Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The respiratory system is continuously exposed to the external environment via the air inhaled, which contains numerous potentially harmful agents. Lipopolysaccharide (LPS), a major proinflammatory glycolipid component of the gram-negative bacterial cell wall, is one of the agents ubiquitously present as contaminant on airborne particles, including air pollution (1), organic dusts (2, 3), and cigarette smoke (4). In ambient breathing air, LPS is measured at very low concentrations (± 0.4 ng/m³) (5). Under these conditions, the lung has efficient defense mechanisms against airborne LPS, including the mucociliary system, antimicrobial compounds in the epithelial lining fluid, and alveolar macrophages. Experimental inhalation of a high LPS dose, however, provokes the innate immune system in healthy human subjects, resulting in an acute inflammatory response, which manifests itself at both the pulmonary and the systemic level. In addition, this response is accompanied by clinical symptoms, including fever and airflow decline (6, 7). Extensive studies investigating acute inflammation using laboratory animals have demonstrated that LPS activates alveolar macrophages via LPS-binding protein (LBP)/CD14/Toll-like receptor (TLR)-4-dependent pathway to produce specific cytokines, resulting in a rapid but transient neutrophil infiltration into the lung (interstitium, alveoli, and airway) (8).

In contrast to short-term LPS exposure, chronic exposure to significant levels of LPS is reported to be associated with the development and/or progression of many types of lung diseases, including asthma, chronic bronchitis, and progressive irreversible airflow obstruction, all characterized by chronic inflammatory processes in the lung. Michel and colleagues reported that the concentration of LPS in the domestic setting is associated with the clinical severity of asthma (9). Moreover, individuals with asthma (10) and those with chronic bronchitis (11) develop airflow obstruction at lower concentrations of inhaled LPS compared with healthy subjects, and thus respond more sensitively to LPS. Chronic occupational exposure to LPS contained in organic dusts, such as grain dust and swine dust, is known to result in airflow obstruction among agricultural workers. In addition, the concentration of inhaled LPS (not dust) in the workplace bioaerosol is strongly and consistently related to respiratory symptoms and airflow obstruction (12, 13). Furthermore, follow-up studies have shown that long-term average exposure to LPS-laden dusts resulted in a profound pulmonary inflammatory response related to longitudinal decline in lung function (14), and thus irreversible and progressive lung pathology. Recently, bioactive LPS is reported to be present at high levels in cigarette smoke (4), which suggests that LPS is an important pathogenic substance in cigarette smoke contributing to the pulmonary diseases that develop in susceptible cigarette smokers. Taken together, these studies indicate that LPS may have a larger role in the pathogenesis of chronic lung diseases than previously realized. However, the extent to which inflammatory processes contribute to lung pathology observed in the lungs of these patients is still poorly understood.

In the present study, pathologic effects of long-term LPS exposure to the lung were investigated in detail. To this end, we developed a murine model in which mice were exposed to multiple intratracheal instillations of Escherichia coli LPS. We show that repeated intratracheal LPS instillation in mice results in persistent chronic pulmonary inflammation with altered cytokine expression, accompanied by airway and alveolar alterations. Interestingly, the observed inflammatory and pathologic changes mimic changes observed in human subjects with chronic pulmonary inflammatory disorders, especially chronic obstructive pulmonary disease (COPD), suggesting that this murine model is applicable to dissect the role of inflammation in the pathogenesis of these disease conditions.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Antibodies

Monoclonal antibody (mAb) Gr-1 (rat anti-mouse Ly-G6) was obtained from PharMingen (San Diego, CA). mAb Moma-2 (rat anti-mouse macrophage/monocyte) was a kind gift from Prof. G. Kraal (Free University, Amsterdam). mAb 6D5 (rat anti-mouse CD19) was obtained from Serotec (Oxford, UK). Hybridoma cell lines GK1.5 (rat anti-mouse CD4), 53-6.72 (rat anti-mouse CD8), and TIB120 (rat anti-murine major histocompatibility complex [MHC] class II) were obtained from the American Type Culture Collection (Rockville, MD). mAb -SMA-FITC (fluorescein isothiocyanate-conjugated mouse anti-human smooth muscle actin) was obtained from Sigma Chemical Co. (St. Louis, MO). mAb MIB-1 (mouse anti-human peptide Ki-67) was obtained from Immunotech (Marseille, France). Control mAb GL113 (rat IgG) was a gift from Dr. Savelkoul (Erasmus University, Rotterdam, The Netherlands).

