| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
B Activation1
Frederick Cancer Research and Development Center [R. N., P. T., C. E. C., G. A. H., M. R. Y., N. H. C.], National Cancer Institute and IRSP, Science Applications International Corporation-Frederick, Frederick, Maryland 21702 [T-C. H.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B (NF-B) when transformation is suppressed, suggesting that NF-B may also play a role in neoplastic transformation. We report here that transactivation of NF-B is inducible by tumor promoters in P+ but not in P- JB6 cells. Inhibition of NF-B using a nondegradable mutant of IB
suppressed inducible anchorage-independent transformation of P+ JB6 cells, suggesting that NF-B activation is required for tumor promotion. Induced degradation of IB occurred in both P+ and P- JB6 cells, indicating that failure to activate NF-B in P- JB6 cells cannot be attributed to failure to degrade IB. Slightly higher levels of nuclear p65 were seen in P+ than in P- JB6 cells. The p65-specific DNA binding activity was also higher in P+ cells upon induction by tumor necrosis factor-, suggesting that differential NF-B activation may be attributable to changes in p65 activity. Transactivation of p65 protein was substantially higher in P+ than in P- JB6 cells, as determined by assay of Gal4-p65 fusion constructs. Thus activated, p65 may be a limiting factor for NF-B activation and transformation responses. Stable expression of p65 in P- JB6 cells conferred not only inducible NF-B and AP-1 activation but also transformation response to tumor promoters. Therefore, p65/NF-B appears to be not only necessary for but also sufficient to confer tumor promotion response. Although stable expression of p65 in P- cells produced p65 increases in whole cell extracts, only the transfectants exhibiting increased nuclear p65 showed transformation response. Thus, elevation of nuclear p65 appears to be a necessary step for a transformation response. The P-/p65 transfectants showing acquired transformation response also showed elevated p65-specific transactivation response, thus recapitulating the NF-B phenotypes seen in P+ cells. Expression of a transactivation-deficient mutant of Jun or dominant-negative extracellular signal-regulated kinase suppressed both AP-1 activation and p65-specific transactivation in JB6 cells, suggesting that AP-1 activity is needed for p65 transactivation and consequently for NF-B activation. Thus, the transformation nonresponsive P- JB6 cells owe their resistance to lack of NF-B activation and p65 transactivation that appears in turn to be attributable to insufficient AP-1 activation. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, epidermal growth factor (4
, 10, 11, 12)
, or other tumor promoters (13
, 14) . These clonal variants of JB6 cells have been studied extensively to identify gene expression events that may contribute to their susceptibility (P+) or resistance (P-) to tumor promoter-induced transformation (4
, 14
, 15)
. Clonal variants of JB6 cells showing differential transformation response also show differential response to the activation of AP-1 dependent transcription where inducible AP-1 response was detected in P+ but not P- JB6 cells (4 , 16) . Although expression of a transactivation-deficient mutant of c-Jun (TAM67) suppresses the inducible AP-1 activation and transformation responses in P+ JB6 cells (6) , overexpression of c-Jun, a differentially expressed AP-1 protein (17) , failed to restore AP-1 or 6transformation response in P- JB6 cells (16) . This result suggests that AP-1 proteins are necessary but not sufficient to promote neoplastic transformation. Similar observations are applicable to other models, including transgenic mice and human and mouse keratinocyte models, where TAM67 expression also suppresses or reverses their respective progression phenotypes (18, 19, 20, 21) .
As shown previously, inhibition of NF-B activation in P+ JB6 cells using a relatively nonspecific chemical inhibitor, pyrrolidine dithiocarbamate, was accompanied by inhibition of AP-1 activation when the transformation response was also suppressed (10)
. Inhibition of AP-1 by TAM67 in human keratinocytes further demonstrated the associated inhibition of NF-B during suppression of the transformed phenotype (10)
. Because the reduction of AP-1 and NF-B activities occur concurrently when the activity of either one is suppressed, NF-B may contribute instead of or in addition to AP-1 to neoplastic transformation. Interaction between p65 of NF-B and c-Fos/c-Jun of AP-1 was reported previously (22)
, providing a possible mechanism for coordinate regulation of AP-1 and NF-B signaling in JB6 cells.
NF-B is a transcription factor complex originally identified as binding to an element in the intronic enhancer region of the immunoglobulin light chain gene in mature B- and plasma cells (23)
. NF-B has been implicated in gene regulation related to cell proliferation, apoptosis, adhesion, immune, and inflammatory responses (24)
. Dimers of structurally related NF-B proteins are retained in the cytoplasm when associated with IB family proteins that act as inhibitors of NF-B activation and nuclear localization (25, 26, 27)
. Activation of NF-B by tumor promoters or other stimuli begins with phosphorylation and degradation of IB followed by nuclear localization of NF-B dimers for subsequent transcriptional regulation. Six members of the inhibitory components of NF-B have been identified, whereas IB, the most active inhibitor, (25
, 26)
appears to be the predominant isoform in JB6 cells (data not shown). NF-B family proteins consist of two subgroups distinguished by the presence or absence of a transactivation domain within sequences COOH-terminal to the RHD (24
, 28
, 29) . The subgroup (NFkB) of p50/p105 (NFkB1) and p52/p100 (NFkB2) lacks the COOH-terminal transactivating region and generally does not function as a transcriptional activator. Another subgroup (rel), including relA/p65, relB and c-rel, contains the transactivation domain, and these are therefore operative for transcriptional activation. The formation of dimers is required for the subsequent DNA binding activity. Homodimers of RelA/RelA, p50/p50, and p52/p52 and heterodimers among p65, c-Rel, RelB, p50, and p52 have been identified.
