INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Intratracheal instillation of lipopolysaccharide (LPS) produces a well-characterized model of acute lung inflammation, leading to the activation of alveolar macrophages and massive tissue infiltration of neutrophils (1). The stimulated leukocytes secrete increased amounts of oxygen free radicals, arachidonic acid metabolites, and various inflammatory mediators such as interleukin-1, proteases, and tumor necrosis factor (2, 3). These mediators may lead to epithelial lesions and increase the alveolocapillary barrier permeability.

Although the response of different inflammatory cell types to LPS has been well characterized, little is known about the changes in bronchiolar epithelium in animals challenged with LPS. In a morphologic study of the acute pulmonary lesions induced by LPS in rats, Ooi and colleagues (4) described in the bronchiolar epithelium alterations of Clara cells, consisting in a pseudostratified rearrangement and an enhanced production of secretory granules. These observations suggest that Clara cells are involved in the development of inflammatory processes. The possible link between the Clara cell and inflammation could be the 10-kD Clara cell protein (abbreviated as CC16, CC10, or CCSP), which is the major constituent of secretory granules of Clara cells. Although the exact biologic role of CC16 is still unclear, there is growing evidence from in vitro and in vivo studies suggesting that this protein plays an important protective role in the lung as an anti- inflammatory agent (5).

Independent of its biologic function, CC16 is now recognized as a new peripheral sensitive marker of lung injury, allowing researchers to evaluate the integrity of the air-blood barrier or, when the latter is intact or slightly compromised, the integrity/number of Clara cells. For instance, CC16 has been found to increase in serum of humans acutely exposed to lung irritants such as smoke from open fires (8) as well as in patients with acute lung injury, pulmonary fibrosis, or sarcoidosis (6, 9). By contrast, decreased levels of CC16 in serum and BALF have been observed in subjects chronically exposed to Clara cell toxicants such as tobacco smoke (9).

The purpose of this study was to characterize the response of Clara cells and CC16 to LPS-induced lung injury. The Clara cell density, the CC16 gene expression in the lung, and the CC16 concentrations in lung homogenate, BALF, and serum were determined and compared with indices of lung inflammation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Treatments

Adult male Sprague-Dawley rats (B&K Universal Limited, North Umberside, UK), weighing approximately 250 to 300 g were used in this study. The animals were housed in an air-conditioned room (25° C, 50% relative humidity) with a 12-h light/dark cycle. Rats were anesthetized by an intraperitoneal injection of sodium pentobarbital (50 mg/kg) and lung injury was induced by intratracheal instillation of LPS (Escherichia coli 055:B5; Sigma Chemical, St. Louis, MO) at doses of 10, 100, or 200 µg/100 g body weight. Twenty-four hours after LPS instillation, the rats were killed and a bronchoalveolar lavage (BAL) was performed. In the kinetic study, rats were instilled with 100 µg LPS and BAL was performed at 3, 6, 24, 48, 72, and 168 h after LPS challenge. LPS was dissolved in 200 µl of sterile phosphate-buffered saline (Dulbecco's PBS; Gibco BRL, Life Technologies LTD, Paisley, UK). Control rats were instilled with 200 µl PBS. Dexamethasone (Sigma) was administered intraperitoneally at a dose of 2 mg/kg, 2 h before LPS instillation (100 µg).

Bronchoalveolar Lavage

Bronchoalveolar lavage was performed as previously described (10). Briefly, the animals were anesthetized intraperitoneally with sodium pentobarbital (50 mg/kg) and the lung was lavaged with normal saline using a total volume of 20 ml. The BAL fluid (BALF) was centrifuged (1,000 g for 10 min at 4° C) and the resultant cell-free supernatant was analyzed for the different biochemical parameters. The cell pellet was resuspended in 1 ml saline. Aliquots of the cell suspension were used to determine the cell number and viability (trypan blue exclusion) using a hemocytometer. Slides were prepared using a cytospin 3 (Shandon Ltd, Runcorn, Cheshire, UK) for the determination of cell differentials after staining using Diff-Quik (Dade, Brussels, Belgium). Before BAL, whole blood was collected by cardiac puncture and stored at 4° C for 3 h. After centrifugation of the clotted blood at 2,000 g for 10 min, the sera were removed and stored at 80° C until analysis.

