The forkhead transcription factor FOXO1, a downstream target of phosphatidylinositol-3-kinase/Akt signalling pathway, regulates cyclic differentiation and apoptosis in normal endometrium, but its role in endometrial carcinogenesis is unknown. Screening of endometrial cancer cell lines demonstrated that FOXO1 is expressed in HEC-1B cells, but not in Ishikawa cells, which in turn highly express the FOXO1 targeting E3-ubiquitin ligase Skp2. FOXO1 transcript levels were also lower in Ishikawa cells and treatment with the proteasomal inhibitor was insufficient to restore expression. Lack of FOXO1 expression in Ishikawa cells was not accounted for by differential promoter methylation or activity, but correlated with increased messenger RNA (mRNA) turnover. Comparative analysis demonstrated that HEC-1B cells proliferate slower, but are more resistant to paclitaxel-mediated cell death than Ishikawa cells, which were partially reversed upon silencing of FOXO1 in HEC-1B cells or its re-expression in Ishikawa cells. We further show that FOXO1 is required for the expression of the growth arrest- and DNA-damage-inducible gene GADD45α. Analysis of biopsy samples demonstrated a marked loss of FOXO1 and GADD45α mRNA and protein expression in endometrioid endometrial cancer compared to normal endometrium. Together, these observations suggest that loss of FOXO1 perturbs endometrial homeostasis, promotes uncontrolled cell proliferation and increases susceptibility to genotoxic insults.
Endometrial cancer is the most common malignancy of the female reproductive tract and its incidence is on the increase in North America and Europe (Amant et al., 2005; Shang, 2006). Endometrial carcinomas are divided into two groups; endometrioid (type I) and non-endometrioid (type II) endometrial cancers. The common type I tumours, which account for 80% of all endometrial carcinomas, are oestrogen-related, low-grade, affect pre- and perimenopausal women, and are often preceded by complex atypical endometrial hyperplasia. In contrast, type II tumours are not oestrogen-driven and mostly develop in atrophic endometrium. Loss or mutation of the tumour-suppressor gene PTEN is the earliest detectable genetic defect in type I endometrioid endometrial cancer (EEC) (Kong et al., 1997). PTEN encodes a lipid phosphatase that specifically dephosphorylates the D3 position of phosphatidylinositol 3,4,5-triphosphate, and thereby, functionally antagonizes the phosphatidylinositol-3-kinase/Akt (PI3K/Akt) signalling pathway. Consequently, loss of PTEN function results in constitutive activation of the PI3K/Akt signal transduction pathway, a hallmark of many cancers (Cully et al., 2006).
The FOXO subfamily of Forkhead transcription factors is a direct downstream target of the PI3K/Akt pathway (Brunet et al., 1999; Kops et al., 1999). This family comprises of three functionally related members, FOXO1, FOXO3a and FOXO4. FOXO proteins are evolutionarily conserved transcriptional activators of genes involved in cell cycle inhibition (for example, p27kip) (Dijkers et al., 2000), apoptosis (for example, BIM, FAS ligand and TRAIL) (Sunters et al., 2003), defence against oxidative stress (for example, manganese superoxide dismutase (MnSOD) and catalase), and DNA repair (for example, growth arrest- and DNA damage-inducible protein α of 45 kDa (GADD45α)) (Kops et al., 2002; Nemoto et al., 2004). PI3K/Akt-dependent phosphorylation of FOXO proteins triggers their export from the nucleus in a CRM1 transporter-dependent manner and promotes binding of these transcription factors to the 14-3-3 chaperone proteins in the cytosol. PI3K/Akt signalling further antagonizes FOXO-dependent transcription by earmarking the phosphorylated proteins for proteosomal degradation (Plas and Thompson, 2003). In addition to Akt, other kinases, such as SGK1 (serum- and glucocorticoid-inducible kinase 1), CK1 (casein kinase 1) and DYRK1A (dual-specificity tyrosine-phosphorylated and regulated 1A), have also been implicated in FOXO phosphorylation and nuclear export (Brunet et al., 2001; Woods et al., 2001; Rena et al., 2002).
FOXO1 expression in human endometrium increases during the secretory phase of the cycle. In contrast, FOXO3a is repressed in differentiating endometrium and FOXO4 appears not to be expressed in this tissue, at least not at protein level (Labied et al., 2006). Recently, we demonstrated that FOXO1 plays a major role in endometrial homeostasis by regulating cyclic differentiation (decidualization) and apoptosis of stromal cells in response to the rise and fall in ovarian progesterone levels (Kajihara et al., 2006). FOXO 1 is also abundantly expressed in endometrial epithelial cells, but its function in this cellular compartment is as yet unknown. In this study, we demonstrate that FOXO1 serves as a tumour suppressor in endometrial cancer cells, involved in normal growth control and maintenance of genomic stability.
Differential expression of FOXO1 in endometrial carcinoma cell lines
We hypothesized that inactivation of FOXO transcription factors could constitute a molecular link between dysregulation of the PTEN/PI3K/Akt signal cascade and endometrial tumorigenicity. To examine the role of FOXO1 in endometrial cancer, we first screened a panel of cell lines. Western blot analysis showed that FOXO1 is expressed in HEC-1B, HEC1A cells, but not in HEC-50B, HEC-108 and Ishikawa cells (data not shown). HEC-1B and Ishikawa cells are widely used for the study of endometrial cancer and chosen for further analysis. Although both cell lines are deficient in PTEN expression, PI3K/Akt activity was considerably higher in Ishikawa cells, as reflected by the phosphorylated Akt levels (Figure 1a). In addition, Ishikawa cells also express higher levels of Skp2, an oncogenic subunit of the Skp1/Cul1/F-box protein ubiquitin complex that promotes ubiquitination and degradation of phosphorylated FOXO1 (Huang et al., 2005). These observations indicated that lack of detectable FOXO1 levels in Ishikawa cells could be a consequence of constitutive PI3K/Akt-dependent phosphorylation of this transcription factor and proteasomal degradation. However, inhibition of the ubiquitin/proteasome pathway with MG132 resulted in accumulation of FOXO1 in HEC-1B cells, but not in Ishikawa cells (Figure 1b). This prompted us to examine FOXO1 transcript levels, which were eightfold lower in Ishikawa cells when compared to HEC-1B cells (Figure 1c). The results suggested that impaired gene expression, rather than targeted phosphorylation and proteasomal degradation, accounts for loss of FOXO1 in Ishikawa cells.
