Regulation of the Varicella-zoster Virus ORF3 Promoter by Cellular and Viral Factors

Abstract

The varicella zoster virus (VZV) immediate early 62 protein (IE62) activates most if not all identified promoters of VZV genes and also some minimum model promoters that contain only a TATA box element. Analysis of the DNA elements that function in IE62 activation of the VZV ORF3 promoter revealed that the 100 nucleotides before the translation start site of the ORF3 gene contains the promoter elements. This promoter lacks any functional TATA box element. Cellular transcription factors Sp1, Sp3 and YY1 bind to the promoter, and mutation of their binding sites inhibited ORF3 gene expression. VZV regulatory proteins, IE63 and ORF29, ORF61 and ORF10 proteins inhibited IE62-mediated activation of this promoter. Mutation of the Sp1/Sp3 binding site in the VZV genome did not alter VZV replication kinetics. This work suggests that Sp family proteins contribute to the activation of VZV promoters by IE62 in the absence of functional TATA box.

INTRODUCTION

Varicella-zoster virus (VZV) is a human alphaherpesvirus that belongs to the family Herpesviridae. It is the causative agent of two diseases, varicella (chickenpox) and herpes zoster (shingles). The VZV genome is a linear double-stranded DNA molecule of 125 Kb and encodes at least 71 genes which are identified numerically according to the location of their open reading frames (ORF) relative to the 5′ end of the VZV genome (Cohen et al., 2007). Upon its entry into the infected cell nucleus, the VZV genome is transcribed by the host cell RNA polymerase II (RNA Pol II) and the general transcription apparatus of the cell. By analogy with herpes simplex virus-1 (HSV-1), the VZV ORFs are presumed to be expressed as three distinct kinetic classes during viral infection: immediate early (IE), early and late. All of the IE gene products appear to be transcriptional regulatory proteins. Expression of these genes permits transcription of the early genes that are involved in DNA replication and nucleotide metabolism and the late genes encoding structural proteins, such as the capsid proteins and glycoproteins, which are required for assembly of virions (Cohen et al., 2007).

The expression of VZV early and late genes is thought to be mediated primarily through functions of the IE62 protein acting in conjunction with viral and cellular factors. Although several other proteins encoded by VZV, including the IE4, IE63, ORF10, and ORF61 proteins have been shown to be capable of transactivating and/or transrepressing specific viral promoters, (Baudoux et al., 2000; Defechereux et al., 1997; Moriuchi et al., 1992; Moriuchi et al., 1993; Moriuchi et al., 1994; Moriuchi et al., 1995; Sato et al., 2003; and Spengler et al., 2000; Wang 2010), IE62 is considered to be the primary viral transactivator driving expression of genes from all three putative kinetic classes (Kinchington et al., 2000).

To date, only a few VZV promoters have been analyzed in enough detail to determine their functional elements. However, these analyses as well as predictions of the positions of putative promoters within the VZV genome indicate that they are relatively typical RNA Pol II promoters with TATA-like elements and binding sites for ubiquitous cellular transcription factors upstream of the TATA elements (Smale et al., 2003). No promoters showing dependence on the presence of an initiator element (INR) at or just downstream of the start site of transcription are known. All of the VZV promoters studied thus far have been shown to depend on either canonical or non-canonical TATA elements for activation by IE62 (Peng et al., 2003; Ruyechan et al., 2003; and Yang et al., 2004) except for the VZV ORF10 promoter as shown by Che et al., (2007). The ORF10 promoter required the presence of a USF site and the mutation of this cis-element inhibited the IE62-mediated transactivation of this promoter. This circumstance most likely reflects the fact that only a few VZV promoters have been analyzed at a functional level rather than that VZV lacks these possible gene regulatory mechanisms.

The biology of VZV infection indicates that control of expression of the viral genes is essential to the ability of the virus to replicate productively in cell types that it targets during primary infection, which include skin and T cells, and to establish and preserve latency in neurons (Cohen et al., 2007). However, the molecular mechanisms by which VZV gene expression is regulated are not well understood. Functional analysis of additional VZV promoters representing all three putative kinetic classes of VZV genes and exhibiting differential regulation of viral gene expression is needed to provide more information about this fundamental process. Based on this consideration, the work presented here examines aspects of the regulation of the ORF3 promoter, which results in expression of ORF3 protein, a late gene product.

VZV ORF3 encodes a predicted 179 amino acid protein with no known function. ORF3 protein was dispensable for VZV replication in melanoma cells and skin organ culture in vitro as well as in human T cells in thymus/liver xenografts in SCID-hu mice infected in vivo (Zhang et al., 2007, Zhang et al. 2010). UL55, the HSV homologue of ORF3, has been shown to be dispensable for viral replication and for establishment of latency (Nash and Spivack, 1994).

In the work presented here, we analyzed the ORF3 promoter to better understand the molecular mechanism of its activation by IE62. The results demonstrated that ORF3 promoter lacks any functional TATA box and requires the cellular transcription factors Sp1, Sp3 and YY1 for its activation. However, the ORF3 promoter was down regulated by the presence of a suppressor cis-acting element located in the nucleotide segment from 100 to 120 before the ORF3 gene translation start site. VZV ORF10, ORF29, ORF61 and IE63 proteins had inhibitory effects on IE62-mediated transactivation of the ORF3 promoter. Mutating the Sp1/Sp3 site in the ORF3 promoter in the context of the VZV genome had no effect on viral replication in MeWo cells. This work suggests that an IE62 and Sp1/Sp3 mechanism is an alternative to VZV promoter activation by IE62 through interaction with a TATA-binding protein (TBP).

