Function of Polo-like Kinase 3 in NF-κB-mediated Proapoptotic Response*

RelA, the p65 subunit of NF-κB transcription factors, plays a key role in regulation of antiapoptotic and proapoptotic responses. However, the downstream target genes regulated by RelA-NF-κB in the initiation of proapoptotic signaling were not identified. We previously showed that RelA-NF-κB functioned as a proapoptotic factor by activating the p53-signaling pathway in response to doxycycline-induced superoxide. In the present study, we demonstrate that the ability of doxycycline/superoxide to induce expression of polo-like kinase 3 (Plk3) depends on NF-κB activity. We identified a κB binding site in the promoter of Plk3, and this κB site is directly involved in its induction by the RelA-NF-κB complex. Plk3 formed a complex with p53 and was involved in the phosphorylation of p53 on Ser-20 in response to superoxide. Inhibition of Plk3 expression by Plk3 small interfering RNA suppressed the doxycycline/superoxide-mediated apoptosis. Overexpression of wild-type Plk3 in HCT116 p53+/+ cells induced rapid apoptosis, whereas overexpression of wild-type Plk3 in HCT116 p53–/– cells and the kinase-defective mutant Plk3K91R in p53+/+ cells induced delayed onset of apoptosis. Furthermore, mutagenesis of Plk3 showed that the N-terminal domain (amino acids 1–26) is essential for the induction of delay onset of apoptosis. These data show that Plk3 is a RelA-NF-κB-regulated gene that induces apoptosis in both p53-dependent and -independent signaling pathways, suggesting a possible mechanism for RelA-NF-κB-regulated proapoptotic responses.

RelA, the p65 subunit of the NF-B transcription factor, plays a key role in protecting cells from proapoptotic stimuli (1)(2)(3). Many studies have shown that proapoptotic signals can induce NF-B, which in turn induces the expression of the genes involved in suppressing apoptotic signals (4 -6). However, a proapoptotic aspect of RelA activity has also been reported; recently, NF-B was shown to induce cell death after after the engagement of T-cell receptor or exposure to DNAdamaging agents (7)(8)(9). Other reports have shown that NF-B activation is required for the onset of apoptosis induced by alphavirus or kainic acid (10,11). The opposing roles of NF-B in regulating apoptosis have also been shown within individual cells. For example, in the same cells in which inhibition of NF-B promotes the induction of apoptosis by glucocorticoids, NF-B is required for the induction of apoptosis by stimulation of phorbol ester and ionomycin for mimicking T-cell activation (12). These findings suggest that whether the function of NF-B is proapoptotic or antiapoptotic in a given cell depends on the cell type, extent of NF-B activation, and nature of the apoptotic signals.
Doxycycline, like chemotherapeutic agents, radiation, cytokines, and other apoptosis inducers, causes structural and morphologic damage to the cell, such as cell shrinkage, chromatin condensation, and DNA fragmentation (13)(14)(15). Kroon and colleagues (16,17) showed that tetracycline preferentially inhibited mitochondrial protein synthesis in cancer cells, including cytochrome c oxidase, the key component of the electron transport chain. Our previous report showed that doxycycline induces superoxide formation, suggesting that decreased synthesis of mitochondrial protein disrupts electron transport function and results in electron leakage from the respiratory chain to O 2 , thus elevating levels of superoxide radicals (18). However, the molecular mechanism by which doxycycline induces apoptosis is not yet completely understood.
