Introduction

Insulin-like growth factor I (IGF-I) is a mitogenic polypeptide with similar molecular structure to insulin. The structure and elementary endocrine function of IGF-I are highly conserved among vertebrates (Moriyama et al. 2000). In fish, apart from being involved in metabolism (Castillo et al. 2004), reproduction (Weber and Sullivan 2000), development (Pozios et al. 2001), and osmoregulation (McCormick 2001), IGF-I is a primary mediator of the effects of growth hormone (GH) (commonly known as the GH-IGF-I axis). It plays a key part in the regulation of systemic growth and hastens growth on almost every cell in the organism (Yakar et al. 2002). Considering its critical biological functions, since the first fish IGF-I was isolated from coho salmon (Cao et al. 1989), many studies have reported the characterizations of IGF-I from different fish species (Clay et al. 2005; Sciara et al. 2008; Xu et al. 2014).

Compensatory growth (CG), also known as catch-up growth, is a phase of accelerated growth when favorable conditions are reinstated after a period of growth restraint mainly induced by complete or partial food deprivation (Ali et al. 2003). Because CG results in enhanced food efficiency and growth rate, the studies for CG in cultivated fish have been carried out in last decades (Won and Borski 2013). Although the underlying mechanisms are still poorly understood, the relevant suites of physiological responses are largely governed by the GH-IGF-I axis during the episodes of food deprivation (Cameron et al. 2007; Imsland et al. 2008). Taking this into account, it is of great significance to examine the time-course responses of the IGF-I to food deprivation and refeeding.

Yellow catfish (Pelteobagrus fulvidraco) distributes in lakes and river channels in eastern Asia from Laos and Vietnam to southeastern Siberia. Due to its great economic value, tolerance for long distance transportation, and widespread availability of fry, it has now become one of the most important freshwater finfish species in Chinese aquaculture (Wu et al. 2010; Dong et al. 2011; Jia et al. 2016). In 2016, its production was 355,700 tons in China according to the data of Ministry of Agriculture of the People’s Republic of China (CFSY, 2016). Therefore, this study was conducted to characterize the full-length cDNAs of the IGF-I of Yellow catfish. The relationship between IGF-I expressions and growth indexes of this species, body weight (BW), body length (BL), and condition factor (CF), was also investigated during starvation and refeeding periods. The data obtained here might illustrate the effects of IGF-I in the compensatory growth of the P. fulvidraco. It might also benefit the designation of appropriate feeding strategies to promote compensatory growth in future studies.

Materials and methods

Fish and tissue collection

Juvenile and adult of P. fulvidraco were collected from the Freshwater Fisheries Research Institute of Jiangsu Province (Jiangsu, China). Prior to experiments, fish were maintained in a recirculating aquaculture system in the laboratory under the following conditions: water temperature, 25–27 °C; dissolved oxygen (DO), 5.0–6.0 mg/L; pH, 7.2–7.6; photoperiod, 12:12 h (dark:light). Fish were fed with concentrated feed (Ningbo Tech-bank, China), which contains 38% protein and 7% lipid for fortnight.

The samples were anesthetized by diluted MS-222 (tricaine methanesulfonate, Sigma, USA) at the concentration of 100 mg/L. After euthanization, certain tissues from each sample (weight of 101.09 ± 20.2 g) were immediately collected and stored in liquid nitrogen for further study.

Feeding regimes experiment

The juvenile fish (initial weight3.80 ± 0.78 g) were randomly divided into two groups (experimental group and control group). Each group had three replicates, including 30 individuals per subgroup. Whole experiment lasted for 6 weeks, and it included two stages, the starvation stage (weeks 1 to 3) and refeeding stage (weeks 4 to 6). The fish in the control group(C) were fed three times a day (6:30, 12:00, and 17:30) with the commercial diet (Ningbo Tech-bank, China) until satiation throughout the experiment. After the starvation stage, fish in the experimental group were fed to satiety three times a day (6:30, 12:00, and 17:30) in the refeeding stage. Fish were sampled at 0, 7, 14, and 21 days during fasting and at 7, 14, and 21 days during refeeding, respectively. At each sampling time point, three individuals were collected randomly from each subgroup of experiment group (totally nine individuals). After the weight and total length of sampled fish were measured individually, liver tissues were subsequently frozen in liquid nitrogen. Growth indices including final weight and CF were calculated according to the formula CF = 100 × (weight/total length3).

