YTHDF1 promotes breast cancer progression by facilitating FOXM1 translation in an m6A-dependent manner

Abstract

Background: N6-methyladenosine (m⁶A) is the most common post-transcriptional modification at the RNA level. However, the exact molecular mechanisms of m6A epigenetic regulation in breast cancer remain largely unknown and need to be fully elucidated.

Methods: The integrating bioinformatics analyses were used to screen clinical relevance and dysregulated m6A "reader" protein YTHDF1 in breast cancer from TCGA databases, which was further validated in a cohort of clinical specimens. Furthermore, functional experiments such as the CCK-8 assay, EdU assay, wound healing assay, transwell invasion assay and cell cycle assay were used to determine the biological role of YTHDF1 in breast cancer. RIP, m6A-IP, and CLIP assays were used to find the target of YTHDF1 and further verification by RT-qPCR, western blot, polysome profiling assay. The protein-protein interaction between YTHDF1 and FOXM1 was detected via co-immunoprecipitation.

Results: Our study showed that YTHDF1 was overexpressed in breast cancer cells and clinical tissues specimens. At the same time, the high expression level of YTHDF1 was positively correlated with tumor size, lymph node invasion, and distant metastasis in breast cancer patients. YTHDF1 depletion repressed the proliferation, invasion and epithelial-mesenchymal transformation (EMT) and induced G0/G1 phase cell cycle arrest of breast cancer cells in vitro and in vivo. We also demonstrated that FOXM1 is a target of YTHDF1. Through recognizing and binding to the m6A-modified mRNA of FOXM1, YTHDF1 accelerated the translation process of FOXM1 and promoted breast cancer metastasis. Whereas overexpression of FOXM1 in breast cancer cells partially counteracted the tumor suppressed effects caused by YTHDF1 silence, which further verified the regulatory relationship between YTHDF1 and FOXM1.

Conclusion: Our study reveals a novel YTHDF1/FOXM1 regulatory pathway that contributes to metastasis and progression of breast cancer, suggesting that YTHDF1 might be applied as a potential biomarker and therapeutic target. That also advances our understanding of the tumorigenesis for breast cancer from m6A epigenetic regulation.

Keywords: Breast cancer; EMT; FOXM1; YTHDF1; m6A.

Fig. 1

YTHDF1 was highly expressed in breast cancer samples and cells. A YTHDF1 expression in breast cancer tissues and normal tissues from TCGA-BRCA dataset. Relative YTHDF1 mRNA expression in B collected breast cancer tissues and adjacent normal tissues samples, C MCF-10A and MCF-7 cells from GSE71862 dataset and D normal human breast cell line MCF-10A and breast cancer cell line MCF-7, MDA-MB-231, SKBR3, T47D and BT-549. Kaplan–Meier curves of OS in the TCGA dataset E and GSE29272 dataset F of all breast cancer patients with high and low YTHDF1 expression. G Immunohistochemical staining of YTHDF1 in a breast cancer sample. YTHDF1 protein expression in H breast cancer tissues and adjacent normal tissues and I normal human breast cell line MCF-10A and breast cancer cell line MCF-7, MDA-MB-231, SKBR3, T47D and BT-549. All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 2

YTHDF1 knockdown inhibits the growth ability and arrests the cell cycle of breast cancer cells. A Relative YTHDF1 mRNA expression levels after transducing scramble shRNA (sh-NC), sh-YTHDF1#1, sh-YTHDF1#2 or sh-YTHDF1#3 in MDA-MB-231 or MCF-7 cells. B CCK-8 assay, C EdU proliferation assay and D cell cycle analysis in MDB-MB-231 or MCF-7 cells transduced with sh-NC, sh-YTHDF1#1 or sh-YTHDF1#2. All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 3

Silence of YTHDF1 restrain cell migration, invasion ability and EMT. A Transwell invasion assay, B wound healing assay and C western blot analysis for EMT-related proteins in MDA-MB-231 or MCF-7 cells transduced with sh-NC, sh-YTHDF1#1 or sh-YTHDF1#2 (sh#1:sh-YTHDF1#1; sh#2:sh-YTHDF1#2). All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 4

