Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Dec 1.
Published in final edited form as: Biochim Biophys Acta. 2017 Aug 24;1863(12):3170–3182. doi: 10.1016/j.bbadis.2017.08.017

Growth arrest and DNA damage-inducible 45α protects against nonalcoholic steatohepatitis induced by methionine- and choline-deficient diet

Naoki Tanaka 1,2,3, Shogo Takahashi 1, Xiao Hu 2, Lu Yu 2, Naoyuki Fujimori 4, Srujana Golla 1, Zhong-Ze Fang 1,5, Toshifumi Aoyama 2, Kristopher W Krausz 1, Frank J Gonzalez 1
PMCID: PMC5675002  NIHMSID: NIHMS914058  PMID: 28844958

Abstract

Growth arrest and DNA damage-inducible 45 α (Gadd45α) is a stress-inducible protein that plays an important role in cell survival/death and DNA repair, but its contribution to the development of nonalcoholic steatohepatitis (NASH) has not been investigated. C57BL/6 Gadd45a-null and wild-type (WT) mice were treated with methionine- and choline-deficient diet (MCD) for eight weeks and phenotypic changes examined. Gadd45a-null mice had more severe hepatic inflammation and fibrosis, higher levels of mRNAs encoding pro-inflammatory, pro-fibrotic, and pro-apoptotic proteins, and greater oxidative and endoplasmic reticulum (ER) stress compared with WT mice. Indeed, Gadd45a mRNA was induced in response to ER stress in primary hepatocytes. Lipidomic analysis of NASH livers demonstrated decreased triacylglycerol (TG) in MCD-treated Gadd45a-null mice, which was presumably associated with increased mRNAs encoding phospholipase D (Pld1/2), phosphatidic acid phosphatase type 2A, and choline/ethanolaminephosphotransferase 1 (Cept1), involved in the phosphatidylcholine-phosphatidic acid-diacylglycerol cycle, and decreased mRNAs encoding fatty acid (FA)-binding protein 1 (Fabp1) and FA transport protein 5. Treatment of cultured primary hepatocytes with tumor necrosis factor α, transforming growth factor β, and hydrogen peroxide led to the corresponding induction of Fabp1, Pld1/2, and Cept1 mRNAs. Collectively, Gadd45α plays protective roles against MCD-induced NASH likely due to attenuating cellular stress and ensuing inflammatory signaling. These results also suggest an interconnection between hepatocyte injury, inflammation and disrupted glycerophospholipid/FA metabolism that yields a possible mechanism of decreased TG accumulation with NASH progression (i.e., “burned-out” NASH).

Keywords: oxidative stress, ER stress, lipidomics, glycerophospholipid metabolism, burned-out NASH

Graphical abstract

graphic file with name nihms914058u1.jpg

1. Introduction

Nonalcoholic steatohepatitis (NASH) is a major chronic liver disease defined by the presence of macrovesicular steatosis, hepatocyte ballooning, and hepatic inflammation regardless of no ethanol consumption. Similar to alcoholic steatohepatitis, NASH can progress to liver cirrhosis, hepatocellular carcinoma, and hepatic failure. It was documented that the accumulation of toxic lipids, such as saturated fatty acid (FA), free cholesterol, and diacylglycerol (DG) in hepatocytes augmented oxidative stress and endoplasmic reticulum (ER) stress, leading to enhanced inflammatory signaling, hepatocyte apoptosis, and Kupffer cell/stellate cell activation [1,2]. Therefore, oxidative and ER stress are thought to be key contributors in the progression from steatosis to steatohepatitis.

Growth arrest and DNA damage-inducible 45 α (Gadd45α) is a 17–18 kilodalton protein that is linked to several cellular events, such as cell survival, DNA repair, chromatin assembly, and genome stability [3,4]. For example, Gadd45α is markedly induced after the administration of dimethylbenzanthracene (DMBA), a genotoxic carcinogen in mice, and decreased DNA repair, increased mutation frequency, and enhanced tumorigenesis were observed in DMBA-treated Gadd45a-null mice compared with wild-type (WT) mice [5]. Additionally, Gadd45α is regulated by p53 and stress-responsive transcription factors, such as activating transcription factor (ATF) 2 [4,6]. These findings indicate a possible link between Gadd45α and chronic liver injury, but its contribution to the pathogenesis of NASH remains unclear.

To address this issue, Gadd45a-null and WT mice were treated with methionine- and choline-deficient diet (MCD), a conventional NASH-inducible diet in rodents, for eight weeks.

2. Materials and Methods

2.1. Mice and treatment

All studies were conducted according to Institute of Laboratory Animal Resource guidelines and approved by the National Cancer Institute Animal Care and Use Committee. The mice were housed in a specific pathogen-free environment controlled for temperature and light (25 °C, 12-hour light/dark cycle) and maintained with NIH31 regular chow and tap water ad libitum. The MCD and control MCS diets were purchased from Dyets Inc. (Bethlehem, PA; #518810 and #518754, respectively). The compositions of these diets were described previously [7,8]. Before starting the experiments, the NIH31 chow was replaced with control MCS for acclimatization. After five-day acclimatization, male C57BL/6NCr WT and Gadd45a-null mice at 8–12 weeks of age (n = 6–8/group) [5] were moved to new cages and the respective diet was given for eight weeks. To examine time course of Gadd45a mRNA expression in MCD-induced NASH, liver tissues obtained from male C57BL/6NCr WT mice fed a MCS or MCD for three days and one, two, and four weeks reported previously [7–9] were used (n = 5–7/group). Throughout all experiments, mice were weighed and killed after a four- to six-hour fasting. Blood was collected using Serum Separator Tubes (Becton, Dickinson and Company, Franklin Lakes, NJ) and centrifuged for 10 minutes at 8,000 g at 4 °C. Liver was isolated, weighed, and divided into the two parts. One part of the liver tissue (a neighboring lobe to gallbladder) was immediately soaked in 10% neutral formalin for histological examination. Sera and the remaining liver were immediately frozen in liquid nitrogen and kept at −80 °C until use.

2.2. Serum metabolomic analysis

Samples were prepared and serum metabolomics was performed using ultraperformance liquid chromatography-electrospray ionization-quadrupole time-of-flight mass spectrometry (UPLC-ESI-QTOFMS) as described previously [7,8,10]. All samples were analyzed in a randomized fashion and MassLynx software (Waters Corp., Milford, MA) was used to acquire the chromatogram and mass spectrometric data. Centroided and integrated chromatographic mass data were processed by MarkerLynx (Waters) to generate a multivariate data matrix. Pareto-scaled MarkerLynx matrices including information on sample identity were analyzed by principal component analysis (PCA) and supervised orthogonal projection to latent structure (OPLS) analysis using SIMCA-P +12 (Umetrics, Kinnelon, NJ). The ions were searched using METLIN metabolite database. The identity of ions was confirmed by tandem mass spectrometry MS/MS fragmentation patterns. The abundance of metabolites was determined by calculating peak area, normalized to that of chlorpropamide, and expressed as fold changes relative to that of MCS-treated WT mice.

