Liver Biology and Pathobiology
Article first published online: 23 MAR 2006
Copyright © 2006 American Association for the Study of Liver Diseases
Volume 43, Issue 4, pages 817–825, April 2006
How to Cite
Sekine, S., Lan, B. Y.-A., Bedolli, M., Feng, S. and Hebrok, M. (2006), Liver-specific loss of β-catenin blocks glutamine synthesis pathway activity and cytochrome p450 expression in mice. Hepatology, 43: 817–825. doi: 10.1002/hep.21131
See Editorial on Page 650.
Potential conflict of Interest: Nothing to report
- Issue published online: 23 MAR 2006
- Article first published online: 23 MAR 2006
- Manuscript Accepted: 4 JAN 2006
- Manuscript Received: 28 OCT 2005
- Pilot Feasibility grant from the University of California–San Francisco Liver Center. Grant Number: P30-DK26743
- National Institutes of Health
- Third Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labor, and Welfare of Japan
There is accumulating evidence that Wnt/β-catenin signaling is involved in the regulation of liver development and physiology. The presence of genetic alterations resulting in constitutive β-catenin stabilization in human and murine liver tumors also implicates this pathway in hepatocyte proliferation. In the present study, we generated hepatocyte-specific β-catenin knockout mice to explore the role of β-catenin in liver function. Conditional knockout mice were born at the expected Mendelian ratio and developed normally to adulthood, indicating β-catenin is dispensable for essential liver function under normal breeding conditions. However, the liver mass of knockout mice was 20% less than those of mice in the control groups. Expression analysis revealed loss of genes required for glutamine synthesis in knockout mice. Loss of the liver glutamine synthesis pathway did not affect the blood ammonia level in mice fed a standard diet, yet, knockout mice showed significantly elevated blood ammonia levels with high-protein dietary feeding. Furthermore, the expression of two cytochrome P450 enzymes, CYP1A2 and CYP2E1, was almost completely abolished in livers from hepatocyte-specific β-catenin knockout mice. Consequently, these mice were resistant to acetaminophen challenge, confirming the requirement of these cytochrome P450 enzymes for metabolism of xenobiotic substances. In conclusion, in addition to regulating hepatocyte proliferation, β-catenin may also control multiple aspects of normal liver function. (HEPATOLOGY 2006;43:817–825.)
β-Catenin is an adapter protein that fulfills two distinct roles in cells. As a part of an adherens junction complex, it interacts with transmembrane proteins of the cadherin family to promote cell–cell adhesion. In addition, β-catenin is an integral part of the canonical Wnt signaling pathway known to regulate cell proliferation, differentiation, and stem cell maintenance in a wide variety of tissues.1–3 Wnt ligands are secreted proteins that bind to cognate frizzled receptors expressed in the cell membrane of Wnt-responsive cells. In the absence of Wnt signals, cytoplasmic β-catenin is phosphorylated by glycogen synthase kinase 3β, a modification that triggers rapid proteosomal degradation of β-catenin. Upon stimulation with Wnt ligands, cytoplasmic β-catenin is stabilized and translocates to the nucleus, where it forms active transcription complexes with members of the TCF/LEF1 transcription factor family.
There is accumulating evidence that indicates a role for Wnt/β-catenin signaling in hepatocyte proliferation. Micsenyi et al. reported nuclear/cytoplasmic localization of β-catenin in hepatocytes during early developmental stages, which gradually decrease with progression of organ development. During liver development, the level of nuclear/cytoplasmic β-catenin expression correlates well with the proliferative activity of hepatocytes.4 Suppression of β-catenin by antisense oligonucleotides in ex vivo liver cultures results in decreased cell proliferation and increased apoptosis of hepatocytes.5 Conversely, transgenic expression of N-terminally truncated β-catenin results in increased hepatocyte proliferation and hepatomegaly,6 a finding that provides direct evidence that stabilization of β-catenin positively regulates hepatocyte proliferation in vivo.
