Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Lipopolysaccharide (LPS), a component of bacterial endotoxin, is a potent inducer of cell activation and induces the secretion of a large number of inflammatory mediators, including cytokines (e.g., interleukin [IL]-1, tumor necrosis factor [TNF]-, and macrophage inhibitory factor [MIF]), chemokines (e.g., IL-8, regulated on activation, normal T cells expressed and secreted [RANTES], and monocyte chemotactic protein [MCP]-1) and lipid mediators (e.g., platelet activating factor [PAF] and leukotriene B₄ [LTB₄]), which have key roles in mediating leukocyte recruitment (1).

Accumulating evidence suggests that LPS may regulate the bronchoconstriction and eosinophil function in asthma. Inhaled LPS leads to airway inflammation with symptoms including cough, bronchospasm, and nonspecific bronchial reactivity in normal human subjects (2, 3). Recently, it has been demonstrated that subjects with asthma have increased sensitivity to inhaled LPS compared with individuals without asthma (4). Although the mechanisms involved in this increased sensitivity are not clear, the involvement of eosinophils is a possibility, prompting a growing interest in the possible effects of LPS on eosinophil activation, recruitment, and survival. Indeed, LPS induces eosinophil accumulation in vivo in experimental animals (5) and in humans (8), primes for chemoattractant-induced eosinophil recruitment (9), and increases eosinophil survival in vitro (10, 11).

We have previously demonstrated that the intrathoracic (i.t.) injection of LPS into mice and rats induces not only influx of neutrophils in the pleural cavities, but also an intense and long-lasting eosinophil accumulation (12). In contrast to the eosinophilia observed in allergic reactions, the eosinophil accumulation induced by LPS is independent of IL-5 (12), but depends on the neosynthesis of an unidentified soluble heat-stable protein with specific eosinophilotactic activity and a molecular weight (M.W.) ranging between 10 and 50 kD (13). The LPS-induced eosinophil accumulation is an indirect mechanism (13), and requires the cooperation of T lymphocytes and resident macrophages (6, 14). However, the full mechanism involved in LPS-induced eosinophil recruitment and activation is unknown.

Chemokines are low molecular weight chemoattractant cytokines that play an important role in leukocyte trafficking to the inflammatory site. Chemokines are divided into four structurally different families according to the number and position of the conserved cysteine domains:  or CXC chemokines, such as IL-8 and IP-10, which predominantly attract neutrophils and lymphocytes;  or CC chemokines, such as eotaxin, RANTES, and macrophage inflammatory protein (MIP)-1, which are chemotactic for eosinophils, monocytes and lymphocytes; the C chemokine, lymphotactin, which attracts T cells; and the CX₃C chemokine, fractalkine, which is a chemoattractant for lymphocytes and monocytes (15).

In the current study, we have investigated the production of CC chemokines in the inflammatory response to LPS, evaluating mRNA and chemokine levels in the pleural cavity. The involvement of CC chemokines in the eosinophil influx was also investigated, by means of in vitro and in vivo treatments with antibodies against eotaxin and RANTES, and with a poxvirus 35-kD CC chemokine- binding protein (vCKBP) able to selectively neutralize the CC chemokine activity. Our results indicate that one or more of the LPS-induced CC chemokines plays a role in T lymphocyte migration, but the eosinophilotactic protein produced after LPS stimulation is not RANTES, eotaxin, or any of the vCKBP-inhibitable CC chemokines. However, our data strongly suggest that LPS-induced eosinophilotactic protein acts through CCR3 receptor.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

C57BL/6 mice of either sex weighing 20 to 25 g, and male Wistar rats weighing 150 to 180 g, were obtained from Oswaldo Cruz Foundation breeding unit (Rio de Janeiro, Brazil) and caged with free access to food and fresh water in a room with temperature ranging from 22 to 24°C and a 12-h light/dark cycle in the Department of Physiology and Pharmacodynamic experimental animal facility until used.

Antibodies and Reagents

LPS (from Escherichia coli serotype 0127:B8), BSA, ovalbumin, RPMI, phosphate-buffered saline (PBS), and HEPES were purchased from Sigma Chemical Co. (St. Louis, MO) and pertussis toxin was obtained from Calbiochem (San Diego, CA). Ficoll-Paque Plus was obtained from Pharmacia Biotech (Sweden) and 6% Dextran 70 was purchased from McGaw, Inc. (Irvine, CA). Murine recombinant eotaxin was obtained from PeproTech (London, UK). For in vivo neutralizing studies, antimurine eotaxin, antimurine RANTES, and antimurine CC chemokine receptor 3 (CCR3) monoclonal Abs (mAbs) were obtained from R&D Systems (Minneapolis, MN). For in vitro assays, antimurine eotaxin polyclonal Abs (pAbs) were kindly provided by Steven L. Kunkel (University of Michigan, Ann Arbor, MI), whereas antihuman CCR3 mAbs 7B11 were provided by Dr. Paul D. Ponath, LeukoSite Inc. (Cambridge, MA). VV Lister 35 kD protein (vCKBP) was kindly provided by Dr. J.A. Symons, Sir William Dunn School of Pathology, University of Oxford, UK (16).

