Volume 27, Issue 6 pp. 2567-2574
ORIGINAL ARTICLE
Full Access

Comparison of selenomethionine and hydroxyselenomethionine on tissue selenium retention, and antioxidative capacity of giant grouper, Epinephelus lanceolatus, fed diet with soybean meal

Yu-Hung Lin

Corresponding Author

Yu-Hung Lin

Department of Aquaculture, National Pingtung University of Science and Technology, Neipu, Pingtung, Taiwan

Correspondence

Yu-Hung Lin, Department of Aquaculture, National Pingtung University of Science and Technology, 1 Shuefu Road, Neipu, Pingtung 912, Taiwan.

Email: [email protected]

Search for more papers by this author
Tzu-Han Lin

Tzu-Han Lin

Department of Aquaculture, National Pingtung University of Science and Technology, Neipu, Pingtung, Taiwan

Search for more papers by this author
First published: 23 October 2021
Citations: 3

Abstract

The study aims to evaluate the effects of two organic selenium (Se) sources, selenomethionine (SeMet) and hydroxyselenomethionine (OH-SeMet), on tissue Se retention, and antioxidative capacity of giant grouper, Epinephelus lanceolatus, fed diets with soybean meal (SBM). The control diet containing 32.92% SBM (replaced 40% fish meal protein) was supplemented with 0.3, 0.6 and 1.0 mg Se/kg of SeMet and OH-SeMet. A reference diet with fish meal (FM) as main protein source was also included for comparison. The eight experimental diets were each fed to triplicate groups of grouper juveniles (initial weight: 25.99 ± 0.04 g) in a recirculation system with seawater for 8 weeks. The Se concentrations in whole body, muscle and liver were dependent on dietary Se concentrations, but not on its supplemental forms. Higher hepatic glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT), but lower malondialdehyde (MDA) content, were observed in fish fed the reference diet than in fish fed the control diet. The results indicated that dietary FM protein replaced by SBM at 40% depressed growth performance, Se retention and antioxidative capacity of giant grouper. Both Se sources supplementation enhanced Se retention and antioxidative capacity, but not growth performance of the fish.

1 INTRODUCTION

Grouper exhibits high economic values because of its meat quality and richness in collagen and has become a popular culture species in Taiwan and Asia (Zhuo et al., 2016). Adequate protein levels for grouper are high at 45%–50% and are largely dependent on fish meal diet (FM; Chou et al., 2008). Due to the limited marine resources, FM production tends to be stable, but FM prices are increasing (FAO, 2020), necessitating the search for a less expensive and readily available alternative to FM. Soybean meal (SBM) shows the advantages of stable yield, rather low price and better amino acid composition, compared with other plant protein feedstuffs (NRC, 2011). High levels of replacement of FM by SBM in the diet of giant grouper have been demonstrated to induce enteritis and oxidative stress and to depress growth performance and nutrient digestibility (Lin & Cheng, 2017; Lin & Lu, 2020). These negative effects were possibly influenced by nutritive imbalance and antinutritional factors in SBM.

Selenium (Se) is an essential microelement which is the component in an antioxidant enzyme and glutathione peroxidase (Rotruck et al., 1973). This enzyme can catalyse hydrogen peroxide or fatty acid hydroperoxides to water or fatty acid alcohol through glutathione to maintain integrated cell membrane (Watanabe et al., 1997). Based on the available literature, Se requirements or bioavailability of organic and inorganic Se sources for fish is well-documented (Jaramillo et al., 2009; Küçükbay et al., 2009; Lin, 2014; NRC, 2011). Fish is considered an excellent Se source, particularly in comparison with plants, because the level of Se in plants is usually low and is greatly dependent on Se concentration in soils (Reilly, 1998). Therefore, Se supplementation must be given careful attention when dietary FM is replaced by SBM.

