S100A8/9 modulates perturbation and glycolysis of macrophages in allergic asthma mice

Cell culture and modeling

Mouse alveolar macrophages MH-S (iCell-m078, iCell Bioscience, Shanghai, China) were cultured in complete medium containing DMEM, 10% fetal bovine serum (13011-8611, Every Green, Taiwan), 0.05 mM β-mercaptoethanol (M6250, Sigma-Aldrich, St. Louis, MO, USA) and 1% penicillin/streptomycin (C0222, Beyotime, Suzhou, China). In addition, Lipofectamine 2000 (11668019, Invitrogen, Carlsbad, CA, USA) with sh-S100A8 or sh-S100A9 was incubated with cells (2  × 10⁵ cells/well) in 6-well plates for 48 h. Furthermore, macrophage polarization was induced for 8 h with OVA (Wang et al., 2021). Lipopolysaccharide (LPS) (100 ng/mL, HY-D1056, MedChemExpress, Boston, MA, USA) as a TLR4 agonist was incubated with MH-S cells for 8 h (Alhouayek et al., 2013) with PBS as a control. The cells were divided into Control, OVA, OVA + sh-S100A8, OVA + sh-S100A9, OVA + sh-S100A8 + LPS, and OVA + sh-S100A9 + LPS groups. Cell experiments of cells were repeated independently at least thrice.

Animals with allergic asthma and grouping

Wild-type male BALB/c mice (approximately 19 g), aged 6–8 weeks, from Lingchang Biotech Co., Ltd, (Shanghai, China). The mice were kept in ventilated cages in a pathogen-free animal facility and provided free access to food and water. All animal protocols were performed by professionals blinded to the group assignment in compliance with the Guidelines for the Humane Treatment of Laboratory Animals and were approved by the Animal Experimentation Ethics Committee of Zhejiang Eyong Pharmaceutical Research and Development Center (approval number: ZJEY-20221205-02).

We injected mice with physiological saline, lentivirus-sh-S100A8/A9, or lentivirus-S100A9 (1  × 10⁹ viral particles per mouse) through the tail vein a week before OVA sensitization. The mice were sensitized and challenged with OVA (A5503, Sigma, Burlington, MA, USA) as previously described (Li et al., 2021a). Briefly, we injected mice intraperitoneally 10 µg of OVA in 1 mg aluminum hydroxide (239186, Sigma, Burlington, MA, USA) with 20 g Freund’s adjuvant (77140, Thermo Fisher Scientific, Waltham, MA, USA) on days 0, 7 and 14. We subsequently challenged the mice with an OVA aerosol using an ultrasonic nebulizer (Pari Proneb nebulizer, Midlothian, WA, USA) with a 1% (wt/vol) OVA solution in saline for 20 min on days 15–21, once a day. Additionally, we intraperitoneally injected 3-bromopyruvate (3-BP, 5 mg/kg/day) into the mice once every 2 days (Fang et al., 2019) to observe glycolysis.

Randomized allocation of animals using the random number table method. We divided 24 mice into four groups to observe the effects of S100A8 and S100A9 knockdown on allergic asthma (six mice per group): negative control (NC), OVA, OVA + sh-S100A8, and OVA + sh-S100A9 groups. Additionally, 24 mice were divided into four groups to observe glycolysis (six mice per group): OVA, OVA + 3-BP, OVA + OE-S100A9, and OVA + S100A9 + 3-BP. In animal experiments, if the animals suffered infections and showed a sudden weight loss of more than 20% in one week, they were euthanized by CO₂ asphyxiation. In this study, the body weight, nutrition, and activity levels of the animals were normal. Therefore, none of these animals were excluded from the study. On day 22, a small-animal ventilator (flexiVent, SCIREQ, Montreal, Canada) was used to assess respiratory function in mice administered pentobarbital sodium for anesthesia. After mice were anesthetized with isoflurane on day 23, orbital blood was sampled to detect OVA-specific IgE and lactic acid. Following this, the mice were euthanized with carbon dioxide. Subsequently, bronchoalveolar lavage fluid (BALF) was collected for inflammatory cell counts, flow cytometry, and detection of inflammatory factors. After lavage, the lungs were isolated to fix for immunohistochemistry and histological observation or frozen at −80 °C for detection of gene and protein levels.

