Article first published online: 24 JUN 2011
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 54, Issue 1, pages 185–195, July 2011
How to Cite
Nakagawa, H., Hirata, Y., Takeda, K., Hayakawa, Y., Sato, T., Kinoshita, H., Sakamoto, K., Nakata, W., Hikiba, Y., Omata, M., Yoshida, H., Koike, K., Ichijo, H. and Maeda, S. (2011), Apoptosis signal-regulating kinase 1 inhibits hepatocarcinogenesis by controlling the tumor-suppressing function of stress-activated mitogen-activated protein kinase. Hepatology, 54: 185–195. doi: 10.1002/hep.24357
Potential conflict of interest: Nothing to report.
S.M. and M.O. were supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan (#17209026 and #19390205). K.T., T.S., and H.I. were supported by Strategic Approach to Drug Discovery and Development in Pharmaceutical Sciences, Global Center of Excellence Program, Grant-in-Aid for Scientific Research (S) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation. K.K. was supported by Health Sciences Research Grants of the Ministry of Health, Labour and Welfare of Japan (Research on Hepatitis).
- Issue published online: 24 JUN 2011
- Article first published online: 24 JUN 2011
- Accepted manuscript online: 12 APR 2011 08:17AM EST
- Manuscript Accepted: 2 APR 2011
- Manuscript Received: 22 DEC 2010
The stress-activated mitogen-activated protein kinases (MAPKs), c-Jun NH2-terminal kinase (JNK), and p38 have been implicated in hepatocarcinogenesis. Although the many interrelated functions of JNK and p38 are precisely regulated by upstream signaling molecules, little is known about upstream regulators. We investigated the role of apoptosis signal-regulating kinase 1 (ASK1), a major player in the regulation of JNK and p38 activities, in hepatocarcinogenesis using a mouse hepatocellular carcinoma (HCC) model. ASK1-deficient (ASK1−/−) and wildtype (WT) mice were treated with diethylnitrosamine on postnatal day 14. Strikingly, after 7 months, approximately three times as many tumors developed in ASK1−/− mice as in WT mice. Although JNK and p38 activation were attenuated in ASK1−/− HCCs relative to WT HCCs, cell proliferation was comparable in HCCs from both types of mice. On the other hand, both cancer cell apoptosis and hyperphosphorylation of BimEL, a proapoptotic Bcl-2 family member, were suppressed in the ASK1−/− HCCs. ASK1−/− mice showed remarkable resistance to Fas-induced hepatocyte apoptosis in vivo, probably because of attenuated JNK-mediated BimEL phosphorylation and mitochondrial apoptotic pathway activation. The reintroduction of ASK1 to ASK1−/− mouse liver using an adenoviral vector restored Fas-induced hepatocyte death and phosphorylation of JNK and BimEL. Similar findings were obtained in tumor necrosis factor alpha-induced hepatocyte apoptosis. Furthermore, ASK1 was involved in DNA damage-induced p21 up-regulation through a p38 pathway. Conclusion: ASK1 is involved in death receptor-mediated apoptosis and DNA-damage response by way of stress-activated MAPK in the liver, and thus acts as a tumor suppressor in hepatocarcinogenesis. This study provides new insight into the regulation of stress- activated MAPK signaling in hepatocarcinogenesis. (HEPATOLOGY 2011;)
Hepatocellular carcinoma (HCC) is the third most common cause of cancer mortality; thus, understanding the molecular carcinogenic mechanism is an important issue.1 Several molecular pathways have been reported to play important roles in hepatocarcinogenesis. In particular, clinical and experimental studies have implicated the stress-activated mitogen-activated protein kinase (MAPK) cascades that converge on c-Jun NH2- terminal kinase (JNK) and p38 as key regulators of hepatocarcinogenesis.2-6
JNK and p38 have complex functions and modulate a wide range of cellular effects, including apoptosis, proliferation, differentiation, migration, and inflammation.7 Evidence implicating the JNK and p38 signaling pathways in the development of various types of cancer is strong, although certain cells use these signaling pathways to combat cancer development, whereas others use these pathways as cancer promoters.8, 9 Crosstalk between the JNK and p38 pathways further complicates the roles of these pathways in carcinogenesis.7 Although determining the mechanisms regulating these complex and multifunctional signaling pathways is essential for the development of new therapeutic approaches, these mechanisms are not yet well understood.
