help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on August 20, 2007, doi:10.1165/rcmb.2007-0162OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2007-0162OCv1
38/2/168    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 Matsuda, A.
Right arrow Articles by Saito, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Matsuda, A.
Right arrow Articles by Saito, H.
American Journal of Respiratory Cell and Molecular Biology. Vol. 38, pp. 168-175, 2008
© 2008 American Thoracic Society
DOI: 10.1165/rcmb.2007-0162OC

Th1/Th2 Cytokines Reciprocally Regulate In Vitro Pulmonary Angiogenesis via CXC Chemokine Synthesis

Akio Matsuda1, Shuhei Fukuda1, Kenji Matsumoto1 and Hirohisa Saito1

1 Department of Allergy and Immunology, National Research Institute for Child Health and Development, Tokyo, Japan

Correspondence and requests for reprints should be addressed to Akio Matsuda, Ph.D., Department of Allergy and Immunology, National Research Institute for Child Health and Development, 2-10-1 Okura, Setagaya-ku, Tokyo, 157-8535, Japan. E-mail: amatsuda{at}nch.go.jp


   Abstract
Top
 Abstract
CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Hypervascularity is known as an important element of airway remodeling in bronchial asthma. However, it remains obscure how allergic inflammation relates to angiogenesis in the lung. In this study, we examined the in vitro effects of inflammatory cytokines on endothelial cell functions, particularly angiogenesis. Human microvascular endothelial cells from normal lung (HMVEC-Ls) were cultured with TNF-{alpha}, IFN-{gamma}, IL-4, a combination of TNF-{alpha} and IFN-{gamma}, or a combination of TNF-{alpha} and IL-4, and the cell proliferation and tube-forming activities were evaluated. IL-4 slightly enhanced the proliferation of HMVEC-Ls in the presence of vascular endothelial growth factor (VEGF), whereas TNF-{alpha} and IFN-{gamma} tended to inhibit it. Synergistic inhibition was observed when TNF-{alpha} and IFN-{gamma} were simultaneously added to the culture medium. The combination of IL-4 and TNF-{alpha} markedly promoted tube formation by HMVEC-Ls, even in the absence of VEGF. The IL-4 and TNF-{alpha} combination induced autocrine production of CXCR2 chemokines, which are known to have angiogenic activity, whereas the production of angiostatic CXCR3 chemokines was dramatically up-regulated when TNF-{alpha} and IFN-{gamma} were present. The marked IL-4– and TNF-{alpha}–induced tube formation was inhibited by a selective CXCR2 antagonist. These results suggest that, in the presence of TNF-{alpha}, IL-4 and IFN-{gamma} reciprocally regulate tube formation by HMVEC-Ls through autocrine synthesis of CXCR2 and CXCR3 chemokines, respectively. Of note, the CXCR2 chemokine-induced tube formation was independent of VEGF. Therefore, CXCR2 chemokines may represent potential therapeutic targets for bronchial asthma.

Key Words: airway remodeling • angiogenesis • CXC chemokine • cytokine • Th1/Th2



   CLINICAL RELEVANCE
Top
 Abstract
 CLINICAL RELEVANCE
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Our findings revealed that autocrine CXCR2 chemokines are indispensable for lung angiogenesis, and might shed new light on the potential of CXCR2 as a novel therapeutic target for airway hypervascularity, particularly in patients with severe asthma.

 
Bronchial asthma is a chronic inflammatory disease of the airways. T cells, especially type 2 helper T (Th2) cells play a critical role in the pathogenesis of asthma through the production of a variety of cytokines (1). Th2-type cytokines such as IL-4, -5, and -13 are known to influence a wide range of events associated with chronic allergic inflammation in local tissues.

In some patients with astma, such chronic inflammation results in persistent structural changes in the airway wall, including shedding of epithelial cells, smooth muscle cell hypertrophy/hyperplasia, matrix deposition, and accelerated angiogenesis (2). These structural changes are collectively described as airway remodeling, which is thought to cause irreversible airflow obstruction and exacerbation of asthma. A number of studies have suggested that, among the elements of airway remodeling, hypervascularity occurs even in the early stage of chronic asthma (3–5). Furthermore, a significant correlation has also been demonstrated between the number of blood vessels in the airway wall and the severity of asthma (4, 6). However, hypervascularity of the airway wall is not found in chronic obstructive pulmonary disease (COPD), another chronic inflammatory disease of the lung (5).

Vascular endothelial growth factor (VEGF), a proangiogenic factor, is thought to play a major role in new vessel formation in the asthmatic airway (7, 8). Indeed, elevated expression of VEGF has been reported in asthmatic airways compared with those of normal control subjects (7–10). It has also been reported that a variety of cells in the asthmatic airway, such as airway epithelial cells (11), smooth muscle cells (12, 13), and mast cells (7), produce VEGF. Recently, lung-specific and inducible overexpression of VEGF in a mouse model of bronchial asthma has been shown to enhance not only angiogenesis in the airways, but also Th2 inflammation, myocyte hyperplasia, and airway hyperresponsiveness (14). More importantly, the biological effects of overexpression of VEGF, such as angiogenesis, were reversed by cessation of the transgenic VEGF elaboration in that murine model. In addition, in patients with mild asthma, inhaled corticosteroid treatment, which is now widely recognized as the first-line therapy for bronchial asthma, was found to effectively reduce VEGF production by the airway epithelial cells and smooth muscle cells, thereby reducing the submucosal vascularity (7, 11, 15). Thus, the effects of VEGF on the development of airway hypervascularity in asthma appear to be dependent upon its expression level.

However, in many patients with severe corticosteroid-dependent asthma, and in those treated with inhaled corticosteroids for more than 5 years, increased airway vascularity does exist despite the long-term treatment with corticosteroids (6, 16). Therefore, we hypothesized that additional factor(s) besides VEGF must contribute to the development of hypervascularity of the airways, particularly in patients with severe asthma.

