Skip to main content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Gastroenterology. Author manuscript; available in PMC 2014 Jul 7.
Published in final edited form as:
Gastroenterology. 2011 Aug; 141(2): 707–718.e5.
Published online 2011 May 4. doi: 10.1053/j.gastro.2011.04.051
PMCID: PMC4084974
NIHMSID: NIHMS294063
PMID: 21679710

Wnt–β-catenin Signaling Protects against Hepatic Ischemia and Reperfusion Injury in Mice

Nadja Lehwald,*,1,2 Guo-Zhong Tao,*,1 Kyu Yun Jang,1,3 Michael Sorkin,4 Wolfram T. Knoefel,2 and Karl G. Sylvester1,5
Author information Copyright and License information PMC Disclaimer

Associated Data

Supplementary Materials

Abstract

Background & Aims

Ischemia and reperfusion injury are common causes of oxidative tissue damage associated with many liver diseases and hepatic surgery. The Wnt–β-catenin signaling pathway is an important regulator of hepatic development, regeneration, and carcinogenesis. However, the role of Wnt signaling in the hepatocellular response to ischemia–reperfusion (I/R) injury has not been determined.

Methods

Hepatic injury following ischemia or I/R was investigated in hepatocyte-specific, β-catenin–deficient mice, as well as Wnt1 overexpressing and wild-type (control) mice.

Results

Wnt–β-catenin signaling was affected by the cellular redox balance in hepatocytes. Following ischemia or I/R, mice with β-catenin–deficient hepatocytes were significantly more susceptible to liver injury. Conversely, mice that overexpressed Wnt1 in hepatocytes were resistant to hepatic I/R injury. Hypoxia inducible factor (HIF)-1α signaling was reduced in β-catenin–deficient liver but increased in hepatocytes that overexpressed Wnt1 under hypoxia and following I/R, indicating an interaction between β-catenin and HIF-1α signaling in the liver. The mechanism by which Wnt signaling protects against liver injury involves β-catenin’s role as a transcriptional co-activator of HIF-1α signaling, which promotes hepatocyte survival under hypoxic conditions.

Conclusion

Cellular redox balance affects Wnt–β-catenin signaling, which protects against hypoxia and I/R injury. These findings might be used to develop strategies for protection of hepatocytes, regeneration of liver, and inhibition of carcinogenesis.

Keywords: TCF, LEF, liver damage, transplantation, reactive oxygen species, oxidative stress

Introduction

A common feature of many liver diseases as well as liver resection or transplantation is tissue hypoxia and reoxygenation that result in increased cellular reactive oxygen species (ROS), macromolecular damage and permanent hepatocellular injury1,2. Since hepatocellular metabolism requires uninterrrupted oxygen delivery for homeostasis, hepatocytes have evolved protective mechanisms in order to mitigate oxidative injury3.

The Wnt signaling pathway is an established molecular regulator of hepatic development, regeneration and carcinogenesis4–6. The canonical Wnt signaling pathway is regulated through post-translational modifications of β-catenin. Wnt signaling is initiated through Wnt ligand binding to two membrane bound receptors, Frizzled, and the co-receptor Lipoprotein receptor Related Proteins 5 and 6 (LRP-5/6)7,8. Wnt ligand binding disables GSK3β preventing β-catenin N-terminal serine/threonine phosphorylation resulting in β-catenin accumulation and translocation to the nucleus where it activates the transcription complex T cell factor/lymphoid enhancer factor (TCF/LEF). Additional post-translational modifications (i.e. acetylation) of β-catenin have been described that significantly affect its activity as a transcriptional regulator9,10. Previously, in vitro evidence has been provided suggesting that β-catenin can be diverted as a transcriptional activator from TCF/LEF to adaptive pathways like HIF-1α (hypoxia inducible factor-1α) in order to mediate the cellular response to oxidative stress11–14. HIF-1, a key regulator of hypoxia, is a heterodimer consisting of a hypoxia-stabilized α-subunit (HIF-1α) and a constitutively expressed β-subunit (HIF-1β). Under hypoxic conditions, stabilized HIF-1α translocates to the nucleus, dimerizes with HIF-1β and binds hypoxia response elements (HRE) to activate target genes to promote angiogenesis and cellular metabolic changes15–17.

Since both Wnt/β-catenin and HIF-1α signaling have established roles in regulating cell metabolism and survival, the present study was designed to investigate whether Wnt/β-catenin signaling is required for an effective response to ischemia-reperfusion (I/R) injury in the liver. Herein, we demonstrate that hepatocyte-specific β-catenin knockdown mice are significantly more sensitive to liver I/R injury resulting in severe necrosis and apoptosis. Conversely, we show that in mice with hepatocyte-specific Wnt/β-catenin stabilization, there is strong hepatic resistance to hypoxia and I/R injury. Together, these data provide the first in vivo evidence that β-catenin is a key component of an effective, tissue-specific reponse to I/R and that the molecular mechanism is mediated by cellular redox balance.

Material and Methods

Animals

To investigate the role of Wnt/β-catenin signaling in the liver, two hepatocyte-specific mouse models were created: for the β-catenin knockdown mouse, a previously described mouse possessing the liver enriched activator (LAP) promoter CEBP/β driving tetracycline-transactivating (tTA) was crossed with a mouse possessing a tetracycline response element driving Cre (tetO-Cre)18,19. Subsequent progeny were further bred with a β-cateninflox/flox mouse6. The resultant triple transgenic mouse LAP-tTA/tetO-Cre/β-cateninflox/flox (LT2-KD) demonstrated responsiveness to the tetracycline analog doxycycline (dox) for β-catenin deletion. For wild-type controls (LT2-WT), their littermates (LAP-tTA−/tetO-Cre−/β-cateninflox/flox, LAP-tTA−/tetO-Cre−/β-cateninwt/flox, LAP-tTA+/tetO-Cre−/β-cateninwt/wt) were used. Dox-water was given in all breeding cages and removed at the age of 6–8 weeks. After dox-withdrawal for 4 weeks mice were used for experiments. For the conditional Wnt1 overexpressing mouse, a tetracycline-responsive tetO-Wnt1-Luc mouse was crossed to the LAP-tTA mouse described above18–20. This double transgenic mouse LAP-tTA/tetO-Wnt1-Luc (Wnt1+) enables hepatocyte-specific activation of the transgene Wnt1-Luc in a dox-dependent manner. For experiments, sex-matched 6–8 weeks old littermates (LAP-tTA+/tetO-Wnt1-Luc+) were kept continually with (Wnt1-WT) or removed from dox-water (Wnt1+) for additional 3 weeks. Experiments were conducted under a protocol approved by Stanford University School of Medicine Institutional Animal Care and in accordance with NIH guidelines.

