Molecular characterization of superoxide dismutase and catalase genes, and the induction of antioxidant genes under the zinc oxide nanoparticle-induced oxidative stress in air-breathing magur catfish (Clarias magur)

The deduced amino acid sequences from the complete cDNA coding sequences of three antioxidant enzyme genes (sod1, sod2, and cat) demonstrated that phylogenetically the magur catfish (Clarias magur) is very much close to other bony fishes with complete conservation of active site residues among piscine, amphibian, and mammalian species. The three-dimensional structures of three antioxidant enzyme proteins are very much similar to mammalian counterparts, thereby suggesting the functional similarities of these enzymes. Exposure to ZnO NPs resulted in an oxidative stress as evidenced by an initial sharp rise of intracellular concentrations of hydrogen peroxide (H2O2) and malondialdehyde (MDA) but decreased gradually at later stages. The level of glutathione (GSH) also increased gradually in all the tissues examined after an initial decrease. Biochemical and gene expression analyses indicated that the magur catfish has the ability to defend the ZnO NP-induced oxidative stress by inducing the SOD/CAT enzyme system and also the GSH-related enzymes that are mediated through the activation of various antioxidant-related genes both at the transcriptional and translational levels in various tissues. Furthermore, it appeared that the stimulation of NO, as a consequence of induction nos2 gene, under NP-induced oxidative stress serves as a modulator to induce the SOD/CAT system in various tissues of magur catfish as an antioxidant strategy. Thus, it can be contemplated that the magur catfish possesses a very efficient antioxidant defensive mechanisms to defend against the oxidative stress and also from related cellular damages during exposure to ZnO NPs into their natural environment.


Introduction
Parallel to the increase of application of nanotechnology in recent years, more utilization of metal oxide nanoparticles (NPs) have received significant attention mainly because of their remarkable and multidimensional potentials. These metal oxide NPs are one of the most widely used nanomaterials in industrial and domestic applications (Aitken et al. 2006). But, at the same time, a large-scale beneficial use of Abstract The deduced amino acid sequences from the complete cDNA coding sequences of three antioxidant enzyme genes (sod1, sod2, and cat) demonstrated that phylogenetically the magur catfish (Clarias magur) is very much close to other bony fishes with complete conservation of active site residues among piscine, amphibian, and mammalian species. The three-dimensional structures of three antioxidant enzyme proteins are very much similar to mammalian counterparts, thereby suggesting the functional similarities of these enzymes. Exposure to ZnO NPs resulted in an oxidative stress as evidenced by an initial sharp rise of intracellular concentrations of hydrogen peroxide (H 2 O 2 ) and malondialdehyde (MDA) but decreased gradually at later stages. The level of glutathione (GSH) also increased gradually in all the tissues examined after an initial decrease. Biochemical and gene expression analyses indicated that the magur catfish has the ability to defend the ZnO NP-induced oxidative stress by inducing the Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s10695-021-01019-3. different metal oxide NPs has resulted in an uncontrolled release of these particles into the environment (Ray et al. 2009;Lin et al. 2010;Smita et al. 2012), thus causing an alarming increase of NPs into the surrounding environment as abiotic stress factors. Following the environmental release, the NPs are likely to enter into the aquatic ecosystem, which ultimately are causing a considerable load of different NPs in water bodies (Wigginton et al. 2007). Various pieces of evidence suggested that the NPs are potent aquatic pollutants since aquatic ecosystems are the final destination for almost all anthropogenic discharges that flow through rainwater, which ultimately may have their adverse physiological effects on aquatic biota, including the fish species (Scown et al. 2010;Shaw and Handy 2011;Handy et al. 2011;Cazenave et al. 2019).
ZnO NPs are noncombustible, white, and odorless metal oxide powders. In recent years, ZnO NPs are known to be utilized extensively in industries and personal products due to their unique physicochemical characteristics (Zhang et al. 2013). ZnO NPs have been recognized as "extremely toxic" to aquatic organisms (Kahru and Dubourguier 2010); however, the toxicity level varies considerably to aquatic organisms belonging to various trophic levels (Blinova et al. 2010;Wong et al. 2010;Hanna et al. 2013). Several earlier studies demonstrated the embryotoxicity and developmental abnormalities under ZnO NPs exposure in zebrafish (Danio rerio), marine medaka (Oryzias melastigma), and daphnia (Daphnia magna) (Choi et al. 2016;Cong et al. 2017;Shin et al. 2018). Although many studies have elucidated the potential hazards of ZnO NPs to aquatic life and their environmental impacts, not much of information is available about the ZnO NP-induced cellular toxicity resulted due to overproduction of reactive oxygen species (ROS) and the possible endowed protective strategies. Some recent studies on common carp (Cyprinus carpio) and Nile tilapia (Oreochromis niloticus) have revealed the effects of ZnO NPs on bioaccumulation, oxidative stress, and the subsequent activation of the antioxidant defense system (Hao and Chen 2012;Kaya et al. 2015). In aquatic ecosystems, NPs are taken up by fish and transported to the various tissues and organs through the blood (Handy et al. 2008). In both in vitro and in vivo studies, it was demonstrated that ZnO NPs could induce oxidative stress by triggering ROS formation (Shahzad et al. 2019;Yang et al. 2020). Induction of oxidative stress causes extensive damage to biomolecules like DNA, proteins, lipids, and cellular membranes (Halliwell 1999). Oxidative stress in the cellular system is generated as a consequence of an imbalance between oxidative and reductive processes and is typically induced when the physiological antioxidant defense system can no longer counteract the elevated ROS level (Zhao et al. 2009;Prieto et al. 2009), or as a result of the cellular incompetency to repair oxidative damages (Dorval 2003). Like other vertebrates, teleosts might have also evolved a wide array of enzymatic and non-enzymatic antioxidant defense systems to convert ROS to harmless metabolites to protect and restore normal cellular homeostasis and functions (Lushchak et al. 2001;Sinha et al. 2015). These include the low molecular weight antioxidants, such as glutathione (GSH), and high molecular weight defenses that have enzymes like superoxide dismutase (SOD), catalase (CAT) as the first line of defense, glutathione metabolism-related enzymes such as glutathione peroxidase (GPx), glutathione reductase (GR) and glutathione-S-transferase (GST), and glutathione reductase (GR) (Valavanidis et al. 2006;Lushchak 2011). Furthermore, the involvement of nitric oxide (NO), generally produced by nitric oxide synthase (NOS), to provide some protection to the cellular system against the oxidative stress has been reported in goldfish (Hansen and Jensen 2010). However, the role of NO in defending against the NPinduced oxidative stress in different cellular systems of teleost fish is yet to be established except the recent report in primary hepatocytes of magur catfish under in vitro condition .
The facultative air-breathing magur catfish (Clarias magur, previously known as Clarias batrachus) is found predominantly in tropical Southeast Asia. They usually live in stagnant and slow-flowing sewage-fed polluted water bodies. They are likely to get exposed to NPs, including the ZnO NPs, in their natural habitats, mainly due to an infestation of industrial wastes and wide-scale anthropogenic activities. The magur catfish is more resistant to various environmental challenges such as high environmental ammonia, hypoxic, and desiccation stresses compared to any other typical teleosts (for reviews, see Saha and Ratha, 1998;. Regardless of previous studies, the defensive strategies against the ZnO NPinduced oxidative stress in air-breathing catfishes and other teleosts have not been well studied. Thus, the primary purpose of the present study was (i) to determine the molecular characterization of SOD and CAT genes that are known to be involved as the first line of defense against oxidative stress; (ii) to investigate the level of oxidative stress caused due to exposure to ZnO NPs by quantifying the indices of oxidative stress such as hydrogen peroxide (H 2 O 2 ), malondialdehyde (MDA), and reduced glutathione (GSH) in different cellular systems of magur catfish; and also (iii) to ascertain the antioxidant defense system in response to ZnO NPs exposure in different cellular systems/organs of magur catfish by analyzing the expression of certain antioxidant genes and their translated products along with the possible involvement of NOS/NO system in such defensive system against the ZnO NP-induced oxidative stress.

