Introduction

Muscle is the principal consumable organ in the teleost fresh water Indian stinging fish Heteropneustes fossilis. This fish is well known in India for nutritional as well as medicinal values of its flesh (Bailly 2013; Erfanullah and Jafri 1999). Therefore, this fish represents one of the most important fishery resources in India (Bailly 2013; Menon 1999). The majority of its habitats such as ponds, ditches, marshes and muddy rivers dry in summer season due to prolonged drought (Bailly 2013; Menon 1999). Nevertheless, H. fossilis can hide itself in own burrows and live at least for 30 days inside mudflats to avoid total dehydration in summer season (Saha et al. 2001; Talwar and Jhingran 1991). H. fossilis can also survive up to 20–70 h under air exposure conditions (Paital 2013; Saha and Ratha 1989). Air exposure can initiate dehydration and hypoxia condition in air breathing marine and fresh water fishes (Bailly 2013; Bridges 2013; Graham 1997; Talwar and Jhingran 1991; Thurston and Gehrke 1993).

During commercial handling or in their natural habitat in summer season (inside mudflats), water running through gills of H. fossilis becomes interrupted. This can limit O2 supply through gills of the fish (Munshi et al. 1978). As a result, the tissues of the fish may respond differentially to oxidant (reactive oxygen species: ROS) generation at compromised O2 availability condition (Paital 2013). Limited O2 supply due to change in abiotic environmental factors to alter oxidative stress (OS) pathways in tissues of aquatic animals has already been noticed (Paital and Chainy 2010, 2012, 2013a; Romero et al. 2007). Reactive oxidants or ROS such as superoxide radical (O ·−2 ), H2O2 and hydroxyl radical (·OH) are generated as by-product in normal oxidative metabolism. Under non-stressed conditions, about 1–2 % of O2 consumed by mitochondria are incompletely reduced to O ·−2 by the electrons that are leaked at respiratory complex I and III enzymes of electron transport chain (ETC) (Abele et al. 2007; Halliwell and Gutteridge 2001). Eventually, O ·−2 becomes responsible for production of the other ROS such as H2O2 and the highly reactive ·OH (Halliwell and Gutteridge 2001). All the above molecules have high oxidant activity and if not neutralized, can oxidize biological macromolecules including lipids, proteins and nucleic acids. O ·−2 is catalyzed to H2O2 by superoxide dismutase (SOD, EC 1.15.1.1) (Paital et al. 2013). H2O2 is neutralized to H2O and O2 by the enzyme catalase (CAT, EC 1.11.1.6) (Paital et al. 2011). H2O2 is also broken down by glutathione peroxidase (GPx, EC 1.11.1.9) enzyme in presence of reduced glutathione (GSH). The oxidized glutathione (GSSG) is reduced back to GSH by the enzyme glutathione reductase (GR) with the help of NADPH. ·OH and other ROS are neutralized by small molecular weight non-enzymatic antioxidants such as ascorbic acid (AA), GSH, non-protein sulfhydryl (–SH), vit-E and vit-C. Under normal physiological conditions, balance is maintained between ROS and antioxidants in cells of aerobic organisms. Lower level of ROS especially H2O2 can be also useful for animals due to its role in various signal transduction processes (Takano et al. 2003). Insufficient level of antioxidants can be resulted in increased lipid peroxidation (LPx), protein oxidation (carbonylation: PC), nucleic acid adducts and finally, a disorder condition known as OS (Abele et al. 2011; Halliwell and Gutteridge 2001).

Exploring the mechanisms of air exposure and hypoxia-induced lipid and protein oxidation in consumable aquatic animals can be useful in aquaculture industry. At environmental low oxygen level, species-specific biochemical adaptation mechanisms are proposed for long-term survival in aquatic animals (Welker et al. 2013). For example, the long-term survival of common carp Cyprinus carpio in hypoxia condition is correlated with the high glycogen level and exclusively anaerobic glycolysis-mediated ATP generation in tissues (Lushchak et al. 2005). At certain conditions, induction of metabolic depression also contributes for long-term survival of aquatic animals under low O2 level or anoxia state (Storey and Storey 2004). On the other hand, metabolic damage by high ROS level at hypoxia and constitutively enhancement of antioxidant defenses associated with hypoxia followed by regeneration to restore the required cellular redox level in tissues are recorded in fishes (Lushchak et al. 2001, 2005; Lushchak and Bagnyukova 2006). This particular physiological response in animals now has been referred as “preparation for oxidative stress” (Welker et al. 2013). Although this is a well-studied phenomenon in non-air breathing fishes such as goldfish Carassius auratus and common carp C. carpio (Lushchak et al. 2001, 2005; Lushchak and Bagnyukova 2006), it is not yet well established in air breathing fishes such as H. fossilis. Theoretically, consistent increase in hypoxia condition induced by constant exposure of air breathing fishes to air may also be associated with dynamic change in their cellular oxidative metabolism. However, studies describing a complete picture of redox metabolism under air exposure conditions are limited in any air breathing fish models in general and H. fossilis in particular (Maiti et al. 2010; Zaccone et al. 1985). Recently, oxidative metabolic responses in one of the most metabolically active tissues, i.e., in brain of air breathing fish H. fossilis in relation to long-term survival under dehydration state is reported (Paital 2013). The present study aimed to analyze the biochemical level of redox metabolism associated parameters in one of the low metabolically active tissue, i.e., in muscle of H. fossilis under air exposure condition. This is also the first report describing a composite picture of ROS, OS and antioxidant parameters along with mitochondrial respiratory complex enzyme activity in muscle tissue of H. fossilis. The data can be useful to understand the fish muscle physiology under hypoxia state as air exposure can limit O2 supply in fishes.

