A role for the receptor for advanced glycation end products in idiopathic pulmonary fibrosis

Abstract

Idiopathic pulmonary fibrosis (IPF) is a severely debilitating disease associated with a dismal prognosis. There are currently no effective therapies for IPF, thus the identification of novel therapeutic targets is greatly needed. The receptor for advanced glycation end products (RAGE) is a member of the immunoglobulin superfamily of cell surface receptors whose activation has been linked to various pathologies. In healthy adult animals, RAGE is expressed at the highest levels in the lung compared to other tissues. To investigate the hypothesis that RAGE is involved in IPF pathogenesis, we have examined its expression in two mouse models of pulmonary fibrosis and in human tissue from IPF patients. In each instance we observed a depletion of membrane RAGE and its soluble (decoy) isoform, sRAGE, in fibrotic lungs. In contrast to other diseases in which RAGE signaling promotes pathology, immunohistochemical and hydroxyproline quantification studies on aged RAGE-null mice indicate that these mice spontaneously develop pulmonary fibrosis-like alterations. Furthermore, when subjected to a model of pulmonary fibrosis, RAGE-null mice developed more severe fibrosis, as measured by hydroxyproline assay and histological scoring, than wild-type controls. Combined with data from other studies on mouse models of pulmonary fibrosis and human IPF tissues indicate that loss of RAGE contributes to IPF pathogenesis.

Figure 1

mRAGE is highly expressed in the lung. Membrane fractions from each of the indicated tissues were prepared from untreated wild-type C57BL/6 mice and analyzed by Western blot for mRAGE expression. The PVDF membrane was then stripped and reprobed for β-actin expression as a loading control. Note that mRAGE expression is highest in the lungs.

Figure 2

mRAGE and sRAGE expression are decreased after asbestos injury. C57BL/6 mice were treated with 0.1 mg of crocidolite asbestos or titanium dioxide (TiO₂) as an inert particulate control. Fourteen days later, lungs were extracted to obtain both membrane and soluble protein fractions. a: mRAGE, sRAGE, and aquaporin-5 expression were analyzed by Western blot analysis. PVDF membranes were stripped and reprobed for β-tubulin as a loading control. b: Shown to the right is the densitometric analysis, where the protein of interest band intensities are normalized to β-tubulin band intensities for each lane. RNA was isolated from the lungs of mice 14 days after they were treated with asbestos and compared with normal lungs from mice treated with titanium dioxide for 24 hours. RAGE mRNA expression was measured by real-time PCR and normalized to GAPDH. c: Results are reported as a percent relative quantity compared to the TiO₂-treated group. A 1.5-fold decrease in RAGE expression was seen after asbestos treatment. **P < 0.01 and *P < 0.05 compared to TiO₂-treated controls.

Figure 3

RAGE transcripts and mRAGE/sRAGE protein are down-regulated in IPF lungs. a and b: Microarray analysis was performed on human IPF lung samples or control lung samples to compare expression of RAGE transcripts between IPF and control lungs. a: Summary of microarray data. Each point represents a single gene, plotted by its average expression in control lungs versus its average expression in IPF lungs. Points in color represent genes whose difference in expression in control versus IPF lungs reached significance. The point representing RAGE transcripts is indicated. b: Plot demonstrating the RAGE transcript level in each patient sample (represented by a single point). RAGE transcript levels are significantly decreased in IPF lungs compared to control lungs (P = 0.0000054). c: Western blot analysis on a different set of patient samples demonstrates a reduction in mRAGE (*P < 0.05) and sRAGE (**P < 0.01) protein levels in fibrotic areas of IPF lungs. Densitometry is shown with band intensities of mRAGE and sRAGE normalized to β-actin as a loading control.

Figure 4

Aged RAGE knockout (KO) mouse lungs spontaneously develop features of pulmonary fibrosis. Lung sections from 19- to 24-month-old wild-type (a, c, e) and RAGE KO (b, d, f) mice were subjected to immunohistochemical analysis for RAGE (a, b) and collagen type I (c, d) as well as Sirius Red staining for total collagen (e, f). RAGE knockout lungs lack immunoreactivity for RAGE (b) and show increased staining for collagen (d, f). g: Right lungs from 48-week-old wild-type (n = 3) or RAGE knockout mice (n = 4) were subjected to hydroxyproline analysis as an indicator of the extent of fibrosis. Forty-eight-week-old RAGE KO mice have significantly increased levels of hydroxyproline compared to age-matched wild-type controls (*P < 0.05).

Figure 5

RAGE knockout (KO) mice develop more severe fibrosis after asbestos treatment. RAGE KO mice and C57BL/6 wild-type mice were treated with 0.1 mg of crocidolite asbestos or titanium dioxide (TiO₂) as an inert particulate control. Fourteen days later, lungs were either inflation-fixed with 10% formalin for histology (n = 3 per group) or removed and dried for hydroxyproline quantification (n = 5 per group). Light microscopic analysis demonstrates increased alveolar thickening in the RAGE-null mice (b, d) compared to wild-type (a, c). e: Histological scoring by a pathologist blinded to genotype and treatment demonstrates more severe fibrosis in the asbestos-treated RAGE KO group compared to wild-type controls. No fibrosis was appreciated in either of the TiO₂-treated groups. f: Asbestos treatment resulted in increased fibrosis in the RAGE KO mice as determined by hydroxyproline levels compared to wild-type mice. g: RAGE KO mice were subjected to asbestos treatment along with intraperitoneal injections of either bovine sRAGE or bovine serum albumin (50 μg/day). There was no difference in the degree of fibrosis between the groups (n = 5 per group) as measured by total lung hydroxyproline. **P < 0.01 and *P < 0.05 compared to TiO₂-treated controls. Scale bars: 30 μm (a, b); 6 μm (c, d).