Redistribution of Cyclophilin A to Viral Factories during Vaccinia Virus Infection and Its Incorporation into Mature Particles

Cells and viruses.

Monolayer cultures of BSC-40 cells were grown at 37°C in Dulbecco's modified Eagle medium (Gibco-BRL Laboratories) supplemented with 8% calf serum and 2% heat-inactivated fetal bovine serum as described previously (17). VV strain WR was propagated and titrated in BSC-40 cells as described elsewhere (17). VV mutant ts53 was kindly provided by R. Condit (University of Florida) and amplified in BSC-40 cells at 31°C (54).

Antibodies and CsA.

The polyclonal anti-CypA antibody was purchased from Affinity Bioreagents Inc., Golden, Colo., and Upstate Biotechnology, Lake Placid, N.Y. Monoclonal anti-α-tubulin (clone B-5-1-2) was purchased from Sigma Chemical Co., St Louis, Mo. Monoclonal anti-Hsp70 was a generous gift from M. A. Rebello (IPPMG-UFRJ, Rio de Janeiro, Brazil). The following antibodies against VV proteins were used: rabbit polyclonal antiserum raised against total VV structural proteins, which has been previously described (16); rabbit anti-F11L (18); monoclonal anti-A18R (a gift from R. Condit, University of Florida); polyclonal anti-D12L and anti-D1R (a gift from E. Niles, State University of New York, Buffalo); monoclonal anti-D8L (a gift from G. Smith, Imperial College, London, United Kingdom); rabbit anti-A12L (a gift from D. Hruby, Oregon State University); and polyclonal anti-H3L (a gift from B. Moss, National Institutes of Health).

CsA and the analog d-Ala(3-amino)8-cyclosporine (8′A-Cs) were a generous gift from Novartis Pharma AG, Basel, Switzerland. CsA was dissolved in 100% dimethyl sulfoxide and stored at −20°C as a 1,000×-concentrated stock solution. The analog 8′A-Cs was dissolved in 100% methanol (4 mg/ml) and stored at 4°C.

Preparation of mock- and VV-infected cell extracts.

Confluent monolayers of BSC-40 cells (6.5 × 10⁶ cells) were infected with VV (multiplicity of infection [MOI] of 10) for 30 min at 37°C. The inoculum was then removed and replaced with complete medium (zero time of infection). At different times postinfection (p.i.), the monolayers were washed twice with phosphate-buffered saline (PBS), scraped from the dish, and centrifuged at 1,000 × g for 5 min at 4°C. The cells were then resuspended in lysis buffer [10 mM Tris-HCl, 150 mM NaCl, 0.5% Triton X-100, 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride (Calbiochem), 20 μg of leupeptin (Sigma) per ml (pH 7.5)] and placed on ice for 25 min. The postnuclear fraction was obtained by centrifugation at 7,200 × g for 15 min at 4°C. The supernatant was removed and stored at −70°C. Protein concentration was determined as previously described (3). For Western blot analysis, equal amounts of protein were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by immunoblotting.

For metabolic-labeling experiments, mock- or VV-infected cells were pulse-labeled for 1 h at 37°C with 100 μCi of [³⁵S]methionine (Amersham/Pharmacia) per ml in methionine-free medium at the indicated times before extract preparation. For pulse-chase assays, mock- or VV-infected monolayers were [³⁵S]methionine labeled as described above at 2 h p.i. (pulse). After this period, the cells were washed twice in PBS containing a 20-fold excess of unlabeled methionine and incubated with fresh medium supplemented with an excess of unlabeled methionine for the indicated times (chase). The extracts were prepared as described above. Protein concentration was determined as previously described (3).

Western blot analysis.

