Introduction

The potential economic benefits of transgenic technology to aquaculture are enormous, following the introduction of novel desirable traits to farmed fishes (Chen et al. 1996). With the expansion of the global population and overfishing, advanced aquaculture techniques are needed to meet the increasing demand for fish protein. Transgenic technology offers the opportunity to improve both the quantity and quality of conventional fish strains currently exploited in aquaculture (Fu et al. 2005). It is also an important methodology for studying the function of genes and genomes in model animals (Kikuta and Kawakami 2009). Successful production of transgenic fish was first demonstrated in goldfish (Zhu et al. 1985) and 3 years later in zebrafish (Stuart et al. 1988). More than 30 fish species, including many of the major aquaculture species like carp, tilapia, catfish and salmonids, have been genetically engineered with most efforts targeted to enhancing growth and feed conversion efficiency through the transfer of growth hormone (GH) gene constructs (Zhu and Sun 2000; Wu et al. 2003; Devlin et al. 2006). GH transgenesis has shown to result in different effects of growth enhancement in host fish depending on different genetic backgrounds (Devlin et al. 2001, 2009). For instance, one report showed that GH transgenesis is effective to growth only in hemizygous but not in homozygous individuals of those transgenic zebrafish (Studzinski et al. 2009) and we have never obtained fast-growing transgenic zebrafish in our laboratory by GH transgenesis.

It is well known that GH is the major regulator of postnatal growth and metabolism (Lichanska and Waters 2008) via GH receptor (GHR) signaling pathways (Rowland et al. 2005). The GHR is a type I cytokine receptor consisting of extracellular, transmembrane and intracellular domains. GH activates the GHR by realigning two identical receptor subunits in a constitutive dimer through binding with their extracellular domains, leading to the activation of JAK2 kinase associated with the intracellular domain of GHR (Herrington and Carter-Su 2001; Waters et al. 2006). STATs 5a, and 5b bind to particular phospho-tyrosines in the cytoplasmic domain of the GHR following their phosphorylation by JAK2, are themselves tyrosine phosphorylated by JAK2, dimerize, and translocate to the nucleus to activate igf1 transcription in particular. JAK2 also directly activates STAT1 and 3 by tyrosine phosphorylation, and again, these STATs translocate to the nucleus, bind to particular STAT responsive elements and together with other trans-factors, activate the transcription of target genes such c-fos (Cesena et al. 2007; Ihle and Gilliland 2007). Intriguingly, in our previous study, GH-independent activation of GHR can be achieved in cell culture by fusing the transmembrane and intracellular domains of GHR to the leucine zippers to achieve an active dimer orientation. This stabilizes the receptor dimer in a conformation that holds the box 1 sequences in proximity, facilitating signaling (Behncken et al. 2000). Here we have utilized the leucine zipper constructs to demonstrate for the first time enhanced growth by a molecularly designed constitutively activated GHR (CA-GHR) in zebrafish, which we have found is normally resistant to GH transgene-mediated growth.

In previous studies, the expression profiles of GH and GHR have been analyzed either during development or in different tissues of zebrafish (Zhu et al. 2007; Di Prinzio et al. 2010). However, a comparative expression analysis of GH, GHR and the signaling targets, c-fos and igf1, in zebrafish, has not been described. In our present study, we utilized zebrafish to study the comparative expression of GH signaling factors during embryonic development and in different tissues of adults. By setting up an in vivo model to evaluate the GH-signal activation (GHSA) levels in zebrafish embryos, furthermore, we were able to perform functional analysis of GH signaling during zebrafish early development by overexpression of GH or GHR.

Materials and methods

Fish maintenance and embryo injection

Fish were cultured in the fish culture facility, Institute of Hydrobiology, according to the Zebrafish Book (Westerfield 2000). Embryos were obtained from artificial fertilization and microinjected with indicated reagents with a pressure microinjector as described (Liu et al. 2005). mRNA was injected in the yolk below the first developing cell and DNA was injected inside the cytoplasm at 1-cell stage.

