Dioscin Improves Pyroptosis in LPS-Induced Mice Mastitis by Activating AMPK/Nrf2 and Inhibiting the NF-κB Signaling Pathway

Dioscin, a natural steroid saponin, has been shown to have anti-inflammatory effects, but its protective mechanism against mastitis is still unknown. NLRP3 inflammasome and pyroptosis play important roles in the pathogenesis of many inflammatory diseases, including mastitis. The purpose of this study was to explore the effect of dioscin on lipopolysaccharide- (LPS-) induced mastitis in vivo and in vitro and its mechanism of action. In vivo experiments, dioscin can reduce the inflammatory lesions and neutrophil motility in mammary tissue. Moreover, dioscin also can reduce the production of proinflammatory factors such as interleukin-1 beta (IL-1β) and inhibit the activation of NLRP3 inflammasome in LPS-induced mice mastitis. In vitro experiments, the results showed that dioscin inhibited the inflammatory response and the activation of NLRP3 inflammasome, but the survival rate of mouse mammary epithelial cells (mMECs) induced by LPS+ATP is increased. Subsequently, the experiment convinces that dioscin can reduce LPS+ATP-induced mMEC pyroptosis by adding Ac-DEVD-CHO (a caspase-3 inhibitor). Further mechanistic studies demonstrate that dioscin can activate AMPK/Nrf2 to inhibit NLRP3/GSDMD-induced mMEC pyroptosis. In summary, this paper reveals a novel function of dioscin on mMEC pyroptosis and provides a new potential therapy of dioscin for the treatment and prevention of mastitis.


Introduction
Cow mastitis is a significant disease that affects the dairy products in production and quality [1]. A lot of factors can cause mastitis, especially the pathogen infections and what we usually found is Gram-negative bacteria causes the mastitis in cow [2,3]. Therefore, lipopolysaccharide (LPS), as the main pathogenic component of Gram-negative bacteria, has been used in a variety of mastitis animal models [4]. Previous studies have shown that mammary epithelial cells have immune activity, which can activate NOD-like receptor protein 3 (NLRP3) inflammasome [5] and nuclear factor kappa B (NF-κB) signaling pathways [6] during mastitis, and that can release proinflammatory cytokines such as interleukin-1 (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor alpha (TNF-α). Thereby, the effects of expanding the inflam-matory response and promoting the exacerbation of inflammation become obvious [7,8]. Recent studies have shown that pyroptosis induced by the activation of NLRP3 inflammasome is a significant cause of proinflammatory cytokine release [9,10]. Therefore, inflammasomes and pyroptosis may play key roles in the inflammatory process of mastitis.
Pyroptosis has long been mistaken for a type of apoptosis caused by an identified pathogen [11]. However, recent studies have demonstrated that pyroptosis as a form of inflammatory regulatory cell death is distinct from apoptosis [12,13]. There is an increasing evidence that inflammasomes can induce pyroptosis in a variety of cells, and because of this, many infectious diseases will be caused. This process plays an important role in a wide range of infectious diseases [14,15]. Inflammasomes, including NLRP3, are a group of protein complexes and also act as pathogen recognition receptors to recognize danger signals and promote inflammation [16,17]. NLRP3, the most characteristic inflammasomes, usually is activated by a variety of "danger signals," such as ATP and reactive oxygen species (ROS) that recruit ASC and promote caspase-1 activation. The activated caspase-1 cleaves Gasdermin D (GSDMD) protein to the N-and C-terminals and leads to enzymatic hydrolysis of inflammatory factors, for instance, IL-1β. Then, the Nterminal fragment of GSDMD results in membrane perforation and the release of proinflammatory cytokines and lactate dehydrogenase (LDH) [10,[18][19][20]. However, whether the activation of NLRP3 inflammasome in mammary epithelial cells induces pyroptosis has not been reported.
At present, antibiotics are generally used in the treatment of cow mastitis, and the former practices of combining antibiotics with anti-inflammatory drugs are good examples to heal with the mastitis. Therefore, the selection of antiinflammatory drugs still is a main therapy in curation of mastitis [21][22][23]. Dioscin, extracted from the tubers of Dioscorea japonica, is a natural compound. As for pharmacological effects, such as antioxidant, anti-inflammatory and antiapoptosis are also its characteristics [24][25][26][27][28]. Previous studies provide many documents in which dioscin, activated nuclear factor erythroid 2-related factor 2 (Nrf2), has an antiinflammatory effect and is an inhibitor of NLRP3 inflammasome [25,29]. However, whether dioscin prevents or mitigates the pathogenesis of mastitis through inhibiting pyroptosis caused by inflammasomes of mammary epithelial cells is still unknown. Therefore, in this study, the protective role and mechanism of dioscin in mice model of mastitis induced by LPS are investigated.

