A phospholipase D2 inhibitor, CAY10594, ameliorates acetaminophen-induced acute liver injury by regulating the phosphorylated-GSK-3β/JNK axis

We examined the role of phospholipase D2 (PLD2) on acetaminophen (APAP)-induced acute liver injury using a PLD2 inhibitor (CAY10594). 500 mg/kg of APAP challenge caused acute liver damage. CAY10594 administration markedly blocked the acute liver injury in a dose-dependent manner, showing almost complete inhibition with 8 mg/kg of CAY10594. During the pathological progress of acute liver injury, GSH levels are decreased, and this is significantly recovered upon the administration of CAY10594 at 6 hours post APAP challenge. GSK-3β (Serine 9)/JNK phosphorylation is mainly involved in APAP-induced liver injury. CAY10594 administration strongly blocked GSK-3β (Serine 9)/JNK phosphorylation in the APAP-induced acute liver injury model. Consistently, sustained JNK activation in the cytosol and mitochondria from hepatocytes were also decreased in CAY10594-treated mice. Many types of immune cells are also implicated in APAP-induced liver injury. However, neutrophil and monocyte populations were not different between vehicle- and CAY10594-administered mice which are challenged with APAP. Therapeutic administration of CAY10594 also significantly attenuated liver damage caused by the APAP challenge, eliciting an enhanced survival rate. Taken together, these results indicate that PLD2 is involved in the intrinsic response pathway of hepatocytes driving the pathogenesis of APAP-induced acute liver injury, and PLD2 may therefore represent an important therapeutic target for patients with drug-induced liver injury.


Results
CAY10594 administration blocks liver damage of the APAP overdose-induced acute liver injury model. At first, we investigated the role of PLD2 on acute liver injury using a CAY10594. Injection of APAP (500 mg/kg) caused marked liver injury, which was measured by hematoxylin and eosin staining of the livers (Fig. 1A). APAP-induced liver injury was almost completely blocked by the administration of a CAY10594 in a dose-dependent manner (Fig. 1A). APAP-induced hepatocyte death was measured by the TUNEL assay. Data are expressed as the mean ± SEM (n = 5 for B,C, n = 7 for D). *P < 0.05; by t-test compared with treatment of APAP in vehicle control mice (B,C). Survival was analyzed by log-rank test. **P < 0.01 (D). Data are representative of two independent experiments. Hepatocyte apoptosis was induced by APAP, which was also markedly decreased in CAY10594-administered mice compared with vehicle-treated mice (Fig. 1B). The protective effects of the CAY10594 against hepatocyte apoptosis were strongly induced at 4 or 8 mg/kg (Fig. 1B).
We then measured aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activity in serum collected from APAP + vehicle or APAP + CAY10594 administered groups. Injection of APAP caused marked increase in both AST and ALT levels in serum, which were significantly decreased by CAY10594 in dose-dependent manners (Fig. 1C). Administration of 8 mg/kg CAY10594 almost completely blocked the increase of AST and ALT levels in APAP mice serum (Fig. 1C). However, PLD1 inhibitor (VU0155069) injection did not block increased AST and ALT levels upon APAP administration (data not shown). A lethal dose of APAP (750 mg/kg) elicited severe mice mortality within 72 hours (Fig. 1D). However, administration of CAY10594 significantly blocked the mortality, showing 100% survival for 72 hours (Fig. 1D). Taken together, these results indicate that PLD2 inhibitor, but not PLD1 inhibitor, shows strong protective effects against APAP-induced acute liver injury. CAY10594 treatment causes rapid recovery of APAP-induced decreased GSH levels but decreases the sustained activation of JNK in the liver. GSH levels have been previously reported to be decreased during acute liver injury 18 . Administration of an overdose of APAP rapidly depletes GSH from the liver of mice 18 . We also observed that GSH levels were strongly decreased at 3 or 6 hours post APAP challenge ( Fig. 2A). However, GSH levels were markedly recovered in CAY10594 administered mice at 3 or 6 hours after the APAP challenge ( Fig. 2A). Presumably, inhibition of PLD2 could repair liver injury or block hepatic cell death signaling during APAP consumption in the DILI mouse model. JNK phosphorylation has been reported to be mainly associated with increased acute liver injury 8 . In this study, we also found that the administration of 500 mg/kg APAP strongly induced JNK phosphorylation in the liver of vehicle-injected mice (Fig. 2B). At 2 hours post APAP challenge, JNK phosphorylation was apparent in the entire liver tissue, and this was markedly attenuated by CAY10594 treatment (Fig. 2B). In addition to JNK, extracellular signal-regulated kinase (ERK) is also known as a key mediator of inflammation and oxidative stress. Inhibition of ERK activation can attenuate APAP-induced liver injury 19 . Therefore, we investigated the role of PLD2 on APAP-induced MAPK phosphorylation. APAP treatment promoted the levels of phosphorylated-ERK in vehicle-injected mice (Fig. 2B). However, APAP-induced ERK phosphorylation was significantly attenuated by CAY10594 treatment (Fig. 2B).
