The superoxide dismutase mimetic MnTBAP prevents Fas-induced acute liver failure in the mouse

GASTROENTEROLOGY2001;121:1451-1459

The Superoxide Dismutase Mimetic MnTBAP Prevents Fas-lnduced Acute Liver Failure in the Mouse BENO/T MALASSAGNE,¢ ,*,t" PIERRE-JACQUES FERRET,§ RENAUD H A M M O U D , * , t MICHELINE TULLIEZ,II SASSIA B E D D A , * HI~LI~NE TRI~BI~DEN, § PATRICK JAFFRAY,~ YVON CALMUS, # BERNARD WEILL,* and FRI~DI~RIC BATTEUX* *Laboratoire d'lmmunologie, IILaboratoire d'Anatomie Pathologique, and #Laboratoire de Biologie Cellulaire, Faculte Cochin, Universite Paris V; tService de Chirurgie Digestive, CHU Mondor, Universit6 Paris XII, Cr6teil; §INSERMUnit6 477 and ¶Laboratoire de Biochimie A, H6pital Cochin, Paris, France

Background &Aims: Acute liver failure (ALF) of viral or-

igin results from massive hepatocyte apoptosis induced by the interaction between Fas expressed on hepatocytes and Fas ligand on activated T lymphocytes. Because Fas-induced apoptosis of hepatocytes involves mitochondrial damages and potential reactive oxygen species (ROS) overproduction, we investigated whether manganese III tetrakis (5,10,15,20 benzoic acid) (MnTBAP), a nonpeptidyl mimic of superoxide dismutase (SOD), can inhibit Fas-induced ALF. Methods: An agonist anti-Fas monoclonal antibody was used to induce hepatocyte apoptosis in vitro and ALF in vivo. Results: Preventive and curative treatments by MnTBAP significantly increased survival rates and significantly reduced aspartate aminotransferase levels and parenchymal lesions. ROS generation was suggested by those beneficial effects and significant increases in SOD and Gpx activities after anti-Fas injection. Flow cytometry of isolated hepatocytes incubated with anti-Fas monoclonal antibody showed that ROS production was associated with the collapse of transmembrane potential and loss of cardiolipin content. After injection of anti-Fas monoclonal antibody, mitochondrial Bcl-2 was decreased, cytochrome c released, and caspase-3 activated. Mitochondrial alterations and their consequences were abrogated by MnTBAP. Conclusions: ROS are key executioners in Fas-induced hepatocyte apoptosis. This finding explains why a nonpeptidyl mimic of SOD can cure ALF in a model of viral hepatitis, pointing out the potential interest of this molecule in humans.

cute liver failure (ALF), a devastating condition that results from massive hepatocyte apoptosis, is responsible for about 1800 deaths each year in the United States. So far, the foremost cause of ALF has been viral hepatitis, accounting for about 70% of cases. The only treatment that can be applied to virus-induced ALF is liver transplantation. 1 Therefore, many lives could be saved and transplantation spared for some patients if a

A

drug could be found that would inhibit the pathophysiological mechanisms leading to hepatocyte apoptosis. Fas-Fas ligand (FasL) interaction plays a major role in hepatocyte injury during viral hepatitis. Cytotoxic T lymphocytes recognize viral antigens expressed on hepatocytes and kill virus-infected hepatocytes at least partially through Fas-induced activation of caspases. 2 If this process is exaggerated, the major part of, if not the whole, liver is destroyed by cytotoxic T cells, resulting in ALF. 3,4 Hepatocyte apoptosis is currently observed by pathologists on liver specimens from patients with viral hepatitis, and it is originally recognized as "acidophilic bodies." Similar lesions can be observed in mice with ALF after intravenous (IV) injection of agonist anti-Fas monoclonal antibody (mAb), which is a relevant model of ALF of viral origin. 5 Novel approaches for the treatment of ALF of viral origin, such as peptidic caspase inhibitors, have been reported, and they are effective protectors against hepatocyte apoptosis. 6,7 However, in hepatocytes from caspase-1, -3, -6, and -9 knockout mice, Fas can induce cell death through compensatory pathways involving the activation of other caspases, s On the other hand, depending on the cell type, activation of the Fas-signaling pathway leads to cell death either through direct caspase activation (type I ceils) or through mitochondrial disruption (type II cells). 9 11 Because hepatocytes have both type I (slow) and type II (fast) apoptotic pathways after "~Deceased. Abbreviations used in this paper: ALF, acute liver failure; DHE, dihydroethidium; FasL, Fas ligand; mAb, monoclonal antibody; MnTBAP,manganeseIII tetrakis (5,10,15,20 benzoicacid); NAO,nonyl acridine orange; ROS, reactive oxygenspecies; SOD, superoxidedismutase; TUNEL, terminal deoxynucleotidyltransferase-mediated deoxyuridinetriphosphate nick-end labeling. © 2001 by the American GastroenterologicalAssociation

0016-5085/01/$35.00 doi:10.1053/gast.2001.29590

1452

MALASSAGNEETAL.

