Hepatocyte signaling through CXC chemokine receptor-2 is detrimental to liver recovery after ischemia/reperfusion in mice†
Article first published online: 7 JUL 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 48, Issue 4, pages 1213–1223, October 2008
How to Cite
Kuboki, S., Shin, T., Huber, N., Eismann, T., Galloway, E., Schuster, R., Blanchard, J., Edwards, M. J. and Lentsch, A. B. (2008), Hepatocyte signaling through CXC chemokine receptor-2 is detrimental to liver recovery after ischemia/reperfusion in mice. Hepatology, 48: 1213–1223. doi: 10.1002/hep.22471
Potential conflict of interest: Nothing to report.
- Issue published online: 26 SEP 2008
- Article first published online: 7 JUL 2008
- Accepted manuscript online: 7 JUL 2008 12:00AM EST
- Manuscript Accepted: 6 JUN 2008
- Manuscript Received: 20 MAR 2008
- National Institutes of Health. Grant Numbers: AG025881, DK56029
CXC chemokines and their receptor, CXC chemokine receptor-2 (CXCR2), are important components of the hepatic inflammatory response to ischemia/reperfusion (I/R). However, direct effects of CXC chemokines on hepatocytes during this response have not been studied. Wild-type and CXCR2−/− mice were subjected to 90 minutes of partial hepatic ischemia followed by up to 96 hours of reperfusion. CXCR2−/− mice had significantly less liver injury at all reperfusion times compared with wild-type mice. Early neutrophil recruitment (12 hours) was diminished in CXCR2−/− mice, but within 24 hours it was the same as that of wild-type mice. Hepatocyte proliferation and regeneration was accelerated in CXCR2−/− mice compared with wild-type mice. These effects were associated with increased activation of nuclear factor κB and signal transducers and activators of transcription-3, despite there being no difference in the expression of proliferative factors such as tumor necrosis factor α, interleukin-6, and hepatocyte growth factor. To establish whether the accelerated proliferation and regeneration observed in CXCR2−/− mice was due to effects on hepatocytes rather than just a generalized decrease in acute inflammatory injury, mice were treated with the CXCR2 antagonist, SB225002, after neutrophil recruitment and injury were maximal (24 hours after reperfusion). SB225002 treatment increased hepatocyte proliferation and regeneration in a manner identical to that observed in CXCR2−/− mice. Treatment of primary wild-type hepatocytes with macrophage inflammatory protein-2 revealed that low concentrations protected against cell death, whereas high concentrations induced cell death. These effects were absent in hepatocytes from CXCR2−/− mice. Conclusion: Our data suggest that hepatocyte CXCR2 regulates proliferation and regeneration after I/R injury and reveal important differences in the role of this receptor in liver regeneration and repair induced under different conditions that may be related to ligand concentration. (HEPATOLOGY 2008.)
Ischemia/reperfusion (I/R) of the liver is a primary complication of liver resection surgery, transplantation, and trauma.1 Extended periods of hepatic ischemia and subsequent reperfusion can lead to liver injury and dysfunction through the initiation of a biphasic inflammatory response.2 The initial phase of this response is characterized by activation of Kupffer cells and their subsequent production and release of reactive oxygen species, leading to mild hepatocellular injury.3, 4 The later phase of injury is initiated by proinflammatory mediators released by activated Kupffer cells and hepatocytes. These mediators include the cytokines interleukin-12 (IL-12), tumor necrosis factor α (TNF-α) and IL-1.5–8 The expression of these cytokines induces the production of CXC chemokines and endothelial cell adhesion molecules that facilitate the adhesion and transmigration of neutrophils from the vascular space into the hepatic parenchyma.9–12 These activated neutrophils then directly injure hepatocytes and vascular endothelial cells through their release of oxidants and proteases.13
Previous studies have demonstrated that CXC chemokines mediate neutrophil infiltration during I/R injury of several organs, including the liver.10, 11 The receptor responsible for these effects is CXC chemokine receptor-2 (CXCR2). Blockade of either individual CXCR2 ligands10, 11 or CXCR2 itself14 prevents neutrophil infiltration and attenuates hepatic I/R injury. In addition, other studies have suggested that CXC chemokines may have direct effects on hepatocytes. CXCR2 has been shown to be expressed by hepatocytes,15 and in models of partial hepatectomy or acetaminophen toxicity, CXCR2 ligands were shown to promote hepatocyte proliferation and liver regeneration.16–18 In contrast, other studies have suggested that CXCR2 ligands are directly hepatotoxic.19, 20
The manner in which the liver recovers and regenerates after a severe ischemic insult is not well characterized. Previous studies from our laboratory have shown that after prolonged (90-minute) ischemia in mice, hepatocyte proliferation and cell cycle progression are initiated within 24–48 hours of reperfusion.21 Whether CXC chemokine signaling through CXCR2 regulates this reparative response is unknown. Therefore, we sought to determine whether deletion or antagonism of CXCR2 had any effect on the recovery from liver damage during hepatic I/R injury.
