www.sciencemag.org/cgi/content/full/science.aaa2657/DC1

Supplementary Material for

foxl3 is a germ cell-intrinsic factor involved in sperm-egg fate decision in medaka

Toshiya Nishimura, Tetsuya Sato, Yasuhiro Yamamoto, Ikuko Watakabe, Yasuyuki

Ohkawa, Mikita Suyama, Satoru Kobayashi, Minoru Tanaka*

*Corresponding author. E-mail: mtanaka@nibb.ac.jp

Published 11 June 2015 on Science Express DOI: 10.1126/science.aaa2657

This PDF file includes:

Materials and Methods Figs. S1 to S14 Full Reference List

Materials and Methods Animals All treatments of animals in this study followed the guidelines of the National Institute for Basic Biology and were approved by the Institutional Animal Care and Use Committee of National Institutes of Natural Sciences. The OKcab strain and sox9b-DsRed/olvas-EGFP transgenic medaka (14) were used in this study. TALEN-induced mutagenesis TALEN target sites of foxl3 were searched using the TALEN Targeter program (https://tale-nt.cac.cornell.edu/node/add/talen)(17) using the following parameters: spacer length of 15–18 bp, repeat array of 16–18 bp, and upstream base of T only. TALEN assembly followed a modified version (18) of the original protocol (19). TALEN plasmids were linearized by NotI digestion and used as templates for in vitro RNA synthesis with the mMESSAGE mMACHINE® T7 transcription kit (Life technologies). TALEN mRNAs (250 ng/µl left and right) were injected into one- or two-cell stage embryos. The F0 founders were crossed with sox9b-DsRed/olvas-EGFP transgenic and non-transgenic medaka. Mutant alleles, N∆17 and FH∆8, were identified from the F1 adults by sequencing the PCR products amplified by following primer sets: foxl3-F 5’-ACTCCAGACATCGCTTTTGC-3’, foxl3-R 5’-GCTGTTCTGCCATCCTTTCT-3’. Unless otherwise indicated, all experiments were performed using the N∆17 mutants (fig. S4). In situ hybridization, immunohistochemistry, and histology Whole–mount in situ hybridization and immunohistochemistry were performed as previously described (20, 21). cDNA clones for foxl3 (clone name: olgi46a18), shippo1 (olte43m14), kif17 (olte3a15) and t-tectin (olte12i08) were obtained from NBRP medaka (http://www.shigen.nig.ac.jp/medaka/). FOXL3 polyclonal antibody was generated by immunizing rabbits with C-terminal peptide (CAPYGRQTESPALGFQSD, amino acids 246–262). For detection of FOXL3 protein, samples were treated with 5 mg/ml proteinase K (Roche) and 2N HCl for 15 min, respectively, prior to blocking. Can Get Signal solution B (TOYOBO) was used as the immunoreactive reagent (1:100). In addition, serum or antibodies specific for the following proteins were used: medaka OLVAS (20) (1:100), EGFP (1:100; mouse; Life Technologies), and DsRed (1:100; rabbit; Life Technologies). Secondary reagents were Alexa Fluor 488–, 568–, and 647–conjugated antibodies (1:100; Molecular Probes). For PAS staining, whole ovaries and testes were fixed in Bouin solution, and 4-µm–thick plastic sections were prepared using Technovit® 8100 (Heraeus Kulzer). The staining procedures were performed as previously described (22). EdU treatment Stage 35 embryos were treated with 100 µM EdU for 72 hours in balanced salt solution (BSS) (23), followed by EdU detection using the Click-iT EdU labeling kit (Invitrogen) and immunohistochemistry for FOXL3 and OLVAS.

