This Article Abstract Full Text (PDF) Other Versions of this Article: JVI.01579-07v1 82/1/40    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Guan, H. Articles by Ricciardi, R. P. Search for Related Content PubMed PubMed Citation Articles by Guan, H. Articles by Ricciardi, R. P.

Tumorigenic Adenovirus Type 12 E1A Inhibits Phosphorylation of NF-B by PKAc, Causing Loss of DNA Binding and Transactivation

Department of Microbiology, School of Dental Medicine,¹ the Abramson Cancer Center, School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104²

Received 6 June 2007/ Accepted 16 October 2007

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Human adenovirus type 12 (Ad12) E1A protein (E1A-12) is the key determinant of viral tumorigenesis. E1A-12 mediates major histocompatibility complex class I (MHC-I) shutoff by inhibiting the DNA binding of the transcriptional activator NF-B (p50/p65) to the class I enhancer. This enables Ad12 tumorigenic cells to avoid class I recognition and lysis by cytotoxic T lymphocytes. In this study, we demonstrate that the phosphorylation of p50 and p65 by the catalytic subunit of protein kinase A (PKAc) is essential for NF-B DNA binding and transactivation activity. Treatment with H89 and knockdown of PKAc in cells led to the inhibition of phosphorylation at p50 Ser³³⁷ and p65 Ser²⁷⁶ and loss of DNA binding by NF-B. Importantly, NF-B phosphorylation by PKAc was repressed by tumorigenic E1A-12, but not by nontumorigenic Ad5 E1A (E1A-5). The stable introduction of E1A-12 into Ad5 nontumorigenic cells resulted in a decrease in the phosphorylation of NF-B, loss of NF-B DNA binding, and the failure of NF-B to activate a target promoter, as well as diminution of MHC-I transcription and cell surface expression. Significantly, the amount and enzymatic activity of PKAc were not altered in Ad12 tumorigenic cells relative to its amount and activity in nontumorigenic Ad5 cells. These results demonstrate that E1A-12 specifically prevents NF-B from being phosphorylated by PKAc.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Human adenovirus type 12 (Ad12) is capable of causing tumors when introduced into immunocompetent adult rodents (4, 25, 26, 38). The viral immediate-early gene product E1A-12 is the key determinant of tumorigenesis (38). Importantly, the E1A-12 oncoprotein mediates diminished expression of major histocompatibility complex class I (MHC-I) antigens on the cell surface by inhibiting the transcription of all class I genes (1, 7, 10, 29, 33). This enables Ad12 tumorigenic cells to evade cytotoxic T-lymphocyte (CTL) immunosurveillance (3, 39). In contrast, E1A proteins of nontumorigenic adenoviruses, such as Ad5, are unable to suppress MHC-I expression, leaving cells transformed by Ad5 vulnerable to recognition and elimination by host CTLs (3, 7, 29, 39).

The transcription factor NF-B, a heterodimer composed of subunits p50 and p65, activates MHC-I transcription by binding to the MHC-I enhancer region (25, 38). In most resting and unstimulated cells, NF-B is retained in the cytoplasm by IB proteins. Upon activation by stimuli such as cytokines, mitogens, and bacterial lipopolysaccharides, IB is phosphorylated by IB kinase and subsequently ubiquitinated and degraded by the 26S proteasome (2, 23, 24, 44). This leads to the phosphorylation and nuclear translocation of NF-B, which in turn binds to target genes and transactivates gene expression.

Phosphorylation plays a crucial role in NF-B DNA binding and transcriptional activation. In particular, the phosphorylation of specific serine residues, such as Ser²⁷⁶, Ser⁵²⁹, and Ser⁵³⁶ of the p65 subunit, is important for NF-B transactivation (2, 23, 24, 28, 34-37, 43, 44). It has been proposed that the phosphorylation of p65 Ser²⁷⁶ by the protein kinase A catalytic subunit (PKAc) promotes transactivation activity by releasing the C-terminal transactivation domain from an intramolecular masking by the N-terminal region (44). By contrast, the p50 subunit does not contain a transactivation domain. However, the phosphorylation of specific serine residues of p50, e.g., Ser³³⁷, is critical for DNA binding (15, 16, 18, 19). Previous work showed that p50 is differentially phosphorylated in tumorigenic Ad12- and nontumorigenic Ad5-transformed cells (18). Specifically, p50 was shown to be hypophosphorylated in Ad12 tumorigenic cells, leading to diminished NF-B DNA binding and the downregulation of MHC-I transcription (18). Conversely, p50 was found to be highly phosphorylated in Ad5 nontumorigenic cells, consistent with the pronounced binding of NF-B to the MHC-I enhancer region in these cells. Most recently, it was discovered that in Ad5 nontumorigenic cells, p50 is constitutively phosphorylated by PKAc and that NF-B DNA binding was dependent upon PKAc activity (13). Significantly, the phosphorylation of p50, as well as its DNA binding activity, was suppressed by the substitution of alanine for Ser³³⁷ and by PKAc-specific inhibitors (13).

In this study, we demonstrate that tumorigenic E1A-12 impedes the ability of PKAc to phosphorylate NF-B at Ser²⁷⁶ of p65 and Ser³³⁷ of p50. As a consequence, NF-B DNA binding and transactivation activities are decreased, MHC-I transcription is downregulated, and class I antigen expression on the cell surface is diminished.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell culture. Mouse cell lines KAd5 and F10-12, transformed by Ad5 and Ad12, respectively, were cultured in Eagle's minimal essential medium (Biowhittaker) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, and 50 µg/ml gentamicin sulfate. Hooded Lister rat cell lines DP5-2 and 12-1 transformed by Ad5 and Ad12, respectively, and COS-7 cells were grown in Dulbecco's modified Eagle's medium (Invitrogen), 10% fetal bovine serum, 2 mM L-glutamine, 100 units/ml penicillin, and 100 µg/ml streptomycin.

