Endometrial adenocarcinoma (EndoCA) is the most common gynecologic cancer type in the United States, and its incidence is increasing. The majority of patients are disease-free after surgical resection of stage I tumors, which is often followed by radiotherapy, but most patients with advanced disease recur and have a poor prognosis, largely because the tumors become refractory to cytotoxic chemotherapies. PTEN, a commonly mutated tumor suppressor in EndoCAs, is well known for its ability to inhibit the AKT/mTOR signaling pathway. Nuclear functions for PTEN have been proposed as well, but whether those affect EndoCA development, progression, or outcomes is not well understood. Using immunohistochemistry, nuclear PTEN expression was observed in approximately half of EndoCA patient tumors, independent of grade and cytoplasmic PTEN expression. Higher levels of the DNA damage response (DDR) marker, γH2AX, were observed by immunohistochemistry and immunofluorescence in human EndoCA tumor sections that were PTEN-negative, in murine EndoCA tissues that were genetically modified to be PTEN-null, and in Ishikawa EndoCA cells, which do not express endogenous PTEN. Overexpression of exogenous PTEN-WT or PTEN-NLS, a modified PTEN with an added nuclear localization signal, significantly improved both DDR and G2–M transition in Ishikawa cells treated with a DNA-damaging agent. Whereas PARP inhibition with Olaparib was not as effective in Ishikawa cells expressing native or PTEN-NLS, inhibition with Talazoparib was not affected by PTEN overexpression. These results suggest that nuclear PTEN subcellular localization in human EndoCA could be diagnostic when considering DDR therapeutic intervention. Mol Cancer Ther; 17(9); 1995–2003. ©2018 AACR.

Endometrial adenocarcinoma (EndoCA), a cancer that arises from the endometrial epithelium (1), is the most common gynecologic tumor in developed countries and has the second highest incidence in developing countries after cervical cancer (2). Although EndoCA is diagnosed during or after menopause with abnormal uterine bleeding in 90% of cases, the rest are diagnosed in premenopausal women, 5% of whom are younger than 40 years old. The standard clinical management strategy for this cancer is hysterectomy (3). Although the 5-year survival rate of EndoCA patients without metastatic disease is 82%, for the rest recurrence is almost always deadly (4). In addition, and unlike most other cancers, the incidence of EndoCA is increasing, which is likely to be related to exposure to estrogens associated with increasing rates of obesity and metabolic syndrome (1).

Analyses of The Cancer Genome Atlas (TCGA) reveal EndoCA has four distinct molecular subgroups: POLE ultramutated, microsatellite instability hypermutated, copy-number-low, and copy-number-high. The first three subgroups are characterized by endometrioid histology with a high rate of tumor-suppressor phosphatase and tensin homolog (PTEN) mutations (Supplementary Fig. S1A). Copy-number-low cancers are mostly with serous histology and have a high rate/frequency of mutation in the TP53 tumor suppressor (Supplementary Fig. S1A). Though both PTEN and TP53 genomic alternations contribute to pathogenesis of EndoCA, only 11.1% of cases have been associated with the cooccurrence of both mutations. Overall, loss of PTEN protein expression due to loss-of-function mutations is the most common molecular signature of EndoCA (5).

PTEN encodes a 403-amino acid protein comprised of an amino (N)-terminal, dual specificity phosphatase domain capable of removing phosphate groups from both proteins and lipids and a carboxyl (C)-terminal C2 domain (6). PTEN was originally thought to act as a tumor suppressor solely by dephosphorylating the lipid second messenger phosphatidylinositol-3,4,5-triphosphate (PIP3), a potent activator of the AKT kinase (7). PIP3 is produced by the phosphorylation of phosphatidylinositol-4,5-bisphosphate (PIP2) by Phosphoinositde 3 Kinase upon activation during receptor tyrosine kinase signaling. The AKT kinase then interacts with PIP3 for its activation, targets many downstream signal transduction pathways preventing apoptosis signaling, and promotes cell growth (8). Thus, loss of the PTEN lipid phosphatase checkpoint function leads to overactivation of the PI3K/AKT pathway and stimulation of cell proliferation and tumorigenesis (9). The majority of the EndoCA-specific mutations are observed in the phosphatase domain of PTEN, inactivating its phosphatase function, but a significant number of mutations are also found outside of this domain (Supplementary Fig. S1B), many of which result in truncated mutants lacking C-terminal PTEN (5, 10). Although plasma membrane–localized PTEN function is central to suppression of the PI3K/AKT pathway, recent studies have indicated that PTEN has PI3K/AKT-independent tumor-suppressor functions (11). Cells without functional PTEN display increased chromosome instability with a high frequency of DNA double-strand breaks (DSB), suggesting that its nuclear function acts as a caretaker of the genome (12, 13).

Although the majority of PTEN studies have focused on its canonical phosphatase activities, the noncanonical functions have been demonstrated to play a critical role in the nucleus through various mechanisms such as chromatin remodeling (14), protein–protein interactions (15), and maintenance of DNA replication fork stability (15, 16). However, there is a significant knowledge gap between the dysregulation of nuclear functions of PTEN and tumorigenesis. Here, we hypothesize that loss of nuclear PTEN from the endometrial epithelium could induce defective DNA damage response (DDR) that might be associated with or contribute to tumor development, progression, or potential response to chemotherapies.

Human and mouse samples

Deidentified, paraffin-embedded, human EndoCAs and adjacent normal tissues were obtained from the Spectrum Health Universal Biorepository or the Van Andel Institute Pathology Core, Grand Rapids, Michigan following institutional guidelines with a Not Human Subject Research determination and approval from the Michigan State University IRB and examined by a pathologist for EndoCA grade. Uterine tissue sections from 8-week-old control PgrCre and Pten conditional knockout mice (17) were kindly provided by Dr. Jeong, Michigan State University.