Depletion of YTHDF1 inhibits the growth of breast cancer cells in vivo. NOD/SCID mice (5 per group) were transplanted with MCF-7 cells (2 * 10⁶) transduced with sh-NC or sh-YTHDF1#2. A Tumor volumes in sh-NC or sh-YTHDF1#2 group at indicated days. B Representative images and C the weight of tumors in sh-NC or sh-YTHDF1#2 group at day 49. D The expression of EMT-related proteins and E cell cycle analysis in xenograft that transduced with sh-NC or sh-YTHDF1. F The number of visible lung metastasis nodules and representative images of lung HE staining slices. (T stands for metastasis and N for normal lung tissues). All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 5

FOXM1 is a direct target of YTHDF1. Normalized expression of target genes detected by RT-qPCR after A RIP, B m6A-IP or C YTHDF1-CLIP assay. D Correlation between FOXM1 protein expression and YHTDF1 mRNA expression in TCGA-BRCA dataset. E FOXM1 protein expression in breast cancer tissue and normal tissue from The Human Protein Atlas. F The protein and expression of YTHDF1 and FOXM1 in MDA-MB-231 and MCF-7 cells transduced with sh-NC, sh-YTHDF1#1 or sh-YTHDF1#2. G The correlation between YTHDF1 and FOXM1 in protein levels by immunohistochemical analysis in 9 samples. H The mRNA expression of YTHDF1 and FOXM1 in MDA-MB-231 and MCF-7 cells transduced with sh-NC, sh-YTHDF1#1 or sh-YTHDF1#2. I Polysome profiling assay in MCF-7 cells transduced with sh-NC or sh-YTHDF1#2 (Left). FOXM1 and 18 s rRNA expression in different polysome fractions (Right). All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 6

YTHDF1 regulates FOXM1 expression in an m⁶A-dependent manner. A The analysis of qPCR for FOXM1 expression in RIP assay (anti-Flag) in MDB-MA-231 and MCF-7 cells. B FOXM1 protein expression detected by western blot in MDB-MA-231 and MCF-7 cells transfected with YTHDF1-WT or YTHDF1-MUT plasmids. C Schematic image of FOXM1-WT plasmids, and FOXM1-MUT plasmids containing m⁶A motif mutations in the CDS region. D Protein expression of HA tagged FOXM1 in MDA-MB-231 and MCF-7 cells co-transfected with empty vector, YTHDF1-WT or YTHDF1-MUT, and FOXM1-WT or FOXM1-MUT plasmids. All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 7

The anti-tumor effect of YTHDF1 silence could be abrogated by FOXM1 overexpression. A Western blot analysis for FOXM1 and YTHDF1 in MDB-MA-231 and MCF-7 cells with sh-YTHDF1-2 (shYTH) or sh-YTHDF1#2 + pcDNA-FOXM1 plasmid (shYTH + FOXM1) B CCK-8 assay, C Transwell invasion assay, D wound healing assay and E Western blot assay for EMT-related proteins in MDB-MA-231 and MCF-7 cells with shNC, sh-YTHDF1#2 (shYTH) or sh-YTHDF1#2 + FOXM1 (shYTH + FOXM1). All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated

Fig. 8

YTHDF1 overexpression could not reverse the anti-tumor effects caused by FOXM1 knockdown in MCF-7 cells. A CCK-8 assays in shNC, shYTHDF1#2 + FOXM1-WT and sh-YTHDF1#2 + FOXM1-MUT MCF-7 cells. B Western blots for EMT-related proteins and C Transwell invasion assays in shNC, shNC + Vector, sh-YTHDF1#2 + FOXM1-MUT and sh-YTHDF1#2 + FOXM1-WT MCF-7 cells. D FOXM1 protein expression in shNC, sh-FOXM1#1 and sh-FOXM1#2 transduced cells. E Western blots for EMT-related proteins, F CCK-8 assays and G Transwell invasion assays in shNC + Vector, sh-YTHDF1#2 + FOXM1-MUT and sh-YTHDF1#2 + FOXM1-WT MCF-7 cells. All data were presented as means ± SD of at least three independent repetitions. Values are significant at *P < 0.05, **P < 0.01, ***P < 0.001 as indicated