Fas ligation, 12 Fas-induced death of hepatocytes could depend p r e d o m i n a n t l y on mitochondrial disruption rather than on direct cleavage of caspase-3 by caspase-8.13 Because mitochondrial generation of reactive oxygen species (ROS) may be involved in mitochondrial damages, we were p r o m p t e d to investigate whether manganese III tetrakis (5,10,15,20 benzoic acid) (MnTBAP), a superoxide dismutase (SOD) m i m i c also endowed with catalase-like activity, 14 can inhibit apoptosis in this particular type of cells. Preventive and curative treatments by M n T B A P significantly increased the survival rates of mice with Fasinduced ALF and dramatically reduced serum aspartate aminotransferase (AST) activities and parenchymal lesions. Those beneficial effects of M n T B A P were consistent with the enzymatic properties of the molecule. Indeed, M n T B A P abrogated the ROS overproduction observed in hepatocytes after in vivo and in vitro exposure to anti-Fas mAb. As a consequence of ROS detoxification, mitochondrial damages, cytochrome c release, and caspase-3 activation were all inhibited.

M a t e r i a l s and M e t h o d s Animals Six to 8-week-old female BALB/c mice (Iffa Credo, L'Arbresle, France) were used in all experiments. They received humane care in compliance with institutional guidelines.

Experimental Procedures Fas-induced ALF was achieved in 27 mice by IV injection of a lethal dose (0.25 txg/g) of a purified hamster agonist anti-Fas mAb (Clone Jo2; PharMingen, San Diego, CA) diluted in 100 btL of sterile saline. In the preventive protocol, 15 other mice were treated intraperitoneally with 10 mg/kg MnTBAP in phosphate-buffered saline (PBS) (Alexis Biochemicals, Paris, France) 2 hours before the induction of acute liver failure by anti-Fas mAb. The curative protocol consisted of the same dose of MnTBAP administered intraperitoneally 2 hours after anti-Fas mAb to 12 other mice. Two control groups of 16 animals each received either PBS or 10 mg/kg MnTBAP alone. For the survival time courses, mice were killed when they appeared moribund. The moribund state was appreciated in a blinded manner by 2 observers who were not aware of which animals had received which treatment. Surviving animals and controls were killed 24 hours after anti-Fas mAb injection. All ex vivo assays were performed at the time of death or 24 hours after anti-Fas mAb injection in surviving mice treated with MnTBAP and in controls injected with MnTBAP alone. At the time of sacrifice, mice were bled by cardiac puncture for measurement of serum AST activities, and livers were removed from the peritoneal cavities for histopathological and biochemical studies. A sublethal dose of 0.15 I~g/g of agonist anti-Fas mAb was injected to 3 groups

GASTROENTEROLOGYVol. 121, No. 6

of 10 mice each. Two groups were preventively and curatively treated by MnTBAP, respectively, following the above-mentioned protocols. Two additional groups of 5 mice each received either PBS or MnTBAP alone. All the animals were killed 12 hours after the injection ofanti-Fas mAb. Serum AST levels, SOD and Gpx activities, cytochrome c release, caspase-9 and -3 activities, and mitochondrial Bcl-2 content were determined in liver extracts. Histologic examination and terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling staining. Livers were cut into small pieces (<0.1 cm 3) and fixed overnight at 4°C in 4% paraformaldehyde in PBS. The tissue was then dehydrated, paraffin-embedded, cut at 4-b~m thickness, and stained with H&E. Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL) staining was performed on dewaxed paraffin-embedded sections that were stained with fluorescent anti-TdT (termino-deoxynucleotidyl transferase) using the In Situ Cell Death Detection kit from Boehringer Mannheim (Mannheim, Germany). Fluorescent nuclei were enumerated on liver sections from mice injected with anti-Fas mAb and preventively treated or not by MnTBAP. Isolation of primary mouse hepatocytes. Livers were washed with 10 mmol/L HEPES, pH 7.6, until complete liver decoloration, then perfused with a 0.5% Collagenase H solution (pH 7.4, 37°C, constant inflow 3 mL/min, 30 min) (Boehringer Mannheim). Perfused liver was minced in M199 medium (GIBCO BRL, Cergy-Pontoise, France) and filtered through a 70-1~m mesh filter. .5 Viable hepatocytes were separated by Percoll centrifugation. Viability of recovered cells exceeded 90% as judged by trypan blue exclusion. ROS production, mitochondrial damages, and apoptosis in isolated hepatocytes. Hepatocytes (5 × 105 cells), preincubated for 2 hours with 1 mg/mL MnTBAP or with culture medium, were washed and then exposed to anti-Fas mAb for 6 hours at 37°C. Cells were incubated with the appropriate dyes (all from Molecular probes, Leiden, The Netherlands) for the last 30 minutes at 37°C. After washing, cells were incubated with 1 I,tg/mL PI (propidium iodide) or 2 I~mol/L YO-PRO-1 and 5 I~g/mL annexin V-Cychrome for 15 minutes. Fluorescence-activated cell sorter analysis was performed on PI- or YO-PROI-negative and annexin V-positive cells (early stage of apoptosis), using a Coulter (Villepinte, France) Epics XL Flow cytometer. ROS production was assessed by oxidation of 5 btmol/L 2',7'-dichlorodihydrofluorescein diacetate (CMDCFDA), 5 I~mol/L dihydroethidium (DHE), or 5 b~mol/L anthracen-9,10-dipropionic acid (Anthracen). Variations in the mitochondrial transmembrane potential dit~m were studied using 0.1 p~mol/L of chloromethyl X-rosamine (CMXRos). Alterations in hepatocyte mitochondrial cardiolipin content were evaluated by incubation with 2 b~mol/L nonyl acridine orange (NAO). 16 Kinetics ofO2 ° (anthracen staining), apoptosis (annexin V staining), and cell death (YO-PRO-1 staining) were studied at various time points

