AKEX0011 was obtained from Ark Biopharmaceutical Co. Crystalline silica particles were obtained from Sigma-Aldrich (S5631,1-5um, purity 99%). Particulates were baked at 200°C for 3 h and then suspended with phosphate buffer saline (PBS). Silica suspensions were vibrated by sonication for 30 min before use.

Specific pathogen-free (SPF) C57BL/6J (male, 10–11 weeks old, average weight 24–26 g) mice were obtained from Sipeifu Biotechnology (Beijing, China). Mice were fed under SPF conditions with room temperature 20–24°C, humidity 35–55%, and 12/12 h light–dark cycles. These animals were euthanized using intraperitoneal pentobarbital. All animal experiments were approved by the Animal Care and Use Committee of the China-Japan Friendship Hospital (Approval certificate number: zryhyy21-21-01-07).

The silicosis mice model was generated via intratracheal instillation of silica using a rodent laryngoscope, which has been previously published (Cao et al., 2020). To establish the early therapeutic model, we randomly divided mice into four experimental groups at 10 animals per group: PBS control, silica + vehicle, silica + AKEX0011, and silica + PFD. PFD was used as the positive control. In the PBS control group, mice were intratracheally instilled with PBS and gavaged with the vehicle (0.5% MC) twice a day. In the silica + vehicle group, mice were instilled with silica and gavaged with the vehicle (0.5% MC) twice a day. In the silica + AKEX0011 group, mice were instilled with silica and gavaged with 100 mpk of AKEX0011 twice a day. In the silica + PFD group, mice were instilled with silica and gavaged with 150 mpk of PFD twice a day. The mice in treatment groups were treated with AKEX0011 or PFD at the day of silica instillation and dosed consecutively for 28 days; then, they were euthanized at day 28.

To perform the advanced therapeutic model, the similar grouping method as early therapeutic models was adapted. We randomly divided mice into four experimental groups at 10 animals per group: PBS control, silica + vehicle, silica + AKEX0011, and silica + PFD. In the PBS control group, mice were intratracheally instilled with PBS and gavaged with the vehicle (0.5% MC) twice a day. In the silica + vehicle group, mice were instilled with silica and gavaged with the vehicle (0.5% MC) twice a day. In the silica + AKEX0011 group, mice were instilled with silica and gavaged with 100 mpk of AKEX0011 twice a day. In the silica + PFD group, mice were instilled with silica and gavaged with 150 mpk of PFD twice a day. Either the vehicle or AKEX0011 or PFD was dosed to mice at day 14 after silica exposure, and then, the drug was dosed consecutively for 28 days; then, they were euthanized at day 42.

The preparation of lung sections for histological analysis was performed, as previously described (Cao et al., 2020). Left lungs were fixed, dehydrated, paraffin-embedded, and then cut into 5 μm sections. Then, the sections were stained by hematoxylin–eosin staining (H&E staining), Masson’s trichrome staining, and Sirius red staining, respectively. H&E staining was undertaken for evaluating the degree of inflammation using Szapiel’s method which consists of no inflammation (grade 0), mild (grade 1+), moderate (grade 2+), and severe (grade 3+) inflammation. The Masson’s trichrome stain was conducted to assess pulmonary fibrosis and was quantified by King’s method (Cao et al., 2020), which is a specific generic method for assessing the silicosis fibrosis scale. Specifically, silicotic nodules on the entire section were defined into five grades (0-5), according to their degree of lesion. Then, the fibrosis score was calculated as the grades (0-5) multiplied by their corresponding percentages of fibrotic area over the total area of the tissue section.

Immunohistochemistry (IHC) analysis was used to detect protein expression levels in lung sections. The lung sections were deparaffinized, rehydrated, and then heated in ethylenediaminetetraacetic acid (EDTA) or citrate buffer. These tissues were blocked by endogenous peroxidase using 0.3% H₂O₂ in 5% serum and then incubated with a mixture of anti-collagen I antibody (ab21286, Abcam, United States), anti-fibronectin antibody (ab2413, Abcam, United States), anti-Arginase-1 antibody (Arg1) (ab233548, Abcam, United States), and anti-iNOS antibody (ab178945, Abcam, United States) at 4°C for 12 h and finally added the secondary antibody for 1 h. The positive areas (%) of collagen I, fibronectin, arginase 1, and iNOS staining were measured by Image-Pro Plus 6.0.

