Inositol-requiring enzyme 1α links palmitate-induced mTOR activation and lipotoxicity in hepatocytes

Abstract

Hepatic lipotoxicity, hepatocyte dysfunction/cell death induced by saturated fatty acids (SFA), plays a central role in the pathogenesis of nonalcoholic fatty liver disease (NAFLD); however, the underlying mechanisms remain unclear. Palmitate is the most abundant SFA in the circulation. In this study, via a small-scale screening of chemical inhibitors using AML12 hepatocytes, we identified mechanistic target of rapamycin (mTOR) complex 1 (mTORC1) to be a culprit in palmitate-induced cell death in hepatocytes in that mTOR inhibition is protective against palmitate-induced cell death. The protective effect of mTORC1 inhibition is independent of autophagy induction, as autophagy inhibition failed to ablate the mTORC1 inhibitor-conferred protection. We have previously reported that the endonuclease activity of inositol-requiring enzyme 1α (IRE1α), one of three canonical signaling pathways of endoplasmic reticulum (ER) stress, was implicated in palmitate-induced cell death in hepatocytes. The continuous mechanistic investigation in this study uncovered that IRE1α is a downstream target of mTORC1 activation upon palmitate exposure and the inhibition of either its endonuclease activity or kinase activity protects against the lipotoxic effect of palmitate. Our research further revealed that protein palmitoylation is potentially involved in palmitate-induced mTORC1 activation and lipotoxicity in hepatocytes. 2-Bromopalmitate, a protein palmitoylation inhibitor, ameliorated palmitate-triggered mTORC1 activation, concomitant with the protection of lipotoxicity in hepatocytes. Collectively, our data have identified that mTORC1 and ER stress are coordinately implicated in hepatocyte cell death in response to palmitate exposure and suggest that this pathway may potentially serve as a therapeutic target for the treatment of NAFLD as well as other metabolic disorders involving lipotoxicity.

INTRODUCTION

Nonalcoholic fatty liver disease (NAFLD) is the general name for a group of pathological liver conditions ranging from steatosis and steatohepatitis to fibrosis/cirrhosis and hepatocellular carcinoma and considered to be a hepatic manifestation of metabolic syndrome. Despite intensive research and much progress being made over the last several decades, detailed cellular and molecular mechanisms underlying its initiation and progression remain unclear. The lack of understanding on its exact pathogenesis greatly stagnates the development of efficacious therapies (18, 29, 30).

Lipotoxicity, a term created to depict the harmful impact of lipid accumulation in nonadipose tissues (ectopic lipid deposition), represents an underlying cause of NAFLD (3, 13). NAFLD development is intimately associated with obesity and insulin resistance, two prominent components of metabolic syndrome, with elevated levels of circulatory free fatty acids (FFAs) being one of its major features (4, 16). Unlike unsaturated fatty acids, saturated fatty acids (SFA) are toxic to cells, inducing cellular toxicity/cell death in a broad range of mammalian cell types, including hepatocytes (27, 28, 31). Accumulated evidence supports that it is not triglyceride formation, but incremental plasma levels of FFAs, especially SFA, that contribute to NAFLD pathogenesis (26). NAFLD development is associated with increased plasma concentrations of palmitate (a 16-carbon SFA), one of the most abundant FFA found in human circulation (2). The mechanisms underpinning palmitate-induced lipotoxicity in hepatocytes remain elusive and obviously multifaceted. Factors/pathways implicated in palmitate-triggered cellular lipotoxic effect include ceramides, lysophosphatidylcholine, endoplasmic reticulum (ER) stress, c-Jun NH₂-terminal kinase (JNK) activation, and oxidative stress, among others (8, 10, 13, 27, 31, 33, 44).

Among all proposed mechanisms, ER stress represents a central mechanism implicated in palmitate-induced lipotoxicity. ER is an organelle controlling cellular protein folding and assembly. In response to disrupted balance between the unfolded proteins and the capacity of ER folding machinery inside the ER lumen, ER stress ensues, leading to the unfolded protein response (UPR) aimed to facilitate the restoration of ER homeostasis. UPR consists of three canonical signaling pathways: protein kinase R (PKR)‐like ER kinase (PERK), inositol-requiring enzyme 1α (IRE1α), and activating transcription factor 6 (ATF6), with IRE1α being the most conserved one, possessing both endoribonuclease activity and kinase activity, leading to generation of X-box binding protein 1 (XBP1s) and JNK activation, respectively (42, 43). Palmitate exposure induces ER stress in a variety of cell types including hepatocytes (8, 33, 44), while alleviation of ER stress using chemical ER chaperones or blockade of UPR pathways via either pharmacological or genetic approach protects against palmitate-induced lipotoxicity (6, 21, 32).

Mechanistic target of rapamycin (mTOR) is an evolutionarily conserved protein serine/threonine kinase and a master regulator in promoting growth and cellular anabolic processes in response to growth factors and nutrient excess, including proliferation, differentiation, and protein synthesis. It forms two functional complexes, mTOR complex 1 (mTORC1) and 2 (mTORC2), with distinctively different components and substrates. The mTORC1 is sensitive to rapamycin and regulates anabolic processes via directly phosphorylating ribosomal p70S6 kinase (p70S6K) and eIF4E-binding protein (24, 45).

