Mechanism of primitive duct formation in the pancreas and submandibular glands: a role for SDF-1.

BACKGROUND: The exocrine pancreas is composed of a branched network of ducts connected to acini. They are lined by a monolayered epithelium that derives from the endoderm and is surrounded by mesoderm-derived mesenchyme. The morphogenic mechanisms by which the ductal network is established as well as the signaling pathways involved in this process are poorly understood. RESULTS: By morphological analyzis of wild-type and mutant mouse embryos and using cultured embryonic explants we investigated how epithelial morphogenesis takes place and is regulated by chemokine signaling. Pancreas ontogenesis displayed a sequence of two opposite epithelial transitions. During the first transition, the monolayered and polarized endodermal cells give rise to tissue buds composed of a mass of non polarized epithelial cells. During the second transition the buds reorganize into branched and polarized epithelial monolayers that further differentiate into tubulo-acinar glands. We found that the second epithelial transition is controlled by the chemokine Stromal cell-Derived Factor (SDF)-1. The latter is expressed by the mesenchyme, whereas its receptor CXCR4 is expressed by the epithelium. Reorganization of cultured pancreatic buds into monolayered epithelia was blocked in the presence of AMD3100, a SDF-1 antagonist. Analyzis of sdf1 and cxcr4 knockout embryos at the stage of the second epithelial transition revealed transient defective morphogenesis of the ventral and dorsal pancreas. Reorganization of a globular mass of epithelial cells in polarized monolayers is also observed during submandibular glands development. We found that SDF-1 and CXCR4 are expressed in this organ and that AMD3100 treatment of submandibular gland explants blocks its branching morphogenesis. CONCLUSION: In conclusion, our data show that the primitive pancreatic ductal network, which is lined by a monolayered and polarized epithelium, forms by remodeling of a globular mass of non polarized epithelial cells. Our data also suggest that SDF-1 controls the branching morphogenesis of several exocrine tissues.

Threshold levels of hepatocyte nuclear factor 6 (HNF-6) acting in synergy with HNF-4 and PGC-1alpha are required for time-specific gene expression during liver development.

During liver development, hepatocytes undergo a maturation process that leads to the fully differentiated state. This relies at least in part on the coordinated action of liver-enriched transcription factors (LETFs), but little is known about the dynamics of this coordination. In this context we investigate here the role of the LETF hepatocyte nuclear factor 6 (HNF-6; also called Onecut-1) during hepatocyte differentiation. We show that HNF-6 knockout mouse fetuses have delayed expression of glucose-6-phosphatase (g6pc), which catalyzes the final step of gluconeogenesis and is a late marker of hepatocyte maturation. Using a combination of in vivo and in vitro gain- and loss-of-function approaches, we demonstrate that HNF-6 stimulates endogenous g6pc gene expression directly via a synergistic and interdependent action with HNF-4 and that it involves coordinate recruitment of the coactivator PGC-1alpha. The expression of HNF-6, HNF-4, and PGC-1alpha rises steadily during liver development and precedes that of g6pc. We provide evidence that threshold levels of HNF-6 are required to allow synergism between HNF-6, HNF-4, and PGC-1alpha to induce time-specific expression of g6pc. Our observations on the regulation of g6pc by HNF-6 provide a model whereby synergism, interdependency, and threshold concentrations of LETFs and coactivators determine time-specific expression of genes during liver development.

The Onecut transcription factors HNF-6/OC-1 and OC-2 regulate early liver expansion by controlling hepatoblast migration.

Liver development in mammals is initiated by the formation of a hepatic bud from the ventral foregut endoderm. The hepatic cells then proliferate and invade the septum transversum mesenchyme, and further differentiate to give rise to hepatocytes and biliary cells. By analyzing mice that are knockout for the transcription factors Hepatocyte Nuclear Factor-6 (HNF-6)/Onecut-1 (OC-1) and OC-2, we show here that these factors redundantly stimulate the degradation of the basal lamina surrounding the liver bud and promote hepatoblast migration in the septum transversum. Gene expression analysis indicates that HNF-6 and OC-2 belong to a gene network comprising E-cadherin, thrombospondin-4 and osteopontin, which regulates liver bud expansion by controlling hepatoblast migration and adhesion. This network operating at the onset of liver development contains candidate genes for investigation of liver carcinogenesis.

Role of metalloproteinases at the onset of liver development.