Both NF-B and p53 are activated in response to many stimuli, including DNA damage and oxidative stress (19,20). Reactive oxygen species-induced phosphorylation of p53 on Ser-20 is mediated in part by polo-like kinase 3 (Plk3) 1 (21), a member of a conserved family of serine/threonine protein kinases that are important in cell cycle regulation and stress response signaling. Phosphorylation of p53 on Ser-15 and Ser-20 plays an important role in the p53-mediated apoptotic pathway (22). Our previous report showed that NF-B-dependent phosphorylation of p53 on Ser-20 is an important event in initiating doxycycline-induced apoptosis (18). However, the proapoptotic downstream target genes induced by NF-B have not been identified. We hypothesized that one of the NF-Bregulated genes that encode the key proteins involved in the phosphorylation of p53 on Ser-20 initiates apoptosis in response to an increased level of superoxide. In this report, we show that the expression of Plk3 is induced by doxycycline and that this induction depends on the activation of NF-B. Plk3 is involved in the phosphorylation of p53 on Ser-20 in response to doxycycline stimulation. Overexpression of wild-type Plk3, but not the N terminus-truncated Plk3 (⌬1-141), induced rapid apoptosis in p53 ϩ/ϩ cells, whereas apoptosis was delayed in p53 Ϫ/Ϫ cells expressing wild-type Plk3 and in p53 ϩ/ϩ cells expressing the kinase-defective mutant Plk3 K91R . Small interfering RNA (siRNA)-mediated inhibition of Plk3 expression suggests that Plk3 is essential for superoxide-induced cell death, and deletion analysis of Plk3 showed that the N-terminal domain (amino acids 1-26) is essential for induction of delayed onset of apoptosis. Taken together, our results suggest that Plk3 is an NF-B-regulated kinase that mediates p53-dependent and -independent proapoptotic response in reaction to elevated levels of superoxide.
Transient Transfection and Luciferase Assays-The Plk3 promoter firefly-and TK-Renilla-luciferase reporter gene plasmids were transfected into the HEK293, MDAPanc28, and MDAPanc-28/IB␣M and WT and RelA-null MEF cell lines by Fugene 6 (Roche Applied Science) or HCT116 and HCT116p53 Ϫ/Ϫ cells with the Lipofectamine methods (Invitrogen) as previously described (26). The activity of both the firefly and the Renilla luciferase, without or with stimulation with doxycycline (50 g/ml), was determined at 24 and 48 h; the experiments were performed in triplicate. The firefly luciferase activities were normalized to the activity of the Renilla luciferase, which served as an internal control.
siRNA Assay-The Plk3 siRNA SMARTpool and nonspecific pooled control (Dharmacon) were transfected into the indicated cells using TransIT-TKO transfection reagent according to the manufacturer's protocol. Twenty-four hours after siRNA transfection, cells were treated with 50 g/ml doxycycline for the indicated times and subjected to subsequent analyses.
Immunoprecipitation and in Vitro Kinase Assay-The protein extracts (1 mg) were subjected to immunoprecipitation with 200 ng of anti-FLAG (M2) antibody linked to agarose beads at 4°C overnight, followed by Western blotting with anti-p53 and anti-Ser-20-phosphorylated p53 antibodies. For the kinase assay, the anti-FLAG (M2) immunoprecipitates were incubated with purified p53 protein (1-342 aa) in 20 l of kinase buffer as previously described (27,31).
DNA Fragmentation-The level of DNA fragmentation in apoptotic cells was determined by gel electrophoresis. The cells were washed with phosphate-buffered saline and incubated with DNA fragment isolation buffer (10 mM Tris-HCl, 20 mM EDTA, 0.5% Triton X-100) for 30 min on ice. The extracts were centrifuged to remove the cell debris, and the low molecular weight DNA fragments were extracted from the supernatant with phenol/chloroform, precipitated with an equal volume of 2-propanol. The DNA pellets were washed with 75% ethanol, resuspended in TE buffer, and analyzed by separation on a 1.5% agarose gel.
Immunocytochemistry and Fluorescence Microscopy-HEK293 cells were seeded onto coverslips and transfected with EGFP-PLK3 or -PLK3 ⌬N fusion construct or control vector using Fugene 6 (Roche Applied Science). Twenty-four hours posttransfection, the cells were fixed with 4% paraformaldehyde in phosphate-buffered saline followed by treatment with 0.2% Triton X-100 in phosphate-buffered saline. Immunostaining of ␥-tubulin was performed using a Cy3conjugate anti-␥-tubulin antibody (Sigma). The coverslips were mounted with the aqueous mounting medium ProLong Gold antifade reagent with 4Ј,6-diamidino-2-phenylindole (Molecular Probes, Inc., Eugene, OR) and analyzed by fluorescence microscopy. The images were captured on the Metamorph imaging system (Universal Imaging Corp., Downingtown, PA).
Colony Formation Assay-Vector control plasmid and wild-type and mutated Plk3 expression plasmids were transfected into HEK293 cells by Fugene 6 (Roche Applied Science), or HCT116 p53 ϩ/ϩ and HCT116p53 Ϫ/Ϫ cells were transfected by Lipofectamine 2000 (Invitrogen). After a 24-h transfection, cells were selected with 500 g/ml G418 for colony formation. Colonies were washed with phosphate-buffered saline and stained with Coomassie Blue G-250. The number of colonies was directly counted. Alternatively, stained colonies were lysed with 1% SDS in phosphate-buffered saline. Colony lysates were diluted in 0.1% SDS, and the A 615 value was measured.
