p53 target gene AEN is a nuclear exonuclease required for p53-dependent apoptosis

Abstract

DNA degradation is one of the biochemical hallmarks detected in apoptotic cells, and several nucleases have been reported to function cooperatively in this process. It has also been suggested that different sets of nucleases are activated by different stimuli, and induce distinct patterns of DNA degradation. Here we report that apoptosis-enhancing nuclease (AEN) is a novel direct target gene of p53. AEN is induced by p53 with various DNA damage, and its expression is regulated by the phosphorylation status of p53. We demonstrate that AEN is a typical exonuclease with conserved exonuclease domains Exo I–III, and it targets both single- and double-stranded DNA and RNA. AEN induces apoptosis by itself, and the conserved domains are essential for both AEN nuclease activity and its apoptosis-inducing ability. AEN possesses nuclear and nucleolar localization signals, and it translocates from the nucleolus to nucleoplasm upon apoptosis induction. We also show the dislocation of nucleophosmin in conjunction with the translocation of AEN to the nucleoplasm, indicating the ability of AEN in nucleolus disruption. In addition, AEN is shown to be required for efficient DNA fragmentation in p53-dependent apoptosis. These results suggest that AEN is an important downstream mediator of p53 in apoptosis induction.

Introduction

Apoptosis is an essential process in the development of organs, maintenance of appropriate cell numbers and proper elimination of unnecessary or damaged cells. Chromosome fragmentation is a biochemical hallmark of apoptosis and several nucleases have been reported to be involved in this process (Samejima and Earnshaw, 2005), including apoptotic nucleases such as DNase II, DFF40/CAD, Endo G and AEN. DNase II is a lysosomal nuclease, and functions in digesting chromosomal DNA in apoptotic cells engulfed by phagocytes (Kawane et al., 2006). The other three nucleases function prior to DNase II, in the early stage of apoptosis when apoptotic DNA degradation occurs (Parrish and Xue, 2006). DFF40/CAD is a cytoplasmic nuclease, and creates 3′ hydroxyl DNA breaks, leading to the generation of 50–300 kb cleavage products and subsequent internucleosomal DNA fragmentation (Nagata et al., 2003). Endo G is a mitochondrial endonuclease that is released from mitochondria to translocate into the nucleus during apoptosis (Li et al., 2001). Endo G also generates characteristic apoptotic DNA ladders. AEN has been reported as a nuclear apoptosis-enhancing nuclease, and has been shown to have exonuclease activity against various DNA substrates (Lee et al., 2005). These nucleases have different characteristics, that is, subcellular localization or enzymatic activity, and are suggested to act cooperatively during apoptosis.

Tumor suppressor gene p53 is the most frequently mutated tumor suppressor gene in human cancer, and it encodes a transcription factor (Levine, 1997; Prives, 1998; Vogelstein et al., 2000; Vousden and Lu, 2002; Hofseth et al., 2004; Shmueli and Oren, 2005; Liu and Chen, 2006). The p53 protein exerts its tumor suppressive function by inducing the target genes involved in apoptosis induction and cell cycle arrest (El-Deiry, 1998; Oda et al., 2000; Ohki et al., 2000). We have been focusing on the functional regulation of p53 through the N-terminal transactivation domain (TAD) of p53. The p53 TAD contains multiple phosphorylation sites, and phosphorylation on these sites is essential for the full activation of p53 (Cordenonsi et al., 2007; Sun et al., 2007). It has been reported that mutations on Ser residues that undergo phosphorylation impair apoptotic activity of p53, and several pro-apoptotic p53 target genes particularly require phosphorylation of Ser residues within the p53 TAD (Unger et al., 1999; Chao et al., 2003; Kaeser et al., 2004; MacPherson et al., 2004; Mayo et al., 2005). We therefore screened for p53 target genes that require p53 phosphorylation, with the intention to identify pro-apoptotic p53 target genes. We performed microarray expression analysis using Saos2-derived cell lines stably expressing temperature-sensitive wild-type p53 and p53 carrying point mutations at all serine residues within the TAD (TAD-S/A; Ohki et al., 2007). In these cell lines, approximately 80% of genes induced by wild-type p53 were not induced by TAD-S/A, demonstrating the significance of phosphorylation within the TAD for the transactivation ability of p53. In addition, as expected, most of the pro-apoptotic genes among the previously reported p53 target genes were not induced by TAD-S/A.

Here we demonstrate that a previously reported apoptosis-enhancing nuclease gene, AEN, is a novel direct target gene of p53, and the expression of AEN is regulated by the phosphorylation status of p53. Furthermore, we have shown that AEN expression leads to apoptosis, and AEN is a mediator of p53-dependent apoptosis.

Results

AEN is a p53-inducible gene

Among the phosphorylation-dependent genes, AEN gene was one of the genes strongly (7.7-fold) induced by temperature-sensitive wild-type p53 (Figure 1a). AEN mRNA was induced by the exogenous expression of p53 in Saos2 and H1299 cells (Figure 1b), and was also induced in a manner dependent on endogenous p53 in HCT116 p53 (+/+) cells and MRC5 cells upon γ-ray irradiation and 5-FU treatment (Figures 1c and d). In addition, various DNA-damaging agents, including adriamycin and 5-FU, induced AEN mRNA expression in many cell lines (Figures 1e and 3a).

