Follicle-stimulating hormone (Fsh) is a major regulator of spermatogenesis, targeting somatic cell functions in the testes. We reported previously that zebrafish Fsh promoted the differentiation of type A undifferentiated spermatogonia (Aund) by stimulating the production of factors that advance germ cell differentiation, such as androgens, insulin-like peptide 3 (Insl3) and insulin-like growth factor 3 (Igf3). In addition, Fsh also modulated the transcript levels of several other genes, including some belonging to the Wnt signaling pathway. Here, we evaluated if and how Fsh utilizes part of the canonical Wnt pathway to regulate the development of spermatogonia. We quantified the proliferation activity and relative section areas occupied by Aund and type A differentiating (Adiff) spermatogonia and we analyzed the expression of selected genes in response to recombinant proteins and pharmacological inhibitors. We found that from the three downstream mediators of Fsh activity we examined, Igf3, but not 11-ketotestosterone or Insl3, modulated the transcript levels of two β-catenin sensitive genes (cyclinD1 and axin2). Using a zebrafish β-catenin signaling reporter line, we showed that Igf3 activated β-catenin signaling in type A spermatogonia and that this activation did not depend on the release of Wnt ligands. Pharmacological inhibition of the β-catenin or of the phosphoinositide 3-kinase (PI3K) pathways revealed that Igf3 activated β-catenin signaling in a manner involving PI3K to promote the differentiation of Aund to Adiff spermatogonia. This mechanism represents an intriguing example for a pituitary hormone like Fsh using Igf signaling to recruit the evolutionary conserved, local β-catenin signaling pathway to regulate spermatogenesis.
Long-term spermatogenesis relies on a population of undifferentiated germ cells including spermatogonial stem cells (SSCs). They reside in a specific microenvironment, the SSC niche, formed by somatic cells such as Sertoli cells (SCs), Leydig cells (LCs) and myoid cells, which provide SSCs with signals relevant to modulate their behavior (Oatley et al. 2009, Ding et al. 2011, Chen et al. 2016a, Lord & Oatley 2017, De Rooij 2017). SSCs can be quiescent or proliferate either to self-renew thereby producing more SSCs or undergo a differentiating proliferation to eventually produce spermatozoa. A balanced self-renewal/differentiation activity is critical to maintain long-term spermatogenesis and testis tissue homeostasis (De Rooij 2015, 2017).
The pituitary gonadotropin Fsh is a major regulator of spermatogenesis. In the mammalian testis, the FSH receptor gene (FSHR) is expressed exclusively by SCs. FSH stimulates the production of growth factors important for spermatogonial proliferation and differentiation, such as FGF2 or IGF1 (Mullaney & Skinner 1992, Pitetti et al. 2013). In fish, the androgen-producing LCs also express the fshr gene and Fsh is a potent steroidogenic hormone (Ohta et al. 2007, García-López et al. 2010). Similar to other vertebrates, androgens stimulate spermatogenesis in fish (Miura et al. 1991, Leal et al. 2009). However, zebrafish Fsh also stimulates spermatogonial differentiation in an androgen-independent manner, for example by stimulating the release of Igf3 mainly from SCs (Nóbrega et al. 2015) and of Insl3 from LCs (Assis et al. 2016, Crespo et al. 2016). Moreover, Fsh suppresses the production of the SC-derived anti-Müllerian hormone (Amh) (Skaar et al. 2011), which would otherwise exert inhibitory effects on spermatogenesis (Morais et al. 2017).
Crespo et al. (2016) identified ~200 testicular transcripts responding to zebrafish Fsh in an androgen-independent manner, including genes belonging to the Wnt signaling pathway. Similar results were reported for rainbow trout testis tissue (Sambroni et al. 2013). The Wnt signaling system is a conserved cell-to-cell communication system that consists of canonical and non-canonical pathways. The former is also referred to as β-catenin-dependent pathway and relies on inhibiting the β-catenin destruction complex (Salic et al. 2000), thereby increasing the levels of cytoplasmic β-catenin (Nusse 1999). After its translocation into the nucleus, β-catenin interacts with transcription factors of the T cell factor/lymphoid enhancer factor (TCF/LEF) family, inducing the expression of Wnt reporter genes such as cyclinD1 and axin2 (Tetsu & McCormick 1999, Shtutman et al. 1999, Jho et al. 2002, Van Amerongen et al. 2012, Takase & Nusse 2016).
The β-catenin-dependent pathway modulates spermatogenesis in mammals (Yeh et al. 2012, Kerr et al. 2014, Takase & Nusse 2016, Tokue et al. 2017), but studying mice carrying β-catenin mutations provided inconsistent results. Kerr et al. (2014) reported that loss of β-catenin in all germ cells depleted spermatocytes and spermatids without affecting the number of PLZF-positive (a marker for undifferentiated spermatogonia) germ cells. In contrast, when β-catenin knockout was mediated by Stra8-Cre (expressed by differentiating spermatogonia), mutant mice showed normal spermatogenesis (Rivas et al. 2014). Still, a later report described that the canonical WNT reporter gene Axin2 is expressed specifically in undifferentiated spermatogonia and that β-catenin knockout in Axin2-positive spermatogonia reduced proliferation activity and number of PLZF-positive spermatogonia in mice, resulting in the loss of differentiating spermatogonia (Takase & Nusse 2016). Moreover, inhibiting WNT secretion by knocking out Gpr177 (the mouse orthologue of Drosophila wls) that codes for a WNT ligand chaperone critical for exocytosis, either in germ cells or in SCs, had no or limited effects on spermatogenesis (Chen et al. 2016), challenging previous conclusions.
