EGR-1 forms a complex with YAP-1 and upregulates Bax expression in irradiated prostate carcinoma cells

Abstract

In this study, we investigated the functional role of early growth response-1 (Egr1 gene) in the regulation of radiation-induced clonogenic inhibition and apoptosis in p53 wild-type and mutant prostate cancer cells 22Rv1 and DU145, respectively. 22Rv1 cells were more sensitive to irradiation compared with DU145 cells, and the sensitivity was enhanced by overexpression of EGR-1 in both cells. Dominant-negative EGR-1 mutant (dnEGR-1) or repressor of EGR-1, NGFIA binding protein 1 (NAB1), increased radioresistance of these cells. Significant activation of caspases 3 and 9 and Bcl2-associated X (Bax) with increased poly(ADP-ribose) polymerase (PARP) cleavage and cytochrome c release was observed in radiation-exposed EGR-1 overexpressing cells. Gel shift analysis and chloramphenicol acetyl transferase (CAT) reporter assays indicate that EGR-1 transactivates the promoter of the Bax gene. Interaction of EGR-1 and Yes kinase-associated protein 1 (YAP-1) through the WW domain of YAP-1 enhances the transcriptional activity of EGR-1 on the Bax promoter as shown by chromatin immunoprecipitation and reporter assays. Irradiation of PC3 cell xenografts that were treated with adenoviral EGR-1 showed significant regression in tumor volume. These findings establish the radiation-induced pro-apoptotic action of EGR-1, in a p53-independent manner, by directly transactivating Bax, and prove that alters the B-cell CLL/lymphoma 2 (Bcl-2)/Bax ratio as one of the mechanisms resulting in significant activation of caspases, leading to cell death through the novel interaction of EGR-1 with YAP-1.

Introduction

Loss of the apoptotic response to ionizing irradiation (IR) by overexpression of pro-survival or downregulation of pro-apoptotic genes has been linked to a radioresistant phenotype (Kyprianou, 1994). In this study, we selected the Egr-1 gene to understand its role in radiation resistance, because (a) upon orchiectomy Egr-1 was rapidly induced, leading to apoptosis of androgen-dependent prostate cells (Buttyan et al., 1988; Day et al., 1993); (b) IR upregulates EGR-1 expression (Datta et al., 1992); and (c) EGR-1 protein upregulates pro-apoptotic gene expression (Ahmed et al., 1997; Nair et al., 1997; Das et al., 2001).

Alterations in p53 function has been attributed to increased radioresistance (Lee and Bernstein, 1993). EGR-1 protein upregulates p53 in response to thapsigargin-induced apoptotic stimuli (Nair et al., 1997). Expression of dominant-negative EGR-1 mutant (dnEGR-1) in melanoma cells (wt p53) caused radio resistance (Ahmed et al., 1996). Despite the absence of p53 protein, the cells overexpressing EGR-1 protein were sensitive to IR, which was associated with upregulation of tumor necrosis factor-α protein through EGR-1-mediated transactivation (Ahmed et al., 1997). In addition, Egr-1−/− mouse embryonic fibroblast cells were highly radioresistant as compared with Egr-1+/− mouse embryonic fibroblast cells, where the radiation-induced pro-apoptotic function of EGR-1 was solely mediated by the target genes such as p53 and retinoblastoma (Das et al., 2001). Thus, EGR-1 plays an important role in the regulation of radiation-induced apoptosis through p53, retinoblastoma and tumor necrosis factor-α target genes.

Yes kinase-associated protein-1 (YAP-1), a transcriptional co-activator of p73, is a 65-kDa protein, also referred to as YAP-65. YAP-1, originally identified as an interaction partner of Src family tyrosine kinase, Yes (Sudol, 1994), contains one WW domain, a PDZ-interacting motif, SH3-binding motif and a coiled-coil domain. A differentially spliced isoform of YAP-1, with two WW domains known as YAP-2 also exists (Sudol et al., 1995). Phosphorylation of YAP by Akt or Lats1 attenuates the p73-mediated transcriptional activity and apoptosis (Basu et al., 2003; Oka et al., 2008). Whether EGR-1, like p73, is regulated through a similar interaction is not known.

Understanding the role played by radiation-induced EGR-1 in apoptosis could be crucial in designing therapy for prostate tumor control. We investigated the function of Egr-1 in radiation-induced apoptosis using wt-p53 (22Rv1), mutant p53 (DU145), P53 null (PC3) prostate cancer cell line and PC3 xenografts in mice. Overexpression of EGR-1 protein transactivates Bax causing an imbalance in the Bcl-2/Bax protein ratio and rendering enhanced susceptibility to radiation-induced apoptosis. Regulation of Bax expression by the interaction of EGR-1 with YAP-1 is suggestive of a novel mode of regulation of EGR-1 targets that could play a crucial role in the apoptosis mediated by EGR-1.

Results

Overexpression of EGR-1 increases sensitivity to radiation-induced clonogenic inhibition and apoptosis

Radiation-induced activity of EGR-1 was minimal in both DU145 cells (3%) and 22Rv1 (8%) cells. Overexpression of EGR-1 by transfection or adenoviral infection in DU145 and 22Rv1 cells, respectively, significantly induced the EBS (EGR-1 binding site)-CAT activity in the untreated and irradiated cells (7.5- and 10.7-fold (DU145 cells), and 4- and 4.5-fold (22Rv1 cells) (Figure 1a). DU145 and 22Rv1 cells show induction of EGR-1 protein by 5 Gy radiation as early as 15–30 min. The protein expression correlated with the EBS-CAT activity with more activity in 22Rv1 cells than in DU145 cells (Supplementary Figure 1a, Figure 1a). Although EGR-1 expression is more in 22Rv1 cells compared with DU145 cells, the functional activity is not significantly different (Figure 1a, Supplementary Figure 1a and c). The 22Rv1 cells were found to be more radiosensitive compared with the DU145 cells by clonogenic assay (Supplementary Figure 1b). As basal EGR-1 transactivation function of EGR-1 was low in DU145 cells (Supplementary Figure 1c), ectopic overexpression of EGR-1 was achieved by stable transfections with CMV EGR-1 (a 2.5-fold increase in protein expression, Supplementary Figure 2a). Further, infection with Ad-EGR-1 in 22Rv1 cells causing a nine-fold increase in EGR-1 protein expression (Supplementary Figure 2b) was also used. Overexpression of NGFIA binding protein 1 (NAB1), an EGR-1 repressor, reduced the endogenous EGR-1 in 22Rv1 cells, as EGR-1 regulates it own transcription (Supplementary Figure 2b). Overexpression or repression of EGR-1, respectively, led to an increased or decreased EBS-CAT and EGR-1-binding activities (electrophoreticmobility shift assay) (Supplementary Figure 1c and 2c).

