Disrupting the TGF-β–Wnt Balance to Control the Polarity of the Cystic Fibrosis Airway Epithelium

Although cystic fibrosis (CF) is a multiorgan disease, its pulmonary manifestations are the major cause of morbidity and mortality associated with the disease. Decreased water content in the airway surface liquid contributes to clogging of the airways, chronic bacterial infections, and inflammation that ultimately results in disseminated bronchiectasis and lung failure (1).

One of the reasons for the increase in bacterial infections is a loss of polarity in the CF airway epithelium. Several groups have reported that loss of polarity is related to cytoskeleton disorganization, and disruption of junctional complexes that leads to fibronectin deposition at the apical surface of epithelial cells (2, 3). This form of airway remodeling, which involves partial epithelial–mesenchymal transition (4), is driven by different signaling mechanisms, with preeminent roles for pathways induced by transforming growth factor β (TGF-β) and Wnt.

TGF-β is a major regulator of pulmonary homeostasis, and its dysregulation is associated with CF lung disease. Reports of increased TGF-β1 levels first described TGF-β as a gene modifier that contributed to more severe lung disease in patients with the CF transmembrane conductance regulator (CFTR) F508del mutation (5). TGF-β can also inhibit CFTR expression and function (6), an effect that is reflected in the negative impact of TGF-β treatment on the beneficial effects CFTR modulator therapies, including lumacaftor (7) and triple-combination elexacaftor–tezacaftor–-ivacaftor (ETI) (8). The downregulation of CFTR expression and function very likely occurs through the TGF-β–induced upregulation of microRNAs, such as miR-145 (9, 10). TGF-β also drives goblet cell hyperplasia, contributing to the mucociliary dysfunction, aberrant inflammatory responses, and enhanced fibrosis (11), all hallmarks of CF lung disease.

Wnt is also involved in the regulation of airway differentiation. Canonical activation of the Wnt pathway leads to the stabilization of β-catenin, which, in turn, contributes to the transactivation of Wnt target genes that are responsible for regulating proliferation and differentiation. CFTR can interact with Wnt signaling components, an association that is lost for F508del CFTR (12). This interaction triggers an inverse relationship, with overexpression of CFTR contributing to shorter Wnt activity and increased differentiation (13).

In this issue of the Journal, Idris and colleagues (pp. 81–94) explore in depth the mechanisms that determine altered airway epithelium polarity in CF, including how TGF-β and Wnt pathways are involved in these processes (14). To identify genes that are differentially expressed and involved in epithelial phenotype disruption, the authors assessed wound healing by RNA sequencing on primary cultures of human airway epithelial cells obtained from either healthy donors or persons with CF (pwCF). Among the genes with higher expression in cells from pwCF, the authors identified TGF-β1 and dickkopf-related protein 1 (DKK1), which encode secreted proteins that promote hyperactive TGF-β and decreased Wnt signaling, respectively. This observation was confirmed at the protein level in the secretome of these cells.

Using the immortalized Calu-3 cell line with CFTR knockdown (KD), the authors confirmed an increase in TGF-β1 mRNA and in secreted TGF-β1 and its downstream effectors, including phosphorylated Smad2/Smad3, total Smad2, and total Smad7. These alterations in TGF-β led to an increase in fibronectin deposition in CFTR KD cells, which is consistent with previous findings from this group (15, 16).

The causal relationship between CFTR and DKK expression was also assessed in Calu-3 cells, with an increase in DKK1 evident in the secretome of CFTR KD cells. This manifested in the inhibition of the canonical Wnt pathway through a decrease in total, membrane-bound, and nuclear β-catenin. CFTR KD cells also exhibited impaired proliferation, suggesting that increased DKK1 may lead to β-catenin loss and unbalanced proliferation and polarization processes.

Exploring the mechanisms behind ectopic apical fibronectin accumulation in CFTR KD cells, the team found that TGF-β inhibition decreased fibronectin and fully inhibited its apical deposition. Similarly, DKK1 inhibition led to inhibition of fibronectin deposition, and, in this case, it occurred alongside an increase in total and membrane-bound β-catenin.

Because different signaling pathways interact to promote epithelial homeostasis, the possible link between TGF-β and Wnt pathways was also investigated. Inhibition of TGF-β led to a decrease in DKK1 and an increase in membrane-bound β-catenin. TGF-β also caused a decrease in total Akt in CFTR KD cells. Moreover, Akt inhibition reduced TGF-β and DKK1 secretion, Smad2 phosphorylation, and fibronectin deposition.

In primary cultures of human airway epithelial cells from healthy donors and F508del homozygous CF donors, the previous observations were validated and reinvestigated under treatment with the triple-combination ETI. It is interesting that, although treatment with modulators restored CFTR expression and its colocalization with ZO-1 at the cell membrane, in CF cells, it did not “rescue” hypersecretion of TGF-β and DKK1 or apical deposition of fibronectin.

Although this study reports some discrepancies between observations in primary cultures from pwCF and in Calu3 cells, it is a significant advance in our understanding of how the interplay between TGF-β and Wnt signaling pathways is unbalanced in CF. A more thorough validation using specific downregulation of the individual molecular players (e.g., by siRNA) would further elucidate and validate the findings reported.

The identification of a central role for the Akt pathway in TGF-β and Wnt signaling pathways interplay highlights a novel and potentially important molecular mechanism in the CF airway epithelium. Clearly, this process is quite intricate, and many details are still to be identified, such as the identity and role of feedback loops between fibronectin deposition and Akt activation.

Finally, this study provides important observations that modulator treatment (at least for the short duration of treatments reported here) is not sufficient to rescue the apicobasal epithelial polarity in CF, specifically with the F508 del CFTR mutation. This agrees with reports that lung infections persist in pwCF, even under ETI treatment. Thus, as the CF field continues its search for ways to increase treatment efficacy, modulation of signaling pathways involving TGF-β and Wnt are worthy of attention.