Previous Article | Next Article ![]()
Journal of Virology, March 2005, p. 3163-3168, Vol. 79, No. 5
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.5.3163-3168.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Janet S. May, and
Philip G. Stevenson*
Division of Virology, Department of Pathology, University of Cambridge, Cambridge, United Kingdom
Received 5 October 2004/ Accepted 1 December 2004
|
|
|---|
|
|
|---|
The human gammaherpesviruses, EBV and the Kaposi's sarcoma-associated herpesvirus (KSHV), do not readily enter the lytic cycle in vitro and exhibit very narrow species tropisms in vivo. These limitations upon analysis have made murine gammaherpesvirus 68 (MHV-68), a gammaherpesvirus isolated from free-living rodents (3), a key experimental tool for understanding in vivo gammaherpesvirus gene functions (12, 13, 15). MHV-68 is closely related to KSHV (7, 16). Moreover, despite genetic divergence in latency genes among MHV-68, KSHV, and EBV, many lytic cycle genes show a high level of conservation among all three viruses. Consequently, the study of MHV-68 lytic genes has the potential to define features of in vivo pathogenesis that are broadly relevant to both human pathogens. Here we have addressed the role of MHV-68 open reading frame 11 (ORF11), which has homologs in both KSHV (ORF11) and EBV (Raji LF2).
MHV-68 ORF11 is a late gene (2, 6, 10). KSHV ORF11 has been placed in the kinetic class of primary lytic genes (8). We identified p43, the MHV-68 ORF11 gene product, as a component of virions, both as native protein and when fused to cyan-fluorescent protein (CFP). ORF11 was dispensable for viral lytic replication in vitro but contributed to lytic replication in vivo, implying a host interaction function. It was not required for the establishment of latency, consistent with the idea that this process is essentially independent of lytic replication.
Generation of ORF11-deficient viruses. We generated two ORF11 mutants using an MHV-68 bacterial artificial chromosome (BAC) (1) (Fig. 1A). First, we ligated the complementary oligonucleotides 5'GAGTGGCAGACCCTCTAGCTAGGATCCGAATTC and 5'GAATTCGGATCCTAGCTAGAGGGTCTGCCACTC into a blunted NcoI site (genomic coordinate 23503) in a HinDIII-I genomic clone (7). This ligation maintained the overlapping 3' end of ORF10 and introduced a translational stop site as the 14th amino acid residue of ORF11, as well as diagnostic EcoRI and BamHI restriction sites (underlined). The correct insert orientation was identified by DNA sequencing. Mutant ORF11 (ORF11STOP) was subcloned into the pST76K-SR shuttle plasmid and recombined into MHV-68 BAC by established protocols (1). We caused the mutant BAC to revert (mutant REV) in a similar way, repairing the ORF11 locus with an unmutated HinDIII-I genomic clone. To make an independent mutant, we used RecE/T recombination to replace genomic coordinates 23548 to 23699 of the MHV-68 BAC with a kanamycin resistance gene (Kanr gene) flanked by flp recombinase target (FRT) sites. The Kanr gene was then excised by transient flp recombinase expression, to leave a single FRT site plus short, flanking plasmid sequences (166 bp in all). The residual insert included diagnostic EcoRI and SacI restriction sites and terminated normal ORF11 translation after 20 amino acid residues (ORF11FRT). Southern blots (Fig. 1B) confirmed the predicted genomic structure of each virus.
![]() View larger version (67K): [in a new window] |
FIG. 1. Generation of ORF11-disrupted viruses. (A) ORF11 was disrupted near its N terminus, just downstream of the overlapping ORF10 stop codon. In the ORF11STOP mutant (STOP), an oligonucleotide was inserted into an NcoI site just within ORF10 but was designed so as to reconstitute the ORF10 C terminus and then halt ORF11 translation after 13 amino acid residues (16). The oligonucleotide also introduced diagnostic BamHI and EcoRI restriction sites. In the ORF11FRT mutant (FRT), a 166-bp insert, incorporating diagnostic EcoRI and SacI restriction sites, replaced 150 bp of the ORF11 coding sequence, changed amino acid residues 21 to 29, and introduced a stop codon as the 30th residue. These mutations were nonoverlapping. (B) Southern blots confirmed the predicted structure of the ORF11STOP (STOP) and ORF11FRT (FRT) mutants plus a revertant (REV) of the ORF11STOP mutant. Viral DNA was digested with EcoRI, BamHI, or SacI as indicated, electrophoresed, and hybridized with a 32P-labeled HinDIII-I probe (genomic coordinates 21965 to 26711) according to standard protocols (4). The predicted EcoRI bands for wild-type (WT) virus were 11,370 plus 5,159 bp. In the ORF11STOP mutant, the 11,370-bp band was cut into 9,234- plus 2,136-bp fragments; in the ORF11FRT mutant, the 1,1370-bp band was cut into 9,429- plus 1,939-bp fragments. The 5,159-bp band incorporates the left-hand internal repeats, which are variably lost in the BAC without obvious consequence for viral fitness (1). Thus, this band lies between 5,159 and 4,300 bp in BAC-derived viruses. The predicted BamHI bands for wild-type virus were 4,295 plus 845 bp. In the ORF11STOP mutant, the 4,295-bp band was cut into 2,304- plus 1,911-bp fragments. The predicted SacI band for wild-type virus was 6,953 bp. Again, this band was somewhat smaller because of some loss of internal-repeat copies. In the ORF11FRT mutant, the 6,953-bp SacI fragment was cut into 4,636-bp (this band incorporates the internal repeats and so was somewhat smaller) plus 2,315-bp fragments.
