Requirement of the N-Terminal Activation Domain of Herpes Simplex Virus ICP4 for Viral Gene Expression

Abstract

ICP4 is the major activator of herpes simplex virus (HSV) transcription. Previous studies have defined several regions of ICP4 that are important for viral gene expression, including a DNA binding domain and transactivation domains that are contained in the C-terminal and N-terminal 520 and 274 amino acids, respectively. Here we show that the N-terminal 210 amino acids of ICP4 are required for interactions with components of TFIID and mediator and, as a consequence, are necessary for the activation of viral genes. A mutant of ICP4 deleted for amino acids 30 to 210, d3-10, was unable to complement an ICP4 null virus at the level of viral replication. This was the result of a severe deficiency in viral gene and protein expression. The absence of viral gene expression coincided with a defect in the recruitment of RNA polymerase II to a representative early promoter (thymidine kinase [TK]). Affinity purification experiments demonstrated that d3-10 ICP4 was not found in complexes with components of TFIID and mediator, suggesting that the defect in RNA polymerase II (Pol II) recruitment was the result of ablated interactions between d3-10 and TFIID and mediator. Complementation assays suggested that the N-terminal and C-terminal regions of ICP4 cooperate to mediate gene expression. The complementation was the result of the formation of more functional heterodimers, which restored the ability of the d3-10-containing molecules to interact with TFIID. Together, these studies suggest that the N terminus contains a true activation domain, mediating interactions with TFIID, mediator, and perhaps other transcription factors, and that the C terminus of the molecule contains activities that augment the functions of the activation domain.

INTRODUCTION

Herpes simplex virus (HSV) transcription is dependent on the cellular RNA polymerase II (Pol II) transcription machinery and is activated by the viral protein ICP4 (1, 2). ICP4 has a DNA binding activity (3–5) that is crucial for activation (6–8). In addition, ICP4 interacts directly with components of TFIID, such as TATA-binding factor (TBP) and TAF1, and has recently been found in complex with mediator (9, 10). As a consequence, ICP4 stabilizes transcription initiation complexes on viral promoters (11–13). Interactions between ICP4 and cellular transcription factors, such as TFIID, most likely occur on different surfaces of the modular ICP4 molecule. Two distinct regions of the molecule have been implicated in protein-protein interactions; these include the N-terminal 251 amino acids and the C-terminal 520 amino acids (6, 8, 14–17).

An ICP4 mutant lacking the C-terminal activation domain, n208, expresses slightly reduced levels of early (E) genes, but is severely impaired for late (L) gene expression (15). Chromatin immunoprecipitation (ChIP) assays performed on n208-infected cells have shown that RNA Pol II and TBP occupancy was reduced on an early promoter and L promoter occupancy was undetectable, probably accounting for the observed levels of E and L gene expression (13). Interactions between ICP4 and TAF1 of TFIID have been mapped to the C-terminal 520 amino acids of ICP4, suggesting that the transcriptional defects observed in n208-infected cells may in part be the result of the molecule being less able to stabilize TFIID onto promoters (9). In addition, the same 520 amino acids of ICP4 are responsible for the multimerization of ICP4 on DNA, which increases the affinity of ICP4 for DNA (18). The additional removal of the N-terminal transactivation domain results in a mutant, X25, which is completely defective for activation, despite retaining the ability to bind to DNA (19). Together, the previous studies suggest that ICP4 has two regions containing activation domains, one in the N-terminal 274 amino acids and the other in the C-terminal 520 amino acids. However, the activities of the C-terminal transactivation domain in the absence of the N-terminal transactivation domain have not been determined.

To examine the activities of the C-terminal activation domain in the absence of the N-terminal activation domain and to determine the defects resulting from the complete absence of the N-terminal activation domain, a mutant lacking amino acids 30 to 210 (d3-10) was characterized with respect to viral gene expression, interactions with TFIID and mediator, and the ability to enhance Pol II promoter occupancy. The resulting data support that ICP4 has one activation domain contained within the N-terminal 210 amino acids, which can function on its own. We suggest that the C-terminal 520-amino-acid region, while not functioning as an activation domain on its own, augments the activity of the N-terminal activation domain. This hypothesis was further supported by the ability of d3-10 and n208 to complement each other via the formation of more functional heterodimers.

MATERIALS AND METHODS

Cells and viruses.

Vero, E5, D14, and I3 cells (African Green monkey kidney cells) were cultured in Dulbecco's modified Eagle's medium (DMEM) with 5% fetal bovine serum (FBS) and maintained as suggested by ATCC. E5 cells were stably transformed with a wild-type (wt) ICP4-encoding plasmid, pK1-2 (14), and pSV2neo and have been described previously (14). D14 cells were constructed by transfecting Vero cells with 6 μg of a plasmid containing the d3-10 mutant and 2 μg pSV2neo (described below). I3 cells were constructed by transfecting Vero cells with 6 μg of a plasmid encoding tandem affinity-purified (TAP)-d3-10 and 2 μg pSV2neo (described below).

