Cardiotrophin-1 is an essential factor in the natural defense of the liver against apoptosis


  • Potential conflict of interest: Nothing to report.


We previously reported that exogenous cardiotrophin-1 (CT-1), a member of the IL-6 family of cytokines, exerts hepatoprotective effects. Because CT-1 is expressed in the normal liver, we hypothesized that this cytokine may constitute an endogenous defense of the liver against proapoptotic stimuli. Here, we found that CT-1−/− mice died faster than wild-type animals after challenge with a lethal dose of the Fas agonist Jo-2. At sublethal doses of Jo-2, all wild-type mice survived whereas CT-1−/− animals developed extensive hepatocyte apoptosis with 50% mortality at 24 hours. Pretreatment with CT-1 improved survival and reduced injury in both CT-1−/− and wild-type animals. Upon Fas ligation the activation of STAT-3, a molecule that defends the liver against apoptosis, was lower in CT-1−/− mice than in wild-type animals despite similar IL-6 up-regulation in the 2 groups. Analysis of liver transcriptome in CT-1−/− and wild-type mice showed that 9 genes reported to be associated with cell survival/death functions were differentially expressed in the 2 groups. Four of these genes [IGFBP1, peroxiredoxin3, TNFR1, and calpastatin (endogenous inhibitor of calpain)] could be validated by real-time PCR. All of them were down-regulated in CT-1−/− mice and were modulated by CT-1 administration. Treatment of CT-1−/− animals with the calpain inhibitor MDL28170 afforded significant protection against Fas-induced liver injury. Conclusion: CT-1−/− mice are highly sensitive to Fas-mediated apoptosis due in part to deficient STAT-3 activation and inadequate control of calpain activity during the apoptotic process. Our data show that CT-1 is a natural defense of the liver against apoptosis. This cytokine may have therapeutic potential. (HEPATOLOGY 2007;45:639–648.)

Programmed cell death or apoptosis is an essential process for normal tissue homeostasis. This process is disturbed in many pathophysiological conditions. The consequences of elevated hepatocyte apoptosis go beyond the simple loss of functional liver mass. Enhanced apoptosis may stimulate fibrogenesis,1, 2 and increased cell turnover in the context of chronic inflammation may create a favorable scenario for cancer development.3

Apoptosis can be triggered by many different stimuli that provoke cell stress. All these stimuli converge at the activation of caspase 3 that leads to internucleosomal DNA degradation, chromatin condensation, cell shrinkage, and formation of small apoptotic bodies that are phagocytosed by neighboring macrophages.4 Although different cellular organelles can be the origin of signals leading to caspase-3 activation, the death receptors expressed on the cell membrane are critical molecules that convey external proapoptotic stimuli and activate the cell death machinery. Hepatocytes express death receptors at high levels, and these molecules constitute important mediators of hepatocyte apoptosis in many different conditions such as viral infections, alcoholic and non-alcoholic hepatitis, cholestatic conditions, autoimmune liver diseases, ischemia/reperfusion injury, and graft rejection.5 In particular, the liver is very sensitive to Fas-induced apoptosis in such a manner that administration to mice of anti-Fas antibody leads to rapid death of the animal due to fulminant hepatitis.6

Apoptosis is a powerful and essential process that needs to be controlled by defensive forces. Diverse cytokines and growth factors, including IL-6, have been shown to protect the liver against proapoptotic insults. IL-6 is produced by nonparenchymal liver cells and exerts hepatoprotective effects by inducing STAT-3 (signal transducer and activator of transcription-3) activation.7 Activated STAT-3 translocates to the nucleus, where it induces the expression of genes that promote cell survival and regeneration.8 Recently, we have shown that cardiotrophin-1 (CT-1), a member of the IL-6 family of cytokines, is generated within the liver and exerts potent antiapoptotic effects on hepatocytes.9 Binding of CT-1 to its receptor on hepatocytes is followed by activation of the cell survival pathways STAT-3, AKT, and ERK1/2. Administration of exogenous CT-1 has been shown to protect the liver in rat models of fulminant hepatic failure and in mice models of concanavalin-A–induced hepatitis.9 Based on these data, we have hypothesized that CT-1 might be a component of the defensive machinery of the liver against the proapoptotic injury. To test this hypothesis, we have analyzed the sensitivity of CT-1–deficient mice to a Fas-agonist antibody. Our data reveal an as yet unrecognized role of CT-1 in the natural defense of the liver against apoptosis.


CT-1, cardiotrophin-1; SOCS-3, suppressor of cytokine signaling-3; STAT-3, signal transducer and activator of transcription-3; TUNEL, terminal deoxynucleotide transferase-mediated dUTP nick-end labeling; WT, wild-type.

Materials and Methods

Animal Studies.

