Effect of ionizing radiation on AP-1 binding activity and basic fibroblast growth factor gene expression in drug-sensitive human breast carcinoma MCF-7 and multidrug-resistant MCF-7/ADR cells.

We studied the effect of ionizing radiation on the activation of the AP-1 transcription factors and the regulation of basic fibroblast growth factor (bFGF) gene expression in drug-sensitive human breast carcinoma (MCF-7) cells and its drug-resistant variant (MCF-7/ADR) cells. Northern blot and gel mobility shift assays showed that 135 cGy of ionizing radiation induced c-jun and c-fos gene expression, AP-1 binding activity, as well as bFGF gene expression in MCF-7/ADR cells. In MCF-7 cells, however, we observed little/no induction of bFGF gene expression and AP-1 binding activity after the stress. Nevertheless, MCF-7 cells transfected with plasmids containing c-jun gene contain high levels of bFGF protein. H-7 (60 μg/ml), a potent protein kinase C (PKC) inhibitor, inhibited the stress-induced AP-1 binding activity and bFGF gene expression in MCF-7/ADR cells. Corroborating this observation, overexpression of PKCα induced bFGF gene expression in MCF-7 cells. Taken together, these results suggest that stress-induced bFGF gene expression is mediated through the activation of PKC and AP-1 transcription factors. Differences in the levels of PKC activity and AP-1 binding factors may be responsible for differential expression of bFGF among breast cancer cell lines. Although there are large differences in response to ionizing radiation between MCF-7 and MCF-7/ADR cell lines, we observed no significant differences in radiocytotoxicity between them.

It has been known for many years that alteration of the micro-or macroenvironment of a cell can trigger the highly complex cellular stress management system. Recent studies have shown that, in addition to the well known stress proteins, environmental stresses such as ischemia/hypoxia (Plate et al., 1993;Millauer et al., 1994), tumor promoters (Winkles et al., 1992), and radiation (Witte et al., 1989;Haimovitz-Friedman et al., 1991) are all stimuli for triggering the synthesis of angiogenic factor(s). The observation of the induction of these angio-genic proteins under these circumstances could have significant implications on the process of tumorogenesis. Upon these stresses, tumor cells may induce neovascularization by synthesizing and releasing diffusible tumor-derived angiogenic factors such as vascular endothelial growth factor and bFGF 1 (Shing et al., 1985;Klagsbrun et al., 1986;Plate et al., 1993;Soutter et al., 1993).
A fundamental question which remains unanswered is how the stresses stimulate the synthesis of these angiogenic factors. DNA sequencing studies suggest that angiogenic factor-related genes such as bFGF contain AP-1 cis-acting regulatory elements (TPA response element: TRE) (Kim et al., 1989a(Kim et al., , 1989bShibata et al., 1991). In vitro DNA binding and in vivo footprinting experiments demonstrated that these regulatory elements are recognized by AP-1 transcription factors (Jun and Fos family proteins) (Angel et al., 1988;Deng and Karin, 1993). The activity of AP-1 transcription factor is regulated by the induction of jun and fos gene transcription and post-translational modification of their products Boyle et al., 1991). Several researchers have reported that c-jun and c-fos genes are expressed in response to a wide range of stresses including heat shock exposure (Andrews et al., 1987;Bukh et al., 1990), UV irradiation (Angel et al., 1985;Hollander and Fornace, 1989;Stein et al., 1992), ionizing radiation exposure (Sherman et al., 1990;Hallahan et al., 1991a), and treatment with chemical agents (Andrews et al., 1987;Hollander and Fornace, 1989;Shibanuma et al., 1990) in mammalian cells. Moreover, these stresses also increase AP-1 binding activity (Piette et al., 1988;Hallahan et al., 1993). Thus, we hypothesized that the stress-induced activation of AP-1 transcription factors is responsible for the expression of the bFGF gene as a result of exposure to ionizing radiation. In this study, we also investigated differences between drug-resistant and -sensitive human breast carcinoma MCF-7 cells in response to ionizing radiation. We observed that the radiation-induced expression of jun and fos genes, AP-1 binding activity, as well as bFGF gene expression occurred more prominently in drugresistant MCF-7/ADR cells compared to drug-sensitive MCF-7 cells. Nonetheless, there was no significant differences in the cytotoxic effects of ionizing radiation between these two cell lines.

