Interaction between BEND5 and RBPJ suppresses breast cancer growth and metastasis via inhibiting Notch signaling

Abstract

High frequent metastasis is the major cause of breast cancer (BC) mortality among women. However, the molecular mechanisms underlying BC metastasis remain largely unknown. Here, we identified six hub BC metastasis driver genes (BEND5, HSD11B1, NEDD9, SAA2, SH2D2A and TNFSF4) through bioinformatics analysis, among which BEND5 is the most significant gene. Low BEND5 expression predicted advanced stage and shorter overall survival in BC patients. Functional experiments showed that BEND5 could suppress BC growth and metastasis in vitro and in vivo. Mechanistically, BEND5 inhibits Notch signaling via directly interacting with transcription factor RBPJ/CSL. BEN domain of BEND5 interacts with the N-terminal domain (NTD) domain of RBPJ, thus preventing mastermind like transcriptional coactivator (MAML) from forming a transcription activation complex with RBPJ. Our study provides a novel insight into regulatory mechanisms underlying Notch signaling and suggests that BEND5 may become a promising target for BC therapy.

Keywords: BEND5; Breast cancer; Notch signaling; bioinformatics; metastasis.

Figure 1

Identification of metastasis driver genes (MDGs) and prognostic signature in BC. (A) The volcano plot showing differentially expressed genes (DEGs) in MDA-MB-231 and its lung metastatic subpopulations from GSE138122 dataset. (B) The volcano plot showing DEGs in 112 paired BC tissues and adjacent tissues from TCGA-BC dataset. (C) The common differentially up/down-expressed genes identified as MDGs. (D) Top 10 enriched GO-BP pathways based on the MDGs. (E) The forest plot exhibiting MDGs that significantly correlates with overall survival (OS) based on univariable Cox regression analysis. (F) The distribution of risk score and survival status of BC patients from TCGA-BC dataset. (G) ROC curve plotted for the prognostic model with 1-, 3- and 5-year OS in BC patients. (H) The comprehensive nomogram for 1-, 3- and 5-year overall survival prediction of BC patients.

Figure 2

The validation of BEND5 in TCGA-BC dataset. (A-D) The associations between BEND5 mRNA expression and clinicopathological features in BC patients from TCGA-BC dataset. (E and F) The association between BEND5 gene expression and overall survival (E) and disease-free survival (F) in BC patients from TCGA-BC dataset.

Figure 3

BEND5 suppresses proliferation, migration and invasion in MDA-MB-231 cells. (A) MDA-MB-231 cells were transfected with FLAG-tagged BEND5 or empty vector and cultured for a specified time. CCK8 assays were used to detect cell numbers, and immunoblot was used to detect the expression of BEND5 in MDA-MB-231 cells. β-actin was used as a loading control. (B) Colony formation assays for MDA-MB-231 cells transfected as in (A). (C and D) Wound-healing assays and transwell assays for MDA-MB-231 cells transfected as in (A). (E) MDA-MB-231 cells were transfected with control shRNA, BEND5 shRNA or BEND5 shRNA plus shRNA-resistant BEND5 (BEND5-R). CCK8 assay was used to detect cell numbers, and immunoblot was used to detect the expression of BEND5 in MDA-MB-231 cells. (F) Colony formation assays for MDA-MB-231 cells transfected as in (E). (G and H) Wound-healing assays and transwell assays for MDA-MB-231 cells transfected as in (E). Data shown are mean ± SD of triplicate measurements with similar results (*P < 0.05, **P < 0.01 versus empty vector).

Figure 4

BEND5 eliminates Notch signaling-induced BC cell proliferation, migration and invasion. (A) GSEA was conducted to predict the function of BEND5 in BC. (B) MDA-MB-231 cells were transfected with empty vector or FLAG-tagged BEND5 and treated with/without Notch signaling activator DLL4 (10 ng/ml). Immunoblot was used to detect Notch pathway downstream targets and EMT-related proteins. (C and D) CCK8 assays and colony formation assays for MDA-MB-231 cells transfected and treated as in (B). (E and F) Wound-healing assays and transwell assays for MDA-MB-231 cells transfected and treated as in (B). Data shown are mean ± SD of triplicate measurements with similar results (*P < 0.05, **P < 0.01).

Figure 5

BEND5 inhibits Notch signaling via binding to RBPJ. (A) MDA-MB-231 or ZR75-1 cells were immunoprecipitated with anti-BEND5 or normal IgG, and the precipitates were analyzed by immunoblot with the indicated antibodies. (B) HEK293T cells were co-transfected with MYC-RBPJ and FLAG-BEND5 or FLAG-BEND5 △BEN as indicated. Cell lysates were immunoprecipitated with anti-FLAG, followed by immunoblot with anti-MYC. (C) HEK293T cells were co-transfected with FLAG-BEND5 and MYC-RBPJ, MYC-RBPJ △CTD, MYC-RBPJ △BTD, or MYC-RBPJ △NTD. Cell lysates were immunoprecipitated with anti-MYC, followed by immunoblot with anti-FLAG. (D) Co-IP analysis of MDA-MB-231 cells and ZR75-1 cells transfected with FLAG-tagged BEND5 and MYC-tagged RBPJ. Cell lysates were immunoprecipitated with anti-MYC, followed by immunoblot with the indicated antibodies. (E) Immunoblot was used to detect Notch pathway downstream targets and EMT-related proteins in MDA-MB-231 and ZR75-1 cells transfected with empty vector, FLAG-BEND5, or FLAG-BEND5 △BEN. Data shown are mean ± SD of triplicate measurements with similar results (*P < 0.05, **P < 0.01).

Figure 6

Knockdown of BEND5 promotes BC tumor growth and metastasis in mice. (A) MDA-MB-231 cells stably transfected with BEND5 shRNA or control shRNA were injected subcutaneously in the right flank of nude mice, and tumor volume was measured with vernier-caliper at the indicated times. (B) Immunoblot analysis of representative excised tumor tissues from (A). (C) Representative bioluminescence image at 30 days of nude mice injected by tail vein with MDA-MB-231 cells expressing firefly luciferase and the indicated constructs (n = 6). The luminescence signal is represented by an overlaid false-color image with the signal intensity indicated by the scale (right panel). (D) Representative lung tissues and H&E-stained sections of the lung tissues from (C). The number of tumor nodules are shown (right panel). Data are shown as mean ± SD (n = 6) (*P < 0.05, **P < 0.01 versus control shRNA). (E) MDA-MB-231 cells stably transfected with BEND5 shRNA or control shRNA were injected into mammary fat pad on the right side of nude mice (n = 6). Representative bioluminescence images are shown one month after injection. Red arrow indicates tumor growth and green arrow tumor metastasis. Tumor growth was compared between MDA-MB-231 cells expressing BEND5 shRNA and control shRNA (**P < 0.01 versus control shRNA). (F) A proposed model for BEND5 modulation of BC growth and metastasis. BEND5 inhibits Notch signaling by disrupting MAML-RBPJ interaction via competing with MAML for binding RBPJ.