Materials and methods

Chemicals

Chemicals for the present experiment were purchased locally from Sigma Chemical Company, Missouri, USA (cumene hydroperoxide, dithiobisnitrobenzoic acid, homovanilic acid, horse radish peroxidase, GR, thiobarbituric acid and duroquinone), Himedia, Mumbai, India (riboflavin, phenylmethane sulphonyl fluoride, hydroxylamine hydrochloride, l-methionine, GSSG, GSH, NADPH, NADH, butylated hydroxytoluene and trichloro acetic acid) and Sisco Research Laboratory, Mumbai, India (analytical grade of all other chemicals).

Animals, experimental set-up and tissue collection

The tissue for the present study was collected and analyzed from the earlier experiment (Paital 2013). Briefly, H. fossilis (15–20 g) was purchased from the local fish market (Chowkaghat, Varanasi, India) and were disinfected by dipping them in 500 ppm potassium permanganate solution for 2–3 min. Fishes were then washed adequately under tap water and were released to plastic tanks containing fresh tap water and were acclimatized in laboratory conditions in 12: 12 h light and dark period. Opaque water tanks were used to make least disturbances to the animals. Water (having dissolved O2 7.1 ± 0.45 mg L−1, pH 7.13 ± 0.12, water hardness 24.5 ± 2.19 mg L−1 and water temperature 25.12 ± 3.21 °C) of the tank was changed in every alternative day. Fishes were fed with minced goat liver at every alternative day but on the day water was not changed. Fishes were acclimatized in the above laboratory conditions for a month. During acclimatization around 10 % mortality rate was observed up to the first 5 days which was arrested thereafter.

Only active male fishes (23.44 ± 1.63 g) were selected for experiment. Fishes were randomly divided into five groups (n = 10 in each) and four groups of fish were transferred onto four plastic trays (40 × 30 × 6 cm3) and were exposed to air at 25 °C. Care was taken to have no significant difference in their body weight among the groups. Tap water was sprayed on them with 3 h interval to avoid total dryness of their skin. Fish were then sacrificed after exposing them to air for 3, 6, 12 and 18 h. After 20 h exposure, fish were started to die and >90 % were died at 24 h exposure time. After 18 h air exposure, the revival of animals was observed when they get back into water. Therefore, maximum air exposure time was set to 18 h. One group of fish (n = 10) was sacrificed immediately after collecting them from the water tank. This group of fish was treated as control for the rest of the air-exposed groups. Recovery or reoxygenation (submerging fish into water after air exposure) experiment was out of the objective and scope of this article and, therefore, was not performed. After sacrificing the fishes by cryoanesthesia, muscle tissue was dissected immediately, washed in ice-cold normal saline (0.78 %, w/v) solution, blotted, weighed and flash frozen in liquid nitrogen. Then the tissues were stored at −20 °C for further analyses.

Tissue processing and biochemical analyses

A 10 % (w/v) homogenate of muscle tissue was prepared at 4 °C in homogenizing buffer (50 mM tris, 1 mM EDTA, 0.25 M mannitol and 1 mM phenylmethane sulphonyl fluoride, pH 7.6) with the help of pre-cooled mortar and pestle by keeping on ice. The crude homogenate was subjected to centrifugation at 1,500×g for 15 min at 4 °C in a cooling centrifuge (REMI, Mumbai, India, model: CPR-24) to sediment nuclei and tissue debris. The crude supernatant obtained was referred as post-nuclear fraction (PNF). PNF was subjected to centrifugation at 14,000×g for 15 min at 4 °C to obtain the clear supernatant which was referred as post-mitochondrial fraction (PMF). Succinate dehydrogenase activity was measured in PMF to check its purity (Lambowitz 1979). The mitochondrial pellet (mitochondrial fraction: MF) obtained at the above step was washed twice in homogenizing buffer and was used to assay activities of mitochondrial complex enzymes. Protein concentration in samples was determined according to Bradford (1976) using bovine serum albumin as standard.