Materials and Methods
Hepatic I/R Injury Model.
Male wild-type (Balb/c) and CXCR2−/− mice (on a Balb/c background) from The Jackson Laboratory (Bar Harbor, ME) weighing 22–28 g were used in the experiments. This project was approved by the University of Cincinnati Animal Care and Use Committee and was in compliance with National Institutes of Health guidelines. The animals underwent either sham operation or I/R. Partial hepatic ischemia was induced as described.11 Sham control mice underwent the same protocol without vascular occlusion. Some mice were injected intraperitoneally with 3% dimethylsulfoxide solution or 4 mg/kg SB225002 (N-(2-hydroxy-4-nitrophenyl)-N′-(2-bromophenyl)urea) (Calbiochem, San Diego, CA), a nonpeptide antagonist of CXCR2,22 dissolved in dimethylsulfoxide at 24-hour intervals starting after 24 hours of reperfusion.
Electrophoretic Mobility Shift Assay.
Nuclear extracts of liver tissue were prepared by the method of Deryckere and Gannon23 and were analyzed via electrophoretic mobility shift assay (EMSA). Briefly, double-stranded consensus oligonucleotides to nuclear factor κB (NF-κB) (Promega, Madison, WI) were end-labeled with γ[32P]ATP (3,000 Ci/mmol at 10 mCi/mL) (Amersham, Arlington Heights, IL). Binding reactions (total volume: 15 μL) containing equal amounts of nuclear protein extract (20 μg) and 35 fmols (≈50,000 counts per minute, Cherenkov counting) of oligonucleotide and were incubated at room temperature for 30 minutes. Binding reaction products were separated in a 4% polyacrylamide gel and analyzed via autoradiography.
Western Blot Analysis.
Liver samples were homogenized in lysis buffer (10 mM 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid [pH 7.9], 150 mM NaCl, 1 mM ethylene diamine tetraacetic acid, 0.6% Nonidet P-40, 0.5 mM phenylmethylsulfonyl fluoride, 1 μg/mL leupeptin, 1 μg/mL aprotonin, 10 μg/mL soybean trypsin inhibitor, 1 μg/mL pepstatin) on ice. Homogenates were sonicated and centrifuged at 10,000 rpm to remove cellular debris. Protein concentrations of each sample were determined. Samples containing equal amounts of protein in equal volumes of sample buffer were separated in a denaturing 10% polyacrylimide gel and transferred to a 0.1-μm pore nitrocellulose membrane. Nonspecific binding sites were blocked with Tris-buffered saline (40 mM Tris [pH 7.6], 300 mM NaCl) containing 5% nonfat dry milk for 1 hour at room temperature. Membranes were then incubated with antibodies to phosphorylated signal transducers and activators of transcription-3 (STAT3) or β-actin (Santa Cruz Biotechnology, Santa Cruz, CA) in Tris-buffered saline with 0.1% Tween 20. Membranes were washed and incubated with secondary antibodies conjugated to horseradish peroxidase. Immunoreactive proteins were detected via enhanced chemiluminescence.
