Silencing of METTL3 effectively hinders invasion and metastasis of prostate cancer cells

Abstract

Background: Since primary prostate cancer (PCa) can advance to the life-threatening metastatic PCa, exploring the molecular mechanisms underlying PCa metastasis is crucial for developing the novel targeted preventive strategies for decreasing the mortality of PCa. RNA N⁶-methyladenosine (m⁶A) is an emerging regulatory mechanism for gene expression and its specific roles in PCa progression remains elusive. Methods: Western blotting, quantitative real-time PCR and immunohistochemical analyses were used to detect target gene expression in PCa cells in vitro and prostate tissues from patients. RNA immunoprecipitation was conducted to analyze the specific binding of mRNA to the target protein. Migration and invasion assays were used to assess the migratory capacities of cancer cells. The correlation between target gene expression and survival rate of PCa patients was analyzed based the TCGA database. Results: We found that total RNA N⁶-methyladenosine (m⁶A) modification levels were markedly upregulated in human PCa tissues due to increased expression of methyltransferase like 3 (METTL3). Further studies revealed that the migratory and invasive capacities of PCa cells were markedly suppressed upon METTL3 knockdown. Mechanistically, METTL3 mediates m⁶A modification of USP4 mRNA at A2696, and m⁶A reader protein YTHDF2 binds to and induces degradation of USP4 mRNA by recruiting RNA-binding protein HNRNPD to the mRNA. Decrease of USP4 fails to remove the ubiquitin group from ELAVL1 protein, resulting in a reduction of ELAVL1 protein. Lastly, downregulation of ELAVL1 in turn increases ARHGDIA expression, promoting migration and invasion of PCa cells. Conclusions: Our findings highlight the role of METTL3 in modulating invasion and metastasis of PCa cells, providing insight into promising therapeutic strategies for hindering PCa progressing to deadly metastases.

Keywords: METTL3; m6A; metastasis; prostate cancer; therapeutic strategies.

Figure 1

The m⁶A mRNA methylation mediates prostate cancer (PCa) progression. A, Relative m⁶A levels in RWPE-1, PC3, DU145, and LNCaP cells were measured by ELISA-based m⁶A quantitative analyses. Data were presented as means ± SEM (n = 3), *p < 0.05. B-C, Cellular m⁶A levels were assessed by immunofluorescence staining, and fluorescence intensity was presented as means ± SEM (n = 3), *p < 0.05. D-E, Protein levels of m⁶A regulatory enzymes in indicated cells were determined by western blotting and quantitatively analyzed. Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the RWPE-1 cells. F-J, The METTL3 mRNA levels were analyzed based on 4 different GSE datasets and the TCGA database (*p < 0.05). PT: prostate tissue; PC: prostate cancer; CRPC: castration-resistant prostate cancer. K-N, Association of METTL3, FTO, METTL14, and ALKBH5 mRNA levels with survival of PCa patients was performed using Kaplan-Meier survival curve analysis methods based on the TCGA database. O-R, Immumohistochemical staining was performed to evaluate the m⁶A levels and expression of METTL3 in 25 paired human PCa tissues and their adjacent normal prostate tissues. The m⁶A levels (Q) and METTL3 protein levels (R) were analyzed by calculating the integrated optical density per area (IOD/area). Data were presented as means ± SEM (n = 25), *** p < 0.001.

Figure 2

METTL3 promotes prostate cancer (PCa) metastasis. A-C, Two independent shRNA sequences targeting METTL3 (sh-METTL3#1 and sh-METTL3#2) were separately transfected into PC3 or DU145 cells. Protein levels of METTL3 were determined by western blotting and quantitatively analyzed (A, B), and cellular m⁶A levels were measured by ELISA-based m⁶A quantitative analyses (C). D-F, The migration and invasion abilities of indicated cells were evaluated. Representative images (D, E) and quantification (F) of the cell migration and invasion assay results were shown. G PCa cells were stained with rhodamine phalloidin (red) and DAPI (blue). H-L, Overexpression constructs of METTL3 (O/E-METTL3) were stably transfected into PC3 and DU145 cells, respectively. Protein levels of METTL3 were determined by western blotting (H), and cellular m⁶A levels were measured by ELISA-based m⁶A quantitative analyses (I). The migration and invasion abilities of indicated cells were assessed, and the representative images of the cell migration and invasion assay results were shown (J-L). Data were presented as means ± SEM where relevant, *p < 0.05 vs. the control cells.

