Previous Article | Next Article ![]()
Journal of Virology, March 2008, p. 2555-2559, Vol. 82, No. 5
0022-538X/08/$08.00+0 doi:10.1128/JVI.01853-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Department of Microbiology and Immunology, McGill University, Montreal, QC H3A 2B4, Canada,1 Department of Molecular Sciences, University of Tennessee Health Sciences Center, Memphis, Tennessee 381632
Received 22 August 2007/ Accepted 13 December 2007
|
|
|---|
|
|
|---|
While human immunodeficiency virus type 1 (23, 37) and amphotropic 10A1 murine leukemia virus (MLV) (18, 24, 27) use a pH-independent pathway, other retroviruses, including mouse mammary tumor virus, avian leukosis and sarcoma virus (ASLV) subgroups A and B, equine infectious anemia virus, foamy virus, and ecotropic Moloney MLV (MoMLV), use a pH-dependent pathway (3, 7, 15, 24, 26, 27, 29, 32, 33). Mouse mammary tumor virus and equine infectious anemia virus appear to use a classical pH-dependent entry mechanism similar to that of VSV, SFV, and influenza A virus, i.e., direct triggering of envelope (Env) glycoprotein conformation changes by low pH (3, 15, 32, 33), while MoMLV entry is more similar to the nonclassical pH-dependent pathway recently reported for Ebola virus and severe acute respiratory syndrome coronavirus (16). Interestingly, ASLV entry involves a unique two-step fusion mechanism in which cellular receptor-induced changes in ASLV Env conformation are required for priming of a subsequent low-pH-triggered step (26).
Jaagsiekte sheep retrovirus (JSRV) is an acutely transforming retrovirus, with its native Env protein functioning as an oncogene to cause lung tumors in sheep and goats (12). It uses ovine and human hyaluronidase 2 (Hyal2) as an entry receptor but is unable to use mouse Hyal2 (31). The pathway used by this simple betaretrovirus during infection has not yet been determined. Here, using pseudovirions of MoMLV, we show that JSRV Env-mediated entry is a relatively slow and nonclassical pH-dependent process that requires dynamin-associated endocytosis.
JSRV entry is weakly inhibited by lysosomotropic agents but is severely impaired by bafilomycin A1. Lysosomotropic agents are widely used for study of pH-dependent virus entry because of their ability to neutralize the acidic environment of endosomal compartments (25). In mouse NIH 3T3 cells expressing human Hyal2 (NIH 3T3/LH2SN), ammonium chloride (NH4Cl) and chloroquine lowered positive control VSV glycoprotein (VSV-G) pseudovirion infection by 90% (P = 0.004-0.007), whereas JSRV and 10A1 Env-mediated infection were reduced by 20 to 30% (P = 0.015 to 0.060) and 10 to 20%, respectively (P = 0.076 to 0.097) (Fig. 1A and B). Surprisingly, bafilomycin A1 (BafA1), an inhibitor of the vacuolar H+-ATPase (2), reduced JSRV infection by 80% (P = 0.002) at 25 nM, comparable to its inhibition of VSV-G pseudovirions (P = 0.045), but had very little effect on that of 10A1 (P = 0.867) (Fig. 1C). A prolonged treatment of cells with 5 or 10 mM of NH4Cl for up to 24 h almost abolished transduction by VSV-G pseudovirions but reduced that of JSRV and 10A1 Env pseudovirions by only 30 and 25%, respectively.
![]() View larger version (25K): [in a new window] |
FIG. 1. JSRV entry is profoundly blocked by BafA1 but is weakly inhibited by lysosomotropic agents. Mouse NIH 3T3 cells overexpressing human Hyal2 (NIH 3T3/LH2SN) were pretreated with NH4Cl (A), chloroquine (B), or BafA1 (C) for 2 h at the indicated concentrations and were exposed to GFP-encoding MoMLV pseudovirions bearing JSRV Env, G protein of VSV (VSV-G), or 10A1 MLV Env in the presence of the agents for 4 h. Noninternalized viral particles were inactivated by citrate buffer (pH 3.0) (JSRV and 10A1) or by 0.25% trypsin (VSV), and cells were incubated for an additional 2 h in the presence of the agents. Following several washes with phosphate-buffered saline, cells were incubated in regular medium for 48 h, and percentages of GFP-positive cells were determined by flow cytometry. The percent infection was calculated relative to results for dimethylsulfoxide-treated cells, and values are the means for two to five independent experiments performed in duplicate ± standard deviations. Experiments were also performed with human HTX and 293 cells expressing endogenous Hyal2, with similar results, except that no inhibitory effect on JSRV by chloroquine was found in HTX cells. Note that comparable multiplicities of infection, typically 0.1 to 0.5, of different pseudovirions were used in a single experiment. The Student t test was used for statistical analysis in these and other experiments throughout this study.
