fax: (36) 52 314989
Liver Biology and Pathobiology
Tissue transglutaminase (TG2) acting as G protein protects hepatocytes against Fas-mediated cell death in mice†
Article first published online: 17 AUG 2005
Copyright © 2005 American Association for the Study of Liver Diseases
Volume 42, Issue 3, pages 578–587, September 2005
How to Cite
Sarang, Z., Molnár, P., Németh, T., Gomba, S., Kardon, T., Melino, G., Cotecchia, S., Fésüs, L. and Szondy, Z. (2005), Tissue transglutaminase (TG2) acting as G protein protects hepatocytes against Fas-mediated cell death in mice. Hepatology, 42: 578–587. doi: 10.1002/hep.20812
Potential conflict of interest: Nothing to report.
- Issue published online: 22 AUG 2005
- Article first published online: 17 AUG 2005
- Manuscript Accepted: 3 JUN 2005
- Manuscript Received: 22 NOV 2004
- Hungarian National Research Fund. Grant Numbers: OTKA T049445, T043083, TS44798, T034191
- Hungarian Ministry of Health. Grant Number: ETT 100/2003
- EU. Grant Number: QLK3-CT-2002-02017
Tissue transglutaminase (TG2) is a protein cross-linking enzyme known to be expressed by hepatocytes and to be induced during the in vivo hepatic apoptosis program. TG2 is also a G protein that mediates intracellular signaling by the alpha-1b-adrenergic receptor (AR) in liver cells. Fas/Fas ligand interaction plays a crucial role in various liver diseases, and administration of agonistic anti-Fas antibodies to mice causes both disseminated endothelial cell apoptosis and fulminant hepatic failure. Here we report that an intraperitoneal dose of anti-Fas antibodies, which is sublethal for wild-type mice, kills all the TG2 knock-out mice within 20 hours. Although TG2−/− thymocytes exposed to anti-Fas antibodies die at the same rate as wild-type mice, TG2−/− hepatocytes show increased sensitivity toward anti-Fas treatment both in vivo and in vitro, with no change in their cell surface expression of Fas, levels of FLIPL (FLICE-inhibitory protein), or the rate of I-κBα degradation, but a decrease in the Bcl-xL expression. We provide evidence that this is the consequence of the impaired AR signaling that normally regulates the levels of Bcl-xL in the liver. In conclusion, our data suggest the involvement of adrenergic signaling pathways in the hepatic regeneration program, in which Fas ligand-induced hepatocyte proliferation with a simultaneous inhibition of the Fas-death pathway plays a determinant role. (HEPATOLOGY 2005.)
Transglutaminases1, 2 are a family of thiol- and Ca2+-dependent acyl transferases that catalyze the formation of a covalent bond between the γ-carboxamide groups of peptide-bound glutamine residues and various primary amines, including the ϵ-amino group of lysine in certain proteins. Eight distinct, enzymatically active transglutaminases have been described.1–3 One of them, TG24, is ubiquitously expressed in mammalian tissues,5 found both extracellularly at the cell surface in association with the extracellular matrix6 and intracellularly in membrane-associated as well as cytosolic forms. The enzyme has been implicated in a variety of cellular processes including signaling,7 cell adhesion, and spreading,8 wound healing,9 and tissue mineralization.10
In addition to its cross-linking activity, TG2 is also a guanosine triphosphate (GTP)-binding protein that mediates intracellular signaling by the alpha-1b-adrenergic receptor in the liver,11 via coupling signals to the phospholipase C-δ112 and to hepatocyte proliferation.13
TG2 is also induced during the in vivo apoptotic program.14–18 Because intracellularly the cross-linking activity of TG2 is inhibited by GTP, Zn2+, and nitric oxide, the accumulation of TG2 inside the cell is not necessarily associated with the activation of its cross-linking activity.19 However, Ca2+-dependent activation of the enzyme in cells undergoing programmed cell death leads to the formation of a detergent-insoluble cross-linked protein scaffold20 that may stabilize the integrity of the dying cells before their clearance by phagocytosis, thus preventing the nonspecific release of harmful intracellular components and consequently inflammatory responses and scar formation in bystander tissues.21
To assess the role of TG2 in programmed cell death, TG2 knock-out mice have been generated.22 When we were studying the thymic apoptosis program in these mice,23 we have noticed that TG2−/− mice are much more susceptible to Fas-mediated killing than the wild-types.
