1. Introduction

In aquaculture, sufficient concentrations of dissolved oxygen are required for glycolysis in the cytoplasm, the continual functionality of the mitochondrial electron transport chain, and energy production via oxidative phosphorylation [1]. These metabolic processes involve cytochrome oxidase, which catalyzes the reduction of molecular oxygen to water. More than 90% of the oxygen consumed by the organism is used in these reactions, which also involve the formation of various by-products [2]. These by-products are reactive oxygen species (ROS), such as the superoxide radical ($$), hydroxyl radical (˙OH), hydrogen peroxide (H₂O₂), single oxygen, lipid peroxides, peroxynitrite (ONOO⁻), and nitric oxide (NO). ROS are generated due to electrons leaking from oxidase reactions and the electron transport chain [3]. Low levels of ROS are related to signal transduction [4], and the regulation of various cellular activities such as cell cycle checkpoint, energy metabolism, redox balance [5], apoptosis, and regulation of protein and gene expression [6]. However, high levels of ROS cannot be tolerated by the organism and can lead to oxidative stress. Consequently, macromolecules such as proteins, lipids, and DNA are damaged, leading to the formation of malondialdehyde (MDA) and lipid peroxidation (LPO) [7].

The removal of radical and non-radical oxygen-derived species, as well as the mitigation of oxidative stress, is facilitated by enzymatic defense mechanisms involving superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPx), glutathione-S-transferase (GST), and glutathione reductase (GR). Non-enzymatic ROS cleavages include glutathione (GSH), vitamin E, and ascorbate. These cleavages work together to remove ROS in cells [7]. SODs are metalloenzymes that catalyze the dismutation of the superoxide anion into hydrogen peroxide (H₂O₂) and molecular oxygen (O₂) [8]. H₂O₂ is reduced to water by GPx in the cytosol and by CAT in peroxisomes utilizing reduced GSH as a cofactor. GST is a crucial member of of xenobiotic detoxification enzymes that function in pro-oxidant environments to facilitate the detoxification of hazardous compounds created by oxidative stress. GR utilizes nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) to convert oxidized glutathione (GSSG), into its reduced form GSH, maintaining cellular redox homeostasis t. Oxidative stress induces the production of SOD, CAT, GPx, GST, and GR. The levels of these enzymes are directly related to the amount of oxidative stress in the cells. GSH and glutathione disulfide (GSSG) are crucial intracellular thiols for biological processes [7, 9]. MDA is an oxidative product generated by the oxidation of polyunsaturated fatty acids in the cell membrane structure and by free radicals-induced LPO [10]. The activity levels of these antioxidant enzymes and the amounts of components directly or indirectly linked to antioxidant capacity are important for the antioxidant defense system and the detoxification mechanism. Impairment or inhibition of these enzymes, which are crucial for antioxidant defense, may lead to mass fish mortality [11].

Organisms in aquatic environments are temporarily or continuously subject to variations in oxygen levels, on both daily and seasonal scales, due to factors such as temperature, water flow, atmospheric conditions, eutrophication, algal photosynthetic activity, organismal respiratory, and increased biomass and production in aquaculture [12–14]. For this reason, oxygen concentration is one of the most important environmental parameters for aquatic organisms. Aquatic hypoxia in freshwater environments is defined as dissolved O₂ concentrations below 5–6 mg O₂/L [15]. Hyperoxia is commonly defined as a higher than normal O₂ tension in water. An average of 12 mg O₂/L is defined as a hyperoxic condition [16, 17]. Chronic exposure to hypoxia and hyperoxia can cause oxidative stress in fish [18]. Rainbow trout (Oncorhynchus mykiss) are recognized as a fish species particularly sensitive to hypoxia, especially when dissolved oxygen levels fall below 7 mg/L [19]. Hypoxia has been associated with increased respiratory rate and amplitude, as well as elevated aortic blood pressure, ultimately resulting in bradycardia [20] and a reduction in ATP concentration [21]. In addition, hypoxia reduces weight gain, specific growth rate (SGR), and survival rate (SR), thereby negatively affecting overall growth parameters. It also triggers stress responses and suppresses certain physiological processes [18]. In the literature, many studies related to hypoxia or hyperoxia can be found [22–24]. However, these studies often focus on protein levels or enzymatic activity, which we consider insufficient for understanding physiological responses in fish. To better understand both the acute and chronic effects of mild hypoxia and hyperoxia on metabolism, this study aimed to investigate the effects of these exposures on gene expression, enzyme activities, and antioxidant-related metabolites in the liver, a metabolically active organ.

