Effect of priming on activation and localization of phospholipase D-1 in human neutrophils Karen A. Cadwallader 1 , Mohib Uddin 1 , Alison M. Condliffe 1 , Andrew S. Cowburn 1 , Jessica F. White 1 , Jeremy N. Skepper 2 , Nicholas T. Ktistakis 3 and Edwin R. Chilvers 1 1 Respiratory Medicine Division, Department of Medicine, University of Cambridge School of Clinical Medicine, Addenbrooke’s and Papworth Hospitals, Cambridge, UK; 2 Department of Anatomy, University of Cambridge, UK; 3 Department of Signalling, Babraham Institute, Cambridge, UK Phospholipase D (PLD) plays a major role in the activation of the neutrophil respiratory burst. However, the repertoire of PLD isoforms present in these primary cells, the precise mechanism of activation, and the impact of cell priming on PLD activity and localization remain poorly defined. RT-PCR analysis showed that both PLD1 and PLD2 iso- forms are expressed in human neutrophils, with PLD1 expressed at a higher level. Endogenous PLD1 was detected by immunoprecipitation and Western blotting, and was predominantly membrane-associated under control and primed/stimulated conditions. Immunofluorescence showed that PLD had a punctate distribution throughout the cell, which was not altered after stimulation by soluble agonists. In contrast, PLD localized to the phagolysosome membrane after ingestion of nonopsonized zymosan particles. We also demonstrate that tumour necrosis factor a greatly potentiates agonist-stimulated PLD activation, myelo- peroxidase release, and superoxide anion generation, and that PLD activation occurs via a phosphatidylinositol 3- kinase-sensitive and brefeldin-sensitive ADP-ribosylation factor GTPase-regulated mechanism. Moreover, propran- olol, which causes an increase in PLD-derived phosphatidic acid accumulation, caused a selective increase in agonist- stimulated myeloperoxidase release. Our results indicate that priming is a critical regulator of PLD activation, that the PLD-generated lipid products exert divergent effects on neutrophil functional responses, that PLD1 is the major PLD isoform present in human neutrophils, and that PLD1 actively translocates to the phagosomal wall after particle ingestion. Keywords: inflammation; lipid mediators; neutrophils; second messengers; signal transduction. Neutrophils play a critical role in host defence against invading pathogens, but have also been implicated in the mechanism of a wide range of inflammatory diseases. The extent of neutrophil activation is influenced by the prior exposure of these cells to agents such as tumour necrosis factor a (TNFa), granulocyte macrophage colony stimu- lating factor or platelet activating factor (PAF). These Ôpriming agentsÕ promote a dramatic increase in the functional responses evoked by subsequent exposure to secretagogue agonists such as fMLP or interleukin-8, and this excessive activation is thought to be one of the key events underlying neutrophil-mediated tissue damage in vivo [1]. Despite the recognition that priming is such an important regulator of neutrophil physiology, compar- atively little is known of the signalling mechanisms underlying this process. Neutrophil activation induced by an array of G-protein-coupled receptors leads to an increase in phospholipase D (PLD) activity and the hydrolysis of PtdChotoPtdOHandcholine[2];PtdOHisthen metabolized to diacylglycerol (DAG) by the enzyme phosphatidate phosphohydrolase (PAP). Both PtdOH and DAG have been proposed to act as important second messengers linking cell stimulation to various effector functions including phagocytosis [3], degranulation [4], and respiratory burst activity [5,6]. It is now known that mammalian cells contain two major PLD isoforms, PLD1 and PLD2, as well as additional splice variants. Both isoforms have an absolute requirement for the lipid phosphatidylinositol 4,5-bisphosphate, but activation of PLD1 also requires interaction with ADP-ribosylation factor (ARF), RhoA or protein kinase Ca [7], whereas PLD2 has no such requirement. Although the mechanisms of PLD activation have been studied extensively in many cells including neutrophils [2,6,8], much of this work has been conducted in transformed cells using overexpression systems. Hence to date, PLD expression has yet to be demonstrated at a Correspondence to K. Cadwallader, Respiratory Medicine Division, Department of Medicine, University of Cambridge School of Clinical Medicine, Level 5, Box 157, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2QQ, UK. Fax: + 44 1223 762007, Tel.: + 44 1223 762007, E-mail: kc220@cam.ac.uk Abbreviations:TNFa, tumour necrosis factor a; PAF, platelet acti- vating factor; PLD, phospholipase D; DAG, diacylglycerol; ARF, ADP-ribosylation factor; PI3-kinase, phosphatidylinositol 3-kinase; MPO, myeloperoxidase; O 2 À , superoxide anion; PtdCho, phosphat- idylcholine; [ 3 H]PtdBut, [ 3 H]phosphatidylbutanol; CHAPS, 3-[(3- cholamidopropyl)dimethylammonio]-1-propanesulfonate; GEF, guanine nucleotide exchange factor. (Received 15 October 2003, revised 5 May 2004, accepted 6 May 2004) Eur. J. Biochem. 