Figure 1
figure1

Ectopic early growth response-1 (EGR-1) expression radiosensitizes prostate carcinoma cells. (a) Overexpression of EGR-1 in DU145 and 22Rv1 cells induces EBS-CAT reporter activity that is further enhanced by radiation. DU145 and 22Rv1 cells were transiently transfected with EBS-CAT reporter activity. Protein extracts from these cells were used for CAT assay. Transfections with gro-CAT or gro-CAT plus IL1 were used as positive controls. Relative percent CAT activity is represented as histograms. (b) Dose–response curves of DU145 cells stably transfected with vector alone, dnEGR-1 or EGR-1 as assessed by clonogenic inhibition assay. SF2 value for DU145 cells transfected with vector alone was found to be 0.609 with a D0 value of 400 cGy. SF2 value for dominant-negative EGR-1 mutant (dnEGR-1)-transfected cells was 0.66 with a D0 value of 509 cGy. SF2 value for EGR-1 transfected cells was 0.34 with a D0 value of 164 cGy. (c) Dose–response curves of 22Rv1 cells infected with Ad-GFP, Ad-NAB1 or Ad-EGR-1 as assessed by clonogenic inhibition assay. SF2 value for 22Rv1 cells infected with Ad-GFP was found to be 0.45 with a D0 value of 270 cGy. SF2 value for Ad-NAB1-infected cells was 0.63 with a D0 value of 375 cGy. SF2 value for Ad-EGR-1-infected cells was 0.29 with a D0 value of 151 cGy.

Using colony-forming assay, SF2 and D0 values of irradiated DU145/Vector cells were 0.609 and 400 cGy, respectively (Figure 1b). On the other hand, the SF2 value for dnEGR-1 was 0.66 with the D0 value of 509 cGy, suggesting that the cells transfected with the dnEGR-1 were more radioresistant (P<0.001). The SF2 value for CMV-EGR-1 was 0.34 with the D0 value of 164 cGy, suggesting enhanced radiation sensitivity (P<0.0001) with EGR-1 overexpression (Figure 1).

Similarly, terminal deoxynucleotidyl transferase nick end labeling (TUNEL) and flow cytometric analysis indicated that DU145/dnEGR-1 cells were resistant (P<0.001) and DU145/CMV-EGR-1 cells were significantly sensitive (P<0.0001) to IR-inducible apoptosis (Supplementary Figure 3a and b).

In 22Rv1 cells, the SF2 value of cells infected with Ad-green fluorescent protein (Ad-GFP) was 0.45 with a D0 value of 270 cGy (Figure 1c). On the other hand, the SF2 value for Ad-NAB1-infected cells was 0.63 with the D0 value of 375 cGy, suggesting that the cells infected with the EGR-1 repressor NAB1, were more radioresistant (P<0.001). Overexpression of EGR-1 (by Ad-EGR-1) showed an SF2 value of 0.29 with the D0 value of 151 cGy, suggesting enhanced radiosensitivity (P<0.0001). TUNEL assay and flow cytometric analysis also indicated that Ad-NAB1-infected cells were radioresistant (P<0.001) and Ad-EGR-1-infected cells were significantly more radiosensitive (P<0.0001) (Supplementary Figure 3c and d). Together these findings suggest that EGR-1 is required for IR-induced apoptosis, and overexpressed EGR-1 potentiates the effects of IR.

EGR-1 induces the expression of BAX

On the basis of above results, we hypothesized that the pro-apoptotic proteins might have been altered by EGR-1. A significant upregulation of Bax protein in both untreated and irradiated DU145/CMV-EGR-1 cells with a slight reduction in the Bcl-2 protein level in EGR-1-overexpressing cells was observed (Figure 2a). Significant induction of Bax mRNA was observed in cells overexpressing EGR-1 as compared with that of cells transfected with dnEGR-1 or vector alone in both the untreated and irradiated cells (Figure 2b). Similar results were obtained with 22Rv1 cells, suggesting a role for EGR-1 in regulating the expression of Bax (Figure 2c). In order to show that expression of EGR-1 leads to activation of Bax, oligomerization and activation of Bax were performed in 22Rv1 cells. Overexpression and/or irradiation lead to increased Bax oligomerization, suggesting an activation of Bax without any change in the levels of β-actin used as an internal control (Figure 2d). As Bax affects mitochondrial permeability, causing the release of cytochrome c, leading to caspase activation (Tsujimoto, 1998), specific activity of caspases 3 and 9 were analyzed at different time intervals after radiation. DU145 cells transfected with EGR-1 showed elevated caspases 3 and 9 activity (Supplementary Figure 4a and b). Western blot analysis of inactive form of caspase 3 shows a decrease with time in irradiated cells transfected with EGR-1. Pro-apoptotic events mediated by EGR-1 are further reflected in the cleavage of 116-kDa poly(ADP-ribose) polymerase (PARP) (Supplementary Figure 4c). Overexpression of EGR-1 by Ad-EGR-1 caused an increase in the release of cytochrome c levels in cytosolic versus mitochondrial fractions in untreated and irradiated cells. In contrast, overexpression of NAB1 caused a decreased cytochrome c release (Supplementary Figure 4d). Promoter analysis of Bax gene indicated two putative overlapping Egr-1-binding sites from −21 to −13 (IndexTermGCGGCGGCG) and from −15 to −7 (IndexTermGCGGGAGCG) with one base variation when compared with the consensus Egr-1-binding site (Figure 3a). To ascertain whether these sites could bind to EGR-1 protein, we performed electrophoreticmobility shift assay and CAT reporter assays (Figure 3a). Electrophoreticmobility shift assay indicated that radio-labeled EGR-1 consensus oligo probe (from Bax promoter sequence) was competitively reduced with cold probe. Furthermore, EGR-1 binding to probe was reduced in supershift assays (Figure 3b).

