- Email: [email protected]
EGFR signaling loop promotes growth of hepatocellular carcinoma cells
Steroids xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
The ER-α36/EGFR signaling loop promotes growth of hepatocellular carcinoma cells Hui You, Kun Meng, Zhao-Yi Wang
⁎
Beijing Shenogen Biomedical Co., Ltd, Beijing, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords: EGFR ER-α36 Rapid estrogen signaling Hepatocellular carcinoma
Hepatocellular carcinoma (HCC) is the common primary liver cancer and the third leading cause of cancer related mortality worldwide. It is generally thought that the estrogen-signaling pathway is not related to the development and progression of human HCC. However, accumulating evidences indicate the existence of a rapid estrogen signaling in HCC cells that is able to promote cell growth. However, the receptor that mediates the rapid estrogen signaling in HCC cells has not been established. Previously, our laboratory identified a variant of ER-α, ER-α36, and found that ER-α36 mediates the rapid estrogen signaling such as the activation of the MAPK/ERK signaling in breast carcinoma cells. Our current experiments studied the role of the rapid estrogen signaling mediated by ER-α36 in growth of HCC HepG2 and PLC/PRF/5 cells that highly express ER-α36 and found these cells were strongly responsive to the rapid estrogen signaling. Knockdown of ER-α36 expression in these HCC cells using the shRNA method attenuated their responsiveness to estrogen and destabilized EGFR protein. ERα36 mediated estrogen-induced phosphorylation of Src and the MAPK/ERK as well as cyclin D1 expression. In addition, there existed an ER-α36/EGFR positive regulatory loop in HCC cells that was important for the maintenance and positive regulation of HCC tumorsphere cells. Our results thus indicated that the rapid estrogen receptor is mediated by ER-α36 in HCC cells through the EGFR/Src/ERK signaling pathway and suggested that the ER-α36/EGFR signaling loop is a potential target to develop novel therapeutic approaches for HCC treatment.
1. Introduction Although the widespread implementation of different screening programs, improvement of diagnostic tools and the progress in therapeutic methods, the survival rate of hepatocellular carcinoma (HCC) remains poor and the morbidity still has been steadily increasing. A number of risk factors have been related to the pathogenesis of HCC, including HBV and HCV infections, aflatoxins, alcohol, diabetes and nonalcoholic fatty diseases [1]. It has been known for a long time that estrogen signaling is related to the pathogenesis of HCC although liver is not a classical estrogen target organ [1]. However, the molecular mechanism by which the estrogen signaling promotes hepatocellular carcinogenesis is largely unclear. Estrogen regulates cell growth through estrogen receptors (ERs), ER-α and β, both of which are members of the nuclear receptor superfamily and function as transcription factors for target genes [2]. The classical estrogen-signaling pathway includes estrogen binding to the receptors, then the receptors dimerization and binding to the estrogen response elements (ERE) of the target gene promoters,
⁎
regulating gene transcription [2]. Besides this classical genomic action, there is another signaling pathway called “membrane-initiated estrogen signaling” or “rapid estrogen signaling” that utilizes growth factors, their receptors, cytoplasmic proteins and other membrane signaling pathways [3–5]. Several studies demonstrated that a rapid estrogen signaling such as activation of IP3-PKC-α, the AKT and the MAPK/ERK contributes to the induction of cyclin D1 expression and DNA synthesis in HCC cells [6–9], indicating the existence of the rapid estrogens signaling in HCC. In 2015, our laboratory isolated and cloned a 36-kDa variant of ERα named ER-α36 to differentiate it from the full-length 66-kDa ER-α, ER-α66. ER-α36 was predominantly detected at the plasma membrane as well as the cytoplasm. The MAPK/ERK and the PI3K/AKT signaling pathways are typical pathways involved in the ER-α36-mediated rapid estrogenic signaling [10,11]. ER-α36 is generated from a promoter located in the first intron of the ER-α66 gene [12], indicating that ER-α36 expression is regulated differently from ER-α66, which is consistent with the findings that ER-α36 is expressed in ER-negative breast cancer [13–15]. Previously, ER-α36 expression was detected in specimens
Corresponding author at: Beijing Shenogen Biomedical Co., Ltd, #29 Life Science Park Road, Beijing, PR China. E-mail address: [email protected] (Z.-Y. Wang).
https://doi.org/10.1016/j.steroids.2018.02.007 Received 14 September 2017; Received in revised form 4 January 2018; Accepted 20 February 2018 0039-128X/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: You, H., Steroids (2018), https://doi.org/10.1016/j.steroids.2018.02.007
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 1. Estrogen and EGF stimulate growth of HCC cells. (A). The expression of ER-α variants and EGFR in HCC PLC/PRF/5 and HepG2. (B). The effect of E2 or EGF on the growth rate of HCC cells. Cells maintained for three days in the phenol red-free DMEM plus 2.5% dextran-charcoal-stripped fetal calf serum were treated with or without 0.1 nM E2 or 10 ng/ml EGF for different time periods. The cell numbers were examined using an automatic cell counter. Five dishes were used for each time point and experiments were repeated more than three times. The mean cell numbers are shown.
from HCC patients but not in the non-tumoral tissues while ER-α66 was mainly detected in the non-tumoral tissues and was decreased or became undetectable in HCC specimens [16,17], suggesting that ER-α36 may contribute to the development and progression of HCC. In this study, we investigated ER-α36-mediated rapid estrogen signaling in growth of HCC cells and in the tumorsphere cells derived from these HCC cells. We found the stimulatory effects of a positive feedback regulatory loop between ER-α36 and EGFR on the growth of HCC cells.
Fig. 2. Estrogen elicits rapid signaling and induces gene expression in HCC cells. (A). Western blot analysis of the effects of indicated concentrations of E2 on the phosphorylation levels of the MAPK/ERK1/2 and the AKT. The columns represent the means of three experiments ± S.D. *, P < 0.05 for cells treated with DMSO (0) vs cells treated with indicated concentrations of E2 for 30 min. (B). E2 induces expression of ER-α36, EGFR and cyclin D1 in HCC cells. Cells were treated with indicated concentrations of E2 or DMSO (0) as a control for 12 h. The membranes were stripped and re-probed with different antibodies.
2. Methods and materials
MAPK, Src and EGRR inhibitors U0126, PP2 and AG1478 were acquired from Calbiochem (San Diego, CA). The ER-α36 antibody was prepared as described before [11]. CyclinD1 antibody was purchased from Santa Cruz (Santa Cruz, CA). Antibodies for β-actin, ER-α66, ERK1/2, pERK1/2, Src, p-Src, AKT, p-AKT and EGFR, p-EGFR, anti-mouse IgGHRP, anti-rabbit IgG-HRP were all purchased from Cell Signaling Technology (Danvers, MA).
2.1. Reagents The 17β-estradiol (E2) and insulin was purchased from Sigma Chemical (St Louis, MO). B-27 was obtained from Gibco (Carlsbad, CA). Basic FGF and EGF were from Peprotech (East Brunswick, NJ). The
2
Steroids xxx (xxxx) xxx–xxx
H. You et al.
(FBS), 1% non-antibiotic-antimycotic from Gibco. HCC cells were cultured at 37 °C and 5% CO2. To establish ER-α36 knocked-down stable cell lines from PLC/PRF/ 5 and HepG2 cells, we cloned the DNA oligonucleotides 5-GATGCCA ATAGGTACTGAATTGATATCCGTTCAGT ACCTATTGGCAT-3′ targeting a sequence in the 3′UTR of ER-α36 gene into the pRNAT-U6.1/Neo expression vector (GenScript Corp, Piscataway, NJ) to construct an ERα36 specific shRNA expression vector. Briefly, the empty expression vector and ER-α36 shRNA expression vector were transfected into the HCC cells and the transfected cells were selected with 600 μg/ml G418 (Gibco) for three weeks. After then, cells were pooled from at least 25 clones. For cell growth assay, cells at the logarithmic phase were collected and 10,000 cells were seeded onto 6-well plates and cultured with phenol red-free DMEM/F12 (Gibco), respectively, containing 1% charcoal-stripped fetal calf serum (Gibco), 1% non-antibiotic-antimycotic. Media were changed every two days and treated with 0.1 nM of E2 or 10 ng/ml of EGF, respectively. Every two days, the cell numbers were examined using the Countess®II automatic cell counter (Thermo Fisher Scientific, Waltham, MA).
