Alagille syndrome is a developmental disorder caused predominantly by mutations in the Jagged1 (JAG1) gene, which encodes a ligand for Notch family receptors. A characteristic feature of Alagille syndrome is intrahepatic bile duct paucity. We described previously that mice doubly heterozygous for Jag1 and Notch2 mutations are an excellent model for Alagille syndrome. However, our previous study did not establish whether bile duct paucity in Jag1/Notch2 double heterozygous mice resulted from impaired differentiation of bile duct precursor cells, or from defects in bile duct morphogenesis.
Here we characterize embryonic biliary tract formation in our previously described Jag1/Notch2 double heterozygous Alagille syndrome model, and describe another mouse model of bile duct paucity resulting from liver-specific deletion of the Notch2 gene.
Citation: Lozier J, McCright B, Gridley T (2008) Notch Signaling Regulates Bile Duct Morphogenesis in Mice. PLoS ONE 3(3): e1851. doi:10.1371/journal.pone.0001851
Editor: Hernan Lopez-Schier, Centre de Regulacio Genomica, Spain
Received: January 15, 2008; Accepted: February 21, 2008; Published: March 26, 2008
Copyright: © 2008 Lozier et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by grants for the National Institute of Diabetes and Digestive and Kidney Diseases (DK066387; TG) and the National Institute of Neurological Disorders and Stroke (NS036437; TG). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The primary functional cells of the mammalian liver are the hepatocytes and the epithelial bile duct cells, or cholangiocytes (for recent reviews, see references –). During liver development, both hepatocytes and cholangiocytes differentiate from bipotential progenitor cells termed hepatoblasts , . Hepatoblasts located in the liver parenchyma differentiate into hepatocytes, while hepatoblasts located at the interface of the portal mesenchyme (which surrounds the portal vein) and the liver parenchyma differentiate into the biliary epithelial cells. Initially, biliary epithelial cells form a continuous single cell layer termed the ductal plate (reviewed in ). The ductal plate subsequently undergoes morphogenesis and remodeling to generate the epithelial bile ducts. Defects in bile duct formation can lead to an impairment of bile duct flow (cholestasis), and result in a diverse group of both genetic and acquired biliary tract disorders termed cholangiopathies (reviewed in , ).
The Notch signaling pathway is an evolutionarily conserved intercellular signaling mechanism (reviewed in , ), and mutations in its components disrupt embryonic development in diverse organisms and cause inherited disease syndromes in humans. Mutations in the JAG1 gene, which encodes a ligand for Notch family receptors, cause Alagille syndrome , . Alagille syndrome (OMIM #118450) is a pleiotropic developmental disorder characterized by cholestasis and jaundice caused by intrahepatic bile duct paucity, congenital heart defects, vertebral defects, eye abnormalities, facial dysmorphism, and kidney abnormalities –. Alagille syndrome exhibits autosomal dominant inheritance, and analysis of the types of JAG1 mutations in Alagille syndrome patients suggest JAG1 haploinsufficiency as the primary cause of Alagille syndrome.
We have described previously a mouse model for Alagille syndrome . Mice heterozygous for a Jag1 null allele, which have the same genotype as Alagille syndrome patients, exhibited haploinsufficient eye defects but did not exhibit other phenotypic abnormalities characteristic for Alagille syndrome . However, mice doubly heterozygous for a Jag1 null allele and a Notch2 hypomorphic allele exhibited most of the clinically relevant features of Alagille syndrome, including bile duct paucity . Our previous studies of these mice concentrated on analysis of late embryonic and postnatal livers, and did not establish whether bile duct paucity in Jag1/Notch2 double heterozygous mice was due to defects in differentiation of bile duct precursors from the bipotential hepatoblast, or defects in morphogenesis of the ductal plate.
A recent study of Hairy and enhancer of split 1 (Hes1)-null mice suggested that the role of Notch signaling during biliary development was in the control of biliary tract morphogenesis, rather than in a hepatocyte-cholangiocyte cell fate specification decision . However, other genes encoding Hes-related bHLH proteins are also Notch targets, raising the possibility that Hes1-null mice may not reflect the full extent of the role played by the Notch signaling pathway during biliary development. In addition, since Hes1-null mice die perinatally from defects unrelated to the liver defects , morphogenesis and maturation of the intrahepatic biliary system cannot be followed during the early postnatal period when major biliary tract remodeling and maturation events take place .
