanding of the very complex nature of the initial steps in the life cycle of HIV-1. A better knowledge of cellular factors relevant for the initiation of infection is crucial for the development of new pharmacological approaches targeting HIV-1 permissive cells. Considering that statins, a family of drugs commonly used for the treatment of hypercholesterolemia, selectively blocked LFA-1-mediated adhesion to ICAM-1 (50), it can be postulated that statins should be considered as additional compounds in the current arsenal that is used for the treatment of HIV-1-infected individuals. Interestingly, two members of the statins family (i.e., lovastatin and simvastatin) were recently shown to diminish the process of HIV-1 infection by impairing the association between virus-associated host ICAM-1 and cell surface LFA-1 (51). Statins were also found to reduce viral load and increase CD4+ T cells counts in chronically HIV-1-infected patients (52), thus providing additional credence to the importance of interactions between ICAM-1 and LFA-1 in the pathogenesis of HIV-1 infection.
Acknowledgments
We are grateful to S. Méthot for editorial assistance and M. Imbeault for his assistance with the Metamorph software. We thank Dr. M. Dufour for performing flow cytometric analyses.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was financially supported by an operating grant (to M.J.T.) from the Canadian Institute of Health Research (CIHR) HIV/AIDS Research Program (Grant No. HOP-14438). M.R.T. holds a Doctoral Award from the CIHR. M.J.T. is the recipient of the Canada Research Chair in Human Immuno-Retrovirology (senior level).
2 Address correspondence and reprint requests to Dr. Michel J. Tremblay, Laboratory of Human Immuno-Retrovirology, Research Center in Infectious Diseases, RC709, CHUL Research Center, 2705 Laurier Boulevard, Quebec, Canada, G1V 4G2. E-mail address: michel.j.tremblay@crchul.ulaval.ca
3 Abbreviations used in this paper: HIV-1, HIV type-1; rh, recombinant human; PKC, protein kinase C; PLC, phospholipase C; PIP2, phosphatidylinositol-4',5'-biphosphate; 3D, three dimensional.
Received for publication October 25, 2004. Accepted for publication May 10, 2005.
References
Fortin, J. F., R. Cantin, G. Lamontagne, M. Tremblay. 1997. Host-derived ICAM-1 glycoproteins incorporated on human immunodeficiency virus type 1 are biologically active and enhance viral infectivity. J. Virol. 71: 3588-3596.
Fortin, J. F., R. Cantin, M. J. Tremblay. 1998. T cells expressing activated LFA-1 are more susceptible to infection with human immunodeficiency virus type 1 particles bearing host-encoded ICAM-1. J. Virol. 72: 2105-2112.
Bounou, S., J. E. Leclerc, M. J. Tremblay. 2002. Presence of host ICAM-1 in laboratory and clinical strains of human immunodeficiency virus type 1 increases virus infectivity and CD4+-T-cell depletion in human lymphoid tissue, a major site of replication in vivo. J. Virol. 76: 1004-1014.
Tardif, M. R., M. J. Tremblay. 2003. Presence of host ICAM-1 in human immunodeficiency virus type 1 virions increases productive infection of CD4+ T lymphocytes by favoring cytosolic delivery of viral material. J. Virol. 77: 12299-12309.
Fortin, J. F., R. Cantin, M. G. Bergeron, M. J. Tremblay. 2000. Interaction between virion-bound host intercellular adhesion molecule-1 and the high-affinity state of lymphocyte function-associated antigen-1 on target cells renders R5 and X4 isolates of human immunodeficiency virus type 1 more refractory to neutralization. Virology 268: 493-503.(Fortin, J. F., R. Cantin,)
Besides interactions between the viral envelope glycoproteins with cell surface receptors, interactions between cell-derived molecules incorporated onto virions and their ligand could also modulate HIV type-1 (HIV-1) entry inside CD4+ T lymphocytes. Although incorporation of host ICAM-1 within HIV-1 increases both virus attachment and fusion, the precise mechanism through which this phenomenon is occurring is still unclear. We demonstrate in this study that activation of primary human CD4+ T lymphocytes increases LFA-1 affinity and avidity states, two events promoting the early events of the HIV-1 replication cycle through interactions between virus-embedded host ICAM-1 and LFA-1 clusters. Confocal analyses suggest that HIV-1 is concentrated in microdomains rich in LFA-1 clusters that also contain CD4 and CXCR4 molecules. Experiments performed with specific inhibitors revealed that entry of HIV-1 in activated CD4+ T cells is regulated by LFA-1-dependent ZAP70, phospholipase C1, and calpain enzymatic activities. By using laboratory and clinical strains of HIV-1 produced in primary human cells, we demonstrate the importance of the LFA-1 activation state and cluster formation in the initial step of the virus life cycle. Overall, these data provide new insights into the complex molecular events involved in HIV-1 binding and entry.
