Omeprazole increases survival via different mechanisms in two rat models of liver injury

Omeprazole (OMZ) is a proton pump inhibitor (PPI) that is used to reduce gastric acid secretion, but little is known about its possible liver protective effects. This study investigated whether OMZ has benecial effects in rat septic models of lipopolysaccharide (LPS)-induced liver injury after D-galactosamine (GalN) treatment and 70% hepatectomy (PH), and to determine the mechanisms of OMZ in an in vitro model of liver injury. In the in vivo models, the effects of OMZ were examined 1 h before treatment. OMZ increased survival and decreased tumor necrosis factor-alpha, inducible nitric oxide synthase, cytokine-induced neutrophil chemoattractant 1, interleukin (IL)-6, and IL-1β mRNA expression, and increased IL-10 mRNA expression in the livers of both GaIN/LPS- and PH/LPS-treated rats. Necrosis and apoptosis were inhibited by OMZ in GaIN/LPS rats, but OMZ had no effects on necrosis in PH/LPS rats. Primary rat hepatocytes were treated with IL1-β in the presence or absence of OMZ (in vitro model). OMZ inhibited iNOS induction partially through suppression of NF-κB signaling in hepatocytes. Furthermore, OMZ inhibited the induction of several inammatory mediators, resulting in the prevention of LPS-induced liver injury after GalN liver failure and PH, although OMZ showed different doses and mechanisms in the two models. of previous studies The regulation of inammatory reactions during the perioperative period is important to prevent organ damage and complications. In this study, we investigated the hepatoprotective effects of OMZ using in vivo and in vitro liver injury models. Our experiments showed that OMZ prevents proinammatory mediator expression (iNOS, TNF-α, CINC-1, IL-6, and IL-1β) by suppressing NF-κB activation. In addition, OMZ increased survival in GalN/LPS and PH/LPS rats. These results suggest that OMZ may have a role in preventing liver injury, and further in-depth studies are needed to explore its possible therapeutic applications.


Introduction
Sepsis is a condition in which the dysfunction of various organs such as the liver, heart, lungs, and kidneys occurs as a result of an infectious disease. Sepsis is characterized by systemic in ammatory responses induced by infection 1 . A particularly severe form of sepsis that presents with severe circulation and abnormal cell metabolism is "septic shock." Infection with Gram-negative bacteria leads to excess bacterial endotoxin (lipopolysaccharide, LPS), endotoxemia, in the blood. In Japanese intensive care units, frequent sites of infection are the pulmonary and intra-abdominal regions in severe sepsis patients 2 .
The incidence sepsis has gradually increased. Over the last decade, the number of reported sepsis diagnoses in the emergency department has tripled, exceeding the number of diagnoses of myocardial infarction. The reason for this increase is thought to be aging and the advancement of medical care. Many older people suffer from chronic disease and are susceptible to infectious diseases that are likely to become severe. In addition, with the progress of medical treatment, the number of cases in which treatments that suppress immunity are applied is increasing with the rise in transplant surgery and chemotherapy for cancer. Furthermore, patients treated for immunosuppression are more susceptible to infections. Most importantly for patient outcome, early diagnosis of sepsis and effective antibiotic treatment are essential.
There are several experimental animal models of endotoxemia and sepsis with liver failure, including two that we reported previously: simultaneous administration of D-galactosamine and LPS (GaIN/LPS) [3][4][5] and a partial (70%) hepatectomy followed by LPS administration (PH/LPS) 6, 7 . In our previous studies, high doses of LPS, such as ≥50 µg/kg and ≥250 µg/kg, resulted in poor survival (less than 10%) in two models of GalN/LPS and PH/LPS, respectively. In this study, we decreased the levels of LPS injected after GalN and PH treatment to closely re ect the conditions in human cases.
Omeprazole (OMZ), the rst clinically approved proton pump inhibitor (PPI), is a substituted benzimidazole that interacts with the gastric proton pump (H + , K + -ATPase) in the secretory membrane, resulting in potent long-acting inhibition of gastric acid secretion [10][11][12] , and it is extensively metabolized by the liver 13 . PPIs inhibit hydrogen potassium adenosine triphosphatase, which in turn leads to reduced gastric acid secretion from parietal cells 14 . Many studies have proposed other mechanisms by which PPIs exert their anti-in ammatory effects [15][16][17][18][19][20][21][22][23][24] . PPIs were demonstrated to be a revolutionary treatment for acid-related diseases, and they minimized the need for elective surgery for ulcers or re ux when introduced to clinical practice 25,26 .
