LPCAT3/LPLAT12 deficiency in the liver ameliorates acetaminophen‐induced acute liver injury

Acetaminophen (APAP) is a double‐edged sword, mainly depending on the dosage. A moderate dose of APAP is effective for fever and pain relief; however, an overdose induces acute liver injury. The mechanism underlying APAP‐induced acute liver failure is unclear, and its treatment is limited. A recent report has shown that several oxidized phospholipids are associated with APAP‐induced acute liver failure. Lysophosphatidylcholine acyltransferase 3 (Lpcat3, Lplat12), which is highly expressed in the liver, preferentially catalyzes the incorporation of arachidonate into lysophospholipids (PLs). In the present study, we investigated the roles of Lpcat3 on APAP‐induced acute liver injury using liver‐specific Lpcat3‐knockout mice. Hepatic Lpcat3 deficiency reduced the degree of APAP‐induced necrosis of hepatocytes around Zone 3 and ameliorated the elevation of hepatic injury serum marker levels, and prolonged survival. Lipidomic analysis showed that the accumulation of oxidized and hydroperoxidized phospholipids was suppressed in Lpcat3‐knockout mice. The amelioration of APAP‐induced acute liver injury was due not only to the reduction in the lipid synthesis of arachidonic acid PLs because of Lpcat3 deficiency, but also to the promotion of the APAP detoxification pathway by facilitating the conjugation of glutathione and N‐acetyl‐p‐benzoquinone imine. Our findings suggest that Lpcat3 is a potential therapeutic target for treating APAP‐induced acute liver injury.


| INTRODUCTION
Acetaminophen (also known as N-acetyl-p-aminophenol [APAP] or paracetamol) is the most widely used drug for the relief of fever and pain.The administration of APAP at therapeutic doses is generally safe; however, an overdose can cause severe liver injury, leading to acute liver failure.Recently, APAP has been the most common cause of acute liver failure in the United States and United Kingdom. 1,2t therapeutic doses, approximately 60% of APAP is glucuronidated and approximately 35% is sulfated in the liver and then excreted through the kidneys.A minor fraction is metabolized to N-acetyl-p-benzoquinone imine (NAPQI) via cytochrome P450 2E1.NAPQI is a highly toxic metabolite that is readily detoxified by conjugation with reduced glutathione (GSH) and excreted in the urine as mercapturic acid.However, when APAP overdose exceeds the capacity for glucuronidation, sulfation, and GSH conjugation, it leads to an overaccumulation of NAPQI.Excess NAPQI binds to mitochondrial proteins, causing oxidative stress inducing hepatotoxicity.Currently, no effective treatment is available for APAP-induced hepatotoxicity except for administration of N-acetyl-L-cysteine (NAC), a precursor of GSH. 3 However, NAC is only effective when administered relatively early after acetaminophen overdose.Therefore, it is necessary to understand the underlying mechanisms to develop novel preventive and therapeutic approaches for APAP-induced liver injury.
In addition to technological advancements in chromatography and mass spectrometry, the identification of genes associated with lipid synthesis has provided new insights into the changes in lipid profiles.While phospholipids are newly synthesized from glycerol 3-phosphates through de novo Kennedy's pathway, 4 they gain fatty acid diversity and sn-1/2 asymmetric distribution via the Lands' remodeling cycle. 57][8][9] LPCAT3 prefers polyunsaturated fatty acyl-CoAs (18:2-acyl-CoA, 20:4-acyl-CoA) as acyl donors and catalyzes the incorporation of linoleate and arachidonate into lysophosphatidylcholine (LysoPC), with small amounts of lysophosphatidylethanolamine (LysoPE) and lysophosphatidylserine (LysoPS).In both humans and rodents, LPCAT3 is ubiquitously expressed; however, it is highly expressed in the liver and small intestine.LPCAT3 is the most abundant LPCAT member in the liver, accounting for 80%-90% of the total LPCAT activity. 10Previously, we and others have reported that global Lpcat3-knockout mice exhibit reduced arachidonoyl PC/ PE and impaired lipoprotein secretion. 10,11cently, it has been reported that 70 oxygenized PCcontaining epoxy and hydroperoxide groups are generated in the early phase of APAP-induced acute liver injury, using high-resolution mass spectrometry-based non targeted analysis. 12Oxidized arachidonate PC, oxidized linoleonate PC, and oxidized docosahexaenoate PC, which were among 70 oxygenized PC identified, were upregulated after APAP. 12However, it has not been clear that which oxygenized PC plays an important role during APAP-induced liver injury.Therefore, in the present study, we focused on arachidonate PC and used liver-specific Lpcat3-knockout mice (Lpcat3 L-KO) to investigate whether differences in phospholipid acyl chain composition influence acute liver failure.

