1 INTRODUCTION

Vitamin E (VE), consisting of tocopherols, tocotrienols and the derivatives, serves as a lipid-soluble antioxidant providing nutrients to animals including fish (Bramley et al., 2000). Its primary function is to act as a free radical scavenger that protects cells and tissues against the deleterious effects of free radicals (Liu et al., 2018; Naderi et al., 2019; Wang et al., 2019). Besides, VE can increase antioxidant ability of fish by enhancing the antioxidant enzyme activities containing superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GSH-Px), such as largemouth bass (Micropterus salmoides) (Li et al., 2018), hybrid snakehead (Channa argus × Channa maculata) (Zhao et al., 2017) and grass carp (Ctenopharyngodon idellus) (Li et al., 2014). Moreover, as an antioxidant, VE supplement could inhibit the oxidation and rancidity of unsaturated fatty acids in animal feed or tissues and improve the growth and feed efficiency of fish (Min et al., 2017; Wassef et al., 2001). Previous studies demonstrated that VE could protect polyunsaturated fatty acids (PUFA) from auto-oxidation, which contributes to increasing PUFA contents in muscle, thereby improving the nutritional quality of meat (Belles et al., 2018; Hamre, 2011; Navarro et al., 2012). Furthermore, VE also has an antioxidant property in the post-mortem tissues and dietary VE supplementation improved lipid stability of cooked meat, which were reported in terrestrial animals such as rabbit (Tres et al., 2008) and chicken (Cortinas et al., 2005). However, similar effects of VE on the amount and type of fatty acid in the cooked flesh have not yet been reported in fish. In addition, several studies have shown that VE can affect fat deposition by regulating genes responsible for lipid metabolism, such as lipogenesis, lipolysis and transport in lambs and broiler chickens (Gonzalez-Calvo et al., 2017; Zhang et al., 2020). Also, VE has been reported to influence hepatic PUFA desaturation and elongation in Zebrafish (Danio rerio) (Lebold et al., 2011). Nevertheless, information on the effects of dietary VE on PUFA β-oxidation and synthesis of fish and other aquatic animals is still limited.

Vitamin E supplementation is considered essential for most fish species, as they cannot synthesize it (Li et al., 2018). Accumulating data indicate that dietary VE deficiency in fish usually results in a number of deficiency symptoms such as muscle degeneration, internal haemorrhaging, retarded growth and depressed immunity (Chen et al., 2004; Head et al., 2021; Kocabas & Gatlin, 1999; Pan et al., 2017). In contrast, overdoes of VE disturbs normal metabolism and consequently resulted in metabolic disorder (Kaewsrithong et al., 2002; Zhang et al., 2016). Considering this, it is necessary to establish an optimal dietary VE level for cultured species. To date, dietary VE requirement has been demonstrated in many fish species, which varying with each other, for example 28 mg/kg diet for hybrid striped bass (Morone chrysops female × Morone saxatilis male) (Kocabas & Gatlin, 1999), 78 to 111 mg/kg for cobia (Rachycentron canadum) (Zhou et al., 2013) and 451 mg/kg for meagre (Argyrosomus regius) (Lozano et al., 2017). So, it is difficult to extrapolate from one fish species to another, and thus, investigating dietary VE requirements for each cultured species is necessary.

Golden pompano (Trachinotus ovatus) has been one of the important cultured marine fish in China because of its fast growth, efficient feed conversion, preferred flesh quality and high economic value (Yin et al., 2015; Zhang, Wang, et al., 2019). To date, some studies have been conducted on the nutrient requirement for this fish, such as optimal dietary carbohydrate, protein and lipid levels (Wang et al., 2013; Zhou et al., 2015; Li et al., 2019; Zhao et al., 2020). However, its requirement on VE remains unknown. Thus, this study was to determine the dietary VE requirement of juvenile T. ovatus and evaluate the effect of dietary VE on PUFA β-oxidation by investigating the effects of dietary VE levels on the growth performance, antioxidant capacity, tissue fatty acid composition, fatty acid stability in muscle during cooking and genes responsible for lipid metabolism of this species. The results obtained here will help to increase the database of the nutritional requirements of T. ovatus and promote the development of nutritionally balanced compound feed.

