CNK1 and other scaffolds for Akt/FoxO signaling

Identification of CNKSR1 as a biomarker for "cold" tumor microenvironment in lung adenocarcinoma: An integrative analysis based on a novel workflow

Background

Therapies targeting PD1/PD-L1 pathway have revolutionized the treatment of lung cancer. However, anti-PD1/PD-L1 therapies have proven beneficial for only a select group of lung adenocarcinoma (LUAD) patients and generally do not work for immuno-cold tumors characterized by a lack of immune cell infiltration. Identifying novel biomarkers is vital to broad therapeutic options for LUAD patients with no response to anti-PD1/PD-L1 immunotherapies.

Methods

Our study has developed a novel strategy to identify a promising biomarker that addresses the limitations of anti-PD1/PD-L1 immunotherapy in treating immunological cold tumors. We exacted LUAD RNA-seq data from the Cancer Genome Atlas database (TCGA). Using several machine learning methods, we identified the candidate biomarker. Based on the expression level of PD-L1 and the identified biomarker, samples were categorized into four groups. We further used ESTIMATE, ssGSEA, and CIBERSORT algorithms to calculate the immune infiltration level of each group. The results were validated in three independent bulk datasets and one scRNA-seq dataset. Immunohistochemistry (IHC) assessments were performed in clinical samples to further evaluate the coexpression of CNKSR1 and PD-L1, and to compare CD8 + T cell infiltration among groups.

Results

After comprehensive analyses, CNKSR1 was identified as a novel promising biomarker for immuno-cold LUAD. CNKSR1 mRNA expression levels exhibited a negative correlation with both PD-L1 mRNA expression and the extent of immune cell infiltration in LUAD. Besides, in contrast to the significant association between the expression of PD-L1 and the majority of other well-established or widely studied immune checkpoint molecules, a mutually exclusive expression pattern is observed between CNKSR1 and these molecules. The aforementioned results were consistent in validation datasets. The prognostic model built based on the CNKSR1 coexpression module also showed robust predictive performance. Additionally, IHC assessments have confirmed that the coexpression of CNKSR1 and PD-L1 is rare in LUAD samples. Notably, LUADs in the high-CNKSR1 group, characterized by high CNKSR1 but low PD- L1 expression, demonstrated reduced infiltration of CD8+ T cells.

Conclusions

In summary, CNKSR1 emerges as a promising biomarker for immune-cold LUADs, and the study into CNKSR1 modulating T-cell infiltration may lead to the identification of compensatory molecules to enhance the effectiveness of current immunotherapy for LUAD.

CNKSR1 serves as a scaffold to activate an EGFR phosphatase via exclusive interaction with RhoB-GTP

CNKSR1 functions as a scaffold protein for activation of an EGFR phosphatase, PTPRH, at the plasma membrane through the exclusive interaction with RhoB-GTP which is constitutively degraded by the CUL3/KCTD10 E3 complex.

Epidermal growth factor receptor (EGFR) and human EGFR 2 (HER2) phosphorylation drives HER2-positive breast cancer cell proliferation. Enforced activation of phosphatases for those receptors could be a therapeutic option for HER2-positive breast cancers. Here, we report that degradation of an endosomal small GTPase, RhoB, by the ubiquitin ligase complex cullin-3 (CUL3)/KCTD10 is essential for both EGFR and HER2 phosphorylation in HER2-positive breast cancer cells. Using human protein arrays produced in a wheat cell-free protein synthesis system, RhoB-GTP, and protein tyrosine phosphatase receptor type H (PTPRH) were identified as interacting proteins of connector enhancer of kinase suppressor of Ras1 (CNKSR1). Mechanistically, constitutive degradation of RhoB, which is mediated by the CUL3/KCTD10 E3 complex, enabled CNKSR1 to interact with PTPRH at the plasma membrane resulting in inactivation of EGFR phosphatase activity. Depletion of CUL3 or KCTD10 led to the accumulation of RhoB-GTP at the plasma membrane followed by its interaction with CNKSR1, which released activated PTPRH from CNKSR1. This study suggests a mechanism of PTPRH activation through the exclusive binding of RhoB-GTP to CNKSR1.

