Up-regulation of WNT-4 signaling and dosage-sensitive sex reversal in humans

Abstract

Wnt-4, a member of the Wnt family of locally acting secreted growth factors, is the first signaling molecule shown to influence the sex-determination cascade. In mice, a targeted deletion of Wnt-4 causes the masculinization of XX pups. Therefore, WNT-4, the human homologue of murine Wnt-4, is a strong candidate gene for sex-reversal phenotypes in humans. In this article, we show that, in testicular Sertoli and Leydig cells, Wnt-4 up-regulates Dax1, a gene known to antagonize the testis-determining factor, Sry. Furthermore, we elucidate a possible mechanism for human XY sex reversal associated with a 1p31-p35 duplication including WNT-4. Overexpression of WNT-4 leads to up-regulation of DAX1, which results in an XY female phenotype. Thus, WNT-4, a novel sex-determining gene, and DAX1 play a concerted role in both the control of female development and the prevention of testes formation. These observations suggest that mammalian sex determination is sensitive to dosage, at multiple steps in its pathway.

Figure 1

Genomic structure and localization of WNT-4. a, Genomic structure of WNT-4. The lengths of the exons appear above corresponding boxes, and the lengths of the introns appear below. b, Alignment of human and mouse Wnt-4 predicted protein sequences. The differences are denoted by white letters. c, FISH with a BAC probe containing WNT-4, on a normal metaphase spread. For easier comparison with the chromosome 1 ideogram, only chromosome 1 is shown. d, Ideogram of chromosome 1, illustrating normal GPG-banding. The guideline indicates the locus that corresponds to the WNT-4 FISH signal on chromosome 1.

Figure 2

FISH mapping of the patient’s duplication. Larger, filled white arrows indicate the chromosome 1 carrying the duplication; smaller arrows indicate the normal chromosome 1. a, Metaphase spread hybridized with whole chromosome 1–specific paint (Coatasome 1 Total Chromosome Probe; Oncor). b, Metaphase spread and interphase nucleus hybridized with probes specific to cytogenetic band 1p31 (BAC 395G6). c, Metaphase spread hybridized with nonoverlapping 3–4-kb probes specific to genomic sequences of WNT-4. d, Metaphase spread hybridized with probes specific to the 1p terminus (BAC 397E07).

Figure 3

Comparison of expression of WNT-4 in RNA from patient fibroblasts and from control fibroblasts. Lanes denoted “+RT” contain RT; lanes denoted “−RT” lack RT. Arrows indicate the 587-bp β-actin RT-PCR product and the 360-bp WNT-4 RT-PCR product.

Figure 4

Regulation of Dax1 by Wnt-4 and SF1. In a, the upper panel shows expression of Dax1 in Sertoli (TM4) cells, by RT-PCR after transfection with Wnt-4 (lanes 1–3), Wnt-4 and SF1 (lanes 4–6), SF1 (lanes 7–9), and Renilla luciferase control vector (lane 10), and the lower panel shows expression of β-actin from the same samples. In b, the upper panel shows expression of Dax1 in Leydig (MLTC-1) cells, by RT-PCR after transfection with Wnt-4 (lanes 1 and 2), Wnt-4 and SF1 (lanes 3 and 4), and Renilla luciferase control vector (lane 5), and the lower panel shows expression of β-actin from the same samples. The templates for positive and negative controls are mouse ovarian RNA and water, respectively. In c, the upper panel shows expression of Dax1 in Sertoli (TM4) cells, by northern blot after transfection with Wnt-4 (lanes 1–3), Wnt-4 and SF1 (lanes 4–6), SF1 (lanes 7–9), and pcDNA3.1 (lanes 10–12), as well as RNA from a mock-transfected control (lane 13), and the lower panel shows expression of β-actin on the same blot.