Frozen sections were blocked by 5% BSA and permeabilized by 0.2% Triton X-100. Then, they were incubated with the anti-alpha smooth muscle actin (α-SMA) antibody (Abcam, ab124964, United States) overnight at 4°C. The fluorescent secondary antibody was added and incubated for 1 h in dark and then mounted with DAPI. The fluorescence microscope (ZEISS, Germany) was used for observation and analysis.

The flexiVent FX (SCIREQ, Montreal, Quebec, Canada) system was used to assess mouse’s lung functions. On day 28 or 42, after 2% pentobarbital (0.2 ml/100 g) intraperitoneal injection, endotracheal intubations were inserted into tracheas, and a mouse was connected to the ventilator. Parameters of lung functions were obtained, including the inspiratory capacity (IC), respiratory resistance (Rrs), elastic resistance (Ers), and tissue damping (G) (Wu et al., 2020).

Mouse computed tomography scans were performed according to the manufacturer’s instructions with analytic parameters set at 90 kV, 80 mm FOV, 150 mA, and 1.00 mm slice thickness. Cell culture and group.

The mouse macrophage RAW 264.7 cell line was obtained from Procell Biology (CL-0190, Wuhan, China). They were cultured in DMEM (Gibco, 12100046, Carlsbad, United States) with 10% FBS (Gibco, 10091-148, Carlsbad, United States) at 37 C in 5% CO₂.

As cellular experiments, RAW 264.7 cells were divided into the following: PBS Control, PBS + AKEX0011 (200 μg/ml), silica, silica + AKEX0011 (200 μg/ml), and silica + AKEX0011 (100 μg/ml). As for the pre-treatment experiment, cells in the treatment group were pre-incubated with AKEX0011 for 6 h, and then, silica (100 μg/cm²) was added and incubated for 24 h before analysis; cells in the silica group were pre-incubated with PBS for 6 h, and then, silica (100 µg/cm2) was added and incubated for 24 h before analysis. Subsequent mechanism experiments were also conducted in this cell model. As for the post-treatment experiment, cells in the treatment group were incubated with silica (100 μg/cm²)for 6 h, and then, AKEX0011 was added and incubated for 24 h before analysis; cells in the silica group were incubated with silica for 30 h.

The protein expression levels of IL-1β, TNF-α, IL-6, TGF-β, IL-4, and IL-10 in the cell supernatant or BALF and hydroxyproline (HYP) in lung homogenates were measured using ELISA kits. Detailed descriptions are listed in the supplementary materials (Supplementary Table S1).

Total proteins of the lung tissue were extracted through RIPA (R0010, Beyotime, Shanghai, China). Their concentrations were determined via the BCA Kit (23223, Thermo Scientific, Waltham, United States). For the analysis, 20 µg protein was separated by 8% or 10% SDS-PAGE gels and then electroblotted onto PVDF membranes. The PVDF membranes containing proteins were blocked by 1% BSA for 3 h and then probed using specific primary antibodies at 4°C for 12 h; the antibodies’ information is enumerated in supplementary materials (Supplementary Table S2). Then, we washed membranes using TBST and incubated membranes with secondary antibodies for 1 h. A ChemiDoc XRS + imaging system was used for detection.

Total RNAs were extracted from lung tissues and RAW 264.7 via TRIzol (15596026, Invitrogen, Carlsbad, United States). Next, we reverse transcribed total RNA into cDNA via PrimeScript RT (RR036A, Takara, Shiga, Japan). The qPCR was amplified with a SYBR Green Master Mix Kit (RR420Q Takara, Shiga, Japan) with Bio-Rad IQ5. Related primers are listed in the supplementary information (Supplementary Table S3).

The apoptotic level in lung tissues was detected by TUNEL staining (C1088, Beyotime, China). Briefly, frozen sections were incubated with 50 μL mixed TUNEL working solution in dark after serum blocking and permeabilization, and then, sections were mounted with DAPI (ab104139, Abcam, United States). A fluorescence microscope (ZEISS, Germany) was used for observation.