The viruses KOS, d120 (20), n208 (15), and n12 (15) have been described previously and with the exception of KOS were propagated on E5 cells.

Production of stably expressing d3-10 and TAP-d3-10 cell lines.

A total of 2 × 10⁶ Vero cells were transfected with 2 μg pSV2neo and 6 μg d3-10 or TAP-d3-10 plasmid DNA using 15 μl Lipofectamine 2000 according to the manufacturer's suggestions (Invitrogen). The d3-10 plasmid contains the ICP4 d3-10 deletion mutant driven by its own promoter (8). The TAP-d3-10 plasmid was constructed using a Red-mediated recombination system as described by the Osterrieder lab (21, 22) and contains a streptavidin binding peptide and a calmodulin binding peptide (Stratagene) linked in frame to the N terminus of d3-10 ICP4 under the control of the ICP4 promoter. Two days posttransfection, the cells were trypsinized and counted, and approximately 10⁴ cells were plated on 100-mm dishes in DMEM with 10% FBS. The following day the medium was replaced with DMEM containing 10% FBS and 800 μg/ml G418 (Invitrogen), allowing for selection of G418-resistant cells. The medium was replaced every 3 to 4 days for 17 days until colonies of cells of sufficient sizes were visible. Forty colonies were isolated from 9 plates of cells, expanded, and characterized for d3-10 protein expression. Eighteen colonies were isolated from 10 plates of cells, expanded, and characterized for TAP-d3-10 expression. The cultures from isolated colonies were maintained in DMEM with 5% FBS and 500 μg/ml G418. Nine of the 40 clones were positive for expression of the d3-10 protein, and 5 of the 18 clones were positive for expression of the TAP-d3-10 protein, as determined Western blot analysis. An isolate expressing similar levels of the d3-10 protein as wt ICP4 from KOS-infected cells was utilized for the assays described herein. The isolate expressing the most TAP-d3-10 (I3) was utilized for the affinity purification assays.

SDS-PAGE and Western blot analyses.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analyses were performed as previously described (23). Briefly, 10% polyacrylamide Tris-HCl gels (Bio-Rad) were loaded with samples diluted in Laemmli buffer (0.05 M Tris-HCl, 2% SDS, 5% β-mercaptoethanol, 0.01% bromophenol blue, 5% sucrose). Electrophoresis was carried out using Tris-glycine electrode buffer (pH 8.5) (25 mM Tris-HCl, 192 mM glycine, 1% SDS) at 140 V for 1 h. Samples subjected to Western blot analyses were transferred to a nitrocellulose or polyvinylidene fluoride (PVDF) membrane in transfer buffer (25 mM Tris-HCl, 192 mM glycine, 20% methanol) at 250 mA for 1.5 h. Immunoblots were blocked in a 5% milk–TBS-T (Tris-buffered saline-Tween) (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20 [pH 7.5]) overnight. The immunoblots were incubated with primary antibodies for ICP4 (58S at 1:500 or N15 at 1:500), Med1 (sc-8998 at 1:250), TBP (M233R [Covance] at 1:500), or TAF1 (sc-735 at 1:500) for 1 h at room temperature. The probed immunoblots were washed 4 times for a total of 30 min in 1% milk in TBS-T, then incubated with either IRDye-conjugated or horseradish peroxidase (HRP)-conjugated secondary antibodies in TBS-T (1:10,000 or 1:5,000, respectively) for 1 h, rinsed 4 times for a total of 30 min in TBS-T, rinsed twice for 5 min in TBS (50 mM Tris-HCl, 150 mM NaCl [pH 7.5]), and finally visualized using the LI-COR Odyssey infrared imager or developed using the ECL enhanced chemiluminescence kit according to the manufacturers' instructions (Amersham/GE Healthcare).

Metabolic labeling of polypeptides.

Metabolic labeling experiments were performed as previously described (23, 24) with the following modifications. A total of 5 × 10⁵ Vero, E5, or D14 cells were infected with d120 at a multiplicity of infection (MOI) of 10 PFU/cell for 1 h at room temperature. The infections were allowed to progress at 37°C for 3.5, 7.5, and 11.5 h, at which time the cysteine- and methionine-free medium was removed and cells were incubated with 22 μCi [³⁵S]methionine (>1,000 Ci/mM [MP Biomedical]) in TBS for 30 min at 37°C. The labeled proteins were harvested as described above and separated through a stacking gel (4% acrylamide, 0.2% N,N′-diallyltartardiamide [DATD], 0.1% SDS, 119 mM Tris-HCl [pH 7.0]) and then through a separation gel (9% acrylamide, 0.2% DATD, 375 mM Tris-HCl, 0.1% SDS [pH 8.8]) at 100 V. The gel was fixed in fixing solution (water-methanol-acetic acid at a 6:3:1 ratio) and exposed to X-ray film (Amersham/GE Healthcare).

Immunofluorescence assays.

