HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 34, 26024-26031, August 25, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/34/26024    most recent M002455200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hsing, A. Articles by Vaziri, C. Search for Related Content PubMed PubMed Citation Articles by Hsing, A. Articles by Vaziri, C. Social Bookmarking                      What's this?

DNA-damaging Aryl Hydrocarbons Induce Mdm2 Expression via p53-independent Post-transcriptional Mechanisms^*

Andrew Hsing, Douglas V. Faller, and Cyrus Vaziri

From the Cancer Research Center, Boston University School of Medicine, Boston, Massachusetts 02118

Received for publication, March 22, 2000, and in revised form, May 17, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

During previous studies, we found that mdm2 mRNA levels were elevated in benzo[a]pyrene (BaP, a polycyclic aryl hydrocarbon)-treated cells under conditions of DNA damage-induced cell cycle arrest (Vaziri, C., and Faller, D. V. (1997) J. Biol. Chem. 272, 2762-2769). We have identified potential aryl-hydrocarbon receptor-binding sites in the mdm2 promoter. However, we show that induction of mdm2 mRNA by BaP is entirely dependent upon aryl-hydrocarbon-induced genotoxicity and does not involve direct aryl-hydrocarbon receptor-mediated transcriptional activation of the mdm2 gene. Heterologous mdm2 promoter-reporter constructs containing p53-response elements were not responsive to BaP treatment. Therefore the p53-response elements in the mdm2 promoter are insufficient to confer DNA damage-dependent expression of mdm2. Furthermore, mdm2 transcripts were induced by BaP in p53 null cells from transgenic mice (although both basal and BaP-induced mdm2 expression levels were reduced in these cells relative to p53^+/+ cultures). These data show that p53-mediated mechanisms cannot account for BaP/DNA damage-induced mdm2 expression. Mdm2 promoter-reporter gene assays and nuclear run-off analyses of nascent mdm2 transcripts showed that transcriptional induction was unable to account for the large changes in mdm2 transcript levels following BaP treatment. However, mdm2 mRNA half-life measurements showed stabilization of the mdm2 transcript (from ~1 h to >4 h) in response to BaP. To our knowledge, this is the first report of control of mdm2 at the post-transcriptional level and in a p53-independent manner. Transient ectopic expression of mdm2 strongly augmented aryl-hydrocarbon-induced apoptosis, demonstrating that mdm2 levels can have a profound effect on the cellular response to DNA damage. Overall, our results suggest a potentially important link between DNA damage signaling and RNA stability that may be relevant to cell cycle regulation, tumor suppression, and environmental carcinogenesis.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Polycyclic aryl-hydrocarbons (PAHs,¹ typified by the ubiquitous pollutant benzo[a]pyrene or BaP) are potent mutagens that elicit tumors in experimental animals and transform normal cells to malignancy in vitro (1). Epidemiological data have also suggested an association between PAH exposure and increased incidence of certain cancers in humans (2, 3). Many PAHs, including BaP, bind to and activate a ligand-dependent transcription factor termed the aryl-hydrocarbon receptor or AhR (4). Ligand-bound AhR translocates to the nucleus and associates with an AhR-related protein termed AhR nuclear transporter (ARNT). Ligand-activated AhR·ARNT complexes can interact with specific promoter elements (termed dioxin-response elements or xenobiotic-response elements), which are present in the 5'-region of a subset of cellular genes including those encoding cytochromes P450 CYP 1A1 and CYP1A2 (4, 5). The proteins encoded by these CYP genes are termed "substrate inducible," because many of the PAHs that induce their expression (via AhR) are also substrates for the metabolic oxidation reactions carried out by these enzymes. Cytochrome P450-mediated metabolism of PAHs such as BaP generates electrophilic species (such as benzo[a]pyrene dihydrodiol epoxide, or BPDE) that react with cellular macromolecules, including DNA, to generate covalently linked adduct moieties (6, 7). Metabolically activated species (typified by BPDE) are considered to be the "ultimate carcinogens" resulting from PAH metabolism, because these mediate the mutagenic and carcinogenic actions of the parent aryl-hydrocarbons (6-8).

Potentially, misreplication of aryl-hydrocarbon-adducted genes during S-phase may result in the "fixation" of mutations. Any such mutations that activate cellular proto-oncogenes or mutations that inactivate tumor suppressor genes might contribute to initiation or progression of malignancy. Indeed, both proto-oncogenes (such as c-ras) and tumor suppressor genes (such as p53) are known to be susceptible to oncogenic mutations in DNA damage "hot spots" targeted by PAHs (8).

Because cells are subject to DNA damage from both intrinsic and extrinsic genotoxic agents, they have evolved mechanisms to prevent the mutagenic effects of DNA damage. Cell cycle "checkpoints" are mechanisms that exert negative controls over cell cycle progression in response to DNA damage (9, 10). Checkpoints are hypothesized to allow cells to repair damaged DNA prior to resuming cell cycle progression (9, 10).

In mammalian cells, the product of the p53 tumor suppressor gene is considered to play an important role in DNA damage-induced cell cycle checkpoint responses (11, 12). The p53 gene encodes a transcriptional activator that is stabilized and activated (via post-translational mechanisms) in cells containing damaged DNA (11, 12). When activated, p53 is believed to regulate the expression of a battery of DNA damage-responsive genes. The p21 cyclin-dependent kinase inhibitor was initially identified as a p53 inducible transcript and the p21 gene was shown to contain p53-response elements (13). p21 encodes a protein that binds to and inhibits the activities of cylin·cyclin-dependent kinase complexes that normally mediate cell cycle progression. Therefore, when expressed in response to DNA damage, p21 can elicit cell cycle arrest, thereby mediating the p53-regulated checkpoint (13, 14). The importance of p53-dependent checkpoints in tumor suppression is highlighted by the existence of cancer-prone individuals with congenital defects in p53 (e.g. Li-Fraumeni patients with inactivating germline mutations in the p53 gene).