MATERIALS AND METHODS

Cells and viruses

MeWo cells, a human melanoma cell line were grown in Eagle’s minimal essential medium supplemented with 10% fetal bovine serum (Spengler et al., 2000). VZV strains MSP and pOka were propagated in MeWo cell monolayers as described by Lynch et al. (2002) and Peng et al. (2003).

Nuclear and whole cell lysate preparation and immunoblot analysis

Nuclear extracts of VZV infected MeWo cells were prepared as previously described (Lynch et al., 2002). MeWo cells were incubated in buffer A (10 mM HEPES, pH 7.9, 1.5 mM MgCl₂, 10 mM KCl, 0.5 mM dithiothreitol) at 4 °C on ice for 15 min to lyse the cells and release the cytoplasmic fraction. After centrifugation, the crude nuclear pellet was incubated on ice in buffer C (20 mM HEPES, pH 7.9, 25% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol). After centrifugation the nuclear extract was dialyzed against buffer D (20mM HEPES, pH 7.9, 20% (v/v) glycerol, 0.1 M KCl, 0.2 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol).

Whole cell lysates of VZV infected and pCMV-ORF3 transfected MeWo cells were prepared in lysis buffer (50 mM Tris-HCl, pH 7.5, 0.15 M NaCl, 1 mM EDTA, 0.1% Triton X-100 and protease inhibitor cocktail (Roche, Mannheim, GE) added per the manufacturer’s instructions) and analyzed for ORF3 protein by immunoblot (10% SDS-PAGE) using a rabbit polyclonal antiserum against a GST fusion full length ORF3 protein (peng et al., 2003, and Yang et al., 2006) and IE63 protein using rabbit polyclonal antibody against full length IE63 protein (Zuraniski et al., 2005). Rabbit polyclonal antibody against β-tubulin was obtained from Santa Cruz Biotechnology (Santa Cruz, CA.) and mouse monoclonal antibody against α-tubulin was obtained from Sigma-Aldrich. Quantification of the relative amounts of ORF3, IE63 α-tubulin and β-tubulin was done using a BioRad GS700 Imaging Densitometer (BioRad Hercules, CA). Statistical significance was determined by one-way ANOVA analysis of variance followed by Tukey’s post hoc test.

Plasmids

A set of luciferase reporter plasmids containing the ORF3 promoter flanked by firefly luciferase was constructed using the pGL2 basic vector (Promega, Madison, WI). The 336 bp intergenic region between ORF3 and ORF4 was amplified by PCR using these two primers containing a HindIII restriction site at the 5′ end and a XhoI restriction site at the 3′ end respectively; the primer sequences were 5′-ATCAAGCTT TAATTAAACGTTCGGTACACGTCT-3′ and 5′-ATCCTCGAGAAATAAAAAATACCTT TTTCATGC-3′. The PCR product was digested and inserted into the pGL2 basic vector multiple cloning sites between the HindIII and XhoI restriction sites. The ORF3 promoter truncation that contained the 120 nucleotides from the translation start codon of ORF3 gene was cloned by amplification of the first 120 bp by PCR using the first primer as above; the second primer was: 5′-ATCCTCGAGTTTTTAAGGCGACGTTG GGGATAT-3′. This PCR product was inserted into the basic pGL2 plasmid. The other ORF3 promoter truncations containing 87, 94 and 100 nucleotides from the translation start codon were constructed from the 120 nucleotides truncation construct using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, LaJolla, CA).

The plasmids containing mutations of the Sp1/Sp3 and YY1 sites and TATA box within the ORF3 promoter were generated from the wild type pGL2-ORF3 plasmid containing the ORF3/ORF4 intergenic region using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, LaJolla, CA). The primer sets for these mutations were: Sp1/Sp3 site: 5′-TGGTTTGAAAGCAATGTAATCCTTCCCATATATCCCCAACGTCGC-3′ and 5′-GCGACGTTGGGGATATATGGGAAGGATTACATTGCTTTCAAACCA-3′; YY1 site: 5′-TGAAAGCAATGTAATCCCGCCCGTATATCCCCAACGTCGCCTTAA-3′ and 5′-TTAAGGCGACGTTGGGGATATACGGGCGGGATTACATTGCTTTCA-3′; TATA box1: 5′-AGTACCGGAATGCCAAGCTTTAGCTAAACGTTCGGTACACGTCTG-3′ and 5′-CAGACGTGTACCGAACGTTTAGCTAAAGCTTGGCATTCCGGTACT-3′; TATA box2: 5′-TGAAAGCAATGTAATCCCGCCCATATCCCCCCAACGTCGCCTTAA-3′ and 5′-TTAAGGCGACGTTGGGGGGATATGGGCGGGATTACATTGCTTTCA-3′. The mutated nucleotides are indicated in bold. All primers were synthesized by IDT (Coralville, IA). The mutations were verified by sequencing at the Roswell Park Cancer Institute sequencing facility, Buffalo NY.

The pCMV62 plasmid expressing ORF62 under the control of the cytomegalovirus immediate-early (IE) promoter and the pCMV-ORF63 and pCMV-ORF29 constructs have been described previously (Perera et al., 1992 and 1993; Stevenson et al., 1996; and Zuraniski et al., 2005). The pCMV-ORF10, pCMV-ORF61 and pCMV-ORF4 constructs were kindly provided by P. R. Kinchington (University of Pittsburgh, Pittsburgh, PA).

The pCMV-ORF3 was constructed using pcDNA empty vector (Invitrogen, Carlsbad, CA). The 540 bp of ORF3 gene was amplified by PCR using these two primers containing a HindIII restriction site at the 5′ end and a XhoI restriction site at the 3′ end respectively; the primer sequences were 5′-ATCAAGCTTATGGATACAAC GGGAGCTTCCGAA-3′ and 5′-ATCCTCGAGTCATAGTCCGCCGACAGCCGCTCG-3′ The PCR product was digested and inserted into the pcDNA basic vector multiple cloning sites between the HindIII and XhoI restriction sites.

