Introduction

Large yellow croaker (Pseudosciaena crocea) is one of the commercially important marine species, and the annual yield from large yellow croak aquaculture is larger than that of any other net cage-farmed marine fish species in China (Liu et al. 2013). At present, larval rearing of large yellow croak mainly takes place in Ningde city, Fujian province. Then, the larvae are expanding into other coastal regions of southern China. However, environmental factors of aquaculture water in different regions are changeable, particularly the salinity. Furthermore, most of large yellow croaker are farmed in embayment; the seawater salinity is easily affected by rainstorm and surface flow, causing high mortality of cultured fish (Wang et al. 2013). It is well known that salinity stress can impose oxidative stress on organisms by accelerating the generation of reactive oxygen species (ROS) (Martinez-Alvarez et al. 2002), which leads to disturb fish’s normal physiological function. However, little information is available on the effects of salinity stress on the antioxidant system in this species by now.

To prevent damage caused by ROS, organisms have evolved an antioxidant defense system, which is composed of a complex network of protective antioxidant enzymes and nonenzymatic free radical scavengers, interacting via both direct and indirect mechanisms (Regoli and Giuliani 2014). For example, antioxidant enzyme superoxide dismutase (SOD) can detoxify superoxide anions, catalase (CAT) can scavenge H2O2, glutathione peroxidase (GPx) can scavenge both H2O2 and organic peroxides dependent on glutathione reaction, and glutathione reductase (GR) can catalyze oxidized form of glutathione (GSSG) generated by GPx to regenerate glutathione (GSH) (Halliwell 2012); some nonenzymatic antioxidants, such as GSH, can also have an effect on antioxidant enzyme activities (Srikanth et al. 2013). Extensive studies have been carried out to investigate the effects of salinity stress on lipid peroxidation and the activities of antioxidant enzymes in fish (Martinez-Alvarez et al. 2002; Liu et al. 2007). Some studies have focused on the effects of salinity stress on the expression levels of antioxidant genes in fish (Choi et al. 2008). However, these studies mentioned have only been addressed to investigate a single aspect of antioxidant responses (mRNA or enzymatic levels), and the underlying mechanisms of response to salinity stress at both enzymatic and molecular levels are poorly known in fish. On the other hand, numerous mammalian studies have shown that the NF-E2-related nuclear factor 2 (Nrf2) and Kelch-like ECH-associated protein 1 (Keap1) signaling molecules play an intermediary role in defending against oxidative stress, by orchestrating the gene transcriptions of antioxidant enzymes (Dodson et al. 2015). Although few studies concerning Nrf2 antioxidant defense induced by high stocking density, PFOS or copper in fish has been conducted (Shi and Zhou 2010; Sahin et al. 2014; Jiang et al. 2015); little information is available regarding the role of Nrf2–Keap1 signaling molecules in salinity stress-induced oxidative stress in fish.

In the present study, antioxidant responses were evaluated by determining lipid peroxidation and by investigating the enzyme activities and expression of genes involved in oxidative stress. To our knowledge, this is the first study on the role of Nrf2–Keap1 signaling in fish under salinity stress and also the first time to determinate antioxidant responses to salinity stress both at enzymatic and molecular levels in large yellow croaker. The objective of this study was to elucidate the effects of abrupt salinity stress on hepatic antioxidant responses in large yellow croaker, which will contribute to further understanding the defense characteristics of the large yellow croaker against environmental stress.

