Iron mobilisation pattern in the airway epithelium: tipping the immune balance in asthma

To the Editor:

The barrier functions of the airway epithelium are centrally involved in regulating tissue homeostasis in the lung, and abnormal responses of the airway epithelium to the environment, including metal exposure, underlies the rise in allergic conditions and chronic airway diseases [1, 2]. Notably, injury-induced epithelial cell apoptosis and injury with resultant lack of normal restitution of epithelial barriers are increasingly recognised as critical events in the pathogenesis of airway diseases such as asthma [1, 3]. Evidence accumulated in recent years points to a connection between iron regulation and asthma pathogenesis [4, 5]. However, the role of iron in asthmatic airway epithelium remains incompletely understood.

As a starting point for investigating the potential role of iron level and regulation in the asthmatic airway epithelium, we identified 14 essential iron transport-related genes in the airway epithelium through combined UniProt database screening and “iron transport” genesets annotated by GeneOntology. Next, we assessed their expression to reflect iron metabolism by integrating different omics data sources. The data expression profiles of asthma were searched in the Gene Expression Omnibus (GEO) database, and two independent mRNA datasets of research on the airway epithelium in human asthma, GSE67472 and GSE43696, respectively, were selected. Differential mRNAs were screened in airway epithelial samples obtained from biopsy in patients with and without asthma. By probing the GSE67472 dataset which included 43 healthy subjects and 62 patients with asthma, we were surprised to find that all upregulated genes (TFR2, SLC39A8 and SLC39A14) were associated with iron uptake. Intriguingly, an external validation in the GSE43696 dataset (comprising 108 subjects) was performed, and the same conclusions were obtained (figure 1a). The aggregate of the above results together suggests that an iron “input mode” exists in the asthmatic bronchial epithelium. Our attempts (data not shown) to detect iron levels in the airway epithelium using Prussian blue staining were unsuccessful (no colouration in the epithelium). Variability in tests of iron content has been an issue in the field, and direct evidence on this issue is still lacking.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Effects and mechanisms of perturbed iron homeostasis on airway epithelium. a) Differential expression analysis of iron transport-related genes in discovery and validation datasets (GSE67472 and GSE43696). b) The iron-overload and -reversal cellular model was constructed using ferric ammonium citrate (FAC) combined with/without deferoxamine (DFO) incubated for 24 h. c) The combined analysis yielded pathway–gene network graphs. d) Protein–protein interaction network showing asthma-associated immune and hub genes regulated by iron. TNF: tumour necrosis factor; IL-17: interleukin-17. *: p<0.05, **p<0.01 versus the control group.

To evaluate the effects of the disturbances in iron metabolism on airway epithelium, epithelial cell iron-overload and -reversal models were respectively constructed (figure 1b) using ferric ammonium citrate (FAC) with/without deferoxamine (DFO) in human bronchial epithelial (HBE) cells (obtained from Shanghai Fuheng Biology Co., Ltd, Shanghai, China). The model was optimised by parameters including different concentrations (10 μM, 50 μM or 100 μM) of iron supplement (FAC) and DFO, and incubation time, to reach overload and followed complete reversal. Intracellular total iron and ferrous ion levels were quantified using a colourimetric assay (Elabscience, Wuhan, China). As can be seen in figure 1b, FAC significantly influenced the total intracellular content of iron and ferrous iron. It can be seen in figure 1b that DFO administration (100 μM) significantly reversed the concentration changes (including total iron and ferrous ions) after FAC treatment. RNA sequencing among the three groups was performed to gain insights into the underlying molecular mechanisms. The trending analysis indicated that most data points after screening (p<0.05; |FC|>1.2) varied dynamically over iron concentration. These “iron-responsive” genes (IRGs) exhibited dynamic expression as cellular iron concentrations changed. The combined analysis yielded 473 results, including 225 iron-activated and 248 iron-repressed genes. Kyoto Encyclopaedia of Genes and Genomes (KEGG) analysis and followed network plot for the pathways’ interaction revealed that IRGs were significantly enriched in apoptosis, interleukin (IL)-17 signalling and tumour necrosis factor (TNF) signalling pathways (figure 1c).

To identify causative genes during iron mobilisation that could be involved in the pathogenesis of asthma, we performed a cross-comparison of IRGs with asthma-associated immune genes from the GeneCards, DisGeNT and Immport databases. A total of 12 intersected genes in four groups were extracted, as shown in figure 1d. The change trends in only three genes (HMOX1, ACTG1, VDR) were consistent with the cellular iron change trend, while the opposite was observed for the remaining genes. Moreover, protein–protein interaction analysis using the STRING database revealed three effective hub genes (CXCL8, CCL2, CSF2) located at the core of the interaction network (figure 1d). To further validate the results obtained from the screening analysis, we next performed quantitative reverse transcriptase PCR in two types of human bronchial epithelial cell lines (HBE and BEAS-2B). Six genes (CCL2, HMOX1, IRF7, TFRC, VDR and MAPK8), which vary dynamically over iron concentration, were confirmed with quantitative PCR in both cell types (data not shown). These provide a degree of validation for both sets of sequencing results. Notably, most of these alterations show a trend towards improved inflammation status and retard asthma progression. Contrary to our expectations, these observations appear to support a beneficial effect of iron mobilisation in the airway epithelium of individuals with asthma rather than a detrimental action.

To our knowledge, this is the first study to provide evidence about the existence of iron mobilisation mode in asthmatics’ airway epithelium. Disordered iron metabolism appears critical in maintaining the airway epithelial barrier function, regardless of whether the self-protection mechanism or normal iron metabolism is disrupted by allergen stimulation. In previous studies, the asthmatic bronchial epithelium has been recognised to be more susceptible to oxidant-induced apoptosis, which is considered a crucial factor in asthma development [6, 7]. Based on our in vitro studies, iron mobilisation patterns in the airway epithelium against the background of asthma pathogenesis may play a pivotal role in the progression of cellular apoptosis. Many human studies have observed increased TNF-α and IL-17A secretion in asthmatic airways [8, 9]. Iron overload induced by upregulated proteins involved in iron trafficking in the epithelium of asthmatic individuals may further aggravate these processes. Available evidence indicates that iron may assume differential or even opposite roles in different lung cells (e.g., macrophages, fibroblasts and epithelial cells) during asthma progression. This may explain some of the previous diverging results regarding iron supplementation in animal and cell models [4, 10, 11]. Given the study's exploratory nature, we believe that our results will lay the foundation for further research.