Background
Idiopathic pulmonary fibrosis (IPF) is a chronic and ultimately fatal disease characterized by a progressive decline in lung function [
1,
2]. In the US, IPF affects over 150,000 people, and 40,000 patients die from IPF each year [
3,
4]. It is widely thought that IPF causes loss of alveolar epithelium and accumulation of fibroblasts and myofibroblasts that in turn result in collagen deposition and fibrosis. These changes cause disruption of the gas permeability barrier and stiffen lung tissue, thereby impairing ventilation of nearby normal lung tissue [
1,
5,
6]. IPF patients eventually experience dyspnea at rest and, without a lung transplant, have a 5-year average survival rate of only 20 % [
3,
4].
A wide range of therapies have failed in IPF clinical trials [
7,
8]. Pirfenidone (PFD) is an orally administered small molecule drug with anti-fibrotic and anti-inflammatory activities [
9‐
12]. It slows the progressive loss of lung function in IPF and was recently approved by the U.S. Food and Drug Administration (FDA) to treat IPF [
13‐
16]. Unfortunately, the therapeutically effective dose of PFD is very high (>30 mg/kg, daily) [
17,
18], which can cause significant side effects such as gastro-esophageal reflux, phototoxicity, dizziness, and fatigue, thereby limiting its therapeutic effectiveness [
16]. Understanding the molecular mechanisms that contribute to PFD effects could improve its safety and efficacy. Activated fibroblasts/myofibroblasts are key pathological features of IPF and are critical effector cells associated with the progression of fibrosis [
5,
6,
19,
20]. Several studies have indicated that PFD reduces growth factor-driven fibroblast proliferation, differentiation, and extracellular matrix production [
10,
21]. However, the specific molecular mechanisms whereby PFD produce these diverse effects to achieve its clinical benefits in IPF remain unknown [
15].
In the studies presented here, we used the GeneChip microarray method to identify RGS2 as an early response gene elevated in response to PFD treatment of human lung fibroblast cells. RGS2 selectively inhibits the magnitude and duration of G
q protein-coupled receptor-induced signaling [
22‐
24]. G protein-coupled receptors (GPCRs) have been shown to play important roles in the pathogenesis of chronic lung diseases including IPF [
25‐
28]. There is also evidence that RGS2 is functionally important in regulating the pathogenesis of fibrosis. For example, RGS2 was found to inhibit the progression of kidney fibrosis following unilateral ureteral obstruction in mice [
29]. We recently found that mice lacking RGS2 exhibit increased peribronchial fibrosis in an acute mouse model of asthma induced by administration of intranasal interleukin 13 (5 μg/day) for 3 days [
30]. However, the role of RGS2 in the onset and progression of IPF remains unknown. Using cellular and animal models, we demonstrate here that endogenous RGS2 exhibits anti-fibrotic functions and that early upregulation of RGS2 contributes to PFD amelioration of pulmonary fibrosis.
Methods
Reagents and cells
PFD (5-methyl-1-phenyl-2-(1H)-pyridone) and human thrombin were purchased from Sigma-Aldrich (St. Louis, MO). Unless indicated otherwise, all other reagents were from Sigma-Aldrich or Thermo Fisher Scientific (Waltham, MA). The human fetal lung fibroblast cell line (HFL1) was purchased from ATCC (CCL-153™, Manassas, VA). Primary human lung fibroblasts were isolated and cultured from lung tissues obtained during open lung biopsy from patients who were undergoing lung transplantation or at time of death. Control primary human lung fibroblast (CPHLF) cell lines were established by Dr. Reynold Panettieri’s lab (University of Pennsylvania) from three patients with brain-related disease but no history of pulmonary fibrosis. Diseased primary human lung fibroblast (DPHLF) cell lines were established by Dr. Carol Feghali-Bostwick (Medical University of South Carolina) [
31,
32] from two patients with IPF who had a confirmed diagnosis based on the criteria established by the American Thoracic Society. The biopsies had no identifiers, and the protocols for cell isolation were approved by the respective universities’ Institutional Review Boards.