Cell culture, plasmids, transfections, and treatments

Ishikawa cells were obtained from Sigma (Sigma Aldrich). EM-E6/E7 hTERT1 cells have been previously described (18). Ishikawa cells were grown in high glutamine MEM, and EM-E6/E7 hTERT1 cells were grown in high glutamine DEMEM/F12 media supplemented with 10% FBS and antibiotics, and maintained at 37°C with 5% CO2.

All pEGFP-PTEN plasmids [empty vector (pEGFP-EV), PTEN wild-type (PTEN-WT), PTEN nuclear localization signal (PTEN-NLS), and PTEN nuclear exclusion signal (PTEN-NES); ref. 19] were kindly provided by Dr. Trotman at Cold Spring Harbor Laboratories. TranslT-LT1 (Mirus Bio) was used for all cellular transfections. All treatments were done at 24 hours posttransfection time point. Zeocin (ThermoFisher Scientific) at 100 μg/mL (IC50 dose, determined prior to experiments) was used to induce DNA damage in cells. PARP inhibitors, Olaparib and Talazoparib (Selleck Chemicals), were used at the indicated concentrations. The IC50 for each inhibitor in Ishikawa cells was determined to be 64 μmol/L Olaparib and 2.5 μmol/L Talazoparib in preliminary MTT assay experiments.

Immunohistochemistry, immunofluorescence, and Western Blot analyses

Briefly, formalin-fixed, paraffin-embedded human and mice tissue sections were deparaffinized and rehydrated prior to antigen retrieval with a citrate-based antigen unmasking buffer (Vector Laboratories), followed by 3% hydrogen peroxide-methanol treatment. Tissue sections were blocked in 10% normal goat serum (Vector Laboratories) for 1 hour at room temperature (RT) and incubated with primary antibodies as indicated. For immunohistochemistry (IHC), biotin-conjugated secondary antibodies followed by streptavidin-horseradish peroxidase (Jackson ImmunoResearch) were used. Scoring was performed by histologic assessment of staining intensity of PTEN (0, 1+, 2+, or 3+) and extended to semiquantitative histo (H)-score (0–300) that was calculated using the formula [1 × (% cells 1+) + 2 × (% cells 2+) + 3 × (% cells 3+); ref. 20]. Human endometrial stroma in the tumor sections was used as a positive control for PTEN expression. AlexaFlour-conjugated secondary antibodies (ThermoFisher Scientific) were used for immunofluorescence.

For immunocytochemistry, cells were grown on coverslips and transfected for 24 hours with each plasmid. Cells were fixed in ice-cold methanol for 10 minutes, washed with PBS, and blocked with PBS containing 1% BSA, 10% goat serum, and 0.1% Triton X-100 for permeabilization, for 1 hour at RT. Cells were then incubated with primary antibody for 1 hour at RT, followed by AlexaFluor-conjugated secondary antibody for 45 minutes at RT. Coverslips were mounted on glass slides and imaged.

Ishikawa and EM-E6/E7 hTERT1 cell lysates were prepared with 1X RIPA buffer in the presence of protease inhibitor or using the Subcellular Protein Fractionation Kit for Cultured Cells (ThermoFisher Scientific). Lysates were resolved on NuPAGE 4%–12% Bis-Tris gels (ThermoFisher Scientific), transferred to polyvinylidene difluoride membranes, and probed with indicated primary antibodies.

Primary antibodies for above experiments were as follows: anti-PTEN, Histone H3, and -H2AX (Cell Signaling Technology), -phosphoH2AX (Ser139; Millipore), GAPDH (GenTex) and tubulin (Sigma Aldrich).

Flow cytometry

Ishikawa cells were transfected with the indicated plasmids and treated with Zeocin as described above. At the designated time points, cells were harvested from culture plates by TrypLE Express (ThermoFisher Scientific), passed through a 100 μm cell strainer, and washed with growth media and staining buffer (PBS, 1% FBS, 0.1% Triton X-100), then resuspended in 100 μL staining buffer with 10 μL anti-H2AX pS139 conjugated with Allophycocyanin (APC; γH2AX; Milentyi Biotech) and incubated in the dark for 15 minutes at RT. Following another wash with staining buffer, cells were resuspended in flow cytometry buffer (PBS, 0.5% BSA, 2 mmol/L EDTA) with 300 nmol/L DAPI, incubated overnight at 4°C, and analyzed on a CytoFLEX Flow Cytometer (Beckman Coulter). Events were gated initially by forward and side scatter, then for singlets (forward scatter height × area, then DAPI height × area), and finally for γH2AX-APC expression × DAPI expression using FlowJo (FlowJo). Fluorescence minus one was used as the gating control for γH2AX and DAPI. Cell-cycle analysis was performed using the Cell Cycle platform in FlowJo.

Cell viability assay

Cell viability was determined by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) tetrazolium (Sigma Aldrich) reduction assay (21). Briefly, 5 × 103 cells were plated in a 96-well plate. After PARP inhibitor treatment, cells were incubated for 3 hours at 37°C with MTT solution (final concentration of 0.45 mg/mL). Purple formazan crystals were dissolved in solubilization buffer [40% (v/v) dimethylformamide, 2% (v/v) glacial acetic acid, 16% (w/v) sodium dodecyl sulfate, and pH adjusted to 4.7], and absorbance was measured at 570 nm using a microtiter plate reader.

Statistical analysis

Statistics for the experiments in Fig. 1 and Supplementary Figs. S3 and S5 were calculated using GraphPad Prism 7 and with specific tests indicated in figure legends. For Figs. 3 and 4, statistics were performed using R v 3.3.0 (https://cran.r-project.org) with specific tests indicated in figure legends.

Figure 1.

