The N¹‐methyladenosine methyltransferase TRMT61A promotes bladder cancer progression and is targetable by small molecule compounds

Dear Editor, Bladder cancer (BLCA) is the most common malignant tumour of the urinary system and has a high recurrence rate.1, 2 N1-methyladenosine (m1A) methylation is a key mechanism of post-transcriptional regulation.3-5 m1A levels in the urine of BLCA patients are higher than in the urine of normal people, and TRMT61A is highly expressed in human BLCA tissues.6, 7 Although some reports are consistent with our previous results,8 the biological functions and potential mechanisms of action of m1A methylation in BLCA remain unknown. Here, the role of TRMT61A in BLCA was assessed in vitro and in vivo. Silencing TRMT61A inhibited 5637 cell proliferation, clonogenicity, migration and invasion, while its overexpression was promoted in T24 cells (Figure S1A,B and Figure 1A–F). Total m1A levels in RNA decreased and increased following TRMT61A knocking-down, and over-expression, respectively. Quantitative ultra-performance liquid chromatography/mass spectrometry analysis confirmed these results (Figure 1G–J). In vivo, tumour growth was effectively suppressed, as reflected by the significant reduction in tumour size and weight in the short hairpin RNA (shRNA) TRMT61A group. Immunohistochemistry staining showed a marked decrease in the proportion of tumour cells positively stained by anti-Ki-67 antibody in the shRNA TRMT61A group (Figure S1C and Figure 1K–M). Overall, TRMT61A promotes m1A modification and BLCA progression. To identify the target mRNAs of TRMT61A in BLCA, we conducted MeRIP-seq and RNA-seq analyses between control shRNA and TRMT61A shRNA. Gene Ontology and heatmap analysis revealed enrichment of genes involved in the regulation of mRNA processing and RNA splicing with decreased m1A methylation after TRMT61A knockdown; Kyoto Encyclopedia of Genes and Genomes pathway analysis indicated that transcripts with decreased m1A methylation were significantly related to the spliceosome and cancer pathways (Figure S2A–D). According to MeRIP-seq analysis, m1A peaks were particularly abundant in the vicinity of 5′ untranslated regions near coding sequence regions and were found throughout the genome and across all chromosomes (Figure 2A and Figure S2A). HMOX2 mRNA was evidently m1A de-methylated in the MeRIP-seq data and downregulated in RNA-seq data after TRMT61A knockdown, and overexpressed in human BLCA tissues; the same result was obtained with The Cancer Genome Atlas (TCGA) database (Figure 2B,C and Figure S2E). AGGCUGGA was the top m1A-modified motif; the HMOX2 mRNA m1A peak diminished upon TRMT61A knockdown (Figure S2F,G). Therefore, further investigations focused on HMOX2 as a TRMT61A target. We observed that HMOX2 was downregulated following TRMT61A knockdown in 5637 cells and upregulated in T24 cells overexpressing TRMT61A (Figure 2D). Using samples from BLCA patients, a positive correlation was observed between HMOX2 and TRMT61A expression; HMOX2 expression decreased in tumour tissues from mice xenografted with TRMT61A shRNA and increased in tumour tissues from BBN-driven urinary BLCA mice (Figures S2G–K and S3). TRMT61A promoted BLCA progression by upregulating the HMOX2 level (Figure S4). Overall, TRMT61A promotes BLCA progression by upregulating HMOX2 expression. Mechanistically, HMOX2 mRNA was enriched by the m1A antibody and m1A modification was reduced after TRMT61A knockdown in 5637 cells. RNA immunoprecipitation (RIP) analysis confirmed the interaction between TRMT61A protein and HMOX2 mRNA in 5637 cells. TRMT61A knockdown decreased the stability of HMOX2 mRNA (Figure 2E–G). YTHDF1 was screened through human BLCA tissues and the TCGA database (Figure 2H,I). RIP assays confirmed the interaction between YTHDF1 protein and HMOX2 mRNA; the interaction was attenuated upon TRMT61A knockdown. RNA pull-down assays confirmed that HMOX2 mRNA could directly bind to TRMT61A and YTHDF1 proteins. In mutant HMOX2, m1A modification was abrogated by the replacement of adenine with guanosine in the m1A consensus sequences. Luciferase assays revealed a significant increase in relative luciferase activity in cells transfected with the wild-type HMOX2 construct, but there was no significant change in luciferase activity in cells transfected with the mutant HMOX2 construct (Figure 2J–L). Next, we explored why TRMT61A is abnormally overexpressed in BLCA. Eleven transcriptional regulators were identified after intersecting transcription