| Animal experiments
Alpha-fetoprotein (Afp)-Cre transgenic mice and Lpcat3(Lplat12) flox/flox mice (Acc No. CDB0653K) 10 (all with C57BL/6J background) were crossed to generate liver-specific Lplat12-knockout mice.Afp-Cre mice were kindly donated by Dr. Fuyuki F Inagaki and Dr. Klaus H. Kaestner.APAP (Tokyo Chemical Industry Co., Ltd., Japan) was dissolved in warm saline solution.The mice were intraperitoneally injected with APAP (200 or 400 mg/kg) or vehicle after fasting.All animal experiments were performed following the guidelines of the Animal Research Committee of the National Center for Global Health and Medicine (approval number: 21034).

| Measurement of hepatic TG and cholesterol levels
Hepatic cholesterol and TG levels were measured enzymatically using FUJIFILM WAKO Lab Assay kits (FUJIFILM WAKO Pure Chemical Industries, Osaka, Japan), according to the manufacturer's instructions.

| Histological analysis
Frozen liver sections (8 μm) were prepared using a Microm HM525NX cryostat (Thermo Fisher Scientific), mounted on APS-coated glass slides (Matsunami Glass, Inc.), and fixed in 4% paraformaldehyde.Sections were then stained with hematoxylin and eosin (H&E), and photomicrographs were obtained using KEYENCE BZ-X710, X700 (KEYENCE, Osaka, Japan).The degree of liver injury was quantified based on a previously reported histopathological score. 13

| Quantitative real-time polymerase chain reaction
Hepatocytes and cholangiocytes were isolated from the liver according to the methods of our previous report. 14otal RNA was extracted from the liver using RNeasy Mini kit (QIAGEN, Venlo, Germany) according to the manufacturer's protocol.cDNA was synthesized using SuperScript III Reverse Transcriptase (Thermo Fisher Scientific, Waltham, MA).Real-time quantitative polymerase chain reaction (PCR) was performed using Fast SYBR Green Master Mix and Step One Plus Real-Time PCR System (Thermo Fisher Scientific).The expression levels of the target genes were normalized to those of mouse 36B4.The primer sequences included: m36B4-Forward (ctgagattcgggatatgctgttg), m36B4-Reverse (aaagcctggaagaaggaggtctt), mLpcat3-Forward (agatggaattcctcattgttatcgt), and mLp-cat3-Reverse (gaagggctgtagggcagtga).

| Lipid composition analysis
For phospholipid analysis, methanol was added to crushed frozen liver tissues and centrifuged.The supernatant was diluted with methanol to adjust the concentration to 0.1 mg tissue/mL.Individual liver phospholipids were measured by liquid chromatography/mass spectrometry (LC-MS/MS) using a Nexera UHPLC system and triple quadrupole mass spectrometer LCMS-8060 (Shimadzu Corporation).Products were separated on a ACQUITY UPLC BEH C8 column ( The peak areas of all the detected species were summed to obtain the total area.Additionally, the individual peak areas were normalized to the total area and indicated as a percentage of the total area. For fatty acid analysis using gas chromatography with flame ionization detection (GC-FID), we used a previously reported method. 15Briefly, the phospholipid fraction was separated after both fractions of neutral lipids and free fatty acids were removed from the lipids, which were extracted from the tissue using the Bligh and Dyer method with solid-phase extraction using InertSep NH2 aminopropyl columns (GL Sciences Inc., Tokyo, Japan).The fatty acids were then methylated and purified using fatty acid methylation and fatty acid methyl ester purification kits (Nacalai Tesque, Kyoto, Japan).Individual fatty acid-derived PLs were measured using a GC-2010 Plus system (Shimadzu Corporation) equipped with a FID.Fatty acid methyl ester samples were separated on a capillary column (FAMEWAX, 30 m, 0.25 mm ID, 0.25 μm, Restek Corporation, Bellefonte, PA).Each fatty acid was identified and quantified using a mixture of fatty acid methyl ester standards (Supelco 37 Component FAME Mix and DPA (n-3) (Sigma-Aldrich), DPA (n-6) (Nu-Chek Prep Inc., Elysian, MN, USA), and DTA (n-6) (Cayman Chemical, Michigan, USA) for calibration.
For eicosanoid analysis using a triple quadrupole mass spectrometer LCMS-8060 (Shimadzu Corporation), the lipids extracted from liver tissues were purified by solid-phase extraction using an Oasis HLB column (Waters Corporation, Milford, MA, USA).The measurement and analysis methods have been described previously. 16

