Breaking Human Cytomegalovirus Major Immediate-Early Gene Silence by Vasoactive Intestinal Peptide Stimulation of the Protein Kinase A-CREB-TORC2 Signaling Cascade in Human Pluripotent Embryonal NTera2 Cells

Cells and viruses.Human NTera2/D1 cells (NT2 cells) (21) were grown in Dulbecco's modified Eagle's medium (DMEM) (containing 4 mM glutamine and 4.5 g/liter glucose) supplemented with 10% knockout serum replacement (Life Technologies, Rockville, MD) and 3 to 4% charcoal-treated fetal bovine serum, as described previously (54). NT2 cells were free of mycoplasm per the MycoAlert mycoplasm detection assay (Lonza Walkersville, Inc., Walkersville, MD). Background levels of NT2 cell differentiation and HCMV MIE expression were kept to a minimum by daily changes of the growth medium and avoidance of trypsin, cell aggregation, overly confluent adherent cell growth, and low-density seeding. When indicated, knockout serum replacement was removed from the medium to further lower the frequency of NT2 cells permitting MIE expression. Differentiation of NT2 cells was achieved by the addition of 10 μM RA (Sigma, St. Louis, MO) to DMEM plus 10% fetal bovine serum for ≥7 days. Cellular differentiation medium was replaced with NT2 growth medium at least 1 day prior to performing experiments. Human foreskin fibroblasts (HFF) were isolated and grown as described previously (56).

HCMV strain Towne was propagated on HFF. HCMV-GFP (Towne strain) represents rΔ−640/−1108gfp (49), which contains the green fluorescent protein gene driven by the viral native UL127 promoter, which is expressed with early/late kinetics. rWT and rCRE_M5 (Towne strain) were characterized previously (36). All HCMV isolates were partially purified by centrifugation of infected HFF supernatant filtrate (0.45-μm filter) through a 20% sorbitol cushion in phosphate-buffered saline (PBS); viruses were resuspended in DMEM or RPMI without serum. Virus absorption was carried out for 90 min, and cells were subsequently washed twice or thrice with Hanks' balanced salt solution without calcium and magnesium.

FSK (10 mM) and H-89 (10 mM) stocks (Sigma, St. Louis, MO) were prepared in dimethyl sulfoxide. Phorbol 12-myristate 13-acetate stock (PMA) (50 μM; Sigma) was prepared in ethanol. VIP (100 μM stock; Calbiochem/EMD Biosciences) was dissolved in PBS containing 1% human serum albumin.

Plasmids, transfection, and RNAi.Human TORC2 cDNA was cloned into the SgfI-MluI restriction sites of pCMV6Entry vector (Myc/Flag tagged) (Origene Technologies, Inc., Rockville, MD) to produce TORC2 that is C-terminally tagged with Myc/Flag. DNA sequencing of the resultant constructs confirmed the correct TORC2 cDNA sequence and fused gene configuration. The S171A substitution in TORC2-Myc/Flag was created using QuikChange multisite-directed mutagenesis (Stratagene) and the oligonucleotide CGA CTA CCA TCT GCA CTT AAC AGG ACA AGC gCT GAC TCT GCC CTT CAT. RNA interference (RNAi) experiments entailed the use of Suresilencing short hairpin RNA (shRNA) plasmids (SA BioSciences, Frederick, MD) containing the simian virus 40 promoter-puromycin cassette and the U1 promoter driving expression of shRNA against CREB (5′-TCATCTGCTCCCACCGTAACT-3′), ATF-1 (5′-AGTACAGGGACTTCAGACATT-3′), the PKA catalytic subunit (g, 5′-GTGGGTTGCTTGCTAGAATTTTT-3′; and y, 5′-GCCCTGGGGGTTCTTATCTCTATGA-3′), PKCδ (5′-CACCCAGAGACTACAGTAACT-3′), or TORC2 (r, 5′-CCAGGTTTCTCTAAGGAGATT-3′; b, 5′-TGAAGTCCCTGGAATTAACAT-3′; y, 5′-GCAGCGAGATCCTCGAAGAAT-3′; and g, 5′-CAGCAGCTGCCCAAACAGTTT-3′). Negative control shRNA consisted of the sequence 5′-GGAATCTCATTCGATGCATAC-3′.

