Previous Article | Next Article ![]()
Journal of Virology, December 2005, p. 14489-14497, Vol. 79, No. 23
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.23.14489-14497.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Molecular Genetics and Biochemistry,1 Renal-Electrolyte Division and Department of Cell Biology and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 152612
Received 23 July 2005/ Accepted 13 September 2005
|
|
|---|
|
|
|---|
The results of these studies to date have revealed two predominant post-receptor-binding entry pathways accessed by enveloped animal viruses. These distinct pathways are designated pH-independent and pH-dependent entry. In the pH-independent entry pathway, enveloped virus binding to specific receptor triggers a direct fusion of the viral and cellular membrane at the extracellular pH, without a requirement for an acidic environment. This pH-independent pathway is utilized by enveloped viruses such as hepatitis B virus (26), Sendai virus (50), and the human and simian immunodeficiency viruses (15, 41). In the pH-dependent pathway, the receptor binding directs the virus into an intracellular compartment in which an acidic environment is required for fusion of the viral and cellular membranes (21, 28, 54). The pH-dependent pathway is utilized by enveloped viruses such as Semliki Forest virus (29), West Nile virus (11), Hantaan virus (32), vesicular stomatitis virus (41), influenza virus (25), avian leukosis virus (16), and salmon anemia virus (19). Viral entry of amphotropic and ecotropic murine leukemia viruses can be both pH dependent and pH independent, conditional on the virus strain (34, 41, 43). Interestingly, some enveloped viruses, like herpes simplex virus, may enter target cells via more than one pathway (3).
As demonstrated by the summary above, individual members of the retrovirus family apparently can utilize different entry pathways. Among the oncoviruses examined, the murine leukemia viruses enter by a pH-independent pathway, while the avian leukosis viruses enter by pH-dependent endocytosis. To date, examination of lentivirus entry mechanisms has indicated that human and simian immunodeficiency viruses enter target cells by pH-independent pathways. However, more recent studies have reported that human immunodeficiency virus type 1 (HIV-1) entry specificity may be dependent on the specific target cell, as HIV-1 evidently can establish productive infection in certain cell types by receptor-specific, clathrin-mediated endocytosis (15). In addition, other studies indicate that HIV-1 entry may or may not require internalization of CD4 receptor (36-38, 46) and that HIV-1 can infect certain cultured CD4-negative human fibroblast cells, suggesting alternatives to CD4-mediated entry (8, 48, 56). These apparent variations reported from independent studies of HIV-1 entry pathways naturally raise the question of the type of entry mechanisms used by other animal lentiviruses, especially those with defined cell receptors.
Equine infectious anemia virus (EIAV), an exclusively macrophage-tropic lentivirus, causes a uniquely dynamic disease in horses that provides a novel model for examining the diverse pathologies associated with lentivirus infection of monocytes and macrophages (42). We recently cloned and characterized a functional cellular receptor for EIAV, designated as equine lentivirus receptor-1 (ELR1) (60). The ELR1 protein is a member of the tumor necrosis receptor protein family and appears to be sufficient for mediating productive virus infection in the absence of any coreceptor, in contrast to human, simian, and feline lentiviruses, which typically require coreceptors. Thus, it was of interest to examine the entry pathway utilized by EIAV in naturally susceptible cells and in nonsusceptible cells transduced with the ELR1 protein.
In the present study, we describe a series of experiments in which we examine the ability of selective drug treatments to inhibit EIAV entry, leading to productive infection of equine macrophages, permissive equine fibroblastic cells, and ELR1-transduced mouse cells. The results of these studies demonstrate for the first time that EIAV receptor-mediated infection occurs via a pH-dependent endocytic pathway. These results are in general agreement with the conclusions drawn by Brindley and Maury in independent experiments with different strains of EIAV, as described in the companion paper (7).
|
|
|---|
Cell viability assay. A 5-mg/ml solution of MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) was freshly prepared and filtered. Cells were incubated with MTT for 4 h at 37°C, washed, and resuspended in isopropanol-HCl solution. The level of color production was assayed on an enzyme-linked immunosorbent assay plate reader by measuring optical density at 570 nm. Concentrations of drugs used in inhibition studies were limited to those levels maintaining at least 90% of the viability observed in untreated cell cultures; drug concentrations further reducing cell viability were excluded from additional consideration.
