Hederasaponin C (>98%) was isolated from Pulsatilla chinensis (Bunge) Regel in our laboratory and determined by HLPC. (4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), LPS (Escherichia coli, serotype 0111: B4), Ca²⁺ detector of Fluo-3/AM, were purchased from Sigma-Aldrich (St. Louis, MO, USA). PierceTM Protein A/G Magmetic Beads (#88802) were obtained from Thermo Fisher Scientific. Dulbecco’s modified eagle medium (DMEM), 1640, fetal bovine serum (FBS) were acquired from Gibco (Grand Island, NY, USA). ATP were purchased from Macklin (Shanghai, China). IL-1β (#224603-014), IL-6 (#225266-022), and TNF-α (#234889-001) ELISA kits were obtained from Invitrogen (Vienna, Austria). Antibodies against PLCγ2(#3872, western blotting:1:1000), IP3 Receptor1 (#8568, western blotting:1:1000), DAG Lipaα (#13626, western blotting:1:1000), TAK1 (#4505, western blotting:1:1000), p-TAK1 (#9339, western blotting:1:1000), PKCα (#2056S, western blotting:1:1000), p- PKCα (#9375S, western blotting:1:1000), p65 (#8242T, western blotting:1:1000), p-p65 (#3033, western blotting:1:1000), poly (ADP-ribose) polymerase (PARP) (#9532, western blotting:1:1000), NLRP3 (#15101, western blotting:1:1000), ASC (#13833, western blotting:1:1000), IL-1β (#31202, western blotting:1:1000), Cleaved-Caspase-1 (#89332 and #4199S, western blotting:1:1000), Cleaved-IL-1β (#63124 and #83186S, western blotting:1:1000), GAPDH (#5174, western blotting:1:1000) and the secondary antibodies including anti-rabbit IgG, HRP-linked Antibody (#7074) and anti-mouse IgG, HRP-linked Antibody (#7076) were obtained from Cell Signaling Technology (Beverly, MA, USA). Caspase-1 (ab1872, western blotting:1:1000) were obtained from Abcam (Cambridge, MA, USA). PIP2 (#53412, western blotting:1:1000) were obtained from Senta (Santa Cruz Biotechnology, USA).

The Cell Bank of the Chinese Academy of Sciences (Shanghai, China) provided RAW264.7 and THP-1 cells, the Kunming Cell Bank of the Chinese Academy of Sciences (Kunming, China) provided J774A.1 cells. RAW264.7 and J774A.1 cells were grown in DMEM (10% FBS). THP-1 cells were cultured in RPMI 1640 medium (10% FBS). All cells were cultured in a CO₂ concentration of 5% with a temperature of 37°C.

RAW264.7 (4×10⁴ cells/well), J774A.1 (2×10⁴ cells/well), and THP-1 (2×10⁴ cells/well) were plated into 96-well plates overnight. The cells were pro-treated with HSC (0, 20, 40, and 80 μM) for 24 h. Each well was added MTT solution and incubated for 4 h. The plate was detected by a microplate reader with an absorbance of 570 nm.

In 12-well plates, RAW264.7 (1.5×10⁵ cells per well), J774A.1 (1×10⁵ cells per well), and THP-1 (1×10⁵ cells per well) cells were seeded overnight. Following a 4 h pre-treatment with HSC (0, 20, 40, and 80 μM), the cells were co-treated with another 6 h with or without LPS (1 μg/ml). Fluo-3/AM (1 mM, 1 h) probes was used to measure the level of intracellular Ca²⁺. An automated flow cytometer (Becton-Dickinson, Franklin Lakes, NJ, USA) was used to detect the fluorescence intensity. For the analysis, 1×10⁴ cells were collected and completely gated.

The concentration of IL-6 and TNF-α was determined by using ELISA kits according to the manufacturer’s specifications. In 12-well plates, RAW264.7 (1.5×10⁵ cells per well), J774A.1 (1×10⁵ cells per well), and THP-1 (1×10⁵ cells per well) cells were cultivated overnight. The cells were pre-processed with HSC (0, 20, 40, and 80 μM) for 4 h, then stimulated with LPS (1 μg/ml) for 18 h. The supernatant was collected and the concentration of IL-6 and TNF-α was measured by using an ELISA. The concentration of IL-1β was determined according to the manufacturer’s specifications. In 12-well plates, J774A.1 (1×10⁵ cells per well) and THP-1 (1×10⁵ cells per well) cells were grown overnight. The cells were pre-processed with HSC (0, 20, 40, and 80 μM) for 4 h, the cells were stimulated with LPS (1 μg/ml) for 6 h, and then co-treated with ATP (5 mM) for further 30 minutes. The supernatant was collected to detect the concentration of IL-1β.