The activities of JNK and p38 are tightly regulated by upstream MAPK kinases and MAPK kinase kinases (MAP3Ks). Acting far upstream in the intracellular MAPK signaling cascade, MAP3Ks respond to intracellular and extracellular stimuli and determine cell fate.10 Apoptosis signal-regulating kinase 1 (ASK1), a ubiquitously expressed MAP3K, selectively activates the JNK and p38 signaling pathways in response to a variety of stimuli, including reactive oxygen species and cytokines, and has been widely accepted as a major player in the modulation of JNK and p38 activities regulating cell death.11 In liver disease, ASK1 is involved in acetaminophen-induced acute liver injury.12 Furthermore, recent reports revealed that ASK1 participates in colon and skin cancer development through the regulation of apoptosis and inflammation.13, 14 However, involvement of ASK1 in hepatocarcinogenesis has not been reported.
In this study we examined whether ASK1 plays a role in hepatocarcinogenesis using a diethylnitrosamine (DEN)-induced mouse HCC model. We found that ASK1 deficiency promoted the development of HCC, and ASK1 inhibited hepatocarcinogenesis by controlling the tumor-suppressing function of stress-activated MAPK.
Materials and Methods
Male ASK1-deficient (ASK1−/−), JNK1−/−, JNK2−/−, and C57BL/6 wildtype (WT) mice (Clea Japan, Tokyo, Japan) were used in the experiments. ASK1−/−, JNK1−/−, and JNK2−/− mice were generated as described12, 15 and backcrossed into the C57BL/6 strain at least 14 times. Mice were maintained under conventional conditions under a light/dark cycle.
All of the experimental protocols were approved by the Ethics Committee for Animal Experimentation and conducted in accordance with the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals.
Drug Administration and Experimental Design in an In Vivo Model.
In the DEN-induced HCC model, DEN (Sigma, St. Louis, MO) dissolved in phosphate-buffered saline (PBS) was injected intraperitoneally into mice (25 mg/kg) on postnatal day 14. Mice were sacrificed after 7 months and their livers were removed and examined for visible tumors. In the DEN-induced acute liver injury model, mice were injected intraperitoneally with 100 mg/kg DEN. In the Fas-induced liver injury model, mice (8-10 weeks old) were injected intraperitoneally with the agonistic anti-Fas antibody Jo2 (0.4 μg/g body weight; BD Pharmingen, CA) dissolved in PBS. In the lipopolysaccharide (LPS)/D-galactosamine (GalN)-induced liver injury model, mice were injected intraperitoneally with LPS (20 μg/kg; Sigma) and GalN (1,000 mg/kg; Wako). Some mice were pretreated with JNK inhibitor SP600125 (25 mg/kg; Biomol, PA) or p38 inhibitor SB203580 (25 mg/kg; Wako, Osaka, Japan) dissolved in PBS containing 10% dimethyl sulfoxide. Inhibitors were administered intraperitoneally 1 hour before Jo2 or DEN injection. Histological analyses, RNA extraction, real-time polymerase chain reaction (PCR), and generation of bone marrow chimeric mice were performed as described in the Supporting Information.
Cells and RNA Interference.
The human HCC cell lines HuH7 (Human Science, Tokyo, Japan) and PLC/PRF/5 (Riken, Tsukuba, Japan) and a human normal hepatocyte line (ACBRI, Kirkland, WA) were cultured in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum. Cell numbers were determined using a Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan). RNA oligonucleotides were synthesized by Qiagen (Hilden, Germany), and small interfering RNA (siRNA) transfections were performed using RNAiMAX (Invitrogen, Carlsbad, CA). Ultraviolet (UV) irradiation was performed using a UVB lamp (UVP, Upland, CA).
Immunoblotting and Coimmunoprecipitation Analysis.
Infection of Recombinant Adenovirus.
Recombinant adenoviruses encoding β-galactosidase (LacZ) and HA-tagged ASK1 (Ad-ASK1) were constructed as described.18 Adenoviruses were diluted in PBS and injected into the tail vein 48 hours before Jo2 administration (1 × 108 plaque-forming units [PFU]/mouse).