Reporter gene assays

Luciferase reporter gene assay experiments were performed in MeWo cells as previously described (Yang et al., 2004). 2 × 10⁵ MeWo cells were seeded in each well of 12-well plates 24 h before transfection. Cells were transfected with one microgram of each reporter vector (pGL2-ORF3) using Lipofectamine reagent (Invitrogen, Carlsbad, CA), along with 5 ng of pEF1α-RL plasmid (Promega, Madison, WI) as a control for transfection efficiency. The cells were superinfected with VZV MSP 24 h post transfection (0.4 infected cells per 1 uninfected cell).

In the experiments done in the presence of IE62, the reporter plasmids were co-transfected along with 5 ng of pCMV-ORF62. Cells were transfected with IE63, IE4, ORF10, ORF29 and ORF61 plasmids (0.5 μg) separately or with pCMV-ORF62. Dual luciferase activities were normalized to the Renilla luciferase activities. pcDNA was transfected along with the pCMV62 plasmid to equalize the amounts of both total DNA and the promoter construct in each set of transfections.

The cells were lysed 48 h post transfection or super-infection in 250 μl of lysis buffer (50 mM HEPES, pH 7.4, 250 mM NaCl, 1% NP-40, 1 mM EDTA). Control experiments without transfection of pCMV-ORF62 or VZV infection were done for each plasmid to determine basal expression levels. Dual-luciferase assays were performed according to the manufacturer’s instruction. Transfection experiments were repeated at least three times.

EMSA and supershift analyses

36 bp oligonucleotide probes (IDT, Coralville, IA) containing wild type and mutant Sp1/Sp3 and YY1 binding sites were used in electrophoretic mobility shift assays (EMSAs). Probes were end labeled with ATP [α-32P] using T4 kinase (Invitrogen, Carlsbad, CA). One hundred femtomoles of the 36 bp labeled probes containing either the wild type or the mutant sequences were incubated with 15 μg of VZV infected MeWo cell nuclear extract in a 10 μl reaction mixture in binding buffer (40 mM HEPES, pH 7.9, 100 mM NaCl, 10 mM MgCl₂, 200 μg/ml bovine serum albumin (BSA), 12% glycerol, 0.05% NP-40, 1 mM dithiothreitol, and 3 μg poly (dI.dC)). The samples were analyzed by electrophoresis on a 5% polyacrylamide (37.5:1 acrylamide/bisacrylamide) gel followed by autoradiography. In competition assays, the ratio of cold probe to labeled probe was 100:1.

In the supershift assays, anti-Sp2, Sp3 and Sp4 antibodies obtained from Santa Cruz Biologicals (Santa Cruz, CA) or anti-Sp1 antibodies (Upstate, Temecula, CA) and rabbit polyclonal YY1 antiserum, generously supplied by Dr. Te-Chung Lee (University of Buffalo) were added in 4μg (Sp family) or 2 μl (YY1) aliquots to reaction mixtures containing infected cell nuclear extracts and the 36 bp probe containing the Sp1/Sp3 and YY1 binding sites. The samples were then analyzed by electrophoresis on a 5% polyacrylamide (37.5:1 acrylamide/bisacrylamide) gel followed by autoradiography.

Magnetic Bead Recruitment Assays

Magnetic bead recruitment assays were performed essentially as described by Lynch et al. (2002). A 100-bp biotinylated DNA fragment containing the minimal ORF3 promoter elements was purchased from IDT (Coralville, IA). 20 pmol of biotinylated DNA was added to 100 μl of M-280 streptavidin-coated magnetic beads (Dynal, Lake Success, NY). The mixture was incubated for 15 min at room temperature. The beads were collected using a magnetic particle concentrator (Dynal), washed twice with 1× B&W buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 1 M NaCl), and twice with TEN buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 0.1 M NaCl) and incubated with 500 μg of VZV-infected MeWo nuclear extract for 1 h at 4 °C in the presence of 3 μg poly (dI.dC). After washing three times with TEN buffer, the beads were boiled in 2X SDS-PAGE loading buffer for 10 min and analyzed by 10% SDS-PAGE and immunoblot.

Mutation of the ORF3 promoter Sp1/Sp3 site in pOka VZV

Recombinant viruses were generated using overlapping cosmids derived from parental Oka (pOka) designated Fsp73 (pOka nucleotides [nt] 1 to 33128), Spe14 (pOka nt 21795 to 61868), Pme2 (pOka nt 53755 to 96035), and Spe23 (pOka nt 94055 to 125124) (Niizuma et al., 2003). The ORF3 promoter is located within the sequence of the cosmid Fsp73. An 11.5 Kb XhoI/SacI fragment from Fsp73 cosmid was cloned into pLitmus28 to generate pLitmus XhoI/SacI. Four primers were designed to amplify the 11.5 Kb fragment in two PCR products and ligate them together with the pLitmus 28 in three PCR product ligation reaction. The primers sequences for the first PCR product: 5′-CCTCTGACTTGAGCGTCGATTTTTG-3′, and 5′-ATTGCTTTCAAACCAATATCTTTG C-3′, and the primers sequence to amplify the second PCR product which contain the Sp1/Sp3 site mutation in the ORF3 promoter: 5′-GTAATCCTTCCCATATATCCCCAA C-3′ and 5′-GCCATCTTCCACGGGTCATGC-3′. Mutated nucleotides are indicated in bold. The first nucleotide of the second primer amplifying the first PCR product (A) is adjacent to the first nucleotide of the first primer amplifying the second PCR product (G). PCR reactions were done with Accuprimer Pfx polymerase (Invitrogen). The left PCR product was digested with XhoI and the right PCR product was digested with NcoI. The digested PCR products were cloned into XhoI/NcoI digested pLitmus XhoI/SacI plasmid to generate pLitmus XhoI/SacI ORF3 promoter. Mutations were verified by DNA sequencing (Quintara Biosciences, South San Francisco, CA). The XhoI/SacI fragment from the pLitmus XhoI/SacI ORF3 promoter construct was to introduce the ORF3 promoter mutation into the Fsp73 cosmid.