2 MATERIALS AND METHODS

3 RESULTS

3.1 Growth performance and feed utilization

The FCR, PER and survival rate were no significant differences (p > .05) among all groups. However, the final weight, WG and SGR all increased as VE levels increased from 34.56 to 61.16 mg/kg (D1–D3), but decreased significantly (p < .05) with further increasing levels (Table 3). Based on the second-order polynomial regression analysis of WG against VE levels, the optimal VE level for juvenile T. ovatus was 90.75 mg/kg (Figure 1).

TABLE 3. Growth performance and feed utilization of T. ovatus fed diets with different VE levels

	Diets					Pooled SEM	p-Value	Regression (p, r2)
	D1	D2	D3	D4	D5			Linear		Quadratic
Initial weight (g)	13.27	13.33	13.67	13.47	13.27	0.10	.737	.820	.004	.524	.102
Final weight (g)	45.56a	46.78ab	52.64c	49.49b	46.36a	0.78	.002	.885	.002	.019	.484
WG1 (%)	243.58a	251.14a	285.60b	267.48ab	249.44a	5.07	.023	.954	.000	.046	.401
SGR2 (%/day)	2.20a	2.24a	2.41b	2.32ab	2.23a	0.03	.025	.983	.000	.045	.403
FCR3	1.68	1.46	1.53	1.39	1.41	0.04	.221	.084	.212	.105	.313
PER4	30.50	34.75	32.96	36.36	36.02	0.89	.205	.072	.228	.104	.314
SUR5 (%)	100	100	100	100	100	–	–	–	–	–	–

Note

• Data represent the means of three biological replicates in each group (n = 3) and SEM is pooled of standard error of mean. Without sharing a common superscript letter in the same row indicate significant differences (p < .05).
• ¹ Weight gain (WG, %) = 100 × (final weight − initial weight)/initial weight;
• ² Specific growth rate (SGR, %/day) = [Ln(final weight (g)) − Ln(initial weight (g))]/days × 100;
• ³ Feed conversion ratio (FCR) = total feed fed/(final weight − initial weight);
• ⁴ Protein efficiency ratio (PER, %) = wet weight gain (g)/total protein fed (g);
• ⁵ Survival rate (SUR, %) = survived fish number/total fish number × 100.

3.3 Vitamin E content and the antioxidant indices

As can be seen from Figure 2, serum CAT and hepatic SOD activities showed no significant differences (p > .05) among the treatments. Hepatic VE content increased significantly (p < .05) with increasing VE levels up to 144.88 mg/kg. The activities of SOD, GSH-Px and T-AOC in serum and GSH-Px, CAT and T-AOC in liver were increased significantly (p < .05) as VE levels increased from 34.56 to 61.16 mg/kg, but decreased with further increasing levels (except for hepatic CAT activities and T-AOC).

3.4 Tissue fatty acid composition

As can be seen from Table 5, hepatic linolenic acid (ALA), saturated fatty acids (SFA) and monounsaturated fatty acids (MUFA) contents showed no significant differences (p > .05) among all the treatments. Hepatic linoleic acid (LA), 20:4n-3, arachidonic acid (ARA), docosahexaenoic acid (DHA), n-6 PUFA and n-3 PUFA contents, as well as n-3/n-6 ratio all increased significantly (p < .05) as VE levels increased from 34.56 to 61.16 mg/kg, while decreased with further increasing levels, but no significant difference (except for 20:4n-3 and ARA content and n-3/n-6 ratio). Additionally, hepatic eicosapentaenoic acid (EPA) content increased significantly (p < .05) as VE levels increased from 34.56 to 91.06 mg/kg, but decreased significantly (p < .05) with further increasing levels.

