help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on November 7, 2007, doi:10.1165/rcmb.2007-0079OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2007-0079OCv1
38/4/414    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, J.-Q.
Right arrow Articles by Roth, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yang, J.-Q.
Right arrow Articles by Roth, M.
American Journal of Respiratory Cell and Molecular Biology. Vol. 38, pp. 414-422, 2008
© 2008 American Thoracic Society
DOI: 10.1165/rcmb.2007-0079OC

Cell Density and Serum Exposure Modify the Function of the Glucocorticoid Receptor C/EBP Complex

Jing-Qing Yang1, Jochen J. Rüdiger1, J. Margaret Hughes2, Stephanie Goulet1, Mesut M. Gencay-Cornelson1, Pieter Borger1, Michael Tamm1 and Michael Roth1,3

1 Pneumology, Pulmonary Cell Research, University Hospital Basel, Basel, Switzerland; 2 Faculty of Pharmacy, University of Sydney, Sydney, Australia; and 3 Molecular Medicine, The Woolcock Institute for Medical Research, Camperdown, Australia

Correspondence and requests for reprints should be addressed to Michael Roth, PhD, Pulmonary Cell Research, Dept. Research, University Hospital Basel, Hebelstrasse 20, CH-4031 Basel, Switzerland. E-mail: Rothmic{at}uhbs.ch


   Abstract
Top
 Abstract
CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The glucocorticoid receptor (GR) is a major control factor for proliferation, differentiation, and inflammation. Our knowledge about the GR is focused on its function as a transcription regulator. However, cells do not always respond to steroids in the same way or develop resistance. The mechanism underlying such a modified steroid response is not well understood, and may depend on the microenvironment of the cells or on the stage of their differentiation. Therefore, we studied the effect of cell density and inflammatory conditions on the expression, compartmentalization, activation, and the anti-proliferative function of the GR in primary human lung fibroblast cultures. In subconfluent cells the GR was located perinuclear, while in confluent cells it was ubiquitously expressed. Serum stimulation up-regulated the level of GR mRNA and protein under all conditions. In subconfluent cells dexamethasone activated the nuclear accumulation and DNA binding of the GR persistently, while in confluent cells its activity declined after 6 hours. In subconfluent cells, but not in confluent cells, the GR interacted with a 42-kD, but not the 30-kD C/EBP-{alpha} isoprotein, which resulted in an up-regulation of p21(Waf1/Cip1) expression and suppression of proliferation. In confluent cells, glucocorticoids induced p27(Kip1) expression via p38 mitogen-activated protein kinase and a 52-kD C/EBP-β isoprotein. However, p27(Kip1) did not mediate the antiproliferative effect of glucocorticoids, but simultaneous inhibition of p21(Waf1/Cip1) and p27(Kip1) unlocked contact inhibition in confluent cells. Our results indicate that cell density and serum exposure alter the localization and function of the GR.

Key Words: glucocorticoid receptor • primary human lung fibroblasts • proliferation control • tissue remodeling



   CLINICAL RELEVANCE
Top
 Abstract
 CLINICAL RELEVANCE
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Cell differentiation and inflammation affects composition and function of the glucocorticoid receptor–C/EBP complex. This implies that monitoring of long-term steroid therapy can reduce side effects and the development of steroid resistance.

 
Glucocorticoids are expressed by all known cell types and control cell differentiation, organ development, and function. Glucocorticoids regulate immune responses, fat metabolism, renal function, and vascular leakage in most organisms (1, 2). Glucocorticoids bind to the inactive glucocorticoid receptor (GR) in the cytosol, followed by restructuring of the GR–multiprotein complex. Two active GR molecules form a dimer that couples to FKBP52 and dynein, which mediate the transported GR into the nucleus (3, 4). In the nucleus the GR-dimer binds to a DNA sequence, the glucocorticoid response element (GRE), and can either stimulate or silence genes (1, 5). Long-term steroid treatment significantly down-regulated GR expression, suggesting a self-limiting control mechanism, and may result in tachyphylaxis (6–8).

Beside its action as a transcription factor, the GR mediates its effect also by binding to and modifying the function of other transcription factors including C/EBP-{alpha}, -β, Stat1/5, NF-{kappa}B, or AP-1 (9–14). Specifically of interest for proliferation control are the complexes formed by the GR with the transcription factors C/EBP-{alpha} or -β, as they are essential to initiate the expression of p21(Waf1/Cip1) and thereby mediate the antiproliferative effect of steroids (10, 15, 16). Fibroblast proliferation is inhibited by glucocorticoids also via the activation of p21(Waf1/Cip1), which requires C/EBP-{alpha} activity (15, 17, 18). However, glucocorticoids do not always block proliferation in fibroblasts, and the reason for this is unknown (19–21). The antiproliferative efficacy of glucocorticoids may vary with the mitogenic stimulus (18), might be a species-specific effect (20), or may be affected either by the organ of origin of the fibroblasts or by the underlying disease (19, 21). We described earlier that a cell type–specific lack of C/EBP-{alpha} in bronchial smooth muscle cells of patients with asthma is responsible for the enhanced proliferative capacity of these cells and possibly leads to airway muscle cell hyperplasia in asthma (22–24). In addition to airway smooth muscle hyperplasia, the airway wall of patients with asthma and chronic obstructive pulmonary disease (COPD) patients is also characterized by increased fibroblast proliferation and extracellular matrix deposition in the lamina propria (2, 25). The effect of steroids on these pathologies is controversial and their effect may depend on drug dosage, length of therapy, and severity of the disease (2). We have recently demonstrated that the effect of steroids on extracellular matrix and collagen deposition is modified by the presence of serum in human lung fibroblasts. In resting cells or in the presence of TGF-β1, steroids down-regulated the deposition of extracellular matrix, while in the presence of serum (5%) steroids further increased extracellular matrix deposition (26).

