The roles of transforming growth factor-β1 and vascular endothelial growth factor in the tracheal granulation formation

Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

Contents lists available at ScienceDirect

Pulmonary Pharmacology & Therapeutics journal homepage: www.elsevier.com/locate/ypupt

The roles of transforming growth factor-b1 and vascular endothelial growth factor in the tracheal granulation formation Yu-Chin Lee a, d,1, Ming-Hui Hung a, Lung-Yu Liu a, Kuo-Ting Chang a, Tec-Ying Chou c, d, Yi-Chung Wang c, Yu-Chung Wu b, d, Chun-Liang Lai e, f, Cheng-Chien Tsai a, Kang-Cheng Su a, Diahn-Warng Perng a, d, *,1 a

Department of Chest Medicine, Taipei Veterans General Hospital, Taipei, Taiwan Division of Chest Surgery, Taipei Veterans General Hospital, Taipei, Taiwan Department of Pathology, Taipei Veterans General Hospital, Taipei, Taiwan d School of Medicine, National Yang-Ming University, Taipei, Taiwan e Section of Chest and Critical Care, Department of Internal Medicine, Buddhist Dalin Tzu Chi General Hospital, Chiayi, Taiwan f School of Medicine, Buddhist Tzu Chi University, Hualien, Taiwan b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 March 2010 Received in revised form 27 October 2010 Accepted 29 October 2010

Background: Acquired tracheal stenosis is common in patients with a long-term tracheostomy and granulation is one of the most commonly observed lesions in benign airway stenosis. The aim of this study was to investigate the mechanisms of tracheal granulation formation and find the potential therapeutic targets to prevent the granulation formation. Results: In granulation tissue obtained from patients during interventional bronchoscopy for the relief of airway obstruction, increased expression of transforming growth factor (TGF)-b1 and vascular endothelial growth factor (VEGF), as well as increased numbers of fibroblasts, was found by immunohistochemical staining. TGF-b1 expression was detected in both the epithelial and submucosal layers. The highest levels of VEGF and vimentin expression occurred in the submucosal layers. In comparison with the control, significantly increased numbers of small vessels were observed in the submucosal layers of the granulation tissue. In vitro, TGF-b1 stimulated production of VEGF by cultured fibroblasts at both the mRNA and protein level. VEGF siRNA treatment resulted in a significant decrease of TGF-b1-induced VEGF production. SIS3, a selective Smad3 inhibitor, and UO126 both inhibited p44/42 MAP kinase phosphorylation and attenuated subsequent VEGF production by fibroblasts. A low concentration of erythromycin (1 mg/ml), but not dexamethasone (100 mM), inhibited TGF-b1-induced VEGF production. Conclusion: This study provides important information that facilitates an understanding, at least in part, of the mechanisms of granulation formation. Targeting these mediators and cells may help to prevent the formation of granulation tissue in long-term tracheostomy or prolonged endotracheal intubation patients. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Airway stenosis Fibroblast TGF-b1 Tracheal granulation VEGF

1. Introduction Acquired tracheal stenosis is common in patients with a longterm tracheostomy. Granulation (tissue) is one of the most commonly

Abbreviations: GAPDH, Glyceraldehyde-3-phosphate dehydrogenase; FGF, Fibroblast growth factor; PDGF, Platelet-derived growth factor; QPCR, Real-time quantitative polymerase chain reaction; RT-PCR, Reverse transcription-polymerase chain reaction; siRNA, Small interference RNA; TGF, Transforming growth factor; VEGF, Vascular endothelial growth factor. * Corresponding author. School of Medicine, National Yang-Ming University, Department of Chest Medicine, Taipei Veterans General Hospital, 201, Section 2, Shih-Pai Road, Taipei 11217, Taiwan, Tel.: þ886 2 28712121x3194; fax: þ886 2 28761009. E-mail address: [email protected] (D.-W. Perng). 1 Dr Lee and Dr Perng made equal contributions to this study. 1094-5539/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.pupt.2010.10.016

observed lesions in obstructive airways [1]. Microscopic examination of granulation tissue reveals an increase in angiogenesis and deposition of extracellular matrix [2,3]; however, the mechanism for the development of granulation tissue in patients with prolonged tracheotomy or endotracheal intubation needs to be elucidated. Mechanical force may play a critical role in tissue remodeling. Apical compression of cultured bronchial epithelial cells can trigger the release of transforming growth factor (TGF)-b and endothelin [4], and in response to compressive stress, the tracheal epithelium can induce TGF-b1 production [5]. Cyclic stretching can upregulate IL-8 and TGF-b1 expression in alveolar cells [6]. Mechanical force may contribute to the development of granulation tissue through some tissue-remodeling mediators. Increased expression of TGF-b1 in the airway epithelium has been associated with the formation of stent-related airway stenosis