Animals

Male Swiss mice ~ 12 wk old were obtained from Charles River Breeding Laboratories (Heidelberg, Germany). Animals were housed individually in standard laboratory cages and allowed food and water ad libitum throughout the experiments. The studies were performed under a protocol approved by the Institutional Animal Care Committee of the Maastricht University.

Experimental Protocol

Mice were repeatedly challenged with LPS (E. coli, serotype O55:B5: Sigma Chemical Co.) twice a week for a period of 12 wk by intratracheal instillation in an attempt to induce a chronic pulmonary inflammation. The dose of LPS used was 5 µg/instillation/mouse, which corresponds to the LPS dose delivered to the human lung by smoking of ~ 25 cigarettes (4). No signs of an overall toxic effect at the dose of LPS employed were observed in the trachea, airways, and lungs. Sham mice were instilled intratracheally with LPS-free sterile 0.9% NaCl, whereas control mice received no treatment. Intratracheal instillation was performed by a nonsurgical technique as described previously (8). In brief, mice were anesthetized by intraperitoneal injection of xylazin/ketamin. A volume of 50 µl was instilled intratracheally via a canule, followed by 0.1 ml of air. After intratracheal treatment, the mice were kept in an upright position for 10 min to allow sufficient spreading of the fluid throughout the lungs. Thirty mice were divided into four experimental groups. LPS+1 group (n = 10): intratracheal exposure to 5 µg LPS; mice were killed 1 wk after the final intratracheal instillation (1-wk recovery period) to allow acute inflammatory changes that occur after each LPS exposure to subside. LPS+8 group (n = 5): intratracheal exposure to 5 µg LPS; mice were killed 8 wk after the final intratracheal instillation (8-wk recovery period) to study the ongoing character of the induced pathogenesis. Saline group (n = 10): intratracheal instillation of 50 µl sterile saline; mice were killed 1 wk after the final intratracheal instillation. Control group (n = 5): age-matched control mice.

Histologic Analysis

After thoracotomy, the left lung was prepared for light microscopy. Lung tissue was inflated with 10% phosphate-buffered formalin (pH = 7.4) at a pressure of 20 cm H₂O through the trachea for 15 min and subsequently fixed in formalin for 24 h. After paraffin embedding, 4-µm sections were cut and stained with hematoxylin/eosin (H&E) to evaluate general morphology; or periodic acid-Schiff (PAS)/Alcian blue to determine the presence of mucin glycoconjugates.

Immunohistochemistry

Inflammatory cells were characterized by immunohistochemistry on frozen sections. Lung tissue of the right lung was infused with 20% sucrose/50% TissueTek OCT compound (Sakura Finetek Europe BV; Zoeterwoude, The Netherlands) and slowly frozen in TissueTek OCT on dry ice. Consecutive sections of 5-7 µm thickness were cut, fixed in acetone, and stored at 20°C until use. Sections were rehydrated in tris-buffered saline, pH 7.8, and endogenous peroxidase was quenched with 0.3% H₂O₂. To prevent nonspecific binding of antibodies, sections were blocked with 5% normal goat serum. Sections were stained with mAbs for detection of neutrophils (Gr-1), macrophages (Moma-2), CD4⁺ T-lymphocytes (GK1.5), CD8⁺ T-lymphocytes (53-6.72), B-lymphocytes (6D5), and cells expressing MHC class II antigen (TIB120). After washing, peroxidase-conjugated goat anti-rat IgG Ab (Jackson, West Grove, PA) preincubated with 5% normal mouse serum was applied as the secondary detection Ab. Enzymatic reactivity was visualized with 3-amino-9-ethylcarbazole (AEC). Sections were counterstained with hematoxylin and mounted. No significant staining was detected in slides incubated with control mAb (GL113) instead of the primary detecting mAb, indicating the absence of significant background staining.

