Antioxidative function of L-FABP in L-FABP stably transfected Chang liver cells

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Liver fatty acid binding protein (L-FABP) contains amino acids that are known to possess antioxidant function. In this study, we tested the hypothesis that L-FABP may serve as an effective endogenous cytoprotectant against oxidative stress. Chang liver cells were selected as the experimental model because of their undetectable L-FABP mRNA level. Full-length L-FABP cDNA was subcloned into the mammalian expression vector pcDNA3.1 (pcDNA-FABP). Chang cells were stably transfected with pc-DNA-FABP or vector (pcDNA3.1) alone. Oxidative stress was induced by incubating cells with 400 μmol/L H2O2 or by subjecting cells to hypoxia/reoxygenation. Total cellular reactive oxygen species (ROS) was determined using the fluorescent probe DCF. Cellular damage induced by hypoxia/reoxygenation was assayed by lactate dehydrogenase (LDH) release. Expression of L-FABP was documented by regular reverse transcription polymerase chain reaction (RT-PCR), real-time RT-PCR, and Western blot. The pcDNA-FABP–transfected cells expressed full-length L-FABP mRNA, which was absent from vector-transfected control cells. Western blot showed expression of 14-kd L-FABP protein in pcDNA-FABP–transfected cells, but not in vector-transfected cells. Transfected cells showed decreased DCF fluorescence intensity under oxidative stress (H2O2 and hypoxia/reoxygenation) conditions versus control in inverse proportion to the level of L-FABP expression. Lower LDH release was observed in the higher L-FABP–expressed cells in hypoxia/reoxygenation experiments. In conclusion, we successfully transfected and cloned a Chang liver cell line that expressed the L-FABP gene. The L-FABP–expressing cell line had a reduced intracellular ROS level versus control. This finding implies that L-FABP has a significant role in oxidative stress. (HEPATOLOGY 2005;42:871–879.)

Cellular oxidative stress is one of the factors responsible for the propagation of liver diseases, such as hepatitis, cirrhosis, and hepatoma.1 Several primary antioxidant defense systems such as superoxide dismutase (SOD), catalase, glutathione (GSH), and glutathione peroxidase are present intracellularly. These systems scavenge reactive oxygen species (ROS) such as hydrogen peroxide, superoxide, lipid peroxides, and free radicals. Exposure to oxidative stress may, however, deplete the cellular antioxidant capacity. Therefore, other antioxidant defense systems are expected to play an important role in oxidative stress.

Liver fatty acid binding protein (L-FABP) is a 14-kd protein found abundantly in the cytoplasm and the nucleus of hepatocytes.2, 3 L-FABP is very likely to be an effective endogenous antioxidant, because it has high affinity and capacity to bind long-chain fatty acid oxidation products.4, 5 Hepatocyte L-FABP concentration could be as high as 0.4 mmol/L,6 and it contains a large number of reducing amino acid residues (1 cysteine and 7 methionine residues) in its molecular structure. With an accessible volume enclosed by the molecular surface of L-FABP of 28,600 Å3, 7 the concentration of total methionine residues in L-FABP could be as high as approximately 400 mmol/L. Methionine and cysteine amino acids are regarded as cellular scavengers of activated xenobiotics and involved in antioxidation.8 However, the direct role of L-FABP in oxidative stress has not been investigated. In this report, we used an L-FABP stably transfected Chang liver cell line to test the hypothesis that hepatocyte L-FABP is an important cellular antioxidant during oxidative stress induced by hydrogen peroxide and hypoxia/reoxygenation.

Abbreviations

SOD, superoxide dismutase; GSH, glutathione; ROS, reactive oxygen species; L-FABP, liver fatty acid binding protein; DMEM, Dulbecco's modified Eagle medium; TNF-α, tumor necrosis factor alpha; RT-PCR, reverse transcription polymerase chain reaction; PBS, phosphate-buffered saline; DCF, dichlorofluorescin; H2DCDFA, 2,7-dichloro-fluorescin diacetate; FBS, fetal bovine serum; LDH, lactate degyfrogenase; DMSO, dimethylsulfoxide.