Figure 1
figure1

AEN is a p53-inducible gene. (ac and e) AEN expression was analysed by northern blotting. Methylene blue staining of 28S ribosomal RNA confirmed equal loading of RNA in each lane. (a) The ts-wild-type p53- or ts-TAD-S/A-expressing Saos2 cell line was tested for the ability to induce AEN upon shift to the permissive temperature with or without γ-ray irradiation. Cells were subjected to γ-ray irradiation after 2 h of temperature shift to 32 °C. Cells were collected at 5 and 14 h post-temperature shift. Raw northern blotting data are shown in the top panel. Signal intensities were calculated using BAS2500, and shown in the bottom panel. (b) Saos2 and H1299 cells were transiently transfected with either pcDNA3 empty vector or pcDNA3-p53. Cells were harvested 24 h post transfection. (c) HCT116 p53 (+/+), HCT116 p53 (−/−), MRC5 cells transduced with control lentivirus or lentivirus expressing p53 shRNA were subjected to γ-ray irradiation (20 Gy). The shRNA targeting p53 contains a 19-nt sequence derived from the ORF of human p53 (nucleotides 775–793). Cells were harvested 6 h post-irradiation. (d) HCT116 p53 (+/+) cells and HCT116 p53 (−/−) cells were treated with 5-FU (0.38 mM) for 24 h. Expression of AEN together with representative p53 target genes (hdm2 and pig3) was analysed by RT–PCR. (e) MCF7, U2OS, A549 and MRC5 cells were treated with adriamycin (1 μM) for 18 h.

AEN is a direct target gene of p53

Since AEN is induced by p53 upon DNA damage, we then explored whether AEN is a direct target gene of p53. The consensus binding motif of p53 is composed of two copies of 5′-RRRCWWGYYY-3′ separated by 0–13 base pairs (R, purine; Y, pyrimidine; W, A or T; El-Deiry et al., 1992). As shown in Figure 2a, we found three putative p53 responsive elements (REs), RE1 in the promoter region, and RE2 and RE3 in intron 1 of the AEN gene. The promoter region and intron 1 of AEN was PCR amplified and cloned upstream of the luciferase reporter gene. We first analysed if the 1.7 kb fragment containing RE1, 1.1 kb fragment containing RE2 and RE3 and 1.5 kb fragment containing RE3 show p53 responsiveness by luciferase reporter promoter assay. While the fragment containing RE2 and RE3 showed no response to wild-type p53, the fragment containing RE1 showed a strong response to p53 (Figure 2b). In order to identify the p53-responsive element within the 1.7 kb fragment, we next analysed p53 responsiveness of the shorter 0.5 kb fragment containing RE1 and the same fragment with four point mutations within the p53 consensus binding site. Although the 0.5 kb fragment containing RE1 showed strong p53 responsiveness, the mutant version of this fragment (RE1 mut) lost its responsiveness. On the other hand, the p53 mutant, V143A, which has no transcription activation ability, did not activate the promoter, confirming that activation of the AEN promoter by wild-type p53 is dependent of the transcriptional activity of p53. We also performed a chromatin immunoprecipitation (ChIP) assay to analyze if p53 protein binds to the AEN promoter in vivo. PCR primers to amplify the AEN promoter region were designed so as to contain RE1 and, as shown in Figure 2c, temperature-sensitive p53 bound to the AEN promoter only when p53 protein was in the active form (at the permissive temperature). Collectively, these results identify AEN as a bona-fide target gene of p53.

Figure 2
figure2

AEN is a direct target gene of p53. (a) Schematic representation of the reporter constructs used in the assay is shown. AEN promoter region or intron 1 containing putative p53 binding sites (RE13) was cloned upstream of the firefly luciferase reporter gene. PCR-amplified fragments were cloned into the Sma I site of the pGL3-promoter vector (Promega, Madison, WI, USA). (b) Luciferase reporter assay was performed. The constructs were tested for transactivation by wild-type p53 and a transactivation-deficient mutant of p53 (V143A). Saos2 cells were seeded in 96-well dishes and co-transfected with 30 ng of firefly luciferase reporter gene and 1 ng of each p53 gene cloned in pcDNA3 vector together with 7.5 ng of renilla luciferase expression vector (pGL4 (TK-Rluc) vector) as an internal control for transfection efficiency. Cells were harvested 24 h post transfection. Data are represented as mean fold activation ±s.d. of three independent experiments. (c) ChIP assay was performed for p21 and AEN promoters. The positions of the PCR primers within the AEN promoter region are shown at the top. (d) Full activation of the AEN promoter requires phosphorylation of p53, especially at Ser15. Luciferase assay was performed using luciferase reporter gene containing −448 to +25 region of the AEN promoter. Saos2 cells were co-transfected with 30 ng of firefly luciferase reporter gene and 1 ng of each p53 gene cloned in pMX-puro retroviral vector (each construct is expressed from the retroviral LTR promoter) together with 7.5 ng of renilla luciferase expression vector pGL4 (TK-Rluc) vector. Cells were harvested 24 h post transfection. Data are represented as the mean fold activation ±s.d. of three independent experiments. (e, f) WI-38 and AT2KY fibroblast cells were subjected to γ-irradiation (5 Gy). Cells were harvested 3 and 6 h post-irradiation. Western blotting was performed to detect p53 expression and activation (Ser15 phosphorylation of p53). To show equal loading of protein on each lane, β-actin was also detected (e). Expression levels of endogenous AEN were analysed by semi-quantitative RT–PCR. The intensity of PCR products was quantified by NIH Image software (f).