Liver Ischemia and Reperfusion Model

We used a murine model of 70% partial warm liver ischemia-reperfusion (75 minutes/6 hours) injury21. Details are provided in Supplementary data.

Cell Culture

Primary hepatocytes from 10–12 weeks old mouse livers were isolated using the two-step collagenase perfusion22. Hepatocytes were cultured in DMEM/F12 supplemented with 1% FCS.

The differentiated non-transformed mouse hepatocyte AML12 and human hepatocellular carcinoma HepG2 cell lines were cultured in either DMEM/F-12 or DMEM supplemented with 10% FCS23 at 37°C with 5% CO2. For hypoxia, cells were grown at 1% O2 in a custom-designed incubator (XVivo Hypoxia Chamber; BioSpherix) for 24 hours. In order to simulate in vivo liver I/R, cells were incubated in 1% hypoxia for 24 hours followed by reoxygenation (H/R) for 2 hours.

Plasmid and Transfection

To activate canonical Wnt/β-catenin signaling, we utilized the plasmid pcDNA3S33Y resulting in robust TCF-dependent transcriptional activation compared to wild-type β-catenin22. Stable S33Y mutants conferring β-catenin gain-of-function were derived through retroviral transfection in AML12 hepatocytes.

Electrophoretic Mobility Shift Assay (EMSA)

Nuclear extracts were prepared using the nuclear and cytoplasmic extraction kit (Thermo Scientific, Rockford, IL). DNA-protein binding reaction was performed using an Epo/HIF-1α probe according to the manusfacturer’s instructions (Panomics, Fremont, CA).

Cell Proliferation and Viablity Assays

BrdU ELISA and MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assays were performed according to the manufacturer’s instructions (Roche Applied Science, Indianapolis, IN).

Apoptosis and Necrosis Assays

TUNEL staining was performed using the In Situ Cell Death Detection Kit (Roche Applied Science) according to the manusfacturer’s instructions. K18D237 antibody (AnaSpec, Fremont, CA) was used for detecting caspase-cleaved Keratin 18 (K18) by immunoblot24. Caspase-Glo 3/7 assay (Promega, Madison, WI) was performed according to the manusfacturer’s instructions. Microscopic quantification analyses were performed in 10 high power fields (HPF) per specimen at 400×.

Luciferase Reporter Assay

Dual-Light Reporter Gene Assay (Applied Biosystems, Foster City, CA) was performed using the pMegaTOPFLASH, pMegaFOPFLASH, 5x-HRE or LacZ plasmids as previously described12,25.

Determination of intracellular ROS levels

Dichlorofluorescein diacetate (DCF-DA; Invitrogen, Carlsbad, CA) was used to monitor intracellular ROS generation. After incubation with 10 µmol/L DCF-DA for 30 minutes, cells were analyzed by flow cytometry26. In addition, in situ ROS detection was performed by dihydroethidium (DHE; Invitrogen) labeling of frozen liver sections with 3 µM DHE at 37°C for 30 minutes.

Statistics

All experiments were performed in at least triplicate unless otherwise indicated. Data are expressed as mean ± standard deviation, and evaluated by Student’s t test. Significance was defined as p<.05; „n.s.“ indicates not significant.

Results

Oxidative stress inhibits β-catenin/TCF signaling in hepatocytes

Since Wnt/β-catenin signaling and cellular redox balance are integral to liver homeostasis, we sought to determine the effect of ROS generation on Wnt signaling in hepatocytes. Exposure of AML12 cells to either hypoxia or H/R resulted in a significant decrease in β-catenin/TCF reporter activity without a change in total β-catenin protein (Fig.1A) or transcript level (not shown). However, β-catenin/TCF target gene expression was significantly reduced in response to hypoxia (supplementary Fig.S1A) in-line with the observed signal activity repression (Fig.1A). Moreover, the hypoxia-induced reduction in TCF signaling appeared to be ROS dependent as the antioxidant N-acetylcysteine (NAC), a ROS scavenger, was able to decrease intracellular ROS (Fig.1B) and diminish the suppressive effect of hypoxia-derived ROS on β-catenin/TCF activity (Fig.1C). Together, these data demonstrate that β-catenin signal transduction is significantly impacted by cellular redox balance.

An external file that holds a picture, illustration, etc. Object name is nihms294063f1.jpg
β-catenin/TCF signaling is impacted by cellular redox changes

  1. β-catenin/TCF reporter activity significantly decreases in response to 1% hypoxia for 24 hours or H/R (24/2 hours) in AML12 hepatocytes without a change in total β-catenin protein.
  2. Intracellular ROS levels in AML12 hepatocytes are significantly elevated by hypoxia as measured by DCF-DA flow cytometry. One hour pre-treatment with NAC (2 mmol/L) significantly prevents ROS production under hypoxia.
  3. One hour pre-treatment with NAC (2 mmol/L) prevents hypoxia-induced inhibition of β-catenin/TCF signaling in AML12 hepatocytes. RLU= relative light units. *p<0.05, **p<0.01.

