Hypermethylation of PTGER2 confers prostaglandin E2 resistance in fibrotic fibroblasts from humans and mice

Abstract

Idiopathic pulmonary fibrosis (IPF) is a devastating lung disease that is characterized by excessive proliferation of fibroblasts. The lipid mediator prostaglandin E2 (PGE2) has the capacity to limit fibrosis through its inhibition of numerous functions of these fibroblasts; however, in the setting of fibrosis, fibroblasts have been shown to be resistant to PGE2. We have linked such resistance to decreased expression levels of the E prostanoid 2 (EP2) receptor. In this study, in fibroblasts from both mice and humans with pulmonary fibrosis, we show that DNA hypermethylation is responsible for diminished EP2 expression levels and PGE2 resistance. Bisulfite sequencing of the prostaglandin E receptor 2 gene (PTGER2) promoter revealed that fibrotic fibroblasts exhibit greater PTGER2 methylation than nonfibrotic control cells. Treatment with the DNA methylation inhibitors 5-aza-2'-deoxycytidine and zebularine as well as DNA methyltransferase-specific siRNA decreased PTGER2 methylation, increased EP2 mRNA and protein expression levels, and restored PGE2 responsiveness in fibrotic fibroblasts but not in nonfibrotic controls. PTGER2 promoter hypermethylation was driven by an increase in Akt signal transduction. In addition to results described for the PTGER2 promoter, fibrotic fibroblasts also exhibited increased global DNA methylation. These findings demonstrate that the down-regulation of PTGER2 and consequent PGE2 resistance are both mediated by DNA hypermethylation; we identified increased Akt signal transduction as a novel mechanism that promotes DNA hypermethylation during fibrogenesis.

Figure 1

DNA methylation of Ptger2 promoter in mouse lung fibroblasts. Fibroblasts from mice treated intratracheally with saline (n = 5) or bleomycin (n = 10) were cultured and assayed for Ptger2 promoter methylation by bisulfite conversion and pyrosequencing. A: Schematic of the mouse Ptger2 gene and promoter, which has 13 CpG sites in the 420 bp upstream of the transcription start site. B: Percent methylation at each CpG site is shown graphically in cells from saline- and bleomycin-treated animals. The mean methylation values at all 13 sites for each group are shown in the inset. *P < 0.05.

Figure 2

Methylation of PTGER2 promoter in IPF fibroblasts. A: Schematic of the human PTGER2 gene and promoter, showing 23 CpG dinucleotides in the first 150 bp upstream of the transcriptional start site. B: DNA methylation analysis by bisulfite conversion and pyrosequencing of the 23 CpG dinucleotides was performed in four different nonfibrotic and five different IPF fibroblast cell lines; the mean ± SE values for the group of four nonfibrotic lines and for each of the IPF lines are depicted. Methylation levels represent two to four independent experiments for each line, with mean ± SE shown for each line. The mean methylation values at all 23 sites for each line are shown in the inset. The mean methylation at all 23 sites for the groups of nonfibrotic and fibrotic cell lines are also shown in the adjacent inset.

Figure 3

EP2 mRNA and protein expression in fibroblasts treated with DNA methylation inhibitors. A: Lung fibroblasts from saline- and bleomycin-treated mice were cultured and treated for 72 hours with or without 5-aza-2′-deoxycytidine (5-aza) or zebularine (zeb). EP2 mRNA expression, relative to β-actin, was measured by real-time RT-PCR with levels from untreated cells in the saline group normalized to 1 (n = 4–6 for each treatment). **P < 0.05 relative to untreated cells in saline group, *P < 0.05 relative to untreated cells in bleomycin group. B: EP2 protein expression was assayed by immunoblotting, with mean densitometry relative to GAPDH shown in graphs underneath the representative blot (n = 4 for both saline and bleomycin groups). *P < 0.05. C: Fibroblasts cultured from nonfibrotic lung tissue and from the lungs of patients with IPF were treated for 72 hours with 5-aza and assayed for expression of EP2 mRNA by real-time RT-PCR. Expression of different IPF cell lines, relative to levels in untreated nonfibrotic cells that were set at 1, is shown. D: Representative immunoblot of EP2 in nonfibrotic (n = 3) and IPF fibroblasts (3 lines with n = 2–4 for each line) is shown, with densitometry values normalized to untreated cells depicted in the bar graph. *P < 0.05.

Figure 4

Expression of EP2 in fibrotic lung fibroblasts treated with DNMT-specific siRNA. Fibroblasts from bleomycin-injured mice were treated with DNMT-specific siRNA and assayed at 72 hours for EP2 mRNA (A, n = 3) and protein (B) expression. Representative blot of three independent experiments is shown. C: Fibroblasts from patients with IPF were treated with DNMT-specific siRNA and assayed at 72 hours for EP2 protein expression. Representative blot from two cell lines is shown with mean densitometric values of three different cell lines shown graphically in D. *P < 0.05.

Figure 5

Effect of DNA methylation inhibitors on PGE₂ suppression of collagen expression and cell proliferation in lung fibroblasts. Lung fibroblasts were pretreated for 72 hours with 5-aza-2′-deoxycytidine (5-aza, μM) or zebularine (zeb, μM). Cells were then treated with PGE₂ (500 nmol/L) for 20 hours and assayed for collagen I expression by immunoblot analysis or proliferation. A: Representative immunoblot of cells from saline- or bleomycin-treated mice (n = 4 each) is shown. C: Densitometric analysis of collagen immunoblots normalized to α-tubulin (n = 4) in bleomycin fibroblasts treated ± PGE₂ were performed, with results expressed as a percentage of no-PGE₂ control. B: Representative immunoblot from nonfibrotic (n = 3) and IPF fibroblasts (three lines with n = 2–4 for each line) is shown, with mean effects of PGE₂ as determined from densitometic analysis shown in D. E: Mean proliferative response to PGE₂ in bleomycin fibroblasts is expressed as a percentage of the no-PGE₂ control (n = 3). F: Mean proliferative response to PGE₂ in IPF fibroblasts is expressed as a percentage of the no-PGE₂ control (n = 4). *P < 0.05 for all graphs.

Figure 6

Ptger2 DNA methylation in Pten^−/− fibroblasts and in fibrotic cells treated with a PI3K inhibitor or Akt inhibitor. A: Bisulfite sequencing of the Ptger2 promoter was performed in Pten^−/− and wild-type murine embryonic fibroblasts (MEF). The percent methylation of individual CpG sites is depicted. *P < 0.05. B: Fibroblasts from mice treated with bleomycin were cultured and treated with the PI3K inhibitor, LY294002 (10 μmol/L), or the Akt inhibitor, Akt-I (5 μmol/L). Methylation of the Ptger2 promoter was determined by bisulfite pyrosequencing, with mean methylation of the 13 CpG sites shown (n = 3). *P < 0.05. C: Bisulfite sequencing of the human PTGER2 promoter was performed in IPF fibroblasts treated for 48 hours with and without the PI3K inhibitor, LY294002 (10 μmol/L), or the Akt inhibitor, Akt-I (5 μmol/L). Shown is the percent methylation of individual CpG sites from a representative line, with the mean methylation from two IPF lines shown on the right.

Figure 7

Global DNA methylation levels in fibrotic and nonfibrotic lung fibroblasts. A: Fibroblasts from saline- (n = 8) and bleomycin-treated (n = 11) mice were cultured and assayed for global DNA methylation as described in Materials and Methods. B: Global DNA methylation levels were assayed in patient-derived nonfibrotic (n = 3) and IPF (n = 5) fibroblasts. *P < 0.05.