Introduction
Entry of HIV type-1 (HIV-1)3 into susceptible target cells is a complex and dynamic process involving several microevents. Besides the normal association between the external envelope gp120 and the appropriate cell surface receptors, it has been proposed that interactions between host-derived molecules present on the virion and their cognate receptors on the target cell surface can also contribute to the formation of a fusion-competent environment. It is well-established that HIV-1 incorporates a vast array of host cell membrane molecules during budding, including the ICAM-1 (1, 2, 3). Interestingly, engagement of ICAM-1 with its natural counterligand LFA-1 increases viral infectivity by several folds. In CD4+ T lymphocytes that express functional LFA-1, this phenomenon is associated with an enhancement of both HIV-1 attachment and virus-cell fusion. It has been shown that the presence of host ICAM-1 onto virions favors fusion rather than endocytosis, the former being associated with a more productive HIV-1 infection process (4). Moreover, CD4 and CXCR4 molecules are essential to achieve an efficient infection with viruses lacking host ICAM-1 as well as with isogenic ICAM-1-bearing virions (4). The presence of virus-anchored host ICAM-1 confers also resistance to neutralization by Abs and to the fusion inhibitor T-20 (5, 6). This reduced susceptibility to the antiviral activity of T-20 illustrates the critical role played by ICAM-1/LFA-1 interaction in the initial steps of the virus replication cycle. The exact molecular process by which ICAM-1-bearing virions are more efficiently entering LFA-1-expressing cells remains to be more clearly defined.
LFA-1-mediated adhesion is playing a key role in immune surveillance and the mounting of a potent immune response. More specifically, LFA-1 is involved in the arrest of rolling lymphocytes along blood vessels facilitating extravasation and migration of the activated T cells to infection sites. The 2 integrin LFA-1 participates to the formation of the immunological synapse between T cells and APCs. These events require a rapid modulation of adhesion/deadhesion and the adhesion needs to be stable for hours to sustain the immunological synapse. Hence, activation of LFA-1 must be tightly regulated. Two mechanisms have been proposed for the regulation of LFA-1-mediated adhesion that include modulation of affinity and avidity (7, 8, 9). Affinity regulation refers to changes in individual L2 heterodimer that are linked to modifications in LFA-1 conformation. Recently, it has been discovered that LFA-1 can take three binding affinities, i.e., low, intermediate and high (10). In the low affinity conformation, the integrin adopts a bent state for which the head domain points toward cell membrane and cannot interact with its ligand. The acquisition of the intermediate and high affinity conformational states involves the unbending of both subunits to form the extended state. The intermediate binding affinity is induced by inside-out signaling triggered by an immunological synapse, chemokine, or phorbol ester treatment (11), whereas the induction of LFA-1 under a high affinity state is achieved upon ligand binding and involves separation of the and subunits at their cytoplasmic, transmembrane, and leg domains (12, 13). The high affinity state can be also generated in vitro by exposing cells to divalent cations such as manganese or magnesium (8). The avidity state of LFA-1 refers to distribution of the integrins in plasma membrane. Reorganization of LFA-1 into micro- and macroclusters on the cell surface follows ligand binding and requires the release of the integrin tail to the actin cytoskeleton. Engagement of ICAM-1 induces outside-in signaling that triggers lateral diffusion of heterodimers in plasma membrane, a process regulating the adhesion strengthening (14, 15, 16, 17, 18). In naive T cells, LFA-1 is maintained under a low affinity/avidity state (19, 20). Upon T cell activation, the inside-out signaling generates the intermediate affinity state and activates the cysteine protease calpain that releases LFA-1 from its cytoskeleton constraint and favors lateral diffusion of LFA-1 (12, 18, 21, 22). These phenomena trigger homotypic adhesion which induces in turn the formation of micro- and macroclusters of LFA-1 at cell-cell contact interfaces (11). While affinity and avidity are differently controlled, the two mechanisms act in a concerted manner to achieve and maintain stable cell-to-cell adhesion (23, 24, 25). Furthermore, ligand-mediated signaling events have been shown to be important for the creation of new linkages with the actin cytoskeleton that reinforce and maintain stable adhesion (18).
Considering that LFA-1 is tightly regulated and its activation state is crucial to permit a firm and stable ICAM-1 binding, we addressed whether the LFA-1 activation state, intermediate lateral mobility and clustering can lead to a more efficient virus attachment and entry. Our results show that capture and internalization of ICAM-1-bearing virions are more efficient in cells expressing activated LFA-1 molecules. Moreover, LFA-1-mediated cytoskeleton remodeling and signaling events were found as key events to achieve an increase in HIV-1 entry in primary CD4+ T lymphocytes. We provide evidence that the noticed augmentation in virus entry is associated with an enrichment of CD4 and CXCR4 molecules in LFA-1 clusters, a process that increases the likelihood of virus-cell fusion upon stable HIV-1 adhesion.