Conditional β-catenin knockdown sensitizes the liver to ischemia-reperfusion injury

Previous studies have reported heightened hepatic sensitivity to injury stimuli in the absence of β-catenin27. However, the response to a clinically relevant source of ROS generation like hypoxia or I/R has not been evaluated. To determine β-catenin’s role in the adaptive response to I/R in vivo, we developed a novel mouse model for conditional and regulatable β-catenin deletion from hepatocytes in order to overcome compensatory changes that may occur when β-catenin is deleted during liver development in cells with transcriptionally active albumin (Alb-Cre)6. The triple transgenic mouse LAP-tTA/tetO-Cre/β-cateninflox/flox (LT2-KD) was engineered to conditionally delete β-catenin in response to dox-withdrawal (Fig.2A; supplementary Fig.S2A)18,19. Four weeks of dox-withdrawal resulted in a significant reduction in β-catenin expression in hepatocytes at mRNA and protein level (Fig.2A). Untreated LT2-WT and KD mice did not show any significant difference in transaminases or liver histology (supplementary Fig.S2B, Fig.S2C). To determine if β-catenin knockdown exacerbates oxidative injury in vivo, we subjected LT2-KD mice and littermates (LT2-WT) to either liver ischemia or I/R respectively. Liver injury as assessed by serum transaminases was significantly increased in KD mice after ischemia and dramatically exacerbated after I/R compared to LT2-WT mice (Fig.2B). The genotypic difference in liver damage was specific for ischemia and I/R, as transaminases were similar in sham-treated LT2-WT and KD mice (supplementary Fig.S2D). Histology (Fig.2C) of KD livers revealed severe hypoxic injury around the central vein with hepatocyte swelling, necrosis and karyolysis, while signs of ischemia were substantially attenuated in wild-type mice. H&E staining after I/R showed extensive hepatocellular injury with wide necrosis, congestion and swelling in KD hepatocytes. Conversely, LT2-WT livers showed relatively well-preserved histological architecture with less hepatocyte damage. TUNEL staining and quantification showed more severe apoptosis in both ischemia and I/R-challenged KD livers (Fig.2C, Fig.2D). Immunoblot for caspase-cleaved K18 also indicated increased hepatocellular apoptosis following I/R in LT2-KD mice (Fig.2D). Morphometry revealed necrosis in ~50% of I/R-subjected KD livers, while only ~25% of LT2-WT livers showed signs of necrosis (Fig.2E). DHE staining for the in situ detection of ROS (Fig.2C) revealed a pronounced increase in LT2-KD livers when compared to controls after I/R. As evidence of an overall increase in hepatocellular oxidative injury, LT2-KD livers demonstrated a uniform induction of the compensatory anti-oxidant genes GST, SOD1 and GPX1 following I/R (supplementary Fig.S2E). Given these phenotypic effects in the LT2-KD mice and a potential crosstalk between β-catenin and HIF-1α that has been previously demonstrated in vitro11–13, we sought further evidence for the effect of β-catenin deletion on HIF signal transduction in response to hypoxia and I/R.

An external file that holds a picture, illustration, etc. Object name is nihms294063f2.jpg
β-catenin knockdown mice are more susceptible to hepatic I/R injury

  1. Schematic strategy for the generation of LT2-KD mouse. qRT-PCR demonstrates a significant reduction of β-catenin mRNA in LT2-KD livers after 4 weeks of dox-withdrawal. Immunoblot shows a remarkable knockdown at its protein level in LT2-KD hepatocytes.
  2. Hepatocellular injury, as evidenced by elevated serum transaminases, is significantly increased in LT2-KD mice after ischemia and I/R.
  3. Severe liver damage, increased apoptosis and elevated ROS is detected in LT2-KD mice after ischemia and I/R. Representative liver histology of sham, ischemia or I/R-treated livers. Apoptosis is measured by TUNEL staining after ischemia or I/R. Intracellular ROS levels are detected by DHE staining after I/R.
  4. LT2-KD livers are more susceptible to I/R-induced apoptosis. Quantification of TUNEL-positive cells/10 HPF in ischemia or I/R-treated livers (see Fig.2C). Immunoblot shows more caspase-cleaved K18Asp237 in total liver lysates of LT2-KD mice after I/R.
  5. β-catenin deficient livers show increased necrosis by quantification of necrotic areas in ischemia and I/R-treated livers (see Fig.2C).
  6. Reduced HIF-1α protein and its target gene (VEGF) mRNA expression in LT2-KD mice. VEGF mRNA expression was measured in sham, ischemia or I/R-treated livers by qRT-PCR. HIF-1α induction after ischemia is significantly impaired in LT2-KD livers as shown by immunoblot (n=5). *p<0.05, **p<0.01.

In response to ischemia, HIF-1α protein significantly increased, whereas HIF-2α expression seemed not inducible (Fig.2F). Intriguingly, a significantly impaired induction in HIF-1α protein (Fig.2F, KD vs LT2-WT) and its target genes VEGF and HO-1 (supplementary Fig.S2F) was observed in β-catenin deficient hepatocytes after ischemia and/or I/R. Further, isolated LT2-KD hepatocytes demonstrated both reduced TCF and HIF signal activity under hypoxia (supplementary Fig.S3A, Fig.S3B) compared to LT2-WT. Together, these results demonstrate that β-catenin has a protective role during liver ischemia and I/R that dramatically reduces the degree of hepatocellular injury. These data further suggest that HIF-1α signaling is impaired in the absence of β-catenin in response to ischemia or I/R.

