¹Department of Medical Biochemistry and Genetics, Center for Cancer Biology and Nutrition, Institute of Biosciences and Technology, Texas A&M University System Health Science Center, 2121 W. Holcombe Blvd., Houston, Texas, TX 77030, USA

²Department of Immunology, Baylor College of Medicine, Houston, Texas, TX 77030, USA

³Department of Anatomical Sciences and Neurobiology, School of Medicine, University of Louisville, Louisville, Kentucky, KY 40202, USA

⁴Shanghai Genomics, Inc., Zhangjiang Hi-Tech Park, Pudong, Shangai 201204, P.R.C.

Correspondence to: M Liu, Department of Medical Biochemistry and Genetics, Center for Cancer Biology and Nutrition, Institute of Biosciences and Technology, Texas A&M University System Health Science Center, 2121 W. Holcombe Blvd., Houston, Texas, TX 77030, USA

Immune cell-specific adaptor proteins create various combinations of multiprotein complexes and integrate signals from cell surface receptors to the nucleus, modulating the specificity and selectivity of intracellular signal transduction. Grap2 is a newly identified adaptor protein specifically expressed in lymphoid tissues. This protein shares 40-50% sequence homology in the SH3 and the SH2 domain with Grb2 and Grap. However, the Grap2 protein has a unique 120-amino acid glutamine- and proline-rich domain between the SH2 and C-terminal SH3 domains. The expression of Grap2 is highly restricted to lymphoid organs and T lymphocytes. In order to understand the role of this specific adaptor protein in immune cell signaling and activation, we searched for the Grap2 interacting protein in T lymphocytes. We found that Grap2 interacted with the hematopoietic progenitor kinase 1 (HPK1) in vitro and in Jurkat T cells. The interaction was mediated by the carboxyl-terminal SH3 domain of Grap2 with the second proline-rich motif of HPK1. Coexpression of Grap2 with HPK1 not only increased the kinase activity of HPK1 in the cell, but also had an additive effect on HPK1 mediated JNK activation. Furthermore, over expression of Grap2 and HPK1 induced significant transcriptional activation of c-Jun in the JNK signaling pathway and IL-2 gene reporter activity in stimulated Jurkat T cells. Therefore, our data suggest that the hematopoietic specific proteins Grap2 and HPK1 form a signaling complex to mediate the c-Jun NH₂-terminal kinase (JNK) signaling pathway in T cells. Oncogene (2001) 20, 1703-1714.

adaptor protein Grap2; HPK1; JNK; T-cells

Introduction

Activation of resting T cells through the T-cell antigen receptor triggers a cascade of intracellular signaling events that lead to enhanced gene transcription, cellular differentiation and proliferation (Cantrell, 1996; Weiss and Littman, 1994; Chan and Shaw, 1996; Crabtree and Clipstone, 1994; Wange and Samelson, 1996). Although components of the T-cell receptor complex have no intrinsic kinase activity, the earliest detectable biochemical events following stimulation of T cell receptor (TCR) are the activation of several protein tyrosine kinases (PTKs), resulting in transient tyrosine phosphorylation of numerous intracellular proteins. Two types of cytoplasmic PTKs are known to catalyze these phosphorylations. The Src family PTKs, Lck and Fyn, associate with the cytoplasmic portions of the CD3/ and CD4/CD8 co-receptors, respectively, and phosphorylate the paired tyrosine residues within the immunoreceptor tyrosine-based activation motifs (ITAM) of the TCR and CD3 cytoplamic tails. The ZAP-70 tyrosine kinase is recruited to the activated complex by binding phosphorylated ITAM motifs via its tandem SH2 domains. Sequential or combinatorial activation of Src family PTKs and the ZAP-70 leads to phosphorylation of many intracellular proteins, including the SH2 domain-containing leukocyte protein of 76 kDa (SLP-76), the linker for activation of T cells (LAT), and the guanine nucleotide exchange factor Vav, phospholipase C-1 (PLC-1), and the protooncoproteins c-Cbl (Jackman et al., 1995; Bustelo, 2000; Park et al., 1991; Weiss et al., 1991; Clements et al., 1998b; Donovan et al., 1994; Weber et al., 1998; Zhang et al., 1998). Two direct consequences of protein tyrosine phosphorylation have been suggested in TCR-mediated signal tranduction. First, phosphorylation of tyrosine residues can alter the catalytic activity of enzymes. Examples are ZAP-70, which must be phosphorylated for full kinase activity (Chan et al., 1995; Wange et al., 1995) and PLC-, which hydrolyzes phosphoinositides and generating second messengers that elevate intracellular Ca²⁺ and activate protein kinase C upon phosphorylation and activation. Second, tyrosine phosphorylation creates sites for binding by proteins with SH2 domains (Pawson, 1995), providing a mechanism for the assembly of protein complexes that participate in transducing antigen receptor signals to downstream cellular targets. However, a detailed explication of the machinery that links ligand occupancy of the TCR to changes in cell behavior and responses remain unclear.

Hematopoietic progenitor kinase 1 (HPK1) is a mammalian Ste20-related protein kinase (Hu et al., 1996; Kiefer et al., 1996). HPK1 is predominantly expressed in hematopoietic organs and cells. It has been demonstrated that transient expression of HPK1 activates the c-Jun N-terminal kinase (JNK) signaling pathway, but does not stimulate ERK or p38 kinase pathways (Hu et al., 1996; Kiefer et al., 1996). HPK1 is found in the signaling complex formed by the JNK interacting protein (JIP-1) and by the adaptor proteins Crk and CrkL (Whitmarsh and Davis, 1998; Ling et al., 1999). The JNK/stress-activated protein kinase signaling cascade is critical for cells to transmit extracellular stimuli into the nucleus, thereby leading to appropriate cellular responses such as proliferation, differentiation, and apoptosis (Ip and Davis, 1998; Minden and Karin, 1997; Whitmarsh et al., 1998; Chen and Tan, 2000). JNK signaling is activated by a number of extracellular stimuli, including mitogens, epidermal and transforming growth factors (EGF and TGF-), cytokines, and environmental stresses. In lymphocytes, HPK1 has been reported to be activated by antigen receptors in T and B cells (Liou et al., 2000). Furthermore, HPK1 was shown to interact with adaptor proteins Grb2 and Crk in vitro and in transfected cell model (Anafi et al., 1997; Ling et al., 1999; Oehrl et al., 1998). In T cells, Grb2 forms complexes with signaling molecules LAT, SLP-76, SOS, Cbl, Shc, and the cytoplasmic tail of T-cell costimulatory receptor CD28 (Schneider et al., 1995; Clements et al., 1999). However, the signaling specificity mediated by Grb2 in T cells and the pathways that link extracellular stimuli to JNK activation have not been fully understood.

