- Email: [email protected]
Organogenesis: Making Pancreas from Liver
Current Biology, Vol. 13, R96–R98, February 4, 2003, ©2003 Elsevier Science Ltd. All rights reserved.
Organogenesis: Making Pancreas from Liver Valerie A. McLin1 and Aaron M. Zorn2
Making endocrine pancreas cells at will is one of the major goals of cellular based therapies for diabetes. The experimentally induced conversion of hepatocytes into pancreatic cells, using a modified version of the transcription factor Pdx-1, may provide an alternative to stem cell approaches.
The promise of tissue replacement therapy — that is, of replacing a body’s damaged or diseased cells with healthy or genetically corrected cells— has driven an intensive search for methods to produce the desired cells or tissue types in vitro. For example, the ability to generate insulin-producing pancreatic β cells in vitro, which could be transplanted into patients, would be invaluable in treating hundreds of millions of diabetics worldwide. Most of the research aimed at producing therapeutically useful cells has focused on the identification, isolation and in vitro differentiation of stem cells. As reported recently in Current Biology, Horb and co-workers [1] have taken an alternative approach, attempting to exploit the phenomenon of ‘transdifferentiation’ — the conversion of one differentiated cell type into another. They have been able to induce the conversion of liver cells into both endocrine and exocrine pancreatic cells by expressing in hepatocytes a modified version of the pancreatic homeobox gene Pdx-1. We shall review their results, highlighting the significant advances and commenting on some of the future hurdles that need to be tackled before this type of approach might reach its therapeutic potential. Transdifferentiation is the term used to describe the process whereby one differentiated cell type transforms into another differentiated cell or tissue [2]. In vivo, this phenomenon is more commonly known as metaplasia and is frequently deemed a premalignant lesion [3,4]. Transdifferentiation between the liver and pancreas has been observed in human autopsy and biopsy specimens. For example, transformation of many endodermal organs into hepatic tissue has been observed in pathology specimens of digestive tract tumors from adenocarcinoma of the stomach to pancreatic islet cell and ductular tumors [5]. Liver to pancreas transformations have also been observed in liver cirrhosis, where the regenerating liver adopts an acinar histology resembling exocrine pancreatic tissue [6]. Experimental models for transdifferentia -tion between liver and the exocrine pancreas have Cincinnati Children’s Hospital Medical Center, 1Division of Gastroenterology, Hepatology and Nutrition, 2Division of Developmental Biology, 3333 Burnet Avenue, Cincinnati, Ohio 45229-3039, USA. E-mail: [email protected]
PII S0960-9822(03)00036-8
Dispatch
also been documented. For example, liver can be converted into exocrine pancreas tissue in rats treated with polychlorinated biphenyls [7] or in fish treated with diethylnitrosamine [8]. But in all reported cases of pancreas transdifferentiated from liver, only exocrine tissue and never endocrine pancreas tissue has been observed. With the aim of converting liver cells into pancreas tissue, including endocrine cells, Horb et al. [1] developed a transgenic strategy using Xenopus and expressed a superactive form of the pancreatic homeobox gene Pdx-1 (also known as Xlhbox8) [9] in the Xenopus embryonic liver [1]. Pdx-1 is a homeobox transcription factor which is thought to be a key determinant of pancreatic development. Targeted deletion of Pdx-1 in mouse results in a failure of the pancreas to form [10,11] and Pdx-1 mutations in humans have been linked to pancreatic agenesis [12]. These results suggested that Pdx-1 might be a ‘master regulatory’ gene specifying the pancreatic lineage [13], though earlier attempts to produce ectopic pancreas tissue by overexpressing Pdx-1 were unsuccessful [14,15]. Horb et al. [1] reasoned that the earlier attempts to induce a complete conversion of liver into pancreatic tissue might have failed because