This article is mentioned in:

Introduction

Tamoxifen (TAM), a selective estrogen receptor modulator (SERM), has been studied for over 20 years as an anti-estrogenic drug used to treat estrogen receptor (ER)-positive breast cancer. By competing for binding at the ER, TAM has been proven to reduce the risk of developing breast cancer by 49% through its antagonistic properties (1). Binding of TAM results in decreased expression of genes that affect cell proliferation leading to diminished cell divisions in breast cancer (2). However, in the endometrium, TAM has been shown to act as a partial estrogen agonist, thus leading to the inappropriate proliferation of endometrial tissues. Consequently, the risk of developing endometrial cancer is increased by 3-fold in TAM-treated women (3). In addition, the agonistic effect of TAM has been shown to exhibit high variability depending upon cell type, promoter context, ambient estradiol concentration and ER subtype (α or β) (4,5). Furthermore, TAM has been shown to have significant in vitro cytoxic and cytostatic effects on ER (+) and ER (−) breast cancer cells (6,7). A study by Reddel et al indicated that the effects of TAM are dose- and ER status-dependent (8). Specifically, their study showed that at low doses TAM effects are ER-mediated, whereas at higher doses TAM effects appeared to be ER-independent.

The antagonistic and agonistic properties inherent to TAM are also present in its numerous metabolites, specifically 4′-hydroxy-tamoxifen (4-OH-TAM). With the addition of a hydroxyl group, 4-OH-TAM has been shown to have a higher potency than TAM both in vitro and in vivo corresponding to a higher affinity for the ER (9). Of interest is the effect of 4-OH-TAM in ER-negative tissues, such as the rat (10) and mouse livers (11), where TAM-DNA adducts and 4-OH-TAM-induced DNA adducts have been demonstrated. Moreover, a study by Shibutani et al demonstrated TAM-DNA adduct formation in ER-positive tissues, such as the human endometrium (12). These results suggest the possibility of an ER independent pathway for TAM and its metabolites to induce cell proliferation.

A study by Castro-Rivera and Safe determined that ER-positive HEC-1A human adenocarcinoma cells treated with TAM exhibited both agonistic and antagonistic tendencies with respect to cell proliferation depending on the concentration used (13). Another study by Perry et al demonstrated the apopotic effects of TAM in breast cancer cells regardless of ER status (14). The ER status of HEC-1A and HEC-1B cells has been debated in the literature. A recent study by Acconcia et al unequivocally demonstrated the presence of ER-α in both cell lines (15). However, their immunofluoresence data showed differences in the ER-α subcellular distribution, where HEC-1A ER-α was observed in both the cytosol and the nucleus, whereas, in HEC-1B cells ER-α was mostly located in the cytosol. The aim of the present study was to investigate the in vitro cytotoxicity of 4-OH-TAM and E2 in human endometrial adenocarcinoma HEC-1A and HEC-1B cancer cell lines. We observed an obvious decrease in HEC-1B and HEC-1A cell growth noted at 10 and 100 μM 4-OH-TAM and 100 μM E2. Furthermore, these micromolar concentrations of 4-OH-TAM induced a non-apoptotic cytotoxic effect.

Materials and methods

Cell lines and tissue culture conditions

HEC-1B and HEC-1A human endometrial adenocarcinoma cell lines were a generous gift from Dr. Cheryl Walker at M.D. Anderson (Smithville, TX, USA). HEC-1B and HEC-1A cells were maintained in MEM (cat. #11095-080) or McCoy’s 5A (cat. #11095-080) purchased from Life Technologies (Grand Island, NY, USA) respectively supplemented with 10% fetal bovine serum (FBS, cat. #SH3039603) or charcoal stripped FBS (cat. #SH3006803; lacks endogenous steroid hormones, CSFBS) (both purchased from Hyclone Laboratories, Logan, UT, USA), 2% L-glutamine (cat. #G-7513) and 1% sodium pyruvate (cat. #S8636) (both obtained from Sigma, St. Louis, MO, USA). Cells were grown in 25 and 75 cm2 culture flasks in a humidified atmosphere of 5% CO2 at 37°C.