One of the outstanding issues upon TCR activation is the identity of signaling proteins that couple proximal protein kinases with downstream pathways, such as the activation of the small GTP-binding proteins Ras/Rac/Rho, the activation of the PLC-Ca²⁺-dependent calcineurin pathway, and the activation of the JNK signaling pathway. The SH3/SH2 adaptor proteins play essential roles in the formation of intracellular signaling complexes, relaying extracellular signals from the plasma membrane to the nucleus of the cell. The SH2 domains can bind tightly to phosphotyrosine residue, and the SH3 domains can interact with proline-rich sequences (Songyang et al., 1994; Pawson, 1995). With multiple binding sites and the potential to create various combinations of multi-protein complexes, these adaptor proteins are well suited to integrate signals from cell surface receptors to the nucleus, determining the specificity of signaling pathways in immune cell activation. An appreciation of the importance of these adaptor proteins was derived from the studies of the growth factor receptor kinase signaling pathway, where initial binding and phosphorylation of the Shc adaptor protein creates a binding site for a second adaptor, Grb2. Then, Grb2 directs Sos, a guanine nucleotide exchange factor, to the membrane, and activate the Ras signaling pathway (Rozakis-Adcock et al., 1992; Cheng et al., 1998). In T cells, Grb2 recruitment also occurs via its SH2 domain, by binding T-cell receptor ITAM motifs directly or indirectly through Shc or p36/38. However, reconstitution of lymphoid receptors and kinases in nonlymphoid cells fails to activate these signaling pathways, suggesting that additional lymphoid-specific proteins are involved in the signal transduction and activation of T cells.

Most recently, we have identified a novel human leukocyte-specific Grb2-related adaptor protein, Grap2, from human bone marrow (Qiu et al., 1998). Grap2 was also independently identified and named as Gads, GrapL, Mona, Grf40, Grid (Asada et al., 1999; Liu and McGlade, 1998; Law et al., 1999; Ellis et al., 2000). The Grap2 protein is a member of the Grb2 family of adaptor proteins that contain amino- and carboxy-terminal SH3 domain and a central SH2 domain (Lowenstein et al., 1992; Rudd, 1999; Feng et al., 1996; Trub et al., 1997; Qiu et al., 1998). Grap2 shares 40-50% sequence homology in the SH3 domains and 57% homology in the SH2 domain with Grb2 and Grap. However, unlike Grb2 and Grap, Grap2 contains a unique 120-amino acid glutamine- and proline-rich domain between the SH2- and C-terminal SH3 domains. Database search finds this unique 120-amino acid domain has no apparent homology with other proteins. Furthermore, the expression of Grap2 is highly specific to lymphoid organs, T cells and monocytes/macrophages, while Grb2 is ubiquitously expressed. In T lymphocytes, Grap2/Gads has been reported to interact with SLP-76 and LAT to regulate nuclear factor of activated T cell activation (NF-AT) (Liu et al., 1999; Law et al., 1999; Zhang et al., 2000). In addition, Grap2 also interacts with the macrophage colony-stimulating factor receptor and the activated T cell costimulatory receptor CD28 (Bourette et al., 1998; Ellis et al., 2000). Therefore, Grap2 is likely to have distinct functions and serve as an important adaptor protein in T cell signaling and activation, although Grb2 and Grap have been proposed to serve overlapping functions in the cell (Cheng et al., 1998).

In this report, we demonstrate that Grap2 directly interacted with the hematopoietic progenitor kinase 1 (HPK1) in vitro and in mammalian cells. The interaction was mediated by the carboxyl-terminal SH3 domain of Grap2 and the second proline-rich motif of HPK1. Coexpression of Grap2 and HPK1 not only increased HPK1 kinase activity, but also had an additive effect on HPK1-mediated activation of c-Jun N-terminal kinase (JNK). In addition, cotransfection of Grap2 and HPK1 in mammalian cells induced transcriptional activation of c-Jun in the JNK signaling pathway and IL-2 gene reporter activity in stimulated Jurkat T cells. Together, our data suggest that hematopoietic proteins Grap2 and HPK1 can form a signaling complex to mediate the JNK signaling pathway in T lymphocytes.

Results

Grap2 interacts with HPK1 in vitro and in mammalian cells

To identify proteins that interact with Grap2 in T lymphocytes, we used an in vitro GST-fusion protein 'pull-down' assay in Jurkat T cells. GST-Grap2 fusion proteins were expressed in E. coli and purified by glutathione agarose column. Jurkat T cells were stimulated with anti-CD3 antibody (OKT3). Cell lysates from unstimulated or stimulated cells were prepared and incubated with GST-Grap2 fusion protein. The protein complexes were separated on SDS-PAGE and Western blotted with either an anti-phosphotyrosine antibody (4G10) or an anti-HPK1 antibody. As shown in Figure 1a, a number of tyrosine phosphorylated proteins were detected in GST-Grap2 complex following stimulation by anti-CD3 antibodies. Among those proteins, a 100 kDa HPK1 protein was detected by an anti-HPK1 antibody (Figure 1b), indicating that Grap2 interacts with HPK1 in Jurkat T cells. To test whether the interaction is dependent on the phosphorylation status of HPK1, we stimulated the Jurkat T cells with an anti-CD3 antibody (OKT3) and examined the tyrosine phosphorylation status of the HPK1 protein. As shown in Figure 1c, HPK1 phosphorylation increased approximately twofold in Jurkat cells stimulated with OKT3 antibodies, indicating that HPK1 is phosphorylated upon T cell stimulation. We then compared the binding of the HPK1 to GST-Grap2 in stimulated or unstimulated cells. As shown in Figure 1b, GST-Grap2 bound HPK1 regardless of TCR stimulation, although a slight increase of binding was observed upon T-cell stimulation. These data suggest that activation of TCR stimulated the phosphorylation of HPK1 and increased the binding affinity of HPK1 to Grap2 in stimulated Jurkat T cells.

To confirm that Grap2 indeed interact with HPK1 in mammalian cells, Grap2 was cotransfected with Flag-tagged HPK1 in COS-7 cells. Protein extracts from the transfected cells were immunoprecipitated using polyclonal rabbit anti-Grap2 antibodies. The interaction of HPK1 with Grap2 was detected by Western blot using an anti-HPK1 antibody or an anti-Flag monoclonal antibody (M2) in the Grap2 immuno-complex. Figure 1d shows that HPK1 coprecipitated with Grap2 complex in cells cotransfected Grap2 and HPK1, indicating that the two proteins can form a complex in the cell.

The carboxyl-terminal SH3 domain of Grap2 interacts with HPK1

To determine which domain(s) of Grap2 is responsible for its interaction with HPK1, we constructed different domains of Grap2 in the yeast Gal4 DNA-binding plasmid (pAS2), including the N-terminal SH3 and SH2 domain (N-SH3-SH2), the carboxyl-terminal unique domain and SH3 domain (QC), the glutamine-/proline-rich (Q/P-) domain, and the C-terminal SH3 domain (SH3-C). Each domain of Grap2 was fused in-frame into Gal4-BD in pAS2-1 expression vector by homologous recombination in yeast cells (Hua et al., 1997). To test which domain of Grap2 interacts with HPK1, yeast cells were cotransformed with expression vectors containing individual domains of Grap2 and the C-terminal domain of HPK1 (HPK1-CD) in Gal4-AD plasmid, pACT-2. Interaction of the Grap2 domain with HPK1 was monitored by the growth of yeast cells on SD plates with selective medium (-Trp/-Leu/-His) and by the activation of the LacZ reporter gene. As shown in Figure 2a, yeast cells cotransformed with either the QC domain or the SH3-C domain of Grap2 together with HPK1-CD grew on selective medium. Cotransformation of other Grap2 domains (NSH3-SH2 and Q/P domain) did not confer the growth of yeast cells. Since the QC domain contains the C-terminal unique domain and the SH3-C domain, these data suggest that the C-terminal SH3 domain is responsible for the interaction of Grap2 with HPK1 in yeast cells. To quantitate the interaction of Grap2 with HPK1, we measured the activation of the LacZ reporter gene under the control of the Gal4 promoter. As shown in Figure 2b, cotransformation of the Grap2-QC or Grap2 SH3-C domain with HPK1-CD strongly activated the LacZ activity, while other cotransformations had no significant LacZ activity. P53 and large T antigen were used as a positive control in our yeast two-hybrid assays. These results indicate that the interaction of Grap2 with HPK1 is mediated by the C-terminal SH3 domain of Grap2.