| Protein isolation
For the detection of LPCAT3 protein from the whole liver, tissue samples were homogenized and lysed in icecold T100 buffer (100 mM Tris-HCl (pH 7.4), 300 mM sucrose, and 1 × cOmplete (Roche Applied system, Basel, Switzerland)) using a sonicator (OHTAKE Works, Japan) and then centrifuged at 800× g for 10 min.To isolate the membrane fractions, the supernatant was recentrifuged at 100 000× g for 60 min.The pellet was suspended in TSE buffer (20 mM Tris-HCl [pH 7.4]), 300 mM sucrose, and 1 mM ethylenediaminetetraacetic acid).Protein concentrations were determined using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA).

| Measurement of LPCAT activity
The method for measuring LPCAT activity has been described previously. 10Briefly, 0.01 μg/tube proteins were added to a mixture of 25 μM deuterium-labeled 16:0 LPC, 1 μM 16:0-CoA, 1 μM 18:1-CoA, 1 μM 18:2-CoA, 1 μM 20:4-CoA, and 1 μM 22:6-CoA and incubated at 37°C for 10 min.After the reaction was stopped by adding 300 μL of chloroform/methanol, DLPC was added as an internal standard.Lipids were extracted from the mixtures using the Bligh and Dyer method and measured by LC-MS.The products were separated on an ACQUITY UPLC BEH C8 column (1.7 μm, 2.1 × 30 mm, Waters Corporation, Milford, MA) using a linear gradient of solvent B (acetonitrile, Wako) over solvent A (20 mM NH 4 HCO 3 /water, Wako), with an ACQUITY ultra performance liquid chromatography (UPLC) system (Waters Corporation, Milford, MA).The flow rate was set at 800 μL/min.The gradient began with 55% solvent B, linearly increased to 95% solvent B over 4.5 min, and was maintained for 1.5 min.The detection was performed using a TSQ Vantage triple-stage quadrupole mass spectrometer (Thermo Fisher Scientific) with selected reaction monitoring.The transition was [M + H] + → 184.1 for PC.The resulting sample signals were compared with the calibration curves of non-labeled standards.

| Western blot analysis
The abovementioned membrane protein fraction was used as the sample analyzed by western blotting, which was performed using 10% sodium dodecyl sulfate-polyacrylamide gels and electrophoretically transferred to nitrocellulose (GE Healthcare) and polyvinylidene fluoride membranes (Merck Millipore) using a semi-dry transfer cell (Bio-Rad Laboratories).Ponceau S (Sigma-Aldrich) was used to confirm protein transfer from the gel to the membrane.The mouse anti-LPCAT3 antibody used in this study has been previously described. 10Anti-mouse horseradish peroxidase (1:1000) (GE Healthcare) was used as the secondary antibody.An enhanced chemiluminescence-selected western blot detection system (GE Healthcare) was used for chemiluminescence, which was detected using ImageQuant LAS500 (GE Healthcare).

| Measurement of malondialdehyde and glutathione levels
Malondialdehyde (MDA) levels in the liver tissue were measured using an OxiSelect TBARS Assay kit (Cell Biolabs Inc., San Diego, CA, USA) according to the manufacturer's instructions.The concentrations of GSH and oxidized glutathione (GSSG) in liver tissues were quantified using a GSSG/GSH Quantification Kit (DOJINDO LABORATORIES, Kumamoto, Japan) according to the manufacturer's instructions.