Immunohistochemistry for -smooth muscle actin (-SMA) was performed as described previously (15). In brief, 4-µm paraffin sections were deparaffinized and endogenous peroxidase was quenched with 0.3% H₂O₂ in 100% methanol. -SMA was detected using FITC-labeled -SMA mAb followed by peroxidase-labeled anti-FITC pAb (Boehringer Mannheim BV, Almore, The Netherlands), and developed with the chromogen 3,3'-diaminobenzidine. Sections were counterstained with hematoxylin and mounted.

Proliferating cells were detected by immunohistochemistry on 4-µm paraffin sections using SuperFrost Plus slides (Menzel-Gläser, Braunschweig, Germany). After deparaffinization, endogenous peroxidase was quenched with 0.3% H₂O₂ in 100% methanol. Antigen was unmasked by heating the sections for 15 min in 10 mM citrate buffer pH 6.0 using a microwave, followed by 60 min cooling in the same buffer. Proliferating cells were detected by incubating the sections with mAb Ki-67 (MIB-1) for 30 min at 37°C. M.O.M. immunodetection kit (Vector Laboratories, Burlingame, CA) was used according to the manufacturer's instructions. The M.O.M. kit significantly reduced background staining caused by the inability of the anti-mouse secondary antibody to distinguish between the mouse primary antibody and endogenous mouse immunoglobulins in the tissue. The peroxidase-labeled secondary antibody was visualized with AEC. Sections were counterstained with hematoxylin and mounted.

Morphologic and Morphometric Analyses

Lymphocytic aggregates (cutoff value > 50 lymphocytes/aggregate) were counted on H&E-stained lung sections in the hilar region (magnification ×50) by two independent observers unaware of the experimental conditions. Results are expressed as the number of aggregates/section (corrected for surface area).

Intra-alveolar macrophages recognized by general morphology on H&E-stained paraffin sections were enumerated for each section on ten randomly chosen fields (divided into 10 × 10 squares), which represents a total area of 2.5 mm². This area was sufficient to obtain a mean value per subject that remained rather constant after further increase of the number of fields examined. Intra-alveolar macrophages were counted at a magnification of ×200 by two independent observers unaware of the experimental conditions. Results are expressed as the number of macrophages/mm².

Lymphocyte subsets recognized by specific immunohistochemical staining on frozen sections were scored on a 4-point scale by one independent observer unaware of the experimental conditions (0 = 0-2 cells; 1 = 3-6 cells; 3 = 7-10 cells; 4 = >10 cells per field). Positive cells were counted in four representative fields by high-power field magnification (×200).

Airway wall thickening was determined using standard morphometric technique on -SMA-stained paraffin section cut from the upper part of the left lung. Conducting airways (width > 190 µm) were captured at ×20 with a digital camera and the smooth muscle cell area surrounding the airways was quantified by computerized morphometry using the Quantimet 570 C imaging analysis system (Leica Microsystems, Cambridge, UK). Increased width of the smooth muscle layer was taken as evidence of airway wall thickening.

Standard morphometric technique was used to determine the presence of emphysematous changes in the lungs. In brief, H&E-stained paraffin sections cut from the upper part of the left lung were used, and 10 randomly selected fields were sampled by projecting a microscopic image of the lung section on a screen with a square reference lattice containing one horizontally and one vertically placed test line. The number of intersections of alveolar walls on the test lines were quantified by computerized morphometry using the Quantimet 570 C imaging analysis system (Leica Microsystems) and used to quantify alveolar mean linear intercept (L_M, the average distance between alveolar walls). Increased L_M was taken as evidence of alveolar enlargement.