Statistical analyses were performed using Student's t test or analysis of variance (ANOVA), followed by Dunnett's test where appropriate. P < 0.05 was considered statistically significant.
Loss of ASK1 Accelerates Chemically Induced Hepatocarcinogenesis.
To determine the role of ASK1 in hepatocarcinogenesis, male WT and ASK1−/− mice were injected with 25 mg/kg DEN on postnatal day 14. After 7 months, untreated WT and ASK1−/− mice revealed no spontaneous liver dysfunction or tumor formation, whereas all mice given DEN developed typical HCCs. Strikingly, the number of detectable tumors was approximately three times higher in ASK1−/− mice than in WT mice, and the tumor- occupied areas were also more extensive in ASK1−/− mice than in WT mice (Fig. 1B,C). The maximum tumor size tended to be larger in ASK1−/− mice, but the difference was not statistically significant (Fig. 1B). DEN-induced liver tumors were histologically similar to well-to-moderately differentiated human HCCs, and the pathological characteristics of the tumors from WT and ASK1−/− mice were similar (Fig. 1C). Thus, loss of ASK1 accelerated DEN-induced HCC development.
Role of ASK1 in Cancer Cell Proliferation and Apoptosis.
We compared the characteristics of DEN-induced HCCs in WT and ASK1−/− mouse livers. The phosphorylation level of JNK, but not of p38, was higher in HCCs than in nontumor tissues, and JNK and p38 phosphorylation levels were lower in ASK1−/− HCCs than in WT HCCs (Fig. 2A). However, important downstream substrates of stress- activated MAPK involved in cell-cycle and tumor promotion, such as c-Jun and cyclin D1, were expressed at comparable levels in WT and ASK1−/− mice (Fig. 2A). Additionally, the frequency of cells positive for proliferating cell nuclear antigen (PCNA), a marker of cell proliferation, was similar for the WT and ASK1−/− HCCs (Fig. 2B). Because ASK1 appeared to be expressed at slightly higher levels in HCCs than in nontumor tissues (Fig. 3A), we examined whether ASK1 affects cancer cell proliferation in vitro by treating the HCC cell line HuH7 with ASK1-specific siRNA. ASK1-silencing decreased JNK phosphorylation (but not p38 phosphorylation) and c-Jun expression, decreased cyclin D1 expression slightly, and inhibited cell proliferation slightly (Fig. 3C,D), suggesting that the ASK1–JNK pathway weakly enhances HCC cell proliferation. A similar result was also observed in the PLC/PRF/5 HCC cell line (Fig. 3D). However, as discussed above, the WT and ASK1−/− HCCs exhibited similar c-Jun expression and cell proliferation rates in vivo, suggesting that other compensatory pathways promote c-Jun expression and cell proliferation in ASK1−/− HCCs. Based on these results, we conclude that the loss of ASK1 does not promote cancer cell proliferation and that there are other reasons for accelerated hepatocarcinogenesis in ASK1−/− mice.
Next, we compared the numbers of apoptotic cells in the WT and ASK1−/− mice livers using the terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay. As shown in Fig. 3A,B, significantly fewer apoptotic tumor cells were found in ASK1−/− HCCs than in WT HCCs. Consistent with this, caspase-3 activation was significantly attenuated in ASK1−/− HCCs (Fig. 3C). Messenger RNA (mRNA) levels for the death ligands tumor necrosis factor-α (TNF-α) and FasL and the death receptor TRAIL-R2/DR5 were higher in HCCs than in nontumor tissues, but did not differ significantly between WT and ASK1−/− HCCs (Fig. 3D). These findings indicate that death receptor pathways were activated in DEN-induced HCC tissues, but ASK1 does not regulate the expression of the main modulators. Furthermore, the expression levels of Bcl-2 families were almost identical in WT and ASK1−/− mice, as shown by western blot analysis (Fig. 3C). However, slower migration of the proapoptotic Bcl-2 family member BimEL band, indicating hyperphosphorylation of BimEL, was more predominant in WT HCCs than with ASK1−/− HCCs (Fig. 3C). JNK-mediated Bim phosphorylation has been reported to play an important role in death receptor-mediated apoptosis in the liver,19, 20 and defective death receptor signaling is considered to be a cause of tumor immune escape.21 Based on these results, we hypothesized that the loss of ASK1 might accelerate hepatocarcinogenesis by allowing cells to escape death receptor-mediated apoptosis.