At the onset of liver development, the hepatic precursor cells, namely, the hepatoblasts, derive from the ventral foregut endoderm and form a bud surrounded by a basement membrane (BM). To initiate liver growth, the hepatoblasts migrate across the BM and invade the neighboring septum transversum mesenchyme. In the present study, carried out in the mouse embryo, we searched for effectors involved in this process and we examined the role of matrix metalloproteinases (MMPs). We found expression of a broad range of MMPs, among which MMP-2 was predominantly expressed in the septum transversum and MMP-14 in the hepatoblasts. Using a new liver explant culture system we showed that inhibition of MMP activity represses migration of the hepatoblasts. We conclude that MMPs are required to initiate expansion of the liver during development and that our culture system provides a new model to study hepatoblast migration.

A vHNF1/TCF2-HNF6 cascade regulates the transcription factor network that controls generation of pancreatic precursor cells.

Generation of pancreatic precursor cells in the endoderm is controlled by a network of transcription factors. Hepatocyte nuclear factor-6 (HNF6) is a key player in this network, because it controls the initiation of the expression of pancreatic and duodenal homeobox 1 (Pdx1), the earliest marker of pancreatic precursor cells. To further characterize this network, we have investigated how the expression of HNF6 is controlled in mouse endoderm, by using in vitro and in vivo protein-DNA interaction techniques combined with endoderm electroporation, transgenesis, and gene inactivation in embryos. We delineated Hnf6 regulatory regions that confer expression of a reporter gene in the embryonic endoderm but not in extraembryonic visceral endoderm. HNF6 expression in the embryonic endoderm was found to depend on an intronic enhancer. This enhancer contains functional binding sites for the tissue-specific factors of the forkhead box A and HNF1 families. Among the latter, variant HNF1 (vHNF1)/TCF2, which is expressed before HNF6 in the endoderm, was found to be critical for HNF6 expression. Therefore, the sequential activation of vHNF1, HNF6, and Pdx1 in the endoderm appears to control the generation of pancreatic precursors. This cascade may be used to benchmark in vitro differentiation of pancreatic precursor cells from embryonic stem cells, for cell therapy of diabetes.

Eph receptors and their ephrin ligands are expressed in developing mouse pancreas.

Pancreas development involves branching morphogenesis concomitantly to differentiation of endocrine, exocrine and ductal cell types from a single population of pancreatic precursors. These processes depend on many signals and factors that also control development of the central nervous system. In the latter, Eph receptors and their class-A (GPI-anchored) and class-B (transmembrane) ephrin ligands control cell migration and axon-pathfinding, help establish regional patterns and act as labels for cell positioning. This raised the question as to whether and where Ephs and ephrins are expressed during pancreas development. Here we have identified the Eph and ephrin genes that are expressed in mouse embryonic pancreas, as detected by RT-PCR analysis. In situ hybridization experiments showed that Ephs and ephrins are mainly expressed in the burgeoning structures of the epithelium which differentiate into exocrine acini. Binding experiments on whole pancreas demonstrated the presence of functional Eph receptors. They showed that EphBs are expressed by the pancreatic epithelium at embryonic day (e) 12.5 and that, from e14.5 on, Ephs of both classes are expressed by the pancreatic epithelium and then become restricted to developing acini. We conclude that specific members of the Eph/ephrin family are expressed in embryonic pancreas according to a dynamic temporal and regional pattern.

6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase: head-to-head with a bifunctional enzyme that controls glycolysis.

Fru-2,6-P2 (fructose 2,6-bisphosphate) is a signal molecule that controls glycolysis. Since its discovery more than 20 years ago, inroads have been made towards the understanding of the structure-function relationships in PFK-2 (6-phosphofructo-2-kinase)/FBPase-2 (fructose-2,6-bisphosphatase), the homodimeric bifunctional enzyme that catalyses the synthesis and degradation of Fru-2,6-P2. The FBPase-2 domain of the enzyme subunit bears sequence, mechanistic and structural similarity to the histidine phosphatase family of enzymes. The PFK-2 domain was originally thought to resemble bacterial PFK-1 (6-phosphofructo-1-kinase), but this proved not to be correct. Molecular modelling of the PFK-2 domain revealed that, instead, it has the same fold as adenylate kinase. This was confirmed by X-ray crystallography. A PFK-2/FBPase-2 sequence in the genome of one prokaryote, the proteobacterium Desulfovibrio desulfuricans, could be the result of horizontal gene transfer from a eukaryote distantly related to all other organisms, possibly a protist. This, together with the presence of PFK-2/FBPase-2 genes in trypanosomatids (albeit with possibly only one of the domains active), indicates that fusion of genes initially coding for separate PFK-2 and FBPase-2 domains might have occurred early in evolution. In the enzyme homodimer, the PFK-2 domains come together in a head-to-head like fashion, whereas the FBPase-2 domains can function as monomers. There are four PFK-2/FBPase-2 isoenzymes in mammals, each coded by a different gene that expresses several isoforms of each isoenzyme. In these genes, regulatory sequences have been identified which account for their long-term control by hormones and tissue-specific transcription factors. One of these, HNF-6 (hepatocyte nuclear factor-6), was discovered in this way. As to short-term control, the liver isoenzyme is phosphorylated at the N-terminus, adjacent to the PFK-2 domain, by PKA (cAMP-dependent protein kinase), leading to PFK-2 inactivation and FBPase-2 activation. In contrast, the heart isoenzyme is phosphorylated at the C-terminus by several protein kinases in different signalling pathways, resulting in PFK-2 activation.

