Binding of Receptor-recognized Forms of α2-Macroglobulin to the α2-Macroglobulin Signaling Receptor Activates Phosphatidylinositol 3-Kinase*

Ligation of the α2-macroglobulin (α2M) signaling receptor by receptor-recognized forms of α2M (α2M*) initiates mitogenesis secondary to increased intracellular Ca2+. We report here that ligation of the α2M signaling receptor also causes a 1.5–2.5-fold increase in wortmannin-sensitive phosphatidylinositol 3-kinase (PI3K) activity as measured by the quantitation of phosphatidylinositol 3,4,5-trisphosphate (PIP3). PIP3 formation was α2M* concentration-dependent with a maximal response at ∼50 pm ligand concentration. The peak formation of PIP3 occurred at 10 min of incubation. The α2M receptor binding fragment mutant K1370R which binds to the α2M signaling receptor activating the signaling cascade, increased PIP3 formation by 2-fold. The mutant K1374A, which binds very poorly to the α2M signaling receptor, did not cause any increase in PIP3 formation. α2M*-induced DNA synthesis was inhibited by wortmannin. 1,2Bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acetoxymethylester a chelator of intracellular Ca2+, drastically reduced α2M*-induced increases in PIP3 formation. We conclude that PI3K is involved in α2M*-induced mitogenesis in macrophages and intracellular Ca2+ plays a role in PI3K activation.

The ␣-macroglobulins are part of a large super family including human ␣ 2 -macroglobulin (␣ 2 M) 1 (1,2). Proteolytic attack on the bait region or direct nucleophilic attack on the thiol ester bonds of human ␣ 2 M subunits triggers a major conformational change that exposes receptor recognition sites present in each of the four ␣ 2 M subunits (2,3). Two receptors bind ␣ 2 M*, namely, LRP/␣ 2 MR and a recently discovered ␣ 2 M signaling receptor (␣ 2 MSR) (4 -13). LRP/␣ 2 MR is a scavenger receptor that binds a wide variety of ligands. Binding of ␣ 2 M* to LRP/ ␣ 2 MR is followed by uptake and degradation in lysosomes but not activation of a signaling cascade (7,8,12). By contrast, binding of ␣ 2 M* or RBF to ␣ 2 MSR triggers classical signaling cascades and regulates cell proliferation (6 -14).
The agonist-induced entry of Ca 2ϩ from the extracellular medium is of major importance in the cytosolic Ca 2ϩ signals that link activation of various receptors on the cell surface with the initiation and control of cell functions (15)(16)(17). Elevated cytosolic Ca 2ϩ modulates specific cell cycle events and DNA synthesis (18 -23). Binding of ␣ 2 M* to ␣ 2 MSR raises p21 RASGTP levels 2-3-fold in macrophages and pretreatment with wortmannin, a specific inhibitor of PI3K, does not affect ␣ 2 M*induced increases in p21 RASGTP levels (24). Cellular 3-phosphoinositides are generated through the action of a family of PI3Ks (25, 26). PI3K activity was first reported in association with v-SRC and v-RAS oncoproteins, where it catalyzes phosphorylation of inositol at the D-3 position of phosphatidylinositol (PtdIns), PtdIns 4-phosphate, and PtdIns 4,5-bisphosphate (25-27). Several down stream protein substrates for PI3K have been identified, which include certain protein kinase C isoforms (PKC␦, PKC⑀, PKC, PKC) (25-27) and the plekstrin homology domain containing protein kinases cAKT and BTK (28). An increase in the intracellular concentration of PtdIns 3,4-bisphosphate and PtdIns 3,4,5-trisphosphate is observed in several cell types on stimulation with growth factors, cytokines, insulin, f-Met-Leu-Phe, agents that activate RAS, and viral transformation (25-27). Signaling by 3-phosphoinositides regulates diverse functions such as mitogenesis, cell growth, membrane ruffling, chemotaxis, oxidant production, secretory responses, insulin-mediated membrane translocation of the glucose transporter, membrane trafficking of growth factor receptors, cell adhesion, and Na/H ϩ exchange (25-27). Since many of the cellular responses elicited upon ligation of ␣ 2 MSR with receptor-recognized forms of ␣ 2 M are similar to those elicited upon binding of growth factors to their receptors, we studied the activity of PI3K by measuring the formation of PtdIns 3,4,5-trisphosphate (PIP 3 ), in murine macrophages stimulated with ␣ 2 M*. Ligation of ␣ 2 MSR increases the wortmannin-sensitive formation of PIP 3 2-3-fold in a concentration-dependent manner and that the agonist-induced formation of PIP 3 is influenced by [Ca 2ϩ ] i levels.

