key: cord-0786408-33nyuht4
authors: Tanaka, Ken-Ichiro; Kubota, Maho; Shimoda, Mikako; Hayase, Tomoko; Miyaguchi, Mamika; Kobayashi, Nahoko; Ikeda, Mayumi; Ishima, Yu; Kawahara, Masahiro
title: Thioredoxin-albumin fusion protein prevents urban aerosol-induced lung injury via suppressing oxidative stress-related neutrophil extracellular trap formation()
date: 2020-10-07
journal: Environ Pollut
DOI: 10.1016/j.envpol.2020.115787
sha: 34cf9326319692f4df6e7e3b96da6fbf053b3ca0
doc_id: 786408
cord_uid: 33nyuht4

The number of deaths from air pollution worldwide is estimated at 8.8 million per year, more than the number of deaths from smoking. Air pollutants, such as PM(2.5), are known to induce respiratory and cardiovascular diseases by inducing oxidative stress. Thioredoxin (Trx) is a 12-kDa endogenous protein that exerts antioxidant activity by promoting dithiol disulfide exchange reactions. We previously synthesized human serum albumin-fused thioredoxin (HSA-Trx), which has a longer half-life in plasma compared with Trx, and demonstrated its efficacy against various diseases including respiratory diseases. Here, we examined the effect of HSA-Trx on urban aerosol-induced lung injury in mice. Urban aerosols induced lung injury and inflammatory responses in ICR mice, but intravenous administration of HSA-Trx markedly inhibited these responses. We next analyzed reactive oxygen species (ROS) production in murine lungs using an in vivo imaging system. The results show that intratracheal administration of urban aerosols induced ROS production that was inhibited by intravenously administered HSA-Trx. Finally, we found that HSA-Trx inhibited the urban aerosol-induced increase in levels of neutrophilic extracellular trap (NET) indicators (i.e., double-stranded DNA, citrullinated histone H3, and neutrophil elastase) in bronchoalveolar lavage fluid (BALF). Together, these findings suggest that HSA-Trx prevents urban aerosol-induced acute lung injury by suppressing ROS production and neutrophilic inflammation. Thus, HSA-Trx may be a potential candidate drug for preventing the onset or exacerbation of lung injury caused by air pollutants.

ABSTRACT 23

The number of deaths from air pollution worldwide is estimated at 8.8 million per year, more 24 than the number of deaths from smoking. Air pollutants, such as PM 2.5 , are known to induce 25 respiratory and cardiovascular diseases by inducing oxidative stress. Thioredoxin (Trx) is a 26 12-kDa endogenous protein that exerts antioxidant activity by promoting dithiol disulfide 27 exchange reactions. We previously synthesized human serum albumin-fused thioredoxin 28 (HSA-Trx), which has a longer half-life in plasma compared with Trx, and demonstrated its 29 efficacy against various diseases including respiratory diseases. Here, we examined the effect of 30 HSA-Trx on urban aerosol-induced lung injury in mice. Urban aerosols induced lung injury and 31 inflammatory responses in ICR mice, but intravenous administration of HSA-Trx markedly 32 inhibited these responses. We next analyzed reactive oxygen species (ROS) production in murine 33 lungs using an in vivo imaging system. The results show that intratracheal administration of urban 34 aerosols induced ROS production that was inhibited by intravenously administered Finally, we found that HSA-Trx inhibited the urban aerosol-induced increase in levels of 36 neutrophilic extracellular trap (NET) indicators (i.e., double-stranded DNA, citrullinated histone 37 H3, and neutrophil elastase) in bronchoalveolar lavage fluid (BALF). Together, these findings 38 suggest that HSA-Trx prevents urban aerosol-induced acute lung injury by suppressing ROS 39 production and neutrophilic inflammation. Thus, HSA-Trx may be a potential candidate drug for 40 preventing the onset or exacerbation of lung injury caused by air pollutants. 41

Air pollution is a major health hazard that causes an estimated 8.8 million deaths per year 47 aerosol-induced lung injury in mice. We also analyzed the mechanism by which HSA-Trx 94 prevents urban aerosol-dependent lung injury, focusing on oxidative stress and neutrophilic 95 inflammation. 96