271, 2755–2764 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04204.x protein level in any primary cell. Uncertainty also exists over the nature of the PLD isoforms expressed in neutrophils [9,10] and the cosignals required for PLD activity, in particular the role of phosphatidylinositol 3-kinase (PI3-kinase), which we and others have shown to play a crucial role in the activation of NADPH oxidase [11,12]. Hence we have recently reported that the metabolic product of PI3-kinase, phosphatidylinositol 3-phosphate, can activate the NADPH oxidase complex by binding to the PX domain of the p40 phox component [13]. Of note, the PX domain of p47 phox has also been shown to possess a binding site for PtdOH, although the relevance of this to membrane localization and activation of the oxidase complex has yet to be determined [14]. In this study, we show for the first time that priming with TNFa causes a substantial up-regulation of agonist-stimu- lated PLD enzymatic activity in neutrophils which parallels the enhanced functional responses observed. We identify PLD1 as the major PLD isoform present in human neutrophils and reveal that PLD localizes to the phago- somal membrane after particle ingestion but not to the plasma membrane after stimulation with soluble agonists. Furthermore, we demonstrate that PLD activation occurs via a PI3-kinase-sensitive and brefeldin-sensitive ARF GTPase-regulated mechanism and provide evidence that the lipid products formed after PLD activation have an unexpected and differential effect in supporting degranula- tion and O À 2 responses. Materials and methods Materials Cytochrome c, superoxide dismutase, fMLP, 3-[(3- cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS) and phosphate-buffered saline (NaCl/P i with or without CaCl 2 and MgCl 2 ) were purchased from Sigma Chemical Company (Poole, Dorset, UK). Percoll, dextran and the ECL detection kit were obtained from Amersham Biosciences (Amersham, Bucks., UK). Human recombinant TNFa was supplied by R & D Systems (Abingdon, Oxon, UK). Wortmannin, brefeldin A and propranolol were obtained from Calbiochem (Nottingham, UK). Fetal bovine serum, L -glutamine and RPMI 1640 medium were from Gibco (Flow Laboratories, Rickmansworth, Herts., UK). The pan-PLD1/2 antibody was generated in rabbits using the C-terminal third of hPLD1 (amino acids 770– 1075) as antigen. The PLD1-specific antibody was generated as described previously [15]. Secondary antibodies were obtained from Dako (Ely, Cambs., UK). Nonopsonized zymosan was supplied by Molecular Probes (Eugene, OR, USA). Isolation of human neutrophils Human neutrophils were isolated from venous blood of normal healthy volunteers using dextran sedimentation followed by centrifugation on plasma-Percoll gradients as previously detailed [16]. The viability of cells, as assessed by trypan blue exclusion, was >97% and the purity of neutrophil preparations was routinely >96% with <0.1% mononuclear cell contamination. Measurement of degranulation Agonist-stimulated myeloperoxidase (MPO) release was determined by the 3,3-dimethoxybenzidine method as described previously [17] with the following minor modifi- cations. Human neutrophils (10 6 ) were suspended in NaCl/ P i with Ca 2+ and Mg 2+ (80 lL) and incubated with TNFa (200 UÆmL )1 )orNaCl/P i at 37 °C for 30 min followed by fMLP (100 n M )orNaCl/P i for 10 min. For inhibitor studies, cells were preincubated with compound or vehicle for 10 min before agonist stimulation. The reactions were terminated by using ice-cold NaCl/P i . Supernatants were incubated with phosphate buffer (pH 6.2), containing 3,3-dimethoxybenzidine (0.125 mgÆmL )1 )andH 2 O 2 (0.001%, v/v), for 20 min, at 37 °C. NaN 3 (0.1%, w/v) was added, and the amount of MPO released was measured spectrophotometrically (460 nm) and expressed as a per- centage of the total MPO activity present in 0.2% (v/v) Triton X-100-lysed cells. Measurement of O À 2 generation Respiratory burst activity was assessed by measuring the generation of O À 2 using superoxide dismutase-inhibitable reduction of cytochrome c as described previously [18]. Determination of PLD activity PLD activity was assayed in [ 3 H]lyso-PAF (1-O-[ 3 H]oct- adecyl-sn-glycero-3-phosphocholine)-labelled neutrophils by measuring the formation of [ 3 H]phosphatidylbutanol ([ 3 H]PtdBut) in the presence of 0.3% (v/v) butan-1-ol as described previously [19]. Human neutrophils (5 · 10 6 per 0.24 mL) prelabelled with [ 3 H]lyso-PAF were suspended in NaCl/P i with Ca 2+ and Mg 2+ in the presence of 0.3% (v/v) butan-1-ol. Cells were incubated for the time periods indicated, and reactions terminated with 0.94 mL ice-cold chloroform/methanol (1 : 2, v/v). Total lipids were extrac- ted, and [ 3 H]PtdBut formation was measured by liquid- scintillation counting after TLC separation as detailed previously [20]. RT-PCR procedures Total RNA was isolated using RNeasy mini spin columns (Qiagen, Crawley, West Sussex, UK). RNA (2 lg) was transcribed into cDNA using oligo(dT) primers (Invitrogen Life Technologies, Paisley, UK) and 50 U reverse transcrip- tase (Promega, Southampton, UK). PCR amplification was performed using primer sets specific for PLD1 (sense: 5¢-ATGAGACACCCGGATCATGT; antisense: 5¢-ACT CACTGGACGGGTGAAAG; 496 bp product) and PLD2 (sense: 5¢-CTGCACCCCAACATAAAGGT; anti- sense: 3¢-GTTCTCCAGAGTCCCTGCTG; 594 bp prod- uct). For a control reaction, a specific primer set for b-actin (sense: 5¢-GTGGGGCGCCCCAGGCACCA; antisense: 3¢-CTCCTTAATGTCACGCAGCACGATTTC; 548 bp product) was used. PCR (35 cycles) was performed using 2UampliTaq DNA polymerase (Bioline, London, UK). PCR products were analyzed by 1% agarose gel electro- phoresis and imaged with ethidium bromide under UV light. 2756 K. A. Cadwallader et al.(Eur. J. Biochem. 