Previously, we reported the liver-protective effects of lansoprazole (LPZ) 5 , a PPI. However, few studies have examined whether other PPIs in uence the expression of proin ammatory mediators and survival in animal models of liver injury or septic shock. In this study, we investigated the hepatoprotection of OMZ in in vivo and in vitro models. We rst used two rat models of liver injury induced with GaIN/LPS or PH/LPS and examined if OMZ in uences survival and various in ammatory mediators. Next, to determine the mechanisms of these hepatoprotective effects, we examined if OMZ inhibits iNOS induction and NO production in primary cultured rat hepatocytes 27 .

Results
Effects of LPS on survival in rat GalN/LPS and PH/LPS models. Previously, we applied GalN (500 mg/kg)/LPS (50 µg/kg) and PH (70% hepatectomy)/LPS (250 µg/kg) in two rat liver injury (septic) models 5,28,29 . In the GalN/LPS model, a mixture of GalN and LPS was simultaneously injected into the penile vein (intravenously, i.v.), while LPS was injected into the penile vein (i.v.) 2 days after 70% hepatectomy in the PH/LPS model. More than 90% of rats had died 72 h after GalN/LPS or LPS injection in these models. In the current study, we examined the effects of lower doses of LPS on survival without changing other conditions to obtain milder survival curves.
Effects of omeprazole on increased survival in rat GalN/LPS and PH/LPS models. OMZ is a potent PPI of H + -ATPase in gastric parietal cells and is widely used for the treatment of peptic ulcer disease and re ux esophagitis 30 . Approximately half of an oral dose of OMZ is systemically available because of substantial rst-pass elimination by the liver 13 . OMZ has been shown to be metabolized by the cytochrome P450 enzymes 3A4 (CYP3A4) and 2C19 (CYP2C19), 31 and the major metabolites found in plasma are hydroxy OMZ and OMZ sulfone 13 .
Effects of omeprazole on nuclear factor (NF)-κB activation in the livers of rat liver injury models. In the GalN (500 mg/kg)/LPS (2.5 µg/kg) and PH/LPS (25 µg/kg) models, electrophoretic mobility shift assay (EMSA) experiments revealed that OMZ (180 mg/kg and 100 mg/kg) inhibited the activation of NF-κB at Effects of omeprazole on pathological changes in the livers of GalN/LPS and PH/LPS models. In both rat models, the areas of focal necrosis with in ammatory cell in ltration and massive hemorrhage were increased in the positive controls at 1 and 6 h (or 4 h), while it was reduced by OMZ at 1 and 6 h (or 4 h) (Fig. 6). In myeloperoxidase (MPO) staining (necrosis), OMZ decreased MPO-positive cells compared with the positive controls in both models ( Fig. 6; B5 and E5). In terminal deoxynucleotidyl transferasemediated deoxyuridine triphosphate-digoxigenin nick-end labeling (TUNEL) staining (apoptosis), OMZ also decreased TUNEL-positive cells as compared with the positive controls at 6 h in the GalN/LPS model ( Fig. 6; C3), but no differences were observed with OMZ in the PH/LPS models ( Fig. 6; F3).
Next, we examined the mechanisms by which OMZ promoted these hepatoprotective effects in primary cultured rat hepatocytes stimulated with IL-1β to represent an in vitro model of liver injury 27 . In the liver during in ammation, in addition to the production of various in ammatory cytokines, the induction of iNOS gene expression is enhanced. Overproduction of NO by iNOS is considered a hepatic disorder, and suppression of iNOS induction is important for the alleviation of hepatic injuries. We demonstrated that NO produced by iNOS was an index of liver damage. We analyzed the organ-protective effects of various clinical drugs, conventional therapeutic drugs, herbal medicines, and functional drugs. Drugs showing inhibition of iNOS induction/NO production were examined in septic animal models to con rm their protective effects on survival.
Effects of omeprazole on nitric oxide production, iNOS protein expression, and in ammatory mediator mRNA expression in primary cultured rat hepatocytes. IL-1β stimulates iNOS mRNA/protein expression and NO production in primary cultured rat hepatocytes 32 . In this study, IL-1β also stimulated the expression of pro-in ammatory cytokines and chemokines such as TNF-α, IL-1β, and CINC-1.
In primary cultured rat hepatocytes, OMZ inhibited the production of nitric oxide (NO) (Fig. 7A, upper) and the expression of iNOS protein (Fig. 7A, middle) in a dose-dependent manner. OMZ showed no cellular toxicity at the indicated concentrations, as evaluated by lactate dehydrogenase (LDH) release and trypan blue exclusion (data not shown). OMZ also reduced the mRNA expression of iNOS, TNF-α, IL-1β, and CINC-1 (Fig. 7B), indicating that OMZ affects these genes at the transcriptional and/or posttranscriptional levels.
