Methoxypoly (ethylene glycol) (mPEG, Mw = 2kD and 5kD) was purchased from Sigma-Aldrich (St. Louis, MO, United States). Artesunate (ART) was obtained from Macklin (Shanghai, China). Dihydroartemisinin is a product of Herbest (Baoji, Shaanxi, China). Bis(2-hydroxyethyl) disulfide was provided by Alfa Aesar (Shanghai, China). Succinic anhydride (SA), 4-dimethylaminopyridine (DMAP) and pyrene were purchased from Aladdin (Shanghai, China). N,N′-dicyclohexylcarbodiimide was obtained from Adamas (Shanghai, China). Lipopolysaccharide (LPS) and esterase was purchased from Sigma-Aldrich. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), fetal bovine serum (FBS), dulbecco’s modified eagle medium (DMEM) and trypsin-EDTA were obtained from Gibco (Langley, OK, United States). Tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6) and interleukin-1β (IL-1β) were obtained from R&D systems (Minneapolis, MN, United States). Myeloperoxidase (MPO) was purchased from Lianke-Bio (Hangzhou, Zhejiang, China). All other chemicals were of analytical or HPLC grade and were used as received.

mPEG_2k-ART and mPEG_5k-ART were synthesized by coupling of the carboxyl group of ART and the hydroxyl group of mPEG_2k or mPEG_5k, respectively. Briefly, 1 mM of mPEG_2k or mPEG_5k was reacted with an excess of ART by activation with an excess of dicyclohexylcarbodiimide (DCC) and 4-dimethylaminopyridine (DMAP) in 10 ml dichloromethane (DCM) and allowed to react at RT for 24 h. The reaction solution was filtrated through 0.22 μm filter to remove the precipitates. The solvent in the resultant solution was then removed by rotary evaporation at 50°C. The gel like product was dissolved in acetone and subjected to a dialysis bag (MWCO 1000) for purification against DI water for 48 h, and was lyophilized.

mPEG_2k-SS-ART and mPEG_5k-SS-ART were prepared through the following procedure. Firstly, 1 mM of mPEG_2k and excess of SA were activated by DCC/DMAP following the similar protocol as aforementioned to obtain carboxyl group terminated mPEG_2k-COOH. Then the product was purified using dialysis bag (MWCO 1000) in DI water for 48 h and was lyophilized. Secondly, 1 mM ART and excess of bis(2-hydroxyethyl) disulfide (HEDS) were activated by excess of DCC/DMAP in 10 ml DCM at RT for 24 h. The solution was filtrated and the filtrate was reacted with activated mPEG_2k-COOH under the excess amount of DCC/DMAP at RT for24 h mPEG_2k-SS-ART was obtained by purification as depicted as aforementioned. The synthesis of mPEG_5k-SS-ART was followed with a similar protocol as that of mPEG_2k-SS-ART.

¹H NMR spectra of polymers were obtained by a BrukerAscend^TM 600 MHz NMR spectrometer (Billerica, MA, United States) in CDCl₃ at 25°C.

The drug loading content (LC) was determined by Agilent 1,260 high performance liquid chromatography (HPLC, Santa Clara, CA, United States) equipped with an Agilent C₁₈ column (4.6 × 250 mm, 5 μm, column temperature 30°C), with a mobile phase of acetonitrile/phosphoric acid solution (pH 3.0) (52:48, v/v) and a flow rate of 1.0 ml/min. A weighed number of lyophilized micelles was dissolved in methanol and subjected to HPLC detection via a UV detector at 210 nm, as compared to a calibration curve of ART in the mobile phase. The LC was calculated according to the following equation: L C = w e i g h t   o f   d r u g   i n   t h e   p o l y m e r w e i g h t   o f   p o l y m e r   p r o d r u g × 100 %

The as-prepared polymer prodrugs could be self-assembled into polymer micelles in aqueous milieu. The polymer micelles were prepared by the thin-film hydration according to previous protocol (Chen et al., 2017).