Improving the Dysregulation of FoxO1 Activity Is a Potential Therapy for Alleviating Diabetic Kidney Disease

A substantial proportion of patients with diabetes will develop kidney disease. Diabetic kidney disease (DKD) is one of the most serious complications in diabetic patients and the leading cause of end-stage kidney disease worldwide. Although some mechanisms have been revealed to contribute to the understanding of the pathogenesis of DKD and some drugs currently in use have been shown to be beneficial, prevention and management of DKD remain tricky and challenging. FoxO1 transcriptional factor is a crucial regulator of cellular homeostasis and posttranslational modification is a major mechanism to alter FoxO1 activity. There is increasing evidence that FoxO1 is involved in the regulation of various cellular processes such as stress resistance, autophagy, cell cycle arrest, and apoptosis, thereby playing an important role in the pathogenesis of DKD. Improving the dysregulation of FoxO1 activity by natural compounds, synthetic drugs, or manipulation of gene expression may attenuate renal cell injury and kidney lesion in the cells cultured under a high-glucose environment and in diabetic animal models. The available data imply that FoxO1 may be a potential clinical target for the prevention and treatment of DKD.

The effects of SIRT1/FoxO1 on LPS induced INS-1 cells dysfunction

Oxidative stress is one of the key mechanisms of sepsis related organ dysfunction including stress hyperglycemia. Silent mating type information regulation 2 homolog 1 (SIRT1) could regulate glucose metabolism through its deacetylase activity. In this study, we aimed to investigate the role of SIRT1/forkhead box protein 1 (FoxO1) pathway on lipopolysaccharide (LPS) induced INS-1 cells dysfunction from aspects of oxidative stress and apoptosis. After being treated with 1 mg/L LPS together with or without SIRT1 activator resveratrol (RSV) or SIRT1 inhibitor EX527, cell viability, ROS generation, malondialdehyde (MDA), superoxide, insulin secretion, and activity of superoxide dismutase (SOD) in INS-1 cells were measured by specific assays. Protein expression of SIRT1, FoxO1, toll-like receptor 4 (TLR4), and acetylated FoxO1 (ac-FoxO1) were detected by western blot analysis. Nuclear and cytoplasmic protein was extracted respectively to analyze SIRT1 and FoxO1 redistribution. Mitochondrial potentials and apoptosis were detected by flow cytometry or observed under fluorescence microscope. Results showed that LPS decreased cell viability and insulin secretion, increased ROS, MDA, and superoxide generation, whereas inhibited SOD activity and FoxO1 nuclear transportation. Activation of SIRT1 by RSV down-regulated TLR4 expression, SIRT1 and FoxO1 nuclear protein expression increased after RSV pretreatment. Additionally, LPS induced decreased mitochondrial membrane potentials and structural abnormalities, which could be partially reversed by RSV. SIRT1/FoxO1 may be one of potential targets which could resist against LPS-induced INS-1 cells from oxidative stress damage and mitochondrial dysfunction.

Regulation of CNKSR2 protein stability by the HECT E3 ubiquitin ligase Smurf2, and its role in breast cancer progression

BACKGROUND

Smurf2 E3 ubiquitin ligase physically associates with and regulate the stability of distinct cellular protein substrates. The multi-functional scaffold protein Connector enhancer of kinase suppressor of ras 2 (CNKSR2) plays a key role in regulating cell proliferation, and differentiation through multiple receptor tyrosine kinase pathways. The aim of this study was to investigate whether the interaction between Smurf2 and CNKSR2 has any significant role in the post transcriptional regulation of CNKSR2 expression in breast cancer.

METHODS

Here we demonstrate a novel interaction of CNKSR2 with Smurf2 by co-immunoprecipitation, indirect immunofluorescence studies, and surface plasmon resonance (SPR) analysis, which can ubiquitinate, but stabilize CNKSR2 by protecting it from proteasome mediated degradation.

RESULTS

CNKSR2 protein levels were significantly increased upon forced overexpression of Smurf2, indicating the role of Smurf2 in regulating the stability of CNKSR2. Conversely, Smurf2 knockdown resulted in a marked decrease in the protein level expression of CNKSR2 by facilitating enhanced polyubiquitination and proteasomal degradation and reduced the proliferation and clonogenic survival of MDA-MB-231 breast cancer cell lines. Tissue microarray data from 84 patients with various stages of mammary carcinoma, including (in order of increasing malignant potential) normal, usual hyperplasia, fibrocystic changes, fibroadenoma, carcinoma-in-situ, and invasive ductal carcinoma showed a statistically significant association between Smurf2 and CNKSR2 expression, which is also well correlated with the ER, PR, and HER2 status of the tissue samples. A comparatively high expression of Smurf2 and CNKSR2 was observed when the expression of ER and PR was low, and HER2 was high. Consistently, both Smurf2 and CNKSR2 showed an integrated expression in MCF10 breast progression model cell lines.

