Discussion
Our study demonstrates an impaired epithelial barrier function in CRSwNP
patients along with decreased protein and mRNA expression of ZO-1,
occludin and Cldn3. Long-term exposure of IL-13 altered hNECs
composition in conjunction with nasal epithelial barrier dysfunction and
enhanced mucosal inflammation, as well as reduced capacity for antiviral
response against acute RV infection. These observations extend the
knowledge of dysregulation of TJs in chronic airway diseases such as
CRSwNP. Most importantly, ALI-cultured hNECs are shown to form
functional barriers similar to that found in nasal biopsies. Using the
IL-13-matured hNECs model, we have demonstrated that pro-inflammatory
mediator IL-13 regulates the formation TJs and thereby disrupts the
nasal epithelial barrier functions.
Nasal epithelial cells form a functional barrier which is mainly
regulated by TJs. Recent studies have highlighted multiple defective TJs
in patients with chronic airway diseases26 and
reported that ZO-1, claudin-1, claudin-4 and occludin contributed to the
leaky barrier of airway epithelium14,27,28. In our
study, we found that the expression of Cldn3, similar to ZO-1 and
occludin, is significantly lower in sinonasal tissues of CRSwNP as
compared to healthy controls. Moreover, we noticed that occludin and
Cldn3 protein expression level negatively correlates with MUC5AC
expression and positively correlates with βIV-tubulin expression in
nasal biopsies specimens. It appears that alteration of nasal epithelial
homeostasis during inflammation may impair TJs integrity. For this
purpose, we used ALI-cultured hNECs with IL-13 stimulation which mimics
the remodeling of nasal epithelium during chronic inflammation to
explore the formation and disruption of TJs and thereby the
IL-13-induced barrier damage.
We demonstrated that long-term exposure of hNECs to IL-13 results in
overproduction of mucus and cilia loss along with impaired epithelial
barrier function which is evident by increased epithelium permeability
and decreased TEER. ZO-1, occludin and Cldn3 structures were also
disrupted by IL-13 treatment during hNECs differentiation. In addition
to enhanced inflammation, defective barrier function allows foreign
substance to infiltrate the sinonasal submucosa and causes aggravation
of airway inflammation and remodeling. Hence, TJs play an important in
the pathophysiology of chronic airway disease. Taken together, these
data indicate that impaired formation of TJs protein ZO-1, occludin and
Cldn3 is largely associated with barrier dysfunction.
Although previous studies have examined TJs expression following various
cytokine insults, differentiated epithelial cell lines or primary airway
epithelial cells were commonly used14. However, these
cell models were not able to elucidate the processes of TJs formation
during epithelium differentiation. Therefore, we investigate the effect
of IL-13 on TJs protein formation during the time course of ALI
cultivation. We found that positively stained ZO-1 and occludin were
observed as early as Day7 and Day3 respectively at early stage of
differentiation of hNECs with or without IL-13 treatment. Interestingly,Cldn3 gene expression only increased when ciliated cells were
first observed at Day11 and is expressed only on ciliated cells of
hNECs. Hence, the reduced levels of Cldn3 may be attributed to loss of
cilia in airway epithelium of chronic inflammatory airway disease.
Additionally, as key barrier function proteins, claudins serve as
paracellular barrier29 and are shown to have
differential tissue-specific expression patterns which account for the
differences in paracellular tightness and ion
selectivity11,30,31. Although Cldn3 is expressed in
the airways, its role in airway barrier function has not been fully
defined. Studies in lower airways have demonstrated that the expression
level and function of Cldn3 are not comparable between type-I and
type-II alveolar cells32-34. Therefore, consistent
with study which reported varying TJ composition in different airway
epithelial cell type, the differential expression of Cldn3 in our study
implies that Cldn3 may be involved in regulation of epithelial barrier
as well as in epithelial differentiation. However, involvement Cldn3 in
ciliogenesis during differentiation is currently unknown. Thus, further
studies will be needed to investigate how TJ composition and epithelial
cell differentiation interrelate for regulation of barrier permeability
within stratified epithelia.
As the nasal airway is the primary target site for most respiratory
viral infections, impaired
epithelial barrier could lead to greater susceptibility against viral
infection and dysregulation of host innate immune
responses35. RV is the most prevalent respiratory
virus in CRSwNP patients and is the most commonly associated with
exacerbation of chronic airway disease16,17,20. As RV
infection may have significant implications in regulating the epithelial
barrier function and mucosal inflammation of CRSwNP,
we further investigate the effect
of RV infection on TJs of IL-13-matured hNECs and the effects of IL-13
on the host responses of hNECs against RV infection. We found that the
altered hNECs composition (cilia loss and mucus overproduction) in the
presence of IL-13 is associated with the reduced RV replication (viral
RNA level) and viral particle formation as compared to RV infection
without IL-13 at the same initial infectious dose. As our previous study
found that RV almost exclusively infected ciliated cells but not goblet
and basal cells in in vitro hNECs24, cilia loss
due to IL-13 may impede RV infection by reducing target cells for viral
replication. However, despite lower viral replication and production, RV
infection of IL-13-treated hNECs worsened the mucociliary function by
further inducing loss of cilia as shown by reduction of Foxj1mRNA level and IF staining. While studies have reported that RV
infection disrupted TJs in primary airway epithelial
cells16,36, our data showed that RV infection induced
minimal alteration to TJs proteins in hNECs with and without IL-13
treatment. In addition, we investigated the effects of IL-13 on innate
immune responses of hNECs against RV infection. We found that viral
entry receptor ICAM-1 , RV-induced host pathogen sensor (TLR3) and
antiviral immune responses (IFN-λ1 and CXCL10) were upregulated in both
untreated and IL-13-treated hNECs, suggesting that RV infection induced
immune surveillance and antiviral responses even in inflammatory hNECs
model. However, the capacity for interferon activation and chemokine
signaling were impaired when hNECs is predisposed with IL-13 environment
as shown by lower IFN-λ1 and CXCL10 expression as compared
untreated hNECs. In contrast, RV-induced greater upregulation ofTSLP expression in IL-13-treated hNECs suggesting that RV
infection in nasal epithelium predisposed with type-2 cytokine
environment could lead to enhanced allergic inflammation which may
further drive inflammation during RV-induced exacerbation of disease.
Our current study investigated the effects of mild respiratory virus
RV-16, which commonly associated to chronic respiratory disease
exacerbations, on epithelium barrier function and immune capacity of
hNECs in type-2 cytokine environment. While RV infection induced minimal
change to TJs dysfunction, the impairment of
efficient antiviral response in nasal epithelium may be attributed to
IL-13-induced change in hNECs composition. More pathogenic respiratory
viruses could be studied using this model to assess the epithelia
barrier function and antiviral responses of chronic inflammatory airway
disease.
In conclusion, IL-13, a typical type-2 cytokine, contributes to
diminished barrier function and airway inflammation that are seen in
CRSwNP patients. Knowledge about the dysregulation of TJs will help to
better understand the pathophysiology of CRSwNP and define the specific
mechanisms that link allergic inflammation and antiviral responses,
which could lead to new strategies for the prevention and treatment of
the disease.