Previous studies suggested a requirement of NF-B for maintenance of tumor phenotype or for inhibition of apoptosis (27
, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40)
. Inhibition of NF-B by expression of IBmut (a non-degradable mutant of IB) or antisense RNA or by gene knockout results in tumor regression (25
, 32
, 38 , 39)
. However, none of these reports or others (41)
showed that any NF-B component is sufficient to confer inducible transformation response. We report here that tumor promoters induce p65 transactivation and NF-B activation in transformation susceptible (P+) but not in resistant (P-) JB6 cells. Inhibition of NF-B by IBmut suppresses transformation susceptibility in P+ cells. P+ cells show elevated levels of nuclear p65 protein, p65-specific DNA binding activity, and p65-specific transactivation. Stable expression of relA in P- JB6 cells not only enhances their tumor promoter-induced NF-B activity but also confers transformation response. These observations suggest that activated p65 is a limiting factor for the NF-B response in JB6 cells, and that p65/NF-B plays a causal role in promoting neoplastic transformation. p65 may thus constitute a suitable molecular target for cancer prevention.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(data not shown). The transformation phenotypes of JB6 cells were determined every 3 weeks by their induced anchorage-independent growth on soft agar as described later. All other cell culture reagents were purchased from BioWhitaker or Life Technologies, Inc.
Plasmids.
AP-1 or NF-B luciferase reporter plasmids consisting of luciferase reporter gene driven by the promoter harboring the appropriate element were constructed as described previously (6
, 10)
. The AP-1 reporter plasmids consist of firefly luciferase genes driven by an AP-1 responsive promoter containing four copies of flanked AP-1 consensus sequence (TCGACTATGATGAGTCATGGGGC) from GCN4 and a minimal albumin promoter region with TATA box (AAGCTTAGAATCTAGTATATTAGAGCGAGTCTTTCTGCACACAGATCACCTTTCCTATCAACCCCACTACCATACCCTTCCTCCATCTATACCACCCTACTCTGCAGGTCGAC). The NF-B reporter plasmids consist of firefly luciferase reporter genes driven by a minimal NF-B responsive region from interleukin-6 promoter containing two copies of NF-B responsive elements in a sense orientation (GACTCTAGAGGATCAAATGTGGGATTTTCCCATGTGGGATTTTCACATGATCATGGGAAAATCCCACATGAAAATCCAATTTCCGGCC). Because there are no other known responsive cis elements identified in the above sequences, any cross-family activation should occur at the level of protein-protein interaction. The human ODC-luciferase reporter plasmid, a kind gift from Dr. Ajit Verma, University of Wisconsin (7)
, consists of firefly luciferase genes driven by a minimal promoter region from human ODC gene. The expression vectors including mouse IBmut (pMEIKB67CJ), human p65/relA and mouse p65dC (p65 lacking the transactivation domain) are generous gifts from Dr. Nancy Rice (National Cancer Institute, Frederick, MD). The ptkRL (Promega Corp., Madison, WI) is a Renilla luciferase reporter driven by a tk promoter used here as control for transfection efficiency. The luciferase activity shown is firefly luciferase activity normalized to the respective Renilla luciferase activity. The Gal4-p65 fusion constructs (p65-
DBD and p65-TA+) are generous gifts from Dr. M. L. Schmitz (German Cancer Research Center, Heidelberg, Germany; Ref. 42
).
Reporter Transfection and Luciferase Assay.
Cell transfections were performed according to the LipofectAMINE protocol from Life Technologies, Inc. In brief, 0.2–0.5 µg of reporter plasmids were mixed with 2 µl of LipofactAMINE solution in 50 µl of reduced medium Opti-MEM (Life Technologies, Inc.), followed by an incubation with 5 x 104 cells for 24-well plates in Opti-MEM. A minimum of 5 h incubation at 37°C was needed for transient transfection, followed by 12–18 h incubation in EMEM with 4% FBS for recovery. The transfected cells were starved and also synchronized in EMEM with 0.2% FBS for >24 h. The resulting quiescent cells were induced for growth using serum (EMEM with 4% FBS), 10 ng/ml TPA, or 10 ng/ml TNF- in EMEM with 0.2% FBS as described in the Fig. legends. The stimulated cells were collected and lysed at 3 or 18 h after induction. The resulting cell lysates were assayed for luciferase activity using Dual-Luciferase Assay kit (Promega Corp.) and DYNEX Luminometer (DYNEX Technologies, Chantilly, VA). Each firefly luciferase activity driven by a specific promoter was normalized to its respective Renilla luciferase activity driven by tk promoter as a control for transfection efficiency.
Stable Transfection and Transfectants.
Stable transfections were carried out in a similar fashion as transient transfection, except that a 24- to 48-h transfection period was used. G418 selection was performed on the transfectant population using 500 µg/ml G418 for Cl 41 and 200 µg/ml for Cl 30-7b to select stable transfectants. Twenty clonal transfectants of Cl 41 P+ JB6 cells stably overexpressing human IBmut were obtained and denoted 41IkB1-1 to 1-8, 41IkB2-1 to 2-5 and 41IkB3-1 to 3-7. All of the tested P+/IBmut stable transfectants showed resistance to tumor promoter-induced transformation and exhibited suppressed NF-B activation while no apparent cell death was observed. The expression levels of IBmut in the stable transfectants range from 2- to 5-fold greater than the endogenous IB levels in P- JB6 cells (data not shown). Clonal transfectants of Cl 30-7b P- JB6 cells stably overexpressing human p65 were also obtained (Figs. 3
and 8
). The clonal transfectants including 30p65S11, 30p65S13, 30p65S32 and 30p65S34 acquired inducible transformation responses, whereas 30p65S14 was inert to tumor promoters. The level of p65 expression was slightly lower in S14 than in the other four stable transfectants. However, all of the five p65/P- transfectants showed elevated p65 expression relative to the vector control stable transfectant or parental Cl 30-7b cells (data not shown).
|
|
Western Blots.