2

Chimeric analysis foxl3–/– or foxl3+/–; sox9b-DsRed/olvas-EGFP transgenic embryos were used as donors, and non-transgenic embryos were used as hosts. The transplantation procedure was performed as previously described (24). Briefly, embryonic cells at the blastula stage were transplanted into embryos at the same stage. In this method, the contribution of somatic cells to chimeric gonads is rare, but cannot be totally ruled out. To exclude the possibility that a few somatic cells contributing to chimeric gonads affect the sexual fate of germ cells, only sox9b-DsRed–negative gonads were examined. Embryos with EGFP-positive germ cells were screened at st.35 and raised to 15 dph. In this chimeric analysis, we expected that approximately 85% of chimeric gonads without sox9b-expressing cells would not contain any gonadal somatic cells derived from donors, for two reasons: sox9b-expressing cells account for approximately 50% of total gonadal somatic cells in wild-type gonads at around 15 dph, and contamination of sox9b-expressing cells to chimeric gonads was less than 15% on average (i.e., 5 out of 34 chimeric gonads contained sox9b-expressing cells). Rescue construction The olvas promoter (5.1 kb) (25) and the 3’UTR (26) were amplified from a fosmid (GOLWFno476_m14) containing the olvas gene and inserted upstream and downstream of EGFP in hsGBA-NKm vector (27) using In-Fusion (Takara) (fig. S7). The resultant vector was further amplified using the following primer pair: foxl3-BAC-F, 5’-aaatggcctgataaggtaactgcaagtcacttacaaaacaaaaaatgtatccaatagaatgagtaatggt-3’; foxl3-BAC-R, 5’-atttgaggggattaaactgttggttaaagcacgactagagcgcgcgcacggtcgaccagttggtgattttg-3’. The amplified fragment was inserted downstream of the foxl3 locus and upstream of TMEM201 (ENSOELG00000005435) in the BAC (ola-212j01). The resultant rescue construct, which allowed germline integration of the transgenes to be monitored, was injected into progeny of foxl3+/– females crossed with foxl3–/– males (non-transgenic line) at the one-cell stage. Embryos with EGFP-positive germ cells were screened at st.35 and raised to 15 dph. RT-PCR Total RNA was extracted from adult gonads of wild-type, foxl3+/–, and foxl3–/– medaka using TriPure Isolation Reagent (Roche). Two testes and one ovary were used in each experiment. cDNA was produced from 800 ng total RNA using SuperScript III (Invitrogen) and used as the template for RT-PCR. PCR conditions and primer sets for dmrt1, foxl2, p45011β, and aromatase were described previously (13). Primer sets for foxl3 and β-actin were as follows: foxl3-F 5’-GACAGTCGGGACCAGAGAAA-3’, foxl3-R 5’- TGTGCCTTATGCTGTTCTGC-3’, β-actin-F 5’- TGGCGCTTGACTCAGGATTT-3’, β-actin-R 5’- GCAGATGCCTGGGGTGTTTA-3’. Two independent experiments were performed using independent pools of gonads. Artificial fertilization A gonad was dissected from 2- or 3-month-old foxl3–/– XX or wild-type XY medaka and minced with forceps to suspend sperm in 90 µl balanced salt solution (BSS) (23). In each experiment, 10–27 eggs were subsequently inseminated by the resultant sperm. The

3

inseminated eggs were incubated at room temperature for 15 min, with gentle mixing by pipetting every 5 min. After replacing the BSS, fertilization rate was assessed by monitoring activation of the egg membrane. The eggs were incubated in BSS at 29°C until hatching. The hatching rate was calculated based on the number of fertilized eggs. Fertility determination of female and male foxl3–/– mutants For fertility of female mutants, six pairs of 3-month-old foxl3+/– and –/– XX females were crossed with a foxl3+/– XY male. Eggs were collected every day for 3 days. The spawned and fertilized eggs were counted each day, and these counts were used as indicators of fecundity. The hatching rate was calculated based on the number of fertilized eggs. For fertility of male mutants, 10 pairs of foxl3–/– XY and six pairs of wild-type males were crossed with a wild-type female, and eggs were collected for 1 day. Fadrozole (FAD) and tamoxifen (TAM) treatment To block estrogen production or signaling, 15 dph larvae from foxl3+/– XX and –/– XY parents were kept in 1 L of water containing 100 ng/ml (0.38µM) fadrozole or 0.1 µM, 0.5 µM, or 2.5 µM tamoxifen (TAM) for 20 days. The water was changed every day. After 5 days under normal conditions, fox3–/– XX gonads at 40 dph were examined by immunohistochemistry. All larvae treated with 2.5 µM TAM were dead 1 day after the treatment was initiated. E2 treatment Embryos (0–7 dpf) were treated with 200 ng/ml E2, and larvae were treated with 100 ng/ml E2 for 5 days. The water was changed every day. After treatment, the larvae were kept under normal conditions for 1 week and examined by immunohistochemistry at 12 dph.

