The atypical ubiquitin ligase RNF31 stabilizes estrogen receptor α and modulates estrogen-stimulated breast cancer cell proliferation

Abstract

Estrogen receptor α (ERα) is initially expressed in the majority of breast cancers and promotes estrogen-dependent cancer progression by regulating the transcription of genes linked to cell proliferation. ERα status is of clinical importance, as ERα-positive breast cancers can be successfully treated by adjuvant therapy with antiestrogens or aromatase inhibitors. Complications arise from the frequent development of drug resistance that might be caused by multiple alterations, including components of ERα signaling, during tumor progression and metastasis. Therefore, insights into the molecular mechanisms that control ERα expression and stability are of utmost importance to improve breast cancer diagnostics and therapeutics. Here we report that the atypical E3 ubiquitin ligase RNF31 stabilizes ERα and facilitates ERα-stimulated proliferation in breast cancer cell lines. We show that depletion of RNF31 decreases the number of cells in the S phase and reduces the levels of ERα and its downstream target genes, including cyclin D1 and c-myc. Analysis of data from clinical samples confirms correlation between RNF31 expression and the expression of ERα target genes. Immunoprecipitation indicates that RNF31 associates with ERα and increases its stability and mono-ubiquitination, dependent on the ubiquitin ligase activity of RNF31. Our data suggest that association of RNF31 and ERα occurs mainly in the cytosol, consistent with the lack of RNF31 recruitment to ERα-occupied promoters. In conclusion, our study establishes a non-genomic mechanism by which RNF31 via stabilizing ERα levels controls the transcription of estrogen-dependent genes linked to breast cancer cell proliferation.

Figure 1

RNF31 depletion inhibits cell proliferation and increases G1 arrest in MCF-7 cells. (a) MCF-7 cells were transfected with siRNF31 or siControl and knockdown efficacy was determined by western blot analysis for RNF31, using GAPDH as internal standard for control (Left panel), and qPCR (Right panel). (b) The WST-1 assay was used to determine the cellular metabolic activity at indicated time points after transfection. Cells are treated for indicated times with 10 nM E2 or vehicle. Experiments were done in triplicates. *P<0.05; **P<0.01 for siRNF31 E2 versus siControl E2. (c) RNF31 knockdown induces arrest in the G1 phase and inhibits the E2-mediated progression into the S phase. The effects of RNF31 knockdown were compared with the effects of ERα knockdown in MCF-7 cells. Cells are treated for 24 h with 10 nM E2 or vehicle. The proportion of cells in each phase was measured by fluorescent-activated cell sorting (FACS). Experiments were done in triplicates. *P<0.05; **P<0.01; ***P<0.001 for siRNF31 versus siControl. All values are mean±s.d. (n=3). (d) Overexpression of ERα partially reverses the reduced number of cells in the S phase following RNF31 depletion. MCF-7 cells were cultured in 10% fetal bovine serum (FBS). The proportion of cells in each phase was measured by FACS. Experiments were done in triplicates. *P<0.05; **P<0.01 for siRNF31 versus siControl; siRNF31 versus siRNF31 plus ERα overexpression. All values are mean±s.d. (n=3).