In this study, we focused on the CXC chemokine family as a likely candidate for such factor(s) for the following two reasons. First, CXC chemokines are divided into two subgroups based on their disparate properties regarding angiogenesis: an angiogenic CXCR2 chemokine subgroup, such as CXC chemokine ligand 8 (CXCL8), and an angiostatic CXCR3 chemokine subgroup, such as CXCL10 (17–22). Second, in patients with severe asthma, local production of CXC chemokines, including CXCL8, has been reported (23, 24). However, it remains to be determined whether or not CXCR2 chemokines are involved in the development of hypervascularity of asthmatic airways. In this context, we examined whether representative cytokines inducing Th2 inflammation (i.e., combination of TNF-{alpha} and IL-4) are associated with angiogenesis, especially the formation of tube-like structures. We employed human microvascular endothelial cells derived from the normal lung (HMVEC-Ls) in vitro and found that autocrine CXCR2 chemokines play a crucial role in Th2 inflammatory cytokine–induced tube formation.


   MATERIALS AND METHODS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Reagents
Recombinant human TNF-{alpha}, IFN-{gamma}, IL-4, and CXCL8 were purchased from PeproTech (Rocky Hill, NJ). Recombinant human VEGF was purchased from R&D Systems (Minneapolis, MN). SB225002 was purchased from Calbiochem (La Jolla, CA). Human Universal Reference (HUR) total RNA was purchased from BD Biosciences (Palo Alto, CA).

Primary Human Endothelial Cell Culture
HMVEC-Ls were purchased from Cambrex (Walkersville, MD) and maintained exactly as recommended by the manufacturer. We obtained two different lots from individual donors, and all data in this study were reproducible between these two lots of HMVEC-Ls. All the experiments described in this study were performed on cells in the second or third passage. Cells were usually cultured with the EGM-2-MV BulletKit (Cambrex) containing growth factors such as epidermal growth factor (EGF), insulin-like growth factor-1 (IGF-1), fibroblast growth factor-2 (FGF2), and VEGF, which are essential for maintaining the phenotype of endothelial cells as confirmed by the expressions of von Willebrand factor and surface platelet endothelial cell adhesion molecule (PECAM/CD31) on the cells.

Proliferation Assay
HMVEC-Ls were suspended in the EGM-2-MV medium not containing growth factors (BulletKit; Cambrex) and cultured for 24 hours in 96-well culture plates (5 x 103 cells per well). After the culture under growth factor–deprived conditions, growth factors were added to the culture either alone or in combination at equivalent concentrations as recommended by the manufacturer. The cells were cultured for a further 48 hours. To examine the effects of cytokines on HMVEC-L proliferation, HMVEC-Ls were cultured for 72 hours with 50 ng/ml TNF-{alpha}, 50 ng/ml IFN-{gamma}, 50 ng/ml IL-4, a combination of TNF-{alpha} and IFN-{gamma} (10–50 ng/ml), or a combination of TNF-{alpha} and IL-4 (10–50 ng/ml), in the presence or absence of FGF2 and VEGF (BulletKit; Cambrex). Cell proliferation was determined by measuring [3H] thymidine (Amersham, Roosendal, The Netherlands) incorporation (pulsed at 1 µCi/well for 6 h).

In Vitro Tube Formation Assay
HMVEC-Ls were pretreated with cytokines for 48 hours before the in vitro tube formation assay. Growth factor–reduced Matrigel (BD Biosciences, Bedford, MA) was thawed at 4°C, and then a 150-µl aliquot of Matrigel was added to each well of 48-well plates and polymerized for 30 minutes at 37°C immediately before addition of the HMVEC-L suspensions (4 x 104 cells/500 µl) in growth factor–deprived EGM-2MV medium. In some experiments, various concentrations of recombinant human VEGF, CXCL8, or a selective CXCR2 antagonist, SB225002, were added to the wells simultaneously with the cells, as described. After 18 hours of incubation, each well was scanned over an equal area and number (5 fields; 1 center field and 4 around the center field/well) of fields under a light microscope (IX70; Olympus, Tokyo, Japan) equipped with a digital camera (DP11; Olympus). The images were printed at a constant magnification, and the length of the tubes formed was measured manually, followed by calculation of the total relative length of the tube-like structures formed per field.

RT-PCR and Real-Time RT-PCR
Primer sets for the following 11 genes were synthesized at Sawady Technology (Tokyo, Japan): CXCL1 (sense, 5'-AAAGCTTGCCTCAATCCTGCAT-3'; antisense, 5'-TCCTCCCTTCTGGTCAGTTGGA-3'), CXCL5 (sense, 5'-GCCGCTTAAGCTTTCAGCTC-3'; antisense, 5'-CCAGTGATTCCTGGCTCACA-3'), CXCL6 (sense, 5'-TGAAGAGTGTGGGGGAAAGC-3'; antisense, 5'-GGTCAATTGCCAAAGGGTTC-3'), CXCL8 (sense, 5'-GTCTGCTAGCCAGGATCCACAA-3'; antisense, 5'-GAGAAACCAAGGCACAGTGGAA-3'), CXCL9 (sense, 5'-AAGGGACTATCCACCTACAATCC-3'; antisense, 5'-CCCATTCTTTTGCTTTTTCTTTT-3'), CXCL10 (sense, 5'-GCCAATTTTGTCCACGTGTTG-3'; antisense, 5'-AGCCTCTGTGTGGTCCATCCT-3'), CXCL11 (sense, 5'-CTTGGCTGTGATATTGTGTGCT-3'; antisense, 5'-CTTGCTTCGATTTGGGATTTAG-3'), CXCR2 (sense, 5'-GGCCTTCCTTTGTTGGCTCT-3'; antisense, 5'-CACCAGGGCAAGCTTTCTAA-3'), CXCR3-B (sense, 5'-TGGCGGGGACAGTTATAGGA-3'; antisense, 5'-AACCTCGGCGTCATTTAGCA-3'), VEGF (sense, 5'-ATGAACTTTCTGCTGTCTTGGGTGC-3'; antisense, 5'-TCACCGCCTCGGCTTGTCAC-3') (28), and β-actin (sense, 5'-CCCAGCCATGTACGTTGCTAT-3'; antisense, 5'-TCACCGGAGTCCATCACGAT-3'). Total RNA samples were isolated from the HMVEC-Ls using RNeasy (Qiagen, Valencia, CA) and digested with RNase-free DNase I (Qiagen) in accordance with the manufacturer's instructions. First-strand cDNA was synthesized using Oligo(dT) (12–18 mers) primer (Invitrogen, Carlsbad, CA) and Superscript III (Invitrogen) as the reverse transcriptase. For RT-PCR, cDNA generated from 50 ng of total RNA was amplified using rTaq polymerase (Toyobo, Osaka, Japan) under the following cycling conditions: 94°C/30 seconds, 60°C/30 seconds, 72°C/30 seconds; 35 cycles for CXCR2, CXCR3-B, and VEGF, and 20 cycles for β-actin. Real-time quantitative RT-PCR analysis was performed with an Applied Biosystems Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA) using iTaq SYBR Green Supermix with ROX (Bio-Rad, Hercules, CA), as previously reported (25). To determine the exact copy numbers of the target genes, quantified concentrations of the purified PCR products of CXCL1, 5, 6, 8, 9, 10, 11, and β-actin were serially diluted and used as standards in each experiment. Aliquots of cDNA equivalent to 5 ng of the total RNA samples were used for each real-time RT-PCR. The levels were normalized to the β-actin level in each sample.

