The phosphatidylinositol 3-kinase (PI3K) signal transduction pathway is implicated in the airway remodeling associated with asthma. The class IA PI3K isoforms are known to be activated by growth factors and cytokines. Because this pathway is a possible site of pharmacological intervention for treating the disease, it is important to know which isoforms contribute to this process. Therefore, we used a pharmacological approach to investigate the roles of the three class IA PI3K isoforms (p110α, p110β, and p110δ) in airway remodeling using airway smooth muscle (ASM) cells derived from asthmatic subjects and ASM cells and lung fibroblasts from nonasthmatic subjects. These studies used the inhibitors N′-[(E)-(6-bromoimidazo[1,2-a]pyridin-3-yl)methylidene]-N,2-dimethyl-5-nitrobenzenesulfonohydrazide (PIK75) (which selectively inhibits p110α), 7-methyl-2-(4-morpholinyl)-9-[1-(phenylamino)ethyl]-4H-pyrido[1,2-a]pyrimidin-4-one (TGX221) (which selectively inhibits p110β), and 2-[(6-amino-9H-purin-9-yl)methyl]-5-methyl-3-(2-methylphenyl)-4(3H)-quinazolinone (IC87114) (which selectively inhibits p110δ). Cells were stimulated with transforming growth factor-β (TGFβ) and/or 10% fetal bovine serum in the presence or absence of inhibitor or vehicle control (dimethyl sulfoxide). PIK75, but not TGX221 or IC87114, attenuated TGFβ-induced fibronectin deposition in all cell types tested. PIK75 and TGX221 each decreased secretion of vascular endothelial growth factor and interleukin-6 in nonasthmatic ASM cells and lung fibroblasts, whereas TGX221 was not as effective in asthmatic ASM cells. In addition, PIK75 decreased cell survival in TGFβ-stimulated asthmatic, but not nonasthmatic, ASM cells. In conclusion, specific PI3K isoforms may play a role in pathophysiological events relevant to airway wall remodeling.
Asthma is characterized by airflow obstruction that may not be fully reversible because of chronic structural changes within the airway wall, termed remodeling. This remodeling includes an increase in airway smooth muscle (ASM) bulk, altered extracellular matrix (ECM) deposition, and angiogenesis. The events leading to the increased mass of ASM and the altered deposition of ECM proteins have been the focus of many studies. ASM cells in culture secrete ECM proteins, including collagen I and fibronectin (Johnson et al., 2006), both of which are known to be increased in the airway wall of asthmatic subjects (Roche et al., 1989). These ECM proteins can modulate cellular function, including proliferation (Hirst et al., 2000; Chan et al., 2006), and it is known that ASM cells derived from asthmatic subjects proliferate more than those from nonasthmatic subjects (Johnson et al., 2001; Trian et al., 2007). However, the precise mechanisms for this are not fully understood. We and others have previously reported significant differences in key processes such as proliferation, ECM deposition, cytokine and chemokine release, and mitochondrial biogenesis (Johnson et al., 2001, 2004; Chan et al., 2006; Trian et al., 2007) in structural cells derived from nonasthmatic and asthmatic subjects. The effects of inhibitors of signal transduction pathways differ in nonasthma and asthma-derived ASM cells, with both the extracellular-regulated kinase (ERK) and phosphatidylinositol 3-kinase (PI3K) pathways playing equal roles in the proliferation of nonasthmatic ASM cells, but the PI3K pathway having a more dominant role in asthmatic cells (Burgess et al., 2008). This suggests the PI3K pathway may be a potential target for treating the ASM bulk in asthma. Furthermore, transforming growth factor-β (TGFβ), which is elevated in the bronchoalveolar lavage fluid of asthmatic subjects (Redington et al., 1997; Vignola et al., 1997), can induce the deposition of the ECM proteins collagen type I and fibronectin from ASM cells (Johnson et al., 2006; Moir et al., 2008). This deposition is regulated through several signaling pathways, including ERK and PI3K, with PI3K playing a more prominent role in asthmatic ASM cells, again suggesting there may be differential up-regulation of PI3K in asthma (Johnson et al., 2006). The PI3K signaling pathway is also involved in cytokine and growth factor release; for example, inhibition of PI3K decreases the release of vascular endothelial growth factor (VEGF), a potent regulator of angiogenesis, from ASM cells (Shin et al., 2009). Thus inhibition of PI3K might offer a site for therapeutic intervention in asthma.
