Th 2 cytokines-DUOX 2-ROS-HMGB 1 translocation axis is important in the 6 pathogenesis of allergic rhinitis 7 8

The function of high-mobility group box 1 (HMGB1) varies according to its location. However, the translocation mechanism behind HMGB1 remains unclear. We hypothesize that Th2 cytokines are involved in the translocation of HMGB1 in the upper airway epithelium. We investigated the mechanism behind HMGB1 translocation using Th2 cytokine stimulation and examined the clinical significance of HMGB1 translocation in allergic rhinitis (AR). Cytoplasmic and extracellular HMGB1 were increased in AR. Inhibiting HMGB1 translocation with glycyrrhizic acid (GA) decreased the level of antigen-specific IgE, the degree of periodic acid-Schiff (PAS), and Sirius red staining in the murine model. The in vivo ROS level in the nasal mucosa was higher in the mice with AR than in the controls. Th2 cytokine-induced up-regulation of the reactive oxygen species (ROS) and translocation of HMGB1 by Th2 cytokines was dependent on the generated ROS. The ROS level also increased in the murine model. We suggest that the Th2 cytokine-DUOX2-ROS-HMGB1 translocation axis is important in AR pathogenesis.


INTRODUCTION 57
High mobility group box 1 (HMGB1) is a nuclear protein ubiquitous in a variety of cells, 58 such as monocytes, macrophages, and epithelial cells. Structurally, HMGB1 is composed of 59 215 amino acids that are organized into two DNA binding domains and a negatively charged     119 Primary HNE cells were incubated with human cytokine IL-4 (10 ng/ml, BD Biosciences, 120 East Rutherford, NJ, USA) and IL-13 (10 ng/ml, BD Biosciences) for 24 hours. They were 121 then harvested and lysed with RIPA lysis buffer (Thermo, Rockford, IL, USA). After lysis, 122 Downloaded from http://portlandpress.com/clinsci/article-pdf/doi/10.1042/CS20201212/902230/cs-2020-1212.pdf by guest on 31 January 2021 Clinical Science. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/CS20201212 the cells were centrifuged for 15 minutes at 13,000 rpm and 4°C, and a bicinchoninic acid 123 assay (Thermo) was used to obtain and quantify the supernatant. We prepared equal amounts 124 of protein samples and western blot assays were performed using 12% SDS-PAGE. Primary    Clinical Science. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/CS20201212 secondary antibody (Invitrogen, Waltham, MA, USA) was added to the cells for 1 hour at RT. 145 Finally, the cell-containing membrane of the Transwell® insert was removed and mounted. 146 An FV1000 confocal microscope (Olympus, Tokyo, Japan) was used to photograph the 147 fluorescent images.      210 We quantified the amount of HMGB1 protein using an HMGB1 ELISA Kit (Shino-Test Corp.,

211
Tokyo, Japan) per the manufacturer's protocol. We performed an HDM-specific IgE ELISA using the sandwich method with biotin-coated anti-mouse IgE (BD Biosciences).

214
Statistics 215 Data are shown as the mean ± standard deviation. The differences between the two groups 216 were evaluated by independent two-sample t-tests. The Shapiro-Wilk test was used to 217 evaluate the collected PAS staining and Sirius red staining data, which were not normally 218 distributed. The data were then compared using the Kruskal-Wallis test followed by the

225
Th2 cytokines induce HMGB1 translocation 226 First, we evaluated the effect of Th2 cytokines on HMGB1 translocation. Using the culture 227 medium from the primary HNE cells, we found that treatment with IL-4 or IL-13 induced 228 HMGB1 translocation into the extracellular area, as shown by western blot assay ( Fig. 1 A, 229 B). We used an ELISA assay to confirm the translocation of HMGB1. Its level in the apical 230 secretion was significantly elevated after stimulation with IL-4 or IL-13 (Fig. 1C).  233 We collected human nasal mucosa samples and lavage fluids (Table 1)  finding that HMGB1 was translocated to the cytoplasm and ultimately the extracellular area 235 in Th2-dominant inflammatory conditions. We hypothesized that the HMGB1 translocation 236 might be involved in AR pathogenesis. We performed immunohistochemistry staining in 237 human nasal mucosa samples and found that HMGB1 was positively stained in both the 238 epithelial and subepithelial areas. Cytoplasmic HMGB1 was more frequently observed in AR 239 patients than in control subjects ( Fig. 2A). The ELISA performed using human nasal lavage 240 fluids demonstrated that the HMGB1 level was significantly higher in AR patients than in the 241 controls (Fig. 2B). 244 To determine if HMGB1 translocation plays a role in AR pathogenesis, we created an AR 245 murine model using HDM extracts with and without GA pretreatment (60 mg/kg), which is a 246 direct HMGB1 translocation inhibitor. The HDM-specific serum IgE level was significantly 247 elevated in the AR murine model, which was suppressed by GA pretreatment (Fig. 3A).