Fas (APO-1, CD95) is a membrane protein of the tumor necrosis factor/nerve growth factor receptor family, which, on interaction with its ligand or on triggering by anti-Fas antibodies, usually acts as an inducer of apoptosis.24 Injection of anti-Fas antibodies into mice leads to a very rapid and massive destruction of hepatocytes and to death within a few hours.25 The liver damage is mediated by bid cleavage and the mitochondrial cell death pathway, leading to activation of procaspase-9,26 and can be inhibited by overexpressing bcl-227 or by deleting bid.28
Although the first reports analyzing the in vivo effects of Fas engagement implied that the death of mice is related entirely to the massive hepatic injury,25 the fact that the death of animals was not prevented if only the livers were protected by selective targeting of bcl-2 suggested that the lethality results also from stimulation of Fas receptors present on other target organs or cells.27 In support of this, a recent report showed that Fas engagement in vivo causes rapid, extensive, and disseminated endothelial cell apoptosis that leads to loss of vascular integrity throughout the body.29
Because TG2−/− mice were more susceptible to Fas-mediated killing than the wild-type mice, the effect of in vivo Fas engagement on the liver of TG2−/− mice was investigated in this study. Although our first suspicion was that incomplete protein crosslinking within TG2−/− apoptotic hepatocytes led to the release of the cell content, resulting in further damage in the surrounding tissues, enhanced further by the impaired phagocytosis,23 the data presented in this paper demonstrate that impaired alpha-1b adrenergic signaling that develops in the absence of TG2 is responsible for the increased apoptotic sensitivity of TG2−/− hepatocytes.
Materials and Methods
Male 4-week-old, TG2+/+ (FVB) and TG2−/−22 or α-1b-adrenergic receptor (AR)+/+(C57BL/6) and AR-deficient mice30 received a single intraperitoneal injection of 1 or 0.3 μ/g body weight Jo2 anti-Fas antibody (PharMingen, Franklin Lakes, NJ), respectively. In some experiments TG2+/+ mice were treated for a week with a daily dose of chloroethylclonidine (Sigma, St. Louis, MO), an AR antagonist,31 to detect changes in the Bcl-xL expression and the in vitro Fas sensitivity of hepatocytes. To test the effect of the α1-adrenergic receptor agonist phenylephrine (Sigma) on Fas sensitivity, TG2+/+ and C57BL/6 mice were injected intraperitoneally with 10 μg/g body weight phenylephrine at 48, 24 and 6 hours before injection of a lethal dose of 2 μg/g and 0.3 μg/g body weight of anti-Fas antibodies, respectively. Study protocols were approved by the Animal Care Committee of the University of Debrecen.
Hepatocyte Culture Experiments.
Hepatocytes were isolated from TG2+/+ and TG2−/− mice by the collagenase perfusion method32 and suspended in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS), 2 mmol/L glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin. After washing, hepatocytes were diluted to a concentration of 106 cells/mL and cultured at 37°C/5% CO2 for 24 hours in the presence or absence of increasing concentrations of the anti-Fas antibodies. The percentage of apoptotic cells was determined by counting the percentage of annexin-V-FITC-positive and propidium iodide–negative cells, using the kit provided by Sigma.
Determination of Cell Surface Expression of Fas on Hepatocytes of Wild-Type and TG2−/− Mice.
