Figure 1.
PRL-3 is phosphoryated in vitro by the Src tyrosine kinase, and endogenous PRL-3 is tyrosine-phosphorylated in cells.
A) Schematic diagram of PRL-3 showing the relative positions of the six tyrosines evaluated in these studies, as well as the C-terminal CaaX motif. The numbers shown reflect the probability of phosphorylation at each site based on predictions by NetPhos 2.0. B) PRL-3 (“WT”) or a mutant in which all six tyrosines were substituted by phenylalanine (“All_F”) was fused to GST, purified from bacteria, and subjected to in vitro phosphorylation with purified Src. WT PRL-3 was phosphorylated by Src while the “All_F” PRL-3 mutant was not. Src itself is visible on the phosphotyrosine blot because it becomes autophosphorylated on Y416 (pSrc). Coomassie staining demonstrates equal amounts of PRL-3 protein in each reaction. C) To determine whether endogenous PRL-3 is also tyrosine phosphorylated in cells, SW480 cells were treated with the tyrosine phosphatase inhibitor pervanadate (VO4, 100 µM, 1 h) to enhance detection of transient tyrosine phosphorylation events. Endogenous PRL-3 was immunoprecipitated using anti-PRL-3 antibody, followed by SDS-PAGE and western blot analysis for either PRL-3 or phosphotyrosine.
Figure 2.
Src is required for PRL-3 tyrosine phosphorylation in cells.
A) SW480 cells were co-transfected with both WT PRL-3 tagged with streptavidin binding protein (SBP) and WT Src in the absence of pervanadate, and then analyzed for phosphotyrosine by western blot following pull down with streptavidin beads. Src kinase activity was evaluated by analysis of the Src autophosphorylation site, Y416. B) WT PRL-3 was transfected into mouse embryo fibroblasts in which expression of Src, Yes and Fyn was simultaneously ablated genetically (SYF cells) or into SYF cells into which only Src had been reintroduced (SYF+Src). Cells were treated with pervanadate (VO4) and the phosphorylation of PRL-3 on tyrosines was then analyzed by western blot following pull down of PRL-3. C) SW480 cells were treated with the Src family kinase inhibitor SU6656 (10 µM or 1 µM, 1 h) prior to treatment with pervanadate and western blot analysis for PRL-3 and phosphotyrosine.
Figure 3.
PDGF stimulates PRL-3 phosphorylation in a Src-dependent manner.
Phosphorylation of endogenous PRL-3 in SW480 cells was evaluated as in Figure 1B following PDGF stimulation (20 ng/mL, 20 min). siRNA was used to reduce Src expression (“Src siR”), or the Src inhibitor SU6656 (5 µM) was used to inhibit Src activity (“SU”). Phosphotyrosine on PRL-3 was analyzed five days after siRNA transfection, and the degree of phosphorylation compared to that seen with control nontargeted siRNA (“Cont. siR”).
Figure 4.
Residue 53 is the primary site of PRL-3 tyrosine phosphorylation.
A) To identify the sites of phosphorylation, SW480 cells were transfected with expression constructs encoding wild-type PRL-3 or mutants in which all six or specific individual tyrosines were mutated to nonphosphorylatable phenylalanines (All_F, Y14F, Y53F, Y126F). Each was expressed as a protein fused to streptavidin-binding protein (SBP). Following pervanadate treatment, SBP-tagged PRL-3 was precipitated using streptavidin-Sepharose, and analyzed by SDS-PAGE and western blot for either PRL-3 or phosphotyrosine. A non-specific band (NSB) appears in all VO4-treated lanes at slightly a faster mobility than PRL-3. B) Conversely, PRL-3 mutants in which all but one tyrosine was mutated to phenylalanine, and which therefore contain a single tyrosine only at position 14 (14Y-only), 53 (53Y-only) or 126 (126Y-only), were analyzed similarly.
Figure 5.
PRL-3 promotion of invasion and cell motility requires an intact Y53.
A) SW480 cells stably expressing WT PRL-3 (WT) or a phosphorylation-deficient mutant lacking tyrosine 53 (Y53F) were subjected to Matrigel invasion assays. Data from three independent assays are shown. For each sample, 5x104 cells were placed in the top chamber in the absence of serum; the bottom chamber contained medium supplemented with 10% serum. After 72 h, invaded cells were fixed, stained and counted. Equal expression of WT HA-tagged PRL-3 and the Y53F mutant was confirmed by HA blot. B) H1299 cells were transiently transfected with EGFP-tagged PRL-3, the Y53F mutant or vector only. The motility rate of individual cells was evaluated using a BioStation IM live cell recorder. Data shown are from three independent assays, and include a total of at least 35 cells for each sample. Images of representative cells expressing each EGFP-tagged protein are shown below the graph. All images were captured under identical microscopic and photographic settings. Data in both A and B are shown +/− SEM, and p-values were calculated using a two-tailed Student's t-test.
Figure 6.
PRL-3 requires the activity of a Src family kinase to promote invasion and RhoC activation.
A) SW480 cells expressing WT PRL-3 were subjected to Matrigel invasion analysis as in Figure 5. The Src family kinase inhibitor SU6656 was added to both the top and bottom chambers at the indicated concentrations. Cells expressing phosphatase-inactive PRL-3 (C104S) served as a negative control. Data are shown normalized to basal levels of invasion of vector-expressing cells. Results from four assays are shown +/− SD. B) SW480 cells expressing WT PRL-3 were treated with the Src inhibitor, SU6656, and analyzed for levels of active, GTP-bound RhoC using a Rhotekin-RBD pull down assay. Images from a representative assay are shown. Lanes containing samples not relevant to this work have been removed. All panels were taken from the same assay using the same photographic exposure time. Phosphatase-inactive PRL-3 (C104S) served as a negative control. C) Data quantitated from at least three pull down assays as in panel B are shown +/− SEM.