MATERIALS AND METHODS
Cell Culture-Human breast carcinoma (MCF-7) and its multidrugresistant subline (MCF-7/ADR) cells (S. H. Kim, Henry Ford Hospital, Detroit, MI) were cultured in McCoy's 5a medium with 10% ironsupplemented bovine calf serum (HyClone Laboratories, Logan, UT) and 26 mM sodium bicarbonate for monolayer cell culture. Two or 3 days prior to the experiment, cells were plated into 35-mm Petri dishes, T-150 flasks, or T-75 flasks. The Petri dishes/flasks containing cells were kept in a 37°C humidified incubator with a mixture of 95% air and 5% CO 2 .
Radiation Exposure-Cells were irradiated at room temperature with a GE Maximar 250-III orthovoltage x-ray unit (General Electric, Chicago, IL) at a dose rate of 1.17 Gy/min. The tube voltage is 250 kVp, current was 15 mA, and added filtration was 0.25 mm Cu.
Northern Blot Analysis-Relative levels of c-jun, c-fos, and bFGF mRNA were determined using the Northern blot technique. Total cellular RNA was extracted by the LiCl-urea method of Tushinski et al. (1977). For RNA analysis, 30 g of total RNA was electrophoresed in a 1% agarose-formaldehyde gel (Lehrach et al., 1977). The RNA was blotted from the gels onto nitrocellulose membrane, and baked at 80°C for 2 h in a vacuum oven. Membranes were prehybridized at 42°C in 50% formamide, 1 ϫ Denhardt's solution, 25 mM KPO 4 (pH 7.4), 5 ϫ SSC (1 ϫ SSC ϭ 150 mM NaCl, 15 mM Na 3 C 6 H 5 O 7 ), 50 g/ml denatured and fragmented salmon sperm DNA. Hybridizations were conducted at 42°C in prehybridization solution containing 10% dextran sulfate and radiolabeled appropriate cDNA probes (Oncogene Science, Uniondale, NY) at a concentration of 4 ϫ 10 6 cpm/ml. After hybridization, blots were washed and placed into a stainless steel cassette with intensifying screen and then autoradiographed. The films were scanned with a computerized laser scanning densitometer (model 300A; Molecular Dynamics, Sunnyvale, CA). Quantitative measurement was performed with this instrument.
Preparation of Nuclear Proteins-Cells in T-150 culture flasks were rinsed twice with cold phosphate-buffered saline (PBS) and scraped into 10 ml of PBS. Cells were collected by centrifugation at 250 ϫ g for 5 min. The pellet was resuspended in buffer A (10 mM HEPES, pH 8.0, 0.5 M sucrose, 50 mM NaCl, 1 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine, 1 mM phenylmethylsulfonyl fluoride, 7 mM ␤-mercaptoethanol, and 0.5% Triton X-100) and incubated on ice for 10 min. The nuclei were collected by centrifugation for 10 min at 1000 ϫ g. The nuclear pellet was rinsed twice with 5 ml buffer A and resuspended in 0.5 ml of buffer B containing 100 mM NaCl (10 mM HEPES, pH 8.0, 25% glycerol, 1 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine, 1 mM phenylmethylsulfonyl fluoride, and 7 mM ␤-mercaptoethanol). After 15 min of gentle agitation on ice, the nuclei were centrifuged at 200 ϫ g for 10 min and resuspended in 0.6 ml of buffer B containing 0.5 M NaCl. After 30 min of incubation on ice with occasional gentle agitation, the nuclei were pelleted at 2000 ϫ g for 10 min and supernatant was brought to 45% saturation with (NH 4 ) 2 SO 4 . The precipitated proteins were collected by centrifugation in a microcentrifuge at 14,000 rpm for 30 min, and finally the pellet was resuspended in 100 l buffer B containing 100 mM NaCl. The protein content of the resulted nuclear extract was determined by Bradford assay (Bradford, 1976).