Figure 2
figure2

Ectopic early growth response-1 (EGR-1) induces the expression of Bax protein in prostate carcinoma cells. (a) Western analysis. DU145 cells transfected with vector, dnEGR-1 or EGR-1 were either left untreated (UT) or treated with 5 Gy X-ray irradiation. Protein extracts were prepared from these cells after 1, 2, 3, 6 and 12 h post irradiation. Blots of these proteins were probed for expressions of Bax, Bcl2 and β-actin. (b) Reverse transcriptase-PCR analysis. DU145 cells transfected with vector, dnEGR-1 or EGR-1 were either untreated or treated with 5 Gy X-ray irradiation. Total RNA was prepared from these cells 1, 3 and 6 h post irradiation. The mRNA levels of Bax and G6PDH as a function of EGR-1 activity are shown. (c) Western analysis. 22Rv1 cells infected with Ad-GFP, Ad-NAB1 or Ad-EGR-1 were either left untreated (UT) or treated with 5 Gy X-ray irradiation. Protein extracts were prepared from these cells 1, 3, 6 and 24 h post irradiation. Blots of these proteins were probed for expressions of Bax, Bcl2 and β-actin. (d) Oligomerization of BAX. PC3 cells transfected with vector or EGR-1 were either untreated or irradiated with 5 Gy radiation and the cells were crosslinked with bis-(sulfosuccinimidyl) suberate and disuccinimidyl suberate (Pierce, Rockford, IL, USA) as per the method of Antonsson et al. (2001). Whole lysates were prepared in RIPA buffer and separated by non-reducing SDS–PAGE and blotted onto polyvinylidene fluoride (PVDF) membrane. The membrane was probed using anti-Bax antibody.

Figure 3
figure3

Bax is a direct target of early growth response-1 (EGR-1). (a) Diagrammatic representation of Bax promoter in a reporter vector. Bax promoter used in our studies shows binding sites for EGR-1 (EBS1 and EBS2). TSS indicates transcription start site. TATA box and p53 binding sites are shown. Reporter construct deleted for EGR-1-binding sites is represented by sequence between −823 and −56. (b) EMSA showing binding of EGR-1 to BAX promoter oligos. Nuclear extracts from DU145 cells stably overexpressing EGR-1 were used. The arrow indicates bound complex. Competition suggests competition of hot probe by the cold probe. Supershift was performed by incubating the EGR-1 probe mixture with antibody against EGR-1. Control shows shift assay with probe and protein complex alone. (c) CAT reporter assay in DU145 cells. DU145 cells were transiently transfected with different combinations of plasmid outlined in the figure. CAT activity in each transfection was reported as percent conversion in the figure. (d) CAT reporter assay in 22Rv1 cells. 22Rv1 cells infected with Ad-GFP, Ad-NAB1 or Ad-EGR-1 were transiently transfected with different combinations of plasmid outlined in the figure. CAT activity in each transfection was reported as percent conversion in the figure.

DU145 cells showed a modest background level of CAT activity (Figure 3c) that increased significantly with EGR-1 overexpression. On the other hand, CAT activity induced by EGR-1 was reduced by dnEGR-1, suggesting that EGR-1 transactivates, and the dnEGR-1 transrepresses, the Bax promoter. EBS-deleted pBax-CAT construct showed attenuation of basal expression driven by the promoter, and ectopically expressed EGR-1 caused no induction of CAT activity (Figure 3c). Similarly, overexpression of EGR-1 significantly induced CAT activity in 22Rv1 cells. On the other hand, significant reduction in CAT activity was observed with NAB1. The pΔBax-CAT vector could not be induced with Ad-EGR-1 (Figure 3d). Thus, these findings suggest that EGR-1 transactivates the Bax through the EGR-1-binding site in its promoter.

WW domain-binding motif PPxY regulates the transactivation function of EGR-1

p300/CBP proteins bind to transcription factors through their cysteine- and histidine-rich (C/H) domains to form active transcriptional complexes (Kalkhoven, 2004). A potential acetylation of KDKK motif in EGR-1 is required for its interaction with p300 (Matheny et al., 1994). YAP-1, a WW domain-containing protein, by binding to p73 recruits p300 to p73 target genes, leading to the induction of these genes (Basu et al., 2003). We reasoned that a similar mechanism might be responsible for recruitment of YAP-1 and p300 to the EGR-1-binding sites. Analysis of the primary sequence of EGR-1 revealed the presence of a WW domain-interacting PPxY motif (235YPPPAY240) in EGR-1 (Figure 4a). To find out whether the PPxY motif is critical in the activation of EGR-1, EGR-1 (Y240A) mutant was constructed. Aromatic amino acid residue (Y235) present at the N terminus of the PPxY motifs, which is a crucial determinant of the WW domain-binding activity, was also mutated (Macias et al., 2002). Both the mutations (Y235,240A) lead to a decrease in the EBS-CAT reporter activity in PC3 cells, suggesting that this motif is required for maximal transcriptional activation of EGR-1 (Figure 4b). Approximately 18 and 39% reduction in EGR-1 activity was observed with EGR-1 (Y240A) and EGR-1 (Y235,240A) mutations, respectively. An average of 24 and 38% reduction of activity was observed in irradiated cells (Figure 4b). Similar reduction in the EBS-GFP-positive cells was observed in PC3 cells transfected with EGR-1 (Y240A) (48.5%) and EGR-1 (Y235,240A) (45.5%) mutations (Figure 4c). Dramatic reduction in the BAX-CAT reporter activity was observed in 22Rv1 cells transfected with EGR-1 (Y240A) (76%) and EGR-1 (Y235,240A) (86%) mutations (Figure 4d).