Augmented β-catenin signaling protects mice from liver I/R Injury

To further investigate the role of Wnt/β-catenin signaling in injury protection in vivo, we created a hepatocyte-specific Wnt1 overexpressing mouse responsive to doxycycline. Bioluminescent imaging showed increased Wnt1-Luciferase activity after dox-withdrawal with a maximum after 21 days (Fig.3A). Longer dox-withdrawal (>3 weeks) did not further increase Wnt1 expression. A significant increase in Wnt1 and Cyclin D1 protein together with a moderate increase in total β-catenin (Fig.3A) was observed in Wnt1+ livers along with several established canonical β-catenin downstream target genes after 3 weeks of Dox-withdrawal (supplementary Fig.S4A). Unteated Wnt1-WT and Wnt1+ mice did not show any difference in transaminases or liver histology (supplementary Fig.S4B, Fig.S4C). To determine if Wnt-mediated β-catenin gain-of-function has a protective effect against hypoxic injury in vivo, we subjected Wnt1+ and Wnt1-WT mice to hepatic ischemia and I/R. Hepatocellular injury as measured by serum transaminases was substantially attenuated in Wnt1+ mice after ischemia and I/R (Fig.3B) when compared to controls indicating significant protection against oxidative injury in vivo. Sham-treated mice demonstrated transaminases in the normal range (supplementary Fig.S4D). Liver histology (Fig.3C) showed remarkable hepatocyte protection against ischemia in Wnt1+ livers with well-preserved hepatocellular architecture. Ischemic Wnt1-WT livers revealed severe hepatocyte injury with swelling and cytoplasmic vacuolization. In response to I/R, Wnt1-WT livers demonstrated extensive hepatocyte damage with signs of liver congestion and degenerative changes. Conversely, the hepatocellular architecture in Wnt1+ livers was still well-preserved. Correspondingly, Wnt1+ mice were also more resistant to apoptosis (Fig.3C, Fig.3D) and showed less necrosis (Fig.3C, Fig.3E) after ischemia and I/R. DHE staining (Fig.3C) showed less ROS production in Wnt1+ livers after I/R when compared to controls. Interestingly and contrary to LT2-KD mice, notable HIF-1α protein and its target genes VEGF (Fig.3F), Epo and Glut1 (supplementary Fig.S4E) were uniformly increased more in Wnt1+ livers after ischemia and/or I/R when compared to controls. However, HIF-2α protein did not change in response to ischemia or between Wnt1-WT and Wnt1+ mice. The augmented response in HIF-1α and correspondingly increased target gene expression is a potentially significant contributor to the observed increase in hepatocyte resistance to hypoxic injury. In further support of a Wnt1-mediated resistance to oxidative injury, a uniform decrease in anti-oxidant gene expression was also observed in Wnt1+ livers following I/R (supplementary Fig.S4F). Taken together, these data demonstrate that Wnt1 overexpression confers significant hepatic protection against I/R injury through augmented HIF-1α signaling and an overall reduction in redox imbalance.

An external file that holds a picture, illustration, etc. Object name is nihms294063f3.jpg
Wnt1 gain-of-function provides strong protection against hepatic I\R Injury

  1. Bioluminescence imaging of Wnt1+-Luciferase mice show increased liver-specific Luciferase activity upon Dox-withdrawal. Increased Wnt1 and β-catenin target gene (Cyclin D1) protein are expressed in liver lysates of Wnt1+ mice along with a moderate increase in total β-catenin after 3 weeks of dox-removal.
  2. Reduced hepatocellular injury as evidenced by lower transaminases is detected in Wnt1+ mice after ischemia and I/R.
  3. Minor hepatic damage with reduced apoptosis and lower ROS was determined in Wnt1+ livers after ischemia and I/R. Representative liver histology (H&E) of sham, ischemia or I/R-treated livers. Apoptosis was measured by TUNEL staining after ischemia or I/R. Intracellular ROS levels were detected by DHE staining after I/R.
  4. Wnt1+ livers are more resistant to ischemia and I/R-induced apoptosis. Quantification of TUNEL-positive cells/10 HPF in ischemia and I/R-treated livers (see Fig.3C). Immunoblot of liver lysates demonstrates diminished levels of caspase-cleaved K18Asp237 in Wnt1+ mice after I/R.
  5. Wn1+ livers show less necrosis in ischemia and I/R-treated livers by quantification of necrotic areas (see Fig.3C).
  6. HIF-1α protein and its target gene (VEGF) expression are significantly induced in Wnt1+ livers. VEGF mRNA expression in sham, ischemia and I/R-treated livers was measured by qRT-PCR. HIF-1α but not HIF-2α is significantly augmented in Wnt1+ livers after ischemia as shown by western blot (n=5). *p<0.05, **p<0.01.

Hepatocytes overexpressing β-catenin resist hypoxia-induced apoptosis through augmented HIF-1α signaling

To further explore the mechanism by which Wnt/β-catenin affords increased protection against hypoxia or H/R, mutant AML12 hepatocytes carrying an amino terminus phosphorylation resistant point mutation (S33Y) in β-catenin were created. S33Y mutants demonstrated increased β-catenin level, robust TCF signal transduction and increased proliferation compared to control cells (supplementary Fig.S5A, Fig.S5B). When cells were exposed to hypoxia, β-catenin mutants demonstrated death resistance and continued proliferation despite hypoxic stress (Fig.4A, supplementary Fig.S5B, Fig.S5C). Moreover, S33Y mutants showed reduced sensitivity to oxidative stress with lower intracellular ROS levels and less apoptosis compared to controls (Fig.4B, Fig.4C, Fig.4D). Reoxygenation after hypoxia resulted in increased apoptosis, but β-catenin mutants remain less susceptible to H/R-induced injury than controls (Fig.4C). HIF-1α protein was not detectable under normoxia, but was stabilized by hypoxia and H/R (Fig.4C). Intringuingly, β-catenin mutants displayed increased HIF-1α in response to both hypoxia and H/R compared to control hepatocytes. Taken together, these data demonstrate that β-catenin stabilization protects hepatocytes against hypoxic injury in association with diminished ROS production and augmented HIF-1α signaling resulting in increased cell survival.