RAW264.7 and THP-1 cells were plated into 96-well plates at densities of 4×10⁵ and 2×10⁵cells per well, respectively overnight. The cells were pre-treated with HSC (80 μM) for 4 h before being co-cultured with or without LPS for another 6 h. The cells were labeled with Fluo-3/AM (10 μM) for 1 h. The fluorescence images were captured by a fluorescence microscope (Leica, DMi8, Wetzlar, Germany).

As previously mentioned, the immunofluorescence of NF-κB/p65, PIP2, PLCγ2, NLRP3, Caspase-1, ASC, and Ca²⁺ was investigated (20). Briefly, 2×10⁵ RAW264.7 cells were seeded into a confocal dish overnight (SPL, Pocheon, Korea). The cells were pre-treated with HSC (80 μM) for 4 h before being stimulated with LPS (1 μg/mL) for further 2 h. Incubation with anti-NF-κB/p65 antibody (1:100) overnight at 4°C, followed by incubation with goat anti-rabbit Alexa Fluor 594 secondary antibody (1:200) for 1 h at room temperature. After that, the cells were fixed and stained for 30 minutes with Hoechst 33342 (1 μM) before being imaged. The immunofluorescence of Caspase-1, PIP2, PLCγ2, and ASC was conducted using the same method as with NF-κB/p65, except the different dilutions of anti-Caspase-1 (1:200), anti-PIP2 (1:100), anti-PLCγ2 (1:100), anti-ASC (1:100). Goat anti-mouse IgG Fluor^® 488 Conjugate (#S0017, 1:200), Alexa Fluor 488 (1:200) secondary antibody. The cells were imaged with a confocal laser microscope (Leica, Wetzlar, Germany).

J774A.1 cells (3×10⁵ cells per well) were plated into 6-well plates overnight. Plasmids of EGFP-Ca²⁺ and EGFP-NLRP3 were transfected for 48 h by using turboFect transfection reagents (#R0531, Thermo Fisher Scientific, Grand Island, NY, United States). Then the cells were seeded in a confocal dish (SPL, Pocheon, Korea) overnight. The immunofluorescence of Ca²⁺ and NLRP3 was conducted by using the above method.

RAW264.7, J774A.1, and THP-1 cells were plated into a dish at a density of 1×10⁶ cells overnight. Pre-treatment with HSC (0, 20, 40, and 80 μM) for 4 h was followed by stimulation with LPS. RIPA lysis buffer was used to extract the proteins. The cytoplasmic and nuclear proteins were extracted by using a nuclear and cytoplasmic protein extraction kit (Beyond time, Shanghai, China) according to the manufacturer’s instructions. The quantities of proteins were measured using a BCA protein kit (Waltham, MA, USA).

The antigen sample was treated with the specified antibody overnight at 4°C with mixing for IP experiments, and then the protein A/G magnetic beads were washed and collected using a magnetic stand before being mixed with the antigen/antibody combination. It was necessary to wash the beads three times with IP buffer. To do western blot analysis, denatured proteins were separated using 8%, 10%, or 12% SDS-PAGE gels and transferred to a PVDF membrane (Millipore, Billerica, MA, USA). After 2 h of blocking with 5% non-fat milk, the PVDF membrane was incubated with antibodies against PIP2 (1:1000), PLCγ2 (1:1000), IP3 Receptor1 (1:1000), DAG Lipaα (1:1000), TAK1 (1:1000), p-TAK1 (1:1000), PKCα (1:1000), p- PKCα (1:1000), p65 (1:1000), p-p65 (1:1000), PARP (1:1000), NLRP3 (1:1000), ASC (1:1000), IL-1β (1:1000), Caspase-1 (1:1000), Cleaved-Caspase-1 (1:1000), Cleaved-IL-1β (1:1000), and GAPDH (1:1000) for more than 12 h at 4°C and secondary antibody anti-rabbit IgG, HRP-linked Antibody and anti-mouse IgG, HRP-linked Antibody (1:5000) for 2 h at room temperature. The protein band signals were detected using Super Signal West Femto maximum sensitivity substrate (Pierce Biotechnology, USA) under the help of ChemiDoc MP Imaging System (Bio-Rad, Hercules, CA, USA).