PI3Ks constitute a family of intracellular signaling molecules that regulate many cellular processes including survival, differentiation, and proliferation (Shepherd et al., 1998). They are divided into three classes (I, II, and III) based on their structural characteristics and substrate specificity. The class I PI3Ks are the most commonly studied of the three classes. They are heterodimers consisting of a catalytic subunit and a regulatory subunit and can be subdivided into class IA and class IB. Class IA consists of a p110α, p110β, or p110δ catalytic subunit and a regulatory subunit (p85α, p55α, or p50α) and is activated by both receptor tyrosine kinases and G protein-coupled receptors. Class IB consists of the catalytic subunit p110γ and the regulatory subunit p101 and is regulated by G protein-coupled receptors. Little is currently known about the role of the specific isoforms in airway disease. A range of well characterized isoform-selective PI3K inhibitors has been developed (Chaussade et al., 2007). These inhibitors have been used to investigate the role of PI3K isoforms in glucose metabolism (Knight et al., 2006), cytokine release (Dagia et al., 2010), fat cell differentiation (Kim et al., 2009), bone function (Grey et al., 2010), and appetite regulation (Tups et al., 2010). The conclusion of these studies is that the involvement of different PI3K isoforms, in particular cellular responses, can vary greatly between cell types and needs to be assessed on a case-by-case basis. A number of isoform-selective inhibitors of PI3K are in clinical development, and it is possible these could have utility in treating asthma (Knight et al., 2006; Rückle et al., 2006; Wong et al., 2010). Therefore, we have used these well characterized reagents to investigate the role of the different PI3K isoforms in fibronectin deposition, cell survival, and cytokine secretion in ASM and lung fibroblast cells and to compare cells from nonasthmatic and asthmatic subjects.
Materials and Methods
Approval for the harvesting of primary pulmonary cells from human lung specimens was provided by the human ethics committees of The University of Sydney and the Sydney South West Area Health Service. All subjects provided written informed consent. Human airway cells were obtained from nonasthmatic (57.4 ± 2.8 years) and asthmatic (43.5 ± 47.9 years) subjects undergoing lung resection for either lung transplant or carcinoma or from bronchial biopsy samples (Table 1). ASM cells were dissected free from the surrounding tissue and grown as explants as described previously (Johnson et al., 2001). ASM cell characterization was confirmed by light microscopy and immunohistochemistry for smooth muscle α-actin and calponin expression. Lung fibroblasts were obtained from finely diced lung parenchyma. Cells were used at passages 3 to 8, and for all experiments they were seeded at a density of 104 cells/cm2.
The PI3K inhibitors LY294002 (2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one) (pan inhibitor; Calbiochem, San Diego, CA), PIK75 (N′-[(E)-(6-bromoimidazo[1,2-a]pyridin-3-yl)methylidene]-N,2-dimethyl-5-nitrobenzenesulfonohydrazide), TGX221 (7-methyl-2-(4-morpholinyl)-9-[1-(phenylamino)ethyl]-4H-pyrido[1,2-a]pyrimidin-4-one), and IC87114 (2-[(6-amino-9H-purin-9-yl)methyl]-5-methyl-3-(2-methylphenyl)-4(3H)-quinazolinone) (p110α, p110β, and p110δ, respectively; Symansis, Auckland, New Zealand) were dissolved in DMSO and diluted in cell culture medium before use. LY294002 was used in the range of 1 to 10 μM. Isoform-specific inhibitors PIK75, TGX221, and IC87114 were used in the range of 0.01 to 1 μM. The reported IC50 values for the specific target isoforms are 0.0078 μM for PIK75 for p110α, 0.0085 μM for TGX221 for p110β, and 0.062 μM for IC87114 for p110δ (Chaussade et al., 2007). The IC50 of these for nontarget isoforms is approximately 100-fold higher, indicating there is a significant range of concentrations of up to 1 μM over which these inhibitors will be selective for the target isoform. In all experiments using the PI3K inhibitors, cells were preincubated with inhibitor or vehicle (DMSO) for 30 min before stimulation.