We previously identified a cell cycle checkpoint elicited by BaP-induced DNA damage (15, 16). Interestingly, BaP-induced cell cycle arrest occurred in p53-deficient cells as well as in p53-proficient cell lines. We hypothesize that the BaP-induced DNA damage checkpoint(s) represent potentially important mechanisms that protect against acquisition of mutations and neoplasia. In previous experiments in which we tested the expression of known p53-responsive genes in BaP-treated cells, we detected strong induction of the mdm2 transcript.

The mdm-2 oncogene was first identified as an amplified DNA present on double minute particles in a spontaneously transformed murine 3T3 cell line (17). MDM2 has since been shown to be overexpressed in several human malignancies such as osteogenic sarcomas (40-60%) and soft tissue sarcomas (30%) (18). The oncogenic properties of the MDM2 protein result in large part from negative regulation of the p53 protein. MDM2 has been shown to bind the transactivation domain of p53, thereby inhibiting p53-mediated gene regulation (19). Additionally, MDM2 targets p53 for ubiquitin-mediated proteolysis (20). Thus mdm2 is an important negative regulator of p53-mediated biological effects. Indeed, homozygous deletion of mdm2 results in embryonic lethality because of increased (p53-mediated) apoptosis (21). However, the phenotype of mdm2^/ mice can be rescued by simultaneous deletion of p53, thereby demonstrating the importance of the p53-mdm2 regulatory loop (21). The mdm2 gene contains two p53-response elements, suggesting a reciprocal role for p53 in mdm2 regulation (22). Consistent with this hypothesis, exogenous expression of p53 has been shown to induce mdm2 promoter-driven reporter genes (22). Moreover, p53 null cells express reduced levels of mdm2 transcripts relative to cells from p53-proficient lines (23). Thus p53 and MDM2 participate in an intricate autoregulatory loop in which the balance of MDM2 and p53 protein levels/activities can have profound effects on cell cycle regulation.

In addition to regulating p53 activity and protein levels, MDM2 has been shown to interact physically and functionally with another important tumor suppressor protein, namely pRB (24). The interaction between Rb and MDM2 results in derepression of the E2F transcription factor by Rb, thereby generating a proliferative signal. Moreover, MDM2 makes physical contact with E2F1·DP1 complexes, further stimulating transcription of E2F-dependent genes (25).

Perturbation of Rb and p53 signaling are frequently causative events in human cancers (11). Clearly Mdm2 is an important regulator of these tumor-suppressive pathways. Our previous data showed a novel stimulatory effect of PAHs upon mdm2 expression that may be relevant to aryl-hydrocarbon-induced tumorigenesis. Therefore, in this report we have investigated the mechanism of mdm2 induction by PAHs. Surprisingly, we show that induction of mdm2 transcripts by genotoxic aryl-hydrocarbons is largely p53-independent and involves post-transcriptional mechanisms. Furthermore, we show that levels of mdm2 expression can dramatically influence the cellular response to aryl-hydrocarbon-induced DNA damage. Our data suggest a potentially important link between DNA damage surveillance and RNA stability that is likely to be important for normal cell cycle regulation and tumor suppression.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cells and Culture-- Swiss 3T3 and Rat1 fibroblasts were obtained from the American Type Culture Collection and maintained as described previously (16). Primary p53^+/+ and p53^/ mouse fibroblasts were kindly provided by Earlene Schmitt and Tyler Jacks (Massachusetts Institute of Technology, Cambridge, MA). AhR^/ fibroblasts were cultured from trypsinized day 11 mouse embryos. For drug treatment, confluent monolayers of quiescent cells were treated with fresh 10% serum-containing medium containing the indicated doses of aryl-hydrocarbons. All aryl-hydrocarbons were obtained from the NCI Chemical Carcinogen Repository. These were added to cell cultures as 1000× stocks dissolved in Me₂SO. Control cultures received Me₂SO with no aryl-hydrocarbon.

Northern Blot Analysis-- RNA extraction and analysis was performed as described previously (15, 16). Ten micrograms of total cellular RNA/sample were fractionated in 1% agarose gels containing formaldehyde and µg of ethidium bromide. Equivalent loading of RNA samples was confirmed by visualization of 18 S and 28 S rRNA. Gels were treated with 0.2 M NaOH for 30 min and neutralized with 50 mM Tris-HCl, pH 7.4, 10 mM NaCl for 30 min before transfer to Duralon-UV nylon membranes (Stratagene) by capillary transfer with 20× SSC. Membranes were cross-linked by UV irradiation using a Stratalinker (Stratagene) and hybridized with random-primed ³²P-labeled cDNA probes. Hybridization was overnight at 42 °C with 5× SSC, 1% SDS, 25× Denhardt's solution, 50% formamide, and 0.1 mg/ml salmon sperm DNA. After hybridization membranes were washed with 0.01% sarkosyl, 0.4× SSC twice at 65 °C and twice at 42 °C. To quantitate relative changes in mRNA expression, regions of the developed nylon filters corresponding to transcripts of interest were excised. The amount of ³²P radioactivity (cpm) associated with each excised area was determined by Cerenkov counting. Similarly sized regions of "blank" areas on the blots were also quantitated to obtain background cpm.

Nuclear Run-off-- Preparation of nuclei, in vitro extension of nascent transcripts in the presence of [-³²P]UTP, and hybridization of labeled transcripts with immobilized plasmid DNA were performed exactly as described in a previous publication (26).

