Leukocyte trafficking to inflammatory sites is a gradual process, which is dominated in its early phases by chemokine- and cytokine-mediated neutrophil recruitment. The chemokine regulated on activation normal T cell expressed and secreted (RANTES) has been shown to be highly expressed in the joints of patient with rheumatoid arthritis and to promote leukocyte trafficking into the synovial tissue. In this study, we investigated the effect of RANTES in a murine model of peritoneal chemotaxis, and we found that RANTES dose-dependently induces neutrophil recruitment. Then, through morphological and histological analyses, we observed that activated neutrophils represent the major infiltrating population in response to RANTES chemotactic stimulus. Furthermore, we demonstrated that oral administration of either nonisoform-specific phosphoinositide 3-kinase (PI3K) inhibitor LY294002 (morpholin-4-yl-8-phenylchromen-4-one) or selective PI3Kγ inhibitor AS041164 (5-benzo[1,3]dioxol-5-ylmethylene-thiazolidine-2,4-dione) blocks RANTES-induced chemotaxis and reduces the level of AKT phosphorylation. Because the two compounds showed a similar pharmacokinetic profile in terms of bioavailability and half-life after oral route administration, the selective inhibition of the PI3Kγ-isoform pathway through AS041164 was three times more potent in reducing neutrophil recruitment. Finally, to confirm the blockade of neutrophil infiltration that occurs in the early phase of the inflammatory response, AS041164 was also tested in a model of carrageenan-induced paw edema in rats. Therefore, the PI3Kγ pathway plays an important role in controlling neutrophil chemotaxis during early steps of inflammation.
During inflammation, neutrophils are rapidly recruited at sites of acute infection and dominate the initial influx of leukocytes (Issekutz and Movat, 1980). Later, during the progression of inflammation, monocytes and macrophages replace neutrophils, suggesting a bimodal recruitment pattern involving a switch from neutrophils to monocytes (Doherty et al., 1988; Henderson et al., 2003). The first step in approaching a site of insult requires neutrophils to transmigrate across endothelial barriers, a process that depends on chemokines (Yoshie et al., 2001). In response to a chemotactic gradient, CXC and CC chemokines activate leukocytes by binding to seven transmembrane receptors coupled to G proteins (Proudfoot, 2002) linked to heterotrimeric G protein complexes (Proudfoot, 2002). Upon stimulation, the G protein complex dissociates and subsequently recruits various signaling components, such as nucleotide exchange factors, phospholipid lipases, and lipid kinase phosphoinositide-3′OH-kinase isoforms, such as phosphoinositide 3-kinase (PI3K) (Akasaki et al., 1999). Neutrophils have been shown to express different chemokine receptors, including CXCR2 and CCR1 (Lee et al., 1995; Zhang et al., 1999). Furthermore, the blockade of the chemokine receptor CXCR2 or its ligands IL-8 and macrophage inflammatory protein (MIP)-2, or alternatively of the CCR1/MIP-1α interaction, has been shown to inhibit neutrophil migration in murine models of inflammation (Tessier et al., 1997).
Neutrophils are responsible for driving inflammatory response in local tissues. Upon binding of TLR2 or TLR4 ligands, neutrophils up-regulate the expression of chemokines, down-regulate some chemokine receptors, and change their expression of adhesion molecules and respiratory burst mediators. Consequently, they leave the blood and migrate to sites of infection in a multistep process mediated through adhesive interactions that are regulated by macrophage-derived cytokines and chemokines (Muller and Randolph, 1999).
The CC chemokine RANTES/CCL5 (ligand for CCR1, 3, and 5) is up-regulated in the joints of patients with rheumatoid arthritis (Shahrara et al., 2003) and has been reported to play a role in the in vivo pathogenesis of experimental arthritis (Barnes et al., 1998; Shahrara et al., 2003). Accordingly, i.p. injection of recombinant human RANTES (r-hRANTES) has been demonstrated to produce a marked recruitment of inflammatory cells in the peritoneal cavity in mice (Proudfoot et al., 2003).