|
The kinetics of JSRV Env-mediated entry is relatively slow and remains sensitive to BafA1. We next examined the entry kinetics to determine if BafA1 acted on an early stage of infection. In NIH 3T3/LH2SN cells, internalization of virus carrying VSV-G or 10A1 Env was rapid, reaching the half-maximal level within the first minute for VSV-G and 15 min for 10A1 Env and reaching maximal levels by 30 min for both (Fig. 2A). Unexpectedly, JSRV Env-mediated infection did not reach half-maximal levels for almost 70 min and did not reach maximal levels even by 4 h (Fig. 2A), indicating that internalization of bound JSRV pseudovirions is a relatively slow process compared to the rate for VSV-G or 10A1 Env.
![]() View larger version (14K): [in a new window] |
FIG. 2. Kinetics of JSRV entry is relatively slow. (A) NIH 3T3/LH2SN cells were bound by JSRV, 10A1, or VSV pseudovirions (MOIs of 0.1 and 0.5, respectively) for 2 h at 4°C to prevent internalization, unbound viral particles were removed with cold phosphate-buffered saline, and the virus-cell complexes were warmed to 37°C (t = 0) to initiate viral internalization and infection. Noninternalized viruses were inactivated at the indicated times after the temperature shift, and viral infectivity was measured 48 h postinfection by flow cytometry. The percent infection was calculated relative to the maximal infection at 4 h following the temperature shift to 37°C. (B) Virus-cell complexes were formed as described for panel A, but 25 nM BafA1 was added instead of inactivating noninternalized virus at 10 min, 30 min, 1 h, 2 h, 3 h, and 4 h after initiation of infection. Note that the total infection period was 6 h in all cases. The percent infection was calculated relative to results for mock-treated cells, and values are averages of two-independent experiments performed in duplicate ± standard deviations.
|
Low pH conditionally overcomes a BafA1-mediated entry block on JSRV. In the classical pH-dependent pathway, a BafA1-imposed fusion block may be rescued by a transient exposure of virus-bound cells to a pulse of acidic extracellular pH (9). We examined if this would be the case for JSRV. A brief pH 5.0 pulse had no effect on the BafA1-induced reduction in JSRV infectivity (Fig. 3A, middle columns). Unexpectedly, the VSV-G pseudovirions were not rescued either. Others previously reported that low-pH rescue of BafA1-blocked VSV is cell type specific, with no rescue occurring on BHK cells and NIH 3T3 cells for reasons that were not clear (20).
![]() View larger version (20K): [in a new window] |
FIG. 3. Low pH conditionally overcomes a BafA1 block on JSRV infection. NIH 3T3/LH2SN cells were given a mock pretreatment (left columns) or pretreated with 25 nM BafA1 for 2 h (middle and right columns) and then were prebound with MoMLV pseudovirions (MOI = 0.1 to 0.5) bearing JSRV Env (A), VSV-G (B), or 10A1 MLV Env (C) for 1 h at 4°C in the absence or presence of the drug. Unbound viruses were removed by cold phosphate-buffered saline, and virion-cell complexes were treated with a pH 7.0 or pH 5.0 pulse at 37°C for 10 min, followed by a continued incubation at 37°C for 4 h in the presence of 25 nM BafA1 (middle columns). Alternatively, virion-bound cells were preincubated at 37°C for 1 h prior to the pH 7.0 or 5.0 pulse, followed by an additional 3-h incubation in the presence of 25 nM BafA1 (right columns). In both cases, the total infection period was 4 h. After 4 h, noninternalized viral particles were inactivated, and viral infectivity was determined 48 h later. The percent infection was calculated relative to results for the DMSO/pH 7.0-treated cells, and values are means of three independent experiments performed in duplicate ± standard deviations.