The critical micelle concentration (CMC) of the polymer prodrugs was determined by fluorescence spectroscopy using pyrene as a probe (Chen et al., 2017). The emission wavelength was set at 390 nm with excitation slit and emission slit of 1.0 and 10 nm, respectively. The excitation intensities at wavelength range from 300 to 360 nm were measured using a HITACHI fluorescence spectrophotometer (F-2k, Tokyo, Japan). The concentration of the polymers solution was varied from 1.0 × 10⁻⁵ to 1.0 mg/ml containing 6 × 10⁻⁸ M of pyrene. The fluorescence intensity ratio of I₃₃₆/I₃₃₄ was calculated against polymer concentration to determine CMC. The size and size distribution of the micelles were measured with a Malvern Zetasizer (Nano ZS90, Worcestershire, United Kingdom) with a detection angle at 90°. The stability of the as-prepared polymer micelles was also monitored on the Zetasizer under physiological conditions at 25 and 37°C, respectively. The morphology and structure of the polymer micelles were observed using a HITACHI H7650 (Tokyo, Japan) transmission electron microscope operated at an acceleration voltage of 80 kV. The samples were prepared by dipping 0.1 wt% sample solution onto TEM grids and air dried.

The release of ART from the prodrugs was characterized according to our previous study (Chen et al., 2017). In short, a fixed amount of mPEG_2k-ART or mPEG_5k-ART micelle (with an ART dosage of 2 mg) was reconstituted in 2 ml PBS with or without the addition of esterase, and was sealed in a dialysis tubing (MWCO 1000) and incubated in 20 ml PBS at pH 7.4. The drug release was carried out at 37°C in an incubator shaking horizontally at 100 rpm. The release medium was completely removed and supplemented with fresh PBS at predetermined time intervals. The ART concentration in the medium was recorded with HPLC with a UV detector operated at 210 nm. The cumulative released drug was integrated from each measurement.

The release of ART from mPEG_2k-SS-ART or mPEG_5k-SS-ART prodrugs was same to that of mPEG_2k-ART except that the esterase was replaced by 10 mM glutathione (GSH).

Mouse RAW264.7 cells (the Cell Resource Center of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, Beijing, China) were used and maintained in DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin at 37°C, 5% CO₂ and 95% humidity.

RAW264.7 cells were cultured in DMEM supplemented with 10% fetal bovine serum and 1% antibiotics (100 U/mL penicilin and 100 μg/ml streptomycin, Life Technologies) at 37°C with 5% CO₂. Cells were seeded in 24-well plates with a density of 2 × 10⁵/well and were treated with PEGylated prodrugs at various ART concentrations (0, 1, 5, 10, 20, 50 μg/ml). 1 h after treatment the cells were challenged with LPS (200 ng/ml) for 6 h. Cell culture supernatants were collected and centrifuged at 10,000 rpm for 10 min at 4°C, then subjected to ELISA of cells secreted TNF-α, IL-1β, IL-6 and total GSH, according to the manufacturer’s instructions (Multisciences Biotech, Hangzhou, China).

Kunming mice (male, 4–5 weeks, Beijing Vital River Laboratory Animal Technology Co., Ltd., Beijing, China) were bred in specific pathogen free (SPF) environment. Food and water were available ad libitum. Animal studies were conducted according to the Regulation on Experimental Animals of China Academy of Chinese Medical Sciences.

LPS-induced lung inflammation was investigated on Kunming mice according to protocols reported previously (Ferretti et al., 2003; Bohr et al., 2017). 35 mice were randomly divided into five groups, seven mice per group. Mice were dosed intranasally with 10 mg/kg of ART, mPEG_2k-ART, mPEG_5k-ART, mPEG_2k-SS-ART, mPEG_2k-SS-ART (1 µL drug solution/g of mice weight), or saline alone, respectively. 1 h later, the mice were nasal administered with LPS solution (1.25 mg/kg) using a pipet by pacing the mice on their backs and allowing them for inhalation of the solution through the nostrils (1 µL LPS solution/g of mice weight). 0.9% NaCl solution was administered in the saline group. The mice were euthanized 6 h later via an over dose of pentobarbital (i.p. injection of 200 μL of pentobarbital solution, 20 mg/ml) for bronchoalveolar lavage (BAL), and the trachea was exposed from an incision and cannulated with a 22-gauge catheter (BD biosciences). Subsequently, BAL was performed by flushing the lungs with 1.0 ml PBS for 2 times and was used for protein quantification and cell counting/differentiation. The BAL was centrifuged at 2 k rpm for 10 min at 4°C, the supernatant was stored at −80°C for protein quantification, and the pellets were pooled and resuspended in fresh PBS.

