To optimize the transfection of human neutrophils, we used the β-galactosidase expression vector pCMV-βgal, previously described 18. After transfection, cells were incubated for 2, 4, 6 or 8 hours at 37°C and at the appropriate times were centrifuged, washed twice with PBS and lysed by sonication in 2% Nonidet P-40. β-Galactosidase activity was then measured using the β-gal Reporter Gene assay kit (Roche). Maximum galactosidase activity was observed at 2 h after gene transfer when using the proprietary transfection solution ''T'' (amaxa biosystems, Germany) and an electrical setting corresponding to program T27 (Fig. 1A). The transfection efficiency under these experimental conditions was 0.4 to 1%. The viability of neutrophils after the electrical pulse and after the 2-hour recovery period was 78.8 ± 2.5% and 78.5 ± 2.3%, respectively, evaluated using Trypan Blue exclusion. Next, we compared the efficiency of transfection of human neutrophils under various electrical settings corresponding to a series of second generation programs for amaxa nucleofectors. In figure 1B, we show that the electrical settings corresponding to programs U14 and Y01 resulted in higher transfection efficiency for similar experimental conditions based on flow cytometric analysis of neutrophils transfected with the vector pmaxGFP. Also in figure 1B, we show that the second generation programs (U14 and Y01) better preserve cell viability when evaluated by propidium iodide. For consistency and since the initial experiments were performed using program T27, this electrical setting was employed in subsequent experiments unless stated otherwise. Importantly, no differences in the subcellular distribution of the fluorescent chimeras were observed when using program Y01 instead of T27. Next, we used confocal microscopy to examine the transfection and expression of enhanced green fluorescent proteins encoded by the pEGFP (EGFP, enhanced green fluorescent protein) expression vector (Clontech^®) in neutrophils. To demonstrate further that the population of transfected neutrophils was fully functional, we transfected neutrophils with the NADPH oxidase cytosolic factor p47^phox as an EGFP chimera, and we stimulated cells with phorbol ester or with the formylated peptide fMLP which mimics bacteria-derived peptides and stimulates neutrophils through binding to the membrane receptor formyl peptide receptor 1 (FPR1) 19. In figure 2, we show that transfected neutrophils stimulated with PMA or fMLP can undergo morphological changes represented by the appearance of protrusions of the plasma membrane, presumably membrane ruffles. These membrane structures were enriched in EGFP-p47^phox (Fig. 2), which is in agreement with previous reports describing the localization of NADPH oxidase factors at membrane ruffles after activation in non-primary cell lines 20. The distribution pattern for EGFP was different from that of EGFP-p47^phox. EGFP was detected in the nucleus and cytosol, and was also detected in membrane protrusions in stimulated cells (Fig. 2).

To evaluate the effect of nucleofection on NADPH oxidase activation, we first analyzed the ability of nucleoporated neutrophils to produce ROS using the luminol-dependent chemiluminescence detection system. In figure 3A, we show that electroporated cells have relatively low basal luminol-dependent chemiluminescence activity in the absence of stimuli. Importantly, they maintain their capacity to respond to stimuli. Nucleoporated neutrophils show ROS production kinetics similar to that of non-electroporated cells in response to PMA (Fig. 3A). Furthermore, electroporated neutrophils expressing an exogenous protein respond to stimulus (Fig. 3B). The response was detected using the nitroblue tetrazolium reaction. Formazan, the blue-black precipitate generated by superoxide anion-dependent NBT reduction, was evident in stimulated cells but undetectable in unstimulated, transfected cells (Fig. 3B). These results suggest that electroporated neutrophils conserve an intact NADPH oxidase. To further support this idea, we transfected neutrophils with a vector expressing wild type p47^phox downstream of EGFP and then stimulated the cells with PMA. The translocation of p47^phox to the plasma membrane was followed in real time by confocal microscopy analysis (Fig. 4). The results indicate that EGFP-p47^phox, but not the EGFP control, undergoes translocation during activation further supporting the idea that electroporated neutrophils are functional and responsive.