Determination of OS and ROS levels

For measurement of LPx, muscle tissue was homogenized in 1.15 % ice-cold KCl to obtain PNF as described above. LPx was measured in the form of thiobarbituric acid reactive substances (TBARS) according to Ohkawa et al. (1979). TBARS (nmol mg−1 pr) were calculated from its molar extinction coefficient as 1.56 × 105 M−1 cm−1. Protein carbonyl content was determined in PNF of the tissue (Levine et al. 1994). Carbonyl content (nmol carbonyl mg−1 pr) was calculated from the molar extinction coefficient of aliphatic hydrazones, i.e., 22 × 103 M−1 cm−1. H2O2 content was measured fluorimetrically in PNF samples according to Staniek and Nohl (1999) and modified by Paital and Chainy (2010). Briefly, PNF sample was immediately precipitated with 5 % (w/v) trichloroacetic acid and centrifuged at 12,000×g. The supernatant was neutralized to ~pH 7.4 with 6 M K2HPO4. H2O2 was measured fluorimetrically in the neutralized supernatant using a method in which horseradish peroxidase-mediated homovanillic acid dimmer formation by H2O2 is established. Result was compared with the standard curve of H2O2 made in similar way to the above and was expressed as ng of H2O2 per g of wet tissue. The methods for the above parameters are described in details earlier (Paital and Chainy 2010).

Antioxidant enzyme assays

Activity of antioxidant enzymes was measured at 25 °C in PMF samples as described in details earlier (Paital and Chainy 2010). SOD (EC 1.15.1.1) activity was measured according to Das et al. (2000) and result was expressed as unit mg−1 protein. CAT (EC 1.11.1.6) activity was measured by monitoring the decrease in absorbance of 25 mM H2O2 at 240 nm in 3 mL 50 mM phosphate buffer, pH 7 (Aebi 1974). Enzyme activity was calculated from the extinction coefficient of H2O2 as 43.6 M−1cm−1 and was expressed as one mole of H2O2 consumed s−1 mg−1 protein (nKat mg−1 pr). GPx activity (EC 1.11.1.9) was measured according to Paglia and Valentine (1967) in which cumene hydroperoxide (7.5 mM) was used as substrate. GR (EC 1.6.4.2) activity was measured according to Massey and Williams (1965). For GR, the rate of conversion of GSSG (120 mM) to GSH was estimated by monitoring oxidation of NADPH (4.5 mM) in the assay system. For both GPx and GR, utilization of NADPH (4.5 mM) by glutathione system was recorded at 340 nm. Results of both the enzymes were calculated from the extinction coefficient of NADPH as 6.22 mM−1 cm−1 and expressed as nmol of NADPH oxidized min−1 mg−1 pr.

Determination of non-enzymatic antioxidants

AA and –SH group were measured in supernatant (14,000×g, 15 min) of trichloroacetic acid (5 %, w/v) precipitated PNF. AA was estimated according to Mitsui and Ohta (1961) by monitoring the reduction of phosphomolybdate by AA in acidic medium at 660 nm. Results were calculated from the standard curve of AA (2.5–50 nmol) and were expressed as nmol of AA g−1 wet tissue. Reaction product of –SH content of sample with Ellman’s reagent (0.01 M dithiobisnitrobenzoic) was measured at 412 nm (Sedlak and Lindsay 1968). Results were calculated from the standard curve of GSH (5–50 nmol) and expressed as nmol of –SH g−1 wet tissue.

Assay of electron transport chain enzymes

MF was placed in an ice bath in aliquots and sonicated at a protein concentration of ~1 mg/100 μl. Sonication was done by giving four bursts for 15 s with each burst being followed by an interval of 30 s. Sonicated samples were centrifuged at 10,000×g for 10 min at 4 °C to separate unbroken mitochondria. Activities of respiratory chain enzymes were measured in the obtained supernatant at 25 °C. The detailed protocols for estimating ETC enzyme (complex I. II, II–III and V) level are described in Paital and Chainy (2012).