ASK1 Is Involved in Fas-Induced Hepatocyte Apoptosis.
To evaluate whether ASK1 plays a role in Fas-mediated hepatocyte apoptosis, WT and ASK1−/− mice were injected intraperitoneally with agonistic anti-Fas antibody (Jo2), which causes severe liver damage through apoptotic Fas signaling. Because a recent report showed that the death pathway in hepatocytes loses its dependence on mitochondria when the cells are cultured on plates,22 we assessed the role of ASK1 in Fas-mediated apoptosis using an in vivo model. As shown in Fig. 4A, the liver from WT mice turned dark red at 5 hours after injection, which was indicative of widespread hemorrhage. In contrast, the liver from ASK1−/− mice showed only slight reddening. The histological examination revealed extensive hepatic apoptosis and hemorrhage in WT mice, but only focal apoptotic change in ASK1−/− mice (Fig. 4B,C). Consistent with these observations, serum alanine aminotransferase (ALT) levels in ASK1−/− mice were significantly lower than those in WT mice (Fig. 4D).
On the other hand, secondary inflammatory responses have been reported to modulate Jo2-induced liver injury.23 To rule out the possibility that ASK1 may be involved in Jo2-induced secondary inflammatory responses, we performed Jo2-induced liver injury experiments using bone marrow chimeric mice. WT mice transplanted with ASK1−/− or WT mouse-derived bone marrow cells showed similar extents of liver injury after Jo2 injection (Fig. 4E,F). These results suggest that ASK1 is involved in Fas-mediated direct hepatocyte apoptosis.
ASK1 Is Required for Activation of the JNK-Bim Pathway in Fas-Induced Apoptotic Signaling.
We observed ASK1 phosphorylation after Jo2 administration in WT mouse liver, suggesting that ASK1 was activated in Fas signaling in vivo (Fig. 5A). Expression levels of antiapoptotic proteins which have been reported to be implicated in Fas-induced liver injury were not affected by the absence of ASK1 (Fig. 5B). On the other hand, Jo2-induced JNK, p38, and caspase-3 activations were significantly attenuated in ASK1−/− mice compared with WT mice (Fig. 5B).
Bim is phosphorylated by JNK and subsequently cleaved by caspase-3, and becomes a hyperactive inducer of cytochrome c release, leading a positive amplification loop in apoptosis.24, 25 In western blot analysis of liver proteins, Jo2 injection induced slower migration of the BimEL band in WT mice, whereas the change in BimEL migration was significantly attenuated in ASK1−/− mice, as also seen in HCC tissues (Fig. 5B). Additionally, we analyzed the activation of the mitochondrial apoptotic pathway, which is essential for Fas-induced apoptosis of hepatocytes (so-called type II cells). The mitochondrial Bax translocation was slightly lower and the cytosolic release of cytochrome c was significantly reduced in ASK1−/− mice compared with WT mice at 2 and 3.5 hours after Jo2 administration (Fig. 5C). At 5 hours after Jo2 administration, marked phosphorylation and subsequent degradation of BimEL and reduction of the cytochrome c level in the mitochondrial fraction were seen in WT mice, whereas these changes were significantly suppressed in ASK1−/− mice (Fig. 5D). As reported,19 administration of a JNK inhibitor reduced Jo2-induced BimEL phosphorylation and serum ALT elevation. However, administration of a p38 inhibitor had no detectable effect on BimEL phosphorylation or liver injury (Fig. 5E,F). These results suggest that ASK1 plays an important role in Fas-induced activation of the JNK–Bim–mitochondrial apoptotic pathway.
Next, to examine whether ASK1 may be involved in a Fas-induced mitochondria-independent apoptotic pathway, we used primary thymocytes, which are independent of mitochondria for Fas-induced apoptosis (so-called type I cells). Fas-induced activation of JNK and p38 was reduced in ASK1−/− thymocytes, whereas caspase-3 activation and cell viability were comparable between WT and ASK1−/− thymocytes (Supporting Fig. 1A,B), suggesting that ASK1 is not required for the mitochondria-independent apoptotic pathway.