In the present study, we investigated if canonical Wnt signaling participates in mediating stimulatory effects of Fsh on zebrafish spermatogenesis, using a primary testis tissue culture system. We report for the first time in a vertebrate that Fsh recruits the β-catenin signaling system via Igf signaling to regulate germ cell development.
Materials and methods
Adult WT male zebrafish (AB strain) and Tg(7xTCF-Xla.Siam:GFP)ia4 male zebrafish (Moro et al. 2012) between 4 and 12 months of age were used in this study. All experiments followed the Dutch National regulations for animal care. The experimental protocols were approved by the Utrecht University Experimental Animal Committee (2013.III.06.045 and NVWA 10800).
To study the role of the canonical Wnt signaling system, adult zebrafish testes were dissected for tissue culture experiments using a previously described system (Leal et al. 2009), in which one testis was incubated under control conditions and the other testis under experimental conditions.
We have reported previously that 100 ng/mL Fsh increased the proliferation of both Aund and Adiff spermatogonia. Since Fsh also elevated androgen production (García-Lopez et al. 2010) and the production of the growth factors Insl3 and Igf3 ( Nóbrega et al. 2015, Assis et al. 2016, Crespo et al. 2016), we studied if androgen, Insl3 or Igf3 use the β-catenin-dependent pathway to stimulate spermatogenesis. To this end, zebrafish testes were incubated under basal conditions (Leal et al. 2009) or in the presence of 11-ketotestosterone (11-KT; Sigma; 200 nM in 0.01% ethanol; García-López et al. 2010) of Insl3 (100 ng/mL; Crespo et al. 2016) or of Igf3 (100 ng/mL; Nóbrega et al. 2015) for 4 or 5 days.
In a parallel experiment, zebrafish testes were incubated for 5 days in the presence of Fsh (100 ng/mL) with or without NVP-AEW541 (10 µM in medium containing 0.1% DMSO; Selleckchem; Morais et al. 2013), an Igf1 receptor inhibitor. The production of biologically active steroids was blocked by including trilostane (25 µg/mL in medium containing 0.1% dimethyl sulfoxide; Chemos), an inhibitor of 3β-hydroxysteroid dehydrogenase activity, in all experiments with Fsh, a potent steroidogenic hormone in zebrafish (Garcia-Lopez et al. 2010).
To study if Igf3 directly stimulates β-catenin signaling, zebrafish testes were incubated for 5 days under basal conditions or in the presence of Igf3 (100 ng/mL) or in the presence of Igf3 (100 ng/mL) with or without 50 µM IWP-12 (Sigma-Aldrich) in medium containing 0.05% DMSO. IWP-12 inhibits an enzyme required to release Wnt ligands (porcupine) in zebrafish (Chen et al. 2009, Dodge et al. 2012). Since no statistical differences were found as regards BrdU incorporation (P = 0.718 for Aund and P = 0.951 for Adiff; unpaired Student’s t-test) and proportion of area (P = 0.563 for Aund and P = 0.595 for Adiff; unpaired Student’s t-test) after incubation with Igf3 in both experiments, we treated them as a single group to compare three different conditions (basal, Igf3 and Igf3 with IWP-12).
To study Igf3 effects on β-catenin signaling, zebrafish testes were incubated for 5 days under basal conditions (Leal et al. 2009) or in the presence of Igf3 (100 ng/mL). In parallel experiment, testes were incubated in the presence of Igf3 (100 ng/mL) with or without 10 µM XAV939 (Sigma-Aldrich) (Shimizu et al. 2012) or 12 µM IWR-1 (Sigma-Aldrich) (Shimizu et al. 2012, Wehner et al. 2014) in medium containing 0.01% or 0.012% DMSO, respectively. XAV939 and IWR-1 are compounds inhibiting β-catenin signaling by increasing the stability of Axin, a key protein of the β-catenin destruction complex, thereby decreasing the levels of cytoplasmatic β-catenin (Huang et al. 2009, Shimizu et al. 2012). Since no statistical differences were found as regards BrdU incorporation and proportion of area (one-way ANOVA, followed by Tukey’s post hoc test) occupied by type A spermatogonia after incubation with Igf3 in the three experiments, we treated them as a single group to compare four different conditions (basal, Igf3 and Igf3 with XAV939 or IWR-1).
To further characterize the effect of XAV939 on Igf3-stimulated spermatogenesis, testes were incubated in the presence of Igf3 (100 ng/mL) alone or in the additional presence of 10 µM XAV939 for 5 days to analyze the expression of selected genes.
