| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Pathology, University of Michigan, Ann Arbor, MI 48109-0602;
Unit of Industrial Toxicology and Occupational Medicine, Université Catholique de Louvain, Brussels, Belgium; and
Department of Pathology, McMaster University, Hamilton, Ontario, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression was evident. Purified lung eosinophils from blm-treated IL-5TG mice stimulated 
-smooth muscle actin and collagen expression in mouse lung fibroblasts, without affecting proliferation. Furthermore instillation of purified eosinophils into murine lungs resulted in extension of blm-induced lung fibrosis, thus confirming a role for eosinophils. However, lung T lymphocytes from blm-treated IL-5-/- mice were able to stimulate fibroblast proliferation but not -smooth muscle actin or collagen expression. Blocking T cell influx by anti-CD3 Abs abrogated lung fibrosis, thus also implicating T lymphocytes as a key participant in fibrosis. Pulmonary fibrosis in IL-5TG mice was preferentially associated with type 2 cytokines (IL-4 and IL-13), whereas fibrotic lesions in IL-5-/- animals were accompanied by proinflammatory cytokine (TNF-, IL-1, and IFN-
) expression. We suggest that eosinophils and T cells contribute distinctly to the development of blm-induced lung fibrosis potentially via their production of different cytokine components, which ultimately induce TGF- expression that is intimately involved with the fibrosis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lung fibrosis can be defined as uncontrolled or abnormal wound healing characterized by a vigorous replication of mesenchymal cells, accumulation of myofibroblasts, and exuberant deposition of extracellular matrices (3). Leukocytic infiltration is also a characteristic of lung fibrosis and is believed to contribute, at least in part, to the fibrotic response (1). Among these recruited cells, macrophages and their profibrotic activity have been the focus of intense research over the past 20 years. It is now clear that macrophages generate and modulate inflammation and fibrosis extension as well as alterations in the cytokine milieu (4). However, the exact role of other cells, such as lymphocytes, neutrophils, and eosinophils also recruited during the establishment of many forms of fibrosis, is still not fully delineated and remains a controversial topic in the published literature (5, 6, 7, 8). Interestingly, all of these immune cells are, as the macrophages, potent sources of mediators capable of activating fibroblast functions, such as cytokines; hence, these immune cell populations have the potential of playing comparable roles.
The importance of cytokines produced during fibrosis is well documented, demonstrating that these mediators have the potential of orchestrating and amplifying inflammation and fibrotic responses (9). For instance, several studies have demonstrated the key role of proinflammatory cytokines such as TNF- and IL-1 (10, 11) as well as growth factors such as TGF- and platelet-derived growth factor (PDGF)3 (12, 13). Exaggerated lung responses to these cytokines may represent key events in the development of pulmonary fibrosis (14). Recently, human and experimental studies have highlighted the potential profibrotic roles of additional cytokines, including IL-4, IL-10, and IL-13, all classified as belonging to the class of Th2 type cytokines (15, 16, 17).
Recent reports have identified the importance of another Th2 cytokine, IL-5, in eosinophil recruitment to the lung in bleomycin (blm)-induced lung inflammation and fibrosis (7, 18). To clarify the role of IL-5/eosinophils, further in vivo experiments were performed with transgenic mice overexpressing IL-5 (IL-5TG), with IL-5 expressing adenoviral construct (AdIL-5) and with mice deficient in IL-5 (IL-5-/-). Analysis of the amplitude of fibrosis demonstrated the presence of marked blm-induced lung fibrosis in IL-5TG and in AdIL-5 mice but paradoxically also in blm-treated IL-5-/- mice. Pulmonary fibrosis was accompanied in IL-5 overexpression models by a massive influx of eosinophils, whereas the lung response of knockout mice was characterized by specific T cell accumulation. To define the exact role in fibrosis of these two recruited immune cell populations, we dissected their separate biological activity both in vitro and in vivo as well as their specific cytokine production profiles.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 (IL-5+/+) and IL-5-/- mice on C57BL/6 background (19) were obtained from The Jackson Laboratory (Bar Harbor, ME). IL-5TG mice were obtained from Glaxo Wellcome (Stevenage, U.K.) (20). Their controls, CBA/Ca (wild-type IL-5) were purchased from Harlan Olac (Bichester, U.K.). Blm (Blenoxane, Mead Johnson, NJ) was suspended in sterile PBS at 1 U/ml. Either 0.02 or 0.05 U was administered by transoral instillation (1 or 2.5 µl/g of mouse, respectively).
Bronchoalveolar lavage (BAL) and whole lung homogenates
BAL was performed with 1 ml of sterile 0.9% NaCl. The BAL fluid (BALF) was centrifuged (1200 rpm, 10 min, 4°C) and the cell-free supernatant used for albumin measurements. BAL was then repeated twice with 2 ml of sterile 0.9% saline. Cells from all the lavage fractions were combined and cell differential counts determined (Diff-Quik staining; Dade Behring, Deerfield, IL). Separately, nonlavaged whole lungs were excised and homogenized on ice in 2 ml of cold 0.9% NaCl. After centrifugation at 4°C (10,000 rpm, 15 min), the supernatants were kept frozen at -80°C until use.
Collagen assays
Collagen deposition was estimated by measuring the hydroxyproline and soluble collagen contents of whole lung homogenates. Hydroxyproline was assessed by colorimetric analysis as previously described (15). Soluble collagen levels were estimated by Sircol collagen assay following the manufacturer’s protocols (Biocolor, Westbury, NY).
