Rac1 and Rac2 are capable of stimulating superoxide production in vitro, but their targeting and functional mechanisms are still unknown. In the present study, we found that Rac1, 2, and 3 all accumulate at the phagosome during FcγR-mediated phagocytosis, and that the order of accumulation (Rac1 > Rac3 > Rac2) depends on the net positive charge in their polybasic (PB) regions (183–188 aa). Although all GFP-tagged prenylated PB regions of Rac isoforms (GFP-Rac(PB)) and GFP-tagged prenylated 6 Ala (GFP-6A) accumulated during phagocytosis, GFP-Rac2(PB) and GFP-6A showed weak accumulation at the phagosome through a linear structure connecting the phagosome and endomembranes. The PB region of Rac1 showed strong phospholipid interaction with PI(3)P, PI(4)P, PI(5)P, PI(3,4,5)P3, and phosphatidic acid, however, that of Rac2 did not. Constitutively active Rac2, GFP-Rac2(Q61L), was predominantly localized at the endomembranes; these endomembranes fused to the phagosome through the linear structure during phagocytosis, and this accumulation mechanism did not depend on positive charge in the PB region. Our conclusion is that Rac1 directly targets to the phagosome using the positively charged PB region and this accumulation mechanism is likely enhanced by the phospholipids. In addition to this mechanism, Rac2 has a positive charge-independent mechanism in which Rac2 initially targets to endomembranes and then these endomembranes fuse to the phagosome.

The small GTPase Rac is a member of the Rho subfamily of Ras proteins and plays a central role in actin remodeling, chemotaxis, and superoxide (O2)3 production. There are three Rac isoforms that are encoded by different genes. Rac1 and Rac3 are ubiquitously expressed. Although Rac2 expression is restricted to hemopoietic cells and is the predominant isoform in human neutrophils (75–90% of total Rac (1)), murine neutrophils express similar amounts of Rac1 and Rac2 (2). The three isoforms share high amino acid identity in both humans and mice (mRac1 shares 92% identity with mRac2, mRac1 shares 93% identity with mRac3, mRac2 shares 88% identity with mRac3) and differ primarily in the C-terminal 10 residues, comprising a polybasic (PB) sequence (183–188 aa) adjacent to a CAAX sequence (189–192 aa) for isoprenylation. The PB region of Rac1 has six contiguous basic amino acids, while those in Rac2 and Rac3 have four and three basic amino acids, respectively, due to neutral amino acid substitution. These differences among the Rac isoforms may account for their isoform-specific targeting mechanisms.

FcγR-mediated phagocytosis activates NADPH oxidase to produce O2, a precursor of reactive oxygen species with microbicidal activity (3). NADPH oxidase in phagocytes is a multiprotein complex that is assembled from a membrane-spanning cytochrome b558 (gp91phox and p22phox) and four cytosolic factors (p47phox, p67phox, p40phox, and Rac) that translocate to the cytochrome b558 to generate the active enzyme. Early studies revealed that the addition of either Rac1-GTP or Rac2-GTP was essential for high-level O2 production in a cell-free system (4, 5). Results using neutrophils from Rac2−/− mice demonstrated that Rac2 is involved in O2 production in response to IgG-opsonized SRBC (6), fMLP (2, 6), and PMA (6, 7), but not opsonized zymosan (6). Rac1 is not required for fMLP-stimulated O2 production in neutrophils (1); however, Roberts et al. (7) suggested that Rac1 may compensate for Rac2 in PMA-stimulated O2 production in Rac2−/− neutrophils. More recently, Zhao et al. (8) showed that human monocytes use Rac1, but not Rac2, in O2 production stimulated by opsonized zymosan, PMA, and fMLP. Taken together, these results suggest that the isoform-specific function of Rac is determined by both the cell type and the stimulus. As Rac needs to localize to its site of action, and the C-terminal 10 residues may target Rac in an isoform-specific manner, we hypothesized that the isoform-specific targeting of Rac is dictated by its PB region. To test this hypothesis, we constructed chimeric Rac proteins, exchanging the PB region of one Rac isoform for that of another, and examined their accumulation during FcγR-mediated phagocytosis in RAW 264.7 macrophages. Additionally, we examined the ability of the chimeras to support O2 production in a cell-free system. Finally, to elucidate the isoform-specific targeting mechanism of Rac, we determined the specific phospholipid binding of each PB region and analyzed the accumulation of the GFP-tagged prenylated PB region of each isoform during FcγR-mediated phagocytosis.

In this study, we describe the isoform-specific targeting mechanisms of Rac1 and Rac2 during FcγR-mediated phagocytosis. Rac1 accumulation requires the highly positively charged PB region. In contrast, Rac2 has a PB-independent pathway in which it initially concentrates in endomembranes that subsequently fuse into the phagosome. Thus, Rac1 and Rac2 use different accumulation mechanisms for the assembly of NADPH oxidase.

Texas Red-conjugated wheat germ agglutinin, ER-Tracker Blue-White DPX, and Lyso-Tracker Red DND-99 were purchased from Molecular Probes.

RAW 264.7 macrophages were maintained in DMEM (Invitrogen Life Technologies) supplemented with 10% heat-inactivated FBS (Invitrogen Life Technologies) and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin) at 37°C in 5% CO2.

We isolated DNA fragments encoding murine Rac1 and Rac2 from total RNA of the RAW cells by RT-PCR. cDNA encoding murine Rac3 was described previously (accession no. AB040819, GeneBank). The cDNAs encoding constitutive active forms of Rac1 and Rac2, Rac1(Q61L), and Rac2(Q61L), respectively, and a dominant-negative form of Rac1, Rac1(T17N), were generated by PCR-mediated site-directed mutagenesis. DNA fragments encoding the various chimeras (Rac1-2-1, Rac1-3-1, Rac2-1-2, Rac3-1-3, and Rac1-6A-1; see Fig. 1) were generated by PCR with appropriate reverse primers. The following DNA fragments, Rac1(C189S), Rac2(C189S), Rac1-6A-1(C189S), and Rac2-6A-2(Q61L), were also generated by PCR-mediated site-directed mutagenesis with appropriate primers (Fig. 1). These RCR products were subcloned into pGEM-teasy (Promega), and then cloned into the EcoRI site in pEGFP-C1 (BD Clontech) (Fig. 1). Forward and reverse oligonucleotides for C-terminal 10 residues of each Rac isoform containing the PB region (six residues) and CAAX motif (four residues), and those for AAAAAACLLL were annealed and cloned into the EcoRI/SalI site in pEGFP-C1, and designated GFP-Rac1(PB), GFP-Rac2(PB), GFP-Rac3(PB), and GFP-6A, respectively (Fig. 1). All constructs were sequenced to confirm their identities.

FIGURE 1.

Structure of GFP-tagged and GST-tagged Rac constructs used in these study. The PB region and CAAX motif (prenylation site) are indicated. The PB sequences for Rac1 (KKRKRK, six basic amino acids, open box), Rac2 (RQQKRP, three basic amino acids, filled box), Rac3 (KKPGKK, four basic amino acids, checked box), and the nonbasic AAAAAA (hatched box) were used to make the indicated chimeras. GFP constructs having the PB region and CAAX motif (GFP-Rac(PB)) were used for confocal imaging. GST constructs having the PB region only (GST-Rac(PB)) were used for protein-lipid overlay assay. C189S and Q61L mutation show a prenylation-defective and a constitutively active form, respectively. Chimera nomenclature: functional domain: PB region: CAAX motif with the numbers designating the parental Rac isoform.

FIGURE 1.

Structure of GFP-tagged and GST-tagged Rac constructs used in these study. The PB region and CAAX motif (prenylation site) are indicated. The PB sequences for Rac1 (KKRKRK, six basic amino acids, open box), Rac2 (RQQKRP, three basic amino acids, filled box), Rac3 (KKPGKK, four basic amino acids, checked box), and the nonbasic AAAAAA (hatched box) were used to make the indicated chimeras. GFP constructs having the PB region and CAAX motif (GFP-Rac(PB)) were used for confocal imaging. GST constructs having the PB region only (GST-Rac(PB)) were used for protein-lipid overlay assay. C189S and Q61L mutation show a prenylation-defective and a constitutively active form, respectively. Chimera nomenclature: functional domain: PB region: CAAX motif with the numbers designating the parental Rac isoform.

