Genetic Analysis of the Vaccinia Virus I6 Telomere-Binding Protein Uncovers a Key Role in Genome Encapsidation

Materials.

Restriction endonucleases, Taq DNA polymerase, T4 DNA ligase, calf intestinal alkaline phosphatase, pancreatic RNase, Escherichia coli DNA polymerase I, DNase I, and DNA molecular weight standards were purchased from either Roche Molecular Biochemicals (Indianapolis, Ind.) or New England Biolabs (Beverly, Mass.). [³⁵S]Met and ³²P-labeled nucleoside triphosphates were purchased from PerkinElmer Life Sciences (Boston, Mass.). Lipofectamine Plus, geneticin (G418 sulfate), and ¹⁴C-labeled protein molecular weight markers were purchased from Invitrogen (Carlsbad, Calif.). Rifampin (RIF) and 1-β-d-arabinofuranosylcytosine (araC) were obtained from Sigma (St. Louis, Mo.). Isatin-β-thiosemicarbazone (IBT) was obtained from Pfaltz and Bauer (Waterbury, Conn.). Oligonucleotides were purchased from IDT (Coralville, Iowa).

Cells and viruses.

BSC40 African green monkey kidney cells were maintained as monolayer cultures in Dulbecco modified Eagle medium (DMEM) (Invitrogen) containing 5% fetal bovine serum (FBS). Stocks of wild-type (wt) vaccinia virus (WR strain) and the viruses described herein, which contained mutant alleles of the I6 gene, were prepared by ultracentrifugation of cytoplasmic extracts through a 36% sucrose cushion.

Site-directed mutagenesis and cloning of the I6 gene into pUC/Neo.

Eight regions of the I6 gene were chosen for clustered charge-to-alanine mutagenesis (Fig. 1) (7, 32). Mutations were generated by overlap PCR using the HindIII I fragment of the vaccinia virus genome as the template. In the first round of PCR, one reaction was performed using an upstream primer (U) and a downstream primer carrying the engineered nucleotide substitutions (#mut-3); a second reaction was performed using an upstream primer carrying the desired substitutions (#mut-5) and a downstream primer (D). The two PCR products from this first round of amplification overlapped by 17 bp. A mixture of the two products served as the template for a second round of PCR performed with primers U and D. The final PCR product contained the desired mutations flanked by 300 bp of upstream and downstream sequences. Primers U and D introduced BamHI recognition sites at the termini of the amplified fragments. The final PCR products were gel purified, digested with BamHI, and cloned into pUC/Neo DNA that had been previously digested with BamH1 and treated with calf intestinal alkaline phosphatase. This plasmid carries the neomycin resistance gene (Neo) under the control of the constitutive viral p7.5 promoter (21, 24). The constructs were subjected to automated DNA sequencing to ensure the presence of the desired nucleotide substitutions and the absence of any additional mutations. The primers used are listed in Table 1.

Transient dominant selection and isolation of I6 recombinant viruses.

Transient dominant selection was used to replace the endogenous I6 gene with the mutant allele (10, 21, 24, 32). Dishes (35-mm diameter) of BCS40 cells were infected with wt virus at a multiplicity of infection (MOI) of 0.03 PFU/cell at 37°C. At 3 h postinfection (hpi), 3.5 μg of the desired pUC/Neo-I6 plasmid was introduced into cells by transfection using Lipofectamine Plus reagent (Invitrogen) and cells were transferred to 32°C. 32°C was utilized for all subsequent steps in the generation of the recombinant viruses. At 16 hpi, geneticin (G418 sulfate) was added to a final concentration of 3 mg per ml to select for the spread and replication of virus that had incorporated the plasmid into the genome. Cells were harvested 2 days after infection, and virus was released by three cycles of freezing and thawing and two 15-s bursts of sonication. After two rounds of plaque purification in the presence of geneticin, incorporation of the plasmid into the drug-resistant viral isolates was confirmed by PCR using primers specific for the Neo gene. These “parental” viruses have the entire plasmid incorporated into the I6 locus and therefore have tandem repeats of a portion of the I5, I6, and I7 regions flanking the Neo^r gene. Drug selection was then removed, and several rounds of plaque purification were performed to enable recombination to occur within these repeats. Such recombination leads to the loss of the Neo^r gene and the retention of either the wt or mutant allele of I6. The loss of the Neo^r gene was monitored by PCR. The I6 allele was amplified by PCR from those viruses shown to have undergone recombinational resolution, and automated DNA sequence analysis was used to distinguish viruses carrying the desired mutant I6 alleles from wt isolates. The appropriate plaques were expanded, and purified viral stocks (designated vI6-1, vI6-2, vI6-4, vI6-5, vI6-6, vI6-9, vI6-12, and vI6-13) were prepared by ultracentrifugation of cytoplasmic extracts through a 36% sucrose cushion.

