Anti‐fibrotic effect of ciglitazone in HRV‐induced airway remodelling cell model

Abstract Fibrosis is an important phenomenon as it can occur early in the pathogenesis of asthma; it may be associated with disease severity and resistance to therapy. There is a strong evidence that infection caused by human rhinovirus (HRV) contributes to remodelling process, but there is lack of studies clearly explaining this pathway. Synthetic peroxisome proliferator‐activated receptor (PPAR) γ presents immunomodulatory and anti‐inflammatory features. In this study, we examined immunomodulatory properties of ciglitazone – PPAR‐γ agonist, in development and modulation of airway remodelling. Epithelial cells (NHBE) and two lines of fibroblasts (WI‐38, HFL1) were stimulated with ciglitazone and rhinovirus. The expression of genes related to airway remodelling process were analysed in the cells; moreover NF‐κB, c‐Myc and STAT3 were silenced in order to estimate potential pathways involved. Ciglitazone decreased mRNA expression of MMP‐9 and TGF‐β. It also modified the expression of α‐SMA and collagen after rhinovirus infection. Transcription factors knockdown altered the levels of expression. The results suggest possible anti‐fibrotic activity of PPAR‐γ agonist in human airway cells. Ciglitazone has been shown to be dependent on NF‐κB‐ and STAT3‐related pathways, thus, the PPAR‐γ agonist may have therapeutic potential for the treatment of airway remodelling in asthma.


| INTRODUC TI ON
Bronchial asthma is a common chronic inflammatory disorder of the airways, caused by immune and inflammatory responses and is characterized by reversible airflow obstruction. 1 One of the central feature of asthma is airway remodelling; defined as changes in airway wall structure, including extensive epithelial damage, airway smooth muscle hypertrophy, collagen deposition, subepithelial basement membrane thickness and fibrosis. Asthma is an incurable nonetheless preventable disease; currently available treatments control airway inflammation and symptoms, but only to a certain degree. 2 It has been demonstrated previously that human airway cells infected with HRV may induce a number of growth factors as well as chemotactic agents connected to airway remodelling, that can drive mesenchymal cells to migrate towards the epithelial layer. 3 Minor et al 4 directly demonstrated that two distinct HRV serotypes are also capable of inducing EMT-related epithelial phenotypic and morphological alterations, further demonstrating the potential contribution of HRV infections to airway remodelling.
A number of mediators, including TGFβ, MMP-9, and ADAM33, have been identified to control airway remodelling. 5 TGFβ is a crucial profibrotic cytokine, which promotes the development of fibroblasts into myofibroblasts, that release growth factors and collagen.
TGFβ expression correlates with thickness of basement membrane, fibroblast numbers, and severity of the disease. 6 The complicated interaction between TGF-β1 and STAT3 appears to serve a variety of biological settings. Numerous fibrotic tissues have shown STAT3 activation; this occurs as a result of the inflammatory response, which is the initial phase of wound healing. 7 However, the molecular processes which would clarify the role of STAT3 in the initiation and development of fibrosis are still unknown.
STAT3 as well as NF-κB both regulate the expression of a large number of downstream genes that control immune functions, stress responses, cell proliferation and survival. Additionally, the two transcription factors are engaged in both positive and negative crosstalk. 8 Several studies demonstrate contribution of STAT3 to fibrosis through induction of ECM (extracellular matrix) production. [9][10][11] TGFβ has a major impact on the production of the ECM via its downstream effectors and it is crucial in signalling net leading to fibrosis. 12 NF-κB signalling in asthma has been demonstrated by increased IκB (inhibitor of NF-κB) phosphorylation or degradation, increased NF-κB nuclear localization or DNA binding, and increased IκB kinase complex (IKK)β expression in the airway tissue. 13 Moreover, increased NF-κB has been also detected in induced sputum inflammatory cells of the asthmatic patients, and inhaled budesonide has been effective to decrease the levels of activated NF-κB in human airway wall tissue. 14 The superfamily of ligand-activated nuclear receptors includes transcriptional factors called peroxisome proliferator-activated receptors (PPARs). There have been recognized as three main types: PPARα, PPARβδ, and PPARγ, and the last one is known to play a crucial role in airway inflammatory responses regulation. Previous studies demonstrated the role of PPARγ in asthmatic airway remodelling and inflammation. 15,16 PPARγ expression is increased in the airway mucosa of asthmatics in comparison to healthy subjects, 17 moreover, PPARγ inhibits expression of many cytokines and activation of inflammatory cells which are increased in asthma. 16 The thiazolidinedione class of anti-diabetic medications as well as naturally occurring ligands may both activate PPARγ, which is up-regulated by TGF-β1. The sensitivity of fibroblasts to PPARligands is greatly increased by transient overexpression of PPARγ.
15d-PGJ2 as well as troglitazone (PPARγ agonists) inhibit TGFβinduced transcriptional activation of type I collagen production and abolish α-SMA expression. 18 The PPARγ ligands have distinct and PPARγ dependent inhibitory effects on these profibrotic reactions. Moreover, the degree of cellular Smad3 or Smad7 expression is unaffected by PPARγ activation in fibroblasts. This might mean that PPARγ prevents TGFβ from inducing profibrotic responses in healthy fibroblasts. 18 There is a proof that PPAR-activation may control the ECM deposition involved in the remodelling of the airway wall. In response to antigen sensitization and challenge, ciglitazone reduced basement membrane thickness and airway collagen deposition, which was accompanied by a decrease in TGFβ production. 19 Ciglitazone has also been demonstrated to block TGFβ signalling in cultured lung fibroblasts. 20 Moreover, previous work by Chima et al. indicated that ciglitazone ameliorated lung inflammation by modulating the IKK/NF-κB pathway which was mediated through inhibition of the IKK/NF-κB pathway. 21 The aim of the study was to examine the signalling mechanisms for airway remodelling development, the role of chosen transcription factors -NF-κB, c-myc and STAT3, and to determine the effect of PPARγ agonist -ciglitazone influence on the expression of airway remodelling-related genes in the presence of rhinoviruses (HRV-2 and HRV-16) infections -factors triggering airway remodelling changes. Blue (BD Pharmingen) and the absorbance was measured at 570 nm.