A large body of evidence indicates a central role of the phosphoinositide 3-kinase class IB isoform (PI3Kγ) in chemokine-induced leukocyte recruitment. Thus, PI3Kγ-deficient mice as well as mice treated with specific PI3Kγ inhibitors showed significantly reduced 3′-phosphorylated phosphoinositide production by leukocytes, following stimulation with various chemoattractants (e.g., N-formyl-l-methionyl-l-leucyl-l-phenylalanine, C5a, IL-8) (Hirsch et al., 2000). Accordingly, chemokines failed to stimulate the phosphoinositide-dependent kinase pathway (e.g., protein kinase B/AKT) in cells isolated from PI3Kγ-deficient mice (Hirsch et al., 2000). Indeed, the recruitment of neutrophils, monocytes, and macrophages in response to chemokines in vitro and in vivo in PI3Kγ-deficient mice was significantly reduced, whereas the response to pleiotropic inflammatory stimuli, such as carrageenan in the air-pouch model, was unaffected (Stephens et al., 1998; al-Aoukaty et al., 1999; Hirsch et al., 2000). Furthermore, blockade of PI3Kγ was shown to suppress joint inflammation and damage in a mouse model of rheumatoid arthritis (Camps et al., 2005).
The aim of the present study was to investigate the role of PI3K/PI3Kγ in r-hRANTES-induced neutrophil chemotaxis. To this end, the nonselective PI3K inhibitor, LY294002 (described in Vlahos et al., 1994), and the selective PI3Kγ isoform inhibitor, AS041164 (synthesis described in Rückle et al., 2004) (for details of both inhibitors, see Tables 1 and 2), were orally administered into RANTES-injected mice. Finally, to confirm the blockade of neutrophil infiltration that occurs in the early phase of the inflammatory response, AS041164 was also tested in a model of carrageenan-induced paw edema in rats, a model in which neutrophils play a crucial role (Vinegar et al., 1987). According to Vinegar et al. (1987), this model can be described as a biphasic edematous response during the first 3 h, and two inflammatory phases can be distinguished: the first nonphagocytic and the second phagocytic (PIR). Their findings show that 180 min after carrageenan challenge, a large number of neutrophils are observed histologically in and around the small blood vessels in the dermis initiating the PIR phase. The agent responsible for neutrophils' diapedesis to the injured dermal cells is unknown. Our findings illustrate the mechanism by which RANTES promotes neutrophil chemotaxis and suggest PI3Kγ inhibition as a potential treatment for inflammatory diseases that involve neutrophil recruitment.
Materials and Methods
Animals. Female BALB/c mice of approximately 8 to 12 weeks of age (18–22 g b.wt.) from Charles River Italia (Calco, Italy) were used for RANTES-induced neutrophil peritoneal recruitment, and male Wistar rats (100–150 g b.wt.) from Charles River Italia were used for carrageenan-induced inflammation model. Mice and rats were housed under the following constant environmental conditions: temperature, 22 ± 2°C; relative humidity, 55 ± 10%; and 15 to 20 air changes/h (filtered on HEPA 99.99%) and artificial light with a 12-h circadian cycle (7:00 AM–7:00 PM). All in vivo studies were performed according to the European Council Directive 86/609/EEC and the Italian Ministry guidelines for the care and use of experimental animals (decree no. 116/92). All the experimental protocols were authorized by the Italian Ministry of Health.