Chromatin Immunoprecipitation Assay-The chromatin immunoprecipitation assay was performed with a chromatin immunoprecipitation assay kit (Upstate Biotechnology, Inc., Lake Placid, NY) as previously described (18). Two PCR primers, 5Ј-TGCAATTCCCAGCCAGGCAA-G-3Ј and 5Ј-TGCAATTCCCAGCCAGGCAAG-3Ј, were used to amplify a 176-bp fragment, which corresponded to the predicted NF-B enhancer region in the Plk3 promoter.

Doxycycline-induced Plk3 Expression Is NF-B-dependent-
EMSAs, immunoblotting for IB␣ degradation, RT-PCR, and Northern blot analyses for Plk3 expression were performed using MDAPanc28/Puro and MDAPanc28/IB␣M cells stimulated with doxycycline (50 g/ml) at the indicated times ( Fig.  1). Within 24 h after doxycycline stimulation, both NF-B DNA binding activity (Fig. 1a, lanes 1-10) and IB␣ degradation (Fig. 1b, lanes 1-8) were detected in the MDAPanc28/Puro cells but not the MDAPanc28/IB␣M cells. The doxycycline-induced expression of Plk3 mRNA in MDAPanc28/Puro cells increased with the time of treatment and reached its maximum level by 48 h (Fig. 1, c and d, lanes 1-4), whereas doxycycline-induced Plk3 expression in MDAPanc28/IB␣M cells was inhibited (Fig. 1, c and d, lanes [5][6][7][8]. Quantification of the induction of Plk3 expression is shown in Fig. 1e. These results suggest that Plk3 is a target gene regulated by NF-B. Plk3 Expression Is Induced by Various Stimuli-To determine whether Plk3 expression is induced by other NF-B inducers in different cells, we stimulated HEK293/Puro and HEK293/IB␣M cells with doxycycline (50 g/ml), MDAPanc-28/Puro, and MDAPanc-28/IB␣M with TNF-␣ (10 ng/ml) and NIH3T3 cells with fibroblast growth factor-1 (20 ng/ml) and phorbol 12-myristate 13-acetate (30 g/ml) for various times, as indicated, and isolated total RNA for measuring the levels of Plk3 expression (Fig. 2, a-c). Within 12 h of doxycycline stimulation of HEK293/Puro cells, the expression of Plk3 was induced and reached a maximal level at 48 h of doxycycline stimulation (Fig. 2a, lanes 1-4). In contrast, the expression of Plk3 was not induced in the doxycycline-stimulated HEK293/ IB␣M cells (Fig. 2a). These results are consistent with the doxycycline-induced Plk3 expression in MDAPanc-28 cells in Fig. 1, c and d. In MDAPanc-28/Puro, but not in MDAPanc-28/ IB␣M cells, the expression of Plk3 was induced by TNF-␣ stimulation. Interestingly, the TNF-␣-induced Plk3 expression appears to be biphasic and occurred at 1 and 24 h of TNF-␣stimulation (Fig. 2b). This is consistent with the previous reports demonstrating biphasic NF-B activation (27,32,33). Fig. 2c shows that the expression of Plk3 is also induced by fibroblast growth factor-1 and phorbol 12-myristate 13-acetate. These results suggest that the expression of Plk3 is induced by various NF-B inducers in different cell lines.