Loss of FOXO1 expression correlates with increased messenger RNA turnover
We hypothesized that the low level of FOXO1 messenger RNA (mRNA) expression in Ishikawa cells could reflect epigenetic silencing through hypermethylation of its promoter, a common mechanism of inactivating tumour suppressor genes in cancer. The proximal FOXO1 promoter harbours a CpG-rich region, defined as a stretch of DNA that has both a >50% GC content and an observed overexpected frequency of CpG dinucleotides of >0.6 (Gardiner-Garden and Frommer, 1987). We first used bisulphite genomic sequencing to determine the methylation status of a 300-bp region of the FOXO1 promoter region (−617 to −317 relative to the transcription start site), which encompasses 29 CpG sites. However, no differences were found in FOXO1 promoter methylation between HEC-1B and Ishikawa cells as determined by bisulphite genomic sequencing (Figure 2a). To confirm that the low expression of FOXO1 transcripts in Ishikawa cells was not due to promoter hypermethylation, we treated these cells with the DNA demethylating agent 5-aza-2′-deoxycytidine (5-aza-dC) for 24 or 48 h. As expected, real-time quantitative–PCR (RTQ–PCR) showed no difference in FOXO1 mRNA levels between treated and untreated cells (Figure 2b). Next, we transfected HEC-1B and Ishikawa cells with the FOXO1 promoter (−1845 to –7) coupled to luciferase and found that basal promoter activity was comparable between these cell lines (Figure 2c). Together, these data did not support our conjecture that FOXO1 transcription is impaired and suggested that FOXO1 transcripts are subjected to post-transcriptional regulation in endometrial cancer cells. To test this, we used actinomycin D to block new RNA synthesis in HEC-1B and Ishikawa cells and measured the decay of FOXO1 transcripts over a 5h period. As shown in Figure 2d, the half-life of FOXO1 mRNA in Ishikawa cells was 1.8±0.05 h compared to 4.5±0.18 h in HEC-1B cells. Thus, lack of FOXO1 expression in Ishikawa cells is accounted for by increased FOXO1 mRNA turnover but not by lack of transcription initiation.
FOXO1 attenuates endometrial cancer cell proliferation
To provide insights into the functional consequences of FOXO1 deficiency in endometrial carcinogenesis, we first compared the growth curves of HEC-1B and Ishikawa cells. As shown in Figure 3a, Ishikawa cells proliferate considerably faster than HEC-1B cells, as determined by the relative increase in cell viability. Analysis of growth curves based on cell number showed a doubling-time of 7.8±1.8 h for Ishikawa cells and 17.5±3.5 h for HEC-1B cells (data not shown). To examine whether FOXO1 is involved in growth control, we transfected HEC-1B cells with either non-targeting (NT) short interfering RNA (siRNA) or FOXO1 siRNA. To ensure adequate FOXO1 knockdown, growth analysis was commenced 48 h after siRNA transfection. In addition, Ishikawa cells were stably transfected with an empty expression vector (pcDNA3.1) or with pcDNA3.1/FOXO1(A3), a plasmid that encodes for a constitutively active FOXO1 mutant in which the three conserved PI3K/Akt phosphorylation sites are changed to alanine (Christian et al., 2002) (Figure 3b). FOXO1 knockdown in HEC-1B cells increased cellular proliferation determined by MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt) assay shown in Figure 3c, whereas overexpression of FOXO1(A3) in Ishikawa cells, even at relatively modest levels, inhibited proliferation (Figure 3d), increasing the doubling-time twofold. Notably, silencing or expression of FOXO1 in these endometrial cell lines did not trigger apoptosis as determined by flow cytometry of propidium iodide (PI)-stained cells or by western blot analysis of cleaved poly(ADP-ribose) polymerase-1 (PARP) (see Figures 5c, d and 6).
FOXO1 confers resistance to paclitaxel-induced apoptosis
FOXO proteins have been shown to mediate programmed cell death in response to paclitaxel in other cell systems, including breast cancer cells (Sunters et al., 2003, 2006). To determine whether FOXO1 plays a role in mediating cellular sensitivity to cytotoxic drugs in endometrial cancer cells, we first treated HEC-1B and Ishikawa cells with increasing concentrations of paclitaxel, ranging from 0.1 to 100 nM, for 24 h. As shown in Figure 4a, loss of cell viability was much more pronounced in Ishikawa cells than HEC-1B cells when exposed to paclitaxel at concentrations between 10 and 100 nM. Flow cytometry confirmed that the fraction of dead or dying cells containing <2n DNA was considerably higher in Ishikawa cells than in HEC-1B cells after 24 h of treatment with 10 nM paclitaxel (Figure 4b). Interestingly, paclitaxel markedly enhanced FOXO1 promoter activity (Figure 4c), mRNA expression (Figure 4d) and protein levels (Figure 4e) in HEC-1B cells, but only weakly so in Ishikawa cells. In addition, western blot analysis of cellular fractions demonstrated that paclitaxel also induced nuclear accumulation of FOXO1 in HEC-1B cells, which peaked around 24 h of stimulation (Figure 4f). The nuclear increase correlates with a transient reduction in phosphorylated (Ser256) FOXO1 levels. Next, we determined the apoptotic response to paclitaxel in HEC-1B and Ishikawa cells transfected with FOXO1 siRNA and pcDNA3.1/FOXO1(A3), respectively. FOXO1 knockdown in HEC-1B resulted in increased programmed cell death in response to paclitaxel, as determined by loss of cell viability (Figure 5a), flow cytometry of PI-stained cells (Figure 5c) and cleaved PARP expression (see Figure 6a). Conversely, expression of FOXO1(A3) in Ishikawa cells conferred resistance to paclitaxel-induced cell death (Figures 5b, d and 6b). Together, these data provide unequivocal evidence that FOXO1 regulates the apoptotic responses in endometrial cancer cells exposed to cytotoxic stress.
