Inhibition of Phosphatidylinositol 3-Kinase by c-Abl in the Genotoxic Stress Response*

Activation of phosphatidylinositol (PI) 3-kinase by growth factors results in phosphorylation of phosphatidylinositol lipids at the D3 position. Although PI 3-ki- nase is essential to cell survival, little is known about mechanisms that negatively regulate this activity. Here we show that the c-Abl tyrosine kinase interacts directly with the p85 subunit of PI 3-kinase. Activation of c-Abl by ionizing radiation exposure is associated with c-Abl-dependent phosphorylation of PI 3-kinase. We also show that phosphorylation of p85 by c-Abl inhibits PI 3-kinase activity in vitro and in irradiated cells. These findings indicate that c-Abl negatively regulates PI 3-kinase in the stress response to DNA damage. The product c- abl gene chemilumi-nescence glu-tathione-Sepharose lized

The product of the c-abl gene is a nonreceptor tyrosine kinase that shares certain structural features, including SH3, SH2, and catalytic domains, with the Src kinase family. In addition, c-Abl contains nuclear localization signals, a bipartite DNAbinding domain, and F-and G-actin binding domains (reviewed in Ref. 1). Cells exposed to ionizing radiation (IR) 1 and certain other DNA-damaging agents respond with activation of c-Abl (1)(2)(3)(4). Signals downstream of c-Abl activation include induction of the stress-activated protein kinase and p38 mitogenactivated protein kinase (1,2,5). The activation of c-Abl by genotoxic stress is also associated with interaction of c-Abl with the p53 tumor suppressor protein in the G 1 arrest response (6,7). Other studies have provided support for involvement of c-Abl in DNA damage-induced apoptosis (8).
One issue concerning the genotoxic stress response is how DNA damage is converted into informational intracellular signals that affect cell behavior. Whereas IR induces DNA doublestrand breaks (9), the DNA-dependent protein kinase (DNA-PK) is activated by double-strand breaks and other DNA lesions (10 -12). Recent studies have demonstrated that DNA-PK phosphorylates and activates c-Abl (13). In a potential feedback mechanism, c-Abl phosphorylates and inhibits DNA-PK activity (13). Other work has shown that c-Abl interacts with the product of the ataxia telangiectasia mutated (ATM) gene and that ATM may also be responsible for activating c-Abl as a consequence of genotoxic stress (14,15). The findings that cells defective in DNA-PK or ATM are hypersensitive to the lethal effects of IR (16,17), whereas c-Abl-deficient cells are resistant to IR-induced killing (8), has suggested that functional interactions between c-Abl and DNA-PK or ATM may contribute to cell fate.
The DNA-PK catalytic subunit and ATM are related to members of the PI 3-kinase family, such as Tor1p, Tor2p, FKBPrapamycin-associated protein, and Schizosaccharomyces pombe Rad 3, involved in cell cycle control and DNA repair (18). PI 3-kinase is activated by growth factor receptors and has been implicated in transducing survival signals (19). Inositol lipids phosphorylated at the D3 position by PI 3-kinase function in part in the activation of the Akt (protein kinase B) serine/threonine protein kinase (20 -23). The PI 3-kinase/Akt pathway inhibits the induction of apoptosis by serum withdrawal (24) and the c-Myc protein (25). Activation of Akt thus promotes survival, whereas inhibition of PI 3-kinase with wortmannin induces apoptosis (24,26). These findings suggest that signaling mechanisms that down-regulate PI 3-kinase would be pro-apoptotic. However, whereas autophosphorylation of PI 3-kinase on serine inhibits PI 3-kinase activity (27,28), little is known about other signals that negatively regulate this kinase.
The present studies demonstrate that c-Abl associates with PI 3-kinase. Phosphorylation of the p85 subunit of PI 3-kinase by c-Abl inhibits PI 3-kinase activity. The results also show that c-Abl phosphorylates and inhibits PI 3-kinase in IRtreated cells.
Phosphorylation of PI 3-Kinase-Recombinant kinase active c-Abl * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. was prepared from baculovirus-infected insect cells (32). GST-p85 was incubated with heat-inactivated (HI) or kinase-active c-Abl and [␥-32 P]ATP in kinase buffer (20 mM HEPES, pH 7.5, 10 mM MgCl 2 and 10 mM MnCl 2 ) for 15 min at 30°C. The reaction products were analyzed by 8% SDS/polyacrylamide gel electrophoresis and autoradiography.
Lipid Kinase Assays-PI 3-kinase activity was measured as described (35) in lipid kinase buffer (10 mM Tris, pH 7.5, 100 mM NaCl, 1 mM EDTA, 100 M sodium vanadate, 1 mM dithiothreitol) containing 2 g/ml phosphatidylinositol (Avanti Polar Lipids, Alabaster, AL) and [␥-32 P]ATP. The reaction was stopped by the addition of 6 N HCl and chloroform/methanol (1:1). The organic layer was separated and spotted on a Silica Gel-60 plate (Sigma). Phosphorylation was assessed by thin layer chromatography and autoradiography.