Biochemical and Immunologic Assays

Lactate dehydrogenase (LDH) activity in BALF was determined immediately after sampling by monitoring, at 340 nm, the reduction of NAD⁺ (5 mM; Sigma) in the presence of lactate (20 mM in 20 ml of 0.25 M TRIS/HC1 buffer at pH 9; Sigma). Total protein in BALF was assayed spectrophotometrically with a Technicon RA-1000 (Bayer, Brussels, Belgium). Rat albumin was determined by BALF by an automated latex immunoassay (11) using the antibody from USB (Cleveland, OH) and, as standard, purified rat albumin from Sigma. CC16 concentrations in BALF and serum were measured by an automated latex immunoassay using rabbit antibodies against purified rat CC16 as previously described (12). The performances of the assay in various rat biologic fluids, including BALF and serum, are similar to that of the immunoassay of human CC16 (13, 14).

Immunohistochemistry

Lungs were fixed in 4% formaldehyde or in Bouin's fluid and embedded in Paraplast Plus paraffin. Serial 5-µm sections were obtained from blocks and mounted on silane-coated glass slides, deparaffinized with graded alcohol, and processed for immunohistochemistry. Cells containing CC16 were localized in lung sections with a rabbit polyclonal antibody raised against rat CC16 using a slightly modified streptavidin-biotin immunoperoxidase method (15). Briefly, before immunostaining, rehydrated sections were immersed in 0.4% hydrogen peroxide for 5 min and pretreated by incubation in a wet autoclave at 120° C for 20 min for antigen retrieval (16). The pretreated sections were washed in PBS buffer and incubated in 5% normal goat serum (NGS) in PBS for 20 min. Slides were then rinsed and incubated sequentially at room temperature in the following solutions: (1) rabbit polyclonal antirat CC16 (diluted 1:4,000) for 1 h (2) biotinylated goat antirabbit IgG (diluted 1:50) for 20 min, and (3) ABC complexes for 20 min. Bound peroxidase activity was visualized by incubation with 0.02% 3,3'-diaminobenzidine-0.01% H₂O₂ in PBS. The different solutions were prepared in 5% NGS-PBS buffer (pH, 7.4) and after each step of the immunostaining procedure the sections were rinsed in the same buffer. The sections were finally counterstained with hemalum and luxol fast blue, dehydrated, and mounted with a permanent mounting medium.

Control of the specificity of immunolabeling included the omission of the primary antibody or the substitution of the primary antibody with nonimmune serum. In each case, these assays confirmed the specificity of the observed immunolabeling. In addition, we checked that the antiserum used in this study gave clear-cut immunostaining patterns on control lung sections, i.e., labeling restricted to Clara cells.

Morphometric analysis was performed by a computer-assisted approach as described previously (17). Briefly, the procedures utilized a hardware consisting of a high-resolution color video camera (JVC Ky-15) mounted on a Zeiss Axioplan microscope and connected to an IBM-compatible PC (Compaq 80386). The software was specifically designed for color analysis and morphometry (SAMBA system: Systéme d'Analyses Microscopiques à Balayage Automatique; Alcatel TITN Answer, Grenoble, France). The analysis of CC16-immunoreactive cells was performed on one lung section per experimental animal by scanning at magnification ×200 10 bronchiolar profiles (terminal bronchioles) taken at random in the tissue section. This represented a total scanned surface of approximately 40,000 µm² of bronchiolar epithelium. After interactive selections to define the bronchiolar surface to be analyzed, the area of the bronchiolar epithelium was automatically measured with an appropriate program. For each animal, the total epithelial area was obtained by pooling the values recorded for each of the 10 bronchiolar profiles. The number of CC16 positive epithelial cells was also recorded for each bronchiolar profile and these values pooled for one section. The density of CC16 immunoreactive cells was then calculated for each animal and expressed as the number of positive cells per 100 mm² of bronchiolar epithelium. Individual data were pooled per experimental group.

RNA Extraction and Northern Blot Analysis

Rat lungs were collected at the indicated time and total lung RNA was isolated with a one-step guanidium-phenol-cholorform extraction procedure using Trizol Reagent (Gibco BRL, Merelbeke, Belgium). RNA was separated by electrophoresis on 1% agarose and transferred onto a Hybond-N membrane for analysis (Amersham Life Science, Buckinghamshire, UK). Membranes were prehybridized and hybridized using Express Hyb hybridization solution according to the instructions of the manufacturer (Clontech Laboratories Inc., Palo Alto, CA).