AEN expression is regulated by phosphorylation status of p53

As shown in Figure 1a, we found that AEN expression is severely defective in cells expressing ts-TAD-S/A. When the luciferase reporter assay was performed using a reporter construct containing p53 RE1, TAD-S/A showed 1.8-fold activation, while wild-type p53 showed 3.1-fold activation (Figure 2d). We therefore analysed which Ser residue within the TAD contributes to AEN expression. We constructed p53 phosphorylation-deficient mutants carrying single amino acid conversions from Ser to Ala on all Ser residues within the TAD. As shown in Figure 2d, while p53 Ala mutants at Ser 6, 9, 20, 33, 37, 46 showed activity similar to that of wild-type p53, the mutant with Ser 15 showed decreased activity similar to that of TAD-S/A (2.1-fold activation). On the other hand, phosphorylation-mimic mutant S15D showed enhanced promoter response (6-fold activation). It has been reported that Ser 15 of p53 is phosphorylated by ATM kinase upon γ-ray irradiation (Banin et al., 1998; Canman et al., 1998). To further address the significance of Ser15 phosphorylation on the expression of AEN, we analysed endogenous AEN expression in ATM-proficient normal human fibroblast WI-38 and ATM-deficient AT2KY fibroblast in response to γ-ray irradiation. As shown in Figure 2e, upon γ-ray irradiation, Ser 15 of p53 is efficiently phosphorylated in WI-38 cells, whereas diminished and delayed phosphorylation is detected in AT2KY cells. Moreover, in concordance with the strength of p53 phosphorylation at Ser 15, strong induction of AEN is achieved in WI-38 cells, while decreased induction is seen in AT2KY cells (Figures 2e and f). It is of note that the p53 protein level does not correlate with AEN expression (compare lanes 2 and 6 of Figures 2e and 2f). Taken together, phosphorylation of p53 at Ser15 is required for effective induction of AEN. It has been reported that phosphorylation on Ser15 is required for efficient phosphorylation of Ser 9 and 20 (Saito et al., 2003). Although the single amino acid phosphorylation of Ser 9 and 20 does not have a dramatic effect on the AEN promoter, it may work in combination with Ser 15, owing to the fact that TAD-S/A showed slightly lower activation ability than S15A (Figure 2d). In addition, interaction with p300/CBP, a transcriptional co-activator of p53, also requires Ser 15 phosphorylation (Lambert et al., 1998). The requirement of Ser 15 phosphorylation for activation of the AEN promoter indicates the possible involvement of p300 in AEN induction. In agreement, the results from ChIP analysis showed that TAD-S/A was able to bind to the AEN promoter as efficiently as wild-type p53, but was still unable to fully activate the promoter (Figure 2c, lanes 5 and 6). Phosphorylation status of p53 may affect the recruitment of co-factors to the promoters, and thereby efficient activation of the promoters.

AEN is required for p53-dependent apoptosis

Since AEN is induced by p53 and reported to enhance apoptosis, we analysed whether AEN is involved in p53-dependent apoptosis (Lee et al., 2005). The human colorectal cancer cell line RKO has wild-type p53, and is known to undergo p53-dependent apoptosis upon DNA damage. Therefore, we knocked down AEN by introducing siRNA duplexes targeting AEN to the RKO cell line, and treated with 5-FU. As shown in Figure 3a, siRNA efficiently downregulated AEN expression. We next analysed the proportion of cells undergoing apoptosis by measuring cells in sub-G1 after DNA damage. As shown in Figure 3b, compared to cells treated with control siRNA, cells in sub-G1 decreased when treated with AEN siRNAs. Since the appearance of cells in the sub-G1 region is a consequence of cells with DNA fragmentation, and AEN ablation led to decreases in the sub-G1 region, we next performed a DNA fragmentation assay (Tounekti et al., 1995). Compared to cells treated with control siRNA, DNA fragmentation diminished in cells treated with AEN siRNAs (Figure 3c). Moreover, expression of cleaved poly (ADP-ribose) polymerase (PARP) in cells treated with AEN siRNAs was reduced compared to control cells (Figure 3d). These results show that AEN is involved in the p53-dependent apoptosis pathway.

Figure 3
figure3

AEN is required for p53-dependent apoptosis. Control or AEN-targeting siRNAs (ON-target plus siRNAs purchased from Dharmacon Research, Lafayette, CO, USA) were introduced into RKO cells using lipofectamin 2000 (Invitrogen, Carlsbad, CA, USA). RKO cells were treated with 5-FU (0.13 mM) at 36 h post transfection. (a) Cells were harvested at 43 h post 5-FU treatment, and endogenous AEN expression was analysed by northern blotting. (b) Cells were harvested at 43 h post drug treatment and fixed with 70% ethanol. Cells stained with 1 μg ml−1 propidium iodide analysed by FACS. The proportion of cells with sub-G1 DNA content was analysed and shown as a graph. (c) Cells were harvested at 0, 24 and 43 h post –5-FU treatment and DNA fragmentation assay was analysed. (d) Cells were harvested 36 h post 5-FU treatment and expression of cleaved PARP (detected with anti-cleaved PARP (Asp214) antibody, CST) was analysed.

AEN is an exonuclease that targets both single- and double-stranded DNA and RNA

When the amino acid sequence of AEN was analysed in detail, we found that AEN possesses three characteristic sequence motifs, termed ExoI, ExoII and ExoIII, which are found in typical 3′–5′ exonuclease (Figure 4a). These domains are clustered around the active site and contain a catalytically active tyrosine and four negatively charged residues that serve as ligands for the two metal ions required for catalysis (Moser et al., 1997). Within these conserved residues, AEN possesses four negatively charged residues, Asp114, Glu116, Asp198 and Asp258. Among exonucleases, it is known that some do not possess the conserved Tyr, in place of which they possess Ser and the preceding His. AEN belongs to this type of nuclease, and has His253 and Ser254. Since these residues are known to be essential for nuclease activity, we constructed a mutant with amino acid conversions D114A, E116A and D258A within ExoI and ExoIII (Exo-mut, shown in Figure 4a). We first analysed the biochemical activity of each AEN protein by in vitro nuclease assay using Flag-tagged AEN proteins immunoprecipitated from transiently transfected 293T cells (Figures 4b and c). Wild-type AEN was previously reported to possess exonuclease activity against various DNA substrates (Lee et al., 2005), however, in addition to nuclease activity against DNA substrates, we have shown that AEN has nuclease activity against both single- (ss)- and double-stranded (ds)RNA substrates (Figure 4c). The physiological significance of the RNase activity of AEN is currently unknown and is an interesting future issue to be clarified. On the other hand, Exo-mut showed no activity against either of the substrates under the conditions tested (Figure 4c). These results demonstrate that AEN is a typical exonuclease and the amino acids Asp114, Glu116 and Asp258 are residues essential for AEN exonuclease activity.