Loss of nuclear PTEN in human EndoCA. A and B, PTEN immunostaining of normal tissues and primary endometrial tumors (EndoCA), respectively. Representative micrographs are shown. Black arrow heads and brackets indicate PTEN nuclear-positive epithelial cells and regions of cells, respectively, in normal endometrial glands (EG) in A. Brackets indicate PTEN nuclear-negative regions of cells in EndoCA in B. C and D, Quantification of staining intensities by histo-scoring (H-score) that represents expression of PTEN in cytoplasm and nuclei of epithelial cells, respectively. Intensities are expressed as median with 95% confidence interval (CI), and statistical significance is determined by Wilcoxon matched-pair signed rank test. n = 29 normal and EndoCA samples. EG, Images demonstrating the variability of staining in EndoCA tumors. Scale bar, 100 μm. Black arrow indicates a positive EG surrounded by negative EGs in E. Black arrow heads in F indicate nuclear and cytoplasmic positive epithelial cells intermixed with negative cells. Black and blue arrow heads indicate nuclear-positive and -negative epithelial cells, respectively, in G.

Figure 1.

Loss of nuclear PTEN in human EndoCA. A and B, PTEN immunostaining of normal tissues and primary endometrial tumors (EndoCA), respectively. Representative micrographs are shown. Black arrow heads and brackets indicate PTEN nuclear-positive epithelial cells and regions of cells, respectively, in normal endometrial glands (EG) in A. Brackets indicate PTEN nuclear-negative regions of cells in EndoCA in B. C and D, Quantification of staining intensities by histo-scoring (H-score) that represents expression of PTEN in cytoplasm and nuclei of epithelial cells, respectively. Intensities are expressed as median with 95% confidence interval (CI), and statistical significance is determined by Wilcoxon matched-pair signed rank test. n = 29 normal and EndoCA samples. EG, Images demonstrating the variability of staining in EndoCA tumors. Scale bar, 100 μm. Black arrow indicates a positive EG surrounded by negative EGs in E. Black arrow heads in F indicate nuclear and cytoplasmic positive epithelial cells intermixed with negative cells. Black and blue arrow heads indicate nuclear-positive and -negative epithelial cells, respectively, in G.

Close modal

Depletion of nuclear PTEN expression is associated with human EndoCA

Initial studies were performed to determine the relative expression levels of PTEN in primary human endometrial tumors along with the matched adjacent normal endometrial tissues as control. We observed that total expression of PTEN was clearly lower in the epithelia of endometrial tumors compared with the adjacent normal tissues (Supplementary Fig. S2A and S2B). In contrast, stromal cells were PTEN positive for both the tumors and the adjacent normal endometria and hence could serve as the internal control for the expression of PTEN. This observation was previously well documented (22) and is also supported by analysis of PTEN levels as measured by reverse phase arrays of EndoCAs in TCGA for endometrial carcinoma (Supplementary Fig. S2E). Confirmation of activated PI3K signaling with the absence of PTEN was observed by IHC with higher phospho-AKT(Ser473) staining in the tumor tissues compared with the adjacent normal (Supplementary Fig. S2C and S2D). To ensure uniformity and rule out possible artifact staining associated with a specific antibody, IHC was performed on serial sections of human normal endometrial tissues with three different monoclonal PTEN antibodies, and consistent results were observed among the different antibodies (Supplementary Fig. S2F–S2H).

Because PTEN exhibits distinct functions in either cytoplasm or nucleus (23), we next determined its localization in each cellular compartment of endometrial cancer and normal endometrial tissue specimens (Fig. 1) by H-score to determine whether there was any correlation between tumors and nuclear localization. Though the median expression of PTEN in the cytoplasm of the epithelia (Fig. 1C) was not significantly different in the tumors (Fig. 1B) compared with normal tissues (Fig. 1A), it was significantly lower or absent in the nuclei of the epithelia (Fig. 1D). Similar results were seen when tumors were broken down by grade (Supplementary Fig. S3A–S3C). The observation that a decrease in nuclear PTEN is associated with tumor tissues suggests that, in addition to its role in helping regulate PI3K signaling, loss of nuclear PTEN function could contribute to progression of EndoCA. It is important to note that although H-scoring revealed a difference (in nuclear) or lack of difference (in cytoplasmic) in median PTEN expression, we observed some variability in EndoCA samples that may not be obvious by H-scoring. For example, variable nuclear PTEN expression was observed in 24 of 58 tumor samples examined, with some glands positive for PTEN in a field of negative glands (Fig. 1E), or positive epithelial nuclei scattered throughout a microscopic field of negative nuclei (Fig. 1F), and the converse, some nuclei negative for PTEN scattered throughout a microscopic field of positive nuclei (Fig. 1G).

Loss of PTEN correlates with increased endogenous DNA damage in human and murine endometrium

Endogenous DNA damage is a hallmark of EndoCA with ultra- and hypermutation phenotypes (24). We investigated whether there was a correlation between phosphorylation of H2AX at Ser139 (γH2AX), a marker of DSBs (25), with loss of PTEN in EndoCA. DNA damage provokes phosphorylation of a histone variant, H2AX, at its C-terminal Ser139 residue by DNA damage–associated kinase Ataxia Telangiectasia Mutated (26) and thus serves as a marker for DNA strand breaks. When we examined uteri from mice that develop EndoCA with conditional deletion of Pten by PgrCre (17), we observed increased levels of γH2AX in their endometria in comparison with control PgrCre mice (Fig. 2A–D). Similarly, γH2AX levels were elevated in human EndoCA in comparison with adjacent normal epithelia in patient tissues (Fig. 2E–H). Furthermore, the absence of coimmunofluorescence of PTEN and γH2AX in human endometria suggests their mutual exclusiveness (Fig. 2I–R). Finally, Western blot analysis of a nontumorigenic endometrial cell line, EM-E6/E7-hTERT1 (18), and the PTEN-null tumorigenic endometrial Ishikawa cell line (27) revealed that loss of PTEN expression correlated with a higher level of endogenous γH2AX (Fig. 2S). Together, these results suggest a direct correlation between the loss of PTEN with increased vulnerability of the endometrial cell genome.

Figure 2.