2. Materials and Methods

3. Results

3.1. Glutathione Dependent Parameters

Acute hypoxia and hyperoxia treatments significantly increased the tGSH level compared to the control (normoxia) (p  < 0.05). Conversely, chronic exposures significantly decreased it (p  < 0.05). The GSH level was initially high with hypoxia treatment, but the chronic effect decreased the level (p  < 0.05). GSSG levels increased inversely to GSH in acute exposures, but chronic exposure did not change the GSSG level. The level of the GSH/GSSG ratio, an oxidative stress indicator, was affected by hypoxia and hyperoxia and its level generally decreased. It was expected that GSH was significantly correlated with GSH/GSSG levels (Figure 1, p  < 0.01). The OSI increased in comparison to control groups after 48 h, 72 h, and 28 days following exposure. Hyperoxia generally increased OSI compared to control groups. Interestingly, hypoxia affected the stress index after 48 h. Acute hypoxia and hyperoxia exposures did not change the MDA and H₂O₂ levels, and no correlation was found between them (Table 2, Figure 1).

Table 2. Antioxidant-dependent metabolite levels in the liver of rainbow trout (O. mykiss) at experimental times.

	Initial level	Exposure	Acute exposures					Chronic exposure
			6 hr	12 hr	24 hr	48 hr	72 hr	28 days
MDA (nmole/mL)	0.021	Normoxia	0.028 ±0.0005aB	0.023 ±0.0008aCD	0.022 ±0.0009aD	0.032 ±0.0003aA	0.023±0.0010bCD	0.024±0.0009bC
		Hypoxia	0.021 ±0.0005bC	0.024 ±0.0007aB	0.025 ±0.0010aB	0.016 ±0.0008bD	0.026±0.0012abB	0.028±0.0008aA
		Hyperoxia	0.025 ±0.0019aA	0.024 ±0.0032aA	0.024 ±0.0030aA	0.011 ±0.0018cB	0.027±0.0006aA	0.030±0.0010aA
H2O2 (pmole/µL)	0.023	Normoxia	0.33 ± 0.01aC	0.19 ± 0.01bD	0.20 ± 0.01abD	0.30 ± 0.01aC	0.44 ± 0.01aB	0.68 ± 0.01aA