Plasmids. Human and mouse p50 cDNA clones pCMV-hp50 and pGEX-mp50 were described previously (16). pCMV-hp65 was cloned by inserting human p65 full-length cDNA into the HindIII-XbaI sites of pRc/CMV (Invitrogen). pET-6xHis-p65 expressing N-terminal six-His-tagged p65 was created by subcloning the human p65 full-length cDNA into pET100/D-TOPO vector (Invitrogen). The pSG424-p65 plasmid expressing the Gal4-p65 fusion protein was constructed by subcloning the human p65 full-length cDNA into the BamHI-XbaI sites of pSG424 vector, which encodes a Gal4 protein (residues 1 to 147) (27). pSG424-p65S276A was constructed similarly, first by generating a substitution of alanine for serine at position 276 of p65, followed by subcloning into pSG424. Plasmids pCMV-Ad5E1A and pCMV-Ad12E1A were constructed by subcloning E1A-5 and E1A-12 13S cDNAs, respectively, into the HindIII-XbaI sites of pRc/CMV. To generate GST fusion proteins, the E1A-5 and E1A-12 cDNAs were subcloned into pGEX4T-1 (Amersham Pharmacia Biotech). To construct N-terminally FLAG-tagged E1A-5 (p2B-Ad5E1A) and E1A-12 (p2B-Ad12E1A), the cDNAs were subcloned into the BamHI and ApaI sites of pCMV-2B (Stratagene). The mouse PKAc cDNA clone pCMV-mPKAc was described previously (13). NF-B-driven firefly luciferase reporter p4R1-luc was constructed by cloning four tandem repeats of the NF-B binding site of the MHC-I enhancer R1 sequences (5'-GGGGAATCCCCT-3'), which were placed upstream of the basic promoter sequence of mouse MHC-I (H-L^d; nucleotides –122 to –1), into pGL3 vector (Promega). The Gal4-Luc reporter plasmid and pSG424 were kindly provided by T. D. Gilmore (Boston University, Boston, MA).

Treatment of cells with kinase inhibitors. Cells were treated with kinase inhibitor(s) for 5 h in culture medium containing 5% fetal bovine serum. H89 was purchased from Alexis Biochemicals. U0-126, PD98059, and SB203580 were purchased from Sigma.

Cell transfection and RNA interference. Transfection of COS-7 was performed by using either GeneJammer transfection reagent (Stratagene) or FuGENE 6 (Roche) according to the manufacturers' instructions. For the coexpression of p50 and p65 with FLAG-tagged E1A-5 or E1A-12 in COS-7 cells, 60 ng of pCMV-hp50, 300 ng of pCMV-hp65, and increasing amounts (200 ng and 500 ng) of p2B-Ad5E1A or p2B-Ad12E1A were used for transfection on six-well plates. Transfection of KAd5 and DP5-2 cells was performed using Lipofectamine 2000 transfection regent (Invitrogen) based on the manufacturer's protocol. For RNA interference analysis, double-stranded DNA (dsDNA) molecules containing a T7 promoter at both ends were first amplified by PCR from pRc/CMV vector (nonspecific), cDNAs of mouse mitogen-and-stress-activated protein kinase-1 (MSK1), and PKAc. The dsDNA templates were then used for in vitro transcription to generate dsRNA by T7 RNA polymerase (Promega). The dsRNA was digested with Dicer enzyme (Stratagene) to produce small interfering RNA (siRNA) according to the manufacturer's protocol. Purified siRNA (125 pmol) was then used for transfection with Oligofectamine (Invitrogen) on six-well plates according to the manufacturer's instructions. Cells were harvested at 48 h posttransfection, and target protein knockdown was confirmed by Western blot analysis.

Generation of stable cell lines. To establish stable Ad5 cell lines expressing E1A-12, KAd5 and DP5-2 cells were transfected with plasmid pCMV-Ad12E1A. Single-cell colonies resistant to Geneticin (Invitrogen) were selected. The expression of E1A-12 was confirmed by Western blot analysis. As a control, KAd5 and DP5-2 cells were also transfected with empty plasmid pRc/CMV.

RNA preparation and Northern blot analysis. Total RNA was prepared from cells by using an RNeasy kit (QIAGEN) according to the manufacturer's instructions. Twenty micrograms of total RNA from each sample was used for Northern blot analysis as previously described (14).

EMSA. The electrophoretic mobility shift assay (EMSA) was conducted by incubating nuclear extracts or whole-cell extracts with ³²P-labeled ds MHC-I R1 oligonucleotides, which contain an NF-B recognition site, as previously described (16). Nuclear and whole-cell extracts were prepared as described previously (16, 18).

In vitro kinase assay. PKAc purified from bovine heart (Sigma) was used for an in vitro kinase assay as described previously, with modifications (16). Briefly, six-His-p65 or GST-p50 proteins (0.4 µg) were first mixed with 1 or 2 µg of GST, GST-E1A-5, or GST-E1A-12 proteins and kept at room temperature for 1 h. The mixtures were then incubated with 2 U of PKAc at 30°C for 45 min in kinase buffer (25 mM HEPES, pH 7.5, 10 mM MgCl₂, 10 mM NaF, 1 mM Na₃VO₄, 1 mM CaCl₂, 5 mM dithiothreitol, 20 µM ATP). Phosphorylated p50 and p65 were analyzed by Western blotting using specific phosphorecognition antibodies as described below.

Immunoprecipitation and Western blotting analyses. Immunoprecipitation and Western blotting analyses were carried out as described previously (13, 18). Rabbit anti-E1A-5 (13 S-5) and goat anti-p50 (C-19) and p65 [(A)-G] were purchased from Santa Cruz Biotechnology. Rabbit anti-E1A-12 was previously described (30). Rabbit anti-PKAc antibody NT was purchased from Upstate Biotechnology. Mouse monoclonal anti-CREB (86B10) antibody, rabbit antiphospho-p65 (Ser²⁷⁶) antibody (no. 3037), and rabbit monoclonal antibodies against phospho-PKA substrate (RRXS/T and 100G7) and phospho-CREB (Ser¹³³) (87G3) were obtained from Cell Signaling Technology, Inc. Rabbit anti-p50 (no. 1613) and p65 (no. 1226) were described previously (18). Monoclonal FLAG (M2) and β-actin (AC-15) antibodies were acquired from Sigma.