Close modal

RT-PCR was performed as previously described (9). One microgram of total RNA was subjected to each reverse transcription (RT) reaction. For the PCR amplification, cDNA products from the RT reaction were denatured for 2 min at 94°C before 30 cycles at 94°C for 30 s, 57°C for 30 s, and 72°C for 45 s. The primers used for amplification were 5′-ATGCAGGCCATCAAGTGTGTGG-3′ and 5′-TTACAACAGCAGGCATTTTCTCTTCCTC-3′ for murine Rac1, 5′-ATGCAGGCCATCAAGTGTGTGG-3′ and 5′-CTAGAGCAGGCTGCAGGGGC-3′ for murine Rac2, and 5′-ATGCAGGCCATCAAGTGCGTG-3′ and 5′-CTAGAATACAGTGCACTTCTTGCCTG-3′ for murine Rac3. The identities of these PCR fragment were confirmed by sequencing.

Cells were lysed in homogenizing buffer (10) by sonication. The total cell lysates (100 μg/lane) were probed by Western blotting using polyclonal Ab for Rac1 (C14, 1:500; Santa Cruz Biotechnology) or polyclonal Ab for Rac2 (C11, 1:500; Santa Cruz Biotechnology).

Two-micrometer glass beads (2.5 ± 0.5 μm) were obtained from Duke Scientific. Fluorescently labeled and IgG-opsonized glass beads (BIgG) were prepared as described previously (11).

A total of 1.0 × 105 RAW cells were seeded on 35-mm glass bottom dishes (MatTek Chambers) and transfected using Superfect (Qiagen). Forty to 48 h after the transfection, the culture medium was replaced with 800 μl of HBSS++ (11), and 200 μl of HBSS++ containing targets (five per cell) were added to each plate. Images were collected using a LSM 510 invert (Carl Zeiss) confocal laser scanning fluorescence microscope with a heated stage and objective (×40 oil or ×63 oil) as described previously (12). The images were collected at 10-s intervals for 10 min.

To detect endogenous Rac, cells were fixed 3 min after adding targets using 4% paraformaldehyde in 0.1 M phosphate buffer (10). After permeabilization, cells were stained using primary Ab (mAb of Rac1, diluted 1/200; Upstate Biotechnology) for 2 h at 22°C. Primary Ab were visualized using Alexa 488-conjugated anti-mouse IgG, (Molecular Probe; 1/2000, 0.5 h at 22°C) and confocal imaging.

The accumulation of GFP-tagged Rac and various chimeras at the phagosomal cup/phagosome (p) were normalized by comparing it to the value of the surrounding cytoplasm (c) using the ratio [(pc)/c]. Results are expressed as means ± SD (∗, p < 0.05).

The following cDNA fragments, Rac1, Rac2, Rac3, Rac2-1-2, Rac3-1-3, Rac1-2-1, Rac1-3-1, and Rac1-6A-1, containing the Q61L and the C189S substitutions to avoid the effects of intrinsic GTPase activity and prenylation, respectively (13), were generated by PCR-mediated site-directed mutagenesis and cloned into the BamHI/EcoRI site of pProEx-Htb (Invitrogen Life Technologies). All the constructs were sequenced to confirm their identities. Proteins fused to (His)6 were expressed in Escherichia coli strain BL21(DE3) and purified by BD Talon (BD Biosciences). The membrane fraction of human neutrophils was prepared as described previously (14). The membranes (9.4 μg of protein/ml) were mixed with His-tagged p47phox (202 nM), His-tagged p67phox (202 nM), and His-tagged Rac proteins (0∼566 nM) preloaded with 100 μM GTPγS followed by incubation with 100 μM SDS for 2.5 min at RT in 100 mM potassium phosphate, pH 7.0, containing 75 μM cytochrome c, 10 μM FAD, 1.0 mM EGTA, 1.0 mM MgCl2, and 1.0 mM NaN3. The reaction was initiated by the addition of NADPH (1 mM) to the reaction mixture. The NADPH-dependent O2-producing activity was measured by determining the rate of O2 dismutase-inhibitable ferricytochrome c reduction at 550–540 nm with a dual wavelength spectrophotometer (Hitachi 557; Hitachi High Technologies). Results were expressed as micromoles per minute per milligram of membrane proteins (14).

Forward and reverse oligonucleotides for the PB region (six residues) of Rac isoform, Rac1(KKRKRK), Rac2(RQQKRP), Rac3(KKPGKK), and 6A(AAAAAA), were annealed and cloned into the BamHI/EcoRI site of pGEX-2T (Amersham Biosciences), and designated GST-Rac1(PB), GST-Rac2(PB), GST-Rac2(PB), and GST-6A, respectively (Fig. 1). All constructs were sequenced to confirm their identities. Protein fused to GST was expressed in E. coli strain BL21(DE3) and purified by glutathione-Sepharose-4B (Amersham Biosciences). diC16-PI(3)P, -PI(4)P, -PI(5), -PI(4,5)P2, -PI(3,4)P2, -PI(3,5)P2, and -PI(3,4,5)P3 were obtained from Echelon Biosciences, and phosphatidic acid (PA) was from Biomol. Protein-lipid overlay assay was performed using phosphatidylinositol phosphate (PIP)-strip (Echelon Biosciences) and Hybond-C extra membrane (Amersham Biosciences) spotted with assorted phospholipid (2–500 pmol) according to the reported methods (15). Briefly, phospholipid-spotted membranes were blocked in 3% fatty acid-free BSA in TBST buffer (50 mM Tris-HCl, 150 mM NaCl, and 0.1% Tween 20, pH 7.5) for 1 h at 22°C. The membranes were then incubated for 3 h at 22°C in the same solution with 20 nM (= 0.5 μg/ml) GST-tagged protein. The membranes were washed five times for 1 h (each time) in TBST buffer and then incubated for 1 h with 1/1000 dilution of anti-GST polyclonal Ab (Sigma-Aldrich). The membranes were washed as before being incubated for 1 h with 1/5000 dilution of anti-rabbit HRP conjugate (Jackson ImmunoResearch Laboratories). After washing, the bound protein was detected by ECL (Amersham Biosciences).

We examined the Rac isoforms expressed in RAW macrophages. RT-PCR revealed the expression of Rac1, Rac2, and Rac3 mRNA, with the Rac3 message less than that for Rac1 and Rac2 (Fig. 2,A, left). Rac1 and Rac2 proteins were detected using isoform-specific Abs (Fig. 2 A, right), and Rac3 protein expression could not be determined because specific anti-Rac3 Ab are not available.

FIGURE 2.

A, Expression of Rac isoforms in RAW 264.7 macrophages. Left, RAW cells express mRNA for Rac1, Rac2, and Rac3. Right, Rac1 and Rac2 protein were detected by Western blotting. M, marker; cont, no reverse transcriptase. Representative of two independent experiments. B, Different extent of accumulations of GFP-Rac isoforms (Rac1 > Rac3 > Rac2) during FcγR-mediated phagocytosis. GFP-Rac1 strongly accumulates (a, arrow), GFP-Rac2 weakly accumulates (b, arrow), GFP-Rac3 moderately accumulates at the phagosomal cup/phagosome (c, arrow). Dominant-negative form of GFP-Rac1, GFP-Rac1(T17N), does not show any accumulation (d, arrows). C, Accumulation of endogenous Rac1 during FcγR-mediated phagocytosis. Endogenous Rac1 was present in the cytoplasm and concentrated weakly (arrowhead) and strongly (arrow) at the phagosome. Right, DIC image showing the phagocytosed BIgG and the phagosome (arrow and arrowhead).

FIGURE 2.

A, Expression of Rac isoforms in RAW 264.7 macrophages. Left, RAW cells express mRNA for Rac1, Rac2, and Rac3. Right, Rac1 and Rac2 protein were detected by Western blotting. M, marker; cont, no reverse transcriptase. Representative of two independent experiments. B, Different extent of accumulations of GFP-Rac isoforms (Rac1 > Rac3 > Rac2) during FcγR-mediated phagocytosis. GFP-Rac1 strongly accumulates (a, arrow), GFP-Rac2 weakly accumulates (b, arrow), GFP-Rac3 moderately accumulates at the phagosomal cup/phagosome (c, arrow). Dominant-negative form of GFP-Rac1, GFP-Rac1(T17N), does not show any accumulation (d, arrows). C, Accumulation of endogenous Rac1 during FcγR-mediated phagocytosis. Endogenous Rac1 was present in the cytoplasm and concentrated weakly (arrowhead) and strongly (arrow) at the phagosome. Right, DIC image showing the phagocytosed BIgG and the phagosome (arrow and arrowhead).