NT2 cells in 30-mm² dishes or 12-well plates were transfected using endotoxin-free plasmid DNA and a combination of Lipofectamine LTX and Plus reagent (Invitrogen) according to the manufacturer's directions. Adherent cells widely dispersed as single cells yield a transient transfection efficiency of 60 to 80%. Approximately 0.5 to 1 million cells in a 30-mm² dish were transfected with 2.0 to 2.5 μg of plasmid DNA, and the numbers were scaled down proportionately for 12-well plates. The DNA-lipid complexes were added to growth medium lacking antibiotics and incubated with cells for 6 and 24 h for TORC2-F and shRNA studies, respectively. For RNAi studies, HCMV infection was performed either 2 h before or 48 h after transfection with plasmids expressing shRNA. In the former situation, VIP stimulation was carried out from 48 to 72 h postinfection (p.i.). In the latter situation, VIP stimulation was carried out from 2 to 26 h p.i. In both situations, the RNAi-mediated knockdown was carried out for 72 h and the VIP stimulation was performed for 24 h. Knockout serum replacement was omitted from the growth medium throughout the stimulation period. The conditions for shRNA-medicated CREB and ATF-1 knockdowns were determined empirically to achieve the end point of reductions in levels of CREB and ATF-1 individually, without depletion of ATF-1 and CREB, respectively. Attention to this detail was necessary because large reductions in CREB produced by the CREB-specific shRNA were found to also mildly decrease the ATF-1 level, and aggressive knockdown of ATF-1 by ATF-1-specific shRNA mildly decreased the CREB level. Similar results were obtained by using shRNA molecules targeting different sequences in CREB or ATF-1 mRNA, and these results did not seem to correspond to the degree or characteristic of CREB and ATF-1 shRNA nucleic acid sequence similarities between counterpart ATF-1 and CREB mRNA sequences, respectively. TORC2 expression plasmids were transfected for 48 h and, in some experiments, subjected to infection and VIP stimulation at the indicated times.

Nucleic acid analyses.HCMV DNA from each sample was quantified in quadruplicate by a TaqMan real-time PCR method using ABI Prism 7700 and 7000 sequence detection systems (Applied Biosystems, Foster City, CA). A MIE enhancer-specific primer set and probe was used (37). Nuclei and DNA were prepared using previously described methods (56). Primers and probes for quantification of cellular glyceraldehyde-3-phosphate dehydrogenase DNA were purchased from Applied Biosystems. The standard curve method was applied to determine the relative concentration of target DNA in relationship to the cycle threshold value.

Whole-cell RNA was isolated according to the method of Chomczynski and Sacchi (10). cDNA was made using Superscript II RNase H⁻ reverse transcriptase (RT) (Life Technologies, Gaithersburg, MD) and random hexamers, and then each sample was quantified in quadruplicate by real-time PCR, using previously reported primers, probe, and amplification conditions (55). The primers target MIE exons 1 and 2 to produce an MIE amplicon spanning intron A. The cDNAs derived from ribosomal 18S, TORC1, and TORC2 RNAs were amplified and detected with reagents supplied by Applied Biosystems. The standard curve method was applied to determine the relative concentration of target cDNA in relationship to the cycle threshold value.

Protein analyses.Western blots of whole cell, cytoplasmic, or nuclear extracts were performed using methods reported previously (37, 56). Proteins were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 6, 7.5, or 10% Tris-glycine gels and transferred to Protran BA85 nitrocellulose membranes (Schleicher & Schuell BioScience, Keene, NH). Blots were incubated overnight at 4°C with primary antibodies in 1× PBS containing 5% dried milk, incubated with peroxidase-conjugated goat anti-rabbit or -mouse immunoglobulin G (IgG) antibody at room temperature for 90 min or at 4°C overnight (1:1,000 to 1:5,000 dilution in PBS containing 5% dried milk), treated with SuperSignal West Femto maximum sensitivity substrate (Pierce Biotechnology, Inc., Rockford, IL), and exposed to autoradiography film. Primary antibodies were used at 1:500 to 1:2,000 dilutions and included rabbit polyclonal anti-CREB (Upstate Biotechnology, Lake Placid, NY), rabbit monoclonal anti-CREB (Epitomics, Burlingame, CA), rabbit polyclonal anti-phospho-CREB (Upstate Biotechnology), monoclonal murine anti-ATF-1 (25C10G) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), polyclonal rabbit anti-PKA catalytic subunit (Santa Cruz Biotechnology), murine monoclonal anti-beta tubulin (E7) (University of Iowa Hybridoma Bank, Iowa City, IA), rabbit monoclonal anti-Oct4 (Epitomics), rabbit monoclonal anti-DAXX (Epitomics), rabbit polyclonal anti-CRTC2 (anti-TORC2; ProteinTech Group, Inc., Chicago, IL), and murine monoclonal anti-Flag (OriGene Technologies, Inc., Rockville, MD). HCMV IE1 p72 and IE2 p86 were detected using monoclonal murine antibody MAB810 (Chemicon International), which reacts to an epitope in both proteins. The rabbit polyclonal anti-phospho-CREB antibody against phospho-Ser133, in CREB residues 126 to 136, cross-reacts with ATF-1 phospho-Ser63 (Upstate Biotechnology) (37). Densitometry analysis was performed using the NIH ImageJ 1.34s program.