Virus stocks. The parental pathogenic proviral molecular clone EIAVUK has been described in detail by Cook et al. (13). ED cells chronically infected with EIAVUK were cultured, and the supernatants from these cells were harvested every 3 days, clarified by centrifugation, aliquoted, and frozen at 80°C. The viral titer on ED cells was 1.66 x 105 IU/ml, as titrated by a standard viral infectious center assay (27).
Measurement of EIAV entry into target cells. To detect virus entry, we initially employed a quantitative real-time PCR assay designed to specifically amplify early reverse transcriptase (early RT) DNA products produced soon after virus internalization. As previously shown in our lab, this assay is a sensitive and specific indicator of EIAV receptor-mediated entry (33, 60).
Pharmacological inhibition of EIAV entry. Specific drug treatments included the use of lysosomotropic agents (ammonium chloride and chloroquine) and vacuolar H+-ATPase (V-ATPase) inhibitors (bafilomycin A1 [BafA1] and concanamycin A [CA]) to inhibit acidification of endosomal compartments (5, 6, 12, 53, 57-59) and sucrose to block clathrin-coated endocytosis (11, 32, 44). Unless otherwise noted, all of the lysosomotropic agents and V-ATPase inhibitors were purchased from Sigma (St. Louis, MO). Working solutions of BafA1 and CA were prepared in dimethyl sulfoxide and stored at 20°C. Stock solutions of ammonium chloride, chloroquine, and sucrose were prepared daily in distilled water and sterilized through a 0.2-µm filter.
To assess the effects of specific drug treatments on EIAV entry, approximately 4 x 105 cells per well were seeded in six-well plates and incubated overnight. Virus infection was performed in duplicate wells. Cells were infected with EIAVUK at a multiplicity of infection of 0.08 for 2 h at 37°C in the presence of various concentrations of V-ATPase inhibitors, lysosomotropic agents, and dimethyl sulfoxide or medium (diluent controls). After infection, unabsorbed virus was removed by washing three times with phosphate-buffered saline (PBS), and the cells were cultured in fresh drug-free medium for an additional 4 h. Controls included mock-infected cells in the presence of the respective drug or inoculated cells with either medium or dimethyl sulfoxide in place of specific drug treatment. In some cases, cells were incubated with agents for 30 min at 37°C prior to infection.
At 6 h postinfection (hpi), the total DNA was extracted from the infected and treated cells using a DNeasy tissue kit (QIAGEN, Valencia, CA) and subjected to quantitative real-time PCR. Individual samples were evaluated in duplicate in at least two separate experiments. Virus infectivity was determined based on production of early reverse-transcribed viral DNA copies using quantitative real-time PCR. The copy number was normalized for copies of the gapdh gene, as described elsewhere (33). The relative infectivity for the culture treated with different agents was normalized against the same cells infected in the absence of agents.
Assays of selective drug inhibition on virus production. To examine the effect of specific agents on virus production, drug treatments were performed as described above, except that the postinfection period was extended to 1 to 6 days. To prevent sequential rounds of virus infection from infected cells, the reverse transcription inhibitor 3'-azido-3'-deoxythymidine (AZT; Sigma), was added at a final concentration of 1 µM to selected cultures at 24 hpi. The culture supernatants were collected, clarified by centrifugation to remove cell debris, and stored at 80°C prior to measuring RT activity as described previously (10).
Assays of EIAV penetration of target cells. To assess the effects of drug treatments specifically on the penetration step in virus infection, infectious EIAVUK virus particles were prebound to target cells for 1 h at 4°C in the absence of drugs. The infectious supernatant was then removed, and the cells were fed with fresh medium in the presence or absence of agents and shifted to 37°C for 2 h. The medium was removed, and the cells were washed three times with cool PBS and cultured in fresh medium without drugs for an additional 4 h. At 6 hpi, the cells were harvested and total cellular DNA was extracted and subjected to quantitative real-time PCR to measure early RT products.