Cells cultured in the presence or absence of TGFβ (1 ng/ml) for 15 min or 48 h were lysed [200 mM Tris-HCl, pH 7.4, 3% SDS, 2 mM EDTA, 10% glycerol, 200 μM sodium orthovanadate, 2 mM phenylmethylsulfonyl fluoride, and 10% (v/v) protease inhibitors (Protease inhibitor cocktail set III; Calbiochem)]. Protein was separated by SDS/polyacrylamide gel electrophoresis on 10% gels and electroblotted onto polyvinylidene fluoride membrane (Millipore Corporation, Billerica, MA). The membrane was incubated with rabbit anti-p110α (BD Biosciences Transduction Laboratories, Lexington, KY), rabbit anti-p110β (Cell Signaling Technology, Danvers, MA), rabbit anti-p110δ (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), rabbit anti-phospho-AKT, rabbit anti-AKT, rabbit anti-phospho-p70 S6K (threonine 389), rabbit anti-p70 S6K, rabbit anti-phospho-p44/p42, or rabbit anti-p44/p42 (Cell Signaling Technology) antibodies overnight at 4°C or anti-mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Santa Cruz Biotechnology Inc.) for 2 h at room temperature. After washing with Tris-buffered saline-containing Tween 20 (0.05%) bound antibody was visualized, using horseradish peroxidase-conjugated goat anti-rabbit IgG or horseradish peroxidase-conjugated anti-mouse IgG antibody (Santa Cruz Biotechnology Inc.) and enhanced chemiluminescence, and imaged (Image Station 4000MM; Kodak Digital Science, New Haven, CT).
Mitochondrial activity was assessed after stimulation with TGFβ with or without inhibitors for 48 h using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) assay as described previously (Moir et al., 2008). Absorbance was measured at a test wavelength of 570 nm and a reference wavelength of 690 nm. For each primary cell culture, results from three to six wells from each treatment were averaged, and data were expressed as absorbance 570 to 690 nm.
Fibronectin deposition from ASM cells and fibroblasts was measured using an ECM ELISA as described previously (Johnson et al., 2000, 2006). In brief, cells were seeded in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum for 24 h, quiesced in Dulbecco's modified Eagle's medium containing 0.1% bovine serum albumin for 24 h, and then stimulated with or without TGFβ (1 ng/ml) ± inhibitors for another 48 h. ECM free of cells was prepared by treatment with hypotonic ammonium hydroxide, and fibronectin was measured. For each primary cell culture, results from triplicate wells from each treatment were averaged and the data were expressed as absorbance at 450 nm.
VEGF and IL-6 levels in ASM and fibroblast cell supernatants collected after a 48-h incubation with 10% fetal bovine serum containing TGFβ (1 ng/ml) ± inhibitors or vehicle (DMSO) control were measured using specific ELISAs according to the manufacturer's instructions (R&D Systems, Minneapolis, MN). The detection limit of these assays was 15.6 pg/ml.
Data were expressed as mean ± S.E.M. obtained from cells cultured from n subjects and analyzed using Student's paired t test or repeated measures one-way analysis of variance with Bonferroni post hoc correction where appropriate. A probability (p) value of less than or equal to 0.05 was considered significant.
Airway Smooth Muscle Cells and Lung Fibroblasts Express PI3K Isoforms.
We first needed to investigate which isoforms of class IA PI3 kinase were expressed in our primary cell cultures. Western blotting studies in ASM cells from nonasthmatic and asthmatic subjects and nonasthmatic-derived fibroblasts showed that all of these express p110α, p110β, and p110δ in the presence and absence of TGFβ (1 ng/ml) stimulation (Fig. 1).
TGFβ Increased Phosphorylation of p70 S6K in Human Airway Smooth Muscle Cells and Fibroblasts.
Stimulation of ASM cells with TGFβ (1 ng/ml) increased the phosphorylation of p70S6K, a downstream target in the PI3K pathway (Fig. 2, A–C). Data showing PIK75, TGX221, and IC87114 reduced phosphorylation of both AKT and p70S6K, but not p44/p42 (ERK), is shown in Fig. 2, D to F. In fact, PIK75 increased the TGFβ-induced phosphorylation of p44/p42 at 15 min.
TGFβ-Induced Fibronectin Deposition.