RAW264.7 cells group as mentioned earlier. The apoptosis detection kit (556547, Becton Dickinson Company, United States) was used following manufacturer’s instructions. Shortly, cells at suitable concentrations were incubated with 5 μl FITC Annexin V and 5 μl PI for 15 min in dark. Then, we analyzed apoptosis by FACS within 1 h. Cells that were stained with FITC Annexin V (+) and PI(+) or (-) were considered apoptotic.

Cells were washed using 3 ml 1% BSA-PBS. Then, we suspended them in 1% BSA-PBS and counted their number with a cell counter, followed by pre-incubation with mouse TruStain FcX PLUS (anti-mouse CD16/32)(Biolegend, 156604, CA, USA)for 10 min. The cell suspension was stained with anti-mouse F4/80 BV605 (743281, Becton Dickinson, San Diego, CA, United States), APC anti-mouse CD163 (155,306, Biolegend, CA, United States), and anti-mouse CD86 BV421(105031, Biolegend, CA, United States) at 4°C for 30 min in the dark. After washing and resuspending in PBS, the cells were analyzed by FACS. The gating strategy was listed in (Supplementary Figure S1).

Statistics analyses were performed by GraphPad Prism 6.0. Significance of multiple groups used ANOVA, followed by Tukey’s post hoc test. A p-value less than 0.05 was considered statistically significant. Data were presented as mean ± SEM.

To evaluate the protective effect of AKEX0011 (Figure 1A) in the early therapeutic silicosis model (Figure 1B), multiple functional and pathophysiological indicators were assessed including the lung function, micro-CT, and the levels of inflammation and fibrosis. The results from silica-instilled mice were compared to PFD-treated mice, which served as a positive control. We chose the 100 mpk dose of AKEX0011 for the next experiment. AKEX0011 at this dose had a better protective effect and did not alter the lung pathology and lung function, indicating that the 100MPK dose of AKEX0011 had no toxic effect on the lungs (Supplementary Figures S2, S3).

Pulmonary function tests and chest CT are common means of clinical assessment of silicosis. In our study, AKEX0011 treatment alleviated impairment of the lung function and improved lung ventilation and compliance, as indicated by increased inspiratory capacity (IC), decreased respiratory resistance (Rrs), elastic resistance (Ers), and tissue damping (G) (Figures 1C–F). Experimental silicosis mouse micro-chest CT showed round nodules, often symmetrically distributed in the lungs; some nodules fused into a patchy shadow. These damages reduced the lung volume and transmittance and imaging damages that have been prominently alleviated with AKEX0011 therapy (Figure 1G).

Persistent inflammation is an important characteristic of silicosis. We evaluated effects of AKEX0011 in the inflammatory reaction of silicosis. H&E staining showed that compared to the PBS control group, the lungs of silica + vehicle mice exhibited nodules in different sizes, alveolar collapse, thickened alveolar septa, inflammatory cell infiltration, and interstitial edema. These pathological changes were alleviated after AKEX0011 treatment (Figures 2A,B). Moreover, AKEX0011 treatment reduced the protein expression of pro-inflammatory factors including TNF-α, IL-6, and IL-1β in the bronchoalveolar lavage fluid (BALF), compared with the silica + vehicle group (Figures 2C–E). These findings were consistent with their mRNA expressions in lung tissues (Figures 2F–H). Taken together, our results indicate that the early therapy of AKEX0011 attenuated an inflammatory response in the silicosis mouse model.

We next evaluated the extent of pulmonary fibrosis between the untreated and AKEX0011-treated groups using several methods. Masson’s trichrome and Sirius red staining demonstrated less extracellular matrix (ECM) deposition in silica + AKEX0011 mice than in silica + vehicle mice (Figures 3A,D,F). Accordingly, the amount of HYP in the tissue homogenate was downregulated upon AKEX0011 treatment (Figure 3G). Furthermore, decreased deposition levels of collagen I, fibronectin, and α-SMA in IHC or IF staining indicated the reduction of fibrosis (Figures 3A–I). Consistent with these results, protein and/or mRNA levels of collagen I and fibronectin additionally confirmed the attenuation of fibrosis in the AKEX0011 treatment group in comparison to the untreated group (Figures 4A–F). It is known that TGF-β is tightly linked to fibroblast activation and fibrogenesis, whereas TGF-β is tightly linked to fibroblast activation and fibrogenesis (Du et al., 2019); meanwhile, IL-4 and IL-10 are important pro-fibrotic factors. ELISA results demonstrated a reduced protein expression of TGF-β, IL-4, and IL-10 in the BALF in AKEX0011-treated mice compared to the untreated ones (Figures 4G–I).