Transfections and Reporter Gene Assay-- For stable transfections, Swiss 3T3 cells were transfected with 30 µg of the mdm2 promoter-CAT construct designated 2.9CAT0 (kindly provided by Dr. Moshe Oren, Weizman Institute) and 1 µg of pClneo (Promega) using calcium phosphate co-precipitation. Stably transfected cells were selected in medium containing 0.5 mg/ml G418. G418-resistent colonies were pooled and used for studies of reporter gene analysis. Preparation of cell extracts and CAT assays were performed according to standard protocols exactly as described in our previous publications (16).

Protein Extraction and Immunoblotting-- Cell extracts were prepared exactly as described previously (16). After normalizing for protein content, extracts were separated on 9% SDS-polyacrylamide gels, transferred to nitrocellulose filters, and probed with commercially available antibodies against phosphoserine 15 of p53 (Oncogene Sciences), Mdm2, and p21 (Santa Cruz). Antibody incubations and detection were performed exactly as described in a previous publication (16).

Adenovirus-- AdCon and AdGFP were described previously (16). Admdm2 was generated similarly by subcloning the murine mdm2 cDNA into the pACCMV shuttle vector and co-transfection of the resulting pACmdm2 construct with pJM17 into 293T cells. The identity of the resulting recombinant adenovirus was verified by restriction mapping and Southern blotting. Large scale purification of adenovirus was performed by polyethylene glycol precipitation, CsCl density gradient centrifugation, and Sephadex G-25 chromatography as described previously (16). For adenovirus infections, 5 × 10¹⁰ plaque-forming units of purified adenovirus particles were added to confluent cultures of Rat1 cells in 10-cm plates. This concentration of virus was sufficient to infect all the cells in the population, as evidenced by fluorescence microscopy of parallel cultures infected with a similar dose of AdGFP.

Cell Cycle Analysis-- Cells were trypsinized, fixed in ethanol, and stained with propidium-iodide as described previously (16). Propidium-iodide-stained nuclei were analyzed for DNA content using a fluorescence-activated cell sorter (Becton-Dickinson) as described previously (16). The distribution of cells between different phases of the cell cycle was determined using the Cellquest program.

Sequence Analysis of the mdm2 Promoter-- We designed an antisense primer corresponding to the translational start site of the mdm2 cDNA. This was used to sequence the 2.9CAT0 plasmid. We obtained ~2 kilobases of sequence immediately upstream of the AUG start site of the mdm2 gene. Sequencing was performed by the Boston University Core Sequencing Facility using a semi-automated instrument (ABI).

Reproducibility-- All data shown are representative results of experiments that were performed at least three times with similar results on each separate occasion.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Aryl-hydrocarbon-induced Genotoxicity Mediates Induction of mdm2 Transcripts-- We previously demonstrated elevated levels of mdm2 transcripts in aryl-hydrocarbon (BaP)-treated cells (15). Furthermore, our sequence analysis identified potential AhR·ARNT sites in the 5'-promoter region of the mdm2 gene. These putative AhR sites (which are in a region of the mdm2 gene that has not previously been sequenced) are described in Table I. It seemed likely therefore, that aryl-hydrocarbons might induce mdm2 expression via direct AhR-mediated transcriptional activation. However, our previous studies have identified cellular responses that result indirectly from BaP metabolism and acquisition of aryl-hydrocarbon-induced DNA damage (15). Additionally, other groups have shown that genotoxic agents can indeed induce mdm2 expression (23, 27). Therefore, we performed experiments to distinguish between a potential mdm2 induction mechanism involving direct AhR-mediated transcriptional activation and an alternative possible mechanism mediated by genotoxic metabolites.

We compared the effects of genotoxic aryl-hydrocarbons (BaP and BPDE) and a poorly metabolized (and therefore nongenotoxic) AhR ligand (2,3,7,8-tetrachloro-p-dioxin (TCDD) or dioxin) on mdm2 expression. Cultures of Swiss 3T3 cells were treated with BaP, BPDE, or TCDD (or Me₂SO for controls) for 24 h. Northern blot analysis of RNA samples extracted from the cultures showed that mdm2 transcript levels were increased by 3- and 3.2-fold in response to BaP (2 µM) and BPDE (0.3 µM), respectively. However, TCDD treatment had no effect on mdm2 transcript levels (Fig. 1). In previous studies, we have shown that both BaP and TCDD elicit nuclear translocation and activation of AhR in these cells (28). Therefore AhR activation by nongenotoxic aryl-hydrocarbons is insufficient to elicit an mdm2 response, suggesting that mdm2 expression resulting from BaP treatment is instead mediated by genotoxic metabolites.

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Induction of mdm2 mRNA by genotoxic aryl-hydrocarbons in AhR+/+ and AhR/ cell lines. AhR+/+ Swiss 3T3 cells (upper panel) or AhR/ 3T3 cultures (lower panel) were stimulated with 1.0 µM BaP (BP), 0.3 µM BPDE (DE), 10 nM TCDD, or Me2SO () for 24 h. Mdm2 transcripts were detected by Northern blot analysis and quantitated as described under "Materials and Methods."

To further test the role of AhR in aryl-hydrocarbon-induced mdm2 expression, we performed additional experiments using AhR null cell lines from transgenic mice. AhR^/ fibroblasts were treated with BaP or BPDE (or received no treatment for control). 24 h after treatment with aryl-hydrocarbons, we analyzed mdm2 expression by RNA blotting. As in AhR-proficient cell lines, BPDE elicited high level mdm2 expression relative to control cultures that received no aryl-hydrocarbon (Fig. 1). Unlike BPDE, BaP failed to induce mdm2 expression in AhR null cells. We have previously shown that AhR is necessary for metabolism of BaP and acquisition of DNA damage (15). Therefore, the lack of mdm2 induction by BaP in AhR null cells is consistent with a DNA damage-mediated mechanism for mdm2 regulation by aryl-hydrocarbons.

