Activation of GPER suppresses epithelial mesenchymal transition of triple negative breast cancer cells via NF-κB signals

Abstract

The targeted therapy for triple-negative breast cancer (TNBC) is a great challenge due to our poor understanding on its molecular etiology. In the present study, our clinical data showed that the expression of G-protein coupled estrogen receptor (GPER) is negatively associated with lymph node metastasis, high-grade tumor and fibronectin (FN) expression while positively associated with the favorable outcome in 135 TNBC patients. In our experimental studies, both the in vitro migration and invasion of TNBC cells were inhibited by GPER specific agonist G-1, through the suppression of the epithelial mesenchymal transition (EMT). The G-1 treatment also reduced the phosphorylation, nuclear localization, and transcriptional activities of NF-κB. While over expression of NF-κB attenuated the action of G-1 in suppressing EMT. Our data further illustrated that the phosphorylation of GSK-3β by PI3K/Akt and ERK1/2 mediated, at least partially, the inhibitory effect of G-1 on NF-κB activities. It was further confirmed in a study of MDA-MB-231 tumor xenografts in nude mice. The data showed that G-1 inhibited the in vivo growth and invasive potential of TNBC via suppression of EMT. Our present study demonstrated that an activation of GPER pathway elicits tumor suppressive actions on TNBC, and supports the use of G-1 therapeutics for TNBC metastasis.

Keywords: EMT; G-1; GPER; NF-κB; TNBC.

Figure 1

The roles of GPER in TNBC clinical samples. (A) Immunohistochemical analysis of GPER in TNBC samples. The carcinomas showed four distinct patterns of expression: no expression (a), low expression (b), medium expression (c), and high expression (d); (B) Kaplan–Meier analysis of cancer specific survival for all patients.

Figure 2

Activation of GPER inhibits in vitro migration and invasion of TNBC cells. (A) The expression of GPER in human cancer cells were measured by Western blot analysis; Confluent monolayers of MDA‐MB‐231 (B) or BT‐549 (C) cells were scraped by a pipette tip to generate wounds and then were treated with or without 1 μM G‐1 for 0, 24, and 48 h, respectively; quantitative analysis of wound healing assay was showed in the right column; (D) MDA‐MB‐231 and BT‐549 cells were allowed to invasive transwell chambers for 48 h in the presence or absence of 1 μM G‐1. Then invaded cells were fixed, stained, and photographed; (E) MDA‐MB‐231 were transfected with si‐NC or si‐GPER for 24 h, scraped by a pipette tip to generate wounds, and then treated with or without 1 μM G‐1 for 48 h; (F) MDA‐MB‐231 cells were transfected with si‐NC or si‐GPER for 24 h and then allowed to invasive transwell chambers for 48 h in the presence or absence of 1 μM G‐1. Data represented the average of three independent experiments. *p < 0.05 compared with control, **p < 0.01 compared with control. Scale bar = 100 μm.

Figure 3

Activation of GPR30 suppresses EMT of TNBC cells while has no effect on the related transcription factors. (A) MDA‐MB‐231 cells were treated with various concentrations of G‐1 for 48 h, and then the cell morphological changes were recorded by a phase contrast microscope. MDA‐MB‐231 cells were treated with 1 μM G‐1 for the indicated times. Then the expression of epithelial maker E‐cad and mesenchymal marker FN were detected by Western blot analysis (B) or qRT‐PCR (C), respectively. (D) MDA‐MB‐231 cells were treated with 1 μM G‐1 for 24 or 48 h, then expression of FN and E‐Cad (green) was analyzed by immunofluorescence staining. Nuclei were visualized with DAPI staining. (E) MDA‐MB‐231 cells were treated with 1 μM G‐1 for the indicated times, and then protein levels of Snail, Slug, Twist, and ZEB1 were detected by Western blotting; (F) MDA‐MB‐231 cells were treated with 1 μM G‐1 for the indicated times, nuclear and cytoplasmic cellular fractions were isolated by differential lysis. The levels of Snail and Slug in nuclear and cytoplasmic cellular fractions were detected by Western blot analysis. Data represented the average of three independent experiments. *p < 0.05 compared with control, **p < 0.01 compared with control. Scale bar = 100 μm.

