Potential conflict of interest: Nothing to report.
Acetaminophen (APAP) overdose causes severe, fulminant liver injury. The underlying mechanism of APAP-induced liver injury (AILI), studied by a murine model, displays similar characteristics of injury as those observed in patients. Previous studies suggest that aside from APAP-induced direct damage to hepatocytes, the hepatic innate immune system is activated and may contribute to the overall pathogenesis of AILI. The current study employed the use of two murine natural killer (NK) cells with T-cell receptor (NKT) cell knockout models (CD1d−/− and Jα18−/−) to elucidate the specific role of NKT cells in AILI. Compared to wild-type (WT) mice, NKT cell-deficient mice were more susceptible to AILI, as indicated by higher serum alanine transaminase levels and mortality. Increased levels of cytochrome P450 2E1 (CYP2E1) protein expression and activities, which resulted in increased APAP protein adduct formation, were observed in livers of APAP-treated NKT cell-deficient mice, compared to WT mice. Compared to WT mice, starvation of NKT cell-deficient mice induced a higher increase of ketone bodies, which up-regulate CYP2E1 through protein stabilization. Conclusion: Our data revealed a novel role of NKT cells in regulating responses to starvation-induced metabolic stress. Elevated ketone body production in NKT cell-deficient mice resulted in increased CYP2E1-mediated APAP biotransformation and susceptibility to AILI. (HEPATOLOGY 2013)
Acetaminophen (APAP) is a commonly used antipyretic and analgesic known to be safe and effective at therapeutic doses (1-4 g/day).1 However, severe liver injuries have been observed after an acute or cumulative overdose of APAP (10-15 g/day).1
APAP-induced hepatocyte damage is initiated by formation of the reactive metabolite, N-acetyl-p-benzoquinone imine (NAPQI).2 NAPQI rapidly depletes glutathione (GSH) within the liver and covalently binds to cellular macromolecules. Impairment of macromolecules results in mitochondrial dysfunction, loss of adenosine triphosphate (ATP), and centrilobular necrosis.3 In addition to APAP-induced direct hepatotoxicity, activation of innate immune cells and their production of pro- and anti-inflammatory mediators may further influence the severity of APAP-induced liver injury (AILI).4-8
Natural killer (NK) T (NKT) cells are a unique subset of T lymphocytes that express NK cell markers and represent one subset of innate immune cells within the liver (30%-50% of liver lymphocytes).9 NKT cells possess an invariant T-cell receptor (Vα14-Jα18) and recognize glycosphingolipids.9 NKT cell activation by glycolipid antigens occurs through the major histocompatibility complex class I–like molecule, CD1d, which presents glycolipid antigens to the T-cell receptor.10 Although the role of NKT cells in AILI has not been examined directly, an earlier study using anti-NK1.1 antibody (Ab), which depletes both NK and NKT cells, demonstrated a protoxicant role of the combination.11 The data suggested that interferon (IFN)-γ secretion from NK/NKT cells was responsible for induction of inflammatory mediators, enhanced leukocyte recruitment, and Fas ligand expression. A more-recent study revealed that this pathogenic role of NK and NKT cells in AILI was likely the result of dimethyl sulfoxide (DMSO) that was used as a solvent for APAP.12 It was found that DMSO increased the number and activation of hepatic NK and NKT cells. As such, the role of NK and/or NKT cells in AILI remains to be elucidated. In the present study, we aimed to investigate the specific role of NKT cells in AILI by the use of mouse models of genetic deletion of NKT cells (CD1d−/− and Jα18−/−).
Our data showed that both CD1d−/− and Jα18−/− mice developed higher degrees of liver injury than wild-type (WT) mice after APAP challenge. This increased susceptibility in NKT cell-deficient mice was the result of their increased expression and activity of cytochrome P450 2E1 (CYP2E1), resulting in enhanced APAP metabolism and protein adduct formation.