Freshly isolated hepatocytes were washed twice with phosphate-buffered saline (PBS) supplemented with 1% bovine serum albumin and 0.1% sodium azide (Buffer A). Next, 106 cells were resuspended in 100 μL Buffer A containing 10 μg anti mouse-FcγR Ab (IgG) and incubated for 15 minutes at room temperature to saturate nonspecific binding sites. For detection of cell surface Fas, cells were stained with a PE-labeled anti-Fas antibody (PharMingen). Fluorescence was analyzed on an Epix Coulter FACScan.
Determination of Bcl-xL or Fas-FADD-bound Caspase-8–Inhibitory ProteinL Expression in the Hepatocytes of Various Mouse Strains.
For detection Bcl-xL or FLIPL protein expression, cells were washed with PBS, lysed, dissolved in Laemmli buffer,33 and analyzed with Western blot technique. For staining the blot, anti-mouse Bcl-xL (Transduction Lab, Lexington, KY) or anti-mouse-FLIP (Alexis, San Diego, CA) antibody was added. Equal loading of protein was demonstrated by probing the membranes with an anti-β-actin antibody (Sigma). Relative densities of protein bands were determined by AlphaImiger 2200 software, Alpha Innotech Corporation, San Leandro, CA).
Determination of Cytochrome c Release, and Cleavage of Procaspase-9 and I-κB in Fas-Treated Livers.
Livers from anti-Fas–treated mice were perfused with ice-cold PBS to wash blood away, homogenized in a buffer 0.1 mol/L Tris pH 8.3, containing 0.25 mol/L sucrose, 0.5 mmol/L EDTA, 100 μg/mL aprotinin, 100 μg/mL leupeptin, and 1 mmol/L phenymethyl-sulphonyl-fluoride, and centrifuged for 5 minutes at 500g to remove cell debris. To obtain a mitochondria-free fraction, the supernatant was further centrifuged at 10,000g for 30 minutes at 4°C. Protein expression was analyzed by Western blot technique. To exclude mitochondrial contamination, cytosolic fractions were first probed by anti–cytochrome c oxydase (Santa Cruz Biotechnology) antibodies. Protein bands were visualized by using anti–cytochrome c (PharMingen), anti–caspase-9 (PharMingen), or anti-I-κB-α (Santa Cruz Biotechnology) antibodies. Equal loading of protein was demonstrated by probing the membranes with an anti–β-actin antibody (Sigma).
Histological Examinations of Liver.
Livers were dissected under anesthesia, both from anti-Fas antibody–injected animals and from saline-treated control littermates. Representative blocks of tissue were formalin-fixed, embedded in paraffin, and stained with hematoxylin-eosin. For transmission electron microscopy (TEM) analysis, livers were fixed with a modified Karnovsky fixative [2% glutaraldehyde + 4% buffered formalin (0.1 mol/L phosphate buffer)] for 2 hours, followed by osmication (2% OsO4 for 2 hours). Tissue samples were then dehydrated and embedded in Araldit resin. Ultrathin sections were “stained” with uranyl acetate and lead nitrate and observed under a JEOL 1010 transmission electron microscope.
Estimation of the In Vivo Rate of Apoptosis After Anti-Fas Injection in the Liver.
Livers from anti-Fas antibody–treated mice were carefully removed, washed with physiological saline, formalin fixed, embedded in paraffin, and stained with the APOPTAG in situ apoptosis detection kit (Intergen Discovery, Gaithersburg, MD), following the manufacturer's instructions.
Determination of Serum Aspartate Aminotransferase and Alanine Aminotransferase Levels.
To determine the extent of liver injury after anti–Fas antibody injection, serum was collected, in the case of TG2−/− or AR−/− mice, at the time of their death, whereas in the case of wild-type mice, at the times at which the knock-out mice died or at 24 hours after treatment. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were determined by routine clinical chemical methods.
Determination of the Rate of Thymocyte Apoptosis In Vivo After Anti-Fas Treatment.
