Zonula occludens-1 expression is reduced in nasal epithelial cells of allergic rhinitis patients

Introduction

Allergic rhinitis (AR) is an IgE-mediated inflammation disorder of the nasal mucosa caused by infiltration of allergens (Varshney & Varshney, 2015). This leads to imbalanced immunological reaction resulting in clinical manifestations of the disease including sneezing, nasal obstruction, nasal pruritus, rhinorrhoea or conjunctivitis (Sin & Togias, 2011; Wheatley & Togias, 2015; Brozek et al., 2017; Siti Sarah et al., 2020). AR is a common disease affecting 10% to 30% of adults and up to 40% of children (Mims, 2014). AR is classified into four groups based on the Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines, which are mild intermittent, mild persistent, moderate-severe intermittent and moderate-severe persistent (Klimek et al., 2019). Exposure to allergens such as grass, tree pollen, house dust mites (HDMs), animal dander, foods, insect venoms and medicines induce the production of IgE that subsequently triggers allergic reaction (Galli, Tsai & Piliponsky, 2008; Ong & Chew, 2009; Yang et al., 2018; Fassio & Guagnini, 2018; Wang et al., 2018b). HDMs and pollen allergens have protease activity that contribute to the disruption of nasal epithelial barrier (Wan et al., 1999; Roche et al., 2000; Okano, 2009; Matsumura, 2012). The decreased of epithelial barrier integrity will result in an increased in allergen invasion and contribute to the pathogenesis of AR (Steelant et al., 2016; Wang et al., 2020a; Wang et al., 2020b). Thus, the decrease of ZO-1 level is responsible for symptoms of AR. Pollen-derived proteases have been shown to degrade occludin, and facilitate the allergen delivery through epithelia by disrupting the epithelial tight junction (TJ) components (Matsumura, 2012). Dermatophagoides pteronyssinus (D. pteronyssinus) delivers cysteine proteases that increases the epithelial permeability by disrupting claudin-1, occludin and ZO-1 molecules (Baker et al., 2003; Matsumura, 2012; Henriquez et al., 2013).

Nasal airways are exposed to the environmental allergens caused by the breakdown of the nasal epithelial barrier that blocks the passage to allergens. The gap between the epithelial cells are closed by TJ molecules and form a complex structure to protect the body from infiltration of external allergens (Beutel et al., 2019). TJ molecules consist of transmembrane proteins (e.g., occludin, claudins and tricellulin) and junctional adhesion molecules (Steelant et al., 2016; Heinemann & Schuetz, 2019). The scaffold adaptor protein zonula occludens (ZO) forms a TJ assembly attached to actin cytoskeleton and TJ molecules (Beutel et al., 2019). ZO proteins are expressed in epithelial cells throughout the airway where they remain localized to the cell borders (Smith, Koval & Levy, 2020).

ZO proteins are from the MAGUK family members including ZO-1, ZO-2 and ZO-3, and they are crucial in maintaining the epithelial barrier integrity (Itoh et al., 1999; Chelakkot, Ghim & Ryu, 2018). Decreased expression of ZO-1 occur in the nasal epithelium cells of AR patients (Lee et al., 2016). ZO-1 and ZO-2 expression were also reduced in patients with chronic rhinosinusitis (CRS) without nasal polyps (Soyka et al., 2012). Defects of these ZO proteins may contribute to the opening of the gap between epithelium cells and the dysfunction of nasal epithelial barrier.

Histone deacetylase (HDAC) is a group of enzymes that can repress the transcription of gene through deacetylation of histone molecules. HDAC acts by removing an acetyl group from lysine on histones (Jiang et al., 2015). HDACs regulate expression of genes involved in multiple cellular activities including inflammatory responses leading to allergies (Bhavsar, Ahmad & Adcock, 2008; Barnes, 2013). A study using a mouse model with AR fed with HDAC inhibitor (HDACi) sodium butyrate (NaB), showed positive effects to treat AR by decreasing the expression of HDAC1 and HDAC8 (Wang et al., 2020a; Wang et al., 2020b). Higher expression of HDAC1 can cause further defects of ZOs expression (Lei et al., 2010; Zhou et al., 2015). HDAC1 has been found to suppress Trek1 and the exposure to the signature T helper 2 (Th2) cytokine, interleukin (IL)-4, upregulated the expression of HDAC1 resulting in significantly suppressed expression of Trek1 in the nasal mucosa (Jiang et al., 2015). These findings highlighted the importance of HDAC1 as a central player in the mechanism regulating the outcome of epithelial barrier function, and HDAC1 is a potential target for inhibition to ameliorate nasal epithelial barrier dysfunction for better outcome in AR. HDAC1 inhibition had been shown to increase the TJ expression while decreasing epithelial barrier integrity defects (Wawrzyniak et al., 2017). ZO-1 mRNA expression level was found to be lower in AML-12 murine hepatocyte cells with the presence of overexpressed HDAC1 (Lei et al., 2010). HDAC2 inhibited claudin-1 expression in malignant epithelial cells (Krishnan et al., 2010) and HDACs are known to suppress TJs expression in other disease contexts (Sin & Togias, 2011; Brozek et al., 2017). Human lens epithelial cells treated with Trichostatin-A (TSA), a HDACi decreased HDAC2 and increased the mRNA expression of ZO-1 (Ganatra et al., 2018). In addition, treatment of NCM460 cells with CAY 10683, a HDAC2 inhibitor increased ZO-1 mRNA and protein levels (Wang et al., 2018a).