Igf1 receptor activation mainly triggers the MAPK pathway and the PI3K pathway (LeRoith 2008). To study which mechanism Igf3 uses to stimulate β-catenin signaling, we examined if a MAPK inhibitor (50 µM PD98059 (InvivoGen) in medium containing 0.05% DMSO; Nóbrega et al. 2015) or a PI3K inhibitor (20 µM LY294002 (Sigma-Aldrich) in medium containing 0.02% DMSO; Pozios et al. 2001) modulated Igf3-induced changes in spermatogenesis or in the transcript levels of two β-catenin-sensitive genes or four germ cell markers, following an incubation period of 5 days.
In case of the 5-day incubation experiment, the medium was changed once after 3 days of incubation. Testis tissue derived from the experiment described above was fixed for morphological analysis or snap-frozen in liquid nitrogen at the end of the incubation period and stored at −80°C until RNA extraction.
The relative transcript levels of germ cell-marker genes and other genes of interest (Table 1) were analyzed by real-time, quantitative polymerase chain reaction (qPCR) assays.
Primers used for gene expression studies (Fw = forward; Rv = reverse) and to generate DNA templates for DIG-labeled cRNA probe syntheses for in situ hybridization.
|Target genes||Primers name||Sequence (5′-3′)||Gene information|
|elf1a||2476 (Fw)||GCCGTCCCACCGACAAG||Reference gene (Morais et al. 2013)|
|cyclinD1||5231 (Fw)||CAAGCAGATACTGCGCAAACA||This paper|
|axin2||4615 (Fw)||ACCCTCGGACACTTCAAGGAA||This paper|
|foxa2||5741 (Fw)||GTCAAAATGGAGGGACACGAAC||Potential marker for type A undifferentiated spermatogonia (Safian et al. 2017)|
|nanos2||4817 (Fw)||AAACGGAGAGACTGCGCAGAT||Expressed in type Aund spermatogonia (Beer et al. 2013, Bellaiche et al. 2014)|
|piwil1||2542 (Fw)||GATACCGCTGCTGGAAAAAGG||Expressed in all generations of type A spermatogonia (Houwing et al. 2008)|
|piwil2||2994 (Fw)||TGATACCAGCAAGAAGAGCAGATCT||Expressed in all germ cell type except type Aund and spermatozoa (Houwing et al. 2008)|
|dazl||3104 (Fw)||AGTGCAGACTTTGCTAACCCTTATGTA||Expressed by B spermatogonia and primary spermatocytes (Chen et al. 2013)|
|4984 (Rv)b||T7Rpp-TGGATTAACCAGTAACATGAGCAATCACTCA||Igf1 receptor a (This paper)|
|4985 (Rv)b||T7Rpp-CAGATGCAAGATTTCTCCTTCACCACAGATGA||Igf1 receptor b (This paper)|
aPrimer 4982 and 4983 contain the T3 RNA polymerase promoter sequence (underlined) at its 5′-end (T3Rpps; 5′-GGGCGGGTGTTATTAACCCTCACTAAAGGG-3′); bPrimer 4984 and 4985 contain the T7 RNA polymerase promoter sequence (underlined) at its 5′-end (T7Rpps; 5′-CCGGGGGGTGTAATACGACTCACTATAGGG-3′).
Total RNA was isolated from the tissue using an RNAqueous Micro kit (Ambion), according to the manufacturer’s protocol. cDNA synthesis from total RNA and quantification of transcript levels were carried out as described previously (Bogerd et al. 2001). In brief, 2 µg of total RNA were reverse transcribed using 250 U of Superscript II RNase¯ reverse transcriptase (Life Technologies). qPCRs were performed in SYBR Green assay mix (Applied Biosystems), specific qPCR primers (900 nM each) and 5 µL cDNA in a total volume of 20 µL. The quantification cycle (Cq) values were determined in a ViiA7 Real-Time PCR System (Applied Biosystems) using default settings. The relative amounts of mRNA in the cDNA samples were calculated using the arithmetic comparative method (ΔΔCt method; Livak & Schmittgen 2001), as described in Bogerd et al. (2001). Transcript levels of the elongation factor 1a (ef1a) were stable (inset in Fig. 1A and B, 3I, and 4C and D) and therefore, ef1a was used to normalize gene expression. All results were expressed as fold induction with respect to the control group.
Quantification of spermatogonial proliferation and of proportions of section surface areas
To quantify the proliferation activity of Aund and Adiff spermatogonia, 100 μg/mL of the proliferation marker 5-bromo-2′-deoxyuridine (BrdU; Sigma-Aldrich) was added to the tissue culture medium during the last 6 h of the incubation period. After fixation in methacarn (60% (v/v) absolute ethanol, 30% chloroform and 10% acetic acid), the samples were dehydrated in graded ethanol (70, 96 and 100%), embedded in Technovit 7100 (Heraeus Kulzer) and sectioned at a thickness of 4 μm. To determine the proliferation activity, one set of sections was used to localize BrdU as described previously (Leal et al. 2009). The BrdU labeling index was determined by analyzing 100 cysts (Adiff spermatogonia) or 100 Aund cells, discriminating between BrdU-positive and -negative cysts and cells, respectively. In addition, the number of BrdU-positive ‘free’ SCs (apparently not associated with germ cells), SCs contacting BrdU-positive and -negative Aund, and SCs contacting BrdU-positive and -negative Adiff was determined using ten randomly chosen, non-overlapping fields at ×400 magnification.