Cytokine assays and ELISA
TNF-, IL-1, IFN-, IL-4, and IL-13 concentrations were measured by ELISA kits (R&D Systems, Minneapolis, MN) following the manufacturer’s protocols. TGF- was measured by an assay using mink lung epithelial cells stably transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct (generous gift of Dr. D. B. Rifkin, New York University, New York, NY) as previously described (21). The relative mRNA expression of 20 cytokines was analyzed in freshly purified eosinophils and T cells (see below) with chemiluminescent GEArray gene arrays (SuperArray, Bethesda, MD) according to the manufacturer’s protocol. The gene expression of the following cytokines or cytokine receptors were analyzed: IFN-, IL-12p40, IL-12p35, IL-18, TNF-, IL-1, IL-4, IL-5, IL-9, IL-10, IL-13, IL-16, IL-3, eotaxin, eotaxin-2, RANTES, monocyte chemoattractant protein (MCP)-1, MCP-3, TGF-1, -2, and -3, CCR2, CCR3.
Albumin levels were measured in BALF by ELISA quantification kit provided by Bethyl Laboratories (Montgomery, TX). Fibronectin, type I collagen and -smooth muscle actin (-SMA) were measured using a standardized ELISA already detailed in a previous study (15).
Isolation and culture conditions of pulmonary eosinophils and T lymphocytes
Lungs from mice were digested enzymatically to obtain cell suspension as already detailed elsewhere (15). For eosinophil purification, the digestion solution also contained 5 ng/ml IL-5 and GM-CSF (R&D Systems). Lymphocytes and granulocytes were isolated by density centrifugation in 40% Percoll. Eosinophils and T cells were further isolated using immunomagnetic beads and the magnetic cell separation system (MACS; Miltenyi Biotec, Auburn, CA). Eosinophils were purified by negative selection using rat IgG Abs directed against cells such as: erythrocytes (anti-Ter 119; PharMingen, BD Biosciences, Bedford, MA), lymphocytes (anti-CD3, anti-CD4, anti-CD19; Caltag Laboratories, Burlingame, CA; anti-CD8a, anti-CD90 Thy1.2, NK1.1; PharMingen, BD Biosciences), and neutrophils (anti-GR-1; PharMingen, BD Biosciences). Goat anti-rat IgG microbeads were used to recognize primary rat IgG Ab. The purity of eosinophil preparations was >95% (Diff-Quik staining, Dade Behring). T lymphocytes were isolated by positive selection with anti-CD90 (Thy-1.2) magnetic beads. The resulting lymphocyte purity was >90%. Purified eosinophils and T cells were resuspended at 2 x 106/ml in complete RPMI medium supplemented with 10% FBS and antibiotics then plated at 0.2 ml/well in 96-well plates with calcium ionophore (A23187, Sigma-Aldrich, St. Louis, MO) and anti-CD3 Ab (BD Biosciences), respectively, as stimulators. After 24 (eosinophil cultures) and 48 h (lymphocyte cultures), supernatants of cell cultures were collected and analyzed by ELISA for cytokine secretion.
Adenoviral constructs
AdIL-5-expressing and control adenoviral (AdCTL) constructs were prepared as previously described (22). A dose of 0.5 x 108 PFU of AdIL-5 or AdCTL vector was diluted in 50 µl of PBS and delivered in mice by transoral instillation 4 days after blm treatment (0.02 U/mouse). This approach allowed us to maintain high lung expression of IL-5 and eosinophilia, until at least day 14 after blm treatment. For example, on day 14, the lung IL-5 levels in mice transfected with AdIL-5 were 648 ± 69 pg/lung compared with 7 ± 3 pg IL-5/lung in mice transfected with AdCTL. The corresponding lung eosinophil counts at the same time point were 2.5 ± 0.3 x 106/lung and 0.1 ± 0.04 x 106/lung, respectively, for AdIL-5 and AdCTL transfected mice.
In vivo injection of eosinophils and anti-CD3 treatment
Eosinophils from blm-treated IL-5TG mice were purified as previously described. Naive and blm-treated (0.01 U/mouse) wild-type mice were administrated by transoral instillation with 5 x 106 eosinophils at days 7 and 14 after blm treatment. Intensity of pulmonary fibrotic lesions was analyzed at day 21.
To deplete T cells, hamster anti-CD3 mAb (clone 145-2C11) and control hamster IgG anti-TNP-PE (clone A19-3) were purchased from PharMingen (BD Biosciences). Groups of 5–10 mice were injected i.p. with a total of 150 µg of anti-CD3 or hamster IgG as control (3 x 50 µg, every 7 days and starting 1 day before blm instillation). This dosing regimen is known to cause T lymphocyte depletion for up to 2 wk (5). Lung fibrosis was assessed 21 days after blm administration.
Mouse lung fibroblast culture
Mouse lung fibroblasts were isolated from lung tissue by mincing and enzymatic digestion as previously described (15). Cells were cultured in complete medium composed of DMEM (Life Technologies, Grand Island, NY) supplemented with 10% plasma-derived serum (Cocalico Biologicals, Reamstown, PA), human recombinant PDGF-BB (5 ng/ml, R&D Systems), recombinant human epidermal growth factor (10 ng/ml, R&D Systems), insulin, transferring and selenium liquid media supplement (1:100, Sigma-Aldrich), and antibiotics. Fibroblasts used after the first passage were seeded into 24-well or 96-well plates at 40 or 10 x 103 cells/wells, respectively. Subconfluent cell monolayers were treated for 24 h with various concentrations of recombinant cytokines (R&D Systems) or with eosinophils and T cells suspended in medium supplemented with 0.5% plasma-derived serum. Cocultures were performed with 2 x 106/ml of purified lung eosinophils or T cells. Fibroblast proliferation was estimated by [3H]thymidine incorporation in 96 wells. Type I collagen and -SMA were measured by ELISA after sonication of the lung fibroblasts cultivated in 24-well plates.
