Identification of STAT3 as a substrate of receptor protein tyrosine phosphatase T

STAT3 Is a Substrate of PTPRT.

To identify potential substrates of PPTRT, we generated two ecdysone-inducible HEK293T cell lines: one expressing the intracellular part containing the two phosphatase domains of PTPRT and the other expressing the extracellular portion of the protein. We used the proteomic approach developed by Rush et al. (9) to globally profile tyrosine phosphopeptides in the two cell lines and parental HEK293T cells (see Materials and Methods). Phosphopeptides derived from five different proteins were present in parental HEK293T cells and cells expressing the extracellular portion of PTPRT but not in cells expressing the intracellular portion. We chose STAT3 of the five candidate substrates for further study, because STAT3 was known to be involved in tumorigenesis.

Importantly, the phosphorylation site identified in our proteomic profiling was Y705, the residue known to be critical for activation of STAT3 signal transduction. To confirm that PTPRT regulates phosphorylation of Y705, lysates from the inducible cell lines were probed with antibodies specific for the phosphorlyated form of Y705. As shown in Fig. 1A, STAT3 phosphorylation was reduced ≈5-fold by induction of the phosphatase domain PTPRT but was unaffected by induction of the extracellular domain. Expression of full-length PTPRT also led to STAT3 dephosphorylation of Y705 in HEK293T cells (Fig. 1B). To test whether PTPRT regulates STAT3 phosphorylation under physiological conditions, we knocked down PTPRT expression by ≈80% in MCF-7 cells by using siRNA specifically against PTPRT. As shown in Fig. 1C, STAT3 phosphorylation increased ≈3-fold in cells transfected with PTPRT siRNA compared with cells transfected with a scrambled siRNA control.

Although the initiators of STAT3 signal transduction in some cell types, particularly of hematopoietic lineage, are well known, STAT3 transduction has not been studied in depth in colorectal epithelial cells. In particular, ligands that trigger STAT3 activation in such cells were not known. To address this issue, we used the CRC cell line HCT116 to assess the effects of cytokines and growth factors that were known to induce STAT3 activation in other cell types (see Materials and Methods). Among the factors examined, the only one that activated STAT3 was IL-6. Because IL-6 plays a direct role in promoting the growth of CRC cells (10, 11), it was important to determine whether PTPRT regulates STAT3 activation in CRC cells stimulated with IL-6. As shown in Fig. 1D, expression of PTPRT reduced STAT3 activation by ≈3-, 8-, and 10-fold that of GFP controls after 60, 120, and 180 min of IL-6 stimulation, respectively. Note that only the phosphorylation status of STAT3, not its total levels, was affected by PTPRT, thereby excluding protein degradation as a cause of these effects. Other controls were provided by STAT1 and STAT5. No STAT1 protein could be detected in these cells, with or without exposure to IL-6, and STAT5 phosphorylation was not altered by IL-6 or PTPRT (Fig. 1E). The activation of STAT3 by IL-6 was not confined to HCT116, as three additional CRC cell lines each showed dramatic induction of phosphorlyated STAT3 by IL-6 without any change in the level of total STAT3 proteins (Fig. 1 F–I).

STAT3 Is a Direct Substrate of PTPRT.

We have demonstrated unambiguously that PTPRT regulates STAT3 phosphorylation, but did not prove that the dephosphorylation was directly mediated by PTPRT rather than through another phosphatase downstream of PTRPT. To address this point, we used a substrate-trapping assay modified from that described for investigation of an unrelated phosphatase (PTP1B) (12). Three GST fusion proteins were constructed for this purpose. All contained both phosphatase domains of PTPRT, but two of them (TRAP-1 and TRAP-2) contained mutations predicted to result in substrate trapping (D1074A in TRAP-1 and C1106S and D1074A in TRAP-2). The recombinant GST fusion proteins were expressed in Escherichia coli, and equal amounts of each were attached to beads (Fig. 2A). The beads were then incubated with cell lysates from SW480 CRC cells. The phosphorylated form of STAT3 was specifically bound by the two substrate trapping mutants but not by the WT or control GST proteins. In contrast, STAT5 was not bound by the substrate trapping mutants (Fig. 2B). Similar results were also observed with cell lysates from HEK293T cells and HT29 CRC cells (Fig. 2C). Furthermore, to test whether PTPRT directly dephosphorylates pSTAT3 in vitro, we incubated pSTAT3 that immunoprecipitated from HEK293T cells as phosphatase substrates with equal amounts of WT, TRAP-2 mutant, or GST proteins with or without phosphatase inhibitor Na₃VO₄. STAT3 was dephosphorylated by WT but not the phosphatase dead mutant (TRAP-2) or GST alone (Fig. 2D). This activity was also inhibited by Na₃VO₄, indicating that it was phosphatase dependent (Fig. 2D).