A total of 2 × 10⁵ Vero, E5, or D14 cells were grown on glass coverslips in 12-well trays. The cells were infected at the indicated MOIs for 1 h at room temperature, after which time the inoculum was removed and fresh medium was added. The infections were carried out at 37°C for the indicated period of time. Cells were washed in PBS and then fixed in 4% paraformaldehyde (PFA) for 10 min. The fixed cells were permeabilized in PBS containing 1% bovine serum albumin (BSA) and 0.4% Triton X-100 for 30 min and then blocked in PBS containing 1% BSA for an additional 30 min. The cells were exposed to primary antibodies for 1 h (ICP4-N15 at 1:500, 58S at 1:500, and promyelocytic leukemia protein [PML] at 1:500 [Santa Cruz; sc-5621]), washed in PBS plus 1% BSA for 30 min, and washed in PBS 3 times for a total of 30 min. The samples were then incubated with Alexa Fluor-conjugated secondary antibodies at a 1:500 dilution for 45 min, subsequently washed 6 times for a total of 60 min in PBS, and then mounted onto glass slides using Immu-Mount (Thermo-Fisher). Fluorescence was visualized using the Olympus Fluoview 1000.

Complementation assays.

A total of 5 × 10⁵ Vero, E5, or D14 cells were infected at an MOI of 5 PFU/cell for 1 h at room temperature. The inoculum was removed, the monolayers were washed three times in TBS, and fresh medium was added. At 24 h postinfection (hpi), the infected cells were harvested into the medium, freeze-thawed three times, and sonicated for 45 s, and the viral lysates were clarified by low-speed centrifugation. Viral yields were determined by plaque assay on E5 cells.

Affinity purification.

For affinity purification, 1.5 × 10⁷ Vero, E5, D14, or I3 cells were infected at an MOI of 10 PFU/cell with KOS, n208, TAP-ICP4, TAP-n208, or d120 for 1 h at 37°C. The inoculum was removed, fresh medium was added, and the infected cells were placed at 37°C for an additional 5 h. The monolayers were washed once with TBS plus 0.1 mM Nα-p-tosyl-l-lysine-chloromethyl ketone (TLCK) and scraped into 10 ml TBS plus 0.1 mM TLCK. Nuclear extracts were collected as previously described by Dignam (25). Briefly, the cells were pelleted at 3,000 × g for 5 min and resuspended in 5 ml hypotonic buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl₂, 10 mM KCl). The cells were allowed to swell on ice for 10 min before being Dounce homogenized. Trypan blue staining for intact nuclei was performed to assess the efficiency of cell lysis. The nuclei were pelleted for 10 min at 3,000 × g. Nuclei were resuspended in ½ the pellet volume of low-salt buffer (20 mM HEPES [pH 7.9], 20% glycerol, 1.5 mM MgCl₂, 20 mM KCl), and ⅓ the total volume of high-salt buffer (20 mM HEPES [pH 7.9], 20% glycerol, 1.5 mM MgCl₂, 1.4 M KCl) was added dropwise. Extracts were rotated end over end for 30 min at 4°C and centrifuged at 12,500 × g for 30 min. The nuclear extracts were diluted with 2× low-salt streptavidin binding buffer (20 mM HEPES [pH 7.9], 20 mM KCl, 4 mM EDTA, 0.2% NP-40) and were then combined with 100 μl of washed streptavidin beads (Pierce) and rotated overnight at 4°C. The beads were washed in streptavidin binding buffer (SBB) (20 mM HEPES [pH 7.9], 200 mM KCl, 2 mM EDTA, 0.1% NP-40) 6 times for 10 min, and 500 μl Laemmli buffer was added to the beads to elute and solubilize the complexes. Alternatively, 2 mM biotin in SBB was used to elute the complexes from the beads after the washing and before the solubilization steps. These samples were subjected to SDS-PAGE and Western blot analysis.

RNA extraction and quantification.

RNA extraction and quantification experiments were carried out as previously described with modifications (26). Vero, E5, or D14 cells were infected at an MOI of 10 PFU/cell with the indicated viruses. RNA was extracted from the cells using the RNAqueous-4PCR kit from Ambion and the included protocol.

One microgram of RNA was reverse transcribed to cDNA using the Retroscript kit from Ambion following the included protocol. Briefly, 1 μg total RNA was combined with 2 μl oligo(dT) in a final volume of 12 μl and heated to 85°C for 3 min. The RNA-oligo(dT) mixture was combined with 2 μl 10× reverse transcriptase (RT) buffer, 4 μl (dNTP) mix, 1 μl RNase inhibitor, and 1 μl Moloney murine leukemia virus reverse transcriptase (MMLV-RT). The RNA was reverse transcribed at 42° for 1 h. The MMLV-RT was inactivated by heating the reaction mixture to 92°C for 10 min.

ChIP.

