Role of the Human T-Cell Leukemia Virus Type 1 PTA
http://www.100md.com
病菌学杂志 2006年第7期
Institute for Molecular Virology
Departments of Diagnostic and Biological Sciences and Microbiology
Cancer Center, University of Minnesota, Minneapolis, Minnesota 55455
Comprehensive Cancer Center
Department of Neuroscience, Ohio State University, Columbus, Ohio 43210
ABSTRACT
Human T-cell leukemia virus type 1 (HTLV-1) Gag is targeted to the plasma membrane for particle assembly and release. How HTLV-1 Gag targeting occurs is not well understood. The PPPY and PTAP motifs were previously shown to be involved in HTLV-1 particle release with PTAP playing a more subtle role in virus budding. These L domains function through the interaction with host cellular proteins normally involved in multivesicular body (MVB) morphogenesis. The plasma membrane pathway rather than the MVB pathway was found to be the primary pathway for HTLV-1 particle release in HeLa cells. Intriguingly, disruption of the PTAP motif led to a defect in the targeting of Gag from the plasma membrane to CD63-positive MVBs. Particles or particle buds were observed to be associated with MVBs by electron microscopy, implying that Gag targeting to the MVB resulted in particle budding. Blocking clathrin-dependent endocytosis was found not to influence localization of the HTLV-1 Gag PTAP mutant, indicating that Gag did not reach the MVBs through clathrin-dependent endocytosis. Our observations imply that the interaction between Gag and TSG101 is not required for Gag targeting to the MVB. Overexpression of dynamitin p50 increased particle release, suggesting that there was an increase in the intracellular transport of MVBs to the cell periphery by the utilization of the dynein-dynactin motor complex. Intriguingly, virus particle release with this mutant was reduced by 20-fold compared to that of wild type in HeLa cells, which is in marked contrast to the less-than-twofold defect observed for particle production of the HTLV-1 Gag PTAP mutant from 293T cells. These results indicate that the role of the PTAP motif in L domain function is cell type dependent.
INTRODUCTION
Many retroviruses initiate particle assembly and budding at the plasma membrane where Gag proteins associate with raft microdomains of the inner leaflet of the plasma membrane (15, 34). The Gag polyprotein is the main driving force for assembly and budding. The targeting of Gag proteins to the cell membrane and stable membrane association are accomplished using a bipartite membrane binding signal consisting of the fatty acid myristate (7, 32) added cotranslationally to the N terminus of Gag in concert with a patch of basic residues within the matrix (MA) domain) (12, 17, 22, 40, 41). Myristate provides a hydrophobic interaction with the lipid membrane, while the basic residues strengthen the interaction by forming the electrostatic interactions with acidic phospholipids that are enriched at the cytoplasmic face of the plasma membrane (26, 41).
The release of the nascent virions from the cell is mediated by the late (L) domain (9, 13). The PTAP-interacting protein TSG101 functions by recruiting the intracellular machinery, ESCRT complexes, to the site of assembly at the plasma membrane (13, 15, 21, 36). These ESCRT complexes are normally involved in multivesicular body (MVB) biogenesis (1, 2). The formation of MVBs, a late endosomal compartment, is characterized by the invagination of limiting membranes to bud small vesicles. Topologically, both virus budding at the plasma membrane and the vesicle budding into MVBs share striking similarities: they occur away from the cytosol, and they involve lipid microdomains, inositol derivatives, and a particular cohort of proteins such as TSG101 and ubiquitin (9, 13).
Previous studies have investigated the role of the human T-cell leukemia virus type 1 (HTLV-1) PPPY and PTAP motifs in virus budding at the plasma membrane (5, 6, 16, 20, 31, 37). Both motifs have been implicated in budding of HTLV-1 from the plasma membrane (6, 37). The PPPY and PTAP motifs function through the interaction with the Nedd4 family of proteins (E3 ubiquitin ligases) and the TSG101 protein (an E2-like ubiquitin-conjugating enzyme), respectively (5, 6, 16). The PPPY motif was observed to be more important than PTAP in particle release (6, 16, 37). However, it remains unclear what different roles these two motifs play in HTLV-1 assembly and release. It has been suggested that PPPY and PTAP act successively in the HTLV-1 assembly process to ensure proper Gag trafficking through the endocytic pathway to late endosomes for particle release (5).
Despite the view that many retroviruses recruit the cellular machinery normally responsible for MVB formation to the plasma membrane for their assembly and budding, it has been long known that in macrophages human immunodeficiency virus type 1 (HIV-1) particles accumulate in an intracellular, vacuole-like compartment (25). A number of recent reports have identified the vacuole-like compartment as a CD63-positive MVB, indicating that HIV-1 particles may be released into the MVB (23, 27, 28). Such observations were further extended to standard tissue culture cell lines previously believed to support budding only at the plasma membrane (23, 24, 33). The intracellularly assembled HIV-1 particles are likely released into the extracellular environment through the fusion of virus-containing compartments directly with the cell surface in an exocytic fashion (23, 24, 27, 28, 33). Such exosome-assisted release of proteins is best documented for the major histocompatibility complex class II (MHC class II) (29, 39). It is currently unclear how retroviral Gag is targeted to MVBs.
Many intracellular vesicles are transported along polarized cytoskeletal elements (i.e., microtubules) by utilizing molecular motor proteins (18). Kinesin moves towards the fast-growing end (or plus end), away from the microtubule organizing center (MTOC), and cytoplasmic dynein moves toward the minus end. In the case of dynein, vesicle binding is thought to be mediated by the interaction of the dynein intermediate chains with the 150-kDa dynactin subunit Glued. Overexpression of the 50-kDa "dynamitin" subunit of the dynactin complex can dissociate the complex (18). Under these conditions dynein is released from its cargo sites, resulting in inhibition of minus-end-directed vesicle transport. The dynein-dynactin motor complex has been reported to be involved in the intracellular transport of the MHC class II-containing compartment (MIIC), and these MIICs can be released as exosomes (29, 38, 39).
To date, the use of the MVB pathway for budding has mainly been reported for HIV release from macrophages. In the present study, the plasma membrane pathway rather than the MVB pathway was found to be the primary pathway for HTLV-1 particle release in HeLa cells. Intriguingly, when PTAP is mutated, Gag strongly colocalizes with CD63, an MVB marker. Particles and particle buds were observed with intracellular vacuoles, suggesting that Gag targeting to the MVB resulted in particle budding. When clathrin-dependent endocytosis was blocked, localization of the HTLV-1 Gag PTAP mutant was not altered, indicating that Gag did not reach the MVBs through clathrin-dependent endocytosis. These findings imply that the interaction between Gag and TSG101 is not required for Gag targeting to the MVB. Overexpression of dynamitin p50 increased particle release, suggesting that there was an increase in the intracellular transport of MVBs to the cell periphery by the utilization of the dynein-dynactin motor complex. Virus particle release with this mutant was reduced by 20-fold compared to that of wild type (wt) in HeLa cells, which is in marked contrast to the less-than-twofold defect observed for particle production of the HTLV-1 Gag PTAP mutant from 293T cells. These results indicate that the role of the PTAP motif in L domain function is cell type dependent.
MATERIALS AND METHODS
Cell culture and transfection. HeLa cells were maintained in Dulbecco's modified Eagle's medium (GIBCO BRL, Gaithersburg, MD) supplemented with 10% Fetal Clone III serum (HyClone, Logan, UT). Transfection of cells was performed by the calcium phosphate precipitation method. Typically, 100-mm petri dishes were used in this study unless otherwise noted. A total of 10 μg of purified plasmid DNA was mixed with 50 μl 2.5 M CaCl2 and 500 μl 2x HEPES-buffered saline and then added to cells that were split the day before transfection and at 50 to 70% confluence. At 12 to 15 h after transfection, medium was replaced with medium with 10 mM sodium butyrate and 20 mM HEPES. After 8 h of incubation, medium was replaced with fresh Dulbecco's modified Eagle's medium with 20 mM HEPES.
Plasmids. The plasmid pCMV-HT1 was derived from an infectious molecular clone of HTLV-1 by replacing the 5' long terminal repeat (LTR) with a cytomegalovirus immediate-early promoter joined to a small fragment from the R region of the LTR that contains the major splice donor site (8). Eps-DIII-GFP (green fluorescent protein) and Eps-D32-GFP were kindly provided by Alexandre Benmerah (Developpement Normal et Pathologique de Systeme Immunitaire, Paris, France). The dynamitin p50 expression construct was kindly provided by Jacques Neefjes (Netherlands Cancer Institute, The Netherlands).
Antibodies. The following antibodies were used: mouse monoclonal anti-HTLV-1 p19 (Zeptometrix, Buffalo, NY), goat polyclonal anti-HTLV-1 p24 (Advanced Biotechnologies, Columbia, Maryland, and Advanced BioScience Laboratories, Inc., Kensington, Maryland), mouse monoclonal anti-CD63 (BD Biosciences), mouse monoclonal anti-LAMP-1 (Southern Biotechnology Associates, Inc., Birmingham, AL), mouse monoclonal anti-EEA1 (BD Biosciences), mouse monoclonal anti-GM130 (Transduction Laboratories), mouse monoclonal anti--tubulin (Sigma, Saint Louis, Missouri), and mouse monoclonal antidynactin p50 (BD Biosciences). Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G (IgG), Alexa Fluor 568-conjugated rabbit anti-goat IgG, and Alexa Fluor 647-conjugated goat anti-mouse IgG were purchased from Molecular Probes (Molecular Probes, Eugene, OR). Nanogold-conjugated rabbit anti-mouse IgG (NanoProbes Inc., Yaphank, NY) was used for electron microscopy immunolabeling.