PTPRT Regulates STAT3 Activity.

Phosphorylated STAT3 is thought to translocate to the nucleus and activate transcription of its target genes, including Bcl-XL and SOCS3 (4). To determine whether dephosphorylation of STAT3 by PTPRT affects expression of such target genes, CRC cells were exposed to IL-6 and then treated with the PTPRT adenovirus. As shown in Fig. 3A, Bcl-XL and SOCS3 protein expression was induced by IL-6 in HCT116 CRC cells, as predicted. Moreover, their expression levels were down-regulated >4-fold by PTPRT.

PTPRT consists of an extracelluar domain, a juxtamembrane region, and two phosphatase domains called D1 and D2 (Fig. 3B). To determine which of the phosphatase domains is required for STAT3 dephosphorylation, adenoviruses expressing PTPRT devoid of the second domain (Del D2) or devoid of both domains (Del D1D2) were generated (Fig. 3B). Each of these PTPRT proteins contained a V5 epitope tag at its C terminus to allow comparison of PTPRT expression levels after infection of HCT116 CRC cells. Expression of the WT PTPRT led to a substantial reduction of phosphoryation of pSTAT3 (Fig. 3C) and an equivalent reduction of expression of the STAT3 targets Bcl-XL and SOCS3 (Fig. 3D). It is generally believed that the second phosphatase domains (D2) of receptor PTPs do not have functionally significant phosphatase activity (13). However, the results in Fig. 3 C and D show that deletion of D2 was associated with a decrease of PTPRT activity compared with WT. This reduction could result from either the loss of an intrinsic enzymatic activity of D2, the disruption of the functional communication between D2 and D1 (14, 15), or a perturbation of the structure of the PTPRT protein upon deletion of its D2 domain.

PTPRT Regulates STAT3 Cellular Localization.

Phosphosphorylated STAT3 is known to dimerize and translocate from the cytoplasm to the nucleus before binding to the promoter of its target genes. To examine the effect of PTPRT on the subcellular localization of STAT3, we examined the cells described in the above experiment with immunofluorescence. Note that the adenoviruses used for these experiments were constructed with the AdEasy system (16), so each coexpressed GFP, allowing distinction between infected and noninfected cells. STAT3 proteins translocated into nuclei after IL-6 stimulation in cells infected with the control GFP virus (Fig. 4). However, STAT3 staining remained diffuse, throughout the cytoplasm, after expression of WT PTPRT, whether or not the cells were treated with IL-6. In contrast, there was no effect on STAT3 translocation after IL-6 stimulation in cells expressing the PTPRT devoid of both its phosphatase domains (Fig. 4).

Regulation of Y705 phosphorylation is critical for STAT3 activation (6). Although the kinases that phosphorylate Y705 of STAT3 have been studied extensively, the phosphatases that dephosphorylate this critical residue have not been clearly defined. It has been reported that the T cell PTP (TC-PTP) can regulate STAT3 phosphorylation (17), but it was not clear whether TC-PTP specifically dephosphorylates the pY705 residue. Most recently, the low molecular weight-dual specificity phosphatase (LMW-DSP2) has been shown to regulate Y705 phosphorylation of STAT3 (18). However, no evidence indicates that STAT3 is a direct substrate of LMW-DSP2. Furthermore, neither of the two previously studied phosphatases plays a role in epithelial cell growth. From the results of the current study, it is clear not only that STAT3 is a direct substrate of PTPRT but also that PTPRT specifically regulates Y705 phosphorylation of STAT3, the ability of STAT3 to transcriptionally activate its target genes and the subcellular localization of STAT3.

How a membrane-localized PTPRT gains access and dephosphorylates STAT3 is an interesting question raised by this study. We propose that PTPRT dephosphorylates pSTAT3 through three possible mechanisms. One is that dephosphorylation of pSTAT3 by PTPRT occurs at the time when STAT3 is activated by receptor-associated kinases, such as Jaks and SRC. The second is that PTPRT dephosphorylates the pSTAT3 pool by taking advantage of the STAT3 shuttling mechanism. A recent study has indicated that STAT3 constitutively shuttles between the cytoplasm and nucleus (19). Finally, PTPRT, like the notch proteins (20), could be cleaved by a protease after activation and the intracellular portion containing the phosphatase domains could translocate into nucleus and dephosphorylate STAT3. Further study is needed to define the mechanism through which PTPRT gains access to pSTAT3.