Material and Methods
2.1. Animal. BABL-/-c (8/9 weeks old) mice were purchased from the Experimental Animal Center of Bethune Medical College of Jilin University and were distributed in a small cage with two males and two females at 22°C-23°C, 12 h light and dark cycle, and ad libitum access to food and water. When the female mouse was pregnant, it was transferred to a separate small cage. This study is approved by the development of Animal Care and Use Committee of Jilin University, and all animal care and experimental procedures are under the supervision of this committee (approved on 27 February 2015, Protocol No. 2015047).

Mastitis
Model. The female mice who were delivering after 5-7 days were randomly assigned to 5 groups. The groups are as follows: control group, dioscin group, LPSinduced mastitis group, and LPS-induced mastitis groups treated with dioscin or dexamethasone. Except for the NT group and the dioscin group, the mouse model of mastitis was induced by LPS in other groups. Dioscin (45 mg/kg/day) was administered orally, and dexamethasone (5 mg/kg/day) was injected intraperitoneally, starting from 3 days before LPS-induced mastitis. Before the mastitis was induced for 3 h, the female mice with their pups were separated. Mice were anesthetized by pentobarbital, and their fourth pair of nipples was disinfected with 75% alcohol. The 1 mm of the nipples was cut off with sterilizing scissors. Then, the mice were injected with 10 μg LPS (0.2 mg/ml) into their nipples with 32G needle. Mastitis was induced 24 h later, and all mice were sacrificed, and samples of mammary tissues were taken to detect.
2.3. Cell Culture and Treatment. mMECs were purchased from the American Type Culture Collection (ATCC, ATCC® CRL-3063™), cultured in DMEM (Gibco, Grand Island, NY 14072, USA) with 10% FBS (Clark Bioscience, Richmond, VA, USA), and maintained at 37°C in a humidified chamber of 5% CO 2 . The mMECs' density reached 80%-90% in the cell culture flask. The cells were seeded into 60 mm × 15 mm cell culture dishes, 6-well plates, or 96-well plates. When the mMECs' density reached 70%, they were preprotected with dioscin for 2 h, then stimulated with LPS (1 μg/ml) for different time (1 h, 6 h, and 24 h), 30 min or 1 h before LPS stimulation was completed, plus ATP (5 mM) stimulation. After stimulation, cell samples were collected for subsequent experiments.
2.4. Histopathological examination. The mammary tissues were fixed in 4% formaldehyde for 24 hours, then placed in different concentrations of alcohol and xylene in turn, fixed with paraffin, and cut into 5-micron thick sections. They were stained with eosin and hematoxylin and we observed histopathological changes under a light microscope. Three skilled pathological section analysts were invited to score the paraffin sections. Five fields of view were selected for each section to observe the degree of neutrophil aggregation, acinar integrity, and changes in acinar wall thickness. The scoring criteria are varying from 0 points to 4 points in no injury, mild injury, moderate injury, severe injury, and extreme injury, respectively.
2.5. Immunohistochemistry. For immunohistochemical detection of GSDMD on paraffin sections of mammary tissue, 5 μm thick sections of mammary tissue fixed with paraffin were cut. Dewaxed and hydrated, the sections were digested with PBS containing 0.01% trypsin (Sigma) for 5 minutes, and then, the sections were heated in a 10 mmol/L citrate buffer (pH 6.0) to 90°C and stopped heating and cooled to 70°C, repeated 2-3 times. And cooling the section waits for down to the normal temperature. After treatment, sections were treated with 3% catalase to block endogenous peroxidase activity and subsequently blocked with 0.5% blocking agent for 30 minutes to reduce the number of nonspecific reactions. Sections were incubated with antibodies against GSDMD (diluted 1 : 500 in buffer) (Santa Cruz, USA) at 4°C overnight. Subsequently, wash 3 times with PBS for 5 minutes, and a biotinylated linking antibody solution (Biozol, Eching, Germany) was applied (100 μL/section) for 20 min at 37°C. Sections should be kept in incubation with antibodies against GSDMD (diluted 1 : 500 in buffer) (Santa Cruz, USA) at 4°C for one night. After washing with PBS again, slides were stained with DAB (IBL, Germany) counterstained with hematoxylin. The negative control used affinity-purified preimmune IgG instead of the primary antibody.     Oxidative Medicine and Cellular Longevity the LDH kit (Beyotime Institute of Biotechnology, China) was used for the test.
2.14. ROS Detection. The ROS kit (Beyotime Institute of Biotechnology, China) was used to detect the intracellular reactive oxygen content.