Tf uptake. The transferrin (Tf) uptake assay was performed using Alexa Fluor 546-labeled transferrin (Molecular Probes, Eugene, OR). HeLa cells were incubated with 0.1 μM Alexa Fluor 546-labeled transferrin at 37°C for 15 min for binding and then washed and fixed for immunofluorescence analysis.
Confocal microscopy. Cells were grown on coverslips and fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 in phosphate-buffered saline (PBS), quenched in 100 mM glycine-PBS, and blocked in PBS containing 3% bovine serum albumin. The cells were then incubated with various antibodies diluted in the blocking buffers. Cells were washed and stained with fluorescent secondary antibodies, washed extensively, and mounted on slides. Images were collected with Zeiss 510 (Carl Zeiss, Germany) and Olympus FV500 (Olympus America Inc.) confocal microscopes. Quantitation of colocalization in single-plane images was determined using Adobe Photoshop CS2 software.
Transmission electron microscopy. Transmission electron microscopy was performed as previously described (37). For thin sectioning, transiently transfected HeLa cell pellets were fixed with 2.5% glutaraldehyde. After dehydration in a grade series of cold ethanol, the samples were embedded in Poly/Bed 812 resin (Pelco International, Redding, CA). Ultrathin sections (90 nm) were then stained with uranyl acetate. The stained sections were observed with a Philips CM 12 electron microscope.
Immunoelectron microscopy. Transfected HeLa cells were fixed by 4% paraformaldehyde in PBS for 20 min at room temperature. Immunoelectron microscopy was done as previously described (14). Briefly, fixed cells were permeabilized and blocked in PBS containing 3% bovine serum albumin and 0.1% saponin for 30 min. Cells were incubated with primary antibody in the blocking buffer followed by incubation with Nanogold-conjugated rabbit anti-mouse IgG. Cells were postfixed with 1.6% glutaraldehyde in PBS followed by incubation in a silver enhancement solution for 20 min. Cells were then postfixed with a 1% OsO4 solution for 30 min before being embedded, dehydrated, and sectioned. Thin sections were examined with a Philips CM 12 electron microscope.
Analysis of virus particle release. Methods for preparing cell and virus-like particle (VLP) lysates have been detailed previously (37). Briefly, 3 days after transfection, cells were lysed in radioimmunoprecipitation assay buffer and immunoprecipitated with anti-HTLV-1 p24 polyclonal antibody (Advanced BioScience Laboratories, Inc., Kensington, Maryland). Supernatant collected from transfected cells was subjected to ultracentrifugation at 40,000 x g for 1 h to obtain the VLP pellets. HTLV immunoblot analysis was done using an anti-HTLV-1 p19 monoclonal antibody as primary antibody and a horseradish peroxidase-conjugated anti-mouse Ig (Amersham, Arlington Heights, IL, and Roche Applied Science, Indianapolis, IN) as secondary antibody with the ECL Western analysis kit (Amersham). The efficiency of VLP production was normalized for cell-associated Gag. Real-time quantitation of band intensities was done using the Quantity One software package with the Chemi Doc XRS Documentation System (Bio-Rad, Richmond, Calif.).
RESULTS
PTAP motif plays a role in targeting Gag to the plasma membrane. We previously observed that the PTAP motif plays a less important role in 293T cells than the PPPY motif in virus release since mutation of PTAP led to a modest reduction in particle production (60% of wild-type level) and could be replaced by PPPY or YPDL without affecting efficiency of particle production (37). We recently reported that the Gag localization pattern for the Gag mutants in 293T cells was comparable to that of wt Gag (37). In HeLa cells, we found that the Gag PTAP mutants are localized more in the perinuclear region than along the cell periphery, as is the case with the Gag PPPY mutants and wt Gag (Fig. 1A and B). We observed that approximately three-fourths of the cells expressing the Gag PTAP mutants have a perinuclear staining pattern while approximately three-fourths of the wt Gag-expressing cells analyzed had a peripheral staining phenotype. These observations indicate that the HTLV-1 Gag PTAP mutants have a Gag localization pattern distinct from that of wt HTLV-1 Gag.
To test if the mutant Gag proteins might be associated with MVBs, transfected HeLa cells were labeled for the MVB marker CD63, a tetraspanin found in late endosomes and MVBs (particularly in the membranes of the internal vesicles). A high degree of colocalization (88% ± 1%) of mutant PPPY-PTRP and CD63 was observed (Fig. 1D and 1J) and to a lower degree with wt Gag (30% ± 7%) (Fig. 1C and 1J). The PPPY-PPPY mutant was found to be similar to PPPY-PTRP Gag. The mutant Gags were also observed to have strong colocalization with another MVB marker, LAMP-1 (data not shown). Less than 1% colocalization was observed between the Gag of PPPY-PTRP or PPPY-PPPY and an endoplasmic reticulum marker (data not shown); the Golgi marker GM130 (Fig. 1E and 1H); and an early endosome marker, EEA1 (Fig. 1F and 1I). These data indicate that Gag is associated with MVBs but not with the other organelles tested.
Localization of HTLV-1 particles by electron microscopy. To visualize HTLV-1 particles, we transfected the wt Gag-expressing construct pCMV-HT1 or the PPPY-PTRP derivative into HeLa cells for analysis by electron microscopy. Cells were fixed, and ultrathin sections were prepared for transmission electron microscopy 48 h posttransfection. For wt Gag, we observed large amounts of HTLV-1 particles budding at the plasma membrane, as we previously observed with 293T cells (37). For the PPPY-PTRP mutant, we found that the transfected cells had intracellular compartments containing what appeared to be HTLV-1 particles (Fig. 2A). Similar to the plasma membrane particles, the intracellular particles contained a spherical, electron-dense core and were of 100 to 110 nm in diameter, which corresponds to the size of authentic HTLV-1 particles (Fig. 2E). Virus particles were not readily observed in MVBs for wt Gag and for the APPY-PTAP mutant (data not shown), though this does not preclude the possibility that particles are released into the MVB.
The accumulation of HTLV-1 virus particles inside the lumen of MVBs for the PPPY-PTRP mutant could be explained by three possible mechanisms. First, the newly assembled virions could enter the lumen of MVBs by a budding process from the cytosol. Second, viral particles could be released by the cells on the plasma membrane and then reinternalized and targeted to these compartments via endocytosis, a phenomenon reported previously for HTLV-1 (30). Third, MVB-associated virus may have begun the assembly process at the plasma membrane but ended up in MVBs via membrane recycling. Further examination of ultrathin sections of preparations of transfected HeLa cells revealed immature ring-like particles inside the lumen and the budding profiles of immature particles (Fig. 2B to D). This observation supports the hypothesis that Gag protein assembled at the limiting membrane of MVBs.
Immunoelectron microscopic analysis was performed to further test the origin of these intracellular HTLV-1 particles. Labeling of particles with p19 found within MVB structures was observed (Fig. 2F). The particles appeared to also be labeled with CD63 (Fig. 2G), suggesting that these particles incorporated CD63 into their membrane as they assembled at the limiting membrane of MVBs.
Blocking endocytosis does not influence HTLV-1 Gag PTAP mutant distribution. The possibility of endocytosis was further excluded by examining the effect of overexpression of an endocytic mutant on the distribution of the PPPY-PTRP mutant Gag. The rationale for this is that, if the Gag ended up in the MVBs through the endocytic pathway, blocking of endocytosis will prevent the accumulation of Gag inside MVBs. Eps15 (epidermal growth factor receptor pathway substrate clone 15) is a constituent of plasma membrane clathrin-coated pits. Overexpression of a dominant-negative mutant of Eps15, Eps-DIII-GFP, but not another mutant, Eps-D32, was shown to inhibit the clathrin-dependent endocytosis by inhibiting the clathrin-coated pit assembly (3, 4). The inhibitory effect of Eps-DIII-GFP on endocytosis was first tested with transferrin. HeLa cells were transfected with the Eps-DIII-GFP or Eps-D32-GFP construct. Transfected cells were incubated with Alexa Fluor-547-labeled Tf, a small molecule that is normally taken up into cells through clathrin-mediated endocytosis. Cells were fixed and then analyzed by confocal microscopy. The distribution of Eps-DIII-GFP was mainly cytosolic (Fig. 3A) as reported previously (3). The punctate staining of Tf indicated that Tf molecules were taken up by cells through endocytosis and localized in intracellular vesicles in the cells that did not express Eps-DIII-GFP (Fig. 3A). However, the cells expressing Eps-DIII-GFP (Fig. 3A) dramatically reduced the uptake of Tf. In contrast, overexpression of Eps-D32-GFP did not inhibit the endocytosis of Tf into HeLa cells (Fig. 3A). The construct Eps-DIII-GFP was then used to test the effect of DIII overexpression on Gag localization. HeLa cells were cotransfected with either wt, PPPY-PTRP, or PPPY-PPPY and Eps-DIII-GFP. Cells were then stained for Gag and CD63 and analyzed by confocal microscopy. In the cell expressing DIII-GFP, wt Gag was observed to colocalize at low levels with CD63 as indicated by the merged light pink color (Fig. 3B). Strong colocalization was observed with the PPPY-PTRP and the PPPY-PPPY Gag mutants (Fig. 3C and D). The punctate staining of Gag was found to be less in these experiments when Gag was coexpressed with Eps-DIII-GFP construct. The overexpression of the Eps-D32-GFP construct with wt and mutant Gag led to results similar to that with overexpression with Eps-DIII-GFP with Gag (data not shown). Taken together, these observations support the interpretation that Gag did not reach the MVBs through endocytosis and that the presence of virus particles or virus buds inside MVBs is the result of targeting and assembly of Gag at the MVBs. Although our data would rule out virus uptake by clathrin-dependent endocytosis, it is formally plausible that a clathrin-independent endocytic pathway could be involved in virus uptake.