All experiments were performed according to the institutional guidelines for the use of laboratory animals. Studies were performed on male (20–25 g) CT-1−/− mice backcrossed into a C57BL6 background for 11 generations (provided by Dr. Pennica, Genentech). The C57BL/6 mice, male (20–25 g) were originally obtained from Jackson Laboratory (Bar Harbor, ME). The CT-1 genotype was confirmed in the breeders by genomic DNA isolation and polymerase chain reaction as previously described.10

Acute Liver Injury.

For survival experiments, male 8- to 10-week-old CT-1−/− and wild-type (WT) animals received a single intraperitoneal injection of the agonistic anti-Fas monoclonal antibody Jo-2 (BD Pharmingen, San Diego, CA) at the lethal dose (0.4 μg/g body weight) or sublethal dose (0.15 μg/g body weight) diluted in sterile saline. The mice were monitored for up to 30 hours, and the time of death was recorded.

The Fas injury model was induced in WT and CT-1−/− mice with a single intraperitoneal injection of Jo-2 at the dose of 0.15 μg/g weight. At the indicated time points (up to 24 hours) after Fas ligation, mice were bled and killed by cervical dislocation. Serum was analyzed for ALT and AST, and livers were snap frozen in liquid nitrogen or formalin-fixed and paraffin embedded for histopathological studies. Where indicated mice were injected intravenously (retroorbital plexus) with rat recombinant CT-1 isolated and purified in our laboratory11 at the dose of 5 μg/mouse 30 minute before Jo-2.

In other experiments, animals were injected intraperitoneally with IGF-binding protein1 (IGFBP1) (R&D Systems Inc., Minneapolis, MN) 0.3 μg/g body weight 30 minutes before the Fas-agonist antibody or with a potent cell-permeable calpain inhibitor, MDL28170 from Sigma (St. Louis, MO) 60 μg/g body weight 1 hour after the injection of Jo-2 monoclonal antibody.

Primary Liver Cells.

Primary hepatocytes and nonparenchymal cells from mice C57BL/6 were obtained by the conventional two-step collagenase perfusion in combination with low-speed centrifugation from Pharmakine (Bilbao, Spain).

RNA Isolation and Gene Expression Analysis.

Total RNA was isolated12 and treated with DNase I (Gibco BRL, Gaithersburg, MD) before reverse transcription with M-MLV Reverse Transcriptase (Gibco BRL).

Some primers were from Sigma and designed to distinguish between genomic and complementary DNA amplification (Table 1). Real-time PCR was performed using an iCycler (Bio-Rad) and the IQSYBR Green Supermix (Bio-Rad). Reagents for real-time PCR analysis of IGFBP1, cystatin C, peroxiredoxin3, calpain10, and GAPDH (Assays-on-Demand, TaqMan Universal PCR Master mix) were purchased from Applied Biosystems (Foster City, CA). Amplification and detection of those specific products were performed with the ABI PRISM 7000HT Sequence Detection System (Applied Biosystems).The amount of each transcript was expressed as the n-fold difference relative to the control gene GAPDH (2ΔCt, where ΔCt represents the difference in threshold cycle between the control and target gene).

Table 1. Primers for RT-PCR
GeneSense Primer (5′ → 3′)Antisense Primer (5′ → 3′)

Terminal Deoxynucleotidyl Transferase-mediated Nick-End Labeling Assay.

The terminal deoxynucleotidyl transferase-mediated nick-end labeling (TUNEL) test was performed using the In Situ Cell Death Detection Kit (Boehringer Mannheim, Mannheim, Germany) according to manufacturer's instructions.

Western Blot Analysis.

Homogenates from liver samples were subjected to Western blot analysis as described.9 Antibodies used were phosphorylated STAT-3 (Tyr705) from Cell Signaling (Beverly, MA) and STAT-3 (Upstate Biotechnology, Charlottesville, VA), Bcl-2 y Mcl-1 from Santa Cruz Biotechnology (Santa Cruz, CA), and GAPDH from biogenesis (Poole, UK). Densitometric analyses were performed by chemiluminescence detection using the ImageQuant ECL system and the ImageQuant TL software (GE Healthcare Life Sciences, Buckinghamshire, UK). The expression levels are presented as fold increase versus control (=1) using the ratios between the densities of the antiapoptotic proteins bands to the corresponding GAPDH bands and P-STAT-3 band to the corresponding STAT-3 band.

cDNA Microarrays-based Gene Expression Analysis.

Microarray hybridization was performed at Progenika Biopharma Inc. (Bilbao, Spain) using tools obtained from Affymetrix and according to the manufacturer's protocol (Affymetrix, Santa Clara, CA). Six Affymetrix Mouse Genome 430 2.0 were used to analyze three livers from WT and three livers from CT-1−/− mice. Analysis of the gene expression data and functional classification according to the ontological criteria of the Gene Ontology Consortium13 thus generated was performed with GARBAN (

Statistical Analysis.

