Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Instillation of bacterial lipopolysaccharide (LPS) into rat airways causes a large accumulation of neutrophils in a manner that depends on availability of tumor necrosis factor- (TNF-), E-selectin, intercellular adhesion molecule-1 (ICAM-1), CD11a/CD18, and CD11b/CD18 (1, 2). In lung inflammatory responses, it appears that the generation of TNF- by pulmonary macrophages induces the upregulation of lung vascular ICAM-1 and E-selectin (3), which are required for adhesive interactions between neutrophils and lung-vascular endothelial cells. In lung, there are several known sources of ICAM-1. Alveolar epithelial cells are known to constitutively express ICAM-1; stimulation of these cells results in no further increase in ICAM-1 expression (6). Alveolar macrophages express low levels of ICAM-1, which can be upregulated after the application of appropriate stimuli (7, 8). In the IgG immune-complex model of lung injury in rats, we have shown a relationship between production of TNF- and upregulation of vascular ICAM-1 (4). Accordingly, in lung there are at least two separate compartments for ICAM-1 expression. In the current study, we sought to define changes in lung ICAM-1 after airway instillation of LPS. The findings indicate that increases in whole-lung ICAM-1 message and protein require the availability of TNF- and interleukin-1 (IL-1). With the use of [¹²⁵I]anti-ICAM-1 antibody, it was apparent that both vascular and airway ICAM-1 were substantially upregulated in a TNF--dependent manner after airway instillation of LPS. The chief cellular sources of upregulated airway ICAM-1 appeared to be bronchiolar epithelial cells and adjacent smooth-muscle cells. These data indicate that after airway instillation of LPS, lung ICAM-1 is upregulated in both vascular and airway compartments. The functional significance of upregulated airway ICAM-1 remains to be determined.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Antibody to ICAM-1

Recombinant rat ICAM-1 containing the lymphocyte function-associated antigen-1 (LFA-1) and leukocyte integrin-1 (MAC-I) binding sites (8, 9) was produced through the bacterial expression vector, Pet 14b (Invitrogen, San Diego, CA), into which was ligated a complementary DNA (cDNA) for rat ICAM-1 containing 620 bp (Figure 1). This recombinant protein was used to immunize (in the presence of complete Freund's adjuvant) New Zealand white rabbits (10). The animals were boosted every fourth week until a high antibody titer was achieved. IgG was purified from whole serum by the use of protein A column chromatography (11).

View larger version (35K): [in this window] [in a new window]   Figure 1.   Ethidium bromide staining of cDNA preparation of rat ICAM-1 after gel electrophoresis.

Animal Model of Lung Injury

LPS-induced lung injury was produced as described elsewhere (11). Briefly, 275- to 300-g male Long-Evans rats (Haran Industries, Rochester, MI) were anesthetized with ketamine and lung injury was induced by intratracheal instillation via an intratracheal catheter of 150 µg LPS (Escherichia coli serotype 055:B5; Sigma Chemical Co., St. Louis, MO) in 300 µl of phosphate-buffered saline (PBS). Animals were killed at indicated time points. To prepare lung homogenates, the vascular and bronchoalveolar areas of lungs were flushed with sterile saline before removal of the lungs.

Lung Homogenates

LPS-induced lung injury was produced according to the protocol described earlier. The rats were sacrificed at 0, 2, 4, 6, and 8 h. After being flushed three times with 10 ml of sterile saline, the lungs were homogenized in a lysing buffer containing 10 mM 3-[cholamidopropyl-dimethylamino]- 1-propanesulfonate (CHAPS) (Sigma), with 20 µg/ml aprotinin (Boehringer Mannheim, Inc., Indianapolis, IN), 20 µg/ml bestatin (Sigma), 20 µg/ml leupeptin (Boehringer Mannheim), and 20 µg/ml 0.1 M phenylmethylsulfonyl fluoride (PMSF) (Sigma) (12). The cocktail of protease inhibitors for lung homogenates was used at the whole (×1) or half (×0.5) of the stated concentrations. The lung samples were pulse-homogenized (Ultra-Turrax TP18/1051; Tekmar Co., Cincinnati, OH) at maximum speed. After centrifugation at 15,000 rpm for 30 min, the supernatant fluids were collected and stored at 20°C.

