G-Protein-coupled Receptors and Fcγ-receptors Mediate Activation of Akt/Protein Kinase B in Human Phagocytes*

Activation of the serine/threonine kinase Akt, also called protein kinase B (PKB), was investigated in human neutrophils. Stimulation of the cells with the chemoattractant fMet-Leu-Phe or the chemokines IL-8 and GROα leads to the rapid and transient activation of PKB. Maximum PKB activation correlates with the well documented kinetics of respiratory burst and exocytosis. Wortmannin, a selective inhibitor of phosphoinositide 3-kinases (PI 3-kinases) in neutrophils, abrogates PKB activation. Similarly homo and heterotypic cross-linking of FcγIIA and FcγIIIB causes a transient activation of PKB that is sensitive to wortmannin treatment. Kinase activity measurements in immunoprecipitates from lysates of the myelocytic GM-1 cells or GM-1/CXCR1 cells, which are transfected with the IL-8 receptor 1, confirmed the transient activation of PKB observed in neutrophils. Stimulation of human monocytes with the CC chemokine RANTES (regulated on activation normal T cell expressed and secreted) also results in the activation of PKB. Preincubation of monocytes and neutrophils with Bordetella pertussis toxin inhibits fMet-Leu-Phe and RANTES-stimulated PKB activation, demonstrating that coupling of the receptors to heterotrimeric Gi-protein is required. The data show, that activation of PKB by Gi-protein-coupled receptors is mediated by PI 3-kinase and suggest that PKB is a constituent of neutrophil activating pathways.

gous defect in the CCR5 gene, that encodes for the receptor of the CC chemokines RANTES, MIP1␣, and MIP1␤, was shown to provide resistance to HIV infection of multiply exposed individuals (33). It was further reported that RANTES and RANTES antagonists prevent infection of lymphocytes with macrophage tropic HIV isolates (33), indicating that binding of the HIV envelope protein gp120 the co-receptor is essential for virus uptake. However, the contribution of signal transduction stimulated by gp120 binding to CD4 and the G i -protein coupled co-receptors for HIV infection is still controversial (34,35).
We show here for the first time that in phagocytes activation of PKB can be stimulated by a variety of G i -protein-coupled receptor-agonists and by cross-linking of Fc␥-receptors. The transient PKB activation observed in neutrophils correlates with other typical responses of the cells, such as the respiratory burst and exocytosis. The findings implicate that PKB is part of the signal cascade involved in neutrophil activation.

Materials
RPMI 1640 media, Hank's balanced salt solution, antibiotics, and additional cell culture supplements were obtained from Life Technologies, Inc. Protein A-Sepharose CL-4B was purchased from Pharmacia Biotech Inc., Immobilon-P (PVDF) was from Millipore Corp., histone 2B was from Fluka, [␥-32 P]ATP was from Amersham Life Science, Inc. and B. pertussis toxin was from List Biochemicals. Alkaline phosphataseconjugated goat anti-rabbit IgG was obtained from Bio-Rad, and the goat anti-mouse (G␣M) IgG F(abЈ) 2 fragments were from Dianova. All other reagents were of molecular biology grade. The anti-RAC 469 -480 antibodies were prepared as described previously (23). The monoclonal antibodies against Fc␥ receptors IIA (mAb IV.3) and IIIB (mAb SD2) were kindly provided by Dr. A. J. Verhoeven. Chemokines IL-8, GRO␣, and RANTES were chemically synthesized by Dr. I. Clark-Lewis (Vancouver, Canada). The CXCR1 (IL-8R1) gene was kindly supplied by Dr. B. Moser, and 17-hydroxywortmannin was a gift from Dr. T. Payne.
Neutrophil and Monocyte Cell Preparation-Neutrophils were isolated from freshly drawn blood of healthy volunteers (37) or from buffy coats of citrated donor blood stored up to 20 h at 4 -10°C provided from the Swiss Red Cross (Bern, Switzerland) (38). Monocytes were prepared from buffy coats as described (39).