Given the complex signaling pathways surrounding both ER stress/UPR and mTOR, it is conceivable that there must be extensive cross talk between these two seemingly independent pathways. Surprisingly, it is only recently that evidence has emerged suggesting that mTORC1-mediated signaling events and UPR pathways are intertwined (7, 22). For example, ATF6, one of three canonical arms of UPR, activates mTORC1 at least partially through upregulation of Ras homolog enriched in brain (RHEB) gene expression, a physiological activator of mTORC1 (7). Furthermore, both tunicamycin and thapsigargin (two well-established ER stress inducers) strongly activate mTORC1 while mTORC1 inhibition prevents ER stress-induced cell death, suggesting that the UPR and mTORC1 conspire to regulate cell death (22).

We have previously reported that the endoribonuclease activity of the IRE1α pathway of UPR was attributable to palmitate-triggered cell death in HepG2 cells, a human hepatoma cell line (40). In this study, we performed in vitro experiments primarily in AML12 cells, a nontransformed mouse hepatocyte cell line, to explore the role of mTORC1 in palmitate-induced lipotoxicity in hepatocytes and the potential mechanisms underlining this. We discovered that palmitate activates mTORC1 via protein palmitoylation. Importantly, mTORC1 inhibition alleviated lipotoxic cell death induced by palmitate in hepatocytes, indicating that mTORC1 activation plays a critical role in regulating lipotoxicity in hepatocytes. Further investigation revealed that mTORC1 inhibition prevented palmitate-triggered IRE1α activation and inhibition of either its endoribonuclease or kinase activity conferred protection against palmitate-induced cell death. Together, our results provide evidence that the mTORC1-IRE1α pathway mechanistically contributes to lipotoxic cell death in hepatocytes.

MATERIALS AND METHODS

Reagents.

Cell culture and transfection reagents were purchased from Invitrogen (Carlsbad, CA). HepG2 human hepatoma cells and AML12 murine hepatocytes were purchased from the American Type Culture Collection (ATCC, Manassas, VA). Palmitate, oleate, and bafilomycin A1 were purchased from Sigma-Aldrich (St. Louis, MO). SP600125 was from Biovision Inc. (Milpitas, CA). SBI0206965, SC79, Torin-1, and rapamycin were purchased from APExBIO Technology LLC (Houston, TX). STF083010 was purchased from Tocris Bioscience (Ellisville, MO). For Western blotting, antibodies for phosphorylated (p-)4E-BP1 (T37/46) (product no. 2855s), 4E-BP1 (product no. 9452), p-S6 (S235/236) (product no. 4858s), S6 (product no. 2317s), p-P70S6 (T389) (product no. 2708s), P70S6 (product no. 9234), LC3B (product no. 2775), p-SAPK/JNK (T183/Y185) (product no. 9251), SAPK/JNK (product no. 9252), p-AKT (S473) (product no. 4058), and β-actin (product no. 4970) were purchased from Cell Signaling Technology (Beverly, MA). Anti-p-IRE1α (Ser724) (catalog no. PA1-16927) was purchased from Thermo Fisher Scientific (Rockford, IL) and IRE1α (catalog no. NB100-2323) antibody from NOVUS Biologicals, LLC (Centennial, CO). The secondary antibody anti-IgG goat polyclonal antibody (IRDye 800CW) (catalog no. 102673-330), was from VWR (Batavia, IL). Anti-membrane palmitoylated protein 1 (MPP1) antibody (catalog no. PB10078) was from Boster Biological Technology (Pleasanton, CA).

Cell culture.

AML12 hepatocytes were cultured as monolayers with Dulbecco’s modified Eagle’s medium (DMEM)-F12 supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, 0.1 mg/mL streptomycin, 40 ng/mL dexamethasone, and ITS (containing 5 mg/L insulin, 5 mg/L transferrin, and 5 μg/L selenium). HepG2 cells were cultured in DMEM medium containing 10% FBS. Cells were grown in 75-cm² flasks and kept at 37°C in a humidified atmosphere of air and 5% CO₂. Cells were grown at 80% of confluence before the exposure to drugs in the various experiments.

Lactate dehydrogenase release assay.

Cells (1 × 10⁵/mL) were seeded in 24-well plates and cultured overnight. After indicated treatments, culture medium was collected and detected with a Pierce LDH Cytotoxicity Assay kit (Thermo Scientific Inc., Rockford, IL) according to the manufacturer’s instructions. The absorption at OD510 was measured with a microplate spectrophotometer (SPECTRAmax 340PC; Molecular Devices Corp., Sunnyvale, CA). The relative lactate dehydrogenase (LDH) release levels were scaled as folds of the controlled untreated group.

MTT assay.

