The 19S Proteasome Activator Promotes Human Cytomegalovirus Immediate Early Gene Expression through Proteolytic and Nonproteolytic Mechanisms

Cells and viruses.

Human foreskin fibroblasts (HFs), telomerase-immortalized human foreskin fibroblasts (tHFs), and cells depleted of Daxx by short hairpin RNA targeting Daxx (shDaxx) (32) were cultured as previously described (18). Cells were infected with HCMV-AD169 as previously described (18). For small interfering RNA (siRNA) experiments, the cell count of a companion scrambled (Scr) well was used to determine viral inocula for the desired multiplicity of infection (MOI) for all other wells. Cell cycle analysis by flow cytometry has been previously described (33).

Inhibitors and antibodies.

Trichostatin A (TSA); 100 ng/ml in dimethyl sulfoxide ([DMSO]) (Millipore) was added 20 h prior to infection; lactacystin (20 μM in DMSO; Calbiochem) was added at the start of infection. Antibodies used for immunoblotting and chromatin immunoprecipitation (ChIP) assays are listed in Table 1. Goat anti-mouse and goat anti-rabbit secondary antibodies conjugated with horseradish peroxidase were purchased from Chemicon and used for visualizing immunoblots with film as previously described (18). Immunoblots for semiquantitative analysis were also analyzed using a Li-Cor Odyssey Fc imaging system as previously described (13). All quantitative immunoblotting graphs represent the means ± standard errors (SE) of at least three biological replicates and have been analyzed with two-tailed paired t tests.

RNA interference.

Cells were transfected with small interfering RNA oligonucleotides from Dharmacon (Table 1). Transfections were carried out using Lonza nucleofection reagents (VPI-1002; Lonza) or Lipofectamine RNAiMax (13778150; Invitrogen), according to the manufacturers' protocols. In all experiments, cells were serum starved for 24 h after transfection with medium containing 0.1% fetal bovine serum and infected at 72 h posttransfection. When cells were transfected with Lonza reagents, an equal number of HFs were transfected with 80 pmol of siRNA/10⁶ cells (for Rpn1 and β5) and equally distributed among culture dishes. When Lipofectamine RNAiMax was used, an equal number of HFs were transfected with 20 pmol (for Rpn1 and β5), 40 pmol (Rpn11), or 80 pmol (Rpt2) of siRNA/10⁶ cells. In all experiments, concentrations were balanced with scrambled siRNA to equalize siRNA delivery.

Chromatin immunoprecipitations.

Equal numbers of cells were fixed with 1% formaldehyde at room temperature for 8 min. Cross-linking reactions were then quenched with 125 mM glycine for 5 min, and cells were washed with phosphate-buffered saline (PBS). Nuclei were isolated with ChIP lysis/wash buffer (150 mM NaCl, 50 mM Tris-HCl [pH 7.4], 5 mM EDTA [pH 8.0], 0.5% NP-40, 1% Triton X-100, protease inhibitors) and then lysed with ChIP lysis buffer (50 mM Tris [pH 8.1], 10 mM EDTA, 1% SDS, protease inhibitors) and subjected to sonication. Cellular debris was removed by centrifugation, and supernatants were diluted with ChIP dilution buffer (16.7 mM Tris [pH 8.1], 167 mM NaCl, 1.2 mM EDTA, 1.1% Triton X-100, 0.01% SDS, protease inhibitors) for overnight incubation with ChIP antibodies at 4°C. Magnetic protein A/G beads (Millipore) were added for 1 h, and beads were washed once with a low-salt buffer (20 mM Tris [pH 8.1], 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, 0.1% SDS), a high-salt buffer (20 mM Tris [pH 8.1], 0.5 M NaCl, 2 mM EDTA, 1% Triton X-100, 0.1% SDS), and a LiCl buffer (10 mM Tris [pH 8.1], 0.25 M LiCl, 1 mM EDTA, 1% NP-40, 1% DOC); they were washed twice with TE buffer (10 mM Tris [pH 8.0], 1 mM EDTA). Beads were then resuspended in water and incubated at 95°C for 10 min. The reaction mixtures were then reverse cross-linked for at least 30 min with ∼80 mM NaCl and 10 μg of proteinase K at 55°C. Reverse cross-linking was terminated at 95°C for 10 min, and DNA was isolated using a QIAquick PCR purification kit (28104; Qiagen) according to the manufacturer's protocol. Isolated DNA was analyzed by real-time PCR.

Real-time PCR.

DNA and RNA were isolated using DNA and RNA extraction kits (IB47201 and IB47322; IBI Scientific) according to the manufacturer's protocol. To make cDNA, isolated nucleic acids were quantified, and equal amounts of extracted RNA were treated with DNase (M6101; Promega) and converted into first-strand cDNA using a SuperScript III first-strand synthesis system (18080-051; Invitrogen). Primer and probe sets for real-time analysis are listed in Table 2. All real-time graphs represent the means ± SE of at least three independent experiments and have been analyzed for statistical significance using a permutation test (Mstat, version 6.1.2) that utilizes each independent experiment analyzed in triplicate by quantitative PCR (qPCR).

19S RP subunit knockdown inhibits both Daxx degradation and IE1 transcript accumulation.