Materials and Methods
Abs and reagents
The anti-LFA-1 mAb MEM25 (anti-CD11a) was provided by Dr. V. Horejsi (Institute of Molecular Genetics, Prague, Czech Republic). The anti-LFA-1 Ab mAb24, which is specific for an activation epitope, was obtained from Dr. N. Hogg (Cancer Research U.K. London Research Institute, London, U.K.). Recombinant human SDF-1 (rhSDF-1) was purchased from PeproTech Canada. The anti-CXCR4 12G5 and the hybridoma cell line that produces the anti-CD4 SIM.2 mAb were provided by the AIDS Repository Reagent Program (Germantown, MD) while polyclonal anti-CD4 Abs were produced in rabbits following immunization with recombinant soluble CD4 (kindly provided by Dr. R. Sweet, SmithKline Beecham, King of Prussia, PA). The hybridoma cell lines producing the anti-CD3 OKT3 and anti-ICAM-1 R6.5 mAbs were obtained from the American Type Culture Collection (ATCC). Purified anti-CD28 Ab (clone 9.3) was kindly provided by Dr. J. A. Ledbetter (Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ). The goat anti-mouse was purchased from Jackson ImmunoResearch Laboratories. The hybridomas that produce 183-H12-5C and 31-90-25, two Abs recognizing different epitopes of the HIV-1 major viral core protein p24, were supplied by the AIDS Repository Reagent Program and ATCC, respectively. The recombinant-PE-conjugated goat anti-mouse IgG was purchased from Jackson ImmunoResearch Laboratories. Alexa 488-, Alexa 546-, and Cy5- conjugated Abs were obtained from Molecular Probes. Jasplakinolide, calpeptin, U73122, piceatannol, and PP2 were obtained from Calbiochem and cytochalasin D from Sigma-Aldrich. The fusion inhibitor T-20 was provided by Roche Bioscience.
Cells
293T cells were provided by Dr. W. C. Greene (The J. Gladstone Institutes, San Francisco, CA) and were cultured in DMEM supplemented with 10% FCS. PBMCs from healthy donors were isolated by Ficoll-Hypaque gradient centrifugation and CD4+ T cells were purified from freshly isolated PBMCs by immunomagnetic negative selection as indicated by the manufacturer (Stem Cell Technologies). The purity of CD4+ T cells was determined by cytofluorometry analysis and was always >97%. Purified CD4+ T cells were cultured for 2 days in RPMI 1640 medium supplemented with 10% FCS in the presence of PHA (1 μg/ml) and rhIL-2 (50 U/ml).
Plasmids and virus production
pNL4-3 is a full length infectious molecular clone of HIV-1. This vector was provided by the AIDS Repository Reagent Program. The pCD1.8 is a eukaryotic expression vector containing the entire human ICAM-1 (a generous gift from Dr. T. Springer, The Center for Blood Research, Boston, MA). Viruses differing only by the absence or the presence of host-encoded ICAM-1 proteins on their surface were produced by the calcium phosphate coprecipitation method in 293T cells as described previously (1). Production of laboratory (NL4-3) and clinical (92HT599) X4-tropic isolates of HIV-1 was achieved by infecting PBMCs for 7 days. Virus preparations were normalized for virion content by using an in-house enzymatic assay specific for the major viral p24 protein as described previously (3). In this test, 183-H12-5C and 31-90-25 are used in combination to quantify p24 levels.
Virus capture assay
The presence of host-encoded ICAM-1 on the surface of HIV-1 particles either produced by 293T cells or PBMCs was investigated using magnetic beads (BioMag, Fc specific; Perspective Diagnostics) coated with an anti-ICAM-1 (R6.5) or an isotype-matched irrelevant Ab (IgG2a). Briefly, 12.5 μl of beads were incubated with virions (3 ng of p24) in a final volume of 1 ml of binding buffer (PBS supplemented with 0.5% BSA) and incubated for 1 h at 4°C on a rotating plate. Beads were then washed three times with binding buffer with a magnetic separation unit and resuspended in 200 μl of binding buffer. The amount of captured viruses was evaluated by the p24 assay.
