Dinucleotide Phosphate Oxidase Activation of Nicotinamide Adenine ...

The Journal of Immunology

A Critical Role of Protein Kinase C␦ Activation Loop Phosphorylation in Formyl-Methionyl-Leucyl-Phenylalanine-Induced Phosphorylation of p47phox and Rapid Activation of Nicotinamide Adenine Dinucleotide Phosphate Oxidase1 Ni Cheng,* Rong He,* Jun Tian,* Mary C. Dinauer,† and Richard D. Ye2* Generation of superoxide by professional phagocytes is an important mechanism of host defense against bacterial infection. Several protein kinase C (PKC) isoforms have been found to phosphorylate p47phox, resulting in its membrane translocation and activation of the NADPH oxidase. However, the mechanism by which specific PKC isoforms regulate NADPH oxidase activation remains to be elucidated. In this study, we report that PKC␦ phosphorylation in its activation loop is rapidly induced by fMLF and is essential for its ability to catalyze p47phox phosphorylation. Using transfected COS-7 cells expressing gp91phox, p22phox, p67phox, and p47phox (COS-phox cells), we found that a functionally active PKC␦ is required for p47phox phosphorylation and reconstitution of NADPH oxidase. PKC␤II cannot replace PKC␦ for this function. Characterization of PKC␦/PKC␤II chimeras has led to the identification of the catalytic domain of PKC␦ as a target of regulation by fMLF, which induces a biphasic (30 and 180 s) phosphorylation of Thr505 in the activation loop of mouse PKC␦. Mutation of Thr505 to alanine abolishes the ability of PKC␦ to catalyze p47phox phosphorylation in vitro and to reconstitute NADPH oxidase in the transfected COS-phox cells. A correlation between fMLF-induced activation loop phosphorylation and superoxide production is also established in the differentiated PLB985 human myelomonoblastic cells. We conclude that agonist-induced PKC␦ phosphorylation is a novel mechanism for NADPH oxidase activation. The ability to induce PKC␦ phosphorylation may distinguish a full agonist from a partial agonist for superoxide production. The Journal of Immunology, 2007, 179: 7720 –7728.

A

n important aspect of innate immunity is the elimination of pathogenic microbes by activated phagocytes. NADPH oxidase activation is primarily responsible for the production of superoxide anion O2. and other reactive oxygen species within phagosomes (1, 2). When this function becomes defective as seen in patients with chronic granulomatous disease, the ability of phagocytes to eliminate invading microbes is compromised and persistent infections occur (3, 4). Spontaneous production of O2. , which can cause vascular injury, is prevented in resting cells due to an effective segregation of the cytosolic factors (p67phox, p47phox, p40phox, and the Rac small GTPase) from membrane-associated flavocytochrome b558. Upon activation, the Cterminal autoinhibitory region of p47phox is phosphorylated, resulting in structural changes that expose its N-terminal binding sites for phosphatidylinositol 3,4-bisphosphate (PI(3,4)P2)3 and phosphatidic acid (5–7). Interaction of p47phox with PI(3,4)P2 and phos-

*Department of Pharmacology, College of Medicine, University of Illinois, Chicago, IL 60612; and †Herman B. Wells Center for Pediatric Research, Riley Hospital for Children, Indiana University School of Medicine, Indianapolis, IN 46020 Received for publication June 19, 2007. Accepted for publication September 28, 2007. 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. 1 This work was supported by Grants AI033503 and HL077806 (to R.D.Y.) and HL045635 and HL069974 (to M.C.D.) from the National Institutes of Health. R.H. is a recipient of a Scientist Development Grant from the American Heart Association, Greater Midwest Affiliate. 2 Address correspondence and reprint requests to Dr. Richard Ye, Department of Pharmacology, College of Medicine, MC868, University of Illinois, 835 South Wolcott Avenue, Chicago, IL 60612. E-mail address: [email protected]

phatidic acid (8) is critical for its membrane translocation and docking to p22phox, a subunit of flavocytochrome b558. Membrane translocation of p47phox and other cytosolic factors is necessary for the assembly of a functional NADPH oxidase. Published studies have shown that several PKC isoforms, mainly PKC␣, ␤II, ␦, and ␨, are present in neutrophils and can phosphorylate p47phox (9 –13). The observation that the phorbol ester PMA could stimulate p47phox phosphorylation reinforces the notion that PKC promotes the assembly of the NADPH oxidase complex. However, it is still unclear whether the PKC isoforms have redundant functions or play specific roles in NADPH oxidase activation. In neutrophils, multiple activation mechanisms are present and can be triggered by agonists such as fMLF, PMA, and phagocytosis of particles. Thus, studies of the individual PKC isoforms may be facilitated by biochemical dissection of the pathways including the use of reconstitution assays. Using cytosoldepleted neutrophil “cores”, Yaffe and colleagues (14) reconstituted NADPH oxidase activity and found PKC␦ to be one of the critical components for O2. production. Several investigators reported that the PKC␦-selective inhibitor rottlerin could significantly reduce O2. production in eosinophils (15), monocytes (16), and neutrophils (17) that are stimulated with leukotriene B4, opsonized zymosan, and fMLF, respectively. In addition, inhibition of PKC␦ expression through the use of antisense oligonucleotides also reduced O2. production (16). We recently reported that in transgenic COS-phox cells (COS-7 stably expressing gp91phox, p22phox, p47phox, and p67phox) (18), reconstitution of fMLF-induced O2. production is

activating factor; phox, phagocyte oxidase; FPR, formyl peptide receptor; HA, hemagglutinin.

3

Abbreviations used in this paper: PI(3,4)P2, phosphatidylinositol 3,4-bisphosphate; PKC, protein kinase C; PDK-1, 3-phosphoinositol-dependent kinase 1; PAF, plateletwww.jimmunol.org

Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00

The Journal of Immunology PKC␦-dependent (17). These findings suggest that PKC␦ plays a unique role in NADPH oxidase activation. PKC␦, along with PKC␧, ␪, and ␩, belongs to the subfamily of novel PKC (19, 20). The regulatory domains of these novel PKCs contain structural determinants for diacylglycerol binding, which is responsible for allosteric modulation of the catalytic activity. Unlike classic PKC, novel PKC is not subject to regulation by calcium. The catalytic domains of all PKC isoforms contain multiple serine and threonine residues that are phosphorylated, a process important for a newly synthesized PKC to acquire catalytic activity (reviewed in Ref. 21). These phosphorylation sites are located in the activation loop, the turn motif, and the hydrophobic motif within the catalytic domain (21). Phosphorylation of the threonine residue in the activation loop, which may be catalyzed by 3-phosphoinositol-dependent kinase 1 (PDK-1), is believed to be important for PKC autophosphorylation and its catalytic activity (22–25). However, several published reports suggest that activation loop phosphorylation may not be required for the catalytic activity of PKC␦, as recombinant PKC␦ expressed in Escherichia coli and PKC␦ with a mutation at Thr505 are not phosphorylated yet retain the ability to catalyze histone and pseudosubstrate-related peptide phosphorylation in in vitro kinase assays (26 –28). These findings cast doubt on the importance of PKC␦ activation loop phosphorylation in its biological functions. To further delineate the mechanism by which PKC␦ regulates NADPH oxidase activation, we designed experiments to use fulllength p47phox both as a substrate for in vitro kinase assay and as an essential component for reconstitution of fMLF-induced O2. production in COS-phox and COS91/22 (COS-7 cells expressing gp91phox and p22phox). The COS-phox cell line has been successfully used in the reconstitution of NADPH oxidase activation induced by arachidonic acid and PMA (18), fMLF (17), and IgGcoated particles (29). As we reported previously, reconstitution of fMLF-induced O2. production in this cell line requires exogenous expression of PKC␦ (17). We have explored this property to study the structure and function relationship of PKC␦-catalyzed phosphorylation of p47phox. Results obtained with this functionally coupled system indicate that fMLF-induced O2. production is accompanied by a rapid and biphasic phosphorylation of PKC␦ at Thr505. A mutation that eliminates Thr505 phosphorylation abolishes the ability of PKC␦ to phosphorylate p47phox in vitro and the ability of fMLF to induce O2. production in transfected cells. A positive correlation between PKC␦ activation loop phosphorylation and NADPH oxidase activity is also observed in differentiated PLB-985 human myelomonoblastic leukemia cell line (30), suggesting that activation loop phosphorylation is a mechanism for the rapid induction of O2. production in fMLF-stimulated cells.