To quantify the proportion of section area occupied by type A spermatogonia, another set of sections was stained with toluidine blue and 10 randomly chosen, non-overlapping fields were photographed at ×400 magnification with a digital camera. The images were analyzed quantitatively based on the number of points counted over the germ cell types investigated (Aund and Adiff), using the ImageJ freeware (National Institute of Health, Bethesda, MD, USA, http://rsbweb.nih.gov/ij) with a 540-point grid, i.e. 5400 data points were analyzed per individual.
Immunofluorescence on paraffin sections
To detect β-catenin-dependent pathway activation, testes from a transgenic zebrafish line (Tg(7xTCF-Xla.Siam:GFP)ia4) expressing gfp upon β-catenin activation were used (Moro et al. 2012). Testes from this line were incubated in basal medium or in the presence of Igf3 (100 ng/mL), IWP-12 (50 µM) or Igf3 (100 ng/mL) in combination with XAV939 (10 µM) or IWP-12 (50 µM). After fixation in phosphate-buffered 4% paraformaldehyde at 4°C overnight, the tissue was dehydrated and embedded in paraplast (Sigma-Aldrich). Sections of 5 µm thickness were dewaxed and rehydrated, according to conventional techniques. After antigen retrieval (10 mM sodium citrate, 0.05% Tween 20, pH 6.0; at 98°C for 10 min) non-specific antibody binding was blocked by pre-incubating slides with 5% normal goat serum (Sigma-Aldrich) and 1% acetylated bovine serum albumin (BSA; Sigma-Aldrich) at room temperature for 30 min. Then, sections were incubated with a primary antibody raised in rabbit against GFP (Nóbrega et al. 2015; 1 µg/mL; TP401, Torrey Pines Biolabs Inc.) in PBS containing 1% BSA at 4°C overnight. After PBS washes, sections were incubated with secondary goat anti-rabbit Alexa Fluor 488 (8 µg/mL, A11034, Sigma-Aldrich) containing 10% normal horse serum (Sigma-Aldrich) at room temperature for 1 h and 30 min. After PBS washes, the sections were incubated for 3 min with propidium iodide (PI; 1 µg/mL, Sigma-Aldrich). Sections were mounted in Vectashield (H-1000, Vector) and images were taken using a confocal laser scanning microscope (LSM 700, Zeiss). Images were acquired and processed using the same settings (pinhole size: 1 AU, same laser power: 12, and gain: 550 mV for Alexa 488 and 400 mV for PI) in all experiments analyzed.
Gfp quantification by Western blot
To detect Wnt ligand release inhibition in response to IWP-12, testes from a transgenic zebrafish line (Tg(7xTCF-Xla.Siam:GFP)ia4) were incubated under basal conditions or in the presence of IWP-12 (50 µM) for 3 days. After the incubation period, tissue was processed for Western blot analysis to detect Gfp protein production. Proteins were extracted according to Skaar et al. (2011) and protein concentrations were determined in a Bradford protein assay (BIO-RAD). Whole testis protein (30 µg) was resolved in 10% SDS-PAGE and Western blot assays were performed according to Fuentes et al. (2013) with minor modifications. Briefly, proteins were transferred to polyvinylidene difluoride membranes (Millipore) and blocked for 1 h at room temperature in 5% skimmed milk powder (Millipore) in PBS. Primary GFP antibody incubation was carried out overnight at 4°C (2 µg/mL; TP401, Torrey Pines Biolabs Inc.). After PBS washes, secondary horseradish peroxidase-conjugated goat anti-rabbit (0.2 µg/mL; 111-035-003, Jackson ImmunoResearch Laboratories Inc.) was incubated for 1 h at room temperature. Protein detection was performed using Pierce ECL Plus Substrate (Thermo Scientific) according to the manufacturer’s instructions. Films were scanned and densitometric analysis of the bands was performed with ImageJ freeware. Gfp analysis was normalized to Coomassie blue staining of total protein.
In situ hybridization on zebrafish testis cryosections
To identify the cell types responding to Igf3, WT zebrafish testes were used to study the expression of igf1 receptor a (igf1ra) and b (igf1rb) by in situ hybridization on cryosections as described previously (Assis et al. 2016). For the production of zebrafish igf1ra and igf1rb DIG-labeled riboprobes (see primers in Table 1), we followed a procedure as reported by Good-Ávila et al. (2009).
Statistical analyses were carried out using the GraphPad Prism 5 software package. Since our tissue culture system compares in most experiments the two testes of a given fish incubated under control vs experimental conditions, we applied Student’s t-test for paired observation to estimate the statistical significance of fold changes between treated and control conditions. These data are presented as fold of basal (mean ± standard error of the mean (s.e.m.)). In Figs 2G and H and 3A and B, three and four conditions respectively were compared, so that fold-changes cannot be calculated. In these cases, the BrdU indices (Figs 2G and 3A) or proportion of area values (Figs 2H and 3B) were processed statistically using one-way ANOVA, followed by Tukey’s post hoc test.
