N6-methyladenosine modification of MEF2A weakens cetuximab sensitivity in colorectal cancer via PD-L1/SOX12 axis

Abstract

Colorectal cancer (CRC) treatment is still a challenge due to chemoresistance. We explored MEF2A function and underlying mechanism on cetuximab sensitivity in CRC. In this study, cancer tissues and adjacent non-cancerous samples were harvested from CRC patients. Cell viability, proliferation and apoptosis in CRC cells were tested by CCK-8, EdU, colony formation, and flow cytometry assays. The binding of MEF2A on the PD-L1 promoter was validated using luciferase reporter assay, CHIP, and EMSA, while the relationship of PD-L1 and SOX12 mRNA, as well as RBM15/IGF2BP1 and MEF2A mRNA, was verified by RIP, RNA pull-down, or FISH combined with immunofluorescence. m6A modification level of MEF2A mRNA was assayed by MeRIP. The expressions of key genes and proteins, including MEF2A, PD-L1, SOX12, RBM15, IGF2BP1, apoptosis- and cell cycle-related proteins, were determined with RT-qPCR, western blot, or immunohistochemistry. In vivo function of MEF2A was validated by establishing a xenograft nude mice model. The results showed that MEF2A was increased in CRC cells and tissues, while it was higher in cetuximab-resistant CRC tissues. Silencing MEF2A improved the sensitivity of cetuximab in CRC cells and xenograft mice. MEF2A binds to PD-L1 promoter to transcriptionally upregulate PD-L1 expression. Increased cetuximab sensitivity was observed in PD-L1 knockout (KO) CRC cells. PD-L1 overexpression reversed the enhanced cetuximab sensitivity induced by MEF2A knockdown. PD-L1 binds to SOX12 mRNA to stabilize its expression. PD-L1 knockdown augmented cetuximab sensitivity, which was overturned by SOX12 overexpression. The m6A modification mediated by RBM15/IGF2BP1 upregulated MEF2A expression in cetuximab-resistant CRC tissues. In conclusion, m6A-modified MEF2A alleviated cetuximab sensitivity in CRC via PD-L1/SOX12 mRNA axis, indicating that MEF2A might function as a promising therapeutic target against cetuximab-resistant CRC.

Fig. 1. MEF2A was increased in CRC tissues, cetuximab-resistant CRC tissues, as well as CRC cell lines.

CRC patients (n = 20, including 8 cetuximab-sensitive and 12 cetuximab-resistant specimens) were recruited for the collection of cancerous tissues and adjacent non-cancerous tissues. RT-qPCR for measuring MEF2A mRNA levels in normal tissues and cancerous tissues (A), as well as in cetuximab-resistant and cetuximab-sensitive CRC tissues (B). MEF2A protein level in clinical samples was detected with immunohistochemistry. Scale bar: 50 μm and 10 μm (C). D The association between 3-year and 5-year survival rates of CRC patients and MEF2A level showed by bioinformatics analysis (Kaplan–Meier Plotter). E In normal colonic epithelial cell line (FHC) and CRC cell lines (HCT 116, SW480, SW620, Caco-2, and LoVo), MEF2A protein levels were determined with western blot. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 2. MEF2A knockdown enhanced cetuximab sensitivity in CRC cells.

A, B sh-MEF2A was transfected into CRC cells (SW480 and LoVo) for 48 h. MEF2A mRNA and protein expressions were determined using RT-qPCR (A) and western blot (B). CRC cells were transfected with sh-MEF2A, before cetuximab exposure. CCK-8 assay was used for measuring cell viability (C), and IC50 values were calculated (D). Clonal formation assay (E) and EdU staining (F) for detecting cell proliferation. Scale bar: 100 μm. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 3. MEF2A bound to PD-L1 promoter to transcriptionally upregulate PD-L1 expression.

The binding motif for MEF2A (A) and MEF2A binding sequences at PD-L1 promoter region (B) were forecasted through JASPAR database. Luciferase reporter assay (C), CHIP (D), and EMSA (E) for analyzing the binding of MEF2A protein on PD-L1 promoter in CRC cells. F, G sh-MEF2A or MEF2A overexpression vector was transfected into CRC cells. RT-qPCR (bF) and western blot (G) for testing PD-L1 mRNA and protein expressions. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 4. Knockdown of MEF2A boosted cetuximab sensitivity in CRC by inhibiting PD-L1 transcription.