Luciferase assay. To measure NF-B-driven firefly luciferase activity, subconfluent cells grown on 24-well plates were transfected with 100 ng of p4R1-luc plasmids together with 10 ng of the Renilla luciferase reporter pRL-TK (Promega), which was used as an internal control of transfection efficiency. At 24 h posttransfection, the cells were lysed, and firefly and Renilla luciferase activities were measured with a Promega dual-luciferase assay kit on a Genios Pro plate reader (Tecan). Each transfection was performed in duplicate, and each experiment was repeated three times. To test p65 transactivation activity, subconfluent cells grown on six-well plates were transfected with 200 ng of Gal4-Luc reporter; 350 ng of either pCMV-FLAG-E1A5 or pCMV-FLAG-E1A12; and 250 ng of either empty vector pSG424, pSG424-p65, or pSG424-p65S276A, along with 30 ng of pRL-TK. Luciferase activities were measured at 40 h posttransfection as described above.

FACS analysis. Cells were incubated with fluorescein isothiocyanate-conjugated monoclonal anti-mouse MHC-I antibody (34-1-2S; eBioscience) in phosphate buffered saline containing 1% bovine serum albumin (PBSB) for 45 min on ice. After being washed with PBSB three times, the cells were fixed in 0.5 ml of 4% paraformaldehyde in PBS and then subjected to fluorescence-activated cell sorting (FACS) analysis.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PKAc fails to phosphorylate NF-B in Ad12 tumorigenic cells. The phosphorylation of NF-B p65 Ser²⁷⁶ by PKAc is essential for the transactivation activity of the nuclear factor (42-44). We also previously reported that the phosphorylation of Ser³³⁷ of the NF-B p50 subunit by PKAc is important for constitutive DNA binding in Ad5 nontumorigenic cells (13, 16). In contrast to the hyperphosphorylation of p50 and strong DNA binding of NF-B in Ad5 nontumorigenic cells, Ad12 tumorigenic cells exhibit a dramatic reduction in p50 phosphorylation and NF-B DNA binding (18). This led us to speculate that PKAc fails to phosphorylate NF-B in Ad12 tumorigenic cells. To test this hypothesis, we examined the phosphorylation levels of PKAc sites on p50 and p65 in two haplotypically matched Ad12 (12-1) and Ad5 (DP5-2) cell lines. As shown in Fig. 1A, Western blot analysis using an antibody recognizing phospho-p65 Ser²⁷⁶ revealed that the phosphorylation of p65 at this PKAc site was dramatically reduced in tumorigenic 12-1 cells in comparison with its phosphorylation in nontumorigenic DP5-2 cells. Consistent with this finding, the phosphorylation of p50 at its PKAc site (Ser³³⁷) was also greatly inhibited in 12-1 cells in comparison with its phosphorylation in DP5-2 cells (Fig. 1B). This was confirmed by Western blot analyses using the phospho-p65 Ser²⁷⁶ antibody (left panel), which was demonstrated to crossreact with phospho-p50 Ser³³⁷, and an antibody against phospho-PKAc substrate (RRXS/T) following the immunoprecipitation of p50 from the two cell lines (right panel). To investigate if the phosphorylation of proteins other than NF-B by PKAc could also be inhibited in Ad12 tumorigenic cells, we examined the phosphorylation status of the PKAc site (Ser¹³³) on the cAMP response element binding protein (CREB) in 12-1 and DP5-2 cells. As shown in Fig. 1C, comparable phosphorylation levels of CREB Ser¹³³ were observed in the two cell lines (lanes 1 and 2). Since CREB Ser¹³³ can also be phosphorylated by several kinases, including mitogen-activated protein kinase, extracellular signal-regulated kinase, MSK, and ribosomal S6 kinase, in addition to PKAc (17), we tested if PKAc, which fails to phosphorylate NF-B in 12-1 cells (Fig. 1A and B), is truly responsible for phosphorylating CREB Ser¹³³. To this end, 12-1 cells were treated with inhibitors of the kinases mentioned above. As shown in Fig. 1C, the phosphorylation of CREB Ser¹³³ was greatly inhibited by PKAc inhibitor H89 (Fig. 1C, lane 6), but was not affected by mock treatment (lane 4) or the chemicals SB203580, PD98059, and U0-126 either jointly (lane 5) or separately (data not shown). A combination of SB203580, PD98059, and U0-126 inhibits the kinases of mitogen-activated protein kinase, extracellular signal-regulated kinase, MSK, and ribosomal S6 kinase altogether. Similar results were obtained when DP5-2 cells were treated with these inhibitors (data not shown). These data indicate that PKAc is functionally active in Ad12 tumorigenic cells and is the kinase responsible for phosphorylating CREB Ser¹³³ but that phosphorylation of NF-B by PKAc is specifically inhibited in Ad12 cells.

View larger version (30K): [in this window] [in a new window]   FIG. 1. Phosphorylation of NF-B by PKAc is specifically inhibited in Ad12 tumorigenic cells. (A) The PKAc site (Ser276) on p65 is not phosphorylated in Ad12 tumorigenic cells. The same amounts of protein extract from Ad5 (DP5-2) and Ad12 (12-1) cells were analyzed by Western blotting using an antibody recognizing phospho-p65 Ser276. The same blot was stripped and reprobed with an antibody recognizing total p65. (B) The PKAc site (Ser337) on p50 is not phosphorylated in Ad12 tumorigenic cells. Left, levels of phospho-p50 Ser337 in DP5-2 and 12-1 cells as revealed by Western blotting using the phospho-p65 Ser276 antibody, which was demonstrated to cross-react with phospho-p50 Ser337. The same blot was reprobed with an antibody recognizing total p50. Right, levels of phospho-p50 Ser337 in DP5-2 and 12-1 cells as determined by Western blotting using an antibody against phospho-PKAc substrate (RRXS/T) following immunoprecipitation of p50 from the two cell lines. As a loading control, the same blot was reprobed with p50 antibody. IgG-h, goat immunoglobulin G heavy chain. (C) The PKAc site (Ser133) of CREB is phosphorylated to a similar extent in Ad5 and Ad12 cells. Whole-cell extracts of DP5-2 and 12-1 cells which were left untreated or treated with PKAc inhibitor H89 (10 µM) or other kinase inhibitors (SB+PD+U; 10 µM each) were analyzed by Western blotting using an antibody against phospho-CREB Ser133. The same blots were then stripped and reprobed with an antibody recognizing total CREB. A circled P indicates phosphorylation. DMSO, dimethyl sulfoxide; SB, SB203580; PD, PD98059; U, U0-126.

