Dmy initiates masculinity by altering Gsdf/Sox9a2/Rspo1 expression in medaka (Oryzias latipes)

Abstract

Despite identification of several sex-determining genes in non-mammalian vertebrates, their detailed molecular cascades of sex determination/differentiation are not known. Here, we used a novel RNAi to characterise the molecular mechanism of Dmy (the sex-determining gene of medaka)-mediated masculinity in XY fish. Dmy knockdown (Dmy-KD) suppressed male pathway (Gsdf, Sox9a2, etc.) and favoured female cascade (Rspo1, etc.) in embryonic XY gonads, resulting in a fertile male-to-female sex-reversal. Gsdf, Sox9a2, and Rspo1 directly interacted with Dmy, and co-injection of Gsdf and Sox9a2 re-established masculinity in XY-Dmy-KD transgenics, insinuating that Dmy initiates masculinity by stimulating and suppressing Gsdf/Sox9a2 and Rspo1 expression, respectively. Gonadal expression of Wt1a starts prior to Dmy and didn't change upon Dmy-KD. Furthermore, Wt1a stimulated the promoter activity of Dmy, suggesting Wt1a as a regulator of Dmy. These findings provide new insights into the role of vertebrate sex-determining genes associated with the molecular interplay between the male and female pathways.

Figure 1. In vitro and in vivo validation of knockdown strategy.

(A) In vitro validation. The specificity and off target effects of pmDMY-AS construct was validated in vitro using COS7 cells via co-transfection of pmDMY-AS construct and the ORF plasmids of either Dmy (specific, solid circles), Dmrt1 (partially specific, solid squared), Sox9a2 (non-specific, grey circles) Gsdf (non specific, open circles) or Sf1 (non specific, open squared). The mean absolute copy numbers of respective gene per 5 ng of RNA are plotted in the graph along with SEM. The significance is indicated by a, b, c, in which different letters indicate the significant difference from other group at p < 0.05. (B,C) In vivo validation. Olvas-eGFP medaka were electroporated with pEGFP-AS plasmid (carries an antisense eGFP sequence) and continuous microscopic visualization (from 3days after fertilization (daf) to 5 days after hatching (dah)) of GFP expression in individual embryos was performed to ascertain the changes of GFP production in control (B, 6 daf) and pEGFP-KD (pEGFP-AS electroporated) groups (C, 6 daf). (D,E) Sharp fall in Gfp mRNA production was assessed through real-time PCR at 20 dah (days after hatching). Non-specific effects of pEGFP-AS construct were analysed by measuring the mRNA amount of Dmy, Dmrt1, Spo11, and ERα. Data are presented as both individual values (with dots and triangles) and mean of 10 individual (white and grey columns for control and knockdown groups, respectively) fish of each sex.

Figure 2. Effect of Dmy knockdown on gonad development.

(A) The total number of mitotic germ cells counted (control XX, white; Dmy-KD XY, grey; control XY, black) at 5 dah (5H) and 10 dah (10H), was used as a marker for gonadal sex. Data are presented as mean of 10 individuals from both control and Dmy-KD groups. Error bars indicate SEM. Letters (a-c) above the bars indicate that these groups differ significantly (p < 0.05) from each other. (B–G) The candidate HE-stained sections of control XX females (B,E), control XY males (C,F), and Dmy-KD XY fish (D,G). Female-like gonad development is seen in Dmy-KD XY fish, which later forms a fully developed ovary. Note: B, C, and D represent gonads at 10 dah, while E, F, and G represent gonadal structure at 50 dah. (H,I) Dmy knockdown directed femininity was persistent during adulthood, which was confirmed by female-like external appearances (H) and gonad structure (I). The femininity was further confirmed by ISH with female (Cyp19a1, J; Foxl2, L; Rspo1, N), and male (Dmy, K; Cyp11a, M); Cyp11b, O) associated genes. (Scale bars, 20 μm.)

Figure 3. Dmy-knockdown induced transcriptional alterations of male and female associated genes.

(A,B) Ontogenic changes in the expression of male (Dmy, Gsdf, Sox9a2, and Dmrt1) (A) and female (Rspo1, Foxl2, and Spo11) (B) associated genes from 5 daf (5F) to 10 dah (10H) were analysed by real-time PCR. Data were normalised with Ef1a expression. Each column in the graph represents mean ± SEM of 10 individuals of control XY (white columns) and Dmy-knockdown XY (grey columns) fish. Letters above the bars indicate that these groups differ significantly (p < 0.05) from each other.

Figure 4. Wt1, an upstream candidate for Dmy regulation.

(A) Dmy and Wt1a expression were analysed using several early stage samples (S4-S18) to assess the precise start point of zygotic expression (XX, white columns and XY, grey columns). (B) Differences in Wt1a expression between control (white columns) and Dmy-KD (grey columns) XY embryos during gonadal differentiation (S20-S39) were also analysed by real-time PCR. (C,D) Wt1a specific whole mount ISH of Dmy-KD XY (C) and control XY (D) embryos (stage 22) showed no virtual difference. (E) Luciferase assays using HEK293 cells showed a dose dependent activation of Dmy promoter by Wt1a KTS (−). Graphical data are presented as an average of three independent experiments, each containing triplicates of individual sampling group. Error bars represent SEM. Letters (a–c) above the bars indicate that these groups differ significantly from each other at p < 0.05.

Figure 5. Gsdf and Sox9a2 are potential downstream targets of Dmy.

(A) In vivo chromatin immunoprecipitation analysis of candidate Dmy targets. Relative expression to control input was calculated using real-time PCR and plotted on the graph. Data are presented as means ± SEM of 4 separate experiments and significances (p < 0.05) are denoted by different letters. (B) Effects of Dmy overexpression on Gsdf and Sox9a2 promoter activity (upper panel) and mRNA transcription (lower panel) were ascertained. (C) Differences in Dmy expression between control (white columns) and olvas-eGFP-Dmy-KD co-injected with Gsdf-cherry and Sox9a2-cyan (grey columns) XY embryos during gonadal differentiation (S20-S39) were assessed using real-time PCR. (D) Representative images showing the germ cell distribution in Olvas-eGFP-control XX, Olvas-eGFP-Dmy-KD-XY embryos at S39 after control (upper panel) and Gsdf/Sox9a2 (lower panel) mRNA injection. Isolated and cluster germ cells are marked with ‘*’ and ‘#’, respectively. Note: Clustered germ cells are representative of female type gonadal development (N = 9/ group). (E) The total number of germ cells was plotted (Y-axis) against different injected groups (X-axis). Data are presented as means of 10 individuals (error bars represent SEM). Different letters (a–d) indicate significant differences from other groups at p < 0.05. Note: C, G, S, and G + S represents PBS, Gsdf-cherry, Sox9a2-cyan, and Gsdf-cherry/Sox9a2-cyan mRNA injected embryos, respectively. (F) The gonadal sexuality of olvas-eGFP-Dmy-KD embryos co-overexpressed with Gsdf-Sox9a2 (N = 7) was assessed at maturity. (Scale bar, 20 μm).