In this paper, we characterize embryonic biliary tract formation in the previously described Jag1/Notch2 double heterozygote mouse model of Alagille syndrome. We also describe another mouse model of bile duct paucity resulting from liver-specific deletion of the Notch2 gene. Our data demonstrate a requirement for Jag1/Notch2-mediated signaling in bile duct formation in mice, and support a model in which bile duct paucity in Notch pathway loss of function mutant mice results from defects in bile duct morphogenesis rather than cell fate specification.
Analysis of bile duct morphogenesis during embryogenesis in Jag1dDSL/+ Notch2del1/+ double heterozygous mice
Our previous study  analyzed late embryonic and postnatal livers, and did not establish whether bile duct paucity in mice doubly heterozygous for a Jag1 null allele (Jag1dDSL)  and a Notch2 hypomorphic allele (Notch2del1)  was due to defects in differentiation of bile duct precursors from the bipotential hepatoblast, or whether it was due to defects in morphogenesis of the ductal plate. Therefore, we analyzed livers of Jag1dDSL/+ Notch2del1/+ double heterozygous mice by cytokeratin immunostaining from embryonic day (E) 16.5 through postnatal day (P) 7. At E16.5 in control littermate embryos, cytokeratin immunostaining revealed the presence of a partly bilayered ductal plate at the interface of the portal mesenchyme and the liver parenchyma (Fig. 1A). Over the next several days, the ductal plate remodels by a process in which focal dilations appear between the two cell layers of the plate (Fig. 1C,E). By P7, some of these focal dilations give rise to patent epithelial bile ducts incorporated into the portal mesenchyme (Fig. 1G), while the remainder of the ductal plate involutes. Cytokeratin immunostaining of liver sections from Jag1dDSL/+ Notch2del1/+ double heterozygous mice revealed that they were very similar to control littermate sections through at least P0. In the Jag1dDSL/+ Notch2del1/+ mice, a ductal plate formed (Fig. 1B) and focal dilations appeared (Fig. 1D,F). However, postnatal remodeling to form a patent epithelial bile duct did not occur. Instead, as we reported in our initial study , by P7 only ductal plate remnants remained in most portal tracts (Fig. 1H). These results indicate that in the Jag1/Notch2 double heterozygote mouse, bile duct paucity results from defects in bile duct morphogenesis, not from defects in differentiation of bile duct precursors from the bipotential hepatoblast.
Cytokeratin immunostaining of control littermate and Jag1dDSL/+ Notch2del1/+ liver sections at the indicated ages. A,B. At E16.5, both control (A) and Jag1dDSL/+ Notch2del1/+ (B) have formed a partly bilayered ductal plate (arrowheads). C–F. Over the next several days, focal dilations (arrowheads) form in the ductal plate of both control and mutant embryos. Other regions of the ductal plate begin to regress. G,H. At P7, the focal dilations have formed epithelial bile ducts incorporated into the portal mesenchyme (arrows) in the control liver (G), while the Jag1dDSL/+ Notch2del1/+ liver (H) exhibits only ductal plate remnants (arrowheads).
Liver-specific Notch2 deletion results in defects in bile duct morphogenesis, but not ductal plate formation
Our previous study was the first to implicate a critical role for the Notch2 gene in bile duct formation . The Notch2 protein is expressed in periportal hepatoblasts near or adjacent to Jag1-expressing cells surrounding the portal veins in mice , , . Further support for a critical role for the Notch2 gene in bile duct formation and/or maintenance comes from recent studies on Alagille syndrome patients. While improved mutation detection protocols can now identify JAG1 mutations in approximately 94% of patients diagnosed with Alagille syndrome , there are still some Alagille syndrome patients in whom no JAG1 mutations can be identified. Recently, heterozygous NOTCH2 mutations were identified in a subset of Alagille syndrome patients who lack JAG1 mutations .
To specifically assess the role of the Notch2 gene in bile duct formation in mice, we disrupted Notch2 function in the liver utilizing mice expressing Cre recombinase under the control of the Albumin 1 promoter (Alb1-Cre) , . We crossed Notch2flox/Notch2flox mice with mice doubly heterozygous for the Alb1-Cre transgene and either the Notch2del2 or Notch2del3 alleles. Both of these Notch2 mutant alleles behave genetically as null alleles . Offspring with the genotypes Alb1-Cre/+; Notch2del2/Notch2flox or Alb1-Cre/+; Notch2del3/Notch2flox were analyzed. Since no differences were detected in the phenotypes of the Alb1-Cre/+; Notch2del2/Notch2flox and Alb1-Cre/+; Notch2del3/Notch2flox mice, mice of both genotypes were designated Notch2-cko (for Notch2 conditional knockout) in this report. Excision of the Notch2flox allele was observed in liver DNA of Notch2-cko mice, but not in kidney DNA of these mice (Fig. 2).