Pdx-1 normally needs other pancreas-specific cofactors to transactivate its target genes and hence initiate pancreatic development. They sought to overcome this by fusing the potent VP16 transcriptional activation domain from Herpes simplex virus [16] to Pdx-1, thus making a super-active form of the protein, which they call Xlhbox8-VP16, that might not require cofactors. Using the liver-specific transthyretin (TTR) promoter, Xlhbox8-VP16 was expressed in the liver buds of transgenic embryos [1]. The transgene construct also included a green fluorescent protein (GFP) reporter under control of the pancreas-specific elastase promoter, which directs expression in both the exocrine and endocrine pancreas. This made it possible for Horb et al. [1] to monitor the transition of liver to pancreas tissue in real time. They found that TTRXlhbox8:VP16 transgenic animals indeed expressed ‘pancreatic-specific’ GFP in the liver bud, suggesting the liver was transformed to pancreatic tissue. In support of this, they observed ectopic expression of endogenous endocrine pancreas markers, insulin and glucagon, in what is normally liver tissue, with a reciprocal loss of endogenous hepatic markers [1]. It is unclear, however, if the cells expressing ectopic pancreas markers go on to form histologically recognizable pancreatic structures indicating functional exocrine and endocrine tissue formation. Importantly, Horb et al. [1] found that the Xlhbox8VP16 transgene was not expressed until a day or so after the liver bud had started expressing hepatic markers. They argue this indicates that they are observing a transdifferentiation event, rather than
Current Biology R97
redirecting the development of some naïve endodermal precursor common to the liver and pancreatic lineage. To further test this, the authors transfected the human hepatoma-derived cell line HepG2, which exhibits a number of characteristics of differentiated hepatocytes, with the TTR-Xlhbox8:VP16 construct [17]. In agreement with their observations on transgenic embryos, they found that Xlhbox8:VP16 induced in the HepG2 cells ‘pancreatic’ GFP expression from the elastase:GFP reporter, expression of endogenous exocrine and endocrine pancreatic markers, and downregulation of liver gene expression [1]. The take-home message from these ‘proof-ofprinciple’ experiments is that it might be possible to convert a patient’s own hepatocytes into endocrine pancreatic cells (Figure 1) as a means to alleviate diabetes. Obviously, DNA delivery to the tissue of interest is one of the major hurdles to overcome before such gene-therapy becomes a viable therapeutic alternative. Horb et al. [1] make the interesting point that, in the transgenic experiments, the TTR:Xlhbox8-VP16 transgene is shut off as the hepatocytes adopt their new pancreatic fate. Thus, while the transgene expression is transient, the ectopic pancreatic tissue persists. Hypothetically, this has exciting therapeutic possibilities, in that no permanent genetic modification would be necessary if one had a means of transiently transfecting such a construct into an adult liver. The transfected DNA would eventually be lost, but the new endocrine cells might persist. Is this approach applicable to other cell types? Transdifferentiation phenomena are not limited to endodermal organs and have been observed in both mesodermal and ectodermal tissues [18,19]. It appears, though, that the closer the embryological relationship between two cell types, the more likely one is to observe a conversion from one to the other [2,19]. Indeed, the liver and pancreatic lineages arise from a common embryonic endoderm cell population in the anterior foregut. The decision between a hepatic or pancreatic fate is thought to be regulated by only few key developmental steps, such as fibroblast growth factor from the adjacent heart promoting the hepatic over a pancreatic fate [20], and the subsequent expression of Pdx-1 in the presumptive pancreatic lineage (Figure 1). The ability ectopically to activate one of the key developmental regulators underpins the success of the experiments reported by Horb et al. [1]. In order for