Cell culture and drug treatments
Cells were routinely cultured at 37 °C with 5 % CO2 in 1:1 mixture of Dulbecco's Modified Eagle's Medium (DMEM) and Ham's F-12 Nutrient Mixture (F12) supplemented with 10 % FBS and were used at passage <10 for experiments. For dose–response experiments, human lung fibroblast cells were seeded into 12-well plates, starved in serum-free DMEM/F12 medium for 24 h, and then treated with the indicated concentrations of PFD for the indicated times.
Quantitative PCR
Total RNA extraction and quantitative real-time PCR were conducted as described previously [
33]. The RNA samples were treated with DNase I (Thermo Scientific, Waltham, MA) to remove contaminating genomic DNA prior to experiments. The housekeeping gene β-actin plasmid was purchased from ATCC (MGC-10559). Human RGS2 was constructed into pcDNA3.1 (Invitrogen Life Technology) [
30]. For RGS2, standard curves were generated by plotting the threshold cycle (C
T) against the natural log of the copy number of plasmid molecules. The equations from the graphs were used to calculate the copy numbers of cDNA molecules present in the unknown samples based on the corresponding C
T values. For connective tissue growth factor (CTGF), standard curves were generated by plotting the C
T against serial dilutions of sample. The primers for CTGF were: 5’-CAGCATGAAGACATACCGAGC-3’ and 5’-GACAGTTGTAATGGCAGGCAC-3’.
GeneChip microarray
Total RNA of control and PFD-treated HFL1 cells pooled from three individual experiments were analyzed by the Genome Sequencing Facility, University of Kansas Medical Center using Human Gene 2.0 ST Arrays (Affymetrix, Santa Clara, CA) and standard Affymetrix protocols [
34]. GeneChip microarray analysis was replicated with two different pooled samples and the results were expressed as the ratio of PFD-treated vs. vehicle control.
Western blot analysis
Cell protein extraction and western blot were conducted as described previously [
30,
35]. Protein was extracted from cells using 1x RIPA lysis buffer (Santa Cruz, Dallas, TX). Samples were electrophoresed and subjected to western blot using primary antibodies against RGS2 (Proteintech, Chicago, IL) and β-actin. IRDye-800 secondary antibodies (LI-COR, Lincoln, NE) were used to capture images with a LI-COR Odyssey.
Overexpression of recombinant RGS2 in HFL1 cells
The medium of HFL1 cells (50 % confluence) was replaced with growth medium containing human Adenovirus Type5 expressing RGS2 under the ubiquitin C promoter with co-expression of an mCherry reporter under a separate cytomegalovirus promoter (Ad-RGS2/mCherry) (SignaGen Laboratories, Rockville, MD) at a multiplicity of infection of 50 plaque-forming units. Cells infected with adenovirus expressing mCherry reporter (Ad-mCherry) were used as a control. The cells were incubated for 24 h at 37 °C and re-seeded for various experiments.
Measurement of intracellular Ca2+ ([Ca2+]i)
HFL1 cells seeded into 96-well plates at 1 × 105cells/well were cultured in serum-free medium for 24 h. Thrombin-induced changes in intracellular Ca2+ concentration were measured with the Fluo-8 No Wash Calcium Assay kit (Abcam, Cambridge, MA) according to kit instructions. The plates were transferred to a FLEX Station II benchtop scanning fluorometer chamber (Molecular Devices, Sunnyvale, CA). The cells were excited at 490 nm and Ca2+-bound fluo-8 emission was recorded at 525 nm. The fluo-8 fluorescence was expressed as Fmax/F0 where Fmax was the maximum and F0 was the baseline fluorescence measured.
Cell proliferation assay
HFL1 cells were treated with thrombin (1 U/ml, Sigma-Aldrich, St. Louis, MO) for 6 h and then labeled with 10 μM 5-bromo-2’-deoxyuridine (BrdU) (BD Pharmigen, San Jose, CA) for 18 h. Cells were stained for nuclei and BrdU using 4’,6-diamidino-2-phenylindole (DAPI) and anti-BrdU antibody (Cell Signaling, Danvers, MA) [
36]. Results are expressed as the percentage of DAPI-stained cells that were also BrdU-positive.