After electrophoresis, the proteins were transferred to nitrocellulose membranes and the blots were processed for immunodetection as previously described (14). Primary antibodies were diluted 1:1,500 (anti-CypA and anti-α-tubulin; 18 h), 1:2,000 (anti-Hsp70 and anti-VV; 1 h), 1:800 (anti-F11L; 18 h), 1:5,000 (anti-A18R; 1 h), 1:500 (anti-D12L, anti-H3L, anti-A12L, and anti-D1R; 1 h), and 1:100 (anti-D8L; 16 h). The bound antibodies were detected with horseradish peroxidase-conjugated goat anti-rabbit (Amersham/Pharmacia or Santa Cruz Biotechnology) or goat anti-mouse (Amersham/Pharmacia) immunoglobulin G (IgG), followed by incubation with the ECL Western blotting detection reagents (Amersham/Pharmacia), as suggested by the manufacturer.

Affinity chromatography.

Anhydrous coupling of 8′A-Cs to Affi-Gel 10 (Bio-Rad) resin was carried out essentially as previously described (16). Control columns were prepared by the same procedure, except that CsA replaced 8′A-Cs. Postnuclear cell extracts (200 μg) were loaded onto affinity minicolumns, and unbound proteins were recovered by washing the columns with TBAz buffer (20 mM Tris-HCl [pH 7.5], 5 mM β-mercaptoethanol, 0.033% sodium azide). Proteins specifically bound to the matrix were eluted with 2 mg of CsA per ml in 50% TBAz-50% methanol (16). After precipitation with cold acetone, proteins were resolved by SDS-13.5% PAGE. The gels were dried and exposed to X-ray films. In pulse-labeling assays, quantitation of ³⁵S-labeled CypA was performed by densitometric analysis of the Coomassie blue-stained gels and the respective autoradiograms with the image processing and analysis program Scion Image (Beta Release 4; Scion Corporation). For each time point, the value obtained by analysis of the autoradiogram was corrected by the corresponding amount of CypA recovered in the Coomassie blue-stained gel. The ratios obtained for the chase lanes were then divided by the ratio obtained for the pulse lane. The resulting values are presented as percentages of the baseline with reference to the amount of CypA synthesized at 2 h (pulse).

Virus purification.

VV particles were purified from cytoplasmic extracts of infected BSC-40 cells by high-speed centrifugation through a 36% sucrose cushion followed by sedimentation in 25 to 40% sucrose gradient, as previously described (14). For further purification, 2.4 × 10¹¹ purified particles obtained in the first gradient sedimentation were banded on two successive 25 to 40% sucrose gradients. After centrifugation, 450-μl fractions were collected from the bottom of the tube and washed with 10 mM Tris-HCl (pH 7.5). VV particles were recovered by centrifugation at 12,000 × g for 30 min at 4°C and resuspended in 10 mM Tris-HCl (pH 7.5). The number of virus particles was determined from the optical density at 260 nm (OD₂₆₀) with the formula 1 U of OD₂₆₀ = 1.2 × 10¹⁰ particles/ml. The fractions were analyzed by SDS-PAGE and Western blotting as already described.

Detergent extraction and protease treatment of virions.

Purified intracellular mature virus (IMV; 7.5 × 10⁹ particles) was incubated for 60 min at 37°C in a reaction mixture containing 0.5% NP-40 and 10 mM Tris-HCl (pH 7.5) in the presence or absence of 50 mM dithiothreitol (DTT). The soluble envelope proteins and insoluble core protein fractions were collected by centrifugation at 12,000 × g for 30 min at 4°C. Viral cores obtained after 0.5% NP-40 treatment were resuspended in DOC buffer (0.2% deoxycholate, 100 mM Tris-HCl [pH 7.5], 250 mM NaCl, 10 mM DTT) and incubated at 4°C for 30 min. The insoluble protein fraction was separated from the solubilized core fraction by centrifugation at 12,000 × g for 10 min at 4°C. The proteins in each fraction were resolved by SDS-PAGE, followed by immunoblotting as described above, and visualized by silver staining.

Purified IMV (7.5 × 10⁹ particles) was digested with either 5 or 50 μg of proteinase K (Gibco-BRL) per ml in 10 mM Tris-HCl (pH 7.5) for 10 min on ice. Alternatively, purified IMV was also digested with 50 μg of chymotrypsin (Sigma) per ml for 30 min at 37°C. The reactions were stopped by addition of 5 mM phenylmethylsulfonyl fluoride (Sigma), and the reaction mixtures were quick-frozen in liquid nitrogen immediately. SDS-PAGE loading buffer was added to the samples still frozen. Proteins were analyzed by electrophoresis, followed by Western blotting, as described above.

