Phosphatase PRL-3 is a Direct Regulatory Target of TGF β in Colon Cancer Metastasis

Metastasis causes most deaths from cancer yet mechanistic understanding and therapeutic options remain limited. Overexpression of the phosphatase PRL-3 is associated with metastasis of colon cancer. Here we show here that PRL-3 is a direct target of signaling by transforming growth factor β (TGF β ), which is broadly implicated in progression and metastasis. We found that suppression of PRL-3 expression by TGF β was mediated by Smad-dependent inhibition of PRL-3 transcription at the level of promoter activity. PRL-3 activation stimulated PI3K/AKT signaling which caused resistance to stress-induced apoptosis. PRL-3 overexpression promoted metastatic colonization in an orthotopic mouse model of colon cancer, whereas PRK-3 knockdown reduced metastatic potential. Altered metastatic phenotypes were not derivative of primary tumor development or local invasion but could be attributed to PRL-3-mediated cell survival. Our findings suggest that inhibiting PRL-3 expression might be an important mechanism through which TGF β suppresses metastasis in colon cancer. Additionally, our findings suggest that loss of TGF β signaling, which occurs commonly during colon cancer progression, is sufficient to activate a PRL-3-mediated cell survival pathway that can selectively promote metastasis. Therefore, a major implication of our findings is that PRL-3 antagonists may offer significant value for anti-metastatic therapy in patients with colon cancer. of PRL-3 up-regulation in colon cancer. In addition, our studies show that PRL-3 activates AKT and maintains AKT activation under growth factor deprivation stress (GFDS). Ectopic expression of PRL-3 increases cell survival under GFDS and promotes metastasis in vivo , whereas knockdown of PRL-3 expression sensitizes colon cancer cells to stress-induced apoptosis and reduces metastasis in an orthotopic model. (TUNEL) The apoptosis was determined quantitatively by counting the number of positively stained apoptotic bodies per field at 20x magnification. Six animals were analyzed for each type of cells. Three histologically similar fields were randomly selected from each slide for analysis. P values were calculated using Student’s t -test. key role for PRL-3 distant colony formation, indicating a strong potential for treatment of patients with colon cancer metastases by PRL-3 antagonism. In summary, we have identified PRL-3 as a direct target of TGF β signaling. Loss of TGF β signaling leads to up-regulation of PRL-3 expression, which is likely a mechanism of increased PRL-3 expression in colon cancer. Up-regulation of PRL-3 expression leads to activation of the PI3K/AKT axis, enhanced survival and resistance to stress-induced apoptosis and increased metastatic potential. Our studies suggest that PRL-3 may be a promising target for cancer treatment, especially in patients with defective TGF β signaling, which occurs in 30-50% of colon cancer patients This is highly important considering that TGF β signaling may actually promote metastasis at a late stage of carcinogenesis which makes restoring TGF β signaling complicated or impossible as a therapeutic strategy. D, Relative PRL-3 promoter activity in above cells treated with TGF β . Inhibition of promoter activity by TGF β was calculated as percentage of suppression relative to the control. The data are presented as the mean ± SD of triplicate experiments. * P < 0.02.


INTRODUCTION
Transforming growth factor β (TGFβ) plays an important role in tumorigenesis and metastasis. Upon ligand binding, TGFβ type II receptor (RII) recruits and activates TGFβ type I receptor (RI), which then activates Smad2 and Smad3. Activated Smad2 and Smad3 form complexes with Smad4 and translocate to the nucleus, where they regulate gene expression (1). We and others have demonstrated that TGFβ suppresses tumor initiation in a variety of cancers including colon cancer, and that loss of TGFβ signaling leads to malignancy (2)(3)(4)(5). However, the role of TGFβ signaling in metastasis has been controversial. Although many studies have shown that TGFβ promotes metastasis (6), others have demonstrated that TGFβ suppresses metastasis (7;8). Recently, studies of human tumor samples indicate that loss or reduction of TGFβ signaling in human colorectal tumors is associated with development of metastasis (9;10).
Our previous studies indicate that abrogation of TGFβ signaling enables increased survival under stress in colon cancer cells (11). In addition, we have shown that loss of TGFβ signaling is associated with increased metastasis, whereas enhanced TGFβ signaling suppresses metastasis in an orthotopic model of colon cancer (Simms et al., unpublished data). These results suggest that endogenous TGFβ increases stress-induced apoptosis to prevent metastatic progression. In contrast, abrogation of TGFβ signaling leads to activation of oncogenic signals that promote survival and protect tumor cells from stress-induced apoptosis, thereby increasing their metastatic potential. Identification To determine how TGFβ signaling might regulate the promoter activity of PRL-3, a reporter plasmid of PRL-3 promoter (Prl-3-F5) was transfected into FET cells followed by treatment of TGFβ. Luciferase assays showed that TGFβ inhibited the activity of this PRL-3 promoter by approximately 55% (Fig. 1D,  Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

