Effects Of Saturated Fatty Acid-induced Endoplasmic Reticulum Stress On Lipoapoptosis In Human Liver Cells And Its Underlying Molecular Mechanism

BACKGROUND AND PURPOSE:Nonalcoholic fatty liver disease (NAFLD) is a chronic metabolicdisorder characterized by fat accumulation in the liver, which is not due toexcessive alcohol consumption and definite factors damaged to liver. Theincidence of NAFLD has increased rapidly. Thus, NAFLD has emerged asa major public health issue in China. Clinically, NAFLD encompasses abroad spectrum of hepatic derangements from steatosis to nonalcoholicsteatohepatitis (NASH). The latter is characterized by hepatic fataccumulation coincident with inflammation, reduced liver function, fibrosis,and eventually liver cirrhosis. Many recent studies have reported thatNASH is a risk factor for cardiovascular disease, such as atherosclerosis,cerebral infarction and myocardial infarction. Therefore, NASH is crucialfor the prognosis of NAFLD. However, to date, the pathogenesis of NASHis still poorly understood. The charactiristic of NASH involves increasedlevels of free fatty acids (FFAs), hepatic fat accumulation and apoptosis of liver cells. The excess of FFAs overwhelms the capacity of the liver tooxidative and esterify FFAs leading to intrahepatic lipid deposition andactivation of inflammatory factors and apoptosis signal pathway, aphenomenon termed lipotoxicity, which includes lipoapoptosis. The latter isa prominent feature of NASH and is associated with severity of the disease.The molecular mechanism responsible for NASH and FFAs-inducedhepatocyte lipoapoptosis remains undefined. Endoplasmic reticulum (ER)is a highly dynamic organelle that synthesizes and processes secretory andtransmembrane proteins. The ER also serves important functions in calciumstorage and signaling, lipid biosynthesis, steroid hormone sythnesis and celldetoxification. A number of biochemical and physiologic stimuli associatedwith liver injury can disrupt ER homeostasis, imposing stress to the ER(ER stress), which includes unfolded protein response (UPR) andnon-unfolded protein response (non-UPR). Disruption of ER homeostasiscauses aberrant accumulation of unfolded or misfolded proteins in the ERlumen, triggering an evolutionarily conserved response, termed the UPR.Proximal sensors of UPR include protein kinase RNA-like ER kinase(PERK), inositol-requiring protein1(IRE1), and activating transcriptionfactor6(ATF6). Non-UPR involves phosphatidylinositide-3-OHkinase/protein kinase B (PI3K/Akt), mitogen-activated proteinkinase/extracellular signal-regulated protein kinases (MEK/ERK),glycogen synthase kinase-3(GSK-3), nuclear transcription factor (NF-κB), Toll-like receptor (TLR), which contributes to the apoptosis, inflammationresponse and immune response，respectively. The mild ER stress serves toovercome the stress stimulus; however, prolonged ER stress will promotecell apoptosis. Although ER stress is related to hepatocytes lipoapoptosis ofNASH, the definite mechanism responsible for the regulation of ER stresson NASH needs to be fully investigated. Furthermore, This study treatedhuman liver L02and HepG2cell lines with sodium palmitate, a saturatedfatty acid, to establish a steatosis model. In this model, we examined theeffects of sodium palmitate on the levels of proteins that are known to beassociated with ER stress, and the regulation of UPR and non-UPR onlipoapoptosis, to provide further mechanistic insights into the coreapoptotic machinery during saturated FFA-mediated lipotoxicity, whichcontributes to clearfy the definite mechanism of NASH.METHORDS:1. Sodium palmitate-induced L02and HepG2cells to lipoapoptosis1） Establishment of the steatosis model of L02and HepG2cells inducedby sodium palmitateThis study treated human liver L02and HepG2cell lines with variousdoses of sodium palmitate, a saturated fatty acid, for12，24and48h. MTTassay was used for cell viability, to adopt the optimal concentration ofsodium palmitate for the following experiments. Measurement ofintracellular triglyceride and oil red O staining was used to detect altered accumulation of lipid droplets and triglyceride in L02and HepG2cells.2） Sodium palmitate induced lipoapoptosis in L02and HepG2cells.Flow cytometry with Annexin V-PE binding and7-amino-actinomycinD (7-AAD) staining and Hoechst33258staining