Constructs design and mRNA synthesis

The CA-GHR construct was described previously (Behncken et al. 2000). As shown in Fig. 1a, the transgene codes for a porcine GHR signal peptide (amino acids 1-27), the mouse c-jun leucine zipper (amino acids 277–315) fused upstream of the porcine GHR transmembrane and cytoplasmic domains (amino acids 251–638), cloned into the pcDNA3.1(+) vector. The transgene expression cassette is driven by a CMV promoter and utilizes a bovine GH polyA (bGH pA) (Fig. 1c) and the construct pGHR-cJun-pcDNA3.1 is hereafter termed as Jun-GHR. cDNA of 3 dpf (day-post-fertilization) zebrafish was used to amplify the full length cDNA and the sequence coding the extracellular and transmembrane domains (dominant negative C-terminal truncated protein, ΔC-GHR, Fig. 1b) of GHRa via high-fidelity PCR by using Kod-plus enzyme (Toyobo, Japan) and appropriate primers (Table 1). The primers were designed according to zebrafish GHRa cDNA sequence available in GenBank (NM_001083578). Common carp GH (cGH) cDNA was amplified from cDNA pool of common carp pituitary with primers cGH_P1 and cGH_P2 (Table 1). For overexpression of cGH and zGHR, the PCR products were cloned into pCS2+ expression vector after double digestion of the PCR products. For mRNA synthesis, ΔC-GHR-pCS2+ construct was linearized with KpnI and capped mRNA was synthesized using the Message Machine kit (Ambion Inc., USA) as previously described (Chen et al. 2009). All the DNA constructs have the same constitutive CMV promoter.

Fig. 1
figure 1

Schematic description of leucine zipper transgene chimera Jun-GHR (a) and C-terminal truncated zebrafish GHR (b) and the plasmid pGHR-cJun-pcDNA3.1(+) (c). Numbers in parentheses refer to the amino acid sequence from which the respective cDNA segments were taken. SP signal peptide, pGHR porcine GHR, zGHR zebrafish GHR, ECD extracellular domain, TMD transmembrane domain, BGH pA bovine GH polyA

Full size image
Table 1 Primers used in the study
Full size table

RT-PCR (reverse-transcription PCR) analysis of GHR signaling related genes

Total RNA of 50 zebrafish embryos, pooled tissues from 3 wildtype individuals or tissues from one transgenic individual was extracted using TRIzol® reagent (Invitrogen, USA) according to the manufacturer’s instructions. All RNA samples were treated with RNase-Free DNase (Promega, USA) according to manufacturer’s protocol. cDNA was synthesized by reverse transcription (RT) from 1 μg of total RNA using the ReverTra Ace enzyme (TOYOBO, Japan), dNTPs and Oligo(dT)20 RT primer (TOYOBO, Japan). The RT reaction was performed at 42ºC for 60 min followed by 95ºC for 5 min. PCR was performed to study the expression of different genes, such as GH, GHR, c-fos and igf1 in different samples. The primer sequences are shown in Table 1. Each pair of primer sequences locate in different exons of the respective gene, to avoid the amplification of genomic DNA potentially contaminated in the samples. PCR was carried out in a 25 μl reaction volume containing 2.5 μl of 10× PCR buffer, 0.2 μM of each primer, 0.2 mM of each dNTP, 0.75 mM of MgCl2, 0.5 unit of Taq DNA polymerase (Fermentas, Canada) and 50 ng of cDNA solution. The PCR reaction consists of a denaturation at 94°C for 5 min, 35 cycles of 30 s at 94°C, 30 s at 56°C and 45 s at 72°C, with a final elongation step of 10 min at 72°C. β-actin primer pairs (Table 1) were used to normalize the cDNA concentration of all samples. The products were run on 1% agarose gels stained with ethidium bromide (0.5 μg/ml), and amplified bands were visualized by ultraviolet transillumination and semi-quantified by Glyko Bandscan software (Novato, CA).