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) method was used to determine the effects of torin-1 and rapamycin protection against palmitate-induced cell death in hepatocytes. Briefly, 5 × 10³ cells/well were evenly distributed and incubated on 96-well plates overnight. After indicated treatments, the medium in each well was replaced with 20 µL of MTT (5 mg/mL in PBS) and incubated at 37°C for 4 h. The purple-blue formazan precipitate was dissolved in 100 µL dimethyl sulfoxide and the optical density was measured at a wavelength of 490 nm on a microplate spectrophotometer reader (SPECTRAmax 340PC; Molecular Devices Corp., Sunnyvale, CA).

Cell lysates and Western blotting detection.

Total protein from hepatocytes was obtained using lysis buffer from Thermo Scientific Inc. (Rockford, IL). Samples were incubated on ice with frequent vortex for 15 min and centrifuged for 10 min at 12,000 g. The protein concentration was determined using an Enhanced BCA Protein Assay Kit from Thermo Scientific Inc. (Rockford, IL) in accordance with the manufacturer’s instructions. Equal amounts of protein (30 µg) were subjected to 10% or 8% SDS-PAGE depending on the molecular weight of the desired proteins and were transferred to a nitrocellulose transfer membrane (Pall Corporation, New Port Richey, FL). After transfer, membranes were blocked in 5% (wt/vol) nonfat dry milk in PBS-0.1% Tween 20 (PBST) and probed with specific antibodies. After incubation with primary antibodies, membranes were washed with PBST and then incubated in fluorescein-conjugated secondary antibody (catalog no. 102673-330 from VWR) at 1:10,000 dilution in blocking buffer at room temperature for 1 h, followed by washing with PBST again. Immunoreactive bands of predicted molecular mass were visualized using a LI-COR Odyssey CLx system and quantified with Image Studio version 4.0.

siRNA transfection.

Cultured cells were transfected with mouse ATG5 siRNA (Santa Cruz, sc-41446). Cells in exponential growth phase were plated in 6-well plates or 24-well plates and allowed to adhere for 24 h before being transfected with siRNA. Transfection was performed according to Invitrogen’s Lipo 2000 protocol. In brief, cells were cultured to 60–70% confluence in a 6-well culture plate and transfected with 8 μL of siRNA (20 mM) using 8 μL Lipo 2000 or a 24-well plate transfected with 2 μL of siRNA (20 mM) using 2 μL Lipo 2000. As a control, cells were transfected with scrambled RNA under the same condition. The result of transfection was identified by RT-PCR after 12-h siRNA transfection.

Quantitative real-time RT-PCR.

Total RNA from hepatocytes was isolated via phenol-chloroform extraction. For each sample, 1.0 μg of total RNA was reverse transcribed using a high-capacity cDNA reverse transcription kit (Applied Biosystems, Vilnius, Lithuania). The cDNA was amplified in MicroAmp Optical 96-well reaction plates with a SYBR Green PCR Master Mix (Applied Biosystems, Warrington, UK) on a Life Technologies ABI 7500 FAST sequence detection system. Relative gene expression was calculated after normalization by a housekeeping gene (mouse GAPDH mRNA).

Primers are as follows: ATG5: 5′-AAGTGAGCCTCAACCGCATCCT-3′ and 5′CTTGCATCAAGTTCAGCTCTTCC-3′; GAPDH: 5′-ATGCCAGTGAGCTTCCCGTTCAG-3′ and 5′-CATCACTGCCACCCAGAAGACTG-3′; XBP1: 5′-TAGACCTCTGGGAGTTCCTCCA-3′ and 5′-TGGACTCTGACACTGTGCCTC-3′; XBP1s: 5′-GAGGCAACAGTGTCAGAGTCC-3′ and 5′-TGCTGAGTCCGCAGCAGGTG-3′; XBP1u: 5′-GAGGCAACAGTGTCAGAGTC-3′ and 5′-CTCAGACTATGTGCACCTCTGC-3′.

Autophagic flux.

The autophagic flux was measured as described previously (25). In brief, cells were pretreated with bafilomycin A1, an endosome acidification inhibitor, before unc-51-like kinase 1 (ULK1) inhibitor addition or after overnight ATG5 siRNA transfection in the presence or absence of torin-1. The autophagic flux was determined by detecting LC3‐II expression by Western blot analysis.

Statistics.

All data are expressed as means ± SD. Statistical analysis was performed using a one-way ANOVA and was analyzed further by post hoc test with Fisher’s least significant difference (LSD) for statistical differences.

RESULTS

Palmitate activates mTORC1 in hepatocytes.

