Assembly of the neutrophil respiratory burst oxidase: A direct interaction between p67^PHOX and cytochrome b₅₅₈

Abstract

Activation of the phagocyte NADPH oxidase complex requires the assembly of the cytosolic factors p47^PHOX, p67^PHOX, p40^PHOX, and Rac1 or Rac2, with the membrane-bound cytochrome b₅₅₈. Whereas the interaction of p47^PHOX with cytochrome b₅₅₈ is well established, an interaction between p67^PHOX and cytochrome b₅₅₈ has never been investigated. We report here a direct interaction between p67^PHOX and cytochrome b₅₅₈. First, labeled p67^PHOX recognizes a 91-kDa band in specific granules from a normal patient but not from a cytochrome b₅₅₈-deficient patient. Second, p67^PHOX binds to cytochrome b₅₅₈ that has been bound to nitrocellulose. Third, GTP-p67^PHOX bound to glutathione agarose is able to pull down cytochrome b_558. Rac1-GTP or Rac1-GDP increased the binding of p67^PHOX to cytochrome b₅₅₈, suggesting that at least one of the oxidase-related functions of Rac1 is to promote the interaction between p67^PHOX and cytochrome b₅₅₈.

The NADPH oxidase of phagocytes plays a pivotal role in host defense against microbial infection. It catalyzes the reduction of oxygen to O at the expense of NADPH. The O generated is the precursor of potent oxidants that are used to kill the invading microorganisms (1). The importance of the NADPH is evidenced by patients with chronic granulomatous disease (CGD), in which an inherited mutation in one of the components of the enzyme impairs its ability to produce superoxide, leading to recurrent bacterial and fungal infections (2).

The NADPH oxidase is a highly regulated enzyme complex composed of a number of proteins: in the cytosol, p67^PHOX, p47^PHOX, p40^PHOX, and Rac1/Rac2, and in the membrane, gp91^PHOX and p22^PHOX, which, together, comprise cytochrome b_558. When activation takes place, the cytosolic components migrate to the membranes, where they associate with the membrane-bound components to assemble the catalytically active oxidase (3). Cytochrome b₅₅₈, which contains both flavin and heme groups and a putative consensus sequence for the binding of NADPH (4), likely is responsible for electron transfer from NADPH to oxygen upon activation of the cells, although it was demonstrated that p67^PHOX also possesses an NADPH-binding site (5). The activation of electron flow in cytochrome b₅₅₈ probably is regulated by the association with p47^PHOX, p67^PHOX, and Rac1/Rac2. P47^PHOX, which is phosphorylated extensively during oxidase activation (6), probably initiates enzyme assembly, whereas p67^PHOX appears to activate catalysis by cytochrome b₅₅₈, although the mechanism of activation is obscure (7). The function of Rac remains unclear, although it is absolutely required for NADPH oxidase activation (8). The function of p40^PHOX also is not well defined. Some results suggest that p40^PHOX is an inhibitory subunit (9), but others showed an increase of O production (10) or an increased affinity for p47^PHOX (11).

During activation, multiple protein–protein interactions occur. Both p47^PHOX and p67^PHOX contain two SH3 domains, which mediate specific interactions between the factors: the C-terminal SH3 domain of p67^PHOX interacts with p47^PHOX, and the N-terminal SH3 domain of p47^PHOX interacts with p22^PHOX (12). Currently, it is proposed that in resting state, p47^PHOX is folded in a masked conformation involving intramolecular interactions between the two SH3 domains. Upon activation, phosphorylation of p47^PHOX disrupts the SH3-mediated intramolecular interaction and p47^PHOX adopts a conformation that allows it to interact with the p22^PHOX (13), bringing p67^PHOX in proximity with cytochrome b₅₅₈. Furthermore, the interaction of Rac with p67^PHOX also is considered crucial for NADPH oxidase activation, because mutant Rac proteins with defective interactions cannot activate the oxidase (14). p40^PHOX interacts with p67^PHOX, although its functional role is not clear (15). Other data suggest a cytochrome b₅₅₈–p67^PHOX interaction. Studies with synthetic peptides of gp91^PHOX suggest not only that p47^PHOX associates with cytochrome b₅₅₈ at multiple sites during the activation but that p67^PHOX also associates with cytochrome b₅₅₈. Indeed, some gp91^PHOX peptides inhibit translocation of p67^PHOX more than that of p47^PHOX in a cell-free system (16). Furthermore, that NADPH oxidase could be activated by exposing cytochrome b₅₅₈ to a high concentration of p67^PHOX in the absence of p47^PHOX (but in the presence of Rac) suggests a direct interaction between cytochrome b₅₅₈ and p67^PHOX (17).

