Previous Article | Next Article ![]()
Journal of Virology, August 2002, p. 7777-7789, Vol. 76, No. 15
0022-538X/02/$04.00+0 DOI: 10.1128/JVI.76.15.7777-7789.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Laboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland
Received 28 February 2002/ Accepted 2 May 2002
|
|
|---|
|
|
|---|
Of the seven known proteins associated with IEV- or EEV-specific membranes, the ones encoded by the F13L and B5R open reading frames (ORFs) are required for the wrapping of IMV to form IEV (3, 9, 51). The B5R product is a 42-kDa type I integral membrane component of the EEV (8, 23). The F13L ORF encodes a nonglycosylated, palmitylated protein that is the most abundant component of the EEV membrane (15, 16, 18, 19, 40). The F13L protein contains a variant of the HKD (His-Lys-Asp) motif that is conserved in a superfamily of phospholipases and phospholipid synthases (25, 33, 45) and has been reported previously to exhibit broad-specificity lipase activities in vitro (1). In addition, mutant vaccinia viruses with amino acid substitutions of either the conserved Lys or Asp of the phospholipase motif exhibited wrapping defects that inhibited IEV formation (37, 45).
To better understand the role of the F13L protein in membrane wrapping, we fused the F13L ORF to that of the enhanced green fluorescent protein (GFP) so that it could be visualized by microscopy. The GFP moiety had no deleterious effect, as the fusion protein complemented a mutant vaccinia virus with a deleted F13L gene and was localized in the IEV, CEV, and EEV membranes (22). When expressed by transfection in the absence of other viral gene products, F13L-GFP was localized in Golgi membranes and post-Golgi vesicles that contained early and late endosomal markers (22). Under similar transfection conditions, the B5R protein was targeted to juxtanuclear Golgi membranes (24, 29, 48). However, coexpression of F13L-GFP and B5R resulted in the colocalization of the two proteins in endosomal vesicles (22). Furthermore, colocalization was dependent on both an unmutated phospholipase motif and the palmitylation site in the F13L protein. These results were intriguing because phospholipase D (PLD) regulates the budding of vesicles from trans-Golgi membranes (2, 6, 10, 26, 41, 42). We therefore proposed a similar role for the F13L protein. To further examine this hypothesis, we posed the following questions. Is the colocalization of B5R with F13L specific, or will other Golgi transmembrane proteins colocalize with F13L? Will B5R or other viral proteins colocalize with PLD, instead of F13L, in post-Golgi vesicles? Can PLD complement an F13L deletion mutant? Will a PLD inhibitor prevent the localization of F13L in endosomes and inhibit extracellular vaccinia virus formation?
|
|
|---|
, originally referred to as vRB12 (3), has most of the F13L ORF replaced with the Escherichia coli xanthine guanine phosphoribosyltransferase ORF. Recombinant vF13L-GFP contains the GFP coding sequence at the C terminus of the F13L ORF (22). For virus titration and analysis of plaque size, infected BS-C-1 monolayers were fixed and stained with 0.1% crystal violet in 20% methanol.
Plasmid constructions.
Plasmids pF13L-GFP containing the GFP coding sequence at the C terminus of the F13L ORF and pB5R containing the B5R ORF were described previously (22). Plasmid pSVGL1, containing wild-type vesicular stomatitis virus (VSV) G protein, and VSVG-GFP, containing a GFP-tagged ts045 VSV G protein, were generously provided by J. Rose and J. Lippincott-Schwartz (34), respectively. Plasmids pCGN-hPLD1b and pCGN-hPLD1b (K898R) containing wild-type and mutated PLD ORFs (17), respectively, were kindly provided by M. Frohman. Additional plasmids were constructed by standard procedures. Briefly, the F13L-GFP fusion ORF in plasmid pGF13L (22) was replaced by PLD1-HA or PLD1-GFP fusion ORFs containing the hemagglutinin (HA) tag or GFP sequence at the N terminus. The HA tag or GFP ORF was joined with PLD1 coding sequence by two-stage PCR, and the final PCR products were cloned into plasmid pGF13L, yielding plasmids pGPLD-LR and pGPLD-LR2, respectively. These two plasmids contain PLD1-HA and PLD1-GFP fusion constructs flanked by
600 bp of left and right flanking sequences of the F13L gene. Similarly, the GFP coding sequence was appended at the N terminus of PLD1 in plasmid pCGN-hPLD1b. Relevant portions of all plasmids were verified by sequencing.
Recombinant virus construction.