Immunofluorescence staining for α-smooth muscle actin
HFL1 cell α-smooth muscle actin (α-SMA) was visualized with an anti-α-actin primary antibody and an Alexa Fluor 488-labeled secondary antibody [
37]. The results are expressed as the percentage of DAPI-stained cells that also have clear α-SMA–positive stress fibers.
Gel contraction assay
HFL1 cells (2.5×10
5 cells/ml) were suspended in type I collagen from rat tail tendon (1.25 mg/ml, BD Bioscience, Bedford, MA). The collagen-cell suspension was added to 24-well plates (300 μl /well) and allowed to polymerize for 45 min at 37 °C. After incubating with DMEM containing 10 % fetal bovine serum (FBS) for 4 h, the medium from the polymerized gels was changed to serum-free medium. Polymerized gels were incubated overnight and were then stimulated with thrombin (1 U/ml) for 24 h. To initiate collagen gel contraction, polymerized gels were gently released from the underlying culture plate and digitally photographed. The gel areas were analyzed using Image-Pro Plus 6.0 (Media Cybernetics, Rockville, MD) and expressed as arbitrary area units [
37].
Measurement of collagen production and deposition
The amount of lung collagen was measured using a Sirius red total collagen detection kit (Chondrex, Redmond, WA) as described [
38]. Briefly, to determine collagen production from fibroblasts in vitro, HFL1 cells seeded into 6-well plates (5 × 10
5cells/well) were starved for 24 h and then stimulated with 1 U/ml thrombin for another 24 h. Cells were harvested and the amount of collagen in cell lysates was determined. To measure collagen deposition in vivo, mouse lung tissues were minced and subjected to the same analysis of the amount of collagen [
30].
Bleomycin-induced mouse pulmonary fibrosis and inhalation delivery of PFD
All animal studies were approved by the Institutional Animal Care and Use Committee at Creighton University. Age- (~11 weeks) and weight- (~20 g) matched RGS2 wild-type (+/+) and knockout (−/−) C57BL/6 J mice [
39] were obtained by in-house breeding and are descendants from the original breeding pairs provided by the University of Texas Southwestern Medical Center with approval from the originator, Dr. David Siderovski (West Virginia University).
Anesthetized mice were administered a single dose of bleomycin sulfate dissolved in 0.9 % saline by intratracheal instillation with an IA-1C MicroSprayer aerosolizer (PennCentury, Wyndmoor, PA) to induce pulmonary fibrosis [
18,
40]. Pulmonary fibrosis was evaluated 21 days later. A lower dose of bleomycin (1.0 U/kg body weight) was used for comparing alterations in lung function and histology between RGS2+/+ and RGS2−/− mice. To compare long-term survival rates, 2.0 U/kg of bleomycin was administered.
To examine therapeutic effects of PFD, mice were administrated bleomycin (1.5 U/kg body weight) as described above and divided into four groups (4 mice/group): bleomycin in RGS2+/+, bleomycin + PFD in RGS2+/+, bleomycin in RGS2−/−, bleomycin + PFD in RGS2−/−. One week after bleomycin administration, PFD solution was administered using a nose-only aerosol exposure tower (Data Sciences International, St. Paul, MN). Aerosol was delivered with the Aeroneb Lab Micropump Nebulizer. Mice received daily 20-minute exposures to 0.4 ml of aerosolized 5 mM PFD or saline (vehicle control) for 14 successive days. Actual administered doses were calculated according to the following formulas [
41]:
$$ Estimation\ of\ inhaled\ volume\ per\ mouse\ (V1) = \left(N/VA\right) \times TV \times RR \times T $$
$$ Estimation\ of\ PFD\ administered\ dose\ D = \left(C \times V1\right)/W $$
In these formulas, N is nebulization rate (300 ml/min), VA is air flow (0.2 L/min), TV is tidal volume (estimated at 170 ml), RR is respiratory rate (estimated at 150 breaths/min), T is the duration of treatment, C is PFD concentration (926 mg/L), and W is mouse weight. Based on these parameters, the estimated administered dose was 0.35 mg PFD/kg daily.
Mouse lung function analysis
Mouse lung dynamic compliance (C
dyn) was recorded using a FinePointe system (Buxco, St. Paul, MN) as previously described [
30,
35].