The Fsh-modulated Igf3 stimulates the activation of β-catenin signaling in type A spermatogonia
We have found that Fsh modulated the expression of Wnt-related genes (Crespo et al. 2016), opening the possibility that one or more downstream mediators of Fsh action stimulate the canonical, β-catenin-dependent pathway. To further investigate this possibility, we examined the effect of three of these mediators, 11-KT (Garcia-Lopez et al. 2010), Insl3 (Crespo et al. 2016) and Igf3 (Nóbrega et al. 2015), on the transcript levels of cyclinD1 and axin2, two genes known to be sensitive to β-catenin signaling also in zebrafish (Neal et al. 2013, Wehner et al. 2014). In response to Igf3 incubation the transcript levels of both genes increased, while 11-KT and Insl3 were unable to do so (Fig. 1A). Next, testes were incubated in the presence or absence of Fsh or in the presence of Fsh with or without an Igf receptor inhibitor (NVP-AEW541). The transcript levels of cyclinD1 did not change, while axin2 transcript increased and decreased in response to Fsh or to Fsh and NVP-AEW541, respectively (Fig. 1B). These experiments suggest that Fsh-stimulated Igf3 release activates β-catenin signaling.
To further study the effect of Igf3 on the β-catenin-dependent pathway, testes from Tg(7xTCF-Xla.Siam:GFP)ia4 zebrafish were incubated under basal conditions or in the presence of Igf3. A weak signal for Gfp protein was detectable in the absence of Igf3 (Fig. 1C, D and E) but was strongly induced in type A spermatogonia in response to Igf3 (Fig. 1F, G and H). Importantly, an inhibitor of Wnt ligand release (IWP-12) did not modulate the Igf3-mediated induction of Gfp (Fig. 2A, B, C, D, E and F), of the proliferation activity or of the proportion of area occupied by type A spermatogonia (Fig. 2G and H). These experiments show that Igf3-stimulated, β-catenin-mediated spermatogonial proliferation and differentiation was independent of Wnt release. The capacity of IWP-12 to block the release of Wnt ligands also in zebrafish testis tissue was demonstrated by the IWP-12-induced decrease in Gfp protein levels in tissue from Tg(7xTCF-Xla.Siam:GFP)ia4 zebrafish (Fig. 2I and J).
Igf3 uses the β-catenin-dependent pathway to promote differentiation
Previously, we reported that Igf3 increased the proliferation of both types of A spermatogonia and decreased the proportion of Aund while increasing the one for Adiff spermatogonia (Nóbrega et al. 2015, Safian et al. 2017), i.e. Igf3 promoted the differentiating proliferation of type A spermatogonia. Here, we examined directly the effect of XAV939 and IWR-1, inhibitors of the β-catenin-dependent pathway, on Igf3-stimulated differentiation of type A spermatogonia. XAV939 and IWR-1 reduced Igf3-stimulated proliferation of both type A spermatogonia to basal levels (Fig. 3A). Regarding the proportion of area, XAV939 and IWR-1 blocked the pro-differentiation effect of Igf3 on type Adiff spermatogonia (Fig. 3B). This suggests that the Igf3-induced transition of Aund to Adiff was compromised by both inhibitors, associated with a certain accumulation of Aund spermatogonia. Immunocytochemical studies using testes of Tg(7xTCF-Xla.Siam:GFP)ia4 zebrafish showed that Igf3-induced, β-catenin-mediated Gfp protein expression in type A spermatogonia was blocked by XAV939 (Fig. 3C, D, E, F, G and H). Moreover, transcript quantification (yielding reduced cyclinD1 and axin2 levels) confirmed inhibition of the β-catenin pathway. Increased transcript levels of type A spermatogonia-marker genes (foxa2 (Safian et al. 2017) and nanos2 (Beer et al. 2013, Bellaiche et al. 2014) and decreased transcript levels of marker genes for more differentiated germ cells (piwil2 (Houwing et al. 2008) and dazl (Chen et al. 2013) (Fig. 3I) are in line with the morphological data (Fig. 3B).
Cellular localization of Igf1 receptors and downstream pathways used by Igf3 to activate β-catenin-dependent pathway
To better understand the mechanism used by Igf3 to stimulate β-catenin-dependent processes, we studied the cellular localization of igf1 receptors a and b by in situ hybridization. Moreover, we tested the effects of a MAPK (PD98059) and of a PI3K (LY294002) inhibitor on the expression of β-catenin-sensitive and germ cell marker genes.
In situ hybridization revealed that igf1ra was expressed by SCs contacting all germ cell generations from spermatogonia to spermatids, while igf1rb was expressed by both SCs and germ cells (Fig. 4A and B).