Southern blot of SacI/EcoRI- digested DNA isolated at P4. Excision of the Notch2flox allele (asterisk indicates the excised allele) was observed only in liver DNA of Notch2-cko mice (lane 2), but not in kidney DNA of Notch2-cko mice (lanes 5,6). Genotype of Notch2-cko mice is Notch2flox/Notch2del2; Alb1-Cre/+.
Notch2-cko mice were smaller than their littermates (Fig. 3A), and at P8-P9 were approximately 19% lighter than their littermates (4.3±0.1 grams Notch2-cko versus 5.3±0.1 grams control littermates). This weight difference was maintained through at least 4–5 weeks of age, when Notch2-cko mice were approximately 15% lighter than their littermates. Gross examination of the livers of the Notch2-cko mice revealed that Notch2-cko livers exhibited focal areas of necrosis (Fig. 3C). These necrotic areas may arise from disruption of bile acid flow, since bile acids are strong detergents and buildup in the liver can lead to necrosis, fibrosis and cirrhosis .
A. Notch2-cko mouse (bottom) and control littermate (top) at P3. The Notch2-cko mouse exhibits jaundice and growth retardation. B,C. Livers at P18. The Notch2-cko liver exhibits focal areas of necrosis (arrowheads).
The biliary tract defects exhibited by Notch2-cko mice were very similar to those exhibited by Jag1dDSL/+ Notch2del1/+ mice (compare Fig. 1 with Fig. 4). Examination of histological sections of the livers of Notch2-cko mice revealed that few morphologically identifiable bile ducts were present (Fig. 4B). Analysis of Dolichos biflorus agglutinin (DBA) lectin expression, a cholangiocyte marker , revealed that DBA-positive cells formed patent bile ducts adjacent to the portal veins in littermate control mice (Fig. 4C,E). In Notch2-cko mice, DBA-positive cells were present in small numbers adjacent to the portal veins, but these cells were not arranged into patent epithelial ducts (Fig. 4D,F). Biliary tract defects were similar using either the Notch2del2 (Fig. 4B,D) or Notch2del3 (Fig. 4F) allele in combination with the Notch2flox allele. Similarly to portal tracts of Jag1dDSL/+ Notch2del1/+ mice, cytokeratin immunostaining revealed that by P7 only ductal plate remnants were detected in most Notch2-cko portal tracts (Fig. 4H), while well-formed bile ducts incorporated into the portal mesenchyme were present in the littermate controls (Fig. 4G).
A,B. Hematoxylin and eosin-stained sections at P7 of control littermate (CT) and Notch2-cko mice using the Notch2del2 allele. Bile ducts (arrow) are observed in the periportal region of the control littermate (A), but not the Notch2-cko mouse (B). C–F. DBA lectin staining. C,D. Control littermate and Notch2-cko mice using the Notch2del2 allele at P7. E,F. Control littermate and Notch2-cko mice using the Notch2del3 allele at P3. DBA-positive cells form patent bile ducts (arrows) adjacent to the portal veins in control mice (C,E). In Notch2-cko mice using either the Notch2del2 (D) or Notch2del3 (F) allele, small numbers of DBA-positive cells (arrowheads) are present adjacent to the portal veins, but these cells have not formed patent ducts. G,H. Cytokeratin immunostaining of control littermate and Notch2-cko mice (Notch2del2 allele) at P7. The ductal plate of the control liver (G) has remodeled into epithelial bile ducts (arrows), while the Notch2-cko liver (H) exhibits only ductal plate remnants (arrowheads).
At 4–5 weeks of age, clinical chemistry analysis of serum revealed that, as a group, Notch2-cko mice had elevated levels of alkaline phosphatase, alanine aminotransferase, and total bilirubin (Table 1). Elevated levels of these parameters are indicative of liver and biliary dysfunction. However, some Notch2-cko mice had alkaline phosphatase, alanine aminotransferase, and total bilirubin levels within the normal range. We also tested blood urea nitrogen levels, which when elevated is indicative of kidney dysfunction. As expected, blood urea nitrogen levels in Notch2-cko mice were not elevated (Table 1), in contrast to Jag1dDSL/+ Notch2del1/+ mice .