| Virus preparation and cell infection
Two serotypes of human rhinoviruses (HRV) 16 and HRV-2 were purchased from the European Collection of Authenticated Cell Cultures (ECACC)). Ohio HeLa cell line, purchased from Sigma-Aldrich was infected until cytopathic effect was observed multiplicity of infection (MOI) of 1, established on the base of literature. 22,23 HRV specimens were exposed to the temperature of 58°C for 1 h in order to inactivate the virus particles, 24 which was subsequently confirmed by a lack of HRV replication.
The target fibroblast and epithelial cells were infected by the addition of 50 μL vehicle (medium) or HRV16. The cells were incubated for 24 h (33°C, 5% CO 2 ). 25

| Experimental procedure
The cultures were exposed to the HRV-2 (minor serotype) and HRV-16 (major serotype) virus for 24 h (33°C, 5% CO 2 ). Following this, the cells were incubated with ciglitazone (3 μM) for 24 h (37°C, 5% CO 2 ). Some of the wells were treated the other way -first ciglitazone was added (3 μM) for 24 h (37°C, 5% CO 2 ), and then, the cells were infected by HRV-2 or HRV-16. The aim of such action was to investigate whether there were differences at the molecular level due to the order in which ciglitazone was added. Namely, whether ciglitazone can prevent HRV-induced changes if it is added before the infection, or whether it can inhibit HRV action when it added after cells were infected. Ciglitazone added to cells before infection may have a preventive effect on the changes subsequently induced by HRV. On the other hand, ciglitazone added to already infected cells may have a mitigating effect; therefore, one of our goals was to analyse the effect of the order in which ciglitazone was added to cells in the context of HRV infection.
The controls contained medium only. We have chosen one HRV serotype from each group in order to evaluate potential differences between them. All the experiments were performed three times in duplicate (passages 3 to 9). The time-point chosen by out of three (6, 12 and 24 h, data not shown), consistent with the literature.

| RNA isolation and cDNA synthesis
Total RNA was isolated from the cells by using a Total RNA mini kit (A&A Biotechnology). The RNA was next purified and stored at −80°C. Reverse transcription was performed using a High Capacity cDNA kit (Applied Biosystems), using 1 μg of total RNA. The procedures were performed according to the producer's protocols.

| Gene expression analysis
The changes in the expression of metalloproteinase -9, transforming growth factor β1(TGF-β1), collagen I, disintegrin and metallopro-

| Statistical analyses
The obtained results were analysed with software Statistica software (StatSoft). The Shapiro-Wilk test and Levene's test, were respectively utilized to check the distribution of data as well as the equality of variances. Significant changes were calculated using the anova test with the appropriate post hoc tests as a multiple comparison procedure. p < 0.05 were considered to be statistically significant.