Determination of 24-h viral yields.

Dishes (35-mm-diameter) of confluent BSC40 cells were infected at an MOI of 2 with wt or vI6 viruses and incubated at either 32 or 40°C. At 24 hpi, cells were harvested by scraping, collected by centrifugation, resuspended in phosphate-buffered saline (PBS) (140 mM NaCl, 2 mM KCl, 10 mM Na₂HPO₄, 1 mM KH₂PO₄ [pH 7.4]), and disrupted by three cycles of freezing and thawing and two 15-s bursts of sonication. Viral yields were determined by plaque assays performed at 32°C on BSC40 monolayers.

Marker rescue.

Dishes (35-mm-diameter) of BSC40 cells were infected with tsI6-12 at an MOI of 0.006 and incubated at 32°C. At 3 hpi, 5 μg of linearized plasmid DNA was introduced into cells with Lipofectamine Plus reagent. Plasmids contained no insert, the wt I6 open reading frame (ORF), the genomic HindIII I fragment, or the genomic HindIII G fragment (negative control). At 6 hpi, dishes were fed with fresh DMEM-5%FBS and transferred to 40°C. At 2 dpi, cells were harvested and the yield of temperature-insensitive virus was determined by titration at 40°C.

Analysis of viral protein synthesis.

Confluent BSC40 cells were infected with wt virus or tsI6-12 at an MOI of 10 and incubated at 32 or 40°C. At the appropriate times, cells were rinsed with prewarmed methionine-free DMEM, incubated for 45 min in the same medium supplemented with 100 μCi of [³⁵S]Met/ml, and then harvested. Cells were harvested at 2, 3.5, 5, 6.5, and 8 hpi; the proteins within whole-cell lysates were resolved by electrophoresis on a sodium dodecyl sulfate (SDS)-12% polyacrylamide gel and then visualized by autoradiography.

Analysis of the proteolytic processing of viral structural proteins.

Cells were infected with wt virus or tsI6-12 at an MOI of 5 in the presence or absence of RIF (100 μg/ml) and incubated at 32 or 40°C. At 8 hpi, cells were labeled as described above for 45 min; samples were either harvested immediately (pulse) or rinsed and fed with prewarmed DMEM-5%FBS and harvested at 24 hpi (chase).

Analysis of viral DNA synthesis by Southern dot blot hybridization.

Dishes (35-mm-diameter) of confluent BSC40 cells were infected with wt virus or tsI6-12 at an MOI of 5 and incubated at 32 or 40°C. At 3, 6, 9, 12, and 24 hpi, cultures were harvested and analyzed in duplicate by dot blot DNA hybridization as previously described (32). Viral DNA was detected by hybridization with a radiolabeled probe representing the HindIII E and HindIII G fragments of the vaccinia virus genome. Data were quantitated by phosphorimager analysis and plotted using Sigma Plot software.

Genome resolution assay.

Cells were infected with wt virus or tsI6-12 at an MOI of 2 and incubated at 32 or 40°C. As controls, cells were also infected with wt virus (MOI of 2) in the presence of 60 μM IBT or with vROG8 virus (a recombinant in which expression of the G8 gene is dependent upon inclusion of IPTG in the culture medium) (41) at an MOI of 15. At 18 hpi, cells were harvested and cytoplasmic DNA was prepared, digested with BstEII, and resolved on a Tris-acetate-EDTA-agarose gel. The DNA was transferred to a ZetaProbe membrane, and Southern blot hybridization was performed using a radiolabeled probe representing a 1.2-kb PvuII-EcoRI fragment released from the pSV9 plasmid (25, 26). This probe hybridizes to the genomic telomeres and adjacent sequences. The blot was then visualized by autoradiography.