Similar procedures were performed in animals pretreated with blocking antibodies to TNF- or IL-1 (using 200 µg rabbit IgG infused intravenously) 6 h prior to airway instillation of LPS. For these experiments, 200 µg of purified IgG anti-ICAM-1 antibody or IgG antibody to either TNF- or IL-1 was injected intravenously. Positive controls received an injection of 200 µg of normal rabbit IgG.

Quantitation of Vascular and Airway ICAM-1 in Lungs

Rabbit polyclonal antibody to rat ICAM-1 was labeled with ¹²⁵I by a standard technique, using chloramine T. A control rabbit IgG (normal IgG) was also labeled and used in parallel with the ¹²⁵I-labeled polyclonal antibody. As indicated earlier, male Long-Evans rats were injected with LPS intratracheally under ketamine-induced anesthesia. Negative control groups received intratracheal PBS instead of LPS. At 5.45 h after airway instillation of LPS, animals were injected intravenously with 500 µl PBS containing 3 µCi of either ¹²⁵I-labeled anti-ICAM-1 antibody or ¹²⁵I-labeled rabbit IgG. A second group was injected intratracheally with 250 µl PBS containing 3 µCi of ¹²⁵I-labeled anti-ICAM-1 antibody or ¹²⁵I-labeled IgG at 5.45 h after airway instillation of LPS. The rats were killed and the pulmonary vascular system was flushed with saline. Intact lungs of animals receiving intratracheal ¹²⁵I-labeled anti-ICAM-1 antibody or ¹²⁵I-labeled IgG were subjected to 12 bronchoalveolar lavages, using 5 ml of sterile saline for each lavage. Remaining whole-lung radioactivity was then determined with a gamma counter, and the anti-ICAM-1 antibody binding values were corrected by subtraction of the radioactivity produced by instillation of ¹²⁵I-labeled rabbit IgG.

Quantitation of ICAM-1 Expression on Pulmonary Macrophages

Pulmonary macrophages were collected through bronchoalveolar lavage (BAL). To eliminate the neutrophils in BAL fluids (BALFs), macrophages were isolated by Ficoll purification. A modified protocol was used as described previously (13, 14). An aliquot of 1 × 10⁷ cells was incubated with 0.3 µCi ¹²⁵I-labeled anti-ICAM-1 antibody or ¹²⁵I- labeled rabbit IgG in a final volume of 400 µl for 1 h at 37°C. One hundred microliters of this cell suspension was layered onto 250 µl of silicone oil in microcentrifuge tubes. The tubes were centrifuged for 2 min at 10,000 × g and quick frozen in liquid nitrogen, and the pellets were obtained by transectioning of the tips of the centrifuge tubes with a razor blade. Radioactivity present in the cell pellets was then determined.