2.3. Tumorsphere formation and estrogen stimulation HCC cells at 10,000 cells/ml were seeded onto the ultra-low attachment 6-well plate (Corning Incorporated, Pasadena, CA) and cultured for five days in the tumorsphere medium: phenol-red free DMEM/ F12 medium (Gibco) containing 10 ng/ml basic FGF, 4 μg/ml insulin (Sigma), 1×B-27 (Gibco), After five days, the tumorspheres were photographed with the Nikon Eclipse TE2000-S microscope. Tumorspheres were collected, washed, dissociated with Trypsin, and then re-suspended. The cell number was determined using the Countess®II automatic cell counter (Thermo Fisher Scientific). For estrogen stimulation assays, 0.1 nM E2 or vehicle (DMSO) was used to treat tumorspheres. All the experiments were repeated at least three times.
2.4. Western blots analysis HCC cells were harvested and washed with cold PBS. The RIPA buffer was used to lyse cells and the cell lysates was boiled in loading buffer. After electrophoresis, the proteins were transferred onto PVDF membranes (EMD Millipore, Danvers, MA), then blocked in the buffer containing 5% BSA, and the primary antibodies β-actin (1:3000), ERα36 (1:2000), ER-α66, ERK1/2, p-ERK1/2, Src, p-Src, AKT, p-AKT and EGFR, p-EGFR (all 1:1000), incubated overnight at 4 °C. The filters were incubated with appropriate anti-mouse IgG or anti-rabbit IgG at room temperature for two hours. Protein bands were visualized with ECL kit (EMD Millipore, Danvers, MA) and photographed with Chemiluminescence Imaging System (LI-COR, Lincoln, NE). All the experiments were repeated at least three times and analyzed with IMAGE J.
Fig. 3. ER-α36 mediates rapid estrogen signaling in HCC cells. (A). Western blot of ERα36, ER-α66 and EGFR expression in different HCC cell variants, control cells (PLC/RPF/ 5/C and HepG2/C); HCC cells transfected with the empty expression vector and ER-α36 expression knocked-down cells (PLC/PRF/5/Sh36 and HepG2/Sh36). (B). Western blot analysis of the effects of indicated concentrations of E2 on the phosphorylation level of the MAPK/ERK1/2 (30 min) and cyclin D1 expression (12 h) in different HCC cell variants. β-actin was used for a loading control for cyclin D1 expression (C). The effects of E2 on the growth of different HCC cell variants. Cells were treated with 0.1 nM E2 or DMSO (vehicle) as a control. Five dishes were used for each concentration and experiments were repeated more than five times. The mean cell numbers ± S.D. are shown. *, P < 0.05 for cells treated with DMSO (vehicle) vs cells treated 0.1 nM E2.
2.5. Statistical analysis Results were expressed as the mean ± standard deviation (S.D.). Statistical analysis was performed using GraphPad InStat software program with paired-samples t-test, or ANOVA followed by the Student’s t-test and the significance was accepted for P < 0.05 levels.
2.2. Cell culture, plasmids and growth assays PLC/PRF/5 and HepG2 cell lines were originally purchased from Japanese Cancer Research Bank (Tokyo). HCC cells were cultured in the complete DMEM medium (Gibco) containing 5% fetal bovine serum
3
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 4. The EGFR/Src/ERK axis is involved in estrogen induction of cyclin D1 and cell growth in HCC cells. (A). Western blot analysis of phosphorylation levels of the ERK1/2, the Src-Tyr416 and the EGFR-Tyr-845 in HCC cells. The HCC cells were treated with 0.1 nM E2 for different time periods. Total proteins were used as loading controls for the phosphorylated ERK1/2 and the Src-Tyr416. β-actin was used as a loading control for EGFR expression (B). Western blot analysis of cyclin D1 expression in HCC cells treated with 0.1 nM of E2 with or without the Src inhibitor PP2, the MAPK inhibitor UO126 and the EGFR inhibitor AG1478 for 12 h. All experiments were repeated at least three times, and the representative results are shown. (C). Cell growth assays of the HCC cells treated with 0.1 nM E2 together with or without the indicated inhibitors as well as the indicated inhibitors alone. The mean cell numbers ± SD are shown. *, P < 0.05 for cells treated with DMSO (vehicle) vs cells treated 0.1 nM E2.
3. Results
in HCC cells (Fig. 1A). As a control, we also examined ER-α66 expression and found that both cell lines express trace amounts of ER-α66 (Fig. 1A). We then decided to determine if estrogen stimulates growth of these HCC cells. As shown in Fig. 1B, the cells treated with 17βestradiol (E2) exhibited a remarkably increased growth rate compared to the cells treated with the vehicle (Fig. 1B). We further found that EGF also potently promoted growth of both HCC cell lines (Fig. 1B). Our data thus indicated that both estrogen and EGF signaling pathways are involved in growth of these HCC cells.
3.1. Estrogen and EGF stimulate growth of HCC cells It has been reported that samples from primary HCC exhibited a decreased ER-α66 expression and an increased ER-α36 expression [16,17]. To determine if established HCC cells retain ER-α36 expression, we used two well-known hepatoma cell lines HepG2 and PLC/ PRF/5. The result showed that ER-α36 and EGFR were highly expressed
4
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 5. Knockdown of ER-α36 expression attenuates EGF signaling of HCC cells. (A). Western blot analysis of the phosphorylation levels of the MAPK/ERK and the AKT in the HCC cell variants treated with 10 ng/ml EGF for 30 min. Columns: means of three independent experiments; bars, S.D. *, p < 0.05, for cells treated with PBS vs EGF. (B). Western blot analysis of cyclin D1 expression in the HCC cell variants treated with 10 ng/ml EGF for 12 h. (C). Cell growth assays of the HCC cell variants treated with EGF (10 ng/ml). Five dishes were used and experiments were repeated three times. The mean cell numbers are shown; bars, S.D. *, p < 0.05, for cells treated with PBS vs EGF.
level of ER-α36 using the shRNA method. Empty expression vector were also transfected into PLC/PRF/5 and HepG2 cells to serve as controls. Western blot analysis indicated that ER-α36 expression was lessened about 80% compared with the control cells (Fig. 3A), while ER-α66 expression was intact in these cells (Fig. 3A). Consequently, EGFR expression was dramatically down-regulated in the HCC cells with knocked-down level of ER-α36 expression (Fig. 3A), E2 treatment failed to induce ERK1/2 phosphorylation, cyclin D1 expression and cell growth in the HCC cell lines with knocked-down level of ER-α36 expression (Fig. 3B&C). These results indicated that ER-α36 mediates the rapid and mitogenic estrogen signaling in these HCC cells.
3.2. Rapid estrogen signaling and induction of ER-α36/EGFR expression in HCC cells The finding that HCC cells express ER-α36 suggested that these cells might harbor the rapid estrogen signaling. To determine whether E2 induced the MAPK/ERK1/2 phosphorylation in HCC cells, different concentrations of E2 were used to treat HCC cells for 30 min to analyze phosphorylation of the ERK1/2 using Western blot analysis. Fig. 2A reveals that E2 induced ERK phosphorylation in both cell lines (Fig. 2A). In addition, E2 also induced the phosphorylation of the AKT in these HCC cells (Fig. 2A). Thus, our results indicated that these HCC cells exists the rapid estrogen signaling. Consistent with the previous report [17], E2 could induce the expression levels of both ER-α36 and EGFR in these HCC cells (Fig. 2B), suggesting the existence of the ERα36/EGFR positive regulatory loop previously reported in breast cancer cells. Consequently, E2 could also induce cyclin D1 expression in these HCC cells (Fig. 2B).
