Subacute ruminal acidosis induces pyroptosis via the mitophagy-mediated NLRP3 inflammasome activation in the livers of dairy cows fed a high-grain diet

High-grain (HG) feeding can trigger subacute ruminal acidosis (SARA) and subsequent liver tissue injury. This study investigated pyroptosis and NLRP3 inflammasome activation in SARA-induced liver injury, and the role of mitophagy during this process. Twelve mid-lactating Holstein cows equipped with rumen fistulas were randomly divided into 2 groups: a low-grain (LG) diet group (grain: forage = 4:6) and a HG diet group (grain: forage = 6:4). Each group had 6 cows. The experiment lasted for 3 weeks. The ruminal fluid was collected through the rumen fistula on experimental d 20 and 21 and the pH immediately measured. At the end of the experiment, all animals were euthanized, and peripheral blood and liver tissue were collected. The ruminal pH was lower in the HG group than that in the LG group at all time points (On d 20: diet, P < 0.001; time, P = 0.02. On d 21: diet, P < 0.001; time, P = 0.002). In addition, the ruminal pH in the HG group was lower than 5.6 at 3 consecutive time points after feeding (4, 6, and 8 h on d 20; 2, 4, and 6 h on d 21), indicating that HG feeding induced SARA. The content of lipopolysaccharide ( P = 0.016), interleukin 1 β (IL-1β; P < 0.01), and apoptosis-related cysteine protease 1 (caspase-1; P < 0.01) and the activity of alanine aminotransferase ( P = 0.026) and aspartate aminotransferase ( P = 0.002) in the blood plasma of the HG group were all significantly increased. Hepatic caspase-1 activity ( P < 0.001) was increased in the livers of the HG group. The increased expression levels of pyroptosis-and NLRP3 inflammasome-related genes


INTRODUCTION
Increasing the milk yield is the primary goal of the dairy farming industry because of the urgent demand for dairy products.To achieve this goal, feeding a high-grain (HG) diet that contains more grains and less forage has become a universal strategy.A HG diet provides a large amount of energy for lactating cows that are prone to energy deficits because of high milk yield (Plaizier et al., 2008).However, this strategy has an adverse impact.Due to their high rumen fermentability, long-term feeding of HG diet can lead to the accumulation of a high concentration of ruminal organic acids and decreased rumen buffering (Kleen et al., 2003, Stone, 2004, Rustomo et al., 2006), further decreasing ruminal pH and causing subacute ruminal acidosis (SARA).Long-term exposure to SARA is recognized to affect feed intake, milk production, and rumen digestion and increases the risk of diseases such as mastitis, liver abscess, diarrhea, and laminitis, constituting a Subacute ruminal acidosis induces pyroptosis via the mitophagy-mediated NLRP3 inflammasome activation in the livers of dairy cows fed a high-grain diet Hongzhu Zhang, 1 Huimin Shi, 1 Wan Xie, 1 Meijuan Meng, 1 Yan Wang, 1 Nana Ma, 1 Guangjun Chang, 1 and Xiangzhen Shen 1 * major threat to the health of dairy cows (Nocek, 1997, Kleen et al., 2003, Plaizier et al., 2008).The liver is the "first pass" organ as this collects digestive tractderived blood containing pathogens and toxins for detoxification.During SARA, the rumen epithelium can be damaged and lipopolysaccharide (LPS), histamine, and many other pathogenic metabolites in the rumen can be absorbed into the blood and cleared by the liver (Chang et al., 2015, Aschenbach et al., 2019).However, long-term exposure to LPS causes impaired hepatic function, enabling the diffusion of LPS in peripheral circulation, which further produces multiple organ injuries and systemic inflammatory responses (Stone, 2004, Aschenbach et al., 2019).
Pyroptosis is a caspase-1-dependent programmed cell death that is characterized by cell membrane rupture and the release of proinflammatory cytokines (Cookson andBrennan, 2001, Jorgensen andMiao, 2015).The onset of pyroptosis requires the activation of the inflammasome, among which the NLR family pyrin domain-containing 3 (NLRP3) inflammasome is the most extensively studied.NLRP3 inflammasome activators include extracellular activators such as viruses, LPS, and damaged cells and intracellular activators such as oxidized mitochondrial DNA (Latz et al., 2013, Rathinam andFitzgerald, 2016).The activation of NLRP3 inflammasome induces the cleavage of gasdermin D (GSDMD) and maturation of pro-interleukin 1 β (pro-IL-1β) and pro-IL-18 (Latz et al., 2013).Subsequently, GSDMD-N-terminal (GSDMD-NT), the activated form of GSDMD, oligomerizes and causes the rupture of the cell membrane.Proinflammatory cytokines including IL-1β and IL-18 enter the extracellular space through the damaged cell membrane, triggering tissue injury (Cookson and Brennan, 2001, Liu et al., 2016, Shi et al., 2017).The NLRP3 inflammasome-mediated pyroptosis has been shown to be an important causative factor of tissue injury.Activated NLRP3 increased gene and protein expression of IL-1β and IL-18, inducing liver inflammation and fibrosis in mice (Wree et al., 2014).In addition, pyroptosis is recognized to be related to liver failure and nonalcoholic fatty liver disease (Beier andBanales, 2018, Wang et al., 2020).However, whether pyroptosis participates in SARA-induced liver injury in dairy cows has not yet been investigated.
Pyroptosis is intimately related to reactive oxygen species (ROS) and mitophagy.As the main site of ROS production, mitochondria play a pivotal role in maintaining cellular homeostasis (Zhou et al., 2011, Pickles et al., 2018).Damaged mitochondria are prone to ROS production (Ashrafi and Schwarz, 2013), and once damage occurs, phagocytic vesicles can engulf and degrade damaged mitochondria in a process known as mitophagy (Narendra et al., 2008).Inhibition of mitophagy causes the accumulation of damaged mitochondria and consequently excessive levels of ROS that activate the NLRP3 inflammasome and trigger pyroptosis (Han et al., 2021).Liu et al. (2022) reported that LPS treatment inhibited mitophagy, increased ROS levels, and triggered pyroptosis in splenic macrophages.Another study showed that activating mitophagy decreased the accumulation of ROS, inhibited NLRP3 inflammasome activation, and suppressed pyroptosis, thereby alleviating kidney injury (Lin et al., 2019).However, the role of mitophagy in SARA-induced liver injury in dairy cows remains unclear.
Therefore, we hypothesized that SARA induces abnormal mitophagy, activates NLRP3 inflammasome, and triggers pyroptosis to cause liver injury in dairy cows fed a HG diet.This study sought to investigate whether SARA can induce pyroptosis and to demonstrate the underlying mechanism.Our research may provide a novel insight into the mechanism whereby SARA induces liver injury in dairy cows.