Induction of mdm2 in p53 Null Cells-- DNA-damaging agents (UV and -radiation) have been shown to induce mdm2 expression (23, 27). It generally believed that p53-dependent mechanisms mediate mdm2 transcription in response to DNA damage. Indeed, the mdm2 gene does contain p53-response elements (22), and p53 deficiency results in reduced expression of mdm2 (23). However, we have shown that p53-independent mechanisms mediate some of the cellular responses to BaP-induced DNA damage. It was therefore of interest to us to determine the p53 dependence of BaP-induced mdm2 expression.

To test the p53 dependence of BaP-induced mdm2 expression, we compared BaP-induced mdm2 expression in p53-proficient cells and in p53-deficient lines from p53^/ transgenic mice. As in previous experiments, RNA samples from control and BaP-treated p53^+/+ and p53^/ cells were analyzed for mdm2 expression by Northern blot analysis. Consistent with results from other laboratories, p53^/ cells expressed low levels of mdm2 relative to p53-proficient lines (Fig. 2). Interestingly, however, mdm2 transcripts were induced in response to BaP in p53-deficient cells (albeit to a lesser level than that attained in p53-proficient cells). The relatively low basal and BaP-induced levels of mdm2 in p53^/ cells are consistent with an important role for p53 in maintenance of mdm2 transcription (22). However, our data suggested the interesting possibility that BaP-induced mdm2 expression may also occur via p53-independent mechanisms. Whereas this experiment does not formally rule out the possibility that other p53-related proteins (e.g. p63 and p73) compensate for p53 deficiency and regulate mdm2 transcription in response to DNA damage, data described below will demonstrate that p53-response elements in the mdm2 promoter cannot account for BaP-induced mdm2 expression.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Effect of p53-deficiency on BaP-induced mdm2 expression. Cultures of p53+/+ (left panels) and p53/ (right panels) cells were stimulated with 1.0 mM BaP or received Me2SO (for controls) for 24 h. Mdm2 transcripts were detected by Northern blot analysis as described under "Materials and Methods." Please note that the autoradiogram in the right panels (p53/ samples) represents a filter that was exposed to film for approximately three times as long as that shown in the left panel (p53+/+ cells).

Induction of mdm2 by BaP Does Not Involve p53-response Elements in the mdm2 Gene and Occurs Post-transcriptionally-- The data described above suggested that p53 did not mediate increased expression of mdm2 mRNA in response to BaP-induced DNA damage. To investigate the possibility that alternative DNA damage-response elements in the mdm2 gene might mediate the observed BaP-induced changes in mdm2 transcript levels, we tested the responsiveness of a heterologous mdm2 promoter-reporter construct to BaP treatment. We obtained an mdm2 promoter-reporter plasmid designated 2.9CAT0 (kindly provided by Dr. Moshe Oren). In this reporter plasmid, a 2.9-kilobase fragment of the mdm2 gene (including the 5'-promoter region as well as internal promoter containing p53-response elements) is linked to a CAT cDNA (22). This construct has previously been shown (by Oren and colleagues (22)) to be responsive to ectopically expressed p53 in transient co-transfection experiments (22). We have also observed trans-activation of the 2.9CAT0 construct by ectopically expressed p53 in 3T3 cells (Fig. 3a). We generated a stably transfected cell line containing 2.9CAT0. CAT activity was readily detectable in extracts from the 2.9CAT0 stable transfectant cell line (Fig. 3b). Moreover, the fragment of the mdm2 gene present in the 2.9CAT0 plasmid also conferred the modest mitogen-induced expression of mdm2 during late G₁ (which we originally described in Ref. 15). Therefore, as expected the 2.9CAT0 construct responded to appropriate activating signals in 3T3 cells.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of aryl-hydrocarbons on mdm2 promoter-driven reporter (CAT) activity. a, 3T3 cells were transfected transiently with 10 µg of 2.9CAT0 plus 10 µg of "empty" vector (pcDNA) or 10 µg of a CMV-driven p53 expression vector (CMV-p53). 48 h after transfection, the cells were harvested, normalized for protein content, and assayed for expression of CAT enzymatic activity as described under "Materials and Methods." The figure is an autoradiogram of the developed TLC plate. The arrows indicate the position of migration of the acetylated [14C]chloramphenicol reaction products. b, RNA samples from quiescent and 10% serum-stimulated 3T3 cells (15 h stimulation) were analyzed for mdm2 expression by Northern blotting. The serum-induced mdm2 transcript is indicated. In a parallel experiment quiescent S2.9CAT0 cells (which contain stably integrated mdm2 promoter-CAT plasmid) were treated with 10% serum (or left untreated for controls). After 15 h, duplicate plates of control and serum-stimulated cells were harvested and assayed for CAT enzymatic activity as described above. The figure is an autoradiogram of the developed TLC plate (right panel). c, Swiss 3T3 cells stably expressing the 2.9CAT0 plasmid were treated with 1.0 µM BaP (BP), 0.3 µM BPDE (DE), 10 nM TCDD, or Me2SO () for 24 h. Extracts from the resulting cells were assayed for CAT activity as described above.

We grew the Swiss 3T3 2.9CAT0 transfectants to confluence and treated them with various aryl-hydrocarbons as described above. After 24 h the aryl-hydrocarbon-treated (or control) cultures were harvested and assayed for reporter enzyme (CAT) activity. Interestingly, these assays indicated that there was no change in the levels of mdm2 promoter-driven reporter enzyme expression in extracts from aryl-hydrocarbon (BaP, BPDE, or TCDD)-treated cells (Fig. 3c). In control RNA blotting experiments in which we analyzed RNA from parallel cultures of the 2.9CAT0-expressing cells we did observe the expected induction of mdm2, showing that these cells indeed retained BaP-responsive induction of the endogenous mdm2 transcript (these data are identical to those presented in Fig. 1 but are not shown to avoid duplication of this result).