Palmitate (a 16-C saturated fatty acid) and oleate (an 18-C monounsaturated fatty acid) are the two most abundant free fatty acids in the circulation. Their effects on mTORC1 activation in hepatocytes were directly examined in cultured AML12 cells. The dose-dependent effects of palmitate exposure on mTORC1 activation are shown in Fig. 1A. During a 16-h exposure period, palmitate at both 0.2 and 0.4 mM triggered mTORC1 activation, evidenced by a noticeable increase in protein abundance of phosphorylated S6K1 (p-S6K1) and 4E-BP1 (p-4E-BP1), two primary downstream targets of mTORC1 activation. The time course effects of palmitate on mTORC1 activation were determined through treating AML12 cells with 0.4 mM palmitate. Protein samples were collected at 0, 2, 4, 8, and 16 h after palmitate addition, and protein abundance of p-S6K1, p-S6 (a direct target of S6K1 activation), and p-4E-BP1 was measured by Western blotting. As shown in Fig. 1B, while an observable elevation of both p-S6K1 and p-4E-BP1 was detected at the 4 h time point, a palpable increase of p-S6 was detected 2 h after palmitate exposure, which reached its peak at the 8-h time point and was maintained over the whole 16-h experimental period. Whereas palmitate exposure strongly induced mTORC1 activation, oleate at the same concentration (0.4 mM) exerted no effect on mTORC1 activity (Fig. 1C).

Fig. 1.

Palmitate activates mechanistic target of rapamycin complex 1 (mTORC1). A: AML12 cells were treated with the indicated concentrations of palmitate for 16 h. Protein abundance of phosphorylated and total S6K1 and p-4E-BP1 was detected by Western blotting; The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Student’s t test was used for statistical evaluation (*P < 0.05, **P < 0.01, ***P < 0.0001 versus corresponding control). B: AML12 cells were treated with 0.4 mM palmitate for the indicated experimental durations. Protein abundance of phosphorylated and total S6K1, S6, and p-4E-BP1 was determined by Western blotting; The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Student’s t test was used for statistical evaluation (*P < 0.05, **P < 0.01 versus time 0). C: AML12 cells were treated with palmitate (PA, 0.4 mM), oleate (OA, 0.4 mM), and torin-1 (0.25 μM) for 16 h. Protein abundance of p-S6 and total S6 protein was determined by Western blotting. The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Student’s t test was used for statistical evaluation (*P < 0.05, ***P < 0.0001 versus corresponding control). Alb, albumin; UT, untreated.

mTORC1 inhibition protects against palmitate-induced cell death in hepatocytes.

Previous studies, including ours, documented that palmitate exposure induced cell death in hepatocytes (25, 39, 41). To determine whether mTORC1 activation contributes to palmitate-induced hepatocyte cell death, we pretreated both AML12 and HepG2 cells with torin-1, a chemical inhibitor for both mTORC1 and mTORC2, or rapamycin, a specific inhibitor for mTORC1, for 2 h before palmitate addition (0.4 mM). The inhibitory efficacy of torin-1 on mTORC1 activation is shown in Fig. 1C. Cell death was determined after a 16-h palmitate exposure by both LDH release and MTT measurements. As shown in Fig. 2, mTOR inhibitors attenuated cell death (Fig. 2A for AML12 cells; Fig. 2B for HepG2 cells), suggesting that mTORC1 activation contributes to palmitate-induced lipotoxicity in hepatocytes.

Fig. 2.

Mechanistic target of rapamycin complex 1 (mTORC1) inhibition protects against palmitate-induced cell death in hepatocytes. A: AML12 cells were pretreated with either torin-1 (0.25 µM) or rapamycin (0.25 µM) for 2 h before palmitate (0.4 mM) exposure. Cell death was determined via both lactate dehydrogenase (LDH) release assay and MTT measurement after a 16-h exposure. B: HepG2 cells were pretreated with torin-1 (0.25 µM) for 2 h before palmitate (0.4 mM) exposure. Cell death was determined after a 16-h exposure by measuring medium LDH release. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). UT, untreated.

The protective effect of mTORC1 inhibition against lipotoxicity is autophagy independent.

Autophagy induction protects against palmitate-induced hepatocyte cell death (25, 39, 41). mTORC1 is a prominent physiological inhibitor of autophagy, partially through phosphorylating the autophagy regulatory complex containing unc-51-like kinase 1 (ULK1), the mammalian Atg13 protein, and focal adhesion kinase interacting protein of 200 kDa (FIP200) (24, 45). Based on these previous findings, it is conceivable that mTORC1 inhibition may prevent palmitate-induced cell death via inducing autophagy. To test the potential involvement of autophagy induction in the protective effect of mTORC1 inhibition on cell death, we disrupted autophagy induction process by either pretreating AML12 hepatocytes with SBI0206965, a ULK1 inhibitor, or transfecting cells with ATG5 siRNA for overnight before palmitate exposure. The overnight ATG5 siRNA transfection led to an ∼60% decrease of ATG5 expression (Fig. 3A), and both ULK1 inhibitor and ATG5 siRNA knockdown resulted in impaired autophagy induction in AML12 cells, which was confirmed by autophagic flux analysis (Fig. 3B). The level of cell death in response to a 16-h palmitate exposure in control cells was comparable to ATG5 siRNA knockdown cells (Fig. 3C). Unexpectedly, a subtle but significant protection with ULK1 inhibitor was observed (Fig. 3D). Nevertheless, the protective effects of mTORC1 inhibition by both torin-1 and rapamycin were retained in these autophagy-deficient cells (Fig. 3, C and D), suggesting that mTORC1 inhibition protects against lipotoxicity via an autophagy-independent mechanism.

Fig. 3.