Influence of dynamitin p50 overexpression on Gag localization. We observed that virus particles for the PPPY-PTRP mutant accumulated inside MVBs. To determine whether dynamitin p50 could influence the localization of Gag, dynamitin p50 was overexpressed together with wt Gag or derivatives. The effect of dynamitin p50 on the distribution of Gag was analyzed by confocal microscopy. Overexpression of dynamitin p50 resulted in the disappearance of the punctate staining pattern for Gag in about three-fourths of the cells viewed for both wt and mutant Gag constructs, such as those in Fig. 4B and C. Overexpression of dynamitin p50 was also found to result in the disappearance of the asymmetric Gag perinuclear staining that is likely at the MTOC (Fig. 4A) with the PPPY-PTRP and PPPY-PPPY Gag mutants (Fig. 4C and D). The lack of Gag perinuclear staining was also seen with wt Gag (Fig. 4B). The colocalizations of wt Gag and PPPY-PTRP were comparable (84% ± 13% and 87% ± 13%, respectively) (Fig. 4E). The PPPY-PTRP and PPPY-PPPY Gag mutants were observed to still be associated with CD63-positive MVBs (data not shown). Taken together, these observations suggest that overexpression of dynamitin p50 leads to a peripheral distribution of both wt and mutant Gag proteins.
Influence of cell type and MVB movement on HTLV-1 particle release. In order to assess whether the redistribution of HTLV-1 Gag in cells in the various experiments described above would influence virus particle release from HeLa cells, we investigated particle production in the different experiments using an immunoprecipitation-immunoblot procedure with real-time signal acquisition. First, to help define the linear range of detection for the assay, a dilution series was made and tested (Fig. 5A). We found that the linear range of detection was at least 256-fold, based on the dilution series tested.
We next analyzed, in parallel, particle production of HTLV-1 particles with wt Gag or the PPPY-PTRP mutant. Interestingly, we observed that particle release of the PPPY-PTRP mutant from HeLa cells was about 20-fold lower than that observed with the wt Gag (Fig. 5B and C, lanes 1 and 2). This is in contrast with what we have observed with the same mutant in 293T cells, where particle production was reduced by less than twofold compared to wt. Also, the observation that the PTAP motif can play an important role in particle release is different from observations made by others (6, 20, 37). Our observations indicate that the PTAP motif plays an important role in HTLV-1 particle release and that this effect is cell type dependent. This conclusion is consistent with observations made with the HIV-1 PTAP motif (10).
Cotransfection with the dynamitin p50 was found to not significantly influence virus release for the wt but did increase (10-fold) virus release of the PPPY-PTRP mutant (Fig. 5B and C, lanes 1 to 4). The expression of dynamitin p50 was not altered by coexpression of the different viral constructs (Fig. 5B). Based on the ability of dynamitin p50 to move MVBs away from the expected position of the MTOC, this confirms the prediction that overexpression of this construct could increase virus particle release, particularly with the PPPY-PTRP mutant.
DISCUSSION
The Gag protein for most retroviruses is thought to be targeted to the plasma membrane, where particle assembly and release occur. In this study, we have observed that HTLV-1 Gag can colocalize with an MVB marker. Particles or particle buds were also observed to be associated with MVBs by electron microscopy, implying that Gag targeting to the MVB resulted in particle budding. Three possible mechanisms could explain these observations. First, HTLV-1 particles could be secreted by cells and then be reinternalized and targeted to MVBs via the endocytic pathway, a phenomenon that has been reported previously (30). Second, HTLV-1 Gag could be targeted to MVBs and assembled into virus particles at the limiting membrane in a manner similar to the biogenesis of the internal vesicles of MVBs. Third, MVB-associated virus may have begun the assembly process at the plasma membrane but ended up in MVBs via membrane recycling. Several lines of evidence argue in favor of the second possibility. First, we have observed immature ring-like particles inside the lumen of MVB and the electron-dense Gag aggregates that are in the process of assembly or budding along with mature particles. These observations suggest that Gag was targeted to MVBs and assembled into virions. Second, the intracellular Gag showed no obvious colocalization with an early endosomal marker, EEA1, suggesting that Gag proteins are not transported to the MVB from early or recycling endosomes, which would be the case if the virus particles were internalized through endocytosis. Third, immunogold labeling of preembedded cells showed HTLV-1 particles inside MVBs decorated with CD63, a protein that is normally found in MVB membranes. These findings suggest that HTLV-1 particles budded into the MVB, incorporating CD63 into their membrane envelope as they assembled at the limiting membrane. It should be noted that CD63 can also be found at the plasma membrane, and virion incorporation of CD63 does not exclude the possibility that CD63-containing particles were released from plasma membrane.
We found that the Gag mutant PPPY-PTRP was primarily directed to MVBs. This observation supports the hypothesis that the PTAP motif is involved in targeting Gag to the plasma membrane. This hypothesis is further supported by the observation of MVB localization of the PPPY-PPPY (Fig. 1F) or PPPY-YPDL (data not shown) mutant. In contrast, the PPPY motif does not appear to influence Gag targeting in HeLa cells since Gag of mutant APPY-PTAP was targeted primarily to plasma membrane. The role of the PTAP motif in HTLV-1 Gag targeting involves interaction with TSG101, since it was reported that overexpression of the N-terminal fragment of TSG101 redirected the particle assembly inside cytoplasmic vacuoles (6). Consistent with our findings, the role of the PTAP-TSG101 interaction in recruiting viral protein to the plasma membrane for assembly has been suggested for Ebola virus (35). Interestingly, both HTLV-1 and Ebola virus contain both the PPPY and PTAP motifs. Another interesting observation relevant to our observations is that the deletion of the HIV-1 p6 protein of Gag (which contains the PTAP motif) does not affect HIV-1 Gag targeting to the MVB in HeLa cells and in macrophages (24). This implies that MVB targeting is not driven by an interaction between p6 and the MVB sorting machinery (e.g., TSG101).
Our study suggests that HTLV-1 particles in MVBs can be released from cells. Such an exocytic pathway has been reported in a variety of hematopoietic cell types such as B lymphocytes and dendritic cells (19, 29). In these cells, exocytosis results in the secretion of the intralumenal membrane vesicles, called exosomes. The exocytic process is best documented for the MIICs. These compartments identified as late endosomes/MVBs have been previously shown to serve as transport organelles, which guide MHC class II to the cell surface (38, 39). Microtubules are involved in the intracellular transport of MIIC. Two molecular motor proteins, kinesin and dynein, are in control of the movement of MIIC away or toward the MTOC region (38). Our observations in this study support the conclusion that virus-containing MVBs can traffic to the cell surface and that virus release likely occurs by MVB fusion with the plasma membrane. Recent studies have suggested that AP-3 is involved in targeting HIV-1 Gag to the MVB (11). It is not currently known what role AP-3 plays in HTLV-1 Gag targeting to the MVB.
We discovered in this study that the PTAP motif can strongly influence HTLV-1 particle release. Specifically, we found that an HTLV-1 Gag mutant with an intact PPPY motif but a mutated PTAP motif (i.e., PPPY-PTRP) led to a dramatic 20-fold reduction in virus particle release. This is in striking contrast to the same mutant in 293T cells, which we previously reported to reduce particle release by less than twofold as well as not influence the distribution of Gag in cells (37). Our findings indicate that (i) PTAP function is cell type dependent and (ii) PTAP plays a role (direct or indirect) in Gag targeting to the plasma membrane. We are currently investigating why particle reduction for this mutant in HeLa cells is greatly reduced. Ono and Freed (24) observed that MVB localization of HIV-1 Gag was still observed in the absence of p6 (PTAP is located in p6), suggesting that an interaction between Gag and TSG101 is not required for Gag targeting to the MVB. This observation is consistent with our findings with HTLV-1 Gag. The observations with HIV-1 were used to hypothesize that Gag targeting involves two different cellular binding partners, one on the plasma membrane and the other in the MVB. Our current studies are directed at testing this hypothesis as well as elucidating the mechanistic basis for the cell type dependence of PTAP function.
ACKNOWLEDGMENTS
We thank K. Wolken of the Campus Microscopy and Imaging Facility at Ohio State University for technical assistance and E. Freed and A. Ono for helpful discussion.
This work was supported by NIH grants AI053155 and AI057735.
These authors contributed equally to this work.
REFERENCES
Babst, M., D. J. Katzmann, E. J. Estepa-Sabal, T. Meerloo, and S. D. Emr. 2002. Escrt-III: an endosome-associated heterooligomeric protein complex required for MVB sorting. Dev. Cell 3:271-282.
Babst, M., D. J. Katzmann, W. B. Snyder, B. Wendland, and S. D. Emr. 2002. Endosome-associated complex, ESCRT-II, recruits transport machinery for protein sorting at the multivesicular body. Dev. Cell 3:283-289.