Western immunoblotting was carried out according to the ECL protocol from Amersham (Amersham Co., Arlington Height, IL). The antibodies used here were either generous gifts from Dr. Nancy Rice (National Cancer Institute-Frederick) or purchased from Santa Cruz Biotechnology (Santa Cruz, California) and Sigma Chemical Co. (San Louis, MO). In brief, 10 µg of nuclear extracts or 20–40 µg of whole cell lysates were boiled and denatured in sample buffer containing SDS and DTT (Novex, San Diego, CA), followed by gel electrophoresis using NuPage 10% Bis-Tris prepacked gel (Novex) in MES (50 mM MES at pH 7.2, 50 mM Tris, 0.1% SDS, and 1 mM EDTA) buffer. The proteins were electrotransferred to nitrocellulose membrane (Schleicher & Schuell, Keene, NH) using a semidry transfer blotting system from Enprotech Co. (Hyde Park, MA). The resulting protein-bound membrane was blotted with selected antibodies as described previously (43)
and visualized using ECL reagents (Amersham). The band intensities were monitored by Kodak digital camera (DC120) and analyzed by its image-analyzing program (Kodak 1D). The protein levels shown in Figs. 3
and 8
were normalized at the sample loading step where an identical amount of total protein was loaded into each sample lane. The cells for all samples were cultured and induced at the same time, and the protein extracts were also prepared at the same time. Protein concentration was determined by Coomassie Blue Protein Assay Reagent from Pierce Corp. (Rockford, IL).
EMSA.
Oligonucleotides containing NF-B consensus sequence (AGTTGAGGGGACTTTCCCAGG) were purchased from Santa Cruz or synthesized by Life Technologies, Inc., followed by annealing. Fifty ng of double-stranded oligos were end-labeled using [32P]ATP and T4 polynucleotide kinase (Roche, Mannheim, Germany). The reaction mixture for EMSA consists of 50,000 cpm of labeled and purified oligonucleotides, 1 µg/ml of poly(deoxyinosinic-deoxycytidylic acid) (Roche), and 1–2 µg of nuclear extracts in binding buffer containing 20 mM of Tris-HCl at pH 7.5, 50 mM KCl, 1 µM DTT, 2 µM EDTA, and 4% glycerol. The reaction was carried out at room temperature for 30 min. The protein-DNA complexes were resolved on a 6% TBE (89 mM Tris base, 89 mM boric acid, 2 mM EDTA, pH 8.3)/Gel Retardation gel (Novex) and visualized by autoradiograph. Antibody supershifts were performed by mixing the selected antibody with nuclear extracts for a 15-min incubation at 4°C, followed by the addition of labeled nucleotides for another 15-min incubation at room temperature. The autoradiograph was monitored by Kodak digital camera (DC120) and analyzed by its image-analyzing program (Kodak 1D). The specificity of the DNA-protein interaction was determined by including unlabeled oligonucleotide in >500-fold excess, where addition of unlabeled oligonucleotide abolished the protein interaction with labeled oligonucleotide.
Anchorage-independent Transformation Assay (Soft Agar Assay).
Transformation assays were performed as described previously (8
, 10)
. In a 60-mm tissue culture dish, 10,000 JB6 cells were resuspended in 1.5 ml of 0.33% agar in EMEM with 10% FBS and layered over 7 ml of 0.5% agar in EMEM with 10% FBS. Both layers of agar were supplemented with DMSO (control), 10 ng/ml TPA, or 10 ng/ml TNF- for tumor promotion. The cells were cultured at 36°C for 14 days, and the resulting colonies, if any, were counted by an automated image analysis system supported by Image Pro-Plus (version 3.0.1) software (Media Cybernetics). The transformation responses are presented as colonies per 10,000 cells per 60-mm tissue culture dish. Neither Cl 41 nor Cl 30-7b cells showed any significant reduction of total viable cell numbers during the time periods of soft agar assay, indicating that the differential response in P+ and P- JB6 cells is not a result of cell death or apoptosis.
Gal4 Transactivation Assay.
The Gal4 expression plasmids consist of sequences from the transactivation domain of p65 (Fig. 5A)
fused to the DNA binding domain of Gal4. The selected fusion construct (50 ng) was cotransfected into JB6 cells with 25 ng of luciferase reporter construct driven by the Gal4-responsive promoter sequence. The transfected cells were measured for luciferase activity. In selected experiments, the cells were also cotransfected with 0.5 µg of a protein expression vector to determine the effect of this protein on p65 transactivation. Transfection efficiency was normalized using Renilla luciferase driven by tk promoters as described in the previous section.
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B. or epidermal growth factor; Refs. 4
, 10
, and 17
). Because inhibition of AP-1 activation by TAM67 in JB6 cells, human or mouse keratinocytes is invariably accompanied by suppression of NF-B activity (5
, 44)
, we examined the role of NF-B in transformation of JB6 cells. As shown in Fig. 1A
, clonal variants of JB6 cells differentially respond to activation of NF-B as determined by IL6-NF-B reporter. There is no other known responsive cis element sequence identified in this NF-B reporter (see "Materials and Methods"). The Cl 41 (P+) JB6 cells showed elevated basal and promoter-induced NF-B activation, whereas the Cl 30-7b (P-) JB6 clones exhibited low basal NF-B activity and were inert to induction by tumor promoters. Other P+ and P- variants of JB6 cells also exhibited similar differential responses with Cl 22 (P+) showing inducible NF-B response and Cl 30-2a (P-) being inert to NF-B activation (data not shown). These observations indicate that differential activation of NF-B (Fig. 1A)
in addition to AP-1 (Fig. 1B
; Refs. 4
and 21
) is associated with the differential transformation response in JB6 clonal variants.
|
B Is Required for Tumor Promoter-induced Transformation.