|
70% of the control infection level (P = 0.007), whereas a control pH 7 treatment in parallel had no effect (Fig. 3A, right columns). A similar pH 5 treatment was much less effective in relieving the BafA1 block on VSV infection (P = 0.08) (Fig. 3B). In contrast, infection by negative control 10A1 pseudovirions was not significantly affected by BafA1 or any of these treatments (Fig. 3C, right columns), in agreement with previous reports that it enters using a pH-independent pathway not affected by BafA1 (24, 27). Similar results were also obtained for JSRV in 293 cells overexpressing human Hyal2, albeit to a lesser extent, whereas a stronger rescue was observed for VSV in this cell line (data not shown). JSRV pseudovirions are resistant to acidic pH inactivation. Characteristically, the infectivity of classical pH-dependent viruses is inactivated by low pH due to premature conformational changes in their viral fusion proteins (9). The exception is VSV, which is believed to resist inactivation because the G protein is able to refold into its original conformation after the pH is returned to neutral (30). Thus, we added SFV-E1 pseudovirions to this study as the positive control (4). Surprisingly, JSRV infectivity was almost as resistant to low-pH inactivation as the VSV-G pseudovirions, in sharp contrast to the positive control SFV-E1 pseudotypes, which were inactivated at pH 5.6 (Fig. 4). The effect of either pH 5.6 or pH 5 treatment and the 50% inactivation by pH 3.6 treatment on JSRV were not statistically significant (P = 0.363, 0.177, and 0.145, respectively). A pH 3.0 treatment resulted in an almost complete inactivation of JSRV infectivity (P = 0.006) (Fig. 4). These results indicated that the JSRV pseudovirions were resistant to an acidic inactivation, suggesting that additional events besides low pH are involved in JSRV fusion and entry. Alternatively, low-pH-induced conformational changes of the JSRV Env protein are reversible, similar to the case with VSV-G.
![]() View larger version (21K): [in a new window] |
FIG. 4. JSRV pseudovirions are resistant to low-pH inactivation. MoMLV pseudovirion stocks bearing JSRV Env, VSV-G, or SFV E1 were warmed at 37°C for 5 min, and then PBS-NaOH or PBS-HCl (pH 7.6, 7.0, 5.6, 5.0, 3.6, or 3.0) was added and virions were incubated for an additional 30 min at 37°C, after which samples were neutralized by rapid addition of equal volumes of regular medium containing 25 mM HEPES in phosphate-buffered saline (pH 7.4). NIH 3T3/LH2SN cells were exposed to serial dilutions of each sample for 8 h, and viral infectivity was determined by flow cytometry 48 h later. The values are calculated relative to the pH 7.6-treated viral stocks and are averages for two to four independent experiments performed in duplicate ± standard deviations. The P values (calculated using Student's t test) for VSV-G pseudovirions were 0.259 at pH 5.6, 0.303 at pH 5.0, and 0.052 at pH 3.0, respectively.
|
100-kDa cell membrane-associated GTPase that is essential for both clathrin- and caveola-mediated endocytosis (14) and facilitates release of endocytic vesicles from the plasma membrane (22). VSV uses dynamin-dependent endocytosis for entry (38). We asked if dynamin was important to JSRV entry. Due to the low transfection efficiency of NIH 3T3/LH2SN cells, we chose to use the highly transfectable 293 cells in this experiment. Expression of wild-type (Dyn-WT) and dominant-negative (Dyn-K44A) dynamin in transfected 293 cells was verified by Western blotting, and their levels of expression were similar (Fig. 5A). Their expression was not cytotoxic over the course of experiments. While expression of Dyn-WT had negligible effects (Fig. 5B), Dyn-K44A expression inhibited JSRV transduction by approximately 70% (P = 0.015) but had no effect on that of 10A1 pseudovirions (P = 0.20). Dyn-K44A inhibition of JSRV entry was comparable to its inhibition of the control VSV-G pseudovirions (P = 0.011). No significant difference in the Hyal2 expression levels was observed between cells expressing the K44A and wild-type forms, indicating that Dyn-K44A did not down-regulate Hyal2 expression on the cell surface (data not shown). Taken together, these results strongly suggested that a dynamin-dependent endocytosis pathway is involved in JSRV entry.