Measurement of Chemokines in the Culture Supernatant
The chemokine concentrations in the supernatant were measured with enzyme-linked immunosorbent assay (ELISA) kits specifically recognizing CXCL1, 5, 6, 8, 9, 10, and 11 (R&D Systems), in accordance with the manufacturer's instructions.

Statistical Analysis
All data are presented as the mean ± SD unless otherwise indicated. Differences between groups were analyzed using the Mann-Whitney U-test and were considered to be significant when P < 0.05.


   RESULTS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Proliferation of HMVEC-Ls In Vitro
HMVEC-Ls are usually cultured in a growth medium supplemented with growth factors such as VEGF, FGF2, IGF-1, and EGF, as described in MATERIALS AND METHODS. Thus, in the first line of experiments, we determined the effects of those growth factors, either alone or in combination, on the proliferation of the HMVEC-Ls in vitro. We found that VEGF and FGF2 promoted the proliferation of the HMVEC-Ls to an equal degree, and when used in combination their effects were additive. On the other hand, EGF and IGF-1 induced no responses in the HMVEC-Ls (Figure 1A). To evaluate the effects of inflammatory cytokines on the VEGF plus FGF2-mediated proliferation of the HMVEC-Ls, the HMVEC-Ls were cultured for 72 hours with TNF-{alpha}, IFN-{gamma}, IL-4, a combination of TNF-{alpha} and IFN-{gamma}, or a combination of TNF-{alpha} and IL-4 in the presence or absence of VEGF and FGF2. IL-4 slightly enhanced the proliferation of HMVEC-Ls in the presence of VEGF, while TNF-{alpha} and IFN-{gamma} tended to inhibit it. Synergistic inhibition was observed when TNF-{alpha} and IFN-{gamma} were simultaneously added (Figure 1B).


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		Figure 1. Proliferation assay of human microvascular endothelial cells derived from the normal lung (HMVEC-Ls). (A) Vascular endothelial growth factor (VEGF) and fibroblast gorwth factor (FGF)2 equally promoted proliferation of the HMVEC-Ls. To determine the effects of growth factors (BulletKit; Cambrex) on the proliferation of the HMVEC-Ls, the cells were cultured with or without growth factors for 48 hours, as indicated in the figure, after growth factor deprivation for 24 hours. (B) Effects of type 1 helper T (Th1) and Th2 cytokines on proliferation of HMVEC-Ls. HMVEC-Ls were cultured with various concentrations of the cytokines in the presence or absence of VEGF and FGF2 for 72 hours. Cell proliferation was determined by measurement of [3H] thymidine incorporation (pulsed at 1 µCi/well for 6 h). The results are shown as the mean ± SD of triplicate wells and are representative of the results obtained in at least three independent experiments. *P < 0.05.

 
In Vitro Tube Formation by HMVEC-Ls
To evaluate the effects of inflammatory cytokines on the tube-forming activity of the HMVEC-Ls, the cells were plated on Matrigel in the presence or absence of 50 ng/ml VEGF after first being primed with cytokines for 48 hours. Then, after an additional 18 hours of incubation, the tube length in each well was measured (Figure 2). As expected, VEGF promoted the in vitro tube-forming activity of the HMVEC-Ls in the control culture. In the presence of VEGF, neither TNF-{alpha} alone nor IFN-{gamma} alone showed any significant effects on the tube formation; however, IL-4 significantly promoted the tube-forming activity. Of note, combination of TNF-{alpha} and IL-4 promoted the tube formation by HMVEC-Ls to a far greater degree, even in the absence of exogenous VEGF. On the other hand, combination of TNF-{alpha} and IFN-{gamma} showed only a marginal effect on the in vitro tube-forming activity of the cells, even in the presence of VEGF (Figure 2).


Figure 2
View larger version (51K):
[in this window]
[in a new window]

  		Figure 2. Effects of cytokines on the in vitro tube-forming activity of HMVEC-Ls. (A) HMVEC-Ls were pretreated for 48 hours with 10 ng/ml of each cytokine in EGM-2MV complete medium. Then, HMVEC-L suspensions (4 x 104 cells/500 µl) in growth factor–deprived EGM-2MV medium containing the same concentrations of the cytokines as those used in the pretreatment were plated on Matrigel in the presence or absence of 50 ng/ml of recombinant human VEGF. Images of the center field of the wells after 18 hours of incubation are shown. (B) The relative tube length per field in five fields from each well was calculated. The results are shown as the mean ± SD of five independent fields in each well and are representative of the results obtained in at least three independent experiments.