Figure 2

RNF31 depletion decreases ERα protein levels and ERα signaling. (a) RNF31 depletion reduces ERα protein levels. MCF-7 cells were transfected with siRNF31 or siControl and treated with 10 nM E2 or vehicle for 72 h. ERα and RNF31 levels were determined by western blot analysis. GAPDH was used as internal control. (b) RNF31 depletion or overexpression affects ERα-dependent expression of an ERE-luciferase reporter gene. MCF-7 cells were transfected with siRNF31 or siControl or with plasmids expressing Myc-tagged RNF31 or Myc-tag vector alone or together with the ERE reporter plasmid. Subsequently, cells were treated with 10 nM E2 or vehicle. Luciferase activity was measured 48 h after transfection. Shown are data from triplicate measurements. ***P<0.001 for siRNF31 versus siControl and RNF31 overexpression versus control. (c) RNF31 depletion reduces the expression of endogenous ERα target genes. MCF-7 cells were transfected with siRNF31 or siControl. 48 h after transfection, cells were treated with 10 nM E2 or vehicle for 6 h. The expression levels of RNF31 and of endogenous ERα target genes (ADORA1, pS2, cyclinD1) were determined by qPCR from triplicate experiments. **P<0.01; ***P<0.001 for siRNF31 versus siControl. (d) RNF31 depletion decreases ERα recruitment to endogenous target gene promoters. MCF-7 cells were transfected with siRNF31 or siControl. Forty-eight hours post-transfection, cells were treated with 10 nM E2 or vehicle for 30 min and chromatin immunoprecipitation (ChIP) assays were performed with ERα antibody or rabbit immunoglobulin G (IgG) and quantified by qPCR. **P<0.01; ***P<0.001 for siRNF31 versus siControl. (e) Heat map of ERα-regulated genes changed by RNF31 depletion in MCF-7 cells. P<0.001 and fold change >2 was set as cutoff to derive regulated genes. All values are mean±s.d. (n=3).

Figure 3

RNF31 is highly expressed and is correlated to ERα target genes in tumour samples. (a) RNF31 is highly expressed in breast tumors. The expression of RNF31 was determined for 72 breast tumors and 37 adjacent breast tissues using qPCR. The Student's t-test was used for comparison. Whiskers, 10th and 90th percentile; box boundaries, 75th and 25th percentile; line within box, median. Dots above the boxes show sample maximum values. asterisk (*) indicates outliers. (b) Correlation between expression of RNF31 and expression of the classical positively regulated ERα target genes GREB1 and pS2 (TFF1). Expression levels are derived from the TCGA mRNA database, test statistics from Pearson product-moment correlation. (c) Heat map showing correlation between RNF31 mRNA levels and mRNA levels of known ERα target genes. Included ERα target genes correspond to genes displaying significant correlation with RNF31 in Supplementary Table S2. RNA expression data are from the TCGA RNA-sequencing database.

Figure 4

RNF31 associates with ERα and increases its stability. (a) Co-IP assays reveal associations between endogenous RNF31 and ERα in MCF-7 cells. (b) Over-expression of RNF31 increases endogenous ERα protein levels in MCF-7 cells. MCF −7 cells were transfected with plasmids expressing Myc-tagged RNF31 or the Myc-tag alone. Forty-eight hours after transfection, whole-protein extracts were prepared and subjected to western blot analysis. The levels of RNF31, ERα and the internal control GAPDH were determined by western blot analysis. The predicted molecular weights are indicated. (c) Depletion of RNF31 decreases endogenous ERα protein stability in MCF-7 cells. Cells were transfected with siRNF31 or siControl. After 72 h, cells were treated with 100 μmol/l cycloheximide for different times as indicated before whole protein extraction. The levels of ERα and the internal control GAPDH were determined by western blot analysis. The experiments were done in triplicates. Image J was applied to quantify ERα and GAPDH density. (d) Depletion of RNF31 decreases endogenous ERα protein stability in MCF-7 cells in the presence of E2. Cells were transfected with siRNF31 or sicontrol. After 72 h, cells were treated with 10 nM E2 for 24 h. Then cells were treated with 100 μmol/l cycloheximide for different times as indicated before whole protein extraction. The levels of RNF31, ERα and the internal control GAPDH were determined by western blot analysis. The experiments were done with triplicates. Image J was applied to quantify the ERα and GAPDH density. (e) Overexpression of RNF31 does not further increase the stability of ERα in the presence of the proteasome inhibitor MG132. HEK-293 cell were transfected with plasmids expressing Myc-tagged RNF31 or the Myc-tag alone. Forty-eight hours post transfection, cells were treated with 100 nmol/l MG132 or vehicle. Cells were harvested 6 h after MGM132 treatment and whole protein extracts were prepared. The levels of RNF31, ERα and the internal control GAPDH were determined by western blot analysis.