271) Ó FEBS 2004 Western blot analysis After the above treatments, human neutrophils (10 · 10 6 cellsÆmL )1 ) were washed twice in NaCl/P i without Ca 2+ and Mg 2+ , lysed in 1 mL detergent lysis buffer [50 m M Tris/ HCl, pH 7.5, 150 m M NaCl, 5 m M EDTA, 1% (v/v) Nonidet P40 and 0.5% (v/v) CHAPS supplemented with 1 tablet per 50 mL lysis solution of a broad-spectrum proteinase (Roche Applied Science, Lewes, East Sussex, UK (Complete tablets)] and left on ice for 30 min. Cells were homogenized or briefly sonicated and then spun (5 min, 15 000 g) to remove insoluble material. The supernatant was collected and immunoprecipitated with protein A– Sepharose and the pan-PLD1/2 antibody for 2 h. After being washed and boiled in sample buffer, samples were analyzed by SDS/PAGE (10% gel). For fractionation experiments, cells were lysed in 1 mL hypotonic lysis buffer [10 m M Tris/HCl, pH 7.5, 5 m M MgCl 2 ,1m M dithiothreitol, 1 m M phenylmethanesulfonyl fluoride plus a broad-spectrum proteinase inhibitor tablet (see above)] and left on ice for 30 min. Cells were homogenized or briefly sonicated and then spun to remove insoluble material (5 min, 1500 g). The supernatants were collected and respun (100 000 g, 30 min), and the pellet (membrane fraction) was resuspended in membrane lysis buffer (50 m M Tris/HCl, pH 7.5, 50 m M NaCl, 5 m M MgCl 2 ,1m M dithiothreitol, 1 m M phenylmethanesulfonyl fluoride plus a broad-spectrum proteinase inhibitor tablet). CHAPS was added to both fractions to a final concentra- tion of 0.5% (v/v). Cytosolic and membrane fractions were immunoprecipitated and Western blotted as above. Anti- body-bound proteins were detected by ECL. The specificity of the pan-PLD1/2 antibody was confirmed using whole cell lysates of several cell lines that express PLD1 (U937, NIH 3T3 and CCL39) or PLD2 (Rat1) only. Differences in the molecular mass of the band corresponding to PLD were observed according to the presence of PLD1 or PLD2 in these cell lines (data not shown). CHO cells transfected with human PLD1 were also used as a positive control [15]. Immunofluorescent staining for PLD Nonopsonized zymosan was sonicated and added to neutrophils (25 · 10 6 cellsÆmL )1 in NaCl/P i containing CaCl 2 and MgCl 2 ) in a 5 : 1 particle to cell ratio at 37 °C. After 10 min, cells were diluted 10-fold in autologous serum and immediately cytospun (28 g,5min).Thecytospinswere fixed [4% (v/v) paraformaldehyde, 10 min] and permeabi- lized [0.1% (v/v) Triton X-100, 10 min] before blocking with NaCl/P i /0.5% BSA/1% goat serum. The pan-PLD1/2 anti- body was used at a 1 : 100 dilution, and the secondary fluorescein isothiocyanate goat anti-rabbit Ig used at a dilution of 1 : 300. The same method was used to examine PLD distribution under primed/stimulated conditions (see above). Electron microscopy Neutrophils were fixed in 4% (v/v) formaldehyde in 0.1 M Pipes buffer for 1 h, cryoprotected in 25% (v/v) polypro- pylene glycol, frozen in propane, and freeze-substituted in methanol containing 0.01% (w/v) uranyl acetate at )90 °C. They were subsequently embedded in Lowicryl HM20 at )50 °C [21]. Thin sections mounted on nickel film were incubated in primary antibodies to PLD diluted 1 : 100 in NaCl/P i containing 2% BSA, 0.1% Triton X-100 and 0.1% Tween 20 (w/v/v/v) for 16 h. Primary antibody-binding sites were visualized by incubation with colloidal gold particles (10 nm) conjugated to species-specific secondary antibodies. The sections were stained with uranyl acetate and lead citrate and viewed in a Philips CM 100. Statistical analysis All values are expressed as mean ± SEM from n separate experiments. Where appropriate, results were analyzed by analysis of variance followed by Student–Newman–Keuls post-test with the statistical program PRISM (GraphPad Software, San Diego, CA, USA). Differences were consid- ered significant when P<0.05. Results Effect of TNFa priming on O À 2 generation, MPO release, and PLD activation Before investigating the effect of TNFa priming on PLD activity, we wished to confirm that our experimental system was optimal for demonstrating priming-mediated up-regu- lation of neutrophil effector functions. Figure 1 shows that both O À 2 generation and myeloperoxidase (MPO) release were minimal when either priming agent (TNFa)or activating agent (fMLP) were added alone. However, if cells were incubated with TNFa (200 UÆmL )1 ,30min) before fMLP stimulation (100 n M ,10 min),O À 2 and MPO responses are significantly increased. Detailed time-course analysis of fMLP-stimulated O À 2 release showed that maximal O À 2 generation occurred by 1 min and that levels were back to basal by 2 min. MPO release in response to fMLP was likewise extremely rapid (Fig. 1B) with more than 90% of the response occurring within 5 min. PLD has a unique characteristic in that it can catalyse a transphosphatidylation reaction such that, in the presence of the primary alcohol butan-1-ol, PtdBut is formed rather then PtdOH. PtdBut is a relatively stable product, and the amount formed can be used as a measure of PLD activity. Figure 2A shows that fMLP stimulated a concentration- dependent accumulation of [ 3 H]PtdBut in TNFa-primed cells (EC 50 ¼ 2.8 ± 0.08 n M ) with a maximal response achieved with 100 n M fMLP. TNFa priming significantly enhanced fMLP-stimulated PLD activity compared with the levels observed with TNFa or fMLP alone (Fig. 2B,C) with > 90% of the [ 3 H]PtdBut accumulating within the first 5 min of stimulation. Effect of PI3-kinase and ARF inhibitors on O À 2 and PLD activation Selective pharmacological inhibitors were used to assess the role of PI3-kinase and ARF proteins on the activation of PLD. As illustrated in Fig. 3A,B, the PI3-kinase inhibitor wortmannin (100 n M ) [11] markedly attenuated both fMLP-stimulated O À 2 generation and [ 3 H]PtdBut accumulation. Brefeldin A (100 lgÆmL )1 ), an inhibitor of Ó FEBS 2004 PLD1 activation in neutrophil priming (Eur. J. Biochem. 