INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cytokine-induced neutrophil chemoattractant (CINC) is a rat cytokine chemoattractant that was first purified from medium conditioned with a cytokine-induced epithelial clone of normal rat kidney (1). It is within the family of the C-X-C chemokines, which in the rat include macrophage-inflammatory protein-2 (MIP-2). CINC shares the greatest sequence homology (69%) with the human C-X-C chemokine melanoma growth-stimulating activity/GRO (MGSA/GRO) and is less homologous with human interleukin-8 (IL-8) (47%) (2). The production of CINC is enhanced by the cytokines IL-1 and with tumor necrosis factor- (TNF-) (1). CINC has been shown to be a most potent inducer of neutrophil chemotaxis and infiltration in vitro and in vivo (3, 4), and may therefore be involved in the evolution of acute neutrophilic inflammation.

Because of the potent chemoattractant effects of CINC for rat neutrophils, we determined the potential role played by CINC in ozone-induced effects on the lung. Ozone is an important component of the photochemical oxidative products of air pollution involving substrates emitted from automobile engines. Experimental exposure to ozone induces airway inflammation in the form of neutrophil influx, airways obstruction, and airway hyperresponsiveness to bronchoconstrictor agents in both humans and animals (5). The mechanisms underlying the influx of neutrophils into the airways are unclear. Leukocyte accumulation in vivo is mediated by a series of adhesive interactions between circulating leukocytes and vascular endothelial cells (9), and by the local generation of chemoattractants. For example, a blocking antibody to a leukocyte adhesion molecule (Mo-1) in the dog inhibited neutrophil influx in the lungs (10), indicating the importance of such adhesion molecules. However, C-X-C chemokines have been shown to be also expressed in the lung following ozone exposure, as in the case of CINC in the rat and MIP-2 in the mouse (11, 12), although the contribution of these chemokines to neutrophil infiltration after ozone exposure is unknown. Studies of other models in the rat have indicated that CINC can contribute to the lung neutrophil influx following endotoxin and IL-1 exposures (13) and to the neutrophil influx observed in immune-complex glomerulitis in the rat (16). We therefore determined the role of CINC in ozone-induced airway neutrophilia by examining the effect of ozone on CINC messenger RNA (mRNA) expression and the effect of an anti-CINC antibody on ozone-induced neutrophil influx. Because ozone also induces bronchial hyperresponsiveness (BHR), we also investigated the link between neutrophil influx and the development of BHR in this model.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Protocol

Virus-free inbred male Brown-Norway (BN) rats weighing 200 to 250 g (Harlan Olac, Bicester, Oxon, UK) were used. They were kept in a special caging system with its own air circulation (Maximizer; Thorens Caging Systems Inc., Hazelton, PA). We studied the effect of an anti-CINC antibody in rats exposed to either 1 ppm or 3 ppm ozone for 3 h.

Exposure to 1 ppm ozone. Animals were divided into five groups as follows in order to investigate the effects of intratracheal and intravenous administration of anti-CINC antibody: (1) Naive animals exposed to filtered laboratory air (n = 6). (2) Intrathecal controls (n = 7). These animals were given control goat IgG (230 µg in 100 µl) intratracheally 15 min before exposure to ozone. (3) Intratracheal anti-CINC group (n = 7). These animals were given a goat anti-CINC polyclonal antibody (17) (230 µg in 100 µl) 15 min before exposure to ozone. This dose was chosen on the basis of two previous studies (13, 15). (4) Intravenous control group (n = 5). Animals were injected with control goat IgG (1 mg in 430 µl) via a tail vein 15 min before exposure to ozone. (5) Intravenous anti-CINC group (n = 8). Animals were injected with anti-CINC antibody (1 mg in 430 µl) via a tail vein 15 min before exposure to ozone. This dose was chosen on the basis of a previous study (16).

Exposure to 3 ppm ozone. Rats were divided into four groups as follows in order to study the effect of intravenous administration of anti-CINC antibody: (1) Naive rats exposed to filtered laboratory air (n = 6). (2) Rats exposed to ozone without pretreatment (n = 6). (3) Intravenous control group (n = 7). Animals were injected with control IgG (1 mg in 430 µl) via a tail vein 15 min before exposure to ozone. (4) Intravenous anti-CINC antibody group (n = 8). Animals were injected with anti-CINC antibody (1 mg in 430 µl) via a tail vein 15 min before exposure to ozone.