CONCLUSIONS

Altogether, our findings reveal that Smurf2 is a novel positive regulator of CNKSR2 and suggest that Smurf2-CNKSR2 interaction may serve as a common strategy to control proliferation of human breast cancer cells by modulating CNKSR2 protein stability.

Resveratrol controlled the fate of porcine pancreatic stem cells through the Wnt/β-catenin signaling pathway mediated by Sirt1

Porcine pancreatic stem cells (PSCs) are considered promising transplant materials that may be used to treat diabetes, but some problems, such as insufficient cell number and low differentiation efficiency, should be solved before its clinical application. Resveratrol is a natural polyphenolic compound that can alleviate the complications of diabetes. In this study, we aimed to explore the specific effect of resveratrol on porcine PSCs. We treated porcine PSCs with 10 μM, 25 μM resveratrol to explore the effect of resveratrol on porcine PSCs. We found that 10 μM resveratrol improved the proliferation of porcine PSCs, increased the expression of A-β-catenin (active β-catenin), Pcna, C-Myc, Bcl-2 and sirtuin-1 (Sirt1), and decreased the expression of P53, Caspase3. While 25 μM resveratrol had almost opposite effect compared with 10 μM resveratrol group. The utilization of Dickkopf-related protein 1 (DKK1, Wnt signaling pathway inhibitor) and nicotinamide (Sirt1 inhibitor) suggested that resveratrol regulated cell proliferation by controlling Wnt signaling pathway and this effect was mediated by Sirt1. Our results further revealed that 10 μM resveratrol promoted the formation of β-like cells regulated by Wnt/β-catenin signal pathway. Relatively low-dose resveratrol could improve porcine PSCs fate. It lays theoretical foundation for diabetes treatment with cell transplantation in future.

Membrane localization of acetylated CNK1 mediates a positive feedback on RAF/ERK signaling

Spatiotemporal control is a common mechanism that modulates activity and function of signal transducers in the signaling network. We identified acetylation of CNK1 (connector enhancer of kinase suppressor of Ras-1) as a late step in the activation of CNK1 signaling, accompanied with prolonged stimulation of extracellular signal-regulated kinase (ERK). We identified the acetyltransferase CREB (cyclic adenosine 3',5'-monophosphate response element-binding protein)-binding protein and the deacetylase SIRT2 (sirtuin type 2) as novel binding partners of CNK1, modulating the acetylation state of CNK1. Acetylation of CNK1 at position Lys414 located in the pleckstrin homology domain drives membrane localization of CNK1 in growth factor-stimulated cells. Inhibition of ERK signaling abolishes CNK1 acetylation. Cosmic database search identified CNK1 mutants at position Arg426 near the acetylation site in several human tumor types. These mutants show constitutive acetylation and membrane localization. CNK1 mutants substituting Arg426, the acetylation mimetic mutant CNK1-K414Q, and membrane-anchored CNK1 mutants all interact with the protein kinase CRAF and stimulate ERK-dependent cell proliferation and cell migration. In RAS-transformed cells, CNK1 is acetylated and membrane-bound and drives cell proliferation. Thus, growth factor-stimulated ERK signaling induces CNK1 acetylation, and acetylated CNK1 promotes ERK signaling, demonstrating a novel function of CNK1 as positive feedback regulator of the RAF/MEK (mitogen-activated protein kinase kinase)/ERK pathway. In addition, acetylation of CNK1 is an important step in oncogenic signaling, promoting cell proliferation and migration.