Probes were generated by PCR starting with clones rat CC16 cDNA or rat -actin into PCR 2.1 plasmid (Invitrogen, San Diego, CA). The specific primers used for the amplification of the 375 pb CC16 probe were 5'-GGGCTTTAGCGTAGAATATCT-3' and 5'-CATCAGCCCACATCTACAGAC-3'. To amplify the 572 bp rat -actin probe, the specific primers 5'-CCATCCTGCGTCTGGACCTG-3' and 5'-CTCATCGTACTCCTGCTTGC-3' were used. The probes were labeled with the multiprime DNA labeling system (Amersham) using (-³²P) dCTP (specific activity > 3,000 Ci/mmol/L; Amersham) according to manufacturer's instructions. The membrane was then washed three times at room temperature for 40 min in 2× SCC, 0.05% SDS and for 40 min at 50° C in 0.1× SSC, 0.1% SDS. The membrane was exposed to X-OMAT film (Kodak) with intensifying screens at 80° C. The same blots were hybridized with CC16 probe and then rehybridized with -actin probe after stripping. mRNA expression was determined with a phosphorimager and accompanying ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Values are expressed as the ratio of CC16/ actin mRNA.

Statistical Analysis

Results are expressed as means ± SE. Differences between groups were assessed by one-way analysis of variance (ANOVA) followed by Dunett's multiple comparison test. All statistical tests were performed on log-transformed data; p values below 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dose-Effect Analysis

Effects of LPS on CC16 expression were examined in rats instilled with LPS at doses of 10, 100, or 200 µg/100 g body weight. Twenty-four hours after LPS challenge, the total cell number and the concentrations of LDH, albumin, and total protein in BALF were increased in all treatment groups (Table 1). The increased cell influx in BALF caused by LPS was dose-dependent until the dose of 200 µg, whereas the albumin and total protein leakage in BALF peaked at the dose of 100 µg LPS. The elevation of LDH was not related to the dose.

The CC16 concentrations in BALF, by contrast, were markedly reduced, averaging only 40, 28, and 25% of control values at doses of 10, 100, and 200 µg, respectively (Figure 1a). In serum, the CC16 concentrations showed an opposite pattern, increasing in parallel with the elevation of albumin in BALF, with a maximum at the dose of 100 µg (Figure 1b).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Changes of CC16 concentrations in BALF (a) and in serum (b) 24 h after instillation with LPS at doses of 10, 100, or 200 µg/100 g body weight. Mean ± SE, n = 6; *p < 0.05.

By immunohistochemistry, CC16 was localized in nonciliated cells lining the terminal bronchioles and corresponding to Clara cells (Figure 2a). No staining was observed in alveolar tissue or in ciliated epithelial cells of the airways. After LPS instillation, the abundance and the intensity of immunoreactive cells were reduced (Figure 2b). Morphometric analysis showed, however, that the density of cells containing CC16 was significantly reduced only at the dose of 200 µg LPS (Figure 3). On the basis of these observations, the dose of 100 µg LPS was selected as that eliciting the highest response of CC16 and of other lung inflammation markers without changes in the number of Clara cells.

View larger version (89K): [in this window] [in a new window]   View larger version (94K): [in this window] [in a new window]   Figure 2.   Immunohistochemical localization of CC16 in Clara cells in the terminal bronchioles. (a) Control rat (original magnification: ×320). (b) Rat instilled with LPS (200 µg/100 g body weight) 24 h before being killed (original magnification: ×250).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Density of Clara cells in the bronchiolar epithelium of control rats and rats instilled with LPS at doses of 10, 100, or 200 µg/100 g body weight. The animals were killed 24 h after LPS challenge. Mean ± SE, n = 4; *p < 0.05.

Time Course Study

The kinetics of changes in the levels of CC16 in BALF, lung homogenate, or serum and of other lung injury indices was studied in rats instilled with 100 µg LPS and examined 3, 6, 24, 48, 72, and 168 h after LPS challenge. The total cell influx as well as the percentage of PMN in BALF showed an almost immediate response after LPS challenge, increasing already after 3 h to reach maximal values at 6 h. This inflammatory response progressively resolved to disappear after Day 3 (total cell count) or Day 4 (PMN). As revealed by the cytologic analysis, the increased cell recovery in BALF was essentially due to an infiltration of neutrophils. The concentration of albumin and total protein in BALF increased rapidly after LPS challenge like inflammatory cells, but the peak values were observed later, around Day 2. The LDH level was significantly increased between Days 1 and 3 post-instillation (Table 2).

