Abundance of Early Embryonic Primordial Germ Cells Promotes Zebrafish Female Differentiation as Revealed by Lifetime Labeling of Germline
Teleost sex differentiation largely depends on the number of undifferentiated germ cells. Here, we describe the generation and characterization of a novel transgenic zebrafish line, Tg(piwil1:egfp-UTRnanos3)ihb327Tg, which specifically labels the whole lifetime of germ cells, i.e., from primordial germ cells (PGCs) at shield stage to the oogonia and early stage of oocytes in the ovary and to the early stage of spermatogonia, spermatocyte, and spermatid in the testis. By using this transgenic line, we carefully observed the numbers of PGCs from early embryonic stage to juvenile stage and the differentiation process of ovary and testis. The numbers of PGCs became variable at as early as 1 day post-fertilization (dpf). Interestingly, the embryos with a high amount of PGCs mainly developed into females and the ones with a low amount of PGCs mainly developed into males. By using transient overexpression and transgenic induction of PGC-specific bucky ball (buc), we further proved that induction of abundant PGCs at embryonic stage promoted later ovary differentiation and female development. Taken together, we generate an ideal transgenic line Tg(piwil1:egfp-UTRnanos3)ihb327Tg which can visualize zebrafish germline for a lifetime, and we have utilized this line to study germ cell development and gonad differentiation of teleost and to demonstrate that the increase of PGC number at embryonic stage promotes female differentiation.
Keywordspiwil1 Spermatogenesis Oogenesis Primordial germ cell Spermatogonia Oogonia
In a large amount of teleosts, the number of undifferentiated germ cells strongly contribute to sex differentiation and sexual dimorphism. For instance, in zebrafish (Danio rerio), medaka (Oryzias latipes), Nile tilapia (Oreochromis niloticus), three-spined stickleback (Gasterosteus aculeatus), and even the gynogenetic all-female fish, Carassius gibelio, depletion of primordial germ cells usually leads to all-male development (Houwing et al. 2007; Lewis et al. 2008; Li et al. 2017; Saito et al. 2007; Siegfried and Nusslein-Volhard 2008). Previous studies have shown that zebrafish, although possesses a polygenetic sex determination system, shows different germ cell proliferations between definitive male and definitive female in larval stage at 14 days post-fertilization (dpf) (Liew et al. 2012; Tong et al. 2010; Tzung et al. 2015). A strong male-biased sex ratio was observed in primordial germ cell (PGC)-less population, indicating that a threshold number of PGCs is required for ovarian differentiation (Tzung et al. 2015). Transplantation of a single PGC into a germ cell-deficient embryos generates males exclusively in zebrafish (Saito et al. 2008). However, whether the initial PGC number at early embryonic stage contributes to zebrafish sex differentiation remains largely unknown.
The gonadal differentiation process of teleost has been thoroughly described using zebrafish as a model system. Zebrafish gonad in 17–21 dpf larva is generally considered to be undifferentiated (Rodriguez-Mari et al. 2005; Tzung et al. 2015; Wang et al. 2007). Primary testis and ovary are morphologically and molecularly distinguishable at around 30 dpf (Rodriguez-Mari et al. 2005). During testis differentiation, a “juvenile ovary-to-testis” transformation process usually occurs from 30 to 40 dpf (Maack and Segner 2003; Wang et al. 2007), which is accompanied by the Tp53-mediated apoptosis of oocytes (Rodriguez-Mari et al. 2010; Uchida et al. 2002). The cysts containing spermatogonia, spermatocytes, and spermatids can be observed in testes of males at 2 months post-fertilization (mpf) (Leu and Draper 2010; Maack and Segner 2003). The differentiation of ovary is considered to start earlier and to last for longer than that of the testis. The perinuclear oocytes (stage 1B) can be visible from 25 to 40 dpf (Lau et al. 2016; Siegfried and Nusslein-Volhard 2008), the cortical alveolar oocytes (stage II) are identical from 50 dpf (Krovel and Olsen 2004), and the matured oocytes form at 3 mpf (Yang et al. 2017). However, most of the previous results rely on chemical staining or immunostaining on gonad sections; the gonad development of zebrafish has never been investigated at the intact-gonad level with high resolution for a lifetime duration, which largely relies on a transgenic line labeling the whole lifetime of germ cells.
