Naturally occurring mutations in the human HNF4alpha gene impair the function of the transcription factor to a varying degree

Abstract

The hepatocyte nuclear factor (HNF)4alpha, a member of the nuclear receptor superfamily, regulates genes that play a critical role in embryogenesis and metabolism. Recent studies have shown that mutations in the human HNF4alpha gene cause a rare form of type 2 diabetes, maturity onset diabetes of the young (MODY1). To investigate the properties of these naturally occurring HNF4alpha mutations we analysed five MODY1 mutations (R154X, R127W, V255M, Q268X and E276Q) and one other mutation (D69A), which we found in HepG2 hepatoma cells. Activation of reporter genes in transfection assays and DNA binding studies showed that the MODY1-associated mutations result in a variable reduction in function, whereas the D69A mutation showed an increased activity on some promoters. None of the MODY mutants acted in a dominant negative manner, thus excluding inactivation of the wild-type factor as a critical event in MODY development. A MODY3-associated mutation in the HNF1alpha gene, a well-known target gene of HNF4alpha, results in a dramatic loss of the HNF4 binding site in the promoter, indicating that mutations in the HNF4alpha gene might cause MODY through impaired HNF1alpha gene function. Based on these data we propose a two-hit model for MODY development.

Figure 1

Naturally occurring mutations within the human HNF4α gene. The domain structure of the HNF4α protein is given (1). The D69A mutation found in HepG2 cells affects an amino acid within the P-box of the first zinc finger of the DNA-binding domain. The diabetes-associated mutations whose functional properties are analysed in this paper are given.

Figure 2

The MODY3 mutation of the HNF4 binding site in the HNF1α promoter results in a loss of function. (A) Schematic representation of the hHNF1(–325/+138)lucII construct containing the human HNF1α promoter is given. The HNF4 binding site with the MODY3 mutation at –58 of the human HNF1α promoter (29) is marked. (B) Comparison of transactivation of the wild-type (wt) and mutant (mut) HNF1α promoter in transient transfection assays by increasing amounts of expression vector encoding wild-type HNF4α in HeLa cells. The total amount of transfected DNA was adjusted by adding pOP13 vector DNA. Fold induction refers to the activity seen without HNF4α expression vector. The error bars indicate the standard deviation of six determinations. (C) Gel shift experiments were done with 0.5 µl of rat liver extract using a labelled HNF4 binding site as probe. The retarded DNA–protein complex and the supershift obtained by an HNF4α-specific antibody are marked with an arrow and an arrowhead, respectively. The addition of increasing amounts of unlabelled wild-type or mutated oligonucleotide representing the HNF4 binding site in (A) is given.

Figure 3

Transactivation potential of wild-type and mutated HNF4α in transient transfection experiments at saturation. The reporter constructs hHNF1(–325/+138)lucII (A) and H4-tk-luc (B) were co-transfected into HeLa cells with saturating amounts of expression vector (300 ng) encoding wild-type or the mutated HNF4α. Fold induction refers to the activity without any HNF4 derivative. The error bars indicate standard deviation of six determinations.

Figure 4

Saturation curves of transactivation potential of wild-type and mutated HNF4α in transfection assays. (A) Increasing amounts of the expression vectors encoding the HNF4α mutants as given were co-transfected with the reporter construct hHNF1(–325/+138)lucII. The dotted and the solid lines represent the activity of the mutant and the wild-type, respectively. (B) The expression vector encoding the HNF4α mutant D69A was co-transfected with either the Xenopus HNF1(–594/207) promoter construct or the rat apoAI (–1000/+14)lucI reporter. Fold induction refers to the activity without any HNF4 derivative and the error bars indicate standard deviation of six determinations.

Figure 5

Transactivation potential of co-transfected wild-type and mutant HNF4α on the human HNF1 promoter. Co-transfection of wild-type HNF4α and either R127W (A) or R154X (B) was done with the amount of expression vector given. In each experiment the amount of reporter gene construct and total DNA transfected was constant. Fold induction refers to the activity with no HNF4α derivative and the error bars indicate standard deviation of six determinations.

Figure 6

Western blot and gel shift assay with nuclear extracts from transfected 293 cells. 293 human embryonic kidney cells were transfected with expression vectors encoding wild-type or mutant HNF4α and nuclear salt extracts were prepared after 24 h for western blots (A) or gel retardation assays (B). The pellet fractions of the salt extracts were solubilised by DNase I digestion and used for western blots. Only the pellet fractions of E276Q and Q268X are shown. To visualise the transfected HNF4α variants the myc-specific antibody 9E10 was used in the western blots, whereas the monoclonal antibody H4/55, raised against amino acids 1–114 of rat HNF4α (18) was used in the gel retardation assays. The band of <36 kDa detected in the western blot of the pellet fractions is non-specific as it is also seen in untransfected cells.