Gel Mobility-shift Assay-A double-stranded oligonucleotide, the TRE-like sequence (GAG TTT AAA CTT TTA AAA GTT GAG TCA CGG CTG GTT G) of the human bFGF gene (Shibata et al., 1991), was used in the binding reaction. The binding assay contained 2-5 g of nuclear extract, 1 g of poly(dI-dC)⅐poly(dI-dC), and 0.5 ng of 32 P-labeled double-stranded oligonucleotide in 10 mM HEPES, pH 8.0, 15% glycerol, 2 mM EDTA, 0.5 mM spermidine, 20 mM NaCl, 4 mM MgCl 2 , 2 mM dithiothreitol. The assay mixture was incubated at room temperature for 15 min in a final volume of 25 l. After the incubation, 5 l of 6 ϫ dye solution (0.1% bromphenol blue, 30% glycerol) was added to the reactions and the samples were immediately loaded and electrophoresed on a nondenaturing 4.5% polyacrylamide gel for 2.5 h at 140 V in 0.5 ϫ TBE (1 ϫ TBE: 89 mM Tris, 89 mM boric acid, and 2.5 mM EDTA). After electrophoresis, gels were dried in a slab gel dryer for 1.5 h at 80°C and autoradiographed on Fuji RX x-ray film.
Immunoprecipitation of AP-1 Binding Factor-Involvement of c-Jun in AP-1 binding activity was determined by immunoprecipitation technique. Affinity purified polyclonal antibody (0.5 g) against c-Jun protein (Oncogene Science) was added to 5 g of nuclear extract and incubated at room temperature for 1 h. After the incubation, 2.5 g of affinity-purified goat anti-rabbit IgG was added to the samples and incubated overnight at 4°C. Protein-antibody complexes were removed by centrifugation in a microcentrifuge at top speed for 10 min, and the supernatant was used in gel mobility shift assays.
Western Blot-Proteins that were separated by SDS-PAGE were transferred onto a nitrocellulose membrane by electroblotting. The transfer was performed at a current of 0.12 A and 30 V for overnight. The membrane was incubated in blocking solution (3% BSA) for 1 h at 25°C, washed, and then incubated with the v-Jun polyclonal antibody (Oncogene Science, 1:130 dilution), the bFGF monoclonal antibody (Santa Cruz Biotech, Santa Cruz, CA, 1:50 dilution), or the protein kinase C (PKC) ␣ polyclonal antibody (Life Technologies, Inc., 1:500 dilution). After incubation with the primary antibody, the membrane was washed, and incubated with alkaline phosphatase-conjugated rabbit anti-mouse IgG or goat anti-rabbit IgG in the diluting solution (1:2,000) for 1-2 h. The membrane was then stained using nitro blue tetrazolium and 5Ј-bromo-4-chloro-3-indolyl phosphate.
Measurement of PKC Activity-Amersham's PKC enzyme assay system was used to determine PKC activity. Cells were mixed with extraction buffer (50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 10 mM EGTA, 0.3% w/v ␤-mercaptoethanol, 50 g/ml phenylmethylsulfonyl fluoride, 1 g/ml leupeptin), sonicated, and then centrifuged. Supernatants were saved and mixed with glycerol for storage. Protein content was determined by the Bradford method (1976). The reaction buffer (calcium buffer, lipid, peptide buffer, and dithiothreitol buffer) was added to the sample (50 -100 g of protein). The reaction was started by adding [␥-32 P]ATP and continued for 15 min at 25°C. The reaction was terminated by adding the stop reagent.
c-jun Gene Transfection-Exponentially growing cells were plated 2 days before experiments at a concentration of 4 ϫ 10 5 cells/60-mm culture dish. Cells were transfected with an expression vector (10 g) containing the c-jun gene (pRSV-c-Jun or pCMV-c-Jun) from Dr. Robert Tjian's laboratory (Baichwal and Tjian, 1990) by calcium phosphate transfection technique and incubated for 12 h at 37°C. After transfection, cells were rinsed with PBS, incubated for 36 h, and then harvested.