An external file that holds a picture, illustration, etc. Object name is nihms294063f4.jpg
β-catenin stabilized hepatocytes are resistant to hypoxia-induced apoptosis in vitro by augmented HIF-1α signaling

  1. S33Y mutant hepatocytes are stress-resistant and proliferate despite hypoxic stress as measured by BrdU ELISA assay.
  2. Under hypoxia, β-catenin mutants show reduced intracellular ROS as determined by DCF-DA flow cytometry.
  3. S33Y hepatocytes are highly resistant to hypoxia or H/R-induced apoptosis as determined by caspase-cleaved K18Asp237. β-catenin mutants show more HIF-1α induction under hypoxia or H/R compared to controls.
  4. Less apoptosis is detected in β-catenin mutants under hypoxia by Caspase-Glo 3/7 activity assay.
  5. TCF reporter activity is decreased in S33Y mutants after hypoxia to a comparable degree as control cells despite a heightened β-catenin/TCF baseline activity.
  6. Augmented HIF-1α reporter activity is detected in β-catenin mutants under hypoxia. RLU= relative light units. *p<0.05.

When exposed to hypoxia, TCF signal transduction in S33Y cells decreased comparable to controls despite a higher baseline elevation (Fig.4E). Interestingly, β-catenin mutants also demonstrated significantly increased HIF activity compared to control cells (Fig.4F). These results support the in vivo findings above and further extend these observations to demonstrate that augmented HIF-1α signaling occurs with β-catenin stabilization, yet independent of the canonical β-catenin/TCF pathway. Moreover, these data suggest that the ROS-mediated dampening of β-catenin/TCF signaling is not dependent on phosphorylation of β-catenin.

In response to hypoxia, β-catenin diverts from TCF binding to HIF-1α to support cell survival

Given the findings above, we next sought to determine if β-catenin complexes directly with TCF or HIF-1α according to cellular conditions that may favor proliferation or survival respectively. To investigate whether the inhibition in TCF signaling is a result of a β-catenin switch as a transcriptional activator, co-immunoprecipitation assays were performed. The hepatocellular carcinoma cells HepG2 were utilized to investigate this phenomenon in a human cell line with clinical relevance and established β-catenin gain-of-function. Under normoxia, β-catenin was found to primarily complex with TCF but alternatively predominantly binds HIF-1α in response to hypoxia (Fig.5A). To extend the in vitro crosstalk above, we further provide in vivo evidence that β-catenin binds to HIF-1α in ischemic livers to promote injury protection as shown by co-immunoprecipitation (Fig.5B). These results provide additional evidence that in Wnt1 overexpressing livers, β-catenin/HIF-1α binding is significantly increased when compared to Wnt1-WT mice.

An external file that holds a picture, illustration, etc. Object name is nihms294063f5.jpg
β-catenin signaling protects hepatocytes against hypoxic injury through augmented HIF-1 signaling

  1. β-catenin alternatively binds TCF (normoxia) or HIF-1α (hypoxia) depending on oxygen availability. Binding switch was verified by co-immunoprecipitation on HepG2 cell lysates after exposure to normoxia (N) or hypoxia (H) using anti-β-catenin antibody followed by immunoblotting for HIF-1α and TCF4.
  2. Wnt1+ livers show more β-catenin/HIF-1α binding as verified by co-immunoprecipitation in sham or ischemia-treated liver lysates using anti-β-catenin antibody followed by HIF-1α immunoblotting.
  3. Augmented HIF-1α promotor binding in β-catenin mutants after 24 hours hypoxia. For EMSA, 5 µg of nuclear extracts were used for HIF-1/HRE DNA-probe binding reactions.
  4. β-catenin/TCF reporter activity can be reduced by HIF-1α stabilization with CoCl2 (150 µmol/L) under normoxia or hypoxia treatment in AML12 hepatocytes. HIF-1α induction was verified by HRE reporter assay and immunoblot. RLU= relative light units.
  5. HIF-1α inhibition under hypoxia by YC-1 pre-treatment (100 µmol/L) for 1 hour results in more apoptosis in AML12 hepatocytes as measured by MTT assay and immunoblot for caspase-cleaved K18Asp237. *p<0.05.

Next, we determined the effect of β-catenin gain-of-function mutation (β-cateninS33Y) on HIF-1 DNA binding activity. EMSA was performed using a HRE probe, derived from the erythropoietin enhancer, along with nuclear extracts from control cells and β-catenin mutants treated with normoxia and hypoxia (Fig.5C). Significantly more HIF-1/HRE probe binding was observed in β-cateninS33Y vs control lysates under hypoxia. This binding specificity of HIF-1 was confirmed by adding an unlabeled competitor probe that resulted in the disappearance of the band (not shown). To investigate if β-catenin/TCF signaling can be dampened by HIF-1α directly, we stabilized HIF-1α expression under normoxia by the hypoxia-mimetic cobalt chloride (CoCl2) (Fig.5D). HIF stabilization by CoCl2 under normoxia resulted in a significant reduction in TCF and increase in HIF-1α signal activity and protein level in AML12 hepatocytes (Fig.5D). To further elucidate whether the protective effect against hypoxia-induced apoptosis is specific to HIF-1α signaling, hepatocytes were treated with YC-1, a known HIF inhibitor. HIF-1α inhibition under hypoxia significantly reduced HIF signal activity (supplementary Fig.S6A) and caused more apoptosis (Fig.5E).

Taken together, these data support a model in which β-catenin binds to HIF-1α under oxygen-limiting conditions to support hepatocyte survival (Fig.6). Evidence is also provided to suggest that a phosphorylation independent event modulates β-catenin’s role as a transcriptional activator in response to hypoxia.

An external file that holds a picture, illustration, etc. Object name is nihms294063f6.jpg
Working model explaining β-catenin’s role in hepatocyte protection to I/R Injury

  1. Under normoxic, favorable conditions (upper panel), β-catenin complexes with TCF to promote hepatocyte proliferation.
  2. Under unfavorable conditions like hypoxia or I/R, β-catenin diverts from TCF and preferably associates with HIF-1α to favor cell adaptation and survival (lower, left panel). However, in the absence of β-catenin (lower, middle panel), neither sufficient TCF nor effective HIF-1α signaling can be activated leading to cell death and enhanced liver injury under hypoxic stress. With Wnt/β-catenin signaling gain-of-function (lower, right panel), signaling through both TCF and HIF-1α occurs resulting in increased resistance to I/R and continued proliferation.