271) 2757 ARF localization and the guanine nucleotide exchange factors (GEFs) for ARF (BIG-1 and 2), was also found to inhibit O À 2 generation and PLD activity (Fig. 3C,D). O À 2 generation and PLD activity were also measured across a range of brefeldin A (0–300 lgÆmL )1 )and wortmannin (0–100 n M ) concentrations (inserts in Fig. 3). These results indicate that PI3-kinase-sensitive and brefeldin A-sensitive GEFs (ARF1 or ARF3) play an important role in the activation of PLD. Functional role of PLD-derived second messengers To determine the extent to which the activation of neutrophil functional responses was dependent on individ- ual PtdCho-derived second messengers, experiments were performed using butan-1-ol, which diverts a proportion of PtdCho hydrolysis into the formation of PtdBut rather than PtdOH, and propranolol, which at 200 l M , sequesters Fig. 1. Effect of TNFa priming on fMLP-stimulated respiratory burst activity and degranulation. Neutrophils were incubated with TNFa (200 UÆmL )1 )orNaCl/P i for 30 min at 37 °C and then stimulated with fMLP (100 n M ) for 10 min or the times indicated. O À 2 generation (A) and MPO release (B) were determined as described in Materials and methods. The data in the inserts represent mean ± SEM from at least three experiments performed in triplicate. Data points for the time courses in (A) and (B) show a representative experiment performed in triplicate. ***P < 0.001, significant increase in O À 2 generation and MPO release over fMLP-stimulated levels. Fig.2.EffectofTNFa priming on fMLP-stimulated PLD activity. Cells were incubated with TNFa (200 UÆmL )1 )orNaCl/P i for 30 min at 37 °C and then stimulated with various concentrations of fMLP for 10 min (A). [ 3 H]PtdBut accumulation was determined as described in Materials and Methods. (B) Time course of [ 3 H]PtdBut accumulation after treatment with TNFa (200 UÆmL )1 )orNaCl/P i for 30 min at 37 °C and then fMLP (100 n M ) for the times indicated. (C) [ 3 H]PtdBut accumulation after TNFa (200 UÆmL )1 )orNaCl/P i for 30 min at 37 °C and then fMLP (100 n M ) for 10 min. The data represent mean ± SEM from at least three experiments performed in triplicate. ***P < 0.001, significant increase in [ 3 H]PtdBut accumulation over fMLP-stimulated levels. 2758 K. A. Cadwallader et al.(Eur. J. Biochem. 271) Ó FEBS 2004 PtdOH, preventing its conversion into DAG by PAP activity and thus leads to PtdOH accumulation [5,22]. Figure 4 illustrates that, whereas preincubation with prop- ranolol (200 l M ) suppressed fMLP-stimulated O À 2 gen- eration in TNFa-primed cells by 30%, it markedly potentiated MPO release under both fMLP only and TNFa-primed/fMLP-stimulated conditions (by 169 ± 37% and 71 ± 9.2%, respectively). In contrast, butan-1- ol (0.3%, v/v) caused a near complete inhibition of TNFa- primed/fMLP-stimulated MPO release under conditions in which O À 2 generation was only marginally suppressed (Fig 4). Butan-1-ol alone, up to a concentration of 3%, had no direct inhibitory effect on the MPO assay (data not shown). These data suggest that important differences exist with regard to the lipid repertoire required to support MPO release and O À 2 generation, with the former response being more dependent on PtdCho-derived PtdOH. Identification and localization of PLD isoforms in neutrophils To identify the PLD isoform(s) present in neutrophils, PCR primers were designed to unique regions of PLD1 or PLD2 as detected by sequence analysis (data not shown). With the use of a semiquantitative RT-PCR technique, PLD1 was identified as the predominant mRNA present in freshly isolated neutrophils with far lower levels of expression of PLD2 (Fig. 5A). Identical data were obtained using eosi- nophil-depleted neutrophils, confirming that these signals were not consequent on the 1–5% eosinophil contamination present in our granulocyte preparations. Identification of PLD in human neutrophils at a protein level was notably more difficult. Hence PLD could only be detected in immunoprecipitates prepared from whole cell lysates in the presence of 1% Nonidet Fig. 3. Effect of PI3-kinase and ARF inhibitors on fMLP-stimulated respiratory burst and PLD activity in TNFa-primed neutrophils. O À 2 generation and [ 3 H]PtdBut accumulation were assessed after fMLP (100 n M , 10 min) stimulation of TNFa-primed (200 UÆmL )1 , 30 min) cells. The inhibitors wortmannin (100 n M ) (A, B) and brefeldin A (100 lgÆmL )1 ) (C, D) or vehicle were added 10 min before stimulation. Data represent mean ± SEM from three experiments each performed in triplicate. **P < 0.01, ***P < 0.001, significant inhibition of O À 2 generation and [ 3 H]PtdBut accumulation over untreated controls. Inserts show the concentration effects of brefeldin A or wortmannin on O À 2 generation and PLD activity. Data represent a single experiment representative of three independent experiments performed in triplicate. Ó FEBS 2004 PLD1 activation in neutrophil priming (Eur. J. Biochem. 271) 2759 P40 and 0.5% CHAPS (Fig. 5B). Initial PLD immuno- precipitates were generated with an antibody specific for either pan-PLD1/2 or PLD1. Western blotting of these immunoprecipitates with the pan-PLD1/2 antibody con- firmed the presence of PLD1 in human neutrophils (Fig. 5B) with an approximate size of 120 kDa. Lysates of CHO cells transfected with human PLD1 were used as positive controls. The subcellular distribution of PLD and the conse- quences of priming and activation were determined using immunoprecipitation and Western blot analysis of neutro- phil cytosol and membrane fractions using the pan-PLD1/2 antibody (Fig. 5C). PLD was found to be membrane associated under all conditions, with no overt change in the membrane/cytoplasm ratio after TNFa priming and/or fMLP stimulation. With the use of confocal microscopy and the PLD1- specific antibody, PLD was found to exhibit a punctate pattern of distribution (characteristic of granule membrane staining) which did not alter after priming and/or stimula- tion with soluble stimuli (Fig. 6A). However, intense PLD immunostaining was apparent at the margin of the phag- olysosome formed after ingestion of nonopsonized zymosan particles (Fig. 6B). Identical results were observed using the pan-PLD1/2 antibody and two other independent PLD1 Fig. 4. Effect of butan-1-ol and phosphatidate phosphohydrolase inhi- bition on fMLP-stimulated respiratory burst and degranulation in TNFa-primed neutrophils. The inhibitors propranolol (200 l M )and butan-1-ol (0.3%, v/v) or vehicle were added 10 min before stimula- tion. O À 2 generation (A) and MPO release (B) were then assessed after fMLP (100 n M , 10 