Towards a global initiative for fibrosis treatment (GIFT)

Genetic network

IPF is a complex genetic disorder [6–8]. At least eight different genes (MUC5B, TERT, TERC, RTEL1, PARN, SFTPC, SFTPA1 and SFTPA2) and 11 gene variants in novel loci have been associated with the disease [9]. A common gain-of-function promoter variant in the MUC5B gene (rs35705950) accounts for 30–35% of the risk for developing IPF and can potentially help identify subjects with a higher risk or patients with preclinical pulmonary fibrosis [9]. Mechanistically, this MUC5B variant appears to decrease mucociliary clearance, which can then enhance cell injury and alter wound repair [9].

Another important hub in the genetic IPF network relates to mechanisms of cell ageing. The discovery of telomerase and telomere functions received the Nobel Prize in Medicine in 2009 [10], and the role of telomerase and telomere attrition in ageing and other diseases has been well established [11, 12]. Telomere length decreases with cell division (i.e. with physiological ageing). If telomeres are shortened, cells age. Conversely, if telomerase activity is high, telomere length is better maintained and cellular senescence is delayed. For instance, this is the case for cancer cells, which can be considered immortal. In IPF, telomerase gene mutations and reduced telomere length are highly prevalent [13, 14]. Telomerase dysfunction and telomere shortening are associated with increased cell senescence and cell damage [15]. Interestingly, telomerase gene mutations have been associated with poor prognosis in IPF [13, 14], and the severity of telomere attrition correlates with disease progression and outcome [16]. Further, age-related cell perturbations found in epithelial cells and fibroblasts of patients with IPF are not present in normal lungs of similar age individuals [17]. However, while it is well established that familial forms of IPF are linked to dysfunctional telomerase activity and mutations in surfactant proteins, the relationship of ageing to the majority of sporadic IPF cases is currently under study. Finally, in addition to telomere attrition, there are other genetic footprints associated with accelerated ageing, including genomic instability and epigenetic changes [18], which can also be relevant to the pathobiology of IPF. Interventions to modify these age-related characteristics are currently being investigated in IPF [19], since the lung-ageing process appears to be associated with the generation of a vulnerable alveolar epithelium, as well as a reduction of enzymes involved in telomere maintenance [20].

Cellular network

Traditionally, the two cellular actors considered most relevant in IPF were alveolar epithelial cells (AECs) and myofibroblasts, respectively. Recent research applying single cell sequencing approaches provides evidence for the existence of several subpopulations of AECs and a variety of interstitial fibroblasts and other mesenchymal cells. Furthermore, mesenchymal stem cells (MSCs) have also been shown to be important cellular actors involved in the pathogenesis of lung fibrosis.

AECs are essential for normal lung function, forming the tight functional barrier required for gas exchange [21]. Under normal conditions, alveolar type II (ATII) cells act as progenitor cells that initiate alveolar epithelial repair and restoration, giving rise to either new ATII cells or differentiation into alveolar type I cells [22]. However, activated AEC's can participate in the fibrogenic response by secreting mesenchymal proteins [23]. In IPF, reprogramming of these cells results in a number of different phenotypes and functions, which include proliferation and bronchiolisation, apoptosis, and acquisition of mesenchymal features [24, 25].

Fibroblasts and myofibroblasts are considered the key effector cells in the fibrogenic response in IPF and a hub within the cellular network because they are responsible for the large amount of ECM production in the fibrotic tissue [26]. Yet, there are still many unanswered questions in relation to fibroblasts. For instance, we do not know if there is only one specific phenotype of fibrotic fibroblast or a variety [27], what the origin may be of the accumulating myofibroblasts in the IPF lung, and the identity of the main pathways driving proliferation and differentiation of these heterogeneous cells [28]. From a network perspective, given that fibroblasts in the alveolus live in close contact with AECs, it is crucial to understand what mechanisms link epithelial injury and fibroproliferation.