Plk3 Promoter Is Inducible With NF-B-To further investigate how NF-B mediates Plk3 induction by doxycycline stimulation in MDAPanc28 cells, we next cloned the promoter (1.8 kb) of the human Plk3 gene from the BAC clone RP11-269F19 (accession number AL592166) from human chromosome 1, which contains the whole genomic sequence of Plk3. Sequence analysis revealed that the Plk3 promoter has a GC-rich region located from bp Ϫ78 to Ϫ328 upstream of the translation initiation site. Of all of the nine reported transcripts found in the nucleotide and expressed sequence tag sequence data bases at NCBI, there are three mRNAs (accession numbers BC013899, AJ293866, and NM_004073) and six expressed sequence tags (accession numbers BM921223, BQ065567, BQ066297, BG437825, BF205939, and CD109602), with different length in the 5Ј-untranslated sequence initiated from the Plk3 promoter region (Ϫ63 to Ϫ239 bp). These transcripts were possibly due to the lack of a definite transcription start site, a characteristic of the GC-rich promoter. A B enhancer element (TGGGAGT-TCC) from Ϫ1130 to Ϫ1121 bp was identified, and other previously identified cis-regulatory elements, such as the binding site for basal transcription factor Sp-1, are also present (Fig.  3a). To confirm that the predicted B motif in the Plk3 promoter is a bona fide NF-B binding site, we performed EMSA for NF-B DNA binding activity using 25-bp oligonucleotides with a WT B motif (5Ј-CCTGAGGTTGGGAGTTCCAGAC-CAG-3Ј) and mutant motif (5Ј-CCTGAGGTTAAAAGTTCCA-GACCAG-3Ј). As shown in competition and anti-p65 antibody supershift assays (Fig. 3b), the predicted WT NF-B binding motif, but not the mutant, bound to p65-p50 and p50-p50 NF-B complexes. These data suggest that this is indeed an NF-B DNA-binding site.
To determine whether p65/NF-B binds to the B site identified in the Plk3 promoter in vivo, we performed chromatin immunoprecipitation assays. As shown in Fig. 3c, the Plk3 Twenty micrograms of total RNA isolated from these cell lines at the indicated time points were analyzed using a human Plk3 cDNA probe, and the levels of 28 and 18 S ribosomal RNA were used as controls for RNA loading. c, Western blot analyses for Plk3 expression in NIH3T3 cells stimulated with fibroblast growth factor-1 (20 ng/ml) and phorbol 12-myristate 13-acetate (50 g/ml) for the indicated times. Fifty micrograms of protein extracts were probed with anti-Plk3 antibody. The control for protein loading was determined by using anti-␤-actin antibody.
promoter DNA sequence was amplified from the anti-p65/ NF-B chromatin immunoprecipitates in the PCR using the specific PCR primers flanking the B site in the Plk3 promoter region. The results suggest that p65/NF-B directly interacted with the B site in the Plk3 promoter region in MDAPanc-28/ Puro cells, but this interaction was inhibited in MDAPanc-28/ IB␣M cells. To determine whether the B site identified in the Plk3 promoter is functional, we cloned the 1.8-kb promoter region of Plk3 into the pGL2-basic luciferase reporter vector (Plk3pLucWT) and generated the mutated B site in Plk3 promoter luciferase reporter vector (Plk3pLucMT). The luciferase reporter activity was significantly stimulated by doxycy-cline in MEF cells transfected with the Plk3pLucWT but not with the Plk3pLucMT reporter gene construct (Fig. 3d). Stimulation with doxycycline induced high levels of luciferase activity in MDAPanc28/Puro and MEF cells but not in MDAPanc28/IB␣M or RelA-null MEF cells, transfected with the Plk3pLucWT reporter gene construct (Fig. 3, e and f). These data indicate that the induction of Plk3 gene expression by doxycycline is probably mediated directly by NF-B and the B site present in the Plk3 promoter.
Doxycycline-induced Phosphorylation of p53 on Ser-20 and Expression of p21 waf1 Are Plk3-dependent-We previously showed that the doxycycline-induced phosphorylation of p53 on Ser-20 in MDAPanc28 cells is NF-B-dependent, and doxycycline-induced apoptosis was detected at 48 h in HCT116 p53 ϩ/ϩ cells but significantly delayed in HCT116 p53 Ϫ/Ϫ cells (18). To determine whether Plk3 is involved in the phosphorylation of p53 on Ser-20 after doxycycline stimulation, we inhibited Plk3 expression using Plk3 siRNA (Fig. 4a). Doxycycline-induced phosphorylation of p53 upon Ser-20 and p21 waf1 expression was completely inhibited (Fig. 4, a and b). To confirm the involvement of Plk3 in the phosphorylation of p53, we performed immunoprecipitation assays using FLAG-Plk3-transfected MDAPanc28/Puro cells without and with doxycycline stimulation. As shown in Fig. 4c, doxycycline induced complex formation between Plk3 and p53. The in vitro kinase assay using FLAG-Plk3 and FLAG-Plk3 K91R immunoprecipitated from doxycycline-stimulated cells, with purified p53 protein (aa 1-342) as a substrate, suggests that Plk3 phosphorylated p53 on Ser-20 (Fig. 4d). Taken together, these results suggest that Plk3 is directly involved in the phosphorylation of p53 upon Ser-20 response to doxycycline stimulation.