 
To examine the possibility that endogenous VEGF acts in an autocrine manner to contribute to the marked tube-forming activity seen under Th2 inflammatory conditions (a combination of TNF-{alpha} and IL-4), the mRNA expression of each VEGF isoform was assessed by semiquantitative RT-PCR. We used a pair of PCR primers that would allow us to distinguish all the six known alternative transcripts of VEGF (26), and we found that the HMVEC-Ls showed only constitutive expression of the VEGF-189, VEGF-165, and VEGF-121 variants regardless of the addition of inflammatory cytokines, even TNF-{alpha} and IL-4 (Figure 3). In addition, to examine the possibility that the enhanced expression of VEGF receptor upon IL-4 stimulation contributes to the VEGF-induced tube-forming activity (Figure 2), we assessed the expression of mRNA for VEGF receptor 1 (Flt-1) and VEGF receptor 2 (KDR). The expression of both Flt-1 and KDR mRNA in the cells was not altered even in the presence of inflammatory cytokines (data not shown). Taken together, it was concluded that the marked tube-forming activity of HMVEC-Ls in culture medium containing TNF-{alpha} and IL-4 was not mediated by VEGF autocrine activity. We then focused on the CXC chemokine family, another regulator of angiogenesis.


Figure 3
View larger version (44K):
[in this window]
[in a new window]

  		Figure 3. Expression of VEGF mRNAs in the HMVEC-Ls. HMVEC-Ls were cultured with 10 ng/ml of each cytokine for 72 hours, and the endogenous total VEGF expression was determined by semiquantitative RT-PCR. cDNA for Human Universal Reference (HUR) total RNA was used as the positive control, and the results are representative of at least three independent experiments.

 
Expression of CXC Chemokines in HMVEC-Ls
Two subgroups of CXC chemokines have been found to possess disparate angiogenic activities, depending upon the presence or absence of a common structural/functional motif (Glutamic acid-Leucine-Arginine; ELR motif) (17–20). The CXC chemokines with the ELR motif, including growth-related oncogene {alpha} (Gro-{alpha})/CXCL1, Gro-β/CXCL2, Gro-{gamma}/CXCL3, epithelial neutrophil-activating peptide 78 (ENA78)/CXCL5, granulocyte chemotactic protein 2 (GCP-2), and IL-8/CXCL8, promote angiogenesis (17). Their angiogenic activity is mediated via their common receptor, CXC chemokine receptor 2 (CXCR2), which is expressed on microvascular endothelial cells (21, 22), but not via CXCR1 (22). In contrast, the CXC chemokines that are interferon-inducible and lack the ELR motif, including monokine induced by IFN-{gamma} (MIG)/CXCL9, IFN-inducible protein 10 (IP-10)/CXCL10, and IFN-inducible T cell {alpha} chemoattractant (I-TAC)/CXCL11, inhibit angiogenesis (17), and their angiostatic activity is mediated via their common receptor, CXCR3-B, which is also expressed on microvascular endothelial cells (27).

The effects of Th1 and Th2 cytokines on the expression of members of the CXC chemokine family in the presence of various cytokines were evaluated by measuring their mRNA and protein levels by real-time PCR and ELISA. Regarding CXCR2 chemokines, TNF-{alpha} alone induced CXCL1 and CXCL8 production, whereas IFN-{gamma} alone and IL-4 alone had no effect on the production of any of the CXCR2 chemokines (Figure 4A). In combination with TNF-{alpha}, IFN-{gamma} showed a synergistic effect on CXCL5 production, but no additive effects of these cytokines were found on the production of any of the other CXCR2 chemokines tested for. In contrast, IL-4 showed significant synergistic effects on the production of all of the CXCR2 chemokines tested for. The protein expression levels of all of the CXCR2 chemokines in the culture supernatants correlated closely with their respective mRNA levels (Figure 4A).


Figure 4
View larger version (27K):
[in this window]
[in a new window]

  		Figure 4. Expression of members of the CXC chemokine family in HMVEC-Ls. Th1 and Th2 cytokine–induced up-regulation of the mRNAs (open bars) and proteins (shaded bars) of the CXCR2 chemokine family (A) and CXCR3 chemokine family (B) in the HMVEC-Ls is shown. HMVEC-Ls were cultured with 10 ng/ml of each cytokine for 72 hours. The mRNA levels of CXCL1, 5, 6, 8, 9, 10, and 11 were determined by SYBR Green-based real-time quantitative PCR. The concentrations of the chemokines in the supernatants were measured by ELISA. The results are shown as the mean ± SD of triplicate samples and are representative of the results obtained in at least three independent experiments. *P < 0.05.

 
In contrast to the CXCR2 chemokines, while IFN-{gamma} alone significantly induced the production of CXCR3 chemokines, TNF-{alpha} alone and IL-4 alone showed no effect. On the other hand, combined addition of TNF-{alpha} and IFN-{gamma} exhibited a marked synergistic effect on the production of all three CXCR3 chemokines, while the addition of IL-4 showed little effect. The protein levels correlated well with the respective mRNA levels, except for CXCL11 (Figure 4B). Collectively, CXCR2 chemokine production was predominantly induced under Th2 inflammatory conditions rather than under Th1 inflammatory conditions (combination of TNF-{alpha} and IFN-{gamma}), whereas CXCR3 chemokine production was predominantly induced under Th1 inflammatory conditions compared with under Th2 inflammatory conditions. These results raise the possibility that reciprocal production of CXCR2/CXCR3 chemokines by Th2/Th1 cytokines, respectively, regulates the tube-forming activity of HMVEC-Ls. Therefore, we next investigated the expression of the receptors for the CXC chemokines in the HMVEC-Ls.

Expression of CXC-Chemokine Receptors in HMVEC-Ls
Semiquantitative RT-PCR was performed to examine the mRNA expression of CXCR2 and CXCR3-B in the HMVEC-Ls. The mRNAs for both CXCR2 and CXCR3-B were constitutively expressed in the HMVEC-Ls even in the absence of any cytokines (Figure 5). Although TNF-{alpha} alone and IFN-{gamma} alone did not affect the mRNA expression level of CXCR2 in the HMVEC-Ls, more abundant expression of the mRNA for CXCR2 was found when the cells were cultured with IL-4 alone, under both Th1 and Th2 inflammatory conditions. On the other hand, the mRNA for CXCR3-B was constitutively expressed in the HMVEC-Ls, and its expression level did not change in the presence of cytokines (Figure 5).


