Physical and functional interactions between STAP-2/BKS and STAT5

Title Physical and functional interactions between STAP-2/BKS and STAT5. Author(s) Sekine, Yuichi; Yamamoto, Tetsuya; Yumioka, Taro; Sugiyama, Kenji; Tsuji, Satoshi; Oritani, Kenji; Shimoda, Kazuya; Minoguchi, Mayu; Yoshimura, Akihiko; Matsuda, Tadashi Citation Journal of Biological Chemistry, 280(9): 8188-8196 Issue Date 2005-03-04 Doc URL http://hdl.handle.net/2115/28113 Rights Copyright © 2005 by the American Society for Biochemistry and Molecular Biology Type article (author version) File Information JBC280-9.pdf


INTRODUCTION
Cytokine signaling is predominatly activated through the Janus kinase (Jak) / signal transducers and activators of transcription (STAT) pathway (1,2,3). The cytokine signal is initiated when Jak is transphosphorylated in response to a receptor juxaposition upon ligand binding. The activated Jak subsequently phosphorylates a number of substrates including members of the STAT protein family.
The regulation of Jaks and STATs is important for the control of cytokine signaling. Because of the critical role of cytokines in mediating inflammation and immunity, it could be proposed that the upregulated activation of Jaks could contribute to hematopoietic disorders, autoimmunity, and inflammatory diseases. In these regards, recent studies have identified several molecules that regulate cytokine signaling. Suppressors of cytokine signaling (SOCS), cytokine-inducible SH2containing protein (CIS) and protein inhibitor of activated STAT (PIAS) are a novel Src homology 2 (SH2)-containing protein family, which is induced by various cytokines suppress Jak-STAT signaling through a direct interaction with Jak kinases or cytokine receptors (18,19). CIS is involved in regulating cytokine signal transduction mediated by STAT5 (18,20,21,22) through binding to the tyrosine-phosphorylated IL-3 and EPO receptors causing a negative regulation of their signals (20). CIS transgenic mice exhibited a phenotype similar to STAT5 deficient mice, supporting its importance in cytokine signaling by directly influencing the STAT5 signal (22).
PIAS family proteins also inhibit DNA binding activity of activated STAT and subsequent gene expression (23). Regulation of cytokine signaling has also been shown to be controlled by several protein tyrosine phosphatases (PTPs) (24,25,26). The mechanisms that terminate or downmodulate the Jak-STAT pathway are not fully understood. However, it has also been suggested that specific regulation of cytokine signaling may involve additional SH2-containing molecules.
Recently, we have cloned a novel adaptor molecule, STAP-2 (Signal Transducing Adaptor Protein-2) as a c-fms interacting protein (27) and STAP-1 as a c-kit interacting protein (28). STAP-2 and STAP-1 contain an N-terminal Pleckstrin homology (PH) and a region distantly related to the Src homology 2 (SH2) domain (overall 33% amino acid identity) (28). However, STAP-2 has a C-terminal proline-rich region that is not present in STAP-1. The N-terminal PH domain of STAP-2 and STAP-1 shared 36% identity and 58% similarity. The central region of STAP-2 is distantly related to the SH2 domain. This region of STAP-2 shared 40% sequence identity with that of STAP-1 and 29% sequence identity with the SH2 domain of human PLCg2 (27,28). Human STAP-2 is identical to a recently cloned adaptor molecule, BKS, a substrate of BRK (breast tumor kinase) tyrosine kinase (27). STAP-1 has been shown to have hematopoietic specific expression and associate with STAT5 (28). STAP-1 was also identified as a Tec-interacting protein, which is tyrosine phosphorylated in response to B-cell receptor (BCR) stimulation and termed as BRDG1 (BCR Downstream Signaling 1) (29). Unlike STAT1, STAP-2/BKS is expressed in a variety of tissues and its C-terminal region contains the proline-rich, tyrosine phosphorylation motifs and a YXXQ motif. We previously showed STAP-2/BKS expression was strongly induced in hepatocytes in response to lipopolysaccharide (LPS) or inflammatory cytokines like interleukin-6 (IL-6), while its expression of myeloid cells is constitutive.
In the present study, we address the involvement of STAP-2/BKS with the STAT5-mediated signaling. We demonstrate STAP-2/BKS physically and functionally interacted with STAT5.
Furthermore, thymocytes from STAP-2/BKS-deficient mice showed an enhanced IL-2-dependent growth. These results indicate that STAP-2/BKS is a negative modulator in the STAT5-mediated signaling.
established as described previously (35) and maintained in the above medium in the presence of G418 (1mg/ml). Human T cell lymphoma, HUT78 was maintained in RPMI1640 medium supplemented with 10% (FCS). Human embryonic kidney carcinoma cell line, 293T, was maintained in DMEM containing 10% FCS and transfected by the standard calcium precipitation protocol (36). The cells were harvested 48 hrs after transfection and lysed in 100 ml of PicaGene Reporter Lysis Buffer (Toyo Ink, Tokyo, Japan) and assayed for luciferase and b-galactosidase activities according to the manufacturer's instructions. Luciferase activities were normalized to the b-galactosidase activities. Three or more independent experiments were carried out for each assay.