Recombinant vaccinia virus expressing PLD1-GFP under the F13L late promoter was constructed according to the procedure described previously (22). Briefly, HeLa cells were infected with vF13L
and immediately transfected with plasmid pGPLD-LR2. After 2 days at 37°C, cells were harvested and the lysate was analyzed by plaque assay on BS-C-1 cells. Isolated green plaques were purified and amplified, and the recombinant virus called vPLD1-GFP was characterized as described earlier (22).
Antibodies. A Golgi sampler kit containing monoclonal antibodies (MAbs) to marker proteins of cis- and trans-Golgi compartments and early endosomal antigen 1 (EEA1) were purchased from Transduction Laboratories. Rabbit anti-ß-coat protein (anti-ß-COP) polyclonal antibody was from Affinity Bioreagents; MAb HA.11, which recognizes an influenza virus HA epitope, was from Babco. Secondary antibody conjugates were purchased from Jackson ImmunoResearch Laboratories. Rabbit polyclonal antibody recognizing a peptide sequence at the C terminus of the A36R protein (53), 192C rat MAb against the B5R protein (39), and MAbI1 anti-VSV G (28) have been described previously. MAb to lysosome-associated membrane protein 2 (LAMP2) was a gift of Thomas August.
Transfections and infections. For transfection, plasmids were prepared with the Qiagen plasmid Midi preparation kit. HeLa cells were grown on glass coverslips until they reached 80 to 90% confluence. Routinely, 2 to 10 µg of Lipofectamine 2000 (Invitrogen) and 0.5 to 2 µg DNA were diluted separately in Opti-MEM I medium (Invitrogen), mixed, incubated at room temperature for 20 min, and added to the cells for 4 to 5 h at 37°C. The Lipofectamine-DNA complex was replaced with Dulbecco's modified Eagle's medium supplemented with 10% FBS, and the incubation was continued for a total of 24 h.
For infections, virus stocks were diluted in growth medium supplemented with 2.5% FBS and added to cell monolayers in wells or on coverslips. After 2 h of incubation at 37°C, virus inocula were replaced with fresh medium containing 2.5% FBS and incubated for a further 17 to 18 h.
Confocal microscopy. At 24 h after transfection, cells were fixed with cold 4% paraformaldehyde in phosphate-buffered saline (PBS) and then incubated at room temperature for 20 min and permeabilized with 0.2% Triton X-100 in PBS for 5 min at room temperature. The permeabilized cells were incubated with primary antibodies diluted in 10% FBS in PBS for 1 h followed by secondary antibody diluted in 10% FBS in PBS for 30 min at room temperature. When two different antibodies were used, staining was sequential. The Golgi apparatus was visualized by staining with mouse anti-p115 MAb unless otherwise stated. Stained cells were washed extensively with PBS, and coverslips were mounted in 20% glycerol and sealed with rubber cement. Fluorescence was examined with a Leica TCS NT inverted confocal microscope, and images were overlaid by using Adobe Photoshop version 5.0.2.
Electron microscopy. For immunoelectron microscopy, RK13 cells were grown in 60-mm-diameter dishes and infected with recombinant vaccinia virus expressing PLD1-GFP at a multiplicity of 10. After 24 h, the cells were prepared for freezing, and ultrathin sections were cut, collected, immunostained, and viewed as previously described (53). The rabbit anti-GFP polyclonal antibody (Clontech) was used at a dilution of 1:100.
Effects of alcohols on intracellular localization and virus formation. Butanol-1, butanol-2, or propanol-2 (0.5 or 1%) was added to the medium overlaying cells at 4 h after transfection, and the cells were incubated for a total of 24 h at 37°C. Alternatively, BS-C-1 cells were infected with vaccinia virus for 2 h at 37°C and then the inoculum was replaced with fresh medium supplemented with butanol-1, butanol-2, or propanol-2 and incubated for 48 h at 37°C. Cells were stained with crystal violet to visualize the plaques. To measure the effect of butanol-1 on virus yields, HeLa cells were infected with the IHD-J strain of vaccinia virus and treated with alcohols as described above. Cells and medium were harvested separately, adjusted to equal volumes, and analyzed by plaque assay. As a control, medium of untreated, infected cells was harvested and treated with butanol-1 at 37°C for 24 h and then analyzed by plaque assay.
|
|
|---|
with a deleted F13L gene (Fig. 1). When cells transfected with pF13L-GFP were infected with vF13L
, about 45% of the A36R localized with F13L (Fig. 1). Thus, the F13L protein dramatically modulated the intracellular distribution of both the B5R protein as previously described (22) and the A36R protein as shown here.
