Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Differential Regulation of the Inhibitor of Apoptosis ch-IAP1 by v-
rel
and the Proto-Oncogene c-
rel
Jarmila Kralova,
1,2Andrew S. Liss,
1William Bargmann,
1Cullen Pendleton,
1Janani Varadarajan,
1Emin Ulug,
1and Henry R. Bose, Jr.
1*
Section of Molecular Genetics and Microbiology and the Institute of Cellular and Molecular Biology, University of Texas
at Austin, Austin, Texas 78712-1095,1and Institute of Molecular Genetics, Academy of Sciences of the Czech
Republic, 166 37 Prague 6, Czech Republic2
Received 2 May 2002/Accepted 23 August 2002
The v-rel oncogene encoded by reticuloendotheliosis virus is the acutely transforming member of the Rel/NF-B family of transcription factors. v-Rel is a truncated and mutated form of c-Rel and transforms cells by inducing the aberrant expression of genes regulated by Rel/NF-B proteins. The expression of ch-IAP1, a member of the inhibitor-of-apoptosis family, is highly elevated in cells expressing v-Rel and contributes to the immortalization of cells transformed by this oncoprotein. In this study we demonstrate that the elevated expression of ch-IAP1 in v-Rel-expressing cells is due to an increased rate of transcription. The ch-IAP1 promoter was isolated, and four Rel/NF-B binding sites were identified upstream of the transcription start site. Two B sites proximal to the transcription start site were required for v-Rel to activate the ch-IAP1 promoter. While c-Rel also utilized these sites, a third more-distalB site was required for its full activation of the ch-IAP1 promoter. Differences in the transactivation domains of v-Rel and c-Rel are responsible for their different abilities to utilize these sites and account for their differential activation of the ch-IAP1 promoter. Although c-Rel was a more potent activator of the ch-IAP1 promoter than v-Rel in transient reporter assays, cells stably overexpressing c-Rel failed to maintain high levels of ch-IAP1 expression. The reduction of ch-IAP1 expression in these cells correlated with the efficient regulation of c-Rel by IB␣. The ability of v-Rel to escape IB␣regulation allows for the gradual and sustained elevation of ch-IAP1 expression directly contributing to the transforming properties of v-Rel.
The Rel/NF-B family of transcription factors regulates the expression of genes involved in immune and stress responses, differentiation, proliferation, and apoptosis (10, 17, 29). Mem-bers of this family, which include c-Rel, v-Rel, RelA, RelB, NF-B1 (p50/p105), and NF-B2 (p52/p100), contain a highly conserved 300-amino-acid region in their N termini termed the Rel homology domain (RHD). The RHD contains sequences that allow these proteins to form homo- and heterodimers, bind DNA, and associate with inhibitory proteins (IBs) such as IB␣. In most cell types, Rel/NF-B dimers are normally sequestered as inactive complexes in the cytoplasm of cells by association with IBs. Appropriate extracellular stimuli induce the phosphorylation, ubiquitination, and subsequent degrada-tion of IBs, allowing for the nuclear translocation of active Rel/NF-B dimers (18). In the nuclei of cells, Rel/NF-B dimers exert their regulatory effects by binding to sequences (B sites) present in promoters and enhancers of target genes. The v-reloncogene is the only acutely transforming member of the Rel/NF-B family. Viruses expressing v-Rel induce a rapid and invariably fatal lymphoma in young birds 7 to 10 days after infection (3). v-relwas derived by a nonhomologous re-combination event between turkey c-reland reticuloendothe-liosis-associated virus strain A (REV-A). While c-Rel is weakly transforming, v-Rel has acquired a number of structural changes that account for its high transformation potential (9).
The transduction of c-relinto REV-A resulted in the deletion of sequences encoding 118 C-terminal amino acids of c-Rel, removing its most potent transactivation domain. This region of c-Rel also contains a cytoplasmic retention element, the loss of which contributes to the increased nuclear access of v-Rel (5). In addition, v-Rel has acquired a number of amino acid substitutions and deletions that alter its DNA-binding speci-ficity and make it refractory to IB␣regulation (5, 24, 27). The increased nuclear access and altered DNA binding and trans-activation properties of v-Rel contribute to its transforming potential by allowing the inappropriate activation or suppres-sion of genes normally regulated by Rel/NF-B family mem-bers (7, 9, 28).
The ability of a cell to escape apoptosis is a critical step in the transformation process, and several studies have demon-strated that the expression of v-Rel inhibits cell death induced by a variety of apoptotic stimuli (26, 30, 31, 35, 36). Apoptosis results from the activation of a conserved, constitutively ex-pressed group of cysteine proteases called caspases (1, 11). The activity of caspases is regulated, in part, by their interaction with members of the inhibitor-of-apoptosis (IAP) family. Our laboratory identified and cloned ch-IAP1, an avian IAP family member, and demonstrated that its expression was highly ele-vated in v-Rel-transformed cells but only modestly eleele-vated in cells transformed by the overexpression of c-Rel (32). Studies using lymphoid cells transformed by a temperature-sensitive v-Rel mutant revealed that induction of ch-IAP1 mRNA and protein correlated with the expression of v-Rel, suggesting that ch-IAP1 was directly regulated by v-Rel (32). Furthermore, ectopic expression of ch-IAP1 prevented these cells from
un-* Corresponding author. Mailing address: Section of Molecular Ge-netics and Microbiology, University of Texas at Austin, Austin, TX 78712-1095. Phone: (512) 471-5525. Fax: (512) 471-2130. E-mail: [email protected].
11960
on November 8, 2019 by guest
http://jvi.asm.org/
dergoing apoptosis at the nonpermissive temperature, indicat-ing that ch-IAP1 mediates, at least in part, the ability of v-Rel to inhibit apoptosis (32).
The studies reported here characterize the regulation of ch-IAP1 expression by v-Rel and c-Rel. A region of chicken genomic DNA encompassing the ch-IAP1 promoter was iso-lated and sequenced. Four Rel/NF-B binding sites were iden-tified within 386 bp of the transcription start site. The contri-bution of each site to the differential regulation of the ch-IAP1 promoter by v-Rel and c-Rel was evaluated. Although c-Rel was a more potent activator of ch-IAP1 transcription in re-porter assays, kinetic analyses revealed that v-Rel, but not c-Rel, induced sustained elevated levels of ch-IAP1. Our re-sults suggest that the high levels of ch-IAP1 in v-Rel-trans-formed cells result from the ability of v-Rel to escape the regulatory effects of⌱B␣.
MATERIALS AND METHODS
Cells and viruses.Chicken embryo fibroblast (CEF) cultures were prepared from embryonated SPAFAS or SC eggs (Charles River Laboratories, Preston, Conn.) as previously described (22). The avian leukosis virus-transformed chicken B-cell line DT40, the v-Rel-transformed B-cell line 7.4.1 (originally named RECC-UT1W41), and the Marek’s disease virus-transformed T-cell line MSB-1 have been previously described (2, 15, 34). All cell cultures were main-tained in Dulbecco’s modified Eagle’s medium supplemented with 2% chicken serum (Life Technologies, Gaithersburg, Md.), 3% newborn calf serum, 3% fetal calf serum (HyClone Laboratories, Logan, Utah), penicillin (100 U/ml), and streptomycin (50g/ml) and incubated in a 37°C incubator containing 8% CO2. REV-A, chicken syncytial virus (CSV), and REV-based retroviruses express-ing v-Rel (REV-TW) or c-Rel (REV-C) have been described previously (15). Retroviral stocks were prepared by transfecting CEF cultures with plasmids encoding these retroviruses and harvesting supernatant fluids after 5 to 7 days. Titers of REV-TW and REV-C viral stocks were determined by dot blot analysis, using REV-TW previously quantitated by an immunohistochemical assay as a standard (15, 25).
