Background
Fatty acid binding protein (FABP) was first discovered in 1969 by Levi et al. [
1]. Liver fatty acid binding protein (FABP1), named after the tissue in which it was first identified, is approximately a 14–15 kDa protein mainly present in the cytoplasm of hepatocytes but expressed in many other tissues [
2,
3]. FABP1’s main function is thought to be the intracellular transport of lipophilic substrates such as long chain fatty acids. FABP1 also interacts indirectly with the peroxisome proliferator-activated receptor alpha (PPARα), which is a key regulator of lipid homeostasis in hepatocytes and a target for fatty acids and hypolipidemic drugs, by transporting PPARα agonists to its site. Thus, FABP1 acts as a cytosolic gateway by directing PPAR ligands to the nucleus [
4].
A new antioxidant property of FABP1 has recently been uncovered [
5,
6]. In its primary structure FABP1 contains seven methionine and one-cysteine amino acids. These groups are regarded as cellular scavengers of activated xenobiotics and are involved in trapping of free radicals [
7]. Thus, in this study we used an acetaminophen induced toxicity model in cultured cells to assess the antioxidant function of FABP1.
Acetaminophen (AAP), 4-hydroxyacetanilide, is a widely utilized drug known for its analgesic and antipyretic properties. When used at therapeutic levels it is safe, however, an acute or cumulative overdose can cause severe liver injury with the potential of liver failure [
8]. At therapeutic doses, AAP is primarily detoxified by glucuronidation and sulfation with a small fraction metabolized by cytochrome P-450-dependent mixed function oxidase system to a highly reactive N-acetyl-p-benzo-quinonemine (NAPQ1) metabolite [
9]. The metabolite reacts with glutathione (GSH) spontaneously or is catalyzed by glutathione-S-transferases to form a GSH-adduct which is mainly excreted into bile through Mrp2 without significant toxicity [
10]. After an AAP overdose, however, glucuronidation and sulfation are insufficient to detoxify AAP. A large fraction of the drug becomes available for metabolism by cytochrome P450, leading to a rapid depletion of hepatic GSH levels. Once GSH is exhausted, any remaining NAPQI formed will react with alternative targets, in particular cellular proteins [
11]. In addition to NAPQI, other cellular effects of AAP toxicity further exacerbates cellular oxidative stress which in turn contributes to the cell injury process [
12].
The present study investigates the hepatoprotective effect of FABP1 in acetaminophen-induced toxicity using the Chang cell line. Chang cells were originally thought to be derived from normal liver tissue, but subsequently found to have been established via HeLa cell contamination. The rationale for using this cell line, however, rests in the fact that these cells were shown to be devoid FABP1 [
5], yet have the metabolic enzymes responsible for metabolizing AAP to the NAPQ1 reactive species [
12]. Thus, it is an ideal cell line to study the biological and bioprotective effects of FABP1.
Discussion
Acetaminophen (AAP) is an effective over the counter medication for relief of minor pain and fever. A small fraction of the AAP dose is metabolized by CYP 2E1 to a highly reactive N-acetyl-p-benzoquinone imine (NAPQI) metabolite. This metabolite is inactivated by glutathione or GSH within the liver. However, if the concentration of NAPQI is high, liver damage can occur. In severe cases liver failure and death occur. Despite its widespread use, the mechanism of AAP’s hepatocellular injury is still being investigated. Many studies have reported impaired mitochondrial respiration [
13,
14], depletion of hepatocellular ATP levels [
15,
16], opening of the mitochondrial membrane permeability transition pore [
17], and increased levels of glutathione disulfide (GSSG) or the ratio of GSSG:GSH, suggesting the involvement of an oxidant stress following AAP overdose [
15,
18]. In response to high levels of ROS the nascent hepatocyte antioxidants may not provide sufficient capacity to inactivate them, other antioxidant defence systems are expected to take effect. FABP1, with its high affinity and capacity to bind lipophilic oxidative products [
19,
20], is a likely candidate for further protecting hepatocytes from ROS.