Figure 5

The RNF31 RBR domain is required for association with and stabilization of ERα. (a) RNF31 domain structure and deletion mutants used in this study (full-length, ΔUBA and ΔRBR). ZnF_RBZ, putative zinc-finger ubiquitin-binding domain; UBA, ubiquitin-associated domain, mediates interaction with RBCK1/LUBAC; RING-IBR-RING (RBR), atypical E3 ubiquitin ligase domain. (b) RNF31 association with ERα requires the RBR domain. HEK-293 cells were transfected with ERα together with plasmids expressing Myc-tagged full-length RNF31, variants deleting the UBA or RBR domains, respectively, or the Myc-tag alone. Cells were harvested 24 h after transfection and whole-cell extracts were prepared for co-IP. The predicted molecular weights of the RNF31 derivatives and ERα are indicated. (c) The RBR domain is necessary for the RNF31-mediated increase of endogenous ERα protein levels. MCF-7 cells were transfected with plasmids expressing Myc-tagged RNF31 derivatives or the Myc-tag alone, as indicated. 48 h after transfection, whole-cell extracts were prepared and levels of ERα protein assayed by western blot analysis. The predicted molecular weights of RNF31 variants, ERα and the loading control GAPDH are indicated. (d) The RBR domain of RNF31 is required to increase ERα signaling. MCF-7 cells were transfected with plasmids expressing Myc-tagged RNF31 derivatives, or the Myc-tag alone, as indicated, along with an ERE-luciferase reporter plasmid. Luciferase activity was measured 48 h after transfection and calculated from experiments performed in triplicates. Data are shown as mean±s.d. (n=3). ***P<0.001 for wt-RNF31 versus RNF31 deletion domains/empty vector; NA, P>0.05 for wt-RNF31 versus RNF31 deletion domains.

Figure 6

RNF31 triggers ERα mono-ubiquitination. (a) Detection of a potentially mono-ubiquitinated form of ERα upon RNF31 overexpression. HEK-293 cells were transfected with ERα together with plasmids expressing Myc-tagged RNF31 or the Myc-tag alone. Forty-eight hours after transfection, whole-cell extracts were prepared and levels of ERα protein assayed by western blot analysis. The predicted molecular weights of RNF31, ERα, mono-ubiquitinated ERα and of the internal control GAPDH are indicated. (b) Detection of endogenous mono-ubiquitinated ERα upon RNF31 depletion. MCF-7 cells were transfected with siRNF31 or siControl. Forty-eight hours after transfection, whole-cell extracts were prepared and levels of ERα protein assayed by western blot analysis. The predicted molecular weights of RNF31, ERα, mono-ubiquitinated ERα and the internal control GAPDH are indicated. (c) Deletion of the RNF31 RBR domain abolishes the potentially mono-ubiquitinated form of ERα. HEK-293 cells were transfected with ERα together with plasmids expressing Myc-tagged full-length RNF31 derivatives or the Myc-tag alone. Forty-eight hours post transfection, cell extracts were prepared and ERα forms were detected by western blot analysis. The predicted molecular weights of RNF31, ERα, mono-ubiquitinated ERα and of the internal control GAPDH are indicated. (d) Direct evidence for ERα mono-ubiquitination. Immunoprecipitation of ubiquitinated proteins from MCF-7 cell extracts upon overexpression of RNF31. Ubiquitinated ERα species were detected by western blots using anti-ERα, identifying a prominent 75 kDa mono-ubiquitinated ER form. (e) ERα mono-ubiquitination requires the RNF31 RBR domain. Plasmids expressing Myc-tagged RNF31 derivatives were transfected into HEK-293 cells together with the ERα expression plasmid. Whole-cell extracts were subjected to immunoprecipitation of ERα and subsequently analyzed for ubiquitinated ERα forms by western blot analysis using anti-ubiquitin. The predicted molecular weight of mono-ubiquitinated ERα is indicated.