With particular importance for the systems biology and cross-network perspective, alveolar macrophages seem to be particularly affected by the pro-fibrotic environment of IPF lungs, which deregulates functionality in wound-healing and repair [29], leading to studies on the downstream implications of high levels of oxidised mitochondria.

Finally, another key consideration relates to the role of MSCs in IPF [30]. As discussed earlier, mechanisms associated with ageing participate in the pathobiology of IPF. Part of this ageing process includes the exhaustion of MSCs and the changes in the function of bone marrow (BM)-MSCs that depend on age [17, 31]. Interestingly, harvested aged MSCs are less effective in preventing fibrotic changes than MSCs from young animals [32]. In addition, MSCs from IPF patients have smaller mitochondria and undergo accelerated senescence [18, 31]. The functional and clinical implications of these observations merit further investigation [18].

Metabolic regulation network

Metabolism is a key player in biological complexity [33]. For example, in cancer it is well established that rapid cell proliferation is associated with increased glucose uptake (which can be detected by increased uptake of 18F-FDG (2-fluoro-2deoxy-d-glucose) on positron emission tomography (PET)) [34]. IPF lungs also present increased uptake of 18F-FDG [35, 36], and recent studies have shown that metabolic reprogramming plays a key role in fibroproliferation and myofibroblast differentiation [37, 38]. Classically, several growth factors, cytokines and hormones have been reported to act in cell-to-cell and cell-to-extracellular matrix (ECM) cross-interactions, inducing different cell metabolic and fibrotic changes. Furthermore, mechanical ECM properties, such as increased stiffness, also contribute to influencing cell phenotype. More recently, a number of hallmarks of ageing such as accumulation of misfolded proteins and dysfunctional mitochondria, dysregulation of miRNA expression, and deficient autophagy have been described to contribute to increasing the pro-fibrotic response and modifying cell behaviour and metabolism [12, 18].

Environmental network

Cells and tissues (hence gene, proteins and metabolites) are constantly exposed to dynamic micro- and macro-environmental changes (e.g. smoking, the microbiome, pollution and/or gastro-oesophageal reflux (GOR)) that can modulate their interaction, changing the endotype and, consequently, the phenotype [39–41]. Smoking exposure increases the probability of the associated emphysema. The treatment of GOR is recommended in the updated IPF guidelines since, when present, it may increase AEC cell injury [2, 39]. The most recent advances in the knowledge of microbiome postulate that the bacterial signature induces a host response, which may play a role in disease behaviour and therefore represent another target to prevent or treat [41].

Standard operating procedures

Interoperability of diverse data is critical. Current ongoing studies, such as PROFILE [45] and the European IPF network registry [46] collect different biological samples to work collaboratively on IPF diagnosis and monitoring, and are already using a shared set of predefined rigorous standard operating procedures (SOPs).

Cooperative biobank

Accessing a large number of biological samples sampled and stored according to SOPs, particularly those obtained in the early phases of the disease, has the potential to identify novel key pathobiological mechanisms and biomarkers that facilitate the discovery of new treatments to prevent disease progression. By combining efforts in sample collection at different academic centres, a larger and more comprehensive biobank could be made available to all IPF researchers, after appropriate assessment of the scientific value and priority of the project proposed [47]. The Global IPF Collaborative Network (www.ucdenver.edu/academics/colleges/medicalschool/departments/medicine/globalipf/Pages/GlobalIPF.aspx) with a DNA consortium is a successful example of this type of initiative [48], with various types of IPF samples (e.g. surgical lung biopsies, cryobiopsies and cryopreserved cells).

Education and support on novel bioinformatics techniques

Systems biology approaches rely heavily on novel bioinformatics analyses. Most basic and clinical research centres do not yet have the necessary educational framework and bioinformatics expertise in place to take advantage of rapidly evolving bioinformatics approaches, which is also crucial for interpretation of results.