Gene transfer into mouse prepancreatic endoderm by whole embryo electroporation.

CONTEXT: Understanding gene function in the developing pancreas is a major issue for pancreatic cell therapy. The in vivo analysis of gene function has essentially been performed by modulating gene expression in transgenesis. A faster and easier method is electroporation of mouse embryos. This technique, coupled with whole embryo culture, enables one to deliver genes and analyze their effects in a spatially and temporally regulated manner. OBJECTIVE: We wanted to adapt the electroporation technique for gene transfer of whole e8.5 mouse embryos into the endoderm to allow expression of transgenes in the pancreas or liver. RESULTS: Using two platinum plate electrodes, low voltage and a precise positioning of the embryo in the electroporation cuvette we could target and express DNA constructs in the prepancreatic or prehepatic territories, identified with cell markers. We also demonstrated that this technique is a valuable tool in the study of transcriptional regulation in the developing endoderm. CONCLUSIONS: Targeted electroporation of whole embryos is a useful method of characterizing the gene network which controls pancreatic development.

Cloning and embryonic expression pattern of the mouse Onecut transcription factor OC-2.

Onecut (OC) transcription factors are evolutionarily conserved proteins with important developmental functions. They contain a bipartite DNA-binding domain composed of a single cut domain associated with a divergent homeodomain. The human genome contains three Onecut paralogues, Hnf6 (also called Oc1), Oc2 and Oc3. We describe here the cloning of mouse (m) OC-2 and its expression pattern in the mouse embryo. The mOc2 gene was localized on chromosome 18. Analysis of the mOC-2 amino acid sequence revealed overall identities of 67% with mHNF-6 and of 56% with mOC-3, and the presence of functional domains delineated earlier in HNF-6. The sequence of the 153 residue-long cut-homeodomain was very conserved, as it was 92% identical to that of mHNF-6 and 89% identical to that of mOC-3. In situ hybridization showed expression of mOc2 in the developing nervous system and gut endoderm. Like Hnf6, Oc2 was expressed in developing liver and pancreas. As many genes that are targeted by Onecut factors are recognized by both OC-2 and HNF-6, this overlap of expression patterns may have functional implications.

Fifty years ago: the quest for steroid hormone receptors.

In 1963 Peter Karlson put forward the revolutionary "hormone-gene" hypothesis, which would change drastically the way in which steroid hormones were thought to act at the time. From a historical perspective, this review relates the acceptance of this initially controversial idea, the discovery of the steroid receptors and the key experiments that have led to the current understanding of the mechanism of steroid hormone action. It shows how, over 50years, the field has widened beyond all expectation and has contributed to major advances not only in endocrinology, but also in molecular biology, pharmacology and therapeutics.

Role of the Onecut transcription factors in pancreas morphogenesis and in pancreatic and enteric endocrine differentiation.

The Onecut (OC) transcription factor HNF-6 (OC-1) is required during embryogenesis for pancreatic specification, morphogenesis and endocrine differentiation. In mammals, HNF-6 has two paralogs, OC-2 and OC-3, which share DNA-binding and transcriptional activation properties and have expression patterns that overlap with that of HNF-6. This suggested that OC-2 and OC-3 play redundant roles with HNF-6 in pancreas development. Here, we have addressed this hypothesis by analyzing the phenotype of mice knockout for the Onecut factors. We found that neither OC-2 nor OC-3 is required for pancreas specification. However, OC-2 plays partially redundant roles with HNF-6 in pancreas morphogenesis and in the differentiation of endocrine precursors. As similar molecular events drive endocrine differentiation in the pancreas and gastrointestinal tract, we also investigated if Onecut factors are involved in enteroendocrine differentiation. OC-2 and OC-3 were found to delineate specific antero-posterior regions of the gut around embryonic day 12.5. Later on, OC2 was expressed in several gut cell types, whereas OC-3 behaved as a specific marker of the enteroendocrine lineage. However, OC-2 and OC-3, alone or in combination, were dispensable for gut development and enteroendocrine differentiation. In conclusion, our data reveal partially redundant roles for HNF-6 and OC-2 in developing pancreas and identify new markers for antero-posterior patterning of the gut and for enteroendocrine differentiation.