Histologic analysis
Sections of paraformaldehyde-fixed mouse lungs were analyzed by hematoxylin and eosin (H&E) or Masson’s trichrome staining to assess fibrotic changes in the lungs [
30,
42]. Photos of 20 fields from multiple sections of each mouse were taken at 100X magnification and scored separately by the modified Ashcroft method (score range 0–8) [
43,
44]. The pulmonary fibrosis histopathology score of each mouse is expressed as the mean score of 20 photos.
Statistical analysis
Data are expressed as means ± SEM. Groups were compared using Student’s t test for unpaired observations or two-way ANOVA with the Bonferroni correction for multiple comparisons. p < 0.05 was taken as statistically significant.
Discussion
Although PFD is approved by the FDA for treatment of IPF, its cellular targets and molecular mechanisms of action remain poorly understood. In the present study, we took the novel approach of screening for genes whose expression changed rapidly upon exposure of HFL1 cells to PFD. This screen identified RGS2 as one of the genes most highly upregulated in response to short-term PFD treatment. Quantitative RT-PCR and western blot analyses confirmed PFD induction of RGS2 expression in five other primary human lung fibroblast cell lines isolated from patients without or with IPF, confirming that this is a general phenomenon. This induction of RGS2 expression is a previously unrecognized genomic response to PFD, and our further studies strongly suggest that RGS2 upregulation is a significant contributor to the pulmonary fibrosis protection induced by PFD.
RGS2 inhibits both the amplitude and duration of signals mediated by G
q-coupled GPCRs [
21‐
23]. Interestingly, several G
q-coupled GPCRs and their ligands are important drivers of pulmonary fibrosis, including PAR1, lysophosphatidic acid receptor 1, and endothelin receptors [
25‐
28]. Inhibition of these signals by RGS2 that is upregulated in response to PFD thus provides a mechanistic rationale for the beneficial effects of both PFD and RGS2 in terms of reducing fibrotic responses of lung fibroblasts. Indeed, when HFL1 cells were exposed to thrombin, a protease elevated in bronchial alveolar lavage fluid of IPF patients [
50], PFD treatment or direct overexpression of recombinant RGS2 to levels similar to those induced by PFD treatment can trigger several anti-fibrotic responses in HFL1 cells. For example, PFD treatment or overexpression of RGS2 suppressed thrombin-induced proliferation and differentiation of HFL1 cells, key components of IPF [
37]. Thrombin-induced collagen production, CTGF expression, and gel contraction were also inhibited by PFD treatment or RGS2 overexpression. Altogether, our study demonstrates for the first time that RGS2 is induced by PFD treatment and that it exhibits multiple anti-fibrotic effects. In addition, we found that PFD treatment or RGS2 overexpression significantly reduced thrombin-stimulated intracellular Ca
2+ signaling. Since thrombin elevation of [Ca
2+]
i promotes fibroblast proliferation and differentiation, our study provides a molecular mechanism to explain the anti-fibrotic effects of RGS2.
Interestingly, animal studies suggest an important role of RGS2 in regulation of the progression of IPF. Excess deposition of collagen by fibroblasts in the lung and decreased lung compliance are hallmarks of human IPF and of bleomycin-induced mouse pulmonary fibrosis [
6,
11]. Our data showed a significant increase in collagen deposition in the lungs of bleomycin-treated RGS2 knockout (RGS2−/−) mice compared to that in bleomycin-treated wild-type (RGS2+/+) mice. Because the accumulation of collagen increases lung stiffness, lungs of bleomycin-treated RGS2−/− mice exhibited much lower compliance than lungs of similarly treated RGS2+/+ mice. The RGS2−/− mice also exhibited a decreased survival rate compared to RGS2+/+ mice, consistent with the high rate of mortality for IPF patients. Altogether, our study indicates that endogenous RGS2 itself plays a protective role and that loss of RGS2 augments the development of pulmonary fibrosis.