Plk3 Induces p53-dependent and -independent Apoptosis-Previously, we showed that NF-B functioned as a proapoptotic  factor by activating the p53-signaling pathway in response to superoxide (18). Superoxide-induced apoptosis was significantly delayed in p53-null HCT116 and MDAPanc-28/IB␣M cells, suggesting that NF-B and p53 are involved in superoxide-induced apoptosis in these cells (18). To determine the role of Plk3, a downstream target gene regulated by NF-B, in the proapoptotic signaling cascade, we transfected Plk3 wild type and mutants into HCT116p53 ϩ/ϩ or HCT116p53 Ϫ/Ϫ cells and measured the DNA fragmentation. Consistent with our previous finding (18), superoxide-induced apoptosis was delayed in the absence of p53 function in HCT116p53 Ϫ/Ϫ cells (Fig. 5a). Similarly, Plk3-mediated apoptosis was delayed in HCT116p53 Ϫ/Ϫ cells but not delayed in HCT116p53 ϩ/ϩ cells (Fig. 5b). The apoptosis induced by overexpression of a kinasedefective Plk3 mutant (Plk3 K91R ) (mutated ATP-binding site in the Plk3 catalytic domain) in HCT116p53 ϩ/ϩ cells was also delayed, whereas an N terminus deletion mutant (amino acids 1-141) of Plk3 (Plk3 ⌬N ) failed to induce apoptosis in HCT116p53 ϩ/ϩ cells (Fig. 5c). These results suggest that Plk3induced rapid onset of apoptosis is p53-dependent and that Plk3-mediated delayed onset of apoptosis is p53-independent and that Plk3 kinase activity may be partially involved in the induction of apoptosis.
Plk3-induced Delayed Onset of Apoptosis Is Independent of p53-To further determine the effects of Plk3 on p53-independent cell death, we performed colony formation assays. Overexpression of Plk3 and Plk3 K91R , but not Plk3 ⌬N , in HEK293 cells completely inhibited colony formation (Fig. 6, a and b). Overexpression of Plk3 greatly inhibited colony formation in both HCT116 p53 ϩ/ϩ and HCT116 p53 Ϫ/Ϫ cells (Fig. 6, c and d). The quantification of colony formation was summarized in Fig. 6, e and f. Overexpression of Plk3 in HEK293 cells significantly inhibited cell growth by inducing apoptosis, and overexpression of Plk3 K91R decreased but did not abolish this effect, whereas Plk3 ⌬N totally abolished this effect (Fig. 6, h and i). These results are consistent with those of the colony formation assays (Fig. 6, a-f). Taken together, these results further suggest that overexpression of Plk3 can mediate p53-independent apoptosis.
The N-terminal Domain of Plk3 Is Required for Inducing Apoptosis-To determine the domains in Plk3 that play an essential role in the delayed proapoptotic signaling cascade, we carried out mutagenesis of Plk3. Since the N terminus deletion mutant (amino acids 1-141) of Plk3 (Plk3 ⌬N ) failed to induce apoptosis in HCT116p53 ϩ/ϩ cells (Fig. 5c), we focused our analysis on the N terminus of Plk3. The Plk3 deletion mutants generated were summarized in Fig. 7a. Overexpression of wildtype Plk3 and Plk3 deletion mutant D52-107 inhibited colony formation (Fig. 7, c and d). On the other hand, overexpression of Plk3 deletion mutant D1-26 failed to inhibit colony formation (Fig. 7, b and e). Quantification of the colony formation is shown in Fig. 7f. Taken together, these results suggest that the N-terminal domain of 1-26 amino acid in Plk3 is required for its apoptotic function. Deletion of these 26 amino acids destroys three Src homology 3 domains on Plk3, which in turn may inhibit crucial interactions between Plk3 and other proapoptotic signaling molecules.
