Acute lung injury results in damage to the alveolar epithelium, leading to leak of proteins into the alveolar space and impaired gas exchange. Lung function can be restored only if the epithelial layer is restored. The process of reepithelialization requires migration of lung epithelial cells to cover denuded basement membranes. The factors that control the migration of lung epithelial cells are incompletely understood. We examined isolated murine type II alveolar epithelial cells (AECs) for expression of CC chemokine receptor 2 (CCR2) and functional consequences of the binding of the main CCR2 ligand monocyte chemoattractant protein-1 (MCP-1). We found that primary AECs bound MCP-1 and expressed CCR2 mRNA. These cells demonstrated functional consequences of CCR2 expression with migration in response to MCP-1 in chemotaxis/haptotaxis assays. Primary AECs cultured from mice lacking CCR2 did not respond to MCP-1. Monolayers of AECs lacking CCR2 demonstrated delayed closure of mechanical wounds compared with AEC monolayers expressing CCR2. Delayed closure of mechanical wounds of wild-type AECs was also demonstrated in the presence of anti-MCP-1 antibody. These data demonstrate for the first time that AECs express CCR2 and are capable of using this receptor for chemotaxis and healing of wounds. CCR2-MCP-1 interactions may be important in the process of reepithelialization after lung injury.
- monocyte chemoattractant protein-1
acute lung injury involves loss of alveolar epithelial cells (AECs) and impaired barrier function. Wound healing is believed to be important for recovery from lung injury. Normal healing is dependent on reepithelialization, a complex process requiring epithelial cell survival, proliferation, adhesion, and migration. Of these processes, cell migration is likely to be a critical component. However, there is limited information about the factors that control AEC migration after lung injury.
Monocyte chemoattractant protein-1 (MCP-1), also known as JE in the mouse, is a chemokine that is abundant in the lung. It plays an important role in the outcome of lung injury (2, 11). The receptor for MCP-1 is the CC chemokine receptor 2 (CCR2), a seven-transmembrane G protein-coupled domain receptor. MCP-1 is the major ligand for CCR2 and binds with high affinity (3, 12, 22), although other ligands, such as MCP-2, MCP-3, MCP-4, MCP-5, and HIV Tat, also use this receptor (1, 4, 7-10, 19, 21).
MCP-1 binding to CCR2 mediates chemotaxis of leukocytes and endothelial cells; however, there is no information about the ability of AECs to respond to MCP-1 or whether these cells express CCR2. We have found that CCR2 mRNA is expressed by murine AECs. Furthermore, we now report that MCP-1 binding to AEC CCR2 stimulates AEC migration and wound healing in vitro.
MATERIALS AND METHODS
Mice. CCR2+/+ (B6129F2/J; Jackson Laboratory, Bar Harbor, ME) and CCR2-/- mice (B6129F2-Cmkbr2tm1Kuz), bred at the University of Michigan, were housed under specific pathogen-free conditions in isolator cages. Clean water and food were given ad libitum. The animal care committees at the University of Michigan and the Ann Arbor Veterans Administration Medical Center approved all protocols.
AEC isolation. Primary cultures of murine type II AECs were isolated as previously described (5). Mice were sedated with pentobarbital sodium by intraperitoneal injection and secured to a dissecting board; the inferior vena cava was cut, and the animal was exsanguinated. The anterior thorax and ribs were removed, and the left ventricle was cut with scissors. The pulmonary vasculature was perfused with normal saline via a direct right ventricular puncture until the lungs were visually free of blood. The trachea was cannulated with a 20-gauge intravenous catheter, which was secured with a suture. The lungs were filled with 1-2 ml of dispase via a syringe connected to the tracheal catheter. The syringe was removed, the lungs were allowed to return to a resting volume, and low-meltingpoint agarose (1%, 0.45 ml prewarmed to 45°C) was slowly added via the tracheal catheter. The lungs were suspended in ice-cold PBS for 2 min and then transferred to a conical tube containing 2 ml of dispase solution at room temperature for 45 min. The lungs were chilled briefly in ice-cold PBS and then transferred to a sterile petri dish containing DMEM with 0.01% DNase. The lung tissue was teased away from the airways with forceps. The solution was transferred to a trypsinizing flask with a magnetic stir bar and gently stirred for 10 min. The suspension was filtered successively through 100-, 40-, and 25-μm nylon mesh filters to create a single-cell suspension. The cells were collected by centrifugation, counted, and resuspended in DMEM with biotinylated anti-CD32 (FcγR, 0.65 μg/106 cells) and anti-CD45 (common leukocyte antigen, 1.5 μg/106 cells). After incubation at 37°C for 30 min, the cells were pelleted, counted, resuspended in 7 ml of DMEM, and added to prewashed streptavidin-coated magnetic particles. The mixture was incubated for 30 min at room temperature with gentle rocking. The tube was attached to a magnetic separator for 15 min to remove bone marrow-derived cells. The cells not bound with magnetic particles were recovered from the tube with a glass pipette, pelleted, and suspended in culture medium. Viability was >97% by trypan blue exclusion. The cells were plated overnight in 60-mm culture plates. Nonadherent cells were recovered, counted, and plated as described below.
