Introduction
Lipopolysaccharide (LPS) is a component of the outer membrane of gram-negative bacteria, capable of inducing severe lung injury in sepsis or bacterial pneumonia upon release. Both conditions are predisposing causes of the acute respiratory distress syndrome (ARDS). Instillation of LPS into the lungs results in an acute inflammatory response orchestrated by the coordinated function of cytokines, chemokines, and adhesion molecules.
In vitro and
in vivo models of acute lung inflammation have been extensively used to elucidate the molecular mechanisms in endotoxin-induced lung injury [
1‐
3]. However, only sparse data exist about the exact role of effector cells such as alveolar macrophages (AM) in the respiratory compartment in this type of injury.
The airway compartment with alveolar macrophages and epithelial cells as the predominant cell types is a physiological barrier to a variety of environmental agents including gases, particulates and microbes. An injury to lung tissue involving the airway compartment can lead to serious illness and life-threatening conditions such as ARDS. AM are located at the air-tissue interface in the lung and are therefore the first cells, which interact with inhaled organisms and antigens [
4]. During the development of an acute inflammatory lung injury, AM play a very important role upon their activation. It has been shown in several lung injury models that activated pulmonary macrophages release the cytokines tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) as well as the chemokines monocyte chemoattractant protein-1 (MCP-1) and macrophage inflammatory protein-1β (MIP-1β) [
5]. Intercellular adhesion molecule-1 (ICAM-1, CD54) and vascular cell adhesion molecule-1 (VCAM-1, CD106) are adhesion molecules of the immunoglobulin superfamily. Adhesion to vascular endothelium mediated through ICAM-1/VCAM-1-integrin interactions is a key step in emigration of white blood cells to sites of inflammation [
6]. It has also been shown that they play a central role in adherence of effectors cells such as AM to the respiratory epithelium [
7]. All these inflammatory mediators together play a crucial role in the orchestration of an inflammatory response, particularly in neutrophil recruitment. Recent studies suggest that cytokines and chemokines do not only originate from AM, but also from other cells such as epithelial cells [
8].
In order to study the role and function of AM regarding inflammatory mediators in endotoxin-induced lung injury, AM were depleted by intratracheal application of dichloromethylene diphosphonate-(Cl
2MDP)-liposomes (clodronate liposomes) as phagolysozymes. Phagocytosis of clodronate-liposomes has been shown to result in the selective destruction of macrophages with an elimination rate up to 80% [
9]. Cellular and molecular immunopathologic characteristics in endotoxin-induced lung injury of non-depleted and AM-depleted animals were investigated, focusing on neutrophil recruitment and expression of inflammatory mediators. The hypothesis was tested if AM upon LPS stimulation have an anti-or proinflammatory character.
Materials and methods
Animals
Male pathogen-free 300 g Wistar rats (RCC Ltd, Füllinsdorf, Switzerland) were anesthetized with subcutaneous fentanyl-fluanisone (0.25 ml/kg) and medetomidine hydrochloridum (0.25 ml/kg). The animal care committees at the University of Zurich approved the experimental protocols.
Clodronate Liposomes
Clodronate liposomes were prepared as described [
10]. To verify any possible contamination with LPS, endotoxin levels were measured using an
in vitro analysis of endotoxin (Limulus-Test; Endotell, Allschwil, Switzerland). With the negative results of these tests any biologically relevant contamination with LPS could be ruled out. Clodronate is a synthetic bisphosphonate which has been used to selectively eliminate macrophages in various tissues. Uptake of liposomal-encapsulated clodronate has been demonstrated by macrophages and monocytes, undergoing apoptosis in response to clodronate. Each animal received a dose of 500 μg liposome-encapsulated clodronate. Empty liposomes were used as controls. Liposomes were diluted in 300 μl phosphate-buffered saline (PBS) and instilled into the trachea of the animals.
LPS Lung Injury
72 hours after liposome application animals were anesthetized again and lung injury was induced by intratracheal application of 150 μg LPS (Escherichia coli serotype 055:B5; Sigma Chemical Co, St. Louis, MO) in 300 μl PBS. Animals were sacrificed 4 hours or 24 hours later and lungs were analyzed.
For inhibition experiments rats received 100 μg intratracheally applied polyclonal neutralizing rabbit anti-rat MCP-1 antibodies or control IgG (both from Torrey Pines Biolabs, Houston, TX) 30 minutes after intratracheal instillation of LPS or PBS.