The normality of quantitative variables was assessed on residuals with Shapiro-Wilks test. When necessary, homoscedasticity was evaluated with Levene test. Normally distributed variables are represented as mean ± standard deviation (SD), and differences between groups were determined by ANOVA test followed by Dunnett test (to compare all treated groups vs. control) or orthogonal contrasts. Statistical analysis for the densitometric data were performed using a one-sample t test to analyze significant differences between specific groups to the control (reference value = 1) and two independent samples t test to analyze significant differences between the Jo-2–treated animals (WT vs. CT-1−/−). Non-normally distributed variables are expressed as median and interquartile range, and differences between groups were analyzed by the Kruskal-Wallis test followed by the Mann-Whitney U test. Survival curves were calculated using the Kaplan-Meier method and compared using the log-rank test. All statistical analyses were carried out with SPSS (SPSS Inc., Chicago, IL). All P values were two-tailed.


Expression of CT-1 in Isolated Liver Cells and in Liver Tissue.

Parenchymal and nonparenchymal liver cells were isolated from normal mice to analyze CT-1 expression in these cell populations. We found that both hepatocytes and nonparenchymal liver cells expressed CT-1 mRNA at approximately the same levels. This was in contrast with IL-6, which was found to be expressed mainly by nonparenchymal cells with negligible contribution of hepatocytes (Fig. 1A). Differences were seen with respect the expression pattern of these two cytokines in response to liver damage induced by Fas agonistic monoclonal antibody (Jo-2). The administration of a sublethal dose of this antibody (0.15 μg/g) was followed by elevation of serum aminotransferases, which reached high values at 10 hours after Jo-2 injection (Table 2). Together with this, we observed a brisk up-regulation of IL-6 mRNA at 10 hours after Jo-2 challenge (Table 2), whereas the levels of CT-1 mRNA tended to decrease after the apoptotic insult; the values at 6, 10, and 24 hours were significantly lower than basal levels (Fig. 1B). These findings indicate that hepatocellular damage may reduce the expression of CT-1 in the liver because of loss of hepatocytes, in keeping with the contribution of parenchymal liver cells to the hepatic synthesis of this cytokine.

Figure 1.

Expression of CT-1 and IL-6 mRNA in isolated liver cells and changes in hepatic CT-1 expression after Fas ligation. (A) Gene expression was assessed by real-time PCR in parenchymal and nonparenchymal cells (NPCc). (B) Gene expression of CT-1 in the livers of wild-type mice at various time points after Jo-2 antibody injection as determined by real-time PCR. The amount of the transcript is expressed as the n-fold difference relative to the control gene GAPDH (2ΔCt, where ΔCt represents the difference in threshold cycle between the control and target gene). Results are expressed as means ± SD of one representative experiment with six mice in each time point. *P < 0.05 versus control mice (C).

Table 2. Serum Aminotransferases and IL-6 mRNA Levels in the Livers of Mice After Administration of Fas-Agonist Antibody Jo-2 (0.15 μg/g) at Different Time Points
AnalyteBasal (0 hours)2 Hours4 Hours6 Hours10 Hours24 Hours
  • NOTE. Results are expressed as median (IQR).

  • *

    P < 0.05 versus basal values (Kruskal-Wallis followed by Mann-Whitney U test).

AST (U/L)130 (120–170)250 (210–370)335 (230–680)600 (217–1265)15515 (210–29582)*1180 (200–1875)
ALT (U/L)60 (50–130)140 (110–270)160 (97–322)275 (87–982)13775 (195–24155)*1145 (225–2550)
IL-6 mRNA (arbitrary units)2.0 (1.4–2.8)2.3 (2.0–5.7)4.5 (3.3–6.9)3.9 (2.1–18.5)17.3 (4.3–20.0)*1.1 (0.7–1.5)

Decreased Survival and Increased Hepatocyte Apoptosis in CT-1−/− Mice After Injection of Anti-Fas Antibody.

We previously reported that the administration of CT-1 exerts a marked anti-apoptotic effect on hepatocytes.9 To determine whether CT-1 may be a critical endogenous factor in the natural defense of the liver against proapoptotic stimuli, we determined the sensitivity of mice lacking CT-1 to Fas-induced apoptosis. To this aim we injected CT-1−/− and WT mice with a single intraperitoneal lethal dose (0.4 μg/g) of Jo-2. Although the lethality was similar in the two groups of animals (6/6) at 24 hours after Jo-2 injection, CT-1−/− mice died faster (Fig. 2A), showing 50% mortality at 4 hours after challenge whereas WT animals were still alive at this time point (P = 0.006).

Figure 2.