Statistical Analysis.
The experimental data were analyzed by GraphPad Prism7 (Manufacturer, La Jolla, CA, USA). Differences between groups were analyzed by oneway analysis of variance (ANOVA), and pairwise comparisons were performed by LSD. p values of less than 0.05 were considered statistically significant ( # significant compared with the NT group; * p < 0:05; * * p < 0:01).

Dioscin Ameliorates LPS-Induced Mastitis in Mice.
In order to investigate the protective effect of dioscin on mastitis mice, the following tests were experimented on the mammary tissue of mice in different treatment groups. The result showed that dioscin administration reduces swelling and congestion of mammary tissue induced by LPS ( Figure 1(a)). HE staining showed that LPS stimulation caused hyperemic edema in the acinar cavity, and acinar cavity was infiltrated with a large amount of polymorphonuclear leukocytes; the addition of dioscin significantly reduced the histopathological changes induced by LPS (Figures 1(b) and 1(d)). The MPO test further proved that the addition of dioscin could significantly reduce the inflammatory cell infiltration induced by LPS (Figure 1(c)). In addition, the results also witnessed that LPS increased the expression and secretion of proinflammatory cytokines such as IL-1β (Figures 1(b) and 1(h)), IL-6 (Figures 1(f) and 1(i)), and TNF-α (Figures 1(g) and 1(j)), but the pretreated of dioscin significantly reduced them.

Dioscin Ameliorates Inflammation and Death of mMECs
Induced by LPS+ATP. To investigate the anti-inflammatory mechanism of dioscin, the experiments were designed to detect the effect of dioscin on LPS+ATP-stimulated mMECs.
The study also has been found that less than 800 ng/ml of dioscin has no cytotoxicity on mMECs (supplementary

11
Oxidative Medicine and Cellular Longevity S1). Stimulation of mMECs with LPS+ATP significantly increased gene expression of the proinflammatory mediators IL-1β, IL-6, TNF-α, INOS, and COX-2 (Figures 2(a)-2(e)) and protein expression of INOS and COX-2 (Figures 2(f) and 2(h)). As expected, dioscin significantly inhibited these responses in a dose-dependent manner. In addition, pretreatment with dioscin significantly reduced LPS+ATP-induced cell death (Figures 2(i) and 2(j)) and LDH release (Figure 2(k)). These results suggest that dioscin has a positive effect on cell viability of mMECs induced by LPS+ATP.

Dioscin Reduces LPS+ATP-Induced Activation of NF-κB
Signaling Pathway and NLRP3 Inflammasome in mMECs. NF-κB signaling pathway and NLRP3 inflammasome play an important role in regulating the production of proinflammatory cytokines in the course of inflammation. The NF-κB signaling pathway and NLRP3 inflammasome in LPS+ATP-induced mMECs were investigated. The results showed that dioscin significantly ameliorated the phosphorylation of NF-κB p65 induced by LPS+ATP in mMECs (Figure 3(b)). Previous studies have shown that the activation of NF-κB can also lead to nuclear translocation of NF-κB p65 subunit [32]. The experimental results from the studies of this paper did show that LPS+ATP induced nuclear translocation of NF-κB p65, and pretreatment of dioscin significantly inhibited nuclear translocation of NF-κB p65 (Figure 3(a)). Detection of inflammasome found that LPS+ATP significantly increased the expression of NLRP3, ASC, and caspase-1 of mMECs at the protein level (Figures 3(e)-3(g)), and NLRP3 inflammasome gene level test results were consistent with protein levels (Supplementary Fig. S2a-d). To our surprised, pretreatment dioscin also reduced the protein levels of cleaved caspase-1 and cleaved caspase-3 (Figures 3(g) and  3(h)). These results indicated that dioscin not only inhibited  The cell LDH release was detected by LDH kit. (i) The production of IL-1β was detected by ELISA. * p < 0:05 and * * p < 0:01 compared to the LPS+ATP group; & p < 0:05 and && p < 0:01 compared to the dioscin group. 12 Oxidative Medicine and Cellular Longevity  14 Oxidative Medicine and Cellular Longevity the expression of NLRP3 inflammasome but also hindered its activation. Besides, LPS caused an increase in GSDMD-N, a key protein of pyroptosis, in LPS+ATP-induced mMECs at protein and gene levels, and dioscin significantly improved this phenomenon (Figure 3(c), Supplementary Fig. S2e).