December 2001

during 18 hours in 6 experiments. Forward and side scatters were used to establish size gates and exclude cellular debris from the analysis. Cells were examined at a flow rate of 100-200 events per second, and 10,000 events were analyzed per sample. Enzymatic activities. After lethal injection of antiFas mAb, enzymatic activities were determined and all other ex vivo assays performed at the time of death or 24 hours after the injection in surviving mice treated with MnTBAP and in controls injected with PBS or MnTBAP alone. When anti-Fas mAb was administered at a sublethal dose, the assays were performed 12 hours later. Serum AST activities were measured using a standard clinical automatic analyzer (type 747; Hitachi, Meylan, France). Cellular glutathione peroxidase (Gpx) activity was evaluated in extracts of mouse livers by measuring the rate of reduced nicotinamide adenine dinucleotide phosphate oxidation at 340 nm (spectrophotometric assay; Calbiochem, La Jolla, CA). Total cellular superoxide dismutase (Cu/Zn and Mn SOD) activities were measured using the Bioxytech SOD-525 spectrophotometric assay from Oxis International (Portland, OR). Kits were used as recommended by the manufacturers. Bcl-2 and cytochrome c determinations by Western blotting. Bcl-2 was determined by Western blots on enriched mitochondrial (HM) fractions of mouse liver extracts, as described by Yang et al. iv Cytochrome c determination was performed on the cytosolic (Sl00) and HM fractions. The amount of proteins in each lysate was assayed using the bovine serum albumin microburet assay from Pierce (Bezons, France). Proteins were electrophoresed on 15% sodium dodecyl sulfate-polyacrylamide gels, transferred to Hybond-P membranes (Amersham-Pharmacia Biotec, Saclay, France), and incubated with anti-Bcl-2 or anticytochrome c mAbs (clones 3F11 and 7H8.2C12, respectively; PharMingen). Cytochrome c and Bcl-2 bands were revealed by chemiluminescence (Amersham), then scanned and quantified using the Bio-profil system (Vilbert Lourmat, Marne la Vall4e, France). Caspase-9 and -3 activities. Caspase-9 and -3 activities were determined using 400 I~mol/L chromogenic substrates Ac-LEHD-pNA and Ac-DEVD-pNA (Calbiochem), respectively, 200 b~g of cytosolic fraction (S100) in 200 btL of 100 mmol/L HEPES, pH 7.4, 20% glycerol, and protease inhibitors. Absorbances were measured at 405 nm for 3 hours. The reaction mixture without substrate or with specific inhibitors (200 btmol/L Ac-LEHD-CHO and 200 I~mol/L AcDEVD-CHO for caspase-9 and -3, respectively) were used as controls.

Statistical Analysis The statistical significance of differences between experimental treated groups and untreated controls was analyzed by X2 test for incidence data, by the Student t test for comparison of means, and by the Mann-Whitney U test for TUNEL determinations. A level of P < 0.05 was accepted as significant.

SOD MIMIC PREVENTSFAS-INDUCEDACUTELIVER FAILURE 1453

Results MnTBAP Protects Mice From Fas-lnduced ALF Seven hours after IV injection of 0.25 ~g/g of agonist anti-Fas mAb, no animal survived. At that time, the survival rate of mice preventively treated by MnTBAP10 was 80% (X2 = 24.25; P < 0.001 vs. untreated mice) (Figure 1A). Twenty-four hours after anti-Fas challenge, 55% of mice preventively treated with MnTBAP10 were still alive (X2 = 16.16; P < 0.001 vs. untreated mice). When MnTBAP10 had been injected 2 hours after antiFas mAb, 40% of curatively treated mice were alive 7 hours after anti-Fas injection (X2 = 21.6; P < 0.001 vs. untreated mice) and 25% of animals were alive 24 hours after anti-Fas injection (X2 = 16.26; P < 0.001 vs. untreated mice). No death was observed in control mice. At the time of death, after lethal injection of antiFas mAb, serum AST activities had significantly risen in mice injected with anti-Fas m A b (15,458 IU/L + 2309; mean --- SEM) compared with mice killed 24 hours after injection of PBS (34 lUlL _+ 27; P < 0.001) or MnTBAP10 alone (109 lUlL --- 30; P < 0.001). After a lethal dose of anti-Fas mAb, M n T B A P inhibited the increase in serum AST activities vs. controls (7648 + 788; P < 0.02) when preventively, but not curatively, administered. Twelve hours after injection of a sublethal dose (0.15 I ' g / g ) of anti-Fas mAb, all the animals were alive. The levels of serum AST were significantly decreased by preventive (P < 0.001) or curative (P < 0.01) administration of MnTBAP vs. nontreated animals.