Effects of omeprazole on NF-κB activation and iNOS mRNA levels in primary cultured rat hepatocytes. Although OMZ had no effects on the degradation of IκBα (Fig. 8A), OMZ inhibited NF-κB activation at 2, 3, and 4 h ( Fig. 8B1 and 8B2). iNOS mRNA expression is regulated through activation of the iNOS promoter by transcription factors such as NF-κB and through post-transcriptional modi cations such as mRNA stabilization 33 . Transfection was performed using pRiNOS-Luc-SVpA and pRiNOS-Luc-3′UTR, which detected iNOS promoter activation (i.e., mRNA synthesis) and mRNA stability, respectively 34 . IL-1β increased the luciferase activity of these vector constructs, and these effects were inhibited by OMZ (Fig.  8C).

Discussion
In this study, we investigated the liver-protective effects of OMZ using two septic rat models (GaIN/LPS and PH/LPS) as in vivo liver injury models. We also attempted to clarify the protective mechanisms of OMZ in IL-1β-stimulated rat hepatocytes in an in vitro liver injury model 27 . OMZ demonstrated hepatoprotective effects in both in vivo models, and our experiments in the in vitro liver injury model indicated several possible mechanisms for these effects.
PPIs such as OMZ are clinically used for the treatment of gastric acid-related disorders including gastroduodenal ulcers, re ux esophagitis, and non-steroidal anti-in ammatory drug (NSAID)-induced gastric lesions. PPIs are also effective on elements of the immune system including monocytes, neutrophils, and endothelial cells 35 . PPIs suppress neutrophil functions such as chemotaxis, superoxide production, and degranulation via IL-8 36 . In addition, P-type proton-ATPase inhibitors have antiin ammatory effects by reducing neutrophil adhesion molecules and free oxygen radicals 37 . They are also known to activate heme oxyegenase-1 (HO-1), an endogenous antioxidant 38 . We have previously reported these hepatoprotective effects of LPZ, a PPI, because of its ability to induce an anti-oxidative stress response in the liver 5 . Sepsis is a major cause of death and is associated with hypotension (i.e., septic shock) and multiple organ failure, including liver failure 1 . However, the etiology of sepsis has not been completely elucidated and there is no speci c treatment. Therefore, determination of the cause of sepsis is especially important in a clinical situation.
In rats, GalN-treatment or 70% hepatectomy with a sublethal dose of LPS increases the sensitivity to endotoxin. Therefore, these rats induce liver failure [39][40][41] . NO in the serum starts to increase at 3 h and further increases until 6 h after LPS injection 4 . NF-κB, a transcription factor involved in in ammation and apoptosis 42 , mediates this induction, including iNOS stimulation 3,4,6 . The resulting cytokine storm provokes multiple organ failure, including liver failure, which is the result of apoptosis of hepatocytes induced by TNF-α 43,44 . Upregulation of iNOS, TNF-α, and other in ammatory mediators in in amed hepatocytes is central to liver in ammation. In response to interactions with pathogenic bacteria, in ammatory cells increase the production of these proin ammatory mediators, which in turn activate other processes that promote in ammation.
In rat models of GalN/LPS and PH/LPS, we reduced the doses of LPS after GalN treatment and 70% hepatectomy to 2.5 and 25 µg/kg, respectively. In these positive controls (without OMZ), lower LPS did not have any effect on survival (less than 10% or approximately 0%) in the GalN/LPS model, but increased survival (20%-40%) in the PH/LPS model (Fig. 1). Under such conditions, 180 and 100 mg/kg OMZ enhanced the cumulative survival of GalN/LPS and PH/LPS rats, respectively (Fig. 2). Biochemical analyses showed that OMZ inhibited the activation of NF-κB (Fig. 3), decreased the mRNA expression of in ammatory mediators (TNF-α, iNOS, CINC-1, IL6 and IL-1β), increased IL-10 mRNA expression in the liver (Fig. 4), and decreased the production of NO, ALT/AST, TNF-α, IL-6, and IL-1β in serum (Fig. 5). Further histopathological analyses in the liver also showed that OMZ reduced the areas of focal necrosis with in ammatory cell in ltration and massive hemorrhage in GalN/LPS and PH/LPS rats (Fig. 6), whereas MPO experiments demonstrated that OMZ reduced necrosis in both models (Fig. 6B5 and 6E5).
However, in TUNEL staining (apoptosis), OMZ reduced apoptosis in GalL/LPS but not PH/LPS rats.
