SET8 promotes epithelial-mesenchymal transition and confers TWIST dual transcriptional activities

Abstract

SET8 is implicated in transcriptional regulation, heterochromatin formation, genomic stability, cell-cycle progression, and development. As such, it is predicted that SET8 might be involved in the development and progression of tumour. However, whether and how SET8 might be implicated in tumourigenesis is currently unknown. Here, we report that SET8 is physically associated with TWIST, a master regulator of epithelial-mesenchymal transition (EMT). We demonstrated that SET8 and TWIST are functionally interdependent in promoting EMT and enhancing the invasive potential of breast cancer cells in vitro and in vivo. We showed that SET8 acts as a dual epigenetic modifier on the promoters of the TWIST target genes E-cadherin and N-cadherin via its H4K20 monomethylation activity. Significantly, in breast carcinoma samples, SET8 expression is positively correlated with metastasis and the expression of TWIST and N-cadherin and negatively correlated with E-cadherin. Together, our experiments revealed a novel role for SET8 in tumour invasion and metastasis and provide a molecular mechanism underlying TWIST-promoted EMT, suggesting SET8 as a potential target for intervention of the metastasis of breast cancer.

Figure 1

SET8 is physically associated with TWIST. (A) Mass spectrometry analysis of TWIST-associated proteins. Whole cellular extracts from FLAG–TWIST stably expressing MCF-7 cells were subjected to affinity purification with anti-FLAG antibody that was immobilized on the agarose beads. The purified protein complex was resolved on SDS–PAGE and silver stained, and the bands were retrieved and analysed by mass spectrometry. (B) Western blotting examination of SET8 protein level and its H4K20 monomethyltransferase activity in the FLAG–TWIST-purified fractions using the indicated antibodies. (C) Co-fractionation of TWIST and SET8 complex by FPLC. MCF-7 cell nuclear proteins were extracted, concentrated, and then 6 mg of nuclear extract was fractionated on Superose 6 size exclusion columns. Chromatographic elution profiles and immunoblotting analysis of the chromatographic fractions are shown. The elution positions of calibration proteins with known molecular masses (kDa) are indicated, and an equal volume from each fraction was analysed. (D) SET8 interacts with TWIST in vivo. Whole cell lysates from FLAG–TWIST stably expressing MCF-7 cells or from MDA-MB-231 cells were prepared and immunoprecipitation was performed with anti-FLAG or anti-TWIST followed by immunoblotting with anti-SET8. (E) SET8 interacts directly with TWIST in vitro. GST pull-down assays were performed with the indicated GST-fused SET8 or SET9 protein and in vitro transcribed/translated TWIST. Coomassie brilliant blue (CBB) staining of the GST-fused proteins was shown. (F) Mapping the interface of SET8 interacting with TWIST by GST pull-down experiments using GST-fused TWIST and in vitro transcribed/translated full-length SET8 or its deletion constructs.

Figure 2

Functional interplay of SET8 and TWIST in promoting EMT of breast cancer cells. (A) SET8 cooperates with TWIST to induce morphological changes in MCF-7 cells. MCF-7 cells were transfected with vector, SET8 or/and TWIST, and the morphological alterations of MCF-7 cells were observed by phase-contrast microscopy. Immunofluorescence staining of epithelial (E-cadherin and β-catenin) and mesenchymal (N-cadherin and fibronectin) markers was visualized by confocal microscopy (Green). DAPI staining was included to visualize the cell nucleus (Blue). (B) Immunoblotting examination of the epithelial and mesenchymal proteins (left) and qPCR analyses of their mRNA expression (right) in MCF-7 cells overexpressing SET8 or/and TWIST. (C) SET8 or TWIST was silenced in vector, TWIST, or SET8 stably transfected MCF-7 cells, and the epithelial and mesenchymal markers were measured by immunoblotting (left) and qPCR (right). Each bar indicates mean±s.d. of three independent experiments. P-values were determined by Student's t-test. ^*P<0.05; ^#P<0.01.

Figure 3

Functional interdependence of SET8 and TWIST in enhancing the invasive potential of breast cancer cells. (A) The endogenous protein or mRNA expression of SET8 and TWIST was examined by western blotting (left) or qPCR (right). (B) MDA-MB-231 cells were transfected with empty vector, TWIST or SET8 expression construct, control siRNA, TWIST or SET8 siRNA. After 48 h of transfection, cells were starved for 18 h before cell invasion assays were performed using Matrigel transwell filters. The invaded cells were stained and counted. The images represent one field under microscopy in each group. Each bar indicates mean±s.d. of a representative experiment performed in triplicate. P-values were determined by Student's t-test. ^*P<0.05; ^#P<0.01.