Increased susceptibility to Fas-mediated liver injury of CT-1−/− mice. (A) Survival of WT and CT-1−/− animals after a lethal dose of Jo-2 monoclonal antibody (0.4 μg/g weight) (P = 0.006). (B) Survival after sublethal dose of Jo-2 (0.15 μg/g weight) of WT, CT-1−/− mice (P = 0.029) and CT-1−/− animals treated with recombinant CT-1 (rCT-1) 30 minutes before challenge. (C) Serum aminotransferases in WT and CT-1−/− mice at 6 and 10 hours after sublethal dose of Jo-2 antibody. (D) Hematoxylin-eosin staining (H&E; original magnification ×200) and TUNEL assay (original magnification ×200) of liver biopsies of WT and CT-1−/− mice after sublethal dose of Fas agonist. (E) Representative Western blot of basal levels of Fas receptor (R) in the liver of WT and CT-1−/− mice. *P < 0.001 versus WT at the same time point after Jo-2 administration.

Next we analyzed the effect of a sublethal dose of Jo-2 antibody (0.15 μg/g) on the survival of WT and CT-1−/− mice (n = 6). With this lower dose of Jo-2, all WT animals survived whereas 50% of CT-1−/− mice were dead at 24 hours after challenge (P = 0.029) (Fig. 2B). To test whether the higher sensitivity of CT-1−/− animals to Fas ligation was due to lack of CT-1 or to any other abnormality present in these mice, we tested whether the administration of recombinant CT-1 could protect CT-1−/− mice from Fas-induced death. We found that CT-1−/− mice injected intravenously with a single dose of CT-1 (5 μg/mouse) 30 minutes before anti-Fas antibody (0.15 μg/g) showed 100% survival, thus behaving in the same way as WT animals (Fig. 2B).

We then evaluated the degree of hepatocellular damage in WT and CT-1−/− mice in response to the sublethal dose of Jo-2. We found that serum aminotransferases at 6 hours after Jo-2 administration were similar in WT and CT-1−/− mice, but at 10 hours the values of AST and ALT were much higher in CT-1−/− mice than in WT animals (P < 0.001) (Fig. 2C), again pointing to a higher susceptibility to Fas ligation of mice lacking CT-1. In agreement with the biochemical findings, histological examination of liver samples obtained at 10 hours after sublethal dose of Jo-2 showed a higher degree of liver injury and the presence of parenchymal hemorrhages in CT-1−/− mice (Fig. 2D). Similarly, TUNEL assay of the liver sections demonstrated more abundant apoptotic nuclei in CT-1−/− mice than in WT controls (Fig. 2D).

The different response to Jo-2 observed in WT and CT-1−/− mice was not attributable to reduced hepatic expression of Fas receptor, because the basal levels of Fas as determined by Western blotting were similar in the livers of the two groups of animals (Fig. 2E).

Deficient STAT-3 Activation on Fas Ligation in CT-1−/− Mice.

STAT-3 is regulated by IL-6 and related cytokines. Moreover, STAT-3 plays an essential role in cell protection against Fas-mediated liver injury.15 To investigate the molecular changes underlying the enhanced sensitivity of CT-1−/− mice to Fas-induced liver damage, we analyzed the activation of STAT-3 after a sublethal dose of Jo-2. We observed that STAT-3 phosphorylation (Tyr750) was induced at 6 hours after Fas ligation and it was maintained at 10 hours in both WT and CT-1−/− animals (Fig. 3A and data not shown). When we compared samples at 6 hours after challenge from animals with similar values of serum aminotransferases, we observed that CT-1−/− mice showed at this time point a reduction of activated STAT-3 (Fig. 3A, B). We also found that administration of recombinant CT-1 induced STAT-3 phosphorylation at similar levels in WT and CT-1−/− mice both in basal situation and after Fas ligation (Fig. 3A, B). Because suppressor of cytokine signaling-3 (SOCS-3) is a target gene for activated STAT-3 and a marker of STAT-3 transcriptional activity,16 we determined its expression. We found that SOCS-3 mRNA correlated directly with the level of STAT-3 activation showing lower up-regulation after Fas ligation in CT-1−/− as compared with WT animals (Fig. 3C). This finding corroborates the existence of a defective STAT-3 response on anti-Fas challenge in CT-1−/− mice. Because both WT and CT-1−/− mice showed similar values of IL-6 mRNA with the same liver damage at 6 hours after Jo-2 administration (Fig. 3D), and this cytokine has been reported to be responsible for the STAT-3 activation after hepatectomy and after CCl4 treatment,17 IL-6 cannot induce full STAT-3 activation in the absence of CT-1 in this model of liver damage.

Figure 3.