Rats were studied 24 h after exposure to ozone because in preliminary studies we found that neutrophil recovery in bronchoalveolar lavage fluid (BALF) was maximal by 12 to 24 h, with BHR peaking by 6 h but maintained at 24 h.

Drug Administration

Intratracheally administered drugs were given through a nylon cannula (1.0 mm O.D. and 30 mm long) placed in the trachea under sedation from a subcutaneous injection of 0.4 ml/kg Hypnorm (0.315 mg/ ml of fentanyl citrate and 10 mg/ml of fluanisone). Rats also received intravenous injections of the anti-CINC antibody under similar sedation with Hypnorm.

Ozone Exposure

Ozone was generated by passing laboratory air through a Sander Ozonizer (Model 500; Erwin Sander GmbH, Uetz-Eltz, Germany). The output (0.2 L/min) was diluted with compressed air (8 L/min) controlled by gas flow-meter (Platon Flow Control, Basingstoke, UK) and fed into a purpose-designed Perspex box (60 × 25 × 20 cm). The concentration of ozone was determined by means of specific gas-sampling tubes (Dräger Tube; Drägerwerk, Aktiengesellschaft Lübeck, Lübeck, Germany) and maintained at 1 or 3 ppm by regular measurement at the output port of the box. Rats were placed in the box for 3 h.

Anti-CINC Antibody

The goat anti-CINC polyclonal antibody was prepared and purified according to the method previously described by Wittwer and coworkers (17). The antibody has been tested for specificity against MIP-2 and CINC-2 in an assay involving in vitro neutrophil chemotaxis. In the chemotaxis assay, the activity of 10 ng /ml of CINC is completely blocked by 1 µg/ml of the antibody, whereas the activity of 10 ng /ml of CINC-2 or MIP-2 is blocked by 30 to 100 µg /ml of the antibody. The anti-CINC antibody was therefore selective for CINC, but not totally specific.

RNA Extraction and Northern Blot Analysis

Three control (air-exposed) and three ozone-exposed (1 ppm for 3 h) animals were killed with a lethal dose of pentobarbital (Expiral, 200 mg/kg; Sanofi Animal Health Ltd, Herts, UK) at 2 h, 8 h, and 24 h after exposure. Lungs were rapidly removed, snap-frozen in liquid nitrogen, and stored at 80° C until used. Total cellular RNAs were isolated according to the method of Chomzynski and Sacchi (18). Poly (A)⁺ RNA was prepared with the PolyTract mRNA Isolation System kit (Promega, Southhampton, UK) according to the manufacturer's instructions. Samples of mRNA were size-fractionated on a 1% agarose/formaldehyde gel containing 20 mM morpholinosulfonic acid (MOPS), 5 mM sodium acetate, and 1 mM ethylenediamine tetraacetic acid (EDTA) (pH 7.0), and were blotted onto Hybond-N filters (Amersham, Buckinghamshire, UK) by capillary action using 20× standard saline citrate (SSC), 1× SSC, 0.15 mM NaCl, and 0.015 M sodium citrate at pH 7.0. Random primer labeling was done with a 205-bp CINC complementary DNA (cDNA) and 1.3-kb PstI fragment from glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA using [-³²P]dCTP (3,000 Ci /mmol). Prehybridization and hybridization were done at 42° C with labeled probes (approximately 1.5 × 10⁶ cpm/ ml) in buffer containing 50% formamide, 50 mM Tris-HCl (pH 7.5), 5× Denhardts solution, 0.1% sodium dodecyl sulfate (SDS), 5 mM EDTA, and 250 µg/ml denatured salmon sperm DNA. Following hybridization, the blots were washed to a stringency of 0.1× SSC/0.1% SDS for 30 min at 60° C and exposed at 80° C for 1 to 7 d to X-OMAT S film (Kodak, Rochester, NY). Autoradiographic bands were quantified by laser densitometry (PDI, New York, NY). To account for possible differences in the loading or transfer of RNA, data were expressed as CINC/GAPDH ratios. The CINC and GAPDH cDNAs were prepared from a Brown-Norway rat lung treated with Escherichia coli lipopolysaccharide (LPS), using the reverse transcriptase- polymerase chain reaction (RT-PCR) as described previously (11).