Optogenetic clustering of CNK1 reveals mechanistic insights in RAF and AKT signalling controlling cell fate decisions

Scaffold proteins such as the multidomain protein CNK1 orchestrate the signalling network by integrating and controlling the underlying pathways. Using an optogenetic approach to stimulate CNK1 uncoupled from upstream effectors, we identified selective clusters of CNK1 that either stimulate RAF-MEK-ERK or AKT signalling depending on the light intensity applied. OptoCNK1 implemented in MCF7 cells induces differentiation at low light intensity stimulating ERK activity whereas stimulation of AKT signalling by higher light intensity promotes cell proliferation. CNK1 clustering in response to increasing EGF concentrations revealed that CNK1 binds to RAF correlating with ERK activation at low EGF dose. At higher EGF dose active AKT binds to CNK1 and phosphorylates and inhibits RAF. Knockdown of CNK1 protects CNK1 from this AKT/RAF crosstalk. In C2 skeletal muscle cells CNK1 expression is induced with the onset of differentiation. Hence, AKT-bound CNK1 counteracts ERK stimulation in differentiated but not in proliferating cells. Ectopically expressed CNK1 facilitates C2 cell differentiation and knockdown of CNK1 impaired the transcriptional network underlying C2 cell differentiation. Thus, CNK1 expression, CNK1 clustering and the thereto related differential signalling processes decide on proliferation and differentiation in a cell type- and cell stage-dependent manner by orchestrating AKT and RAF signalling.

AKT-dependent phosphorylation of the SAM domain induces oligomerization and activation of the scaffold protein CNK1

Scaffold proteins are hubs for the coordination of intracellular signaling networks. The scaffold protein CNK1 promotes several signal transduction pathway. Here we demonstrate that sterile motif alpha (SAM) domain-dependent oligomerization of CNK1 stimulates CNK1-mediated signaling in growth factor-stimulated cells. We identified Ser22 located within the SAM domain as AKT-dependent phosphorylation site triggering CNK1 oligomerization. Oligomeric CNK1 increased the affinity for active AKT indicating a positive AKT feedback mechanism. A CNK1 mutant lacking the SAM domain and the phosphorylation-defective mutant CNK1_S22A antagonizes oligomerization and prevents CNK1-driven cell proliferation and matrix metalloproteinase 14 promoter activation. The phosphomimetic mutant CNK1_S22D constitutively oligomerizes and stimulates CNK1 downstream signaling. Searching the COSMIC database revealed Ser22 as putative target for oncogenic activation of CNK1. Like the phosphomimetic mutant CNK1_S22D, the oncogenic mutant CNK1_S22F forms clusters in serum-starved cells comparable to clusters of CNK1 in growth factor-stimulated cells. CNK1 clusters induced by activating Ser22 mutants correlate with enhanced cell invasion and binding to and activation of ADP ribosylation factor 1 associated with tumor formation. Mutational analysis indicate that EGF-triggered phosphorylation of Thr8 within the SAM domain prevents AKT binding and antagonizes CNK1-mediated AKT signaling. Our findings reveal SAM domain-dependent oligomerization by AKT as switch for CNK1 activation.

Pits, a protein interacting with Ttk69 and Sin3A, has links to histone deacetylation

Histone deacetylation plays an important role in transcriptional repression. Previous results showed that the genetic interaction between ttk and rpd3, which encodes a class I histone deacetylase, is required for tll repression. This study investigated the molecular mechanism by which Ttk69 recruits Rpd3. Using yeast two-hybrid screening and datamining, one novel protein was found that weakly interacts with Ttk69 and Sin3A, designated as Protein interacting with Ttk69 and Sin3A (Pits). Pits protein expressed in the early stages of embryos and bound to the region of the tor response element in vivo. Expanded tll expression patterns were observed in embryos lacking maternal pits activity and the expansion was not widened by reducing either maternal ttk or sin3A activity. However, in embryos with simultaneously reduced maternal pits and sin3A activities or maternal pits, sin3A and ttk activities, the proportions of the embryos with expanded tll expression were significantly increased. These results indicate that all three gene activities are involved in tll repression. Level of histone H3 acetylation in the tll proximal region was found to be elevated in embryo with reduced these three gene activities. In conclusion, Ttk69 causes the histone deacetylation-mediated repression of tll via the interaction of Pits and Sin3A.