Recombinant viruses were made by transfection of melanoma cells with either the wild type or the mutated Fsp73 cosmid and the other three intact cosmids, Spe14, Spe23 and Pme2. To confirm the Sp1/Sp3 site mutation, genomic DNA was extracted from virus-infected melanoma cells with DNAzol (Invitrogen, Carlsbad, CA). A PCR fragment covering the mutated region was amplified from the genomic DNA, gel-purified with a QIAquick gel extraction kit (Qiagen, Inc., Valencia, CA), and sequenced (Elim Biopharm, Inc., Hayward, CA).

The replication kinetics of the recombinant viruses was assessed by an infectious focus assay with immunostaining using anti-VZV monoclonal antibody (Meridian) to detect plaques (Chaudhuri et al., 2008; and Moffat et al., 1995). Statistical differences in growth kinetics were determined by Student’s t test.

RESULTS

Mapping of the ORF3 promoter and identification of the cis-acting elements that regulate this promoter

To map the ORF3 promoter structure, luciferase reporter gene assays were done using constructs that included portions of the intergenic region between ORF3 and ORF4 genes (Fig. 1A and B). The intergenic region consisted of the 336 bp between the ORF3 translation start site and the stop codon of the ORF4 gene. Expression of the ORF3 reporter gene was assessed in the context of VZV super-infection and pCMV-62 transfection, using the pGL2-ORF3 plasmid that expressed firefly luciferase under the control of the ORF3 promoter or ORF3 promoter truncations. The activity of the promoter constructs in the presence of VZV infection or pCMV-62 was determined as induction (n-fold) of the luciferase activity compared to the basal activity observed without infection or pCMV-62 transfection respectively.

Fig. 1.

The ORF3-ORF4 intergenic region. A). Schematic diagram shows the structure of the VZV genome and the position of the intergenic region between ORF3 and ORF4 genes includes the ORF3 promoter. UL and US are the unique long and unique short segments of the genome respectively. TRL and IRL are the terminal and internal repeat sequences of UL. TRS and IRS are the terminal and internal repeat sequence of the US. The numbers below the ORF3-ORF4 intergenic region represent the nucleotide positions of the beginning and end of the region in the VZV MSP strain. B) Nucleotide sequence of the ORF3-ORF4 intergenic region. The arrows indicate the truncations used in the reporter gene assays for mapping the intergenic region for the presence of ORF3 promoter elements. TBP sites are underlined, Sp1/Sp3 site is boxed and YY1 site is in italics.

The analysis of the intergenic region showed that the 100 nucleotides immediately adjacent to the ORF3 translation start site displayed the maximum luciferase activity in context of both pCMV-62 transfection and VZV super-infection (Fig. 2A and 2B). These results suggested that this 100 bp region contained the promoter cis-acting elements for ORF3 gene and included the essential elements required for ORF3 promoter activation. ORF3 promoter activation was low in absence of IE62 or VZV superinfection. The first two constructs containing 87 and 94 nucleotides respectively from the ORF3 gene translation start site showed particularly low expression levels (~10–15%) compared to the maximum activity observed with the 100 bp construct. In contrast, the two constructs which included the nucleotides between 100 and 120 had ~2-fold less activity than the 100 bp construct, suggesting that this 20 bp segment contained repressive cis-elements regulating the ORF3 promoter activity as mediated either by IE62 alone or VZV infection. Bioinformatics analysis of these 20 nucleotides for putative cellular transcription factors binding sites suggested binding sites for the Sp family and NFκB proteins but experimental analysis did not show binding of these factors to 30 bp oligonucleotides containing this sequence (data not shown).

Fig. 2.

Mapping the cis-acting elements that influence ORF3 promoter activity. Reporter gene assays were done to determine the effects of truncation constructs containing segments of the ORF3-ORF-4 intergenic region on ORF3 promoter activation. Experiments were done in triplicate with A) pCMV-ORF62 transfection and B) VZV super-infection. The promoter activities resulting from the presence of pCMV-ORF62 or VZV super-infection are reported as induction (n-fold) of luciferase activity over basal (without ORF62 transfection or VZV super-infection). Statistical significance was determined by a one-way ANOVA analysis of variance followed by Tukey’s post hoc test. Error bars indicate standard error.

The 100 bp segment immediately upstream of the ORF3 gene translation start site was then analyzed for the presence of cellular transcription factor binding sites using the AliBaba2.1 and Patch programs (http://www.gene-regulation.com/pub/programs.html; Biobase). A GC-rich sequence was identified from 88 to 94 bp before the ORF3 gene translation start site. This GC-rich sequence contained predicted binding sites for Sp family cellular transcription factors. A potential binding site (5′-CCCATATAT-3′) for the cellular transcription factor YY1 was also identified that partially overlapped the GC-rich sequence as well as two postulated TATA box sites. The first one was immediately adjacent to the translation start codon of the ORF3 gene and the second partially overlapped the predicted YY1 binding site. Based on the bioinformatics results and the reporter gene assays, experiments were done using EMSA, supershift, and mutational analyses in order to determine whether the GC-rich sequence and the YY1 site were bona fide binding targets for the specific cellular transcription factors.

