Differential Signaling after beta 1 Integrin Ligation Is Mediated Through Binding of CRKL to p120CBL and p110HEF1*

(Received for publication, September 9, 1996, and in revised form, February 14, 1997)

Martin Sattler Dagger , Ravi Salgia Dagger , Gautam Shrikhande Dagger , Shalini Verma Dagger , Naoki Uemura Dagger , Susan F. Law §, Erica A. Golemis § and James D. Griffin Dagger

From the Dagger  Division of Hematologic Malignancies, Dana-Farber Cancer Institute, Boston, Massachusetts 02115 and the § Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES


ABSTRACT

CRKL is an SH2-SH3-SH3 adapter protein that is a major substrate of the BCR/ABL oncogene. The function of CRKL in normal cells is unknown. In cells transformed by BCR/ABL we have previously shown that CRKL is associated with two focal adhesion proteins, tensin and paxillin, suggesting that CRKL could be involved in integrin signaling. In two hematopoietic cell lines, MO7e and H9, we found that CRKL rapidly associates with tyrosine-phosphorylated proteins after cross-linking of beta 1 integrins with fibronectin or anti-beta 1 integrin monoclonal antibodies. The major tyrosine-phosphorylated CRKL-binding protein in the megakaryocytic MO7e cells was identified as p120CBL, the cellular homolog of the v-Cbl oncoprotein. However, in the lymphoid H9 cell line, the major tyrosine-phosphorylated CRKL-binding protein was p110HEF1. In both cases, this binding was mediated by the CRKL SH2 domain. Interestingly, although both MO7e and H9 cells express p120CBL and p110HEF1, beta 1 integrin cross-linking induces tyrosine phosphorylation of p120CBL (but not p110HEF1) in MO7e cells and of p110HEF1 (but not p120CBL) in H9 cells. In both cell types, CRKL is constitutively complexed to C3G, SOS, and c-ABL through its SH3 domains, and the stoichiometry of these complexes does not change upon integrin ligation. Thus, in different cell types CRKL and its SH3-associated proteins may form different multimeric complexes depending on whether p120CBL or p110HEF1 is tyrosine-phosphorylated after integrin ligation. The shift in association of CRKL and its SH3-associated proteins from p120CBL to p110HEF1 could contribute to different functional outcomes of "outside-in" integrin signaling in different cells.


INTRODUCTION

Integrins play a role in cell movement and apoptosis and also act as costimulatory molecules. The integrin receptors are alpha /beta heterodimeric transmembrane proteins that mediate cell-cell or cell-extracellular matrix interactions. Activation of integrin receptors leads to the formation of focal adhesions where integrin cytoplasmic domains are connected with actin-containing cytoskeleton components, thereby providing a link between the extracellular environment and intracellular elements. Tyrosine phosphorylation of cellular proteins is an early event after integrin receptor stimulation and is believed to initiate a series of signaling events involving protein-protein interactions leading to changes in viability, proliferation, or other functions in various cells (1, 2). One tyrosine kinase that is localized to the focal adhesion and is activated after integrin ligation has been identified as p125FAK (3). This kinase may have a negative regulatory role in the formation of focal adhesions (4). Also, another non-receptor tyrosine kinase (related adhesion focal tyrosine kinase) has been found to be partially associated with the actin cytoskeleton and is activated by integrins (5, 6). Recently, investigators have begun to identify the major cellular proteins that are tyrosine-phosphorylated after cross-linking of integrins by ligands. For example, p120CBL is tyrosine-phosphorylated after beta 1 integrin ligation in the human B cell line Nalm-6 and after beta 1 and beta 2 integrin ligation in the megakaryoblastic cell line MO7e (7, 8).

In many signal transduction pathways activated by tyrosine kinases, adapter molecules have been shown to play a key role in mediating transient protein-protein interactions. We have previously shown that the adapter protein CRKL is associated with the focal adhesion protein paxillin in cells transformed by the oncogenic tyrosine kinase BCR/ABL (9). CRKL is a 39-kDa protein with one SH2 and two SH3 domains (10). CRKL has a high homology to c-CRK-II and belongs to the CRK family of adapter proteins, which includes v-CRK, c-CRK-II, and c-CRK-I (11-13). The CRK and CRKL SH3 domains have been shown to specifically bind to c-ABL, SOS, or C3G (14-19). The SH2 domain of CRKL has been shown to bind to p120CBL in cells transformed by oncogenic tyrosine kinases (19, 20), and CRKL binds p120CBL inducibly after epidermal growth factor receptor stimulation (21) or after T cell receptor stimulation (22).

In this study, we examined the involvement of CRKL in signal transduction pathways activated after cross-linking of beta 1 integrins in two hematopoietic cell lines, the megakaryoblastic cell line MO7e and a T cell line, H9. In both cell lines, beta 1 integrin stimulation resulted in the rapid association of CRKL with a single major tyrosine-phosphorylated cellular protein. Surprisingly, however, this protein was of a different apparent molecular mass in the two cell lines. We found that p120CBL was the major tyrosine-phosphorylated CRKL-binding protein in MO7e cells, and p110HEF1 was the major tyrosine-phosphorylated CRKL-binding protein in H9 cells. In both cases the binding was mediated through the CRKL SH2 domain, while proteins constitutively associated with the CRKL SH3 domain, including C3G, SOS, and c-ABL, did not appear to be affected by beta 1 integrin stimulation. These results indicate that CRKL and its associated signaling proteins can interact with more than one signaling pathway activated by beta 1 integrin ligation.