The water content of the lungs was measured following the administration of LPS. The right lungs were excised, blotted, and weighed to obtain the wet weight, then desiccated at 80°C for 24 h to obtain the dry weight. The wet/dry weight ratio was calculated as an assessment of tissue edema.

The lung tissues were collected and fixed in 4% paraformaldehyde fixing solution. The tissues were dehydrated, embedded in paraffin, sliced into 5 μm sections. The sections were stained with hematoxylin and eosin (H&E) reagent and visualized under a light microscope.

Lung tissues were harvested, rinsed, homogenized, and centrifuged, then the supernatants were collected and subjected to ELISA for determination of myeloperoxidase (MPO) activity using the MPO activity colorimetric assay kit (Biovision, Zurich, Switzerland).

The number of neutrophils in each lavage was counted by Wright’s staining using hemocytometer slides. Bronchoalveolar lavage fluid (BALF) samples were collected by flashing the lung three times with 4 ml PBS through a tracheal cannula placed into each mouse under anesthesia. Briefly, animals were sacrificed and the chest was opened, a median sternotomy was performed, and the trachea was isolated using a blunt dissection. Next, a suitable small-caliber tube was inserted into the airway and secured. Then PBS solution was infused slowly into the lungs, and the BALF was withdrawn into the tube. The fluid recovery rate was > 80%. The lavage samples were centrifuged at 1,500 g for 10 min at 4°C.

The protein levels of IL-6, IL-1β, and TNF-α in the BAL were quantified using ELISA. BAL supernatants were stored at −20°C and were used directly or diluted (1:3, v/v) in assay diluent from the kit, and the samples were prepared following the instructions illustrated in the protocol of the kit. All samples were measured in triplicate.

The concentration of TNF-α, IL-1β, and IL-6 in BALF was measured using a commercial ELISA kit. The results were expressed as pg/mL of BALF.

The lung tissue of mice was quickly frozen with liquid nitrogen and ground into powder.The total GSH content of BALF and lung tissue from mice in different treatment groups were determined as the same as the aforementioned.

Data were processed using GraphPad Prism 5.01 software (La Jolla, CA, United States) and presented as mean ± standard deviation (SD). Statistical analysis was performed using one-way analysis of variance (ANOVA). Tukey’s post-hoc comparisons were used for statistical comparisons. The difference was regarded statistically significant as ^# p < 0.05 and ^* p < 0.05, very significant as ^## p < 0.01, and extremely significant as ^### p < 0.001.