The antilipotoxic effect of mechanistic target of rapamycin complex 1 (mTORC1) inhibition is autophagy independent. A: AML12 cells were transfected with ATG5 siRNA overnight before palmitate (0.4 mM) exposure. ATG5 gene expression was quantified by real-time RT-PCR. Data are expressed as means ± SD; n = 3 separate experiments. Differences between 2 groups were determined by Student’s t test (*P < 0.05 versus control siRNA). B: AML12 cells were either pretreated with ULK-1 inhibitor SBI0206965 (5 µM) for 2 h or transfected with ATG5 siRNA overnight, followed by bafilomycin A1 (BAF, 100 nM) ± torin-1. LC3-II protein abundance was determined by Western blotting. C: AML12 cells were pretreated with ULK-1 inhibitor SBI0206965, torin-1, or rapamycin for 2 h, followed by 0.4 mM palmitate exposure. Cell viability was determined by lactate dehydrogenase (LDH) release measurement after a 16-h exposure. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis. Bars with different superscript letters differ significantly (P < 0.05). D: AML12 cells were transfected with ATG5 siRNA overnight and pretreated with either torin-1 or rapamycin for 2 h, followed by 0.4 mM palmitate exposure. Cell death was determined by LDH release measurement after a 16-h exposure. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis. Bars with different superscript letters differ significantly (P < 0.05). UT, untreated.

JNK activation contributes to mTORC1-mediated lipotoxicity.

The critical involvement of JNK activation in palmitate-induced lipotoxicity has been well documented in many cell types, including hepatocytes (8, 22). In line with these previous reports, including ours, both time course (Fig. 4A)- and dose (Fig. 4B)-dependent activation of JNK upon palmitate exposure were observed. Also, pretreatment of AML12 cells with JNK inhibitor attenuated palmitate-induced cell death (Fig. 4C). Despite being well reported, the exact mechanism underlying the palmitate-instigated JNK activation remains unclear. Given the fact that both mTORC1 inhibitor and JNK inhibitor exerted protection against cell death induced by palmitate, we examined whether mTORC1 represents a critical link between palmitate exposure and JNK activation. JNK phosphorylation status was examined in AML12 cells exposed to palmitate for 16 h with and without a 2-h pretreatment with mTORC1 inhibitors. As shown in Fig. 4, D and E, both mTORC1 inhibitors attenuated palmitate-triggered JNK activation, suggesting that mTORC1 activation contributes to palmitate-induced JNK activation.

Fig. 4.

c-Jun NH₂-terminal kinase (JNK) activation contributes to mechanistic target of rapamycin complex 1 (mTORC1)-mediated lipotoxic effect. A: AML12 cells were treated with 0.4 mM palmitate for the indicated times. JNK activation was determined by Western blot detection of phosphorylated (p-) and total JNK protein abundance. The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Student’s t test was used for statistical evaluation (*P < 0.05, **P < 0.01 versus time 0). B: AML12 cells were treated with the indicated concentrations of palmitate for 16 h. JNK activation was determined by Western blot detection of phosphorylated and total JNK protein abundance; The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Student’s t test was used for statistical evaluation (*P < 0.05, **P < 0.01, ***P < 0.0001 versus corresponding control). C: AML12 cells were pretreated with JNK inhibitor SP600125 (20 µM) for 2 h before palmitate (0.4 mM) addition. Cell death was determined by lactate dehydrogenase (LDH) release measurement after a 16-h palmitate exposure. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (****P < 0.0001). D and E: AML12 cells were pretreated with either rapamycin (D) or torin-1 (E) for 2 h before palmitate (0.4 mM) exposure. Phospho-JNK and total JNK protein abundance were determined by Western blotting. The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (*P < 0.05, **P < 0.01, ***P < 0.001). UT, untreated.

IRE1α activation links mTORC1 and JNK activation.

One of the well-established signaling pathways to activate JNK is the induction of IRE1α, one of three sensors activated during ER stress/UPR, via its kinase activity (21, 22). The inhibitory impact of mTOR inhibition on palmitate-induced JNK activation (Fig. 4D) encouraged us to look further at whether IRE1α activation is attributable to mTORC1-induced JNK activation in response to palmitate exposure. IRE1α has both endoribonuclease and kinase activities, activation of which leads to transcription factor XBP1 splicing and JNK activation, respectively (Fig. 5A). We have previously reported that inhibition of the endoribonuclease activity of IRE1α with STF083010 protected HepG2 cells against palmitate-induced cell death (40). In this study, a comparable effect was observed in AML12 cells. STF083010 attenuated palmitate-induced lipotoxicity in AML12 hepatocytes (Fig. 5B). Furthermore, chemical inhibitors for other two canonical pathways activated in response to ER stress, GSK2606414 for PERK and Ceapin-A7 for ATF6, failed to protect against palmitate lipotoxicity (Fig. 5B). In addition to inducing XBP1 splicing, IRE1α activation also instigates JNK activation by interacting with TRAF2 and ASK1 during ER stress (21, 22). The effect of palmitate exposure with and without torin-1 pretreatment on IRE1α activation was subsequently examined. Palmitate exposure activated IRE1α, which was made evident by both the enhanced IRE1α protein phosphorylation (at Ser 724) (Fig. 5C) and increased XBP1s-to-XBP1u ratio (Fig. 5D). However, this activation was ablated when cells were pretreated with torin-1 (Fig. 5, C and D), suggesting that IRE1α activation links mTORC1 signaling and JNK activation in the process of cell death in response to palmitate exposure. Together, these data suggest that IRE1α is the downstream target of mTORC1 activation in response to palmitate exposure and the activations of both its endoribonuclease and kinase activity contribute to palmitate lipotoxicity in hepatocytes.