Defective STAT-3 activation in CT-1−/− mice after Fas challenge. (A) Representative Western blot for STAT-3 phosphorylation (Tyr705) in the livers from WT (+/+) and CT-1−/− (−/−) animals in basal situation, 6 hours after administration of rCT-1 (5 μg/mouse), 6 hours after Jo-2 administration (0.15 μg/g) and 6 hours after Jo-2 in animals given rCT-1 30 minutes before Jo-2 challenge. CT-1−/− mice exhibit deficient STAT-3 activation after Fas challenge but show high levels of phosphorylated STAT-3 when treated with rCT-1. (B) Results from the densitometric analysis are presented as median and interquartile range. (C) SOCS-3 mRNA and (D) IL-6 mRNA were determined by quantitative real-time PCR in the livers from WT (+/+) and CT-1−/− (−/−) mice in controls (C) and at 6 hours after Jo-2 administration. Values are means ± SD of five animals from the representative experiment (experiment performed 3 times). *P < 0.05 versus control (C). **P < 0.05 between the 2 groups.

CT-1 Administration Attenuates Fas-mediated Liver Injury in WT and CT-1−/− Animals.

To determine whether CT-1 can limit the extent of liver injury, we examined the effect of CT-1 administration in mice challenged with Jo-2 antibody. WT and CT-1−/− mice (n = 5) were pre-treated with a single dose of intravenous CT-1 (5 μg/mouse) 30 minutes before a sublethal dose of Jo-2 and samples were collected 10 hours after challenge. We observed that CT-1 treatment decreased liver damage in the two groups of animals as manifested by a reduction of histopathological damage, diminished apoptotic nuclei in the liver sections as estimated by the TUNEL technique, and a marked decrease of serum aminotransferases (Fig. 4A–D).

Figure 4.

CT-1 administration prevents Fas-induced liver damage in WT and CT-1−/− animals. (A) Effect of CT-1 (5 μg/mouse) pre-treatment 30 minutes before Jo-2 injection in WT mice. Liver damage was analyzed at 10 hours after Jo-2 injection (0.15 μg/g) by serum aminotransferases and (B) liver histology (H&E staining and TUNEL assay; original magnification ×200) in liver sections. (C) Serum aminotransferases and (D) liver histology in CT-1−/− mice pretreated or not with rCT-1 given 30 minutes before anti-Fas administration (0.15 μg/g). Determinations were performed at 10 hours after Jo-2 administration. (E) Representative Western blot and densitometric analysis for Bcl-2 and Mcl-1 in liver extracts from WT (+/+) and CT-1−/− (−/−) mice under the same conditions. (n = 5 per group in all experiments). *P < 0.05 versus control (C). **P < 0.05 between the 2 groups.

In another set of experiments rCT-1 was injected at two different doses (5 and 10 μg/mouse) and at two different time points (3 and 6 hours) after sublethal dose of Jo-2 (n = 5 each group). Liver damage was analyzed at 10 hours after Fas challenge. Only the group that received 10 μg of CT-1 per animal at 3 hours after Jo-2 administration showed protection as assessed by serum aminotransferases (ALT, 8,850 ± 4,358 in controls given saline versus 1,187.5 ± 912.5 in rCT-1–treated mice) and histology (data not shown). These findings indicate that CT-1 is able not only to prevent Fas-induced apoptosis but also to protect against ongoing liver damage when given early after the insult.

Bcl-2 family proteins inhibit apoptosis induced by a variety of stimuli, including Fas-mediated apoptosis.18 We assessed by Western blotting the expression of anti-apoptotic proteins Bcl-2, Mcl-1, and Bcl-xL and of pro-apoptotic proteins Bax y Bak, 10 hours after injection of anti-Fas antibody in WT and CT-1−/− mice. No changes were observed in the levels of Bcl-xL, Bax y Bak (data not shown). However, in contrast with WT animals, we found that both Bcl-2 and Mcl-1 proteins tended to decrease on Fas ligation in CT-1−/− mice. Pretreatment of animals with recombinant CT-1 restored both Bcl-2 and Mcl-1 protein levels (Fig. 4E).

Differential Gene Expression in the Liver of WT and CT-1−/− Mice: Effect of Recombinant CT-1 on Liver Transcriptome.

To further delineate possible gene products that could be responsible for the increased sensitivity of CT-1−/− mice to Fas-induced apoptosis, livers from these animals were compared with WT by cDNA microarray-based gene expression analysis in basal conditions. For this purpose we used mRNA from three WT and three CT-1−/− mice. Microarray data were normalized and ANOVA-test was applied with a P < 0.001. Based on these criteria, CT-1−/− mice as compared with WT animals showed different values with 72 probe-sets of which 33 indicated down-regulation and 39 up-regulation of the corresponding genes (Table 3). Of these genes, nine have been reported to be involved in cell death/survival functions. These genes include: IGFBP1, calpastatin, tumor necrosis factor receptor 1 (TNFR1), peroxiredoxin3, fibrinogen, cartilage link protein 1 (crtl1), bone morphogenic protein receptor (BMP-receptor), calpain10, and protein tyrosine phosphatase 4a2 (Ptp4a2). We were able to validate by real-time PCR only the first 4 genes of the list, and we found that all of them showed lower values in mice lacking CT-1. We also observed that these four genes, except peroxiredoxin3, were up-regulated when the CT-1−/− mice were treated with recombinant CT-1 (5 μg/mouse) (Fig. 5A). Interestingly, administration of CT-1 also induced a 2.3-fold decrease of calpain10 mRNA in the CT-1 null mice (Fig. 5A).