Pathologic Study of Livers From Mice Injected With Anti-Fas mAb Livers from mice injected with a lethal dose of anti-Fas m A b showed extensive damages involving the whole parenchyma (data not shown). Almost all hepatocytes showed chromatin condensation and nuclear shrinkage. In contrast, treatment with MnTBAP10 protected the livers. A l t h o u g h the lesions were less limited after curative than after preventive treatment, they were no longer confluent and the preserved areas had retained a normal parenchymal structure at 24 hours. Twelve hours after a sublethal dose of anti-Fas mAb, severe lesions were still observed (Figure 1B, upper left quadrant). Preventive and curative M n T B A P preserved the parenchymal structure in most areas (Figure 1B, middle and lower left quadrants). After sublethal anti-Fas mAb, the number of apoptotic hepatocytes in T U N E L experiments was significantly

1454 MALASSAGNEET AL.

GASTROENTEROLOGYVol. 121, No, 6

A PBS or MnTBAP

2

80 a :)

"E ,q

=

Preventive MnTBAP

60 t-I

O

a.

40

1 - ~

_

Curative MnTBAP

ii

im

im

aL

ah

A

II

II

20 'j

Anti.Fas mAb A

0

2

4

6

8

aL

10

12

14

A

16

18

20

22

24

time after anU-Fas Injection (hours)

B

Figure 1. Survival rates and histologic data. (A) Survival rates in mice injected with a lethal (0.25 t~g/g) dose of agonist anti-Fas mAb (e; n = 27); in mice injected with MnTBAP (10 mg/kg) 2 hours before antiFas mAb ( . ; n = 15) (P < 0.001 vs. untreated mice) or 2 hours after anti-Fas mAb (11; n = 12) (P < 0.001 vs. untreated mice), in control mice treated with PBS (©; n = 16), or with MnTBAP alone (A; n = 16). Curves are the compiled results of 3 experiments. Survival rates in the groups treated preventively and curatively were still significantly increased at 24 hours (P < 0.001 vs. untreated mice). (B) Microscopic views of livers 12 hours after injection of a sublethal (0.15 ~g/g) dose of anti-Fas mAb. H&E staining (250-fold magnification), left quadrants; TUNEL staining (400fold magnification), right quadrants; anti-Fas mAb alone, upper quadrants; anti-Fas mAb and preventive injection of MnTBAP, middle quadrants; and anti-Fas mAb and curative injection of

December2001

lower in mice preventively (47 + 17/mm2; P < 0.01) and curatively (78 + 16/mm2; P < 0.05) treated with MnTBAP, than in untreated mice (185 -+- 53/mm 2) (Figure 1B, right quadrants). ROS Production, Mitochondrial Damages, and Apoptosis in Isolated Hepatocytes Treated With Anti-Fas mAb Incubating isolated hepatocytes with anti-Fas led to the production of ROS as assessed by the marker CM-DCFDA (Figure 2A), and that production was abolished by preincubation with 1 mg/mL MnTBAP. Since MnTBAP is a mimic of SOD that acts on 02 ° - , we used a marker of superoxide anions, DHE. Increased staining by DHE after anti-Fas mAb in vitro treatment was prevented by MnTBAP (Figure 2B). Similar results were obtained with another 0 2 fluorigenic marker, anthracen 18 (Figure 2C). Using the cationic dye CMXRos, we observed a marked decrease in AtOm after treatment with anti-Fas mAb (Figure 2D). N A O staining showed a significant decrease in mitochondrial cardiolipin under the same conditions (Figure 2E). Again, incubation of hepatocytes with MnTBAP abrogated those mitochondrial alterations. ROS production and the rates of apoptosis and cell death were studied at different time points during 18 hours in isolated hepatocytes exposed to anti-Fas mAb. Anthracen staining showed a significant increase in O2 ° production after 3 hours. At that time point, no significant increase in hepatocyte apoptosis (annexin V staining) or cell death (YO-PRO-1 staining) could be observed. They were significantly augmented only 6 and 9 hours after exposure to anti-Fas mAb, respectively (Figure 3). Enzymatic Activities of Antioxidant Proteins A 2-fold increase in both SOD-like and Gpxlike activities was observed in the livers of normal mice 24 hours after injection of M n T B A P vs. controls injected with PBS (Table 1). A lethal dose of anti-Fas mAb induced 3-fold and 2-fold increases in SOD and Gpx activities, respectively, at the time of death. In the livers of surviving mice injected with anti-Fas antibody and preventively treated with MnTBAP, a 4-fold increase in SOD-like activity was observed at 24 hours vs. controls. After a sublethal dose of anti-Fas mAb, significant increases in SOD-like activity followed preventive and curative injections of M n T B A P vs. mice injected with anti-Fas mAb but not treated with MnTBAP.