Close modal

To verify the isoform-specific subcellular localization of Rac isoforms, GFP-Rac1, GFP-Rac2, and GFP-Rac3 were transfected into RAW cells. GFP-Rac1 was expressed in the cytoplasm and nucleus (Fig. 2,Ba). GFP-Rac2 was localized in the cytoplasm with a slight accumulation at the Golgi region (Fig. 2,Bb). This accumulation at the Golgi was confirmed by colocalization of the GFP signal with Texas Red-conjugated wheat germ agglutinin (data not shown), consistent with a previous study (16). The localization of GFP-Rac3 had features of both GFP-Rac1 and GFP-Rac2, with weak localization both at the nucleus and at the Golgi region (Fig. 2,Bc). Next we tried to detect the subcellular localization of endogenous Rac (Rac1 and Rac2) in RAW cells. mAb of Rac1 showed that endogenous Rac1 was localized in the cytoplasm and slightly in the nucleus (Fig. 2 C). Control study (no Rac1 Ab and secondary Ab) showed no detectable signals from cells (data not shown). We could not get clear information about endogenous Rac2 due to the lack of specific anti-murine Rac2 Ab good for immunohistocytochemistry.

Real-time confocal imaging of GFP conjugates was used to follow the movement of the three Rac isoforms during FcγR-mediated phagocytosis. Although all isoforms of GFP-Rac accumulated at the phagosomal cup/phagosome, the extent of the accumulation varied (i.e., Rac1 > Rac3 > Rac2; Figs. 2,B and 3,B). The subcellular localization of GFP-Rac1 and the accumulation of GFP-Rac1 during ingestion of BIgG mimicked those of the endogenous protein (Fig. 2,C), suggesting that the GFP-conjugated Racs reflect their endogenous counterparts. Next we asked whether accumulation required Rac1 to be in its active (i.e., GTP-bound) form. We tested a dominant-negative form of Rac1, Rac1(T17N), which is locked in the GDP-bound form, and found that it did not accumulate during BIgG ingestion (Fig. 2 Bd).

FIGURE 3.

The extent of accumulation of Rac during FcγR-mediated phagocytosis is determined by the PB region of the Rac isoform. A, Chimeras with the Rac1 PB region (i.e., Rac2-1-2 (a, arrow) and Rac3-1-3 (b, arrow)) exhibit the highest level of phagosomal fluorescence, while that has the Rac2 PB region (Rac1-2-1, c, arrows) exhibits the lowest concentration and that has the Rac3 PB region (Rac1-3-1, d, arrows) exhibits moderate concentration. The absence of basic residues in the PB region (Rac1-6A-1, e, arrows) abolishes Rac1 accumulation. B, Quantitation of phagosomal fluorescence with each of the Rac proteins shown in Figs. 2,B and 3 A. Data are shown as means ± SD. Rac1, n = 25; Rac2-1-2, n = 27; Rac3-1-3, n = 22; Rac2, n = 26; Rac1-2-1, n = 42; Rac3, n = 20; Rac1-3-1, n = 23; Rac1-6A-1, n = 15 from 8–15 independent experiments. ∗, p < 0.01.

FIGURE 3.

The extent of accumulation of Rac during FcγR-mediated phagocytosis is determined by the PB region of the Rac isoform. A, Chimeras with the Rac1 PB region (i.e., Rac2-1-2 (a, arrow) and Rac3-1-3 (b, arrow)) exhibit the highest level of phagosomal fluorescence, while that has the Rac2 PB region (Rac1-2-1, c, arrows) exhibits the lowest concentration and that has the Rac3 PB region (Rac1-3-1, d, arrows) exhibits moderate concentration. The absence of basic residues in the PB region (Rac1-6A-1, e, arrows) abolishes Rac1 accumulation. B, Quantitation of phagosomal fluorescence with each of the Rac proteins shown in Figs. 2,B and 3 A. Data are shown as means ± SD. Rac1, n = 25; Rac2-1-2, n = 27; Rac3-1-3, n = 22; Rac2, n = 26; Rac1-2-1, n = 42; Rac3, n = 20; Rac1-3-1, n = 23; Rac1-6A-1, n = 15 from 8–15 independent experiments. ∗, p < 0.01.

Close modal

Because each Rac isoform differs primarily in the PB region (Rac1, Rac2, and Rac3 have six, three, and four basic amino acids in their PB regions, respectively), we tested the hypothesis that the Rac isoform-specific localization/accumulation is dictated by the PB region. Rac chimeras were generated by exchanging the PB region of one Rac isoform for that of a different isoform (Fig. 1). We quantified the GFP accumulation of the Rac chimeras and compared them to the accumulation of the intact isoforms. Consistent with our hypothesis, the phagosomal accumulation of the Racchimeras reflected the identity of the PB region. That is chimeras having the Rac1 PB region (GFP-Rac2-1-2 and GFP-Rac3-1-3) showed the strongest accumulation, GFP-Rac1-3-1 and GFP-Rac3 showed moderate accumulation, and the Rac proteins containing the Rac2 PB region (GFP-Rac1-2-1 and GFP-Rac2) showed the least concentration (Fig. 3). As predicted, GFP-Rac1-6A-1, having 6 Ala and no basic residues in the PB region, showed no accumulation during BIgG ingestion (Fig. 3). Taken together, these results demonstrated that the accumulation of Rac isoform during FcγR-mediated phagocytosis requires activation and depends on the PB region, most probably on net positive charge in the PB region.

As Rac is a component of the NADPH oxidase system, which is activated during FcγR-mediated phagocytosis, we investigated the ability of Rac isoforms/chimeras to support O2 production using an amphiphile-activated cell-free assay system. Nonprenylated (C189S) and constitutively active (Q61L) recombinant Rac proteins were used to exclude effects due to prenylation or intrinsic differences in GTPase activity. Consistent with the rank order of phagosomal accumulation, Rac1 produced the maximal O2 production, with Rac3 supporting 31.8% and Rac2 producing only 12.8% of the O2 generated by Rac1; Rac1-6A-1 did not enhance O2 production over control (Fig. 4). Substitution of the Rac1 PB region for that of Rac3 (Rac3-1-3) resulted in O2 production similar to that of intact Rac1 (Fig. 4 B). The reverse chimera, i.e., Rac3 PB region in the context of Rac1 (Rac1-3-1) produced a response similar to Rac3. Although the activity for O2 production of Rac2-1-2 was not completely restored in the same level of Rac1 (68.4% of Rac1), it was markedly enhanced (5.32-fold) over Rac2. The reverse chimera (Rac1-2-1) was not completely reduced in the same level of Rac2; however, it also markedly reduced the activity for O2 production (26.3% of Rac1). These results suggested that the accumulation of the Rac isoforms during BIgG ingestion, namely that the identity of the PB region dictates the localization to the phagosome.

FIGURE 4.

The ability of Rac and Rac chimeras to stimulate O2 production in the amphiphile-activated cell-free system is determined by the identity of the PB region. A, O2 production was conducted as described in Materials and Methods using the indicated concentration of His-tagged Rac(Q61L, C189S)-proteins (0–566 nM). The ability to support O2 production paralleled the concentration of the Rac proteins to phagosomes (compare with Fig. 3 B). Rac proteins with the Rac1 PB region (Rac1, Rac2-1-2, and Rac3-1-3) show the strongest ability to produce O2, while Rac proteins with the Rac2 (Rac2 and Rac1-2-1) or Rac3 PB region (Rac3 and Rac1-3-1) show weaker ability to produce O2. B, Comparison of the ability of Rac or Rac chimeras to support O2 production, in the concentration 566 nM. Three independent experiments were performed. Data are shown as means ± SD.

FIGURE 4.

The ability of Rac and Rac chimeras to stimulate O2 production in the amphiphile-activated cell-free system is determined by the identity of the PB region. A, O2 production was conducted as described in Materials and Methods using the indicated concentration of His-tagged Rac(Q61L, C189S)-proteins (0–566 nM). The ability to support O2 production paralleled the concentration of the Rac proteins to phagosomes (compare with Fig. 3 B). Rac proteins with the Rac1 PB region (Rac1, Rac2-1-2, and Rac3-1-3) show the strongest ability to produce O2, while Rac proteins with the Rac2 (Rac2 and Rac1-2-1) or Rac3 PB region (Rac3 and Rac1-3-1) show weaker ability to produce O2. B, Comparison of the ability of Rac or Rac chimeras to support O2 production, in the concentration 566 nM. Three independent experiments were performed. Data are shown as means ± SD.

Close modal

It has been reported that Rac proteins have two membrane localization motifs, prenylation and the PB region, with the PB region dominating in vitro (17). To examine the importance of Rac prenylation in subcellular localization, prenylation-defective mutants GFP-Rac1(C189S), GFP-Rac2(C189S), and Rac1-6A-1(C189S) were used (13). GFP-Rac1(C189S) concentrated in the nucleus and GFP-Rac2(C189S) was localized in intracellular vesicles (Fig. 5,A). GFP-Rac1-6A-1(C189S) was localized in the cytoplasm and nucleus (Fig. 5,A) similar to vector controls expressing GFP only (data not shown). These results are quite different from those obtained using the intact isoforms (Fig. 2 B), suggesting that prenylation is required for normal localization of Rac in cells. As the main sequence divergence among Rac isoforms is in the PB region, these results suggested that the PB region of Rac1 and Rac2 may function as a nuclear localization and as an intracellular vesicle localization motif, respectively.