Flow cytometry
The following rat anti-mouse mAbs were used: anti-CD3, anti-CD4, anti-CD19 (Caltag Laboratories), anti-CD8a, anti-CD90 (Thy1.2), anti-NK1.1, and anti-CD62L (PharMingen, BD Biosciences). Cell-fixed Abs were recognized by a secondary R-PE-conjugated goat anti-rat Ig polyclonal Ab (PharMingen, BD Biosciences). R-PE-conjugated hamster anti-mouse 
- and -chain T cell receptor (PharMingen, BD Biosciences) were also used. After staining, cells were fixed in paraformaldehyde (1.25%) and 104 cells/sample were analyzed on a FACScan apparatus (BD Biosciences). Analysis of the lymphocyte population was undertaken with appropriate gating according to side and forward light scatter to exclude granulocytes, macrophages, and dead cells.
Histology
Animals were euthanized and perfused via the right ventricle with saline. Lungs were inflated with 1 ml 10% neutral-buffered formalin and fixed overnight. After dehydration in 70% ethanol, the lungs were then processed using standard procedures and embedded in paraffin. Sections were cut, mounted on slides, and stained with H&E or Masson Trichrome stain.
Statistics
Treatment-related differences were evaluated using Student’s t test or one-way ANOVA, followed by pairwise comparisons using the Student-Newman-Keuls test, as appropriate. For flow cytometry data, statistical analyses were performed by Mann-Whitney U test for unpaired values using Instat software (GraphPad Software, San Diego, CA). Statistical significance was considered at p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We first determined by ELISA the level of IL-5 in lung homogenates of mice administered with standard doses of blm. Seven days following instillation of 0.02 U/mouse of blm, a significant increase in pulmonary IL-5 content was noted in CBA/Ca wild-type (IL-5WT) mice in comparison to saline-treated control mice (Fig. 1A). At all other time points studied, IL-5 levels were comparable to the saline values. In blm-treated IL-5TG animals, IL-5 expression was dramatically increased (>10-fold increase over wild type) at each time point studied in comparison to the transgenic mice injected with saline (Fig. 1A). In contrast and as expected, IL-5 was undetectable in lungs of blm-treated IL-5 knockout mice (IL-5-/-), whereas their matching wild-type C57BL/6 controls (IL-5+/+) showed significant elevations at days 3 and 7 after blm treatment (Fig. 1B).
|
To assess the amplitude of lung injury and inflammation induced by blm (0.02 U/mouse), we compared the lung wet weight, the lung concentration of albumin, and the number of inflammatory/immune cells present in the lung at 7 days after blm instillation. The lung weight and BALF albumin levels were increased by blm treatment, but the increase was significantly higher in IL-5TG mice than in the treated IL-5WT animals (Table I). Analysis of lung immune cell subset distribution revealed an abnormally high eosinophil accumulation in the BALF of blm-treated IL-5TG mice in comparison to their corresponding treated wild-type mice (Table I). BALF neutrophils were absent in IL-5TG mice and macrophages as well as lymphocyte numbers were slightly increased compared with that observed in IL-5WT (Table I). We also compared the response of IL-5TG mice vs IL-5WT mice to administration of a higher dose of blm, which induced lethal lung injury. Following instillation of 0.05 U/mouse blm, mice of both strains began to die by day 10. However by day 21 post-blm instillation, cumulative mortality in IL-5TG mice was greater than that noted in IL-5WT animals (100% in IL-5TG compared with 80% in IL-5WT). All together, these data indicate that in response to blm treatment, IL-5 overexpressing transgenic mice exhibited more acute pulmonary lesions associated with a specific/selective and massive eosinophil accumulation.
|
and - epitopes) did not differ significantly between the two strains treated with blm (data not shown). However, when absolute counts for the different cell subsets were analyzed, the number of T cells (CD90.1.2+, CD3+, CD4+, and CD8+, TCR- and - T cells) in lung interstitium of blm-treated IL-5-deficient mice was significantly higher than those in the corresponding wild-type mice (Table II). No significant difference in the number of lung tissue NK and B cells was observed between the two strains of mice treated with blm (Table II). IL-5-/- mice were relatively resistant to lethal pulmonary injury induced by high-dose blm (0.05 U/mice) compared with treated IL-5+/+ mice. Thus the mortality rate in the deficient animals at day 21 was only 50%, whereas it reached 100% in IL-5+/+ mice. Collectively these results suggest that deletion of the IL-5 gene had an important effect on the lung response resulting in significantly reduced blm-induced lung injury and mortality. The lung responses in IL-5-deficient animals were also characterized by a relative absence of eosinophilia but a more pronounced T cell influx.
|
The amplitude of the pulmonary fibrosis induced by a standard dose (0.02 U/mouse) of blm was determined at days 7, 14, 21, and 28 by measuring lung hydroxyproline, soluble collagen, fibronectin, and TGF- contents as well as by histology. In comparison to blm-treated IL-5WT, treated IL-5TG mice showed more severe lung fibrosis. Indeed at each time point examined, levels of lung hydroxyproline, soluble collagen, fibronectin, and TGF- were higher in blm-treated IL-5TG relative to IL-5WT mice (Fig. 2, A–D). Histological examination also showed more extensive blm-induced fibrotic lesions and collagen deposition (Masson Trichrome stain) in transgenic mice in comparison to wild type (data not shown). Thus increased IL-5 was associated with an enhanced fibrotic response, suggesting a role for this cytokine in pulmonary fibrosis.
|
Collectively, these results showed that blm-induced lung fibrosis was increased paradoxically in both IL-5TG and IL-5-/- mice in comparison to their respective blm-treated wild-type mice.