Dioscin Reduces Pyroptosis of mMECs Induced by LPS+ATP.
To investigate the unique effect of dioscin on apoptosis or pyroptosis, Ac-DEVD-CHO was used, a specific caspase-3 inhibitor, to inhibit caspase-3-induced apoptosis.
The results showed that after adding Ac-DEVD-CHO, the activation of caspase-3 stimulated by LPS+ATP was significantly inhibited (Figures 4(b), 4(c), and 4(e)), but caspase-1 had no effect (Figures 4(a), 4(c), and 4(d)). Dioscin still significantly reduced the activation of caspase-1 induced by LPS+ATP after using Ac-DEVD-CHO (Figures 4(c) and 4(d)). Flow cytometry was chosen to observe how dioscin affected cell death. Interestingly, the results were consistent with caspase-1. Dioscin significantly reduced LPS+ATPinduced cell death, while caspase-3 had no effect on cell death (Figures 4(f) and 4(g)). These results led to speculate that dioscin ameliorated pyroptosis, not apoptosis. IL-1 secretion and LDH release are important indicators of pyroptosis, and the results showed that the inhibition of IL-1 secretion and LDH release by dioscin has no dependence on caspase-3 (Figures 4(h)-4(j)). Therefore, the hypothesis is that dioscin can reduce the mMEC pyroptosis induced by LPS+ATP.

Dioscin Activates AMPK-Nrf2 in LPS+ATP-Induced mMECs.
Previous studies have shown that activating Nrf2 is a significant factor in the activation of NLRP3 inflammasome and phosphorylation of NF-κB. Dioscin has been reported to be an antioxidant [25,33], so the Nrf2 signaling pathway is the main work to do. The dioscin treated mMECs

17
Oxidative Medicine and Cellular Longevity for different time periods (0, 0.5, 1, 3, 6, and 12 h), and the results showed that dioscin increased the protein expression of total Nrf2 and nuclear Nrf2 and phosphorylation of AMPK and HO-1 (Figures 5(a) and 5(c)-5(f)), decreased the protein of Nrf2 in the cytoplasm, and reduced reactive oxygen (Figures 5(a), 5(b), and 5(g)). Previous studies suggested that AMPK is an upstream of Nrf2 [34]. The results showed dioscin also promoted AMPK phosphorylation. Then compound C and ML385 (AMPK phosphorylation inhibitor and Nrf2 inhibitor) were added in the experiments, and what was found was that dioscin can reduce ROS production by increasing AMPK phosphorylation to increase Nrf2 activation (Supplementary Fig. 3s). Further research showed that dioscin could provoke Nrf2 and HO-1 to express after LPS+ATP stimulated mMECs (Figures 5(i)-5(k)). As predicted, dioscin enhanced nuclear translocation of the Nrf2 protein and caused significant activation of Nrf2 ( Figure 5(h)).
To our surprise, we subsequently found that ML385 significantly reduced the inhibitory effect of dioscin on GSDMD-N (Figures 6(a) and 6(d)). LDH and flow cytometry experiments showed that ML385 also significantly reduced the improvement effect of dioscin on pyroptosis of mMECs (Figures 6(k) and 6(m)). The secretion of IL-1β is consistent with the above results (Figure 6(n)). In summary, diocin reduces LPS-induced pyroptosis of mMECs by activating Nrf2.