In regard to cell proliferation control it is important to note that cell density and the presence of a mitogen modified the expression and function of p21(Waf1/Cip1) in fibroblasts (27, 28). Furthermore, translocation of nonactivated p21(Waf1/Cip1) into the nucleus of pancreatic fibroblasts was associated with differentiation into myofibroblasts, which expressed muscle cell actin and were more susceptible to apoptotic signals (29). In an animal model for adipocyte differentiation, cell density and serum affected the function of dexamethasone and the expression of C/EBP-{alpha} and -β (30). Taking into account that at least active C/EBP-{alpha} is required for the antiproliferative effect of glucocorticoids, we hypothesized that cell density, as well as mitogen, and cell differentiation change the function of the GR. Furthermore, glucocorticoids also activate a second cell cycle control protein, p27(Kip), but its role in the drug's effect is unclear and may be cell type or species specific (15, 31).

Therefore, we investigated the effect of cell density and the presence of serum on the activation and translocation of the GR in response to dexamethasone in cultures of primary human lung fibroblasts. In an earlier study (26), we used subconfluent fibroblasts as a model of injured tissue undergoing wound repair, and confluent cell layers to model intact tissue. High serum concentration (10%) was used to mimic vascular leakage as it occurs during early stages of inflammation, while low serum (0.1%) represented the situation in nondamaged tissue. We further investigated the role of the GR and C/EBP-{alpha} and -β isoforms on expression of two major antiproliferative proteins—p21(Waf1/Cip1) and p27(Kip)—and their activation by dexamethasone.


   MATERIALS AND METHODS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Chemicals
All chemicals were from Calbiochem (La Jolla, CA) if not otherwise noted. Complete protease inhibitor was from Roche Diagnostics (Basel, Switzerland), and gradient poly-acryl amid gels (PAGE) from Bio-Rad (Hercules, CA). Minimal essential medium (MEM), RPMI 1640, and phosphate-buffered saline (PBS w/o Ca2+ and Mg2+) were from Cambrex Bio Science Verviers (Verviers, Belgium). PVDF membranes were from Millipore (Bedford, MA), Ponceau from Sigma (Buchs, Switzerland), enhanced chemiluminescence (ECL) from Pierce (Rockford, IL) and X-ray films from Kodak (Eastman Kodak, Rochester, NY). Antibodies and small inhibitory (si)RNAs were from Santa Cruz Biotechnology (Santa Cruz, CA). Cy3-labeled goat anti-rabbit immunoglobulin G (IgG) was from Jackson ImmunoResearch Laboratories (West Grove, PA). GRE oligonucleotides were from Affinity Bioreagents, Inc. (Golden, CO), and all other DNA oligonucleotides were from MWG-Biotech GmbH (Ebersberg, Germany).

Fibroblast Culture
Primary fibroblasts were established from lung tissue samples after written consent and approval by the ethical committee (University Hospital Basel). Fibroblasts were grown in RPMI 1640 supplemented with 10% fetal calf serum (FCS) and 1% MEM vitamins. All experiments were performed between cell passages two and six. Fibroblasts were seeded (1 x 104 cells/cm2) and either used directly (subconfluent), or grown to 100% confluence for 2 days. Before the experiments, all cells were serum starved in 0.1% FCS for 24 hours. For siRNA was used at a final concentration of 1 to 10 nMol for 24 hours before experiments. No specific transfection medium was used because fibroblasts readily absorbed siRNA within 24 hours, as confirmed by immunoblotting.

Proliferation
Manual cell counts in a Neubaur hemocytometer were used to determine the proliferative effect of serum and its inhibition by drugs or siRNAs on subconfluent fibroblast cultures. Fibroblasts were seeded at a density of 1 x 104 cells/ml and were grown for 3 days in the various conditions as described in RESULTS (12, 22).

Protein Extraction, Electrophoresis, Co-Immunoprecipitation, and Immunoblotting
Cytosolic and nuclear protein fractions were isolated as described earlier, and total protein concentration was determined by Bradford's method (12, 17). Protein (10 µg) was dissolved in Laemmli buffer, denatured (95°C, 5 min), chilled on ice (5 min), centrifuged (13,000 x g, 50 s), and applied to electrophoresis (4–15% SDS-PAGE). Proteins were transferred onto a PVDF membrane by semi-dry electro-transfer, which was confirmed by Ponceau's staining. Membranes were washed three times with PBS, blocked with 5% skimmed milk in PBS (4°C, overnight), and incubated with one of the primary antibodies (anti-GR antibody: 0.2 ng/ml, sc-1003; p21(waf1/Cip1): 0.2 ng/ml, sc-817; C/EBP-{alpha}: 0.2 ng/ml, sc-61; p27(Kip): 0.5 ng/ml, sc-1027). Unbound antibodies were washed off before membranes were incubated with horseradish-labeled species-specific secondary antibodies for 1 hour at room temperature. After washing (3 x 15 min) the membranes with blocking buffer, bound antibody signals were detected by ECL substrate and documented on X-ray film (12, 17). Co-immunoprecipitation for GR–C/EBP complexes was performed with the same antibodies used for immunoblotting and followed the protocol described earlier (12).

Electrophoretic Mobility Shift Assay
Electrophoretic mobility shift assay (EMSA) was performed using a [32P]-labeled GRE oligonucleotide (sc-2545) as described earlier (22, 26). Specificity of the GRE–protein complex was characterized by pre-incubating protein extracts with 50-fold excess of unlabeled GRE.

Immunochemistry and Confocal Microscopy
Fibroblasts were grown on coverslips (VWR International AG; Dietikon, Switzerland), treated, and then fixed in methanol/acetic acid (3:1 vol:vol, 15 min, room temperature). Unspecific signals were blocked in PBS, 5% donkey serum, and 0.3% Triton-X-100 by overnight incubation at room temperature. For immunofluorescence, slides were incubated with a primary rabbit anti-GR antibody followed by incubation with a Cy3-labeled goat anti-rabbit IgG antibody (blocking buffer, 1 h, 4°C), followed by two washes with PBS and one with distilled water (5 min), then embedded in Fluorsave. Identical conditions and antibody concentration were used for control slides stained with a primary isotype IgG rabbit antibody. Images were taken by a Zeiss LSM510 confocal microscope (Carl Zeiss AG, Jena, Germany) and analyzed using ImageJ v. 1.33.