Binding of MCP-1. Freshly isolated AECs were incubated with mouse serum for 15 min at 40°C. Cells were centrifuged to remove serum. Ten microliters of biotinylated MCP (R&D Systems) were added to 25 μl of the washed cell suspension in a 12 × 75-mm borosilicate tube. As a negative staining control, an identical sample of cells was stained with 10 μl of biotinylated soybean trypsin inhibitor as a negative control. The cells were incubated for 60 min at 20°C, and 10 μl of avidin-FITC reagent were added to each tube. This mixture was incubated for 30 min at 20°C in the dark. The cells were washed twice with 2 ml of 1× RDF1 buffer (supplied by manufacturer, R&D Systems) to remove unreacted avidin-fluorescein and resuspended in 0.2 ml of 1× RDF1 buffer for flow cytometric analysis. For specificity testing, 20 μl of anti-mouse JE blocking antibody were mixed with 10 μl of mouse JE-biotin and allowed to incubate for 15 min at room temperature. The stained cells were pretreated with purified mouse Ig (10 μl of 1 mg/ml per 106 cells) for 15 min at room temperature to block Fc-mediated interactions. Cells were added to the tube containing the anti-mouse JE-blocking antibody-fluorokine mixture. Cells were incubated for 60 min, and avidin-FITC reagent was added to each tube as described above. The final control consisted of unstained cells to which avidin-FITC was added. Cells were analyzed for chemokine receptor expression using an EPICs XL cytometer (Coulter, Palo Alto, CA). In each sample, ≥5,000 cells were analyzed. Results are expressed as percentage of MCP-1-bound positive AECs.
Chemotaxis/haptotaxis. Standard Boyden chamber assays were performed with freshly isolated AECs as previously described (16) with minor modifications. The chamber filters (12-μm pore size, Nucleopore) were precoated with fibronectin for 1 h. AECs were placed in the upper chamber. For chemotaxis experiments, the chemoattractant was placed in the lower chamber. For haptotaxis experiments, the filters were precoated with chemoattractant for 1 h and then washed before addition of AECs. Chambers were incubated with cells for 1 h at 37°C in 7.5% CO2. The filters were removed, fixed, stained, and counted as described elsewhere (16). Data are expressed as cells per high-power field.
RT-PCR. AECs were purified from CCR2+/+ or CCR2-/- mice as described above. Total RNA was prepared from 500,000 AECs, and 1 μg of total RNA was subjected to RT-PCR analysis for CCR2 expression using the following primers: GTATGACTACGATGATGGTGAGCCTTG (sense) and GAGCCTCACAGCCCTGTGCCTCTTC (antisense) for CCR2 and GTGGGGCTCCCCAGGCACCA (sense) and GCTCGGCCGTGGTGGTGAAGC (antisense) for β-actin. After cDNA synthesis, CCR2 mRNA was amplified for 35 cycles at 95°C for 1 min, 55°C for 1 min, and 68°C for 1 min following the directions of the Promega Access RT-PCR kit. β-Actin was amplified for 25 cycles. RT-PCR products were separated on a 1.5% agarose gel and transferred to Zetaprobe nylon membranes (Bio-Rad) using vacuum suction and 0.4 N NaOH. Southern blots were performed with 32P-labeled internal oligonucleotide primers. Probe primer for CCR2 was CAAACACAGCCACCACCCAAGTG. The β-actin probe sequence is GGGACGACATGGAGAAGATCTGG. Products were visualized by autoradiography.