was used to detectapoptosis. Transmission electron microscopy was used to detectmorphological changes.2. The regulation of UPR on sodium palmitate-induced lipoapoptosis inL02and HepG2cells1） The expression levels of ER stress marker and UPR related proteins insodium palmitate-treated L02and HepG2cellsThe total RNAs and proteins were collected in L02and HepG2cellstreated with sodium palmitate for up to48h. Western blot analysis wasused to detect protein expression of glucose-regulated protein78(GRP78),phosphorylated PKR-like ER kinase (p-PERK), activating transcriptionfactor4(ATF4) and C/EBP-homologous protein (CHOP). RT-PCR wasused to detect mRNA expression of unspliced X-box binding protein-1(uXBP-1) and X-box binding protein-1splicing (XBP-1s).2） The effect of PERK gene knockdown on sodium palmitate-inducedPERK/ATF4/CHOP pathway and lipoapoptosisL02and HepG2cells were grown and transiently transfected withPERK shRNA or negative control shRNA from using transfection reagentPoly JetTMaccording to the manufacturer’s instructions. Twenty-four hours after transfection, the medium was changed to regular medium and the cellswere treated with sodium palmitate for an additional48h. Western blotanalysis was used to detect protein expression of total PERK, ATF4andCHOP. Flow cytometry was used to detect apoptosis. The proteinexpression levels and the number of apoptotic cells were comparedbetween the PERK shRNA plus sodium palmitate group and ControlshRNA plus sodium palmitate group.3. The regulation of GSK-3β (non-UPR) on sodium palmitate-inducedlipoapoptosis in L02and HepG2cells1） The expression levels of GSK-3β after sodium palmitate treatment andthe effect of Inhibition of GSK-3β activity or expression on sodiumpalmitate-induced apoptosis in steatotic L02and HepG2cells.L02and HepG2cells were treated with sodium palmitate for up to48h, and then subjected to protein extraction and western blot analyses. Theprotein expression levels of phosphorylated GSK-3β and total GSK-3βwere compared between model group and Control group.L02and HepG2cells were treated with sodium palmitate for up to48h, in the presence or absence of lithium chloride, a GSK-3β inhibitor, andthen subjected to protein extraction, western blot analyses, and flowcytometry with Annexin-PE and7AAD double staining. The proteinexpression levels of phosphorylated GSK-3β and the number of apoptoticcells were compared between inhibitor group and model group at the same time point.The cells were grown and transfected with GSK-3β shRNA ornegative control shRNA for24h, then treated with sodium palmitate for anadditional48h, and subjected to protein extraction, western blot analysesand flow cytometry with Annexin-PE and7AAD double staining. Theprotein expression levels of total GSK-3β and the number of apoptotic cellswere compared between the GSK-3β shRNA plus sodium palmitate groupand Control shRNA plus sodium palmitate group.2） Inhibition of GSK-3β activity or expression on regulation of sodiumpalmitate-induced c-Jun NH2-terminal kinase (JNK) phosphorylation,Bcl-2-associated X protein (Bax) upregulation, Caspase-3activity andactivation of UPR in L02and HepG2cells.L02and HepG2cells were treated with sodium palmitate for up to48h, in the presence or absence of lithium chloride, a GSK-3β inhibitor, andthen subjected to protein extraction, western blot analyses, and caspase-3activity measurements with a caspase-3colorimetric assay kit.The protein expression levels of phosphorylated JNK, Bax and Caspase-3activity were compared between inhibitor group and model group at thesame time point.The cells were grown and transfected with GSK-3β shRNA ornegative control shRNA for24h, then treated with sodium palmitate for anadditional48h, and subjected to protein extraction and western blot analyses. The protein expression levels of phosphorylated JNK, Bax,GRP78, phosphorylated inositol-requiring protein1(IRE1),phosphorylated PERK and Caspase-3activity were compared between theGSK-3β shRNA plus sodium palmitate group and Control shRNA plussodium palmitate group.RESULTS:1. Sodium palmitate-induced L02and HepG2cells to lipoapoptosis1） MTT assay data showed that sodium palmitate had very little effect atlow dose and short exposure time, while higher doses (i.e.,144and180μmol/L sodium palmitate) inhibited the growth of