Blood and Tissue Analysis.
Blood was obtained via cardiac puncture for analysis of serum alanine aminotransferase (ALT) as an index of hepatocellular injury. Measurements of serum ALT were made using a diagnosis kit via bioassay (Wiener Laboratories, Rosario, Argentina). Serum and tissue levels of TNF-α, IL-6, macrophage inflammatory protein-2 (MIP-2) and keratinocyte chemokine (KC) were measured by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instructions (R&D Systems, Minneapolis, MN). Liver neutrophil accumulation was determined via myeloperoxidase (MPO) content and chloroacetate esterase staining. MPO content was assessed using methods described elsewhere.24 Chloroacetate esterase staining was performed on paraffin-embedded liver tissue using the Naphthol-ASD Chloroacetate Esterase Kit (Sigma, St. Louis, MO) according to the manufacturer's instructions. Stained liver sections were scored as follows: 0, none; 1+, faint to moderate; 2+, moderate to strong; 3+, strong; 4+, very strong. For histological examination, liver tissues were fixed in 10% neutral-buffered formalin, processed, and embedded in paraffin prior to staining with hematoxylin-eosin.
Proliferating Cell Nuclear Antigen Staining.
Immunohistochemical staining for proliferating cell nuclear antigen (PCNA) was performed on paraffin-embedded liver tissue with anti-PCNA antibody using a DakoCytomation ARK kit (Dako, Copenhagen, Denmark). Briefly, a three-step peroxidase method was performed according to the manufacturer's instructions. PC-10 monoclonal antibody (Santa Cruz Biotechnology) was used at a dilution of 1:50 for 15 minutes at room temperature. The sections were counterstained with hematoxylin. Evaluation of PC-10 immunostaining was performed based on the percentage of positive nuclei of 400-600 hepatocytes from 4-6 highest positive fields at high power (×400), and was expressed as PCNA labeling index.
Primary Hepatocyte Studies.
Hepatocytes from wild-type and CXCR2−/− mice were isolated as described elsewhere.25 Isolated hepatocytes in Williams media were distributed onto 96-well flat-bottomed plates (TTP, Trasadingen, Switzerland) at a concentration of 1.5 × 104 cells/200 μL/well, and incubated overnight to allow cell adherence. Media was changed and cells were incubated for 24 hours in Williams medium with or without recombinant murine MIP-2 or KC (Peprotech, Rocky Hill, NJ) at 1–10,000 ng/mL. Cytotoxicity was determined by lactate dehydrogenase release according to the manufacturer's instructions (Roche, Mannheim, Germany), and percent cell death was calculated. Hepatocyte apoptosis was determined by analyzing cell lysates for active caspase-3 activity via ELISA (R&D Systems, Minneapolis, MN). Hepatocyte proliferation was determined via DNA incorporation of 5-bromo-2′-deoxyuridine using the Biotrak cell proliferation ELISA system (GE Healthcare, Buckingham, UK). Data are expressed as a ratio compared with 0 ng/mL. All experiments were repeated at least twice to confirm these results.
All data are expressed as the mean ± standard error of the mean (SEM). Data were analyzed with a one-way analysis of variance with subsequent Student-Newman-Keuls test. Differences were considered significant when P < 0.05.
Signaling Through CXCR2 Auguments Liver Injury After I/R.
Because we and others have established the importance of CXCR2 ligands for acute neutrophil recruitment and injury after I/R in the liver,10, 11 we first assessed the response of wild-type and CXCR2−/− mice to ischemia with prolonged periods of reperfusion. As expected, after 12 hours of reperfusion, there was significant neutrophil accumulation in wild-type mice, but not CXCR2−/− mice (Fig. 1A). Interestingly, after 24 or 48 hours of reperfusion, there was no difference in neutrophil accumulation. After 96 hours of reperfusion, there was significantly lower MPO content in the livers from CXCR2−/− mice compared with wild-type mice, but neutrophil scores were not different between these groups (Fig. 1A).