The CC16 concentration in BALF was significantly decreased 6 h after LPS injection and reached a minimum after 48 h before gradually returning to control levels (Table 3). A similar time course was observed in the CC16 content of lung homogenate, although in that case the reduction was already significant after 3 h. It is of interest to note that CC16 levels in BALF and lung homogenate were reduced to a very similar extent. In serum, the increase of CC16 was already maximal 3 h after LPS instillation and the elevation remained until Day 3. 

Northern Blot Analysis

To determine whether the LPS-induced reduction of CC16 in lung and BALF was mediated by a downregulation of CC16 gene expression, we compared the mRNA levels for CC16 and the housekeeping gene -actin between control and treated rats at different times after the LPS challenge (100 µg). After LPS administration, CC16 mRNA levels decreased continuously until Day 2 after which they progressively returned to control levels (Figure 4a). This effect was not observed with the housekeeping gene -actin, thus indicating that it was not a general effect of LPS on gene expression. This was confirmed by the ratio of CC16 to -actin mRNA values, which was significantly decreased at 6 and 24 h (Figure 4b).

View larger version (27K): [in this window] [in a new window]   Figure 4.   Time course of CC16 and -actin mRNA levels in the lung of control and LPS-treated rats. (a) Northern blotting. (b) Ratios of the CC16/-actin densitometric signals. Mean ± SE, n = 3; *p < 0.05.

Effects of Dexamethasone Pretreatment

Pretreatment by dexamethasone abolished almost completely the inflammatory response to LPS as evidenced by the cell infiltration and the albumin leakage in BALF. Dexamethasone, however, failed to prevent the LPS-induced changes in CC16 levels in BALF and in serum (Figure 5). Administered alone, dexamethasone had no effect on the levels of albumin, CC16, and total cells in BALF or on the concentration of CC16 in serum.

View larger version (25K): [in this window] [in a new window]   Figure 5.   Effect of dexamethasone pretreatment on the LPS-induced changes of CC16 concentration in serum and in BALF and of the influx of cell and albumin in BALF. Dexamethasone was administered intraperitoneally (2 mg/kg) 2 h before LPS instillation (100 µg/ 100 g body weight). BALF and serum samples were collected 24 h after LPS challenge. Mean ± SE, n = 4; *significantly different from control; $significantly different from LPS.

Correlations

Determinants of CC16 variations in serum were examined by stepwise regression analysis, testing as dependent variable serum CC16 and as independent variables, the concentrations of CC16, albumin, or total protein in BALF. In both the dose- effect and the kinetics experiments, the concentration of CC16 was independent of the CC16 concentration in BALF but correlated very significantly with the concentration of albumin (partial r² = 0.54 and 0.26, p = 0.0001 and 0.0022, respectively) and even better of total protein in BALF (Figure 6).

View larger version (17K): [in this window] [in a new window]   Figure 6.   Correlations between the concentrations of CC16 in serum and of total protein in BALF from control and LPS-treated rats. Data are from the dose-effect experiment (a) or from the kinetic experiment (b).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we investigated the effects of acute lung inflammation induced by the intratracheal instillation of LPS on the CC16 expression. The results showed that LPS treatment results in a decrease of CC16 concentrations in BALF and lung homogenate as well as of the CC16 mRNA levels in lung tissue. At the highest dose, LPS also reduced the number of CC16-positive cells in the bronchiolar epithelium. These changes were associated with an increased leakage of CC16 in serum.

The decrease of CC16 concentrations in BALF and lung homogenates could be due to a decreased production/secretion of the protein or a reduction of number of Clara cells. The kinetic study in rats challenged with 100 µg LPS showed that the concentrations of CC16 in lung homogenates and BALF as well as the CC16 mRNA levels in lung are already markedly diminished at 3 or 6 h postinstillation, whereas the LDH release in BALF is increased only after 24 h postinstillation. These effects were found in the absence of changes in the number of Clara cells, which indicates that the reduction of CC16 expression and secretion is not the consequence of cytotoxic effects causing a decrease of Clara cell population. They probably reflect a downregulation of CC16 production by mediators released in the airways during inflammation. At the highest dose of LPS (200 µg), however, the decrease of CC16 mRNA and protein levels in BALF was associated with a reduction of the number of Clara cells and thus is a consequence of damage to Clara cells secondary to pulmonary inflammation, as suggested by the morphologic study of Ooi and colleagues (4). The intravascular leakage of CC16 across the disrupted bronchoalveolar blood barrier probably also further contributes to the reduction of CC16 concentration in BALF.