Nuclear run-on analysis.Nuclear run-on reactions were performed as previ-ously described (19). Briefly, genomic DNA (2g) isolated from each cell line and linearized plasmids (1g) encoding full-length c-junor ch-IAP1 or empty pBluescript SK(⫺) (Stratagene, La Jolla, Calif.) were denatured, neutralized, and slot blotted onto Hybond N⫹(Amersham Pharmacia Biotech, Piscataway,
N.J.). Nuclei were isolated from 108cells by using a modified NP-40 lysis protocol and were employed for the synthesis of run-on RNA transcripts in the presence of [32P]UTP (New England Nuclear, Boston, Mass.) (6). Equivalent amounts of radioactivity (2⫻107cpm) from each nuclear run-on reaction were hybridized with the prepared membranes at 40°C for 72 h. The membranes were washed extensively, and bound radioactivity was quantitated by phosphorimager analysis. RNA half-life and Northern blot analysis.Northern blot analysis was per-formed as previously described (28). Probes for ch-IAP1,-actin, and glyceral-dehyde-3-phosphate dehydrogenase (GAPDH) were prepared from full-length cDNA templates and labeled with [32P]dCTP (New England Nuclear) by using the Prime-a-Gene random-priming labeling kit (Promega, Madison, Wis.). For mRNA half-life analysis, DT40 cells infected with REV-A or REV-TW were treated with actinomycin D (2.5g/ml), and total RNA was isolated at various times after the start of treatment with the RNAwiz reagent according to the manufacturer’s directions (Ambion, Austin, Tex.). RNA (10g) was fractionated on a 1% agarose gel containing formaldehyde, transferred by capillary action to nylon membranes (Hybond N⫹; Amersham), and hybridized with a
ch-IAP1-specific probe by using the ULTRAhyb reagent according to the manufacturer’s directions (Ambion) (22). ch-IAP1 RNA levels were quantitated by phospho-rimager analysis. The blot was stripped and reprobed with an actin-specific probe to normalize the signals obtained. Due to the low level of expression of ch-IAP1 in DT40 cells infected with REV-A, longer exposure times were required for these blots.
Primer extension analysis.Primer extension was performed by using the Primer Extension System-AMV Reverse Transcriptase kit according to the man-ufacturer’s directions (Promega). A primer (5⬘-CGTTTGAGGTGCCTACGC-3⬘) complementary to nucleotides between 110 and 93 of the ch-IAP1 cDNA was end labeled with [32P]ATP using T4 polynucleotide kinase. This primer was
hybridized with poly(A)⫹RNA (1g) isolated from 7.4.1 or MSB-1 cells by
using the Oligotex mRNA mini kit (Qiagen, Valencia, Calif.). Extension prod-ucts were resolved by electrophoresis on a 7 M urea–acrylamide gel in parallel to a DNA sequencing reaction. Conventionally, the location of the transcription start site within the genomic clone is determined by comparing the size of a primer extension product to that of a sequencing reaction performed using the genomic clone as a template and the primer extension primer as the sequencing primer. However, this was not possible, since an intron is present between the annealing site for the primer extension primer and the sequence corresponding to the 5⬘ end of the ch-IAP1 cDNA. Therefore, a sequencing reaction was performed with pGEM-3Zf(⫹) and the M13 forward primer using thefmolDNA cycle sequencing system (Promega), which provided standards of known lengths. Isolation of a ch-IAP1 genomic clone.A chicken genomic library in bacterio-phage lambda (ATCC 37501; American Type Culture Collection, Manassas, Va.) was screened with a 650-bpApaI fragment from the 5⬘end of the ch-IAP1 cDNA as a probe. Two independently isolated clones containing sequences correspond-ing to the 5⬘end of the ch-IAP1 cDNA were isolated. Both clones appeared to contain fragments of identical size and composition, as determined by restriction mapping and Southern blot hybridization with the same probe. A fragment of approximately 10 kb flanked byEcoRI sites was subcloned into pBluescript SK(⫺) (E10), and both strands were sequenced by automated DNA sequencing at the University of Texas DNA Core Facility.
Reporter plasmids and site-directed mutagenesis.The 1.2-kbPstI fragment of E10 was subcloned into thePstI site of pBluescript SK(⫺) to generate P1.2. This fragment was excised by digestion withXhoI andSacI and ligated into the corresponding sites of the pGL3-Basic luciferase reporter vector (Promega) to prepare IAP1-luc. IAP1-luc was digested withBstXI andBsu36I, blunt ended with T4 DNA polymerase, and self-ligated to make IAP2-luc. AnXhoI-SacI fragment of IAP1-luc was digested withPspGI, blunt ended by treatment with Klenow, and then digested withHindIII. The resulting fragment was subcloned intoSmaI andHindIII sites of the pGL3-Basic vector to generate IAP3-Luc. The 1.2-kbPstI fragment of E10 was digested withBsrBI, and a 0.55-kbBsrBI/PstI fragment was purified, blunt ended with T4 DNA polymerase, and ligated into theSmaI site of pGL3-Basic to generate IAP4-luc. A 0.35-kbNcoI-PstI fragment of P1.2 was blunt ended with T4 DNA polymerase and subcloned into theSmaI site of pGL3-Basic to make IAP5-luc. All constructs were sequenced to verify proper orientation of the subcloned fragments.
Site-directed mutagenesis was performed using either the Altered Sites directed mutagenesis kit (Clontech, Palo Alto, Calif.) or the Gene Editor site-directed mutagenesis kit (Promega) in accordance with the manufacturers’ in-structions. To mutate each of the fourB sites in IAP2-luc,primersB1m (5⬘-GGCGCGGGGTTCtgCACGCAGAACGT-3⬘),B2m (5⬘-CGGGACGGC GGGTGCTTTgCTGGCGGCGCGG–3⬘), B1/2m (5⬘-CGGGACGGCGGGT GCTTTgCTGGCGGCGCGGGGGTTCtgCACGCAGAACGT-3⬘),B3m (5⬘-CGCGTCGCTGcaGCTTTCCGGC-3⬘), andB4m (5⬘-CGCTCCCCTCGcAG TTTCgCGGGCCGGGG-3⬘) were employed. The underlined sequences indicate the location of theB site, and lowercase letters indicate mutated nucleotide positions.
Transient-transfection and luciferase assays.For transient-transfection as-says, the expression vector pRc/RSV (Invitrogen, Carlsbad, Calif.) and pRc/RSV encoding v-Rel, c-Rel, c-Rel⌬T, or ttc-Rel were employed. The construction of c-Rel⌬T and ttc-Rel has been described previously (14). CEF cultures were plated in 60-mm tissue culture dishes (7.5⫻105cells/plate) 24 h before trans-fection. Cells were transfected using a calcium phosphate precipitation protocol (20). A total of 10g of DNA was used per transfection: 0.5g of a reporter plasmid, 0.1g of an expression plasmid, 0.5g of pRL-TK coreporter plasmid (Promega), and 8.9g of empty pBluescript plasmid DNA as a carrier. At 32 h posttransfection, cells were rinsed once with phosphate-buffered saline and har-vested by adding of 500l of 1⫻reporter lysis buffer (Promega) and scraping with a rubber policeman. Protein concentrations of the lysates were determined by using the Bio-Rad protein assay reagent (Bio-Rad, Hercules, Calif.). Lysates (6g of protein) were analyzed with the Dual Luciferase assay system (Promega) on a Dynex luminometer (Chantilly, Va.). Luciferase activity in each assay was normalized for transfection efficiency by measuring pRL-TK coreporter activity. The means (⫾standard errors) of two to six independent experiments are presented. Statistical analysis was performed by using Student’sttest. APvalue of⬍0.05 was considered significant.