To confirm our observation in yeast cells, we also performed an in vitro binding assay with GST-fusion proteins corresponding to different domains of Grap2 (Figure 3a). Cell lysates from COS-7 cells transfected with Flag-tagged HPK1 were incubated with the GST-Grap2 fusion proteins. The association of HPK1 with individual Grap2 domains was detected by Western blot using an anti-Flag monoclonal antibody (M2). Figure 3b shows that only the C-terminal SH3 domain of Grap2 interacted with HPK1. GST-fusion proteins that contain the N-terminal SH3 and the SH2 domains did not associate with the HPK1 protein in our in vitro assay. Similar data were obtained when Jurkat T cell lysates containing endogenous HPK1 were used in the binding assays (data not shown). Together, these results suggest that the C-terminal SH3 domain of Grap2 interacts specifically with the HPK1 protein in the cell.

Grap2 interacts with the second proline-rich motif of HPK1

To determine which region(s) of HPK1 is involved in the interaction with the Grap2 protein, Flag-tagged wild-type HPK1 and four mutant HPK1 expression constructs were generated. Figure 4a shows the domain structures of five Flag-tagged HPK1 expression constructs, including the wild-type HPK1, the kinase-deficient mutant HPK1 (HPK1-M46), the HPK1 kinase domain (HPK1-KD), the HPK1 carboxyl domain (HPK1-CD), and the HPK1 proline-rich domain (HPK1-PR). Proteins were generated by in vitro transcription and translation method and labeled with [³⁵S]methionine. The translation of HPK1 and HPK1 mutant proteins were resolved on SDS-PAGE and detected by autoradiography (Figure 4b). To examine which domain(s) of HPK1 is involved in the interaction with Grap2, we incubated the GST-Grap2-SH3(C) domain with the in vitro translated HPK1 proteins. Proteins bound to Grap2 were washed extensively and separated by SDS-PAGE. As shown in Figure 4c, the full-length HPK1, the HPK1-M46, the HPK1-CD domain, and the HPK1-PR domain bound with the Grap2 C-terminal SH3 domain while the HPK1 kinase domain did not bind Grap2, suggesting that the proline-rich region of HPK1 is essential for the interaction with Grap2.

It has been known that the SH3 domain binds to the proline-rich region with consensus motif (PXXPXK/R) in the protein (Songyang et al., 1994; Pawson, 1995). Examination of the HPK1 sequence reveals that the protein contains four proline-rich regions, therefore, four putative Grap2-SH3 domain-binding motifs, PR1, PR2, PR3, and PR4 (Figure 4a). To investigate whether the four individual HPK1 proline-rich motifs are involved in the interaction with Grap2, we synthesized four peptides corresponding to the putative SH3-binding motifs in HPK1, termed PR1, PR2, PR3, and PR4 (Figure 4a). Using the four proline-rich peptides, we performed competition-binding assays by adding individual HPK1 peptides to the Grap2-SH3(C) and HPK1 complexes. As shown in Figure 4d, the HPK1 second proline-rich peptide (PR2) completely blocked the formation of Grap2-HPK1 complexes. The first and the fourth proline-rich peptides (PR1 and PR4) of HPK1 also showed some inhibitory effect on the complex, whereas the third proline-rich peptide (PR3) had no effect on the formation of Grap2-HPK1 complex. These data suggest that the second proline-rich peptide of HPK1 is the key site for interaction with the Grap2-SH3(C) domain, while the role of the fourth proline-rich motif in the interaction with Grap2 is not certain since the fourth proline-rich domain is not conserved between human and mouse. The first proline-rich domain of HPK1 may also play some role in facilitating the formation of the Grap2-HPK1 complex.

Grap2 activates HPK1 and has additive effects with HPK1 in the activation of JNK

To determine the functional significance of this interaction, we examined whether Grap2 could regulate HPK1 activity in the cell. The kinase activity of HPK1 was measured by an immunocomplex kinase assay in the presence or absence of Grap2. As shown in Figure 5a, cotransfection of Grap2 increased the kinase activity of HPK1 by approximately 2-3-fold, suggesting that the hematopoietic adaptor protein Grap2 may recruit HPK1 to the proximity of other HPK1 activators in the cell.

It has been reported that HPK1 is involved in the JNK signaling cascade (Hu et al., 1996; Kiefer et al., 1996; Ling et al., 1999; Zhou et al., 1999). Overexpression of HPK1 activates JNK kinase activity in transfected cells (Ling et al., 1999). To examine whether coexpression of Grap2 and HPK1 has any effect on the JNK signaling pathway, we assayed the JNK activity in the presence of Grap2, HPK1, or both. Cotransfection of HPK1 and HA-tagged JNK activates JNK activity by about threefold and is inconsistent with previous reports (Ling et al., 1999). Overexpression of Grap2 with JNK slightly increased the activation of JNK activity compared to HA-tagged JNK alone. However, in the presence of Grap2 and HPK1, JNK activity was increased approximately 4-5-fold (Figure 5b), indicating that coexpression of Grap2 with HPK1 in the cell can increase HPK1-mediated activation of the JNK pathway.

Grap2 and HPK1 have an additive effect on the activation of c-Jun in the JNK signaling pathway

To determine the effect of the Grap2-HPK1 interaction on the stimulation of the JNK signaling pathway, we employed a PathDetect trans-reporting system with the activation domain of c-Jun fused to the DNA-binding domain of Gal4. Cells were cotransfected with the pathway-specific c-Jun transcription activator plasmid and a reporter vector containing luciferase (pFR-Luc). As shown in Figure 6a, cotransfection of JNK1, a known activator of the JNK signaling pathway, increased the c-Jun coupled luciferase activity by about 10-fold. Overexpression of either Grap2 or HPK1 increased c-Jun transcriptional activity by 3-4-fold, respectively. However, cotransfection of Grap2 and HPK1 with the c-Jun reporter system increased the transcriptional activity of c-Jun by at least eightfold, similar to the activation by overexpression of JNK1 in the cell. These data indicate that coexpression of Grap2 and HPK1 has an additive effect on the activation of the JNK signaling pathway. To further confirm the specific effect of Grap2-HPK1 interaction, two deletion mutants, Grap2-dSH3C (C-terminal SH3 deleted) and Grap2-dSH2 (SH2 domain deleted), were cotransfected together with HPK1 in the cell, respectively. Both mutants failed to activate the JNK signaling pathway like the wild-type Grap2 protein (Figure 6a), suggesting that the additive effect is mediated by the specific Grap2-HPK1 interaction.