Immunostaining was performed using the following mAbs from PharMingen (San Diego, CA): FITC-conjugated hamster IgG antimurine CD3, FITC-conjugated rat IgG2a antimurine CD4, PE-conjugated rat IgG2a antimurine CD8 and PE-conjugated hamster IgG antimurine TCR.

Pleurisy

Pleurisy was induced by an i.t. injection under anesthesia of either LPS (250 ng/cavity), eotaxin (30 pmol/cavity), or LPS pleural wash (1:2 vol/vol) diluted in sterile PBS to a final volume of 100 µl. Control group received an i.t. injection of 100 µL of sterile PBS. At specific time points after the stimuli, the animals were killed by an excess of carbon dioxide and their thoracic cavity was rinsed with 1 ml of saline containing heparin (10 UI/ml).

Allergic Pleurisy

Active sensitization was achieved by a subcutaneous injection of 0.2 ml of a mixture of ovalbumin (50 µg) and aluminum hydroxide (5 mg). Fourteen days later, animals were challenged by an i.t. injection of ovalbumin (12 µg/cavity) and killed 24 h later for pleurisy evaluation as described above. Sensitized mice challenged with PBS vehicle alone were used as the negative control group.

Treatments

In designated experiments, animals received an intraperitoneal (i.p.) injection of neutralizing antimurine eotaxin (15 or 30 µg/mouse), antimurine CCR3 (15 or 30 µg/mouse), or antimurine RANTES (10 µg/mouse) mAbs 30 min before stimulation with LPS, ovalbumin or eotaxin as indicated. Purified rat IgG was used as a control. In another set of experiments, animals received a 10 or 50 pmol dose of vCKBP diluted in the same preparation as stimuli, and incubated with the stimuli for 5 min at 37°C prior i.t. injection.

Leukocyte Counts

Total leukocyte counts were performed using a Neubauer chamber under an optical microscope, after dilution in Türk fluid (2% acetic acid). Differential counts of mononuclear cells, neutrophils, and eosinophils were made under an oil immersion objective, using stained cytospins by the May-Grünwald-Giemsa method (Cytospin 3, Shandon Inc., Pittsburgh, PA). Counts are reported as numbers of cells per cavity.

Immunofluorescent Staining and Flow Cytometric Analysis

Samples of 10⁶ cells recovered from mesenteric lymph nodes or pleural cavities were labeled with the appropriate concentration of FITC or PE-conjugated mAbs to CD3, CD4, CD8, or TCR for 30 min at 4°C after incubation with rat serum to block nonspecific binding. Cells were then washed with PBS/0.1% azide and surface marker analysis performed using the Cell Quest program in a FASCalibur flow cytometer (Becton Dickinson, San Jose, CA). At least 10⁴ lymphocytes were acquired per sample. All data were collected and displayed on a log scale of increasing green and red fluorescence intensity. Data were presented as two-dimensional dot plots. To determine the percentages of the lymphocyte subpopulations, lymphocytes were specifically gated. Counts are reported as numbers of cells per cavity.

Pleural Wash Transfer Assay

Six hours after the i.t. injection of LPS (250 ng/cavity) or saline into donor mice, the pleural cavity was washed with 200 µL of sterile PBS. The pleural wash was centrifuged (400 × g, 10 min) to remove cells and heated for 30 min at 100°C. The pleural wash was then submitted to ultracentrifugation (30,000 × g, 30 min) and filtered (Millipore filters, 0.22 µm; Bedford, MA). For shape change assays, the pleural wash from LPS-injected animals (LPW) was passed through Sep-Pak Plus C-18 cartridges (Waters Corporation; Milford, MA) and lyophilized.

Isolation of Eosinophil Chemoattractant Activity by High Performance Liquid Chromatography

To obtain larger quantities of LPW for purification, rats were injected i.t. with LPS (250 ng/cavity) and washed at 6 h with 3 ml sterile PBS. After centrifugation, heating and filtration as described above, the LPW was adjusted with TFA to pH 2.0, passed through C18 Sep Pak cartridges, eluted with acetonitrile (ACN)/ 0.08% TFA, and lyophilized. The extract was dissolved in 0.08% TFA and applied to a wide pore reversed phase HPLC column (C18 Vydac, 4.6 × 250 mm, 300 Å). The column was then eluted with a linear gradient of acetonitrile (0-80% ACN in 0.08% TFA) at 1 ml/min, for 80 min. Fractions were collected each minute. Bioactivity was determined by an eosinophil shape change assay as described below. Aliquots of 240 µL of each fraction (or 240 µL from each of two adjacent fractions) were lyophilized in the presence of a carrier protein (BSA) and redissolved in 400 µL of assay buffer.