Conventional immunofluorescence and confocal laser scanning microscopy.

BSC-40 cells, grown on 13-mm-diameter round glass coverslips in 24-well plates, were infected with VV at an MOI of 10 and incubated at 37°C as described above. In some experiments, the cells were infected with wild-type (WT) or ts53 mutant VV (MOI of 5) and incubated at 31 or 40°C. When indicated, the cells received fresh medium containing 35 μM CsA, 40 μg of cytosine arabinoside (AraC) per ml, 10 mM hydroxyurea (HU), or 100 μg of cycloheximide (CHX) per ml. Procedures for immunofluorescence microscopy were similar to those described previously (7). At the indicated times p.i., the cells were fixed in 4% paraformaldehyde in PHEM buffer (60 mM PIPES, 20 mM HEPES, 10 mM EGTA, 50 mM MgCl₂, 70 mM KCl [pH 7.2]) for 10 min at room temperature and 50 min at 4°C and washed in PBS. The cells were then permeabilized in 0.5% Triton X-100 for 3 to 5 min and rinsed in PBS. The coverslips were incubated for 1 h in P-BSA (PBS-1% bovine serum albumin [BSA]) and then for 30 min in 50 mM NH₄Cl. The coverslips were washed in P-BSA and incubated for 1.5 h with the anti-CypA (1:100), anti-D8L (1:50), anti-α-tubulin (1:50) primary antibodies. The cells were then rinsed with P-BSA and incubated for 2 h with Cy3-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch) diluted 1:800 and/or Alexa Fluor 488 goat anti-mouse IgG (Molecular Probes) diluted 1:500. Cellular and viral DNAs were stained with 1 μg of 4′,6′-diamidino-2-phenylindole (DAPI) per ml for 1 min. The coverslips were washed in P-BSA and then in PBS, mounted in 0.2 M n-propyl gallate, and analyzed with a Zeiss conventional fluorescence microscope (Axiophot). Confocal microscopy was also performed with Zeiss confocal laser scanning microscopes (CLSM 310 and CLSM 510). Each channel was collected separately, and then the channels were merged.

Electron microscopy. (i) Immunogold labeling of thin sections of VV-infected cells.

BSC-40 cells were infected with VV (MOI = 10), and at 20 h p.i., the monolayers were processed for immunolocalization as previously described (40). Briefly, the cells were washed in PBS and fixed with a solution containing 0.4% glutaraldehyde type I, 4% paraformaldehyde, 1% picric acid, 5 mM CaCl₂, and 0.5% sucrose in PHEM buffer for 10 min at room temperature and for 50 min at 4°C. The cells were scraped, washed in PBS, and incubated with 50 mM NH₄Cl for 30 min on ice. After washing in PBS, the cell pellets were dehydrated in a graded ice-cold ethanol series (30 to 100%) and infiltrated with Unicryl (British Biocell) at −20°C under UV illumination. Ultrathin sections on 400-mesh nickel grids were immunolabeled as follows. The grids were treated with 200 mM NH₄Cl in PBS for 20 min, rinsed in PBS, and after incubation with blocking solution (1.5% BSA, 0.5% fish skin gelatin, 0.01% Tween 20 in PBS [pH 7.2]) for 1 h, anti-CypA antibody diluted in blocking solution was added and the mixture was incubated overnight at 4°C. The grids were washed in blocking solution and then incubated for 1 h with 10- or 15-nm colloidal gold-conjugated goat anti-rabbit IgG (BBInternational). The material was rinsed in deionized water, stained with uranyl acetate and lead citrate, and observed in a Zeiss EM 900 transmission electron microscope.

(ii) Immunogold labeling and negative staining of purified IMV particles.