The poor solubility severely limits the bioavailability of ART both in vitro and in vivo. At present, special agents or solvents such as NaHCO₃ have been used to prepare well-dissolved ART stock solutions on account of their poor solubility but shows low bioavailability. The significant effect of PEG-modification may pave new roads in tackling these problems (Turecek et al., 2016). The use of reduction-sensitive disulfide bond for GSH mediated intracellular release has already been studied (Giri et al., 2005) and proved (Wang et al., 2011) to be very effective in reducing off-targeting effect. In this scenario, mPEG_2k-ART and mPEG_5k-ART were synthesized by coupling of the carboxyl group of ART and the hydroxyl group of mPEG_2k and mPEG_5k, using dicyclohexylcarbodiimide and 4-dimethylaminopyridine (DCC/DMAP) as catalysts to facilitate the synthesis of ART prodrugs and achieve desirable drug grafting ratio, respectively (Supplementary Figure S1A). mPEG_2k-SS-ART and mPEG_5k-SS-ART were synthesized by conjugation of ART with mPEG_2k or mPEG_5k by coupling with a disulfide bond containing molecule using DCC/DMAP linkage chemistry via the formation of ester bond, respectively (Supplementary Figure S1B). The molar ratio of ART to total matrices was set as a minimum of 3.5:1 (mol/mol) to ensure the successful ligation of one ART molecule to one mPEG chain (Supplementary Table S1). The chemical structures of the prodrugs were characterized by ¹H NMR spectroscopy, the appearance of peaks at 5.4 ppm (a, 1H), 5.8 ppm (b, 1H) and 2.8 ppm (g, S-CH₂-CH₂-), 4.4 ppm (f, S-CH₂-CH₂-) were attributed to ART and HEDS, respectively. Particularly, a prominent single peak at 3.4 ppm (e, -OCH₃) appeared in the spectrum of prodrugs, demonstrating the conjugation of ART and HEDS with mPEG, respectively. The characterization with ¹ H NMR spectroscopy revealed the ART prodrugs were successfully synthesized (Supplementary Figure S2). The hydrophobic core composed of ART was stabilized by the cloaking mPEG chains, conferring “stealth” properties to the nanosystems in the blood against clearance by the mononuclear phagocyte system for prolonged circulation (Suk et al., 2016). Noted that the inflamed tissues are typically associated with increased reductive and oxidative stress (García et al., 2017). Therefore, the release of ART from mPEG_2k-SS-ART or mPEG_5k-SS-ART prodrugs could be mediated by inflammation-associated reductive microenvironment. It is anticipated to realize both prolonged systemic circulation and on-demand drug release, therefore can improve the anti-inflammatory purposes.

The self-assembly behavior of the ART prodrugs in aqueous milieu was evaluated by pyrene fluorescence spectrometry to analyze the critical micelle concentration (CMC) of the prodrugs, due to its high sensitivity to the surrounding polar microenvironment. The intensity ratio (I336/I334) was monitored as a function of ART prodrugs concentration. The intensity ratio (I336/I334) exhibited significant increase with the increase of the prodrug concentration, corresponding to the formation of aggregates together with the accumulation of pyrene into hydrophobic domains. It should be noted that the as-prepared prodrugs exhibited very low CMC values. The CMC values of mPEG_2k-SS-ART and mPEG_5k-SS-ART were 0.072 and 0.256 mg mL⁻¹, respectively (Supplementary Figure S3). However, the CMC values of mPEG_2k-ART and mPEG_5k-ART were not obtained from pyrene fluorescence spectrometry. The results suggested the self-assembly ability of the prodrugs in aqueous medium at low copolymer concentrations. The hydrodynamic diameter of the particles was monitored using dynamic light scattering (DLS). The diameters of ART prodrugs with ester linkages were fairly small with 3.72 and 5.18 nm for mPEG_2k-ART and mPEG_5k-ART, respectively (Supplementary Table S2). However, the diameters of mPEG_2k-SS-ART and mPEG_5k-SS-ART increased significantly after disulfide bond conjugation, with diameters of 41.04 and 106.35 nm, respectively. The influence of the physiological environment on the size of the prodrug micelles was investigated in PBS (pH 7.4), it can be observed that the sizes of mPEG_2k-SS-ART and mPEG_5k-SS-ART micelles decreased significantly in PBS (Supplementary Figure S4), indicating the electrostatic screening effect on the backbone of the prodrugs, leading to more coiled molecular structure in PBS solution. As a comparison, the diameter of mPEG_2k-ART and mPEG_5k-ART prodrugs changed slightly after replacement of the medium from DI water to PBS. Transmission electron microscopy (TEM) revealed the spherical structure of the as-prepared micelles, which was in consistence with the results observed using DLS (Supplementary Figure S5).