Survival Determination-Cell survival was assayed by the colonyforming ability of individual cells to obtain quantitative dose survival curves. Each experiment was repeated four times and plating efficiency was in the range of 59 -77%. Figs. 1A and 2A show the level of c-jun and c-fos mRNA after x-ray exposure in adriamycin-sensitive (MCF-7) and -resistant (MCF-7/ADR) cells. Note that differences in RNA loading were examined by rehybridizing the nitrocellulose membranes with the house keeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe. The level of c-jun and c-fos mRNA increased rapidly and reached a maximal value within 0.25-0.5 h after irradiation with 135 cGy (Fig. 1) or 75 cGy (Fig. 2) in MCF-7/ADR cells. Data from densitometer analysis (Figs. 1B and 2B) show that MCF-7/ADR cells contained a 2-6-fold elevated level of c-jun mRNA 0.5 h after irradiation. Similar results were observed in MCF-7 cells. However, the level of both mRNAs in MCF-7 was lower than that detected in MCF-7/ADR cells ( Figs. 1 and 2).

Radiation-induced c-jun and c-fos Gene Expression-Northern blots in
Radiation-induced AP-1 Binding Activity-Gel mobilityshift assays in Fig. 3 show that AP-1 binding activity to a TPA responsive element (TRE; AP-1 binding site) sequence of the bFGF promoter was enhanced by radiation or TPA. We also observed that the level of radiation-induced AP-1 binding activity in MCF-7/ADR cells was much higher than that of MCF-7 cells. The AP-1 binding activity was shown to be highly specific to TRE in competition assays when the addition of 200 molar excess of unlabeled AP-1 abolished the radioactive signal from 32 P-labeled AP-1 transcription factors and TRE complex (AP-1 in Fig. 3). The specificity of AP-1 binding activity was also shown by failure to bind to unlabeled nonspecific oligonucleotides, e.g. SP1 consensus binding sequence (SP-1 in Fig. 3).
To confirm the presence of c-Jun, which binds to the TRE in the bFGF promoter, anti-c-Jun antibody was used to precipitate AP-1 complexes from nuclear extracts. Protein-antibody complexes were removed by centrifugation. The AP-1 binding activity of nuclear extracts from x-irradiated cells was decreased by adding anti-c-Jun antibody followed by a secondary antibody (Fig. 4). These results confirmed that c-Jun protein is involved in the observed AP-1 binding activity.
Radiation-induced bFGF Gene Expression-Data presented in Fig. 5 show that x-radiation induces bFGF mRNA. Although the bFGF gene is known to be a single copy and is located on human chromosome 4 (Mergia et al., 1986), primer extension analysis indicates that one start site is used to transcribe three sizes of bFGF mRNA (1.4, 3.7, and 7 kb) (Shibata et al., 1991). Our data from Northern blot assays confirmed the presence of three sizes of bFGF mRNA. Radiation elevated the level of bFGF mRNA in MCF-7/ADR cells. However, bFGF gene expression was not detected in MCF-7 cells. This variance may be due to the differences in the level of AP-1 transcription factor, and PKC␣. This hypothesis was tested as described below.
Involvement of c-Jun in the Regulation of bFGF Gene Expression-Data from transfection experiments demonstrated that MCF-7 cells that were transfected with the plasmids pRSV-c-Jun or pCMV-c-Jun containing the human c-jun gene con-tained an elevated level of bFGF protein (Fig. 6). These results were consistent with observations in MCF-7/ADR cells, which contain relatively high levels of c-Jun and bFGF proteins (ADR in Fig. 6). Note that multiple sizes of bFGF protein originate from the use of the AUG codon as well as novel CUG translation initiation codons (Florkiewicz and Sommer, 1989). Consistent results were also obtained from MCF-7/ADR cells transfected with the pCMV-c-Jun DNA Ϫ vector, which contains a DNA binding domain defective c-Jun mutant gene (data not shown). The elevated level of mutant c-Jun protein resulted in the reduction of the AP-1 binding activity and bFGF protein level (data not shown). Fig. 7A demonstrated that the amount of PKC␣ in MCF-7/ADR cells was 6.5-fold higher than that in MCF-7 cells as determined by densitometer. Fig. 7B also showed a 5-7-fold elevation in the level of PKC activity in MCF-7/ADR cells compared to that of MCF-7 cells. The possible The autoradiography of c-jun mRNA was analyzed with a densitometer (panel B). C, RNA from untreated control cells. GAPDH, GAPDH probe was used to verify the equivalent amounts and integrity of RNAs loaded in each lane. Autoradiograms were all from the same blot, which was stripped and rehybridized with different probes.