Subcellular fractionation and immunoblotting.Nuclear and cytoplasmic frac-tions from CEF cultures infected with CSV, REV-C, or REV-TW were prepared by lysing cells in hypotonic buffer (50 mM Tris-HCl [pH 8.0], 1.1 mM MgCl2, and 0.5% Triton X-100) in the presence of protease inhibitors (23). Whole-cell extracts were prepared by lysing cells directly in sodium dodecyl sulfate-polyac-rylamide gel electrophoresis (SDS-PAGE) loading buffer. Proteins from cell
VOL. 76, 2002 REGULATION OF ch-IAP1 BY v-Rel AND c-Rel 11961
on November 8, 2019 by guest
http://jvi.asm.org/
lysates (5⫻105 cells) were resolved by SDS-PAGE and electrophoretically transferred to Optitran BA-reinforced nitrocellulose (Schleicher & Schuell, Keene, N.H.). Membranes were stained with Ponceau S to determine the quality of transfer and the equivalence of protein loading. Immunoblotting was per-formed with rabbit polyclonal antisera specific to ch-IAP1 or with monoclonal antibodies specific for v-Rel and c-Rel (HY87) (15). Proteins were detected using the Renaissance Western blot chemiluminescence reagent (Perkin-Elmer Life Sciences, Boston, Mass.).
EMSAs.Electrophoretic mobility shift assays (EMSAs) and the preparation of nuclear extracts from CEF cultures have been described previously (14, 28). The top strand of the wild-type (WT) and mutant (Mut) oligodeoxynucleotides used as probes for EMSAs were as follows:B1-WT (5⬘-CGTTCTGCGTGGGGAA CCCCGCGCCGCCAA-3⬘);B1-Mut (5⬘-CGTTCTGCGTGGGGAACtgCGC GCCGCCAA-3⬘);B2-WT (5⬘-GCGCCGCCAAGGAAAGCAgCCGCCGTCT CG-3⬘); B2-Mut (5⬘-GCGCCGCCAAGGAAAGCACCCGCCGTCTCG-3⬘) B3-WT (5⬘-GCGCGTCGCTGGGGCTTTCCGGCCGCTTCC-3⬘); B3-Mut (5⬘-GCGCGTCGCTGcaGCTTTCCGGCCGCTTCC-3⬘);B4-WT (5⬘-CGCTC CCCTCGGAGTTTCCCGGGCCGGGG-3⬘);B4-Mut (5⬘-CGCTCCCCTCGc AGTTTCgCGGGCCGGGG-3⬘). Underlined sequences indicate the location of the B sites, and lowercase letters represent mutated positions. The bottom strand of the probe was synthesized by annealing a 9-bp oligonucleotide com-plementary to the 3⬘end of the top strand and extending with Klenow in the presence of [32P]dCTP (14). Antibodies against v-Rel, NF-B1, and NF-B2 used in supershift analyses have been previously described (14, 21, 24). The c-Rel-specific antiserum (2348-6) was kindly provided by M. Hannink.
Nucleotide sequence accession number.The sequence of the ch-IAP1 genomic clone was submitted to GenBank (accession number AF311289).
RESULTS
The transcription rate of ch-IAP1 is increased in v-Rel-expressing cells. Previous results from this laboratory have demonstrated that the steady-state level of ch-IAP1 RNA is significantly elevated in v-Rel-transformed cells (32). Elevated ch-IAP1 RNA levels may result from an increased rate of their transcription and/or a decreased rate of degradation. To define whether the transcription rate of ch-IAP1 is increased in v-Rel-expressing cells, nuclear run-on assays were performed (19). Nuclei were isolated from DT40 cells, DT40 cells infected with the helper virus REV-A (DT40-REV-A), or DT40 cells in-fected with a retrovirus expressing v-Rel (DT40-TW) and used in nuclear run-on reactions containing [32P]UTP to produce
radiolabeled pools of RNA. This RNA was hybridized with membranes onto which the ch-IAP1 cDNA had been immobi-lized. This analysis demonstrated that the transcription rate of ch-IAP1 was elevated approximately 10-fold in DT40-TW cells relative to the rate in DT40 or DT40-REV-A cells (Fig. 1A). The cDNA of c-junwas included as a positive control for these experiments. The transcription of c-jun was elevated in DT40-TW cells relative to control cells, consistent with earlier findings that c-junis transcriptionally upregulated by v-Rel (7, 8, 21). Hybridization of the labeled RNA to genomic DNA demonstrated that comparable incorporation of radioactivity occurred in the nuclear run-on reactions, while negligible hy-bridization occurred with the pBluescript SK(⫺) negative con-trol. This experiment was performed two additional times with similar results. Nuclear run-on experiments also revealed a 10-fold increase in the rate of ch-IAP1 transcription following overexpression of v-Rel in DT95 cells, an avian B-cell line (data not shown).
Analysis of the ch-IAP1 RNA half-life was performed to determine if changes in RNA stability also contribute to the increased steady-state level of ch-IAP1 in v-Rel-expressing cells. Total RNA was isolated from DT40-REV-A and DT40-TW cultures at various times after treatment with
acti-nomycin D (2.5g/ml) and subjected to Northern blot analysis with a ch-IAP1-specific probe (Fig. 1B, top panel). Northern blots were quantitated by phosphorimager analysis and nor-malized to the signal obtained after stripping and hybridization with an actin-specific probe (Fig. 1B, lower panel). The ch-IAP1 RNA exhibited a half-life of approximately 1 to 2 h in DT40 cells expressing v-Rel, in comparison to 4 to 6 h in control REV-A-infected DT40 cultures, indicating that the turnover of ch-IAP1 RNA is also enhanced in v-Rel-trans-formed cells. Taken together, these experiments indicate that the elevated levels of ch-IAP1 RNA in v-Rel-expressing cells are due to an increase in the rate of transcription, rather than an increase in RNA stability.
[image:3.603.302.542.71.309.2]Isolation and characterization of ch-IAP1 promoter se-quences.To determine whether v-Rel directly contributes to the increased transcription of ch-IAP1, the promoter of the ch-IAP1 gene was isolated. A chicken genomic library was screened with a probe specific to the 5⬘ end of the ch-IAP1 cDNA. Two identical phage clones were isolated after screen-ing approximately 250,000 plaques. Southern blot analysis, re-striction mapping, and sequencing of a 10-kbEcoRI fragment
FIG. 1. Mechanism of increased ch-IAP1 RNA levels in cells ex-pressing v-Rel. DT40 cells were infected with the helper virus REV-A or a retrovirus expressing v-Rel (REV-TW). (A) Nuclear run-on anal-yses of ch-IAP1 expression. Nuclei isolated from DT40 cells or DT40 cells infected with REV-A or REV-TW were used to produce run-on transcripts in the presence of [32P]UTP. Equivalent amounts of
radio-activity from these reactions were hybridized to membranes containing genomic DNA from each cell type, pBluescript plasmid DNA, and plasmids encoding ch-IAP1 and c-jun. Bound radioactivity was visual-ized by phosphorimager analysis. (B) Half-life analysis of ch-IAP1 RNA in DT40 cells infected with REV-A or REV-TW. Total RNA (10 g) was isolated from cells treated with actinomycin D (2.5g/ml) for 0, 1, 2, 4, 6, 8, and 10 h and analyzed for ch-IAP1 by Northern blot analysis (top panel). Due to low levels of ch-IAP1 mRNA expression in REV-A-infected DT40 cells, a longer exposure of the Northern blot from these cells is shown than for REV-TW-infected cells. ch-IAP1 RNA expression was normalized for loading by comparing the expres-sion of-actin (lower panel).
on November 8, 2019 by guest
http://jvi.asm.org/
revealed the junction of the 5⬘ untranslated region of the ch-IAP1 cDNA and at least 5 kb of upstream sequences en-compassing the putative promoter region. This fragment also contains a significant, though incomplete, portion of the coding region of ch-IAP1. The sequence of approximately 2 kb of this genomic clone is shown in Fig. 2.