Figure 5
View larger version (43K):
[in this window]
[in a new window]

  		Figure 5. Expression of CXCR2 and CXCR3-B mRNA in HMVEC-Ls. HMVEC-Ls were cultured with 10 ng/ml of each cytokine for 72 hours, and the expressions of CXCR2 and 3-B were determined by semiquantitative RT-PCR analysis. cDNA for peripheral blood monocytes (PBMC) was used as the positive control, and the results are representative of least three independent experiments.

 
Functional Analysis of CXCR2 in HMVEC-Ls
Functional analysis of CXCR2 in the HMVEC-Ls was performed to examine the actual involvement of the CXCR2 chemokine/CXCR2 system in the tube-forming activity of these cells.

First, we examined the effect of exogenous CXCL8 on the in vitro tube-forming activity of the HMVEC-Ls. It was found that recombinant human CXCL8 significantly promoted the tube-forming activity of the cells in a dose-dependent manner (Figure 6A). The tube-forming activity induced by CXCL8 was almost comparable to that of VEGF at the optimal concentration of 50 ng/ml.


Figure 6
View larger version (11K):
[in this window]
[in a new window]

  		Figure 6. Functional analysis of CXCR2 expressed on HMVEC-Ls. (A) HMVEC-Ls were suspended (4 x 104 cells/500 µl) in growth factor–deprived EGM-2MV medium containing different concentrations of recombinant human CXCL8 (solid triangles) or VEGF (solid circles) and plated on Matrigel. The relative tube length per field in five fields from each well was calculated. The results are shown as the mean ± SD of five independent fields in each well. The values are representative of the results obtained in three independent experiments. (B) SB225002 specifically inhibited the CXCL8-induced tube-forming activity of the HMVEC-Ls. HMVEC-L suspensions (4 x 104 cells/500 µl) containing 50 ng/ml of recombinant human CXCL8 (solid bars) or VEGF (open bars) were plated on Matrigel, and different concentrations of SB225002 were added to the wells. The relative tube length per field in five fields from each well was calculated, and the % inhibition was evaluated by comparing the relative tube length per field in each experiment with that in the control (no SB225002). The results are shown as the mean ± SD of the % inhibition calculated for five equivalent fields. The values are representative of the results obtained in three independent experiments. (C) SB225002 partially inhibited the tube-forming activity of the HMVEC-Ls stimulated with a combination of TNF-{alpha} and IL-4. HMVEC-Ls were pre-treated with 10 ng/ml of each cytokine for 48 hours in the presence (closed bars) or absence (open bars) of 200 nM of SB225002. HMVEC-L suspensions (4 x 104 cells/500 µl) containing the same concentrations of the cytokines and SB225002 as those used in the pre-treatment were plated on Matrigel. The relative tube length per field in five fields from each well was calculated. Representative results are shown from three independent experiments. * P < 0.05.

 
Second, to test whether or not the in vitro tube-forming activity induced by exogenous CXCL8 is mediated by CXCR2, we added SB225002, a selective antagonist of CXCR2 (28, 29), to the culture medium containing 50 ng/ml of VEGF or CXCL8 (Figure 6B). SB225002 caused marked attenuation of the CXCL8-induced tube-forming activity of the HMVEC-Ls in a dose-dependent manner. This result indicates that CXCR2 is functionally expressed on the HMVEC-Ls. SB225002 also showed modest attenuation of the VEGF-induced tube-forming activity, although the degree of inhibition of the VEGF-induced tube-forming activity was much less than that of the CXCL8-induced tube-forming activity.

Finally, we examined whether or not the tube-forming activity in the presence of TNF-{alpha} and IL-4 was mediated by CXCR2 chemokines. Based on the results in Figure 6B, we used 200 nM SB225002 for specific inhibition of CXCR2. The tube-forming activity of the HMVEC-Ls in the presence of TNF-{alpha} and IL-4 was partially but significantly inhibited by the addition of 200 nM SB225002 (Figure 6C). This result strongly suggests involvement of the CXCR2 chemokines and their receptors in the tube-forming activity of the HMVEC-Ls under Th2 inflammatory conditions in an autocrine manner.


   DISCUSSION
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Since VEGF-mediated airway hypervascularity in patients with mild asthma is largely reversed by corticosteroid treatment (7, 11, 15), we hypothesized that additional factor(s) other than VEGF must contribute to the development of steroid-insensitive hypervascularity in patients with severe asthma. It should be stressed that marked in vitro tube formation by HMVEC-Ls was observed when both IL-4 and TNF-{alpha} were present and that it occurred in a VEGF-independent manner. On the other hand, in the presence of VEGF, even IL-4 alone induced tube formation (Figure 2). When VEGF-independent tube formation was observed in the presence of both IL-4 and TNF-{alpha}, angiogenic CXCR2 chemokines were produced in an autocrine manner. On the other hand, IL-4 alone failed to induce CXCR2 chemokine production even when tube formation was observed in the presence of VEGF (Figure 4A). These results highlight the importance of TNF-{alpha} for VEGF-independent/CXCR2 chemokine family-dependent angiogenesis in vitro. Indeed, recent reports indicated increased expression of TNF-{alpha} in severe and refractory asthma (30, 31). Taken together, microvascular endothelial cells are suggested to play a crucial role in TNF-{alpha}–associated angiogenesis in patients with severe asthma. Of note, our findings also revealed for the first time that autocrine CXCR2 chemokines are indispensable for in vitro tube formation of microvascular endothelial cells in the presence of TNF-{alpha} and IL-4. Such angiogenic CXCR2 chemokines also act as chemoattractants for neutrophils. Neutrophils were recently reported to be poorly responsive to corticosteroids and to play a far more important role in the inflammation of asthma than had been thought, particularly in patients with severe disease (24). Therefore, our findings might shed new light on the potential of CXCR2 as a novel therapeutic target for airway hypervascularity as well as neutrophilic inflammation (32), particularly in patients with severe and/or steroid-insensitive asthma.