RT-PCR
RNeasy Mini kit (Qiagen, Basel, Switzerland) was used to extract total RNA (1 µg), which was transcribed into cDNA using murine leukemia virus reverse transcriptase (37°C, 60 min) (Promega, Madison, WI). PCR was performed for GR: forward 5'-CACCCTCACTGGCTGTCGCTTCTC-3', reverse 5'-TGACAAACGAAAGAGGAG ACCGCC-3', 23 cycles; denaturation (98°C, 20 s), annealing (58°C, 22 s), and extension (72°C, 30 s); p21 (primer set; R&D Systems Inc, Minneapolis, MN), 30 cycles; denaturation (94°C, 45 s), annealing (55°C, 45 s), and extension (72°C, 45 s); β2-microglobulin (β2-M): forward 5'-CTCGCGCTACTCTCTCTCTTTCT-3', reverse 5'-TTAAGTGGGATCGAGACATGTAAGC-3', 23 cycles; denaturation (98°C, 30 s), annealing (60°C, 60 s), and extension (72°C, 60 s). PCR products were size fractionated by electrophoresis in a 1% agarose gel and products were visualized by ethidium bromide (12, 17).

Active and Inactive GR by Enzyme-Linked Immunosorbent Assay
GR to GRE binding was assessed using an enzyme-linked immunosorbent assay (ELISA) kit (Orgenium Laboratories, Helsinki, Finland) distinguishing total from active GR as described by the distributor. GR activation was determined in the same protein extracts used for EMSA and immunoblotting.

Cell Cycle Analysis
The DNA content of the cells was analyzed by flow cytometry as described previously (33). Fibroblasts were seeded at 1 x 104 cells/ml and grown for 24 and 48 hours in various conditions. Cells were then harvested by trypsin treatment, and 105 cells per condition were permeabilized and stained using 0.5% wt/vol saponin and 0.1% wt/vol serum albumin in PBS containing propidium iodide (50 µg/ml) and RNase A (50 µg/ml) for 20 minutes at 4°C. Cells were acquired on a FACS Calibur Sort and their DNA profiles analyzed using Modfit software (Becton Dickinson, Franklin Lakes, NJ). (32).

Statistics
Results of protein activation kinetics were compared by paired ANOVA. The effect of signal protein inhibitors and siRNAs were compared by paired two-tailed Student's t test. The null hypothesis was equality of response between groups. Results were considered to be significantly different when P < 0.05.


   RESULTS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Cell Density and Serum Modifies the Cell Compartmental Distribution of the GR
Cell compartmental distribution of the GR was determined by three independent methods: confocal microscopy, immunoblotting, and GRE-ELISA. Serum (10%) did not significantly increase the cell compartment–specific expression of the GR in subconfluent fibroblasts within 24 hours, and the GR was located in the cytosol in a perinuclear location (Figure 1A). In contrast, in confluent serum-starved cells the GR could be detected in the cytosol and the nucleus, and was homogenously distributed and was significantly up-regulated by serum (10%) in both cell compartments as early as 3 hours (Figure 1A). This effect of cell density on the compartmental distribution of the GR was confirmed by immunoblotting of cytosolic and nuclear protein extracts as depicted in Figure 1A. The basal level of total GR protein, as determined by ELISA, was increased in confluent cells, which expressed 4.2 ± 1.6 times more GR compared with subconfluent cells when grown in 0.1% serum.


Figure 1
Figure 1
View larger version (64K):
[in this window]
[in a new window]

  		Figure 1. (A) The left three representative confocal images show the compartment localization of the glucocorticoid receptor (GR) (green) in the presence and absence of fetal calf serum (FCS) in subconfluent fibroblasts. The nucleus was counterstained with Hoechst dye (blue). The three confocal images on the right are representative of GR distribution in confluent fibroblasts. The densitometric analysis of immunoblots of GR in cytosolic and nuclear protein extracts are summarized as kinetics charts. Representative immunoblots for the compartmental location of GR are depicted below the kinetic graphs. (B) The effect of serum stimulation on GR mRNA expression. All data points represent the mean ± SEM obtained in six different primary fibroblast cell lines. Open symbols, 0.1% FCS; solid symbols, 10% FCS; diamonds, subconfluent; squares, confluent.

 
The serum-stimulated increase of GR protein expression in confluent fibroblasts was preceded by de novo mRNA synthesis (Figure 1B). In confluent cells serum (10%) significantly increased the GR mRNA signal after 0.5 hours, peaking at 3 hours and declining thereafter, while in subconfluent cells the increase in GR mRNA was only significantly elevated at 1 hour (Figure 1B). GR mRNA signals were suppressed by actinomycin D (10 µM), but not cycloheximide (10 µM), confirming that mRNA de novo synthesis rather than accumulation was involved (data not shown).

Activation of the GR by Steroids Is Prolonged in Subconfluent Cells
When subconfluent fibroblasts were treated with dexamethasone (10–8–10–6 M), the GR was concentration-dependently activated and translocated completely into the nucleus within 3 hours (immunoblotting analysis). The level of the nuclear GR increased significantly within 0.5 hours compared with start levels (P < 0.001) and remained at a high level over 24 hours, with no significant difference in the kinetics comparing serum-starved and serum (10%)-stimulated cells (Figure 2A). In confluent fibroblasts the translocation of the GR into the nucleus was slower compared with subconfluent cells, peaked at 1 hour, and declined thereafter to basal levels within 24 hours (Figure 2A). Similar results were obtained using confocal microscopy and immunoblotting. The functional activation of the nuclear GR was confirmed by GRE-specific EMSA and ELISA; both methods showed that the increase of the GR in the nucleus was paralleled by binding of the GR to a synthetic GRE oligonucleotide and confirmed the decline of GR activity in the presence of serum within 24 hours (Figure 2B). A summary of the densitometric analysis of the three independent EMSAs is depicted in Figure 2B.


Figure 2
Figure 2
View larger version (65K):
[in this window]
[in a new window]

  		Figure 2. (A) The effect of dexamethasone on the cell compartmental localization of the GR in subconfluent and confluent fibroblasts and its modification by serum (FCS). Data points represent the mean ± SEM obtained in six different primary fibroblast cell lines. Open symbols, cytosol; solid symbols, nucleus; squares, 0.1% FCS; triangles, 10% FCS. (B) A representative EMSA of the GR/GRE complex and the optical density analysis of three independent EMSA (mean ± SEM) obtained in confluent cells from three different fibroblast lines. Open symbols, 0.1% FCS; solid symbols, 10% FCS; circles, without dexamethasone; squares, + 10–8 M dexamethasone. *P < 0.01 for GR level based on Student's t test (paired, two tailed) compared with start level.