Loss of PTEN is associated with defective DDR in murine and human endometria. A and B, Immunostaining of γH2AX in control (PgrCre; n = 4) and conditional knockout of Pten (PgrCre;Ptenfl/fl; n = 4) in endometrium of mouse tissue, respectively. C and D, Magnified images of the boxed areas in A and B, respectively. E and F, Immunostaining of γH2AX in adjacent normal tissue (ANT; n = 4) and EndoCA (n = 4), respectively. G and H, Magnified images of the boxed areas in E and F, respectively. IP, Coimmunofluorescence images of PTEN and γH2AX in ANT (n = 3) and EndoCA (n = 3). Q and R, Magnified images of the boxed areas in O and P, respectively. EG, endometrial glands; LE, luminal epithelia. Arrowheads indicate γH2AX-positive epithelial cells. S, Western blot analysis for the expression of γH2AX in PTEN-null, tumorigenic Ishikawa, and nontumorigenic EM-E6/E7-hTERT1 endometrial cells.

Figure 2.

Loss of PTEN is associated with defective DDR in murine and human endometria. A and B, Immunostaining of γH2AX in control (PgrCre; n = 4) and conditional knockout of Pten (PgrCre;Ptenfl/fl; n = 4) in endometrium of mouse tissue, respectively. C and D, Magnified images of the boxed areas in A and B, respectively. E and F, Immunostaining of γH2AX in adjacent normal tissue (ANT; n = 4) and EndoCA (n = 4), respectively. G and H, Magnified images of the boxed areas in E and F, respectively. IP, Coimmunofluorescence images of PTEN and γH2AX in ANT (n = 3) and EndoCA (n = 3). Q and R, Magnified images of the boxed areas in O and P, respectively. EG, endometrial glands; LE, luminal epithelia. Arrowheads indicate γH2AX-positive epithelial cells. S, Western blot analysis for the expression of γH2AX in PTEN-null, tumorigenic Ishikawa, and nontumorigenic EM-E6/E7-hTERT1 endometrial cells.

Close modal

Nuclear PTEN rescues DNA damage–associated cellular responses in PTEN-null EndoCA cells

To determine whether nuclear PTEN can directly protect against DNA damage in EndoCA, a series of experiments were performed. Nuclear and nonnuclear PTEN isogenic cell models were created by overexpression of plasmids containing GFP fusion constructs with PTEN wild-type (PTEN-WT), PTEN tagged with nuclear localization signal (PTEN-NLS), PTEN tagged with nuclear export signal (PTEN-NES) along with control GFP empty vector (pEGFP-EV; ref. 19) in PTEN-null Ishikawa EndoCA cells. After overexpression of the PTEN constructs, cells were examined by live GFP reporter expression (Fig. 3A, D, G, and J) and by immunofluorescence for PTEN (Fig. 3B, E, H, and K) to ensure that PTEN-WT was located throughout the cells (Fig. 3D–F), PTEN-NLS was mostly detected in the nuclei (Fig. 3G–I), and PTEN-NES was cytoplasmic and reduced in the nuclei (Fig. 3J–L). Cells transfected with control GFP reporter–only vector showed live GFP (Fig. 3C) but did not express PTEN (Fig. 3B). Cells were also examined by flow cytometry to assess transfection efficiency (Fig. 3M–P). In addition, cellular fractionation was performed on transfected cells yielding the cytoplasmic, soluble nuclear, and chromatin-bound nuclear fractions. Western blot for PTEN confirmed that PTEN-NLS was enriched in the nuclear fractions, whereas PTEN-NES was enriched in the cytoplasmic fraction (Fig. 3Q). To determine whether expression of exogenous nuclear PTEN can help protect cells from DNA damage, transfected Ishikawa cells were treated with Zeocin, a bleomycin family glycopeptide that causes DSBs (28), for 16 hours followed by a 24-hour recovery period (Fig. 4A). Levels of γH2AX were examined by coimmunofluorescence with PTEN and quantified by flow cytometry along with cell-cycle analysis. We observed an initial increase in γH2AX-positive cells following Zeocin exposure (+Zeocin) for all four plasmids compared with undamaged (-Zeocin), transfected cells (Supplementary Fig. S4Q) which was also visualized by immunofluorescence (Supplementary Fig. S4A–S4P). At 0-hour recovery, there was no significant difference in γH2AX-positive cells (normalized to undamaged transfected cells) between the four plasmids (Fig. 4B). However, after 24-hour recovery, Zeocin-treated cells transfected with PTEN-WT, -NLS, and -NES had a significantly lower percentage of γH2AX-expressing cells compared with pEGFP-EV–transfected cells (q = 0.001, q = 0.005, and q = 0.007; respectively; Fig. 4C). Cells transfected with PTEN-WT or -NLS also had greater recovery, as seen by a larger decrease in γH2AX levels from 0 to 24 hours, compared with pEGFP-EV–transfected cells (q < 0.001 for both; Fig. 4D). This suggests that PTEN can partially rescue the DDR in PTEN-null Ishikawa cells; moreover, nuclear PTEN is as efficient as wild-type PTEN (Fig. 4D).

Figure 3.

Compartmentalization of GFP-PTEN fusion proteins by live GFP and PTEN antibody. Expression of GFP-PTEN fusion proteins demonstrated localization of (A) control empty vector (pEGFP-EV), (D) PTEN-WT, (G) nuclear PTEN (PTEN-NLS), and (J) cytoplasmic PTEN (PTEN-NES) in Ishikawa cells by live GFP. PTEN localization was confirmed by anti-PTEN antibody (PTEN-Ab) staining (B, E, H, and J) and by cellular fractionation and Western blot for PTEN (Q). Transfection efficiency was assessed by flow cytometry for pEGFP-EV (M), pEGFP-PTEN-WT (N), pEGFP-PTEN-NLS (O), and pEGFP-PTEN-NES (P). Scale bar, 10 μm. WCL, whole cell lysate; CP, cytoplasmic fraction; Sol Nuc, soluble nuclear fraction; Ch Nuc, chromatin-bound nuclear fraction. GAPDH and Histone H3 serve as fractionation controls.

Figure 3.