Since phosphoryated GSK-3β mediates the early phase of APAP-induced liver injury, we examined the effects of CAY10594 on the phosphorylation of GSK-3β, which triggers mitochondrial dysfunction in the liver. Sustained activation of JNK in the cytosol can lead to translocation of activated JNK to the mitochondria, which is associated with the initiation of mitochondrial dysfunction that can lead to hepatocyte necrosis 8,20 . As expected, inhibition of PLD2 led to blockage of phosphorylated-JNK translocation to the mitochondria (Fig. 2C). The results suggest that inhibition of PLD2 could prevent mitochondrial dysfunction by inhibiting JNK translocation. Collectively, inhibition of PLD2 may show protective effects against acute liver injury by attenuating JNK phosphorylation in the APAP challenge. CAY10594 administration regulates cytokine and chemokine production but not immune cell recruitment after APAP overdose. Overdose drug-induced liver injury is caused by cytosolic reactive oxygen species and intrinsic signaling molecules, GSK-3β, JNK and mitochondrial reactive oxygen species, which subsequently induce hepatic cell death 9 . Extrinsic signaling mediators, pro-inflammatory cytokines, and damage-associated molecular patterns, have been also closely associated with APAP-induced liver failure 21 . In this study, we tested the effects of the CAY10594 on the expression levels of inflammatory cytokines and chemokines in APAP-induced acute liver injury. The administration of 500 mg/kg APAP strongly increased the levels of TNF-α, IL-6, IL-1β, and CCL2 in the sera from the vehicle administered group. However, the increase of pro-inflammatory cytokines and chemokine by APAP were restrained by treatment with CAY10594 (Figs. 3A).
Many types of immune cells are largely involved in hepatic injury through the secretion of pro-or anti-inflammatory cytokines such as TNF-α, IL-6, IL-1β and IL-10 22 . These cytokines serve to promote inflammation during APAP-induced hepatotoxicity 22 . Here, we investigated the effects of CAY10594 on the composition of immune cell population in the liver of APAP-challenged mice. The total number of myeloid cells, including neutrophils (CD11b hi Ly6G + ), mononuclear cells (CD11b hi Ly6C + ), and Kupffer cells (CD11b int F4/80 hi ) showed no significant difference in both vehicle-and CAY10594-treated mice after the APAP challenge ( Fig. 3B,C).

CAY10594 shows decreased APAP-induced hepatotoxicity in vitro.
Acute liver injury can be induced by hepatocyte apoptosis in response to APAP challenge in vitro. In this study, we tested the effects of CAY10594 on hepatocyte toxicity in response to APAP in primary hepatocytes. Isolated primary hepatocytes were stimulated with 10 mM APAP for several lengths of time. We measured LDH release, which is a biomarker for cytotoxicity in damaged cells. Hepatocytes were found to be damaged by APAP after 4 hours, and approximately 40% of the cells were damaged at 16 hours. However, CAY10594 treatment significantly decreased the www.nature.com/scientificreports www.nature.com/scientificreports/ APAP-induced hepatocyte damage (Fig. 4A). The addition of APAP into hepatocytes markedly increased the death of hepatocytes showing PI + cells, which were decreased by CAY10594 treatment (Fig. 4B). In vitro stimulation of primary hepatocytes with APAP (10 mM) also induced GSK-3β phosphorylation (Fig. 4C). Addition of CAY10594 markedly blocked APAP-induced GSK-3β phosphorylation in primary hepatocytes (Fig. 4C). Taken together, the results suggest that PLD2 plays a key role in the intrinsic response pathway of hepatocytes to APAP-induced hepatotoxicity.