Experimental animal
The magur catfish (weighing 140-160 g body mass) belonging to an age group of 18-24 months were purchased from a single source that was bred, cultured in selected commercial ponds of Nilbagan Fish Seed Farm situated in Barpeta, Assam, India. Fishes were acclimatized in the laboratory for approximately 1 month at 27 ± 2 °C with 12 h:12 h light and dark photoperiod before experiments. No sex differentiation was done while performing these studies. Minced dry fish and rice bran (5% of body wt) were given as food every day, and the water, collected from a natural stream, was changed on alternate days. The study was approved by the Institutional Animal Ethics Committee (IAEC) of North-Eastern Hill University, Shillong, India (NEC/IEC/2018/016).

Total RNA extraction and cDNA synthesis
The total RNA was isolated from 50 mg of each tissue using TRI® Reagent (Sigma-Aldrich, St. Louis, USA), following the method of Rio et al. (2010), and quantified spectrophotometrically at 260 nm with the help of QIAxpert (Qiagen, Germany). The first-strand cDNA was synthesized from 400 ng of total RNA in a total volume of 20 μl with an iScript cDNA synthesis kit (Bio-Rad, USA) as per the standard protocol.
Sequencing and structural analysis of three antioxidant genes Degenerate primers for three antioxidant enzyme genes (sod1, sod2, and cat) were designed from the conserved region of published sequences of different fish species. Partial sequencing of each gene was performed in 3130 Genetic Analyzer (Applied Biosystems, USA) using BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA), followed by full-length sequences of sod1, sod2, and cat genes following 5′-and 3′-RACE-PCR with several gene-specific primers using the SMARTer RACE 5′/3′ Kit (Takara Clontech, Japan). All the primers used are described in Supplementary  Table T1. The open reading frame (ORF) was determined using the NCBI ORF finder (https:// www. ncbi. nlm. nih. gov/ orffi nder/). The sequences were submitted to the GenBank with the following accession number: sod1 (KF444052), sod2 (MH891501), and cat (MW752164).
UTR motifs from the untranslated regions (UTRs) of the cDNAs were predicted by the UTRscan tool (Pesole et al. 2000). A Conserved Domain Database search (CD-search) was used for the identification of functional domains of the three antioxidant enzyme genes (sod1, sod2, and cat) (Marchler-Bauer et al. 2011). We used ProtParam (Gasteiger et al. 2005), ScanProsite (de Castro et al. 2006), and Motif Scan (Sigrist et al. 2010) tools on the ExPASy Server for determining the physicochemical properties and motifs of the translated amino acid (aa) sequences. The subcellular localization of the three genes was predicted by the MitoFates webserver (Fukasawa et al. 2015). Homology modeling of the three antioxidant enzymes was performed by SWISS-MODEL workspace (Schwede et al. 2003), and for quality assessment of the protein model, QMEAN4 (Benkert et al. 2011) and PROCHECK (Laskowski et al. 1993) tools were used.

Multiple sequence alignments and phylogenetic analyses
Multiple sequence alignments of the inferred amino acids of the three enzymes from piscine, amphibian, and mammalian sequences using the M-coffee tool were performed (Moretti et al. 2007), followed by using of Gblocks program (Talavera et al. 2007) to eliminate poorly aligned positions and divergent regions of the alignments. Unrooted phylogenetic trees were constructed using the maximum-likelihood principle-based phylogeny with Smart Model Selection (PhyML + SMS) (Lefort et al. 2017) in the NGPhylogeny.fr web server (Lemoine et al. 2019).