TABLE 5. Fatty acid composition of liver of T. ovatus fed diets with different VE levels (mg/g sample, dry weight)

Fatty acids	Diets					Pooled SEM	p-Value	Regression (p, r2)
	D1	D2	D3	D4	D5			Linear		Quadratic
18:2n-6 (LA)	3.61a	4.12ab	4.64b	4.89b	4.42ab	0.15	.037	.125	.171	.004	.594
18:3n-3 (ALA)	3.16	3.17	3.64	3.38	3.19	0.14	.822	.954	.000	.667	.065
20:4n-3	1.76a	2.12ab	2.56b	2.09ab	2.02a	0.09	.047	.952	.000	.228	.218
20:4n-6 (ARA)	0.34a	0.45b	0.60c	0.50b	0.48b	0.02	.002	.295	.084	.032	.435
20:5n-3 (EPA)	0.21a	0.26ab	0.27bc	0.38d	0.33c	0.02	.000	.006	.453	.000	.809
22:6n-3 (DHA)	1.65a	3.93b	4.65b	4.46b	4.53b	0.33	.001	.034	.302	.003	.620
∑SFA	31.55	32.94	41.46	32.27	31.36	1.57	.200	.565	.026	.505	.108
∑MUFA	39.66	42.96	55.00	42.13	39.62	2.07	.073	.507	.035	.312	.177
∑n-6 PUFA	5.52a	6.06ab	7.07b	6.94b	6.51ab	0.20	.043	.192	.127	.016	.500
∑n-3 PUFA	7.21a	10.36b	13.01c	11.15bc	11.10bc	0.57	.002	.177	.136	.022	.472
n-3/n-6 ratio	0.96a	1.33b	1.50c	1.38bc	1.31b	0.05	.000	.251	.100	.007	.563

Note

• Data represent the means of three biological replicates in each group (n = 3), and SEM is pooled of standard error of mean. Without sharing a common superscript letter in the same row indicate significant differences (ANOVA analysis; p < .05).
• ΣSFA, ΣMUFA, Σn-6 PUFA and Σn-3 PUFA are consistent with Table 2.

The fatty acid composition of raw and cooked muscle was shown in Figure 3. The contents of LA, ARA, EPA and n-6 PUFA in raw muscle were no significant differences (p > .05) among all groups. The contents of ALA, DHA and n-3 PUFA in raw muscle of fish fed diets with moderate dietary VE level (61.16 and 91.06 mg/kg) were higher than those of other groups. Moreover, fatty acid contents in cooked muscle of fish fed diets with 61.16 mg/kg VE were all significantly (p < .05) higher than those of fish fed diets containing 34.56, 47.35 and 144.88 mg/kg VE. In addition, the contents of LA, ALA, ARA, EPA, DHA, n-6 PUFA and n-3 PUFA in the cooked muscle of fish fed diets with low VE levels (34.56 and 47.35 mg/kg) were significantly lower than those of raw muscle. However, fatty acid contents in raw and cooked muscle were not significantly different (p > .05), as VE levels increased from 61.16 to 144.88 mg/kg.

3.5 Hepatic lipid metabolism-related gene expression

The expression levels of cpt1, l-fabp, pparα, elongation of very-long-chain fatty acids 4 (elovl4) and elongation of very-long-chain fatty acids 5 (elovl5) in liver all decreased as VE levels increased from 34.56 to 91.06 mg/kg, but increased significantly (p < .05) with further increasing levels. In addition, the expression levels of fatty acid synthase (fas) and delta 6 fatty acid desaturase (Δ6 fad) both were decreased as VE levels increased from 34.56 to 61.16 mg/kg, but increased significantly (p < .05) with further increasing levels (Figure 4).

4 DISCUSSION

Nutritional balances have profound influences on fish health and growth, particularly during larval and juvenile stages. In the present study, the final weight, WG and SGR of T. ovatus markedly increased with increasing dietary VE levels from 34.56 to 61.16 mg/kg. Results from the parabola model indicated that VE requirement of T. ovatus juveniles for optimum growth was considered to be 90.75 mg/kg, which is higher than optimum VE requirement of most fish species, for example 55.53 mg/kg for blunt snout bream (Megalobrama amblycephala) (Zhang et al., 2017), 40 mg/kg for parrot fish (Oplegnathus fasciatus) (Galaz et al., 2010) and 31 mg/kg for red drum (Sciaenops ocellatus) (Peng & Gatlin, 2009). However, there was no effect on the growth with the increase in dietary VE levels in Japanese eel (Anguilla japonica) (Shahkar et al., 2018) and meagre (Lozano et al., 2017). This might be because the changes in VE requirements depending on fish species, size, age and culture conditions (Huang et al., 2020; Lozano et al., 2017). In addition, negative effects on growth performance were also observed in T. ovatus, showing that significantly decreased performance in higher VE exceeds 91.06 mg/kg. Similar cases also appeared in several previous studies conducted in cobia (Zhou et al., 2013), yellow catfish (Pelteobagrus fulvidraco) (Lu et al., 2016) and sea cucumber (Apostichopus japonicus) (Wang et al., 2015). This result may be attributed to that high dietary VE level could result in toxic effects, thus inhibiting the growth of fish (Galaz et al., 2010; Kiron et al., 2004).

