RAW264.7 cells, provided by Servicebio (Wuhan, China), were cultured in Dulbecco’s modified Eagle medium (Gibco 11965092) supplemented with 1% (v/v) penicillin (100 U/mL), streptomycin (100 μg/mL) and 10% fetal bovine serum (FBS; Gibco, 10099141C), followed by incubation at 37°C in a 5% CO₂ incubator.

A Cell Counting Kit-8 (NCM Biotech C6005, China) was used to assess cell viability. Briefly, RAW 264.7 cells were cultured in 96-well culture plates at a density of 2 × 10⁴ cells/well and incubated for 24 h before removal of the cell culture medium. Afterward, the cells were pretreated with 1 μg/mL lipopolysaccharide (LPS) (Beyotime S1732, China) for 1 h and then coincubated with the oridonin derivatives for 24 h. A mixture of 10 μL of CCK-8 reagent with 90 μL of cell culture medium was added to each well for incubation at 37°C for 50 min before the absorbance of the solution was measured at 450 nm by using a BioTek Synergy HTX Multi-Mode Reader.

The production of NO was determined with a NO assay kit (Beyotime S0021M, China) according to the manufacturer’s instructions. Briefly, the cell culture supernatant was mixed with Griess Ⅰ and Griess II in sequence and then incubated at room temperature for 10 min as described in our previous report (Zhang et al., 2024; Li et al., 2024). The absorbance was measured at 540 nm with a BioTek Synergy HTX Multi-Mode Reader. ELISAs for IL-1β and IL-6 were performed via ELISA kits (Boster EK0394, EK0411, China) according to the manufacturer’s instructions.

The total protein was extracted from the cells and lung tissues with RIPA lysis buffer (Beyotime P0013, China). The proteins were separated by SDS‒PAGE and transferred to a PVDF membrane before 1 h of blocking with 5% skim milk at room temperature. The membranes were further incubated overnight at 4°C with primary antibodies (Cell Signaling Technology: β-actin, 8H10D10; COX-2, D5H5; NLRP3, D4D8T; IL-6, D5W4V; P-IκBα, 14D4; IκBα, L35A5; P-NF-κB, 93H1; NF-κB, D14E12; and GAPDH: D16H11) at a dilution of 1:1,000. Afterward, the membranes were incubated with secondary antibodies (Cell Signaling Technology: anti-mouse: 7,076; and anti-rabbit: 7,074) at a 1:1,000 dilution for 1 h at room temperature. The membranes were subsequently exposed to a ChemiDoc imaging system (Bio-Rad) after treatment with enhanced chemiluminescence (ECL) substrate.

BALB/c male mice aged 6–8 weeks and weighing 18–22 g were purchased from Zhuhai Bestest Biotech Co., Ltd. (certificate SCXK20200051; Zhuhai, China). The animals were housed in an environment with constant room temperature on a 12/12 h light−dark cycle and had free access to food and water. The mice were randomly divided into four groups: control (PBS with 10% DMSO), LPS (1.25 mg/kg in PBS), LPS (1.25 mg/kg in PBS) + oridonin (20 mg/kg in PBS with 10% DMSO), and LPS (1.25 mg/kg in PBS) + 4c (20 mg/kg in PBS with 10% DMSO). Oridonin and 4c were injected intraperitoneally, followed by LPS infusion into the nasal cavity 1 h later. The mice were euthanized by carbon dioxide (100%) asphyxiation 24 h after LPS administration, and then lung tissue samples were collected for analysis. Histological analysis of the lung tissues was performed after the tissues were immersed in 4% paraformaldehyde, embedded in paraffin, cut into 4 μm sections, and stained with hematoxylin‒eosin (HE) before examination with a microscope (Nikon, Japan).