Our data showed that p53 (and AhR) sites present in the mdm2 promoter were insufficient to confer BaP-induced expression of CAT. This result suggested either that PAH-induced expression of mdm2 required additional cis-acting elements (that were absent from the 2.9-kilobase promoter fragments used for our reporter gene assays) or that BaP-induced changes in mdm2 expression occurred via post-transcriptional mechanisms.

To test whether PAH induction of mdm2 occurred through increases in transcription, we performed nuclear run-off analysis of nascent mdm2 transcripts in nuclei from control and PAH-treated cells. Confluent cultures of Swiss 3T3 cells were treated for 24 h with aryl-hydrocarbons (BaP or BPDE) or were left untreated (for controls). We isolated nuclei from the cultures and generated in vitro run-off transcripts in the presence of [³²P]UTP, as described under "Materials and Methods." Equivalent amounts (cpm) of the resulting ³²P-labeled run-off transcripts from each experimental condition were hybridized with membranes containing immobilized mdm2 cDNA as well as pGEM3 (as a negative control for hybridization) and genomic DNA (as a positive control for hybridization to the filter).

The top panel of Fig. 4 shows an autoradiogram of the washed and RNase-treated membranes following hybridization with labeled transcripts. As expected, there was no hybridization of transcripts to the pGEM3 control DNA (demonstrating lack of nonspecific interactions between labeled transcripts and immobilized DNAs), and transcripts derived from all experimental conditions hybridized strongly to genomic DNA. There was no increase in the amount of ³²P-labeled mdm2 transcripts generated by nuclei from aryl-hydrocarbon-treated cells relative to control, untreated fibroblasts. In the same experiment, mdm2 mRNA levels were strongly induced by both BaP and BPDE (Fig. 4, lower panel). These data corroborated our earlier reporter gene analyses and demonstrated that PAH-induced mdm2 expression did not result from transcriptional mechanisms. These data were also consistent with our previous observation that regulation of mdm2 levels in response to PAHs was not p53-mediated.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Nuclear run-off analysis of mdm2 transcripts in aryl-hydrocarbon-treated cells. Swiss 3T3 cells were treated with 1.0 µM BaP (BP), 1.0 µM BPDE (DE), or Me2SO () for 24 h. Nuclei prepared from the resulting cells were used to generate [32P]UTP-labeled nascent "run-off" transcripts. Equivalent cpm of radioactive transcripts were hybridized extensively with filters containing 1 µg of immobilized genomic DNA, mdm2 cDNA, and pGEM3 DNA. The resulting filters were washed and Rnase-treated prior to being exposed to film (upper panel). In the same experiment, RNA was harvested from some plates of cells and analyzed for mdm2 by Northern blotting (lower panel).

Induction of mdm2 by BaP Results from Increases in RNA Stability-- The results from our promoter-reporter gene analyses and the nuclear run-off experiments showed that mdm2 induction by BaP·DNA damage is p53-independent and not regulated at the level of transcription. It seemed likely therefore that mdm2 induction may result from increased transcript stability following acquisition of DNA damage. To test this hypothesis, we determined the stability of the mdm2 transcript in control and BaP-treated cells.

Confluent Swiss 3T3 cells were treated with BaP (to induce mdm2 expression) or were left untreated (for controls). 24 h later, the cultures received the RNA synthesis inhibitor actinomycin D for various times. RNA samples isolated from the cells at various times after actinomycin D treatment were tested for mdm2 expression by Northern blot analysis. As shown in Fig. 5, the mdm2 transcripts in control and BaP-treated cultures decayed at markedly different rates.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effect of BaP on mdm2 mRNA stability. Swiss 3T3 cells were treated with 1.0 µM BaP (BP) or Me2SO () for 24 h. All cultures then received 10 µg/ml actinomycin D. RNA samples were collected from the cultures at various times (0-7 h) after actinomycin D addition. RNAs were analyzed for mdm2 expression by Northern blotting (top panels). To quantify mdm2 mRNA levels remaining at different times after actinomycin D treatment, the bands corresponding to labeled mdm2 were excised from the filter. The amount of radioactivity associated with each excised band was determined by Cerenkov counting. A background cpm value (corresponding to a similarly sized blank region of the blot) was subtracted from the resulting counts. The amount of mdm2 radioactivity remaining at each time after actinomycin D treatment was expressed as a percentage of the mdm2-associated label at the time of actinomycin D addition (time 0). These data are represented graphically in the lower panel.

To quantify the relative levels of mdm2 expression in this experiment, sections of the nitrocellulose filter corresponding to mdm2 transcript were excised from the blot. The amount of ³²P associated with each excised band was measured by Cerenkov counting. The percentage of mdm2 remaining at each time point following actinomycin D treatment (for control and BaP-treated cells) is shown graphically in Fig. 5. These data show an increase in mdm2 half-life (from ~1 h in control cultures to ~4 h in BaP-treated cells) in response to PAH-induced genotoxicity. In three independent experiments, BaP treatment increased the half-life of the mdm2 transcript from 1 ± 1/4 h to 4-5 h. Our data strongly suggest that changes in mdm2 expression in response to PAH-adducted DNA post-transcriptionally via mechanisms involving changes in mRNA stability. It is formally possible that the high levels of mdm2 mRNA in aryl-hydrocarbon-treated cells might overwhelm the putative factors that mediate transcript degradation, thereby resulting in artifactual increases in our measurements of mRNA half-life. Although we cannot unambiguously exclude this possibility, the data presented in Figs. 3 and 4 show that transcriptional mechanisms are unlikely to account for the DNA damage-induced changes in mdm2 mRNA. Therefore, the simplest explanation for our data is that the mdm2 transcript is indeed stabilized in cells containing DNA damage.