Depletion of the Daxx protein during HCMV infection, either naturally through degradation by tegument-delivered pp71 or artificially through RNA interference, stimulates viral IE gene expression (18, 34, 35). Correspondingly, stabilization of Daxx through genetic alteration of pp71 or proteasome inhibition inhibits viral IE gene expression (18). Because the 19S RP is required for Daxx degradation (13), we asked if it was also required for viral IE gene expression. Independent knockdown of three distinct 19S RP subunits (Rpn1, Rpt2, and Rpn11) stabilized Daxx and impaired IE1 protein accumulation during HCMV infection (Fig. 1A to D). 19S RP subunit knockdown was confirmed by Western blotting, and inhibition of 19S RP function was confirmed by observing the stabilization of p53, a constitutively degraded 26S proteasome substrate (36). Consistent with the known role for Daxx in suppressing IE1 transcription (18, 32, 37), 19S RP knockdown suppressed the accumulation of IE1 transcripts (Fig. 2A and B). We found no defects in HCMV genome delivery (Fig. 2C and D) or cell cycle synchronization by serum starvation (Fig. 2E) in 19S RP subunit knockdown cells that could explain the observed inhibition of IE gene expression. Thus, we conclude that 19S RP subunit depletion inhibits HCMV IE1 transcription.

Maintenance of the intrinsic defense through 19S RP knockdown is not the sole reason for depressed IE1 transcription.

Daxx institutes an intrinsic defense against HCMV that silences viral IE gene expression. It is inactivated through Daxx degradation by tegument-delivered pp71 or pharmacological histone deacetylase (HDAC) inhibition (18) and is maintained by proteasome impairment through small-molecule inhibition of the 20S CP or 19S RP subunit knockdown (13, 18). The silencing of IE gene expression observed upon 20S CP inhibition can be reversed by Daxx knockdown or HDAC inhibition (18), indicating that the intrinsic defense is the sole or major impediment to IE gene expression in 20S CP-inhibited cells. In contrast, Daxx knockdown (Fig. 3A) or HDAC inhibition (Fig. 3B) failed to rescue IE1 protein accumulation in 19S RP knockdown cells. While HDAC inhibition could rescue IE1 transcription during 20S CP knockdown, it was unable to promote IE1 transcription during 19S RP knockdown (Fig. 3C). We conclude that in addition to maintaining the Daxx- and HDAC-mediated intrinsic defense by disabling pp71-mediated Daxx degradation, 19S RP knockdown triggers an additional restriction to HCMV IE1 transcription.

The 19S RP is recruited to the MIE locus at the onset of HCMV infection and promotes transcriptional elongation.

Inhibition of both the 20S CP and the 19S RP impairs protein degradation and HCMV IE gene expression. The observations that Daxx knockdown or HDAC inhibition can rescue IE gene expression only upon 20S CP but not 19S RP inhibition implies a nonproteolytic role for the 19S RP in HCMV transcription. Indeed, a nonproteolytic role for the 19S RP in cellular gene expression is emerging (38). While mechanisms remain elusive, 19S RP subunits have been observed at transcriptionally active loci (25, 29), and their presence promotes transcriptional elongation (26, 29). Likewise, using immunoprecipitation (ChIP), we found the endogenous 19S RP subunit Rpn7 but not the 20S CP subunit β7 present at the MIE locus, both at the promoter (Fig. 4A) and within exon 3 (Fig. 4B), in HCMV-infected cells. At this time during infection, the UL47 and UL48 genes are not transcribed. Encoded by the same DNA strand, these genes at late times after infection generate transcripts with unique 5′ ends but identical 3′ termini (39). Rpn7 showed a very modest recruitment to the 5′ (promoter) region of the UL47 gene (Fig. 4C) that may represent poising for future transcription, as has been postulated for cellular genes (40). Rpn7 was not recruited to the 3′ end of the UL48 gene (Fig. 4D). Thus, the 19S RP is recruited more substantially to the actively transcribed MIE locus than the transcriptionally inactive UL47/48 locus. The ability of the β7 antibody to immunoprecipitate the endogenous protein (Fig. 4E) instills confidence in the accuracy of the negative ChIP experiments.

We also utilized ChIP experiments to gauge transcriptional elongation by quantifying RNAPII occupancy at three sites within the MIE locus (Fig. 5A and 6A): the promoter, exon 4, and exon 5. Furthermore, we used an antibody that recognizes the nonphosphorylated form of RNAPII (8WG16) or one (H5) that recognizes RNAPII only when it is phosphorylated on serine 2, a mark of a transcriptionally elongating polymerase (41). Knockdown of the 19S RP subunit Rpn1 decreased the occupancy of total RNAPII at the MIEP, exon 4, and exon 5 (Fig. 5B) and of serine 2-phosphorylated RNAPII at exon 5 (Fig. 5C). In contrast, knockdown of the 20S CP subunit β5 impaired RNAPII recruitment to the MIEP but not occupancy at exon 4 or exon 5 (Fig. 6B) and had no effect on exon 5 occupancy of serine 2-phosphorylated RNAPII (Fig. 6C). Western blot controls indicated that proteasome subunit knockdown was achieved and impaired proteasome activity (detected by stabilization of p53) but that it did not affect the steady-state levels of the relevant forms of RNAPII (Fig. 7). From these data we conclude that the 19S RP, but not the 20S CP, facilitates RNAPII transcriptional elongation through the MIE locus in HCMV-infected cells. In total, our work uncovers proteolytic and nonproteolytic roles for the 19S RP in promoting viral IE gene expression at the onset of HCMV lytic infection.