Blm lung response after adenovirus construct for IL-5 administration
To clarify the role of IL-5 and to support the observations in IL-5TG mice, we studied the effects of enhanced lung IL-5 expression (as a result of treatment with an AdIL-5) on blm-responses in IL-5WT as well as in low responder (to blm-induced pulmonary fibrosis) BALB/c mice. The effects of AdIL-5 treatment on blm-induced pulmonary fibrosis were evaluated by measuring lung content of hydroxyproline, soluble collagen, type I collagen, and fibronectin at day 21. The results showed that blm-induced fibrosis, as assessed using all criteria, was significantly increased in IL-5WT mice treated with AdIL-5 relative to those treated with AdCTL (Table III). In addition increasing the dose of adenovirus (108 PFU) caused a greater increase in the mortality rate of IL-5WT mice treated with AdIL-5 relative to AdCTL (50% vs 80% in AdIL-5 and AdCTL, respectively). Treatment of BALB/c mice with blm did not cause significant fibrosis (data not shown), but additional treatment with AdIL-5 caused significant fibrosis, although the level of fibrosis was still less than that seen in IL-5+/+ mice (Table III). No significant differences were noted between the saline-treated control group of IL-5WT and BALB/c mice receiving either construct (data not shown). Collectively, these data support our observations obtained in blm-treated IL-5TG mice and suggest a key role for IL-5 in the extension of blm-induced pulmonary fibrosis. However these results appear to contradict the findings using the IL-5-deficient mice, wherein reduced eosinophils due to IL-5 deficiency but enhanced lung T lymphocyte influx are associated with enhanced pulmonary fibrosis.
|
To clarify this apparent paradox, the potential roles of selective influx of eosinophils or T lymphocytes in blm-induced fibrosis in IL-5 transgenic and deficient animal models were examined in vitro using coculture studies. These pulmonary immune cells were isolated and purified from lung tissue at day 7 (eosinophils) or day 14 (T cells) after instillation of blm to donor animals. They were then cultivated in coculture with murine lung fibroblasts obtained from respective naive (not treated with blm) wild-type mice (IL-5WT and IL-5+/+). The capacity of these immune cells to regulate fibroblast function was then examined in terms of cell proliferation ([3H]thymidine incorporation), collagen production, and myofibroblast differentiation (expression of -SMA) in culture. Unfortunately, the cell purification techniques used in this study did not permit isolation of sufficient viable eosinophils from the parenchyma tissue of IL-5WT mice treated with saline or blm. In contrast, numerous eosinophils were obtained from saline and blm-treated IL-5TG mice. Thus, we compared only the effect of eosinophils from the transgenic mice purified at days 7 and 14 after blm or saline treatment. Eosinophils from both saline- and blm-treated mice and at either time point, did not significantly affect fibroblast proliferation in coculture (Fig. 3A). However, eosinophils purified from blm-treated IL-5TG at day 7 and to a lesser extent at day 14, significantly induced myofibroblast differentiation as estimated by measuring -SMA and increased type I collagen expression (Fig. 3, B and C). Although eosinophils from saline-treated IL-5TG weakly stimulated both parameters as compared with the medium alone, this effect was not statistically significant. PDGF and TGF- were used as positive controls for fibroblast proliferation and -SMA (as well as type I collagen synthesis), respectively. These data showed that in vitro purified blm-IL-5TG eosinophils could directly activate certain fibroblast functions.
|
-SMA expression, Fig. 3E) or collagen production (Fig. 3F). The lung fibroblasts however were responsive to TGF-, which was selected as the positive control for induction of myofibroblast differentiation. Thus eosinophils have the potential to directly affect fibrosis by stimulating myofibroblast differentiation, whereas T cells do not have this role. Effects of anti-CD3 treatment on acute lung injury and pulmonary fibrosis
Because enhanced T lymphocyte influx in the lung in the absence of eosinophilia is associated with enhanced blm-induced pulmonary fibrosis in IL-5-deficient mice, the role of eosinophils and T cells in vivo was investigated. To further characterize the biological activity of eosinophils and T lymphocytes in blm-induced lung fibrosis, we investigated the effects on blm-induced lung fibrosis of 1) instillation of purified eosinophils from blm-treated IL-5TG mice into the lungs of wild-type mice and 2) T cell depletion by anti-CD3 Ab treatment in IL-5+/+ and IL-5-/- mice. Wild-type mice were administrated at day 0 with saline or blm at low dose (0.01 U/mouse) to induce submaximal levels of lung injury and fibrosis. At days 7 and 14, these mice received by oral instillation 5 x 106 purified lung eosinophils from blm-treated IL-5 transgenic mice. Eosinophil injection into saline-treated recipient mice slightly increased the level of lung hydroxyproline at day 21 (Fig. 4). This effect of eosinophil injection was much more marked and significant in blm-treated recipient mice. Thus lung hydroxyproline was significantly higher in blm-treated mice receiving eosinophils than in mice receiving blm only (Fig. 4).
