The special reagents included emodin and sodium taurocholate (Solarbio Science Technology, Beijing, China), rat CIRP enzyme-linked immunosorbent assay (ELISA) kit (cat#OM546910), anti-ASC (cat#OM204428), anti-CXCL1 (cat#OM248569, Omnimabs, Poway, CA, United States), amylase and IL-1β ELISA kits (Shanghai Lengton Biotech, Shanghai, China), antibodies against Ly6G (cat#GTX40912, Genetex, Dallas, TX, United States), CIRP (cat#A6080), p-P65 (cat#AP0123), P65 (cat#A16271), IKBα (cat#A16929), p-IKBα (cat#AP0707), NLRP3 (cat#A14223), Caspase-1 (cat#A0964), GSDMD (cat#A18281), IL-1β (cat#A11369), β-actin (cat#AC026 Abclonal Biotech, Wuhan, China), TAK242 and recombinant rat IL-1β (Absin Bioscience, Shanghai, China), HRP-conjugated goat anti-rabbit IgG (H + L) (cat#31460, Thermo, Waltham, MA, United States), mouse anti-F4/80 antibody (cat# SC-52664, Santa Cruz Biotechnology, Santa, Cruz, CA, United States), rabbit anti-glucagon antibody (cat#GB11097), mouse anti-insulin antibody (cat#GB13121), Cy3-conjugated goat anti-rabbit IgG (H + L) (cat#GB21303), Cy5-conjugated goat anti-mouse IgG (H + L) (cat#GB27301), FITC-conjugated goat anti-mouse IgG (H + L) (cat#GB22301, Servicebio Technology, Wuhan, China). C23 and recombinant rat CIRP (Genscript Biotech, Nanjing, China). Others were RNA extraction reagent, RNAex Pro Reagent (Accurate Biotechnology, Changsha, China), the reverse transcription kit and quantitative real-time PCR kit (Vazyme Biotech, Nanjing, China), CIRP, NLRP3, CXCL1, and IL-1β primers (Synbio Technologies, Suzhou, China).

All experiments were performed in accordance with the experimental protocol approved by the Committee for Research and Animal Ethics of Dalian Medical University (Dalian, China). Male Sprague-Dawley rats, 180–220 g, were obtained from the Experimental Animal Center of Dalian Medical University. The rats were housed in a specific pathogen-free room with consistent temperature of 20–22°C and a cycle of 12 h light-dark and allowed free access to standard rodent chow and water.

The rat model of SAP-ALI was established as descripted previously (Xu et al., 2020). Briefly, 40 rats were randomly divided into the groups: the Control (CON), SAP (SAP), C23 (C23), and emodin (EMO) groups, respectively, (n = 10 per group). The experimental rats were retrograde-infused with 5.0% sodium taurocholate (0.1 ml/100 g body weight, 0.1 ml/min) into their biliopancreatic duct while the CON group of rats received the equivalent volume of 0.9% sterile saline. 2 h later, the C23 and EMO groups of rats were treated with C23 (8 mg/kg body weight, intravenous injection) and emodin (40 mg/kg body weight, by gavage) (Xu et al., 2021), respectively. The EMO group of rats were repeated the same dose of emodin at 12 h post-operation. 24 h after the injection with sodium taurocholate, all rats were anesthetized with 1% pentobarbital sodium (4 mg/100 g bodyweight, Merck KGaA, Darmstadt, Germany), and their arterial blood samples were collected from their abdominal aorta. Their pancreatic and lung tissues were dissected and one part of the tissues were frozen in liquid nitrogen, and then stored at −80°C. The remaining pancreatic and lung tissues were fixed in 10% formalin and embedded in paraffin. Some arterial blood samples were used for arterial blood gas analysis, and the remaining blood samples were centrifuged to prepare serum samples.

The levels of serum CIRP, amylase and IL-1β in individual rats were analyzed by ELISA using specific kits, according to the manufacturer’s instructions. The blood gas analysis was performed in an automatic blood gas analyzer (RapidPoint 500, Siemens, Berlin and Munich, Germany).

The paraffin-embedded pancreatic and lung tissue sections (5 μm) were routine-stained with hematoxylin and eosin (HE). The pathological changes in each section was scored as 0 to 4 for the degree of acinar necrosis, inflammation, hemorrhage, and edema in the pancreas (Rongione et al., 1997), and as 0–3 for the degrees of edema, leukocyte infiltration, and hemorrhage in the lung in a blinded manner (Osman et al., 1998).