Benmerah, A., M. Bayrou, N. Cerf-Bensussan, and A. Dautry-Varsat. 1999. Inhibition of clathrin-coated pit assembly by an Eps15 mutant. J. Cell Sci. 112:1303-1311.
Benmerah, A., C. Lamaze, B. Begue, S. L. Schmid, A. Dautry-Varsat, and N. Cerf-Bensussan. 1998. AP-2/Eps15 interaction is required for receptor-mediated endocytosis. J. Cell Biol. 140:1055-1062.
Blot, V., F. Perugi, B. Gay, M. C. Prevost, L. Briant, F. Tangy, H. Abriel, O. Staub, M. C. Dokhelar, and C. Pique. 2004. Nedd4.1-mediated ubiquitination and subsequent recruitment of Tsg101 ensure HTLV-1 Gag trafficking towards the multivesicular body pathway prior to virus budding. J. Cell Sci. 117:2357-2367.
Bouamr, F., J. A. Melillo, M. Q. Wang, K. Nagashima, M. de Los Santos, A. Rein, and S. P. Goff. 2003. PPPYEPTAP motif is the late domain of human T-cell leukemia virus type 1 Gag and mediates its functional interaction with cellular proteins Nedd4 and Tsg101. J. Virol. 77:11882-11895.
Bryant, M., and L. Ratner. 1990. Myristoylation-dependent replication and assembly of human immunodeficiency virus 1. Proc. Natl. Acad. Sci. USA 87:523-527.
Copeland, K. F. T., A. G. M. Haaksma, D. Derse, and J. L. Heeney. 1994. Detection of human T-cell leukemia virus 1 permissive cells using cell lines producing selectable recombinant virions. J. Virol. Methods 50:219-226.
Demirov, D. G., and E. O. Freed. 2004. Retrovirus budding. Virus Res. 106:87-102.
Demirov, D. G., J. M. Orenstein, and E. O. Freed. 2002. The late domain of human immunodeficiency virus type 1 p6 promotes virus release in a cell type-dependent manner. J. Virol. 76:105-117.
Dong, X., H. Li, A. Derdowski, L. Ding, A. Burnett, X. Chen, T. R. Peters, T. S. Dermody, E. Woodruff, J. J. Wang, and P. Spearman. 2005. AP-3 directs the intracellular trafficking of HIV-1 Gag and plays a key role in particle assembly. Cell 120:663-674.
Freed, E. O. 1998. HIV-1 Gag proteins: diverse functions in the virus life cycle. Virology 251:1-15.
Freed, E. O. 2002. Viral late domains. J. Virol. 76:4679-4687.
Gilerovitch, H. G., G. A. Bishop, J. S. King, and R. W. Burry. 1995. The use of electron microscopic immunocytochemistry with silver-enhanced 1.4-nm gold particles to localize GAD in the cerebellar nuclei. J. Histochem. Cytochem. 43:337-343.
Gottlinger, H. G. 2001. The HIV-1 assembly machine. AIDS 15(Suppl. 5):S13-S20.
Heidecker, G., P. A. Lloyd, K. Fox, K. Nagashima, and D. Derse. 2004. Late assembly motifs of human T-cell leukemia virus type 1 and their relative roles in particle release. J. Virol. 78:6636-6648.
Hill, C. P., D. Worthylake, D. P. Bancroft, A. M. Christensen, and W. I. Sundquist. 1996. Crystal structures of the trimeric human immunodeficiency virus type 1 matrix protein: implications for membrane association and assembly. Proc. Natl. Acad. Sci. USA 93:3099-3104.
Hirokawa, N. 1998. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science 279:519-526.
Kolodziej, D., and E. Pajtasz-Piasecka. 2003. Role of dendritic cells in recognizing antigens: their binding, transformation and presentation to T lymphocytes. Postepy Hig. Med. Dosw. 57:149-170. (In Polish.)
Le Blanc, I., M. C. Prevost, M. C. Dokhelar, and A. R. Rosenberg. 2002. The PPPY motif of human T-cell leukemia virus type 1 Gag protein is required early in the budding process. J. Virol. 76:10024-10029.
Martin-Serrano, J., T. Zang, and P. D. Bieniasz. 2003. Role of ESCRT-I in retroviral budding. J. Virol. 77:4794-4804.
Massiah, M. A., M. R. Starich, C. Paschall, M. F. Summers, A. M. Christensen, and W. I. Sundquist. 1994. Three-dimensional structure of the human immunodeficiency virus type 1 matrix protein. J. Mol. Biol. 244:198-223.
Nydegger, S., M. Foti, A. Derdowski, P. Spearman, and M. Thali. 2003. HIV-1 egress is gated through late endosomal membranes. Traffic 4:902-910.
Ono, A., and E. O. Freed. 2004. Cell-type-dependent targeting of human immunodeficiency virus type 1 assembly to the plasma membrane and the multivesicular body. J. Virol. 78:1552-1563.
Orenstein, J. M., M. S. Meltzer, T. Phipps, and H. E. Gendelman. 1988. Cytoplasmic assembly and accumulation of human immunodeficiency virus types 1 and 2 in recombinant human colony-stimulating factor-1-treated human monocytes: an ultrastructural study. J. Virol. 62:2578-2586.
Parent, L. J., C. B. Wilson, M. D. Resh, and J. W. Wills. 1996. Evidence for a second function of the MA sequence in the Rous sarcoma virus Gag protein. J. Virol. 70:1016-1026.
Pelchen-Matthews, A., B. Kramer, and M. Marsh. 2003. Infectious HIV-1 assembles in late endosomes in primary macrophages. J. Cell Biol. 162:443-455.
Raposo, G., M. Moore, D. Innes, R. Leijendekker, A. Leigh-Brown, P. Benaroch, and H. Geuze. 2002. Human macrophages accumulate HIV-1 particles in MHC II compartments. Traffic 3:718-729.
Raposo, G., H. W. Nijman, W. Stoorvogel, R. Liejendekker, C. V. Harding, C. J. Melief, and H. J. Geuze. 1996. B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183:1161-1172.
Romero, I. A., M. C. Prevost, E. Perret, P. Adamson, J. Greenwood, P. O. Couraud, and S. Ozden. 2000. Interactions between brain endothelial cells and human T-cell leukemia virus type 1-infected lymphocytes: mechanisms of viral entry into the central nervous system. J. Virol. 74:6021-6030.
Sakurai, A., J. Yasuda, H. Takano, Y. Tanaka, M. Hatakeyama, and H. Shida. 2004. Regulation of human T-cell leukemia virus type 1 (HTLV-1) budding by ubiquitin ligase Nedd4. Microbes Infect. 6:150-156.
Schultz, A. M., and A. Rein. 1989. Unmyristylated Moloney murine leukemia virus Pr65gag is excluded from virus assembly and maturation events. J. Virol. 63:2370-2373.
Sherer, N. M., M. J. Lehmann, L. F. Jimenez-Soto, A. Ingmundson, S. M. Horner, G. Cicchetti, P. G. Allen, M. Pypaert, J. M. Cunningham, and W. Mothes. 2003. Visualization of retroviral replication in living cells reveals budding into multivesicular bodies. Traffic 4:785-801.
Swanstrom, R., and J. W. Wills. 1997. Synthesis, assembly, and processing of viral proteins, p. 263-334. In J. M. Coffin, S. H. Hughes, and H. E. Varmus (ed.), Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Timmins, J., G. Schoehn, S. Ricard-Blum, S. Scianimanico, T. Vernet, R. W. Ruigrok, and W. Weissenhorn. 2003. Ebola virus matrix protein VP40 interaction with human cellular factors Tsg101 and Nedd4. J. Mol. Biol. 326:493-502.
von Schwedler, U. K., M. Stuchell, B. Muller, D. M. Ward, H. Y. Chung, E. Morita, H. E. Wang, T. Davis, G. P. He, D. M. Cimbora, A. Scott, H. G. Krausslich, J. Kaplan, S. G. Morham, and W. I. Sundquist. 2003. The protein network of HIV budding. Cell 114:701-713.
Wang, H., N. J. Machesky, and L. M. Mansky. 2004. Both the PPPY and PTAP motifs are involved in human T-cell leukemia virus type 1 particle release. J. Virol. 78:1503-1512.
Wubbolts, R., M. Fernandez-Borja, I. Jordens, E. Reits, S. Dusseljee, C. Echeverri, R. B. Vallee, and J. Neefjes. 1999. Opposing motor activities of dynein and kinesin determine retention and transport of MHC class II-containing compartments. J. Cell Sci. 112:785-795.
Wubbolts, R., M. Fernandez-Borja, L. Oomen, D. Verwoerd, H. Janssen, J. Calafat, A. Tulp, S. Dusseljee, and J. Neefjes. 1996. Direct vesicular transport of MHC class II molecules from lysosomal structures to the cell surface. J. Cell Biol. 135:611-622.
Yuan, X., X. Yu, T. H. Lee, and M. Essex. 1993. Mutations in the N-terminal region of human immunodeficiency virus type 1 matrix protein block intracellular transport of the Gag precursor. J. Virol. 67:6387-6394.