To summarize, these findings revealed that AKEX0011 improved lung dysfunction, decreased inflammation, and attenuated fibrosis in the early therapeutic silicosis model. Noticeably, AKEX0011 treatment had a stronger effect in reducing the IL-1β level compared to PFD treatment. For other inflammatory and fibrotic biomarkers, AKEX0011 treatment tends to have a comparative effect to the positive control of PFD.

In clinical practice, early diagnosis of silicosis is very challenging; hence, silicosis patients are often diagnosed long after the silica exposure. Therefore, we further investigated whether advanced therapy of AKEX0011 could ameliorate experimental silicosis. So we gavaged mice with AKEX0011 or PFD at day 14 after silica exposure and kept drug dosing for the subsequent 28 days. The same strategy, as mentioned earlier, was used to evaluate the long-term efficacy of AKEX0011 (Figure 5A). In this study, AKEX0011 significantly alleviated the impairment of the lung function, as indicated by the amelioration of IC, Rrs, Ers, and G parameters (Figures 5B–E). Similarly, H&E staining revealed that inflammatory infiltration was significantly reduced upon AKEX0011 treatment compared to the silica vehicle-only mice (Figures 5F,G). The silica-induced increase of the levels of inflammatory factors (TNF-α, IL-6, and IL-1β) in the BALF and lung tissues was remarkably reduced in the AKEX0011 treatment group (Figure 5H-5M). Furthermore, Masson’s trichrome and IHC staining and HYP measurements demonstrated that ECM deposition was decreased after AKEX0011 treatment (Figures 6A–E), indicating attenuated pulmonary fibrosis. In line with this observation, attenuated silicosis was confirmed by the decreased expression of fibronectin, collagen I, α-SMA, and TGF-β (Figures 6F–K). In conclusion, our results indicate that AKEX0011 improved the lung function, alleviated pulmonary inflammation, and delayed fibrosis even in the advanced therapeutic silicosis model.

After confirmation of the therapeutic effect of AKEX0011 in treatment of silicosis, we further investigated its mechanisms of action. MAPKs perform a vital regulatory role in the development of inflammation and fibrosis. p38, a subclass of MAPKs, is regulated by apoptosis signal-regulating kinase 1 (ASK1), which is an upstream activator of p38. ASK1/p38 is involved in several fibrotic diseases including silicosis (Du et al., 2019). We studied AKEX0011 effects on the ASK1-p38 MAPK signal transduction pathway. Our studies showed that WB results demonstrated that exposure to silica significantly increased the phosphorylation of ASK1 and p38 in mice, whereas AKEX0011 treatment inhibited these changes (Figures 7A–C). These findings suggest that AKEX0011 might exert anti-inflammatory and anti-fibrotic effects by blocking the ASK1-p38 MAPK pathway.

Since apoptosis closely linked to silicosis, we performed TUNEL staining to detect apoptosis in the lung sections. We observed that the number of TUNEL-positive cells increased in silica + vehicle mice, while it decreased in silica + AKEX0011 mice (Figures 7D,E).

Macrophages play a key role in the development of silicosis as many previous studies demonstrated that activation of macrophages and their polarization into M1 and M2 subtypes promoted silicosis development (Carneiro et al., 2017; Zhao et al., 2020; Liu et al., 2021). To verify if this was also the case in the silicosis mouse model, we stained iNOS and Arg1, specific markers of M1 and M2, respectively. IHC on lung sections confirmed increased activation of both M1 and M2 macrophages in the silica group; the increase in the M1 subtype was more pronounced; administration of AKEX0011 reduced the accumulation of M1 and M2 macrophages, as compared to the silica group (Figures 8A–D). Furthermore, mRNA and protein levels of iNOS and Arg1 were significantly increased in the silica group, in comparison to the PBS control group, and AKEX0011 decreased the expression of these markers (Figures 8E–I). Additionally, the expressions of M1-related inflammatory factors (IL-1β, IL-6, and TNF-α) and M2-related fibrotic factors (IL-4, TGF-β, and IL-10) were also reduced upon AKEX0011 treatment.