FIGURE 5.

A, Role of isoprenylation for subcellular localization of Rac isoform in RAW cells. Nonprenylated Rac1, GFP-Rac1(C189S), is primarily located in the nucleus; in contrast, nonprenylated Rac2, GFP-Rac2(C189S), is primarily located in the intracellular vesicles. Nonprenylated Rac1-6A-1, GFP-Rac1-6A-1(C189S), is localized at the cytoplasm and nucleus the same as GFP only was. Bar, 3 μm. B, Localization and accumulation of the PB region of Rac isoforms during FcγR-mediated phagocytosis. a, GFP-Rac1(PB) is primarily localized at the plasma membrane and accumulates at the phagosomal cup (arrows). b, GFP-Rac2(PB) is weakly localized at the plasma membrane with the localization at the Golgi region and intracellular vesicles, and accumulates at the phagosome (arrow). Note a linear structure having GFP-Rac2(PB) (arrowheads) between the phagosome and the Golgi region. (Inset), An enlargement of the linear structure. c, GFP-Rac3(PB) is localized at the plasma membrane with localization at the Golgi region, and accumulates at the phagosomal cup/phagosome (arrows). d, GFP-6A shows a reticular localization in the cytoplasm with localization at the Golgi region. Upper panel, Note a linear structure containing GFP-6A (big arrowheads) that connects the phagosomal cup (arrow) and the Golgi region. Lower panel, Small arrowheads mark an ingested BIgG. Inset, An enlargement of the linear structure (medium-sized arrowheads) toward the phagosomal cup (small arrowheads). See the detail of the GFP-6A subcellular localization in Fig. 7 Bb.

FIGURE 5.

A, Role of isoprenylation for subcellular localization of Rac isoform in RAW cells. Nonprenylated Rac1, GFP-Rac1(C189S), is primarily located in the nucleus; in contrast, nonprenylated Rac2, GFP-Rac2(C189S), is primarily located in the intracellular vesicles. Nonprenylated Rac1-6A-1, GFP-Rac1-6A-1(C189S), is localized at the cytoplasm and nucleus the same as GFP only was. Bar, 3 μm. B, Localization and accumulation of the PB region of Rac isoforms during FcγR-mediated phagocytosis. a, GFP-Rac1(PB) is primarily localized at the plasma membrane and accumulates at the phagosomal cup (arrows). b, GFP-Rac2(PB) is weakly localized at the plasma membrane with the localization at the Golgi region and intracellular vesicles, and accumulates at the phagosome (arrow). Note a linear structure having GFP-Rac2(PB) (arrowheads) between the phagosome and the Golgi region. (Inset), An enlargement of the linear structure. c, GFP-Rac3(PB) is localized at the plasma membrane with localization at the Golgi region, and accumulates at the phagosomal cup/phagosome (arrows). d, GFP-6A shows a reticular localization in the cytoplasm with localization at the Golgi region. Upper panel, Note a linear structure containing GFP-6A (big arrowheads) that connects the phagosomal cup (arrow) and the Golgi region. Lower panel, Small arrowheads mark an ingested BIgG. Inset, An enlargement of the linear structure (medium-sized arrowheads) toward the phagosomal cup (small arrowheads). See the detail of the GFP-6A subcellular localization in Fig. 7 Bb.

Close modal

Taken together, our results suggest that both prenylation and the PB region are important for Rac localization. To determine which regions control Rac accumulation at the phagosome, we used the GFP-tagged PB region of Rac isoform with prenylation signal: GFP-Rac1(PB), GFP-Rac2(PB), GFP-Rac3(PB), and GFP-6A (Fig. 1). In resting cells, GFP-Rac1(PB) was predominantly localized at the plasma membrane (Fig. 5,Ba), and GFP-Rac3(PB) was expressed in the cytosol with slight accumulations at the plasma membrane and Golgi region (Fig. 5,Bc). In sharp contrast, a small amount of GFP-Rac2(PB) was localized at the plasma membrane, with localization at the Golgi region and intracellular vesicles (Fig. 5,Bb). These vesicles did not colocalize with either early endosome Ag-1 (EEA-1) or a marker for the lysosome (data not shown). GFP-6A stained reticular structures in RAW cells with a slight accumulation at the nuclear membrane and strong localization at the Golgi region (Figs. 5,Bd and 7 Bb). This result was consistent with the previous study that the prenylation targets proteins to endomembranes (endoplasmic reticulum (ER), Golgi complex, and nuclear membrane) (18). Because GFP-Rac(PB) does not form a complex with RhoGDI, we propose that the prenylated PB region may accumulate at the site where full-length Rac protein is localized in the active form. Furthermore, the net positive charge in the PB region may determine localization of the active form; Rac1 at the plasma membrane and Rac2 mainly at the endomembranes and partly at the plasma membrane.

FIGURE 7.

A, Accumulation of GFP-Rac1(Q61L) at the phagosome during FcγR-mediated phagocytosis. GFP-Rac1(Q61L) is primarily localized at the plasma membrane (arrow) and intracellular vesicles (double arrowheads). GFP-Rac1(Q61L) shows retained localization at the phagosome (double arrows) after engulfment of BIgG. B, Localization of Rac2(Q61L), GFP-6A, and GFP-Rac2-6A-2(Q61L) in RAW cells. GFP-Rac2(Q61L) (a) shows localization at the endomembranes both in the cytoplasm and in Golgi region (∗) and is weakly localized at the plasma membrane (arrows). Inset, An enlargement of the endomembrane localization of GFP-Rac2(Q61L) in the cytoplasm. GFP-6A (b) and GFP-Rac2-6A-2(Q61L) (c) show Golgi (∗), reticular, and nucleus membrane (arrowhead) staining and no plasma membrane localization. Bars, 3 μm. C, Changes in the distribution of GFP-Rac2(Q61L) using a linear structure during FcγR-mediated phagocytosis. (Before stimulus) In resting cells, GFP-Rac2(Q61L) is predominantly seen in endomembranes in the Golgi region (∗); little expression is apparent at the plasma membrane (arrows) and intracellular vesicles (double arrowheads). (After stimulus) Note a linear structure containing GFP-Rac2(Q61L) (arrowheads) between the phagosome (double arrows) and the endomembranes at Golgi region (∗). This structure apparently supplies Rac2(Q61L) to the phagosome. This is more apparent in Supplemental Movie 1. D, Delivery of GFP-Rac2-6A-2(Q61L) to the phagosome via a linear structure during FcγR-mediated phagocytosis. (Before stimulus) GFP-Rac2-6A-2(Q61L) showed a reticular staining pattern with Golgi localization (∗). (After stimulus) During BIgG ingestion, GFP-Rac2-6A-2(Q61L) accumulates, albeit weakly, at the phagosome (double arrows). As with Rac2(Q61L), the accumulation at the phagosome coincides with the appearance of a linear GFP-Rac2-6A-2(Q61L)-containing structure (arrowheads) between the phagosome (double arrows) and the Golgi region (∗).

FIGURE 7.

A, Accumulation of GFP-Rac1(Q61L) at the phagosome during FcγR-mediated phagocytosis. GFP-Rac1(Q61L) is primarily localized at the plasma membrane (arrow) and intracellular vesicles (double arrowheads). GFP-Rac1(Q61L) shows retained localization at the phagosome (double arrows) after engulfment of BIgG. B, Localization of Rac2(Q61L), GFP-6A, and GFP-Rac2-6A-2(Q61L) in RAW cells. GFP-Rac2(Q61L) (a) shows localization at the endomembranes both in the cytoplasm and in Golgi region (∗) and is weakly localized at the plasma membrane (arrows). Inset, An enlargement of the endomembrane localization of GFP-Rac2(Q61L) in the cytoplasm. GFP-6A (b) and GFP-Rac2-6A-2(Q61L) (c) show Golgi (∗), reticular, and nucleus membrane (arrowhead) staining and no plasma membrane localization. Bars, 3 μm. C, Changes in the distribution of GFP-Rac2(Q61L) using a linear structure during FcγR-mediated phagocytosis. (Before stimulus) In resting cells, GFP-Rac2(Q61L) is predominantly seen in endomembranes in the Golgi region (∗); little expression is apparent at the plasma membrane (arrows) and intracellular vesicles (double arrowheads). (After stimulus) Note a linear structure containing GFP-Rac2(Q61L) (arrowheads) between the phagosome (double arrows) and the endomembranes at Golgi region (∗). This structure apparently supplies Rac2(Q61L) to the phagosome. This is more apparent in Supplemental Movie 1. D, Delivery of GFP-Rac2-6A-2(Q61L) to the phagosome via a linear structure during FcγR-mediated phagocytosis. (Before stimulus) GFP-Rac2-6A-2(Q61L) showed a reticular staining pattern with Golgi localization (∗). (After stimulus) During BIgG ingestion, GFP-Rac2-6A-2(Q61L) accumulates, albeit weakly, at the phagosome (double arrows). As with Rac2(Q61L), the accumulation at the phagosome coincides with the appearance of a linear GFP-Rac2-6A-2(Q61L)-containing structure (arrowheads) between the phagosome (double arrows) and the Golgi region (∗).