Diff-Quik staining solution was purchased from Sysmex (Kobe, Japan). Antibodies against 99 citrullinated histone H3 (citrulline R2 + R8 + R17) and neutrophil elastase were purchased from 100 Abcam (Cambridge, UK). Luminal-based chemiluminescent probe (L-012) and isoflurane were 101 obtained from Fujifilm Wako Pure Chemical Corporation (Tokyo, Japan). The RNeasy ® kit was 102 obtained from Qiagen (Hilden, Germany), PrimeScript ® 1 st strand cDNA Synthesis kit was 103 obtained from Takara Bio (Ohtsu, Japan), and THUNDERBIRD ® SYBR qPCR Mix was obtained 104 from Toyobo (Ohtsu, Japan). Novo-Heparin (5000 units), suitable for injection, was purchased 105 from Mochida Pharmaceutical (Tokyo, Japan). RAW264 cells were purchased from Riken 106 BioResource Center (Tsukuba, Japan). ICR mice (6-7 weeks old, male or female) were 107 purchased from Charles River (Yokohama, Japan). The experiments and procedures described 108 here were carried out in accordance with the Guide for the Care and Use of Laboratory Animals 109 Studies (NIES;, Tsukuba, Japan). These urban aerosol particles were collected in air filters from 118 the central ventilation systems of buildings in central Beijing. Analysis of the urban aerosol 119 particles composition was reported in a previous paper (Mori et al., 2008) and NIES website 120 (https://www.nies.go.jp/labo/crm-e/aerosol.html). For the animal studies, urban aerosol particles 121 were suspended in 0.5% methylcellulose solution and administered to mice. For the cell 122 experiments, urban aerosol particles were suspended in ultrapure water, and the suspension was 123 directly added to the medium. 124 125 Production of HSA-Trx fusion protein 126 HSA, Trx, and the HSA-Trx fusion protein were produced following previously reported 127 methods (Kodama et al., 2013; Tanaka et al., 2013) . Transformed Pichia pastoris cells were 128 incubated in 1.25 L of BMGY liquid media (growth phase) for 2 days (OD 600 = 2) at 30 ºC and 129 then cultured in 800 ml of BMMY media (protein induction phase) for 3 days at 30 ºC. The 130 fusion protein was purified by chromatography on a Blue Sepharose 6 Fast Flow column and 131

HiTrap Phenyl HP column for hydrophobic chromatography. The fusion protein was analyzed by 132 SDS-PAGE using a 10% polyacrylamide gel, with Coomassie blue R250 staining. The purity of 133 the fusion protein was estimated to be in excess of 95%. 134 135

Mice were intravenously administered HSA-Trx (3.5 nmol protein/mouse) in sterile saline 137 immediately before and 24 h after the intratracheal administration of urban aerosol particle 138 suspensions. After anesthetization with isoflurane, mice were administered urban aerosol 139 particles (200 µg/mouse) in 0.5% methylcellulose solution intratracheally using a P200 140 J o u r n a l P r e -p r o o f micropipette. During administration, the nostrils of the mice were blocked so that the solutions 141 were inhaled from the mouth into the respiratory tract as the mice breathed. The control group 142 received sterile saline intravenously and 0.5% methylcellulose solution intratracheally, the urban 143 aerosol treatment group received sterile saline intravenously and urban aerosol particle 144 suspension in 0.5% methylcellulose solution intratracheally, and the HSA-Trx group received 145 intravenous HSA-Trx and intratracheal urban aerosol particle suspensions. Additionally, we 146 examined the effect of HSA-Trx alone (intravenously administered HSA-Trx and intratracheally 147 administered 0.5% methylcellulose solution with timing as above), and found that it did not affect 148 protein concentration or number of leukocytes in the BALF, nor did it affect the levels of reactive 149 oxygen species (ROS) in the lung ( Supplementary Fig. S2 ). 150 151

BALF was collected by cannulating the trachea and lavaging the lung twice with 1 ml of sterile 153 saline containing 50 units/ml heparin. Approximately 1.8 ml of BALF was routinely recovered 154 from each mouse, and the total cell number in the BALF was counted using a hemocytometer. 155