| Analysis of oxidized phospholipids
Methanol, isopropanol, and chloroform of UPLC/MS quality were obtained from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan).Ultrapure water was obtained from a Milli-Q water system (Millipore, Milford, MA, USA).
For lipid preparation, lipidome analysis was performed according to the Lipidome lab Oxidized Lipid Scan package (Lipidome lab, Akita, Japan) using liquid chromatograph orbitrap mass spectrometry (LC-OrbitrapMS).Total lipids, including oxidized phospholipids, were extracted from 10 mg mouse liver samples using the modified Bligh-Dyer method with 5% dibutylhydroxytoluene as an antioxidant.An aliquot of the lower/organic phase was evaporated to dryness under N2 and the residue was dissolved in methanol for LC-MS/MS measurements.
We performed instrumental analysis and data processing as described below.LC-electrospray ionization-MS/MS analysis was performed using a Q Exactive Plus mass spectrometer equipped with an UltiMate 3000 LC system (Thermo Fisher Scientific).Samples were separated on L-column3 C18 metal-free column (2.0 μm, 2.0 mm × 100 mm i.d.) at 40°C using a gradient solvent system: mobile phase A (isopropanol/methanol/water (5/1/4 v/v/v) supplemented with 5 mM ammonium formate and 0.05% ammonium hydroxide (28% in water))/ mobile phase B (isopropanol supplemented with 5 mM ammonium formate and 0.05% ammonium hydroxide (28% in water)) ratios of 60%/40% (0 min), 40%/60% (0-1 min), 20%/80% (1-9 min), 5%/95% (9-11 min), 5%/95% (11-22 min), 95%/5% (22-22.1 min), 95%/5% (22.1-25 min), 60%/40% (25-25.1 min), and 60%/40% (25.1-30 min).The injection volume was 10 μL and the flow rate was 0.1 mL/min.The heated electrospray ionization (HESI-II) source conditions were as follows: ionization mode, positive or negative; sheath gas, 60 arbitrary units; auxiliary gas, 10 arbitrary units; sweep gas, 0 arbitrary units; spray voltage, 3.2 kV in positive and −3.0 kV in negative mode; heater temperature, 325°C; ion transfer capillary temperature, 300°C in positive and −320°C in negative mode; and S-lens RF level, 50.The Orbitrap mass analyzer was operated at a resolving power of 70 000 in full-scan mode (scan range 200-1, 800 m/z in positive and 190-1, 800 m/z in negative mode; automatic gain control (AGC) target 1e6 in positive and 3e6 in negative mode) and resolving power of 17 500 in positive and 35 000 in negative mode in the Top 20 data-dependent MS2 mode (stepped normalized collision energy 20, 30, and 40; isolation window 4.0 m/z; AGC target 1e5) with dynamic exclusion setting of 10.0 s.Post-processing of the raw data files for oxidized phospholipids were done using Xcalibur 4.2.47 software (Thermo Fisher Scientific).Identification of oxidized phospholipids was carefully annotated manually after the interpretation of the molecular substructures by checking typical product ions derived from the PC head group and two fatty acyl groups in negative and positive ion modes.The annotation method used in this study corresponds to "Fatty Acyl/Alkyl Level" defined by the Lipidomics Standard Initiative. 17In addition, in this method, biological matrix effects cannot be normalized to all detected peaks because it is impossible to prepare appropriate internal standards corresponding to all detected peaks.Relative values were calculated as the ratio of the chromatographic peak area of each analyte to that of the internal standard.

| Statistical analysis
Statistical analyses were performed using GraphPad Prism, version 8 (GraphPad Software, La Jolla, CA, USA).Student's t-test was used to compare two groups.One-way analysis of variance (ANOVA), followed by Tukey's test, was used for comparisons among multiple groups.The cumulative probability of survival was analyzed using the Kaplan-Meier survival curve analysis, and the results were compared using the log-rank test.A p-value of <.05 was considered statistically significant.