Detection of mRNAs by quantitative real-time PCR

Through lysis and centrifugation, total RNAs of fresh cells or tissues (frozen at −80 °C) were extracted using a total RNA small extraction kit (B618583-0100, Sangon, China). Total RNA was treated with RNase-free DNaseI. RNase-free water (20 µL; Am9932, Thermo Fisher Scientific, Waltham, MA, USA) was used to dissolve the total RNA, and an ultra-micro spectrophotometer (Nanodrop One, Thermo Fisher Scientific, Waltham, MA, USA) was used to determine its concentration and purity. Samples with ratios of 260/280 nm values between 1.9 and 2.1 and 260/230 nm values greater than 2.0, were used for subsequent experiments. HiFiScript cDNA Synthesis Kit (CW2569, CWBIO, Jiangsu, China) and SYBR Green qPCR Kit (11201ES08, Yeasen, Shanghai, China) were used for reverse transcription and stored at −20 °C pending analysis. Subsequently, quantitative real-time PCR (qRT-PCR) was performed using a LightCycler96 qRT-PCR instrument (Roche, Basel, Switzerland). The cycling conditions were determined according to the manufacturer’s instructions. β-actin was used as an endogenous control. All data were processed by relative quantitative method (2^−△△Ct). The primer sequences are listed in Table 1. Genomic DNA contamination was not detected in the qRT-PCR products.

Detection of inflammatory factors, lactic acid content, and glycolysis

Cell supernatant was used to detect the concentration of IL-1β (MM-0040M2, MEIMIAN, Jiangsu, China), IL-6 (MM-0163M2, MEIMIAN, Jiangsu, China), and TNF-α (MM-0132M2, MEIMIAN, Jiangsu, China) using ELISA. In addition, after testing respiratory function, blood was collected via orbital blood collection, isoflurane-anesthetized mice were euthanized with CO₂, the lungs were lavaged twice with one mL of cold PBS, and BALF was collected. Serum OVA-specific IgE and lactic acid levels were detected using ELISA (MM-45386M2, MEIMIAN, Jiangsu, China) and a lactic acid content detection kit (BC2230, Solarbio, Beijing, China). Supernatant of BALF was counted to detect the concentration of IL-4 (MM-01065M2, MEIMIAN), IL-13 (MM-0173M2, MEIMIAN), TGF-β1 (MM-0135M2, MEIMIAN), TNF-α, and INF-γ (MM-0182M2, MEIMIAN). The extracellular acidification rate (EACR) kit (BB48311, BestBio, Shanghai, China), phosphofructokinase (PFK) kit (BC0530, Solarbio, Beijing, China), and hexokinase (HK) kit (BC0745, Solarbio, Beijing, China) were used to observe glycolysis in cells. Assays were performed using a multifunctional microplate reader (CMaxPlus, San Francisco, CA, USA).

Inflammatory cell count

BALF cells were centrifuged (Cytospin 500, Sandton, UK) at 3,000 rpm for 5 min at 22 °C using cell centrifugation. Then, the cells on slides were stained with the Wright-Giemsa Staining kit (D010-1-2, Nanjing Jiancheng, China), air-dried, and fixed with Permount Mounting Medium (MM1411, MKBio, China). The slides were observed under a microscope, and at least 300 cells were counted for each preparation.

Polarization and pyroptosis detection by flow cytometry

To confirm MH-S cells and lung macrophage polarization and pyroptosis were analyzed by flow cytometry. After processing the cells according to the grouping, MH-S cells were prepared into 2  × 10⁷ cells/mL PBS suspensions. Subsequently, the FITC anti-F4/80 (ab60343, Abcam, Cambridge, UK), APC anti-CD86 (ab218757, Abcam, Cambridge, UK), and PE/Cy7 anti-mannose receptor [15-2] (CD206) (ab270682, Abcam, Cambridge, UK) antibodies were incubated with MH-S cells at 4 °C for 15 min. Additionally, the BALF was centrifuged at 1,000 rpm for 5 min to collect the cells. Cells from BALF resuspended by PBS to  2 × 10⁷ cells/mL and were incubated at 4 °C for 15 min with antibodies including Alexa Fluor^® 700 anti- CD45 (157210, Biolegend, San Diego, CA, USA), FITC anti-F4/80, APC anti-CD86, PE/Cy7 anti-mannose receptor [15-2] (CD206), and PE anti-CD11b (ab25175, Abcam, Cambridge, UK). The gating strategy of BALF is described in Fig. S1. For pyroptosis, the FLICA 660 Caspase-1 (9122, ImmunoChemistry, Bloomington, MN, USA) and PI staining (556547, BD, Franklin Lakes, NJ, USA) were used to incubate the cells at 37 °C for 15 min. The percentage of cells that were double-positive for caspase-1 and PI was used to indicate pyroptosis. All cells were screened through a 200-mesh sieve and flow cytometry (NovoCyte, Agilent, Santa Clara, CA, USA) was used to detect the proportion of cells.