SOD MIMIC PREVENTSFAS-INDUCEDACUTELIVERFAILURE 1455

Effects of MnTBAP on Cytochrome c Release, Caspase Activation, and Mitochondrial Bcl-2 Content in the Livers of Mice With ALF Preventive but not curative treatment by MnTBAP inhibited the release of cytochrome c 24 hours after a lethal dose of anti-Fas mAb (Table 1 and Figure 4A), caspase-3 activation, and the decrease in mitochondrial Bcl-2 (Table 1). Preventive and curative MnTBAP significantly inhibited cytochrome c release, caspases-9 and -3 activities, and the decrease in mitochondrial Bcl-2 content 12 hours after a sublethal dose of anti-Fas mAb (Table 1 and Figure 4A and B). Discussion This report describes the beneficial effects of a nonpeptidyl mimic of SOD in the treatment of Fasinduced ALF in mice. The correlations between survival curves, serum AST activities, and histologic data indicate that severe hepatitis related to hepatocyte apoptosis occurred in the animals injected with a lethal dose of agonist anti-Fas mAb. The beneficial effects of MnTBAP on all the parameters tested after a lethal injection of anti-Fas mAb suggest that the molecule can inhibit Fas-induced apoptosis of hepatocytes. Although less pronounced than those of preventive administration, the effects of curative treatment remain highly significant and are particularly remarkable because MnTBAP was administered only 1 hour before the death of the first animal. That period of time was too short to prevent the increase in serum AST levels. However, 12 hours after sublethal administration of anti-Fas mAb and preventive or curative MnTBAP, serum AST levels were significantly decreased vs. animals untreated by MnTBAP. Because MnTBAP is a molecule known to detoxify 02 ° - and H202,14 our results indirectly suggest the involvement of ROS in this model of hepatocyte apoptosis. 19 The generation of 02 ° - and H , O , is also suggested by the increases in SOD and Gpx activities in the livers after in vivo anti-Fas mAb challenge. A direct evidence of ROS production, especially of 02 °-, has been provided in vitro by flow cytometric analysis of isolated mouse hepatocytes. ROS production was significantly diminished by preincubation with MnTBAP as early as 3 hours after exposure to anti-Fas mAb, at a time when apoptosis and cell death were not observable. Therefore, it is likely that the beneficial effects of MnTBAP result from the detoxification of ROS generated after Fas engagement and causing apoptosis and cell death. Indeed, MnTBAP retains its enzymatic properties in vivo, as shown by the

1456

MALASSAGNE ET AL.

GASTROENTEROLOGY Vol, 121, No. 6

Without MnTBAP

A

With MnTBAP

250 -~

25O

200

< 10050 ~

.

~

~

0

0,1

B ~

1

~

'

10

100

0

1000

.... ~

0,1

150

150

100

100

[

1

10

100

1000

I 1

10

100

1000

;

I

100

1000

5O 0

"1

0,1

C

10

100

1000

250

200

200

150

150

~

100

100

<

50

50

0 0,1

:~

~

1

10

100

1000

0 0,1

5o

50

40

40

30

30

I 1

10

20

c..)

10

10 0

1

0,1

E

0 -'-0,1

250 "~

D

1

10

.~

300

"~

250

j

~

0

200 150 100 5O 0 0,1

1

10

100

1000

0 0,1

1

10

100

1000

300 250 2OO 150 100 5O 100

1000

0,1

1

10

100

1000

Figure 2. Flow cytofluorometry analysis of isolated hepatocytes after anti-Fas challenge (gray lines) and in controls without anti-Fas mAb (black lines), after preincubation or not with MnTBAP. The various parameters were plotted as fluorescence intensity on the abscissa and cell counts on the ordinate. All determinations were done on hepatocytes at an early stage of apoptosis (annexin V-positive, PI, or YO-PRO1-negative cells). ROS production was assessed by (A) CMDCFDA, (B) 02° production by DHE, and (C) anthracen staining. (D) Variations in A~m were evaluated using the cationic dye CMXRos, and (E) alterations in cardiolipin content were evaluated by NAO incorporation. Data from 1 of 4 representative experiments with similar results are presented.

December 2001

A

100

75

SOD MIMIC PREVENTS FAS-INDUCED ACUTE LIVER FAILURE

+PBS --I--MnTBAP "--Ii~--Anti-Fas ---X---Anti-Fas+ MnTBAP

~*W

~

***

0

0

3

6

9

12

18

Hours

100 :

-....I.--PBS

**

--I--MnTBAP

75

~Anti-Fas --X~Anti-Fas+ M n T B A P

~ ~

T _.....~ l

25

0

0

3

6

9

12

18

Hours

C

100 _w

75

PBS --W-- MnTBAP ~Anti-Fas •--K---Anti-Fas+ MnTBAP +

*** A ~

.L

. . / - z o.