FOXO1 regulates GADD45α expression
To further explore the loss of FOXO1 in endometrial cancer cells, we examined HEC-1B and Ishikawa cells for the expression of GADD45α, a putative FOXO target gene involved in DNA repair and maintaining genomic stability (Tran et al., 2002). As shown in Figures 6a and b, GADD45α was readily detectable in HEC-1B, but not in Ishikawa cells by western blot analysis. Transfection of FOXO1 siRNA attenuated GADD45α levels in HEC-1B cells (Figure 6a), whereas expression of FOXO1(A3) was sufficient to induce its expression in Ishikawa cells (Figure 6b). Paclitaxel treatment for 24 h enhanced GADD45α levels in HEC-1B cells and induced its expression in Ishikawa cells. Interestingly, while the response to paclitaxel was inhibited in HEC-1B cells transfected with FOXO1 siRNA, it was independent of FOXO1(A3) expression in Ishikawa cells. In fact, paclitaxel-mediated induction of GADD45α in Ishikawa cells in the absence of overexpressed FOXO1 coincided with increased cell death as determined by PARP cleavage (Figure 6b).
Loss of FOXO1 and GADD45α expression in EEC
Our in vitro data suggested that loss of FOXO1 expression could be an important event in endometrial carcinogenesis. To test this hypothesis, we first examined the expression of FOXO1 transcripts in 14 normal endometrial samples, taken randomly in the cycle, and in 25 type I EECs by RTQ–PCR. In addition, we examined the expression of transcripts that encode for the other FOXO family members, FOXO3a and FOXO4. As shown in Figure 7a, FOXO3a and FOXO4 mRNA levels did not differ between normal and malignant endometrium. In contrast, FOXO1 transcripts were approximately sixfold less abundant in EEC than in cycling endometrium. To substantiate this finding, we examined FOXO1 immunoreactivity in 15 normal endometrial samples and 35 EECs (Figure 7b). FOXO1 immunostaining in proliferative endometrium was confined to epithelial cells and its expression in this compartment increased further during the secretory phase of the cycle. In addition, decidualizing stromal cells in secretory endometrium were also strongly FOXO1 immunopositive. FOXO1 was not detectable in 77% (27/35) of EECs (Figure 7b). In the remaining cases, FOXO1 immunoreactivity was weak and, in contrast to cycling endometrium, often confined to the cytoplasm. To confirm these findings, 31 EEC needle core biopsies were spotted on a tissue microarray (TMA). FOXO1 immunoreactivity was undetectable in 10 out of 13 (77%) grade I EECs, 6 out of 8 (75%) grade II and 8 out of 10 (80%) grade III tumours (data not shown). Together, these data point towards a marked reduction in FOXO1 activity in EEC, which is, at least partially, accounted for by impaired FOXO1 mRNA expression.
To further determine the in vivo relevance of loss of FOXO1 in endometrial cancer cells, we also compared the levels of GADD45α transcripts in normal endometrium and EECs. RTQ–PCR analysis demonstrated a fourfold reduction in GADD45α mRNA expression in cancer samples when compared to normal endometrium (Figure 7c), paralleling the downregulation of FOXO1 transcripts. This was further validated by immunohistochemistry. Like FOXO1, GADD45α was expressed in normal endometrial epithelial cells throughout the cycle, but staining in the stromal compartment was confined to decidualizing cells, during the secretory phase of the cycle (Figure 7d). GADD45α immunoreactivity was weak or absent in a majority of EECs, especially in cases with lots of necrosis, and the findings confirmed by staining the TMA. Thus, the loss of GADD45α immunoreactivity was comparable, albeit less pronounced, to that of FOXO1 (Figure 7b).
Finally, we used methylation-specific PCR analysis to examine in vivo the methylation status of the FOXO1 promoter in a limited number of normal endometrial samples and EECs. As shown in Figure 8, the FOXO1 promoter region was unequivocally unmethylated in the seven endometrial biopsies obtained from normal cycling women. The results of this analysis in EEC were much more heterogeneous, showing partial promoter methylation in five cases, lack of methylation in four cases and complete methylation in a single case.