For immunofluorescence assay (IFA) of HCMV MIE protein and TORC2-Flag, duplicate samples of cells were fixed in ice-cold methanol for 5 min, permeabilized with 0.3% Triton X-100 in PBS for 5 min, and blocked with 10% goat serum in PBS for 30 min at room temperature. Murine monoclonal antibody against the Flag tag (1:200 dilution) (OriGene Technologies, Inc.) or the HCMV IE1 p72 and IE2 p86 proteins (MAB810) (Chemicon International, Charlottesville, VA) (1:1,000 to 1:2,000 dilution) was applied in 10% goat serum in PBS for 1 h at 37°C or overnight at 4°C. Secondary goat anti-mouse antibody conjugated to Alexa Fluor 555 (Molecular Probes Invitrogen, Eugene, OR) was applied at 1:1,000 to 1:5,000 dilutions for 1 h at room temperature. Anti-HCMV IE1 p72 and IE2 p86 antibody conjugated to Alexa Fluor 488 (Chemicon International) was applied after detection of TORC2-Flag by indirect immunofluorescence for the colocalization of these proteins. Cells were counterstained with 4′,6-diamidino-2-phenylindole (1 mg/ml; DAPI). Images were captured by an inverted Olympus IX 51 fluorescence microscope equipped with an X-Cite 120 fluorescence illumination system or by an inverted Zeiss 519 laser scanning confocal microscope. The ratio of MIE-expressing cells (Alexa Fluor 555 positive) to total cells (DAPI positive), with calculated mean and standard deviation (SD), was determined with NIH ImageJ 1.34s software applied to three to eight representative captured images, each comprising the entire field of original magnification at ×20. In the case of manual counting, paired images of MIE- and DAPI-positive NT2 cells were divided into 50 zones of equal size. Ten zones were randomly selected for manual counting of MIE- and DAPI-positive cells. The mean and SD were calculated for the percentage of MIE-positive cells among DAPI-positive cells.

VIP activates HCMV MIE gene expression.The screening of biological agonists of the cAMP/CRE signaling pathway for the ability to activate HCMV gene expression in the NT2 model identified VIP as a candidate. VIP is a 28-amino-acid peptide induced by various stimuli for limited production in multiple human organs and tissues (24, 46) wherein HCMV also replicates. Its levels are increased in disease states (4, 16, 32) in which HCMV also reactivates with greater frequency (26, 33, 40, 43, 51, 80). VIP receptors are expressed on progenitors of neuronal, hepatic, retinal pigment epithelial, and hematopoietic cells (6, 7, 35, 82). On the basis of this information, VIP was further studied in the HCMV reactivation model.

Quiescent HCMV infection was established in NT2 cells after inoculation with HCMV particles partially purified by zonal centrifugation (37, 54). Infection by this method has been shown to yield >98% of NT2 nuclei containing HCMV pp65 at 1 h p.i. at a multiplicity of infection (MOI) of 3 to 5 (37) and approximately three HCMV genome equivalents per nucleus at 24 h p.i. at an MOI of 10 (56). As depicted in Fig. 1A, quiescently infected NT2 cells (MOI of 3 to 5) were subjected to stimulation with VIP at 24 h p.i., and the level of HCMV MIE RNA or protein production was assessed thereafter. Increasing VIP concentrations in the range of 1 to 500 nM resulted in commensurate increases in the level of expression of HCMV spliced MIE RNA (Fig. 1B). As little as 1 nM VIP increased this viral message 3.6-fold at 6 h poststimulation. The 100 nM VIP concentration, which produced a submaximal response (16.4-fold), was applied in subsequent experiments. As shown in Fig. 1C, MIE RNA levels rose 9.9-fold as early as 2 h after VIP treatment. The increase in MIE RNA was greatest at 4 h posttreatment (34.2-fold) and waned at 6 h (19.4-fold), possibly in response to negative autoregulation by the viral MIE gene product, IE2 p86. VIP's actions translate into an approximately 23.5-fold increase in the proportion of infected NT2 cells expressing HCMV MIE protein at 24 h poststimulation (∼21% of cells express MIE [MIE⁺ cells]) compared to the unstimulated infected NT2 population, which was composed of 0.9% MIE⁺ cells at baseline in this particular experiment (Fig. 1D). The degree of change between these two groups increases when the background level of MIE⁺ cells in the unstimulated NT2 population decreases, as illustrated in subsequent studies. The overall profile of findings closely resembles that for 10 μM FSK stimulation (37), although VIP appears to generate a lower level of maximal output for MIE gene expression (Fig. 1B and C).

• Open in new tab
• Download powerpoint

FIG. 1.