 
Cell Density Affects Serum-Dependent C/EBP Expression and GR Complex Formation
Lung fibroblasts expressed two of the four known C/EBP-{alpha} isoforms, a 42- to 45-kD and a 30-kD protein (Figure 3A). In serum-deprived cells, both C/EBP-{alpha} isoforms were located in the cytosol and independent of cell density. Interestingly, in subconfluent cells serum induced a fast increase of the 30-kD C/EBP-{alpha} isoform within 3 hours, while the larger isoform was not significantly upregulated before 18 hours (Figure 3A, upper panel). In contrast, in confluent fibroblasts C/EBP-{alpha} was expressed at a much lower level and serum induced a significant increase of the large kD C/EBP-{alpha} isoform at 6 hours, which then declined to nearly basal levels within 24 hours (Figure 3A, lower panel). The 30-kD isoform was up-regulated at 18 hours and declined also at 24 hours (Figure 3A, lower panel). In serum-stimulated subconfluent fibroblasts, dexamethasone (10–8 M) shifted the large C/EBP-{alpha} into the nucleus and the total amount of C/EBP-{alpha} was lower at 24 hours compared with the start level (Figure 3B). The cell compartmental distribution of the 30-kD isoform was not shifted by dexamethasone, but its expression was reduced in both compartments (Figure 3B).


Figure 3
View larger version (57K):
[in this window]
[in a new window]

  		Figure 3. (A) The effect of cell density on the kinetics of the expression pattern of two C/EBP-{alpha} isoform (30, 42 kD) in cytosolic protein extracts of human primary lung fibroblasts, depicted as a representative immunoblot. Similar results were obtained in five different fibroblast lines. (B) The dexamethasone-induced activation/translocation of the two C/EBP-{alpha} isoforms from the cytosol into the nucleus in a representative immunoblot. The ratio of the nuclear/cytosol location of the 42-kD C/EBP-{alpha} was determined by immunoblotting and the data are shown as mean ± SEM obtained in five different primary fibroblast cell lines. Open squares, 0.1% FCS; open triangles, 10% FCS; solid squares, 0.1 + dexamethasone; solid triangles, 10 + dexamethasone. * indicates a significant increase over base line with P < 0.001. (C) A representative immunoblot of the kinetics of C/EBP-β isoform expression in cytosolic protein extracts of confluent fibroblasts. Similar results were obtained in three other confluent and four subconfluent fibroblast lines. (D) Representative immunoblots of C/EBP-{alpha} and -β isoforms after GR was targeted and immnuoprecipitated with an anti-GR antibody from nuclear protein extracts 12 hours after the addition of the drug. Similar results were obtained in three other cell lines.

 
In addition, we characterized the expression pattern of C/EBP-β, which was expressed as three isoproteins at 45, 40, and 30 kD, with a major band at 45 kD only in the cytoplasma of subconfluent and confluent fibroblasts (Figure 3C). The expression level of all three C/EBP-β isoforms increased when stimulated with serum and peaked at 6 hours, slowly declining thereafter. The expression of all three C/EBP-β isoforms was higher in confluent cells (n = 4) compared with subconfluent fibroblasts (n = 4), but neither their expression nor their cell compartmental distribution was affected by dexamethasone (data not shown).

Using an anti-GR antibody to capture the GR–C/EBP complex, followed by immunoblot analysis of the C/EBP-isoforms in the precipitates, indicated that the complex composition changed over time. At 6 hours a low amount of the 42-kD C/EBP-{alpha} and more of the 40/45 C/EBP-β proteins were co-precipitated with the GR (Figure 3D). At 24 hours the level of the 42-kD C/EBP-{alpha} was increased, while that of the two C/EBP-β isoforms was reduced in the co-precipitates of the GR (Figure 3D).

Cell Density and Serum Modifies the Expression of p21(Waf1/Cip1)
The kinetics of the GR and C/EBP-{alpha} activation overlapped for several hours and should therefore up-regulate the expression of p21(Waf1/Cip1); therefore, we determined its expression in the two cell compartments and calculated its nuclear to cytosolic ratio. In subconfluent serum-starved fibroblasts most of the p21(Waf1/Cip1) was consistently expressed in the nucleus over the observation period, and serum (10%) significantly increased its nuclear accumulation within 3 hours (P < 0.05), declining afterward (Figure 4A). In serum-starved confluent cells p21(Waf1/Cip1) levels were unchanged over 24 hours, while serum (10%) increased its accumulation in the nucleus significantly within 3 hours (P < 0.01) and maintained it at a high level until 24 hours (Figure 4A). The observed increase of p21(Waf1/Cip1) expression by serum or the steroid was preceded by de novo synthesis of mRNA (data not shown).


Figure 4
Figure 4
View larger version (42K):
[in this window]
[in a new window]

  		Figure 4. (A) The effect of serum on the compartmental distribution of p21(Waf1/Cip1) and (B) the effect of dexamethasone on p21(Waf1/Cip1) location. Data points in A and B represent the mean ± SEM obtained in five different primary fibroblast cell lines. Open symbols, subconfluent; solid symbols, confluent; diamonds, 0.1% FCS; squares, 10% FCS. (C) The role of GR in the nuclear accumulation of p21(Waf1/Cip1) and its modification by the presence of serum was determined at the 6-hour time point. GR-decoy oligonucleotides = GR-decoy. Data points in A–C represent the mean ± SEM obtained in five different primary fibroblast cell lines. Open bars, subconfluent; solid bars, confluent. (D) The effect of dexamethasone, C/EBP-{alpha}, - β, and p21(Waf1/Cip1) on serum (FCS)-stimulated fibroblast proliferation at the third day of culture. Data points represent the mean ± SEM of cell counts obtained in six different primary fibroblast cell lines. (E) The role of p21(Waf1/Cip1) on steroid-dependent inhibition of fibroblast proliferation assessed by cell cycle analysis on a flow cyometer. *P < 0.01 based on analysis by paired, two-tailed Student's t test. Data points represent the mean ± SEM obtained in five different primary fibroblast cell lines. Small inhibitory RNA = "si", c-si = control (scrambled) small RNA, dexamethasone = dexa.