Materials and Methods Reagents The N-formyl peptide fMLF, PMA, and isoluminol were purchased from Sigma-Aldrich. HRP and superoxide dismutase were obtained from Roche. The anti-hemagglutinin (HA) Ab and mouse mAb to ␤-actin were acquired from Santa Cruz Biotechnology. Rabbit polyclonal Abs against the nonphosphorylated PKC␦ and phospho-PKC␦ (Thr505) were obtained from Cell Signaling Technology.

Preparation of PKC and p47phox expression constructs The mouse cDNAs for PKC␣, ␦, ␧, and ␨ were provided by Dr. I. B. Weinstein (Columbia University, New York, NY). The mouse cDNA for PKC␤II was a gift from Dr. C. L. Aschendel (Purdue University, West Lafayette, IN). The mouse PKC␪ cDNA was a gift from Dr. Z. Sun (University of Illinois, Chicago, IL). An HA tag was placed in the C termini of these constructs. The cDNAs were subcloned into the pcDNA3 vector (Invitrogen Life Technologies). PKC chimeras and truncation mutants were prepared by PCR amplification of the selected PKC␦ and PKC␤II

7721 fragments using overlapping oligonucleotide primers with the following sequences: PKC␦-8 (forward) 5⬘-GCCCAATTGGCACCCTTCCTGCGCA TCT-3⬘; PKC␦-2044 (reverse) 5⬘-GCGCAATTGAATGTCCAGGAATTG CTCAAACTT-3⬘; PKC␤-43 (forward) 5⬘-TAACAATTGGCTGACCCGG CTGCGGG-3⬘; PKC␤-2076 (reverse) 5⬘-GCGCAATTGGCTCTTGACT TCAGGTTTTAAAAATT-3⬘; PKC-C (forward) 5⬘-AAAGGCAGCTTTG GCAAGGT-3⬘; PKC-C (reverse) 5⬘-ACCTTGCCAAAGCTGCCTTT-3⬘; and PKC␦-995 (forward) 5⬘-GCGCAATTGAACAACGGGACCTATG GCAAG-3⬘. PCR was performed with denaturing at 94°C (30 s), annealing at 55°C (30 s), and extension at 72°C (3 min), for a total of 25 cycles. The PCR products were subcloned into the pcDNA3 vector. Site-directed mutagenesis of the PKC␦ and p47phox genes was conducted with the QuikChange kit from Stratagene. Selected serine residues in p47phox were mutated to alanine (see Fig. 1B). The cloned PCR products and mutated cDNAs were sequenced to confirm accuracy.

Cell culture and transfection The transgenic COS-phox cells were generated as previously described (18). The cells were maintained at 37°C with 5% CO2 in DMEM supplemented with 10% heat-inactivated FBS, 2 mM L-glutamine, 100 IU/ml penicillin, 50 ␮g/ml streptomycin, 0.2 mg/ml hygromycin (SigmaAldrich), 0.8 mg/ml geneticin (Invitrogen Life Technologies), and 1 ␮g/ml puromycin (Calbiochem). LipofectAMINE 2000 reagent (Invitrogen Life Technologies) was used for transient transfection of 4 ␮g of DNA (2.5 ␮g of an expression vector for human formyl peptide receptor (FPR), 0.5 ␮g of a G␣i2 expression vector, and 1 ␮g of an expression vector for the PKC isoform of interest) into COS-phox cells in a 100-mm culture dish (0.5–1 ⫻ 106 cells). A transfection efficiency of 45–50% was achieved in a typical experiment. Cells were analyzed 21–24 h after transfection. The PLB-985 human myelomonoblastic leukemia cell line (30) was maintained in RPMI 1640 medium supplemented with 2 mM L-glutamine, 10% FBS, 100 IU/ml penicillin, and 50 ␮g/ml streptomycin. Exponentially growing cells at a starting density of 2 ⫻ 105/ml were differentiated for 6 days with 0.5% dimethylformamide in RPMI 1640 containing 1% Nutridoma-SP and 0.5% FCS, with a medium change on day 3. Approximately 95% of the cells underwent granulocytic differentiation by day 6, based on morphologic analysis after May-Gru¨nwald-Giemsa staining. The differentiated cells were used in studies with fMLF and platelet-activating factor (PAF; Biomol).

Measurement of O2. production O2. production in COS-phox cells was determined by an isoluminol-ECL assay (31), in 6-mm diameter wells of a 96-well, flat-bottom, white tissue culture plates (E & K Scientific), as detailed in a previous publication (17). The chemiluminescence count per second was continually recorded at 1-min intervals. Samples containing 250 U of superoxide dismutase, in addition to the stimuli, were run in parallel. The relative level of O2. produced was calculated based on the integrated chemiluminescence. In some experiments, integrated chemiluminescence in a given period of time after fMLF stimulation was expressed as detailed in figure legends.

Analysis of protein expression Whole cell extracts were generated by lysis of transfected COS-phox cells with 200 –500 ␮l of PAGE buffer containing protease inhibitors (Protease Inhibitor Mixture Set I; Calbiochem). Samples were sonicated for 15 s on ice using a Model 60 Sonic Dismembrator (Fisher Scientific), and heated at 95°C for 5 min. Whole cell extracts were analyzed on a 10% denaturing gel. Resolved protein samples were transferred to nitrocellulose membranes (Hybond ECL; Amersham Biosciences) for Western blotting using ECL detection (Pierce).

In vitro kinase assay The coding sequence of a full-length p47phox cDNA was fused to a GST gene in the pGEX4T1 vector (GE Healthcare), generating a GST-p47phox fusion construct. The fusion protein was expressed in E. coli (DH5␣ DE3), and purified on glutathione affinity column according to the manufacturer’s instructions. In vitro kinase assays were performed using immunoprecipitated, recombinant PKC␦ constructs. Briefly, HEK 293T cells in 100-mm dishes were transfected with the HA-tagged expression constructs of PKC␦, or with the control vector (pCDNA3) (2 ␮g/dish). Cellular proteins were extracted with PKC extraction buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 0.1% Tween 20, 1 mM EDTA, 2.5 mM EGTA, 10% glycerol) containing protease inhibitors (10 ␮g/ml aprotinin, 10 ␮g/ml leupeptin, 0.1 mM PMSF) and phosphatase inhibitors (1 mM NaF, 0.1 mM Na3VO4, 10

7722

MECHANISM OF PKC␦-MEDIATED OXIDASE ACTIVATION mM ␤-glycerophosphate). HA-tagged PKC proteins were immunoprecipitated from 300 ␮g of cell extracts using 3 ␮g of an anti-HA Ab and 30 ␮l of protein A/G-Sepharose after 3 h of incubation at 4°C. The immunoprecipitates were washed twice with PKC extraction buffer and once with the PKC kinase buffer (50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM DTT, 2.5 mM EGTA, 1 mM NaF, 0.1 mM Na3VO4, 10 mM ␤-glycerophosphate), and resuspended in 20 ␮l of PKC kinase buffer. The kinase assay was initiated by adding 40 ␮l of the kinase buffer containing 10 ␮g of GST-p47phox fusion protein or histone H1.1 (Santa Cruz Biotechnology) and 5 ␮Ci of [␥-32P] ATP. The reactions were performed at 30°C for 30 min and terminated by adding Laemmli sample buffer and boiled for 5 min. Samples were electrophoresed on 10% SDS-polyacrylamide gel, transferred to nitrocellulose membrane, and exposed to a PhosphorImager screen. The autoradiograph data were quantified using the ImageGauge software (v.3.12; Fuji Photo Film). The blots were then subjected to Western blotting with anti-HA Ab to verify equal loading of the proteins in each reaction.

