Ceramide Nanoliposomes as Potential Therapeutic Reagents for Asthma

Ceramides are an emerging class of anti-inflammatory lipids, and nanoscale ceramide-delivery systems are potential therapeutic strategies for inflammatory diseases. This study investigated the therapeutic effects of ceramide nanoliposomes (CNL) on type 2 inflammation-based asthma, induced by repeated ovalbumin (OVA) challenges. Asthmatic mice intratracheally treated with ceramide-free liposomes (Ghost) displayed typical airway remodeling including mucosal accumulation and subepithelial fibrosis, whereas, in CNL-treated mice, the degree of airway remodeling was significantly decreased. Compared to the Ghost group, CNL treatment unexpectedly failed to significantly influence formation of type 2 cytokines, including IL-5 and IL-13, known to facilitate pathogenic production of airway mucus predominantly comprising MUC5AC mucin. Interestingly, CNL treatment suppressed OVA-evoked hyperplasia of MUC5AC-generating goblet cells in the airways. This suggests that CNL suppressed goblet cell hyperplasia and airway mucosal accumulation independently of type 2 cytokine formation. Mechanistically, CNL treatment suppressed cell growth and EGF-induced activation of Akt, but not ERK1/2, in a human lung epithelial cell culture system recapitulating airway goblet cell hyperplasia. Taken together, CNL is suggested to have therapeutic effects on airway remodeling in allergic asthma by targeting goblet cell hyperplasia. These findings raise the potential of ceramide-based therapies for airway diseases, such as asthma.


Introduction
Asthma is a chronic inflammatory disease characterized by inflammation-associated airway remodeling and hyperresponsiveness [1,2]. Asthma pathogenesis is associated with type 2 inflammation together with key effecter cells, such as T-helper 2 (Th2) cells, dendritic cells, group 2 innate lymphoid cells (ILC2s), airway epithelial cells, and eosinophils [3]. These cells contribute to the multiple features of allergic inflammation by secreting a myriad of proinflammatory mediators that deteriorate vasodilation, vascular permeability, airway smooth muscle contraction, mucus secretion, and immune cell recruitment [4]. Thus, given the important roles of type 2 cytokines such as interleukin (IL)-4, IL-5, and IL-13, in pathogenesis, they are attractive therapeutic targets [2,3,5].

CNL Preparation
Preclinical development of the CNL has previously been described [23]. Briefly, lipids (distearoylphosphatidylcholine, dioleoylphosphatidylethanolamine, distearoylphos phatidylethanolamine-polyethylene glycol (PEG) 2000, PEG750-C 8 -ceramide with or without C 6 -ceramide to form the CNL or Ghost, respectively) were dissolved and mixed in chloroform, dried to a thin film under nitrogen, and then hydrated by addition of saline at 60 • C with sonication and vortexing. Lipid solutions were then extruded at 60 • C by passing through 100 nm polycarbonate filters. Size and charge were validated using a Malvern Zetasizer Nano (Malvern Panalytical, UK).

Sensitization and Challenges for Asthmatic Model
The asthmatic model was established as previously reported [28]. Briefly, 5-weekold BALB/c mice (Japan SLC, Hamamatsu, Japan) were sensitized by i.p. injections with OVA adsorbed to Al(OH) 3 at a dose of 50 µg of OVA/2 mg Al(OH) 3 /0.5 mL of saline/animal/time on days 0, 14 and 28. The sensitized mice were intratracheally challenged with OVA-free PBS or OVA (5 µg/25 µL/animal/time/day) on days 33, 34, 35, and 38. Intratracheal administration was conducted by insertion of a polyethylene tube from oral cavity under inhalation anesthesia with isoflurane as reported previously [29,30]. PBS, Ghost and CNL were intratracheally administered according to the schedule ( Figure 1). Amounts of 10 or 30 µg CNL correspond to 1.4 or 4.2 µg C 6 -ceramide, respectively.

Ethics Statement
All animal studies were approved by the Experimental Animal Research Committee at Setsunan University (Hirakata, Japan).

Ethics Statement
All animal studies were approved by the Experimental Animal Research Committee at Setsunan University (Hirakata, Japan).