50

0

3

6

9

12

1B

Hours

Figure 3. Flow cytometry analysis of isolated hepatocytes submitted to 1 i~g/mL of agonist anti-Fas mAb after preincubation or not with 1 mg/mL MnTBAP. (A) ROS production as assessed by inhibition of anthracen staining. (B) Early apoptosis as assessed by annexin V. (C) Cell death as assessed by YO-PRO-I. Data are presented as means _+ SEM of 6 independent experiments. Anti-Fas + MnTBAP vs. anti-Fas alone: * P < 0.05; * * P < 0.02; * * * P < 0.O1; and * * * * P < 0.001.

increase in SOD- and Gpx-like activities after its administration to normal mice. Thus, MnTBAP adds its intrinsic enzymatic properties to the SOD and Gpx activities elicited in the animals by the injection of anti-Fas mAb. We then investigated the relationship between ROS production and the caspase activation that follows crosslinking of Fas. ROS overproduction can induce severe mitochondrial injuries that lead to cytochrome c leakage into the cytoplasm. ~v In our hands, the generation of ROS by hepatocytes treated with agonist anti-Fas mAb is

1457

associated in vitro with mitochondrial membrane damages such as the collapse of transmembrane potential A~m and the loss of cardiolipin content. In vivo, injection of anti-Fas mAb leads to a decrease in mitochondrial Bcl-2 content in liver extracts and to the release of cytochrome c into the cytoplasm of injured cells. Cytochrome c release is responsible for the activation of caspase-9 through APAF12° that leads to hepatocyte apoptosis. The detoxification of ROS by MnTBAP and the simultaneous inhibition of apoptosis are consistent with the abrogation of mitochondrial alterations and subsequent prevention of caspase-9 and -3 activation, as observed in our in vivo and in vitro experiments. To study comparable numbers of animals at the same time point, the effects of MnTBAP were also investigated 12 hours after injection of a sublethal dose of anti-Fas mAb. Preventive and curative administrations of MnTBAP significantly inhibited the increase in AST levels and cytochrome c release, caspase-9 and -3 activities, and mitochondrial Bcl-2 decrease. Those data confirm that MnTBAP can inhibit the cascade of events leading to hepatocyte injury and apoptosis. Our data contrast with recent experiments showing that MnTBAP prevents ROS-mediated apoptotic death of T lymphocytes, but not Fas-induced apoptosis. H Reciprocally, inhibitors of caspases prevent Fas-induced T-cell death but not ROS-mediated mitochondrial disruption. Thus, Fas and ROS pathways of apoptosis are strictly independent in T lymphocytes classified as type I cells. ~3 In type II, Fas-mediated cell death primarily depends on mitochondrial disruption and cytochrome c release, whereas caspase activation is only a subsequent phenomenon) 2 Hepatocytes seem to behave as type II cells, as shown by the primary role of mitochondria during Fas-mediated apoptosis of those ceils in Bcl-2-transgenic mice. Bcl-2 blocks programmed cell death by inhibiting cytochrome c release from mitochondria. 17 Transgenic expression of Bcl-2 in the liver prevents Fas-mediated apoptosis of hepatocytes, 21 whereas Bcl-2 transgenic T lymphocytes are susceptible to Fas-induced cell death. 22,23 In line with this finding, we have observed a significant decrease in mitochondrial Bcl-2 content in livers of mice injected with anti-Fas mAb, as already described in isolated mouse hepatocytes. 24"25 Mitochondrial release of Bcl-2 was inhibited in animals treated with MnTBAP. Thus, our data not only confirm the pivotal role of mitochondria during Fas-mediated hepatocyte apoptosis, but also show that ROS are key executioners in mitochondrial membrane disruption, cytochrome c release, and hepatocyte apoptosis.

1458

MALASSAGNE ET AL.

GASTROENTEROLOGYVol. 121, No. 6

Table 1. Effects o f MnTBAP on t h e Levels o f SOD, Gpx, C a s p a s e - 9 and -3 Activities; on M i t o c h o n d r i a l Bcl-2 Content, and on M i t o c h o n d r i a l (HM) and Cytosolic ( $ 1 0 0 ) C y t o c h r o m e c C o n t e n t in Liver Extracts; on AST Levels in t h e Sera From M i c e Injected With Anti-Fas mAb Lethal anti-Fas (0.25 i,g/g -1)

Sublethal anti-Fas (0.15 i,g/g -1) PBS (n = 15) SOD Gpx Caspase-9 activity Caspase-3 activity Bcl-2 Cytochrome c (HM) Cytochrome c ($100) AST

363 222 33 36 151 102 14 134

MnTBAP (n = 15)

_+ 51 _+ 21 _+ 5 + 6 ± 4 ± 7 -+ 4 ± 27

589 427 34 41 142 109 11 109

+ 68 + 61 -+ 7 ± 4 ± 5 ± 10 ± 2 +_ 30

PBS (n = 10) 663 257 133 159 106 49 58 5034

MnTBAP P (n = 10)

+ 40 1101 +_ 34 301 ± 11 55 -+ 13 67 + 5 140 + 8 89 ± 8 19 + 619 983

MnTBAP C (n = 10)

_+ 89 d 898 - 43 ns 279 --- 9 d 79 + 10 d 114 ± 4d 129 ± 7d 72 + 4d 29 ± 145 d 1989

PBS (n = 10)

MnTBAP P (n = 8)

MnTBAP C (n = 3)