Chemicals and Reagents. All chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise specified. Antibodies used in flow cytometry were obtained from BD PharMingen (San Diego, CA), except the antibody against CCR3, purchased from R&D Systems (Minneapolis, MN). Recombinant wild-type RANTES (Proudfoot at al., 2003), LY294002 (described in Vlahos et al., 1994), and AS041164 (synthesis described in Rückle et al., 2004) were from the Serono Pharmaceutical Research Institute (Geneva, Switzerland). For the in vivo studies, both compounds were suspended in 0.5% carboxymethylcellulose/0.25% Tween 20 as vehicle.
r-hRANTES-Induced Neutrophil Peritoneal Recruitment. Mice received i.p. 200 μl of lipopolysaccharide-free saline or r-hRANTES (0.05–0.15–0.5–1.5 mg/kg, n = 6/group). Four and 18 h postinjection, mice were asphyxiated by CO2 inhalation, and their peritoneal cavity was washed three times with 5 ml of ice-cold PBS. The total lavage was then pooled for individual mouse, and 200 μl was processed for morphological cell analysis with cytospin (Proudfoot et al., 2003). In brief, 200 μl of peritoneal lavage was cytocentrifuged (45g, 10 min) onto slides, air dried, and stained with May-Grünwald-Giemsa. Total cells from washing collections were counted with a Beckman Coulter AcT 5diff (Beckman Coulter, Fullerton, CA).
Antichemotactic Effect of PI3K Inhibitors in RANTES-Induced Neutrophil Peritoneal Recruitment. Lipopolysaccharide-free saline or r-hRANTES (0.5 mg/kg) were injected i.p. into mice, orally pretreated (30 min before r-hRANTES injection) with LY294002 (30–100–300 mg/kg, n = 6/group), or orally pretreated AS041164 (3–10–30 mg/kg, n = 6/group). Four hours after r-hRANTES injection, mice were sacrificed, the peritoneal cavity was washed, and cells were processed as mentioned above.
Flow Cytometric Analysis. Cells obtained from peritoneal lavages were washed twice in PBS, counted, and resuspended in FACS buffer (1% bovine serum albumin in PBS containing 0.01% NaN3). For phenotype analysis, cells (0.2–1.0 × 106 cells/stain) were initially incubated with CD16/32 (2.4G2, Fc block) for 20 min at 4°C. Subsequently, cells were incubated with the appropriate antibody against cell surface markers: CD45 (30-F11), Gr1 (RB6–8C5), CCR5 (C34–3448), CD62L (MEL-14; BD PharMingen), and CCR3 (83101) from R&D Systems. All incubations were performed in ice for 20 min followed by two washes with FACS staining buffer (1% bovine serum albumin in PBS). Appropriate isotype controls were used in all cases. For flow cytometric analysis, a typical forward and side scatter gate was set to exclude dead cells and aggregates; a total of 104 events in the gate were collected and analyzed using a FACSCalibur and Cell Quest software (BD Biosciences, San Jose, CA).
Immunohistochemistry Analysis. Paraformaldehyde-fixed and paraffin-embedded mesenteric tissues were sectioned at approximately 4 to 5 μm of thickness and deparaffinized/rehydrated for immunoperoxidase staining using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA). In brief, antigen unmasking was performed by incubation in 10 mM sodium citrate buffer, pH 6.0, and endogenous peroxidase was quenched with 1% H2O2 for 10 min. Nonspecific immunoglobulin binding sites were blocked by incubating for 1 h with normal goat serum, and then sections were incubated with the primary antibody anti-phospho-c-AKT (Ser473) and anti-AKT (Cell Signaling Technology, Beverly, MA) overnight at 4°C. Sections were successively incubated for 30 min with a biotinylated secondary antibody solution followed by 30-min incubation with ABC reagent (Vectastain Elite ABC kit; Vector Laboratories). Immunoglobulin complexes were visualized by incubation with 3,3′-diaminobenzidine and then washed, counterstained with hematoxylin, cleared, dehydrated, mounted, and examined by light microscopy. Ten fields were observed for each sample. As negative control for the immunohistochemical staining, tissue sections were treated with normal serum instead of primary antibodies. Conventional histological observation was performed using hematoxylin and eosin staining.