The rate of thymic apoptosis in vivo was evaluated at 20 hours after anti-Fas antibody injection by analyzing changes in the percentage of the apoptosis-sensitive CD4+CD8+ population. Thymocytes were isolated, washed twice, and resuspended in ice-cold PBS before staining either with PE-labeled anti-CD4 and FITC-conjugated anti-CD8 (PharMingen) antibodies. Cell-bound fluorescence was analyzed by using an Epix Coulter FACScan instrument.
Thymocyte Apoptosis In Vitro.
Thymocyte suspensions were prepared by mincing thymus from 4-week-old mice in RPMI 1640 medium supplemented with 5% FCS, 2 mmol/L glutamine, 1 mmol/L Na-pyruvate, 100 units/mL penicillin, and 100 μg/mL streptomycin. After washing, thymocytes were diluted to a concentration of 106 cells/mL and cultured at 37°C / 5% CO2. Apoptosis was induced by addition of 1 μg/mL anti-CD95 monoclonal antibody. After 24 hours, the extent of cell death was determined by 7-aminoactinomycin D uptake.34
Loss of TG2 Does Not Affect the Rate of Thymocyte Apoptosis In Vitro From Exposure to Anti-Fas Antibodies, Although the Rate Observed In Vivo Is Slightly Delayed.
Previous work in our laboratory has shown that intraperitoneal administration of anti-Fas antibodies at a dose of 1 μg/g body weight does not kill wild-type mice, but the rate of Fas-mediated thymocyte apoptosis in vivo can be studied.17 To monitor the rate of thymocyte apoptosis in vivo, the decrease in the percentage of CD4+CD8+ (apoptosis-sensitive) population was determined in the wild-type and in the few knock-out animals that survived 20 hours after anti-Fas injection. As shown in Fig. 1A, after anti-Fas injection, the size of the CD4+CD8+ thymocyte pool decreased in both types of animals, and the disappearance of TG2−/− thymocytes was slightly delayed. However, no significant difference was found in the rate of apoptosis in vitro induced by anti-Fas antibodies determined at 24 hours, which was 56.5 ± 4.7 and 55.4 ± 3.9 % in the wild-type and knock-out thymocytes, respectively. These data suggest that loss of TG2 does not affect Fas sensitivity of TG2−/− thymocytes. Despite the lack of changes in Fas sensitivity of thymocytes, or even the delayed rate of their death in vivo, TG2−/− mice were much more susceptible to Fas-induced killing. Whereas all the wild-type animals survived the 1 μg/g body weight intraperitoneal dose of Jo2 antibody, none of the knock-out mice survived for more than 20 hours after administration of the same dose of anti-Fas antibodies (Fig. 1B).
Wild-Type Hepatocytes Die Primarily by Necrosis, Whereas TG2 Knock-Out Cells Die by Apoptosis After Fas Engagement In Vivo by a Nonlethal Dose of Anti-Fas Antibodies.
Because it was previously shown that higher doses of anti-Fas antibodies kill mice by inducing massive liver apoptosis,25 histological sections of livers from both wild-type and TG2−/− mice were investigated after anti-Fas antibody injection. In the wild-type liver at 5 hours after anti-Fas injection, the structure investigated by light microscope (LM) seemed to be only slightly altered, as compared with the untreated controls (Fig. 2A), with dilated central veins visible and with perivenous hepatocytes displaying a pale, granulated cytoplasm (Fig. 2C). Electron microscopically, however, necrotic hepatocytes could be detected displaying vacuolized plasma and disorganized mitochondria (Fig. 3C). Nineteen hours later, significant cellular damage could be detected by LM, with markedly dilated blood vessels, single-cell or confluent necrosis with a marked hemorrhagic component, and with a few apoptotic cells being visible (Fig. 2D). These changes were significantly more conspicuous when the samples were examined by TEM (Fig. 3E). In accordance with the lack of significant death induction, at early time points terminal deoxynucleotidyl transferase–mediated nick-end labeling (TUNEL) staining, which detects single-strand DNA produced by activated DNases, was detected only in the endothelial cells (Fig. 4A-B), and even at later points relatively few TUNEL-positive hepatocytes could be detected (Fig. 4C). Lack of significant apoptosis induction was further confirmed by the lack of cytochrome c release and procaspase-9 cleavage in wild-type livers (Fig. 5).