24

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

[7]. TGF-b1 plays an important role in tissue repair and remodeling [8,9], and can mediate these responses by inducing the release of a variety of cytokines, including vascular endothelial growth factor (VEGF) [10]. Enhanced VEGF expression has been observed in some exuberant tracheal granulation tissue in children [3]. VEGF is a potent mediator of angiogenesis, which has potential effects on lung development and in the pathogenesis of some chronic airway disorders [11]. This study attempted to investigate the mechanism of tracheal granulation formation in patients with prolonged tracheotomy or endotracheal intubation. We examined the expression of TGF-b1 and VEGF by immunohistochemistry in the exuberant tracheal granulation tissue obtained from patients with benign tracheal stenosis, and TGF-b1 was applied to cultured human lung fibroblasts to induce VEGF production. We hypothesized that TGF-b1, released from epithelial cells, may stimulate fibroblasts to produce VEGF, which in turn induces angiogenesis and contributes to the formation of granulation tissue. 2. Materials and methods We intended to compare the expression of TGF-b1 and VEGF in tracheal granulation tissues with those in normal tracheobronchial tissues, and determined the production of VEGF from fibroblasts after stimulation by TGF-b1 and the possible mechanisms involved. 2.1. Subjects Twelve patients with benign airway stenosis were recruited to this study. Granulation tissue was obtained from patients with tracheal stenosis during interventional bronchoscopy for the relief of upper airway obstruction. These granulation tissues were associated with long-term tracheostomy or prolonged endotracheal intubation. Normal tracheobronchial tissues, obtained from patients who had undergone surgical resection because of malignancy, were used as a control. Written informed consent was obtained from all subjects prior to the commencement of the study. 2.2. Immunohistochemistry Immunohistochemical analysis was used to detect TGF-b1, VEGF and vimentin expression in granulation tissue. Samples were fixed in formalin, embedded in paraffin blocks and cut into 3-mm-thick sections. Briefly, the sections were then deparaffined, hydrated, and incubated in 3% H2O2 blocking solution for 5 min. Epitopes were retrieved by microwaving at 95e100  C with citrate buffer (pH 6.0) for 15 min. Sections were incubated with mouse anti-VEGF antibody (Zymed Laboratories; San Francisco, USA) at a dilution of 1:75, anti-TGF-b1 (Acris Antibodies, Hiddenhausen, Germany) at a dilution of 1:100, or anti-vimentin antibody (Dako, Glostrup, Denmark) at a dilution of 1:600. Biotinylated horse anti-mouse IgG (Novocastra Laboratories, Newcastle upon Tyne, UK) was used as a secondary antibody and applied for 25 min at room temperature. Color was developed using DAB (3,30 -diaminobenzidine tetrahydrochloride) peroxidase substrate for 10 min. Microscopic fields with the positive immunoreactivity were chosen for analysis and reviewed by two senior pathologists. At least 1000 cells were analyzed in each case A numeric intensity score was set from 0 to 3 (0, no staining; 1þ, weak staining; 2þ, moderate staining; 3þ, strong staining)[3,7]. 2.3. Human lung fibroblast culture Normal human lung parenchyma, obtained by surgical lobectomy for lung cancer, was rinsed several times with Leibovitz’s L-15

Table 1A Demographic data of patients with benign airway stenosis. Patient Age, Sex, Diagnosis number years female/ male, F/M 1 2 3 4 5 6 7 8 9 10 11 12

50 29 92 51 79 74 72 79 78 79 56 52

F M M M F F M F F M M M

Duration of intubation/ tracheostomy (months)

Tracheostomy-related stenosis Tracheostomy-related stenosis Tracheostomy-related stenosis Postintubation-related stenosis Tracheostomy-related stenosis Tracheostomy-related stenosis Postintubation-related stenosis Tracheostomy-related stenosis Tracheostomy-related stenosis Postintubation-related stenosis Tracheostomy-related stenosis Foreign body-associated granulation

123 3.5 28 1.5 144 3 11.5 2.5 2 4.5 4.8 25

medium containing penicillin (100 U/ml), streptomycin (100 mg/ml), and amphotericin B (0.25 mg/ml). The tissue was cut into 1- to 2-mm2 pieces, and 3e4 pieces of tissue were planted onto six-well culture plates. The culture medium consisted of antibiotics/antimycotics, glutamine (2 mM), and 10% FBS in DMEM. The purity of the fibroblasts appeared to be more than 98%, as determined by morphology and by immunocytochemistry with antibodies against vimentin for fibroblasts, myosin for smooth muscle cells and cytokeratin for epithelial cells. Cells were detached from the plates by trypsinization and seeded onto 24-well culture plates for the following studies. 2.4. Assay of VEGF production The levels of VEGF released from lung fibroblasts after TGF-b1 stimulation were measured. The concentration of VEGF in the culture supernatants was assayed by means of a colorimetric enzyme-linked immunosorbent assay according to the manufacturer’s instructions (R&D Systems, Abingdon, UK). To suppress the effect of VEGF release induced by TGF-b1, cells were treated with dexamethasone (1e100 mM), erythromycin (1e100 mg/ml), UO126 10 mM (1,4-diamino-2,3-dicyano-1,4-bis (2-aminophenylthio) butadiene, a selective p44/42 inhibitor from Promega, USA) or SIS3 (a novel specific Smad3 inhibitor, 6,7-dimethoxy-2-((2E)-3-(1methyl-2- phenyl-1H-pyrrolo[2,3-b]pyridin-3-yl-prop-2-enoyl))1,2,3,4-tetrahydroisoquinoline, 1e100 mM, Calbiochem, Germany) [12]. Supernatants were collected at each time point and stored at 80  C prior to assessment for mediators. Cell viability was determined by light microscopy and dye exclusion with trypan blue. Table 1B Demographic data of patients who had undergone surgical resection because of malignancy. Normal tracheobronchial tissues were used as a control. Patient Age, Sex, Diagnosis number years female/ male, F/M