Wounding assay. AECs (2 × 105 cells/well) were placed in a fibronectin-coated 96-well culture plate for 2 days in medium consisting of DMEM, penicillin-streptomycin, and 10% FBS at 37°C in 7.5% CO2. The medium was removed, and the monolayers were washed with PBS. The monolayer was wounded using a single pass with a sterile P-1000 pipette tip. The medium was returned to the wells. In some experiments, anti-MCP antibody (8 μg/ml) or control antibody was added to the medium after wounding of the monolayers. Individual wells were viewed with the ×4 objective on an Olympus CK2 inverted microscope. An image of the entire wound area was captured using an Olympus DP10 digital camera, and the plate was returned to the incubator. Digital images of each well were repeated at 24 and 48 h. The area of the wound was measured from the digital image using NIH Image 1.62 software. This method of injury led to a consistent baseline injury area (e.g., 74.6 ± 6.7 arbitrary units, n = 24), so data from all wells were included in the analysis. The data are expressed as percentage of baseline. Each condition was examined in six wells.
Statistics. Values are means ± SE. Groups were compared using a two-tailed t-test for unpaired samples with Statview 4.5. Comparisons between multiple groups were performed using ANOVA. Comparisons were deemed statistically significant at P < 0.05.
MCP-1 binds to AECs. We previously showed that MCP-1 plays an important role in the resolution of acute lung injury (14). Although MCP-1 is known for its role as a chemoattractant for mononuclear cells and is produced by AECs, it is unknown whether AECs can recognize MCP-1. To explore this idea, AECs isolated from CCR2+/+ mice were incubated with biotinylated MCP-1 or a biotinylated irrelevant control protein control and then in streptavidin-FITC and analyzed by flow cytometry. AECs bound MCP-1 (Fig. 1). Specificity was demonstrated by blocking MCP-1 binding with a specific anti-MCP-1 antibody. The entire population shows a mean fluorescence intensity shift from ∼30 to 200. This indicates that AECs express a receptor that binds MCP-1.
CCR2 is expressed by AECs. CCR2 is the main receptor for MCP-1. RT-PCR was used to demonstrate expression of CCR2 on AECs (Fig. 2). CCR2 mRNA is expressed from AECs. As a control, AECs were isolated from CCR2-/- mice. Isolation of cells from these mice resulted in yield and purity similar to CCR2+/+ mice (data not shown). As expected, these cells did not express CCR2 mRNA (Fig. 2).
AECs exhibit chemotaxis in response to MCP-1. Chemotaxis assays were performed to evaluate the functional consequences of CCR2 expression on AECs. AECs from CCR2+/+ mice demonstrated chemotaxis in response to MCP-1 (Fig. 3A). A checkerboard analysis demonstrated that this response was due to directed migration, and not to chemokinesis. There was a dose-dependent response that appeared to be maximal at 1 ng/ml. The isolation procedure provides highly enriched populations of type II AECs (5) that include <5% nonepithelial cells by vimentin staining. To ensure that cells found at the leading edge of the chemotactic front were not macrophages, we stained filters with the lectin BS-1, which binds murine alveolar macrophages (17). Less than 1% of the cells at the leading edge of chemotaxis stained with BS-1 (data not shown), indicating that migration observed in response to MCP-1 was not due to a minority population. Thus AECs demonstrate chemotaxis in response to MCP-1.
AEC chemotaxis is dependent on expression of CCR2. To evaluate the role of CCR2 in AEC chemotaxis, we isolated AECs from CCR2-/- mice. When CCR2-/- AECs were subjected to the chemotaxis assay, there was no response to MCP-1 over the entire dose range (Fig. 3B). CCR2-/- AECs did migrate in response to granulocyte-macrophage colony-stimulating factor (GM-CSF, 10 ng/ml), a different chemotactic signal (13 ± 2 and 2.2 ± 0.9 cells/high-power field for GM-CSF and control, respectively, P < 0.001). This demonstrates that the defect in chemotaxis in CCR2-/- AECs is specific for MCP-1.