Compounds

The compounds 4-OH-TAM (cat. #H-7904) and E2 (cat. #E-2758) were purchased from Sigma and dissolved in 100% ethanol, stored and protected from light in stock solutions of 1 mM at −20°C. The final concentration of ethanol in the culture media was consistently <0.1% (v/v).

Exposure to compounds and determination of cell growth

The media corresponding to HEC-1B and HEC-1A cells were replaced using 5% CSFBS 24 h prior to plating. Both cell lines were seeded (10,000 cells/well) in triplicates into 96-well plates. After 24 h, cells were incubated with 200 μl of various 4-OH-TAM or E2 concentrations for 1–3 days at the conditions described above. Cells grown in the absence of 4-OH-TAM and E2 were used as a control. Media were changed on day 2 to ensure proper nutrient content and effective drug concentration.

Cell viability was assessed using CellTiter 96®AQueous One Solution Cell Proliferation Assay (cat. #G3580; Promega, Madison, WI, USA) every 24 h. Briefly, 20 μl of 3-(4,5-dimethyl-thiazol-2-yl)-5-(3-carboxymethoxphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (inner salt; MTS) and an electron-coupling reagent (phenazine ethosulfate; PES) were added to each well. Plates were incubated for 1–3 h at 37°C in a humidified 5% CO2 atmosphere. Absorbance was read at 490 nm using an ELISA plate reader (BioRad, Hercules, CA, USA). Statistical analysis was conducted using a two-way ANOVA with post-hoc comparisons.

DNA fragmentation assay

DNA extracts were prepared from adherent and detached HEC-1B and HEC-1A cells treated with 10 μM 4-OH-TAM or E2 (in 5% CSFBS media) for 24 h or cells treated with 10 μM 4-OH-TAM or 50 μM E2 (in 5% CSFBS media) for 6 or 12 h at the incubating conditions described above. DNA extracts from HEC-1B and HEC-1A cells grown in the absence of 4-OH-TAM or E2 were used as a control. In brief, DNA was isolated using the Wizard® Genomic DNA purification kit (cat. #A1120; Promega) and DNA laddering analysis was performed using a DNA Laddering Assay kit (cat. #660990; Cayman, Ann Arbor, MI, USA) following the manufacture’s instructions. DNA fragments were then separated on a 2% Agarose E-Gel® (cat. #G5018-02; Life Technologies) and visualized with ethidium bromide under ultraviolet light.

Western blot analysis

Caspases

HEC-1A and HEC-1B cells were seeded in 25 cm2 culture flasks and allowed to grow to 85% confluency. Cells were harvested and protein extracts were prepared from cells (adherent and detached cells) grown in the absence (control) or presence of 1, 10 or 100 μM 4-OH-TAM or E2 (in 5% CSFBS media) for 24 h at the incubating conditions previously described followed by western blot analysis. In brief, whole cell lysates were prepared from log phase cells with 4X sample buffer (40% v/v glycerol, 4% SDS, 0.5% w/v bromophenol blue and 0.16 M Tris pH 7.0) plus 10% β-mercaptoethanol, subjected to gel electrophoresis on precast 12% SDS-polyacrylamide gels (cat. #456-1043; BioRad) and transferred to a polyvinyldene fluoride (PVDF) membrane (cat. #IPVH304F0; Merck Millipore Ltd., Co. Cork, Ireland). The membrane was blocked for 1 h with 5% milk in PBS and probed with 0.25 μg/ml mouse anti-caspase-8 monoclonal antibody (cat. #551242; BD Biosciences, San Diego, CA, USA) or with 2 μg/ml mouse anti-caspase-3 monoclonal antibody (cat. #35-1600Z; Life Technologies) for 1 h at room temperature with constant agitation followed by a series of washes and incubated for 1 h with goat anti-mouse HRP conjugate (1:3,000 dilution; cat. #170-6516; BioRad). A chemiluminescent HRP signal detection system Amersham ECL™ Prime Western Blotting Detection Reagent (cat. #RPN2232; GE Healthcare, Buckinghamshire, UK) was used to detect the signal.