Chromatin immunoprecipitation (ChIP) assays were carried out as previously described (10, 13, 26–28). Briefly, 5 ×10⁶ Vero, E5, or D14 cells were plated onto 100-mm dishes. Cells were infected with the indicated viruses at an MOI of 10 PFU/cell for 1 h at room temperature. Following adsorption, fresh DMEM plus 5% FBS was added to each plate, and the plates were incubated at 37°C for 3 h and 15 min, at which time the medium was removed and formaldehyde was added to the medium to give a final concentration of 1%. The samples were further incubated for 10 min at 37°C. The formaldehyde was quenched by the addition of glycine to the medium to a final concentration of 125 mM. The monolayers were washed twice in TBS and then scraped into TBS. The cells were pelleted by low-speed centrifugation, resuspended in 500 μl SDS-lysis buffer, (1% SDS, 10 mM EDTA, 50 mM Tris HCl [pH 8.1]), and incubated on ice for 30 min. The samples were then sonicated 6 times for 10 s each at a 30% output using a Cole-Palmer ultrasonic processor with microtip. The cell debris was pelleted at 12,500 × g for 15 min, and the sonicated chromatin was diluted 11 times in ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 167 mM NaCl). The samples were precleared twice for 2 h using 120 μl of protein A agarose plus single-stranded DNA (ssDNA) beads (Millipore) each time. The precleared samples were divided into five samples, and the immunoprecipitations were performed overnight using 30 μl protein A agarose plus ssDNA beads and either 5 μl of 58S antibody for ICP4 or 2 μl 8WG16 antibody (Covance) for RNA polymerase II. Beads containing the precipitated protein/DNA were washed twice for 4 min each in low-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 150 mM NaCl), twice for 4 min each in high-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 500 mM NaCl), once for 4 min in LiCl buffer (0.25 mM LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl [pH 8.1]), and three times for 4 min each in Tris-EDTA (TE) buffer (10 mM Tris-HCl [pH 8.0], 1 mM EDTA). The samples were eluted from the beads in 250 μl of elution buffer (1% SDS, 0.1 M sodium bicarbonate) twice for 15 min each. Twenty microliters of 5 M NaCl was added to each 500-μl sample, and samples were then incubated at 65°C overnight to reverse the cross-links. Thirty-one microliters of a proteinase K mixture (1:2:1 0.5 M EDTA–Tris-HCl [pH 6.5]–20 mg/ml proteinase K) was added to each tube, and the samples were incubated at 55°C for 2 h. The DNA was purified using phenol-chloroform-isoamyl alcohol (25:24:1) extractions. The purified DNA was precipitated with ethanol, resuspended in 60 μl DNase/RNase-free water, and analyzed by quantitative PCR (qPCR) (described below).

Real-time qPCR.

qPCRs were set up such that the PCR mixtures contained 7.5 μl SYBR green, 2× PCR mix, a final concentration of 0.3 μM primers (see below), and 3 μl cDNA (from RNA experiments) or DNA (from ChIP experiments) in a total volume of 15 μl. The reactions were carried out in a Step One Plus real-time PCR machine from Applied Biosystems under the following conditions: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. A melt curve was also included under the following conditions: 95°C for 15 s and 60°C for 1 min, followed by +0.3°C to 95°C for 15 min. For cDNA or DNA quantification the primers used were 5′ACCCGCTTAACAGCGTCAACA3′ and 5′CCAAAGAGGTGCGGGAGTTT3′ for TK downstream sequence, 5′ACTTAATCAGGTTGTTGCCG3′ and 5′GAAGTTGTGGACTGGGAAGG3′ for ICP4, or 5′CAGCTGCTTCATCCCCGTGG3′ and 5′AGATCTGCGGCACGCTGTTG3′ for the TK promoter.

RESULTS

In a previous study, we showed that a number of deletions within a region between amino acids 30 and 210 had an effect on the ability of ICP4 to activate transcription, but none completely abolished activation in the context of the otherwise intact ICP4 molecule (23). This region comprises the majority of the amino-terminal transactivation domain without interfering with the DNA binding domain (7, 8, 15) (Fig. 1A). Deletion of the entire region between amino acids 30 and 210 completely eliminated activation, but this was only observed in the context of a molecule truncated at amino acid 774 (23). Repeated attempts to construct a virus with amino acids 30 to 210 deleted in the otherwise intact ICP4 gene were unsuccessful, possibly because the deletion interferes with the action of wt ICP4 provided by the complementing cells. Accordingly, a cell line stably transfected with a plasmid encoding the d3-10 deletion under the control of the wt ICP4 promoter was constructed. Several clonal isolates were tested for expression of the d3-10 protein by Western blot analysis (data not shown), and the D14 clonal isolate was chosen for use in these studies because it expressed a similar level of the mutant ICP4 protein to wt ICP4 expressed in KOS-infected Vero cells (Fig. 1B).

Fig 1.

Characterization of d3-10-expressing cells. (A) Schematic of the primary structure of ICP4 mutant molecules. The sequences in green and red represent the N-terminal transactivation domain and the C-terminal 520 amino acids, respectively. (B) Western blot analysis of a d3-10-expressing cell line (D14 cells). The indicated cells were mock infected or infected with KOS or d120 (MOI, 10 PFU/cell). Infected whole-cell extracts prepared at 6 h postinfection were subjected to Western blot analyses using an antibody against the C terminus of ICP4 (58S). (C) The localization of wt ICP4 and d3-10 was compared to the PML distribution by immunofluorescence assays of the indicated cells infected with d120 at an MOI of 1 PFU/cell. (D) ChIP assay. ChIP was performed as described in the text with antibodies for ICP4 and primers specific to the TK promoter. The data are presented as the percentage of input genomes bound by ICP4. (E) ChIP results as in panel D from 3 independent biological samples.