The cyclical waves of proliferation, differentiation, shedding and regeneration of human endometrium, which occur on average 400 times during reproductive life, are unparalleled in any other tissue of the body. Endometrial cancer is the most common malignancy of the female reproductive tract (Amant et al., 2005; Shang, 2006), yet the molecular safeguards that ensure endometrial homeostasis are poorly understood. FOXO transcription factors have emerged as key mediators of cell fate decisions through their ability to regulate seemingly opposing cellular responses, such as differentiation, apoptosis, cell cycle arrest and defence responses against oxidative and genotoxic stress (Dijkers et al., 2000; Kops et al., 2002; Sunters et al., 2003; Nemoto et al., 2004). Recent studies in human endometrium have provided further evidence for this ability of FOXO transcription factors to regulate diverse gene programmes in response to changing hormonal and environmental cues (Kajihara et al., 2006; Labied et al., 2006). For example, FOXO1 critically regulates the expression of decidual marker genes, such as PRL and IGFBP-1, in differentiating human endometrial stromal cells in response to elevated progesterone levels, but elicits BIM (BCL2L11) expression, a BH3 domain protein that induces apoptosis of decidualized cells upon progesterone withdrawal.
In contrast to stromal cells, FOXO1 is constitutively expressed in normal endometrial epithelial cells throughout the cycle. Several strands of existing evidence indicated that loss of FOXO1 activity in epithelial cells would be an integral event in neoplastic transformation of endometrium, especially in EECs. First, loss or mutation of the tumour-suppressor gene PTEN is the earliest detectable genetic defect in EECs (Kong et al., 1997) and leads to unrestrained PI3K/Akt signalling, which, in turn, would phosphorylate and inactivate FOXO transcription factors (Brunet et al., 1999; Kops et al., 1999). Second, Skp2, the oncogenic subunit of the Skp1/Cul1/F-box protein complex that directs FOXO1 ubiquitination and proteasomal degradation, is overexpressed in EECs (Lahav-Baratz et al., 2004). Third, the expression of known FOXO target genes, such as the cyclin-dependent kinase inhibitor p27kip, is dramatically lower in malignant endometrium (Bamberger et al., 1999). Moreover, unopposed oestrogen signalling, a well-recognized risk factor for EEC (Amant et al., 2005; Shang, 2006), has been shown to enhance PI3K/Akt activity through binding of oestrogen receptor α (ERα) to the p85α regulatory subunit of PI3K (Sun et al., 2001). Conversely, recent studies in heterozygote Pten+/− mice, which develop EECs with full penetrance, demonstrated that enhanced PI3K/Akt signalling leads to targeted phosphorylation of ERα on Ser167 and ligand-independent activation, thus mimicking a hyperestrogenic environment (Vilgelm et al., 2006). Furthermore, FOXO1 engages in the transcriptional cross-talk with ERα through physical interaction (Schuur et al., 2001), suggesting an important role for FOXO1 in mediating oestrogen responses in target tissues.
This study shows that FOXO1 activity is indeed lost in EECs. Using relevant cell line models, we demonstrate that FOXO1 not only limits proliferation of endometrial cancer cells, but also that its induction and activation confers resistance to paclitaxel. This appears to be in contrast to breast cancer cells, where paclitaxel treatment results in c-Jun NH2-terminal kinase 1/2-dependent activation of FOXO3a, but not FOXO1, which in turn increases BIM expression and elicits apoptosis (Sunters et al., 2003, 2006). It is possible that altered sensitivity to paclitaxel mediated by FOXO1 expression is secondary to different cellular growth. Toxicity to anticancer drugs such as paclitaxel involves generation of reactive oxygen species, resulting in indiscriminate damage to proteins, lipids and DNA (Varbiro et al., 2001; Laurent et al., 2005). Therefore, another mechanism whereby FOXO1 could limit cytotoxity involves induction of free radical scavengers, such as the mitochondrial MnSOD (Kops et al., 2002; Nemoto et al., 2004), which is indeed the case in endometrial cancer cell lines (data not shown). Other mechanism involves the expression of DNA-repair genes. GADD45α expression is induced in many cell types by cellular stress and various DNA-damaging agents, such as ultraviolet light, ionizing radiation and alkylating agents, and causes G2/M cell cycle arrest and DNA repair (Tran et al., 2002; Brunet et al., 2004; Lal et al., 2006). GADD45 proteins can also trigger apoptosis of cells damaged beyond repair. We have now shown that GADD45α is constitutively expressed in normal endometrial epithelial cells, suggesting that it has an important role safeguarding genomic stability during the cyclical waves of proliferation and shedding. In addition, we demonstrated that basal GADD45α expression in endometrial cells is dependent on FOXO1 activity and, in agreement, a marked reduction in GADD45α levels was observed in ECCs, paralleling the loss of FOXO1 expression. However, transcriptional induction of GADD45α by cytotoxic stress also involves other transcription factors such as p53 and simultaneous de-repression from factors such as c-myc and ZBRK1 (Zheng et al., 2000; Tran et al., 2002). These alternative mechanisms may explain the induction of GADD45α in paclitaxel-treated Ishikawa cells, which appeared to be independent of FOXO1.
An important observation is that loss of FOXO1 in EECs may not primarily reflect increased PI3K/Akt signalling. Approximately 75% of both low- and high-grade tumours were devoid of FOXO1 staining, indicating that loss of this transcription factor may be an early event in endometrial carcinogenesis. In addition, we found no correlation between FOXO1 and PTEN staining (data not shown), but admittedly a larger clinicopathological study is required to validate these findings, to examine the expression of FOXO1 in hyperplastic endometrial lesions and to assess if lack of FOXO1 correlates with the presence of PTEN mutations. However, we did find a sixfold reduction in FOXO1 transcript levels in neoplastic, when compared to cycling endometrium. This observation concurs with the findings of a recent gene microarray study, demonstrating lower FOXO1 mRNA expression not only in EEC, but also in non-endometrioid cancers when compared to cycling endometrium (Risinger et al., 2003). However, this study also reported lower FOXO4 transcript levels in EEC, an observation that we were unable to confirm by RTQ–PCR. The mechanism responsible for lack of FOXO1 expression in EEC is not entirely clear. In contrast to normal endometrium, the FOXO1 promoter was methylated in 6 out of 10 EECs, although the degree varied considerably between tumour samples. Thus, our findings suggest that methylation events may not be causal to the loss of FOXO1 expression in EECs, although the results are not conclusive. Comparative analysis of HEC-1B and Ishikawa cells demonstrated that loss of FOXO1 expression did correlate with increased decay of its mRNA. It is interesting to note that microRNAs complementary to FOXO1 have been identified in several species, including humans (http://microrna.sanger.ac.uk/). Furthermore, FOXO1 mRNA has an extended 3′-UTR that contains several AUUA pentamers, indicating potential regulation by RNA-binding proteins (Yeap et al., 2002).