Immunoprecipitation and immunoblotting
The immunoprecipitation and Western blotting assays were performed as described previously (36).
The cells were harvested and lysed in lysis buffer (50 mM Tris-HCl, pH 7.4, 0.15 M NaCl, containing 1% NP-40, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride and 10 mg/ml each of aprotinin, pepstatin and leupeptin). The immunoprecipitates from cell lysates were resolved on SDS-PAGE and transferred to Immobilon filter (Millipore; Bedford, MA). The filters were then immunoblotted with the respective antibody. Immunoreactive proteins were visualized using an enhanced chemiluminescence detection system (Amersham Pharmacia Biotech).

Indirect Immunofluorescence
Monkey COS7 or human HeLa cells were maintained in DMEM containing 10% FCS transfected with FLAG-STAT5a and Myc-STAP-2/BKS together with Jak2WT or Jak2KE, EPOR by the calcium phosphate precipitation protocol (37). Forty-eight hrs after transfection, cells were fixed with a solution containing 4% paraformaldehyde and reacted with anti-FLAG antibody or anti-Myc antibody. The cells were then reacted with FITC-conjugated anti-rabbit IgG or rhodamineconjugated anti-mouse IgG (CHEMICON, Temecula, CA) and observed under a confocal laser fluorescent microscope. Images were obtained by using a Zeiss LSM 510 laser scanning microscope with an Apochromat x63/1.4 oil immersion objective and x4 zoom.

Association of STAP-2/BKS with STAT5 in vivo
Studies examining the members of the family of STAP proteins have suggested that association with the STAT family of latent transcriptional factors may have regulational significance (27,28). In these regards, STAP-2/BKS association with STAT3 up-regulated the ability of STAT3 to induce transcription (27). The observation that STAP-1 interacts with STAT5 (28)  To examine the physiological interaction between STAP-2/BKS and STAT5, human T cell lymphoma, HUT78, which endogenously expressed both proteins, were employed. HUT78 cells were unstimulated or stimulated with IL-2 for 15 min and the cells were lysed and immunoprecipitated with control antibody or anti-STAP-2/BKS antibody. The same lysates were also immunoprecipitated with anti-STAT5 antibody. In HUT78 cells, endogenous STAT5 was well tyrosine phosphorylated by IL-2 stimulation (Fig. 1D, lower panel). Furthermore, the immunoprecipitate with anti-STAT5 antibody contained STAP-2/BKS ( Fig. 1D; upper panel). It is worthy to note, the amounts of STAP-2/BKS bound to STAT5 decreased as a result of STAT 5 activation/phosphorylation in response to IL-2. The dissociation of STAP-2/BKS from activated STAT5 was also observed in 293T cells overexpressing Jak2. Increasing amounts of tyrosine phosphorylated STAT5 resulted in a decrease of the STAP-2/BKS bound to STAT5 (data not shown). Therefore, STAP-2/BKS can directly associate with STAT5.
However, their interaction is reduced when STAT5 is tyrosine-phosphorylated.

Co-localization of STAP-2/BKS with STAT5
We next examined co-localization of STAP-2/BKS with STAT5a in COS7 cells. To activate STAT5a in COS cells, we transfected expression vectors for wild-type Jak2 (Jak2WT) or Jak2KE, an inactive form of Jak2. COS7 cells were transfected with FLAG-tagged STAT5a and Myc-tagged STAP-2/BKS together with Jak2KE or Jak2WT. Forty-eight hrs after transfection, the transfectants were fixed and reacted with rabbit anti-FLAG polyclonal antibody or mouse anti-Myc monoclonal antibody, and visualized with rhodamine-conjugated anti-rabbit antibody or fluorescein isothiocyanateconjugated anti-mouse antibody. As shown in Fig. 2A, STAT5a remained localized to the cytoplasm in the presence of Jak2 KE, whereas STAT5a translocated into nucleus in the presence of Jak WT. STAP-2/BKS was observed in both the cytoplasm and nucleus in the Jak2WT-and Jak2KEtransfected cells. We also examined co-localization of STAP-2/BKS with STAT5 after EPO stimulation in COS7 cells. COS7 cells were transfected with FLAG-tagged STAT5a and Myc-tagged STAP-2/BKS together with an expression vector for EPO receptor. Forty-eight hrs after transfection, the cells were stimulated with EPO for 30 min and fixed, and then stained with rabbit anti-FLAG polyclonal antibody or mouse anti-Myc monoclonal antibody, and visualized with rhodamineconjugated anti-rabbit antibody or fluorescein isothiocyanate-conjugated anti-mouse antibody. As shown in Fig. 2B, only small amounts of STAP-2/BKS translocated into nucleus, while most of STAT5 existed in nucleus after EPO-stimulation. Therefore, unphosphorylated STAT5 co-localized with STAP-2/BKS in cytoplasm, but phosphorylated STAT5 translocated into nucleus without STAP-2/BKS-association.

