Stabilization of IGF2BP1 by USP10 promotes breast cancer metastasis via CPT1A in an m6A-dependent manner

Abstract

Metastasis leads to the vast majority of breast cancer mortality. Increasing evidence has shown that N6-methyladenosine (m6A) modification and its associated regulators play a pivotal role in breast cancer metastasis. Here, we showed that overexpression of the m6A reader IGF2BP1 was clinically correlated with metastasis in breast cancer patients. Moreover, IGF2BP1 promoted distant metastasis in vitro and in vivo. Mechanistically, we first identified USP10 as the IGF2BP1 deubiquitinase. USP10 can bind to, deubiquitinate, and stabilize IGF2BP1, resulting in its higher expression level in breast cancer. Furthermore, by MeRIP-seq and experimental verification, we found that IGF2BP1 directly recognized and bound to the m6A sites on CPT1A mRNA and enhanced its stability, which ultimately mediated IGF2BP1-induced breast cancer metastasis. In clinical samples, USP10 levels correlated with IGF2BP1 and CPT1A levels, and breast cancer patients with high levels of USP10, IGF2BP1, and CPT1A had the worst outcome. Therefore, these findings suggest that the USP10/IGF2BP1/CPT1A axis facilitates breast cancer metastasis, and this axis may be a promising prognostic biomarker and therapeutic target for breast cancer.

Keywords: Breast cancer; CPT1A; IGF2BP1; USP10; m6A; metastasis.

Figure 1

Elevated IGF2BP1 expression resulted in poor prognosis of BC. (A) The mRNA levels of m6A-related enzymes in 112 BC tissues and paired normal breast tissues from TCGA database (n=112, p<0.05, paired t test). (B) The mRNA levels of m6A-related enzymes in BC tissues and paired normal breast tissues from GEO datasets (GSE22820 and GSE86374). (C) Overlap of the differentially expressed genes of interest from TCGA dataset and the GEO datasets (GSE22820 and GSE86374). (D) Kaplan-Meier OS curves based on IGF2BP1, FTO and WTAP mRNA expression using the online bioinformatics tool based on TCGA. (E) RNAs isolated from MDA-MB-231, LM2, BM6, and 1833 cells were used in dot blot assays with an anti-m6A antibody. MB (methylene blue) served as a loading control. (F) Protein levels of m6A-related enzymes were measured in MDA-MB-231, LM2, BM6, and 1833 cells by western blotting. (G) Representative IHC staining images of the BC samples probed with an anti-IGF2BP1 antibody (scale bars=100 µm or 50 µm) (left panel). Calculated IHC staining indices of BC samples (right panel, n=11, p=0.0002, paired t test). (H) Kaplan-Meier DFS and OS analyses of IGF2BP1 expression in patients with BC (n=80, p<0.01, log-rank test). (I) Representative IHC staining images of the primary BC samples with or without metastasis probed with an anti-IGF2BP1 antibody (scale bars=50 µm) (left panel). Calculated IHC staining indices of BC samples (right panel, n=7, p=0.0031, unpaired t test). The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 2

IGF2BP1 promoted breast cancer metastasis in vitro and in vivo. (A) The protein level of IGF2BP1 in LM2 cells with stable downregulation of IGF2BP1 measured by western blotting (upper panel). The grey values of the images were calculated using ImageJ (bottom panel). (B) Cell migration and invasion assays of LM2 cells deficient in IGF2BP1. Representative images and quantification of the cell migration and invasion assays are shown. (C) The protein levels of HA-tag and IGF2BP1 in MDA-MB-231 cells overexpressing IGF2BP1 measured by western blotting (upper panel). The grey values of the images were calculated using ImageJ (bottom panel). (D) Migration and invasion of MDA-MB-231 cells overexpressing IGF2BP1. Representative images and quantification of cell migration and invasion are shown. (E-F) IGF2BP1 downregulation led to markedly decreased BC lung metastasis in female nude mice (total n=9). Representative bioluminescence images and quantification of bioluminescence (E) of mice and HE staining (scale bars=275 µm) (F) of pulmonary metastases 8 weeks after tail vein injection of IGF2BP1 downregulated or vector control LM2 cells. The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 3