Fig. 5.

Inositol-requiring enzyme 1α (IRE1α) activation links mechanistic target of rapamycin complex 1 (mTORC1) and c-Jun NH₂-terminal kinase (JNK) activations. A: 3 canonical pathways activated by endoplasmic reticulum (ER) stress and specific inhibitors employed. IRE1α activation leads to X-box binding protein 1 (XBP1) splicing and JNK activation via its endonuclease and kinase activities. B: AML12 cells were pretreated with the inhibitors for 2 h before palmitate (0.4 mM) addition. Cell death was determined by measuring lactate dehydrogenase (LDH) release after a 16-h exposure. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis. Bars with different superscript letters differ significantly (P < 0.05). C: AML12 cells with or without a 2-h pretreatment with torin-1 were exposed to palmitate (0.4 mM) for 16 h. Protein abundance of total and phosphorylated (p-)IRE1α was determined by Western blotting. The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (*P < 0.05). D: AML12 cells with or without a 2-h pretreatment with torin-1 were exposed to palmitate (0.4 mM) for 16 h. mRNA levels of spliced Xbp1 (Xbp1s) and unspliced Xbp1 (Xbp1u) were determined by RT-qPCR and Xbp1s-to-Xbp1u ratios calculated. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (*P < 0.05). ATF, activating transcription factor; PERK, protein kinase R (PKR)‐like ER kinase. UT, untreated.

Protein palmitoylation potentially contributes to palmitate-induced mTORC1 activation and lipotoxicity.

The exact mechanisms underlying palmitate-triggered mTORC1 activation and resultant lipotoxicity in hepatocytes remain ambiguous. Protein palmitoylation is required for the complex mTORC1 activation processes (35, 36). To test the hypothesis that protein palmitoylation is attributable to palmitate-induced mTORC1 activation and cell death in hepatocytes, both AML12 and HepG2 cells were exposed to palmitate with or without a 2-h pretreatment with 2-bromopalmitate (2-Br), a chemical inhibitor of protein palmitoylation. Cell death was determined after a 16-h cell stimulation. As shown in Fig. 6, 2-Br pretreatment protected against palmitate-induced cell death in both AML12 (Fig. 6A) and HepG2 cells (Fig. 6B). Considering that 2-Br was also reported to be an inhibitor of fatty acid β-oxidation (9, 20), we subsequently examined whether the antilipotoxic activation of 2-Br was derived from its inhibitory effect on fatty acid β-oxidation. To test this, we pretreated both AML12 and HepG2 cells with etomoxir, a specific inhibitor of carnitine palmitoyltransferase (CPT)-1, an enzyme required for long-chain fatty acid β-oxidation, for 2 h before palmitate exposure. Cell death was determined after 16-h treatment. As shown in Fig. 6, C and D, etomoxir failed to exhibit protection against palmitate lipotoxicity in hepatocytes. To determine the effect of 2-Br on protein palmitoylation, we subsequently examined the effect of 2-Br pretreatment on protein palmitoylation using anti-membrane palmitoylated protein 1 (MPP1) antibody. As shown in Fig. 6E, palmitate exposure led to a slight, but significant, increase in MPP1, which was ameliorated by 2-Br pretreatment. Finally, we examined the effect of 2-Br pretreatment on palmitate-triggered mTORC1 activation. Our data showed that the protective effect of 2-Br was associated with an alleviated mTORC1 activation in response to palmitate exposure (Fig. 6F). Together, these results suggest that protein palmitoylation is potentially implicated in palmitate-triggered mTORC1 activation and lipotoxicity.

Fig. 6.

Protein palmitoylation is implicated in palmitate-triggered mechanistic target of rapamycin complex 1 (mTORC1) activation. A: AML12 cells were pretreated with 2-bromopalmitate (2-Br, 100 μM) for 2 h before palmitate (0.4 mM) addition. Cell death was determined by measuring lactate dehydrogenase (LDH) release 16 h later. B: HepG2 cells were pretreated with 2-Br (100 μM) for 2 h before palmitate (0.4 mM) addition. Cell death was determined by measuring LDH release after a 16-h exposure. C: AML12 cells were pretreated with etomoxir (10 μM) for 2 h before palmitate (0.4 mM) addition. Cell death was determined by measuring LDH release after a 16-h exposure. D: HepG2 cells were pretreated with etomoxir (10 μM) for 2 h before palmitate (0.4 mM) addition. Cell death was determined by measuring LDH release after a 16-h exposure. Data are expressed as means ± SD; n = 4 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (**P < 0.01, ***P < 0.001; ns, not significant). E and F: membrane palmitoylated protein-1 (MPP) (E) as well as phosphorylated (p-) and total S6 protein abundance (F) were determined by Western blotting. AML12 cells with or without a 2-h pretreatment with 2-Br (10 μM) were exposed to palmitate (PA, 0.4 mM) for 16 h. The signal of each protein band was measured by densitometry and then divided by the signal of its corresponding actin abundance in the same sample. Data are expressed as means ± SD; n = 3 separate experiments. Differences between groups were determined by 1-way ANOVA analysis (*P < 0.05, **P < 0.01). UT, untreated.