Figure 4

SET8 cooperates with TWIST to regulate E-cadherin and N-cadherin transcription. (A) Recruitment of SET8 and TWIST on E-cadherin and N-cadherin promoters. ChIP and ChIP/ReIP experiments were performed in MDA-MB-231 cells with indicated antibodies. (B) SET8 is recruited by TWIST to the E-cadherin or N-cadherin promoter. SET8 or TWIST was silenced individually in MDA-MB-231 cells; qChIP assays were performed with indicated antibodies. (C) SET8 or TWIST silencing led to increase or decrease in E-cadherin or N-cadherin mRNA level measured by qPCR. (D) SET8 or TWIST depletion led to an elevation or a reduction in E-cadherin or N-cadherin protein level, respectively, as measured by western blotting. (E) SET8 interplays with TWIST to regulate E-cadherin promoter- or N-cadherin promoter-driven luciferase activity. MDA-MB-231 cells were transfected with promoter luciferase constructs together with SET8 or/and TWIST expression or silencing molecules. Luciferase activities were measured and normalized to those of Renilla. Each bar indicates mean±s.d. of three independent experiments. P-values were determined by Student's t-test. ^*P<0.05; ^#P<0.01.

Figure 5

SET8-directed H4K20me1 is a dual epigenetic mark on the E-cadherin and N-Cadherin promoters. SET8 recruitment on E-cadherin (A) or N-cadherin (B) promoter is associated with the monomethylation status of histone H4K20. After transfection of MDA-MB-231 cells with siRNAs of control, TWIST or SET8 for 72 h, qChIP assays were performed with mono-, di-, and tri-methylated H4K20-specific antibodies with anti-H4 normalized as 100%. (C) MDA-MB-231 cells were treated with control or SET8 siRNAs, and western blotting analysis was performed with antibodies against the indicated proteins. (D) Western blotting was conducted in MDA-MB-231 cells overexpressing the FLAG-tagged vector (control), FLAG–SET8 construct or a dominant-negative FLAG-tagged SET8 R295G point mutant (SET8 R295G). (E) Catalytically inactive SET8 R295G mutant resulted in derepression or deactivation of E-cadherin or N-cadherin transcription, respectively, as measured by qPCR. (F) Catalytically inactive SET8 R295G mutant resulted in derepression or deactivation of E-cadherin or N-cadherin promoter activity. Each bar indicates mean±s.d. of three independent experiments. P-values were determined by Student's t-test. ^*P<0.05.

Figure 6

SET8 cooperates with TWIST to promote breast cancer metastatic potential in orthotopic mouse model. (A) The effect of SET8 gain-of-function or loss-of-function on spontaneous lung metastasis of orthotopic breast cancer cells. (B, C) Cooperation of SET8 and TWIST on spontaneous lung metastasis of orthotopic breast cancer cells. MDA-MB-231-Luc-D3H2LN cells were co-infected with retroviruses and lentiviruses carrying empty, SET8 or/and TWIST expression vector, and control shRNA, SET8 or TWIST shRNA-containing vector. These cells were inoculated into the left abdominal mammary fat pad (2 × 10⁶ cells) of 6-week-old immunocompromised female BALB/c mice. Tumour size was measured on day 42 (mammary tumours, n=10). The presence of circulating tumour cells (intravasation, n=10) was assessed by qPCR of human GAPDH expression relative to murine β2-microglobulin in 1 ml of mouse blood perfusate. Lung metastases were quantified using bioluminescence imaging (n=10) after 6 weeks of initial implantation, and representative in vivo bioluminescent images are shown. (D) Examination of SET8 and/or TWIST overexpression and/or knockdown in the cell used in animal experiments by IB. In all panels, each bar indicates mean±s.d. P-values were determined by Student's t-test. ^*P<0.05; ^#P<0.01.

Figure 7

Correlation of SET8 with the metastasis of human breast cancers. The level of SET8 (A) or TWIST (B) mRNA is positively correlated with breast cancer metastasis. Total RNAs from the sample of breast normal tissue (Normal), non-metastatic breast cancer (NMBC), and metastatic breast cancer (MBC) were extracted and the expression of SET8 and TWIST mRNA was measured by qPCR with GAPDH as the reference gene. (C) The level of SET8 mRNA is negatively correlated with E-cadherin mRNA and positively correlated with N-cadherin mRNA in MBCs. The level of SET8, E-cadherin, and N-cadherin mRNAs was measured by qPCR and the relative level of SET8 mRNA was plotted against that of E-cadherin or N-cadherin. (D) The level of SET8 mRNA is positively correlated with that of TWIST mRNA in MBC samples. Total RNAs from NMBC and MBC samples were extracted, and the expression of SET8 and TWIST was measured by qPCR. The relative level of SET8 mRNA was plotted against that of TWIST. (E) Immunohistochemical analyses of SET8 or TWIST expression in NMBC, MBC (the black square represents invasive cancer cells and the red square indicates cancer embolus in circulation), and MLN samples. Representative images are shown. Scale bar, 50 μm.