 
When subconfluent fibroblasts were treated with dexamethasone (10–8 M), the nuclear accumulation of p21(Waf1/Cip1) was unchanged for 6 hours, and decreased to 50% of its initial level (P < 0.01) 24 hours after the addition of the steroid in subconfluent cells, and this effect was not altered by the presence of 10% serum (Figure 4B). In confluent serum-deprived cells the steroid caused a significant increase of nuclear p21(Waf1/Cip1) at 6 hours (P < 0.01), but the level decreased to basal at 24 hours (Figure 4B). However, in the presence of 10% serum, dexamethasone (10–8 M) significantly increased nuclear p21(Waf1/Cip1) accumulation at 3 hours (P < 0.01), and this high level was maintained over 24 h (Figure 4B). Blocking steroid signaling by pretreating the cells for 30 minutes with either the GR antagonist RU486 (10–6–10–8 M) or for 1 hour with a GRE containing decoy oligonucleotide sequence (10 µM) inhibited the expression of p21(Waf1/Cip1) as determined at the 6-hour time point (Figure 4C).

We further show that steroid-induced (10–6–10–8 M) inhibition of fibroblast proliferation involved the action of C/EBP-{alpha} and p21(Waf1/Cip1), while down-regulation of C/EBP-β or control siRNA for C/EBP-{alpha} or p21(Waf1/Cip1) had no significant effect (Figure 4D). The requirement of p21(Waf1/Cip1) for the antiproliferative effect of steroids was further confirmed by cell cycle analysis using flow cytometry (Figure 4E). Serum induced a significant drop in cell in the G0/G1-phase of the cell cycle, while it increased the number of cells in the S-phase; this shift was significantly inhibited by dexamethasone (10–8 M), and the inhibitory effect of the steroid was not observed any more when cells were pretreated with siRNA targeting p21(Waf1/Cip1) (Figure 4E). Preincubation of cells with small inhibitory RNA targeting p27(Kip) had no effect on steroid-dependent cell cycle inhibition (data not shown).

Cell Density and p38 Mitogen-Activated Protein Kinase Regulate Steroid-Dependent Expression of p27(Kip1)
Only in confluent fibroblasts did steroid treatment (10–8 M) induce the expression of a second negative cell cycle regulator, p27(Kip), and this effect was independent of the presence of serum (Figure 5A). RU486 (10–6 M) or GRE decoy oligonucleotides (10 µM) diminished the effect of dexamethasone, while siRNA for C/EBP-{alpha} or C/EBP-β (1 µM) or decoy for NF-{kappa}B (1 µM) had no effect (Figure 5A). Interestingly, siRNA for p38 mitogen-activated protein (MAP) kinase, but not for Erk1/2 MAP kinase (1 µM), significantly reduced the expression of p27(Kip1) in dexamethasone-treated cells (Figure 5B). Furthermore, in the presence of 10% FCS the expression of p27(Kip1) was up-regulated when the Erk1/2 signaling pathway was blocked (Figure 5B).


Figure 5
View larger version (42K):
[in this window]
[in a new window]

  		Figure 5. (A) The role of GR, C/EBP-{alpha}, and -β signaling on the expression of p27(Kip) in nuclear cell extracts 12 hours after stimulation in confluent fibroblasts. (B) The role of Erk1/2 MAP kinase and p38 MAP kinase signaling on the expression of p27(Kip) in nuclear cell extracts 12 hours after stimulation in confluent fibroblasts. (C) The role of the GR, C/EBP-{alpha}, C/EBP-β, p21(Waf1/Cip1), and p27(Kip) on fibroblast proliferation at the third day. Data points shown in A, B, and C represent the mean ± SEM obtained in at least four different primary fibroblast cell lines. Small inhibitory RNA = "si", c-si = control (scrambled) small RNA, dexamethasone = dexa. (D) Summary of the results presented here and interpretation of the function of GR–C/EBP complexes as a consequence of cell density.

 
We further assessed the role of p21(Waf1/Cip1) and p27(Kip) on proliferation control in confluent contact inhibited cells with and without steroids. As expected, cell counts of subconfluent fibroblast stimulated with 10% FCS increased significantly within 3 days, and a single dose of dexamethasone at Day 1 reduced the proliferation dose-dependently (10–6, 10–8 M) by maximal 38.2 ± 4.7% (P < 0.001). The inhibition of C/EBP-{alpha} or p21(Waf1/Cip1) siRNA diminished the antiproliferative effect of the steroid, while siRNA for C/EBP-β or p27(Kip1) had no such effect (data not shown). In confluent fibroblasts 10% serum did not significantly increase cell numbers within 3 days, and treatment with neither dexamethasone (10–6 M, 10–8 M), p21(Waf1/Cip1), nor C/EBP-{alpha} siRNA had any significant effect on cell numbers (Figure 5C). However, when confluent fibroblasts were treated with siRNA for p27(Kip), the confluent cell numbers further increased significantly (P < 0.02) over a period of 3 days (Figure 5C). When p21(Waf1/Cip1) was blocked in addition to p27(Kip), the contact inhibition of cell proliferation was abolished in the presence and absence of the steroid (Figure 5C).


   DISCUSSION
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The presented data showed that the expression and the cell compartmental distribution of the GR, and the complex it forms with C/EBP-{alpha} and -β, is affected by cell density and the presence of serum. The steroid-dependent expression of p21(Waf1/Cip1) requires the presence of 42 kD C/EBP-{alpha}, and only in confluent fibroblasts did the steroid induce the expression of p27(Kip) via the activation of p38 MAP kinase and a 45/40-kD isoform of C/EBP-β. Furthermore, contact inhibition of confluent fibroblasts involved the combined action of p21(Waf1/Cip1) and p27(Kip).