VIP activates HCMV MIE gene expression, resembling the outcome of FSK stimulation. (A) Schematic diagram of experimental design using the quiescently infected NT2 cell model. The conditions and limitations of this model have been detailed previously (37, 54). Infections were carried out at an MOI of 3 to 5 PFU per cell. (B and C) Dose-response (B) and time course (C) data for VIP-induced expression of HCMV-spliced MIE RNA, as measured by real-time RT-PCR and normalized to cell 18S RNA (MIE RNA_N). For panel C, VIP (100 nM) and FSK (10 μΜ) were used at the indicated concentrations. (D) IFA for determination of NT2 cell population expressing HCMV MIE proteins (IE1 p72 and IE2 p86) at 24 h of VIP stimulation, with DAPI staining of cell nuclei. A mean of 21.2% (±4.9%) of VIP-stimulated NT2 cells expressed MIE proteins, whereas 0.9% (±0.2%) of unstimulated NT2 cells expressed these proteins, as determined by manual counting (see Materials and Methods).

Viral CRE are required for VIP-induced HCMV MIE gene expression.The importance of the viral MIE CRE in VIP-induced MIE RNA expression was gauged by comparison of wild-type (WT) HCMV and rCRE_{5 M}. rCRE_{5 M} differs from the parental WT in having only two base substitutions in each of the five CRE in the MIE enhancer (36) (Fig. 2A). The experimental controls for these studies revealed that WT HCMV and rCRE_{5 M} yield equivalent amounts of MIE protein in acutely infected HFF and NT2-D cells (Fig. 2B and D) and of viral DNA in nuclei (Fig. 2C) of unstimulated NT2 cells, in accord with previously reported results (36, 37). In contrast, in quiescently infected NT2 cells, the MIE CRE mutations greatly diminished the levels of VIP-induced expression of MIE protein (IE1 p72 and IE2 p86) (Fig. 2B) and RNA (Fig. 2C), as well as the incidence of MIE⁺ NT2 cells (Fig. 2D). The CRE have a less prominent role in mediating MIE gene activation in response to stimulation with PMA (Fig. 2D), a known activator of classical and novel PKC isoforms having the capability of stimulating CRE-mediated gene expression in other systems. This result suggests that PMA acts through both CRE-dependent and -independent mechanisms in reversing MIE gene silence.

• Open in new tab
• Download powerpoint

FIG. 2.

VIP's effect requires the HCMV MIE CRE. (A) Schematic diagram of HCMV rCREB_{5 M} (5M) compared to WT virus. rCREB_{5 M} has mutations in all five copies of the CRE in the MIE enhancer. The proximal CRE is a variant AP1/CRE site that functions like the four consensus CRE in response to FSK stimulation (83). (B to D) NT2 cells, HFF, and differentiated NT2 (NT2-D) cells were inoculated with equivalent titers of WT and rCREB_{5 M} HCMV (MOI of 4 for NT2 and NT2-D cells and 0.1 for HFF). MIE proteins (IE1 p72 and IE2 p86) (B) and spliced MIE RNA (C) were analyzed by Western blotting and real-time RT-PCR, respectively, 24 h after stimulation with VIP (100 nM), FSK (10 μM), or nothing. Duplicate experimental samples were pooled for Western blot analysis. MIE proteins were analyzed at 24 h p.i. for HFF and at 48 h p.i. for NT2-D and NT2 cells. The MIE RNA level was normalized to the 18S RNA level (MIE RNA_N). Means and SD for triplicate experimental samples depict changes in HCMV MIE RNA_N relative to that in infected NT2 cells receiving no stimulation. HCMV DNA in cell nuclei at 24 h p.i. was also quantified by real-time PCR and normalized to cellular 18S DNA (HCMV DNA_N). Means and SD for triplicate experimental samples depict changes in the HCMV DNA_N level. (D) IFA of MIE⁺ NT2 cell density after stimulation with VIP (100 nM), PMA (20 nM), or nothing for 24 h. Original magnification, ×20. The bar graph depicts ratios of MIE⁺ cells to all cells. Means and SD were determined from three random fields, using NIH ImageJ 1.34s software.

Thus, the viral CRE are required for VIP-induced MIE gene expression in quiescently infected NT2 cells, whereas they are dispensable for MIE gene expression in actively infected HFF and NT2-D cells.