Materials and methods

Abrupt salinity stress

Healthy large yellow croaker with an initial weight of 74.6 ± 4.2 g (mean ± SD) were obtained from a local offshore fish cage in the Yuyang Fisheries Co. Ltd., Zhejiang Province, China. Large yellow croaker were held in 500-L circular fiberglass tanks of a static aquarium system for a 2-week acclimatization. The fish were fed twice daily with a commercial diets (lipid and protein contents of 11 and 48% on a dry matter basis, respectively) input in slight excess of satiation during the acclimatization period. Afterward, the fish were acclimated to the target ambient salinities [12 (hypo-salinity), 26 (control, natural seawater), and 40 (hyper-salinity)] with seven salinity changes per hour, each salinity group with three replicates, and 20 fish per tank. During the challenge test, the fish were not fed, and the salinities of the seawater were measured using a handheld YSI 556 Multi-Probe System (YSI, USA) two times per day. The water was prepared at least 1 day before changing and aerated continuously to maintain dissolved oxygen near saturation levels, and 100% of the water volume was renewed daily. The salinity gradients were obtained by adding either dechlorinated tap water or marine salt to the natural seawater. The experiment was exposed to a natural photoperiod and performed at ambient temperature. The ranges of the water quality parameters were as follows: temperature was 25.3 ± 2.9 °C, salinity was 26.4 ± 0.19, dissolved oxygen was 7.38 ± 0.26 mg L−1, pH was 7.54 ± 0.31, and the hardness of the water was 134 mg L−1 as CaCO3.

Sampling and analysis

After the 6, 12, 24 and 48 h of exposure, six individuals in each group were sacrificed via anesthesia overdose by immersing the animal in benzocaine (Sigma); the liver tissues were sampled following decapitation and frozen immediately in liquid nitrogen until biochemical determinations and RNA extraction. We assured that all experiments, animal care, and protocols followed the ethical guidelines of the Zhejiang Ocean University for the care and use of laboratory animals.

Lipid peroxidation and activity of antioxidant enzyme

Liver samples were homogenized in tenfold volumes of ice-cold buffer (1 mmol L−1 of EDTA, 1 mmol L−1 of DTT, 0.5 mol L−1 of saccharose, and 0.15 mol L−1 of KCl; pH = 7.6). The homogenates were centrifuged at 900g at 4 °C for 10 min to precipitate large particles and centrifuged again at 12,000g at 4 °C for 20 min (Zeng et al. 2015). The resultant supernatant was collected and stored at −80 °C until it was measured for biochemical analysis.

Lipid peroxidation was determined by the thiobarbituric reactive species (TBARS) assay, which measured the production of malondialdehyde (MDA) that reacted with thiobarbituric acid, according to the method described by Livingstone et al. (1990). GSH level was measured as previously described by Vandeputte et al. (1994). The MDA level and GSH level were expressed as nanomole of MDA and microgram of GSH per milligram of soluble protein, respectively.

Superoxide dismutase (SOD, EC 1.15.1.1) activity was measured as previously described by Beauchamp and Fridovich (1971) based on aerobic reduction of NBT at 535 nm by superoxide radicals. Catalase activity (CAT, EC 1.11.1.6) was determined as previously described by Aebi (1984). The activity of glutathione peroxidase (GPx, EC 1.11.1.9) was determined as previously described by Drotar et al. (1985). Glutathione reductase activity (GR, EC 1.6.4.2) was assayed as previously described by Tanaka et al. (1994). Protein concentration was determined by the Bradford method (Bradford 1976) using bovine serum albumin (BSA) as the standard (Bradford protein assay kit, Tiangen Biotech, Beijing, China). The assays were run in triplicate. All enzyme activities were expressed as U (units) per milligram of soluble protein.

Gene expression

Total RNA was isolated from liver tissue using TRIzol® reagent (Invitrogen) according to the manufacturer’s instruction. The RNA quality was analyzed using 1.0% agarose gel electrophoresis, and the RNA concentration and purity were determined using an ultramicro spectrophotometer (Thermo Scientific) at 260 and 280 nm. Two milligrams of total RNA was used as templates to synthesize the first-strand complementary DNA (cDNA) using a first-strand cDNA synthesis kit (Fermentas). The cDNA synthesis reactions were diluted to 200 μL in water. Gene expression levels were determined by quantitative real-time PCR that was conducted on an Applied Biosystems Prism 7500 Sequence Detection System (Applied Biosystems, USA). The primer sequences of genes used in this analysis are shown in Table 1. The design was based on the genomic sequences in the large yellow croaker genome data (Wu et al. 2014). The qPCR amplifications were performed in a 20-μL reaction volume containing 10 μL SYBR® Premix Ex Taq Master Mix (Takara), 2.0 μL of cDNA, and 0.2 μM of each primer. PCR amplification was performed in duplicate, using the following thermal profile: initial denaturation of 1 min at 95 °C, followed by 45 cycles of 5 s at 95 °C and 10 s at 57 °C, and a final extension of 30 s at 72 °C. Each reaction was verified to contain a single product of the correct size using agarose gel electrophoresis, and dissociation curve analysis was performed after each assay to determine target specificity. Standard curves were constructed for each target gene to explain the differences in amplification efficiency between different RNA samples. A reaction was carried out without the cDNA template to ensure the absence of contamination in the reagents and primers used. A set of four housekeeping genes (EF 1α, GAPDH, β-actin, and HPRT) were selected from the literature to test their transcription stability. According to the geNorm software (Vandesompele et al. 2002), GAPDH and β-actin were used as a housekeeping gene, both of which did not change over the course of the experimental treatments. The relative expression of each gene was determined using the “delta–delta Ct” method (Pfaffl 2001), using a geNorm method to normalize the geometric mean of the best combination of two genes (GAPDH and β-actin).