Briefly, activity of NADH: duroquinone oxidoreductase (complex I, EC 1.6.5.3) was assayed by monitoring the decrease in absorbance (at 340 nm) due to reduction of NADH (13 mM) in the assay mixture (Gassner et al. 1997). Enzyme activity was calculated using molar extinction coefficient of NADH as 6.22 × 103 M−1 cm−1 and result was expressed as nmol NADH oxidized min−1 mg−1 protein. Activity of succinate: 2, 6-dichlorophenolindophenol oxidoreductase (complex II, EC 1.3.99.1) was assayed by monitoring the reduction rate of dichlorophenolindophenol (1.125 mM) at 600 nm (Lambowitz 1979). Complex II activity was expressed as unit min−1 mg−1 protein. Succinate: cytochrome c oxidoreductase (complex III, EC 1.3.2.1) activity was assayed by following the reduction rate of oxidized cytochrome c (1 %, w/v) at 550 nm (Tisdale 1967). Enzyme activity was calculated using molar extinction coefficient for cytochrome c (21 × 103 M−1cm−1) and expressed as nmol cyt-c reduced min−1 mg−1 protein. ATPase (complex V, EC 3.6.1.34) activity was assayed according to Cormier et al. (2001) and Chen et al. (1956). First, externally supplied ATP molecules (50 mM) were hydrolyzed by MF sample and then iP produced were measured using a color reagent (6 N H2SO4, 2.5 % ammonium molybdate and 10 % AA in the ratio of 2:1:1, v/v). Absorbance of the color product was read at 820 nm and KH2PO4 was taken as standard. Complex V activity was expressed as μmol iP formed h−1 mg−1 protein.

Statistical analyses

Each set of data (mean ± S.D., n = 10) were tested for homogeneity of variance (Levene test) and normality (Lilliefors test). Means were compared and analyzed by one-way analysis of variance (ANOVA) followed by Duncan’s new multiple range test for normally distributed data. A non-parametric test (Kruskal–Wallis ANOVA) was used for the data which did not reveal homogeneity of variance and normal distribution (CAT, GPx, AA, and GSH). Differences among the means were considered significant at p < 0.05 level. Correlation coefficients (r) between OS parameters and air exposure time were determined at 5 % significant level using Microsoft excel program. Discriminant function analysis (DFA) was performed according to Garson (2008) and Jennrich (1977), to evaluate the contribution of the variables (OS physiology parameters and ETC enzymes) on the groups. Mean values of the biochemical parameters between the groups were compared to calculate the absolute change in percentage. Loss in body weight of the fish was compared by two-tailed paired t test using Microsoft excel program.

Results

Change in body weight under air exposure condition

A significant (p ≤ 0.05, two-tailed t test) loss of 5, 9, 14 and 15 % in body weight of the fish was recorded after air exposure than before exposure time points at 3, 6, 12 and 18 h, respectively (data not shown). The loss in body weight observed at 3 h and onwards air exposure time in H. fossilis indicates induction of dehydration condition in the fish.

Effects of dehydration on levels of OS and ROS indices

TBARS level in muscle of H. fossilis was increased after 3 h air exposure (Fig. 1). TBARS value was equally the highest in 6, 12 and 18 h groups in comparison to 0 and 3 h groups. From the control set, 100, 232, 243 and 257 % higher (p ≤ 0.001, one-way ANOVA, Fig. 1a) TBARS level was recorded at 3, 6, 12 and 18 h exposure groups, respectively. PC value was increased and maintained at least ≥7 % higher (one-way ANOVA, p = 0.05, Fig. 1b) level than control and 3 h groups in comparison to 6, 12 and 18 air exposure groups. No difference in PC level was observed between 0 and 3 h groups and among 6, 12 and 18 h groups. H2O2 was increased in the tissue at 3 h group than control set and then was maintained equally in rest of the groups. At least 120 % higher (one-way ANOVA, p ≤ 0.001, Fig. 1c) H2O2 level was recorded in air exposure groups than control set. The tissue was able to retain equal level of H2O2 at 3 h onwards air exposure groups than control group.

Fig. 1
figure 1

Effects of air exposure on oxidative stress indices and ROS (H2O2) level in muscle of the teleost Heteropneustes fossilis. a Lipid peroxidation level, b protein carbonylation level, c H2O2 level. Data are presented as mean ± SD. (n = 10). Different superscripts indicate statistical (one-way ANOVA followed by Duncan’s post hoc test) difference between the mean values at p ≤ 0.05