Notch signaling regulates bile duct morphogenesis independently of HNF6 and HNF1β expression
Previous studies have shown that biliary tract morphogenesis is dependent on the transcription factors Hepatocyte Nuclear Factor-6 (Hnf6; Onecut1 – Mouse Genome Informatics) and HNF1β (Tcf2 – Mouse Genome Informatics). Mice homozygous for a targeted null mutation of the Hnf6 gene , or with liver-specific deletion of the Hnf1b gene , fail to properly remodel the ductal plate to form patent bile ducts and exhibit persistence of ductal plate remnants. HNF1β expression was strongly downregulated in livers of Hnf6-null mice, indicating that the Hnf6 gene functioned upstream of the Hnf1b gene .
We tested by immunohistochemistry whether the HNF6 and HNF1β proteins were expressed in the periportal region of Jag1dDSL/+ Notch2del1/+ and Notch2-cko mice. HNF1β protein expression was observed in the periportal region of Notch2-cko mice at P0 and P7 (Fig. 5B,D). Similarly, HNF6 protein expression was observed in the periportal region of Jag1dDSL/+ Notch2del1/+ mice at P0 (Fig. 5F). These data suggest that the etiology of the bile duct morphogenesis defects in the Notch pathway mutants is independent of the function of the HNF6 and HNF1β proteins. Independent functioning of the Notch pathway and the HNF6/HNF1β pathway is supported by the finding that Jag1 and Hes1 expression is unaffected in fetal livers of Hnf6-null mice .
A–D. HNF1β expression in Notch2-cko mice at P7 and P0. E,F. HNF6 expression in Jag1dDSL/+ Notch2del1/+ mice at P0. The HNF1β and HNF6 proteins were expressed similarly in the periportal region (arrowheads) of both control littermate and mutant livers.
The Notch signaling pathway is frequently utilized to specify cell fate during bipotential cell fate decisions , , so an attractive model to explain the defects in bile duct formation in Jag1dDSL/+ Notch2del1/+ mice was reduced differentiation of cholangiocytes from the bipotential hepatoblast. The first indication that this model was likely incorrect came from analysis of mice homozygous for a null mutation of the Hes1 gene, which encodes a basic helix-loop-helix protein that is a downstream effector of the Notch pathway. The Hes1-null mice formed a relatively-normal ductal plate consisting of cytokeratin- and DBA-positive cholangiocyte precursors, suggesting that the primary defect in these mice was not in the initial bipotential cell fate decision of the hepatoblast . However, by P0 in wildtype littermates, patent bile ducts were beginning to form, while none were evident in the Hes1-null mice. Unfortunately, Hes1-null mice die at birth from severe central nervous system defects , precluding the analysis of later stages of ductal plate remodeling and bile duct morphogenesis in these mice.
Our analysis of Jag1dDSL/+ Notch2del1/+ mice supports the model that Notch signaling regulates ductal plate remodeling and bile duct morphogenesis rather than cholangiocyte differentiation, and suggests that the Jag1/Notch2-mediated signal responsible for bile duct morphogenesis acts, at least in part, by modulating Hes1 expression. Our analysis of bile duct formation in Notch2-cko mice is consistent with this model. While the Alb1-Cre transgene does not delete early enough during embryogenesis to study the role of Notch2 gene function during ductal plate formation , the essentially identical biliary tract defects exhibited by the Jag1dDSL/+ Notch2del1/+ and Notch2-cko mice at late embryonic and postnatal stages strongly suggest that these defects arise by the same mechanism in both mouse models. However, it remains possible that Notch signaling may play some role in cholangiocyte differentiation, since none of the three mouse models analyzed (Hes1-null mice, Jag1dDSL/+ Notch2del1/+ mice, and Notch2-cko mice) are likely to be entirely deficient in Notch signaling when the cholangiocyte-hepatocyte cell fate decision is made.
In contrast to Notch2 deletion, deletion of the Jag1 gene in liver hepatoblasts did not lead to defects in bile duct development , suggesting that Jag1 expression in endothelial cells and/or vascular smooth muscle cells was sufficient for signaling to Notch2-expressing hepatoblasts during ductal plate remodeling and bile duct morphogenesis. Interestingly, this study also demonstrated that in mice that were compound heterozygotes for a Jag1 null allele and the Jag1 conditional allele deleted in hepatoblasts, a subset of animals exhibited bile duct proliferation . Other recent studies support a model in which cholangiocyte differentiation is controlled by a gradient of Activin/TGFβ signaling that is controlled by the expression of Onecut-family transcription factors, such as HNF6 and Onecut2 (OC2) , . Our results suggest that Notch signaling regulates bile duct morphogenesis independently of the Activin/TGFβ/Onecut pathway.
