Fig 1: NF90‐NF45 complex impairs m6A modification of pri‐mir‐7‐1 by METTL3/14 owing to preferential binding of NF90 to pri‐miRNAs in vitro. (A and B) Predicted structural models of pri‐mir‐7‐1/pre‐mir‐7‐1 (A) and pri‐mir‐200a/pre‐mir‐200a (B) by RNAfold (http://rna.tbi.univie.ac.at/cgi‐bin/RNAWebSuite/RNAfold.cgi) using the minimum free energy (MFE) model. Temperature conditions and salt concentration were 37 °C (default) and 1.021 m (default), respectively. The models were represented by FORNA (http://rna.tbi.univie.ac.at/forna). (C and E) In vitro m6A modification assay performed using pri‐mir‐7‐1 (C) and pri‐mir‐200a (E) probes and recombinant METTL3/14, NF90 and NF45 proteins. The levels of m6A modification on pri‐mir‐7‐1 or pri‐mir‐200a were detected by immunoblotting and standardized by methylene blue staining. The spot intensities were measured by densitometry and presented as a bar graph. Data are presented as a scatter plot and expressed as the mean ± SD [n = 4 (C) or 3 (E) per group]. (D) The level of m6A modification in pri‐mir‐7‐1 probes was analyzed using the m6A RNA methylation assay quantification kit in accordance with the manufacturer's protocol (ab185912; Abcam). The assay was performed using a pri‐mir‐7‐1 probe and recombinant METTL3/14, NF90 and NF45 proteins. Data are presented as a scatter plot and expressed as the mean ± SD (n = 5 per group). *P < 0.05, **P < 0.01 relative to control using a two‐tailed Welch's t test. ‘+’ and ‘++’ indicate 25 and 50 ng of recombinant protein, respectively.
Fig 2: Competition between the NF90‐NF45 and METTL3/14 complexes affects miR‐7 biogenesis. (A) HEK293 cells were transfected with pri‐mir‐7‐1, NF90‐NF45 and/or METTL3/14 expression plasmids. The expression levels of NF90, NF45, METTL3 and METTL14 were detected by immunoblotting. GAPDH was used as an internal control. Experiments were performed independently in triplicate, and representative results are shown. (B and C) RNA levels of pri‐mir‐7‐1 (B) and mature miR‐7 (C) in cells transfected with the indicated expression plasmids analyzed by qRT‐PCR. β‐actin and RNU6B were used as internal controls to normalize the data. Data are presented as a scatter plot and expressed as the mean ± SD [n = 3 (B) or 4 (C) per group]. *P < 0.05, **P < 0.01 relative to the sample indicated in the Figure, using a two‐tailed Welch's t test.
Fig 3: Proposed model for the regulatory mechanism of pri‐miRNA m6A modification by NF90‐NF45.
Fig 4: Co‐expression of METTL3/14 overcomes the NF90‐NF45‐induced accumulation of pri‐mir‐7‐1. (A) METTL3/14 and NF90‐NF45 expression levels detected by immunoblotting. GAPDH was used as an internal control. Experiments were performed independently in triplicate, and representative results are shown. (B and C) Mature miR‐7 (B) and pri‐mir‐7‐1 (C) levels in cells transfected with METTL3/14 and/or NF90‐NF45 expression plasmids or mock plasmids analyzed by qRT‐PCR. RNU6B and β‐actin were used as an internal control to normalize the data. Data are presented as a scatter plot (n = 5 per group) and expressed as the mean ± SD. *P < 0.05, **P < 0.01 relative to the mock transformant according to a two‐tailed Welch's t test. n.s., non‐significant. (D) Potential targeted m6A motif sites in pri‐mir‐7‐1 according to the SRAMP online website. The horizontal and vertical axes represent the nucleotide positions within pri‐miR‐7‐1 (1–203 nucleotides) and the probability score for m6A modification (combined score), respectively. The red arrow indicates an adenosine at position 175 (A175) within pri‐mir‐7‐1 with a high probability of m6A modification.
Supplier Page from Abcam for Recombinant human METTL3 + METTL14 protein (Active)