Measurement of Airway Responsiveness to Acetylcholine

Airway responsiveness was measured 18 to 20 h after exposure to ozone. Rats were anesthetized with an initial dose of 80 mg/kg pentobarbital sodium injected intraperitoneally. Additional pentobarbital was administered as required to maintain adequate anesthesia. A tracheal cannula (10 mm in length and of 1.3 mm I.D.) was inserted into the lumen of the cervical trachea through a tracheostomy, and was tied snugly with suture material. A polyethylene catheter was inserted into the left carotid artery to monitor blood pressure and heart rate with a pressure transducer. The right external jugular vein was cannulated for administration of intravenous drugs and fluids. Animals were then connected to a small-animal respirator (Harvard Apparatus, Ltd, Edenbridge, Kent, UK) and ventilated with air at 10 ml/kg at a rate of 90 strokes/min. Transpulmonary pressure was measured with a pressure transducer (Model FCO 40; ± 1,000 mm H₂O; Furness Controls, Ltd, Bexhill, Sussex, UK) with one side attached to an air-filled catheter that was inserted into the right pleural cavity and the other side attached to a catheter connected to a side port of the intratracheal cannula. The ventilatory circuit had a total volume of 20 ml. Airflow was measured with a pneumotachograph (Model F1L; Mercury Electronics, Ltd, Glasgow, UK) connected to a transducer (Model FCO 40; ± 20 mm H₂O; Furness Controls, Ltd). The signals from the transducers were digitized with a 12-bit analog-to-digital board (NB-MIO-16; National Instruments, Austin, TX) connected to a Macintosh IIcx computer (Apple Computer Inc., Cupertino, CA) and analyzed with software (LabView; National Instruments) that was programmed to instantaneously calculate lung resistance (RL) according to the method of von Neergard and Wirz (19). Transpulmonary pressure and mean blood pressure were also monitored throughout the experiments. Acetylcholine (ACh) aerosol was generated with an ultrasonic nebulizer (Model 2511; PulmoSonic, DeVilbiss Health Care Inc., Hazleton, PA). The mean mass diameter of the aerosol was 3.8 µm, with a geometric standard deviation of 1.3 µm, as measured with a laser droplet and particle analyzer (Model 2600C; Malvern Instruments, Derbyshire, UK).

Animals were initially injected with propranolol (1 mg/kg intravenously) to inhibit adrenergic effects, and with suxamethonium (1.5 mg/kg) to stop spontaneous breathing. A dose of inhaled saline solution was administered for 45 breaths, and the subsequent RL value was used as baseline for subsequent ACh administration. Starting 3 min after saline exposure, increasing half-log₁₀ concentrations of ACh were administered by inhalation (45 breaths), with the initial concentration set at 10³ mol/L. Increasing half-log concentrations were administered at 5-min intervals, with one hyperinflation of twice the tidal volume (VT) applied between each ACh concentration, which was accomplished by manually blocking the outflow of the ventilator. The challenge was stopped when an increase in RL that exceeded the initial baseline by more than 200% was obtained. The logPC_200ACh, which is the log₁₀ transformation of the provocation concentration of acetylcholine producing a 200% increase in RL, was calculated by log-linear interpolation of the concentration-response curves.

Bronchoalveolar Lavage and Cell Counting

After measurement of airway responsiveness, rats were given a lethal dose of pentobarbital (200 mg/kg intravenously), and the lungs were lavaged 10 times with 2-ml aliquots of saline solution through the tracheostomy. The lavage fluid was centrifuged (500 × g for 10 min at 4° C) and the cell pellet was resuspended in 1 ml of Hanks' balanced salt solution (HBSS). Total cell counts were made by adding 10 µl of the cell suspension to 90 µl of Kimura stain, with cell counting in a Neubauer chamber (American Optical Corp., Southbridge, MA) under a light microscope. Differential cell counts were made on cytospin preparations, which were provided by centrifuging at 300 rpm for 5 min and staining with May-Grünwald stain. Cells were identified as macrophages, neutrophils, eosinophils, lymphocytes, basophils, and epithelial cells according to standard morphology. Five hundred cells were counted under ×400 magnification, and the percentage and absolute number of each cell type were calculated.

Drugs

We used the following agents and drugs: acetylcholine (Sigma Chemical Co., Poole, Dorset, UK), propranolol (Inderal; ICI, plc Macclesfield, Cheshire, UK), suxamethonium (Antigen Pharmaceuticals Ltd, Roscrea, Ireland), pentobarbital sodium (60 mg/ml; Sagatal; May & Baker, Ltd, Dagenham, UK), pentobarbital sodium (200 mg/ml for euthanasia; Expiral), midazolam (Hypnovel; Roche Products Ltd, Herts, UK), and fentanyl citrate and fluanisone (Hypnorm; Jannssen Pharmaceuticals Ltd, Oxon, UK).