similar approaches to be used with other cell types, a detailed knowledge of their key developmental regulators would be required. Another important parameter in transdifferentiation is the developmental stage or differentiation state of the cells. It appears that the more immature, fetal or transformed the cell type, the more readily transdifferentiation can occur — as mentioned above, it is often associated with premalignant neoplasias [3]. It could be argued that both the liver bud and the tumorderived HepG2 cells are not true differentiated hepatocytes, but only partially differentiated hepatoblast- type cells which may be more plastic in their ability to transdifferentiate. Thus, to be therapeutically useful it will be important in the future to test if adult hepatocytes can
Normal development
Pluripotent embryonic endoderm +FGF
- FGF
Pancreatic precursors (Pdx-1)
Hepatoblasts
Bile duct cells
Hepatocytes
Endocrine cells
Exocrine cells
Transdifferentiation
Pancreatic precursors (Pdx-1)
Hepatoblasts
Hepatocytes + Pdx1-VP16
Endocrine cells
Exocrine cells Current Biology
Figure 1. (A) Normal embryonic development of the closely related liver and pancreas lineages. Pluripotent endoderm tissue in the developing foregut that receives a fibroblast growth factor (FGF) signal from adjacent heart tissue is specified along a liver lineage and forms hepatoblasts. Nearby cells of the same pluripotent endoderm, which do not receive the cardiac FGF signal, adopt a pancreatic fate and express the homeobox gene Pdx-1. Hepatoblasts go on to differentiate into hepatocytes and bile duct cells, while the pancreatic precursor ultimately gives rise to all the endocrine and exocrine cell types of the pancreas. (B) Experimentally induced transdifferentiation of liver into pancreas. Expression of recombinant Pdx1-VP16 in differentiating hepatocytes converts those cells into pancreatic precursors which go on to give rise to both endocrine and exocrine tissue.
be induced to form functioning pancreatic tissue. A test of this might be to express Pdx1-VP16 in the liver of adult mice with an inducible transgenic strategy and determine if the mature hepatocytes can be converted to pancreas. Clearly, there are still numerous steps to be rigorously examined before this approach can be considered a viable therapeutic option. However, Horb et al. [1] have demonstrated an important and interesting alternative to stem cell approaches, indicating that transdifferentiation may indeed be an important route to effective tissue replacement therapy in the future.
Dispatch R98
References 1. Horb, M.E., Shen, C.-N., Tosh, D. and Slack, J.M.W. (2003). Experimental conversion of liver to pancreas. Curr. Biol. 20 January issue. 2. Tosh, D. and Slack, J.M. (2002). How cells change their phenotype. Nat. Rev. Mol. Cell Biol. 3, 187–194. 3. Zhang, Z., Yuan, X.M., Li, L.H. and Xie, F.P. (2001). Transdifferentiation in neoplastic development and its pathological implication. Histol. Histopathol. 16, 1249–1262. 4. Alison, M.R., Poulsom, R., Forbes, S. and Wright, N.A. (2002). An introduction to stem cells. J. Pathol. 19, 419–423. 5. Paner, G., Thompson, K. and Reyes, C. (2000). Hepatoid carcinoma of the pancreas. Cancer 88, 1583–1589. 6. Wolf, H.K., Burchette, J.L., Jr., Garcia, J.A. and Michalopoulos, G. (1990). Exocrine pancreatic tissue in human liver: a metaplastic process? Am. J. Surg. Pathol. 14, 590–595. 7. Rao, M.S., Bendayan, M., Kimbrough, R.D. and Reddy, J.K. (1986). Characterization of pancreatic-type tissue in the liver of rat induced by polychlorinated biphenyls. J. Histochem. Cytochem. 34, 197–201. 8. Lee, B.C., Hendricks, J.D. and Bailey, G.S. (1989). Metaplastic pancreatic cells in liver tumors induced by diethylnitrosamine. Exp Mol Path 50, 104–113. 9. Wright, C.V., Schnegelsberg, P. and De Robertis, E.M. (1989). XlHbox 8: a novel Xenopus homeo protein restricted to a narrow band of endoderm. Development 105, 787–794. 10. Jonsson, J., Carlsson, L., Edlund, T. and Edlund, H. (1994). Insulinpromoter-factor 1 is required for pancreas development in mice. Nature 371, 606–609. 