CAY10594 shows therapeutic effects against APAP overdose-induced acute liver injury. Clinically,
it is important to develop therapeutic agents against liver damage due to drug overdose. Therefore, we investigated whether CAY10594 shows therapeutic effects against APAP-induced acute liver injury by administering CAY10594 after the APAP challenge. Administration of CAY10594 (8 mg/kg) 3 h post APAP challenge markedly blocked APAP-induced liver damage, which was measured by histological analysis after hematoxylin and eosin staining (Fig. 5A). Mouse survival rate was also highly increased upon 8 mg/kg of CAY10594 administration at 3 h after APAP challenge, showing an 50% survival rate, respectively (Fig. 5B). Under the same experimental conditions, the vehicle administered group after APAP challenge showed a 12.5% survival rate at 72 h (Fig. 5B). These results strongly indicate that CAY10594 has therapeutic effects against APAP-induced acute liver injury.

Discussion
PLD regulates lipid metabolism by hydrolyzing phosphatidylcholine to PA and choline 23 . PLD2 has been known to have a far higher basal activity than PLD1, and PLD2 mediates various unique protein interactions 24 . Moreover, PLD2 has different metabolic properties from PLD1, and most production of cyclic PA is To determine hepatocyte death, hepatocytes were stained against propidium iodide (PI) and the number of PI positive cells were counted at x200 magnification by fluorescence microscopy. Scale bar, 100μm (B). (C) Phosphorylation of GSK-3β and total GSK-3β from primary hepatocytes was determined by Western blot analysis. All data are represented as the mean ± SEM. The data are representative of two independent experiments. **P < 0.01; by t-test compared with APAP-treated cells given the vehicle (A).
www.nature.com/scientificreports www.nature.com/scientificreports/ dependent on PLD2 rather than PLD1 25 . Previously, it was reported that accumulation of PA can enhance the regeneration of liver after APAP-induced liver injury 17 . A separate report demonstrated that inhibition of PA production with FSG67, an inhibitor of glycerol 3-phosphate acyltransferase, did not affect the severity of APAP-induced acute liver injury at the injury phase 17 . However, in this study, we have observed preventive effects of CAY10594, showing 100% survival against the APAP-induced acute liver injury mice model (Fig. 1D). CAY10594 also induced strong therapeutic effects in APAP-challenged mice (Fig. 5). The results suggest an important role for PLD2, which crucially mediates APAP-induced liver injury. Because PLD2 has basal activity, administration of CAY10594 would block the generation of PA, the product of PLD2 enzymatic activity, in an experimental acute liver injury model. Therefore, it might reasonable to assume that the protective and therapeutic effects of CAY10594 in an acute liver injury model would be mediated by blocking the generation of PA. Based on our findings, we suggest that PA may have a pathological role in the disease progress of acute liver injury.
In APAP-induced acute liver injury, phosphorylation of JNK is a central player in inducing hepatic damage 26 . Sustained phosphorylation of JNK and translocation of JNK to the mitochondria exacerbates mitochondrial dysfunction, and causes hepatic necrosis. Moreover, previous reports already demonstrated that JNK inhibitors such as leflunomide, SP600125 or D-JNKI1 inhibited APAP-induced liver injury 27,28 . JNK phosphorylation can be differentially controlled in two different phases, the early and late phase. Phosphorylated-GSK-3β/JNK axis is a major source of APAP-induced liver injury at the very early phase. The early phase of JNK activation is regulated by GSK-3β, which can activate JNK through MEKK-1 or mixed lineage kinase-dependent pathways. On the other hand, the late phase of JNK activation is believed to mediate apoptosis signal-regulating kinase 1. Recently, p53 up-regulated modulator of apoptosis has been reported to be induced downstream of JNK and mediate APAP-induced necrosis and liver injury 29 . GSK-3β silencing has been reported to have beneficial effects against APAP-induced liver injury. Therefore, we tested whether PLD2 modulates phosphorylated-GSK-3β in the early phase following the APAP challenge. CAY10594 almost completely blocked GSK-3β phosphorylation when compared with vehicle administered mice (Fig. 2B). Our results suggest that CAY10594 may block JNK phosphorylation at the early phase by inhibiting GSK-3β phosphorylation (Fig. 6). At about 30 min to 6 h after the APAP challenge, NAPQI, the metabolite of APAP, induces the dysfunction of mitochondria and production of reactive oxygen species that initiate liver injury through JNK