248
There was a significantly increased level of translocated extracellular HMGB1 in the nasal 249 lavage fluid in the AR mice and a decreased level in the mice pretreated with GA (Fig. 3B).

250
The immunohistochemistry results identified significant differences in PAS (p < 0.05) and 251 Sirius red staining (p < 0.05) between the AR group and the AR group pretreated with GA 252 (Fig. 3C-E). The AR group pretreated with GA demonstrated significantly fewer goblet cells 253 and eosinophils than the AR group. These findings suggest that extracellularly translocated 254 HMGB1 is essential to AR pathogenesis.  257 We next hypothesized that ROS might be up-regulated after Th2 cytokine stimulation, which 258 is important in HMGB1 translocation. We checked the ROS levels in a dose-and time-259 dependent manner and found that Th2 cytokines induced an up-regulation in ROS level after 260 12 hours and 10 ng/mL of treatment ( Fig. 4A-C). Western blot assays were performed to 261 evaluate the DUOX expression, which are known ROS generators in upper-airway epithelial 262 cells. We found that DUOX2 expression significantly increased after 12 hours of Th2 263 cytokine stimulation (Fig. 4D, E). 266 To determine whether ROS could mediate HMGB1 translocation, we pretreated primary 267 HNE cells with NAC, a ROS scavenger, before stimulating them with Th2 cytokines. ROS 268 elevation was noted with the Th2 cytokine treatment and suppressed by NAC pretreatment, as 269 expected (Fig. 5A, B). We measured the HMGB1 level in the apical secretion of primary 270 HNE cells by ELISA assay and found that the HMGB1 translocation induced by Th2 271 cytokines decreased with NAC pretreatment (Fig. 5C). This result suggests that the  276 To confirm the role of DUOX2 in ROS-dependent HMGB1 translocation, we used shRNA  Figure 1). The Th2 cytokine-induced up-regulation of ROS decreased when 280 DUOX2 was down-regulated by shRNA (Fig. 6A, B). Furthermore, Th2 cytokine-induced 281 HMGB1 translocation in the apical secretion, measured by ELISA assay, also decreased with 282 DUOX2 down-regulation (Fig. 6C). 285 Finally, we used in vivo imaging to confirm that the basal level of in vivo ROS may increase 286 in the AR murine model. We used a chemiluminescent nanoprobe (BioNT) with appropriate 287 physicochemical properties, such as high ROS sensitivity, no background noise, and 288 deliverability to the respiratory tract, for in vivo ROS imaging 16,17 . AR mice exposed and not 289 exposed to GA pretreatment (60, 120 mg/kg) were prepared and subjected to the in vivo ROS 290 imaging with BioNT. The total serum IgE levels in the prepared models confirmed that the 291 AR developed properly and that pretreatment with GA suppressed AR development 292 ( Supplementary Fig. 2). We then compared the in vivo ROS levels in the nasal mucosa and 293 found the ROS level was significantly elevated in the AR mice compared with the control 294 mice (Fig. 7A, B). In the GA-pretreated mice, the ROS level showed a smaller, nonsignificant 295 increase than in the AR mice. HMGB1translocation. This is the first report to our knowledge to evaluate the translocation 307 mechanism of HMGB1 in nasal epithelial cells under allergic inflammatory conditions.

359
In the lower airway, the local HMGB1 level was clinically associated with disease severity, 360 suggesting that HMGB1 could be a noninvasive biomarker reflecting disease status. The 361 sputum HMGB1 level was positively associated with the total serum IgE level in children 362 with asthma, and a significant inverse correlation was observed between sputum HMGB1 363 levels and pulmonary function indices 8 . However, we did not find a significant relationship 364 between HMGB1 and IgE levels in humans or our mouse AR model (data not shown). The upper airway is continuously exposed to various stimuli that affect the ROS level and the 367 function of HMGB1 can change with oxidative stress.

368
Our study has several limitations. First, our study did not identify the downstream 369 immunological mechanisms after HMGB1 translocation. HMGB1 activates dendritic cells, of subjects, and other inflammation-associated factors, such as smoking, were not considered.

375
A large, population-based human study is needed to confirm our conclusions. Third, although 376 we found that IL-4/13 directly increased DUOX2 expression at the protein level, we did not