Results PKC␦-catalyzed phosphorylation of p47phox is necessary for fMLF-induced O2. production

FIGURE 1. PKC␦-mediated p47phox phosphorylation and reconstitution of fMLF-induced NADPH oxidase in COS-phox cells. A, COS-phox cells were transfected with expression constructs for FPR, G␣i2, and either one of the listed PKC isoforms or a control vector as indicated. Twenty-four hours after transfection, cells were assayed for O2. production based on isoluminol-enhanced chemiluminescense in counts per second (cps). The fMLF-induced increase in O2. generation was inhibited in the presence of superoxide dismutase (data not shown). A representative set of data, from one of five experiments that generated similar results, is shown. B, Amino acid sequence of the C-terminal region of p47phox (residues 301–390) is shown. Potential phosphorylation sites to be mutated are underlined. C, Results from in vitro kinase assay using wild-type (WT) and mutant p47phox proteins fused to GST as substrates. The substrates (10 ␮g each) were incubated with 0.2 ␮g of recombinant PKC␦ and [␥-32P]ATP at 30°C

We recently found that reconstitution of fMLF-induced NADPH oxidase activation in the transgenic COS-phox cells requires not only the heterologous FPR but also signaling molecules, such as PKC␦, that are abundant in neutrophils but scanty in COS cells (17). This observation is consistent with results from the use of neutrophil cores, PKC␦-selective inhibitors and antisense depletion (14 –17), all of which suggest an important role of PKC␦ in NADPH oxidase activation. The major isoforms of PKC in neutrophils are PKC␣, ␤II, ␦, and ␨, which belong to three different subclasses of the PKC family and are known to phosphorylate p47phox, an essential component of the phagocyte NADPH oxidase (11, 32). In transfected COS-phox cells, however, only the ␦ isoform can reconstitute fMLF-induced NADPH oxidase activation (Fig. 1A). This result suggests that PKC␦ is important for FPRmediated signaling leading to the catalysis of p47phox phosphorylation and O2. production. We examined whether there is a causal relationship between PKC␦-catalyzed p47phox phosphorylation and fMLF-induced O2. production using COS-phox cells as a functionally coupled system for study of the PKC␦ activation mechanism. The C-terminal autoinhibitory region of p47phox contains multiple serine residues that can be phosphorylated by PKC isoforms and other serine/ threonine kinases (9, 10, 12, 13). Based on a previous report (33), we mutated seven serine residues (underlined in Fig. 1B) to alanine, either individually or in combinations. The resulting p47phox mutants were fused to GST and used as substrates for in vitro kinase assay with recombinant PKC␦. Among the residues selected for mutagenesis, Ser303, Ser304, and Ser328 represent consensus phosphorylation sites for PKC (34). Mutations of Ser303 and Ser304 to alanine reduced phosphorylation of p47phox by ⬃40% (Fig. 1C). Mutation of other selected serine residues reduced p47phox phosphorylation by 25–30%. Additional reduction in p47phox phosphorylation was observed with the triple mutations (S303/304/345A for 30 min. The samples were analyzed by SDS-PAGE, and the autoradiogram was quantified. Several experiments (n ⬎ 3) were performed, and duplicate data (mean ⫾ SD) from a representative experiment are shown. D, The same set of p47phox mutants were individually expressed in the COS91/22 cells, cotransfected with expression vectors for FPR, PKC␦, and p67phox, and assayed for O2. production as described in A. The level of O2. was based on integrated chemiluminescence (int. CL) collected during the first 20 min after fMLF (1 ␮M) stimulation. The expression levels of the p47phox mutants were detected by Western blotting (bottom) using an anti-p47phox Ab, showing equal expression and loading of all samples. A representative set of data is shown as mean ⫾ SEM, and similar results were obtained in two other repeating experiments.

The Journal of Immunology

7723 and S303/304/379A) and the quadruple mutations (S303/304/345/ 379A). Interestingly, mutation of Ser345, a consensus phosphorylation site for MAPKs (35), also impaired p47phox phosphorylation by PKC␦. The effect of p47phox mutations on fMLF-induced O2. production was determined in COS-7 cells that express gp91phox and p22phox but not p67phox and p47phox (COS91/22). Heterologous expression of FPR, p67phox, p47phox, and PKC␦ reconstituted fMLF-induced O2. production (Fig. 1D, first group). All p47phox mutants described were expressed equally well in the transfected cells (Fig. 1D, Western blot). There was a positive correlation between reduced phosphorylation of the p47phox mutants (Fig. 1C) and compromised O2. production (Fig. 1D). Moreover, expression of several p47phox mutants, including S328A, S345A and the double mutants S359/370A and S345/S379A, lowered the basal level of O2. production. The presence of residual activities in the O2. production assay and in vitro kinase assay suggests the presence of additional phosphorylation sites in p47phox that contribute to NADPH oxidase activation. Together, these results demonstrate that fMLF-induced PKC␦ activation is functionally coupled to O2. production through p47phox phosphorylation.

FIGURE 2. Characterization of PKC␦/PKC␤II chimeras and the catalytic and regulatory domains of PKC␦. A, Representation of the chimeric PKC constructs, showing the swapped catalytic and regulatory domains of PKC␤⌱⌱ and PKC␦. B, Expression of the PKC chimeras in COS-phox cells was determined by Western blotting (top), using an anti-HA Ab against the tagged PKC isoforms and another Ab against GAPDH for equal loading control. The fMLF-induced changes in O2. production are shown as a function of time vs counts per second (cps) in the transfected COS-phox cells expressing the chimeras (bottom). At least three independent experiments were performed, and similar results were obtained. C, Deletion mutants were generated to test the roles of the regulatory PKC␦-R and catalytic PKC␦-C domains of PKC␦ in O2. production. The catalytic and regulatory domains of PKC␦ constructs were schematically shown (top), and their expression in the transfected COS-phox cells was confirmed by Western blotting using an anti-HA Ab against the tagged protein constructs (bottom). D, Twenty-four hours after transfection, the catalytic and regulatory domains of PKC␦ constructs were analyzed in O2. production assays in the absence or presence of fMLF (1 ␮M) stimulation. The changes in O2. production were quantified based on integrated chemiluminescence (Int. CL) collected during the first 20 min after agonist stimulation. Data shown are mean ⫾ SEM from three experiments, each performed in duplicate.