4

Fig. S1. foxl3 expression analysis by in situ hybridization during gonadal development. Cross-sections of the gonadal region. The gonad is encircled by black dotted lines. (A and B) No foxl3 signal is detected in either XX or XY gonads at st.33, the onset of gonadal formation. (C and D) foxl3 signals are detected in a subset of germ cells (arrows) in both XX and XY at st.35, the time of sex differentiation. (E and D) Signals were detected in single isolated type I germ cells (arrows) and cystic type II germ cells (blue dotted lines) in both XX and XY embryos at the hatching stage (h0d). (G to J), Whereas foxl3 signals disappear by 10 dph in XY gonads (H and J), they are continuously

5

detected in germ cells afterwards in XX gonads (G and I). Scale bars: 20 µm. (n) nephric duct. (g) hindgut.

6

Fig. S2. FOXL3 expression by immunohistochemistry during gonadal development. Ventral view of gonads (dotted lines). Germ cells (green) are detected by OLVAS expression. (A and B) FOXL3 protein is not detected in either XX (A) or XY (B) gonads at st.33. (C and D) FOXL3 signals (red) start to be detected in nuclei of germ cells (red arrowheads) at st.35 in both XX (C) and XY (D). (E, G and I) In XX gonads, FOXL3-positive germ cells are continuously detected from h0d (E) onwards (5dph in G and 10dph in I). (F, H and J) In XY gonads, FOXL3-positive germ cells are present at h0d (F) and 5 dph (H) but disappear by 10 dph (J), when only type I germ cells are present. Note that non-specific signals in nephric ducts (strong red behind the gonads) are detected by the FOXL3-antibody. Scale bars: 10 µm. st.33 XX: n = 8; XY: n = 8. st.35 XX: n = 15; XY: n = 15; h0d (hatching stage) XX: n = 19; XY: n = 20. 5 dph XX: n = 13, XY: n = 13. 10 dph XX: n = 12; XY: n = 13.These data are related to Fig. 1B.

7

Fig. S3. FOXL3 is expressed in mitotically active type I germ cells. Long-term EdU treatment from st.35 for 72 hr until hatching (h0d). (A) XX gonad at h0d. Insets show EdU (green) and FOXL3 (red) double-positive type I (right, dotted lines) and FOXL3 double-negative type I (left, arrowheads) germ cells. (B) XY gonad at h0d. The inset shows EdU and FOXL3 double-negative germ cells (arrowheads). (C and D) The number (C) and rate (D) of FOXL3-positive cells in mitotically quiescent [EdU(–)] and active [EdU(+)] type I germ cells at h0d. XX: n = 10, XY: n = 14. Scale bars: 10µm.

8

Fig. S4. Design of TALEN targeting foxl3. (A) foxl3 gene and target sites of TALENs (arrowheads). The numbers indicate the base positions starting at the adenine nucleotide of the initiation codon. (B and C) Nucleotide and amino-acid sequences of TALEN target sites (red), and the resultant deletions. Deletions of 17 bp upstream of the forkhead domain (B, N∆17) and 8 bp at the forkhead domain (C, FH∆8) were obtained. (D) Predicted amino-acid sequences of N∆17 and FH∆8 alleles. Asterisks indicate stop codons. (E to H) FOXL3 signals (red) are detected in foxl3+/– mutants for both N∆17 (E) and FH∆8 (G) at 7dph, but not in foxl3–/– mutants for both N∆17 (F) and FH∆8 (H) at 7 dph. Spermatogenesis (yellow dotted-lines) is observed in foxl3–/– XX mutants for FH∆8 at 7 dph (H). Blue cells are germ cells, detected by the presence of OLVAS. Scale bars: 10 µm.

