Warm temperature promotes gonadal transformation in ricefield eel.

(A) The distribution of 4 geographic populations of ricefield eels in China, showing the average annual temperature of each. (B) Bar graph showing the percentage of intersex animals in 2-year-old wild-caught ricefield eels from Hainan (Hainan province) and Wuhan area (Hubei Province). n=200 for each group. (C) Diagram showing the design of long term temperature experiments. Two temperatures were used: 25 °C (cool temperature), 33 °C (warm temperature). n=400 for each group. (D) Representative H&E staining images showing the gonad types of animals that were reared at cool and warm temperatures at the indicated time points. Bar: 200 µm. (E) Bar graph showing the percentage of intersex animals after 180 days of cool and warm temperature treatment. The experiments were repeated at least two times.

trpv4 is highly responding to environmental temperatures.

(A) Relative expression levels of trpv4 in 10 different tissues in adult ricefield eels. B: brain, H: heart, E: eye, K: kidney, S: spleen, L: liver, M:muscle, O: ovary, OT: ovotestis, T: testis. n=3. (B) ISH images showing trpv4 expression in ovaries, ovotestes, and testes. Black arrows pointing to trpv4 expressing cells. Bar: 200 µm. (C) The expression of the indicated trp and sex-biased genes of in vitro cultured ovaries at cool and warm temperatures. n=5 per group. (D) qPCR results showing the expression patterns of trpv4 and male sex genes in repeated temperature shifting experiments of in vitro cultured ovaries. n=5 per group. (E) Confocal images showing the calcium signaling in cultured ovarian cells at the indicated conditions. After shifting to 34 °C for 2 h, Cal-520 was added and calcium signal imaging was performed. The Trpv4 agonist GSK1016970A and antagonist RN1734 were administrated in the culture medium, and 2 h later, calcium signal imaging was performed. Bar: 100 µm. (F) qPCR results showing the dynamic expression of the indicated genes in gonads of female ricefield eel at 25 °C, and at the indicated time points after shifting to 34 °C. day 1: day 1 after shifting to 34 °C. n=5 per group. (G) ISH images showing the dynamic expression of trpv4 at the indicated time points before and after shifting to 34 °C. At 25 °C, trpv4 was moderately expressed in follicles of various stages of developing oocytes, and interstitial cell types. After shifting to 34 °C, trpv4 signals became much stronger compared to 25 °C. Bar: 200 µm. (H) Representative IF images showing the co-localization of Trpv4– and Foxl2-expressing cells. Bar: 50 µm. n=10. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. All experiments were repeated at least three times.

Warm temperature-induced male gene expression depends on Trpv4.

(A) Cartoon showing the design of animal experiments. Female eels kept at cool (25 °C) and warm (34 °C) temperatures were injected with the Trpv4 agonist GSK1016790A and antagonist RN1734 into the ovaries, respectively. After 2-3 days of injection, the ovaries were isolated and processed for the subsequent experiments. n=40. (B) qPCR results showing the relative expression levels of the sex-biased genes at the indicated conditions, based on the animal experiments. n=5 per group. (C) Representative WB images showing the expression of the indicated markers at the indicated conditions. GSK: GSK1016790A. (D) Relative quantification of the indicated proteins in Panel C. WB were repeated 3 times. (E) Representative IF images showing the expression of the indicated markers at the indicated conditions. Vimentin was used for show all cell types in ovaries. GSK: GSK1016790A. Bar: 200 µm. n=10, and 9/10 showed induced expression of Dmrt1/Sox9a. (F) quantification of panel E. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. All experiments were repeated at least three times.

siRNA-mediated trpv4 knockdown abolishes the abnormal up-regulation of male genes by warm temperature treatment.

(A) qPCR results showing the relative expression of the indicated genes at the indicated conditions in animal experiments. (B) Representative WB images showing the expression of sex biased proteins at the indicated conditions in animal experiments. (C) Quantification of panel B. (D) qPCR results showing the relative expression of the indicated genes at the indicated conditions. 10 µM RN1734 and DMSO was injected into gonads of male fish reared at 25/26 °C. Female fish injected with DMSO were used as control. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. All experiments were repeated at least three times.

The JAK/Stat3 signaling is downstream of Trpv4.

(A) Heat map showing the expression of the indicated genes of different groups. (B) IF images showing the expression of pStat3 at the indicated conditions in animal experiments. The white arrows indicated the location of pStat3 expressing cells. The experiments were repeated at least two times. Bar: 200 µm. n=12, and 10/12 showed increased expression of pStat3. (C) Quantification of panel B. (D) Bar graph showing the relative calcium signals at the indicated conditions. Ovarian explants were cultured at the indicated conditions, and calcium signals were determined by calcium indicator dye Cal-520 acetoxymethyl ester. (E) Representative WB images showing the expression of the indicated makers in ovaries, early ovotestes, and middle ovotestes. n=5 per group. (F) Quantification of panel E for relative expression of Amh and pStat3. (G) pStat3 levels after the addition of A23187 and BAPTA-AM in cultured ovarian cells at 25 °C and 34 °C conditions. (H) Quantification of panel G. All experiments were repeated at least two times.