a

1 2 3

64 70 80

M M M

LLL RUL RML

*pT2N0M0 stage IB pT2N2M0 stage IIIA pT1N1M0 stage IIA

4 5

80 52

M F

RUL RLL

pT4N0M1 pT4N2M0

stage IV stage IIIB

6 7 8 9

56 79 69 55

F M F M

RML RUL LUL LUL

pT1N0M0 pT2N0M0 pT2N0M0 pT1N0M0

stage stage stage stage

Adenocarcinoma Adenocarcinoma Squamous cell carcinoma Adenocarcinoma Poorly-differentiated carcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma

Site Stage

IA IB IB IA

a Site; LUL: left upper lobe. LLL: left lower lobe. RUL: right upper lobe. RML: right middle lobe. RLL: right lower lobe. *p indicates the pathological stage.

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

25

Fig. 1. Expression of transforming growth factor (TGF)-b1, vascular endothelial growth factor (VEGF) and vimentin in granulation tissue. Granulation tissue was comprised of mucosal and submucosal layers. Anti-mouse IgG antibody was used as a negative control ( 400) (A). Increased expression of TGF-b1 was detected in both the epithelial and submucosal layers ( 400) (B). The highest expression levels of VEGF ( 400) (C) and vimentin ( 400) (D) were detected in the submucosal layers. VEGF expression was observed in the cytoplasm of infiltrating cells in the submucosa and endothelial cells.

2.5. Semiquantative reverse transcription-polymerase chain reaction (RT-PCR) and real-time quantitative polymerase chain reaction (QPCR) for VEGF mRNA expression Following the removal of supernatants for mediator detection, total cellular RNA was isolated from fibroblasts using a High Pure RNA Isolation Kit (Roche Molecular Biochemicals, Mannheim, Germany). The RNA (1 mg) was reverse transcribed into cDNA using Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega, Madison, WI, USA) and oligo d(T) as a primer at 42  C for 60 min followed by heating at 94  C for 3 min. An aliquot of cDNA was then subjected to 35 cycles of PCR for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using a standard procedure: denaturing at 94  C for 2 min, hybridizing at 55  C for 30 s, and elongating at 72  C for 1.5 min. Another aliquot of cDNA was subjected to 35

Table 2 Immunostaining of VEGF, vimentin and TGF-b1 in granulation tissue. Control subjects (*n ¼ 9) a

Staining VEGF mucosa submucosa Vimentin mucosa submucosa TGF-b1 mucosa submucosa

Granulation (n ¼ 12)

0

1þ

2þ

3þ

0

1þ

2þ

3þ

9 8

0 1

0 0

0 0

9 1

3 4

0 2

0 5

9 9

0 0

0 0

0 0

12 0

0 1

0 6

0 5

9 8

0 1

0 0

0 0

4 3

3 6

3 1

2 2

*n indicates the number of patients enrolled. a Staining; 0, no staining; 1þ, weak staining; 2þ, moderate staining; 3þ, strong staining.

cycles of PCR for VEGF using a standard procedure: denaturing at 95  C for 2 min, annealing at 56  C for 40 s, and elongating at 72  C for 5 min. The sense primer was 50 -CGAAGT GGTGAAGTTCATGGATG-30 ; the anti-sense primer 30 -TTCTGTATCAGTCTT TCCTGGTGAG-50 . RTPCR of mRNA encoding the 121-, 165- and 189-amino acid isoforms of VEGF resulted in PCR products of 403, 535 and 607 bp, respectively. The constitutively expressed gene, GAPDH, was used as an internal control. The primers for GAPDH were 50 primer ATCAAGAAGGTGGTGAAGCAGG and 30 primer GCAACTGTGAGGAGGGGAGATT, generating a 385-bp PCR product. The amplified products were electrophoresed in a 2% agarose gel containing ethidium bromide (0.5 mg/ml) and viewed under an UV illuminator. Each band was quantified using densitometry by calculating the ratio of the target cDNA signal to the GAPDH control, and mRNA expression was presented as a percentage of the GAPDH signal. Real-time QPCR was carried out in a MicroAmp Optical 96-well plate using power SYBR Green PCR Master Mix (Applied Biosystems Inc., Foster City, CA, USA), with 2 ml cDNA in each well. PCR reactions were monitored in real-time using the ABI PRISM 7900 Real-time PCR System (Applied Biosystems Inc.). The PCR primers for human VEGF were designed using the Primer ExpressÒ program and synthesized by Applied Biosystems (Foster City, CA). The sequences used were (forward) 50 -TCTACCTCCACCATGCCAAGT-30 and (reverse) 50 -TGCGCTGAT AGA TATCCATGA-30 , generating a 102-bp PCR product. The thermal cycling conditions for real-time PCR were 50  C for 2 min, then 95  C for 10 min, followed by 40 cycles of denaturing (95  C, 15 s) and annealing/ extension (60  C, 60 s). After examination of the PCR cycle dissociation curve, the relative VEGF mRNA expression (normalization to GAPDH) was determined using the DDCt method, according to the manufacturer’s directions.