Angiogenesis is defined as the growth of new vessels from preexisting ones (33). The process of angiogenesis is regulated by a complex interaction of activators and inhibitors, and inappropriate production of these factors may cause aberrant angiogenesis. A recent report indicated the importance of the CXC chemokine-mediated angiogenic/angiostatic balance in bleomycin-induced pulmonary fibrosis. That study, using the corneal micropocket assay, clearly showed that a CXCR3 chemokine, CXCL11, could inhibit CXCL8-mediated angiogenesis (34). Thus, aberrant vascularity might be regulated by the CXC chemokine-mediated angiogenic/angiostatic balance. In our study, the CXCR2 chemokines were predominantly induced under Th2 inflammatory conditions rather than under Th1 inflammatory conditions. In contrast, the CXCR3 chemokines were induced by IFN-{gamma} alone, with TNF-{alpha} showing a synergistic effect, but IL-4 showed only a marginal effect (Figure 4). These results indicated that, in the presence of TNF-{alpha}, Th2/Th1 cytokines induce different sets of CXC chemokines, that is, those with and without ELR motifs, respectively. We also found that IL-13, another Th2 cytokine whose signal transduction pathway overlaps with that of IL-4, showed significant synergistic effects on the production of the CXCR2 chemokines in the presence of TNF-{alpha} (data not shown). Therefore, we surmised that reciprocal production of CXCR2/CXCR3 chemokines by Th2/Th1 cytokines, respectively (Figure 4), resulted in the different tube-forming activity (Figure 2). Indeed, we found that CXCR2 was functionally expressed on the HMVEC-Ls, and the IL-4 and TNF-{alpha} combination-induced tube formation was mediated via CXCR2 (Figure 6), suggesting that, in the presence of TNF-{alpha}, IL-4 and IFN-{gamma} reciprocally regulate tube formation by HMVEC-Ls via autocrine synthesis of CXCR2 and CXCR3 chemokines, respectively.

Although TNF-{alpha} induced substantial production of CXCL1 and CXCL8 by HMVEC-Ls (Figure 4A), TNF-{alpha} did not promote the in vitro tube-forming activity (Figure 2). These observations would appear to contradict our proposal that inflammatory cytokine-mediated autocrine synthesis of CXCR2 chemokines contributes to the tube-forming activity. Importantly, TNF-{alpha} did not enhance the mRNA level of CXCR2, whereas IL-4 and the combination of IL-4 and TNF-{alpha} significantly enhanced the expression of mRNA for CXCR2 (Figure 5). These results are consistent with a recent report of induction of functional CXCR2 by IL-4 in human monocytes (35). Therefore, the additive effects of both the CXCR2 chemokines and their receptor, CXCR2, may play an important role in the marked tube-forming activity seen under Th2 inflammatory conditions; however, such additive effects cannot be expected with TNF-{alpha} alone.

As shown in Figure 1A, VEGF and FGF2 promoted the proliferation of the HMVEC-Ls to an equal degree, whereas EGF and IGF-1 had no such effect. This cytokine-response profile was different from that of human umbilical vein endothelial cells (HUVECs), which were significantly induced to proliferate by FGF2 and EGF but not VEGF (36, and data not shown). The responses to different cytokines vary depending on the origin of the endothelial cells (37). Thus, to clarify the cytokine regulation of angiogenesis in the lung, HMVEC-Ls should be tested. IL-4 slightly enhanced the proliferation of HMVEC-Ls in the presence of VEGF, whereas TNF-{alpha} and IFN-{gamma} tended to inhibit it. IL-4 also showed a modest increase in the exogenous VEGF-dependent (Figure 2) and CXCR2 chemokine-independent (Figure 4A) tube-forming activity of the HMVEC-Ls. The mRNA expression of both Flt-1 and KDR, which are VEGF receptors, was not altered at all by IL-4 when compared with the control culture (data not shown). Furthermore, we found that IL-4 supported substantial survival of HMVEC-Ls, resulting in protection from TNF-{alpha}–induced apoptosis (data not shown). Therefore, the IL-4–induced survival of HMVEC-Ls may contribute to both the VEGF plus FGF2-mediated proliferation (Figure 1B) and the VEGF-mediated tube-forming activity (Figure 2).

Proliferation of the HMVEC-Ls was completely inhibited under Th1 inflammatory conditions, even in the presence of VEGF (Figure 1B). This observation is fully compatible with the pathological findings in COPD, a typical Th1 cytokine-mediated inflammatory disease of the lung, in which hypervascularity is unusual (5, 38). Proliferation of the HMVEC-Ls was also significantly inhibited in a dose-dependent manner under Th2 inflammatory conditions; however, the magnitude of inhibition was much less than that under Th1 inflammatory conditions (Figure 1B). Overproduction of CXCR3 chemokines (Figure 4B) as well as complete inhibition of cell proliferation (Figure 1B) may be involved in the pathogenesis of COPD through inhibition of angiogenesis, and they may facilitate hypoxia under Th1 inflammatory conditions (38). In addition, CXCR3 chemokines are highly induced in bronchial epithelial cells and other cells during bacterial and viral infections, which are well-known triggers of exacerbation of COPD (39). CXCR3 chemokines and/or their receptors were proposed as potential therapeutic targets for COPD because they recruit inflammatory cells into the lung (40), but our results indicate a new role for this system since it also inhibits angiogenesis in the lung.

In conclusion, our results clearly indicate that Th1/Th2 cytokines reciprocally regulate the tube-forming activity of human lung microvascular endothelial cells via induction of CXC chemokine synthesis and suggest that this system may be a good target for development of therapeutic interventions for severe asthma and COPD. Future studies need to examine whether or not CXCR2/CXCR3 chemokines play roles in the in vivo development and inhibition of angiogenesis in patients with asthma and COPD, respectively. It would also be interesting to test for correlations between the local production of these chemokines and the degree of hypervascularity or hypovascularity in the airways in these diseases.


