Subcellular expression of CYP2E1 in HepG2 cells impacts response to free oleic and palmitic acid

Abstract

Aims: Cytochrome P450 2E1 (CYP2E1) is a mammalian monooxygenase expressed at high levels in the liver that metabolizes low molecular weight pollutants and drugs, as well as endogenous fatty acids and ketones. Although CYP2E1 has been mainly studied in the endoplasmic reticulum (ER, microsomal fraction), it also localizes in significant amounts to the mitochondria, where it has been far less studied. We investigated the effects of CYP2E1 expression in mitochondria, endoplasmic reticulum, or both organelles in transgenic HepG2 cells exposed to free oleic and palmitic acid, including effects on cytotoxicity, lipid storage, respiration, and gene expression.

Results: We found that HepG2 cells expressing CYP2E1 in both the ER and mitochondria have exacerbated levels of palmitic acid cytotoxicity and inhibited respiration. CYP2E1 expression did not impact lipid accumulation from fatty acid exposures, but mitochondrial CYP2E1 expression promoted lipid droplet depletion during serum starvation. In contrast to HepG2 cells, differentiated HepaRG cells express abundant CYP2E1, but they are not sensitive to palmitic acid cytotoxicity. Oleic acid exposure prompted less cytotoxicity, and CYP2E1 expression in the ER prevented an oleic-acid-induced increase in respiration. HepG2 cells exposed to mixtures of palmitic and oleic acid are protected from palmitic acid cytotoxicity. Additionally, we identified that CYP2E1 was decreased at the gene and protein level in hepatocellular carcinoma. Moreover, patients with tumors that had higher CYP2E1 expression had a better prognosis compared to patients with lower CYP2E1 expression.

Innovation: This study has demonstrated that transgenic CYP2E1 subcellular localization plays an important role in sensitivity to cytotoxicity, lipid storage, and respiration in the hepatoma cell line HepG2 exposed to palmitic and oleic acid. HepaRG cells, in contrast, were insensitive to palmitic acid. This work demonstrates the clear importance of CYP2E1 in dictating lipotoxicity and differential roles for the mitochondrial and ER forms of the enzyme. Additionally, our data supports a potentially unique role for CYP2E1 in cancer cells.

Conclusion: There lies a role for CYP2E1 in altering lipotoxicity, and since CYP2E1 is known to be upregulated in both liver disease and hepatocellular carcinoma, it is important to better define how the role of CYP2E1 changes during disease progression.

Keywords: CYP2E1; Free fatty acids; Mitochondria; Oleic acid; Palmitic acid.

Graphical abstract

Fig. 1

Lentiviral transduction of HepG2 cells with CMV-CYP2E1 results in high expression of mRNA and modest induction of protein levels. To express CYP2E1 in HepG2 cells, lentiviral transduction was carried out at a MOI of 10, and transduced cells were selected with puromycin. Monoclonal lines were isolated and lines with similar mRNA and protein levels were selected for comparison. Panel A, mRNA expression of CYP2E1 determined by qPCR (normalized to B-actin). Data represent three independent experiments. Asterisks represent significance in one-way ANOVA post-testing with Dunnett’s multiple comparison test compared to HepG2: ***, p < 0.001, ****, p < 0.0001. Panel B, immunofluorescent imaging of fixed cells stained for anti-CYP2E1 (Abcam 28146) and counterstained with DAPI. Panel C, relative fluorescence determined using ImageJ and normalized to the DAPI signal to control for number of cells. Data represent two independent experiments. Asterisks represent significance in one-way ANOVA post-testing with Dunnett’s multiple comparison test compared to HepG2: * p < 0.05, **, p < 0.01.