Figure 7

The ubiquitin ligase activity of RNF31 is required for ERα stabilization, signaling and mono-ubiquitination. (a) The mono-ubiquitination function of RNF31 is required for the RNF31-mediated increase of endogenous ERα protein levels. MCF-7 cells were transfected with plasmids expressing Myc-tagged RNF31, Myc-tagged RNF31 R1/2M, in which cysteine residues responsible for the transfer of ubiquitin to substrates have been mutated or the Myc-tag alone, as indicated. Forty-eight hours after transfection, whole-cell extracts were prepared and levels of ERα protein assayed by western blot analysis. The predicted molecular weights of RNF31 variants, ERα and the loading control GAPDH are indicated. (b) The mono-ubiquitination function of RNF31 is required to increase ERα signaling. MCF-7 cells were transfected with plasmids expressing Myc-tagged RNF31, Myc-tagged RNF31 R1/2M or the Myc-tag alone, as indicated, along with an ERE-luciferase reporter plasmid. Luciferase activity was measured 48 h after transfection and calculated from experiments performed in triplicates. Data are shown as mean±s.d. (n=3). ***P<0.001 for wt-RNF31 versus RNF31 R1/2M or Myc-tag. RNF31 deletion domains/empty vector; NA, P>0.05 for wt-RNF31 versus RNF31 deletion domains. (c, d) Mutation of the RNF31 ubiquitin ligase domain abolishes mono-ubiquitination of ERα. (c) HEK-293 cells were transfected with ERα together with plasmids expressing Myc-tagged RNF31, RNF31 R1/2M or the Myc-tag alone. Forty-eight hours post transfection, cell extracts were prepared and ERα was detected by western blot analysis. The predicted molecular weights of RNF31, ERα, mono-ubiquitinated ERα and the internal control GAPDH are indicated. (d) HEK-293 cells were transfected with ERα together with plasmids expressing Myc-tagged full-length RNF31, RNF31 R1/2M or the Myc-tag alone. After 48 h, whole-cell extracts were subjected to immunoprecipitation of ERα and subsequently analyzed for ubiquitinated ERα forms by western blot analysis using anti-ubiquitin antibodies. The predicted molecular weights of RNF31, IgG, ERα and mono-ubiquitinated ERα are indicated.

Figure 8

Intracellular localization analysis and model of crosstalk between RNF31 and ERα signaling in breast cancer cells. (a) MCF-7 cells were treated with 10 nM E2 or vehicle for 30 min before fixation. Intracellular localization of RNF31 (red) and ERα (green) was determined by immunofluorescence staining. Nuclei (blue) were stained with 4',6-diamidino-2-phenylindole (DAPI). Shown are representative images. For quantitative analysis see Supplementary Figure S5. (b) Co-IP assay reveals the interaction between RNF31 and ERα in the cytoplasm. The Subcellular Protein Fractionation Kit (Thermo Scientific, 78840) was used for extraction of cytoplasmic and nuclear proteins from MCF-7 cells. Vinculin and Histone 3 were used to identify the quality of cytoplasmic and nuclear fractions, respectively (left panel). Co-IP assays were performed with RNF31 antibody for precipitation and ERα antibody for detection (right panel). (c) Hypothetical model for the functional interplay of RNF31 with ERα signaling in breast cancer cells. RNF31 associates with ERα predominantly in the cytoplasm and promotes mono-ubiquitination of ERα, thereby counteracting proteasome-mediated degradation. The so elevated ERα protein levels cause increased genomic ERα signaling, for example, transcription of E2-target genes linked to the proliferation of breast cancer cells. RNF31, as part of the LUBAC ubiquitin ligase complex, has an established independent function in NFκB signaling. To which extent the two signaling pathways also crosstalk remains to be investigated.