Growth inhibition vs. cell death

In order to differentiate between the cytostatic and cytotoxic effects of 4-OH-TAM and E2, medium corresponding to HEC-1B and HEC-1A cells was replaced using 5% CSFBS 24 h prior to plating. Both cell lines were then seeded (50,000 cells/well) in duplicates into 6-well plates. Cell counts of two random wells were performed during plating to confirm uniformity of cell distribution and to establish the actual initial number of cells plated and 24 h later before drug exposure to confirm the population doubling time (PDT). At this time, cells were incubated with 2 ml of 10 μM 4-OH-TAM or 100 μM E2 for 24 h in 5% CSFBS media at the conditions previously described. Cells grown in the absence of 4-OH-TAM or E2 were used as a control. Cell viability was determined by trypan blue dye exclusion assay.

Results

Determination of cell growth after exposure to the compounds

HEC-1B and HEC-1A cells were exposed to varying concentrations of 4-OH-TAM and E2 to determine the effects of these compounds on cell survival.

When comparing the different 4-OH-TAM or E2 concentrations (10–1000 nM) used to treat the HEC-1B cells, no significant difference in the percent survival was observed (p≤0.260; p≤0.755) (Fig. 1A and B). However, as shown in Fig. 2A, HEC-1B cells exposed to higher concentrations (1–100 μM) of 4-OH-TAM showed a significant difference in the percent cell survival between concentrations (p≤0.0001). Furthermore, at the concentrations of 10 and 100 μM, a definitive decrease in the percent survival was noted when compared to the untreated cells (percent survival indicated by solid line and set to 100%). In contrast, as shown in Fig. 2B, HEC-1B cells treated with the same concentrations of E2 underwent an initial proliferative response observed after 24 h at 1 and 10 μM, followed by an apparent decrease in cell survival that did not appear to be different from the non-treated cells. Treatment with 100 μM E2 resulted in complete cell death. Furthermore, a significant difference in percent survival was noted between E2 concentrations (p≤0.0001).

Figure 1 Percentage of survival of HEC-1B endometrial cancer cells treated with nanomolar concentrations (10–1,000 nM) of 4-hydroxy-tamoxifen (4-OH-TAM) (A) or E2 (B) for 1–3 days. Untreated cells (control) are represented by the solid line. HEC-1B cells treated with nanomolar concentrations of 4OH-TAM or E2 resulted in no significant difference in percent survival between concentrations for either treatment.

Figure 2 Percentage of survival of HEC-1B endometrial cancer cells treated with micromolar concentrations (1–100 μM) of 4-hydroxy-tamoxifen (4-OH-TAM) (A) or E2 (B) for 1–3 days. Untreated cells (control) are represented by the solid line. HEC-1B cells treated with micromolar concentrations of 4-OH-TAM or E2 resulted in a significant difference in percent survival between varying concentrations for either treatment. A definitive decrease in percent survival was observed with 10 and 100 μM concentrations of 4-OH-TAM in comparison to the control (A). Cells exposed to the same concentrations of E2 produced an initial proliferative response after 24 h at 1 and 10 μM followed by an apparent decrease in growth that did not appear to be different from the control, whereas a definitive decrease in percent survival was noted at 100 μM E2 (B).

When HEC-1A cells were treated with the same high concentrations (1–100 μM) as HEC-1B cells of 4-OH-TAM or E2, we observed a significant difference in the percentage of cell survival between concentrations (p≤0.0001, similar to HEC-1B cells. When comparing the percent survival of treated cells with the control, 4-OH-TAM appeared to decrease cell survival at 10 μM and complete cell death was observed at 100 μM (Fig. 3A). Similarly, as shown in Fig. 3B, E2 caused complete cell death at 100 μM and there were significant differences between concentrations.