Shape Change Assay

The shape change assay was previously described (17). In brief, polymorphonuclear leukocytes (PMNL) were obtained by peripheral vein puncture from healthy normal or atopic subjects under no systemic medication. PMNL were purified by dextran sedimentation followed by Ficoll-Paque discontinuous gradients. Any erythrocyte contamination of the PMNL pellet was lysed by hypotonic shock. Purified PMNL were re-suspended and preincubated at 37°C for 30 min in a solution of PBS (10 mM), HEPES (10 mM), glucose (10 mM), and BSA (0.1%) at pH 7.4. Aliquots of 5 × 10⁵ cells (of which 3-10% were eosinophils) were incubated in a final volume of 100 µL in the presence or absence of the agonists in a 37°C shaking water bath for 6 min, after which the reaction was stopped by the addition of 250 µL of ice-cold fixative solution (CellFIX, Mountain View, CA) and placed on ice until analysis.

In some experiments, the cell aliquots were preincubated with antihuman-CCR3 mAbs (80 ng/ml) or murine IgG1 (MOPC 21) for 10 min or pertussis toxin (100 ng/ml) for 1 h. In another set of experiments, agonists were preincubated for 10 min with either antimurine eotaxin pAbs (1:10 dilution), normal rabbit serum or vCKBP (10 pmol/tube).

Samples were immediately analyzed on a FACScalibur flow cytometer (Becton Dickinson). Acquisition was set using a FL-2 fluorescence channel, through which human eosinophils can be distinguished from neutrophils by means of their different autofluorescence characteristics. Forward scatter (FSC-H), side scatter (SSC-H), and FL-2 data were saved. Five hundred eosinophils were acquired for each of the duplicate samples. As measurement of shape change, data are reported as percentage change in FSC-H compared with a baseline of 100% for buffer-treated cells.

Enzyme-Linked Immunosorbent Assay

Measurement of RANTES, eotaxin, and MIP-1  in the pleural fluid were performed by sandwich enzyme-linked immunosorbent assay (ELISA) using matched antibody pairs from R&D Systems (Minneapolis, MN) according to the manufacturer's instructions.

RNAse Protection Assay

Multiple mRNA chemokine expression analysis was performed on mice leukocytes recovered from pleural cavities 6 h after the i.t. injection of LPS or vehicle. Cells were resuspended in Ultraspec total RNA isolation reagent (Biotecx Laboratory Inc., Houston, TX) (10⁶ cells/ml) and total RNA purified as recommended by the manufacturer. Ten micrograms of total RNA were applied per lane. mRNA expression was evaluated with the multiple chemokine RNAse protection assay mCK-5 multiprobe template set, according to the manufacturer's instructions (PharMingen, San Diego, CA).

Statistical Analysis

Data are reported as the mean ± SEM and were analyzed statistically by means of analysis of variance (ANOVA) followed by Newman-Keuls-Student test or Student's t test. Values of P  0.05 were regarded as significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Analysis of Leukocyte Accumulation Induced by LPS in the Pleural Cavity of Mice

The i.t. injection of LPS into the pleural cavity of C57BL/6 mice (250 ng/cavity) induced a twofold increase in the numbers of total leukocytes in the pleural cavity 6 h after stimulation (Table 1). This increase was due to a marked neutrophil migration at the acute phase of the LPS-induced response, with no changes observed in the numbers of other cell types. As previously reported by Bozza and colleagues (12), an intense influx of mononuclear cells and eosinophils was observed 24 h after LPS administration that remained significantly elevated until 72 h (data not shown). FACS analysis showed an accumulation of T lymphocytes within 24 h, but not 6 h, after LPS administration. As previously reported, ⁺ T cells were significantly increased 24 h after LPS administration (14, Table 1).

Analysis of Chemokine Production in the In Vivo Response to LPS

LPS is a potent stimulus for the production of a wide variety of inflammatory mediators, including cytokines and chemokines (18, 19). To determine the RNA expression of CC chemokines in the pleural cavity after LPS stimulation, we used a multiprobe RNAse protection assay (mCK-5; PharMingen). Assay of RNA purified from cells obtained from the pleural cavity of mice 6 h after i.t. injection of LPS revealed the increased expression of mRNA for a wide range of chemokines, including RANTES, eotaxin, MIP-1, MIP-1, MIP-2, IP-10, and MCP-1 (Figure 1). The production of eotaxin and RANTES was confirmed by the detection of these chemokines in 6-h cell-free pleural washes from LPS-injected mice (LPW). Eotaxin and RANTES levels in mice pleural washes recovered 24 h after LPS stimulation were substantially decreased compared with the levels found at 6 h. Importantly, these chemokines could not be detected by ELISA after heating LPW to 100°C for 30 min (Figure 2). By contrast, even though mRNA expression for MIP-1 was increased after LPS stimulation, we were unable to detect increased levels of this chemokine in the 6-h LPW (2.5 ± 0.6 fmol/cavity in SPW group compared with 3.1 ± 0.6 fmol/cavity in LPW; NS).