From a clinical perspective, dysregulation of FOXO transcription factors has been described in several other cancers, including breast cancer, prostate cancer and leukaemia (Arden, 2006). This suggests that they may be useful therapeutic targets and a screen of small molecules has identified several compounds capable of enhancing FOXO1 activity by disabling its nuclear export (Kau et al., 2003). If our observations in endometrial cell lines are recapitulated in vivo then inhibiting FOXO1 mRNA turnover may be necessary to enhance the efficacy of such compounds in the treatment of EEC.
Materials and methods
Human endometrial cancer cells (HEC-1B and Ishikawa cells) were maintained in Dulbecco's modified Eagle's medium/F12 (GIBCO Industries Inc., Carlsland, CA, USA) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT, USA), 2 mM L-Glutamine and 100 U/ml penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA), in a humidified incubator in an atmosphere of 5% CO2 at 37°C, and routinely tested for mycoplasma infection. Paclitaxel was obtained from Sigma-Aldrich (Poole, UK) and dissolved in dimethyl sulphoxide (DMSO).
Western blot analysis
Whole-cell extracts or nuclear and cytoplasmic protein fractions were immunoblotted as described previously (Brosens et al., 1999; Zoumpoulidou et al., 2004). Antibodies to FOXO1, phospho-FOXO1 (Ser256), Akt, phospho-Akt (Ser473) (Cell Signaling Technology, Danvers, MA, USA), Skp2 (Invitrogen), GADD45α (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and PARP Cleavage Site (214/215) (Biosource, Carmavillo, CA, USA) were used at 1:1000, whereas the antibody to β-actin (Abcam, Cambridge, UK) was diluted 1:100 000.
Total RNA was isolated from endometrial cancer cell lines and tissue samples, complementary DNA prepared and amplified as described previously (Labied et al., 2006). The following gene-specific primer pairs were used: L19-sense (5′-IndexTermGCG GAA GGG TAC AGC CAA T-3′) and L19-antisense (5′-IndexTermGCA GCC GGC GCA AA-3′); FOXO1-sense (5′-IndexTermTGG ACA TGC TCA GCA GAC ATC-3′) and FOXO1-antisense (5′-IndexTermTTG GGT CAG GCG GTT CA-3′); FOXO3a-sense (5′-IndexTermCCC AGC CTA ACC AGG GAA GT-3′) and FOXO3a-antisense (5′-IndexTermAGC GCC CTG GGT TTG G-3′); FOXO4-sense (5′-IndexTermCCT GCA CAG CAA GTT CAT CAA-3′) and FOXO4-antisense (5′-IndexTermTTC AGC ATC CAC CAA GAG CTT-3′); GADD45α-sense (5′-IndexTermTCT CGG CTG GAG AGC A-3′) and GADD45α-antisense (5′-IndexTermGGC TTT GCT GAG CAC T-3′). L19, a non-regulated ribosomal housekeeping gene, served as an internal control and was used to normalize for differences in input RNA. All measurements were performed in triplicate.
Bisulphite genomic sequencing and demethylation
The proximal FOXO1 promoter region was first analysed for the presence of CpG islands using the CpG Island Searcher (www.cpgislands.com). Next, genomic DNA was extracted from HEC-1B and Ishikawa cells, using the Dneasy Tissue Kit (Qiagen, Crawley, UK) and modified with sodium bisulphite using the CpGenome DNA Modification Kit (Chemicon, Temecula, CA, USA). PCR amplification of the FOXO1 promoter region (−617 to −317 relative to the transcription start site) was performed using Hot Start Taq Polymerase (Qiagen) with the following primers designed using the MethPrimer program (Li and Dahiva, 2002): 5′-IndexTermGAA AAT ATT AAA TTA AAA TAA AAT TTA T-3′ (forward) and 5′-IndexTermTTT AAT TAC TAA AAA ACA AAC CAA C-3′ (reverse). The amplicons were purified from a 2% agarose gel with QIUquick gel extraction kit (Qiagen) and sequenced. Universally Unmethylated DNA and Universally Methylated DNA (Chemicon) were used as negative and positive controls, respectively.
Cloning of the FOXO1 promoter
The proximal human FOXO1 promoter region, −1845 to −7 bp relative to the transcription start site, was amplified by PCR from genomic DNA extracted from MCF-7 cells using the following primers with overhangs to incorporate, respectively, Kpn1 and HindIII restriction sites (underlined): FOXO1pro(−7) 5′-IndexTermGGC CGG TAC C CCT AAT TTT TCC TTT TTT CCC CTC-3′ and FOXO1pro(−1845) 5′-IndexTermCCG GAA GCT TGA GTG GAA GCG CGA GCC-3′. The PCR product was restriction digested and cloned into the Kpn1 and HindIII sites of the pGL3-basic promoter luciferase reporter vector (Promega, Madison, WI, USA) and verified by sequence analysis.