   Acknowledgments
 
The authors thank Dr. Jun Abe of the Department of Allergy and Immunology, National Research Institute for Child Health and Development, and Dr. Koji Higashi and Dr. Ryota Ebata of the Department of Pediatrics, Chiba University School of Medicine, for their valuable suggestions.


   Footnotes
 
This work was supported in part by grants from the National Institute of Biomedical Innovation (ID05–24 and ID05–41) and a grant from the Japan Health Science Foundation (KH51046).

Originally Published in Press as DOI: 10.1165/rcmb.2007-0162OC on August 20, 2007

Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form May 8, 2007

Accepted in final form June 28, 2007


   References
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Kay AB. Allergy and allergic diseases: first of two parts. N Engl J Med 2001;344:30–37.[Free Full Text]
  2. Busse W, Elias J, Sheppard D, Banks-Schlegel S. Airway remodeling and repair. Am J Respir Crit Care Med 1999;160:1035–1042.[Free Full Text]
  3. Li X, Wilson JW. Increased vascularity of the bronchial mucosa in mild asthma. Am J Respir Crit Care Med 1997;156:229–233.[Abstract/Free Full Text]
  4. 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]
  5. Hashimoto M, Tanaka H, Abe S. Quantitative analysis of bronchial wall vascularity in the medium and small airways of patients with asthma and COPD. Chest 2005;127:965–972.[CrossRef][Medline]
  6. Vrugt B, Wilson S, Bron A, Holgate ST, Djukanovic R, Aalbers R. Bronchial angiogenesis in severe glucocorticoid-dependent asthma. Eur Respir J 2000;15:1014–1021.[Abstract]
  7. Chetta A, Zanini A, Foresi A, D'Ippolito R, Tipa A, Castagnaro A, Baraldo S, Neri M, Saetta M, Olivieri D. Vascular endothelial growth factor up-regulation and bronchial wall remodelling in asthma. Clin Exp Allergy 2005;35:1437–1442.[CrossRef][Medline]
  8. Hoshino M, Takahashi M, Aoike N. Expression of vascular endothelial growth factor, basic fibroblast growth factor, and angiogenin immunoreactivity in asthmatic airways and its relationship to angiogenesis. J Allergy Clin Immunol 2001;107:295–301.[CrossRef][Medline]
  9. Asai K, Kanazawa H, Otani K, Shiraishi S, Hirata K, Yoshikawa J. Imbalance between vascular endothelial growth factor and endostatin levels in induced sputum from asthmatic subjects. J Allergy Clin Immunol 2002;110:571–575.[CrossRef][Medline]
  10. Asai K, Kanazawa H, Kamoi H, Shiraishi S, Hirata K, Yoshikawa J. Increased levels of vascular endothelial growth factor in induced sputum in asthmatic patients. Clin Exp Allergy 2003;33:595–599.[CrossRef][Medline]
  11. Bandi N, Kompella UB. Budesonide reduces vascular endothelial growth factor secretion and expression in airway (Calu-1) and alveolar (A549) epithelial cells. Eur J Pharmacol 2001;425:109–116.[CrossRef][Medline]
  12. Wen FQ, Liu X, Manda W, Terasaki Y, Kobayashi T, Abe S, Fang Q, Ertl R, Manouilova L, Rennard SI. TH2 Cytokine-enhanced and TGF-beta-enhanced vascular endothelial growth factor production by cultured human airway smooth muscle cells is attenuated by IFN-gamma and corticosteroids. J Allergy Clin Immunol 2003;111:1307–1318.[CrossRef][Medline]
  13. Kazi AS, Lotfi S, Goncharova EA, Tliba O, Amrani Y, Krymskaya VP, Lazaar AL. Vascular endothelial growth factor-induced secretion of fibronectin is ERK dependent. Am J Physiol Lung Cell Mol Physiol 2004;286:L539–L545.[Abstract/Free Full Text]
  14. Lee CG, Link H, Baluk P, Homer RJ, Chapoval S, Bhandari V, Kang MJ, Cohn L, Kim YK, McDonald DM, et al. Vascular endothelial growth factor (VEGF) induces remodeling and enhances TH2-mediated sensitization and inflammation in the lung. Nat Med 2004;10:1095–1103.[CrossRef][Medline]
  15. Feltis BN, Wignarajah D, Reid DW, Ward C, Harding R, Walters EH. Effects of inhaled fluticasone on angiogenesis and vascular endothelial growth factor in asthma. Thorax 2007;62:314–319.[Abstract/Free Full Text]
  16. Tanaka H, Yamada G, Saikai T, Hashimoto M, Tanaka S, Suzuki K, Fujii M, Takahashi H, Abe S. Increased airway vascularity in newly diagnosed asthma using a high-magnification bronchovideoscope. Am J Respir Crit Care Med 2003;168:1495–1499.[Abstract/Free Full Text]
  17. Strieter RM, Polverini PJ, Kunkel SL, Arenberg DA, Burdick MD, Kasper J, Dzuiba J, Van Damme J, Walz A, Marriott D, et al. The functional role of the ELR motif in CXC chemokine-mediated angiogenesis. J Biol Chem 1995;270:27348–27357.[Abstract/Free Full Text]
  18. Angiolillo AL, Sgadari C, Taub DD, Liao F, Farber JM, Maheshwari S, Kleinman HK, Reaman GH, Tosato G. Human interferon-inducible protein 10 is a potent inhibitor of angiogenesis in vivo. J Exp Med 1995;182:155–162.[Abstract/Free Full Text]
  19. Keane MP, Arenberg DA, Moore BB, Addison CL, Strieter RM. CXC chemokines and angiogenesis/angiostasis. Proc Assoc Am Physicians 1998;110:288–296.[Medline]
  20. Strieter RM, Belperio JA, Burdick MD, Keane MP. CXC chemokines in angiogenesis relevant to chronic fibroproliferation. Curr Drug Targets Inflamm Allergy 2005;4:23–26.[CrossRef][Medline]