More liver injury was observed in wild-type mice at virtually every time point measured. After ischemia and 12 or 24 hours of reperfusion, CXCR2−/− mice had significantly less liver injury than wild-type mice as determined by serum ALT levels (Fig. 1B). After 48 hours of reperfusion, serum ALT levels in CXCR2−/− mice returned to baseline values, whereas levels in wild-type mice remained significantly elevated (Fig. 1B). By 96 hours of reperfusion, serum ALT levels were normal in both groups. Histological examination confirmed our biochemical data. Both wild-type and CXCR2−/− mice undergoing sham operation had normal liver architecture (Fig. 1C). After 12 hours of reperfusion, both wild-type and CXCR2−/− mice had large areas of liver necrosis, but CXCR2−/− mice had less hepatocellular necrosis and little evidence of neutrophil accumulation (Fig. 1C). However, after 24 and 48 hours of reperfusion, neutrophil accumulation in wild-type and CXCR2−/− mice was similar, but there was far less hepatocellular necrosis in CXCR2−/− mice (Fig. 1C). After 96 hours of reperfusion, CXCR2−/− mice had nearly normal liver structure, except for a few small necrotic areas infiltrated with neutrophils (Fig. 1C). In contrast, wild-type mice still had large necrotic areas with severe neutrophil infiltration.
CXCR2 Signaling Does Not Alter Cytokine Expression.
The acute inflammatory response induced by I/R in the liver is driven primarily by TNF-α and is regulated by IL-6.6, 26 To determine if lack of CXCR2 expression altered either of these cytokines and therefore the injury response, we assessed serum levels of TNF-α and IL-6 over the time course of injury and recovery. No differences were observed in TNF-α or IL-6 between wild-type and CXCR2−/− mice at any time point (Fig. 2A). We next evaluated the expression of the two primary CXCR2 ligands in mice, MIP-2 and KC. Both chemokines were significantly increased in the serum and livers of CXCR2−/− mice compared with wild-type mice throughout the time course of injury and recovery (Fig. 2B,C). This suggests that signaling via CXCR2 provides negative feedback for chemokine production.
Enhanced Hepatocyte Proliferation in CXCR2−/− Mice After I/R.
Because histopathologic analysis suggested that CXCR2−/− mice recovered faster from I/R injury, we evaluated hepatocyte proliferation after I/R. Liver sections were stained for PCNA, which is regarded as a marker of S phase entry. Sham-operated wild-type or CXCR2−/− mice had almost no PCNA-positive hepatocytes (Fig. 3A). After 24 or 48 hours of reperfusion (48-hour data not shown), livers from wild-type mice had very few PCNA-positive hepatocytes, whereas significantly more PCNA-positive hepatocytes were detected in livers from CXCR2−/− mice (Fig. 3A). After 96 hours of reperfusion, PCNA staining was increased in wild-type mice but was much more abundant in CXCR2−/− mice (Fig. 3A). Fig. 3B shows quantified PCNA data. The results suggest that the proliferative response after I/R injury occurs much earlier in CXCR2−/− mice than in wild-type mice. Because soluble factors such as hepatocyte growth factor (HGF) are also important stimuli for hepatocyte proliferation, we measured HGF protein levels in the liver. No significant difference was observed in HGF expression between wild-type mice and CXCR2−/− mice (data not shown).
Activation of NF-κB and STAT3 Is Increased in CXCR2−/− Mice After I/R.
Because NF-κB and STAT3 are known to promote hepatocyte proliferation,27–29 we assessed the activation of these two factors. In wild-type mice, NF-κB activation (as measured by EMSA) was significantly decreased after 12 hours of reperfusion and recovered to baseline after 96 hours of reperfusion. Interestingly, NF-κB activation was significantly increased in CXCR2−/− mice compared with wild-type mice at all reperfusion time points measured (Fig. 4A). STAT3 activation, as determined by STAT3 phosphorylation, was markedly increased within 12 hours of reperfusion and was maximal after 24 hours in wild-type mice (Fig. 4B). STAT3 activation was significantly higher in CXCR2−/− mice compared with wild-type mice after 12, 24, and 48 hours of reperfusion (Fig. 4B). After 96 hours of reperfusion, there was no difference in STAT3 activation between wild-type and CXCR2−/− mice.