Reverse Transcription-Polymerase Chain Reaction

Total RNA was isolated from snap frozen lung tissue of the right lung using a commercially available kit (SV Total RNA Isolation System; Promega, Madison, WI). Total RNA concentration and purity was ascertained by electrophoresis on an ethidium bromide-stained 0.8% agarose gel followed by UV illumination and UV spectrophotometric analysis at wavelengths of 260 and 280 nm. Five micrograms of total RNA was reverse transcribed in a 20-µl volume using oligo (dT) primers and Moloney murine leukemia virus reverse transcriptase (M-MLVRT; Life Technologies, Paisley, UK) according to the supplier's recommendations. PCR was performed in a 25-µl reaction volume containing 100 µM of each dNTP, 200 nM sequence-specific primers, and 0.5 U Taq DNA polymerase (Perkin Elmer/Cetus) during 35-40 cycles under the following conditions: 95°C 30 min, annealing temperature (AT) 45 min, 72°C 30 min. PCR primers used in RT-PCR for interleukin (IL)-6 (318 bp, AT 56°C) were sense primer (5'-TGG GAAATCGTGGAAATGAGA-3') and antisense primer (5'-GAGAGCATTGGAAATTGGGGT-3'), whereas PCR primers for tumor necrosis factor (TNF)- (307 bp, AT 63°C), interferon (IFN)- (243 bp, AT 63°C), IL-18 (582 bp, AT 63°C), and -actin (348 bp, AT 60°C) were designed as previously described (16, 17). A mock PCR (without cDNA) was included to exclude contamination. cDNA samples were standardized based on the content of -actin cDNA as housekeeping gene. -Actin cDNA was evaluated by performance of a -actin PCR on multiple dilutions of each cDNA sample. The amount of amplified product was estimated by densitometry of gelstar (FMC Bioproducts, Rockland, ME) stained 1.2% agarose gels using a CCD camera and Imagemaster VDS software (Pharmacia). Relative mRNA levels for TNF-, IL-6, IFN-, and IL-18 were calculated by comparison of band intensities of the RT-PCR products to standard curves prepared by PCR amplifications on dilution series of a highly concentrated murine lung cDNA.

Statistical Analysis

Statistical analysis was performed by means of one-way ANOVA, and the Mann-Whitney U test was used as post hoc test (Statistical Package for the Social Sciences, version 7.5 for Windows; SPSS Inc., Chicago, IL). Data are expressed as mean ± standard error of the mean (SEM). A P value < 0.05 denotes the presence of a significant statistical difference.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Persistent Chronic Pulmonary Inflammation after Long-Term LPS Exposure

Histological examination of H&E-stained lung tissue sections of mice repeatedly exposed to LPS (5 µg/mouse) twice/wk for 12 wk demonstrated a striking inflammatory pattern after a 1-wk recovery period. First, lungs of LPS-challenged mice contained dense lymphocytic aggregates, which were observed ubiquitously in all parts of the lungs (Figures 1A and 1B). As illustrated in Figure 1A, aggregates appeared to be located around the airways (peribronchial and peribronchiolar) and large vessels (perivascular), usually without affecting the ciliated bronchial epithelial cells. Incidentally, lymphocytic cells infiltrated the bronchial epithelial layer, resulting in local bronchiolitis (data not shown). These lymphocytic aggregates or infiltrates were not noted in lungs of saline-treated and age-matched control animals. Second, a profound intra-alveolar accumulation of mononuclear leukocytes was clearly noticeable in parenchyma of LPS-exposed mice, which consisted predominantly of macrophages (Figures 1C and 1D). These inflammatory changes were not observed in the parenchyma of saline-treated and age-matched control mice. Interestingly, pulmonary infiltration of neutrophilsobserved to occur rapidly after each LPS exposure (data not shown)was not observed after 1-wk recovery period, indicating that acute inflammatory changes were resolved at this time point.

View larger version (67K): [in this window] [in a new window]   Figure 1.   Persistent pulmonary inflammation induced by long-term intratracheal LPS instillation. Mice were repeatedly instilled with 5 µg LPS for 12 wk, and killed 1 wk [LPS+1] or 8 wk [LPS+8] after the final intratracheal instillation. Age-matched control mice were repeatedly exposed to saline [S] for 12 wk or not treated [C]. (A) H&E staining of paraffin sections demonstrated persistent peribronchial and perivascular lymphocytic infiltrates (arrows) in LPS-challenged mice and the lack of these inflammatory patterns in control mice. Bar = 200 µm. (B) Quantification of lymphocytic aggregates, expressed as aggregates/section (mean ± SEM, n = 5 to 10 mice/ group). *P < 0.05 versus control; NS, not significant. (C) Higher-power view of H&E-stained paraffin section demonstrating accumulation of macrophages (arrows) in parenchyma of LPS-treated mice. Bar = 50 µm. (D) Quantitative distribution of intra-alveolar macrophages, expressed as macrophages/mm2 (mean ± SEM, n = 5 mice/group). *P < 0.05 versus control, #P < 0.05 versus LPS.