(Lanes 2 and 4), expression of IBmut inhibits TPA- or TNF-induced activation of NF-B in P+ cells and also reduces the basal NF-B activity in P- JB6 cells. Expression of IBmut, however, neither alters the transcriptional promoter activity of human ODC nor impairs cell viability (data not shown), suggesting that expression of IBmut does not inhibit general transcription machinery or confer cell death on JB6 cells but specifically targets NF-B activation. The AP-1 activity was substantially suppressed by the NF-B-specific inhibitor IBmut (Fig. 1B
, Lanes 2 and 4). This suggests a potential cross-talk between NF-B and AP-1 signaling pathways because the inhibition of AP-1 by IBmut would need to be indirectly mediated through NF-B components.
Because an elevated NF-B response appears to be associated with a tumor promotion-susceptible phenotype, we examined whether NF-B activation is required for the transformation response. To determine whether inhibition of NF-B would suppress the transformation response in JB6 cells, we generated P+ JB6 cells stably expressing IBmut. As shown in Fig. 2
, stable expression of IBmut in three independent lines of P+ JB6 cells reduces NF-B and AP-1 activity (Fig. 2, A and B)
and results in suppression of inducible transformation responses (Fig. 2, C and D)
. Thus, NF-B activation, along with AP-1 activation, is apparently required for tumor promoter-induced neoplastic transformation. The P+ JB6 cell transfectant (M3C) stably expressing a transactivation-deficient mutant of c-Jun, TAM67, also showed suppression of its NF-B activation and transformation responses. The observations made with IBmut and TAM67-expressing cells suggest that activation of NF-B and of AP-1 are mutually necessary for each other’s full activation. Furthermore, neither NF-B nor AP-1 activation can be excluded as required for induced neoplastic transformation.
|
B in P+ and P- JB6 Cells Is Not Attributable to Differential IB Degradation.B begins with the induced phosphorylation and the subsequent degradation of cellular IB, followed by nuclear translocation of NF-B proteins leading to target gene activation. We first examined the levels of cellular IB in P+ and P- JB6 cells after induction by TPA or TNF-. TPA treatment did not induce IB degradation in either P+ or P- cells within the time course conducted (data not shown). TNF-, on the other hand, induced the degradation of IB with a similar time course in P+ and P- JB6 cells, where endogenous IB was completely abolished at 15 min after induction and resynthesized at 45 min after induction (Fig. 3A
, Lanes 2–4, respectively). It is therefore apparent that induced degradation of IB is not the limiting factor responsible for differential activation of NF-B in P+ and P- JB6 cells. Although the total IB protein appears to be higher in P+ than P- JB6 cells (Fig. 3A
, Lanes 1), paralleling the elevated nuclear p65 in P+ cells (Fig. 3B)
, the tumor promoter-induced degradation of IB and its dissociation from NF-B are not differential between the two cell types.
Increased Levels of NF-B p65 and p52, but not p50, in the Nuclei May Contribute to Elevated NF-B Response.
The similar time course and pattern for induced degradation of IB in P+ and P- JB6 cells suggest that failure to dissociate NF-B from IB upon induction is not the explanation for the lack of inducible activation of NF-B in P- JB6 cells. We next examined the levels of NF-B proteins in the nuclei to determine whether nuclear concentration could be limiting. As demonstrated in Fig. 3B
and further confirmed in Fig. 8B
, TNF- induces nuclear localization of NF-B proteins at 3 or 18 h after addition of TNF-, whereas TPA only minimally induces this nuclear translocation activity. Slightly (1.5–2-fold) but consistently elevated levels of p65 and p52 are detected in the nuclei of TNF--induced P+ JB6 cells as compared with their respective induced levels in P- cells (Fig. 3B
, Lanes 2 and 7). The nuclear p50, on the other hand, remains at similar levels in P+ and P- JB6 cells. This similar level of p50 in P+ and P- cells may serve as an internal loading control for the differential levels of p65 in these two cell types. Neither c-Rel nor RelB was detected (data not shown), suggesting that p65/RelA is the only transcriptionally active NF-B factor in JB6 cells. The elevated level of nuclear p65 was paralleled by elevated p65 in whole cell extracts (Fig. 8A
, Lanes 1 and 2) of P+ relative to P- JB6 cells after addition of TNF-. Corresponding increases of IB expression were also observed in P+ cells, a result that is expected because expression of IB is regulated by p65/NF-B. Taken together these observations suggest that small elevations of nuclear p65 may contribute to elevated NF-B responses in transformation susceptible JB6 cells.
Elevated NF-B DNA Binding Activity and Increased Levels of p65-containing DNA Binding Complexes Were Detected in P+ Relative to P- JB6 Cells.