Ethics statement
This study was approved by the Animal Ethics Committee for the Use and Care of Nanjing Agricultural University (No. SYXK-2017-0027), and all animalbased operations were performed in strict accordance with the guideline of the Ministry of Science and Technology's Law on Experimental Animals (2006, Beijing, China).

Animals, diet, and experimental design
A total of 12 mid-lactating Holstein dairy cows (parity: 2-3; body weight: 651 ± 54 kg; lactation days: 233 ± 16 d) equipped with rumen fistulas were used in the study and had been raised in Taizhou dairy herds (Jiangsu, China).During a 2-week adaptation period, all cows were fed a LG diet.Subsequently, the 12 cows were randomly divided into a HG group (forage: grain = 4:6, n = 6) and a LG group (forage: grain = 6:4, n = 6), and were respectively fed HG and LG diets for 3 weeks.Each group was fed at 04:00, 12:00, and 20:00 each day.During the experimental period, cows were housed in individual tiestalls and had free access to water.The body condition of cows was monitored daily by evaluating the rectal temperature, respiratory rate, and feed intake.All cows were healthy during the experimental period.The specific ingredients in the diet are listed in Table 1.The DMI is listed in Supplemental Table S1 (https: / / data .mendeley.com/datasets/ 6f5x5v7dp9/ 1).