In summary, we demonstrate here that similar defects in bile duct formation were observed in both Jag1dDSL/+ Notch2del1/+ and Notch2-cko mice. However, Jag1dDSL/+ Notch2del1/+ mice exhibit defects in many organ systems other than the biliary tract, such as the heart and the kidney . We suggest that liver-specific deletion of the Notch2 gene in Notch2-cko mice represents an improved and more specific model than Jag1dDSL/+ Notch2del1/+ mice for studying the role of Notch signaling during bile duct morphogenesis and remodeling.
Materials and Methods
Jag1dDSL, Notch2del1, Notch2del2, Notch2del3, and Notch2flox mice were described previously , , , . Albumin-Cre (Alb1-Cre) mice ,  were obtained from the Jackson Laboratory. To produce Alb1-Cre/+; Notch2flox/- mice (referred to as Notch2-cko, for Notch2 conditional knockout), Notch2flox/flox mice were mated to mice heterozygous for both the Alb1-Cre transgene and either the Notch2del2 or Notch2del3 alleles. Animal maintenance and experimental procedures were in accordance with the NIH Guidelines for Animal Care and Use and the principles of the Helsinki Declaration, and were approved by the Institutional Animal Care and Use Committee of the Jackson Laboratory.
Immunohistochemistry and lectin binding
The antibodies and lectins used in these studies were rabbit polyclonal anti-human cytokeratins (Dako, Cat. A0575); rabbit polyclonal anti-HNF1β (Santa Cruz, Cat. sc-22840); rabbit polyclonal anti-HNF6 (Santa Cruz, Cat. sc-13050); and biotinylated Dolichos biflorus agglutinin (DBA) lectin (Vector Laboratories, Cat. B-1035). Mutant sections were either Jag1dDSL/+ Notch2del1/+ or Notch2-cko. Other genotypes, with the exception of Jag1dDSL/+, were used as littermate controls. No differences were noted in the phenotypes exhibited by the different control genotypes.
Conceived and designed the experiments: TG JL BM. Performed the experiments: JL. Analyzed the data: TG JL. Wrote the paper: TG.
- 1. Zaret KS (2002) Regulatory phases of early liver development: paradigms of organogenesis. Nat Rev Genet 3: 499–512.
- 2. Lemaigre F, Zaret KS (2004) Liver development update: new embryo models, cell lineage control, and morphogenesis. Curr Opin Genet Dev 14: 582–590.
- 3. Duncan SA (2003) Mechanisms controlling early development of the liver. Mech Dev 120: 19–33.
- 4. Strick-Marchand H, Weiss MC (2003) Embryonic liver cells and permanent lines as models for hepatocyte and bile duct cell differentiation. Mech Dev 120: 89–98.
- 5. Lemaigre FP (2003) Development of the biliary tract. Mech Dev 120: 81–87.
- 6. Carlton VE, Pawlikowska L, Bull LN (2004) Molecular basis of intrahepatic cholestasis. Ann Med 36: 606–617.
- 7. Lazaridis KN, Strazzabosco M, Larusso NF (2004) The cholangiopathies: disorders of biliary epithelia. Gastroenterology 127: 1565–1577.
- 8. Bray SJ (2006) Notch signalling: a simple pathway becomes complex. Nat Rev Mol Cell Biol 7: 678–689.
- 9. Ehebauer M, Hayward P, Martinez-Arias A (2006) Notch, a universal arbiter of cell fate decisions. Science 314: 1414–1415.
- 10. Li L, Krantz ID, Deng Y, Genin A, Banta AB, et al. (1997) Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nature Gen 16: 243–251.
- 11. Oda T, Elkahloun AG, Pike BL, Okajima K, Krantz ID, et al. (1997) Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nature Gen 16: 235–242.
- 12. Spinner NB, Krantz ID (2003) JAG1 and the Alagille Syndrome. In: Epstein CJ, Erickson RP, Wynshaw-Boris A, editors. Inborn errors of development. New York: Oxford University Press. pp. 461–469.