In summary, our data suggest that the STAT3 pathway may play an important role in the growth of CRC cells. Moreover, our results have obvious therapeutic implications, as one could imagine inhibiting CRC growth through inhibition of STAT3 tyrosine phosphorylation. Finally, the proteomic approach described in this study may be applicable to the identification of substrates of other PTPs.

Cell Lines.

HCT116, SW480, HT29, DLD1, MCF-7, and HEK 293T cells were obtained from the American Type Culture Collection (Manassas, VA). CRC cells were maintained in McCoy 5A media plus 10% FBS. HEK293T and MCF-7 cells were maintained in DMEM media plus 10% FBS.

Establishment of Cell Lines Stably Expressing PTPRT Fragments.

The DNA sequences encoding the extracellular or intracellular fragment of PTPRT (see Results) were cloned into the mammalian expression vector pMZI such that each polypeptide carried a tandem tag consisting of a calmodulin binding peptide and three consecutive FLAG tags at its C terminus (21). The pMZI plasmid was a kind gift of J. Greenblatt (University of Toronto, Toronto, ON, Canada). The pVgRXR plasmid was used to express the ecdysone receptor. HEK 93T cells were cotransfected with pVgRxR and pMZI plasmid at a 1:1 ratio and selected in the presence of 0.4 mg/ml of geneticin and 0.2 mg/ml of zeocin. To induce expression of PTPRT, the cell lines were treated with 3 μM ponasterone A for 15 h and PTPRT expression was assessed via Western blotting using an anti-FLAG antibody. Individual clones expressing equivalent amounts of PTPRT polypeptides were chosen for further analysis (Fig. 1A).

Profiling pY Peptides.

Cell lines were treated with 3 μM ponasterone A for 15 h and lysates were prepared under denaturing conditions in the presence of phosphatase inhibitors [20 mM Hepes (pH 8.0), 9 M urea, 1 mM sodium vanadate]. After trypic digestion, peptides were concentrated and partitioned into three fractions by reversed-phase solid-phase extraction. The phosphopeptides from each fraction were bound to agarose beads conjugated with the phosphotyrosine-specific antibody pTyr-100. After thorough washing, peptides were eluted from the immobilized antibody with dilute acid and analyzed by nanoflow liquid chromatography–tandem MS with an ion trap mass spectrometer. Lists of credible phosphopeptide sequence assignments were assembled. Peptides from five proteins were found to be present in parental HEK293T cells expressing the extracellular part of PTPRT but not in cells expressing the intracellular, enzymatically active part. The five proteins and corresponding peptides were: STAT3 (YCRPESQEHPEADPGAAPyLK), Claspin (ESALNLPyHMPENK), Paxillin (FIHQQPQSSSPVyGSSAK), Vimentin (SLyASSPGGVYATR), and LOC439965 (HNVyIHVESK).

Adenoviruses.

Viruses expressing WT or deleted forms of PTPRT were constructed by using the AdEasy system (16). In brief, a fragment containing a human PTPRT cDNA was fused in-frame with a C-terminal V5 tag in the pcDNA3.1/V5-His vector. After recombination with the pAdEasy-1 vector, high-titer viruses were generated in 293 T cells. Viruses were purified by CsCl₂ gradient centrifugation at 32,000 rpm (SW41 rotor; Thermo Scientific, Waltham, MA), and titers were assayed through measurement of the number of GFP-positive cells after infection of HEK293 cells.

Stimulation of CRC Cells with Growth Factors and Cytokines.

HCT116 cells were starved for 15 h and stimulated with either IL-6 (10 ng/ml), EGF (200 ng/ml), VEGF (10 ng/ml), and PDGF (10 ng/ml) for 10, 30, or 60 min. Western blots were performed to detect pSTAT3.

Western Blot Analysis.

Cells were lysed in RIPA buffer with complete protease inhibitor mixture and phosphatase inhibitors (50 mM Tris·HCl, pH 8.0/0.5% Triton X-100/0.25% sodium deoxycholate/150 mM sodium chloride/1 mM EDTA/1 mM sodium orthovnadate/50 mM NaF/80 μM β-glyerophosphate/20 mM sodium pyrophosphate). Western blots were performed essentially as described (22). Antibodies used included anti-FLAG antibody (M2; Sigma, St. Louis, MO); anti-pY705 STAT3 antibody, anti-STAT3 antibody, anti-pSTAT5 antibody and anti-STAT5 antibody (Cell Signaling Technology, Danvers, MA); anti-BCL-XL antibody and anti-SOCS3 antibody (Santa Cruz Biotechnology, Santa Cruz, CA).