Human and murine tumor genetic studies further imply a role for β-catenin in regulation of hepatocyte proliferation. It has been reported that approximately 30% of hepatocellular carcinomas and more than 80% of hepatoblastomas harbor genetic alterations that impair β-catenin degradation, including mutations in β-catenin, APC, Axin1, and Axin2.7–12 Mutations in these genes cause inhibition of glycogen synthase kinase 3β–mediated phosphorylation of β-catenin and prevent subsequent degradation of β-catenin, resulting in constitutive activation of TCF/LEF1-dependent transcription. Hepatocellular carcinomas with stabilized β-catenin show marked proliferative activity.13, 14
In addition to a role in hepatocyte proliferation, studies on human and murine liver tumors also suggested that Wnt/β-catenin signaling is likely to regulate additional aspects of liver function. Several liver enzymes are upregulated in liver tumors with β-catenin mutations, including three genes involved in glutamine synthesis: glutamine synthetase (GS), ornithine aminotransferase (OAT) and glutamate transporter 1 (GLT-1), as well as cytochrome P450 enzymes.15–17
Until now, there have been no functional studies addressing the requirement for Wnt/β-catenin signaling in liver in vivo. We report the generation of hepatocyte-specific β-catenin knockout mice to clarify the physiological roles of β-catenin in the liver. Our findings demonstrate a requirement for β-catenin in hepatocyte proliferation and metabolism of ammonia and xenobiotic substances.
Material and Methods
Generation of albumin promoter-driven Cre recombinase transgenic mice (Alb-Cre mice) and mice carrying the floxed allele of β-catenin (β-cateninloxP/loxP mice) has been previously described.18, 19 Alb-Cre mice express Cre recombinase specifically in hepatocytes during perinatal to postnatal stages.18, 20 Alb-Cre and β-cateninloxP/loxP mice were crossed to obtain hepatocyte-specific β-catenin knockout mice (Alb-Cre;β-cateninloxP/loxP mice). Genotyping was performed via polymerase chain reaction (PCR) analysis using genomic DNA isolated from the tail tip as previously described.18, 19 Mice with heterozygous β-catenin hepatocytes (Alb-Cre;β-cateninloxP/+) and mice lacking the Cre recombinase (β-cateninloxP/loxP and β-cateninloxP/+) behaved similarly throughout the study. Therefore, mice with these 3 genotypes were grouped together as control littermates unless specified otherwise.
Mice used in the present study were maintained in the barrier facility according to protocols approved by the Committee on Animal Research at the University of California–San Francisco. Mice had free access to food and water and were kept on a 12-hour light/dark cycle unless specified otherwise. For analysis, mice were sacrificed between 10:00 A.M. and 12:00 A.M. to avoid effects of diurnal variation.
Liver tissue samples were fixed overnight in zinc-containing neutral buffered formalin, embedded in paraffin, cut into 5-μm-thick sections, and placed on Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA). Sections were subjected to hematoxylin-eosin and immunohistochemical staining. For immunohistochemistry, deparaffinized and rehydrated slides were subjected to autoclave antigen retrieval in a 10-mmol/L citric acid buffer (pH 6.0) and allowed to cool for 30 minutes at room temperature. Slides were blocked with 0.3% H2O2 for 20 minutes, washed in phosphate-buffered saline, then blocked with avidin and biotin blocking agent (Vector Laboratories, Burlingame, CA) and protein blocking reagent (DAKO Cytomation, Glostrup, Denmark). Slides were incubated with diluted primary antibodies overnight at 4°C. The following primary antibodies were used: mouse anti–β-catenin (1:500 dilution) (Becton and Dickinson, Franklin Lakes, NJ)16, 17; mouse anti-GS (1:500 dilution, Becton and Dickinson)16, 17; rabbit anti–GLT-1 (1:500 dilution) (gift from Masahiko Watanabe)21, 22; sheep anti-CYP1A2 (1:1,000 dilution) (Chemicon, Temecula, CA)23; rabbit anti-CYP2E1 (1:1,000 dilution) (gift from Magnus Ingelman-Sundberg).16, 24 Biotinylated anti-rabbit (Vector Laboratories) and anti-sheep (Jackson Immunoresearch, West Grove, PA) antibodies were used as secondary antibodies at a 1:200 dilution. Mouse monoclonal primary antibodies were detected with an M.O.M. kit (Vector Laboratories) according to the manufacturer's protocol. 3-3′-Diaminobenzidine tetrahydrochloride was used as a chromogen.