To study if LPS-induced pulmonary inflammation is an ongoing process, mice were killed 8 wk after cessation of LPS challenge. Histologic examination demonstrated that lymphocytic aggregates observed at 1 wk after the final LPS exposure were still present after this long recovery period, and appeared to be of similar size and density (Figure 1A-B), suggesting persistence of the inflammatory reaction. Furthermore, parenchymal tissue of LPS-exposed mice showed a significant increase of macrophages at this time point when compared with control mice (Figures 1C and 1D). However, macrophage numbers tended to be slightly but significantly reduced when compared with LPS-treated animals after 1-wk recovery period. Taken together, these data demonstrate that repeated intratracheal instillations of LPS result in a persistent chronic pulmonary inflammatory response, characterized by lymphocytic aggregates around airways and vessels and parenchymal accumulation of mononuclear leukocytes.

Characterization of Cellular Infiltrates

The specific nature of cellular infiltrates in airways and parenchyma was further characterized by immunohisto-chemistry on frozen lung sections. To this end, a panel of monoclonal antibodies against murine T lymphocytes (CD4⁺ or CD8⁺), B lymphocytes, macrophages, and neutrophils was used. As shown in Figure 2A, immunohistochemical staining identified CD4⁺ T lymphocytes as the principal component of the aggregates present in lungs of LPS-exposed mice after a short recovery period. Other mononuclear cells included CD8⁺ T lymphocytes and B lymphocytes, which were present to lesser extent at the periphery of these aggregates. Thorough examination of serial sections revealed that aggregates were not organized in T cell and B cell zones, thus lacking any resemblance with lymph nodes or bronchus-associated lymphoid tissue (BALT). In addition, neither neutrophils nor macrophages were noted in these aggregates.

View larger version (110K): [in this window] [in a new window]   Figure 2.   Characterization of cellular infiltrates in LPS-challenged mice. Frozen lung sections of LPS-challenged mice after 1-wk recovery period were evaluated by immunohistochemistry using mAbs for macrophages (M), T lymphocytes (CD4+ or CD8+), and B lymphocytes (CD19+). (A) Composition of lymphocytic aggregates (circle). Note the presence of lymphocytes (CD4+, CD8+, and CD19+), and absence of macrophages in these aggregates. (B) Cellular infiltrates in parenchyma. Note the accumulation of macrophages and CD8+ T cells in the parenchymal area. (C) Expression of MHC class II antigen. Note the low staining for MHC II in control animals [C], and the intense staining for MHC II in intra-alveolar mononuclear leukocytes as well as lymphocytic cells in the aggregates (circle) in LPS-treated animals after 1-wk recovery period [LPS+1]. Bar = 50 µm. Results are representative for three separate experiments.

As shown in Figure 2B, immunohistochemical staining confirmed the prominent accumulation of macrophages as well as the absence of infiltrated neutrophils in the parenchyma of LPS-challenged mice. In addition, positive staining for lymphocytes was observed in the alveolar area of mice repeatedly exposed to LPS; this was not clearly noticeable in H&E-stained paraffin sections. Interestingly, the number of CD8⁺ T cells, but not CD4⁺ T cells and CD19⁺ B cells, in the parenchyma was significantly increased relative to controls (Table 1). Furthermore, evaluation lungs from LPS-exposed mice killed after an 8-wk recovery period demonstrated that cessation of LPS challenge did not result in an altered composition of the lymphocytic aggregates and the inflammatory pattern in the parenchyma (Table 1).

Additionally, we analyzed MHC class II antigen expression, which is considered to be a marker for cell activation that correlates with the intensity of the immune response (18). As shown in Figure 2C, low staining for MHC II was noted in alveolar macrophages from control animals. In LPS-challenged animals, mononuclear leukocytes present in the parenchyma as well as lymphocytic cells located in the aggregates displayed positive staining for MHC II after 1-wk (Figure 2C) and 8-wk (data not shown) recovery periods. The staining intensity was strongly increased in LPS-challenged mice when compared with controls, indicating that infiltrated inflammatory cells in both airway and parenchymal regions were activated. Taken together, these data demonstrated CD4⁺ T cells to be the principal component of peribronchial and perivascular lymphocytic aggregates observed after 1-wk and 8-wk recovery periods, whereas macrophages and CD8⁺ T cells were identified to accumulate in the parenchymal area of LPS-exposed mice. Furthermore, infiltrated inflammatory cells were activated in respect of MHC II expression.