ELISA Quantitation of ICAM-1 in Lung Homogenates and in BALFs

A sandwich enzyme-linked immunosorbent assay (ELISA) technique was developed, employing a polyclonal antibody to capture ICAM-1 and a second antibody to allow development of the reaction product (15, 16). For this technique, polyclonal antibodies were raised in rabbits as follows: for the primary (capture) antibody, purified soluble ICAM-1 was administered, whereas for the development antibody, rabbits were immunized with ICAM-1 present in sodium dodecyl sulfate (SDS)-polyacrylamide gels, from which the area containing ICAM-1 was removed and used for immunization. The assumption was that these antibody preparations would recognize different epitopes on ICAM-1. Optimal concentrations of antibodies used for capture and development were determined (17). The secondary antibody (used for development of the reaction product) was biotinylated, using a kit available for this purpose (NHS-LC Biotinylation Kit; Pierce Co., Rockford, IL) (18, 19). Ninety-six-well microtiter plates were coated with 50 µl per well of the capture anti-ICAM-1 antibody, diluted to 1 µg/ml in borate-buffered saline (pH 9.0), and incubated overnight at 4°C. Each washing step was performed with 200 µl PBS (pH 7.4) containing 0.2% Tween-20. The plate was blocked for 30 min at room temperature with 1% bovine serum albumin (BSA) and 0.05% Tween-20 containing PBS with a volume of 200 µl per well. The standards were prepared in PBS with CHAPS (50 µl per well). Fifty microliters per well of lung homogenates or 50 µl of undiluted BALFs were loaded (for each time point, the experiment was performed in triplicate with lung homogenates and BALFs from three different animals). The ELISA preparation was incubated for 1 h at 37°C. After a washing step, the secondary (biotinylated) antibody was added at a dilution of 5 µg/ml for 1 h at 37°C. After washing again, plates were incubated for 15 min with streptavidin-horseradish peroxidase (Pierce) at a concentration of 0.1 mg/ml (100 µl per well), and developed with 100 µl per well of O-phenylenediamine dihydrochloride. The reaction was stopped with the same volume of 3 M sulfuric acid, and the optical density at 490 nm was determined in an ELISA reader (Bio-Tek Instruments, Inc., Winooski, VT).

For the BAL experiments, animals underwent bronchoalveolar lavage with 10 ml PBS (without calcium or magnesium), which was instilled into the lungs, withdrawn, and reinstilled three times. Protease inhibitors in amounts described earlier (at 1× the stated concentrations) were added and the samples were centrifuged at 10,000 × g for 10 min. Supernatant fluids were frozen at 70°C for subsequent ELISA measurement of ICAM-1, as described earlier.

Northern Blot Analysis

Rats were killed at 0, 2, 4, 6, and 8 h after airway instillation of LPS. The pulmonary vascular system was flushed with PBS and the lungs snap frozen in liquid nitrogen. Using the guanidinium-isothiocyanate RNA extraction method described previously (20, 21), total RNA (10 µg) was fractionated electrophoretically on a 1% formaldehyde gel and blotted onto a nylon membrane (Cunco Laboratories, Meriden, CT). Methylene blue staining of the 18S and 28S ribosomal RNA (rRNA) bands showed equal loading. Full-length cDNA was random labeled with [³²P]dCTP. The membrane was hybridized with probes containing [³²P]dCTP at 10⁶ cpm/ml at 65°C for 15 h. The radiogram was developed on X-Omat film (Kodak, Rochester, NY), and densitometry was done with an Ambis Image Analysis System (Ambis Systems, Inc., San Diego, CA). Northern blot analysis was also performed on lung extracts from animals initially treated with anti-TNF- or anti-IL-1 antibody and then subjected to airway instillation of LPS and killed at 4 h unless otherwise indicated.

Western Blot Analysis

Characterization of several proteins was done with SDS- polyacrylamide gel (5% acrylamide) under reducing (2-mer-captoethanol) and nonreducing conditions with Laemmli's buffer. Five microliters of Laemmli's buffer were added to 20 µl of protein sample and the resulting solution loaded onto the gel. After electrophoresis, the gels were stained with Coomassie blue or the gel products were transferred to nylon membranes. The membranes were blocked with 3% BSA in Tris-buffered saline (TBS), incubated with the murine or rabbit anti-ICAM-1 antibody at a dilution of 1:500, and then incubated with a 1:3,000 dilution of goat antimurine or goat antirabbit alkaline phosphatase-conjugated antibody (Biorad, Inc., Hercules, CA), followed by treatment with color reagents A and B (Biorad).