Figure 4

NF‐κB mediates the suppression effects of G‐1 on EMT. (A) MDA‐MB‐231 cells were treated with 1 μM G‐1 for the indicated times, and then the expression of phosphorylation and total p65 were measured by Western blot analysis; (B) MDA‐MB‐231 cells were treated with 1 μM G‐1 for 24 or 48 h, nuclear and cytoplasmic cellular fractions were isolated by differential lysis. The levels of p65 in nuclear and cytoplasmic cellular fractions were detected by Western blot analysis; (C) MDA‐MB‐231 cells were transfected with a luciferase reporter construct containing 5 copies of the κB site plasmid for 24 h and treated with 1 μM G‐1 for the indicated times, then the lysates were assayed. Shown are relative luciferase activities normalized to Renilla activities; (D) MDA‐MB‐231 or BT‐549 cells were allowed to invade transwell chambers for 48 h in the presence or absence of10 μM BAY 11‐7028, then invaded cells were fixed and counted; (E) MDA‐MB‐231 or BT‐549 cells were treated with or without 10 μM BAY 11‐7028 for 48 h, the expression of FN, E‐Cad, and Vim were detected by Western blot analysis; (F) MDA‐MB‐231 or BT‐549 cells were transfected with pcDNA3.1 (Vector control) or pcDNA/p65 for 24 h and then treated with 1 μM G‐1 for further 48 h. Then the expression of mesenchymal marker FN and epithelial maker E‐cad were detected by Western blot analysis. Data represented the average of three independent experiments. *p < 0.05 compared with control, **p < 0.01 compared with control.

Figure 5

Phosphorylation of GSK‐3β by PI3K/Akt and ERK1/2 participates the process of G‐1 suppressed NF‐κB activities. (A) MDA‐MB‐231 were treated with 1 μM G‐1 for the indicated times, and then the phosphorylation and total protein levels of PKC, ERK1/2, Akt, and GSK‐3β were detected by Western blot analysis; (B) MDA‐MB‐231 cells were transfected with si‐NC or GSK‐3β siRNAs for 24 h, and then the p‐GSK‐3β, p‐p65, p65, and FN were detected by Western blot analysis; (C) MDA‐MB‐231 cells were treated with LiCl (GSK‐3β inhibitor) or G‐1 for 30 min, and then the p‐GSK‐3β, p‐p65, p65, and FN were detected by Western blot analysis; (D) MDA‐MB‐231 cells were pretreated with 10 μM ERK1/2 inhibitor PD98059 (PD), PI3K inhibitor LY294002 (LY), or PKC inhibitor GF109203X (GF) for 90 min, and then exposed to 1 μM G‐1 for further 15 min, the phosphorylation and total protein levels of GSK‐3β were detected by Western blot analysis; (E) MDA‐MB‐231 cells were treated with G‐1 for 12 h, and then GSK‐3β and Akt were immunoprecipitated, respectively, from equal amount of lysates and the associated GSK‐3β or Akt were detected by Western blot analysis. Data represented three independent experiments.

Figure 6

Activation of GPER inhibits the growth and metastasis of TNBC in vivo. (A) The tumor volumes of G‐1 (n = 13) and control (n = 11) group at the end of experiment; (B) The total and phosphorylated proteins related to the metastasis suppression effects of G‐1 were determined by Western blot analysis in the tumor lysates from the control and G‐1 treated group; (C) The tumor tissue sections of control and G‐1 group were subjected to IHC detection of FN and E‐Cad; (D) H&E examination of metastasis in lung tissue sections of the control and G‐1 group. Scale bar = 100 μm.

Figure 7

A proposed model to illustrate the mechanisms of GPER mediated EMT suppression of TNBC cells.