Antagonism of CXCR2 Augments Liver Proliferation and Recovery Despite Fulminant Inflammatory Injury.
Our experiments show that CXCR2−/− mice have less neutrophil accumulation and liver injury than wild-type mice (Fig. 1). These experiments do not eliminate the possibility that the accelerated regeneration and recovery observed in CXCR2−/− mice might be due to the reduced acute neutrophil accumulation and injury in these mice. Therefore, we examined whether the reduced liver injury and accelerated liver regeneration in CXCR2−/− mice during the late phase of reperfusion was a result of reduced neutrophil accumulation and injury or elimination of a direct hepatoxic effect of CXC chemokines. To do this, wild-type mice were injected intraperitoneally with 3% dimethylsulfoxide (control) or 4 mg/kg SB225002, a nonpeptide antagonist of CXCR2, 24 hours after reperfusion and every 24 hours thereafter. Because neutrophil accumulation in the liver peaks by 24 hours, administration of SB225002 at this time does not alter acute neutrophil recruitment. Treatment with SB225002 had no effect on neutrophil accumulation after 48 or 96 hours of reperfusion (Fig. 5A). However, SB225002 significantly reduced liver injury after 48 hours of reperfusion, as measured with serum ALT levels (Fig. 5B). After 96 hours of reperfusion, ALT levels in both groups returned to baseline values. Histopathologic analysis revealed that control mice undergoing sham operation had normal hepatic architecture (Fig. 5C). However, control mice after 48 and 96 hours of reperfusion had severe hepatocellular necrosis and neutrophilic infiltrates (Fig. 5C). After 48 or 96 hours of reperfusion, livers from mice treated with SB225002 showed high levels of neutrophil infiltration but reduced areas of necrosis (Fig. 5C). These results suggested that CXC chemokines have direct hepatotoxic effects that are independent of neutrophil infiltration.
Blockade of CXCR2 Promotes Hepatocyte Proliferation After I/R in Association with Increased NF-κB and STAT3 Activation.
We next assessed whether treatment with SB225002 had any effect on hepatocyte proliferation after I/R. Immunohistochemical staining for PCNA revealed that sham-operated control mice had almost no PCNA-positive hepatocytes (Fig. 6A). After 48 hours of reperfusion, PCNA staining was increased in both control and SB225002-treated mice, but no differences were seen between the groups (Fig. 6A). In contrast, PCNA staining after 96 hours of reperfusion was significantly greater in mice treated with SB225002 compared with control mice (Fig. 6A). Quantification of PCNA data revealed a significant increase in PCNA staining after 96 hours of reperfusion (Fig. 6B).
Analysis of NF-κB and STAT3 activation revealed that treatment with SB225002 had similar effects as a knockout of CXCR2. EMSA analysis of liver nuclear extracts demonstrated that treatment with SB225002 increased activation of NF-κB compared with the vehicle control after 48 and 96 hours of reperfusion (Fig. 7A). Western blot analysis revealed that SB225002 also increased STAT3 activation 48 hours after reperfusion (Fig. 7B). After 96 hours of reperfusion, STAT3 phosphorylation was similar between the two groups.
CXCR2 Signaling Can Be Hepatoprotective or Hepatotoxic Depending on Ligand Concentration.
Previous studies have shown that treatment of primary hepatocytes with the CXCR2 ligand, MIP-2, induces proliferation.16 Given our current findings suggesting that signaling via CXCR2 is harmful after I/R, we examined whether signaling through CXCR2 may directly affect hepatocyte cell death. Primary mouse hepatocytes were cultured in the presence or absence of 1–10,000 ng/mL of MIP-2, and hepatotoxicity was determined via release of lactate dehydrogenase. As shown in Fig. 8A, low concentrations of MIP-2 (<100 ng/mL) protected wild-type hepatocytes against cell death, whereas high concentrations of MIP-2 (>100 ng/mL) induced cell death. These effects were mediated by CXCR2, because hepatocytes from CXCR2−/− mice showed no significant differences at any dose of MIP-2 (Fig. 8B). To determine if MIP-2 may also affect hepatocyte apoptosis, active caspase-3 activity was assessed after treatment with 10 or 10,000 ng/mL of MIP-2. Only the high dose of MIP-2 (10,000 ng/mL) induced a significant increase in apoptosis (Fig. 8C).