Transient and stable transfections
To assess basal FOXO1 promoter activity, HEC-1B and Ishikawa cells were plated in 96-well dishes (approximately 7000 cells/well) and transfected the following day using Lipofectamine 2000 (Invitrogen), according to the manufacturer's instruction. The transfections were performed in quadruplicates, using 20 ng of reporter construct and 5 ng of pRLTK Renilla plasmid (Promega), and cultures lysed 24 h later in 20 μl/well of Passive Lysis Buffer (Promega). Promoter activity was determined using the Dual-Luciferase Reporter Assay kit (Promega), according to manufacturer's instructions. Luciferase activity was normalized by dividing the luciferase reading by the Renilla reading. For silencing experiments, HEC-1B cells cultured in six-well plates were transfected with 50 nM of FOXO1 siGENOME SMARTpool or NT siRNA pool (Dharmacon, Lafayette, CO, USA) using Lipofectamine 2000 (Invitrogen), according to the manufacture's specifications. FOXO1 knockdown was confirmed by western blot analysis in all experiments. Ishikawa cells were transfected with control pcDNA3.1 or with pcDNA3.1/FOXO1(A3) (Christian et al., 2002) using Lipofectamine2000 (Invitrogen), re-plated and incubated with 600 μg/ml G418 sulphate (Invitrogen). Individual G418-resistant colonies were isolated after 2 weeks of selection and expanded in the presence of G418 (100 μg/ml). Three clonal lines stably expressing FOXO1 were mixed to eliminate interclonal bias and used for functional analyses.
Analysis of FOXO1 mRNA stability
Ishikawa and HEC-1B cells were incubated with the transcriptional inhibitor actinomycin D (Sigma-Aldrich) at a final concentration of 5 μg/ml and total mRNA extracted either immediately (designated time point 0 h) or at hourly intervals over the next 5 h. RTQ–PCR was used to determine the amount of FOXO1 mRNA, relative to the abundance of L19 transcripts, at each time point and the half-life of FOXO1 mRNA determined by calculating linear regressions to a log-linear plot of mRNA expression versus time.
Cell proliferation assay and flow cytometry
Endometrial cancer cells were plated in 24-well plates (2 × 103 cells/cm2) for growth curve analysis. At the indicated time points, cells were trypsinized and detached from the plates, and cell number was counted under a microscopy using a hemocytometer. Each experiment was performed in quadruplicate. For MTS assay, cells were seeded onto 96-well plates, at approximately 2000 or 10 000 cells/well for proliferation and cytotoxicity assays, respectively, and allowed to attach overnight. Cell viability was determined by MTS assay using the CellTiter 96 Aqueous One Solution Cell Proliferation Assay (Promega), according to manufacturer's instructions. To study the effects of paclitaxel on cell proliferation, cells were treated with various doses of paclitaxel for 24 h. After completion of the treatment, the percentage absorbance was calculated against untreated cells. Flow cytometry analysis was used to quantify apoptosis in endometrial cancer cells by evaluating the sub-G1 fraction (<2n) after PI staining of ethanol-fixed cells.
The participating Local Research and Ethics Committees approved the study and patient consent was obtained before tissue collection. Snap-frozen endometrial cancer samples were obtained from patients undergoing hysterectomy without preoperative chemotherapy or radiation and histologically validated for type and grade. Normal endometrial samples were obtained from premenopausal women awaiting in vitro fertilization treatment for either tubal or male-factor infertility or at the time of laparoscopy. These samples were immersed in RNAlater (Ambion Inc, Austin, TX, USA) and stored at −80°C until use.
Paraffin-embedded, formalin-fixed endometrial specimens were immunostained for FOXO1 and GADD45α using the Universal LSAB Plus Kits (DAKO North America, Inc., Carpinteria, CA, USA) as described previously (Kajihara et al., 2006; Labied et al., 2006), using primary antibodies against FOXO1 (Cell Signalling Technology Inc.; 1:50 dilution) and GADD45α (Santa Cruz Biotechnology Inc.; 1:100 dilution). Histological and immunohistochemical assessments were performed by two independent pathologists. Every tumour was given a score by multiplying the intensity of the staining (no staining=0; low staining=1; medium staining=2; strong staining=3) with the percent of stained epithelial cells (0%=0; under 10%=1; 1–50%=2: 51–80%=3; over 80%=4). The final immunoreactive score was graded as follows: negative, 0; low, 1–3; medium, 4–7; and high, 8–12. To validate the findings, needle core biopsies of 31 EECs were used to construct a TMA and stained for FOXO1 and GADD45α.