Lung fibroblasts play a key role in the pathology of several human diseases including fibrosis, sarcoidosis, COPD, emphysema, and asthma (22, 30, 33, 34), and glucocorticoids are the major therapeutic drugs used to control the inflammatory and the profibrotic aspects of these diseases (1). However, the contribution of fibroblasts to the pathology of those diseases is not clear, nor has the benefit of inhaled glucocorticoids on the profibrotic processes been documented (35). It is widely accepted that the pathologic remodeling of the fibrotic lung is similar to badly controlled wound healing. In this study we used subconfluent cell layers combined with high serum levels to represent the initial phase of wound repair, while confluent fibroblasts and low serum represented intact undamaged tissue. The addition of serum to confluent fibroblasts then mimicked the condition found in tissue adjacent to a wound.

Our data suggest that cell density affects the cell compartmental distribution of the GR in fibroblasts and also modulates its biological availability. The GR is encoded by a single gene that is regulated by three independent promoters producing several GR isoforms, which are further modified by post-translational modifications. The expression of the various GR isoforms is assumed to be cell type, tissue, or species specific; however, their specific functions are not well defined (36). The best studied GR isoforms are GR{alpha} and GRβ. It was assumed that the GRβ isoform is only expressed in the nucleus, where it competes with the GR{alpha} for GRE binding (37). However, the expression of GRβ is species specific and may be cell type specific (34–36). Based on its characteristic molecular weight, we observed only the GR{alpha} isoform under all conditions, including confluent fibroblasts and glucocorticoid activation in human fibroblasts. However, in confluent fibroblasts the nuclear fraction of the GR was significantly increased compared with subconfluent cells. Nevertheless, the GRE-binding capacity of the nuclear GR isolated from confluent cells was not increased compared with subconfluent cells as assessed by EMSA and ELISA. Dexamethasone increased the GR–GRE binding capacity under all conditions indicating that the nuclear location of the GR in confluent fibroblasts was not caused by increased activation. Cell type– and cell layer location–specific distribution and activation of the GR is not new and had been described in the placenta (38). Interestingly, the authors observed that extended treatment with glucocorticoids resulted in a 90% down-regulation of GR expression in placenta fibroblasts, which is comparable with the effect we observed in lung fibroblast cultures. In our experiments the consequence of GR depletion was the loss of the antiproliferative effect of the glucocorticoid.

In fibroblasts the complexes that are formed by GR with members of the C/EBP family are affected by cell density and the composition of the complex seems to regulate the effect of steroids on p21(Waf1/Cip1). While a GR–C/EBP-{alpha} complex induces p21(Waf1/Cip1) expression, a complex of the GR with C/EBP-β activates p27(Kip), and interestingly, the 30-kD isoforms of both C/EBPs do not bind to GR. At this stage we are unable to conclude if the cell density affects the ratio of C/EBP-{alpha} and C/EBP-β isoforms, which subsequently form different complexes with GR, thereby partly controlling its different functions in subconfluent versus confluent fibroblasts. However, for C/EBP-β isoforms, opposing functions in regard to cell proliferation have been demonstrated (39, 40).

The antiproliferative effect of glucocorticoids is clearly linked with the activation of C/EBP-{alpha} and p21(Waf1/Cip1) (10, 15–18), while the necessity of p53 for p21(Waf1/Cip1) gene expression is not clear (10), and could also be a feedback mechanism of p53 on GR expression (31, 41). The findings are that the steroid-induced nuclear accumulation of p21(Waf1/Cip1) extends the life-span of senescent human fibroblasts (42) and cells of patients with Cushing syndrome (18). Furthermore, in pancreatic fibroblasts nuclear accumulation of p21(Waf1/Cip1) was associated with a phenotypic switch into myofibroblasts (29). These observations suggest that p21(Waf1/Cip1) in confluent fibroblasts may have different properties compared with proliferating subconfluent cells or may need additional factors to control proliferation.

In addition to p21(Waf1/Cip1), dexamethasone induced the expression of a second cell cycle control protein p27(Kip), which has been reported in other cell types before (43) and may depend on residue-specific phosphorylation or dimerization of the GR (44). Again the interaction of p21(Waf1/Cip1) with p27(Kip) may be cell type or species specific (3, 15, 36, 45). In our experiments deprivation of p21(Waf1/Cip1) expression alone in confluent serum-exposed fibroblasts was insufficient to overcome contact inhibition. Only when p27(Kip) was also down-regulated did confluent serum-stimulated fibroblasts lose contact inhibition and grow in multiple layers. The effect could not be stopped by dexamethasone treatment. The interaction of p27(Kip) with p21(Waf1/Cip1) in regard to cell proliferation control has been described by others (15, 37) and may even make cell cycle arrest independent of cyclin-dependent kinase 2 (3, 45).

In Figure 5D we provide a schemata that summarizes our findings, which imply that in subconfluent fibroblasts the GR forms a complex with the 42-kD C/EBP-{alpha}, which leads to p21(Waf1/Cip1) expression and cell cycle arrest, while in confluent cells GR binds preferably to the 45/40-kD C/EBP-β isoform, which induces p27(Kip) expression. These data are consistent with the function of GR being modified by cell differentiation, vascular leakage and inflammation and needs to be further addressed in the context of pathologies that either developed steroid resistance or that did not respond well to steroid therapy.


   Acknowledgments
 
The authors thank Mr. S'ng CT for reading and editing the manuscript.


   Footnotes
 
This research was supported by the Swiss National Foundation (3200B0–105737/1) and ASTRA Zeneca, Lund, Sweden.