Table 1 Primer sequences used for real-time PCR
Full size table

Statistical analysis

Prior to statistical analysis, all data were tested for normality of distribution using the Kolmogornov–Smirnov test. The homogeneity of variances among the different treatments was tested using Barlett’s test. The results were subjected to one-way ANOVA and Tukey’s multiple range test. Nonparametric Spearman’s correlation analysis was used to examine the relationship between different parameters. Analysis was performed using the SPSS 16.0 for Windows (SPSS, Michigan Avenue, Chicago, IL, USA). For all analyses, significant levels were set at P < 0.05. All data were presented as the means ± SEM.

Results

Survival rate and MDA and GSH content

The survival rates of fish were 100% except those in the hyper-salinity group with a significantly lower survival rate (83.33 ± 5.77%, P < 0.05). The lowest and highest peaks of MDA content in the hypo-salinity group were recorded at 6 and 12 h, respectively (Fig. 1a). Compared with the control group, fish in the hyper-salinity group remarkably decreased MDA content at 6 h followed by a remarkable increase towards the end of the exposure. In comparison with the control group, fish in the hypo-salinity group significantly elevated GSH content at 6 and 24 h and sharply reduced GSH content at 12 h (Fig. 1b). Compared with the control group, fish in the hyper-salinity group remarkably increased GSH content at 6 h followed by a sharp reduction towards the end of the exposure.

Fig. 1
figure 1

Effects of acute salinity stress on MDA content and GSH content in the liver of large yellow croaker. Values are means ± SEM (n = 3). Different letters indicate significant difference at P < 0.05

Full size image

Antioxidant enzyme activity

In comparison with the control group, fish in the hypo-salinity group remarkably increased SOD activity at 6 and 24 h (Fig. 2a), GPx activity at 6 and 24 h (Fig. 2c), and GR activity at 6 and 24 h (Fig. 2d) and had no effect on CAT activity (Fig. 2b). Compared with the control group, fish in the hyper-salinity group significantly enhanced SOD activity at 6 h, GPx activity at 6 and 12 h, and GR activity at 6 h, sharply inhibited SOD activity at 48 h, GPx activity at 48 h, and GR activity at 48 h, and had no effect on CAT activity.

Fig. 2
figure 2

Effects of acute salinity stress on activities of SOD (a), CAT (b), GPx (c), and GR (d) in the liver of large yellow croaker. Values are means ± SEM (n = 3). Different letters indicate significant difference at P < 0.05

Full size image

Expression of antioxidant genes

In comparison with the control group, fish in the hypo-salinity group remarkably increased Cu/Zn-SOD expression at 6 and 24 h (Fig. 3a), Mn-SOD expression at 6 and 24 h (Fig. 3b), CAT expression at 6 and 24 h (Fig. 3c), GPx1a expression at 6 and 24 h (Fig. 3d), GPx1b expression at 6 h (Fig. 3e), GR expression at 6 and 24 h (Fig. 3f), Nrf2 expression at 6 and 24 h (Fig. 3g), and Keap1 expression during 12–48 h (Fig. 3h) and sharply reduced Cu/Zn-SOD expression at 12 h and Mn-SOD expression at 12 h. Compared with the control group, fish in the hyper-salinity group significantly stimulated Cu/Zn-SOD expression at 6 h, CAT expression at 6 h, GPx1a expression during 6–12 h, GPx1b expression during 6–12 h, GR expression at 6 h, Nrf2 expression during 6–12 h, and Keap1 expression during 24–48 h and remarkably inhibited Cu/Zn-SOD expression at 48 h, Mn-SOD expression at 48 h, CAT expression at 48 h, GPx1a expression during 48 h, GR expression during 24–48 h, Nrf2 expression at 48 h, and Keap1 expression at 6 h.