Previous studies demonstrated that the NF-κB (p65)/IκBα pathway, of which IκBα is the major NF-κB inhibitor protein, controls M1 macrophage polarization ((Zhao et al., 2020; Carneiro et al., 2017; Liu et al., 2021)). Hence, we explored whether AKEX0011 could regulate this pathway. WB analysis revealed that silica significantly increased the ratio of the phosphorylated NF-κB to the total NF-κB level (P-NF-κB/NF-κB) and reduced the expression level of IκBα. AKEX0011 treatment reduced the P-NF-κB/NF-κB ratio and elevated the expression of IκBα, indicating that AKEX0011 might inhibit M1 macrophage polarization through blocking of NF-κB signaling (Figures 8E,J,K). Also, PPAR-γ is a vital molecule controlling M2 macrophage polarization ((Carneiro et al., 2017; Zhao et al., 2020; Liu et al., 2021)). We demonstrated that PPAR-γ was significantly upregulated in the silica group, and AKEX0011 reversed this upregulation (Figure 8E), thus surmising that AKEX0011 attenuates M2 macrophage polarization through PPAR-γ inhibition. PPAR-γ and NF-κB can affect each other. Usually, PPAR-γ activation causes NF-kB inhibition. We believe that AKEX affects these two pathways independently, resulting in an interaction between the two pathways which is not obvious.

To further explore AKEX0011 mechanisms of action at the cellular level, we performed in vitro experiments. In view of the important role of macrophages in the pathogenesis of silicosis and the protective effect of AKEX0011 on macrophage in vivo experiments, we used RAW264.7 macrophages for in vitro experiments. We used silica-stimulated macrophages and treated them with different concentrations of AKEX0011 at different timings, as described in the materials and methods.

To study preventive and therapeutic effects of AKEX0011 on the secretion of cytokines in the macrophages, we measured the levels of major cytokines in the cellular supernatant by ELISA. Our results demonstrated that AKEX0011 effectively inhibited silica-induced macrophages from the secretion of IL-1β, IL-6, TNF-α, and TGF-β; the therapeutic effects of AKEX0011 are better in the high-dose group, so we chose high-dose AKEX0011 treatment for the next experiment (Figures 9A–D). We investigated whether AKEX0011 could inhibit macrophages from silica-induced apoptosis. FACS analysis showed that silica can induce macrophage apoptosis, especially late apoptosis. In the pre-treated experiment, it can be seen that a small proportion of macrophages undergo early apoptosis, while in the post-treated experiment, almost all of macrophages undergo late apoptosis. Pre- and post-AKEX0011 treatment specifically decreased silica-induced cell apoptosis (Figure 9E), suggesting that AKEX0011 can ameliorate silicosis by inhibiting apoptosis.

Previous studies have shown that silica inducing secretion of cytokines in macrophages is partly through ASK1-p38 MAPK pathways (Carneiro et al., 2017; Zhao et al., 2020; Liu et al., 2021). In our study, WB analysis revealed that AKEX0011 effectively blocked the activation of the ASK1-p38 MAPK signaling pathway in macrophages exposed to silica (Figures 9F–H). So we speculate that AKEX0011 ameliorates silicosis by inhibiting ASK1-p38 MAPK pathways in macrophages.

Since silica can induce polarization of RAW264.7 macrophages toward certain M1 subtypes, FACS analysis revealed that the ratio of F480 + CD86 ⁺ M1 macrophages significantly increased after silica exposure for 24 h, while the ratio of F480 + CD163 + M2 macrophages did not change (Figures 9I–J) compared to the PBS control group. Consistently, mRNA and protein expressions of iNOS increased after silica exposure. We treated macrophage cells with AKEX0011 and found that it decreased the ratio of F480 + CD86 ⁺ M1 macrophages (Figures 9I–J) and the expression of iNOS, as compared to the silica group (Figures 9K–M). WB analysis revealed that AKEX0011 effectively blocked the activation of the NF-κB signaling pathway in macrophages exposed to silica (Figure 9N). Meanwhile, we have shown that the level of M1-related inflammatory factors IL-1β, IL-6, and TNF-α was reduced by AKEX0011 treatment as well. Altogether, our findings indicated that AKEX0011 can inhibit M1 polarization via the NF-κB signaling pathway in silica-induced macrophages.