Sp1/Sp3 bind to the GC-rich sequence and YY1 binds to the predicted YY1 binding site

In the first series of experiments, EMSA and supershift assays were performed using 36 bp duplex oligonucleotides containing both the GC-rich sequence and the predicted YY1 binding site in the presence of a nuclear extract from VZV infected MeWo cells. Numerous complexes were formed using this 36 bp oligonucleotide, including two major slowly migrating complexes (Fig. 3B). Antibody supershift assays were then done to assess the binding of the Sp family members and the YY1 cellular transcription factors to these sequences. The anti-Sp1 antibody supershifted part of the upper major complex and the anti-Sp3 antibody completely supershifted the lower major complex (Fig. 3B). The anti-YY1 antibody also completely supershifted one of the minor complexes formed using these 36 bp oligonucleotides (Fig. 3C). In contrast, no supershifts or loss of complexes were observed in the presence of anti-Sp2 or anti-Sp4 antibodies.

FIG. 3.

EMSA and supershift assays using 36 bp oligonucleotides containing the wild-type GC rich sequence and the YY1 site of the ORF3 promoter. (A) Sequence of the 36 bp oligonucleotide. The GC rich sequence is underlined and the YY1 site is boxed. Supershift assays were done in the presence of VZV-infected cell nuclear extracts. B) Polyclonal antibodies against Sp1, Sp2, Sp3 and Sp4 were used in the supershifts as indicated. C) Polyclonal antibody against YY1 was used in the supershifts as indicated. The YY1 supershift assay gel was exposed in two stages with the upper part being developed for a shorter time in order to better visualize the two separate complexes of the newly formed supershifted YY1 complex and the original Sp1 containing complex.. Lanes: FO, free oligonucleotide; I, VZV-infected nuclear extract. The positions of the original major complexes are indicated by arrows and the positions of the supershifted bands are indicated by asterisks.

Next, EMSAs were performed using oligonucleotides containing mutations in the consensus Sp1 binding site located in the GC-rich sequence and in the YY1 binding site. Site-specific mutagenesis was used to substitute CG residues with TT in the GC-rich sequence and the first A following the GC-rich sequence was mutated to G. These mutations were predicted to abolish binding of Sp1 and Sp3 and YY1 respectively. As shown in Fig. 4B and C, both mutations ablated the formation of the YY1 complex even though altering the GC-rich sequence was not predicted to affect YY1 binding to the sequence. Quantitative analyses showed that the substitutions in the GC-rich sequence inhibited the formation of the Sp1 and Sp3 complexes by about 85% while the substitution mutation of the YY1 binding site inhibited the formation of these Sp1 and Sp3 complexes by ~20%.

FIG. 4.

EMSA assays using wild type, Sp1/Sp3 mutant and YY1 mutant duplex oligonucleotides. (A) Sequences of the oligonucleotides containing the wild type and mutant oligonucleotides. The GC rich sequence is underlined and the YY1 site is boxed. Typical EMSA results showed binding to wild type probe as well as the mutant probe in experiments using B) Sp1/Sp3 mutant probe or C) the YY1 mutant probe. Lanes: FO, free oligonucleotide; I, VZV-infected nuclear extract. The positions of the original complexes are indicated by arrows.

To test the specificity of the formation of these complexes, competition EMSA experiments were done using the unlabeled 36-bp containing the GC-rich sequence and the predicted YY1 binding site as the specific competitor and a 30-bp oligonucleotide containing the origin binding protein (OBP) box A sequence as the nonspecific competitor (Khalil et al., 2008). As shown in Fig. 5B, the cold specific competitor efficiently competed away the formation of these complexes, while the presence of the nonspecific competitor had no effect on the pattern of the shifted bands. These experiments indicated that cellular transcription factors Sp1, Sp3 and YY1 bind specifically to the ORF3 promoter element.

FIG. 5.

EMSA and competition assays using a 36-bp oligonucleotide containing the wild type Sp1/Sp3 and YY1 sites in the presence of VZV-infected cell nuclear extracts. A) Sequences of the two oligonucleotides used in the assays. The Box A oligo was used as the nonspecific competitor. B) Competition assays using a 100-fold excess of cold wild type specific competitor and nonspecific competitor. The positions of the Sp1, Sp3 and YY1 complexes are indicated by arrows. C) Magnetic-bead recruitment assays showing the binding of IE62, Sp1, Sp3 and YY1 to the 100 bp biotinylated duplex oligonucleotides containing the ORF3 promoter cis-acting elements. α-TBP was used as a negative control.

Next, magnetic bead recruitment assays were done to identify the viral and cellular proteins that were involved in gene transcription and bound to the 100 bp oligonucleotides containing the ORF3 promoter. As shown in Fig. 5C, the cellular transcription factors Sp1, Sp3 and YY1 and IE62 were able to bind to the ORF3 minimal promoter sequence. In contrast, the TBP protein did not bind to this sequence under these experimental conditions.

Sp1/Sp3 and YY1 site mutations inhibited ORF3 promoter activity

In order to test the effects of the binding of Sp1, Sp3 and YY1 to these sequences on ORF3 promoter activation, the mutations that were determined to inhibit binding were introduced in the full length reporter construct containing the complete intergenic region between ORF3 and ORF4 genes. Reporter gene assays were done in the context of VZV super-infection and with pCMV-62 transfection. Luciferase activities obtained from each reporter plasmid in the absence of VZV super-infection and ORF62 transfection represent the basal levels from this plasmid and were normalized to 1. Reporter gene activities in the presence of VZV super-infection and ORF62 transfection were reported as induction (n-fold) of luciferase activities in reference to the basal activity without VZV super-infection or ORF62 transfection.