Total RNA was prepared from liver tissue samples using Trizol (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. Reverse-transcriptase PCR was performed using a Superscript First-Strand Synthesis System (Invitrogen). PCR reactions were performed in a 25-μL reaction mixture containing 1 × SYBR Green PCR master mix (Applied Biosystems, Foster City, CA) and 300 nmol/L of each primer. Amplification was performed via initial polymerase activation for 10 minutes at 95°C and 40 cycles of 94°C for 30 seconds followed by elongation for 1 minute at 60°C. Primer sequences used are listed in Table 1. To exclude contamination with nonspecific PCR products, melting curve analysis was applied to all final PCR products after the cycling protocol. RNA samples without reverse transcription were also subjected to PCR to exclude contamination of genomic DNA. Expression of liver genes was compared with the expression level of β-glucuronidase as previously described.25
Blood Ammonia Measurement.
Six- to seven-week-old Alb-Cre;β-cateninloxP/loxP (n = 4) and control (n = 4) mice were fed a high-protein diet containing 60% protein (LabDiet, Richmond, IL) for 10 days. At the end of the 10-day feeding period, the mice were sacrificed, the trunk blood was collected into an EDTA-coated tube, the plasma was separated, and ammonia concentration was determined (IDEXX, Westbrook, MA). Mice fed a standard diet containing 20% protein (LabDiet) were used as controls.
Acetaminophen Liver Toxicity.
After overnight fasting, mice received an intraperitoneal injection of acetaminophen (300 mg/kg dissolved in phosphate-buffered saline) or vehicle alone (phosphate-buffered saline). Five Alb-Cre;β-cateninloxP/loxP mice and 5 control group mice were used in each group. Animals were sacrificed 24 hours after acetaminophen administration. Blood samples were collected from the inferior vena cava, and serum alanine aminotransferases (ALT) and aspartate aminotransferase (AST) levels were determined (IDEXX). Liver tissue samples were collected, processed as described in the Histological Analysis section, and evaluated histologically on hematoxylin-eosin–stained sections.
The results are presented as the mean ± SD. Statistical significance was determined using the Student's t test; a P value of less than .05 was considered significant.
Generation of Liver-Specific β-Catenin Knockout Mice.
Alb-Cre;β-cateninloxP/loxP mice were generated by breeding β-cateninloxP/loxP mice with Alb-Cre transgenic mice.18, 19 The resulting Alb-Cre;β-cateninloxP/+ mice were mated to β-cateninloxP/loxP mice to obtain the Alb-Cre;β-cateninloxP/loxP mice and control mice with genotypes that include Alb-Cre;β-cateninloxP/+, β-cateninloxP/loxP, and β-cateninloxP/+. Alb-Cre;β-cateninloxP/loxP pups were born in the expected Mendelian distribution and developed normally to adulthood. Liver tissue from Alb-Cre;β-cateninloxP/loxP mice were histologically normal (Fig. 1A-B). No alterations in liver architecture or hepatocyte morphology were noted. Immunohistochemistry confirmed the loss of β-catenin expression in hepatocytes from Alb-Cre;β-cateninloxP/loxP mice. In control group mice, hepatocytes showed uniform membranous expression of β-catenin (Fig. 1C,E). In Alb-Cre;β-cateninloxP/loxP mice, β-catenin was completely absent from cell borders between hepatocytes, confirming hepatocyte-specific loss of β-catenin (Fig. 1D,F). Weakly positive staining was observed in sinusoidal endothelial cells along the surface of hepatocyte trabeculae.
Hepatocyte-Specific Deletion of β-Catenin Reduces Liver Size.
Whereas the hepatocyte-specific ablation of β-catenin did not have a measureable effect on whole body weight (Fig. 2A), the liver weights of Alb-Cre;β-cateninloxP/loxP mice were significantly lower than those of control mice (Fig. 2B). Comparison of the liver/body weight ratios revealed an approximately 20% reduction in liver mass in Alb-Cre;β-cateninloxP/loxP mice (Fig. 2C). Thus, hepatocyte-specific loss of β-catenin results in reduced liver size.