Western Blotting Analysis. Cells obtained from peritoneal lavages were lysed with ice-cold lysis buffer [62.5 mM Tris-HCl, pH 6.8, at 25°C, 2% (w/v) SDS, 10% glycerol, 50 mM dithiothreitol, and 0.01% (w/v) bromphenol blue or phenol red]. After sonication, lysates were centrifuged, protein concentration was determined, and 20 μg of proteins was separated by electrophoresis on 10% SDS-polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride-plus membrane. After blocking with 5% milk, the immunoblots were probed with a 1:100 dilution anti-phospho-c-AKT (Ser473) antibody (Cell Signaling Technology) overnight at 4°C, followed by a 1-h incubation at room temperature with the corresponding secondary antibodies. The blots were visualized with ECL-Plus reagent. Phospho-AKT immunoblots were then stripped with strip buffer at 50°C for 30 min and reblotted for total AKT (Cell Signaling Technology). The volume of the protein bands was quantified by a Bio-Rad ChemiDoc EQ densitometer and a Bio-Rad Quantity One software (Bio-Rad Laboratories, Hercules, CA). Total AKT was used as loading control. Phosphorylation of AKT was measured as a ratio of phospho-AKT versus total-AKT and expressed as -fold change over the saline-treated control.
Carrageenan-Induced Inflammation. Male Wistar rats (100–150 g b.wt.) were injected subplantarly with 0.1 ml of a 1% λ-carrageenan (Sigma-Aldrich) suspension in sterile saline. Animals were fasted for approximately 15 h before experiment starting. Paw volume was determined by water displacement, i.e., measuring the volume of displaced fluid by means of a plethysmometer (Ugo Basile, Comerio, Italy) 0, 1, 3, 5, and 24 h after carrageenan challenge. Edema volumes were determined as the difference between paw volume at each indicated time point and preinjection values (time 0). Paw-swelling time-dependent variations in each treatment group were then compared with control animals receiving vehicle alone. Compounds were administered 0.5 h before carrageenan challenge.
Statistical Analysis. All values in the text and figures are presented as mean ± S.E.M. of (n) independent experiments. All data were analyzed by one-way ANOVA followed by Tukey test. p Values < 0.05 were considered statistically significant. For peritoneal chemotaxis experiments, a dose-response curve was plotted from the inhibition values obtained for each dose group at the peak effect, and when possible, the relative ED50 value was calculated using S-Plus 2000 version 4.6 statistical software (Mathsoft Inc., Seattle, WA).
r-hRANTES Promotes Rapid Recruitment of Gr1+Neutrophils. To characterize the cell subsets recruited by r-hRANTES, peritoneal exudate cells (PECs) were harvested 4 and 18 h after i.p. injection. Total cells from washing collections were counted with a Beckman Coulter AcT 5diff and stained for cytospin analysis with May-Grünwald-Giemsa (Fig. 1, A–C). A significant dose-dependent infiltration of neutrophils, with no changes in macrophage counts, was observed 4 h postinjection (Fig. 1A). At the 18-h time point, neutrophil presence was less evident (significant only at 1.5 mg/kg dose), and more macrophages were observed, although this trend was not statistically significant compared with saline-treated mice (Fig. 1A). To further characterize the cell composition, the peritoneal infiltrate cells were analyzed by flow cytometry (Fig. 1D). This analysis confirmed that 4-h r-hRANTES administration resulted in recruitment of Gr1+ neutrophils, characterized by a low side scatter (Fig. 1D). Accordingly, a histological analysis of the mesenteric tissue confirmed the neutrophil infiltration 4 h post-r-hRANTES injection, which is not present after saline injection (Fig. 2, A and B). These results suggest that i.p. r-hRANTES injection results in a rapid neutrophil recruitment, which gradually resolves in the absence of additional local inflammatory signals.