View larger version (78K): [in this window] [in a new window]   Figure 1.   Chemokine mRNA profile of total leukocytes recovered from mouse pleural cavities 6 h after saline or LPS stimulation. Chemokine expression was determined by a multiprobe RNAse protection assay, using RNA from two pools of cells from saline-injected mice (lanes 1 and 2); and three pools of cells from LPS-injected mice (lanes 3, 4, and 5). The constitutive expressed genes mL32 (murine ribosomal protein L32) and mGAPDH (murine GAPDH) were used as control. Ten micrograms of RNA were applied per lane.

View larger version (10K): [in this window] [in a new window]   Figure 2.   Time-dependent generation of RANTES (A) and eotaxin (B) in LPS-injected mice. Displayed are chemokine protein levels determined by ELISA in pleural washes recovered at 6 and 24 h after LPS stimulation (LPW), compared with control groups injected with saline (SPW), and 6 h heated LPW (30 min, 100°C). Results are expressed as the mean ± SEM from six animals. Statistically significant differences (P  0.05) are indicated by an asterisk.

Effect of Neutralization of Eotaxin and RANTES upon Eosinophil Migration and Activation

Eotaxin and RANTES are potent chemoattractants for eosinophils (20, 21). Because we demonstrated the mRNA expression and secretion of these two chemokines during the inflammatory reaction induced by LPS, we decided to investigate their putative involvement in the eosinophil influx triggered by LPS. We performed an in vivo pretreatment with a neutralizing antimurine eotaxin mAb, followed by the i.t. administration of LPS (250 ng/cavity). As shown in Figure 3A, neutralization of eotaxin (antimurine eotaxin mAb 15 and 30 µg/mouse) failed to inhibit the eosinophil influx induced by LPS. It should be noted that at the lower dose used, the anti-eotaxin mAb inhibited the eosinophil influx induced by eotaxin itself by 84% (Figure 3A, inset). Furthermore, no changes were observed between LPS-injected untreated and anti-eotaxin-treated mice in either mononuclear cells (5.50 ± 0.40 × 10⁶ cells/cavity in treated group versus 4.70 ± 0.40 × 10⁶ cells/cavity in untreated group; NS) or neutrophils (1.56 ± 0.24 × 10⁶ cells/cavity in treated group versus 1.35 ± 0.15 × 10⁶ cells/cavity in untreated group; NS). Similarly, anti-RANTES mAb treatment, which blocked the 24-h eosinophil influx induced by ovalbumin in presensitized animals, failed to inhibit the LPS-induced influx of eosinophils (Figure 4), mononuclear cells (1.35 ± 0.08 × 10⁶ cells/ cavity in treated group versus 1.60 ± 0.25 × 10⁶ cells/cavity in untreated group; NS), neutrophils (0.53 ± 0.05 × 10⁶ cells/ cavity in treated group versus 0.71 ± 0.12 × 10⁶ cells/cavity in untreated group; NS), or T cells (8.11 ± 0.96 × 10³ cells/cavity in treated group versus 7.30 ± 1.57 × 10³ cells/ cavity in untreated group; NS).

View larger version (27K): [in this window] [in a new window]   Figure 3.   (A) Effect of anti-eotaxin mAb treatment (15 or 30 µg/ cavity) or purified rat IgG as control (30 µg/cavity) on the 24-h eosinophil accumulation induced by LPS (250 ng/cavity) into mouse pleural cavity. The insert shows the effect of anti-eotaxin or IgG control (15 µg/cavity) on eotaxin (30 pmol/cavity)-induced eosinophil accumulation. Treatments were administered i.p. 30 min before stimuli. Results are expressed as the mean ± SEM from 6-10 animals. Statistically significant differences (P  0.05) are indicated by an asterisk. (B) Eosinophil shape change induced by mu-eotaxin (10 nM) or RP-HPLC LPW bioactive fractions collected at minutes 21-22, 26-27, and 35-36. Mixed human PMNL were treated for 6 min at 37°C with agonists in the presence or absence of antimurine eotaxin pAbs (1:10). Results are expressed as the percentage increase in FSC induced by each agonist compared with that of buffer-stimulated cells. Data represent the mean ± SEM of at least three experiments, each using cells from a different donor.

View larger version (21K): [in this window] [in a new window]   Figure 4.   Effect of anti-RANTES (10 µg/cavity) i.p. treatment on the 24-h eosinophil accumulation induced by ovalbumin (OVA) (12 µg/cavity) (A) or LPS (250 ng/cavity) (B) into mouse pleural cavity. Results are expressed as the mean ± SEM performed with at least six animals. Statistically significant differences (P  0.05) are indicated by an asterisk.