Experimental setup
Initially, the magur catfish were exposed to different concentrations of ZnO NPs to determine the 96 h LC 50 value, which was found to be 50 mg/L. Hence, it was decided to perform all the subsequent experiments with five times lesser concentration (10 mg/L) than the LC 50 value, where no mortality was noticed even after 30 days of exposure.
Three sets of fish with 5 fish in each group were kept in three different plastic tumblers and administered to the aqueous suspension of ZnO NPs at a 10 mg/L concentration. Another set of 5 fish was held in another plastic tumbler containing 4 L of bacteriafree filtered stream water, which served as a control. The ZnO NP suspension was prepared in bacteriafree filtered stream water, where the NPs were dispersed by sonication for 6 h in a bath-type sonicator (100 W, 40 kHz). The ZnO NPs suspension and the bacteria-free filtered stream water were changed on alternate days at a fixed time. After 3, 7, and 14 days, 5 fish from treated and 5 control fish were removed and killed by decapitation after anesthetizing in neutralized MS-222 (0.2 g/L) for 5 min. The liver, kidney, brain, muscle, and gills were dissected out and plunged into liquid nitrogen before storing at − 80 °C. All analyses were completed within 3 weeks of collecting tissues. The experimental room temperature was maintained at 26 ± 2 °C with 12 h:12 h day and night photoperiods.
Quantification of H 2 O 2 , MDA, GSH, and NO concentrations For quantification of H 2 O 2 , MDA, GSH, and NO concentrations in different tissues of both the control and treated fish, a 10% homogenate of each tissue were prepared in 50 mM Tris-HCl buffer (pH 7.4) containing a cocktail of protease inhibitor (Roche, Germany) using a motor-driven Potter-Elvehjem glass homogenizer with a Teflon pestle, followed by centrifugation at 10,000 × g for 10 min at 4 °C. The supernatant was used for various measurements. The H 2 O 2 and MDA concentrations in the supernatant were measured by using Amplex® Red Hydrogen Peroxide/Peroxidase Assay Kit (Thermo Fisher, USA) and TBARS assay kit (DTBA-100, BioAssay Systems, USA), respectively, following manufacturers' instructions. The reduced glutathione (GSH) was measured in the supernatant, according to Ellman (1959). The yellowcolored mercapto-2-nitrobenzoic acid (TNB), formed as a product of the reaction between DTNB and GSH, was measured at 412 nm with a UV-visible spectrophotometer (Cary 60, Agilent, USA). To determine the concentration of NO, the supernatant was further treated with 5% perchloric acid (PCA) in a 1:0.5 ratio to precipitate out the proteins, followed by centrifugation again at 10,000 × g for 5 min. The concentration of NO in the supernatant was determined by the Griess reaction described by Sessa et al. (1994). The purple azo product was measured spectrophotometrically at 540 nm with a UV-visible spectrophotometer (Cary 60, Agilent) against the reagent blank. The TBARS and NO concentrations were calculated using MDA and sodium nitrite, respectively, as reference standards.
The protein concentration in different tissues was determined using the Bradford reagent (Bradford 1976).
Enzyme assay A 10% homogenate of each tissue was prepared in a homogenizing buffer containing 50 mM Tris-HCl (pH 7.5), 300 mM mannitol, 1 mM dithiothreitol (DTT), 1 mM ethylenediamine tetra-acetic acid (EDTA), and a cocktail of protease inhibitor (Roche, Germany). The homogenate was treated with 0.5% Triton-X 100 in a 1:1 ratio for 30 min, followed by a mild sonication for 30 s × 2. The homogenates were then centrifuged at 10,000 × g for 10 min, and the supernatant was used for assaying different enzyme activities. All the above steps were carried out at 4 °C.
The superoxide dismutase (SOD) and catalase (CAT) activities were assayed spectrophotometrically following Paoletti et al. (1986) and Beers and Sizer (1952), respectively. One unit of enzyme activity was defined as the amount of enzyme required to cause 50% inhibition of NBT photoreduction for SOD, and the amount of enzyme required to catalyze 1 μmole of H 2 O 2 for CAT per min at 26 °C.
The glutathione-S-transferase (GST), glutathione peroxidase (GPx), and thioredoxin reductase (TrxR) activities were assayed spectrophotometrically following Habig et al. (1974), Chiu et al. (1976), and Holmgren and Bjornstedt (1995), respectively. One unit of enzyme activity was defined as the amount of enzyme required to form 1 μmol of GSH-CDNB conjugate for GST, the amount of enzyme required to transform 1 μmol of NADPH to NADP + for GPx, and the amount of enzyme required to form 1 μmol of TNB for TrxR per min at 26 °C.
The inducible nitric oxide synthase (iNOS) activity was assayed spectrophotometrically following the method of Knowles and Salter (1998) with certain modifications made by Choudhury and Saha (2012). Citrulline, so formed as the reaction product, was estimated spectrophotometrically at 490 nm in a UV-visible spectrophotometer (Cary 60, Agilent, USA) following the method of Moore and Kauffman (1970) and expressed as enzyme activity. One unit of iNOS enzyme activity was defined as that amount of enzyme required to catalyse the formation of 1 μmole of citrulline per h at 26 °C.