MATERIALS AND METHODS

Cells

The human megakaryoblastic cell line MO7e (obtained from Dr. Steve Clark, Genetics Institute, Cambridge, MA) was maintained in Dulbecco's modified Eagle's medium (Mediatech, Washington, D. C.), 10 ng/ml granulocyte-macrophage colony-stimulating factor (Genetics Institute), and 20% (v/v) fetal calf serum (PAA Laboratories Inc., Newport Beach, CA) at 37 °C with 10% CO2. The BCR/ABL-expressing MO7e cell line MO7/p210 was generated by transfection with the pGD vector containing the sequence for the p210BCR/ABL cDNA (obtained from Dr. George Daley, MIT, Cambridge, MA). For stimulation studies, MO7e cells were washed with Dulbecco's phosphate-buffered saline (DPBS)1 and deprived of growth factors for 20 h at 37 °C in serum-free medium with 1% (w/v) bovine serum albumin (Sigma). The human T cell line H9 (obtained from Dr. Jerome Ritz, Dana-Farber Cancer Institute) was maintained in RPMI 1640 (Mediatech) and 10% (v/v) fetal calf serum (PAA Laboratories Inc.) at 37 °C with 5% CO2. Starved H9 cells were prepared by washing with DPBS and were deprived of serum for 2 h at 37 °C in serum-free medium.

Stimulation of Cells and Preparation of Cellular Lysates

Starved MO7e or H9 cells were first incubated for 15 min on ice with antibodies against CD29/beta 1 integrin (4B4, obtained from Dr. C. Morimoto, Dana-Farber Cancer Institute), CD3 (OKT3, Coulter Corp., Miami, FL), or an irrelevant antibody (3C11C8, an anti-interferon-gamma murine monoclonal antibody) and then stimulated by cross-linking using affinity-purified rabbit anti-mouse Ig (Dako Corp., Carpinteria, CA) at 37 °C for 10 min. For alpha  integrin subunit cross-linking, starved MO7e and H9 cells were incubated for 30 min on ice with antibodies against alpha 4 integrins (8F2 from Dr. C. Morimoto and B5G10 from Dr. M. Hemler, Dana-Farber Cancer Institute) or against alpha 5 integrins (2H6 from Dr. C. Morimoto and A5-PUJ2 from Dr. M. Hemler) and then stimulated by cross-linking for 20 min as described above. Either starved H9 cells or MO7e cells (washed three times in DPBS after starvation and resuspended in Dulbecco's modified Eagle's medium) were used for stimulation with fibronectin (Life Technologies, Inc.) in the same fashion. Cell lysates were prepared as described (23).

Immunoprecipitation and Western Blotting

Western blotting using a chemiluminescence technique was performed as described (23). Immunochemical detection of tyrosine-phosphorylated proteins in Western blots utilized monoclonal antibody 4G10 (kindly provided by Dr. B. Druker, Oregon Health Science University, Portland, OR). Polyclonal rabbit antisera against p120CBL (Santa Cruz Biotechnology, Santa Cruz, CA), CRKL (Santa Cruz Biotechnology), p110HEF1 (HEF1 SB) (24), and mouse monoclonal antibodies against c-ABL (AB-3 from Oncogene Science, Manhasset, NY) and CRKL (the mouse monoclonal was generated as described elsewhere (25) and only used for Western blotting) were used for this study. The pGEX vector containing the SH2 and SH3-SH3 domains of CRKL was obtained from Dr. J. Groffen, Children's Hospital, UCLA, Los Angeles, CA. The GST-fusion proteins were expressed in Escherichia coli (DH-5alpha ) by isopropyl-1-thio-beta -D-galactopyranoside induction and isolated from sonicated bacterial lysates using glutathione-Sepharose beads (Pharmacia Biotech Inc.) according to the manufacturer's directions.

Flow Cytometry Analysis

MO7e and H9 cells (0.5 × 106 cells/sample) were incubated with murine monoclonal antibodies against integrin receptors including alpha 1 (TS2/7), alpha 2 (A2-2E10), alpha 3 (A3-2F5), alpha 4 (B5G10), alpha 5 (A5-PUJ2), alpha 6 (A6-ELE) (all anti-alpha integrin receptor antibodies were obtained from Dr. M. Hemler, Dana-Farber Cancer Institute), beta 1 (4B4), or an irrelevant monoclonal antibody (3C11C8) for 20 min on ice and then washed once with DPBS. Cells were incubated with fluorescein isothiocyanate-conjugated goat anti-mouse serum (Southern Biotechnology Assoc., Birmingham, AL) for an additional 20 min and subsequently washed twice in DPBS before analysis using a Coulter Epics XL flow cytometer (Coulter Corp.) for analysis.

Far-Western Blotting

Using previously established techniques (26), far-Western blotting was performed as described previously (19). In brief, immunoprecipitated proteins were transferred after SDS-PAGE to Immobilon-P (polyvinylidene difluoride) membrane (Millipore) and blocked with 5% nonfat dry milk in 0.1% Tween 20 in phosphate-buffered saline, pH 7.4. The specific direct in vitro binding was evaluated by probing the membrane with GST-fusion proteins and visualized with a combination of anti-GST monoclonal antibody (Santa Cruz Biotechnology) and horseradish peroxidase-coupled anti-mouse IgG antibody by chemiluminescence.


RESULTS

CRKL Binds to p120CBL after Fibronectin Stimulation in MO7e Cells and to p110HEF1 in H9 Cells

To investigate the potential role of CRKL in integrin signaling we looked for tyrosine-phosphorylated proteins that coprecipitate with CRKL, since tyrosine phosphorylation of cellular proteins is an early event following integrin receptor ligation. To determine if CRKL associates with tyrosine-phosphorylated proteins, we investigated two different hematopoietic cell lines with known differences in tyrosine phosphorylation of cellular proteins after beta 1 integrin stimulation.