cells by more than50%at48h. Thus, we adopted the lower cytotoxic dose of108μmol/L sodiumpalmitate for the following experiments.2） Oil Red O staining and triglyceride quantification revealed prominentlipid accumulation in the cytoplasm and increased triglyceride levels,respectively.3） There was no significant difference in the number of apoptotic cells inL02and HepG2cells treated with sodium palmitate for24in comparisonwith the control. However, incubation of hepatocytes with108μmol/Lsodium palmitate for48h caused a significant increase in the number ofapoptotic cells, compared to the control (P <0.05). The Hoechst33258stainand electron microscopic revealed that L02and HepG2cells treated withsodium palmitate for48h displayed apoptotic cells with nuclear fragmentation, chromatin condensation and formation of apoptotic bodies,respectively.2. The regulation of UPR on sodium palmitate-induced lipoapoptosis inL02and HepG2cells1） The data showed that GRP78expression was obviously upregulated ina time-dependent manner in L02cells. In contrast, GRP78expressionreached a peak at24h and slightly decreased24h after treatment withsodium palmitate in HepG2cells (P <0.05).2） The data showed that the levels of phosphorylated PERK increased insodium palmitate-treated L02and HepG2cells over time. The same is truefor the expression of the transcription factor ATF4and CHOP (P <0.05).uXBP-1expression was stimulated only after the12h treatment withsodium palmitate, as compared to the control HepG2cells. In L02cells,there was no significant difference in uXBP-1expression in either of thesodium palmitate treatment groups in comparison with the control. Cellstreated with sodium palmitate did not have the spliced version of XBP-1.3） PERK1shRNA plasmid had the optimal silencing effect among thethree constructed PERK shRNA plasmids. In L02and HepG2cells inwhich PERK shRNA had been transiently transfected, PERK proteinexpression was significantly suppressed, as compared to the cells of thenegative control. ShRNA-targeted knockdown of PERK also suppressedATF4and CHOP induction by palmitate sodium (P <0.05). 4） The rate of apoptosis induced by sodium palmitate treatment was alsoreduced by PERK shRNA transfection, compared to the control shRNA (P<0.05).3. The regulation of GSK-3β（non-UPR）on sodium palmitate-inducedlipoapoptosis in L02and HepG2cells1） our data showed that sodium palmitate induced the dephosphorylationof GSK-3β at Ser-9in a time-dependent manner, suggesting GSK-3βactivity progressively increased after sodium palmitate treatment (P <0.05).2） Compared with the model group, the level of p-GSK-3β (Ser9) ininhibitor group had significantly increased (P<0.05). Our data suggestedthat lithium chloride inhibited GSK-3β activity by induction ofphosphorylation of GSK-3β at Ser-9. In addition, lithium chloride reducedsodium palmitate-induced apoptosis in steatotic L02and HepG2cells, ascompared with the model group (P <0.05).3） GSK-3β1shRNA plasmid had the optimal silencing effect among thethree constructed GSK-3β shRNA plasmids. The rate of apoptosis inducedby sodium palmitate treatment was reduced by GSK-3β shRNAtransfection, compared to the control shRNA (P <0.05).4） Our data showed that phosphorylated JNK levels progressivelyincreased in sodium palmitate-treated L02and HepG2cells, while Bax, apro-apoptotic effector downstream of JNK, was also significantlyupregulated at24to48h after sodium palmitate treatment. The same is true for the Caspase-3activity. Inhibition of GSK-3β expression or activitysuppressed sodium palmitate-induced JNK phosphorylation, Baxupregulation and Caspase-3activity (P <0.05).5） Our data showed that treatment of L02and HepG2cells with sodiumpalmitate promoted a significant increase in GRP78expression andphosphorylation of PERK and IRE1(P <0.05), but inhibition of GSK-3βexpression using GSK-3β shRNA transfection did not affect GRP78expression or sodium palmitate-induced phosphorylation of PERK andIRE1(P>0.05).CONCLUSION:1. Saturated fatty acid-induced L02and HepG2cells to lipoapoptosis.2. Saturated fatty acid-induced lipoapoptosis in L02and HepG2wereenacted through the PERK/ATF4/CHOP signaling pathway.3. Saturated fatty acid-induced lipoapoptosis of L02and HepG2cells wasregulated by GSK-3β activation, which may depend on JNKphosphorylation and Bax upregulation.