Several transgenic lines have been generated to visualize different types of germ cells at different stages of zebrafish. For undifferentiated germ cells, the askops (kop) promoter has been used to label the migrating PGCs until 3 dpf (Blaser et al. 2005), the Tg(ddx4:ddx4-EGFP)zf45Tg (previous name: Tg(vasa:EGFP)) can be used to label PGCs from 1 dpf (Krovel and Olsen 2002), the piwil1 promoter in Tg(piwil1:EGFP)uc1Tg (previous name: Tg(ziwi:EGFP)) can specifically drive EGFP expression in the undifferentiated germ cells after 7 dpf (Leu and Draper 2010), and by using a Gal4/UAS system and the kop promoter, our previous study has shown that the undifferentiated germ cells could be visible from the shield stage to 25 dpf (Xiong et al. 2013). For female germ cells, the Tg(-0.5zp3b:GFP) m1032Tg (previous name: Tg(zpc0.5:EGFP)) labels all stages of oocytes (Onichtchouk et al. 2003), the vasa promotor can be used to label the early stages of oocyte but not oogonia (pre-meiotic germ cells) (Krovel and Olsen 2004; Leu and Draper 2010; Wang et al. 2007), and Tg(piwil1:EGFP)uc1Tg could specifically label the oogonia and oocytes (Leu and Draper 2010), whereas the Tg(alk8:EGFP) only labels the early chromatin nucleolus-stage oocytes (stage IA oocytes) (Payne-Ferreira et al. 2004). For male germ cells, both Tg(ddx4:ddx4-EGFP)zf45Tg and Tg(piwil1:EGFP)uc1Tg label the spermatogonia and spermatocytes (Leu and Draper 2010). To the best of our knowledge, there is no report of transgenic fish labeling the whole lifetime of germ cells, from the early PGCs to the fully differentiated gametes.
In order to accomplish a lifetime visualization of zebrafish germ cells, early post-transcription regulation elements like nanos3 3′-UTR and later transcriptional regulation sequences such as piwil1 promoter might be coordinately combined. Piwil1 belongs to Argonaute protein family which interacts with Piwi-interacting RNA (piRNA) to mediate the post-transcriptional regulation of mRNA and transposon silencing in the animal germ line (Houwing et al. 2007), and a 4.7-kb promoter region could direct the expression of EGFP in germ cells from 7 dpf (Leu and Draper 2010). Nanos3 is one of the germ plasm components and is essential for maintaining oocyte production (Draper et al. 2007; Kosaka et al. 2007). During early embryogenesis, the specific region of 3′-UTR of nanos3 can be recognized by the miR430 which mediate maternal mRNA degradation (Mickoleit et al. 2011; Mishima et al. 2006), and Dnd1 can bind to the Uracil-rich region of 3′-UTR of nanos3 to protect the mRNA from miR430-mediated degradation in PGCs (Kedde et al. 2007; Mishima et al. 2006). Thus, the nanos3 3′-UTR has been widely used to mediated PGC-specific gene expression in either transient overexpression experiment or transgenic study (Saito et al. 2006; Xiong et al. 2013).
In this study, by using a combination of the piwil1 promoter and nanos3 3′-UTR, we successfully generated the transgenic zebrafish line, Tg(piwil1:egfp-UTRnanos3)ihb327Tg, which could label the whole lifetime of zebrafish germ cells, from early PGCs to oocytes in females and spermatids in males. Furthermore, we carefully traced the development of gonad in fish with different numbers of PGCs from the embryonic stage until adulthood, and proved that the numbers of PGCs became variable at as early as 1 dpf and abundant PGCs at embryonic stage promoted later ovary differentiation and female development.
Materials and Methods
Embryos were obtained from the natural mating of zebrafish of the AB genetic background (from the China Zebrafish Resource Center, CZRC, Wuhan, China; Web: http://zfish.cn) and maintained, raised, and staged as previously described (Kimmel et al. 1995). The density of fish is 50 larvae/L for age from 5 dpf to 15 dpf, 20 fish/L for age 15 dpf to 1 mpf, 10 fish/L for age from 1 mpf to 2 mpf, and 5 fish/L for age older than 2 mpf. The experiments involving zebrafish were performed under the approval of the Institutional Animal Care and Use Committee of the Institute of Hydrobiology, Chinese Academy of Sciences.