26

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

Fig. 2. Effect of TGF-b1 on VEGF generation by human lung fibroblasts. (A) VEGF level was determined in supernatants of fibroblasts incubated for 3, 6 and 24 h with buffer alone or varying concentrations of TGF-b1. Mean VEGF values (SEM) are shown for four to six experiments performed in duplicate. *p < 0.05, **p < 0.005 compared with cells incubated with buffer alone. (B) RT-PCR analysis of expression of mRNA encoding the 121-, 165- and 189-amino acid isoforms of VEGF and the housekeeping gene GAPDH in fibroblasts following incubation for 3 and 6 h with varying concentrations of TGF-b1. (C) Relative VEGF mRNA expression normalized to internal control GAPDH. Data are shown for 3 experiments performed in duplicate and expressed as fold increase of those of cells without TGF-b1 stimulation. *p < 0.05, **p < 0.005 compared with cells incubated with buffer alone.

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

27

Fig. 3. Effect of TGF-b1 on Smad3 and p44/42 MAP kinase activation. Cells were treated with TGF-b1 (10 ng/ml) for the durations indicated and harvested for Western blotting. Immunodetection was performed using specific antibodies to phosphorylated Smad3 (Smad3-p) and p44/42 (p44/42-p), or pan-Smad3 (Smad3) and pan-p44/42 (p44/42). Blots presented are representative of three separate experiments.

Inhibition of VEGF production was performed using VEGFdirected siRNA reagents (human VEGF, siGENOME SMARTpool siRNA; Dharmacon, Lafayette, CO). Human fibroblast cells were seeded onto 24-well plates 24 h before transfection. Cells were 60% confluent in DMEM medium. Fibroblasts were transfected with siRNA duplexes using DharmaFECT1 (Dharmacon) according to the manufacturer’s instructions. VEGF siRNA was designed and synthesized by the Dharmacon Inc. siRNA consisted a mixture of (1) sense strand: 50 GCAGAAUCAUCACGAA GUG30 , anti-sense: 50 CACUUCGUGAUGAUUCUGC 30 ; (2) sense strand: 50 CAA CAAAUGUGAAUGCAGA 30 , anti-sense strand: 50 UCUGCAUUCACAUUUGUUG 30 ; (3) sense strand: 50 GGAGUACCCUGAUGAG AUC 30 , anti-sense strand: 50 GAU CUCAUCAGGGUAC UCC 30 ; and (4) sense strand: 50 GAUCAAACCUCACCAA GGC 30 , anti-sense strand: 50 GCCUUGGUGAGGUUUGAUC 30 (Accession #: NM_001025370; GI:71051584; VEGFA, transcript variant 6, mRNA). Protein levels were measured at 48 h posttransfection.

aprotinin (0.2 U/ml), leupeptin (0.5 mg/ml), and 1 mM Na3VO4. The protein concentration was determined using a bicinchoninic acid protein assay (Pierce Chemicals, Rockford, IL, USA) with bovine serum albumin as the standard. Equal amounts of total cell lysates (15 mg) were solubilized in a sample buffer by boiling for 10 min, fractionated on 7.5% SDS-polyacrylamide gel, and transferred onto a nitrocellulose membrane. The membrane was washed with Trisbuffered saline (TBS) supplemented with 0.1% Tween 20 and incubated in a blocking buffer (TBS containing 5% non-fat dry milk and 0.1% Tween 20). Anti-phospho-p46/54 (SAPK/JNK, Thr183/Tyr185) antibody, anti-phospho-p44/42 (Thr202/Tyr204) antibody, antiphospho-Smad3 antibody or anti-phospho-p38 (Thr180/Tyr182) antibody (Cell Signaling Technology) was then applied at a 1:1000 dilution at 4  C overnight, with gentle shaking. After washing three times with TBS, blots were incubated with a 1:2000 dilution of a horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology) for 1 h. The protein bands were viewed using enhanced chemiluminescence (Amersham Pharmacia Biotech, Sunnyvale, CA, USA) and autoradiography with Kodak X-ray film.