|
Cytokine profile of purified eosinophils and T cells
To determine the mechanism by which eosinophils and T cells participate to the extension of fibrosis, we then explored their expression of cytokines implicated in the pathogenesis of fibrosis. GEArrays gene arrays allowed us to profile the expression of 20 preselected genes known to activate fibroblast functions and to be produced specifically by both these immune cell types. Transcript levels of type 2 cytokines such as IL-4, IL-13, and IL-10 were significantly increased in eosinophils purified from IL-5TG mice administrated with blm in comparison to eosinophils from saline-treated IL-5TG animals. Similarly higher expression of TGF-1 and -3 was found in eosinophils from blm- vs saline-treated mice (data not shown). Analysis of T cells using a similar array revealed that blm-treated IL-5-/- and IL-5+/+ mice significantly expressed TNF-, IL-1, and IFN-, although expression was higher in deficient cells. Similarly, TGF-1 was more highly induced in T cells from IL-5-/- vs IL-5+/+ mice (data not shown).
On the basis of these array analyses, we further focused our work on selected cytokines. Isolated and purified eosinophils and T lymphocytes were analyzed for TNF-, IL-1, IFN-, IL-4, IL-13, and TGF- production in the presence of calcium ionophore (eosinophils) or anti-CD3 Abs (T cells). Consistent with the array data, purified and activated eosinophils from saline and blm-treated IL-5TG mice expressed IL-4, IL-13, and TGF- (Table IV). Eosinophils from blm-treated transgenic mice however produced significantly higher levels of these cytokines than those from saline-treated transgenic mice. Very low or undetectable levels of TNF-, IL-1, and IFN- were expressed by all eosinophil cultures, and no difference was noted between cells from blm- vs saline-treated mice (Table IV). Without in vitro stimulation, eosinophils from both blm- and saline-treated mice did not produce detectable levels of cytokines albeit freshly purified cells expressed the transcripts. This observation may reflect a lack of sensitivity of our ELISA or denote the relative incapacity of eosinophils to produce spontaneously cytokines in vitro without strong stimulating factors such as calcium ionophore.
|
, IL-1, and IFN- expression were slightly decreased in cultures from blm-treated IL-5-/- mice relative to cultures from blm-treated IL-5+/+ mice (Table IV), whereas TGF- levels were significantly increased. Anti-CD3-induced expression of IL-4 and IL-13 were dramatically diminished in IL-5-/- T cells in comparison to IL-5+/+ T cells.
In summary, these data showed that activated saline and blm transgenic eosinophils produced preferentially type 2 cytokines (IL-4 and IL-13) and TGF-, and the expression level was higher in cells from blm-treated animals. T cells from blm-treated IL-5-/- and IL-5+/+ mice expressed both proinflammatory/type 1 and type 2 cytokines as well as TGF-. However T cells from IL-5-deficient mice expressed significantly less type 2 cytokines relative to T cells from wild-type mice.
To ensure that the cytokine profile observed in vitro was also apparent in vivo, we measured the levels of the studied cytokines in whole lung homogenates. TNF- or IFN- proteins were not detectable in all samples examined. However IL-4 (data not shown) and IL-13 (Fig. 5A) proteins were markedly induced in lungs of blm-treated IL-5TG mice in comparison to similarly treated IL-5WT mice. This increase was maximal in IL-5TG at day 7 after blm treatment. In contrast, IL-1 levels were significantly lower at days 14 and 21 post-blm treatment in IL-5TG relative to IL-5WT animals (Fig. 5B). The opposite effects on cytokine levels were noted in IL-5-/- mice in comparison to IL-5+/+ mice. Thus IL-13 and IL-4 (data not shown) levels were decreased in deficient animals 7 days after blm treatment, whereas IL-1 lung contents were higher in IL-5-/- vs IL5+/+ mice at days 14 and 21 (Fig. 5, C and D). These data indicate that the more severe fibrotic lesions observed in IL-5TG and IL-5-/- relative to their respective wild-type controls, were associated with different patterns of cytokine expression. Transgenic mice produced preferentially type 2 cytokines such as IL-4 and IL-13 in response to blm, whereas deficient animals developed pulmonary responses characterized by a pronounced presence of proinflammatory/type 1 cytokines. However, the lung response to blm in both strains was associated with an increased level of TGF- expression (Fig. 2, D and H).