The stability of the ART prodrugs was investigated by measurement of the micelles’ sizes during incubation at 25 and 37°C according to our previous protocol (Hao et al., 2020). As shown in (Supplementary Figure S6), the mPEG_2k-ART and mPEG_5k-ART micelles showed high stability in hydrodynamic diameter within 48 h. However, significant increase in the size of prodrugs was observed after elongation of incubation to 72 h at 25°C, demonstrating the formation of large aggregates. The mPEG_2k-SS-ART and mPEG_5k-SS-ART micelles were stable at least within 12 h at both 25 and 37°C, respectively. However, the sizes of the two prodrugs increased significantly after 12 h incubation. The differences in the stability of the prodrugs may be ascribed to the different speed of degradation of ester linkage and disulfide bond (Qiao et al., 2011).

The ester bond can be degraded by esterase, which is abundant in inflammatory tissues (Xu et al., 2008). The nanoparticle carriers stably constrain their drug payloads during systemic circulation but perform esterase-triggered rapid release of drugs in inflammatory tissues (Wang et al., 2021). Dihydroartemisinin (DHA) is an active metabolite of ART. The ART loading content (LC) of the as-prepared prodrugs could be determined by measurement of the DHA content in the presence of porcine liver esterase. As shown in Supplementary Table S1, the LC of ART prodrugs were ranged from 1.4 to 5.7% as determined by HPLC, respectively, which was not in consistence with the result calculated from the ¹H NMR data, hinting the incomplete dissociation of ART from the respective prodrugs in the current incubation milieu. Therefore, longer incubation time or more esterase would be needed to completely degrade the prodrugs for ART release. However, the solubility of ART was greatly enhanced by the current mPEG modification strategy.

The sustained drug release profile from therapeutic nanoparticles after intravenous administration is particularly important for improving the pharmacokinetics and biodistribution of therapeutic agent (Attia et al., 2013). The in vitro release study is important to evaluate whether or not that the prodrugs could release their payload after taken up into cells (Qiao et al., 2021). Considering the formation of drug conjugates via an ester bond or disulfide bond, in this study, the release behavior of ART from the prodrugs was investigated via dialysis method in PBS at 37°C (pH 7.4, with or without esterase for mPEG-ART, with or without GSH for mPEG-SS-ART) to mimic the in vivo physiological conditions of blood and cells, respectively. As shown in Supplementary Figure S7, a burst release was observed from mPEG_2k-ART in PBS with esterase in 8 h and approximately 75% of the drug released. As a comparison, only 27.1% of drug was released from mPEG_2k-ART in PBS without esterase even after 24 h of incubation, indicating the accelerated drug release profile in the assistance of esterase. However, similar release profiles were observed for mPEG_5k-ART in PBS with or without esterase, with cumulative drug released amount of 55.6 and 52.1%, respectively. This can be explained that the more coiled structure of mPEG_5k-ART may impede the access of esterase to the ester bond between mPEG and ART. Subsequently, the drug release profiles of the redox-responsive prodrugs of mPEG_2k-SS-ART and mPEG_5k-SS-ART were also investigated. A burst release profile was observed for mPEG_2k-SS-ART in PBS with the existence of GSH after incubation for 8 h with 50.7% of drug was released, due to the disintegration of disulfide bond under GSH. In comparison, only 26.8% of drug was released from mPEG_2k-SS-ART in PBS without GSH even after 24 h of incubation. As a contrast, similar release profiles of mPEG_5k-SS-ART were observed in PBS with or without the addition of GSH, with cumulative drug released amount of 47.2 and 34.3%, respectively, which should be ascribed to the more entangled molecular structure of mPEG_5k.