2.7. Western blot analysis of Smad3 and MAP kinases

2.8. Statistics

Cells were exposed to TGF-b1 in the presence or absence of inhibitors of Smad3 or MAP kinase activity. After treatment, cells were lysed on ice in lysis buffer containing 50 mM Tris $ HCl at pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, pepstatin A (1 mg/ml),

Data are expressed as means  SE. Statistical analysis for multiple comparisons was performed with ANOVA for parametric data. Student’s t-test (for the VEGF assay) or the paired Student’s t-test (for the mRNA expression data) was employed. P values <0.05 were considered significant for all tests.

2.6. Small interference RNA (siRNA) targeting VEGF

28

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

Fig. 4. Effects of erythromycin, dexamethasone and UO126 on TGF-b1-induced p44/42 MAP kinase activation and VEGF production. Cells were treated with TGF-b1 (10 ng/ml) in the presence or absence of varying concentrations of erythromycin (1, 10 and 100 mg/ml; Ery 1, Ery 10 and Ery 100, respectively), dexamethasone (100 mM) or UO126 (10 mM) for 3, 6 and 24 h, then the supernatant was collected for VEGF measurement. (A) Cells were treated with TGF-b1 (10 ng/ml) alone or in combination with erythromycin (1 mg/ml), dexamethasone (100 mM) or UO126 (10 mM), and then harvested for Western blotting after 1 h of stimulation. Representative immunoblots for phosphorylated p44/42 are shown (B). Data are presented for three separate experiments. Significant inhibition (*p < 0.05, **p < 0.005) was detected as compared with cells treated with TGF-b1 alone.

3. Results 3.1. Increased TGF-b1, VEGF and vimentin expression in the granulation tissues

were observed in the submucosal layers of granulation tissues as compared with the control tissues (Fig. 1C). TGF-b1 expression was detected in both the epithelial and submucosal layers (Fig. 1B). 3.2. Effect of TGF-b1 on VEGF release from human lung fibroblasts

The demographic data of the patients with benign airway stenosis are shown in Table 1A. Non-specific anti-mouse IgG antibody was used as a negative control (Fig. 1A). Increases in expression of TGF-b1 (Fig. 1B), VEGF (Fig. 1C) and vimentin (Fig. 1D) in the granulation tissues of patients with benign airway stenosis were observed (Table 2). Immunostaining for the above factors was nearly undetectable in the specimens from the control subjects without stenotic airways (Table 1B), and only one specimen from a control subject showed weak expression of VEGF and TGF-b1. VEGF expression was increased significantly in eleven out of the twelve subjects in the granulation group. Vimentin is the major subunit protein of the intermediate filaments and is a characteristic marker for mesenchymal cells. The highest levels of VEGF and vimentin expression were detected in the submucosal layers. Most VEGF expression was observed in the cytoplasm of infiltrating cells in the submucosa and endothelial cells. Significantly increased numbers of small vessels

The effect of TGF-b1 on the generation of VEGF by human lung fibroblasts is shown in Fig. 2A. The basal secretion of VEGF from fibroblasts was detectable in culture supernatants at the 3-h time point, and the maximal concentration of VEGF (2638  296 pg/ml vs. 1173  126 pg/ml, stimulated vs. control, p ¼ 0.003) was detected in the presence of TGF-b1 (10 ng/ml) after 24 h of incubation. Reverse transcription-polymerase chain reaction was employed to detect the mRNA expression of VEGF. After stimulation of fibroblasts with TGF-b1 (10 ng/ml), the VEGF mRNA expression reached a peak at 3 h, and faded at 6 h (Fig. 2B). GAPDH was used as an internal control. To further confirm the increase in VEGF mRNA expression, real-time QPCR was performed. The significant expression relative to that of the GAPDH housekeeping gene was approximately 9.54-fold that of unstimulated cells 3 h after TGF-b1 (10 ng/ml) stimulation (Fig. 2C).

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

29

Fig. 5. Effect of Smad3 inhibitor SIS3 on p44/42 MAP kinase activation and VEGF production. Cells were treated with TGF-b1 (10 ng/ml) in the presence or absence of varying concentrations of SIS3 (1, 10, and 50 mM) and then harvested for Western blotting after 1 h of stimulation. Representative immunoblots for phosphorylated p44/42 are shown (A). Supernatant was collected for VEGF measurement in the presence or absence of UO126 (10 mM) or SIS3 (1, 10, and 50 mM) after 6 h of stimulation with TGF-b1 (10 ng/ml). Data are presented for three separate experiments. Significant inhibition of VEGF production (*p < 0.05, **p < 0.005) was detected as compared with cells treated with TGF-b1 alone.

3.3. TGF-b1-induced Smad3 and p44/42 MAP kinase phosphorylation We investigated whether TGF-b1 induced VEGF production through Smad3 activation and MAP kinase phosphorylation. TGF-b1 induced Smad3 phosphorylation at 15 min, which peaked at 90 min (Fig. 3A), while p44/42 MAP kinase (Erk1/Erk2) phosphorylation started at 15 min, peaked at 60 min, and declined at 120 min (Fig. 3B). There were no observable differences in p38 and p46/54 (SAPK/JNK) phosphorylation in either the presence or absence of TGF-b1 stimulation (data not shown).

