Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive fibrosing lung disease that is estimated to affect over 5 million people worldwide (1). It is characterized by the accumulation of activated fibroblasts and the excessive and erratic deposition of collagens and other connective tissue matrices in lung, thereby leading to the destruction of gas-exchanging units and, ultimately, respiratory failure. Although advances in the diagnosis and management of this disease have been made over the past 20 years, its cause and the cellular and molecular mechanisms that lead to its development and progression remain unclear. Two drugs, nintedanib and pirfenidone, have recently been shown to slow down the decline in lung function in IPF, but no strategy has been formally proven to improve survival (2). Currently, the average survival of patients with IPF remains at 50% three years after diagnosis.

It is generally considered that IPF is triggered by injury to the lung epithelium resulting in cellular dysfunction, oxidative stress, and the production of profibrotic growth factors, among other mechanisms that drive tissue fibrogenesis. These “repair” mechanisms appear to be exaggerated in aging and, together with continuing injury (e.g., smoking), stem cell exhaustion, and immune dysregulation, contribute to disease progression (3, 4). However, despite the emerging importance of the epithelium, few question the critical role of fibroblasts in IPF, as their proliferation and uncontrolled production of extracellular matrix are responsible for much of the downstream histological and physiological derangements observed in the disease (5). These cells emerge from the proliferation of resident fibroblasts, the transdifferentiation of incoming stem cells, and the mesenchymal transformation of epithelial cells. Far from acting as innocent bystanders, however, activated fibroblasts harvested from IPF lungs demonstrate a profibrotic phenotype characterized by overproduction of growth factors such as transforming growth factor β (TGF-β) and platelet-derived growth factor, and by their ability to promote lung fibrosis when administered to immunosuppressed animals, further emphasizing their role in driving tissue fibrogenesis (6, 7).

Lung fibroblasts also produce molecules with antifibrotic activity, such as cyclooxygenase-2, Thy-1, and Caveolin-1 (8–10). In IPF, however, there is an imbalance in the production of these molecules that favors fibrogenesis. One such counterregulatory factor is the C-X-C motif chemokine 10 (CXCL10, also known as IFN-γ–induced protein 10 or small inducible cytokine B10), which has been implicated in tissue repair (11). CXCL10 is antiangiogenic, is a chemoattractant for immune cells, and has antitumor activity. It acts by binding CXCR3 chemokine receptors and is produced by many cell types, including monocytes, endothelial cells, and fibroblasts, in response to IFN-γ, among other stimulants. Consistent with a role in lung fibrosis, CXCL10 inhibits lung fibroblast migration, and in the bleomycin model of lung fibrosis, CXCL10 expression is decreased, whereas experimental inhibition of this molecule (or inhibition of its receptor) increases pulmonary fibrosis (12, 13). Interestingly, CXCL10 production has been found to be decreased in patients with IPF (13). These observations led to the hypothesis that dysregulation (in this case, downregulation of CXCL10 expression) contributes to the development of IPF, but the factors that control CXCL10 expression in lung remain unclear.

In this issue of the Journal, Coward and colleagues (pp. 449–460) explore how epigenetic mechanisms resulting in histone methylation affect CXC10 gene expression (14). This group previously showed that methylation of histone H3 lysine 27 (H3K27 me3) is involved in repression of the CXCL10 gene and is increased in lung fibroblasts harvested from patients with IPF (15). H3K27 me3 methylation is catalyzed by the histone lysine methyltransferase Enhancer of Zest Homolog 2 (EZH2), which is also increased in IPF. Other histone lysine methyltransferases—G9a and G9a-like protein—catalyze methylation of H3K9 me2/3, and G9a has been shown to methylate H3K27. These observations and the ability of G9a to mediate EZH2 recruitment to gene promoters suggested potential interactions between these pathways. In the current article, the investigators examine this hypothesis. Their work reveals that EZH2 and G9a physically interact with each other at the CXCL10 gene promoter in fibroblasts harvested from patients with IPF, and this interaction promotes CXCL10 repression. Interestingly, reduction of EZH2 using chemical inhibitors or overexpression of specific demethylases reversed the effect, unveiling possible targets for intervention. EZH2- and G9a-mediated histone hypermethylation also appears to contribute to epigenetic signaling in tumor-suppressor genes, again suggesting a mechanistic link between carcinogenesis and fibrogenesis.

Another interesting observation is that inhibition of EZH2 and G9a reduced their activity as well as the methylation of relevant histones. Later, the authors confirmed a physical interaction between these two histone lysine methyltransferases, and the interdependence of their activities. As the authors point out, this appears to be the first report to suggest that a functionally interdependent cross-talk between these two histone lysine methyltransferases regulates the epigenetic repression of antifibrotic genes. Furthermore, inhibition of EZH2 not only reduced methylation, it enhanced histone H3 and H4 acetylation, suggesting that these methyltransferases can affect two important epigenetic regulatory mechanisms. Finally, they tested the effect of TGF-β in these processes, considering its documented role in lung fibrosis. They treated normal lung fibroblasts with TGF-β and observed an increased association of EZH2 and G9a with the CXCL10 promoter, which decreased CXCL10 expression in a manner similar to that observed in IPF fibroblasts. These and other experiments suggest a key role for histone methylation and an interdependent role for EZH2 and G9a in TGF-β–mediated lung fibrosis.

An important strength of this work is the fact that it was performed in primary lung fibroblasts harvested from patients with IPF or from normal lung tissue obtained from organ donors. This not only ensures the evaluation of mechanisms more relevant to the human condition but also emphasizes that the “IPF phenotype” of these cells is sustained ex vivo. Another strength is the sophisticated methodology used in the study, which included a state-of-the art evaluation of histone modifications and their regulation of gene expression. Of course, as highlighted by the authors, a clear limitation of their study is that they explored epigenetic events at the end of the disease process. Unfortunately, this is not a unique problem and is a common feature of most, if not all, studies of humans with IPF. Although this is not the first article in the literature to indicate a role for epigenetic mechanisms and, in particular, histone modifications in the expression of genes linked to IPF pathogenesis (as recently reviewed in References 16–18), it adds to our current understanding of this field and further reveals the complexity of the mechanisms involved. Many questions logically emerge from this study. For example, what other mechanisms contribute to this process? Are similar mechanisms operational in other chronic fibrosing pulmonary disorders? How are these events affected by aging and environmental exposures? Importantly, does this represent a common and targetable mechanism of antifibrotic gene repression in IPF? Other studies pointing to similar mechanisms involved in the regulation of antifibrotic genes (e.g., COX-2) suggest that the latter is indeed the case (8).

While research in IPF continues to uncover new genetic variants that contribute to IPF initiation and progression, this work points to epigenetic modifications as another important pathogenic pathway. It also provides a glimpse into the great complexity of these mechanisms while pointing to reversible pathways that could be targeted at the clinic in the not-too-distant future (19).

References

Section:

ChooseTop of pageReferences <<CITING ARTICLES