The catalytic domain of PKC␦ contains a structural determinant regulated by fMLF for NADPH oxidase reconstitution PKC isoforms differ in their regulatory domains between the subfamilies. These differences underline the distinct roles of diacylglycerol and Ca2⫹ in regulating the catalytic activity of classic vs novel PKC (20). In comparison, significant structural homology exists between the catalytic domains of different PKC isoforms. To understand whether the catalytic or regulatory domain of PKC␦ determines the ability of this kinase to reconstitute fMLF-induced O2. production in reconstituted COS-phox cells, we prepared chimeric PKC constructs by domain swap between PKC␦ and PKC␤II, the latter being unable to reconstitute NADPH oxidase under the same experimental conditions (Fig. 1A). Two chimeras, one with the regulatory domain of PKC␤II fused to the catalytic domain of PKC␦ (PKC␤II/␦), and the other with the catalytic domain of PKC␤II fused to the regulatory domain of PKC␦ (PKC␦/ ␤II), were generated (Fig. 2A). These chimeras were examined individually in transfected COS-phox cells. Although both

FIGURE 3. Agonist-induced phosphorylation of PKC␦ at Thr505. A, A representative Western blot (n ⫽ 3) showing fMLF-induced (1 ␮M) Thr505 phosphorylation of the truncated PKC␦ catalytic domain (PKC␦-C) as a function of time, using an anti-phospho-Thr505 Ab. The assay was performed 24 h after transfection of COS-phox cells. The expression level of PKC␦ catalytic domain was determined with an anti-PKC␦ Ab. Quantification of the Western blot shown is presented below. B and C, Western blot showing fMLF-induced (1 ␮M) Thr505 phosphorylation of the full-length PKC␦. C, Transfected COS-phox cells were treated with pertussis toxin (500 ng/ml, 4 h) before fMLF stimulation. The samples were analyzed on the same gel with the untreated samples (B). The extent of Thr505 phosphorylation was expressed as the ratio of the signal detected by the anti-phospho-Thr505 Ab to the signal detected by the anti-PKC␦ Ab, which was used for normalization of sample loading. Data shown are representative of at least three experiments, with similar patterns of Thr505 phosphorylation.

7724

MECHANISM OF PKC␦-MEDIATED OXIDASE ACTIVATION

constructs were properly expressed after transfection, only the PKC␤II/␦ construct was able to restore fMLF-stimulated O2. production (Fig. 2B). Our results suggest that the catalytic domain of PKC␦ contains structural features required for the reconstitution of NADPH oxidase upon fMLF stimulation. This possibility was further examined through individual expression of the catalytic and regulatory domains of PKC␦ in COS-phox cells (Fig. 2C). Both PKC␦ domains were adequately expressed with the expected sizes (Fig. 2C). When examined for O2. production, the COS-phox cells transfected with the regulatory domain (PKC␦-R) failed to respond to fMLF, whereas cells expressing the catalytic domain (PKC␦-C) produced a significant amount of O2. without agonist stimulation (Fig. 2D). Of interest was the observation that fMLF stimulation resulted in an additional (⬃53%) increase in O2. production compared with the unstimulated cells that express the catalytic domain of PKC␦. fMLF induces phosphorylation of Thr505 in PKC␦-expressing COS-phox cells Based on our findings, we examined the possibility that the fMLFinduced signaling pathway acts on the catalytic domain and modulates its activity. It is known that several serine and threonine residues in the catalytic domain are phosphorylated before a newly synthesized PKC acquires its catalytic potency, a process sometimes referred to as maturation of PKC (reviewed in Ref. 21). Activation loop phosphorylation may be catalyzed by PDK-1 (22, 23). However, whether activation loop phosphorylation is necessary for the catalytic potency of PKC␦ remains controversial as conflicting results were obtained from several studies (23, 26 –28). Using an anti-phospho-PKC␦ Ab recognizing a phosphorylated Thr505, we determined the level of activation loop phosphorylation as a function of time. In COS-phox cells expressing the catalytic domain of PKC␦ (PKC␦-C), fMLF stimulation caused increases in Thr505 phosphorylation that peaked at 30 and 180 s (Fig. 3A). A similar experiment was conducted with transfected COS-phox cells expressing the full-length PKC␦ (Fig. 3B). A transient increase in the level of Thr505 phosphorylation was also observed 30 s after fMLF stimulation, and was followed by a decrease at 60 and 120 s. A prominent increase in Thr505 phosphorylation was again detected 180 s after fMLF stimulation (Fig. 3B). A comparison with the kinetics of fMLF-induced O2. production (Figs. 1A and 2B) suggested to us that the first increase in Thr505 phosphorylation corresponds to the initial rise of O2. production in reconstituted COS-phox cells. Basal phosphorylation at Thr505 was detected in cells expressing the catalytic domain of PKC␦ as well as the full-length PKC␦. Densitometry analysis showed a basal phosphorylation level of 0.66 ⫾ 0.03 for PKC␦-C and 0.48 ⫾ 0.02 for the full-length PKC␦, based on data from three independent experiments. Although cells expressing the catalytic domain of PKC␦ displayed higher basal level phosphorylation, induced phosphorylation of Thr505 was more evident in transfected COS-phox cells expressing the full-length PKC␦. Pertussis toxin, which ADPribosylates the ␣ subunit of Gi/o proteins on a C-terminal cysteine and prevents its interaction with the receptor (36), reduced fMLFinduced phosphorylation of Thr505 (Fig. 3C), confirming that the fMLF-induced PKC␦ activation loop phosphorylation is a G protein-mediated event. It was notable that pertussis toxin also reduced the basal phosphorylation level. Basal phosphorylation at Thr505 may result from the constitutive (ligand-independent) activity of FPR, which became more evident in cells overexpressing FPR (37). Because both constitutive and agonist-induced activation of FPR involves Gi coupling, they are subjected to inhibition by pertussis toxin.

FIGURE 4. A comparison of fMLF and PAF for their induction of O2. production, Ca2⫹ flux, and PKC␦ phosphorylation in PLB-985 cells. PLB-985 or dimethylformamide-differentiated PLB-985 (6 days after addition of 0.5% dimethylformamide; Diff.) were stimulated with fMLF or PAF (100 nM) for their responses in O2. production (A and B), Ca2⫹ flux (C and D) and activation loop phosphorylation at Thr507 (equivalent to Thr505 in mouse PKC␦) (E and F). The O2. production assays were conducted with real-time measurement of isoluminol-ECL as described in Materials and Methods. Each sample contains ⬃1 ⫻ 106 cells. Ca2⫹ flux was measured with 5 ⫻ 105 cells/sample using Indo-1/AM as an indicator, and expressed as the ratio of emission fluorescence measured at 405 nm and 485 nm (relative fluorescence). Phosphorylation assays were performed after a 24-h serum starvation of PLB-985 and differentiated PLB-985 (n ⫽ 3). Graphs represent the level of Thr507 phosphorylation, determined as the ratio of the band density detected by the anti-phospho-Thr507 Ab vs the band density detected by the anti-PKC␦ Ab. The latter serves as a loading control.

Induction of activation loop phosphorylation correlates with fMLF stimulation of O2. production The relationship between PKC␦ activation loop phosphorylation and NADPH oxidase activation was investigated using PLB-985, a human myelomonoblastic leukemia cell line (30). After 6 days of differentiation with 0.5% dimethylformamide, PLB-985 acquired the ability to respond to fMLF with O2. production (Fig. 4A). In comparison, PAF, a priming agent that does not directly activate NADPH oxidase in neutrophils, did not stimulate a significant production of O2. in differentiated PLB-985 cells (Fig. 4B). Both fMLF and PAF induced a rapid Ca2⫹ flux, suggesting the presence of functional receptors on the cell surface that could mediate proximal signaling (Fig. 4, C and D). However, only fMLF induced biphasic phosphorylation of human PKC␦ at Thr507 (equivalent to Thr505 in mouse PKC␦). These results demonstrate a correlation between induction of PKC␦ activation loop phosphorylation and the ability of the agonist to stimulate O2. production.