Materials and Methods

Materials

Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum, pyruvate, penicillin, streptomycin, geneticin (G-418), amphotericin B, and Lipofectamine (Invitrogen, Carlsbad, CA) were purchased from GIBCO/BRL. Recombinant human tumor necrosis factor alpha (TNF-α) was purchased from ProSpec-Tany TechnoGene (Rehovot, Israel). Human hepatoma cell lines—Chang liver cell, HepG2, and PLC/PRF/5—were purchased from American Type Culture Collection (ATCC, Rockville, MD), Huh7 were obtained from Dr. G. Minuk (University of Manitoba, Canada). The pcDNA3.1/v5-his vector was purchased from Invitrogen. Marathon-ready human liver cDNA and Marathon polymerase chain reaction (PCR) amplification kits were purchased from Clontech Laboratories (Mountain View, CA). L-FABP polyclonal antibody was generated in our laboratory as previously reported.9

Cell Culture Conditions

Cells were grown in DMEM supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, 10% fetal bovine serum, 1% fungizone, and 0.011% sodium pyruvate in a humidified, 37°C incubator in an atmosphere of 95% air and 5% CO2. Neomycin-resistant transfected Chang cells were maintained in the presence of G418 at concentration of 200 mg/L.

Reverse Transcription PCR Analysis of FABP mRNA Expression

Gene-specific PCR primers for human L-FABP were designed with Oligo 5.1 program from the cDNA sequence obtained from GenBank (GenBank M10050). Primers were synthesized by Invitrogen (Burlington, ON, Canada). Total RNA was isolated from hepatoma cells by TriZOL reagent (Life Technologies). The first stranded DNA was performed with CLONTECH First-cDNA reverse transcription (RT)-PCR kit. The reaction products (5 μL) were used for PCR. The oligonucleotide primers were 5′-TGT CGG AAA TCG TGC AG-3′ (sense), encompassing nucleotides 155 to 172, and 5′-GAT TAT GTC GCC GTT GAG TT-3′ (anti-sense), encompassing nucleotides 350 to 370. The product length was 215 bp. The PCR amplification was carried out in 30 cycles of denaturation (94°C, 45 seconds), annealing (56°C, 30 seconds), elongation (72°C, 120 seconds), and with an additional 7-minute final extension at 72°C. The PCR product was analyzed by running on 1.5% agarose gel.

Construction of Mammalian Expression Plasmid of Human L-FABP

Human L-FABP cDNA was generated by PCR using CLONTECH's Marathon-Ready cDNA kit. The primers were 5′-CTA TTG CCA CCA TGA GTT-3′ (sense), encompassing nucleotides 5 to 23, and 5′-AAT AAT ATG AAA TGC AGA CTT G-3′ (antisense), encompassing nucleotides 401 to 423. The PCR product (418 bp) containing the full-length L-FABP cDNA was subcloned into PCR-Blunt II-TOPO plasmid vector. The plasmid was then cut with HindIII and XbaI to generate a digested product of 530-base pair cDNA containing full-length L-FABP coding region, which was then subcloned into pcDNA3.1/V5-His B mammalian expression vector (Invitrogen). The produced plasmid was referred as pcDNA-FABP.

Selection of Stably Transfected Chang Liver Cells

Chang liver cells were stably transfected with pcDNA-FABP plasmid. In brief, 1 × 106 Chang liver cells in 3.5-cm culture dishes were transfected with 1 μg linearized pcDNA-FABP or pcDNA3.1/V5-His using Lipofectamine according to manufacturer's instructions. Stably transfected cells were established in the presence of G418 (800 μg/mL) after 4 weeks of incubation. G418-resistant clones were then isolated with cloning cylinders and maintained in culture medium containing G418 (200 μg/mL). Clones were analyzed individually by RT-PCR for the expression of L-FABP.

L-FABP mRNA Expression Assay With Quantitative Real-Time RT-PCR

Quantitative real-time PCR was performed in a LightCycler instrument (Roche Diagnostics, Mannheim, Germany) using the RNA Master SYBR Green I kit (Roche Diagnostics). Results were analyzed using the LDCA software supplied with the instrument. L-FABP mRNA standard curve was constructed by in vitro transcription assay with Ribprobe in vitro Transcription Kit (Promega, Madison, WI) and the pc DNA-FABP plasmid. A 215-bp fragment of L-FABP mRNA was amplified using primers as mentioned earlier. Each 20-μL PCR reactant contained 100 ng total RNA, 3.5 mmol/L MgCl2, 0.5 μmol/L of each primer, and 2 μL of the master mix supplied with the kit. The following real-time PCR protocol was employed: reverse transcription at 61°C for 20 minutes; PCR reaction started at 95°C for 30 seconds; 45 cycles in 4 steps: 95°C for 1 second, 56°C for 5 seconds, 72°C for 13 seconds, and 72°C for 25 seconds.