Real-time quantitative PCR analysis of c-fos and igf1

Fertilized zebrafish zygotes were microinjected with the DNA constructs Jun-GHR, zebrafish GHRa (zGHRa), carp GH (cGH), cGH and zGHRa at a concentration of 50 ng/μl for each construct. To illustrate the importance of functional GHR, cGH expression construct of 50 ng/μl and ΔC-GHR mRNA of 800 ng/μl were co-injected into zebrafish embryos. Each embryo was injected with 1 nl of DNA or RNA sample and each sample was injected with 300 embryos. cDNA from the manipulated zebrafish embryos at 2-cell stage, 1 and 3 dpf and from different tissues of adult transgenic zebrafish was synthesized as described above and analyzed with real-time quantitative PCR for c-fos and igf1 expression level. Real-time PCR was performed on an ABI PRISM® 7000 Sequence Detector (Applied Biosystems, Inc. USA) according to the manufacturer’s instructions. Reactions were performed in a 20 μl volume with 150 ng of cDNA, 0.2 μM primers and 10 μl of SYBR® Green Realtime PCR Master Mix (TOYOBO, Japan). The primer pairs were shown in Table 1. All reactions were run using the following conditions: 1 min denaturation at 95°C followed by 40 cycles of 95°C denaturation for 15 s, 55°C annealing for 15 s, and 72°C extension for 45 s. Detection of the fluorescent product was carried out at the end of the 72°C extension period. The data (Ct value), obtained from the software (7000 system SDS software), was transferred to Windows Excel (Microsoft co.) sheet and fold-change was calculated using 2−∆∆Ct method, with β-actin as the calibrator. To validate the stability of β-actin for real-time PCR normalization, several pairs of primers specific for PSB7, eTIF-2B, eEF-1A and GAPDH were designed as described in our previous study and used for GeNorm analysis (Vandesompele et al. 2002; Pei et al. 2007). Three samples were run for each analysis and all real-time PCR reactions were run in triplicates. The significance of the mean differences between various experimental groups was determined by one way ANOVA followed by Duncan’s multiple range test analyses. A P value <0.05 was considered statistically significant.

In vivo luciferase assay of spi2.1 promoter activity

Each zebrafish zygote was co-injected with spi2.1-luc (Jiao et al. 2006) of 50 pg and a constitutively expressed TK-Renilla luciferase construct (Promega) of 0.5 pg and the manipulated embryos were subsequently injected with indicated DNA or RNA samples. Embryos were allowed to develop until 3 dpf, then sets of 20 embryos were lysed in passive lysis buffer (Dual-Luciferase Reporter Assay System, Promega) and the luciferase activity were measured with a Berthold luminometer and relative luciferase activity was calculated as described (Liu et al. 2009). Each sample was analyzed in triplicate, and mean value and standard deviation were calculated. The significance of the mean differences between various experimental groups was determined by one way ANOVA followed by Duncan’s multiple range test analyses. A P value < 0.05 was considered statistically significant.

Generation, screening and growth trial of transgenic zebrafish

Fertilized zebrafish zygotes were microinjected with Jun-GHR construct of 50 ng/μl. Transgenic embryos and fish were raised according to standard procedure (Westerfield 2000). Transgene positive P0 fish were screened out by PCR assay of the total DNA of caudal fin. DNA samples were extracted from caudal fin samples or embryo samples by means of treatment with DNA extraction buffer and phenol/chloroform purification, followed by ethanol precipitation as described (Sun et al. 2005). PCR reactions were performed in a reaction mix of 25 μl, containing 50 ng of total DNA, 0.5 unit of Taq DNA polymerase (Fermentas, Canada), 1× Taq buffer, 0.5 μl of 10 mM dNTPs (Generay Biotech, China), 1 μl of each 10 μM primer, and a suitable amount of sterile deionized water. The PCR protocol used for transgene detection was as follows: 5 min denaturation at 95°C, 30 cycles of 30 s at 95°C, 30 s at 55°C and 45 s at 72°C and a final elongation step for 7 min at 72°C. Jun-GHR_P1 and Jun-GHR_ P2 (Table 1) were used as PCR primers. All PCR products were separated by 1% agarose gel electrophoresis and visualized by ultraviolet transillumination.