Figure 4
figure4

WW domain-binding motif PPxY is important regulator of early growth response-1 (EGR-1) activity. (a) Diagrammatic representation of PPxY motif in EGR-1. PPxY motif in EGR-1 is present in the repression module II of EGR-1. Residues 237–240 represent the PPxY motif (PPAY). Aromatic residues at the N terminus are known to have an effect on binding. Therefore, the mutants that were generated include EGR-1 Y240A and EGR-1 Y235-240A. (b and c) PPxY motifs are essential for the transactivation function of EGR-1 in PC3 cells. Both the EGR-1 mutants EGR-1 Y240A and EGR-1 Y235-240A were found to possess reduced activity as assessed by EBS-CAT (b) as well as EBS-GFP (c) reporter activity. (d) PPxY motifs are essential for the activity of EGR-1 in 22Rv1 cells. Both the EGR-1 mutants EGR-1 Y240A and EGR-1 Y235-240A were found to possess reduced activity as assessed by BAX-CAT reporter activity. (e) Yes kinase-associated protein 1 (YAP-1) binding to EGR-1 is required for the clonogenic cell death mediated by EGR-1. Colony-forming assay indicates that YAP-1 binding to EGR-1 through PPxY motif is required for EGR-1-mediated clonogenic cell death. EGR-1 (wild type) significantly reduces cell survival, which is reversed with EGR-1 mutant lacking YAP-1-binding motifs.

YAP-1 binding to EGR-1 is required for the clonogenic cell death mediated by EGR-1

PC3 cells stably transfected with pCB6 EGR-1 or EGR-1 mutants (EGR-1 Y240A; EGR-1 Y235,240A) were used for clonogenic assay. EGR-1 was found to decrease cell survival with SF2 value of 0.154±0.0018 when compared with vector alone (0.292±0.0049). This was reversed when YAP-1-binding mutants of EGR-1 were used. EGR-1 Y240A-transfected cells showed SF2 value of 0.21±0.008. Mutation of EGR-1 at Y235,240A was more effective as the cells transfected with EGR-1 Y235,240A shoed SF2 value of 0.25±0.008 (Figure 4e). YAP-1 siRNA caused pronounced cell survival in both untreated and irradiated PC3 cells as measured by real-time cell electronic signal system (data not shown).

Protein array of WW, SH3 and Tudor domains using PPxY motif identifies YAP-1 as the protein interacting with EGR-1

Owing to a significant role of PPxY motifs in EGR-1 activity, we reasoned that EGR-1 might be interacting with a WW domain-containing protein. Glutathione-S-transferase (GST) fusion proteins harboring the listed WW, SH3 and Tudor domains were arrayed (Espejo and Bedford, 2004). Biotinylated peptide motifs encompassing the PPxY motifs in EGR-1 or with mutants were used to identify the proteins interacting with these motifs. The wild-type peptides showed strong interaction with WW domain of YAP and this interaction was weak or absent with the mutant peptides. Interaction of the biotinylated peptides with the WW domain-containing proteins such as Itck and RSP II were also detected. Other domain-containing proteins including the SH3 domain or Tudor domain-containing fusion proteins did not show any interaction (Figure 5a).

Figure 5
figure5

Early growth response-1 (EGR-1) interacts with Yes kinase-associated protein 1 (YAP-1) through WW domain-interacting PPxY motif. (a) WW protein array detects YAP-1 as binding partner of EGR-1. WW, SH3 and TUDOR domain-containing GFP fusion protein array was probed with biotinylated PPxY encompassing motif from the EGR-1 protein. This motif was either wild type or mutant with aromatic tyrosine residues mutated to alanine. WW protein, YAP-1, was found to interact with the PPxY motif from EGR-1. This interaction was reduced or absent in mutants. (b) EGR-1 interacts with YAP-1. Extracts from PC3 cells were reciprocally immunoprecipitated with both EGR-1 or YAP-1 antibodies and the presence of YAP-1 and EGR-1 proteins, respectively, in these precipitates was analyzed by Western immunoblot analysis. Results show specific interaction of EGR-1 with YAP-1. (c) EGR-1–YAP-1 complex is present at the BAX promoter. Chromatin immunoprecipitation showed enrichment of BAX promoter in both the EGR-1 and YAP-1 ChIPs in the untreated and as well as in the irradiated PC3 cells. Irradiated EGR-1 ChIP showed more enrichment compared with untreated sample. Lack of GAPDH enrichment in ChIPs suggests the specificity of the EGR-1/YAP-1 localization at the BAX promoter.

EGR-1 interacts with YAP-1 through its PPxY motif

Reciprocal immunoprecipitation was carried out to confirm the interaction between EGR-1 and YAP-1 seen in the above mentioned GST-fusion protein array. PC3 cell protein extracts were immunoprecipitated using anti-EGR-1 or anti-YAP-1 and the immunoprecipitates were probed with anti-YAP-1 or anti-EGR-1, respectively. Immunoprecipitation of YAP-1 with EGR-1 antibody and EGR-1 with YAP-1 antibody is suggestive of interaction of these proteins. Furthermore, a negative control, constituting an immunoprecipitation using nonspecific rabbit IgG, did not show the presence of either EGR-1 or YAP-1 (Figure 5b). To show that these proteins are engaged on the BAX promoter as a complex, chromatin immunoprecipitation was carried out using anti-EGR-1 and anti-YAP antibodies on a nucleoprotein extract. PCR amplified the BAX promoter using the DNA immunoprecipitated by these antibodies, indicating enrichment of BAX promoter but not the nonspecific GAPDH gene suggestive of specific interaction of EGR-1 and YAP-1 at the BAX promoter (Figure 5c).

Radiation induces the regression of tumor volumes of PC-3 xenografts infected with Ad-EGR-1

To find out whether our in vitro findings translate in animal models of prostate cancer, PC-3 xenografts developed in nude mice were exposed to different treatment regimens outlined in Materials and methods section. Tumors treated with Ad-EGR-1 plus radiation showed a significant regression in tumor volume compared with those treated with Ad-GFP or Ad-EGR-1 or Ad-GFP plus radiation or radiation alone. No significant differences were observed between untreated and Ad-GFP-infected tumors, suggesting that adenoviral infection alone cannot change tumor volume. Interestingly, infection with Ad-EGR-1 alone significantly increased tumor volume as compared with radiation alone or Ad-GFP plus radiation or Ad-EGR-1 plus radiation, suggesting that in the absence of radiation, EGR-1 might potentially act as a pro-survival protein (Figure 6).