Past studies have focused on the association of ZO-1 and HDAC1 gene expressions in moderate/severe HDM-induced AR patients; these HDMs include Dermatophagoides farinae (D. farinae), D. pteronyssinus and Blomia tropicalis (B. tropicalis), which are the most common aeroallergens affecting AR population in Malaysia (Mohd Ashari et al., 2014; Lim et al., 2015; Azid et al., 2019; Sani et al., 2019; Nur Husna et al., 2021). There is lack of literature in association of ZO-2, ZO-3 and HDAC2 in moderate /severe HDM-induced AR patients. Moreover, data associating ZO and HDAC to environmental or lifestyle factors such as having pets remain scarce.

Our gene expression studies and their associations with clinico-demographical and environmental parameters in AR patients are divided into two parts. The first part of this study focused on ZO proteins and HDACs, while the second part focused on occludin, claudins and desmogleins. In this study, we aimed to investigate the mRNA expression of ZOs (ZO-1, ZO-2 and ZO-3) and HDACs (HDAC1 and HDAC2) in AR patients compared to healthy controls in subjects recruited from Hospital Universiti Sains Malaysia, Kelantan, Malaysia. The associations of the expression levels of ZOs and HDACs with the clinico-demographical parameters of AR patients and healthy controls, as well as patients’ lifestyle were measured.

Materials & Methods

Results

Expression profile of ZOs and HDACs in AR patients compared to healthy controls

To address potential role of TJ molecules in maintaining the epithelial barrier integrity, ZO-1, ZO-2, ZO-3, HDAC1 and HDAC2 mRNA expression were measured in NECs of AR patients and healthy controls. Shapiro–Wilk normality test showed that the data were not normally distributed. Thus, Mann–Whitney U test was used. ZO-1 expression was significantly lower in AR patients (p = 0.010) compared to controls (Fig. 1A). No difference was found in the expression of ZO-2 (p = 0.868) and ZO-3 (p = 0.351) in AR patients compared to controls (Figs. 1B and 1C).

In terms of HDACs expression, no significant difference in the expression of HDAC1 (p = 0.748) and HDAC2 (p = 0.693) in AR patients compared to healthy controls (Figs. 1D and 1E).

Discussion

AR is a chronic and prevalent nasal mucosa inflammatory disorder, with the pathophysiologic mechanisms responsible for the severity of disease being only partially understood (Wang et al., 2020a; Wang et al., 2020b). Nasal epithelial barrier defects in AR patients have been hypothesized to result in excessive allergen exposure and activation of inflammatory cells (Steelant et al., 2018). In this context, the role of the nasal epithelium in protecting the nasal airway from allergens exposure is still not clearly understood.

In this study, we demonstrated that decreased expression of ZO-1 occurred in HDM-sensitized moderate-severe AR patients compared to healthy controls. This was in line with findings of previous studies in AR patients that showed impaired epithelial TJ function and lower ZO-1 mRNA expression in both nasal epithelial cells and nasal biopsy specimens (Steelant et al., 2016; Wang et al., 2020a; Wang et al., 2020b). Independent studies also showed significantly decreased expression of ZO-1 protein in nasal epithelium of AR patients compared with healthy controls (Wang et al., 2020a; Wang et al., 2020b). Immunofluorescence staining demonstrated a relatively poor ZO-1 arrangement in biopsy specimens of HDM-induced AR patients compared to healthy controls (Steelant et al., 2016). HDMs are known to contain abundant proteases (Reithofer & Jahn-Schmid, 2017), and protease activity of HDMs increases the permeability of epithelial barrier that promotes the penetration of allergens into nasal mucosa by targeting TJs (Wan et al., 1999; Roche et al., 2000; Okano, 2009). Reduced expression of ZO-1 was also observed in patients with CRS with nasal polyps, asthmatic patients, eosinophilic esophagitis (EoE) and these impaired epithelial TJ function are considered to be part of the pathophysiology (De Benedetto et al., 2011; Salim & Söderholm, 2011; Xiao et al., 2011; Soyka et al., 2012; Katzka et al., 2014). The airway epithelium is known as the first line of defense that acts as a protective barrier against environmental allergens (Nadeem et al., 2019). Impaired expression of ZO-1 may thus facilitate the infiltration of allergens into the epithelial barrier of AR patients.