To evaluate the effects of MIP-2 on hepatocyte proliferation, DNA incorporation assays were conducted. MIP-2 stimulated hepatocyte proliferation at virtually every dose tested (Fig. 9A). These effects appeared to be mediated specifically by CXCR2 as no increases in proliferation were observed in hepatocytes from CXCR2−/− mice (Fig. 9B).
Finally, because several ligands for CXCR2 may have overlapping function, we assessed the effects of combined stimulation of wild-type hepatocytes with MIP-2 and KC. There appeared to be synergistic effects, because cell death was significantly reduced by treatment of hepatocytes with 5 ng/mL each of MIP-2 and KC (Fig. 10). The reduction in cell death was nearly twice that observed with 10 ng/mL of only MIP-2 (Fig. 8A). Likewise, treatment of hepatocytes with 5,000 ng/mL each of MIP-2 and KC resulted in nearly twice the amount of cell death as 10,000 ng/mL of MIP-2 (Fig. 10).
The present study is the first to examine the effects of CXCR2 ligands on liver recovery after I/R injury. The most significant finding of this report is the demonstration that signaling through CXCR2 is detrimental to hepatocyte proliferation and regeneration in this model. This is in stark contrast to studies in models of partial hepatectomy or acetaminophen toxicity in which CXCR2 was found to promote hepatocyte proliferation and liver regeneration.16–18 While the precise mechanism of the divergent effects of CXCR2 ligands on liver regeneration between I/R and partial hepatectomy or acetaminophen models is unclear, it is plausible that the differences are related to the amount of CXC chemokines produced during these insults. Our laboratory has found that levels of MIP-2 and KC are increased three-fold and five-fold, respectively, 48 hours after 70% hepatectomy (S. Kuboki and A. B. Lentsch, unpublished data), which is consistent with previous studies in rats.16 In models of acetaminophen toxicity, MIP-2 and KC levels increase no more than 2.5-fold 48 hours after challenge.17 In contrast, our data show that after ischemia and 48 hours of reperfusion MIP-2 and KC levels are 50-fold and 25-fold higher, respectively, than sham-operated controls. Adenoviral-mediated overexpression of KC in the liver (>100-fold) results in massive hepatocellular necrosis within 48 hours.20 Collectively, these studies suggest that moderate increases in CXCR2 ligands, as occurs after partial hepatectomy or acetaminophen toxicity, may promote liver regeneration, whereas much larger increases in expression of CXCR2 ligands, as occurs after I/R injury, may be hepatotoxic and/or oppose hepatocyte proliferation and regeneration. Our in vitro data support this concept. We found that low concentrations of the CXCR2 ligand, MIP-2, had hepatoprotective effects in hepatocytes from wild-type mice, whereas high concentrations induced significant cytotoxicity. These divergent effects were absent in hepatocytes isolated from CXCR2−/− mice, indicating that CXCR2 may mediate both protective and cytotoxic signaling. Finally, we found that when hepatocytes were treated with both MIP-2 and KC, the effects were far greater than just MIP-2 at equal concentrations, suggesting that multiple CXCR2 ligands may induce additive or synergistic effects through the receptor.
Alternatively, the divergent effects of CXCR2 signaling on liver regeneration between partial hepatectomy and I/R may be related to the impact of the insult on the hepatocyte. After partial hepatectomy, the remaining liver comprises almost entirely normal hepatocytes. Conversely, after I/R all hepatocytes are exposed to an injurious stimulus, and those that survive are highly stressed. It has been reported that CXC chemokines greatly enhance acute liver injury, whereas their hepatotoxic effects were less in healthy hepatocytes.19 Therefore, CXC chemokines might be mitogenic to normal healthy hepatocytes, but hepatotoxic to injured hepatocytes. Further study is needed to reveal the differences in CXC chemokine behavior under different conditions.