In the present study, liver VE deposition increased directly in response to the increased level of dietary VE in T. ovatus as reported for other fish containing rainbow trout (Salmo gairdneri) (Puangkaew et al., 2005), coho salmon (Oncorhynchus kisutch) (Huang et al., 2004) and red sea bream (Pagrus major) (Gao et al., 2012). The improvement of VE content in the liver would be expected to increase the antioxidant ability of the biological system (Zhao et al., 2017). Lu et al. (2016) reported that dietary VE can enhance antioxidant ability of organisms by increasing the activities of SOD and CAT. In this study, the activities of serum SOD, GSH-Px and liver GSH-Px, as well as T-AOC in serum and liver of juvenile T. ovatus all increased remarkably as dietary VE levels increased from low-to-moderate levels, then decreased notably with further increasing VE levels. This indicated that optimal VE level might enhance liver antioxidant ability of T. ovatus. This was supported by the following facts: (i) SOD and GSH-Px are important antioxidant enzymes closely involved in the removal of superoxide anions (Dandapat et al., 2000); (ii) T-AOC could directly reflect the antioxidant capacity of fish, and low T-AOC means a decreased antioxidant capacity (Tan et al., 2017). Meanwhile, high VE levels resulted in low activities of liver antioxidant enzymes in fish, which could adversely affect liver antioxidant ability. This might be partly due to that a high dietary VE dose could induce toxic effects, thus inhibiting the activities of liver antioxidant enzymes (Galaz et al., 2010; Kiron et al., 2004).

Previous studies have shown that VE can affect fat deposition (Gonzalez-Calvo et al., 2017; Zhang et al., 2020). In this study, whole-body protein and muscle lipid contents of juvenile T. ovatus showed no significant differences among all the treatments. However, whole-body lipid contents were significantly increased by elevating the VE levels. This might due to that VE functions as a lipid-soluble antioxidant, protecting biological membranes and lipoproteins against oxidation (Zhang et al., 2017), which leads to an increase in lipid contents in fish. In addition, the lowest muscle protein content was found in fish fed the diets with 144.88 mg/kg VE, which was in line with the results obtained in rohu (Labeo rohita), where fish fed with the highest VE levels had the lowest protein content (Sau et al., 2004). Nevertheless, the specific reason for the change caused by excessive doses of dietary VE in muscle protein is unclear and supposed to do further investigations.