min) stimulation of TNFa-primed (200 UÆmL )1 , 30 min) cells. Data represent mean ± SEM from three experiments performed in triplicate. ***P < 0.001, significant inhibi- tion of MPO release; P<0.001, significant enhancement of MPO release over untreated controls. Fig. 5. PLD isoform expression and localization in human neutrophils. (A) Total mRNA was extracted from freshly isolated human neu- trophils using RNeasy spin columns (Qiagen), and RT-PCR was carried out as described in Materials and methods. PLD1 was identi- fied as the predominant isoform with only marginal expression of PLD2 in two independent experiments. Equal amplification of b-actin transcripts confirmed identical total RNA content of the samples. (B) Human neutrophils (10 · 10 6 ) were lysed in whole cell lysis buffer and immunoprecipitated using either the pan-PLD1/2 antibody or a PLD1-specific antibody as described. Samples were separated by SDS/ PAGE and analyzed by Western Blotting using the pan-PLD1/2 antibody. Control lysates from CHO cells transfected with human PLD1 or vector alone were also run. (C) Cells were incubated for 30mininthepresenceorabsenceofTNFa (200 UÆmL )1 )beforesti- mulation with fMLP (100 n M , 1 min). Cells were lysed and fraction- ated into cytosolic and membrane fractions as described in Materials and methods. Immunoprecipitations were performed with the pan- PLD1/2 antibody. SDS/polyacrylamide gels were then blotted with the pan-PLD1/2 antibody. s, Cytosolic fraction; p, membrane fraction. 2760 K. A. Cadwallader et al.(Eur. J. Biochem. 271) Ó FEBS 2004 antibodies (Cell Signaling, Beverly, MA, USA; data not shown). The distribution of immunogold label for PLD was similar in all four treatments (Fig. 7). Gold label was seen over both intact and degranulated vesicles and diffusely over the cytoplasm of neutrophils. Label was absent over the nuclei and plasma membrane. No difference in the distri- bution was observed in any of the four treatment groups. With the use of commercially available PLD2-specific antibodies, PLD2 could not be detected by immunoprecip- itation or immunofluorescence techniques under any treat- ment condition. Discussion The recent cloning of the mammalian forms of PLD has led to renewed interest in the regulation and downstream effects of PLD. Although neutrophil priming has been previously shown to result in a small up-regulation of agonist- stimulated PLD activation [6,8,23], the underlying mecha- nisms and the relationship between this event and the functional consequences of priming have not been explored. Our data confirm that the extent of agonist-stimulated PLD activity, O À 2 generation, and MPO release in neutrophils is exquisitely sensitive to whether these cells have been previously primed. Moreover, we show that the onset and maximum rate of PLD activity occurs concurrently with O À 2 generation and MPO release. PI3-kinase has been shown to have an essential role in a number of neutrophil functions including adhesion [24,25], phagocytosis [26], chemotaxis [27], degranulation and respiratory burst activity [11,12]. Here we show that wortmannin, a PI3-kinase inhibitor, abolished both fMLP-stimulated O À 2 generation and [ 3 H]PtdBut genera- tion, providing further evidence that PI3-kinase is upstream of both these responses. Yasui & Komiyama [28] reported that NADPH oxidase is activated by PI3-kinase in the early phase and by PLD in a later phase. Brefeldin A, an inhibitor of ARF GTPases and therefore ARF activation, likewise caused significant inhibition of both fMLP-stimulated PLD activity and O À 2 generation. Moreover, this inhibitor also abolished the oxidative burst in response to nonopsonized zymosan, whereas there was no effect on O À 2 generation with opsonized zymosan (data not shown). Together these observations suggest a selective role for a brefeldin-sensitive ARF–GEF complex in regulating granulocyte responses to soluble stimuli [29,30]. Brefeldin-sensitive PLD activation and O À 2 generation have previously been described [31,32] and are thought to be mediated by Class 1 ARFs such as ARF1 and ARF3. More recently, ARF6 has been implicated in activating PLD and functional responses in chromaffin cells [33], macrophages [34] and epithelial cell lines [35], and in neutrophil-like PLB- 985 cells a specific role for ARF6 (controlled by brefeldin- insensitive GEFs) has been implicated in the activation of NADPH oxidase after fMLP stimulation [36]. We recog- nize, however, that brefeldin A has also been reported to block ARF binding to Golgi membranes and the trans- location of proteins from the endoplasmic reticulum to the Golgi and to cause disassembly of the Golgi complex [37] and hence might be influencing processes other than through a direct effect on ARF itself. Having established a link between PLD activation and the secretory and oxidative responses, we examined the differential effects of the immediate PLD products, PtdOH and DAG, on these processes. Butan-1-ol leads to the preferential formation of PtdBut over PtdOH and thereby reduces agonist-stimulated PtdOH accumulation. Propranolol sequesters PtdOH away from the enzyme phosphatidate phosphohydrolase and thereby augments the accumulation of PtdOH. Hence the ability of propranolol to enhance fMLP-stimulated MPO release in both primed and unprimed cells suggests that PtdOH has a pivotal role in supporting degranulation responses [4,38]. It is of particular interest that propranolol increased the degranulation response observed with fMLP alone to levels normally only reached in primed/stimulated cells. In contrast, our data would suggest that PLD-generated DAG is more essential for supporting O À 2 generation [39]. Fig. 6. PLD1 localizes to the phagosomal membrane on exposure to particulate stimuli. (A) Neutrophils were treated with TNFa (200 UÆmL )1 , 30 min) or fMLP (100 n M , 1 min), fixed and permea- bilized before incubation with the PLD1-specific antibody. Immuno- staining was imaged using confocal microscopy. (B) For immunofluorescence studies, neutrophils were incubated with non- opsonized zymosan in a 5 : 1 particle to cell ratio, fixed, and perme- abilized before incubation with the pan-PLD1/2 antibody. Ó FEBS 2004 PLD1 activation in neutrophil priming (Eur. J. Biochem. 