![]() View larger version (66K): [in a new window] |
FIG. 1. Effect of F13L-GFP on the intracellular localization of A36R in transfected or infected cells. HeLa cells were transfected or infected for 24 or 18 h, respectively, with the name of the plasmid or virus indicated at the left of each row. First row, cells transfected with pA36R were stained with anti-p115 MAb followed by indodicarbocyanine (Cy5)-conjugated anti-mouse immunoglobulin antibody and then with anti-A36R polyclonal antibody followed by Texas red-conjugated anti-rabbit immunoglobulin antibody. Second row, cells were cotransfected with pA36R and pF13L-GFP and stained for A36R as in the first row. Third row, cells infected with vF13L-GFP (1 PFU/cell) were stained with anti-A36R polyclonal antibody followed by tetramethyl rhodamine isothiocyanate-conjugated anti-rabbit immunoglobulin antibody. Fourth row, cells infected with vF13L (1 PFU/cell) were stained with anti-p115 MAb followed by tetramethyl rhodamine isothiocyanate-conjugated anti-mouse immunoglobulin antibody and then stained with anti-A36R polyclonal antibody followed by Alexa 488-conjugated anti-rabbit immunoglobulin antibody. Fifth row, cells were transfected with plasmid pF13L-GFP; after 24 h the cells were infected with vF13L as in the fourth row and stained with anti-A36R polyclonal antibody followed by tetramethyl rhodamine isothiocyanate-conjugated anti-rabbit immunoglobulin antibody. Cells were analyzed by confocal microscopy. Green, GFP or Alexa 488; red, Texas red or tetramethyl rhodamine isothiocyanate; yellow, overlap of green and red. White arrowheads show the colocalization of F13L-GFP with A36R in vesicular structures.
|
![]() View larger version (36K): [in a new window] |
FIG. 2. Time course of VSV G colocalization with F13L-GFP. HeLa cells were cotransfected with plasmids pF13L-GFP and pVSVGts045 and incubated at 39°C. After 24 h, the cells were shifted to 31°C and chased for the time indicated on the left of each row. Cells were fixed, permeabilized, stained with anti-VSV G MAb followed by rhodamine red-conjugated anti-mouse immunoglobulin antibody, and analyzed by confocal microscopy. Green, GFP; red, rhodamine red; yellow, overlap of green and red.
|
![]() View larger version (69K): [in a new window] |
FIG. 3. Colocalization of PLD1-GFP with cellular markers. HeLa cells were transfected with plasmid pPLD1-GFP for 24 h and then fixed, permeabilized, and stained with the indicated antibodies. Transfected HeLa cells were stained with mouse MAbs to p115, p230, LAMP2, or EEA1 or rabbit polyclonal anti-ß-COP antibodies followed by rhodamine red-conjugated anti-mouse immunoglobulin antibody or tetramethyl rhodamine isothiocyanate-conjugated anti-rabbit immunoglobulin antibody, respectively. Cells were analyzed by confocal microscopy. Green, GFP; red, tetramethyl rhodamine isothiocyanate; yellow, overlap of green and red. White arrowheads indicate vesicles containing PLD1-GFP and endosomal marker LAMP2 or EEA1.
|
![]() View larger version (66K): [in a new window] |
FIG.4. Colocalization of PLD1 with F13L-GFP and B5R in transfected cells. HeLa cells were transfected with plasmids indicated at the left of each row and incubated for 24 h. First row, cells transfected with pF13L-GFP alone (left), pCGN-hPLD1b alone (middle), or pB5R alone (right) were unstained or stained with mouse anti-HA MAb or rat anti-B5R MAb followed by rhodamine red-conjugated anti-mouse immunoglobulin antibody or fluorescein isothiocyanate-conjugated anti-rat immunoglobulin antibody. Second row, cells cotransfected with plasmid pF13L-GFP and pCGN-hPLD1b were stained with mouse anti-HA followed by rhodamine red-conjugated immunoglobulin antibody. Third row, cells cotransfected with plasmids pCGN-hPLD1b and pB5R were stained for HA and then for B5R as in the first row. Fourth and fifth rows, cells cotransfected with pPLD1-GFP and pB5R or pPLD1(K898R)-HA and pB5R, respectively, were stained for B5R or HA and examined by confocal microscopy as in the third row. Green, GFP or fluorescein isothiocyanate; red, rhodamine red; yellow, overlap of green and red.
|
. Whereas B5R and A36R were predominantly juxtanuclear in cells infected with the mutant virus, colocalization with vesicles containing PLD1-HA was found in the transfection-infection experiment (Fig. 5).