Lpcat3-knockout mice
Global Lpcat3-knockout mice are neonatally lethal owing to hypoglycemia and malnutrition caused by severe damage to intestinal epithelial cells. 10,11Hence, to examine the role of LPCAT3 in the adult liver, we generated conditional Lpcat3-knockout mice by mating alpha-fetoprotein (AFP)-Cre mice and LPCAT3 flox/flox mice.Recent report revealed that biliary epithelial cells undergo direct conversion to hepatocytes using a murine lineage tracing model. 18A TAA-induced liver injury model was used in their experiments, and it was unclear whether direct conversion is also seen in the APAP liver injury model.However, considering the possibility of occurring the trans-differentiation, we used AFP-Cre mice instead of Alb-Cre mice in the present study.AFP-Cre − LPCAT3 flox/flox mice were used as controls.Unlike the global Lpcat3-knockout mice, Lpcat3 L-KO mice are not neonatally lethal and survive to adulthood.As AFP is expressed in liver progenitor cells during development, knockout of LPCAT3 using AFP-Cre mice was expected to be achieved in hepatocytes and cholangiocytes.To confirm this, we isolated hepatocytes and cholangiocytes from adult control mice and adult AFP-Cre-driven KO mice, and the expression level of LPCAT3 were examined.In AFP-Cre-driven LPCAT3 KO mice, the expression levels of LPCAT3 were significantly reduced in both hepatocytes and cholangiocytes, compared to control mice (Figure S1A).These deficiencies led to the result that more than 91% of LPCAT3 mRNA expression could be suppressed in the whole liver of LKO mice (Figure S1B).On the other hand, in another group's study using Alb-Cre mice, LPCAT3 expression in the liver was suppressed by about 70%. 11Moreover, the amount of LPCAT3 protein in the liver was significantly reduced in L-KO mice compared with that of control mice (Figure S2), without apparent compensatory upregulation of other LPCAT family members, such as Lpcat1 (Lplat8) and Lpcat2 (Lplat9) (Figure S3).To examine whether the enzymatic activity of lysophosphatidylcholine acyltransferase was lost in L-KO mice, LPCAT activity in liver lysates was measured using LC-MS.In particular, LPCAT activity for PLPC (16:0/18:2) and PAPC (16:0/20:4) was significantly decreased in the livers of L-KO mice compared with that of control mice (Figure S4).These results indicate that Lpcat3 deficiency in the liver inhibits the incorporation of arachidonate and linoleate into LPC.
We also examined the phenotype of Lpcat3 L-KO mice under physiological conditions.LC-MS lipidomic analysis revealed that the total PC content in the liver did not differ significantly between control and L-KO mice (Figure S5, right upper).However, the PC composition differed between the two groups.Specifically, the PC 36:4 (total carbon number: double bonds of sn-1 and sn-2 fatty acids), PC 36:5, and PC 38:4 fractions, which may contain arachidonic acid, were markedly reduced, whereas those of PC 32:0, PC 32:1, PC 34:3, PC 34:1, PC 36:1, PC 38:6, PC 40:6, and PC 40:7 slightly increased.LPCAT3 deficiency reduced the incorporation of linoleate into LPC in vitro; however, the fractions of PC34:2 and PC36:2, which may contain linoleate, were similar between the two groups in vivo (Figure S5).A similar trend was observed for the PE (Figure S6).
Serum lipid analysis showed that the levels of total cholesterol, TG, PL, and NEFA did not differ between the two groups (Figure S7).There were no significant differences in the hepatic cholesterol or TG levels between the two groups (Figure S8).Similarly, serum AST and ALT levels were almost similar in Lpcat3 L-KO and normal mice (Figure S9).These results indicate that there were few differences between control and Lpcat3 L-KO mice under physiological conditions, except that PLs containing arachidonate were decreased in Lpcat3 L-KO mice.

| Deficiency of Lpcat3 in liver mitigates APAP-induced liver injury
To elucidate the role of Lpcat3 in acute liver injury induced by APAP, we monitored survival after the intraperitoneal injection of high-dose APAP (400 mg/kg) at a single time point in control and Lpcat3 L-KO mice.All control mice died within 60 h of APAP administration; however, half of Lpcat3 L-KO mice survived 96 h after injection (Figure 1A).Therefore, to investigate differences in sensitivity to APAP, mice were examined 3 h after APAP (200 mg/kg) injection.Serum AST and ALT levels were elevated by APAP administration in both Lpcat3 L-KO and control mice; however, the degree of elevation in Lpcat3 L-KO mice was significantly milder than that in control mice (Figure 1B).While severe centrilobular zonal necrosis was observed in the livers of APAP-treated control mice, significantly fewer necrotic areas were observed in the livers of APAP-treated Lpcat3 L-KO mice (Figure 1C).The histopathology score, which reflects the severity of hepatotoxicity, was significantly lower in APAP-treated Lpcat3 L-KO mice than that in APAP-treated control mice (Figure 1D).These results indicated that Lpcat3 deficiency ameliorated APAP-induced liver injury.