Knockdown of PTPRT Expression by siRNA.

MCF-7 cells were transfected with 40 nM of siRNA against PTPRT (sense GGAGGAGCGACUUCAGAAGUCACGG and antisense CCGUGACUUCUGAAGUCGCUCCUCCUU) by using the Lipofectamine RNAiMAX transfection reagent (Invitrogen, Carlsbad, CA). Knockdown of PTPRT expression was examined by RT-PCR with primers 5′-TCCGGTACATGGCCCACAGAACGTG-3′ and 5′-GCCTTGTAGTTGATCTCGTAGAGCGTG-3′. The transfected cells were serum-starved overnight 24 h posttransfection and then treated with IL-6 for 2 h. Cell lysates were made as described above.

Construction, Expression, and Purification of GST Recombinant Proteins.

The DNA fragment coding for the two phosphatase domains of PTPRT was cloned in-frame with GST into pGEX6p-1 plasmid. Substrate trapping mutants, D1074A or C1106S and D1074A doubt mutant, were made by site-directed mutagenesis as described (2). The WT and substrate trap mutants of PTPRT were expressed in E. coli BL21(DE3) cells. Cells were grown to an A₆₀₀ = 0.7 and induced with 1 mM isopropyl 1-thio-β-d-galactopyranoside at 37°C for 1 h. The harvested cell pellet was lysed in resuspension buffer (0.1 M NaCl/10 mM Tris·HCl, pH 8.0/1 mM EDTA/1% Triton X-100/1 mM PMSF/1 mM DTT/1× Complete protease inhibitor mixture) and sonicated. GST fusion proteins were isolated from the cleared supernatant by using glutathione Sepharose 4B beads.

Phosphatase Substrate Trapping Assay.

Substrate trapping was performed as described (12) and modified as follows. Ten million cells were treated with 100 μM pervanadate for 30 min and collected by centrifugation. The cell pellet was lysed with 1 ml of lysis buffer (25 mM Hepes, pH 7.4/150 mM NaCl/1% Nonidet P-40/1× Complete protease inhibitor mixture/1 mM EDTA/1 mM Benzmidine), treated with 5 mM iodoacetic acid on ice for 5 min, neutralized by addition of 10 mM DTT for 15 min, and subjected to centrifugation at 16,000 × g for 30 min to remove debris. GST-PTPRT bound beads were incubated with this lysate at 4°C for 1 h. The beads were pelleted and washed three times for 5 min with lysis buffer supplemented with 1 mM DTT. The beads were then boiled and aliquots analyzed by SDS/PAGE and Western blotting.

In Vitro Phosphatase Assay.

HEK293T cells overexpressing FLAG-tagged STAT3 proteins were treated with 50 μM pervanadate for 30 min and lysed in RIPA buffer. STAT3 proteins were immunoprecipitated with anti-FLAG antibody-conjugated agarose beads. The immune complexes were washed twice in wash buffer (150 mM NaCl/50 mM Tris·HCl pH 7.4/5 mM EDTA/1% Nonidet P-40) with the phosphatase inhibitors (10 mM NaF and 2 mM Na₃VO₄), twice in the same buffer without the phosphatase inhibitors, once in ST buffer (150 mM NaCl/50 mM Tris·HCl pH 7.4) and once in phosphatase assay buffer (150 mM NaCl/50 mM Tris·HCl/5 mM DTT). STAT3 immunocomplexes were then incubated with equal amounts of either GST alone, GST-PTPRT, or the TRAP-2 mutant in the absence or presence of 10 mM Na₃VO₄ at 37°C for 45 min. Western blots were performed to quantitate pSTAT3.

Immunofluorescent Staining.

HCT116 cells were seeded on glass coverslips and grown to 50% confluence. Cells were then infected with virus for 6 h and serum-starved for 18 h. A subset of virus-infected cells was treated with 10 ng/ml of IL-6 for 30 min following fixation with 4% paraformaldehyde for 30 min at room temperature. The fixed cells were permeabilized with 0.2% Triton X-100 at room temperature for 5 min and then blocked with Image-iT FX signal enhancer (Invitrogen) at room temperature for 30 min. Immunofluorescent staining was performed with anti-STAT3 antibody (Santa Cruz Biotechnology) and the Alexa 594-conjugated anti-rabbit secondary antibody (Invitrogen). Nuclei were stained with DAPI (1 μg/ml) at room temperature for 20 min. Images were captured with a Nikon (Tokyo, Japan) fluorescence microscope.