Each matured P0 transgenic zebrafish was mated with wildtype zebrafish and a pool of 10 F1 embryos from each cross was checked by PCR detection of the transgene. Each batch of embryos showing transgene positive was considered as an individual F1 transgenic batch. At 1 month-post-fertilization (mpf), tail fin DNA was extracted to check the presence of transgene in each F1 fish. 100 transgenic individuals from one batch of F1 embryos and 100 non-transgenics were raised in aquaria at the same condition. The body length and weight were measured from 20 individuals of transgenic and non-transgenic zebrafish after random selection from 1 to 4 mpf. Student’s t-test was applied for statistical comparison and the differences were considered significant at P < 0.01. All values in the text and figures refer to mean ± standard deviation (SD).

Results

Spatio-temporal expression of GH signaling components

RNA from various tissues was extracted and RT-PCR was used to analyze the transcription of GH, GHR, c-fos (Cesena et al. 2007) and igf1 (Wood et al. 2005), which represent the ligand, receptor and targets in the signaling, in different tissues as well as in a series of developmental stages of wildtype zebrafish (Fig. 2a, b).

Fig. 2
figure 2

The transcription analysis of GH, GHR, c-fos and igf1 in wildtype zebrafish. a RT-PCR analysis in different tissues of adult zebrafish. 1 brain, 2 heart, 3 liver, 4 kidney, 5 ovary, 6 muscle, 7 fin, and M DNA Ladder (DL 2000). b RT-PCR analysis in different developmental stages of zebrafish embryo. 1 2-cell, 2 high, 3 80%-epiboly, 4 early somite, 5 1 dpf, 6 2 dpf, 7 3 dpf, and M DNA ladder (DL 2000). β-actin was used as the internal control

Full size image

In adult fish (Fig. 2a), GH was expressed in brain, liver, ovary and muscle with the highest expression level in the brain that includes pituitary gland by analysis with Glyko Bandscan software. GHR transcript was found in all tissues examined except for heart. The highest levels of GHR expression were observed in liver, followed by brain and ovary. In the case of c-fos, the highest expression was observed in ovary followed by brain, and in all other tissues except the kidney. igf1 transcripts were present in all tissues except kidney with the highest expression level in liver.

During different stages of embryonic development (Fig. 2b), GH transcript was detected in 2-cell stage embryos but not at later stages, reappearing from 1 dpf, and increasing with development. This indicates that GH is maternally expressed and its zygotic expression starts from 1 dpf. GHR is also maternally expressed although the expression level might be not as high as that of GH. The embryonic expression of GHR starts from the early somite stage, and in contrast to GH, the expression of GHR decreased from the early somite stage to 3 dpf. Although c-fos transcript was in 2-cell embryos, from the early somite stage expression of c-fos increased gradually with the time, mimicking the GH expression profile. igf1 was minimally expressed maternally, with expression from 1 dpf, and increasing thereafter.

Activation of GHR signaling by combined overexpression of GH and GHR in zebrafish embryos

In order to quantify GH signaling in transgenic fish, we measured expression of the GH target genes c-fos and igf1. Since c-fos reached a rather high expression level at 1 dpf and igf1 reached its highest expression level at 3 dpf, we used 1 dpf and 3 dpf embryos to check the expression levels of c-fos and igf1, respectively, by normalizing values to 2-cell embryo levels, since these should not be altered by the transgene. By GeNorm analysis, we found that β-actin is the most stable gene among all the candidate reference genes (data not shown) and it could be used to quantify the relative expression levels of c-fos and igf1. First, we injected cGH expression construct into zebrafish embryos and found that c-fos expression was 1.7-fold and igf1 expression was 3.1-fold in the GH-injected embryos in comparison to wildtype embryos (Fig. 3a, b), suggesting significant GHSA in the injected embryos. Second, we overexpressed zebrafish GHR and found 1.7-fold of c-fos expression and 3.5-fold of igf1 expression, demonstrating that overexpression of GHR could also activate GH signaling in the presence of endogenous GH. Third, we used a synergetic expression of GH and GHR, and this overexpression showed 2.0-fold of c-fos expression and 4.5-fold of igf1 expression, higher than these constructs separately. In contrast, overexpression of ΔC-GHR efficiently attenuated the GHSA activity of GH overexpression, as shown by a lower expression level of igf1 (1.5-fold) than the overexpression of GH alone (3.1-fold). These results clearly indicate the importance of GHR level in GH signaling, and by implication, fish growth, since Igf1 is one of the major effectors of growth performance (Eppler et al. 2010).