Both the Sp1/Sp3 and the YY1 site mutations inhibited ORF3 promoter activation by about 80% in the experiment done using pCMV-ORF62 transfection and ~85–90% in VZV super-infection experiments (Fig. 6A and 6B). In contrast, these mutations did not influence ORF3 promoter activity significantly in the absence of IE62 or VZV superinfection. These results suggested the involvement of these cellular transcription factors in the activation of the ORF3 promoter mediated by IE62 or by VZV super-infection. In contrast, the mutations of the two putative TATA boxes had no effect on ORF3 promoter activation in the context of either VZV infection or pCMV-ORF62 transfection (Fig. 6A and B).

Fig. 6.

The effect of mutations of the Sp1/Sp3 and YY1 sites and putative TATA boxes in the ORF3 promoter on ORF3 gene expression. Results of triplicate assays assessing the effects of the presence of the Sp1/Sp3 and YY1 site mutations on the expression levels of the firefly luciferase reporter gene present at the position of the ORF3 gene in the context of A) pCMV-ORF62 transfection and B) VZV super-infection. The promoter activities resulting from the presence of transfected ORF62 or VZV super-infection are reported as induction (n-fold) of luciferase activity over basal (without ORF62 transfection or VZV super-infection respectively). Statistical significance was determined by a one-way ANOVA analysis of variance followed by Tukey’s post hoc test. Error bars indicate standard error.

Transactivation of the ORF3 promoter by VZV regulatory proteins

In these experiments, the pGL2-ORF3 reporter that included the complete 336 bp intergenic region between ORF3 and ORF4 genes was used in reporter gene assays. The plasmids encoding the VZV regulatory proteins IE62, IE63, ORF61, ORF10, IE4 and ORF29 were co-transfected individually into MeWo cells with the reporter pGL2-ORF3 plasmid. The transactivation results were presented as n-fold induction compared to the luciferase activity with the empty vector, which was normalized to 1. As expected, IE62 had potent transactivation effects on the ORF3 promoter (about 50-fold activation). In contrast none of other VZV regulatory proteins, IE4, ORF61, and IE63, ORF10 and ORF29, transactivated the ORF3 promoter (Fig. 7).

FIG. 7.

Transactivation of the ORF3 promoter by VZV proteins. MeWo cells were transfected with the full-length pGL2-ORF3 promoter and plasmids expressing IE4, ORF10, ORF61, IE62, IE63, ORF29 or IE62 together with IE4, ORF61, ORF10, ORF29 or IE63. Cells were harvested at 48 h post-transfection, and the dual luciferase assay was performed. The promoter activities resulting from the presence of transfected VZV proteins are reported as induction (n-fold) of luciferase activity over basal (without VZV protein expression). Statistical significance was determined by a one-way ANOVA analysis of variance followed by Tukey’s post hoc test. Error bars indicate standard error.

To further evaluate the regulatory roles of the other transactivators on IE62-mediated activation of the ORF3 promoter, IE62 was co-expressed with IE4, ORF61, IE63, ORF10 or ORF29. IE4 had no significant influence on IE62-mediated transactivation of the ORF3 promoter. However, the other VZV regulatory proteins inhibited IE62-mediated activation with inhibition of ~2-fold by ORF29 and ORF10 proteins and~3-fold by IE63 (Fig. 7). Of interest, ORF61 protein showed maximum inhibition of the IE62-mediated activation of the ORF3 promoter of ~10-fold. These results suggested a possible mechanism for downmodulation of IE62-mediated transactivation of late gene expression mediated by other VZV regulatory proteins.

Mutating the Sp1/Sp3 site in the ORF3 promoter did not affect VZV replication in vitro

Introducing the Sp1/Sp3 site mutation into the VZV genome using the pOka cosmids yielded recombinant virus with the expected sequence change compared to the recombinant generated with intact pFsp73 cosmid (recombinant wild type virus). When the growth kinetics of pOka and the ORF3 promoter mutant were compared in MeWo cells, titers were equivalent over the first five days post infection with a slight decrease for the ORF3 promoter mutant at day 6 (Fig. 8A). The expression of ORF3 protein in the whole cell lysate of MeWo cells transfected with pCMV-ORF3 expressing plasmid showed that the presence of 19 KDa protein equivalent to the expected molecular weight of ORF3 protein (179 amino acids). Then we evaluated the influence of this ORF3 promoter mutation on the expression of the ORF3 protein in infected MeWo cells. The promoter mutation was associated with a significant reduction in ORF3 protein synthesis of about four-fold based on densitometer analyses done at 36 hrs post-infection, compared to pOka (Fig. 8B). Also, the expression level of IE 63 was determined in the wild type as well the ORF3 mutant viruses. We have found that equal expression of the IE63 protein in the two viruses. The expression level of IE63 protein here was used as a control for the level of infection in the wild type and ORF3 mutant virus preparations.

Fig. 8.

The effect of VZV pOka-Sp1/Sp3 site mutation in the ORF3 promoter on VZV replication in MeWo cells. A) Growth kinetics of pOka and pOka-Sp1/Sp3 mutant virus in MeWo cells. Cells were inoculated at 10³ PFU/ml with wild type and mutant viruses and infectious virus yields were determined for 6 days after inoculation. B) Effect of Sp1/Sp3 site mutation on ORF3 expression levels in MeWo cells during VZV infection and the expression of ORF3 protein in MeWo cells transfected with pCMV-ORF3 plasmid. Western blot analyses show the expression levels of ORF3, IE63, α-tubulin and β-tubulin at 36 hrs post-infection. α-tubulin and β-tubulin were used as a loading control in the experiments. The blots were scanned by densitometry to obtain quantitative data (in triplicate). Statistical significance was determined by a one-way ANOVA analysis of variance followed by Tukey’s post hoc test.