Dioscin Inhibits Activation of NF-κB Signaling Pathway and NLRP3 Inflammasome in LPS-Induced Mice Mastitis.
In vivo, we further confirmed the inhibitory effect of dioscin on NF-κB signaling pathway and NLRP3 inflammasome. The result showed that feeding dioscin significantly reduced LPS-induced NF-κB p65 phosphorylation in mastitis mice (Figures 7(a) and 7(b)). Subsequently, we tested the NLRP3 inflammasome in mice mammary tissue and found that NLRP3, ASC, and caspase-1 were significantly increased at the gene level and protein level under LPS stimulation (Figures 7(a) and 7(b)-7(h)). The trend is the same as in vitro results; dioscin markedly reduced NLRP3 inflammasome expression and production. As predicted, dioscin treatment also reduced the protein levels of cleaved caspase-1 and cleaved caspase-3 (Figures 7(a), 7(h), and 7(i)). Dioscin also inhibited LPS-induced GSDMD expression of gene and GSDMD-N protein in LPS-induced mice (Figures 7(k) and 7(l)). Immunohistochemistry further confirmed our results of GSDMD in mammary tissue of mice (Figure 7(j)).

Dioscin promotes the activation of Nrf2 in LPS-induced
Mastitis Mice. In vitro, dioscin has been shown to activate AMPK and Nrf2. For further expounding the mechanism of dioscin, the effect of dioscin on Nrf2 was also detected in LPS-induced mastitis mice. As was our anticipation, phosphorylation of AMPK and protein expressions of HO-1 and Nrf2 increased in mammary tissues of mice after intragastric administration with dioscin ( Figure 8).

Discussion
Cow mastitis is a disease that will cause a great damage to dairy industry, and the therapy always relies on antibiotics mostly. However, the use of antibiotics is a promoter to raise