Bronchoalveolar Lavage (BAL) Fluid Collection
BAL fluid analysis was performed as per previous studies [28,31]. Total leukocyte numbers were counted following hemolysis with ammonium-chloride-potassium (ACK) hemolysis buffer. Cells were settled on glass slides and stained with Diff-Quik solution (Sysmex International Reagent, Kobe, Japan). Eosinophils and neutrophils were observed by light microscope.

Immunohistochemistry
After the final challenge, mice were perfused with 20 mL PBS and 50 mL 10% formalin under anesthesia with pentobarbital and xylazine. Left lung lobes were isolated and cut into 3 sections. Formalin-fixed tissues were embedded in paraffin, and the 4-µm sections were stained using antibodies specific for MUC5AC, phospho-p38, and phospho-JNK along with the VECTASTAIN Elite ABC kit. Paraffin sections were deparaffinated with xylene and hydrophilized by 70% to 100% ethanol. After washing with water for 10 min, antigen retrieval was performed using HistoVT one according to the manufacture's instruction. After washing with PBS containing 0.05% Tween 20 (PBS/0.05% Tween 20), the sections were treated with 3% hydrogen peroxide-methanol for 30 min. After washing with PBS/0.05% Tween 20, the sections were blocked with rabbit or goat normal serum for 30 min at room temperature. The sections were then treated with primary antibodies (MUC5AC, 1:100; phospho-p38, 1:100; and phospho-JNK 1:100) overnight at 4 • C. After washing with PBS/0.05% Tween 20, the sections were treated with biotin-conjugated antigoat IgG antibody or anti-rabbit IgG antibody at room temperature for 30 min. After washing with PBS/0.05% Tween 20, lung sections were treated with avidin-HRP for 30 min at room temperature. After washing, the sections were incubated with ImmPACT™ DAB and counter-stained with hematoxylin. After staining, cell numbers were counted using the Hybrid Cell Count application (Keyence, Osaka, Japan).

Lung Histology
Histological examination was performed according to previous reports [28,31]. In brief, after the final OVA challenge, the left lobes of the lung were fixed with 10% neutral buffered formalin. Tissues were embedded in paraffin, and 4-µm sections were stained with Masson trichrome and periodic acid-Schiff (PAS). Histological changes were semi-quantitatively scored as described [31]. The degree of epithelial thickening was quantified using PASstained sections as follows: the bronchus (the diameter > 150 µm) was photographed, and then epithelial area (µm 2 ) and basement membrane length were measured using version 1.42 Image J (NIH, Bethesda, MD, USA). The epithelial area was then divided by the basement membrane length. For quantification of the epithelial layer mucus accumulation, PAS-stained areas of the epithelium (µm 2 ) were measured using Adobe Photoshop (Adobe Systems, San Jose, CA, USA) and version 1.42 Image J. The PAS-stained area was then divided by the length of the basement membrane. To assess subepithelial fibrosis, Masson trichrome staining was performed. Subepithelial fibrosis was assessed by the extent of the blue fibrotic area, and scored on a 7-point graded scale from 0 to 3 based on the distance that the blue fibrotic area extended underneath the basement membrane: zero, no blue area underneath the basement membrane; then 0.5, 1, 1.5, 2, 2.5, and 3, corresponding extensions of less than 15 µm, 15-30 µm, 30-45 µm, 45-60 µm, 60-75 µm, or more than 75 µm, respectively. All specimens were analyzed in a blind manner.

ELISA
ELISA was performed according to previous reports [28,31]. Briefly, the left lung lobes in T-PER containing protease inhibitor cocktails were homogenized with the polytron homogenizer. After centrifugation, supernatants were stored at −80 • C until measurement of type 2 cytokines (IL-4, IL-5, and IL-13) by ELISA. Type 2 cytokine concentrations were normalized relative to protein.