11. Offield, M.F., Jetton, T.L., Labosky, P.A., Ray, M., Stein, R.W., Magnuson, M.A., Hogan, B.L. and Wright, C.V. (1996). PDX-1 is required for pancreatic outgrowth and differentiation of the rostral duodenum. Development 122, 983–995. 12. Stoffers, D.A., Zinkin, N.T., Stanojevic, V., Clarke, W.L. and Habener, J.F. (1997). Pancreatic agenesis attributable to a single nucleotide deletion in the human IPF1 gene coding sequence. Nat. Genet. 15, 106–110. 13. McKinnon, C.M. and Docherty, K. (2001). Pancreatic duodenal homeobox-1, PDX-1, a major regulator of beta cell identity and function. Diabetologia 44, 1203–1214. 14. Ferber, S., Halkin, A., Cohen, H., Ber, I., Einav, Y., Goldberg, I., Barshack, I., Seijffers, R., Kopolovic, J., Kaiser, N. et al. (2000). Pancreatic and duodenal homeobox gene 1 induces expression of insulin in liver and ameliorates streptozotocin-induced hyperglycemia. Nat. Med. 6, 568–571. 15. Grapin-Botton, A., Majithia, A.R. and Melton, D.A. (2001). Key events of pancreas formation are triggered in gut endoderm by ectopic expression of pancreatic regulatory genes. Genes Dev. 15, 444–454. 16. Sadowski, I., Ma, J., Triezenberg, S. and Ptashne, M. (1988). GAL4VP16 is an unusually potent transcriptional activator. Nature 335, 563–564. 17. Hoekstra, R. and Chamuleau, R. (2002). Recent developments on human cell lines for the bioartificial liver. Int. J. Artif. Organs 25, 182–191. 18. Strutz, F. and Muller, G. (2000). Transdifferentiation comes of age. Nephrol. Dial. Transplant. 15, 1729–1731. 19. Okada, T.S. (1991) Transdifferentiation: flexibility in cell differentiation. Oxford: Clarendon Press. 20. Deutsch, G., Jung, J., Zheng, M., Lora, J. and Zaret, K.S. (2001). A bipotential precursor population for pancreas and liver within the embryonic endoderm. Development 128, 871–881.
Organogenesis: Making Pancreas from Liver Valerie A. McLin1 and Aaron M. Zorn2
Making endocrine pancreas cells at will is one of the major goals of cellular based therapies for diabetes. The experimentally induced conversion of hepatocytes into pancreatic cells, using a modified version of the transcription factor Pdx-1, may provide an alternative to stem cell approaches.
The promise of tissue replacement therapy — that is, of replacing a body’s damaged or diseased cells with healthy or genetically corrected cells— has driven an intensive search for methods to produce the desired cells or tissue types in vitro. For example, the ability to generate insulin-producing pancreatic β cells in vitro, which could be transplanted into patients, would be invaluable in treating hundreds of millions of diabetics worldwide. Most of the research aimed at producing therapeutically useful cells has focused on the identification, isolation and in vitro differentiation of stem cells. As reported recently in Current Biology, Horb and co-workers [1] have taken an alternative approach, attempting to exploit the phenomenon of ‘transdifferentiation’ — the conversion of one differentiated cell type into another. They have been able to induce the conversion of liver cells into both endocrine and exocrine pancreatic cells by expressing in hepatocytes a modified version of the pancreatic homeobox gene Pdx-1. We shall review their results, highlighting the significant advances and commenting on some of the future hurdles that need to be tackled before this type of approach might reach its therapeutic potential. Transdifferentiation is the term used to describe the process whereby one differentiated cell type transforms into another differentiated cell or tissue [2]. In vivo, this phenomenon is more commonly known as metaplasia and is frequently deemed a premalignant lesion [3,4]. Transdifferentiation between the liver and pancreas has been observed in human autopsy and biopsy specimens. For example, transformation of many endodermal organs into hepatic tissue has been observed in pathology specimens of digestive tract tumors from adenocarcinoma of the stomach to pancreatic islet cell and ductular tumors [5]. Liver to pancreas transformations have also been observed in liver cirrhosis, where the regenerating liver adopts an acinar histology resembling exocrine pancreatic tissue [6]. Experimental models for transdifferentia -tion between liver and the exocrine pancreas have Cincinnati Children’s Hospital Medical Center, 1Division of Gastroenterology, Hepatology and Nutrition, 2Division of Developmental Biology, 3333 Burnet Avenue, Cincinnati, Ohio 45229-3039, USA. E-mail: [email protected]