B. pertussis Toxin Treatment of Neutrophils and Monocytes-Pretreatment of neutrophils and monocytes with B. pertussis toxin was as described previously (40). Briefly, neutrophils or monocytes (1.35 ϫ 10 7 /ml) were incubated for 90 min at 37°C in the absence or presence of 1 g/ml B. pertussis toxin in a balanced salt solution (110 mM NaCl, 10 mM KCl, 1 mM MgCl 2 , 10 mM glucose, 30 mM HEPES, and 0.1% w/v bovine serum albumin).

Cell Stimulation
Chemoattractant Stimulation-The cells were washed twice with phosphate-buffered saline, and resuspended in Hank's balanced salt solution containing 20 mM HEPES, pH 7.4 (6 ϫ 10 6 cells/ml for neutrophils or monocytes). Aliquots (500 l) were incubated for 10 -15 min at 37°C in the absence or presence of 100 nM wortmannin prior to stimulation with the indicated stimuli. The reactions were stopped by the addition of trichloroacetic acid to a final concentration of 10%. Precipitated protein was recovered by centrifugation, washed twice in cold acetone, air dried, and resuspended in sample buffer for SDS-PAGE (37).
Fc␥-receptor Cross-linking-Hetero or homotypic cross-linking (41) was performed as follows. Neutrophils or serum-starved GM-1 cells (10 7 /ml) were incubated for 25 min on ice with 10 g/ml monoclonal antibodies for the Fc␥-receptors IIA (mAb IV.3) and/or IIIB (mAb SD2). Excess antibodies were removed by centrifugation, and the cells were resuspended in Hank's balanced salt solution containing 20 mM HEPES, pH 7.4 (1.2 ϫ 10 7 cells/ml for immunoprecipitations or 6 ϫ 10 6 cells/ml for total protein precipitations). Aliquots (500 or 250 l) were incubated for 10 -15 min at 37°C in the absence or presence of 100 nM wortmannin prior to stimulation with cross-linking G␣M IgG F(abЈ) 2 fragments. The reactions were terminated either with trichloroacetic acid, as above, or ice-cold lysis buffer (described below).

Immunoprecipitation and in Vitro Kinase Assay
GM-1/CXCR1 and IFN␥ differentiated GM-1 cells were serumstarved overnight or for 3 h at 37°C in RPMI 1640. The cells were washed twice with phosphate-buffered saline and resuspended in Hank's balanced salt solution containing 20 mM HEPES, pH 7.4 (1.2 ϫ 10 7 cells/ml). Aliquots (500 l) were incubated for 10 -15 min at 37°C in the absence or presence of 100 nM wortmannin prior to stimulation with the indicated stimuli or as described above for Fc␥-receptor crosslinking. The reactions were stopped by the addition of concentrated lysis buffer giving final concentrations of 50 mM Tris-HCl, pH 7.5, 1% Nonidet P-40, 120 mM NaCl, 1 mM EDTA, 50 mM NaF, 40 mM ␤glycerophosphate, 0.1 mM sodium vanadate, 0.5 mM phenylmethylsulfonyl fluoride, 1 mM benzamidine, and 2 M microcystin-LR and were subsequently frozen in liquid nitrogen. Samples were unthawn on ice, centrifuged for 10 min at 12,000 ϫ g at 4°C and precleared with protein A-Sepharose. Endogenous PKB was immunoprecipitated (5 ϫ 10 6 cells) for 2 h at 4°C with an affinity-purified rabbit antibody specific for the conserved carboxyl terminus (anti-RAC 469 -480 (23)) coupled to protein A-Sepharose. Immunoprecipitates were washed twice with lysis buffer and once with kinase buffer (50 mM Tris-HCl, pH 7.5, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, and 1 mM benzamidine). In vitro kinase assays were performed as described previously (20), with the exception that histone 2B was used as a substrate. The incorporation of 32 P into histone 2B was determined from 13.5% SDS-PAGE using a PhosphorImager (Molecular Dynamics). The PKB content of the immunoprecipitates were measured on Western blots (see below).