The Journal of Immunology

7725

FIGURE 6. The kinase activity of wild-type (WT) and mutant PKC␦ constructs using GST-p47phox as substrate. In vitro kinase assays were performed using immunoprecipitated PKC␦ mutants from transfected HEK 293T cells and a recombinant GST-p47phox fusion protein as substrate, as detailed in Materials and Methods. After incubation at 30°C for 30 min in the presence of [␥-32P]ATP, the reaction was terminated. The samples were analyzed by SDSPAGE and autoradiography. Two phosphorylated bands were identified: Top, representing autophosphorylated PKC␦ (apparent molecular mass ⬇78 kDa). Bottom, representing phosphorylated GST-p47phox (apparent molecular mass ⬇72 kDa). The activation loop substitutions E500A (A) and T505A (B), the turn motif mutant S643A (C), and the hydrophobic motif mutant S662A (D) were compared. The GST-p47phox content in each kinase assay was determined by Western blotting using an anti-GST Ab (bottom). The K376R mutant was used as a negative control in the kinase assays. At least three independent kinase assays were performed, and similar results were obtained. A representative set of data is shown.

FIGURE 5. Different effects of PKC␦ catalytic domain mutations in fMLF-induced O2. production. Site-directed mutagenesis was performed in the catalytic domain of PKC␦, as schematically shown in A and detailed in Materials and Methods. Each mutant of PKC␦ was analyzed along with PKC␦K376R (deficient in ATP binding) in transfected COS-phox cells. B, Characterization of the activation loop substitutions, E500A and T505A. C, Characterization of the turn motif mutant S643A, and the hydrophobic motif mutant S662A. The expression level of the wild-type (WT) and mutant PKC␦ was examined with anti-HA for detection of the tagged PKC␦ constructs. The fMLF-induced changes in O2. production was shown based on integrated chemiluminescence (Int. CL) determined in the first 20 min after agonist (1 ␮M fMLF) stimulation. Data shown are mean ⫾ SEM from three experiments for each mutant.

Mutation of Thr505 in the activation loop impairs fMLF-induced NADPH oxidase activation Phosphorylation of a PKC in the catalytic loop, turn motif and hydrophobic motif is thought to be a prerequisite for its catalytic potency (21, 38). Previously published reports indicate that PKC␦ is an exception in that its activation does not require activation loop phosphorylation (26 –28). These studies, however, used pseudosubstrate peptide or histone rather than a physiologically relevant protein as substrates in the in vitro phosphorylation assay. Because PKC␦ is essential for the reconstitution of NADPH oxidase in COS-phox cells, and p47phox is a substrate for PKC-catalyzed phosphorylation, we examined the effect of PKC␦ phosphorylation in its catalytic domain on fMLF-induced O2. production. Phosphorylation of Ser643 in the turn motif (39) was determined using an Ab against the phospho-Ser643 of PKC␦. Persistent phosphorylation of Ser643 was observed in the absence of fMLF and over the entire course of fMLF stimulation in PKC␦-transfected COS-phox cells (data not shown). Therefore, it is unlikely that

Ser643 phosphorylation plays a critical role in the initiation of fMLF-induced NADPH oxidase activation in the transfected COSphox cells. The effects of serine/threonine phosphorylation in the catalytic domain of PKC␦ were further examined using site-directed mutagenesis. Point mutations were individually introduced into the full-length cDNA of PKC␦, producing alanine substitutions at Glu500, Thr505 (activation loop), Ser643 (turn motif) and Ser662 (hydrophobic motif) in the catalytic domain as illustrated in Fig. 5A. In addition, Lys376, an ATP binding site (40), was replaced with an arginine, and the resulting construct (PKC␦K376R) was used as a negative control for the reconstitution assay. When individually examined in the transfected COS-phox cells, the above PKC␦ mutants produced different effects in fMLF-induced O2. production. The E500A substitution resulted in a significant ( p ⱕ 0.01) decrease in the basal level O2. production (Fig. 5B). However, fMLF was still able to stimulate a significant increase in O2. production. In contrast, cells transfected with the T505A mutant responded poorly to fMLF in the reconstituted COS-phox cells, generating only a fraction of O2. when compared with cells expressing the wild-type PKC␦. Reduced basal level O2. production was also observed in cells expressing the T505A mutant. An analysis of data from the transfection experiments found a ⬃35% decrease in O2. production with the S643A mutation, indicating that Ser643 is required for maximal O2. production. In comparison, no change in O2. production was observed with the S662A mutation. Collectively, these results indicate a critical role of Thr505 phosphorylation in basal and induced O2. production in the transfected COS-phox cells. Mutation of Thr505 abrogates PKC␦ autophosphorylation and its ability to phosphorylate p47phox A major function of PKC in NADPH oxidase activation is the catalysis of p47phox phosphorylation (12, 41, 42). To determine

7726

MECHANISM OF PKC␦-MEDIATED OXIDASE ACTIVATION Fig. 6, different results were obtained when histone H1.1 was used as a substrate (Fig. 7). Although autophosphorylation of PKC␦T505A remained minimal, a significant phosphorylation of histone H1.1 was observed with the T505A mutant of PKC␦ (Fig. 7B). This finding may help to reconcile the difference in PKC␦ catalytic specificity and suggest substrate-specific phosphorylation by PKC␦T505A.

Discussion

FIGURE 7. The kinase activity of wild-type (WT) and mutant PKC␦ constructs using histone H1.1 as substrate. In vitro kinase assays were performed using immunoprecipitated PKC␦ mutants from transfected HEK 293T cells, as detailed in Materials and Methods. Recombinant histone H1.1 (Santa Cruz Biotechnology) was used as a substrate in place of GSTp47phox. The activation loop substitutions E500A and T505A (A and B), the turn motif mutant S643A (C), and the hydrophobic motif mutant S662A (D) were compared. The histone protein content in each kinase assay was determined by SDS-PAGE using Coomassie brilliant blue staining (lower panels). The K376R mutant was used as a negative control in the kinase assays. At least three independent in vitro kinase assays were performed, and similar results were obtained from these experiments. A representative experiment is shown.

whether one or more of the mutations alters the ability of PKC␦ to phosphorylate p47phox, we conducted in vitro kinase assay using as substrate a full-length p47phox fused to GST, as described in the legend for Fig. 1. The wild-type PKC␦, expressed in transfected HEK 293T cells and immunoaffinity purified from the cell lysate, strongly phosphorylated the GST-p47phox substrate (lower band in the phosphorimages in Fig. 6) as well as the kinase itself (autophosphorylation; upper band in the phosphorimages in Fig. 6). Mutation of Glu500 to alanine resulted in a small and statistically insignificant decrease ( p ⬎ 0.05) in p47phox phosphorylation and PKC␦ autophosphorylation (Fig. 6A). In contrast, the T505A mutation caused a marked reduction in PKC␦ autophosphorylation and a complete loss of p47phox phosphorylation (Fig. 6B). We also examined Ser to Ala mutations in the turn motif (PKC␦S643A) and hydrophobic motif (PKC␦S662A). Although no significant changes were observed with the alanine substitution at Ser662 (Fig. 6D), a small reduction in PKC␦ autophosphorylation and p47phox phosphorylation were observed with the S643A mutation (Fig. 6C). The phosphorylation patterns of these PKC␦ mutants are consistent with their abilities to reconstitute fMLF-induced NADPH oxidase in the transfected COS-phox cells, shown in Fig. 5. The lack of autophosphorylation in PKC␦T505A suggests that Thr505 phosphorylation is a prerequisite for autophosphorylation at Ser643 and Ser662. The catalytic activity of PKC␦T505A is substrate-specific As noted, different views exist with regard to a role of activation loop phosphorylation in the catalytic activity of PKC␦ (26 –28). To determine whether the use of different substrates has contributed to the discrepancy in these studies, we conducted in vitro kinase assay with histone H1.1, a substrate used in one of the previous studies mentioned. Under the same experimental conditions as in

