Comparative effects of dietary supplementation with maggot meal and soybean meal in gibel carp (Carassius auratus gibelio) and darkbarbel catfish (Pelteobagrus vachelli): growth performance and antioxidant responses

A 6‐week growth trial was conducted to investigate the effect of dietary supplementation with maggot meal (MGM) and soybean meal (SBM) on the growth performance and antioxidant responses of gibel carp (GC) and darkbarbel catfish (DC). The basal diet was formulated to contain 114 g kg−1 fish meal (FM) and 200 g kg−1SBM. The basal diet was supplemented with either 280 g kg−1FM (Control), 390 g kg−1MGM or 450 g kg−1SBM to obtain three isonitrogenous (crude protein: 380 g kg−1) and isocaloric (gross energy: 16 kJ g−1) diets. For GC, a significant decrease in specific growth rate (SGR) was only observed in fish fed the SBM diet compared with the control (P < 0.05). Principal components analysis (PCA) of GC showed a higher similarity in antioxidant response to dietary supplementation with MGM and SBM proteins between liver and intestine, but the DC did not. The present results suggest that supplementing 390 g kg−1MGM protein to basal diet cause an enhancement of the antioxidant capacity in GC, but supplementing 390 g kg−1MGM and 450 g kg−1SBM proteins to basal diets resulted in a significant attenuation of the antioxidant capacity in DC.

antioxidant capacity; Carassius auratus gibelio; Pelteobagrus vachelli; principal component analysis; protein sources

Protein is the most expensive component in fish feeds, and the quantity and quality of dietary protein are primary factors influencing fish growth and feed costs (Luo et al. [23] ). Due to the scarcity and cost of FM in the world market, it is imperative to reduce feed cost by exploring cheaper alternative protein in fish diets (Ogunji et al. [30] ). To date, a large number of studies investigated several feed ingredients including both animal protein and plant protein sources. In plant protein sources, SBM is a widely available, economical protein source with relatively high digestible protein and energy contents and good amino acid profile (Hertrampf & Piedad‐Pascual [18] ). There have been many studies on the replacement effects of SBM in diets for many fish species, such as rainbow trout (Refstie et al. [36] ), Atlantic salmon (Refstie et al. [37] ), cobia (Zhou et al. [48] ), cuneate drum (Wang et al. [42] ), tilapia (Shiau et al. [39] ) and Asian seabass (Tantikitti et al. [40] ). On the other hand, many studies in recent years have also shown that many rendered animal protein, such as meat and bone meals, poultry by‐product meal and other animal by‐products are useful for fish feed formulation and comparatively less expensive than FM (Fasakin et al. [13] ; Wang et al. [43] , [44] ; Goda et al. [15] ; Hu et al. [19] ). In addition, based on cost‐effectiveness, availability and crude protein content, MGM has considerable potential in fish feeds (Ogunji et al. [30] , [31] ). As a dietary protein source, MGM has been tested in tilapia and catfish (Fasakin et al. [12] ; Ajani et al. [1] ; Ogunji et al. [30] , [31] ).

However, in the context of researching alternative protein sources for fish, growth, feed utilization and whole body composition were the main parameters for evaluation. Dietary changes in protein sources often cause no grossly observable signs, but they may severely influence the organism's health status, which would not emerge from nutritional parameters (Yang et al. [47] ). The effect of different protein sources on antioxidant capacity should also be taken into consideration, but this has been seldom addressed (Ogunji et al. [30] ). Like all terrestrial animals, fish are also susceptible to the effects of reactive oxygen species (ROS). Free radicals generated in fish tissues are effectively scavenged by the antioxidant defence system that constitutes antioxidant enzymes, such as glutathione peroxidase (GPx), glutathione reductase (GR), superoxide dismutase (SOD) and catalase (CAT), and various antioxidant substances, such as reduced glutathione (GSH). When the activity of the antioxidant defence system decreases or ROS production increases, an oxidative stress may occur (Packer [32] ; reviewed in Martínez‐Álvarez et al. [24] ). Thus, the health of fish is linked to the overproduction of ROS and the production of antioxidants that scavenge ROS and protect cell membranes against these free radicals. In addition, there is considerable variation in the nutritional value or quality of alternative protein sources, especially between animal and plant protein sources (Friedman [14] ), so fishes may differ in their ability to utilize dietary alternative protein sources. Therefore, changes of antioxidant capacity should be considered as one important aspect for evaluating the nutritive value of alternative proteins for animal and plant protein sources including SBM and MGM.