- 13. Emerick KM, Rand EB, Goldmuntz E, Krantz ID, Spinner NB, et al. (1999) Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology 29: 822–829.
- 14. Alagille D, Estrada A, Hadchouel M, Gautier M, Odievre M, et al. (1987) Syndromic paucity of interlobular bile ducts (Alagille syndrome or arteriohepatic dysplasia): review of 80 cases. J Pediatr 110: 195–200.
- 15. McCright B, Lozier J, Gridley T (2002) A mouse model of Alagille syndrome: Notch2 as a genetic modifier of Jag1 haploinsufficiency. Development 129: 1075–1082.
- 16. Xue Y, Gao X, Lindsell CE, Norton CR, Chang B, et al. (1999) Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1. Hum Mol Genet 8: 723–730.
- 17. Kodama Y, Hijikata M, Kageyama R, Shimotohno K, Chiba T (2004) The role of Notch signaling in the development of intrahepatic bile ducts. Gastroenterology 127: 1775–1786.
- 18. Ishibashi M, Ang SL, Shiota K, Nakanishi S, Kageyama R, et al. (1995) Targeted disruption of mammalian hairy and Enhancer of split homolog-1 (HES-1) leads to up-regulation of neural helix-loop-helix factors, premature neurogenesis, and severe neural tube defects. Genes Dev 9: 3136–3148.
- 19. McCright B, Gao X, Shen L, Lozier J, Lan Y, et al. (2001) Defects in development of the kidney, heart and eye vasculature in mice homozygous for a hypomorphic Notch2 mutation. Development 128: 491–502.
- 20. Tanimizu N, Miyajima A (2004) Notch signaling controls hepatoblast differentiation by altering the expression of liver-enriched transcription factors. J Cell Sci 117: 3165–3174.
- 21. Warthen DM, Moore EC, Kamath BM, Morrissette JJ, Sanchez P, et al. (2006) Jagged1 (JAG1) mutations in Alagille syndrome: increasing the mutation detection rate. Hum Mutat 27: 436–443.
- 22. McDaniell R, Warthen DM, Sanchez-Lara PA, Pai A, Krantz ID, et al. (2006) NOTCH2 mutations cause Alagille syndrome, a heterogeneous disorder of the notch signaling pathway. Am J Hum Genet 79: 169–173.
- 23. Postic C, Shiota M, Niswender KD, Jetton TL, Chen Y, et al. (1999) Dual roles for glucokinase in glucose homeostasis as determined by liver and pancreatic beta cell-specific gene knock-outs using Cre recombinase. J Biol Chem 274: 305–315.
- 24. Postic C, Magnuson MA (2000) DNA excision in liver by an albumin-Cre transgene occurs progressively with age. Genesis 26: 149–150.
- 25. McCright B, Lozier J, Gridley T (2006) Generation of new Notch2 mutant alleles. Genesis 44: 29–33.
- 26. Shiojiri N, Nagai Y (1992) Preferential differentiation of the bile ducts along the portal vein in the development of mouse liver. Anat Embryol (Berl) 185: 17–24.
- 27. Clotman F, Lannoy VJ, Reber M, Cereghini S, Cassiman D, et al. (2002) The onecut transcription factor HNF6 is required for normal development of the biliary tract. Development 129: 1819–1828.
- 28. Coffinier C, Gresh L, Fiette L, Tronche F, Schutz G, et al. (2002) Bile system morphogenesis defects and liver dysfunction upon targeted deletion of HNF1beta. Development 129: 1829–1838.
- 29. Sund NJ, Ang SL, Sackett SD, Shen W, Daigle N, et al. (2000) Hepatocyte nuclear factor 3beta (Foxa2) is dispensable for maintaining the differentiated state of the adult hepatocyte. Mol Cell Biol 20: 5175–5183.
- 30. Loomes KM, Russo P, Ryan M, Nelson A, Underkoffler L, et al. (2007) Bile duct proliferation in liver-specific Jag1 conditional knockout mice: effects of gene dosage. Hepatology 45: 323–330.
- 31. Clotman F, Jacquemin P, Plumb-Rudewiez N, Pierreux CE, Van der Smissen P, et al. (2005) Control of liver cell fate decision by a gradient of TGF beta signaling modulated by Onecut transcription factors. Genes Dev 19: 1849–1854.
- 32. Clotman F, Lemaigre FP (2006) Control of hepatic differentiation by activin/TGFbeta signaling. Cell Cycle 5: 168–171.