VIP triggers a wave of CREB Ser133 and ATF-1 Ser63 phosphorylation but does not change levels of Oct4, hDaxx, and HCMV genome penetration.The potential mechanisms by which VIP overcomes HCMV MIE gene silence were investigated. VIP stimulation increases the levels of phosphorylated CREB Ser133 (pCREB) and ATF-1 Ser63 (pATF-1) but not the levels of total CREB and ATF-1 (Fig. 3A and B). pCREB and pATF-1 levels increased within 10 to 15 min of VIP stimulation and then declined to their respective baseline values by 3 to 8 h poststimulation. The VIP-induced pATF-1 level peaked greatly above its very low baseline value in unstimulated NT2 cells. In contrast, VIP stimulation only modestly increased the pCREB level above the high level of pCREB at baseline. The HCMV MIE proteins, IE1 p72 and IE2 p86, were not detected in unstimulated NT2 cells by Western blotting but became evident by 8 h following VIP stimulation. VIP does not lower the level of cellular Oct4 during this time frame, thus indicating that the cellular differentiation program is not activated (50) and therefore is not responsible for increasing MIE protein expression. It was reported previously that HCMV pp71 is sequestered in the NT2 cytoplasm but that when it is allowed into the nucleus, pp71 targets cellular hDaxx for inactivation and degradation and thereby relieves repression of the MIE genes (69). We show in Fig. 3B that VIP does not appreciably change the hDaxx level in the NT2 model of quiescent infection. Whether VIP modifies hDaxx or pp71 function has not been determined. Lastly, VIP does not appear to enhance viral genome penetration, because the amount of HCMV DNA reaching cell nuclei was unchanged by this stimulus (Fig. 3C).

• Open in new tab
• Download powerpoint

FIG. 3.

CREB Ser133 and ATF-1 Ser63 phosphorylation levels increase, whereas levels of HCMV genomes in nuclei, cellular Oct4, and cellular hDaxx do not appreciably change. (A and B) Western blot analyses of the indicated proteins in NT2 cells in relation to various times of exposure to VIP (100 nM). Duplicate samples were pooled for analysis. The amino acid residues surrounding Ser133 of CREB (43 kDa) and Ser63 of ATF-1 (38 kDa) are nearly identical, thus allowing use of the same phospho-specific antibody to detect both pCREB and pATF-1. MK, mock-infected cells. (C) HCMV DNA in cell nuclei at 24 h of VIP (100 nM) stimulation, as determined by quantitative PCR and normalized to cellular glyceraldehyde-3-phosphate dehydrogenase DNA (HCMV DNA_N). Means and SD for triplicate experimental samples depict changes in HCMV DNA_N relative to that in infected NT2 cells receiving no stimulation (none).

This set of experimental results strengthens the perspective that VIP stimulates a signaling cascade that acts through the viral CRE to reverse MIE gene silence. The action is accompanied by a wave of phosphorylation of CREB Ser133 and ATF Ser63 that is possibly involved in CRE-dependent MIE gene expression.

Cellular PKA and CREB are required for VIP-induced HCMV MIE gene expression.While VIP's effect on NT2 cells has not been reported, VIP is known to activate target cellular genes through signaling pathways that involve PKA, PKC, or both, depending on cell type (9, 27). The type of stimulus and the cellular phenotype determine whether PKA (64) or PKC (30, 77) acts through the CRE to increase MIE enhancer/promoter activity in transient assays.

Two different experimental approaches were applied to determine the roles of PKA and PKC in mediating VIP's actions in quiescently infected NT2 cells. We showed that H89, a selective inhibitor of PKA activity, greatly attenuates VIP's ability to increase the MIE⁺ NT2 population density (Fig. 4A) and MIE protein expression across all cells, in a dose-dependent manner (Fig. 4B). H89 also returned VIP-induced pCREB and pATF-1 levels to baseline values (Fig. 4B). As expected, PKA levels were not changed by H89 (Fig. 4B). Pharmacologic inhibition of classical and novel PKC isoforms did not appreciably decrease VIP's ability to elevate MIE protein expression (data not shown).

• Open in new tab
• Download powerpoint

FIG. 4.

Either pharmacologic inhibition or depletion of cellular PKA greatly attenuates VIP-induced HCMV MIE gene expression. (A and B) H89, an inhibitor of PKA, was added 30 min before and throughout VIP (100 nM) stimulation. MIE proteins (IE1 p72 and IE2 p86) were detected by IFA (A) or Western blotting (B) at 24 h of VIP or no (none) stimulation. MK, mock-infected cells. (B) VIP-induced IE1 p72 levels were decreased 3.1- and 50.9-fold by 2 and 5 μM H89, respectively, when normalized to β-tubulin. PKA, pCREB, and pATF-1 levels were analyzed in relationship to MIE protein levels. (C to E) NT2 cells were transfected with a plasmid expressing shRNA against the PKA catalytic subunit (PKAi_g or PKAi_y), PKCδ (PKCi), or negative control (NCi) for 72 h. HCMV infection was carried out after (C and D) or before (E) PKA knockdown, as described in Materials and Methods. Expression of IE1 p72 and IE2 p86 in response to VIP stimulation (100 nM) for 24 h was analyzed by Western blotting (D and E) and IFA (C). (C) Original magnification, ×20. The bar graph depicts the ratio of MIE⁺ cells to all cells. Means and SD were determined from three random fields. (D) PKAi_g and PKAi_y decreased PKA levels 7.6- and 4.4-fold, respectively; PKCi decreased the PKCδ level 3.8-fold. VIP-induced IE1 p72 levels decreased 37.4-, 23.3-, and 2.3-fold as a result of PKAi_g, PKAi_y, and PKCi, respectively. Values were normalized to β-tubulin. Duplicate experimental samples were pooled for Western blot analyses.