EXPERIMENTAL PROCEDURES
Materials-Human ␣ 2 M, ␣ 2 M-methylamine (␣ 2 M*), RBF and its mutants K1370A and K1374R were prepared as described (13). The sources of thioglycollate-elicited macrophages and cell culture requirements have been described previously (7)(8)(9)(10). PtdIns 4-phosphate (PIP), PtdIns 4,5-bisphosphate (PIP 2 ) and PtdIns 3,4,5-trisphosphate (PIP 3 ), were from Biomol (Plymouth Meeting, PA). Insulin, wortmannin, thapsigargin, fatty acid-free bovine serum albumin (BSA) and molybdenum blue spray were from Sigma. Fura 2/AM and BAPTA/AM were from Molecular Probes (Eugene, OR). tion in murine peritoneal macrophages was measured essentially according to the method of Okada et al. (29) except that [ 3 H]myoinositol was used to label inositol lipids in place of 32 P i . Briefly thioglycollateelicited macrophages (ϳ8 ϫ 10 6 /well) were collected in Hanks' balanced salt solution (HHBSS) containing 10 mM HEPES, pH 7.4, and were allowed to adhere for 2 h in RPMI 1640 medium containing 2 mM glutamine, 12.5 units of penicillin/ml, and 6 g of streptomycin/ml, and 5% fetal bovine serum at 37°C in a humidified CO 2 (5%) incubator. Nonadherent cells were removed with cold HHBSS, and a volume of RPMI 1640 medium was added containing the additions listed above except that 0.2% fatty acid-free BSA was substituted for the serum. To each well [ 3 H]myoinositol, 30 Ci/ml, was added, and the cells were incubated as above for 20 h. The monolayers were washed four times with the above RPMI 1640 medium, a volume of the medium added to each well, and the cells preincubated for 3 min at 37°C before stimulation with different agonists for 10 min. In experiments where the effect of wortmannin on agonist-induced formation of PIP 3 was studied, it was incubated (30 nM) with samples for 30 min at 37°C prior to addition of agonists. In experiments where the effects of modulation of intracellular Ca 2ϩ by thapsigargin (100 nM) and BAPTA/AM (10 M) were to be studied on PIP 3 formation, the former was added 10 min and the latter 30 min before the addition of the agonist. The reaction was terminated by aspirating the medium, a volume of chilled methanol was added to each well, and the lipids were extracted and separated on oxalate-impregnated silica gel G plates as described by Okada et al. (29). Authentic standards of PIP, PIP 2, and PIP 3 were co-chromatographed with each run. The chromatoplates were air-dried and phospholipid spots detected by lightly spraying with molydenum blue spray (30). The R F values obtained under the experimental conditions for PIP, PIP 2 , and PIP 3 were 0.63, 0.23, and 0.12, respectively. Gel areas corresponding in R F values to PIP 3 were scraped into scintillation vials and the radioactivity determined by liquid scintillation counting. In preliminary experiments, the identity of 3 H-labeled PIP, PIP 2, and PIP 3 on chromatoplates was established by 1) autoradiography of developed chromatoplates on Kodak BioMax film using BioMax TranScreen-LE intensifying screen (Eastman Kodak Co.) at Ϫ70°C for 10 days and comparing the R F values of radioactive spots with authentic standards co-chromatographed and (2) by adding authentic standards of PIP, PIP 2, and PIP 3 (15 g each) to samples prior to chromatography and spraying the developed chromatoplates with molydenum blue spray (30).