The levels of CIRP and Ly6G expression in pancreatic and lung tissue sections were determined by immunohistochemistry using antibodies against CIRP (1:600) and ly6G (1:500) as described previously (Ramos-Vara, 2017).

Rat alveolar macrophage NR8383 cells, were obtained from the Zhong Qiao Xin Zhou Biotechnology (Shanghai, China) and cultured in Ham’s F-12K medium (Thermo) supplemented with 15% fetal bovine serum (FBS) in a humidified atmosphere of 5% CO₂ at 37°C.

To determine the effect of CIRP, NR8383 cells were treated with vehicle as the Control or 1.5 μg/ml recombinant rat CIRP for 6 h as the CIRP group. NR8383 cells were pre-treated with TAK-242 (500 µM) for 24 h (Wu et al., 2018) and treated with 1.5 μg/ml CIRP for 6 h. In addition, NR8383 cells were pre-treated with 300 ng/ml C23 for 1 h (Zhang et al., 2018) and treated with 1.5 μg/ml CIRP for 6 h as the CIRP + C23 group. Moreover, NR8383 cells were treated with 1.5 μg/ml CIRP and 20 µM emodin for 6 h as the CIRP + EMO group (Xiang et al., 2017a). To examine the role of IL-1β in the CIRP-increased CXCL1 expression in AMs, NR8383 cells were treated with vehicle as the Control or 1.5 μg/ml CIRP for 6 h as the CIRP group. Furthermore, NR8383 cells were transfected with control siRNA or IL-1β-specific siRNA (Table 1) for 48 h and treated with 1.5 μg/ml CIRP for 6 h as the NC or Si-IL-1β group, respectively. To determine the effect of IL-1β on CXCL1 expression, NR8383 cells were treated with various concentrations (0, 1 ng/ml, 5 ng/ml, 10 ng/ml) of recombinant rat IL-1β for 24 h or treated with 10 ng/ml IL-1β for varying time periods (0, 6, 12, 24 h), respectively.

NR8383 cells were transfected with control siRNA or IL-1β specific siRNA (Keygen, Nanjing, China, Table 1) using the Lipofectamine TM 2000 (Invitrogen, Carlsbad, CA, United States), according to the manufacturer’s instruction.

Total RNAs were extracted from rat pancreas, lung samples and the different groups of NR8383 cells using the RNAex Pro Reagent (AG), and the resulting RNA samples (1 µg each) were transcribed reversely into cDNA using the HiScript® II Reverse transcriptase (Vazyme). The relative levels of target gene mRNA transcripts to the control GAPDH were quantified by qPCR using the ChamQ universal SYBR® qPCR Master Mix (Vazyme) and specific primers (Table 2). The data were analyzed by the 2^−ΔΔCT method.

Fresh pancreatic and lung tissues from individual rats were homogenized and lyzed in RIPA lysis buffer (Keygen). Similarly, the different groups of NR8383 cells were lyzed in RIPA lysis buffer. After quantification of protein concentrations, individual lysate samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on 10–15% gels (Solarbio, Beijing, China.) and transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% fat-free dry milk in TBST for 1 h and incubated with primary antibodies, including anti-CIRP (1:1000), anti-NLRP3 (1:1000), anti-ASC (1:300), anti-caspase-1 (1:1000), anti-IL-1β (1:1000), anti-GSDMD (1:1000), anti-p-P65 (1:1000), anti-P65 (1:1000), anti-IKBα (1:1000), anti-p-IKBα (1:1000), anti-CXCL1 (1:500), and anti-β-actin (1:20,000) at 4°C overnight. After being washed, the bound antibodies were detected with goat anti-rabbit IgG (H + L) secondary antibody (1:5000) and visualized using the enhanced chemiluminescent reagent (Keygen). The bands were imaged using the Tanon 5200 imaging system. The relative levels of the target protein to the control β-actin were quantified by densitometric analysis using the Image Pro Plus 5.1 software.