Fig. 3
figure 3

Analysis of GH-signal activation levels in zebrafish embryos receiving different constructs. a Relative expression of igf1 in zebrafish embryos at 3 dpf by real-time PCR analysis. b Relative expression of c-fos in zebrafish embryos at 1 dpf by real-time PCR analysis. c Relative promoter activities of spi2.1 in zebrafish embryos at 3 dpf by luciferase assay. Control refers to wildtype embryos that were only injected with spi2.1-luc and TK-Renilla constructs. In a, b and c, each bar presents the mean value and the corresponding standard deviation from triplicate analysis. A statistically significant difference (P < 0.05) is indicated by a different letter above the bar, as determined by one-way ANOVA followed by Duncan’s test

Full size image

On the other hand, we conducted luciferase assay by utilizing spi2.1-luc construct (Jiao et al. 2006), to evaluate the overall GHSA level in zebrafish embryos. As shown in Fig. 3c, cGH overexpression induced 2.3-fold GHSA in comparison with wildtype embryos, zGHR overexpression induced 3.7-fold GHSA and the synergetic overexpression of GH and GHR could even result in a 6.1-fold GHSA. Thus the in vivo analysis of GHR signaling activity is validated by the luciferase assay of spi2.1 promoter activity. Both expression analysis of c-fos and igf1 and promoter activity analysis of spi2.1 showed that higher GHR signaling activation could be achieved by synergetic expression of GH and GHR.

Elevated activation of GHR signaling by Jun-GHR overexpression in zebrafish embryos

To test whether our designed Jun-GHR constructs could activate GH signaling, we injected Jun-GHR construct into zebrafish embryos and checked the expression of c-fos in 1 dpf and igf1 in 3 dpf embryos (Fig. 3a, b). The relative expression levels of c-fos and igf1 were 3.9-fold and 10.8-fold, respectively, in Jun-GHR transgenic embryos, relative to the expression level of wildtype embryos. By contrast, the synergetic overexpression of GH and GHR, which showed the highest relative expression levels of c-fos and igf1 in our earlier injection combinations, were only 2.0-fold and 4.5-fold of wildtype embryos. We also checked the promoter activity of spi2.1 in Jun-GHR injected embryos. The relative luciferase activity of the Jun-GHR injected embryos was 26.8-fold, significantly higher than that of the cGH + zGHR injected embryos, which was only 6.1-fold. Herein the overexpression of Jun-GHR gave a higher efficiency of activation of signaling in zebrafish embryos when compared with the synergetic overexpression of GH and GHR. This result has been revealed by the stimulation of two GH target genes, c-fos and igf1, and the activation of GH signaling responsive spi2.1 promoter. Thus it is expected that we would obtain transgenic zebrafish with accelerated growth performance using Jun-GHR construct as the transgene.

Activation of GH targets and accelerated growth of transgenic fish

As shown by RT-PCR analysis during development of transgenic embryos, the transcription of Jun-GHR starts from high stage, reached the highest level in 3 dpf embryos (Fig. 4a). By PCR screening of transgenic P0 generation we obtained 16 Jun-GHR transgene positive founders. We further sampled one P0 Jun-GHR transgenic fish, and Jun-GHR transcript expression was found in almost all the tissues tested, although expression in the brain was weak (Fig. 4b). More interestingly, by both bandscan analysis of the agarose gel bands (Fig. 4b) and real-time PCR assay (Fig. 4c, d), we detected higher expression of GH target genes, c-fos and igf1 in different tissues of transgenic zebrafish. For instance, higher expression of c-fos was observed in all the tissues except kidney (Fig. 4b, c). Ectopic expression of igf1 was even found in the fin tissue of transgenic fish, since there was nearly no expression of igf1 in the fin tissue of wildtype fish (Fig. 4b, d). These results strongly indicate that Jun-GHR transgene is faithfully transcribed in transgenic fish, and as a result strongly activates GH/GHR target genes in various tissues.