  21. Addison CL, Daniel TO, Burdick MD, Liu H, Ehlert JE, Xue YY, Buechi L, Walz A, Richmond A, Strieter RM. The CXC chemokine receptor 2, CXCR2, is the putative receptor for ELR+ CXC chemokine-induced angiogenic activity. J Immunol 2000;165:5269–5277.[Abstract/Free Full Text]
  22. Heidemann J, Ogawa H, Dwinell MB, Rafiee P, Maaser C, Gockel HR, Otterson MF, Ota DM, Lugering N, Domschke W, et al. Angiogenic effects of interleukin 8 (CXCL8) in human intestinal microvascular endothelial cells are mediated by CXCR2. J Biol Chem 2003;278:8508–8515.[Abstract/Free Full Text]
  23. Gibson PG, Simpson JL, Saltos N. Heterogeneity of airway inflammation in persistent asthma: evidence of neutrophilic inflammation and increased sputum interleukin-8. Chest 2001;119:1329–1336.[CrossRef][Medline]
  24. Jatakanon A, Uasuf C, Maziak W, Lim S, Chung KF, Barnes PJ. Neutrophilic inflammation in severe persistent asthma. Am J Respir Crit Care Med 1999;160:1532–1539.[Abstract/Free Full Text]
  25. Kato A, Homma T, Batchelor J, Hashimoto N, Imai S, Wakiguchi H, Saito H, Matsumoto K. Interferon-alpha/beta receptor-mediated selective induction of a gene cluster by CpG oligodeoxynucleotide 2006. BMC Immunol 2003;4:8.[CrossRef][Medline]
  26. Stimpfl M, Tong D, Fasching B, Schuster E, Obermair A, Leodolter S, Zeillinger R. Vascular endothelial growth factor splice variants and their prognostic value in breast and ovarian cancer. Clin Cancer Res 2002;8:2253–2259.[Abstract/Free Full Text]
  27. Lasagni L, Francalanci M, Annunziato F, Lazzeri E, Giannini S, Cosmi L, Sagrinati C, Mazzinghi B, Orlando C, Maggi E, et al. An alternatively spliced variant of CXCR3 mediates the inhibition of endothelial cell growth induced by IP-10, Mig, and I-TAC, and acts as functional receptor for platelet factor 4. J Exp Med 2003;197:1537–1549.[Abstract/Free Full Text]
  28. Catusse J, Liotard A, Loillier B, Pruneau D, Paquet JL. Characterization of the molecular interactions of interleukin-8 (CXCL8), growth related oncogen alpha (CXCL1) and a non-peptide antagonist (SB 225002) with the human CXCR2. Biochem Pharmacol 2003;65:813–821.[CrossRef][Medline]
  29. White JR, Lee JM, Young PR, Hertzberg RP, Jurewicz AJ, Chaikin MA, Widdowson K, Foley JJ, Martin LD, Griswold DE, et al. Identification of a potent, selective non-peptide CXCR2 antagonist that inhibits interleukin-8-induced neutrophil migration. J Biol Chem 1998;273:10095–10098.[Abstract/Free Full Text]
  30. Howarth PH, Babu KS, Arshad HS, Lau L, Buckley M, McConnell W, Beckett P, Al Ali M, Chauhan A, Wilson SJ, et al. Tumour necrosis factor (TNFalpha) as a novel therapeutic target in symptomatic corticosteroid dependent asthma. Thorax 2005;60:1012–1018.[Abstract/Free Full Text]
  31. Berry MA, Hargadon B, Shelley M, Parker D, Shaw DE, Green RH, Bradding P, Brightling CE, Wardlaw AJ, Pavord ID. Evidence of a role of tumor necrosis factor alpha in refractory asthma. N Engl J Med 2006;354:697–708.[Abstract/Free Full Text]
  32. Barnes PJ. New molecular targets for the treatment of neutrophilic diseases. J Allergy Clin Immunol 2007;119:1055–1062.[CrossRef][Medline]
  33. Risau W. Mechanisms of angiogenesis. Nature 1997;386:671–674.[CrossRef][Medline]
  34. Burdick MD, Murray LA, Keane MP, Xue YY, Zisman DA, Belperio JA, Strieter RM. CXCL11 attenuates bleomycin-induced pulmonary fibrosis via inhibition of vascular remodeling. Am J Respir Crit Care Med 2005;171:261–268.[Abstract/Free Full Text]
  35. Bonecchi R, Facchetti F, Dusi S, Luini W, Lissandrini D, Simmelink M, Locati M, Bernasconi S, Allavena P, Brandt E, et al. Induction of functional IL-8 receptors by IL-4 and IL-13 in human monocytes. J Immunol 2000;164:3862–3869.[Abstract/Free Full Text]
  36. Wu LW, Mayo LD, Dunbar JD, Kessler KM, Baerwald MR, Jaffe EA, Wang D, Warren RS, Donner DB. Utilization of distinct signaling pathways by receptors for vascular endothelial cell growth factor and other mitogens in the induction of endothelial cell proliferation. J Biol Chem 2000;275:5096–5103.[Abstract/Free Full Text]
  37. Shimizu M, Pelisek J, Nikol S. Vasculogenesis and angiogenesis depend on the developmental origin in the arterial tree. Curr Med Chem 2002;9:1619–1630.[Medline]
  38. Jeffery PK. Remodeling and inflammation of bronchi in asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2004;1:176–183.[Abstract/Free Full Text]
  39. Burge PS. Prevention of exacerbations: how are we doing and can we do better? Proc Am Thorac Soc 2006;3:257–261.[Abstract/Free Full Text]
  40. Panina P, Mariani M, D'Ambrosio D. Chemokine receptors in chronic obstructive pulmonary disease (COPD). Curr Drug Targets 2006;7:669–674.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2007-0162OCv1
38/2/168    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 Matsuda, A.
Right arrow Articles by Saito, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Matsuda, A.
Right arrow Articles by Saito, H.

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