ART prodrugs downregulate LPS-induced inflammatory mediators in RAW264.7 cells. The release of inflammatory cytokines plays an essential role in inflammatory process, therefore, the mediation effect of ART prodrugs on the released cytokines from macrophages was investigated. RAW264.7 cells were pretreated with various concentrations of ART prodrugs for 1 h, and stimulated with LPS for 6 h, after which the secretion of TNF-α, IL-1β, and IL-6 in the culture medium was analyzed by ELISA. A significant increase of intracellular inflammatory cytokine levels was observed in the macrophages after stimulation with 200 ng/ml of LPS, as compared to that of the non-stimulated cells. The cytokine levels decreased significantly after treatment by free ART, demonstrating the anti-inflammatory effect of the free drug (Figure 1). Significant downregulation of the cytokines such as TNF-α, IL-6 and IL-1β was observed in the cells treated by mPEG_2k-ART, mPEG_5k-ARTand mPEG_2k-SS-ART, respectively. However, mPEG_5k-SS-ART showed negligible anti-inflammatory effect at all concentrations investigated. The TNF-α expression can be significantly downregulated by ART and mPEG_2k-SS-ART at 5 μg mL⁻¹ (p <0.001) (Figure 1A), with 0.86 and 0.82 fold decrease, respectively. The ART, mPEG_2k-ART and mPEG_2k-SS-ART groups showed potent inhibitory effect on IL-6 even at 1 μg mL⁻¹ (p <0.01) (decreased by 0.90, 0.91, and 0.90 fold, respectively). mPEG_5k-ART and mPEG_5k-SS-ART exhibited slight inhibition on IL-6 expression at 20 μg mL⁻¹ and 50 μg mL⁻¹ (p <0.001, p <0.01) (Figure 1B). For IL-1β mediation, mPEG_2k-SS-ART showed significant modulation effect as compared to free ART, whereas both ART and mPEG_2k-SS-ART exhibited potent inhibitory effects at some concentrations (10 μg mL⁻¹, 5 μg mL⁻¹, p <0.001). However, mPEG_5k-SS-ART exhibited inhibitory effects only at higher concentrations (50 μg mL⁻¹, p <0.05) (Figure 1C). The present study showed that the cytokine levels were downregulated more efficiently by the mPEG_2k-SS-ART than that of the free ART and other ART prodrugs, hinting the inflammation can be effectively inhibited by mPEG_2k-SS-ARTprodrug at concentrations as low as 10 μg mL⁻¹.

ART prodrugs protect cells by elevation of intracellular GSH level. Glutathione (GSH) is the most abundant non-protein thiol that defends against oxidative stress but suffers depletion during the inflammatory responses. As shown in Figure 2, the intracellular GSH level was significantly decreased upon LPS stimulation, but that decrease was significantly attenuated in groups co-treated with ART or mPEG_2k-SS-ART. GSH levels in groups co-treated with ART and mPEG_2k-SS-ART were 1.79 and 1.99 fold increased compared with LPS group, respectively (at 20 μg mL⁻¹). However, no significant upregulation effect on GSH level was observed in the groups of mPEG_2k-ART, mPEG_5k-ART and mPEG_5k-SS-ART, respectively, which might be ascribed to the steric hindrance of mPEG chains which hinders the cellular uptake (Męczyńska-Wielgosz et al., 2016). Our results demonstrated that ART and mPEG_2k-SS-ART can effectively elicit compensation to oxidative stress by upregulation of GSH levels.

For ALI treatment, nanodrugs are mainly administrated through intrapulmonary or the intravenous route. For the intrapulmonary route, the increase of lung permeability in the pathological environment of ALI may contribute to nanodrug-mediated passive targeted delivery (Newman, 2017). Nanodrugs are easy to penetrate the mucus layer and pass through the cell membrane (Lin and Dean, 2011). Therefore, the anti-inflammatory effect of the as-prepared ART prodrugs was investigated on LPS-induced ALI model via nasal administration for effective delivery to lung, as compared to free ART. Significant pulmonary edema was observed in mice after stimulation with LPS. The wet to dry (W/D) ratio of the lungs in all treatment groups were measured 6 h after LPS challenge. Compared with the saline group, lung W/D ratio was significantly increased in the LPS-treated group. The W/D ratio of the mice was significantly decreased in mPEG_2k-SS-ARTgroup, as compared to that of the LPS group (Figure 3) (p <0.01). It was shown that stimulation of the lung epithelial cells by LPS could induce the release of chemokines and cytokines (Welbourn and Young, 1992). These cytokines could lead to the accumulation of numerous neutrophils into the lung tissues and lead to lung edema. In addition, the overproduction of these inflammatory cytokines leads to the injury of the lung tissues.