Figure 4
figure4

AEN has nuclease activity against ds/ssDNA and ds/ssRNA. (a) Schematic representation of wild-type or mutant AEN protein used in the study. Each AEN was cloned in pcDNA3 vector and C-terminally tagged with the Flag epitope. (b, c) Exonuclease activity of each AEN was measured. Each AEN proteins were purified by immunoprecipitating Flag-tagged AENs from 293T cells, and nuclease assay was performed. AEN proteins immunoprecipitated from 8 × 104 and 2.5 × 105 cells were used for samples loaded on lanes 2, 4, 6 and 3, 5, 7, respectively. For dsDNA substrates, immunoprecipitated protein obtained from 2.5 × 105 (lanes 2, 4, 6) and 8 × 105 (lanes 3, 5, 7) cells was used. The substrates used in the assay were dsDNA with hairpin loop, 5′-TATAGATCTCCTACCTGAGAACGTCCACATTAATCACTGATCCTTCGCTTCAGTGATTAATGTGGACGTTCTCAGGTAGGAGATCTATA-3′; ssDNA with a terminal hairpin loop, 5′-TATAGATCTCCTACCTGAGAACGTCCACATTAATCACTGATCCTTCGCTTCAGTGATTAAT-3′; dsRNA, RNA duplex purchased from Dharmacon Research (nonspecific control X); ssRNA, boiled RNA duplex. The amounts of substrates used in the assay were, 5 pmol and 60 pmol for DNAs and RNAs, respectively. (d) Detecting formation of ssDNA by nondenaturing BrdU staining assay. U2OS cells were transfected with pcDNA3, pcDNA3-AEN-Flag or pcDNA3-Exo-mut-Flag. Cells were harvested 36 h post transfection.

To further examine the role of AEN as an exonuclease within the cell, we performed nondenaturing BrdU staining assay to analyse whether AEN expression leads to the formation of ssDNA within the cell. As shown in Figure 4d, while cells transfected with Exo-mut were negative for BrdU staining, cells expressing wild-type AEN were positive for BrdU staining; therefore, ssDNA was formed within the cell in a manner dependent on the exonuclease activity of AEN. Thus, in addition to the data obtained by biochemical analysis, it was shown that AEN digests DNA and forms ssDNA within the cell.

Apoptosis induction by AEN requires exonuclease activity

It has been reported that AEN induces apoptosis, and apoptosis-inducing ability is dependent on its nuclease activity (Lee et al., 2005); however, in our study, the expression of the previously described nuclease activity-deficient mutant of AEN was significantly low compared to that of wild-type AEN (data not shown). Therefore, to assess the apoptosis-inducing ability of nuclease activity-deficient AEN, we tested the efficiency of apoptosis induction by Exo-mut. First, in order to evaluate the expression level of each AEN protein, control empty pcDNA3, pcDNA3-AEN-Flag or pcDNA3-Exo-mut-Flag vectors were transfected to 293T cells. As shown in Figure 5a, each AEN protein was expressed at a similar level. In addition, subcellular localization of wild-type and Exo-mut AEN proteins were analysed in U2OS cells, and both proteins showed similar subcellular localization, showing that subcellular localization of AEN does not require exonuclease activity (Figure 6d and discussed below in detail). As shown in Figures 5b and c, when AEN-expressing cells were observed, they showed typical apoptotic phenotypes revealed by 4′-6-diamidino-2-phenylindole (DAPI), and were positive for annexin V and caspase-3 activity. On the other hand, Exo-mut-expressing cells did not show these phenotypes. Then, control empty pcDNA3, pcDNA3-AEN-Flag or pcDNA3-Exo-mut-Flag vectors were transfected to U2OS, HCT116 p53 (+/+) and HCT116 p53 (−/−) cells together with TK-Rluc vector expressing renilla luciferase under a constitutively active thymidine kinase promoter, and luciferase activity-based cell survival analysis was performed. Similar results were obtained with either of the cell lines tested, and luciferase activity of pcDNA3-Exo-mut-Flag-expressing cells was similar to that of control cells, showing that the conditions of the cells expressing Exo-mut-Flag were similar to cells without its expression (Figure 5d). On the other hand, luciferase activity was reduced to 40% of that of control cells in AEN-Flag-expressing cells, indicating that approximately 60% of cells co-transfected with AEN underwent cell death under the conditions tested. In addition, as shown in Figure 5d, AEN showed cell death-inducing ability in both p53-positive HCT116 p53 (+/+) cells and p53-negative HCT116 p53 (−/−) cells, indicating that this ability is independent of p53 status, consistent with the notion that AEN is a downstream mediator of p53 in apoptosis induction. Collectively, these results show that AEN induces apoptosis, and the ability is dependent on its exonuclease activity.

Figure 5
figure5

AEN-induced apoptosis requires exonuclease activity. (a) Expression of each AEN was analysed by western blotting. 293 T cells (1 × 106 cells) were transfected with 1 μg of pcDNA3, pcDNA3-AEN-Flag or pcDNA3-Exo-mut-Flag. Cells were harvested at 24 h post transfection. (b, c) U2OS cells (5 × 104 cells) were transfected with 200 ng of pcDNA3 or pcDNA3-AEN-Flag. Cells were stained with annexin V and DAPI (b), or anti-cleaved caspase-3 (detected with anti-cleaved caspase-3 antibody (Asp178), CST) and DAPI (c), 48 h post transfection. (d) Luciferase gene co-transfection cell survival analysis was performed as described (Ohki et al., 2007). U2OS, HCT116 p53 (+/+) and (−/−) cells (2.5 × 104 cells) were transfected with 70 ng of pcDNA3, pcDNA3-AEN-Flag or pcDNA3-Exo-mut-Flag together with 4.5 ng of renilla luciferase expression vector (pGL4 (TK-Rluc)) control vector. Cells were harvested at 24 and 56 h post transfection, and analysed using Luciferase Reporter Assay Reagent (Promega). Luciferase activities were indistinguishable at 24 h post transfection, showing that transfection was successful in all samples (data not shown). Fold survival rate (Fold luciferase activities) at 56 h post transfection is shown as a graph.