Cytokine mRNA Expression

Because the chronic character of an inflammatory response is possibly due to alteration in expression levels of Th1 cytokines (19), we investigated mRNA levels of the Th1 type cytokines TNF-, IL-6, and IFN- in lungs of LPS-exposed mice and controls using RT-PCR. Additionally, IL-18 mRNA levels were determined, which is known to be involved in IFN- expression and to be associated with chronic inflammation. After correction for -actin levels, lungs of LPS-exposed mice demonstrated enhanced mRNA expression for TNF-, IFN-, and IL-18, whereas these mRNA levels were low or negligible in saline-exposed and noninstilled lungs (Figure 3A). Remarkably, expression levels for all three cytokines further increased after 8 wk when compared with levels after a 1-wk recovery period. In contrast, lungs of saline-exposed and control mice demonstrated constitutive mRNA levels for IL-6, but these levels were not increased after long-term LPS exposure. These observations were confirmed by the calculation of the densitometric ratio for each mRNA expression (Figure 3B).

View larger version (22K): [in this window] [in a new window]   Figure 3.   Transcription of inflammatory mediators in murine lung tissue as determined by RT-PCR. Total RNA was isolated from lung tissue of age-matched control mice [C], saline-treated mice [S], and LPS-challenged mice after 1-wk [LPS+1] and 8-wk [LPS+8] recovery periods. (A) mRNA expression levels of TNF- (307 bp), IL-6 (318 bp), IFN- (243 bp), IL-18 (582 bp), and -actin (348 bp) were analyzed using RT-PCR. Equal intensity of -actin mRNA bands confirmed equal loading of cDNA in RT-PCR. One representative experiment from a series of three is shown. (B) Densitometric ratios (n = 3 mice/group).

Airway and Alveolar Alterations after Long-Term LPS Exposure

Lungs of LPS-challenged mice demonstrated a variety of abnormalities in airways and parenchyma. First, lung sections stained with PAS/Alcian blue showed a marked increase of mucus producing cells numbers after a 1-wk recovery period, compared with saline-treated controls (Figure 4A). Metaplasia of airway goblet cell was evident in the larger airways of the respiratory tract, especially at sites where lymphocytic infiltrates were underlying the bronchial epithelium. Secretory cell metaplasia was also observed 8 wk after cessation of LPS challenge, but to a lesser extent, suggesting that these metaplastic events may be reversible. Second, airway walls of LPS-exposed mice were thickened when compared with controls. Immunohistochemical staining indicated the presence of -SMA in small, elongated cells arranged circumferentially in a single layer in the airway submucosa of control mice. On the other hand, airways of LPS-challenged mice were characterized by a patchy -SMA staining, and commonly of two layers of nuclei in the airway submucosa. As shown in Figure 4B, width of the smooth muscle layer was significantly increased after both 1-wk and 8-wk recovery periods compared with controls.

View larger version (35K): [in this window] [in a new window]   Figure 4.   Airway and alveolar abnormalities after long-term intratracheal LPS instillation. (A) Mucus cell metaplasia. Representative lung histology (PAS/Alcian blue staining) of mice repeatedly exposed to saline [S] or LPS after 1-wk recovery period [LPS+1]. Note the mucus producing bronchial epithelial cells in LPS-challenged mice (arrows). Bar = 25 µm. (B) Morphometric analysis of airway wall thickening. Paraffin sections were stained for -SMA, and width of smooth muscle layer is expressed in µm. *P < 0.05 versus control; NS, not significant. (C) Alveolar enlargement. H&E staining of paraffin sections demonstrated destruction of alveolar walls in LPS-instilled mice after 1-wk recovery period [LPS+1] compared with control mice [C]. Bar = 100 µm. (D) Morphometric analysis of alveolar enlargement. Alveolar mean linear intercept (LM) values are expressed in µm (mean ± SEM, n = 5 mice/group). *P < 0.05 versus control; NS, not significant.