To further evaluate the anti-inflammatory effect of ART prodrugs, immunohistochemical analysis was performed to evaluate the effect of ART formulations on the LPS-induced histopathological changes. As shown in Figure 4, the structure of normal lung tissues was exhibited in the saline group. Significant pathological changes were observed in the lung tissues of LPS group, including neutrophils infiltration, edema, significant thickening of alveolar walls and alveolar disarray. However, the histopathological deterioration was significantly alleviated after treatment by ART, mPEG_2k-SS-ART and mPEG_5k-SS-ART, respectively. And the most potent therapeutic effect was observed in mPEG_2k-SS-ART group.

MPO was abnormally produced in neutrophils under inflammatory conditions (Ndrepepa, 2019), therefore the MPO activity in lung was spectrophotometrically assayed using test kits. The lung neutrophil penetration marker of MPO activity was detected 6 h later after LPS stimulation. As illustrated in Figure 5, the lung MPO activity was significantly increased in the LPS group compared with the saline group. However, the MPO activity was dramatically inhibited by free ART and mPEG_2k-SS-ART, with better therapeutic effect for the mPEG_2k-SS-ART group. However, negligible inhibitory effect on MPO activity was observed in mPEG_5k-SS-ART group, which might due to the impedance of longer mPEG chain on the release of the drug (Sanchez et al., 2017).

The pathogenesis of ALI was also associated with the vasculature inflammation and neutrophil-mediated inflammation (Thompson et al., 2017). The severity and development of ARDS/ALI was manipulated by the migration of neutrophil into the lungs in response to activated macrophages (Williams and Chambers, 2014). In the current study, the total neutrophil counts of BALF were all significantly down-regulated, suggesting the remarkable therapeutic effect of ART preparations to combat ALI. And best therapeutic efficacy was observed in the mPEG_2k-SS-ART group, as compared to that of free ART (Figure 6), which might be ascribed to the accelerated phagocytosis by macrophages (Kim et al., 2018), and enhanced neutrophil uptake for smaller sized nanoparticles with reduced cell viability (Yu et al., 2020).

The inflammatory cytokines played an important role in the development of lung injury (Kim et al., 2019). The inflammation associated cytokines including TNF-α, IL-6 and IL-1β were detected in the BALF of ALI mice by ELISA to investigate the anti-inflammatory effects of ART prodrugs in vivo. It has been shown that the level of proinflammatory cytokines such as TNF-α, IL-6 and IL-1β in the lung was significantly increased after LPS stimulation on mice, which is consistent with the anti-inflammatory properties of ART prodrugs in vitro. It can be seen that the BALF levels of TNF-α, IL-6 were significantly reduced in the “LPS + ART” group as compared to the LPS group. Moreover, the BALF levels of TNF-α, IL-6 and IL-1β all exhibited significant decrease post mPEG_2k-SS-ART treatment, demonstrating the better anti-inflammatory effect of mPEG_2k-SS-ART prodrug in vivo studies (Figure 7).

Furthermore, the levels of GSH in the BALF and the lung tissues were measured in order to explore the possible mode of action of the ART prodrugs. Interestingly, increased GSH levels in the BALF and the lung tissues were observed in free ART and mPEG_2k-SS-ART group after LPS stimulation. In the mPEG_2k-SS-ART group, the GSH levels in the BALF and in the lung tissue increased significantly (2.09 and 2.10 fold, respectively). However, in the free ART group, the GSH level only significantly increased in the BALF (1.55 fold) solution. On another hand, mPEG_5k-SS-ART showed a similar trend of increased GSH level but not statistically significant, with a comparison to the LPS stimulated group. Therefore, it could be indicated that mPEG_2k-SS-ART may be the best ART prodrugs on account of anti-inflammatory applications (Figure 8).