Close modal

During BIgG ingestion, strong accumulation of GFP-Rac1(PB), moderate accumulation of GFP-Rac3(PB), weak accumulations of GFP-Rac2(PB), and faint accumulation of GFP-6A were observed at the phagosomal cup/phagosome (Fig. 5,B). Interestingly, a linear structure connecting the phagosome and endomembranes containing GFP-Rac2(PB) or GFP-6A was observed during BIgG ingestion (Fig. 5 B, b and d). These results suggested that the accumulation of Rac isoform during BIgG ingestion depend on the PB region of each isoform.

The presence of polyphosphoinositides in the phagosome and that the phosphoinositide composition of phagosome changes with time has been well documented (19, 20, 21, 22, 23). The positively charged PB region of Rac may thus interact with the negatively charged phosphoinositide species, providing a potential mechanism for their selective accumulation. If true, we would predict that the phosphoinositide binding would correlate with the basic charge on the PB region. Thus, to further define the mechanism of accumulation of the PB region, the ability of the PB region to bind phospholipids was examined by a protein-lipid overlay assay. Using PIP-strip (Echelon Biosciences), GST-Rac1(PB) showed binding to PA and phosphoinositides (PIP, and PIP3), but not to phos-phatidylinositol, phosphatidylethanolamine, phosphatidylcholine, phosphatidylserine, lysophosphatidic acid, lysophosphocholine, or sphingosine-1-phosphate (data not shown). To confirm these results, the concentration-dependent interaction between GST-Rac(PB) and PIP, PIP2, PIP3, and PA was examined. GST-Rac1(PB) strongly interacted with PI(3)P, PI(4), PI(5)P, PI(3,4,5)P3, and PA, but not with PI(4,5)P2: GST-Rac3(PB) had weak interactions with these phospholipids (Fig. 6). Although GST-Rac2(PB) showed interactions with PA and PI(3,4,5)P3, these were not significant when compared with the GST control (Fig. 6). These GST-Rac2(PB) data further support the results in Fig. 5 Bb demonstrating that intracellular vesicles observed upon the expression of GFP-Rac2(PB) did not colocalize with EEA-1, a marker of PI(3)P. The binding capacity of our PIP(4,5)2 was confirmed using the PH domain of PLCδ (data not shown). Taken together, these results suggest that the PB regions of both Rac1 and Rac3 have the ability to interact with specific phospholipids, which may explain their selective association with the phagosome.

FIGURE 6.

Binding of the PB region of Rac isoform to polyphosphoinositides. The GST-tagged Rac PB region (without CAAX motif) was used in lipid overlay assays to identify PB region-binding lipids. GST-Rac1(PB) binds PI(3)P, PI(4)P, PI(5)P, PI(3,4,5)P3, and PA with weaker binding to PI(3,4)P2 and PI(3,5)P2; GST-Rac1(PB) does not associate with PI(4,5)P2. GST-Rac3(PB) shows weak binding to PI(3)P, PI(4)P, PI(5)P, PI(3,4,5)P3. In contrast, GST-Rac2(PB) shows slight binding to PI(3,4,5)P3 only. GST-Rac1(PB) but neither GST-Rac2(PB) nor GST-Rac3(PB) shows the significant binding to PA compared with the GST control. Representative of at least three independent experiments.

FIGURE 6.

Binding of the PB region of Rac isoform to polyphosphoinositides. The GST-tagged Rac PB region (without CAAX motif) was used in lipid overlay assays to identify PB region-binding lipids. GST-Rac1(PB) binds PI(3)P, PI(4)P, PI(5)P, PI(3,4,5)P3, and PA with weaker binding to PI(3,4)P2 and PI(3,5)P2; GST-Rac1(PB) does not associate with PI(4,5)P2. GST-Rac3(PB) shows weak binding to PI(3)P, PI(4)P, PI(5)P, PI(3,4,5)P3. In contrast, GST-Rac2(PB) shows slight binding to PI(3,4,5)P3 only. GST-Rac1(PB) but neither GST-Rac2(PB) nor GST-Rac3(PB) shows the significant binding to PA compared with the GST control. Representative of at least three independent experiments.

Close modal

Our results are consistent with a model in which the accumulation of Rac1 and Rac3 during FcγR-mediated phagocytosis depends on the net positive charge and specific phospholipid binding characteristics of the PB region. Rac2 appears to use an alternative pathway. We compared the localization and movement of Rac1 and Rac2 during phagocytosis using constitutively active mutants of Rac1 and Rac2, GFP-Rac1(Q61L) and GFP-Rac2(Q61L). In resting cells, GFP-Rac1(Q61L) resided predominantly at the plasma membrane and the membrane of a vesicle, which may have originated from endocytosis (Fig. 7,A). In contrast, GFP-Rac2(Q61L) localized weakly to the plasma membrane and membrane of vesicles (Fig. 7, Ba and C). Differences in the localization between GFP-Rac2(Q61L) and GFP-Rac1(Q61L) included the expression of GFP-Rac2(Q61L) at endomembranes and less localization at the plasma membrane compared with GFP-Rac1(Q61L) (Fig. 7, Ba and C). The endomembranes labeled with GFP-Rac2(Q61L) were predominantly at the Golgi and perinuclear regions, an area that colocalized with ER staining (data not shown). GFP-6A and GFP-Rac2-6A-2(Q61L) showed reticular and nuclear membrane localization with strong localization at the Golgi complex (Fig. 7,B, b and c). However, the plasma membrane localization was not observed in GFP-6A or GFP-Rac2-6A-2(Q61L), which is an active GTP-bound form of Rac2 (Fig. 7 B, b and c). Compared with GFP-6A and GFP-Rac2-6A-2(Q61L), GFP-Rac2(Q61L)-transfected cells had less Golgi and reticular labeling and more fine reticular staining in the cytoplasm. When Rac1(G12V) and Rac2(G12V) were used instead of Rac1(Q61L) and Rac2(Q61L), the same results were obtained (data not shown). These results suggested that Rac1 primarily functions at the plasma membrane; in contrast, Rac2 primarily functions at the endomembranes.

To further test this model, the accumulation of GFP-Rac1(Q61L) and GFP-Rac2(Q61L) at the phagosome during of BIgG ingestion was examined. The extent of the accumulation of GFP-Rac1(Q61L) at the phagosome was greater than that of GFP-Rac2(Q61L) (data not shown), and retained longer after complete engulfment of BIgG than GFP-Rac1 (Fig. 7,A). An intriguing difference was observed between the behavior of GFP-Rac1(Q61L) and GFP-Rac2(Q61L) during BIgG ingestion. A linear structure that delivered GFP-Rac2(Q61L) from endomembranes at the Golgi region to the phagosome (Fig. 7,C and Supplemental Movie 1)4 was seen in GFP-Rac2(Q61L), but not GFP-Rac1(Q61L) expressing cells. These results suggest that the accumulations of Rac1 and Rac2 are regulated differently: Rac1 accumulates directly at the phagosome using its highly positively charged PB region; Rac2 also accumulates (albeit weakly) using its weakly charged PB region. Additionally, Rac2 targets to endomembranes that subsequently fuse with the phagosome. Although neither GFP-Rac1-6A-1(Q61L) (data not shown) nor GFP-Rac2-6A-2(Q61L) (Fig. 7,D) showed the localization at the plasma membrane, both of them weakly accumulated at the phagosome with the linear structure during BIgG ingestion (Fig. 7 D). These results suggested that the accumulation at the phagosome by a Rac1-mediated mechanism requires the highly positive charge in the PB region and, in contrast, that Rac2 uses a different mechanism that does not require a positively charged PB region.