J774A. 1 cells were pretreated for 4 h with HSC (20, 40, and 80 μM) and subsequently with LPS (1 μg/ml). Total RNA was extracted, and 1 g of RNA was tested using qRT-PCR. The PCR amplification process included SYBR green. TNF-α, IL-6, PLCγ2, PIP2, IP3 Receptor 1, DAG Lipase α and GAPDH oligonucleotide primers were as follows:

This study was conducted under the approval of the Ethics Committee for Experimental Animal Management of Guangxi University of Chinese Medicine (NO. SYXK-GUI-2019-0001). All animals were well cared for in accordance with the Local Guidelines for the Care and Use of Laboratory Animals of Guangxi University of Chinese Medicine. Healthy BALB/c mice (male, 6-8 weeks old, weighing 18-22 g) were purchased from Hunan SJA Laboratory Animal Co. Ltd (Changsha, Hunan, China) and acclimated for 3 days. All animals were housed under standard specific pathogen-free (SPF) conditions and given free access to water and a safe diet at room temperature (20-26°C, daily temperature difference ≤4°C) and relative humidity (40%-70%).

Each experiment was carried out at least three times. The statistical significance analysis was conducted using the GraphPad Prism 6.0 software (GraphPad Software, San Diego, CA). To compare groups, a one-way ANOVA with Dunnett’s multiple comparisons test was performed. Significant differences were defined as those with a P value less than 0.05 (P < 0.05).

As shown in Figure 1A , HSC is a triterpenoid saponin. Our findings showed that HSC (20, 40, and 80 μM) had no significant toxicity in RAW264.7, J774A.1, and THP-1 cells, respectively ( Figure 1B ). Epigenetic regulation is a key determinant of gene expression in sepsis (23). At the beginning of the infection, host-pathogen interactions usually lead to epigenetic changes in host cells that favor pathogen survival. At the same time, the host inflammatory response is characterized by epigenetic alterations in key regulatory genes, including TNF-a and IL-6 (24). TNF-α and IL-6 levels significantly increased in LPS-stimulated RAW 264.7 cells (P<0.001), while HSC (20, 40, and 80 μM) suppressed the level of TNF-α and IL-6 (Fig 1C and D) (F=1494, P<0.0001; F=1882, P<0.0001), which were further confirmed by J774A.1 (F=682.6, P<0.0001; F=2046, P<0.0001) and THP-1 cells (F=1756, P<0.0001; F=91.33, P<0.0001) (Fig 1E-H). Simultaneously, the qRT-PCR results indicated that HSC suppressed TNF-α and IL-6 at the gene expression level as well (F=555.6, P<0.0001; F=98.56, P<0.0001) ( Figure S1A ).

NF-κB, a transcription factor, is a heterodimer composed of p50 and p65, of which p65 is its most common form of transcriptionally activated component transcription factor (25), and has a crucial role in LPS-induced inflammation (26). The production of inflammatory mediators TNF-a and IL-6 as a result of LPS-induced nuclear translocation of NF-κB/p65 has been shown (27). The p-p65 protein expression was increased in LPS-stimulated macrophages, while pre-treatment with HSC dramatically reduced p-p65 protein expression ( Figures 2A–C ). Furthermore, LPS promoted the dissociation of NF-κB/p65 in the cytoplasm across the nuclear membrane into the nucleus in RAW264.7 cells, and HSC prevented NF-κB/p65 from translocation into the nucleus ( Figure 2D ), which was further confirmed by the immunofluorescence data ( Figure 2E ). Thus, we have concluded that HSC could reverse the LPS-induced nuclear translocation of NF-κB/p65.

NF-κB could be activated by multiple kinase signaling cascades and regulated by multiple factors (28). A previous study indicated that the protein kinase PKC could regulate the phosphorylation of p65 (29). Furthermore, Bing-C Chen et al. demonstrated that Ro 31-8220 (a PKC inhibitor) reduced the nuclear translocation of P65 that had been activated with LPS in J774A.1 cells (30), and Thu-Huyen Pham et al. also demonstrated that PKC is an upstream regulator of NF-κB (31). J. Wu et al. found that inhibition of p-TAK1 could further inhibit downstream activation of the signaling cascade, and thus NF-κB activation (32). As shown in Figures 2A–C , HSC significantly inhibited phosphorylation of TAK1 and PKCα in LPS-induced macrophages. Taken together, the results suggested that HSC inhibited NF-κB nuclear translocation in LPS-stimulated macrophages via the TAK1/PKCα pathway.