Western Blot Analysis of PKB
Immunoprecipitated or total protein samples were resolved by 8.0% SDS-PAGE, transferred to PVDF membranes, and incubated with the anti-RAC 469 -480 antibody. Detection was performed using alkaline phosphatase-conjugated anti-rabbit antibody.

Activation of PKB by Chemoattractants-
The determination of PKB activity by substrate phosphorylation from chemoattractant-stimulated neutrophils gave inconsistent results. The high proteolytic and phosphatase activities found in detergent lysates of neutrophils cause the rapid degradation and dephosphorylation of PKB, even in the presence of high concentrations of protease and phosphatase inhibitors. However, activation of PKB was demonstrated to coincide with the phosphorylation on Ser 473 and Thr 308 , resulting in an altered electrophoretic mobility (42,43) which is detected on Western blots using an antibody directed against the carboxyl terminus of the kinase (23). Fig. 1 shows the transient activation of PKB in human neutrophils stimulated by fMet-Leu-Phe, IL-8, and GRO␣. Phosphorylated PKB with reduced elctrophoretic mobility is detectable as early as 10 s after stimulation of the neutrophils and shows a maximum of activation at ϳ40 -80 s. Stimulation with fMet-Leu-Phe leads to a marked phosphorylation of PKB that is still apparent after 5 min. IL-8, which binds with high affinity to both IL-8 receptors CXCR1 and CXCR2, elicits a more transient activation of PKB. Stimulation with GRO␣, which binds only to CXCR2 with high affinity (2), causes an even less pronounced response. Pretreatment of neutrophils with wortmannin abolishes the phosphorylation of PKB in all cases, confirming previous observations that PKB is a downstream effector of PI 3-kinase (20,28).
To measure PKB activity we used promyeolocytic GM-1 cells that can be differentiated with IFN␥ to respond to stimulation with fMet-Leu-Phe (36). Endogenous PKB was immunoprecipitated with affinity purified antibodies from detergent lysates of GM-1 cells and subjected to Western blot analysis. Like in neutrophils, stimulation with fMet-Leu-Phe results in the wortmannin-sensitive transient phosphorylation of PKB (Fig.  2, left). The electrophoretic mobility shift of endogenous PKB in serum-starved GM-1 cells can be detected as early as 10 s, shows a maximum at ϳ40 -80 s, and is still apparent 5 min after stimulation. We assessed protein kinase activity from the same anti-PKB immunoprecipitates from lysates of IFN␥-differentiated cells using histone 2B as substrate. Exposure of serum-starved GM-1 cells to fMet-Leu-Phe stimulates ϳ3-fold the activity of endogenous PKB. The increase in PKB activity shows a maximum at 40 -80 s and is still measurable after 5 min. Incubation of the anti-PKB immunoprecipitates with wortmannin did not inhibit the activity of the kinase, confirming that PKB is not a target. However, pretreatment of the cells with wortmannin blocks agonist-dependent PKB activity, indicating that PI 3-kinase is an upstream regulator. In agreement with previous results (42), incubation of the immunoprecipitates with protein phosphatase 2A (PP2A) abolished PKB activity and the electrophoretic mobility shift (data not shown).
Differentiation induced with IFN␥ may alter signaling pathways which could regulate PKB activation. To exclude this possibility, we introduced the IL-8 receptor CXCR1 into GM-1 cells and cloned stable transfectants. GM-1 cells expressing CXCR1 (GM-1/CXCR1) showed a typical rise in cytosolic free calcium upon stimulation with IL-8 (data not shown). The cells were then used for testing IL-8-dependent activation of PKB. Lysates of control and IL-8-stimulated cells were treated with anti-PKB antibodies, and the immunoprecipitates were resolved on SDS-PAGE. Like in neutrophils, stimulation of serum-starved GM-1/CXCR1 cells results in the transient phosphorylation of PKB (Fig. 2, right). The effect is fully prevented by pretreatment of the cells with 100 nM wortmannin. Measurements of kinase activity associated with the immunoprecipitates from GM-1/CXCR1 cells confirm the transient IL-8stimulated activation of PKB. The kinase activity shows a maximum at 40 s after stimulation. As in the case of neutrophils, stimulation with IL-8 results in a less pronounced and more transient activation of PKB than stimulation with fMet-Leu-Phe. Measurements of PKB activity stimulated by different concentrations of IL-8 displays a bell-shaped dose-response curve (Fig. 3). Maximum stimulation, about 3-fold, is obtained with 100 nM IL-8. At higher IL-8 concentrations, a slightly weaker activation of PKB is observed (Fig. 3). This effect could be mediated by an increase of phosphatase activities, which are more potently stimulated at high agonist concentrations. However, stimulation of kinase activity can be detected in anti-PKB immunoprecipitates from cells exposed to concentrations as low as 1 nM IL-8, suggesting a close coupling of PKB activation to the chemokine receptor.