Bronchoalveolar Lavage Fluid (BALF)
Bronchoalveolar lavage was performed with 10 ml cold PBS. BALF was centrifuged and the supernatant aliquoted and stored at -20°C. Cells were counted and differentiated using a cytospin and Diff-Quick staining (Dade Behring AG, Düdingen).
Myeloperoxidase (MPO) Assay
In order to determine the parenchymal content of neutrophils, a MPO assay was performed as previously described [
11]. The vascular and respiratory compartment were flushed with PBS in order to eliminate blood cells, notably neutrophils.
ELISA Quantitation of TNF-α, MCP-1, and Chemokine-Inducible Neutrophil Chemoattractant-1 (CINC-1) in BALF
TNF-α, MCP-1, and CINC-1 were assessed in the BALF of animals using a standard enzyme-linked immunosorbent assay (ELISA) purchased from PharMingen, San Diego, CA (TNF-α and MCP-1) and from R&D Systems, Minneapolis, MN (CINC-1 Duoset).
Western Blot Analysis of TNF-α, MCP-1, MIP-1β, and Macrophage Inflammatory Protein-2 (MIP-2) in BALF
BALF of animals were loaded and electrophoresed in a 7.5% SDS-polyacrylamide gel. After separation, the proteins were transblotted to a nitrocellulose membrane. Incubation with a polyclonal goat anti-rat TNF-α, anti-rat MCP-1, and anti-MIP-2 antibody (all from Santa Cruz Biotechnology, Santa Cruz, CA) or with a polyclonal rabbit anti-rat MIP-1β (Santa Cruz Biotechnology, Santa Cruz, CA) was performed overnight at 4°C. A secondary horseradish peroxidase-labeled anti-goat or anti-rabbit IgG (1:5000) was added for 30 minutes at room temperature. Signals were detected by enhanced chemiluminescence.
Isolation and Reverse Transcription PCR Amplification of Lung mRNA
The vascular and respiratory compartments were flushed before using the lungs for determination of mRNA of various genes. Total RNA from lungs was extracted as previously described and analyzed by Reverse Transcription Polymerase Chain Reaction (RT-PCR) (Perkin-Elmer, Branchburg, NJ) [
7]. The primers used for gene analysis are summarized in Tab.
1. PCR was also performed with 18S primers to ensure equal loading and mRNA/18S ratios were calculated.
Table 1
Optimized conditions for RT-PCR
Gene | Primers | Fragments (bp) | Thermocycle Condition |
TNF-α | sense | 5'-ACT GAA CTT CGG GGT GAT TG-3' | 334 | 23 cycles; Tm: 57°C |
| antisense | 5'-GTG GGT GAG GAG CAG GTA GT-3' | | |
IL-1β | sense | 5'-AGC TGC ACT GCA GGC TTC GAG ATG-3' | 339 | 28 cycles; Tm: 68°C |
| antisense | 5'-GAA CTG TGC AGA CTC AAA CTC CAC-3' | | |
MCP-1 | sense | 5'-TAT GCA GGT CTC TGT CAC GC-3' | 255 | 24 cycles; Tm: 58°C |
| antisense | 5'-TTC CTT ATT GGG GTC AGC AC-3' | | |
MIP-1β | sense | 5'-CGT GTC TGC CTT CTC TCT CC-3' | 220 | 24 cycles; Tm: 58°C |
| antisense | 5'-CAC AGA TTT GCC TGC CTT TT-3' | | |
ICAM-1 | sense | 5'-AGG TAT CCA TCC ATC CCA CA-3' | 660 | 22 cycles; Tm: 58°C |
| antisense | 5'-CTT CAG AGG CAG GAA ACA GG-3' | | |
VCAM-1 | sense | 5'-CGG TCA TGG TCA AGT GTT TG-3' | 570 | 24 cycles; Tm: 57°C |
| antisense | 5'-ACC CTC ATG TAG CCT TGT GG-3' | | |
In vitro Assay with Alveolar Epithelial Cells (AEC)
AEC from rat lungs were collected following previously described protocols [
12]. Cell purity of AEC was more than 95%. AM were harvested following an earlier protocol [
7]. AEC were kept for 5 days in culture. Most of the cells showed type I phenotype. At the time of co-culture with AM, AEC were confluent. To each well of a 96-well plate with AEC 1 × 10
7 AM were added. The relation of AM to AEC was 1:2 (1 × 10
7 and 2 × 10
7), when the experiments were performed, according to previous studies [
13]. AM were allowed to settle on AEC for 30 min prior to the addition of LPS. As a control AEC and AM were stimulated separately. Supernatants were collected and ELISA (TNF-α, MCP-1) or Western blot (MIP-1β) was performed as described above.