| Ciglitazone effect on airway remodellingrelated genes expression
Ciglitazone showed no cytotoxicity towards any of the cell lines used (Figure 1). The effect of ciglitazone caused noticeable changes in MMP-9 and TGF-β1 mRNA expressions, resulting in reduced rhinovirus-induced mRNA expression (HRV-16 and HRV-2), for all cell types tested (Figure 2A 3.79 after HRV-16+ciglitazone (p < 0.05)), with the effect not observed when ciglitazone was added prior infection ( Figure 2D).
Protein expression analysis showed statistical significance for HFL1 cells -ciglitazone decreased HRV-2 and HRV-16 induced MMP-9 protein expression (p < 0.05, Figure 3A). In addition, statistical significance was observed for TGF-β1 protein expression -for HFL1 cells ( Figure 3B), ciglitazone decreased HRV-16-induced protein expression; both in the experimental system when ciglitazone was added to the existing infection and when it was added before the infection (p < 0.05), while for WI-38 cells -a significant change appeared only when ciglitazone was added to the existing infection (p < 0.05, Figure 3C). The other expression values of the tested proteins showed no statistical significance (p > 0.05).   Additionally, the evaluation of PPARγ concentrations in all the cell lines used (Figure 9). The data showed that PPAR-is reduced under the influence of rhinovirus, but ciglitazone w can modulate these concentration changes.

| DISCUSS ION
Airway inflammation and remodelling are the two characterized pathological features of asthma. PPARγ exerts significant influence on inflammatory responses through acting on T cells, macrophages, dendritic cells, and mast cells, as indicated by the literature. [26][27][28] When activated, PPARγ in smooth muscle cells exhibit antiinflammatory effects as opposed to corticosteroid effects. 29 PPARγ agonists suppress the production of pro-inflammatory cytokines such IL-1, TNFα, and IL-6 in monocytes and macrophages. 30 Taking into consideration the anti-inflammatory potential of PPARγ, we aimed to evaluate its agonist -ciglitazone in airway remodelling phe- Instead, activation of PPARγ-represses MMP-9 production, at least in part, through inhibiting the NF-κB signalling pathway. In fact, it was discovered that PPARγ activation by ox-LDL led to considerably less NF-κB binding to the κB site, moreover, LXR/RXR heterodimers were proved to inhibit expression of MMP-9 gene in a similar