Full size image

Antioxidant enzyme responses to air exposure condition

Air exposure has caused a significant decrease in activities of the studied four antioxidant enzymes in muscle of H. fossilis (Fig. 2). Dehydration up to 18 h was found to responsible to diminish SOD activity to the lowest level (at least 55 % in comparison to the other air-exposed groups, p ≤ 0.001, one-way ANOVA, Fig. 2a) in the tissue of the teleost fish. On the other hand, SOD had at least 28 % higher (p ≤ 0.05, one-way ANOVA) activity in the control set in comparison to the other air exposure groups. From the control set, 37, 47, 47 and 55 % lower (p ≤ 0.01, one-way ANOVA) SOD activity was observed at 3, 6, 12 and 18 h air exposure groups, respectively. CAT activity initially was found to decrease linearly from 0 h group up to 12 h group and then remained unchanged at 18 h group than 12 h group. Its activity was 93, 136, 202 and 228 % higher (p ≤ 0.001, Kruskal–Wallis ANOVA, Fig. 2b) at 3, 6, 12 and 18 h groups than control set. GPx activity was decreased at least by 39 and by 22 % (p ≤ 0.01, Kruskal–Wallis ANOVA, Fig. 2c) from the control set at 3, 6 and 12 h and at 18 h air exposure groups, respectively. A mild induction (at least 28 % higher, p ≤ 0.01, Kruskal–Wallis ANOVA) in GPx activity level was observed at 18 h group in comparison to 3, 6 and 12 h groups. The biochemical activity of GR was found to be reduced due to dehydration (Fig. 2d). Air exposure up to 3, 6 and 12 h and 18 h was responsible to reduce (p ≤ 0.001, one-way ANOVA) GR activity almost double and triple times than 0 h group, respectively. A 54, 54, 53 and 67 % (p ≤ 0.001) reduction in GR activity was observed in 3, 6, 12 and 18 h exposure groups than 0 h group, respectively. After reduction of GR activity at 3 h than 0 h group, its level was maintained unchanged up to 6 and 12 h groups then again reduced at 18 h group.

Fig. 2
figure 2

Activities of antioxidant enzymes in muscle of the teleost Heteropneustes fossilis under air exposure condition. a Superoxide dismutase, b catalase, c glutathione peroxidase d glutathione reductase. Data are presented as mean ± SD. (n = 10). Different superscripts indicate statistical (one-way ANOVA followed by Duncan’s post hoc test, Kruskal–Wallis ANOVA for CAT and GPx only) difference between the mean values at p ≤ 0.05

Full size image

Change in levels of small antioxidant molecules

Dehydration states seem to have no effects on AA level in muscle of H. fossilis up to 12 h exposure time (Fig. 3a, p > 0.05, Kruskal–Wallis ANOVA). However, ~58 % (p = 0.05) higher AA level was observed at 18 h group in comparison to 3 and 6 h groups. Free non-protein –SH concentration in the muscle was linearly increased (p = 0.05, Kruskal–Wallis ANOVA, Fig. 3b) from 3, 6 and 18 h groups with an insignificant change in its level between 3 and 12 h and between 0 and 12 h groups. Free –SH accumulation in the tissue was observed at 18 h (at least 51 % higher, p = 0.05) in comparison to the other groups. It was ~156 % higher (p = 0.01) at 0 and 6 h groups in comparison to 3 and 12 h groups.

Fig. 3
figure 3

Variation of non-enzymatic antioxidant molecules in muscle of the teleost Heteropneustes fossilis as a function of air exposure time. a Ascorbic acid and b non-protein sulfhydryl group (−SH). Data are presented as mean ± SD. (n = 10). Different superscripts indicate statistical (Kruskal–Wallis ANOVA) difference between the mean values at p ≤ 0.05

Full size image

Effects of air exposure on activities ETC enzymes

ETC enzymes and ATPase in muscle tissue of H. fossilis have shown significant changes in their level in response to dehydration condition (Fig. 4). Complex I level was increased linearly up to 6 and 12 h groups then again deceased at 18 h group. Activity of complex I was significantly (one-way ANOVA) increased by 37 (p = 0.05) and 74 % (p = 0.01) from the control set at 3 and 6 h groups, respectively, followed by no change between 6 and 12 h groups (Fig. 4a). Its activity was reduced by ~1.5 times at 18 h group than 6 h group. A linear increase in complex II activity was recorded with respect to time of air exposure (Fig. 4b). It had 31, 70, 93 and 127 % (one-way ANOVA, p ≤ 0.05) increased activity at 3, 6, 12 and 18 h groups than 0 h group, respectively. Reduction in activity of both complexes III and V was observed in response to air exposure of the fish. Reduction of 28 (p = 0.05), 25 (p = 0.05), 55 (p = 0.01) and 70 % (p = 0.001) (one-way ANOVA for all p values) of complex III activity was observed at 3, 6, 12 and 18 h air exposure time than 0 h, respectively (Fig. 4c). Similarly, ATPase activity was diminished linearly (Kruskal–Wallis ANOVA) by 33 (p = 0.05), 74 (p = 0.001), 81 (p = 0.003) and 89 % (p = 0.0002) at 3, 6, 12 and 18 h sets than 0 h, respectively (Fig. 4d).