DISCUSSION

This analysis mapped the ORF3 promoter and identified cis-acting elements that affect activation of the ORF3 promoter activation mediated by IE62, the major VZV trans-activator. This work shows that ORF3 activation occurs by an alternative mechanism that is independent of the putative TATA box site(s) in the ORF3-ORF4 intergenic region. In the case of ORF10 promoter, we suggested that the absence of the TATA element might be overcome by the recruitment of the TBP to the ORF10 promoter through the physical interaction with IE62 or USF (Peng et al., 2003; and Yang et al., 2006). However, the binding of TBP to ORF10 promoter was not assessed whereas these experiments showed that TBP does not bind to ORF3 promoter.

The VZV genome is transcribed by the host cell RNA polymerase II (RNA Pol II). RNA polymerase II transcription is initiated through the assembly of the pre-initiation complex (PIC) consisting of the general transcription factors (GTFs, TFIID, TFIIA, TFIIB, TFIIF, TFIIE and TFIIH) (Orphanides et al., 1996). The efficiency of the formation of the PIC is a critical step in determining the rate of transcription and thus is a common site of both positive and negative transcriptional regulation (Choy and Green, 1993). TFIID is a large protein complex consisting of the TBP and its associated factors TAFs (Nakajima et al., 1988). In eukaryotes, TBP is responsible for the assembly of the transcription initiation machinery at TATA box containing promoters as a component of the TFIID complex. Moreover, a TFIID complex containing TBP is essential for transcription even when the eukaryote gene promoter lacks a TATA box (Pugh and Tjian, 1991). VZV promoters are characterized by the fact that the TATA-elements required and/or predicted to be positioned correctly for expression are frequently atypical and therefore less likely to be bound with high affinity by the TATA-binding protein (TBP) (Ruyechan et al., 2003). Although some VZV transcripts are generated using canonical TATA sequences (TATAAA), as in the case of ORF61, other transcripts arise from non-canonical TATA sequences like TTTTAA (ORF62), ATAAAA (ORF67), ATTTAAATT (ORF14) or TATGTAAA (ORF28) (Stevenson et al., 1992; Mckee and Preston, 1991; Ling et al., 1992; and Meier and Straus, 1993).

In the present experiments, we found that the first 100 nucleotides before the translation start site of the ORF3 gene were required for ORF3 expression in a reporter construct. These 100 nucleotides have two putative TATA boxes but mutation of these sites did not affect ORF3 promoter activation and TBP did not bind to a 100 bp oligonucleotide sequence containing the ORF3 promoter cis-elements. These observations indicate that IE62 utilizes a different mechanism to activate the ORF3 promoter. The cellular transcription factors Sp1, Sp3 and YY1 were shown to bind to this sequence and mutation of their proposed binding sites inhibited interaction and also inhibited transactivation of ORF3 promoter mediated by IE62 or VZV superinfection significantly. These results pointed to a role for Sp1, Sp3 and YY1 interactions with IE62 in absence of any functional TATA box in achieving recruitment of basal transcription machinery to the ORF3 promoter, leading to its activation.

Many viral and cellular promoters transcribed using RNA polymerase II lack any obvious TATA box predicted to bind TFIID. These cellular and viral genes include those encoding for dihydrofolate reductase, c-H-ras, adenine deaminase, TGF-α, thymidylate synthase and SV-40 late gene products and EBV latent membrane protein1 LMP1 (Dynan, 1986; Swick et al., 1989; Carcamo et al., 1989; Mitchell et al., 1987; and Sadler and Raab-Traub, 1995). These TATA-less promoters are characterized by the presence of GC boxes which are the binding site for the Sp family members of transcription factors and their transcription initiation is critically dependent on Sp1 (Pugh and Tjian, 1991). On the other hand, many other TATA-less genes are regulated during cell growth and their transcription is stimulated during cell proliferation (Azizkhan et al., 1993). Among these, two TATA-less promoters in the rat for haploid-specific Oxct2b gene and catalase required cAMP responsive element (CRE) and CCAAT/enhancer binding protein-beta respectively (Somboonthum et al., 2005; and Taniguchi et al., 2005). In HSV-1, an alphaherpesvirus closely related to VZV, UL9 and ICP34.5 promoters do not have a canonical TATA box site (Deb et al., 1993 and Sarisky and Weber, 1994). However, these promoters may have a non-canonical TATA box. The ICP34.5 promoter sequence has a predicted non-canonical TATA-box site that is similar to the TATA site shown to be functional for the activation of UL38 promoter (Sarisky and Weber, 1994). The promoter of HSV-1 UL55, the ORF3 homologue, has not been analyzed.

The physical and functional interaction between IE62 and Sp1 has been demonstrated previously (Narayanan et al., 2007; and Peng et al., 2003). Analyses of several VZV promoters have identified functional GC rich elements that are required for binding of the Sp family of cellular transcription factors. Some of these sites have the canonical Sp1 site (5′-GGGCGGG-3′) identified in the SV40 early promoter by Dynan and Tjian (1983). Sp1 regulates the gI and gE promoters in transient transfection assays and as demonstrated by mutagenesis in the VZV genome; GC rich elements have been identified in IE4 and IE63 promoters (Beraraducci et al., 2007; He et al., 2001; Ito et al., 2003; Kinchington et al., 1994; Rahaus et al., 1999; and Rahaus and Wolff, 2000). Sp1 sites are also identified in the ORF61 promoter and disrupting these sites in the VZV genome affected not only the expression of ORF61 but also virus growth in skin xenografts in vivo (Wang et al., 2009). In addition, bioinformatics analysis reveals a high frequency of predicted Sp1 sites within VZV gene promoters, indicating a minimum of 18 beyond those that have already been authenticated (Ruyechan et al., 2003).