The distributions of insulin, glucagon and CIRP expression in the pancreatic tissue sections were characterized using the tyramide signal amplification (TSA) method (Francisco-Cruz et al., 2020). Briefly, the pancreatic tissue sections were incubated with rabbit anti-glucagon antibody (1:10,000) overnight at 4°C, and the bound antibodies were detected HRP-conjugated goat anti-rabbit IgG (H + L) (1:500). After being washed, the sections were incubated with FITC-labelled TSA reagents. Subsequently, the sections were probed with mouse anti-insulin antibody (1:300) and rabbit anti-CIRP antibody (1:100) at 4°C overnight. The sections were further incubated with Cy3-conjugated goat anti-rabbit IgG (H + L) (1:300) and Cy5-conjugated goat anti-mouse IgG (H + L) (1:300). The fluorescent signals were scanned using the Pannoramic DESK and the images were captured using the software provided (3DHISTECH, Budapest, Hungary).

In addition, the distribution of F4/80 and NLRP3 or F4/80 and CXCL1 was analyzed by immunofluorescence using two sets of antibodies, as described previously (Mori and Cardiff, 2016). Briefly, the lung tissue sections were incubated with mouse anti-F4/80 antibody (1:100) and rabbit anti-NLRP3 (1:300) or rabbit anti-CXCL1 antibody (1:200) overnight at 4°C. The bound antibodies were probed with FITC-conjugated goat anti-mouse IgG (H + L) (1:200) and Cy3-conjugated goat anti-rabbit IgG (H + L) (1:300). The fluorescent signals were captured using an Olympus BX63 fluorescence microscope (Olympus, Tokyo, Japan) at 200× magnification.

To test the effect of emodin or C23 on the development of SAP-ALI, SD rats were retrograde-infused with sodium taurocholate into their biliopancreatic duct to induce SAP-ALI. 2 h later, the rats were randomized and treated with vehicle (SAP group), emodin or C23. 24 h post induction, the rats were euthanized and their pancreatic and lung tissues were examined by pathology (Figure 1A). While the control rats displayed healthy pancreas and lung tissue morphology, the SAP group of rats exhibited severe edema, inflammation, hemorrhage, and acinar necrosis in their pancreases and edema, leukocyte infiltration and hemorrhage in their lungs, which were less remarkable in the EMO or C23 group of rats. Semi-quantification indicated the histological scores for the pancreas and lung sections in the SAP group were significantly higher than that in the CON group, but were significantly reduced in the EMO or C23 group of rats (P < 0.001 for both, Figures 1B,C). A similar pattern of serum amylase and IL-1β levels was detected among the groups of rats, indicating that treatment with emodin or C23 mitigated the SAP and inflammation in rats (Figures 1D,E). Analyses of arterial gas revealed that compared with the CON group, the levels of PaO2 were significantly reduced while the levels of PaCO2 were elevated in the SAP group of rats, which were significantly mitigated in the EMO or C23 group of rats (P < 0.001, Figures 1F,G). Such data indicated that treatment with emodin or C23 significantly mitigated the severity of SAP-ALI in rats.

CIRP is an inflammatory mediator and can bind to the TLR4 to induce pro-inflammatory cytokine production. To understand the molecular mechanisms underlying the action of emodin, we characterized the expression of CIRP in the different groups of rats. As shown in Figure 2A, while there was a little CIPR expression in the pancreatic and lung tissues of the control rats, the levels of CIRP expression significantly increased in the pancreatic islets and lungs of the SAP group of rats. In contrast, the levels of CIRP expression in the pancreatic islets and lung tissues were obviously reduced in the C23 or EMO group of rats, particularly in the EMO group. Furthermore, the levels of serum CIRP in the C23 or EMO group of rats were significantly lower than that in the SAP, but remained higher than that in the CON group of rats (P < 0.001 for all, Figure 2B). A similar pattern of CIRP mRNA transcripts and protein expression was detected in the pancreatic and lung tissues in the different groups of rats (Figures 2C–F). Finally, immunofluorescent analysis revealed that CIRP expression was induced in β and α cells in the pancreatic islets of the SAP group of rats, which was obviously reduced in the C23 group, particularly in the islet β cells and further decreased in the islets of the EMO group of rats (Figure 2G). Collectively, the results demonstrated that treatment with emodin, like C23, significantly mitigated the SAP-up-regulated CIRP expression in the pancreatic islets and lung tissues of rats.