To evaluate whether nucleofection induces more subtle degrees of activation in neutrophils, we examined the surface expression of CD11b, a human leukocyte integrin subunit that is mainly present in the membrane of secretory vesicles and tertiary granules in resting neutrophils. In figure 5, we show that nucleoporated neutrophils have, in fact, a marked increase in CD11b molecules on their cellular surface. The level of CD11b at the plasma membrane after nucleoporation was even larger than that triggered by fMLP stimulation. The increase in the surface expression of CD11b was evident using either program T27 or the second generation program Y01 (Fig. 5A and 5B). In some experiments, two populations of cells were identified according to their level of surface expression of CD11b after nucleoporation (Fig. 5B). Cells expressing GFP were mainly distributed with the subpopulation of cells showing higher surface expression of CD11b (Fig. 5C). These results suggest that nucleofection induces the mobilization of secretory vesicles and, probably, tertiary granules, which are the subpopulation of neutrophil secretory organelles with the highest tendency to undergo exocytosis.

The cytosolic factors p47^phox and p40^phox each contain a PX domain at their amino terminus 16. PX domains are also present in proteins involved in vesicular trafficking and cellular signaling including Vam7p 21 and PI3K-C2 2223, respectively. PX domains are phosphoinositide binding modules 16. The affinity of the various PX domains for several phosphoinositides differs from protein to protein. In particular, the PX domain of p47^phox has been shown to bind preferentially to PI(4)P, PI(3,4)P₂ 16, and PI(3,4,5)P₃ 20. The mechanism mediated by the PX domain of p47^phox in the activation of the oxidase is controversial. One group has suggested that the phosphoinositide-binding activity of the p47^phox PX domain is essential for the membrane translocation of this protein and activation of the phagocyte NADPH oxidase 11. Other groups have indicated that the translocation of the p47^phox PX domain to the plasma membrane is not due to interactions with phospholipids but to association with the actin cytoskeleton 24 and have shown in a variety of cell lines that the PX domain of p47^phox targets to cell membranes via interaction with moesin 2425. Furthermore, while some researchers propose that a single residue substitution in the PX domain of p47^phox decreases its affinity to phosphoinositides and its binding to cell membranes in resting or PMA-stimulated cells 11, others show that the PX domain of p47^phox translocates to the plasma membrane after activation in a phosphoinositide-independent manner in COS cells 24. The subcellular localization of the PX domain of p47^phox in human neutrophils has not yet been shown. To analyze this, we transfected human neutrophils with a vector for the expression of a chimera composed of EGFP upstream of the p47^phox PX-domain (EGFP-p47-PX, Fig. 6). The visualization of the subcellular distribution of the PX domain is somewhat difficult due to the relatively small size of these cells. However, the fluorescence intensity plots helped to identify an increase in the distribution of fluorescence towards the edge of the cells. We concluded that the p47^phox PX domain partially localizes at the plasma membrane in nucleoporated neutrophils. Localization of EGFP-p47-PX in intracellular structures was also evident, a phenomenon previously described in COS cells 20. Differently from the fusion protein EGFP-p47^phox(full-length) which required phorbol ester-mediated neutrophil activation for plasma membrane localization (Fig. 4), the subcellular distribution of EGFP-p47-PX was not markedly affected by cellular stimulation (Fig. 6). Although the concentrations of PI(3,4)P₂ 16, and PI(3,4,5)P₃ in resting neutrophils are considered to be relatively low 26, it could be possible that PI3-kinases undergo some level of activation during nucleoporation thus increasing the membrane distribution of the PX domain in unstimulated cells. However, pretreatment of neutrophils with the PI3-kinase inhibitor LY294002 did not alter the pattern of protein distribution in EGFP-p47-PX transfected neutrophils or prevent the increase in the surface expression of CD11b after nucleoporation (not shown). Another possibility is that once exposed (the PX domain is masked in unphosphorylated full-length p47^phox), the PX domain could recognize pre-existent structures at the plasma membrane, possibly moesin, PI(4)P 16, phosphatidic acid 27 or even basal levels of PI 3-kinase products. From our experiments, it seems that the plasma membrane of nucleoporated neutrophils has the necessary molecular machinery to sequester, at least in part, the p47^phox PX-domain. The molecular basis of this mechanism remains to be clarified.

