LTBP2 is secreted from lung myofibroblasts and is a potential biomarker for idiopathic pulmonary fibrosis

Abstract

Although differentiation of lung fibroblasts into α-smooth muscle actin (αSMA)-positive myofibroblasts is important in the progression of idiopathic pulmonary fibrosis (IPF), few biomarkers reflecting the fibrotic process have been discovered. We performed microarray analyses between FACS-sorted steady-state fibroblasts (lineage (CD45, TER-119, CD324, CD31, LYVE-1, and CD146)-negative and PDGFRα-positive cells) from untreated mouse lungs and myofibroblasts (lineage-negative, Sca-1-negative, and CD49e-positive cells) from bleomycin-treated mouse lungs. Amongst several genes up-regulated in the FACS-sorted myofibroblasts, we focussed on Ltbp2, the gene encoding latent transforming growth factor-β (TGF-β) binding protein-2 (LTBP2), because of the signal similarity to Acta2, which encodes αSMA, in the clustering analysis. The up-regulation was reproduced at the mRNA and protein levels in human lung myofibroblasts induced by TGF-β1. LTBP2 staining in IPF lungs was broadly positive in the fibrotic interstitium, mainly as an extracellular matrix (ECM) protein; however, some of the αSMA-positive myofibroblasts were also stained. Serum LTBP2 concentrations, evaluated using ELISA, in IPF patients were significantly higher than those in healthy volunteers (mean: 21.4 compared with 12.4 ng/ml) and showed a negative correlation with % predicted forced vital capacity (r = -0.369). The Cox hazard model demonstrated that serum LTBP2 could predict the prognosis of IPF patients (hazard ratio for death by respiratory events: 1.040, 95% confidence interval: 1.026-1.054), which was validated using the bootstrap method with 1000-fold replication. LTBP2 is a potential prognostic blood biomarker that may reflect the level of differentiation of lung fibroblasts into myofibroblasts in IPF.

Keywords: Biomarker; Idiopathic pulmonary fibrosis; LTBP2; Microarray; Myofibroblast.

Figure 1. FACS and microarray analyses

(A) Representative images of FACS analysis for steady-state fibroblasts (FB) from untreated mice (UM) and activated FB/myofibroblasts (MFB) from bleomycin-treated mice (BM). (B) Fluorescence intensities of FITC-conjugated anti-αSMA antibody in steady-state FB, activated FB, and MFB (n=4 per group). Statistical comparisons by t test were performed using the value of steady-state FB as a control. *P<0.05. (C) Heatmap generated by microarray analyses. Genes up-regulated in MFB were selected according to the following criteria: (i) five-fold greater than in steady-state FB; (ii) five-fold greater than in total lung homogenates; and (iii) two-fold greater than the mean value of all genes in MFB. Conversely, genes up-regulated in steady-state FB were defined according to the following criteria: (i) five-fold greater than in MFB; (ii) five-fold greater than in total lung homogenates; and (iii) two-fold greater than the mean value of all genes in steady-state FB. The listed genes are the top 20 genes with high signals in each group and are available in the NCBI Reference Sequence.

Figure 2. Results of RT-PCR using mouse lung cells isolated by FACS

(A,B) Ltbp2 and Acta2 mRNA levels in mouse lung steady-state fibroblasts (FB), activated FB, and myofibroblasts (MFB) (n=3 per group). Statistical comparisons by t test were performed using the value of steady-state FB as a control. *P<0.05, **P<0.01, ***P<0.001. (C) Ltbp2 mRNA levels in lineage cells from untreated and bleomycin-treated mouse lungs (n=3 per group). There were no statistically significant differences between the two mouse groups (t test).

Figure 3. Results of RT-PCR and ELISA using human lung fibroblasts (TIG-1-20 and HPF-a)

(A,B) LTBP2 and Acta2 mRNA levels in human lung fibroblasts with or without 48-h external administration of TGF-β1 (n=3 per group). Statistical comparisons by t test were performed using the value of the untreated group as a control. *P<0.05, **P<0.01, ***P<0.001. (C) Protein concentrations, measured by ELISA, of LTBP2 in culture supernatant with or without 48-h external administration of TGF-β1 (n=3 per group). Statistical comparisons by t test were performed using the value of the untreated group as a control. *P<0.05, **P<0.01, ***P<0.001.