Differential tyrosine phosphorylation controls the function of CNK1 as a molecular switch in signal transduction

Scaffold proteins are multidomain proteins without enzymatic function that play a central role in coordinating signaling processes. The scaffold protein CNK1 interacts with pathway-specific signaling proteins and thereby regulates these respective pathways. Here, we revealed tyrosine phosphorylation as a critical regulation mechanism to control the function of CNK1. We identified Tyr 26 as a PDGF-induced and, additionally, Tyr 519 and Tyr 665 as SRC-induced tyrosine phosphorylation sites. Phosphomimetic mutants indicate that phosphorylation of Tyr 519 recruits CNK1 to the nucleus and additional phosphorylation of Tyr 26 enables CNK1 to promote SRE-dependent gene expression. Contrary, mutants preventing tyrosine phosphorylation promote matrix metalloproteinase MMP14 promoter activity. CNK1-driven cell proliferation partially depends on its tyrosine phosphorylation. Upon PDGF stimulation, CNK1 is recruited to the plasma membrane mediated by SRC. Knock down of CNK1 prevents PDGF-induced SRE-dependent gene expression, MMP14 promoter activity and cell proliferation. Thus, tyrosine phosphorylation is an important mechanism to control the subcellular localization of CNK1 and its distinct biological functions.

KSR1 is coordinately regulated with Notch signaling and oxidative phosphorylation in thyroid cancer

Kinase suppressor of RAS1 (KSR1) is a scaffold protein implicated in RAS-mediated RAF activation. However, the molecular function of KSR in papillary thyroid cancer (PTC) is unknown. Thus, this study aimed to characterize the role of KSR1 in patients with PTC. qRT-PCR and immunohistochemistry (IHC) revealed inter-tumor heterogeneities in the expression of KSR1 in PTC tissues. Interestingly, BRAFV600E-positive PTC showed higher KSR1 mRNA expression than BRAFV600E-negative PTC (P<0.001). Gene Set Enrichment Analysis (GSEA) using public repositories showed that high KSR1 expression coordinately upregulated Notch signaling (nominal P=0.019, false discovery rate (FDR) q-value=0.165); this finding was supported by GeneNetwork analysis, indicating that KSR1 expression is positively correlated with NOTCH1 expression (ρ=0.677, P=6.15×10(-9)). siRNA against KSR1 (siKSR1) significantly decreased ERK phosphorylation induced by BRAFV600E, resulting in reduced expression of NOTCH1 and HES1, targets of Notch signaling. GSEA revealed that high KSR1 expression was also associated with downregulation of genes related to oxidative phosphorylation (OxPhos). Consistent with this, electron microscopy showed that PTCs with high KSR1 expression exhibited structural defects of the mitochondrial cristae. Furthermore, siKSR1-transfected BCPAP and 8505C cells generated fewer colonies in colony-forming assays. In addition, GSEA showed that high expression of KSR2 and connector enhancer of KSR1 (CNKSR1) also coordinately upregulated Notch signaling (KSR2: nominal P=0.0097, FDR q-value=0.154 and CNKSR1: nominal P<0.0001, FDR q-value=0.00554), and high CNKSR2 was associated with downregulation of the OxPhos gene set (nominal P<0.0001, FDR q-value <0.0001). In conclusion, KSR1 is coordinately regulated with Notch signaling and OxPhos in PTC, because its scaffold function might be required to sustain the proliferative signaling and metabolic remodeling associated with this type of cancer.

Turning off AKT: PHLPP as a drug target

Precise control of the balance between protein phosphorylation, catalyzed by protein kinases, and protein dephosphorylation, catalyzed by protein phosphatases, is essential for cellular homeostasis. Dysregulation of this balance leads to pathophysiological states, driving diseases such as cancer, heart disease, and diabetes. Aberrant phosphorylation of components of the pathways that control cell growth and cell survival are particularly prevalent in cancer. One of the most studied tumor suppressors in these pathways is the lipid phosphatase PTEN (phosphatase and tensin homolog deleted on chromosome ten), which dephosphorylates the lipid second messenger phosphatidylinositol 3,4,5-trisphosphate (PIP3), thus preventing activation of the oncogenic kinase AKT (v-akt murine thymoma viral oncogene homolog). In 2005, the discovery of a family of protein phosphatases whose members directly dephosphorylate and inactivate AKT introduced a new negative regulator of the phosphoinositide 3-kinase (PI3K) oncogenic pathway. Pleckstrin homology domain leucine-rich repeat protein phosphatase (PHLPP) isozymes comprise a novel tumor suppressor family whose two members, PHLPP1 and PHLPP2, are deleted as frequently as PTEN in cancers such as those of the prostate. PHLPP is thus a novel therapeutic target to suppress oncogenic pathways and is a potential candidate biomarker to stratify patients for the appropriate targeted therapeutics. This review discusses the role of PHLPP in terminating AKT signaling and how pharmacological intervention would impact this pathway.