Previous reports have described that β-catenin can transactivate transcription of some genes that promote cell proliferation, including CCND1, c-myc, and EGFR.26–28 We sought to determine whether downregulation of these genes was responsible for reduced liver size in Alb-Cre;β-cateninloxP/loxP mice. However, quantitative PCR analysis did not reveal significantly altered expression of these genes (Fig. 2D). We attempted to determine differences in hepatocyte proliferation by using immunohistochemistry against Ki-67 and BrdU staining on sections of 6-week-old mouse livers. Although the proliferative activity in both control and transgenic mice was too low for reliable quantitative analysis by Ki-67 and BrdU (<0.5% range), no obvious differences were observed between Alb-Cre;β-cateninloxP/loxP and control mice. Immunohistochemistry against cleaved caspase-3 and terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling were performed to determine the level of apoptosis. Few apoptotic cells were detected via staining for cleaved caspase-3 and in terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling, while no significant differences were noted between Alb-Cre;β-cateninloxP/loxP and control mice (data not shown).
β-Catenin Is Indispensable for Expression of Genes Involved in Glutamine Synthesis.
Analysis of human and mouse liver tumors suggested β-catenin–dependent upregulation of some genes in the glutamine synthesis pathway.15–17 To determine whether expression of these genes depends on β-catenin function, we determined the transcript levels of 3 pathway genes, GS, GLT-1, and OAT, and a gene involved in ammonia uptake, rhesus blood group B glycoprotein (RhBG).
Quantitative PCR revealed almost complete loss of all 4 genes in liver tissue isolated from Alb-Cre;β-cateninloxP/loxP mice (Fig. 3A). To confirm that reduction in messenger RNA correlates with protein reduction, we performed immunohistochemistry for GS and GLT-1. In the control littermates, expression of GS and GLT-1 is confined to a few cell layers of hepatocytes surrounding the central veins (Fig. 3B,D). In contrast, expression of these proteins was almost completely lost in the Alb-Cre;β-cateninloxP/loxP mice (Fig. 3C,E).
Given the almost complete ablation of critical components of the glutamine synthesis pathway, we next sought to determine the functional consequence of these alterations. Circulating glutamine is mostly derived from skeletal muscle and lung, and the relative contribution of liver to the pool of circulating glutamine is considered minor.29 However, in the liver, the glutamine synthesis pathway is involved in the detoxification of blood ammonia, a by-product of protein metabolism. Ammonia production increases with increased protein intake; thus, we decided to determine serum ammonia levels in Alb-Cre;β-cateninloxP/loxP and control mice that were fed a standard (20% protein) or high-protein diet (60% protein). No changes in plasma ammonia levels were found in Alb-Cre;β-cateninloxP/loxP and control mice fed the standard diet (Fig. 3F). In contrast, high-protein dietary feeding significantly increased the plasma ammonia levels of Alb-Cre;β-cateninloxP/loxP mice, while ammonia levels in control mice remained unaffected. Thus, loss of β-catenin signaling in hepatocytes severely disrupts the glutamine synthesis pathway and results in impaired ammonia detoxification.
β-Catenin Is Required for CYP1A2 and CYP2E1 Expression.
A recent study reported elevated expression of a subset of cytochrome P450 enzymes in some liver tumors marked by the presence of β-catenin mutations.16 We therefore sought to determine whether β-catenin also regulates the expression of cytochrome P450 enzymes in nonneoplastic liver tissue. We examined the expression of 5 genes coding for cytochrome p450 enzymes via quantitative PCR. CYP1A2 and CYP2E1 were found to be downregulated by ten- and 200-fold, respectively, in liver tissue from Alb-Cre;β-cateninloxP/loxP compared with control samples (Fig. 4A). In contrast, CYP2C29 expression was reduced by 50%, and CYP1A1 and CYP3A11 expression were unaffected.
Immunohistochemistry was used to confirm that ablation of CYP1A2 and CYP2E1 messenger RNA transcripts resulted in loss of protein expression. Liver tissue from control mice showed expression of CYP1A2 and CYP2E1 in a centrilobular pattern (Fig. 4B,D). Both enzymes were found in cells reaching from the perivenous zone to the midhepatic lobule, with CYP2E1 showing a broader pattern. In liver tissue isolated from Alb-Cre;β-cateninloxP/loxP mice, expression of both CYP1A2 and CYP2E1 was almost completely abolished (Fig. 4C,E). Thus, β-catenin is essential for proper expression of a subset of cytochrome P450 enzymes.