Compartmentalization of GFP-PTEN fusion proteins by live GFP and PTEN antibody. Expression of GFP-PTEN fusion proteins demonstrated localization of (A) control empty vector (pEGFP-EV), (D) PTEN-WT, (G) nuclear PTEN (PTEN-NLS), and (J) cytoplasmic PTEN (PTEN-NES) in Ishikawa cells by live GFP. PTEN localization was confirmed by anti-PTEN antibody (PTEN-Ab) staining (B, E, H, and J) and by cellular fractionation and Western blot for PTEN (Q). Transfection efficiency was assessed by flow cytometry for pEGFP-EV (M), pEGFP-PTEN-WT (N), pEGFP-PTEN-NLS (O), and pEGFP-PTEN-NES (P). Scale bar, 10 μm. WCL, whole cell lysate; CP, cytoplasmic fraction; Sol Nuc, soluble nuclear fraction; Ch Nuc, chromatin-bound nuclear fraction. GAPDH and Histone H3 serve as fractionation controls.

Close modal
Figure 4.

Nuclear PTEN rescues DDR in PTEN-null Ishikawa cells A, Time line and experimental setup for the DNA damage recovery experiments. BD, Quantification of γH2AX-positive cell population by flow cytometry in cells overexpressing indicated PTEN-constructs (n = 3, ±SEM). EG, Quantification of G2–M population by flow cytometry in the same set of cells analyzed in C to E (n = 3, ±SEM). Statistical analyses were done using linear mixed-effect model with FDR adjusted linear contrast was fit to test-specific hypothesis related to the percentage of γH2AX-positive and G2–M cells after normalization to undamaged control. Effect of four different PTEN-specific constructs on DDR was considered at two time points (0-hour and 24-hour recovery). Homoscedasticity was verified using Levenne test, and normality of residuals was confirmed visually using quantile–quantile plots. All tests were two-sided with significance level set at 0.05. Results are reported as FDR q values.

Figure 4.

Nuclear PTEN rescues DDR in PTEN-null Ishikawa cells A, Time line and experimental setup for the DNA damage recovery experiments. BD, Quantification of γH2AX-positive cell population by flow cytometry in cells overexpressing indicated PTEN-constructs (n = 3, ±SEM). EG, Quantification of G2–M population by flow cytometry in the same set of cells analyzed in C to E (n = 3, ±SEM). Statistical analyses were done using linear mixed-effect model with FDR adjusted linear contrast was fit to test-specific hypothesis related to the percentage of γH2AX-positive and G2–M cells after normalization to undamaged control. Effect of four different PTEN-specific constructs on DDR was considered at two time points (0-hour and 24-hour recovery). Homoscedasticity was verified using Levenne test, and normality of residuals was confirmed visually using quantile–quantile plots. All tests were two-sided with significance level set at 0.05. Results are reported as FDR q values.

Close modal

In response to DSBs, cells activate G2–M checkpoint proteins, thereby preventing cells from entering mitosis to help protect genomic integrity (29). Therefore, we analyzed the cell cycle of Ishikawa cells transfected with the various PTEN constructs by flow cytometry with the experimental setup described previously (Fig. 4A). Similar to what was seen with γH2AX, there was an increase in the percentage of cells in G2–M with Zeocin treatment (+Zeocin) compared with untreated cells (-Zeocin) in all four plasmid groups (Supplementary Fig. S4R). When comparing between the plasmid groups, there were no significant changes in the percentages of cells in G2–M phase at 0 hour with different constructs of PTEN (Fig. 4E). However, after 24-hour recovery, Zeocin-treated cells harboring PTEN-WT or -NLS plasmids demonstrated a significantly lower percentage of cells in G2–M phase compared with cells expressing pEGFP-EV (q = 0.003 for both; Fig. 4F). Cells transfected with PTEN-WT or -NLS also had lower percentage of cells in G2–M phase after 24-hour recovery compared with 0 hour, relative to cells transfected with pEGFP-EV plasmid, demonstrating greater overall recovery (q = 0.04 and q = 0.037, respectively; Fig. 4G). This indicates that both WT and nuclear PTEN can rescue the DNA damage–associated G2–M cell cycle checkpoint in PTEN-null Ishikawa cells.

Nuclear PTEN protects Olaparib-mediated cell killing in PTEN-null Ishikawa cells

DDR defects caused by PTEN-deficient cells can be targeted by DNA repair enzyme PARP inhibitors (30). Upon DNA single-strand break, PARP1 enzyme binds to DNA and initiates the repair process through poly ADP-ribosylation (PARylation) of itself (autoPARylation) and DNA repair effector proteins using β-NAD+ cofactor (31). PARP inhibitors (PARPi) are small molecules that prevent the binding of β-NAD+ to the catalytic domain of PARP1/2 enzymes, thus inhibiting PARylation and leading to cell death due to their inability to effectively repair DNA (32). In this context, we investigated the effect of the PARPis, Olaparib and Talazoparib, on DNA repair function in Ishikawa cells transfected with pEGFP-EV (control), PTEN-WT, PTEN-NLS, and PTEN-NES. Indicated PTEN constructs were overexpressed in PTEN-null Ishikawa cells that were then treated with increasing concentrations of Olaparib for 48 hours prior to measuring cell viability by MTT assay. The change in cell death per doubling of Olaparib for cells transfected with PTEN-NLS was significantly less than those transfected with PTEN-NES (comparisons of slopes on log2 scale; q = 0.036; Fig. 5). At the highest dose, the viability of cells transfected with PTEN-NLS and PTEN-WT was significantly higher compared with pEGFP-EV–transfected cells (q = 0.036 for both). The inhibiting effect of nuclear PTEN on PARPi treatment was not observed in Ishikawa cells treated with Talazoparib (Fig. 5), another much more potent PARPi (32), suggesting that PTEN is not able to affect Talazoparib therapy and that Talazoparib could be a better option for women with endometrial cancer, regardless of PTEN status.

Figure 5.