Western Blot

Protein extracts were prepared from cell cultures by RIPA Buffer (50 mmol/L Tris-HCl pH 7.4, 150 mmol/L NaCl, 1 mmol/L PMSF (phenylmethanesulfonylfluoride), 1 mmol/L EDTA, 5 μg/mL aprotinin, 5 μg/mL leupeptin, 1% Triton x-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate) lysis. Briefly, cell pellets (∼108 cells) were washed once with ice-cold phosphate-buffered saline (PBS) and incubated with 1 mL RIPA buffer on ice for 20 minutes and vortexed 2 to 3 times. The lysate was centrifuged for 5 minutes at 4°C at 20,000g in a microfuge tube. Supernatant was transferred into a clean tube. Protein concentration was measured using the Lowry protein assay. Samples were stored at −80°C until used. The protein sample (20 μg) was subjected to SDS-PAGE, with the resulting band analyzed by enhanced chemiluminescence (ECL system, Amersham, Sydney, Australia). The optical density values of each target protein band were determined by using NIH Imaging software.

Cellular Glutathione Content and Superoxide Dismutase Activity Assay

Cells were harvested by scraping with a rubber policeman and were disrupted by sonication at 4°C. The homogenate was centrifuged for 10 minutes at 3000g for glutathione (GSH) assay or 1500g for superoxide dismutase (SOD) assay samples. The supernatant was collected and stored on ice. Protein concentrations were measured using the Bradford assay.

GSH content was determined by using a colorimetric GSH assay kit (GSH-400, OxisReseach). Supernatant (100 μL) was added to 800 μL of reaction buffer and then 50 μL of solution R1 and 50 μL of solution R2 were added sequentially as described in the assay kit. The reaction mixture was incubated at room temperature for 10 minutes in the dark. Absorbance was measured at 400 nm. Cellular GSH content was expressed as μg/mg cellular protein.

SOD activity was determined using an SOD assay kit (Cayman Chemical, Ann Arbor, MI). The assay was performed on a manufacturer-provided 96-well plate. Sample (10 μL) was added to 200 μL of the diluted radical detector. The reaction was initiated by adding 20 μL diluted xanthine oxidase to each well using a multi-channel pipette. The plate was carefully shaken and incubated at room temperature for 20 minutes. Absorbance was recorded at 450 nm using a plate reader. SOD activity was expressed as U/mg cellular protein.

Inducing Oxidative Stress and Dichlorofluorescin Fluorescence Assay

ROS produced in oxidative stress were assessed by the fluorescence intensity of DCF using a fluoro-microplate reader. Briefly, a 10 mmol/L (4.873 mg/mL) stock solution of 2,7-dichloro-fluorescin diacetate (H2DCFDA) was prepared daily in ethanol, stored at −20°C, and diluted to 100 μmol/L with DMEM medium before each study. This cell-permeable compound is converted into a nonfluorescent product (H2DCF), after intracellular deacetylation by intracellular esterases, and is rapidly oxidized to a highly fluorescent compound dichlorofluorescin (DCF). The DCF and H2DCF are retained more effectively in the cytosol than H2DCFDA.

Oxidative Stress Induced by H2O2.

Chang-transfected cells with different L-FABP expression levels were cultured in a 96-well culture plate (Costar no. 3603, Corning Inc.) at a cell density of 25,000 cells per well. After 8 hours of incubation in DMEM-10% fetal bovine serum (FBS), cultures were washed twice with PBS and then incubated with 100 μmol/L H2DCFDA in 95% air/5% CO2 for 30 minutes at 37°C. Extracellular H2DCFDA was removed by washing the culture twice with warm PBS. Cellular oxidative stress was induced by incubating cells with 400 μmol/L H2O2 in PBS containing Ca++ and Mg++ for 20 minutes at 37°C in the dark. Negative controls were performed using the same conditions but without H2O2. Cellular fluorescence intensity of each well was measured and immediately recorded.

Oxidative Stress Induced by Hypoxia/Reoxygenation.