During particle engulfment, the NADPH oxidase complex is assembled at the membrane of the phagosome so that superoxide anion is pumped into the core of the phagosome where the engulfed microorganism is killed. In particular, p47^phox has been shown to localize towards nascent and mature phagosomes in neutrophils after phagocytosis of opsonized zymosan particles 10. This poses the question of whether the PX domain of p47^phox, which has been proposed to mediate the translocation of p47^phox towards the cytochrome b₅₅₈-containing plasma membrane and to participate in the activation of the NADPH oxidase in cells stimulated with soluble stimuli 11, could also be implicated in the assembly of the oxidase at the phagosome membrane. In an attempt to clarify this, we analyzed the distribution of the PX domain of p47^phox during phagocytosis. We transfected neutrophils with the expression vector pEGFP-p47-PX or with cDNA encoding a chimera of EGFP and RhoB, a small GTPase thought to function in the regulation of endocytosis 28 and phagocytosis 29. Then, we exposed the transfected cells to Texas Red-conjugated opsonized zymosan A particles, and protein distribution during phagocytosis was evaluated by confocal microscopy. In Figure 7, we show that neutrophils expressing EGFP-RhoB or EGFP-p47-PX can undergo phagocytosis. From those experiments, it becomes clear that nucleoporated neutrophils conserve their ability to phagocytose opsonized particles. The accumulation of green fluorescence around the phagosome is evident in EGFP-RhoB-expressing neutrophils but was not observed in cells expressing EGFP-p47-PX (Fig. 7). Similar results were observed in HL-60 promyelocytic cells when differentiated to granulocytes (Fig. 8). These data suggest that the mechanisms of translocation of p47^phox to the plasma membrane and to the phagosome membrane are different. It is likely that the translocation of p47^phox to the phagosome does not involve the PX domain. One possibility is that the SH3 domains of p47^phox, which have been largely shown to bind to the membrane-localized cytochrome b₅₅₈ through interaction with the cytosolic domain of p22^phox, are sufficient to translocate p47^phox to the phagosome. However, p47^phox was clearly detected in forming phagosomes in X-linked CGD neutrophils which lack cytochrome b₅₅₈, indicating that the recruitment of this protein to the nascent phagosome is independent of cytochrome b₅₅₈ 10. In the same work, p47^phox was undetectable in phagosome membranes of mature phagosomes in neutrophils from X-linked CGD patients 10, suggesting that p47^phox can not be retained at the phagosome in the absence of cytochrome b₅₅₈. Likewise, the PX domain of p47^phox localizes at the plasma membrane in the proximity of the opsonized particle in the nascent phagosome (Fig. 7), but is not retained in the membrane of mature phagosomes (Fig. 7 and 8) despite the fact that moesin, the actin-binding protein shown to bind to the PX domain of p47^phox, is present in phagosome membranes 30. Therefore, the reason why the PX domain of p47^phox is not retained in the mature phagosome may reside in the differential role that the products of class I and class III phosphatidylinositol 3-kinase play during phagosome formation and maturation 31. The product of class I PI 3-kinase, PI(3,4,5)P3, has been shown to be rapidly synthesized during phagosomal formation 3233, but the accumulation was sharply restricted to the phagosomal cup 32. Conversely, PI(3)P, a product of class III PI 3-kinase, was only detected in sealed phagosomes 31. These data correlate with our observation that the PI(3,4,5)P3-binding domain of p47^phox (PX domain) is present in membranes in the forming phagosome but is absent from the membrane of the mature phagosome. Although not explored here, it is also possible that the PI(3)P-binding domain of p40^phox (p40^phox-PX domain) is implicated in maintaining the cytosolic complex of the NADPH oxidase at the mature phagosome.