Analysis of ILC2, Treg, and Tr1 Cells
The left lungs were rinsed with PBS and tissues digested by 870 units/mL collagenase type I for 1 h at 37 • C. Cells were dispersed with a syringe then filtered through a 108 µm nylon mesh. After centrifugation, cells were treated with ACK lysis buffer to remove erythrocytes. The total leukocyte cell number was counted by trypan blue staining. ILC2s were defined as Lineage -CD45 + CD278 + CD90.2 + ST2 + cells [28]. Treg cells were defined as CD4 + CD3 + CD25 + Foxp3 + cells [32]. Tr1 cells were defined as CD4 + CD3 + IL-10 + Foxp3 − cells [32] after the OVA stimulation.
For Treg cell detection, lung cells were treated with anti-CD16/32 antibodies. After washing with PBS/2% FBS, cells were treated with FITC-conjugated anti-mouse CD4, PE/Cy7-conjugated anti-mouse CD3, and PE-conjugated anti-mouse CD25 antibodies. After incubation for 20 min at 4 • C, cells were washed with PBS/2% FBS and then incubated with Fixation/Permeabilization Concentrate for 12 h. After washing with permeabilization buffer, cells were stained with eFluor ® 660-conjugated anti-mouse Foxp3 antibody for 30 min at 4 • C. After washing with PBS/2% FBS, the stained cells were analyzed using FACSAria™ Fusion.
To detect Tr1 cells, lung cells were seeded on a plate. The cells were stimulated with 10 mg/mL OVA for 6 h at 37 • C followed by treatment with 2 µM monensin sodium salt. After washing with PBS/2% FBS, cells were treated with an anti-CD16/32 antibody. The cells were further stained with FITC-conjugated anti-mouse CD4 and PE/Cy7-conjugated anti-mouse CD3 antibodies for 20 min at 4 • C. After washing with PBS/2% FBS, cells were incubated with Fixation/Permeabilization Concentrate for 12 h. After washing with permeabilization buffer, the cells were stained with eFluor ® 660-conjugated anti-mouse Foxp3 antibody and PE-conjugated anti-mouse IL-10 antibody for 30 min at 4 • C. After washing with PBS/2% FBS, the cells were analyzed using the FACSAria™ Fusion.

Cell Culture
Human lung epithelial A549 cells were grown in DMEM supplemented with 10% FBS. The cells were maintained at <80% confluence under standard incubator conditions (humidified atmosphere, 95% air, 5% CO 2 , and 37 • C). Mycoplasma contamination was not observed in the cell lines.

Cell Viability Assay
A549 cells (4 × 10 4 cells/well) were seeded on 96-well culture plate. Cell viability was determined using a CellTiter-Glo luminescent cell viability assay according to the manufacturer's protocol.

Statistical Analysis
Statistical analysis was performed with Graphpad Prism version 8. A one-way analysis of variance was performed. If significant differences were detected, individual differences were determined by Dunnett's test. The unpaired t-test was used to compare two groups.