2.3. Liver lipidomic analysis

Approximately twenty-five mg of liver tissue was homogenized in H2O/methanol (300 μL/400 μL) containing chlorpropamide and aminopimelic acid (final concentrations of 5 μM and 10 μM, respectively). The homogenates were added to 800 μL of chloroform containing heptadecanoyl-lysophosphatidylcholine (17:0-LPC, 2 μM), heptadecanoyl-phosphatidylcholine (17:0-PC, 1 μM), heptadecanoyl-sphingomyelin (2 μM), and heptadecanoyl-ceramide (2 μM), incubated at 37 °C while shaking for 20 min, and then centrifuged at 10,000 g for 20 min. Organic phases were carefully collected, evaporated under nitrogen flow around 1 hour, and dissolved with 100 μL of methanol/chloroform (1:1). After 50-fold dilution with injection buffer (isopropanol: acetonitrile: H2O=2:1:1), samples were subjected to using UPLC-ESI-QTOFMS. The samples were separated and analyzed using a Waters Acquity UPLC system coupled to a Waters Synapt HDMS Quadrupole-Time of Flight (Q-TOF) mass spectrometer operating under the following conditions: capillary volts 2.8kV, sample cone 30V, source temperature 150 °C, desolvation temperature 400 °C, cone and desolvation gas flow 50 and 850 L/hr, respectively. Data was acquired in centroid mode in both positive and negative electrospray ionization modes, using sulfadimethoxine (m/z 311.0814+, 309.0658−) as the lock mass. Mass range acquired was 100–1200 m/z at 0.3 second scans. Chromatography was carried out using a Waters Acquity CSH C18 column (2.1×100mm) under acidic conditions buffered with 10mM ammonium formate using the following compositions: (A) 60% acetonitrile in water; (B) 10% acetonitrile in isopropanol. The following gradient (6=linear, 1=ballistic) was used: initial conditions 60% (A) to 80% (A) at 6.5 minutes (6), to 50% (A) at 2.1 minutes (1), to 46% (A) at 12.0 minutes (6), to 30% (A) at 12.1 minutes (1), to 1% (A) at 18.0 minutes (6), to 60% (A) at 18.1 minutes (6), held for 5 minutes for column equilibration before the next injection. Total run time was 21 minutes. Flow rate was maintained at 0.4mL/min throughout the run and column temperature was maintained at 55 °C. All samples were injected at 5 μL, using partial loop with needle overfill. Centroided and integrated chromatographic mass data were processed by MarkerLynx (Waters) to generate a multivariate data matrix. Pareto-scaled MarkerLynx matrices including information on sample identity were analyzed by PCA and supervised OPLS analysis using SIMCA-P+12 (Umetrics, Kinnelon, NJ). The OPLS loading scatter S-plot was used to determine the lipids that contributed significantly to the separation between MCD-treated WT and Gadd45a-null mice. The lipid metabolite structures were determined using the METLIN metabolite database. The identity of LPC in the liver was confirmed by tandem mass spectrometry MS/MS fragmentation patterns (Supplementary Fig. 1). The abundance of lipid metabolites was measured by calculating peak area, normalized to that of chlorpropamide, and expressed as fold changes relative to that of MCS-treated WT mice.

2.4. Quantitative polymerase chain reaction (qPCR) analysis

After extraction of total RNA from liver tissue using a TRIzol Reagent (Invitrogen, Carlsbad, CA), cDNA was generated from 1 μg RNA with a SuperScript IITM Reverse Transcriptase kit and random oligonucleotides (Invitrogen) and qPCR was performed using SYBR green PCR master mix and ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA) [7–9]. The primer pairs were designed sing qPrimerDepot (http://mouseprimerdepot.nci.nih.gov/) and listed in Supplementary Table 1. The mRNA levels were normalized to those of 18S ribosomal RNA and expressed as fold changes relative to those of MCS-treated WT mice.

2.5. Histological analysis

Small pieces of liver tissue were fixed in 10% neutral formalin, dehydrated by serial ethanol/xylene, and embedded in paraffin. The sections (4 μm thick) were stained by the hematoxylin and eosin [11] and picrosirius red staining methods [12]. The percentage of fat droplet area to hepatocyte area in the hematoxylin and eosin-stained sections was automatically determined using A BIOREVO BZ-9000 microscope (Keyence, Osaka, Japan) and Dynamic cell count BZ-II analysis application (Keyence) as described previously [13]. More than 10 fields were randomly examined in each section and the final value was expressed as the percentage of total fat areas to total hepatocyte areas. Additionally, liver tissues from some mice were subjected to making frozen sections (10 μm thick) for oil red O staining [14].

2.6. Biochemical analysis

Serum levels of alanine aminotransferase (ALT) and alkaline phosphatase (ALP) were measured with assay kits for ALT and ALP (Catachem, Bridgeport, CT), respectively. Hepatic contents of triacylglycerol (TG) and non-esterified fatty acids (NEFA) were quantified using Wako TG kit and NEFA kit (Wako Chemicals USA Inc.) as described elsewhere [7–10]. Hepatic levels of PC, DG, and malondialdehyde (MDA) were measured by Phosphatidylcholine assay kit, Diacylglycerol assay kit, and OxiSelect™ TBARS assay kit, respectively (Cell Biolabs, Inc., San Diego, CA). Hepatic phospholipase D (PLD) activity was determined by use of a colorimetric assay kit purchased from BioVision (Milpitas, CA). Mitochondial fraction was isolated by differential centrifugation of sucrose and hydrogen peroxide (H2O2) and glutathione (GSH) measured using PeroxiDetect kit and glutathione assay kit, respectively (Sigma-Aldrich, Inc., St. Louis, MO) [9]. Measurement of protein concentrations was carried out using BCA™ protein assay kit (Thermo Scientific, Rockford, IL).