F I G U R E 4
The effect of ciglitazone treatment on remodelling process induced by rhinovirus, under conditions of siRNA knock down of NF-κB transcription factor. Rhinovirus 2 (H2, minor serotype) and 16 (H16, major serotype) were used. Ciglitazone (Ci) has been added before rhinovirus infection (Ci+H16, Ci+H2) or after rhinovirus infection (H16+Ci, H2+Ci). NF-κB silencing eliminated effects of HRV and ciglitazone in MMP-9 (A) and TGFβ (B), (p > 0.05). The knockdown reversed changes introduced by ciglitazone in the expression of MMP-9 (A), COL I (C), α-SMA (E), and partly in TGFβ (B) and ADAM 33 (D), in comparison to cells expressing NF-κB (p < 0.05). *p < 0.05 in comparison to the control sample, +p < 0.05 in comparison to the rhinovirus sample (HRV-2 or HRV-16, respectively); C -control sample with medium only. manner. 33 As PPARγ inhibits the oxidative radicals, it may be also assumed that MMP-9 expression is also dependent on ROS. Walter et al demontrated recently, that MMP-9 expression correlates with the reduced ROS levels. Previous studies reported such dependence in cancer cells and astrocytes. 34,35 TGFβ and growth factors, among other inflammatory cytokines, are essential for the synthesis and release of MMP-9. 36 TGFβ is a profibrotic cytokine, its isoforms are implicated in the changes of ECM observed during fibrosis. TGFβ expression is increased in asthmatic airways and appears to correlate with disease severity and degree of subepithelial fibrosis. 37 In vitro, TGFβ is able to induce secretion of a numerous extracellular matrix proteins, e.g. collagen I and III, proteoglycans and fibronectin, in fibroblasts. 15 It is also a essential cytokine for the induction of collagens after tissue injury. It is known that that collagen I, is the most abundant protein of extracellular matrix, deposited in the asthmatic airways and a hallmark of fibrosis. 40 In our study, collagen expression increased upon rhinovirus stimulation, however ciglitazone hampered this action in epithelial cells but not in fibroblasts. This is quite unexpected as fibroblasts are the major players in the production of collagen I during fibrosis. We thus hypothesize that this might be due to the low ciglitazone concentration, or the limitation resulting from monoculture of fibroblasts and epithelial cells. Current knowledge recognizes airway fibroblasts and epithelial cells as the two cell types fundamentally engaged in the processes leading to tissue healing following damage. Their close contacts with other parenchymal and/or inflammatory cells appear to enhance their significant metabolic and immunologic features. 41 Furthermore, collagen gene expression in fibroblasts is suppressed by PPARγ receptor agonists. Effective inhibitors of TGF-β1-induced collagen synthesis and COL1A2 gene promoter activity include 15d-PGJ2 and thiazolidinedione. Additionally, dominantnegative PPARγ expression vectors prevent 15d-PGJ2 from inhibiting TGFβ stimulated COL1A2 promoter activity. 18 Next to collagen I, α-SMA is widely used as indicator of fibrosis, including lung fibrosis. In this study it was used as marker of myofibroblast differentiation. The expression pattern of α-SM actin in cells has been related to a transition from proliferative to differentiated state and cell-cell contact or communication. 42 Previous chinese study demonstrated that inhibition of alpha-SMA expression alleviate airway remodelling in mice. 43 We demonstrate that ciglitazone dicrease rhinovirus-induced α-SMA expression, thus showing possible potential of this drug. Surprisingly, the expression of α-SMA appeared only in fibroblasts, after preincubaiton with ciglitazone, however, regarding fibroblasts, these data are consistent with the literature -TGF-β1 has been shown to induce ECM and α-SMA expression by activating the PI3 kinase/Akt pathway and PPARγ agonists suppress TGF-β1-induced synthesis of ECM and α-SMA by blocking Akt phosphorylation. 44 They are also internally consistent, as ciglitazone simultaneously reduced the expression of the studied genes closely involved in airway remodelling but also influencing each other in this process. It is, however important to consider the data with caution, as previous studies have shown 44 57 We hypothesize that collagen I may have more than one expression pathways which may be swiched, however, it is difficult to support it, as no data is available regarding this issue. siRNA inhibition of STAT3 transcription factor, reversed the changes in TGFβ expression caused by ciglitazone and rhinoviruses, it also significantly reduced the expression of ADAM33 expression, which might mean that this transcription factor is pivotal for the ex-

| CON CLUS ION
In conclusion, in this study we found that ciglitazone stimulation attenuated expression of some genes strongly connected to airway remodelling. Moreover, ciglitazone decreased rhinovirus-induced F I G U R E 8 The result of siRNA knockdown of STAT3 on collagen I and α-SMA protein expression. Ciglitazone (Ci) has been added before rhinovirus infection (Ci+H16, Ci+H2) or after rhinovirus infection (H16+Ci, H2+Ci). The confirmation of mRNA expression on protein level was observed in COL I (A, B) and α-SMA (C, D). Ciglitazone (Ci) reversed the HRV-induced augmentation, regardless the virus serotype and the order of stimulation. α-SMA protein expression has been decreased significantly only after HRV-16 (H16) stimulation (p < 0.05). The bands presented were obtained from Western Blot analysis. The concentrations were measured in supernatants utilizing immunoassay in triplicate. Data presented as % of change, normalizing to control sample ± SD. +<0.05.

F I G U R E 9
Effects of ciglitazone and two Rhinovirus serotypes on PPAR-levels in fibroblast and epithelial cells. Ciglitazone caused an increase in PPARγ while both rhinovirus serotypes decreased it. Ciglitazone added after HRV infection increased PPARγ concentrations in HFL1 fibroblasts and epithelial cells. Ciglitazone (Ci) has been added before rhinovirus infection (Ci+H16, Ci+H2) or after rhinovirus infection (H16+Ci, H2+Ci). Data presented as mean ± SEM, *p < 0.05 in comparison to control, #p < 0.05 in comparison to HRV16 or HRV2. expression of the most important fibrotic genes -MMP-9, TGFβ and collagen I suggesting that the therapeutic effect on airway remodelling exerted by ciglitazone, was associated with activation of PPARγ and its down-stream pathways. In this process, NF-κB, c-Myc and STAT3 transcription factors may contribute. Further investigation into the mechanisms through which PPARγ inhibits airway remodelling are needed.

ACK N O WLE D G E M ENTS
Not applicable.

FU N D I N G I N FO R M ATI O N
This work was supported by a grants from the National Science

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors have no relevant financial or non-financial interests to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy restrictions.