Within the sections of livers from the non-treated knock-out animals, no alterations in the liver structure were found even at electron microscopical level (Fig. 3B). However, at 5 hours after Fas engagement in vivo, a significantly increased number of apoptotic hepatocytes could be detected by LM (Fig. 2B). This was accompanied by massive hemorrhagic suffusions, suggesting also the development of disseminated endothelial cell damage. Appearance of apoptotic cells was confirmed by TEM (Fig. 3D). In addition, a peculiar finding is demonstrated in Fig. 3F. In several samples, the formation of labyrinthine, curvilinear structures that resemble folded up or spiraling rough/smooth endoplasmic reticulum was detected. These structures often appeared in isolation from hepatocytes, without a clearly definable outer limiting membrane. Induction of apoptosis in hepatocytes was confirmed by an intense TUNEL staining all over on the tissue sections (Fig. 4D), and by demonstrating the release of cytochrome c and cleavage of procaspase-9 (Fig. 5).
The increased damage of TG2−/− hepatocytes after anti-Fas treatment in vivo was further confirmed by the significantly elevated serum AST and ALT levels in TG2−/− mice as compared with wild-type mice determined at the time points of the death of knock-out animals. However, as the time passed, AST and ALT levels increased further in the surviving wild-type mice as well, demonstrating a continuous and time-dependent liver damage following the anti-Fas treatment (Table 1).
|Mice||Treatment||ALT (U/mL)||AST (U/mL)|
|TG2+/+||Saline||85 ± 49||176 ± 92|
|Anti-Fas†||2,416 ± 2,012*||1,841 ± 1,432*|
|Anti-Fas‡||7,008 ± 4,493||5,420 ± 3,421|
|TG2−/−||Saline||96 ± 45||317 ± 237|
|Anti-Fas||11,096 ± 6,214||7,969 ± 4,549|
These data implied that TG2−/− livers are more sensitive to an otherwise nonlethal dose of anti-Fas antibodies, that the type of hepatocyte death induced by the nonlethal dose of anti-Fas antibodies is different in the wild-type and knock-out animals, and TG2, as a cross-linking enzyme, might be required for the proper apoptotic morphology.
TG2−/− Hepatocytes Show Increased Sensitivity Toward Anti-Fas Treatment In Vitro in Correlation With Their Lower Bcl-xL Expression.
To determine whether TG2−/− hepatocytes are more sensitive to anti-Fas treatment than wild-type cells also in vitro, hepatocytes were isolated, cultured, and exposed to increasing concentrations of anti-Fas antibodies. The percentage of apoptotic hepatocytes was determined 24 hours later. As shown in Fig. 6A, wild-type hepatocytes survived well under the experimental conditions and were relatively resistant to Fas-mediated death, as previously shown.35 TG2−/− hepatocytes, however, demonstrated an increased sensitivity to Fas-mediated death, as compared with wild-type cells. The increased in vitro sensitivity to Fas antibodies suggested an intrinsic change in the Fas cell death pathway in TG2−/− hepatocytes. This finding was not related to an enhanced cell surface expression of Fas (Fig. 6B). Neither did we find a difference in the hepatic expression of FLIPL, which interferes with the autoproteolytical maturation of Fas-FADD-Bound Caspase-8 (FLICE) proximal in the Fas apoptotic pathway (Fig. 6C,F).36
Nuclear factor-κB (NF-κB) is involved in the transcriptional control of many genes that protect cells against apoptosis. Notably, an anti-apoptotic role of NF-κB especially in hepatocytes was suggested by the embryonic death of NF-κB p65 knock-out mice from extensive apoptosis in the liver.37 NF-κB was shown to inhibit Fas-mediated apoptosis of hepatocytes,38 and Fas receptor stimulation was shown to be coupled to NF-κB activation in a pathway that involved caspase-8.39 Though the core mechanism of NF-κB activation is phosphorylation of its inhibitor (I-κB) by which I-κB is targeted to ubiquitation and consequent proteosomal degradation,40 recently it was shown that TG2 also can polymerize I-κBα, leading to its dissociation from NF-κB and to NF-κB activation.41 To test whether I-κBα polymerization by TG2 leading to NF-κB activation could be involved in the protection