phosphorylation. Therefore, the regulation of reactive oxygen species is important in APAP-induced liver injury. These reactive oxygen species can be removed by GSH, but APAP overdose induces GSH depletion in an hour 30 . We found that CAY10594 prevents phosphorylation of JNK (Fig. 2B) and the regulation of GSK-3β phosphorylation, that might contribute to this mechanism (Fig. 2B). APAP challenge-induced GSH depletion was rapidly recovered upon CAY10594 administration ( Fig. 2A). Since acute liver injury induced by APAP overdose has been reported to be mediated by the generation of excess oxidative stress 31 , it is important to regulate the expression of anti-oxidant genes in the liver to endure oxidative stress 32 . We also tested for a possible effect of CAY10594 on the regulation of anti-oxidant genes. The expression of several anti-oxidant genes, including FGF21, Nrf2, HO-1 and NQO1, were highly expressed upon APAP overdose. However, anti-oxidant gene expression was not increased by CAY10594, with the expression levels remaining low compared to vehicle mice at 6 hours post APAP challenge (data not shown). Consequently, it appears that CAY10594-induced protective effects against APAP-induced liver injury is not mediated by upregulation of www.nature.com/scientificreports www.nature.com/scientificreports/ anti-oxidant genes. Collectively, our results suggest that CAY10594 may modulate early hepatopathology to prevent APAP-induced liver injury by rapid recovery of GSH levels without affecting anti-oxidant gene expression.
The pathological progress of acute liver injury caused by APAP administration can be mediated by early recruitment and activation of several immune cell types. This immune cell infiltration to the liver in a short period of time is regulated by several damaged-associated molecular patterns such as the S-100A families and HMGB-1 from hepatocytes 21 . Immune cells infiltrated into the liver can induce the secretion of inflammatory cytokines and chemokines, which contributes greatly to liver damage. In our study, we found that administration of CAY10594 did not affect the number of CD11b hi Ly6C + , CD11b hi Ly6G + , and CD11b int F4/80 + cells in APAP-challenged mice (Fig. 3C). These results suggest that PLD2 is not involved in immune cell population changes during APAP-induced liver damage.
If liver damage is suppressed at early time phases, inflammatory stimuli would not be secreted. TNF-α and IL-1β are known as major pro-inflammatory cytokines which play roles in DILI. TNF-α induces mitochondrial dysfunction via the TNF-receptor and induces hepatocyte death through caspase activation 33 . IL-1β is also an inflammatory cytokine that induces liver damage and is produced by activation of the inflammasome 34 . CAY10594 efficiently reduced secretion of pro-inflammatory cytokines and the CCL2 chemokine (Fig. 3A). Since we also observed that CAY10594 administration suppresses liver damage at the early phase, our results suggest that the protective effects of CAY10594 against acute liver injury would be directly mediated by regulation of hepatocytes but not by regulation of immune cell activity. Our findings of the direct inhibitory effects of CAY10594 on the hepatotoxicity and GSK-3β phosphorylation caused by APAP treatment (Fig. 4) further support our idea that CAY10594 may show protective effects against acute liver injury by modulating hepatocyte activity (Fig. 6). On the action mode of CAY10594, the inhibitor can inhibit GSK-3β/JNK axis to regulate ROS generation in mitochondria of hepatocytes (Fig. 6).
In APAP-induced liver injury, pharmacological treatment options are highly limited. For APAP overdose patients, a prescription of N-acetylcysteine is the only treatment since the 1970's 35 . Metformin has also been revealed as a therapeutic candidate for APAP-induced liver injury, however, it must be administered at a concentration higher than 300 mg/kg like N-acetylcysteine. However, in our study, we demonstrate that CAY10594, which has been highlighted as being therapeutic without unacceptable clinical side effects 14 , has great therapeutic performed in accordance with the Korea Food and Drug Administration guidelines. Protocols were approved by the Animal Care and Use Committee, Sungkyunkwan University (Suwon, Korea). Mice were fasted for 16 hours before APAP injection. APAP (500 mg/kg) was administered with oral gavage in mice. CAY10594 (N-[2-(4-oxo-1-phenyl-1,3,8-triazaspiro [4,5]dec-8-yl)ethyl]-2-naphthalene carboxamide) 23 was dissolved in 1% DMSO and intraperitoneally administered to mice 30 minutes prior to APAP injection for examining protective effects or after 3 hours from APAP challenge for investigating therapeutic effects of CAY10594.