Data Analysis

PC_200ACh data were log₁₀-transformed and reported as geometric means. Nonparametric analysis of variance (ANOVA, Kruskal-Wallis method) was used to determine significant variance among the groups. We used the Mann-Whitney U test to analyze for significant differences, between individual groups, and a value of p < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CINC mRNA Expression

Figure 1 summarizes the mean CINC/GAPDH ratios. Northern blot analysis revealed a significant increase in steady-state levels of CINC mRNA at 2 h after 1 ppm ozone exposure for 3 h, whereas the CINC mRNA signal was barely detectable in RNA obtained from control animals exposed to filtered laboratory air. The increase in CINC mRNA was still significant at 8 h after exposure, but it returned to baseline values by 24 h.

View larger version (44K): [in this window] [in a new window]   Figure 1.   Effect of air and ozone exposures on steady-state levels of CINC mRNA expression in BN rat lungs. Lung mRNA was hybridized sequentially to CINC and GAPDH cDNA, the latter used as an internal control. The upper panel shows a representative Northern blot analysis of CINC and GAPDH mRNA in oxone-exposed rats at 2 h, 8 h, and 24 h after exposure. The lower panel shows the mean ratio of CINC to GAPDH mRNA as measured with laser densitometry for three rats in each group. **p < 0.01 and *p < 0.05 compared with air-exposed rats.

Cell Counts in BALF

The recovery rates of BALF were similar in all groups (approximately 90% of instilled fluid; data not shown). Compared with that for naive animals, the number of neutrophils recovered in BALF was increased after 1 ppm ozone exposure in both the intrathecal control and intravenous control groups. Intravenous anti-CINC antibody pretreatment completely inhibited the 3.8-fold greater neutrophil influx into airways of treated animals (p < 0.01 compared with control animals, but not significantly different from naive animals), whereas intratracheal administration resulted in only a partial inhibition (p < 0.05 compared with naive animals). There were no significant differences in the neutrophil recoveries for the control groups treated with intravenous or intrathecal goat IgG. There were no significant differences in other cell counts between groups (Table 1).

After exposure to 3 ppm ozone, there was a 5.5-fold increase in neutrophil recovery (p < 0.05), which was not significantly affected by the intravenous-goat IgG pretreatment. However, intravenous anti-CINC antibody administration significantly inhibited ozone-induced neutrophil recovery (Table 1).

Airway Responsiveness to ACH

Exposure to 1 ppm ozone for 3 h caused significant BHR to ACh in the groups of animals treated with control goat IgG as compared with naive animals. Neither intravenous nor intratracheal administration of anti-CINC antibody showed an inhibitory effect against ozone-induced BHR (Figure 2). Similar results were obtained with rats exposed to 3 ppm ozone (Figure 2).

View larger version (34K): [in this window] [in a new window]   Figure 2.   Effect of anti-CINC antibody on ozone-induced BHR to Ach, measured as PC200, which is the concentration of Ach needed to increase baseline resistance by 200%. The upper panel shows the data regarding exposure to 1 ppm ozone (O3) for 3 h. Administration of anti-CINC antibody intratracheally (IT CINC) or intravenously (IV + CINC) did not alter BHR as compared with the control groups that received goat IgG (IT IgG and IV IgG). The lower panel shows the data after exposure to 3 ppm ozone for 3 h. There was no difference in PC200 values for ozone-exposed rats treated with anti-CINC antibody intravenously (O3 + CINC) or with control IgG intravenously (O3 + IgG). *p < 0.05) compared with animals exposed to air.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have shown that ozone (1 ppm for 3 h) induces a marked increase in lung CINC mRNA expression within 2 h after exposure, and that an anti-CINC antibody is effective in inhibiting neutrophil influx induced after exposure to ozone at 1 or 3 ppm for 3 h. The higher intravenous dose was more effective than the lower intratracheal dose administered. However, despite the reduction in ozone-induced neutrophil counts, there was no effect on ozone-induced BHR. Our studies therefore indicate that CINC plays an important role in neutrophil chemoattraction following ozone exposure.