Hydroxyselenomethionine (OH-SeMet) is a new type of organic Se source, which is a hydroxyanalogue selenomethionine (SeMet) (Briens et al., 2013). The metabolic pathway of OH-SeMet is similar to that of SeMet; however, the stability of OH-SeMet is better than SeMet during storage (Surai et al., 2018). The European Commission has approved the use of OH-SeMet for animals as a feed additive (EFSA, 2013). Much attention has been given to the use of OH-SeMet feed additives for land animals such as broilers (Briens et al., 2013, 2014), dairy cows (Sun et al., 2017; Wei et al., 2018) and pork (Jlali et al., 2014). For aquatic animals, specifically gilthead seabream (Mechlaoui et al., 2019) and rainbow trout (Wischhusen et al., 2019), it has been reported that supplementation of OH-SeMet improved growth, hepatic morphology and stress resistance. However, the comparison of bioavailability between OH-SeMet and SeMet for fish is still unclear. Thus, an 8-week feeding trial was conducted to compare effects of the two organic Se sources on growth performance, tissue Se retention and antioxidant capacity of giant grouper, Epinephelus lanceolatus, fed diets with SBM.

2 MATERIALS AND METHODS

2.1 Diets preparation

Formulation and composition of the experimental diets are listed in Table 1. All the diets were maintained in isonitrogenous and isolipidic form. The diet containing 329.2 g/kg soybean meal without Se additive (TTET Union, Taiwan; replaced 40% fish meal protein) was used as a control diet. The control diet was incorporated with 0.3, 0.6 and 1.0 mg Se/kg of selenomethionine (Availa® Se, Zinpro, USA; 0.3 SeMet, 0.6 SeMet and 1.0 SeMet) and hydroxyselenomethionine (Selisseo®, Adisseo, France; 0.3 OH-SeMet, 0.6 OH-SeMet and 1.0 OH-SeMet), respectively. The FM protein replacement level (40%) was referred to our previous studies (Lin & Cheng, 2017; Lin & Lu, 2020), which found that the level caused poor growth for giant grouper. The reference diet containing 580.5 g/kg fish meal (Pesquera Diamante, Peru) as the main protein source was used for comparison. Feed preparation followed the protocol described in our previous study (Lin, 2014). The dietary Se concentrations were analysed by using atomic-absorption spectrophotometry with a hydride-generation system (ZA-3000, Hitachi Co., Tokyo, Japan), and the analysed values were 1.07 (all fish meal), 0.77 (control), 1.00 (0.3 SeMet), 1.30 (0.6 SeMet), 1.75 (1.0 SeMet), 1.03 (0.3 OH-SeMet), 1.33 (0.6 OH-SeMet) and 1.71 mg Se/kg diet (1.0 OH-SeMet). Proximate composition was measured based on the methods described by AOAC (1995).

TABLE 1. Formulation, proximate composition and selenium (Se) concentration of experimental diets
Reference Control Selenomethionine (SeMet) Hydroxyl-selenomethionine (OH-SeMet)
0.3 mg/kg 0.6 mg/kg 1.0 mg/kg 0.3 mg/kg 0.6 mg/kg 1.0 mg/kg
Ingredient (g/kg)
Fish meal 580.5 348.3 348.3 348.3 348.3 348.3 348.3 348.3
Soybean meal 0 329.2 329.2 329.2 329.2 329.2 329.2 329.2
Fish oil 28.0 45.5 45.5 45.5 45.5 45.5 45.5 45.5
Soybean oil 19.7 15.6 15.6 15.6 15.6 15.6 15.6 15.6
Alpha starch 140 45.1 45.1 45.1 45.1 45.1 45.1 45.1
Methionine 0 2.2 2.2 2.2 2.2 2.2 2.2 2.2
Lysine 0 4.5 4.5 4.5 4.5 4.5 4.5 4.5
Glycine 6.7 0 0 0 0 0 0 0
SeMet 0 0 0.29 0.58 0.97 0 0 0
OH-SeMet 0 0 0 0 0 0.15 0.30 0.50
Premix 191 191 191 191 191 191 191 191
Cellulose 40.8 25.4 25.1 24.8 24.4 25.2 25.1 24.9
Composition (%, wet basis)
Moisture 77.9 82.2 86.9 79.0 79.0 78.7 76.6 78.0
Ash 117.2 99.7 95.5 100.0 100.7 98.8 100.3 100.7
Crude protein 472.6 466.8 472.8 465.7 484.1 481.6 481.1 487.2
Ether extract 99.0 100.9 101.5 101.0 102.2 90.5 101.8 97.1
Se (mg Se/kg diet) 1.07 0.77 1.00 1.30 1.75 1.03 1.33 1.71
  • 1 Premix (g/kg): soy lecithin, 10; squid liver meal, 50; scallop meal, 50; wheat gluten, 40; yeast, 10; vitamin premix, 10; mineral premix, 20; choline chloride, 1. Vitamin and mineral (without Se) premix were prepared according to Lin and Cheng (2017).