For our studies, we isolated neutrophils from healthy human donors as previously described 34 and resuspended them in phosphate-buffered saline (PBS). Neutrophils were then stored on ice for 30 min or less before use. Immediately before transfection, cells were centrifuged at 1,800 rpm (800 × g) and 4°C for 5 min then resuspended in transfection buffer as indicated below. Ninety μL of neutrophil suspension (2 × 10⁶ cells) were transferred to a nucleoporation cuvette (amaxa Biosystems, Germany), and 5 μg of the indicated cDNA were added to complete a final volume of 100 μl. We found that this concentration of cells is essential to achieve maximum transfection efficiency. Cells were transfected in an amaxa nucleofector apparatus then immediately transferred to 8-well poly-L-lysine–coated chambered glass slides (Lab-Tek) containing RPMI medium supplemented with 10% fetal calf serum (50,000 to 100,000 cells per well in 400 μl of medium). Neutrophils were maintained for 2 h at 37°C in 5% CO₂/air then fixed with 3.7% paraformaldehyde for 10 min. Fixed cells were washed three times with PBS and stored in Fluoromount-G (Southern Biotechnology, CA) until analysis by laser-scanning confocal microscopy on either a Bio-Rad MRC1024 attached to a Zeiβ Axiovert S100TV microscope or a Zeiss (BioRad) Radiance 2100 Rainbow laser scanning confocal microscope (LSCM) attached to a Nikon TE2000-U microscope with infinity corrected optics. Images were collected using the Bio-Rad LaserSharp (v3.2) software. Images were taken at constant exposure times pre-determined to be sub-saturating for the brightest sample. The images were processed and analyzed for the distribution of the fluorescence intensity using the NIH image processing and analysis program IMAGE/J software, IMARIS software (Bitplane AG) and Image-pro plus (MediaCybrnetics^®). In some experiments, neutrophils were stimulated 2 h after transfection using the formylated peptide fMLP (1 μM) or PMA (0.1 μg/ml) for 5 or 20 min, respectively, at 37°C. Where indicated, the subcellular localization of the EGFP-p47^phox chimera was followed in real time using a Zeiss Radiance 2100 Rainbow LSCM attached to a Nikon TE2000-U microscope equipped for viewing live specimens with a Neue temperature and CO₂ controlled live chamber (LiveCell Inc., PA) and a Bioptechs Objective heater (Biotechs, Inc., PA).

Neutrophils were transfected as described above with the expression vectors EGFP-RhoB or EGFP-p47^phox-PX. After transfections, the cells were maintained for 2 h at 37°C in RPMI medium at a cellular concentration of 100,000 neutrophils per 400μl of RPMI in 8-well poly-L-lysine – coated chambered glass slides (Lab-Tek). Then, the cells were incubated in the presence of Texas Red-conjugated zymosan A (S. cerevisiae) BioParticles (Invitrogen, CA), previously opsonized using the Fluorescent Particles Opsonizing Reagent (Invitrogen, CA) as described by the manufacturer. The fluorescent particles were added in a ratio of particles to phagocytes 15:1 or 100:1, the slides were immediately spun down at 1,500 rpm for 5 min at 4°C then incubated for 10 or 15 min at 37°C. Cells were fixed, washed with PBS and stored in Fluoromount-G (Southern Biotechnology) at 4°C until analyzed by laser-scanning confocal microscopy as described above. The quantification of the fluorescent intensity (F_I) at the phagosomal membrane versus the cytosolic distribution of the fluorescent chimera for the green channel was assessed using the NIH image processing and analysis program IMAGE/J software. Briefly, two lines were drawn on each phagosome using the straight line tool. These generated four points where the lines intersected the phagosomal membrane. The intensity of fluorescence at the end of the lines (~ 1 μm outside the phagosome into the cytosol) was subtracted from the fluorescence intensity at the point where the lines intersected the phagosome (ΔF_I). The four independent values were averaged and used as representative of the ΔF_I for that particular phagosome. At least three phagosomes from different cells were analyzed for each chimera.

Human neutrophils were nucleoporated as described above and maintained in RPMI medium for 2 h at 37°C in 8-well chambered coverglass slides at 50.000 cells/well). Then, cells were incubated with 1 mg/ml Nitroblue tetrazolium (NBT) (Bio-Rad Laboratories, CA) in the presence or absence of 0.1 μg/ml PMA, for 30 min at 37°C. After stimulation, cells were washed with PBS, fixed and analyzed by confocal microscopy.

Neutrophils were nucleoporated in solution "T" using an EGFP expression vector and an electrical setting corresponding to program T27. The cells were recovered in serum-free RPMI containing 0.1% gelatin (Sigma, MO) for 2 h at 37°C. Neutrophils (2 × 10⁶) were resuspended in PBS containing 5 mM glucose and 0.1% gelatin. Luminol was added to a final concentration of 1 μM. Cells were left untreated or stimulated with 0.1 μg/ml PMA. Luminol-dependent chemiluminescence was continuously recorded for 65 minutes.