CINC and MIP-2 are two of the most recent members of the C-X-C chemokine family with neutrophil chemoattractant properties to be isolated in the rat (3, 20), and both CINC and MIP-2 mRNA are rapidly expressed in rat lung following ozone inhalation (12, 21). We have previously shown that exposure to a higher concentration of ozone (3 ppm for 3 h) led to a large expression of CINC and MIP-2 mRNA in the lungs of rats (11, 21), and we now confirm that this can also occur at a lower concentration of ozone exposure. In mice, only concentrations of ozone of 1 ppm or above were found to enhance the expression of MIP-2 mRNA in the lungs (12). The alveolar macrophage appears to be one of the cells capable of expressing the genes for both CINC and MIP-2 following ozone exposure (11, 12, 21).

The mechanisms by which CINC expression is upregulated following ozone exposure are unclear, but may be direct or indirect. Proinflammatory cytokines such as IL-1 and TNF- have been shown to stimulate CINC production by epithelial cells in vitro and also in vivo when the cytokines are administrated intratracheally (15, 22). In the rat, ozone exposure leads to an enhancement of IL-1 and TNF- expression in alveolar macrophages (23), and it is therefore conceivable that these cytokines may be the stimuli for CINC expression following ozone exposure. This possibility is supported by the observation that an anti-CINC antibody can block IL-1-induced inflammation in the lung (13, 15). Alternatively, ozone, as a powerful oxidizing agent, may directly induce an increase in the binding of the regulatory transcription factor nuclear factor kappa-B (NF-B) (24) to the NF-B motif of the promoter region of the CINC gene (25), as we have previously observed (11). However, TNF- and IL-1 can also induce NF-B (26, 28), and it is likely that both direct and indirect mechanisms are involved in the upregulation of CINC expression by ozone.

Our data indicate that CINC mediates the influx of neutrophils into the lungs and airways following ozone exposure. Intratracheal administration of recombinant rat CINC to rats has been shown to induce dose-dependent neutrophil influx into the airspaces (29). Other chemokines, such as MIP-2 or adhesion molecules, may contribute to ozone-induced neutrophilia. It is of interest to note that CINC itself can increase the expression of CD11b/CD18 (Mac-1) integrin on rat neutrophils (29). Apart from ozone exposure, CINC also appears to be involved in neutrophil infiltration following exposure to endotoxin and IL-1 in the lung (13, 15, 29). The participation of CINC is also not tissue-specific, because an anti-CINC antibody inhibits neutrophil infiltration in immune-complex- induced glomerulonephritis (16). Recent studies with an anti-human IL-8 antibody in rats have supported a role for an IL-8-like molecule in mediating the migration of neutrophils in acute inflammation (30). However, it is unlikely that the effect of the anti-CINC antibody we used could be attributed to the participation of an IL-8-like CINC homologue. The anti-CINC antibody did not cross-react significantly with human IL-8 or granulocyte chemoattractant (GRO), which are the close human homologues of CINC (17, 31). In addition, we used an anti-human IL-8 antibody in three rats exposed to ozone and did not observe any inhibition of ozone-induced neutrophilia (unpublished data). We cannot rule out the contribution of MIP-2 to ozone-induced neutrophil influx, because although the anti-CINC antibody we used was quite selective for CINC, it was not entirely specific, as assessed with in vitro neutrophil chemotaxis assays. However, the degree of selectivity of the anti-CINC antibody, and the suppression of the neutrophil influx observed, indicate that CINC is likely to contribute more strongly than MIP-2 to ozone-induced neutrophil influx.

The increase in bronchial responsiveness induced by ozone was still present when neutrophil influx in BALF was prevented by the anti-CINC antibody, indicating that ozone-induced BHR is unlikely to be secondary to neutrophil migration into the lungs. The temporal relationship between the presence of neutrophils and the induction of BHR has been the subject of many studies (5, 10, 32). A temporal dissociation between these two events can be demonstrated, such as by accompaniment of the remission of ozone-induced BHR by neutrophil infiltration in the airways of the guinea-pig (5), and the induction of ozone-induced BHR in the absence of tracheal neutrophils in the rat (32). Moreover, depletion of neutrophils by cyclophosphamide in the guinea-pig (33), or by a monoclonal antibody to a leukocyte adhesion molecule, did not attenuate ozone-induced BHR (10), and depletion of neutrophils with hydroxyurea did not inhibit ozone-induced BHR in the dog (34). Therefore, BHR may be induced in the presence or absence of neutrophils, indicating that their presence is not essential for the development of BHR. Other mechanisms, such as those involving activating on of neural pathways, may be involved in the development of BHR (35).