In the present study, we show a direct interaction between p67^PHOX and cytochrome b₅₅₈ and find that this interaction increases when the proteins are incubated in the presence of Rac1-GTP/GDP.

Experimental Procedures

Expression and Purification of Recombinant p67^PHOX and Rac1 in Escherichia coli.

P67^PHOX was expressed in E. coli as a glutathione S-transferase (GST) fusion protein and purified with glutathione-Sepharose beads (18). The protein was separated from GST while on the beads by cleavage with PreScission protease (Amersham Pharmacia) or eluted from the beads by incubating for 20 min at 4°C with 1 ml of 50 mM Tris⋅HCl, pH 8/5 mM glutathione/150 mM NaCl. Rac 1 was expressed in E. coli as a GST fusion protein and purified with glutathione-Sepharose beads (Amersham Pharmacia) followed by thrombin cleavage.

Purification of Cytochrome b₅₅₈.

Cytochrome b₅₅₈ was purified from human neutrophil plasma membranes by heparin and hydroxyapatite chromatography and subsequently relipidated and reflavinated as described by Cross et al. (19).

Overlay, Dot–Blot, and Affinity Precipitation Studies.

For overlay studies, specific granules (2.5 × 10⁷ cells equivalent), prepared as described (20) from a normal subject or a CGD patient deficient in gp91^PHOX, were resolved by SDS/PAGE and transferred to nitrocellulose. The nitrocellulose was blocked in 50 mM Tris, pH 7.5/100 mM NaCl/2 mM EDTA/5% low-fat milk/0.1% Nonidet P-40 for 1 h at room temperature and then incubated in binding buffer (20 mM Hepes, pH 7.5/0.5% low-fat milk/1% Nonidet P-40/10 mM NaCl/1 mM EGTA) in the presence of p67^PHOX (1.5 μg/ml), which had been phosphorylated (see below) in the presence of [γ³²-P]ATP, for 2 h at room temperature. Subsequently, the nitrocellulose was washed five times with wash buffer (50 mM Tris⋅HCl, pH 7.6/50 mM NaCl/0.05% Tween 20/0.1% azide/0.2% low-fat milk), dried, and subjected to autoradiography. The same blots were then stripped in 62.5 mM Tris⋅HCl, pH 6.7/100 mM 2-mercaptoethanol/2% SDS at 50°C for 30 min, and a Western blot was performed by using a mAb directed against gp91^PHOX (a kind gift of U. Knaus, The Scripps Research Institute, La Jolla, CA).

For dot-blot assays, 8.6 pmol each of pure cytochrome b₅₅₈, p47^PHOX, carbonic anhydrase, or BSA was applied directly to nitrocellulose filters, which then were incubated with p67^PHOX (1.5 μg/ml) in binding buffer for 2 h at room temperature. Filters then were washed extensively with wash buffer and incubated with anti-p67^PHOX antibody for detection of binding.