Table 3. Differentially Expressed Probesets Identified in Livers from CT-1 Knockout Animals, with P Value < 0.001.
Gene NameFoldΔUnigene No.
  1. NOTE. The fold change column indicates the ratio between knockout and the wild type. The expression values used in the fold change calculation are the mean expression levels of each class.

Down-regulated probesets
Dynein, axonemal, intermediate chain 10.065Mm.79127
F-box protein FBX150.077Mm.28369
Cartilage link protein 1 (Crtl1)0.094Mm.266790
Insulin-growth factor binding protein 10.108Mm.21300
AMP-activated, gamma 3 non-catatlytic subunit0.138Mm.166501
Somatostatin receptor 50.188Mm.353282
ATPase, cass I, type 8B, member 30.198Mm.52511
Similar to retinoblastoma-binding protein 90.210Mm.24216
Zinc finger, FYVE domain containing 270.288Mm.285369
Profilin 30.296Mm.348015
Mitogen-activated protein kinase 70.323Mm.3906
Cystatin C0.427Mm.4263
Sodium-calcium exchanger (Slc8a2)0.451Mm.241147
Lysosomal apyrase-like10.512Mm.291443
Monocyte to macrophage differentiation-associated0.535Mm.277518
Ligand of numb protein X2 (lnx2)0.586Mm.272203
Spactic paraplegia 4 homolog0.586Mm.19804
WD repeat domain 370.592Mm.284654
Protein tyrosine phosphatase, receptor-type, F0.612Mm.29855
Carboxypeptidase D0.618Mm.276736
Similar to phosphotidylinositol 4-kinase type II0.635Mm.117037
ATP-binding cassette, subf B (MDRTAP), member 40.638Mm.297825
Similar to mannan-binding lectin serine protease 20.672Mm.378962
Tumor necrosis factor I receptor (TNFR1)0.676Mm.235328
BMP receptor0.742Mm.142570
Guanine nucleotide exchange factor (Larg)0.756Mm.275366
Cysteine-rich protein 2 (Csrp2)0.772Mm.2020
Similar to solute carrier fam 17 (sodium phosphate)0.775Mm.24030
Similar to for protein disulfie isomerase-related0.827Mm.71015
Hippocampus abundant gene transcript 10.860Mm.280077
Fibrinogen beta chain0.899Mm.30063
Up-regulated probests
Transmembrane emp24-like trafficking protein 101.074Mm.381246
Sufeit gene 41.160Mm.300594
Adaptor-related protein complex 2, beta 1 subunit1.183Mm.378939
Low density lipoprotein B1.206Mm.25923
Nuclear export receptor for tRNAs1.293Mm.359408
Poly (A) binding protein, nuclear 11.337Mm.360551
Calpain 101.387 
Cofactor for Sp1 transcriptional activation, subunit 71.428 
Similar to phosphoinositol 3-phosphate-binding pr 21.507Mm.385225
ATP citrate lyase1.535 
Actinin alpha 11.718Mm.379778
RAB4A, member RAS oncogene family (Rab4a)1.943 
Amyloid beta (A4) precursor protein2.000Mm.34706
Small prolin-rich protein 2H2.037Mm.386997
Similar to phosphoinositol 3-phosphate-binding pr22.176Mm.25042
Protein tyrosine phosphatase 4a22.204Mm.352216
Bruno-like4, RNA binding protein2.292 
Similar to WD40 and FYVE-domain containing pr22.853Mm.253564
Nicotinamide nulceotide transhidrogenase3.129 
ATROIN1 (Atrogin1)3.204 
High mobility group b23.231Mm.327214
Kv channel-interacting protein 23.294Mm.358843
Interferon alpha family, gene 73.761Mm.208004
Ankyrin repeat domain 124.491Mm.378962
Leucine-rich repeat transmembrane neuronal 1 (Lrrtm1)4.520 
Phospholipase A2, group VII (Pla2g7)4.740Mm.30649
Glycogen synthase 15.293 
Solute carrier family 12, member 15.897Mm.275654
Proline-serine-threonine phosphatase-interacting p18.622Mm.270783
Lysozyme-like 18.645 
Retinitis pigmentosa 9 homolog (human) (Rp9h)9.235Mm.370845
Adaptor-related protein complex 1, sigma 2 subunit9.440 
Gene model 106, (NCBI) (Gm106)10.875 
Glucagon-like peptide 1 receptor11.572 
Nfia protein mRNA12.219Mm.10693
CytochromeP450, family 2, subfamily c3958.476Mm.369187
Figure 5.