Purified IMV particles (1.5 × 10⁸) were adsorbed to glow-discharged, Formvar-coated 400-mesh nickel grids for 15 min. Subsequently, the grids were incubated or not with 0.5% NP-40-50 mM DTT in 10 mM Tris-HCl (pH 7.5) for 45 min at 37°C. Alternatively, the virus-covered grids were fixed with 4% paraformaldehyde in PHEM buffer for 45 min and then treated or not with 0.2% Triton X-100 in PBS for 15 min at room temperature. After both detergent treatments, the grids were incubated with 20 mM glycine in PBS for 20 min and blocked and immunolabeled as described above, except for the primary antibodies, which were added for 1 h of incubation. After a final wash in PBS, samples were fixed with 2.5% glutaraldehyde in PBS for 15 min and rinsed in deionized water. Virus particles were negatively stained with 5% uranyl acetate for 5 min and analyzed in a Zeiss EM 900 transmission electron microscope.

CypA accumulation and synthesis in VV-infected BSC-40 cells.

Previous work indicated the involvement of CypA in the antiviral effect of CsA on VV replication. This suggestion was based on the analysis of a panel of CsA analogs with different levels of affinity to CypA. Analogs with strong CypA-binding activity also exhibited a strong antiviral effect. These data led to the proposition that CypA could be required for progression of the VV replicative cycle (16). To gain insight into this issue, we first evaluated the expression of CypA throughout VV infection. BSC-40 cells were infected with VV (MOI = 10) and collected at different times p.i. for preparation of postnuclear extracts. Equal amounts of protein were analyzed by Western blotting. As shown in Fig. 1A, the accumulation of CypA remained practically invariable from early to late times of infection, with levels comparable to those of mock-infected cells. Minor variations in CypA expression detected at some time points were also noticed in α-tubulin expression, which was used as an endogenous control in double-labeled blots. A parallel immunoblot with anti-F11L (early viral protein) and anti-A18R (early/late viral protein) antibodies was performed as a positive control for the progression of VV infection (Fig. 1A).

To investigate CypA synthesis, mock- and VV-infected BSC-40 monolayers were metabolically pulse-labeled at various times p.i. and postnuclear extracts were prepared. The ³⁵S-labeled extracts were loaded onto 8′A-Cs affinity columns that specifically retain CsA-binding proteins, in particular, the major Cyp isoform CypA (16). Proteins specifically bound to 8′A-Cs columns were eluted with an excess of CsA and analyzed by SDS-PAGE and autoradiography. Coomassie blue staining of the gel confirmed the specificity of 8′A-Cs column binding for CypA, which was not retained in control columns (Fig. 1B) as previously reported (16). Consistent with the results shown in Fig. 1A, VV-infected cells accumulated CypA to levels of mock-infected cells (Fig. 1B). However, the autoradiogram analysis revealed a drastic inhibition of CypA synthesis at late times of infection (Fig. 1C). At 2 h p.i., CypA was synthesized to control levels, but after 8 h of infection, only a faint band could be seen (Fig. 1C). This result was not totally unexpected because infection of cells with VV is known to induce a severe shutoff of host protein synthesis (17, 41). This effect is observed in Fig. 1D, which shows the diminished background of host proteins synthesized during late times of infection and eluted from the columns in the flowthrough fractions.

The data in Fig. 1A and C indicate that VV infection may alter CypA stability. To investigate this possibility, we measured CypA turnover in mock- and VV-infected cells by metabolic labeling of the cells at 2 h p.i. (pulse) and chasing for the indicated times with unlabeled medium. The extracts were loaded onto 8′A-Cs affinity columns, and the CypA eluted with CsA was analyzed by SDS-PAGE and autoradiography (Fig. 1E). To quantify the ³⁵S-labeled CypA at each time point, we performed densitometric analyses of the gels and autoradiograms (Fig. 1F). To properly express the amount of ³⁵S-labeled CypA, the values obtained for each lane of the autoradiogram were first corrected by the total amount of CypA in the corresponding lane of the Coomassie blue-stained gel (Fig. 1E). The values presented in Fig. 1F were calculated as described in Materials and Methods and are expressed with reference to the pulse time point (baseline). Figure 1F shows that in mock-infected cells, the [³⁵S]methionine-labeled CypA was progressively degraded, reaching 50% of the baseline value after an 18-h chase. In contrast, when the cells were infected with VV, the half-life of pulse-labeled CypA was markedly extended, with 100% of the labeled protein recovered after an 18-h chase.