We next tested the effect of depletion of the endogenous PKA catalytic subunit on VIP-induced MIE gene expression by using plasmids expressing shRNA that specifically targets the PKA catalytic subunit. The depletion of the PKA catalytic subunit by this RNAi method markedly attenuated VIP's ability to increase the MIE⁺ NT2 population density (Fig. 4C) and MIE protein production (Fig. 4D) compared to negative control RNAi. This outcome was unchanged by using different RNAi target sequences to deplete the PKA catalytic subunit (Fig. 4D) or by introducing the RNAi after the infection (Fig. 4E). Knockdown of PKCδ by RNAi lowered VIP-induced MIE gene expression only 2.3-fold, in contrast to the 23.3- to 37.4-fold inhibition resulting from PKA knockdown (Fig. 4D), which might reflect cross talk between these two signaling pathways. These combined results indicate that PKA activation is one of the primary drivers in VIP-induced MIE gene expression, whereas PKC appears to contribute little.

Whether CREB, ATF-1, or both are involved in CRE-dependent MIE gene expression resulting from VIP or FSK (37) stimulation is not known. Therefore, CREB and ATF-1 were selectively depleted by RNAi to assess the relative roles of each of these transcription factors, using experimental conditions detailed in Materials and Methods. As shown in Fig. 5A, the CREB-specific shRNA (CREBi) lowered the CREB level 3.6-fold and did not decrease the ATF-1 level, yet it produced a 13-fold decrease in VIP-induced MIE gene expression compared to negative control shRNA (NCi). The CREB knockdown did not perturb the hDaxx level (Fig. 5A). Infecting cells prior to CREB knockdown also blocked VIP's ability to activate MIE gene expression (Fig. 5B), indicating that CREB functions at the postentry stage of infection to increase MIE gene expression. The depletion of CREB translated into lower pCREB levels, as expected (Fig. 5B and data not shown). In contrast, the knockdown of ATF-1 to an undetectable level via ATF-1-specific shRNA (ATFi) resulted in negligible to minimal changes in VIP-induced MIE protein expression compared to that with NCi or a dissimilar ATF-1-specific shRNA that fails to deplete ATF-1 (ATFi_NC) (Fig. 5C). The CREB level was not appreciably changed by ATFi or ATFi_NC. The VIP-induced increase in MIE⁺ NT2 population density was likewise diminished by CREB knockdown and changed negligibly to minimally by ATF-1 knockdown (Fig. 5D and data not shown). Taken together, the results point to CREB having the major role in VIP-induced MIE gene expression. They do not eliminate the possibility that CREB also acts via an indirect mechanism.

• Open in new tab
• Download powerpoint

FIG. 5.

VIP's effect requires cellular CREB. (A to D) NT2 cells were transfected with a plasmid expressing shRNA against CREB (CREBi), ATF-1 (ATFi), or negative control (NCi) for 72 h. HCMV infection was performed after (A, C, and D) or before (B) CREB/ATF knockdown, as described in Materials and Methods. Expression of IE1 p72 and IE2 p86 in response to VIP stimulation (100 nM) for 24 h was analyzed by Western blotting (A to C) and IFA (D). Duplicate samples were pooled for Western blot analyses. (A) CREB, ATF-1, and hDaxx levels were analyzed in relationship to MIE protein levels. CREBi decreased the VIP-induced CREB level 3.6-fold, the IE1 p72 level 13.0-fold, and the ATF-1 level 0.21-fold compared to NCi and normalized to tubulin. (B) pCREB levels were analyzed in relationship to MIE protein levels. (C) ATF-1 and CREB levels were analyzed in relationship to MIE protein levels. Compared to NCi, ATFi decreased IE1 p72 and IE2 p86 levels 1.2- and 2.1-fold, respectively. ATFi_NC decreased IE1 p72 and IE2 p86 levels 1.2- and 1.8-fold, respectively. Results were normalized to the CREB level. (D) RNAi-mediated depletion of CREB and ATF-1, performed using the methods described for panels A and C. Original magnification, ×20. The bar graph depicts the ratio of MIE⁺ cells to all cells. Means and SD were determined from three random fields.