Sample collection
On d 20 and 21 of the experimental period, rumen fluid was collected from the ventral sac of the rumen through the rumen fistula before feeding in the morning (0 h) and at 2, 4, 6, 8, and 12 h after feeding.After filtering the rumen fluid through 4 layers of gauze, the rumen fluid pH value was immediately checked with a pH meter (HI 9125,Hanna Instruments,Italy).Milk samples were collected on d 20 and 21 of the experimental period for milk composition analysis (Supplemental Table S1, https: / / data .mendeley.com/datasets/ 6f5x5v7dp9/ 1).On d 21 of the experimental period, peripheral blood was collected with a heparin sodium vacuum tube, and plasma was collected after centrifugation (1,000 × g, 15 min) and stored at −80°C.At the end of the experiment, cows were anesthetized with a captive bolt and killed by carotid artery bleeding.After removing the head, hooves, and tail, the liver was immediately separated from the body and was rinsed with sterile saline to remove the blood clot on the surface.Then, the liver was diced into approximately 1 cm 3 cubes with a scalpel, that were then stored in liquid nitrogen.All kits used in our study are listed in Table 2.

RNA extraction, cDNA synthesis, and real-time quantitative PCR
RNA extraction was described in our previous study (Zhang et al., 2023).Briefly, liver samples were soaked in liquid nitrogen and ground into powder with a pestle.A total of 100 mg of tissue powder and 1 mL of RNAiso Plus (Cat # 9108, TaKaRa, Shiga, Japan) were added to a 1.5-mL nuclease-free tube and mixed for 10 min.Then, chloroform and isopropanol were added and centrifuged at 16,000 × g, 4°C for 15 min.The RNA precipitate was collected and dissolved in 50 μL of nuclease-free water.The RNA concentration and purity were measured with a NanoDrop ND-1000 spectrophotometer (Thermo Fisher, San Jose, CA, USA).The optical density ratio at 260 and 280 nm (OD 260 / OD 280 ) of RNA was between 1.9 and 2.0.RNA integrity was checked via electrophoresis (Supplemental Figure S1, https: / / data .mendeley.com/datasets/ 6f5x5v7dp9/ 1).A reverse transcription kit (Cat # R323, Vazyme, Nanjing, China) was used to synthesize cDNA with a Mastercycler Nexus (Eppendorf, Hamburg, Germany).
cDNA was diluted with nuclease-free water.Then, 5 μL of 2 × ChamQ Universal SYBR qPCR Master Mix (Cat # Q711, Vazyme, Nanjing, China), 2 μL of cDNA, 0.4 μL each of 10 μM forward and reverse primers, and 2.6 μL of nuclease-free water were added to a 200-μL nuclease-free tube.Reactions were conducted in an ABI 7300 Real-Time PCR system (Thermo Fisher).Primers used in this experiment were designed with Oligo 7 software (Molecular Biology Insights, Colorado Springs, USA) and are listed in Table 3. Actin β (ACTB) and glyceraldehyde phosphate dehydrogenase (GAPDH) were used as the internal reference.The relative abundance of target genes was normalized to the geomean of ACTB and GAPDH and analyzed through the 2 −△△Ct method.ACTB and GAPDH were stably expressed in different groups (Supplemental Figure S2, https: / / data .mendeley.com/datasets/ 6f5x5v7dp9/ 1).

Mitochondrial protein isolation
A mitochondrial extraction kit (G008-1, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) was used for mitochondrial protein isolation from liver tissue according to the manufacturer's instructions.Briefly, a total of 200 mg of liver powder and 1.5 mL of lysis buffer were mixed.After centrifugation at 800 × g, 4°C for 5 min, 0.5 mL of supernatant was transferred to a 1.5-mL tube and mixed with 0.5 mL of solution A. After another centrifugation (15,000 × g, 4°C, 10 min), the protein precipitate (mitochondrial protein) was collected for subsequent Western blot analysis.LG, low-grain; HG, high-grain.