		Hypoxia	0.11 ± 0.01bC	0.46 ± 0.00aAB	0.28 ± 0.03aBC	0.41 ± 0.13aABC	0.41 ± 0.15aABC	0.61 ± 0.12aA
		Hyperoxia	0.30 ± 0.05aB	0.31 ± 0.09abB	0.14 ± 0.02bB	0.92 ± 0.28aA	0.41 ± 0.15aB	0.60 ± 0.15aAB
tGSH (ng/mL)	1.42	Normoxia	1.42 ± 0.02cB	1.41 ± 0.02cB	1.44 ± 0.02bB	1.47 ± 0.02abAB	1.45 ± 0.01bB	1.52 ± 0.01aA
		Hypoxia	1.94 ± 0.02aaA	1.50 ± 0.01bBC	1.57 ± 0.04aB	1.52 ± 0.0aBC	1.55 ± 0.02aBC	1.47 ± 0.02bC
		Hyperoxia	1.49 ± 0.01bB	2.01 ± 0.02aA	1.49 ± 0.01abB	1.45 ± 0.0bB	1.50 ± 0.03abB	1.45 ± 0.02bB
GSH (ng/mL)	1.04	Normoxia	0.74 ± 0.02bA	0.73 ± 0.02bA	0.74 ± 0.02abA	0.74 ± 0.02aA	0.75 ± 0.02aA	0.76 ± 0.0aA
		Hypoxia	1.17 ± 0.02aA	0.76 ± 0.01bBC	0.79 ± 0.05aB	0.69 ± 0.0bC	0.75 ± 0.02aBC	0.69 ± 0.0bC
		Hyperoxia	0.74 ± 0.0bB	1.18 ± 0.03aA	0.67 ± 0.01bCD	0.65 ± 0.0bD	0.72 ± 0.03aBC	0.67 ± 0.02bCD
GSSG (ng/mL)	1.51	Normoxia	0.68 ± 0.0cD	0.68 ± 0.0cD	0.70 ± 0.0cC	0.73 ± 0.0cB	0.70 ± 0.01cC	0.76 ± 0.0aA
		Hypoxia	0.77 ± 0.0aC	0.74 ± 0.0bD	0.78 ± 0.0bC	0.83 ± 0.0aA	0.81 ± 0.01aB	0.78 ± 0.01aC
		Hyperoxia	0.75 ± 0.01bD	0.83 ± 0.01aA	0.81 ± 0.0aB	0.79 ± 0.0b	0.78 ± 0.01b	0.78 ± 0.0a
GSH/GSSG	1.38	Normoxia	1.09 ± 0.03bA	1.07 ± 0.03bA	1.05 ± 0.04aA	1.02 ± 0.02aA	1.07 ± 0.03aA	1.00 ± 0.01Aa
		Hypoxia	1.52 ± 0.02aA	1.02 ± 0.02bB	1.02 ± 0.06aB	0.83 ± 0.0bC	0.92 ± 0.02bC	0.90 ± 0.01bC
		Hyperoxia	0.99 ± 0.01cB	1.41 ± 0.05aA	0.83 ± 0.01bCD	0.82 ± 0.02bD	0.93 ± 0.04bBC	0.85 ± 0.02bCD
OSI	84.21	Normoxia	95.65 ± 1.36bA	96.48 ± 1.60aA	97.39 ± 1.68bA	99.24 ± 1.08bA	96.74 ± 1.48bA	100.04 ± 0.40bA
		Hypoxia	79.39 ± 0.51cD	99.05 ± 0.92aC	99.38 ± 3.08bC	109.36 ± 0.23aA	104.38 ± 1.24aB	105.45 ± 0.50aAB
		Hyperoxia	100.40 ± 0.64aB	82.94 ± 1.84bC	109.20 ± 0.74aA	109.64 ± 0.97aA	103.74 ± 2.22aB	108.21 ± 1.47aA