3.4. The EGFR/Src/ERK axis is involved in estrogen induction of cyclin D1 and cell growth in HCC cells Recently, we reported that EGFR, Src and the MAPK/ERK all contributed to estrogen-stimulated growth of breast carcinoma cells [17]. We decided to determine if HCC cells also harbor the same signaling axis. E2 at 0.1 nM was added into the HCC cell lines for different time periods. Phosphorylation levels of the ERK1/2 and Src were assessed using Western blot analysis. Fig. 4 indicates that E2 elicited phosphorylation of the MAPK/ERK in 10 min. (Fig. 4A). In addition, E2 also
3.3. ER-α36 mediates the rapid estrogen signaling in HCC cells To determine if ER-α36 mediates the rapid estrogen signaling of HCC cells, we established HCC cell sub-lines to express knocked-down
5
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 6. HCC cells exist an ER-α36/EGFR positive regulatory loop. (A). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells with ER-α36 expression knocked-down transfected with an ER-α36 expression vector containing the ER-α36 cDNA without the 3′UTR. (B). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells with ER-α36 expression knocked-down transfected with an EGFR expression vector. (C). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells treated with EGF (10 ng/ml) for 12 h.
Fig. 5. (continued)
induced phosphorylation of Src-Tyr-416 in these HCC cells (Fig. 4A). We also found that E2 induced EGFR-Tyr-845 phosphorylation as well as the expression level of EGFR protein (Fig. 4A). Thus, our results indicated that the EGFR/Src/ERK axis is involved in the rapid estrogen signaling of these HCC cells. To further confirm these findings, we utilized the chemical inhibitors for EGFR (AG1478), Src (PP2) and the MAPK/ERK (U0126). We found that all inhibitors potently diminished E2-induced cyclin D1 expression (Fig. 4B) and cell growth (Fig. 4C) in these HCC cells. Our results thus demonstrated that the EGFR/Src/ERK signaling axis is involved in the ER-α36-mediated estrogen induction of cyclin D1 expression.
cells (Fig. 5B). Consequently, the growth of the HCC cells with ER-α36 knock-down failed to response to EGF stimulation (Fig. 5C). Thus, the EGF signaling is attenuated in the HCC cells with ER-α36 expression knocked-down presumably due to down-regulated EGFR expression. 3.6. HCC cells exist an ER-α36/EGFR positive regulatory loop To further confirm there exists a positive regulatory loop between ER-α36 and EGFR in HCC cells, we introduced the ER-α36 and EGFR expression vectors into the HCC cells with knocked-down levels of ERα36 expression. We found that re-introduction of recombinant ER-α36 that lacks the 3′UTR into these cells increased the expression level of EGFR protein (Fig. 6A). In addition, we also observed that recombinant EGFR expression in the HCC cells with knocked-down levels of ER-α36 also increased endogenous ER-α36 expression (Fig. 6B), although the induction was modest probably due to the existence of the ER-α36 shRNA. Previously, we reported that ER-α36 expression is subjected to the positive regulation of EGFR signaling [18]. We then treated HCC cells with 10 ng/ml of EGF and then examined ER-α36 and EGFR expression by Western blot. We found that EGF treatment was able to induce both ER-α36 and EGFR expression in HCC cells (Fig. 6C). Thus,
3.5. Knockdown of ER-α36 expression attenuates EGF signaling of HCC cells Previously, we reported that there is a positive regulatory loop between EGFR and ER-α36 in breast cancer cells; EGFR signaling activates the ER-α36 promoter activity and ER-α36 stabilizes EGFR protein [18]. Here, our Western blot results revealed that EGFR expression was remarkably reduced in HCC cells with ER-α36 expression knocked-down. We decided to determine whether the EGF signaling is impacted in these HCC cells. The results indicated that the MAPK/ERK and AKT phosphorylation induced by EGF was significantly attenuated in the HCC cells with ER-α36 expression knocked-down (Fig. 5A). In addition, EGF induced cyclin D1 expression was also diminished in these HCC
6
Steroids xxx (xxxx) xxx–xxx
H. You et al.
whereas ER-α66 expression remained unchanged (Fig. 8B). Finally, we observed that 0.1 nM E2 treatment potently increased tumorsphere cell number and the size from the control HCC cells but not from the HCC cells with ER-α36 expression knocked-down (Fig. 8C&D). We also observed that in the absence of 0.1 nM E2, the HCC cells with knockeddown ER-α36 expression also formed less number and smaller size tumorspheres in comparison with the control cells (Fig. 8C&D). Our results demonstrated that the ER-α3/EGFR positive regulatory loop is important for maintenance and regulation of cancer stem-like cells in HCC. 4. Discussion In this report, HCC PLC/PRF/5 and HepG2 cells were used to study the effect of the rapid estrogen signaling mediated by ER-α36 on the growth of HCC cells. Here, we found that HCC cells expressing trace amount of ER-α66 but high level of ER-α36 exhibited a potent mitogenic estrogen signaling. Previously, HCC cells such as HepG2 have been shown to rapidly respond to estrogen through activation of the PKC-α, the MAPK/ERK and the AKT as well as induction of cyclin D1 expression and DNA synthesis [6–9]. However, the mediator of these rapid estrogen activities in HCC has not been established. In this paper, we used an ER-α36specific shRNA to knock-down ER-α36 expression in these HCC cells and found that the PLC/PRF/5 and HpeG2 cells with knocked-down level of ER-α36 expression exhibited no responsiveness to the mitogenic estrogen signaling, indicating that ER-α36 is the receptor that mediates the rapid estrogen signaling in HCC cells. Previously, we reported that EGFR and ER-α36 positively regulate each other’s expression in breast cancer cells; the promoter activity of ER-α36 is activated by EGFR signaling and ER-α36 stabilizes the steady state levels of EGFR protein [18]. In this paper, we also found the existence of the ER-α36/EGFR positive regulatory loop in HCC cells that plays an important role in cell growth. We showed that estrogen induced the MAPK/ERK phosphorylation and cyclin D1 expression through a pathway that includes ER-α36 and the EGFR/Src axis. Previously, Our laboratory demonstrated that ER-α36 interacted strongly with the EGFR/Src/Shc complex and that estrogen primarily induced the EGFR-Tyr-845 phosphorylation [17,18]. Our laboratory also demonstrated that estrogen activated phosphorylation of Src-Tyr-416 and the AG1478 as well as PP2 blocked estrogen-induced cyclin D1 expression. In the current study, we identified that the EGFR/Src complex plays an important role in the rapid estrogen signaling of the HCC cells that express ER-α36. It is worth to note that E2 treatment also augmented the expression level of EGFR protein in HCC cells with 30 min, suggesting the enhanced transcription may not be involved, which is well in accordance with our previous report that ER-α36 functions at post-transcription level to stabilize EGFR protein [19]. In the past decade, a small number of tumor cells with stem cell properties have been described in different kind of cancers [20]. These stem-like cancer cells are responsible for the initiation, recurrence and metastasis of tumor and also for the failure of the current therapies [20,21]. Recently, it has been reported that HCC tumorsphere cells are rich in cancer stem-like cells [22]. Based on this, we cultured tumorspheres from both HCC cell lines and used them to study their responses to estrogen. Here, our results showed that tumorsphere cells derived from the established HCC cells exhibited enhanced expression and activities of the ER-αβ6/EGFR positive regulatory loop and the HCC cells with ER-α36 expression knocked-down generated smaller size and less number tumorspheres compared with the control cells, suggesting that this loop plays a critical role in these cancer stem-like cells from HCC. We also documented that E2 treatment could increase
Fig. 7. Tumorsphere cells exhibit enhanced growth-promoting signals. (A). Western blot analysis of the ER-α36, ER-α66 and EGFR expression in the tumorsphere cells derived from the HCC cells. The membranes were stripped and re-probed with different antibodies. (B). Western blot analysis of the phosphorylation levels of the MAPK/ERK and the AKT in the tumorsphere cells derived from the HCC cells.
