Alpha-tocopherol exerts protective function against the mucotoxicity of particulate matter in amphibian and human goblet cells

Exposure to particulate matter (PM) in ambient air is known to increase the risk of cardiovascular disorders and mortality. The cytotoxicity of PM is mainly due to the abnormal increase of reactive oxygen species (ROS), which damage cellular components such as DNA, RNA, and proteins. The correlation between PM exposure and human disorders, including mortality, is based on long-term exposure. In this study we have investigated acute responses of mucus-secreting goblet cells upon exposure to PM derived from a heavy diesel engine. To this end, we employed the mucociliary epithelium of amphibian embryos and human Calu-3 cells to examine PM mucotoxicity. Our data suggest that acute exposure to PM significantly impairs mucus secretion and results in the accumulation of mucus vesicles in the cytoplasm of goblet cells. RNA-seq analysis revealed that acute responses to PM exposure significantly altered gene expression patterns; however, known regulators of mucus production and the secretory pathway were not significantly altered. Interestingly, pretreatment with α-tocopherol nearly recovered the hyposecretion of mucus from both amphibian and human goblet cells. We believe this study demonstrates the mucotoxicity of PM and the protective function of α-tocopherol on mucotoxicity caused by acute PM exposure from heavy diesel engines.

The embryonic epithelium of amphibians is composed of mucus-secreting goblet cells and multi-ciliated cells which protect the animal from environmental challenges such as infectious microbes and contaminants [15][16][17][18] . Similarities between the embryonic mucociliary epithelium of amphibians and the human airway epithelium enables relevant in vivo approaches which are not feasible using other experimental models such as mice or immortalized cell lines. Amphibian embryos can be easily obtained because a single female can be induced to ovulate by injecting human chorionic gonadotrophin (HCG), after which it will lay several hundred eggs. In vitro fertilization synchronizes the developmental process, and the mucociliary epithelium develops within 2 days after fertilization. Additionally, the mucociliary epithelium is exposed to the outer skin, making it an experimental model of choice. Previously, we have shown that the embryonic epithelium of Xenopus laevis is an alternative in vivo model to study the pathophysiology of mucociliary epithelium and perform high-throughput drug screening for muco-active reagents 16 .
In this study, we have examined the acute toxicity of PM from a heavy diesel engine to the mucociliary epithelium using amphibian embryos and human goblet cells. Our data demonstrate that a reduction in mucus secretion from goblet cells is a conserved and acute response to PM exposure, the response of which may be relieved by α-tocopherol.

Results
the mucociliary epithelium of Bombina orientalis is favorable for detecting acute mucus secretion response to exogenous stimuli. Previous studies demonstrated that exposure to PM results in DNA damage by increasing ROS. In addition, long-term exposure to PM was shown to damage cardiovascular systems, respiratory tracts, and increase the risk of cancer and mortality 6,[19][20][21][22] . However, the response of the respiratory tract to acute PM exposure is not fully understood. A recent study suggests that transcriptional responses to acute PM exposure significantly change the gene expression profiles of human bronchial epithelial cell lines 23 . This suggests that the defensive function of the mucociliary epithelium may be compromised by acute PM exposure before the accumulation of oxidative damage by ROS and associated consequences of long-term PM exposure. However, the current research model is not an appropriate system to examine acute responses of mucociliary epithelium due to its limited availability and the complexity of analysis tools.
In a previous study, we developed an alternative research model for studying mucus secretion and successfully identified potential muco-active reagents using the embryonic mucociliary epithelium of the amphibian X. laevis 16 . To evaluate the effect of PM in the environment, we used another amphibian embryo model to analyze the acute response of mucociliary epithelium upon exposure to PM from a heavy diesel engine. Like X. laevis, the embryonic epithelium of Bombina orientalis was examined and found to have mucus-secreting goblet and multi-ciliated cells very similar to the human mucociliary epithelium in the airway tract (Fig. 1A,B). We further examined if the mucociliary epithelium is physiologically similar to that of human airway epithelium by treating with known muco-active reagents. The mucus secretion level was measured by WGA-HRP (HRP-conjugate wheat germ agglutinin) as previously described 16 .
Known muco-regulators such as bicuculline and phorbol 12,13-dibutyrate significantly upregulated mucus secretion from the embryonic epithelium of B. orientalis (Fig. 1C,D). Narasin also significantly reduced mucus secretion as previously shown (Fig. 1E) 16 . These data suggest that the embryonic epithelium of B. orientalis offers similar utility as X. laevis for studying the physiology of mucociliary epithelium. We have also found that B. orientalis embryos are more favorable for embryonic manipulation due to their large size compared to X. laevis embryos.