|
![]() View larger version (28K): [in a new window] |
FIG. 2. Growth of ORF11-disrupted viruses in vitro and in vivo. (A) BHK-21 cells were infected at a low multiplicity of infection (0.01 PFU/cell) with wild-type, ORF11STOP, or revertant virus as indicated. The growth of virus with time was monitored by plaque assay on BHK-21 cells. (B) Growth in BHK-21 cells after a high multiplicity of infection (5 PFU/cell) was monitored by plaque assay as described for panel A. (C) Growth in murine embryo fibroblasts after a low multiplicity of infection (0.01 PFU/cell). Again, the titer of virus was determined by plaque assay on BHK-21 cells. We also assayed the ORF11FRT mutant. (D) Six- to 8-week-old female BALB/c mice were infected intranasally with 2 x 104 PFU of virus as previously described (12). The titers of infectious virus in lung homogenates were determined by plaque assay on mouse embryonic fibroblasts (5). Each bar shows mean titers ± standard errors of the means of results for five individual mice. ORF11STOP virus titers were significantly reduced compared to those of the wild type at days 3, 5 (P < 0.005 by Student's t test), and 7 (P < 0.03) postinfection. (E) Latent virus in spleens was measured by infectious center assay (5). Preformed infectious virus, measured by infectious-center assay of freeze-thawed samples, was always <0.1% of the infectious center titer (data not shown). The ORF11STOP virus showed significantly reduced initial seeding to the spleen at days 5 and 7 (P < 0.02). Titers thereafter were equivalent to that of the wild type. Again, each point shows the mean ± standard error of the mean of results for five mice per group. ICs, infectious centers. (F) Infectious virus titers in lungs were measured by plaque assay after intranasal infection of BALB/c mice with two independently derived ORF11FRT mutants. Each point represents the result for one mouse. Both mutants showed a significant reduction in titers compared to those of the wild-type virus at days 3 and 5 (P < 0.005). (G) C57BL/6Jmice were infected intranasally with 2 x 104 PFU of ORF11STOP, the revertant, ORF11FRT or wild-type MHV-68 as indicated. Infectious virus in lung homogenates was measured by plaque assay. Each point represents the result for one mouse. Both ORF11 mutants showed a significant reduction in titer compared to that of the wild-type virus (P < 0.05 at each time point), whereas the revertant did not.
|
![]() View larger version (80K): [in a new window] |
FIG. 3. Identification of the ORF11 gene product. (A) Cell-associated (cells) and supernatant (sup) stocks of ORF11STOP (11) and wild-type (WT) MHV-68 were immunoblotted with rabbit anti-ORF11-GST serum. The ORF11-specific product of approximately 43 kDa is indicated by an arrowhead. The predicted size of the ORF11 gene product is 42.5 kDa. (B) The same virus stocks were immunoblotted with rabbit anti-MHV-68 serum as a loading control. This process did not detect an ORF11-specific band. (C) Purified virions were immunoblotted with rabbit anti-ORF11-GST serum as described for panel A. Virions were recovered from supernatants of virus-infected cells by high-speed centrifugation (38,000 x g, 90 min) and then resuspended in phosphate-buffered saline, sonicated, layered over 5 to 15% Ficoll gradients, centrifuged again (30,000 x g, 90 min), recovered as a distinct band, washed in phosphate-buffered saline, and stored at 70°C. In the ORF11-CFP virus, ORF11 is tagged in situ with CFP at its C terminus. The predicted size of the ORF11-CFP fusion protein is 70 kDa. (D) Purified virions were lysed or not lysed by adding Triton X-100 to a final concentration of 1% and incubating the virions on ice for 30 min. All samples were then incubated for 1 h at 37°C with or without 20 µg of trypsin per ml. Digestion was stopped by adding protease inhibitors (Roche Diagnostics) and returning the samples to ice. The samples were then boiled in Laemmli's buffer, resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and immunoblotted with anti-ORF11 or anti-MHV-68 rabbit sera as indicated. The open arrowhead indicates the native ORF11 band; the filled arrowhead indicates the ORF11-CFP fusion band. (E) ORF11-CFP virions were treated as described for panel D but immunoblotted with either an anti-GFP rabbit serum (Molecular Probes) to detect the ORF11-CFP fusion protein or the anti-gB monoclonal antibody T7H9 (9). gB in virions is predominantly cleaved from a 120-kDa precursor, and T7H9 recognizes the 65-kDa C-terminal fragment (9).