Nuclear PTEN protects against Olaparib-mediated cell killing in PTEN-null Ishikawa cells. Dose-dependent response of Olaparib or Talazoparib (PARP inhibitors) after transfection with control empty vector (pEGFP-EV), PTEN-WT, nuclear PTEN (PTEN-NLS), and cytoplasmic PTEN (PTEN-NES) in Ishikawa cells determined by MTT assay (n = 3, ±SEM). Linear regression was used to determine if the combination of the indicated inhibitor and plasmids affected cell viability differently relative to EV. Data were normalized to dosage 0 for each plasmid. Two models were fit, one was used to test if the change in cell viability per additional dose of inhibitor was different for any of the plasmids compared to EV (comparison of slopes). The second was used to determine if the plasmids have different cell viability at the maximum dose included in the experiment (64 μmol/L, Olaparib; 2.5 μmol/L, Talazoparib). FDR-adjusted P values (q value) are reported.

Figure 5.

Nuclear PTEN protects against Olaparib-mediated cell killing in PTEN-null Ishikawa cells. Dose-dependent response of Olaparib or Talazoparib (PARP inhibitors) after transfection with control empty vector (pEGFP-EV), PTEN-WT, nuclear PTEN (PTEN-NLS), and cytoplasmic PTEN (PTEN-NES) in Ishikawa cells determined by MTT assay (n = 3, ±SEM). Linear regression was used to determine if the combination of the indicated inhibitor and plasmids affected cell viability differently relative to EV. Data were normalized to dosage 0 for each plasmid. Two models were fit, one was used to test if the change in cell viability per additional dose of inhibitor was different for any of the plasmids compared to EV (comparison of slopes). The second was used to determine if the plasmids have different cell viability at the maximum dose included in the experiment (64 μmol/L, Olaparib; 2.5 μmol/L, Talazoparib). FDR-adjusted P values (q value) are reported.

Close modal

Nuclear PTEN localization is positively correlated to tumor suppression in several cancers including primary cutaneous melanoma (33) and colorectal cancer (34). A missense mutation (K289E) in the noncatalytic C2 domain of PTEN leads to its aberrant nuclear localization (19). Stalling and collapse of DNA replication forks by activation of oncogenes lead to DSBs in both precancerous lesions and cancers (35). PTEN antagonizes the activation of the oncogene, AKT, by dephosphorylating PIP3 (7) in the cytoplasm and also maintains the stability of DNA replication fork by facilitating chromatin loading of RAD51 (16) in the nucleus. Therefore, in the current study, higher levels of γH2AX, an indicator of DNA damage, in murine and human endometrial tissues in the context of PTEN depletion might be caused by dysregulation of DNA replication machinery. EndoCAs often have DNA errors created by endogenous error-prone processes such as DNA POLE proofreading defects or DNA mismatch repair deficiency. Also, nuclear PTEN is biochemically and functionally more stable compared with its cytoplasmic counterpart (19). All of which suggest that nuclear PTEN may protect the genome from endogenous DNA damage in those endometrial epithelial cells that are positive for nuclear PTEN and make those cells more resistant to DNA-damaging therapies. For example, our results suggest that EndoCA patients with nuclear PTEN localization in their tumors would be more resistant to Olaparib therapy, and our findings suggest that loss of nuclear PTEN in EndoCA could be an important diagnostic feature for clinical management of the disease.

PTEN also plays important roles in maintaining genomic stability through cell-cycle regulation (36). It promotes G1 arrest to check chromosomal instability and uncontrolled proliferation by regulating the expression and stability of cell-cycle inhibitor, p27Kip1 (37). Also, upon genotoxic stress, PTEN activates the DNA damage checkpoint by suppressing AKT-dependent CHK1 phosphorylation (38). Moreover, PTEN can form a complex with cyclin B1 and CDK1, and accumulates in the nucleus to prevent premature transition from G2 to M phase of the cell cycle (39). We propose that reconstitution of WT or nuclear PTEN in PTEN-null Ishikawa cells activates DNA damage checkpoint G2–M upon Zeocin treatment (at 0-hour time point) and rescues G2–M recovery after Zeocin withdrawal and successful DNA repair.

PARP inhibitors can target tumor cells with defective homologous recombination (HR) that compromise DSB repair process, suggesting a druggable synthetic lethal interaction between them (40). With PARP inhibition, unrepaired single strand breaks persist, and when encountered by a replication fork, lead to replication fork collapse, creating DSBs (41). More recently, PARPi were also shown to “trap” PARP1 on DNA interfering with its catalytic cycle (32). Therefore, PARP inhibition in cancers with defects in other factors required for DDR, especially repair of DSBs, can result in synthetic lethality and a promising therapeutic strategy. For example, a phase I clinical trial of the PARP inhibitor, Olaparib, on patients with germline BRCA1 and BRCA2 mutations demonstrated a clinical benefit (42). PTEN-deficient tumor cells display a defect in HR due to defective expression and chromatin loading of RAD51 (12, 16, 43) and hence could be targeted by PARP inhibitors (30, 43, 44). In fact, PTEN deficiency sensitizes endometroid endometrial cancer to PARP inhibition alone (45) or in combination with PI3K pathway inhibition (46). In the present study, we confirmed these previous reports, but importantly, we found that Ishikawa EndoCA cells lacking nuclear PTEN are more sensitive to Olaparib, whereas cells with nuclear PTEN show an attenuated affect. This discovery suggests that the type of PTEN mutation and its resulting cellular localization may provide clinical benefit to certain patients but not to others. In particular, our data suggest that immunolabeling of patient tumors with PTEN to determine its cellular location (as seen in Fig. 1) could give insight into the responsiveness of those tumors to treatment with Olaparib. These data show the importance of nuclear-specific PTEN functions and provide a rationale for designing therapeutic strategies based on localization of PTEN in EndoCA.

No potential conflicts of interest were disclosed.