| Lpcat3 deficiency suppresses the accumulation of the end products of lipid peroxidation after APAP injection
2][3] Subsequently, we focused on the end products of lipid peroxidation, malondialdehyde (MDA), and 8-iso PGF2α, 19 to clarify the mechanism by which Lpcat3 deficiency attenuates APAP-induced acute liver injury.Intracellular MDA was generated from a lipid peroxide precursor using a thiobarbituric acid assay.8-iso PGF2α is a biomarker of nonenzymic oxidation products of arachidonic acid.Both hepatic MDA and 8-iso PGF2α levels in APAP-treated Lpcat3 L-KO mice were lower than that in APAP-treated control mice (Figure 2A,B).These data indicated that APAP-treated Lpcat3 L-KO mice accumulated fewer end products of lipid peroxidation than APAP-treated control mice, leading to the attenuation of acute liver injury.

| Deficiency of Lpcat3 suppresses the accumulation of oxidized arachidonate PL and peroxidized arachidonate PL after APAP injection
To investigate which fatty acids in the liver were associated with the lower accumulation of lipid peroxidation endproducts in APAP-treated Lpcat3 deficient mice, the ratios of saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids (PUFAs) derived from PL in the APAP-treated liver were analyzed using GC-FID.The PUFA ratio was almost the same in APAP-treated control and Lpcat3 L-KO mice (Figure 3A).However, analysis of the fatty acid composition of PL-derived PUFAs revealed that arachidonic acid (C20:4) and eicosapentaenoic acid (C20:5) levels were significantly decreased and docosahexaenoic acid (C22:6) was significantly increased in Lpcat3 L-KO mice after APAP injection compared with control mice after APAP injection (Figure 3B).
Recently, a relationship between iron-dependent cell death (ferroptosis) and APAP-induced acute liver failure was reported. 20,21In another report, genome-wide CRISPER-based genetic screening and microarray analysis of cancer tissues suggested that Lpcat3 is an essential component of ferroptosis associated with arachidonate and peroxidized arachidonate PE via 15LOX/PEBP1. 20,21herefore, we examined the content of arachidonate PE, oxidized arachidonate PE, and peroxidized arachidonate PE using LC-MS/MS.PE 36:4 (PE 16:0/20:4) and PE 38:4 (PE 18:0/20:4) were significantly reduced in APAPtreated Lpcat3 L-KO mice compared with APAP-treated control mice (Figure 3C).In addition, the production of oxidized arachidonate PE and peroxidized arachidonate PE was significantly suppressed in APAP-treated Lpcat3 L-KO mice compared with APAP-treated control mice (Figure 3D).Arachidonate PC showed a similar tendency as arachidonate PE (Figure 3E,F).Our results suggested that Lpcat3/Lpcat12 deficiency reduces the production of oxidized and peroxidized arachidonate PLs, possibly leading to the amelioration of ferroptosis-related liver injury.

| Deficiency of Lpcat3 after APAP injection relates to the accumulation of GSH level regardless of phospholipid composition
To elucidate whether ferroptosis is the sole reason for reduced APAP-induced liver injury in Lpcat3 L-KO mice, examined another major polyunsaturated fatty acid, docosahexaenoic acid (C22:6) PL.Under physiological conditions, the levels of docosahexaenoic acid PL increased significantly in Lpcat3 L-KO mice compared with control mice (Figures S5 and S6).After APAP injection, docosahexaenoic acid PL levels were significantly higher in L-KO mice than in control mice, which was similar to the physiological conditions (Figure 4A,B).Nevertheless, the levels of oxidized docosahexaenoic acid PC and PE and hydroperoxy docosahexaenoic acid PC and PE in Lpcat3  L-KO mice were significantly lower than those in control mice (Figure 4C,D).Therefore, we investigated the upstream pathways of APAP metabolism.The total hepatic glutation levels in the APAP-treated Lpcat3 L-KO group were significantly higher than those in the APAP-treated control group (Figure 4E).Hepatic GSH levels in Lpcat3 L-KO mice were significantly higher than those in control mice (Figure 4F), whereas the levels of hepatic GSSG were almost the same between the two groups after APAP injection (Figure 4G).GSH is converted to GSSG by glutathione peroxidase 4 (GPx4).Therefore, the protein levels of hepatic GPx4 were examined, and no significant difference was observed between the L-KO and control groups (data not shown).These results suggested that the same amount of peroxidized PL was reduced to produce hydroxy lipids because there was no difference in the ability to reduce lipid peroxides, as GSSG and GPx4 were almost at the same level between APAP-treated Lpcat3 L-KO mice and APAP-treated control mice.NAPQI, an APAP metabolite, is detoxified by its conjugation with GSH.However, when the intracellular GSH levels are reduced, NAPQI directly binds to mitochondrial proteins and induces reactive oxygen species (ROS) production.These results suggested that higher GSH levels in APAP-treated Lpcat3 L-KO mice compared with control mice may allow GSH to bind to NAPQI, resulting in less ROS production and less oxygenated PL.Overall, hepatic Lpcat3 deficiency prevents APAPinduced acute liver injury by suppressing the accumulation of oxygenated compounds by promoting conjugation between GSH and NAPQI.