Subcellular localization of CypA in VV-infected cells.

To further investigate the possible involvement of CypA in the VV life cycle, we next examined the distribution pattern of CypA in mock- and VV-infected BSC-40 cells by confocal microscopy (Fig. 2). Mock-infected cells presented a slight punctate staining superimposed upon a fainter diffuse background dispersed throughout the cytoplasm, in agreement with the cytosolic localization of CypA (28, 37). Notably, a weaker CypA immunoreactivity was also found in the nucleus (Fig. 2A). Similar nuclear labeling has also been described by others (37, 47). Infection of BSC-40 cells with VV for 20 h induced a striking reorganization of the distribution pattern of CypA in the cytoplasm. CypA was mostly relocated to regions juxtaposed to the nuclear periphery (Fig. 2E, arrows), and this strong anti-CypA labeling fully colocalized with viral factories, as defined by DNA staining (Fig. 2F, asterisks). Colocalization was clearly evident when the two images were merged (Fig. 2G). It is important to note that the complete coalignment of CypA with VV factories was consistently detected in distinct experiments, corresponding to 96.3% of the cells counted (n = 600) in random fields.

To gain additional understanding of the redistribution of CypA during VV infection, a time course experiment was performed. BSC-40 cells were infected with VV, and at the indicated times p.i., the monolayers were processed for indirect immunofluorescence microscopy as described in Materials and Methods. As shown in Fig. 3A, no changes in the CypA distribution pattern were visible at 2 h p.i. compared to that in mock-infected cells (Fig. 3C). Also, cytoplasmic viral factories were not revealed by DAPI staining (Fig. 3B). As the infection proceeded, CypA distribution shifted to a mainly punctate, granular pattern, with the appearance of intensely labeled spots (Fig. 3D and G, arrows). Virosomes were clearly distinguished by DAPI staining, initially as round structures at 4 h p.i. (Fig. 3E, asterisk), evolving to irregularly shaped bodies later in infection (Fig. 3H, asterisk). The granular structures intensely decorated with CypA were dispersed within the cytoplasm and did not colocalize with viral factories at 4 and 6 h p.i. (Fig. 3F and I, arrows), although most of these spotty structures were in close proximity to the virosomes. At 12 h p.i., full colocalization of CypA and viral factories was clearly evident (Fig. 3J to L), as observed at 20 h p.i. (shown in Fig. 2).

The immunosuppressant CsA was reported to severely inhibit VV progeny production (15), and this effect was correlated with its ability to bind CypA (16). Therefore, we were most interested in investigating the effect of CsA on CypA rearrangement during infection. As shown in Fig. 3M, in the presence of 35 μM CsA, CypA no longer colocalized with viral factories after 12 h of infection. In contrast, staining was dispersed throughout the cytoplasm, except for the virosome region, which appeared to be excluded from CypA labeling. Some labeling outlined viral factories, suggesting a halo-type pattern surrounding the virosomes in the presence of CsA (Fig. 3M and O, open arrows). Additionally, we observed that DNA factories were smaller in CsA-treated cells (Fig. 3N, asterisk) than in untreated cells (Fig. 3K), which is consistent with previous reports that viral DNA replication is inhibited in VV-infected cells treated with CsA (15, 16). Under these conditions, virus progeny production was inhibited to a level 87.4% lower than that in untreated VV-infected BSC-40 cells. Interestingly, although CypA was mainly distributed throughout the cytoplasm, granular staining was still detected in VV-infected cells treated with CsA, although less frequently than in untreated cells at early stages of infection. This pattern of CypA was not visible in mock-infected cells treated with CsA (Fig. 3P) or not treated with CsA (Fig. 3C), indicating that CsA inhibited the relocation of CypA to viral factories induced by VV infection but could not restore the regular distribution pattern observed in uninfected cells.

Specificity of CypA redistribution in VV-infected cells.