Effect of drug treatment on ELR1 expression. To examine the effect of drug treatments on EIAV receptor protein expression, 3T3-RHA cells expressing HA-tagged ELR1 were infected with virus in the presence of the indicated concentrations of CA for 2 h at 37°C. Cells were fixed with 2% paraformaldehyde and permeabilized with 0.1% Triton X-100. Subsequently, cells were stained with rat anti-HA antibody (clone 3F10; Roche) followed by staining with anti-rat immunoglobulin G-fluorescein isothiocyanate-conjugated antibody (Roche) at 4°C. The level of protein expression was analyzed by flow cytometry using a BD FACSCalibur (BD Biosciences, San Jose, Calif.).
|
|
|---|
As summarized in Fig. 1, the results of these inhibition assays revealed substantial reduction of early RT products, and thus virus entry, in all cell types by treatments with BafA1, CA, or ammonium chloride. In contrast, no significant reduction in early RT levels was produced by the presence of either chloroquine or sucrose during virus infection of the same target cells. Specifically, a 97% reduction in early RT products relative to untreated cell infections resulted from treatment of the ED and FEK cells (Fig. 1A and B) with either 5 nM CA or 0.3 µM BafA1, while virus infection in the presence of 50 mM ammonium chloride reduced early RT product levels by about 43% relative to control cell infections. Importantly, a similar pattern of inhibition of virus entry was observed in equine macrophages (Fig. 1C), the natural target cell for EIAV, although relatively higher drug concentrations were required to achieve similar levels of inhibition observed in the ED or FEK cells. For example, 50 nM CA produced a 99% reduction in virus entry, 1 µM BafA1 produced an 83% reduction, and 20 mM ammonium chloride produced a 25% reduction (Fig. 1C). Cell viability measurements using MTT assays (data not shown) confirmed that the cells incubated with the indicated drug concentrations were at least 95% viable, as observed with untreated cells.
![]() View larger version (16K): [in a new window] |
FIG. 1. Effect of endocytosis inhibitors on EIAVUK infectivity in FEK (A), ED (B), or equine macrophages (C). The specific inhibitors at the indicated concentrations were added to target cells at the indicated concentrations during the initial 2-h incubation with infectious virus. FEK and ED cells were treated with 0.3 M sucrose, 200 µM chloroquine, 50 mM NH4Cl, 5 nM CA, or 0.3 µM BafA1. Equine macrophages were treated with 150 mM sucrose, 10 µM chloroquine, 20 mM NH4Cl, 50 nM CA, or 1 µM BafA1. At 6 hpi, the cells were harvested, and the total cellular DNA was isolated and subjected to quantitative real-time PCR to assay EIAV early RT products. The data are representative of several independent experiments. Individual treatments were performed in triplicate, and the error bars show deviations from the means. The level of early RT DNA observed in untreated control cells infected with EIAVUK was set to 100% and used for comparison to early RT levels in cells infected in the presence of drugs to calculate a measure of relative infectivity.
|
Taken together, these initial observations of substantial inhibition of virus entry by a lysosomotropic agent and two V-ATPase inhibitors suggest that EIAV entry is by a pH-dependent endocytic pathway. In addition, the absence of inhibition by sucrose is consistent with EIAV entry being independent of clathrin-mediated uptake.
Inhibition of EIAV entry is dose dependent. Having identified specific inhibitors of EIAV entry in the preceding screening assays, we next examined the dose dependence of the observed inhibitory effects by BafA1, CA, and ammonium chloride. The data in Fig. 2 demonstrate that all three agents inhibited virus entry in a dose-dependent manner, and they also reveal the relative potency of the various drugs in suppressing EIAV entry into FEK or ED cells. For example, from Fig. 2, it can be estimated that a greater-than-90% inhibition of virus entry into either the FEK or ED cells required at least 100 nM BafA1, but only about 3 to 4 nM CA. In contrast, only about 75% virus entry was inhibited in both cell types by 50 mM ammonium chloride, the maximum concentration that could be used without significant cytotoxicity. Based on the relative specific activities of the various agents evaluated, we chose the V-ATPase inhibitors for subsequent mechanistic studies of the inhibition of EIAV entry into target cells.