• Note: Different letters denote significant differences from each other at p  < 0.05 by one-way ANOVA with Duncan’s post hoc test. The statistically significant difference is indicated by different lowercase letters at each period and by different uppercase letters in each treatment from each other.
• Abbreviations: GSH, glutathione; GSSG, glutathione disulfide; MDA, malondialdehyde; OSI, oxidative stress index; tGSH, total glutathione.

3.4. Correlation Analysis and Heat Map Indicators

The significant increase, decrease, or insignificant changes of all parameters in our study are presented in Table 5. Providing all the parameters clearly in this table helps us see the whole picture at once. Correlation analysis and Heat Map indicators showing the correlations and changes in the antioxidant-dependent metabolites, enzyme activities, and gene expressions of antioxidants are shown in Figures 1 and 2, respectively.

Table 5. Antioxidant parameters with significant increase, decreases, and non-significant changes with regard to hypoxia and hyperoxia exposures and over time.

Time	6 h		12 h		24 h		48 h		72 h		28 d
	Hypoxia	Hyperoxia	Hypoxia	Hyperoxia	Hypoxia	Hyperoxia	Hypoxia	Hyperoxia	Hypoxia	Hyperoxia	Hypoxia	Hyperoxia
tGSH	↑	↑	↑	↑	↑	ns	ns	ns	↑	ns	↓	↓
GSH	↑	ns	ns	↑	ns	ns	↓	↓	ns	ns	↓	↓
GSSG	↑	↑	↑	↑	↑	↑	↑	↑	↑	↑	ns	↓
GSH/GSSG	↑	↓	ns	↑	ns	↓	↓	↓	↓	↓	↓	ns
OSI	↓	↑	ns	↓	ns	↑	↑	↑	↑	↑	↑	↑
MDA	↓	ns	ns	ns	ns	ns	↓	↓	ns	↑	↑	↑
H2O2	↓	ns	↑	ns	ns	ns	ns	ns	ns	ns	ns	ns
SOD exp.	↓	↓	↓	ns	↑	↑	ns	↓	↓	↓	↓	↓
CAT exp.	↓	↓	↓	↑	↑	↑	↑	ns	↓	↓	↓	↓
GPx exp.	↑	↑	↓	↓	ns	ns	ns	ns	ns	↑	↑	↓
GST exp.	↓	↓	↓	↑	↑	↑	↑	ns	↓	↓	↓	↓
GR exp.	ns	ns	↓	↑	ns	↑	↓	↓	ns	↑	↓	ns
SOD act.	↓	↓	↓	ns	ns	ns	ns	↓	↓	↓	ns	ns
CAT act.	↓	↑	↓	↑	↑	↑	↑	↑	ns	↑	↑	↓
GPx act.	ns	↓	↓	ns	↑	↓	↓	↓	↓	↑	ns	↑
GST act.	↓	↓	↑	↓	↑	↓	↑	↓	↑	↓	↑	↑
GR act.	↓	↓	↓	↑	↓	ns	↓	↓	ns	↓	↓	ns

• Note: (↑) symbols with red color indicates the increase, (↓) symbols with green color the decrease, and (ns) not significant indicates that the levels have not changed. The increase and decrease are statistically significant (p  < 0.05).
• Abbreviations: act., activity; CAT, catalase; exp., expression; GR, glutathione reductase; GSH, glutathione; GSSG, glutathione disulfide; GST, glutathione-S-transferase; MDA, malondialdehyde; OSI, oxidative stress index; SOD, superoxide dismutase; tGSH, total glutathione.

4. Discussion

The examination of the genomic composition of rainbow trout (O. mykiss) and its susceptibility to hypoxic and hyperoxic conditions has become a critical area of focus in recent aquaculture research. This highlights its significance in understanding environmental stress adaptation and the management of fish health. The potential implications of such research, including the enhancement of fish health management, and the promotion of environmentally sustainable and economically viable aquaculture practices, are of considerable importance [18, 22, 37]. The effects of hypoxia and hyperoxia conditions on the antioxidant defense system of rainbow trout are closely linked to the organism’s biological adaptation mechanisms for maintaining oxygen homeostasis [19]. The initial increase in GSH levels under hypoxia likely reflects an acute adaptive response to oxidative stress, as GSH is a crucial non-enzymatic antioxidant that neutralizes

ROS [38]. However, the subsequent decline in GSH levels after prolonged exposure suggests that the antioxidant defense system becomes overwhelmed, leading to redox imbalance and oxidative damage [22]. In our study, GSH levels initially increased under hypoxia but gradually decreased after 48 h, especially following chronic exposure. Similarly, the reduced to oxidized glutathione (GSH/GSSG), which was elevated in the early stages, showed a progressive decrease in both acute and chronic hypoxia and hyperoxia conditions. These findings suggest that prolonged oxidative stress impairs the cellular redox balance, leading to the depletion of antioxidant reserves.

GSH level, initially high due to hypoxia treatment, gradually decreased after 48 h, including as a result of chronic exposure. Although the ratio of reduced to oxidized glutathione (GSH/GSSG) increased at the beginning, a decrease in progressive acute and chronic exposure occurred in both hypoxia and hyperoxia in our study. Therefore, GSH and GSH/GSSG levels are expected to show a positive correlation with each other in both hypoxia (p  < 0.01, r = 0.999) and hyperoxia (p  < 0.01, r = 0.999). Decreased GSH/GSSG ratio and GSH concentration, as well as increased OSI and GSSG levels, are also good biomarkers of oxidative stress and evidence of redox imbalance [7]. Although OSI and GSSG were correlated under hypoxia (p  < 0.05, r = 0.48), no correlation was found under hyperoxia. In this context, it is not surprising that OSI showed a significant negative correlation to GSH and GSH/GSSG in both exposures. GSH is required for GPX to reduce hydrogen peroxide. Moreover, in animal organisms, GSH can also scavenge ROS directly and, therefore, GSH depletion contributes to oxidative stress [38].