our results strongly indicated that HCC cells also contain the ER-α36/ EGFR positive regulatory loop previously reported in breast cancer cells. 3.7. The ER-α36/EGFR positive regulatory loop plays an important role in tumorspheres derived from HCC cells Accumulating experimental and clinical studies indicated that cancer stem-like cells contribute the development and progression of human cancer [20]. Our laboratory found that OCT-4 and Nanog were highly expressed in the tumorsphere cells derived from HCC cells (data not shown). To examine the effect of the ER-α36/EGFR positive regulatory loop on growth of cancer stem-like cells derived from HCC cells, we cultured HCC cells in tumorsphere medium that lacks EGF to form tumorspheres. We found that both ER-α36 and EGFR expression as well as the downstream signals such as the MAPK/ERK and AKT phosphorylation were significantly enhanced in HCC tumorsphere cells while ER-α66 expression was without significant changes (Fig. 7A&B). We further found that E2 treatment also augmented the MAPK/ERK and the AKT phosphorylation in these tumorsphere cells (Fig. 8A), suggesting that tumorsphere cells derived from HCC cells retain the rapid estrogen signaling. Consequently, E2 treatment increased the expression levels of ER-α36, EGFR and cyclin D1 in these tumorsphere cells
7
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 8. The rapid estrogen signaling expands the tumorsphere cells derived from the HCC cells. (A). Western blot analysis of phosphorylation levels of the MAPK/ERK1/2 and the AKT in the tumorsphere cells derived from the HCC cells treated E2. The tumorsphere cells were treated with indicated concentrations of E2 for 30 min. The columns represent the means of three experiments; bars, S.D. *, P < 0.05 for cells treated with vehicle (0) vs cells treated different concentrations of E2. (B). Western blot analysis of ER-α36, EGFR and cyclin D1 expression in the tumorsphere cells derived from the HCC cells treated with E2 for 12 h. The tumorsphere cells were treated with different concentrations of E2 for 12 h. (C). ER-α36-mediated estrogen signaling positively regulates the size and number of tumorspheres from HCC cells. Representative tumorspheres from HCC cell variants treated with vehicle or 0.1 nM E2 for seven days. Magnification: X200, and X400 for the inserted pictures. (D). The numbers of cells from dissociated tumorspheres of different HCC cell variants were determined. The columns represent the means of three experiments; bars, S.D. *, P < 0.05 for cells treated with vehicle vs cells treated with 0.1 nM E2. #, P < 0.05 for the control HCC cells vs the HCC cells with ER-α36 expression knocked-down.
Thus, ER-α36 is a novel player in the rapid estrogen signaling during HCC tumorigenesis. The finding that the ER-α36/EGFR regulatory loop plays an important role in the maintenance and positive regulation of cancer stem-like cells from HCC also highlights the importance of targeting this positive regulatory loop as a novel approach to treat human HCC.
tumorsphere cell numbers and sizes, suggesting estrogen was able increase the pool of cancer stem-like cells in HCC cells via the ER-α36mediated signaling. In summary, we have shown that ER-α36 expressing HCC cells exhibited quick responses to estrogen, indicating that the rapid estrogen signaling contributes to development and progression of HCC.
8
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 8. (continued)
Conflict of interest
References
The authors declare no conflict of interest.
[1] E. Villa, Role of estrogen in liver cancer, Women’s Health 4 (1) (2008) 41–50. [2] E.H. Kong, A.C. Pike, R.E. Hubbard, Structure and mechanism of the oestrogen receptor, Biochem. Soc. Trans. 31 (2003) 56–59. [3] J.H. Segars, P.H. Driggers, Estrogen action and cytoplasmic signaling cascades. Part I: membrane-associated signaling complexes, Trends Endocrinol. Metabol. 13 (2002) 349–354. [4] P.H. Driggers, J.H. Segars, Estrogen action and cytoplasmic signaling pathways. Part II: the role of growth factors and phosphorylation in estrogen signaling, Trends Endocrinol. Metab. 13 (2002) 422–427. [5] M.J. Kelly, E.R. Levin, Rapid actions of plasma membrane estrogen receptors, Trends Endocrinol. Metab. 12 (2001) (2001) 152–156.
Acknowledgement This work was supported by “Significant New Drug Development” National Science and Technology major project (2013ZX09401004).
9
Steroids xxx (xxxx) xxx–xxx
H. You et al.
[6] M. Marino, V. Pallottini, A. Trentalance, Estrogen cause rapid activation of IP3-PKC-α signal transduction pathway in HepG2 cells, Biochem. Biophys. Res. Comm. 245 (1998) 251–258. [7] M. Marino, E. Distefano, S. Caporali, G. Ceracchi, V. Pallottini, A. Trentalance, β-estradiol stimulation of DNA synthesis requires different PKC isoforms in HepG2 and MCF7 cells, J. Cell. Phys. 188 (2001) 170–177. [8] M. Marino, F. Acconcia, F. Bresciani, A. Weisz, A. Trentalance, Distinct nongenomic signal transduction pathways controlled by 17b-Estradiol regulate DNA synthesis and cyclin D1 gene transcription in HepG2 cells, Mol. Biol. Cell 13 (2002) 3720–3729. [9] M. Marino, F. Acconcia, A. Trentalance, Biphasic estradiol induced AKT phosphorylation is modulated by PTEN via MAP kinase in HepG2 cells, Mol. Biol. Cell 14 (2003) 2583–2591. [10] Z. Wang, X. Zhang, P. Shen, B.W. Loggie, Y. Chang, T.F. Deuel, Identification, cloning, and expression of human estrogen receptor-alpha36, a novel variant of human estrogen receptor-alpha66, Biochem. Biophys. Res. Commun. 336 (4) (2005) 1023–1027. [11] Z. Wang, X. Zhang, P. Shen, B.W. Loggie, Y. Chang, T.F. Deuel, A variant of estrogen receptor-{alpha}, hER-{alpha}36: transduction of estrogen- and antiestrogen-dependent membrane-initiated mitogenic signaling, Proc. Natl. Acad. Sci. U.S.A. 103 (24) (2006) 9063–9068. [12] Y. Zou, L. Ding, M. Coleman, Z. Wang, Estrogen receptor-alpha (ER-alpha) suppresses expression of its variant ER-alpha 36, FEBS Lett. 583 (8) (2009) 1368–1374. [13] L.M. Lee, J. Cao, H. Deng, P. Chen, Z. Gatalica, Z.Y. Wang, ER-alpha36, a novel variant of ER-alpha, is expressed in ER-positive and -negative human breast carcinomas, Anticancer Res. 28 (1B) (2008) 479–483. [14] L. Shi, B. Dong, Z. Li, Y. Lu, T. Ouyang, J. Li, T. Wang, Z. Fan, T. Fan, B. Lin, Z. Wang,
[15]
[16]
[17]
[18]
[19]
[20] [21] [22]
10
Y. Xie, Expression of ER-{alpha}36, a novel variant of estrogen receptor {alpha}, and resistance to tamoxifen treatment in breast cancer, J. Clin. Oncol. 27 (21) (2009) 3423–3429. S. Vranic, Z. Gatalica, H. Deng, S. Frkovic-Grazio, L.M. Lee, O. Gurjeva, Z.Y. Wang, ERalpha36, a novel isoform of ER-alpha66, is commonly over-expressed in apocrine and adenoid cystic carcinomas of the breast, J. Clin. Pathol. 64 (1) (2011) 54–57. V. Miceli, L. Cocciadiferro, M. Fregapane, M. Zarcone, G. Montalto, L.M. Polito, B. Agostara, O.M. Granata, G. Carruba, Expression of wild-type and variant estrogen receptor alpha in liver carcinogenesis and tumor progrssion, OMICS 15 (5) (2011) 313–317. J. Zhang, J.W. Ren, J.M. Wei, C.C.N. Chong, D.G. Yang, Y.L. He, G.G. Chen, P.B.S. Lai, Alternative splicing of estrogen receptor alpha in hepatocellular carcinoma, BMC Cancer 16 (2016) 926. X.T. Zhang, H. Deng, Z.Y. Wang, Estrogen activation of the mitogen-activated protein kinase is mediated by ER-α36 in ER-positive breast cancer cells, J. Steroid Biochem. Mol. Bio. 143 (2014) 434–443. X.T. Zhang, L.G. Kang, L. Ding, S. Vranic, Z. Gatalic, Z.Y. Wang, A positive feedback loop of ER-α36/EGFR promotes malignant growth of ER-negative breast cancer cells, Oncogene 30 (2011) 770–780. T. Reya, S.J. Morrison, M.F. Clarke, I.L. Weissman, Stem cells, cancer, and cancer stem cells, Nature 414 (6859) (2001) 105–111. M. Dean, Cancer stem cells: redefining the paradigm of cancer treatment strategies, Mol. Interv. 6 (3) (2006) 140–148. X.L. Zhang, Q. Jia, L. Lv, T. Deng, J. Gao, Tumorspheres derived from HCC cell are enriched with cancer stem-like cells and present high chemoresistance dependent on the akt pathway, Anti-Cancer Agent Med. Chem. 15 (2015) 755–763.