2.7. Immunoblot analysis

Preparation of whole liver homogenates was conducted as described previously [15–17] and their protein concentrations were measured colorimetrically with the BCA™ protein assay kit. Whole liver homogenates (30–80 μg of protein in each lane) were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The polyacrylamide concentration in the gel was 10% or 12.5%, which was dependent on molecular weight of the target protein. After electrophoresis, the proteins were transferred to a nitrocellulose membrane (Amersham Hybond-P, GE Health Care, Little Chalfont, UK). The membranes were blocked for 1 hour with 3% non-fat dry milk in Tris-buffered saline and incubated overnight with the respective primary antibody. Primary antibodies against FA-binding protein 1 (FABP1) were described previously [18]. The following primary antibodies were purchased from Abcam (Cambridge, MA): choline/ethanolamine phosphotransferase 1 (CEPT1), #107415, 1:50 dilution; serine palmitoyltransferase (SPT), #23696, 1:300 dilution; phosphatidic acid (PA) phosphatase type 2A (PPAP2A), #198280, 1:500 dilution; and β-actin, #8227, 1:1000 dilution. After four washes, the membranes were incubated with alkaline phosphatase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, #93785, 1:1000 dilution) and treated with 1-step NBT/BCIP substrate (Pierce, Rockford, IL). The true band positions were determined by co-electrophoresing molecular weight standards (Bio-Rad, Hercules, CA) and the band of β-actin was used as a loading control. Immunoblotting was repeated twice for each protein and each band intensity was quantified using NIH image J (National Institutes of Health, Bethesda, MD), normalized by those of loading control, and subsequently expressed as a fold change relative to those of WT mice fed MCS diet.

2.8. Isolation and treatment of mouse primary hepatocytes

Mouse primary hepatocytes were isolated from C57BL/6NCr WT mice at the age of 8–12 weeks and treated with as thapsigargin (250 and 500 nM, Sigma-Aldrich), H2O2 (150 and 300 μM, Sigma-Aldrich), tumor necrosis factor α (TNFα, 100 ng, Sigma-Aldrich), and transforming growth factor β (TGFβ, 3 ng, R&D Systems, Minneapolis, MA), respectively, as described previously [7,9,10]. The cDNA samples used in the previous report [9] were subjected to qPCR analysis.

2.9. Statistical analysis

Quantitative data were expressed as mean ± standard deviation. Statistical analyses were performed using the two-tailed Student’s t-test, Dunnet’s test, and ANOVA test with Bonferroni’s correction. Correlation analysis was done using Spearman’s test. The number of samples was 7, 6, 7, and 8 obtained from MCS-treated WT and Gadd45a-null mice and MCS-treated WT and Gadd45a-null mice, respectively. A P value of less than 0.05 was considered to be statistically significant.

3. Results

3.1. Gadd45a mRNA is induced by ER stress

The trigger of Gadd45α induction was examined using primary hepatocyte cultures. Treatment with 100 ng of TNFα and 3 ng of TGFβ did not change Gadd45a mRNA levels, while treatment with 300 μM of H2O2 led to marginal increases (Supplementary Fig. 2). Thapsigargin (250 and 500 nM) treatment increased Gadd45a mRNA levels (Supplementary Fig. 2), indicating that Gadd45α is induced in response to cellular stress, especially ER stress.

3.2. Increased inflammation and fibrosis in the livers of Gadd45a-null mice after 8-week MCD treatment

Since preliminary experiments showed that Gadd45a mRNA levels were significantly increased after 8-week MCD treatment (Supplementary Fig. 3), WT and Gadd45a-null mice were treated with MCD for eight weeks and phenotypes compared. While liver-to-body weight ratios were not different between the groups (Fig. 1A), serum ALT and ALP levels were significantly increased in MCD-treated Gadd45a-null mice compared with MCD-treated WT mice (Fig. 1B). Liver histology showed marked infiltration of inflammatory cells and some apoptotic hepatocytes in MCD-treated Gadd45a-null mice (Fig. 1C). As revealed by oil red O staining, lipid droplets were smaller and fewer in the MCD-treated Gadd45a-null mice compared with similarly-treated WT mice (Fig. 1C). The percentage of total fat droplet areas to total hepatocyte areas were significantly decreased in MCD-treated Gadd45a-null mice compared with similarly-treated WT mice (Fig. 1D). Additionally, picrosirius red staining demonstrated more dense perisinusoidal/periportal fibrosis in these mice (Fig. 2A). Increased inflammation and fibrosis in the pathological findings were corroborated by higher levels of mRNAs encoding CD68 (Cd68) and integrin alpha M (Itgam), which are expressed in macrophages and inflammatory cells, TNFα (Tnf) and TGFβ1 (Tgfb1), typical pro-inflammatory and pro-fibrotic cytokines, respectively, and alpha smooth muscle actin (Acta2), collagen 1a1 (Col1a1), and keratin 19 (Krt19) in MCD-treated Gadd45a-null mice compared with WT counterparts (Fig. 1D and 2B). Connective tissue growth factor and galectin 3 (encoded by Ctgf and Lgals3, respectively) are the main regulators for promoting fibrogenesis [19,20]. These mRNA levels were also increased in MCD-treated Gadd45a-null mice (Fig. 2B). All results indicate that Gadd45a-null mice exhibit more severe inflammation and fibrosis after 8-week MCD treatment.

Fig. 1. More severe hepatocyte damage and hepatitis in the livers of Gadd45a-null mice after 8-week MCD treatment.

Fig. 1

Open in a new tab

Male 8- to 12-week-old C57BL/6NCr wild-type (WT) and Gadd45a-null (KO) mice were treated with a methionine- and choline-deficient diet (MCD) or control methionine- and choline-supplemented MCD diet (MCS) for 8 weeks (n = 6–8/group).

(A) Liver weight, expressed as a percentage of body weight (BW).

(B) Serum alanine aminotransferase (ALT) and alkaline phosphatase (ALP) activities.

(C) Representative histological appearance of liver. Hematoxylin and eosin staining and Oil Red O staining; Bar = 200 μm (×100 magnification) or 100 μm (×200 magnification).

(D) The percentage of fat droplet area to hepatocyte area.

(E) qPCR analysis of genes associated with inflammation. The mRNA levels were normalized to those of MCS-treated WT mice.

Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. *P<0.05; **P<0.01; ***P<0.001 between MCD-treated WT and KO mice. #P<0.05; ##P<0.01; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

Fig. 2. More severe fibrosis in the livers of Gadd45a-null mice after 8-week MCD treatment.

Fig. 2

Open in a new tab

Male 8- to 12-week-old C57BL/6NCr wild-type (WT) and Gadd45a-null (KO) mice were treated with a methionine- and choline-deficient diet (MCD) or control methionine- and choline-supplemented MCD diet (MCS) for 8 weeks (n = 6–8/group).

(A) Representative photomicrographs of picrosirius red staining. Bar = 200 μm (×100 magnification). Dense pericellular/periportal fibrosis was detected in MCD-treated KO mice.