Activation of PKB by Fc␥-receptor Cross-linking-Cross-linking of Fc␥-receptors has been reported to induce the activation of PI 3-kinase (44). We have therefore tested if homo or heterotypic cross-linking of Fc␥-receptors IIA and/or IIIB (41) leads to activation of endogenous PKB. Fig. 4 shows the effect of goat anti-mouse F(abЈ) 2 -stimulated cross-linking of Fc␥-receptors in neutrophils that were preincubated with receptor-specific monoclonal antibodies, mAb IV.3 (anti-Fc␥RIIA) and/or mAb SD2 (anti-Fc␥RIIIB). The phosphorylation-induced electrophoretic mobility shift of PKB was absent if the cells were pretreated with 100 nM wortmannin or if the cells were incubated only with receptor-specific antibodies or cross-linking F(abЈ) 2 . In contrast, following cross-linking, Fc␥RII induces the transient activation of PKB which shows a maximum at ϳ90 -120 s and was nearly absent after 5 min. Similarly, heterotypic

FIG. 2. Time course of PKB activation in differentiated GM-1 and GM-1/CXCR1 cells. Top panel, PKB activation in IFN␥-differentiated GM-1 (left) or GM-1/CXCR1 (right) cells was detected as retarded electrophoretic mobility on
Western blots. Cells were pretreated with buffer or with 100 nM wortmannin (WT) and then stimulated with either 100 nM fMet-Leu-Phe (left) or 100 nM IL-8 (right) for the indicated times. Following stimulation, lysates were prepared, PKB was immunoprecipitated, and the kinase was resolved on SDS-PAGE. Bottom panel, PKB activity was measured in immunoprecipitates from lysates of differentiated GM-1 cells stimulated with 100 nM fMet-Leu-Phe (left) or GM-1/CXCR1 stimulated with 100 nM IL-8 (right). Open triangles show activity measurements from cells treated with 100 nM wortmannin. Squares indicate PKB activity in the absence of primary antibody. Phosphorylated histone 2B was resolved on SDS-PAGE, and the 32 P content was determined using a PhosphorImager. Results are representative of at least two other experiments.

cross-linking of both receptors causes the transient phosphorylation of PKB.
Because of the rapid degradation of PKB in neutrophil lysates, we used GM-1 cells to assess Fc␥-receptor-stimulated kinase activity in anti-PKB immunoprecipitates. Cross-linking of Fc␥RII results in 2.5-3-fold stimulation of PKB activity (Fig.  5). Pretreatment of the cells with wortmannin inhibits Fc␥RIIstimulated PKB activity. Incubation of GM-1 cells with mAb IV.3 or cross-linking F(abЈ) 2 alone does not stimulate PKB activity. Like U937 cells, the GM-1 express very low levels of Fc␥RIII (36,41), which could account for why we were unable to measure consistently Fc␥RIII-mediated stimulation of PKB activity in lysates of GM-1 cells.