View larger version (35K): [in this window] [in a new window]   FIG. 2. PKAc protein levels in nontumorigenic Ad5 and tumorigenic Ad12 cells are similar. Whole-cell extracts of cells transformed with Ad5 (DP5-2 and KAd5) and Ad12 (12-1 and F10-12), of which DP5-2 versus 12-1 and KAd5 versus F10-12 are haplotypically matched, were immunoblotted using PKAc antibody. As a quantitative control, the same blot was reprobed with β-actin antibody.

Phosphorylation by PKAc is critical for NF-B DNA binding.

View larger version (29K): [in this window] [in a new window]   FIG. 3. PKAc activity is essential for NF-B phosphorylation and DNA binding. (A) Inhibition of NF-B phosphorylation by H89. DP5-2 and KAd5 cells were treated with dimethyl sulfoxide (DMSO; mock) or PKAc inhibitor H89 (10 µM) for 5 h. Whole-cell extracts were then prepared and probed with an antibody specifically recognizing phosphorylated p65 Ser276 and p50 Ser337. As a loading control, the same blots were reprobed with antibodies against total p65 and p50. A circled P indicates phosphorylation. (B) Repression of NF-B DNA binding by PKAc knockdown. Nuclear extracts of KAd5 cells treated with nonspecific, MSK1, or PKAc siRNA were subjected to EMSA using a 32P-labeled dsDNA probe that contains the NF-B recognition site from the MHC-I enhancer. The indicated positions of p50/p65 heterodimers and p50/p50 homodimers are based on previous supershift analyses using p50 and p65 antibodies (13, 15, 18, 20). The protein levels of PKAc, p65, and p50 in the cells as determined by Western blot analysis are shown under the EMSA.

E1A-12 inhibits PKAc from phosphorylating NF-B. We next asked why PKAc, which is equally expressed in Ad5 and Ad12 cells (Fig. 2), is unable to phosphorylate NF-B in Ad12 tumorigenic cells. We hypothesized that an inhibitor resides in Ad12 tumorigenic cells and specifically prevents PKAc from phosphorylating NF-B; such an inhibitor is absent in Ad5 nontumorigenic cells. Since Ad12's tumorigenicity is solely determined by the viral oncoprotein E1A-12, we examined if E1A-12 per se interferes with the ability of PKAc to phosphorylate NF-B. To this end, in vitro kinase assays were conducted by incubating six-His-p65 or GST-p50 with purified PKAc and ATP in the presence or absence of GST, GST-E1A-5, or GST-E1A-12. Western blot analyses were then performed using specific phosphoantibodies to probe phosphorylated NF-B proteins. As shown in Fig. 4A, p65 Ser²⁷⁶ was phosphorylated by PKAc to a similar extent regardless of the presence of GST or GST-E1A-5 (lanes 1 to 4). By contrast, the phosphorylation of p65 Ser²⁷⁶ by PKAc was dramatically suppressed by GST-E1A-12 (Fig. 4A, lanes 5 to 6). Similar results were obtained when GST-p50 was used as a substrate for the in vitro kinase assay (Fig. 4B). These data strongly suggest that E1A-12, but not E1A-5, inhibits the ability of PKAc to phosphorylate NF-B.

View larger version (27K): [in this window] [in a new window]   FIG. 4. E1A-12 prevents PKAc from phosphorylating NF-B in vitro. Purified PKAc (2 units) was incubated with 0.4 µg of either six-His-p65 (A) or GST-p50 (B) in the presence of indicated amounts of GST, GST-E1A-5 (13S), or GST-E1A-12 (13S) in an in vitro kinase assay. After the reaction was performed, phosphorylated p65 and p50 were analyzed by Western blotting using antibodies against phospho-p65 Ser276 and phospho-PKAc substrate (RRXS/T), respectively. As a gel-loading control, the same blots were reprobed with regular anti-p65 and anti-p50 antibodies. A GST antibody was used to detect GST, GST-E1A-5, and GST-E1A-12 that were added in the assays. A circled P indicates phosphorylation.

Inhibition of NF-B phosphorylation by E1A-12 leads to diminished DNA binding.

View larger version (44K): [in this window] [in a new window]   FIG. 5. E1A-12 suppresses cellular NF-B DNA binding activity. (A) EMSA of NF-B DNA binding in the presence of E1A-12 and E1A-5. p50 and p65 plasmids (60 ng and 300 ng, respectively) were cotransfected with increasing amounts (200 ng and 500 ng) of plasmids expressing FLAG-tagged 13S products of E1A-12 or E1A-5 in COS-7 cells. After 24 h, total cell protein extracts were used for EMSA analysis with a 32P-labeled dsDNA probe that contains the MHC-I NF-B recognition site. Positions of p50/p65 heterodimers and p50/p50 homodimers are indicated. +, present; –, absent. (B) Western blot of the same extracts used in the EMSA shown in panel A probed with antibodies against p50, p65, phospho-p65 Ser276, and FLAG tag (for detection of E1A-12 and E1A-5). To probe phospho-p50 Ser337, p50 was first immunoprecipitated from the extracts and then analyzed by Western blotting using an antibody against phospho-PKAc substrate (RRXS/T). A circled P indicates phosphorylation.