2.2 Experimental procedure

The animal experiments in this study were in full compliance with the principles specified by the Institutional Animal Care and Use Committee (IACUC) in National Pingtung University of Science and Technology (NPUST), Approval No. NPUST 108-073.

The Elanceolatus juveniles were bought from a local hatchery (Pingtung, Taiwan). When the fish were shipped to the wet laboratory in NPUST, they were acclimated in a 3500 L-FRP tank for 4 weeks until the body weight reached the desired size. During the acclimation period, the fish were fed the commercial feed for the first 2 weeks, and then, they were fed the control diet for another 2 weeks. At the beginning of the feeding trial, 192 juveniles (body weight: 25.99 ± 0.04 g) were randomly stocked in twenty-four 70 L-FRP tanks. The eight experimental diets were estimated in triplicate (n = 3). Each tank was a part of a recirculation system equipped with primary filter, biofilter, ultraviolet and temperature control system (28 ± 1°C). One air-stone was placed in each tank for aeration to maintain the dissolved oxygen concentration ≥6 mg/L. The seawater used in the system was filtered and disinfected (28–30 g/L salinity). The automatically controlled photoperiod was 12 h light (08:00–20:00) and 12 h dark (20:00–08:00). The ambient Se concentration of 1.91 ± 0.14 μg/L was analysed weekly.

Feed was delivered to fish twice per day (9:00 and 17:00) at 3% of their wet weight. Every 2 weeks, fish were weighed in bulk to adjust feed amounts. The feeding status of all fish was recorded every day. Any uneaten feed was collected by siphon and dried to calculate the actual feed intake. After the 8-week feeding trial, all the fish were weighed individually to measure growth parameters. Calculation of weight gain, feed efficiency and survival were presented as follows:

Weight gain (%) = (final body weight–initial body weight)/initial body weight × 100.

Feed efficiency = (final body weight–initial body weight)/feed intake.

Survival (%) = final fish number/initial fish number × 100.

2.3 Sampling and analysis

All fish in each tank were weighed individually for the final weight after 24 h of starvation. After weighing, two fish were collected randomly from each tank and pooled for whole body Se concentration analysis. Another three fish were collected from each tank for liver and muscle analysis. The livers of these fish were dissected and washed three times with sterile saline, and then pooled for further analysis. Muscle was also sampled from the same fish.

Approximately 0.5 g of feed, liver, muscle and whole body was separately weighed and placed in respective test tubes for digestion with 5 ml 70% HNO3 and 4 ml 70% HClO4 (Echo Chemical CO., LTD, Taiwan) in 100°C water bath for 6 h. Finally, 2 ml 37% HCl (J. T. Baker, USA) was added to each tube to reduce the valence of Se. The Se concentrations were measured by atomic-absorption spectrophotometry with a hydride-generation system (ZA-3000, Hitachi Co., Tokyo, Japan). Whole body Se retention was calculated as: (final body weight × final whole body Se)–(initial body weight × initial whole body Se)]/(feed intake × feed Se)× 100. Hepatic antioxidant enzymes, including glutathione peroxidase (GPx), total superoxide dismutase (SOD) and catalase (CAT), were measured by commercial kits (Randox Crumlin, UK). The protein content in the liver homogenate was analysed using folin phenol reagent (Lowry et al., 1951). Malondialdehyde (MDA) contents were determined to evaluate oxidative status in liver in accordance with the methods of Uchiyama and Mihara (1978). All analyses were performed and evaluated in triplicate.

2.4 Statistical analysis

The assignment of diet and animal trial was performed using a completely randomized design. Normality and variance homogeneity of all data were assessed by using the Kolmogorov–Smirnov test and Bartlett's test, respectively. The factor effect was analysed with one-way ANOVA (analysis of variance) by SAS/PC statistical software (SAS Institute, Cary, NC, USA). A post hoc assessment (Duncan's multiple test) was used to test significant differences among mean values at a 5% probability level (< .05). Excluding all fish meal groups, two-way ANOVA was used to examine the two factors, that is Se source and Se level as well as their interaction. Dose-dependent (linear) effects of graded levels SeMet or OH-SeMet supplementation in SBM-based diets were tested by polynomial orthogonal contrasts with SigmaPlot (10.0 version, Systat Software, Inc, USA).