Cytochrome P450 enzymes are involved in the metabolism of xenobiotic agents, and both CYP1A2 and CYP2E1 are known to metabolize acetaminophen.30, 31 Cytochrome P450-mediated modification of acetaminophen results in formation of highly reactive metabolites that cause liver cell necrosis. To examine the functional significance of the loss of CYP1A2 and CYP2E1, we challenged Alb-Cre;β-cateninloxP/loxP and control mice with acetaminophen. Release of intracellular liver enzymes, including AST and ALT, serve as a quantitative indicator of liver damage upon acetaminophen challenge. Twenty-four hours after acetaminophen administration, serum AST and ALT levels were considerably elevated in control mice, whereas Alb-Cre;β-cateninloxP/loxP mice showed only slight elevation of both liver enzymes (Fig. 5A-B). Concordant with the elevation of serum liver enzymes, control mice developed confluent necrosis of hepatocytes in the centrilobular areas (Fig. 5C). Remarkably, there were virtually no signs of liver injury in Alb-Cre;β-cateninloxP/loxP mice (Fig. 5D). Thus, Alb-Cre;β-cateninloxP/loxP mice are highly resistant to acetaminophen-induced liver toxicity.
Wnt/β-catenin signaling has been shown to play pivotal roles in cell proliferation and stem cell maintenance in several organs.3 For instance, mice lacking TCF4, a transcription factor activated by β-catenin, completely lack the proliferative compartment of the intestinal crypts.32 Previous studies on liver organ culture and liver tumors suggested a similar role for β-catenin in hepatocyte proliferation in both physiological and neoplastic conditions.4, 5, 13 Transgenic expression of a constitutively active form of β-catenin resulted in hepatocyte proliferation and hepatomegaly.6 Consistent with these findings, our studies show that Alb-Cre;β-cateninloxP/loxP mice develop low liver weight compared with mice in the control group. However, the effects of β-catenin ablation on liver size are relatively small. Liver weight was reduced by 20% in Alb-Cre;β-cateninloxP/loxP mice, yet changes in liver architecture were not found. Furthermore, these mice are viable and fertile, and no changes in feeding behavior or growth under normal breeding conditions were observed. These findings suggest that β-catenin loss does not interfere with essential liver functions. In addition, 3 previously reported β-catenin target genes that regulate cell proliferation, including one identified in the liver, were not significantly downregulated in Alb-Cre;β-cateninloxP/loxP mice.
A potential explanation for the limited effects on liver formation and hepatocyte proliferation comes from the findings that Cre recombinase in Alb-Cre mice is only expressed during the late stages of embryonic development. At birth, only 40% of hepatocytes have undergone recombination, and complete excision of loxP flanked elements is accomplished by 6 weeks after birth.20 Therefore, although our findings in Alb-Cre;β-cateninloxP/loxP mice partially confirm the positive role of β-catenin in hepatocyte proliferation during the later stages of liver development, the role of β-catenin function during the early stages of liver development cannot be assessed. Liver Cre strains that direct expression of the recombinase at earlier stages would be required to determine the role of β-catenin during early liver development.
Ammonia metabolization is one of the most important liver functions. Ammonia produced by protein degradation in the gut is detoxified by two major pathways present at different locations in the liver: the urea synthesis and the glutamine synthesis pathways.33 Ammonia enters the liver through the portal vein and is first eliminated by the high-capacity urea synthesis pathway that is active in periportal to midzonal hepatocytes. In the second system, glutamine synthesis is confined to the perivenous area and serves to detoxify the remaining ammonia that has not been converted to urea.33 Thus, the sequence of a low affinity, but high capacity system (urea synthesis) followed by a high affinity system (glutamine synthesis) ensures effective and complete elimination of blood ammonia.
Three genes involved in glutamine synthesis—GS, OAT, and GLT-1—have previously been reported as β-catenin target genes in liver tumors.15 In Alb-Cre;β-cateninloxP/loxP mice, the expression of these genes is almost completely lost, indicating that β-catenin is required for transcription of these genes under physiological conditions. In addition, expression of RhBG, another molecule involved in ammonium uptake,34 is also abolished in Alb-Cre;β-cateninloxP/loxP mice. Thus, β-catenin function is essential to ensure expression of genes required for glutamine synthesis in hepatocytes.