Figure 6
figure6

Radiation sensitizes Ad-EGR-1-infected PC3 xenograft tumor cells leading to regression of tumor volume. Six different groups were used in these studies. Tumor volumes (represented by Y axis) were calculated at different time intervals (represented on the X axis) by the Attria and Weiss formula.

Immunohistochemical analysis of the PC3 xenograft tissue sections showed induction of Bax with both radiation or, Ad-EGR-1 alone. Elevated levels of Bax were found in Ad-EGR-1-infected tumors treated with radiation (Supplementary Figure 5). Thus, the tumor regression in response to combinatorial therapy involving ectopic EGR-1 and radiation could be mediated by Bax and the downstream apoptotic machinery.

Discussion

Expression of EGR-1 positively correlates with the Gleason score, exclusively in prostate tumors (Eid et al., 1998). EGR-1 overexpression is correlated with the loss of its co-repressor NAB2 in primary prostate carcinoma causing elevated EGR-1 transcriptional activity (Abdulkadir et al., 2001a). Egr-1−/− TRAMP mice showed significant delay in prostate tumor formation compared with Egr-1+/+ TRAMP mice, suggesting that the loss of EGR-1 failed to prevent tumor initiation, but delayed the progression of tumors (Abdulkadir et al., 2001b). On the contrary, EGR-1 can suppress growth when overexpressed or re-expressed in transformed cells. EGR-1 can induce apoptosis either by stimulating p53 activity or by IR (Ahmed et al., 1996, 1997), or PTEN expression (Virolle et al., 2001). Thus, it appears that EGR-1 can act as a tumor suppressor in some cells and as a growth stimulator in other cell types.

Regulation of p73 activity by acetylation of p73 and interaction with p300 resembles that of EGR-1 activity. YAP-1 functions as a co-activator of p73 (Basu et al., 2003). The role of YAP-1 in the process of tumorigenesis remains unclear. The unique feature of EGR-1 is its radio-inducibility, which would make it distinct from the p73. Zender et al. (2006) reported that YAP-1 is amplified in hepatocarcinoma and cooperates with Myc in transformation, implying that YAP-1 is an oncoprotein. Similarly, Dong et al. (2007) suggested that YAP-1 plays a role in size control mechanism in Drosophila and mammals . In contrast, other studies have shown that YAP-1 has antitumoral properties (Basu et al., 2003; Matallanas et al., 2007; Strano and Blandino, 2007). Furthermore, YAP-1 mRNA upregulation correlates with improved survival in breast cancer patients (Matallanas et al., 2007). Although it appears that YAP-1 may play a crucial role in the process of tumorigenesis, it may play an opposite role during radiation enabling the process of EGR-1-mediated apoptosis. EGR-1 mutants defective in binding to YAP-1 showed decrease in three BAX reporters tested (Figures 4b–d). We clearly show that the radiation-induced pro-apoptotic function of EGR-1 leading to clonogenic cell death is mediated through its interaction with YAP-1 (Figure 4e). Furthermore, physical interaction of EGR-1 and YAP-1 as shown by WW domain array, immunoprecipitation and ChIP (Figures 5b and c) suggests a role for EGR-1/YAP-1 complex in the regulation of BAX gene. We also found that Puma, a pro-apoptotic protein, was regulated by EGR-1/YAP-1 (Reeves A and Ahmed MM. EGR-1 mediated upregulation PUMA is regulated by PPxY motif in the transactivation domain, unpublished results). This suggests that in addition to Bax activation, other mechanisms such as transactivation of Puma may mediate radiation-induced apoptotic function of EGR-1. In addition, marginal reductions in the levels of Bcl2 in cells expressing ectopic EGR-1 could be because of inhibition of nuclear factor-κB by binding of p65 subunit with EGR-1. The induction of Bax expression by EGR-1 with YAP-1 functioning as co-activator in irradiated cells, as shown in this study, is a significant step in understanding the pro-apoptotic mode of EGR-1 function during radiotherapy. Although overexpression of EGR-1 by adenoviral EGR-1 therapy increased in tumor volume, overexpression of EGR-1 with X-ray irradiation significantly regressed the tumor, suggesting that modulation of appropriate regulators of EGR-1, including YAP-1, by DNA-damaging agents such as IR could be useful in tumor control.

It was not clear how EGR-1 could mediate an apoptotic response in PC3 cells despite the lack of p53 protein in our earlier studies. Our present studies using two different prostate cancer cell lines, 22Rv1 (p53 wt) and DU145 (p53 mutant), show that the apoptotic function could be mediated directly by EGR-1 through the transcriptional upregulation of Bax mRNA, protein, and increased oligomerization and activation. It is likely that p73 with YAP-1 may also contribute to the transactivation of Bax. However, the radio-inducibility of EGR-1 makes it unique for mediating this function. Thus, it appears that IR primes EGR-1 for an important role in mediating apoptosis, independent of p53 function, through a novel mode of regulation by YAP-1

Materials and methods

Plasmid constructs

The plasmid CMV-EGR-1 encoding full-length EGR-1 protein and dominant-negative EGR-1 was described earlier (Ahmed et al., 1997). EGR-1 mutants, EGR-1Y240A and EGR-1Y235,240A, were constructed by site-directed mutagenesis. The reporter construct, EBS-CAT, contains three EGR-1 binding sites (IndexTermCGCCCCCGC) placed in tandem upstream of a minimal c-fos promoter and CAT cDNA. The full-length Bax promoter region was PCR amplified from human genomic DNA, using sense primer (5′-IndexTermTTTTTCTAGACAGCACAGATTAGTTTCTGCCAC-3′, −959 to −937) and the antisense primer (5′-IndexTermTTTTAAGCTTCACCGCCGCTCCCGCCGC-3′, −18 to −1) containing built-in sites (underlined) for XbaI and HindIII restriction sites, respectively. The Bax promoter without EGR-1-binding site was constructed using the sense primer (5′-IndexTermTTTTTCTAGAATGCTTGAGTCTGGGAGTTCA-3′, −823 to −807) and the antisense primer (5′-IndexTermTTTTAAGCTTCCCGGGTCACGTGAGAGC-3′ −73 to −56). The full-length Bax promoter (pBax-CAT) with or without EBS was cloned in the pG-CAT reporter plasmid. The EBS-GFP reporter plasmid was constructed by subcloning the promoter sequence into the AseI–NheI cloning sites. Adenoviral construct expressing EGR-1 was a gift from Dr Jeffrey Milbrandt (Svaren et al., 2000).