Recently, Fas signaling was reported to play a role in not only cancer cell apoptosis, but also cancer cell proliferation.26 JNK has also been shown to be one of the main mediators of Fas-mediated proliferative signals. To investigate whether ASK1 participated in Fas-mediated hepatocyte proliferation, we injected Jo2 to WT and ASK1−/− mice after partial hepatectomy, which is known to convert Fas signaling from apoptotic to proliferative.27 As reported,26, 27 Jo2 injection after partial hepatectomy induced JNK phosphorylation and accelerated hepatocyte proliferation without liver injury (Supporting Fig. 2A,B). Although liver regeneration after partial hepatectomy and Jo2-induced JNK phosphorylation were slightly impaired in ASK1−/− mice (especially the upper band corresponding to JNK2), there was no significant difference in Jo2-mediated acceleration of hepatocyte proliferation (Supporting Fig. 2A,B). Thus, ASK1 seemed to regulate the apoptotic, but not proliferative, function of JNK in Fas signaling.
Restoration of Fas Sensitivity in ASK1-Reintroduced ASK1−/− Mouse Liver.
To further confirm the involvement of ASK1 in Fas-induced hepatocyte apoptosis, we examined whether the reintroduction of ASK1 to ASK1−/− mouse liver restored sensitivity to Fas. We injected an adenoviral vector encoding either Ad-ASK1 or LacZ into the tail vein of ASK1−/− mice. ASK1 protein was successfully expressed in ASK1−/− mouse liver, as much as that in WT mouse liver, at 48 hours after Ad-ASK1 injection (Fig. 6A). Immunohistochemical analysis using anti-HA antibody revealed that ≈70%-80% of hepatocytes were transduced with the ASK1 gene (Fig. 6B). The reintroduction of ASK1 did not affect the serum ALT level or liver histology. These mice were injected intraperitoneally with Jo2 5 hours later; only mild serum ALT elevation and histological liver damage were found in LacZ-injected mice, whereas Ad-ASK1-injected mice revealed marked serum ALT elevation and severe histological damage (Fig. 6C,D). Furthermore, reintroduction of ASK1 restored Jo2-induced phosphorylation of JNK and BimEL in the liver (Fig. 6E).
Role of ASK1 in TNF-α-Induced Hepatocyte Apoptosis.
To examine whether ASK1 is required for TNF-α-induced apoptosis of hepatocytes in vivo, we used an LPS/GalN liver injury model that depends on TNF-α-induced apoptosis.28 At 6 hours after LPS/GalN administration, WT mice exhibited marked ALT elevation, severe histological liver damage, and hepatocyte apoptosis, whereas these changes were significantly attenuated in ASK1−/− mice (Fig. 7A-C). As expected, LPS/GalN-induced phosphorylation of JNK and BimEL and cleavage of caspase-3 were significantly attenuated in ASK1−/− mice, as well as in Fas-induced liver injury (Fig. 7D). On the other hand, WT and ASK1−/− mice exhibited no significant difference in serum TNF-α levels (Fig. 7E). These findings provide further support for the hypothesis that ASK1 is required for death receptor-mediated hepatocyte apoptosis by way of the JNK–Bim-mediated mitochondrial apoptotic pathway. Furthermore, ASK1 silencing by siRNA attenuated TNF-α-induced sustained JNK and p38 activation, BimEL cleavage, and apoptosis in the HCC cell line HuH7 (Supporting Fig. 3A,B). Thus, resistance to death signaling may be a predominant cause of accelerated hepatocarcinogenesis in ASK1−/− mice.
Involvement of the ASK1-p38 Pathway in DNA Damage Response.
Because DEN-induced acute phase reaction in the liver is known to be associated with future HCC development, we assessed the involvement of ASK1 in this phase.29 Although the DEN-induced activation of JNK was slightly attenuated in ASK1−/− mouse livers, the increases in serum ALT levels were statistically similar in the WT and ASK1−/− mice (Fig. 8A, Supporting Fig. 4A). Bromodeoxyuridine labeling revealed that the numbers of compensatory proliferating hepatocytes in WT and ASK1−/− mice were similar after DEN administration (Supporting Fig. 4B). Furthermore, the level of DEN-induced p53 activation was similar in both groups (Fig. 8A). These findings suggest that DEN induces a similar extent of hepatocyte death, DNA damage, and compensatory proliferation in WT and ASK1−/− mice.