Neutrophils were nucleoporated in solution "T" using an electrical setting corresponding to programs T27, U14 or Y01 in the presence or absence of the expression vector pmaxGFP (amaxa biosystems). The cells were recovered in serum-free RPMI containing 0.1% gelatin (Sigma, MO) for 2 h at 37°C. Untransfected cells were also incubated in RPMI and used as a control. Where indicated, control cells were stimulated with fMLP (1 μM). The cells were spun down, washed and resuspended in flow cytometer diluent buffer (PBS containing 0.5% BSA and 3 mM NaN₃). Cells were incubated with a specific antibody anti-CD11b or with an isotype-matched control (BD Pharmingen, CA). Next, they were incubated with a fluorescein isothiocyanate-conjugated anti-mouse antibody (Jackson ImmunoResearch Laboratories, PA) then fixed in 1% paraformaldehyde. Expression of CD11b antigen on the surface of treated and untreated neutrophils was analyzed by flow cytometry (FACSCalibur BD Biosciences, CA). The data was analyzed using CellQuest™ software (Becton Dickinson, CA). For flow cytometry based viability analysis, cells were labeled for 15 min with propidium iodide (final conc. 10 μg/ml) and analyzed by flow cytometry. GFP fluorescence was detected in the FL-1 channel and propidium iodide using the FL-3 channel.

The various steps in the cloning of the constructs used in this work were performed by standard techniques, and all constructs were verified by sequencing using an automated fluorescent dye-terminator sequencer. The full-length p47^phox and the PX domain of p47^phox (residues 1–130) cDNA were amplified from the full-length cDNA using pfu polymerase (Stratagene, La Jolla, CA) and the following primers: 5' primer

GAATTC

GGTACC

GGTACC

The promyelocytic leukemia human HL-60 cell line (American Type Culture Collection (ATCC), VA) was cultured in Dulbecco's Modified Eagle Medium (D-MEM) (Gibco) supplemented with 20% fetal bovine serum (Hyclone), 0.292 mg/ml glutamine, 50 units/ml penicillin and 50 μg/ml streptomycin at 37°C in 5% CO₂/air. HL-60 cells were differentiated to granulocytes by incubation in the presence of 1.3% DMSO for 48 h. For transfections, 5 × 10⁶ cells were resuspended in 100 μl of Solution "V" (amaxa biosystems) in the presence of 5 μg of the vectors expressing EGFP-p47-PX domain or EGFP-RhoB and nucleofected in the nucleoporator apparatus (amaxa Biosystems, Germany) using the T01 electrical setting. The cells were then re-plated in complete medium in the presence of 1.3% DMSO, incubated at 37°C in 5% CO₂/air and used for analysis 24 h post-transfection. For phagocytosis assays, transfected HL-60 cells were seeded at 70% confluence in an eight-well chambered coverglass (pre-treated with 0.01% poly-L-lysine in PBS) in D-MEM medium for 30 min at 37°C and incubated in the presence of Texas Red^®-conjugated zymosan A (S. cerevisiae) BioParticles^® (Invitrogen), that had been opsonized using the Fluorescent Particles Opsonizing Reagent (Invitrogen) as described by the manufacturer. The fluorescent particles were added in a ratio of particles to phagocytes approximately 15:1 and the slides were immediately spun down at 1,500 rpm for 5 min at 4°C then incubated for 15 min at 37°C. Next, cells were fixed with 3.7% PAF, washed with PBS, and stored in Fluoromount-G (Southern Biotechnology) at 4°C until analysis by laser-scanning confocal microscopy

HL-60 granulocytes were transfected with the expression vectors pEGFP, pEGFP-p47^phox or pEGFP-p47^phox-PX as described above. The cells were lysed using M-PER mammalian protein extraction reagent (Pierce) in the presence of anti-proteases and 40 μg of total protein was resolved by SDS-PAGE, transferred to nitrocellulose and detected using a monoclonal antibody raised against the GFP tag (B-2, Santa Cruz Biotechnology).

All procedures regarding human subjects have been reviewed and approved by the Human Subjects Research Committee at The Scripps Research Institute and were conducted in accordance with the requirements set forth by the mentioned Human Subjects Research Committee.