19
Oxidative Medicine and Cellular Longevity the drug resistance of mastitis. Because of the drug resistance and high cost of antibiotics, the researchers are trying to select an appropriate combination of antibiotics and antiinflammatory drugs to cure mastitis with a low cost but more effective. Therefore, that is why dioscin is into the eyes of researcher. This natural compound, dioscin, through the experiments, has many pharmacological effects including anti-inflammatory and also has virtues of high efficiency and low price [35]. To detect the therapeutic effect and mechanism of dioscin on mastitis, this experiment was designed in vivo and in vitro. In summary, in vivo, the classic model of LPS-induced mouse mastitis was introduced to the experiments [36]. In vitro, the application of LPS and ATP is to stimulate mMECs and to construct an in vitro mastitis model. Based on existent reports, two main interdependent processes will aggravate the mastitis inflammation: one is recruitment of inflammatory cells and the other is upregulation of proinflammatory cytokines [37]. In vivo, these experiments demonstrated that the dioscin reduces the mastitis symptoms induced by LPS in mice. HE staining and MPO 21 Oxidative Medicine and Cellular Longevity of mouse mammary tissue intended that dioscin significantly ameliorates the pathological and histological changes of mammary and neutrophil infiltration caused by LPS. Previous studies have proved that reducing the production of proinflammatory cytokines, such as IL-1β, IL-6, and TNF-α, can reduce the effect of mastitis [38,39]. Consistent with these studies, the dioscin obviously inhibited the production of proinflammatory cytokines IL-1β, IL-6, and TNF-α in vivo and in vitro.
After confirming the anti-inflammatory effect of dioscin in mastitis mice, the anti-inflammatory mechanism of dioscin was explored in the next step. During the occurrence of various inflammations, the NF-κB pathway and NLRP3 inflammasome affect the subsequent development of inflammation through the regulation of inflammatory factors [40,41]. The former way, NF-κB pathway, relies on the regulation of expression on proinflammatory cytokines. To illustrate intensively, NF-κB, a transcription factor mainly composed of p50/p65 heterodimer, regulates the expression of proinflammatory cytokines [42]. Activation of NF-κB by LPS leads to the phosphorylation and nuclear translocation of NF-κB p65, which not only increases the expression of trigger cytokine precursors, but also as an important initial step in the activation of NLRP3 inflammasome [43,44]. But the latter, different with the NF-κB pathway, NLRP3 inflammasome promotes maturation and secretion of cytokine. The expression of NLRP3, ASC, and caspas-1 increased significantly under the stimulation of LPS. Then, NLRP3 inflammasome is activated under the stimulation of damage-associated molecular patterns (DAMPs), such as ROS and ATP; then, ASC is recruited, caspase-1 is activated, and pro-IL-1β is processed into mature form [45]. Two independent and distinct forms of signals together induce the production of proinflammatory cytokines, for instance, IL-1β. This experiment ascertains that dioscin simultaneously inhibits the activation of NF-κB signaling pathway and NLRP3 inflammasome, which implies that dioscin may have a stronger effect than the drug that only has single function way in antiinflammatory.
Besides the production of intracellular proinflammatory cytokines, in the process of inflammation, the release of proinflammatory cytokines is also the key to trigger the inflammation and pyroptosis and is the main way to release the proinflammatory cytokine IL-1β from the cell [46]. Previous studies have certified that LPS stimulation leads to pyroptosis of multiple tissues and organs, such as intestinal epithelium, vascular endothelium, and adipose tissue [47][48][49]. However, whether LPS can induce pyroptosis in mMECs or the relationship between pyroptosis of mMECs and mastitis has not been reported. In this experiment, we demonstrated that dioscin can reduce LPS+ATP-induced mMEC death, but whether the reduction is due to pyroptosis is still unclear. It is worth noting that there have been studies justifying that dioscin regulates apoptosis in various diseases [28,50]. Therefore, after the detection of caspas-3, we found that dioscin could reduce the expression and activation of caspas-3. In this study, we used Ac-DEVD-CHO, a specific inhibitor of caspase-3, to exclude the negative effect of caspase-3mediated apoptosis on mMECs. Subsequent detection of LDH and IL-1β proved that dioscin reduced LPS+ATPinduced mMEC pyroptosis. Studies have proven that GSDMD is an indispensable factor in the occurrence of cell pyroptosis [51]. The N-terminal GSDMD fragment mediates the formation of cell membrane pores and directly triggers the occurrence of cell pyroptosis [46]. We speculate whether GSDMD is the crucial to dioscin regulating pyroptosis. Our results indicate that dioscin reduces the cell pyroptosis of mMECs by inhibiting the production of GSDMD-N. It has been reported that the transcription of GSDMD can be regulated by NF-κB [49], so the speculation is raised that the dioscin is likely to inhibit GSDMD by regulating the activation of the NF-κB pathway. This conjecture needs further verification.
After exploring, one thing can be sure that dioscin has antioxidant effects that can scavenge the production of ROS

22
Oxidative Medicine and Cellular Longevity by activating Nrf2 [25]. While continuing and uncontrolled inflammation is closely related to oxidative stress, LPS promotes the infiltration of inflammatory cells and proinflammatory mediators to evoke more production of ROS; in turn, excessive ROS inflicts more severe inflammation, exacerbating tissue destruction [52][53][54]. In this experiment, dioscin inhibited the production of ROS by activating AMPK-Nrf2 in mastitis mice and mMECs. Recent studies exemplified that Nrf2 activation can inhibit inflammation-mediated damage through its antioxidant effect [55]. However, the effects of Nrf2 on the NF-κB signaling pathway, NLRP3 inflammasome, and cell pyroptosis remain controversial [56,57]. By adding ML385, Nrf2 inhibitors, the effect of Nrf2 on the NF-κB signaling pathway, NLRP3 inflammasome, and cell pyroptosis in LPS+ATP-stimulated mMECs was investigated. The results intended that dioscin had an inhibitory effect on NF-κB and NLRP3 in a Nrf2-dependent manner on one hand. On the other hand, dioscin may have other inhibitory pathways besides inhibiting mMEC pyroptosis by activating Nrf2.
In conclusion, dioscin improves pyroptosis in LPSinduced mice mastitis by activating AMPK/Nrf2 and inhibiting the NF-κB signaling pathway. We demonstrated that dioscin significantly alleviated LPS-induced mastitis in mice, and the underlying mechanisms of this protective effect include antioxidant and anti-inflammatory. In summary, the results of this study provide experimental basis for the application of dioscin in the treatment of mastitis and will lead the way for the therapy of mastitis caused by Gramnegative bacterial infection. The further study of dioscin should focus its inhabitation function in mastitis and other similar diseases.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.