Animal studies provide further compelling support for the importance of PFD-induced RGS2 upregulation as part of PFD's anti-fibrotic action. PFD treatment significantly upregulated endogenous pulmonary RGS2 expression in RGS2+/+ mice. More importantly, PFD effectively reduced collagen deposition in the lung following bleomycin administration and ameliorated the bleomycin-induced decrease in lung compliance of RGS2+/+ mice. In marked contrast, PFD treatment had no effect in RGS2−/− mice, indicating that upregulated RGS2 is a crucial mediator of the anti-fibrotic effects associated with PFD treatment. Interestingly, RGS2 has been reported to be increased in IPF patient samples [
51], which is consistent with the notion that upregulated RGS2 functions as a potential feed-back regulator of pulmonary fibrosis. Further investigation of the relative expression levels of RGS2 in patients with IPF following PFD treatment will be necessary to determine the clinical importance of RGS2 upregulation in PFD treatment of patients with IPF.
It should be noted that although western blot analyses showed upregulation of RGS2 protein expression in lungs of RGS2+/+ mice after PFD administration, the expression pattern of RGS2 induction in mouse lungs is unclear. We have performed immunohistochemical staining for RGS2 protein in paraffin-embedded mouse lung tissues. However, several RGS2 antibodies we tested showed positive staining in RGS2−/− mice. Thus, better antibodies will be needed to establish in which pulmonary cell types RGS2 up-regulation by PFD occurs in mice or in human patient samples. However, RGS2 is highly expressed in pulmonary fibroblast cells. In the bleomycin-induced pulmonary fibrosis mouse model, RGS2−/− mice had significantly increased pulmonary deposition of collagen compared to RGS2+/+ mice, and this collagen is predominately produced by differentiated pulmonary fibroblast cells, suggesting that endogenous RGS2 in pulmonary fibroblast cells should function as an anti-fibrotic gene. Indeed, overexpression of recombinant RGS2 or PFD treatment significantly reduced thrombin-induced collagen production and expression of CTGF in cultured human lung fibroblast cells. Interestingly, PFD treatment selectively upregulates RGS2 expression without effects on the other 20 RGS gene family members in human pulmonary fibroblast cells. More importantly, PFD treatment also increased RGS2 expression in primary pulmonary fibroblast cells isolated from patients with IPF. Thus, it is likely that induction of RGS2 by PFD in pulmonary fibroblast cells contributes, at least in part, to the anti-fibrotic effects of PFD in vivo.
Because PFD has a short half-life of only 2.5 h following oral administration [
17], the therapeutically effective doses of PFD for human IPF and bleomycin-induced mouse pulmonary fibrosis are very high (>30 mg/kg, daily) [
17,
18]. Such high doses can cause significant side effects that in turn limit therapeutic effectiveness [
16]. Theoretically, lowering the dose of PFD could result in reduction of side effects. In the current study, we administered PFD by nose-only nebulization at a dose of 0.35 mg/kg. Our data show that PFD can be an effective anti-fibrotic agent at much lower concentrations if the drug is administered by inhalation directly to airways, at least in mice. Thus developing an effective inhaled delivery form of PFD could improve its safety and efficacy in IPF patients.
Abbreviations
[Ca2+]i, intracellular Ca2+; BrdU, 5-bromo-2’-deoxyuridine; Cdyn, dynamic compliance; CPHLF, control primary human lung fibroblast; CT, Threshold cycle; CTGF, connective tissue growth factor; DAPI, 4’,6-diamidino-2-phenylindole; DMEM, Dulbecco's modified eagle's medium; DPHLF, diseased primary human lung fibroblast; F12, Ham's F-12 Nutrient Mixture; FBS, fetal bovine serum; FDA, food and drug administration; GPCRs, G protein-coupled receptors; H&E, hematoxylin and eosin; HFL1, human fetal lung fibroblast; IPF, idiopathic pulmonary fibrosis; PAR1, proteinase-activated receptor 1; PFD, pirfenidone; RGS2, regulator of G-protein signaling 2; α-SMA, α-smooth muscle actin
Acknowledgments
We gratefully acknowledge Dr. Carol Feghali-Bostwick at Medical University of South Carolina and Dr. David Siderovski at West Virginia University for providing primary human lung fibroblasts and the original RGS2 knockout breeding mice, respectively.