Total protein isolation and Western blot analysis
The total protein isolation method was described in our previous study (Zhang et al., 2023).Briefly, 100 mg of liver tissue powder, 10 μL of protein phosphatase inhibitor, 10 μL of protease inhibitor, and 1 mL of protein lysates (Cat # PC101, Epizyme Biotech, Shanghai, China) were added to a 1.5-mL Eppendorf tube and mixed well for 15 min.After centrifugation at 16,000 × g, 4°C for 15 min, the protein concentration in the supernatant was measured with a BCA Protein Assay Kit (Cat # 23225, Thermo Fisher).All samples were diluted to the same concentration and mixed with 5 × protein loading buffer (Cat # LT101, Epizyme Biotech) to maintain the stability of protein molecules and indicate the location of the protein during electrophoresis.
Different concentrations of polyacrylamide gels (10% and 12.5%, Cat # PG112 and PG113, Epizyme Biotech) were used for electrophoresis analysis of different proteins.After electrophoresis, the separated proteins were transferred to polyvinylidene fluoride membranes (Cat # p2938, Millipore, Bedford, MA, USA).Subsequently, membranes were soaked in 5% bovine serum albumin (Cat # A8010, Solarbio, Beijing, China) for analysis of phosphorylated proteins or 5% skim milk for all other blots at room temperature for 2 h followed by incubation with the primary antibody at 4°C for 12 h.The antibodies used in this study are listed in Table 4.After 6 washes with Tris-buffer saline/Tween (TBST), membranes were soaked in horseradish peroxidaseconjugated secondary antibodies at room temperature for 1 h.After another 6 washes with TBST, the membranes were incubated in chemiluminescence reagent (Cat # E422, Vayzme).The chemiluminescence of bands were detected using a ChemDoc XRS+ Imaging System (Bio-Rad, Hercules, CA, USA) and were analyzed with Image J software (National Institutes of Health, Bethesda, MD, USA).

Measurement of plasma LPS
The LPS concentration in blood plasma was measured with a Chromogenic End-point Tachypleus Amebocyte Lysate Assay Kit (EC80545S, Chinese Horseshoe Crab Reagent Manufactory Co. Ltd., Xiamen, China).Plasma was diluted 10-fold with endotoxin-free water before analysis.A total of 100 μL of endotoxin-free water, 100 μL of diluted plasma, and 100 μL of tachypleus amebocyte lysate was added to a pyrogen-free tube and mixed.After incubation at 37°C for 30 min, 100 μL of chromogenic matrix solution was added followed by incubation at 37°C for 6 min.Then, 500 μL each of azotization reagent I, II, and III was added in order and mixed.After incubating at room temperature for 5 min, The optical density (OD) was measured at 545 nm using a microplate reader (Thermo Fisher).The standard curve of endotoxin concentration (y) and OD value (x, at 545 nm) was y = 0.0707x − 0.006, R 2 = 0.9981.The endotoxin concentration of the standard curve consisted of 0.01, 0.0025, 0.05, and 0.1 EU/mL, and were all measured in triplicate.The lowest y-value of the standard curve was greater than that of the nega- tive control, and the average value of the samples to be measured was within the range of the standard curve.

Determination of caspase-1 activity
A caspase-1 activity assay kit (Cat # C1101, Beyotime Biotech, Beijing, China) was used to analyze hepatic caspase-1 activity.Briefly, 5 mg of tissue powder and 100 μL of lysate were mixed and fully homogenized on ice.After centrifugation (16,000 × g, 4°C, 15 min), 50 μL of supernatant, 40 μL of buffer, and 10 μL of Ac-YVAD-pNA were mixed and incubated at 37°C for 1 h.The OD was measured at 405 nm using a microplate reader (Thermo Fisher).

Dihydroethidium staining assay
ROS were detected with a dihydroethidium (DHE) kit (Cat # 50102ES02) supplied by Yeasen Biotech  (Shanghai, China).Briefly, frozen liver samples were cut into slices with a cryostat (CM1950, Leica, Germany).Each frozen slice was incubated with 200 μL of 10 μM DHE solution at 37°C for 1 h.After 3 PBS washes, the fluorescence was detected and quantitated with a LSM 710 confocal laser microscope system (Zeiss, Oberkochen, Germany).

Statistical analysis
All statistical analyses were performed using SPSS 26.0 (IBM Inc., Armonk, NY, USA).Data were tested for normality of distribution with the Shapiro-Wilk test.Data for hepatic ROS, activities of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in blood plasma, and IL-1β and NLRP3 gene expression were not normally distributed and were analyzed with a nonparametric test using the Mann-Whitney U test.The rumen fluid pH values were analyzed using 2-way ANOVA with a univariate general linear model.Other data were analyzed using an independent-sample t-test.The results are expressed as the mean and standard error of the mean (mean ± SEM).The data were considered statistically significant at P < 0.05.