|
Immunofluorescence with the anti-p43 serum showed ORF11-specific cytoplasmic staining (Fig. 4A and B). There was also nonspecific staining of some dead and dying cell nuclei (Fig. 4A). We used the ORF11-CFP virus as an independent means of localizing p43, without the complication of background staining (Fig. 4C to F). Unmodified GFP, expressed from the MHV-68 BAC cassette (1), showed a predominantly nuclear localization in infected cells (Fig. 4G and H). In contrast, ORF11-CFP was perinuclear. At 24 h postinfection, cyan fluorescence was concentrated in perinuclear spots (Fig. 4D and E). This appearance was consistent with the immunofluorescence data and presumably reflected p43 accumulating at sites of virion assembly. There was weaker cyan fluorescence on infected-cell surfaces, probably from p43 in mature virions that were attached to the outer surface of the plasma membrane (Fig. 4D). CFP spots that appeared to be within the nucleus (Fig. 4D) appeared to be within protrusions of the perinuclear space rather than the nuclear substance itself. The cyan fluorescence at 6 h postinfection (Fig. 4F) was relatively weak, consistent with ORF11 being a late gene of MHV-68, but was again perinuclear in distribution.
![]() View larger version (76K): [in a new window] |
FIG. 4. p43 localization in MHV-68-infected cells. (A) BHK-21 cells were infected overnight with the wild type (WT), ORF11STOP (11), or REV as indicated. After fixation in methanol (20°C, 5 min) and permeabilization with 0.1% Tween 20, the cells were stained with rabbit anti-ORF11 serum plus Alexa488-coupled goat anti-rabbit immunoglobulin G (IgG) serum (green) or a gp150-specific monoclonal antibody plus Alexa594-coupled goat anti-mouse IgG serum (red). The nuclei of gp150-stained cells were counterstained with DAPI (4',6'-diamidino-2-phenylindole) (blue). Fluorescence was visualized on an Olympus microscope with a Hamamatsu digital camera. The anti-ORF11 serum stained the nuclei of some dead or dying cells nonspecifically but showed ORF11-specific cytoplasmic staining. (B) Perinuclear fluorescence in cells infected with wild-type virus and stained with anti-ORF11 serum plus Alexa488-coupled goat anti-rabbit IgG serum, visualized on a Leica confocal microscope. (C) Uninfected BHK-21 cells whose nuclei are stained with with DRAQ-5 (blue). (D) BHK-21 cells 24 h after infection with the ORF11-CFP virus, with cyan fluorescence (green) concentrated in perinuclear spots (arrowheads). (E) ORF11-CFP-infected BHK-21 cells as described for panel D. Arrowheads indicate CFP spots that appeared to be intranuclear but were probably infoldings of the perinuclear space. (F) BHK-21 cells 6 h after infection with ORF11-CFP virus. (G) BHK-21 cells 6 h after infection with wild-type MHV-68 expressing unconjugated GFP under the control of a human cytomegalovirus IE1 promoter as part of the left-end BAC cassette. (H) BHK-21 cells infected with GFP+ virus as described for panel G but after 24 h.
|
A three- to fivefold decrease in lytic replication might seem rather insignificant for viral fitness, particularly with normal latency establishment. However, the key role for lytic replication probably lies not in primary infection but in reactivation from latency and transmission to new hosts. Despite viral immune evasion, host immunity leaves only a limited window for productive reactivation, so even a threefold reduction in lytic replication would severely reduce long-term infectivity. We are only just beginning to uncover the many host functions that complex viruses such as MHV-68 modulate to maximize the infectious particle yield.
Present address: Division of Immunology, Department of Pathology, University of Cambridge, Cambridge, United Kingdom. ![]()
|
|
|---|
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»