The hypoxia-reoxygenation experiments were done according to the protocol described by Hasinoff.10 The hypoxia apparatus was a cylindrical plastic container that was slightly larger than a 35-mm culture dish. The container had a small hole at the center of the top and the bottom of the dish chamber through which gas was passed. The flow of gas was directed onto the top surface of liquid in the culture dish. The gas was humidified by bubbling it through water in a gas-washing tower to prevent evaporative loss of water from the culture dish. The whole apparatus was housed in a 37°C incubator. pcDNA-FABP–transfected Chang liver cells with different L-FABP expression levels were parallel seeded in 35-mm culture dish at the cell density of one million cells per dish. After 8 hours of incubation in DMEM–10% FBS, the culture medium was replaced 3 times with 1 mL DMEM–0% FBS to minimize the background lactate dehydrogenase (LDH) levels. Culture dishes were placed the hypoxia apparatus and made hypoxic by passing 95% N2/5% CO2 over the surface of the liquid for 3 hours at a flow rate of 80 mL/min. Cells were subsequently reoxygenated with medical-grade 95% O2/5% CO2 for 3 hours at the same flow rate. To maximize gas exchange during hypoxia, the hypoxia apparatus was gently shaken at 40 rpm/min. In negative control studies, samples were kept in an incubator with 95% air and 5% CO2 for 6 hours. At the end of hypoxia-reoxygenation, a 100-μL sample of the supernatant was collected and centrifuged for 5 minutes at 500g to remove any floating cells. Samples were kept at −80°C until LDH assay requirement. After the supernatant was sampled, plates were loaded with 1 mL 100 μmol/L H2DCFDA for 30 minutes in the incubator. Plates were then washed twice with PBS, and the cultures were lysed with 0.5 mL dimethylsulfoxide (DMSO) for 15 minutes at room temperature in dark. Two hundred microliters lysate was collected for immediately assay of DCF fluorescence.

DCF Fluorescence Measurement

H2DCFDA-loaded cells were placed in a FLUOstar Galaxy plate reader (BMG LABTECH, Offenburg, Germany). The excitation filter was set at 485 nm and the emission was 520 nm. Temperature was maintained at 25°C. Fluorescence in each well was captured, digitized, and stored on a computer using FLUOstar Galaxy V4.11. DMSO was tested to ensure that DMSO had no quenching effect on DCF fluorescence. Fluorescence signals were stable for more than 1 hour in DMSO.

LDH Release Assay

LDH activity was measured spectrophotometrically as previously described.11 The substrate for LDH reaction was made by mixing 10% (vol/vol) of 2.5 mmol/L NADH and 10% (vol/vol) 25 mmol/L pyruvate in a Tris-KCl buffer (50:150 mmol/L, pH 7.40) on the day of the assay and equilibrated to 25°C in a water bath before use. Quadruplicate 20-μL aliquots of the culture supernatants collected from hypoxia-reoxygenation sample (containing LDH) were added to a quartz cuvette at 25°C that contained 800 μL of the substrate solution in a Cary Win UV spectrometer. Change in absorbance at 340 nm was directly proportional to LDH activity in the supernatant samples.

Inducing Apoptosis and Caspase Activity Assay

Apoptosis Induced by H2O2 and TNF-α.

Chang-transfected cells with different L-FABP expression levels were cultured in 6-well culture plates at a cell density of approximately 1 million cells per well. After 24 hours of incubation in DMEM-10% FBS, cultures were washed with PBS and then incubated with 1 mL of 400 μmol/L H2O2 in PBS containing Ca++ and Mg++ for 20 minutes at 37°C in the dark. Cultures were washed with PBS to remove H2O2, and then incubated with culture medium for 3 hours in a 37°C incubator. The other group of cell cultures after 24 hours of incubation was treated with recombinant human TNF-α (40 ng/mL) in DMEM–0% FBS medium for 12 hours for inducing apoptosis.

Apoptosis Induced by Hypoxia–Reoxygenation.

Cellular apoptosis was induced by 3 hours of hypoxia and 3 hours of reoxygenation treatment as described.

Caspase-1 Activity Assay.