Results
The potential of liposomal ceramides, CNL, to act as an anti-asthmatic reagent was investigated. Ceramide-free ghost liposomes (Ghost) or CNL were intratracheally administered to mice ( Figure 1). Firstly, airway inflammation was assessed by infiltration of inflammatory cells to BAL fluids. PBS/OVA-challenges increased the number of neutrophils, mononuclear cells and eosinophils in BAL fluids. At 10 or 30 µg/mouse CNL, OVA-induced infiltration of leukocytes such as neutrophils and mononuclear cells, was suppressed in comparison to the Ghost group (Figure 2A,B). In 10 µg/mouse CNL-treated mice, there was less eosinophil infiltration compared to the Ghost group ( Figure 2C). The number of eosinophils in BAL fluids of 30 µg/mouse CNL group were comparable with the Ghost group.
Airway inflammatory responses are associated with airway histological changes, referred to as airway remodeling. Airway remodeling is a pathological feature of asthma, characterized by goblet cell hyperplasia-governed mucosa accumulation, bronchial epithelial thickening, subepithelial fibrosis, and hyperplasia of airway smooth muscle cells. To assess CNL effects on airway remodeling, lung tissues were subjected to PAS and Masson trichrome staining ( Figure 3A). OVA-challenged mice displayed airway remodeling characterized by up-regulation of bronchial epithelial thickening, epithelial mucus accumulation, and subepithelial fibrosis ( Figure 3B-D). CNL treatment significantly suppressed the progression of airway remodeling, though significant dose-dependent responses to CNL were not observed. suppressed in comparison to the Ghost group (Figure 2A,B). In 10 μg/mouse CNL-treated mice, there was less eosinophil infiltration compared to the Ghost group ( Figure 2C). The number of eosinophils in BAL fluids of 30 μg/mouse CNL group were comparable with the Ghost group. Airway inflammatory responses are associated with airway histological changes, referred to as airway remodeling. Airway remodeling is a pathological feature of asthma, characterized by goblet cell hyperplasia-governed mucosa accumulation, bronchial epithelial thickening, subepithelial fibrosis, and hyperplasia of airway smooth muscle cells. To assess CNL effects on airway remodeling, lung tissues were subjected to PAS and Masson trichrome staining ( Figure 3A). OVA-challenged mice displayed airway remodeling characterized by up-regulation of bronchial epithelial thickening, epithelial mucus accumulation, and subepithelial fibrosis ( Figure 3B-D). CNL treatment significantly suppressed the progression of airway remodeling, though significant dose-dependent responses to CNL were not observed.
CD4 + regulatory T (Treg) cells are important in promoting immune tolerance to allergens and preventing allergic diseases [33]. Treg cells are classified into two subsets, Foxp3 + Treg cells or Foxp3 − Tr1 cells, which strongly produce the anti-inflammatory cytokine IL-10 [34]. In previous studies using the OVA asthmatic mouse model, Tr1 cells were shown to suppress infiltration of eosinophils and neutrophils into BAL fluids, airway remodeling, and IL-5 up-regulation [32]. We wondered if CNL affected immune tolerance. In response to OVA challenges, the number of Foxp3 + Treg and Tr1 cells in lung tissues increased (Table 1). However, CNL treatment had no significant effect on the number of cells, instead slightly decreasing numbers. Therefore, Treg subsets are unlikely to contribute to the anti-asthmatic effect of CNL.   ILC2s are implicated in the pathogenesis of type 2 inflammation-associated allergic asthma [35]. In response to airway epithelial cell-derived cytokines such as IL-33 and CD4 + regulatory T (Treg) cells are important in promoting immune tolerance to allergens and preventing allergic diseases [33]. Treg cells are classified into two subsets, Foxp3 + Treg cells or Foxp3 − Tr1 cells, which strongly produce the anti-inflammatory cytokine IL-10 [34]. In previous studies using the OVA asthmatic mouse model, Tr1 cells were shown to suppress infiltration of eosinophils and neutrophils into BAL fluids, airway remodeling, and IL-5 up-regulation [32]. We wondered if CNL affected immune tolerance. In response to OVA challenges, the number of Foxp3 + Treg and Tr1 cells in lung tissues increased (Table 1). However, CNL treatment had no significant effect on the number of cells, instead slightly decreasing numbers. Therefore, Treg subsets are unlikely to contribute to the anti-asthmatic effect of CNL. Table 1. Effects of ceramide nanoliposomes (CNL) on the number of ILC2, Treg, and Tr1 cells in lung tissues. OVA-sensitized mice were challenged with OVA-free PBS or OVA. OVA-challenged asthmatic mice were treated with 30 µg/animal ceramide-free Ghost or CNL. Cell numbers of ILC2, Treg, and Tr1 cells isolated from lungs were determined. Data are shown as mean ± S.E. ILC2s are implicated in the pathogenesis of type 2 inflammation-associated allergic asthma [35]. In response to airway epithelial cell-derived cytokines such as IL-33 and thymic stromal lymphopoietin, ILC2s produce large quantities of type 2 cytokines [28,36,37]. OVA challenges evoked significant increases in the number of lung ILC2s, though CNL treatment had no significant effect (Table 1).