Primer extension analysis was performed to define the tran-scription start site for the ch-IAP1 gene. For these experi-ments, poly(A)⫹RNA was isolated from a v-Rel-transformed
cell line (7.4.1) that expresses high levels of ch-IAP1 and a chicken T-cell line (MSB-1) that expresses very low amounts of ch-IAP1 mRNA. The differential expression of ch-IAP1 by these cell lines was demonstrated by Northern blot analysis (Fig. 3A). A primer complementary to the 5⬘ untranslated region of the ch-IAP1 cDNA was end labeled with [␥-32P]ATP,
hybridized to 1g of poly(A)⫹RNA, and extended with
re-verse transcriptase. The products were analyzed by electro-phoresis on a sequencing gel with a DNA sequencing ladder that permitted the sizing of the extension products to 1-bp resolution (Fig. 3B). A primer extension product was not de-tected when mRNA from MSB-1 cells was analyzed (lane 1), consistent with the low levels of ch-IAP1 detected by Northern blot analysis. However, reactions containing mRNA isolated from v-Rel-transformed cells produced multiple extension products (lane 2). The most abundant primer extension prod-uct is 119 bases long and represents 35% of the total primer extension products. The corresponding nucleotide in the
se-quence of the ch-IAP1 genomic clone has been denoted as⫹1 (Fig. 2). Consistent with this, a putative TATA box resides 33 nucleotides upstream of this transcription start site. The sizes of other primer extension products corresponded to transcrip-tion start sites upstream of the cloned cDNA, while additranscrip-tional minor products corresponded to transcription start sites in sequences found at the beginning of exon 2 (Fig. 2). The same pattern of primer extension products was observed in studies employing RNA isolated from DT95 cells expressing v-Rel (data not shown).
c-Rel stimulates the ch-IAP1 promoter more efficiently and through different B sites than v-Rel does. The exogenous expression of v-Rel and c-Rel correlates with the induction of ch-IAP1 RNA in fibroblasts and lymphoid cells (32). Transient reporter assays were employed to evaluate the effect of v-Rel and c-Rel on ch-IAP1 promoter activity and to map the min-imal ch-IAP1 promoter element that could confer inducibility to reporter genes by these Rel/NF-B proteins (Fig. 4). Five promoter constructs derived from a 1.2-kbPstI fragment of the ch-IAP1 genomic clone were introduced in the pGL3-Basic luciferase reporter vector (Promega). CEF cultures were co-transfected with these constructs and the Rc/RSV expression plasmid or Rc/RSV encoding either v-Rel or c-Rel. Luciferase activity in extracts of the transfected cells was assayed after 32 h, a time when maximal activity was observed. The largest reporter construct (IAP1-luc), containing 1,001 bases up-stream of the major transcription start site, exhibited
approx-FIG. 2. Sequence analysis of the ch-IAP1 promoter. The sequence of a ch-IAP1 genomic clone from⫺1001 to⫹977 is shown. The location of the major (arrow) and minor (asterisks) transcription start sites are indicated. The sequence is numbered relative to the major transcription start site (⫹1). The gray box identifies a potential TATA box at position⫺33. The nucleotide corresponding to the 5⬘end of the ch-IAP1 cDNA is indicated by a black box. Lowercase letters represent sequences encoding the first intron of ch-IAP1, identified by comparison to the ch-IAP1 cDNA. Four potentialB sites (B1 to -4) are underlined and labeled. Sequences complementary to the primer employed for primer extension are indicated by a double underline. Bold letters indicate the twoPstI restriction sites employed to create the IAP1-luc reporter vector.
VOL. 76, 2002 REGULATION OF ch-IAP1 BY v-Rel AND c-Rel 11963
on November 8, 2019 by guest
http://jvi.asm.org/
imately twofold-greater reporter activity when cotransfected with a vector encoding v-Rel than when a vector control was used (Fig. 4A). This level of induction is comparable to that observed previously in studies demonstrating the direct activa-tion of the avianikbapromoter by v-Rel (20). Successive de-letions of up to 822 nucleotides from the 5⬘ end of this pro-moter sequence (IAP2-luc to IAP5-luc) did not significantly alter the v-Rel-inducible response of the promoter relative to that in Rc/RSV controls. These results indicate that sequences
in the ch-IAP1 promoter within 179 bases upstream of the transcription start site are sufficient for the induction of ch-IAP1 expression by v-Rel.
c-Rel is a more potent activator of transcription than v-Rel, and the ectopic expression of c-Rel increased reporter activity of IAP1-luc and IAP2-luc by approximately fourfold (Fig. 4A). In contrast to v-Rel, sequential deletion of sequences between
⫺367 and⫺179 (IAP3-luc to IAP5-luc) reduced the ability of c-Rel to induce reporter activity by half. ch-IAP1 promoter constructs lacking these sequences were activated by c-Rel and v-Rel to a comparable extent. These experiments establish that v-Rel and c-Rel differentially activate the ch-IAP1 promoter and that these proteins affect this activation through different sequences.
Inspection of the 1.2-kb PstI fragment of the ch-IAP1 genomic clone revealed at least four potential B sites up-stream of the transcription start site (Fig. 2 and 4). The loca-tions of these B sites corresponded to sequences important for the Rel-inducible expression of the ch-IAP1 promoter (Fig. 4A). OneB site (B1) was found in the region of the ch-IAP1 promoter necessary for full activation by c-Rel, whileB3 and
B4 were found in the smallest reporter construct that is re-sponsive to both v-Rel and c-Rel (IAP5-luc). To confirm that these sequences represent authenticB binding sites, EMSAs were performed (Fig. 5). Oligonucleotides corresponding to these sequences were synthesized, labeled with [32P]dCTP, and
used in EMSA reactions with nuclear extracts prepared from CEF cultures infected with the helper virus CSV or with ret-roviruses expressing v-Rel or c-Rel (Fig. 5A). All sites bound proteins from nuclear extracts of cells expressing v-Rel. Com-plexes from these cells most efficiently bound toB1,B3, and
B4 and most weakly bound toB2. Although nuclear extracts from cells overexpressing c-Rel bound the B sites with a similar relative affinity, considerably lower levels of binding activity were observed. The differences in DNA binding activity observed in nuclear extracts from v-Rel- and c-Rel-overex-pressing cells is more dramatic than that represented in Fig. 5, since fivefold more protein was used in the EMSA analysis for c-Rel-overexpressing cells. The binding to these probes was specific for theB sites, since mutations in these sites abol-ished or greatly reduced binding in these assays. Additional studies revealed that the DNA binding complexes from v-Rel-expressing cells were supershifted by the addition of antisera specific for v-Rel and that complexes from cells overexpressing c-Rel failed to bind DNA in the presence of antisera specific for c-Rel, thereby demonstrating the presence of v-Rel and c-Rel in these DNA binding complexes (Fig. 5B). As the su-pershift analysis for all potential B sites provided similar results, only those for theB3 site are presented.
To define whether the use of theseB sites by v-Rel or c-Rel can account for their ability to activate the ch-IAP1 promoter, site-directed mutagenesis was performed (Fig. 4B). Each B site was mutated individually or in combination in the context of the smallest reporter vector that conferred maximum re-sponse to v-Rel and c-Rel (IAP2-luc). Reporter assays were then performed following coexpression of these constructs with vectors expressing c-Rel or v-Rel (Fig. 4C). Mutations in
B1 (IAP6-luc) or B2 (IAP7-luc) alone or in combination (IAP11-luc) did not significantly alter the ability of v-Rel to activate the reporter construct. However, mutations in B3
FIG. 3. Identification of the ch-IAP1 transcription start sites. (A) Expression of ch-IAP1 mRNA in lymphoid cells. Poly(A)⫹RNA
(1g) from MSB-1 (lane 1) or 7.4.1 (lane 2) cells was analyzed for ch-IAP1 expression by Northern blot analysis (top panel). The expres-sion of GAPDH in these cells is shown in the lower panel. (B) Primer extension analysis of the ch-IAP1 mRNA. A probe specific to the ch-IAP1 cDNA was hybridized with poly(A)⫹ RNA (1 g) from
MSB-1 (lane 1) or 7.4.1 (lane 2) cells and extended with reverse transcriptase. Products of these reactions and DNA sequencing stan-dards were analyzed by denaturing polyacrylamide gel electrophoresis and visualized by phosphorimager analysis. The locations of the major (arrow) and minor (asterisks) primer extension products are indicated.
on November 8, 2019 by guest
http://jvi.asm.org/
FIG. 4. v-Rel and c-Rel differentially activate the ch-IAP1 promoter. CEF cultures were transfected with a ch-IAP1 reporter construct, the pRL-TK coreporter vector, and an empty expression plasmid (Rc/RSV) as a control, or Rc/RSV encoding v-Rel or c-Rel. Cells were harvested 32 h after transfection and analyzed for luciferase activity. Luciferase activity from the ch-IAP1 promoter constructs was normalized for transfection efficiency by measuring pRL-TK coreporter activity. The numbers presented are the mean (⫾ standard error) of two to six independent experiments. Reporter activities in extracts from cells overexpressing v-Rel or c-Rel that are significantly different (P⬍0.05) from Rc/RSV controls are indicated by an asterisk. (A) Deletion constructs of the ch-IAP1 promoter were subcloned into the pGL3-Basic reporter vector as described in Materials and Methods. Diagrams of the reporter constructs are shown to the left of the luciferase assay results. The relative positions of theB sites (I) and the major transcription start site (arrow) are indicated. Luciferase assay results are presented as relative luciferase units per microgram of protein. (B) The sequence of the wild-typeB sites (top row) and mutantB sites (bottom row) are shown. (C) Reporter constructs with wild-type (I) or mutant (⫻)B sites were constructed in pGL3-Basic as described in Materials and Methods. The results are presented as fold activation relative to results obtained with an empty expression plasmid control.