It is known that both Rac1 and Rac2 are capable of stimulating O2 production in cell-free systems (24, 25). It has been shown that Rac2 and Rac1 are main components of the NADPH oxidase complex in neutrophils (6) and in other cells (8, 26), respectively. As the Rac proteins are cytosolic and assemble into the active, membrane-localized NADPH oxidase complex during FcγR-mediated phagocytosis, we sought to elucidate the mechanism by which Rac localizes to membranes during phagocytosis. Using GFP-conjugated Rac1, 2, and 3, we demonstrated that all three accumulated at the phagosomal cup/phagosome during FcγR-mediated phagocytosis in RAW 264.7 macrophage cells. Furthermore, all isoforms stimulated O2 production in a cell-free system albeit to different levels (Rac1 ≫ Rac3 > Rac2). These results suggest that all Rac isoforms are capable of stimulating O2 production in intact cells if they contain a membrane-targeting domain that would localize them to the NADPH oxidase-containing membranes.

Joseph et al. (27) reported that the C-terminal domain of Rac1, but not Rac2, is a plasma membrane targeting motif that is not sequence-specific, but related to the presence of a PB motif. Using Rac-chimeric proteins, we showed that the differential accumulation among the Rac isoforms at the phagosome is indeed dependent on the PB regions, specifically the number of basic residues in that region. Rac1, with six basic residues, is most strongly localized to the phagosome followed by Rac3 (four basic amino acids) then Rac2 (three basic residues) (Figs. 2,B and 3,B). The PB region is responsible for membrane targeting as substituting the PB of Rac1 with 6 Ala (Rac1-6A-1) abrogates its accumulation at the phagosome and ability to support O2 production. Conversely, substitution of the PB region of Rac1 with the PB 6 Lys (GFP-Rac1-6K-1) did not alter its accumulation at the phagosome compared with GFP-Rac1 (data not shown). These results indicate that it is the overall charge of the PB region rather than an exact sequence that is critical for membrane targeting. That membrane association of Rac is critical for O2 production was demonstrated in our cell-free system in which the ability to localize to membranes correlated with the stimulation of O2 production (compare Figs. 3,B and 4 B). That is, the accumulation of GFP-Rac1 at the phagosomal cup/phagosome during FcγR-mediated phagocytosis was much stronger than that of GFP-Rac2. Likewise, the ability of Rac1 to stimulate O2 production was much stronger than that of Rac2. Our results are consistent with those of Kreck et al. (28) who reported that nonprenylated Rac1 had a significantly stronger association with the oxidase complex (15- to ∼20-fold) than nonprenylated Rac2 in the amphiphile-activated cell-free O2 production assay. Taken together, the differences between Rac1 and Rac2 with respect to their accumulation during FcγR-mediated phagocytosis in RAW cells and their ability to activate O2 production are likely derived from the positive charge on their PB regions, which facilitates membrane targeting.

As the PB region of Rac targets the protein to the phagosome, it must bind to specific molecules that are produced at the phagosome during BIgG ingestion. The negatively charged polyphosphoinositides are likely candidates for interacting with the positively charged PB region. Indeed, the PB region of Rac1, but not that of Rac2, strongly binds to PI(3)P, PI(4), PI(5)P, PI(3,4,5)P3, and PA. The accumulation of GFP-Rac1–6K-1 and binding of GST-6K to phospholipids showed the same pattern as those of GFP-Rac1 and GST-Rac1(PB), respectively (data not shown). These results are consistent with a model in which the PB domain of Rac binds to negatively charged phospholipids in the membrane, an ionic interaction that is charge, but not residue, dependent. It was reported that PA may function to localize p47phox to the membrane (22, 29). It is also reported that PI(3)P and PI(3,4,5)P3 are generated at the phagosome during phagocytosis (19, 20). We have also confirmed that PA, PI(3)P, and PI(3,4,5)P3 are produced at the phagosome during FcγR-mediated phagocytosis in RAW cells using the following indicators; GFP-PLD2, GFP-p40phox(PX), and GFP-Akt(PH), respectively (our unpublished data). GFP-Akt is reported to concentrate at the forming phagosome, before activation of Rac1 and Rac2 (30). Recently, generation of PI(5)P by dephosphorylation of PI(4,5)P2 during phagocytosis of Salmonella typhimurium was reported (23). Although PI(4)P is present in the plasma membrane in the resting state (31), the formation of these negatively charged phospholipids (PA and PI3K products) during phagocytosis may provide docking sites for the PB region that would target Rac to the forming phagosome. There is a discrepancy about PI(3)P between our results and a previous report that showed a Rac1 binding to PI(3,4)P2 as well as PI(3,4,5)P3, but not to PI(3)P (32). One possible explanation is that they used PI(3)P from a mixture of crude brain phosphoinositides; in contrast, we used synthesized diC16-PI(3)P. In most systems, Rac1 activation is dependent on PI3K (30). There are reports that PI3K inhibitor inhibits phagosome closure, but not formation of phagosomal cup (33). More recently, the same group reported that both Rac1 and Rac2 accumulate and are activated during the formation of the phagosome in RAW cells: Rac1 activation was biphasic (during pseudopod extension but not formation of the phagosomal cup, and during the phagosomal closure), and Rac2 activation is during phagosomal closure. They also suggested that the first phase of Rac1 activation was associated with pseudopod extension, and the late phase associated with NADPH oxidase (30).

For Ras family proteins, it has been reported that prenylation of the CAAX motif targets proteins to endomembranes (ER and Golgi complex), and that a second signal is required for the plasma membrane localization for Ras proteins (18). In K-Ras4B, that second signal is a PB region adjacent to the CAAX motif. To function as a plasma membrane targeting motif, the PB region of K-Ras4B requires a net positive charge of four or more (34). Using these criteria, the PB region of Rac1 and Rac3, but not Rac2, contains plasma membrane targeting signals. Michaelson et al. (16) reported that GFP-Rac2 was weakly localized at the plasma membrane with the major concentration in the Golgi complex and the ER. Although localization of GFP-Rac2 at the plasma membrane was not observed in the present study, the results from GFP-Rac2(PB) vs GFP-6A and GFP-Rac2(Q61L) vs GFP-Rac2-6A-2(Q61L) indicated that Rac2 may contain a weak membrane targeting signal in the PB region in activated state. GFP-Rac2(Q61L) showed less localization at the Golgi complex and more diffuse localization at the endomembranes compared with GFP-Rac2-6A-2(Q61L). The similar pattern of localization of GFP-Rac2(PB) and GFP-Rac2(C189S) (which lacks the prenylation sequence and thus contains only the PB region) to intracellular vesicles is consistent with a model in which the PB region of Rac2 targets the protein to intracellular vesicles. Taken together, our results suggest that prenylation alone (without positive charge in the PB region) promotes Rac binding to the Golgi complex and ER; increasing the net positive charge in the PB region causes a shift from the Golgi to intracellular vesicles; four or more basic amino acids in the PB region promotes Rac association with the plasma membrane.

In resting cells, Rac is complexed with RhoGDI in the cytoplasm. Rac becomes activated upon cell stimulation, disso-ciating from RhoGDI and translocating to the membrane (25). During phagocytosis of BIgG, GFP-Rac2(PB), GFP-6A, GFP-Rac2(Q61L), and GFP-Rac2-6A-2(Q61L) (which are localized at the membranes due to prenylation signal and/or predominantly active form that cannot bind RhoGDI) accumulated at the phagosome through a linear structure between the phagosome and endomembranes. Localization of GFP-Rac2(PB), GFP-Rac2(Q61L), and GFP-Rac2-6A-2(Q61) at endomembranes suggests that Rac2 translocates to endomembranes when it is activated. Moreover, because the linear structure was observed even in GFP-6A- and GFP-Rac2-6A-2(Q61L) (which is prenylated and localized at endomembranes, but does not have positive charge)-expressing cells during BIgG ingestion, we would suggest that the linear structure supplying Rac2 to the phagosome does not require the positive charge in the PB region. It has widely been held that the phagosomal membrane is derived both by invagination of the plasma membrane and by stimulus-dependent exocytosis into the forming phagosome (35). The sources of the exocytotic membrane may include the ER, intracellular granules, or early endosome (35, 36). Gagnon et al. (37) reported that ER-mediated phagocytosis is a general mechanism in macrophages. Because GFP-Rac2(Q61L) was localized at endomembranes with weak localization at the plasma membrane, it is suggested that Rac2 accumulates at the phagosome using two different mechanisms: 1) directly using its weakly positive charged PB region as Rac1 does, and 2) indirectly, initially accumulating at the endomembranes, which then fuse into the phagosome.

