Wilms' tumor protein Wt1 is an activator of the anti-Müllerian hormone receptor gene Amhr2

Abstract

The Wilms' tumor protein Wt1 plays an essential role in mammalian urogenital development. WT1 mutations in humans lead to a variety of disorders, including Wilms' tumor, a pediatric kidney cancer, as well as Frasier and Denys-Drash syndromes. Phenotypic anomalies in Denys-Drash syndrome include pseudohermaphroditism and sex reversal in extreme cases. We have used cDNA microarray analyses on Wt1 knockout mice to identify Wt1-dependent genes involved in sexual development. The gene most dramatically affected by Wt1 inactivation was Amhr2, encoding the anti-Müllerian hormone (Amh) receptor 2. Amhr2 is an essential factor for the regression of the Müllerian duct in males, and mutations in AMHR2 lead to the persistent Müllerian duct syndrome, a rare form of male pseudohermaphroditism. Here we show that Wt1 and Amhr2 are coexpressed during urogenital development and that the Wt1 protein binds to the promoter region of the Amhr2 gene. Inactivation and overexpression of Wt1 in cell lines was followed by immediate changes of Amhr2 expression. The identification of Amhr2 as a Wt1 target provides new insights into the role of Wt1 in sexual differentiation and indicates, in addition to its function in early gonad development and sex determination, a novel function for Wt1, namely, in Müllerian duct regression.

FIG. 1.

Amhr2 expression is lost in gonads of Wt1^−/− embryos. Expression of Amhr2 (A, B, E, and F) and Lhx9 (C and D) in urogenital ridges of Wt1^+/+ (A, C, and E) and Wt1^−/− (B, D, and F) embryos with 12 (A to D), 15 (E), or 16 (F) tail somites (E11.0 to E11.5) was analyzed by RNA in situ hybridization on paraffin sections (A to D) and whole urogenital ridges (E and F). (G) Quantitative real-time RT-PCR analysis of Amhr2 and Lhx9 expression in Wt1^−/− urogenital ridges of the 12-tail-somite stage compared to wild-type tissue. Expression levels were normalized to Actb. For comparison, expression in the wild-type urogenital ridge was set as 1. In order to make the value for Amhr2 expression in Wt1^−/− tissue visible, the scale of the y axis is logarithmic. Numbers above white bars indicate the ratio between mutant and wild-type expression levels. Data represent means and standard deviations of one representative experiment, measured in triplicate. da, dorsal aorta; dm, dorsal mesentery; g, gonad; m, mesonephros; v, subcardial vein.

FIG. 2.

Wt1 and Amhr2 show overlapping expression patterns in the region of the developing Müllerian duct as well as in embryonic and adult gonads. Expression of Wt1 (A, D, F, H, and J), Amhr2 (B, E, G, I, and K), and Sf1 (C) in the urogenital region of E13.5 (A to C), E14.5 (D, E, H, and I), and E15.5 (F, G, J, and K) wild-type male (A to C and H to K) and female (D to G) embryos was investigated by RNA in situ hybridization on paraffin sections. Insets in A, B, and C show areas marked by dashed boxes in higher magnification. Expression of Wt1 (L and N) and Amhr2 (M and O) was also analyzed in adult ovary (L and M) and testis (N and O). tc, testis cord; g, gonad; m, mesonephros; arrowhead, Wolffian duct; arrow, Müllerian duct; *, primary follicle; **, secondary follicle; ***, tertiary follicle.

FIG. 3.

Amhr2 expression is altered by Wt1 knockdown and Wt1 induction. (A) M15 cells were treated with two different siRNAs directed against Wt1, an siRNA directed against luciferase, or transfection reagent only. Expression of Amhr2, Wt1, and Gapdh was measured 48 h after transfection by real-time RT-PCR and normalized to Actb expression. Expression levels in luciferase siRNA-treated cells were set as 1. Data represent the means and standard deviations of two independent experiments. Asterisks indicate statistically significant differences (P < 0.05; Student's t test) from expression in both controls (luciferase siRNA and cells not treated with siRNA). (B) Western blot analysis of Wt1 protein levels in siRNA-treated cells 48 h after transfection. Lactate dehydrogenase (Ldh) served as a loading control. (C) Western blot analysis of cells harboring an inducible allele of Wt1(−KTS) or Wt1(+KTS) 24 h after induction. (D) AMHR2 expression in uninduced and induced cells was measured by real-time RT-PCR 10 h, 20 h, and 48 h after induction of Wt1(−KTS) or Wt1(+KTS) in UB27 and UD28 cells. Expression levels were normalized to ACTB, and expression levels in uninduced cells were set as 1. Data represent the means and standard deviations of two independent experiments. Asterisks indicate statistically significant differences (*, P < 0.05; **, P < 0.01; Student's t test) from expression in uninduced cells.