AECs undergo haptotaxis in response to MCP-1. MCP-1 can bind to matrix proteins. We have demonstrated measured increased levels of MCP-1 by ELISA in fibrotic lungs (14) and have seen increased staining for MCP-1 in fibrotic lesions (data not shown). To repair epithelial wounds, cells migrate over a provisional matrix likely to contain bound chemokines induced in response to injury. We therefore tested the capacity of CCR2+/+ and CCR2-/- AECs to migrate on MCP-1 bound to fibronectin in a haptotaxis assay. CCR2+/+ AECs demonstrated significant haptotaxis on membranes sequentially coated with fibronectin and MCP-1 (P < 0.0001 compared with fibronectin alone). No haptotaxis was observed with CCR2-/- AEC (Fig. 4). These data demonstrate that AECs can migrate in response to MCP-1 bound to extracellular matrix proteins.
CCR2 expression accelerates AEC wound healing in vitro. Covering exposed basement membrane is thought to be a critical event in normal healing after acute lung injury. This reepithelialization is a complex process requiring many cellular events. However, migration to cover denuded surfaces is a critical step. To model this process in vitro, we developed a model of mechanical wounding of AEC monolayers. In preliminary experiments, we found that closure of standard wounds required ∼48 h. Using this assay, we performed wounding experiments with confluent monolayers of CCR2+/+ and CCR2-/- AECs. Monolayers of CCR2-/- AECs were visually indistinguishable from CCR2+/+ AECs. Wounded monolayers of CCR2+/+ AECs covered the denuded area faster than CCR2-/- AECs measured at 1 day (36.9 ± 3.4 vs. 25.5 ± 3.1%, P < 0.05) and 2 days (95.2 ± 1.2 vs. 76.9 ± 3.4%, P < 0.0005; Fig. 5) after wounding of the monolayers. Thus CCR2 participates in the normal closure of mechanical wounds in epithelial cell cultures.
Blocking MCP-1 inhibits AEC wound healing in vitro. Having demonstrated a role for CCR2 in the closure of a mechanical wound in epithelial cell monolayers, we next assessed the role MCP-1 may play in this process. Two experimental approaches were used to address this question. MCP-1 was added to wounded monolayers but did not change the rate of closure of mechanical wounds over a wide range of concentrations (data not shown). This may be explained by the presence of high levels of MCP-1 at baseline in AEC cultures (18). Indeed, significant MCP-1 was measured in conditioned medium obtained from wounded and control wells at 24 h (1.5 ± 0.1 vs. 1.0 ± 0.05 ng/ml, n = 6) and 48 h (2.0 ± 0.1 vs. 1.9 ± 0.2 ng/ml, n = 6). This concentration of MCP-1 is similar to the concentration that led to maximal chemotaxis (Fig. 3A). However, when MCP-1 antibodies were added to wounded monolayers, the wound closure was significantly delayed: 52.1 ± 1.8 and 30.1 ± 4.5% for anti-MCP-1 and control, respectively, on day 1 (P < 0.002) and 99.4 ± 0.4 and 73.5 ± 7.0% for anti-MCP-1 and control, respectively, on day 2 (P < 0.005; Fig. 6). Thus blocking endogenous MCP-1 significantly inhibited the closure of mechanical wounds in AEC culture.
In these studies, we examined the expression and functional significance of CCR2 expression on murine type II AECs. There are four key findings. 1) AECs express CCR2. 2) CCR2 expression on AECs mediates chemotaxis/haptotaxis in response to MCP-1. 3) Healing of a mechanical wound in AEC monolayers is delayed in the absence of CCR2. 4) Blocking endogenous MCP-1 inhibits closure of a mechanical wound in CCR2+/+ AEC monolayers. Taken together, these findings suggest a role for MCP-1/CCR2 in the process of reepithelialization that may have important implications in determining the outcome of acute lung injury.
These results are the first to demonstrate CCR2 expression in murine AECs. Expression of CCR2 has been described on a variety of cell types, including leukocytes, endothelial cells, and epithelial cells. There is a single report of expression of CCR2 in airway epithelial cells in patients with chronic obstructive lung disease (6). The present studies are the first to describe CCR2 expression on epithelial cells in the distal lung.