Treatment of nonasthmatic and asthmatic ASM cells and nonasthmatic lung fibroblasts for 48 h with the profibrotic stimulus TGFβ (1 ng/ml) induced fibronectin deposition (Fig. 3). To clarify the role of PI3K and determine which isoforms are involved in TGFβ-induced fibronectin deposition we used the pan inhibitor LY294002 (1–10 μM) and the inhibitors of individual PI3K isoforms at concentrations at which they are isoform-specific (Chaussade et al., 2007). These were the p110α inhibitor PIK75 (0.01–1 μM), the p110β inhibitor TGX221 (0.01–1 μM), and the p110δ inhibitor IC87114 (0.01–1 μM). TGFβ-induced fibronectin deposition was significantly inhibited in both nonasthmatic and asthmatic ASM cells and lung fibroblasts by LY294002 (10 μM; nonasthmatic ASM, 69.5% reduction; asthmatic ASM, 66.9% reduction; fibroblasts, 44.1% reduction; p < 0.05; Fig. 3, A–C). PIK75 decreased fibronectin deposition from both nonasthmatic ASM cells (n = 4, p < 0.05; Fig. 3D) and asthmatic ASM cells (n = 4, p < 0.05; Fig. 3E) and fibroblasts (PIK75, 1 μM, n = 4, p < 0.05; Fig. 3F). TGX221 (0.01–1 μM; n = 3, nonasthmatic ASM; n = 4, asthmatic ASM; n = 7, fibroblasts) and IC87114 (0.01–1 μM; n = 5, nonasthmatic ASM; n = 3, asthmatic ASM; n = 5, fibroblasts) had no effect on TGFβ-induced fibronectin deposition from ASM cells or fibroblasts (data not shown).
LY294002 (10 μM) had no significant effect on nonasthmatic ASM cells (Fig. 4A); however, it caused a small, but significant, decrease in mitochondrial activity as measured by the MTT assay in TGFβ-stimulated asthmatic ASM cells (11.3 ± 2.6% reduction, n = 6; p < 0.05) (Fig. 4B) and unstimulated and TGFβ-stimulated lung fibroblasts (unstimulated, 21.4 ± 5.0% reduction; TGFβ stimulated, 20.0 ± 2.5% reduction; n = 5; p < 0.05) (Fig. 4C). In nonasthma-derived and asthma-derived ASM cells, and lung fibroblasts, TGX221 and IC87114 had no significant effect on mitochondrial activity in the presence or absence of TGFβ (data not shown). However, 1 μM PIK75 significantly decreased mitochondrial activity in unstimulated nonasthmatic ASM cells, asthmatic ASM cells, and lung fibroblasts (p < 0.05; Fig. 4, D–F). Furthermore, PIK75 decreased mitochondrial activity in TGFβ-stimulated asthmatic cells, whereas in nonasthmatic cells it had no effect (Fig. 4, D and E). At these concentrations PIK75 specifically targeted p110α. In fibroblasts PIK75 decreased mitochondrial activity only at the highest concentration (1 μM; Fig. 4F). We confirmed in an independent series of experiments, our previous finding (Moir et al., 2008) that stimulation with TGFβ (1 ng/ml) has no significant effect on the mitochondrial activity of ASM cells (Fig. 4, A, B, D, and E; p > 0.05) or on lung fibroblasts (Fig. 4, C and F).
PIK75 and TGX221 significantly reduced TGFβ-induced VEGF in nonasthmatic ASM cells (n = 11) and fibroblasts (n = 8). Results for nonasthmatic ASM cells are shown in Fig. 5A. In fibroblasts, PIK75 at 0.1 and 1 μM produced 63.0 and 83.5% reductions, respectively (p < 0.05; data not shown), and TGX221 at 1 μM produced a 64.9% reduction (p < 0.05; data not shown). However, in asthmatic cells PIK75 at 0.1 and 1 μM reduced TGFβ-induced VEGF 43.9 and 53.8%, respectively (p < 0.05; n = 4; Fig. 5B), and TGX221 produced no change (Fig. 5B). IC87114 had no effect on ASM cells (Fig. 5, A and B) or fibroblasts (data not shown). LY294002 had no effect on VEGF secretion from nonasthmatic cells (13,629 ± 1458 pg/ml TGFβ + LY294002 (10 μM) versus 14,765 ± 1568 pg/ml TGFβ + DMSO; n = 11) or asthmatic cells (16,651 ± 4046 pg/ml TGFβ + LY294002 (10 μM) versus 17,411 ± 4357 pg/ml TGFβ + DMSO; n = 4).
