The varicella-zoster virus portal protein is essential for cleavage and packaging of viral DNA

Abstract

The varicella-zoster virus (VZV) open reading frame 54 (ORF54) gene encodes an 87-kDa monomer that oligomerizes to form the VZV portal protein, pORF54. pORF54 was hypothesized to perform a function similar to that of a previously described herpes simplex virus 1 (HSV-1) homolog, pUL6. pUL6 and the associated viral terminase are required for processing of concatemeric viral DNA and packaging of individual viral genomes into preformed capsids. In this report, we describe two VZV bacterial artificial chromosome (BAC) constructs with ORF54 gene deletions, Δ54L (full ORF deletion) and Δ54S (partial internal deletion). The full deletion of ORF54 likely disrupted essential adjacent genes (ORF53 and ORF55) and therefore could not be complemented on an ORF54-expressing cell line (ARPE54). In contrast, Δ54S was successfully propagated in ARPE54 cells but failed to replicate in parental, noncomplementing ARPE19 cells. Transmission electron microscopy confirmed the presence of only empty VZV capsids in Δ54S-infected ARPE19 cell nuclei. Similar to the HSV-1 genome, the VZV genome is composed of a unique long region (UL) and a unique short region (US) flanked by inverted repeats. DNA from cells infected with parental VZV (VZVLUC strain) contained the predicted UL and US termini, whereas cells infected with Δ54S contained neither. This result demonstrates that Δ54S is not able to process and package viral DNA, thus making pORF54 an excellent chemotherapeutic target. In addition, the utility of BAC constructs Δ54L and Δ54S as tools for the isolation of site-directed ORF54 mutants was demonstrated by recombineering single-nucleotide changes within ORF54 that conferred resistance to VZV-specific portal protein inhibitors. Importance: Antivirals with novel mechanisms of action would provide additional therapeutic options to treat human herpesvirus infections. Proteins involved in the herpesviral DNA encapsidation process have become promising antiviral targets. Previously, we described a series of N-α-methylbenzyl-N'-aryl thiourea analogs that target the VZV portal protein (pORF54) and prevent viral replication in vitro. To better understand the mechanism of action of these compounds, it is important to define the structural and functional characteristics of the VZV portal protein. In contrast to HSV, no VZV mutants have been described for any of the seven essential DNA encapsidation genes. The VZV ORF54 deletion mutant described in this study represents the first VZV encapsidation mutant reported to date. We demonstrate that the deletion mutant can serve as a platform for the isolation of portal mutants via recombineering and provide a strategy for more in-depth studies of VZV portal structure and function.

FIG 1

Isolation of an ORF54 stable cell line. (A) The full-length ORF54 gene was cloned into the Gentarget lentivirus vector pLenti-EF-1a-Rsv-Puro. Control regions include the 5′ long terminal repeat (LTR) and 3′ LTR, packaging sequence (ψ), Rev-responsive element (RRE), central polypurine tract (cPPT), human elongation factor 1 alpha promoter (EF1a), Rous sarcoma virus promoter (RSV), and woodchuck hepatitis posttranscriptional regulatory element (WPRE). Open reading frames include the VZV open reading frame 54 (ORF54) and puromycin resistance gene (PURO). (B) ARPE-19 cells were transduced with ORF54 lentivirus, and selection was performed with puromycin. (C) Genomic integration of ORF54 was validated by PCR using genomic DNA from the stable cell clone. Genomic integration of ORF54 into human retinal epithelial cells (ARPE19 cells) was validated by performing PCR with ORF54-specific primers and analyzing the products on a 1% agarose gel. Lane 1, lambda HindIII DNA molecular weight markers. Lane 2, ARPE19 genomic DNA. Lane 3, ORF54-specific PCR of ARPE54 genomic DNA.

FIG 2

Predicted genome structures and features of parental and mutant viruses in the ORF54 region. (A) VZV_LUC parental virus indicating shared coding regions of ORF53 and ORF54. The ORF55 promoter is predicted to fall within the 3′ coding sequence of ORF54. (B) Δ54L, or large 54 deletion, indicates a complete (2,310-bp) deletion and replacement of ORF54 sequences with galK sequences, resulting in deletion of the ORF53 promoter and 3′ coding sequences. The predicted ORF55 promoter is also deleted. (C) Δ54S, or small 54 deletion, indicates an internal deletion of 1,223 bp that conserves ORF53 and ORF55 coding regions and promoters. Arrows, promoters; galK, E. coli galactokinase gene.