To determine if deletion of amino acids 30 to 210 interfered with the ability of the molecule to localize and bind to viral genomes, immunofluorescence and chromatin immunoprecipitation (ChIP) assays were performed. It has been previously reported that viral genomes localize to PML bodies in infected cells and that ICP4 localizes juxtaposed to viral genome-containing PML bodies (29, 30). Accordingly, Vero, E5, or D14 cells were infected with d120 at an MOI of 1 PFU/cell and processed for immunofluorescence at 2 hpi (Fig. 1C). The relatively low MOI used in this experiment was necessary to optimize for the observation of ICP4 juxtaposed to PML bodies prior to the degradation of PML mediated by ICP0. The white arrows in the merged panels indicate the juxtaposed ICP4 and PML foci. In most fields, a d3-10 focus was observed next to a PML focus in infected D14 cells, indicating that d3-10 ICP4 localized to sites of viral genome deposition in the same way as wt ICP4.

Chromatin immunopreciptation (ChIP) assays were also performed to determine if d3-10 bound viral DNA in infected cells. Vero and D14 cells were infected with d120, at an MOI of 10 PFU/cell for 3.25 h. KOS-infected Vero cells were used as a positive control in this assay, while i13- and d120-infected Vero cells served as negative controls. The viral mutant i13 produces a full-length ICP4 protein that is defective in binding to DNA due to a 2-amino-acid insertion within the DNA binding domain (31). A monoclonal antibody (58S) that reacts with an epitope in the C terminus of ICP4 (32) was used in the ChIP experiment, and the percentage of genomes containing ICP4 immunoprecipitated with the TK promoter was determined. The results from a representative assay are shown in Fig. 1D. The ChIP assays were performed on three independent biological samples (Fig. 1E). d3-10 and wt ICP4 bound more genomes at the TK promoter than the negative controls in each independent sample. Collectively (Fig. 1), the Western blot, immunofluorescence, and ChIP analyses demonstrated that D14 cells produced the d3-10 protein to similar levels as wt ICP4 in a KOS infection and that the d3-10 protein localized and bound to viral DNA.

The contributions of the region deleted in d3-10 to viral transcription and replication were addressed by complementation, mRNA, and SDS-PAGE analyses. The ability of d3-10 to support viral growth was determined by comparing the yields of d120 on Vero, E5, and D14 cells in a single-step yield experiment. The cells were infected with d120 at a multiplicity of 10 PFU/cell, cultures were harvested 24 hpi, and viral yields were determined by plaque assay on the complementing E5 cell line. Figure 2A shows average viral yields from this assay. d120 infection of the wt ICP4-producing E5 cells produced nearly 10⁸ PFU/ml, while infection of either Vero cells or D14 cells produced about 10⁴ PFU/ml, demonstrating that the d3-10 protein provided no ICP4 complementing activity.

Fig 2.

Activities of the d3-10 mutant. (A) Yield from d120-infected Vero, E5, and D14 cells collected 24 hpi and quantified by plaque assay on the E5 complementing cell line. (B) TK mRNA accumulation in d120-infected Vero, E5, and D14 cells at 4 hpi. RNA was reverse transcribed into cDNA, measured by qPCR as described in Materials and Methods, and the values were normalized to those seen in d120-infected E5 cells. (C) ICP4 mRNA expression levels in infected cells. RNA was harvested from indicated cells infected with the indicated virus (MOI of 10 PFU/cell) at 4 hpi. ICP4 mRNA was quantified as in panel B. (D) Protein synthesis in d120-infected cells. Vero, E5, or D14 cells were infected with d120 at an MOI of 10 PFU/cell for 3.5, 7.5, or 11.5 h, or D14 cells were mock infected (M). The cultures were then labeled with [³⁵S]methionine and analyzed by SDS-PAGE as described in Materials and Methods. Some of the infected cell polypeptides (e.g., ICP22 and ICP27) are identified by number on the right side of the gel. 4* denotes the d3-10 ICP4. The other proteins are abbreviated as follows: 1/2, VP1/2; gB, glycoprotein B; 0, 4, 5, 6, 8, and 25, ICPs 0, 4, 5, 6, 8, and 25 (VP16), respectively.

The inability of the d3-10 protein to complement d120 suggested that d3-10 may be defective for the activation or repression of viral transcription. To investigate this, qRT-PCR was used to determine the relative abundance of TK and ICP4 mRNA present in infected cell extracts at 4 h. Figure 2B shows the percentage of TK mRNA expressed from the d120 genome in infected Vero and D14 cells relative to that in E5 cells. Only about 4% of the amount of TK mRNA expressed in E5 cells was expressed in Vero cells. This represents transcription independent of ICP4 activation and is similar to previously reported levels of TK RNA expression in the absence of functional ICP4 (33). Similarly, infection of D14 cells yielded only about 4% of the amount of TK mRNA as E5 cells, indicating that the d3-10 protein was not capable of activating TK gene transcription.