Virus entry assay
PHA/IL-2-treated CD4+ T lymphocytes (5 x 105) were resuspended in 0.1 ml of culture medium containing isogenic HIV-1 particles either lacking (called NL4-3) or bearing host-derived ICAM-1 (called NL4-3/ICAM-1) (2.5 ng of p24 per 105 cells) or laboratory and clinical viruses produced in PBMCs (20 ng of p24 per 105 cells) and were incubated at 37°C for 1 or 2 h. To monitor the role played by the cytoskeleton and signaling proteins in the process of HIV-1 entry, cells were pretreated for 30 min at 37°C with various inhibitors before adding viruses. As a control, cells were pretreated with DMSO, the dissolving agent for the tested compounds. In some experiments, PHA activated CD4+ T cells were pretreated with an anti-LFA-1 (MEM25), an anti-CD4 Ab (i.e., SIM.2 at 20 μg/ml), rhSDF-1 (500 ng/ml) or T-20 (10 μg/ml) for 30 min at 37°C to block either ICAM-1/LFA-1 interaction, the CD4 primary receptor, the CXCR4 coreceptor or virus fusion. In other conditions, resting CD4+ T cells were incubated with viruses and rhSDF-1 (1 ng/ml), Mn2+ (2 mM) or OKT3/9.3/goat anti-mouse combination (1:1:10 μg/ml) to activate LFA-1. After an incubation of 2 h with viruses at 37°C, cells were washed and trypsinized for 5 min to remove uninternalized viruses. Next, cells were first washed once with RPMI 1640 supplemented with 10% FCS and then twice with PBS before lysis in 200 μl of ice-cold lysis buffer (20 mM HEPES (pH 7.4), 150 mM NaCl, 0.5% Triton X-100). The amount of viruses entering cells was estimated by the p24 assay. This virus entry assay has been shown to allow measurements of viruses entering target cells through both cytosolic delivery (i.e., productive infection) and endocytosis (i.e., abortive infection) (4).
FACS analysis
To monitor LFA-1 affinity change in PHA/IL-2-treated CD4+ T lymphocytes, cells were incubated at 37° C for 15 min with mAb24 or an isotype-matched irrelevant Ab in the presence or absence of Mn2+. Cells were next washed with PBS and then incubated with a secondary Ab conjugated to recombinant-PE for 30 min at 4° C. After two washes, cells were fixed in 2% paraformaldehyde and analyzed by cell sorting (Epics ELITE ESP; Coulter Electronics).
Confocal microscopy
To evaluate the role of the cytoskeleton and of some signaling proteins in LFA-1 distribution in PHA/IL-2-treated CD4+ T lymphocytes, cells were either left untreated or treated with various drugs for 30 min at 37° C. Cells were then washed with binding buffer before incubation with an anti-LFA-1 (MEM25) (5 μg/ml) for 30 min on ice followed by a FITC-conjugated secondary Ab. For colocalization of LFA-1 and CD4, cells were also incubated with a polyclonal anti-CD4 (5 μg/ml) followed by an Alexa 633-conjugated secondary Ab. Next, cells were washed three times with binding buffer, fixed in 2% paraformaldehyde for 20 min at 4°C and slides were mounted in 90% glycerol in PBS. To localize bound and internalized HIV-1 particles, cells (1 x 106 cells) were incubated for 90 min at 37°C with isogenic NL4-3 particles either lacking or bearing host ICAM-1 (100 ng of p24). Cells were fixed, permeabilized with 0.05% saponin, and incubated with rhodamine-conjugated phalloidin (Molecular Probes) and pooled human serum from HIV-1-positive patients for 45 min on ice to stain the actin cytoskeleton and HIV-1, respectively. Cells were next washed and incubated for 30 min with goat anti-human IgG secondary Ab conjugated to Alexa 488. Slides were mounted as described above. To colocalize LFA-1 and HIV-1, CD4 and HIV-1, or CXCR4 and HIV-1, infected cells were incubated with anti-LFA-1 (mouse), anti-CD4 (rabbit), or anti-CXCR4 (mouse) Abs for 30 min on ice followed by the appropriate secondary Ab (i.e., Cy5-, FITC- or Alexa 633-conjugated). Cells were washed three times with binding buffer, fixed in 2% paraformaldehyde for 20 min and permeabilized for 5 min at 37°C with 0.025% saponin in binding buffer. HIV-1 particles were stained as described above except for the secondary Ab which was conjugated with Alexa 546 instead of Alexa 488. LFA-1, CD4, CXCR4, and bound/internalized viruses were visualized by confocal laser scanning microscopy (Olympus Fluoview FV300) and digital images were processed with Adobe Photoshop. All the images were taken under similar experimental conditions (i.e., exposure time, magnification, and intensification) and the processing was also the same for all the images shown. Colocalization was analyzed with the Metamorph Offline software version 6.1 (Universal Imaging). The size of the slices used to acquire the confocal images was 1 μM.
Statistical analysis
Results presented are expressed as percentage of inhibition compared with untreated cells from three to five experiments. Statistical significance between groups was computed by Wilcoxon analysis. Calculations were made with Instat3 software. Values of p <0.05 were considered statistically significant.