Alveolar alterations were examined by morphometric measurements of alveolar enlargement and evaluation of proliferation. Morphometric evaluation of alveolar dimensions revealed emphysematous changes in LPS-challenged mice at 1-wk recovery period (Figures 4C and 4D). In contrast, repeated saline exposure did not result in alveolar enlargement when compared with controls. Importantly, lungs from LPS-exposed mice sacrificed after an 8-wk recovery period still displayed emphysematous changes, indicating permanent enlargement induced by long-term LPS exposure. Furthermore, the expression of Ki-67 antigen, a marker of active cell proliferation, was evaluated to investigate presence of repair processes. Alveolar epithelial cells of LPS-exposed mice did not display Ki-67 staining, neither after 1-wk nor after 8-wk recovery periods (data not shown), indicating the absence of active repair processes. Taken together, these data show that long-term LPS exposure results in pathologic changes in airways and parenchyma, characterized by goblet cell metaplasia and airway wall thickening, and irreversible alveolar enlargement.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study investigated inflammatory and pathologic alterations after long-term LPS exposure to the lung. Mice were repeatedly exposed to LPS and killed after 1-wk or 8-wk recovery periods. We have previously demonstrated that single intratracheal LPS challenge results in rapid but transient infiltration of neutrophils from the pulmonary vascular bed into the respiratory air spaces (8) associated with acute lung injury, such as increased pulmonary vascular permeability and pulmonary edema. The present study is the first to characterize chronic inflammatory events induced by repeated intratracheal LPS instillation. We demonstrate that long-term intratracheal LPS exposure resulted in a persistent chronic inflammatory reaction, which is accompanied by alterations in airways and parenchyma.

Lungs of LPS-challenged mice contained dense peribronchial and perivascular lymphocytic aggregates, which did not appear to be organized in lymphoid follicles. In addition, aggregates were shown to consist of mainly T lymphocytes and less B lymphocytesno macrophagessupporting the observation that these lymphocytic aggregates show no similarity with normal lymph nodes or BALT present in the murine lung (22). Origin of these aggregates is uncertain, but they may originate from small T and B cell clusters that have been described in the murine lung in the vicinity of bronchioles and veins (22). Interestingly, this type of mononuclear inflammatory cell accumulation in the peribronchial area was also noted in transgenic mice overexpressing specific proinflammatory cytokines in the airway epithelial cells (20, 23, 24). Especially peribronchial infiltrates in human IL-6 transgenic mice show striking similarity to our observations, e.g., significant numbers of CD4⁺ and CD8⁺ T lymphocytes and B lymphocytes, but lacking alveolar macrophages (20).

Long-term LPS exposure also resulted in a chronic inflammatory response in the parenchymal area, which was comprised of activated macrophagesindicated by increased expression of MHC class II molecules on the cell surfaceand CD8⁺ T cells. The absence of neutrophils at 1 wk after the last LPS exposure indicates that neutrophil infiltration is rather a transient process triggered by each LPS exposure instead of being a chronic factor. Recently, Ofulue and colleagues demonstrated that long-term exposure (up to 6 mo) to cigarette smoke in rats resulted in a similar increase of intra-alveolar macrophages (25). This result is in accordance with our observations, assuming that LPS is an important contaminant of cigarette smoke (4).

An interesting result of the present study is that the characteristic inflammatory pattern in the lung induced by repeated intratracheal LPS instillation continued after cessation of the LPS challenge, which supports the important role for LPS in chronic lung inflammation. Studies using transgenic mice (19) suggest that altered expression levels of Th1-type cytokines like TNF-, IL-6, and IFN- may contribute to the ongoing inflammatory response and structural changes in airways and parenchyma. We therefore investigated mRNA levels of these specific Th1-type cytokines in lungs of LPS-exposed mice, and also mRNA levels of IL-18, which is involved in IFN- expression and has been associated with chronic inflammation. The inflammatory response in the LPS-treated animals was characterized by significant increases in mRNA levels for TNF-, IL-18, and IFN-, which further increased after cessation of LPS challenge. These data support the idea that ongoing inflammatory processes induced by long-term LPS exposure results from an altered cytokine balance, involving cytokines derived from both macrophages and lymphocytes. Interestingly, long-term LPS exposure did not influence the constitutive mRNA levels for IL-6 in saline-exposed and control lungs. This was unexpected, since the pattern of pulmonary inflammation in IL-6 transgenic mice showed striking similarities with our observations (20). However, IL-6 is documented to be a downregulator of LPS-induced TNF- and IL-1 expression in vitro and a strong inhibitor of LPS-induced acute inflammation in rats (26, 27). We speculate that the lack of IL-6 upregulation in lungs of mice repeatedly exposed to LPS may contribute to an altered cytokine balance. The exact mechanism by which long-term LPS exposure alters the cytokine balance, however, is unclear and awaits further investigation.