Grap2 and HPK1 potentiate IL-2 activation by the TCR and PMA

TCR ligation initiates distinct signaling cascades that ultimately combine to activate transcription of the interleukin-2 (IL-2) gene, a key growth and differentiation factor for T cells (Weiss and Littman, 1994; Karin and Hunter, 1995). IL-2 induction involves the simultaneous activation of the nuclear factor of activated T cells (NF-AT) by the calcium/calmodulin-dependent phosphatase calcineurin, and of the AP-1 complex (Verweij et al., 1990; Clipstone and Crabtree, 1992; Crabtree and Clipstone, 1994). It has been shown that cotransfection of HPK1 with a reporter construct (5´TRE-CAT) increased AP-1 activity dramatically (Hu et al., 1996). To examine whether Grap2 and HPK1 are directly or indirectly involved in the IL-2 activation, we cotransfected Grap2 and HPK1 with an IL-2-driven luciferase reporter and subsequently stimulated with an anti-CD3 antibody and PMA. As shown in Figure 6b, Grap2 increased TCR-stimulated IL-2 activity by about 2-3-fold, while HPK1 had limited effects on the IL-2 activation. Cotransfection of Grap2 and HPK1 together with the IL-2 reporter increased the IL-2 activity by approximately fourfold, suggesting Grap2-HPK1 may directly or indirectly participate in the activation of IL-2 gene in T-cell activation.

Discussion

The selectivity and specificity of intracellular signaling are achieved by compartmentalization and regional organization of signaling components via molecular scaffold and adaptor proteins (Whitmarsh et al., 1998; Rudd, 1999; Clements et al., 1999; Zhang and Samelson, 2000). In the past few years, a number of scaffold and adaptor proteins have been identified to form signaling complexes and play an important role in determining the specificity and selectivity of signal transduction in the cell. These protein complexes may result from physical interaction between components of particular signaling pathways or by the assembly of signaling molecules on anchor or scaffold proteins that localize their binding partners to specific subcellular compartments or to specific substrates. Examples include the MAP kinase complexes coordinated by the scaffold proteins Ste5p and Pbs2p in the yeast Saccharomyces cerevisiae (Choi et al., 1994), and the JNK kinase signaling complexes by the scaffold protein, JIP-1 (Whitmarsh et al., 1998). In immune cells, a variety of adaptor molecules play critical functions in linking receptor proximal signal transduction events such as the activation of receptor-associated kinases to downstream effector molecules to promote the transcriptional induction of multiple genes. A number of lymphocyte specific adaptor proteins, such as LAT, SLP-76, FYB/SLAP, SKAP55, 3BP2, and BLNK, have been identified to interact with signaling molecules, regulating lymphocyte activation and development (Weber et al., 1998; Clements et al., 1998a; Zhang et al., 1998; Pivniouk et al., 1998; da Silva et al., 1997; Deckert et al., 1998; Fu et al., 1998; Pappu et al., 1999). Therefore, these adaptor proteins regulate immune cell signaling and activation by selectively forming signaling complexes with different molecules (Clements et al., 1999; Rudd, 1999). Grap2 is a new member of the Grb2 family of adaptor proteins (Qiu et al., 1998; Law et al., 1999; Xia et al., 2000). Grap2 shares similar structural organization as Grb2 and Grap with the N- and C-terminal SH3 domains and a central SH2 domain. However, unlike Grb2 and Grap, Grap2 contains a unique 120-amino acid glutamine-/proline-rich domain, and the expression of the Grap2 protein is highly specific in hematopoietic tissues and in T cells (Qiu et al., 1998). In this report, we demonstrated that the Grap2 protein interacted with the hematopoietic progenitor kinase 1 (HPK1) to form signaling complex both in vitro and in mammalian cells. The Grap2-HPK1 interaction is mediated by the carboxy-SH3 domain of Grap2 and the proline-rich regions of HPK1. Among the four proline-rich regions of HPK1, the second proline-rich motif was identified to be critical for the binding of HPK1 to Grap2. Although our data show that peptides from the first and the fourth proline-rich motifs also partially blocked the interaction of HPK1 and Grap2, the role of these two proline-rich motifs in the functional interaction of HPK1 and Grap2 is uncertain and is under investigation. The interaction between Grap2 and HPK1 is independent of the stimulation status of the cells although we observed that HPK1 has high affinity for interaction with Grap2 upon T-cell stimulation. Therefore, we cannot rule out the possibility that Grap2-HPK1 form a functional signaling complex only upon stimulation of T cells.

Cotransfection of Grap2 and HPK1 directly activated HPK1 in our kinase assays and Grap2 also cooperated with HPK1 to activate JNK (Figure 5). These results indicate that the Grap2 protein not only physically interacts with HPK1 but also functionally modulates HPK1-mediated JNK activation. Furthermore, overexpression of Grap2 and HPK1 activates the JNK-mediated c-Jun transcriptional activity, suggesting that the Grap2-HPK1 complex plays an important role in JNK-mediated signal transduction in T lymphocytes. HPK1 receives the upstream signals through its interaction with the SH2/SH3 adapter proteins and subsequently transmits these signals to downstream targets (MEKK1 and TAK1) via its kinase domain or distal regulatory region (Ling et al., 1999; Anafi et al., 1997). HPK1 has been identified as a substrate for protein tyrosine kinases and is activated by lymphocyte antigen receptors, TGF-, or by erythropoietin (Zhou et al., 1999; Anafi et al., 1997; Nagata et al., 1999; Liou et al., 2000). Both the receptor tyrosine kinases (EGF, PDGF) and the Src family of cytoplasmic tyrosine kinases has been shown to induce tyrosine phosphorylation of HPK1 (Anafi et al., 1997; Wang et al., 1997). It is tempting to speculate that upon stimulation of TCR, Grap2 may recruit HPK1 to the proximity of the receptors, facilitate the phosphorylation and activation of HPK1 by protein tyrosine kinases, and subsequently transmit these signals to downstream targets, leading to the activation of the JNK signaling pathway.

In T-cell activation, optimal IL-2 induction requires two major signaling pathways triggered by costimulation of the T-cell receptor (TCR) and CD28 or mimicked by incubating T cells with phorbol ester (i.e., PMA) and Ca²⁺ ionophore (Su et al., 1994). Early studies indicate that AP-1 binds to the IL-2 promoter (Muegge et al., 1989) and forms a complex with NF-AT to induce IL-2 activation (Northrop et al., 1993; Ullman et al., 1993). JNK is involved in the integration of the costimulatory signals CD3 plus CD28 or PMA plus Ca²⁺ ionophore in T cells (Su et al., 1994). As a hematopoietic upstream activator of JNK, HPK1 may be involved in IL-2 induction in T cells. Using IL-2 luciferase reporter assays, we found that coexpression of Grap2 and HPK1 enhanced IL-2 activation by T-cell stimuli (anti-CD3 antibody and PMA) in Jurkat T cells (Figure 6b). Like some of the well-known adaptor proteins, Grap2 may form signaling complexes with distinct signaling molecules in T cells to mediate cell signaling and activation. Grap2 may couple the TCR and CD28 with protein tyrosine kinases, such as Src family of kinases, to HPK1 to activate HPK1-mediated JNK signaling pathway during T-cell activation.