Female and male Balb/cJ WT, CD1d−/−, C57Bl/6J WT (Jackson Laboratories, Bar Harbor, ME), and Jα18−/− mice13 (gift from Dr. Laurent Gapin, National Jewish Health, Denver, CO) were maintained in the Center for Animal Care. Mice (7-10 weeks old) were allowed food and water ad libitum until experimental use. Before treatment, food was withheld overnight (16 hours). APAP (Sigma-Aldrich, St. Louis, MO), dissolved in warm phosphate-buffered saline, was administered by intraperitoneal (IP) injection and food restored. After various time points, blood and liver tissues were collected.
Livers were sonicated in 0.1 N of perchloric acid (1:20, w/v). Glutathione (GSH) was measured by high-performance liquid chromatography (HPLC) equipped with electrochemical detection, using a CoulArray system (ESA, Chelmsford, MA).
Measurement of Mitochondrial Membrane Potential and Reactive Oxygen Species.
Mitochondria were isolated by homogenization of liver tissue (0.5 g), followed by two centrifugation steps at 650×g and 5,400×g. JC-1 dye (5 μM; Molecular Probes, Grand Island, NY) or MitoSOX dye (10 μM; Invitrogen, Grand Island, NY) was added to mitochondrial pellets (1 mg/mL). Membrane potential and reactive oxygen species (ROS) were detected by fluorescence excitation/emission spectra of 490/590 and 485/520 nm, respectively.
CYP2E1 and Proteasomal Activity.
CYP2E1 activity of microsomal protein was measured by hydroxylation of p-nitrophenol, as previously described.14 Proteasomal activity of liver homogenates were assayed for chymotrypsin-like (CT-L) and trypsin-like (T-L) activity, as previously described.15
Ketone Body Measurement.
Serum 3-hydroxybutyrate (BOH) was measured using the EnzyChrom Ketone body assay kit (BioAssay Systems, Hayward, CA). Absorbance was measured at 340 nm.
Statistical analysis was performed using the Student t test. Differences in values were considered significant at P < 0.05.
CD1d−/− Mice Are Significantly More Susceptible to AILI Than WT Mice.
Female WT and CD1d−/− mice were IP injected with APAP (385 mg/kg). CD1d−/− mice displayed significantly greater serum alanine aminotransferase (ALT) levels than WT mice at 8 and 24 hours post-APAP challenge (Supporting Fig. 2). Moreover, a significant decrease in survival was also observed in CD1d−/− mice, compared to WT mice, starting at 8 hours post-APAP challenge. Only 25% of CD1d−/− mice survived at 24 hours, whereas all the WT mice survived (Fig. 1A). When a lower dose of APAP (350 mg/kg) was administered, marked increases in serum ALT levels were observed in CD1d−/− mice, compared to WT mice, at 24 and 48 hours post-APAP challenge (Fig. 1B). Blinded histopathological evaluation of hematoxylin and eosin (H&E)-stained liver tissue samples was performed. Histological analysis revealed more-dramatic liver injury in CD1d−/− mice, compared to WT mice, 48 hours post-APAP challenge (Fig. 1E, F). To determine whether increased susceptibility of CD1d−/− mice to AILI is gender specific, we further compared susceptibilities of male WT and CD1d−/− mice to AILI. Similar to female mice, a decrease in survival was observed in male CD1d−/− mice, compared to WT mice, starting at 8 hours with no mice surviving at 48 hours post-APAP challenge (235 mg/kg; Fig. 1C). At this dose, CD1d−/− mice displayed significantly greater serum ALT levels at 8 hours, compared to WT mice; however, there were not enough mice that survived to measure ALT at 24 and 48 hours post-APAP challenge (Supporting Fig. 2). At a lower dose, in which both groups of mice survive, ALT levels were significantly higher in CD1d−/− mice, compared to WT mice, at 24 and 48 hours post-APAP challenge (230 mg/kg; Fig. 1D).
Increased APAP-Protein Adduct Formation in CD1d−/− Mice, Compared to WT Mice.
The mechanism of AILI involves APAP biotransformation into NAPQI, which depletes GSH in the liver. Upon GSH depletion, NAPQI binds to hepatocellular proteins, forming APAP protein adducts.16 To assess whether differential amounts of APAP protein adducts are formed in WT and CD1d−/− mice after APAP challenge, female WT and CD1d−/− mice were treated for 2 hours with APAP (350 mg/kg). Levels of hepatic protein adducts were significantly increased in CD1d−/− mice, compared to WT mice (Fig. 2A,C). Studies have shown that adduct formation in the mitochondria is essential in APAP toxicity, because this leads to induction of mitochondrial ROS formation and mitochondrial permeability transition.17 Therefore, mitochondria were isolated after 2-hour APAP challenge to measure APAP protein adducts. Levels of mitochondrial protein adducts were significantly higher in CD1d−/− than WT mice (Fig. 2B,D).
Increased Mitochondrial ROS and Dysfunction in CD1d−/− Mice Than WT Mice.
Mitochondrial ROS induction has been demonstrated after APAP challenge.17 In agreement with these findings, we observed a significant increase in mitochondrial superoxide after 1-hour APAP challenge in WT mice. Importantly, CD1d−/− mice exhibited significantly higher superoxide levels in mitochondria, compared to WT mice (Fig. 3A). Interestingly, we also observed a significant increase in superoxide levels after 16-hour starvation of CD1d−/− mice, but not in WT mice (Fig. 3A). Associated with the increase in mitochondrial ROS, there was a significant decrease in mitochondrial membrane potential (MMP) in CD1d−/− mice, compared to WT mice, after starvation as well as 1 and 2 hours after APAP challenge (Fig. 3B). These data indicate that CD1d−/− mice are uniquely susceptible to mitochondrial oxidative stress and dysfunction after starvation and APAP challenge.
Similar GSH Levels in WT and CD1d−/− Mice Before and After APAP Challenge.
Covalent binding of NAPQI by GSH represents an important defense mechanism against APAP toxicity. To assess whether GSH levels were innately different between WT and CD1d−/− mice, liver GSH levels in naïve mice were measured, and data showed similar levels in WT and CD1d−/− mice. Starvation of mice for 16 hours caused a similar reduction in GSH levels (approximately 50%) in WT and CD1d−/− mice (Fig. 4). After APAP challenge, GSH levels in WT and CD1d−/− mice decreased to the lowest level at 2 hours and began to rebound at 8 and 19 hours. GSH levels were lower in CD1d−/− than WT mice at 8 hours post-APAP challenge, perhaps the result of enhanced hepatotoxicity in these mice.
CYP2E1 Protein Expression and Activity Are Higher in CD1d−/− Mice, Compared to WT Mice, After Starvation.
CYP2E1 is the major metabolizing enzyme in the biotransformation of APAP into NAPQI.3 Therefore, we compared expression levels of CYP2E1 in WT and CD1d−/− mice. No difference in CYP2E1 protein levels between naïve WT and CD1d−/− mice was observed. However, after 16-hour starvation, CYP2E1 protein levels were significantly higher in CD1d−/− mice, compared to WT mice (Fig. 5A,B,D). Although immunoblotting analysis did not show changes in CYP2E1 protein levels in WT mice upon starvation, immunohistochemical (IHC) staining revealed a significant increase in CYP2E1 expression in WT mice; however, starvation-induced up-regulation of CYP2E1 was markedly greater in CD1d−/− mice. CYP2E1 activity was higher in CD1d−/− mice than WT mice after 16-hour starvation (Fig. 5C). To confirm that increased susceptibility of CD1d−/− mice was the result of starvation-induced up-regulation of CYP2E1, we treated naïve nonstarved female WT and CD1d−/− mice with APAP (700 mg/kg). The results showed no difference in serum ALT levels between WT and CD1d−/− mice (Supporting Fig. 1).
Similar Cyp2e1 Messenger RNA and Proteasomal Degradation, but Significantly Higher Ketone Body Levels, in CD1d−/− Mice, Compared to WT Mice.
CYP2E1 regulation occurs at both transcriptional and post-translational levels.18, 19 To explore the mechanism of increased CYP2E1 protein expression in CD1d−/− mice after starvation, we examined hepatic Cyp2e1 messenger RNA (mRNA) levels. Real-time polymerase chain reaction (PCR) analysis of naïve and starved WT and CD1d−/− mice demonstrated that starvation did not up-regulate Cyp2e1 mRNA and that mRNA levels were similar between WT and CD1d−/− mice (Fig. 6A). Furthermore, proteasome peptidase activities, measured by CT-L and T-L activity analyses, were similar between WT and CD1d−/− mice after starvation (Fig. 6B,C), suggesting that differential proteasomal degradation cannot explain the increased expression level in CD1d−/− mice. Post-translational stabilization mediated by substrate binding is another possible mechanism accounting for the increased protein expression of CYP2E1.20 BOH represents a main ketone body produced in the liver, which is freely converted into acetoacetate and broken down into acetone, two molecules reported to stabilize CYP2E1 post-translationally.18, 21 Our data demonstrated that naïve WT and CD1d−/− mice exhibited similar levels of BOH in serum. Starvation of mice increased BOH levels in WT and CD1d−/− mice. A significantly greater elevation of BOH levels was observed in CD1d−/− mice, compared to WT mice (Fig. 6D).
Increased Susceptibility of Jα18−/− Mice to AILI.
To determine whether increased susceptibility of CD1d−/− mice to AILI was the result of NKT cell depletion, but not an unexpected effect of CD1d deletion, we examined another strain of NKT cell-deficient mice (i.e., Jα18−/− mice). Female WT and Jα18−/− mice were injected with APAP (350 mg/kg). A marked increase in serum ALT levels was observed in Jα18−/− mice, compared to WT mice, at 8, 24, and 48 hours post-APAP challenge (Fig. 7A). Similar to female mice, male Jα18−/− mice developed a greater degree of injury, compared to WT mice (Fig. 7B). Significantly higher APAP protein adducts and a significant decrease in MMP were observed in Jα18−/− mice after 1-hour APAP challenge, compared to WT mice (Fig. 7 C,D). A trending decrease in membrane potential was observed after starvation in Jα18−/−, compared to WT mice (data not shown). Similar to CD1d−/− mice, starvation resulted in a significant increase in CYP2E1 protein levels in Jα18−/− mice than WT mice (Fig. 7E). Last, starvation caused greater elevations of serum BOH levels in Jα18−/− mice, compared to WT mice (Fig. 7F). These data confirm that it was the deletion of NKT cells that rendered mice more susceptible to AILI.
Our data demonstrate that NKT cell-deficient mice are more susceptible to AILI than WT mice. This is the result, in part, to starvation-induced up-regulation of CYP2E1 protein expression and activity, which is associated with marked increases in hepatic APAP protein adduct formation. Starvation also caused greater elevations of ketone bodies in NKT cell-deficient mice, which may account for the increase in CYP2E1 protein levels.
Upon activation, NKT cells rapidly produce cytokines, such as interleukin (IL)-4 and IFN-γ.9 Many studies have shown both protective and pathological functions of these cytokines in liver disease models.22, 23 Based on these findings, we examined whether differential production of these cytokines between WT and CD1d−/− mice may explain the increased susceptibility of NKT cell-deficient mice to AILI. However, message levels of a number of cytokines were similar in liver tissues and isolated liver mononuclear cells (in which NKT cells are enriched) from APAP-treated WT and CD1d−/− mice (data not shown). These results suggest that APAP treatment does not trigger NKT cells to produce protective cytokines.
It is established that APAP metabolism to NAPQI and its covalent modification of liver proteins are essential in triggering hepatocyte damage.16 Because a series of downstream events, such as mitochondrial dysfunction, ATP depletion, and DNA damage, take place before ALT release, there is a delay between NAPQI generation and increase of serum ALT levels. Compared to WT mice, CD1d−/− mice had significantly higher levels of APAP protein adducts as early as 2 hours post-APAP (Fig. 2); however, depending on the dose of APAP, a significantly higher ALT level was not observed until 8 (Supporting Fig. 2) or 24 hours after APAP challenge (Fig. 1).
GSH plays a pivotal role in AILI through scavenging of NAPQI. It has been demonstrated that mice deficient in both IL-10 and IL-4 (IL-10/4−/− mice) are more susceptible to AILI, compared to WT mice. This appears to be the result of lower GSH levels in IL-I0/4−/− mice before, and more dramatically after, APAP challenge.24 We observed no differences in total GSH levels in livers of naïve or starved WT and CD1d−/− mice (Fig. 4). Although there appears to be a slight delay in GSH rebound in the CD1d−/− mice at 8 hours, GSH levels were similar in WT and CD1d−/− mice at 19 hours after APAP treatment. Furthermore, we did not observe significant differences in expression and holoenzyme formation of glutamate cysteine ligase, the rate-limiting enzyme for GSH synthesis (data not shown). These data suggest that increased susceptibility of NKT-deficient mice to AILI, compared to WT mice, is not caused by differential levels of GSH stores or synthesis.
Although CYP2E1 protein expression is similar in naïve WT and NKT cell-deficient mice, it is significantly higher in CD1d−/− and Jα18−/− mice than WT mice upon starvation (Figs. 5A,B,D and 7E). CYP2E1 activity is induced under a variety of physiological and pathological conditions, including chronic alcohol consumption, nonalcoholic steatohepatitis, and diabetes.20 CYP2E1 protein levels, but not mRNA levels, have been shown to increase 2- to 8-fold after treatment with ethanol, acetone, pyrazole, and isoniazid. In pathological conditions, such as diabetes, and obesity, CYP2E1 levels have been observed to increase 3- to 8-fold at both mRNA and protein levels.20 The elevation of CYP2E1 after these conditions has been attributed to changes in metabolism, specifically, the increase of ketone bodies during these states.25 Our data demonstrated no significant difference in transcriptional activation in starved WT and CD1d−/− mice (Fig. 6A). WT and CD1d−/− mice displayed similar amounts of proteasomal activity after starvation (Fig. 6B,C), indicating that a change in overall proteasomal function was not responsible for increased CYP2E1 protein and activity. CYP2E1 substrates, such as acetone, pyrazole, and ethanol, have been reported to enhance CYP2E1 protein expression through increasing of protein stability.26 Studies of in vivo protein labeling in rats revealed a biphasic turnover of CYP2E1 at 7 and 32 hours. Acetone treatment resulted in loss of the 7-hour degradation of CYP2E1, a process termed “substrate-induced stabilization.”18 Computational modeling of a predicted cytosolic domain of CYP2E1 identified a potential ubiquitylation site, which may also serve as a site for substrate interaction. This finding provides a possible mechanism for the ability of substrate to bind and shield the enzyme from proteasomal degradation.27 Additional CYP enzymes have been shown to be regulated by substrate-induced stabilization. For example, CYP3A protein is stabilized by troleandomycin.28
Ketone bodies are produced primarily in the liver and serve as a source of energy during starvation. Our data demonstrated that, after 16-hour starvation, NKT cell-deficient mice produced significantly higher amounts of BOH than WT mice (Figs. 6D and 7F). The correlation of increased ketone bodies to induction of CYP2E1 is supported by many reports. In a rat model of streptozocin-induced hyperketonemia, increased CYP2E1 protein expression and activity were observed.29 Diabetic rats with severe ketosis, consisting of high BOH in plasma, were found to have significantly higher CYP2E1 than nondiabetic control mice.25 Furthermore, treatment of cultured mouse hepatocytes with acetoacetate stabilizes CYP2E1 protein expression in vitro.21 Acetone has also been implicated in the induction of CYP2E1 activity. When administered to rats in drinking water, acetone induced CYP2E1 2-fold higher, compared to control.18 Aside from the up-regulation of CYP2E1, starvation of CD1d−/− mice significantly affected mitochondria, as evident by increased ROS production and decreased mitochondrial viability (Fig. 3). This may be attributable to starvation-induced elevation in ketone bodies, because they are known to be produced in the mitochondrial matrix of hepatocytes and have been shown to induce mitochondrial ROS and dysfunction.30 Elevation of ketone bodies (acetoacetate) has been associated with decreased GSH levels in diabetic patients as well as in vitro cell-culture models.31 Because GSH is a potent ROS scavenger, reduction in GSH levels is important in causing mitochondrial dysfunction. Mitochondrial impairment was dramatically worsened in CD1d−/− and Jα18−/− than WT mice upon APAP challenge, which likely contributes to increased susceptibility of CD1d−/− mice to AILI (Figs. 3B and 8D).
Increased ketone body production in NKT cell-deficient mice suggests an underlying role of NKT cells in metabolism. Several lines of evidence support a link between NKT cells and metabolism. Patients with abetalipoproteinemia, a rare Mendelian disorder characterized by a lack of functional microsomal triglyceride transfer protein, also exhibits reduced number of NKT cells and impaired functionality of these cells.32 In murine models of obesity (ob/ob mice), NKT cells are decreased in number.33 Upon adoptive transfer of NKT to ob/ob mice, a significant reduction in liver steatosis was observed, coinciding with marked improvement in glucose sensitivity.34 Furthermore, stimulation and expansion of NKT cell populations by means of norepinephrine or glucocerebroside injection has been shown to decrease size and fat accumulation in the liver and decrease overall hepatic injury.35
The mechanisms by which NKT cells regulate metabolism during conditions of energy deficit or oversupply remain largely unknown, despite several recent studies on this topic.36, 37 We hypothesize that intrinsic IL-4 production by NKT cells may be critical in maintaining metabolic homeostasis. A recent report suggests that IL-4 activation of signal transducer and activator of transcription 6 in hepatocytes can regulate fatty acid (FA) oxidation by suppression of peroxisome proliferator-activated receptor alpha.38 It is also reported that IL-4 increases thermogenic gene expression, FA mobilization, and energy expenditure by means of stimulating alternatively activated macrophages.39 Another study demonstrated that IL-4 produced by eosinophils in adipose tissue is important in protecting mice from high-fat-diet–induced obesity.40 It is our plan for future studies to examine the role of endogenous IL-4 production by NKT cells in metabolic regulation, which will require the use of IL-4-reporter mice.
In conclusion, our data demonstrate that NKT cells protect mice from AILI because genetic deletion of these cells causes significantly higher ketone body production upon starvation. Increased ketone bodies lead to mitochondrial stress even before APAP challenge, rendering mice more susceptible to another insult. Increased ketone bodies also stabilize CYP2E1 protein, resulting in a marked increase of APAP bioactivation to generate the hepatotoxic metabolite, which causes liver injury (Fig. 8). We found that message levels of a number of cytokines were similar in liver tissues and liver mononuclear cells (in which NKT cells are enriched) isolated from APAP-treated WT and CD1d−/− mice (data not shown). These results suggest that APAP treatment does not trigger NKT cells to produce protective cytokines. Our data do not support an active protective role for NKT cells, but rather that the lack of NKT cells renders mice more susceptible to AILI. This is the first study to examine the specific role of NKT cells in AILI. The findings provide further insights into the underlying mechanisms of drug-induced liver injury, as well as other liver conditions in which CYP2E1-mediated ROS generation plays an important pathological role.41 Aside from genetic conditions, such as abetalipoproteinemia, lipid antigens, bacterial, and viral pathogens have been demonstrated to activate NKT cells, which leads to decreased cell number.42 Under such situations, NKT cell deficiency may result in increased susceptibility to metabolic stress, as well as hepatotoxin-induced liver injury.
The authors thank Drs. Chris Franklin and Don Backos for their assistance with glutathione cysteine ligase western blotting analysis. The authors thank Casey Trambly for conducting the proteasome and CYP2E1 activity assays and Dr. James Galligan for assistance in CYP2E1 IHC. Special thanks to Dr. Sean Colgan for the generous use of HPLC instrumentation and Brittelle Bowers and Adrianne Burgess for their technical assistance with HPLC setup.