In both models, OMZ had similar liver-protective effects. However, there were some differences between these models. For example, in the case of EMSA (NF-κB activation), positive control rats in the PH/LPS model exhibited less effective increases in NF-κB activation at 4 h, and OMZ had no effect (Fig. 3B). In contrast, at both 1 and 6 h in the GalN/LPS model, high increases in NF-κB activation were observed, which were inhibited by OMZ, and IL-6 and IL-1β were increased at 1 h in PH/LPS and at 6 h in GalN/LPS rats, respectively, which was also inhibited by OMZ. These differences may demonstrate an important indicator for the clinical use of OMZ in the future.
We next compared OMZ effects in different septic (liver injury) models. The dose and administration method of OMZ (40-240 mg/kg, i.p.; single administration) used in this study differed to the standard clinical use (20 mg/50 kg, i.v.; single administration). These doses were calculated according to those previously used in our experimental studies (100 mg/kg LPZ) 5 .
In addition, from the results obtained in in vitro primary cultured rat hepatocytes ( Fig. 7 and 8), we con rmed that OMZ inhibited the induction of iNOS in dose-dependent manner, followed by the blockade of excess NO production, which is one of the factors involved in organ injury including that of the liver 8, 9,27 . In vitro experiments also revealed that OMZ reduced the mRNA expression of other proin ammatory mediators (TNF-α, IL-1β, and CINC-1), in part through the inhibition of NF-κB activation. OMZ decreased the expression of iNOS mRNA and protein through the inhibition of both promoter transactivation (mRNA synthesis) and mRNA stabilization. These ndings are consistent with the results of previous studies 45,46 .
The regulation of in ammatory reactions during the perioperative period is important to prevent organ damage and complications. In this study, we investigated the hepatoprotective effects of OMZ using in vivo and in vitro liver injury models. Our experiments showed that OMZ prevents proin ammatory mediator expression (iNOS, TNF-α, CINC-1, IL-6, and IL-1β) by suppressing NF-κB activation. In addition, OMZ increased survival in GalN/LPS and PH/LPS rats. These results suggest that OMZ may have a role in preventing liver injury, and further in-depth studies are needed to explore its possible therapeutic applications.

Methods
Ethics statement. Animal care and experiments were performed in accordance with the standards in the ARRIVE 47 and PREPARE 48 guidelines. In addition to these, our study was in accordance with the relevant guidelines and regulations, which was approved by the Animal Care Committee of Kansai Medical University (19-009 and 20-059). Animal models were created in compliance with the criteria of the ARRIVE 47 and PREPARE 48 guidelines. All methods proposed in these studies were also carried out according to the standards of relevant institutional guidelines and regulations.

Animals. Male Wistar and
The rats that were randomly assigned to receive OMZ were injected (i.p.) with various doses of OMZ (40-100 mg/kg) 1 h before LPS treatment. Survival was monitored for 5 days. The rats were killed when they appeared weak and moribund because of the progression of liver failure, congestion, and multi-organ failure. We used the NIH O ce of Animal Care and Use 51 score and severity assessment to assess the animals following liver resection 52 . Liver and blood samples were collected from the rats 1 and 6 h after GaIN/LPS treatment, and 1 and 4 h after LPS treatment in PH/LPS. EMSA. EMSA was performed as described previously 53,54 with a minor modi cation, as described elsewhere 6, 55 . Nuclear extracts were prepared from frozen liver at −80°C or cultured hepatocytes. Binding reactions were undertaken by incubating the nuclear extracts in reaction buffer (20 mM HEPES-KOH, pH 7.9, containing 1 mM EDTA, 60 mM KCl, 10% glycerol, and 1 µg poly[dI-dC]) with a probe (40,000 dpm) for 20 min at room temperature. Products were electrophoresed on a 4.8% polyacrylamide gel in high-ionicstrength buffer, and dried gels were analyzed by autoradiography. An NF-κB consensus oligonucleotide (5′-AGTTGAG GGGA-CTTTCCCAGGC) from the mouse immunoglobulin κ light chain was purchased and labelled with [γ-32 P]-ATP (PerkinElmer, Tokyo, Japan) and T4 polynucleotide kinase. Protein was measured using the Bradford method 56 . Bands corresponding to NF-κB were quanti ed by densitometry using ImageJ 57 .