For comparison, we also examined the kinetics of decay of another DNA damage-induced transcript, namely the p21 cyclin-dependent kinase inhibitor. As we found for mdm2 mRNA, p21 transcripts were stabilized in response to PAH-induced DNA damage (Fig. 6). p21 message half-life was increased from ~1 to >3 h after BaP treatment. The increased expression of mdm2 and p21 transcripts in BaP-treated cells did not reflect a general change in cellular transcripts. In fact, in a recent DNA chip gene array experiment we compared the expression of 18,000 transcripts in control and BaP-treated cells. In this experiment the levels of expression of only 48 mRNAs, including mdm2 and p21, varied in abundance (reflecting increased as well as decreased levels in response to BaP) by a significant amount (2.5-fold or greater, data not shown). Therefore, aryl-hydrocarbon·DNA damage-induced changes in mRNA levels are restricted to a subset of RNA transcripts. Overall, our data suggest that regulation of transcript stability in response to BaP-adducted DNA is important for determining expression levels of mdm2 and p21, and perhaps other DNA damage inducible transcripts.

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Effect of BaP on p21 mRNA stability. The RNA samples used for the experiment described in Fig. 5 were analyzed for p21 expression by Northern blotting and hybridization with a random-primed p21 cDNA probe (upper panels). Quantification of relative amounts of p21 mRNA between different samples (lower panel) was performed exactly as described in the legend to Fig. 5.

Effect of Aryl-hydrocarbons on p53, mdm2, and p21 Protein-- It was of interest to determine whether the aryl-hydrocarbon-induced changes in mdm2 and p21 mRNA levels were sufficiently large to affect expression of the corresponding proteins. Therefore we performed immunoblot analysis of cell lysates from control and aryl-hydrocarbon-treated 3T3 cells. As shown in Fig. 7, Mdm2 protein levels were increased by ~3.2-fold in response to BPDE. This fold induction correlates well with the aryl-hydrocarbon-induced change in mdm2 mRNA levels. For comparison, the same filter was also probed with antisera against p21. Interestingly, and in contrast with Mdm2, p21 protein levels were not increased in response to BPDE (despite increases in p21 mRNA relative to control cultures under the same conditions, as shown in Fig. 6). However, as shown by other workers, p21 levels are high in mitogen-treated cells. It is possible that p21 protein is expressed maximally as a result of serum stimulation in the control cultures, thereby precluding additional increases in response to aryl-hydrocarbon. Although our data suggest that p53 does not play a major role in the induction of mdm2 in response to aryl-hydrocarbons, we have previously demonstrated modest increases in p53 levels in BaP-treated cells (15). We have extended our earlier results by performing immunoblot analysis with anti-phoshpho-p53 antibodies. These results, which are presented in Fig. 7, show increased phosphorylation of serine 15 of p53 in response to aryl-hydrocarbon treatment. Serine 15 is known to be phosphorylated by the ataxia telangactasia-mutated protein. Therefore our data suggest a possible role for ataxia telangactasia-mutated protein in cellular responses to aryl-hydrocarbon-induced DNA damage (although, we have no evidence for involvement of ataxia telangactasia-mutated protein in mdm2 regulation).

View larger version (34K): [in this window] [in a new window]   Fig. 7.   Immunoblot analysis of mdm2, p21, and serine 15-phosphorylated p53. Quiescent cultures of Swiss 3T3 cells were treated with 1 µM BPDE (or Me2SO for control). Cultures then received 10% serum for 20 h. Extracts from the resulting cells (50 µg protein/lane) were separated by SDS-polyacrylamide gel electrophoresis on a 9% gel. After transfer to nitrocellulose, appropriate sections of the same filter was probed with antibodies against Mdm2, p21, and serine 15-phosphorylated p53 (p53(P-ser 15)).

Ectopically Expressed mdm2 Augments BPDE-induced Apoptotic Response-- We tested the effect of mdm2 overexpression on the cellular responses to aryl-hydrocarbon-induced DNA damage. We generated a recombinant adenovirus encoding the mdm2 cDNA. The resulting virus (designated Admdm2) and a control virus (designated AdCon) were used to infect cultures of quiescent BPDE-treated (and control untreated) Rat1 fibroblasts. The resulting cells were stimulated to enter the cell cycle by the addition of mitogen (10% serum). 24 h after the addition of serum, the cells were harvested, fixed, and stained with propidium iodide for FACScan analysis. The fluorescence-activated cell sorter profiles obtained in this experiment are shown in Fig. 8. As expected, quiescent cells displayed a predominantly G₀/G₁ DNA content at the time of serum addition (Fig. 8, top panel). Serum stimulation of AdCon-infected cells resulted in the expected progression through S-phase and accumulation of cells in G₂/M. As shown in a previous publication (16), BPDE treatment of AdCon-infected cells elicited S-phase arrest, as evidenced by increased numbers of cells in S-phase and decreased G₂/M populations. In agreement with studies from other laboratories (32), BPDE also elicited an apoptotic response (~12.2% apoptosis), as shown by the increase in the sub-G₁ population indicated by marker M1 in Fig. 8. Interestingly, Admdm2 infection in the absence of DNA damage resulted in cell cycle acceleration, as shown by the increased numbers of G₂/M phase cells relative to AdCon-infected cultures. Remarkably, however, mdm2 overexpression in BPDE-treated cells resulted in a large increase in the apoptotic response (Fig. 8). Approximately 60.7% of the BPDE + Admdm2-treated cells death underwent apoptosis within a 24-h period following serum stimulation. By 48 h post-serum stimulation, the majority of the AdCon + BPDE-treated cells had recovered from S-phase arrest and progressed into G₂/M. In contrast, The entire population of Admdm2 + BPDE-treated cells had undergone apoptosis at this time (data not shown). These data demonstrate that mdm2 expression levels can have a profound effect on the cellular responses to aryl-hydrocarbon-induced DNA damage.