Electron microscopy.

Dishes (60-mm-diameter) of BSC40 cells were infected with tsI6-12 virus at an MOI of 2 and incubated at 32 or 40°C. At 17 hpi, cells were rinsed with PBS and fixed in situ with 1% glutaraldehyde-0.1 M PBS (pH 7.4) on ice for 30 min. Cells were scraped gently, collected by centrifugation (800 × g for 10 min), and then processed for conventional electron microscopy as described previously (7, 36). Purified viral particles were concentrated by sedimentation (13,000 × g for 5 min) and then processed as described above. Samples were examined on a Hitachi H-600 transmission electron microscope.

Purification of virus particles.

Four 15-cm-diameter dishes of BSC40 cells were infected with either wt virus or tsI6-12 at an MOI of 2 and kept at 32 or 40°C. At 24 hpi, cells were harvested and disrupted by Dounce homogenization. Virus particles were purified from cytoplasmic extracts by ultracentrifugation through a 36% sucrose cushion (43,000 × g for 80 min) followed by banding on a 25 to 40% sucrose gradient (6,000 × g for 45 min). After the gradients were photographed, 20 fractions (250 μl each) were collected beginning at the bottom of the tube. Aliquots from each fraction were solubilized with 0.25% NP-40-0.25% deoxycholate-250 mM NaCl-0.5 mM dithiothreitol, and the protein content was quantitated by a Bradford assay (Bio-Rad, Hercules, Calif.). Fractions containing detectable amounts of protein which corresponded to those containing a light-scattering band were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) and silver staining.

Immunoblot analysis.

Protein samples were resolved by SDS-PAGE and electrophoretically transferred to nitrocellulose membranes in CAPS [3-(cyclohexylamino)-1-propanesulfonic acid] buffer (10 mM CAPS [pH 11.3], 10% methanol). Membranes were incubated with sera generated in our laboratory which react with the L4, F18, and I1 proteins (21, 24) or with sera directed against the I7 and L1 proteins, which were generous gifts from S. Shuman (Sloan-Kettering Institute, New York, N.Y.) and D. Hruby (Oregon State University, Corvallis, Oreg.), respectively. Horseradish-peroxidase-conjugated goat anti-rabbit secondary antibody was used, and immunoreactive proteins were visualized after chemiluminescent development.

Southern dot blot analysis of purified viral particles.

Fractions 7 to 18 (5 μl each) from the 25 to 40% sucrose gradient (see above) were diluted in 50 μl of water, boiled for 5 min, and applied (using a Bio-Dot microfiltration apparatus [Bio-Rad]) to a hydrated ZetaProbe membrane (Bio-Rad). Samples were denatured and renatured in situ, and the blot was then subjected to Southern dot blot hybridization using 10⁶ cpm of radiolabeled probe/ml representing the vaccinia virus HindIII E and HindIII F genomic fragments. Data were analyzed on a phosphorimager.

Computer analysis.

Autoradiographic films and electron micrographic negatives were scanned using a SAPHIR scanner (Linotype-Hell Co., Hauppauge, N.Y.). Plaque assays were photographed using an AlphaImager (Alpha Innotech, San Leandro, Calif.) documentation system. Data for the Southern dot blot analyses were acquired using a Storm PhosphorImager (Molecular Dynamics, Sunnydale, Calif.) and quantitated using ImageQuant software (Molecular Dynamics). Images were adjusted with Adobe Photoshop 5.0 software (Adobe Systems Inc., San Jose, Calif.), data were plotted using SigmaPlot 7.1 software (SSPS, Chicago, Ill.), and final figures were assembled and labeled with Canvas 8.0 software (Deneba Systems, Miami, Fla.).