Fig. 3
figure 3

Effects of acute salinity stress on mRNA levels of Cu/Zn-SOD (a), Mn-SOD (b), CAT (c), GPx1a (d), GPx1b (e), GR (f), Nrf2 (g), and Keap1 (h) in the liver of large yellow croaker. Values are means ± SEM (n = 6). Different letters indicate significant difference at P < 0.05

Full size image

Correlation analysis

MDA content was negatively related to activities of GSH, SOD, GPx, and GR (Table 2). Pearson’s correlation coefficients between enzymatic activities and mRNA levels of according genes and between the transcription factor Nrf2 and mRNA levels of genes involved in oxidative stress are presented in Table 3. Positive correlations were observed between activity and expression of GR. There also were positive relationships between SOD activity and Cu/Zn-SOD expression, between GPx activity and GPx1a expression, and between GPx activity and GPx1b expression. No correlation was observed between SOD activity and Mn-SOD expression and between activity and expression of CAT. Nrf2 expression was positively correlated with the mRNA levels of Cu/Zn-SOD, Mn-SOD, CAT, GPx 1a, GPx 1b, and GR and was negatively related to the mRNA level of Keap1.

Table 2 Pearson’s correlation coefficients between MDA content and antioxidant enzyme activities
Full size table
Table 3 Pearson’s correlation coefficients between enzymatic activities and mRNA levels of according gene and between the transcription factor Nrf2 and mRNA levels of antioxidant genes
Full size table

Discussion

Studies involved in the potential signaling molecules of antioxidant system in fish under salinity stress have not received much attention. The present study provides new experimental evidence for antioxidant defense in large yellow croaker under salinity stress and also elucidates a central role of the transcription factor Nrf2 in the salinity stress-induced oxidative stress for the first time.

The exact mechanism involved in salinity stress-induced oxidative stress is poorly understood, but it may be closely involved in changes in the activities of antioxidant enzymes (Martinez-Alvarez et al. 2002; Liu et al. 2007; Choi et al. 2008). Among the antioxidant enzymes, SOD–CAT and SOD–GPx are considered to be the vital first line of defense against oxidative stress (Winston and Di Giulio 1991), since SOD catalyzes superoxide anion radical (O2 −) and H+ into O2 and H2O2, which is subsequently transformed into nontoxic H2O by CAT and GPx. Thus, cytosolic CAT and/or GPx activities are often induced concomitantly with the activation of SOD to protect cells from oxidative stress (Di Giulio et al. 1989). In the present study, a coordinated increase was observed in activities of SOD and GPx in fish under salinity stress at 6 h, suggesting a protective role against damage by salinity stress during the early stage of exposure. However, salinity stress had no effect on CAT activity. It is known that superoxide radicals can directly inhibit CAT activity. Moreover, the antioxidant capacity of CAT could be compensated by an increase in the activity of GPx (Kang et al. 2005). The increase in GR activity suggested that a pool of reduced GSH is maintained by enhanced rates of turnover (re-conversion of oxidized GSSG to GSH) and is available as substrate for GPx. Thus, the enhancement of GR activity implied a protective and adaptive response to salinity stress in the liver and may be associated with an increase in GSH (Halliwell 2012). For example, the significant increase in GR was accompanied by a remarkable increment of GSH content in the hypo-salinity group at 6 and 24 h and in the hyper-salinity group at 6 h. In the present study, lipid peroxidation (MDA) remained relatively constant or significantly reduced while activities of antioxidant enzymes significantly increased during the early stage of exposure in both groups, further supporting the protective efficiency of antioxidant enzymes against salinity stress-induced oxidative stress. Negative relationships between lipid peroxidation and activities of antioxidant enzymes further confirmed the fact mentioned above. However, there is an accumulated risk of oxidative damage because of increased lipid peroxidation when the antioxidant system could not neutralize or eliminate the excess of ROS, which in turn inhibit the activities of antioxidant enzymes or even degrade the enzymes (Lushchak 2011). This notion was supported by the increase in lipid peroxidation and the reduction in activities of SOD, GPx, and GR at 48 h in the hyper-salinity group.