ICP4 mRNA abundance in n12-infected Vero, E5, and D14 cells was analyzed to determine whether amino acids 30 to 210 contributed to the repression of viral transcription. ICP4 binds to a site within its own promoter (4, 5, 34) and prevents the activation of transcription through the formation of a complex with TFIIB and TBP, thus causing reduced accumulation of ICP4 mRNA (35, 36). The viral mutant n12 expresses only the first 251 amino acids of the ICP4 protein, and as such, it lacks the DNA binding domain, rendering it defective for the repression of transcription (15). n12 infection of Vero cells yielded approximately 3 times more ICP4 mRNA than did infection of E5 cells, demonstrating the expected defect in repression (Fig. 2C). n12 infection of D14 cells yielded similar amounts of ICP4 mRNA to n12 infection of E5 cells, indicating that the d3-10 protein retained repression activity.

While mRNA quantification data indicated amino acids 30 to 210 were dispensable for the repression functions but necessary for the activation functions of ICP4, these conclusions were drawn based on two representative genes. To obtain a better understanding of global expression patterns, protein synthesis profiles from d120-infected Vero, E5, and D14 cells were analyzed. Cells were infected at a multiplicity of 10 PFU/cell, and the synthesized proteins were metabolically labeled and analyzed by SDS-PAGE, as described in Materials and Methods. Consistent with Fig. 2C, there were reduced levels of immediate-early (IE) protein synthesis in d120-infected D14 cells compared to Vero cells (Fig. 2D; see ICP4, ICP0, ICP22, and ICP27). However, little to no E and L protein expression was no evident in d120-infected D14 cells (Fig. 2D; see VP1/2, ICP5, ICP8, gB, and VP16 [ICP25]). Taken together (Fig. 2), the complementation, mRNA abundance, and protein expression data indicated that amino acids 30 to 210 of the N-terminal transactivation domain are necessary for the activation of transcription, but not IE gene repression.

To address the basis for the reduced levels of viral gene expression in d3-10-expressing cells, the ability of Pol II to bind to a representative E promoter was examined using a ChIP assay. Vero and D14 cells were infected with d120, i13, or KOS at an MOI of 10 PFU/cell for 3.25 h. ChIP was performed with an antibody directed against Pol II, and the immunoprecipitated DNA was amplified with primers specific to the TK promoter. A larger percentage of genomes were bound by Pol II at the TK promoter in KOS-infected cells than in D14 cells (Fig. 3A). The percentage of genomes bound by Pol II at the TK promoter in D14 cells was similar to that of the d120 and i13 negative controls. ChIP assays were performed in triplicate on independent biological samples, and the results are shown in Fig. 3B. The absence of Pol II on the representative E promoter in D14 cells suggests that while d3-10 ICP4 can bind to viral DNA (Fig. 1), it does not recruit Pol II, most likely explaining the defects observed in gene expression.

Fig 3.

Binding of RNA Pol II in d3-10-infected cells to the TK promoter. (A) ChIP analysis of the percentage of genomes bound by RNA Pol II at the TK promoter. ChIP was performed as described in Materials and Methods using an antibody to RNA Pol II and primers specific to the TK promoter. (B) ChIP results as in panel A from 3 independent biological samples.

TFIID and mediator are involved in recruiting and stabilizing RNA Pol II to TATA box-containing promoters, and wt ICP4 can be isolated in complexes with these factors (10). To determine whether amino acids 30 to 210 of ICP4 are important for interactions with TFIID and mediator, affinity purification assays using TAP-tagged d3-10-expressing cells were performed. The d3-10 protein was engineered to contain a tandem affinity purification (TAP) tag at the N terminus of the protein, and a stable cell line (I3) expressing the TAP-d3-10 under the control of the wt ICP4 promoter was created. The TAP-tagged protein was purified using a streptavidin-conjugated matrix and eluted with biotin. As a control a TAP-tagged version of wt ICP4 was constructed within the context of the viral genome and also purified over a streptavidin-conjugated matrix. Untagged versions of wt ICP4 from KOS-infected Vero cells and d3-10 from d120-infected D14 cells served as negative controls.

To identify whether components of TFIID or mediator were present in d3-10-purified samples, Western blots with antibodies directed against ICP4, TBP, TAF1, and Med1 were performed (Fig. 4). ICP4, TBP, TAF1, and Med1 were present in all of the nuclear extracts (NE) (Fig. 4). The presence of immunoreactive bands in the pulldown (PD) at the correct molecular weights on the immunoblot of ICP4 indicated that TAP-ICP4 and TAP-d3-10 were isolated using this procedure. However, TBP, TAF1, and Med1 were only present in the PD of the TAP-ICP4 samples and not in the TAP-d3-10 (I3) samples. Collectively these data suggest that the N-terminal 210 amino acids of ICP4 are required for ICP4 to form a complex with TFIID and mediator. Thus, despite the fact that d3-10 ICP4 can bind to the viral genome (Fig. 1), it does not recruit Pol II (Fig. 3), because it cannot form a complex with TFIID and mediator (Fig. 4), most likely explaining the defect in gene expression.