PII S0960-9822(03)00036-8
Dispatch
also been documented. For example, liver can be converted into exocrine pancreas tissue in rats treated with polychlorinated biphenyls [7] or in fish treated with diethylnitrosamine [8]. But in all reported cases of pancreas transdifferentiated from liver, only exocrine tissue and never endocrine pancreas tissue has been observed. With the aim of converting liver cells into pancreas tissue, including endocrine cells, Horb et al. [1] developed a transgenic strategy using Xenopus and expressed a superactive form of the pancreatic homeobox gene Pdx-1 (also known as Xlhbox8) [9] in the Xenopus embryonic liver [1]. Pdx-1 is a homeobox transcription factor which is thought to be a key determinant of pancreatic development. Targeted deletion of Pdx-1 in mouse results in a failure of the pancreas to form [10,11] and Pdx-1 mutations in humans have been linked to pancreatic agenesis [12]. These results suggested that Pdx-1 might be a ‘master regulatory’ gene specifying the pancreatic lineage [13], though earlier attempts to produce ectopic pancreas tissue by overexpressing Pdx-1 were unsuccessful [14,15]. Horb et al. [1] reasoned that the earlier attempts to induce a complete conversion of liver into pancreatic tissue might have failed because Pdx-1 normally needs other pancreas-specific cofactors to transactivate its target genes and hence initiate pancreatic development. They sought to overcome this by fusing the potent VP16 transcriptional activation domain from Herpes simplex virus [16] to Pdx-1, thus making a super-active form of the protein, which they call Xlhbox8-VP16, that might not require cofactors. Using the liver-specific transthyretin (TTR) promoter, Xlhbox8-VP16 was expressed in the liver buds of transgenic embryos [1]. The transgene construct also included a green fluorescent protein (GFP) reporter under control of the pancreas-specific elastase promoter, which directs expression in both the exocrine and endocrine pancreas. This made it possible for Horb et al. [1] to monitor the transition of liver to pancreas tissue in real time. They found that TTRXlhbox8:VP16 transgenic animals indeed expressed ‘pancreatic-specific’ GFP in the liver bud, suggesting the liver was transformed to pancreatic tissue. In support of this, they observed ectopic expression of endogenous endocrine pancreas markers, insulin and glucagon, in what is normally liver tissue, with a reciprocal loss of endogenous hepatic markers [1]. It is unclear, however, if the cells expressing ectopic pancreas markers go on to form histologically recognizable pancreatic structures indicating functional exocrine and endocrine tissue formation. Importantly, Horb et al. [1] found that the Xlhbox8VP16 transgene was not expressed until a day or so after the liver bud had started expressing hepatic markers. They argue this indicates that they are observing a transdifferentiation event, rather than
Current Biology R97
redirecting the development of some naïve endodermal precursor common to the liver and pancreatic lineage. To further test this, the authors transfected the human hepatoma-derived cell line HepG2, which exhibits a number of characteristics of differentiated hepatocytes, with the TTR-Xlhbox8:VP16 construct [17]. In agreement with their observations on transgenic embryos, they found that Xlhbox8:VP16 induced in the HepG2 cells ‘pancreatic’ GFP expression from the elastase:GFP reporter, expression of endogenous exocrine and endocrine pancreatic markers, and downregulation of liver gene expression [1]. The take-home message from these ‘proof-ofprinciple’ experiments is that it might be possible to convert a patient’s own hepatocytes into endocrine pancreatic cells (Figure 1) as a means to alleviate diabetes. Obviously, DNA delivery to the tissue of interest is one of the major