Isolation of full-length IGF-I cDNA

Total RNA was extracted from the liver tissue of an adult P. fulvidraco using Trizol Reagent (Invitrogen, USA), and the remaining DNA was removed by DNase digestion. Primer IGF-I-F and IGF-I-R were designed based on the highly conserved region s of IGF-I from some closely related teleost species to obtain an IGF-I cDNA fragment of the P. fulvidraco (Table 1). First-strand cDNA was synthesized by reverse transcription according to the instructions of Clontech SMARTer™ RACE cDNA Amplification kit (Takara, China).The specific primer was designed based on the obtained IGF-I fragment to perform the 3′/5′-cDNA end amplification (RACE), using SMARTer™ RACE cDNA Amplification Kit (Invitrogen, USA).The obtained PCR products were cloned into a pGEM T-Easy vector (Promega) and sequenced from both directions, which were used to assemble the full-length IGF-I cDNA for the P. fulvidraco.

Table 1 Nucleotide sequences and positions of primers used in the experiment
Full size table

Phylogenetic analysis with homologous genes

Homology search was carried out from NCBI database by the BLASTN and BLASTP (Altschul et al. 1997), and multiple alignments were performed by CLUSTAL X (Thompson et al. 1997). The putative amino acid sequences of the identified IGF-I cDNA and other related sequences were used to build a topological tree via MEGA6.0 with the neighbor-joining (NJ) method (Tamura et al. 2013) with 1000 times bootstrap test.

QPCR analysis

The spleen, anterior intestine, posterior intestine, liver, head kidney, skin, gill, brain, ovary, stomach, testis, muscle, and adipose tissues were collected from six adult P. fulvidraco (3 males and 3 females). Total RNA was extracted by Trizol Reagent (Invitrogen, USA), and the quality and quantity of RNA were detected by Nanodrop (Nanodrop Technology, USA). The residual DNA was digested through DNase digestion. QPCR primers for P. fulvidraco were designed by the Primer 3. As an internal reference gene, β-actin was used to calibrate the cDNA template. First-strand cDNA was synthesized using SYBR® PrimeScript™ RT-PCR Kit. The amplification program contained 40 cycles of 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s. The 2−△△Ct method was used for the analysis of the relative gene expressions of the IGF-I in P. fulvidraco (Livak and Schmittgen 2001).

Statistical analysis

Data about IGF-I mRNA expression and the growth performance were recorded, and the normality of distribution and homogeneity of variances was estimated utilizing Kolmogorov-Smirnov test and Levene’s test, respectively. One-way ANOVA followed by Duncan’s test was employed to estimate the starvation/refeeding effects on growth indices when variances were homogenous. Kruskall-Wallis test was employed if variances were not uniform. Dunnett’s T3 test was utilized to analyze the post hoc multiple comparisons of mean values. t test or Mann-Whitney U test was used to compare the two means depending on homogeneity of variances between treatments, whenever appropriate. IBM SPSS Statistics version 19 was used to perform all the above-mentioned statistical analyses, and significant difference was considered between different treatments if P was smaller than 0.05.

Results

IGF-I cDNA sequence features and protein sequence homology with other species

The cDNA of P. fulvidraco, with 879 bp in length, containing the open reading frame, the sequence encoding IGF-I and a poly (A) tail, was identified (KX434878). The coding region of 480 bp, which encodes a polypeptide comprising 597 amino acids, was preceded by a 270 bp 5′-UTR and followed by a 134 nt 3′-UTR (Fig. 1). It is composed of a signal peptide (41 aa), domain B (29 aa), domain C (12aa), domain A (21 aa), domain D (9 aa), and domain E (47 aa) (Fig. 2). The predicted mature IGF-I protein has 71 amino acid residues with a calculated molecular mass of 17.58 kD. We compared the protein sequences encoded by IGF-I of P. fulvidraco with some closely related species in GenBank. The percent identity between deduced amino acids of IGF-I in P. fulvidraco and those of other species was as follows: Ictalurus punctatus (87%, NP_001187224.1), Ctenopharyngodon idella (85%, ABU40947.1), Cyprinus carpio (78%, ABQ08938. 1), Danio rerio (77%, NP_571900.1), Elopichthys bambusa (77%, AEA72297.1), Carassius auratus (76%, ABG75920.1), Esox lucius (73%, XP_010902153.1), Pimephales promelas (73%, AAT02176.1), and Salmo salar (Atlantic salmon) (71%, ABO36526.1). IGF-I domains starting locations are indicated at the top of the graphic (B, C, A, D, E). The domain of the IGF genes was underlined. Asterisks indicate the residues of the conserved feature of the IGFBP binding surface on conserved domain IGF.