For initial experiments we stimulated MO7e and H9 cells with fibronectin, a natural ligand for some integrin receptors including VLA-4 (alpha 4beta 1) and VLA-5 (alpha 5beta 1), which are the major beta 1 integrin receptors in MO7e cells as well as in H9 cells. In MO7e cells, fibronectin induced association of CRKL with a prominent 120-kDa tyrosine phosphoprotein (Fig. 1A, left panel). We have previously shown that beta 1 integrin ligation induces tyrosine phosphorylation of p120CBL in MO7e cells. We therefore asked if the 120-kDa protein coprecipitating with CRKL in MO7e cells was p120CBL. This blot was stripped, and the phosphoprotein was identified as p120CBL by immunoblotting (Fig. 1A, upper right panel). The same results were obtained when the immunoprecipitation and blotting antibodies were reversed (data not shown). The lower right panel in Fig. 1A demonstrates that equal amounts of CRKL were loaded in each lane.


Fig. 1. CRKL forms a stable complex with p120CBL in MO7e cells and p110HEF1 in H9 cells after fibronectin stimulation. Unstimulated (CTRL) or fibronectin (FN)-stimulated MO7e cells or H9 cells (20 × 106 cells) were used for immunoprecipitation. A, growth factor-deprived MO7e cells were stimulated for 15 min with fibronectin (FN). Cell lysates were immunoprecipitated with antisera to CRKL and immunoblotted with either anti-phosphotyrosine antibody (p-TYR), antisera to p120CBL (CBL), or CRKL as indicated. B, serum-starved H9 cells were stimulated for 15 min with fibronectin (FN). Cell lysates were immunoprecipitated with antisera to CRKL and immunoblotted with either anti-phosphotyrosine antibodies (p-TYR), antisera to p110HEF1 (HEF1), or CRKL as indicated.
[View Larger Version of this Image (31K GIF file)]

In H9 cells, CRKL was also found to associate with tyrosine-phosphorylated proteins. However, a 110-kDa tyrosine-phosphorylated protein coprecipitated with CRKL after fibronectin stimulation (Fig. 1B, left panel). This protein did not react with a p120CBL antibody (data not shown). Based on its molecular mass and the presence of multiple potential CRKL SH2 binding motifs (Tyr-X-X-Pro), we examined p110HEF1 for possible coprecipitation with CRKL. The blot was stripped, and the phosphoprotein was identified as p110HEF1 by immunoblotting (Fig. 1B, upper right panel). The lower right panel in Fig. 1B demonstrates that comparable amounts of CRKL were loaded. These results demonstrate that integrin receptor activation with fibronectin can induce the formation of a CRKL-p120CBL complex in MO7e cells and a CRKL-p110HEF1 complex in H9 cells. However, we did not detect significant association of p120CBL with CRKL in H9 cells or with p110HEF1 in MO7e cells at any time points tested between 0 and 60 min (data not shown).

p120CBL and p110HEF1 Are Differentially Tyrosine-phosphorylated after beta 1 Integrin Ligation in MO7e Cells and H9 Cells

Since we observed differential association of tyrosine-phosphorylated p120CBL and p110HEF1 with CRKL in MO7e cells or H9 cells, respectively, we asked if these proteins were differentially tyrosine-phosphorylated after beta 1 integrin ligation in these cells. Stimulation of the megakaryocytic MO7e cells or the T cell line H9 with a monoclonal antibody to cross-link beta 1 integrins induced rapid tyrosine phosphorylation of cellular proteins compared with unstimulated cells (Fig. 2A, left panel). Mock stimulation with an irrelevant antibody (3C11C8, an anti-interferon-gamma murine monoclonal antibody) did not induce tyrosine phosphorylation (data not shown). The major tyrosine-phosphorylated proteins in MO7e cells include proteins with apparent molecular masses of 145, 120, 95, 70, and 40 kDa, whereas in H9 cells two prominent proteins of 110 and 95 kDa were tyrosine-phosphorylated. H9 cells treated with an irrelevant antibody also did not induce tyrosine phosphorylation of cellular proteins. The tyrosine phosphorylation pattern induced by fibronectin was virtually identical to the beta 1 integrin-induced pattern.


Fig. 2. Differential activation of p120CBL in MO7e cells and p110HEF1 in H9 cells. A, unstimulated (-) or beta 1 integrin-stimulated (+) MO7e cells or H9 cells (20 × 106 cells) were used for immunoprecipitation. Cell lysates were immunoprecipitated with anti-phosphotyrosine antibodies (p-TYR, PY20) and immunoblotted with either anti-phosphotyrosine antibodies (p-TYR, 4G10), antisera to p120CBL (CBL), or antisera to p110HEF1 as indicated. B, total cell lysate of MO7e cells or H9 cells was separated by SDS-PAGE, and protein expression of p120CBL and p110HEF1 was detected by Western blotting. C, unstimulated MO7e cells or MO7/p210 cells (15 × 106 cells) were used for immunoprecipitation. Cell lysates were immunoprecipitated with anti-phosphotyrosine antibodies (p-TYR, PY20) and immunoblotted with either anti-phosphotyrosine antibodies (p-TYR, 4G10) or antisera to p110HEF1 as indicated. D, H9 cells (15 × 106 cells) incubated with an irrelevant antibody (CTRL) or CD3 receptor-stimulated were used for immunoprecipitation. Cell lysates were immunoprecipitated with antisera to p120CBL (CBL) and immunoblotted with either anti-phosphotyrosine antibodies (p-TYR) or antisera to p120CBL (CBL) as indicated. E, lysates of MO7e cells or H9 cells (30 × 106 cells) incubated with an irrelevant antibody (-) or alpha 4 integrin-stimulated (+) were used for immunoprecipitation. Cell lysates were immunoprecipitated with antisera to CRKL and immunoblotted with either anti-phosphotyrosine antibodies (p-TYR), antisera to p120CBL (CBL), antisera to p110HEF1 (HEF1), or an antibody to CRKL. F, expression of alpha  integrins and beta 1 integrins on MO7e cells (gray bars) and H9 cells (black bars) was analyzed by flow cytometry using specific monoclonal antibodies or an irrelevant monoclonal antibody (CTRL) as a negative control. The relative fluorescence intensities are indicated as mean values.
[View Larger Version of this Image (36K GIF file)]