We download the genomic sequence of piwil1 from Ensembl (ENSDARG00000041699) and designed the primer pairs to amplify the 5 kb transcriptional regulatory element containing the exon 2 and its upstream sequence according to a previous report (Leu and Draper 2010). The forward primer is 5′-CAAAGAACGAACACCTGGTTGG-3′ and the reverse primer is 5′-ACCGGTGCTTTACAAATGCTGA-3′. The 5-kb fragment was inserted into the upstream of EGFP coding sequence of construct pTol2(EGFP-UTRnanos3) (Kwan et al. 2007; Xiong et al. 2013) to generate the transgenic plasmid, pTol2(piwil1:EGFP-UTRnanos3). The transgenic construct of pTol2(UAS:buc-UTRnanos3) was made based on the pTol2(UAS:mRFP-UTRnanos3) (Xiong et al. 2013). The CDS region of bucky ball (buc) was amplified from zebrafish ovary cDNA and inserted into the pTol2(UAS:mRFP-UTRnanos3) construct to replace the mRFP coding sequence. In-fusion cloning kit (TaKaRa) was used for plasmid construction. All constructs were verified by Sanger sequencing.
Generation of Transgenic Fish
To generate transgenic zebrafish Tg(piwil1:egfp-UTRnanos3), 50 ng of transgenic construct pTol2(piwil1:egfp-UTRnanos3) was injected into the animal pole of 1-cell embryo and 100 pg transposase mRNA was injected into the yolk. The F0 embryos were screened under fluorescence stereomicroscope (AF205, Leica). The embryos with bright fluorescence were picked and raised up. The F0 fish were mated with wildtype fish to generate F1 fish. The 2 dpf F1 embryos with bright EGFP-labeled PGCs were selected to raise up. At 14 dpf, the F1 larvae were screened to exclude non-specific EGFP expression at other parts of the body. The F1 adult fish were mated with wildtype fish to generate F2 transgenics. The F2 embryos were also screened at 2 dpf and 14 dpf to exclude non-specific EGFP labeling. Finally, the Tg(piwil1:egfp-UTRnanos3)ihb327Tg was established and deposited in CZRC.
RT-qPCR primer list
Reverse-Transcription Quantitative PCR Analysis
Embryos at 1 dpf were collected for RNA extraction and reverse transcription. The BioRad CFX Connect Real-Time System was used for transcript quantification. Samples were tested in triplicate for each gene, and resultant Cq values were averaged. Primer efficiencies and gene expression levels were calculated according to the previous study (Pfaffl 2001). β-actin was selected as a reference gene. Data was processed using 2−ΔΔCq method. All reverse-transcription quantitative PCR (RT-qPCR) gene-specific primers are listed in Table 1.
Whole-Mount In Situ Hybridization
Digoxigenin-labeled antisense RNA probe of zebrafish vasa was synthesized by in vitro transcription. Whole-mount in situ hybridization (WISH) was performed as described (Wei et al. 2014).
Whole-mount immunofluorescence was performed as described previously with some modifications (Ye and Lin 2013). Incubation of the first antibody was performed at 4 °C overnight. Incubation of secondary antibody was performed at room temperature (23 °C) for 3 h. Samples were immerged in 50% glycerol-PBS at 4 °C overnight before confocal microscopy. The rabbit polyclonal antibody against zebrafish Vasa was used (GTX128306, Gentex, 1:200). Anti-rabbit Alexa Fluor 568 was used as secondary antibody (Molecular Probes).
Transgenic larvae at specific stages were dissected by fine tweezers (5SA-JP, VETUS). The gut and liver were removed to expose the swim bladder. The gonad was just located on both sides of the swim bladder. The swim bladder exposed larvae were fixed in 4% methanol-free PFA at 4 °C overnight. After washing with 1× PBS/Triton X-100 (0.1%), the gonads were isolated under fluorescence stereomicroscope (AF205, Leica). The gonads were stained by phalloidin-Alexa Fluor 568 (Molecular Probes) and DAPI, to visualize F-actin and nuclear DNA, respectively. The gonads were then infiltrated by 50% glycerol-PBS at 4 °C overnight. Confocal images were acquired using a laser-scanning confocal inverted microscope (SP8, Leica) with a LD C-Apo 40×/NA 1.1 water objective. Z-stacks were generated from images taken at 0.5-μm intervals, using the following settings (2048 × 2048 pixel, 400 MHz). EGFP intensity and nuclear diameter were measured by Fiji software (Schindelin et al. 2012). The Cell Counter plugin was used to count the number of oogonia, oocyte IA, and oocyte IB. All data were exported to Excel for statistical analysis. The graphs were generated and the Student t test was performed using GraphPad Prism software.