Negative controls were performed concurrently. Apoptosis was evaluated by caspase-1 activity. Caspase-1 activity was assayed using a Caspase-1/ICE colorimetric assay kit (BioVison, Exton, PA). After these treatments, cell cultures were washed once with PBS and lysed with chilled cell lysis buffer for 10 minutes on ice. The lysate was centrifuged at 10,000g for 1 minute. The supernatant (cytosolic extract) was saved and stored at −80°C until required. Protein concentrations were determined using the Bradford protein assay. Cell lysate (50 μL) containing 100 μg cellular protein was used for caspase-1 activity assay according to the manufacture protocol. Absorption at 405 nm was directly proportional to the activity of caspase-1.

Statistical Analysis

Results are expressed as mean ± SEM. Appropriate statistical analysis included Student t test (unpaired) where 2 groups are to be compared. Two-way ANOVA was used for multiple comparisons. Statistical significance was considered at P < .05. The n value refers to the number of experimental assays in each study.

Results

Transfection of Chang Liver Cells With L-FABP cDNA

To study the role of L-FABP in cellular oxidative stress, the best in vitro model is a hepatoma cell line that has no or low L-FABP expression. The human hepatoma cell lines of Chang, Huh7, HepG2, and PLC/PRF/5 were examined for the expression of L-FABP. As shown in Fig. 1, L-FABP mRNA was detected in all cell lines with the exception of Chang cells. After stable transfection with pcDNA-FABP, more than 30 cell clones were selected for L-FABP expression identification by RT-PCR; 77% of the clones expressed the full-length transcript mRNA of L-FABP. Colonies with different expression levels of L-FABP were selected for further study. L-FABP mRNA levels in the selected clones were demonstrated using regular RT-PCR and real-time RT-PCR (Fig. 2A). Western blot studies showed expression of 14 kd L-FABP protein in pcDNA-FABP–transfected Chang liver cells with different expression levels, whereas control cells transfected with vector (pcDNA3.1) did not express L-FABP (Fig. 2B). All L-FABP–expressing clones selected for subsequent study were morphologically similar to vector-transfected control cell.

Figure 1.

Expression of L-FABP cDNA in hepatoma cell lines by RT-PCR. Total RNA was isolated from different cell cultures. RNA (1 μg) was subject to RT-PCR. The expected PCR product size was 215 bp. Lane 1: Ladder; Lanes 2 and 6: Chang; Lane 3: Hep G2; Lane 4: Huh 7; Lane 5: PLC; Lane 7: Negative controls; Lane 8: Blank. Chang cells did not show any L-FABP expression. L-FABP, liver fatty acid binding protein; RT-PCR, reverse transcription polymerase chain reaction.

Figure 2.

(A) L-FABP mRNA expression levels in the selected clones. (i) RT-PCR images of L-FABP and loading control GAPDH in the selected clones. (ii) L-FABP mRNA concentration in total cellular RNA in the selected clones. (B) Western blot analysis of L-FABP expression in the experimental pcDNA-FABP–transfected Chang cells. Panel a is the expression of the 14-kd L-FABP protein. Panel b is the expression of the 36-kd GAPDH loading control. The relative L-FABP expression was calculated by normalizing L-FABP optical density against GAPDH (ii). Data represent mean ± SEM, n = 4. V represents vector-transfected cell; H, high L-FABP expression cell; L1, low L-FABP expression cell; L2, lower L-FABP expression cell. Vector (pcDNA3.1)-transfected Chang cells did not show any L-FABP expression. **P < .01, n = 4. L-FABP, liver fatty acid binding protein; RT-PCR, reverse transcription polymerase chain reaction.

GSH Contents and SOD Activities of L-FABP–Transfected and Vector-Transfected Chang Liver Cells

GSH contents and SOD activities of Chang-transfected cells were determined and are shown in Table 1. The L-FABP transfection overexpressed in Chang cells did not alter cellular GSH level or SOD activity. Thus, we assumed that the antioxidant effect observed in this study was mainly directed to cellular L-FABP expression.

Table 1. GSH Contents and SOD Activities in the Experimental Cell Lines
Cell LinesVectorL2L1H
  1. NOTE. Results are mean ± SEM of 6 determinations. P > .05. No significant difference was seen among these cell lines in GSH content and SOD activity. H represents high L-FABP expression cell; L1, L2, low and lower L-FABP expression cells, respectively. Abbreviations: GSH, glutathione; SOD, superoxide dismutase.