Immunohistochemical Analysis of Lung Tissue

Frozen sections of LPS-injured lungs were placed on poly-L-lysine-coated slides, fixed in acetone, and incubated with the polyclonal anti-ICAM-1 antibody (20 µg/ml) for 60 min at room temperature. Further steps were performed (22) with the Vectastain biotin-avidin-peroxidase system (Vector Laboratories, Inc., Burlingame, CA).

In Situ Hybridization

In situ hybridization studies were done according to previously described methods (18). After lung injury was induced with LPS, the lungs of anesthetized rats were filled with 10 ml of embedding medium for frozen tissue specimens (O.C.T. Compound; Miles, Inc., Elkhard, IN), frozen in liquid nitrogen, and stored at 80°C. Frozen sections (8 to 10 mm thick) were melted on poly-L-lysine-treated slides and stored at 80°C, using techniques that have previously been described (23, 24). Prior to use, frozen sections were thawed briefly, fixed in 4% paraformaldehyde, rinsed in PBS, and acetylated in PBS containing 0.1 M triethanolamine and 0.25% acetic anhydride. The slides were then rinsed again, dehydrated in graded ethanol, and air-dried. [³²P]UTP-labeled sense and antisense riboprobes were generated from the previously described rat ICAM-1 cDNA by in vitro transcription of linearized pCR^TM11 plasmid. Unincorporated nucleotides were removed with ProbeQuant^TM G-50 microcolumns (Pharmacia Biotech, Inc., Piscataway, NJ). One million counts of sense and antisense probe in hybridization buffer were added to each section, the buffer containing 50% deionized formamide, 5× standard saline citrate (SSC), 50% dextran sulfate, 5× Denhardt's solution (Sigma), 10 mg/ml salmon sperm DNA (Sigma), 100 mg/ml transfer RNA (tRNA) (Boehringer Mannheim), and 100 mM dithiothreitol (DTT). The slides were covered with parafilm and incubated overnight at 50°C in a slide box humidified with 50% formamide. After hybridization, the slides were washed as follows: 2 times 10 min in 4× SSC, 30 min at 37°C in 2× SSC containing 25 µg/ml ribonuclease A (RNase A) (Boehringer Mannheim), 2 times 10 min each in 2×, 1×, and 0.5× SSC, and finally 1 h in 0.1× SSC at 50°C. The slides were then dehydrated in ethanols and air-dried. The dried slides were dipped in Kodak nitroblue tetrazolium-2 (NBT-2) emulsion diluted 1:1 with deionized water, air-dried, and stored at 4°C in foil-wrapped slide boxes containing desiccant. After 2 wk of exposure, the slides were developed for 2.5 min in Kodak D19 developer, rinsed in distilled water for 30 s, fixed for 3 min, rinsed in distilled water for 15 min, and air-dried. The sections were stained with hematoxylin for 30 s, rinsed in tap water, immersed for 30 s in Eosin Y, and then immersed in 95% and 100% ethanol. The sections were finally mounted in Permount (Fisher Scientific, FairLawn, NY).

Statistical Analysis

Data sets were examined with one- and two-way analysis of variance (ANOVA), and individual group means were compared through the use of Student's t test. Multigroup comparisons were made through the use of Schaffer's t test as well as with Fisher's protected least-significant-difference test. To calculate percent protection, all positive values were adjusted by subtraction of negative control values. Untreated positive controls were then compared with treated positive controls.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Time Course of Neutrophil Accumulation

These studies were performed in order to provide a basis of reference for studies described subsequently in which lung ICAM-1 upregulation was quantitated. After airway instillation of 150 µg LPS into rats, neutrophil accumulation in BALFs was evaluated as a function of time after instillation of LPS. As shown in Figure 2, neutrophil content rose from 0.2 × 10⁶ cells/ml in normal lung (time 0) to a peak of 3 × 10⁶ cells/ml at 6 h, followed by a gradual decline thereafter.