Zhou, W., L. J. Parent, J. W. Wills, and M. D. Resh. 1994. Identification of a membrane-binding domain within the amino-terminal region of human immunodeficiency virus type 1 Gag protein which interacts with acidic phospholipids. J. Virol. 68:2556-2569.(Irene J. Dorweiler, Susan)
Departments of Diagnostic and Biological Sciences and Microbiology
Cancer Center, University of Minnesota, Minneapolis, Minnesota 55455
Comprehensive Cancer Center
Department of Neuroscience, Ohio State University, Columbus, Ohio 43210
ABSTRACT
Human T-cell leukemia virus type 1 (HTLV-1) Gag is targeted to the plasma membrane for particle assembly and release. How HTLV-1 Gag targeting occurs is not well understood. The PPPY and PTAP motifs were previously shown to be involved in HTLV-1 particle release with PTAP playing a more subtle role in virus budding. These L domains function through the interaction with host cellular proteins normally involved in multivesicular body (MVB) morphogenesis. The plasma membrane pathway rather than the MVB pathway was found to be the primary pathway for HTLV-1 particle release in HeLa cells. Intriguingly, disruption of the PTAP motif led to a defect in the targeting of Gag from the plasma membrane to CD63-positive MVBs. Particles or particle buds were observed to be associated with MVBs by electron microscopy, implying that Gag targeting to the MVB resulted in particle budding. Blocking clathrin-dependent endocytosis was found not to influence localization of the HTLV-1 Gag PTAP mutant, indicating that Gag did not reach the MVBs through clathrin-dependent endocytosis. Our observations imply that the interaction between Gag and TSG101 is not required for Gag targeting to the MVB. Overexpression of dynamitin p50 increased particle release, suggesting that there was an increase in the intracellular transport of MVBs to the cell periphery by the utilization of the dynein-dynactin motor complex. Intriguingly, virus particle release with this mutant was reduced by 20-fold compared to that of wild type in HeLa cells, which is in marked contrast to the less-than-twofold defect observed for particle production of the HTLV-1 Gag PTAP mutant from 293T cells. These results indicate that the role of the PTAP motif in L domain function is cell type dependent.
INTRODUCTION
Many retroviruses initiate particle assembly and budding at the plasma membrane where Gag proteins associate with raft microdomains of the inner leaflet of the plasma membrane (15, 34). The Gag polyprotein is the main driving force for assembly and budding. The targeting of Gag proteins to the cell membrane and stable membrane association are accomplished using a bipartite membrane binding signal consisting of the fatty acid myristate (7, 32) added cotranslationally to the N terminus of Gag in concert with a patch of basic residues within the matrix (MA) domain) (12, 17, 22, 40, 41). Myristate provides a hydrophobic interaction with the lipid membrane, while the basic residues strengthen the interaction by forming the electrostatic interactions with acidic phospholipids that are enriched at the cytoplasmic face of the plasma membrane (26, 41).
The release of the nascent virions from the cell is mediated by the late (L) domain (9, 13). The PTAP-interacting protein TSG101 functions by recruiting the intracellular machinery, ESCRT complexes, to the site of assembly at the plasma membrane (13, 15, 21, 36). These ESCRT complexes are normally involved in multivesicular body (MVB) biogenesis (1, 2). The formation of MVBs, a late endosomal compartment, is characterized by the invagination of limiting membranes to bud small vesicles. Topologically, both virus budding at the plasma membrane and the vesicle budding into MVBs share striking similarities: they occur away from the cytosol, and they involve lipid microdomains, inositol derivatives, and a particular cohort of proteins such as TSG101 and ubiquitin (9, 13).
Previous studies have investigated the role of the human T-cell leukemia virus type 1 (HTLV-1) PPPY and PTAP motifs in virus budding at the plasma membrane (5, 6, 16, 20, 31, 37). Both motifs have been implicated in budding of HTLV-1 from the plasma membrane (6, 37). The PPPY and PTAP motifs function through the interaction with the Nedd4 family of proteins (E3 ubiquitin ligases) and the TSG101 protein (an E2-like ubiquitin-conjugating enzyme), respectively (5, 6, 16). The PPPY motif was observed to be more important than PTAP in particle release (6, 16, 37). However, it remains unclear what different roles these two motifs play in HTLV-1 assembly and release. It has been suggested that PPPY and PTAP act successively in the HTLV-1 assembly process to ensure proper Gag trafficking through the endocytic pathway to late endosomes for particle release (5).
Despite the view that many retroviruses recruit the cellular machinery normally responsible for MVB formation to the plasma membrane for their assembly and budding, it has been long known that in macrophages human immunodeficiency virus type 1 (HIV-1) particles accumulate in an intracellular, vacuole-like compartment (25). A number of recent reports have identified the vacuole-like compartment as a CD63-positive MVB, indicating that HIV-1 particles may be released into the MVB (23, 27, 28). Such observations were further extended to standard tissue culture cell lines previously believed to support budding only at the plasma membrane (23, 24, 33). The intracellularly assembled HIV-1 particles are likely released into the extracellular environment through the fusion of virus-containing compartments directly with the cell surface in an exocytic fashion (23, 24, 27, 28, 33). Such exosome-assisted release of proteins is best documented for the major histocompatibility complex class II (MHC class II) (29, 39). It is currently unclear how retroviral Gag is targeted to MVBs.
Many intracellular vesicles are transported along polarized cytoskeletal elements (i.e., microtubules) by utilizing molecular motor proteins (18). Kinesin moves towards the fast-growing end (or plus end), away from the microtubule organizing center (MTOC), and cytoplasmic dynein moves toward the minus end. In the case of dynein, vesicle binding is thought to be mediated by the interaction of the dynein intermediate chains with the 150-kDa dynactin subunit Glued. Overexpression of the 50-kDa "dynamitin" subunit of the dynactin complex can dissociate the complex (18). Under these conditions dynein is released from its cargo sites, resulting in inhibition of minus-end-directed vesicle transport. The dynein-dynactin motor complex has been reported to be involved in the intracellular transport of the MHC class II-containing compartment (MIIC), and these MIICs can be released as exosomes (29, 38, 39).
To date, the use of the MVB pathway for budding has mainly been reported for HIV release from macrophages. In the present study, the plasma membrane pathway rather than the MVB pathway was found to be the primary pathway for HTLV-1 particle release in HeLa cells. Intriguingly, when PTAP is mutated, Gag strongly colocalizes with CD63, an MVB marker. Particles and particle buds were observed with intracellular vacuoles, suggesting that Gag targeting to the MVB resulted in particle budding. When clathrin-dependent endocytosis was blocked, localization of the HTLV-1 Gag PTAP mutant was not altered, indicating that Gag did not reach the MVBs through clathrin-dependent endocytosis. These findings imply that the interaction between Gag and TSG101 is not required for Gag targeting to the MVB. Overexpression of dynamitin p50 increased particle release, suggesting that there was an increase in the intracellular transport of MVBs to the cell periphery by the utilization of the dynein-dynactin motor complex. Virus particle release with this mutant was reduced by 20-fold compared to that of wild type (wt) in HeLa cells, which is in marked contrast to the less-than-twofold defect observed for particle production of the HTLV-1 Gag PTAP mutant from 293T cells. These results indicate that the role of the PTAP motif in L domain function is cell type dependent.
MATERIALS AND METHODS
Cell culture and transfection. HeLa cells were maintained in Dulbecco's modified Eagle's medium (GIBCO BRL, Gaithersburg, MD) supplemented with 10% Fetal Clone III serum (HyClone, Logan, UT). Transfection of cells was performed by the calcium phosphate precipitation method. Typically, 100-mm petri dishes were used in this study unless otherwise noted. A total of 10 μg of purified plasmid DNA was mixed with 50 μl 2.5 M CaCl2 and 500 μl 2x HEPES-buffered saline and then added to cells that were split the day before transfection and at 50 to 70% confluence. At 12 to 15 h after transfection, medium was replaced with medium with 10 mM sodium butyrate and 20 mM HEPES. After 8 h of incubation, medium was replaced with fresh Dulbecco's modified Eagle's medium with 20 mM HEPES.
Plasmids. The plasmid pCMV-HT1 was derived from an infectious molecular clone of HTLV-1 by replacing the 5' long terminal repeat (LTR) with a cytomegalovirus immediate-early promoter joined to a small fragment from the R region of the LTR that contains the major splice donor site (8). Eps-DIII-GFP (green fluorescent protein) and Eps-D32-GFP were kindly provided by Alexandre Benmerah (Developpement Normal et Pathologique de Systeme Immunitaire, Paris, France). The dynamitin p50 expression construct was kindly provided by Jacques Neefjes (Netherlands Cancer Institute, The Netherlands).
Antibodies. The following antibodies were used: mouse monoclonal anti-HTLV-1 p19 (Zeptometrix, Buffalo, NY), goat polyclonal anti-HTLV-1 p24 (Advanced Biotechnologies, Columbia, Maryland, and Advanced BioScience Laboratories, Inc., Kensington, Maryland), mouse monoclonal anti-CD63 (BD Biosciences), mouse monoclonal anti-LAMP-1 (Southern Biotechnology Associates, Inc., Birmingham, AL), mouse monoclonal anti-EEA1 (BD Biosciences), mouse monoclonal anti-GM130 (Transduction Laboratories), mouse monoclonal anti--tubulin (Sigma, Saint Louis, Missouri), and mouse monoclonal antidynactin p50 (BD Biosciences). Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G (IgG), Alexa Fluor 568-conjugated rabbit anti-goat IgG, and Alexa Fluor 647-conjugated goat anti-mouse IgG were purchased from Molecular Probes (Molecular Probes, Eugene, OR). Nanogold-conjugated rabbit anti-mouse IgG (NanoProbes Inc., Yaphank, NY) was used for electron microscopy immunolabeling.