Results
Entry of laboratory and clinical strains of HIV-1 produced in primary human cells within activated CD4+ T cells is promoted by cell surface LFA-1
We initially monitored the presence of host-derived ICAM-1 in progeny virus produced in human embryonic kidney 293T cells, expressing or not expressing ICAM-1, and in laboratory (NL4-3) and clinical (92HT599) X4-tropic variants of HIV-1 produced in PBMCs through the use of a virus capture assay. As expected, ICAM-1 is efficiently acquired by viruses produced in ICAM-1-expressing 293 T cells but not in ICAM-1-negative parental 293T cells (Fig. 1A). Moreover, both virus strains that were amplified in PBMCs were found to incorporate host-derived ICAM-1 (Fig. 1B). Next, to address whether the interaction between virus-anchored ICAM-1 and cell surface LFA-1 is contributing to the process of virus entry, a neutralizing anti-LFA-1 Ab (i.e., MEM25) was used to abolish this interaction. It should be noted that purified CD4+ T cells were used as targets to parallel in vivo situations. Because it has been shown that infection of quiescent peripheral CD4+ T cells by HIV-1 is nonproductive and results in incomplete, labile, reverse transcripts, those target cells were treated with the lectin PHA, a powerful mitogenic agent used to mimic the activation from the immunological synapse. This treatment induces the LFA-1 intermediate affinity state and its release from the cytoskeleton, two events known to be required for an optimal interaction with ICAM-1. Results depicted in Fig. 1C indicate that entry of virions carrying host-derived ICAM-1 into PHA-stimulated primary CD4+ T cells is higher compared with virions lacking this adhesion molecule. Furthermore, internalization of viruses produced in ICAM-1-expressing 293T cells is significantly diminished upon treatment with an Ab that blocks the ICAM-1/LFA-1 interaction (i.e., MEM25), in contrast to viruses produced in ICAM-1-negative 293T cells. Interestingly, entry of virions amplified in primary human cells was also reduced upon treatment with the blocking anti-LFA-1 Ab (Fig. 1D). We evaluated also the possible role played by CD4 and CXCR4 in entry of the studied virus preparations. Results demonstrate that entry of virions lacking ICAM-1 in activated CD4+ T cells are not affected by agents that block gp120/CD4 (i.e., SIM.2) and gp120/CXCR4 interactions (i.e., SDF-1) as well as by a fusion inhibitor (i.e., T-20) (Fig. 2). These data suggest that, at least under the tested experimental conditions (i.e., virus entry monitored 2 h following infection), virions lacking host ICAM-1 are mainly internalized by endocytosis and not through fusion in the studied target cells. In contrast, isogenic ICAM-1-bearing viruses were found to be sensitive to all blocking agents studied including MEM25. These results are in agreement with our previous work demonstrating that the presence of host ICAM-1 favors entry of HIV-1 particles through fusion, a process known to result in productive infection (4). Overall, our data indicate that the association between virus-anchored host ICAM-1 and cell surface LFA-1 plays a key role in the initial steps of HIV-1 life cycle in activated CD4+ T cells by increasing virus attachment and fusion as described previously (4, 26).
Cysteine protease calpain is important for HIV-1 internalization in CD4+ T cells
Cluster formation requires lateral mobility of LFA-1 molecules, a process that increases the frequency of interactions with dispersed ICAM-1. The cysteine protease calpain is crucial for promoting rapid LFA-1-mediated adhesion by allowing lateral movement of this integrin on the plasma membrane when ICAM-1 is present at a low density (33). It was reported that treatment of T cells with a calpain inhibitor reduces both PMA- and TCR-stimulated cell-to-cell adhesion (21). Furthermore, it was shown that calpain is essential in regulating LFA-1 affinity by cleavage of talin, a protein that links the integrin with the actin cytoskeleton and induces LFA-1 activation (12, 22). To assess the implication of calpain-mediated LFA-1 activation and lateral mobility in the observed enhancement of virus entry, target cells were treated with the calpain inhibitor calpeptin after PHA treatment. Calpeptin does not affect the distribution of LFA-1 in clusters on the surface of PHA-activated CD4+ T lymphocytes (Fig. 7A). Interestingly, entry of ICAM-1-bearing NL4-3 particles is reduced by 24% while internalization of isogenic viruses lacking ICAM-1 is not affected (Fig. 7B). Similarly, entry of NL4-3 and 92HT599 that were produced in PBMCs was diminished by 43 and 32%, respectively (Fig. 7C). This confirms that entry of ICAM-1-bearing virions in primary CD4+ T cells requires a lateral mobility of the integrin LFA-1.