During the preparation of this study for publication, a study describing the interaction of Gads and HPK1 was independently reported (Liu et al., 2000). Our studies not only independently confirm the finding that Grap2 interacts with HPK1 in T cells, but also demonstrate that the interaction modulates the activation of HPK1 and HPK1-mediated JNK signaling pathway. While the fourth proline-rich domain of HPK1 may participate in the interaction with Grap2, the second proline-rich domain is identified to play more critical role in the interaction of HPK1 and Grap2 in our study. Taken together, our data suggest that the hematopoietic adaptor protein Grap2 forms a signaling complex with HPK1 to activate the JNK signaling pathway in TCR-mediated signal transduction and activation. Since Grap2 contains multiple domains for protein-protein interaction, Grap2 may function as a central linker protein to interact with multiple molecules and form different signaling complexes during T-cell activation. Further experiments are under investigation to test the hypothesis that different domains of Grap2 may be involved in the interactions with various signaling molecules to determine the signaling specificity and selectivity in T-cell activation and development.

Materials and methods

Plasmid constructs

To generate HA- or Flag-Tagged Grap2 expression plasmids, PCR was used to amplify different domains of Grap2 using the full-length of human Grap2 cDNA as a template. Pfu (Strategene) was used in all PCR subcloning, and the sequences of different Grap2 constructs were confirmed by DNA sequencing. To generate the GST (glutathione S-transferase)-Grap2 fusion constructs, cDNA fragments corresponding to different domains of Grap2 were subcloned into pGEX-2TK or pGEX-4T1 (Pharmacia Biotech, Inc.) for expression as a fusion protein of GST and various lengths of Grap2. pCIneo-FLAG-tagged vectors of wild-type HPK1 and various HPK1 mutants, including a kinase-dead mutant (HPK1-K46M), the HPK1 kinase domain alone (HPK1-KD; amino acids 1-291), the HPK1 carboxyl domain (HPK1-CD; amino acids 292-833), and the HPK1 proline-rich domain (amino acids 288-482) were described previously (Hu et al., 1996; Ling et al., 1999). Hemagglutin (HA)-tagged JNK and GST-cJun (1-79) were described previously (Diener et al., 1997; Hu et al., 1996).

To generate Grap2 domain specific expression vectors in yeast cells, sequences corresponding to the SH3(N)-SH2, the SH2, the SH3(C), and the unique Q/P domains were amplified by PCR and fused in-frame to the Gal4 BD of the pAS-2 vector by homologous recombination in yeast cells. The carboxyl-terminal domain of HPK1 (HPK1-CD, amino acids 292-833) was subcloned into the Gal4-AD plasmid, pACT-2 using the PCR technique. The luciferase reporter constructs (IL-2 luciferase and NF-AT luciferase) were obtained from Dr G Crabtree (Stanford University). The c-Jun Path-Detect Assay system was purchased from Stratagene (CA, USA).

Recombinant proteins

The pGEX plasmids encoding various GST-fusion proteins were used to transform E. coli strain BL21 (DE3) together with 2tRNA plasmid. After induction of protein expression with 1 mM IPTG (isopropyl-1-thio-D-galactopyranoside) for 3-4 h, the bacteria were resuspended in lysis buffer (50 mM Tris/HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, leupeptin, 0.4 mg/ml lysozyme) and sonicated briefly on ice. Triton X-100 (1%) was added to lysates before centrifugation at 10 000 g for 30 min. In certain cases, proteins in inclusion bodies were solubilized with RIPA buffer (1´PBS, 1% Igepal CA-630, 0.5% sodium deoxycholate, 0.1% SDS, pH 7.4). The GST-Grap2 fusion proteins were bound to glutathione agarose (Sigma) beads, and non-specific proteins were washed away with buffer containing 25 mM Tris/HCl, pH 7.5, 100 mM NaCl, 1.0 mM EDTA, 1.0 mM dithiothreitol and protease inhibitors. Protein purity and concentrations were assessed on SDS-PAGE by Coomassie blue staining.

Cell culture, transfection, and luciferase assays

COS-7 cells were grown in Dulbecco modified Eagle's medium with 10% fetal bovine serum (FBS). The day before transfection, cells (2´10⁵ cells per well) were plated in a six-well plate. Transfection was then performed by the GenePorter transfection reagents (Gene Therapy, Inc.) with various plasmids as indicated in the figures. After 48 h, the cells were harvested and lysed in lysis buffer (150 mM NaCl, 20 mM HEPES, pH 7.4, 2 mM EGTA, 50 mM glycerophosophate, 1% Triton X-100, 10% glycerol, 500 M phenylmethylsulfonyl fluoride [PMSF], 5 g/ml leupeptin, 3 g/ml aprotinin). Cell lysates were used for binding assays and kinase assays.

Jurkat cells in log-phase were washed with serum-free RPMI 1640 once and electroporated in triplicate with a total of 30 g of plasmid at 240 V, 960 F using a Gene Pulser (Bio-Rad) at 10⁷ cells/400 l serum-free RPMI 1640. Sonicated salmon sperm DNA (Sigma Chemical Co.) was used to maintain the total DNA at 30 g whenever necessary. After electroporation, cells were incubated in fully supplemented RPMI 1640 for 14-18 h at 37°C, 5% CO₂. Cells were then transferred into 96-well plates at 2-5´10⁵ cells/well in a total of 100 l of medium. Cells were stimulated in triplicate OKT3 antibodies for 6-8 h at 37°C. Control cells were maintained in culture medium. After stimulation, cells were lysed in 50 l Reporter Lysis buffer (Promega) for 30 min at room temperature. Luciferase activity was quantified by adding 50 l of the lysate to 25 l of luciferase assay substrate (Promega) and immediately measured with a GenProbe LeaderTM 1 luminometer (Wallac Inc.). To normalize for variations in transfection efficiency, luciferase activities in different transfection conditions were expressed as fold increase compared to the control experiments.

Peptides, antibodies, immunoprecipitation and immunoblot analysis

Four wild-type HPK1 proline-rich peptides were synthesized using Fmoc (N(9-fluorenyl)methoxycarbonyl) chemistry and purified by high pressure liquid chromatography (HPLC). Polyclonal antibodies to Grap2 were prepared by immunization of rabbits with GST-Grap2-QC domain fusion protein. Grap2-specific antibodies were affinity-purified by passage of the antisera over immobilized GST-agarose beads. Anti-HPK1 antibodies were generated against the HPK1 peptide (amino acids 341-366). Other antibodies used in these studies include: anti-CD3 mAb OKT3 (American Type Culture Collection); anti-phosphotyrosine mAb 4G10 (Upstate Biotechnology); anti-phosphotyrosine mAb PY-20 (Sigma); anti-Flag mAb, M2 (Sigma); anti-HA monoclonal antibodies (12CA5; Boehringer Mannheim).