Conception and design: A. Mukherjee, A.L. Patterson, J.W. George, J.I. Risinger, J.M. Teixeira

Development of methodology: A. Mukherjee, A.L. Patterson, J.M. Teixeira

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Mukherjee, A.L. Patterson, T.J. Carpenter, J.I. Risinger

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A. Mukherjee, A.L. Patterson, Z.B. Madaj, J.I. Risinger, J.M. Teixeira

Writing, review, and/or revision of the manuscript: A. Mukherjee, A.L. Patterson, Z.B. Madaj, J.I. Risinger, J.M. Teixeira

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.W. George, T.J. Carpenter

Study supervision: J.M. Teixeira

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Mutter
GL
,
Prat
J
.
Pathology of the female reproductive tract
. 3rd ed.
Edinburgh
:
Churchill Livingstone
2014
.
2.
Siegel
RL
,
Miller
KD
,
Jemal
A
. 
Cancer statistics, 2017
.
CA Cancer J Clin
2017
;
67
:
7
30
.
3.
Morice
P
,
Leary
A
,
Creutzberg
C
,
Abu-Rustum
N
,
Darai
E
. 
Endometrial cancer
.
Lancet
2016
;
387
:
1094
108
.
4.
Howlader
N
,
Noone
AM
,
Krapcho
M
,
Miller
D
,
Bishop
K
,
Kosary
CL
, et al
SEER cancer statistics review, 1975–2014
.
Bethesda, MD
:
National Cancer Institute
. https://seer.cancer.gov/csr/1975_2014/ April 2017.
5.
Cancer Genome Atlas Research
N
,
Kandoth
C
,
Schultz
N
,
Cherniack
AD
,
Akbani
R
,
Liu
Y
, et al
Integrated genomic characterization of endometrial carcinoma
.
Nature
2013
;
497
:
67
73
.
6.
Song
MS
,
Salmena
L
,
Pandolfi
PP
. 
The functions and regulation of the PTEN tumour suppressor
.
Nat Rev Mol Cell Biol
2012
;
13
:
283
96
.
7.
Maehama
T
,
Dixon
JE
. 
The tumor suppressor, PTEN/MMAC1, dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5-trisphosphate
.
J Biol Chem
1998
;
273
:
13375
8
.
8.
Mundi
PS
,
Sachdev
J
,
McCourt
C
,
Kalinsky
K
. 
AKT in cancer: new molecular insights and advances in drug development
.
Br J Clin Pharmacol
2016
;
82
:
943
56
.
9.
Stambolic
V
,
Suzuki
A
,
de la Pompa
JL
,
Brothers
GM
,
Mirtsos
C
,
Sasaki
T
, et al
Negative regulation of PKB/Akt-dependent cell survival by the tumor suppressor PTEN
.
Cell
1998
;
95
:
29
39
.
10.
Eng
C
. 
PTEN: one gene, many syndromes
.
Hum Mutat
2003
;
22
:
183
98
.
11.
Yu
S
,
Yang
F
,
Shen
WH
. 
Genome maintenance in the context of 4D chromatin condensation
.
Cell Mol Life Sci
2016
;
73
:
3137
50
.
12.
Shen
WH
,
Balajee
AS
,
Wang
J
,
Wu
H
,
Eng
C
,
Pandolfi
PP
, et al
Essential role for nuclear PTEN in maintaining chromosomal integrity
.
Cell
2007
;
128
:
157
70
.
13.
Bassi
C
,
Ho
J
,
Srikumar
T
,
Dowling
RJ
,
Gorrini
C
,
Miller
SJ
, et al
Nuclear PTEN controls DNA repair and sensitivity to genotoxic stress
.
Science
2013
;
341
:
395
9
.
14.
Chen
ZH
,
Zhu
M
,
Yang
J
,
Liang
H
,
He
J
,
He
S
, et al
PTEN interacts with histone H1 and controls chromatin condensation
.
Cell Rep
2014
;
8
:
2003
14
.
15.
Song
MS
,
Carracedo
A
,
Salmena
L
,
Song
SJ
,
Egia
A
,
Malumbres
M
, et al
Nuclear PTEN regulates the APC-CDH1 tumor-suppressive complex in a phosphatase-independent manner
.
Cell
2011
;
144
:
187
99
.
16.
He
J
,
Kang
X
,
Yin
Y
,
Chao
KS
,
Shen
WH
. 
PTEN regulates DNA replication progression and stalled fork recovery
.
Nat Commun
2015
;
6
:
7620
.
17.
Kim
TH
,
Franco
HL
,
Jung
SY
,
Qin
J
,
Broaddus
RR
,
Lydon
JP
, et al
The synergistic effect of Mig-6 and Pten ablation on endometrial cancer development and progression
.
Oncogene
2010
;
29
:
3770
80
.
18.
Kyo
S
,
Nakamura
M
,
Kiyono
T
,
Maida
Y
,
Kanaya
T
,
Tanaka
M
, et al
Successful immortalization of endometrial glandular cells with normal structural and functional characteristics
.
Am J Pathol
2003
;
163
:
2259
69
.
19.
Trotman
LC
,
Wang
X
,
Alimonti
A
,
Chen
Z
,
Teruya-Feldstein
J
,
Yang
H
, et al
Ubiquitination regulates PTEN nuclear import and tumor suppression
.
Cell
2007
;
128
:
141
56
.
20.
Hirsch
FR
,
Varella-Garcia
M
,
Bunn
PA
 Jr.
,
Di Maria
MV
,
Veve
R
,
Bremmes
RM
, et al
Epidermal growth factor receptor in non-small-cell lung carcinomas: correlation between gene copy number and protein expression and impact on prognosis
.
J Clin Oncol
2003
;
21
:
3798
807
.
21.
Mosmann
T
. 
Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays
.
J Immunol Methods
1983
;
65
:
55
63
.
22.