factors from the two prediction tools; nuclear factor kappa B (NF-κB) p56 was the transcriptional regulator most significantly associated with TRMT61A (Figure S5A). Real-time polymerase chain reaction (RT-PCR), western blot and IHC results showed that NF-κB was highly expressed in cancer tissues and positively correlated with TRMT61A expression (Figure 3A–C). Subsequently, 5637 cells were treated with SN50, an NF-κB inhibitor, which decreased p65 protein expression and phosphorylation (p-NF-κB, p-p65) and reduced TRMT61A protein expression. Conversely, p-p65 and TRMT61A levels increased after treatment with tumour necrosis factor-alpha (TNF-α), an NF-κB activator, in T24 cells (Figure 3D and Figure S5B,C). However, TRMT61A knockdown or overexpression did not affect p65 and p-p65 expression in 5637 and T24 cells (Figure S5D–F). Bioinformatics prediction by JASPAR revealed two NF-κB binding sites; chromatin immunoprecipitation confirmed strong binding of NF-κB at the -1105 to -1094 and -649 to -638 bp sites of the TRMT61A promoter. Experiments with dual-luciferase reporter genes confirmed the targeting relationship (Figure 3E–G). TRMT61A lacks a catalytic nuclear localization sequence.9 Thus, we examined whether NF-κB bound to TRMT61A protein and regulated its nuclear translocation. NF-κB activation facilitated the nuclear translocation of TRMT61A (Figure 3H–L). Finally, we identified TRMT61A inhibitors using a virtual screening strategy and two compounds (abbreviated as CMP) were potentially effective in inhibiting TRMT61A (Figure 4A,B, Figures S6–S8 and Table S1). Treatment with CMP1 or CMP9 significantly reduced m1A levels; RIP assays with anti-TRMT61A and anti-m1A antibodies and RT-qPCR analysis revealed that treatment with CMP1 or CMP9 significantly reduced TRMT61A protein binding to HMOX2 mRNA, HMOX2 mRNA m1A modification and HMOX2 mRNA expression (Figure 4C–F). In vivo, CMP1 and CMP9 showed strong anti-tumor effects, similar to DDP and Thiram (Figure 4G,H). CMP9 showed a good safety profile, while CMP1 increased serum Urea and Crea levels, indicating kidney toxicity exerted by CMP1 (Figure 4I,J). Consistent with the results of in vitro experiments, m1A levels and HMOX2 mRNA expression in tumour tissues decreased after CMP1 and CMP9 treatment (Figure 4K,L). Additional studies to demonstrate the anticancer efficacy of CMP9 in different BLCA mouse models and pharmacodynamics and pharmacokinetics studies are warranted to facilitate its potential applications. Overall, this study elucidates the molecular mechanisms via which NF-κB increases the transcription of TRMT61A, which accelerates m1A modification of HMOX2 mRNA and enhances its stability via a YTHDF1-dependent mechanism, ultimately promoting BLCA progression. The small molecule compound TRMT61A inhibitor CMP9 is a promising novel anticancer agent (Figure 4M). Jianjian Yin designed the study, performed experiments, analyzed and interpreted the data and wrote the manuscript. Dongkui Song, Lirong Zhang and Tao Liu conceived the study and reviewed and edited the manuscript. Fengmin Shao supervised the study and reviewed the manuscript. Xudong Zhang and Lei Jin supervised the experiments. Xin Fan, Qi Chang and Linlin Yang performed experiments and interpreted the data. Yuanheng Dai, Tao Wang, Lei Shi and Xiaoming Yang collected clinical specimens. All authors have read and approved the final manuscript. We thank Yongjie Lu (Centre for Molecular Oncology, Barts Cancer Institute, Queen Mary University of London, UK) for the helpful discussions. The authors declare no conflict of interest. This work was funded by the National Natural Science Foundation of China (Grant Nos. U1904162 and 82373077). This study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University, and written informed consent was obtained from the patients or their relatives prior to the study. All animal experiments were approved by the Zhengzhou University Animal Care and Ethics Committee (ZZUIRB2022-143). The original contributions presented in the study are included in the article/Supporting Information. The publicly available datasets analyzed in the current study are available from TCGA (https://www.cancer.gov/ccg/research/genome-sequencing/tcga). The RNA and MeRIP sequence datasets presented in this study can be found in online repositories (GSE255629). Further information and requests for resources and reagents should be directed to and will be fulfilled by the corresponding authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.