Immunoprecipitation and immunoblot analysis were carried out as described previously (Xia et al., 2000). In brief, cells were lysed with a cell extraction buffer (1% NP-40, 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM Na₃VO₄, 2 mM PMSF and 20 g/ml aprotinin, 3 g/ml leupeptin, and 3 g/ml aprotinin), and immunoprecipitated with the indicated antibodies or GST-fusion proteins. Lysates were rocked 30 min at 4°C, centrifuged (10 000 g, 15 min), and precleared with washed protein A-Sepharose 4B (Pharmacia) or GST-coated glutathione-agarose. Anti-HPK1 and anti-HA immunocomplexes were recovered by using protein-A beads (Sigma). Anti-FLAG immunocomplexes were recovered by using protein-G beads (Santa Cruz Biotechnology). All immunoprecipitates were washed four times with lysis buffer and were separated by SDS-PAGE and then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore). After incubation in TBST containing 2% BSA or 5% dry milk powder, the membranes were probed with the indicated antibodies and visualized using the SuperSignal West Pico detection system (Pierce, IL, USA).

Immunocomplex kinase assay

Assays for JNK activity were performed as described previously (Hu et al., 1996) Briefly, HA-tagged JNK1 from 50 g of transfected COS-7 cell lysates was immunoprecipitated by using an anti-HA monoclonal antibody and protein A-agarose beads in 500 l of lysis buffer. After being washed twice with lysis buffer, twice with LiCl buffer (500 mM LiCl, 100 mM Tris-Cl, pH 7.6, 0.1% Triton X-100), and twice with kinase buffer (20 mM MOPS [morpholine propanesulfonic acid], pH 7.2, 2 mM EGTA, 10 mM MgCl₂, 0.1% Triton X-100), the immunoprecipitates were incubated with 30 l of kinase buffer containing 2 g of GST-c-Jun (amino acids 1-79), 20 M unlabeled ATP, and 20 Ci of [³²P]ATP. The reactions were incubated at 30°C for 30 min and terminated by boiling the samples in 4´SDS sample buffer. Proteins were separated by SDS-PAGE (12%). Similar kinase assays were performed for HPK1 activity. Jurkat T cells and COS-7 cells transiently expressing HPK1 were immunoprecipitated by using an anti-HPK1 antibody or an anti-FLAG monoclonal antibody, and kinase assays were performed with myelin basic protein (MBP).

In vitro translation and binding assays

In vitro transcription and translation of proteins were performed essentially according to the manufacturer's instructions (Promega Biotech, Inc.). Proteins were labeled and detected by the incorporation of [³⁵S]methionine. Plasmids (1-2 g/reaction) were added to a reaction mixture, including 25 l of rabbit reticulocyte lysate, 2 l of reaction buffer, 1 l of T7 RNA polymerase, 1 l of amino acid mixture minus methionine (1 mM), 4 l of [³⁵S]methionine (10 mCi/ml), ribonuclease inhibitor (40 U/l), and nuclease-free H₂O to a final volume of 50 l. The reactions were incubated at 30°C for 90 min. The in vitro translated proteins were then used for in vitro binding assays. The interactions of in vitro translated proteins with GST fusion proteins were directly examined by autoradiography. The ability of individual HPK1 proline-rich peptides to block the interaction of HPK1 with Grap2 was performed as previously described (Ling et al., 1999) or detected by Western blot using anti-HPK1 antibodies

Acknowledgements

This work was supported in part by a Scientist Development Grant 0030160N from the American Heart Association National and the Basil O'Connor Starter Scholar Research Award from March of Dimes Foundation (to M Liu) and NIH grants RO1-AI42532 and RO1-AI38649 (to T-H Tan).

Anafi M, Kiefer F, Gish GD, Mbamalu G, Iscove NN, Pawson T. (1997). J. Biol. Chem. 272, 27804-27811. MEDLINE

Asada H, Ishii N, Sasaki Y, Endo K, Kasai H, Tanaka N, Takeshita T, Tsuchiya S, Konno T, Sugamura K. (1999). J. Exp. Med. 189, 1383-1390. MEDLINE

Bourette RP, Arnaud S, Myles GM, Blanchet JP, Rohrschneider LR, Mouchiroud G. (1998). EMBO J. 17, 7273-7281. MEDLINE

Bustelo XR. (2000). Mol. Cell. Biol. 20, 1461-1477. MEDLINE

Cantrell DA. (1996). Cancer Surv. 27, 165-175. MEDLINE

Chan AC, Dalton M, Johnson R, Kong GH, Wang T, Thoma R, Kurosaki T. (1995). EMBO J. 14, 2499-2508. MEDLINE

Chan AC, Shaw AS. (1996). Curr. Opin. Immunol. 8, 394-401. MEDLINE

Chen YR, Tan TH. (2000). Int. J. Oncol. 16, 651-662. MEDLINE

Cheng AM, Saxton TM, Sakai R, Kulkarni S, Mbamalu G, Vogel W, Tortorice CG, Cardiff RD, Cross JC, Muller WJ, Pawson T. (1998). Cell 95, 793-803. MEDLINE

Choi KY, Satterberg B, Lyons DM, Elion EA. (1994). Cell 78, 499-512. MEDLINE

Clements JL, Boerth NJ, Lee JR, Koretzky GA. (1999). Annu. Rev. Immunol. 17, 89-108. MEDLINE

Clements JL, Ross-Barta SE, Tygrett LT, Waldschmidt TJ, Koretzky GA. (1998a). J. Immunol. 161, 3880-3889.

Clements JL, Yang B, Ross-Barta SE, Eliason SL, Hrstka RF, Williamson RA, Koretzky GA. (1998b). Science 281, 416-419. Article MEDLINE

Clipstone NA, Crabtree GR. (1992). Nature 357, 695-697. MEDLINE

Crabtree GR, Clipstone NA. (1994). Annu. Rev. Biochem. 63, 1045-1083. MEDLINE

da Silva AJ, Li Z, de Vera C, Canto E, Findell P, Rudd CE. (1997). Proc. Natl. Acad. Sci. USA 94, 7493-7498. Article MEDLINE

Deckert M, Tartare-Deckert S, Hernandez J, Rottapel R, Altman A. (1998). Immunity 9, 595-605. MEDLINE

Diener K, Wang XS, Chen C, Meyer CF, Keesler G, Zukowski M, Tan TH, Yao Z. (1997). Proc. Natl. Acad. Sci. USA 94, 9687-9692. MEDLINE

Donovan JA, Wange RL, Langdon WY, Samelson LE. (1994). J. Biol. Chem. 269, 22921-22924. MEDLINE

Ellis JH, Ashman C, Burden MN, Kilpatrick KE, Morse MA, Hamblin PA. (2000). J. Immunol. 164, 5805-5814. MEDLINE

Feng GS, Ouyang YB, Hu DP, Shi ZQ, Gentz R, Ni J. (1996). J. Biol. Chem. 271, 12129-12132. MEDLINE