pM decreases mucus secretion from goblet cells in the mucociliary epithelium of B. orientalis.
Next, we performed a series of experiments to investigate the acute response of mucus secretion to PM exposure. To examine the effects of PM on mucociliary epithelium, we treated B. orientalis embryos with increasing doses of PM from a heavy diesel engine. Secreted mucus from embryonic mucociliary epithelium 3 h after PM treatment in nursing media was adhered to a multi-well plate, and adhered mucus levels were measured by enzyme-linked lectin assay (ELLA) using WGA-HRP, as previously described 16 .
ELLA results indicated that increasing doses of PM significantly reduced secreted mucus levels from embryonic mucociliary epithelium rather than increasing mucus secretion ( Fig. 2A). We further examined whether this reduction in mucus secretion was due to the cytoplasmic accumulation of mucus vesicles by staining with WGA-Alexa 488. Indeed, PM-treated embryos retained an increased number of mucus vesicles in their cytoplasm compared to those of the control (Fig. 2B,C).
We next confirmed the mucotoxicity of PM using X. laevis embryos. Consistent with the B. orientalis data, PM treatment significantly decreased mucus secretion. (Fig. 2D). The reduction in mucus secretion was mainly due to the defective secretion of mucus vesicles as was seen in B. orientalis embryos, in which WGA-labeled mucus vesicles accumulated in the cytoplasm of goblet cells that significantly worsened with increasing doses of PM ( Fig. 2E-G). the impact of acute pM exposure on mucus secretion-related gene expression. We examined the molecular mechanism for PM mucotoxicity in mucociliary epithelium by analyzing changes in gene expression profiles upon PM treatment. RNA-seq and gene enrichment analysis for differentially expressed genes (DEG) after PM treatment revealed that the transcriptome of PM-exposed embryos was significantly changed compared to control embryos (Fig. 3A, Supplementary Table S1). However, GO term pathway analysis on the DEG revealed that mucus-related genes were not significantly affected by the acute PM exposure (Supplementary Table S2). We did confirm that the expression of otogelin, a known mucus component, was slightly reduced by PM treatment, although without statistical significance (Fig. 3B). These data suggest that the mucociliary epithelium is a vulnerable target of acute PM exposure; however, changes to gene expression involving mucus secretion or production were not the major cause of defective mucus secretion induced by exposure to PM.  www.nature.com/scientificreports www.nature.com/scientificreports/ α-tocopherol protects against pM-induced hyposecretion of mucus in X. laevis and B. orientalis.