Inhibition of PI3K Reduces r-hRANTES-Induced Neutrophil Recruitment. To investigate the role of PI3K in r-hRANTES-induced chemotaxis, r-hRANTES was injected in the presence and absence of a nonselective PI3K inhibitor (LY294002; 30–300 mg/kg p.o.) or of a selective PI3Kγ-isoform inhibitor (AS041164; 3–100 mg/kg p.o.). Both inhibitors dose-dependently decreased r-hRANTES-induced neutrophil recruitment (Fig. 3, A and B). ED50 values for LY294002 and AS041164 were 81.59 (95% confidence interval, 26.9–136.28) and 27.35 mg/kg (95% confidence interval, 14.26–68.95), respectively. Because the two compounds showed a similar pharmacokinetic profile in terms of bioavailability and half-life after oral route administration, the selective inhibition of the PI3Kγ-isoform pathway through AS041164 was three times more potent in reducing neutrophil recruitment.
PI3Kγ Inhibition Reduces the Level of AKT Phosphorylation. To further characterize the mechanism by which these inhibitors block neutrophil recruitment, r-hRANTES-injected mice were treated with LY294002 (100 mg/kg p.o.), AS041164 (30 mg/kg p.o.), or with the vehicle. Four hours post-r-hRANTES injection, PECs were harvested, counted, and analyzed by Western blot for the level of phosphorylation of the downstream PI3K substrate Ser/Thr kinase AKT. As previously noted, both compounds significantly decreased the number of recruited neutrophils (Fig. 3). Furthermore, in vivo treatment with both inhibitors resulted in a significantly decreased r-hRANTES-induced AKT phosphorylation (Fig. 4). Accordingly, morphological analysis of mesenteric tissue revealed that r-hRANTES injection increased the number of phospho-AKT-positive neutrophils (Fig. 5, B and F) compared with saline-injected mice (Fig. 5, A and E). Moreover, a complete abrogation of immune cell chemotaxis was observed in r-hRANTES-injected mice treated with LY294002 (Fig. 5, C and G) or AS041164 (Fig. 5, D and H). Finally, a concomitant flow cytometric analysis of PECs defined a Gr1+ neutrophil population in r-hRANTES-injected saline-treated mice, which was not observed in mice treated with either of the PI3K inhibitors (Fig. 5, J–L).
To further characterize the cellular infiltrate in the peritoneal cavity, we performed a flow-cytometric analysis of PECs harvested from r-hRANTES-injected mice. This analysis confirmed that r-hRANTES administration resulted in infiltration of Gr1+ neutrophils. In addition, recruited neutrophils displayed a low side scatter and were positive for L-selectin (CD62L) and negative for CCR3 and CCR5 (Fig. 6). In contrast, peritoneum-resident neutrophils isolated from saline-treated mice that showed a high side scatter and mild expression of the lineage marker Gr1 were negative for CD62L but strongly expressed the two RANTES receptors CCR3 and CCR5. Importantly, the administration of both PI3K inhibitors was able to inhibit the recruitment of this RANTES-induced neutrophil population (Fig. 6).
Inhibition of PI3K Significantly Reduces Paw Swelling after Carrageenan Challenge. The administration of AS041164 in a carrageenan-induced inflammation model in rat showed a significant inhibition of paw swelling. In particular (Fig. 7), AS041164 at the dose of 100 mg/kg p.o. induced a significant reduction of paw thickness that was maximal 3 h after the carrageenan injection (p < 0.01) during the PIR phase when a massive infiltration of neutrophils into the dermal tissue was reported. Indomethacin (at 5 mg/kg p.o.) was included in the experimental design as internal reference control since efficacy of nonsteroidal anti-inflammatory drugs was described in this model (Otterness and Moore, 1988). Indomethacin administration exerted an inhibitory effect on paw edema; in fact, animals belonging to that treatment group presented a significantly reduced (p < 0.001) swelling again 3 h after carrageenan injection.