Tf uptake. The transferrin (Tf) uptake assay was performed using Alexa Fluor 546-labeled transferrin (Molecular Probes, Eugene, OR). HeLa cells were incubated with 0.1 μM Alexa Fluor 546-labeled transferrin at 37°C for 15 min for binding and then washed and fixed for immunofluorescence analysis.
Confocal microscopy. Cells were grown on coverslips and fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 in phosphate-buffered saline (PBS), quenched in 100 mM glycine-PBS, and blocked in PBS containing 3% bovine serum albumin. The cells were then incubated with various antibodies diluted in the blocking buffers. Cells were washed and stained with fluorescent secondary antibodies, washed extensively, and mounted on slides. Images were collected with Zeiss 510 (Carl Zeiss, Germany) and Olympus FV500 (Olympus America Inc.) confocal microscopes. Quantitation of colocalization in single-plane images was determined using Adobe Photoshop CS2 software.
Transmission electron microscopy. Transmission electron microscopy was performed as previously described (37). For thin sectioning, transiently transfected HeLa cell pellets were fixed with 2.5% glutaraldehyde. After dehydration in a grade series of cold ethanol, the samples were embedded in Poly/Bed 812 resin (Pelco International, Redding, CA). Ultrathin sections (90 nm) were then stained with uranyl acetate. The stained sections were observed with a Philips CM 12 electron microscope.
Immunoelectron microscopy. Transfected HeLa cells were fixed by 4% paraformaldehyde in PBS for 20 min at room temperature. Immunoelectron microscopy was done as previously described (14). Briefly, fixed cells were permeabilized and blocked in PBS containing 3% bovine serum albumin and 0.1% saponin for 30 min. Cells were incubated with primary antibody in the blocking buffer followed by incubation with Nanogold-conjugated rabbit anti-mouse IgG. Cells were postfixed with 1.6% glutaraldehyde in PBS followed by incubation in a silver enhancement solution for 20 min. Cells were then postfixed with a 1% OsO4 solution for 30 min before being embedded, dehydrated, and sectioned. Thin sections were examined with a Philips CM 12 electron microscope.
Analysis of virus particle release. Methods for preparing cell and virus-like particle (VLP) lysates have been detailed previously (37). Briefly, 3 days after transfection, cells were lysed in radioimmunoprecipitation assay buffer and immunoprecipitated with anti-HTLV-1 p24 polyclonal antibody (Advanced BioScience Laboratories, Inc., Kensington, Maryland). Supernatant collected from transfected cells was subjected to ultracentrifugation at 40,000 x g for 1 h to obtain the VLP pellets. HTLV immunoblot analysis was done using an anti-HTLV-1 p19 monoclonal antibody as primary antibody and a horseradish peroxidase-conjugated anti-mouse Ig (Amersham, Arlington Heights, IL, and Roche Applied Science, Indianapolis, IN) as secondary antibody with the ECL Western analysis kit (Amersham). The efficiency of VLP production was normalized for cell-associated Gag. Real-time quantitation of band intensities was done using the Quantity One software package with the Chemi Doc XRS Documentation System (Bio-Rad, Richmond, Calif.).
RESULTS
PTAP motif plays a role in targeting Gag to the plasma membrane. We previously observed that the PTAP motif plays a less important role in 293T cells than the PPPY motif in virus release since mutation of PTAP led to a modest reduction in particle production (60% of wild-type level) and could be replaced by PPPY or YPDL without affecting efficiency of particle production (37). We recently reported that the Gag localization pattern for the Gag mutants in 293T cells was comparable to that of wt Gag (37). In HeLa cells, we found that the Gag PTAP mutants are localized more in the perinuclear region than along the cell periphery, as is the case with the Gag PPPY mutants and wt Gag (Fig. 1A and B). We observed that approximately three-fourths of the cells expressing the Gag PTAP mutants have a perinuclear staining pattern while approximately three-fourths of the wt Gag-expressing cells analyzed had a peripheral staining phenotype. These observations indicate that the HTLV-1 Gag PTAP mutants have a Gag localization pattern distinct from that of wt HTLV-1 Gag.
To test if the mutant Gag proteins might be associated with MVBs, transfected HeLa cells were labeled for the MVB marker CD63, a tetraspanin found in late endosomes and MVBs (particularly in the membranes of the internal vesicles). A high degree of colocalization (88% ± 1%) of mutant PPPY-PTRP and CD63 was observed (Fig. 1D and 1J) and to a lower degree with wt Gag (30% ± 7%) (Fig. 1C and 1J). The PPPY-PPPY mutant was found to be similar to PPPY-PTRP Gag. The mutant Gags were also observed to have strong colocalization with another MVB marker, LAMP-1 (data not shown). Less than 1% colocalization was observed between the Gag of PPPY-PTRP or PPPY-PPPY and an endoplasmic reticulum marker (data not shown); the Golgi marker GM130 (Fig. 1E and 1H); and an early endosome marker, EEA1 (Fig. 1F and 1I). These data indicate that Gag is associated with MVBs but not with the other organelles tested.
Localization of HTLV-1 particles by electron microscopy. To visualize HTLV-1 particles, we transfected the wt Gag-expressing construct pCMV-HT1 or the PPPY-PTRP derivative into HeLa cells for analysis by electron microscopy. Cells were fixed, and ultrathin sections were prepared for transmission electron microscopy 48 h posttransfection. For wt Gag, we observed large amounts of HTLV-1 particles budding at the plasma membrane, as we previously observed with 293T cells (37). For the PPPY-PTRP mutant, we found that the transfected cells had intracellular compartments containing what appeared to be HTLV-1 particles (Fig. 2A). Similar to the plasma membrane particles, the intracellular particles contained a spherical, electron-dense core and were of 100 to 110 nm in diameter, which corresponds to the size of authentic HTLV-1 particles (Fig. 2E). Virus particles were not readily observed in MVBs for wt Gag and for the APPY-PTAP mutant (data not shown), though this does not preclude the possibility that particles are released into the MVB.
The accumulation of HTLV-1 virus particles inside the lumen of MVBs for the PPPY-PTRP mutant could be explained by three possible mechanisms. First, the newly assembled virions could enter the lumen of MVBs by a budding process from the cytosol. Second, viral particles could be released by the cells on the plasma membrane and then reinternalized and targeted to these compartments via endocytosis, a phenomenon reported previously for HTLV-1 (30). Third, MVB-associated virus may have begun the assembly process at the plasma membrane but ended up in MVBs via membrane recycling. Further examination of ultrathin sections of preparations of transfected HeLa cells revealed immature ring-like particles inside the lumen and the budding profiles of immature particles (Fig. 2B to D). This observation supports the hypothesis that Gag protein assembled at the limiting membrane of MVBs.
Immunoelectron microscopic analysis was performed to further test the origin of these intracellular HTLV-1 particles. Labeling of particles with p19 found within MVB structures was observed (Fig. 2F). The particles appeared to also be labeled with CD63 (Fig. 2G), suggesting that these particles incorporated CD63 into their membrane as they assembled at the limiting membrane of MVBs.
Blocking endocytosis does not influence HTLV-1 Gag PTAP mutant distribution. The possibility of endocytosis was further excluded by examining the effect of overexpression of an endocytic mutant on the distribution of the PPPY-PTRP mutant Gag. The rationale for this is that, if the Gag ended up in the MVBs through the endocytic pathway, blocking of endocytosis will prevent the accumulation of Gag inside MVBs. Eps15 (epidermal growth factor receptor pathway substrate clone 15) is a constituent of plasma membrane clathrin-coated pits. Overexpression of a dominant-negative mutant of Eps15, Eps-DIII-GFP, but not another mutant, Eps-D32, was shown to inhibit the clathrin-dependent endocytosis by inhibiting the clathrin-coated pit assembly (3, 4). The inhibitory effect of Eps-DIII-GFP on endocytosis was first tested with transferrin. HeLa cells were transfected with the Eps-DIII-GFP or Eps-D32-GFP construct. Transfected cells were incubated with Alexa Fluor-547-labeled Tf, a small molecule that is normally taken up into cells through clathrin-mediated endocytosis. Cells were fixed and then analyzed by confocal microscopy. The distribution of Eps-DIII-GFP was mainly cytosolic (Fig. 3A) as reported previously (3). The punctate staining of Tf indicated that Tf molecules were taken up by cells through endocytosis and localized in intracellular vesicles in the cells that did not express Eps-DIII-GFP (Fig. 3A). However, the cells expressing Eps-DIII-GFP (Fig. 3A) dramatically reduced the uptake of Tf. In contrast, overexpression of Eps-D32-GFP did not inhibit the endocytosis of Tf into HeLa cells (Fig. 3A). The construct Eps-DIII-GFP was then used to test the effect of DIII overexpression on Gag localization. HeLa cells were cotransfected with either wt, PPPY-PTRP, or PPPY-PPPY and Eps-DIII-GFP. Cells were then stained for Gag and CD63 and analyzed by confocal microscopy. In the cell expressing DIII-GFP, wt Gag was observed to colocalize at low levels with CD63 as indicated by the merged light pink color (Fig. 3B). Strong colocalization was observed with the PPPY-PTRP and the PPPY-PPPY Gag mutants (Fig. 3C and D). The punctate staining of Gag was found to be less in these experiments when Gag was coexpressed with Eps-DIII-GFP construct. The overexpression of the Eps-D32-GFP construct with wt and mutant Gag led to results similar to that with overexpression with Eps-DIII-GFP with Gag (data not shown). Taken together, these observations support the interpretation that Gag did not reach the MVBs through endocytosis and that the presence of virus particles or virus buds inside MVBs is the result of targeting and assembly of Gag at the MVBs. Although our data would rule out virus uptake by clathrin-dependent endocytosis, it is formally plausible that a clathrin-independent endocytic pathway could be involved in virus uptake.