We have previously demonstrated that the i.t. injection of LPS generates a protein with selective eosinophil chemoattractant activity in the pleural fluid of mice that can be transferred to naive recipient animals (12, 13). It is important to note that this activity is not species specific, since both the rat and mouse proteins are able to activate human, guinea pig, and mouse eosinophils; in addition, the protein recovered from mice also activates human and rat eosinophils (not shown). Thus, for in vitro studies, we used rat LPW, partially purified by reverse phase (RP)-HPLC in human eosinophil shape change assay. The partial purification of the eosinophil chemoattractant activity by RP-HPLC resulted in bioactive fractions eluting at 21-22, 26- 27, and 35-36 min, as assessed by eosinophil shape change assay. It should be noted that the LPW and its fractions were devoid of LPS contamination as attested by LAL assay (QCL 1000; BioWhittaker, Walkersville, MD) and LPS itself at different concentrations failed to induce eosinophil shape change (data not shown). In agreement with in vivo data (Figure 3A), the eosinophil shape change induced by RP-HPLC LPW bioactive fractions was not inhibited by 10 min preincubation with antimurine eotaxin pAbs, which cross-react with rat eotaxin (data not shown), whereas under the same conditions the response to 3 nM of eotaxin was blocked (Figure 3B).

Involvement of CCR3 in Eosinophil Activation Induced by LPW In Vitro and on LPS-Induced Eosinophil Accumulation In Vivo

Chemokines bind to and activate seven transmembrane receptors that are coupled to heterotrimeric G proteins. As shown in Figure 5A, preincubation of eosinophils with pertussis toxin significantly inhibited the shape change induced by LPW in these cells, suggesting the involvement of G_i proteins in eosinophil activation induced by LPW.

View larger version (24K): [in this window] [in a new window]   Figure 5.   (A) Effect of pertussis toxin on eosinophil shape change induced by mu-eotaxin (3 nM) or pleural washes recovered 6 h after LPS i.t. injection (LPW). Human PMNL were incubated with pertussis toxin (100 ng/ml) for 1 h before the addition of agonists. (B) Effect of anti-hCCR3 mAb or isotype control (MOPC 21) (100 ng/ml) on eosinophil shape change induced by mu-eotaxin or LPW or RP-HPLC LPW bioactive fractions. Cells were incubated with anti-CCR3 mAb for 10 min before stimulation. Data are expressed as one representative experiment from at least two separate experiments, each using cells from a different donor. (C) Effect of vCKBP on eosinophil shape change induced by mu-eotaxin (10 nM) or pleural washes recovered 6 h after the i.t. injection of saline (SPW) or LPS (LPW). Agonists were incubated with vCKBP for 10 min before stimulation. Data represent the mean ± SEM of three separate experiments, each using cells from a different donor. + represents statistically significant differences between stimulated group and vCKBP-treated stimulated group. Results are expressed as the percent increase in FSC induced by each agonist in normal or pretreated cells.

The suggested role for a G protein-coupled receptor during LPW signaling in eosinophils led us to investigate the involvement of the CCR3. CCR3 is highly expressed in eosinophils and is thought to be a central pathway for eosinophil activation by CC chemokines, including eotaxin and RANTES (17, 22). As shown in Figure 5B, 10 min preincubation of eosinophils with mAbs against CCR3 (80 ng/ml), but not its isotype control, significantly blocked the increase in eosinophil FSC-H values induced by 3 nM murine eotaxin. Moreover, preincubation with anti-CCR3 mAbs also inhibited the eosinophil shape change induced by murine LPW and rat RP-HPLC LPW, suggesting the involvement of this receptor in LPW signaling (Figure 5B), even though eotaxin and RANTES seem not to be major mediators in the eosinophil activation and migration induced by LPW in vivo.

The involvement of CCR3 signaling in LPS-induced eosinophil migration was investigated in vivo. As shown in Figure 6, pretreatment with antimurine CCR3 mAb significantly inhibited the eotaxin-induced eosinophil influx to the pleural cavity, thus indicating the neutralizing activity of this antibody for in vivo treatment. Confirming the results obtained in vitro, pretreatment with anti-CCR3 dose-dependently inhibited the eosinophil accumulation induced by LPS in the pleural cavity. It should be noted that at the highest antibody dose used (30 µg/mouse) a complete inhibition of eosinophil influx was seen (Figure 6B).

View larger version (14K): [in this window] [in a new window]   Figure 6.   Effect of anti-muCCR3 (15 or 30 µg/cavity) or purified rat IgG i.p. treatment on the 24-h eosinophil accumulation induced by eotaxin (30 pmol/cavity) (A) or LPS (250 ng/cavity) (B) into mouse pleural cavity. Results are expressed as the mean ± SEM performed with at least six animals. * represents statistically significant differences (P  0.05) between control and stimulated groups, whereas + indicates differences between stimulated and treated groups.