Chemotaxis Assays
1 × 10
5 polymorphonuclear cells (PMN) of calcein-AM labelled neutrophils were given into MultiScreen-MIC filter plates [
14]. Receiver plates were loaded with 150 μl of BALF. To some of these wells, 10 μg/ml polyclonal neutralizing rabbit anti-rat MCP-1 (Torrey Pines Biolabs, Houston, TX) or 10 μg/ml control IgG (Torrey Pines Biolabs, Houston, TX) were added. For experiments with CINC-1 antibodies 10 μg/ml monoclonal mouse anti-rat CINC-1 antibody (R&D Systems, Europe Ltd., Abingdon, Oxon, UK) or 10 μg/ml MOPC-21 as control (mouse IgG, Pharmingen, San Diego, CA) were used. 0.1 μM of the chemotactic substance fMLP (Sigma-Aldrich, Buchs, Switzerland) was taken as positive control and PBS/1% fetal bovine serum (FBS) to determine basal migration. After 2 hours incubation at 37°C the filter was removed and migrated cells were lysed with Triton 1% (Sigma-Aldrich, Buchs, Switzerland). Fluorescence was measured by using an excitation filter at 485 nm and an emission filter at 535 nm. Ratio between stimulated and basal migration was used as indication of PMN chemotaxis.
Statistical Methods
For each group and/or time point 5 animals were used. The relative reduction in mRNA or protein expression was calculated for LPS clodronate-liposome results. LPS control liposome and LPS clodronate-liposome values were calculated relative to their respective PBS liposome values. LPS clodronate-liposome reduction or upregulation was calculated as the relative change compared to LPS control liposome results. ELISA contained three replicates with five different animals. Each data point in the graphs represents mean +/- standard error of mean (SEM). Analysis of variance (ANOVA) with post-ANOVA comparison was performed to assess the statistical significance of differences.
Discussion
AM are considered to play a predominant role in the inflammatory response to lung injury. An effective host defense is primarily dependent on this cell type [
18]. AM do not only act as phagocytes, but have also a potent function as secretory cells. Upon stimulation, AM release a wide variety of biologically active molecules, which are part of the complex regulatory system of inflammatory reactions. A recent study demonstrated that depletion of AM in a murine model of
Klebsiella pneumoniae-induced lung injury decreased survival dramatically by causing 100% lethality 3 days after infection, thereby demonstrating that AM exert a crucial role in this model [
17]. In contrast to these findings, the elimination of AM promoted pulmonary immunity in a model of
Listeria-induced lung injury [
19]. To further examine the role of AM in the pathogenesis of neutrophilic lung inflammation in the early phase of LPS-induced lung injury, macrophages were eliminated by administration of liposome-encapsulated clodronate. Increased neutrophil recruitment was found in depleted LPS-injured animals compared to non-depleted LPS-injured animals. We postulated that beside other neutrophil chemoattractants enhanced concentrations of MCP-1 as detected in the respiratory compartment might be responsible for this increase in neutrophil recruitment. Therefore, AM could have an anti-inflammatory effect at this early time point of injury suppressing MCP-1 production by AEC.
Neutrophil accumulation in the lungs is one of the hallmarks of an acute inflammatory response. The mechanisms by which PMN are recruited into the lung in response to inflammatory stimuli are unclear, but seem to be dependent on the release of chemotactic and activating cytokines and chemokines [
20]. These molecules recruit neutrophils and mediate their migration into the interstitial and finally into the alveolar space. Several models of lung injury have already been explored regarding neutrophil recruitment after AM-depletion. A model of AM-depletion with both intratracheally and intravenously delivered liposome-encapsulated clodronate showed decreased neutrophilic alveolitis in AM-depleted lungs after exposure to aerosolic LPS [
21]. In acute lung injury induced by deposition of immunoglobulin G immune complexes, myeloperoxidase activity was decreased by 60% in AMdepleted lungs compared to AM-competent animals [
16]. Similar results were seen in a model of
Pseudomonas aeruginosa infection [
22]. In a
Klebsiella model of lung injury, an increased number of neutrophils in AM-depleted animals was observed after AM depletion [
17]. This
K. pneumoniae-induced PMN recruitment was shown to be TNF-α-mediated. In our model with an increase of PMN upon AM depletion, beside CINC-1, MCP-1 might be a possible neutrophil chemoattractant in the respiratory compartment. Chemotaxis assays underlined this hypothesis.