We identified the 120-kDa protein as p120CBL in phosphotyrosine immunoprecipitations of stimulated MO7e cells but not H9 cells (Fig. 2A, middle panel). In contrast, the 110-kDa protein in the phosphotyrosine immunoprecipitation of H9 cells was found to be the recently cloned p130CAS-related protein p110HEF1 (Fig. 2A, right panel). In addition, p120CBL and p110HEF1 were also inducibly (but again selectively) tyrosine-phosphorylated with fibronectin stimulation in MO7e and H9 cells, respectively (data not shown). Interestingly, p110HEF1 is not tyrosine-phosphorylated in MO7e cells. The increased tyrosine phosphorylation of p120CBL and p110HEF1 is likely to mediate the specific interaction with CRKL after integrin cross-linking (Fig. 1). The differences in phosphorylation of p120CBL or p110HEF1 could not be attributed to differential expression of p120CBL and p110HEF1 as expression of these proteins in MO7e and H9 cells by Western blotting was comparable (Fig. 2B). In addition to p110HEF1, the antiserum to p110HEF1 recognized a 95-kDa protein in Western blot experiments (Fig. 2B, right panel). The identity of the 95-kDa protein is not known at this time. However, our preliminary data suggest that it may be the SH3 domain-containing p130CAS-related protein p95EFS/SIN (27, 28) (data not shown).

The failure to tyrosine-phosphorylate p110HEF1 in MO7e cells and p120CBL in H9 cells could be due to defects in signaling pathways leading to tyrosine phosphorylation of these proteins. To address this issue, other pathways known to induce phosphorylation of p120CBL and p110HEF1 were examined. We found p110HEF1 in phosphotyrosine immunoprecipitates of MO7e cells expressing the oncogenic tyrosine kinase BCR/ABL but not in untransfected cells, demonstrating apparent phosphorylation of p110HEF1 in response to BCR/ABL (Fig. 2C). Also, p120CBL was found to be inducibly tyrosine-phosphorylated after CD3 cross-linking in H9 (Fig. 2D). These data suggest that p120CBL and p110HEF1 can be tyrosine-phosphorylated in both cell lines by stimuli other than beta 1 integrin receptor cross-linking. Tyrosine-phosphorylated p120CBL and p110HEF1 were also found to be inducibly and selectively associated with CRKL after cross-linking with alpha 4 integrin in MO7e and H9 cells, respectively (Fig. 2E). Cross-linking of alpha 5 integrin in MO7e cells produced similar results; however, the increased association of CRKL with p110HEF1 was very small in H9 cells (data not shown). We further tested if different signaling was due to differences in alpha  integrin or beta 1 integrin receptor expression. We found that both cell lines had comparable expression of the beta 1 integrin as well as alpha 4, alpha 5, and alpha 6 integrins. Expression of alpha 1, alpha 2, and alpha 3 integrins was lower or negligible (Fig. 2F). Overall these results demonstrate that similar integrin receptors can activate distinct signaling proteins in different cell lines.

In Vitro Association of CRKL GST-fusion Proteins with p120CBLand p110HEF1

The above results suggest the potential induction of one or more multimeric protein complexes containing CRKL, p120CBL, or p110HEF1. The binding of CRKL to p120CBL and p110HEF1 appears to require tyrosine phosphorylation of these proteins. Since CRKL has one SH2 and two adjacent SH3 domains, we sought to determine the mechanism of CRKL binding to p120CBL and p110HEF1 using GST-fusion proteins containing various segments of each protein. The SH2 domain of CRKL precipitated p120CBL from lysates of stimulated (but not unstimulated) MO7e cells (Fig. 3A). The blots were stripped and reprobed with antibodies against c-ABL demonstrating that GST-CRKL-SH3 but not GST-CRKL-SH2 constitutively precipitated c-ABL (Fig. 3A). We also found constitutive coprecipitation of C3G and SOS with the ABL-SH3 domain (data not shown). Using lysates from H9 cells, the SH2 domain of CRKL precipitated p110HEF1 after fibronectin stimulation, while the GST-CRKL SH3 domain did not (Fig. 3B).


Fig. 3. Precipitation of p120CBL and p110HEF1 with CRKL GST-fusion proteins. Unstimulated (-) or beta 1 integrin (+)-stimulated MO7e cells or H9 cells were used for precipitations. A, lysates of 7.5 × 106 MO7e cells before (-) and after (+) beta  integrin stimulation (7.5 min, 37 °C) were incubated with 10 µg of GST-fusion protein and GST immobilized on glutathione beads. GST-fusion proteins of the SH2 domain (SH2) and both SH3 domains (SH3-SH3) of CRKL were used for precipitations. Coprecipitation of p120CBL (CBL) or c-ABL was detected by Western blotting. B, lysates of 15 × 106 H9 cells before (-) and after (+) beta 1 integrin stimulation (30 min, 37 °C) were incubated with 10 µg of GST-fusion protein and GST immobilized on glutathione beads. GST and GST-fusion proteins of the SH2 domain (SH2) and both SH3 domains (SH3-SH3) of CRKL were used for precipitations. Coprecipitation of p110HEF1 (HEF1) was detected by Western blotting. C, lysates of 10 × 106 unstimulated (-) or beta  integrin-stimulated cells (+) were incubated with antibodies against CRKL or p120CBL (CBL) as indicated. Proteins were separated by SDS-PAGE and transferred to polyvinylidene difluoride membrane, and specific direct binding of GST-fusion proteins to p120CBL was detected by far-Western blotting. GST and GST-fusion proteins of the CRKL SH2 domain (CRKL-SH2) were used.
[View Larger Version of this Image (23K GIF file)]