Effect of the Neutralization of CC Chemokine on Eosinophil Shape Change Induced In Vitro

As shown in Figure 5C, preincubation of eotaxin with vCKBP substantially decreased the changes observed in eosinophil FSC-H after eotaxin stimulation in vitro. In contrast, vCKBP failed to inhibit the eosinophil shape change induced by LPW in vitro (Figure 5C), suggesting that LPW induces eosinophil activation through a mechanism independent of any vCKBP-inhibitable CC chemokines.

Effect of Neutralization of CC Chemokines by vCKBP on Eosinophil Accumulation In Vivo

Although the CC chemokine neutralization by vCKBP has been extensively shown in vitro, its role in chemokine neutralization during in vivo inflammatory reactions has been less explored. As shown in Figure 7A, the CC chemokine- binding protein vCKBP was able to inhibit eotaxin-induced eosinophil accumulation at 24 h by 80%.

View larger version (18K): [in this window] [in a new window]   Figure 7.   Effect of vCKBP (10 pmol/cavity) on eotaxin- (A) or ovalbumin (B)-induced eosinophil accumulation into mouse pleural cavity. Animals received a concomitant i.t. injection of vCKBP and the stimulus and analysis was performed within 24 h. Eotaxin (30 pmol/cavity) was injected into naive mice, whereas ovalbumin (12 µg/cavity) was administered into presensitized mice, as described in MATERIALS AND METHODS. Results are expressed as the mean ± SEM from at least six animals. Statistically significant differences between control and agonist-stimulated groups are indicated by an asterisk, whereas + represents differences between vCKBP-treated and -untreated groups.

Eosinophil accumulation observed during allergic reactions, such as lung allergic inflammation and asthma, requires the production of eotaxin and other CC chemokines (20, 23). Treatment with the CC chemokine-binding protein vCKBP significantly inhibited the increase in the eosinophil numbers triggered by antigen in the model of allergic pleurisy, suggesting that the production of CC chemokines is also crucial for eosinophil accumulation in this model (Figure 7B).

Administration of the CC chemokine-binding protein vCKBP was used to investigate the involvement of CC chemokines in the eosinophil accumulation induced by LPS in the pleurisy model. In support of our in vitro results, the concomitant treatment with vCKBP, even at doses 5 times higher than the one used to inhibit the eosinophil influx induced by allergen or eotaxin, also failed to inhibit the significant rise in eosinophil counts 24 h after i.t. injection of LPS (Figure 8A). This result suggests that the CC chemokines which can be inhibited by vCKBP are not crucial for the eosinophil migration during the inflammatory response to endotoxin. To confirm these data, LPW was incubated with vCKBP before injection into naive recipient animals. Incubation of LPW with the CC chemokine ligand protein failed to inhibit the LPW-induced eosinophil influx to the pleural cavity (Figure 8B). The vCKBP also failed to inhibit the neutrophil influx observed 4 h after i.t. injection of LPS (0.59 ± 0.09 × 10⁶ cells/cavity in untreated LPS- injected group versus 0.81 ± 0.07 × 10⁶ cells/cavity in vCKBP treated group; NS). By contrast, the administration of vCKBP inhibited the increase in T lymphocyte counts observed 24 h after LPS stimulation (Figure 8C). These results suggest that vCKBP-inhibitable CC chemokines are involved in T lymphocyte, but not eosinophil, recruitment in response to LPS.

View larger version (20K): [in this window] [in a new window]   Figure 8.   Effect of vCKBP (10 or 50 pmol/cavity) on the eosinophil accumulation induced by LPS (250 ng/cavity) (A) or LPW (B). (C) Effect of vCKBP on the T cell accumulation induced by LPS into mouse pleural cavity. Animals were concomitantly injected with vCKBP and the agonist. Results are expressed as the mean ± SEM from one of two separate experiments performed with at least six animals each. Statistically significant differences between control and agonist-stimulated groups are indicated by an asterisk, whereas + represents differences between vCKBP-treated and -untreated groups.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The mechanisms involved in LPS-induced eosinophil accumulation are not fully characterized. Our group and others have previously described that eosinophils are activated and migrate to inflammatory sites after LPS challenge through an indirect effect that involves newly synthesized proteins (11, 13, 24). In clear contrast to allergic and helmith-induced eosinophil trafficking, LPS-induced eosinophil recruitment is independent of IL-5 and CD4 T cell activation but requires macrophage- and T cell-secreted products (6, 14). In the present study, we investigated the role of CC chemokines on LPS-induced eosinophil accumulation in the mouse pleural cavity. We provide evidence that, although LPS can upregulate the local production and release of different CC chemokines, including the eosinophil-activating chemokines RANTES and eotaxin, they are not crucial for the LPS-induced eosinophil accumulation at the pleural cavity.