GSH (μg/mg)36.07 ± 1.7737.12 ± 1.6038.35 ± 1.6536.83 ± 2.29
SOD (U/mg)13.44 ± 1.2112.18 ± 1.3410.72 ± 0.6612.09 ± 1.16

Antioxidant Effect of L-FABP on Oxidative Stress Induced by H2O2

Hydrogen peroxide (H2O2) is a commonly used cell membrane–permeable precursor of various intracellular free radicals. It was used in our study to generate an in vitro model of cellular oxidative stress. Cellular oxidative stress was monitored with the oxidation-sensitive probe 2,7-dichloro-fluorescein diacetate (H2DCFDA). Various ROS and hydroperoxides oxidize H2DCFDA, yielding the fluorescent product DCF. Native or modified albumin reportedly suppresses the fluorescence of oxidized DCF,12 but this was not the case for L-FABP–expressing cells in our experiments (data not shown).

The pcDNA-FABP–transfected Chang cells with different expression levels were exogenously challenged with 400 μmol/L H2O2 for 10, 20, and 30 minutes at 37°C. Total intracellular ROS in transfected cells was measured using DCF fluorescence. We observed that DCF fluorescence intensity in L-FABP–transfected cells was significantly reduced (P < .01) with an increase in L-FABP expression (Fig. 3). All L-FABP–expressed cells had lower DCF fluorescence intensity than the non–L-FABP expressed vector-transfected control cell. The inverse relationship between L-FABP and DCF fluorescence intensity was interpreted to suggest that intracellular L-FABP was able to function as an intracellular antioxidant and had the ability to play a major role in the oxidative stress induced by hydrogen peroxide.

Figure 3.

DCF fluorescence of pcDNA-FABP–transfected Chang cells in oxidative stress induced by hydrogen peroxide. Cells were cultured in 96-well plates. H2DCFDA (100 μmol/L) was loaded onto cells for 30 minutes. Cells were subsequently exposed to 400 μmol/L H2O2 for 10, 20, and 30 minutes. Cellular fluorescence in each well was measured and immediately recorded. Data are represented as mean ± SEM, n = 8, **P < .01, *P < .05, one tail distribution. V represents vector-transfected cell; H, high L-FABP expression cell; L1, low L-FABP expression cell; L2, lower L-FABP expression cell. Hollow bars represent negative controls of L-FABP expressed cells (L1) and no FABP expressed cells (V). No significant differences were found in negative control group. The DCF fluorescence intensity was proportional to the level of cellular reactive oxygen species (ROS). The figure shows reduced DCF intensity with increased L-FABP. L-FABP, liver fatty acid binding protein; H2DCFDA, 2,7-dichloro-fluorescin diacetate.

Antioxidant Effects of L-FABP on Oxidative Stress Induced by Hypoxia/Reoxygenation

Hepatic ischemia–reperfusion (or hypoxia/reoxygenation) is related to the generation of ROS, and results in liver injury through oxidative stress.13 The pcDNA-FABP–transfected cells and vector control cell were exposed to hypoxia for 3 hours then reoxygenated for an additional 3 hours. The DCF fluorescence intensity of L-FABP–expressed cells was lower than non–L-FABP–expressed vector control cells. DCF fluorescence intensity was inversely proportional to the L-FABP expression level; the higher the L-FABP level, the lower was the DCF fluorescence intensity (Fig. 4). These data show that L-FABP expression reduced the level of intracellular ROS generated during cellular oxidative stress induced by hypoxia–reoxygenation in vitro.

Figure 4.

DCF fluorescence in pcDNA-FABP–transfected Chang cells subjected to hypoxia-reoxygenation. Cells were cultured in 30-mm tissue culture dishes and exposed to hypoxia (3 hours in 95% N2/5% CO2)-reoxygenation (3 hours in 95% O2/5% CO2). After sampling 100 μL supernatant, 100 μmol/L H2DCFDA was loaded onto cells for 30 minutes. Cells were lysed with DMSO and harvested for DCF fluorescence measurement. Data are presented as mean ± SEM (n = 4), **P < .01, *P < .05, one tail distribution. Hollow bars represent negative control sample that was performed by incubating cells with 95% O2/5% CO2 for 6 hours. V represents vector-transfected cell; H, high L-FABP expression cell; L1 and L2, low and lower L-FABP expression cells, respectively. In the negative control group, there was no significant difference in cells with different L-FABP expression levels. However, the DCF fluorescence intensity was reduced with increasing cellular L-FABP expression in the experimental group. DCF, dichlorofluorescin; H2DCFDA, 2,7-dichloro-fluorescin diacetate; L-FABP, liver fatty acid binding protein.