9

Fig. S5. foxl3–/– XX gonads express marker genes for spermatogenesis. Signals of each gene are indicated in purple. The insets in (A to D) are magnified in the panels on the right. Cross-sections of gonads at 10 dph are encircled by black dotted lines. (A) Expression analysis of protamine, a transcript encoding a nuclear protein that replaces histones in spermatids. The protamine signals are detected in round spermatids in testes (yellow dotted line in A’). (B) Expression analysis of kif17 in testes, a transcript encoding testis-specific kinesin. The signals are detected in primary spermatocytes (B’ yellow dotted line). (C) Expression analysis of tektin-t in testes, a transcript encoding an axonemal protein in sperm tail. The signals are detected in primary spermatocytes (C’ yellow dotted line). (D) Expression analysis of shippo1 in testes, a transcript encoding a component of the sperm tail. The signals are detected from primary spermatocytes to round spermatids. (E to H) Expression analysis of protamine (E), kif17 (F), tektin-t (G), and shippo1 (H) in foxl3–/– XX gonads at 10 dph. All of these transcripts were detected in the foxl3–/– XX gonads. (I to L) None of the spermatogenic marker genes are expressed in foxl3+/– XX gonads at 10 dph. n ≥ 2. Scale bars: 20 µm.

10

Fig. S6. foxl3–/– germ cells progress through spermatogenesis in wild-type XX gonads. Ventral view of chimeric gonads at 15 dph. foxl3+/– or foxl3–/– cells were transplanted into wild-type embryos at the blastula stage. Germ cells derived from donors expressed EGFP under control of the olvas promoter (green, olvas-EGFP). Note that contribution of gonadal somatic cells was monitored by sox9b-DsRed–expressing cells, and only chimeric gonads without DsRed-positive cells were examined. (A to D) foxl3–/– germ cells in XX (A and C) and XY (B and D) wild-type gonads at 15 dph. Whereas both XX (A) and XY (C) donor germ cells (green) undergo spermatogenesis (yellow dotted lines in insets) in XX gonads, host germ cells (blue) undergo oogenesis. On the other hand, spermatogenesis does not initiate in XY gonads at this stage (B and D). (E to H) foxl3+/– germ cells in wild-type XX (E and G) and XY (F and H) gonads. Donor germ cells (green) behave in the same way as host germ cells (blue) do. The yellow arrowhead in (G) indicates a sox9b-DsRed–expressing cell located outside the chimeric gonad. Scale bars: 20µm.

11

Fig. S7. foxl3-rescue experiments. (A) foxl3-rescue construction. An olvas-EGFP cassette in which EGFP was expressed under the control of a 5.1 kb promoter and 3’UTR of olvas was inserted by homologous recombination into the BAC in the region between foxl3 and TMEM201. The resultant BAC was injected into foxl3+/– or foxl3–/– embryos. In this system, germ cells that integrated the transgene are labeled with EGFP. (B to D) Oocyte formation (dotted lines) in foxl3–/– XX gonad with the rescue transgenes at 15 dph. The white boxes in (B) are magnified in (C) and (D). The arrowhead in (C) indicates FOXL3 protein derived from the transgene. EGFP-negative germ cells undergo spermatogenesis (D). (E and F) foxl3+/– XX gonad with rescue transgenes at 15 dph. The white box is magnified in (F). Both EGFP-positive (yellow-dotted lines) and -negative (white-dotted lines) oocytes are observed. Red and yellow arrowheads indicate FOXL3 protein in EGFP-negative and -positive germ cells, respectively. n = 2. Scale bars: 10 µm.