271) 2761 It should be noted, however, that the modest suppression of fMLP-stimulated O À 2 generation in TNFa-primed cells by propranolol suggests that DAG is not the sole mediator of the respiratory burst response. Our results therefore reveal important differences in the lipid-dependency of the oxida- tive and secretory responses in neutrophils. It is probable, however, that DAG and PtdOH act together to activate components of the NADPH oxidase. This theory is supported by earlier work in intact neutrophils showing that activation of O À 2 generation occurs in parallel with the elevation of both PtdOH and DAG levels [40] and in experiments using cell-free systems where PtdOH and DAG work collectively to activate NADPH oxidase components [41]. It should also be noted that propranolol is a b-adrenoceptor antagonist and has also been shown to inhibit protein kinase C [42]. It is possible therefore that some of the effect observed with propranolol reflected protein kinase C inhibition. Characterization of the PLD isoforms present in human neutrophils and indeed most other primary cells has been a major challenge. PLD1 protein expression has previ- ously been described in neutrophils, but the Western blots were not shown [43], and PLD2 message has been shown in peripheral blood leucocytes [44]. Previous investigators have also described two additional human neutrophil PLDs with varying molecular mass (92 kDa and 350– 400 kDa) [9,10]. Using RT-PCR, we have been able to show the presence of both PLD1 and PLD2 isoforms in human peripheral blood neutrophils, with PLD1 being the more abundant isoform. At a protein level, the major isoform was shown to be PLD1 as immunoprecipitation with a PLD1-specific antibody and blotting with the pan- PLD1/2 antibody gave a band of the same density as immunoprecipitation and blotting with the pan-PLD1/2 antibody. Moreover, no staining was identified using currently available commercial PLD2-specific antibodies despite their ability to detect PLD2 in Rat1 cells (data not shown). Using the pan-PLD1/2 antibody, we show that PLD is associated with the membrane fraction under control conditions and that this localization pattern did not alter on priming or stimulation. Unstimulated neutrophils exhibited a punctate pattern of PLD immunostaining which we predict from our cell fractionation data to represent localization to intracellular granules. Soluble stimuli produced no change in PLD localization. With particulate stimuli, however, PLD was shown to translocate to the phagosomal membrane after engulfment of nonopsonized zymosan particles. Others have shown that an epitope-tagged form of PLD1 expressed in RBL-2H3 cells localizes to secretory granules and lysosomes of unstimulated cells and, on stimulation, translocates to the plasma membrane [45]. Other investiga- tors have reported that PLD1 is localized on the Golgi [46], nucleus [47] and lysosomal/endosomal compartments [48]. It is possible therefore that the subcellular distribution of PLD is cell specific and may depend on the nature of the stimulus. Although previous studies in human neutrophils have demonstrated translocation of ARF1-regulated PLD activity from secretory granules to the plasma membrane after fMLP stimulation [49], no shift in PLD protein was apparent in our studies. This was despite the induction of maximal degranulation (using an optimal priming/activa- Fig. 7. Gold electron microscopy of PLD localization in human neutrophils. Gold label is present over intact vesicles (single arrowheads) and degranulated vesicles (double arrow- heads) and diffusely over the cytoplasm. Label is not present over the nuclei (n). Scale bar is 100 nm. (A) Control; (B) TNFa treatment; (C) fMLP treatment; (D) TNFa and fMLP. 2762 K. A. Cadwallader et al.(Eur. J. Biochem. 271) Ó FEBS 2004 tion strategy) and clear evidence of granule depletion on electron microscopy (data not shown). Given that we found no translocation of PLD from the cytosol to the crude membrane fraction, we investigated the possibility that the uplift in PLD activity reflected translocation of ARF rather than PLD. Preliminary data (not shown) have indicated that under basal and TNFa-primed conditions, ARF1 and ARF6 are distributed equally between the cytosol and membrane fractions and that priming/stimulation increases the amount of membrane-associated ARF1/6. Using im- munogold electron microscopy, we again found no differ- ence in PLD distribution between primed, stimulated or primed/stimulated cells. Gold label was found over both intact and degranulated vesicles and diffusely over the cytoplasm of neutrophils in all treatment groups. In summary, our data show that priming causes a major up-regulation of agonist-stimulated PLD activity which parallels the increase in O À 2 generation and MPO release. The products of PLD activation appear to have a differ- ential effect on these responses, with PtdOH able to support degranulation and DAG being more important for respir- atory burst activity. PLD activation and O À 2 generation are both dependent on PI3-kinase and ARF1/3. We show for the first time that the major PLD isoform present in neutrophils is PLD1 and that only particulate stimuli cause a major relocation of PLD protein within the cell. These studies provide important information on the mechanisms underlying granulocyte activation. Acknowledgements This work was funded by The Wellcome Trust, BBSRC, The Medical Research Council and the British Lung Foundation. M.U. was supported by an MRC studentship. A.M.C holds a Wellome Trust Advanced Fellowship, and J.F.W. an MRC Clinical Training Fellow- ship. K.A.C. is a British Lung Foundation Research Scientist. Electron microscopy was carried out in the MIC which was established with funding from the Wellcome Trust. We thank Dr Nancy Pryer, Onyx Pharmaceuticals, USA for kindly donating the pan-PLD1/2 antibody and Professor Michael Wakelam for helpful discussions. References 1. Condliffe, A.M., Kitchen, E. & Chilvers, E.R. (1998) Neutrophil priming: pathophysiological consequences and underlying mechanisms. Clin. Sci. 94, 461–471. 