An endothelial-mesenchymal relay pathway regulates early phases of pancreas development.

Understanding the tissue interactions that induce pancreatic progenitor cells from the embryonic endoderm provides insights into congenital malformations, tissue repair, and differentiating stem cells to a pancreatic fate. The specification of pancreatic progenitors within the dorsal endodermal epithelium has been thought to involve two phases of mesodermal interactions; first with the lateral plate mesoderm and notochord and then with aortic endothelial cells. Afterwards, branching morphogenesis of the pancreatic bud is induced by Isl-1-positive dorsal mesenchyme cells, whose growth is stimulated by factors in the circulation. Using mouse genetic models and embryo tissue explants, we show that the aortic endothelium and dorsal mesenchyme each possess an additional role in pancreatic induction, prior to the branching morphogenesis step. Specifically, we find that aortic endothelial cells promote the survival of nearby, Isl-1-positive dorsal mesenchyme, independently of factors from the circulation. Furthermore, we find that FGF10 signaling from the mesenchyme cells maintains Ptf1a expression in the dorsal pancreatic bud and appears genetically redundant with a role for the transcription factor gene HNF6 in promoting the induction of Pdx-1-positive dorsal endoderm. Together, these studies reveal a relay pathway from aortic endothelium to dorsal mesenchyme and then to the endoderm, along with functions of the dorsal mesenchyme that promote the initial differentiation of the dorsal pancreatic endoderm, prior to organ morphogenesis.

The transcription factor hepatocyte nuclear factor-6 controls the development of pancreatic ducts in the mouse.

BACKGROUND & AIMS: A number of hereditary polycystic diseases are associated with formation of cysts within the pancreatic ducts. The cysts result from abnormal tubulogenesis, but how normal pancreatic duct development is controlled remains poorly understood. Here, we investigate the transcriptional mechanisms that control pancreatic duct development by addressing the role of the transcription factor hepatocyte nuclear factor (HNF)-6. METHODS: Using immunostaining, we have determined the expression pattern of HNF-6 in pancreatic ducts during mouse development. Hnf6 null mice at various stages of development were studied by immunolocalization methods to assess the morphology, differentiation, and proliferation status of ductal cells. The expression of genes involved in hereditary polycystic diseases was determined by real-time, reverse-transcription polymerase chain reaction (RT-PCR). RESULTS: We show that HNF-6 is expressed in the pancreatic duct epithelium throughout development and that, in the absence of HNF-6, duct morphogenesis is perturbed. Although development of the intercalated ducts is normal, cysts appear within the interlobular and intralobular ducts. This is associated with abnormal development of primary cilia at the apical pole of the duct cells and with reduced expression of a set of genes involved in polycystic diseases, namely those coding for HNF-1beta and for the cilium-associated proteins polyductin/fibrocystin and cystin. CONCLUSIONS: We identify HNF-6 as the first transcriptional regulator of pancreatic duct development and reveal the existence of different regulatory mechanisms in distinct duct compartments. HNF-6 controls a network of genes involved in cilium formation and in hereditary polycystic diseases. Finally, HNF-6 deficiency represents a genetically defined model of pancreatic cystic disease.

Control of liver cell fate decision by a gradient of TGF beta signaling modulated by Onecut transcription factors.

During liver development, hepatocytes and biliary cells differentiate from common progenitors called hepatoblasts. The factors that control hepatoblast fate decision are unknown. Here we report that a gradient of activin/TGFbeta signaling controls hepatoblast differentiation. High activin/TGFbeta signaling is required near the portal vein for differentiation of biliary cells. The Onecut transcription factors HNF-6 and OC-2 inhibit activin/TGFbeta signaling in the parenchyma, and this allows normal hepatocyte differentiation. In the absence of Onecut factors, the shape of the activin/TGFbeta gradient is perturbed and the hepatoblasts differentiate into hybrid cells that display characteristics of both hepatocytes and biliary cells. Thus, a gradient of activin/TGFbeta signaling modulated by Onecut factors is required to segregate the hepatocytic and the biliary lineages.