Discussion

The hepatocellular response to a variety of stimulants involving oxidative stress is a common injury mechanism in nearly all liver pathologies1,2. In this study, we utilized cellular hypoxia and tissue ischemia-reperfusion as clinically relevant models involving profound changes in hepatocyte redox balance to gain a better understanding of the critical molecular regulators of hepatic adaptation and potential liver protection strategies. The pleiotropic effects of Wnt signaling are well established with specific roles identified for the Wnt/β-catenin axis during hepatic development, nutrient metabolism, regeneration and carcinogenesis4–6,28–30. Despite abundant data demonstrating the predominance of biologic activity mediated by Wnt signal transduction through β-catenin and TCF/LEF, more recent evidence supports a diverse role for β-catenin as a transcriptional co-activator independent of TCF. Specifically, β-catenin has been previously shown to interact with HIF-1α and FOXO transcription factors to mediate a cellular response to unfavorable environmental conditions in vitro11,12,14,31. Since the liver is of singular metabolic importance and β-catenin as well as HIF signaling are known critical regulators of cellular metabolism, this study was designed to further investigate the in vivo relevance and molecular mechanisms governing possible β-catenin/HIF crosstalk. Moreover, as tissue hypoxia and Wnt signaling are both known to be critical mediators of diverse biologic processes from development to tissue regeneration and tumorigenesis, the findings reported herein may have far reaching implications for developing novel treatment strategies.

In this study, we report that β-catenin signaling through HIF-1α is critical for an adequate in vivo tissue response in a model of hepatic ischemia-reperfusion injury. We submit that the severe hepatic damage observed in the β-catenin knockdown mouse in response to ischemia and I/R results from the deletion of β-catenin in adult mouse liver. In contrast to all other reports6,27,29,30 using the β-cateninflox/flox/Alb-Cre mouse that results in perinatal β-catenin deletion, our novel β-catenin knockdown approach has the additional advantage of conditional control for the genetic deletion of β-catenin from mature hepatocytes in order to overcome any compensatory adaptive changes that may occur during liver development and post-natally. HIF-1α is a key transcription factor involved in the cellular response to oxygen and nutrient stress signaling. HIF-1α is stabilized under hypoxic conditions resulting in the upregulation of genes involved in angiogenesis, energy metabolism, adaptation, survival, and glycolysis15,16,32,33. Crosstalk between Wnt/β-catenin and HIF-1α signaling has been previously described in different organ systems in vitro12,13,31,34, but to date has not been shown in vivo liver injury. As previously reported by Kaidi and others12,13, the current study complements previous findings that HIF-1α can compete with TCF for binding β-catenin in response to hypoxia. Our in vivo findings further suggest that HIF-1α signaling is specifically impaired and cellular redox balance disrupted as a result of β-catenin depletion from hepatocytes. We have extended the prior observations of others to critically demonstrate that β-catenin signal transduction is significantly impacted by cellular redox balance as shown by changes in ROS level and their modulation by the anti-oxidant NAC.

Taken together, our results demonstrate that an increase in cellular ROS levels that accompany numerous injury stimuli (oxidative, hypoxic, metabolic and genotoxic stress), and thereby relevant hepatic pathologic states, are critical intra-cellular mediators of β-catenin signaling in response to changing conditions. This provides a key mechanistic distinction to deepen our understanding of the proposed model (Fig.6) in which hepatocytes respond to changes in relevant stimuli to enact programs alternatively for proliferation (TCF) when conditions are favorable, or adaptation and survival (HIF-1α) in response to limiting environmental conditions.

Another intriguing finding in this study was the observation that canonical phosphorylation changes in β-catenin did not appear to be the primary molecular mechanism by which Wnt signal transduction is altered in response to changing cellular conditions. Wnt pathway gain-of-function mutations, including commonly identified key β-catenin phosphorylation site mutations that were utilized herein, have been reported in human HCC and mouse liver tumor models28,35. In light of our findings, the interaction of accumulated β-catenin with both TCF and HIF-1α may represent a maladaptive response in which given an abundance of β-catenin, cellular proliferation proceeds despite unfavorable conditions, but also results in enhanced HIF-1α signaling for increased survival as we have demonstrated. Since the phosphorylation resistant mutants (β-cateninS33Y) were responsive to oxidative changes as demonstrated by decreased TCF activity comparable to cells with wild-type β-catenin, we speculate that alternative post-translational modifications (e.g. acetylation) regulating β-catenin are additionally responsible for modulating its activity as a transcriptional regulator9,10,13. Additional studies beyond the scope of this report are needed to further define the post-translational mechanisms governing β-catenin activity in response to ROS.

To strengthen and extend our findings, we also sought to determine whether the Wnt pathway could be utilized to augment HIF-1α signaling to promote cell survival in response to the same injury stimuli that β-catenin deficient mice were sensitized to in vivo. Wnt-1 is an established proto-oncogene and upstream regulator of β-catenin signaling36–38. It has been previously reported that Wnt-1/β-catenin activation via retroviral Wnt-1 injection promotes liver regeneration through enhanced hepatocyte proliferation following partial hepatectomy35. However, to our knowledge Wnt pathway manipulation has not previously been tested for an ability to promote tissue resistance to oxidative injury, and specifically to I/R in the liver. Our findings that Wnt-1 overexpression results in activation of β-catenin/TCF signaling and relevant downstream target genes is also supported by previous experimental observations in other tissues39,40. Interestingly, we observed that Wnt1-mediated β-catenin stabilization resulted in increased hepatocyte survival and strong hepatic resistance to oxidative injury. Intriguingly, we also detected a parallel increase in HIF-1α protein, relevant HIF-1α target genes and a corresponding decrease in the compensatory antioxidants SOD1 and GPX. Surprisingly, HIF-2α did not change at protein level in response to I/R. Together, these findings support the important role of the Wnt axis as a key in vivo modulator of HIF-1α signaling and redox balance for adaptation and cell survival.