(1e100 mg/ml). TGF-b1-induced VEGF generation was not inhibited by pretreatment of cells with dexamethasone, but erythromycin significantly attenuated TGF-b1-induced VEGF production at a concentration of 1 mg/ml; the inhibitory effect was not dosedependent (Fig. 4A). To determine the effect of dexamethasone, erythromycin and UO126 on p44/42 MAP kinase activation, the phosphorylation status of these enzymes was examined. Pretreatment of cells with erythromycin (1 mg/ml) and UO126 (10 mM) resulted in inhibition of the p44/42 MAP kinase phosphorylation activated by TGF-b1 (Fig. 4B).

3.4. Inhibition of TGF-b1 induced VEGF release

3.5. Effect of Smad3 inhibitor on p44/42 MAP kinase activation and VEGF production

To suppress TGF-b1-induced VEGF release, fibroblasts were treated with dexamethasone (1e100 mM) or erythromycin

We observed that the TGF-b1-induced VEGF release from fibroblasts involved p44/42 MAP kinase activation and that the

30

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

phosphorylation of p44/42 MAP kinase was abolished by UO126. The relationship between Smad3 phosphorylation and p44/42 MAP kinase activation was further investigated. Pretreatment of cells with SIS3 (1e100 mM), a specific inhibitor of Smad3, resulted in substantial suppression of p44/42 MAP kinase phosphorylation (Fig. 5A) and substantial inhibition of TGF-b1-induced VEGF production (Fig. 5B). UO126 or SIS3 alone did not affect the levels of VEGF (data not shown). 3.6. Effect of mRNA silencing on VEGF production In order to confirm TGFb1-induced VEGF production, RNA interference was performed. siRNA for VEGF significantly suppressed TGFb1-induced VEGF production to a great extent (Fig. 6). Transfection reagent (FECT) and specific RNA (Dharmacon) for negative control did not affect the levels of VEGF. 4. Discussion The current study demonstrated a mechanism that can be attributed to the development of granulation formation in benign airway stenosis. Increases in expression of TGF-b1 and VEGF, as well as in the numbers of fibroblasts, were observed in granulation tissue obtained from patients with benign airway stenosis. TGF-b1 stimulated production of VEGF by cultured fibroblasts at both the mRNA and protein levels in vitro. VEGF siRNA treatment resulted in a significant decrease of TGF-b1-induced VEGF production. TGF-b1, released from epithelial cells, may stimulate fibroblasts to produce VEGF through Smad3 and p44/42 MAP kinase. Erythromycin and UO126 can inhibit this p44/42 MAP kinase phosphorylation and attenuate subsequent VEGF production to a great extent, whereas dexamethasone failed to inhibit TGF-b1-induced VEGF production. Targeting these mediators and cells may help to prevent the formation of granulation in patients with long-term tracheostomy or prolonged endotracheal intubation. A high incidence of tracheal granuloma formation has been observed in patients with long-term tracheostomy [1]. The tracheal stretching and expansion that occurs during respiration, head movement and irritation may contribute to the development of granulation [13]. Studies performed on trachea and lung explants have demonstrated that airway epithelial damage may induce a fibrotic response [14,15]. Mechanical irritation of the mucosa may produce cytokines that induce formation of granulation tissue

Fig. 6. Effect of RNA interference on VEGF production. siRNA (100 nM) and transfection reagent (1 ml, FECT) were added to the fibroblasts and incubated for 24 h. Cells were then treated with TGF-b1 (10 ng/ml) and culture media were harvested for VEGF measurement 24 h after treatment. Specific RNA from Dharmacon was used as a negative control. Significant inhibition of VEGF production (**p < 0.005) was detected as compared with cells treated with TGF-b1 without RNA interference. Data are presented for three separate experiments, each of which contained triplicate cultures.