|
To determine whether cytokine production by purified eosinophils and T cells can be related to their known biological activity in coculture with fibroblasts (Fig. 3), we assessed the potential direct activity of these two groups of cytokines on pulmonary fibroblasts. We compared the in vitro activity of various concentrations of recombinant mouse IL-4, IL-13, and TGF- on lung fibroblasts isolated from IL-5WT mice, and the effects of TNF-, IL-1, and IFN- on lung fibroblasts isolated from IL-5+/+ mice. Recombinant IL-4, IL-13, and TGF- at different concentrations (1 to 20 ng/ml) stimulated fibroblast proliferation, which was significant at the 20 ng/ml dose for all three cytokines (Fig. 6A). In addition, -SMA and type I collagen expression were also increased after addition of these three cytokines compared with medium alone, with the greatest response noted with TGF- (Fig. 6, B and C). Addition of recombinant TNF-, IL-1, and IFN- also caused increased proliferation in IL-5+/+ fibroblasts (Fig. 6D). In contrast, no major differences in -SMA or type I collagen expression were observed after addition of these proinflammatory type 1 cytokines compared with medium alone (Fig. 6, E and F), whereas these cells did respond to TGF- used as a positive control. Recombinant mouse IL-5 was also tested in culture of fibroblasts obtained from all murine strains. At all concentrations used (1 to 20 ng/ml), IL-5 was unable to significantly affect proliferation rate, -SMA expression and collagen synthesis (data not shown). These results demonstrated that both types of cytokines, type 2 vs proinflammatory type 1, expressed in IL-5 transgenic and deficient mice, respectively, mimicked relatively well the capacity of purified eosinophils and T cells to activate certain functions of fibroblasts in culture.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first significant finding from this study was that the intensity of the pulmonary response to blm depended, at least in part, on an IL-5-dependent eosinophil infiltrate in animals that are IL-5 sufficient. Lung accumulation of eosinophils accentuated not only alveolitis but also the fibrosis induced by administration of blm in these mice. In mouse models of IL-5 overexpression or IL-5 deficiency, we demonstrated that eosinophils contributed to the early blm-induced lung injury and mortality (due to high-dose blm). Although the absence of IL-5 was associated with a relative limitation of early lung injury and a relative protection from lethal dose of blm, exaggerated lung IL-5 expression was accompanied by the opposite effects, namely more severe blm-induced lung injury and mortality. The amplitude of early lesions in our models were specifically linked with the intensity of pulmonary eosinophilia, emphasizing the important role of activated eosinophils in blm-induced acute injury. We can further postulate that these deleterious effects are due to the activated eosinophil ability to release a spectrum of cytotoxic mediators such as eosinophil cationic proteins and oxygen radicals (24), and as demonstrated in other models of lung injury (25, 26).
By studying more chronic effects of blm, we also found that recruited and activated eosinophils can specifically and directly interact with mesenchymal cells by stimulating myofibroblast differentiation and collagen synthesis and thus amplify lung fibrosis. This was observed from in vitro studies where eosinophils were cultivated with lung fibroblasts. Our in vitro observations using lung cells are consistent with the data obtained with eosinophils purified from blood and cultivated with lung or dermal fibroblasts (27, 28, 29), except pulmonary eosinophils from blm-treated IL-5 transgenic mice were unable to affect fibroblast proliferation in contrast to a previous study (28). This notable difference emphasizes the need to consider the functional heterogeneity of eosinophils from different anatomical compartments, or that eosinophils may adopt different phenotypes depending on anatomic localization and/or the pathology of the particular lesion causing the eosinophilia. The in vivo importance of eosinophils was confirmed specifically by the ability of donor eosinophils instilled into recipient mice to enhance the chronic response to blm-induced pulmonary fibrosis. This was additionally confirmed by the ability of AdIL-5 transfection to enhance blm-induced fibrosis in both responder (IL-5+/+) as well as low-responder (BALB/c) mice. By increasing IL-5 and thus eosinophil recruitment in BALB/c mice, we were able to significantly induce pulmonary fibrosis in this strain known to develop minimal (if any) eosinophilia and lung fibrosis in response to blm. Beside its capacity to elaborate toxic and injurious products, the eosinophil has the potential of being an immunomodulatory cell with a remarkable capacity to produce a whole range of potent mediators such as pro-inflammatory (TNF-), Th1 (IL-12), and Th2 (IL-4 and IL-13) cytokines, and thus may dramatically and directly change the phenotype of the immune response (24). In our study, lung eosinophils were identified as a major source of TGF-, IL-4 and IL-13 but not proinflammatory cytokines, consistent with a preferentially Th2-like phenotype in the course of blm-induced inflammation and fibrosis. It is noteworthy that these cytokines are strongly involved in tissue remodeling and fibrosis (30). Thus, we can further postulate that their profibrotic functions are mediated by type 2 cytokines. In addition, these observations well support data obtained by immunohistochemistry and published by our group showing that eosinophils are key cellular sources of the profibrotic mediators TGF- and MCP-1 in the blm model (31, 32, 33). Finally because the lung eosinophil is also an important source of the chemokines RANTES and IL-16 (data not shown), it is probable that it also participates in the initiation of pulmonary inflammation by promoting the recruitment and the activation of macrophages and lymphocytes as demonstrated in another lung injury model (34). This evidence for a role for eosinophils in IL-5-sufficient mice is consistent with a previous study showing suppression of blm-induced fibrosis by eosinophil depletion using anti-IL-5 Abs (7), but appears to contradict another study showing failure of such Abs to suppress blm-induced fibrosis despite suppression of eosinophilia (8). The basis for this contradiction is unclear, but may be related to differences in dosage of blm used, timing of analysis, method for quantitating lung eosinophilia, and the use of the SCID mouse in the latter study.