Fu C, Turck CW, Kurosaki T, Chan AC. (1998). Immunity 9, 93-103. MEDLINE

Hu MC, Qiu WR, Wang X, Meyer CF, Tan TH. (1996). Genes Dev. 10, 2251-2264. MEDLINE

Hua SB, Qiu M, Chan E, Zhu L, Luo Y. (1997). Plasmid. 38, 91-96. Article MEDLINE

Ip YT, Davis RJ. (1998). Curr. Opin. Cell. Biol. 10, 205-219. MEDLINE

Jackman JK, Motto DG, Sun Q, Tanemoto M, Turck CW, Peltz GA, Koretzky GA, Findell PR. (1995). J. Biol. Chem. 270, 7029-7032. MEDLINE

Karin M, Hunter T. (1995). Curr. Biol. 5, 747-757. MEDLINE

Kiefer F, Tibbles LA, Anafi M, Janssen A, Zanke BW, Lassam N, Pawson T, Woodgett JR, Iscove NN. (1996). EMBO J. 15, 7013-7025. MEDLINE

Law CL, Ewings MK, Chaudhary PM, Solow SA, Yun TJ, Marshall AJ, Hood L, Clark EA. (1999). J. Exp. Med. 189, 1243-1253. MEDLINE

Ling P, Yao Z, Meyer CF, Wang XS, Oehrl W, Feller SM, Tan TH. (1999). Mol. Cell. Biol. 19, 1359-1368. MEDLINE

Liou J, Kiefer F, Dang A, Hashimoto A, Cobb MH, Kurosaki T, Weiss A. (2000). Immunity 12, 399-408. MEDLINE

Liu SK, Fang N, Koretzky GA, McGlade CJ. (1999). Curr. Biol. 9, 67-75. Article MEDLINE

Liu SK, McGlade CJ. (1998). Oncogene 17, 3073-3082. MEDLINE

Liu SK, Smith CA, Arnold R, Kiefer F, McGlade CJ. (2000). J. Immunol. 165, 1417-1426. MEDLINE

Lowenstein EJ, Daly RJ, Batzer AG, Li W, Margolis B, Lammers R, Ullrich A, Skolnik EY, Bar-Sagi D, Schlessinger J. (1992). Cell 70, 431-442. MEDLINE

Minden A, Karin M. (1997). Biochim. Biophys. Acta. 1333, F85-F104. MEDLINE

Muegge K, Williams TM, Kant J, Karin M, Chiu R, Schmidt A, Siebenlist U, Young HA, Durum SK. (1989). Science 246, 249-251. MEDLINE

Nagata Y, Kiefer F, Watanabe T, Todokoro K. (1999). Blood 93, 3347-3354. MEDLINE

Northrop JP, Ullman KS, Crabtree GR. (1993). J. Biol. Chem. 268, 2917-2923. MEDLINE

Oehrl W, Kardinal C, Ruf S, Adermann K, Groffen J, Feng GS, Blenis J, Tan TH, Feller SM. (1998). Oncogene 17, 1893-1901. MEDLINE

Pappu R, Cheng AM, Li B, Gong Q, Chiu C, Griffin N, White M, Sleckman BP, Chan AC. (1999). Science 286, 1949-1954. Article MEDLINE

Park DJ, Rho HW, Rhee SG. (1991). Proc. Natl. Acad. Sci. USA 88, 5453-5456. MEDLINE

Pawson T. (1995). Nature 373, 573-580. MEDLINE

Pivniouk V, Tsitsikov E, Swinton P, Rathbun G, Alt FW, Geha RS. (1998). Cell 94, 229-238. MEDLINE

Qiu M, Hua S, Agrawal M, Li G, Cai J, Chan E, Zhou H, Luo Y, Liu M. (1998). Biochem. Biophys. Res. Commun. 253, 443-447. MEDLINE

Rozakis-Adcock M, McGlade J, Mbamalu G, Pelicci G, Daly R, Li W, Batzer A, Thomas S, Brugge J, Pelicci PG. (1992). Nature 360, 689-692. MEDLINE

Rudd CE. (1999). Cell 96, 5-8. MEDLINE

Schneider H, Cai YC, Prasad KV, Shoelson SE, Rudd CE. (1995). Eur. J. Immunol. 25, 1044-1050. MEDLINE

Songyang Z, Shoelson SE, McGlade J, Olivier P, Pawson T, Bustelo XR, Barbacid M, Sabe H, Hanafusa H, Yi T. (1994). Mol. Cell. Biol. 14, 2777-2785. MEDLINE

Su B, Jacinto E, Hibi M, Kallunki T, Karin M, Ben-Neriah Y. (1994). Cell 77, 727-736. MEDLINE

Trub T, Frantz JD, Miyazaki M, Band H, Shoelson SE. (1997). J. Biol. Chem. 272, 894-902. MEDLINE

Ullman KS, Northrop JP, Admon A, Crabtree GR. (1993). Genes Dev. 7, 188-196. MEDLINE

Verweij CL, Guidos C, Crabtree GR. (1990). J. Biol. Chem. 265, 15788-15795. MEDLINE

Wang W, Zhou G, Hu MCT, Yao Z, Tan TH. (1997). J. Biol. Chem. 272, 22771-22775. Article MEDLINE

Wange RL, Guitian R, Isakov N, Watts JD, Aebersold R, Samelson LE. (1995). J. Biol. Chem. 270, 18730-18733. MEDLINE

Wange RL, Samelson LE. (1996). Immunity 5, 197-205. MEDLINE

Weber JR, Orstavik S, Torgersen KM, Danbolt NC, Berg SF, Ryan JC, Tasken K, Imboden JB, Vaage JT. (1998). J. Exp. Med. 187, 1157-1161. MEDLINE

Weiss A, Koretzky G, Schatzman RC, Kadlecek T. (1991). Proc. Natl. Acad. Sci. USA 88, 5484-5488. MEDLINE

Weiss A, Littman DR. (1994). Cell 76, 263-274. MEDLINE

Whitmarsh AJ, Cavanagh J, Tournier C, Yasuda J, Davis RJ. (1998). Science 281, 1671-1674. Article MEDLINE

Whitmarsh AJ, Davis RJ. (1998). Trends Biochem. Sci. 23, 481-485. Article MEDLINE

Xia C, Bao Z, Tabassam F, Ma W, Qiu M, Hua S, Liu M. (2000). J. Biol. Chem. 275, 20942-20948. MEDLINE

Zhang W, Samelson LE. (2000). Semin. Immunol. 12, 35-41. Article MEDLINE

Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. (1998). Cell 92, 83-92. MEDLINE

Zhang W, Trible RP, Zhu M, Liu SK, McGlade CJ, Samelson LE. (2000). J. Biol. Chem. 275, 23355-23361. Article MEDLINE