[13,16]. TGF-b exists in three isofroms called TGF-b1, TGF-b2 and TGF-b3. Most of the studies suggest that compression stress [5] or cyclic stretch [6] induces TGF-b1 production from airway epithelial cells. In addition, Karagiannidis et al. [7] demonstrated a high-level expression of TGF-b1 (but not TGF-b3) in specimens from benign airway stenosis. In contrast, Tschumperlin et al. [4] reported that mechanical stress triggers a release of fibrotic mediator TGF-b2 from bronchial epithelium. In our preliminary results, immunohistochemistry failed to demonstrate a positive staining for TGF-b2 and TGF-b3 in the granulation tissues. Instead of TGF-b2 and TGF-b3, increased expression of TGF-b1 was observed. These findings were also compatible with those reported by Karagiannidis et al. TGF-b1 is a strong extracellular matrix inducer, a chemotactic mediator for fibroblasts and polymorphonuclear cells [17,18] and a mitogen for fibroblasts [19]. Regenerating epithelial cells and fibroblast proliferation appear to be essential in tracheal wound healing following mechanical injury [20]. We found that vimentin-positive cells were recruited to and aggregated in the submucosa of granulation tissue. The fibrogenic mediator TGF-b1, released from epithelial cells, may subsequently stimulate these vimentin-positive cells to increase cell proliferation and matrix deposition. In this study, human lung fibroblasts were applied and we observed that TGF-b1 stimulates fibroblasts to produce VEGF. Vascular development is regulated by endothelial cells in response to various kinds of angiogenic growth factors and proteins, such as fibroblast growth factor (FGF), angiopoietins, platelet-derived growth factor (PDGF), VEGF and matrix metalloproteinase [21]. FGF has been reported to act as a mitogen for both endothelial cells and mural cells [22]. PDGF has a potent synergistic effect in combination with VEGF and FGF in the induction of neoangiogenesis [23]. VEGF plays an important role in stimulating endothelial cell mitogenesis, migration and angiogenesis [24,25]. Production of VEGF indicates neovascularization, which characterizes tissue repair following injury. TGF-b1 signaling within cells through the Smad family of transcriptional activators has been thoroughly investigated [26]. TGF-b1/Smad signaling is strongly associated with MAP kinase signaling cascades [9]. We observed that TGF-b1 activated Smad3 and p44/42 kinase in fibroblasts; Smad3 inhibitor affected the phosphorylation of p44/42 kinase, and inhibition of Smad3 and p44/42 kinase phosphorylation resulted in a significant reduction of VEGF release. Corticosteroids are currently the most popular anti-inflammatory agents used in the management of chronic inflammatory airway diseases; they can inhibit fibroblast proliferation and associated cytokine release in a dose-dependent manner. 100 nM of dexamethasone has been shown to have an inhibitory effect on cell proliferation and cytokine release in fibroblasts [27,28]; however, this was not found to be the case in our study, in which dexamethasone of concentrations up to 100 mM failed to show an inhibitory effect on TGF-b1-induced VEGF production by fibroblasts. Erythromycin can prevent bleomycin-induced lung injury by attenuation of inflammatory cell migration and VCAM-1 expression and by downregulation of neutrophil-derived elastase [29,30]. Yu et al. reported that erythromycin can inhibit TGF-b signaling through inhibition of the Smad2/3 pathway in human lung fibroblasts [31]. In addition, low-dose erythromycin has a greater suppressive effect on intrapulmonary neutrophil influx induced by intratracheal injection of lipopolysaccharides than high-dose erythromycin [32]. We found that erythromycin at a low concentration (1 mg/ml) significantly inhibited TGF-b1-induced p44/42 phosphorylation and subsequent VEGF production. In clinical applications, low-dose erythromycin has been shown to exert a positive effect on various chronic inflammatory airway disorders, including asthma, bronchiectasis and diffuse panbronchiolitis [33].

Y.-C. Lee et al. / Pulmonary Pharmacology & Therapeutics 24 (2011) 23e31

5. Conclusions This study provides important information that facilitates improved understanding of at least part of the mechanism of granulation formation. Therapeutic intervention to stop these interactions may help to prevent the development of granulation tissue and subsequent airway obstruction. Conflicts of interest The authors declare no conflict of interest. Acknowledgements This study was supported by research grants from Taipei Veterans General Hospital (VGH97C1-105). We thank Mo-Tzu Wu and Tun-Yun Hsueh for collection of granulation tissue and preparation of immunostaining. References [1] Law JH, Barnhart K, Rowlett W, de la Rocha O, Lowenberg S. Increased frequency of obstructive airway abnormalities with long-term tracheostomy. Chest 1993;104:136e8. [2] Liu H, Chen JC, Holinger LD, Gonzalez-Crussi F. Histopathologic fundamentals of acquired laryngeal stenosis. Pediatr Pathol Lab Med 1995;15:655e77. [3] Pokharel RP, Maeda K, Yamamoto T, Noguchi K, Iwai Y, Nakamura H, et al. Expression of vascular endothelial growth factor in exuberant tracheal granulation tissue in children. J Pathol 1999;188:82e6. [4] Tschumperlin DJ, Shively JD, Kikuchi T, Drazen JM. Mechanical stress triggers selective release of fibrotic mediators from bronchial epithelium. Am J Respir Cell Mol Biol 2003;28:142e9. [5] Ressler B, Lee RT, Randell SH, Drazen JM, Kamm RD. Molecular responses of rat tracheal epithelial cells to transmembrane pressure. Am J Physiol Lung Cell Mol Physiol 2000;278:L1264e72. [6] Yamamoto H, Teramoto H, Uetani K, Igawa K, Shimizu E. Cyclic stretch upregulates interleukin-8 and transforming growth factor-beta1 production through a protein kinase C-dependent pathway in alveolar epithelial cells. Respirology 2002;7:103e9. [7] Karagiannidis C, Velehorschi V, Obertrifter B, Macha HN, Linder A, Freitag L. High-level expression of matrix-associated transforming growth factor-beta1 in benign airway stenosis. Chest 2006;129:1298e304. [8] Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth factor beta in human disease. N Engl J Med 2000;342:1350e8. [9] Leask A, Abraham DJ. TGF-beta signaling and the fibrotic response. Faseb J 2004;18:816e27. [10] Wen FQ, Liu X, Manda W, Terasaki Y, Kobayashi T, Abe S, et al. TH2 Cytokineenhanced 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;111:1307e18. [11] Voelkel NF, Vandivier RW, Tuder RM. Vascular endothelial growth factor in the lung. Am J Physiol Lung Cell Mol Physiol 2006;290:L209e21.