FIG. 2. Northern blots of c-jun and c-fos mRNA after x-irradiation in MCF-7 or MCF-7/ADR cells (panel A) and quantitative analysis of c-jun mRNA (panel B).
Cells were irradiated with 75 cGy and incubated at 37°C for the intervals (0.25, 0.5, 1.25, 3, 6, and 8 h) indicated at the bottom of each lane. Northern blot analysis was performed as described in Fig. 1. C, RNA from untreated control cells. GAPDH, GAPDH probe hybridized to the same blot to verify RNA uniformity.

FIG. 3. Detection of AP-1 binding activity in nuclear extracts from x-irradiated MCF-7 or MCF-7/ADR cells.
Cells were irradiated with 135 cGy and incubated at 37°C for the intervals indicated at the top of each lane (0.25-3 h). The gel mobility-shift assay was performed with a 32 P-labeled AP-1 oligonucleotide and nuclear extracts (5 g) prepared from irradiated cells. Competition assays were performed by adding a 200-fold molar excess of an unlabeled AP-1 or an unlabeled SP1 oligonucleotide. C, nuclear extracts from untreated control cells. TPA, nuclear extracts from cells treated with 1 g/ml TPA for 1 h. Closed arrow indicates AP-1 binding activity. Open arrow indicates free 32 P-labeled oligonucleotide fragment (FREE). involvement of PKC in the regulation of bFGF gene expression was studied by employing H-7, a potent PKC inhibitor (Figs. 8 -10) or treatment with PKC␣ antisense oligonucleotides (Fig. 11).
The effect of H-7 on c-jun gene expression, AP-1 binding activity, and bFGF gene expression was investigated in MCF-7/ADR cells (Figs. 8 -10). H-7 (60 g/ml) markedly suppressed the levels of c-jun, whereas the drug did not affect the GAPDH mRNA level (H-7 in Fig. 8). Data from gel mobility-shift assay demonstrated that treatment with H-7 (60 g/ml) significantly suppressed the radiation-induced AP-1 binding activity (lane D in Fig. 9). In contrast, HA1004 (60 g/ml), an H-7 analogue that is a potent inhibitor of cAMP and cGMP-dependent protein kinases and a weak inhibitor of PKC, did not suppress the radiation-induced increase in the AP-1 binding activity (lane E in Fig. 9). Northern blot and quantitative analysis in Fig. 10 demonstrated a 2-fold or more reduction of the 7 kb bFGF mRNA level in the presence of 60 g/ml H-7. To confirm the role of PKC␣ in the regulation of bFGF gene expression, MCF-7 cells were transfected with plasmid pSV 2 M(2)6 containing bovine PKC␣ cDNA (Fig. 11). As compared to either parental MCF-7 or MCF-7-vector cells, PKC␣ was overexpressed ap-proximately 5-fold in MCF-7-PKC␣ cells as determined by densitometric analysis (Fig. 11A). Interestingly, MCF-7-PKC␣ cells displayed bFGF gene expression as detected by Northern FIG. 4. Immunospecific inactivation of the AP-1 binding activity in nuclear extracts of MCF-7/ADR cells by anti-c-Jun antibody. Nuclear extracts (5 g) from control (odd numbers) and x-irradiated (135 cGy) cells (even numbers) were reacted with 0.5 g of polyclonal antibody developed against c-Jun (lanes 5, 6, 9, and 10) and precipitated with 2.5 g of goat anti-rabbit 2nd antibody (lanes 9 and 10). Nuclear extracts were used in the assay without 4°C incubation (lanes 1 and 2), with 4°C overnight incubation (lanes 3-10), or with the addition of 2nd antibody only (lanes 7 and 8). Gel mobility-shift assay was performed as described in Fig. 3. Closed arrow indicates AP-1 binding activity. Open arrow indicates free 32 P-labeled oligonucleotide fragment (FREE).  H-7, cells were treated with 60 g/ml H-7 for 1 h before, during, and after x-irradiation. Northern blot analysis was performed as described in Fig. 1. C, RNA from untreated control cells. GAPDH, internal standard GAPDH mRNA. See legend of Fig. 1 for further details. blot analysis (lane C of MCF-7-PKC␣ in Fig. 11B). Elevated levels of c-Jun and c-Fos proteins were also observed in MCF-7-PKC␣ cells (data not shown). Moreover, the level of bFGF mRNA increased after irradiation with 135 cGy in MCF-7-PKC␣ cells.