Hematoxylin and eosin staining and immunohistochemistry

After euthanasia, the lungs were isolated from the mice, fixed with 4% paraformaldehyde for 24 h, and embedded in paraffin blocks. Subsequently, lung histopathology was performed using hematoxylin and eosin (HE) staining. Lung paraffin blocks were sliced into 4 µm sections, dewaxed, and hydrated. Finally, the sections were stained using an HE kit (C0105S, Beyotime, Shanghai, China) to observe morphology and cell morphology under a Nikon Eclipse Ci-L microscope (Tokyo, Japan).

Western blot

Cell or tissue proteins were extracted using RIPA solution (P0013C, Beyotime, Shanghai, China) containing protease inhibitors (CW2200S, CWBIO, Beijing, China). After extraction and denaturation of total proteins, 10% gel electrophoresis was used to separate the proteins, and activated PVDF membranes were used for transfer. Subsequently, the blocked-membranes with 5% skim milk powder were incubated with primary antibodies such like anti-TLR4 (1:1000, 14358S, CST, Boston, MA, USA), anti-MyD88 (1:1000, ab219413, Abcam, Cambridge, UK), anti- TIR domain containing adaptor molecule 1 (TRIF) (1:3000, ab13810, Abcam, Cambridge, UK), anti-NF-κB (1:3000, 8242T, CST, Danvers, MA, USA), anti-IκBα (1:3000, 4814T, CST, Danvers, MA, USA), Anti-ATP citrate lyase antibody (ACLY) (1:10000, ab40793, Abcam, Cambridge, UK), phospho- ACLY (1:1000, 4331T, CST, Danvers, MA, USA), and anti- β-actin (1:20000, 81115-1-RR, Proteintech, Chicago, IL, USA) followed by corresponding secondary antibodies (1:6000, 7074/7076, CST, Danvers, MA, USA) at 25 °C for 2 h. The membrane was incubated with ECL reagents (610020-9Q, Qing Xiang, Shanghai, China) and visualized by ImageJ software.

Statistical analysis

The statistical software SPSS 19.0 (IBM, Armonk, NY, USA) was used to analyze the data, and the continuous variables were presented as mean ± standard deviation. Data from multiple in vivo experiments were analyzed using one-way analysis of variance (ANOVA) with a post-hoc Tukey test. In cases where measurement data were not normally distributed, the Kruskal–Wallis H test was used. Any p-value that resulted in lower than 0.05 was deemed statistically significant.

S100A8/9 knockdown inhibited polarization of ovalbumin-induced MH-S model

We examined the effects of S100A8 and S100A9 knockdowns on MH-S cell polarization and inflammation. As shown in Fig. S2, the S100A9 or S100A8 knockdown cell models were successfully established. The concentrations of the inflammatory factors, IL-1β, IL-6, and TNF-α were significantly increased in the OVA group (p < 0.01), while the knockdown of S100A8 or S100A9 decreased their concentrations (p < 0.01) (Figs. 1A–1C). We detected biomarkers mRNA of M1 macrophage (IL-6, IL-1β, and iNOS) and M2 macrophage (IL-10, Arg1, and Fizz1) and found that they were increased in the OVA group, except for IL-10 (p < 0.01), while knockdown of S100A8 or S100A9 decreased IL-6, IL-1β, iNOS, Arg1 and Fizz1 mRNA and increased IL-10 mRNA (p < 0.01) (Figs. 1D–1I). Furthermore, we detected the proportion of M1 (CD86+) and M2 (CD206+) cells using flow cytometry (Figs. 1J, 1K). In the OVA group, the proportion of M1 (CD86+) and M2 (CD206+) cells was increased, whereas S100A8 or S100A9 knockdown inhibited this increase (p < 0.01) (Figs. 1J, 1K). Additionally, LPS intervention antagonized the effects of S100A8 or S100A9 knockdown on OVA-induced MH-S cell injury models (p < 0.05) (Figs. 1A–1K). These results suggest that the knockdown of S100A8 or S100A9 has an inhibitory effect on OVA-induced macrophage polarization, whereas LPS is a TLR4 agonist that can partially antagonize this inhibitory effect.