Quantitative real-time PCR (RT-qPCR) analysis
The real-time quantitative polymerase chain reaction (RT-qPCR) analysis was performed in a StepOne plus Real-time PCR system (Applied Biosystems, USA) using Premix Ex Taq II SYBR Green (Takara, Japan). The reaction mixture of 10 μl each contained 5 μl of SYBR Green Mix, 0.5 μl of cDNA, 400 nM of each primer, and MilliQ H 2 O. The assay conditions included an initial denaturation step at 95 °C for 30 s, followed by 40 cycles at 95 °C for 5 s and 60 °C for 34 s. The RT-qPCR was performed in triplicate for each sample, and negative controls using no cDNA were run for each gene. Melting curve analysis was used to confirm the amplification of only a single PCR product. The RT-qPCR products were sequenced in ABI 3130 Genetic Analyzer (Applied Biosystems, USA) to re-confirm the amplified product. Two reference genes (actinb and tuba1) were selected, according to Banerjee et al. (2017). Fold changes of each gene were calculated using the modified ΔΔCT method (Livak and Schmittgen 2001). The used primer pairs were designed, and the specificity of each primer pair was checked using the Primer-BLAST tool (Ye et al. 2012) (Supplementary  table T2).

Western blot analysis
Different tissues were homogenized in a lysis buffer (50 mM Tris-HCl pH 7.5, 300 mM sucrose, 1 mM EDTA, 0.1% SDS, 1% Triton X, 1 mM PMSF, protease inhibitor cocktail) and sonicated for 30 × 4 s. Lysates were centrifuged at 10,000 × g for 10 min at 4 °C. For each tissue, 50 μg of cellular protein was resolved in 7.5% SDS-PAGE, followed by transferring the protein bands to the polyvinylidene fluoride (PVDF) membrane. Blots were probed using a specific antibody against the SOD, CAT, GPx, GST, TrxR, and iNOS enzyme proteins (1:5000 dilution), followed by incubation with HRP-conjugated secondary antibodies, and the chemiluminescence was detected using Clarity™ Western ECL substrate (Bio-Rad, USA) in Image Quant LAS 500 system (GE Healthcare Life Sciences, USA). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a protein loading control.

Statistical analysis
The data, collected from different experiments, were statistically analyzed by a one-way ANOVA test to evaluate the significant differences obtained with relation to respective controls using GraphPad Prism software and presented as mean ± SEM (n = 5 in each set of the experiment). Differences with P < 0.05 were regarded as statistically significant. Levene's test was also performed to verify the homogeneity of variance of different parameters between the control and treated groups of each experiment set.

Results
Identification and characterization of three antioxidant enzyme genes Partial sequencing, followed by full-length 5′ and 3′ RACE-PCR of sod1, sod2, and cat produced complete cDNA sequences of these three antioxidant enzyme genes. The sequence of sod1 consisted of 741 nt encoding a 154 aa protein, sod2 consisted of 1033 nt encoding a 224 aa protein. Similarly, the cat sequence consisted of 3187 nt encoding 526 aa protein. The UTRscan tool predicted the presence of an internal ribosome entry site (IRES) in all three genes. Table 1 consisted of the detailed characteristics of full-length cDNAs for the antioxidant enzyme genes.
ScanProsite tool predicted the presence of two motif (copper/zinc superoxide dismutase signature) with a consensus pattern of respectively, in the sod1 aa sequence. The conserved domain search predicted a single conserved domain (Cu-Zn_Superoxide_Dismutase) with several Cu 2+ and Zn 2+ binding sites. In the case of the sod2 aa sequence, a single motif (manganese and iron superoxide dismutases signature) was predicted with a consensus pattern of D- Similarly, the conserved domain search also predicted a single conserved domain (SodA, superoxide dismutase).
Multiple sequence alignment of derived aa sequences of sod1 and sod2 with some piscine, amphibian, bird, and mammalian sequences exhibited conservation of all the motif residues (Figs. 1, 2). The multiple protein alignment of SOD1 (Cu/ Zn-SOD) protein (derived aa sequences of sod1) showed the highest sequence identity (87%) with another air-breathing stinging catfish (Heteropneustes fossilis), followed by two catfishes such as striped catfish (Pangasianodon hypophthalmus) and channel catfish (Ictalurus punctatus) with 86% and 82% identities, respectively, and 83% sequence identity with the Mexican tetra (Astyanax mexicanus). Mito-Fates predicted the presence of a mitochondrial processing peptidase (MPP) cleavage site (26th residue in the cat aa sequence. The conserved domain search also predicted one conserved domain (cata-lase_clade_3) with several heme-binding sites. Multiple sequence alignment of derived aa sequences of cat with some piscine, amphibian, bird, and mammalian sequences exhibited conservation of all the motif residues (Fig. 3). The amino acid sequence alignment Fig. 1 Multiple alignment of amino acid sequences of SOD1. Multiple alignments of deduced amino acid sequences of SOD1 from different vertebrates along with C. magur. Strictly conserved residues are shaded in red exhibited a high sequence identity (96-92%) with the stinging catfish, striped catfish, and channel catfish; 90% sequence identity with the Mexican tetra, zebrafish, and fugu (Takifugu rubripes); 80% identity with the elephant shark (Callorhinchus milii); and 78% identity with the mammalian sequences.

Structural analysis
The ProtParam tool predicted a theoretical pI of 5.94, 8.35, and 8.34 and the MW of 15.9, 25.08, and 59.8 kDa of the derived aa sequences of sod1, sod2, and cat genes, respectively. The homology modeling of SOD1 is predicted as homo-dimer protein. At the same time, SOD2 and CAT were predicted both as homo-tetrameric proteins. The predicted model for SOD1 was based on human-mouse SOD1 chimera (3gtv.1) with 72.37% of sequence homology (Fig. 4a). The predicted model for SOD2 was based on human mitochondrial MN3 + superoxide dismutase (1n0j.1) with 84.3% of sequence homology (Fig. 4a). Similarly, beef liver catalase (7cat.1) with 81.3% sequence homology was used as the CAT model template (Fig. 4a). Structural analysis revealed the putative metal-binding sites of the SOD1 (four residues for Zn 2+ binding: H64, H72, D84, and H81) and SOD2 (four residues for Mn 2+ binding: H52, H100, D185, and H189) (Fig. 4b). The Ramachandran plot of phipsi torsion angles showed that 99.34%, 96.68%, and 94.76% of residues were in the most favored regions of SOD1, SOD2, and CAT models, respectively (Fig. 4c).