Figure 3

ARHGDIA mediates METTL3 promotion of prostate cancer (PCa) metastasis. A-B, PC3 and DU145 cells were transfected with shRNA targeting METTL3 (sh-METTL3), respectively. Western blotting was used to examine protein expression of invasion- and EMT-related genes. C-D, Immumohistochemical staining was performed to evaluate the expression of ARHGDIA in 25 paired human PCa tissues and their adjacent normal prostate tissues, and ARHGDIA protein levels were analyzed by calculating the integrated optical density per area (IOD/area). Data were presented as means ± SEM (n = 25), *** p < 0.001. E-G, The ARHGDIA mRNA levels were analyzed in 2 different GSE datasets and the TCGA database, *p < 0.05. PT: prostate tissue. H, Correlation between ARHGDIA mRNA expression and survival of PCa patients was performed using Kaplan-Meier survival curve analysis methods based on the TCGA database. I-N, PCa cells were transfected with sh-METTL3#1 before transfection with pcDNA3.1-ARHGDIA (O/E-ARHGDIA). Protein levels of ARHGDIA were determined by western blotting and quantitatively analyzed (I, J). The migration and invasion abilities of indicated cells were evaluated. Representative images (K, L) and quantification (M, N) of the cell migration and invasion assay results were shown. Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells, # p < 0.05 vs. the sh-METTL3-treated cells.

Figure 4

ELAVL1 reduces ARHGDIA mRNA stability and prostate cancer (PCa) metastasis. A-E, PC3 cells were transfected with shRNA targeting METTL3 (sh-METTL3). The ARHGDIA mRNA levels in indicated cells were determined by qRT-PCR assay (A). The ARHGDIA promoter constructs were transfected into PC3 cells, and luciferase activity was measured (B). PC3 cells were treated with actinomycin D (5 µg/mL) for 2 h, followed by measurement of ARHGDIA mRNA levels at indicated times (C). Protein levels of ELAVL1, NCL, and HNRNPD were determined by western blotting and quantitatively analyzed (D, E). Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells. F-G, Immumohistochemical staining was performed to evaluate the expression of ELAVL1 in 25 paired human PCa tissues and their adjacent normal prostate tissues (F), and ELAVL1 protein levels (G) were analyzed by calculating the integrated optical density per area (IOD/area). Data were presented as means ± SEM (n = 25), *** p < 0.001. H, Correlation between ELAVL1 mRNA expression and survival of PCa patients was analyzed using Kaplan-Meier survival curve analysis methods based on the TCGA database. I-K, PC3 cells were transfected with sh-ELAVL1. Protein levels of ELAVL1 and ARHGDIA were determined by western blotting (I). The migration and invasion abilities of indicated cells were evaluated. Representative images (J) and quantification of the cell migration and invasion assay results were shown (K). Data were presented as means ± SEM (n = 5), * p < 0.05 vs. the control cells. L-P, PCa cells were transfected with sh-METTL3 before transfection with sh-ELAVL1. Protein levels of METTL3, ELAVL1, and ARHGDIA were determined by western blotting and quantitatively analyzed (L, M). The migration and invasion abilities of indicated cells were evaluated. Representative images and quantification of the cell migration and invasion assay results were shown (N, O). PC3 cells were treated with actinomycin D (5 µg/mL) for 2 h, followed by measurement of ARHGDIA mRNA levels at indicated times (P). Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells, # p < 0.05 vs. the sh-METTL3-treated cells. Q, ELAVL1 was immunoprecipitated, followed by qRT-PCR assay to evaluate the association of the ARHGDIA transcripts with ELAVL1 protein. Data were presented as means ± SEM (n = 3), * p < 0.05.