Effect of B. pertussis Toxin-Chemoattractants, such as fMet-Leu-Phe or IL-8, bind to heptahelical receptors that are coupled to the G i class of heterotrimeric G-proteins (40,45). Uncoupling of the G i -proteins from the receptors is a result of ADP-ribosylation by B. pertussis toxin. We have tested the effect of pertussis toxin treatment on the activation of PKB in phagocytes. Fig. 6 shows that preincubations of neutrophils with the toxin blocks fMet-Leu-Phe-stimulated activation of PKB. Western blot analysis of total cellular proteins from human peripheral blood monocytes reveals the expression of PKB (Fig. 6, bottom panel). Unlike in neutrophils, several immunoreactive PKB forms are detected in resting cells. However, a similar pattern of PKB expression has been described in fibroblasts (42). The predominant form of PKB exhibits the same electrophoretic mobility as the neutrophil enzyme. Stimulation of human monocytes with the CC chemokine RANTES induced the appearance of a slower migrating form, which co-migrates with the phosphorylated PKB observed in neutrophils and GM-1 cells. Pretreatment of the monocytes with pertussis toxin inhibits the RANTES-stimulated phosphorylation of PKB, suggesting that the activation of PKB by the CC chemokine is mediated by a G-protein-coupled receptor. In contrast, activation of PKB by cross-linking of Fc␥-receptors is not affected by preincubation of neutrophils with pertussis toxin. This observation is in line with the notion that Fc␥-receptors are not coupled to heterotrimeric G-proteins. DISCUSSION We have investigated the activation of PKB in phagocytes upon stimulation with chemoattractants that bind to G-protein-coupled receptors and following Fc␥-receptor cross-linking. In response to stimulation with chemoattractants, the kinetics of PKB activation complies with other well characterized neutrophil responses, such as respiratory burst and granule exocytosis (1). The findings suggest that activation of PKB is involved in the regulation of neutrophil responses. The more pronounced activation of PKB, upon stimulation with fMet-Leu-Phe, is in line with the previously reported stronger potency of the formyl-peptide to induce the respiratory burst and exocytosis of neutrophils (40,46). Chemokines, which induce a weaker superoxide production than fMet-Leu-Phe, stimulate equally potent the chemotaxis of neutrophils, a response, how- Cross-linking of Fc␥-receptors (IIA) stimulates PKB activity in GM-1 cells. PKB activity was determined in immunoprecipitates prepared from lysates of cells that were preincubated with primary anti-Fc␥RIIA mAb followed by cross-linking for 60 s with G␣M IgG F(ab)Ј 2 fragments (IgG). Preincubation with 100 nM wortmannin (WT; black bars), as compared with nontreated cells (gray bars), inhibits PKB activity. Maximum activity (100%), from three separate experiments, was determined at 60 s after addition of cross-linking IgG. Controls: (white bars, left to right) (i) mock immunoprecipitates lacking PKB antibody, (ii) cells treated with secondary Ab alone (primary Fc␥RIIA mAbs were omitted), and (iii) non-treated cells in which both the primary anti-Fc␥RIIA and G␣M IgG F(ab)Ј 2 fragments (IgG) were omitted. Phosphorylated histone 2B was resolved on SDS-PAGE, and the 32 P content was determined using a PhosphorImager. ever, that is independent of PI 3-kinase activity and thus may not involve PKB (14).
Our experiments reveal that PI 3-kinase activity is necessary for chemoattractant and Fc-receptor-mediated activation of PKB. Recent reports, however, suggest that wortmannin-insensitive pathway(s) may exist that stimulate PI 3-kinaseindependent activation of PKB. The reports include activation by ␤ 3 -adrenoreceptors (32), which couple to G s -and G i -type heterotrimeric G-proteins (47), by elevation of intracellular cAMP levels (48) or by a stress-induced pathway (49). It is conceivable that the alternative pathways are not linked to the phagocyte responses, which are activated upon stimulation with agonists of G i -linked receptors or during Fc-mediated phagocytosis.
Until now only glycogen synthase kinase 3 (GSK3) has been characterized as substrate, which is phosphorylated by PKB upon stimulation of cells with insulin (27). The present data suggest that PKB is involved in rapid signal transduction pathways that lead to transient secretory responses. Stimulation of platelets with thrombin-receptor activating peptide resulted in comparable kinetics of PKB activation as we describe here for neutrophils (31). However, more experimental work is required to define effector molecules of secretory pathways that are phosphorylated by PKB in response to stimulation with agonists of G-protein-coupled receptors.