Zhou G, Lee SC, Yao Z, Tan TH. (1999). J. Biol. Chem. 274, 13133-13138. MEDLINE

Figure 1 Interaction of Grap2 with HPK1 in vitro and in mammalian cells. (a) Jurkat T cells were either left unstimulated or stimulated with an anti-CD3 antibody (OKT3) and lysed. Lysates were precipitated using GST-Grap2 fusion protein. GST alone was used as a negative control and did not show any significant binding affinity to any phosphotyrosine proteins. The precipitates were blotted with anti-phosphotyrosine antibody 4G10, followed by chemiluminescent detection. (b) HPK1 was detected in the GST-Grap2 precipitates by anti-HPK1 antibody in both stimulated or unstimulated Jurkat cells. GST alone did not bind to HPK1. (c) Treatment of Jurkat cells with OKT3 or pervanadate increased the tyrosine phosphorylation of HPK1. Cell lysates were immunoprecipitated (IP) using an anti-HPK1 antibody. Phosphorylated HPK1 was detected by the anti-phosphotyrosine antibody 4G10. Bottom panel: Western blot shows the amount of HPK1 protein is similar in the assay. A preimmune rabbit serum was used as negative control and did not show any binding to either HPK1 or Grap2. (d) Association of Grap2 with HPK1 in mammalian cells. COS-7 cells were transfected with cDNAs encoding Grap2 or Grap2 and HPK1. Cell lysates were immunoprecipitated using a polyclonal rabbit anti-Grap2 antibody, followed by a goat anti-rabbit antibody conjugated to agarose beads. The immunoprecipitates were blotted by an anti-HPK1 antibody and detected by chemiluminescent method. Similar binding of Grap2 with endogenous HPK1 in Jurkat T cells were obtained (data not shown). For each figure, at least 2-3 independent experiments were performed

Figure 2 Grap2 carboxyl SH3 (SH3-C) domain interacts with HPK1 in yeast cells. (a) Interaction of the QC domain and the SH3-C domain of Grap2 with HPK1 in yeast cells. The QC domain of Grap2 contains the glutamine-/proline-rich domain and the C-terminal SH3 domain. Cotransformation of expression plasmids encoding the QC domain or the SH3-C domain of Grap2 with HPK1-CD allows the yeast cells to grow on selection medium plates (SD/-Trp/-Leu/-His). However, coexpression of Grap2-SH3(N)-SH2 or Grap2-Q/P domain with HPK1 failed to confer yeast cell growth on the same selective plates. Plasmids encoding p53 and large T antigen were used as the positive control. (b) Activation of lacZ reporter gene activity in yeast cells cotransformed with Grap2-QC domain or Grap2-SH3-C domain with HPK1. Like p53 and large T antigen, cotransformation of Grap2-QC or Grap2-SH3-C domain with HPK1-CD domain significantly increased lacZ gene expression, and therefore, the -galactosidase activity. For simplicity, HPK1-CD is labeled as HPK1 in the figures

Figure 3 Mapping the domain of Grap2 responsible for the interaction with HPK1 in vitro. (a) GST-Grap2 fusion proteins, including GST-Grap2 full-length protein, GST-Grap2-SH3(N)-SH2, GST-Grap2-Q/P-rich domain, and GST-Grap2-SH3-C domain. The proteins were constructed in pGEX-4T (Pharmacia) and expressed in BL21 cells. Proteins were purified to 95% homology by glutathione-agarose column as assayed on SDS-PAGE. (b) in vitro binding of GST-Grap2 and Grap2-CSH3 domain with HPK1. COS-7 cells were transfected with Flag-tagged HPK1. GST-Grap2 fusion proteins were incubated with the cell lysates. The protein complexes were precipitated by glutathione-agarose beads and washed extensively. HPK1 was detected by the anti-Flag M2 monoclonal antibody. The same amount of GST-fusion proteins (2 g) was used in the binding assays. Bottom panel: Western blot analysis of HPK1 shows similar amount of proteins used in the assay

Figure 4 The proline-rich domain of HPK1 is responsible for the interaction with Grap2. (a) A schematic diagram showing different HPK1 constructs and four proline-rich peptides. Wild-type HPK1 and four HPK1 mutant constructs, including HPK1-K46M, HPK1-KD (kinase domain), HPK1-CD (the carboxyl domain, from amino acids 292 to 833), and HPK1-PR (proline-rich domain, amino acids 288-482), were used in the assays. Shown on the right are the sequences for the four proline-rich peptides synthesized (PR1, PR2, PR3, and PR4). (b) in vitro transcription and translation of HPK1 full-length and mutant proteins in the presence of [³⁵S]methionine. Proteins were separated by SDS-PAGE and detected by autoradiography. (c) Binding of HPK1 domains to Grap2 protein. [³⁵S]methionine-labeled wild-type HPK1 and its mutant proteins (10 l) were incubated with immobilized GST-Grap2-QC fusion protein. The interacting complexes were washed extensively and resolved by 10% SDS-PAGE. HPK1 and mutants were detected by autoradiography. (d) HPK1 proline-rich peptides differentially block the interaction of HPK1 with Grap2-SH3-C domain. Four HPK1 proline-rich peptides were added to block the interaction of HPK1 with Grap2 in competition binding assays. The amount of HPK1 bound to GST-Grap2 beads was detected by Western blot using an anti-HPK1 antibody

Figure 5 Grap2 activates HPK1 and potentiates the HPK1-mediated JNK activation. (a) Coexpression of Grap2 with HPK1 activated the kinase activity of HPK1 by approximately twofold using MBP as a substrate. COS-7 cells were transfected with HPK1 alone or with HPK1 plus Grap2. HPK1 was immunoprecipitated with the anti-Flag M2 monoclonal antibody, and an equal amount of HPK1 was used for immunocomplex kinase assays. (b) Cotransfection of Grap2 with HPK1 activated JNK activity synergistically. HA-tagged JNK was transfected into COS-7 cells alone or with 1 g of expression plasmids as indicated. Proteins were immunoprecipitated by an anti-HA monoclonal antibody. JNK activity was determined by immunocomplex kinase assays using GST-c-Jun (amino acids 1-79) as a substrate for HA-JNK1. Coexpression of Grap2 and HPK1 has synergistic effect on the activation of JNK. The expression level of JNK in transfected cells was detected by Western blot with anti-HA antibodies

Figure 6 Cooperation between Grap2 and HPK1 in the transcriptional activation of c-Jun and IL-2. (a) Activation of c-Jun transcription in the JNK signaling pathway by cotransfection of Grap2 and HPK1. The activation of c-Jun was assayed by the PathDetect c-Jun trans-reporting system (Stratagene, CA, USA). cDNAs encoding c-Jun/Gal4 fusion protein and the luciferase reporter pFR-Luc were cotransfected into COS-7 cells with individual expression vectors indicated. HA-JNK1 was used as positive controls in the assay. The activation of luciferase activity in the cell was performed according to the manufacturer's manual. Triplicates were performed for the same transfection assay. At least three transfection experiments were performed. (b) Coexpression of Grap2 and HPK1 increased the IL-2 luciferase reporter activity. Jurkat T cells were transfected with a control plasmid encoding lacZ, or expression plasmids encoding Grap2, Grap2 mutants, HPK1, and Grap2 plus HPK1 as indicated. The activation of IL-2 luciferase activity was assayed in unstimulated cells or cells stimulated with an anti-CD3 antibody plus PMA (dark bars). The expression level of Grap2, Grap2 mutants, and HPK1 were similar in each transfection assay as estimated by Western blot using respective antibodies