Goblet
The major cause of cytotoxicity upon PM exposure is caused by excessive ROS. We therefore reasoned that antioxidant treatment may recover the reduced mucus secretion in PM-matter treated embryonic epithelium. For this experiment, we used three well-known antioxidants: Trolox, NAC (N-acetyl cysteine), and α-tocopherol.
To this end, we pretreated X. laevis embryos with various doses of antioxidants for 1 h, followed by 3 h of PM treatment. Mucus secretions were then compared to those of PM-treated embryos without antioxidant pretreatment (Fig. 4). To our surprise, trolox and NAC had minimal effects on mucus secretion recovery; however, α-tocopherol-treated embryos displayed mucus recovery to normal levels of secretion (Fig. 5A).
Next, we examined if α-tocopherol induces mucus secretion recovery by facilitating mucus vesicle secretion. Consistent with previous data, mucus vesicles accumulated in the cytoplasm upon treatment with PM. However, pretreatment with α-tocopherol prior to PM exposure significantly reduced the accumulation of mucus vesicles ( Fig. 5B-D). We also examined the protective effects of α-tocopherol using B. orientalis embryos and obtained similar results (Fig. 6A-C). These data indicate that reductions in mucus secretion may be a common cytotoxic outcome of PM exposure for mucus-secreting goblet cells. Also, our data suggest that α-tocopherol protects mucociliary epithelium from PM mucotoxicity. www.nature.com/scientificreports www.nature.com/scientificreports/ ROS are not the major cause of PM-induced mucus hyposecretion. We speculated that increased ROS may be a direct cause of mucus hyposecretion, and pretreatment with α-tocopherol could recover secretion by reducing ROS levels. To this end, we measured ROS levels induced by PM treatment. Consistent with previous reports, PM increased cellular ROS levels measured by H 2 DCFDA (Supplementary Fig. 1). We then pretreated X. laevis embryos with the antioxidants trolox, NAC, and α-tocopherol for 1 h, followed by 3 h of PM treatment. To our surprise, among the antioxidants, trolox pretreatment effectively recovered PM-induced ROS to normal levels, while α-tocopherol did not affect ROS (Supplementary Fig. 1). Next, we measured other known ROS-dependent cytotoxicities of PM such as cell death and DNA damage by performing TUNEL and γ-H2AX www.nature.com/scientificreports www.nature.com/scientificreports/ immunostaining assays, respectively. However, acute PM exposure did not significantly cause cell death or DNA damage in our in vivo X. laevis model (Supplementary Fig. 2). These data indicate that PM-induced ROS are not directly affecting mucus hyposecretion, per se. Rather, there are unknown components in PM that may cause hyposecretion of mucus from the mucociliary epithelium. www.nature.com/scientificreports www.nature.com/scientificreports/ α-tocopherol protects against PM-induced mucus hyposecretion in Calu-3 human goblet cells.
We next examined the mucotoxic effects of PM in human cells. Calu-3 cells have been frequently used to study the pathophysiology of goblet cells 24 . We treated Calu-3 cells with PM and analyzed the accumulation of cytoplasmic mucus vesicles using WGA-Alexa 488. Consistent with data obtained using embryonic mucociliary epithelium in B. orientalis and X. laevis, acute PM exposure significantly increased cytoplasmic mucus granules (Fig. 7A). We also confirmed that the accumulated cytoplasmic vesicles contained mucin by performing fluorescent immunostaining using an anti-MUC5AC antibody (Fig. 7B). Indeed, MUC5AC-positive mucus vesicles were significantly enriched in PM-treated Calu-3 cells (Fig. 7B,D). Furthermore, pretreating with α-tocopherol rescued the abnormal accumulation of mucus granules, which were examined by either WGA-Alexa 488 or anti-MUC5AC staining (Fig. 7A-D). These data strongly suggest that the abnormal accumulation of mucus granules in goblet cells is a common response to PM exposure, which reduces the level of secreted mucus from the mucociliary epithelium including human airway epithelium.

Discussion
PM in the ambient air is known to contain biologically toxic compounds such as PAH and transition metals. These toxic compounds in PM penetrate mucus barriers in the mucociliary epithelium of the human airway tract and directly affect human health. The major cause of cytotoxicity from PM is a known increase in ROS. Abnormal increases in ROS damage cellular components such as DNA and proteins, thereby causing mutations to accumulate. This cytotoxicity consequently increases the risk of cardiovascular and pulmonary disorders. The correlation between PM exposure and human disorders is based on long-term exposure. However, recent studies also suggest that the acute response to PM also causes significant changes in gene expression and the physiology of live organisms.
The mucociliary epithelium in the human airway tract is an essential tissue that functions as a defense mechanism against environmental challenges such as pathogens and toxic pollutants. Therefore, most respiratory disorders, such as COPD, asthma, and respiratory infections, often disrupt normal airway mucociliary function. In this study, we examined the acute response of the mucociliary epithelium to PM exposure from heavy diesel engines. To this end, we utilized amphibious embryonic skin models. Our data suggest that acute exposure to PM significantly impairs mucus secretion and results in the accumulation of mucus vesicles within the cytoplasm of goblet cells. In addition, mucus hyposecretion from the mucociliary epithelium upon PM exposure may be a consequence of disrupting the mucus vesicle secretory pathway. Consistent with this idea, RNA-seq analysis demonstrated that the expression of genes involved in mucus production pathways were not significantly affected by PM exposure.
Interestingly, pretreatment of α-tocopherol elicited a near complete recovery of mucus hyposecretion from the mucociliary epithelium. Furthermore, consistent with the amphibian data, pretreatment of α-tocopherol completely reversed PM-induced mucus accumulation in human Calu-3 cells. These data imply that PM may cause very common toxicity across mucus-secreting goblet cells, and this mucotoxicity is an immediate response rather than a consequence of long-term exposure to PM.
Although the major cytotoxic components of PM are PAH and transition metals, PM has very diverse composition depending on many environmental factors such as cities, seasons, and air pollutants. Therefore, one limitation of this study is that the composition of PM, exhaust from heavy diesel engines (purchased from Sigma-Aldrich), is not an exact match to naturally occurring PM without a defined chemical composition. For this reason, we were unable to identify the component responsible for mucotoxicity on the mucus-secreting cells. Further research needs to be performed to identify the most important mucotoxic components of PM.
Altogether, we believe this study clearly demonstrates that mucus hyposecretion is a conserved and common response of mucociliary epithelium to PM exposure. Furthermore, our data present clear evidence that α-tocopherol has a protective capacity against the mucotoxicity of PM from heavy diesel engines (Fig. 8).