Although the expression of genes required for glutamine synthesis is abolished in Alb-Cre;β-cateninloxP/loxP mice, plasma ammonia levels remained normal when mice were fed a standard diet containing normal amounts of protein. The glutamine synthesis pathway likely plays a supplementary role in ammonia detoxification, because its function is readily compensated by other pathways under normal feeding conditions. This finding is in agreement with a recent study on the pulmonary trunk banding model which demonstrated how the elevation of serum ammonia is not correlated with the disappearance of GS expression, but with the severity of heart failure.35 This report implied that blood ammonia could be kept low even in the absence of a functional glutamine synthesis pathway. However, our results suggest that the function of this pathway is critical when mice are fed with a high-protein diet. Under these conditions, plasma ammonia level increased significantly in Alb-Cre;β-cateninloxP/loxP mice, suggesting an essential role for the glutamine synthesis pathway as a backup system to other pathways involved in ammonia detoxification. Thus, our studies reveal that β-catenin function is required to regulate plasma ammonia levels under conditions of increased ammonia production.
It is known that hepatocellular tumors harboring β-catenin mutations overexpress cytochrome P450 enzymes.16 Here we demonstrate that β-catenin regulates the expression of several cytochrome P450 enzymes also under physiological conditions. We found that 3 out of 5 cytochrome P450 genes examined are significantly downregulated in Alb-Cre;β-cateninloxP/loxP mice. In particular, expression of CYP1A2 and CYP2E1 is almost completely abolished. Inspection of the promoter regions of CYP1A2 and CYP2E1 identified multiple putative TCF/LEF binding sites. Future studies will address whether transcription of these genes is regulated by the direct binding of TCF/β-catenin complex to their promoters.
Cytochrome P450 enzymes play a major role in the metabolism of xenobiotic agents, and both CYP1A2 and CYP2E1 have been implicated in acetaminophen metabolism.30, 31 Acetaminophen is one of the most widely used analgesic and antipyretic drugs. However, overdose can cause fatal centrilobular hepatic necrosis.36, 37 Cytochrome P450 enzymes transform acetaminophen into a reactive metabolite that covalently binds critical cellular proteins. These interactions impair cellular function and consequently lead to cell death.30, 31 Although both CYP2E1 and CYP1A2 are involved in acetaminophen toxicity, their contributions differ, with CYP2E1 being the enzyme predominantly involved in this process.30 Studies of CYP1A2- and CYP2E1-null mice have elucidated the roles of both enzymes in acetaminophen toxicity. Whereas CYP1A2-null mice are somewhat protected against high concentrations of acetaminophen, CYP2E1-null mice are highly resistant to acetaminophen toxicity.38, 39 Mice lacking both CYP1A2 and CYP2E1 have been shown to be almost completely resistant to acetaminophen challenge.40
As expected by the loss of CYP1A2 and CYP2E1 expression, Alb-Cre;β-cateninloxP/loxP mice are resistant to acetaminophen toxicity. In contrast to the prominent hepatocyte necrosis observed in control mice, Alb-Cre;β-cateninloxP/loxP mice do not show histologically detectable signs of liver injury. In addition, serum ALT and AST are only marginally elevated in Alb-Cre;β-cateninloxP/loxP mice. Thus, our results indicate that modulation of β-catenin function could be used to regulate hepatic response to xenobiotic agents metabolized by CYP1A2 and CYP2E1, including acetaminophen, chloroform, and carbon tetrachloride.39 One advantage of this strategy would be that β-catenin is not required for fundamental liver function as indicated by the present study. Although genetic ablation of β-catenin gene function during acetaminophen toxicity would not be feasible, it should be noted that small chemical inhibitors of β-catenin function have been developed.41 Similar agents might eventually allow efficient and rapid inhibition of β-catenin function.
The authors thank Dr. Magnus Ingelman-Sundberg (Karolinska Institute, Stockholm, Sweden) and Dr. Masahiko Watanabe (Hokkaido University, Sapporo, Japan) for providing antibodies and Pedro Gutierrez for reviewing the manuscript.