View larger version (43K): [in this window] [in a new window]   FIG. 6. E1A-12 suppresses cellular p50 DNA binding activity. (A) EMSA of p50 DNA binding in the presence of E1A-12 and E1A-5. p50 plasmids (60 ng) were cotransfected with increasing amounts (200 ng and 500 ng) of plasmids expressing FLAG-tagged 13S products of E1A-12 or E1A-5 in COS-7 cells. After 24 h, total cell protein extracts were used for EMSA analysis using the 32P-labeled R1 probe. Positions of p50/p50 homodimers are indicated. +, present; –, absent. (B) Western blot of the same extracts used for the EMSA shown in panel A probed with anti-p50 antibody, as well as anti-FLAG antibody to detect E1A-12 and E1A-5.

View larger version (28K): [in this window] [in a new window]   FIG. 7. E1A-12 inhibits cellular p65 transactivation activity. COS-7 cells were cotransfected with Gal4-p65 expression plasmids, which consist of a full-length cDNA of p65 fused to a Gal4 DNA binding domain and a luciferase reporter placed downstream of a Gal4 binding site (Gal4-Luc reporter), in the absence (–) or presence (+) of the 13S products of E1A-12 or E1A-5. Similarly, a p65 mutant containing an alteration of Ser276 to alanine (Gal4-p65S276A) was also analyzed for its ability to transactivate the luciferase reporter with or without the E1A-12 or E1A-5. Plasmids expressing Gal4 only were used as a negative control for the luciferase assay. A Renilla luciferase reporter was included for each assay as an internal control for transfection. At 40 h posttransfection, the cells were lysed and luciferase activities were measured. Data represent the averages of the results of three independent experiments. Error bars show standard deviations. Comparable expression of Gal4-p65 and Gal4-p65S276A was confirmed by Western blotting using an antibody against p65. The same blot was reprobed with β-actin antibody.

Expression of E1A-12 in Ad5 nontumorigenic cells prevents NF-B DNA binding and target gene transcription.

View larger version (41K): [in this window] [in a new window]   FIG. 8. Stable expression of E1A-12 in Ad5 nontumorigenic cells inhibits NF-B DNA binding activity and NF-B-dependent target gene expression. (A) EMSA of NF-B DNA binding obtained from extracts of stable cell lines coexpressing the 13S products of E1A-12 and E1A-5. Stable cell lines expressing both E1A-12 and E1A-5 (DP5-2/E1A-12 and KAd5/E1A-12) were established by transfecting DP5-2 and KAd5 cells with plasmids harboring the 13S cDNA of E1A-12. Nuclear extracts prepared from independent, stable DP5-2/E1A-12 and KAd5/E1A-12 cell lines, as well as from normal and vector-transfected cells (DP5-2/vector and KAd5/vector), were used for EMSA analysis with the 32P-labeled dsDNA probe R1. Nuclear extract of 12-1 cells was used as a negative control for NF-B binding. (B) Western blots of the same extracts used for the EMSA shown in panel A probed with antibodies recognizing phospho-p65 Ser276, phospho-p50 Ser337, p65, p50, E1A-12, and β-actin. A circled P indicates phosphorylation. (C) Northern blot analysis of MHC-I. Total RNA isolated from each cell line described in the legend for panel A was used for Northern blot hybridization with a 32P-labeled MHC-I probe. As a quantitative control, the same blot was rehybridized with a 32P-labeled RNase P probe. (D) FACS analysis of MHC-I surface antigens on the KAd5/E1A-12 cell lines derived from KAd5. MHC-I cell surface antigen levels represent the mean of the results of two independent FACS analyses. (E) NF-B-driven luciferase activity assay. The KAd5/E1A-12 cell lines derived from KAd5 were transfected with a NF-B-driven firefly luciferase reporter together with a Renilla luciferase reporter (internal control). At 24 h posttransfection, the cells were lysed and luciferase activities were measured. Data represent the averages of the results of three independent experiments. Error bars show standard deviations.

We previously reported that, in addition to the inhibition of NF-B activity, a second mechanism involving enhanced binding of the transcriptional repressor COUP-TFII to the class I enhancer region also contributes to MHC-I shutoff in Ad12 tumorigenic cells (31, 32, 40, 41). To exclude the possible repression effect of COUP-TFII on MHC-I expression, we transfected a luciferase reporter that is driven only by NF-B into KAd5 cells, KAd5/vector control cells, and KAd5/E1A-12 cells. As shown in Fig. 8E, the luciferase activity in both of the KAd5/E1A-12 cell lines (bars 3 and 4) was greatly reduced in comparison with the activity in KAd5 cells (bar 1) or KAd5/vector control cells (bar 2). Significantly, the loss of luciferase reporter activity in the KAd5/E1A-12 cell lines correlates with the loss of NF-B phosphorylation and DNA binding (Fig. 8A and B, lanes 9 and 10). These results unquestionably demonstrate that E1A-12 inhibits NF-B phosphorylation, DNA binding, and transactivation to specifically prevent the expression of MHC-I.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The diminished surface expression of MHC-I antigens on Ad12 tumorigenic cells enables them to escape CTL recognition and lysis (1, 3, 7, 10, 29, 33, 39). Downregulation of class I expression results from loss of the binding of NF-B activator to the R1 site of the class I transcriptional enhancer (20), as well as from the gain of COUP-TFII repressor binding to the R2 site of the enhancer (21, 26, 31, 32). The loss of NF-B DNA binding activity in Ad12 tumorigenic cells was thought to involve the hypophosphorylation of the p50 subunit (18). This seminal finding in Ad12 tumorigenic cells led to the broader discovery that the binding of p50 homodimers to DNA requires the phosphorylation of Ser³³⁷ (16) and that PKAc serves as the kinase of Ser³³⁷ (13).

We previously showed that PKAc-specific inhibitors, including H89, were able to block NF-B DNA binding activity and that treatment of E. coli-expressed p50 with PKAc in vitro enhances DNA binding by the NF-B subunit (13, 16). In the present study, we establish that the ability of NF-B to bind DNA in cells is truly dependent on PKAc. This was demonstrated by the finding that the PKAc inhibitor H89 prevents phosphorylation of p50 Ser³³⁷ and p65 Ser²⁷⁶ in cells. In addition, PKAc knockdown by siRNA resulted in a loss of the DNA binding activities of NF-B p50/p65 heterodimers, as well as p50/p50 homodimers, that otherwise exhibit strong binding to DNA. We also demonstrate both in vitro and in vivo that E1A-12 inhibits p50 and p65 from becoming phosphorylated by PKAc. Consequently, NF-B fails to bind the class I transcriptional enhancer and activate class I antigen expression. This E1A-12 function provides an effective immune escape strategy. By contrast, E1A-5 does not employ this escape mechanism, making Ad5 nontumorigenic cells vulnerable to CTL lysis.