Methylation-specific PCR analysis
Sodium bisulphite conversion of FOXO1 5′ CpG island was performed in 3 μg of genomic DNA obtained from 17 frozen samples, 7 normal endometrial biopsies and 10 EECs. Briefly, DNA was denatured by incubation with NaOH (final concentration 0.2 M), for 10 min, at 50°C and mixed with 1 μg of salmon sperm DNA (FLUKA). Freshly made sodium bisulphite solution (2.5 M sodium bisulphite (Sigma, St Louis, MO, USA), 125 mM hydroquinone (Sigma) and NaOH 2 M, pH=5.0) was added to the DNA sample to convert the unmethylated cytosines to uracils, allowing the distinction between methylated and unmethylated DNA. After 3 h incubation in the dark at 70°C, converted DNA was purified, using a commercial Wizard DNA purification resin, and eluted into 45 μl of pre-heated (70°C) water. Modification was completed by NaOH treatment (final concentration 0.3 M) for 10 min at room temperature, followed by neutralization of the reaction with 75 μl of 7.5 M ammonium acetate. DNA precipitation was performed, the pellet washed with 70% ethanol, the DNA dried, resuspended in 40 μl of water and stored at −20°C. DNA samples were amplified by PCR. A total of 400 ng of bisulphite-modified DNA was added to the PCR mix containing × 1 PCR buffer (16.6 mM ammonium sulphate, 67 mM Tris (pH 8.8), 6.7 mM MgCl2, and 10 mM 2-mercaptoethanol), deoxynucleotide triphosphates (each at 1 mM), 0.6 μ M primers, 2% DMSO and 1unit Platinum TaqDNA polymerase (Invitrogen). Gene methylation specific primers for both methylated (FOXO1-M forward: 5′-IndexTermGGTACGGAGAAGGGTTTTTC-3′, FOXO1-M reverse: 5′-IndexTermACGATATCTTTACTAAACGACGT-3′) and unmethylated (FOXO1-U forward: 5′-IndexTermGGTATGGAGAAGGGTTTTTTG-3′, FOXO1-U reverse: 5′-IndexTermCCCACAATATCTTTACTAAACAACATA-3′) FOXO1 promoter were used to recognize and amplify FOXO1 methylated and unmethylated DNA promoter, respectively. Reactions were hot-started at 95°C for 5 min and PCR amplification was carried out in a thermocycler for 38 cycles (30 s at 95 °C, 30 s at 55°C, the annealing temperature, and 45 s at 72°C). A final extension for 10 min at 72°C was performed. Negative controls, without DNA, were performed for all PCR reactions. Each mutagenetically separated-PCR product was directly loaded into a high-resolution precast 2% agarose Seakem gel (Cambrex), and stained with SYBR Gold (Invitrogen).
All values are presented as mean±s.d. where appropriate. Statistical significance between two groups was determined by use of a two-tailed t-test, and values of P<0.05 were considered significant.
endometrioid endometrial cancer
growth arrest- and DNA damage-inducible protein α of 45 kDa
Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I . (2005). Endometrial cancer. Lancet 366: 491–505.
Arden KC . (2006). Multiple roles of FOXO transcription factors in mammalian cells point to multiple roles in cancer. Exp Gerontol 41: 709–717.
Bamberger AM, Riethdorf L, Milde-Langosch K, Bamberger CM, Thuneke I, Erdmann I et al. (1999). Strongly reduced expression of the cell cycle inhibitor p27 in endometrial neoplasia. Virchows Arch 434: 423–428.
Brosens JJ, Hayashi N, White JO . (1999). Progesterone receptor regulates decidual prolactin expression in differentiating human endometrial stromal cells. Endocrinology 140: 4809–4820.
Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS et al. (1999). Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96: 857–868.
Brunet A, Park J, Tran H, Hu LS, Hemmings BA, Greenberg ME . (2001). Protein kinase SGK mediates survival signals by phosphorylating the forkhead transcription factor FKHRL1 (FOXO3a). Mol Cell Biol 21: 952–965.
Brunet A, Sweeney LB, Sturqill JF, Chua KF, Greer PL, Lin Y et al. (2004). Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase. Science 303: 2011–2015.
Christian M, Zhang X, Schneider-Merck T, Unterman TG, Gellersen B, White JO et al. (2002). Cyclic AMP-induced forkhead transcription factor, FKHR, cooperates with CCAAT/enhancer-binding protein beta in differentiating human endometrial stroma cells. J Biol Chem 277: 20825–20832.
Cully M, You H, Levine AJ, Mak TW . (2006). Beyond PTEN mutations: the PI3K pathway as an integrator of multiple during tumourgenesis. Nat Rev Cancer 6: 184–192.
Dijkers PF, Medema RH, Pals C, Banerji L, Thomas NS, Lam EW et al. (2000). Forkhead transcription factor FKHR-L1 modulates cytokine-dependent transcriptional regulation of p27(KIP1). Mol Cell Biol 20: 9138–9148.
Gardiner-Garden M, Frommer M . (1987). CpG islands in vertebrate genomes. J Mol Biol 196: 261–282.
Huang H, Regan KM, Wang F, Wang D, Smith DI, van Deursen JM et al. (2005). Skp2 inhibits FOXO1 in tumour suppression through ubiquitin-mediated degradation. Proc Natl Acad Sci USA 102: 1649–1654.
Kajihara T, Jones M, Fusi L, Takano M, Feroze-Zaidi F, Pirianov G et al. (2006). Differential expression of FOXO1 and FOXO3a confers resistance to oxidative cell death upon endometrial decidualization. Mol Endocrinol 20: 2444–2455.
Kau TR, Schroeder F, Ramaswamy S, Wojciechowski CL, Zhao JJ, Roberts TM et al. (2003). A chemical genetic screen identifies inhibitors of regulated nuclear export of a Forkhead transcription factor in PTEN-deficient tumor cells. Cancer Cell 4: 463–476.
Kong D, Suzuki A, Zou TT, Sakurada A, Kemp LW, Wakatsuki S et al. (1997). PTEN1 is frequently mutated in primary endometrial carcinomas. Nat Genet 17: 143–144.
Kops GJ, Dansen TB, Polderman PE, Saarloos I, Wirtz KW, Coffer PJ et al. (2002). Forkhead transcription factor FOXO3a protects quiescent cells from oxidative stress. Nature 419: 316–321.
Kops GJ, de Ruiter ND, De Vries-Smits AM, Powell DR, Bos JL, Burgering BM . (1999). Direct control of the Forkhead transcription factor AFX by protein kinase B. Nature 398: 630–634.
Labied S, Kajihara T, Madureira PA, Fusi L, Jones MC, Higham JM et al. (2006). Progestins regulate the expression and activity of the forkhead transcription factor FOXO1 in differentiating human endometrium. Mol Endocrinol 20: 35–44.