The role of MCP-1 regarding neutrophil accumulation in inflammatory processes is controversial. The nature of the injury is likely to determine the effect of MCP-1 on PMN. MCP-1 has recently been shown to play a certain role in PMN recruitment in various experimental systems [
23‐
25]. Johnston et al. demonstrated an induction of neutrophil chemotaxis in mesenteric tissue upon MCP-1 superfusion. MCP-1 could also be shown to be involved in neutrophil accumulation during gastric ulceration. Vozzelli et al. performed blocking studies with anti-MCP-1 antibodies in newborn rats, exposed to hyperoxia [
26]. Animals injected with 25 μg of anti-MCP-1 showed reductions in macrophage as well as neutrophil cell counts. Decreasing macrophage and monocyte activation by treatment with anti-MCP-1 might indirectly reduce neutrophil influx by decreasing a significant cellular source of neutrophil chemokines. However, this group showed as well an attenuation of PMN recruitment in hyperoxic rats, treated with 5 μg anti-MCP-1 antibody, with an unchanged number of alveolar macrophages, possibly demonstrating the effect of MCP-1 on neutrophil recruitment. Maus et al. postulated that JE/MCP-1 together with endotoxin, when co-appearing in the alveolar compartment elict an early phase of lung inflammatory injury with increased neutrophil recruitment [
27]. Treatment of mice with only JE/MCP-1 induced a monocyte, but not a neutrophil influx into the bronchoalveolar compartment, with monocyte number peaking at 48 hours. Challenging mice with endotoxin in the presence of JE/MCP-1 led to an increase of neutrophils at 12 hours, while monocyte counts peaked at 48 hours. However, alveolar deposition of both MCP-1 and endotoxin was a prerequisite for the response as MCP-1 without accompanying injury did not provoke neutrophil emigration into the lung. Therefore, these studies do not fully underline a chemoattractant component of MCP-1 regarding PMN recruitment. Another study of Maus et al. reported decreased neutrophil and monocyte recruitment in alveolar macrophage-depleted animals upon combined JE/MCP-1/LPS treatment, also not supporting the concept of MCP-1 as a neutrophil chemoattractant [
28]. Notabely, the earliest time point after LPS application the authors focused on was 6 hours, and never at 4 hours.
While our data demonstrated enhanced PMN accumulation in LPS-injured AM-depleted animals, Koay et al. showed decreased neutrophil recruitment upon LPS stimulation in intratracheally and intravenously applied clodronate mice at 4 hours of injury. However, there are certain essential differences regarding the animal model. 1) Mice instead of rats were used. 2) An intratracheal and intravenous depletion was performed in this study, while we used only the intratracheal one. 3) Intratracheally applied LPS was aerosolized, while we instilled LPS, dissolved in PBS. 4) Much higher doses of LPS were used in mice than in rats, which could also explain a more severe inflammation and therefore an immediate participation of extraalveolar macrophages in the lung. Noteworthy is the observation, that also in the study of Koay MCP-1 levels were increased in bronchoalveolar lavage fluid in LPS-clodronate animals compared to LPS-control liposome animals. Although the methodological approach by Koay et al. and our experimental systems appear similar, two clearly different models were used.
While in the presence of AM the amount of MCP-1 in BALF was low in LPS-injured lungs, it increased dramatically in the absence of AM, leading to the assumption that epithelial cells are responsible for the majority of MCP-1 content in BALF, although a receptor-induced ligand internalization can not be excluded. The production of MCP-1 by epithelial cells has been demonstrated previously [
29]. Our
in vivo data lead to the new hypothesis, that AEC in the presence of AM are modulated in their activity in producing MCP-1. We therefore performed
in vitro assays with primary cultures of AEC and co-incubation of these cells with AM. After exposure to LPS, AEC/AM together produced less MPC-1 protein than LPS-stimulated AEC alone. This lead to the assumption that AM potentially exert a negative control effect on AEC. A similar observation was made with human bronchial epithelial cells (HBEC) and AM in culture, when exposed to ambient pollution particles [
30]. While most of the inflammatory mediators such as TNF-α, IL-1β and IL-6 were significantly increased upon stimulation of the HBEC/AM-co-culture, MCP-1 production was decreased. This
in vitro phenomenon could help to understand the increased levels of MCP-1 protein in the BALF of AM-depleted animals, although limitations of
in vitro models are well known and results have to be interpreted with caution.