The in vitro GST-fusion protein precipitations with p120CBL, c-ABL, and CRKL do not indicate if binding of the SH2 or SH3 domains is direct or indirect. We therefore used a far-Western technique to examine possible direct in vitro interactions. Cellular lysates from unstimulated and beta  integrin-stimulated MO7e cells were used for immunoprecipitations with anti-p120CBL or anti-CRKL antibody. Fig. 3C shows that GST protein alone does not bind to proteins in p120CBL or CRKL immunoprecipitations. Direct binding of a single 120-kDa protein band in CRKL immunoprecipitations using the GST-CRKL SH2 protein as a probe was found only after beta  integrin ligation. This protein was identified as p120CBL in the p120CBL immunoprecipitation and the CRKL SH2 far-Western blot. We also observed binding of the CRKL SH3 domain to a set of proteins between 140 and 160 kDa; however, this interaction was not changed upon beta  integrin ligation (data not shown). The binding of SH2 domains to p120CBL is likely to require phosphotyrosine, since no binding was observed to p120CBL in lysates from unstimulated cells where tyrosine phosphorylation of p120CBL is not induced. These results indicate that in MO7e cells, CRKL is linked through its SH2 domain to a pathway involving p120CBL, whereas in H9 cells CRKL is linked to a pathway involving p110HEF1.

beta Integrin Ligation Does Not Alter the Binding of CRKL to SOS, C3G, or c-ABL

We also asked if integrin ligation changes any complexes of CRKL with SH3-binding proteins. Fig. 4 demonstrates that beta 1 integrin ligation did not alter the coprecipitation of CRKL with c-ABL, C3G, and SOS. The same results were obtained when the immunoprecipitation and blotting antibodies were reversed (data not shown). We did not observe detectable induction of tyrosine phosphorylation of c-ABL after integrin ligation (data not shown). These data demonstrate that integrin cross-linking does not alter the constitutive complexes of CRKL with c-ABL, C3G, and SOS.


Fig. 4. beta 1 integrin ligation does not alter the constitutive association of CRKL with c-ABL, C3G, and SOS. Lysates of unstimulated (CTRL) or beta 1 integrin (beta 1)-stimulated MO7e cells (20 × 106 cells) were used for immunoprecipitation. Cell lysates were immunoprecipitated with preimmune serum (NRS) or antisera to CRKL and immunoblotted with antisera to CRKL, c-ABL, SOS, or C3G as indicated.
[View Larger Version of this Image (22K GIF file)]


DISCUSSION

The biological effects of cross-linking integrins may vary widely from cell to cell, ranging from stimulation of proliferation to induction of apoptosis. When integrins are cross-linked through binding with a natural ligand such as fibronectin, a series of signaling events are initiated. This signaling is associated with the following changes in the actin cytoskeleton: formation of a cytoskeletal complex of proteins that includes actin, vinculin, talin, p125FAK, paxillin, and tensin, activation of tyrosine phosphorylation, and activation of other signal transduction pathways such as the p21RAS pathway. Overall, this outside-in signaling of integrins is likely to be an important part of the signals sent by the microenvironment to influence cell behavior (1, 2).

However, the mechanisms of outside-in signaling are not well understood. This is due in part to the complexity of studying a system with many related receptors (the integrin family) that are expressed heterogeneously on different cell types, coupled with the fact that different integrins may share the same ligand. Since the biological effects of outside-in signaling may vary widely in different cells, it is of interest to determine how integrin cross-linking in one cell may augment proliferation but induce apoptosis in another cell type. It would be anticipated that different integrins may activate different signaling pathways in the same cell and also that the same integrin could potentially activate different pathways in different cells. Despite this prediction, there are few examples of differential signaling by integrins and even fewer examples where specific integrin-activated signaling pathways have been directly linked to a biological event.

In this study we have investigated the specific role of CRKL (an adapter protein that has one SH2 domain and two SH3 domains) in integrin signaling as part of a larger effort to understand the cellular functions of CRKL. During preliminary studies in the human megakaryoblastic cell line MO7e, we had noted that after cross-linking of beta 1 integrins by monoclonal antibody, CRKL was induced to bind through its SH2 domain to a 120-kDa protein identified as p120CBL. p120CBL was shown to be one of the most prominently tyrosine-phosphophorylated proteins induced after integrin activation in these cells (8) and virtually the only tyrosine phosphoprotein coprecipitating with CRKL. However, in another hematopoietic cell line, the T cell line H9, we noted that CRKL did not coprecipitate with p120CBL after integrin cross-linking, despite the fact that the H9 cell line was found to have the same pattern of beta 1 integrin expression as MO7e as well as abundant expression of p120CBL (29, 30). This unexpected result was made more interesting by the finding that CRKL was induced to coprecipitate with another tyrosine phosphoprotein in H9 cells, p110HEF1, which is a signaling protein related to p130CAS. Again, p110HEF1 was virtually the only tyrosine phosphoprotein coprecipitating with CRKL after integrin stimulation, and the interaction was mediated by the CRKL SH2 domain. Like p120CBL, p110HEF1 has multiple copies of potential CRKL SH2 binding motifs (phospho Tyr-X-X-Pro) (7, 24). These combined observations suggest that CRKL is not only involved in integrin-mediated outside-in signaling, it can also participate in different pathways depending on which upstream molecule (p120CBL or p110HEF1) is phosphorylated (probably at the phospho Tyr-X-X-Pro motifs previously shown to represent binding sites for CRK and CRKL SH2 domains). This provides for the possibility of an intracellular signaling "switch" that could couple integrin signaling to different biological effects.

