Fig 1.
Repression of initial follicular growth in ovarian grafts under androgen-excess host conditions.
(A, B) Anti-GATA4 and AMH immunostaining of wild-type ovarian tissues grafted into intact male (XY-host), intact female (XX-host) and castrated male (XY-cast-host) hosts treated with or without testosterone (T), or dihydrotestosterone (DHT). The lower magnified images of GATA4-positive gonadal areas are shown in upper plates in A and B. AMH-positive healthy primary, secondary, and antral follicles were detected in ovaries grafted into female hosts (lower plate in B). (C) Bar graphs indicate the relative numbers of normal healthy follicles (left) and degenerating follicles (right). The data are expressed as means ± SEM (*p<0.05 as compared with non-treated host value in each host group; #p<0.05 as compared between two groups, Steel's test). Each number in parentheses indicates the number of explants used in each host. Scale bars, 100 μm for (A, B).
Fig 2.
Ectopic appearance of SOX9-posititve Sertoli-like cells in ovarian grafts under androgen-excess host conditions.
(A–D) Anti-GATA4, SOX9, FOXL2, DMRT1, or GDNF immunostaining of wild-type ovarian tissues grafted into XY-, XX-, and XY-cast-hosts treated with or without T or DHT on days 20 post-transplantation. The lower magnified images of GATA4-positive gonadal areas are shown in upper plates in a. Normal follicular structures were mostly absent, and tubular structures containing ectopic SOX9-positive cells were evident near the edge of ovaries grafted into XY-host and T/DHT-treated XX- and XY-cast-hosts on day 20 post-transplantation (arrows in A). The ectopic SOX9-positive cells are FOXL2-negative (B), and they are found in the DMRT1-positive/GDNF-positive tubular region (C, D). In c, the inset shows high-magnification images of tubular structures indicated by arrows. (E) Bar graphs indicate the SOX9-positive cells per gonadal area (mm2). The data are expressed as means ± SEM (*p<0.05 as compared with non-treated host value in each host group; #p<0.05 as compared between two groups, Steel's test). Each number in parentheses indicates the number of explants used in each host. Broken lines indicate the border of tubular structures in c and e. Scale bars, 100 μm for (A, C); 20 μm for (B, D).
Fig 3.
Sox8 expression in AMH-positive and -negative degenerating follicles and masculinized phenotypes of the Sox8-null and Amh-null ovaries transplanted into male host mice.
(A, B) In situ hybridization using a Sox8 antisense probe and anti-AMH immunostaining of two serial ovarian tissue sections grafted into XY-host on days 10 (A; The most left panels are lower and the others are higher magnified images) and 20 (B) post-transplantation. Sox8-positive signals (yellow arrowheads) were detected in the granulosa cells of AMH-negative and -positive degenerating follicles on day 10 post-transplantation, as well as in tubular structures, including SOX9-positive Sertoli-like cells on day 20 post-transplantation. (C–F) Anti-GATA4, AMH, SOX9, or GDNF immunostaining of wild-type, Sox8-null (Sox8-/-), or Amh-null (Amh-/-) ovarian tissues grafted into XY-host on day 20 post-transplantation. The gonadal area is identified by GATA4 staining (C). Degenerating follicles were frequently seen in either Sox8-null or Amh-null grafted ovaries even at this stage (C, D). In contrast, the tubular structure formation (arrows or broken outlines) and ectopic appearance of SOX9-positive Sertoli-like cells in GDNF-positive tubular structures are found in all three types of grafted ovaries (E, F). (G) Bar graphs indicate the relative numbers of normal healthy follicles (left), degenerating follicles (center), and SOX9-positive cells (right) per gonadal area (mm2) in wild-type, Sox8-null, or Amh-null grafts in male host mice. The Sox8-null or Amh-null ovarian explants exhibited a large number of degenerating follicles even on day 20 post-transplantation, but no changes in the number of SOX9-positive cells were detected among the three genotypes. The data are expressed as means ± SEM (*p<0.05 as compared with wild-type host value, Dunnett's test). The numbers in parentheses indicate the number of explants used in each host. Broken lines indicate the border of tubular structures. Scale bars, 20 μm for (A, B, F); 100 μm for (C, D).
Fig 4.
Temporal changes in the expression levels of pre-granulosa or Sertoli cell-specific genes in ovarian grafts in the male-host environment.