12

13

Fig. S8. In situ hybridization for sex marker genes during gonadal development. Cross-sections of gonads are encircled by black dotted lines. Signals of each gene are indicated in purple. (A and B) Expression analysis of dmrt1, a male marker gene, at 10 dph (A) and 30 dph (B). dmrt1 is expressed in XY gonadal somatic cells. (C and D) Expression analysis of foxl2, a granulosa marker gene, at 10 dph (C) and 30 dph (D). foxl2 is expressed in somatic cells surrounding spermatogenic cells and oocytes of foxl3–

/– and foxl3+/– XX gonads, respectively. (E and F) Expression analysis of aromatase, estrogen synthetase, at 10 dph (E) and 30 dph (F). The aromatase signals appear on the ventral side of gonads in both foxl3–/– and +/– XX larva at 10 dph, and subsequently surround spermatogenic cells and oocytes at 30 dph. (G to I) Expression analysis of nobox, an oocyte marker gene, at 10 dph (G), 20 dph (H) and 30 dph (I). In foxl3 +/– XX gonads, the nobox signals in oocytes are observed from 10 dph. On the other hand, no signals are detected in foxl3–/– gonads at 10 dph. The signals appear in a subset of cystic germ cells (blue dotted line in H) at 20 dph and a very few oocytes (inset in I) at 30 dph. n ≥ 3. Scale bars: 10 µm.

14

Fig. S9. In situ hybridization for sex-marker and stem-marker genes in adult gonads. Cross-sections of adult gonads. Signals for each gene are indicated in purple. (A to C) Expression analysis of p45011β, 11-ketotestosterone synthetase, in the adult gonads. p45011β is only expressed in interstitial cells of testes (C, arrows). (D to F) Expression analysis of aromatase. The aromatase signals are detected in somatic cells surrounding oocytes of both foxl3+/– and –/– ovaries (D and E, arrows), and in somatic cells surrounding the expanded germinal epithelium filled with spermatogenic cells in foxl3–/– ovaries (E, arrowheads). (G to I) Expression analysis of nanos2, a marker of germline stem cells. The nanos2 signals are detected in a subset of germ cells (arrows) in the germinal cradle in foxl3+/– ovary (G), expanded germinal epithelium in foxl3–/– ovary (H), and lobule of foxl3–/– testis (I). n = 2. Scale bars: 10 µm.

15

Fig. S10. Secondary sex characteristics, gonadal structures, and fertility of foxl3–/– XX and XY fish. (A to D) XX wild-type and foxl3–/– adult fish form ovarian structures (C and D) and show female secondary sex characteristics (A and B), as characterized by the shape of dorsal/anal fins (dotted lines) and urogenital papilla (arrowheads). Asterisks indicate ovarian cavities. (E to H) XY wild-type and foxl3–/– adult fish form testes (G and H) and exhibit typical male secondary sex characteristics (E and F). Asterisks indicate efferent ducts. (I to K) foxl3–/– XX fish produce functional eggs. (I) Numbers of spawned eggs per day from foxl3+/– and –/– XX females paired with foxl3+/– XY males. n = 6 (pairs). (J) Fertilization rates. (K) Hatching rates of fertilized eggs in (J). (L and M) foxl3–/– XY fish produce functional sperm. (L) Fertilization rates of eggs from wild-type females crossed with wild-type (6 pairs) and foxl3–/– males (10 pairs). (M) Hatching rates of fertilized eggs in (L). Scale bars: 1 mm in A, B, E and F, and 100µm in C, D, G and H. Statistical significance was evaluated by two-tailed Student t-test: * p < 0.05, ** p < 0.001.

16

Fig. S11. Schematic representation of cross-sections of wild-type and foxl3–/– ovaries. (A) Wild-type ovary (left) and foxl3–/– ovary (right) (B) Germinal cradle (left) and expanded germinal epithelium (right). In the wild-type ovary, the germinal epithelium harbors the germinal cradles containing the germline stem cells, cystic germ cells, and early diplotene oocytes. As folliculogenesis proceeds, diplotene oocytes exit through the basement membrane towards the stromal compartment where oocyte growth and maturation occur. In the foxl3–/– ovary, early-stage germ cells (red cells in the right panel B), surrounded by sox9b-expressing cells, start spermatogenesis. As a result, sperm expand and fill the germinal epithelium. A few oocytes (orange cells in the right panel B) also develop in the germinal epithelium, possibly in an estrogen-independent manner. These oocytes exit from the germinal epithelium to form follicles, as seen in folliculogenesis of the wild-type ovary.