against Fas-mediated death of hepatocytes, the possible polymerization of I-κBα was followed in the livers of TG2+/+ and TG2−/− mice after Fas engagement. As shown in Fig. 5, in accordance with a possible activation of NF-κB by I-κBα phosphorylation during Fas signaling,39 I-κBα was degraded in both types of livers after Fas crosslinking. Because the kinetics of the disappearance of I-κBα, which is dependent on caspase-8 activation,39 was similar in the 2 types of the liver, caspase-8 must have been activated in the wild-type livers as well. However, formation of I-κB polymers was not detected, implying that TG2 was not involved in the process. By 20 hours, the levels of I-κBα returned to the basal level in the surviving wild-type mice. This finding may be the result of the stimulation of I-κBα synthesis by NF-κB.42
Because in TG2−/− livers the sublethal dose of anti-Fas antibodies did not induce cytochrome c release or procaspase-9 cleavage, which is known to be influenced by the levels of mitochondrial anti-apoptotic proteins, the level of Bcl-xL expression in hepatocytes was also determined. As shown in Fig. 6C-D, Bcl-xL expression was found to be decreased in TG2−/− hepatocytes, as compared with the wild-type cells.
Inhibition of the Alpha-1b-Adrenergic Receptor Leads to Both Downregulation of the Bcl-xL Expression and Increased Fas-Sensitivity of Hepatocytes.
Because TG2 participates as a G protein in the AR-signaling pathway in the liver,11 and it was previously shown that in TG2−/− hepatocytes the AR signaling is impaired,43 we decided to test the possibility that impaired AR signal transduction is related to the observed decrease in Bcl-xL expression in the liver of TG2−/− mice. To achieve this, various doses of chloroethylclonidine, an AR antagonist known to fully block the AR pathway of hepatocytes,31 were injected daily in wild-type mice for a week, and than the Bcl-xL expression of hepatocytes was determined. As shown in Fig. 6E, inhibition of the AR signaling pathway by chloroethylclonidine resulted in a dose-dependent decrease in the Bcl-xL expression of hepatocytes. When these hepatocytes were isolated and cultured, their sensitivity to anti-Fas treatment in vitro and the rate of their spontaneous cell death were similar to that of TG2−/− hepatocytes (Fig. 6A). These data demonstrated that impaired AR signaling can lead to downregulation of Bcl-xL in hepatocytes and to a consequent increase in their Fas sensitivity.
Alpha-1b Adrenergic Receptor–Deficient Mice Are Also More Sensitive to Fas-Induced Killing.
To prove further the role of AR in the regulation of Fas sensitivity of hepatocytes, the Fas sensitivity of AR-deficient mice30 was also determined. The wild-type partners for the AR-deficient mice were C57BL/6 mice, in which the hepatic density of β2-adrenergic receptors is low. As a result, these mice were reported to exhibit increased sensitivity to Fas-mediated apoptosis of hepatocytes.44 Consequently, lower doses of anti-Fas antibodies were found to be sublethal for this mouse strain. Injection of 0.3 μg/g body weight antibody resulted in 80% survival of the wild-type (C56BL/6) mice (Fig. 7A). The same dose of antibody, however, killed 90% of the AR-deficient mice within 2 days. Increased damage of AR-deficient livers as compared with their wild-type counterparts was demonstrated by the higher number of TUNEL-positive liver cells on tissue sections (Fig. 7E-F) and by the higher serum ALT and AST levels (Fig. 7B) after anti-Fas treatment. LM sections of Fas-treated AR+/+ livers revealed that besides the healthy-looking parenchymal cells, many hepatocytes lost their nucleus, contained nuclear fragments, and their cytoplasm was condensed, homogenous, and eosinophilic. These cells represented apoptotic cells. A low number of necrotic cells were also visible (Fig. 7C). On the LM sections of Fas-treated AR−/− livers, the basic structural features of the histological structure of liver tissue were mostly lost. Most parenchymal cells either lost their nucleus or showed apoptotic morphology. In addition, necrotic cells were also visible (Fig. 7D).