To further understand the action of emodin, we investigated the effect of emodin on the NF-κB signaling and NLRP3 inflammasome formation in the lungs of the different groups of rats. In comparison with the CON group, significantly higher levels of NF-κBp65 and IKBα phosphorylation were detected in the lung of the SAP group of rats, which were significantly reduced in the lungs of the EMO or C23 group of rats, indicating that treatment with emodin or C23 mitigated the SAP-activated NF-κB signaling in the lungs of rats (Figure 3A). A similar pattern of NLRP3 and IL-1β mRNA transcripts and protein expression as well as polymer, dimer and monomer ASC, caspase-1, GSDMD-N, pro-IL-1β, cleaved IL-1β (p17) expression was observed in the lungs of different groups of rats (Figures 3B–D). Immunofluorescent analysis using anti-NLRP3 and anti-F4/80, a marker of resident macrophages (Miyabe et al., 2019) indicated that the NLRP3 expression was mainly overlapped with F4/80 in macrophages in the lungs of the SAP group of rats, which was significantly reduced in the lungs of the EMO or C23 group of rats (Figure 3E). Together, such data indicated that treatment with emodin or C23 significantly mitigated the SAP-activated NF-κB signaling and NLRP3 inflammasome formation in the lungs of rats.

CXCL1 is critical for the migration of neutrophils and their infiltration in the lung contributes to the development of ALI (Bhatia et al., 1998; De Filippo et al., 2013; Inoue et al., 1995; Keck et al., 2002). To understand the action of emodin in inhibiting the SAP-ALI, we tested the levels of CXCL1 expression in the lungs of different groups of rats. We found that the levels of CXCL1 mRNA transcripts and protein expression in the lungs of the SAP group were significantly higher than that in the CON group of rats, but were significantly reduced in the EMO or C23 group of rats (Figures 4A,B). Immunofluorescent analysis indicated that the CXCL1 expression was co-localized with F4/80 in the lung tissue sections of the SAP group of rats, but little was detected in the lungs of other groups of rats (Figure 4C). Such data indicated that CXCL1 was mainly expressed by lung resident macrophages during the process of SAP-ALI. Immunohistochemistry analysis using antibodies against Ly6G, a marker of neutrophils (Feng et al., 2017) revealed that there were many Ly6G + neutrophils in the lung and alveolar spaces of the SAP group of rats, which was obviously reduced in the C23 and EMO groups of rats (Figure 4D). Hence, treatment with emodin or C23 significantly reduced the SAP-up-regulated CXCL1 expression and neutrophil infiltration in the lungs of rats.

To explore the effect of emodin on the CIRP-activated NF-κB signaling, inflammasome formation and pyroptosis in macrophages, NR8383 cells were stimulated with recombinant rat CIRP for 6 h. We found that CIRP stimulation significantly increased the levels of IκBα and NF-κBp65 phosphorylation, a hall mark of the activation of the NF-κB signaling in NR8383 cells, which were nearly completely abrogated by TAK242 treatment and significantly mitigated by C23 or emodin treatment (Figure 5A). A similar pattern of NLRP3 and IL-1β mRNA transcripts and NLRP3, ASC, caspase-1 (p20), GSDMD-N, and IL-1β protein expression was observed in the different groups of NR8383 cells (Figures 5B–D). Flow cytometry analysis revealed that CIRP triggered a high frequency of caspase-1+PI + NR8383 cells, a hallmark of cell pyroptosis, which was also completely abrogated by treatment with TAK242, and significantly reduced by treatment with C23 or emodin (Figures 5E,F). Thus, treatment with emodin significantly reduced the CIRP-activated NF-κB signaling, NLRP3 inflammasome formation, and pyroptosis in NR8383 cells in vitro.

Finally, we tested whether CIRP could modulate CXCL1 expression and how treatment with emodin, or C23 could alter its effect in NR8383 cells. First, we found that treatment with CIRP significantly up-regulated CXCL1 mRNA transcripts and protein expression in NR8383 cells, which was abrogated by TAM242 treatment, and significantly mitigated by C23 or emodin treatment (Figures 6A,B). Because CIRP induced the NLRP3 inflammasome formation and CXCL1 and IL-1β expression in NR8383 cells, we tested whether IL-1β was crucial for CIRP to induce CXCL1 expression in NR8383 cells. We found that CIRP only induced very lower levels of CXCL1 expression in IL-1β silencing NR8383 cells, relative to that in the control siRNA-transfected NR8383 cells (Figures 6C,D). Furthermore, treatment with different concentrations of IL-1β for varying periods revealed that IL-1β induced CXCL1 expression in NR8383 cells in a dose- and time-dependent manner (Figures 6E–H). Therefore, treatment with emodin mitigates the CIRP-induced IL-1β-dependent CXCL1 expression in NR8383 cells in vitro.