17

Fig. S12. A few contacts between spermatogenic cells and sox9b-expressing cells in foxl3–/– ovary. Immunohistochemistry of foxl3–/– XX gonads at 60 dph. (A) Early stage germ cells, possibly including germline stem cells, are surrounded by sox9b-expressing cells (yellow arrowheads). While small cluster of cystic germ cells (yellow dotted line) contacts with sox9b-expressing cells but is not thoroughly enclosed by them, only a few contacting sox9b-expressing cells are observed in a large cluster of cystic germ cells (magenta dotted line). (B) Yellow insets are magnified in the panels on the right. Primary spermatocytes at zygotene (upper, yellow dotted line) and pachytene (lower, yellow dotted line) stages appear to be isolated from sox9b-expressing cells. Scale bars: 20 µm.

18

Fig. S13. Inhibition of estrogen production and signaling does not block oocyte formation in foxl3–/– gonads. Larvae were treated with an aromatase inhibitor, fadrozole (FAD), and an estrogen receptor antagonist, tamoxifen (TAM), from 15 dph for 20 days. (A to D) Cross section of gonads. (A’ to D’) Ventral view of gonads. (A and B) FAD-untreated (A, control) and FAD-treated (B) foxl3–/– XX gonads at 40 dph (5 days after completion of the treatment). (C and D) TAM-untreated (C, control) and TAM treated (D) foxl3–/– XX gonads at 40 dph. Note ovarian cavities (asterisks in A and C) develop in control gonads, but not in drug-treated gonads (B and D), indicating that E2 production and activity are blocked by the drug treatments. Arrows in (D) indicate the primordium of gonadal elongation for the formation of ovarian cavity. Arrowheads indicate diplotene oocytes. (E and F) Quantitation of oocyte number in FAD- (E) and TAM- (F) treated gonads. Scale bars: 20 µm.

19

Fig. S14. Estrogen (E2) treatment during gonadal development does not induce oocytes in foxl3–/– gonads. Fertilized eggs were treated with E2 until 5 dph. (A to F) E2-untreated (control) and E2-treated gonads at 12 dph (1 week after completion of the treatment). Oocyte formation is induced in E2-treated foxl3+/– XY gonads (A and D, insets), but not in both foxl3–/– XY (B and E) or XX (C and F) gonads. (G) Quantitation of the number of pachytene and diplotene oocytes (insets in D). Scale bars: 10 µm.

20

References and Notes 1. M. Matsuda, Y. Nagahama, A. Shinomiya, T. Sato, C. Matsuda, T. Kobayashi, C. E. Morrey,

N. Shibata, S. Asakawa, N. Shimizu, H. Hori, S. Hamaguchi, M. Sakaizumi, DMY is a Y-specific DM-domain gene required for male development in the medaka fish. Nature 417, 559–563 (2002). Medline doi:10.1038/nature751

2. I. Nanda, M. Kondo, U. Hornung, S. Asakawa, C. Winkler, A. Shimizu, Z. Shan, T. Haaf, N. Shimizu, A. Shima, M. Schmid, M. Schartl, A duplicated copy of DMRT1 in the sex-determining region of the Y chromosome of the medaka, Oryzias latipes. Proc. Natl. Acad. Sci. U.S.A. 99, 11778–11783 (2002). Medline doi:10.1073/pnas.182314699

3. M. T. Geraldo, G. T. Valente, A. S. Braz, C. Martins, The discovery of Foxl2 paralogs in chondrichthyan, coelacanth and tetrapod genomes reveals an ancient duplication in vertebrates. Heredity 111, 57–65 (2013). Medline doi:10.1038/hdy.2013.19