After centrifugation with a Cytospin ® 4 (Thermo Fisher Scientific, Waltham, MA, USA), the cells 156 were stained with Diff-Quik reagents, and the ratio of neutrophils to the total cell number was 157 determined. The amount of protein or double-stranded DNA (dsDNA) present in the BALF was 158 evaluated by the Bradford method or by using a Quant-iT™ PicoGreen ® dsDNA Assay Kit 159 (Thermo Fisher Scientific). 160 membranes using the iBlot ® 7-Minute Blotting System (Thermo Fisher Scientific). Membranes 165 were blocked with 5% non-fat dry milk at room temperature for 1 h, then incubated overnight 166 with rabbit anti-citrullinated histone H3 antibodies (1:1,000 dilution) or rabbit anti-neutrophil 167 elastase antibodies (1:1,000 dilution) in 5% bovine serum albumin (BSA), 1× Tris-buffered saline 168 (TBS), and 0.1% Tween 20, followed by incubation for 1 h with HRP-linked goat anti-rabbit IgG 169 antibodies (1:2,000 dilution) in 1× TBS containing 0.1% Tween 20. Finally, the protein bands 170 were visualized using SuperSignal™ West Dura Extended Duration Substrate (Thermo Fisher 171 Scientific). Band intensities were quantitated using ImageJ software (version 1.39u). 172 173

In vivo imaging of ROS in mice was performed as previously described (Tanaka et al., 2017) . We 175 used the FUSION chemiluminescence imaging system (Vilber Lourmat, Collégien, France). Mice 176 were intraperitoneally administered the ROS-sensing chemiluminescent probe, L-012, in sterile 177 saline (75 mg/kg) 24.5 h after the administration of urban aerosol particle suspensions. At 10 min 178 after the L-012 administration, the mice were euthanized, and their lungs were promptly dissected 179 and imaged (5-min exposure). All data were analyzed using the FUSION chemiluminescence 180 imaging system software. 181 182

Total RNA was extracted from lung tissue using an RNeasy kit in accordance with the 184 manufacturer's protocol. Samples were reverse-transcribed using the PrimeScript ® kit described 185 above. The synthesized cDNA was used in real-time PCR experiments with THUNDERBIRD ® 186 SYBR qPCR Mix and analyzed with a Bio-Rad (Hercules, CA, USA) CFX96™ real-time system 187 and CFX Manager™ software. Specificity was confirmed by electrophoretic analysis of reaction products and the inclusion of template-or reverse transcriptase-free controls. To normalize the 189 amount of total RNA present in each reaction, glyceraldehyde-3-phosphate dehydrogenase 190 (Gapdh) cDNA was used as an internal standard. Primers were designed using the 191 Primer-BLAST website (https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Primer sequences 192 will be provided upon request. 

We first examined the effect of HSA-Trx on urban aerosol-induced lung injury by measuring 212 the number of leukocytes in BALF 48 h after an intratracheal administration of urban aerosols. 213

The dose of HSA-Trx used was 3.5 nmol protein per mouse, and this optimal concentration 214 was established in previous studies on lung disease (Tanaka et al., 2014; Tanaka et al., 2013) . 215

As shown in Figure 1A and 1B, the total number of leukocytes, and especially the number of 216 neutrophils, in the BALF was increased by urban aerosol treatment. The number of 217 macrophages was not increased by this treatment. In contrast, mice pre-treated with HSA-Trx 218 had significantly lower total numbers of leukocytes, particularly neutrophils, following urban 219 aerosol exposure. The number of neutrophils after treatment with a control, urban aerosols, or 220 urban aerosols plus HSA-Trx was 0.05 ± 0.02, 0.96 ± 0.12, or 0.58 ± 0.12 (x 10 4 cells), 221 respectively. The protein concentration in BALF is an indicator of lung injury and edema. As 222 shown in Figure 1C , treatments with urban aerosols increased the protein concentration in 223 BALF, whereas, pre-treatment with HSA-Trx significantly suppressed the urban 224 aerosol-dependent increase in protein concentration. These results suggest that HSA-Trx 225 suppresses urban aerosol-induced lung injury. 226 aerosol treatment trended higher compared with mock-treated controls, but these differences 236

were not statistically significant. In contrast, pre-treatment with HSA-Trx significantly 237 inhibited the urban aerosol-dependent increases in Tnf-α, Il-6, and Kc expression. While, a 238 trend toward suppression of the urban aerosol-dependent increase in Mip2 and Mcp1 239 expression was observed, these differences were not statistically significant. Together, the data 240 suggest that HSA-Trx inhibits urban aerosol-induced lung injury through the inhibition of 241 inflammatory responses. in Figure 3A , intratracheal the administration of urban aerosols increased ROS levels in the 250 lung. The area that is stained red is the place where high levels of reactive oxygen were 251 generated. In contrast, mice pre-treated with HSA-Trx did not display such red staining after 252 urban aerosol treatment, indicating that HSA-Trx pre-treatment markedly suppressed ROS 253 production. A quantitative analysis using the standard software provided with FUSION 254 showed that HSA-Trx significantly inhibits ROS production ( Figure 3B ). We next used an in 255 vitro system to investigate whether HSA-Trx directly inhibits urban aerosol-dependent ROS 256 production. As shown in Figure 3C , the treatment of RAW264 cells with urban aerosols 257 induced ROS production in a dose-dependent manner. In contrast, urban aerosol-dependent 258 ROS production was significantly inhibited in RAW264 cells that were pre-treated in vitro 259 with HSA-Trx ( Figure 3D ). These results suggest that HSA-Trx inhibits urban 260 aerosol-induced lung injury and inflammatory responses through reducing ROS production.

