*3.1. E*ff*ects of CSE on MMP and TIMP Production in Nasal Fibroblasts*

To determine whether CSE regulates the production of MMPs and TIMPs, we treated fibroblasts with CSE at various concentrations (0–5%). CSE increased MMP-2 expression dose-dependently but did not affect the expression of MMP-9 mRNA and protein (Figure 1A,B). Gelatin zymography, used to evaluate enzymatic activity, showed that only MMP-2 activity was increased by CSE treatment (Figure 1C). The stimulatory effect of CSE on MMP-2 protein was also confirmed by immunofluorescence staining (Figure 1D). CSE decreased TIMP-2 expression but did not affect the expression of TIMP-1 (Figure 1E,F). These results indicate that CSE induced MMP-2 production and decreased TIMP-2 production in nasal fibroblasts.

fibroblasts.

**Figure 1.** Effect of CSE on MMP and TIMP expression in nasal fibroblasts. After treatment of nasal fibroblasts with 5% CSE, the mRNA levels of *MMPs* and *TIMPs* were measured using real-time PCR (**A** and **E**), and MMP and TIMP protein expression levels were determined using ELISA (**B** and **F**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**C**). The expression and localization of MMP-2 protein (green) were observed using immunofluorescence staining (**D**). \**p* < 0.05 vs. control. Scale bar = 100 μm. **Figure 1.** Effect of CSE on MMP and TIMP expression in nasal fibroblasts. After treatment of nasal fibroblasts with 5% CSE, the mRNA levels of *MMPs* and *TIMPs* were measured using real-time PCR (**A**,**E**), and MMP and TIMP protein expression levels were determined using ELISA (**B**,**F**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**C**). The expression and localization of MMP-2 protein (green) were observed using immunofluorescence staining (**D**). \* *p* < 0.05 vs. control. Scale bar = 100 µm.

#### *3.2. Role of ROS in MMP and TIMP Production in CSE-Stimulated Nasal Fibroblasts 3.2. Role of ROS in MMP and TIMP Production in CSE-Stimulated Nasal Fibroblasts*

To investigate the role of ROS in MMP-2 and TIMP-2 production, nasal fibroblasts were pretreated with ROS scavengers 1 h before CSE treatment. In DCFH-DA and Amplex Red staining assays, CSE treatment induced the production of ROS and hydrogen peroxide; however, such production was blocked by the addition of NAC, ebselen, or DPI (Figure 2A,B). Next, the effects of ROS scavengers on mRNA and protein expression were measured using real-time PCR and western blotting. Pretreatment with ROS scavengers blocked the stimulatory effects of CSE on MMP-2 expression (Figure 2C,D). This result was also observed through gelatin zymography (Figure 2E). Inversely, pretreatment with antioxidants blocked the inhibitory effect of CSE on TIMP-2 expression (Figure 2F,G). MMP-9 and TIMP-1 production was not affected by ROS scavengers in nasal To investigate the role of ROS in MMP-2 and TIMP-2 production, nasal fibroblasts were pretreated with ROS scavengers 1 h before CSE treatment. In DCFH-DA and Amplex Red staining assays, CSE treatment induced the production of ROS and hydrogen peroxide; however, such production was blocked by the addition of NAC, ebselen, or DPI (Figure 2A,B). Next, the effects of ROS scavengers on mRNA and protein expression were measured using real-time PCR and western blotting. Pretreatment with ROS scavengers blocked the stimulatory effects of CSE on MMP-2 expression (Figure 2C,D). This result was also observed through gelatin zymography (Figure 2E). Inversely, pretreatment with antioxidants blocked the inhibitory effect of CSE on TIMP-2 expression (Figure 2F,G). MMP-9 and TIMP-1 production was not affected by ROS scavengers in nasal fibroblasts.

**Figure 2.** Effects of ROS on CSE-regulated MMP and TIMP expression in nasal fibroblasts. Total ROS and mitochondrial superoxides were quantified using the 2,7-dichlorofluorescein diacetate assay and Amplex Red assay (**A**,**B**). Fibroblasts were pretreated with NAC (1 mM), ebselen (10 µM), and DPI (2 µM) before being treated with CSE (5%). The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**C**,**F**). MMP and TIMP protein levels were determined using ELISA (**D**,**G**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**E**). \* *p* < 0.05 vs. control; † *p* < 0.05 vs. CSE only. Scale bar = 50 µm.