Genes with cell death/survival functions differentially expressed in CT-1−/− mice that are modulated by CT-1 administration. Effect of IGBP1 and calpain inhibitor, MDL28170, on Fas-induced liver injury in CT-1−/− animals. (A) Changes in the expression of IGF-binding protein 1, calpastatin, peroxiredoxin 3, TNFR1, and calpain10, 2 and 6 hours after CT-1 administration (5 μg/mouse) to CT-1−/− mice. mRNA levels were quantified in the liver by real-time PCR and expressed as fold increased versus basal levels. (B) Serum aminotransferases at 10 hours after Jo-2 administration (0.15 μg/g) to CT-1−/− mice that received saline (n = 6) or recombinant IGFBP1 (n = 7) 30 minutes before challenge. (C) Serum aminotransferases and (D) liver histology at 10 hours after Jo-2 administration (0.15 μg/g) to CT-1−/− mice that received vehicle (DMSO) (n = 6) or the calpain inhibitor MDL28170 (n = 8) 1 hour after anti-Fas administration. *P < 0.05 versus DMSO-treated animals.

Of the genes down-regulated in CT-1−/− mice that are induced on recombinant CT-1 administration, TNFR1 may convey protective and pro-regenerative signals to the cell.19 IGFBP1 has been reported to reduce the levels of proapoptotic signals20 and calpastatin is a potent antiprotease that inhibits calpain. The involvement of calpain in pathological conditions has been well established in ischemia–reperfusion and anoxia-induced liver injury21 and also in the progression of acute liver injury initiated by hepatotoxicants.22 To investigate the role of these genes in the sensitivity to Fas-induced apoptosis observed in CT-1−/− mice, we challenged these animals with Jo-2 in the presence or absence of recombinant IGFBP1 or of the calpain inhibitor MDL 28170. We found that the recombinant IGFBP1 at the dose of 0.3 μg/g body weight (a dose that protects IGFBP1-deficient mice against Fas-induced apoptosis20) given 30 minutes before Jo-2 injection (0.15 μg/g) did not induce any protective effect; the levels of serum aminotransferases at 10 hours after anti-Fas challenge were similar to those of the vehicle-pretreated animals (Fig. 5B). In contrast, when CT-1−/− mice were pretreated with MDL 28170, the number of apoptotic liver cells (as assessed by H&E staining and TUNEL assay) and the rise of serum aminotransferases after Fas ligation were significantly reduced (Fig. 5C, D). Our data suggest that the failure of CT-1−/− mice to control calpain during the apoptotic process of hepatocytes may contribute to the higher susceptibility of these cells to undergo cell death on Fas ligation.


Many different hepatotoxic insults converge at inducing hepatocyte apoptosis. Fas is an important death receptor that conveys proapoptotic signals in diverse forms of liver injury including alcoholic hepatitis,23 non-alcoholic steatohepatitis,24 cholestatic liver injury,23 fulminant hepatitis,25 and T-cell–mediated hepatocellular damage in patients with viral or autoimmune hepatitis.26 Defensive mechanisms that protect liver cells against excessive proapoptotic stimuli are crucial for the preservation of tissue homeostasis. Various growth factors including hepatocyte growth factor,27 ligands of epidermal growth factor receptor such as amphiregulin and HB-EGF,28, 29 and cytokines such as IL-617 have been shown to exert hepatoprotective activities. Previous work from our laboratory showed that administration of recombinant CT-1 markedly ameliorates liver damage.9 However, little is known on the physiological role of CT-1 as part of the natural protective mechanisms of the liver against injury. Thus, in the current work we analyzed the role of endogenous CT-1 in liver defense against apoptosis.

As indicated in results, CT-1 is expressed at similar levels by both hepatocytes and nonparenchymal liver cells whereas IL-6 is produced exclusively by nonparenchymal cells. The dissimilarity in the cellular origin of the two cytokines may explain the differences observed in the expression of IL-6 and CT-1 after Fas ligation. Whereas Jo-2 injection was followed by a brisk elevation of IL-6 mRNA values, the levels of CT-1 mRNA tended to decrease. The fall of the latter cytokine in this model of hepatocellular damage is consonant with the contribution of hepatocytes to the synthesis of CT-1.

Our data show that CT-1−/− mice died faster after lethal doses of Jo-2 and developed more intense liver damage and higher mortality after sublethal doses of the Fas agonist. In parallel, these animals showed decreased hepatic levels of the antiapoptotic Bcl-2 protein after Fas challenge. Together, these findings reveal an enhanced susceptibility of CT-1−/− mice to Fas ligation, indicating that CT-1 is an important player in the natural defense of the liver against proapoptotic insults.