Originally Published in Press as DOI: 10.1165/rcmb.2007-0079OC on November 7, 2007

Conflict of Interest Statement: M.T. from 2003–2005 received a three-year unrestricted research grant from ASTRA/Zeneca (Switzerland) in the amount of $72,000. None of the other authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form March 8, 2007

Accepted in final form October 14, 2007


   References
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Buckingham JC. Glucocorticoids: exemplars of multi-tasking. Br J Pharmacol 2006;147:S258–S268.[CrossRef][Medline]
  2. Dostert A, Heinzel T. Negative glucocorticoid receptor response elements and their role in glucocorticoid action. Curr Pharm Des 2004;10:2807–2816.[CrossRef][Medline]
  3. Smith E, Redman RA, Logg CR, Coetzee GA, Kasahara N, Frenkel B. Glucocorticoids inhibit developmental stage-specific osteoblast cell cycle: dissociation of cyclin A-cyclin-dependent kinase 2 from E2F4-p130 complexes. J Biol Chem 2000;275:19992–20001.[Abstract/Free Full Text]
  4. Wochnik GM, Ruegg J, Abel GA, Schmidt U, Holsboer F, Rein T. FK506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells. J Biol Chem 2005;280:4609–4616.[Abstract/Free Full Text]
  5. Jeffery PK. Remodelling and inflammation of bronchi in asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2004;1:176–183.[Abstract/Free Full Text]
  6. Silva CM, Powell-Oliver FE, Jewell CM, Sar M, Allgood VE, Cidlowski JA. Regulation of the human glucocorticoid receptor by long-term and chronic treatment with glucocorticoid. Steroids 1994;59:436–442.[CrossRef][Medline]
  7. Andreae J, Tripmacher R, Weltrich R, Rohde W, Keitzer R, Wahn U, Paul K, Buttgereit F. Effect of glucocorticoid therapy on glucocorticoid receptors in children with autoimmune diseases. Pediatr Res 2001;49:130–135.[Medline]
  8. Wallace AD, Cidlowski JA. Proteasome-mediated Glucocorticoid Receptor degradation restricts transcriptional signaling by glucocorticoids. J Biol Chem 2001;267:42714–42721.
  9. Aittomaki S, Pesu M, Groner B, Janne OA, Palvimo J, Silvennoinen O. Cooperation among Stat1, glucocorticoid receptor, and PU.1 in transcriptional activation of the high-affinity Fc gamma receptor I in monocytes. J Immunol 2000;164:5689–5697.[Abstract/Free Full Text]
  10. Terada Y, Okado T, Inoshita S, Hanada S, Kuwahara M, Sasaki S, Yamamoto T, Marumo F. Glucocorticoids stimulate p21(CIP1) in mesangial cells and in anti-GBM glomerulonephritis. Kidney Int 2001;59:1706–1716.[CrossRef][Medline]
  11. Mukhopadhyay SS, Wyszomierski SL, Gronostajski RM, Rosen JM. Differential interactions of specific nuclear factor I isoforms with the glucocorticoid receptor and STAT5 in the cooperative regulation of WAP gene transcription. Mol Cell Biol 2001;21:6859–6869.[Abstract/Free Full Text]
  12. Rüdiger JJ, Roth M, Bihl MP, Cornelius BC, Johnso M, Ziesche R, Block LH. Interaction of C/EBPalpha and the glucocorticoid receptor in vivo and in nontransformed human cells. FASEB J 2002;16:177–184.[Abstract/Free Full Text]
  13. Hayashi R, Wada H, Ito K, Adcock IM. Effects of glucocorticoids on gene transcription. Eur J Pharmacol 2004;500:51–62.[CrossRef][Medline]
  14. Ki SH, Cho IJ, Choi DW, Kim SG. Glucocorticoid receptor (GR)-associated SMRT binding to C/EBPbeta TAD and Nrf2 Neh4/5: role of SMRT recruited to GR in GSTA2 gene repression. Mol Cell Biol 2005;25:4150–4165.[Abstract/Free Full Text]
  15. Ramalingam A, Hirai A, Thompson EA. Glucocorticoid inhibition of fibroblast proliferation and regulation of the cyclin kinase inhibitor p21Cip1. Mol Endocrinol 1997;11:577–586.[Abstract/Free Full Text]
  16. Cram EJ, Ramos RA, Wang EC, Cha HH, Nishio Y, Firestone GL. Role of the CCAAT/enhancer binding protein-alpha transcription factor in the glucocorticoid stimulation of p21waf1/cip1 gene promoter activity in growth-arrested rat hepatoma cells. J Biol Chem 1998;273:2008–2014.[Abstract/Free Full Text]
  17. Eickelberg O, Roth M, Lorx R, Bruce V, Rudiger J, Johnson M, Block LH. Ligand- independent activation of the glucocorticoid receptor by beta2-adrenergic receptor agonists in primary human lung fibroblasts and vascular smooth muscle cells. J Biol Chem 1999;274:1005–1010.[Abstract/Free Full Text]
  18. Oddera S, Cagnoni F, Mangraviti S, Giron-Michel J, Popova O, Canonica GW. Effects of triamcinolone acetonide on adult human lung fibroblasts: decrease in proliferation, surface molecule expression and mediator release. Int Arch Allergy Immunol 2002;129:152–159.[CrossRef][Medline]
  19. Dube J, Chakir J, Laviolette M, Saint Martin S, Boutet M, Desrochers C, Auger F, Boulet LP. In vitro procollagen synthesis and proliferative phenotype of bronchial fibroblasts from normal and asthmatic subjects. Lab Invest 1998;78:297–307.[Medline]
  20. Warshamana GS, Martinez S, Lasky JA, Corti M, Brody AR. Dexamethasone activates expression of the PDGF-alpha receptor and induces lung fibroblast proliferation. Am J Physiol 1998;274:L499–L507.[Medline]
  21. Dik WA, Versnel MA, Naber BA, Janssen DJ, van Kaam AH, Zimmermann LJ. Dexamethasone treatment does not inhibit fibroproliferation in chronic lung disease of prematurity. Eur Respir J 2003;21:842–847.[Abstract/Free Full Text]
  22. Roth M, Johnson PR, Borger P, Bihl MP, Rüdiger JJ, King GG, Ge Q, Hostettler K, Burgess JK, Black JL, et al. Dysfunctional interaction of C/EBPalpha and the glucocorticoid receptor in asthmatic bronchial smooth-muscle cells. N Engl J Med 2004;351:560–574.[Abstract/Free Full Text]