Replacement of the endogenous I6 allele with alleles carrying clustered charge-to-alanine substitutions.

A total of 13 clusters of charged residues were identified in the predicted amino acid sequence of the I6 protein (Fig. 1). Of these, eight clusters spread throughout the I6 ORF were chosen for alteration. Overlap PCR was used to change the codons for these residues (shaded) in regions 1, 2, 4, 5, 6, 9, 12, and 13 so that they would instead encode alanine. Mutant alleles were then inserted into the pUC/Neo plasmid, which contains the neomycin resistance gene (Neo^r) under the control of a constitutive vaccinia virus promoter (24), to facilitate their placement within the endogenous I6 locus of the viral genome.

Allelic replacement of the endogenous (wt) I6 gene was accomplished with the use of transient dominant selection (10, 21, 24). pUC/Neo plasmids containing the mutant I6 alleles were introduced into cells infected with wt virus, and following infection-transfection, the cells were maintained at 32°C in the presence of G418 for 2 days. Viruses that had incorporated the plasmid into their genomes via homologous recombination were selected by two rounds of plaque purification performed in the presence of G418. Insertion of the plasmid was confirmed by screening viral isolates for the presence of the Neo^r gene by PCR. These G418^r viruses contain an unstable, tandem duplication of the I6 region flanking the Neo^r gene. When selective pressure is removed by performing additional rounds of plaque purification in the absence of G418, the duplicated sequences readily undergo homologous recombination, resulting in the loss of the Neo^r gene. This resolution reaction was monitored by PCR; viruses in which the resolved genome retained the mutant, rather than the wt, I6 allele were identified by DNA sequence analysis. All eight recombinant viruses, vI6-1, vI6-2, vI6-4, vI6-5, vI6-6, vI6-9, vI6-12, and vI6-13, were successfully isolated at 32°C. Thus, each of the alleles generated could support virus viability at this permissive temperature.

Phenotypic analysis: several I6 alleles affect plaque size and/or viral yield.

The eight viruses generated were tested for their ability to form plaques at 32 and 40°C (Fig. 2). All mutant viruses were able to form plaques at 32°C, although the plaques observed for vI6-1, vI6-2, and vI6-13 were somewhat smaller than those formed by wt virus. The viruses showed distinct differences at 40°C. vI6-4, vI6-5, vI6-6, and vI6-9 formed large plaques comparable in size to those formed by wt virus, whereas vI6-1, vI6-2, vI6-12, and vI6-13 formed smaller plaques. This temperature-dependent decrease in plaque size might reflect a reduction in infectious viral yield during the first and subsequent rounds of infection or might instead reflect a decrease in the ability of the mutant virus to spread to adjacent cells after the first round of infection.

To quantitate the total yield of infectious virus from a single round of infection, BSC40 cells were infected with wt or mutant viruses at an MOI of 2 and maintained at 32 or 40°C for 24 h. As shown in Fig. 3A, wt infections produced similar amounts of infectious virus at both temperatures; similar results were obtained for vI6-4, vI6-5, vI6-6, and vI6-9. vI6-1, vI6-2, vI6-12, and vI6-13 were ts. For vI6-1, vI6-2, and vI6-13, the viral yield at 32°C was 6- to 8-fold lower than that obtained from wt infections performed in parallel and 30- to 40-fold lower than that obtained from wt infections at 40°C. vI6-12 exhibited the tightest ts phenotype; the yield at 32°C was comparable to that obtained with wt virus, whereas a 100-fold reduction in viral yield was seen at 40°C, the nonpermissive temperature. This virus, denoted as tsI6-12, was chosen for further characterization, as described below.

Surprisingly, tsI6-12, which shows a 2-log reduction in 24-h viral yield at 40°C, was still able to form small plaques at this temperature. We therefore monitored the time course of virus production in cells infected with tsI6-12 to more clearly distinguish between reduced and delayed virus production (Fig. 3B). In wt-infected cells, virus production was more rapid at 40 than at 32°C but reached an equivalent plateau at 24 hpi. A similar growth curve was observed for the tsI6-12 infection performed at 32°C, although the viral yield was slightly lower than that obtained for wt virus at 32°C. In contrast, cells infected nonpermissively with tsI6-12 showed an increase in viral yield from 8 to 16 hpi, after which no further virus production was seen. This result demonstrates that the production of infectious virus during a nonpermissive tsI6-12 infection is blocked rather than delayed.