#### *3.3. Involvement of the PI3K*/*Akt Cascade in MMP and TIMP Production in CSE-Stimulated Nasal Fibroblasts*

The fibroblasts were treated with CSE and a PI3K/Akt inhibitor to confirm whether a PI3K/Akt pathway is involved in the expression of MMP-2 and TIMP-2. CSE induced PI3K/Akt phosphorylation, which was inhibited by the PI3K/Akt inhibitor (LY294002) (Figure 3A). Next, we confirmed that the PI3K/Akt inhibitor suppressed MMP-2 mRNA and protein expression and enzymatic activation, which are induced by CSE in nasal fibroblasts (Figure 3B–D). On the contrary, TIMP-2 mRNA and protein expression was significantly increased by treatment with the PI3K/Akt inhibitor in CSE-stimulated nasal fibroblasts (Figure 3E,F). Additionally, we confirmed that ROS inhibition suppressed the activation of PI3K/AKT in nasal fibroblasts (data not shown). Therefore, we could assume that the signaling pathway associated with fibroblast activation under oxidative stress is attributable in part to the activation of the PI3K and AKT signaling pathways.

**Figure 3.** Regulation of PI3K/Akt signaling pathways with CSE regulated MMP and TIMP expression. Nasal fibroblasts were pretreated with LY294002 (PI3K/Akt inhibitor) before treatment with 5% CSE. Levels of phosphorylated (p)-PI3K and p-Akt were determined using western blotting (**A**). The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**B** and **E**). MMP and TIMP protein levels were determined using ELISA (**C** and **F**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**D**). \**p* < 0.05 vs. control; †*p* < 0.05 vs. CSE only. **Figure 3.** Regulation of PI3K/Akt signaling pathways with CSE regulated MMP and TIMP expression. Nasal fibroblasts were pretreated with LY294002 (PI3K/Akt inhibitor) before treatment with 5% CSE. Levels of phosphorylated (p)-PI3K and p-Akt were determined using western blotting (**A**). The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**B**,**E**). MMP and TIMP protein levels were determined using ELISA (**C**,**F**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**D**). \* *p* < 0.05 vs. control; † *p* < 0.05 vs. CSE only.

#### *3.4. Effect of CSE on NF-κB Activation for MMP and TIMP Production 3.4. E*ff*ect of CSE on NF-*κ*B Activation for MMP and TIMP Production*

The activated NF-κB enhanced MMP-2 expression in airway. To determine whether NF-κB activation is involved in MMP-2 and TIMP-2 expression, fibroblasts were pretreated with an NF-κB inhibitor (BAY11-7082) and then stimulated with CSE. Phosphorylated p65, a subunit of NF-κB, was induced by CSE treatment and inhibited by treatment with the NF-κB inhibitor (Figure 4A). NF-κB transcriptional activity, assessed by a luciferase reporter, was increased by CSE and then inhibited by the NF-κB inhibitor (Figure 4B). Immunocytochemical staining showed that CSE induced the translocation of p-p65 to the nucleus and that this was blocked by the NF-κB inhibitor (Figure 4C,D). Treatment with the NF-κB inhibitor inhibited MMP-2 mRNA and protein expression and activation of enzymatic capacity, which were stimulated by CSE (Figure 4E–G), and reversed the change in TIMP-2 mRNA and protein expression that had been inhibited by CSE treatment (Figure 4H,I). MMP-9 and TIMP-1 production was not affected by the NF-κB inhibitor in nasal fibroblasts. The activated NF-κB enhanced MMP-2 expression in airway. To determine whether NF-κB activation is involved in MMP-2 and TIMP-2 expression, fibroblasts were pretreated with an NF-κB inhibitor (BAY11-7082) and then stimulated with CSE. Phosphorylated p65, a subunit of NF-κB, was induced by CSE treatment and inhibited by treatment with the NF-κB inhibitor (Figure 4A). NF-κB transcriptional activity, assessed by a luciferase reporter, was increased by CSE and then inhibited by the NF-κB inhibitor (Figure 4B). Immunocytochemical staining showed that CSE induced the translocation of p-p65 to the nucleus and that this was blocked by the NF-κB inhibitor (Figure 4C,D). Treatment with the NF-κB inhibitor inhibited MMP-2 mRNA and protein expression and activation of enzymatic capacity, which were stimulated by CSE (Figure 4E–G), and reversed the change in TIMP-2 mRNA and protein expression that had been inhibited by CSE treatment (Figure 4H,I). MMP-9 and TIMP-1 production was not affected by the NF-κB inhibitor in nasal fibroblasts.

