Eltrombopag Inhibits Metastasis in Breast Carcinoma by Targeting HuR Protein

Abstract

Eltrombopag is a small molecule TPO-R agonist that has been shown in our previous studies to inhibit tumor growth by targeting Human antigen R (HuR) protein. HuR protein not only regulates the mRNA stability of tumor growth-related genes, but it also regulates the mRNA stability of a variety of cancer metastasis-related genes, such as Snail, Cox-2, and Vegf-c. However, the role and mechanisms of eltrombopag in breast cancer metastasis have not been fully investigated. The purpose of this study was to investigate whether eltrombopag can inhibit breast cancer metastasis by targeting HuR. Our study first found that eltrombopag can destroy HuR-AU-rich element (ARE) complexes at the molecular level. Secondly, eltrombopag was found to suppress 4T1 cell migration and invasion and inhibit macrophage-mediated lymphangiogenesis at the cellular level. In addition, eltrombopag exerted inhibitory effects on lung and lymph node metastasis in animal tumor metastasis models. Finally, it was verified that eltrombopag inhibited the expressions of Snail, Cox-2, and Vegf-c in 4T1 cells and Vegf-c in RAW264.7 cells by targeting HuR. In conclusion, eltrombopag displayed antimetastatic activity in breast cancer in an HuR dependent manner, which may provide a novel application for eltrombopag, hinting at the multiple effects of HuR inhibitors in cancer therapy.

Keywords: HuR; breast cancer; eltrombopag (ELB); lung metastasis; lymph node metastasis; macrophage.

Figure 1

Eltrombopag interrupted the interaction between HuR RRM12 domain and ARE sequences (ARE^Snail, ARE^Cox-2, and ARE^Vegf-c). (A–C) The inhibitory effect of eltrombopag on the binding of HuR RRM12 and ARE^Snail, ARE^Cox-2, and ARE^Vegf-c by EMSA method. (D–F) The IC₅₀ of eltrombopag disrupts binding of HuR RRM12 to ARE (Snail, Cox-2, and Vegf-c) by FP method. n = 3.

Figure 2

The role of eltrombopag in migratory and invasive of 4T1 cells. (A) Impact of eltrombopag to expression of E-cadherin protein in 4T1 cells, representative images and quantitative data of immunofluorescence experiments are shown (Scale bar: 100 μm). (B) Impact of eltrombopag to expression of vimentin protein in 4T1 cells, representative images and quantitative data of immunofluorescence experiments are shown (Scale bar: 100 μm). (C) Scratch wound assay was used to determine the migration effect of eltrombopag on 4T1 cells, representative images and quantitative data are shown (Scale bar: 500 μm). (D) Transwell migration experiment was used to determine the vertical migration effect of eltrombopag on 4T1 cells, representative images and quantitative data are shown (Scale bar: 100 μm). (E) Transwell invasion assay was used to investigate the inhibitory effect of eltrombopag on the invasion of 4T1 cells, representative images and quantitative analyses are shown (Scale bar: 100 μm). Mean ± SD. n = 3. ** p < 0.01 vs. control group (t-test).

Figure 3

The role of eltrombopag in migratory of HuR-knockout 4T1 cells (ΔHuR 4T1) and HuR-rescued ΔHuR 4T1 cells (ΔHuR 4T1 HuROE). (A) Wound healing assay was used to determine the migration of eltrombopag on ΔHuR 4T1 cells, representative images and quantitative data are shown (Scale bar: 500 μm). (B) Transwell migration experiment was used to determine the vertical migration effect of eltrombopag on ΔHuR 4T1 cells, representative images and quantitative data are shown (Scale bar: 100 μm). (C) Wound healing assay was used to determine the migration of eltrombopag on ΔHuR 4T1 HuROE cells, representative images and quantitative data are shown (Scale bar: 500 μm). (D) Transwell migration experiment was used to determine the vertical migration effect of eltrombopag on ΔHuR 4T1 HuROE cells, representative images and quantitative data are shown (Scale bar: 100 μm). Mean ± SD. n = 3. ** p < 0.01 vs. control group (t-test). ns means no significant difference.