Fig. 2

CYP2E1 expression in HepG2 cells sensitizes to palmitic acid cytotoxicity and mitochondrial inhibition.Panel A, cell death following 24-hour exposure to palmitic acid was determined by propidium iodide fluorescence (normalized to total fluorescence after addition of 1 % Triton X-100). ****, p < 0.0001, adjusted for multiple comparisons after 2-way ANOVA. Panels B-C, lipid droplets were measured following a 24 h (fed, panel B) or 5-day (serum starved, panel C) exposure to 0.25 mM PA. Briefly, following the exposure cells were fixed with 4 % paraformaldehyde and stained with Nile Red. **, p < 0.01, ***, p < 0.001, adjusted for multiple comparisons after 2-way ANOVA. Panels D-G, real-time oxygen consumption rate during exposure to 0.25 mM PA was measured using a Resipher device from Lucid Scientific. ****, p < 0.0001 comparing area under the curve compiled from 3 independent replicates. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Endoplasmic reticulum-localized CYP2E1 alters respiration following oleic acid exposure.Panel A, cell death following 24-hour exposure to oleic acid was determined by propidium iodide fluorescence (normalized to total fluorescence after addition of 1 % Triton X-100). **, p < 0.01, ***, p < 0.001, adjusted for multiple comparisons after 2-way ANOVA. Panels B-C, lipid droplets were measured following a 24 h (fed, panel B) or 5-day (serum starved, panel C) exposure to 0.5 mM OA. Briefly, following the exposure cells were fixed with 4 % paraformaldehyde and stained with Nile Red. **, p < 0.01, ***, p < 0.001, adjusted for multiple comparisons after 2-way ANOVA. Panels D-G, real-time oxygen consumption rate during exposure to 0.5 mM OA was measured using a Resipher device from Lucid Scientific. *, p < 0.05, ****, p < 0.0001 comparing area under the curve compiled from 3 independent replicates. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Genes involved in ER stress, lipid handling, and oxidative stress are differentially expressed in CYP2E1-transduced HepG2 cells exposed to oleic acid. Cells were exposed to 0.5 mM oleic acid for 24 h, then cells were harvested and lysed and RNA was isolated. After cDNA synthesis, gene expression was determined by qPCR. The fold change was determined using the 2^−ΔΔCt method (normalized to β-actin and compared to the untreated cells) for each cell line and compared across cell lines. Panel A, spliced (activated) X-box binding protein 1, involved in ER stress response. Panel B, Fatty Acid Synthase, catalyzes the synthesis of fatty acids. Panel C, Acyl-CoA Oxidase 2, the rate-limiting enzyme in β-oxidation of branched, long-chain fatty acids. Panel D, Superoxide Dismutase 1, an enzyme localized throughout the cell that converts superoxide to hydrogen peroxide. Panel E, Superoxide Dismutase 2, an enzyme localized in the mitochondria that converts superoxide to hydrogen peroxide. Data shown are from 3 independent replicates. *, p < 0.05, **, −<0.01, adjusted for multiple comparisons after one-way ANOVA.

Fig. 5

HepG2 cells exposed to mixtures of palmitic and oleic acid are protected from cytotoxicity, and gene expression patterns are similar to oleic acid alone.Panel A, cell death following 24-hour exposure to mixtures of palmitic and oleic acid was determined by propidium iodide fluorescence (normalized to total fluorescence after addition of 1 % Triton X-100). Panels B-C, lipid droplets were measured following a 24 h (fed, panel B) or 5-day (serum starved, panel C) exposure to 0.25 mM PA and 0.5 mM OA. Briefly, following the exposure cells were fixed with 4 % paraformaldehyde and stained with Nile Red. For gene expression analysis, cells were exposed to 0.25 mM PA and 0.5 mM oleic acid for 24 h, then cells were harvested and lysed and RNA was isolated. After cDNA synthesis, gene expression was determined by qPCR. The fold change was determined using the 2^−ΔΔCt method (normalized to β-actin and compared to the untreated cells) for each cell line and compared across cell lines. Panel D, Superoxide Dismutase 1, an enzyme localized throughout the cell that converts superoxide to hydrogen peroxide. Panel E, Thioredoxin, catalyzes the reduction of oxidized cysteine residues and the cleavage of disulfide bonds. Panel F, Acyl-CoA Oxidase 1, the rate-limiting enzyme in peroxisomal strain-chain fatty acid β-oxidation. Panel G, Acyl-CoA Oxidase 2, the rate-limiting enzyme in β-oxidation of branched, long-chain fatty acids. Data shown are from 3 independent replicates. *, p < 0.05, **, −<0.01, adjusted for multiple comparisons after one-way ANOVA. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

HepaRG co-culture system accumulates abundant CYP2E1 protein during differentiation, but is not sensitive to palmitic acid cytotoxicity.Panels A-B, Immunofluorescent staining of CYP2E1 (magenta) and nuclei (DAPI, blue) in differentiated and undifferentiated HepaRG cells. HepaRG cells were cultured according to manufacturer protocols (Biopredic) and differentiated for 30 days on glass coverslips in 6-well plates. Undifferentiated cells were allowed to attach and recover from splitting for 3 days before fixation. Panels C-D, cells were differentiated and harvested for automated capillary-based Western Blot with the Jess system (Protein Simple). Panel C shows a rendering of bands for CYP2E1 at ∼ 53 kDa only in the differentiated cells, while the quantitation (integration of peak areas) is shown in panel D, normalized to the undifferentiated cells. Panels E-F, cell death following 24-hour exposure to palmitic acid was determined by propidium iodide fluorescence (normalized to total fluorescence after addition of 1 % Triton X-100). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7

CYP2E1 expression is decreased in liver hepatocellular carcinomas (LIHC) tumors and influences overall survival.Panel A, CYP2E1 mRNA expressed in transcript per million (TPM) in LIHC tumors and non-cancer control liver tissue. Panel B. CYP2E1 mRNA in LIHC tumors and paired tumor adjacent normal control liver tissue. Panel C. Survival curve of patients with high or low expressing LIHC tumors over 120 months. *p < 0.05, **p < 0.001, ***p < 0.0001, ***p < 0.00001, adjusted for multiple comparisons after one-way ANOVA.