Smurf2 E3 ubiquitin ligase modulates proliferation and invasiveness of breast cancer cells in a CNKSR2 dependent manner

BACKGROUND

Smurf2 is a member of the HECT family of E3 ubiquitin ligases that play important roles in determining the competence of cells to respond to TGF- β/BMP signaling pathway. However, besides TGF-β/BMP pathway, Smurf2 regulates a repertoire of other signaling pathways ranging from planar cell polarity during embryonic development to cell proliferation, migration, differentiation and senescence. Expression of Smurf2 is found to be dysregulated in many cancers including breast cancer. The purpose of the present study is to examine the effect of Smurf2 knockdown on the tumorigenic potential of human breast cancer cells emphasizing more on proliferative signaling pathway.

METHODS

siRNAs targeting different regions of the Smurf2 mRNA were employed to knockdown the expression of Smurf2. The biological effects of synthetic siRNAs on human breast cancer cells were investigated by examining the cell proliferation, migration, invasion, focus formation, anchorage-independent growth, cell cycle arrest, and cell cycle and cell proliferation related protein expressions upon Smurf2 silencing.

RESULTS

Smurf2 silencing in human breast cancer cells resulted in a decreased focus formation potential and clonogenicity as well as in vitro cell migration/invasion capabilities. Moreover, knockdown of Smurf2 suppressed cell proliferation. Cell cycle analysis showed that the anti-proliferative effect of Smurf2 siRNA was mediated by arresting cells in the G0/G1 phase, which was caused by decreased expression of cyclin D1and cdk4, followed by upregulation p21 and p27. Furthermore, we demonstrated that silencing of Smurf2 downregulated the proliferation of breast cancer cells by modulating the PI3K- PTEN-AKT-FoxO3a pathway via the scaffold protein CNKSR2 which is involved in RAS-dependent signaling pathways. The present study provides the first evidence that silencing Smurf2 using synthetic siRNAs can regulate the tumorigenic properties of human breast cancer cells in a CNKSR2 dependent manner.

CONCLUSIONS

Our results therefore suggest a novel relation between Smurf2 and CNKSR2 thereby regulating AKT-dependent cell proliferation and invasion. Owing to the fact that PI3K-AKT signaling is hyperactivated in various human cancers and that Smurf2 also regulates cellular transformation, our results indicate that Smurf2 may serve as a potential molecule for targeted cancer therapy of certain tumour types including breast cancer.

EphrinB1 interacts with CNK1 and promotes cell migration through c-Jun N-terminal kinase (JNK) activation

The Eph receptors and their membrane-bound ligands, ephrins, play important roles in various biological processes such as cell adhesion and movement. The transmembrane ephrinBs transduce reverse signaling in a tyrosine phosphorylation-dependent or -independent, as well as PDZ-dependent manner. Here, we show that ephrinB1 interacts with Connector Enhancer of KSR1 (CNK1) in an EphB receptor-independent manner. In cultured cells, cotransfection of ephrinB1 with CNK1 increases JNK phosphorylation. EphrinB1/CNK1-mediated JNK activation is reduced by overexpression of dominant-negative RhoA. Overexpression of CNK1 alone is sufficient for activation of RhoA; however, both ephrinB1 and CNK1 are required for JNK phosphorylation. Co-immunoprecipitation data showed that ephrinB1 and CNK1 act as scaffold proteins that connect RhoA and JNK signaling components, such as p115RhoGEF and MKK4. Furthermore, adhesion to fibronectin or active Src overexpression increases ephrinB1/CNK1 binding, whereas blocking Src activity by a pharmacological inhibitor decreases not only ephrinB1/CNK1 binding, but also JNK activation. EphrinB1 overexpression increases cell motility, however, CNK1 depletion by siRNA abrogates ephrinB1-mediated cell migration and JNK activation. Moreover, Rho kinase inhibitor or JNK inhibitor treatment suppresses ephrinB1-mediated cell migration. Taken together, our findings suggest that CNK1 is required for ephrinB1-induced JNK activation and cell migration.