264 ROS has been reported to play a major role in the formation of neutrophilic extracellular traps 265 (Mullin et al.) (Mullin et al.; Stoiber et al., 2015) . NETs are composed of decondensed 266 chromatin fibers and cytoplasmic protein, such as neutrophil elastase and peptidylarginine 267 deiminase 4 (PAD4). Citrullination of histones H3 and H4 by PAD4 is essential for the 268 disaggregation of chromatin, which is required for the formation of NETs (Wang et al., 2009) . 269

Thus, we monitored the dsDNA levels in BALF as an indicator of NET formation. As shown 270

in Figure 4A , the levels of dsDNA in BALF markedly increased following the intratracheal 271 administration of urban aerosols. In contrast, HSA-Trx pre-treatment significantly reduced 272 this increase. We then analyzed the expression levels of two other NET indicators, 273 citrullinated histone H3 (Cit-H3) and neutrophil elastase, in BALF. As shown in Figure 4B Figures 1 and 2 , we found that HSA-Trx clearly suppresses urban aerosol-induced 285 lung injury and inflammatory responses. Moreover, focusing on oxidative stress and 286 neutrophilic inflammation, we analyzed the mechanism by which HSA-Trx prevents urban 287 aerosol-dependent lung injury. As shown in Figures 3 and 4 , we found that HSA-Trx 288 suppresses urban aerosol-induced ROS production in vivo and in vitro, and urban 289 aerosol-induced NET formation in vivo. Based on these results, we suggest that HSA-Trx 290 prevents urban aerosol-induced acute lung injury by suppressing ROS production and 291 neutrophilic inflammation. Although it has been known from previous studies that air 292 pollutants such as PM 2.5 produce ROS in vivo and in vitro (Li et al., 2018b; Li et al., 2015; 293 Zhang et al., 2018) , this is the first study to detect the production of ROS by urban aerosol 294 exposure using an in vivo imaging system. In the future, we would like to detect the 295 production of ROS following long-term exposure to urban aerosols and analyze the effects of 296 candidate drugs in animal models. 297

Oxidative stress has been shown to play a major role in the development of various lung 298 injuries via activation of neutrophil inflammation, vascular permeability, and the coagulation 299 system (Sarma and Ward, 2011; Tasaka et al., 2008) . For example, in ARDS patients, ROS 300 levels have been reported to increase in plasma or BALF (Lamb et al., 1999; Quinlan et al., 301 1997; Tasaka et al., 2008) . Recently, Delgado-Roche et al. speculated that oxidative stress 302 may also play an important role in severe acute respiratory syndrome coronavirus 303 (SARS-CoV) infection (Delgado-Roche and Mesta, 2020). Moreover, ROS has also been 304 suggested to be involved in chronic obstructive pulmonary disease (COPD), a respiratory 305 disease with a high number of patients caused by smoking and air pollution. Specifically, the 306 levels of oxidative radicals, 8-OHdG, or lipid peroxide have been reported to be higher in 307 lung tissues and BALF from COPD patients or smokers compared with healthy nonsmokers previously found that HSA-Trx exerts its effects against influenza virus-dependent acute lung 310 injury and bleomycin-dependent pulmonary fibrosis via antioxidant action (Tanaka et al., 2014; 311 Tanaka et al., 2013) . Considering these results, we believe that HSA-Trx may be a potential 312 candidate for drugs to prevent the onset or exacerbation of lung injury caused by air pollutants. 313

NETs play an important role as a defense system against bacterial and viral infections. 314

However, the overactivation of NETs is thought to lead to systemic exacerbation of 315 inflammation, including inflammatory lung disease (Porto and Stein, 2016; Yang et al., 2016) . 316

