Figure 3

Prl3a Interacts with Ddx21 in Zebrafish and Melanoma Cell Nuclei

(A) Experimental overview and intensity quantification of Ddx21 peptides from Prl3a-GST pull-downs. Bars and error bars are mean ± SEM. See also Table S1.

(B) Cellular component GO-enrichment analysis of Prl3a-interacting proteins identified by co-immunoprecipitation and mass-spectrometry (adjusted p values: Benjamini-Hochberg test).

(C) Quantification of melanocytes in a mitfa^vc7 regeneration assay following control or ddx21 morpholino injection, ^∗∗∗∗p < 0.0001 Student's t test.

(D) Co-immunoprecipitation of human HA-tagged PRL3 protein from A375 cells: empty vector (EV) and PRL3-expressing stable transfected cells.

(E) Structured illumination microscopy (SIM) of endogenous PRL3 (magenta) and DDX21 (green) in C092 melanoma cells. Scale bar, 1 μm. Co-localization indicated (arrows); zoomed image shows PRL3-DDX21 co-localization with line scan, and intensity plot profile of line scan showing signal overlap (yellow arrow). DAPI: blue. See also Figure S3.

(F) Quantification of PRL3-DDX21 complexes per nucleus identified by SIM in A375 melanoma cells expressing empty vector or HA-tagged PRL3 (^∗∗∗p < 0.001, error bars: x indicates the mean; Student's t test). See also Figure S3.

(G) Mass spectrometry of DDX21 phosphorylation sites in EV control cells (n = 5 samples) and cells expressing PRL3 (n = 7 samples) (^∗∗∗p < 0.001; n.s., not significant; Bars and error bars are mean ± SEM; Student's t test). See also Table S2.

(H) Quantification of regenerating melanocytes in wild-type or prl3a mutant mitfa^vc7 embryos injected with mRNA encoding DDX21 (WT) or DDX21 S71A (S71A), or an uninjected control (uninj) (^∗∗p < 0.01, ANOVA using Tukey’s analysis).