Effect of H-7 on Radiation-induced Cytotoxicity-Several studies have shown that bFGF protects endothelial cells against the lethal effects of ionizing radiation (Haimovitz-Friedman et al., 1991;Fuks et al., 1994). The radioprotective effect of bFGF is shown to be mediated by PKC activation and it can be abrogated by H-7 (Haimovitz-Friedman et al., 1994). Thus, our results raised a potentially important question. Given the large differences in the level of PKC␣ and the response of AP-1 factors and bFGF to ionizing radiation between MCF-7 and MCF-7/ADR cell lines, are MCF-7/ADR cells resistant to the cytotoxic effects of ionizing radiation? Second, can the effect be blocked by a PKC inhibitor? Data from Fig. 12A showed that there was no inherent differences in radiosensitivity between MCF-7 and MCF-7/ADR cells as displayed by cell survival curves. In addition, survival curves were still essentially superimposable for all radiation doses in the absence/presence of H-7 in both cell lines (Fig. 12, B and C). At the present time, only speculation can be made concerning these discrepancies. The lack of radioprotection in MCF-7/ADR cells suggests that extracellular bFGF rather than intracellular bFGF may have an important role in the radioprotection.

DISCUSSION
Our data from Figs. 1-3 demonstrated that ionizing radiation activated the expression of jun and fos genes and increased AP-1 binding activity in human breast cancer cells, particularly multidrug-resistant MCF-7/ADR cells. Our observations are consistent with results obtained in human epithelial cells (Hallahan et al., 1991a) and human sarcoma cell line RIT-3 (Hallahan et al., 1993). Several researchers have also reported that UV radiation or H 2 O 2 treatment stimulates the expression of both c-jun and c-fos, as well as AP-1 binding activity in HeLa S3 cells (Devary et al., 1991). Transcriptional activation of the c-jun gene is exerted through the distal and proximal AP-1 binding sites (TRE) in its promoter (Angel et al., 1988). In vitro DNA binding and in vivo footprinting experiments demonstrated that TRE is recognized by either c-Jun homodimers (Angel et al., 1988: Deng andKarin, 1993) or c-Jun/activating transcription factor-2 heterodimer (van Dam et al., 1993, Herr et al., 1994. The transcriptional activity of these binding factors is stimulated by the phosphorylation of c-Jun (Binetruy et FIG. 9. Effect of H-7 or HA1004 on the radiation-induced AP-1 binding activity in MCF-7/ADR cells. A, untreated control cells. B, cells were irradiated with 135 cGy and nucleoproteins were extracted immediately after irradiation. C, cells were irradiated and nucleoproteins were extracted 0.5 h after irradiation. D, cells were treated with 60 g/ml H-7 for 1 h before, during, and after irradiation. Nucleoproteins were extracted 0.5 h after irradiation. E, cells were treated with 60 g/ml HA1004 for 1 h before, during, and after irradiation. Nucleoproteins were extracted 0.5 h after irradiation. The gel mobility-shift assay was performed as described in Fig. 3. Closed arrow indicates AP-1 binding activity.  , 1991;Devary et al., 1992;Pulverer et al., 1991;Smeal et al., 1991Smeal et al., , 1992 and possibly activating transcription factor-2 (Abdel-Hafiz et al., 1992). Studies from Datta et al. (1992) suggest that x-ray-induced transcriptional activation of the c-jun gene is mediated through the formation of reactive oxygen intermediates and PKC is involved in the signal pathway. Our data from Fig. 8 and Hallahan et al. (1991a) demonstrated that treatment with isoquinolinesulfonamide derivative H-7, a potent PKC inhibitor, prior to radiation attenuated the increase in c-jun transcripts. The transcriptional regulation of c-fos is not identical to that of c-jun. The induction of c-fos transcription is mediated by two major cis-elements: the serum response element (SRE) and the v-sis conditioned medium induction element. The SRE is recognized and stably bound by a homodimer of serum response factor (Treisman, 1986). The binary SRE-serum response factor complex interacts with another factor, the ternary complex factor (TCF) (Treisman, 1992). The TCF, which belongs to the family of E26 transformation-specific (Ets) proteins (Dalton and Treisman, 1992), is homologous to Elk-1 (Hipskind et al., 1991). The activity of TCF is rapidly increased in response to cell stimulation with various agents, such as growth factors, which lead to mitogen-activated protein kinase activation. Mitogen-activated protein kinase, also known as the extracellular signal-regulated protein kinase (ERK) can be activated by UV and x-irradiation (Radler-Pohl et al., 1993;Stevenson et al., 1994). It appears to be responsible for the phosphorylation of TCF (Gille et al., 1992;Marais et al., 1993) and subsequently its transcriptional activity (Zinck et al., 1993). The activity of v--sis-inducible factor, which binds to serum induction element is also enhanced by phosphorylation. However, it is regulated by a different signaling pathway: a cytoplasmic protein tyrosine kinase (Nordheim et al., 1994).
