Fig 2: Model for MARF1-mediated mRNA decay.MARF1 recognizes target mRNAs via LOTUS domains 3 through 5. Subsequently, RRM1 enhances NYN-mediated cleavage of target mRNAs, potentially by assisting in positioning the NYN domain on target. EDC4 regulates MARF1-mediated repression by interacting with the MARF1 C-terminus, segregating MARF1 to P-bodies and preventing MARF1 from binding target mRNAs, potentially by interfering with LOTUS domain-RNA interactions.

Fig 3: EDC4-MARF1 interaction localizes MARF1 to P-bodies and impairs MARF1 silencing.(A) Schematic model of MARF1 C-terminus competition assay. MARF1 contains a C-terminal motif (red) that interacts with the mRNA decapping machinery, including EDC4, DCP1, DCP2 and XRN1. Co-transfecting a plasmid encoding the MARF1 C-terminal motif may compete with full-length MARF1 in binding to the decapping machinery. (B) Confocal fluorescence micrographs of fixed HeLa cells expressing wild-type λNHA-MARF1 (with or without FLAG-tagged MARF1C-term) and λNHA-MARF1ΔC-term, along with EDC4. Cells were stained with anti‐HA (red) and anti‐EDC4 (green) antibodies. The merged images show the HA signal in red and the EDC4 signal in green. (C) RL activity detected in extracts from HEK293 cells expressing the indicated proteins. Cells were co-transfected with constructs expressing the RL-MAML1 reporter, FL, low amounts of indicated λNHA-MARF1 constructs, along with/without a plasmid coding for MARF1 C-terminal motif. Histograms represented normalized mean values of RL activity from a minimum of three experiments. RL activity values seen in the presence of λNHA-MARF1ΔNYN were set as 100. (D) RNA immunoprecipitation of RL-MAML1 reporter by FLAG-tagged MARF1ΔNYN in HEK293 cells plus/minus EDC4 overexpression.