Marker rescue was performed to confirm that the ts phenotype of tsI6-12 was indeed caused by the engineered mutation and was not due to an unintended, nonfortuitous change in the genome of this virus. Cells were infected at 40°C with tsI6-12 and transfected with linearized plasmid DNA with or without a wt I6 ORF insert; at 48 hpi, the yield of temperature-insensitive virus was titrated at 40°C. Plasmid DNA containing the wt I6 gene, but not empty plasmid, resulted in a significantly increased yield of virus able to form large plaques at 40°C (data not shown), verifying that the ts phenotype was caused solely by the amino acid substitutions within the I6 gene.

Neither viral DNA synthesis nor concatemer resolution is defective in nonpermissive tsI6-12 infections.

Given the affinity of the I6 protein for the viral telomeric hairpins and the importance of these structures for minichromosome replication, we hypothesized that I6 might play a role in DNA replication. To examine this possibility, we quantitated the accumulation of viral DNA in cells infected with wt virus or tsI6-12 at 32 or 40°C. Cell lysates were prepared at 3, 6, 9, 12, and 24 hpi and analyzed by Southern dot blot analysis. As shown in Fig. 4A, the profiles of DNA accumulation were comparable in all of the samples, indicating that the tsI6-12 allele does not cause a ts defect in viral DNA synthesis. We also examined the temporal profile of viral protein synthesis in cells infected with wt virus or tsI6-12 at both 32 and 40°C. Lysates were prepared from cultures metabolically labeled with ³⁵S-Met at various times postinfection, resolved by SDS-PAGE, and visualized by autoradiography; the patterns of protein expression directed by the two viruses were indistinguishable (data not shown). Because DNA synthesis is a prerequisite for the switch to intermediate and late transcription, these data confirm that the tsI6-12 allele does not affect the progression of DNA replication and also indicate that it has no impact on viral gene expression.

Next, we asked whether viral replication intermediates (concatemers) were efficiently resolved into monomeric genomes. Viral DNA was extracted from cells infected with wt or tsI6-12 virus at 32 or 40°C and analyzed by Southern blot hybridization using a probe that differentiates between concatemeric and monomeric genomes. As controls, cells were infected under conditions that compromised late gene expression and hence led to diminished concatemer resolution. These conditions are engendered by infection with wt virus in the presence of the drug IBT or with vROG, a recombinant virus in which the late-transcription factor G8 is repressed when IPTG is omitted from the culture medium (29, 41). As shown in Fig. 4B, the samples prepared from cells infected with vROG (lane 1) or with wt virus in the presence of IBT (lane 2) both accumulated unresolved concatemers (2.6-kb BstII band) whereas monomers (1.3-kb predicted band) predominated in the samples prepared from wt and tsI6-12 infections performed at either 32 or 40°C (lanes 3 to 6). We conclude that genome resolution proceeds normally during nonpermissive infections performed with tsI6-12.

Proteolytic processing of late viral core proteins is incomplete in tsI6-12.