Because increased nuclear localization of transcription factors frequently associates with enhanced transactivation and because DNA binding precedes the interaction with general transcription machinery, we next examined the NF-B DNA binding activity in P+ and P- JB6 cells. As shown in Fig. 4
, higher NF-B DNA binding activity was observed in P+ (Lanes 1) than in P- (Lanes 2) JB6 cells. TNF- significantly enhanced the DNA binding activity of NF-B (Fig. 4C
, Lanes 1 for P+ and Lanes 2 for P-), whereas TPA produced a minimal increase in NF-B DNA binding (Fig. 4B
, Lanes 1 and 2 for P+ and P-, respectively). Bands 1 of P+ (Fig. 4
, Lanes 1) exhibited higher intensity than the corresponding bands 1 of P- cells (Lanes 2). The complex in band 1 consists of dimers containing p65 because antibody specific for p65 was able to up-shift the entire band 1 (Fig. 4
, Lanes 3 and 4) to form band 3.
|
B DNA binding complexes (bands 1 and 2) was analyzed by EMSA and antibody supershifts as demonstrated in Fig. 4
. The slow-migrating complex (band 1) appears to consist predominantly of dimers containing p65 coupling with p52 or p50, because either p65-specific antibodies alone or the combination of p50- and p52-specific antibodies was able to up-shift the entire band 1 to form the band 3 complexes (Lanes 3, 4, 9, and 10). The fast-migrating complex (band 2) is composed predominantly of dimers of p50 or p52 (band 4 in Lanes 9 and 10) because the combination of p50- and p52-specific antibodies up-shifted most of the bands 2 to form bands 4. The higher intensity of the supershifted complex in band 3 of Lane 3 (P+) than the same complex in band 3 of Lane 4 (P-) is also consistent with the observation that P+ JB6 cells have elevated levels of nuclear p65.
Among the three NF-B proteins studied here, p65 contains a functional transactivation domain, whereas p50 and p52 lack this domain (30
, 42)
. Hence, the p65-containing dimers appear to be the active complexes responsible for NF-B transcriptional activation, and the dimers of p50 or p52 may be transcriptionally inert or inhibitory. Either an increase in the active p65-containing complexes (slow-migrating band 1) or a reduction in the inhibitory p50/p52 dimers (fast-migrating band 2) might contribute to the enhanced NF-B response in P+ JB6 cells. Thus, an increase in p65-specific DNA binding activity may contribute to the enhanced NF-B response and the subsequent transformation response seen in P+ cells.
Elevated p65-specific Transactivational Activity in Transformation-responsive P+ JB6 Cells.
The modestly elevated levels of nuclear p65 and p65-specific DNA binding activity appear to contribute but not to suffice to account for the magnitude of elevated NF-B activation observed in P+ JB6 cells. We thus examined the activation of the p65 protein. This used a Gal4 transactivation assay that allows the assessment of the ability of the cells to activate the transactivation domain of a transcription factor without interference from its DNA binding domain. Fragments of p65 protein corresponding to the various regions of its transactivation domain but not to the DNA binding domain (Fig. 5A)
were fused to a DNA binding domain from Gal4 (42)
. These fusion constructs were cotransfected with a luciferase reporter driven by the minimal Gal4-responsive promoter sequence. As shown in Fig. 5B
, significantly elevated basal and TNF--inducible luciferase activities were observed in P+ JB6 cells when using either the p65-DBD (p65 lacking the DNA binding domain) or the p65-AD (containing only TA1 and TA2 of p65; Fig. 5A
). In addition, p65 transactivation is inducible by TNF- in the P+ but not the P- JB6 cells, paralleling the inducible nature of NF-B response and transformation response detected only in P+ JB6 cells. These results suggest that p65 transactivation is a limiting factor in P- JB6 cells that are deficient in NF-B activation. Thus, the elevated NF-B activity in P+ JB6 cells appears to be attributable to the increased nuclear localization (1.5–2-fold increase as shown in Fig. 3
), the enhanced p65-specific DNA-binding activity (3–4-fold increase as shown in Fig. 4
), and the strikingly elevated p65 transactivation (at least 10-fold increase, as shown Fig. 5
), where the activated p65 appears to be the limiting factor.
Ectopic Expression of p65 (relA) Confers NF-B Activation and Transformation Responses on P- JB6 Cells.
Because the deficiency in p65 protein and its activation appears to contribute to the lack of NF-B activation and transformation responses in P- JB6 cells, we examined the effect of overexpressed p65/relA in P- JB6 cells on NF-B activation and tumor promotion. If p65 levels are limiting for NF-B activation and transformation responses in P- JB6 cells, the expression of p65/NF-B should amend the deficiency and enable transformation in P- JB6 cells. Transient expression of p65/relA increased the basal and promoter-induced NF-B and AP-1 activities in P- JB6 cells (Fig. 6, A and C
, Lanes 2 and 3). Expression of transactivation-deficient mutant of p65 (p65dC) and p50 did not enhance either NF-B or AP-1 activation (Fig. 6, A and C
, Lanes 4, and data not shown) showing the specificity of the wild-type p65.
|
B activity in JB6 cells, we then generated stable transfectants of P- JB6 cells overexpressing p65. Stable expression of p65 not only increased basal NF-B activity but also enhanced the promoter-inducible NF-B responses in P- JB6 cells (Fig. 6B
, Lanes 2 and 3 for S34 and S32, respectively). The AP-1 responses were coordinately enhanced in P-/p65 stable transfectants (Fig. 6D
, Lanes 2 and 3). The P-/p65 stable transfectants showing inducible NF-B response developed significant transformation response (Fig. 7, D–F
, and Table 1
). The transfectants not inducible by tumor promoters for NF-B activation (S14; Fig. 6B
, Lanes 4) were also transformation nonresponsive (data not shown). This acquisition of inducible NF-B response appears to be essential for inducible transformation response. The P- JB6 stable transfectants harboring control vectors (30p65EV2; Fig. 7, A–C
, and Table 1
) or p65dC (data not shown) did not develop transformed colonies, indicating the specificity of wild-type p65 in causing transformation. Thus, p65 not only is required but appears to be limiting for NF-B activation and neoplastic transformation in JB6 cells.