The paraffin slides were dewaxed and rehydrated. An antigen retrieval protocol using heat was used to unmask the antigens (30 min in citrate buffer 0.01 M, pH 6.0), then block the slides with serum blocking reagent. The slides were incubated with LCA (CD45) primary antibody (60287-1-lg, Proteintech, United States) overnight at 4°C. After washing with TBST for three times, the slides were incubated with the secondary antibody. Washing slides with TBST for three times then adding DAB Chromogen Solution to cover the entire tissue section and incubate for 10 min. Slides were lightly counterstained with hematoxylin to reveal nuclei, examined and photographed with a microscope.

Total RNA from the cells and lung tissues was extracted with an RNeasy^® RNA extraction kit (Solarbio R1200, China) and dissolved in enzyme-free water. cDNA was synthesized with PrimeScript RT Master Mix (Takara RR036A, Japan) on a Biometra TAdvanced 96 SG (Analytik Jena, Jena, Germany). The total cDNA was used as the starting material for qPCR with GOTaq qPCR Master Mix (Promega A6002, United States) on a StepOnePlus Real-Time PCR system (Thermo Fisher, United States) according to the manufacturer’s instructions. The relative expression of the target genes was calculated via the 2^−ΔΔCT method, with GAPDH used as a control. The mouse cDNA primer sequences for qPCR are listed in the supplementary materials (Supplementary Table S1).

RAW 264.7 cells were seeded in 3 cm dishes, treated with 1 μg/mL LPS for 1 h and then coincubated with oridonin derivatives for 24 h before RNA extraction. The samples were prepared in triplicate. Total RNA was extracted with TRIzol reagent (15596026, Invitrogen) according to the manufacturer’s instructions. Preparation of the RNA library and transcriptome sequencing were conducted by Novogene Co., Ltd. (Beijing, China). Genes with adjusted p values <0.05 and |log2 (fold change)| values > 1 were considered to be significantly differentially expressed.

AutoDockFR (ADFR) software was used for molecular docking. The 3D structure of compound 4c was generated in ChemDraw 22.0, followed by MM2 energy optimization. The X-ray crystal structure of the NLRP3 protein served as the receptor after the elimination of the unnecessary water molecules and small ligands bound to the protein. The Ori-NLRP3 protein interaction sites were defined according to a published paper (He et al., 2018). The binding results were visualized with PyMOL version 2.5.2 software.

GraphPad Prism (Version 9.5.1) was used for statistical analysis. The results are expressed as the means ± SDs of data from at least 3 independent experiments. Student’s t-test or one-way ANOVA was used to determine the statistical significance of differences between two or more groups. A p value < 0.05 was considered to indicate statistical significance.

NO is strongly associated with the pathogenesis of inflammatory diseases. The Griess assay for NO production was considered the gold standard for screening oridonin analogs for the treatment of inflammatory-associated disorders. Initially, the appropriate concentration of LPS to induce inflammation in the RAW264.7 macrophage line was investigated. As shown in Figures 1A, B, LPS at concentrations ranging from 0.125 to 8 μg/mL showed no toxicity to RAW264.7 cells and could stimulate NO production in these cells. Hence, 1 μg/mL LPS was utilized to evaluate the ability of the oridonin derivatives to inhibit NO release. In preliminary experiments, we reported that the highest concentrations of oridonin and its derivatives with no severe cytotoxicity were 5 μM and 0.3125 μM, respectively. Therefore, the initial treatment concentrations of oridonin and its analogs were set to these two concentrations. Notably, treatment of RAW264.7 cells with oridonin and its derivatives significantly suppressed the release of NO induced by LPS at the abovementioned concentrations (Figure 1C). In summary, these oridonin derivatives significantly inhibited LPS-induced NO release from RAW264.7 cells.