In this study, we used transgenic COS-phox and COS91/22 cells to investigate the function of PKC␦ in fMLF-induced O2. production. This approach has allowed us to mutate either p47phox or PKC␦ and examine the effects of these mutations on NADPH oxidase activation. In parallel experiments, we studied the effects of the mutations on the catalytic activity of PKC␦ using in vitro kinase assay, with GST-p47phox as a substrate. The results obtained from these functionally coupled studies have led to two major conclusions: 1) fMLF-induced phosphorylation of PKC␦ in its activation loop is a mechanism for the rapid production of O2. in intact cells and 2) the requirement of activation loop phosphorylation for PKC␦-catalyzed phosphorylation is substrate-specific. Agonist-induced phosphorylation of the activation loop in PKC␦ contributes to the rapid production of O2. The major PKC isoforms in neutrophils are PKC␣, ␤II, ␦, and ␨ (9, 11, 32). Previous studies have shown that all these PKC isoforms are able to phosphorylate p47phox when stimulated with PMA, although the kinetics of PKC␦ binding and membrane translocation differ from those of the classic PKC isoforms (12, 32, 43). Twodimensional phosphopeptide mapping indicates that, with the exception of PKC␨, all major PKC isoforms in neutrophils phosphorylate p47phox with a similar phosphopeptide profile (32). Therefore, the catalytic potency of PKC␦ is probably similar to that of PKC␣ and PKC␤II with respect to p47phox phosphorylation. The mechanism of activation, however, may be different between PKC␦ and the two classic PKC isoforms. It has been well established that diacylglycerol binding to the regulatory domain of a PKC causes allosteric changes in its structure, allowing access of the substrate to the catalytic domain of the PKC. In phagocytes, PMA is a potent activator of NADPH oxidase due to its ability to mimic diacylglycerol in the activation of PKCs. PMA-induced O2. production is sustained but lacks the rapid initial rise as seen in fMLF-stimulated cells. The discrepancy in the kinetics of NADPH oxidase activation may be attributed to the mechanism of PKC activation. Results derived from the current study indicate that fMLF-induced PKC␦ phosphorylation in its activation loop may be such a mechanism. In differentiated PLB-985 human myelomonoblastic cells, fMLF-induced O2. production was accompanied by PKC␦ phosphorylation in its activation loop. In comparison, PAF failed to stimulate O2. production and did not induce PKC␦ phosphorylation in its activation loop. PAF is known as a priming agent and not a direct activator of NADPH oxidase. The priming effect results in part from PAF-induced phosphorylation of p67phox but not p47phox (44). Therefore, the ability to induce PKC␦ phosphorylation in its activation loop may distinguish a full agonist for NADPH oxidase activation from a priming agent. In neutrophils, O2. production results from the activation of multiple signaling pathways that may be triggered simultaneously. In contrast, the approaches that we took focus on an individual kinase and the related signaling pathway to gain a better understanding of the underlying mechanism. Therefore, the functional impact of PKC␦ and its activation loop phosphorylation in neutrophil NADPH oxidase activation may be broad or restricted to a particular pathway downstream of the activated FPR. fMLF stimulation

The Journal of Immunology results in guanine nucleotide exchange on the ␣ subunit of the Gi proteins and the subsequent separation of the ␤␥ subunits. These changes activate two pathways: the PLC␤ pathway leads to hydrolysis of phosphatidylinositol 4,5-bisphosphate and the generation of the second messengers, diacylglycerol and inositol 1,4,5trisphosphate, which are responsible for the activation of classic PKC isoforms and possibly novel PKC isoforms. The PI3K pathway causes phosphorylation of phosphatidylinositol 4,5-bisphosphate at the 3⬘ position of the inositol ring and the production of phosphatidylinositol 3,4,5-trisphosphate. The latter is responsible for a variety of cellular activation processes including the activation of kinases, such as PDK-1 and Akt, and guanine nucleotide exchange factors, such as P-Rex1 (45). PKC␦ activation may result from PDK-1-catalyzed activation loop phosphorylation (23), thus representing a branch of the fMLF-induced signaling pathways. Indeed, both pertussis toxin and PI3K inhibitors are known to block fMLF-induced O2. production and other neutrophil functions supporting the hypothesis that PKC␦ involvement in oxidase activation is a terminal function of a more extensive fMLF signaling network. In COS cells, which are relatively inefficient in PLC␤ activation unless additional G protein ␤␥ subunits are expressed (46), the fMLF-induced O2. generation may rely heavily on PKC␦ activation loop phosphorylation that can be triggered by the PI3K pathway. In the study conducted by Yaffe and colleagues (14), it was found that phosphatidylinositol with 3⬘ position phosphorylation (PI(3,4)P2, phosphatidylinositol 3,4,5-trisphosphate, and phosphatidylinositol (3)-phosphate) as well as PKC␦ are essential for the reconstitution of NADPH oxidase in neutrophil cores. There are also published data that contradict the “restricted role” hypothesis and suggest a broader function of PKC␦ in NADPH oxidase activation and other neutrophil activities. For example, studies showing PKC␦ involvement in oxidant production were conducted not only with fMLF (17) but also with other agonists including leukotriene B4 (15), PMA (14), and opsonized zymosan (16). In the current study, we observed that the first wave of PKC␦ activation loop phosphorylation is followed by a rapid decline in its phosphorylation, suggesting possible activation of a protein phosphatase in response to PKC␦ activation. Moreover, knockout of PKC␦ gene in mice leads to compromised neutrophil responses to fMLF, TNF-␣, and IL-8 in O2. production as well as other activities including adhesion, chemotaxis, and lactoferrin release (47). Therefore, it is possible that PKC␦ phosphorylates not only p47phox but also other proteins that are involved in neutrophil activation. The requirement of activation loop phosphorylation for PKC␦-catalyzed phosphorylation is substrate-specific Studies conducted by Stempka and colleagues (26, 27) led to the conclusion that phosphorylation of Thr505 in the activation loop is not essential for mouse PKC␦ to acquire its catalytic competence, although the same process is known to be important for the activation of other PKC isoforms (22–24, 48). This conclusion was based on two observations. First, recombinant PKC␦ produced in bacteria, which lacked Thr505 phosphorylation due to the absence of a “PKC kinase” such as PDK-1, was nonetheless competent for allosteric modulation by activators such as PMA. Secondly, mutation of Thr505 to an alanine did not abolish its catalytic activity in in vitro kinase assay (26). It was proposed that the independence of PKC␦ activity from its activation loop phosphorylation was due to the presence of Glu500, which carries the negative charge and helps to maintain the catalytic competence of PKC␦. Indeed, Stempka and colleagues found that mutation of Glu500 significantly weakens the catalytic activity of mouse PKC␦ (27). A more detailed structure and function analysis of human PKC␦ was con-