VIP's effect on HCMV MIE gene expression requires cellular TORC2 and is accompanied by PKA-dependent TORC2 nuclear entry and dephosphorylation.Having determined that the MIE gene is a CREB target gene that is silent during pCREB abundance, we focused on the possibility that VIP regulates a transcriptional coactivator. TORC2, a CREB coactivator, is shown in Fig. 6 to participate in VIP-induced MIE gene expression in quiescently infected NT2 cells. Depletion of TORC2 by RNAi lowered VIP's effect on MIE⁺ NT2 population density (Fig. 6A) and MIE protein production (Fig. 6B), even when the TORC2 was depleted after the infection (Fig. 6C). The result was reproduced by using different shRNAs to achieve TORC2 knockdown and was not accompanied by changes in pCREB and TORC1 levels (Fig. 6B).

• Open in new tab
• Download powerpoint

FIG. 6.

VIP's effect requires cellular TORC2. (A to C) NT2 cells were transfected with a plasmid expressing one of four different shRNAs against TORC2 (TORC2i_y, TORC2i_g, TORC2i_r, or TORC2i_b) or a negative control (NCi) for 72 h. HCMV infection was carried out after (A and B) or before (C) TORC2 knockdown, as described in Materials and Methods. (A) IFA shows the MIE⁺ NT2 cell density after VIP stimulation (100 nM) for 24 h. Original magnification, ×20. The bar graph depicts the ratio of MIE⁺ cells to all cells. Means and SD were determined from five random fields. (B and C) The expression of IE1 p72 and IE2 p86 was analyzed by Western blotting after stimulation for 24 h with VIP (100 nM) or nothing. Duplicate samples were pooled for analyses. Panel B also shows levels of TORC2, TORC1, and pCREB from the same experiment; IE1 p72, IE2 p86, TORC2, and pCREB were detected on same blot in relationship to β-tubulin. Changes in TORC2 and IE1 p72 levels resulting from TORC2i could not be determined accurately by densitometric analysis of the exposed films because band intensities were outside the linear range or undetectable. TORC2 knockdown was at least 3.3-fold (TORC2i_y and TORC2i_r) or greater (TORC2i_g and TORC2i_b). IE1 p72 knockdown was at least 25-fold for TORC2i_y, TORC2i_g, TORC2i_r, and TORC2i_b.

In hepatocytes and neuroendocrine cells (38, 72), FSK stimulation results in TORC2 nuclear entry and dephosphorylation. The possibility of VIP also acting in this manner was examined in NT2 cells containing a plasmid expressing Flag-tagged TORC2 (TORC2-F). Both confocal microscopy (Fig. 7A) and Western blot (Fig. 7B) analyses of TORC2-F distribution in cytoplasmic and nuclear compartments indicated that VIP stimulation for 15 min promotes TORC2's translocation from the cytoplasm to the nucleus. In VIP-stimulated NT2 cells, the TORC2-Flag (TORC2-F) in whole-cell extracts fractionated by SDS-PAGE existed as two immunoreactive bands differing slightly in electrophoretic mobility, whereas only the slower-migrating band was evident in unstimulated cells (Fig. 7C). Based on published results for FSK stimulation of Flag-tagged TORC2 in other systems (72), the dephosphorylated TORC2-F was expected to migrate faster by SDS-PAGE than its phosphorylated counterpart. This inference is partly backed by evidence showing that VIP induces the dephosphorylation of TORC2 Ser171 (Fig. 7D). VIP's ability to produce these TORC2 alterations at 15 min poststimulation was unaffected by HCMV infection (Fig. 7C versus D; data not shown). At 48 h p.i., the exogenously expressed TORC2-F was still retained in unstimulated NT2 cytoplasm and the MIE proteins were not made, as determined by IFA (Fig. 7E). In contrast, VIP stimulation for 24 h resulted in TORC2-F and MIE proteins colocalizing in NT2 nuclei (Fig. 7E). MIE proteins were also evident in stimulated neighboring cells expressing endogenous TORC2, which is not detected by the anti-Flag antibody.

• Open in new tab
• Download powerpoint

FIG. 7.

VIP's effect corresponds to TORC2's translocation to the nucleus and dephosphorylation. (A to D) NT2 cells transfected for 48 h with a plasmid expressing TORC2 tagged with Flag (TORC2-F) or with negative control vector (NC) were subjected to IFA plus confocal microscopy (A) or Western blotting (B to D) at 15 min of VIP (100 nM) or FSK (10 μM) stimulation or no stimulation (none). Cytoplasm (Cy) and nuclei (Nu) were fractionated prior to the analysis depicted in panel B. Whole-cell extracts were analyzed in panels C and D. HCMV infection was carried out at 24 h posttransfection for the experiment shown in panel D. TORC2-F (pTORC2 and TORC2) was detected with anti-Flag antibody (A-D), and TORC2 Ser171 phosphorylation (pTORC2 Ser171) was detected with phospho-Ser171 antibody (D). Duplicate samples of whole-cell and nuclear proteins from each experimental group were combined prior to fractionation in 7.5% (C) or 10% (B and D) polyacrylamide gels. (E) HCMV-infected NT2 cells were transfected (at 2 h p.i.) for 48 h with a plasmid expressing TORC2-F and stimulated (at 24 h p.i.) with VIP for 24 h. At 48 h p.i., Flag (red) and MIE (green) antigens were detected by IFA, and the results are shown as merged images. Nuclei were counterstained with DAPI. Original magnification, ×40.