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
The ER-α36/EGFR signaling loop promotes growth of hepatocellular carcinoma cells Hui You, Kun Meng, Zhao-Yi Wang
⁎
Beijing Shenogen Biomedical Co., Ltd, Beijing, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords: EGFR ER-α36 Rapid estrogen signaling Hepatocellular carcinoma
Hepatocellular carcinoma (HCC) is the common primary liver cancer and the third leading cause of cancer related mortality worldwide. It is generally thought that the estrogen-signaling pathway is not related to the development and progression of human HCC. However, accumulating evidences indicate the existence of a rapid estrogen signaling in HCC cells that is able to promote cell growth. However, the receptor that mediates the rapid estrogen signaling in HCC cells has not been established. Previously, our laboratory identified a variant of ER-α, ER-α36, and found that ER-α36 mediates the rapid estrogen signaling such as the activation of the MAPK/ERK signaling in breast carcinoma cells. Our current experiments studied the role of the rapid estrogen signaling mediated by ER-α36 in growth of HCC HepG2 and PLC/PRF/5 cells that highly express ER-α36 and found these cells were strongly responsive to the rapid estrogen signaling. Knockdown of ER-α36 expression in these HCC cells using the shRNA method attenuated their responsiveness to estrogen and destabilized EGFR protein. ERα36 mediated estrogen-induced phosphorylation of Src and the MAPK/ERK as well as cyclin D1 expression. In addition, there existed an ER-α36/EGFR positive regulatory loop in HCC cells that was important for the maintenance and positive regulation of HCC tumorsphere cells. Our results thus indicated that the rapid estrogen receptor is mediated by ER-α36 in HCC cells through the EGFR/Src/ERK signaling pathway and suggested that the ER-α36/EGFR signaling loop is a potential target to develop novel therapeutic approaches for HCC treatment.
1. Introduction Although the widespread implementation of different screening programs, improvement of diagnostic tools and the progress in therapeutic methods, the survival rate of hepatocellular carcinoma (HCC) remains poor and the morbidity still has been steadily increasing. A number of risk factors have been related to the pathogenesis of HCC, including HBV and HCV infections, aflatoxins, alcohol, diabetes and nonalcoholic fatty diseases [1]. It has been known for a long time that estrogen signaling is related to the pathogenesis of HCC although liver is not a classical estrogen target organ [1]. However, the molecular mechanism by which the estrogen signaling promotes hepatocellular carcinogenesis is largely unclear. Estrogen regulates cell growth through estrogen receptors (ERs), ER-α and β, both of which are members of the nuclear receptor superfamily and function as transcription factors for target genes [2]. The classical estrogen-signaling pathway includes estrogen binding to the receptors, then the receptors dimerization and binding to the estrogen response elements (ERE) of the target gene promoters,
⁎
regulating gene transcription [2]. Besides this classical genomic action, there is another signaling pathway called “membrane-initiated estrogen signaling” or “rapid estrogen signaling” that utilizes growth factors, their receptors, cytoplasmic proteins and other membrane signaling pathways [3–5]. Several studies demonstrated that a rapid estrogen signaling such as activation of IP3-PKC-α, the AKT and the MAPK/ERK contributes to the induction of cyclin D1 expression and DNA synthesis in HCC cells [6–9], indicating the existence of the rapid estrogens signaling in HCC. In 2015, our laboratory isolated and cloned a 36-kDa variant of ERα named ER-α36 to differentiate it from the full-length 66-kDa ER-α, ER-α66. ER-α36 was predominantly detected at the plasma membrane as well as the cytoplasm. The MAPK/ERK and the PI3K/AKT signaling pathways are typical pathways involved in the ER-α36-mediated rapid estrogenic signaling [10,11]. ER-α36 is generated from a promoter located in the first intron of the ER-α66 gene [12], indicating that ER-α36 expression is regulated differently from ER-α66, which is consistent with the findings that ER-α36 is expressed in ER-negative breast cancer [13–15]. Previously, ER-α36 expression was detected in specimens
Corresponding author at: Beijing Shenogen Biomedical Co., Ltd, #29 Life Science Park Road, Beijing, PR China. E-mail address: [email protected] (Z.-Y. Wang).
https://doi.org/10.1016/j.steroids.2018.02.007 Received 14 September 2017; Received in revised form 4 January 2018; Accepted 20 February 2018 0039-128X/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: You, H., Steroids (2018), https://doi.org/10.1016/j.steroids.2018.02.007
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 1. Estrogen and EGF stimulate growth of HCC cells. (A). The expression of ER-α variants and EGFR in HCC PLC/PRF/5 and HepG2. (B). The effect of E2 or EGF on the growth rate of HCC cells. Cells maintained for three days in the phenol red-free DMEM plus 2.5% dextran-charcoal-stripped fetal calf serum were treated with or without 0.1 nM E2 or 10 ng/ml EGF for different time periods. The cell numbers were examined using an automatic cell counter. Five dishes were used for each time point and experiments were repeated more than three times. The mean cell numbers are shown.
from HCC patients but not in the non-tumoral tissues while ER-α66 was mainly detected in the non-tumoral tissues and was decreased or became undetectable in HCC specimens [16,17], suggesting that ER-α36 may contribute to the development and progression of HCC. In this study, we investigated ER-α36-mediated rapid estrogen signaling in growth of HCC cells and in the tumorsphere cells derived from these HCC cells. We found the stimulatory effects of a positive feedback regulatory loop between ER-α36 and EGFR on the growth of HCC cells.
Fig. 2. Estrogen elicits rapid signaling and induces gene expression in HCC cells. (A). Western blot analysis of the effects of indicated concentrations of E2 on the phosphorylation levels of the MAPK/ERK1/2 and the AKT. The columns represent the means of three experiments ± S.D. *, P < 0.05 for cells treated with DMSO (0) vs cells treated with indicated concentrations of E2 for 30 min. (B). E2 induces expression of ER-α36, EGFR and cyclin D1 in HCC cells. Cells were treated with indicated concentrations of E2 or DMSO (0) as a control for 12 h. The membranes were stripped and re-probed with different antibodies.
2. Methods and materials
MAPK, Src and EGRR inhibitors U0126, PP2 and AG1478 were acquired from Calbiochem (San Diego, CA). The ER-α36 antibody was prepared as described before [11]. CyclinD1 antibody was purchased from Santa Cruz (Santa Cruz, CA). Antibodies for β-actin, ER-α66, ERK1/2, pERK1/2, Src, p-Src, AKT, p-AKT and EGFR, p-EGFR, anti-mouse IgGHRP, anti-rabbit IgG-HRP were all purchased from Cell Signaling Technology (Danvers, MA).
2.1. Reagents The 17β-estradiol (E2) and insulin was purchased from Sigma Chemical (St Louis, MO). B-27 was obtained from Gibco (Carlsbad, CA). Basic FGF and EGF were from Peprotech (East Brunswick, NJ). The
2
Steroids xxx (xxxx) xxx–xxx
H. You et al.
(FBS), 1% non-antibiotic-antimycotic from Gibco. HCC cells were cultured at 37 °C and 5% CO2. To establish ER-α36 knocked-down stable cell lines from PLC/PRF/ 5 and HepG2 cells, we cloned the DNA oligonucleotides 5-GATGCCA ATAGGTACTGAATTGATATCCGTTCAGT ACCTATTGGCAT-3′ targeting a sequence in the 3′UTR of ER-α36 gene into the pRNAT-U6.1/Neo expression vector (GenScript Corp, Piscataway, NJ) to construct an ERα36 specific shRNA expression vector. Briefly, the empty expression vector and ER-α36 shRNA expression vector were transfected into the HCC cells and the transfected cells were selected with 600 μg/ml G418 (Gibco) for three weeks. After then, cells were pooled from at least 25 clones. For cell growth assay, cells at the logarithmic phase were collected and 10,000 cells were seeded onto 6-well plates and cultured with phenol red-free DMEM/F12 (Gibco), respectively, containing 1% charcoal-stripped fetal calf serum (Gibco), 1% non-antibiotic-antimycotic. Media were changed every two days and treated with 0.1 nM of E2 or 10 ng/ml of EGF, respectively. Every two days, the cell numbers were examined using the Countess®II automatic cell counter (Thermo Fisher Scientific, Waltham, MA).