The i.t. injection of LPS induces an early influx of neutrophils into the pleural cavity of mice that is followed by a long-lasting accumulation of macrophages, lymphocytes, and eosinophils maximal within 24 to 48 h. LPS-induced eosinophil and lymphocyte accumulation could be mimicked by the transfer of heated pleural wash recovered 6 h after LPS stimulation (6, 13). It is important to note that heated LPW failed to induce neutrophil accumulation. To sort out the involvement of chemokines in this phenomenon, we analyzed the mRNA expression of different chemokines in the cells recovered from the mouse pleural cavity 6 h after the i.t. injection of LPS. We observed a remarkable increased expression of mRNA for all the chemokines tested, including MIP-1, MIP-1, MIP-2, MCP-1, IP-10, RANTES, and eotaxin, the latter two known to have significant stimulatory activity on eosinophils and lymphocytes. Our results are in agreement with previous reports showing that LPS induces an increased expression of mRNA for several CXC and CC chemokines in in vitro and in vivo murine models, although different profiles of chemokine expression have been observed depending on dose and cells/tissues involved (18, 19, 25). Increased mRNA expression for RANTES, MIP-1, and MCP-5 were observed in LPS-stimulated macrophages (18). Moreover, increased mRNA expression for CC chemokines in models of endotoxin or sepsis-induced pulmonary inflammation have been reported to correlate with polymorphonuclear and mononuclear cell influx to the lung (19, 26). In addition to the increased mRNA expression for eotaxin and RANTES, we observed significant increased levels of these chemokines by ELISA in the pleural wash after LPS induction. Interestingly, protein levels of eotaxin were three times higher than RANTES, although mRNA expressions for RANTES were much higher than that for eotaxin. This result may be explained by the finding that mesothelial cells are sources of CC chemokines, including eotaxin (27). If mesothelial cells are possibly important sources of eotaxin secreted in the pleural cavity during the LPS response, the failure to detect a high mRNA expression for eotaxin is explained by the fact that mesothelial cells are not recovered in the pleural wash fluid.

Despite apparent functional redundancies, structural homologies, and receptor overlaps, numerous reports indicate nonredundant roles for individual chemokines in controlling leukocyte trafficking during inflammatory conditions (15, 23). The production of CC chemokines during allergic responses is a pivotal event resulting in eosinophil accumulation and activation. Indeed, it has been demonstrated that both eosinophilic (eotaxin, RANTES, and MIP-1) and noneosinophilic chemokines (MCP-1) act in concert to induce eosinophil recruitment to allergic inflamed tissues (23, 28). Eotaxin is a potent and specific eosinophil chemoattractant, first detected in a guinea pig model of allergic airways inflammation (20). Several other reports place eotaxin as a key molecule in attracting eosinophils to sites of allergic and nonallergic inflammation. Indeed, eotaxin expression correlates with eosinophil migration in murine and guinea pig models of allergic lung eosinophilia and both Ab blockade of eotaxin in vivo and the targeted disruption of the eotaxin gene lead to reduced eosinophil infiltration after allergen challenge (29). Similarly, an important role for RANTES in mediating eosinophil influx in allergic inflammation was shown by inhibition of the eosinophil accumulation in mice induced by ovalbumin with met-RANTES or anti-RANTES neutralizing Abs (23, 32). In the present study the individual roles for RANTES and eotaxin in LPS-induced eosinophil recruitment were dissected by their blockade with Abs. We found that pretreatment of animals with anti-eotaxin Ab failed to inhibit LPS-induced eosinophil accumulation in the pleural cavity of mice, indicating a different mechanism for eosinophil influx in allergic and LPS-induced inflammatory responses. Moreover, incubation of LPW with anti-eotaxin pAb failed to inhibit the LPW-induced eosinophil shape change, under conditions in which it completely neutralized eotaxin-induced eosinophil shape change. In addition, in contrast to the eosinophilotactic protein present in LPW, eotaxin is not heat stable, losing its activity in eosinophil shape change assay when submitted to 100°C for 30 min (data not shown). The functional role for RANTES in cell accumulation seems also to be different between LPS-induced and allergic responses. Our results demonstrate that RANTES is not an important mediator for the eosinophil recruitment in the LPS response, since the in vivo neutralization of this chemokine does not affect eosinophil or lymphocyte accumulation. In addition, RANTES was not able to induce shape change in isolated eosinophils (data not shown), excluding its role in eosinophil activation induced by LPW. Interestingly, eotaxin and RANTES were not detected by ELISA in heated LPW, even though LPW was still able to activate and attract eosinophils, adding support to a role for an eosinophilotactic factor different from eotaxin and RANTES on the eosinophil influx observed in LPS-induced pleurisy.