Antioxidant gene expression is considered an accurate estimate of fish antioxidant capacity, where interference of biochemical origin is not involved (Malandrakis et al. 2014). In this sense, it could be of added value to determine the onset of antioxidant gene reaction under salinity stress. In the present study, mRNA levels of Cu/Zn-SOD, Mn-SOD, GRx1a, GRx1b, and GP were upregulated in a time-dependent manner by salinity stress. The enhanced mRNA levels of these enzymes indicated that salinity stress can stimulate antioxidant capacity, which in turn correlates well with increasing activities of antioxidant enzymes in large yellow croaker. However, there was no significant correlation between SOD activity and Mn-SOD expression and between activity and expression of CAT. The possible reasons for the mismatch relationships between mRNA levels and enzyme activities are as follows: first, an SOD gene in large yellow croaker has more than two isoenzymes. The mRNA transcription level is limited to one subtype of the antioxidant gene encoding a single isoenzyme, while enzyme activity detected in the liver is equal to the total enzyme activity of the different isoenzymes. Second, enzyme activity might also be modulated at post-translational level (Sadi et al. 2014). Third, mRNA level of an antioxidant enzyme represents a snapshot of activity at any given time, while there is a time-lag effect between transcription and translation (Nam et al. 2005).

The induction of antioxidant enzyme genes is regulated by several cell signaling pathways and transcription factors (Fiol and Kültz 2007). The transcription factor Nrf2 is a master regulator of the cellular antioxidant response through the Nrf2–Keap1 signaling pathway (Dodson et al. 2015). In particular, Nrf2 is important in protecting the liver, since Nrf2 absence increases hepatic lipid peroxidation (Gong and Cederbaum 2006). In the present study, an increase in Nrf2 expression was observed in fish exposed to the hypo-salinity group at 6 and 24 h and the hyper-salinity group during 6–12 h. Enhanced expression of Nrf2 in salinity stress-induced oxidative stress indicating a protective role of Nrf2. Nrf2 expression was positively related to expression of antioxidant genes, suggesting Nrf2 may play an important role in regulating antioxidant genes. Similar results have previously been observed in Oncorhynchus mykiss (Sahin et al. 2014) and Danio rerio (Wang and Gallagher 2013). However, the persistent accumulation of Nrf2 in the nucleus may have dangerous effects, like free radical damage, apoptosis, and tumorigenesis (Katoh et al. 2005; Kensler et al., 2007). The enhancement of Keap1 expression would increase Nrf2 degradation, leading to a feedback autoregulatory loop which controls Nrf2 abundance (Katoh et al. 2005). Keap1 reversely correlated with the mRNA expression of Nrf2, indicating that Keap1 is a negative regulator to switch off the Nrf2 response. Similar result was also obtained from Anguilla anguilla exposed to H2O2 (Giuliani and Regoli 2014).

In conclusion, our study clearly elucidated abrupt salinity stress-induced antioxidant defenses, depending on salinity concentrations and time course. Compared with the hypo-salinity group, fish in the hyper-salinity group remarkably increased MDA and mortality rate while significantly reducing activities and mRNA levels of antioxidant enzymes during the late stage of exposure, suggesting that fish cannot adapt to abrupt high salinity stress. Positive correlations between gene expression and antioxidant enzyme activities were observed, suggesting transcriptional regulation may play an essential role in defending against oxidative damage of salinity stress. In the process, the Nrf2–Keap1 pathway is required for the induction of antioxidant genes. However, it should be noted that the gene mRNA only provides a portion of the transcriptional information about the de novo syntheses of these factors. Detailed mechanisms should be revealed by analyses of their protein levels and post-translational modifications in the future research.