Rumen fluid pH and LPS concentration in plasma
The rumen fluid pH measured at different time points after feeding on experimental d 20 and 21 is shown in Table 5.The rumen fluid pH values in the HG group were lower than those in the LG group at all time points (On d 20: diet, P < 0.001; time, P = 0.02.On d 21: diet, P < 0.001; time, P = 0.002).Notably, the rumen fluid pH was lower than 5.6 at 4, 6, and 8 h on d 20 and at 2, 4, and 6 h on d 21 in the HG group, whereas the rumen fluid pH was higher than 5.6 at all time points in the LG group.Moreover, the LPS content in peripheral blood was significantly higher in the HG group than that in the LG group (P = 0.016) (Table 6).These results showed that SARA was successfully induced in dairy cows of the HG group.

Expression of pyroptosis-related markers in peripheral blood and liver
As shown in Table 6, the activities of ALT (P = 0026) and AST (P = 0.002) in the HG group were significantly higher than those in the LG group.Also, we assessed the content of pyroptosis-related markers in plasma and liver.As shown in Figure 1, the hepatic caspase-1 activity in the HG group was significantly higher than that in the LG group (P < 0.001).Furthermore, the caspase-1 content in the blood was significantly higher in the HG group than that in the LG group (P < 0.001), whereas the content of IL-18 (P = 0.260) was not significantly different (Table 7).Notably, the content of TNF-α (P < 0.001), IL-1β (P < 0.001), and IL-6 (P < 0.001) were significantly higher in the HG group than that in the LG group, indicating a systemic inflammatory response (Table 7).In addition, Western blot analysis showed that the protein expression of GS-DMD-NT, the effector in pyroptosis, was significantly higher in the HG group than that in the LG group (P = 0.006) (Figure 2A-B).In agreement with the Western blot analysis, the results of real-time quantitative PCR showed that the expression of GSDMD (P = 0.001) was also significantly higher in the HG group than that in the LG group (Figure 2C).

Expression of NLRP3 inflammasome-related genes and proteins in the liver
To investigate whether the NLRP3 inflammasome was activated in the livers of dairy cows with SARA, the expression levels of NLRP3 inflammasome-related genes and proteins were detected.As shown in Figure 3, the expression of NLRP3 (P = 0.002), CASP1 (P < 0.001), ASC (P = 0.001), IL1B (P = 0.002), and IL18 (P < 0.001) was significantly higher in the HG group than that in the LG group.Western blot analysis showed that the protein expression levels of NLRP3 (P = 0.017), ASC (P = 0.016), cleaved-caspase-1 (P = 0.001), cleaved-caspase-11 (P < 0.001), and IL-1β (P < 0.001) were also significantly increased over those in the LG group while the expression of IL-18 (P = 0.076) tended to be higher in the HG group than that in the LG group (Figure 4).

Expression of mitophagy-related proteins in the liver
In terms of mitophagy, Western blot analysis showed that the expression of MAP1LC3-II (P = 0.001) and BECN1 (P = 0.001) was significantly higher in the HG group than that in the LG group, whereas the expression of SQSTM1 (P = 0.039) was significantly lower in the HG group than that in the LG group (Figure 5).Furthermore, the expression levels of Parkin (P < 0.001) and PINK1 (P = 0.006), 2 proteins involved in mitochondrial substrate recognition by autophagy adaptors, were significantly higher in the HG group than those in the LG group (Figure 6).

Content of ROS in the liver
As ROS is related to mitophagy and the activation of NLRP3 inflammasome, we assessed the content of As shown in Figure 7, the fluorescence intensity was significantly enhanced in the HG group compared with that in the LG group (P = 0.002), indicating that the content of ROS in the HG group was significantly more than that in the LG group.