In addition to the persistent inflammatory events, long-term LPS exposure resulted in significant airway and alveolar abnormalities. First, bronchial epithelium of LPS-challenged mice displayed an increase in the mucus-producing goblet cells in the larger airways, especially at sites where lymphocytic infiltrates were underlying the bronchial epithelium. Because only small numbers of goblet cells are present in murine airways, this increase is likely due to cell differentiation (metaplasia) rather than cell division (hyperplasia). In line with our findings, increase of airway mucus cells was reported after single intratracheal LPS instillation in mice (28), and repeated intratracheal LPS exposure in rats (29) and hamsters (30). Whether the mucus cell metaplasia observed by histologic analysis also involves increased expression of mucin genes obviously needs further study. Second, airways of LPS-challenged mice demonstrated thickening of the smooth muscle layer. Recently, George and colleagues (31) demonstrated similar airway wall thickening with patchy actin staining after long-term exposure to corn dust extract in mice, which is known to contain high levels of LPS. In addition, increased bronchial wall thickness was also demonstrated in transgenic mice overexpressing IL-6 or IL-11 in airway epithelial cells (32). Third, long-term intratracheal LPS exposure resulted in alveolar enlargement without any signs of proliferation of alveolar epithelial cells, which was shown to be irreversible. Consistent with our data, previous reports described approximately equal extent of parenchymal destruction in the lungs of hamsters (30, 33) and mice (34) after repeated LPS challenge using different instillation protocols, and after long-term exposure to cigarette smoke (25, 35). However, lungs of transgenic mice overexpressing specific proinflammatory cytokinesinducible or notin alveolar type II cells (TNF- [19]) or airway epithelial cells (IFN- [21], IL-6, or IL-11 [32]) displayed enormous alveolar destruction (L_M >100%).

The pathologic changes we observed in both airways and parenchyma after multiple LPS challenge in mice mimic important features observed in COPD patients, such as goblet cell metaplasia, airway wall thickening, and emphysematous changes. Recent histopathologic studies have demonstrated a prominent infiltration of macrophages and CD8⁺ lymphocytes in bronchial tissue and lung parenchyma in COPD (36), both of which we also observed in the lungs of mice after long-term LPS exposure. Additionally, neutrophil infiltration is shown particularly in the central airways of COPD patients (37). The fact that we focused on peripheral airways and parenchyma of LPS-exposed mice may therefore explain the lack of neutrophils as a chronic factor in our model. Increasing evidence emerges that both macrophages and neutrophils have an important role in the proteolytic degradation of the extracellular matrix in COPD by production of matrix metalloproteinases. The role of CD8⁺ T cells and their mediators in this process is, however, still uncertain but very intriguing as they may be involved in apoptosis and destruction of alveolar-wall epithelial cells through the release of TNF-, perforins, and granzymes (37).

Pulmonary burden with various reagents may potentially result in a response by the respiratory system. Concerning LPS exposure, numerous studies demonstrated that chronic exposure to significant levels of LPS in man is associated with the development and/or progression of many types of chronic inflammatory lung diseases, including asthma, chronic bronchitis, and progressive irreversible airflow obstruction. Additionally, this study showed that long-term LPS exposure in mice induces a chronic mononuclear inflammatory response with specific pathologic changes mimicking COPD. Because the major environmental factor that predisposes patients to COPD is long-term cigarette smoking, we speculate that LPS present in cigarette smoke is a good candidate to trigger smoking-induced accumulation of macrophages, lymphocytes, and neutrophils in respiratory bronchioles, alveolar ducts, and alveoli in susceptible cigarette smokers. We consider this experimental murine model an important means for further elucidation of the role of chronic inflammation in the pathogenesis of chronic pulmonary disorders like COPD.