On the basis of the initial anti-inflammatory activity screening results (Figure 1C), three pairs of oridonin derivatives, 4a and 5a, 4b and 5b, and 4c and 5c, which had high structural similarity and promising anti-inflammatory activity, were further investigated for IC₅₀ determination. The IC₅₀ values of the oridonin derivatives for NO inhibition are shown in Table 1. As shown in Figures 2A–G, oridonin derivatives 4a, 4b and 4c with 7,20-epoxy ent-kaurane diterpenoid scaffolds had greater inhibitory effects on NO release than did 5a, 5b and 5c with 6,20-epoxy ent-kaurane diterpenoid scaffolds, suggesting that DAST rearrangement of the scaffold did not significantly improve the anti-inflammatory activity of the compound in RAW246.7 cells. Compound 4c, with an IC₅₀ of 0.28 ± 0.01 μM, displayed the highest anti-inflammatory activity, with approximately 17-fold greater anti-inflammatory activity than oridonin in the NO release assay.

Oridonin hybrids conjugated with FDA-approved NSAIDs were then synthesized to evaluate if there was a synergistic effect after hybridization. NO release assays (Figure 3) revealed that, compared with oridonin derivative 4c, compounds 4g-j presented lower anti-inflammatory activity.

Compounds 4b and 4c, which had the lowest IC₅₀ values for NO production, were selected for further investigation of their inhibitory effects on IL-6 and IL-1β secretion from RAW 264.7 cells. The results shown in Figure 4A and B reveal that LPS dramatically induced IL-6 and IL-1β secretion. However, both 4b and 4c significantly inhibited the secretion of IL-6 and IL-1β in a dose-dependent manner. Accordingly, the IC₅₀ values of compound 4b against the secretion of IL-1β and IL-6 were 0.22 ± 0.02 μM and 0.23 ± 0.02 μM, respectively, whereas for compound 4c, these values were 0.21 ± 0.02 μM and 0.21 ± 0.03 μM, respectively. These data indicate that the inhibitory effect of 4c on the release of IL-6 and IL-1β was slightly greater than that of 4b.

Next, we evaluated the inhibitory effects of 4b and 4c on inflammatory mediators at the mRNA level. As shown in Figures 5A, B, the qPCR results indicated that LPS enhanced the mRNA expression of the inflammatory genes IL-6, COX-2, IL-1β, TNF-α and iNOS in the RAW 264.7 cell line. Notably, compounds 4b and 4c inhibited the LPS-stimulated expression of the inflammatory genes IL-6, COX-2 and IL-1β. Moreover, 4c, compared with 4b, much more effectively inhibited the expression of these inflammatory genes. Unfortunately, neither 4b nor 4c inhibited the LPS-stimulated expression of the inflammatory genes iNOS and TNF-α. In summary, 4c was more effective than 4b in inhibiting the LPS-stimulated expression of the inflammatory genes IL-6, COX-2, and IL-1β, but not iNOS and TNF-α.

Previous studies have indicated that oridonin is a covalent inhibitor of NLRP3. Therefore, we also measured the expression of inflammation-related proteins, including those involved in the NLRP3 inflammasome and the NF-κB pathway. As shown in Figures 6A–M, 4b and 4c regulated the NF-κB signaling pathway by inhibiting the LPS-induced expression of phosphorylated NF-κB and phosphorylated IκB, as well as their downstream target COX-2. Importantly, 4b and 4c also inhibited the NLRP3 inflammasome by inhibiting the expression of NLRP3 and its downstream effector IL-6. Interestingly, as shown in Figures 2D, F, compounds 4b and 4c significantly inhibited LPS-induced NO production. However, the results of the qPCR assay shown in Figure 5A and B indicated that 4b and 4c could not downregulate iNOS mRNA. However, WB analysis suggested that iNOS protein expression was significantly inhibited by 4c but not by 4b. In summary, 4c was more effective than 4b in inhibiting the LPS-induced expression of inflammatory proteins, including phosphorylated NF-κB, phosphorylated IκB, NLRP3, IL-6 and iNOS.