Influence of dynamitin p50 overexpression on Gag localization. We observed that virus particles for the PPPY-PTRP mutant accumulated inside MVBs. To determine whether dynamitin p50 could influence the localization of Gag, dynamitin p50 was overexpressed together with wt Gag or derivatives. The effect of dynamitin p50 on the distribution of Gag was analyzed by confocal microscopy. Overexpression of dynamitin p50 resulted in the disappearance of the punctate staining pattern for Gag in about three-fourths of the cells viewed for both wt and mutant Gag constructs, such as those in Fig. 4B and C. Overexpression of dynamitin p50 was also found to result in the disappearance of the asymmetric Gag perinuclear staining that is likely at the MTOC (Fig. 4A) with the PPPY-PTRP and PPPY-PPPY Gag mutants (Fig. 4C and D). The lack of Gag perinuclear staining was also seen with wt Gag (Fig. 4B). The colocalizations of wt Gag and PPPY-PTRP were comparable (84% ± 13% and 87% ± 13%, respectively) (Fig. 4E). The PPPY-PTRP and PPPY-PPPY Gag mutants were observed to still be associated with CD63-positive MVBs (data not shown). Taken together, these observations suggest that overexpression of dynamitin p50 leads to a peripheral distribution of both wt and mutant Gag proteins.
Influence of cell type and MVB movement on HTLV-1 particle release. In order to assess whether the redistribution of HTLV-1 Gag in cells in the various experiments described above would influence virus particle release from HeLa cells, we investigated particle production in the different experiments using an immunoprecipitation-immunoblot procedure with real-time signal acquisition. First, to help define the linear range of detection for the assay, a dilution series was made and tested (Fig. 5A). We found that the linear range of detection was at least 256-fold, based on the dilution series tested.
We next analyzed, in parallel, particle production of HTLV-1 particles with wt Gag or the PPPY-PTRP mutant. Interestingly, we observed that particle release of the PPPY-PTRP mutant from HeLa cells was about 20-fold lower than that observed with the wt Gag (Fig. 5B and C, lanes 1 and 2). This is in contrast with what we have observed with the same mutant in 293T cells, where particle production was reduced by less than twofold compared to wt. Also, the observation that the PTAP motif can play an important role in particle release is different from observations made by others (6, 20, 37). Our observations indicate that the PTAP motif plays an important role in HTLV-1 particle release and that this effect is cell type dependent. This conclusion is consistent with observations made with the HIV-1 PTAP motif (10).
Cotransfection with the dynamitin p50 was found to not significantly influence virus release for the wt but did increase (10-fold) virus release of the PPPY-PTRP mutant (Fig. 5B and C, lanes 1 to 4). The expression of dynamitin p50 was not altered by coexpression of the different viral constructs (Fig. 5B). Based on the ability of dynamitin p50 to move MVBs away from the expected position of the MTOC, this confirms the prediction that overexpression of this construct could increase virus particle release, particularly with the PPPY-PTRP mutant.
DISCUSSION
The Gag protein for most retroviruses is thought to be targeted to the plasma membrane, where particle assembly and release occur. In this study, we have observed that HTLV-1 Gag can colocalize with an MVB marker. Particles or particle buds were also observed to be associated with MVBs by electron microscopy, implying that Gag targeting to the MVB resulted in particle budding. Three possible mechanisms could explain these observations. First, HTLV-1 particles could be secreted by cells and then be reinternalized and targeted to MVBs via the endocytic pathway, a phenomenon that has been reported previously (30). Second, HTLV-1 Gag could be targeted to MVBs and assembled into virus particles at the limiting membrane in a manner similar to the biogenesis of the internal vesicles of MVBs. Third, MVB-associated virus may have begun the assembly process at the plasma membrane but ended up in MVBs via membrane recycling. Several lines of evidence argue in favor of the second possibility. First, we have observed immature ring-like particles inside the lumen of MVB and the electron-dense Gag aggregates that are in the process of assembly or budding along with mature particles. These observations suggest that Gag was targeted to MVBs and assembled into virions. Second, the intracellular Gag showed no obvious colocalization with an early endosomal marker, EEA1, suggesting that Gag proteins are not transported to the MVB from early or recycling endosomes, which would be the case if the virus particles were internalized through endocytosis. Third, immunogold labeling of preembedded cells showed HTLV-1 particles inside MVBs decorated with CD63, a protein that is normally found in MVB membranes. These findings suggest that HTLV-1 particles budded into the MVB, incorporating CD63 into their membrane envelope as they assembled at the limiting membrane. It should be noted that CD63 can also be found at the plasma membrane, and virion incorporation of CD63 does not exclude the possibility that CD63-containing particles were released from plasma membrane.
We found that the Gag mutant PPPY-PTRP was primarily directed to MVBs. This observation supports the hypothesis that the PTAP motif is involved in targeting Gag to the plasma membrane. This hypothesis is further supported by the observation of MVB localization of the PPPY-PPPY (Fig. 1F) or PPPY-YPDL (data not shown) mutant. In contrast, the PPPY motif does not appear to influence Gag targeting in HeLa cells since Gag of mutant APPY-PTAP was targeted primarily to plasma membrane. The role of the PTAP motif in HTLV-1 Gag targeting involves interaction with TSG101, since it was reported that overexpression of the N-terminal fragment of TSG101 redirected the particle assembly inside cytoplasmic vacuoles (6). Consistent with our findings, the role of the PTAP-TSG101 interaction in recruiting viral protein to the plasma membrane for assembly has been suggested for Ebola virus (35). Interestingly, both HTLV-1 and Ebola virus contain both the PPPY and PTAP motifs. Another interesting observation relevant to our observations is that the deletion of the HIV-1 p6 protein of Gag (which contains the PTAP motif) does not affect HIV-1 Gag targeting to the MVB in HeLa cells and in macrophages (24). This implies that MVB targeting is not driven by an interaction between p6 and the MVB sorting machinery (e.g., TSG101).
Our study suggests that HTLV-1 particles in MVBs can be released from cells. Such an exocytic pathway has been reported in a variety of hematopoietic cell types such as B lymphocytes and dendritic cells (19, 29). In these cells, exocytosis results in the secretion of the intralumenal membrane vesicles, called exosomes. The exocytic process is best documented for the MIICs. These compartments identified as late endosomes/MVBs have been previously shown to serve as transport organelles, which guide MHC class II to the cell surface (38, 39). Microtubules are involved in the intracellular transport of MIIC. Two molecular motor proteins, kinesin and dynein, are in control of the movement of MIIC away or toward the MTOC region (38). Our observations in this study support the conclusion that virus-containing MVBs can traffic to the cell surface and that virus release likely occurs by MVB fusion with the plasma membrane. Recent studies have suggested that AP-3 is involved in targeting HIV-1 Gag to the MVB (11). It is not currently known what role AP-3 plays in HTLV-1 Gag targeting to the MVB.
We discovered in this study that the PTAP motif can strongly influence HTLV-1 particle release. Specifically, we found that an HTLV-1 Gag mutant with an intact PPPY motif but a mutated PTAP motif (i.e., PPPY-PTRP) led to a dramatic 20-fold reduction in virus particle release. This is in striking contrast to the same mutant in 293T cells, which we previously reported to reduce particle release by less than twofold as well as not influence the distribution of Gag in cells (37). Our findings indicate that (i) PTAP function is cell type dependent and (ii) PTAP plays a role (direct or indirect) in Gag targeting to the plasma membrane. We are currently investigating why particle reduction for this mutant in HeLa cells is greatly reduced. Ono and Freed (24) observed that MVB localization of HIV-1 Gag was still observed in the absence of p6 (PTAP is located in p6), suggesting that an interaction between Gag and TSG101 is not required for Gag targeting to the MVB. This observation is consistent with our findings with HTLV-1 Gag. The observations with HIV-1 were used to hypothesize that Gag targeting involves two different cellular binding partners, one on the plasma membrane and the other in the MVB. Our current studies are directed at testing this hypothesis as well as elucidating the mechanistic basis for the cell type dependence of PTAP function.
ACKNOWLEDGMENTS
We thank K. Wolken of the Campus Microscopy and Imaging Facility at Ohio State University for technical assistance and E. Freed and A. Ono for helpful discussion.
This work was supported by NIH grants AI053155 and AI057735.
These authors contributed equally to this work.
REFERENCES
Babst, M., D. J. Katzmann, E. J. Estepa-Sabal, T. Meerloo, and S. D. Emr. 2002. Escrt-III: an endosome-associated heterooligomeric protein complex required for MVB sorting. Dev. Cell 3:271-282.
Babst, M., D. J. Katzmann, W. B. Snyder, B. Wendland, and S. D. Emr. 2002. Endosome-associated complex, ESCRT-II, recruits transport machinery for protein sorting at the multivesicular body. Dev. Cell 3:283-289.