11965
on November 8, 2019 by guest
http://jvi.asm.org/
(IAP8-luc) orB4 (IAP9-luc) alone abolished the responsive-ness of this promoter element to v-Rel. Similar results were obtained with a reporter construct that had mutations in both of these sites (IAP10-luc). These results indicate that v-Rel activates transcription from the ch-IAP1 promoter throughB sites most proximal to the transcription start site (B3 and
B4), while the most distalB sites (B1 and B2) are not required for ch-IAP1 activation by v-Rel.
The regulation of the ch-IAP1 promoter by c-Rel was more complex than that observed for v-Rel. Mutations in B1 (IAP6-luc) resulted in an approximately 40% reduction in the activation of the ch-IAP1 promoter by c-Rel. Mutations inB2 (IAP7-luc) did not significantly affect the ability of c-Rel to activate the ch-IAP1 promoter. The inability of c-Rel to acti-vate the ch-IAP1 promoter from this site is consistent with the poor DNA binding activity of c-Rel complexes to B2 (Fig. 5A). Mutations in bothB1 andB2 (IAP11-luc) reduced the c-Rel-inducible expression of the reporter only slightly more than that observed forB1 alone. Mutation ofB3 (IAP8-luc) orB4 (IAP9-luc) alone resulted in a 40 to 50% reduction in the ability of c-Rel to activate the ch-IAP1 promoter, relative to that of IAP2-luc. Mutation of both B3 andB4 (IAP10-luc) further reduced the ability of c-Rel to activate the reporter vector. However, when B1, B3, and B4 were mutated in combination (IAP12-luc), the promoter failed to exhibit any responsiveness to c-Rel. These results suggest that c-Rel acti-vates the transcription of ch-IAP1 throughB sites located in two distinct regions of the ch-IAP1 promoter.
Differences between the C termini of v-Rel and c-Rel medi-ate their differential effects on the ch-IAP1 promoter. v-Rel and c-Rel differ with regard to both their DNA binding spec-ificities and their transactivation potentials, and these differ-ences may account for their differential regulation of the ch-IAP1 promoter (9). To determine the structural elements in c-Rel responsible for its ability to more efficiently activate ch-IAP1 promoter activity, a c-Rel protein C-terminally trun-cated to the size of v-Rel (c-Rel⌬T) and a chimeric protein containing the N terminus of v-Rel and the C-terminal
trans-activation sequences of c-Rel (ttc-Rel) were tested for their ability to activate the ch-IAP1 promoter (Fig. 6A and B) (24). While c-Rel was able to induce the activation of the IAP2-luc reporter construct fourfold, c-Rel⌬T activated the reporter at levels similar to v-Rel (twofold). In addition, c-Rel⌬T activated a reporter construct containing a mutatedB1 site at reduced levels relative to c-Rel. These results indicate that the ability of c-Rel to utilize the B1 site to activate ch-IAP1 promoter activity is due to its potent C-terminal transactivation domain. Consistent with this, a v-Rel protein containing the C-terminal transactivation domain of c-Rel (ttc-Rel) was more efficient at activating IAP2-luc than v-Rel. This enhanced activation was mediated, at least in part, through theB1 site, since mutation ofB1 (IAP6-luc) reduced the activation of the promoter by ttc-Rel. Moreover, unlike v-Rel, ttc-Rel was able to activate ch-IAP1 expression when theB3 andB4 sites were mutated (IAP10-luc), albeit at low levels.
[image:7.603.70.519.68.234.2]Surprisingly, ttc-Rel was more efficient than c-Rel at activat-ing IAP2-luc in reporter assays (eightfold versus fourfold). This difference was not due to differences in the expression of these proteins, as each of the Rel proteins was expressed at comparable levels from the Rc/RSV vectors (Fig. 6C). In ad-dition, since these proteins contain the same C-terminal se-quences, differences in the transactivation potentials of these proteins are unlikely to account for their differential activation of ch-IAP1 reporter constructs. Therefore, EMSAs were per-formed to determine whether the abilities of these proteins to activate the ch-IAP1 promoter correlated with their binding affinities to theB1 site (Fig. 6D). EMSAs performed using in vitro-translated v-Rel, c-Rel, c-Rel⌬T, and ttc-Rel demon-strated that proteins containing N-terminal v-Rel sequences (v-Rel and ttc-Rel) bound to B1 more efficiently than did those with N-terminal c-Rel sequences (c-Rel and c-Rel⌬T). The higher DNA binding affinity of v-Rel relative to c-Rel in these experiments indicates that the inability of v-Rel to acti-vate transcription from B sites not in close proximity to the transcription start site is due to its lack of a potent transacti-vation domain.
FIG. 5. MultipleB sites in the ch-IAP1 promoter are bound by v-Rel and c-Rel. (A) Nuclear extracts were isolated from CEF cultures 2 days after infection with the helper virus CSV (H), REV-TW (V), or REV-C (C). EMSAs were performed using wild-type or mutant probes encompassing theB sites found in the ch-IAP1 promoter with 5g of nuclear extracts from CSV- or REV-C-infected cells or 1g of nuclear extracts from REV-TW-infected cells. The migration profile of free probe (F) for each wild-typeB site is shown. (B) Supershift analysis ofB binding complexes. Nuclear extracts from CEF cultures infected with REV-TW (v-Rel) or REV-C (c-Rel) were incubated with normal rabbit serum or antisera specific for v-Rel or c-Rel prior to the addition ofB3 probe.
on November 8, 2019 by guest
http://jvi.asm.org/
v-Rel, but not c-Rel, induces a sustained elevated expression of endogenous ch-IAP1 during transformation.We have pre-viously observed that the levels of ch-IAP1 are higher in cells transformed by v-Rel than c-Rel (32). However, the results described here demonstrate that c-Rel more efficiently acti-vates transcription from the ch-IAP1 promoter than does v-Rel (Fig. 4). To resolve this paradox, ch-IAP1 induction was mon-itored at various times following overexpression of v-Rel and c-Rel. CEF cultures were infected with retroviruses expressing v-Rel or c-Rel, or with the helper virus CSV. RNA was har-vested at 2, 4, 8, and 14 days after infection and analyzed for the expression of ch-IAP1 by Northern blot analysis (Fig. 7A). Two days after infection, the levels of ch-IAP1 RNA were elevated in cells overexpressing v-Rel, relative to control cells. The levels of ch-IAP1 in v-Rel-expressing cells gradually in-creased over time, attaining their highest levels when the cul-tures were morphologically transformed (day 14). Cells over-expressing c-Rel also contained elevated levels of ch-IAP1 RNA by 2 days after infection. However, in contrast to the expression pattern observed in cells expressing v-Rel, ch-IAP1 RNA levels in cells overexpressing c-Rel steadily decreased between 2 and 14 days. Similar changes in ch-IAP1 protein expression were observed in CEF cultures overexpressing v-Rel or c-v-Rel (Fig. 7B, top panel). The changes in ch-IAP1 expression over time were not due to alterations in the levels of v-Rel or c-Rel (Fig. 7B, lower panel). These results indicate that c-Rel transiently induces the expression of ch-IAP1, whereas v-Rel induces a sustained elevated expression of ch-IAP1 in transformed cells.