We verified that these oridonin derivatives, among which 4c was the most promising, showed promising in vitro anti-inflammatory activity. Next, we established a mouse model of ALI to determine the in vivo anti-inflammatory activity of 4c. As shown in Figure 7A, after nasal inhalation of LPS, the inflammatory cell infiltration in the lung tissue of the mice was examined. Notably, inflammatory infiltration was partially alleviated by oridonin (20 mg/kg) and fully alleviated by 4c (20 mg/kg). Moreover, we performed IHC staining of the lung tissue with leukocyte common antigen (LCA, also known as CD45) antibody as a biomarker of inflammatory cells. Figures 7B,C indicated that the percentage of LCA positive inflammatory cells in the lung tissue was significantly upregulated after LPS administration, and 20 mg/kg of 4c treatment could reduce the percentage of LCA positive inflammatory cells more significantly than oridonin. In addition, the expression of the genes IL-6 and TNF-α in lung tissue was detected by qPCR. Oridonin and 4c significantly reduced the LPS-induced expression of the IL-6 and TNF-α genes, and 4c was much more effective than oridonin (Figures 7D, E). WB analysis of the proteins extracted from the lung tissues revealed that both oridonin and 4c significantly inhibited the LPS-induced expression of NLRP3, phosphorylated NF-κB and IL-6, and 4c was more effective than oridonin (Figures 7F–I). In conclusion, compared with oridonin, 4c showed much more promise for alleviating ALI induced by LPS in mice.

ADFR software was employed to elucidate the binding mode of compound 4c with the NLRP3 protein, whose crystal structure (ID: 8etr) was downloaded from the Protein Data Bank (https://www.rcsb.org/) (Ravindranath et al., 2015; McBride et al., 2022) The structure of compound 4c with the lowest energy was prepared with Chem 3D 20.0. The docking results shown in Figure 8 indicate that the α,β-unsaturated carbonyl unit of compound 4c could act as a Michael acceptor that targets the thiol group of the residue Cys279 of the NLRP3 protein to form a stable C-S covalent bond, which had been considered during a previous oridonin target investigation (He et al., 2018). Additionally, the protein‒ligand interaction profiler (PLIP) website revealed that compound 4c could form two hydrogen bonds with residues Tyr143 and Glu511 (marked in yellow), as well as salt bridges with the amino acid residue Arg147 (marked in red). The length of the hydrogen bonds ranged from 3.16 Å to 3.26 Å, implying strong physicochemical forces. These results suggested that the in silico predictions were in high accordance with the aforementioned in vitro and in vivo results and met our initial expectations, which indirectly proved the feasibility of the NLRP3-based structural design strategy.

Both the in vitro and in vivo assays verified that the anti-inflammatory effects of 4c occurred via modulation of the NLRP3 inflammasome and the NF-κB pathway. However, the specific mechanism is still not clear. Therefore, RNA-seq analysis was conducted to determine the genes that were differentially expressed in RAW264.7 cells after 4c treatment. As shown in Figure 9A, the expression of 573 genes was significantly altered after treatment with 4c, among which 303 were upregulated and 270 were downregulated. Through cluster heatmap analysis, we found that 4c could significantly reverse LPS-induced gene regulation (Figure 9B). In addition, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses revealed that inflammation-associated signaling pathways were significantly affected after 4c treatment (Figures 9C–F).

On the basis of the aforementioned RNA-seq results, we further used a qPCR technique to verify the top upregulated and downregulated genes (Table 2). The expression of the top 3 upregulated genes, Trdc, Stfa2 and Gsta2, was significantly inhibited after the 4c treatment (Figures 10A–C). The most downregulated gene, Spib (Figure 10D), and 2 more genes, Nr4a1 and Csf2 (Figures 10E, F), were confirmed to be downregulated after 4c treatment in a dose-dependent manner. However, the expression of Mybpc2 and Heg1 was not significantly affected by 4c treatment (Figures 10G, H). For the first time, it was verified that Trdc, Stfa2, Gsta2, Spib, Csf2 and Nr4a1 are significantly regulated by oridonin derivatives.