To determine whether PKA is involved in VIP's effect on TORC2, PKA was inhibited by H89 or depleted by RNAi prior to VIP stimulation of NT2 cells expressing TORC2-F. We showed that H89 inhibits VIP-induced TORC2 Ser171 dephosphorylation (Fig. 8A) and TORC2 enrichment in nuclei (Fig. 8B). Either H89 or PKA knockdown abrogated the VIP-induced appearance of the TORC2-F band having increased electrophoretic mobility connoting dephosphorylation (Fig. 8A and C).

• Open in new tab
• Download powerpoint

FIG. 8.

PKA is involved in VIP-induced TORC2 dephosphorylation and nuclear localization. (A and B) Western blot analyses of whole cells (A) and isolated nuclei (B) from NT2 cells expressing TORC2-F (TORC2 and pTORC2) for 48 h and stimulated with VIP (100 nM) or not stimulated (none) for 45 min, with or without H89 (10 μM). (C) Western blot analysis of NT2 whole cells cotransfected for 48 h with plasmids expressing TORC2-F and NCi or PKAi, infected for 24 h with HCMV, and stimulated with VIP (100 nM) for the final 15 min. Triplicate samples from each experimental group (A to C) were combined prior to fractionation in 7.5% (A and C) and 10% (B) polyacrylamide gels.

TORC2 S171A desilences the MIE genes in the absence of VIP stimulation.The finding that VIP induces both TORC2 Ser171 dephosphorylation and TORC2 nuclear entry is consistent with the known function of TORC2 phospho-Ser171 in sequestering TORC2 in the cytoplasm. To determine whether TORC2's dephosphorylation and repositioning to nuclei contribute to MIE gene activation, a Flag-tagged TORC2 S171A mutant devoid of phosphorylation at Ser171 was expressed from a plasmid in unstimulated NT2 cells (Fig. 9A). Both IFA and Western blot methods revealed that Flag-tagged TORC2 S171A localized to nuclei in greater amounts than those of WT TORC2-F (Fig. 9A and B). This finding parallels that of TORC2 S171A's ability to render the MIE genes active, as reflected by the coappearance of TORC2-F S171A and MIE protein in NT2 nuclei (Fig. 9C) and the increases in MIE⁺ NT2 population density (Fig. 9D) and MIE protein production across all cells (Fig. 9E). Notably, approximately 25% of nuclei immunostaining for TORC2-F S171A did not stain for MIE protein (data not shown). Heterogeneity in the NT2 cell population (37, 54) is one possible reason for this result. The recent report that dephosphorylation of TORC2 Ser-275 is additionally needed to fully activate CREB-dependent gene expression via cAMP signaling in islet cells (31) may also bear on this result. Despite the system's potential imperfection, the experimental approach applied here using TORC2-F S171A has yielded concordant findings linking desilencing of the MIE genes with TORC2's hypophosphorylation and nuclear translocation.

• Open in new tab
• Download powerpoint

FIG. 9.

TORC2 S171A localizes to the nucleus and activates MIE gene expression in the absence of VIP stimulation. NT2 cells transfected for 48 h with a plasmid expressing TORC2-F (TORC2-F S171) or Flag-tagged TORC2 S171A (TORC2-F S17A) (A to E) and infected with HCMV for 24 h (D and E) or 48 h (C) were analyzed by Western blotting (A and E) or IFA (B to D). TORC2-F S171 and TORC2-F S171A were detected with anti-Flag antibody (A to C and E), and TORC2 Ser171 phosphorylation (pTORC2 Ser171) was detected with phospho-Ser171 antibody (A). For panels A and E, duplicate samples of whole-cell and nuclear proteins from each experimental group were combined prior to fractionation in 7.5% polyacrylamide gels. For panels B to D, NT2 cells were infected at 24 h posttransfection (B and D) or 2 h before transfection (C). Nuclei were counterstained with DAPI. Original magnification, ×20. Flag was detected by indirect IFA (B and C), whereas MIE proteins (IE1 p72 and IE2 p86) were detected by indirect (D) or direct (C) IFA (see Materials and Methods). Merged images of TORC2-F (S171 or S171A) (red) and nuclei (blue) are shown in panel B. A merged image of IFA for TORC2-F S171A (red) and MIE proteins (green) is shown in panel C. The bar graph in panel D depicts the ratio of MIE⁺ cells to all cells. Means and SD were determined from eight random fields.