HIV-1 colocalizes with LFA-1 clusters in activated CD4+ T lymphocytes
Our previous observations suggest that the initial steps in the HIV-1 life cycle when put in contact with activated CD4+ T lymphocytes require a dynamic clustering of LFA-1 to achieve an efficient virus entry process. These results imply that virions are most likely patched on the plasma membrane and colocalized with LFA-1 clusters. This possibility was first tested by analyzing the distribution of viruses either lacking or bearing host ICAM-1 on the studied target cells. Confocal microscopic analyses revealed that a barely detectable amount of virions lacking ICAM-1 is found associated with activated CD4+ T cells, whereas large quantities of ICAM-1-bearing viruses are concentrated in some specific areas on target cells (Fig. 8). Localization of LFA-1, CD4, and HIV-1 was then monitored in an attempt to detect possible colocalization events through the use of the Metamorph Imaging System. This is a powerful system for performing operations such as quantitative measurement of colocalization events. This tool allows measurements for the area, averages gray scale intensity, and integrates gray scale intensity of regions for which different fluorescent probes are overlapped. Results were thus generated using the most recent bioimaging technology. Colocalization images represent the superposition of pixels from two and/or three colors calculated by the Metamorph algorithms. As depicted Fig. 9, very few colocalization events between CD4 and HIV-1 (Fig. 9F, magenta color), CXCR4 and HIV-1 (panel L, white color) CXCR4, CD4 and HIV-1 (panel N, pink color) are seen upon infection of activated CD4+ T lymphocytes with ICAM-1-bearing virions. However, a high proportion of HIV-1 particles carrying host ICAM-1 colocalizes with LFA-1 following incubation with PHA-activated CD4+ T lymphocytes (Fig. 9E, yellow color). Moreover, a large fraction of CD4 molecules is present in LFA-1 clusters (around 40%) (Fig. 9D, white color). However, colocalization of HIV-1 particles with CD4 relies primarily on the colocalization of virions with LFA-1. Indeed, colocalization of LFA-1, CD4, and ICAM-1-bearing viruses (Fig. 9G, pink color) is similar to colocalization of CD4 and virions ((Fig. 9F, magenta color). Given that attachment of virions lacking host ICAM-1 is barely detectable, colocalization events cannot be detected (data not shown). It should be stated that the very low amounts of ICAM-1-negative viruses were not concentrated in LFA-1 clusters but were randomly distributed around surface of CD4+ T lymphocytes (data not shown). Altogether, these data indicate that a high amount of ICAM-1-bearing viruses is captured by LFA-1 under an activated state and only a small fraction of progeny virus is also attached to CD4 and CXCR4.
Discussion
Intermediate affinity, lateral mobility, and clustering of LFA-1 are thought to play a dominant role in cell-to-cell adhesion because these two processes facilitate the interaction between LFA-1 and ICAM-1 as well as the transport of molecules to the cell-to-cell contact area. The data shown in the present work indicate that LFA-1-mediated adhesion is also important for HIV-1 to attach to and gain entry within activated CD4+ T lymphocytes, a cell type considered as a major cellular reservoir in infected individuals. It can be proposed that molecular events involved in attachment of HIV-1 to LFA-1-expressing cells might be different from events required for the more traditional ICAM-1/LFA-1-mediated cell-to-cell adhesion. Indeed, when one considers the small diameter of HIV-1 (100 nm compared with at least 10 μm for most cells of the immune system), viruses carrying host ICAM-1 on their exterior surface might behave as a multivalent soluble ligand. We provide evidence that activation of LFA-1 which results in induction of intermediate binding affinity state, lateral mobility and clustering of the integrins is the mechanism through which PHA-activated CD4+ T cells more efficiently capture ICAM-1-bearing viruses compared with isogenic virions lacking host-derived ICAM-1. The physiological relevance of such findings is high considering that the adhesion molecule ICAM-1 is efficiently acquired by all tested clinical strains of HIV-1 bearing different tropisms (R5 and X4) that were produced in primary human cells (3, 34, 35, 36, 37). Data presented in this work were confirmed with a X4-tropic variant of HIV-1 (92HT599) expanded in PBMCs that was used to inoculate primary CD4+ T lymphocytes. Thus, the overall importance of the interaction between virus-anchored host ICAM-1 and cell surface LFA-1 has been established with a primary isolate of HIV-1, which provides clinical relevance to these observations. An indirect proof of the relative contribution of the ICAM-1/LFA-1 association when natural cellular reservoirs of HIV-1 are infected with primary viral isolates has been obtained in previous works. For example, it was demonstrated that infection of human lymphoid tissue cultured ex vivo with primary R5- and X4-tropic isolates of HIV-1 that were expanded in PBMCs was severely reduced by a treatment with a blocking anti-ICAM-1 Ab (3). More recently, a reduction in virus replication was seen when R5- and X4-tropic field isolates of HIV-1 amplified in PBMCs were used to infect mitogen-stimulated PBMCs that were treated this time with a blocking anti-LFA-1 Ab (3).