The fact that E1A-12 is able to inhibit phosphorylation of both p50 Ser³³⁷ and p65 Ser²⁷⁶ suggests that DNA binding and transactivation activities of NF-B are closely regulated. Our data clearly demonstrate that inhibition of p65 Ser²⁷⁶ phosphorylation by E1A-12 represses NF-B transactivation activity. However, it remains to be established whether E1A-12-mediated reduction in NF-B's DNA binding involves p50 Ser³³⁷ alone or also includes the PKAc recognition site Ser²⁷⁶ of p65. Based on the crystal structures of DNA-bound NF-B (6, 8, 12, 22), both p50 Ser³³⁷ and p65 Ser²⁷⁶ are proximal to the hinge regions that link the DNA binding motifs and the dimerization domains and neither is in direct contact with DNA. Furthermore, it has been found that dimerization of p50 and p65 is not affected by the mutation of p50 Ser³³⁷ to alanine (16). These data suggest that p50 Ser³³⁷ and p65 Ser²⁷⁶ are likely involved in regulating the DNA binding and transactivation activities of NF-B, rather than directly participating in DNA binding and dimerization. This notion is manifested by the finding that phosphorylation of p65 Ser²⁷⁶ by PKAc induces a conformational change that releases an intramolecular masking of the C-terminal transactivation domain by the N-terminal region, thus leading to enhanced binding by the coactivator CBP/p300 (42, 44). In view of the high similarity in sequence and conformation surrounding p50 Ser³³⁷ and p65 Ser²⁷⁶, the phosphorylation of p65 Ser²⁷⁶ may also contribute to NF-B DNA binding. Our previous report revealed that NF-B, when reconstituted in vitro with one subunit isolated from Ad5 cells and the other from Ad12 cells, has a higher DNA binding activity than when both subunits are derived from Ad12 cells (18). This suggests that the phosphorylated p50 and p65 subunits from Ad5 cells both contribute to DNA binding. In addition, our recent preliminary results obtained from the coexpression of p50 and p65 in COS 7 cells indicated that the alteration of p50 Ser³³⁷ to alanine alone is not sufficient to suppress p50/p65 heterodimer DNA binding (unpublished data), though this mutation greatly inhibits the DNA binding of p50/p50 homodimers (16). These data suggest that DNA binding by p50/p65 heterodimers is regulated in a manner that is more complicated than the regulation of p50/p50 homodimers. Unlike p65, p50 does not contain a transactivation domain, and its main function is to bind DNA. It is not known how the phosphorylation of p50 Ser³³⁷ is able to promote DNA binding. It is possible that the phosphorylation of p50 Ser³³⁷ might also trigger a conformational change that allows the subunit to bind DNA more strongly. While our data indicate that the phosphorylation of NF-B at the PKAc consensus sites is critical for DNA binding, the possibility that other potential phosphoresidues of NF-B might also contribute to DNA binding cannot be ruled out.

We previously reported that MHC-I transcription in somatic cell hybrids of Ad5/Ad12-transformed cells was reduced to the same degree as in the original Ad12-transformed cells (11). Consistent with this data, we demonstrate in the present study that the E1A-12 protein is dominant to E1A-5 by virtue of its ability to inhibit phosphorylation of NF-B and subsequent DNA binding and transactivation activities. The mechanism underlying this process has yet to be elucidated. It is tempting to speculate that inhibition of the phosphorylation of p50 and p65 in the heterodimer by E1A-12 could be coregulated. NF-B, PKAc, and IB have been found to form an inactive trimolecular complex in the cytoplasm, and only following IB degradation can PKAc then phosphorylate NF-B (43). Intriguingly, IB degradation occurs constitutively in both Ad12 and Ad5 cells and there is no major difference in the levels of nucleus-translocated p50 and p65 in the two cell lines (20). This indicates that E1A-12-mediated inhibition of NF-B phosphorylation is not due to the repression of IB degradation or nuclear translocation of the transcriptional activator. Rather, our in vitro kinase assays revealed that E1A-12 was able to inhibit PKAc from phosphorylating NF-B. Moreover, our preliminary data indicate that E1A-12, but not E1A-5, can interact with both NF-B and PKAc (data not shown). This implies that E1A-12 could substitute for the constitutively degraded IB to form a trimolecular complex with NF-B and PKAc, thus specifically inhibiting PKAc from phosphorylating NF-B.

While E1A-12 functions to inhibit PKAc from phosphorylating NF-B both in vitro and in vivo, further study is needed to delineate what regions of E1A-12 are essential for this inhibitory effect. Possibly involved are the nonconserved regions of E1A-12 spanning from the N terminus to conserved region CR2, which are thought to be responsible for MHC-I repression (26). Interestingly, the N-terminal region of E1A-12 has been found to physically interact with the regulatory subunit RII of PKA (9). Furthermore, it has been reported that the 52R polypeptide, the smallest splicing product of E1A-12 that does not have any conserved regions, prevents the phosphorylation of c-JUN through direct interaction with the transcription factor (5). This suggests that tumorigenic E1A-12 is able to interfere with the phosphorylation of different transcription factors.

	ACKNOWLEDGMENTS

 
This work was supported by grant CA29797 from the National Cancer Institute (to R.P.R.).