3 RESULTS

Grouper fed the FM-based reference diet exhibited significantly higher (< .05) final weight, weight gain and feed efficiency than fish fed diets containing SBM, regardless of Se sources and Se levels (Table 2). Survival was 100% in all dietary treatments.

TABLE 2. Initial weight, final weight, weight gain, feed efficiency and survival of grouper fed different diets for 8 weeks
Diets Initial weight (g) Final weight (g) Weight gain (%) Feed efficiency Survival (%)
Reference 25.98 147.26b 466.78b 1.03b 100
Control 26.04 124.03a 376.38a 0.93a 100
0.3 SeMet 25.96 125.75a 384.32a 0.95a 100
0.6 SeMet 25.99 117.51a 352.16a 0.97a 100
1.0 SeMet 26.00 129.09a 396.47a 0.96a 100
0.3 OH-SeMet 26.00 128.85a 395.64a 0.95a 100
0.6 OH-SeMet 25.99 125.95a 384.58a 0.93a 100
1.0 OH-SeMet 25.98 127.73a 391.69a 0.97a 100
Pooled SEM 0.01 1.73 6.65 0.01 0
Overall p value .5462 .0012 .0011 .0038 -
Two-way p value
Se source .8574 .5945 .5972 .2370 -
Se level .1845 .2345 .1961 .5604 -
Se source × Se level .7718 .7731 .7959 .4811 -
  • Abbreviations: SeMet, selenomethionine; OH-SeMet, hydroxyl-selenomethionine.
  • 1 All data are presented as mean values (n = 3). Mean values in the same column with different superscript letters are significantly different (p < .05) from Duncan's multiple range test.

Whole body, muscle and hepatic Se concentrations were significantly affected by the dietary groups (Table 3). The concentration of Se in whole body, muscle and liver depended on the Se concentrations in experimental diets, but not on its form of supplementation. Fish fed the reference diet had significantly higher whole body, muscle, and hepatic Se concentrations than fish fed the control diet.

TABLE 3. Whole body, muscle and hepatic Se concentration of grouper fed different diets for 8 weeks
Diets Whole body Se (μg/g) Muscle Se (μg/g) Hepatic Se (μg/g)
Reference 0.35bc 0.17b 1.78cd
Control 0.17a 0.07a 1.14a
0.3 SeMet 0.20a 0.15ab 1.36ab
0.6 SeMet 0.31b 0.21bc 1.65bc
1.0 SeMet 0.46cd 0.30cd 1.98d
0.3 OH-SeMet 0.27ab 0.15ab 1.40ab
0.6 OH-SeMet 0.36bc 0.21bc 1.86cd
1.0 OH-SeMet 0.48d 0.31d 2.02d
Pooled SEM 0.02 0.02 0.07
Overall p value <.0001 .0004 <.0001
Two-way p value
Se source .1588 .9951 .3454
Se level <.0001 <.0001 <.0001
Se source × Se level .7249 .9952 .7786

Note:

  • All data are presented as mean values (n = 3). Mean values in the same column with different superscript letters are significantly different (p < .05) from Duncan's multiple range test.
  • Abbreviations: SeMet, selenomethionine; OH-SeMet, hydroxyl-selenomethionine.

Final Se contents were generally reflected in the total Se intake in all dietary treatments (Table 4). Se retention, in descending order, was highest in fish fed the FM diet, followed by fish fed diets containing 1.0 OH-SeMet, 1.0 SeMet, 0.6 OH-SeMet, 0.3 OH-SeMet, 0.6 SeMet and the lowest in fish fed the control diet. Selenium source, Se level and their interaction were significant (< .05) in whole body Se retention. Whole body Se retention increased with increment of dietary Se levels, whereas Se retention was higher in OH-SeMet group than that in SeMet group.