PIK75 and TGX221 reduced TGFβ-induced IL-6 in nonasthmatic ASM (n = 9; p < 0.05; Fig. 5C). However, in asthmatic cells whereas PIK75 reduced IL-6 (n = 4; p < 0.05; Fig. 5D), TGX221 produced little change (Fig. 5D). IC87114 had no effect on both types of ASM cells (Fig. 5, C and D); however, it did reduce IL-6 secretion by fibroblasts (IC87114 0.1 and 1 μM; n = 5; 48.8 and 54.4% reduction, respectively; p < 0.05; data not shown). LY294002 had no effect on IL-6 secretion from nonasthmatic cells (12,718 ± 2266 pg/ml TGFβ + LY294002 (10 μM) versus 13,534 ± 2391 pg/ml TGFβ + DMSO; n = 11) or asthmatic cells (8777 ± 883 pg/ml TGFβ + LY294002 (10 μM) versus 10,690 ± 1106 pg/ml TGFβ + DMSO; n = 4).
A summary of the results is provided in Table 2.
The advent of the pan PI3K inhibitors wortmannin and LY294002 in the 1990s provided very powerful tools to increase our understanding of PI3K signaling in vitro (Arcado and Wymann, 1993; Vlahos et al., 1994). Considerable evidence now suggests this pathway plays an important role in the structural alterations associated with airway remodeling in asthma. For example, PI3K mediates mitogen-induced proliferation of ASM cells (Scott et al., 1996; Krymskaya et al., 1999) and attenuates subepithelial fibrosis in a murine model of allergic airway disease (Lee et al., 2008). Furthermore, we have previously reported a more prominent role for PI3K in TGFβ-induced collagen and fibronectin in ASM cells derived from asthmatic subjects compared with nonasthma-derived cells (Johnson et al., 2006). However, the role of the individual PI3K isoforms in these processes had not been investigated.
Expression of the class I PI3K isoforms p110α and p110β is widespread, so it is not surprising that ASM cells derived from nonasthmatic subjects express both of these isoforms (Krymskaya et al., 1999). In the present study we confirmed this finding and extended it to asthmatic ASM cells and nonasthmatic lung fibroblasts. We were surprised to find that we demonstrated the expression of p110δ in these cell types, whereas others have reported it to be leukocyte-specific (Vanhaesebroeck et al., 1997).
There have been attempts to understand the role of different PI3K isoforms in animals by using transgenic gene knockout approaches, but it has been difficult to learn much from these studies because homozygous deletion of either p110α (Bi et al., 1999) or p110β (Bi et al., 2002) in a mouse model is embryonic lethal, whereas a p110δ knockout is viable and has no major defects (Okkenhaug et al., 2002). In contrast, the availability of isoform-selective small-molecule inhibitors of PI3K has allowed rapid progress in identifying roles played by different isoforms (Ihle and Powis, 2010).
The structural alterations that constitute remodeling are refractory to current asthma therapies. TGFβ, which is increased in asthma, can induce the deposition of the ECM proteins collagen and fibronectin from ASM cells in vitro. This deposition is mediated through several signaling pathways including PI3K (Johnson et al., 2006; Moir et al., 2008). In the current study we confirmed our previous finding that TGFβ induces phosphorylation of Akt in human ASM cells (Johnson et al., 2006) and demonstrated that in ASM cells derived from asthmatic and nonasthmatic subjects TGFβ induces phosphorylation of p70S6K, a downstream signaling molecule in the PI3K pathway, thus supporting the observation that TGFβ activates the PI3K pathway. This is in agreement with Deng et al. (2010) who also reported TGFβ increased phosphorylation of p70S6K in human ASM cells from lung donor tissue. Furthermore, the elevated, although not significant, phosphorylation of AKT and p70S6K by TGFβ at 15 min was reduced in the presence of the specific inhibitors, suggesting that constitutive levels of PI3K contribute to the cellular functions. In addition, TGFβ induced phosphorylation of p44/p42 (ERK1/2), which was further increased in the presence of PIK75. This increase in phosphorylation of p44/42 after treatment with PIK75 is difficult to explain but may suggest cross-talk between the PI3K and ERK pathways (Dai et al., 2009).