Lifetime Labeling of Zebrafish Germ Cells in Tg(piwil1:egfp-UTRnanos3)ihb327Tg
To verify that the EGFP specifically labeled the PGCs at the embryonic stage, we performed immunofluorescence using Vasa antibody on transgenic embryos at 1 dpf. As shown in Fig. 1 i–k, the Vasa signals were perfectly co-localized with EGFP, suggesting that the EGFP were exclusively expressed in the PGCs. Taken together, we successfully generate a transgenic line which labels germ cell lineage throughout the whole lifetime of zebrafish.
Early Dimorphic Gonads Lead to Male- or Female-Biased Development
The dimorphic sizes became more obvious at 20 dpf. To carefully examine the cellular characters of both small and big gonads, we performed confocal microscopy (Fig. 2e (e1, e4)). Most of the germ cells in both small and big gonads were gonocytes with a visible nucleolus (Fig. 2e (e3, e8)). Occasionally, stage IA oocytes could be seen in big gonads (Fig. 2(e6) “arrow”). Interestingly, we observed that the nuclei of most germ cells in big gonad were of irregular shape rather than round shape while the nuclei in most of the germ cells in small gonad were round (Fig. 2(e2, e3 e5, e7)). These data suggest that small and big gonads may undergo different morphogeneses of germ cells, leading to a phenotype of gonadal dimorphism from 20 dpf.
Abundance of Early PGCs Promotes Female Development
To further verify the above finding, we performed PGC-specific overexpression of buc using Gal4/UAS inductive overexpression method by crossing heterozygote Tg(kop:KalTA4-UTRnanos3,CMV:EGFP)ihb8Tg females with heterozygote Tg(UAS:buc-UTRnanos3)ihb120Tg males. All the offspring with mixed genetic background were raised together in the same tank until adulthood. Their genotypes and genders were individually identified either by PCR or by expression of EGFP. To verify that buc was overexpressed in the double transgenics, we performed RT-qPCR analysis on 1 dpf embryos of double-transgenic fish and Tg(UAS:buc-UTRnanos3)ihb120Tg using the primer pairs amplify the region covering the junction of buc and nanos3 3′-UTR (buc-UTRnanos3) (Fig. 3d). The expression of both the coding region of buc and the junction of buc-UTRnanos3 was remarkably increased in the Tg(UAS:buc-UTRnanos3)ihb120Tg or double-transgenic fish when compared with the wildtype siblings (Fig. 3d). Similar to the result from buc injection, the double-transgenic siblings showed female bias compared to wildtype siblings (Fig. 3e), and the Tg(UAS:buc-UTRnanos3)ihb120Tg siblings also showed pro-female development, due to the reason that the maternally deposited Gal4 proteins in the eggs of heterozygote Tg(kop:KalTA4-UTRnanos3) female could activate the buc expression in Tg(UAS:buc-UTRnanos3)ihb120Tg embryos. Taken together, our data demonstrate that abundant embryonic PGCs facilitate the female differentiation.
Ovary Development in Tg(piwil1:egfp-UTRnanos3)ihb327Tg
Testis Development in Tg(piwil1:egfp-UTRnanos3)ihb327Tg
Visualization of the process of testis development and specific labeling of germ cells by Tg(piwil1:egfp-UTRnanos3)ihb327Tg provided us an opportunity to identify different cell types in intact gonads during spermatogenesis. The different germ cell stages during spermatogenesis are generally divided into main four classes—spermatogonia, primary spermatocytes, secondary spermatocytes, and spermatid (Schulz et al. 2010). We carefully measured diameters of cell nuclei and intensities of different germ cells to characterize germ cells of different developmental stages from 45 dpf (Fig. 5 e(e1–e8)).
Based on both nuclear morphology and EGFP intensity of germ cells on a 45 dpf testis, there are two subtypes of spermatogonia—one with less number and higher EGFP intensity, whereas the other with more cells but lower EGFP intensity. However, there is no significant difference in nuclear diameter between early and late stages of spermatogonia (Fig. 5 e(e1, e2)). Our observation indicates that the spermatogonia with lower EGFP intensity is at a later stage and the one with higher EGFP intensity is at an earlier stage which might be the type A spermatogonia. As meiotic prophase proceeded, the intensity of EGFP increased and reached a peak at an early stage of spermatid, and the average intensity decreased but become aggregated in the late stage of spermatid (Fig. 5 e(e3–e8), g, h). Taken together, the early development of testis and spermatogenesis can be characterized in details at intact-gonad level using Tg(piwil1:egfp-UTRnanos3)ihb327Tg.