Our previous work showed that Chang cells were devoid of FABP1. Transfecting the Chang cell line with FABP1 cDNA gave us an opportunity to study this protein in different models of oxidative stress. Using a hypoxia/reoxygenation as well as a H
2O
2 induced oxidative stress models [
5], we reported that while FABP1 cDNA transfected cells had the same complement of intracellular antioxidant enzymes as the vector transfected cell line, FABP1 cDNA transfected cells were associated with much less ROS levels, suggesting that FABP1 is somehow involved in inactivating free radicals. Yan et al. [
21] investigated the mechanism for the antioxidant protective function of FABP1. Rat FABP1 is known to have seven methionine groups in positions 1, 19, 22, 74, 85, 91, and 113, as well as one cysteine group in position 69. Methionine and cysteine are known to react with ROS. Using a recombinant form of rat FABP1 that was cultured in
E. coli, isolated and purified, the group showed that indeed the methionine groups of FABP1 were associated with reactive oxygen as assessed by MALDI-TOF. Moreover, FABP1 was shown to react with free radicals in both hydrophilic and lipophilic domains of the cell. Using AAPH as a hydrophilic free radical generator the group determined that methionine 1, 19, 22, 91, and 113 were oxidized. In the lipophilic environment, using AMVN as the lipophilic free radical generator, methionine 1, 19, 22, and 113 were only reactive. Methionine 74 and 85 were unreactive in both systems suggesting that these groups may be buried deep within the binding site of FABP1 while the other reactive groups are surface exposed. Interestingly, methionine 91 did not react with any free radicals in the lipophilic domain. This suggested that FABP1 might orientate itself to the membrane allowing other methionine groups access to react with free radicals at the membrane surface. FABP interaction with membrane surfaces has been suggested to occur with FABP2 (intestinal fatty acid binding protein) but not with FABP1 [
22].
In this study we investigated the role of FABP1 in drug-induced liver damage. The data show a rapid onset of an intracellular oxidative stress as early as 3 hrs in cultured cells using 2,7-dichlorofluorescein diacetate (H
2DCFDA) as a marker of intracellular free radical levels. There was a dose-dependent increase in released ROS induced by AAP with DCF fluorescence intensities being significantly lower in FABP1 cDNA transfected cells compared to the vector transfected cells (Figure
2). AAP induced oxidative stress was studied by other groups such as Bajt et al. [
23] who showed that a greater than 10 fold increase in ROS levels between 3.5 and 12.5 hrs resulted following 5 mM AAP treatment in cultured murine hepatocytes. Pretreatment of hepatocytes with 20 mM N-acetylcysteine was shown to enhance cellular glutathione content and suppressed the AAP-induced decrease in cell viability. Bajt concluded that AAP-induced oxidant stress precedes cell necrosis and the oxidant stress is involved in the propagation of cell injury.
The effect of FABP1 on AAP-induced cell damage was assessed using the WST-1 assay in our study, which depends on mitochondrial respiration [
24,
25]. Results from this assay indicated a substantial functional deterioration of hepatocytes following AAP treatment. Similar to the DCF assay, results showed that the cytotoxicity induced by AAP was dose-dependent and statistical differences were observed between FABP1 cDNA and vector transfected cells following 3 hrs drug treatment (Figure
1). Thus, FABP1 protects cellular mitochondrial function in some way. One possibility is by inactivating cytosolic free radicals. These reactive species are not able to interact with the mitochondrial membrane, thus, preserving cellular function. AAP also caused a progressive release of LDH into the culture media, showing cellular damage. The presence of FABP1 attenuated this increase but only at the latter stages of AAP induced cell injury since LDH release only became statistically significant after 12 hrs of treatment. As well as being present in the cytosol, FABP1 is also present in the mitochondria. It is likely that FABP1 plays an early protective role in AAP induced mitochondrial impairment through scavenging free radicals within the mitochondria itself as well as in the cytosol.