Animal experiments showing that warm temperature induced male gene expression depends on pStat3.

(A) Cartoon showing the design of the experiments. Female eels kept at cool (25 °C) and warm (34 °C) temperatures were injected with the pStat3 agonist Colivelin and antagonist HO3867 into the ovaries, respectively. After 2-3 days of injection, the ovaries were isolated and processed for the subsequent experiments. n=50. (B) qPCR results showing the relative expression of the indicated genes at the indicated conditions. n=5 per group. (C) IF images showing the expression of male biased genes at the indicated conditions. Bar: 200 µm. n=10, and 8/10 showed increased expression of pStat3/Dmrt1. (D) Quantification of panel C. n=5 per group. (E) qPCR results showing the relative expression of the indicated genes at the indicated conditions. n=5 per group. (F) qPCR results showing the relative expression of the indicated genes at the indicated conditions. n=5 per group. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. The qPCR experiments were repeated at least three times.

pStat3 binds and activates the kdm6b gene.

(A) Cartoon showing the conserved pStat3 binding motifs upstream the TSS of the kdm6b gene. (B) ChIP experiments showing the enrichment of pStat3 at the kdm6b locus in ovarian tissues of fish reared at cool temperature (CT) and warm temperature (WT) conditions, in the absence and presence of HO-3867. (C) Luciferase assay for kdm6b-luc and kdm6bM-luc activities in 293T cells, in the absence and presence of pStat3 agonist Colivelin. (D) The representative ISH images showing the expression of kdm6b in ovary and testis. Bar: 200 µm. n=3 per group. (E) Cartoon showing how Trpv4 may link environmental temperature to the sex determination cascades via the downstream signaling pathways in ricefield eel. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. ChIP and Luciferase experiments were repeated two times.

Related to Figure 1.

(A) The graph showing the annual temperature dynamic by month for 30 years in Wuhan area, Hubei province, China. The highest and lowest temperatures per month were shown. (B) Schematic of the whole process of sex change, summarizing morphology, length and gonadal histology across time (10 years). The ricefield eel are born as females. After 2 year growth, they reach sexual maturity. After spawning (ranging from May to August each year), females will enter 1-2 years of intersex stage before becoming functional males. The average lifespan of wild ricefield eel is around 10 years. (C) H&E staining images showing the typical cell types in an ovary of a 2-year-old female, before spawning. Bar: 200 µm. (D) The average body length of fish that were raised for 90 and 180 days at the indicated temperatures. (E) The average body weight of fish that were raised for 90 and 180 days at the indicated temperatures. ns: not significant. The experiments were repeated two times.

Related to Figure 2.

(A) The qPCR results showing the expression patterns of the indicated female sex genes in repeated temperature shifting experiments of in vitro cultured ovaries. n=5 per group. (B) The qPCR results showing the expression of trpv4 at the indicated time points of in vitro cultured ovarian explants. n=5 per group. (C) The qPCR results showing the expression of the indicated temperature responding– and sex-biased genes in ovaries of females that reared at 25 °C and after 1 day exposure to 34 °C. n=5 per group. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. All experiments were repeated at least three times.

Related to Figure 3.

(A) qPCR results showing the expression of the indicated male– and female-biased genes in ovarian explants cultured at 25 °C and 34 °C with increasing doses of small molecule RN1734. n=5 per group. (B) qPCR results showing the expression of the sex-biased genes at the indicated conditions, based on in vitro cultured ovarian explants. n=5 per group. (C) WB images showing the expression levels of the indicated proteins at the indicated conditions, based on in vitro cultured ovarian explants. (D) Quantification of panel C. (E) IF images showing the expression levels of Dmrt1 at the indicated conditions, based on in vitro cultured ovarian explants. n=5 per group. (F) Quantification of panel C. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. All experiments were repeated at least three times.

Related to Figure 5.

(A-B) The representative IF images showing the expression of pERK and Amh from ovaries, ovotestes, and testes. n=6 per group. The experiments were repeated at least two times.

Related to Figure 6.

(A) qPCR results showing the expression of the indicated genes at the indicated conditions, based on in vitro cultured ovarian explants. (B) Representative IF images showing the expression of pStat3 at the indicated conditions, based on in vitro cultured ovarian explants. n=6 per group. (C) qPCR results showing the expression of the indicated genes at the indicated conditions, based on in vitro cultured ovarian explants. *: P< 0.05, **: P< 0.01, ***: P< 0.001, and ****: P< 0.0001. ns: not significant. All experiments were repeated at least three times.