USP10 bound to and stabilized IGF2BP1 by deubiquitination. (A-C) USP10 specifically bound to IGF2BP1. HEK293T cells were transfected with FLAG-tagged DUBs and HA-tagged IGF2BP1 plasmids for 72 h, immunoprecipitated with FLAG or HA beads along with IgG control beads and then subjected to immunoblotting (IB) with the indicated antibodies. (D) Endogenous USP10 bound to IGF2BP1 in LM2 cells, MDA-MB-468 cells and HCC1806 cells. Cell lysates from LM2 cells, MDA-MB-468 cells and HCC1806 cells were immunoprecipitated with an anti-IGF2BP1 antibody along with a normal IgG control, followed by IB with the indicated antibodies. (E) Immunofluorescence (IF) assays were used to detect the expression and location of USP10 and IGF2BP1 in LM2 cells (scale bars=10 µm). (F-G) IGF2BP1 protein levels were detected by overexpressing or downregulating USP10 in MDA-MB-231 and LM2 cells. Cells were transfected with USP10 plasmids or siRNAs targeting USP10 for 48 h (upper panel). Densitometry quantification was carried out by ImageJ (bottom panel). (H) LM2 cells were treated with 10 μM Spautin-1 for 6 h. The protein levels of USP10 and IGF2BP1 were measured by western blotting (upper panel). Grey value quantification was carried out by ImageJ (bottom panel). (I) LM2 cells, MDA-MB-468 cells and HCC1806 cells were treated with 10 μM MG132 for 6 h, and the protein levels of IGF2BP1 were measured by western blotting. (J) USP10-overexpressing MDA-MB-231 cells and control cells were treated with 100 μg/ml CHX for the indicated time periods. The protein levels of IGF2BP1 were measured by western blotting (upper panel). The decay curves are shown based on quantification of the grey values by ImageJ (bottom panel). (K) HEK293T cells were transfected with Flag-tagged USP10, Flag-tagged NC, HA-tagged IGF2BP1, and HIS-tagged UB plasmids for 72 h, followed by treatment with 10 μM MG132 for 6 h. IGF2BP1 protein was pulled down by HA beads and then subjected to IB with the indicated Abs. (L) LM2 cells were treated with 10 μM Spautin-1 for 6 h, followed by 10 μM MG132 for 6 h. IGF2BP1 protein was pulled down and then subjected to IB with the indicated Abs. (M-N) USP10 was overexpressed or downregulated in MDA-MB-468 cells and HCC1806 cells followed by treatment with 10 μM MG132 for 6 h. IGF2BP1 protein was pulled down and then subjected to IB with the indicated Abs. The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 4

The USP10/IGF2BP1 axis promoted BC metastasis in vitro and in vivo. (A) USP10 mRNA expression in 112 BC tissues and paired normal breast tissues was analysed using TCGA data (n=112, p<0.001, paired t test). (B) Kaplan-Meier OS curves were analysed based on USP10 mRNA expression from TCGA data using the online bioinformatics tool. (C-D) Kaplan-Meier OS and DFS analyses of IGF2BP1 expression in patients with BC were performed (n=80, p<0.01, log-rank test). (E) The correlation between IGF2BP1 expression and USP10 expression in BC patients was analysed (n=80, p<0.001, Spearman test). (F-G) USP10 knockdown using targeting siRNAs or Spautin-1 effectively inhibited the migration and invasion abilities of LM2 cells or the migration ability of both LM2 and 1833 cells (upper panel). A graphical representation of the quantified data is shown (bottom panel). (H) IGF2BP1 overexpression rescued the reduced migration ability of LM2 cells caused by siRNAs targeting USP10. Representative images of the migration assays (right panel). Western blotting images and corresponding quantification of the migration assay (left panel). (I) Knockdown of IGF2BP1 inhibited the improved migration ability of MDA-MB-231 cells caused by USP10 overexpression. Representative images of migration assays (right panel). Western blotting and the corresponding quantification of migration ability results (left panel). (J) Downregulating IGF2BP1 blocked the enhanced metastatic ability of MDA-MB-231 cells induced by USP10 overexpression in female nude mice (total n=9). Representative bioluminescence images of mice 8 weeks after tail vein injection of treated MDA-MB-231 cells attached to the lung region and the corresponding quantification. (K) Corresponding HE staining images of pulmonary metastases (scale bars=275 µm). The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 5