7727 ducted by Liu and colleagues recently (28). In this study, the authors observed that Phe500 and Phe527, along with the N-terminal extension of the kinase domain, play important roles in maintaining the kinase activity in the absence of activation loop phosphorylation at Thr507, which is equivalent to Thr505 in mouse PKC␦. Structural analysis has shown that interactions of Phe500 with Tyr334 and Phe527 with Ile499 are important for stabilization of the conformation of PKC␦T507A. As a result, PKC␦T507A is able to phosphorylate PKC␣ pseudosubstrate in vitro (28). However, our results indicate that mutation of Thr505 renders the mouse PKC␦ incapable of phosphorylating p47phox. The difference between our results and the findings made by Liu et al. (28) may be attributed to substrate specificity. In their study, Liu and colleagues found that the human PKC␦T507A prefers certain peptide substrates, but displays poor catalytic activity on other substrates, especially when these substrates are used at higher concentrations. Their experimental data suggest that PKC␦T507A-specific product inhibition is a plausible explanation (28). The possibility that PKC␦T505A catalyzes phosphorylation in a substrate-specific manner is further suggested by our observation that mouse PKC␦T505A could phosphorylate histone H1.1, despite minimal autophosphorylation of PKC␦T505A (Fig. 7B). In most published studies, the catalytic activity of PKC␦ was determined using synthetic peptides or histone as substrates. The current study explores the unique features of p47phox in NADPH oxidase activation, and uses this physiologically relevant protein as substrate for the in vitro kinase assay. The same protein is also used in a whole cell-based reconstitution assay, thereby providing a functionally coupled approach for study of PKC␦ activity in vitro and in cells. It is important to note that phosphorylation of substrate in intact cells may differ from phosphorylation assays conducted in test tubes. For instance, Liu and colleagues (28) observed that the human PKC␦T507A mutant is fully capable of phosphorylating the PKC␣ pseudosubstrate peptide in vitro but fails to stimulate NF-␬B and AP-1 reporter activity in cells. Moreover, this dephosphorylated mutant of human PKC␦ is able to induce apoptosis (28). Therefore, the diversity of intracellular substrates for PKC isoforms is a factor to consider when evaluating the catalytic activity of PKC␦ in the cells. As discussed in the study by Liu et al. (28), reduction in PKC␦ catalytic activity at higher substrate concentration may result from either substrate inhibition or product inhibition. With respect to PKC␦-mediated p47phox phosphorylation, the C-terminal autoinhibitory region contains multiple phosphorylation sites and is an excellent substrate for a number of PKC isoforms. A more detailed analysis will be necessary to delineate the structure and function relationship between PKC␦ activation loop phosphorylation and p47phox phosphorylation, and to determine whether failure to phosphorylate p47phox by PKC␦T505A is the result of substrate inhibition or production inhibition. In summary, our results demonstrate that inducible PKC␦ phosphorylation in its activation loop is required for fMLF-stimulated O2. production in transfected COS-phox cells and in differentiated PLB-985 myelomonoblastic cells. These findings provide an explanation for the rapid production of O2. following fMLF stimulation and suggest the possibility that different PKC isoforms may indeed have nonredundant functions in neutrophil activation. The conclusions drawn from this study also emphasize the importance of using physiologically relevant protein substrates in studies of PKC structure and function.

Acknowledgments We thank Drs. C. L. Aschendel, I. B. Weinstein, and Z. Sun for providing the PKC expression vectors used in this study.

7728

Disclosures The authors have no financial conflict of interest.

References 1. Babior, B. M. 1999. NADPH oxidase: an update. Blood 93: 1464 –1476. 2. Segal, A. W. 2005. How neutrophils kill microbes. Annu. Rev. Immunol. 23: 197–223. 3. Dinauer, M. C., and S. H. Orkin. 1992. Chronic granulomatous disease. Annu. Rev. Med. 43: 117–124. 4. Roos, D. 1994. The genetic basis of chronic granulomatous disease. Immunol. Rev. 138: 121–157. 5. Karathanassis, D., R. V. Stahelin, J. Bravo, O. Perisic, C. M. Pacold, W. Cho, and R. L. Williams. 2002. Binding of the PX domain of p47phox to phosphatidylinositol 3,4-bisphosphate and phosphatidic acid is masked by an intramolecular interaction. EMBO J. 21: 5057–5068. 6. Groemping, Y., K. Lapouge, S. J. Smerdon, and K. Rittinger. 2003. Molecular basis of phosphorylation-induced activation of the NADPH oxidase. Cell 113: 343–355. 7. Ago, T., F. Kuribayashi, H. Hiroaki, R. Takeya, T. Ito, D. Kohda, and H. Sumimoto. 2003. Phosphorylation of p47phox directs phox homology domain from SH3 domain toward phosphoinositides, leading to phagocyte NADPH oxidase activation. Proc. Natl. Acad. Sci. USA 100: 4474 – 4479. 8. Qualliotine-Mann, D., D. E. Agwu, M. D. Ellenburg, C. E. McCall, and L. C. McPhail. 1993. Phosphatidic acid and diacylglycerol synergize in a cell-free system for activation of NADPH oxidase from human neutrophils. J. Biol. Chem. 268: 23843–23849. 9. Majumdar, S., L. H. Kane, M. W. Rossi, B. D. Volpp, W. M. Nauseef, and H. M. Korchak. 1993. Protein kinase C isotypes and signal-transduction in human neutrophils: selective substrate specificity of calcium-dependent ␤-PKC and novel calcium-independent nPKC. Biochim. Biophys. Acta 1176: 276 –286. 10. el Benna, J., L. P. Faust, and B. M. Babior. 1994. The phosphorylation of the respiratory burst oxidase component p47phox during neutrophil activation: phosphorylation of sites recognized by protein kinase C and by proline-directed kinases. J. Biol. Chem. 269: 23431–23436. 11. Kent, J. D., S. Sergeant, D. J. Burns, and L. C. McPhail. 1996. Identification and regulation of protein kinase C-␦ in human neutrophils. J. Immunol. 157: 4641– 4647. 12. Nixon, J. B., and L. C. McPhail. 1999. Protein kinase C (PKC) isoforms translocate to Triton-insoluble fractions in stimulated human neutrophils: correlation of conventional PKC with activation of NADPH oxidase. J. Immunol. 163: 4574 – 4582. 13. Chen, Q., D. W. Powell, M. J. Rane, S. Singh, W. Butt, J. B. Klein, and K. R. McLeish. 2003. Akt phosphorylates p47phox and mediates respiratory burst activity in human neutrophils. J. Immunol. 170: 5302–5308. 14. Brown, G. E., M. Q. Stewart, H. Liu, V. L. Ha, and M. B. Yaffe. 2003. A novel assay system implicates PtdIns(3,4)P2, PtdIns(3)P, and PKC␦ in intracellular production of reactive oxygen species by the NADPH oxidase. Mol. Cell. 11: 35– 47. 15. Bankers-Fulbright, J. L., H. Kita, G. J. Gleich, and S. M. O’Grady. 2001. Regulation of human eosinophil NADPH oxidase activity: a central role for PKC␦. J. Cell. Physiol. 189: 306 –315. 16. Bey, E. A., B. Xu, A. Bhattacharjee, C. M. Oldfield, X. Zhao, Q. Li, V. Subbulakshmi, G. M. Feldman, F. B. Wientjes, and M. K. Cathcart. 2004. Protein kinase C␦ is required for p47phox phosphorylation and translocation in activated human monocytes. J. Immunol. 173: 5730 –5738. 17. He, R., M. Nanamori, H. Sang, H. Yin, M. C. Dinauer, and R. D. Ye. 2004. Reconstitution of chemotactic peptide-induced nicotinamide adenine dinucleotide phosphate (reduced) oxidase activation in transgenic COS-phox cells. J. Immunol. 173: 7462–7470. 18. Price, M. O., L. C. McPhail, J. D. Lambeth, C. H. Han, U. G. Knaus, and M. C. Dinauer. 2002. Creation of a genetic system for analysis of the phagocyte respiratory burst: high-level reconstitution of the NADPH oxidase in a nonhematopoietic system. Blood 99: 2653–2661. 19. Nishizuka, Y. 1992. Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C. Science 258: 607– 614. 20. Mellor, H., and P. J. Parker. 1998. The extended protein kinase C superfamily. Biochem. J. 332: 281–292. 21. Newton, A. C. 2003. Regulation of the ABC kinases by phosphorylation: protein kinase C as a paradigm. Biochem. J. 370: 361–371. 22. Dutil, E. M., A. Toker, and A. C. Newton. 1998. Regulation of conventional protein kinase C isozymes by phosphoinositide-dependent kinase 1 (PDK-1). Curr. Biol. 8: 1366 –1375. 23. Le Good, J. A., W. H. Ziegler, D. B. Parekh, D. R. Alessi, P. Cohen, and P. J. Parker. 1998. Protein kinase C isotypes controlled by phosphoinositide 3-kinase through the protein kinase PDK1. Science 281: 2042–2045.