Benmerah, A., M. Bayrou, N. Cerf-Bensussan, and A. Dautry-Varsat. 1999. Inhibition of clathrin-coated pit assembly by an Eps15 mutant. J. Cell Sci. 112:1303-1311.
Benmerah, A., C. Lamaze, B. Begue, S. L. Schmid, A. Dautry-Varsat, and N. Cerf-Bensussan. 1998. AP-2/Eps15 interaction is required for receptor-mediated endocytosis. J. Cell Biol. 140:1055-1062.
Blot, V., F. Perugi, B. Gay, M. C. Prevost, L. Briant, F. Tangy, H. Abriel, O. Staub, M. C. Dokhelar, and C. Pique. 2004. Nedd4.1-mediated ubiquitination and subsequent recruitment of Tsg101 ensure HTLV-1 Gag trafficking towards the multivesicular body pathway prior to virus budding. J. Cell Sci. 117:2357-2367.
Bouamr, F., J. A. Melillo, M. Q. Wang, K. Nagashima, M. de Los Santos, A. Rein, and S. P. Goff. 2003. PPPYEPTAP motif is the late domain of human T-cell leukemia virus type 1 Gag and mediates its functional interaction with cellular proteins Nedd4 and Tsg101. J. Virol. 77:11882-11895.
Bryant, M., and L. Ratner. 1990. Myristoylation-dependent replication and assembly of human immunodeficiency virus 1. Proc. Natl. Acad. Sci. USA 87:523-527.
Copeland, K. F. T., A. G. M. Haaksma, D. Derse, and J. L. Heeney. 1994. Detection of human T-cell leukemia virus 1 permissive cells using cell lines producing selectable recombinant virions. J. Virol. Methods 50:219-226.
Demirov, D. G., and E. O. Freed. 2004. Retrovirus budding. Virus Res. 106:87-102.
Demirov, D. G., J. M. Orenstein, and E. O. Freed. 2002. The late domain of human immunodeficiency virus type 1 p6 promotes virus release in a cell type-dependent manner. J. Virol. 76:105-117.
Dong, X., H. Li, A. Derdowski, L. Ding, A. Burnett, X. Chen, T. R. Peters, T. S. Dermody, E. Woodruff, J. J. Wang, and P. Spearman. 2005. AP-3 directs the intracellular trafficking of HIV-1 Gag and plays a key role in particle assembly. Cell 120:663-674.
Freed, E. O. 1998. HIV-1 Gag proteins: diverse functions in the virus life cycle. Virology 251:1-15.
Freed, E. O. 2002. Viral late domains. J. Virol. 76:4679-4687.
Gilerovitch, H. G., G. A. Bishop, J. S. King, and R. W. Burry. 1995. The use of electron microscopic immunocytochemistry with silver-enhanced 1.4-nm gold particles to localize GAD in the cerebellar nuclei. J. Histochem. Cytochem. 43:337-343.
Gottlinger, H. G. 2001. The HIV-1 assembly machine. AIDS 15(Suppl. 5):S13-S20.
Heidecker, G., P. A. Lloyd, K. Fox, K. Nagashima, and D. Derse. 2004. Late assembly motifs of human T-cell leukemia virus type 1 and their relative roles in particle release. J. Virol. 78:6636-6648.
Hill, C. P., D. Worthylake, D. P. Bancroft, A. M. Christensen, and W. I. Sundquist. 1996. Crystal structures of the trimeric human immunodeficiency virus type 1 matrix protein: implications for membrane association and assembly. Proc. Natl. Acad. Sci. USA 93:3099-3104.
Hirokawa, N. 1998. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science 279:519-526.
Kolodziej, D., and E. Pajtasz-Piasecka. 2003. Role of dendritic cells in recognizing antigens: their binding, transformation and presentation to T lymphocytes. Postepy Hig. Med. Dosw. 57:149-170. (In Polish.)
Le Blanc, I., M. C. Prevost, M. C. Dokhelar, and A. R. Rosenberg. 2002. The PPPY motif of human T-cell leukemia virus type 1 Gag protein is required early in the budding process. J. Virol. 76:10024-10029.
Martin-Serrano, J., T. Zang, and P. D. Bieniasz. 2003. Role of ESCRT-I in retroviral budding. J. Virol. 77:4794-4804.
Massiah, M. A., M. R. Starich, C. Paschall, M. F. Summers, A. M. Christensen, and W. I. Sundquist. 1994. Three-dimensional structure of the human immunodeficiency virus type 1 matrix protein. J. Mol. Biol. 244:198-223.
Nydegger, S., M. Foti, A. Derdowski, P. Spearman, and M. Thali. 2003. HIV-1 egress is gated through late endosomal membranes. Traffic 4:902-910.
Ono, A., and E. O. Freed. 2004. Cell-type-dependent targeting of human immunodeficiency virus type 1 assembly to the plasma membrane and the multivesicular body. J. Virol. 78:1552-1563.
Orenstein, J. M., M. S. Meltzer, T. Phipps, and H. E. Gendelman. 1988. Cytoplasmic assembly and accumulation of human immunodeficiency virus types 1 and 2 in recombinant human colony-stimulating factor-1-treated human monocytes: an ultrastructural study. J. Virol. 62:2578-2586.
Parent, L. J., C. B. Wilson, M. D. Resh, and J. W. Wills. 1996. Evidence for a second function of the MA sequence in the Rous sarcoma virus Gag protein. J. Virol. 70:1016-1026.
Pelchen-Matthews, A., B. Kramer, and M. Marsh. 2003. Infectious HIV-1 assembles in late endosomes in primary macrophages. J. Cell Biol. 162:443-455.
Raposo, G., M. Moore, D. Innes, R. Leijendekker, A. Leigh-Brown, P. Benaroch, and H. Geuze. 2002. Human macrophages accumulate HIV-1 particles in MHC II compartments. Traffic 3:718-729.
Raposo, G., H. W. Nijman, W. Stoorvogel, R. Liejendekker, C. V. Harding, C. J. Melief, and H. J. Geuze. 1996. B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183:1161-1172.
Romero, I. A., M. C. Prevost, E. Perret, P. Adamson, J. Greenwood, P. O. Couraud, and S. Ozden. 2000. Interactions between brain endothelial cells and human T-cell leukemia virus type 1-infected lymphocytes: mechanisms of viral entry into the central nervous system. J. Virol. 74:6021-6030.
Sakurai, A., J. Yasuda, H. Takano, Y. Tanaka, M. Hatakeyama, and H. Shida. 2004. Regulation of human T-cell leukemia virus type 1 (HTLV-1) budding by ubiquitin ligase Nedd4. Microbes Infect. 6:150-156.
Schultz, A. M., and A. Rein. 1989. Unmyristylated Moloney murine leukemia virus Pr65gag is excluded from virus assembly and maturation events. J. Virol. 63:2370-2373.
Sherer, N. M., M. J. Lehmann, L. F. Jimenez-Soto, A. Ingmundson, S. M. Horner, G. Cicchetti, P. G. Allen, M. Pypaert, J. M. Cunningham, and W. Mothes. 2003. Visualization of retroviral replication in living cells reveals budding into multivesicular bodies. Traffic 4:785-801.
Swanstrom, R., and J. W. Wills. 1997. Synthesis, assembly, and processing of viral proteins, p. 263-334. In J. M. Coffin, S. H. Hughes, and H. E. Varmus (ed.), Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Timmins, J., G. Schoehn, S. Ricard-Blum, S. Scianimanico, T. Vernet, R. W. Ruigrok, and W. Weissenhorn. 2003. Ebola virus matrix protein VP40 interaction with human cellular factors Tsg101 and Nedd4. J. Mol. Biol. 326:493-502.
von Schwedler, U. K., M. Stuchell, B. Muller, D. M. Ward, H. Y. Chung, E. Morita, H. E. Wang, T. Davis, G. P. He, D. M. Cimbora, A. Scott, H. G. Krausslich, J. Kaplan, S. G. Morham, and W. I. Sundquist. 2003. The protein network of HIV budding. Cell 114:701-713.
Wang, H., N. J. Machesky, and L. M. Mansky. 2004. Both the PPPY and PTAP motifs are involved in human T-cell leukemia virus type 1 particle release. J. Virol. 78:1503-1512.
Wubbolts, R., M. Fernandez-Borja, I. Jordens, E. Reits, S. Dusseljee, C. Echeverri, R. B. Vallee, and J. Neefjes. 1999. Opposing motor activities of dynein and kinesin determine retention and transport of MHC class II-containing compartments. J. Cell Sci. 112:785-795.
Wubbolts, R., M. Fernandez-Borja, L. Oomen, D. Verwoerd, H. Janssen, J. Calafat, A. Tulp, S. Dusseljee, and J. Neefjes. 1996. Direct vesicular transport of MHC class II molecules from lysosomal structures to the cell surface. J. Cell Biol. 135:611-622.
Yuan, X., X. Yu, T. H. Lee, and M. Essex. 1993. Mutations in the N-terminal region of human immunodeficiency virus type 1 matrix protein block intracellular transport of the Gag precursor. J. Virol. 67:6387-6394.
Zhou, W., L. J. Parent, J. W. Wills, and M. D. Resh. 1994. Identification of a membrane-binding domain within the amino-terminal region of human immunodeficiency virus type 1 Gag protein which interacts with acidic phospholipids. J. Virol. 68:2556-2569.(Irene J. Dorweiler, Susan)