NLRP3 inflammasome plays a crucial role in inflammatory illnesses (33). To examine the effect of HSC on NLRP3 inflammasome, we investigated the NLRP3 inflammasome-associated proteins expression including NLRP3, ASC, caspase-1, and IL-1β. As shown in Figures 3A, B , LPS+ATP increased the expression of cleaved-caspase-1 and cleaved-IL-1β, which was significantly attenuated by HSC pretreatment. ELISA was used to determine the level of IL-1β in the supernatant, which is consistent with the protein expression ( Figures 3C, D ) (F=70.78, P<0.0001; F=117.8, P<0.0001). The inflammasome is a multi-protein complex that recruits pro-caspase-1 via ASC and then proceeds to cleave the cytokine precursors pro-IL-1β and pro-IL-18 into mature IL-1β and IL-18 (34). As shown in Figures 3A, B , LPS+ATP exerted no effects on the expression of ASC and caspase-1. However, immunofluorescence results showed a significant increase in ASC spots (green spots) in LPS+ATP-stimulated THP-1 cells, and the ASC spot (green spot) formation was decreased by HSC (80 μM) ( Figure 3E ). The immunofluorescence results showed that LPS+ATP-triggered J774A.1 cells activated caspase-1 expression, while the activation of caspase-1 was suppressed by HSC ( Figure 3F ).

A two-signal model for how the NLRP3 inflammasome is primed. It could be caused by microbial components or endogenous cytokines to activate the transcription factor NF-κB, resulting in the up-regulation of NLRP3 and pro-IL-1β. The activation signal comes from a variety of factors like extracellular ATP, RNA viruses, and particulate matter. Multiple molecular or the assembly of NLRP3, ASC, and caspase-1 into inflammasome to promote pro-IL-1β uncapping into IL-1β release into the extracellular (35). LPS+ATP increased the expression of NLRP3, which was significantly attenuated by HSC ( Figures 3A, B ). J774A.1 cells were transfected with an EGFP-NLRP3 plasmid. Immunofluorescence intensity showed the same results as western blotting ( Figure 3G ). Using Co-IP assays, we found that HSC significantly decreased the formation of NLRP3 and ASC complexes when LPS+ATP was added to the cells ( Figure 3H ). HSC also dramatically decreased the formation of NLRP3, ASC, caspase-1 complexes ( Figure 3I ). Thus, these results indicated that HSC could inhibit NLRP3 inflammasome activation.

Ca²⁺ has an irreplaceable regulatory role in many cell functions and is a vital signal factor in the inflammatory response (36). The Ca²⁺ release was detected by using immunofluorescent, the level of Ca²⁺ significantly increased in LPS-treated cells, while HSC suppressed the Ca²⁺ release ( Figure 4A ). As shown in Figures 4C, D , the flow cytometry analysis revealed that HSC decreased Ca²⁺ level, which was further confirmed by J774A.1 cells transfected with an EGFP-Ca²⁺ plasmid ( Figure 4B ). Taken together, the results indicated that HSC inhibited LPS-stimulated Ca²⁺ release.

PLC degrades PIP2 lipids and regulates multiple cellular events (12). Activation of PKC and release of intracellular calcium from the endoplasmic reticulum are mediated by DAG and IP3, which are produced by PLCγ2 catalyzing PIP2 (37), leading to the activation of the NF-κB and NLRP3 inflammasome pathways (16, 38). As shown in Figure 5A , LPS stimulation up-regulated PIP2, IP3, and DAG expression, while pretreatment with HSC slightly down-regulated PIP2, IP3, and DAG protein expression. Pretreatment with HSC increased PLCγ2 protein expression in RAW264.7 cells. It was further verified in J774A.1 and THP-1 cells ( Figures 5B, C ). Meanwhile, the qRT-PCR study indicated that PIP2, PLCγ2, DAG, and IP3 were obtained similar at the gene expression level in LPS-induced J774A.1 cells (F=63.25, P<0.0001; F=660.8, P<0.0001; F=129.1, P<0.0001; F=40.13, P=0.0003) ( Figure S1B ). Immunofluorescence results showed that PLCγ2 fluorescence (red) was significantly diminished under LPS-induced RAW264.7 cells, while red fluorescence was stronger in the HSC group, suggesting that HSC could promote PLCγ2 expression ( Figure 5D ). We also further examined the effect on PIP2 expression by using flow cytometry and immunofluorescence. As shown in Figure 5E , LPS-stimulated PIP2 expression in RAW264.7 cells was increased, while HSC inhibited PIP2 expression. The immunofluorescence results showed that PIP2 fluorescence (green) was significantly increased in LPS-stimulated RAW264.7 and THP-1 cells, while the green fluorescence was weaker in the HSC group ( Figure 5F ).