To evaluate the specificity of CypA rearrangement and exclude a general effect of virus infection on cellular proteins under our assay conditions, we analyzed the distribution pattern of another abundant cellular protein, α-tubulin, in VV-infected cells. As previously described (43), the microtubule cytoskeleton was dramatically reorganized during infection (Fig. 4C), but its distribution did not colocalize with the virosomes (Fig. 4E), in contrast to the CypA pattern (Fig. 4D). As a control, the distribution patterns of α-tubulin and CypA in mock-infected cells are shown in Fig. 4A and B, respectively. CypA relocation to viral DNA factories was also detected in mouse embryo fibroblasts (Fig. 4G and H) and HEp-2, RK13, and HeLa cells (data not shown) infected with VV for 20 h.

Alteration of the CypA staining pattern requires VV early gene expression.

In an attempt to relate CypA redistribution to the stages of VV gene expression, we next examined whether the replication of viral DNA and all subsequent postreplicative events are required for CypA rearrangement. BSC-40 cells were infected with VV for 10 h in the presence of 40 μg of AraC per ml to block viral DNA synthesis. As expected, under this condition, the formation of viral DNA factories was completely abolished (Fig. 5B). Nevertheless, CypA changed its labeling pattern, displaying punctate staining dispersed all over the cytoplasm with brightly labeled granular structures. No redistribution of CypA was observed (Fig. 5A). This pattern was already observed at 6 h posttreatment (data not shown) and was quite similar to that observed in untreated cells at early times p.i., as shown in Fig. 3D and G. Infection of the monolayers for 6 h in the presence of 100 μg of CHX per ml did not alter the subcellular localization or staining pattern of CypA (Fig. 5E) compared to that in mock-infected cells (Fig. 3C), indicating the requirement of de novo protein synthesis to launch CypA rearrangement. Together, these data suggest that the occurrence of the first discernible changes in the CypA distribution pattern during VV infection requires only the early stage of the cycle. The subsequent redistribution of CypA to the virosomes appears to depend on the next steps of infection.

Redistribution of CypA to the virosomes requires viral DNA replication and postreplicative protein synthesis.

To further examine whether DNA replication and/or postreplicative protein synthesis are required for the subsequent colocalization of CypA with viral factories, we used two experimental approaches. Initially, we performed the assay outlined in Fig. 6A. BSC-40 cells were infected with VV in the presence of 10 mM HU to block DNA replication, allowing the early phase of infection to occur. After 6 h of infection, the HU-containing medium was removed to release viral DNA synthesis at the same time that postreplicative protein synthesis was inhibited by the addition of 100 μg of CHX per ml to the monolayers and incubation for another 6 h. After this period, CypA localization was analyzed by indirect immunofluorescence microscopy. Confirming previous data obtained with AraC, virus DNA replication was not required for the initial changes in the staining pattern of CypA represented by the appearance of intense punctate labeling with dense granular structures at 6 h p.i. in the presence of HU (Fig. 6B). As expected, removal of HU released the replication of viral DNA (Fig. 6F and J) in the presence or absence of postreplicative protein synthesis, which was monitored by expression of the IMV late protein D8L (Fig. 6H and L). In agreement with previous observations made by others, the size of DNA factories did not increase over time when DNA replication occurred in the presence of CHX (Fig. 6J) (2, 20). When virus postreplicative protein synthesis occurred, CypA was recruited to DNA factories and colocalized completely (Fig. 6E and G). Conversely, in the absence of postreplicative protein synthesis, CypA remained mostly scattered throughout the cytoplasm with bright spots dispersed randomly. On limited occasions, we detected CypA colocalization with a few virosomes per cell, corresponding to 14.9% of the cells (n = 464) counted in random fields (Fig. 6I and K). This indicates that virus postreplicative protein synthesis is necessary for full redistribution of CypA to viral factories. Nevertheless, these observations should be interpreted with caution since the use of metabolic inhibitors affects both viral and cellular functions.

