Fig 2: Validation of the ARAF/MITF interaction.a Identification of endogenous ARAF/MITF complexes by co-immunoprecipitation. ARAF-only cells extracts were immunoprecipitated with an anti-ARAF antibody. Immune complexes (IP) and total extracts (TE) were immunoblotted with anti-MITF and anti-ARAF antibodies. b Identification of endogenous ARAF/MITF complexes by Proximity Ligation Assay. ARAF/MITF complexes were visualized as red dots in ARAF-only cells compared to control cells expressing an shRNA against ARAF by using a fluorescence microscope. Cell nuclei were stained with DAPI. Scatter plots show the average number of dots per cell (at least 148 nuclei were observed) from six microscopic fields and are representative of three different experiments. Means with standard deviations are shown. *p value = 0.0019 compared by unpaired t test with Welch’s correction. Representative pictures from three independent experiments are shown. Scale bar: 100 µm. c Identification of a direct ARAF/MITF interaction by in vitro coimmunoprecipitation. Recombinant MITF and flag-tagged ARAF proteins were incubated in NP40 Buffer. ARAF was immunoprecipitated with an anti-flag antibody. Immune complexes and recombinant proteins were blotted with anti-MITF and anti-ARAF antibodies.

Fig 3: Characterization of RAF/MITF interaction.MITF interaction with ARAF (a), BRAF (b) or CRAF (c). HEK293T cells were cotransfected with the myc-MITF construct and each of the three HA-ARAF, HA-BRAF, or HA-CRAF constructs (panels a, b, c respectively). RAF proteins were immunoprecipitated with anti-HA antibody. Immune complexes and total extracts were immunoblotted with anti-myc or anti-HA antibodies. d HEK293T cells are cotransfected with the HA-MITF construct and each of the three myc-BRAFWT, myc-BRAFV600E or myc-BRAFKD constructs. Cell lysates were immunoprecipitated with anti-myc antibody. Immune complexes and total extracts were revealed with anti-HA and anti-myc antibodies. e HEK293T cells were cotransfected with the myc-MITF construct and HA-Nter or flag-Cter or both constructs. Cell lysates were immunoprecipitated either with anti-flag or anti-HA antibodies, and immune complexes were revealed with anti-MITF, anti-HA or anti-flag antibodies. Transfection efficiency was monitored by direct western blotting of total protein extracts. Coimmunoprecipitations were quantified using Image J software. The ratio of immunoprecipitated MITF over total MITF (IP/TE) was obtained by dividing the measured MITF signal intensity in immunoprecipitation (IP) by the MITF signal in the total extract (TE) for each condition and the ratio was set to 1 for the control condition. Coimmunoprecipitations are representative of at least three independent experiments.

Fig 4: Effect of RAF proteins on MITF subcellular localization.Subcellular localization of MITF in the presence of ARAF (a), BRAF (b) or CRAF (c). HEK293T cells were cotransfected with myc-MITF and HA-ARAF, HA-BRAF or HA-CRAF. Cell immunostaining was performed with anti-myc or anti-MITF and anti-ARAF, anti-BRAF or anti-CRAF. Scatter plots represent the ratio MITFnuc/MITFcyt calculated by measuring the quantity of nuclear MITF over cytoplasmic MITF. Means with standard deviations are shown. Scale bar: 200 µm; **** p value < 0.0001, ** p value = 0.014 or 0.029 (for BRAF and CRAF, respectively) compared by unpaired t test with Welch’s correction.

Fig 5: Effect of RAF proteins on MITF transcriptional activity.HEK293T cells were cotransfected with 5 ng of MITF plasmid and increasing doses of either ARAF, CRAF, BRAFWT, BRAFV600E or BRAFKD constructs in the presence of a TYR-Luc luciferase reporter and a control β-galactosidase reporter. The ratio of luciferase to β-galactosidase activities is shown as the mean with standard deviations of three replicates (n = 3). One-way ANOVA test was performed to compare all conditions and Dunnett’s tests for the multiple comparisons to the no MITF condition (n.s., non-significant; * p value < 0.05; ** p value < 0.01; *** p value < 0.001; **** p value < 0.0001).