+ 56 ¢ 903 _+ 94 1423 ± 201 a 951 ± 39 ns 341 - 23 435 ± 89 ns 363 ± 10 c ND ND _+ 15 b 750 ± 143 310 +_ 59 b 720 ± 5c 77 ± 8 142 ± 4 d 87 + 8c 8 -+ 2 36 +- 5 12 _+ 5 c 104 + 14 71 ± 7 99 _+ 486 c 15,458 ± 2309 7648 _+ 788 b14,789

_+ i 0 5 ns -+ 97 ns ND ± 158 ns ± 7 ns ± 4 ns ± 13 ns ± 2217 ns

NOTE. Assays were performed at the time of death or 24 hours after injection of a lethal (0.25 i~g/g -1) dose of anti-Fas mAb. They were performed 12 hours after injection of a sublethal (0.15 i~g/g -1) dose of anti-Fas mAb. Results are expressed as U/mg protein for SOD and Gpx, as delta DO at 405 × 10 3 nm for caspase activities, and as arbitrary units for Bcl-2 and cytochrome c. AST, aspartate aminotransferase. ap < 0.05, bp < 0.02, cp < 0.01, dp < 0.001 versus mice injected with anti-Fas mAb, but not treated with MnTBAPlo.

cellular and mitochondrial membranes contrasts with the inability of native and recombinant SODs to penetrate into cells. 26 Therefore, SOD mimics such as MnTBAP are potential therapeutic tools that could be used in

MnTBAP has several advantages over SOD. First, it is endowed with catalase and Gpx activities. In addition, aqueous solutions of MnTBAP are effective through both systemic and oral routes (not shown). Its capacity to cross

A

Sublethal anti-Fas

Lethal anti.Fas

m a.

MnTBAP (P) (C)

X

Mnl~l~

(P)

(C) -<~ p15

HM : C~?>":~ :¸¸¸ r:

/: cC ¸. . . .

:

Sl00

:'

....c,i

B

+ ....

,L_

++:

p15

Sublethal anti-Fas ~] D.

MnTBAP (P) (C) I~

281(0

Figure 4. Preventive (P) or curative (C) MnTBAP prevents cytochrome c release from mitochondria and preserves mitochondrial Bcl-2 in livers from mice with Fas-induced ALF. After lethal (0.25 Fg/g) doses of anti-Fas mAb, determinations were performed at the time of death or 24 hours after anti-Fas mAb injection in surviving mice treated with MnTBAP and in controls injected with MnTBAP alone. After sublethal (0.15 i~g/g) doses of anti-Fas, all determinations were performed 12 hours after antibody injection. (A) Western blot analysis of cytochrome c in liver mitochondrial (HM) and cytosolic ($100) fractions from control mice injected with PBS or MnTBAP alone, and from mice injected with anti-Fas mAb, treated or not with MnTBAPlo. (B) Mitochondrial expression of Bcl-2. Western blot analysis of Bcl-2 was performed in enriched mitochondrial fractions of livers.

December 2 0 0 1

humans to cure viral ALF and possibly other diseases caused by cellular losses due to apoptosis. 27

References 1. Caraceni P, Van Thiel DH. Acute liver failure. Lancet 1995;345: 163-169. 2. Seino K, Kayagaki N, Takeda K, Fukao K, Okumura K, Yagita H. Contribution of Fas ligand to T cell-mediated hepatic injury in mice. Gastroenterology 1997;113:1315-1322. 3. Galle PR, Hofmann WJ, Walczak H, Schaller H, Otto G, Stremmel W, Krammer PH, Runkel L. Involvement of the CD95 (APO-1/Fas) receptor and ligand in liver damage. J Exp Med 1 9 9 5 ; 1 8 2 : 1 2 2 3 1230. 4. Kondo T, Suda T, Fukuyama H, Adachi M, Nagata S. Essential roles of the Fas ligand in the development of hepatitis. Nat Med 1997;3:409-413. 5. Ogasawara J, Watanabe-Fukunaga R, Adachi M, Matsuzawa A, Kasugai T, Kitamura Y, Itoh N, Suda T, Nagata S. Lethal effect of the anti-Fas antibody. Nature 1 9 9 3 ; 3 6 4 : 8 0 6 - 8 0 9 . 6. Rouquet N, Pages JC, Molina T, Briand P, Joulin V. ICE inhibitor YVADcmk is a potent therapeutic agent against in vivo liver apoptosis. Curr Biol 1996;6:1192-1195. 7. KOnstle G, Leist M, Uhlig S, Revesz L, Feifel R, MacKenzie A, Wendel A. ICE-protease inhibitors block murine liver injury and apoptosis caused by CD95 or by TNF-alpha. Immunol Lett 1997; 55:5-10. 8. Zheng TS, Hunot S, Kuida K, Momoi T, Srinavasan A, Nicholson DW, Lazebnik Y, Flavell RA. Deficiency in caspase-9 or caspase-3 induces compensatory caspase activation. Nat Med 2000;6: 1241-1247. 9. Vercammen D, Brouckaert G, Denecker G, Van de Craen M, Declercq W, Fiers W, Vandenabeele P. Dual signal of the Fas receptor: initiation of both apoptotic and necrotic cell death pathways. J Exp Med 1 9 9 8 ; 1 8 8 : 9 1 9 - 9 3 0 . 10. Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, Debatin KM, Krammer PH, Peter ME. Two CD95 (APO-1/Fas) signaling pathways. EMBO J 1998;17:1675-1687. 11. Scaffidi C, Kirchhoff S, Krammer PH, Peter ME. Apoptosis signaling in lymphocytes. Curr Opin Immunol 1999;11:277-285. 12. Hatano E, Bradham CA, Stark A, Limuro Y, Lemasters JJ, Brenner DA. The mitochondrial permeability transition augments Fas-induced apoptosis in mouse hepatocytes. J Biol Chem 2000;275: 11814-11823. 13. Yin XM, Wang K, Gross A, Zhao Y, Zinkel S, Klocke B, Roth KA, Korsmeyer SJ. Bid-deficient mice are resistant to Fas-induced hepatocellular apoptosis. Nature 1 9 9 9 ; 4 0 0 : 8 8 6 - 8 9 1 . 14. Patel M, Day BJ. Metalloporphyrin class of therapeutic catalytic antioxidants. Trends Pharmacol Sci 1 9 9 9 ; 2 0 : 3 5 9 - 3 6 4 . 15. de Coupade C, Gillet R, Bennoun M, Briand P, Russo-Marie F, Solito E. Annexin 1 expression and phosphorylation are up-regulated during liver regeneration and transformation in antithrombin III SV40 T large antigen transgenic mice. Hepatology 2000;31: 371-380.