Fig. 4
figure 4

The transcription of Jun-GHR, c-fos and igf1 in Jun-GHR transgenic zebrafish. a Jun-GHR transcription in transgenic embryos at different developmental stages. 1 2-cell, 2 high, 3 shield, 4 80%-epiboly, 5 early somite, 6 1 dpf, 7 2 dpf, 8 3 dpf, 9 wildtype embryos, R template of total RNA without reverse transcription. b The transcription of Jun-GHR transgene, c-fos and igf1 in different tissues of transgenic zebrafish. 1 fin, 2 muscle, 3 brain, 4 heart, 5 liver, 6 kidney, 7 ovary, and M DNA Ladder (DL 2000). β-actin amplified from the same samples was used as the internal control. c The relative expression levels of c-fos in different tissues of wildtype and in transgenic zebrafish. d The relative expression levels of igf1 in different tissues of wildtype and in transgenic zebrafish. In c and d, each bar presents the mean value and the corresponding standard deviation from triplicate analysis

Full size image

All the transgene positive founders were crossed with wildtype zebrafish to produce the F1 generation. Among them, 2 crosses gave transgene positive F1 embryos. At 1 mpf, we screened out 100 transgene positive zebrafish from 1 cross and raised them in the aquaria, with 100 non-transgenic zebrafish in the same condition as control. From 2 mpf, the average body weight and body length (mean ± SD) of Jun-GHR transgenic F1 fish showed to be significantly higher (P < 0.01) than those of the non-transgenics (Fig. 5a, b). At 4 mpf, the transgenic fish weighed 0.56 ± 0.08 g at average, 86.1% higher than the average body weight of controls (0.30 ± 0.08 g) (P < 0.01), and the average body length of the transgenics was 3.61 ± 0.18 cm, 25.2% higher than that of the controls (2.88 ± 0.29 cm) (P < 0.01). As shown in Fig. 5c, the transgenic fish showed wider lateral bodies with wider and thicker dorsal bodies than the non-transgenic control fish as what we found in GH-transgenic common carp (Wang et al. 2001). These demonstrate that the expression of Jun-GHR transgene resulted in growth acceleration of transgenic zebrafish.

Fig. 5
figure 5

Growth trial of F1 transgenic and non-transgenic zebrafish. a The increase of body weight along development. b The increase of body length along development. c The typical phenotype of Jun-GHR transgenic (upper) and non-transgenic (lower) zebrafish. Weight and length of transgenic and non-transgenic zebrafish were statistically compared by student’s t-test (** P < 0.01)