2.3. Tumorsphere formation and estrogen stimulation HCC cells at 10,000 cells/ml were seeded onto the ultra-low attachment 6-well plate (Corning Incorporated, Pasadena, CA) and cultured for five days in the tumorsphere medium: phenol-red free DMEM/ F12 medium (Gibco) containing 10 ng/ml basic FGF, 4 μg/ml insulin (Sigma), 1×B-27 (Gibco), After five days, the tumorspheres were photographed with the Nikon Eclipse TE2000-S microscope. Tumorspheres were collected, washed, dissociated with Trypsin, and then re-suspended. The cell number was determined using the Countess®II automatic cell counter (Thermo Fisher Scientific). For estrogen stimulation assays, 0.1 nM E2 or vehicle (DMSO) was used to treat tumorspheres. All the experiments were repeated at least three times.
2.4. Western blots analysis HCC cells were harvested and washed with cold PBS. The RIPA buffer was used to lyse cells and the cell lysates was boiled in loading buffer. After electrophoresis, the proteins were transferred onto PVDF membranes (EMD Millipore, Danvers, MA), then blocked in the buffer containing 5% BSA, and the primary antibodies β-actin (1:3000), ERα36 (1:2000), ER-α66, ERK1/2, p-ERK1/2, Src, p-Src, AKT, p-AKT and EGFR, p-EGFR (all 1:1000), incubated overnight at 4 °C. The filters were incubated with appropriate anti-mouse IgG or anti-rabbit IgG at room temperature for two hours. Protein bands were visualized with ECL kit (EMD Millipore, Danvers, MA) and photographed with Chemiluminescence Imaging System (LI-COR, Lincoln, NE). All the experiments were repeated at least three times and analyzed with IMAGE J.
Fig. 3. ER-α36 mediates rapid estrogen signaling in HCC cells. (A). Western blot of ERα36, ER-α66 and EGFR expression in different HCC cell variants, control cells (PLC/RPF/ 5/C and HepG2/C); HCC cells transfected with the empty expression vector and ER-α36 expression knocked-down cells (PLC/PRF/5/Sh36 and HepG2/Sh36). (B). Western blot analysis of the effects of indicated concentrations of E2 on the phosphorylation level of the MAPK/ERK1/2 (30 min) and cyclin D1 expression (12 h) in different HCC cell variants. β-actin was used for a loading control for cyclin D1 expression (C). The effects of E2 on the growth of different HCC cell variants. Cells were treated with 0.1 nM E2 or DMSO (vehicle) as a control. Five dishes were used for each concentration and experiments were repeated more than five times. The mean cell numbers ± S.D. are shown. *, P < 0.05 for cells treated with DMSO (vehicle) vs cells treated 0.1 nM E2.
2.5. Statistical analysis Results were expressed as the mean ± standard deviation (S.D.). Statistical analysis was performed using GraphPad InStat software program with paired-samples t-test, or ANOVA followed by the Student’s t-test and the significance was accepted for P < 0.05 levels.
2.2. Cell culture, plasmids and growth assays PLC/PRF/5 and HepG2 cell lines were originally purchased from Japanese Cancer Research Bank (Tokyo). HCC cells were cultured in the complete DMEM medium (Gibco) containing 5% fetal bovine serum
3
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 4. The EGFR/Src/ERK axis is involved in estrogen induction of cyclin D1 and cell growth in HCC cells. (A). Western blot analysis of phosphorylation levels of the ERK1/2, the Src-Tyr416 and the EGFR-Tyr-845 in HCC cells. The HCC cells were treated with 0.1 nM E2 for different time periods. Total proteins were used as loading controls for the phosphorylated ERK1/2 and the Src-Tyr416. β-actin was used as a loading control for EGFR expression (B). Western blot analysis of cyclin D1 expression in HCC cells treated with 0.1 nM of E2 with or without the Src inhibitor PP2, the MAPK inhibitor UO126 and the EGFR inhibitor AG1478 for 12 h. All experiments were repeated at least three times, and the representative results are shown. (C). Cell growth assays of the HCC cells treated with 0.1 nM E2 together with or without the indicated inhibitors as well as the indicated inhibitors alone. The mean cell numbers ± SD are shown. *, P < 0.05 for cells treated with DMSO (vehicle) vs cells treated 0.1 nM E2.
3. Results
in HCC cells (Fig. 1A). As a control, we also examined ER-α66 expression and found that both cell lines express trace amounts of ER-α66 (Fig. 1A). We then decided to determine if estrogen stimulates growth of these HCC cells. As shown in Fig. 1B, the cells treated with 17βestradiol (E2) exhibited a remarkably increased growth rate compared to the cells treated with the vehicle (Fig. 1B). We further found that EGF also potently promoted growth of both HCC cell lines (Fig. 1B). Our data thus indicated that both estrogen and EGF signaling pathways are involved in growth of these HCC cells.
3.1. Estrogen and EGF stimulate growth of HCC cells It has been reported that samples from primary HCC exhibited a decreased ER-α66 expression and an increased ER-α36 expression [16,17]. To determine if established HCC cells retain ER-α36 expression, we used two well-known hepatoma cell lines HepG2 and PLC/ PRF/5. The result showed that ER-α36 and EGFR were highly expressed
4
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 5. Knockdown of ER-α36 expression attenuates EGF signaling of HCC cells. (A). Western blot analysis of the phosphorylation levels of the MAPK/ERK and the AKT in the HCC cell variants treated with 10 ng/ml EGF for 30 min. Columns: means of three independent experiments; bars, S.D. *, p < 0.05, for cells treated with PBS vs EGF. (B). Western blot analysis of cyclin D1 expression in the HCC cell variants treated with 10 ng/ml EGF for 12 h. (C). Cell growth assays of the HCC cell variants treated with EGF (10 ng/ml). Five dishes were used and experiments were repeated three times. The mean cell numbers are shown; bars, S.D. *, p < 0.05, for cells treated with PBS vs EGF.
level of ER-α36 using the shRNA method. Empty expression vector were also transfected into PLC/PRF/5 and HepG2 cells to serve as controls. Western blot analysis indicated that ER-α36 expression was lessened about 80% compared with the control cells (Fig. 3A), while ER-α66 expression was intact in these cells (Fig. 3A). Consequently, EGFR expression was dramatically down-regulated in the HCC cells with knocked-down level of ER-α36 expression (Fig. 3A), E2 treatment failed to induce ERK1/2 phosphorylation, cyclin D1 expression and cell growth in the HCC cell lines with knocked-down level of ER-α36 expression (Fig. 3B&C). These results indicated that ER-α36 mediates the rapid and mitogenic estrogen signaling in these HCC cells.
3.2. Rapid estrogen signaling and induction of ER-α36/EGFR expression in HCC cells The finding that HCC cells express ER-α36 suggested that these cells might harbor the rapid estrogen signaling. To determine whether E2 induced the MAPK/ERK1/2 phosphorylation in HCC cells, different concentrations of E2 were used to treat HCC cells for 30 min to analyze phosphorylation of the ERK1/2 using Western blot analysis. Fig. 2A reveals that E2 induced ERK phosphorylation in both cell lines (Fig. 2A). In addition, E2 also induced the phosphorylation of the AKT in these HCC cells (Fig. 2A). Thus, our results indicated that these HCC cells exists the rapid estrogen signaling. Consistent with the previous report [17], E2 could induce the expression levels of both ER-α36 and EGFR in these HCC cells (Fig. 2B), suggesting the existence of the ERα36/EGFR positive regulatory loop previously reported in breast cancer cells. Consequently, E2 could also induce cyclin D1 expression in these HCC cells (Fig. 2B).
3.4. The EGFR/Src/ERK axis is involved in estrogen induction of cyclin D1 and cell growth in HCC cells Recently, we reported that EGFR, Src and the MAPK/ERK all contributed to estrogen-stimulated growth of breast carcinoma cells [17]. We decided to determine if HCC cells also harbor the same signaling axis. E2 at 0.1 nM was added into the HCC cell lines for different time periods. Phosphorylation levels of the ERK1/2 and Src were assessed using Western blot analysis. Fig. 4 indicates that E2 elicited phosphorylation of the MAPK/ERK in 10 min. (Fig. 4A). In addition, E2 also
3.3. ER-α36 mediates the rapid estrogen signaling in HCC cells To determine if ER-α36 mediates the rapid estrogen signaling of HCC cells, we established HCC cell sub-lines to express knocked-down
5
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 6. HCC cells exist an ER-α36/EGFR positive regulatory loop. (A). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells with ER-α36 expression knocked-down transfected with an ER-α36 expression vector containing the ER-α36 cDNA without the 3′UTR. (B). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells with ER-α36 expression knocked-down transfected with an EGFR expression vector. (C). Western blot analysis of the ER-α36 and EGFR expression in the HCC cells treated with EGF (10 ng/ml) for 12 h.