View larger version (27K): [in this window] [in a new window]   Figure 2.   Analysis of neutrophil content in BALFs obtained at various times after airway instillation of 150 µg LPS into rat airways. For each bar, n = 4.

Western Blot Analysis of Anti-ICAM-1 Preparations

Both the murine monoclonal antirat ICAM-1 antibody (1A29) and the newly developed rabbit polyclonal antirat ICAM-1 antibody were evaluated for reactivity with homogenates of LPS-injured lung (obtained at 6 h). Homogenates were in a volume of 10 ml (containing protease inhibitors at concentrations of 1×, described earlier), from which 20-µl samples were added to slots in gels in the absence (nonreducing conditions) or presence (reducing conditions) of 2-mercaptoethanol. The results are shown in Figure 3. The positions of the reference protein standards are shown (as kD values) in Figures 3A, 3B, and 3C, as are positions of the ICAM-1 bands. Gels in Figures 3A and 3B were run in the same experiment, whereas the gel shown in Figure 3C was run at a different time. When polyclonal anti-ICAM-1 antibody was used, bands developing in homogenate samples were found in a position consistent with a protein of approximately 90 kD (Figure 3A). This band was unaffected by reducing conditions. When the monoclonal anti-ICAM-1 antibody (1A29) was used, distinct bands in the 90-kD position were also found under reducing and nonreducing conditions (Figure 3B). In Figure 3C, homogenates of LPS-treated lungs (obtained at 6 h) were compared under reducing conditions for the effects of the antiprotease cocktail added to the lungs at the time of homogenization, either at 0.5× or 1× concentrations. The findings in Figure 3C indicate that reducing the inhibitor concentrations of antiproteases to half the original levels resulted in a degraded ICAM-1 product that migrated in a position of approximately 55 kD, whereas at full inhibitor concentrations the ICAM-1 band developed in a position indicative of an intact ICAM-1 (approximately 90 kD). At lower concentrations of protease inhibitors, we consistently failed to detect a full-length ICAM-1 product in lung homogenates.

View larger version (60K): [in this window] [in a new window]   Figure 3.   Lung homogenates from LPS-injured lungs (at 6 h), evaluated for reactivity with polyclonal anti-ICAM-1 antibody (A) or monoclonal antirat ICAM-1 (1A29) antibody (B). Homogenate samples (20 µl from a 10-ml volume) were run under reducing (R) or nonreducing (N) conditions. Homogenates were also made up in a cocktail of protease inhibitors, and the gel run under reducing conditions; protease inhibitors were used at two concentrations (0.5× and 1.0×) (C). Arrows indicate positions of standard markers and also indicate the position of ICAM-1. Gels A and B were done together. Gel C was run separately.

Time Course for Upregulation of ICAM-1 mRNA and Protein in Rat Lung

mRNA was obtained from lungs at 0, 2, 4, 6, and 8 h after airway instillation of LPS (150 µg). mRNA content for ICAM-1 was evaluated with Northern blot analysis (inset, Figure 4) and quantitated through densitometry (Figure 4A). By 2 h, there was a clearly defined increase in ICAM-1, peaking at 4 h and declining gradually thereafter. Unless otherwise indicated, mRNA for ICAM-1 was measured in all subsequent experiments at 4 h after LPS instillation, whereas ICAM-1 protein was measured at 6 h.

View larger version (22K): [in this window] [in a new window]   Figure 4.   Analysis of lung mRNA (A) and ICAM-1 protein (B) as a function of time after airway instillation of LPS. mRNA was evaluated through Northern blot analysis 4 h after LPS instillation (A). Inset shows Northern blots for ICAM-1 mRNA. Lung ICAM-1 protein was measured in whole-lung homogenates 6 h after LPS instillation (B). sICAM-1 protein in BALF was also measured at 6 h (C).