Plk3 Is Essential in Doxycycline-induced Apoptosis-To determine whether Plk3 plays an essential role in doxycyclineinduced apoptosis, we inhibited Plk3 expression in doxycyclinestimulated cells using Plk3 siRNA. In both HEK293 and MDAPanc-28/Puro cells, doxycycline-induced-apoptosis was inhibited by the transfection of Plk3 siRNA but not a control siRNA (Fig. 8, a and b). However, no apoptosis was detectable in TNF-␣ (50 ng/ml)-treated MDAPanc-28/Puro cells (Fig. 8c), although Plk3 expression is induced by TNF-␣ (Fig. 2b). Taken together, these data suggest that Plk3 plays a key role in doxycycline-induced apoptosis and that ectopic expression of Plk3 induces apoptosis. Furthermore, the findings imply that whether the function of NF-B is proapoptotic or antiapoptotic may depend on the nature of the stimuli and outcome of the expression of genes in response to such stimuli.
To determine whether the subcellular localization of Plk3 was involved in induction of apoptosis, both EGFP-tagged wild-type Plk3 and the N terminus deletion mutant of Plk3 (Plk3 ⌬N ) were transfected into HEK293 cells and analyzed by fluorescence microscopy. As shown in Fig. 8b, Plk3 mainly localized to cell membrane and a condensed spot in the transfected cell and a much weaker diffuse distribution in the cytoplasm, whereas a very diffused distribution was found in the cells expressing EGFP-Plk3 ⌬N . The control EGFP vector generated a diffused pattern. Double labeling experiments using an anti-␥-tubulin antibody showed that Plk3 is localized around centrosomes in a condensed spot. These subcellular localization patterns suggest that Plk3 may be involved in regulating the microtubule dynamics and that dysregulation of Plk3 may induce mitotic catastrophe. DISCUSSION Transcription factor NF-B can regulate both proapoptotic and antiapoptotic signaling pathways; however, less is known  1-4), and overexpression of Plk3 ⌬N in HCT116p53 ϩ/ϩ cells (lanes 5-8) failed to induce apoptosis. HCT116p53 ϩ/ϩ or HCT116p53 Ϫ/Ϫ cells were transfected with wild-type Plk3, Plk3 K91R , or Plk3 ⌬N expression vector for the indicated times. DNA fragmentation assays were performed, and DNA marker is shown. about the mechanism by which NF-B induces apoptosis. Our results, summarized in Fig. 8c, suggest a possible mechanism by which NF-B regulates proapoptotic-signaling cascades by inducing Plk3 expression, which in turn activates p53-dependent and -independent apoptotic pathways in doxycycline/superoxide-induced cell death.
We previously described a possible role of NF-B signaling cascades in doxycycline/superoxide-induced apoptosis and showed that doxycycline/superoxide-induced apoptosis is inhibited in cells lacking functional NF-B activity and is partly dependent on NF-B-mediated p53 activation (18). In that study, p53 was stabilized primarily by reducing Hdm2 fulllength protein and by inducing a kinase to phosphorylate p53 protein at Ser-20, which in turn induced the expression of its downstream target genes, such as PUMA, for initiating proapoptotic signaling (18). These results suggest that NF-B regulates a kinase that functions as a proapoptotic factor to activate the p53 signaling pathways. Previous studies have shown that Plk3 functionally links DNA damage to cell cycle arrest and apoptosis via the p53 pathway by interacting directly with p53 and phosphorylating p53 on Ser-20 in response to reactive oxygen species and irradiation (21,24,34). Our studies demonstrate the following findings: first, Plk3 is an NF-B downstream target gene in response to doxycycline/ superoxide stimulation; second, the promoter of the Plk3 gene contains a B site that is directly involved in its induction by the NF-B complexes; third, after induction by doxycycline/ superoxide, Plk3 forms a complex with p53 and may be directly involved in the phosphorylation of p53 on Ser-20; fourth, Plk3 plays a key role in doxycycline-induced apoptosis; and fifth, Plk3 induces cell death, possibly by inducing p53-dependent and -independent pathways. Similar to growth factor withdrawal, chemotherapeutic agents, radiation, cytokines, and other apoptosis inducers, doxycycline induces apoptotic cell death with structural and morphologic features such as cell shrinkage, mitochondrial swelling, chromatin condensation, and DNA fragmentation (13)(14)(15). These agents have been evaluated in preclinical cancer models and have entered recent clinical trials in patients with malignant diseases (35)(36)(37). Our previous study showed that doxycycline induced the formation of superoxide, which may in turn induce NF-B activation (18). In this report, we have shown that doxycycline/superoxide-induced apoptosis is partly mediated by phosphorylation of p53 on Ser-20 by Plk3.