Fig. 1
figure 1

The complete nucleotide sequence of the pre-pro IGF-I cDNA of P. fulvidraco. The lowercase letters are the 3′untranslated region and 5′untranslated region. The predicted pre-pro IGF-I peptide contains the signal peptide (residues 1–41, is underlined), the mature peptide (residue 42–112), and the E domain (residues 113–159). Initiation codon (ATG) is boxed and stop codons are shown in italics. IGF domain is in bold letters

Full size image
Fig. 2
figure 2

Comparison of the primary amino acid sequence of mature IGF-I of P. fulvidraco with some closely related species. IGF-I domains starting locations are indicated at the top of the graphic (B, C, A, D, E). The domain of the IGF genes was underlined. Asterisks indicate the residues of the conserved feature of the IGFBP binding surface on conserved domain IGF

Full size image

Phylogenetic trees for IGF-I

To determine the evolutionary relationships between P. fulvidraco and other species, a phylogenetic tree was built based on the results of multiple alignments (Fig. 3).The results present two distinct clades. Amphibians, reptiles, birds, and mammals were in one clade, while Siluriformes, Cypriniformes, Perciformes, Acipenseriforms, and other teleost fish were in the other clade. Among them, P. fulvidraco first clustered with I. punctatus which was also belonging to order Siluriformes with a high bootstrap value of 100%.

Fig. 3
figure 3

Neighbor-joining phylogenetic tree IGF-I from P. fulvidraco and other species. The IGF-I sequences included in this analysis other than P. fulvidraco are Ictalurus punctatus (NP_001187224) and Homo sapiens (nm_000612)

Full size image

Expression of IGF-I mRNA in P. fulvidraco tissues

The expression of IGF-I mRNA was detectable in all tissues involved (Fig. 4). The expression in the liver presents the highest level, while lower expression was found in the brain, spleen, head kidney, and adipose than that of the liver. Detectable amounts of IGF-I mRNA were also found in the posterior intestine, ovary, muscle, skin, gill, anterior intestine, testis, and stomach, with lower expression level.

Fig. 4
figure 4

Reverse transcriptase polymerase chain reaction (RT-PCR) expression analysis on IGF-I gene. For tissue expression, data are referred to the values obtained in the spleen. Each column represents the mean (n = 3) and the vertical bars indicate S.E.M. Lanes correspond to (1) spleen, (2) anterior intestine, (3) posterior intestine, (4) liver, (5) head kidney, (6) skin, (7) gill, (8) brain, (9) ovary, (10) stomach, (11) testis, (12) muscle, and (13) adipose

Full size image

The growth indexes of fish subjected to different feeding regimes

During the experiment, fish in the control group grew gradually, and starvation of experimental group had significant impact to fish growth and CF (Fig. 5a, b, and c). By the end of the 3-week starvation period, the BL, BW, and CF of the test group were dramatically lower than those of the control group (P < 0.01). Specifically, the BL and BW of the starved group were 6.16 cm and 3.01 g in comparison with 7.16 cm and 6.20 g of the control group. The CF values were 1.28 and 1.67 in the experimental group and control group, respectively. After 2 weeks refeeding, the BL and BW of test group increased highly significantly (P < 0.01), while CF increased significantly (P < 0.05). By the end of the 3-week refeeding, no statistically significant difference was found between the test group and control group (P > 0.05).

Fig. 5
figure 5

Changes of body weight (a), body length (b), condition factor (c), and hepatic IGF-I mRNA expression (d) of juvenile P. fulvidraco subjected to different feeding regimes. Fish in the control group (□) were fed three times a day throughout the experiment. Fish in the test group (▲) were starved for 3 weeks and then refed three times a day for the next 3 weeks. Each point represents the means ± S.E.M. (n = 9). Significant differences (P < 0.05) among sampling times within each treatment are indicated by different letters (upper case for control, lower case for starved, and refed group). * indicates a significant difference (P < 0.05) between the two treatments at each sampling time. **P < 0.01

Full size image

mRNA expressions of IGF-I in the liver of P. fulvidraco subjected to different feeding regimes

Initially, hepatic IGF-I expressions showed slight difference within control treatment (P > 0.05). After 1 week, the IGF-I mRNA expressions in the experimental group were significantly lower than those of the control group (P < 0.05). After 2 weeks starving, the hepatic IGF-I expressions were decreasing further (P < 0.01).