Fig. 4
figure 4

Activities of respiratory complex enzymes in muscle of the teleost Heteropneustes fossilis under air exposure condition. a NADH oxidoreductase, b succinate dehydrogenase, c cytochrome c oxidoreductase and d ATPase. Data are presented as mean ± S.E.M. (n = 10). Different superscripts indicate statistical (one-way ANOVA followed by Duncan’s post hoc test, Kruskal–Wallis ANOVA for ATPase) difference between the mean values at p < 0.05

Full size image

Correlation analysis

Values of the analyzed OS and ROS parameters were positively correlated (correlation coefficient was 0.81 for LPx, 0.76 for PC, 0.49 for H2O2) with time period of air exposure in H. fossilis. In contrast to the above, negative correlation coefficient values, i.e., −0.73, −0.73 and −0.68 were observed for SOD, CAT and GR, respectively, with time period of exposure. Although the correlation coefficient values between air exposure time and AA (0.39) and –SH (0.33) level in muscle of H. fossilis were low, they were significant and positive. A strong and positive correlation was observed between complex II (r = 0.88) activity and air exposure time. Conversely, it was negative between activities of complex III (r = −0.88) and V (r = −0.82) enzymes and air exposure time (Supplementary Fig. 1).

Discriminant function analysis

OS physiology parameters and ETC enzymes in muscle of H. fossilis exposed to air for 0, 3, 6, 12 and 18 h had yielded four roots which revealed discrimination among five groups (Table 1 and Fig. 5). Figures for root 1 versus root 2 or root 3 are presented in Fig. 5 and data from statistical output (Coefficient for Canonical variables) of all the four roots are presented in Table 1. The contribution of the variables was clearly observed for separation of the groups when all the parameters studied were analyzed together for DFA (Fig. 5d). DFA between any specified roots indicates a clear separation of the groups without overlapping with each other (Fig. 5d for root 1 versus root 3, data for other roots are not given). Taking OS indices into consideration, DFA has revealed that 6, 12 and 18 h groups had overlapping with each other. However, 0 and 3 h groups had strong non-overlapping relation with each other (Fig. 5a). DFA for only antioxidant parameters indicates that only 0 h group was segregated from the other groups which were strongly over lapped with each other (Fig. 5b). On the other hand, all the groups were separated without any major overlapping when DFA for only respiratory complex enzymes was considered (Fig. 5c). DFA of all antioxidant molecules, OS parameters and complex enzymes taken together exhibited the strongest discriminating function for SOD (via 1st root), –SH (via 4th root), complex III and V (both via 1st root) enzymes. Contribution of GPx and –SH via the second root was also significant. The overall variance was 110 % via the fist two roots (Table 1).

Table 1 Standardized Coefficient for Canonical variables of oxidative stress indices, antioxidants and respiratory complex enzyme activities in muscle of the teleost Heteropneustes fossilis as a function of dehydration
Full size table
Fig. 5
figure 5

Discriminant function analysis of oxidative stress indices, antioxidants and respiratory complex enzyme activity levels in muscle of the teleost Heteropneustes fossilis as a function of dehydration. Groups were generated by root 1 (x axis) versus root 2 (y axis). a Oxidative stress indicators showing overlapping within each other among 0, 3, 6 and 12 h of air exposure groups. b Clear group discrimination was observed for antioxidant molecules among 0, 3, 12 and 18 h groups but overlapping between 3 and 6 h exposure group is very distinct. c For complex enzymes, 0, 3, 6 and 12 h groups are vividly divided into 5 categories. d More distantly 5 groups are obtained when oxidative stress indices, antioxidant molecules and complex enzymes are considered together for discriminate function analysis. H0-control group, H3- 3 h air exposure group, H6- 6 h air exposure group, H12- 12 h air exposure group and H18- 18 h air exposure group

Full size image

Discussion

Tolerance to air exposure or hypoxia-anoxia is an important survival strategy for some specific lower vertebrates, including air breathing fishes. Tissue-specific large glycogen reserves, anaerobic glycolysis, strategy for increased redox status maintenance, metabolic depression with specific biochemical and transductional adjustments and finally reduced ATP generation and energy utilization are few metabolic responses adapted by such fishes for long-term survival under hypoxia state (Lushchak et al. 2001, 2005; Lushchak and Bagnyukova 2006; Welker et al. 2013). Such phenomenons are observed in hypoxia tolerant non-air breathing fish models such as goldfish (C. auratus) and common carp C. carpio (Lushchak et al. 2005; Van den Thillart and van Waarde 1985). For long-term survival and redox status regulation, expected metabolic responses at such compromised O2 availability condition in tissues would be the effort to generate low cellular ROS (Turrens 2003). Simultaneously, reduced energy investment not only for the expression of redox status regulating antioxidant molecules but also for the other ATP utilizing path ways would normally happen under such hypoxia state to slow down different metabolic processes (Welker et al. 2013). These responses are highly dependant on metabolic capacity of tissue and species (Lushchak et al. 2005; Welker et al. 2013). The Indian stinging fish H. fossilis is a highly air exposure (60–70 h) tolerant teleost fish (Saha and Ratha 1989) and at the present experiment, the critical time of survival period under air exposure induced hypoxia state at 25 °C in this fish weighing 23.44 ± 1.63 g was found to be >18 h. Results of the present study indicate that the dynamic changes in redox metabolism may also be associated with such long-term survival of H. fossilis under water deprivation condition.