To our knowledge, this is the first study correlating in acute lung inflammation the changes of CC16 mRNA levels in the lung with those in the concentrations of the protein in BALF and lung homogenate. In 1992, Hackett and colleagues (18) reported that intraperitoneal injection of LPS did not affect the size or abundance of CC16 mRNA in the adult rat lung, a negative finding that was probably due to the use of a different route of LPS administration. In another study, Stripp and colleagues (19) showed that acute naphthalene injury induces a dose- and time-dependent decrease of the CC16-mRNA levels in the lung, but the investigators did not check whether this effect was associated with a reduction of the protein levels in the respiratory tract.

Intratracheal instillation of LPS in rats induces a transient alveolitis characterized by neutrophil and macrophage influx and increased airway permeability to albumin as reported earlier (20, 21). However, our study has shown that this early inflammatory response to LPS is associated with an increased leakage of CC16 from lung into blood that correlates with the passage of albumin and other plasma proteins in the opposite direction. In contrast to the influx of inflammatory cells, the transepithelial leakage of plasma proteins and CC16 observed 24 h after the challenge is not dose-dependent over the three doses of LPS, reaching a maximum at the dose of 100 µg. This phenomenon could be related to the kinetics of changes that could vary according to the LPS dose or to the intervention, at the highest dose, of a mechanism limiting the increase of epithelial permeability to proteins.

An enhanced bidirectional leakage of proteins originating from the plasma or the lung (CC16 and surfactant-associated proteins) across the alveolar blood barrier has been reported in various inflammatory lung disorders characterized by an increased lung epithelium barrier permeability such as sarcoidosis, adult respiratory distress syndrome, and idiopathic pulmonary fibrosis (21). The mechanism by which inflammatory response initiates and sustains epithelial permeability in these diseases are not fully elucidated. Several studies point to inflammatory cells as important mediators of this increased airway permeability (20, 26). Pretreatment of rats by an intravenous injection of neutrophil antibody (20) or by cyclophosphamide (26), which inhibit the leukocyte influx into the airways, prevents the enhanced epithelial permeability induced by endotoxin instillation or ozone exposure. In recent studies, Bernareggi and colleagues (27) and Li and coworkers (28) showed that LPS administration induced an increase of airway epithelial permeability that was significantly diminished by treatment of rats with the nitric oxide synthase inhibitors L-NMMA and dexamethasone, suggesting a participation of nitric oxide in increased epithelial permeability caused by LPS.

The involvement of inflammatory cells in the increased airway permeability is supported by our observations showing that dexamethasone pretreatment attenuates both the leukocyte infiltration and the albumin leakage into airways. The reason why dexamethasone does not completely prevent the increased airway permeability to albumin and has no detectable effect on CC16 leakage is unclear. The explanation for these different responses of CC16 in serum and albumin in BALF most probably lies in the greater sensitivity of serum CC16 to epithelial barrier loss of integrity as compared with albumin in BALF. The leakage of CC16 across the epithelial barrier is indeed greatly facilitated by the huge transepithelial concentration gradient compared with that of albumin (5,000 versus 10). Because of its smaller size CC16 is probably also more sensitive than albumin to a slight enlargement of the transepithelial pathways that would persist after dexamethasone pretreatment. These observations are of interest since they suggest that the assay of serum CC16 allows investigators to detect subtle defects in the epithelial barrier permeability that may be present despite resolved or limited inflammation and are not accurately reflected by the levels of albumin in BALF.

The consequence of CC16 decrease in the lung is difficult to assess since the exact biologic function(s) of CC16 have not been fully established. In vitro studies suggest that CC16 can exert an anti-inflammatory action through its ability to inhibit phospholipase A2 (5, 6), the PDGF-induced chemotaxis of lung fibroblasts (6), and the interferon- (IFN-) production and biologic activity (7). Furthermore, in a recent study, Johnston and colleagues (29) reported an altered pulmonary response to hyperoxia in Clara cell protein-deficient mice, leading to increased mortality, early onset of lung edema, and induction of proinflammatory cytokine mRNAs. In view of this growing evidence of a protective role of CC16 in the lung, one might logically postulate that an early decrease of its production, as observed in the LPS model, may contribute to the development of inflammation and lung injury.

In conclusion, our data show that in the LPS model of lung inflammation, the expression of CC16 is downregulated at both the transcript and the protein level, suggesting an implication of this lung secretory protein in the development of inflammatory processes. This study also further validates serum CC16 as peripheral marker of the permeability of the alveolar-capillary barrier.