Phylogenetic analysis
The molecular phylogenetic analysis of the three antioxidant enzymes was performed using the PhyML algorithm. Phylogenetic analysis of several vertebrate SOD1 and SOD2 enzymes displayed two large clusters, one with the teleost group and another cluster represented the mammalian, amphibian, bird, and elasmobranch (Fig. 5). Furthermore, the phylogenetic analysis of CAT from different vertebrates showed a similar pattern, where all the teleost CAT were clustered together, and another cluster represented the mammalian, bird, and elasmobranch sequences (Fig. 5).
Changes in cellular concentrations of H 2 O 2 , MDA, and GSH Exposure to ZnO NPs led to a sharp rise in the intracellular concentration of H 2 O 2 in different tissues of magur catfish within 3 days, followed by a gradual decrease at later stages, reaching almost to the basal level after 14 days of exposure (Fig. 6a). It increased maximally by 2.3-fold in the liver, 2.2-fold in the kidney, 2.0-fold in the brain and muscle, and 2.2-fold in the gills after 3 days of exposure than in untreated controls. Similarly, a sharp rise of MDA level was also observed in different tissues of magur catfish within 3 days of exposure to ZnO NPs, later followed by a gradual decrease, but, it remained significantly higher than the respective control values even after 14 days of exposure (Fig. 6b). It increased maximally in the liver, kidney, brain, muscle, and gills by 1.7-, 1.8-, 2.7-, 2.6-, 6.5-fold, respectively, after 3 days of exposure compared to untreated controls. Whereas the GSH level in different tissues of ZnO NP-treated magur catfish was found to decrease significantly until 7 days, but on 14 days, it was found to increase considerably in all the tissues (Fig. 6c). It increased significantly by 1.5-, 1.4-, 1.3-, 1.4-, and 1.3-fold in the liver, kidney, brain, muscle, and gills, respectively, after 14 days of treatment.

Changes in activities of antioxidant enzymes
The changes of activities of different antioxidant enzymes in various tissues of magur catfish while exposing to ZnO NPs compared to untreated controls are presented in Fig. 7. The activities of both SOD and CAT enzymes were seen to decrease initially in different tissues within 3 days of treatment, followed by a gradual and significant rise of activities at later stages of treatment. In the case of SOD, the activity increased maximally by 1.3-, 1.5-, 1.4-, 1.9-, and 1.6-fold in the liver, kidney, brain, muscle, and gills, respectively, after 14 days of treatment compared to untreated controls. In the case of CAT, the activity increased maximally in the liver and brain by twofold, followed by in the kidney (1.9-fold), muscle (1.8-fold), and gills (1.6-fold) after 14 days of Fish Physiol Biochem (2021) 47:1909-1932 treatment compared to untreated controls, whereas in the case of GSH metabolism-related enzymes such as the GST, GPx, and TrxR, the activities were found to increase significantly within 3 days of treatment, followed by a further rise of activities at later stages of treatment. Compared to untreated controls, the GST activity increased maximally by 1.8-fold in the liver, followed by in the kidney and muscle (1.6-fold), the brain (1.5-fold), and in the gills (1.4-fold) after 14 days of exposure to ZnO NPs. In the case of GPx, the activity increased maximally in the liver, kidney, brain, muscle, and gills by 1.8-, 1.4-, 2.0-1.7-, and 1.2-fold, respectively, after 14 days of exposure compared to untreated controls. Similarly, the activity of TrxR was also found to increase maximally by 3.1-, 3.0-, 2.5-, 1.8-, and 1.7-fold in the liver, brain, kidney, muscle, and gills, respectively, after 14 days of treatment compared to untreated controls.

Changes in the expression of antioxidant enzyme genes
The relative abundance of mRNAs for different antioxidant genes was measured by RT-qPCR analysis to assess the possible changes of gene expression in various tissues of magur catfish during exposure to ZnO NPs (Fig. 8). In general, the expression of mRNAs for different antioxidant enzyme genes increased significantly in most of the tissues of treated fish compared to untreated controls at variable levels. The maximum increase in mRNA expression for sod1 gene was seen in the muscle and brain by 2.4-and 2.3-fold, respectively, after 3 days, followed by in the kidney (1.9-fold), liver (1.8-fold), and gills (1.6-fold) after 14 days of treatment. The level of sod2 mRNA increased maximally by 3.2-fold in the liver, followed by in the kidney (2.9-fold), gills (2.8-fold), brain (2.6fold), and muscle (2.5-fold) after 14 days of exposure.
In the case of the cat gene, the mRNA expression increased maximally in the brain and gills by 2.5fold, followed by in the kidney (2.4-fold), liver (2.3fold), and muscle (2.1-fold) after 14 days of treatment compared to untreated controls. The expression of mRNA for the gst gene increased maximally in the kidney by 3.4-fold after 7 days, followed by the liver, muscle, brain, and gills by 2.6-, 2.3-, 2.2-, and 1.9fold, respectively. The gpx1 mRNA level increased are plotted as mean ± SEM (n = 5). a,b,c: P values significant at < 0.01, < 0.05 and < 0.001 levels, respectively, compared to respective controls (one-way ANOVA) maximally in the muscle by 3.9-fold, followed by in the brain (3.6-fold), liver (2.3-fold), gills (2.2-fold), and kidney (2.1-fold) after 14 days of exposure. In the case of the txnrd1 gene, the mRNA level increased significantly within 3 days, which later increased further to a maximum level by 3.1-fold in the liver, followed by in the brain (2.6-fold), muscle (2.5-fold), gills (2.1-fold), and kidney (1.9-fold) after 14 days of exposure.