The transcript levels of cyclinD1, but not of axin2, were down regulated by the MAPK inhibitor in Igf3-stimulated testis tissue (Fig. 4C). Concerning germ cell-marker gene expression, the mRNA levels of piwil1 (expressed by all spermatogonia and primary spermatocytes; Houwing et al. 2008) and piwil2 were slightly reduced (Fig. 4C). The PI3K inhibitor, on the other hand, significantly compromised Igf3-mediated increases of both cyclinD1 and axin2 transcript levels (Fig. 4D), increased transcript levels of foxa2 and nanos2 but reduced those of piwil2 (Fig. 4D). Overall, this indicates a shift towards more undifferentiated germ cells when interfering with Igf3 effects by adding MAPK or PI3K inhibitors. The morphological analysis pointed into the same direction (Fig. 4E, F, G and H). The proliferation activity of both Aund and Adiff spermatogonia decreased while an increased proportion of Aund and a reduced proportion of Adiff spermatogonia was recorded in the presence of Igf3 and the PI3K inhibitor. This suggests that in addition to a reduced proliferation of type A spermatogonia, also spermatogonial differentiation was compromised, resulting in an accumulation of type Aund spermatogonia and a partial depletion of Adiff spermatogonia.
Effect of Igf3 on Sertoli cell proliferation
Since SC proliferation is required for fertility in the adult fish testis (De França et al. 2015), we have studied the number of BrdU-positive SCs that were free (i.e. apparently not in contact with germ cells) (Fig. 5A) and that contacted Aund (Fig. 5B and C) or Adiff spermatogonia (Fig. 5D). This was examined under basal conditions or in the presence of Igf3 (100 ng/mL), with or without the inhibitor of PI3K (LY294002, 20 µM) or the β-catenin signaling inhibitor XAV939 (10 µM). Igf3 reduced the number of proliferating SCs contacting BrdU-negative Aund, while it increased the number of proliferating SC contacting BrdU-positive Adiff (Fig. 5E). In the additional presence of the β-catenin signaling inhibitor XAV939, the number of proliferating free SC and SCs contacting BrdU-negative Aund increased while decreasing the proliferation of SCs contacting BrdU-positive Aund (Fig. 5F). Similarly, the proliferation of free SC and SCs contacting BrdU-negative Aund increased in the presence of the PI3K inhibitor (Fig. 5G).
Fsh stimulates spermatogonial differentiation via androgen production in LCs and also by stimulating the transcript levels and release of Igf3, but not those of igf1, igf2a and b (Nóbrega et al. 2015; earlier sometimes referred to as Igf1b) mainly from SCs and of insulin-like peptide 3 (Insl3) from LCs (Assis et al. 2016, Crespo et al. 2016). Since a previous report indicated that Fsh modulated the transcript levels of Wnt pathway-associated genes but did not regulate the expression of Wnt ligands (Crespo et al. 2016), we hypothesized that the β-catenin pathway becomes activated by one or more downstream mediators of Fsh action, potentially in a manner independent of Wnt ligands. Our results show that Fsh-stimulated Igf3 release most likely activated β-catenin signaling in germ cells via igf1rb. This occurred in a manner involving the PI3K pathway but did not depend on the release of Wnt ligands. Activation of β-catenin signaling in type A spermatogonia promoted their differentiating proliferation.
Previous studies have suggested that crosstalk occurs between Igf and the β-catenin pathway in mammals (Playford et al. 2000, Desbois-Mounthon et al. 2001, Ye et al. 2010, Wang et al. 2010, Hu et al. 2012), but somewhat inconsistent results have been found. Desbois-Mounthon et al. (2001), using human liver-derived HepG2 cells, and Wang et al. (2010), using murine chondrocytes, proposed that IGF1 uses the PI3K/AKT pathway to inhibit glycogen synthase kinase 3β (GSK3β), an important subunit of the β-catenin destruction complex. Ng et al. (2009), on the other hand, proposed that core proteins of the β-catenin destruction complex (in particular axin) protect GSK3β from the inhibitory effect of IGF/PI3K/AKT pathway and that Wnt-mediated transcriptional activity was not modulated by IGF/PI3K/AKT in prostate cancer (PC3) and breast cancer (BT549) cell lines. These inconsistencies may be related to the different experimental approaches and cell lines used. To our knowledge, the present study is the first to propose that a growth factor not belonging to the Wnt family activates the β-catenin-dependent pathway in the testis. Our data also support the view of a stimulatory role of PI3K in this activation.
The fact that Igf3 does not require Wnt ligands to activate β-catenin signaling is a feature not addressed previously in mammals. Porcupine and Wntless (GPR177 in mice) are proteins required for the intracellular transport and release of Wnt ligands (Herr et al. 2012). Surprisingly, loss of Gpr177 either in germ cells or SCs is compatible with normal spermatogenesis in mice and only after 8 months of age, germ cell-specific Gpr177 knockout mice showed smaller testis and reduced fertility (Chen et al. 2016). However, the latter study did not examine the state of activity of the β-catenin-dependent pathway in Gpr177 knockout mice. We found previously that Igf3 promoted the proliferation of both type A spermatogonia and stimulated the accumulation of Adiff while reducing the proportion of Aund (Nóbrega et al. 2015, Safian et al. 2017). Here, we report that blocking the β-catenin-dependent pathway inhibits in particular the pro-differentiation effects of Igf3 in zebrafish spermatogonia. A possible explanation for the absence of a clear and immediate effect of the loss of Wnt ligand release on spermatogenesis in mice may be related to our finding that signaling molecules of other pathways, such as Igf ligands, can stimulate β-catenin-dependent signaling in a Wnt ligand-independent manner in germ cells.