TABLE 4. Whole body Se retention of grouper fed different diets for 8 weeks
Diets Initial Se content (μg) Final Se content (μg) Total Se intake (μg) Whole body retention (%)
Reference 3.68 51.41f 137.15c 34.81d
Control 3.69 20.91a 88.46a 19.49a
0.3 SeMet 3.68 25.74b 115.60b 19.08a
0.6 SeMet 3.68 37.00d 143.39c 24.75b
1.0 SeMet 3.68 59.19g 207.80f 26.72b
0.3 OH-SeMet 3.68 34.52c 121.30b 25.43b
0.6 OH-SeMet 3.68 45.63e 157.11d 26.74b
1.0 OH-SeMet 3.68 60.70g 196.56e 29.01c
Pooled SEM 0.00 0.68 2.67 1.01
Overall p value .5776 <.0001 <.0001 <.0001
Two-way p value
Se source .8985 <.0001 .2529 <.0001
Se level .2044 <.0001 <.0001 <.0001
Interaction .7792 <.0001 .0009 .0013
  • 1 Initial Se content (μg) = initial body weight (g) × initial body Se concentration (μg/g); Final Se content (μg) = final body weight (g) × final body Se concentration (μg/g); Total Se intake (μg) = total feed intake (g) × dietary Se concentration (μg/g); Se retention = (final Se content – initial Se content)/(total Se intake) × 100.

Hepatic GPx activity was the higher in fish fed the fish meal, 1.0 SeMet, 0.6 OH-SeMet and 1.0 OH-SeMet diets than in fish fed the control diet (Table 5). Fish fed FM, 1.0 SeMet and 1.0 OH-SeMet diets had higher hepatic SOD activity than fish fed control and 0.6 SeMet diets. Hepatic CAT activity was highest in fish fed FM, 1.0 SeMet and 1.0 OH-SeMet diets, followed by fish fed 0.3 SeMet and 0.3 OH-SeMet diets, and lowest in fish fed control and 0.6 SeMet diets. Hepatic MDA content was highest in fish fed control and 0.6 SeMet diets, followed by fish fed 0.3 SeMet and 0.3 OH-SeMet diets, and lowest in fish fed FM, 1.0 SeMet and 1.0 OH-SeMet diets. Hepatic GPx and SOD activities were significantly affected by Se level, but not Se source. For CAT activity and MDA content, both Se source and Se level, and their interaction were significant (< .05). Hepatic CAT activity increased but MDA content decreased with increment of dietary Se levels, whereas Se retention was higher in OH-SeMet group than that in SeMet group.

TABLE 5. Hepatic glutathione peroxidase (GPx), superoxide dismutase (SOD), catalase (CAT) and malondialdehyde (MDA) concentrations of grouper fed different diets for 8 weeks
Diets GPx (unit/mg protein) SOD (unit/g protein) CAT (unit/mg protein) MDA (nmol/g tissue)
Reference 1.24cd 172.57d 2.82d 43.75a
Control 0.90a 139.16a 1.51a 62.85d
0.3 SeMet 1.10bc 143.27ab 1.97b 51.80c
0.6 SeMet 0.99ab 135.04a 1.58a 62.59d
1.0 SeMet 1.29d 158.95bcd 2.50cd 46.05ab
0.3 OH-SeMet 1.14bcd 146.55abc 1.99b 50.32c
0.6 OH-SeMet 1.18cd 151.00abc 2.22bc 48.12bc
1.0 OH-SeMet 1.32d 162.30cd 2.83d 44.83ab
Pooled SEM 0.03 2.90 0.11 1.52
Overall p value .0008 .0013 <.0001 <.0001
Two-way p value
Se source .1413 .1520 .0087 .0002
Se level <.0001 .0049 <.0001 <.0001
Se source × Se level .3983 .4716 .0429 .0001
  • Abbreviations: SeMet, selenomethionine; OH-SeMet, hydroxyl-selenomethionine.
  • 1 All data are presented as mean values (n = 3). Mean values in the same column with different superscript letters are significantly different (p < .05) from Duncan's multiple range test.

Linear responses of graded levels of SeMet and OH-SeMet on whole body, muscle, hepatic Se concentrations, whole body Se retention, hepatic GPx, SOD, CAT activities and MDA content are presented in Table 6. All parameters, except MDA, showed positive correlation (< .05). When calculating the slope ratio of OH-SeMet and SeMet, the ratios ranged from 1.03 to 1.60.