| DISCUSSION
In the present study, we demonstrated that Lpcat3 is associated with APAP-induced acute liver injury.Liverspecific Lpcat3 deficiency ameliorated APAP-induced     During ferroptosis, iron-mediated intracellular peroxidation of phospholipids induces cytotoxicity, leading to cell death. 23,24Previous reports have indicated that ferroptosis causes various diseases, 25 including nonalcoholic steatohepatitis (NASH), 26 pulmonary ischemia-reperfusion, 27 intestine ischemia-reperfusion, 28 cardiomyopathy, 29 and so on.However, it was recently argued that APAP-induced cell death was not ferroptosis but necrosis because α-tocopherol, a ferroptosis inhibitor, could not improve APAP-induced liver failure.The relationship between APAP and ferroptosis remains unclear. 30ur results showed that Lpcat3 deficiency reduced APAP-induced liver failure and that hydroperoxide arachidonate PE decreased significantly compared with the control group (Figure 3D), indicating that APAP-induced liver injury was partly explained by ferroptosis.
In addition, lipidomic analysis using mass spectrometry showed that oxidized and hydroperoxide docosahexaenoic acid PL accumulated less in the livers of APAP-treated Lpcat3 L-KO mice than in the livers of APAP-treated control mice.The reduced accumulation of most types of oxidized and hydroperoxide PL suggested that the loss of Lpcat3 in the liver strengthened APAP nontoxic pathway by promoting the conjugation of GSH and NAPQI, resulting in reduced production of ROS and less oxygenated PL.However, our study did not reveal how GSH levels in Lpcat3 deficient mice was higher than in control mice after APAP administration.
Lpcat3 regulates the introduction of arachidonic acid into phospholipids.However, our results showed that although Lpcat3 is deficient in the liver (Figure S1), the levels of arachidonic acid-containing PC and PE were not completely suppressed (Figures S5 and S6), as reported previously. 10In addition to Lpca3, Lplat11 (MBOAT7), Lpcat1 (LPLAT8), Lpcat2 (LPLAT9), Lplat5 (AGPAT5), and Lpcat4 (LPLAT10) have been reported to introduce arachidonic acid into PC and PE. 9 Furthermore, oxidized docosahexaenol PL decreased in APAP-treated mice, despite increasing levels of docosahexaenoic acid-containing PL (Figure 4A-D).Agpat3 (Lpaat3) biosynthesizes docosahexaenoic acid-containing PL. 6,8,[31][32][33][34][35] Previously, our group has reported that docosahexaenoic acid-containing PL decreased in hepatic Agpat3 deficient mice, whereas arachidonic acid-containing PL increased. 36The use of APAP-treated Agpat3 deficient mice may further clarify the relationship between APAP-induced acute liver failure and Lpcat3.RNA-Seq or microarray analyses are needed to further examine the pathophysiology of APAP-induced liver injury using both wild type and L-KO mice liver.
Lpcat3 is related to the occurrence and development of several diseases such as atherosclerosis, 37,38 NASH, [39][40][41] skeletal muscle myopathy, 42 and colorectal cancer. 43PCAT3 inhibitors 44 and activators based on the reported structure 45 should be further developed.Our study suggested that LPCAT3 inhibitors may be useful in the treatment of APAP-induced acute liver failure.

AUTHOR CONTRIBUTIONS
Natsuko F. Inagaki designed the study, performed the experiments, analyzed, and interpreted the data, and wrote and edited the manuscript.Hiroki Nakanishi and Takayo Ohto performed the LC-MS/MS experiments, and wrote the Materials and Methods in this part.Hideo Shindou edited the article.Takao Shimizu supervised and edited the manuscript.
22ute liver injury by inhibiting oxidative stress.Lipidomic analysis with mass spectrometry revealed less accumulation of oxidized arachidonate PL and hydroperoxide arachidonate PL in the livers of APAP-treated Lpcat3 L-KO mice.Ferroptosis is a novel form of programmed cell death first reported in RAS-mutated cancer cells.22