Moreover, in this study, the fatty acid composition of the liver was affected by dietary VE levels. Hepatic 20:4n-3, ARA, EPA, DHA, n-6 PUFA and n-3 PUFA contents significantly increased with increasing levels of dietary VE from 34.56 to 91.06 mg/kg. Interestingly, compared with the VE deficiency group (D1), hepatic DHA contents were over twofold increased in VE supplement groups (D2–D5). The positive correlation between liver PUFA and dietary VE levels was also found in channel catfish (Ictalurus punctatus) (Lim et al., 2010) and red sea bream (Gao et al., 2012). According to previous studies, these results might due to the fact that VE can prevent fatty acids against lipid oxidation, thus increasing PUFA contents in the liver (Gao et al., 2012; Hamre & Lie, 1995). With increasing VE levels from 91.06 to 144.88 mg/kg, hepatic 20:4n-3, ARA, DHA, n-6 PUFA and n-3 PUFA contents were slightly reduced. The results may attributed to that high VE levels could result in cell death by increasing cytotoxic activity (Cuesta et al., 2001), as correspondingly increases the degradation of PUFA detached from the cell membrane. Here, in order to characterize the underlying mechanisms related to the regulation of lipid metabolism by VE, molecular investigations were conducted in this study. Hepatic mRNA levels of cpt1, l-fabp, pparα, fas, elovl4 and elovl5 were reduced by the supplementation of dietary VE levels from 34.56 to 91.06 mg/kg. The results indicated a decrease in PUFA β-oxidation and synthesis in the liver of T. ovatus fed the D1–D4 diets. This was supported by the following facts: (i) PPARα activation could accelerate fatty acid oxidation by regulating genes involved in fatty acid transport (fabp), β-oxidation (cpt1) and mitochondrial respiration in various tissues (Lu et al., 2013); (ii) the increased Fas activity could promote body fatty acid biosynthesis (Xu et al., 2018); and (iii) both Elovl4 and Elovl5 are the key elongase involved in PUFA synthesis of fish (Castro et al., 2016; Monroig et al., 2012). The decreased PUFA β-oxidation in liver might be associated with the antioxidant effects of VE (Zhang et al., 2017). It has been proved that dietary VE supplementation could inhibit lipid oxidation in tissues by suppressing the expression of pparα (Zhang et al., 2017). In addition, the decrease in PUFA synthesis is also not surprising since optimal VE supplementation in diets has been demonstrated to inhibit liver fatty acid synthesis by down-regulating the expression of pparγ, which plays an important role in lipid accumulation (Stumvoll & Haring, 2002; Zhang et al., 2017). Zhang et al. (2020) also reported that VE could restrain fatty acid synthesis by down-regulating the expression of fas and acc in broiler chickens. Then, hepatic mRNA levels of pparα, cpt1, l-fabp, fas, Δ6 fad, elovl4 and elovl5 were all significantly increased as VE levels increased from 91.06 to 144.88 mg/kg, suggesting an enhancement of PUFA β-oxidation and synthesis in the liver of T. ovatus. This may be due to the fact that high dietary VE levels can cause cell death by increasing cytotoxic activity (Cuesta et al., 2001), as might reduce lipid stability on the cell membrane, thus increasing PUFA β-oxidation and auto-oxidation. As a result, when PUFA oxidation products increase, they would inevitably promote PUFA synthesis, which may be reflected by the up-regulation of fas, Δ6 fad, elovl4 and elovl5 mRNA levels.

Fish contains high n-3 PUFA levels, especially in EPA and DHA, which are beneficial to human health (Gogus & Smith, 2010). However, PUFA is potentially susceptible to lipid peroxidation (Hamre, 2011), which results in the compromise of the nutritional values of fish. The present results found that suitable dietary VE level (61.16–91.06 mg/kg) could improve the contents of ALA, DHA and n-3 PUFA in raw muscle. This finding was in line with the results obtained by Lozano et al. (2017) and Xu et al. (2019), who reported optimum dietary VE supplements could improve PUFA content in muscle of meagre and cobia, respectively. In addition, the EPA and DHA contents of the cooked muscle in fish fed the diet with low VE levels were significantly lower than those of the raw muscle. However, no significant differences were observed between the raw and cooked muscle, when VE levels increased from 61.16 to 144.88 mg/kg. The results might be attributed to that VE could protect biological membranes containing PUFA against free radicals formation and effectively minimize lipid peroxidation, thus increasing PUFA content and stability in fish (Mourente et al., 2000; Tocher et al., 2003). In general, dietary supplementation of more than 47.35 mg/kg could avoid the reduction in muscle PUFA content and stability, thereby elevating the flesh quality of juvenile T. ovatus.

The results in this study indicated that dietary VE supplementation could improve growth performance in T. ovatus. Based on the second-order polynomial regression analysis of WG against VE levels, the optimal VE levels for juvenile T. ovatus were 90.75 mg/kg in diet. Moreover, increasing dietary VE (from 34.56 to 91.06 mg/kg) increased the antioxidant ability and muscle DHA and n-3 PUFA contents. Besides, supplementation of more than 47.35 mg/kg VE could protect muscle PUFA content from decreasing during cooking. In addition, the optimal dietary VE may inhibit PUFA β-oxidation by suppressing the expression of cpt1, l-fabp and pparα to increase the PUFA contents and restrain PUFA synthesis by down-regulating the expression of fas, Δ6 fad, elovl4 and elovl5 in liver of juvenile T. ovatus.

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.