S100A8/9 knockdown inhibited pyroptosis and glycolysis of ovalbumin-induced MH-S model

We measured pyroptosis and found that the knockdown of S100A8 or S100A9 significantly reduced MH-S cell pyroptosis (p < 0.05), while LPS antagonized it (p  <  0.01) (Figs. 1N, 1O). We also measured the EACR to observe glycolysis in MH-S cells. OVA intervention significantly decreased EACR at 1–63 min (p  < 0.01) (Table 2). S100A8 knockdown significantly decreased ECAR of MH-S cells with OVA intervention, except at 1 and 36 min (p < 0.05) (Table 2). S100A9 knockdown significantly decreased ECAR of MH-S cells with OVA intervention at 9–63 min (p < 0.05) (Table 2). LPS intervention significantly counteracted the inhibitory effect of S100A8 knockdown (p < 0.05) and in OVA-inverted MH-S cells with S100A9 knockdown, LPS intervention significantly increased the ECAR at 9, 18, 27, 45, 54, and 63 min (p < 0.05) (Table 2). Furthermore, the concentrations of lactic acid, FPK, and HK were increased in MH-S cells treated with OVA intervention (p < 0.01) (Figs. 2A–2C). We also measured the mRNA levels of glycolysis-related genes. In OVA-induced MH-S cells, GAPDH, HK2, and LDHA mRNA levels significantly increased, whereas PDH mRNA levels decreased (p < 0.01) (Figs. 2D–2F). S100A8 or S100A9 knockdown inhibited GAPDH, HK2, and LDHA mRNA expression in OVA-induced MH-S cell injury models and increased PDH mRNA levels (p < 0.01); however, LPS intervention antagonized these effects (p < 0.05) (Figs. 2D–2G). The activation of the TLR4/MyD88/TRIF/NF-κB signaling pathway promotes glycolysis (Li et al., 2021b). Therefore, we detected the TLR4/MyD88/TRIF/NF-κB signaling pathway by Western blot. The levels of TLR4, MyD88, TRIF, NF-κB, and p-ACLY/ACLY in MH-S cells were increased and S100A8 or S100A9 knockdown decreased them (p < 0.01), while LPS intervention partially antagonized the inhibition of S100A8 or S100A9 knockdown on TLR4/MyD88/TRIF/NF-κB signaling pathway (p < 0.05) (Figs. 2H–2K, 2M, 2N). Furthermore, IκBα expression levels had the opposite trend to these proteins (p < 0.05) (Figs. 2L, 2N). We found that S100A8 or S100A9 knockdown inhibited macrophage pyroptosis and glycolysis through the TLR4/MyD88/TRIF/NF-κB signaling pathway.

S100A8/9 knockdown improved respiratory function, lung tissue injury, and inflammation of BALF in ovalbumin-sensitized and challenged mice