The data described above indicate that the replication cycle of tsI6-12 proceeds normally through the biochemical processes of viral gene expression, DNA replication, and genome resolution. Therefore, the defect must lie in a subsequent stage of the viral life cycle—affecting either virion morphogenesis or virion infectivity. As an initial test of whether morphogenesis might be affected, we examined the proteolytic processing of the major core proteins. Three abundant viral structural proteins (p4a, p4b, and preL4) are proteolytically cleaved in a manner that is tightly linked to intracellular mature virus (IMV) formation (37). Cells were infected with wt virus or tsI6-12 at 32 or 40°C; as a control, cells were also infected with wt virus in the presence of RIF, a drug that inhibits viral morphogenesis (14, 28). At 8 hpi, cells were pulsed with [³⁵S]-Met for 45 min; half of the cultures were harvested immediately (Fig. 5, lanes p) and the other half were rinsed and fed with medium containing unlabeled methionine and incubated for an additional 16 h (lanes c). During wt infection, cleavage of the three viral precursors into the products 4a, 4b, and L4 was nearly complete at both temperatures (compare lanes 3 and 4 to lanes 5 and 6). In the presence of RIF, however, only minimal cleavage is observed (compare lanes 1 and 2). Processing is somewhat diminished in the tsI6mut-12 infection performed at 32°C, as evidenced by the persistence of detectable levels of uncleaved precursors in the sample harvested after the 16-h chase (compare lanes 7 and 8 to lanes 3 and 4). This diminution of processing is more pronounced at 40°C, at which temperature only ∼30% of the precursor is processed (compare lanes 9 and 10 to lanes 5 and 6). This observation implies that viral morphogenesis is compromised during nonpermissive tsI6-12 infections.

Morphogenesis is blocked at the IV-to-IMV transition during nonpermissive tsI6-12 infections.

Virus assembly proceeds through several visually distinct stages: the appearance of electron-dense regions (termed virosomes) in the cytoplasm, the formation of crescents at the virosomal periphery, the maturation of these crescents into viroplasm-containing immature virions (IV), the development of nucleoid-containing IVN, and transformation into the infectious form known as IMV (reviewed in reference 27). Transmission electron microscopy was used to investigate the progression of virion morphogenesis in cells infected with tsI6-12. In cultures infected at the permissive temperature, all of the typical hallmarks of morphogenesis were observed (data not shown). We also noted the presence of some additional, unusual structures (see below), which were more dramatic and predominant in cells infected at the nonpermissive temperature. The images shown in Fig. 6 illustrate the range of structures seen in cells infected with tsI6-12 at 40°C. Cells contained normal virosomes, crescents, and IV. However, there was a virtual absence of IVN and IMV. A few IV contained several dark internal foci but nothing comparable to a nucleoid. In other IV, the internal electron-dense material was distributed asymmetrically to one hemisphere. The most abundant and distinct structures observed were spherical, electron-dense particles. Another unusual feature was the presence of large crystalloids, often surrounded by membrane. Brick-shaped crystalloids (thought to represent viral DNA) have been documented in infected cells treated with RIF, a drug that arrests morphogenesis prior to IV formation (14). The crystalloids seen here were several times larger than those observed in the presence of RIF and also lacked a precise rectangular shape; however, the internal structure of parallel filaments was retained.

The execution point of the I6-12 allele occurs at late times of infection.

Temperature shift experiments represent classical genetic tools for the determination of the execution point of conditionally lethal mutants. To determine when the tsI6-12 allele exerts its effect during the infectious cycle in a manner that causes an arrest in virus maturation, we performed a series of shift-up and shift-down experiments. When infections were initiated at 32°C and cultures were then shifted up to 40°C at 3, 6, or 9 hpi, the 24-h yield was approximately 2 logs lower than that obtained from cells maintained at the permissive temperature for the full 24 h (Fig. 7A). When cultures were shifted to 40°C at 12 hpi, the 24-h yield was no higher than that obtained from the 32°C cultures harvested at 12 hpi. Thus, even when infections were maintained at the permissive temperature for the first 12 h of infection, the shift to 40°C blocked further production of virus. The execution point was therefore very late in the infectious cycle, during the crucial stages of virus assembly.

The converse experiment was also performed: infections were initiated at 40°C and cultures were shifted down to 32°C at various times postinfection (Fig. 7B). When infected cultures were shifted down at 3 or 6 hpi, the viral yield was equivalent to that obtained from fully permissive infections. Even when cultures were shifted down at 9 or 12 hpi, only a threefold decrease in viral yield was seen. These shift-down experiments confirm that the function of the I6 protein is most crucial at late times of infection.