hurdles to overcome before such gene-therapy becomes a viable therapeutic alternative. Horb et al. [1] make the interesting point that, in the transgenic experiments, the TTR:Xlhbox8-VP16 transgene is shut off as the hepatocytes adopt their new pancreatic fate. Thus, while the transgene expression is transient, the ectopic pancreatic tissue persists. Hypothetically, this has exciting therapeutic possibilities, in that no permanent genetic modification would be necessary if one had a means of transiently transfecting such a construct into an adult liver. The transfected DNA would eventually be lost, but the new endocrine cells might persist. Is this approach applicable to other cell types? Transdifferentiation phenomena are not limited to endodermal organs and have been observed in both mesodermal and ectodermal tissues [18,19]. It appears, though, that the closer the embryological relationship between two cell types, the more likely one is to observe a conversion from one to the other [2,19]. Indeed, the liver and pancreatic lineages arise from a common embryonic endoderm cell population in the anterior foregut. The decision between a hepatic or pancreatic fate is thought to be regulated by only few key developmental steps, such as fibroblast growth factor from the adjacent heart promoting the hepatic over a pancreatic fate [20], and the subsequent expression of Pdx-1 in the presumptive pancreatic lineage (Figure 1). The ability ectopically to activate one of the key developmental regulators underpins the success of the experiments reported by Horb et al. [1]. In order for similar approaches to be used with other cell types, a detailed knowledge of their key developmental regulators would be required. Another important parameter in transdifferentiation is the developmental stage or differentiation state of the cells. It appears that the more immature, fetal or transformed the cell type, the more readily transdifferentiation can occur — as mentioned above, it is often associated with premalignant neoplasias [3]. It could be argued that both the liver bud and the tumorderived HepG2 cells are not true differentiated hepatocytes, but only partially differentiated hepatoblast- type cells which may be more plastic in their ability to transdifferentiate. Thus, to be therapeutically useful it will be important in the future to test if adult hepatocytes can
Normal development
Pluripotent embryonic endoderm +FGF
- FGF
Pancreatic precursors (Pdx-1)
Hepatoblasts
Bile duct cells
Hepatocytes
Endocrine cells
Exocrine cells
Transdifferentiation
Pancreatic precursors (Pdx-1)
Hepatoblasts
Hepatocytes + Pdx1-VP16
Endocrine cells
Exocrine cells Current Biology
Figure 1. (A) Normal embryonic development of the closely related liver and pancreas lineages. Pluripotent endoderm tissue in the developing foregut that receives a fibroblast growth factor (FGF) signal from adjacent heart tissue is specified along a liver lineage and forms hepatoblasts. Nearby cells of the same pluripotent endoderm, which do not receive the cardiac FGF signal, adopt a pancreatic fate and express the homeobox gene Pdx-1. Hepatoblasts go on to differentiate into hepatocytes and bile duct cells, while the pancreatic precursor ultimately gives rise to all the endocrine and exocrine cell types of the pancreas. (B) Experimentally induced transdifferentiation of liver into pancreas. Expression of recombinant Pdx1-VP16 in differentiating hepatocytes converts those cells into pancreatic precursors which go on to give rise to both endocrine and exocrine tissue.
be induced to form functioning pancreatic tissue. A test of this might be to express Pdx1-VP16 in the liver of adult mice with an inducible transgenic strategy and determine if the mature hepatocytes can be converted to pancreas. Clearly, there are still numerous steps to be rigorously examined before this approach can be considered a viable therapeutic option. However, Horb et al. [1] have demonstrated an important and interesting alternative to stem cell approaches, indicating that transdifferentiation may indeed be an important route to effective tissue replacement therapy in the future.