Growth factors, such as EGF and PDGF, were the first demonstrated to activate PKB via PI 3-kinase (24 -26). Wortmannin, a specific inhibitor of PI 3-kinase (6), growth factor receptor mutants, and dominant negative PI 3-kinase forms, prevent stimulus-dependent PKB activation. Expression of constitutively active PI 3-kinases in COS-7 or GM-1 cells on the other side cause the stimulus-independent activation of PKB (20,28). Thus, PKB is clearly a downstream effector of PI 3-kinase. Recent data suggest that PI(3,4)P 2 binds with high affinity to the pleckstrin homology domain of PKB and thereby contributes to the activation (31, 50, 51). PI(3,4)P 2 is rapidly formed upon dephosphorylation of PI(3,4,5)P 3 , the main product of PI 3-kinase (52). The previously reported kinetics of the formation D3 polyphosphoinositides in platelets and neutrophils stimulated with G-protein receptor-coupled agonists (52)(53)(54) is in excellent agreement with the present data on the activation of PKB.
The two receptors CXCR1 and CXCR2 bind IL-8 with high affinity, whereas GRO␣ is only recognized by CXCR2 with high affinity (55). Both receptors mediate chemotaxis and exocytosis of neutrophils, but the respiratory burst is activated solely by CXCR1 (56,57). The signaling pathways of CXCR1 and CXCR2 lead to the same degree of calcium mobilization from intracellular stores via phospholipase C (58,59) and activation of MAP kinase (60). However, only CXCR1 stimulates activation of phospholipase D (56,61). We show here that both chemokine receptors stimulate the activation of PKB via PI 3-kinase.
A role of PI 3-kinase in promoting cell survival under conditions that would mediate apoptosis has been implicated (62). Recently this effect was shown to be mediated by PKB (29,30,50). The observation that chemokines activate PKB may suggest that chemokine receptors could promote the survival of immune cells. However, resistance to apoptosis would also result in improved conditions for the replication of intracellular pathogens. Presently it is not known if binding of HIV to the chemokine receptors expressed on CD4 ϩ cells results in the activation of PKB.
Treatment of L6 myotubules with insulin results in a marked (ϳ12-fold) activation of PKB (43). In CHO cells, which express low levels of endogenous insulin receptors, insulin mediates only a moderate 3-fold activation of PKB activity (26). Thus the 2.5-4-fold stimulation of endogenous PKB in GM-1 and GM1/CXCR1 cells may be limited by the levels of endogenous proteins involved in activation. G-protein-coupled receptor-induced activation of PKB may involve a distinct isoform of PI 3-kinase that is regulated by the ␤␥-subunits of heterotrimeric G-proteins (16,17,63). Activation of PKB by Fc␥-receptor cross-linking is mediated by signal transduction mechanisms, which resemble growth factor-receptor signaling, and involves tyrosine phsophorylations, coupling to adaptor proteins and the activation of the classical PI 3-kinase p85/p110 (11,64,65).
Signal transduction initiated by the glycosyl phosphatidylinositol-anchored Fc␥RIIIB of neutrophils (7) leads to calcium mobilization (66,67), translocation of Src-related tyrosine kinases (68), actin polymerization, and enhanced phagocytosis (69), but does not activate MAP-kinase as shown for the transmembrane splice variant Fc␥RIIIA that is expressed on mononuclear cells (70). The coupling of the GPI-anchored Fc␥RIIIB to cytoplasmic signaling molecules is not established (71). We report here that cross-linking of the Fc␥RIIIB triggers the rapid activation of PKB by a PI 3-kinase-dependent pathway. The rapid activation of PKB by Fc␥-receptor cross-linking suggests that PKB may be involved in early signaling events associated with antibody-mediated phagocytosis.
The precise role of PKB in neutrophil signal transduction remains to be established. Stimulation via Fc␥-receptors and the more pronounced activation of PKB upon stimulation with fMet-Leu-Phe as compared with the chemokines IL-8 and GRO␣, may indicate that PKB is involved in secretory responses, such as the respiratory burst and exocytosis.