Molecular mechanisms of STAP-2/BKS interactions with STAT5
We also determined the interacting domain of STAP-2/BKS on STAT5 using a series of STAT5a or STAT5b deletion mutants. As depicted for STAT5 in Fig. 3A or STAT5b were transiently expressed in 293T cells. The binding potential of these proteins with FLAG-tagged STAT5a or STAT5b was examined by immunoprecipitation with anti-FLAG antibody followed by western blotting with anti-GST. As shown in Fig. 3H, both GST-STAP-2 PH and GST-STAP-2 SH2 interacted with STAT5a or STAT5b. Therefore, both PH and SH2 domains of STAP-2/BKS can bind to STAT5, but the interaction of the PH domain was stronger than that of the SH2 domain.

STAP-2/BKS suppresses EPO-induced tyrosine-phosphorylation and transcriptional activation of STAT5
To elucidate whether the interactions between STAP-2/BKS and STAT5 have functional significance, the interaction was examined in EPO-stimulated or unstimulated 293T cells. 293T cells expressing exogenous EPO receptor were transfected with or without STAP-2/BKS and then activated with EPO. The stimulated cells were harvested and immunoprecipitated with anti-STAT5 antibody followed by western blotting. EPO-induced tyrosine-phosphorylation of STAT5 decreased in the cells transfected with STAP-2/BKS (Fig. 4A), suggesting that STAP-2/BKS interaction with STAT5 influences STAT5 activation. To address whether this reduction in tyrosine phosphorylation correlated to transcriptional regulation, a transient reporter assay was performed using STAT5 reporter constructs, STAT5-LUC or Casein-LUC. 293T cells were transfected with STAT5-LUC or Casein-LUC and/or increasing amounts of STAP-2/BKS. Thirty-six hrs after transfection, the cells were stimulated with EPO. As shown in Fig, 4B, STAP-2/BKS suppressed the STAT5 transcriptional activation by EPO in parallel with its expression. Therefore, STAP-2/BKS regulates STAT5 transcriptional activity.

STAP-2/BKS suppresses IL-3-dependent growth and tyrosine-phosphorylation of STAT5 in Ba/F3 cells
To further assess the functional relevance between STAP-2/BKS and STAT5, we established the stable transformants expressing wild-type STAP-2/BKS or its deletion mutants in an IL-3-dependent murine pro-B cell line, Ba/F3 (Fig. 5A) in which STAT5 has been shown to exert pleiotropic functions regulating cell growth, differentiation and apoptosis (39,40). We examined the effects of wild-type STAP-2/BKS or its mutants on IL-3-dependent cell growth. As shown in Fig. 5B, all transformants expressing wild-type STAP-2/BKS and its mutants showed significant reduction in cell growth Predicting that STAP-2 deficient mice should have a greater response to growth factor because of the reduced negative regulation imposed by STAP-2/BKS, we examined the effect of IL-2/STAT5mediated T-cell growth in the STAP-2/BKS deficient thymocytes. STAP-2/BKS deficient thymocytes showed the enhanced response to IL-2 (Fig. 6A). However, PHA stimulation, which is STAT5independent, induced the similar growth response in the wild-type and STAP-2/BKS deficient thymocytes (Fig. 6B). Together, these results demonstrate a role for STAP-2/BKS in negatively regulating STAT5-signaling in vivo.