Figure 3 Percentage of survival of HEC-1A endometrial cancer cells treated with micromolar concentrations (1–100 μM) of 4-hydroxy-tamoxifen (4-OH-TAM) (A) or E2 (B) for 1–3 days. Untreated cells (control) are represented by the solid line. HEC-1A cells treated with micromolar concentrations of 4OH-TAM or E2 resulted in a significant difference in percent survival between varying concentrations for either treatment. A definitive decrease in percent survival was observed with 10 and 100 μM concentrations of 4-OH-TAM in comparison to the control (A). Cells exposed to the same concentrations of E2 produced an obvious decrease in percent survival at 100 μM E2 in comparison to the control while 1 and 10 μM concentrations did not appear to differ from the control (B).

Absence of DNA laddering

To determine whether apoptosis was the underlying cellular mechanism responsible for the decreased percent cell survival observed in both cell lines exposed to 10 μM or 100 μM 4-OH-TAM or 100 μM E2 (when compared to the control); DNA was extracted from adherent and floating HEC-1B and HEC-1A cells after 24 h of exposure with 10 μM 4-OH-TAM or E2 and subjected to electrophoresis on a 2% agarose gel. No DNA laddering was observed in any of the samples (data not shown). Due to concern with the possibility of having extracted the DNA past a time-point where DNA fragmentation could be observed, HEC-1B and HEC-1A cells were incubated with 10 μM 4-OH-TAM or 50 μM E2 and DNA was extracted at 6 and 12 h after exposure. Cells were exposed to a higher E2 concentration (50 μM), since we thought that 10 μM E2 was not high enough to induce apoptosis (refer to Fig. 2B and 3B). Again, no DNA laddering was observed in any of the samples suggesting that apoptosis was not the mechanism involved (Fig. 4).

Figure 4 DNA fragmentation analysis after 6 and 12 h in HEC-1B (A) and HEC-1A cells (B) after treatment with 4-hydroxy-tamoxifen (4-OH-TAM) or E2. DNA extracts were prepared as described in Materials and methods from cells treated with 10 μM of 4-OH-TAM or 50 μM E2. The DNA was visualized on a 2% agarose gel with ethidium bromide under UV light. A typical DNA laddering pattern (180–200 bp) was not observed.

Absence of caspase-3 and -8 activation as determined by immunoanalysis

To determine whether activation of caspase-8 and -3 had occurred, western blot analysis of the protein extracts collected from adherent and floating HEC-1B and HEC-1A cells grown in the absence (control) or presence of 1 or 10 μM 4-OH-TAM or E2 for 10 and 24 h was performed. We did not observed the expected lower molecular weight bands for caspase-8 or -3 with any of the concentrations used or time-points (data not shown). Caspase-3 exists in cells as an inactive 32-kDa proenzyme and during apoptosis pro-caspase-3 is cleaved into 17- and 12-kDa active subunits by upstream proteases such as caspase-8. Similarly, during apoptosis, the proform of caspase-8a 55-kDa protein, is cleaved into smaller active subunits of 40/36 kDa (doublet) and 23 kDa.

Cytotoxic effect of 4-OH-TAM and estradiol

Due to the absence of apoptotic markers, we wanted to differentiate between the cytostatic and cytotoxic effects of 4-OH-TAM and E2. In order to do that cells were incubated in the presence or absence of 10 μM 4-OH-TAM or 100 μM E2 for 24 h in 5% CSFBS.

Trypan blue dye exclusion assay was used to assess the cytotoxic effect of 4-OH-TAM or E2 (Fig. 5). Cell viability of HEC-1B and HEC-1A cells treated with 10 μM 4-OH-TAM was determined to be <25% in comparison to the number of living cells at the time of exposure. When HEC-1B and HEC-1A cell lines were treated with 100 μM E2, 0% cell survival was observed. Untreated cells of both cell lines exhibited proliferation of greater than 40% in comparison to the number of living cells at the time of exposure. These results confirmed that cell death occurred differentiating 4-OH-TAM and E2 cytotoxic effect from a cytostatic effect.