	FOOTNOTES

 
* Corresponding author. Mailing address: University of Pennsylvania, Levy Research Building, Room 221, 4010 Locust Street, Philadelphia, PA 19104. Phone: (215) 898-3905. Fax: (215) 898-8385. E-mail: ricciardi{at}biochem.dental.upenn.edu

Published ahead of print on 24 October 2007.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Ackrill, A. M., and G. E. Blair. 1988. Regulation of major histocompatibility class I gene expression at the level of transcription in highly oncogenic adenovirus transformed rat cells. Oncogene 3:483-487.[Medline]
2. Baldwin, A. S., Jr. 1996. The NF-kappa B and I kappa B proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649-683.[CrossRef][Medline]
3. Bernards, R., P. I. Schrier, A. Houweling, J. L. Bos, A. J. van der Eb, M. Zijlstra, and C. J. Melief. 1983. Tumorigenicity of cells transformed by adenovirus type 12 by evasion of T-cell immunity. Nature 305:776-779.[CrossRef][Medline]
4. Blair, G. E., and M. E. Blair-Zajdel. 2004. Evasion of the immune system by adenoviruses. Curr. Top. Microbiol. Immunol. 273:3-28.[Medline]
5. Brockmann, D., C. Bury, G. Kroner, H. C. Kirch, and H. Esche. 1995. Repression of the c-Jun trans-activation function by the adenovirus type 12 E1A 52R protein correlates with the inhibition of phosphorylation of the c-Jun activation domain. J. Biol. Chem. 270:10754-10763.[Abstract/Free Full Text]
6. Chen, F. E., D. B. Huang, Y. Q. Chen, and G. Ghosh. 1998. Crystal structure of p50/p65 heterodimer of transcription factor NF-kappaB bound to DNA. Nature 391:410-413.[CrossRef][Medline]
7. Eager, K. B., J. Williams, D. Breiding, S. Pan, B. Knowles, E. Appella, and R. P. Ricciardi. 1985. Expression of histocompatibility antigens H-2K, -D, and -L is reduced in adenovirus-12-transformed mouse cells and is restored by interferon gamma. Proc. Natl. Acad. Sci. USA 82:5525-5529.[Abstract/Free Full Text]
8. Escalante, C. R., L. Shen, D. Thanos, and A. K. Aggarwal. 2002. Structure of NF-kappaB p50/p65 heterodimer bound to the PRDII DNA element from the interferon-beta promoter. Structure 10:383-391.[Medline]
9. Fax, P., C. R. Carlson, P. Collas, K. Tasken, H. Esche, and D. Brockmann. 2001. Binding of PKA-RIIalpha to the adenovirus E1A12S oncoprotein correlates with its nuclear translocation and an increase in PKA-dependent promoter activity. Virology 285:30-41.[CrossRef][Medline]
10. Friedman, D. J., and R. P. Ricciardi. 1988. Adenovirus type 12 E1A gene represses accumulation of MHC class I mRNAs at the level of transcription. Virology 165:303-305.[CrossRef][Medline]
11. Ge, R., X. Liu, and R. P. Ricciardi. 1994. E1A oncogene of adenovirus-12 mediates trans-repression of MHC class I transcription in Ad5/Ad12 somatic hybrid transformed cells. Virology 203:389-392.[CrossRef][Medline]
12. Ghosh, G., G. van Duyne, S. Ghosh, and P. B. Sigler. 1995. Structure of NF-kappa B p50 homodimer bound to a kappa B site. Nature 373:303-310.[CrossRef][Medline]
13. Guan, H., S. Hou, and R. P. Ricciardi. 2005. DNA binding of repressor nuclear factor-kappaB p50/p50 depends on phosphorylation of Ser337 by the protein kinase A catalytic subunit. J. Biol. Chem. 280:9957-9962.[Abstract/Free Full Text]
14. Guan, H., D. A. Smirnov, and R. P. Ricciardi. 2003. Identification of genes associated with adenovirus 12 tumorigenesis by microarray. Virology 309:114-124.[CrossRef][Medline]
15. Hou, S., H. Guan, and R. P. Ricciardi. 2002. In adenovirus type 12 tumorigenic cells, major histocompatibility complex class I transcription shutoff is overcome by induction of NF-B and relief of COUP-TFII repression. J. Virol. 76:3212-3220.[Abstract/Free Full Text]
16. Hou, S., H. Guan, and R. P. Ricciardi. 2003. Phosphorylation of serine 337 of NF-kappaB p50 is critical for DNA binding. J. Biol. Chem. 278:45994-45998.[Abstract/Free Full Text]
17. Johannessen, M., M. P. Delghandi, and U. Moens. 2004. What turns CREB on? Cell. Signal. 16:1211-1217.[CrossRef][Medline]
18. Kushner, D. B., and R. P. Ricciardi. 1999. Reduced phosphorylation of p50 is responsible for diminished NF-B binding to the major histocompatibility complex class I enhancer in adenovirus type 12-transformed cells. Mol. Cell. Biol. 19:2169-2179.[Abstract/Free Full Text]
19. Li, C. C., R. M. Dai, E. Chen, and D. L. Longo. 1994. Phosphorylation of NF-KB1-p50 is involved in NF-kappa B activation and stable DNA binding. J. Biol. Chem. 269:30089-30092.[Abstract/Free Full Text]
20. Liu, X., R. Ge, and R. P. Ricciardi. 1996. Evidence for the involvement of a nuclear NF-B inhibitor in global down-regulation of the major histocompatibility complex class I enhancer in adenovirus type 12-transformed cells. Mol. Cell. Biol. 16:398-404.[Abstract]
21. Liu, X., R. Ge, S. Westmoreland, A. J. Cooney, S. Y. Tsai, M. J. Tsai, and R. P. Ricciardi. 1994. Negative regulation by the R2 element of the MHC class I enhancer in adenovirus-12 transformed cells correlates with high levels of COUP-TF binding. Oncogene 9:2183-2190.[Medline]
22. Muller, C. W., F. A. Rey, M. Sodeoka, G. L. Verdine, and S. C. Harrison. 1995. Structure of the NF-kappa B p50 homodimer bound to DNA. Nature 373:311-317.[CrossRef][Medline]
23. Pahl, H. L. 1999. Activators and target genes of Rel/NF-kappaB transcription factors. Oncogene 18:6853-6866.[CrossRef][Medline]