Stress-induced AP-1 binding activity is due to the posttranslational modification of both newly synthesized and preexisting Jun and Fos proteins (Boyle et al., 1991;Smeal et al., 1991). c-Jun contains three domains: a DNA binding domain, a transcription activation domain, and a regulatory domain. In the unstimulated state, c-Jun is constitutively phosphorylated by glycogen synthase kinase-3 or casein kinase II at serines (Ser-243 and Ser-249) and threonines (Thr-231 and Thr-239) close to its C-terminal DNA binding domain. Phosphorylation in this region markedly reduces the DNA binding and transcription ability of c-Jun (Boyle et al., 1991). Dephosphorylation in this region occurs upon stresses such as phorbol ester tumor promoter TPA treatment (Boyle et al., 1991) and UV irradiation (Devary et al., 1992) and results in enhanced AP-1 binding activity. Boyle et al. (1991) reported that TPA-activated PKC is responsible for site-specific dephosphorylation of c-Jun. It has been known that PKC can be activated by stress such as heat shock (Wooten, 1991) and ionizing radiation (Hallahan et al., 1991b). Our data (Fig. 7) and Lee et al. (1992) showed that MCF-7/ADR cells contain markedly elevated amounts of PKC␣. These observations suggest that differences in the level of PKC␣ and the level of AP-1 transcription factors (c-Jun and c-Fos) may be responsible for the differential effects of stress on various cell types.
Several researchers have suggested that PKC mediates stress-induced bFGF gene expression. Tumor promoters such as phorbol 12-myristate 13-acetate, mezerein, and phorbol 12,13-didecanoate activate PKC. These promoters induce the accumulation of bFGF mRNA and its protein in human dermal fibroblasts (Winkles et al., 1992). The enhancement of bFGF gene expression by these tumor promoters is reduced by treatment with H-7. These results are consistent with our observations, which demonstrate the reduction of x-ray-induced bFGF gene expression by H-7 (Fig. 10). Although the dose of H-7 (60 g/ml) we use is relatively specific for PKC, we cannot rule out H-7 as a general kinase inhibitor. Nonetheless, these results and data from Fig. 11 strongly indicate that the x-ray-induced bFGF gene expression is likely mediated through activation of PKC. Data from Figs. 3, 5, 9, and 10 also show a good correlation between the AP-1 binding activity and an increase in bFGF gene expression. Moreover, Fig. 6 shows that bFGF level was elevated in cells which were transfected with plasmids containing human c-jun cDNA. Taken together, our data have indicated that the enhancement of bFGF gene expression is related to an increase in PKC and AP-1 binding activity after ionizing radiation exposure.
We believe that many critical questions still remain to be answered to understand the mechanisms of regulation of bFGF gene expression after x-irradiation. However, our proposed model will provide important information to understand how environmental stresses induce bFGF gene expression and subsequently lead to tumor angiogenesis. This model will also provide a framework to study the critical steps in tumor development and metastasis.