Transgenic labeling of germ cells by fluorescent proteins provides us with a powerful tool to study the germ cell development and gonad differentiation of teleost. In the present study, we have successfully generated a transgenic zebrafish line Tg(piwil1:egfp-UTRnanos3)ihb327Tg, in which the lifetime germline could be specifically and efficiently labeled for the first time. By using this transgenic line, we were able to visualize the lifetime development of zebrafish germline and to analyze the relationship between initial embryonic PGC number and the later gonad differentiation.
In the previous studies, although different types of germ cells at different stages have been labeled by various transgenic zebrafish, none of them could label the whole lifetime of zebrafish germline. For instance, a similar transgenic line, Tg(piwil1:EGFP)uc1Tg could label the germline from 7 dpf but not the embryonic PGCs before that, probably due to that a different SV40 polyA signal was used in their study (Leu and Draper 2010). In our transgenic fish, the use of piwil1 promoter ensures the maternal expression of EGFP and the use of 3′-UTR of nanos3 ensure its restricted expression in PGCs. Thus, the embryonic PGCs could be labeled as early as the shield stage. During spermatogenesis, unlike the Tg(piwil1:EGFP)uc1Tg, which showed highest EGFP expression level in spermatogonia (Leu and Draper 2010), our Tg(piwil1:egfp-UTRnanos3)ihb327Tg displayed different expression levels in different types of spermatogenic cells, with a relatively high level in spermatocyte and spermatid. Considering that the Tg(piwil1:egfp-UTRnanos3)ihb327Tg has specific expression pattern on labeling different types of germ cells particularly the germline stem cells, it will be of great interest to use this transgenic fish to identify and sort out the early type A spermatogonia, oogonia, and PGCs combining the usage of fluorescent-activated cell sorting (FACS) in the future. Those sorted germline stem cells can be used not only for transcriptome analysis but also for research on surrogate reproduction (Morita et al. 2015).
It is reported that the number of PGCs plays critical roles in sexual development and the depletion of PGCs usually results in infertile male in some fish species such as zebrafish, medaka, and Carassius gibelio (Kurokawa et al. 2007; Liu et al. 2015; Siegfried and Nusslein-Volhard 2008). However, there is no report of the effects of early germ cell induction on later sex differentiation. In our present study, we utilized the Tg(piwil1:egfp-UTRnanos3)ihb327Tg transgenic line to unveil that rich embryonic PGC number led to later ovary differentiation and female development and less embryonic PGC number resulted in later male development. More interestingly, we utilized the method of buc overexpression (Distel et al. 2009), by either transient overexpression with mRNA injection or stable overexpression with Gal4/UAS transgenic system (Xiong et al. 2013) to induce ectopic PGCs at shield stage. The induction of PGCs at shield stage could also promote female differentiation. All these data demonstrate that the initial number of PGCs should greatly contribute to the sex differentiation in zebrafish. In the future, it will be interesting to test whether different amounts of germplasm exist in different oocytes and which mechanism is responsible for this type of germplasm variation. Since the Gal4/UAS system has been successfully utilized for PGC-specific overexpression of buc which leads to female-biased development, it might have great application in female-biased breeding by induction of female development in some commercial fish species with sexual growth dimorphism (Mei and Gui 2015).
We thank Kuoyu Li from China Zebrafish Resource Center (CZRC) for taking care of the zebrafish. We thank Fang Zhou from the Analysis and Testing Center of Institute of Hydrobiology, CAS, for the technical support of confocal imaging.
This work was supported by the National Key R&D Program of China (grant no: 2018YFD0901205), National Natural Science Foundation of China (grant nos. 31721005 and 31872550), Strategic Priority Research Program of Chinese Academy of Sciences (grant no: XDA08010106), and Projects from State Key Laboratory of Freshwater Ecology and Biotechnology (grant nos. 2016FBZ03, 2015FB11).
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflict of interest.
- Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF (1995) Stages of embryonic development of the zebrafish. Dev Dyn 203(3):253–310Google Scholar
- Rodriguez-Mari A, Yan YL, Bremiller RA, Wilson C, Canestro C, Postlethwait JH (2005) Characterization and expression pattern of zebrafish anti-Mullerian hormone (Amh) relative to sox9a, sox9b, and cyp19a1a, during gonad development. Gene Expr Patterns 5:655–667Google Scholar
- Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9:676–682CrossRefGoogle Scholar
- Uchida D, Yamashita M, Kitano T, Iguchi T (2002) Oocyte apoptosis during the transition from ovary-like tissue to testes during sex differentiation of juvenile zebrafish. J Exp Biol 205:711–718Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.