Histopathology and immunohistochemistry. Twelve hours after APAP (500 mg/kg) was administered with oral gavage in mice, the mice were euthanized, and their livers were isolated and fixed with 10% neutral buffered formalin. Liver tissues were embedded in paraffin and stained with hematoxylin and eosin for morphological analysis. The TUNEL assay was performed on paraffin-embedded tissue sections using a standard histological protocol. In brief, section slides were incubated using terminal deoxynucleotidyl transferase dUTP nick end labeling fluorescein for 1 hour at 37 °C in dark. The samples were washed 3 times in PBS for 3 minutes each and incubated with peroxidase conjugated anti-fluorescein antibody for 30 min at 37 °C in the dark. TUNEL positive cells were detected by diaminobenzidine solution. Liver histology and apoptotic cells were observed under a Leica ICC50 HD microscope.

Measurement of serum AST and ALT.
ALT and AST levels were measured using an ALT activity kit and AST activity assay kit (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer's instructions. GSH measurements. Liver tissue (33 mg) was homogenized with 5% metaphosphoric acid (500 μl) and centrifuged at 12,000 rpm for 10 minutes at 4 °C. GSH levels were measured from the supernatants using a glutathione (GSSG/GSH) detection kit according to the manufacturer's instructions (Enzo Life Sciences Inc, Farmingdale, NY, USA). Total GSH levels were measured according to the net slope of the standard curve.
Mitochondria isolation. Liver tissue (35 mg) was homogenized and used for the fractionation of mitochondria using a mitochondria isolation kit (Thermo Fisher Scientific TM , Waltham, MA USA) according to the manufacturer's instructions.
Cytokine measurement. Blood samples were collected from acute liver injury mice 12 hours post-APAP treatment. Cells were removed from the harvested biofluids by centrifugation (12,000 rpm for 1 minute), and the levels of cytokines were measured by ELISA (eBioscience Inc., San Diego, CA, USA).
Flow cytometry. Single cell suspensions of cells from collagenase treated liver samples were generated and blocked with FcγIII antibody and stained for CD11b (M1/70), Ly6G (1A8), Ly6C (HK1.4), F4/80 (BM8) on the cell surface. All those primary antibodies were purchased from eBioscience (San Diego, CA, USA). Cells were analyzed by flow cytometry and data were analyzed by FlowJo_V10.

Isolation of primary hepatocytes and in vitro experiments with APAP treatment. Primary
hepatocytes were isolated from wild-type mice following liver-specific perfusion with 50 ml of a buffer containing 66.7 mM NaCl, 6.7 mM KCl, 50 mM HEPES, 4.8 mM CaCl 2 2H 2 O, collagenase type IV. Cells were centrifuged at 500 rpm for 4 minutes. Primary hepatocytes were separated from dead cells and other cell types by Percoll gradient centrifugation (1,250 rpm for 5 minutes) and seeded into 6-well culture dishes (Thermo Fisher Scientific), and then cells were stimulated with 5~20 mM APAP with time dependency. Stimulated cells were fixed with 4% formaldehyde at room temperature for 10 minutes. Next, the cells were washed twice with 1 × PBS and stained with PI in 1 × PBS at room temperature for 10 minutes. PI positive cells were visualized with a fluorescence microscope (KI-2000F, Korea Lab Tech, Seongnam, Korea) and images were analyzed with OptiView 3.7.
Cytotoxicity assay. Cytotoxicity was assessed by detection of the enzyme lactate dehydrogenase (LDH) in the supernatant of the cell culture. The assay was performed using the Cytotoxicity Detection Kit (Promega, Madison, WI, USA) according to the manufacturer's protocol. The LDH concentration in 50 µl of the cell culture supernatant was determined at a wavelength of 490 nm. Cells treated with lysis solution served as a reference for the maximum possible LDH release (100%, high control). The relative LDH release of a given sample is then defined as the ratio of LDH measured in the supernatant of the sample and the high control value with LDH values under 10% regarded as a nontoxic effect level.