For affinity precipitation, 80 pmol of GST or full-length GST-p67^PHOX was incubated in the presence of 4.3 pmol of cytochrome b₅₅₈ in 200 μl of 20 mM Hepes, pH 7.5/1% Nonidet P-40/10 mM NaCl/1 mM EGTA for 2 h at room temperature. In some assays, phosphorylated GST-67^PHOX was used instead. Forty microliters of glutathione-Sepharose 4B beads (Pharmacia Amersham) was then added followed by incubation for 2 h at 4°C. After washing five times with the same buffer, the beads were pelleted, resuspended in 40 μl of Laemmli sample buffer, and boiled for 5 min. Proteins were separated by SDS/PAGE on a 12% Tris-glycine gel (Bio-Rad) and transferred onto nitrocellulose. The cytochrome b₅₅₈ on the nitrocellulose then was visualized by using a monoclonal anti-gp91^PHOX antibody. To determine the effect of Rac1 and Cdc42 (a kind gift of U. Knaus) on p67^PHOX binding to gp91^PHOX, 800 pmol of Rac 1, Rac1-guanosine 5′-O-(3-thiotriphosphate) (GTP[γS]) or Rac1-guanosine 5′-O-(2-thiodiphosphate) (GDP[βS]), Cdc42, Cdc42-GTPγS, or Cdc42-GDPβS was added to the assay at the same time as p67^PHOX. Rac1 and Cdc42 were loaded with GTP[γS] or GDP[βS] as described by Dang et al. (18).

Western Blotting.

Nitrocellulose membranes were blocked with 5% nonfat dry milk in borate-buffered saline, pH 8.4 (100 mM boric acid/25 mM borax/75 mM NaCl) for 1 h at room temperature and then incubated with 1/5,000 goat anti-p67^PHOX antibody or 1/2,000 mouse monoclonal anti-gp91^PHOX antibody overnight. The membranes then were washed extensively and incubated with alkaline phosphatase-conjugated 1/5,000 anti-rabbit IgG or horseradish peroxidase-conjugated 1/5,000 anti-mouse IgG for 1 h at room temperature. Blots were visualized by using 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium phosphate substrate (Bio-Rad) or enhanced chemiluminescence Western blotting reagents (Amersham Pharmacia).

Phosphorylation.

Phosphorylation of p67^PHOX by MAP kinase and protein kinase C (Calbiochem) in combination was performed by incubating 80 pmol of p67^PHOX in a reaction mixture containing 40 mM Hepes (pH 7.5), 10 mM MgCl₂, 10 mM DTT, 0.6 mM CaCl₂, 5 μg/ml 1,2-dioleoyl-Sn-glycerol (Sigma), 50 μg/ml l-α-phosphatidylserine (Sigma), the indicated kinase, and 50 μM ATP in a total volume of 40 μl for 30 min at 30°C.

Statistical Analysis.

All of the experiments were repeated a minimum of three times. Results are expressed as means ± SE. Statistical analysis was performed by using one-way ANOVA. A P < 0.05 was considered to be statistically significant.

Results

P67^PHOX Interacts Directly with Cytochrome b₅₅₈.

Because previous studies have suggested the existence of an interaction between p67^PHOX and cytochrome b₅₅₈ (21), we aimed to demonstrate this interaction directly by different approaches. In a first approach, an overlay technique was used. Neutrophil-specific granules (used here as a source of cytochrome b₅₅₈) from a normal subject and a CGD patient deficient in gp91^PHOX were separated by SDS/PAGE and transferred to nitrocellulose. The blot then was incubated with p67^PHOX that had been labeled by phosphorylation by using [γ-³²P]ATP. As shown in Fig. 1 Left, the labeled p67^PHOX recognized a major band whose molecular mass of ≈91 kDa corresponds to that of gp91^PHOX. This band was not recognized in specific granules from a CGD patient deficient in gp91^PHOX. To establish clearly the identity of this protein, the same blots were stripped as described under Experimental Procedures and incubated with a monoclonal anti-gp91^PHOX antibody. The anti-gp91^PHOX antibody detected a band that comigrated with the band recognized by the labeled p67^PHOX in the normal specific granules but not in the specific granules from the CGD patient, indicating further that the band recognized by labeled p67^PHOX was the gp91^PHOX subunit of cytochrome b₅₅₈

Figure 1.