DFA was used to identify variables that were important to determine oxidative and redox status in muscle tissue of H. fossilis under air exposure condition. Altogether, the contribution of the studied variables such as ROS, OS indices, antioxidant enzymes, small antioxidant molecules and respiratory complexes of ETC to separate the five sets, i.e., 0, 3, 6, 12 and 18 h air exposure fish to five groups was distinct. It indicates that air exposure has definite role on the dynamic changes in OS physiology in muscle of the fish. Similar observations were recorded for OS physiology parameter due to seasonal effects in crayfish (Borkovic et al. 2008) and due to seasonal environmental O2 concentration in mud crabs (Paital and Chainy 2013b). DFA results for only OS indices point out that 6, 12 and 18 h groups were very strongly overlapped with each other whereas 0 and 3 h groups were separated clearly in a non-overlapping manner. It indicates the increase in ROS level and OS state when the fish were kept out of water for 3 h. Once they were out of water, the above states were maintained almost at a steady state at 6, 12 and 18 h groups. Therefore, the critical time observed for OS and ROS generation by dehydration or hypoxia in muscle of H. fossilis was 3 (for LPx and H2O2) to 6 h (for PC). The unchanged PC level observed in muscle of H. fossilis with increased H2O2 and TBARS level at 3 h air exposure than 0 h may indicate the imbalance between production of oxidized proteins and removal of the same by proteasomes in favor of the former (Halliwell and Gutteridge 2001). Reduced O2 supply to the tissue under air-exposed states could be a reason to limit the scope of producing more ROS in 3 h onward groups (Turrens 2003). This could be a reason why the tissue maintained equal level of H2O2 and its by-products, i.e., TBARS and PC levels at 6, 12 and 18 h air exposure groups and these three groups yielded a non-overlapping DFA result. DFA indicates that both the enzymatic and non-enzymatic redox regulating molecules had the contributions for the observed above pattern for ROS and OS level in the tissue. Other than the control group, 3, 6, 12 and 18 h groups were strongly overlapped with each other in DFA for the studied antioxidant molecules indicating the immediate response (decrease in the present case) of the antioxidant defense in fish to air exposure. Although significant changes within the air exposure groups were observed for all the four antioxidants studied and an inconsistent change were noticed for AA and –SH level, only SOD, GPx and –SH were contributed the highest for discrimination of 0 h groups from the rest four air exposure groups. Overall, DFA suggests that the reduction in antioxidant capacity is responsible for the increase in ROS and OS level in the tissue when the fish were kept out of water. The fact is further supported by the noticed negative correlation coefficients between activities of the studied enzymes and time of air exposure and positive correlation observed between the levels of ROS and OS indices and time of air exposure. Responses such as slow metabolic rate, initiation of proteolysis and conservative investment strategy for enzyme expression in combination or alone could contribute for the collective dysfunction of the studied four redox state maintaining enzymes in the fish under air exposure condition (Welker et al. 2013). Conversely, a tolerable ROS level and OS state in the tissue at air exposure condition could be also responsible for the lowering of the antioxidant enzyme activity level. Along with the dysfunction of redox regulating enzymes, the increased ROS and OS level observed in the tissue under air exposure stress may also be due to the cumulative action of ETC complex enzymes. DFA revealed that all the five groups were separated without any major overlapping for the four studied respiratory complex enzymes. The consistent increase in complex II activity and increase in complex I activity up to 6 h of air exposure may indicate the facilitation of smooth electron supply to ETC of the fish. The significant contribution of complex II in the above process was also revealed by the higher standardized coefficient for canonical variables observed for it in DFA. At the same time, the decreased trend of complex III and V enzymes could be an adaptation in the tissue to slow down further electron flow in ETC and ATP production in air, respectively. The high standardized coefficient for canonical value recorded for complex V also supports the latter case. Combined, DFA and correlation analysis show that the lower activities of antioxidant enzymes, reduced ETC activity with possibly alleviated ATP production rate and tolerable ROS and OS could confer the fish to survive for a long period (>18 h) out of water. On the other hand, the redox metabolic responses especially ETC activity in brain of H. fossilis were found to be different than muscle under such air exposure condition in H. fossilis (Paital 2013).