In this study, we used a murine model of peritoneal chemotaxis to show that r-hRANTES injection induces a dose-dependent neutrophil recruitment via the PI3Kγ pathway. This evidence is confirmed by morphological and histological analyses showing that neutrophils are the major recruited population during the early phase of inflammation. In addition, a cytofluorimetric analysis of cellular peritoneum infiltrate shows that the RANTES-recruited neutrophil population displays a different phenotype and can be distinguished from resident cells. Resident neutrophils display a high side scatter and mild expression of the lineage marker Gr1. They are negative for L-selectin (CD62L); however, they strongly express the two RANTES receptors CCR3 and CCR5 (Proudfoot, 2002). On the contrary, RANTES-recruited neutrophils present a low side scatter and high Gr1 expression, and they are L-selectin-positive and CCR3- and CCR5-negative. Molecules of the selectin family such as L-selectin are constitutively expressed on leukocytes and shed from the cell surface upon activation (Lee et al., 2007). Therefore, it is possible that neutrophils transiently recruited by RANTES in the absence of additional inflammatory signals do not become activated in the absence of local inflammatory stimuli.
Neutrophils are not only the most abundant cell type in inflamed joints, they are also important in the initiation and perpetuation of the inflammation in both human rheumatoid arthritis and mouse models (Kistsis and Weissmann, 1991; Wipke and Allen, 2001). Neutrophils and inflammatory monocytes are recruited rapidly into sites of infection. The initial influx of leukocytes into inflamed tissues is dominated by neutrophils. This dominance may reflect the higher frequency of neutrophils in peripheral blood compared with monocytes (Muller and Randolph, 1999). Later in inflammation, monocytes/macrophages gradually replace neutrophils as the predominant leukocyte subtype. Recruited neutrophils are thought to mediate this switch by releasing soluble factors into the early inflammatory milieu, thus promoting monocyte recruitment (Issekutz and Movat, 1980; Doherty et al., 1988; Henderson et al., 2003). Neutrophils and monocytes migrate across the endothelium into tissues in response to endothelial cell-bound factors, such as chemokines, that deliver activating and chemoattracting signals (Proudfoot, 2002). This critical role of chemokines in the recruitment of myeloid cells is well established in murine in vivo models (Baggiolini et al., 1997). CC chemokines, including RANTES, are known to act on monocytes, lymphocytes, basophils, and eosinophils but do not usually target neutrophils. Traditionally, human PMNs have been thought to express receptors for the CXC (Baggiolini, 1998) or CX3C (Baggiolini et al., 1997) families and to preferentially migrate toward chemokine ligands from these two families (Pan et al., 1997). However, different studies have provided conflicting evidences on whether human PMNs express CCR receptors. Although Xu et al. (1995) found that resting human PMNs possess binding sites for the CC chemokines MIP-1α and monocyte chemoattractant protein-3, it is generally accepted that CC chemokines have no direct functional effect on resting human PMNs. In contrast to findings in human PMNs, experiments performed in several murine models of inflammation have shown that CC chemokines do play a role in PMN chemotaxis (Gao et al., 1997). For example, in vivo studies of lipopolysaccharide- and endotoxemia-associated lung injury in mice have demonstrated that neutralization of MIP-1α attenuates neutrophil infiltration into inflammatory sites (Gao et al., 1997). Furthermore, PMNs isolated from inflammatory exudates in mice display chemotactic migration and calcium flux responses to MIP-1α (Gao et al., 1997). Furthermore, a growing body of literature demonstrates that cytokines regulate the expression of chemokine receptors on a variety of cell subsets. Thus, proinflammatory mediators, such as tumor necrosis factor-α, IL-1β, and lipopolysaccharide, have been shown to down-regulate CCR2 expression by human monocytes (Sica et al., 1997), whereas the immunomodulatory cytokine IL-10 up-regulates CCR1, CCR2, and CCR5 on these cells (Sozzani et al., 1998). PI3K has a central role in the regulation of inflammatory responses, whereas the isoforms α and β are almost ubiquitously expressed and regulate a variety of cell functions as proliferation and survival; the δ and γ isoforms are mostly expressed in the hematopoietic system and have been shown to regulate immune responses (Bi et al., 2002). Furthermore, it has been demonstrated that PI3Kγ plays a central role in chemokine-induced recruitment of leukocytes (Sasaki et al., 2000). In fact, PI3Kγ knockout mice showed a reduced chemoattractant-induced respiratory burst, defective migration of macrophages and neutrophils to inflammation sites, as well as impairments in adaptive immunity in general and in T-cell activation in particular (Del Prete et al., 2004). In addition, impairment in the capacity of chemokines to stimulate the phosphoinositide-dependent kinase cascades (e.g., protein kinase B/AKT) has been shown in cells that lack PI3Kγ expression (Hirsch et al., 2000). Accordingly, PI3Kγ–/– mice are resistant to collagen II-specific antibody-induced arthritis, and the administration of a selective PI3Kγ inhibitor was shown to suppress the progression of joint inflammation and damage in a mouse model of collagen-induced arthritis (Camps et al., 2005).