The observed increase in MDA levels after 24 h, especially under chronic exposure, suggests intensified LPO, a key indicator of oxidative stress. MDA is a byproduct of polyunsaturated fatty acid oxidation, and its accumulation is indicative of oxidative membrane damage [39]. The fact that MDA levels remained stable during early exposure suggests that initial antioxidant responses, including increased GSH activity and enzymatic defenses, were sufficient to mitigate LPO. Increased hepatic LPO following the reduction of GSH levels has also been observed by Jia et al. [39] in black rockfish under hypoxia stress [39]. Another study on rainbow trout reported that sufficient antioxidant defenses limited LPO [28]. This suggests that chronic oxidative stress disrupts membrane stability, potentially impairing cellular functions.

The major reducer of hydrogen peroxide in antioxidant defense is GPx ([40]). However, GPx protects against low levels of oxidative stress by detoxifying hydrogen peroxide, while CAT protects against severe levels of oxidative stress [7]. We found that hydrogen peroxide and GPx activity were negatively correlated in hypoxia (p  < 0.05, r = 0.501), but hydrogen peroxide was not correlated with CAT enzyme activity levels. The change in hydrogen peroxide concentration indicates the presence of oxidative stress [7]. In our study, hydrogen peroxide levels in the liver exhibited considerable variability during the early stages of acute exposure but later returned to levels comparable to those in the control group. This fluctuation was primarily attributed to the high standard error of the mean. A stable hydrogen peroxide level may suggest that antioxidant enzymes, such as CAT and/or GPx, effectively mitigate oxidative stress by neutralizing this peroxide.

How different oxygen levels regulate or alter antioxidant responses can be discovered by examining enzyme activities and gene expression simultaneously. Generally, an increase in intracellular ROS levels leads to an upregulation of antioxidant enzymes in zebrafish [41]. This might result from an increase in ROS caused by the loss of mitochondrial electron transport chain cytochromes and the subsequent leaking of these cytochromes to leftover oxygen molecules [7]. In our study, SOD activity was lower in both treatments compared to the control. A decrease in SOD activity against oxidative stress in mice [42] and expression against environmental stress conditions in rainbow trout [43] was also observed in other studies. These low levels of SOD may not have caused a significant increase in hydrogen peroxide.

Acute hyperoxia enhanced CAT activity in the liver of rainbow trout during all exposure periods. It was reported that CAT activity decreased in chronic hyperoxia, matching data obtained from 48-hour hyperoxia exposure in trout [24]. Hyperoxia increased CAT activity first, which then gradually inhibited. This pattern partly paralleled the gene expression results. There was a significant positive correlation between CAT gene expression and activity levels as a result of hyperoxia exposure (p  < 0.01, r = 0.69).

Enzymes involved in antioxidant defenses exist as a coordinated system and include GSH-dependent enzymes such as GPx which neutralizes H₂O₂, and GR which catalyzes the transformation of GSSG to reduced GSH using NADPH [7]. We determined that the pattern of GPx activity does not correlate with gene expression, a situation observed in other enzymes as well, except for CAT. The incompatibility between activity and gene expression of antioxidant enzymes has been reported in mice and rats [44, 45]. This discrepancy arises because these enzymes are largely controlled post-transcriptionally in zebrafish embryos [46]. Post-transcriptional modifications, the presence of pre-formed antioxidant enzymes or conformational changes can alter the activity levels of this enzyme [47]. MDA and GPx activity showed a significant correlation (p  < 0.05, p = 0.498) in hyperoxia. GPx activity is directly related to LPO [7]. We estimate that the GPX enzyme functions exclusively, rather than the CAT enzyme, to eliminate the peroxides produced under stress conditions. Considering the levels of hydrogen peroxide and LPO, although CAT activity increased, GPX activity was inconsistent. This finding is consistent with those reported by Izawa, Inoue, and Kimura [48] and was also observed in a study conducted on Saccharomyces cerevisiae, indicating that the affinity of GPx for H₂O₂ is higher than that of CAT.