Measurement of DNA Synthesis-DNA synthesis was measured according to Charlesworth and Rozengurt (11,23). Briefly, 2-h-adhered macrophages (4 ϫ 10 5 cells/well) were incubated in a volume of RPMI 1640 medium containing glutamine, penicillin, streptomycin, and 0.2% fatty acid-free BSA. To each well [ 3 H]thymidine (2 Ci/ml) was added followed by the addition of different ligands to the respective wells and the incubations continued as above for 20 h. In experiments where the effects of wortmannin (30 nM) were examined on RBF-induced DNA synthesis, it was added 30 min before adding the ligand, and the incubations were performed as above. The incubations were terminated by aspirating the medium, a volume of 5% trichloroacetic acid was added to each well, and plates were left on ice for 30 min. Trichloroacetic acid was aspirated and cells washed once more with trichloroacetic acid followed by washing three times with cold HHBSS. The cells were lysed in 1 N NaOH and radioactivity determined by liquid scintillating. For protein measurement, identically incubated, but untreated, cells were lysed in 0.1 N NaOH and protein estimated according to Bradford (31).

Measurement of Inositol 1,4,5-Triphosphate and [Ca 2ϩ ] i -IP 3 and [Ca 2ϩ
] i elicited upon exposure of murine peritoneal macrophages to ␣ 2 M* were measured as described (6 -10). In experiments where the effects of wortmannin (30 nM) on ␣ 2 M*-induced changes in [Ca 2ϩ ] i and IP 3 were studied, it was added 30 min before the agonist.

RESULTS AND DISCUSSION
␣ 2 MSR Ligation with ␣ 2 M* Increases PIP 3 Levels-The effect of ␣ 2 M* on the synthesis of PIP 3 in macrophages is shown in Fig. 1. The maximum synthesis of PIP 3 occurred at a ligand concentration of ϳ50 pM (Fig. 1A). The kinetics of PIP 3 synthesis is similar to that noted previously for p21 RASGTP formation (24) in macrophages stimulated with ␣ 2 M*. Since wortmannin treatment had no effect on ␣ 2 M*-stimulated p21 RASGTP synthesis (24), PI3K is downstream of RAS, consistent with the report that PI3K is a substrate for activated RAS (32). The synthesis of PIP 3 stimulated with 100 pM of ␣ 2 M* was maximal after a 10-min period of incubation but declined at longer periods of incubations (Fig. 1B). The ␣ 2 M*-induced synthesis of PIP 3 was comparable with the effect of insulin (20 nM) ( Fig. 2A), a potent activator of PI3K (33). That the increase in PIP 3 formation occurs due to the binding of ␣ 2 M* to ␣ 2 MSR was confirmed by using a RBF of ␣ 2 M and its mutants (Fig. 2B). Both RBF and its mutant K1370R, which bind to ␣ 2 MSR and generate signaling events similar to that of ␣ 2 M* (13), caused a 2-fold increase in PIP 3 synthesis (Fig. 2B). By contrast, the binding site mutant K1374A, which binds poorly to ␣ 2 MSR, does not elicit increases in IP 3 formation or increases in [Ca 2ϩ ] i (13) and failed to stimulate PIP 3 synthesis (Fig. 2B). Wortmannin, a potent and specific inhibitor of PI3K activity (34), completely inhibited ␣ 2 M*-, RBF-, and insulin-induced increases in PIP 3 synthesis ( Fig. 1A and 2). We also tested the effect of 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002)on ␣ 2 M*-induced PIP 3 synthesis. LY294002 is a specific inhibitor PI3K, albeit its EC 50 is greater than wortmannin (35,36). This inhibitor almost completely abolished PIP 3 synthesis in macrophages exposed to ␣ 2 M* ( Table I).