Lungs were homogenized at time points from 0 to 8 h after intratracheal instillation of LPS. The homogenate was evaluated with ELISA for whole-lung ICAM-1 protein. As shown in Figure 4B, low levels of constitutive ICAM-1 were found in lung extracts. After airway instillation of LPS, there was more than a 5-fold increase in ICAM-1 protein 2 h later, with a steady increase (nearly 7-fold) during the first 6 h followed by a decline at 8 h.

BALFs were evaluated for the presence of soluble ICAM-1 (sICAM-1) at the same time points. Very little sICAM-1 was detected in BALFs during the first 2 h (Figure 4C). However, at 4 h there was a large increase (7-fold) (P < 0.01) in sICAM-1 in BALFs, followed by a decline at 6 h and a return to basal levels (time 0) by 8 h. Thus, it would appear that after airway instillation of LPS, sICAM-1 is released into the airways.

Cytokine Requirements for Upregulation of Whole-lung ICAM-1 mRNA and Protein

In rats treated systemically with blocking antibodies to TNF- or IL-1, expression of whole-lung ICAM-1 mRNA at 4 h was decreased by 81% and 37%, respectively (Figure 5A). Six hours after airway instillation of LPS, rat-lung homogenates were assessed for ICAM-1 protein with ELISA. In these experiments, as shown in Figure 5B, there was a large increase in total lung ICAM-1 (rising from 10 ng/ml to 2,000 ng/ml) in the supernatant fluids from lung homogenates. In animals pretreated with either anti-TNF- or anti-IL-1 antibody, upregulation of lung ICAM-1 was decreased by 85% (P < 0.005) and 25% (P < 0.02), respectively. Therefore, under the conditions employed, upregulation in lung of both mRNA and ICAM-1 protein requires availability of TNF- and, to a lesser extent, of IL-1.

View larger version (36K): [in this window] [in a new window]   Figure 5.   Northern blot analysis of ICAM-1 mRNA in lung extracts obtained 4 h after airway instillation of 150 µg LPS or sterile saline (A). Whole-lung ICAM-1 protein (as determined with ELISA) in lung homogenates at 6 h after airway instillation of 150 µg LPS was also determined (B). Some animals were pretreated with anti-TNF- or anti-IL-1 antibody. For each vertical bar, n = 5.

LPS-induced Upregulation of Lung-vascular and Airway ICAM-1

It has previously been shown in the IgG immune-complex model of lung injury that lung-vascular ICAM-1 upregulation depends on the availability of TNF- (4). In the present study, LPS was instilled intratracheally into the airways, and the binding of ¹²⁵I-labeled anti-ICAM-1 antibody (given intravenously) to the lung vasculature was measured at 6 h. As shown in Figure 6A, there was approximately a 4-fold increase in binding of anti-ICAM-1 antibody to the lung vasculature. In animals treated with anti-TNF-, this binding was reduced by 84% (P < 0.003).

View larger version (39K): [in this window] [in a new window]   Figure 6.   Lung vascular ICAM-1 (A) and airway ICAM-1 (B) 6 h after airway instillation of LPS. The vertical axes represent binding of [125I]anti-ICAM-1 antibody to the lung vasculature or airway. For each bar, n = 5. Effects of an anti-TNF antibody (coinstilled intratracheally with the LPS or administered intravenously at time 0) are shown. For each vertical bar, n = 5.

Because Type I alveolar epithelial cells also exhibit constitutive ICAM-1, which is not subject to further upregulation (6, 25, 26), airway binding of [¹²⁵I]anti-ICAM-1 antibody was evaluated following airway instillation of LPS. [¹²⁵I]anti-ICAM-1 antibody was instilled intratracheally 6 h after exposure to LPS, and binding of this antibody to the airways was evaluated after 12 successive lung lavages to remove unbound anti-ICAM-1 antibody. The results are shown in Figure 6B. Following airway instillation of LPS, there was an 11-fold increase in airway binding of [¹²⁵I]anti-ICAM-1 antibody as compared with binding of [¹²⁵I]anti-ICAM-1 antibody to airways of normal lungs. This increase in binding of anti-ICAM-1 antibody to the airways was reduced by 47% (P < 0.05) in animals pretreated intravenously with anti-TNF- antibody and by 78% (P < 0.02) in animals pretreated intratracheally with anti-TNF- antibody. These data indicate that airway exposure to LPS results in greatly increased binding of [¹²⁵I]anti-ICAM-1 antibody to airway structures, in a manner that is TNF--dependent.