**Figure 4.** Effects of NF-κB activation on CSE-regulated MMP and TIMP levels in nasal fibroblasts. Nasal fibroblasts were pretreated with an NF-κB inhibitor (BAY 11-7082) before treatment with 5% CSE. Phospho (p)-p65 was measured using western blotting to determine NF-κB activation (**A**). NFκB transcriptional activation was measured using the luciferase assay (**B**). After stimulation with CSE, the translocation of p-p65 protein (red) was observed by immunofluorescent staining. Magnification, ×400 (**C** and **D**). Nuclei were stained using DAPI (blue). The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**E** and **F**). The secretion levels of MMP and TIMP proteins were determined using ELISA (**H** and **I**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**G**). \**p* < 0.05 vs. control; †*p* < 0.05 vs. CSE only. Each experiment **Figure 4.** Effects of NF-κB activation on CSE-regulated MMP and TIMP levels in nasal fibroblasts. Nasal fibroblasts were pretreated with an NF-κB inhibitor (BAY 11-7082) before treatment with 5% CSE. Phospho (p)-p65 was measured using western blotting to determine NF-κB activation (**A**). NF-κB transcriptional activation was measured using the luciferase assay (**B**). After stimulation with CSE, the translocation of p-p65 protein (red) was observed by immunofluorescent staining. Magnification, ×400 (**C**,**D**). Nuclei were stained using DAPI (blue). The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**E**,**F**). The secretion levels of MMP and TIMP proteins were determined using ELISA (**H**,**I**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**G**). \* *p* < 0.05 vs. control; † *p* < 0.05 vs. CSE only. Each experiment was performed three biological replicates. Scale bar = 100 µm.

#### was performed three biological replicates. Scale bar = 100 μm. *3.5. E*ff*ect of Steroids on CSE-Regulated MMP and TIMP Production*

*3.5. Effect of Steroids on CSE-Regulated MMP and TIMP Production*  Dexamethasone (Dex) and fluticasone propionate (FP) are potent synthetic corticosteroids that are widely used as anti-inflammatory agents to treat respiratory diseases [18]. To assess whether Dexamethasone (Dex) and fluticasone propionate (FP) are potent synthetic corticosteroids that are widely used as anti-inflammatory agents to treat respiratory diseases [18]. To assess whether steroidsinhibited CSE-regulated MMP and TIMP production, fibroblasts were pretreated with dexamethasone

steroids inhibited CSE-regulated MMP and TIMP production, fibroblasts were pretreated with

pathways in fibroblasts.

or fluticasone propionate and then stimulated with CSE. MMP-2 expression was increased by CSE treatment and suppressed by steroids, whereas the opposite pattern was observed for TIMP-2 mRNA (Figure 5A,D) and protein levels (Figure 5B,E). Gelatin zymography showed similar patterns for MMP-2 (Figure 5C). Steroids did not affect MMP-9 and TIMP1 expression. Steroids significantly decreased ROS production, PI3K/Akt phosphorylation, and NF-κB activation in CSE-stimulated nasal fibroblasts (Figure 5F–H). These results suggested that steroids could regulate MMP-2 and TIMP-2 expression by blocking the ROS/PI3K/Akt and NF-κB signaling pathways in fibroblasts. increased by CSE treatment and suppressed by steroids, whereas the opposite pattern was observed for TIMP-2 mRNA (Figure 5A,D) and protein levels (Figure 5B,E). Gelatin zymography showed similar patterns for MMP-2 (Figure 5C). Steroids did not affect MMP-9 and TIMP1 expression. Steroids significantly decreased ROS production, PI3K/Akt phosphorylation, and NF-κB activation in CSE-stimulated nasal fibroblasts (Figure 5F–H). These results suggested that steroids could regulate MMP-2 and TIMP-2 expression by blocking the ROS/PI3K/Akt and NF-κB signaling