Cell
Type 2 cytokines, predominantly produced by ILC2, play key roles in airway remodeling and are emerging therapeutic targets [3]. We assessed CNL effects on the formation of IL-4, IL-5, and IL-13 in OVA-challenged lungs. In OVA-challenged mice, the formation of IL-5 and IL-13, but not IL-4, was significantly increased in lung homogenates, which was unsuppressed by CNL treatment (Figure 4). Those results suggest that CNL prevents mucus secretion and fibrosis in the lung epithelium without affecting levels of mucus secretion-stimulating cytokines IL-5 and IL-13.  Airway mucus is composed of mucus glycoproteins (mucins) including MUC5AC and MUC5B. Among them, MUC5AC accounts for approximately 90% of the mucin content of sputum [38]. Goblet cells in the bronchiolar epithelium predominantly secretes airway MUC5AC, with goblet cell hyperplasia and the associated mucus hypersecretion contributing to asthma airway remodeling [39][40][41]. In our asthma model, goblet cell hyperplasia, represented by an increased number of MUC5AC-positive goblet cells in the airway epithelium, was observed in OVA-challenged mice ( Figure 5A). Importantly, CNL treatment significantly down-regulated MUC5AC-positive goblet cell numbers in airways to approximately 65% that observed after Ghost treatment, demonstrating the inhibitory effect of CNL on goblet cell hyperplasia ( Figure 5B). As intratracheally administered CNL is logically thought to act on bronchiolar epithelial cells, airway goblet cell hyperplasia is a potential anti-asthmatic target for CNL. To model the effects of CNL on goblet cell hyperplasia, human epithelial cells recapitulating bronchial epithelial cells, including goblet cells, were employed. The epithelial cells cultured with 10% FBS showed time-dependent cell growth, which was significantly suppressed by CNL treatment (Figure 5C). Trypan Airway mucus is composed of mucus glycoproteins (mucins) including MUC5AC and MUC5B. Among them, MUC5AC accounts for approximately 90% of the mucin content of sputum [38]. Goblet cells in the bronchiolar epithelium predominantly secretes airway MUC5AC, with goblet cell hyperplasia and the associated mucus hypersecretion contributing to asthma airway remodeling [39][40][41]. In our asthma model, goblet cell hyperplasia, represented by an increased number of MUC5AC-positive goblet cells in the airway epithelium, was observed in OVA-challenged mice ( Figure 5A). Importantly, CNL treatment significantly down-regulated MUC5AC-positive goblet cell numbers in airways to approximately 65% that observed after Ghost treatment, demonstrating the inhibitory effect of CNL on goblet cell hyperplasia ( Figure 5B). As intratracheally administered CNL is logically thought to act on bronchiolar epithelial cells, airway goblet cell hyperplasia is a potential anti-asthmatic target for CNL. To model the effects of CNL on goblet cell hyperplasia, human epithelial cells recapitulating bronchial epithelial cells, including goblet cells, were employed. The epithelial cells cultured with 10% FBS showed time-dependent cell growth, which was significantly suppressed by CNL treatment ( Figure 5C). Trypan blue exclusion assay revealed that 3 µM CNL had no significant effects on cell viability. Whereas 10 µM CNL, slightly but significantly, decreased the cell viability (Supplementary Figure S1). The short-chain C 6 -ceramide is a known apoptotic agent [42,43]. MCF-7 and HL-60 cells are sensitive to the apoptotic effects of C 6 -ceramide in the 3-10 µM range, whereas A549 cells are more resistant [44]. Moreover, liposomal C 6 -ceramide did not induce apoptosis of non-cancerous epithelial cells in vitro or in vivo in previous studies [21]. Therefore, it is conceivable that CNL with 3-10 µM C 6 -ceramide primarily limits growth and survival for lung epithelial A549 cells.  