TABLE 6. Linear responses of grade levels of selenomethionine (SeMet) or hydroxyl-selenomethionine (OH-SeMet) for giant grouper
Parameters Equation p Value OH-SeMet/SeMet slope ratio
Whole body Se concentration (μg/g, wet basis)
SeMet Y = 0.31X−0.08, r = .91 <.0001 1.03 (0.32/0.31)
OH-SeMet Y = 0.32X−0.07, r = .95 <.0001
Muscle Se concentration (μg/g, wet basis)
SeMet Y = 0.23X−0.09, r = .88 .0002 1.09 (0.25/0.23)
OH-SeMet Y = 0.25X−0.10, r = .94 <.0001
Hepatic Se concentration (μg/g, wet basis)
SeMet Y = 0.86X+0.50, r = .90 <.0001 1.14 (0.98/0.86)
OH-SeMet Y = 0.98X+0.43, r = .89 <.0001
Whole body Se retention (%)
SeMet Y = 8.15X+12.31, r = .96 .0001 1.14 (9.30/8.15)
OH-SeMet Y = 9.30X+13.91, r = .91 <.0001
Hepatic glutathione peroxidase activity (unit/mg protein)
SeMet Y = 0.34X+0.66, r = .71 .0091 1.21 (0.41/0.34)
OH-SeMet Y = 0.41X+0.64, r = .84 .0006
Hepatic superoxide dismutase activity (unit/g protein)
SeMet Y = 17.78X+122.67, r = .61 .0353 1.34 (23.78/17.78)
OH-SeMet Y = 23.78X+120.98, r = .68 .0154
Hepatic catalase activity (unit/mg protein)
SeMet Y = 0.84X+0.88, r = .71 .0091 1.60 (1.34/0.84)
OH-SeMet Y = 1.34X+0.51, r = .94 <.0001
Hepatic malondialdehyde content (nmol/g tissue)
SeMet Y = −12.87X+71.33, r = −.62 .0304 1.35 (−17.38/−12.87)
OH-SeMet Y = −17.38X+72.56, r = −.88 .0001

4 DISCUSSION

Poor growth performance was exhibited in fish fed the SBM-based control diet, compared with grouper fed the reference (all FM) diet. These results were consistent with the results of our previous study in the same species (Lin & Cheng, 2017; Lin & Lu, 2020). This finding can be attributed to the limited amino acids and the presence of antinutritional factors in SBM (NRC, 2011). However, the Se concentration/availability in SBM was primarily discussed in the present study. Selenium concentration in the control diet was 0.77 mg Se/kg, which was less than Se concentration in the reference diet (1.07 mg Se/kg). Selenium derived from a fish dietary source is considered to be more readily utilized than that derived from a plant source, because Se concentration in plants is rather low and is largely dependent on soil Se concentration (Reilly, 1998). In addition, Cubadda et al. (2010) indicated that plants are capable of converting the Se ion to seleno-amino acids, such as Se-methyl-selenocysteine (Se-MeSeCys) and γ-glutamyl-Se-methyl-selenocysteine (γ-Glu-MeSeCys). Tie et al. (2015) found the Se-MeSeCys accounted for about 66% of the seleno-amino acid but only a trace in SeMet. In pigs, SeMet has been demonstrated to have higher availability than Se-MeSeCys (Zhang et al., 2020). Therefore, after replacing FM with SBM, the essentiality of Se supplementation in diet should be carefully reconsidered.

Although the Se requirement for giant grouper is yet to be estimated, the amount in the SBM-based control diet was a little lower than the recommended level for Epinephrine malabaricus (0.9–0.98 mg Se/kg; Lin, 2014). However, it should be noted that the increase in the dietary Se level did not improve growth performance when the Se-supplemented diets were included with SBM. It is difficult to clarify the essentiality of Se for giant grouper based on growth data. First, the rearing water containing a Se concentration of 1.91 ± 0.14 μg/L may be a possible source of Se for the fish. Furthermore, our experimental diets were formulated as practical diets, resulting in a control diet that contained a certain amount of Se (0.77 mg Se/kg). An 8-week feeding trial may not be long enough to observe the changes in growth. Therefore, physiological parameters other than growth are required.