FIG. 4.

Amhr2 promoter contains three conserved Wt1(−KTS) consensus binding sites. (A) Graphical representation of the more G-rich strand of the Wt1(−KTS) consensus binding site designed with http://weblogo.berkeley.edu. (B) Alignment of the consensus sequence with the three Wt1(−KTS) consensus binding sites found in the murine Amhr2 promoter. In all cases the more C-rich strand is shown in the 5′ to 3′ direction. Conserved positions are boxed in gray. Positions mutated for EMSA and reporter gene assays (C to A) are underlined. (C) The mouse Amhr2 promoter sequence (accession number AF092445) was aligned with homologous human and rat sequences acquired from http://genome.ucsc.edu/index.html (human assembly, May 2004; rat assembly, November 2004) using the T-coffee program (http://igs-server.cnrs-mrs.fr/Tcoffee/tcoffee_cgi/index.cgi). Conserved transcription factor binding sites for Sf1, Gata, Sp1, and two unknown factors as well as the transcriptional start site (TS) (53) are indicated by dashed boxes. All sequences contain three conserved Wt1(−KTS) consensus binding sites (gray shaded boxes) overlapping with the two described binding sites for unknown factors (53).

FIG. 5.

Wt1(−KTS) and Sf1 bind to the murine Amhr2 promoter in vitro. (A) Schematic representation of the Amhr2 promoter, containing potential transcription factor binding sites, and the fragments used for EMSA. (B) For EMSA analysis, labeled fragments F1 to F6 were incubated without protein and with recombinant GST, GST-Wt1(−KTS), and GST-Sf1. (C) Wild-type and mutated F4 were additionally analyzed by adding GST-Wt1(+KTS) and a mutant form of GST-Wt1(−KTS). F4 was mutated by replacing three of the five conserved cytosines with adenines in all three Wt1 binding sites. (D) Binding of Wt1 to the consensus binding sites was also shown by DNase footprinting analysis using the same recombinant proteins and a 5′-labeled PCR fragment. Sequencing reactions for A and G bases of the same fragment are shown. The gray box indicates the region protected by Wt1(−KTS) spanning all three consensus binding sites. Numbers in panels A and D refer to the murine Amhr2 promoter sequence (53).

FIG. 6.

Wt1(−KTS) transactivates the Amhr2 promoter. (A) M15 and TM4 cells were transfected with the 1.6-kb Amhr2 promoter construct together with expression constructs for Wt1(−KTS), Wt1(+KTS), and Sf1. Subsequently, luciferase reporter assays were performed. (B) Constructs harboring a shorter fragment and mutated forms of this fragment with mutations in binding site 3 (mut3), 1 and 2 (mut1/2), or all three sites together (mut1/2/3) were cotransfected with Wt1(−KTS). The respective Amhr2 promoter fragments are indicated schematically below the diagrams. Results are given as relative activation of the reporter by the expression constructs compared to the empty vector. Data represent the means and standard deviations of three (A) and four (B) independent experiments. Asterisks indicate statistically significant differences of activation compared to cells transfected with an empty expression construct (*, P < 0.05; **, P < 0.01; ***, P < 0.001; analysis of variance).

FIG. 7.

Wt1(−KTS) binds the Amhr2 promoter in living cells. PCR analysis of DNA purified after chromatin immunoprecipitation with (A) uninduced and induced UB27 cells and (B) M15 cells. Two different antibodies against Wt1 were used, Wt1 C-19 (A) and WTc8 (B). Immunoprecipitation using an antibody against c-Myc or without antibody served as negative controls, while antibody against acetylated histone H3 and input DNA served as positive controls. Primer pairs binding to the transcribed region (transc. reg.) of human and murine Amrh2, 2 and 10 kb downstream of the transcriptional start site, respectively, were used as controls for PCR.