The protective effect of L-FABP on hepatic hypoxia/reoxygenation injury was assessed by detecting cellular LDH release. LDH is a cytosolic enzyme present within all mammalian cells. Normal plasma membrane is impermeable to LDH, but damaging cell membrane results in membrane permeability increases and subsequent leakage of LDH into the extracellular fluid. Release of LDH into culture supernatant correlates with reduced cell membrane integrity and cell viability. We used the LDH release assay to evaluate the potential damage of oxidative stress on hepatocytes with different levels of L-FABP expression. As shown in Fig. 5, a significant increase in LDH release was found in all transfected cells after hypoxia/reoxygenation (P < .01) compared with the control cells. As the L-FABP level decreased, more LDH was detected (P < .01) in the culture solution. These observations further reinforced the notion that L-FABP provides cytoprotection against the oxidative stress induced by hypoxia/reoxygenation. This protection is most likely due to scavenging of intracellular ROS by L-FABP.

Figure 5.

LDH release from L-FABP–transfected Chang cells subjected to hypoxia/reoxygenation (3 hours of hypoxia and 3 hours of reoxygenation). Hollow bars represent negative control samples. Solid bars represent hypoxia/reoxygenation samples. Negative control experiment was performed by incubating cells with 95% O2/5% CO2 for 6 hours. LDH activity was measured as absorbance at 340 nm, expressed as the rate of absorbance change per minute. Data are presented as mean ± SEM (n = 4), **P < .01, *P < .05, one tail distribution. V represents vector-transfected cell; H, high L-FABP expression cell; L1 and L2, low and lower L-FABP expression cells, respectively. In negative control samples, there were no significant differences in the cells with different L-FABP expression levels. LDH release was, however, decreased with an increase in cellular L-FABP level. LDH, lactate dehydrogenase; L-FABP, liver fatty acid binding protein.

Susceptibility of Hepatocytes to Apoptosis Induced by ROS and TNF-α in L-FABP–Transfected Chang Cells

To delineate the protective pathway of L-FABP, we evaluated antiapoptotic activity of L-FABP in L-FABP–transfected Chang liver cells. The H2O2 treatment and hypoxia/reoxygenation increased cellular ROS level and initiated apoptosis by increasing caspase-1 activity compared with the negative controls (Fig. 6). Expression of L-FABP in Chang cells reduced cellular apoptotic activity. ROS is known to play a critical role in TNF-α–induced hepatocyte apoptosis. L-FABP inhibited the caspase-1 activity in the TNF-α–treated cell cultures (Fig. 6).

Figure 6.

Antiapoptotic effect of L-FABP in L-FABP–transfected Chang cells. Apoptosis was induced by H2O2, hypoxia/reoxygenation, and TNF-α treatment. Caspase-1 activity was used as the marker of apoptosis. Results are presented as percentage of caspase-1 activity relative to negative controls. Data are shown as mean ± SEM (n = 6), **P < .01, *P < .05, two tail distribution. V represents vector-transfected cell; H, high L-FABP expression cell; L1 and L2, low and lower L-FABP expression cells, respectively. L-FABP, liver fatty acid binding protein; TNF-α, tumor necrosis factor alpha.

Discussion

In the last 3 decades, remarkable advances have been made in understanding the structure and biology of FABPs. Although their complete tertiary structure, gene organization, and binding properties are known, the functional aspects of these proteins are still undergoing investigation. FABP's protective role is evidenced by: (1) controlling the availability of free fatty acids and their metabolites in the cytosol, and which prevents their cellular toxicity14, 15; (2) modulating the interaction of fatty acids with nuclear receptors16; (3) sequestrating or removing cytotoxic drugs17; and (4) trapping or scavenging ROS.4, 18–21 However, no significant protective effects of FABP had been observed for chemical-induced anoxia or transport of intracellular fatty acid using the FABP-transfected kidney cell (MDCK) model.22 Hence, the antioxidant effect of L-FABP is not clear. Furthermore, the extent of the protective function that L-FABP offers against oxidant-induced oxidative stress in cells has not been directly studied.