In contrast to the effects of integrin-induced tyrosine phosphorylation on the binding of the CRKL SH2 domain to signaling molecules, the proteins that were bound to the CRKL SH3 domains were not affected by integrin cross-linking. The known CRKL SH3-binding proteins include c-ABL, C3G, and SOS. These proteins were first described as binding to the CRKII SH3 domain; however, we and others have shown that they also bind to the CRKL SH3 domain (14-19). SOS has known guanine-exchange factor activity for p21RAS; in contrast, C3G appears to have specific guanine exchange activity for p21RAP1. C3G does not have substrate specificity for p21RAS (31), but its substrate p21RAP1 appears to regulate, at least in part, the signal from p21RAS to the RAF kinase. C3G also shows sequence similarity to CDC25 and SOS family proteins (17) and preferentially binds to the N-terminal SH3 domain (16). The exact function of the tyrosine kinase c-ABL is unknown, although c-ABL has been shown to be involved in transcriptional activation (32) and possibly is activated in response to certain types of DNA damage (33). Interestingly, c-ABL can interact with the actin cytoskeleton through an actin binding site in its C terminus (34). During integrin signaling, c-ABL, C3G, or SOS could be linked to either p120CBL or p110HEF1 by CRKL, although no direct evidence of such multimeric proteins was demonstrated in this study.

The protooncoprotein p120CBL (for Casitas B-lineage lymphoma) is a widely expressed 120-kDa protein. It is the cellular homolog of v-Cbl, the oncoprotein in the CAS NS-1 retrovirus (35, 36) that induces pre-B cell lymphomas and myelogenous leukemias in mice (37). The p120CBL homolog Sli-1 in Caenorhabditis elegans is a negative regulator of the epidermal growth factor receptor tyrosine kinase homolog Let-23 (38). p120CBL is also known to be a substrate of tyrosine kinases in response to T cell (39) and B cell (40) activation, FC-gamma receptor cross-linking (41, 42), and growth factors (23, 43-45). In mammalian cells, the function of p120CBL is not known, although several interactions with other signaling proteins have been reported. For example, p120CBL has been shown to associate with active phosphatidylinositol 3-kinase in antigen receptor-stimulated cells or BCR/ABL transformed cells, and it interacts with the SH3 domains of GRB2, NCK, or SRC kinases including LYN and FYN (39, 42, 46). The H9 T cell line, which was derived from the HuT 78 cell line, expresses both a full-size c-CBL protein and a protein containing a C-terminal truncation of c-CBL (47). In H9 cells, we observed tyrosine phosphorylation of full-length p120CBL after CD3 stimulation, indicating that this p120CBL pathway is intact.

p110HEF1 (for human enhancer of filamentation 1) is a tissue-specific protein first identified during cloning of human genes that induce morphological changes in Saccharomyces cerevisiae. Expression of the p110HEF1 C terminus induces pseudohyphae in S. cerevisiae. p110HEF1 shares 64% homology with p130CAS and similarly has an N-terminal SH3 domain. p110HEF1 is also a prominent substrate of oncogenic tyrosine kinases including v-ABL and may function as a docking protein. This protein is structurally related to p130CAS, it appears to be localized to the nucleus and the cell periphery (24). It is not known if p110HEF1 in mammalian cells is also involved in organization of the cytoskeleton. Interestingly, p110HEF1 has several Tyr-X-X-Pro motifs (24) that have been shown to be recognized by the CRKL SH2 domain (9, 48). This is consistent with our findings demonstrating coprecipitation of CRKL with p110HEF1.

Our data demonstrate that beta 1 integrin receptors in MO7e or H9 cells are activating distinct signaling pathways. These differences are probably not mediated through different expression of beta 1 integrins since we demonstrate that the major beta 1 integrin receptors (the fibronectin receptors alpha 4beta 1 (VLA4) and alpha 5beta 1 (VLA5)) are expressed both in MO7e and H9 cells. Further, cross-linking alpha 4 integrin chains with specific monoclonal antibodies in MO7e and H9 cells also resulted in selective phosphorylation of p120CBL and P110HEF1, respectively. However, the beta 1 integrin family consists of four known isoforms (A, B, C, and D) that have the same extracellular domains but differ in their cytoplasmic domains. The major isoform beta 1A is ubiquitously expressed but is substituted in muscle cells by the beta 1D isoform and thus not expressed in hematopoietic cells (49). beta 1B integrin is a minor isoform and is coexpressed with beta 1A in some tissues and cells (50). The beta 1B isoform might negatively regulate adhesion and mobility (51). The beta 1C isoform is expressed in hematopoietic cells but does not appear to colocalize to focal adhesions and has been shown to cause growth arrest and inhibit DNA synthesis when transfected and expressed in fibroblasts (52). It is possible that the cell is using these distinct signaling pathways involving p120CBL or p110HEF1 depending on the differential expression of another regulatory signaling protein. Interestingly, there is at least one additional pattern of CRKL-related signaling in hematopoietic cells. We recently examined several additional cell lines and found that after beta 1 integrin ligation in the B cell line Nalm-6, CRKL binds to both tyrosine-phosphorylated p120CBL as well as p110HEF1.2 This suggests that the mechanisms that activate either pathway are not mutually exclusive. The differential activation of p120CBL and p110HEF1 could be directly mediated through a process that leads to activation of different tyrosine kinases that are specific for p120CBL or p110HEF1, and we are currently investigating this possibility.