On the other hand, p38 activation was significantly attenuated in the ASK1−/− mouse livers (Fig. 8A), and p38 has been reported to play an important role in DNA damage responses, such as cellular senescence, by inducing cyclin-dependent kinase inhibitors through p53-dependent and -independent mechanisms.30 Thus, we next compared induction of cyclin-dependent kinase inhibitors after DEN administration between WT and ASK1−/− mouse livers. As shown in Fig. 8B, p16 and p21 were slightly and remarkably induced after DEN administration, respectively, and p21 induction was significantly attenuated in ASK1−/− mouse livers. Because the p38 inhibitor, but not the JNK inhibitor, suppressed DEN-induced p21 up-regulation (Fig. 8C), we considered that the ASK1-p38 pathway may be involved in DNA damage-induced p21 up-regulation. However, in addition to DNA damage, there are many other kinds of stimuli in the liver after DEN administration in vivo: inflammatory responses, liver regeneration signals, and toxic metabolites of DEN. Thus, we assessed the role of the ASK1-p38 pathway in p21 induction after DNA damage using UVB irradiation, which is a well-known direct DNA damage-inducer. UVB irradiation to the immortalized human normal hepatocyte line induced strong phosphorylation of p38 and very weak phosphorylation of JNK, and ASK1 silencing attenuated UVB-induced p38 phosphorylation, especially in the late phase (Fig. 8D, Supporting Fig. 5). Furthermore, UVB-induced p21 up-regulation was attenuated by ASK1 silencing and p38 inhibition (Fig. 8E F). These results suggest that ASK1 is involved in DNA damage-induced p21 up-regulation through p38 activation, and an impaired DNA damage response may be one reason for increased hepatocarcinogenesis in ASK1−/− mice.
Dysregulation of the balance between cell proliferation and apoptosis plays a critical role in hepatocarcinogenesis.31 Our results suggest that ASK1 plays only a minor role in cancer cell proliferation and a major role in death receptor-mediated apoptosis in the liver through the JNK pathway. Loss of ASK1 appears to cause an imbalance that accelerates chemical hepatocarcinogenesis in ASK1−/− mice. Furthermore, ASK1 is involved in the DNA damage response, through the p38 pathway. This study provides new insight into the regulation of stress-activated MAPK signaling in hepatocarcinogenesis.
JNK (primarily JNK1) has been reported to promote DEN-induced hepatocarcinogenesis by promoting cancer cell proliferation and neovascularization.3, 5 Although JNK activation was attenuated in ASK1−/− mice, the phenotype of ASK1−/− mice and JNK1−/− mice was opposite in hepatocarcinogenesis.3, 5 We suggest that the reasons may be as follows: (1) Although we observed attenuation of JNK activation in ASK1−/− HCC tissues, ASK1 appears to play only a minor role in HCC cell proliferation (Fig. 2). Additionally, vessel density and VEGF expression in ASK1−/− HCC tissues were unaffected (Supporting Fig. 6A,B). Thus, the tumor-enhancing function of JNK1 seems to be preserved in ASK1−/− mice. (2) JNK has also been reported to act as a tumor suppressor by inducing cancer cell apoptosis.32 JNK1 and JNK2 isoforms have distinct or redundant roles in some situations, including apoptosis induction. JNK2, but not JNK1, has been reported to play a major role in TNF-α-mediated hepatocyte apoptosis in vivo.33 In our experiments using a Jo2-induced hepatitis model, the lack of neither JNK1 nor JNK2 resulted in a significant reduction in the ALT elevation or BimEL phosphorylation, unlike ASK1−/− mice or a pan-JNK inhibitor (Supporting Fig. 7A,B). These results suggest that the role of JNK1 and JNK2 in death receptor-mediated hepatocyte apoptosis may be redundant. Furthermore, a recent report demonstrated that JNK1 and JNK2 deficiency in hepatocytes increased DEN-induced HCC.34 Thus, JNK1 and JNK2 have a wide range and redundant or distinct functions, and upstream molecules, such as MAP3Ks, must regulate the complex functions of JNK. We consider that ASK1 plays major roles in tumor-suppressing part of JNK in hepatocarcinogenesis. However, knockdown of ASK1 in HCC cell lines slightly decreased cell proliferation. This finding suggests that ASK1 may weakly promote the proliferation of some HCC cells, which could explain why the WT and ASK1−/− mice did not exhibit significant differences in tumor size.