Gibel carp (Carassius auratus gibelio) and darkbarbel catfish (Pelteobagrus vachelli) have been widely cultured in China. Generally, gibel carp is considered an omnivorous fish and feeds mainly on microorganisms, zooplankton and detritus in natural habitat (Pei et al. [34] ). Darkbarbel catfish is considered a mixture of carnivorous and omnivorous species and usually feeds on small prey such as insects and mollusks, and sometimes feeds on phytoplankton, algae and detritus in natural habitat (Wang et al. [41] ; Li et al. [22] ). However, how dietary animal and plant protein sources affect the antioxidant capacity in the two fishes is still poorly understood. The aim of this study was to investigate the effect of dietary supplementation with MGM and SBM proteins on the growth performance of gibel carp and darkbarbel catfish, and the difference in antioxidant response by different fishes.

Gibel carp and darkbarbel catfish were obtained from the Binzi hatchery farm in Hefei, Anhui, China and acclimated in 10 fibreglass tanks (diameter: 60 cm, water volume: 190 L) for 15 days prior to the experiment. During the acclimation period, all fish were fed to satiation with an equal mixture of three experimental diets for both species twice daily (09:00 and 15:00 h).

Maggot meal used in this study was the processed housefly (Musca domestica) larvae. Housefly larvae were grown on lays of substrate (5–10 cm) with a 1 : 1 : 1 mixture of fermented wheat bran, rice bran and soya bean dregs. The basal diet used in the current study contained 114 g kg−1 white fish meal and 200 g kg−1 SBM. For both species, three approximately isonitrogenous (crude protein: 380 g kg−1) and isocaloric (gross energy: 16 kJ g−1) diets were made with either 280 g kg−1 white fish meal (Control), 390 g kg−1 maggot meal (MGM) or 450 g kg−1 soybean meal (SBM). In the control diet, 280 g kg−1 wheat meal was used as the carbohydrate source, while 240 g kg−1 and 110 g kg−1 wheat meal were used for MGM and SBM groups, respectively, to obtain approximately isocaloric diets for both species. Soybean oil was not included in the MGM diet and 70 g kg−1 soybean oil were added, respectively, to the control and SBM diets as the soybean oil to maintain the amount of total crude lipids at approximately 120 g kg−1 diet for both species. Nutrient composition of three protein sources is shown in Table [NaN] , and the diet formulations and chemical compositions are shown in Table [NaN] . The experimental diets were made into pellets (2–3 mm diameter) using a laboratory pellet presser, oven‐dried at 60 °C and stored at −4 °C prior to use.

Nutrient composition (g kg −1 in dry matter) of three protein sources

1 White fish meal: American Seafood Company, Seattle, Washington, USA.

• 2 Maggot meal: Anhui Jingde Green spring Ecological Breeding Farm, Xuancheng, China.
• 3 Soybean meal: Anhui Huayi Agr‐livestock Technology Co., Ltd., Hefei, China.

Formulation and chemical composition of the experimental diets for gibel carp and darkbarbel catfish (g kg −1 in dry matter)

• 4 *Vitamin premix (mg kg−1 diet, NRC, 1993): Thiamin, 20; riboflavin, 20; pyridoxine, 20; cyanocobalamine, 2; folic acid, 5; calcium patotheniate, 50; inositol, 100; niacin, 100; biotin, 5; starch, 3226; vitamin A, 110; vitamin D3, 20; vitamin E, 100; vitamin K3, 10.
• 5 †Mineral premix (mg kg−1 diet, H440): NaCl, 500; MgSO4.7H2O, 7500; NaH2PO4.2H2O, 12,500; KH2PO4, 16,000; Ca(H2PO4).2H2O,10,000; FeSO4, 1250; C6H10CaO6.5H2O, 1750; ZnSO4.7H2O,176.5; MnSO4.4H2O, 81; CuSO4.5H2O, 15.5; CoSO4.6H2O, 0.5; KI, 1.5; starch, 225.

The experiment was conducted in an indoor recirculation system containing 18 fibreglass tanks (diameter: 60 cm, water volume: 190 L). At the beginning of the trial, all fish were fasted for 24 h. Healthy gibel carp (weighing 12.21 g) and darkbarbel catfish (weighing 10.95 g) of equal size were randomly selected, weighed and stocked in each tank (25 fish per tank). Three tanks were randomly assigned to each diet and each species. For both species, three groups of eight fish each were sampled for determination of the initial body composition. The feeding trial lasted 6 weeks.