Figure 4

Impact of eltrombopag on the expression of Snail, Cox-2, and Vegf-c in 4T1, HuROE 4T1, or ΔHuR 4T1 cells. (A) The mRNA levels of Snail, Cox-2, and Vegf-c in 4T1 cells were tested by qRT-PCR. (B) The protein expression of Snail, Cox-2, and Vegf-c was detected in 4T1 cells via Western blot, photographs, and quantification. (C) The protein expression of Snail, Cox-2, and Vegf-c was detected in HuROE 4T1 cells via Western blot, photographs, and quantification. (D) The protein expression of Snail, Cox-2, and Vegf-c was detected in ΔHuR 4T1 cells via Western blot, photographs, and quantification. Mean ± SD. n = 3. * p < 0.05 (t-test), ** p < 0.01 (t-test), ns means no significant difference.

Figure 5

Eltrombopag regulated the function of the HuR protein in 4T1 cells. (A–C) The mRNA stability of Snail, Cox-2, and Vegf-c in 4T1 cells was impacted by eltrombopag. (D) Schematic diagram of the RIP analysis. (E–H) HuR-binding mRNA (Snail, Cox-2, and Vegf-c) and non-HuR-binding mRNA (Gapdh) levels in 4T1 cells influenced by eltrombopag detected with RIP. (I–K) Luciferase activity mediated by Snail-ARE, Cox-2-ARE, and Vegf-c-ARE in 4T1 cells impacted by eltrombopag. Mean ± SD. n = 3. * p < 0.05 (t-test), ** p < 0.01 (t-test), ns means no significant difference.

Figure 6

Eltrombopag regulated lymphangiogenesis-related gene in RAW264.7 cells by targeting HuR. (A) Relative mRNA level of Vegf-c in RAW264.7 cells impacted by eltrombopag. (B,C) Relative protein expression of Vegf-c in RAW264.7 cells impacted by eltrombopag measured with Western blot, representative images and quantitative analyses. (D) mRNA stability of Vegf-c in RAW264.7 cells impacted by eltrombopag. (E) RIP analysis detected the enrichment of HuR-binding mRNA (Vegf-c) and non-HuR-binding mRNA (Gapdh) in RAW264.7 cells influenced by eltrombopag. Mean ± SD. n = 3. * p < 0.05 vs. the LPS-treated group, ## p < 0.01 vs. the control group, ** p < 0.01 vs. the LPS-treated group, ns means no significant difference.

Figure 7

Eltrombopag regulated macrophage-mediated lymphangiogenesis. (A,B) SVEC4-10 cell migration impacted by supernatants of eltrombopag-treated macrophage, representative images (left) and quantitative analyses (right). (C,D) SVEC4-10 cell migration impacted by eltrombopag directly, representative images (left) and quantitative analyses (right). Scale bar: 500 μm. Mean ± SD. n = 3. ** p < 0.01 (t-test), ns means no significant difference.

Figure 8

Eltrombopag suppressed lung metastasis of 4T1-Luc cells in vivo. (A) Schematic diagram of experimental procedures. (B) Body weight data of different groups after treatment. (C) Bioluminescent imaging data of lung metastasis after treatment with eltrombopag or docetaxel. Two mice in the docetaxel group were dead before the end of experiment, which might be due to the toxicity of docetaxel. (D) H&E staining of lung tissues of 4T1-Luc mice. Scale bar: 1000 μm. (E) SNAIL, COX-2, and VEGF-C expression in lung tissues of different groups with IHC method (Scale bar: 50 μm). Mean ± SD. n = 5. * p < 0.05 (t-test).

Figure 9

Eltrombopag suppressed lymph node metastasis of 4T1-Luc cells in vivo. (A) Experimental timeline of 4T1-Luc lymph node metastasis model. (B) Representative images of popliteal lymph node in a mouse model. (C) The appearance and weight of popliteal lymph node. (D) Body weight of tumor-bearing mice. (E) Bioluminescent data analysis in primary tumor site after treatment with eltrombopag. (F,G) Bioluminescent images and data analysis of popliteal lymph node metastasis after treatment with eltrombopag. (H) Lymphangiogenesis analysis in different sections of lymph node, lymph-vessel analysis IHC using antibodies of LYVE-1 (Scale bar: 100 μm). (I) Sections of popliteal lymph node in different groups, submitted to IHC using antibodies of VEGF-C (Scale bar: 100 μm). (J) Representative images of H&E-stained lymph node sections (Scale bar: 1000 μm). LN: lymph node. Mean ± SD. n = 5. * p < 0.05 (t-test).