Interpreting individual set of data such as ROS and OS, redox regulating molecules and ETC activity can make the above picture clearer. The observed positive correlation between the studied ROS and OS indices (TBARS and PC) in muscle of H. fossilis and air exposure time undoubtedly indicate signs of OS under hypoxia state. Up to threefold increase in TBARS observed in the liver of C. carpio under hypoxia and a similar observation in the budding yeast Saccharomyces cerevisiae support the above fact (Dirmeier et al. 2002; Turrens 2003; Lushchak et al. 2005). A concomitant decreased PC level without any change in LPx level was observed in muscle of a common carp C. carpio under hypoxia state up to 5.5 h (Lushchak et al. 2005). Although no apoptotic markers were analyzed in the present case, initiation and gradual progression of cell decomposition and damage leading to apoptosis could also contribute to the above fact. Altogether, it could be due to the achievement of hypoxia-induced conservative strategy adapted in muscle of the fish to prevent ROS-induced damage and energy dissipation (Welker et al. 2013). Probably such conservative strategy is one of the adaptation mechanisms which help this teleost fish to survive up to so long period (18 h) out of water. To the contrary, the fish were died after a very prolonged air exposure such as 24 h (data not shown). This may indicate the initiation of reduced protein turn over, increased proteolysis and cellular metabolic collapsing and possibly triggering apoptosis in H. fossilis after 3 h air exposure.

The collective loss in activities of SOD, CAT, GPx and GR observed in air exposure groups than control indicates the reduction in redox capacity in the tissue. Lushchak et al. (2005) noticed similar decrease in only GPx activity in muscle of the common carp C. carpio under hypoxia stress up to 5.5 h. On the other hand, an increase in activity of antioxidant enzymes such as CAT and GPx was observed by the above author in brain of carp C. carpio under hypoxia condition for 5.5 h (Lushchak et al. 2005). At 18 h, the redox state in muscle of H. fossilis could be the lowest due to the lowest level of all studied antioxidant enzymes. This could create a scope for ROS to be accumulated in the tissue at 18 h. However, no such result observed at 18 h could be due to the neutralization of surplus ROS by the increased –SH group at 18 h than the rest of the air exposure groups. On the other hand, a constant pattern was not observed for non-protein –SH groups in response to air exposure. It is reviewed that usually hypoxia has no effects on glutathione level in fishes (Lushchak and Bagnyukova 2006). No change in AA level in the tissue up to 12 h air exposure may indicate about the synthesis or accumulation of this antioxidant in the tissue (Drouin et al. 2011; Halliwell and Gutteridge 2001). This is because, AA level in muscle of H. fossilis under OS condition at air exposure condition should be diminished (Parihar and Dubey 1995), if AA is not synthesized in the tissue and the source of accumulated AA level in the tissue is only from dietary supplement (goat liver in the present study). A small but significantly augmented AA level was noticed at 18 h group in comparison to 3 and 6 h groups may indicate the presence of AA synthesizing enzyme, i.e., l-gulonolactone oxidase in muscle or in any other tissue of H. fossilis (Paital 2013). However, the presence of l-gulonolactone oxidase in fresh water teleosts including H. fossilis is still unclear (Drouin et al. 2011).

The higher ROS level than control could contribute to the decreased activity of complex I enzyme after 6 h onwards air exposure due to the removal of iron from the Fe–S center of complex I by ROS as observed in mammalian cells (Powell and Jackson 2003). The decreased activity of complex III noticed in air exposure groups could be anticipated by the tissue to slow down the flow of electrons to reduce the ETC activity. Finally, the decrease in transport of electrons in ETC could serve as an adaptive response to limit overproduction of ROS during hypoxia and to slow the rate of oxygen depletion to prevent anoxia in cells (William and Chandel 2011). Reduction in complex V activity can be correlated with the reduced ATP synthesis rate in the muscle of the fish under air exposure condition because, air exposure or hypoxia found to diminish ATP production in part by lowering the activity of the ETC enzymes through activation of the transcription factors, for example, hypoxia-inducible factor-1 (William and Chandel 2011). This fact is further supported by the observed decrease in activity of all complex enzymes in mammalian cells to hypoxic state for 24 h (Colleoni et al. 2013). Therefore, altogether, the observed cumulative response of ETC in muscle of H. fossilis under air exposure stress could be helpful to adapt the conservative strategy in the tissue to slow down energy production and subsequent utilization (Abele et al. 2011; Brand et al. 1991; Welker et al. 2013). As a result the fish could manage to survive for a longer period out of water.

The critical time of air exposure for the dynamic change in oxidative metabolism in muscle of H. fossilis was mostly found to be 3 h. Lethal air exposure stress period was >18 h after which the cellular physiology might be started collapsing in muscle of H. fossilis and possibly this arrested protein turnover or synthesis leading to proteolysis and ceasing of cellular physiology in the fish. This could be a reason for the gradual decrease in antioxidant functions in the tissue after 3 h air exposure. Redox status regulation and responses of respiratory complexes are very important to adapt specific strategy for hypoxia tolerance in H. fossilis in water deprivation condition. Future studies on the reoxygenation phase should be carried out to better understand antioxidant response in muscle of H. fossilis.