DISCUSSION
SARA is prevalent in the dairy herds of many countries.It has been determined that more than 33% of dairy cows were experiencing SARA in 10 herds of Italy (Morgante et al., 2007) while Kleen et al. (2009) reported a prevalence of 13.8% in 18 dairy herds of Dutch.The factors leading to SARA are complex and diverse.As the HG diet lacks structural fiber and contains excessive amounts of grain, feeding a HG diet causes rapid fermentation of carbohydrates and consequently the accumulation of volatile fatty acids (VFA) in the rumen, thereby decreasing the ruminal pH (Plaizier et al., 2008).The shift of diets is also an important factor inducing SARA.The acid absorption capacity of the rumen is low for the ruminal papillae are short during the dry period and early-lactation period (Kleen et al., 2003).Therefore, the high amounts of VFA caused by feeding the early-lactation diet is a challenge to the rumen.Although many studies have been conducted on SARA, the exact definition of SARA remains in dispute.A ruminal pH drop down to 5.5 has been pro- LG, low-grain; HG, high-grain.LG, low-grain; HG, high-grain.LG, low-grain; HG, high-grain.
posed as the threshold for SARA (Kleen et al., 2003).Gozho et al. (2005) proposed that when SARA occurs, the ruminal pH should be lower than 5.6 for more than 3 h per day.In our study, we measured ruminal pH at different time points after feeding.The ruminal pH values were lower than 5.6 at 4, 6, and 8 h on d 20 and at 2, 4, and 6 h on d 21.In addition, the ruminal pH values were no higher than 5.5 at 4 h on both d 20 and 21.Our results meet the criteria for SARA occurrence proposed by Kleen et al. and Gozho et al.In addition to ruminal pH, systemic inflammatory response is a nonspecific marker of SARA (Gozho et al., 2005).
We observed that feeding a HG diet induced systemic inflammatory responses evidenced by increased content of IL-1β, IL-6, and TNF-α in the blood, which is auxiliary evidence for the occurrence of SARA.Therefore, we suggest that feeding a HG diet induced SARA in this study.
The detrimental effects of SARA include decreased feed intake and milk yield, milk fat depression, and the occurrence of diseases such as diarrhea, laminitis, liver abscess, and mastitis (Kleen et al., 2003).Our previous study detected an inflammatory response in mammary gland in cows with SARA (Meng et al., 2022).In this study, we observed milk fat depression, but no other disease was found, which may be due to the short experiment period.Obvious consequence from SARA could arise with a certain delay from the initial insult (Kleen and Cannizzo, 2012).Although the cause of these detrimental effects remains to be investigated, LPS is considered as a contributory factor (Nagaraja andTitgemeyer, 2007, Plaizier et al., 2012).Under SARA condition, the low ruminal pH increases populations of gram-positive bacteria and decreases populations of gram-negative bacteria (Li et al., 2012).The free ruminal LPS is derived from the lysis of gram-negative bacteria caused by low ruminal pH and could transfer into the blood through damaged rumen epithelium, inducing a systemic inflammatory response (Khafipour et al., 2009, Plaizier et al., 2012, Monteiro and Faciola, 2020).LG, low-grain; HG, high-grain; GSDMD-FL, gasdermin D full length; GSDMD-NT, gasdermin D N-terminal; ACTB, actin β. **P < 0.01.Khafipour et al. (2009) and Zhao et al. (2018) reported an increase in the circulatory level of LPS in cows with SARA compared with that in healthy dairy cows.Our previously published study also demonstrated that feeding HG diet induced hepatic inflammatory injury and a higher content of blood LPS (Guo et al., 2017).In this study, we observed a significant increase in LPS in plasma from 0.24 EU/mL in the LG group to 0.42 EU/mL in the HG group.Furthermore, the activities of ALT and AST, the 2 most commonly used parameters in liver injury assessment, were significantly elevated in the HG group.Stoldt et al. (2015) and Shi et al. (2021) regarded AST and ALT activities as important indicators for judging the occurrence of liver injury in dairy cows.Gonzalez et al. (2011) believed that AST activity higher than 100 U/L is indicative of hepatic injury in cows.The AST activity (mean value: 197 U/L) in our study far exceeds this value, which indicates the occurrence of liver injury.Nevertheless, we cannot conclude whether liver injury is caused by increased LPS in the blood.In addition to LPS, many other pathogenic and immunogenic proteins, such as fimbrial adhesins, heatstable and heat-labile toxins, and inflammatory peptides, are present in the blood during SARA (Plaizier et al., 2008).Furthermore, histamine was reported to trigger an inflammatory response during SARA (Chang et al., 2018).Therefore, our result demonstrated that LPS may be a causative factor in SARA-induced liver injury.Further experiments such as infusion of LPS in the portal vein should be conducted to investigate the role of LPS in SARA-induced liver injury.