A second chief finding was that blm-induced pulmonary fibrosis is also intimately dependent on T lymphocytes and can proceed in the absence of IL-5 and lung eosinophil recruitment. The importance of T cells in the development of lung fibrosis was demonstrated by administration of anti-CD3 Abs, which highly reduced the amplitude of fibrotic lesions in both blm-treated wild-type and IL-5-deficient mice. Our observations are in accordance with previous reports showing that blocking the recruitment of CD4+, CD8+, or CD3+ cells limit the extension of fibrotic lesions in different lung experimental models (5, 35, 36). On the basis of our coculture experiments, we can further propose that profibrotic activity of T lymphocytes may be due, at least in part, to their ability to directly stimulate the proliferation of lung fibroblasts. Because T cells can produce growth factors for fibroblasts such as PDGF, basic fibroblast factor, (37) and as observed in this study, TNF-, it is probable that the effect observed with T cells on fibroblast proliferation is mediated by such growth factors. However, purified T cells were unable to stimulate other aspects of fibroblast activation, namely, myofibroblast differentiation and collagen synthesis, at least under in vitro conditions as described in this study. Thus, T cells appear to play a limited but specific direct role in terms of promoting fibrosis in this model, although indirect roles certainly could play additional and potentially more important roles. Furthermore, a more intense lung T cell recruitment in IL-5-deficient mice could compensate for the loss of lung eosinophil recruitment in the lung fibrotic response to blm. Thus although IL-5-dependent lung eosinophilia plays an important role in blm-induced lung fibrosis, it is not indispensable, especially in transgenically IL-5-deficient mice where compensatory mechanisms have time to develop. This conclusion is consistent with a previous study showing no protection from fibrosis in IL-5-deficient mice (8). Interestingly such compensatory mechanisms are inadequate to fully compensate for the lack of IL-5/eosinophils in the acute injury response to blm. It is noteworthy that induction of lung TGF- expression remained intact in these IL-5-deficient mice. Thus fibrosis in both IL-5TG and IL-5-deficient mice is associated with induction of TGF-, which could account for the observed fibrosis despite differences in Th1 vs Th2 profiles.
Evidence has emerged that lung fibrosis is the result of disordered fibroblast regulation. In fibrotic lesions, numerous histological studies have shown the simultaneous presence of scattered foci of fibroblast proliferation as well as the emergence of myofibroblast phenotype (38). Bearing this in mind, we can postulate that the appearance of the myofibroblast phenotype during fibrosis may be dependent at least in part on recruited eosinophils, whereas fibroblast proliferation could be induced at least in part by recruited T cells. However the potential role of additional cell types, such as mast cells, cannot be excluded from the current study.
Recent observations in humans and animals supported the view that lung fibrosis is a type 2-related disease (30). Nevertheless the type 1 component of the lung response to injury also appears to be important because IFN--deficient mice exhibit attenuated blm-induced pulmonary fibrosis (39). The data presented in this study indicate a possible concomitant presence of both types of immune responses in lung injury and fibrosis. On the basis of the cytokine profile expressed in vitro and in vivo, it is difficult to discriminate between preferential type 1 or type 2 response in blm-induced lung disease because we detected in wild-type mice the expression of both types 1 and 2 cytokines, namely IFN-, TNF-, IL-1, IL-4, and IL-13. In addition, our observations on transgenic and deficient animals support the notion that either exaggerated type 2 or proinflammatory/type 1 response can promote lung fibrosis, as observed respectively in blm-treated IL-5TG and IL-5-/- mice. This clearly argues for a more complicated mechanism, wherein both types of responses can promote fibrosis either individually or in concert. If this could be confirmed in human lung fibrosis, development of new therapeutic approaches based on controlling the activity of each type of immune responses should take this possibility into consideration. This may imply that selective suppression of either one of these responses is unlikely to adequately control disease progression.
In summary, we describe in this study that recruited eosinophils and T cells both can contribute to the genesis of lung fibrosis induced by blm, partly via their ability to directly affect certain fibroblast functions mediated by elaboration of both types 1 and 2 cytokines. These findings may have similar implications for other models or diseases where tissue remodeling/fibrosis is characterized by eosinophil and T cell infiltration.
|
Acknowledgments |
|---|
assay. We also thank Lisa Riggs for her excellent technical assistance. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Sem H. Phan, Department of Pathology, University of Michigan Medical School, Ann Arbor, MI 48109-0602. E-mail address: shphan{at}umich.edu
3 Abbreviations used in this paper: PDGF, platelet-derived growth factor; blm, bleomycin; BAL, bronchoalveolar lavage; BALF, BAL fluid; MCP, monocyte chemoattractant protein; Ad, adenovirus construct; TG, transgenic; WT, wild-type; SMA, smooth muscle actin.