As mentioned earlier, Igf3 activated β-catenin signaling in type A spermatogonia. However, information on the cellular localization of Igf receptors in the zebrafish testis was missing. In rainbow trout, igf1r expression was found in cell fractions enriched in SCs but also in spermatogonia and primary spermatocytes (Le Gac et al. 1996). We report here that both Igf receptors were expressed by SCs but that igf1rb was also expressed by zebrafish germ cells. This opens the possibility that Igf1rb mediates the Igf3 effects to activate the β-catenin-dependent pathway in spermatogonia. Our results also suggest that Igf3 uses PI3K, and not MAPK, to activate β-catenin-dependent signaling. Similarly, Wang et al. (2014) reported that IGF1 stimulated the proliferation of murine spermatogonial stem cells in culture via PI3K and not via MAPK signaling. The PI3K pathway has been implicated previously in modulating spermatogonial proliferation in mammals. Shen et al. (2014) reported that IGF1 activated PI3K and MAPK pathways to stimulate differentiation of germ cells in culture of mice testicular fragments. The present study is the first to report that PI3K and β-catenin pathways interact in testis tissue; however, further investigation will be require to elucidate how exactly the PI3K pathway is involved.
Type Aund spermatogonia are single germ cells enveloped by one or two SCs (Leal et al. 2009). This germ cell population also includes the SSCs in the zebrafish testis (Nóbrega et al. 2010). Self-renewing SSCs produce two independent Aund daughter cells, each of them associating with SCs that support their survival to form new spermatogenic cysts. Consequently, the production of new spermatogenic cysts also requires the availability of ‘free’ SCs. When going through a differentiating division, on the other hand, a single Aund cell produces two smaller Adiff spermatogonia that remain connected via a cytoplasmic bridge within their spermatogenic cyst formed by SCs enveloping the developing germ cell clone. When Adiff spermatogonia continue to go through differentiating mitotic cell cycles, germ cell number and hence volume increases, which also requires SC proliferation. Since for both, the formation of new cysts and the developmental growth of existing cysts, SC proliferation is required in the adult testis (De França et al. 2015), we have also studied the proliferation of Sertoli cells under the influence of Igf3 and the additional presence of β-catenin signaling or PI3K inhibitors. Igf3 reduced the number of BrdU-positive SC contacting BrdU-negative Aund, suggesting a reduced level of activity of the earliest stage of spermatogonial cysts. On the other hand, we recorded a clear increase in the number of BrdU-positive SC contacting BrdU-positive type Adiff spermatogonia. Since this Igf3-mediated stimulation of SC proliferation was independent of PI3K and β-catenin signaling, we suspect it was mediated by the Igf receptors localized on SCs. Interestingly, this Igf3-mediated increase of SC proliferation was observed only in cysts where also the Adiff spermatogonia had incorporated BrdU, suggesting that the effect potentially involves germ-Sertoli cell communication to coordinate SC and germ cell proliferation. Overall, Igf3 seems to support the further growth of existing spermatogenic cysts, i.e. facilitates differentiation.
In conclusion, we have shown that Fsh-triggered Igf3 release stimulates β-catenin-signaling in a manner involving PI3K signaling, and that this pathway operates independent of Wnt ligands but is required to promote differentiation of Aund spermatogonia in zebrafish testis (Fig. 6).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This work was supported by the Research Council of Norway (SALMOSTERILE project n°: 221648/O30), by the European Union Grant LIFECYCLE FP7-222719, and by a scholarship awarded to D Safian from CONICYT/Becas Chile.
The authors thank Henk van der Kant for technical assistance in sectioning and immunohistochemistry, and Dr Enrico Moro from the University of Padova, Italy for sharing embryos of the Tg(7xTCF-Xla.Siam:GFP)ia4 zebrafish line.