In accordance with our previous finding that chloroethylclonidine treatment downregulated Bcl-xL levels, AR-deficient livers also expressed lower levels of Bcl-xL than their wild-type counterparts (Fig. 6C-D). Interestingly AR+/+ (C56BL/6) mice also expressed lower levels of FLIPL and Bcl-xL than the TG2+/+ (FVB) mice, which might explain their increased Fas sensitivity. Conversely, no difference was found in the FLIPL expressions of the AR+/+ or AR−/− hepatocytes (Fig. 6F). Collectively, these data confirm that the impaired AR signaling makes the animals more sensitive to Fas-mediated killing, and this is related to changes in the Fas sensitivity of hepatocytes.
Treatment with phenylephrine, an AR agonist, protects hepatocytes against Fas-mediated apoptosis. Next we decided to test whether stimulation of AR could protect hepatocytes against a lethal dose of anti-Fas antibodies. For this, FVB mice were exposed to phenylephrine, an alpha-1-adrenergic agonist, at 48, 24, and 6 hours before injection of a lethal dose of anti-Fas antibodies. As shown in Fig. 8A, injection of 2 μg/g body weight anti-Fas antibodies to FVB mice resulted in the loss of all of the animals within 25 hours. At 4 hours after anti-Fas antibody injection, massive destruction of the liver was detected with a high number of TUNEL-positive hepatocytes showing apoptotic morphology (Fig. 8D-E). The extent of tissue damage was also demonstrated by the simultaneously high AST and ALT levels (Fig. 8B). Induction of apoptosis was accompanied by the cleavage of procaspase-9 molecules (Fig. 8C). Pretreatment with phenylephrine resulted in induction in the bcl-xL levels and in a complete protection of hepatocytes against Fas-induced apoptosis demonstrated by the lack of procaspase-9 cleavage, by the lack of apoptotic hepatocytes on tissue sections, and by the physiological AST and ALT serum levels at the same time (Fig. 8). Despite the lack of hepatocyte apoptosis, however, mice were not protected by phenylephrine against the death induced by 2 μg/g body weight dose of anti-Fas antibodies, because phenylephrine-treated mice died with the same kinetics as the non-treated ones (Fig. 8A). That is why we also tested the effect of phenylephrine on Fas-mediated death of C56BL/6 mice, for which a 0.3 μ/g bodyweight dose resulted in 20% mortality rate because of increased hepatic Fas sensitivity (Fig. 7). Pretreatment with phenylephrine not only protected hepatocytes of C56BL/6 mice against Fas-mediated death, but all of the 20 mice tested survived the injection of 0.3 μg/g body weight dose of anti-Fas antibodies (data not shown).