DISCUSSION
STAP-2/BKS. STAP-2/BKS is expressed in a variety of tissues, and it has the C-terminal prolinerich and tyrosine phosphorylation motifs in an addition to the motifs defining the STAP family protein.
In our previous study, we also demonstrated STAP-2/BKS interacted with STAT3 through its YXXQ motif and enhanced STAT3 transcriptional activity (27).  3H). These data suggest that STAP-2/BKS contains at least two binding sites for the core fragment of STAT5, and binds to and masks STAT5 protein. The conformational change of the core structure of STAT5 by tyrosine-phophorylation may reduce the binding affinity for STAP-2/BKS. Crystal structural analysis of STAT5-STAP-2/BKS complex will be required to clarify the details of their interactions.
STAT5a-deficient mice showed a marked decreased proliferative response to low concentration of IL-2 (15). There was a much greater decreased proliferation in response to IL-2 in STAT5b-deficient mice than was seen in STAT5a-deficient mice (13). The STAT5a/b double knockout mice showed a profoundly deficiency in peripheral T-cells (17), indicating that STAT5 proteins plays vital role in IL-2 signaling. Recent study using STAT5a/b knockout mice also demonstrated STAT5 is required for embryonic thymocyte production, TCRg gene transcription, and Peyer's patch development (42). In STAP-2/BKS deficient mice, cellularity in the thymus and T-cell development assessed by expression of CD4 and CD8 appeared normal in the STAP-2/BKS-deficient mice (data not shown). Expression levels of CD3 on thymocytes were also indistinguishable between the wild-type and STAP-2/BKSdeficient mice (data not shown). However, STAP-2/BKS deficient thymocytes showed the enhanced response to IL-2. This result suggests that STAP-2/BKS may play a role in the regulation of STAT5 in thymocytes.
Although the molecular mechanism of STAT5 modulation by STAP-2/BKS has not been clarified yet, one of the mechanisms is that STAP-2/BKS simply masks the STAT5 molecule to form a latent STAT5-STAP-2/BKS complex in the absence of ligand stimulation. This model may be supported by our observation that IL-2 stimulation in HUT78 cells can decrease the amounts of STAP-2/BKS bound to STAT5 (Fig. 1C). The other possibility is that STAP-2/BKS recruits other regulatory molecules, such as protein tyrosine phosphatases (PTPs) or ubiquitin ligases. They dephosphorylate tyrosinephosphorylated STAT5 or accelerate the degradation of STAT5 by the ubiquitin-proteasome pathway.
Recently APS, adaptor molecule containing PH and SH2 domains, was shown to be tyrosinephosphorylated by Jak2 at its C-terminal tyrosine residue and interacted with c-Cbl (43).  (46), since these adaptor molecules contain the PH domain at the N-terminal region, a phosphotyrosine-binding domain, such as the PTB domain, or a Met-binding domain (MBD), in the middle, and tyrosine phosphorylation sites in the C-terminal region. The most of these molecules expresses ubiquitously. However, the knockout works showed that some of them are indispensable for the functions in a specific tissue or organ. An ubiquitously expressed adaptor molecule, Gab2 is recently shown to be essential in the FceRI-mediated signaling pathway in mast cells (47). Our previous study demonstrated that STAP-2/BKS affected on the FceRI signaling pathway in rat basophilic leukemia cells (34). In the STAP-2/BKS-deficient thymocytes, we could detect a little bit enhanced proliferation compared to the wild-type thymocytes, when we stimulated them with anti-CD3 and anti-CD28 antibodies (Fig. 6B). These results suggest that STAP-2/BKS may have some effects on TCR signaling. However, at the present time, we have not obtained any results on the FceRI signaling in the STAP-2/BKS-deficient mast cells. The STAP-2/BKS deficient mice will also provide a powerful tool for these investigations.

Overexpression of APS in an EPO
In conclusion, our studies describe a novel function for STAP-2/BKS in regulating STAT5mediated signaling. STAP-2/BKS interacted with STAT5 in vivo and suppressed EPO/STAT5mediated transcriptional activation or IL-3-depedent growth in pro-B cells. Furthermore, thymocytes from STAP-2/BKS-deficient mice showed the enhanced proliferation in response to IL-2. We propose that STAP-2/BKS may play a role in modulating the STAT5-mediated signaling in immune cells.    (B) 293T cells in a 12-well plate were transfected with STAT5-LUC (0.4mg) or Casein-LUC (0.4mg) and/or the increasing amounts of STAP-2/BKS together with EPOR (50ng) as indicated.

FIGURE LEGEND
Thirty-six hrs after transfection, the cells were stimulated with EPO (1U/ml) for additional 12 hrs.
The stimulated cells were harvested, and luciferase activities were measured. The results are indicated as fold induction of luciferase activity from triplicate experiments, and the error bars represent the S.D. Total extracts (5%) from cells transfected with the above constructs were blotted with anti-Myc antibody (lower panel) to monitor the expression of wild-type STAP-2/BKS.