Far Western blot with specific granules from normal or CGD patient overlaid with labeled p67^PHOX. Specific granules (2.5 × 10⁷ cell equivalents) from a normal subject or a CGD patient were resolved by 12% SDS/PAGE and blotted onto nitrocellulose. The blot then was incubated with p67^PHOX labeled by phosphorylation with [γ-³²P]ATP. After washing five times, the blot was exposed to x-ray film (Left). The same blot then was stripped as described in Experimental Procedures and incubated with a mAb against gp91^PHOX (Right).

In a second approach, purified cytochrome b₅₅₈, p47^PHOX (positive control), BSA, and carbonic anhydrase (negative controls) were applied directly onto nitrocellulose, which then was incubated with p67^PHOX (1.5 μg/ml). The dot–blot then was washed extensively and treated with p67^PHOX antibody for detection of binding. As shown in Fig. 2 Top, p67^PHOX bound purified cytochrome b₅₅₈ and p47^PHOX but not carbonic anhydrase or BSA. P67^PHOX was not able to bind to denatured cytochrome b₅₅₈ and bound only weakly to denatured p47^PHOX (Fig. 2 Middle). Furthermore, the signal observed did not represent a nonspecific binding of the antibodies to cytochrome b₅₅₈, because in the absence of p67^PHOX (Fig. 2 Bottom), no signal was observed. Finally, the direct binding of p67^PHOX to cytochrome b₅₅₈ was confirmed by affinity-precipitation experiments. Cytochrome b₅₅₈ was incubated in the presence of GST-p67^PHOX or GST. Fig. 3 shows that cytochrome b₅₅₈ was precipitated with GST-p67^PHOX coupled to glutathione-Sepharose beads but not with GST coupled to glutathione-Sepharose beads.

Figure 2.

Binding of p67^PHOX to immobilized cytochrome b₅₅₈. Eight and six-tenths pmol each of pure cytochrome b₅₅₈, p47^PHOX (positive control), carbonic anhydrase, or BSA was applied directly to nitrocellulose filters, which were incubated with p67^PHOX (1.5 μg/ml) as described in Experimental Procedures. The filters then were washed and incubated with anti-p67^PHOX antibody. (Top) Native proteins. (Middle) Denatured proteins. (Bottom) No p67^PHOX.

Figure 3.

Affinity precipitation of cytochrome b₅₅₈ by GST-p67^PHOX. Eighty picomoles of GST or full-length GST-p67^PHOX were incubated in the presence of 4.3 pmol of cytochrome b₅₅₈ in 200 μl of 20 mM Hepes, pH 7.5/1% Nonidet P-40/10 mM NaCl/1 mM EGTA for 2 h at room temperature. Forty microliters of glutathione-Sepharose 4B beads (Amersham Pharmacia) then were added, followed by incubation for 2 h at 4°C. After washing five times with the same buffer, the beads were pelleted, resuspended in 40 μl of Laemmli sample buffer (22), and boiled for 5 min. Proteins were resolved by 12% SDS/PAGE and transferred onto nitrocellulose. The proteins then were detected by using a monoclonal anti-gp91^PHOX antibody.

Effect of Rac1 on the Interaction of p67^PHOX with Cytochrome b₅₅₈.

Of particular interest is the question of whether Rac is able to alter the interaction between cytochrome b₅₅₈ and GST-p67^PHOX. The results (Fig. 4A) showed a major increase in the amount of cytochrome b₅₅₈ associated with p67^PHOX when a larger amount of Rac1 was present. Of interest was the finding that this effect was observed with both Rac1-GTP[γS] and Rac1-GDP[βS]. Under the same conditions, Cdc42, a GTPase closely related to Rac1 but not capable of inducing NADPH oxidase activation, had no effect (Fig. 4B).

Figure 4.