IGF2BP1 stabilized CPT1A via m6A modification. (A) The distribution of m6A peaks in each transcript segment of IGF2BP1 knockdown LM2 cells and vector control LM2 cells was analysed. (B) Enrichment analyses for an averaged base frequency matrix based on the frequency distribution of a specific motif (RRACH) in IGF2BP1 knockdown and control LM2 cells performed by MeRIP-seq. (C) The overlapping differentially expressed genes determined by RNA-seq and MeRIP-seq performed using LM2 cells with stable knockdown of IGF2BP1 and compared with their corresponding controls. (D) The top 20 enriched KEGG pathways based on the overlapping genes in (C). (E) The m6A abundances detected by MeRIP-seq on CPT1A mRNA transcripts in IGF2BP1 knockdown LM2 cells and the corresponding control cells plotted by Integrative Genomics Viewer (IGV). (F-G) The mRNA expression levels of six candidate genes were examined in IGF2BP1 knockdown LM2/1833 cells (F) and IGF2BP1-overexpressing MDA-MB-231 cells (G) and compared with the corresponding control cells by qRT-PCR. (H) MeRIP-qPCR analysis was employed to demonstrate the decrease or increase in m6A modifications on CPT1A mRNA caused by IGF2BP1 downregulation and overexpression in LM2/MDA-MB-231 cells. (I) The mRNA expression of CPT1A was detected in IGF2BP1 knockdown LM2/1833 cells compared with the corresponding control LM2/1833 cells treated with actinomycin D (2 µg/mL) at the indicated time points by qRT-PCR. (J-K) The protein expression of CPT1A was examined in IGF2BP1 knockdown LM2 cells and 1833 cells and IGF2BP1-overexpressing MDA-MB-231 cells and compared with the corresponding control cells. The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 6

IGF2BP1 promoted BC metastasis by upregulating CPT1A. (A-B) CPT1A knockdown and overexpression efficiency verified by western blotting in LM2 cells (A) and MDA-MB-231 cells (B) (upper panel), and the corresponding quantification of the grey values (bottom panel). (C-D) Migration and invasion of LM2 cells with CPT1A knockdown (C) or MDA-MB-231 cells with CPT1A overexpression (D). Representative images (upper panel) and quantification (bottom panel) of cell migration and invasion are shown. (E&G) Upregulating CPT1A repaired the decreased migration ability of LM2 cells with stable knockdown of IGF2BP1 caused by targeting lentiviruses. Western blotting images and corresponding quantification analysis are shown (E). Representative images of migration assays and the corresponding quantification (G). (F&H) Knocking down CPT1A blocked the increased migration ability in IGF2BP1-overexpressing LM2 cells. The results of western blotting and corresponding quantification are shown (F). Representative migration assay results and the corresponding quantification are shown on the right (H). (I) Kaplan-Meier OS curves were analysed based on CPT1A mRNA expression using the online bioinformatics tool. (J-K) Kaplan-Meier OS and DFS analyses were carried out based on the CPT1A IHC staining indices in patients with BC (n=80, p<0.0001, log-rank test). The data are represented as the means ± SEM. * p<0.05; ** p<0.01; *** p<0.001.

Figure 7

The clinical significance of the USP10/IGF2BP1/CPT1A axis in BC prognosis. (A) Representative IHC staining images of USP10, IGF2BP1, and CPT1A in the cancerous tissues of BC patients with high or low USP10 levels. (B) BC patients with high expression of USP10, IGF2BP1, and CPT1A had the lowest OS and DFS compared with the groups with low and no expression of these indicators. Kaplan-Meier OS and DFS analyses of 80 BC patients were performed (n=80, p<0.01, log-rank test). (C) The graphic illustrates that the USP10/IGF2BP1/CPT1A axis promoted BC metastasis.