4. B. Crespo, O. Lan-Chow-Wing, A. Rocha, S. Zanuy, A. Gómez, foxl2 and foxl3 are two ancient paralogs that remain fully functional in teleosts. Gen. Comp. Endocrinol. 194, 81–93 (2013). Medline doi:10.1016/j.ygcen.2013.08.016

5. D. Baron, J. Cocquet, X. Xia, M. Fellous, Y. Guiguen, R. A. Veitia, An evolutionary and functional analysis of FoxL2 in rainbow trout gonad differentiation. J. Mol. Endocrinol. 33, 705–715 (2004). Medline doi:10.1677/jme.1.01566

6. C. Ottolenghi, S. Omari, J. E. Garcia-Ortiz, M. Uda, L. Crisponi, A. Forabosco, G. Pilia, D. Schlessinger, Foxl2 is required for commitment to ovary differentiation. Hum. Mol. Genet. 14, 2053–2062 (2005). Medline doi:10.1093/hmg/ddi210

7. D. Schmidt, C. E. Ovitt, K. Anlag, S. Fehsenfeld, L. Gredsted, A. C. Treier, M. Treier, The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance. Development 131, 933–942 (2004). Medline doi:10.1242/dev.00969

8. N. H. Uhlenhaut, S. Jakob, K. Anlag, T. Eisenberger, R. Sekido, J. Kress, A. C. Treier, C. Klugmann, C. Klasen, N. I. Holter, D. Riethmacher, G. Schütz, A. J. Cooney, R. Lovell-Badge, M. Treier, Somatic sex reprogramming of adult ovaries to testes by FOXL2 ablation. Cell 139, 1130–1142 (2009). Medline doi:10.1016/j.cell.2009.11.021

9. M.-H. Li, H. H. Yang, M. R. Li, Y. L. Sun, X. L. Jiang, Q. P. Xie, T. R. Wang, H. J. Shi, L. N. Sun, L. Y. Zhou, D. S. Wang, Antagonistic roles of Dmrt1 and Foxl2 in sex differentiation via estrogen production in tilapia as demonstrated by TALENs. Endocrinology 154, 4814–4825 (2013). Medline doi:10.1210/en.2013-1451

10. D. Saito, C. Morinaga, Y. Aoki, S. Nakamura, H. Mitani, M. Furutani-Seiki, H. Kondoh, M. Tanaka, Proliferation of germ cells during gonadal sex differentiation in medaka: Insights from germ cell-depleted mutant zenzai. Dev. Biol. 310, 280–290 (2007). Medline doi:10.1016/j.ydbio.2007.07.039

11. T. Nishimura, M. Tanaka, Gonadal development in fish. Sex Dev. 8, 252–261 (2014). Medline doi:10.1159/000364924

12. S. Nakamura, I. Watakabe, T. Nishimura, J. Y. Picard, A. Toyoda, Y. Taniguchi, N. di Clemente, M. Tanaka, Hyperproliferation of mitotically active germ cells due to defective

anti-Müllerian hormone signaling mediates sex reversal in medaka. Development 139, 2283–2287 (2012). Medline doi:10.1242/dev.076307

13. H. Kurokawa, D. Saito, S. Nakamura, Y. Katoh-Fukui, K. Ohta, T. Baba, K. Morohashi, M. Tanaka, Germ cells are essential for sexual dimorphism in the medaka gonad. Proc. Natl. Acad. Sci. U.S.A. 104, 16958–16963 (2007). Medline

14. S. Nakamura, K. Kobayashi, T. Nishimura, S. Higashijima, M. Tanaka, Identification of germline stem cells in the ovary of the teleost medaka. Science 328, 1561–1563 (2010). Medline doi:10.1126/science.1185473

15. A. Saiki, M. Tamura, M. Matsumoto, J. Katowgi, A. Watanabe, K. Onitake, Establishment of in vitro spermatogenesis from spermatocytes in the medaka, Oryzias latipes. Dev. Growth Differ. 39, 337–344 (1997). Medline doi:10.1046/j.1440-169X.1997.t01-2-00009.x