2. Exton, J.H. (1998) Phospholipase D. Biochim. Biophys. Acta 1436, 105–115. 3. Serrander, L., Fallman, M. & Stendahl, O. (1996) Activation of phospholipase D is an early event in integrin-mediated signalling leading to phagocytosis in human neutrophils. Inflammation 20, 439–450. 4. Zhou, H.L., Chabot-Fletcher, M., Foley, J.J., Sarau, H.M., Tzimas, M.N., Winkler, J.D. & Torphy, T.J. (1993) Association between leukotriene B4-induced phospholipase D activation and degranulation of human neutrophils. Biochem. Pharmacol. 46, 139–148. 5. Agwu, D.E., McPhail, L.C., Sozzani, S., Bass, D.A. & McCall, C.E. (1991) Phosphatidic acid as a second messenger in human polymorphonuclear leukocytes. Effects on activation of NADPH oxidase. J. Clin. Invest. 88, 531–539. 6. Bauldry, S.A., Bass, D.A., Cousart, S.L. & McCall, C.E. (1991) Tumor necrosis factor alpha priming of phospholipase D in human neutrophils. Correlation between phosphatidic acid production and superoxide generation. J. Biol. Chem. 266, 4173– 4179. 7. Hodgkin,M.N.,Clark,J.M.,Rose,S.,Saqib,K.&Wakelam, M.J. (1999) Characterization of the regulation of phospholipase D activity in the detergent-insoluble fraction of HL60 cells by protein kinase C and small G-proteins. Biochem. J. 339, 87–93. 8. Bourgoin, S., Plante, E., Gaudry, M., Naccache, P.H., Borgeat, P. & Poubelle, P.E. (1990) Involvement of a phospholipase D in the mechanism of action of granulocyte-macrophage colony-stimu- lating factor (GM-CSF): priming of human neutrophils in vitro with GM-CSF is associated with accumulation of phosphatidic acid and diradylglycerol. J. Exp. Med. 172, 767–777. 9. Balsinde, J., Diez, E., Fernandez, B. & Mollinedo, F. (1989) Biochemical characterization of phospholipase D activity from human neutrophils. Eur. J. Biochem. 186, 717–724. 10. Horn, J.M., Lehman, J.A., Alter, G., Horwitz, J. & Gomez- Cambronero, J. (2001) Presence of a phospholipase D (PLD) distinct from PLD1 or PLD2 in human neutrophils: immunobiochemical characterization and initial purification. Biochim. Biophys. Acta 1530, 97–110. 11. Arcaro, A. & Wymann, M.P. (1993) Wortmannin is a potent phosphatidylinositol 3-kinase inhibitor: the role of phosphati- dylinositol 3,4,5-trisphosphate in neutrophil responses. Biochem. J. 296, 297–301. 12. Condliffe, A.M., Hawkins, P.T., Stephens, L.R., Haslett, C. & Chilvers, E.R. (1998) Priming of human neutrophil superoxide generation by tumour necrosis factor-alpha is signalled by enhanced phosphatidylinositol 3,4,5-trisphosphate but not inositol 1,4,5-trisphosphate accumulation. FEBS Lett. 439, 147–151. 13. Ellson, C.D., Gobert-Gosse, S., Anderson, K.E., Davidson, K., Erdjument-Bromage, H., Tempst, P., Thuring, J.W., Cooper, M.A., Lim, Z.Y., Holmes, A.B., Gaffney, P.R., Coadwell, J., Chilvers,E.R.,Hawkins,P.T.&Stephens,L.R.(2001)PtdIns(3)P regulates the neutrophil oxidase complex by binding to the PX domain of p40 (phox). Nat. Cell. Biol. 3, 679–682. 14. Karathanassis, D., Stahelin, R.V., Bravo, J., Perisic, O., Pacold, C.M., Cho, W. & Williams, R.L. (2002) Binding of the PX domain of p47 (phox) to phosphatidylinositol 3,4-bisphosphate and phosphatidic acid is masked by an intramolecular interaction. EMBO J. 21, 5057–5068. 15. Manifava, M., Sugars, J. & Ktistakis, N.T. (1999) Modification of catalytically active phospholipase D1 with fatty acid in vivo. J. Biol. Chem. 274, 1072–1077. 16. Haslett, C., Guthrie, L.A., Kopaniak, M.M., Johnston, R.B. Jr & Henson, P.M. (1985) Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacterial lipo- polysaccharide. Am. J. Pathol. 119, 101–110. 17. Henson, P.M., Zanolari, B., Schwartzman, N.A. & Hong, S.R. (1978) Intracellular control of human neutrophil secretion. I. C5a-induced stimulus-specific desensitization and the effects of cytochalasin B. J. Immunol. 121, 851–855. 18. Cohen, H.J. & Chovaniec, M.E. (1978) Superoxide generation by digitonin-stimulated guinea pig granulocytes. A basis for a con- tinuous assay for monitoring superoxide production and for the study of the activation of the generating system. J. Clin. Invest. 61, 1081–1087. 19. Randall, R.W., Bonser, R.W., Thompson, N.T. & Garland, L.G. (1990) A novel and sensitive assay for phospholipase D in intact cells. FEBS Lett. 264, 87–90. 20. Wakelam, M.J., Hodgkin, M. & Martin, A. (1995) The mea- surement of phospholipase D-linked signaling in cells, Methods Mol. Biol. 41, 271–278. 21. Skepper, J.N. (2000) Immunocytochemical strategies for electron microscopy: choice or compromise. J. Microsc. 199, 1–36. 22. Koul, O. & Hauser, G. (1987) Modulation of rat brain cyto- solic phosphatidate phosphohydrolase: effect of cationic Ó FEBS 2004 PLD1 activation in neutrophil priming (Eur. J. Biochem. 271) 2763 amphiphilic drugs and divalent cations. Arch. Biochem. Biophys. 253, 453–461. 23. L’Heureux, G.P., Bourgoin, S., Jean, N., McColl, S.R. & Nac- cache, P.H. (1995) Diverging signal transduction pathways acti- vated by interleukin-8 and related chemokines in human neutrophils: interleukin-8, but not NAP-2 or GRO alpha, stimu- lates phospholipase D activity. Blood 85, 522–531. 24. Metzner, B., Heger, M., Hofmann, C., Czech, W. & Norgauer, J. (1997) Evidence for the involvement of phosphatidylinositol 4,5-bisphosphate 3-kinase in CD18-mediated adhesion of human neutrophils to fibrinogen. Biochem. Biophys. Res. Commun. 