Figure 5

USP4 is an METTL3 downstream effector and mediates ELAVL1 protein stability. A, The ELAVL1 mRNA levels in indicated cells were determined by qRT-PCR assay. B-C, The indicated PC3 cells were pretreated with cycloheximide (CHX, 10 µg/mL) for 3 h, followed by measurement of ELAVL1 protein levels at indicated times. D-E, PC3 cells were treated with cycloheximide (CHX) for 12 h. After washing out CHX, cells were cultured for the indicated times. ELAVL1 synthesis levels were determined by western blotting and quantitatively analyzed. F, PC3 cells were treated with MG132 for 6 h. Lysates from the indicated cells were subjected to coimmunoprecipitation (Co-IP) assay with anti-ELAVL1 antibody, and the blots were then probed with anti-ubiquitin (UB) antibody for detection of ubiquitination of ELAVL1. G-H, Protein levels of ELAVL1, NCL, and HNRNPD in indicated cells were determined by western blotting and quantitatively analyzed. Data were presented as means ± SEM (n = 25), * p < 0.001 vs. the control cells. I-J, Immumohistochemical staining was performed to evaluate the expression of USP4 in 25 paired human PCa tissues and their adjacent normal prostate tissues (I), and USP4 protein levels (J) were analyzed by calculating the integrated optical density per area (IOD/area). Data were presented as means ± SEM (n = 25), *** p < 0.001. K, The USP4 mRNA levels were analyzed in the TCGA database. L, Correlation between USP4 mRNA expression and survival of PCa patients was analyzed using Kaplan-Meier survival curve analysis methods based on the TCGA database. M-R, PC3 cells were transfected with sh-METTL3 before transfection with sh-ELAVL1. Protein levels of METTL3, USP4, ELAVL1, and ARHGDIA were determined by western blotting and quantitatively analyzed. The migration and invasion abilities of indicated cells were evaluated. Representative images and quantification of the cell migration and invasion assay results were shown. The indicated cells were pretreated with CHX for 3 h, followed by measurement of ELAVL1 protein levels at indicated times. S, Lysates from the indicated cells were subjected to Co-IP with anti-ELAVL1, and the blots were probed with anti-USP4 antibody. T, PC3 cells were transfected with indicated plasmids, and ubiquitination of ELAVL1 was measured by Co-IP assay. Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells, # p < 0.05 vs. the sh-METTL3-treated cells. WCL: whole cell lystate.

Figure 6

m⁶A induces decay of the USP4 transcript in prostate cancer (PCa) cell. A-C, PC3 and DU145 cells were transfected with shRNA targeting METTL3 (sh-METTL3), respectively. The USP4 mRNA levels in indicated cells were determined by qRT-PCR assay (A). The USP4 promoter constructs were transfected into indicated cells, and luciferase activity was measured (B). PCa cells were treated with actinomycin D (5 μg/mL) for 2 h, followed by measurement of USP4 mRNA levels at indicated times (C). Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells. D-E, PC3 cells (D) and DU145 cells (E) were treated with indicated siRNAs, and cell lysates were subject to western blotting. F, Lysates from the PC3 cells were subjected to immunoprecipitation with anti-YTHDF2, and the association of the USP4 transcript with YTHDF2 was determined by qRT-PCR. Data were presented as means ± SEM (n = 3), * p < 0.05. G, Schematic representation of the position of m⁶A motifs within USP4 transcript. H, Abundance of USP4 transcript among mRNA immunoprecipitated with anti-m⁶A antibody was measured by qRT-PCR and normalized to input. Data are presented as means ± SEM (n = 3). * p < 0.05 vs. the IgG group. I, Abundance of USP4 transcript among mRNA immunoprecipitated with anti-m⁶A antibody was measured by qRT-PCR. Data are presented as means ± SEM (n = 3). * p < 0.05 vs. the control cells. J, USP4-CDS of the wild-type or mutant (A to G) was fused with a luciferase reporter. K-M, Luciferase activity of USP4-CDS was measured and normalized to Renilla luciferase activity. Data are presented as means ± SEM (n = 3). * p < 0.05. N-O, Lysates from the indicated cells were subjected to immunoprecipitation with anti-HNRNPD (N) or anti-ELAVL1 (O), and the association of the USP4 transcript with each protein was determined by qRT-PCR. Data were presented as means ± SEM (n = 3), * p < 0.05. P-R, PC3 cells and DU145 cells were treated with indicated siRNAs, and cell lysates were subject to western blotting (P). The USP4 mRNA levels in indicated cells were determined by qRT-PCR (Q). Luciferase activity of USP4-CDS in indicated cells was measured and normalized to Renilla luciferase activity (R). Data are presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells.