Figure 5 Growth inhibition vs. cell death of HEC-1B and HEC-1A endometrial cancer cells. HEC-1B and HEC-1A cells were seeded into 6-well plates and allowed to attach for 24 h. Prior to exposure, cells were counted to determine the initial number of living cells (determined by trypan blue dye exclusion assay) at the time of exposure (represented by the line). HEC-1B and HEC-1A cells were then grown in the absence (control) or presence of 10 μM 4-OH-TAM or 100 μM E2 for 24 h. The percentage of living cells for each treatment was determined by dividing the number of living cells after the 24-h hormonal treatment exposure by the number of living cells counted prior to the start of the exposure. The number of living cells in either cell line treated with 100 μM E2 was not detectable (represented by ND).

Discussion

The present study examined the effects of 4-OH-TAM and E2 on the proliferation of HEC-1B and HEC-1A human endometrial adenocarcinoma cells. Since endometrial cell proliferation is the most sensitive marker for differentiating agonistic vs. antagonistic activity (16), endometrial cell lines were subjected to hormonal treatments of 4-OH-TAM and E2 at varying concentrations (nM to μM) for 1–3 days.

At concentrations ranging from 10–1000 nM of 4-OH-TAM or E2, HEC-1B cell proliferation did not differ from that of the untreated cells. Due to the apparent lack of effect of 4-OH-TAM on the growth patterns between treated and untreated HEC-1B cells, we decided to increase the concentrations of 4-OH-TAM and E2 to the micromolar range. When subjecting HEC-1B and HEC-1A cells to 1 μM 4-OH-TAM for 1–3 days, the growth patterns did not differ from the untreated cells, while cells treated with 10 and 100 μM 4-OH-TAM underwent a significant decrease in cell prolifieration when compared to the control. These results differ from a study by Castro-Rivera and Safe (13) which determined that HEC-1A human endometrial adenocarcinoma cells treated with 1 μM TAM underwent a significant decrease in cell proliferation. When using 4-OH-TAM, a decreased proliferation of HEC-1A cells was not observed until the 4-OH-TAM concentration was increased to 10 μM, thus suggesting a difference in the effects of TAM and its metabolite 4-OH-TAM. This is an interesting result considering 4-OH-TAM has been shown to possess a higher potency than TAM (9).

While micromolar concentrations of 4-OH-TAM and E2 appear to be clinically irrelevant, the plasma concentration range of TAM measured in chemotherapy patients has been recorded as 0.1–10 μM (17). Furthermore, patients treated with higher doses of TAM (720 mg per day) have serum levels of up to 3.5 μM with accumulated levels in tissues reaching 16–30 times higher than that of the serum (14). Therefore, the concentrations of 4-OH-TAM utilized in this study mirror actual concentrations of TAM recorded in patients undergoing treatment for breast cancer, thus verifying the application of our chosen concentrations.

Following observation of the HEC-1B and HEC-1A cells exposed to 1 or 10 μM E2 for 1–3 days, the growth patterns did not appear to be different from the untreated cells, whereas 100 μM E2 caused an obvious decrease in cell proliferation in both cell lines. Since HEC-1A and HEC-1B cells contain ER-α (15), estrogen treatment was hypothesized to induce an increase in cell growth in comparison to the untreated cells. However, the concentrations of estadiol used in this study were higher than physiological plasma concentrations of estrogen (~1 nM) (18) and thus may have led to the cytotoxic effect.

In order to determine the mechanism behind the observed decrease in cell proliferation, we looked for DNA laddering 24 h after exposure of the cells to 10 μM 4-OH-TAM or E2. DNA gel electrophoresis failed to show DNA-laddering in any of the samples. We were concerned with the possibility of having extracted the DNA at a time-point when it was too late to observe DNA fragmentation. Therefore, we extracted DNA at 6 and 12 h after exposure to 10 μM 4-OH-TAM or 50 μM E2. Another concern was that the E2 concentration (10 μM) may not have been high enough to induce apoptosis. Therefore, we increased the E2 concentration to 50 μM for the 6 and 12 h exposures. Neither of these measures resulted in the evidence of apoptosis through DNA laddering. Similarly, Dietze et al found that 1 μM TAM, but not an equimolar concentration of 4-OH-TAM, induced apoptosis in ER-positive normal human mammary epithelial cells (HMECs) (19). This suggests that 4-OH-TAM induces cell death through a mechanism other than apoptosis.