CRC cells transfected with sh-MEF2A or PD-L1 overexpression vector were exposed to cetuximab. A Western blot was performed for measuring MEF2A and PD-L1 protein expressions. B CCK-8 for measuring cell viability. Cell proliferation was detected with clonal formation assay (C) and EdU staining (D). Scale bar: 100 μm. E Flow cytometry for assessing apoptosis. F Western blot for measuring cleaved caspase-3, uncleaved PARP, cyclin D1, cyclin E1, and CDK2. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 5. PD-L1 bound to SOX12 mRNA and stabilized its expression.

A PD-L1 protein pulled down by biotinylated SOX12 probe in RNA pull-down assay. B The co-localization of SOX12 mRNA and PD-L1 was measured using FISH combined with immunofluorescence. Scale bar: 25 μm. Representative immunofluorescence (C) and FISH (D) images in the negative control experiments. Scale bar: 25 μm. E RIP for detecting the binding relationship between PD-L1 and SOX12 mRNA. F Actinomycin D was added into CRC cells transfected with sh-PD-L1 or PD-L1 overexpression vector. SOX12 mRNA expressions were tested using RT-qPCR. G SOX12 level in CRC was analyzed using bioinformatics analysis (UALCAN). H SOX12 mRNA levels in normal tissues (n = 20), cetuximab-sensitive (n = 8) and cetuximab-resistant (n = 12) CRC tissues were detected with RT-qPCR. I Western blot for determining SOX12 protein level in CRC cells. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 6. PD-L1 knockdown strengthened cetuximab sensitivity by inhibiting the stability of SOX12 mRNA.

CRC cells transfected with sh-PD-L1 or SOX12 overexpression vector were exposed to cetuximab. A Western blot for testing PD-L1 and SOX12 protein expressions. B CCK-8 assay for cell viability measurement. EdU staining (C) and clonal formation assay (D) for determining cell proliferation. Scale bar: 100 μm. E Apoptosis analyzed by flow cytometry. F Western blot for determining the expressions of cleaved caspase-3, uncleaved PARP, cyclin D1, cyclin E1 and CDK2. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 7. RBM15/IGF2BP1-mediated m6A modification led to upregulation of MEF2A in cetuximab-resistant CRC tissues.

A Total m6A levels in adjacent normal tissues and cancerous tissues (n = 20) were tested by commercial kit. B Total m6A levels in cetuximab-sensitive (n = 8) and resistant (n = 12) CRC tissues. C MeRIP for measuring m6A modification level of MEF2A mRNA in normal tissues (n = 20), cetuximab-sensitive (n = 8) and resistant (n = 12) CRC tissues. D Bioinformatics prediction (ENCORI) showed the recognition of MEF2A mRNA by both “writer” RBM15 and “reader” IGF2BP1. E RT-qPCR for testing the mRNA expressions of RBM15 and IGF2BP1 in normal tissues (n = 20), cetuximab-sensitive (n = 8) and resistant (n = 12) CRC tissues. The combination of RBM15/IGF2BP1 with MEF2A mRNA was validated using RNA pull-down (F) and RIP (G, H) in CRC cells. I Actinomycin D was added into CRC cells transfected with sh-IGF2BP1 or IGF2BP1 overexpression vector. MEF2A mRNA level was determined by RT-qPCR. J CRC cells were transfected with sh-RBM15 or RBM15 overexpression vector. RIP was used to verify IGF2BP1 binding to MEF2A mRNA. K RBM15 overexpression vector or sh-IGF2BP1 was co-transfected into CRC cells, before actinomycin D treatment. MEF2A mRNA expression was tested by RT-qPCR. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 8. MEF2A knockdown promoted cetuximab sensitivity in CRC xenograft nude mice.

CRC cells (SW480 or LoVo) with stable MEF2A knockdown were subcutaneously injected into nude mice, before cetuximab treatment. A Tumors formed in each group. B Tumor size. C Tumor weight. D Immunohistochemistry for observing Ki-67 expression in tumor tissues. Scale bar: 50 μm. E TUNEL staining for assessing apoptosis in tumor tissues. Scale bar: 50 μm. F Immunohistochemistry for measuring MEF2A expression in tumor tissues. Scale bar: 50 μm. G MEF2A, PD-L1, and SOX12 protein levels in tumor tissues were determined using western blot. H The protein expressions of cleaved caspase-3, uncleaved PARP, cyclin D1, cyclin E1, and CDK2 in tumor tissues were tested by western blot. All bar chart analysis of western blot is a repeated experiment three times. *p < 0.05, **p < 0.01, and ***p < 0.001.