ICAM-1 Content of Alveolar Macrophages

Alveolar macrophages were evaluated for fixation of [¹²⁵I]anti-ICAM-1 antibody at 0, 2, 4, 6, and 8 h after airway instillation of LPS. The results are shown in Figure 7. Between 0 and 2 h after airway instillation of LPS, alveolar macrophages showed a 5-fold increase in binding of [¹²⁵I]anti-ICAM-1 antibody, after which all evidence of increased binding of anti-ICAM-1 antibody was lost. In animals pretreated intravenously with anti-TNF- antibody, no significant reduction in binding of [¹²⁵I]anti-ICAM-1 antibody was found at 2 h, but intratracheal instillation of anti-TNF- antibody (200 µg IgG anti-TNF- antibody) reduced the upregulation of ICAM-1 on alveolar macrophages by approximately 80% (P < 0.02). These data suggest that after airway instillation of LPS, upregulation of ICAM-1 occurs on both macrophages and in airway tissues, and that this upregulation depends on availability of TNF- in the airway. The failure to detect the increase in alveolar macrophage ICAM-1 at 4 and 6 h, whereas airway binding of [¹²⁵I]anti-ICAM-1 antibody was significantly increased at 4 and 6 h (Figure 4C), suggests that alveolar macrophages are not the source of the increased ICAM-1 expression found at 6 h after airway instillation of LPS.

View larger version (21K): [in this window] [in a new window]   Figure 7.   In vitro binding of [125I]anti-ICAM-1 antibody to rat alveolar macrophages obtained by BAL at various times after airway instillation of LPS. In one set of animals, anti-TNF antibody (200 µg) was administered at time 0 either intratracheally or intravenously, and the cells were obtained at various time points and evaluated for binding of [125I]anti-ICAM-1 antibody. For each vertical bar, n = 5.

In Situ Hybridization Studies of Lung mRNA for ICAM-1

In situ hybridization studies of sections from normal rat lungs showed little evidence of ICAM-1 mRNA in cells of the airways (Figure 8A), whereas at 4 h after LPS instillation, ICAM-1 mRNA expression increased dramatically in bronchial epithelial cells and in adjacent cells (Figure 8B). Thus, both ICAM-1 message and ICAM-1 protein are upregulated in airway cells of animals exposed to LPS.

View larger version (81K): [in this window] [in a new window]   Figure 8.   Sections of normal rat lung displaying bronchiolar structures (A and C) and rat lung after LPS instillation (B and D). In normal rat lung, no evidence of mRNA for rat ICAM-1, as detected by in situ hybridization, was found (A), whereas in lungs at 4 h after airway instillation of LPS, numerous particles were found in bronchiolar epithelial cells (arrows, C). Normal rat lung immunostained for rat ICAM-1 revealed occasional staining of bronchiolar epithelial cells and underlying smooth muscle (C), whereas lungs at 6 h after airway instillation of LPS contained bronchiolar epithelial cells and underlying smooth muscle with heavy staining for ICAM-1 (D). (A and B, magnification: ×400; C and D, magnification: ×200).