The differential ability of v-Rel and c-Rel to transform cells is due, in part, to their differential regulation by IB␣ (27). IB␣efficiently sequesters c-Rel in the cytoplasm of cells and directly inhibits its ability to bind DNA (4, 5, 27). However, v-Rel has acquired mutations that make it refractory to IB␣
[image:8.603.52.265.71.600.2]regulation (27). To determine if the differential induction of ch-IAP1 expression by v-Rel and c-Rel correlates with the regulatory effects of IB␣, the subcellular localization and DNA binding activity of v-Rel and c-Rel were evaluated at various times after infection (Fig. 8). Nuclear and cytoplasmic fractions were prepared from the CEF cultures described
FIG. 6. Identification of c-Rel sequences responsible for enhanced activation of the ch-IAP promoter. (A) Diagram of v-Rel and c-Rel proteins. The location of the RHD and transactivation domains I (TAD I) and II (TAD II) are indicated for c-Rel. Vertical lines in v-Rel indicate amino acid differences between v-Rel and c-Rel. The black boxes represent the envelope-derived sequences in v-Rel. A c-Rel construct (c-Rel⌬T) lacking the 118 C-terminal amino acids missing in v-Rel and a v-Rel construct (ttc-Rel) containing the C-terminal transactivation sequences of c-Rel are shown. (B) Activation of the ch-IAP1 promoter by Rel proteins. CEF cultures were trans-fected with a ch-IAP1 reporter construct containing wild-type ( ) or mutant (⫻)B sites, the pRL-TK coreporter vector, and an empty expression plasmid (Rc/RSV) as a control, or Rc/RSV encoding v-Rel, c-Rel, c-Rel⌬T, or ttc-Rel. Cells were harvested 32 h after transfection
and analyzed for luciferase activity. Luciferase activity from the ch-IAP1 promoter constructs was normalized for transfection efficiency by measuring pRL-TK coreporter activity. The results presented are the mean fold activation (⫾standard error) of three independent exper-iments. Reporter activities in extracts from cells overexpressing Rel proteins that are significantly different (P⬍0.05) from Rc/RSV con-trols are indicated by an asterisk. (C) Expression of transiently ex-pressed Rel proteins. CEF cultures were transfected with an empty Rc/RSV plasmid (1g) or Rc/RSV encoding v-Rel, c-Rel, c-Rel⌬T, or ttc-Rel. Cells lysates were harvested 32 h after transfection and ana-lyzed for the expression of Rel proteins by Western blotting. Exog-enously expressed c-Rel and ttc-Rel comigrate with endogExog-enously expressed c-Rel, while v-Rel and c-Rel⌬T migrate as lower-molecular-weight proteins. (D) DNA binding activity of chimeric Rel proteins. v-Rel, c-Rel, c-Rel⌬T, or ttc-Rel were translated in vitro, and DNA binding activity was determined by EMSAs using the B1 site as a probe (top panel). Equal amounts of each protein were used in each reaction as determined by phophorimager analysis of parallel in vitro translation reactions performed with [35S]methionine and were carried
out in parallel reactions to show equivalent protein loading in the EMSA analysis (lower panel).
VOL. 76, 2002 REGULATION OF ch-IAP1 BY v-Rel AND c-Rel 11967
on November 8, 2019 by guest
http://jvi.asm.org/
above. Two days after infection, cells infected with retroviruses expressing c-Rel contained nuclear and cytoplasmic c-Rel, al-though the cytoplasmic levels were much higher (Fig. 8A). Between 2 and 14 days after infection, the nuclear levels of c-Rel decreased. This difference was not due to changes in the total amount of c-Rel in cells, since no significant differences in the levels of c-Rel in whole-cell and cytoplasmic extracts were observed (Fig. 7B and 8A). In contrast to c-Rel, cells express-ing v-Rel contained comparable levels of nuclear and cytoplas-mic v-Rel 2 days after infection. Moreover, the nuclear levels of v-Rel decreased more slowly and to a lesser extent than that observed for c-Rel. The more-efficient nuclear localization of v-Rel was not due to a lower induction of IB␣, because both v-Rel and c-Rel efficiently induced⌱B␣expression as early as 2 days after infection (data not shown).
EMSA reactions were performed to determine whether changes in ch-IAP1 expression also correlated with changes in the DNA binding activity of v-Rel and c-Rel.B1 DNA bind-ing activity in nuclear extracts isolated from CEF cultures infected with retroviruses expressing c-Rel rapidly decreased between 2 and 14 days after infection (Fig. 8B). In contrast, the binding activity in nuclear extracts from v-Rel-expressing cells was much stronger and remained relatively steady between 2 and 14 days after infection. Similar results were observed for each of the fourB sites found in the ch-IAP1 promoter (Fig. 8C). These experiments demonstrate that shortly after its over-expression in cells, c-Rel is rapidly prevented from binding
DNA and is efficiently retained in the cytoplasm. In contrast, v-Rel maintains its high nuclear levels and DNA binding ac-tivity over time.
In addition to the sustained level of DNA binding activity in nuclear extracts from cells expressing v-Rel, one additional DNA binding complex was observed bound to each of the four
B sites at later stages of transformation (day 14). The appear-ance of this complex (Fig. 8B and C) coincided with a signifi-cant increase in the levels of ch-IAP1 (Fig. 7). Supershift anal-ysis was employed, using theB1 site as a probe to identify the proteins in this binding complex (Fig. 8D). Whereas each of the binding complexes was efficiently supershifted with v-Rel-specific antiserum, none of the complexes was affected by the addition of antiserum specific for c-Rel or NF-B2 (data not shown). However, the DNA binding complex that increased in abundance over time was supershifted with antiserum specific for NF-B1, indicating that it is composed of v-Rel/NF-B1 heterodimers. Similar experiments employing theB2 andB3 sites as probes confirmed that this additional complex was composed of v-Rel/NF-B1 heterodimers (data not shown).
DISCUSSION
Our previous studies have shown that the expression of ch-IAP1 is elevated in v-Rel-transformed cells (32). In this report, nuclear run-on and RNA half-life analyses demonstrate that the elevated steady-state levels of ch-IAP1 RNA in cells expressing v-Rel is due to an increase in its rate of transcription (Fig. 1). A genomic clone containing the ch-IAP1 promoter was isolated in order to investigate the mechanisms by which v-Rel elevates the rate of ch-IAP1 transcription (Fig. 2). Primer extension experiments identified the major transcrip-tion start site of ch-IAP1 12 bp upstream of the cloned cDNA. Previous studies have cloned and partially characterized a genomic clone of c-IAP2, the human homolog of ch-IAP1 (12, 33). Although the transcription start site of c-IAP2 has not been accurately defined, a genomic sequence upstream of the cloned cDNA that confers Rel/NF-B-inducible expression to reporter constructs has been identified (12). The gross archi-tecture of this genomic sequence and the ch-IAP1 promoter described here are similar, and both contain multiple func-tional Rel/NF-B sites. Results presented here indicate that ch-IAP1 is directly regulated by Rel/NF-B proteins and, moreover, the mechanisms that account for the difference in the ability of v-Rel and c-Rel to enhance the expression of ch-IAP1 have been defined.
The transient overexpression of v-Rel or c-Rel resulted in the activation of reporter vectors containing ch-IAP1 promoter sequences, although c-Rel did so approximately two times more efficiently than v-Rel (Fig. 4A). Three B sites were found within 384 bp of the major transcription start site that conferred Rel-inducible activation of this promoter. The two
B sites most proximal to the transcription start site (B3 and
B4) were the only B sites required for activation of the ch-IAP1 promoter by v-Rel (Fig. 4C). In contrast, the most distalB site (B1) and the two proximalB sites (B3 and
B4) were all required for the full activation of the ch-IAP1 promoter by c-Rel. The additional activation of ch-IAP1 ex-pression by c-Rel relative to v-Rel was largely due to the ability of c-Rel to activate transcription fromB1 (Fig. 4C).