![]() View larger version (56K): [in a new window] |
FIG. 5. Intracellular localization of B5R and A36R in vF13L -infected cells expressing PLD1-HA. First row, HeLa cells were transfected with plasmid pCGN-PLD1b or infected with vF13L . Second and third rows, transfected cells were infected with vF13L and after 18 h were fixed, permeabilized, and stained with anti-HA MAb followed by Alexa 488-conjugated anti-mouse immunoglobulin antibody. Cells were then stained with either anti-B5R MAb or anti-A36R followed by tetramethyl rhodamine isothiocyanate-conjugated anti-rat or anti-rabbit immunoglobulin antibody, respectively. Stained cells were analyzed by confocal microscopy. Green, Alexa 488; red, tetramethyl rhodamine isothiocyanate; yellow, overlap of green and red. White arrowheads show the colocalization of PLD1 with B5R and A36R in vesicular structures.
|
. As an initial measure of complementation, the cells were stained with phalloidin to discern IEV or CEV with actin tails. When cells were transfected with a plasmid expressing F13L, actin tails were present in the majority of cells whereas very few cells transfected with PLD1 had actin tails (Fig. 6A). In addition, we constructed a recombinant vaccinia virus in which the PLD1-GFP ORF replaced that of F13L. This recombinant virus (called vPLD1-GFP) expressed an expected
150-kDa fusion protein as demonstrated by immunoblotting and immunoprecipitation analyses (data not shown). Nevertheless, vPLD1-GFP formed small plaques, and IEV were not detected by electron microscopy. PLD1-GFP was associated with membranes, some of which were in the vicinity of IMV but not associated with them (Fig. 6B). We concluded that PLD1 could not fully substitute for F13L.
![]() View larger version (75K): [in a new window] |
FIG. 6. Failure of PLD1 to complement an F13L deletion mutant. (A) Quantitation of actin tail formation in vF13L -infected cells transfected with either pGF13L or pGPLD1-LR. HeLa cells were infected with vF13L for 2 h and then transfected with plasmid pGF13L or pGPLD1-LR. At 24 h after transfection, the cells were fixed and permeabilized. Cells transfected with plasmid pGPLD1-LR were stained with anti-HA MAb followed by Alexa 488-conjugated anti-mouse immunoglobulin antibody. Plasmid pGF13L has GFP at the C terminus of F13L. Data shown are the averages of three separate experiments. (B) vPLD1-GFP does not produce IEV particles. A confluent monolayer of RK13 cells was infected with vPLD1-GFP. After 24 h, cells were fixed, cryosectioned, and probed with rabbit anti-GFP polyclonal antibodies followed by protein A conjugated to 10-nm colloidal gold particles. Arrowheads point to PLD1-GFP fusion protein on membranes close to IMV particles.
|
![]() View larger version (95K): [in a new window] |
FIG. 7. Effects of butanol-1, butanol-2, and propanol-2 on intracellular localization of F13L-GFP. HeLa cells were transfected with plasmid pF13L-GFP. After 4 h, 1% concentrations of the indicated alcohols were added to the media and the incubations were continued for a total of 24 h at 37°C. Cells were fixed and analyzed by confocal microscopy.
|
![]() View larger version (97K): [in a new window] |
FIG. 8. Effects of butanol-1, butanol-2, and propanol-2 on plaque size of vaccinia virus. BS-C-1 cells, in six-well tissue culture plates, were infected with 50 to 100 PFU of vaccinia virus strain WR or IHD-J/well. After 2 h of adsorption of IHD-J, the virus inocula were replaced with liquid media supplemented with 0.5 or 1.0% concentrations of the indicated alcohol and the incubation was continued for 2 days. For vaccinia virus WR, methylcellulose was included in the overlay and the cells were incubated for 3 days. Plaques were visualized by staining with crystal violet.
|
![]() View larger version (22K): [in a new window] |
FIG. 9. Effect of butanol-1 on the production of IMV and EEV particles. (A) BS-C-1 cells were infected with vaccinia virus strain IHD-J. After 2 h, the inoculum was replaced with medium supplemented with 0.5 or 1.0% butanol-1. After 48 h at 37°C, the medium and cells were harvested separately in similar volumes and analyzed by plaque assay. Plaque numbers are presented as the averages of three separate counts from two different experiments with standard deviations. Filled and unfilled bars refer to extracellular and cell-associated virus, respectively. (B) BS-C-1 cells were infected with vaccinia virus strain IHD-J, and after adsorption, inoculum was replaced with normal medium. After 48 h, the medium was collected; cleared by low-speed centrifugation; and incubated with 0, 0.5, or 1.0% butanol-1 at 37°C for 48 h. Infectious virus was quantified by plaque assay.