(B) qPCR analysis of genes associated with liver fibrosis. The mRNA levels were normalized to those of MCS-treated WT mice.

Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. **P<0.01; ***P<0.001 between MCD-treated WT and KO mice. #P<0.05; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

3.3. Greater oxidative stress and ER stress in the livers of Gadd45a-null mice after 8-week MCD treatment

Ctgf and Lgals3 mRNAs are induced by reactive oxygen species (ROS) and ER stress [19,20]. In support of this finding, liver MDA, a byproduct of lipid peroxidation, and mitochondrial H2O2 levels were markedly increased in MCD-treated Gadd45a-null mice without significant differences in mitochondrial GSH contents (Fig. 3A). Consistent with these findings, the mRNA levels of ROS-generating enzymes, such as p47phox (Ncf1), p40phox (Ncf4), and NADPH oxidase 2 (Cybb), were increased (Fig. 3B), while mRNAs encoding the ROS-detoxifying enzymes, copper/zinc and manganese superoxide dismutases and catalase, were not different between the groups (Supplementary Fig. 4). Peroxiredoxin 3 mRNA levels tended to be increased in MCD-treated Gadd45a-null mice compared with similarly-treated WT mice, but the increase did not reach statistical significance (Supplementary Fig. 4). Additionally, the mRNA levels of ER stress-induced genes, such as DNA-damage inducible transcript 3 (Ddit3, also known as CHOP), heat shock protein 5 (Hspa5), and calreticulin (Calr) were increased in MCD-treated Gadd45a-null mice (Fig. 3C). Oxidative stress and ER stress activate several apoptotic pathways. The mRNA levels of genes encoding pro-apoptotic proteins, such as death receptor 5 (Dr5, encoded by Tnfrsf10b), BCL2-associated X protein (Bax), and BCL2-like 11 (Bcl2l11, also known as Bim), were enhanced in these mice (Fig. 3D) without any alterations in mRNAs encoding anti-apoptotic Bcl2/Bcl2l1 (also known as Bcl-xL) (data not shown). Collectively, greater oxidative stress and ER stress in the livers of MCD-treated Gadd45a-null mice were likely associated with aggravated hepatocyte damage and hepatic inflammation/fibrosis. Considering the results from MCD-treated Gadd45a-null mice and significant Gadd45α induction observed in primary hepatocytes treated with thapsigargin, Gadd45α appeared to protect against MCD-induced hepatocyte injury primarily through ameliorating ER stress.

Fig. 3. Greater oxidative stress, endoplasmic reticulum (ER) stress, and apoptotic signaling in the livers of Gadd45a-null mice after 8-week MCD treatment.

Fig. 3

Open in a new tab

Liver samples obtained from mice presented in Fig. 1 were subjected to the assays (n = 6–8/group). WT, wild-type mice; KO, Gadd45a-null mice; MCS, control diet: MCD, methionine- and choline-deficient diet.

(A) Liver malondialdehyde (MDA) levels and mitochondrial hydrogen peroxides (H2O2) and glutathione (GSH) contents.

(B) qPCR analysis of genes encoding oxidative stress-producing enzymes, p47phox (Ncf1), p40phox (Ncf4), and NADPH oxidase 2 (Cybb).

(C) qPCR analysis of ER stress-induced genes, DNA-damage inducible transcript 3 (Ddit3, also known as CHOP), heat shock protein 5 (Hspa5), and calreticulin (Calr).

(D) qPCR analysis of genes encoding ER stress-associated pro-apoptotic proteins, death receptor 5 (Tnfrsf10b), BCL2-associated X protein (Bax), and BCL2-like 11 (Bcl2l11, also known as Bim).

The mRNA levels were normalized to those of MCS-treated WT mice and statistical analysis was performed using the ANOVA test with Bonferroni’s correction. *P<0.05; **P<0.01; ***P<0.001 between MCD-treated WT and KO mice. #P<0.05; ##P<0.01; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

3.4. Serum metabolomic analysis reveals decreased palmitoyl-lysophosphatidylcholine (16:0-LPC) in MCD-treated Gadd45a-null mice

Metabolomics is a useful tool for understanding disease mechanism [21]. To examine metabolite changes linked to NASH severity in MCD-treated Gadd45a-null mice, a serum metabolomic analysis was conducted. PCA of UPLC-ESI-QTOFMS negative mode data showed clear separation between MCD-treated Gadd45a-null and WT mice (Fig. 4A). The loading plot detected LPCs (1-acyl-sn-glycero-3-phosphocholine), such as palmitoyl-LPC (16:0-LPC), stearoyl-LPC (18:0-LPC), oleoyl-LPC (18:1-LPC), and linoleoyl-LPC (18:2-LPC) as ions that were significantly decreased in MCD-fed Gadd45a-null mice (Fig. 4B). Quantification confirmed a significant reduction in 16:0-LPC in MCD-treated Gadd45a-null mice compared with similarly-treated WT mice (Fig. 4C) that was inversely correlated with serum ALT and hepatic Tnf and Cd68 mRNAs (Fig. 4D). These results indicate that decreased 16:0-LPC in serum reflects hepatitis severity.

Fig. 4. Serum metabolomic analysis revealed decreased palmitoyl-lysophosphatidylcholine (16:0-LPC) in MCD-treated Gadd45a-null mice.

Fig. 4

Open in a new tab

Serum samples obtained from the mice presented in Fig. 1 were subjected to metabolomic analysis (n = 6–8/group). WT, wild-type mice; KO, Gadd45a-null mice; MCS, control diet: MCD, methionine- and choline-deficient diet.

(A) PCA in WT (box) and KO (circle) mice treated with 8-week MCD.

(B) S-plot of OPLA analysis using the same data as (A).

(C) Serum LPC levels. Values were normalized to those of MCS-treated WT mice and were expressed as relative abundance. Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. *P<0.05 between MCD-treated WT and KO mice. ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

(D) Inverse correlation between serum 16:0-LPC and ALT and hepatic Tnf and Cd68 mRNA levels. Statistical analysis was done using Spearman’s test. Correlation coefficients (r) and P values were indicated.