Muta-FIG. 7. Kinetic analysis of ch-IAP1 expression. CEF cultures were infected with the helper virus CSV (H), REV-C (C), or REV-TW (TW). Total RNA and whole-cell extracts were isolated at 2, 4, 8, and 14 days after infection. (A) Northern blot analysis of ch-IAP1 expres-sion. Total RNA (10g) from cultures infected with CSV, REV-C, or REV-TW was analyzed for the expression of ch-IAP1 RNA (top pan-el). The expression of GAPDH in these samples is shown to demon-strate the equal loading of RNA (lower panel). (B) Protein from whole-cell lysates (5⫻105cells) from these cells were resolved by
SDS-PAGE, transferred to nitrocellulose, and analyzed by Western blotting with the antiserum specific for ch-IAP1 (top panel) or anti-Rel monoclonal antibody HY87 (lower panel). The migrations of ch-IAP1,
v-Rel, and c-Rel are indicated.
on November 8, 2019 by guest
http://jvi.asm.org/
[image:9.603.43.286.73.291.2]tions in this sequence diminished the activation of the ch-IAP1 promoter by c-Rel to levels similar to those observed for v-Rel. Interestingly, the binding activity of nuclear extracts from cells expressing v-Rel or c-Rel toB1 was comparable to that of
B3 andB4 (Fig. 5). Therefore, the ability of c-Rel to utilize and activate ch-IAP1 expression from this distal site was due to the presence of its potent C-terminal transactivation domain, rather than a preferential binding affinity. v-Rel lacks this transactivation domain and, presumably, can only execute its weak transactivation potential fromB sites in close proximity to the transcription start site. Consistent with this interpreta-tion, a c-Rel mutant that had this transactivation domain re-moved to the same extent as v-Rel activated reporter con-structs to a similar extent as v-Rel (Fig. 6B). Furthermore, a v-Rel mutant containing the C-terminal transactivation do-main of c-Rel was able to activate reporter constructs from the distalB1 site. This is the first demonstration that c-Rel acti-vates a promoter through aB site distinct from that used by v-Rel.
While the results described in this report demonstrate that c-Rel is a more potent activator of the ch-IAP1 promoter than v-Rel, our previous studies demonstrated higher levels of ch-IAP1 RNA in cells expressing v-Rel than in cells overexpress-ing c-Rel (32). Kinetic analysis of ch-IAP1 expression demon-strated that ch-IAP1 levels were efficiently elevated in cells infected with viruses expressing v-Rel or c-Rel shortly after infection (Fig. 7A). However, ch-IAP1 levels rapidly decrease over a 14-day period in cells overexpressing c-Rel. This de-crease in ch-IAP1 expression was correlated with a dede-crease in c-Rel DNA binding activity and the relocalization of c-Rel to the cytoplasm (Fig. 7B and 8). These results, therefore, suggest that the efficient regulation of c-Rel by IB␣ results in its failure to induce sustained elevated levels of ch-IAP1.
[image:10.603.43.279.67.575.2]In contrast, the levels of ch-IAP1 gradually increase in cells while they undergo transformation mediated by v-Rel (Fig. 7A). The increase in ch-IAP1 expression during transforma-tion is consistent with a secondary activatransforma-tion of the ch-IAP1 promoter by additional factors. Early after infection of cells with retroviruses expressing v-Rel, v-Rel homodimers are the predominant nuclear Rel/NF-B complex (14). Approximately 1 week after infection, v-Rel heterodimers with endogenous Rel/NF-B proteins are found in the nuclei of cells (14). We observed in this study that the appearance of v-Rel/NF-B1 heterodimers in B DNA binding complexes coincided with the increased expression of ch-IAP1 (Fig. 7 and 8). Cotrans-fection of v-Rel and NF-B1 resulted in a small but reproduc-ible increase in ch-IAP1 reporter activity (data not shown), suggesting a possible role for this complex in the enhanced expression of ch-IAP1 during transformation. Although the induction of ch-IAP1 reporter activity by v-Rel requires the presence of functionalB sites in transient-transfection assays,
FIG. 8. Kinetic analysis of the subcellular localization andB bind-ing activity of v-Rel and c-Rel. (A) Subcellular localization of v-Rel and c-Rel. Proteins from cytoplasmic (C) and nuclear (N) extracts (5 ⫻105cells) from cells infected with CSV (top panel), REV-C (middle
panel), or REV-TW (lower panel) were resolved by SDS-PAGE, trans-ferred to nitrocellulose, and analyzed by Western blotting with the anti-Rel monoclonal antibody HY87. The migration of v-Rel and c-Rel are indicated. (B)B binding activity in cells overexpressing v-Rel or c-Rel. Nuclear extracts were isolated from CEF cultures 2, 4, 8, and 14 days after infection with the helper virus CSV, REV-C (c-Rel), or REV-TW (v-Rel). EMSAs were performed using wild-type probes encompassing theB1 site found in the ch-IAP1 promoter with 5g of nuclear extract from CSV- or REV-C-infected cells or 1g of nuclear extract from REV-TW-infected cells. A DNA binding complex that increases in abundance between 2 and 14 days is indicated by an asterisk. (C) EMSAs were performed as described for panel A by using
wild-type probes encompassing theB1,B2,B3, andB4 sites with nuclear extracts isolated from cells 2 or 14 days after infection. (D) Su-pershift analysis ofB binding complexes. Nuclear extracts from CEF cultures isolated 14 days after infection with REV-TW were incubated with normal rabbit serum or antisera specific for v-Rel, NF-B1, or c-Rel prior to the addition ofB1 probe. The binding complex con-taining v-Rel and NF-B1 is indicated.
VOL. 76, 2002 REGULATION OF ch-IAP1 BY v-Rel AND c-Rel 11969
on November 8, 2019 by guest
http://jvi.asm.org/
the contribution of other transcription factors to the elevated levels of ch-IAP1 during transformation cannot be ruled out. DNA binding sites for transcription factors induced by the expression of v-Rel (AP-1 and IRF family members) are found in the ch-IAP1 promoter (13, 21).
Previous studies analyzing the transcriptional regulation of v-Rel target genes have provided insight into the importance of the structural differences between v-Rel and c-Rel for trans-formation. Studies of the regulation ofikbademonstrated that the potent C-terminal transactivation domain of c-Rel was responsible for higher levels of ikba in cells overexpressing c-Rel relative to v-Rel, and this likely contributes to the low transformation potential of c-Rel (28). In addition, the char-acterization of the transcriptional activation of c-junby v-Rel revealed the importance of its DNA binding specificity. The altered DNA binding specificity of v-Rel relative to c-Rel al-lows for a higher affinity for theB site in the c-junpromoter and the more-efficient transcriptional activation of c-jun, a key player in transformation by v-Rel (7, 8, 21). We have recently demonstrated that the elevated expression of ch-IAP1 by v-Rel is critical for transformation and that the differential regulation of ch-IAP1 by v-Rel and c-Rel contributes to their different transformation potentials (unpublished data). The results de-scribed in this study are the first to provide evidence that the more-efficient expression of a v-Rel target gene correlates with the ability of v-Rel to escape IB␣regulation. The inability of IB␣to sequester v-Rel in the cytoplasm allows for the con-tinued presence of v-Rel in the nuclei of cells and the eventual nuclear translocation of v-Rel complexes with higher transac-tivation potentials (14, 16). Therefore, the ability of v-Rel to be maintained at high levels in the nuclei of cells compensates for its weak transcriptional activation potential by leading to the gradual and sustained upregulation of target genes responsible for transformation.
ACKNOWLEDGMENTS
Jarmila Kralova and Andrew S. Liss contributed equally to this work.
We thank Mark Hannink for providing c-Rel antisera.
This work was supported by Public Health Service grant CA33192 from the National Cancer Institute. Jarmila Kralova was supported in part by grant 301/98/K042 from the Grant Agency of the Czech Re-public. Cullen Pendleton was supported as an NIH predoctoral fellow on grant T32 CA09583 from the National Cancer Institute.