Changes in the expression of antioxidant enzyme proteins
Western blot analysis demonstrated that the exposure of magur catfish to ZnO NPs also led to a significant rise in the expression of different antioxidant enzyme proteins in various tissues compared to untreated controls (Fig. 9a, b). The SOD enzyme protein level increased maximally by 2.7-fold in the kidney, followed by in the liver by 2.6-fold, brain, and gills by 2.3-fold, and in the muscle, it increased by 2.2-fold after 14 days of treatment. The increase of the relative abundance of CAT enzyme protein also occurred in all the tissues examined with a maximum rise by 2.2-fold in the liver, followed by in the kidney (2.1-fold), gills (2.0-fold), muscle (1.8-fold), and brain (1.7-fold) after 14 days. The GST enzyme protein also displayed the highest increase in its expression in the kidney by 2.7-fold after 7 days, followed by the liver, muscle, gills, and brain by 2.4-, 2.1-, 1.9-, and 1.7-fold, respectively, after 14 days of exposure. The maximum increase in the expression of GPx enzyme protein was observed in the muscle, brain, gills, liver, and kidney by 3.2-, 3.1-, 2.2-, 2.5-, and 2.1-fold, respectively, after  (sod1, sod2, cat, gst, gpx1, and txnrd1) in different tissues of C. magur during exposure to ZnO NPs for 14 days. Values are plotted as mean ± SEM (n = 5). a,b,c: P values significant at < 0.01, < 0.05, and < 0.001 levels, respectively, compared to respective controls (one-way ANOVA) 14 days of exposure. Similarly, the expression of TrxR enzyme protein also increased significantly with a maximum rise in the liver, muscle, brain, kidney, and gills by 2.4-, 2.2-, 1.9-, 1.7-, and 1.6fold, respectively, after 14 days of exposure.

Changes in the concentration of NO and iNOS activity
A relatively low level of NO was recorded in different tissues of control magur catfish, which increased significantly at variable levels in multiple tissues within 3 days of exposure to ZnO NPs, followed by a further rise at later stages of exposure (Fig. 10a). It increased maximally in the muscle by 5.0-fold, followed by in the gills, kidney, and liver by 4.5-, 2.2-, and 2.1-fold, respectively, after 14 days, and in the brain, it increased maximally by 1.7-fold after 7 days of exposure. The iNOS activity could not be detected in any of the control fish tissues examined. However, after treating with ZnO NPs, a significant level of iNOS activity could be detected in all the tissues examined within 3 days, which increased further at later stages of treatment (Fig. 10b). While comparing the relative level of activity of iNOS in different tissues, the maximum activity was seen in the kidney (18.3 units/g wet wt), followed by in the liver (12.2 units/g wet wt), brain (5.3 units/g wet wt), gills (4.2 units/g wet wt), and muscle (2.4 units/g wet wt) after 3 days of treatment with ZnO NPs. The iNOS activity enhanced further at later stages of treatment with a maximum rise by 3.3-, 2.6-, 3.4-, 4.0-, and 3.7-fold, respectively, in the liver, kidney, brain, muscle, and gills after 14 days while comparing with the 3 days exposed fish. Changes in the expression of nos2 mRNA and iNOS protein Exposure of fishes to ZnO NPs led to a significant increase in nos2 mRNA expression in different tissues within 3 days, followed by a further rise at later stages (Fig. 10c). The nos2 mRNA level increased maximally in the kidney by 12.4-fold, followed by in the muscle (4.6-fold) and brain (4.4-fold) after 7 days, whereas in the liver and gills, it increased maximally by 9.0-and 8.6-fold, respectively, after 14 days of exposure.
We could not detect any visible immunoreactive iNOS protein band in any of the tissues of untreated control fish as examined by Western blotting. However, after exposing the fish to ZnO NPs, prominent immunoreactive bands of approximately 130 kDa could be detected in all the tissues with a further rise of band intensities at later stages of exposure (Fig. 10d). Densitometric analysis of protein bands revealed that the expression of iNOS protein increased maximally in the kidney by 3.7-fold, followed by in the liver, muscle, gills, and brain by 3.4-, 1.7-, and 1.6-fold, respectively, after 14 days of exposure to ZnO NPs (Fig. 10e).