16. A. Suzuki, M. Tanaka, N. Shibata, Expression of aromatase mRNA and effects of aromatase inhibitor during ovarian development in the medaka, Oryzias latipes. J. Exp. Zoolog. A Comp. Exp. Biol. 301A, 266–273 (2004). Medline doi:10.1002/jez.a.20027

17. E. L. Doyle, N. J. Booher, D. S. Standage, D. F. Voytas, V. P. Brendel, J. K. Vandyk, A. J. Bogdanove, TAL Effector-Nucleotide Targeter (TALE-NT) 2.0: Tools for TAL effector design and target prediction. Nucleic Acids Res. 40 (W1), W117–W122 (2012). Medline doi:10.1093/nar/gks608

18. T. Sakuma, S. Hosoi, K. Woltjen, K. Suzuki, K. Kashiwagi, H. Wada, H. Ochiai, T. Miyamoto, N. Kawai, Y. Sasakura, S. Matsuura, Y. Okada, A. Kawahara, S. Hayashi, T. Yamamoto, Efficient TALEN construction and evaluation methods for human cell and animal applications. Genes Cells 18, 315–326 (2013). Medline doi:10.1111/gtc.12037

19. T. Cermak, E. L. Doyle, M. Christian, L. Wang, Y. Zhang, C. Schmidt, J. A. Baller, N. V. Somia, A. J. Bogdanove, D. F. Voytas, Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res. 39, e82 (2011). 10.1093/nar/gkr218 Medline doi:10.1093/nar/gkr218

20. Y. Aoki, I. Nagao, D. Saito, Y. Ebe, M. Kinjo, M. Tanaka, Temporal and spatial localization of three germline-specific proteins in medaka. Dev. Dyn. 237, 800–807 (2008). Medline doi:10.1002/dvdy.21448

21. S. Nakamura, D. Kobayashi, Y. Aoki, H. Yokoi, Y. Ebe, J. Wittbrodt, M. Tanaka, Identification and lineage tracing of two populations of somatic gonadal precursors in medaka embryos. Dev. Biol. 295, 678–688 (2006). Medline doi:10.1016/j.ydbio.2006.03.052

22. I. Quintero-Hunter, H. Grier, M. Muscato, Enhancement of histological detail using metanil yellow as counterstain in periodic acid Schiff’s hematoxylin staining of glycol methacrylate tissue sections. Biotech. Histochem. 66, 169–172 (1991). Medline doi:10.3109/10520299109109964

23. H. Yang, T. R. Tiersch, Current status of sperm cryopreservation in biomedical research fish models: Zebrafish, medaka, and Xiphophorus. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 149, 224–232 (2009). Medline doi:10.1016/j.cbpc.2008.07.005

24. S. Nakamura, I. Watakabe, T. Nishimura, A. Toyoda, Y. Taniguchi, M. Tanaka, Analysis of medaka sox9 orthologue reveals a conserved role in germ cell maintenance. PLOS ONE 7, e29982 (2012). 10.1371/journal.pone.0029982 Medline doi:10.1371/journal.pone.0029982

25. M. Tanaka, M. Kinoshita, D. Kobayashi, Y. Nagahama, Establishment of medaka (Oryzias latipes) transgenic lines with the expression of green fluorescent protein fluorescence exclusively in germ cells: A useful model to monitor germ cells in a live vertebrate. Proc. Natl. Acad. Sci. U.S.A. 98, 2544–2549 (2001). Medline doi:10.1073/pnas.041315498

26. H. Kurokawa, Y. Aoki, S. Nakamura, Y. Ebe, D. Kobayashi, M. Tanaka, Time-lapse analysis reveals different modes of primordial germ cell migration in the medaka Oryzias latipes. Dev. Growth Differ. 48, 209–221 (2006). Medline doi:10.1111/j.1440-169X.2006.00858.x

27. S. Nakamura, D. Saito, M. Tanaka, Generation of transgenic medaka using modified bacterial artificial chromosome. Dev. Growth Differ. 50, 415–419 (2008). Medline doi:10.1111/j.1440-169X.2008.01027.x