REFERENCES
1. Ashkenazi, A., and V. M. Dixit.1998. Death receptors: signaling and mod-ulation. Science281:1305–1308.
2. Baba, T. W., B. P. Giroir, and E. H. Humphries.1985. Cell lines derived from avian lymphomas exhibit two distinct phenotypes. Virology144:139–151. 3. Bose, H. R., Jr.1992. The Rel family: models for transcriptional regulation
and oncogenic transformation. Biochim. Biophys. Acta1114:1–17. 4. Davis, N., S. Ghosh, D. L. Simmons, P. Tempst, H. C. Liou, D. Baltimore,
and H. R. Bose, Jr.1991. Rel-associated pp40: an inhibitor of the Rel family of transcription factors. Science253:1268–1271.
5. Diehl, J. A., T. A. McKinsey, and M. Hannink.1993. Differential pp40IB
inhibition of DNA binding byrelproteins. Mol. Cell. Biol.13:1769–1778. 6. Dyer, R. B., and N. K. Herzog.1995. Isolation of intact nuclei for nuclear
extract preparation from a fragile B-lymphocyte cell line. BioTechniques 19:192–195.
7. Fujii, M., T. Minamino, M. Nomura, K. I. Miyamoto, J. Tanaka, and M. Seiki.1996. Selective activation of the proto-oncogene c-junpromoter by the transforming protein v-Rel. Oncogene12:2193–2202.
8. Fujii, M., T. Minamino, M. Nomura, K. I. Miyamoto, J. Tanaka, and M. Seiki.1997. v-Rel activates the proto-oncogene c-Jun promoter: a correla-tion with its transforming activity. Leukemia11:402–404.
9. Gilmore, T. D.1999. Multiple mutations contribute to the oncogenicity of the retroviral oncoprotein v-Rel. Oncogene18:6925–6937.
10. Gilmore, T. D.1999. The Rel/NF-B signal transduction pathway: introduc-tion. Oncogene18:6842–6844.
11. Green, D. R.2000. Apoptotic pathways: paper wraps stone blunts scissors. Cell102:1–4.
12. Hong, S. Y., W. H. Yoon, J. H. Park, S. G. Kang, J. H. Ahn, and T. H. Lee. 2000. Involvement of two NF-B binding elements in tumor necrosis fac-tor␣-, CD40-, and Epstein-Barr virus latent membrane protein 1-mediated induction of the cellular inhibitor of apoptosis protein 2 gene. J. Biol. Chem. 275:18022–18028.
13. Hrdlickova, R., J. Nehyba, and H. R. Bose, Jr.2001. Interferon regulatory factor 4 contributes to transformation of v-Rel-expressing fibroblasts. Mol. Cell. Biol.21:6369–6386.
14. Hrdlickova´, R., J. Nehyba, and H. R. Bose, Jr.1995. Mutations in the DNA-binding and dimerization domains of v-Rel are responsible for altered B DNA-binding complexes in transformed cells. J. Virol.69:3369–3380. 15. Hrdlickova´, R., J. Nehyba, and E. H. Humphries.1994. v-relinduces
expres-sion of three avian immunoregulatory surface receptors more efficiently than c-rel. J. Virol.68:308–319.
16. Hrdlickova´, R., J. Nehyba, A. Roy, E. H. Humphries, and H. R. Bose, Jr. 1995. The relocalization of v-Rel from the nucleus to the cytoplasm coincides with induction of expression ofikbaandnfkb1and stabilization of IB-␣. J. Virol.69:403–413.
17. Kanegae, Y., A. T. Tavares, J. C. I. Belmonte, and I. M. Verma.1998. Role of Rel/NF-B transcription factors during the outgrowth of the vertebrate limb. Nature392:611–614.
18. Karin, M., and Y. Ben-Neriah.2000. Phosphorylation meets ubiquitination: the control of NF-B activity. Annu. Rev. Immunol.18:621–663. 19. Kourembanas, S., R. L. Hannan, and D. V. Faller.1990. Oxygen tension
regulates the expression of the platelet-derived growth factor-B chain gene in human endothelial cells. J. Clin. Investig.86:670–674.
20. Kralova, J., J. D. Schatzle, A. S. Liss, W. Bargman, and H. R. Bose, Jr.1996. Synergistic stimulation of avian IB␣transcription by rel and fos/jun factors. Oncogene12:2595–2604.
21. Kralova, J., A. S. Liss, W. Bargmann, and H. R. Bose, Jr.1998. AP-1 factors play an important role in transformation induced by the v-reloncogene. Mol. Cell. Biol.18:2997–3009.
22. Kralova, J., J. D. Schatzle, W. Bargmann, and H. R. Bose, Jr.1994. Trans-formation of avian fibroblasts overexpressing the c-relproto-oncogene and a variant of c-rellacking forty C-terminal amino acids. J. Virol.68:2073–2083. 23. Morrison, L. E., G. Boehmelt, and P. J. Enrietto.1992. Mutations in the
rel-homology domain alter the biochemical properties of v-reland render it transformation defective in chicken embryo fibroblasts. Oncogene7:1137– 1147.
24. Nehyba, J., R. Hrdlickova´, and H. R. Bose, Jr.1997. Differences inB DNA-binding properties of v-Rel and c-Rel are the result of oncogenic mutations in three distinct functional regions of the Rel protein. Oncogene 14:2881–2897.
25. Nelson, D. M., J. J. Wahlfors, L. Chen, M. Onodera, and R. A. Morgan.1998. Characterization of diverse viral vector preparations, using a simple and rapid whole-virion dot-blot method. Hum. Gene Ther.9:2881–2897. 26. Rinkenberger, J. L., and S. J. Korsmeyer.1997. Errors of homeostasis and
deregulated apoptosis. Curr. Opin. Genet. Dev.7:589–596.
27. Sachdev, S., and M. Hannink.1998. Loss of IB␣-mediated control over nuclear import and DNA binding enables oncogenic activation of c-Rel. Mol. Cell. Biol.18:5445–5456.
28. Schatzle, J. D., J. Kralova, and H. R. Bose, Jr.1995. Avian IB␣is tran-scriptionally induced by c-Rel and v-Rel with different kinetics. J. Virol. 69:5383–5390.
29. Sonenshein, G. E.1997. Rel/NF-B transcription factors and the control of apoptosis. Semin. Cancer Biol.8:113–119.
30. White, D. W., and T. D. Gilmore.1993. Temperature-sensitive transforming mutants of the v-reloncogene. J. Virol.67:6876–6881.
31. White, D. W., A. Roy, and T. D. Gilmore.1995. The v-Rel oncoprotein blocks apoptosis and proteolysis of IB-␣in transformed chicken spleen cells. Oncogene10:857–868.
32. You, M., P. T. Ku, R. Hrdlickova, and H. R. Bose, Jr.1997. ch-IAP1, a member of the inhibitor-of-apoptosis family, is a mediator of the antiapop-totic activity of the v-Rel oncoprotein. Mol. Cell. Biol.17:7328–7341. 33. Young, S. S., P. Liston, J. Y. Xuan, C. McRoberts, C. A. Lefebvre, and R. G.
Korneluk.1999. Genomic organization and physical map of the human inhibitors of apoptosis: HIAP1 and HIAP2. Mamm. Genome10:44–48. 34. Zhang, J. Y., and H. R. Bose, Jr.1989. Acquisition of new proviral copies in
avian lymphoid cells transformed by reticuloendotheliosis virus. J. Virol. 63:1107–1115.
35. Zong, W.-X., M. Farrell, J. Bash, and C. Ge´linas.1997. v-Rel prevents apoptosis in transformed lymphoid cells and blocks TNF␣-induced cell death. Oncogene15:971–980.
36. Zong, W. X., J. Bash, and C. Gelinas.1998. Rel blocks both anti-Fas- and TNF alpha-induced apoptosis and an intact Rel transactivation domain is essential for this effect. Cell Death Differ.5:963–972.