Localization of Rat ICAM-1 in Lung by Immunostaining

Immunostaining was performed on normal lungs and on lungs at 6 h after airway instillation of LPS. Staining conditions were set so as to minimize background staining in normal lungs. As shown in Figure 8C, faint epithelial staining and occasional staining of bronchiolar smooth-muscle cells was found in sections of normal rat lungs. In airways of LPS-treated animals, there were dramatic increases in ICAM-1 expression in both bronchiolar epithelial cells and adjacent smooth-muscle cells (Figure 8D).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

ICAM-1 expression was increased in both the vascular and airway compartments after intratracheal LPS instillation into rats, with the airway compartment seeming to show a greater net increase than the vascular compartment in ICAM-1 expression. It is possible that this at least partly reflects differences in surface areas between the two compartments. Immunostaining data suggest that most of the ICAM-1 expression in the airway compartment involves bronchiolar epithelial cells and adjacent smooth-muscle cells, which normally show little staining for ICAM-1. Alveolar macrophages also express ICAM-1 (7). Previous studies have described alveolar type II epithelial-cell expression of ICAM-1; these have involved the use of primary cultures of alveolar epithelial cells, which in vitro express high constitutive levels of ICAM-1 and seem incapable of further ICAM-1 expression after stimulation (6, 25). However, phenotypic transition of type II cells to type I cells occurs in vitro, raising the question of which cell type may be more expressive of ICAM-1 (6, 26). Hyperoxia seems to be able to induce expression of ICAM-1 on type II cells of mice (27). Upregulation of ICAM-1 in bronchial epithelial cells, as demonstrated in the present study, might result in enhanced adhesion of inflammatory cells, perhaps putting the bronchial epithelial cells at risk of damage by adherent leukocytes. Collectively, the data provided by the study, together with information in the literature, suggest that the lung contains four separate compartments for ICAM-1 expression: vascular endothelial cells, alveolar macrophages (and lymphocytes), alveolar epithelial cells, and bronchiolar epithelial cells (and underlying smooth muscle). Evidence for inducible ICAM-1 expression is best established for endothelial cells and alveolar macrophages. The data in the present study indicate that it may be necessary to add bronchiolar epithelial cells (and underlying smooth-muscle cells) to this list. It would appear that TNF- is a powerful force for upregulating lung airway ICAM-1, as revealed by the ability of anti-TNF- antibody to greatly attenuate upregulation of airway-related ICAM-1. This effect of TNF- appears to closely resemble its effect in upregulating lung vascular ICAM-1 in rats with intrapulmonary deposition of IgG immune complexes (4).

In the LPS model used in the present study, upregulation of vascular and airway ICAM-1 shows a requirement for TNF-, the blocking of which reduces ICAM-1 expression by > 80%. ICAM-1 has previously been shown to be a requirement for neutrophil recruitment after airway instillation of LPS (1). On the basis of the data in the present study, ICAM-1 expression on macrophages is also regulated by TNF-, but in a different time frame than that of airway ICAM-1. The early peak (at 2 h) of alveolar macrophage expression of ICAM-1, followed by loss of ICAM-1 thereafter, suggests that macrophage ICAM-1 is not responsible for the measured increase in airway ICAM-1 at 6 h after instillation of LPS. Bronchiolar epithelial cells and underlying smooth-muscle cells in lungs instilled with LPS were positive for ICAM-1. Earlier studies have demonstrated ICAM-1 expression on vascular smooth-muscle cells after stimulation in vitro with IL-1, IL-6, and macrophage chemotactic protein-1 (28). The functional significance of ICAM expression by smooth-muscle cells remains unknown. In ischemic canine myocardium, ICAM-1 is upregulated, but staining for ICAM-1 in this model indicates localization of ICAM-1 in the intercalated discs between cardiac myocytes (29), a finding whose meaning is obscure.

The biologic significance of upregulated ICAM-1 expression in peribronchiolar smooth muscle remains to be determined. Upregulation of ICAM-1 on bronchiolar epithelial cells may well facilitate adhesion of recruited leukocytes (especially neutrophils) to these epithelial cells after airway instillation of LPS. Whether this puts the epithelial cells at risk of damage by the neutrophils remains to be determined.