Received for publication April 24, 2003. Accepted for publication September 4, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
to rat lung induces severe pulmonary inflammation and patchy interstitial fibrogenesis with induction of transforming growth factor-1 and myofibroblasts. Am. J. Pathol. 153:825. induces acute lung injury and chronic repair leading to pulmonary fibrosis. J. Clin. Invest. 107:1529.[Medline]
1 induces prolonged severe fibrosis in rat lung. J. Clin. Invest. 100:768.[Medline]
using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal. Biochem. 216:276.[Medline]
in the stimulation of lung fibroblast IL-6-type cytokine production. J. Immunol. 156:4449.[Abstract]
(TGF-). Proc. Natl. Acad. Sci. USA 96:9660.-mediated lung cytokine networking and eosinophil recruitment in pulmonary fibrosis. J. Immunol. 158:954.[Abstract]
expression in bleomycin-induced pulmonary fibrosis. Am. J. Pathol. 147:352.[Abstract]
-deficient mice after intratracheal bleomycin. Am. J. Respir. Cell Mol. Biol. 24:545.This article has been cited by other articles:
|
|
 |
|
  P. L. Simonian, C. L. Roark, F. Wehrmann, A. M. Lanham, W. K. Born, R. L. O'Brien, and A. P. Fontenot IL-17A-Expressing T Cells Are Essential for Bacterial Clearance in a Murine Model of Hypersensitivity Pneumonitis J. Immunol., May 15, 2009; 182(10): 6540 - 6549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. L. Simonian, C. L. Roark, F. Wehrmann, A. K. Lanham, F. Diaz del Valle, W. K. Born, R. L. O'Brien, and A. P. Fontenot Th17-Polarized Immune Response in a Murine Model of Hypersensitivity Pneumonitis and Lung Fibrosis J. Immunol., January 1, 2009; 182(1): 657 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Morschl, J. G. Molina, J. B. Volmer, A. Mohsenin, R. S. Pero, J.-S. Hong, F. Kheradmand, J. J. Lee, and M. R. Blackburn A3 Adenosine Receptor Signaling Influences Pulmonary Inflammation and Fibrosis Am. J. Respir. Cell Mol. Biol., December 1, 2008; 39(6): 697 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zhang, J. Qian, X. Xing, F.-M. Kong, L. Zhao, M. Chen, and T. S. Lawrence Inhibition of the Tumor Necrosis Factor-{alpha} Pathway Is Radioprotective for the Lung Clin. Cancer Res., March 15, 2008; 14(6): 1868 - 1876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. G. Luzina, N. W. Todd, A. T. Iacono, and S. P. Atamas Roles of T lymphocytes in pulmonary fibrosis J. Leukoc. Biol., February 1, 2008; 83(2): 237 - 244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. Wallach-Dayan, R. Golan-Gerstl, and R. Breuer Evasion of myofibroblasts from immune surveillance: A mechanism for tissue fibrosis PNAS, December 18, 2007; 104(51): 20460 - 20465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Pottier, C. Chupin, V. Defamie, B. Cardinaud, R. Sutherland, G. Rios, F. Gauthier, P. J. Wolters, Y. Berthiaume, P. Barbry, et al. Relationships between Early Inflammatory Response to Bleomycin and Sensitivity to Lung Fibrosis: A Role for Dipeptidyl-Peptidase I and Tissue Inhibitor of Metalloproteinase-3? Am. J. Respir. Crit. Care Med., December 1, 2007; 176(11): 1098 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. van den Brule, J. Heymans, X. Havaux, J.-C. Renauld, D. Lison, F. Huaux, and O. Denis Profibrotic Effect of IL-9 Overexpression in a Model of Airway Remodeling Am. J. Respir. Cell Mol. Biol., August 1, 2007; 37(2): 202 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Pochetuhen, I. G. Luzina, V. Lockatell, J. Choi, N. W. Todd, and S. P. Atamas Complex Regulation of Pulmonary Inflammation and Fibrosis by CCL18 Am. J. Pathol., August 1, 2007; 171(2): 428 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. K. Dhillon, D. Pinson, S. Dhillon, O. Tawfik, M. Danley, M. Davis, O. Nemon, M. Mayo, A. Kumar, Y.-J. Tsai, et al. Bleomycin treatment causes enhancement of virus replication in the lungs of SHIV-infected macaques Am J Physiol Lung Cell Mol Physiol, May 1, 2007; 292(5): L1233 - L1240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Studer and N. Kaminski Towards Systems Biology of Human Pulmonary Fibrosis Proceedings of the ATS, January 1, 2007; 4(1): 85 - 91. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Misson, F. Brombacher, M. Delos, D. Lison, and F. Huaux Type 2 immune response associated with silicosis is not instrumental in the development of the disease Am J Physiol Lung Cell Mol Physiol, January 1, 2007; 292(1): L107 - L113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Xu, A. L. Mora, J. LaVoy, K. L. Brigham, and M. Rojas Increased bleomycin-induced lung injury in mice deficient in the transcription factor T-bet Am J Physiol Lung Cell Mol Physiol, October 1, 2006; 291(4): L658 - L667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Huaux, M. Gharaee-Kermani, T. Liu, V. Morel, B. McGarry, M. Ullenbruch, S. L. Kunkel, J. Wang, Z. Xing, and S. H. Phan Role of Eotaxin-1 (CCL11) and CC Chemokine Receptor 3 (CCR3) in Bleomycin-Induced Lung Injury and Fibrosis Am. J. Pathol., December 1, 2005; 167(6): 1485 - 1496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. B. Jones, R. P. Hart, and P. G. Popovich Molecular Control of Physiological and Pathological T-Cell Recruitment after Mouse Spinal Cord Injury J. Neurosci., July 13, 2005; 25(28): 6576 - 6583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Barbarin, Z. Xing, M. Delos, D. Lison, and F. Huaux Pulmonary overexpression of IL-10 augments lung fibrosis and Th2 responses induced by silica particles Am J Physiol Lung Cell Mol Physiol, May 1, 2005; 288(5): L841 - L848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Arras, J. Louahed, J.-F. Heilier, M. Delos, F. Brombacher, J.-C. Renauld, D. Lison, and F. Huaux IL-9 Protects against Bleomycin-Induced Lung Injury: Involvement of Prostaglandins Am. J. Pathol., January 1, 2005; 166(1): 107 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Cavarra, F. Carraro, S. Fineschi, A. Naldini, B. Bartalesi, A. Pucci, and G. Lungarella Early response to bleomycin is characterized by different cytokine and cytokine receptor profiles in lungs Am J Physiol Lung Cell Mol Physiol, December 1, 2004; 287(6): L1186 - L1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Barbarin, M. Arras, P. Misson, M. Delos, B. McGarry, S. H. Phan, D. Lison, and F. Huaux Characterization of the Effect of Interleukin-10 on Silica-Induced Lung Fibrosis in Mice Am. J. Respir. Cell Mol. Biol., July 1, 2004; 31(1): 78 - 85. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