Results of the current study suggest that hepatic oxidative status was induced by grouper fed the SBM-based control diet, and recovery from this stress is achievable through Se supplementation. The high content of MDA in fish fed SBM-based diet has also been observed in previous studies (Lin & Cheng, 2017; Lin & Lu, 2020). The phenomenon may be caused by the antigenic component derived from SBM known as β-glycine, which has been demonstrated to induce oxidative stress in common carp (Zhang et al., 2013). Another possibility is that the phenomenon may be driven by the chelation of copper and zinc by phytic acid from SBM, because the two microelements are central metals of superoxide dismutase. This possible causative interaction is supported by the observed reduction in SOD activity in grouper fed the control diet (Table 5). Interestingly, the oxidative stress was improved by Se dietary supplementation for Elanceolatus, showing a contrary trend with GPx activity, a selenoprotein enzyme, that increased incrementally with Se supplementation levels. It suggested an inadequate Se status, such as low Se concentration or poor Se bioavailability in the SBM-based control diet. Therefore, Se supplementation in a SBM-based diet for giant grouper is recommended. Moreover, it should be noted that the GPx activities increased linearly (< .05) with an increase in supplementation levels (Y = 0.34X + 0.66, r = .71 for SeMet groups; Y = 0.41X + 0.64, r = .84 for OH-SeMet; Table 6). Although an adequate Se level was not obtained by the GPx activities in our study, the results clearly indicated the importance of Se supplementation for grouper fed the SBM-based diet.

Like the responses of GPx and SOD, hepatic CAT, another antioxidant enzyme, also exhibited an increasing trend with dietary Se levels. All three antioxidant enzymes, in combination with MDA content, have commonly been used to establish antioxidative capacity in animals (Ashouri et al., 2015; Han et al., 2011; Lin, 2014). Reactive oxygen species (ROS) are highly reactive chemical molecules formed as a result of the electron receptivity of O2. Superoxide dismutase can convert superoxide anion (O2) to hydrogen peroxide (H2O2), then GPx or CAT can catalyse the reduction of H2O2 to H2O (Tapiero et al., 2003). In addition to GPx, hepatic SOD and CAT were enhanced by Se supplementation in the diet of common carp (Ashouri et al., 2015). Our results are compatible with this finding. It suggests that dietary Se supplementation can lead to a general improvement in antioxidative capacity in giant grouper fed the SBM-based diet.

When looking into the results of two-way ANOVA, a significant difference (p < .05) was found in the Se source for hepatic MDA content (Table 5). It appears that fish fed diets with SeMet exhibited a higher oxidative status than fish fed diets with OH-SeMet. Moreover, a negative correlation (p < .05) was shown between dietary Se levels and hepatic MDA content (Table 6) in both Se source treatments. The slope of linear response in OH-SeMet group was about 135% of that in SeMet group (−12.87 vs −17.38), suggesting the efficacy on antioxidative capacity of OH-SeMet is greater than SeMet for giant grouper. A similar response was also observed in whole body Se retention and hepatic CAT activity of the fish. Although GPx and SOD activities were not influenced by Se sources, the linear responses (p < .05) also showed higher slope in fish fed diets with OH-SeMet than in fish fed diets with SeMet (121% and 134% for GPx and SOD, respectively). Hydroxyl-selenomethionine (2-hydroxy-4-methylselenobutanoic acid) is a hydroxyanalogue of selenomethionine (Briens et al., 2013). Until now, no study compared the utilization and availability between OH-SeMet and SeMet in fish. Without addressing their availability, Geraert et al. (2015) and Surai et al. (2018) indicated that OH-SeMet is more stable than SeMet during storage. Our study firstly demonstrated that OH-SeMet is a superior Se source for enhancing antioxidative capacity and whole body Se retention.

In conclusion, dietary replacement of FM protein with SBM at 40% reduced growth performance, Se retention and antioxidant capacity of giant grouper. Dietary supplementation with both of the tested Se sources enhanced Se retention and antioxidative capacity, but not growth performance of the fish. Based on hepatic CAT activity, MDA content and whole body Se retention, OH-SeMet exhibited greater efficacy than SeMet for E. lanceolatus.

ACKNOWLEDGEMENTS

This research project was fully sponsored by Ministry of Science and Technology, Taiwan with grant number MOST 109-2313-B-020-001-MY3.

    CONFLICT OF INTEREST

    The authors report no declarations of interest.

    Data Availability Statement

    All data used in this study are available from the corresponding authors on reasonable request.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.