The proinflammatory kinase p38 has emerged as a potential asthma therapeutic target [47][48][49]. In experimental models, p38 inhibition prevents allergen-induced goblet cell hyperplasia, mucus hypersecretion, and airway hyperresponsiveness [50,51]. In the lung tissues of OVA-challenged mice, the subepithelial regions where infiltrated leukocytes were observed, were positively stained with antibodies specific for phospho/active-p38, EGF is a potent growth factor for epithelial cells, including goblet cells [45,46]. To investigate the molecular mechanism(s) underlying CNL-suppressed cell growth, the effects of CNL on EGF signaling were examined using human lung epithelial A549 cells. Cells were stimulated with EGF for the indicated periods. Phosphorylation/activation of EGF receptors peaked at 10 min. Downstream kinases such as MAPKs and Akt were also phosphorylated/activated. Surprisingly, CNL had no significant effects on phosphorylation of EGFR and MAPKs including ERK1/2, JNK1/2, and p38 ( Figure 5D and Supplementary Figure S2). Whereas CNL potently reduced Akt phosphorylation at the Ser473 and Thr308 residues ( Figure 5D,E). This suggests selective CNL suppression of the EGF receptorgoverned pathway of phosphatidylinositol-3 kinase (PI3K)-Akt, responsible for cell growth. Considering the goblet cell hyperplasia following the differentiation of airway epithelial cells to goblet cells in response to OVA-induced airway inflammation, CNL appears to target and suppress goblet cell hyperplasia and associated mucus hypersecretion.
The proinflammatory kinase p38 has emerged as a potential asthma therapeutic target [47][48][49]. In experimental models, p38 inhibition prevents allergen-induced goblet cell hyperplasia, mucus hypersecretion, and airway hyperresponsiveness [50,51]. In the lung tissues of OVA-challenged mice, the subepithelial regions where infiltrated leukocytes were observed, were positively stained with antibodies specific for phospho/active-p38, but the epithelial cells showed negative-staining ( Figure 6A). CNL treatment drastically suppressed p38 activation ( Figure 6B). but the epithelial cells showed negative-staining ( Figure 6A). CNL treatment drastically suppressed p38 activation ( Figure 6B). JNK is also a proinflammatory kinase associated with asthma promotion. JNK activation is implicated in airway epithelial cell differentiation into goblet cells, increased mucus production, and epithelial cell proliferation [52]. As with evaluating p38 activation, the subepithelial regions, excluding epithelial cells, were stained with antibodies specific for phospho/active-JNK1/2 in Ghost-treated asthmatic mice ( Figure 6C). The CNL treatment suppressed JNK activation by about 50% (Figure 6D). JNK is also a proinflammatory kinase associated with asthma promotion. JNK activation is implicated in airway epithelial cell differentiation into goblet cells, increased mucus production, and epithelial cell proliferation [52]. As with evaluating p38 activation, the subepithelial regions, excluding epithelial cells, were stained with antibodies specific for phospho/active-JNK1/2 in Ghost-treated asthmatic mice ( Figure 6C). The CNL treatment suppressed JNK activation by about 50% (Figure 6D).