3.5. Liver lipidomic analysis demonstrates significant decreases in TG in MCD-treated Gadd45a-null mice

Decreased serum LPC suggests wide-ranging changes in lipid metabolism since PC is a key intermediate of glycerophospholipid/sphingolipid/FA metabolism [22]. Additionally, histological findings revealed decreased lipid droplets with more severe hepatitis and fibrosis in MCD-treated Gadd45a-null livers (Fig. 1C and D), indicating disrupted neutral lipid metabolism. For the unbiased assessment of altered lipid metabolism in Gadd45a-null NASH livers, a lipidomic analysis was performed. PCA of UPLC-ESI-QTOFMS positive mode data demonstrated clear discrimination between MCD-treated WT and Gadd45a-null mice (Fig. 5A). S-plot of OPLS analysis showed that six lipids were increased in MCD-treated Gadd45a-null mice compared with similarly treated WT mice (Supplementary Table 2 and Fig. 5B) and were assumed to be PCs based on retention time and m/z, while these could not be identified due to the lack of authentic standards. Alternatively, the overall hepatic PC contents were determined using a commercial kit, but no significant changes were detected between the groups (Fig. 5C). Similarly, ten lipids, including 16:0-LPC and 18:0-LPC, were selected as lipids significantly decreased in MCD-fed Gadd45a-null mice (Supplementary Table 2 and Fig. 5B). While half of the lipids could not be confirmed, they were probably TGs consisting of various combinations of FAs with different carbon number and carbon double bond position. Indeed, liver TG contents were determined by biochemical reactions and found significant decreases in MCD-treated Gadd45a-null mice (Fig. 5D). Hepatic levels of DG and NEFA, precursors of TG, were also assayed and the former was decreased in these mice (Fig. 5E and F). Quantification revealed no significant differences in hepatic 16:0-LPC and 18:0-LPC between the MCD-fed WT and Gadd45a-null mice (Fig. 5G).

Fig. 5. Liver lipidomic analysis demonstrated decreases in TG in MCD-treated Gadd45a-null mice.

Fig. 5

Open in a new tab

Liver samples obtained from the mice presented in Fig. 1 were subjected to lipidomic analysis (n = 7–8/group). WT, wild-type mice; KO, Gadd45a-null mice; MCS, control diet: MCD, methionine- and choline-deficient diet.

(A) PCA in WT (circle) and KO (triangle) mice treated with 8-week MCD.

(B) S-plot of OPLA analysis using the same data as (A). Abbreviations of lipids are identical to those used in Supplementary Table 2.

(C-G) Hepatic contents of PC (C), TG (D), DG (E), NEFA (F), and LPC (G). PC, TG, DG, and NEFA were measured using biochemical reaction. The values of LPC were normalized to those of MCS-treated WT mice and were expressed as relative abundance.

Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. *P<0.05; **P<0.01; ***P<0.001 between MCD-treated WT and KO mice. #P<0.05; ##P<0.01; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

3.6. Altered glycerophospholipid metabolism in NASH livers of Gadd45a-null mice

In order to understand the mechanism of diminished TG in the livers of MCD-treated Gadd45a-null mice, the expression of genes encoding enzymes/proteins involved in TG metabolism was determined. The mRNA levels of genes involved in TG synthesis [DG acyltransferase 1 (Dgat1) and 2 (Dgat2)], TG hydrolysis [patatin-like phospholipase domain containing 2 (Pnpla2) and 3 (Pnpla3)], and TG secretion [microsomal TG transfer protein (Mttp) and apolipoprotein B (Apob)] were not changed between the MCD-treated groups (Fig. 6A and Supplementary Fig. 5A). The lack of changes in the expression of these genes prompted us to examine the PC-PA-DG cycle. PC is converted to PA via PLD1–4 and subsequently dephosphorylated to DG by PAPP2A-C and lipin 1–3. Among these, the expression of mRNAs encoding PLD1 (Pld1), PLD2 (Pld2), and PPAP2A (also called phospholipid phosphatase 1, Plpp1) were increased in MCD-treated Gadd45a-null mice (Fig. 6B and Supplementary Fig. 5B). Lastly, DG is converted to PC by CEPT1 and choline phosphotransferase 1 (Chpt1), or converted to phosphatidylethanolamine (PE) by Cept1 and ethanolaminephosphotransferase 1 (Ept1). PE is further converted to PC by phosphatidylethanolamine N-methyltransferase (Pemt), which is reported to play an important role in NASH development [23]. Among these enzymes related to DG utilization, the levels of Cept1 mRNA were increased in MCD-fed Gadd45a-null mice (Fig. 6C). PC is converted from LPC by lysophosphatidylcholine acyltransferase (Lpcat) 1–4. The levels of Lpcat1/2/4 mRNAs were markedly increased in MCD-fed Gadd45a-null mice (Supplementary Fig. 5C). In accordance with abovementioned mRNA measurements, hepatic PLD activity and the expression levels of CEPT1 protein were increased in MCD-fed Gadd45a-null mice compared with similarly-treated WT mice (Fig. 6D–F), but the PPAP2A protein levels could not be addressed due to weak band intensity (data not shown). These changes demonstrate that an enhanced PC-PA-DG cycle may be associated with the reduced TG contents in Gadd45a-null NASH livers.

Fig. 6. Altered expression of enzymes involved in glycerophospholipid metabolism in NASH livers of Gadd45a-null mice.

Fig. 6

Open in a new tab

(A–C) qPCR analysis of genes associated with TG metabolism (A), DG synthesis (B), and conversion from DG to PC (C). The same cDNA samples used in Figs. 1–3 were adopted (n = 6–8/group). The mRNA levels were normalized to those of MCS-treated WT mice.

(D) Hepatic PLD activity. Samples from the same mice presented in Fig. 6B were used for the measurements.

(E–F) Immunoblot analysis of CEPT1. Whole liver homogenates (80 μg of proteins) were loaded in each well and the band of β-actin was used as a loading control. The band intensity was measured densitometrically, normalized to those of β-actin, and expressed as values relative to MCS-treated WT mice. Band intensity values were obtained from two independent immunoblot experiments.

Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. WT, wild-type mice; KO, Gadd45a-null mice; MCS, control diet: MCD, methionine- and choline-deficient diet. *P<0.05; **P<0.01; ***P<0.001 between MCD-treated WT and KO mice. #P<0.05; ##P<0.01; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

3.7. Altered FA metabolism in NASH livers of Gadd45a-null mice

Because FA and FA-CoA are sources of PA/DG/TG as well, the mRNAs encoding enzymes/proteins associated with FA transport and metabolism were also quantified. Among the enzymes/proteins involved in FA uptake/activation (Fig. 7A), de novo FA synthesis (Supplementary Fig. 5D), and FA catabolism (Fig. 7B), FABP1 (Fabp1) and FA transporter 5 (Slc27a5) mRNAs were decreased and long-chain acyl-CoA dehydrogenase (Acadl) and FA translocase (Cd36) mRNAs were increased in the livers of MCD-fed Gadd45a-null mice. Additionally, palmitic acid is used in the de novo ceramide synthesis pathway and the first step is regulated by SPT. The mRNA levels encoding SPT long chain base subunit 1 and 2 (Sptlc1 and Sptlc2, respectively) were enhanced in MCD-treated Gadd45a-null mice (Fig. 7C). Immunoblot analysis revealed significant decreases in FABP1, but not SPT (Fig. 7D and E). Therefore, these changes, except for Cd36 mRNA, may reduce FA supply into the PA/DG/TG cycle, presumably leading to attenuated TG accumulation in the liver.