Pyroptosis is a form of programmed cell death that has been widely investigated in many experimental models but not in SARA-induced liver injury in dairy cows.Pyroptosis is characterized by the rupture of the cell membrane, a process that relies on GSDMD, and the release of proinflammatory cytokines (Cookson and Brennan, 2001).Once activated, the GSDMD is cleaved into the GSDMD-NT and the GSDMD-C terminal (Shi et al., 2017).The GSDMD-NT oligomerizes and attaches to the inner cell membrane to create pores in the cell membrane, and consequently, the cell membrane ruptures and releases proinflammatory cytokines such as IL-1β and IL-18 (Liu et al., 2016).Since IL-1β secretion also increases in general inflammatory responses, IL-18 and GSDMD are specific markers of pyroptosis.Many studies have reported pyroptosis in liver diseases such as hepatic fibrosis, hepatocellular carcinoma, nonalcoholic fatty liver, and viral hepatitis (Szabo andCsak, 2012, Yu et al., 2021).Gaul et al. (2021) reported  the increased protein expression of GSDMD-NT in the liver as well as elevated levels of plasma IL-1β and IL-18 in liver fibrosis.Ruan et al. (2021) also reported pyroptosis in a nonalcoholic fatty liver disease model.In agreement with these studies, our findings showed that the expression of GSDMD mRNA and GSDMD-NT protein were significantly higher in the HG group.Moreover, SARA induced elevated contents of IL-1β, IL-18, and caspase-1 in the blood plasma.Our results suggest that pyroptosis was activated in the livers of dairy cows fed a HG diet, and may exert a vital role in SARA-induced liver injury.
The activation of NLRP3 inflammasome is essential for initiating pyroptosis.Qu et al. (2022) showed that the NLRP3 inflammasome antagonist MCC950 alleviated deoxynivalenol-induced pyroptosis in IPECs and intestinal injury in mice.The overexpression of NLRP3 aggravated LPS-induced pyroptosis in neonatal rat cardiomyocytes (Li et al., 2019).The NLRP3 inflam-masome is a member of the inflammasome family and comprises 3 components: a cytosolic sensor NLRP3, an adaptor ASC, and an effector pro-caspase-1 (De Nardo and Latz, 2011, Rathinam andFitzgerald, 2016).The assembly of NLRP3, ASC, and pro-caspase-1 into the NLRP3 inflammasome induces cleavage of procaspase-1 into caspase-1, which subsequently activates GSDMD (Rathinam and Fitzgerald, 2016).Here, we found increased protein expression of NLRP3, ASC, and caspase-1 in the HG group, which indicated that the NLRP3 inflammasome was activated.We also observed increased protein expression of caspase-11, an effector in the non-canonical pathway of NLRP3 activation, in the HG group.When cells are exposed to gram-negative bacteria or LPS, the pro-caspase-11 is cleaved into active caspase-11, which directly activates GSDMD, and subsequently, the GSDMD-NT creates cell membrane pores and activates NLRP3 inflammasome, causing the maturation of proinflammatory cytokines (He et al., 2015, Kayagaki et al., 2015).Our data indicated that SARA induced the activation of the NLRP3 inflammasome to facilitate the initiation of pyroptosis and release of proinflammatory cytokines in the liver of cows fed a HG diet.Moreover, we found that non-canonical pathways participated in SARA-induced NLRP3 inflammasome activation, which may be attributed to the increased content of LPS in blood.
ROS is produced by the respiratory chain in mitochondria, and accumulation of ROS in the liver can lead to liver damage.Under SARA condition, the increased ROS may come from 2 sources: long-term feeding of a high-energy diet increases the metabolic load of mitochondria, and subsequently causing excessive ROS levels (Rani et al., 2016); in addition, rumen-derived harmful substances such as LPS can impair the antioxidant system in the liver, causing the accumulation of ROS (Mier-Cabrera et al., 2011).Here, we observed a higher content of hepatic ROS in cows fed a HG diet.We consider that both the role of HG diet and the role of LPS cannot be ignored in causing the increased levels of hepatic ROS.It has been demonstrated that ROS mediate the activation of the NLRP3 inflammasome (Zhou et al., 2011).Pretreatment with N-acetyl-L-cysteine, a ROS scavenger, inhibited the NLRP3 inflammasome and attenuated proptosis in LPSexposed Leydig cells (Li et al., 2019).In this study, we observed both increased ROS and activation of NLRP3 inflammasome in the livers of dairy cows with SARA, indicating that increased ROS may mediate the activation of NLRP3 inflammasome in SARA-induced liver injury.Mitophagy maintains the balance between ROS formation and their removal by digesting damaged mitochondria (Ashrafi andSchwarz, 2013, Pickles et al., 2018).Disturbed mitophagy causes the accumulation of ROS (Ko et al., 2021).Therefore, we hypothesized that mitophagy is disturbed in SARA-induced liver injury.To verify our hypothesis, we extracted mitochondrial protein to examine mitophagy and found that the expression of autophagy-related proteins MAP1LC3-II, BECN1, and ATG5 was increased while that of SQSTM1 was decreased in the HG group.Parkin and PINK1 are 2 regulators of mitophagy (Pickrell and Youle, 2015).PINK1 can attach to the outer membrane of damaged mitochondria and lead to the activation of Parkin (Kondapalli et al., 2012).The activated Parkin ubiquitinates proteins on the surface of mitochondria and facilitates the recognition of these proteins by autophagy adaptors (Narendra et al., 2008).The protein expression of Parkin and PINK1 was increased in the HG group, indicating an increased mitophagy in the livers of dairy cows fed a HG diet.We suggest that in the hepatocytes exposed to blood pathogens, mitochondria were damaged and the levels of ROS were increased, and mitophagy was consequently adaptively increased to eliminate excessive ROS.