Figure 6
figure6

AEN has nuclear and nucleolar localization signals, and is localized in the nucleus and nucleolus. (a) Subcellular localizations of AEN proteins were analysed by confocal microscopy. U2OS cells were transfected with pcDNA3, pcDNA3-AEN. Cells were fixed with 4% paraformaldehyde 18 h post transfection. The fixed cells were sequentially incubated with anti-AEN antibody (I, raised against C-terminus of AEN (aa 306–325); II, raised against N-terminus of AEN (aa 1–12)) and AlexaFluor 488 conjugated secondary antibody, and subjected to immunofluorescent microscopic analysis. Nuclei were counterstained with DAPI. (b) Sequence alignment of NoLS of different human and viral protein with AEN NoLS. Clusters of basic amino acids are highlighted in boldface. (c) Schematic representation of wild-type or mutant AEN proteins used in the study. (d) Subcellular localizations of AEN proteins were analysed. U2OS cells were transfected with control vector or Flag-tagged AEN constructs. Cells were fixed and subjected to immunofluorescent microscopic analysis 18 h post transfection. Each AEN protein was detected with anti-Flag antibody. Nuclei were counterstained with DAPI. (e) Subcellular localizations of AEN proteins in cells were quantified by scoring fluorescent signals from immunofluorescent microscopic analysis. The percentages of cell population showing each localization were calculated using more than 300 cells in three independent experiments. The percentages of cells localized in the nucleus (black bars), nucleolus (hatched bars) and cytoplasm (open bars) are shown.

AEN has nuclear and nucleolar localization signals, and is predominantly localized in the nucleolus

In order to reveal the physiological function of AEN, we next analysed the subcellular localization of AEN. AEN was expressed in U2OS cells, and subcellular localization of AEN was detected with two different anti-AEN antibodies that were raised against N- or C terminus of AEN protein. Both antibodies detected similar patterns, showing that AEN is localized in the nucleus (Figure 6a). Interestingly, many large punctate signals were seen in the nucleus, indicating that AEN protein localizes predominantly in nuclear compartments. DAPI-stained cells are also shown in Figure 6a, in which dark crater-like nucleoli are surrounded by the brighter DAPI fluorescence of nucleic acid in the nucleoplasm. In the same cells, AEN is concentrated in the same area where DAPI staining is weak, demonstrating that AEN is predominantly localized in the nucleolus.

It is well known that Lys- and Arg-rich regions function as nuclear or nucleolar localization signals (Figure 6b; Jans et al., 2000; Rowland and Yoo, 2003). Since AEN was localized in the nucleus/nucleolus, we searched for nuclear and/or nucleolar localization signals in AEN, and found that AEN has an Lys- and Arg-rich region at positions aa 27–35 and 165–188 (Figure 6c). We therefore investigated whether those regions of AEN function as nuclear and/or nucleolar localization signals. Since many nucleolar localization signals are often composed of tandemly repeated basic amino acids, we speculated that aa 27–35 functions as a nucleolar localization signal, and aa 165–188 functions as a nuclear localization signal. In order to demonstrate the function of both regions, we constructed three AEN mutants (NoLS-mut, all of the Arg and Lys residues within aa 27–35 were changed to Ala; NoLS/NLS-mut, all of the Arg and Lys residues within aa 27–35 and 165–188 were changed to Ala and NLS-mut, all of the Arg and Lys residues within aa 165–188 were changed to Ala), and each AEN protein was expressed in U2OS cells (Figure 6d). As summarized in Figure 6e, subcellular localization of each AEN was analysed by immunofluorescent microscopic analysis from 300 cells. The localization of NLS-mut was similar to that of wild-type AEN, showing that aa 165–188 is not essential for nuclear and nucleolar localization signals. NoLS-mut localized in the nucleus but not in the nucleolus, showing that aa 27–35 is required for nucleolar localization, but is not essential for nuclear localization. In addition, NoLS/NLS-mut was localized in the cytoplasm, showing that nuclear and nucleolar localization signals are only present within aa 27–35 and 165–188 (Figures 6d and e). Therefore, we identified the motif 27RKRHKRKSR35 as a nuclear and nucleolar localization signal, and the other motif, 165RQHMRK-KLLKGK188, as a nuclear localization signal of AEN. Taken together, AEN has one nucleolar/nuclear localization signal and one nuclear localization signal.

AEN is translocated to the nucleoplasm from the nucleolus upon execution of apoptosis

While we have demonstrated that AEN was predominantly localized in the nucleolus (Figure 6d), a previous report demonstrated that AEN was localized in the nucleus (Lee et al., 2005). To define the precise subcellular localization of AEN, we performed a time course study of the subcellular localization of AEN using immunofluorescence microscopic analysis. Interestingly, we found that AEN changes its subcellular localization during the course of apoptosis induction. When AEN is expressed and cells are not yet undergoing apoptosis, the subcellular localization of AEN is predominantly in the nucleolus (18 h post transfection, shown in Figures 6d, e and 7a); however, when cells are undergoing apoptosis (as shown in Figure 5c), AEN is localized in the nucleoplasm but not in the nucleolus (36 h post transfection, Figures 7b and c). We also compared the subcellular localization of AEN with that of nucleophosmin (NPM). NPM is a representative nucleolar protein, frequently used as a marker of an intact nucleolus (Rubbi and Milner, 2003). As shown in Figure 7a, the localization of wild-type and Exo-mut AEN was indistinguishable from that of NPM, confirming the nucleolar localization of these proteins. On the other hand, localization of NoLS-mut did not overlap with that of NPM, again confirming that this mutant does not localize to the nucleolus. Interestingly, when wild-type or Exo-mut AEN translocates to the nucleoplasm, NPM simultaneously translocates to the nucleoplasm (Figure 7b). Since this NPM translocation was not observed in cells expressing NoLS-mut or cells transfected with an empty vector, it seems that AEN has the ability to translocate NPM, and this ability is independent of the nuclease activity of AEN. Dislocalization of NPM is a criterion for nucleolus disruption, and therefore indicates that AEN has the ability to disrupt the nucleolus.