Wortmannin Inhibits RBF-induced DNA Synthesis-A num-  PI3 Kinase and ␣ 2 Signaling Receptor protein kinase B (PKB also known as cAKT). The latter are activated consequent to PI3K activation in cells treated with growth factors and mitogens, overexpression of PI3K, and inhibited by wortmannin and by dominant negative subunit mutants of PI3K (25- 27,41). Downstream targets of PKB include p70 ribosomal kinase S6 involved in up-regulation of transcripts for ribosomal proteins and elongation factors (43,44). We have assessed the involvement of PI3K in RBF-induced DNA synthesis in macrophages by using wortmannin (Fig. 3A). Incubation of cells with wortmannin (30 nM/30 min/37°C) prior to stimulation with RBF (100 pM) nearly abolished RBF-induced DNA synthesis (Table II), which shows that the PI3K signaling pathway is involved in DNA synthesis in cells stimulated with receptor-recognized forms of ␣ 2 M. The PI3K inhibitor LY294002 also nearly abolished DNA synthesis induced by RBF or ␣ 2 M* (data not shown).
Chelation of [Ca 2ϩ ] i with BAPTA/AM Inhibits ␣ 2 M*-induced PIP 3 Synthesis-We have previously reported the dependence of protein and DNA synthesis on intracellular Ca 2ϩ levels in macrophages stimulated with ␣ 2 M* (11). We have now examined the role of [Ca 2ϩ ] i on PIP 3 synthesis in macrophages stimulated with ␣ 2 M* in several ways: 1) by modulating [Ca 2ϩ ] i with thapsigargin (100 nM/10 min/37°C), an endoplasmic reticulum Ca 2ϩ -ATPase inhibitor that raises [Ca 2ϩ ] i by releasing Ca 2ϩ from both IP 3 -dependent and IP 3 -independent internal Ca 2ϩ pools and 2) by use of BAPTA/AM (10 mM/30 min/37°C) that chelates [Ca 2ϩ ] i . Thapsigargin alone increased PIP 3 synthesis comparable with that seen with ␣ 2 M* or with thapsigargin plus ␣ 2 M* (Fig. 2C). By contrast, BAPT/AM nearly abolished ␣ 2 M*-induced PIP 3 synthesis (Fig. 2C). We have reported previously that manipulating IP 3 and [Ca 2ϩ ] i profoundly alters agonist-induced increases in protein and DNA synthesis (10,11). In light of the importance of [Ca 2ϩ ] i in ␣ 2 M-induced DNA synthesis, and inhibition of DNA synthesis by wortmannin (Fig. 2C) we evaluated the effect of wortmannin on ␣ 2 M*induced synthesis of IP 3 and changes in [Ca 2ϩ ] i (Fig. 3). Wortmannin by itself showed no effect on IP 3 synthesis in macrophages, and when administered before ␣ 2 M*, it only slightly attenuated IP 3 synthesis (about 10 -15%) compared with ␣ 2 M*treated cells (Fig. 3A). As expected, treatment of cells with wortmannin before ␣ 2 M* only slightly attenuated both the IP 3 -induced increase in [Ca 2ϩ ] i as well as Ca 2ϩ entry from the medium (Fig. 3B).
The tyrosine kinase class of receptors, which include growth factor receptors, upon activation, induce mitogenesis via a series of downstream steps that may show cellular variance and include signaling proteins Grb2, Sos, Ras, Raf, MEK, and MAPK (40). We show here, for the first time, that like insulin and other growth factors, receptor-recognized forms of ␣ 2 M, upon binding to the ␣ 2 MSR, also induce the activation of wortmannin-sensitive PI3K. Thus as suggested earlier (10 -14), in addition to being a classical endocytic receptor, ␣ 2 MSR also appears to have an additional role in tissue repair.