Refeeding of the starving fish increased the hepatic IGF-I expressions significantly (P < 0.05). By the end of the 3-week refeeding, the difference between the two groups was not statistically significant anymore.

Discussion

In this study, the cloning and characterization of P. fulvidraco IGF-I mRNA were described. The deduced amino acid sequence revealed high identity with other teleost fish and other vertebrates. The homology between P. fulvidraco and I. punctatus is 87%. Because only the two Siluriformes species IGF-I sequences were obtained so far, it seems to be not enough to conclude the homology of IGF-I among all Siluriformes. More IGF-I of Siluriformes and other species needed in-depth study.

The IGF-I mRNA was detected in all tissues we analyzed in P. fulvidraco in the present study. Similarly, in other teleosts, birds, and even mammals, the highest IGF-I mRNA expressions were also found in hepatic tissue, while the expressions were detectable but relatively lower in those non-hepatic tissues (Duan et al. 1993; Duguay et al. 1994; Inoue et al. 2003; Vong et al. 2003; Biga et al. 2004; Caelers et al. 2004). These results indicated that extrahepatic IGF-I had tissue-specific paracrine/autocrine functions, which was also common in mammals (Sara and Hall 1990; Jones and Clemmons 1995). In this study, the predominant expression was observed in the liver. The transcript levels in the intestine, spleen, head kidney, and mesenteric adipose tissue were also remarkable. The abundant presence of IGF-I mRNA in liver was not surprising in fish (Moriyama et al. 2000) because the liver was commonly accepted as the most important tissue to synthesize and secrete IGF-I (Shamblott et al. 1995; Moriyama et al. 2000). The relatively high expression found in brain was also justifiable since IGF-I play a critical role in the regulation of cell growth and development, especially in nerve cells (Yakar et al. 2002). However, the aforementioned studies mainly focused on mammals; the functions of IGF-I in neurogenes in fish have still not been well understood and should be studied further. As for the gastrointestinal tract, the moderate abundance could indicate the differential physiology of intestine regions and the response to GH (Sciara et al. 2008). However, after the chronic stimulation of GH, analysis of difference in absorption capacity between the anterior and the posterior intestine may provide an answer to this question. Because this tissue is a complex, essential, and highly active metabolic and endocrine organ, the relatively high expression of IGF-I in adipose was also rational (Kershaw and Flier 2004). It responds to afferent signals from traditional hormone systems and the central nervous system, as well as expresses and secretes factors with important endocrine functions (Kershaw and Flier 2004). Due to the fact that little literature is available concerning the potential roles of the spleen and head kidney in endocrine regulations, the moderate IGF-I expressions in both organs are hard to explain, as warrants further in-depth studies.

The comparison of IGF-I expression between starving and refeeding group in juvenile P. fulvidraco was carried out to investigate its potential effects in the systemic growth of fish. Deprivation of food for 1–2 weeks resulted in a remarkable decrease in BW, BL, and CF than those of the fed group (P < 0.05). However, all parameters increased significantly after a refeeding period of 3 weeks, and finally there was no statistical difference between the experimental group and the control group (P>0.05). This was the consequence of compensatory growth, which was a common phenomenon that organisms grow rapidly after a period of slow development (Ali et al. 2003; Jobling 2010). Similar findings have also been observed in Atlantic cod (Gadus morhua), that complete compensation was observed after 3 weeks fasting (Jobling et al. 1994).

Long-term nutrition deprivation prolonged starvation lead to a remarkable reduction of hepatic IGF-I mRNA expression in P. fulvidraco (P < 0.05), and refeeding restored the expression to normal level (P > 0.05). These results suggested that nutritional condition played a vital role in hepatic IGF-I expressions in teleost fish. Together with the results in growth performance, it suggested that (1) starvation reduced IGF-I mRNA expression in the hepatic tissue as well as systemic tissues, and finally results in growth retardation of P. fulvidraco; and (2) refeeding induced an enhanced expression of IGF-I, thus may accelerate the growth rate of this species. However, further investigation concerning the effects of feeding regimes on the entire GH-IGF-I axis is needed to achieve further understanding of functions of this axis, particularly how to regulate the growth of skeleton and the partition of nutrient (Navarro and Gutiérrez 1995; Pérez-Sánchez and Bail 1999).