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Effect of transiently expressed mdm2 on cell cycle and apoptosis. Quiescent cultures of Rat1 cells in 10-cm plates were treated with 0.6 µM BPDE (or Me2SO for controls). After 2 h, the cells received 5 × 1010 plaque-forming units of AdCon or Admdm2. Parallel infections performed using AdGFP indicated that this dose of virus was sufficient to infect the entire population. 2 h later, cultures were stimulated to enter the cell cycle by the addition of 10% serum. All cultures also received 0.2 µg/ml of nocodazole (to simplify cell cycle analysis by "trapping" cells that complete S-phase in a G2/M-arrested state). 24 h after serum addition, cells were harvested for fluorescence-activated cell sorter analysis as described under "Materials and Methods." The marker designated "M1" on the each of the above profiles indicates the sub-G1(apoptotic) population. For quiescent cells that received no serum and no BPDE, M1 = 2.6% of the total population. For AdCon + serum-treated cells, M1 = 1.5%. For AdCon + serum + BPDE-treated cells, M1 = 12.2%. For Admdm2 + serum-treated cells M1 = 3.2%. For Admdm2 + serum + BPDE-treated cells, M1 = 60.7%.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this report we have characterized the mechanism of mdm2 induction by aryl-hydrocarbons. Although we have identified potential binding sites for ligand-activated AhR·ARNT in the 5'-region of the mdm2 gene, our data show that aryl-hydrocarbon-induced genotoxicity is necessary for mdm2 induction. Other genotoxic agents have been shown to elicit increases in mdm2 expression (23, 28), yet this has always been attributed to direct transcriptional activation of the mdm2 gene by activated p53. This reasonable assumption has been supported by the ability of ectopically overexpressed p53 to increase mdm2 promoter-driven reporter activity in transient co-transfection experiments (22) and by the demonstration of increased abundance of mdm2 transcripts originating at the p53-specific internal promoter in response to UV radiation (27). Clearly, however, these results do not preclude an additional post-transcriptional mechanism for mdm2 induction in genotoxin-treated cells. Our data show that p53 is not required for the BaP-induced elevation of mdm2 expression (although consistent with work from other laboratories, we find that p53 plays an important role in maintaining basal and induced levels of mdm2 mRNA). Our reporter gene assays and nuclear run-off analyses show that p53-mediated (or other) transcriptional mechanisms are unlikely to account for BaP-induced mdm2 expression. Instead, our measurements of RNA half-life in control and BaP-treated cells demonstrate that aryl-hydrocarbon induction of mdm2 mRNA results from increased RNA stability.

Although the mdm2 promoter contains several consensus sites for AhR·ARNT, our data show that activation of AhR (by the potent nongenotoxic AhR ligand TCDD) does not confer increases in mdm2 expression. It is formally possible that AhR may contribute to mdm2 regulation in cell lineages other than the mesenchymal cultures used in our experiments. Alternatively it is possible that additional transcriptional activators are required for (putative) AhR-mediated regulation of mdm2 transcription. In this regard, the mdm2 5'-promoter contains potential sites for a number of other transcription factors (including E2F, estrogen receptor) that might cooperate with AhR to regulate gene expression. In unpublished transient transfection experiments we have found that high level overexpression of AhR (in the absence of ligand) can increase mdm2 promoter-driven CAT expression. Interestingly, we have previously shown that AhR expression is up-regulated by mitogenic stimuli (28). Potentially, mitogen-induced increases in AhR levels could activate the mdm2 gene (or other AhR target genes). Further experiments are underway to determine whether AhR plays any physiological role in mdm2 regulation. Nevertheless, our data indicate that AhR is necessary for the metabolic activation of BaP, but plays no direct role in transcriptional activation of the mdm2 gene in cultured rodent fibroblasts.

The p53 protein is a transcriptional activator, which is itself activated and stabilized in response to DNA damage. p53 is generally believed to mediate transcriptional induction of DNA damage-responsive genes such as mdm2 and p21, which contain p53-response elements. However, BaP also induced expression of mdm2 mRNA in p53 null cells (although both basal and induced levels of mdm2 mRNA were reduced in p53 null cells relative to control p53^+/+ cultures). Furthermore, we were unable to detect increased transcription of mdm2 using either reporter gene assays (with mdm2 promoter-CAT constructs) or nuclear run-off analyses. Collectively, these data show that the BaP-induced expression of mdm2 mRNA does not result from p53-mediated transcriptional activation. Instead, our measurements of mdm2 transcript half-life show stabilization of this mRNA in BaP-treated cells. To our knowledge, there is no precedent for control of mdm2 mRNA (by DNA damaging agents or other stimuli) by transcript stabilization.