Full size image

Discussion

We have previously demonstrated that GH and GHR are present from a very early embryonic stage in mice (Pantaleon et al. 1997), and summarized potential contributions of GH to prenatal development in mammals (Waters and Kaye 2002). In this study, the expression of GH, GHR, c-fos and igf1 genes were extensively and comparatively analyzed during 7 stages of embryonic development and in 7 different tissues of adult zebrafish. Transcription of GH, GHR and c-fos was observed at the 2-cell stage, indicating maternal expression of these genes, and detection of their expression in the ovary further suggests that they start to be accumulated in the oocytes during maturation. For most of them, the zygotic expression was detected from the end of gastrulation, i.e., early somite stage, far prior to the formation of functional pituitary gland. Expression of GH increased with expression of igf1 at 3 dpf. The maternal expression and early expression of GH and igf1 in fishes have also been reported by several investigators (Li et al. 2006, 2007). These observations suggest that GH/GHR pathway also plays a role in early development of fish, and in our present study, overexpression of GHR or GH alone led to increased GHSA in 1 and 3 dpf embryos, strongly indicating the existence of functional GH and GHR expression in zebrafish embryos from 1 dpf. Moreover, blocking GHSA in zebrafish during early development by overexpression a dominant-negative GHR, ΔC-GHR, resulted in embryos with early developmental defects (Ishtiaq Ahmed et al., unpublished data). In previous studies, it has been found that GH can stimulate actin rearrangement (Goh et al. 1997), microtubule polymerization (Goh et al. 1998), and the assembly of multiprotein complex involved in cell adhesion and cell movement (Zhu et al. 1998a, b). Given that the fish embryos undergo dynamic cell movement in early development, e.g., gastrulation movements (Chen et al. 2009), it is possible that the GH/GHR pathway may participate in the regulation of cell movement during early development. It was also reported that GH/Stat5b can directly regulate the transcription of a Wnt signaling element (frizzled4) and suppressors of cytokine signaling (Vidal et al. 2007), thus it is also possible that GH/GHR pathway may crosstalk with the other pathways to regulate the early development of fish.

In most of previous studies, the transcriptional activity of GH signaling in fish has been measured in cultured cells or tissues (Björnsson et al. 2002; Jiao et al. 2006). Here we have utilized zebrafish embryos as a host system to analyze the GH signaling target gene igf1 and the promoter activity of spi2.1, after injection of different combination of transgene constructs, such as GH, GHR, GH + GHR, and GH + ΔC-GHR into zygotes. This has allowed us to compare the signaling activity of GH/GHR pathway upon treatment with different GH pathway-related components in vivo. As compared with overexpression of GH or GHR alone, the combined overexpression of GH and GHR increased GHSA, and overexpression of ΔC-GHR which lacks the intracellular signal transducer domain efficiently attenuated the transcriptional activity. These results validate the efficacy of this in vivo system for studying GH/GHR signaling.

Because GH-induced activation of the GHR is the critical step in GH signaling, we utilized a constitutively activated form of GHR, Jun-GHR to ascertain the effect of continuous activation of GHR in the absence of the IGF-1 feedback on GH secretion. Although shown to be highly active in cell cultures (Behncken et al. 2000), this construct has not been studied in vivo. By using zebrafish as a host animal, we found that Jun-GHR transgenics could not only strongly activate the GH/GHR signaling but also stimulate the growth performance of the transgenic fish. Notably, we have never previously been able to stimulate growth of these zebrafish with GH transgenes alone. In terms of GH target gene activation, Jun-GHR increased the transcriptional activity of the GH/GHR pathway very significantly more than the overexpression of GH, GHR, and even GH + GHR. This demonstrates that CA-GHR transgenesis should be much more powerful than the common GH transgenesis in the aspect of signaling activation and growth stimulation. Intriguingly, igf1 was even ectopically activated in the fin tissue of transgenic zebrafish, which tissue did not display detectable igf1 transcription in wildtype zebrafish. Surprisingly, although Jun-GHR was transcribed in the kidney of transgenic fish, we could not detect visible transcription of igf1 or c-fos in the kidney, which suggests that the zebrafish kidney may lack some intracellular transducer for the GH/GHR pathway. The broad distribution of Jun-GHR transcripts in various tissues and strong activation of GH/GHR target genes have presumably contributed to the allometric growth stimulation of transgenic zebrafish, which will be tremendously useful for increasing yields of transgenic farmed fish. Nevertheless, in future studies we should use “all-fish” CA-GHR constructs for production of transgenic commercial fishes, as CMV promoter was used in our present study.

In previous studies of transgenic animals, researchers usually utilized chimeric constructs. Although the promoters and the coding sequences are from different genes (Laible 2009), the final products are usually the wildtype forms of certain proteins. Here we propose that researchers may produce transgenic animals by introducing the conception of molecular designed protein, e.g., the Jun-GHR product does not exist in the nature but show higher efficiency than the natural proteins. As the comprehensive understanding of many pathways has been extensively advanced, it is certain that other transgenes could be usefully and efficiently modified based on protein design.