24. Palombella, V. J., O. J. Rando, A. L. Goldberg, and T. Maniatis. 1994. The ubiquitin-proteasome pathway is required for processing the NF-kappa B1 precursor protein and the activation of NF-kappa B. Cell 78:773-785.[CrossRef][Medline]
25. Ricciardi, R. P. 1999. Adenovirus transformation and tumorigenicity, p. 217-227. In P. Seth (ed.), Adenoviruses: from basic research to gene therapy applications. RG Landes Co., Austin, TX.
26. Ricciardi, R. P., B. Zhao, and H. Guan. 2006. Mechanism of tumorigenesis mediated by adenovirus-12 E1A. In M. Tognon (ed.), Viral oncogenesis. Research Signpost, Kerala, India.
27. Sadowski, I., and M. Ptashne. 1989. A vector for expressing GAL4(1-147) fusions in mammalian cells. Nucleic Acids Res. 17:7539.[Free Full Text]
28. Sakurai, H., H. Chiba, H. Miyoshi, T. Sugita, and W. Toriumi. 1999. IkappaB kinases phosphorylate NF-kappaB p65 subunit on serine 536 in the transactivation domain. J. Biol. Chem. 274:30353-30356.[Abstract/Free Full Text]
29. Schrier, P. I., R. Bernards, R. T. Vaessen, A. Houweling, and A. J. van der Eb. 1983. Expression of class I major histocompatibility antigens switched off by highly oncogenic adenovirus 12 in transformed rat cells. Nature 305:771-775.[CrossRef][Medline]
30. Scott, M. O., D. Kimelman, D. Norris, and R. P. Ricciardi. 1984. Production of a monospecific antiserum against the early region 1A proteins of adenovirus 12 and adenovirus 5 by an adenovirus 12 early region 1A-β-galactosidase fusion protein antigen expressed in bacteria. J. Virol. 50:895-903.[Abstract/Free Full Text]
31. Smirnov, D. A., S. Hou, X. Liu, E. Claudio, U. K. Siebenlist, and R. P. Ricciardi. 2001. Coup-TFII is up-regulated in adenovirus type 12 tumorigenic cells and is a repressor of MHC class I transcription. Virology 284:13-19.[CrossRef][Medline]
32. Smirnov, D. A., S. Hou, and R. P. Ricciardi. 2000. Association of histone deacetylase with COUP-TF in tumorigenic Ad12-transformed cells and its potential role in shut-off of MHC class I transcription. Virology 268:319-328.[CrossRef][Medline]
33. Vasavada, R., K. B. Eager, G. Barbanti-Brodano, A. Caputo, and R. P. Ricciardi. 1986. Adenovirus type 12 early region 1A proteins repress class I HLA expression in transformed human cells. Proc. Natl. Acad. Sci. USA 83:5257-5261.[Abstract/Free Full Text]
34. Vermeulen, L., G. De Wilde, P. Van Damme, W. Vanden Berghe, and G. Haegeman. 2003. Transcriptional activation of the NF-kappa B p65 subunit by mitogen- and stress-activated protein kinase-1 (MSK1). EMBO J. 22:1313-1324.[CrossRef][Medline]
35. Viatour, P., M. P. Merville, V. Bours, and A. Chariot. 2005. Phosphorylation of NF-kappaB and IkappaB proteins: implications in cancer and inflammation. Trends Biochem. Sci. 30:43-52.[CrossRef][Medline]
36. Wang, D., and A. S. Baldwin, Jr. 1998. Activation of nuclear factor-kappaB-dependent transcription by tumor necrosis factor-alpha is mediated through phosphorylation of RelA/p65 on serine 529. J. Biol. Chem. 273:29411-29416.[Abstract/Free Full Text]
37. Wang, D., S. D. Westerheide, J. L. Hanson, and A. S. Baldwin, Jr. 2000. Tumor necrosis factor alpha-induced phosphorylation of RelA/p65 on Ser529 is controlled by casein kinase II. J. Biol. Chem. 275:32592-32597.[Abstract/Free Full Text]
38. Williams, J. F., Y. Zhang, M. A. Williams, S. Hou, D. Kushner, and R. P. Ricciardi. 2004. E1A-based determinants of oncogenicity in human adenovirus groups A and C. Curr. Top. Microbiol. Immunol. 273:245-288.[Medline]
39. Yewdell, J. W., J. R. Bennink, K. B. Eager, and R. P. Ricciardi. 1988. CTL recognition of adenovirus-transformed cells infected with influenza virus: lysis by anti-influenza CTL parallels adenovirus-12-induced suppression of class I MHC molecules. Virology 162:236-238.[CrossRef][Medline]
40. Zhao, B., S. Hou, and R. P. Ricciardi. 2003. Chromatin repression by COUP-TFII and HDAC dominates activation by NF-kappaB in regulating major histocompatibility complex class I transcription in adenovirus tumorigenic cells. Virology 306:68-76.[CrossRef][Medline]
41. Zhao, B., and R. P. Ricciardi. 2006. E1A is the component of the MHC class I enhancer complex that mediates HDAC chromatin repression in adenovirus-12 tumorigenic cells. Virology 352:338-344.[CrossRef][Medline]
42. Zhong, H., M. J. May, E. Jimi, and S. Ghosh. 2002. The phosphorylation status of nuclear NF-kappa B determines its association with CBP/p300 or HDAC-1. Mol. Cell 9:625-636.[CrossRef][Medline]
43. Zhong, H., H. SuYang, H. Erdjument-Bromage, P. Tempst, and S. Ghosh. 1997. The transcriptional activity of NF-kappaB is regulated by the IkappaB-associated PKAc subunit through a cyclic AMP-independent mechanism. Cell 89:413-424.[CrossRef][Medline]
44. Zhong, H., R. E. Voll, and S. Ghosh. 1998. Phosphorylation of NF-kappa B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell 1:661-671.[CrossRef][Medline]

This Article Abstract Full Text (PDF) Other Versions of this Article: JVI.01579-07v1 82/1/40    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Guan, H. Articles by Ricciardi, R. P. Search for Related Content PubMed PubMed Citation Articles by Guan, H. Articles by Ricciardi, R. P.