Another interesting finding of our study is the nature in which CXCR2 regulates neutrophil recruitment following I/R. Consistent with previous studies that used neutralizing antibodies to CXC chemokines,10, 11 the lack of CXCR2 blocked early neutrophil recruitment (<12 hours). Surprisingly, neutrophil recruitment was similar between wild-type and CXCR2−/− mice after 24 and 48 hours of reperfusion. This suggests that CXCR2-mediated neutrophil recruitment is relevant only during the acute inflammatory phase of injury and not for the more chronic reparative phase. These studies also confirm earlier studies regarding the contributions of neutrophils to acute liver injury after I/R.4, 30 CXCR2−/− mice had significantly less liver injury than wild-type mice. This finding represents a confounding variable in our studies with CXCR2−/− mice. Could it be that the improved recovery observed in CXCR2−/− mice was simply due to reduced initial injury? Our studies with the CXCR2 antagonist SB225002 provide strong evidence against this possibility. In wild-type mice, we administered SB225002 after 24 hours of reperfusion, a time at which neutrophil recruitment to the liver was already maximal. Treatment with SB225002 not only reduced hepatocellular necrosis, but accelerated hepatocyte proliferation and ultimate regenerative repopulation of the liver. Combined with our studies in CXCR2−/− mice and our in vitro studies with primary hepatocytes, these data demonstrate that signaling through CXCR2 in hepatocytes after I/R is directly hepatotoxic and delays proliferation and regeneration.
The precise mechanism by which CXCR2 limits hepatocyte proliferation and regeneration after I/R remains to be elucidated. However, our current data suggest that deletion or blockade of CXCR2 signaling is associated with increased activation of NF-κB and STAT3. Both NF-κB and STAT3 are critical for proper liver regeneration.27–29 Primary stimuli for activation of NF-κB and STAT3 in the liver are TNF-α and IL-6. However, we detected no differences in TNF-α or IL-6 expression between wild-type and CXCR2−/− mice. Expression of HGF, a primary mediator of hepatocyte proliferation, was also similar in wild-type mice and CXCR2−/− mice. Therefore, it seems likely that signals transduced through CXCR2 impinge on upstream components of the NF-κB and STAT3 pathways, such as kinases regulating IκBα and/or STAT3 phosphorylation, to limit activation of these transcription factors. However, it should be noted that the current data do not demonstrate any direct connection between CXCR2 signaling pathways and NF-κB or STAT3. It is also interesting that in CXCR2−/− mice or SB225002-treated mice that STAT3 activation was found to be augmented during late injury and early recovery phases (12–48 hours after reperfusion). This increased activation occurs during the initial period of increased hepatocyte proliferation. This could suggest that STAT3 plays an important role in the early stages of liver recovery and regeneration after I/R.
CXCR1 is another receptor for CXC chemokines that has some functional similarities to CXCR2.31 For many years, it was thought that rodents did not express a homolog of human CXCR1. However, the murine homolog to CXCR1 was recently discovered.32, 33 Although any potential role of CXCR1 in liver injury and repair is currently unknown, it is tempting to speculate as to how this receptor might augment or oppose the function of CXCR2 in hepatocytes. Clearly, this will be an important area of research in the near future.
In summary, the present study demonstrates that CXCR2 is an important regulator of hepatocyte proliferation and regeneration after I/R injury. These data reveal important differences in the role of this receptor in liver regeneration and repair induced by different stimuli and suggest that ligand concentration may be an important factor in these differences. These data highlight the importance of future studies aimed at delineating the mechanism by which CXCR2 signal transduction affects activation of transcription factors such as NF-κB and STAT3. Our data combined with those of previous studies suggest that CXCR2 is a rational therapeutic target that may be relevant to a wide variety of liver insults and diseases.