DISCUSSION

In this study, we have demonstrated that mTORC1 activation plays a mechanistic role in regulating palmitate-induced lipotoxic cell death in hepatocytes. The protective effect of mTORC1 inhibition against lipotoxicity is autophagy independent in that autophagy suppression via either genetic or pharmacological approach fails to prevent cell death from palmitate exposure. Instead, mTORC1 inhibition attenuates palmitate-induced activation of IRE1α, one of three canonical pathways activated during ER stress/UPR. Furthermore, inhibition of either endoribonuclease or kinase activity (JNK activation) of IRE1α protects hepatocytes against palmitate-induced cell death. Together, we have provided evidence that via the activation of IRE1α pathway, mTORC1 and ER stress pathway integrate and are coordinately implicated in lipotoxic cell death of hepatocytes in response to palmitate excess (Fig. 7).

Fig. 7.

Schematic illustration of mechanistic target of rapamycin complex 1 (mTORC1) implication in palmitate-induced hepatocyte lipotoxicity. Protein palmitoylation seems to contribute to palmitate-evoked mTORC1 activation. mTORC1 evocation is required for palmitate-induced activation of inositol-requiring enzyme 1α (IRE1α), one of the 3 canonical branches of unfolded protein response (UPR). Both endonuclease and kinase [c-Jun NH₂-terminal kinase (JNK)] activity of IRE1α are involved in palmitate-induced hepatocyte cell death. XBP1, X-box binding protein 1. ? denotes unknown mechanism.

Although accumulated evidence suggests that ectopic lipid deposition and subsequent lipotoxic hepatocyte cell death are attributable to peripheral insulin resistance in NAFLD pathogenesis (13, 18, 30), the underlying cellular/molecular mechanisms of lipotoxicity in hepatocytes remain to be clarified. Using palmitate exposure of mouse AML12 and human HepG2 hepatocytes as in vitro lipotoxicity models, via a small-scale screening of available chemical inhibitors we identified that the mTOR inhibitors torin-1 and rapamycin ameliorated palmitate-induced cell death in hepatocytes. Torin-1 is an ATP-competitive inhibitor of mTOR, inhibiting both mTORC1 and mTORC2, whereas rapamycin is a selective allosteric inhibitor of mTORC1 only (37). The distinct specificities of these two inhibitors suggest that mTORC1 inhibition is the predominant contributor in the observed protective effect, although we cannot completely rule out the role of mTORC2 in this process.

mTORC1 pathway integrates a wide range of extracellular and intracellular signals to regulate cell growth, anabolism, and protein synthesis. Its activation depends on growth factor stimulation and availability of nutrients. The dependence of mTORC1 activation on amino acid availability has been extensively studied and very well documented (16). Similarly, the activation of mTORC1 in response to glucose availability has also been widely reported, especially in the field of cancer research (11). Despite not being as intensively investigated as the other two macronutrients, evidence is emerging that fatty acids, particularly SFA palmitate, can activate mTORC1 (46). In line with these previous reports, in this study we showed that palmitate activates mTORC1 in hepatocytes, whereas oleate, an 18-C monounsaturated fatty acid, displayed an incapability of doing so when the same quantity was employed. The regular FBS contains ∼0.47 mM total FFAs (41). Given the relatively low concentration of total FFAs in our cell culture medium (∼0.047 mM in 10% FBS-containing DMEM medium), the palmitate in the FBS cannot exert a significant interference on the interpretation of the results. The physiological concentration of FFAs in human plasma is ∼0.5 mM, ranging from ∼0.25 mM to as high as being more than 1 mM (1, 15, 38). In mice, the plasma concentration of FFAs is ∼0.4 mM (12, 14), which is much more consistent across different studies than those for human subjects. Therefore, the FFA concentrations in our cell culture system, ∼0.3 mM and 0.5 mM, respectively, can be considered as being physiological and pathophysiological.