We next asked whether the recovery of virus production seen in cultures shifted down to 32°C in the midst of infection depended on de novo morphogenesis and/or DNA synthesis. When RIF was added to the cultures as they were shifted down to the permissive temperature at 12 hpi, the viral yield was reduced to the levels obtained from cultures maintained at 40°C (Fig. 7C). Since RIF blocks morphogenesis at crescent formation prior to the block seen in tsI6-12 infections, we conclude that the viral structures that accumulate during nonpermissive tsI6-12 infections are dead-end products and that de novo morphogenesis is required to reverse the ts arrest. An analogous experiment was performed using araC, an inhibitor of DNA synthesis. Addition of araC to permissive infections at 12 hpi did not reduce the 24-h viral yield, indicating that the pool of DNA synthesized prior to 12 h was sufficient to support normal levels of virus production. However, when araC was added to the 40°C cultures as they were shifted to the permissive temperature at 12 hpi, the viral yield was reduced to the levels obtained from cultures maintained at 40°C. These data show that reversibility of the ts arrest is also dependent on de novo DNA synthesis and suggest that the DNA crystalloids that accumulate at 40°C cannot provide viral genomes for encapsidation into fresh virions.

Morphology, protein composition, and specific infectivity of viral particles purified from nonpermissive tsI6-12 infections.

IMVs produced during wt infections can be purified by ultracentrifugation through a 36% sucrose cushion followed by banding on a 25 to 40% sucrose gradient. We attempted to apply the same protocol to the purification of viral particles from cells infected with tsI6-12 at the nonpermissive temperature. In the tsI6-12 sample, a light-scattering band was observed at a position slightly above that seen with the wt sample (Fig. 8A). Fractions were collected from each gradient, and those containing the highest optical density were assayed for protein content and composition, infectivity (viral titer), and ultrastructural morphology. Protein concentrations were determined on fractions solubilized with mild detergent to ensure complete dissociation of viral particles; the peak tsI6-12 fractions contained one-third to one-half of the total protein obtained from the peak wt fractions. To quantitate the infectivity of the samples, monolayers were infected at 32°C with an equivalent number of particles (normalized by protein content) from the ts or wt peak fractions and plaques were counted after 48 h. The particle/PFU ratio of the tsI6-12 preparations was 80-fold higher than that of the wt preparations. The determination that the specific infectivity of tsI6-12 particles was 1.25% of that of wt particles corresponded well to our initial demonstration that the viral yield from tsI6-12-infected cultures was 2 logs lower than that obtained from wt-infected cultures (Fig. 2). When we examined pooled purified preparations by conventional transmission electron microscopy, the wt sample was shown to consist mainly of IMV whereas the tsI6-12 sample contained deformed particles corresponding to the half-filled and dense spherical particles observed in the infected cells (Fig. 8B; compare with Fig. 6).

Aliquots from the gradients corresponding to the light-scattering band were also resolved by SDS-PAGE and visualized by silver staining (Fig. 8C). wt and tsI6-12 fractions were found to have similar protein profiles once the processing status of the structural proteins was taken into account. p4a, p4b, and pre-L4 predominated in the fractions from the tsI6-12 gradient, whereas the cleaved products predominated in the fractions obtained from the wt sample. To obtain a more refined view of the protein composition of the two particle preparations, equivalent amounts of protein were resolved electrophoretically and subjected to immunoblot analysis using antisera generated against multiple viral proteins (Fig. 8D). All the proteins surveyed (L1, H5, I7, I1, F18, and L4) were present at similar levels in the wt and tsI6-12 samples, except the L4 protein was present primarily as an uncleaved precursor in the ts-I6 sample.

tsI6-12 particles are devoid of viral DNA.

The gradient fractions described above were also subjected to Southern dot blot analysis to quantitate the amount of viral DNA encapsidated in the purified particles (Fig. 9). In the wt sample, a characteristic profile of DNA content was seen, with peak levels found in fraction 12. A dramatically different result was seen for the tsI6-12 sample: no peak of DNA was found. Background levels of hybridization were detected for the whole range of fractions tested. The mutant particles, therefore, appear to be devoid of viral DNA. We conclude that at nonpermissive temperatures, the tsI6-12 mutation leads to a failure in genome encapsidation.