Oxidative stress is involved in the pathogenesis of most chronic disease. In addition to being a cellular signal, ROS may initiate damaging biochemical reactions.23 In response to high levels of ROS, cells protect themselves against oxidative stress by using a variety of free radical scavengers, including SOD, glutathione, and catalase. In some cases, however (e.g., diseases or drug exposure), the titer of these enzymes may be too low to enable adequate protective function.24, 25 Cells likely have another mechanism to deal with such high ROS levels. In this study, we investigated the effect of L-FABP cDNA transfection of Chang liver cells on oxidative stress induced by hydrogen peroxide and hypoxia/reoxygenation. DCF fluorescence intensities were higher in both experiments, suggesting elevated ROS levels in experimental cells. L-FABP–expressed cells demonstrated lower DCF fluorescence intensity; this decrease of fluorescence intensity was correlated with an increased level of cellular L-FABP expression. Moreover, LDH release, a marker of cell damage, was decreased in cells that expressed L-FABP while undergoing hypoxia/reoxygenation oxidative stress. The extent of the reduction in LDH release was proportional to the cellular L-FABP expression. However, LDH release did not provide a clue for the pathway of cell death. Cellular oxidative stress has a causative role in the development of cell apoptosis. Depletion of GSH can sensitize hepatocytes to TNF-α–induced cell death.26 Overexpression of thioredoxin, a small reducing protein, significantly attenuated lipopolysaccharide/d-galactosamine–induced liver injury via its antiapoptotic effect.27 Thus, we studied the antiapoptotic activity of L-FABP in the apoptosis induced by H2O2, hypoxia/reoxygenation, and TNF-α treatment in our transfected model. Overexpressed L-FABP Chang cells reduced the caspase-1 activity in all experiments. These results suggested a cytoprotective function of L-FABP against oxidative stress, and thereby reducing cellular susceptibility to ROS involved apoptosis. The protective mechanism of L-FABP in oxidative stress is likely a result of the protein either inhibiting or scavenging ROS.

L-FABP is very likely to be an effective endogenous antioxidant, because it has high affinity and capacity to bind long-chain fatty acid oxidative products.4, 5 It also has a very favorable protein structure and redox conditions, such as large portion of the intracellular protein pool (approximately 0.4 mmol/L in hepatocyte cytosol),6 a large number of reducing amino acid residues (methionines and cysteine), and a high expression level of methionine sulfoxide reductase.28 Oxidized methionine (methionine sulfoxide) can be reduced back to its reducing from by methionine sulfoxide reductase.29, 30 The cyclic oxidation-reduction of methionine residues of protein may contribute an important antioxidant function by the conversion of ROS to innocuous products.31, 32 Cysteine residue may be involved in the binding of other hydrophobic ligands or serve as an antioxidant participating in S-thiolation/dethiolation.33, 34

In conclusion, peroxisome proliferators upregulate L-FABP expression through activation of peroxisome proliferator activated receptors.35 Peroxisome proliferators also increase intracellular levels of peroxisomes,36 resulting in an overproduction of hydrogen peroxide.37, 38 Hydrogen peroxide, being a precursor of ROS, may contribute to the hepatocarcinogenic effects of peroxisome proliferators.39 In addition to damaging DNA, high levels of hydrogen peroxide may be expected to oxidize hepatic proteins and lipids. Interestingly, few studies have examined the effect of peroxisome proliferators on hepatic protein oxidation and lipid peroxidation.40–42 However, recent studies indicated that hepatic levels of hydrogen peroxide and lipid products are not increased in animals treated with peroxisome proliferators.43, 44 Conversely, nafenopin-pretreatment increased hepatic resistance to cytotoxicity induced by hydrogen peroxide in rats.45 Furthermore, various peroxisome proliferators have been found to be involved in cytoprotection against hepatotoxicity and oxidative stress.46, 47 The protection mechanism that peroxisome proliferators afford is poorly understood. The increased glutathione availability and catalase activity by clofibrate were not the primary protective pathway in the protection against acetaminophen-induced hepatotoxicity and hepatic lipid peroxidation.48–50 Thus, peroxisome proliferators are likely acting through the L-FABP pathway to scavenge high levels of ROS. Collectively, our data indicate that L-FABP is a strong endogenous antioxidant. L-FABP levels could be targeted through appropriate pharmacological treatment to minimize cellular damage.

Ancillary