On the other hand, mice with liver-specific p38 deficiency exhibit increased HCC development similar to ASK1−/− mice.4, 6 The accelerated hepatocarcinogenesis in p38-deficient mice is reportedly attributable to compensatory JNK activation and cancer cell proliferation. Although p38 activation was attenuated in ASK1−/− mice, JNK activation was also attenuated, unlike the liver-specific p38-deficient mice. Thus, the mechanisms of accelerated hepatocarcinogenesis in ASK1−/− mice and liver- specific p38-deficient mice appear to differ. p38 has also been reported to play an important role in DNA damage responses, such as cellular senescence, by inducing cyclin-dependent kinase inhibitors.30 In this study, we showed that ASK1 is involved in DNA damage-induced p21 up-regulation through p38 activation. Furthermore, the ASK1-p38 pathway has been reported to have an inhibitory effect on malignant transformation of fibroblasts by triggering apoptosis in response to oncogene-induced reactive oxygen species (ROS).35 Thus, the ASK1-p38 pathway may play a key role in the inhibition of tumor initiation in hepatocarcinogenesis.
Defective death-receptor signaling is considered a cause of tumor immune escape, so understanding its apoptotic mechanism is very important not only from the point of view of carcinogenesis, but also for cancer therapeutics.21 Several in vitro studies have demonstrated that ASK1 is implicated in the TNF-α- and Fas-mediated apoptotic pathways,11, 36 but the in vivo role of ASK1 has not been determined. Our current findings provide the first evidence that ASK1 plays an important role in TNF-α- and Fas-mediated hepatocyte apoptosis in vivo and suggest that the JNK- Bim-mediated mitochondrial apoptotic pathway is an important downstream target of ASK1. JNK-mediated Bim phosphorylation triggers the proapoptotic activity of Bim by causing its release from sequestration to the microtubular dynein motor complex.25 Bim initiates the mitochondrial apoptotic pathway by activating Bax and Bak directly and indirectly blocking prosurvival Bcl-2 family members.37 Recent reports have shown that Bim plays an important role in Fas- and TNF-α-induced hepatocyte apoptosis.19, 20 Mitochondria are considered an important downstream target of ASK1 in apoptosis because overexpression of the active form of ASK1 in cell lines induces apoptosis through cytochrome c release from mitochondria and activation of caspase-9 and -3.38 In our study, ASK1 was found to be involved in Fas-induced hepatocyte apoptosis but not in thymocyte apoptosis, suggesting that ASK1 is required for mitochondria-dependent apoptosis. Thus, we believe that the ASK1–JNK–Bim–mitochondrial pathway plays an important role in death receptor-mediated hepatocyte apoptosis. The observed attenuation of Bim phosphorylation and caspase-3 activation in ASK1−/− HCC tissues is consistent with the inhibition of death receptor-induced apoptosis.
Recently, death-receptor signaling, such as Fas signaling, has been reported to play a role in not only cancer cell apoptosis, but also cancer cell proliferation.26 Our finding that Jo2-induced acceleration of hepatocyte proliferation after partial hepatectomy was comparable between WT and ASK1−/− mice suggests that ASK1 does not play a major role in Fas-mediated cell proliferation. Furthermore, the finding that WT and ASK1−/− HCCs exhibited no significant differences in cancer cell proliferation rates in vivo also supports this. Thus, ASK1 seemed to regulate the apoptotic, but not proliferative, function of JNK in Fas signaling, and ASK1−/− hepatocytes might alter death-receptor signaling to favor survival by escaping apoptosis. However, this is a relatively new concept, so further study is needed to clarify the role of ASK1 in death receptor-mediated cancer cell proliferation.
In conclusion, ASK1 controls the tumor-suppressing function of stress-activated MAPK signaling, and thus acts as a tumor suppressor in hepatocarcinogenesis.
Additional Supporting Information may be found in the online version of this article.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.