The liver of mice is particularly sensitive in vivo to anti-Fas antibodies Jo2, which cause massive liver apoptosis that is followed by the death of the animals after only a few hours, with a phenotype that resembles fulminant hepatitis.25 Studies related to the mechanism of Fas-mediated signaling in the liver showed that the pathway involves activation of caspase-8, cleavage of bid, and the mitochondrial cell death pathway.26 Accordingly, the Fas-mediated liver injury could be inhibited by overexpressing bcl-2,27 by deleting bid,28 or by administering interleukin-1β -converting enzyme-like protease inhibitors.45
We demonstrated that mice lacking TG2 are more sensitive to Fas-induced liver injury and killing than the wild-type animals. Increased hepatic sensitivity was demonstrated both in vitro and in vivo. As a result, TG2−/− hepatocytes exposed to an otherwise sublethal dose of Jo2 antibody died rapidly by apoptosis. The same dose of Jo2 antibodies caused some liver damage in the wild-type animals as well. However, this seemed to be secondary to the apoptotic damage of endothelial cells, because the endothelial cell death in the long term might have caused disturbances in the oxygen or nutritional transport, which led to the observed necrotic damage in the hepatocytes. In addition to the altered Fas sensitivity, an altered apoptotic morphology of TG2−/− hepatocytes was also described.
The altered sensitivity of hepatocytes to Fas-mediated death was not a generalized phenomenon for each cell type, because TG2−/− thymocytes did not die faster after Fas engagement. Neither was it related to an altered Fas or FLIPL expression, or to an altered NFκB activation in TG2−/− hepatocytes. However, a good correlation was found with the decreased Bcl-xL protein expression. Because TG2 was shown to mediate AR signaling in the liver,11 and in TG2−/− hepatocytes the AR signaling is impaired,43 whether the impaired AR signaling might be related to the observed downregulation of Bcl-xL in TG2−/− hepatocytes was investigated. Wild-type hepatocytes treated for 1 week with chloroethylclonidine, an AR antagonist, indeed expressed lower levels of Bcl-xL and demonstrated increased sensitivity to Fas-induced death in vitro. Hepatocytes of AR-deficient mice also expressed lower levels of Bcl-xL protein and were more sensitive to anti-Fas antibodies than their wild-type counterparts. In addition, AR-deficient mice were also more sensitive to the killing effect of anti-Fas antibodies.
Addition of phenylephrine, an AR agonist, however, induced bcl-xL levels and prevented hepatocytes against Fas-mediated death. The effect of phenylephrine on the Fas-induced animal death was, however, dependent on the dose of antibodies used. Whereas for C57BL/6 mice, in which the animal death seems to be induced primarily by hepatocyte apoptosis at the 0.3 μg/g body weight dose of anti-Fas antibodies, phenylephrine was found to be protective, it failed to prevent the death of FVB mice induced by 2 μg/g body weight of anti-Fas antibodies. This result is in line with the previous finding that selective targeting of bcl-2 to the liver cannot protect animals against Fas-induced killing.27 Altogether, our data provide evidence for an AR-mediated signaling pathway that provides protection against Fas-mediated apoptosis in hepatocytes by regulating the Bcl-xL level.
After various kinds of liver injury, such as partial hepatectomy or CCl4 administration, increases in the level of Fas ligand, hepatocyte growth factor (HGF) and norepinephrine were detected within 2 hours after the treatment.46, 47 Not only ligation of the HGF receptor, but Fas engagement by Fas ligand was also shown to be required for and accelerate liver regeneration after partial hepatectomy, and this effect was related to a simultaneous inhibition of the Fas-mediated cell death pathway in hepatocytes.47 In this context, it is worth noting that both HGF48 and norepinephrine (acting on the beta2-adrenergic receptors)44 were reported to block Fas-mediated death pathway in hepatocytes, and here we show that ligation of AR is also inhibitory. In addition, norepinephrine was shown to both promote HGF production49 and to act as co-mitogen with the produced HGF.50 These data together with ours suggest a complex crosstalk between the HGF, Fas ligand, and norepinephrine-induced signaling pathways, and imply that adrenergic receptor–driven signal transduction pathways might participate in the recovery program of the liver after hepatic injury by promoting both the HGF- and the Fas ligand–driven proliferation of hepatocytes.
- 41Transglutaminase 2 induces Nuclear factor-κB activation via a novel pahway in BV-2 microglia. J Biol Chem 279; 51: 53725-53735., , , , , , et al.