Previous studies have indicated that changes in binding affinity, lateral mobility, and clustering of the 2 integrins such as LFA-1 is influenced by a number of molecular events including polymerization/depolymerization of actin, generation of intracellular second messengers (calcium and diacylglycerol), activation of PKC, PI3K, and the small GTPases Rap1 and RapL (38, 39, 40, 41). The formation and lateral movement of LFA-1 clusters on the plasma membrane are active processes that require the release from cytoskeleton constraints. Conversely, actin polymerization is crucial for stabilizing clusters and maintaining cell-to-cell adhesion (12, 18, 21, 22, 23, 24, 25). Although the involvement of cytoskeleton remodeling in LFA-1 activation is well-established, we wanted to investigate its role in the early steps of the HIV-1 life cycle. We found that alteration of polymerization/depolymerization of actin cytoskeleton disrupts LFA-1 patches and reduces the ability of HIV-1 to enter inside activated CD4+ T lymphocytes. The importance of ICAM-1/LFA-1 interaction in the observed phenomenon is depicted by the observation that entry of isogenic virions lacking ICAM-1 is not modulated by agents that affect the actin cytoskeleton. An intact cytoskeleton is thus essential for maintaining the dynamic structure of LFA-1 clusters, which are subjected to a rapid turnover. Changing the equilibrium between free and cytoskeleton-linked LFA-1 impairs the ability of the integrin to efficiently bind to its natural ligand ICAM-1. LFA-1-mediated signaling also requires a dynamic cytoskeleton to cluster together to enhance T cell adhesion upon ligand binding (30).
It is well-established that engagement of LFA-1 by ICAM-1 triggers a signaling cascade leading to activation of c-Src kinases and ZAP70, tyrosine phosphorylation of PLC1, phospholipid hydrolysis, activation of different isoforms of PKC, and mobilization of intracellular calcium (42, 43, 44, 45, 46, 47). Those signaling events are important for the establishment of new linkages with the actin cytoskeleton that reinforce and maintain stable cellular adhesion through clustering. Our results demonstrate for the first time the contribution of ZAP70 but not c-Src kinases in entry of HIV-1 particles in primary human CD4+ T cells. Therefore, LFA-1-mediated activation of ZAP70 that is mediated upon HIV-1 attachment seems to be important for the entry process. Even though inhibition of ZAP70 has no effect on LFA-1 distribution, it can perturb some more downstream events. PLC1, a known effector of receptor tyrosine kinase signaling, is also recognized as a signal transducer that is acting downstream from ZAP70. This protein is essential for spatial organization of LFA-1 in clusters and the induction of intermediate and high affinity states. Inhibition of PLC1 leads to disruption of LFA-1 clusters and results in a reduced entry of HIV-1 particles harboring ICAM-1 in their envelope. By hydrolyzing phosphatidylinositol-4',5'-bisphosphate (PIP2), PLC1 generates two important second messengers, calcium and diacylglycerol, that are crucial for LFA-1 activation through Rap1 (32). In contrast, PIP2 is an important regulator of the actin cytoskeleton. Several lines of evidence indicate that accumulation of PIP2, following inactivation of PLC1, increases actin polymerization and cytoskeleton constraints of LFA-1 are enhanced through the activation of the actin-linker talin (48). The cleavage of talin by the protease calpain is known to lead to LFA-1 activation (20, 21). Calpain is a cysteine protease activated by calcium, a second messenger produced by the activation of PLC1. Inhibition of calpain has no effect on preformed LFA-1 clusters, but might considerably affect subsequent cluster formation and strengthen adhesion because its inhibition diminishes entry of ICAM-1-bearing virions. Together, our results indicate that clustering of LFA-1 is a key event to allow an efficient HIV-1 attachment to and entry in CD4-expressing T cells. Given that we provide evidence that most ICAM-1-bearing HIV-1 particles colocalize with LFA-1 clusters, it can be postulated that attachment of such virions to cell surface LFA-1 clusters might occur even before engagement of CD4 molecules. Binding of viruses to LFA-1 might in turn trigger outside-in signaling that might favor a firmer docking of the viral entity onto the target cell surface, a process that might facilitate the subsequent encounter with an appropriate number of CD4 and coreceptors.