Effect of Rac1 and Cdc42 on the binding of p67^PHOX to cytochrome b₅₅₈. The experiments were performed as described in the text, except that 800 pmol of Rac-GTP[γS]/GDP[βS] or Cdc42-GTP[γS]/GDP[βS] was used. (A Upper) Affinity precipitation in the presence of Rac. Lanes: 2, plus 800 pmol Rac1; 3, plus 800 pmol Rac1-GTP[γS]; 4, plus 800 pmol Rac1-GDP[βS]. (A Lower) Quantification of the binding by scanning densitometry. The data are expressed as percentage of full-length GST-p67^PHOX and are the mean ± SE of three experiments. (B Upper) Affinity precipitation in the presence of Cdc42. Lanes: 2, plus 800 pmol Cdc42; 3, plus 800 pmol Cdc42-GTP[γS]; 4, plus 800 pmol Cdc42-GDP[βS]. (B Lower) Quantification of the binding by scanning densitometry. The data are expressed as percentage of full-length GST-p67^PHOX and are the mean ± range of two experiments.

Discussion

In the present study, we demonstrated that p67^PHOX is able to bind directly to cytochrome b₅₅₈. First, labeled p67^PHOX was able to recognize a 91-kDa band, which, by Western blot, was shown to comigrate with gp91^PHOX in specific granules from a normal subject but not from a CGD patient deficient in cytochrome b₅₅₈. Second, p67^PHOX was able to bind to pure cytochrome b₅₅₈ that had been applied to nitrocellulose. Third, cytochrome b₅₅₈ could be affinity-precipitated by GST-p67^PHOX but not by GST when either had been coupled to glutathione-Sepharose beads. A direct interaction between p67^PHOX and cytochrome b₅₅₈ is in accord with the idea that p67^PHOX regulates the transfer of electrons from NADPH to the flavin (7) because p67^PHOX then would be in proximity to the flavin center, enabling it to perform a regulatory function in this part of the protein. Recently, one specific domain of p67^PHOX, called the activation domain, has been shown to be involved in this regulation (23).

Perhaps the most interesting result of these experiments was the finding that Rac1, a small GTPase that is necessary for the activation of the oxidase in a cell-free system, was able to promote the interaction between cytochrome b₅₅₈ and GST-p67^PHOX. This finding was specific for Rac1, because Cdc42, a closely related small GTPase that, however, is unable to promote the activation of NADPH oxidase, had no effect on this interaction. The interaction was promoted by both Rac1-GTP[γS] and Rac1-GDP[βS], a finding reminiscent of earlier work by Gabig et al. (24), who provided evidence that the oxidase used both the GTP- and GDP-bound forms of a small GTPase, and more recently by Gorzalczany et al. (25), who found that both the GTP[γS]- and GDP[βS]-bound forms of prenylated Rac1 are active in a cell-free system using membrane or purified cytochrome b₅₅₈ and p67^PHOX in the absence of p47^PHOX. These findings suggest that the function of Rac1 in the activation of NADPH oxidase is to promote the interaction between cytochrome b₅₅₈ and p67^PHOX. Alternatively, it might participate in the recruitment of cytosolic p67^PHOX to the membrane environment, as shown by Gorzalczany et al. (25) and by our own results using a high concentration of Rac1, although simple recruitment to the membrane seems unlikely to explain the results of the affinity-precipitation experiments. We cannot exclude the possibility that the increase in the affinity precipitation of cytochrome b₅₅₈ is mediated by a direct interaction of Rac1-GTP[γS] with cytochrome b₅₅₈, binding the cytochrome to GST-p67^PHOX indirectly; but this, too, would be a significant three-way interaction between the cytochrome, GST-p67^PHOX, and Rac1-GTP[γS].

In conclusion, our data demonstrate a direct interaction between p67^PHOX and cytochrome b₅₅₈. Furthermore, the interaction is promoted by Rac1-GTP[γS] (and Rac1-GDP[βS]), but not by the small GTPase Cdc42.

Abbreviations

CGD

chronic granulomatous disease

GST

glutathione S-transferase

GTP[γS]

guanosine 5′-O-(3-thiotriphosphate)

GDP[βS]

guanosine 5′-O-(2-thiodiphosphate)