We observed a reduction in fibronectin deposition in the presence of PIK75, which may reflect an inability of the cells to produce fibronectin, an ECM protein that has been shown to support ASM cell survival. PIK75 also reduced TGFβ-stimulated VEGF and IL-6 secretion in all cell types tested. This is consistent with findings in human peripheral blood monocytes (Dagia et al., 2010) and could suggest p110α activity is necessary for the production of proinflammatory mediators. However, we also showed that PIK75 significantly reduced the conversion of MTT to formazan in asthmatic and nonasthmatic ASM cells and lung fibroblasts, thus suggesting these cells require p110α for their survival. It has been reported in other cell types that at low concentrations PIK75 is specific for p110α and does not have off-target effects but that at concentrations >1 μM it may inhibit the mammalian target of rapamycin (Knight et al., 2006). Therefore, the differences we observed in this study, after p110α inhibition, may be caused by a reduction in cell number. This highlights the need for isoform-specific inhibitors to target particular cellular functions, because broad-spectrum inhibitors may have detrimental side effects.
Cytokines and growth factors such as VEGF and IL-6, which are secreted by structural cells within the lung, contribute to tissue remodeling via induction of cell proliferation and angiogenesis. TGFβ induces the release of VEGF and IL-6, and PI3K regulates this release (Burgess et al., 2006; Shin et al., 2009). Our study demonstrates that the presence of disease confers a differential effect on p110 isoform inhibition. TGX221 reduced TGFβ stimulated VEGF secretion, although only in ASM cells derived from nonasthmatic subjects. Likewise, TGX221 significantly reduced IL-6 in nonasthmatic ASM cells but had little effect on asthmatic cells. This provides evidence for a previously unsuspected role for p110β in cytokine production and/or secretion. Treatment of cells with the p110δ inhibitor IC87114 had no effect on cell survival, fibronectin deposition, VEGF, or IL-6 secretion, thus supporting the concept that the individual isoforms play specific roles in cell function.
Overall, we demonstrate differences in PI3K signaling in asthmatic ASM cells compared with nonasthmatic ASM cells and lung fibroblasts. Furthermore, these results show that the specific isoforms are involved in particular cellular functions. Because the structural changes in asthmatic airways that constitute airway remodeling can be resistant to current therapy it may be that these features can be modulated with novel drugs that specifically target PI3K isoforms.
Participated in research design: Moir, Trian, Ge, Burgess, Oliver, and Black.
Conducted experiments: Moir and Trian.
Contributed new reagents or analytic tools: Shepherd.
Performed data analysis: Moir, Trian, and Black.
Wrote or contributed to the writing of the manuscript: Moir, Trian, Ge, Shepherd, Burgess, Oliver, and Black.
Other: Burgess, Oliver, and Black acquired funding for the research.
We thank the cardiopulmonary transplant team and pathologists at St. Vincent's Hospital, Sydney, Australia and the thoracic physicians and pathologists at Royal Prince Alfred, Concord, and Strathfield Private hospitals and Davies Campbell and De Lambert Pathology, Rhodes, Sydney, Australia; Dr. Greg King and Dr. Melissa Baraket at the Woolcock Institute of Medical Research and Dr. Edmund Lau and Dr. Rebecca Pearson at the Royal Prince Alfred Hospital for supplying the asthmatic biopsy samples; and Dr. Claire Chaussade for technical advice.
This work was supported by the National Health and Medical Research Council of Australia [Grant 512301]. J.K.B. is supported by a National Health and Medical Research Council R. Douglas Wright Fellowship [Fellowship 402835]. J.L.B. is supported by a National Health and Medical Research Council Senior Principal Research Fellowship [Fellowship 571098].
Conflict of interest: P.R.S. is a founder of Symansis Ltd.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- airway smooth muscle
- phosphatidylinositol 3-kinase
- extracellular matrix
- transforming growth factor-β
- extracellular-regulated kinase
- vascular endothelial growth factor
- chronic obstructive pulmonary disease
- glyceraldehyde-3-phosphate dehydrogenase
- enzyme-linked immunosorbent assay
- Received July 30, 2010.
- Accepted February 23, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics