Preactivation of NKT cells with α-GalCer protects against hepatic ischemia-reperfusion injury in mouse by a mechanism involving IL-13 and adenosine A2A receptor

Zongxian Cao, Youzhong Yuan, Geetha Jeyabalan, Qiang Du, Allan Tsung, David A. Geller, Timothy R. Billiar

Abstract

Hepatic preconditioning has emerged as a promising strategy of activating natural pathways to augment tolerance to liver ischemia-reperfusion (IR) injury. Liver-resident natural killer T (NKT) cells play an important role in modulating the local immune and inflammatory responses. This work was aimed to investigate whether preactivation of NKT cells could provide a beneficial “preconditioning” effect to ameliorate the subsequent hepatic IR injury. To selectively activate NKT cells, C57BL/6 mice were treated intraperitoneally with the glycolipid antigen α-galactosylceramide (α-GalCer) 1 h prior to hepatic ischemia. Significantly reduced liver IR injury was observed in mice pretreated with α- GalCer, and this protective effect was specifically abrogated by a CD1d blocking antibody. Serum TNF-α, IFN-γ, and IL-13 levels were markedly increased shortly after α-GalCer injection. Pretreatment with a neutralizing antibody against TNF-α or IFN-γ did not influence the protective effect of α-GalCer preconditioning, whereas preadministration of an IL-13 neutralizing antibody completely abolished the effect. Treatment with α-GalCer also led to an increased expression of adenosine A2A receptor (A2AR) in the liver, and blockade of A2AR by SH58261 diminished α-GalCer pretreatment-mediated attenuation of liver IR injury. In contrast, administration of the selective A2AR agonist CGS21680 reversed the counteracting effect of the IL-13 neutralizing antibody on α-GalCer preconditioning. Additionally, α-GalCer pretreatment was associated with a decreased neutrophil accumulation in the ischemic liver. These findings provide the first evidence that hepatic preconditioning by preactivation of NKT cells with α-GalCer protects the liver from IR injury via an IL-13 and adenosine A2AR-dependent mechanism.

  • liver protection
  • natural killer T cells
  • glycolipid
  • Th1/Th2 cytokines
  • α-galactosylceramide

liver damage caused by ischemia and reperfusion (IR) is well recognized as a significant cause of morbidity and mortality in a number of clinical settings, including liver transplantation, major hepatic resections, and hypovolemic shock with resuscitation (9, 30, 34). Many strategies have been proposed to mitigate hepatic IR injury either by directly interfering with the pathways of liver IR injury or by a preemptive induction of hepatic tolerance against IR injury, a phenomenon known as hepatic preconditioning (6, 18, 29, 30). Preconditioning intervention, particularly ischemic preconditioning (IPC), has emerged as a powerful method of ameliorating hepatic IR injury in both animal studies and clinical situations (18, 26). Many other experimental strategies of liver preconditioning have also been reported, such as hyperthermic and pharmacological preconditioning (13, 24, 29, 30, 36, 40). However, to date there are no treatments available that can effectively prevent hepatic IR injury.

Studies on the mechanisms of IR-induced injury are likely to provide insights for developing potential therapeutic approaches to limit ischemic tissue damage following reperfusion. Recently, several lines of evidence suggest that natural killer T (NKT) cells play an important role in initiating liver IR injury. The number of hepatic NKT cells was shown to increase during liver IR and reduced liver IR injury was observed in NKT cell-deficient mice (31). NKT cells were shown to be activated following liver IR, and adoptive transfer of wild-type NKT cells restored liver IR injury in RAG1 knockout mice (19). NKT cells represent a group of T cells that are reactive to self- and foreign glycolipids presented by the major histocompatibility complex class I-related molecule CD1d (2, 10). NKT cells are highly abundant in the liver and are considered as a unique component of innate immunity, with the capacity of rapid activation to produce immunoregulatory cytokines, many of which such as TNF-α and IFN-γ are also important mediators of liver IR injury (2, 10, 37). The underlying biological principle of liver preconditioning is that preexposure of the liver with a low level of IR-related stress can trigger endogenous defense mechanisms against the subsequent reperfusion injury (6, 29). The recent insights of the critical role of NKT cells in liver IR injury raise the possibility that this distinct cell population could be harnessed as a potential target to develop a novel hepatic preconditioning strategy to protect liver from IR injury.

In the present study, we treated C57BL/6 mice with α-galactosylceramide (α-GalCer) to selectively activate NKT cells prior to hepatic ischemia and investigated its effect on the subsequent liver reperfusion injury. Our data indicate that NKT cell activation by α-GalCer 1 h prior to the ischemic insult significantly suppressed the severity of liver damage induced by hepatic reperfusion. Furthermore, we provide evidence that the protective effect is mediated by IL-13 and adenosine A2A receptor (A2AR).

EXPERIMENTAL PROCEDURES

Animals and reagents.

Male C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). The animals were maintained in a laminar-flow, specific pathogen-free atmosphere. The research project was approved by the University of Pittsburgh Animal Care and Use Committee and was in compliance with the National Institutes of Health guidelines for the Use of Laboratory Animals. The reagent α-GalCer was purchased from AXXORA (San Diego, CA). DMSO was initially used to dissolve α-GalCer to a concentration of 1 mg/ml, and the solution was then diluted to 0.2 mg/ml in 1×PBS containing 0.5% Tween-20. Neutralizing antibodies for mouse TNF-α, IFN-γ, and IL-13 were obtained from R&D Systems (Minneapolis, MN). The clone 1B1 monoclonal antibody reacting with mouse CD1d was ordered from eBioscience (San Diego, CA). SH58261 and CGS21680 were purchased from Tocris Bioscience (Ellisville, MO).

Mouse model of hepatic ischemia and reperfusion injury.

Mice were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg). A midline laparotomy was performed to expose the liver. The hepatic hilum was carefully dissected and an atraumatic vascular clamp was used to interrupt the portal circulation to the median and left lateral lobes of the liver. The abdomen was covered with a piece of saline-soaked gauze to prevent dehydration. The mice were placed on a heating pad (37°C) to prevent drop of core temperature. After 60 min of partial hepatic ischemia, the clamp was removed to initiate hepatic reperfusion. The abdominal wall was sutured and the animals were returned to their cages. After 6 h of reperfusion, mice were anesthetized with isoflurane inhalation (Baxter Healthcare, Deerfield, IL), the abdomens and thoraxes were opened, and whole blood samples were collected by cardiac puncture before the removal of the left liver lobes for histological analysis. Sham surgeries were performed by following similar procedure but without vascular occlusion.

Animal treatment protocols.

All animal treatment protocols were in accordance with institutional animal care guidelines. To preactivate hepatic NKT cells, mice were treated by intraperitoneal (ip) injection with 40 μg/kg of α-GalCer 1 h prior to the initiation of liver ischemia. Neutralizing antibody against TNF-α or IFN-γ was administered ip at a dose of 150 μg per mouse 30 min before α-GalCer treatment. SH58261 (5 mg/kg) and CGS21680 (2 mg/kg) were given ip 30 min before α-GalCer treatment. All the reagents for ip injection were diluted in sterile normal saline in a final volume of 500 μl.

Measurement of serum ALT levels.

Blood samples were obtained by cardiac puncture at the time of euthanasia and were kept at 4°C overnight to allow the serum to separate from the clot. Serum preparations were collected following centrifugation at 1,000 g for 15 min. Serum alanine aminotransferase (ALT) levels were determined by using the Opera Clinical Chemistry System (Bayer, Tarrytown, NY). The ALT values are expressed as international units per liter.

Histopathology and immunohistochemistry.

Liver tissue slices were cut from the left liver lobes, fixed overnight in 10% buffered formalin, and embedded with paraffin. Standard hematoxylin and eosin staining was performed on the tissue sections (5-μm thickness) to evaluate liver damage. For immunohistochemical staining of hepatic neutrophils, the liver tissue sections were deparaffinized and rehydrated, followed by antigen retrieval using an antigen unmasking solution (Vector Laboratory, Burlingame, CA). The slides were incubated with 0.1% hydrogen peroxide to quench endogenous peroxidase activity. The sections were incubated overnight at 4°C with a primary antibody (clone 7/4, rat IgG2a) specific to mouse neutrophils (Cedarlane, Westbury, NY) diluted to 5 μg/ml. The immune complexes were visualized by use of a rat ABC Staining Kit (Santa Cruz Biotechnology, Santa Cruz, CA). The sections were counterstained with Vector Hematoxylin QS (Vector Laboratory) and were mounted with the Crystal/Mount mounting medium (Biomeda, Foster City, CA). The specimens were examined and images were acquired by using an Olympus Provis microscope with a digital camera.

ELISA assay for serum cytokine levels.

Quantikine ELISA kits (R&D Systems, Minneapolis, MN) specific for mouse TNF-α, IFN-γ, or IL-13 were used to determine the serum cytokine concentrations according to the manufacturer's instructions. All samples were assayed in duplicate.

RT-PCR.

Total cellular RNA was isolated from frozen liver tissues using TRIzol Reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. An aliquot of 1 μg total RNA was used as templates to synthesize the first-strand cDNA by using the TITANIUM One-Step RT-PCR Kit (Clontech, Mountain View, CA). The primers used for semiquantitative RT-PCR for adenosine receptors were described previously (4). PCR was performed for 35 cycles, with each cycle at 94°C for 45 s, 60°C for 45 s, and 68°C for 1 min.

Isolation of hepatic MNCs and flow cytometric analysis.

Mice were anesthetized by isoflurane inhalation, their abdomens and thoraxes were opened, and the blood was drained by cardiac puncture before the removal of the liver. The liver was cut into small pieces and gently pressed through a 200-gauge stainless steel mesh. The liver cell suspension was collected and centrifuged at 50 g for 5 min to remove hepatocytes and tissue debris. The supernatant was then centrifuged at 300 g for 10 min and a red blood cell lysis buffer (eBioscience, San Diego, CA) was added to the cell pellet and incubated for 5 min at room temperature. After being washed twice in PBS, the cells were resuspended in 37% Percoll (GE Healthcare Bio-Sciences, Uppsala, Sweden) in RPMI 1640 medium. The cell suspension was gently overlaid onto 70% Percoll and centrifuged for 25 min at 800 g. Purified hepatic mononuclear cells (MNCs) were collected from the interface, washed twice in PBS, and resuspended in a FACS staining buffer (PBS supplemented with 2% fetal bovine serum and 0.09% sodium azide).

To analyze intracellular IFN-γ expression in NKT cells, the hepatic MNC were first stained with a FITC-conjugated anti-CD3 monoclonal antibody (mAb) and an allophycocyanin (APC)-conjugated NK1.1 mAb (eBioscience) at 4°C for 30 min. The cells were washed twice in the staining buffer and were then fixed in BD Fixation/Permeabilization solution for 20 min at 4°C. After fixation, the cells were washed with BD Perm/Wash solution and resuspended in 50 μl of the solution containing either a phycoerythrin (PE)-Cy7-conjugated anti-IFN-γ mAb or a PE-Cy7-conjugated isotype control IgG for 30 min at 4°C. To analyze neutrophils in the liver by FACS, hepatic MNC were stained with a PE-conjugated CD11b mAb and a Pacific blue-conjugated Gr-1 mAb. The stained cells were analyzed on BD LSR II System (Becton Dickinson, San Jose, CA) with the BD FACSDiva software.

MACS purification of CD4+ and NK1.1+NKT cells and cell cultures.

Mouse liver MNCs were isolated as described above. NKT cells (CD4+/NK1.1+), natural killer (NK) cells (CD4−/NK1.1+), and NK1.1− cells were sorted by use of the MACS system (Miltenyi Biotec, Auburn, CA). Briefly, MNCs pooled from three mouse livers were incubated with an APC-conjugated anti-mouse NK1.1 mAb (eBioscience, San Diego, CA) in 100 μl of MACS buffer for 30 min at 4°C. The cells were washed and then incubated with an anti-APC MicroBeads for 30 min at 4°C. After being washed twice, the cells were washed and transferred onto a LS separation column attached to a MidiMACS separation unit. The cells in the elution are NK1.1− cells. The column-bound NK1.1+ cells were collected and incubated with a Multisort release reagent to remove the MicroBeads. The cells were then incubated with a mouse CD4 MicroBeads and were passed through a separation column. The cells in the elution are NK cells (CD4−/NK1.1+). The column-bound cells are NKT cells (CD4+/NK1.1+). The cells were recovered, counted, and cultured in 100 μl RPMI medium containing 10% fetal bovine serum. Supernatants were collected after 24 h of culture for cytokine determination. The cells were enriched to >98% purity as determined by FACS analysis. The numbers of cells purified from the MNCs pooled from three mouse livers are 2.55×106 of NK1.1− cells, 4.5×105 of NKT cells, and 1.45×105 of NK cells.

Statistical analysis.

All data were expressed as means ± SE. Comparison of means between two groups was performed by Student's t-test. Comparisons between multiple groups were analyzed by one-way analysis of variance with a Dunnett's post hoc test. Statistical significance was inferred by P < 0.05. All the statistical analyses were performed with SPSS statistical software (SPSS, Chicago, IL).

RESULTS

Pretreatment of mice with α-GalCer 1 h prior to an ischemic insult suppressed subsequent hepatic reperfusion injury.

A mouse model of partial hepatic warm IR was used in this work, which consists of 1 h of ischemia followed by 6 h of reperfusion. This model has been optimized and extensively utilized by our group to obtain a reproducible level of hepatic damage. To activate hepatic NKT cells, C57BL/6 mice were treated with an ip injection of α-GalCer, a synthetic glycolipid that has been widely used as a potent agonist to selectively activate NKT cells both in vitro and in vivo (2, 10, 37). On the basis of current literature, α-GalCer was used mostly at a dose of 2 μg per mouse to achieve maximal activation of NKT cells in vivo (3, 7, 25). In the present study, we performed preliminary experiments and chose a dose of 40 μg/kg, which is equivalent to 1 μg per mouse with a body weight of 25 g. Liver NKT cells were efficiently activated in mice treated with α-GalCer at this dose, whereas no detectable liver injury was observed up to 7 h after treatment, which is the time frame for the hepatic IR procedure used in this study. In vehicle-pretreated mice subjected to IR, serum ALT levels were significantly increased compared with the sham-operated control animals (Fig. 1A). The liver damage was confirmed histologically by the presence of hepatic necrosis (Fig. 1B). Treatment of mice with α-GalCer 1 h prior to ischemia significantly reduced serum ALT levels and liver tissue necrosis (Fig. 1). α-GalCer pretreatment alone did not increase serum ALT levels and did not cause appreciable histological alterations in the liver of the sham control animals (Fig. 1). α-GalCer injection induced a rapid hepatic NKT cell activation as shown by elevated intracellular IFN-γ expression, which was abrogated by administration of a CD1d blocking antibody that selectively inhibits CD1d-mediated presentation of α-GalCer to NKT cells (Fig. 2B). Consistent with previous reports (19, 31), we found that CD1d blocking antibody pretreatment reduced serum ALT levels in animals subjected to liver IR (Fig. 2A). Pretreatment with α-GalCer combined with the CD1d blocking antibody did not provide further protection beyond that achieved by CD1d blockade (Fig. 2A). Together, these experiments demonstrated that hepatic preconditioning with α-GalCer protected the liver from IR injury via a specific CD1d-restricted NKT cell activation.

Fig. 1.

α-Galactosylceramide (α-GalCer) pretreatment reduces hepatic ischemia-reperfusion (IR) injury. C57BL/6 mice were pretreated by intraperitoneal injection (ip) with vehicle or 40 μg/kg of α-GalCer for 1 h, followed by 1 h of liver ischemia and 6 h of reperfusion. Sham animals received the same treatments as IR group. A: serum alanine aminotransferase (ALT) levels were assayed and presented as means ± SE of n = 4–6 mice per group. (*P < 0.05 vs. vehicle-treated IR group). B: representative histological images of liver tissue sections stained with hematoxylin and eosin (original magnification ×200).

Fig. 2.

Pretreatment with a CD1d blocking antibody (clone 1B1) abrogates the protective effect of α-GalCer pretreatment on hepatic IR injury. A: mice were pretreated ip with anti-CD1d or a control IgG Ab for 30 min and were then treated with vehicle or 40 μg/kg of α-GalCer for 1 h, followed by 1 h of hepatic ischemia and 6 h of reperfusion. Serum alanine aminotransferase ALT levels were measured and presented as means ± SE (n = 3–5 mice per group; *P < 0.05 vs. the control group treated with PBS plus vehicle). B: liver mononuclear cells were isolated from animals treated as indicated. The cells were stained for natural killer (NK) T (NKT) cell surface marker CD3 and NK1.1, followed by intracellular staining of IFN-γ. The plots show the percentages of IFN-γ-positive cells in NKT cell population gated as double positive for CD3 and NK1.1. Unfilled dark gray curves, intracellular staining with an isotype control IgG; curves filled with light gray shading, intracellular staining with the IFN-γ Ab.

Neither TNF-α nor IFN-γ was involved in α-GalCer pretreatment-mediated protective effect on hepatic IR injury.

It is known that, upon activation by α-GalCer, NKT cells are able to produce a broad range of cytokines, such as IFN-γ, IL-4, TNF-α, IL-10, and IL-13 (3, 10, 37). Since both TNF-α and IFN-γ have been shown to play an important role in hepatic IR injury (5, 19, 27, 34), we sought to investigate whether these two proinflammatory cytokines were involved in the preconditioning effect of α-GalCer on liver IR injury. C57BL/6 mice were treated with 40 μg/kg of α-GalCer over a time course up to 24 h. TNF-α and IFN-γ levels in the sera were measured by ELISA. As shown in Fig. 3A, TNF-α rapidly reached maximal levels 2 h after α-GalCer application and gradually decreased to a lower level 12 h later. Serum IFN-γ peaked at 4 h and remained at a high level up to 24 h after α-GalCer administration. However, administration of a neutralizing antibody to either TNF-α or IFN-γ did not influence the protection against liver IR injury afforded by α-GalCer pretreatment (Fig. 3B). These data suggest that α-GalCer-induced TNF-α and IFN-γ expression were not responsible for the protective effect of α-GalCer preconditioning on hepatic IR injury.

Fig. 3.

TNF-α and IFN-γ are not responsible for the α-GalCer pretreatment induced protection against hepatic IR injury. A: mice were treated ip with 40 μg/kg of α-GalCer and were euthanized at the time points as indicated. TNF-α and IFN-γ levels in the serum were detected by ELISA. B: mice were pretreated with a neutralizing Ab against TNF-α, IFN-γ, or a control IgG Ab for 30 min and were then treated with vehicle or α-GalCer for 1 h, followed by 1 h of hepatic ischemia and 6 h of reperfusion. Serum ALT levels were detected and presented as means ± SE from 2 independent experiments with 3 mice in each group.

The type 2 cytokine IL-13 played an important role in α-GalCer preconditioning-mediated protection against liver IR injury.

One hallmark of NKT cells is their Th0-like cytokine expression pattern, with the capacity of making both Th1-type and Th2-type cytokines simultaneously (1, 3, 10, 35, 37). In this study, we observed that administration of the NKT cell selective agonist α-GalCer resulted in a significant increase in serum IL-13 levels, which peaked at 4 h and remained elevated up to 24 h after the treatment (Fig. 4A). We also demonstrated that NKT cells are the predominant cell population that produces IL-13 upon α-GalCer treatment (Fig. 4B). IL-13 is a type 2 cytokine that has remarkable anti-inflammatory effects in a number of inflammatory processes including hepatic IR injury (16, 22, 23, 39). To determine whether the increased IL-13 levels were responsible for the protective effect of α-GalCer preconditioning on liver IR injury, we pretreated the mice with a monoclonal neutralizing antibody against mouse IL-13. As shown in Fig. 4C, neutralization of IL-13 completely abolished α-GalCer preconditioning-mediated protection against liver IR injury compared with the control group receiving an isotype control antibody. These data therefore demonstrated that NKT cell-derived IL-13 plays an essential role in α-GalCer preconditioning-mediated protection against hepatic IR injury.

Fig. 4.

IL-13 plays an important role in the hepatoprotective effect of α- GalCer preconditioning for liver IR injury. A: mice were treated ip with 40 μg/kg of α-GalCer for the indicated time course. Serum IL-13 levels were measured by ELISA. B: mice were treated with vehicle or 40 μg/kg of α-GalCer for 3 h. NKT cells (CD4+/NK1.1+), NK cells (CD4−/NK1.1+), and NK1.1− cells were sorted by use of the MACS system. The cells were pooled from 3 mice in each group and were cultured in vitro for 24 h. The supernatants were collected for IL-13 ELISA assay. C: mice were pretreated with a neutralizing Ab against IL-13 or a control IgG Ab for 30 min and were then treated with vehicle or α-GalCer for 1 h, followed by 1 h of hepatic ischemia and 6 h of reperfusion. Serum ALT levels were assayed and presented as means ± SE (n = 5 mice per group; *P < 0.05 vs. the group receiving IgG plus α-GalCer treatment).

Adenosine A2AR was involved in the hepatoprotective effect of α-GalCer preconditioning for liver IR injury.

To further explore the mechanisms by which α-GalCer preconditioning confers protection against liver IR injury, we examined the potential role of adenosine and its receptor A2AR, which have been shown to play a critical role in ischemic preconditioning (6, 29). We found that α-GalCer treatment resulted in a rapid increase in A2AR mRNA expression in the liver, whereas transcripts for the A1, A2B, and A3 adenosine receptors remained unchanged (Fig. 5A). These data suggest that α-GalCer pretreatment could lead to an increased adenosine signaling via A2AR in the liver tissues. To block adenosine function through A2AR, mice were pretreated with 5 mg/kg of SH58261, a selective antagonist for the receptor. As shown in Fig. 5B, SH58261 pretreatment did not alter liver IR injury in mice of the vehicle-treated control group. However, SH58261 treatment effectively eliminated the reduction in serum ALT levels afforded by α-GalCer preconditioning. Preadministration of A2AR specific agonist CGS21680 reduced liver IR injury in mice of the control group receiving vehicle treatment, but CGS21680 treatment combined with α-GalCer did not result in further additive protection compared with preconditioning with α-GalCer alone (Fig. 5B). Furthermore, CGS21680 treatment completely reversed the counteracting effect of the IL-13 neutralizing antibody on α-GalCer preconditioning (Fig. 5C). We also found that both NKT cells and hepatocytes express A2AR, and α-GalCer treatment increased A2AR message in both cells (Fig. 5D). However, CGS21680 treatment did not significantly alter NKT-mediated IL-13 production (Fig. 5E). Taken together, these data suggest that, in addition to IL-13, adenosine and its receptor A2AR were important mediators of the protective effect of α-GalCer preconditioning on hepatic IR injury.

Fig. 5.

Role of adenosine A2A receptor in α-GalCer preconditioning-mediated protective effect on liver IR injury. A: mice were treated with vehicle or 40 μg/kg of α-GalCer for the indicated time points. Total cellular RNA was isolated from the liver tissues and subjected to RT-PCR using primer pairs specific for the 4 adenosine receptors. B: mice were pretreated with PBS, 5 mg/kg of SH58261, or 2 mg/kg of CGS21680 for 30 min and were then treated with vehicle or 40 μg/kg of α-GalCer for 1 h, followed by 1 h of liver ischemia and 6 h of reperfusion. Serum ALT levels were assayed and presented as means ± SE (n = 5 mice per group; *P < 0.05 vs. the control group treated with PBS plus vehicle; **P < 0.05 vs. the group receiving PBS plus α-GalCer treatment). C: mice were pretreated with PBS or 2 mg/kg of CGS21680 for 30 min and were then treated with a control IgG Ab or an IL-13 neutralizing Ab for 30 min, followed by treatment with vehicle or α-GalCer for 1 h prior to 1 h of liver ischemia and 6 h of reperfusion. Serum ALT levels were measured and presented as means ± SE (n = 5 mice per group; *P < 0.05 vs. the group receiving α-GalCer treatment only; **P < 0.05 vs. the group receiving anti-IL-13 Ab plus α-GalCer treatment). D: mice were treated with vehicle or 40 μg/kg of α-GalCer for 3 h. Hepatocytes were separated from the nonparenchymal mononuclear cells (MNCs). NKT cells were then purified from the MNCs by using the MACS system. Total cellular RNA isolated from the NKT cells and hepatocytes was subjected to RT-PCR using a primer specific for adenosine A2A receptor. β-Actin was amplified as an internal control. E, top: mice were pretreated with PBS or 2 mg/kg of CGS21680 for 30 min and were then treated with vehicle or 40 μg/kg of α-GalCer for 3 h. Serum IL-13 levels were assayed and presented as means ± SD (n = 3 mice per group). Bottom: NKT cells were purified from the liver MNCs of the mice treated as described above. The cells were pooled from the 3 mice in each group and were cultured for 24 h. The supernatants were collected for IL-13 ELISA assay.

The hepatoprotective effect of α-GalCer preconditioning was associated with a decreased neutrophil accumulation in the liver.

We performed immunohistochemical staining to evaluate the alteration of neutrophilic accumulation in the liver following α-GalCer preconditioning. After 1 h of ischemia and 6 h of reperfusion, a remarkable number of neutrophils were observed in the liver, particularly within the necrotic areas (Fig. 6A). Pretreatment with α-GalCer markedly decreased the presence of neutrophils in the liver of mice underwent IR injury. In contrast, mice received α-GalCer treatment combined with anti-IL-13 Ab or SH58261 exhibited significant neutrophil infiltration following liver IR injury (Fig. 6A). To more accurately quantify the extent of neutrophil accumulation in the liver, we carried out flow cytometric analysis of the liver-derived MNCs by staining the cells with neutrophil surface makers CD11b and Gr-1. There was an approximately twofold decrease in neutrophil infiltration in the ischemic liver of mice pretreated with α-GalCer, and this reduction in neutrophil accumulation was blunted substantially by the combined treatment of α-GalCer with anti-IL-13 Ab or SH58261 and reverted to a similar level to that of the vehicle-treated control group (Fig. 6, B and C). These data indicate that α-GalCer pretreatment attenuated hepatic IR-induced neutrophil accumulation in the liver, which was presumably mediated by IL-13 and adenosine A2A receptor.

Fig. 6.

Alterations of neutrophil accumulation in the liver. A: mice were pretreated with vehicle, α-GalCer, α-GalCer plus anti-IL-13 Ab, or α-GalCer plus SH58261, followed by 1 h of hepatic ischemia and 6 h of reperfusion. Immunostaining of the liver sections was performed by using a specific anti-mouse neutrophil antibody. Neutrophils are presented by a dark brown stain. Images are representative of 3 separate experiments with similar results. B: mice were treated as described above. Hepatic MNCs were isolated from the liver and were stained with neutrophil surface maker CD11b and Gr-1 for FACS analysis. The numbers shown in the figure represent the percentage of CD11b+/Gr-1+ neutrophils in the hepatic MNC population. Shown are representative results of 3 replicate experiments. C: summarized data of the numbers of neutrophils in the liver. The data were based on 3 independent experiments of FACS analysis. (*P < 0.05 vs. the vehicle-treated control group; **P < 0.05 vs. the group receiving α-GalCer treatment only).

DISCUSSION

A transient episode of ischemia-reperfusion to the liver, a process known as IPC, protects against liver tissue damage induced by subsequent more sustained ischemic insults (6, 18, 26, 29). Similarly to the prolonged IR, IPC generates a variety of metabolites and ligands including reactive oxygen species and TNF-α, which are well documented as important mediators of liver damage following prolonged IR (6, 18, 28, 29, 33). Interestingly, the protective effect achieved by IPC on the subsequent IR injury can be mimicked by treatments with H2O2 or a low dose of TNF-α (28, 33). These observations support the notion that preexposure of the liver with a sublethal stress that is usually involved in the pathogenesis of IR injury triggers endogenous adaptive reactions in the liver to increase tissue resistance to ischemia-reperfusion injury. This concept provides a rationale for developing new strategies of hepatic preconditioning based on the underlying mechanisms of liver IR injury.

NKT cells are a group of unique lymphocytes expressing a semi-invariant T cell receptor α-chain that react with glycolipid antigen presented by CD1d (2, 10, 37). Activation of NKT cells has a profound impact on both innate and acquired immunity and has been implicated in a number of disease conditions such as autoimmune disorders, tumors, allergy, and various infectious diseases (2, 10, 37). NKT cells are far more abundant in the liver than in any other organs in mouse (10). These cells constitutively express mRNA encoding a variety of cytokines, which allows their rapid response to stimuli and vigorous cytokine production upon activation (37). Notably, among the various cytokines generated by activated NKT cells, TNF-α, IFN-γ, and IL-6 are well-known mediators important for liver IR injury. Indeed, recent evidence is emerging to support the direct involvement of NKT cells in the initiation of liver IR injury (19, 31). This novel insight together with the aforementioned distinct properties of NKT cells make this cell population an attractive target for developing new ways of hepatic preconditioning to ameliorate liver IR injury.

In the present study, we performed hepatic preconditioning in mice with a suboptimal dose of α-GalCer 1 h prior to hepatic ischemia and asked whether NKT cell preactivation by α-GalCer could provide beneficial preconditioning effects to protect against liver IR injury. We observed a significant reduced hepatic IR injury in mice preexposed with α-GalCer challenge. The protective effect was specifically mediated by a CD1d-dependent NKT cell activation, as evidenced by the inability of α-GalCer preconditioning to protect IR injury in mice pretreated with a CD1d blocking antibody. In agreement with our findings, activation of NKT cells by α-GalCer has been shown to have protective effects in several animal models of autoimmune and inflammatory conditions such as rheumatoid arthritis, experimental autoimmune encephalomyelitis (EAE), and autoimmune uveitis (11, 38). Interestingly, some contradictory results have also been reported regarding the disease-modulating effect of α-GalCer administration. For example, preimmunization of α-GalCer with myelin basic protein (MBP) peptides in an EAE model resulted in disease protection, whereas coimmunization of α-GalCer with MBP exacerbated the disease (38). Administration of α-GalCer at a later stage of an animal model of systemic lupus sythematosus (SLE) exacerbated the disease compared with a disease protective effect when α-GalCer was given at an early stage (38). Mice were protected from the endotoxin shock when they were treated with α-GalCer within 2 h before LPS challenge, whereas no protection was achieved if mice received α-GalCer treatment 6 to 18 h before LPS challenge (32). During our pilot experiments aimed at determining an appropriate protocol for hepatic preconditioning with α-GalCer, we found that the timing schedule of α-GalCer treatment 1 h prior to hepatic ischemia was of crucial importance to achieve a protective effect on the following liver IR injury, because we observed a significantly worsened liver IR injury when α-GalCer was administered 6 h prior to liver ischemia. Therefore, our observations are consistent with the notion that the effects of α-GalCer-mediated NKT cell activation on disease progression are influenced by a variety of parameters particularly the timing of α-GalCer treatment.

The response of NKT cells to α-GalCer is characterized by a rapid production of a large number of cytokines, including IFN-γ, TNF-α, IL-2, IL-4, IL-10, IL-13, and TGF-β (3, 10, 37). We sought to determine which specific cytokines were responsible for the protective effect of α-GalCer preconditioning on hepatic IR injury. The proinflammatory cytokine TNF-α has been well characterized as a critical mediator of liver IR injury, and liver IR injury-associated TNF-α production can be suppressed by many protective interventions including IPC (6, 34). Furthermore, a transient increase in TNF-α levels elaborated by the brief episode of IR seems to contribute to the hepatoprotective effect of IPC (33). In the present study, however, we found that TNF-α does not appear to play a role in α-GalCer preconditioning since neutralization of TNF-α did not affect α-GalCer-mediated preconditioning effect even though we found that α-GalCer preconditioning was correlated with a rapid increase in serum TNF-α levels. The Th1 cytokine IFN-γ is another important cytokine involved in hepatic IR injury, and it was recently shown that NKT cell-derived IFN-γ was particularly important for IR-induced liver damage (5, 19). Consistent with previous studies, we found that α-GalCer treatment resulted in a robust elevation of serum IFN-γ levels. Nevertheless, neutralization of IFN-γ failed to influence the preconditioning effect of α-GalCer either. These data indicate that neither TNF-α nor IFN-γ was responsible for α-GalCer-mediated preconditioning effect, suggesting the underlying mechanisms involved in α-GalCer preconditioning might be different from those of other methods of hepatic preconditioning, including the conventional ischemic preconditioning.

We next explored the possibility that α-GalCer-mediated preconditioning effect might result from the direct hepatocellular protection afforded by certain factors elaborated by the activated NKT cells. Among the various cytokines released by NKT cells upon α-GalCer stimulation, the Th2 cytokine IL-13 seems to be a plausible candidate that might be responsible for the protective effect of α-GalCer preconditioning. IL-13 was previously shown to have pronounced anti-inflammatory effects in several animal models of inflammatory diseases (23, 39). The anti-inflammatory effects of IL-13 have been linked with an inhibition of nuclear factor-κB and downregulating the production of proinflammatory cytokines such as TNF-α, IL-1β, and IL-6 (2123). In this study, we observed a rapid and sustained increase in serum IL-13 levels after NKT cell activation by α-GalCer, and neutralization of IL-13 strikingly abolished the hepatoprotective effect of α-GalCer preconditioning. Therefore, NKT cell-derived IL-13 plays an essential role in α-GalCer preconditioning, presumably by suppressing the inflammatory events associated with liver IR. This finding is in agreement with several recent reports showing an important role of IL-13 in regulating the inflammatory responses during hepatic IR. The work by the Lentsch group showed that the mouse liver displayed much greater degree of injury induced by IR in IL-13-deficient mice (16), whereas exogenous administration of IL-13 protected liver IR injury through a STAT6-dependent mechanism (17, 41).

Neutrophil infiltration into the liver is considered to be a significant cellular event that contributes to the maximal liver injury during the later phase of hepatic IR injury (14, 15, 20). Neutrophil-induced damage to hepatocytes is mediated by the release of reactive oxygen species and proteases by the activated neutrophils. In the present study, we found that the hepatoprotective effect of α-GalCer preconditioning was associated with a decreased neutrophil accumulation in the liver, which was diminished by the cotreatment with an anti-IL-13 Ab or the adenosine A2AR antagonist SH58261. We also found that α-GalCer preconditioning did not significantly alter liver injury following 1 h of reperfusion (data not shown), an earlier phase in which neutrophils are not involved in the injury process. These observations support a notion that hepatic preconditioning with α-GalCer causes an increased IL-13 release and enhanced adenosine signal via A2AR, which in turn leads to a suppressed neutrophil accumulation in the liver and thereby attenuates hepatic damage following liver IR.

Additionally, we showed that α-GalCer treatment was associated with an increased adenosine A2AR expression in the liver and the protective effect of α-GalCer pretreatment on liver IR injury was diminished by an adenosine A2AR antagonist, implicating the involvement of adenosine and A2AR in α-GalCer preconditioning. Adenosine is an endogenous compound produced by ATP metabolism and is well documented as an important factor mediating the hepatoprotective effect of hepatic ischemic preconditioning by specifically activating the A2AR (6, 34). Administration of adenosine or selective A2AR agonists enhances liver tolerance to IR injury (8, 12). Interestingly, a recent study suggested that NKT cells express functional A2AR and administration of a selective A2AR agonist inhibited NKT cell activation and resulted in reduced hepatic IR injury (19). The study we present here provides evidence that adenosine and its receptor A2AR also play a role in the protection of liver IR injury by hepatic preconditioning with α-GalCer. We also found that NKT cells express A2AR, but the agonist CGS21680 treatment did not significantly alter NKT-mediated IL-13 production. Furthermore, we showed that preconditioning with α-GalCer and CGS21680 did not result in further additive protection compared with preconditioning with α-GalCer alone and that administration of the A2AR agonist CGS21680 reversed the counteracting effect of the IL-13 neutralizing antibody on α-GalCer preconditioning. These observations raise the possibility that adenosine A2AR activation might act downstream of IL-13 to relay the hepatoprotective effect of α-GalCer preconditioning. In a recent study, increased adenosine levels were detected in lung tissues of IL-13 transgenic mice, and suppression of adenosine levels by adenosine deaminase enzyme therapy substantially diminished the parenchymal inflammation in the lung tissues of IL-13 transgenic mice (4). This intriguing study therefore demonstrated a direct link between IL-13 and adenosine in regulating lung inflammation. It would be of interest in future studies to investigate whether such a link is also present in mice subjected to hepatic preconditioning with α-GalCer.

In conclusion, our data support that administration of α- GalCer 1 h prior to liver ischemia confers protection against hepatic IR injury in a mouse model. The protective effects depend on a mechanism involving an enhanced IL-13 production and adenosine A2AR signaling, which seems to in turn lead to a suppressed neutrophil accumulation in the liver and thereby attenuates hepatic damage following liver IR. Our study provides useful insights for the design of new hepatic preconditioning strategies to protect liver from IR injury based on treatments with specific NKT ligands.

GRANTS

This work has been supported by the National Institutes of Health research grant R01-GM-50441.

REFERENCES

View Abstract