Fig. 5. (continued)
induced phosphorylation of Src-Tyr-416 in these HCC cells (Fig. 4A). We also found that E2 induced EGFR-Tyr-845 phosphorylation as well as the expression level of EGFR protein (Fig. 4A). Thus, our results indicated that the EGFR/Src/ERK axis is involved in the rapid estrogen signaling of these HCC cells. To further confirm these findings, we utilized the chemical inhibitors for EGFR (AG1478), Src (PP2) and the MAPK/ERK (U0126). We found that all inhibitors potently diminished E2-induced cyclin D1 expression (Fig. 4B) and cell growth (Fig. 4C) in these HCC cells. Our results thus demonstrated that the EGFR/Src/ERK signaling axis is involved in the ER-α36-mediated estrogen induction of cyclin D1 expression.
cells (Fig. 5B). Consequently, the growth of the HCC cells with ER-α36 knock-down failed to response to EGF stimulation (Fig. 5C). Thus, the EGF signaling is attenuated in the HCC cells with ER-α36 expression knocked-down presumably due to down-regulated EGFR expression. 3.6. HCC cells exist an ER-α36/EGFR positive regulatory loop To further confirm there exists a positive regulatory loop between ER-α36 and EGFR in HCC cells, we introduced the ER-α36 and EGFR expression vectors into the HCC cells with knocked-down levels of ERα36 expression. We found that re-introduction of recombinant ER-α36 that lacks the 3′UTR into these cells increased the expression level of EGFR protein (Fig. 6A). In addition, we also observed that recombinant EGFR expression in the HCC cells with knocked-down levels of ER-α36 also increased endogenous ER-α36 expression (Fig. 6B), although the induction was modest probably due to the existence of the ER-α36 shRNA. Previously, we reported that ER-α36 expression is subjected to the positive regulation of EGFR signaling [18]. We then treated HCC cells with 10 ng/ml of EGF and then examined ER-α36 and EGFR expression by Western blot. We found that EGF treatment was able to induce both ER-α36 and EGFR expression in HCC cells (Fig. 6C). Thus,
3.5. Knockdown of ER-α36 expression attenuates EGF signaling of HCC cells Previously, we reported that there is a positive regulatory loop between EGFR and ER-α36 in breast cancer cells; EGFR signaling activates the ER-α36 promoter activity and ER-α36 stabilizes EGFR protein [18]. Here, our Western blot results revealed that EGFR expression was remarkably reduced in HCC cells with ER-α36 expression knocked-down. We decided to determine whether the EGF signaling is impacted in these HCC cells. The results indicated that the MAPK/ERK and AKT phosphorylation induced by EGF was significantly attenuated in the HCC cells with ER-α36 expression knocked-down (Fig. 5A). In addition, EGF induced cyclin D1 expression was also diminished in these HCC
6
Steroids xxx (xxxx) xxx–xxx
H. You et al.
whereas ER-α66 expression remained unchanged (Fig. 8B). Finally, we observed that 0.1 nM E2 treatment potently increased tumorsphere cell number and the size from the control HCC cells but not from the HCC cells with ER-α36 expression knocked-down (Fig. 8C&D). We also observed that in the absence of 0.1 nM E2, the HCC cells with knockeddown ER-α36 expression also formed less number and smaller size tumorspheres in comparison with the control cells (Fig. 8C&D). Our results demonstrated that the ER-α3/EGFR positive regulatory loop is important for maintenance and regulation of cancer stem-like cells in HCC. 4. Discussion In this report, HCC PLC/PRF/5 and HepG2 cells were used to study the effect of the rapid estrogen signaling mediated by ER-α36 on the growth of HCC cells. Here, we found that HCC cells expressing trace amount of ER-α66 but high level of ER-α36 exhibited a potent mitogenic estrogen signaling. Previously, HCC cells such as HepG2 have been shown to rapidly respond to estrogen through activation of the PKC-α, the MAPK/ERK and the AKT as well as induction of cyclin D1 expression and DNA synthesis [6–9]. However, the mediator of these rapid estrogen activities in HCC has not been established. In this paper, we used an ER-α36specific shRNA to knock-down ER-α36 expression in these HCC cells and found that the PLC/PRF/5 and HpeG2 cells with knocked-down level of ER-α36 expression exhibited no responsiveness to the mitogenic estrogen signaling, indicating that ER-α36 is the receptor that mediates the rapid estrogen signaling in HCC cells. Previously, we reported that EGFR and ER-α36 positively regulate each other’s expression in breast cancer cells; the promoter activity of ER-α36 is activated by EGFR signaling and ER-α36 stabilizes the steady state levels of EGFR protein [18]. In this paper, we also found the existence of the ER-α36/EGFR positive regulatory loop in HCC cells that plays an important role in cell growth. We showed that estrogen induced the MAPK/ERK phosphorylation and cyclin D1 expression through a pathway that includes ER-α36 and the EGFR/Src axis. Previously, Our laboratory demonstrated that ER-α36 interacted strongly with the EGFR/Src/Shc complex and that estrogen primarily induced the EGFR-Tyr-845 phosphorylation [17,18]. Our laboratory also demonstrated that estrogen activated phosphorylation of Src-Tyr-416 and the AG1478 as well as PP2 blocked estrogen-induced cyclin D1 expression. In the current study, we identified that the EGFR/Src complex plays an important role in the rapid estrogen signaling of the HCC cells that express ER-α36. It is worth to note that E2 treatment also augmented the expression level of EGFR protein in HCC cells with 30 min, suggesting the enhanced transcription may not be involved, which is well in accordance with our previous report that ER-α36 functions at post-transcription level to stabilize EGFR protein [19]. In the past decade, a small number of tumor cells with stem cell properties have been described in different kind of cancers [20]. These stem-like cancer cells are responsible for the initiation, recurrence and metastasis of tumor and also for the failure of the current therapies [20,21]. Recently, it has been reported that HCC tumorsphere cells are rich in cancer stem-like cells [22]. Based on this, we cultured tumorspheres from both HCC cell lines and used them to study their responses to estrogen. Here, our results showed that tumorsphere cells derived from the established HCC cells exhibited enhanced expression and activities of the ER-αβ6/EGFR positive regulatory loop and the HCC cells with ER-α36 expression knocked-down generated smaller size and less number tumorspheres compared with the control cells, suggesting that this loop plays a critical role in these cancer stem-like cells from HCC. We also documented that E2 treatment could increase
Fig. 7. Tumorsphere cells exhibit enhanced growth-promoting signals. (A). Western blot analysis of the ER-α36, ER-α66 and EGFR expression in the tumorsphere cells derived from the HCC cells. The membranes were stripped and re-probed with different antibodies. (B). Western blot analysis of the phosphorylation levels of the MAPK/ERK and the AKT in the tumorsphere cells derived from the HCC cells.