Di Cristofano
A
,
Ellenson
LH
. 
Endometrial carcinoma
.
Annu Rev Pathol
2007
;
2
:
57
85
.
23.
Worby
CA
,
Dixon
JE
. 
PTEN
.
Annu Rev Biochem
2014
;
83
:
641
69
.
24.
Chadwick
RB
,
Pyatt
RE
,
Niemann
TH
,
Richards
SK
,
Johnson
CK
,
Stevens
MW
, et al
Hereditary and somatic DNA mismatch repair gene mutations in sporadic endometrial carcinoma
.
J Med Genet
2001
;
38
:
461
6
.
25.
Rogakou
EP
,
Pilch
DR
,
Orr
AH
,
Ivanova
VS
,
Bonner
WM
. 
DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139
.
J Biol Chem
1998
;
273
:
5858
68
.
26.
Burma
S
,
Chen
BP
,
Murphy
M
,
Kurimasa
A
,
Chen
DJ
. 
ATM phosphorylates histone H2AX in response to DNA double-strand breaks
.
J Biol Chem
2001
;
276
:
42462
7
.
27.
Nishida
M
. 
The Ishikawa cells from birth to the present
.
Hum Cell
2002
;
15
:
104
17
.
28.
Povirk
LF
,
Austin
MJ
. 
Genotoxicity of bleomycin
.
Mutat Res
1991
;
257
:
127
43
.
29.
Bartek
J
,
Lukas
J
. 
DNA damage checkpoints: from initiation to recovery or adaptation
.
Curr Opin Cell Biol
2007
;
19
:
238
45
.
30.
Mendes-Pereira
AM
,
Martin
SA
,
Brough
R
,
McCarthy
A
,
Taylor
JR
,
Kim
JS
, et al
Synthetic lethal targeting of PTEN mutant cells with PARP inhibitors
.
EMBO Mol Med
2009
;
1
:
315
22
.
31.
Satoh
MS
,
Lindahl
T
. 
Role of poly(ADP-ribose) formation in DNA repair
.
Nature
1992
;
356
:
356
8
.
32.
Lord
CJ
,
Ashworth
A
. 
PARP inhibitors: synthetic lethality in the clinic
.
Science
2017
;
355
:
1152
8
.
33.
Whiteman
DC
,
Zhou
XP
,
Cummings
MC
,
Pavey
S
,
Hayward
NK
,
Eng
C
. 
Nuclear PTEN expression and clinicopathologic features in a population-based series of primary cutaneous melanoma
.
Int J Cancer
2002
;
99
:
63
7
.
34.
Zhou
XP
,
Loukola
A
,
Salovaara
R
,
Nystrom-Lahti
M
,
Peltomaki
P
,
de la Chapelle
A
, et al
PTEN mutational spectra, expression levels, and subcellular localization in microsatellite stable and unstable colorectal cancers
.
Am J Pathol
2002
;
161
:
439
47
.
35.
Halazonetis
TD
,
Gorgoulis
VG
,
Bartek
J
. 
An oncogene-induced DNA damage model for cancer development
.
Science
2008
;
319
:
1352
55
.
36.
Brandmaier
A
,
Hou
SQ
,
Shen
WH
. 
Cell cycle control by PTEN
.
J Mol Biol
2017
;
429
:
2265
77
.
37.
Li
DM
,
Sun
H
. 
PTEN/MMAC1/TEP1 suppresses the tumorigenicity and induces G1 cell cycle arrest in human glioblastoma cells
.
Proc Natl Acad Sci U S A
1998
;
95
:
15406
11
.
38.
Puc
J
,
Keniry
M
,
Li
HS
,
Pandita
TK
,
Choudhury
AD
,
Memeo
L
, et al
Lack of PTEN sequesters CHK1 and initiates genetic instability
.
Cancer Cell
2005
;
7
:
193
204
.
39.
Kim
SJ
,
Lee
HW
,
Baek
JH
,
Cho
YH
,
Kang
HG
,
Jeong
JS
, et al
Activation of nuclear PTEN by inhibition of Notch signaling induces G2–M cell cycle arrest in gastric cancer
.
Oncogene
2016
;
35
:
251
60
.
40.
Lord
CJ
,
Tutt
AN
,
Ashworth
A
. 
Synthetic lethality and cancer therapy: lessons learned from the development of PARP inhibitors
.
Annu Rev Med
2015
;
66
:
455
70
.
41.
Pommier
Y
,
O'Connor
MJ
,
de Bono
J
. 
Laying a trap to kill cancer cells: PARP inhibitors and their mechanisms of action
.
Sci Transl Med
2016
;
8
:
362ps317
.
42.
Fong
PC
,
Boss
DS
,
Yap
TA
,
Tutt
A
,
Wu
P
,
Mergui-Roelvink
M
, et al
Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers
.
N Engl J Med
2009
;
361
:
123
34
.
43.
Minami
D
,
Takigawa
N
,
Takeda
H
,
Takata
M
,
Ochi
N
,
Ichihara
E
, et al
Synergistic effect of olaparib with combination of cisplatin on PTEN-deficient lung cancer cells
.
Mol Cancer Res
2013
;
11
:
140
8
.
44.
McEllin
B
,
Camacho
CV
,
Mukherjee
B
,
Hahm
B
,
Tomimatsu
N
,
Bachoo
RM
, et al
PTEN loss compromises homologous recombination repair in astrocytes: implications for glioblastoma therapy with temozolomide or poly(ADP-ribose) polymerase inhibitors
.
Cancer Res
2010
;
70
:
5457
64
.
45.
Dedes
KJ
,
Wetterskog
D
,
Mendes-Pereira
AM
,
Natrajan
R
,
Lambros
MB
,
Geyer
FC
, et al
PTEN deficiency in endometrioid endometrial adenocarcinomas predicts sensitivity to PARP inhibitors
.
Sci Transl Med
2010
;
2
:
53ra75
.
46.
Bian
X
,
Gao
J
,
Luo
F
,
Rui
C
,
Zheng
T
,
Wang
D
, et al
PTEN deficiency sensitizes endometrioid endometrial cancer to compound PARP-PI3K inhibition but not PARP inhibition as monotherapy
.
Oncogene
2018
;
37
:
341
51
.

Supplementary data