The Onecut transcription factor HNF-6 (OC-1) is required for timely specification of the pancreas and acts upstream of Pdx-1 in the specification cascade.

The pancreas derives from cells in the ventral and dorsal foregut endoderm that express the transcription factor Pdx-1. These specified cells give rise to the precursors of the endocrine, ductal, and exocrine pancreatic cells. The identification of transcription factors that regulate the onset of Pdx-1 expression is therefore essential to understand pancreas development. No such factor that acts both in the ventral and in the dorsal endoderm is known. We showed previously that the Onecut transcription factor HNF-6 promotes differentiation of the endocrine cell precursors in which it stimulates expression of the proendocrine gene Ngn-3. By analyzing the phenotype of HNF-6 null mice, we now demonstrate that HNF-6 also controls an earlier step in pancreas development. Indeed, the pancreas of Hnf6(-/-) mice was hypoplastic. This did not result from decreased proliferation or from increased apoptosis, but from retarded pancreatic specification of endodermal cells. The onset of Pdx-1 expression was delayed both in the ventral and in the dorsal endoderm, leading to a reduction in the number of endodermal cells expressing Pdx-1 at the time of pancreatic budding. In normal embryos, HNF-6 was detected in the endoderm prior to the expression of Pdx-1. Moreover, HNF-6 could directly stimulate the Pdx1 promoter. Our data therefore identify HNF-6 as the first factor known to control Pdx-1 expression both in the ventral and in the dorsal endoderm. We conclude that HNF-6 controls the timing of pancreas specification and that HNF-6 acts upstream of Pdx-1 in this developmental process. Together with the known role of HNF-6 in pancreatic endocrine cell differentiation, our data point to HNF-6 as a key regulator of pancreas development.

The transcription factor hepatocyte nuclear factor-6/Onecut-1 controls the expression of its paralog Onecut-3 in developing mouse endoderm.

During development, the endoderm gives rise to several organs, including the pancreas and liver. This differentiation process requires spatial and temporal regulation of gene expression in the endoderm by a network of tissue-specific transcription factors whose elucidation is far from complete. These factors include the Onecut protein hepatocyte nuclear factor-6 (HNF-6), which controls pancreas and liver development as shown in our previous work on Hnf6 knock-out embryos. In mammals, HNF-6 has two paralogs, Onecut-2 (OC-2) and OC-3, whose patterns of expression in the adult overlap with that of HNF-6. In the present work, we examine the expression profile of the three Onecut factors in the developing mouse endoderm. We show that HNF-6, OC-2, and OC-3 are expressed sequentially, which defines new steps in endoderm differentiation. By analyzing Hnf6 knock-out embryos we find that HNF-6 is required for expression of the Oc3 gene in the endoderm. We show that OC-3 colocalizes with HNF-6 in the endoderm and in embryonic pancreas and liver. Based on transfection, chromatin immunoprecipitation, and whole embryo electroporation experiments, we demonstrate that HNF-6 can bind to and stimulate the expression of the Oc3 gene. This study identifies a regulatory cascade between two paralogous transcription factors, sheds new light on the interpretation of the Hnf6 knock-out phenotype, and broadens the transcription factors network operating during development of the endoderm, liver, and pancreas.

Shh-dependent differentiation of intestinal tissue from embryonic pancreas by activin A.

The pancreas develops from the endoderm to give rise to ducts, acini and islets of Langerhans. This process involves extracellular signals of the Transforming Growth Factor beta (TGFbeta) family. The aim of this work was to study the effects of activin A, a member of this family, whose potential role in pancreas differentiation is controversial. To this end, we used pancreatic explants from E12.5 mouse embryos. In culture these explants exhibited spontaneous growth, epithelial morphogenesis and endocrine and exocrine differentiation. Exposure to activin A did not affect exocrine or endocrine differentiation. Surprisingly, activin A induced in the explants the appearance of a large contractile structure surrounded by a cylindrical epithelium, a thick basal lamina and a smooth muscle layer. This structure, the formation of which was prevented by follistatin, was typical of an intestinal wall. Consistent with this interpretation, activin A rapidly induced in the explants the mRNAs for fatty acid binding proteins (FABPs), which are markers of the intestine, but not of the pancreas. We also found that induction of the FABPs was preceded by induction of Sonic hedgehog (Shh), a known inducer of intestinal differentiation in the endoderm. Activin B induced neither Shh nor intestinal differentiation. The activin A-mediated intestinal differentiation was blocked by cyclopamine, an inhibitor of Hedgehog signaling, and it was mimicked by Shh. We conclude that activin A does not appear to affect the exocrine or endocrine components of the pancreas, but that it can promote differentiation of pancreatic tissue into intestine via a Shh-dependent mechanism. These findings illustrate the plasticity of differentiation programs in response to extracellular signals in the pancreas and they shed new light on the regulation of pancreas and intestinal development.