CCR2 expression on murine AECs has functional signifi-cance. The major ligand for CCR2 is MCP-1. Our data confirm that MCP-1 binds to AECs. AECs respond to MCP-1 with chemotaxis and haptotaxis. Not surprisingly, the response to MCP-1 is dependent on the expression of CCR2, because AECs lacking CCR2 do not respond to MCP-1.
CCR2 expression plays a role in mechanical wound closure in vitro. The epithelium of the lung is subjected to a variety of insults. Repair of the wounded epithelium is a critical event in normal repair after acute lung injury. Closure of mechanical wounds in vitro can provide important information about cellular migration, which is believed to be a key factor in healing epithelial wounds. Our studies show that wounds induced in alveolar epithelial cultures from mice lacking CCR2 healed more slowly than wounds in cultures of cells obtained from wild-type mice. Blocking the major CCR2 ligand MCP-1 had a similar effect. To our knowledge, this is the first report implicating CCR2 in epithelial cell wound healing. We have not measured time to heal skin wounds in CCR2-/- animals and have not informally observed differences in healing time of skin wounds (14). However, our present data suggest that CCR2, perhaps by binding MCP-1 bound to provisional matrix proteins, mediates the migration of epithelial cells in response to mechanical injury.
After lung injury, repair of the epithelium appears to be essential in preventing pulmonary fibrosis (20). Multiple factors in the alveolus may influence the process of reepithelialization. The present in vitro model suggests that MCP-1/CCR2 has an important effect on the rate of wound closure. This may be true in the lung as well; however, other factors are present that can affect the healing process. For instance, GM-CSF is present and plays a protective role in the outcome of lung injury (13). GM-CSF is also a chemoattractant for AECs. Thus healing the epithelium in CCR2-/- mice could be controlled by factors other than MCP-1, resulting in a more normal epithelium. Furthermore, these data demonstrate that AECs have receptors for MCP-1 and open the possibility that this cytokine may have additional effects on AEC activities beyond its effects of haptotaxis and migration. As an example, we previously showed that exposure of AECs to MCP-1 decreased the production of PGE2 (15). This relative deficiency of PGE2 may explain the relative protection from fibrosis in the absence of CCR2 signaling (14). The ability to isolate primary cultures of AECs from knockout mice provides a powerful tool to examine epithelial cell behavior and elucidate the role of these molecules during conditions of wound healing.
We previously showed that CCR2/MCP-1 plays a role in the development of pulmonary fibrosis in bleomycin- and FITC-treated mice (14). The mechanism of this effect is complex and may involve cross talk between epithelial cells and fibroblasts (15). The data presented here suggest another role for CCR2/MCP-1 in epithelial cell migration. Although these data appear to be at odds with previous data showing a protective role for CCR2 in animal models of fibrosis, it may uncover another level of complexity for reepithelialization after lung injury. It is possible that reepithelialization in the lung, where a delicate thin barrier between blood and air space is maintained, requires a highly ordered repair. In the context of severe lung injury, significant amounts of provisional matrix are deposited. Regenerating epithelial cells might use MCP-1 bound in provisional matrix as a signal to migrate in a disordered attempt to repair the damaged epithelial cell surface. This might result in an alveolar surface that is incapable of gas exchange. In the absence of CCR2 (or a deficiency of MCP-1) or the timely resolution of provisional matrix, the regeneration of the epithelial surface proceeds in an orderly fashion. The result is a repaired epithelial cell surface capable of gas exchange. Further studies are needed to examine this possibility.
In summary, CCR2 is expressed on AECs. It can bind MCP-1 and mediate AEC chemotaxis/haptotaxis. Wound closure of a mechanical wound in AEC cultures is delayed in the absence of CCR2 and in the presence of blocking antibodies to MCP-1. These data point to a possible role for CCR2/MCP-1 in mediating reepithelialization and, thereby, affecting the outcome after lung injury. The ability to isolate pure populations of epithelial cells from mice lacking CCR2 represents a powerful tool to explore the role of this receptor in the complex process of reepithelialization after lung injury.
This work was supported by National Heart, Lung, and Blood Institute Merit Review Award 1P50 HL-56402 (to P. J. Christensen and R. Paine III) and Research Enhancement Award Program funds from the Department of Veterans Affairs.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society