The formation of an immunological synapse is always accompanied by clustering of LFA-1. It was recently reported that HIV-1 dissemination through a direct cell-to-cell transfer is efficiently achieved upon the creation of a virological synapse (49). This structure relies on rapid recruitment to the cellular interface of CD4, CXCR4, talin, and LFA-1 on the target cell, and of virus-produced Env and Gag proteins on the effector cell. We hypothesize here that the presence of ICAM-1 onto HIV-1 particles and of LFA-1 clusters on target cells can have a similar positive effect on virus-cell interactions as it does for cell-to-cell transfer of the virus. By using LFA-1 clusters to firmly attach to the surface of target cells, HIV-1 particles may not have to reorganize CD4 and CXCR4 distribution on the cellular membrane and fusion can thus be facilitated. Interestingly, data from confocal analyses aimed at localizing LFA-1 and CD4 molecules on PHA-activated CD4+ T cells revealed that CD4 is primarily located in LFA-1 clusters, independently of viral infection. Therefore, microdomains that contain LFA-1 clusters represent ideal environments on the cell surface membrane to gain entry into target cells. This hypothesis is confirmed by the observation that ICAM-1-bearing virions colocalize with LFA-1 clusters and also partly with CD4. Among viral particles colocalizing with the CD4 glycoprotein, a small proportion also colocalizes with the chemokine coreceptor CXCR4. Once strongly attached to LFA-1, HIV-1 may bind CD4 leading to a conformational change in the viral envelope permitting its engagement with the appropriate coreceptor. Because clustering of LFA-1 might lead to enrichment of CD4 and CXCR4 molecules in close vicinity, the process of fusion of viral and cellular membranes might be facilitated.
In summary, this study confirms the key role played by interactions between virus-associated host ICAM-1 and cell surface LFA-1 in HIV-1 biology. These results provide additional clues for the understanding of the very complex nature of the initial steps in the life cycle of HIV-1. A better knowledge of cellular factors relevant for the initiation of infection is crucial for the development of new pharmacological approaches targeting HIV-1 permissive cells. Considering that statins, a family of drugs commonly used for the treatment of hypercholesterolemia, selectively blocked LFA-1-mediated adhesion to ICAM-1 (50), it can be postulated that statins should be considered as additional compounds in the current arsenal that is used for the treatment of HIV-1-infected individuals. Interestingly, two members of the statins family (i.e., lovastatin and simvastatin) were recently shown to diminish the process of HIV-1 infection by impairing the association between virus-associated host ICAM-1 and cell surface LFA-1 (51). Statins were also found to reduce viral load and increase CD4+ T cells counts in chronically HIV-1-infected patients (52), thus providing additional credence to the importance of interactions between ICAM-1 and LFA-1 in the pathogenesis of HIV-1 infection.
Acknowledgments
We are grateful to S. Méthot for editorial assistance and M. Imbeault for his assistance with the Metamorph software. We thank Dr. M. Dufour for performing flow cytometric analyses.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was financially supported by an operating grant (to M.J.T.) from the Canadian Institute of Health Research (CIHR) HIV/AIDS Research Program (Grant No. HOP-14438). M.R.T. holds a Doctoral Award from the CIHR. M.J.T. is the recipient of the Canada Research Chair in Human Immuno-Retrovirology (senior level).
2 Address correspondence and reprint requests to Dr. Michel J. Tremblay, Laboratory of Human Immuno-Retrovirology, Research Center in Infectious Diseases, RC709, CHUL Research Center, 2705 Laurier Boulevard, Quebec, Canada, G1V 4G2. E-mail address: michel.j.tremblay@crchul.ulaval.ca
3 Abbreviations used in this paper: HIV-1, HIV type-1; rh, recombinant human; PKC, protein kinase C; PLC, phospholipase C; PIP2, phosphatidylinositol-4',5'-biphosphate; 3D, three dimensional.
Received for publication October 25, 2004. Accepted for publication May 10, 2005.
References
Fortin, J. F., R. Cantin, G. Lamontagne, M. Tremblay. 1997. Host-derived ICAM-1 glycoproteins incorporated on human immunodeficiency virus type 1 are biologically active and enhance viral infectivity. J. Virol. 71: 3588-3596.
Fortin, J. F., R. Cantin, M. J. Tremblay. 1998. T cells expressing activated LFA-1 are more susceptible to infection with human immunodeficiency virus type 1 particles bearing host-encoded ICAM-1. J. Virol. 72: 2105-2112.
Bounou, S., J. E. Leclerc, M. J. Tremblay. 2002. Presence of host ICAM-1 in laboratory and clinical strains of human immunodeficiency virus type 1 increases virus infectivity and CD4+-T-cell depletion in human lymphoid tissue, a major site of replication in vivo. J. Virol. 76: 1004-1014.
Tardif, M. R., M. J. Tremblay. 2003. Presence of host ICAM-1 in human immunodeficiency virus type 1 virions increases productive infection of CD4+ T lymphocytes by favoring cytosolic delivery of viral material. J. Virol. 77: 12299-12309.
Fortin, J. F., R. Cantin, M. G. Bergeron, M. J. Tremblay. 2000. Interaction between virion-bound host intercellular adhesion molecule-1 and the high-affinity state of lymphocyte function-associated antigen-1 on target cells renders R5 and X4 isolates of human immunodeficiency virus type 1 more refractory to neutralization. Virology 268: 493-503.(Fortin, J. F., R. Cantin,)