The role of Sp family members other than Sp1 in VZV infection is unknown except for the demonstration of Sp3 binding to the Sp1/Sp3 site in the downstream region of oriS, which supports a role for this transcription factor in the VZV life cycle (Khalil et al., 2008). Of interest, the mutation of this Sp1/Sp3 site in the downstream region of VZV oriS inhibited both origin dependent DNA replication and flanking gene transcription in transient transfection assays (Khalil et al., 2012), suggesting a possible role for Sp3 in replication as well as promoter activation.

YY1 is a 414 amino acid zinc-finger protein capable of repression and activation of gene transcription in several viruses, including adeno-associated virus, human papillomavirus, parvovirus B19, HSV-1 and Moloney murine leukemia virus (Lee et al., 1992; Lee et al., 1998; Pajunk et al., 1997; and Shi et al., 1998). Information about the role of YY1 in VZV replication and infection is limited. We established that YY1 binds to the downstream region of VZV oriS but the mutation of this binding site had no significant effect on either origin-dependent DNA replication or expression of the ORF62 and ORF63 flanking genes. However, mutation of this YY1 site in a VZV recombinant virus was associated with an increase in total virus DNA level in MeWo cells and VZV growth in skin xenografts (Khalil et al., 2008; and Khalil et al., 2012).

YY1 has been shown to physically interact with Sp1 (Lee et al., 1993; Seto et al., 1993; and Shi et al., 1997), raising the possibility that binding could occur between YY1 and Sp1 at their respective sites in the ORF3 promoter. This binding may be important for the recruitment of YY1 to the ORF3 promoter or at least for stabilizing the formation of the YY1 complex. This hypothesis were supported by our finding that mutation of the Sp1/Sp3 site, which was not expected to affect YY1 binding to the ORF3 promoter, not only inhibited the formation of the Sp1/Sp3 containing complexes but also ablated the formation of the YY1 complex. These results also suggest that the effect of the Sp1/Sp3 site mutation on ORF3 expression may be due not only to the inhibition of Sp1 and Sp3 binding to the site but also to the absence of YY1 binding to its site in the ORF3 promoter.

We found that the ORF3 promoter was trans-activated only by IE62 and not by IE4, IE63, ORF10, ORF29 or ORF61 proteins. With the exception of IE4, all of these VZV proteins down-regulated IE62 mediated transactivation of the ORF3 promoter. ORF29 and ORF10 proteins inhibited ORF3 promoter transactivation by about 2-fold and IE63 reduced IE62 activation by about 3-fold. ORF61 exhibited the most significant effect, inhibiting the response to IE62 by about 10-fold. This interplay between the VZV regulatory proteins has been reported in experiments showing that ORF29 and IE63 enhance IE62 activation of the gI promoter in MeWo cells (Boucaud et al., 1998; and He et al., 2001). IE63 also has been shown to up-regulate the IE62-mediated transactivation of the ORF66, ORF61 and gE promoters (Folster et al., 2011; Wang et al., 2009; and Berarducci et al., 2007). In contrast, ORF61 was determined to be the major viral trans-repressor of IE62 trans-activating functions in Vero cells (Nagpal and Ostrove, 1991) even though ORF61 expressed without IE62 activated the ORF4, ORF61 and ORF62 promoters in Vero cells (Moriuchi et al. 1993).

These observations showing that the Sp1/Sp3 and YY1 sites are required for IE62 mediated transactivation of the ORF3 promoter in the absence of any functional TATA box element suggests an alternative mechanism by which IE62 can activate VZV promoters. In this mechanism, the physical interaction between IE62 and cellular transcription factors Sp1 and USF (and probably other cellular transcription factors including Sp3 and YY1) may act as a platform in the presence of IE62 for recruiting the basal transcription machinery to the VZV promoters and the formation of a stable preinitiation complex (PIC) (Fig. 9). This model does not exclude the possibility of the presence of TFIID complex components in the activation of these promoters. This model may help to expand our understanding of the mechanisms used by IE62 to activate transcription of specific VZV genes based on the characteristics of their promoter binding sites for cellular trans-activating factors.

Fig. 9.

A model for the mechanism by which IE62 mediates the transactivation of the VZV promoters during VZV infection. A) In the presence of the TATA box site, the IE62 is able to dimerize, bind to DNA and has the ability to bind to other cellular transcription factors like Sp1, Sp3 and the Mediator complex and activate the initiation of the gene transcription. B) In the absence of a TATA box site, IE62 can utilize other cellular transcription factors binding to sites like that of the Sp1, Sp3 and YY1. These binding sites may act as an alternative to compensate for the absence of the functional TATA box; IE62 binding to DNA and to some of these cellular factors may recruit the basal cellular transcription machinery and activate TATA-less promoters like the ORF3 promoter. This model does not exclude the recruitment of TFIID complex components to the promoter.

Research Highlights.

• We determined the mechanism of ORF3 expression mediated by VZV IE62 protein.

• The ORF3 promoter analysis showed the absence of any functional TATA box.

• The Sp1/Sp3 and YY1 binding sites are required for the ORF3 promoter activation.

• This activation is inhibited by the viral factors ORF10, ORF29, ORF61 and IE63.

• The Sp1/Sp3 site mutant virus has similar growth kinetics to the wild type virus.