Fig 4.

Interaction of d3-10 with TFIID and mediator. wt ICP4 from KOS, TAP-ICP4 from TAP, d3-10 from D14 cells, and TAP-d3-10 from I3 cells were affinity purified using streptavidin-conjugated matrix as described in Materials and Methods. The presence of ICP4, TFIID, and mediator in these complexes was determined by Western blotting with antibodies directed to ICP4, TBP, TAF1, and Med1.

Different ICP4 mutants can function together in the same cell to complement the defects observed within each mutant (31, 37). The formation of a more functional molecule via intragenic complementation can be attributed in part to the dimeric structure of ICP4 (31). We hypothesized that d3-10, which is missing the N-terminal transactivation domain, and n208, which is missing the C-terminal transactivation domain, could form more functional heterodimers when expressed in the same cell and complement the defects in viral gene expression and growth.

To analyze virus growth, Vero, E5, and D14 cells were infected with n208 at a multiplicity of 5 PFU/cell and harvested at 24 hpi, and viral yields were determined by plaque assay on the complementing E5 cell line. In accordance with previously published data, yields from n208-infected Vero cells were reduced by approximately 40-fold compared to yields from the complementing E5 cell line (Fig. 5A) (15, 23). n208 infection of D14 cells yielded similar levels of viral progeny to infection of E5 cells, indicating that n208 and d3-10 complement.

Fig 5.

Complementation of n208 and d3-10. (A) Growth of n208 on Vero, E5, and D14 cells. The indicated cells were infected with n208 at an MOI of 5 for 24 h. Viral lysates were harvested and titrated on the complementing E5 cells. (B) Immunofluorescence with antibodies specific for d3-10 (58S; green) and n208 (N15; red). The indicated cells were infected with the indicated viruses for 6 h and processed for immunofluorescence as described in Materials and Methods.

The formation of globular replication compartments in the nucleus is a hallmark of viral DNA replication and L gene expression (38). n208 has previously been shown to be defective in the formation of replication compartments (15). To determine if replication compartments formed in n208-infected D14 cells, immunofluorescence assays were performed. To eliminate confusion between the two ICP4 mutants, d3-10 was detected with the monoclonal antibody recognizing an epitope in the C terminus of ICP4, 58S, while n208 was detected with the polyclonal N15 antibody, which was raised against the n208 protein. d3-10 or n208 expression alone did not result in the formation of replication compartments (Fig. 5B, top two rows). d3-10 formed small aggregates in the nucleus that only stained with 58S, consistent with prereplication compartments (39, 40), whereas n208 only stained with N15 and had a diffuse nuclear distribution. However, when D14 cells were infected with n208, both d3-10 and n208 were present in replication compartments (Fig. 5B), as indicated by the staining of the replication compartments with both 58S and N15. The replication compartments formed in n208-infected d3-10 cells were similar in morphology to those in previously published work with wt ICP4 (39, 41) and are consistent with the increased viral yields seen in Fig. 5A.

The costaining of newly generated replication compartments in n208-infected d3-10 cells with 58S and N15 suggested complementation between n208 and d3-10 was possibly the result of the formation of more functional heterodimers. To examine this, we took advantage of the TAP-tagged version of d3-10. First, to establish that d3-10 and n208 could form heterodimers, TAP-d3-10-expressing I3 cells were infected with n208. As a negative control, d3-10-expressing D14 cells were also infected with n208. The TAP-d3-10-containing complexes were affinity purified, and the presence of n208 in these complexes was assessed via Western blot analysis (Fig. 6). The presence of both TAP-d3-10 (green) and n208 (red) in the I3 pulldown lane (Fig. 6, lane 8), indicates that TAP-d3-10 and n208 formed heterodimers that could be isolated by the affinity purification approach.

Fig 6.

The formation of heterodimers between d3-10 and n208 restores interactions with TFIID. ICP4 dimers were isolated using affinity purification over a streptavidin matrix as described in Materials and Methods. The presence of d3-10 and n208 in these heterodimers was determined by Western blotting using antibodies directed against the C terminus of ICP4 (58S; green) for detection of d3-10 and the N terminus of the ICP4 (N15; red) for the detection of n208. The presence of TFIID in complex with the heterodimers was determined by Western blotting with antibodies directed to TBP and TAF1. Lanes labeled “NE” represent nuclear extracts, whereas lanes labeled “PD” represent the pulldowns with streptavidin beads. inf., infected.