Figure 7
figure7

AEN is translocated to the nucleoplasm from the nucleolus upon apoptosis induction. (a, b) Subcellular localizations of AEN proteins were analysed. U2OS cells were transfected with control vector or Flag-tagged AEN constructs. Cells were fixed 18 h (a) or 36 h (b) post transfection, and subjected to immunofluorescent microscopic analysis, as in Figure 6d. NPM was detected with anti-NPM B23 (H106) antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA). (c) Subcellular localizations of AEN proteins in cells were quantified as in Figure 6e. (d) Luciferase gene co-transfection cell survival analysis was performed as in Figure 5d. Cells were harvested at 24, 48 and 72 h post transfection. Fold survival rates (luciferase activities) are shown as a graph. (e, f) Clonogenic survival assay was performed. U2OS and HCT116 cells (1 × 106 cells) were transfected with 1 μg of control vector, AEN-Flag, Exo-mut-Flag or NoLS-mut-Flag cloned in pLenti6/V5-DEST vector. Cells were harvested at 24 h post transfection, and expression of each AEN was analysed by western blotting (e). Clonal selection by blasticidin S (U2OS, 8 μg ml−1; HCT116, 4 μg ml−1) was initiated 24 h post transfection. Cells were fixed with methanol and stained with Giemsa 14 days post-drug selection (f).

AEN induces apoptosis in the nucleoplasm

We next investigated the significance of nucleolus localization in the function of AEN. To address this issue, the function of NoLS-mut AEN was investigated. Control empty pcDNA3, pcDNA3-AEN-Flag, pcDNA3-Exo-mut-Flag or pcDNA3-NoLS-mut-Flag vectors were transfected to U2OS and HCT116 cells, and the efficiency of cell death induction was analysed. In both cells, induction of cell death by NoLS-mut AEN was stronger than that by wild-type AEN (Figure 7d). Furthermore, we assayed the effect of each AEN on clonogenic survival when expressed continually over a period of time. Control empty pLenti6/V5-DEST, pLenti6/V5-DEST-AEN-Flag, pLenti6/V5-DEST-Exo-mut-Flag and pLenti6/V5-DEST-NoLS-mut-Flag were expressed in HCT116 and U2OS cells, and a clonogenic survival assay was performed. As shown in Figure 7e, each AEN protein was expressed at a similar level. Transfection of AEN or NoLS-mut, when compared with cells transfected with vector alone, caused a dramatic reduction in the number of colonies (Figure 7f). In addition, the expression of NoLS-mut caused further reduction in the number of colonies compared with cells transfected with wild-type AEN, confirming the result obtained by luciferase gene co-transfection cell survival analysis. On the other hand, while no effect of Exo-mut AEN on cell viability was detected by luciferase gene co-transfection cell survival analysis (Figure 7d), a certain negative effect on cell growth was detected by clonogenic survival assay (Figure 7f). We do not know how and why AEN showed a cell growth-suppressing effect independent of nuclease activity, and this is an interesting issue to be clarified in the future; however, since Exo-mut retains NPM-translocating ability, this ability may be one cause of the growth suppressive function of Exo-mut. Taken together, these results show that AEN displays its apoptosis-inducing function when dislocated from the nucleolus and released into the nucleoplasm.

Discussion

We have previously shown that phosphorylation of p53 within the N-terminal transactivation domain is of great importance (Ohki et al., 2007). In particular, genes involved in apoptosis induction were shown to have a tendency to require p53 phosphorylation. The novel direct target gene of p53, AEN, also requires p53 phosphorylation for efficient induction by p53, and is a mediator of p53-dependent apoptosis.

It was reported previously that AEN is a nuclear DNA exonuclease, and functions in the enhancement of apoptosis (Lee et al., 2005); however, when the biochemical activity of AEN was analysed using various DNA and RNA substrates, we demonstrated that AEN shows exonuclease activity against both ss- and dsDNA and ss- and dsRNA. Although the physiological significance of the RNase activity of AEN is currently unknown, AEN is localized to the nucleolus where RNA biogenesis is active and RNases are engaged in RNA degradation and maturation (Houseley et al., 2006). Accordingly, AEN may act on RNA to modulate RNA synthesis upon genotoxic stresses. The involvement of AEN in the regulation of RNA species is an interesting issue to be clarified in future studies. On the other hand, our results suggest that DNase activity of AEN may be directly involved in p53-induced apoptosis. It has been reported that another exonuclease, Exonuclease-1, forms ssDNA within the cell, and play an important role in the DNA damage response (Schaetzlein et al., 2007). Similarly, when AEN was expressed in cells, it formed ssDNA. Thus, these results indicate that when AEN is induced by p53 upon DNA damage, it digests dsDNA to form ssDNA, and therefore amplifies DNA damage signals leading to the enhancement of apoptosis.

We also showed that AEN induces apoptosis by itself, and its exonuclease activity is essential for AEN-induced apoptosis. In addition, we performed a time course study of AEN localization, and found that AEN is localized in nucleolus pre-apoptosis, and released to the nucleoplasm upon apoptosis induction. It has been reported that many proteins translocate from the nucleolus to nucleoplasm upon DNA damage (Kurki et al., 2004; Mayer and Grummt, 2005). For example, p53 is localized to the nucleolus, and DNA damage markedly decreases the association of p53 with the nucleolus (Karni-Schmidt et al., 2007). Werner helicase also translocates from the nucleolus to nucleoplasm upon DNA damage, and exerts its function in the nucleoplasm (Blander et al., 2002). Thus, it seems that translocation from the nucleolus to nucleoplasm is a common mechanism to regulate the function of proteins involved in the DNA damage response.