Discussion
Insufficient ceramide biosynthesis is a risk factor in non-allergic asthma [9,53]. In the present study, we demonstrated, in a murine model for the first time, the anti-asthmatic effects of intratracheal ceramide administration using nanoliposomal delivery. This mitigated the airway remodeling corresponding to irreversible pathological airway changes. Of the anti-asthmatic effects, noteworthy features of CNL treatment includes the inhibition of both mucosal accumulation and goblet cell hyperplasia in the airways. Mechanistically, CNL was thought to limit cell growth by inactivating EGF-governed AKT pathways, but not MAPK pathways. These findings support ceramide-based anti-inflammatory therapies for novel asthma treatments.
The pathophysiological roles of ceramides in the lung remains poorly understood. Biochemical studies in mouse lungs identified ceramide species compositions and the gene expression profile of ceramide synthases (CERSes) [54]. Very-long-chain C 24 -ceramide is a major species, and the loss of the CerS2 gene, which is responsible for very-long-chain ceramide biosynthesis, leads to significant airway flow obstruction, inflammation, increased lung volume, along with decreased very-long-chain ceramide and increased long-chain ceramide (C 16 -ceramide). Though the physiological roles for distinct ceramide species remain elusive, CERS2-governed biosynthesis and homeostasis of ceramide species are crucial for lung physiology.
Interestingly, Izawa et al. demonstrated that extracellular ceramides interacting with CD300f, an immune inhibitory receptor, limits OVA-induced chronic airway inflammation, characterized by the accumulation of inflammatory granulocytes and goblet cell hyperplasia [20], implicating ceramides as anti-asthmatic lipids. Interaction of CNL-derived ceramides with CD300f might contribute to anti-asthmatic effects of CNL.
In a house dust-mite-challenged asthma mouse model, lung ceramides were increased along with allergic response. Pharmacological inhibition of sphingolipid biosynthesis by intraperitoneal administration of myriocin or fumonisin B1 prevented reactive-oxygen species formation and apoptosis induction in lung tissues, as well as neutrophil recruitment to the lungs in mice [55]. In these studies, the effects of sphingolipid biosynthesis inhibition on asthmatic pathogenesis, such as airway remodeling and airway hyperresponsiveness, were not tested. Moreover, myriocin inhibits SPT, presumably suppressing biosynthesis of all sphingolipids, including the proinflammatory lipid sphingosine-1-phosphate. Sphingosine-1-phosphate and its generating enzyme, sphingosine kinase, are implicated in airway hyperresponsiveness and airway remodeling in asthma [56][57][58]. The molecular mechanisms underlying the inhibitory effects of myriocin on asthmatic inflammation appear complicated. Further extensive studies are needed to clarify the pathobiological roles of sphingolipid metabolism and metabolites in allergic asthma.
It is postulated that bronchial epithelial cells may take up CNL carrying C 6 -ceramide that is subsequently metabolized to sphingosine-1-phosphate. As intratracheal administration of CNL had substantial anti-asthmatic effects, ceramide action appears predominant in airways. Nevertheless, extensive CNL pharmacodynamic analysis is necessary.
Goblet cells are generally sparse in normal lung tissue and goblet cell hyperplasia is evoked by type 2 inflammation [59,60]. CNL is unlikely to significantly affect type 2 cytokine formation in asthmatic lung tissues ( Figure 4). However, CNL acted as a potent inhibitor in goblet cell hyperplasia in asthmatic lung airways ( Figure 5), which suggests that CNL targets asthmatic lung epithelia.
EGFR and IL-13 are believed to stimulate bronchial epithelial cells, including Clara and ciliated cells, to differentiate into goblet cells through the action of SPDEF and FoxA2 [39,61,62]. As CNL was reported to prevent embryonic stem cell differentiation [63], whether CNL suppresses goblet cell differentiation and growth needs to be tested in future studies. Our findings provide insights to establish novel molecular bases to understand the molecular pharmacological roles of CNL in inflammation-governed goblet cell differentiation and hyperplasia. In addition to asthma, goblet-cell hyperplasia is also a critical pathological feature of hypersecretory airway diseases, including chronic obstructive pulmonary disease and cystic fibrosis. Thus, CNL-based therapeutics may be effective against those lung diseases.
The PI3K-Akt pathway is proposed as a target for the growth-suppressing functions of ceramides [64,65]. Akt is activated by PI3K on the plasma membrane [66,67] and inactivated/dephosphorylated by Ser/Thr protein phosphatase PP2A [68]. Ceramides are potent inhibitory lipids toward PI3K, consequently preventing Akt activation [69]. Consistent with our results, C 6 -ceramide treatment selectively suppressed EGF activation of Akt at residues Thr308 and Ser473, but not ERK1/2 in epithelial ovarian cancer cells. Ceramide-activated PP2A is responsible for dephosphorylating Akt [64] and may be involved in the CNL inactivation of Akt. Various ceramide activities coordinately inactivate EGFR-governed PI3K-Akt signaling, which may be the mechanism behind the therapeutic effects of CNL on asthmatic lung epithelia.
Corticosteroid insensitivity is a clinical feature of severe asthma and p38 activation is involved in cellular corticosteroid insensitivity [48,70,71]. Suppression of p38 by CNL may reverse glucocorticoid resistance and improve the anti-inflammatory effects of glucocorticoids. Our future studies will determine therapeutic effectiveness of CNL in severe asthma model.

Conclusions
In conclusion, CNL, in an asthmatic model, was shown to have anti-inflammatory properties therapeutically effective for inhibition of airway remodeling. These novel findings shed light on potential therapeutic strategies for asthma through ceramide-based anti-inflammatory therapies as a novel class of asthma treatment.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cells12040591/s1, Figure S1: Effects of CNL on cell viability in human lung epithelial A549 cells, Figure S2: Effects of CNL on EGF signaling in human lung epithelial A549 cells.