Figure 7

METTL3-USP4-ELAVL1-ARHGDIA regulation axis promotes prostate cancer (PCa) metastasis. PC3 cells were transfected with shRNA targeting METTL3 (sh-METTL3) before transfection with sh-USP4. A-C, Athymic nude mice were subcutaneously injected into the right axillary region of each mouse with indicated cells. Eight weeks after cell injection, the mice were sacrificed and the tumors were surgically removed and photographed (A, B), as well as weighed (C). D-F, The PC3 cells were injected into the SCID mice by tail vein injection. Representative images of metastatic nodules in the lung (D) and the H&E staining results were shown (E), and the number of metastatic nodules were quantitatively analyzed (F). G-K, The lung tissues obtained from SCID mice were subjected to immumohistochemical staining assay for evaluating METTL3 (G), m⁶A modification (H), USP4 (I), ELAVL1 (J), and ARHGDIA (K) levels. L, The m⁶A levels and the expression levels of the target proteins were analyzed by calculating the integrated optical density per area (IOD/area). Data were presented as means ± SEM (n = 6), * p < 0.05 vs. the control cells, # p < 0.05 vs. the sh-METTL3-treated cells.

Figure 8

Enhanced METTL3 expression is associated with reduced METTL3 promoter methylation in prostate cancer (PCa) cells. A, The methylation levels of METTL3 promoter were analyzed based on the TCGA database (*p < 0.05). B, Potential CpG islands in the human METTL3 promoter were analyzed using online MethPrimer software (http://www.urogene.org/methprimer/), and the blue shaded regions indicate the potential CpG islands. C-E, The methylation status of the USP4 promoter at different CpG islands was measured by methylation-specific PCR. Data were presented as means ± SEM (n = 5), * p < 0.05 vs. the RWPE-1 cells. F, Celluarl METTL3 mRNA levels were determined by qRT-PCR. Data were presented as means ± SEM (n = 5), * p < 0.05 vs. the RWPE-1 cells. G-L, PCa cells were treated with Bobcat339 at 50 μM for 48 h. The METTL3 mRNA levels in PCa cells were determined by qRT-PCR (G). The METTL3 protein levels in PCa cells were determined by western blotting and quantitatively analyzed (H, I). The migration and invasion abilities of indicated cells were evaluated, and representative images were shown (J-L). Data were presented as means ± SEM (n = 3), * p < 0.05 vs. the control cells. M, A model for the critical link between METTL3 and PCa metastasis is proposed. Reduced METTL3 promoter methylation increases METTL3 expression by promoting its transcription. Upregulation of METTL3 increases cellular m⁶A mRNA methylation levels, which downregulates USP4 expression by inducing m⁶A-mediated decay of the USP4 transcript. USP4 reduction leads to decreased expression of ELAVL1 by increasing ubiquitination (UB) of ELAVL1, contributing to upregulated expression of ARHGDIA by retarding ELAVL1-mediated decay of the ARHGDIA transcript.