|
|
|
|---|
At the start of this study we posed several questions. The first was whether colocalization of B5R with F13L is specific or whether other Golgi transmembrane proteins colocalize with F13L. In a previous study (22), we demonstrated by cotransfection experiments that the B5R membrane protein colocalized with the F13L protein in endosomal vesicles. In the absence of F13L, B5R was predominantly associated with juxtanuclear Golgi membranes. Attempts to show interaction of the B5R and F13L proteins by coimmunoprecipitation were unsuccessful, however, leading us to suspect that the colocalization did not require sequence-specific interactions. Again using cotransfection methods, we found that another IEV protein, A36R, colocalized with F13L in endosomal vesicles. Moreover, the VSV G protein also colocalized with F13L. We took advantage of a VSV G protein harboring a ts mutation that prevented it from exiting the endoplasmic reticulum at the nonpermissive temperature. When the temperature was lowered, the G protein was first transported to the Golgi network and then into endosomal vesicles within 30 to 45 min. As sequence-specific interactions between vaccinia virus and VSV proteins seemed improbable, we concluded that F13L induced the formation of vesicles that incorporated Golgi membrane proteins relatively nonspecifically.
The second and third questions were whether B5R and A36R colocalize with PLD in place of F13L in post-Golgi vesicles and whether PLD can complement an F13L deletion mutant. Our previous study (22) demonstrated that mutations in the phospholipase motif of F13L abrogated its ability to induce colocalization of B5R. Since the present findings made sequence-specific interactions between F13L and colocalizing proteins unlikely, we considered that PLD1 might at least partially substitute for F13L. This prediction was borne out by cotransfection experiments. PLD1 and F13L colocalized with each other, and PLD1 induced the colocalization of B5R in endosomal vesicles. Moreover, B5R and A36R colocalized with PLD1 when cells expressing PLD1 were infected with an F13L deletion mutant virus. These results are consistent with reports that PLD stimulates release of nascent secretory vesicles from the trans-Golgi network (6, 26). PLD1, however, could not substitute for F13L in the formation of IEV, suggesting that PLD activity is insufficient for virus assembly.
The fourth question was whether a PLD inhibitor would prevent the localization of F13L in endosomes and inhibit extracellular vaccinia virus formation. We used a specific PLD inhibitor to correlate an enzymatic role of F13L in inducing vesicles with the wrapping process needed to form extracellular virus. In the presence of primary but not secondary alcohols, PLD produces phosphatidylalcohols at the expense of phosphatidic acid. Alcohol concentrations of 1 to 3% have been shown elsewhere to be inhibitory in vitro and in vivo (6, 26). We demonstrated that 1% butanol-1 prevented the localization of F13L in endosomal vesicles and severely reduced vaccinia virus plaque size and comet formation. No such effect was seen with the same concentrations of butanol-2 or propanol-2. The specificity of this effect was confirmed by showing that butanol-1 inhibited formation of extracellular virus without significantly reducing the yield of IMV.
The findings that F13L contains a region similar to the catalytic site of PLD (25, 33, 45); mutagenesis of the catalytic motif abrogates F13L function (22, 37, 45); F13L, like PLD, induces post-Golgi vesicle formation; and primary alcohols inhibit F13L and PLD function are all consistent with F13L having an essential PLD activity. Nevertheless, F13L lacks a conserved histidine of the PLD active site motif, and recombinant F13L fusion proteins were found previously only to have phospholipase A and C activities in vitro (1). Although this negative result for PLD activity could be due to a deficiency in the in vitro assay conditions or the absence of an activator, it is possible that the PLD-like effects of F13L seen in vivo are indirect and result from its other described phospholipase activities or other interactions. The inhibition of F13L-mediated vesicle formation by expression of amphiphysins (unpublished data), which specifically bind PLD (27), would be consistent with activation of PLD by F13L. PLD is a highly regulated protein and is activated by ADP-ribosylation factor 1, which indirectly induces the formation of coatomer-coated vesicles (17, 26).
In summary, our present and previous data support a model in which the F13L protein localizes in Golgi membranes and induces vesicle formation by direct or indirect means through a PLD-like activity. Viral membrane proteins present in the Golgi apparatus are included in the induced vesicles. Up to this stage, PLD1 can substitute for F13L. Further steps involving the membrane wrapping of IMV, however, depend on additional properties of the F13L protein that remain to be determined.
|
|
|---|
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»