Lung tissues in patients with systemic sclerosis have gene expression patterns unique to pulmonary fibrosis and pulmonary hypertension

Abstract

Objective: Pulmonary complications, including pulmonary fibrosis (PF) and pulmonary arterial hypertension (PAH), are the leading cause of mortality in patients with systemic sclerosis (SSc). The aim of this study was to compare the molecular fingerprint of lung tissue and matching primary fibroblasts from patients with SSc with that of lung tissue and fibroblasts from normal donors, patients with idiopathic pulmonary fibrosis (IPF), and patients with idiopathic pulmonary arterial hypertension (IPAH).

Methods: Lung tissue samples were obtained from 33 patients with SSc who underwent lung transplantation. Tissues and cells from a subgroup of SSc patients with predominantly PF or PAH were compared to those from normal donors, patients with IPF, and patients with IPAH. Microarray data were analyzed using efficiency analysis for determination of the optimal data-processing methods. Real-time polymerase chain reaction and immunohistochemistry were used to confirm differential levels of messenger RNA and protein, respectively.

Results: Consensus efficiency analysis identified 242 and 335 genes that were differentially expressed in lungs and primary fibroblasts, respectively. SSc-PF and IPF lungs shared enriched functional groups in genes implicated in fibrosis, insulin-like growth factor signaling, and caveolin-mediated endocytosis. Gene functional groups shared by SSc-PAH and IPAH lungs included those involved in antigen presentation, chemokine activity, and interleukin-17 signaling.

Conclusion: Using microarray analysis on carefully phenotyped SSc and comparator lung tissues, we demonstrated distinct molecular profiles in tissues and fibroblasts from patients with SSc-associated lung disease compared to idiopathic forms of lung disease. Unique molecular signatures were generated that are disease specific (SSc) and phenotype specific (PF versus PAH). These signatures provide new insights into the pathogenesis and potential therapeutic targets of SSc-related lung disease.

Figure 1

Gene expression profiles of SSc, IPF and IPAH lung tissues. (A) Efficiency analysis identified 242 genes with a consensus score greater than 50% by the top 20% performing methods. SSc samples were ordered from left to right by lowest (severe PF, no PAH) to highest FVC (no PF, severe PAH) as indicated in Table 1. All genes with J5 and fold change scores in each group can be viewed in supplement table 2. (B) The cluster patterns reveal genes that are up-regulated in SSc and IPF (Cluster 1), up-regulated in SSc-PF and IPF (Cluster 2), and up-regulated in SSc-PAH and IPAH (Cluster 3). Clusters of genes were also down-regulated in SSc-PF and IPF (Cluster 4) and down-regulated in all SSc and IPF (Cluster E). (C) Venn diagrams show number of shared and unique genes in SSc-PF, SSc-PAH, IPF and IPAH groups. Shared and unique genes are shown in supplement tables 5–9. (D) Unsupervised clustering of all lung samples using all 18,630 genes show clustering of samples by disease phenotype (PF or PAH).

Figure 2

Gene expression profiles of SSc, IPF and IPAH fibroblasts. (A) Efficiency analysis identified 335 genes with a consensus score greater than 50% by the top 20% performing methods. All genes and fold change scores in each group can be viewed in supplement table 7. (B) The clustering pattern shown in finer detail on the right panel shows clusters of genes that are up-regulated in IPAH fibroblasts (Cluster 6), in IPF, SSc-PF and IPAH fibroblasts (Cluster 7), and in IPF and SSc-PF (Cluster 8). Clusters of genes were also down-regulated in IPAH fibroblasts (Cluster 9) and coordinately down-regulated in SSc and IPF fibroblasts (Cluster 10). (C) Venn diagrams show the number of shared and unique genes in SSc-PF, SSc-PAH, IPF and IPAH fibroblast groups. Shared and unique genes are listed in supplement tables 14–18. (D) Unsupervised clustering of fibroblast samples using all 18,630 genes.

Figure 3

Enriched functional groups of SSc, IPF and IPAH lungs (panel A) and fibroblasts (panel B). Ingenuity Pathway Analysis and Gene Ontology Tree Machine software programs identified enriched function groups in gene expression patterns of SSc-PF, SSc-PAH, IPF, and IPAH fibroblasts. Functional classes that have a –log(P-value) greater than 1.3 (or P-value of 0.05 or less), as indicated by the dotted line, signifies impacted pathways that may be involved in disease process.

Figure 4

Validation of genes differentially expressed in lung tissues. A: Messenger RNA levels were analyzed by real time PCR. Each sample was performed in triplicate and the average was normalized to b-actin mRNA levels. The relative expression values are displayed as the fold change for each gene relative to normal lungs. B: Localization of IGFBP7, secretoglobin 1a1 (SCGB1a1) and metallothionein in lung tissues (red-brown stain). IGFBP7 has high expression in fibrotic tissues (SSc-PF and IPF) and to a lesser degree in SSc-PAH lungs. SCGB1a1 was detected in all disease groups, but not normal lungs. Metallothionein was detected in SSc-PF airway epithelium and in SSc-PAH and IPAH lungs, but not IPF or normal lungs. Images were taken at 400x.

Figure 5

Validation of genes differentially expressed fibroblasts. A: Real time PCR was performed on mRNA extracted from primary pulmonary fibroblasts. Samples were analyzed in triplicate and average values were normalized to b-actin mRNA levels. Relative expression values are displayed as the fold change compared to normal fibroblasts. B: Western Blot of fibroblast lysates reveals differential protein levels of secreted Frizzled-related protein-1 (sFRP-1), four-and-a-half-LIM-domain-2 (FHL2) and lysyl oxidase.