31

[12] Jinnin M, Ihn H, Tamaki K. Characterization of SIS3, a novel specific inhibitor of Smad3, and its effect on transforming growth factor-beta1-induced extracellular matrix expression. Mol Pharmacol 2006;69:597e607. [13] Balestrieri F, Watson CB. Intubation granuloma. Otolaryngol Clin North Am 1982;15:567e79. [14] Adamson IY, Young L, Bowden DH. Relationship of alveolar epithelial injury and repair to the induction of pulmonary fibrosis. Am J Pathol 1988;130:377e83. [15] Dai J, Gilks B, Price K, Churg A. Mineral dusts directly induce epithelial and interstitial fibrogenic mediators and matrix components in the airway wall. Am J Respir Crit Care Med 1998;158:1907e13. [16] Grillo HC. Stents and sense. Ann Thorac Surg 2000;70:1142. [17] Hannigan M, Zhan L, Ai Y, Huang CK. The role of p38 MAP kinase in TGFbeta1-induced signal transduction in human neutrophils. Biochem Biophys Res Commun 1998;246:55e8. [18] Postlethwaite AE, Keski-Oja J, Moses HL, Kang AH. Stimulation of the chemotactic migration of human fibroblasts by transforming growth factor beta. J Exp Med 1987;165:251e6. [19] Perng DW, Wu YC, Chang KT, Wu MT, Chiou YC, Su KC, et al. Leukotriene C4 induces TGF-beta1 production in airway epithelium via p38 kinase pathway. Am J Respir Cell Mol Biol 2006;34:101e7. [20] Horiba K, Fukuda Y. Synchronous appearance of fibronectin, integrin alpha 5 beta 1, vinculin and actin in epithelial cells and fibroblasts during rat tracheal wound healing. Virchows Arch 1994;425:425e34. [21] Coultas L, Chawengsaksophak K, Rossant J. Endothelial cells and VEGF in vascular development. Nature 2005;438:937e45. [22] Savona C, Javerzat S, Perollet C, Bikfalvi A. [Angiogenesis and neoangiogenesis]. Rev Prat 1997;47:2239e43. [23] Kano MR, Morishita Y, Iwata C, Iwasaka S, Watabe T, Ouchi Y, et al. VEGF-A and FGF-2 synergistically promote neoangiogenesis through enhancement of endogenous PDGF-B-PDGFRbeta signaling. J Cell Sci 2005;118:3759e68. [24] Ferrara N. Vascular endothelial growth factor and the regulation of angiogenesis. Recent Prog Horm Res 2000;55:15e35. discussion 35e16. [25] Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989;246:1306e9. [26] Attisano L, Wrana JL. Signal transduction by the TGF-beta superfamily. Science 2002;296:1646e7. [27] Brenner RE, Felger D, Winter C, Christiansen A, Hofmann D, Bartmann P. Effects of dexamethasone on proliferation, chemotaxis, collagen I, and fibronectinmetabolism of human fetal lung fibroblasts. Pediatr Pulmonol 2001;32:1e7. [28] Sabatini F, Silvestri M, Sale R, Serpero L, Giuliani M, Scarso L, et al. Concentration-dependent effects of mometasone furoate and dexamethasone on foetal lung fibroblast functions involved in airway inflammation and remodeling. Pulm Pharmacol Ther 2003;16:287e97. [29] Azuma A, Furuta T, Enomoto T, Hashimoto Y, Uematsu K, Nukariya N, et al. Preventive effect of erythromycin on experimental bleomycin-induced acute lung injury in rats. Thorax 1998;53:186e9. [30] Li Y, Azuma A, Takahashi S, Usuki J, Matsuda K, Aoyama A, et al. Fourteenmembered ring macrolides inhibit vascular cell adhesion molecule 1 messenger RNA induction and leukocyte migration: role in preventing lung injury and fibrosis in bleomycin-challenged mice. Chest 2002;122:2137e45. [31] Yu C, Azuma A, Li Y, Wang C, Abe S, Usuki J, et al. EM703, a new derivative of erythromycin, inhibits transforming growth factor-beta signaling in human lung fibroblasts. Exp Lung Res 2008;34:343e54. [32] Kadota J, Sakito O, Kohno S, Sawa H, Mukae H, Oda H, et al. A mechanism of erythromycin treatment in patients with diffuse panbronchiolitis. Am Rev Respir Dis 1993;147:153e9. [33] Amsden GW. Anti-inflammatory effects of macrolidesean underappreciated benefit in the treatment of community-acquired respiratory tract infections and chronic inflammatory pulmonary conditions? J Antimicrob Chemother 2005;55:10e21.