Cell Culture

22Rv1 cells were grown in RPMI 1640 medium with 10% fetal bovine serum and 1% penicillin–streptomycin at 37 °C and 5% CO2. DU145 (mutant p53) and PC-3 cells (p53 null) were grown in α-MEM and RPMI, respectively, supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin at 37 °C and 5% CO2. Stable CMV-EGR-1 and dnEGR-1 were selected using G418 sulfate (400 μg/ml). 22Rv1 cells were infected with Ad5/GFP (referred as Ad-GFP), Ad5/GFP-EGR-1 (referred as Ad-EGR-1) and Ad5/GFP-NAB1 (referred as Ad-NAB1) at a viral titer of 1 × 108 plaque-forming units/ml for 2 h.

DNA transfection and CAT assays

Stable transfections were performed using Lipofectamine plus reagent (Gibco BRL, Carlsbad, CA, USA). Transient transfections were performed by the calcium phosphate co-precipitation method or by effectene transfection reagent (Qiagen, Valencia, CA, USA). CAT assays were performed as described earlier (Ahmed et al., 1997).

Irradiation

A 100 kV industrial X-ray machine (Phillips, Amsterdam, The Netherlands) was used to irradiate the cultures. The dose rate with a 2-mm aluminum plus 1-mm beryllium filter was 2.64 Gy/min at a focus-surface distance of 10 cm.

Colony-forming assay

Prostate carcinoma, DU145 and 22Rv1 cells, transiently or stably transfected or infected cells were either left untreated or treated with 1–6 Gy. Stably transfected or infected DU145 and 22Rv1 cells were treated with 5 Gy. PC3 cells were left untreated or treated with 2 Gy radiation.

EBS-GFP reporter assay

PC3 cells transfected with EBS-GFP and EGR-1-WT or EGR-1-Y240A or EGR-1-Y235,240A or empty vector were either untreated or irradiated with 5 Gy dose, and the expression of GFP was observed 24 h later using Zeiss Axiovert 35 fluorescent microscope. The percentage of GFP-positive cells was determined.

Quantitation of apoptosis

Apoptosis was quantified by TUNEL staining and flow cytometry. The ApopTag in situ apoptosis detection kit (Oncor, Gaithersburg, MD, USA) was used for apoptosis as described earlier (Ahmed et al., 1997). To determine the percentage of cells showing apoptosis, approximately 1000 cells were counted in each of the four experiments performed. For flow cytometry, cells were dissociated using non-enzymatic cell dissociation medium (Sigma, St Louis, MO, USA), washed with phosphate-buffered saline, stained with Hoechst (Ho342) and merocyanine (MC540) and analyzed by flow cytometry using a FACStar Plus cell sorter (Ahmed et al., 1997).

Cytochrome C release assay

The 22Rv1 cells were infected with Ad-GFP, Ad-EGR-1 or Ad-NAB1 as described in earlier section for 24 h. The infected cells were then irradiated with 5 Gy. Post-irradiated cells (at 3 and 6 h) were then fractionated for mitochondrial and cytosolic fractions using a kit and instructions (BioVision, Mountain view, CA, USA). After protein estimation using Bradford's dye, samples were prepared for Western immunoblot analysis.

Western blot analysis

Total protein extracts from untreated and irradiated cells at various time intervals were subjected to western blot analysis using rabbit anti-EGR-1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit anti-Bax (Santa Cruz Biotechnology), rabbit anti-Bcl-2 (Santa Cruz Biotechnology), mouse anti-caspase 3 (Sigma, St Louis, MO, USA), mouse anti-PARP antibody (Sigma), rabbit anti-GRP75 (mitochondrial matrix-specific protein) (Santa Cruz Biotechnology), mouse anti-cytochrome c (BioVision) and mouse anti-β-actin antibody (Sigma). Secondary goat anti-mouse IgG and anti-rabbit IgG (Amersham, Piscataway, NJ, USA). The bound immune complexes were detected using chemiluminescence method.

Colorimetric assay for caspases 3 and 9 activities

Activity of caspases 3 and 9 were assayed colorimetrically using DEVD-pNA and LEHD-pNA substrates, respectively. Briefly, untreated and irradiated DU145-transfected cells were lysed in 50 μl of chilled Cell Lysis Buffer (BioVision, Mountain View, CA, USA) and incubated on ice for 10 min. The suspension was centrifuged for 1 min at 10 000 g and supernate (50 μl) was incubated with 50 μl of 2 × reaction buffer and 5 μl of 4 mM LEHD-pNA or DEVD-pNA for 1 h at 37 °C. The activity was measured at 405 nm, and the specific activity of the caspases was calculated using the molar extinction coefficient of pNA (8800/M/cm).

Electrophoretic mobility shift assay

Electrophoretic mobility shift assay was performed using Nushift kit (Geneka Biotechnology, Montreal, Quebec, Canada) and nuclear lysates from untreated and irradiated DU145 cells. The consensus oligo probe that contains Egr-1-binding site (underlined) is 5′-IndexTermGGATCCAGCGGGGGCGAGCGGGGGCGAACG-3′. For supershift experiments, rabbit EGR-1 polyclonal antibody was incubated with binding buffer and nuclear extract for 20 min at 10 °C prior to adding oligo probe. Binding reactions were electrophoresed on a 5% polyacrylamide gel in 1 × Tris-glycine electrophoresis buffer.

Generation of WW, SH3, Tudor domain protein array, peptide synthesis and labeling

GST fusion proteins harboring the listed WW, SH3 and Tudor domains were arrayed (Espejo and Bedford, 2004). The following peptides encompassing the PPxY motif from EGR-1 and its mutants conjugated to biotin label were synthesized (Biosynthesis Inc., Lewisville, TX, USA): biotin-TALQYPPPAYPAAK, biotin-TALQAPPPAYPAAK and biotin-TALQAPPPAAPAAK. Biotinylated peptides (10 attogram) were labeled as described earlier (Espejo and Bedford, 2004). A GenePix 4200A scanner (Axon Inc., Sunnyvale, CA, USA) was used for array analysis.

Reciprocal co-immunoprecipitation of EGR-1 and YAP-1

PC3 cell (80% confluent) lysates in RIPA buffer were incubated with protein A beads to preclear the lysate, which was incubated with anti-EGR-1, -YAP-1 or nonspecific rabbit IgG (at the ratio of 1:500 of IgG to precleared lysate) followed by incubation with protein A beads for 4 h at 4 °C. The beads were washed with RIPA buffer followed by washes with 62.5 mM Tris-HCl (pH 6.8). Elution was performed by adding SDS–polyacrylamide gel electrophoresis sample buffer and boiling at 95 °C for 1 min.

Chromatin immunoprecipitation

Chromatin immunoprecipitation assay was described earlier (Shareef et al., 2007) using 2–5 × 109 PC3 cells. PCR was performed using specific primers for BAX promoter (5′-IndexTermGCCTGGGCAACACAGTGAG-3′ and 5′-IndexTermGCTCCCTCGGGAGGTTTG-3′) that yielded 500 bp product. HG6PD gene served as a negative control.

Isolation of RNA and reverse transcriptase-PCR

Total RNA extracted using TRIZOL reagent (Life Technologies, Foster City, CA, USA) was reverse-transcribed into cDNA. Specific primers for BAX gene (5′-IndexTermCCAGCTCTGAGCAGATCATGAAG-3′ and 5′-IndexTermGCAATCATCCTCTGCAGCTCCAT-3′) were used for reverse transcriptase-PCR. Primers for GAPDH gene were used as internal controls.

PC-3 xenografts

Pathogen-free athymic Balb/c nude mice (6–8 weeks old; Harlan Sprague Dawley, Indianapolis, IN, USA), which were handled according to the guidelines stipulated in ‘Guide for the care and use of laboratory animals’ (DHHS publication NIH-85-23), were used. PC3 cells (5 × 106) were injected subcutaneously on the medial side of thigh just above the stifle. The injected animals were monitored twice a week for subcutaneous tumors till the tumor volume reached at least 0.5 cm3, when the animals were randomized into six groups. The tumor volumes were calculated by the Attria and Weiss formula: a2 × b2 × 0.4, where ‘a’ is the smallest and ‘b’ the largest tumor diameter (measured by calipers). Group I—parental xenografts, no treatment; group II—parental xenografts, 20 Gy of total IR dose; group III—Ad/GFP-treated xenografts; group IV—Ad/GFP-EGR-1-treated xenografts; group V—Ad/GFP-treated xenografts with 20 Gy total dose of IR; and group VI—Ad/GFP-EGR-1-treated xenografts with 20 Gy total dose of IR. Ad/GFP or Ad/GFP-EGR-1 constructs were injected into the tumor at a dose of 100 MOI (multiplicity of infection) (3 × 108 PFU) at the first day of treatment followed by 2 Gy radiation dose per day to the tumor for 5 days. On 7th and 8th days, intra-tumoral injections of Ad/GFP or Ad/GFP-EGR-1 constructs were repeated followed by 2 Gy dose of radiation for 5 days. On the 14th day, the animals received the final 100 MOI of Ad/GFP or Ad/GFP-EGR-1 and were observed for next 7 weeks.

PC-3 xenografts and tissue harvest for paraffin embedding

Tissue, of one mice from each group euthanized on the 1st, 7th, 14th, 21st, 28th, 35th and 40th day of treatment, was fixed in 5% buffered formalin and embedded in paraffin. Paraffin sections were subjected to Bax immunohistochemistry using the avidin–biotin–peroxidase complex method.

Statistical analysis

For in vitro experiments, simple t-test was used. Animal groups were compared using the two-tailed Fisher's exact test.

References

  1. Abdulkadir SA, Carbone JM, Naughton CK, Humphrey PA, Catalona WJ, Milbrandt J . (2001a). Frequent and early loss of the EGR1 corepressor NAB2 in human prostate carcinoma. Hum Pathol 32: 935–939.

    CAS  Article  Google Scholar 

  2. Abdulkadir SA, Qu Z, Garabedian E, Song SK, Peters TJ, Svaren J et al. (2001b). Impaired prostate tumorigenesis in Egr1-deficient mice. Nat Med 7: 101–107.

    CAS  Article  Google Scholar 

  3. Ahmed MM, Sells SF, Venkatasubbarao K, Fruitwala SM, Muthukkumar S, Harp C et al. (1997). Ionizing radiation-inducible apoptosis in the absence of p53 linked to transcription factor EGR-1. J Biol Chem 272: 33056–33061.

    CAS  Article  Google Scholar 

  4. Ahmed MM, Venkatasubbarao K, Fruitwala SM, Muthukkumar S, Wood Jr DP, Sells SF et al. (1996). EGR-1 induction is required for maximal radiosensitivity in A375-C6 melanoma cells. J Biol Chem 271: 29231–29237.

    CAS  Article  Google Scholar 

  5. Antonsson B, Montessuit S, Sanchez B, Martinou JC . (2001) Bax is present as a high molecular weight oligomer/complex in the mitochondrial membrane of apoptotic cells. J Biol Chem 276: 11615–11623.

    CAS  Article  Google Scholar 

  6. Basu S, Totty NF, Irwin MS, Sudol M, Downward J . (2003). Akt phosphorylates the Yes-associated protein, YAP, to induce interaction with 14-3-3 and attenuation of p73-mediated apoptosis. Mol Cell 11: 11–23.

    CAS  Article  Google Scholar 

  7. Buttyan R, Zakeri Z, Lockshin R, Wolgemuth D . (1988). Cascade induction of c-fos, c-myc, and heat shock 70K transcripts during regression of the rat ventral prostate gland. Mol Endocrinol 2: 650–657.

    CAS  Article  Google Scholar 

  8. Das A, Chendil D, Dey S, Mohiuddin M, Mohiuddin M, Milbrandt J et al. (2001). Ionizing radiation down-regulates p53 protein in primary Egr-1−/− mouse embryonic fibroblast cells causing enhanced resistance to apoptosis. J Biol Chem 276: 3279–3286.

    CAS  Article  Google Scholar 

  9. Datta R, Rubin E, Sukhatme V, Qureshi S, Hallahan D, Weichselbaum RR et al. (1992). Ionizing radiation activates transcription of the EGR1 gene via CArG elements. Proc Natl Acad Sci USA 89: 10149–10153.

    CAS  Article  Google Scholar 

  10. Day ML, Wu S, Basler JW . (1993). Prostatic nerve growth factor inducible A gene binds a novel element in the retinoblastoma gene promoter. Cancer Res 53: 5597–5599.

    CAS  PubMed  Google Scholar 

  11. Dong J, Feldmann G, Huang J, Wu S, Zhang N, Comerford SA et al. (2007). Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell 130: 1120–1133.

    CAS  Article  Google Scholar 

  12. Eid MA, Kumar MV, Iczkowski KA, Bostwick DG, Tindall DJ . (1998). Expression of early growth response genes in human prostate cancer. Cancer Res 58: 2461–2468.

    CAS  Google Scholar 

  13. Espejo A, Bedford MT . (2004). Protein-domain microarrays. Methods Mol Biol 264: 173–181.

    CAS  PubMed  Google Scholar 

  14. Kalkhoven E . (2004). CBP and p300: HATs for different occasions. Biochem Pharmacol 68: 1145–1155.

    CAS  Article  Google Scholar 

  15. Kyprianou N . (1994). Apoptosis: therapeutic significance in the treatment of androgen-dependent and androgen-independent prostate cancer. World J Urol 12: 299–303.

    CAS  Article  Google Scholar 

  16. Lee JM, Bernstein A . (1993). p53 mutations increase resistance to ionizing radiation. Proc Natl Acad Sci USA 90: 5742–5746.

    CAS  Article  Google Scholar 

  17. Macias MJ, Wiesner S, Sudol M . (2002). WW and SH3 domains, two different scaffolds to recognize proline-rich ligands. FEBS Lett 513: 30–37.

    CAS  Article  Google Scholar 

  18. Matallanas D, Romano D, Yee K, Meissl K, Kucerova L, Piazzolla D et al. (2007). RASSF1A elicits apoptosis through an MST2 pathway directing proapoptotic transcription by the p73 tumor suppressor protein. Mol Cell 27: 962–975.

    CAS  Article  Google Scholar 

  19. Matheny C, Day ML, Milbrandt J . (1994). The nuclear localization signal of NGFI-A is located within the zinc finger DNA binding domain. J Biol Chem 269: 8176–8181.

    CAS  PubMed  Google Scholar 

  20. Nair P, Muthukkumar S, Sells SF, Han SS, Sukhatme VP, Rangnekar VM . (1997). Early growth response-1-dependent apoptosis is mediated by p53. J Biol Chem 272: 20131–20138.

    CAS  Article  Google Scholar 

  21. Oka T, Mazack V, Sudol M . (2008). Mst2 and Lats kinases regulate apoptotic function of Yes kinase-associated protein (YAP). J Biol Chem 283: 27534–27546.

    CAS  Article  Google Scholar 

  22. Shareef MM, Cui N, Burikhanov R, Gupta S, Satishkumar S, Shajahan S et al. (2007). Role of tumor necrosis factor-alpha and TRAIL in high-dose radiation-induced bystander signaling in lung adenocarcinoma. Cancer Res 67: 11811–11820.

    CAS  Article  Google Scholar 

  23. Strano S, Blandino G . (2007). YAP1 meets tumor suppression. Mol Cell 27: 863–864.

    CAS  Article  Google Scholar 

  24. Sudol M . (1994). Yes-associated protein (YAP65) is a proline-rich phosphoprotein that binds to the SH3 domain of the Yes proto-oncogene product. Oncogene 9: 2145–2152.

    CAS  PubMed  PubMed Central  Google Scholar 

  25. Sudol M, Bork P, Einbond A, Kastury K, Druck T, Negrini M et al. (1995). Characterization of the mammalian YAP (Yes-associated protein) gene and its role in defining a novel protein module, the WW domain. J Biol Chem 270: 14733–14741.

    CAS  Article  Google Scholar 

  26. Svaren J, Ehrig T, Abdulkadir SA, Ehrengruber MU, Watson MA, Milbrandt J . (2000). EGR1 target genes in prostate carcinoma cells identified by microarray analysis. J Biol Chem 275: 38524–38531.

    CAS  Article  Google Scholar 

  27. Tsujimoto Y . (1998). Role of Bcl-2 family proteins in apoptosis: apoptosomes or mitochondria? Genes Cells 3: 697–707.

    CAS  Article  Google Scholar 

  28. Virolle T, Adamson ED, Baron V, Birle D, Mercola D, Mustelin T et al. (2001). The Egr-1 transcription factor directly activates PTEN during irradiation-induced signalling. Nat Cell Biol 3: 1124–1128.

    CAS  Article  Google Scholar 

  29. Zender L, Spector MS, Xue W, Flemming P, Cordon-Cardo C, Silke J et al. (2006). Identification and validation of oncogenes in liver cancer using an integrative oncogenomic approach. Cell 125: 1253–1267.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work was supported by grants, NIH (RO1 CA78471), PCRP-DOD (PC 970526) and Charlotte Geyer foundation to MMA; post-doctoral training grant, PCRP-DOD (WB1XWH-04-1-0039) to MZ; NIH (DK62345) and Pennsylvania Department of Health Grant (SAP4100037378) to MMS and Welch Foundation Grant (G-1495) to MTB.

Author information

Affiliations

Authors

Corresponding author

Correspondence to M M Ahmed.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Zagurovskaya, M., Shareef, M., Das, A. et al. EGR-1 forms a complex with YAP-1 and upregulates Bax expression in irradiated prostate carcinoma cells. Oncogene 28, 1121–1131 (2009). https://doi.org/10.1038/onc.2008.461

Download citation

Keywords

  • EGR-1
  • YAP-1
  • Bax
  • prostate cancer
  • radiation

Further reading

Search