Fig. 7. Altered expression of transporters and enzymes involved in fatty acid (FA) metabolism in NASH livers of Gadd45a-null mice.

Fig. 7

Open in a new tab

(A–C) qPCR analysis of genes involved in FA uptake into hepatocytes and activation (A), FA catabolism (B), and sphingosine acylation (C). The same cDNA samples used in Figs. 1–3 were adopted (n = 6–8/group). The mRNA levels were normalized to those of MCS-treated WT mice.

(D and E) Immunoblot analysis of FABP1 and SPT. Whole liver homogenates (30 and 60 μg of proteins for FABP1 and SPT, respectively) were loaded in each well and the band of β-actin was used as a loading control. The band intensity was measured densitometrically, normalized to those of β-actin, and expressed as values relative to MCS-treated WT mice. Band intensity values were obtained from two independent immunoblot experiments.

Statistical analysis was performed using the ANOVA test with Bonferroni’s correction. WT, wild-type mice; KO, Gadd45a-null mice; MCS, control diet: MCD, methionine- and choline-deficient diet. *P<0.05; **P<0.01 between WT and KO mice in the same treatment. #P<0.05; ##P<0.01; ###P<0.001 vs. MCS-treated mice in the same genotype. NS, not significant.

3.8. Crosstalk between glycerophospholipid/FA metabolism and inflammatory signaling and cellular stress

Whether inflammatory signaling and cellular stress can influence mRNA levels of glycerophospholipid/FA metabolism-related genes was assessed using primary hepatocytes. Treatment of hepatocytes with TNFα down-regulated Fabp1 mRNA levels, and TGFβ up-regulated Pld1/2 mRNAs (Fig. 8A and B). Additionally, treatment with H2O2 increased Cept1 mRNA levels, while thapsigargin treatment did not alter the expression of any of these mRNAs (Fig. 8C and D). Therefore, inflammatory cytokines and oxidative stress might directly affect the expression of some genes involved in glycerophospholipid/FA metabolism.

Fig. 8. Crosstalk between glycerophospholipid/FA metabolism and inflammatory signaling/cellular stress.

Fig. 8

Open in a new tab

Mouse primary hepatocytes isolated from C57BL/6NCr wild-type mice were treated with TNFα (100 ng, 24 hours), TGFβ (3 ng, 24 hours), H2O2 (300 μM, 4 hours), and thapsigargin (500 nM, 6 hours) as described previously [7–10]. The cDNA samples used in the previous report [9] were subjected to qPCR analysis (n = 3–6/group). The mRNA levels of genes altered in MCD-treated Gadd45a-null mice were quantified to see the impact of inflammatory signaling/cellular stress on glycerophospholipid/FA metabolism and were normalized to those of vehicle-treated hepatocytes. Statistical analysis was performed using the Student’s t-test. *P<0.05; **P<0.01; ***P<0.001 vs. vehicle.

4. Discussion

Although Gadd45α is induced after administration of several hepatotoxicants in mice [24], the pathophysiological roles of Gadd45α in liver disease have not been investigated. This study demonstrated that ER stress is a major inducer of Gadd45a mRNA in hepatocytes and that disruption of Gadd45a augmented ER and oxidative stresses and pro-inflammatory and pro-fibrotic signaling in the liver after 8-week MCD treatment. These results indicate that Gadd45α is not only as a stress sensor, but is also a hepatoprotectant in the context of experimentally-induced NASH.

Gadd45α was reported to be induced in cooperation with ATF2 and the breast cancer suppressor BRCA1, and down-regulated in mammary tumors of mice heterozygous for Atf2 and p53 genes [6], suggesting anti-tumor properties of Gadd45α. However, it is unclear whether Gadd45α induction causes or is the result of liver disease. The current study using MCD-fed Gadd45a-null mice revealed a protective role for Gadd45α in NASH. Additionally, more advanced fibrosis was seen in MCD-fed Gadd45a-null mice compared with the similarly-treated WT counterparts. These findings are consistent with a previous report that treatment with a peroxisome proliferator-activated receptor gamma agonist KR62776 suppresses stellate cell activation through Gadd45α induction and attenuates carbon tetrachloride-induced liver fibrosis in rats [25]. Additionally, a recent study revealed that knockdown of Gadd45α in isolated hepatic stellate cells enhanced carbon tetrachloride-induced ROS production [26]. However, this is the first study suggesting that Gadd45α influences NASH, a common chronic liver disease increasing worldwide.

It is known that Gadd45α is up-regulated in response to DNA damage [3,4]. However, while DNA damage is a consequence of various cytotoxic stimuli, it is unclear whether Gadd45α induction is the result of DNA damage. Additionally, the contribution of Gadd45α to apoptotic signaling is controversial [3,4]. Experiments using primary hepatocytes demonstrated that ER stress is an early inducer of Gadd45α. ER stress causes up-regulation of Ddit3 mRNA through ATF6 and unfolded protein response. Ddit3 is a transcription factor inducing pro-apoptotic proteins, Dr5 and Bim [27]. Since MCD-treated Gadd45a-null mice exhibited increased expression of Ddit3, Dr5 and Bim and more severe hepatocyte injury, Gadd45α induction caused by ER stress is likely an adaptive response to attenuate cell damage and pro-apoptotic signaling. Future studies are needed to examine how Gadd45α modulates stress-activated signaling.

MCD-treated Gadd45a-null mice showed more severe fibrosis but milder liver TG accumulation. This finding is similar to well-known phenomenon that steatosis diminishes as fibrosis progresses in NASH (“burned-out” NASH) [28]. However, the mechanism of “burned-out” NASH is not fully understood. Liver lipidomics revealed significant decreases in hepatic TG levels in MCD-treated Gadd45a-null mice. Serum 16:0-LPC was reduced in MCD-treated Gadd45a-null mice with inverse correlation to hepatitis activity, which is consistent with a previous study showing the induction of Lpcat2/4 by TNFα and TGFβ [7]. The induction of Lpcat suggests accelerated conversion from LPC to PC. Additionally in this study, hepatic PLD activity and CEPT1 expression were up-regulated in MCD-treated Gadd45a-null mice, associated with enhanced PC-PA-DG cycle and diminished hepatic TG synthesis. Furthermore, down-regulation of FABP1, involved in FA uptake and supply, was also linked to decreased TG in the livers. Interestingly, TGFβ, TNFα, and H2O2 resulted in similar changes in mRNA levels of Pld1/2, Fabp1, and Cept1, respectively, to those observed in MCD-fed Gadd45a-null mice. These findings suggest disruption of glycerophospholipid/FA metabolism in the process of “burned-out” NASH development and crosstalk between lipid metabolism, inflammatory signaling, and cellular stress.