  23. Woodruff PG, Dolganov GM, Ferrando RE, Donnelly S, Hays SR, Solberg OD, Carter R, Wong HH, Cadbury PS, Fahy JV. Hyperplasia of smooth muscle in mild to moderate asthma without changes in cell size or gene expression. Am J Respir Crit Care Med 2004;169:1001–1006.[Abstract/Free Full Text]
  24. Borger P, Matsumoto H, Boustany S, Cornelsen-Gencay MM, Tamm M, Burgess JK, King GG, Black JL, Roth M. Differential expression and drug dependent regulation of CCAAT/ Enhancer Binding Proteins (C/EBP) in airway smooth muscle cells of asthma and COPD patients. J Allergy Clin Immunol 2007;119:98–105.[CrossRef][Medline]
  25. Salvato G. Quantitative and morphological analysis of the vascular bed in bronchial biopsy specimens from asthmatic and non-asthmatic subjects. Thorax 2001;56:902–906.[Abstract/Free Full Text]
  26. Goulet S, Bihl MP, Gambazzi F, Tamm M, Roth M. Opposite effect of corticosteroids and long-acting beta(2)-agonists on serum- and TGF-beta(1)-induced extracellular matrix deposition by primary human lung fibroblasts. J Cell Physiol 2006;210:167–176.[CrossRef]
  27. Promwikorn W, Hawley SR, Pennington SR. Changes in p21(Cip1) and p27(Kip1) expression are not required for cell cycle entry and progression to S phase in Swiss 3T3 cells. Cell Signal 2000;12:619–627.[CrossRef][Medline]
  28. Cariveau MJ, Kovacs CJ, Allison RR, Johnke RM, Kalmus GW, Evans M. The expression of p21/WAF-1 and cyclin B1 mediate mitotic delay in x–irradiated fibroblasts. Anticancer Res 2005;25:1123–1129.[Abstract/Free Full Text]
  29. Manapov F, Muller P, Rychly J. Translocation of p21(Cip1/WAF1) from the nucleus to the cytoplasm correlates with pancreatic myofibroblast to fibroblast cell conversion. Gut 2005;54:814–822.[Abstract/Free Full Text]
  30. Hausman GJ. The influence of dexamethasone and insulin on expression of CCAAT/enhancer binding protein isoforms during preadipocyte differentiation in porcine stromal-vascular cell cultures: evidence for very early expression of C/EBPalpha. J Anim Sci 2000;78:1227–1235.[Abstract/Free Full Text]
  31. Zhu XY, Liu YJ, Lu J, Xu RB. Knockdown of glucocorticoid receptor expression by RNA interference promotes cell proliferation in murine macrophage RAW264.7 cells. J Steroid Biochem Mol Biol 2004;92:375–382.[CrossRef][Medline]
  32. Johnson PRA, Roth M, Tamm M, Hughes M, Qi G, King G, Black J. Airway smooth muscle cell proliferation is increased in asthma. Am J Respir Crit Care Med 2001;164:474–477.[Abstract/Free Full Text]
  33. Noordhoek JA, Postma DS, Chong LL, Vos JT, Kauffman HF, Timens W, van Straaten JF. Different proliferative capacity of lung fibroblasts obtained from control subjects and patients with emphysema. Exp Lung Res 2003;29:291–302.[CrossRef][Medline]
  34. Garantziotis S, Steele MP, Schwartz DA. Pulmonary fibrosis: thinking outside of the lung. J Clin Invest 2004;114:319–321.[CrossRef][Medline]
  35. Barnes PJ, Hansel TT. Prospects for new drugs for chronic obstructive pulmonary disease. Lancet 2004;364:985–996.[CrossRef][Medline]
  36. Yudt MR, Cidlowski JA. The glucocorticoid receptor: coding a diversity of proteins and responses through a single gene. Mol Endocrinol 2002;16:1719–1726.[Abstract/Free Full Text]
  37. Oakley RH, Jewell CM, Yudt MR, Bofetiado DM, Cidlowksi JA. The dominant negative activity of the human glucocorticoid receptor beta isoform. Specificity and mechanisms of action. J Biol Chem 1999;274:27857–27866.[Abstract/Free Full Text]
  38. Lee MJ, Wang Z, Yee H, Ma Y, Swenson N, Yang L, Kadner SS, Baergen RN, Logan SK, Garabedian MJ, et al. Expression and regulation of glucocorticoid receptor in human placental villous fibroblasts. Endocrinology 2005;146:4619–4626.[CrossRef][Medline]
  39. Bachmeier M, Loeffler G. The effect of platelet-derived growth factor and adipogenic hormones on the expression of CCAAT/enhancer binding protein in 3T3–L1 cells in serum-free conditions. Eur J Biochem 1997;243:128–133.[Medline]
  40. Grimm SL, Conteras A, Barcellos-Hoff MH, Rosen JM. Cell cycle defects contribute to a block in hormone induced mammary gland proliferation in CCAAT/enhancer binding protein (C/EBPbeta)-null mice. J Biol Chem 2005;280:36301–36309.[Abstract/Free Full Text]
  41. Urban G, Golden T, Aragon IV, Cowsert L, Cooper SR, Dean NM, Honkanen RE. Identification of a functional link for the p53 tumor suppressor protein in dexamethasone-induced growth suppression. J Biol Chem 2003;278:9747–9753.[Abstract/Free Full Text]
  42. Mawal-Dewan M, Frisoni L, Cristofalo VJ, Sell C. Extension of replicative lifespan in WI-38 human fibroblasts by dexamethasone treatment is accompanied by suppression of p21 Waf1/Cip1/Sdi1 levels. Exp Cell Res 2003;285:91–98.[CrossRef][Medline]
  43. Wang Z, Garabedian MJ. Modulation of glucocorticoid receptor transcriptional activation, phosphorylation, and growth inhibition by p27Kip1. J Biol Chem 2003;278:50897–50901.[Abstract/Free Full Text]
  44. Rogatsky I, Hittelman AB, Pearce D, Garabedian MJ. Distinct glucocorticoid receptor transcriptional regulatory surfaces mediate the cytotoxic and cytostatic effects of glucocorticoids. Mol Cell Biol 1999;19:5036–5049.[Abstract/Free Full Text]
  45. Reichert M, Eick D. Analysis of cell cycle arrest in adipocyte differentiation. Oncogene 1999;18:459–466.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2007-0079OCv1
38/4/414    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, J.-Q.
Right arrow Articles by Roth, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yang, J.-Q.
Right arrow Articles by Roth, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Crit. Care Med.
Copyright © 2008 American Thoracic Society.