Discussion
In the present study, we report here the full nucleotide and the translated amino acid sequences of three antioxidant enzyme genes, such as the sod1, sod2, and cat, and also some of the molecular characteristics of these genes in magur catfish. Unlike highly structured introns, the UTRs of mRNAs are not generally evolved to adopt single, well-defined structures; however, UTRs contain several gene regulatory elements (Barrett et al. 2012). These UTRs mainly control the translation and RNA decay and act as targets for RNA interference (Halvorsen et al. 2010). The presence of IRES motifs suggests an alternative ribosomal binding site for translation initiation (Le and Maizel 1997), and all the antioxidant enzyme genes (sod1, sod2, and cat) contained this motif. The Musashi protein is an evolutionarily conserved protein necessary for regulating the temporal order of mRNA translation (Charlesworth et al. 2006). The presence of MBE-box in sod1, sod2, and cat suggests that the Musashi protein remains bound to these genes and regulate their translational processes.
The crystal structure of human-mouse SOD1 chimera (3gtv.1), human mitochondrial MN3 + superoxide dismutase (1n0j.1), and beef liver catalase (7cat.1) with very high sequence homologies were identified as the best template for magur catfish SOD1, SOD2, and CAT, respectively, in homology modeling by the Swiss-Model workspace. The amino acids, coordinated for the metal ion binding in SOD1 and SOD2, were found to be highly conserved across species (Lin et al. 2009;Umasuthan et al. 2014). The three antioxidant enzyme protein models' Ramachandran plots showed that the Phi-and Psi-angles for most residues are in the most favored region, demonstrating a highquality model. These findings strongly indicated that the three-dimensional structures of three antioxidant enzyme proteins are like their mammalian counterparts, suggesting the functional similarities of these enzymes in antioxidant defense.
The molecular phylogenetic analysis of SOD1 and SOD2 displayed two large clusters, where all the teleost SODs, including the magur catfish, belonged to one cluster, and mammalian, amphibian, and elasmobranch were in another cluster. The phylogeny of CAT also showed a similar pattern. In all the phylogenetic trees, the nearest neighbor of the magur catfish is another catfish, H. fossilis. Two other catfish (I. punctatus and P. hypophthalmus) were also present as neighbors. Indeed, it is very much evident that a close relationship exists between these two antioxidant enzymes of air-breathing magur catfish with other freshwater catfishes, and to some extent, with other teleosts rather than those of lungfishes and cartilaginous fishes. Moreover, the sequence alignments of amino acids of these three vertebrate antioxidant enzyme proteins showed a high degree of sequence homology between the magur catfish and the other bony fishes (average of 92%). A complete amino acid conservation in functional domains across the species was also observed. Furthermore, the alignments demonstrated complete conservation of active site residues in all the three antioxidant enzyme proteins among piscine, amphibian, and mammalian species, which appeared to be justified since the active site residues are required for the functionality of enzymes.
The results from the present study clearly demonstrated that exposure of magur catfish to a sub-lethal dose of ZnO NPs (10 mg/L) resulted to an oxidative stress, as evidenced by a sharp increase in the intracellular concentrations of both H 2 O 2 and MDA at variable levels in different tissues examined. The excess of H 2 O 2 (and O 2 ֗ ‾), which can be transformed to form highly reactive oxidant hydroxyl radicals (֗ OH) via the Harber-Weiss reaction, generally leads to lipid peroxidation through degradation of polyunsaturated fatty acids (PUFAs). Hence, we could also observe a significant rise in MDA concentration, the oxidative end product of PUFAs, in different tissues of magur catfish examined in response to ZnO NPs exposure with a peak of increase within 3 days, followed by a gradual decrease at later stages of treatment. In general, the nano form of various metal ions produces more ROS than their bulk counterparts in cellular systems due to their unique physicochemical characteristics (Fu et al. 2014). Similar to our findings, elevated production of ROS and subsequent oxidative damages by ROS during exposure to nano-scale of TiO 2 and ZnO have also been reported in zebrafish (Xiong et al. 2011). Several studies on mechanistic pathways illustrating the significant role of dissolved intracellular Zn 2+ in ROS production and subsequent oxidative stress during exposure to ZnO NPs have also been reported (Kao et al. 2012;Buerki-Thurnherr et al. 2013). Similarly, ZnO particle size-dependent cellular damages have also been reported in Nile tilapia and the increase in the levels MDA in different tissues due to increased production of ROS (Kaya et al. 2016). Thus, it is apparent that more accumulation of MDA in various tissues of magur catfish at variable levels, observed during in situ exposure to ZnO NPs, was associated with elevation of lipid peroxidation in response to NP-induced oxidative stress.
GSH is one of the most abundant intracellular antioxidant thiols, which is known to act as a substrate or cofactor for various enzymatic reactions of the glutathione-dependent cycle, decomposition of H 2 O 2 to water, and in the reduction of lipid peroxidation.
Further, it is known to get involved in cellular redox homeostasis by quenching superoxides and is central to defensive mechanisms against toxic agents and oxidant-mediated injury (DeLeve and Kaplowitz 1991; Townsend et al. 2003). Our present study observed that the GSH level decreased at the initial stage of treatment with ZnO NPs in all the tissues of magur catfish examined, but increased at later stages with a peak of increase after 14 days of exposure. The initial depletion of the GSH level was probably associated with the rise of ZnO NP-induced ROS production in different tissues. A drastic drop in the level of GSH at the early stages of exposure indicated a condition of oxidative stress inside the cell as a consequence of imbalance in the ROS formation and antioxidant defense system within the cellular system (Guan et al. 2012). This observation is consistent with the reports that GSH is consumed in many ways, such as oxidation, conjugation, hydrolysis, and GSH can be directly oxidized by ROS and RNS or indirectly during GSH-dependent peroxidase-catalyzed reactions (Kohen and Nyska 2002;Lushchak 2012). Similarly, Sarkar et al. (2017) also demonstrated the depletion of GSH levels in zebrafish liver and kidney under arsenic trioxide-induced oxidative stress. Similar to our findings, exposure to silver nanoparticles (Ag-NPs) was reported to cause an initial decrease in GSH level in brain tissue of two fishes (O. niloticus and Tilapia zillii), followed by a gradual increase of GSH level at later stages (Afifi et al. 2016). Thus, the initial depletion of GSH level in different tissues of magur catfish, observed during exposure to ZnO NPs, was probably associated with to quench the superoxides, followed by a significant elevation of its level at later stages due to induction of GSH synthetic pathway, and also other antioxidant pathways (discussed later) in different tissues of magur catfish mainly to defend against the ZnO NP-induced oxidative stress.