RESULTS AND DISCUSSION
c-Abl interacts with the PI 3-kinase-related DNA-PK and ATM proteins (13,15). To determine if c-Abl associates with PI 3-kinase, anti-c-Abl immunoprecipitates were analyzed by immunoblotting with an antibody against the p85 subunit of PI 3-kinase. The results demonstrate constitutive binding of c-Abl and PI 3-kinase (Fig. 1A). The activation of c-Abl by IR treatment (1) had little effect on the interaction (Fig. 1A). The finding that the p85 protein detected in the anti-c-Abl immunoprecipitates exhibits a slightly decreased electrophoretic mobility compared with that in total cell lysates supports binding of c-Abl to only certain post-translationally modified forms or specific isoforms of p85. Analysis of anti-p85 immunoprecipitates with anti-c-Abl in the reciprocal experiment confirmed binding of c-Abl and PI 3-kinase in control and irradiated cells (Fig. 1B). Incubation of cell lysates with GST fusion proteins demonstrated binding of p85 to full-length c-Abl and to Abl SH3 but not to the N-terminal SH3 domain of Grb2 (Fig. 1C). The c-Abl SH3 domain binds to proline-rich sequences with the PXXP consensus (33,36). The presence of potential sequences for c-Abl binding in the p85 subunit suggested that there may be a direct interaction. The finding that GST-p85 binds to recombinant c-Abl purified from a baculovirus expression system provided evidence for a direct interaction between these proteins (Fig. 1D).
To assess the functional significance of the interaction between c-Abl and p85, we asked if PI 3-kinase is a substrate for c-Abl. Incubation of recombinant kinase-active c-Abl with GST-p85 resulted in phosphorylation of p85 ( Fig. 2A). c-Abl exhibited no detectable phosphorylation of GST alone (data not shown), and there was no phosphorylation of p85 with heatinactivated c-Abl ( Fig. 2A). To determine whether DNA damage induces c-Abl-dependent phosphorylation of p85, anti-P-Tyr immunoprecipitates were assayed from mouse fibroblasts deficient in c-Abl (Abl Ϫ/Ϫ ) (30) and after stable expression of c-Abl in this background (Abl ϩ ) (1). Irradiation of the Abl ϩ fibroblasts was associated with tyrosine phosphorylation of the p85 subunit (Fig. 2B). By contrast, there was no IR-induced phosphorylation of p85 in the Abl Ϫ/Ϫ cells (Fig. 2B). Reprobing the filters with an antibody against the PI 3-kinase p110 subunit demonstrated no detectable reactivity (data not shown). These results indicate that IR induces tyrosine phosphorylation of the p85 subunit of PI 3-kinase by a c-Abl-dependent mechanism. To confirm these findings, we irradiated MCF-7 cells that were stably transfected with empty vector (MCF-7/ pSR) or the dominant negative, kinase-inactive c-Abl(K-R) (6). Anti-p85 immunoprecipitates were analyzed by immunoblotting with anti-P-Tyr. The finding that tyrosine phosphorylation of p85 is substantially abrogated in IR-treated MCF-7/c-

FIG. 1. c-Abl associates with PI 3-kinase.
A and B, U-937 cells (1) were exposed to 20 Gy of IR and harvested at 1 h. Lysates from control (C) and irradiated (IR) cells were subjected to immunoprecipitation with anti-c-Abl or anti-p85. The immunoprecipitates were separated in 8% SDS-polyacrylamide gels, transferred to nitrocellulose, and analyzed by immunoblotting (IB) with anti-p85 or anti-c-Abl. Lysate not subjected to immunoprecipitation was used as a positive control for the immunoblot analysis. C, cell lysate was incubated with 5 g of GST-Grb2 N.SH3, GST-Abl SH3, or GST-c-Abl (full-length) (32) for 2 h at 4°C. The adsorbates were analyzed by immunoblotting with anti-p85. Lysate was used directly as a positive control. D, recombinant c-Abl was purified from GST-c-Abl prepared in baculovirus by cleavage of the GST with thrombin (32). The recombinant c-Abl was incubated with GST or GST-p85. The adsorbates were analyzed by immunoblotting with anti-c-Abl.