|
|
(Lanes 3–5). Only moderate increases of p65 expression were detected in P-/p65 stable transfectants, whereas the expression of endogenous IB was coordinately increased by the overexpression of p65 with the level of p50 expression unaltered (Fig. 8A
, Lanes 3–5 as compared with Lanes 1). The increased levels of p65 expression in P-/p65 stable transfectants match the levels of p65 expression in P+ JB6 cells, supporting the notion illustrated in Fig. 3B
that a slight increase of p65 can contribute to significant functional outcomes. The levels of nuclear p65 were coordinately increased in the clonal transfectants showing transformation responses (Fig. 8B
, Lanes 2, 3, and 5) as compared with transformation-inert clonal variants (Fig. 8B
, Lanes 1 and 4). The P-/p65 clonal transfectants, S14, did not show transformation response, despite their increased cellular expression of p65 (Fig. 8A
, Lanes 4). In this case, the lack of nuclear p65 appears to limit its transformation response (Fig. 8B
, Lanes 4). Because the other components of NF-B, p50 and p52, do not change in accordance with their transformation response, these components may not limit tumor promotion. In addition, stable transfectants expressing p50 did not acquire a transformation response, also supporting the concept that p50 concentration was not limiting for neoplastic transformation (data not shown). The levels of nuclear p65 do vary in parallel with the transformation response of p65-expressing clones, suggesting that levels of nuclear p65 may contribute to regulating neoplastic transformation. Transactivation of p65, as determined using the activation domain (TA1) of p65 (42)
, was greatly elevated and became inducible by TNF- in P- cells stably expressing p65 (Fig. 8C
, Lanes 3; P-/p65) as compared with the parental or the vector control stable P- cells (P-/EC; Fig. 8C
, Lanes 2 and 4, respectively). This result suggests that overexpression of p65 does increase activated p65 in the nucleus and further argues that activated p65 appears to be a limiting factor for transformation response.
Transactivation of p65 Requires Activation of the AP-1 Transcription Factor.
Transactivation of p65 protein appears to be limiting for NF-B activation in P- JB6 cells. As shown in Fig. 9
(Lanes 2) using the Gal4 assay, transactivation of p65 in P+ JB6 cells was abolished by expressing the dominant-negative inhibitor of AP-1 (TAM67), suggesting that transactivation of p65 and consequently activation of NF-B requires AP-1-dependent transcriptional events. Dominant-negative Erk2 significantly suppressed p65-specific transactivation (Fig. 9
, Lane 3) suggesting that NF-B activation may also depend on Erk activity. Erk modulation, however, constitutes an independent mode of regulating AP-1 activity because expression of dominant-negative Erk (DN-Erk) inhibits, and expression of wild-type Erk (WT-Erk) enhances, AP-1 activity (8
, 44)
. Transient expression of WT-Erk2 elevated p65-specific transactivation by 20–40-fold in P+ JB6 cells and 5–7-fold in P- JB6 cells (Fig. 9, A and B
, Lanes 4). This suggests that Erk and hence AP-1 activation is limiting for NF-B activation in P- JB6 cells. Overexpression of p65 also enhanced p65-specific transactivation in P- JB6 cells (7–8-fold) but not in P+ cells (1–2-fold; Fig. 9
, Lanes 5). This result further supports that p65 is a limiting factor only in P- JB6 cells and that overexpression of p65 can facilitate its transactivation. relA/p65 expression also enhanced AP-1 activation (Fig. 6, C and D)
. Thus, increasing the concentration of p65 protein appears to activate the machinery for transactivating p65 through transcriptional events that require AP-1.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B protein p65/relA appears to explain the NF-B and transformation nonresponsiveness of JB6 P- cells. Overexpression of p65/relA enhances p65 transactivation and NF-B activation and confers inducible transformation response on the transformation-inert JB6 cell variants. These observations provide the first evidence suggesting that p65 of NF-B plays a causal role in driving the process of neoplastic transformation. Transactivation of p65 appears to be a limiting factor for activation of NF-B, and NF-B activation is required for neoplastic transformation of JB6 cells. The observed causal effect in tumor promotion of p65/RelA and its activated form suggests a molecular target for cancer prevention.
Inhibition of NF-B by antisense RNA was shown to reverse tumor phenotype in fibroblasts and other cells (37
, 45)
. Studies with p65 null fibroblasts demonstrated the requirement of p65 for H-ras-induced transformation (32)
. However, up-regulation of NF-B activity or proteins was insufficient to convert normal fibroblasts directly to tumor phenotype. The JB6 cell model is valuable for studying gene expression events specifically critical to the rate-limiting tumor promotion stage of carcinogenesis. Among these are induction of ODC (7)
, AP-1 transactivation (6
, 43
, 44)
, and NF-B transactivation. Similar to observations using other cell culture models (28
, 33)
, stable overexpression of p65 did not directly confer tumor phenotype on those JB6 cell variants having a transformation-inert phenotype. These P-/p65 transfectants became transformed only when subsequently induced by tumor promoters. Although induced p65 transactivation and subsequent NF-B activation appear to be necessary for tumor promotion, it is yet unknown whether elevated NF-B/p65 activity is also required for the subsequent tumor progression stage of carcinogenesis.