FOOTNOTES

*   This work was supported by José Carreras International Leukemia Foundation Fellowship FIJC-95/INT (to M. S.), National Institutes of Health Grants CA60821 (to R. S.), R29-CA63366 (to E. A. G.), and CA36167 (to J. D. G.), American Cancer Society Fellowship PF4383 (to S. F. L.), and American Cancer Society Grant CB-74749 (to E. A. G.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
   To whom correspondence should be addressed: Div. of Hematologic Malignancies, Dana-Farber Cancer Inst., 44 Binney St., Boston, MA 02115. Tel.: 617-632-3360; Fax: 617-632-4388.
1   The abbreviations used are: DPBS, Dulbecco's phosphate-buffered saline; GST, glutathione S-transferase.
2   M. Sattler, unpublished data.

REFERENCES

  1. Schwartz, M. A., Schaller, M. D., and Ginsberg, M. H. (1995) Annu. Rev. Cell Dev. Biol. 11, 549-599 [CrossRef][Medline] [Order article via Infotrieve]
  2. Clark, E. A., and Brugge, J. S. (1995) Science 268, 233-239 [Medline] [Order article via Infotrieve]
  3. Lipfert, L., Haimovich, B., Schaller, M. D., Cobb, B. S., Parsons, J. T., and Brugge, J. S. (1992) J. Cell Biol. 119, 905-912 [Abstract]
  4. Ilic, D., Furuta, Y., Kanazawa, S., Takeda, N., Sobue, K., Nakatsuji, N., Nomura, S., Fujimoto, J., Okada, M., Yamamoto, T., and Aizawa, S. (1995) Nature 377, 539-544 [CrossRef][Medline] [Order article via Infotrieve]
  5. Avraham, S., London, R., Fu, Y., Ota, S., Hiregowdara, D., Li, J., Jiang, S., Pasztor, L. M., White, R. A., Groopman, J. E., and Avraham, H. (1995) J. Biol. Chem. 270, 27742-27751 [Abstract/Free Full Text]
  6. Li, J. Z., Avraham, H., Rogers, R. A., Raja, S., and Avraham, S. (1996) Blood 88, 417-428 [Abstract/Free Full Text]
  7. Blake, T. J., Shapiro, M., Morse, H. D., and Langdon, W. Y. (1991) Oncogene 6, 653-657 [Medline] [Order article via Infotrieve]
  8. Manié, S. N., Sattler, M., Astier, A., Phifer, J. S., Canty, T., Morimoto, C., Druker, B. J., Salgia, R., Griffin, J. D., and Freedman, A. S. (1997) Exp. Hematol. 25, 45-50 [Medline] [Order article via Infotrieve]
  9. Salgia, R., Uemura, N., Okuda, K., Li, J.-L., Pisick, E., Sattler, M., de Jong, R., Druker, B., Heisterkamp, N., Chen, L. B., Groffen, J., and Griffin, J. D. (1995) J. Biol. Chem. 270, 29145-29150 [Abstract/Free Full Text]
  10. ten Hoeve, J., Morris, C., Heisterkamp, N., and Groffen, J. (1993) Oncogene 8, 2469-2474 [Medline] [Order article via Infotrieve]
  11. Mayer, B. J., Hamaguchi, M., and Hanafusa, H. (1988) Nature 332, 272-275 [CrossRef][Medline] [Order article via Infotrieve]
  12. Reichman, C. T., Mayer, B. J., Keshav, S., and Hanafusa, H. (1992) Cell Growth & Differ. 3, 451-460 [Abstract]
  13. Matsuda, M., Tanaka, S., Nagata, S., Kojima, A., Kurata, T., and Shibuya, M. (1992) Mol. Cell. Biol. 12, 3482-3489 [Abstract]
  14. Feller, S. M., Knudsen, B., and Hanafusa, H. (1994) EMBO J. 13, 2341-2351 [Abstract]
  15. Matsuda, M., Hashimoto, Y., Muroya, K., Hasegawa, H., Kurata, T., Tanaka, S., Nakamura, S., and Hattori, S. (1994) Mol. Cell. Biol. 14, 5495-5500 [Abstract]
  16. Knudsen, B. S., Feller, S. M., and Hanafusa, H. (1994) J. Biol. Chem. 269, 32781-32787 [Abstract/Free Full Text]
  17. Tanaka, S., Morishita, T., Hashimoto, Y., Hattori, S., Nakamura, S., Shibuya, M., Matuoka, K., Takenawa, T., Kurata, T., Nagashima, K., and Matsuda, M. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 3443-3447 [Abstract]
  18. Feller, S. M., Knudsen, B., and Hanafusa, H. (1995) Oncogene 10, 1465-1473 [Medline] [Order article via Infotrieve]
  19. Sattler, M., Salgia, R., Okuda, K., Uemura, N., Durstin, M. A., Pisick, E., Xu, G., Li, J. L., Prasad, K. V., and Griffin, J. D. (1996) Oncogene 12, 839-846 [Medline] [Order article via Infotrieve]
  20. de Jong, R., ten Hoeve, J., Heisterkamp, N., and Groffen, J. (1995) J. Biol. Chem. 270, 21468-21471 [Abstract/Free Full Text]
  21. Fukazawa, T., Miyake, S., Band, V., and Band, H. (1996) J. Biol. Chem. 271, 14554-14559 [Abstract/Free Full Text]
  22. Reedquist, K. A., Fukazawa, T., Panchamoorthy, G., Langdon, W. Y., Shoelson, S. E., Druker, B. J., and Band, H. (1996) J. Biol. Chem. 271, 8435-8442 [Abstract/Free Full Text]
  23. Sattler, M., Durstin, M. A., Frank, D. A., Okuda, K., Kaushansky, K., Salgia, R., and Griffin, J. D. (1995) Exp. Hematol. 23, 1040-1048 [Medline] [Order article via Infotrieve]