![]() View larger version (23K): [in a new window] |
FIG. 2. Dose dependence of inhibition of EIAVUK entry into FEK or ED cells by treatments with different concentrations of endocytosis inhibitors. Infections of FEK or ED cells with EIAVUK were performed as described in the legend for Fig. 1 in the presence of increasing concentrations of BafA1 (A and D), CA (B and E), or NH4Cl (C and F). After 2 h of infection at 37°C in the presence of drug, the cells were washed with PBS and fed with drug-free medium. At 6 hpi, total cellular DNA was isolated for determining early RT products as a measure of virus infectivity as described in Materials and Methods. The results represent the averages of two independent experiments, with the standard deviations indicated as error bars.
|
![]() View larger version (15K): [in a new window] |
FIG. 3. Kinetics of inhibition of EIAVUK infection by endocytosis inhibitors. Cultures of FEK cells were infected with EIAVUK at 37°C and treated with either 300 nM BafA1 or 5 nM CA at the following times relative to virus inoculation: 0 to 2, 2 to 4, or 4 to 6 hpi. At 6 hpi, cells were harvested and assayed for early RT products as a measure of virus infectivity as described in the preceding figure legends. Early RT levels observed in untreated control cells at 6 hpi were set as 100% and compared to the respective early RT levels observed in cells treated with drugs at the indicated times to calculate a measure of relative infectivity. Data represent the averages of at least two independent experiments with treatments performed in duplicate wells, with standard deviations indicated.
|
![]() View larger version (11K): [in a new window] |
FIG. 4. Effect of V-ATPase inhibitors on EIAVUK penetration into target cells. FEK (A) or ED (B) cells were incubated with EIAVUK for 1 h at 4°C to allow binding but not penetration of virus into target cells. Unbound virus was then removed by washing, and the cultures were shifted to 37°C to initiate virus entry in the presence or absence of the indicated V-ATPase inhibitors (5 nM CA or 300 nM BafA1). At 6 hpi, the cells were harvested, and the total DNA was isolated and assayed for early RT as a measure of infectivity. The early RT levels observed in untreated control cells were set as 100% and compared to early RT levels observed in treated cells as a measure of relative infectivity. The data presented here represent the averages of two independent experiments.
|
![]() View larger version (14K): [in a new window] |
FIG. 5. Inhibition of EAIVUK entry into murine cells transduced to express the ELR1 receptor. (A and B) Murine 3T3-RHA cells were infected with EIAVUK in the presence or absence of either of the indicated doses of CA (A) or ammonium chloride (B) and assayed for early RT products as a measure of virus infectivity, as described in Materials and Methods. Control, nontransduced NIH 3T3 cells were incubated with infectious EIAV to demonstrate lack of significant early RT product production indicative of entry in the absence of the ELR1 receptor, as reported by Zhang et al. (60). (C) To examine the effect of inhibitor treatments on ELR1 expression levels, flow cytometry was used to analyze the levels of ELR1 expression on untreated 3T3-RHA cells compared to cells treated with the indicated concentration of CA for 2 h at 37°C. Following this incubation, the cells were then stained with anti-HA and anti-immunoglobulin G-fluorescein isothiocyanate at 4°C to label the HA-tagged ELR1 protein. NIH 3T3 cells treated with drugs and stained with the above antibodies were measured to determine nonspecific binding by the anti-HA antibody (gray peak). Thin line, 0 nM CA; dark solid line, 5 nM CA; dotted line, 50 nM CA; dashed line, 200 nM CA.
|
![]() View larger version (14K): [in a new window] |
FIG. 6. Effect of CA on virion production in single- and multiple-cycle infections by EIAVUK. Cultures of FEK (A) or ED (B) cells were infected by EIAVUK at 37°C in the presence or absence of 5 nM CA. At 24 hpi, 1 µM AZT was added to one-half of the cell cultures to inhibit subsequent cycles of virus infection, while the remaining cell cultures were continued in the absence of AZT to allow multiple cycles of virus infection. At the indicated days postinfection, culture supernatants were analyzed for RT activity as a measure of virion production. The data presented represent the mean values from duplicate experiments.