Our early attempts to establish P- JB6 cells stably expressing high levels of p65/relA failed. These p65/P- transfectants showed cell death or growth retardation within 4 weeks after transfection (data not shown), possibly because of induction of growth-inhibitory cytokines, as demonstrated in transgenic models (46)
. Additional stable transfectants of p65 were subsequently obtained by transfecting smaller amounts of p65 expression constructs to reduce the level of p65 overexpression. Western analysis confirmed the modest increase of p65 expression in our p65/P- stable transfectants. Additionally, only the transfectants with TPA- or TNF--inducible NF-B responses exhibited a transformation response. The P-/p65 stable transfectants lacking the inducible NF-B response retained their transformation-inert phenotype despite the slightly increased expression of p65. Thus, inducible NF-B as well as inducible AP-1 responses appear to be necessary for tumor promoter-induced transformation response. These observations support the conclusion that induced NF-B activation is required and possibly sufficient for tumor promoter-induced neoplastic transformation of JB6 cells.
Because inhibition of either AP-1 or NF-B in JB6 cells by specific gene inhibitors results in repression of both activities and of subsequent transformation, other approaches are needed to determine whether inhibition of either suffices to prevent transformation. Whether communication between AP-1 and NF-B is the result of a physical interaction between p65/NF-B and AP-1 proteins as described by Stein et al. (22)
needs further determination. These authors suggested a physical interaction between the RHD of p65 and the b-zip domain of c-Fos/c-Jun (22)
, resulting in activation of both factors. It is not clear whether the physical interaction between the two families of transcription factors plays a role in tumor promotion. However, we have found that TAM67, an NH2-terminally truncated c-Jun mutant, interacts with p65 in a human keratinocyte model when elevated AP-1 activity, NF-B activity, and tumor phenotype were suppressed (40)
. The mechanism by which IBmut suppressed AP-1 activity is not known. Because IB has no structural basis for interaction with any of the AP-1 family proteins, the inhibition of AP-1 by IB may be mediated through NF-B family proteins. The interaction between p65 and AP-1 proteins was further suggested by the observation that stable overexpression of p65 restored not only the NF-B response but also the AP-1 response in P- JB cells.
Exogenous expression of p65 in P- JB6 cells appears to induce p65-specific transactivation and consequently to reconstitute NF-B activation and the transformation-responsive phenotype. The details of the mechanism by which increased p65 expression leads to inducible NF-B response remain to be determined, but increasing the level of p65 protein appears to produce a parallel increase in p65-specific transactivation (Fig. 9)
. This p65 transactivation is accompanied by induced AP-1 activation that appears to be required for the p65 transactivation. How can one reconcile the observations that expression of either Erk 2 or p65 appears to be sufficient to restore NF-B activation in P- cells? Erk 2 expression in P- cells sufficed to restore the AP-1 response and the transformation response (8
, 44)
. AP-1 activation is required for p65 activation. P65 expression can activate AP-1, possibly through the reported interactions involving AP-1 b-zip and p65 RHDs (22)
. Thus, either Erk 2 or p65 can activate the AP-1 that appears to be necessary for p65 activation. Erk has been reported to activate p65 by a pathway requiring coactivator p300/CBP (47)
. Although Erk appears to be an effector for p65 activation in JB6 cells, Erk does not appear to regulate p65 transactivation by directly altering p65 phosphorylation.4
Nor is the concentration of p300 limiting for NF-B activation in JB6 cells.4
Whether other coactivators are needed for NF-B activation in this model will be of interest to determine.
In summary, NF-B activation appears to be essential and causal for neoplastic transformation of JB6 cells. The elevated levels of nuclear p65 (1.5–2-fold increase) and the enhanced p65-specific DNA binding activity (3–4-fold) contributed to the elevated NF-B response and the transformation-sensitive phenotype in P+ JB6 cells. However, the most significant determinant appears to be the greatly elevated level of p65-specific transactivation (10–20-fold) in P+ cells relative to P- cells. Because transactivation of p65 appears to depend on activation of AP-1, NF-B activation may function downstream of AP-1 activation that is directly regulated by Erk. Supplementation of either Erk or NF-B component p65 suffices to confer p65-specific transactivation (Fig. 9)
and NF-B response on transformation-resistant cells and to produce a tumor phenotype (8
, 44)
upon induction by tumor promoter.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work is funded in part by Contract No. NO1-CO-56000 from NCI, NIH to SAIC-Frederick. 
2 To whom requests for reprints should be addressed, at Gene Regulation Section, National Cancer Institute, Building 560, Room 21-31, Frederick, MD 21702. Phone: (301) 846-1342; Fax: (301) 846-6143; E-mail: colburn{at}ncifcrf.gov
3 The abbreviations used are: TPA, 12-O-tetradecanoylphorbol-13-acetate; TNF, tumor necrosis factor; AP-1, activator protein-1; NF-B, nuclear factor B; EMEM, Eagle’s minimum essential medium; FBS, fetal bovine serum; tk, thymidine kinase; Erk, extracellular signal-regulated kinase; ODC, ornithine decarboxylase; RHD, rel homology domain; EMSA, electrophoretic mobility shift analysis.
4 T-C. Hsu, unpublished observations.
Received 11/ 7/00. Accepted 3/19/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and activator protein-1 activation block the neoplastic transformation response. Cancer Res., 57: 3569-3576, 1997.B (NF-B)-dependent transcriptional events in carcinogenesis. FORUM Free Radical Biol. Medicine, 28: 1338-1348, 2000.