CONCLUSIONS
Feeding a HG diet induced SARA in dairy cows by decreasing the rumen pH and increasing LPS levels in blood.SARA induced liver injury and systemic inflammatory response by increasing the content of proinflammatory cytokines TNF-α and IL-6 and pyroptosisrelated cytokines IL-1β and IL-18 in blood.SARA also increased protein expression of GSDMD-NT, causing pyroptosis in the liver of dairy cows.In addition, a higher ROS content, increased mitophagy, and activated NLRP3 inflammasome were found in the livers of dairy cows with SARA.
Zhang et al.: SUBACUTE RUMINAL ACIDOSIS AND LIVER HEALTH OF COWS Zhang et al.: SUBACUTE RUMINAL ACIDOSIS AND LIVER HEALTH OF COWSROS in the livers of dairy cows via DHE staining assay.
Figure 2. Expression of pyroptosis-related genes and proteins in liver tissue.(A, B) The relative protein expression of GSDMD-NT in liver tissue was measured via Western blot analysis.(C) The mRNA relative expression of GSDMD was measured via real-time quantitative PCR; ACTB and GAPDH were selected as the reference genes.Each group contained 6 cows.Values are the mean ± SEM.LG, low-grain; HG, high-grain; GSDMD-FL, gasdermin D full length; GSDMD-NT, gasdermin D N-terminal; ACTB, actin β. **P < 0.01.
Figure 6.Expression of Parkin and PINK1 in liver tissue.The relative protein expression of Parkin and PINK1 against the reference COX IV protein.Each group contained 6 cows.Values are the mean ± SEM.LG, low-grain; HG, high-grain; PINK1, PTEN induced kinase 1. **P < 0.01.
Figure 7. DHE staining and fluorescence quantitative analysis of the liver tissue section.ROS production was determined by dihydroethidium staining, and the fluorescence was visualized with a confocal laser scanning microscope.Scale bar, 100 μm.Values are the mean ± SEM.LG, low-grain; HG, high-grain; DHE, dihydroethidium staining image detected at 610 nm; Bright, bright field image detected under white light.**P < 0.01.
Zhang et al.: SUBACUTE RUMINAL ACIDOSIS AND LIVER HEALTH OF COWS

Table 2 .
Zhang et al.: SUBACUTE RUMINAL ACIDOSIS AND LIVER HEALTH OF COWS Information for kits used in experiment Ma et al. (2018)d byMa et al. (2018); 2, 3 the kits was used byZhang et al. (2023); 4 the kit was used by He et al. (2021); 5 the kit was used by Wang et al. (2021); 6 the kit was used by Wei et al. (2020); 7 the kit was used by Zheng et al. (2021); 8 the kit was used by Zhang et al. (2023);

Table 4 .
Antibody information for Western blot analysis

Table 5 .
Zhang et al.: SUBACUTE RUMINAL ACIDOSIS AND LIVER HEALTH OF COWS Rumen pH values on experimental d 20 and 21 1Values are the mean ± SEM.

Table 6 .
The concentration of LPS and the activities of ALT and AST in blood plasma 1EU, endotoxin units. 2

Table 7 .
The contents of IL-1β, IL-18, and caspase-1 in the blood plasma