Materials and Methods embryo manipulation. Xenopus laevis and
Bombina orientalis adult females were ovulated using 380 units of human chorionic gonadotropin (HCG) a day before. For B. orientalis, an adult male was also ovulated using 280 units of HCG a day before for natural mating. For X. laevis egg fertilization, a testicle was dissected from an adult male, ground, and sprayed on the squeezed X. laevis eggs. To remove the jelly layer, a 3% cysteine pH 7.9 solution was used for X. laevis embryos and a 3% cysteine pH 9.5 solution for B. orientalis embryos. Fertilized X. laevis embryos were grown in 1/3× MMR and fertilized. B. orientalis embryos were grown in sterilized water until used. The Institutional Animal Care and Use Committee (IACUC) of Ulsan National Institute of Science and Technology (UNIST) approved this work (Reference number, UNISTIACUC- . All methods were performed in accordance with the relevant guidelines and regulations enzyme-linked lectin assay (eLLA) for secreted mucus detection. Embryos were washed using sterilized 1/3× MMR buffer for X. laevis and sterilized water for B. orientalis to remove secreted mucus particles which remain on the epithelium. After washing, five embryos were incubated with each compound. Treatments consisted of bicuculline and narasin at 10 μM and phorbol 12,13-dibutyrate at 1 μM for 3 h. Antioxidant pretreatments occurred for 1 h, after which diesel PM treatment was carried out for 3 h. Media was collected as a time series into a U-shaped bottom 96-well plate. Collected media was incubated overnight with sodium azide at 4 °C. The next day, the incubated 96-well plates were washed twice using PTW (1 × PBS, 0.1% Tween-20), then blocked using 1% BSA for 1 h. After blocking, HRP-conjugated wheat germ agglutinin (WGA) was diluted to 1/10,000 in 1% BSA and also incubated for 1 h. After incubation, the 96-well plates were washed twice using PTW. For chromogenic detection, one tablet of o-phenylenediamine dihydrochloride (SIGMA P8289) was dissolved (2020) 10  www.nature.com/scientificreports www.nature.com/scientificreports/ cell culture. Calu-3, which is a human lung epithelial cell line (ATCC HRB-55), was cultured using DMEM (Gibco 11995) in 5% carbon dioxide. Cells were harvested using trypsin-EDTA (Gibco, 25200) and spun down at 1,300 rpm for 5 min. For diesel PM treatments, 4 × 10 5 cells were cultured on a coverslip for fixation and immunofluorescence analysis.
Harvested cells were washed with PBS and stained using CM-H2DCFDA (Invitrogen, C6827) per manufacturer's protocol. Experiments were performed using a BD Biosciences LSRFortessa cell analyzer. Data were analyzed using FlowJo software (BD Biosciences). transcriptome analysis. Total RNA was extracted using Trizol reagent (Thermo Fisher, 15596018) according to manufacturer's protocol. To eliminate DNA, samples were purified using PCI/Ethanol purification and lithium chloride methods. Library preparation and RNA sequencing were performed by Theragen using a TruSeq stranded mRNA sample prep kit (Illumina, 20020595) and HiSeq2500 (Illumina). Raw data were aligned to the X. Laevis transcripts downloaded from Xenbase (http://xenbase.org/, RRID:SCR_003280) using BWA (version 0.7.17) and differentially expressed genes (DEG) were analyzed using edgeR (version 3.22.5). DEG were sorted as greater than 2-fold change and less than 0.01 false discovery rate (FDR).  . Schematic of the protective effect of α-tocopherol on particulate matter mucotoxicity. Our data suggest that acute exposure to PM causes mucus hyposecretion, likely through interference with secretory and exocytic processes in goblet cells. α-tocopherol offers a protective function over PM mucotoxicity. PM, particulate matter.