A previous study showed that Plk3 acts downstream of both ATM and Chk2 in this important pathway of DNA damage-de-FIG. 6. Plk3-induced delayed onset of apoptosis is p53-independent. a and b, colony formation of HEK293 cells transfected with control expression vector (Vec-CTL), Plk3 N terminus deletion mutant (Plk3 ⌬N ), and kinase-defective Plk3 (Plk3 K91R ). c and d, HCT116 and HCT116p53 Ϫ/Ϫ cells transfected with control expression vector and WT Plk3. Colonies were stained and counted. e, summary of the colony formation counts in HEK293 cells transfected with control, WT, and mutant Plk3 expression vectors. f, summary of the colony formation counts in HCT116 and HCT116p53 Ϫ/Ϫ cells transfected with vector control (Vec-CTL) and WT Plk3 expression vectors. h, HEK293 cells were transfected with control, WT, and mutant Plk3 expression vectors as indicated. Forty-eight hours after transfection, the cells were harvested, and the numbers of living cells were calculated using trypan blue exclusion staining. i, analysis of the effect of Plk3 on apoptosis by DNA fragmentation assay. HEK293 cells were transfected with WT or mutant Plk3 expression vectors as indicated (lanes 1-3). Forty-eight hours after transfection or stimulation, cells were harvested, and DNA fragmentation assays were performed. pendent activation of p53 and that Chk2 in fact stimulates Plk3 activity in response to DNA damage (38). Plk3 plays an important role in the regulation of microtubule dynamics and centrosomal function in the cell, and the deregulated expression of Plk3 results in cell cycle arrest and apoptosis (39). Plk3 has been shown to localize to the cellular cortex and to the cell midbody during exit from mitosis (40), a finding that is consistent with the notion that Plk3 plays a role in cytokinesis and that suggests that altered expression of Plk3 interferes with cellular proliferation by impeding cytokinesis.
The expression of Plk3 is transiently induced by growth factors and cytokines and does not require protein synthesis for transcriptional activation (23). Three AU-rich elements, the most common RNA-destabilizing elements known in mammalian cells, are present in the 3Ј-untranslated regions of Plk3 and are also present in many labile mRNAs (41,42). These findings suggest that the activity of Plk3 is transcriptionally and post-transcriptionally regulated. Our results further show that overexpression of Plk3 induces apoptosis, suggesting that transcriptional and post-transcriptional regulation of Plk3 plays a key role in the control of Plk3 function. However, the posttranslational regulation of Plk3 and the mechanisms by which Plk3 induces apoptosis still remain unclear.
In summary, our results suggest a mechanism by which NF-B functions as a proapoptotic factor. In response to high   HEK293 cells (a, lanes 1-4) and MDAPanc-28/Puro cells (b, lanes 1-4) were treated with 50 g/ml doxycycline in the presence of control (CTL) and Plk3 siRNA. MDAPanc-28/Puro cells (c, lanes 1-4) were treated with TNF-␣ (50 ng/ml) in the presence of control and Plk3 siRNA. Forty-eight hours after transfection or stimulation, cells were harvested, and DNA fragmentation assays were performed. d, colocalization of overexpressed Plk3 and centrosomes. EGFP control vector, EGFP-wt-PLK3, and EGFP-Plk3 N terminus deletion mutant (Plk3 ⌬N ) were transfected into HEK293 cells. Twenty-four hours after transfection, these cells were fixed, stained with anti-tubulin-␥ antibody and 4Ј,6-diamidino-2-phenylindole (DAPI), and visualized with fluorescence microscopy. e, proposed working model of the NF-B-regulated proapoptotic signaling cascade. In response to doxycycline or other stimulations, NF-B induces Plk3 expression, which in turn activates p53-independent apoptotic signaling and p53-dependent apoptotic signaling by phosphorylation of p53 on Ser-20. levels of superoxide induced by doxycycline, NF-B, a key regulator of the antiapoptotic pathway, is required to induce Plk3 to initiate apoptosis independent of p53 activity, possibly involving the Src homology 3 domain in the N terminus of Plk3, and to phosphorylate p53 protein at Ser-20, which in turn induces the expression of its downstream target genes, such as PUMA, to initiate proapoptotic signaling.