The onecut transcription factor hepatocyte nuclear factor-6 controls B lymphopoiesis in fetal liver.

Mouse genetic models have helped to identify transcription factors that are expressed by hemopoietic cells and control their differentiation into lymphoid cells. However, little is known on transcription factors that are involved in this process, but are expressed in nonhemopoietic cells of the microenvironment. We show in this study that inactivation of the gene coding for hepatocyte nuclear factor-6 (HNF-6) in mice led to B lymphopenia in the bone marrow and spleen. This phenotype disappeared shortly after birth when fetal B lymphopoiesis is no longer active, pointing to a defect in fetal liver. Indeed, the number of B cells was decreased in this organ as well. An analysis of B cell developmental markers in fetal liver cells showed that B lymphopoiesis was impaired just beyond the pre-pro B cell stage. Hemopoietic cells from hnf6(-/-) fetal liver could reconstitute the lymphoid system when injected into scid mice. Because parenchymal cells, but not hemopoietic cells, expressed hnf6 in normal liver, we concluded that HNF-6 controls B lymphopoiesis in fetal liver and that HNF-6 exerts this control indirectly by acting in parenchymal cells. The involvement, in the B cell defect of hnf6(-/-) fetuses, of genes known to exert such an indirect control was ruled out by expression analysis, including microarrays, and by in vivo rescue experiments. This work identifies HNF-6 as the first noncell-intrinsic transcription factor known to control B lymphopoiesis specifically in fetal liver.

OC-3, a novel mammalian member of the ONECUT class of transcription factors.

Transcription factors of the ONECUT class possess a single cut domain and a divergent homeodomain. They regulate gene networks by controlling the expression of other transcription factors and they play an important role in cell differentiation and metabolism. We identified earlier in mammals HNF-6 (ONECUT-1), the founding member of the class, and ONECUT-2 (OC-2). We have now characterized in the mouse a third ONECUT member, which we call OC-3. Its gene is located on chromosome 10. The sequence of OC-3 (490 residues) displays 51% amino acid identity with HNF-6 and 50% with OC-2. OC-3 has a DNA-binding specificity similar to that of HNF-6 and it is a stimulator of gene transcription. OC-3 mRNA is found in brain, stomach, and upper intestine in the adult and embryonic mouse. Our earlier work on HNF-6 and the expression patterns of the three mammalian ONECUT genes suggest that they all participate to the control of organ development from the foregut and midgut endoderm.

Hepatocyte nuclear factor-6 stimulates transcription of the alpha-fetoprotein gene and synergizes with the retinoic-acid-receptor-related orphan receptor alpha-4.

The rat alpha-fetoprotein ( afp ) gene is controlled by three enhancers whose function depends on their interaction with liver-enriched transcription factors. The afp enhancer III, located at -6 kb, is composed of three regions that act in synergy. Two of these regions, called s1 and s2, contain a putative binding site for hepatocyte nuclear factor-6 (HNF-6). This factor is the prototype of the ONECUT family of cut-homoeodomain proteins and is a known regulator of liver gene expression in adults and during development. We show here that the two splicing isoforms of HNF-6 bind to a site in the s1 region and in the s2 region. The core sequence of the s1 site corresponds to none of the known HNF-6 binding sites. Nevertheless, the binding properties of the s1 site are identical with those of the s2 site and of previously characterized HNF-6 binding sequences. The HNF-6 consensus should therefore be rewritten as DRRTCVATND. Binding of HNF-6 to the s1 and s2 sites requires both the cut and the homoeo domains, is co-operative and induces DNA bending. HNF-6 strongly stimulates the activity of the afp enhancer III in transient transfection experiments. This effect requires the stereo-specific alignment of the two HNF-6 sites. Moreover, HNF-6 stimulates the enhancer in synergy with the retinoic-acid-receptor-related orphan receptor alpha (RORalpha), which binds to a neighbouring site in the s1 region. Thus expression of the afp gene requires functional interactions between HNF-6 molecules and between HNF-6 and RORalpha.