While a close relationship between PLD and inflammatory signaling was reported [29], the mechanism on how H2O2 induces Cept1 deserves further investigation. Since oxidative stress and DG activate protein kinase C (PKC) [30], Cept1 might be up-regulated through PKC activation in order to limit these toxicities.

In conclusion, the present study uncovered a protective role for Gadd45α in NASH development to minimize cellular stresses and ensuing inflammatory signaling in the liver. Pharmacological interventions to enhance or activate Gadd45α expression might be useful for attenuating NASH and fibrosis progression.

Supplementary Material

supplement

Highlights.

  1. Gadd45α was induced by endoplasmic reticulum stress in hepatocytes.

  2. MCD-treated Gadd45a-null mice exhibited more severe hepatitis and liver fibrosis.

  3. Liver lipidomics showed decreased TG in MCD-fed Gadd45a-null mice.

  4. Inflammatory signaling and oxidative stress can modulate glycerophospholipid metabolism.

  5. A novel mechanism of decreased fat deposition in advanced stage NASH was proposed.

Acknowledgments

We thank Linda G. Byrd and John Buckley for providing technical assistance with the mouse studies. This study was supported by the National Cancer Institute Intramural Research Program, Center for Cancer Research.

Financial support: Supported by the National Cancer Institute Intramural Research Program and U54 ES16015.

Abbreviations

NASH

nonalcoholic steatohepatitis

FA

fatty acid

DG

diacylglycerol

ER

endoplasmic reticulum

Gadd45

growth arrest and DNA damage-inducible 45

DMBA

dimethylbenzanthracene

WT

wild-type

ATF

activating transcription factor

MCD

methionine- and choline-deficient diet

UPLC-ESI-QTOFMS

ultraperformance liquid chromatography-electrospray ionization-quadrupole time-of-flight mass spectrometry

PCA

principal component analysis

OPLS

supervised orthogonal projection to latent structure

LPC

lysophosphatidylcholine

PC

phosphatidylcholine

qPCR

quantitative polymerase chain reaction

ALT

alanine aminotransferase

ALP

alkaline phosphatase

TG

triglyceride

NEFA

non-esterified fatty acid

MDA

malondialdehyde

H2O2

hydrogen peroxide

GSH

glutathione

TNFα

tumor necrosis factor α

TGFβ

transforming growth factor β

ROS

reactive oxygen species

Dr5

death receptor 5

Bim

BCL2-like 11

PLD

phospholipase D

PA

phosphatidic acid

PPAP

PA phosphatase

PE

phosphatidylethanolamine

CEPT

choline/ethanolaminephosphotransferase 1

FABP

FA-binding protein

SPT

serine palmitoyltransferase

PKC

protein kinase C

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Potential conflict of interest: The authors have declared that no conflict of interest exists.