Lahav-Baratz S, Ben-Izhak O, Sabo E, Ben-Eliezer S, Lavie O, Ishai D et al. (2004). Decreased level of the cell cycle regulator p27 and increased level of its ubiquitin ligase Skp2 in endometrial carcinoma but not in normal secretory or in hyperstimulated endometrium. Mol Hum Reprod 10: 567–572.
Lal A, Abdelmohsen K, Pullmann R, Kawai T, Galban S, Yang X et al. (2006). Posttranscriptional derepression of GADD45alpha by genotoxic stress. Mol Cell 22: 117–128.
Laurent A, Nicco C, Chereau C, Goulvestre C, Alexandre J, Alves A et al. (2005). Controlling tumor growth by modulating endogenous production of reactive oxygen species. Cancer Res 65: 948–956.
Li LC, Dahiva R . (2002). MethPrimer: designing primers for methylation PCRs. Bioinformatics 18: 1427–1431.
Nemoto S, Fergusson MM, Finkel T . (2004). Nutrient availability regulates SIRT1 through a forkhead-dependent pathway. Science 306: 2105–2108.
Plas DR, Thompson CB . (2003). Akt activation promotes degradation of tuberin and FOXO3a via the proteasome. J Biol Chem 278: 12361–12366.
Rena G, Woods YL, Prescott AR, Peggie M, Unterman TG, Williams MR et al. (2002). Two novel phosphorylation sites on FKHR that are critical for its nuclear exclusion. EMBO J 21: 2263–2271.
Risinger JI, Maxwell GL, Chandramouli GV, Jazaeri A, Aprelikova O, Patterson T et al. (2003). Microarray analysis reveals distinct gene expression profiles among different histologic types of endometrial cancer. Cancer Res 63: 6–11.
Schuur ER, Loktev AV, Sharma M, Sun Z, Roth RA, Weigel RJ . (2001). Ligand-dependent interaction of estrogen receptor-alpha with members of the forkhead transcription factor family. J Biol Chem 276: 33554–33560.
Shang Y . (2006). Molecular mechanisms of oestrogen and SERMs in endometrial carcinogenesis. Nat Rev Cancer 6: 360–368.
Sun M, Paciga JE, Feldman RI, Yuan Z, Coppola D, Lu YY et al. (2001). Phosphatidylinositol-3-OH Kinase (PI3K)/AKT2, activated in breast cancer, regulates and is induced by estrogen receptor alpha (ERalpha) via interaction between ERalpha and PI3K. Cancer Res 61: 5985–5991.
Sunters A, Fernandez de Mattos S, Stahl M, Brosens JJ, Zoumpoulidou G, Saunders CA et al. (2003). FOXO3a transcriptional regulation of Bim controls apoptosis in paclitaxel-treated breast cancer cell lines. J Biol Chem 278: 49795–49805.
Sunters A, Madureira PA, Pomeranz KM, Aubert M, Brosens JJ, Cook SJ et al. (2006). Paclitaxel-induced nuclear translocation of FOXO3a in breast cancer cells is mediated by c-Jun NH2-terminal kinase and Akt. Cancer Res 66: 212–220.
Tran H, Brunet A, Grenier JM, Datta SR, Fornace Jr AJ, DiStefano PS et al. (2002). DNA repair pathway stimulated by the forkhead transcription factor FOXO3a through the Gadd45 protein. Science 296: 530–534.
Varbiro G, Veres B, Gallyas Jr F, Sumeqi B . (2001). Direct effect of Taxol on free radical formation and mitochondrial permeability transition. Free Radic Biol Med 31: 548–558.
Vilgelm A, Lian Z, Wang H, Beauparlant SL, Klein-Szanto A, Ellenson LH et al. (2006). Akt-mediated phosphorylation and activation of estrogen receptor alpha is required for endometrial neoplastic transformation in Pten+/− mice. Cancer Res 66: 3375–3380.
Woods YL, Rena G, Morrice N, Barthel A, Becker W, Guo S et al. (2001). The kinase DYRK1A phosphorylates the transcription factor FKHR at Ser329 in vitro, a novel in vivo phosphorylation site. Biochem J 355: 597–607.
Yeap BB, Voon DC, Vivian JP, McCulloch RK, Thomson AM, Giles KM et al. (2002). Novel binding of HuR and poly(C)-binding protein to a conserved UC-rich motif within the 3′-untranslated region of the androgen receptor messenger RNA. J Biol Chem 277: 27183–27192.
Zheng L, Pan H, Li S, Flesken-Nikitin A, Chen PL, Boyer TG et al. (2000). Sequence-specific transcriptional corepressor function for BRCA1 through a novel zinc finger protein, ZBRK1. Mol Cell 6: 757–768.
Zoumpoulidou G, Jones MC, Fernandez de Mattos S, Francis JM, Fusi L, Lee YS et al. (2004). Convergence of interferon-gamma and progesterone signalling pathways in human endometrium: role of PIASy. Mol Endocrinol 18: 1988–1999.
This work was financially supported by the Great Britain Sasakawa Foundation and the IOG Trust. We wish to thank Dr Akihiko Suenaga for his technical support.
The authors have nothing to disclose.
About this article
Forced FoxO1:S249V expression suppressed glioma cell proliferation through G2/M cell cycle arrests and increased apoptosis
Neurological Research (2019)
Assessing Estrogen-Induced Proliferative Response in an Endometrial Cancer Cell Line Using a Universally Applicable Methodological Guide
International Journal of Gynecologic Cancer (2018)
Long noncoding RNA LINC00261 regulates endometrial carcinoma progression by modulating miRNA/FOXO1 expression
Cell Biochemistry and Function (2018)
Seminars in Cancer Biology (2018)
Seminars in Cancer Biology (2018)