(A) Hierarchical clustering of microarray data obtained from the grafted ovaries on days 0, 4, 7, 10, 15, and 20 post-transplantation, based on all 45,101 probes in the Mouse Genome 430 2.0 GeneChip array. Duplicate raw data were generated from transplants on days 0, 10, and 20 post-transplantation. (B) GO terms (p<0.01) for 241 down/576 upregulated genes (|fold| > 2, false discovery rate [q] < 0.3) in grafted ovaries on day 20 post-transplantation. (C) Venn diagrams showing 643 pre-granulosa (orange) and 695 Sertoli cell (blue)-specific gene probes selected from among the altered gene probes at each stage by comparing data from Sertoli cells versus pre-granulosa cells at 12.5 dpc using a published dataset [47] (right circles). (D) Scatterplot showing the number of down/upregulated pre-granulosa/Sertoli cell-specific genes in the grafted ovaries. Each point on the graph shows the grafted ovaries on each day post-transplantation. The vertical or horizontal lines show the total numbers of downregulated pre-granulosa or upregulated Sertoli cell-specific genes (|fold| > 2, p<0.05) at 12.5 dpc from the published dataset [42]. (E) Heatmap showing relative changes in expression levels of the 525 pre-granulosa (containing cluster I) and 402 Sertoli cell-specific genes (containing clusters II/III) in grafted ovaries (signal intensity: high [red], medium [black] to low [green]). (F) GO terms (p<0.05) for cluster I and clusters II/III genes in the ovarian grafts. (G, H) Line graphs show the fold expression changes for several key pre-granulosa (G) and Sertoli cell-specific genes (H) post-transplantation. The expression levels on day 0 post-transplantation were set as 1.0 on the y-axis.
Table 1.
Top 10 genes of the pre-granulosa cell-specific downregulated genes (cluster-I) and Sertoli cell-specific upregulated genes (cluster-II and III)1).
Fig 5.
Temporal changes in the expression levels of each pre-granulosa/Sertoli cell-specific gene encoding a transcription/nuclear factor in grafted ovaries.
(A–B) Bar graphs showing temporal changes in the expression levels (y-axis) of pre-granulosa (A) and Sertoli cell-specific genes (B) in ovarian grafts in male and female host mice on each day post-transplantation (x-axis). The data are expressed as means ± SEM. The expression levels on day 0 post-transplantation (i.e., the ovaries isolated from the 13.0-dpc embryos pretreated with busulfan) were set as 1.0 on the y-axis (*p<0.05 as compared with day 0, Dunnett’s test or Steel’s test). The numbers in parentheses indicate the number of explants used at each stage. (C) Expression profile categories for pre-granulosa or Sertoli cell-specific genes in grafted ovaries during partial masculinization.
Table 2.
List of the potentially 16 down- or 22 up-regulated genes of pre-granulosa or Sertoli cell-specific transcription/nuclear factor, respectively1).
Fig 6.
Expression profiles of AR and its target genes in grafted ovaries.
Bar graphs showing temporal changes in the expression levels (y-axis) of Ar and AR-target genes, Ube2b, B4galnt1 and Tubb3, in ovarian grafts in XY- and XX-host on each day post-transplantation (x-axis). The data are expressed as means ± SEM. The expression levels on day 0 post-transplantation were set as 1.0 on the y-axis (*p<0.05 as compared with day 0; #p<0.05 as compared between different host groups, Steel’s test). The numbers in parentheses indicate the number of explants used at each stage.
Fig 7.
Testosterone-dependent masculinization of mouse fetal ovaries transplanted into the XY host.
Schematic representation showing i) normal sex differentiation of SDSI-positive bipotential supporting cells in the maternal environment (pink background at left side) and ii) the partial masculinization of pre-granulosa cells, together with temporal upregulation of Ar, Amh, Sox8 and Sox9 in ovarian grafts in XY host (light blue background at right side). The global shared transcriptomes with Sertoli cell- and granulosa cell-specific genes between normal sex differentiation and masculinization processes was also shown (i.e., downregulation of 41.2% [643/1,561 gene probes] of the pre-granulosa cell-specific genes, versus upregulation of 37.6% [695/1,849 gene probes] of the Sertoli cell-specific genes [47]). Early downregulation of pro-ovarian genes and subsequent upregulation of pro-testis genes during days 4 to 20 post-transplantation (see Fig 5C) are indicated by a color change from red to purple in the bar. SDSI-positive bipotential state is also indicated as gray bars.