One of the main antioxidant enzymes and the first-line markers of the antioxidant status is GST, which detoxifies various substances by conjugating them with GSH [49]. This antioxidant and detoxification mechanism was shown to be related to the increase in MDA in the freshwater fish Channa punctata [50]. In addition, an oxidative state linked to hypoxia can increase GST activity. Studies conducted on rats [51] and rainbow trout [52] indicate that GST activity decreases in acute exposure to hyperoxia [51, 52]. Zhao et al. [37] found in Schizothorax prenanti that the SOD, CAT, and GPx activity data are compatible with our study after 12 and 24 h of hypoxia [37]. GST gene expression correlated positively with CAT and SOD gene expressions (p  < 0.01). We found that GST expression decreased over time with chronic effects. A similar situation was observed at the transcriptome levels in other enzymes. Jiang et al. [53] also noted that antioxidant genes were expressed more often on Cyprinus carpio under oxidative stress [53]. However, gene expression levels decreased below normal levels in our study. This suggests that the fish can tolerate the oxidative stress effects caused by chronic hypoxia and hyperoxia at the gene expression level.

The regulation of antioxidant defense systems under different oxygen conditions is largely controlled at the molecular level through transcriptional and post-translational modifications [38]. One of the primary regulators of oxidative stress responses is the Nrf2-Keap1-ARE pathway, which modulates the expression of antioxidant enzymes, such as SOD, CAT, GPx, and GST [54]. This mechanism could explain the initial upregulation of SOD and CAT gene expression in our study, followed by their subsequent downregulation as prolonged oxidative stress may lead to feedback inhibition [55]. The interaction between HIF-1 α and Nrf2 plays a key role in hypoxic stress adaptation. Under low oxygen conditions, HIF-1 α enhances glycolysis and antioxidant defenses to reduce ROS accumulation [56]. However, prolonged hypoxia can induce Nrf2 degradation, leading to decreased antioxidant enzyme transcription [46]. This aligns with our findings, where antioxidant enzyme activity and gene expression initially increased but later declined under chronic hypoxia.

The decrease in GPx activity after 48 h of the same treatment was notable. GPx activity induced by hypoxic conditions positively correlated with GSH levels (p  < 0.05, r = 0.50). In addition, GR activity also showed a positive correlation with GSH activity (p  < 0.05, r = 0.56) indicating that these GSH-dependent enzymes act together in the conversion of oxidized and reduced GSH. Under stressful conditions, various enzymatic antioxidants work together, sometimes partially replacing one another [57]. The activities of GSH-dependent enzymes changed as a result of the in vivo inhibition of CAT in the goldfish brain [58, 59]. High levels of oxidized GSH may result from the low activity levels of the GR in hypoxia. GR activity, like other examined parameters, was not correlated with gene expression. A study conducted on zebrafish embryos demonstrated that antioxidant enzymes are regulated by multiple factors and that transcription and translation play a crucial role in the temporary adaptation to oxidative stress [46]. The increase in SOD, CAT, and GST expressions at the 24th h after both exposures indicates a stress-dependent response. This suggests that 24 h may be the maximum defense time at the transcriptomic level for the rainbow trout.

5. Conclusion

This study determined that hypoxia and hyperoxia exposures, compared to normoxia conditions, significantly affected various parameters, including gene expression levels. OSI, GSH, GSSG, and GSH/GSSG levels revealed that hypoxia and hyperoxia caused liver damage and negatively impacted the components of the antioxidant defense system. Acute and chronic exposure to low and high levels of dissolved oxygen resulted in various changes in enzyme activity and gene expression in the liver. A large number of genes undergo expression changes under various stimuli, with some genes being downregulated, others upregulated, some modulated in the early stages of adaptation, and other genes’ expression impacted later on. The roles of hydrogen peroxide and post-transcriptional modifications in this process are inevitable. Changes in enzyme activity, GSH metabolism, hydrogen peroxide, and MDA observed after 6 h of exposure to hypoxia or hyperoxia conditions may represent the organism’s initial response to oxygen-induced ROS. The continued high enzyme activity after chronic exposures indicates that rainbow trout are in constant defense against hypoxia and hyperoxia.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding

This work was funded by the Scientific and Technological Research Projects Funding Program as a part of Türkiye Bilimsel ve Teknolojik Araştırma Kurumu (TUBITAK) (project no. 114O366).