  24. Law, S. F., Estojak, J., Wang, B. L., Mysliwiec, T., Kruh, G., and Golemis, E. A. (1996) Mol. Cell. Biol. 16, 3327-3337 [Abstract]
  25. Uemura, N., Salgia, R., Li, J-L., Pisick, E., Sattler, M., and Griffin, J. D. (1997) Leukemia 11, 376-385 [CrossRef][Medline] [Order article via Infotrieve]
  26. ten Hoeve, J., Kaartinen, V., Fioretos, T., Haataja, L., Voncken, J. W., Heisterkamp, N., and Groffen, J. (1994) Cancer Res. 54, 2563-2567 [Abstract]
  27. Ishino, M., Ohba, T., Sasaki, H., and Sasaki, T. (1995) Oncogene 11, 2331-2338 [Medline] [Order article via Infotrieve]
  28. Alexandropoulos, K., and Baltimore, D. (1996) Genes Dev. 10, 1341-1355 [Abstract]
  29. Bazzoni, G., Carlesso, N., Griffin, J. D., and Hemler, M. E. (1996) J. Clin. Invest. 98, 521-528 [Abstract/Free Full Text]
  30. Nojima, Y., Tachibana, K., Sato, T., Schlossman, S. F., and Morimoto, C. (1995) Cell. Immunol. 161, 8-13 [CrossRef][Medline] [Order article via Infotrieve]
  31. Gotoh, T., Hattori, S., Nakamura, S., Kitayama, H., Noda, M., Takai, Y., Kaibuchi, K., Matsui, H., Hatase, O., Takahashi, H., Kurata, T., and Matsuda, M. (1995) Mol. Cell. Biol. 15, 6746-6753 [Abstract]
  32. Welch, P. J., and Wang, J. Y. (1993) Cell 75, 779-790 [Medline] [Order article via Infotrieve]
  33. Yuan, Z. M., Huang, Y. Y., Whang, Y., Sawyers, C., Weichselbaum, R., Kharbanda, S., and Kufe, D. (1996) Nature 382, 272-274 [CrossRef][Medline] [Order article via Infotrieve]
  34. Van Etten, R. A., Jackson, P. K., Baltimore, D., Sanders, M. C., Matsudaira, P. T., and Janmey, P. A. (1994) J. Cell Biol. 124, 325-340 [Abstract]
  35. Langdon, W. Y., Hyland, C. D., Grumont, R. J., and Morse, H. D. (1989) J. Virol. 63, 5420-5424 [Medline] [Order article via Infotrieve]
  36. Langdon, W. Y., Hartley, J. W., Klinken, S. P., Ruscetti, S. K., and Morse, H. D. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 1168-1172 [Abstract]
  37. Fredrickson, T. N., Langdon, W. Y., Hoffman, P. M., Hartley, J. W., and Morse, H. D. (1984) J. Natl. Cancer Inst. 72, 447-454 [Medline] [Order article via Infotrieve]
  38. Yoon, C. H., Lee, J. H., Jongeward, G. D., and Sternberg, P. W. (1995) Science 269, 1102-1105 [Medline] [Order article via Infotrieve]
  39. Donovan, J. A., Wange, R. L., Langdon, W. Y., and Samelson, L. E. (1994) J. Biol. Chem. 269, 22921-22924 [Abstract/Free Full Text]
  40. Cory, G., Lovering, R. C., Hinshelwood, S., Maccarthymorrogh, L., Levinsky, R. J., and Kinnon, C. (1995) J. Exp. Med. 182, 611-615 [Abstract]
  41. Marcilla, A., Rivero-Lezcano, O. M., Agarwal, A., and Robbins, K. C. (1995) J. Biol. Chem. 270, 9115-9120 [Abstract/Free Full Text]
  42. Tanaka, S., Neff, L., Baron, R., and Levy, J. B. (1995) J. Biol. Chem. 270, 14347-14351 [Abstract/Free Full Text]
  43. Galisteo, M. L., Dikic, I., Batzer, A. G., Langdon, W. Y., and Schlessinger, J. (1995) J. Biol. Chem. 270, 20242-20245 [Abstract/Free Full Text]
  44. Meisner, H., Conway, B. R., Hartley, D., and Czech, M. P. (1995) Mol. Cell. Biol. 15, 3571-3578 [Abstract]
  45. Odai, H., Sasaki, K., Iwamatsu, A., Hanazono, Y., Tanaka, T., Mitani, K., Yazaki, Y., and Hirai, H. (1995) J. Biol. Chem. 270, 10800-10805 [Abstract/Free Full Text]
  46. Rivero-Lezcano, O. M., Sameshima, J. H., Marcilla, A., and Robbins, K. C. (1994) J. Biol. Chem. 269, 17363-17366 [Abstract/Free Full Text]
  47. Blake, T. J., and Langdon, W. Y. (1992) Oncogene 7, 757-762 [Medline] [Order article via Infotrieve]
  48. Andoniou, C. E., Thien, C., and Langdon, W. Y. (1996) Oncogene 12, 1981-1989 [Medline] [Order article via Infotrieve]
  49. Belkin, A. M., Zhidkova, N. I., Balzac, F., Altruda, F., Tomatis, D., Maier, A., Tarone, G., Koteliansky, V. E., and Burridge, K. (1996) J. Cell Biol. 132, 211-226 [Abstract]
  50. Balzac, F., Belkin, A. M., Koteliansky, V. E., Balabanov, Y. V., Altruda, F., Silengo, L., and Tarone, G. (1993) J. Cell Biol. 121, 171-178 [Abstract]
  51. Balzac, F., Retta, S. F., Albini, A., Melchiorri, A., Koteliansky, V. E., Geuna, M., Silengo, L., and Tarone, G. (1994) J. Cell Biol. 127, 557-565 [Abstract]
  52. Meredith, J., Jr., Takada, Y., Fornaro, M., Languino, L. R., and Schwartz, M. A. (1995) Science 269, 1570-1572 [Medline] [Order article via Infotrieve]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.