Contrary to MCP-1, TNF-α protein in BALF of AM-depleted LPS-animals was decreased significantly as well as whole lung mRNA of TNF-α. A study with hyperoxia in AM-depleted rats showed no significant differences in lung TNF-α content [
31]. In contrast to the observation in our experiments in AM-depleted LPS-lungs, an increase of TNF-α in the AM-depleted
Klebsiella pneumonia lungs was observed 2 days after the onset of the injury [
17]. Several factors could explain these differing observations: 1. The difference of TNF-α content might reflect the difference in time points at which the lung injury was assessed. While TNF-α was evaluated at an early time point in the LPS-model, the analysis was performed in the
Klebsiella model after 2 days. Kinetic expression of TNF-α is known to be time-variable [
32]. 2. The mechanism of the inflammatory response to bacteria or toxins might bear different characteristics. 3. Alveolar macrophages have been shown to secrete large amounts of TNF-α, while interstitial macrophages are more efficient in releasing IL-1β and IL-6 [
33]. This finding supports our observation with a markedly reduced content of TNF-α protein in BALF and moderate decrease of whole lung mRNA for TNF-α. In our
in vitro data, AM are the dominant cells secreting TNF-α.
In the pulmonary tissue compartment – in contrast to the airway compartment – elimination of macrophages resulted in decreased production of mRNA for TNF-α, IL-1β, and MIP-1β in LPS-injured lungs, but only to a degree between 10 and 60%. These data suggest that other cells than AM contribute to tissue injury. The discrepant observation of minimal MCP-1 mRNA and protein decrease in whole lung and on the other side a MCP-1 increase in the BAL was observed in other models of lung injury [
21]. The most likely explanation for this discrepancy is that the inflammatory reactions do not bear the same characteristics in different compartments of the lung. mRNA of the adhesion molecules ICAM-1 and VCAM-1 was not changed. Therefore, it can be concluded that AM do not affect the expression of these important inflammatory mediators. This is in accordance with previous data showing that whole lung ICAM-1 is regulated by about 80% through TNF-α [
1]. In our studies, mRNA for TNF-α was only decreased by 32%, which might not be sufficient to elicit an ensuing change in ICAM-1 mRNA.
In order to define the role of AM at a later time point during LPS-induced lung injury, AM depletion was performed and lungs were collected 24 hours after LPS instillation. No difference in neutrophil recruitment between lungs with or without AM was observed. Two conclusions could be drawn from these experiments: 1) AM do not play the same critical functional role any more regarding PMN recruitment at this later time point of injury. As previously shown by Xing et al. AM play a key role in the very early LPS-induced lung inflammation [
34]. 2) New AM were recruited during 24 hours of injury. Depletion of AM has been observed to persist for 5 days, followed by an increase of AM (unpublished data). These data were evaluated without application of any injury. In the present model, however, recruitment of new macrophages in response to the injury can be assumed.
The current study provides evidence for an important protective role of AM very early in the course of LPS-induced lung injury. AM were shown to affect neutrophil recruitment by reducing the accumulation of PMNs, most likely via suppression of alveolar epithelial MCP-1 production. This observation was surprising and not in accordance with our previously stated hypothesis. The precise role and the mechanisms of AM in the up-or downregulation of inflammatory mediators remain to be further investigated. Therapeutic options for the effective management of endotoxin-induced lung injury are still very limited. Therefore, the detailed exploration of an animal model of LPS-induced lung injury with regard to the role of macrophages will be helpful for the determination of the mechanistic and activation pathways.
Authors' Contributions
BBS carried out animal studies and molecular biology studies and helped drafting the manuscript.
RS provided liposomes.
LR and CB performed molecular biology studies.
TP participated in the design of the study and helped drafting the manuscript.
RS and RCS was involved in the design and coordination of the study and helped drafting the manuscript.