As a second experimental strategy to investigate the involvement of postreplicative protein synthesis in CypA relocation to the virosomes, we performed experiments with the temperature-sensitive VV mutant ts53. This mutant has been previously characterized by Condit and coworkers as defective in late protein synthesis but fully competent in DNA replication (30). The mutation was mapped to the 147-kDa subunit of the viral RNA polymerase (54). Infection of BSC-40 cells with either WT or ts53 VV was performed at 31°C (permissive temperature) or 40°C (nonpermissive temperature), and after 12 h of infection, the cells were processed for immunofluorescence assay as described in Materials and Methods. At 31°C, intense colocalization of CypA with the virosomes was observed in ts53-infected cells (Fig. 7I to K) and the pattern of CypA redistribution to viral factories was quite similar to that of WT infections at both 31°C (Fig. 7A to C) and 40°C (Fig. 7E to G). As a control, the D8L protein was used to examine the synthesis of late proteins (Fig. 7D, H, and L). Consistent with previous data (30), cells infected with ts53 at the nonpermissive temperature exhibited large, round virosomes (Fig. 7N) and were unable to synthesize virus late polypeptides (Fig. 7P). Under this condition, CypA remained mostly dispersed all through the cytoplasm (Fig. 7M, open arrow) and its redistribution to the virosomes was scarcely observed (Fig. 7M, white arrow). In some of these few positive cases, only partial colocalization was detected, particularly at the periphery of the virosomes (Fig. 7M, thin arrows). Table 1 shows that the coalignment of CypA with viral factories was distinguished in 19.3% of the ts53-infected cells at 40°C, in contrast to more than 75% of the cells displaying full colocalization at 31°C.

To evaluate whether the redistribution of CypA to viral factories would recover with the resumption of postreplicative protein synthesis, the cells were infected nonpermissively with ts53 for 12 h and then shifted to 31°C for 2 or 4 h. In agreement with Hooda-Dhingra et al. (30), the temperature downshift allowed the restoration of virus late protein synthesis detected by anti-D8L staining (Fig. 7T). Simultaneously, full colocalization of CypA with the virosomes was progressively evident (Fig. 7Q to S). Quantitation of the coalignment reinforced this observation and revealed that at 2 h after the temperature downshift, 48.2% of the cells displayed a clear redistribution of CypA to viral factories. Control levels were reached within 4 h after incubation at 31°C (Table 1). Together, these results suggest that postreplicative protein synthesis is important for accurate colocalization of CypA with virosomes.

Association of CypA with virosomes and virus particles in infected cells.

The immunofluorescence assays described above clearly indicated the specific association of CypA with viral factories during VV infection. To investigate this interaction in greater detail, immunoelectron microscopy was performed on ultrathin sections of VV-infected cells that were incubated with anti-CypA antibody, followed by immunogold labeling. As shown in Fig. 8, strong labeling of CypA was detected in association with an amorphous electron-dense material located in the proximity of IMV particles and assembly intermediates (Fig. 8A, B, and D). These dense labeled masses were not observed in areas of the cytoplasm devoid of VV assembly forms or in mock-infected cells (Fig. 8C). Significant labeling was also seen surrounding and within the viroplasm (Fig. 8E). Remarkably, gold particles were consistently observed decorating distinct viral forms of VV morphogenesis. CypA labeling was detected in close association with the viroplasm underlying the concave side of the viral crescents (Fig. 8F to H). Spherical immature particles (IV) and IVs with nucleoid were mainly labeled in the internal region and also surrounding the electron-dense nucleoid (Fig. 8I and J). In mature particles (IMV and intracellular enveloped virus), CypA labeling was primarily found associated with the core and in the space between the core and the IMV envelope (Fig. 8K to N). At all stages of virus maturation, gold grains were rarely detected in association with the particle membranes. We also observed that mature forms were labeled less extensively than IVs, probably because of reduced access of the antibodies to CypA epitopes in the densely packed cores. Similar observations have been reported by others who used immunoelectron microscopy assays to evaluate the localization of distinct VV proteins, including the internal proteins A4L (10) and A30L (52). Also, it is interesting that the CypA labeling associated with the immature and mature forms of the virus was essentially distributed unevenly and in clusters inside the particles. This distribution pattern was detected in 86.7% of the labeled virus particles (n = 117) counted in random fields. The asymmetrical arrangement could also contribute to the weak detection of CypA in some section planes of mature virions.