FIG 3

Confirmation of VZV BAC genome structures. SapI restriction sites flanking the ORF54 gene and the resulting predicted fragment sizes are shown for VZV_LUC, Δ54L, and Δ54S. Insertion of galK results in an additional SapI site. SapI-digested BAC DNAs were visualized after electrophoresis on a 0.65% agarose gel and staining with EtBr.

FIG 4

Growth kinetics. (A) ARPE54 or ARPE19 cells were infected with cell-free Δ54S, and a representative field was photographed at low and high magnification 5 days postinfection. Monolayers of (B) ARPE19, (C) ARPE54, or (D) MeWo cells were infected in triplicate with VZV_LUC, Δ54S, or 54R. Cells were harvested in luciferase lysis buffer at the indicated time points. Firefly luciferase activity is shown in relative luminescent units (RLU). Each point is the average of results for three independent samples (the standard deviation is provided).

FIG 5

Expression of pORF54 in Δ54S-infected cells. (A) Mock-infected or virus-infected ARPE19 cell extracts were immunoprecipitated with anti-pORF54 rabbit serum. Precipitated proteins were fractionated by SDS-PAGE and analyzed by Western blotting with an anti-pORF54 guinea pig serum. (B) Aliquots of cell extracts precipitated in panel A were immunoblotted with VZV anti-gE (α-gE) monoclonal antibody. The arrow indicates the position of the pORF54 monomer.

FIG 6

DNA cleavage assay. Monolayers of ARPE19 cells were infected with Δ54S or 54R. Cells were harvested at 24 h postinfection. Total infected cell DNA was digested with BamHI, fractionated by agarose gel electrophoresis, and analyzed by Southern blotting with probes that specifically recognize the long (TRL) and short (TRS) terminal repeats. DNA fragments representing uncleaved (2.8 kb), TRL (1.9 kb), or TRS (0.9 kb) are indicated.

FIG 7

Electron microscopy of VZV_LUC-, Δ54S-, and 54R-infected AREP19 cells. Monolayers of ARPE19 cells were infected with (A) VZV_LUC-, (B) Δ54S-, or (C) 54R-infected cell stocks (multiplicity of infection [MOI], ∼0.001). Cells were harvested at 72 h postinfection, fixed, and examined via transmission electron microscopy. Boxed images in the right column represent the boxed regions in the corresponding left column. The solid black arrows in panels A and C indicate intranuclear DNA-filled capsid. All other capsids are unfilled and represent either A- or B-type capsids. No DNA-filled capsids were observed in the nuclei of Δ54S-infected ARPE19 cells (B). Budding virions containing an envelope, capsid, and DNA could be observed at the plasma membrane of VZV_LUC- and 54R-infected ARPE19 cells (star).

FIG 8

Electron microscopy of VZV_LUC- and Δ54S-infected ARPE19 cells. Monolayers of ARPE19 cells were infected with Δ54S-infected cell stocks (MOI, ∼0.001). Cells were harvested at 72 h postinfection, fixed, and examined via transmission electron microscopy. (A) DNA-filled and unfilled empty capsids in the nucleus of a VZV_LUC-infected ARPE19 cell. Δ54S cells contained aggregated and individual empty capsids but no DNA-filled capsids. (B) Reverse images of the exact same fields as those represented in panel A. DNA-filled capsids contain bright white centers (white arrows), while empty capsids have gray or dark interiors. The star represents an empty capsid between the inner and outer nuclear membranes of a Δ54S-infected ARPE19 cell.

FIG 9

Plaque reduction assay. Monolayers of ARPE19 cells were infected in triplicate with VZV_LUC, VZV54₄₈^r, or VZV54₄₀₇^r in the presence of various concentrations of thiourea inhibitor (0, 0.03, 0.06, 0.125, 0.250, 0.500, 1.00, or 2.00 μg/ml). After incubation for 5 days at 37°C, cells were harvested in luciferase lysis buffer and assayed for luciferase activity. Each point is the average of results from three independent samples (the standard error is provided), and the data are presented as percent activity of the maximum RLU recorded for each virus. The dotted line is provided to compare IC₉₀ values between viruses.