Because our cell-free assay system measuring O2 production used nonprenylated Rac proteins, the results obtained reflect membrane targeting by the PB region alone. Heyworth et al. (38) and Kreck et al. (28) reported that prenylated Rac1 and Rac2 had essentially the same activity in cell-free O2 production systems. These results strongly suggested that prenylation plays a crucial role in accumulation of Rac2. In the present study, we demonstrated at the cellular level that Rac2 has another accumulation mechanism that is different from that of Rac1, and this mechanism is not dependent on the positive charge of the PB region, but depends on prenylation. Although ER is not the major source of membrane in neutrophils (37), neutrophils are enriched in a variety of intracellular granules that rapidly fuse with the phagosome during phagocytosis (39). Rac2 has been suggested to have a role as an exocytotic GTPase (40). It was recently reported that Rac2−/− neutrophils are deficient in exocytosis of primary granules in response to fMLP stimulation (41). These reports are particularly intriguing because these results support our idea that activated Rac2 translocates to endomembranes and promotes their exocytosis, delivering activated Rac2 to the phagosome. This model is novel, intriguing, and will be pursued.

Zhao et al. (8) reported that human monocytes use Rac1, but not Rac2, for O2 production stimulated by opsonized zymosan, PMA, and fMLP. More recently, it was reported that macrophages from Rac2−/− mice had decreased O2 production stimulated by PMA (∼30∼50% of control) and IgG-opsonized SRBC (∼50% of control), but not opsonized zymosan (42). However, Rac2−/− macrophages showed the proportional inhibition of FcγR-mediated phagocytosis (∼50%), but not phagocytosis of opsonized zymosan (42). It should be remembered that O2 production occurs at the phagosomal cup and phagosome in cooperation with phagocytosis (12). Although Rac1 is the predominant isoform of Rac in monocytes/macrophages (8, 42), Rac2 may play a role for O2 production in macrophages. Roberts et al. (7) suggested that unactivated-bone marrow neutrophils from Rac2−/− mice clearly impaired PMA-induced O2 production, but PMA with priming by TNF-α ameliorated by 61% of wild-type cells. In murine neutrophils, Rac2, but not Rac1, deficiency reduces fMLP-induced O2 production ∼60%, and deficiency of both Rac1 and Rac2 reduces O2 production below that seen in the Rac2−/− cells (43). During preparation of this study, it was reported that the PB region of Rac2 is critical for O2 production in neutrophils (44, 45). Taken together, these reports and our present study in which all Rac isoforms accumulated at the phagosomal cup/phagosome during FcγR-mediated phagocytosis in RAW cells indicate that Rac2 is absolutely required for O2 production in neutrophils, but in some circumstances, Rac1 may compensate for Rac2; Rac1 predominates in monocytes/macrophages, but all Rac isoforms may function in monocytes/macrophages during FcγR-mediated phagocytosis.

In conclusion, we showed that although both Rac1 and Rac2 accumulate during FcγR-mediated phagocytosis, their targeting mechanisms at the phagosome differ. The selective use of particular Rac isoforms by different cell or different stimulation may reflect differences in their expression and/or differences in their targeting mechanisms. Because Rac2 apparently uses the endomembrane compartment for the accumulation at the phagosome, it may be the predominant and relevant isoform for phagocytosis, particularly in neutrophils. In contrast, Rac1 may be used in other phagocytes and nonphagocytic cells having a capacity for O2 production. Indeed, Nox1, a homologue of gp91phox (Nox2) in colon epithelial cells, required Rac1 for their activation (46, 47).

The authors have no financial conflict of interest.

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 from the 21st Century Center of Excellence Program of the Ministry of Education, Culture, Sports, Science and Technology of Japan; from Core Research for Evolutional Science and Technology; and the Ministry of Education, Culture, Sports, Science and Technology in Japan; by a Grant-in-Aid for Scientific Research on Priority Areas from Ministry of Education, Culture, Sports, Science and Technology in Japan; and from the Sankyo Foundation of Life Science.

3

Abbreviations used in this paper: O2, superoxide; PB, polybasic; BIgG, IgG-opsonized glass bead; PIP, phosphatidylinositol phosphate; PA, phosphatidic acid; RT, reverse transcription; ER, endoplasmic reticulum.

4

The online version of this article contains supplemental material.