To define the molecular basis of the vulnerability of CT-1–deficient mice to Jo-2 administration, we analyzed the activation of STAT-3 after injury. STAT-3 is induced by the members of IL-6 family of cytokines30 and constitutes an essential component of survival pathways that defend hepatocytes against Fas-mediated apoptosis.15 Our data show that on Fas challenge CT-1−/− mice exhibit reduced STAT-3 activation in the liver when compared with WT animals. This is manifested by decreased STAT-3-Tyr750 and diminished mRNA values of SOCS-3, a known target gene of activated STAT-3.16 Interestingly, CT-1−/− mice failed to activate STAT-3 adequately after Fas ligation despite the fact that IL-6 was induced at levels similar to those found in WT animals. It therefore seems that CT-1 is an indispensable factor for full activation of STAT-3 in the Fas-induced liver damage.

Importantly, we found that the administration of recombinant CT-1 induced protection against Fas-induced cell death in both WT and CT-1−/− mice, an effect that was accompanied by enhanced activation of STAT-3. These findings suggest a potential therapeutic application of CT-1 in the prevention of liver damage. Activation of STAT-3 has been shown to increase both Bcl-2 and Mcl-1.31, 32 Thus, the reduced levels of those proteins observed in the CT-1−/− mice after Fas ligation could be explained by the defect in the activation of STAT-3.

We found that treatment with recombinant CT-1 was able not only to prevent Fas-induced apoptosis but it also could reduce liver damage when given up to 3 hours after Fas ligation. This suggests that this cytokine can block the apoptotic process when administered early after injury, a property that may increase its therapeutic potential.

To get further insight into the gene products that mediate the enhanced sensitivity of CT-1−/− animals to Fas-induced apoptosis, we compared the hepatic basal transcriptome of WT and CT-1−/− mice using microarray analysis. Of the genes that were differentially expressed in the two groups of animals, at least nine have been reported to mediate either cell protection or cell injury. Comparing WT and CT-1−/− mice we could confirm by real-time PCR analysis significant differences in the expression levels of four of these genes, namely TNFR1, IGFBP1, calpastatin, and peroxiredoxin3. All these genes were down-regulated in CT-1−/− mice and all of them except peroxiredoxin3 were induced by administration of recombinant CT-1.

TNFR1 has been shown to be important in maintaining normal liver homeostasis. Thus, signaling through this receptor can initiate hepatocyte proliferation, tissue regeneration, and the acute-phase response.19, 33 In fact, TNFR1−/− mice exhibit increased sensitivity to hepatotoxic agents.34 Accordingly, low levels of this receptor in CT-1−/− mice may alter the balance between proapoptotic and antiapoptotic intracellular pathways in favor of the former, thus enhancing Fas-induced damage. IGFBP1, conversely, has been shown to be up-regulated by STAT-335 and to reduce the amplification of proapoptotic signals during Fas-mediated liver injury, mainly by preventing the secondary activation of transforming growth factor beta that occurs in the liver on Fas ligation.20 Despite this important function, administration of recombinant IGFBP1 to CT-1−/− mice challenged with Jo-2 was not able to attenuate liver damage, indicating that down-regulation of IGFBP1 by itself is not a critical factor for the enhanced sensitivity of these mice to Fas ligation. Another gene that was down-regulated in CT-1−/− mice was calpastatin, the endogenous inhibitor of calpain. Interestingly, administration of CT-1 to CT-1−/− mice increased the expression of calpastatin and decreased the levels of the cytoplasmic protease calpain. Recently, ischemic preconditioning, a protective maneuver that activates STAT-3 in the liver, has been shown to be associated with down-regulation of calpain10.36 Accumulating evidence shows that calpain is a critical mediator of apoptosis in different models of liver damage, including anoxia, ischemia–reperfusion,21, 37 and in the progression of acute liver injury after hepatotoxicants.22 During apoptosis, activation of specific proteases results in degradation of cellular components leading to cell death. For many years it was believed that caspases were the only enzymes responsible for the proteolytic cascade in apoptosis. However, evidence shows that other proteases such as lysosomal proteases and calpains are involved in the initiation or execution of the apoptotic program.38 Liver cell injury causes lysosome instability but also provokes a rise of cytosolic calcium leading to calpain activation. Low levels of calpastatin have been shown to facilitate activation of calpain during apoptosis.39 Calpain degrades a large number of cytoskeletal proteins and contributes to the amplification of hepatocellular damage by attacking membrane proteins of the neighboring cells.22 The role of deregulated calpain activity in the enhanced sensitivity to hepatocyte apoptosis of CT-1−/− mice is supported by the effect of the calpain inhibitor in protecting theses animals against Fas ligation as reflected by a reduction in the number of apoptotic nuclei, parenchymal hemorrhage, and serum aminotransferases. Together, these findings indicate a novel role of calpain in Fas-mediated liver damage and suggest that the procurement of adequate gene expression of calpain and calpastatin is one of the biological protective functions of CT-1 in the liver.

In summary, our data demonstrate that CT-1 deficiency is accompanied by a marked enhancement of the sensitivity of the liver to Fas-induced apoptosis, indicating that this cytokine plays an essential role in liver homeostasis.