References

  • 1.Neuschwander-Tetri BA. Hepatic lipotoxicity and the pathogenesis of nonalcoholic steatohepatitis: the central role of nontriglyceride fatty acid metabolites. Hepatology. 2010;52:774–788. doi: 10.1002/hep.23719. [DOI] [PubMed] [Google Scholar]
  • 2.Walenbergh SM, Koek GH, Bieghs V, Shiri-Sverdlov R. Non-alcoholic steatohepatitis: the role of oxidized low-density lipoproteins. J Hepatol. 2013;58:801–810. doi: 10.1016/j.jhep.2012.11.014. [DOI] [PubMed] [Google Scholar]
  • 3.Liebermann DA, Hoffman B. Gadd45 in stress signaling. J Mol Signal. 2008;3:15. doi: 10.1186/1750-2187-3-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Tian J, Locker J. Gadd45 in the liver: signal transduction and transcriptional mechanisms. Adv Exp Med Biol. 2013;793:69–80. doi: 10.1007/978-1-4614-8289-5_5. [DOI] [PubMed] [Google Scholar]
  • 5.Hollander MC, Kovalsky O, Salvador JM, Kim KE, Patterson AD, Haines DC, Fornace AJ., Jr Dimethylbenzanthracene carcinogenesis in Gadd45a-null mice is associated with decreased DNA repair and increased mutation frequency. Cancer Res. 2001;61:2487–2491. [PubMed] [Google Scholar]
  • 6.Maekawa T, Sano Y, Shinagawa T, Rahman Z, Sakuma T, Nomura S, Licht JD, Ishii S. ATF-2 controls transcription of Maspin and GADD45 alpha genes independently from p53 to suppress mammary tumors. Oncogene. 2008;27:1045–1054. doi: 10.1038/sj.onc.1210727. [DOI] [PubMed] [Google Scholar]
  • 7.Tanaka N, Matsubara T, Krausz KW, Patterson AD, Gonzalez FJ. Disruption of phospholipid and bile acid homeostasis in mice with nonalcoholic steatohepatitis. Hepatology. 2012;56:118–129. doi: 10.1002/hep.25630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Tanaka N, Takahashi S, Fang ZZ, Matsubara T, Krausz KW, Qu A, Gonzalez FJ. Role of white adipose lipolysis in the development of NASH induced by methionine- and choline-deficient diet. Biochim Biophys Acta. 2014;1841:1596–1607. doi: 10.1016/j.bbalip.2014.08.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Tanaka N, Takahashi S, Zhang Y, Krausz KW, Smith PB, Patterson AD, Gonzalez FJ. Role of fibroblast growth factor 21 in the early stage of NASH induced by methionine- and choline-deficient diet. Biochim Biophys Acta. 2015;1852:1242–1252. doi: 10.1016/j.bbadis.2015.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Matsubara T, Tanaka N, Krausz KW, Manna SK, Kang DW, Anderson ER, Luecke H, Patterson AD, Shah YM, Gonzalez FJ. Metabolomics identifies an inflammatory cascade involved in dioxin- and diet-induced steatohepatitis. Cell Metab. 2012;16:634–644. doi: 10.1016/j.cmet.2012.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Tanaka N, Moriya K, Kiyosawa K, Koike K, Gonzalez FJ, Aoyama T. PPARalpha activation is essential for HCV core protein-induced hepatic steatosis and hepatocellular carcinoma in mice. J Clin Invest. 2008;118:683–694. doi: 10.1172/JCI33594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Junqueira LC, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem J. 1979;11:447–455. doi: 10.1007/BF01002772. [DOI] [PubMed] [Google Scholar]
  • 13.Fujimori N, Tanaka N, Shibata S, Sano K, Yamazaki T, Sekiguchi T, Kitabatake H, Ichikawa Y, Kimura T, Komatsu M, Umemura T, Matsumoto A, Tanaka E. Controlled attenuation parameter is correlated with actual hepatic fat content in patients with non-alcoholic fatty liver disease with none-to-mild obesity and liver fibrosis. Hepatol Res. 2016;46:1019–1027. doi: 10.1111/hepr.12649. [DOI] [PubMed] [Google Scholar]
  • 14.Qu A, Shah YM, Manna SK, Gonzalez FJ. Disruption of endothelial peroxisome proliferator-activated receptor γ accelerates diet-induced atherogenesis in LDL receptor-null mice. Arterioscler Thromb Vasc Biol. 2012;32:65–73. doi: 10.1161/ATVBAHA.111.239137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Aoyama T, Yamano S, Waxman DJ, Lapenson DP, Meyer UA, Fischer V, Tyndale R, Inaba T, Kalow W, Gelboin HV, Gonzalez FJ. Cytochrome P-450 hPCN3, a novel cytochrome P-450 IIIA gene product that is differentially expressed in adult human liver. cDNA and deduced amino acid sequence and distinct specificities of cDNA-expressed hPCN1 and hPCN3 for the metabolism of steroid hormones and cyclosporine. J Biol Chem. 1989;264:10388–10395. [PubMed] [Google Scholar]
  • 16.Okiyama W, Tanaka N, Nakajima T, Tanaka E, Kiyosawa K, Gonzalez FJ, Aoyama T. Polyenephosphatidylcholine prevents alcoholic liver disease in PPARα-null mice through attenuation of increases in oxidative stress. J Hepatol. 2009;50:1236–1246. doi: 10.1016/j.jhep.2009.01.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Komatsu M, Kimura T, Yazaki M, Tanaka N, Yang Y, Nakajima T, Horiuchi A, Fang ZZ, Joshita S, Matsumoto A, Umemura T, Tanaka E, Gonzalez FJ, Ikeda S, Aoyama T. Steatogenesis in adult-onset type II citrullinemia is associated with down-regulation of PPARα. Biochim Biophys Acta. 2015;1852:473–481. doi: 10.1016/j.bbadis.2014.12.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Aoyama T, Peters JM, Iritani N, Nakajima T, Furihata K, Hashimoto T, Gonzalez FJ. Altered constitutive expression of fatty acid-metabolizing enzymes in mice lacking the peroxisome proliferator-activated receptor α (PPARα) J Biol Chem. 1998;273:5678–5684. doi: 10.1074/jbc.273.10.5678. [DOI] [PubMed] [Google Scholar]
  • 19.Gressner OA, Gressner AM. Connective tissue growth factor: a fibrogenic master switch in fibrotic liver diseases. Liver Int. 2008;28:1065–1079. doi: 10.1111/j.1478-3231.2008.01826.x. [DOI] [PubMed] [Google Scholar]
  • 20.Li LC, Li J, Gao J. Functions of galectin-3 and its role in fibrotic diseases. J Pharmacol Exp Ther. 2014;351:336–343. doi: 10.1124/jpet.114.218370. [DOI] [PubMed] [Google Scholar]
  • 21.Idle JR, Gonzalez FJ. Metabolomics. Cell Metab. 2007;6:348–351. doi: 10.1016/j.cmet.2007.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Prentki M, Madiraju SR. Glycerolipid metabolism and signaling in health and disease. Endocr Rev. 2008;29:647–676. doi: 10.1210/er.2008-0007. [DOI] [PubMed] [Google Scholar]
  • 23.Nakatsuka A, Matsuyama M, Yamaguchi S, Katayama A, Eguchi J, Murakami K, Teshigawara S, Ogawa D, Wada N, Yasunaka T, Ikeda F, Takaki A, Watanabe E, Wada J. Insufficiency of phosphatidylethanolamine N-methyltransferase is risk for lean non-alcoholic steatohepatitis. Sci Rep. 2016;6:21721. doi: 10.1038/srep21721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Liu J, Wu KC, Lu YF, Ekuase E, Klaassen CD. Nrf2 protection against liver injury produced by various hepatotoxicants. Oxid Med Cell Longev. 2013:305861. doi: 10.1155/2013/305861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bae MA, Rhee SD, Jung WH, Ahn JH, Song BJ, Cheon HG. Selective inhibition of activated stellate cells and protection from carbon tetrachloride-induced liver injury in rats by a new PPARgamma agonist KR62776. Arch Pharm Res. 2010;33:433–442. doi: 10.1007/s12272-010-0313-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hong L, Sun QF, Xu TY, Wu YH, Zhang H, Fu RQ, Cai FJ, Zhou QQ, Zhou K, Du QW, Zhang D, Xu S, Ding JG. New role and molecular mechanism of Gadd45a in hepatic fibrosis. World J Gastroenterol. 2016;22:2779–2788. doi: 10.3748/wjg.v22.i9.2779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Cazanave SC, Mott JL, Bronk SF, Werneburg NW, Fingas CD, Meng XW, Finnberg N, El-Deiry WS, Kaufmann SH, Gores GJ. Death receptor 5 signaling promotes hepatocyte lipoapoptosis. J Biol Chem. 2011;286:39336–39348. doi: 10.1074/jbc.M111.280420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Nagaya T, Tanaka N, Suzuki T, Sano K, Horiuchi A, Komatsu M, Nakajima T, Nishizawa T, Joshita S, Umemura T, Ichijo T, Matsumoto A, Yoshizawa K, Nakayama J, Tanaka E, Aoyama T. Down-regulation of SREBP-1c is associated with the development of burned-out NASH. J Hepatol. 2010;53:724–731. doi: 10.1016/j.jhep.2010.04.033. [DOI] [PubMed] [Google Scholar]
  • 29.Speranza F, Mahankali M, Henkels KM, Gomez-Cambronero J. The molecular basis of leukocyte adhesion involving phosphatidic acid and phospholipase D. J Biol Chem. 2014;289:28885–28897. doi: 10.1074/jbc.M114.597146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Brognard J, Newton AC. PHLiPPing the switch on Akt and protein kinase C signaling. Trends Endocrinol Metab. 2008;19:223–230. doi: 10.1016/j.tem.2008.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement
NIHMS914058-supplement.pdf (3MB, pdf)

RESOURCES