Figure 4. Representative immunohistochemistry images of lungs from a healthy subject and a patient with IPF who received surgical lung biopsy

(A,B) Lung section from a healthy subject, stained by Hematoxylin–Eosin (HE) and anti-LTBP2 antibody. (C–F) Images of fibrotic lesions in an IPF lung, stained by HE and anti-LTBP2 antibody. LTBP2 was broadly positive in the fibrotic interstitium but almost negative in the body of fibroblastic foci (arrows), as well as in epithelial cells, interstitial inflammatory cells, and alveolar macrophages. (G,H) Images of fibrotic lesions in an IPF lung, stained by Elastica van Gieson (EVG) and anti-LTBP2 antibody.

Figure 5. Representative immunofluorescence images of a biopsied lung of IPF (color: green = αSMA; red = LTBP2; white = lineage markers (CD45, CD324, CD31, podoplanin, and h-caldesmon))

(A) (Fibroblastic focus): aggregated αSMA-positive, LTBP2-negative, and lineage-negative myofibroblasts (arrows) and αSMA-negative, LTBP2-negative, and lineage (CD324)-positive epithelial cells (triangles). (B) (Fibrotic interstitium): a non-aggregated αSMA-positive, LTBP2-positive, and lineage-negative myofibroblast (arrow). (C) (Fibrotic interstitium): αSMA-positive, LTBP2-negative, and lineage (h-caldesmon)-positive smooth muscle cells at an alveolar wall (arrows). (D) (Bronchiole): αSMA-positive, LTBP2-negative, and lineage (h-caldesmon)-positive smooth muscle cells around a bronchiole (arrows); αSMA-negative, LTBP2-negative, and lineage (CD45)-positive alveolar macrophages (triangles). (E) (Blood vessel) αSMA-negative, LTBP2-negative, and lineage (CD31)-positive vascular endothelial cells (arrows). (F) (Lymphatic vessel): αSMA-negative, LTBP2-negative, and lineage (podoplanin)-positive lymphatic endothelial cells (arrows). All scale bars = 10 μm.

Figure 6. Serum LTBP2 concentrations evaluated by ELISA

(A) Comparison of serum LTBP2 concentrations between patients with IPF (n=116) and healthy controls (n=31). Statistical comparison was performed using the Mann–Whitney U test. (B) ROC curve to discriminate patients with IPF from healthy controls.

Figure 7. Association of LTBP2 with acute exacerbation (AE) of IPF

(A) Comparison of serum LTBP2 concentrations between patients with (n=24) and without (n=92) AE of IPF at IPF diagnosis or within 6 months of IPF diagnosis. Statistical comparison was performed using the Mann–Whitney U test. (B–D) Immunohistochemistry images of diffuse alveolar damage in an autopsied lung from a patient with AE of IPF, stained by Hematoxylin–Eosin, anti-LTBP2 antibody, and anti-αSMA antibody, respectively. Intra-alveolar LTBP2-positive hyaline membranes (arrows) and edematous alveolar walls with many αSMA-positive cells were observed. Hyaline membranes were not stained when a normal rabbit polyclonal IgG was used as the primary antibody. (E–G) Immunofluorescence images of the same lung section (color: green = αSMA; red = LTBP2; white = lineage markers (CD45, CD324, CD31, podoplanin, and h-caldesmon)). (E) LTBP2 positivity was found in or around interstitial myofibroblasts (triangles) and in the hyaline membrane (arrow). (F) In a thickened alveolar wall, diffuse positivity for both αSMA and LTBP2 was observed. (G) Smooth muscle cells (h-caldesmon-positive cells) at a fibrotic interstitium showed positivity for LTBP2 as well as αSMA.

Figure 8. Prognostic analyses using serum LTBP2 concentrations

(A,B) Time-dependent ROC curves for all-cause death and death by respiratory events, respectively. Integrated AUC (iAUC) was calculated for each cut-off value of serum LTBP2 concentration. The cut-off value with the highest iAUC was estimated commonly as 18 ng/ml. (C,D) Survival curves drawn by the Kaplan–Meyer method using the cut-off value of 18 ng/ml for all-cause death and death by respiratory events, respectively. The prognostic difference was statistically significant (P<0.001 by log-rank test in both situations).