*Antioxidants* **2020**, *9*, x FOR PEER REVIEW 10 of 14

**Figure 5.** Effects of steroids on CSE-regulated MMP and TIMP levels in nasal fibroblasts. Nasal fibroblasts were pretreated with or without dexamethasone (Dexa, 2.5 μM) and fluticasone propionate (FP, 2.5 μM) before treatment with 5% CSE. The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**A** and **D**). MMP and TIMP protein secretion levels were determined using ELISA (**B** and **E**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**C**). Total ROS production and mitochondrial superoxides were quantified using the 2,7-dichlorofluorescein diacetate assay and Amplex Red assay (**F** and **G**). Levels of phosphorylated p-PI3K, p-Akt, and p-p65 were determined using western blotting (**H**). \**p* < 0.05 vs. control; †*p* < 0.05 vs. CSE only. . Scale bar = 50 μm **Figure 5.** Effects of steroids on CSE-regulated MMP and TIMP levels in nasal fibroblasts. Nasal fibroblasts were pretreated with or without dexamethasone (Dexa, 2.5 µM) and fluticasone propionate (FP, 2.5 µM) before treatment with 5% CSE. The expression levels of MMP and TIMP mRNAs were determined by real-time PCR (**A**,**D**). MMP and TIMP protein secretion levels were determined using ELISA (**B**,**E**). The enzymatic activities of MMP-2 and MMP-9 were measured by gelatin zymography (**C**). Total ROS production and mitochondrial superoxides were quantified using the 2,7-dichlorofluorescein diacetate assay and Amplex Red assay (**F**,**G**). Levels of phosphorylated p-PI3K, p-Akt, and p-p65 were determined using western blotting (**H**). \* *p* < 0.05 vs. control; † *p* < 0.05 vs. CSE only. Scale bar = 50 µm.

#### **4. Discussion**

The present study showed that CSE induced MMP-2 expression and decreased TIMP-2 expression but did not affect MMP-9 and TIMP-1 expression in nasal fibroblasts. CSE exposure induced ROS production. Treatment with ROS scavengers, such as NAC, ebselen, and DPI, suppressed CSE-regulated MMP-2 and TIMP-2 expression. Additionally, CSE induced PI3K/AKT phosphorylation and NF-κB activation. When CSE-stimulated nasal fibroblasts were treated with PI3K/AKT and NF-κB inhibitors, the CSE-mediated regulation of MMP-2 and TIMP-2 expression was suppressed in nasal fibroblasts. These data indicated that CSE induces MMP-2 expression and inhibits TIMP-2 expression via ROS, PI3K/AKT, and NF-κB signaling pathways in nasal fibroblasts.

Histopathological changes in the nasal mucosa of smokers are reported to differ from those in the nasal mucosa in nonsmokers [19]. In several studies, CSE exposure in airways increased inflammatory responses and tissue remodeling to aggravate chronic upper respiratory inflammation and diseases, such as CRS [2,20,21]. In CRS, histomorphological changes, including epithelial cell hyperplasia, basement membrane thickening, and ECM accumulation, occur in the respiratory system, leading to tissue remodeling [22]. An imbalance in MMPs and TIMPs, which are important factors involved in ECM homeostasis, leads to tissue remodeling [23]. Particularly, gelatinases, MMP-2, and MMP-9 are known to degrade almost all basement membrane components, including collagens, laminins, and gelatins, in tissue remodeling [24]. Bachert et al. showed that MMP-2 and MMP-9 expression was significantly enhanced in CRS patients compared to that in healthy controls [25]. We have previously reported that MMP-2 expression was increased in TGF-β1-stimulated nasal polyp-derived fibroblasts [26]. The present study demonstrated that CSE induces MMP-2 expression and inhibits TIMP-2 expression in nasal fibroblasts.