MECHANISM OF PKC␦-MEDIATED OXIDASE ACTIVATION 24. Chou, M. M., W. Hou, J. Johnson, L. K. Graham, M. H. Lee, C. S. Chen, A. C. Newton, B. S. Schaffhausen, and A. Toker. 1998. Regulation of protein kinase C␨ by PI 3-kinase and PDK-1. Curr. Biol. 8: 1069 –1077. 25. Parekh, D. B., W. Ziegler, and P. J. Parker. 2000. Multiple pathways control protein kinase C phosphorylation. EMBO J. 19: 496 –503. 26. Stempka, L., A. Girod, H. J. Muller, G. Rincke, F. Marks, M. Gschwendt, and D. Bossemeyer. 1997. Phosphorylation of protein kinase C␦ (PKC␦) at threonine 505 is not a prerequisite for enzymatic activity: expression of rat PKC␦ and an alanine 505 mutant in bacteria in a functional form. J. Biol. Chem. 272: 6805– 6811. 27. Stempka, L., M. Schnolzer, S. Radke, G. Rincke, F. Marks, and M. Gschwendt. 1999. Requirements of protein kinase c␦ for catalytic function: role of glutamic acid 500 and autophosphorylation on serine 643. J. Biol. Chem. 274: 8886 – 8892. 28. Liu, Y., N. V. Belkina, C. Graham, and S. Shaw. 2006. Independence of protein kinase C-␦ activity from activation loop phosphorylation: structural basis and altered functions in cells. J. Biol. Chem. 281: 12102–12111. 29. Suh, C. I., N. D. Stull, X. J. Li, W. Tian, M. O. Price, S. Grinstein, M. B. Yaffe, S. Atkinson, and M. C. Dinauer. 2006. The phosphoinositide-binding protein p40phox activates the NADPH oxidase during Fc␥IIA receptor-induced phagocytosis. J. Exp. Med. 203: 1915–1925. 30. Tucker, K. A., M. B. Lilly, L. Heck, Jr., and T. A. Rado. 1987. Characterization of a new human diploid myeloid leukemia cell line (PLB-985) with granulocytic and monocytic differentiating capacity. Blood 70: 372–378. 31. Dahlgren, C., and A. Karlsson. 1999. Respiratory burst in human neutrophils. J. Immunol. Methods 232: 3–14. 32. Fontayne, A., P. M. Dang, M. A. Gougerot-Pocidalo, and J. El-Benna. 2002. Phosphorylation of p47phox sites by PKC␣, ␤II, ␦, and ␨: effect on binding to p22phox and on NADPH oxidase activation. Biochemistry 41: 7743–7750. 33. Johnson, J. L., J. W. Park, J. E. Benna, L. P. Faust, O. Inanami, and B. M. Babior. 1998. Activation of p47phox, a cytosolic subunit of the leukocyte NADPH oxidase: phosphorylation of Ser-359 or Ser-370 precedes phosphorylation at other sites and is required for activity. J. Biol. Chem. 273: 35147–35152. 34. Pearson, R. B., and B. E. Kemp. 1991. Protein kinase phosphorylation site sequences and consensus specificity motifs: tabulations. Methods Enzymol. 200: 62– 81. 35. Dang, P. M., A. Stensballe, T. Boussetta, H. Raad, C. Dewas, Y. Kroviarski, G. Hayem, O. N. Jensen, M. A. Gougerot-Pocidalo, and J. El-Benna. 2006. A specific p47phox-serine phosphorylated by convergent MAPKs mediates neutrophil NADPH oxidase priming at inflammatory sites. J. Clin. Invest. 116: 2033–2043. 36. Bokoch, G. M., and A. G. Gilman. 1984. Inhibition of receptor-mediated release of arachidonic acid by pertussis toxin. Cell 39: 301–308. 37. Wenzel-Seifert, K., C. M. Hurt, and R. Seifert. 1998. High constitutive activity of the human formyl peptide receptor. J. Biol. Chem. 273: 24181–24189. 38. Toker, A. Signaling through protein kinase C. 1998. Front. Biosci. 3: D1134 –D1147. 39. Steinberg, S. F. 2004. Distinctive activation mechanisms and functions for protein kinase C␦. Biochem. J. 384: 449 – 459. 40. Li, W., J. C. Yu, D. Y. Shin, and J. H. Pierce. 1995. Characterization of a protein kinase C-␦ (PKC-␦) ATP binding mutant: an inactive enzyme that competitively inhibits wild type PKC-delta enzymatic activity. J. Biol. Chem. 270: 8311– 8318. 41. Heyworth, P. G., M. L. Karnovsky, and J. A. Badwey. 1989. Protein phosphorylation associated with synergistic stimulation of neutrophils. J. Biol. Chem. 264: 14935–14939. 42. Nauseef, W. M., B. D. Volpp, S. McCormick, K. G. Leidal, and R. A. Clark. 1991. Assembly of the neutrophil respiratory burst oxidase: protein kinase C promotes cytoskeletal and membrane association of cytosolic oxidase components. J. Biol. Chem. 266: 5911–5917. 43. Reeves, E. P., L. V. Dekker, L. V. Forbes, F. B. Wientjes, A. Grogan, D. J. Pappin, and A. W. Segal. 1999. Direct interaction between p47phox and protein kinase C: evidence for targeting of protein kinase C by p47phox in neutrophils. Biochem. J. 344(Pt. 3): 859 – 866. 44. Brown, G. E., M. Q. Stewart, S. A. Bissonnette, A. E. Elia, E. Wilker, and M. B. Yaffe. 2004. Distinct ligand-dependent roles for p38 MAPK in priming and activation of the neutrophil NADPH oxidase. J. Biol. Chem. 279: 27059 –27068. 45. Welch, H. C., W. J. Coadwell, C. D. Ellson, G. J. Ferguson, S. R. Andrews, H. Erdjument-Bromage, P. Tempst, P. T. Hawkins, and L. R. Stephens. 2002. P-Rex1, a PtdIns(3,4,5)P3- and G ␤␥-regulated guanine-nucleotide exchange factor for Rac. Cell 108: 809 – 821. 46. Wu, D., G. J. LaRosa, and M. I. Simon. 1993. G protein-coupled signal transduction pathways for interleukin-8. Science 261: 101–103. 47. Chou, W. H., D. S. Choi, H. Zhang, D. Mu, T. McMahon, V. N. Kharazia, C. A. Lowell, D. M. Ferriero, and R. O. Messing. 2004. Neutrophil protein kinase C␦ as a mediator of stroke-reperfusion injury. J. Clin. Invest. 114: 49 –56. 48. Keranen, L. M., E. M. Dutil, and A. C. Newton. 1995. Protein kinase C is regulated in vivo by three functionally distinct phosphorylations. Curr. Biol. 5: 1394 –1403.