Fig 2: Histological analysis of testes from adult and juvenile Gnathopogon caerulescens.Testes were sampled from spawning (May) and non-spawning (September) adults and from juvenile fish. Images show the gross morphology of the fish and testes; adjacent testicular sections are stained with haematoxylin-eosin (HE) and for Vasa (green) and proliferating cell nuclear antigen (PCNA, red). Nuclei were stained with DAPI (blue). The images shown are representative of ten different male G. caerulescens per group. The insets in each panel show high-magnification images of spermatogonia. Spawning testes (May) were thick and plump with ivory white coloration, whereas non-spawning adult (September) and juvenile (5-month-old) testes were thin and threadlike with translucent white coloration. Experiment reproduced ten times. Scales are indicated in the figure.

Fig 3: GSCs contribute to the germline cell lineage of host embryosto generate zebrafish germline chimeras. (A) The dnd MO abolished the expression of dnd-GFP reporter, whereas the control MO did not. (B) Images showing GFP signals in embryos injected with control and dnd MOs. The GFP-positive PGCs (red arrows) were distributed in the gonadal region of the Tg(piwil1:eGFP) embryos or embryos injected with control MO. No GFP-positive PGCs were observed in the gonadal region of embryos injected with dnd MO. The frequency of embryos with the indicated phenotypes was shown in the bracket of each group. Lateral view. Scale bars, 500 µm. (C) Gel images of RT-PCR products showing the expression of the eel-specific dazl gene in gonads of transplanted zebrafish. ß-actin was used as a reference gene. Red arrows indicate PCR bands amplified by second-round RT-PCR. (D) Alignment of the amplified dazl sequences from gonads of M. albus and gonads of three fGSCs-transplanted zebrafish. (E) IF staining of Vasa in ovaries of control-, dnd MO-, and dnd MO-transplanted with fGSCs zebrafish. White arrows indicate the germ cells. Scale bar = 50 µm.

Fig 4: In vitro spermatogenesis of G. caerulescens in adherent culture.Testicular cells isolated from spawning (May) and non-spawning (September) adult and juvenile testes (5-month-old) were cultured. (a) Bright-field images of testicular cell cultures. The insets in each panel represent high-magnification images of the most advanced germ cell types. In the testicular cell cultures from spawning adults, spermatogenic cells of all developmental stages were observed at days 7, 11, and 21. In the testicular cell cultures from non-spawning adults, germ cell colonies of spermatogonia, spermatocytes, and flagellated sperm were observed at days 7, 11, and 18. In testicular cell cultures from juveniles, colonies of spermatogonia, spermatocytes, and flagellated sperm were observed at days 4, 13, and 19. Sperm flagella are indicated by arrowheads. Experiment reproduced six times. Bars represent 100 µm. (b) Bright-field images (top) of juvenile testicular cell cultures and fluorescent images for 5-ethynyl-2'-deoxyuridine (EdU, green) and Vasa (red). Nuclei were stained with DAPI (blue). EdU-positive and EdU-negative cells are indicated by the arrows/solid lines and arrowheads/dotted lines, respectively. EdU-positive spermatogonia, spermatocytes, and sperm were observed at days 7, 14, and 21, respectively. Experiment reproduced four times. Bars represent 50 µm.

Fig 5: Expression of Sox9 in in vivo and in vitro spermatogenesis.Testes were sampled from spawning (May) and non-spawning (September) adults. Cell aggregates formed in suspension culture of non-spawning adult testicular cells were sampled at first, second, and third week of culture. Sections were stained for Sox9 (green) and Vasa (red). Nuclei were stained with DAPI (blue). Arrowheads indicate Sox9-positive cells. The insets represent high-magnification images of Sox9-positive cells. Experiment reproduced three times. Bars represent 50 µm.