CSE contains high concentrations of oxidants that induce ROS, which play a significant role in the pathogenesis of diseases, such as CRS [27]. In particular, CSE-induced ROS may promote tissue remodeling. Fordham et al. showed that ROS were increased in the epithelia of patients with CRS, and exposure to CSE increased ROS in nasal tissues [27]. Thus, we showed that CSE induced an imbalance in MMP-2 and TIMP-2 through ROS in nasal fibroblasts. Additionally, ROS scavengers, including NAC, ebselen, and DPI, inhibited not only the imbalance in MMP-2 and TIMP-2 production, but also PI3K/Akt phosphorylation and NF-κB activation in CSE-stimulated nasal fibroblasts. PI3K/Akt phosphorylation and NF-κB activation are known to occur in inflammatory response progression [28]. Previous studies suggested that the PI3K/Akt pathway regulates various signaling pathways that lead to NF-κB activation [29]. CSE also caused PI3K/Akt phosphorylation and NF-κB activation in these signaling pathways in lung fibroblasts [30,31]. Intracellular signaling pathways, such as PI3K/Akt and NF-κB, have been shown to modulate MMP-2 and MMP-9 expression in lung tissues [32]. In our study, the inhibition of PI3K/Akt and NF-κB significantly suppressed CSE-enhanced MMP-2 expression and inhibited TIMP-2 expression in nasal fibroblasts, which might be associated with CRS aggravation. These findings correspond well with those of earlier studies, which indicated that the PI3K/Akt and NF-κB signaling pathways play a crucial role in MMP-2 and TIMP-2 expression.

Although, steroids have severe side effects, they are still one of the most important drugs that can treat various diseases with anti-inflammatory and antioxidant properties. Steroids, such as dexamethasone and fluticasone propionate, have been successfully used in the therapy of various human inflammatory diseases, such as CRS. Steroids may also reduce mucosal inflammation and edema in paranasal sinuses and improve symptoms associated with CRS. However, recent evidence showed that glucocorticoids may affect not only the prevention of inflammation, but also the inhibition of tissue remodeling in CRS. It was shown that inhaled steroids inhibited TGF-β1-induced MMP expression in bronchial fibroblasts [33]. Steroids have been known to inhibit inflammation response and tissue remodeling by inhibiting various signaling pathway such as ROS and AKT signaling pathways [34,35]. Our study demonstrated that steroids down-regulated CSE-induced MMP-2 expression through inhibiting the ROS/PI3K/Akt and NF-κB signaling pathways. These results suggested that steroids contributed to the inhibition of MMP-2 expression in nasal fibroblasts.

Our study has some limitations. CRS is a multifactorial disorder, and its pathogenesis involves interactions between environmental insults, infectious loading, and genetic predisposition, it is not possible to directly translate the obtained results into the CRS model. However, indirectly, it may be thought that the regulation of MMP may influence tissue remodeling and thus contribute to the development of CRS. In addition to, we did not check the effect of CSE in CRS mouse model. We will check whether the exposure of CSE can aggravate symptoms of CRS.

### **5. Conclusions**

We have provided here the first evidence that CSE increases MMP-2 production and inhibits TIMP-2 expression in nasal fibroblasts. However, CSE did not affect MMP-9 and TIMP-1 expression in nasal fibroblasts. Additionally, our study demonstrated the role of the ROS/PI3K/Akt and NF-κB signaling pathways in mediating the CSE-regulated MMP-2/TIMP-2 imbalance in nasal fibroblasts, which might contribute to tissue remodeling in CRS. The present study did not demonstrate the effect of CSE on MMP and TIMP expression, which was only examined in one type of submerged cell culture. Additional experiments are needed to clarify the results of the present study.

**Author Contributions:** Conceptualization, J.-H.P., J.-M.S., and H.-M.L.; Data curation, J.-H.P. and J.-G.C.; Funding acquisition, J.-G.C. and I.-H.P.; Investigation, J.-H.P., J.-M.S., and H.-M.L.; Methodology, J.-H.P. and I.-H.P.; Validation, J.-H.P. and H.-W.Y.; Writing—original draft, J.-H.P. and I.-H.P.; Writing—review & editing, T.H.K., S.H.L., J.-M.S., and I.-H.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (2020R1F1A1071985). This research was supported and by a Korea University grant (K1912871).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