1
Glogauer, M., C. C. Marchal, F. Zhu, A. Worku, B. E. Clausen, I. Foerster, P. Marks, G. P. Downey, M. Dinauer, D. J. Kwiatkowski.
2003
. Rac1 deletion in mouse neutrophils has selective effects on neutrophil functions.
J. Immunol.
170
:
5652
-5657.
2
Li, S., A. Yamauchi, C. C. Marchal, J. K. Molitoris, L. A. Quilliam, M. C. Dinauer.
2002
. Chemoattractant-stimulated Rac activation in wild-type and Rac2-deficient murine neutrophils: preferential activation of Rac2 and Rac2 gene dosage effect on neutrophil functions.
J. Immunol.
169
:
5043
-5051.
3
Babior, B. M..
1999
. NADPH oxidase: an update.
Blood
93
:
1464
-1476.
4
Abo, A., E. Pick, A. Hall, N. Totty, C. G. Teahan, A. W. Segal.
1991
. Activation of the NADPH oxidase involves the small GTP-binding protein p21rac1.
Nature
353
:
668
-670.
5
Knaus, U. G., P. G. Heyworth, T. Evans, J. T. Curnutte, G. M. Bokoch.
1991
. Regulation of phagocyte oxygen radical production by the GTP-binding protein Rac 2.
Science
254
:
1512
-1515.
6
Kim, C., M. C. Dinauer.
2001
. Rac2 is an essential regulator of neutrophil nicotinamide adenine dinucleotide phosphate oxidase activation in response to specific signaling pathways.
J. Immunol.
166
:
1223
-1232.
7
Roberts, A. W., C. Kim, L. Zhen, J. B. Lowe, R. Kapur, B. Petryniak, A. Spaetti, J. D. Pollock, J. B. Borneo, G. B. Bradford, et al
1999
. Deficiency of the hematopoietic cell-specific Rho family GTPase Rac2 is characterized by abnormalities in neutrophil function and host defense.
Immunity
10
:
183
-196.
8
Zhao, X., K. A. Carnevale, M. K. Cathcart.
2003
. Human monocytes use Rac1, not Rac2, in the NADPH oxidase complex.
J. Biol. Chem.
278
:
40788
-40792.
9
Ueyama, T., Y. Ren, S. Ohmori, K. Sakai, N. Tamaki, N. Saito.
2000
. cDNA cloning of an alternative splicing variant of protein kinase C δ (PKC δIII), a new truncated form of PKCδ, in rats.
Biochem. Biophys. Res. Commun.
269
:
557
-563.
10
Ueyama, T., Y. Ren, N. Sakai, M. Takahashi, Y. Ono, T. Kondoh, N. Tamaki, N. Saito.
2001
. Generation of a constitutively active fragment of PKN in microglia/macrophages after middle cerebral artery occlusion in rats.
J. Neurochem.
79
:
903
-913.
11
Larsen, E. C., T. Ueyama, P. M. Brannock, Y. Shirai, N. Saito, C. Larsson, D. Loegering, P. B. Weber, M. R. Lennartz.
2002
. A role for PKC-ε in FcγR-mediated phagocytosis by RAW 264.7 cells.
J. Cell Biol.
159
:
939
-944.
12
Ueyama, T., M. R. Lennartz, Y. Noda, T. Kobayashi, Y. Shirai, K. Rikitake, T. Yamasaki, S. Hayashi, N. Sakai, H. Seguchi, et al
2004
. Superoxide production at phagosomal cup/phagosome through βI protein kinase C during FcγR-mediated phagocytosis in microglia.
J. Immunol.
173
:
4582
-4589.
13
Kreck, M. L., D. J. Uhlinger, S. R. Tyagi, K. L. Inge, J. D. Lambeth.
1994
. Participation of the small molecular weight GTP-binding protein Rac1 in cell-free activation and assembly of the respiratory burst oxidase. Inhibition by a carboxyl-terminal Rac peptide.
J. Biol. Chem.
269
:
4161
-4168.
14
Sumimoto, H., K. Hata, K. Mizuki, T. Ito, Y. Kage, Y. Sakaki, Y. Fukumaki, M. Nakamura, K. Takeshige.
1996
. Assembly and activation of the phagocyte NADPH oxidase. Specific interaction of the N-terminal Src homology 3 domain of p47phox with p22phox is required for activation of the NADPH oxidase.
J. Biol. Chem.
271
:
22152
-22158.
15
Dowler, S., G. Kular, D. R. Alessi.
2002
. Protein lipid overlay assay.
Sci. STKE
129
:
PL6
16
Michaelson, D., J. Silletti, G. Murphy, P. D’Eustachio, M. Rush, M. R. Philips.
2001
. Differential localization of Rho GTPases in live cells: regulation by hypervariable regions and RhoGDI binding.
J. Cell Biol.
152
:
111
-126.
17
Gorzalczany, Y., N. Alloul, N. Sigal, C. Weinbaum, E. Pick.
2002
. A prenylated p67phox-Rac1 chimera elicits NADPH-dependent superoxide production by phagocyte membranes in the absence of an activator and of p47phox: conversion of a pagan NADPH oxidase to monotheism.
J. Biol. Chem.
277
:
18605
-18610.
18
Choy, E., V. K. Chiu, J. Silletti, M. Feoktistov, T. Morimoto, D. Michaelson, I. E. Ivanov, M. R. Philips.
1999
. Endomembrane trafficking of ras: the CAAX motif targets proteins to the ER and Golgi.
Cell
98
:
69
-80.
19
Ellson, C. D., K. E. Anderson, G. Morgan, E. R. Chilvers, P. Lipp, L. R. Stephens, P. T. Hawkins.
2001
. Phosphatidylinositol 3-phosphate is generated in phagosomal membranes.
Curr. Biol.
11
:
1631
-1635.
20
Gillooly, D. J., A. Simonsen, H. Stenmark.
2001
. Phosphoinositides and phagocytosis.
J. Cell Biol.
155
:
15
-17.
21
Marshall, J. G., J. W. Booth, V. Stambolic, T. Mak, T. Balla, A. D. Schreiber, T. Meyer, S. Grinstein.
2001
. Restricted accumulation of phosphatidylinositol 3-kinase products in a plasmalemmal subdomain during Fcγ receptor-mediated phagocytosis.
J. Cell Biol.
153
:
1369
-1380.
22
Karathanassis, D., R. V. Stahelin, J. Bravo, O. Perisic, C. M. Pacold, W. Cho, 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.
23
Terebiznik, M. R., O. V. Vieira, S. L. Marcus, A. Slade, C. M. Yip, W. S. Trimble, T. Meyer, B. B. Finlay, S. Grinstein.
2002
. Elimination of host cell PtdIns(4,5)P2 by bacterial SigD promotes membrane fission during invasion by Salmonella.
Nat. Cell Biol.
4
:
766
-773.
24
Heyworth, P. G., U. G. Knaus, X. Xu, D. J. Uhlinger, L. Conroy, G. M. Bokoch, J. T. Curnutte.
1993
. Requirement for posttranslational processing of Rac GTP-binding proteins for activation of human neutrophil NADPH oxidase.
Mol. Biol. Cell
4
:
261
-269.
25
Dinauer, M. C..
2003
. Regulation of neutrophil function by Rac GTPases.
Curr. Opin. Hematol.
10
:
8
-15.
26
Kim, K. S., K. Takeda, R. Sethi, J. B. Pracyk, K. Tanaka, Y. F. Zhou, Z. X. Yu, V. J. Ferrans, J. T. Bruder, I. Kovesdi, et al
1998
. Protection from reoxygenation injury by inhibition of rac1.
J. Clin. Invest.
101
:
1821
-1826.
27
Joseph, G., Y. Gorzalczany, V. Koshkin, E. Pick.
1994
. Inhibition of NADPH oxidase activation by synthetic peptides mapping within the carboxyl-terminal domain of small GTP-binding proteins: lack of amino acid sequence specificity and importance of polybasic motif.
J. Biol. Chem.
269
:
29024
-29031.
28
Kreck, M. L., J. L. Freeman, A. Abo, J. D. Lambeth.
1996
. Membrane association of Rac is required for high activity of the respiratory burst oxidase.
Biochemistry
35
:
15683
-15692.
29
Iyer, S. S., J. A. Barton, S. Bourgoin, D. J. Kusner.
2004
. Phospholipases D1 and D2 coordinately regulate macrophage phagocytosis.
J. Immunol.
173
:
2615
-2623.
30
Hoppe, A. D., J. A. Swanson.
2004
. Cdc42, Rac1, and Rac2 display distinct patterns of activation during phagocytosis.
Mol. Biol. Cell
15
:
3509
-3519.
31
Hokin, L. E..
1985
. Receptors and phosphoinositide-generated second messengers.
Annu. Rev. Biochem.
54
:
205
-235.
32
Missy, K., V. Van Poucke, P. Raynal, C. Viala, G. Mauco, M. Plantavid, H. Chap, B. Payrastre.
1998
. Lipid products of phosphoinositide 3-kinase interact with Rac1 GTPase and stimulate GDP dissociation.
J. Biol. Chem.
273
:
30279
-30286.
33
Araki, N., M. T. Johnson, J. A. Swanson.
1996
. A role for phosphoinositide 3-kinase in the completion of macropinocytosis and phagocytosis by macrophages.
J. Cell Biol.
135
:
1249
-1260.
34
Hancock, J. F., H. Paterson, C. J. Marshall.
1990
. A polybasic domain or palmitoylation is required in addition to the CAAX motif to localize p21ras to the plasma membrane.
Cell
63
:
133
-139.
35
Aderem, A..
2002
. How to eat something bigger than your head.
Cell
110
:
5
-8.
36
Greenberg, S., S. Grinstein.
2002
. Phagocytosis and innate immunity.
Curr. Opin. Immunol.
14
:
136
-145.
37
Gagnon, E., S. Duclos, C. Rondeau, E. Chevet, P. H. Cameron, O. Steele-Mortimer, J. Paiement, J. J. Bergeron, M. Desjardins.
2002
. Endoplasmic reticulum-mediated phagocytosis is a mechanism of entry into macrophages.
Cell
110
:
119
-131.
38
Heyworth, P. G., U. G. Knaus, J. Settleman, J. T. Curnutte, G. M. Bokoch.
1993
. Regulation of NADPH oxidase activity by Rac GTPase activating protein(s).
Mol. Biol. Cell
4
:
1217
-1223.
39
Tapper, H., W. Furuya, S. Grinstein.
2002
. Localized exocytosis of primary (lysosomal) granules during phagocytosis: role of Ca2+-dependent tyrosine phosphorylation and microtubules.
J. Immunol.
168
:
5287
-5296.
40
Brown, A. M., A. J. O’Sullivan, B. D. Gomperts.
1998
. Induction of exocytosis from permeabilized mast cells by the guanosine triphosphatases Rac and Cdc42.
Mol. Biol. Cell
9
:
1053
-1063.
41
Abdel-Latif, D., M. Steward, D. L. Macdonald, G. A. Francis, M. C. Dinauer, P. Lacy.
2004
. Rac2 is critical for neutrophil primary granule exocytosis.
Blood
104
:
832
-839.
42
Yamauchi, A., C. Kim, S. Li, C. C. Marchal, J. Towe, S. J. Atkinson, M. C. Dinauer.
2004
. Rac2-deficient murine macrophages have selective defects in superoxide production and phagocytosis of opsonized particles.
J. Immunol.
173
:
5971
-5979.
43
Gu, Y., M. D. Filippi, J. A. Cancelas, J. E. Siefring, E. P. Williams, A. C. Jasti, C. E. Harris, A. W. Lee, R. Prabhakar, S. J. Atkinson, et al
2003
. Hematopoietic cell regulation by Rac1 and Rac2 guanosine triphosphatases.
Science
302
:
445
-449.
44
Filippi, M. D., C. E. Harris, J. Meller, Y. Gu, Y. Zheng, D. A. Williams.
2004
. Localization of Rac2 via the C terminus and aspartic acid 150 specifies superoxide generation, actin polarity and chemotaxis in neutrophils.
Nat. Immunol.
5
:
744
-751.
45
Yamauchi, A., C. C. Marchal, J. Molitoris, N. Pech, U. Knaus, J. Towe, S. J. Atkinson, M. C. Dinauer.
2004
. Rac GTPase isoform-specific regulation of NADPH oxidase and chemotaxis in murine neutrophils in vivo: role of the C-terminal polybasic domain.
J. Biol. Chem.
280
:
953
-964.
46
Park, H. S., S. H. Lee, D. Park, J. S. Lee, S. H. Ryu, W. J. Lee, S. G. Rhee, Y. S. Bae.
2004
. Sequential activation of phosphatidylinositol 3-kinase, β Pix, Rac1, and Nox1 in growth factor-induced production of H2O2.
Mol. Cell. Biol.
24
:
4384
-4394.
47
Kawahara, T., M. Kohjima, Y. Kuwano, H. Mino, S. Teshima-Kondo, R. Takeya, S. Tsunawaki, A. Wada, H. Sumimoto, K. Rokutan.
2004
. Helicobacter pylori lipopolysaccharide activates Rac1 and transcription of NADPH oxidase Nox1 and its organizer NOXO1 in guinea pig gastric mucosal cells.
Am. J. Physiol.
288
:
C450
-C457.

Supplementary data