TAK1, a highly conserved MAPK kinase, is a key regulator in the innate immune response (39). PIP2 lipids bind directly to TAK1 at W241 and N245. TAK1 activation and the associated generation of pro-inflammatory cytokines are eliminated when PIP2’s binding affinity is impaired or the PIP2 binding site is mutated (12). As shown in Figures 6A, B , LPS-stimulated RAW264.7 and THP-1 cells increased interaction of endogenous TAK1 with endogenous PIP2, while HSC attenuated the combination between TAK1 and PIP2.

In this study, the sepsis model was established to investigate the effect of HSC on ALI. Intratracheal non-invasive instillation of LPS (15 mg/kg, i.t.) was used to successfully establish a sepsis model ( Figure 7A ). In the LPS-induced sepsis group, all mice died within 156 h. However, HSC (5, 10, and 20 mg/kg) rescued the mice’s survival rate (40%, 60%, and 70%) in 156 h ( Figure 7B ), of which the effect was similar to that of the DEX, a positive drug (70%). Furthermore, respiratory mechanics research revealed that the resistance of the lung (RL) and resistance of expiration (Re) were increased in the LPS group, while they were significantly decreased in the HSC-treated group ( Figures 7C, D ) (F=7.321, P<0.0001; F=10.21, P<0.0001). HSC significantly increased respiratory lung compliance (Cdyn) level ( Figure 7E ) (F=11.78, P<0.0001).

To validate HSC’s therapeutic effects on ALI, mice were treated with LPS by non-invasive intratracheal instillation, and the lung index, white blood cells (WBC), neutrophil, and lymphocyte were measured, along with hematoxylin and eosin (H&E) ( Figure 8A ). The findings suggested that LPS increased the lung index and caused damage and enlargement of the lung, which was considerably reduced by the treatment with HSC ( Figure 8B ) (F=61.15, P<0.0001). The excessive rise in white blood cells (WBC), neutrophils, and lymphocyte levels caused by LPS was dramatically reduced after treatment with HSC ( Figures 8C–E ) (F=5.085, P=0.0007; F=17.42, P<0.0001; F=5.098, P=0.0007). And the body weight was not affected by HSC compared to the LPS group ( Figures S2A, B ) (F=28.56, P>0.05). Additionally, H&E staining results revealed that LPS induced alveolar interstitial exudation disrupted the normal structure of alveoli, and resulted in the infiltration of a high number of inflammatory cells. HSC, as well as the DEX, ameliorated lung tissue damages ( Figure 8F ).

LPS-induced lung inflammation is a typical form of ALI with the production of inflammatory cytokines such as TNF-α, IL-1β, and IL-6 (40). To test the effect of HSC on the inflammatory response in ALI, the serum, BALF, and lung tissues were collected, and the level of inflammatory response was determined. Our results indicated that LPS increased the level of TNF-α, IL-1β, and IL-6 in serum, BALF, and lung tissues, while DEX and HSC decreased the level of TNF-α, IL-6, and IL-1β in serum ( Figures 9A–C ) (F=50.84, P<0.0001; F=34.24, P<0.0001; F=94.84, P<0.0001), TNF-α, IL-6, and IL-1β in BALF ( Figures 9D–F ) (F=37.68, P<0.0001; F=26.37, P<0.0001; F=31.71, P<0.0001), and TNF-α, IL-6, and IL-1β in lung tissues ( Figures 9G–I ) (F=36.24, P<0.0001; F=23.57, P<0.0001; F=16.07, P<0.0001). These results demonstrated that HSC had a significant anti-inflammatory effect on ALI.

A recent study shows NF-κB has a critical role in ALI formation and progression (41). LPS stimulates the NF-κB signaling pathway in lung tissue, resulting in the release of pro-inflammatory cytokines such as IL-1β, IL-6, and TNF-α in lung tissue (42). Our results showed that the phosphorylation of TAK1, PKCα, and p65 in lung tissue was significantly increased by LPS, whereas phosphorylation of TAK1, PKCα, and p65 was significantly decreased by DEX and HSC. ( Figure 10A ).

A previous study revealed that the participation of the NLRP3 inflammasome is required for the development of ALI (43). When the NLRP3 inflammasome was activated, cleaved-caspase-1 and cleaved-IL-1β were over-expressed, leading to neutrophil recruitment, which in turn promotes the development of ALI (44). As shown in Figure 10B , LPS increased the expression of NLRP3, cleaved-caspase-1, and cleaved-IL-1β, all of which were reversed by treatment with HSC and DEX. The expression of PIP2, DAG, and IP3 proteins was up-regulated and PLCγ2 protein expression was down-regulated by LPS stimulated, while DEX and HSC significantly reversed the expression of these proteins ( Figures 10C, D ).