SOD MIMIC PREVENTS FAS-INDUCED ACUTE LIVER FAILURE

1459

16. Zamzami N, Marchetti P, Castedo M, Zanin C, Vayssiere JL, Petit PX, Kroemer G. Reduction in mitochondrial potential constitutes an early irreversible step of programmed lymphocyte death in vivo. J Exp Med 1 9 9 5 ; 1 8 1 : 1 6 6 1 - 1 6 7 2 . 17. Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, Cai J, Peng TI, Jones DP, Wang X. Prevention of apoptosis by Bcl-2: release of cytochrome c from mitochondria blocked. Science 1 9 9 7 ; 2 7 5 : 1 1 2 9 1132. 18. Petit PX, Lecoeur H, Zorn E, Dauguet C, Mignotte B, Gougeon ML. Alterations in mitochondrial structure and function are early events of dexamethasone-induced thymocyte apoptosis. J Cell Biol 1995;130:157-167. 19. Jacobson MD. Reactive oxygen species and programmed cell death. Trends Biochem Sci 1 9 9 6 ; 2 1 : 8 3 - 8 6 . 20. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. Cytocbrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1 9 9 7 ; 9 1 : 4 7 9 - 4 8 9 . 21. Lacronique V, Mignon A, Fabre M, Viollet B, Rouquet N, Molina T, Porteu A, Henrion A, Bouscary D, Varlet P, Joulin V, Kahn A. Bcl-2 protects from lethal hepatic apoptosis induced by an anti-Fas antibody in mice. Nat Med 1 9 9 6 ; 2 : 8 0 - 8 6 . 22. Van Parijs L, Biuckians A, Abbas AK. Functional roles of Fas and Bcl-2-regulated apoptosis of T lymphocytes. J Immunol 1998; 160:2065-2071. 23. Van Parijs L, Peterson DA, Abbas AK. The Fas/Fas ligand pathway and Bcl-2 regulate T-cell responses to model self and foreign antigens. Immunity 1998;8:265-274. 24. Woo M, Hakem A, Ella AJ, Hakem R, Duncan GS, Patterson BJ, Mak TW. In vivo evidence that caspase-3 is required for Fasmediated apoptosis of hepatocytes. J Immunol 1999;163: 4909-4916. 25. Kovalovich K, Li W, DeAngelis R, Greenbaum LE, Ciliberto G, Taub R. Interleukin-6 protects against Fas-mediated death by establishing a critical level of anti-apoptotic hepatic proteins FLIP, Bcl-2, and Bcl-xL. J Biol Chem 2 0 0 1 ; 2 7 6 : 2 6 6 0 5 - 2 6 6 1 3 . 26. Petkau A. Scientific basis for the clinical use of superoxide dismutase. Cancer Treat Rev 1 9 8 6 ; 1 3 : 1 7 - 4 4 . 27. Salvemini D, Wang ZQ, Zweier JL, Samouilov A, Macarthur H, Misko TP, Currie MG, Cuzzocrea S, Sikorski JA, Riley DP. A nonpeptidyl mimic of superoxide dismutase with therapeutic activity in rats. Science 1 9 9 9 ; 2 8 6 : 3 0 4 - 3 0 6 .

Received October 6, 2000. Accepted August 23, 2001. Address requests for reprints to: Fr6d~ric Batteux, Ph.D., Laboratoire d'lmmunologie, Pavilion Hardy, H6pital Cochin, 7 5 6 7 9 Paris Cedex 14, France. e-mail: [email protected]; fax: (33) 1-58-4120-08. BenoR Malassagne and Pierre-Jacques Ferret equally contributed to the work. The authors dedicate this article to the memory of BenoR Malassagne, who was a brilliant surgeon, a bright scientist, and the best of friends.