Dispatch R98
References 1. Horb, M.E., Shen, C.-N., Tosh, D. and Slack, J.M.W. (2003). Experimental conversion of liver to pancreas. Curr. Biol. 20 January issue. 2. Tosh, D. and Slack, J.M. (2002). How cells change their phenotype. Nat. Rev. Mol. Cell Biol. 3, 187–194. 3. Zhang, Z., Yuan, X.M., Li, L.H. and Xie, F.P. (2001). Transdifferentiation in neoplastic development and its pathological implication. Histol. Histopathol. 16, 1249–1262. 4. Alison, M.R., Poulsom, R., Forbes, S. and Wright, N.A. (2002). An introduction to stem cells. J. Pathol. 19, 419–423. 5. Paner, G., Thompson, K. and Reyes, C. (2000). Hepatoid carcinoma of the pancreas. Cancer 88, 1583–1589. 6. Wolf, H.K., Burchette, J.L., Jr., Garcia, J.A. and Michalopoulos, G. (1990). Exocrine pancreatic tissue in human liver: a metaplastic process? Am. J. Surg. Pathol. 14, 590–595. 7. Rao, M.S., Bendayan, M., Kimbrough, R.D. and Reddy, J.K. (1986). Characterization of pancreatic-type tissue in the liver of rat induced by polychlorinated biphenyls. J. Histochem. Cytochem. 34, 197–201. 8. Lee, B.C., Hendricks, J.D. and Bailey, G.S. (1989). Metaplastic pancreatic cells in liver tumors induced by diethylnitrosamine. Exp Mol Path 50, 104–113. 9. Wright, C.V., Schnegelsberg, P. and De Robertis, E.M. (1989). XlHbox 8: a novel Xenopus homeo protein restricted to a narrow band of endoderm. Development 105, 787–794. 10. Jonsson, J., Carlsson, L., Edlund, T. and Edlund, H. (1994). Insulinpromoter-factor 1 is required for pancreas development in mice. Nature 371, 606–609. 11. Offield, M.F., Jetton, T.L., Labosky, P.A., Ray, M., Stein, R.W., Magnuson, M.A., Hogan, B.L. and Wright, C.V. (1996). PDX-1 is required for pancreatic outgrowth and differentiation of the rostral duodenum. Development 122, 983–995. 12. Stoffers, D.A., Zinkin, N.T., Stanojevic, V., Clarke, W.L. and Habener, J.F. (1997). Pancreatic agenesis attributable to a single nucleotide deletion in the human IPF1 gene coding sequence. Nat. Genet. 15, 106–110. 13. McKinnon, C.M. and Docherty, K. (2001). Pancreatic duodenal homeobox-1, PDX-1, a major regulator of beta cell identity and function. Diabetologia 44, 1203–1214. 14. Ferber, S., Halkin, A., Cohen, H., Ber, I., Einav, Y., Goldberg, I., Barshack, I., Seijffers, R., Kopolovic, J., Kaiser, N. et al. (2000). Pancreatic and duodenal homeobox gene 1 induces expression of insulin in liver and ameliorates streptozotocin-induced hyperglycemia. Nat. Med. 6, 568–571. 15. Grapin-Botton, A., Majithia, A.R. and Melton, D.A. (2001). Key events of pancreas formation are triggered in gut endoderm by ectopic expression of pancreatic regulatory genes. Genes Dev. 15, 444–454. 16. Sadowski, I., Ma, J., Triezenberg, S. and Ptashne, M. (1988). GAL4VP16 is an unusually potent transcriptional activator. Nature 335, 563–564. 17. Hoekstra, R. and Chamuleau, R. (2002). Recent developments on human cell lines for the bioartificial liver. Int. J. Artif. Organs 25, 182–191. 18. Strutz, F. and Muller, G. (2000). Transdifferentiation comes of age. Nephrol. Dial. Transplant. 15, 1729–1731. 19. Okada, T.S. (1991) Transdifferentiation: flexibility in cell differentiation. Oxford: Clarendon Press. 20. Deutsch, G., Jung, J., Zheng, M., Lora, J. and Zaret, K.S. (2001). A bipotential precursor population for pancreas and liver within the embryonic endoderm. Development 128, 871–881.