![]() View larger version (18K): [in a new window] |
FIG. 5. JSRV entry is blocked by a dominant-negative mutant of dynamin. Human 293 cells were transiently transfected with plasmids encoding amino-terminally hemagglutinin epitope-tagged dynamin wild type (Dyn-WT) or a dominant-negative mutant (Dyn-K44A) (kind gifts of Eric Cohen, Institut de Recherches Cliniques de Montréal, Montréal, Canada) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). (A) Expression of Dyn-WT and Dyn-K44A. Cell lysates were harvested from a portion of the transfected cells and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis using our previously published methods (19), and expression of dynamin was detected by Western blotting using an anti-hemagglutinin antibody (Sigma, St. Louis, MO) to detect the HA-dynamin fusion protein (upper panel). Similar amounts of cell lysates were loaded onto a sodium dodecyl sulfate-polyacrylamide gel and were analyzed for β-actin using an antibody against β-actin (Sigma) (lower panel). (B) Relative infectivity of JRSV pseudovirions in cells expressing Dyn-WT or Dyn-K44A. Twenty-four to forty-eight hours posttransfection, transfected cells were exposed to serial dilutions of alkaline phosphatase (AP)-encoding MoMLV pseudovirions bearing JSRV Env, VSV-G, or 10A1 MLV Env. Viral titers were determined by counting AP-positive foci 48 h postinfection. Values are calculated relative to viral titers obtained from the empty vector-transfected cells and are means for three independent experiments ± standard deviations.
|
One possible reason for the lack of strong inhibition by the weak bases is that the JSRV pseudovirions arrested by NH4Cl or chloroquine in endocytic vesicles are stable and remain infectious, and once these agents are removed from culture medium, infection might quickly resume. We favor this possibility because it is consistent with the reversibility of the weak bases on endocytic acidification (25) and with our findings that JSRV entry is a relatively slow process compared to that of VSV and that JSRV Env is stable enough to resist acidic inactivation. Alternatively, the lack of strong inhibition may result from a pleiotropic effect on an unidentified cellular target of these lysosomotropic agents.
Given that dynamin was shown in this study to be important to JSRV entry (Fig. 5) and that the JSRV receptor, Hyal2, is a glycosylphosphatidylinositol-anchored protein likely localized in lipid rafts, we favor that JSRV utilizes a dynamin-dependent, caveola-mediated endocytosis pathway for entry. Such pathways were previously shown to be critical to simian virus 40 (28), Ebola virus (11), and amphotropic MLV entry (1). Future studies should address if JSRV utilizes a caveola-mediated or clathrin-mediated endocytosis pathway.
Rescue of BafA1 block on JSRV entry by low pH required preincubation of virus-bound cells (Fig. 3A and data not shown). These results were unexpected given our finding in the accompanying paper that JSRV Env induces syncytium formation and cell-cell fusion at low pHs (6). However, syncytium formation has not always correlated viral infection (8, 17, 20), and the ability of a low pH to overcome a BafA1 block can also be cell type dependent (20).
The atypical pH-dependent characteristics reported here suggest that JSRV may use a two-step fusion mechanism similar to that of ASLV (26). Consistent with this possibility, JSRV pseudovirions were resistant to low-pH inactivation (Fig. 4) and low pH alone was not sufficient to overcome a BafA1 block on JSRV entry (Fig. 3A). Future work will investigate if JSRV fusion and cell entry are a two-step process, and if so, the role of the initial receptor-induced conformational changes in Env and of low-pH-induced subsequent conformational triggering.
This work was supported by funding from the Canadian Institute of Health Research (CIHR) to S.-L.L. and a U.S. National Institutes of Health grant to L.M.A. (AI 33410). P.B. was supported by Fonds Québécois de la Recherche sur la Nature et les Technologies (FQRNT) to P.B. and Natural Sciences and Engineering Research Council of Canada (NSERC) and Fonds de la Recherche en Santé du Québec (FRSQ) scholarships to M.C. S.-L. Liu is a Canada Research Chair in Virology and Gene Therapy.
Published ahead of print on 19 December 2007. ![]()
|
|
|---|
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»