BogerdJBlomenröhrMAnderssonEvan der PuttenHHTensenCPVischerHFGrannemanJCJanssen-DommerholtCGoosHJSchulzRW 2001 Discrepancy between molecular structure and ligand selectivity of a testicular follicle-stimulating hormone receptor of the African catfish (Clarias gariepinus). Biology of Reproduction 64 1633–1643. (https://doi.org/10.1095/biolreprod64.6.1633)
ChenSXBogerdJSchoonenNEMartijnJde WaalPPSchulzRW 2013 A progestin (17α, 20β-dihydroxy-4-pregnen-3-one) stimulates early stages of spermatogenesis in zebrafish. General and Comparative Endocrinology 185 1–9. (https://doi.org/10.1016/j.ygcen.2013.01.005)
Desbois-MouthonCCadoretABlivet-Van EggelpoelMJBertrandFCherquiGPerretCCapeauJ 2001 Insulin and IGF-1 stimulate the [beta]-catenin pathway through two signalling cascades involving GSK-3 [beta] inhibition and Ras activation. Oncogene 20 252. (https://doi.org/10.1038/sj.onc.1204064)
DodgeMEMoonJTuladharRLuJJacobLSZhangLSShiHWangXMoroEMongeraA 2012 Diverse chemical scaffolds support direct inhibition of the membrane-bound O-acyltransferase porcupine. Journal of Biological Chemistry 287 23246–23254. (https://doi.org/10.1074/jbc.M112.372029)
FuentesENSafianDEinarsdottirIEValdésJAElorzaAAMolinaABjörnssonBT 2013 Nutritional status modulates plasma leptin, AMPK and TOR activation, and mitochondrial biogenesis: implications for cell metabolism and growth in skeletal muscle of the fine flounder. General and Comparative Endocrinology 186 172–180. (https://doi.org/10.1016/j.ygcen.2013.02.009)
García-LópezÁDe JongeHNóbregaRHDe WaalPPVan DijkWHemrikaWTarangerGLBogerdJSchulzRW 2010 Studies in zebrafish reveal unusual cellular expression patterns of gonadotropin receptor messenger ribonucleic acids in the testis and unexpected functional differentiation of the gonadotropins. Endocrinology 151 2349–2360. (https://doi.org/10.1210/en.2009-1227)
HuQLeeSYO’KuskyJRYeP 2012 Signalling through the type 1 insulin-like growth factor receptor (IGF1R) interacts with canonical Wnt signalling to promote neural proliferation in developing brain. American Society for Neurochemistry 4 AN20120009. (https://doi.org/10.1042/AN20120009)
JhoEHZhangTDomonCJooCKFreundJNCostantiniF 2002 Wnt/β-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Molecular and Cellular Biology 22 1172–1183. (https://doi.org/10.1128/MCB.22.4.1172-1183.2002)
Le GacFLoirMLe BailPYOllitraultM 1996 Insulin-like growth factor (IGF-I) mRNA and IGF-I receptor in trout testis and in isolated spermatogenic and Sertoli cells. Molecular Reproduction and Development 44 23–35. (https://doi.org/10.1002/(SICI)1098-2795(199605)44:1<23::AID-MRD3>3.0.CO;2-V)
MoraisRDVSNóbregaRHGomez-GonzalezNESchmidtRBogerdJFrancaLRSchulzRW 2013 Thyroid hormone stimulates the proliferation of Sertoli cells and single type A spermatogonia in adult zebrafish (Danio rerio) testis. Endocrinology 154 4365–4376. (https://doi.org/10.1210/en.2013-1308)
MoraisRDVSCrespoDNóbregaRHLemosMSVan de KantHJGDe FrançaLRMaleRBogerdJSchulzRW 2017 Antagonistic regulation of spermatogonial differentiation in zebrafish (Danio rerio) by Igf3 and Amh. Molecular and Cellular Endocrinology 454 112–124. (https://doi.org/10.1016/j.mce.2017.06.017)
MoroEOzhan-KizilGMongeraABeisDWierzbickiCYoungRMBourneleDDomenichiniAValdiviaLLumL 2012 In vivo Wnt signaling tracing through a transgenic biosensor fish reveals novel activity domains. Developmental Biology 366 327–340. (https://doi.org/10.1016/j.ydbio.2012.03.023)
MullaneyBPSkinnerMK 1992 Basic fibroblast growth factor (bFGF) gene expression and protein production during pubertal development of the seminiferous tubule: follicle-stimulating hormone-induced Sertoli cell bFGF expression. Endocrinology 131 2928–2934. (https://doi.org/10.1210/endo.131.6.1446630)
PitettiJLCalvelPZimmermannCConneBPapaioannouMDAubryFCederrothCRUrnerFFumelBCrausazM 2013 An essential role for insulin and IGF1 receptors in regulating Sertoli cell proliferation, testis size, and FSH action in mice. Molecular Endocrinology 27 814–827. (https://doi.org/10.1210/me.2012-1258)
SafianDVan der KantHJGCrespoDBogerdJSchulzRW 2017 Follicle-stimulating hormone regulates igfbp gene expression directly or via downstream effectors to modulate Igf3 effects on zebrafish spermatogenesis. Frontier in Endocrinology 8 328. (https://doi.org/10.3389/fendo.2017.00328)
SkaarKSNóbregaRHMagarakiAOlsenLCSchulzRWMaleR 2011 Proteolytically activated, recombinant anti-müllerian hormone inhibits androgen secretion, proliferation, and differentiation of spermatogonia in adult zebrafish testis organ cultures. Endocrinology 152 3527–3540. (https://doi.org/10.1210/en.2010-1469)
TokueMIkamiKMizunoSTakagiCMiyagiATakadaRNodaCKitadateYHaraKMizuguchiH 2017 SHISA6 confers resistance to differentiation-promoting Wnt/β-catenin signaling in mouse spermatogenic stem cells. Stem Cell Reports 8 561–575. (https://doi.org/10.1016/j.stemcr.2017.01.006)
WehnerDCizelskyWVasudevaroMDÖzhanGHaaseCKagermeier-SchenkBRoderADorskyRIMoroEArgentonF 2014 Wnt/β-catenin signaling defines organizing centers that orchestrate growth and differentiation of the regenerating zebrafish caudal fin. Cell Reports 6 467–481. (https://doi.org/10.1016/j.celrep.2013.12.036)