TGFβ signaling inhibits expression of PRL-3 in colon cancer cells
Copyright © 2010 American Association for Cancer Research Smad-independent TGFβ signaling has also been reported in different cell types (27;28).
To determine whether TGFβ-mediated inhibition of PRL-3 expression is Smaddependent or Smad-independent, Smad2 and Smad3 were knocked down individually or simultaneously in FET cells by shRNAs specific for Smad2 or Smad3. Expression of Smad2 and/or Smad3 was reduced efficiently and specifically in Smad2, Smad3 or Smad2/Smad3 knockdown cells whereas Smad4 expression was not affected ( Fig. 2A).
Western blot (Fig. 2B, Fig. S4A), RT-PCR (Fig. 2C, Fig. S4B) and promoter analyses PRL-3 is a direct target of TGFβ/Smad signaling. Two Smad binding element (SBE)-like sites, E1 and E4, were identified in PRL-3 promoter (Fig. 3A). Smad3 and Smad4 recombinant proteins were tested by an EMSA assay to determine whether they bind to E1 and/or E4. The results showed that GST-Smad3 bound to wild type E1 and E4 oligonucleotides in a concentration-dependent manner (Fig. 3B Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Copyright © 2010 American Association for Cancer Research the specificity of these binding sites (Fig. 3B). The binding of Smad3 or Smad4 to E1 and E4 was further confirmed in vivo by chromatin immunoprecipitation (CHIP) assays, which revealed that TGFβ increased the binding of Smad3 or Smad4 to E1 and E4 of PRL-3 promoter in FET cells (Fig. 3C). Furthermore, mutations in E1, E4 or both regions of PRL-3 promoter reversed the inhibitory effect of TGFβ on PRL-3 promoter activity (Fig. 3D). These results demonstrate that Smad3/4 directly binds to E1 and E4 of PRL-3 promoter to inhibit its activity. indicating that PRL-3 also protects FET cells from UV-induced apoptosis.

PRL-3 promotes cell survival under GFDS
Because PI3K/AKT is a major survival pathway in colon cancer cells, we next determined the effect of PRL-3 expression on AKT activation. Phosphorylation of AKT at Thr308 and Ser473 leads to activation of its kinase activity (29). As shown in Figure   4C,

Knockdown of PRL-3 sensitizes colon cancer cells to GFDS-induced apoptosis.
To further define the role of PRL-3 in cell survival, we used an RNA interference approach to knock down expression of PRL-3. Stable transfection of a PRL-3 shRNA vector into FET and GEO cells led to a significant reduction of PRL-3 expression in these cells (Fig. 5A). As a result, there was increased apoptosis in PRL-3 knockdown cells (shPRL3) relative to control cells (pSR) under GFDS (Fig. 5B, Fig.   S5D). These results were confirmed by DNA fragmentation assays (Fig. 5C). These data demonstrate that down-regulation of PRL-3 expression sensitizes colon cancer cells to GFDS-induced apoptosis. In addition, knockdown of PRL-3 expression accelerated decay of AKT phosphorylation under GFDS (Fig. 5B, Figs. S7C & S7D), providing further evidence that PRL-3 regulates AKT signaling.
To test the specificity of PRL-3 shRNA, we ectopically expressed a mutant PRL-3 cDNA in GEO shPRL3 cells (designated GEO resmut) to determine whether it could rescue the phenotypes caused by knockdown of PRL-3 expression. The mutant PRL-3 cDNA harbored a silent mutation in the region targeted by the PRL-3 shRNA so that it is resistant to silencing. Figure 5D showed that PRL-3 expression was restored in GEO resmut cells as compared to GEO shPRL3 cells (left panel). As a result, GEO resmut cells displayed a higher level of AKT phosphorylation under GFDS and were more resistant to GFDS induced apoptosis than GEO shPRL3 cells (Fig. 5D, right panel). DNA fragmentation assays confirmed that re-expression of PRL-3 in GEO shPRL-3 cells decreased GFDS-induced apoptosis to the level that is similar to that of GEO control cells Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.
Copyright © 2010 American Association for Cancer Research (Fig. 5C, right panel). Taken together, these results indicate that PRL-3 activates PI3K/AKT survival signaling and enhances stress resistance in colon cancer cells.  (26).

PRL-3 mediates metastasis of colon cancer cells in an orthotopic
In vivo studies showed that animals implanted with xenografts formed by GEO control, GEO shPRL3 or GEO PRL-3 cells demonstrated 100% primary growth at the site of implantation with clear invasion of the bowel on histological evaluation (Fig. 6A & 6C). In addition, sizes of primary tumors were very similar in animals bearing different types of cells (Fig. 6A). However, compared to control cells, orthotopic implantation of GEO PRL-3 cells gave rise to significantly increased metastatic localization to the liver, whereas implantation of GEO shRPL3 cells in the same model resulted in markedly diminished liver metastasis (Fig. 6B). Moreover, fluorescence imaging of explanted liver showed a remarkable increase in lesion size of liver metastases in the animals implanted with GEO PRL-3 cells relative to control animals (Fig. 6B). The livers of each animal were serially sectioned at 1 mm intervals and the presence or absence of metastatic disease was confirmed by microscopic histological analysis (Fig. 6C). These results indicate that PRL-3 is necessary and sufficient for metastatic colonization in the liver. To

Quantitation of cleaved caspase 3 from western blots analyses is presented in
Supplemental Figure S5C.  Methods. The data are presented as the mean ± SD. *P < 0.001; **P < 0.005.