|
|
|
|---|
In addition to the major finding that EIAV entry into permissive fibroblastic and macrophage cells utilizes pH-dependent endocytosis, the studies described here also provide new information about important specific details of the mechanism of virus entry. In this regard, the observation that CA and ammonium chloride blocked EIAV entry into 3T3-RHA cells provides conclusive proof that pH-dependent endocytosis is associated with EIAV entry via a defined functional receptor (ELR1) and that this entry pathway is not restricted to nonspecific uptake of virus by target cells. Moreover, the kinetics of inhibition of virus entry (Fig. 3 and 4) demonstrate that the inhibitors specifically block early steps in virus infection that follow binding of the virion to the cell receptor and that are completed by about 2 h postexposure; drug treatments initiated at 2 to 4 h postexposure failed to inhibit virus entry. This time window for effective inhibition is consistent with the time required for the formation and acidification of endocytic vesicles that is necessary for pH-dependent virus entry (9, 23).
Despite the consistency of the data from the current EIAV entry studies, there was an inconsistency in the fact that chloroquine treatment failed to inhibit EIAV entry, in distinct contrast to the substantial inhibition observed with the other lysosomotropic agent, ammonium chloride. Both of these agents are weak bases that in relatively high concentrations enter and block the acidification of intracellular vesicles. It does not appear that the failure of chloroquine to inhibit EIAV entry can be attributed to the concentration used in the current study (10 µM), as chloroquine at 0.6 µM has been shown to raise endosomal pH levels (52). However, it is possible that equine cells are more resistant to chloroquine treatment, as previous studies have indicated a cell type dependency of chloroquine sensitivity (14, 30). We observed substantial cytotoxicity in equine macrophages treated with greater than 10 µM chloroquine, which precluded further evaluations of inhibition of EIAV entry by this agent in the current studies. Although the reason for the lack of inhibition by chloroquine is not certain, it is important to note that discrepancies between ammonium chloride and chloroquine as inhibitors of virus entry have been reported previously (14, 30, 47).
While the inhibition of EIAV entry observed with ammonium chloride and the two V-ATPase inhibitors indicates the importance of pH-dependent endocytosis in virus entry, the lack of inhibition by sucrose treatment suggests that this entry pathway is not clathrin dependent, as high sucrose concentrations inhibit clathrin-dependent endocytosis (11, 31, 32, 35, 44). However, this conclusion should at this time be considered tentative, as the data are currently based on a lack of effect by a single drug treatment. The postentry cellular pathways utilized by infecting viruses may follow diverse and complex pathways, and biochemical and imaging studies using additional specific inhibitors of endocytic pathways are indicated to clarify the role of clathrin-mediated endocytosis in EIAV entry.
The current finding that receptor-mediated EIAV infection is by pH-dependent endocytosis is in contrast to the predominantly pH-independent entry reported for HIV-1, HIV-2, and simian immunodeficiency virus (20, 39-41). This apparent difference may reflect the fact that EIAV infection can be mediated by a single receptor (e.g., ELR1), while productive infection by human and nonhuman primate lentiviruses in general requires a combination of coreceptors (e.g., CD4 and CXCR4 or CCR5). Studies on the mechanisms of HIV-1 receptor interactions indicate that sequential receptor binding is necessary to induce the fusion active form of the viral envelope transmembrane protein to mediate fusion of the viral and cellular lipid bilayers (24, 51, 55). Based on the current data, we would propose that EIAV binding to the ELR1 receptor triggers endocytosis of the virion into vesicles in which subsequent acidification activates viral envelope fusion of the viral and vesicle lipid bilayers. The data reported in the companion paper by Brindley and Maury (7) demonstrate an optimization of EIAV infectivity at pH levels of around 5.0, consistent with the proposed pH dependence of the EIAV envelope in mediating fusion and entry into target cells. In this regard, it is also interesting that Clements and colleagues (2) have identified an apparent complex of a 30-kDa chondroitin sulfate proteoglycan and a 45-kDa protein kinase as a functional receptor for visna virus, a sheep lentivirus. These observations raise the possibility that the ungulate lentiviruses of horses, sheep, and goats that exclusively infect monoctyes/macrophages may have a common entry mechanism in which interactions with a single receptor protein induces pH-dependent endocytosis, perhaps reflecting a fundamental difference in the envelope structure-function related to fusogenicity. Thus, future studies are designed to characterize in detail the specificity and mechanisms of EIAV envelope fusion activity.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»