We established a mouse model of allergic asthma using OVA to observe the protective effects of S100A8 and S100A 9 knockdown in the lungs post-OVA-challenged. As shown in Fig. S3, we successfully established OVA-sensitized and -challenged mice with S100A8 or S100A9 knockdown. To verify the role of S100A8 and S100A9 in allergic asthma mice, we assessed respiratory function, observed pathological damage to lung tissue, and measured the levels of inflammatory cells and cytokines in BALF. The respiratory function indicators tidal volume (TV), vital capacity (VC), expiratory volume (EV), minute ventilation volume (MV), forced expiratory volume in 0.1 s (FEV0.1), end inspiratory pause (EIP), peak expiratory flow (PEF), mid expiratory flow (EF50), and dynamic lung compliance (Cdyn) in OVA mice were markedly decreased while S100A8 and S100A9 knockdown reversed them (p < 0.01) (Figs. 3A–3I). Mice with OVA-sensitized and -challenged had increased serum OVA-specific IgE levels (p < 0.01) (Fig. 3J), whereas S100A8 or S100A9 knockdown markedly decreased IgE levels compared to those in the OVA group (p < 0.01) (Fig. 3J). Histological analysis indicated an increase in peribronchial inflammatory infiltrates in the lungs of mice that were sensitized and challenged with OVA, whereas S100A8 or S100A9 knockdown markedly reduced it (Fig. 3K). Furthermore, macrophages, lymphocytes, neutrophils, and eosinophils were counted by Diff-Quik staining, and concentrations of IL-4, IL-13, TGF- β1, TNF-α, and IFN-γ were measured by ELISA. Most of them were significantly increased in the BALF from mice sensitized and challenged with OVA (p < 0.01) (Figs. 3L–3S) but the IFN-γ concentration was decreased (p < 0.01) (Fig. 3T). S100A8 or S100A9 knockdown antagonized the OVA-induced inflammatory response (p < 0.01) (Figs. 3L–3T). In summary, S100A8/9 knockdown improved respiratory function, lung tissue injury, and inflammation in mice sensitized and challenged with OVA.

S100A8/9 knockdown suppressed macrophage polarization in OVA-sensitized and challenged mice

Changes in glycolysis in vivo. Flow cytometry was used to detect macrophage polarization in the BALF. The results showed that the proportions of M1 (CD86+) and M2 (CD206+) cells were increased in mice sensitized and challenged with OVA, whereas S100A8 or S100A9 knockdown decreased them (p < 0.05) (Figs. 4A, 4B). IHC was used to detect the expression of macrophage biomarkers. The total macrophage biomarkers CD68, M1 macrophage biomarker IRF-5, and M2 macrophage biomarker YM-1 were measured, and their positive cells increased in OVA-sensitized and challenged mice (p < 0.01) (Figs. 4C, 4D). However, S100A9 knockdown significantly decreased these levels in OVA-sensitized and challenged mice (p < 0.05) (Figs. 4C, 4D). Furthermore, the mRNA levels of IL-6, iNOS, Arg1, and IL-10 were determined using qRT-PCR. Among these, IL-6 and iNOS are genes related to M1 macrophages, whereas Arg1 and IL-10 are related to M1 macrophages (Zhu et al., 2015). They increased in OVA-sensitized sensitized and challenged mice, whereas S100A8 or S100A9 knockdown decreased them, except for IL-10, which showed the opposite trends (p < 0.01) (Figs. 4E–4F).

S100A8/9 knockdown inhibited glycolysis in the lung of OVA- sensitized and challenged mice

Serum lactic acid concentration was markedly increased in the OVA group. However, S100A8 and S100A9 knockdown inhibited it (p < 0.01) (Fig. 5A), suggesting that glycolysis was inhibited. Pyruvate dehydrogenase (PDH), lactate dehydrogenase (LDH) A, and HK2 are key enzymes involved in glycolysis (Pereverzeva et al., 2022). Their mRNAs increased in OVA-sensitized and challenged mice, whereas S100A8 and S100A9 knockdown significantly decreased their levels (p < 0.01) (Figs. 5A–5D). Furthermore, we used Western blot to measure the levels of LDHA, HK2, TLR4, MyD88, p-NF-κB/NF-κB, p-IκBα/IκBα, gasdermin D-N, and cleaved-caspase-1/caspase-1. OVA-induced allergic asthma increased all levels, whereas S100A8 and S100A9 knockdown decreased the levels in OVA-sensitized and challenged mice (p < 0.01) (Figs. 5E–5M).

S100A9 overexpression had an adverse impact on respiratory function and lung tissue while enhancing inflammation in ovalbumin-sensitized and challenged mice