To address whether the formation of n208/TAP-d3-10 heterodimers yielded more functional dimers, the presence of TFIID in complex with the affinity-purified heterodimers was assessed. TAP-tagged wt ICP4 was used as a positive control, and untagged versions of wt ICP4 (KOS) and d3-10 from D14 cells were used as negative controls. As expected, both TBP and TAF1 were isolated in complex with TAP-wt ICP4 (Fig. 6, bottom two panels, lane 4), whereas neither TBP nor TAF1 was isolated in complex with TAP-d3-10 alone (Fig. 6, bottom two panels, lane 12, and Fig. 4). However, both TBP and TAF1 were isolated in complex with heterodimers of TAP-d3-10 and n208, indicating interactions between the ICP4 heterodimer and components of TFIID (lane 8). The amount of TBP and TAF1 in complex with TAP-d3-10 and n208 (lane 8) was substantially reduced relative to that seen with TAP-wt ICP4 (lane 4). Taken together, these data support a model in which the complementation of n208 and d3-10 occurs through the ability of the two proteins to form more functional heterodimers.

DISCUSSION

ICP4 has two discontinuous regions that are involved in the transactivation of transcription, one N terminal and one C terminal to the DNA binding domain. Previous studies demonstrated that the C-terminal 520 amino acids are dispensable for the activation of early genes. However, late gene expression and viral growth are impaired (15). The deletion of amino acids 30 to 210 in the absence of the C-terminal 520 amino acids completely abolishes activation and viral growth (23). This study further investigates the requirement and activities of the N-terminal transactivation domain with respect to viral gene transcription and the possible sufficiency of the C-terminal transactivation domain. We found that the N terminus of ICP4 was necessary for ICP4 activation of transcription, as the C terminus and the DNA binding domains were unable to activate transcription. We conclude that the region between amino acids 30 and 210 contains the major activation domain of ICP4 and that the C-terminal segment of 520 amino acids, while not functional as an activation domain by itself, specifies activities that augment the function of the activation domain specified by amino acids 30 to 210. These observations and conclusions are further discussed below.

It has been previously demonstrated that ICP4 interacts with components of TFIID and mediator (9, 10) to stabilize the transcription initiation machinery on HSV promoters (11, 13, 42). The data presented herein demonstrate that amino acids 30 to 210 of the N-terminal transactivation domain are required for formation of complexes with components of TFIID and mediator. This region of ICP4 contains several 10- to 20-amino-acid stretches that are highly conserved with the ICP4 homologs in other alphaherpesviruses, which individually contribute to viral growth and activation (23). Therefore, it is likely that the region coded for by amino acids 30 to 210 comprises a complex activation domain that may make multiple contacts with components of TFIID, mediator, and possibly other factors. Only the deletion of the entire region, as in d3-10, results in a molecule that does not activate transcription.

d3-10, which consists of the DNA binding domain with the entire C-terminal region of ICP4, had no activation function. Previous studies have shown the C terminus of ICP4 is required for interaction with isolated TAF1, while the N-terminal 774 amino acids were sufficient for interactions with TBP, both components of TFIID (9, 36). d3-10 did not form a complex with TAF1 in infected cells (Fig. 4 and 6). It is possible that an interaction between the C terminus and TAF1 alone is not sufficient to result in a stable complex in infected cells and that the amino-terminal activation domain is required. Therefore, the interaction with TAF1, which is specified by the C terminus, may augment the function of the amino-terminal activation domain. Alternatively, the conformation of the C terminus in the absence of the amino-terminal activation domain may be different from that of the wt molecule, thus affecting the interaction with TAF1. Another way the C-terminal region of ICP4 may augment the function of the amino-terminal activation domain is to affect the binding of ICP4 to viral genome. We have previously shown that the C-terminal 520-amino-acid region promotes the multimerization of ICP4 on DNA, thus increasing its affinity (18).

ICP4 exists as a dimer within infected cell nuclei (43). Previous studies have shown that individual defects in different ICP4 mutants can be complemented by the formation of more functional heterodimers (31, 37). Some mutants also interfere with the function of wt ICP4 by the formation of less functional heterodimers (19). We hypothesized that if the C-terminal region of ICP4 functions to augment the activities of the amino-terminal activation domain, then mutants individually lacking one of these regions might complement each other by the formation of more functional heterodimers. Infection of d3-10-expressing cells with n208 yielded complementation at the level of viral production (Fig. 5) and importantly resulted in n208/d3-10 heterodimers capable of interacting with TFIID (Fig. 6). This demonstrated not only that one functional N-terminal transactivation domain and one C-terminal transactivation domain are sufficient for viral replication, but also that the N terminus and C terminus of ICP4 cooperate to mediate the activities of ICP4. Additionally, this implies that the N-terminal activation domain and C-terminal region of opposite molecules within the dimer function together, perhaps suggesting a conformation in which they are in close proximity within the dimer.

The data presented here demonstrate that the N-terminal transactivation domain is absolutely necessary for ICP4-mediated transcription. The region between amino acids 30 and 210 specifies interactions with components of TFIID and mediator. The C-terminal region of ICP4 along with the DNA binding domain has no activation function. We propose that the region between amino acids 30 and 210 constitutes a complex activation domain, possibly specifying multiple interactions with cellular factors such as TFIID, mediator, and others. The C-terminal region, while not an activation domain on its own its own, augments the function of the N-terminal activation domain by specifying additional interactions, such as with TAF1, or by affecting the binding of ICP4 to the viral genome.