Chemokines activate and induce cell migration via seven transmembrane receptors expressed on target cells that are coupled to specific G proteins (15, 34). -chemokine receptors are generally linked to G proteins of the G_i class, which can be demonstrated by the inhibition of chemokine-induced signaling events by Bordetella pertussis toxin. To better clarify the intracellular signaling mechanism of LPW on eosinophils, we investigated the involvement of G proteins on eosinophil activation induced by LPW. Both eotaxin- and LPW-induced eosinophil shape change was inhibited by the pretreatment with pertussis toxin, suggesting that LPW might use one or more seven transmembrane domain receptors coupled to G_i proteins. Chemokine receptors are constitutively expressed by some cells, whereas they can be inducible in other cell populations. CCR3 receptor for chemokines is highly expressed on eosinophils, and is thought to mediate most of the actions of CC chemokines on eosinophils (15, 17, 22). Moreover, CCR3 is crucial for eosinophil accumulation during allergic reactions (15). Blocking guinea pig or mouse CCR3 with specific Abs inhibits eosinophil accumulation in eotaxin-injected skin sites (35). Our data shows that blocking eosinophil CCR3 by preincubation with anti-CCR3 mAb drastically inhibited eotaxin- and also LPW-induced shape change. Most importantly, the LPS-induced eosinophil accumulation in vivo was completely suppressed by treatment with neutralizing anti-CCR3 mAbs. This suggests that eotaxin receptor is shared by the eosinophilotactic protein produced during LPS-induced inflammatory response.

Taken together, the inhibitory effect of anti-CCR3 mAb, both in vitro and in vivo, with the lack of inhibition by anti-eotaxin and anti-RANTES Abs would suggest a role for CC chemokines other than RANTES and eotaxin on LPS-induced eosinophil activation and accumulation into inflammatory sites. To better characterize the involvement of different CC chemokines as well as their association in the LPS reaction, we used a potent and selective general inhibitor of CC chemokines produced by poxvirus. The DNA viruses family of poxviruses can encode a variety of immunomodulatory proteins that subvert the chemokine/cytokine network of infected hosts (36). Recently, it has been demonstrated that some species of orthopoxviruses secrete a 35-kD protein that binds with high affinity to virtually all known CC chemokines, inhibiting their biologic activity by competitive inhibition of chemokine interaction with their respective cellular receptors on target cells (16, 37). Indeed, it has been shown that the 35-kD CC chemokine-binding protein (vCKBP), secreted by the vaccinia virus strain Lister, binds with different affinities to virtually all known CCR3 ligands, including human MCP-3, MCP-4, eotaxin-1, eotaxin-2, and RANTES, and rodent MCP-3, eotaxin-1, and RANTES (16, 37-39, and data not shown). Furthermore, the vCKBP inhibited in a dose-dependent manner the accumulation of eosinophils induced by eotaxin in vivo (16). In accordance with this result, we were able to inhibit the eotaxin-induced eosinophil influx into the mouse pleural cavity by the concomitant injection of vCKBP. Moreover, we demonstrated in the present work that vCKBP is also capable of inhibiting the eosinophil accumulation triggered by an allergic reaction, which has been previously characterized as involving a complex network of CC chemokines (23, 40). Interestingly, vCKBP failed to inhibit eosinophil accumulation induced by LPS and LPW in vivo, as well as eosinophil activation in vitro, even at doses five times higher than the one required for eotaxin or allergen inhibition, demonstrating contrasting roles for CC chemokines in the eosinophil influx observed during allergic and LPS-induced inflammation. Collectively our results indicate that LPS induces eosinophil activation through a mechanism independent of the vCKBP-inhibitable CC chemokine known so far. Recently, two CCR3 ligands were described, mu-eotaxin-2 and h-MEC (41, 42), but at present there is no data available for the binding affinity of vCKBP on these newly described CCR3 ligands, nor are there neutralizing antibodies available to these chemokines, which makes difficult the assessment of their role in the LPS-induced eosinophil infiltration. Therefore, at this point we cannot rule out a role for newly described CC chemokines that do not have the affinity for vCKBP assayed, and further experiments are needed to clarify the identity of the LPS-induced eosinophilotactic activity.

Beside their effect on eosinophils, CC chemokines are potent in inducing T lymphocyte subset recruitment (23). In contrast to the observations for eosinophils, vCKBP significantly inhibited the T lymphocyte influx induced either by LPS or LPW. This finding provides evidence for the involvement of CC chemokine in LPS-induced lymphocyte recruitment. Moreover, it also suggests that different molecules regulate the recruitment of lymphocytes and eosinophils induced by LPS. The CC chemokine(s) responsible for T lymphocyte influx induced by LPS are still under investigation. Although RANTES neutralization completely abolished the CD4 and T lymphocyte influx to the pleural cavity in the ovalbumin-induced allergic response (data not shown), it failed to inhibit T lymphocyte recruitment in LPS-injected animals.

In conclusion, data reported here suggest that LPS- induced T cell recruitment, as well as the eosinophil and lymphocyte influx triggered by allergic pleurisy, is mediated by one or more CC chemokines. In contrast, although eosinophil activation induced by factors released by LPS are dependent on CCR3, LPS-induced eosinophil accumulation does not depend on eotaxin, RANTES, or other vCKBP-inhibitable CC chemokines.