Figure 1.
Complete or partial XY sex-reversed phenotype of XY Gadd45g−/− mice on a pure B6 or mixed 129/B6 background.
(A–F) External genitals, (G–L) internal reproductive organs (8x magnification), and (M–X) hematoxylin/eosin (HE)-stained gonad or epididymis sections from young adult Gadd45g−/− and wild type mice. External and internal (C,I) reproductive organs from B6 Gadd45g−/− XY female mice (XY-F) are morphologically indistinguishable from wild type females (B, H). HE-stained ovaries from all B6 XY-F mice appear normal and contain oocytes and follicles at different maturation stages (O), including antral follicles with oocytes (U). Corpora lutea are found in XX-F and XY-F ovaries (N,O,T,U). In contrast, a large spectrum of disorders of sexual development (DSD) is seen in Gadd45g−/− XY mice on the 129/B6 background, which can be divided into three groups. 1) Infertile XY-females (E) with female internal reproductive organs (ovary, oviduct, uteri) (K), although the ovaries usually contain only primordial follicles and interstitial cells (Q,W). 2) Infertile XY-males with male phenotype (D) and reproductive system (J; testis, epididymis, vas deferens and seminal vesicles [not shown in J]) with hypoplastic testis (compare J to G). Seminiferous tubules showed reduced spermatogenesis and interstitial cell hyperplasia (P), and no spermatozoa were present in the cauda epididymis (V) and vas deferens. 3) XY-intersex mice (F) with male and female characteristics. External genitals were male, female, or ambiguous (as in F). One side often developed a hypoplastic testis/epididymis/vas deferens (L) and the contralateral gonad, a uterus and a hypoplastic ovary/oviduct/uterus or an ovotestis with mixed ovarian and testicular tissue (L,R,X). Magnification: 5x (N,O,Q,R) or 10x (M,P,S,T,U,V,W,X).
Figure 2.
Frequency of sex reversal in distinct Gadd45g genotypes on 129/B6 or B6 backgrounds.
(A) SRY genotyping showed that 100% of XY Gadd45g−/− and 3% of Gadd45g+/− mice on the B6 background, and (B) 80% of XY Gadd45g−/− mice on the 129/B6 background were born as sex-reversed XY females. (C) Sex reversal in Gadd45g−/− mice born as females was detected by PCR amplification of a 404-bp band from the Y chromosome SRY gene. (D) Presence of the Y chromosome in SRY-positive Gadd45g−/− XY-F mice was confirmed by chromosome painting (FISH).
Table 1.
Lack of Gadd45g, but not Gadd45a or Gadd45b leads to male-to-female sex reversal in mice.
Figure 3.
Lack of Sertoli cell differentiation and testis cord formation in XY Gadd45g−/− gonads.
(A–L) Confocal optical slices of whole mount immunostained B6 gonads (dashed outline), showing expression of Sertoli cell markers SOX9 (nuclear, blue) and AMH (cytoplasmic, red) and the germ/endothelial cell marker Pecam1 (membrane, green). (A) From 12.5 dpc, wild type male gonads can be distinguished morphologically from female gonads (B) by the appearance of testis cord structures containing SOX9-expressing Sertoli cells. (C) In Gadd45g−/− XY gonads, only transient SOX9 expression was observed in a small number of somatic cells at 12.5 dpc. (D–F) and (J–L) are enlarged details of the images above. (G) At 13.5 dpc, all germ cells in male gonads are enclosed in testis cords and interact closely with a surrounding single layer of SOX9- and AMH-positive Sertoli cells. (H,I,K,L) No testis cord formation or AMH expression was induced in Gadd45g+/+ XX (H,K) or in Gadd45g−/− XY (I,L) gonads.
Figure 4.
Gadd45g is necessary for SRY expression and testis development at the time of primary sex determination of the bipotential embryonic gonad.
(A) RT-PCR quantification of relative Gadd45g expression in B6 wild type XX and XY embryonic gonads (including mesonephros), collected between 10.5 (8 tail somites) and 12.5 dpc (28 tail somites). In all images, the dashed red line denotes the point of maximal SRY expression (11.5 dpc or 18 tail somites). (B) Microarray quantification of relative Gadd45g expression in normal wild type XX and XY embryonic gonads (including mesonephros) from the time of the bipotential gonad (GEO data set GSE6916). (C) Microarray quantification of relative Gadd45g expression in purified somatic support precursor cells during the critical sex determination period (10.5–11.5 dpc) (GEO data set GDS1724). (D) RT-PCR quantification of SRY copy numbers in Gadd45g+/+, Gadd45g+/− and Gadd45g−/− gonads (including mesonephros) between 10.5–12.5 dpc. (E) RT-PCR quantification (mean and SEM) of SRY copy numbers in Gadd45g+/+, Gadd45g+/− and Gadd45g−/− gonads (including mesonephros) at the 17 to 21 tail somite stage. Statistical analysis (Student’s t-test) of all data from Fig. 4D at the 17–18 or 19–20 tail somite stage showed reduced SRY expression in Gadd45g−/− mice and delayed SRY upregulation in Gadd45g+/− mice compared to wild type controls. Numbers for each group (n) are displayed in each column. (F) Main components of the transcriptional network around the time of primary sex determination and suggested mechanism of Gadd45g action. In the absence of SRY (XX gonads), female-specific genetic programs (such as the WNT4/beta-catenin pathway) direct differentiation of the somatic support lineage into granulosa cells and ovary development. In XY gonads, SRY expression at 11.5 dpc induces SOX9 expression, Sertoli cell differentiation and testis cord formation. In males, Sertoli cell-derived anti-Müllerian hormone (AMH) induces Müllerian duct regression, and testosterone induces differentiation of the Wolffian duct into vas deferens, seminal vesicles and epididymis. AMH acts ipsilaterally, which explains the intersex phenotype shown in Fig. 1L. In Gadd45g-deficient mice, SRY expression in XY gonads fails to reach the threshold level necessary for testis differentiation. The dashed red line denotes the point of maximal SRY expression (11.5 dpc).