Interestingly, p21, which was originally identified by Vogelstein and colleagues (13) as a p53-induced transcript, has also recently been shown to be induced via p53-independent pathways. For example p53 null cell lines show induction of p21 in response to DNA damage, although both basal and DNA damage-induced p21 levels are lower in p53 null cells than in p53-proficient lines (29). Our data show that expression of mdm2 displays a similar p53-independent pattern of induction by BaP-induced DNA damage. Recent reports by Holbrook, Gorospe, and colleagues (30, 31) have shown that p21 induction by UVC (presumably resulting from DNA damage) does indeed occur as a result of increased p21 message stability. These workers have identified sequences in the 3'-UTR of the p21 transcript that conferred mRNA stabilization in response to UVC. Interestingly, in the latter study the Elav-type RNA-binding protein HuR bound to these sequences in vitro, and lowering of HuR levels in intact cells (by antisense methods) destabilized p21 mRNA. It is highly likely that similar (or analagous) mechanisms exist to regulate p21 and mdm2 RNA stability in response to BaP-induced DNA damage.

It is not intuitively clear why aryl-hydrocarbons should induce mdm2 expression. Nevertheless we and many other groups have shown that mdm2 is induced in response to a variety of DNA-damaging agents. It has been suggested that mdm2 serves to attenuate p53 activity thereby preventing excess p53 signaling. The lethality of the mdm2^/ mice and the rescue of this phenotype by p53 eficiency (in mdm2^/p53^/ "double knockout" mice) demonstrates that excess p53 activity is indeed detrimental. Although these transgenic studies are compelling and indeed suggest a role for mdm2 in normal cell cycle and development, they do not necessarily address the role of mdm2 in the response to DNA damage. An interesting possibility suggested by our mdm2 overexpression experiments is that mdm2 actually contributes to the cell cycle/apoptotic response to DNA damage. It is known that BPDE-treated cells undergo apoptosis (32). Our results show that mdm2 expression can strongly augment the apoptotic response to BPDE. We have not yet elucidated the mechanism whereby mdm2 contributes to apoptosis. However, mdm2 is known to bind Rb and derepress the E2F family transcription factors (24). In unpublished experiments² E2F1 overexpression resulted in acceleration of mitogen-induced cell cycle progression and augmentation of BPDE-induced apoptosis in a manner similar to that observed with mdm2 (in Fig. 8 of "Results" section). Hence we speculate that the effects of mdm2 on cell cycle progression and apoptosis are E2F-mediated. Nevertheless, regardless of the precise mechanism of the mdm2-induced cell cycle and apoptotic effects, our data show that changes in mdm2 expression levels can dramatically impact the cellular responses to DNA damage. Therefore, aryl-hydrocarbon-induced changes in Mdm2 are likely to be of biological importance. Indeed, elimination of cells containing BPDE-adducted DNA through increased apoptosis is likely to represent an effective tumor-suppressive mechanism.

It is becoming increasingly clear that DNA damage can influence gene expression by altering mRNA stability as well as via regulating changes in gene transcription (e.g. via p53-mediated pathways). Although much progress has been made in understanding the signaling pathways that couple DNA damage to p53 and transcriptional activation, little is known regarding the mechanisms that mediate DNA damage-induced changes in mRNA stability. Mdm2 and p21 play key roles in cell cycle regulation and checkpoint control. The putative signaling mechanisms that regulate the stability of these (and probably other) important transcripts are likely to be relevant to our understanding of normal cell growth as well as checkpoint responses. Experiments are currently underway to elucidate the mechanisms whereby BaP-induced DNA damage controls stability of the mdm2 transcript.

	ACKNOWLEDGEMENTS

We thank Dr. Moshe Oren for the 2.9CAT0 plasmid and Dr. Donna George for the murine mdm2 cDNA. We are also grateful to Dr. Mary Ellen Perry for helpful comments during the course of these studies. Antisera against p53 and mdm2 were generous gifts from Dr. Mark Eller and Dr. Remco Spanjaard, respectively.

	FOOTNOTES

^* This work was funded by National Institutes of Health Grants ES05998 (to C. V.) and CA50454 (to D. V. F.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Cancer Research Center, Boston University School of Medicine, 80 E. Concord St., Boston, MA 02118. Tel.: 617-638-4175; Fax: 617-638-5609; E-mail: cvaziri@acs.bu.edu.

Published, JBC Papers in Press, June 15, 2000, DOI 10.1074/jbc.M002455200

² C. Vaziri, unpublished results.

	ABBREVIATIONS

The abbreviations used are: PAH, polycyclic aryl-hydrocarbons; BaP, benzo[a]pyrene; AhR, aryl-hydrocarbon receptor; ARNT, AhR nuclear transporter; BPDE, benzo[a] pyrene dihydrodiol epoxide; CAT, chloramphenicol acetyltransferase.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				V. Stribinskis and K. S. Ramos Activation of Human Long Interspersed Nuclear Element 1 Retrotransposition by Benzo(a)pyrene, an Ubiquitous Environmental Carcinogen. Cancer Res., March 1, 2006; 66(5): 2616 - 2620. [Abstract] [Full Text] [PDF]

				T. R. Van Vleet, T. L. Watterson, P. J. Klein, and R. A. Coulombe Jr. Aflatoxin B1 Alters the Expression of p53 in Cytochrome P450-Expressing Human Lung Cells Toxicol. Sci., February 1, 2006; 89(2): 399 - 407. [Abstract] [Full Text] [PDF]

				S. P. Deb Cell Cycle Regulatory Functions of the Human Oncoprotein MDM2 Mol. Cancer Res., December 1, 2003; 1(14): 1009 - 1016. [Abstract] [Full Text] [PDF]

				M. H. Wu and B. Y. M. Yung UV Stimulation of Nucleophosmin/B23 Expression Is an Immediate-early Gene Response Induced by Damaged DNA J. Biol. Chem., December 6, 2002; 277(50): 48234 - 48240. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/34/26024    most recent M002455200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hsing, A. Articles by Vaziri, C. Search for Related Content PubMed PubMed Citation Articles by Hsing, A. Articles by Vaziri, C. Social Bookmarking                      What's this?