Furthermore, western blot analysis of caspase-8 and -3 expression failed to unequivocally show caspase activation. The inactive form of caspase-8, a 55-kDa protein, is cleaved into smaller subunits of 40/36 kDa (doublet) and 23 kDa upon activation, whereas procaspase 3 (32 kDa) is cleaved into active 17- and 12-kDa subunits. Our data excluded the death receptor pathway of the apoptotic mechanism (which utilizes caspase-8) as the underlying cellular mechanism of the observed cell death. Therefore, the results of the present study and the lack of DNA laddering suggest that apoptosis was not the underlying mechanism used to cause cell death in the 4-OH-TAM and E2-treated HEC-1B and HEC-1A cells.

Although apoptosis is the most common form of programmed cell death there are alternative non-apoptotic mechanisms that can lead to regulated cell death such as autophagy, necroptosis or the non-regulated pathway of necrosis. In fact, mammalian cells have been shown to express cell death proteases even when they are not undergoing apoptosis (20). Furthermore, characteristics assumed to be unique to apoptosis, such as chromatin condensation and even DNA fragmentation, may not be strictly indicative of apoptosis (21). On a kinetic scale, morphological changes in cell structure may occur over a wide time range depending on cell type as noted in hepatocytes (2–3 h) and lymphocytes (36–48 h) (22). Considering all the data from these studies, a form of cross-talk appears to exist between apoptosis and other cytotoxic mechanisms ending in necrosis in which DNA laddering and caspase-8 activation may not be strictly characterized as apoptotic events. Several in vitro studies have shown that treatment of breast cancer cells with various antiestrogens or aromatase inhibitors induces cell death via unknown mechanisms (23–26). Therefore, although DNA laddering and western blot analysis of caspase-8 and -3 failed to demonstrate the activation of an apoptotic mechanism as a means of cell death, a cytotoxic effect mediated strictly by mechanisms resulting in necrosis may not be accurate. As a result, subsequent studies may focus on determining whether 4-OH-TAM-induced DNA adducts contribute to the observed cytotoxic effect or whether autophagy or neroptosis are at play. Furthermore, future aims may also include using endometrial cancer cells to replicate studies conducted with breast cancer cells. When analyzing TAM- and 4-OH-TAM-treated normal HMECs, apoptosis was found in TAM-treated, but not 4-OH-TAM-treated HMECs (19). It would be of interest to investigate whether similar results are noted with endometrial cancer cells, thus elucidating the differences between TAM and 4-OH-TAM cytotoxic effects in the endometrium.

Due to the obvious decrease in cell survival observed in both HEC-1B and HEC-1A cells exposed to 10 μM 4-OH-TAM or 100 μM E2 for 1–3 days in comparison to the untreated cells and the lack of apoptotic markers, we wanted to distinguish whether the decline observed was due to a cytostatic (growth inhibition) or cytotoxic (cell death) effect. After exposure to the above concentrations of 4-OH-TAM and E2, a cytotoxic effect was observed as determined by trypan blue dye exclusion assay. Indeed, the number of living cells decreased to <25 and 0% for HEC-1B and HEC-1A cells treated with 4-OH-TAM and E2, respectively. Therefore, the micromolar concentrations of 4-OH-TAM and E2 induced a cytotoxic effect resulting in extensive to complete cell death.

Overall, this study suggests that micromolar concentrations of 4-OH-TAM induce a non-apoptotic cytotoxic effect in the endometrium. However, subsequent studies are needed to elucidate the underlying mechanism involved in the cytotoxic effect of 4-OH-TAM and E2.

Acknowledgements

This study was supported by the MERCK-AAAS grant awarded to the Biology and Chemistry Department, Sam Taylor Fellowship and the Fleming Fund of Southwestern University awarded to Dr Cuevas. The authors would like to thank Dr Maria Todd for her editorial feedback, Taylor Vickers for his help with the figures, Dr Cherryl Walker (MD Anderson, Smithville, TX) for providing the HEC-1A and HEC-1B cell lines and Dr Romi Burks for her statistical expertise.

References