Evidence proved that the expression level of S100A8 is regulated by S100A9 (Hobbs et al., 2003). We found knockdown of S100A9 ameliorated injury in allergic mice and inhibited glycolysis in macrophages and lung tissues. To clarify the ability of S100A9 to regulate glycolysis, we used allergic asthmatic mice overexpressing S100A9 to observe the promotion of glycolysis by S100A9 overexpression. The glycolysis inhibitor, 3-BP, can inhibit HK2 (Zhong et al., 2022). Using 3-BP treatment, we explored the potential mechanisms by which S100A9 regulates glycolysis. S100A9 mRNA levels in the OVA + OE-S100A9 group were 2.94 times higher than those in the OVA group, and 3-BP intervention did not affect the mRNA levels (p < 0.01) (Fig. 6A). Serum OVA-specific IgE in OVA-sensitized and challenged mice with the 3-BP intervention was significantly decreased compared to that in the OVA group (p < 0.01), while S100A9 overexpression markedly increased IgE compared to that in the OVA group (p < 0.01) (Fig. 6B). Compared with the OVA group, TV, VC, EV, MV, FEV0.1, EIP, PEF, EF50, and Cdyn in the OVA + 3-BP group were markedly increased, whereas S100A9 overexpression increased them (p < 0.05) (Figs. 6C–6K). Histological analysis indicated a decrease in peribronchial inflammatory infiltrates in the lungs of OVA-sensitized and challenged mice with 3-BP intervention, while S100A9 overexpression enhanced lung tissue damage (Fig. 6L). Furthermore, in the BALF of the OVA + 3-BP group compared to that of the OVA group, macrophages, lymphocytes, neutrophils, eosinophils, and the concentrations of IL-4, IL-13, TGF-β1, and TNF-α were significantly decreased (p < 0.01), while their levels were increased in OVA-sensitized and challenged mice with S100A9 overexpression (p < 0.01) (Figs. 6M–6T). However, compared to the OVA group, IFN-γ concentration was increased in the BALF of the OVA + 3-BP group, whereas it was decreased in the OE-S100A9 group (p < 0.01) (Fig. 6U). Furthermore, 3-BP treatment antagonized the effects of S100A9 overexpression on inflammation. S100A9 overexpression had detrimental effects on respiratory function and lungs in OVA-sensitized and -challenged mice, while also enhancing inflammation by promoting glycolysis.

S100A9 overexpression promoted macrophage polarization in ovalbumin-sensitized and challenged mice

Flow cytometry results showed that the proportion of M1 (CD86+) and M2 (CD206+) macrophages was decreased in OVA-induced allergic asthma mice with 3-BP intervention, whereas they increased in OVA-sensitized and challenged mice with S100A9 overexpression (p < 0.01) (Figs. 7A, 7B). Immunohistochemistry was performed to observe the expression levels of macrophage biomarkers. CD68 is a macrophage biomarker, IRF-5 is an M1 macrophage biomarker, and YM-1 is an M2 macrophage biomarker. CD68, IRF-5, and YM-1 positive cells were all decreased in OVA-sensitized and challenged mice with 3-BP administration, but were significantly increased after S100A9 overexpression (p < 0.01) (Fig. 7D). In the lungs, IL-6, iNOS, and Arg1 mRNA levels decreased in OVA-sensitized and challenged mice with 3-BP intervention and increased in OVA-sensitized and challenged mice with S100A9 overexpression, whereas IL-10 showed the opposite trend. Furthermore, results showed that 3-BP intervention antagonized the stimulatory effects of S100A9 overexpression on macrophage polarization in OVA-sensitized and -challenged mice (p < 0.01) (Fig. 7).

S100A9 overexpression inhibited glycolysis in the lung of ovalbumin-sensitized and challenged mice

Compared to the OVA group, the serum lactic acid concentration in OVA-sensitized and -challenged mice was inhibited by 3-BP intervention and were increased in the OVA + OE-S100A9 group (p < 0.01) (Fig. 8A). PDH, LDH, and HK2, the key enzymes of glycolysis, were measured by qRT-PCR and were decreased in mice post-OVA challenge with 3-BP intervention, whereas they were increased in OVA-sensitized and challenged mice by S100A9 overexpression (p < 0.05) (Figs. 8B–8D). The expression levels of LDHA, HK2, TLR4, MyD88, p-NF-κB/NF-κB, p-IκBα/IκBα, gasdermin D-N, and cleaved-caspase-1/caspase-1 levels were measured by Western blot. Compared to the OVA-sensitized and challenged mice, they all decreased in the OVA + 3-BP group and increased in the OVA + OE-S100A9 group (p < 0.01) (Figs. 8E–8L). Additionally, 3-BP antagonized the promotion of S100A9 overexpression in serum lactic acid, glycolysis-related enzymes, and genes.