FIG. 2. c-Abl phosphorylates PI 3-kinase in vitro and in irradiated cells.
A, HI and kinase-active recombinant c-Abl were incubated with GST-p85 and [␥-32 P]ATP. Phosphorylated proteins were separated in 8% SDS-polyacrylamide gels and analyzed by autoradiography. B, Abl ϩ and Abl Ϫ/Ϫ cells were exposed to 20 Gy of IR and harvested at the indicated times. Cell lysates were subjected to immunoprecipitation with anti-P-Tyr. The immunoprecipitates were analyzed by immunoblotting (IB) with anti-p85. Lysates from Abl ϩ and Abl Ϫ/Ϫ cells not subjected to immunoprecipitation were used as controls for the immunoblot analysis. C, MCF-7/p/SR and MCF-7/c-Abl(K-R) cells were exposed to 20 Gy of IR and harvested at 1 h. Lysates were subjected to immunoprecipitation with anti-p85, and the precipitates were analyzed by immunoblotting with anti-P-Tyr.
The p85 subunit of PI 3-kinase is required for the lipid kinase activity of p110 (37). To determine if c-Abl-mediated phosphorylation of p85 affects PI 3-kinase activity, anti-p85 immunoprecipitates were incubated with a stoichiometric excess of purified recombinant c-Abl. Heat-inactivated c-Abl had no effect on phosphorylation of PI, whereas kinase-active c-Abl inhibited PI 3-kinase activity (Fig. 3A). To assess the effects of IR treatment, anti-c-Abl immunoprecipitates from control and irradiated cells were analyzed for PI 3-kinase activity. The precipitates from control cells exhibited phosphorylation of PI, whereas this activity was inhibited in IR-treated cells (Fig. 3B). Analysis of anti-p85 immunoprecipitates from IR-treated Abl ϩ cells also demonstrated down-regulation of PI 3-kinase activity (Fig. 3C). However, IR treatment of Abl Ϫ/Ϫ cells had little if any effect on PI phosphorylation compared with that in control Abl Ϫ/Ϫ cells (Fig. 3C). These results suggest that c-Abl-mediated phosphorylation of p85 inhibits PI 3-kinase activity.
The available evidence indicates that control of PI 3-kinase is important in determining cell fate. However, other than downregulation of PI 3-kinase by autophosphorylation on serine (27,28), there are no known negative regulatory effectors of this kinase. The present findings demonstrate that the SH3 domain of c-Abl binds constitutively to PI 3-kinase. The results also support the hypothesis that activation of c-Abl by DNA damage (1,3,6,13,38) down-regulates PI 3-kinase activity by phosphorylation of p85. Importantly, transforming variants of c-Abl, such as Bcr-Abl and v-Abl, have been found to activate PI 3-kinase (39,40). Thus, the present finding that c-Abl inhibits PI 3-kinase activity is in contrast to the effects of oncogenic c-Abl variants. The molecular basis for this difference could be important to our understanding of transformation by genetic alterations of the c-abl gene.
The demonstration that cells deficient in c-Abl are resistant to IR-induced apoptosis has supported a pro-apoptotic function for c-Abl in the cell death response to DNA damage (8). By contrast, Bcr-Abl expression inhibits the apoptotic response to DNA-damaging agents (41). The Akt/protein kinase B kinase is a downstream effector of PI 3-kinase that participates in suppression of apoptosis (20,21,24,25,42,43). Taken together with the functional interaction found between c-Abl and PI 3-kinase in the present studies, these findings suggest that c-Abl could contribute to the regulation of cell fate through down-regulation of PI 3-kinase. The present results also provide the first evidence for a tyrosine kinase that negatively regulates PI 3-kinase.
FIG. 3. c-Abl inhibits PI 3-kinase activity. A, U-937 cell lysate was subjected to immunoprecipitation with anti-p85 and divided into two aliquots. The immunoprecipitates were incubated with HI and kinaseactive recombinant c-Abl in the presence of ATP. The immune complexes were washed and assayed for PI 3-kinase activity. The origin and position of PI 3-phosphate are indicated. The anti-p85 aliquots were also analyzed by immunoblotting (IB) with anti-p85 to demonstrate equal amounts of protein. B, U-937 cells were exposed to 20 Gy of IR and harvested at 1 h. Lysates from control (C) and irradiated (IR) cells were subjected to immunoprecipitation with anti-c-Abl. The immunoprecipitates were analyzed for PI 3-kinase activity and by immunoblotting with anti-p85. C, Abl ϩ and Abl Ϫ/Ϫ fibroblasts were exposed to 20 Gy of IR and harvested at 1 h. Lysates from control and irradiated cells were immunoprecipitated with anti-p85. The immunoprecipitates were analyzed for phosphorylation of PI and for protein by immunoblotting with anti-p85.

Down-regulation of PI 3-Kinase by c-Abl
23487