In summary, this study provides the first in vivo evidence that β-catenin is a key component of a tissue-specific response to I/R and that the molecular mechanism appears to be mediated by cellular redox balance and β-catenin’s role as a transcriptional activator. Our observations have significant clinical relevance since ischemia and reperfusion are common features of many liver diseases. In light of the data presented here, this study may further support the promise and clinical application of Wnt1 manipulation or other modulators of Wnt/β-catenin signaling in clinically relevant settings involving a hepatic oxidative stress response to effect liver injury protection, repair and regeneration.

Supplementary Material

01

Acknowledgements

Grant Support:

This work was supported by grants from the American College of Surgeons, American Pediatric Surgical Association, Oak Foundation, Packard Foundation, and the National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases (pilot grant DK56339: Digestive Disease Center at Stanford) to K.G.S. and the YAEL Foundation and the German Research Foundation (DFG NL 2509/2-1) to N.L.

We thank Dr. Geoffrey Gurtner (Stanford University) for providing the HRE and Dr. Eric Fearon (University of Michigan) for the pcDNA3S33Y plasmids for these studies. We thank Dr. Unsal Kuscouglu for creating the Wnt1+ mouse and generating the S33Y mutants. We would like to thank Dr. Dean Felsher (Stanford University) for providing the LAP-tTA/tet-O-Cre mice and Dr. L.A. Chodoff (University of Pennsylvania) for the TetO-Wnt1-Luc mice. We are grateful to the entire staff at the Veterinary Service Center for the excellent care of research animals.

Abbreviations

B2Mβ2-microglobulin
CoCl2Cobalt Cloride
DCF-DAdichlorofluorescein diacetate
DHEdihydroethidium
Epoerythropoetin
FACSfluorescence activated cell sorting
GAPDHglyceraldehyde 3-phosphate dehydrogenase
Glut1glucose transporter 1
GPX1glutathione peroxidase 1
GSTglutathione S-transferase
HCChepatocellular carcinoma
HO-1heme oxygenase 1
HIF-1αhypoxia inducible factor-1α
HPFhigh power field
H/Rhypoxia reoxygenation
HREhypoxia response element
I/Rischemia reperfusion
LAPliver enriched activator protein (a.k.a. CEBP/β)
LEFlymphoid enhancer factor
NACN-acetylcysteine
SOD1superoxide dismutase 1
TCFT cell factor
tetO-Cretetracycline response element driving Cre
tTatetracycline transactivating
VEGFVascular Endothelial Growth Factor

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The authors declare no conflict of interest.

Author contributions:

N.L., G.Z.T. and K.G.S. designed research; N.L., G.Z.T., M.S. and K.Y.J. performed research; W.T.K., K.Y.J., and G.Z.T. contributed new reagents/analytic tools; N.L., G.Z.T., W.T.K. and K.G.S. analyzed data; and N.L., G.Z.T. and K.G.S. wrote the manuscript.