The translocation of mTOR to the surface of lysosomes is an obligatory step for mTORC1 activation. Although the exact mechanisms underlying the lysosomal translocation and membrane localization of mTOR remain to be fully understood, protein palmitoylation appears to play a critical role during this process (35, 36). Protein S-palmitoylation is the posttranslational addition of palmitate to protein cysteine residues, facilitating membrane localization of specific proteins (23). Recent studies have revealed that many proteins in the mTORC1 activation pathway are palmitoylated, including LAMTOR1, a key component of the Regulator complex, required for mTORC1 translocation to lysosomes (35, 36). Using 2-bromopalmitate, which has been reported to inhibit protein palmitoylation and to attenuate palmitate lipotoxicity in different cell types (5, 17), the present study appears to support that protein palmitoylation in response to excess palmitate exposure plays a mechanistic role in its inductive effect on mTORC1 activation and resultant lipotoxicity in hepatocytes. It is noteworthy that 2-bromopalmitate is not a selective, let alone specific, inhibitor of protein palmitoylation. Although we excluded the involvement of its inhibitory effect on fatty acid β-oxidation in this process, we must acknowledge that this chemical may interfere with many unknown processes involved in lipid metabolism. Future studies are warranted to confirm the role of protein palmitoylation in palmitate-triggered mTOR activation.

One of the best-described functions of mTORC1 belongs to its regulatory role in autophagy, an evolutionarily conserved process for the bulk degradation of aged/dysfunctional cellular components (organelles and proteins) during starvation, leading to nutrient redistribution for maintenance of cellular energy homeostasis (24, 45). It has been well documented that autophagy induction is protective against palmitate-induced cell death in various cell types, including hepatocytes (25, 39, 47). mTORC1 is the major physiological inhibitor of autophagy and mTORC1 inhibitors, including torin-1 and rapamycin, stimulate autophagy in a variety of cells and tissues, including hepatocytes and the liver (24, 45). Therefore, it is rational to postulate that the protective effect of mTORC1 inhibition against lipotoxicity might derive from autophagy induction. As expected, torin-1 treatment induces autophagy in hepatocytes. To confirm the potential role of autophagy in this process, autophagy inhibition was achieved by pretreating cells with a chemical inhibitor of ULK1 or siRNA knockdown of autophagy protein ATG5, which is essential for autophagy induction. Our data have clearly shown that the protection of mTOR inhibition on lipotoxicity persists even under the circumstance of autophagy inhibition, suggesting that the protective effect of mTORC1 inhibition on hepatocytes lipotoxicity occurs by an autophagy-independent mechanism. It is important to note that, unexpectedly, ULK1 inhibitor indeed mitigates palmitate lipotoxicity slightly but significantly. Currently, we have no plausible explanation for this observation, and further investigations are warranted to delineate the mechanism behind this scenario.

We previously reported that the inhibition of IRE1α endonuclease activity protects HepG2 cells against palmitate-induced cell death (40). IRE1α possesses both endonuclease activity and kinase activity (JNK) (42, 43). In this study, we observed that inhibition of either its endonuclease activity or JNK activation endows protection in AML12 cells, suggesting that IRE1α activation plays an important role in regulating palmitate-induced lipotoxic effect. Importantly, we have shown that mTORC1 inhibition mitigates IRE1α activation. Although our data provided evidence that IRE1α is a downstream target of mTORC1 activation during palmitate-induced hepatoxicity, further investigations are required to elucidate the mechanistic link between mTORC1 and IRE1α activation. Whereas the critical role of JNK activation in palmitate-induced lipotoxicity has been well documented (8, 22), the mechanism(s) regarding how IRE1α endonuclease activity contributes to palmitate-triggered hepatocyte cell death remains unclear. In addition to inducing nonconventional splicing of the transcription factor Xbp-1, IRE1α activation also promotes the degradation of mRNAs encoding mostly ER-targeted proteins via a process named RIDD (regulated IRE1α-dependent decay) (19). It will be interesting to confirm whether Xbp-1 splicing and/or RIDD contributes to hepatic lipotoxicity and the mechanisms that underlie this in future investigations.

Hepatic lipotoxicity plays a central role in the pathogenesis of NAFLD. The critical involvement of mTORC1 activation in palmitate-induced lipotoxicity in hepatocytes suggests the hepatoprotective potential via blocking mTORC1 in NAFLD. This is particularly interesting in that rapamycin, the specific inhibitor of mTORC1, is clinically utilized for the treatment of certain cancers. However, as mTORC1 activity is essential for cell growth, proliferation, and, metabolism, among others, inhibiting mTORC1 by rapamycin has been reported to have an adverse effect on the regenerative or repair process in organs and tissues, including the liver. Thus, further investigations will be necessary to clarify the underpinning mechanisms behind palmitate-induced mTORC1 activation as well as the molecular link between mTORC1 and IRE1α activation in hepatocytes and to explore its clinical relevance.

GRANTS

This work was funded in part by NIH National Institute on Alcohol Abuse and Alcoholism Grants R21AA025363 (to Z.S.) and R01AA026603 (to Z.S.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.C., A.G., and Z.S. conceived and designed research; Y.C., A.G., J.W., T.Z., and Q.S. performed experiments; Y.C., A.G., J.W., T.Z., and Q.S. analyzed data; Y.C., A.G., and Z.S. interpreted results of experiments; Y.C., A.G., J.W., and Q.S. prepared figures; Y.C. and A.G. drafted manuscript; A.G. and Z.S. edited and revised manuscript; Y.C., A.G., J.W., T.Z., Q.S., and Z.S. approved final version of manuscript.