Various antioxidant defense systems have been evolved in different organisms to fight against the free radical-mediated cellular damages and also to maintain redox homeostasis. Out of those, specific enzymatic systems are known to work as antioxidants in different cellular systems; these include the SOD/CAT enzymes as the first line of defense and also certain glutathione-dependent enzymes like the GST, GPx, and GR (Birben et al. 2012). In the present study, we observed an initial decrease of SOD and CAT activities in different tissues within 3 days of exposure to ZnO NPs, thus causing a higher accumulation of free radicals, and hence an increased production and accumulation of H 2 O 2 and MDA in various tissues. But, the activities of both the enzymes increased significantly at later stages of exposure to ZnO NPs at variable levels in different tissues examined, which eventually were able to scavenge the elevated levels of ROS and ameliorate the ZnO NP-induced oxidative damages to a certain extend, thereby reducing the levels of H 2 O 2 and MDA in different tissues at later stages. Other than the SOD and CAT, we also observed a gradual increase in the activities of glutathione-dependent antioxidant enzymes like GST and GPx in different tissues during exposure to ZnO NPs. Exposure to ZnO NPs also caused an increase in the activity of thioredoxin reductase (TrxR) in various tissues, which we speculated as an alternative to glutathione reductase (GR) for the maintenance of the GSH pool in this magur catfish. The transcriptome analysis of this magur catfish revealed the absence of any gene for the GR enzyme, but the presence of TrxR (txnrd1) gene . Further, we also could not detect the GR activity in any of the tissues of magur catfish examined. The annotated genome sequence of a closely related channel catfish (GCF_001660625.1) also revealed the absence of any gene for GR (Liu et al. 2016). Studies in Drosophila melanogaster and Saccharomyces cerevisiae demonstrated the role of TrxR as a substitution for GR (Kanzok 2001;Tan et al. 2010). Sakurai et al. (2005) also reported that TrxR is an essential component of the Trx system in the cytosol of bovine endothelial cells, which is induced under oxidative stress.
Interestingly, the increase of antioxidant enzyme activities was accompanied by a significant rise in the expression of all the antioxidant enzyme genes (sod1, sod2, cat, gst, gpx1, and txnrd1) during exposure to ZnO NPs, as evidenced by more abundance of mRNA levels of all the genes at variable levels in different tissues of magur catfish. Out of two isoforms of SOD genes, sod2 (mitochondrial) was found to get more induced than the sod1 (cytosolic) in all the tissue examined, similar to the observation made in mammals under a variety of intracellular and environmental cues (Zelko et al. 2002). The whole transcriptome analysis, carried out under ammonia-induced oxidative stress in the liver and brain tissues of magur catfish, also demonstrated the possible induction of a set of antioxidant genes . The increase in concentrations of mRNAs for different antioxidant genes was also supplemented by a significant rise in the abundance of various antioxidant enzyme proteins at variable levels in all the tissues examined during exposure to ZnO NPs. Thus, it may be contemplated that both the transcriptional and translational activation of different antioxidant genes took place for the active scavenging of free radicals and also to counteract the redox imbalance in this catfish under NP-induced oxidative stress. However, it is to be noted that the increase of mRNA levels of most of the antioxidant genes was relatively higher compared to the rise in enzyme protein levels and also the enzymatic activities under NP-induced oxidative stress. Hence, it may be contemplated that some translational and post-translational regulatory mechanisms are existing in this magur catfish for regulation of antioxidant enzymes. Nonetheless, the present findings clearly demonstrated that the magur catfish has the ability to defend itself against the ZnO NPinduced oxidative stress by inducing different antioxidant-related genes as a unique biochemical adaptational strategy.
More recently, the antioxidant activity of NO against the ammonia-and ZnO NP-induced oxidative stress has been demonstrated in primary hepatocytes of magur catfish Hasan et al. 2020). Therefore, in the present study, we also looked at the possibility of induction of NO synthesis via the NO/NOS system in various tissues of magur catfish while exposing to a sub-lethal dose of ZnO NP. Interestingly, exposure to ZnO NPs also led to a continuous increase of NO concentration at variable levels in different tissues throughout the experiment, which was supplemented by the stimulation of iNOS activity, up-regulation of the nos2 gene expression, and more abundance of its translated enzyme protein level at variable levels in different tissues of magur catfish. In our previous studies, we have fairly demonstrated that the induction of both SOD and CAT genes under ammonia-and NP-induced oxidative was mediated by the NO, the production of which was stimulated via the NFκB-mediated induction of NO/ NOS system as an antioxidant strategy Hasan et al. 2020). The induction of NO synthesis via the NOS/NO system has also been reported in non-hepatic tissues of this magur catfish under hyper-ammonia stress, and also while treating with lipopolysaccharide (Choudhury et al. 2018;Kumari et al. 2019). Therefore, it is logical to think that NO is also playing a vital role in defending against the NPinduced oxidative stress in various tissues of magur catfish mainly by stimulating the SOD/CAT system. However, one has also to look into the involvement of NO in the induction of GSH pathway-related genes in this catfish under NP-and probably under other environmentally induced oxidative stress.

Conclusion
The sequence alignments of amino acids of three antioxidant enzyme proteins (SOD1, SOD2, and CAT) of magur catfish, deduced from nucleotide sequences, showed a high degree of sequence homology with that of other bony fishes along with complete conservation of active site residues in all the three enzyme proteins among piscine, amphibian, and mammalian species. The three-dimensional structures of these three antioxidant enzyme proteins were found to be similar with their mammalian counterparts, thereby suggesting the functional similarities of these enzymes in antioxidant defensive mechanisms. Furthermore, it is strongly believed that the magur catfish has the ability to reduce the ZnO NP-induced oxidative stress by inducing the SOD/CAT enzymatic system and also the GSH pathway-related enzymes as a consequence of induction of related genes both at the transcriptional and translational levels in various tissues. Our results further suggest that the induction of SOD/CAT system in various tissues was mediated through NO; the production of which was stimulated due to induction nos2 gene under NP-induced oxidative stress as one of the major antioxidant strategies. Thus, it may be contemplated that the magur catfish possesses a very efficient antioxidant defensive mechanisms to defend against the NP-induced oxidative stress and also from related cellular damages.