our results strongly indicated that HCC cells also contain the ER-α36/ EGFR positive regulatory loop previously reported in breast cancer cells. 3.7. The ER-α36/EGFR positive regulatory loop plays an important role in tumorspheres derived from HCC cells Accumulating experimental and clinical studies indicated that cancer stem-like cells contribute the development and progression of human cancer [20]. Our laboratory found that OCT-4 and Nanog were highly expressed in the tumorsphere cells derived from HCC cells (data not shown). To examine the effect of the ER-α36/EGFR positive regulatory loop on growth of cancer stem-like cells derived from HCC cells, we cultured HCC cells in tumorsphere medium that lacks EGF to form tumorspheres. We found that both ER-α36 and EGFR expression as well as the downstream signals such as the MAPK/ERK and AKT phosphorylation were significantly enhanced in HCC tumorsphere cells while ER-α66 expression was without significant changes (Fig. 7A&B). We further found that E2 treatment also augmented the MAPK/ERK and the AKT phosphorylation in these tumorsphere cells (Fig. 8A), suggesting that tumorsphere cells derived from HCC cells retain the rapid estrogen signaling. Consequently, E2 treatment increased the expression levels of ER-α36, EGFR and cyclin D1 in these tumorsphere cells
7
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 8. The rapid estrogen signaling expands the tumorsphere cells derived from the HCC cells. (A). Western blot analysis of phosphorylation levels of the MAPK/ERK1/2 and the AKT in the tumorsphere cells derived from the HCC cells treated E2. The tumorsphere cells were treated with indicated concentrations of E2 for 30 min. The columns represent the means of three experiments; bars, S.D. *, P < 0.05 for cells treated with vehicle (0) vs cells treated different concentrations of E2. (B). Western blot analysis of ER-α36, EGFR and cyclin D1 expression in the tumorsphere cells derived from the HCC cells treated with E2 for 12 h. The tumorsphere cells were treated with different concentrations of E2 for 12 h. (C). ER-α36-mediated estrogen signaling positively regulates the size and number of tumorspheres from HCC cells. Representative tumorspheres from HCC cell variants treated with vehicle or 0.1 nM E2 for seven days. Magnification: X200, and X400 for the inserted pictures. (D). The numbers of cells from dissociated tumorspheres of different HCC cell variants were determined. The columns represent the means of three experiments; bars, S.D. *, P < 0.05 for cells treated with vehicle vs cells treated with 0.1 nM E2. #, P < 0.05 for the control HCC cells vs the HCC cells with ER-α36 expression knocked-down.
Thus, ER-α36 is a novel player in the rapid estrogen signaling during HCC tumorigenesis. The finding that the ER-α36/EGFR regulatory loop plays an important role in the maintenance and positive regulation of cancer stem-like cells from HCC also highlights the importance of targeting this positive regulatory loop as a novel approach to treat human HCC.
tumorsphere cell numbers and sizes, suggesting estrogen was able increase the pool of cancer stem-like cells in HCC cells via the ER-α36mediated signaling. In summary, we have shown that ER-α36 expressing HCC cells exhibited quick responses to estrogen, indicating that the rapid estrogen signaling contributes to development and progression of HCC.
8
Steroids xxx (xxxx) xxx–xxx
H. You et al.
Fig. 8. (continued)
Conflict of interest
References
The authors declare no conflict of interest.
[1] E. Villa, Role of estrogen in liver cancer, Women’s Health 4 (1) (2008) 41–50. [2] E.H. Kong, A.C. Pike, R.E. Hubbard, Structure and mechanism of the oestrogen receptor, Biochem. Soc. Trans. 31 (2003) 56–59. [3] J.H. Segars, P.H. Driggers, Estrogen action and cytoplasmic signaling cascades. Part I: membrane-associated signaling complexes, Trends Endocrinol. Metabol. 13 (2002) 349–354. [4] P.H. Driggers, J.H. Segars, Estrogen action and cytoplasmic signaling pathways. Part II: the role of growth factors and phosphorylation in estrogen signaling, Trends Endocrinol. Metab. 13 (2002) 422–427. [5] M.J. Kelly, E.R. Levin, Rapid actions of plasma membrane estrogen receptors, Trends Endocrinol. Metab. 12 (2001) (2001) 152–156.
Acknowledgement This work was supported by “Significant New Drug Development” National Science and Technology major project (2013ZX09401004).
9
Steroids xxx (xxxx) xxx–xxx
H. You et al.
[6] M. Marino, V. Pallottini, A. Trentalance, Estrogen cause rapid activation of IP3-PKC-α signal transduction pathway in HepG2 cells, Biochem. Biophys. Res. Comm. 245 (1998) 251–258. [7] M. Marino, E. Distefano, S. Caporali, G. Ceracchi, V. Pallottini, A. Trentalance, β-estradiol stimulation of DNA synthesis requires different PKC isoforms in HepG2 and MCF7 cells, J. Cell. Phys. 188 (2001) 170–177. [8] M. Marino, F. Acconcia, F. Bresciani, A. Weisz, A. Trentalance, Distinct nongenomic signal transduction pathways controlled by 17b-Estradiol regulate DNA synthesis and cyclin D1 gene transcription in HepG2 cells, Mol. Biol. Cell 13 (2002) 3720–3729. [9] M. Marino, F. Acconcia, A. Trentalance, Biphasic estradiol induced AKT phosphorylation is modulated by PTEN via MAP kinase in HepG2 cells, Mol. Biol. Cell 14 (2003) 2583–2591. [10] Z. Wang, X. Zhang, P. Shen, B.W. Loggie, Y. Chang, T.F. Deuel, Identification, cloning, and expression of human estrogen receptor-alpha36, a novel variant of human estrogen receptor-alpha66, Biochem. Biophys. Res. Commun. 336 (4) (2005) 1023–1027. [11] Z. Wang, X. Zhang, P. Shen, B.W. Loggie, Y. Chang, T.F. Deuel, A variant of estrogen receptor-{alpha}, hER-{alpha}36: transduction of estrogen- and antiestrogen-dependent membrane-initiated mitogenic signaling, Proc. Natl. Acad. Sci. U.S.A. 103 (24) (2006) 9063–9068. [12] Y. Zou, L. Ding, M. Coleman, Z. Wang, Estrogen receptor-alpha (ER-alpha) suppresses expression of its variant ER-alpha 36, FEBS Lett. 583 (8) (2009) 1368–1374. [13] L.M. Lee, J. Cao, H. Deng, P. Chen, Z. Gatalica, Z.Y. Wang, ER-alpha36, a novel variant of ER-alpha, is expressed in ER-positive and -negative human breast carcinomas, Anticancer Res. 28 (1B) (2008) 479–483. [14] L. Shi, B. Dong, Z. Li, Y. Lu, T. Ouyang, J. Li, T. Wang, Z. Fan, T. Fan, B. Lin, Z. Wang,
[15]
[16]
[17]
[18]
[19]
[20] [21] [22]
10
Y. Xie, Expression of ER-{alpha}36, a novel variant of estrogen receptor {alpha}, and resistance to tamoxifen treatment in breast cancer, J. Clin. Oncol. 27 (21) (2009) 3423–3429. S. Vranic, Z. Gatalica, H. Deng, S. Frkovic-Grazio, L.M. Lee, O. Gurjeva, Z.Y. Wang, ERalpha36, a novel isoform of ER-alpha66, is commonly over-expressed in apocrine and adenoid cystic carcinomas of the breast, J. Clin. Pathol. 64 (1) (2011) 54–57. V. Miceli, L. Cocciadiferro, M. Fregapane, M. Zarcone, G. Montalto, L.M. Polito, B. Agostara, O.M. Granata, G. Carruba, Expression of wild-type and variant estrogen receptor alpha in liver carcinogenesis and tumor progrssion, OMICS 15 (5) (2011) 313–317. J. Zhang, J.W. Ren, J.M. Wei, C.C.N. Chong, D.G. Yang, Y.L. He, G.G. Chen, P.B.S. Lai, Alternative splicing of estrogen receptor alpha in hepatocellular carcinoma, BMC Cancer 16 (2016) 926. X.T. Zhang, H. Deng, Z.Y. Wang, Estrogen activation of the mitogen-activated protein kinase is mediated by ER-α36 in ER-positive breast cancer cells, J. Steroid Biochem. Mol. Bio. 143 (2014) 434–443. X.T. Zhang, L.G. Kang, L. Ding, S. Vranic, Z. Gatalic, Z.Y. Wang, A positive feedback loop of ER-α36/EGFR promotes malignant growth of ER-negative breast cancer cells, Oncogene 30 (2011) 770–780. T. Reya, S.J. Morrison, M.F. Clarke, I.L. Weissman, Stem cells, cancer, and cancer stem cells, Nature 414 (6859) (2001) 105–111. M. Dean, Cancer stem cells: redefining the paradigm of cancer treatment strategies, Mol. Interv. 6 (3) (2006) 140–148. X.L. Zhang, Q. Jia, L. Lv, T. Deng, J. Gao, Tumorspheres derived from HCC cell are enriched with cancer stem-like cells and present high chemoresistance dependent on the akt pathway, Anti-Cancer Agent Med. Chem. 15 (2015) 755–763.