232, 719–723. 25. Capodici, C., Hanft, S., Feoktistov, M. & Pillinger, M.H. (1998) Phosphatidylinositol 3-kinase mediates chemoattractant-stimu- lated, CD11b/CD18-dependent cell-cell adhesion of human neu- trophils: evidence for an ERK-independent pathway. J. Immunol. 160, 1901–1909. 26. Baggiolini, M., Dewald, B., Schnyder, J., Ruch, W., Cooper, P.H. & Payne, T.G. (1987) Inhibition of the phagocytosis-induced respiratory burst by the fungal metabolite wortmannin and some analogues. Exp. Cell Res. 169, 408–418. 27. Sotsios, Y. & Ward, S.G. (2000) Phosphoinositide 3-kinase: a key biochemical signal for cell migration in response to chemokines. Immunol. Rev. 177, 217–235. 28. Yasui, K. & Komiyama, A. (2001) Roles of phosphatidylinositol 3-kinase and phospholipase D in temporal activation of super- oxide production in FMLP-stimulated human neutrophils. Cell Biochem. Funct. 19, 43–50. 29. Brown, H.A., Gutowski, S., Moomaw, C.R., Slaughter, C. & Sternweis, P.C. (1993) ADP-ribosylation factor, a small GTP-dependent regulatory protein, stimulates phospholipase D activity. Cell 75, 1137–1144. 30. Cockcroft, S., Thomas, G.M., Fensome, A., Geny, B., Cunning- ham,E.,Gout,I.,Hiles,I.,Totty,N.F.,Truong,O.&Hsuan,J.J. (1994) Phospholipase D: a downstream effector of ARF in gran- ulocytes. Science 263, 523–526. 31. Kaldi, K., Szeberenyi, J., Rada, B.K., Kovacs, P., Geiszt, M., Mocsai, A. & Ligeti, E. (2002) Contribution of phopholipase D and a brefeldin A-sensitive ARF to chemoattractant-induced superoxide production and secretion of human neutrophils. J. Leukoc. Biol. 71, 695–700. 32. Shome, K., Nie, Y. & Romero, G. (1998) ADP-ribosylation factor proteins mediate agonist-induced activation of phospholipase D. J. Biol. Chem. 273, 30836–30841. 33. Caumont, A.S., Galas, M.C., Vitale, N., Aunis, D. & Bader, M.F. (1998) Regulated exocytosis in chromaffin cells. Translocation of ARF6 stimulates a plasma membrane-associated phospholipase D. J. Biol. Chem. 273, 1373–1379. 34. Zhang, Q., Cox, D., Tseng, C.C., Donaldson, J.G. & Greenberg, S. (1998) A requirement for ARF6 in Fcgamma receptor-mediated phagocytosis in macrophages. J. Biol. Chem. 273, 19977–19981. 35. Criss, A.K., Silva, M., Casanova, J.E. & McCormick, B.A. (2001) Regulation of salmonella-induced neutrophil transmigration by epithelial ADP-ribosylation factor 6. J. Biol. Chem. 276, 48431– 48439. 36. Dana, R.R., Eigsti, C., Holmes, K.L. & Leto, T.L. (2000) A reg- ulatory role for ADP-ribosylation factor 6 (ARF6) in activation of the phagocyte NADPH oxidase. J. Biol. Chem. 275, 32566–32571. 37. Mendez, A.J. (1995) Monensin and brefeldin A inhibit high den- sity lipoprotein-mediated cholesterol efflux from cholesterol- enriched cells. Implications for intracellular cholesterol transport. J. Biol. Chem. 270, 5891–5900. 38. Stutchfield, J. & Cockcroft, S. (1993) Correlation between secre- tion and phospholipase D activation in differentiated HL60 cells. Biochem. J. 293, 649–655. 39. Bonser, R.W., Thompson, N.T., Randall, R.W. & Garland, L.G. (1989) Phospholipase D activation is functionally linked to superoxide generation in the human neutrophil. Biochem. J. 264, 617–620. 40. Catz, S.D. & Sterin-Speziale, N.B. (1996) Bradykinin stimulates phosphoinositide turnover and phospholipase C but not phos- pholipase D and NADPH oxidase in human neutrophils. J. Leukoc. Biol. 59, 591–597. 41. Palicz, A., Foubert, T.R., Jesaitis, A.J., Marodi, L. & McPhail, L.C. (2001) Phosphatidic acid and diacylglycerol directly activate NADPH oxidase by interacting with enzyme components. J. Biol. Chem. 276, 3090–3097. 42. Sozzani, S., Agwu, D.E., McCall, C.E., O’Flaherty, J.T., Schmitt, J.D., Kent, J.D. & McPhail, L.C. (1992) Propranolol, a phos- phatidate phosphohydrolase inhibitor, also inhibits protein kinase C. J. Biol. Chem. 267, 20481–20488. 43. Marcil, J., Harbour, D., Houle, M.G., Naccache, P.H. & Bour- goin, S. (1999) Monosodium urate-crystal-stimulated phospholi- pase D in human neutrophils. Biochem. J. 337, 185–192. 44. Saqib, K.M. & Wakelam, M.J. (1997) Differential expression of human phospholipase D genes. Biochem. Soc. Trans. 25, S586. 45. Brown, F.D., Thompson, N., Saqib, K.M., Clark, J.M., Powner, D., Thompson, N.T., Solari, R. & Wakelam, M.J. (1998) Phos- pholipase D1 localises to secretory granules and lysosomes and is plasma-membrane translocated on cellular stimulation. Curr. Biol. 8, 835–838. 46. Freyberg, Z., Sweeney, D., Siddhanta, A., Bourgoin, S., Frohman, M. & Shields, D. (2001) Intracellular localization of phospholipase D1 in mammalian cells. Mol. Biol. Cell. 12, 943–955. 47. Balboa, M.A. & Insel, P.A. (1995) Nuclear phospholipase D in Madin-Darby canine kidney cells. Guanosine 5¢-O-(thiotriphos- phate)-stimulated activation is mediated by RhoA and is down- stream of protein kinase C. J. Biol. Chem. 270, 29843–29847. 48. Toda, K., Nogami, M., Murakami, K., Kanaho, Y. & Nakayama, K. (1999) Colocalization of phospholipase D1 and GTP-binding- defective mutant of ADP-ribosylation factor 6 to endosomes and lysosomes. FEBS Lett. 442, 221–225. 49. Morgan, C.P., Sengelov, H., Whatmore, J., Borregaard, N. & Cockcroft, S. (1997) ADP-ribosylation-factor-regulated phos- pholipase D activity localizes to secretory vesicles and mobilizes to the plasma membrane following N-formylmethionyl-leucyl- phenylalanine stimulation of human neutrophils. Biochem. J. 325, 581–585. 2764 K. A. Cadwallader et al.(Eur. J. Biochem. 271) Ó FEBS 2004 . Effect of priming on activation and localization of phospholipase D-1 in human neutrophils Karen A. Cadwallader 1 , Mohib Uddin 1 , Alison M. Condliffe 1 ,. P<0.05. Results Effect of TNFa priming on O À 2 generation, MPO release, and PLD activation Before investigating the effect of TNFa priming on PLD activity,