In fact, a clinical trial of asthma, which is a lung disease characterized by airflow limitation, 317 detected neutrophil elastase, an indicator of NETs, in 67% of sputum samples from asthmatic 318 patients (compared with 0% in the control group) (Simpson et al., 2007) . An analysis using 319 sputum from COPD patients revealed that the presence of large amounts of NETs is 320 associated with disease severity. Specifically, such high NET amounts were observed in more 321 than 90% of subjects with exacerbated COPD, in more than 45% of subjects with stable 322 COPD, and in more than 25% of smoking control subjects, but in less than 5% of nonsmoking 323 subjects (Grabcanovic-Musija et al., 2015) . Furthermore, it has also been shown that the 324 NET-DNA concentration correlates with disease severity in patients with gastric 325 aspiration-induced ARDS (Li et al., 2018a) . In contrast, to the best of our knowledge, no 326 studies have previously shown that urban aerosols induce NETs in the lungs of humans or 327 mice. Therefore, it is notable that we have demonstrated here for the first time that urban 328 aerosols induce NETs in vivo and that HSA-Trx inhibited the NET formation in this study. 329

Considering the future potential clinical application of HSA-Trx, it is important to analyze 330 the effect of HSA-Trx on urban aerosol-induced exacerbations of various respiratory diseases, 331 not only on lung injury caused directly by administration of urban aerosols. For example, a 332 meta-analysis published in 2016 shows that short-term exposure to air pollutants (e.g., O 3 , CO, 333 expiratory volume in 1 second (FEV1), and forced vital capacity (FVC)) and were associated 337 with methacholine responsiveness (PC 20 ) in children with asthma (Ierodiakonou et al., 2016) . 338

Moreover, in an animal study, PM 2.5 exposure was found to aggravate cigarette 339 smoke-induced pulmonary emphysema changes (airspace enlargement) and inflammatory 340 responses in mice (Zhao et al., 2019) . The findings from these reports indicate that it is 341 important to examine the effect of HSA-Trx on the exacerbation of various respiratory 342 diseases by urban aerosols. 343 344

We revealed here that the intravenous administration of HSA-Trx prevents urban 346 aerosol-dependent lung injury and inflammatory responses. Moreover, HSA-Trx treatment 347 significantly suppressed urban aerosol-dependent ROS production and neutrophilic 348 inflammation, specifically NET induction. Therefore, we believe that HSA-Trx may be a 349 potential candidate drug to prevent the onset or exacerbation of lung injury caused by air Mori, I., Sun, Z., Ukachi, M., Nagano, K., McLeod, C.W., Cox, A.G., Nishikawa, M., 2008. 441 Development and certification of the new NIES CRM 28: urban aerosols for the 442 determination of multielements. Anal Bioanal Chem 391, 1997 -2003 Pinamonti, S., Leis, M., Barbieri, A., Leoni, D., Muzzoli, M., Sostero, S., Chicca, M.C., 454 Carrieri, A., Ravenna, F., Fabbri, L.M., Ciaccia, A., 1998 Male ICR mice were intravenously administered with HSA-Trx (3.5 nmol protein/mouse in 533 sterile saline) or sterile saline alone, immediately before and 24 h after intratracheal 534 administration of urban aerosol particle suspensions administration (200 µg/mouse) or 0.5% 535 methylcellulose solution alone (Cont). Total RNA was extracted from the lungs 48 h after 536 urban aerosol particle suspensions administration and subjected to real-time RT-PCR using a 537 specific primer set for each gene. Values were normalized to Gapdh and are expressed relative 538 to the Control. Values are the mean ± S.E.M.; # p < 0.05; ** or ## p < 0.01 (*, vs Control; #, vs 539 UA). 

Histone hypercitrullination mediates 507 chromatin decondensation and neutrophil extracellular trap formation

New 510 Insights into Neutrophil Extracellular Traps: Mechanisms of Formation and Role in 511 Inflammation

particles-induced lung injury in an Nrf2-dependent manner

517 lungs were imaged using a FUSION chemiluminescence imaging system. (B) The summed 548 pixel intensity of the ROS signal was determined using standard software for FUSION

The cells were then treated with 550 urban aerosols (3.8-60 ng/cm 2 ) (C) or HSA-Trx (0.3-2.5 µM) prior to the addition of urban 551 aerosols (30 ng/cm 2 ) (D) to the medium

microplate reader. Values represent the mean ± S.E.M. # p <

# vs UA (30 ng/cm 2 ) )