References

1. Kaplowitz N. Mechanisms of liver cell injury. J Hepatol. 2000;32:39–47. [PubMed] [Google Scholar]
2. Fausto N. Liver regeneration and repair: hepatocytes, progenitor cells, and stem cells. Hepatology. 2004;39:1477–1487. [PubMed] [Google Scholar]
3. Schneider M, Van Geyte K, Fraisl P, et al. Loss or silencing of the PHD1 prolyl hydroxylase protects livers of mice against ischemia/reperfusion injury. Gastroenterology. 138:1143–1153. e1-2. [PubMed] [Google Scholar]
4. Calvisi DF, Factor VM, Ladu S, et al. Disruption of beta-catenin pathway or genomic instability define two distinct categories of liver cancer in transgenic mice. Gastroenterology. 2004;126:1374–1386. [PubMed] [Google Scholar]
5. Ober EA, Verkade H, Field HA, et al. Mesodermal Wnt2b signalling positively regulates liver specification. Nature. 2006;442:688–691. [PubMed] [Google Scholar]
6. Tan X, Behari J, Cieply B, et al. Conditional deletion of beta-catenin reveals its role in liver growth and regeneration. Gastroenterology. 2006;131:1561–1572. [PubMed] [Google Scholar]
7. Nusse R. Wnts and Hedgehogs: lipid-modified proteins and similarities in signaling mechanisms at the cell surface. Development. 2003;130:5297–5305. [PubMed] [Google Scholar]
8. Nelson WJ, Nusse R. Convergence of Wnt, beta-catenin, and cadherin pathways. Science. 2004;303:1483–1487. [PMC free article] [PubMed] [Google Scholar]
9. Lim JH, Park JW, Chun YS. Human arrest defective 1 acetylates and activates beta-catenin, promoting lung cancer cell proliferation. Cancer Res. 2006;66:10677–10682. [PubMed] [Google Scholar]
10. Levy L, Wei Y, Labalette C, et al. Acetylation of beta-catenin by p300 regulates beta-catenin-Tcf4 interaction. Mol Cell Biol. 2004;24:3404–3414. [PMC free article] [PubMed] [Google Scholar]
11. Hoogeboom D, Burgering BM. Should I stay or should I go: beta-catenin decides under stress. Biochim Biophys Acta. 2009;1796:63–74. [PubMed] [Google Scholar]
12. Kaidi A, Williams AC, Paraskeva C. Interaction between beta-catenin and HIF-1 promotes cellular adaptation to hypoxia. Nat Cell Biol. 2007;9:210–217. [PubMed] [Google Scholar]
13. Lim JH, Chun YS, Park JW. Hypoxia-inducible factor-1alpha obstructs a Wnt signaling pathway by inhibiting the hARD1-mediated activation of beta-catenin. Cancer Res. 2008;68:5177–5184. [PubMed] [Google Scholar]
14. Essers MA, de Vries-Smits LM, Barker N, et al. Functional interaction between beta-catenin and FOXO in oxidative stress signaling. Science. 2005;308:1181–1184. [PubMed] [Google Scholar]
15. Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer. 2003;3:721–732. [PubMed] [Google Scholar]
16. Graeber TG, Osmanian C, Jacks T, et al. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature. 1996;379:88–91. [PubMed] [Google Scholar]
17. Pouyssegur J, Dayan F, Mazure NM. Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature. 2006;441:437–443. [PubMed] [Google Scholar]
18. Beer S, Komatsubara K, Bellovin DI, et al. Hepatotoxin-induced changes in the adult murine liver promote MYC-induced tumorigenesis. PLoS One. 2008;3:e2493. [PMC free article] [PubMed] [Google Scholar]
19. Shachaf CM, Kopelman AM, Arvanitis C, et al. MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer. Nature. 2004;431:1112–1117. [PubMed] [Google Scholar]
20. Debies MT, Gestl SA, Mathers JL, et al. Tumor escape in a Wnt1-dependent mouse breast cancer model is enabled by p19Arf/p53 pathway lesions but not p16 Ink4a loss. J Clin Invest. 2008;118:51–63. [PMC free article] [PubMed] [Google Scholar]
21. Abe Y, Hines IN, Zibari G, et al. Mouse model of liver ischemia and reperfusion injury: method for studying reactive oxygen and nitrogen metabolites in vivo. Free Radic Biol Med. 2009;46:1–7. [PMC free article] [PubMed] [Google Scholar]
22. Li WC, Ralphs KL, Tosh D. Isolation and culture of adult mouse hepatocytes. Methods Mol Biol. 633:185–196. [PubMed] [Google Scholar]
23. Wu JC, Merlino G, Fausto N. Establishment and characterization of differentiated, nontransformed hepatocyte cell lines derived from mice transgenic for transforming growth factor alpha. Proc Natl Acad Sci U S A. 1994;91:674–678. [PMC free article] [PubMed] [Google Scholar]
24. Tao GZ, Li DH, Zhou Q, et al. Monitoring of epithelial cell caspase activation via detection of durable keratin fragment formation. J Pathol. 2008;215:164–174. [PubMed] [Google Scholar]
25. Hu M, Kurobe M, Jeong YJ, et al. Wnt/beta-catenin signaling in murine hepatic transit amplifying progenitor cells. Gastroenterology. 2007;133:1579–1591. [PubMed] [Google Scholar]
26. Diehn M, Cho RW, Lobo NA, et al. Association of reactive oxygen species levels and radioresistance in cancer stem cells. Nature. 2009;458:780–783. [PMC free article] [PubMed] [Google Scholar]
27. Behari J, Yeh TH, Krauland L, et al. Liver-specific beta-catenin knockout mice exhibit defective bile acid and cholesterol homeostasis and increased susceptibility to diet-induced steatohepatitis. Am J Pathol. 176:744–753. [PMC free article] [PubMed] [Google Scholar]
28. de La Coste A, Romagnolo B, Billuart P, et al. Somatic mutations of the beta-catenin gene are frequent in mouse and human hepatocellular carcinomas. Proc Natl Acad Sci U S A. 1998;95:8847–8851. [PMC free article] [PubMed] [Google Scholar]
29. Apte U, Singh S, Zeng G, et al. Beta-catenin activation promotes liver regeneration after acetaminophen-induced injury. Am J Pathol. 2009;175:1056–1065. [PMC free article] [PubMed] [Google Scholar]
30. Sekine S, Gutierrez PJ, Lan BY, et al. Liver-specific loss of beta-catenin results in delayed hepatocyte proliferation after partial hepatectomy. Hepatology. 2007;45:361–368. [PubMed] [Google Scholar]
31. Mazumdar J, O'Brien WT, Johnson RS, et al. O2 regulates stem cells through Wnt/beta-catenin signalling. Nat Cell Biol. 12:1007–1013. [PMC free article] [PubMed] [Google Scholar]
32. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer. 2004;4:891–899. [PubMed] [Google Scholar]
33. Simon MC, Keith B. The role of oxygen availability in embryonic development and stem cell function. Nat Rev Mol Cell Biol. 2008;9:285–296. [PMC free article] [PubMed] [Google Scholar]
34. Giles RH, Lolkema MP, Snijckers CM, et al. Interplay between VHL/HIF1alpha and Wnt/beta-catenin pathways during colorectal tumorigenesis. Oncogene. 2006;25:3065–3070. [PubMed] [Google Scholar]
35. Nejak-Bowen KN, Thompson MD, Singh S, et al. Accelerated liver regeneration and hepatocarcinogenesis in mice overexpressing serine-45 mutant beta-catenin. Hepatology. 51:1603–1613. [PMC free article] [PubMed] [Google Scholar]
36. Shang XZ, Zhu H, Lin K, et al. Stabilized beta-catenin promotes hepatocyte proliferation and inhibits TNFalpha-induced apoptosis. Lab Invest. 2004;84:332–341. [PubMed] [Google Scholar]
37. Monga SP, Monga HK, Tan X, et al. Beta-catenin antisense studies in embryonic liver cultures: role in proliferation, apoptosis, and lineage specification. Gastroenterology. 2003;124:202–216. [PubMed] [Google Scholar]
38. Zeng G, Apte U, Cieply B, et al. siRNA-mediated beta-catenin knockdown in human hepatoma cells results in decreased growth and survival. Neoplasia. 2007;9:951–959. [PMC free article] [PubMed] [Google Scholar]
39. Young CS, Masckauchan TN, Kitajewski J. Beta-catenin/Tcf activation partially mimics the transforming activity of Wnt-1 in Rat-1 fibroblasts. Differentiation. 2003;71:477–485. [PubMed] [Google Scholar]
40. Hsu SC, Galceran J, Grosschedl R. Modulation of transcriptional regulation by LEF-1 in response to Wnt-1 signaling and association with beta-catenin. Mol Cell Biol. 1998;18:4807–4818. [PMC free article] [PubMed] [Google Scholar]
-