In the present work, we evaluated the effect of two PI3K inhibitors with different specificity and similar pharmacokinetic profiles, namely LY294002 (described in Vlahos et al., 1994), a nonisoform-selective PI3K inhibitor with low affinity for the γ-isoform (Ward et al., 2003), and AS041164 (synthesis described in Rückle et al., 2004), a selective PI3Kγ inhibitor (for more details, see Tables 1 and 2). We found that AS041164-mediated inhibition of RANTES-induced chemotaxis is three times more potent compared with that of LY294002, suggesting that the PI3Kγ isoform has a specific role in RANTES-induced chemotaxis. To further support these data, the level of AKT phosphorylation following RANTES injection was investigated in the presence and absence of the two inhibitors. We observed that both LY294002 and AS041164 decreased the level of AKT phosphorylation in PECs. Finally, we demonstrated that inhibition of PI3Kγ pathway in vivo results in the reduction of inflammatory swelling in the model of carrageenan-induced paw edema and that maximal effect is obtained at time points in which neutrophils are massively recruited.
Together, our data suggest that the mechanism by which RANTES induces neutrophil recruitment involves PI3K/AKT pathway activation, via class 1B G-protein coupled receptors. The administration of both PI3K inhibitors specifically blocks RANTES-mediated cellular recruitment as well as inflammatory infiltrates induced by a flogistic agent such as carrageenan. Interestingly, suppression of joint inflammation in murine models of arthritis by PI3Kγ inhibition has been reported to occur through a reduced presence of neutrophils in joint lesions (Camps et al., 2005).
In conclusion, the present findings suggest that PI3Kγ plays an important role in controlling neutrophil chemotaxis in early steps of inflammation. Therefore, pharmacological interventions targeting PI3Kγ could represent an interesting and promising new therapeutic approach to treat inflammatory diseases.
We thank Alberto Franzino for technical support and Amanda Proudfoot for providing the biochemical tools for the validation of the chemotaxis peritoneal recruitment model.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: PI3K, phosphoinositide 3-kinase; IL, interleukin; MIP, macrophage inflammatory protein; RANTES, regulated on activation normal T cell expressed and secreted; r-hRANTES, recombinant human regulated on activation normal T cell expressed and secreted; LY294002, morpholin-4-yl-8-phenyl-chromen-4-one; AS041164, 5-benzo[1,3]dioxol-5-ylmethylene-thiazolidine-2,4-dione; PIR, phagocytic response; PBS, phosphate-buffered saline; FACS, flow analyzer cell sorter; PEC, peritoneal exudate cell; PMN, polymorphonuclear neutrophil; AKT, protein kinase B.
- Received March 20, 2007.
- Accepted May 24, 2007.
- The American Society for Pharmacology and Experimental Therapeutics