Moreover, by analysing the localization of NPM, we found several interesting features of cells expressing AEN. We found that NPM translocates from the nucleolus to the nucleoplasm simultaneously with AEN translocation. This migration of NPM was also observed in cells expressing Exo-mut, showing that the ability to translocate NPM does not require the nuclease activity of AEN. On the other hand, NPM translocation was not observed in cells expressing NoLS-mut, demonstrating that nucleolar localization is essential for NPM translocation ability. Translocation of NPM is often a criterion for nucleolus disruption, and nucleolar disruption is observed upon various cellular stresses (Rubbi and Milner, 2003; Mayer and Grummt, 2005). Our results indicate that besides exonuclease activity, AEN has another function in nucleolus disruption, and this function may also be involved in the DNA damage response downstream of p53. Accordingly, we propose the model shown in Figure 8. Upon DNA damage, AEN is first trapped in the nucleolus, but then AEN disrupts the nucleolus and is released into the nucleoplasm. Subsequently, AEN in the nucleoplasm degrades DNA and forms ssDNA, therefore amplifying the DNA damage response, leading to apoptosis induction. Although the molecular mechanism of nucleolar disruption by AEN is currently unknown, identification of AEN-associating molecules in the nucleolus or nucleoplasm may reveal the mechanism underlying this phenomenon.

Figure 8
figure8

Novel p53-dependent apoptosis pathway mediated by AEN.

In summary, AEN is a direct target gene of p53, and is an apoptosis inducer required for p53-dependent apoptosis. Since apoptosis is efficiently executed by the coordination of apoptotic nucleases, the cooperation of AEN with other nucleases is an important future issue to be clarified. Our results also suggest that AEN has a tumor-suppressive function by regulating apoptosis, and therefore analysing the involvement of AEN in tumorigenesis is also an interesting future research direction.

Materials and methods

Information for northern blotting analysis, western blotting analysis, antibodies, ChIP assay, reverse transcription (RT)–PCR, plasmids, transfection, luciferase reporter assay, flow cytometry, cell death assay, DNA fragmentation assay, immunoprecipitation and immunofluorescence are available as Supplementary Information on the Oncogene website.

Nuclease assay

Nuclease assay was performed by incubating nucleic acid substrates with immunoprecipitated AENs in 50 mM Hepes-KOH (pH 7.0), 5% glycerol, 50 mM NaCl, 1 mM MnCl2, 0.01% Triton and 0.1 mM DTT. After 10 h, the reactions were stopped by adding 0.1% SDS and 10 mM EDTA. Samples were run on PAG Mini DAIICHI 15/25% (Daiichi Pure Chemicals, Tokyo, Japan), and stained with SYBR green I (FMC BioProducts, Rockland, ME, USA).

Detection of single-strand DNA by nondenaturing BrdU staining

Cells were subjected to transfection with the indicated plasmids. Culture medium was changed to medium containing 10 μg ml−1 BrdU 2 h post transfection. Cells were fixed 36 h after transfection and immunostained using FITC-conjugated anti-BrdU antibody without any preceding DNA denaturation or nuclease treatment. Cells were counterstained with DAPI. The images were obtained with HS ALL-in-one Type Fluorescence Microscope BZ-9000 Series (Keyence, Osaka, Japan).

References

  1. Banin S, Moyal L, Shieh S, Taya Y, Anderson CW, Chessa L et al. (1998). Enhanced phosphorylation of p53 by ATM in response to DNA damage. Science 281: 1674–1677.

  2. Blander G, Zalle N, Daniely Y, Taplick J, Gray MD, Oren M . (2002). DNA damage-induced translocation of the Werner helicase is regulated by acetylation. J Biol Chem 277: 50934–50940.

  3. Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K et al. (1998). Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science 281: 1677–1679.

  4. Chao C, Hergenhahn M, Kaeser MD, Wu Z, Saito S, Iggo R et al. (2003). Cell type- and promoter-specific roles of Ser18 phosphorylation in regulating p53 responses. J Biol Chem 278: 41028–41033.

  5. Cordenonsi M, Montagner M, Adorno M, Zacchigna L, Martello G, Mamidi A et al. (2007). Integration of TGF-beta and Ras/MAPK signaling through p53 phosphorylation. Science 315: 840–843.

  6. El-Deiry WS . (1998). Regulation of p53 downstream genes. Semin Cancer Biol 8: 345–357.

  7. El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW, Vogelstein B . (1992). Definition of a consensus binding site for p53. Nat Genet 1: 45–49.

  8. Hofseth LJ, Hussain SP, Harris CC . (2004). p53: 25 years after its discovery. Trends Pharmacol Sci 25: 177–181.

  9. Houseley J, LaCava J, Tollervey D . (2006). RNA-quality control by the exosome. Nat Rev Mol Cell Biol 7: 529–539.

  10. Jans DA, Xiao CY, Lam MH . (2000). Nuclear targeting signal recognition: a key control point in nuclear transport? Bioessays 22: 532–544.

  11. Kaeser MD, Pebernard S, Iggo RD . (2004). Regulation of p53 stability and function in HCT116 colon cancer cells. J Biol Chem 279: 7598–7605.

  12. Karni-Schmidt O, Friedler A, Zupnick A, McKinney K, Mattia M, Beckerman R et al. (2007). Energy-dependent nucleolar localization of p53 in vitro requires two discrete regions within the p53 carboxyl terminus. Oncogene 26: 3878–3891.

  13. Kawane K, Ohtani M, Miwa K, Kizawa T, Kanbara Y, Yoshioka Y et al. (2006). Chronic polyarthritis caused by mammalian DNA that escapes from degradation in macrophages. Nature 443: 998–1002.