The mechanism of AAP-induced cell death is not completely understood. Of the events leading to apoptotic and/or necrotic cell death, Bax seems to be central for the mitochondrial-dependent mechanisms. Bax is a pro-apoptotic Bcl-2 family member that plays a pivotal role in the formation of the mitochondrial permeability pore. Higher levels of Bax are associated with changes in the outer membrane permeability, which are responsible for changes in the inner mitochondrial membrane that leads to disruption in cell function (e.g., membrane potential, cell swelling, leakage). Deletion of Bax has been shown to be associated with dramatic reduction in necrotic injury during myocardial infraction [
26]. Thus, Bax may modulate necrosis through mechanisms that are distinct from apoptosis [
26,
27]. In our study, decreased Bax levels were seen in the AAP treated FABP1 cDNA transfected group at all time points (Figure
4). In the vector-transfected groups the Bax level started to statistically increase at 3 hrs and peaked at 12 hrs. This was not seen in the FABP1 cDNA transfected cells. In the FABP1 cDNA transfected group the Bax level increased only after 12 hr incubation with AAP. Thus, FABP1 shows a higher protective ability at the early time points (3 and 6 hrs) by preventing the increase in Bax level. At the late time points (12 and 24 hrs), Bax level in the FABP1 cDNA transfected group was lower than in the vector-transfected group but still higher than at the 6 hrs FABP1 cDNA transfected group. FABP1 levels, however, were also declining during the late time points. A rationale explanation for the declining FABP1 levels is not apparent. Studies aimed at assessing the time course of FABP1 mRNA activity would be necessary to explain the decrease in FABP1 levels. Presence of necrosis may be a possibility that leads to the release of FABP1. Nevertheless a strong correlation exists between FABP1 and Bax levels, showing that FABP1 prevents cell death through a reduction in Bax activity. Although our work cannot discern whether the higher (or lower) levels of Bax is associated with Bax translocation to the mitochondria, it is reasonable to speculate that the increased levels would lead to increased translocation. Bajt had earlier reported that following 300 mg/Kg intraperitoneal AAP injection to C57BL/6 mice, Bax translocation from the cytosol to mitochondria occurs as early as 1 hr [
28]. Thus, FABP1 likely protects cells against AAP hepatotoxicity during the early time periods of hepatocyte injury.
LDH release is indicative of cell death. Membranes become leaky through a variety of processes including that of oxidative damage. Since release of LDH in our study reached statistical significance only after 12 hrs of AAP treatment, this showed that although earlier time points were associated increased Bax level, later time points involved necrosis. As stated above apoptosis and necrosis are not completely independent processes [
29]. These two processes share a common mitochondrial permeability transition (MPT) pathway. When the MPT occurs abruptly, ATPase becomes activated which depletes ATP levels leading to membrane rupture and oncotic necrosis. However, the ATP level can remain constant (baseline condition) when the MPT proceeds relatively slowly or the ATPase is inhibited by glycolysis or oligomycin. Under these conditions, necrosis can be blocked and apoptosis occurs. At any time, ATP depletion can supervene to cause secondary necrosis. A new term, necrapoptosis has been introduced to describe a death process that begins with death signals or toxic stress, proceeds by shared pathways, but results in either cell lysis (oncotic necrosis) or programmed cellular resorption (apoptosis) depending on other factors, such as ATP [
29]. In our study FABP1 attenuated the increased LDH release. It is not clear on the exact mechanism (direct or indirect) for the FABP1’s hepatoprotective effect but we speculate that it may in part be likely due to its antioxidant role.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
YG participated in study design, carried out the cell culture work and drafted the manuscript. YC performed some of the fluorescence studies and helped in the drafting of the manuscript. JY helped in the design of the studies. GW, YG, and FJB were instrumental in conceiving the study design, obtaining the necessary research funds to carry out the work. All authors read and approved the final manuscript.