Reverse transcriptase-polymerase chain reaction (RT-PCR). Total RNA was extracted from the frozen liver samples or cultured hepatocytes using TRIzol ® reagent (the guanidinium thiocyanate-phenol-chloroform mixture) 58 . cDNA was synthesized from 1 µg total RNA from each sample with Oligo(dT)20 Primer (25 ng/µL), 5× RT Buffer (5 µL), 10 mM dNTPs Mixture (2.5 µL), RNase Inhibitor (20 units/0.5 µL), Rever Tra Ace (100 units/µL), and UltraPure™ DNase/RNase-free distilled water (total volume, 25 µL). The conditions of thermal cycling using iCycler (Bio-Rad Laboratories, Hercules, CA, USA) were 42°C for 60 min and 95°C for 5 min. Real-time PCR was performed using SYBR Green and primers for each gene. Primer sequences were synthesized by Euro ns Genomics (Tokyo, Japan) ( Table 1). The conditions of thermal cycling using a Rotor-Gene Q (Qiagen, Stanford, VA, USA) were 95°C for 5 min followed by 40 cycles of 95°C for 5 s and 60°C for 10 s. Collection and analyses of data were undertaken using the system software. mRNA expression levels of each gene were measured as CT threshold levels and normalized to those of eukaryotic elongation factor-1α. The cDNA sequence for rat NOS2 mRNA was deposited in the DNA Data Bank of Japan/European Bioinformatics Institute/GenBank under accession number AB250951.
Serum biochemical analyses. Serum ALT and AST levels were quanti ed using commercial kits. The serum levels of nitrite and nitrate (stable metabolites of NO) were measured using a commercial kit (Roche, Mannheim, Germany) according to the Griess method 59 .
Histopathological analyses. Excised liver specimens from the Sprague-Dawley rats were collected and xed in 10% formalin and embedded in para n. Sections of 3-5 µm in size were cut and stained with HE. Neutrophil in ltration was evaluated by staining with MPO using anti-MPO antibodies (A0398; DAKO, Glostrup, Denmark) before HE staining. Apoptotic bodies in the hepatocyte nuclei were detected by TUNEL staining using an in-situ Apoptosis Detection Kit (MK500; Takara Bio Inc., Kusatsu, Shiga, Japan). The number of MPO-and TUNEL-positive cells per square millimeter was counted by analysts who were blinded to the treatment arm.
Treatment of the cultured hepatocytes with OMZ. OMZ was dissolved in Williams' medium E under sterile conditions. On day 1 after cell culture, the hepatocytes were washed with fresh serum-and hormone-free Williams' medium E and incubated with IL-1β (1 nmol/L) in the same medium, either in the presence or absence of OMZ (dose range, 0.1-0.5 mmol/L).
Determination of NO production and LDH activity in the cultured hepatocytes. The amount of nitrite (a stable metabolite of NO) in the cell culture medium of the hepatocytes was measured using the Griess method 59 . Cell viability was measured on the basis of LDH activity using a commercial kit (Cytotoxicity LDH Assay Kit-WST; Dojindo Inc., Tokyo, Japan).
Western blotting in the cultured hepatocytes. Total cell lysates were obtained from the cultured hepatocytes using a previously described method with minor modi cations 4,46 . Immunostaining was performed with primary antibodies against mouse iNOS (A nity BioReagents, Golden, CO, USA), human inhibitor of κB alpha (IκBα; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and rat β-tubulin. Immunoreactive proteins were visualized by an enhanced chemiluminescence detection kit (GE Healthcare Biosciences, Piscataway, NJ, USA).
Transfection and luciferase assay in the cultured hepatocytes. Transfection of the cultured hepatocytes was performed using a previously described method 63 . Hepatocytes were cultured at 3 × 10 5 cells/dish (35 × 10 mm) in Williams' medium E with serum, dexamethasone, and insulin for 7 h before undergoing magnet-assisted transfection. Reporter constructs pRiNOS-Luc-SVpA (for detecting the transactivation of the NOS2 promoter) or pRiNOS-Luc-3′UTR (for detecting the stability of mRNA) (1 µg) and the cytomegalovirus promoter-driven β-galactosidase plasmid pCMV-LacZ (1 ng; internal control) were mixed with a magnet-assisted transfection reagent (1 µL; IBA Lifesciences, Göttingen, Germany) in fresh serumand hormone-free Williams' medium E (1.5 mL), followed by incubation with cultured cells. After a 15-min incubation period on a magnetic plate at room temperature, the medium was replaced with fresh Williams' medium E with serum. The cells were then cultured overnight and treated with IL-1β in the presence or absence of OMZ.
Statistical analyses. Quantitative results were obtained from three to four independent experiments for each of the various analyses, and the mean values and their standard deviations were calculated. Differences between groups and survival rates were identi ed using the Student's t-test and log-rank test, respectively (JMP® 14, SAS Institute Inc., Cary, NC, USA). P <0.05 was considered signi cant.