CypA is incorporated in purified virions and packaged in viral cores.

To further examine the association of CypA with VV particles, intracellular virions were purified by three successive sucrose gradient centrifugations and fractions from the third gradient were collected and subjected to SDS-PAGE, followed by Western blotting as already described. Figure 9C shows that CypA detection coincided with the virus-containing fractions, as determined by OD measurement (Fig. 9A) and immunoblotting of VV proteins with antiserum raised against virion structural proteins (Fig. 9B). The copurification of CypA with VV particles was specific since we did not detect other cellular proteins, such as α-tubulin or Hsp70, in association with the virion-containing fractions. To estimate the amount of CypA present in the virus particles, we analyzed the virion proteins by SDS-PAGE, followed by Western blotting. Parallel lanes of the blot contained increasing amounts (10 to 150 μg) of purified CypA that were used as reference samples in the evaluation. Densitometric analysis of the CypA bands was performed, and taking into account the CypA molecular mass, we determined that approximately 156 ± 0.9 CypA molecules were present per VV particle (two separate assays run in duplicate).

To determine the localization of CypA within the virion, we used three different approaches. First, we treated the purified VV particles with 0.5% NP-40 in the presence or absence of 50 mM DTT and then separated the core and envelope fractions by centrifugation as described in Materials and Methods. The proteins from each fraction were analyzed by SDS-PAGE, followed by silver staining (Fig. 10A) or immunoblotting. As shown in Fig. 10B, CypA was not extracted by the NP-40 treatment, indicating a core localization, and was only partially released from the virions when treated with the detergent plus the reducing agent. As a control, the immunoblots were stripped and reprobed with antibodies to the viral core proteins A18R, D12L, D1R, and A12L. Consistent with their core location, these polypeptides remained associated with the insoluble fraction after NP-40 treatment. H3L was used as a control for IMV envelope protein and was considerably extracted from VV particles treated with NP-40, as reported before (12). The core fraction was further treated with deoxycholate-containing buffer to separate soluble and insoluble core proteins. Similarly to A18R, D12L, and D1R, CypA was also found entirely associated with the insoluble core fraction after treatment of cores with deoxycholate. In contrast, the core protein A12L was completely extracted to the soluble core fraction.

We also investigated the subviral location of CypA after treatment of purified virions with proteinase K and chymotrypsin, followed by protein analysis by Western blotting. As shown in Fig. 10C, even the highest concentration of proteinase K or chymotrypsin treatment did not produce digestion of CypA, consistent with its association to the viral core. A similar pattern was observed with other core proteins, such as A18R and D12L. As a control for membrane proteins susceptible to protease digestion, the blots were stripped and reprobed with anti-H3L (proteinase K treatment) and anti-D8L (chymotrypsin treatment). As expected, both proteins were digested into smaller products.

As a third strategy to verify the internal location of CypA in virions, we performed immunolabeling experiments with purified VV particles. After adsorption to nickel grids, the IMV particles were permeabilized with 0.5% Triton X-100 or not permeabilized and then incubated with anti-CypA, followed by a secondary antibody conjugated to 10-nm colloidal gold. Alternatively, the virus-covered grids were incubated with 0.5% NP-40-50 mM DTT to remove the viral membranes, as described in Materials and Methods. Figure 11 shows that intact untreated virus particles were not labeled significantly with anti-CypA (Fig. 11A and C). As expected, untreated IMV particles were labeled with antibodies to H3L and D8L, both envelope proteins (Fig. 11E). After the detergent treatments, immunogold labeling of IMV particles with anti-CypA was detected, implying that CypA is not exposed to the virion surface (Fig. 11B and D). As a control, permeabilized virus was also labeled with anti-D12L (Fig. 11G), a core protein, whereas the intact virus showed no significant labeling (Fig. 11F). Together, these data indicate that CypA is packaged within the virions and is specifically located in the viral core.