In contrast to ZO-1, expression levels of ZO-2 and ZO-3 did not significantly differ between AR patients and controls in our study. Recent study demonstrated that lower expression of ZO-2 and ZO-3 were observed in nasal biopsy samples of atopic patients in Turkey (Yılmaz et al., 2019). Epidemiological studies showed that cigarette smoke is one of a risk factor for asthma and CRS with polyps. Incubation of 16HBE14 cells and human bronchial epithelial cells (pHBECs) with cigarette smoke extract decreased ZO-2 expression and thus disrupted TJ barrier (Fukuoka & Yoshimoto, 2018). However, there are still no studies investigating the expression of ZO-2 and ZO-3 molecules in AR patients.

MAGUKs (ZO-1, ZO-2 and ZO-3) are thought to play important roles in producing and maintaining specialized membrane domains in various types of cells (Inoko et al., 2003). Similar to ZO-1, both ZO-2 and ZO-3 are localized at TJs in epithelial cells and bind directly to the actin filaments (Itoh et al., 1997; Fanning et al., 1998). Nevertheless, the contribution of both ZO-2 and ZO-3 to the nasal epithelial barrier remains unclear due to lack of studies on the involvement of both ZO-2 and ZO-3 in AR pathogenesis.

Pertaining to HDACs, we did not observe significant difference in the expression levels of HDAC1 and HDAC2 in AR patients compared to controls. HDAC1 can suppress TWIK-related potassium channel-1 (Trek1) in the nasal mucosa essential in preserving the epithelial barrier function (Bittner et al., 2013; Jiang et al., 2015), establishing HDAC1 as one of the key players in deregulating epithelial barrier. Recent therapeutic research has demonstrated that HDACi enhance ZO proteins expression, leading to improved epithelial barrier integrity (Yılmaz et al., 2019). HDACi such as NaB and JNJ-26481585 inhibit the activity of HDAC1 resulting in enhanced nasal epithelial barrier integrity (Steelant et al., 2019; Wang et al., 2020a; Wang et al., 2020b). Furthermore, treatment of fetal human lens epithelial cells with the HDACi TSA increased the expression of ZO-1 and improved the epithelial barrier integrity (Ganatra et al., 2018). Previous study reported higher levels of HDAC1 in AR patients compared to healthy controls in China (Wang et al., 2015). This may be due to prevalence of seasonal allergens in China (e.g., pollens) (Zhang et al., 2009) but distinct from perennial allergens (e.g., HDMs) in Malaysia.

Interestingly, we observed that higher expression of HDAC2 in AR patients was significantly associated with lower frequency of changing bedsheet. HDMs mainly come into contact with human through mattresses and bedsheet (Abu Khweek et al., 2020). A study in China also showed that high concentration of D. farinae was detected on rhinitis children’s beddings compared to controls (Huang et al., 2019). Hypersensitivity to HDMs also contribute to atopic sensitization in 50–85% of asthmatics, and are strong inducers of causing allergy in worldwide population (Abu Khweek et al., 2020). HDAC activity has been identified as a key factor of allergic inflammation and TJ dysfunction (Steelant et al., 2019). Increased production of Th2 cytokines caused by allergic inflammation leads to increased HDAC activities in epithelial cells that augment mucosal permeability (Steelant et al., 2018). This is in line with higher expression of HDAC2 which was significantly associated with our cohort of AR patients sensitized to D. farinae. Collectively, this suggests that HDACi administration is a potential therapeutic strategy for AR patients sensitized to D. farinae and this is subject to future investigations.

In this study, we observed that the expression of ZO-2 was higher in healthy controls who had pets (Table 5). Exposure to pets at early stage may decrease the possibility of developing allergic sensitization to animal dander or other inhaled allergens (Shargorodsky et al., 2017). It has been shown that living with cats and dogs was associated with lower risk of developing atopy during childhood and young adulthood (Päivi & Darryl, 2009). Frequent exposure to allergens may later protect from sensitization by adapting to transient increase of Th2 cytokines levels and subclinical responses (Shargorodsky et al., 2017; Steelant et al., 2018). However the role of Th1 is still not clearly understood (Steelant et al., 2018).

We acknowledge the limitations of the study as follows: (1) We focused on HDM-sensitized AR patients only without the inclusion of other allergens during SPT. It is plausible that our cohort of AR patients were also allergic to other allergens which may contribute to the TJs disruption; (2) In our study, the gene expression was sufficient to demonstrate the presence of ZO and HDAC. Although protein expression data is important to confirm the existence of ZO and HDAC gene, however we could not perform the protein analysis in this study due to budget constraint. (3) We did not measure specific IgE tests for all the allergens. In the future, we would like to further investigate more on protein expression levels, which will help to strengthen our research.

Conclusions

In summary, our data indicate that ZO-1 is a key TJ molecule whose reduced expression is associated with defective nasal epithelial barrier integrity in HDM-sensitized AR patients. These results support development of therapies that restore ZO-1 expression in nasal epithelial cells of AR patients. The expression profile of ZO-2, ZO-3 and HDAC2 were demonstrated in AR patients relative to non-allergic subjects for the first time. Collectively, dysregulated ZO-1 and HDAC2 expression levels may play key roles in the onset of AR through disruption of nasal epithelial barrier integrity.

Supplemental Information