  14. Kurki S, Peltonen K, Laiho M . (2004). Nucleophosmin, HDM2 and p53: players in UV damage incited nucleolar stress response. Cell Cycle 3: 976–979.

  15. Lambert PF, Kashanchi F, Radonovich MF, Shiekhattar R, Brady JN . (1998). Phosphorylation of p53 serine 15 increases interaction with CBP. J Biol Chem 273: 33048–33053.

  16. Lee JH, Koh YA, Cho CK, Lee SJ, Lee YS, Bae S . (2005). Identification of a novel ionizing radiation-induced nuclease, AEN, and its functional characterization in apoptosis. Biochem Biophys Res Commun 337: 39–47.

  17. Levine AJ . (1997). p53, the cellular gatekeeper for growth and division. Cell 88: 323–331.

  18. Li LY, Luo X, Wang X . (2001). Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412: 95–99.

  19. Liu G, Chen X . (2006). Regulation of the p53 transcriptional activity. J Cell Biochem 97: 448–458.

  20. MacPherson D, Kim J, Kim T, Rhee BK, Van Oostrom CT, DiTullio RA et al. (2004). Defective apoptosis and B-cell lymphomas in mice with p53 point mutation at Ser 23. EMBO J 23: 3689–3699.

  21. Mayer C, Grummt I . (2005). Cellular stress and nucleolar function. Cell Cycle 4: 1036–1038.

  22. Mayo LD, Seo YR, Jackson MW, Smith ML, Rivera Guzman J, Korgaonkar CK et al. (2005). Phosphorylation of human p53 at serine 46 determines promoter selection and whether apoptosis is attenuated or amplified. J Biol Chem 280: 25953–25959.

  23. Moser MJ, Holley WR, Chatterjee A, Mian IS . (1997). The proofreading domain of Escherichia coli DNA polymerase I and other DNA and/or RNA exonuclease domains. Nucleic Acids Res 25: 5110–5118.

  24. Nagata S, Nagase H, Kawane K, Mukae N, Fukuyama H . (2003). Degradation of chromosomal DNA during apoptosis. Cell Death Differ 10: 108–116.

  25. Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T et al. (2000). Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 288: 1053–1058.

  26. Ohki R, Kawase T, Ohta T, Ichikawa H, Taya Y . (2007). Dissecting functional roles of p53 N-terminal transactivation domains by microarray expression analysis. Cancer Sci 98: 189–200.

  27. Ohki R, Nemoto J, Murasawa H, Oda E, Inazawa J, Tanaka N et al. (2000). Reprimo, a new candidate mediator of the p53-mediated cell cycle arrest at the G2 phase. J Biol Chem 275: 22627–22630.

  28. Parrish JZ, Xue D . (2006). Cuts can kill: the roles of apoptotic nucleases in cell death and animal development. Chromosoma 115: 89–97.

  29. Prives C . (1998). Signaling to p53: breaking the MDM2-p53 circuit. Cell 95: 5–8.

  30. Rowland RR, Yoo D . (2003). Nucleolar-cytoplasmic shuttling of PRRSV nucleocapsid protein: a simple case of molecular mimicry or the complex regulation by nuclear import, nucleolar localization and nuclear export signal sequences. Virus Res 95: 23–33.

  31. Rubbi CP, Milner J . (2003). Disruption of the nucleolus mediates stabilization of p53 in response to DNA damage and other stresses. EMBO J 22: 6068–6077.

  32. Saito S, Yamaguchi H, Higashimoto Y, Chao C, Xu Y, Fornace Jr AJ et al. (2003). Phosphorylation site interdependence of human p53 post-translational modifications in response to stress. J Biol Chem 278: 37536–37544.

  33. Samejima K, Earnshaw WC . (2005). Trashing the genome: the role of nucleases during apoptosis. Nat Rev Mol Cell Biol 6: 677–688.

  34. Schaetzlein S, Kodandaramireddy NR, Ju Z, Lechel A, Stepczynska A, Lilli DR et al. (2007). Exonuclease-1 deletion impairs DNA damage signaling and prolongs lifespan of telomere-dysfunctional mice. Cell 130: 863–877.

  35. Shmueli A, Oren M . (2005). Life, death, and ubiquitin: taming the mule. Cell 121: 963–965.

  36. Sun P, Yoshizuka N, New L, Moser BA, Li Y, Liao R et al. (2007). PRAK is essential for ras-induced senescence and tumor suppression. Cell 128: 295–308.

  37. Tounekti O, Belehradek Jr J, Mir LM . (1995). Relationships between DNA fragmentation, chromatin condensation, and changes in flow cytometry profiles detected during apoptosis. Exp Cell Res 217: 506–516.

  38. Unger T, Sionov RV, Moallem E, Yee CL, Howley PM, Oren M et al. (1999). Mutations in serines 15 and 20 of human p53 impair its apoptotic activity. Oncogene 18: 3205–3212.

  39. Vogelstein B, Lane D, Levine AJ . (2000). Surfing the p53 network. Nature 408: 307–310.

  40. Vousden KH, Lu X . (2002). Live or let die: the cell's response to p53. Nat Rev Cancer 2: 594–604.

Download references

Acknowledgements

This study was supported by the program for the promotion of Fundamental Studies in Health Sciences of the Pharmaceuticals and Medical Devices Agency (PMDA; to TO and HI), a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to YT), a Grant-in-Aid for Third Term Comprehensive Control Research for Cancer from the Ministry of Health, Labor and Welfare, Japan (to YT).

Author information

Correspondence to R Ohki or Y Taya.

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

Kawase, T., Ichikawa, H., Ohta, T. et al. p53 target gene AEN is a nuclear exonuclease required for p53-dependent apoptosis. Oncogene 27, 3797–3810 (2008). https://doi.org/10.1038/onc.2008.32

Download citation

Keywords

  • p53
  • exonuclease
  • apoptosis
  • p53 phosphorylation

Further reading