B transactivation and suppresses anchorage independent growth of HPV immortalized human keratinocytes. Oncogene, 16: 2711-2721, 1998.[Medline]
B p65 and Fos/Jun transcription factors produces potentiated biological function. EMBO J., 12: 3879-3891, 1993.[Medline]
B and IB proteins: an overview. Semin. Cancer Biol., 8: 63-73, 1997.[Medline]
B masks the nuclear localization signal(s) of p50 and c-Rel homodimers. Mol. Cell. Biol., 18: 2640-2649, 1998.B proteins control NF-B activity. Proc. Natl. Acad. Sci. USA, 96: 49-54, 1999.B/IB family: intimate tales of association and dissociation. Genes Dev., 9: 2723-2735, 1995.B. Science (Wash. DC), 256: 367-370, 1992.B p65 transcriptional activator and an alternatively spliced derivative. Mol. Cell. Biol., 12: 444-454, 1992.B and IB proteins: new discoveries and insights. Annu. Rev. Immunol., 14: 649-683, 1996.[Medline]
B regulation by antisense RNA expression leads to malignant transformation. Oncogene, 9: 3189-3197, 1994.[Medline]
B transcriptional activity, which is required for cellular transformation. J. Biol. Chem., 272: 24113-24116, 1997.B/IB proteins and cancer. Oncogene, 13: 1367-1378, 1996.[Medline]
B blocks tumorigenicity and causes tumor regression. Proc. Natl. Acad. Sci. USA, 90: 9901-9905, 1993.B activation prevents cell death. Cell, 87: 565-576, 1996.[Medline]
B and IB factors in oncogenesis. Seminar Cancer Biol., 8: 103-111, 1997.
B/Rel expression and the pathogenesis of breast cancer. J. Clin. Investig., 100: 2952-2960, 1997.[Medline]
-induced apoptosis by NF-B[see comments]. Science (Wash. DC), 274: 787-789, 1996.B/IB family: intimate tales of association and dissociation. Genes Dev., 9: 2723-2735, 1997.
B and AP-1 target genes and suppresses tumor phenotype in human keratinocytes. Mol. Carcinog., 29: 159-169, 2000.[Medline]
B subunit p65 to transform fibroblasts. Oncogene, 9: 2391-2398, 1994.[Medline]
B. J. Biol. Chem., 270: 15576-15584, 1995.B transcription factor in oncogenesis. Anticancer Res., 16: 589-596, 1996.[Medline]
B function in transgenic epithelial tissue demonstrate a growth inhibitory role for NF-B. Proc. Natl. Acad. Sci. USA, 95: 2307-2312, 1998.B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell, 1: 661-671, 1998.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. Golovine, P. Makhov, R. G. Uzzo, T. Shaw, D. Kunkle, and V. M. Kolenko Overexpression of the Zinc Uptake Transporter hZIP1 Inhibits Nuclear Factor-{kappa}B and Reduces the Malignant Potential of Prostate Cancer Cells In vitro and In vivo Clin. Cancer Res., September 1, 2008; 14(17): 5376 - 5384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-S. Piao, Y.-C. Du, H. Oshima, J.-C. Jin, M. Nomura, T. Yoshimoto, and M. Oshima Platelet-type 12-lipoxygenase accelerates tumor promotion of mouse epidermal cells through enhancement of cloning efficiency Carcinogenesis, February 1, 2008; 29(2): 440 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Yang, B. J. Misner, R. J. Chiu, and F. L. Meyskens Jr Redox effector factor-1, combined with reactive oxygen species, plays an important role in the transformation of JB6 cells Carcinogenesis, November 1, 2007; 28(11): 2382 - 2390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Gray and J. J. Pestka Transcriptional Regulation of Deoxynivalenol-Induced IL-8 Expression in Human Monocytes Toxicol. Sci., October 1, 2007; 99(2): 502 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. P. Singh, S. Dhanalakshmi, S. Mohan, C. Agarwal, and R. Agarwal Silibinin inhibits UVB- and epidermal growth factor-induced mitogenic and cell survival signaling involving activator protein-1 and nuclear factor-{kappa}B in mouse epidermal JB6 cells Mol. Cancer Ther., May 1, 2006; 5(5): 1145 - 1153. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Feng, Y. Ohmori, and P.-L. Chang Production of chemokine CXCL1/KC by okadaic acid through the nuclear factor-{kappa}B pathway Carcinogenesis, January 1, 2006; 27(1): 43 - 52. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hu and N. H. Colburn Histone Deacetylase Inhibition Down-Regulates Cyclin D1 Transcription by Inhibiting Nuclear Factor-{kappa}B/p65 DNA Binding Mol. Cancer Res., February 1, 2005; 3(2): 100 - 109. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hu, H. Nakano, H. Sakurai, and N. H. Colburn Insufficient p65 phosphorylation at S536 specifically contributes to the lack of NF-{kappa}B activation and transformation in resistant JB6 cells Carcinogenesis, October 1, 2004; 25(10): 1991 - 2003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Dhar, J. M. Brindley, C. Stark, M. L. Citro, L. K. Keefer, and N. H. Colburn Nitric oxide does not mediate but inhibits transformation and tumor phenotype Mol. Cancer Ther., December 1, 2003; 2(12): 1285 - 1293. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Cooper, J. MacGowan, J. Ranger-Moore, M. R. Young, N. H. Colburn, and G. T. Bowden Expression of Dominant Negative c-jun Inhibits Ultraviolet B-Induced Squamous Cell Carcinoma Number and Size in an SKH-1 Hairless Mouse Model Mol. Cancer Res., September 1, 2003; 1(11): 848 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. G. Uzzo, P. Leavis, W. Hatch, V. L. Gabai, N. Dulin, N. Zvartau, and V. M. Kolenko Zinc Inhibits Nuclear Factor-{kappa}B Activation and Sensitizes Prostate Cancer Cells to Cytotoxic Agents Clin. Cancer Res., November 1, 2002; 8(11): 3579 - 3583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Thompson, J. MacGowan, M. R. Young, N. Colburn, and G. T. Bowden A Dominant Negative c-jun Specifically Blocks Okadaic Acid-induced Skin Tumor Promotion Cancer Res., June 1, 2002; 62(11): 3044 - 3047. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, basal; 
, TNF-
) and TNF