As previously mentioned, fish were hand‐fed to apparent satiation twice daily (09:00 and 15:00 h). The daily feed supplied was recorded, and uneaten feed was siphoned 1 h after feeding, dried and weighed. The leaching rate of uneaten feed was estimated by placing weighed feed in tanks without fish for 1 h and were then collected, dried and reweighed. The average leaching rate was used to calibrate the amount of uneaten feed.

During the experiment, the photoperiod was 12 h light/12 h dark, with the light period from 08:00 to 20:00 h. The light intensity at the water surface was approximately 250 lx. Oxygen was supplied by aeration, with the minimum level at 7.0 mg L−1. The water temperature was recorded daily and kept constant at 26 ± 1.0 °C; pH was approximately 7.0. Ammonia‐N was measured weekly and kept at <0.1 mg L−1.

At the end of the experiment, the test fish were pooled and weighed individually following 24 h of food deprivation and then the fish were immediately anaesthetized with MS‐222 (200 mg L−1, Sigma‐Aldrich, St. Louis, MO, USA). Six gibel carp or darkbarbel catfish from each tank were randomly selected for analysis of the whole body composition. For each species, another eight fish per tank were randomly selected, and fork length (FL) was measured first (from the tip of the snout to the fork of the tail) for each fish. The fish were then killed by a blow to the head, and immediately dissected viscera were removed quickly and weighed to determine the viscerosomatic index (VSI). The liver was removed quickly, rinsed in 0.65% ice‐cold physiological saline and weighed after being blotted on a filter paper to determine the hepatosomatic index (HSI). The foregut, midgut and hindgut (approximately 3 in each gut section for darkbarbel catfish and 5 cm in each gut section for gibel carp) of each fish were removed quickly, rinsed in 0.65% ice‐cold physiological saline and mixed after being blotted on a filter paper, and these samples were then immediately frozen in liquid nitrogen and stored at −80 °C until analyses. Condition factor (CF), HSI, VSI and inter‐individual variation coefficients of final body weight (CVf) within tanks were calculated.

The initial and final fish samples were pooled, autoclaved at 120 °C for 20 min, homogenized and oven‐dried at 60 °C prior to whole body composition analysis. Dry matter, crude protein, crude lipid, ash and gross energy contents of fish samples and diets were determined as described by Dong et al. ([10] ).

Liver and intestine samples were homogenized in ice‐cold, 0.65% physiological saline using a tissue homogenizer. The homogenates were centrifuged (15 000 ×g) for 15 min at −4 °C. The supernatant was used to determine SOD (EC 1.15.1.1), CAT (EC 1.11.1.6), GPx (EC 1.11.1.9), GR (EC 1.6.4.2), γ‐glutamyl transpeptidase (γ‐GT, EC 2.3.2.2) activities, GSH concentrations, total antioxidative capacity (T‐AOC) and soluble protein contents. All these indices were determined using commercial kits, and the commercial kits were supplied by Nanjing Jiancheng Bioengineering Institute, Nanjing, China. SOD activity was assayed according to a SOD kit protocol (No. A001) based on the method described by Bayer & Fridovich ([2] ). CAT activity was assayed according to a CAT kit protocol (No. A007) based on the method described by Claiborne ([7] ). GPx and GR activities were assayed according to GPx or GR kit protocol (No. A005 or No. A062) based on the method described by Wheeler et al. ([45] ). γ‐GT activity was assayed according to a γ‐GT kit protocol (No. C017) based on the method described by Meister et al. ([27] ). T‐AOC was assayed according to T‐AOC kit protocol (No. A015) based on the method described by Miller et al. ([28] ). GSH concentration was assayed according to GSH kit protocol (No. A006) based on the method described by Griffith ([16] ). SOD, CAT, GPx, γ‐GT activities and T‐AOC were expressed as U (units) per microgram of soluble protein, and GR activity was expressed as U per gram of soluble protein, and GSH concentration was expressed as microgram per gram of soluble protein. Soluble protein content of tissue homogenates was determined according to the method of Bradford ([3] ) using bovine serum albumin as the standard protein. At least duplicate analyses were conducted for each sample.

All data were presented as mean ± SE. The normality and homogeneity of variances among groups were tested and all data were subjected to one‐way anova. If significances (P < 0.05) were identified, Duncan's multiple range tests were used to determine the differences between experimental groups. A principal component analysis was performed to compare the correlations among antioxidant enzymes, T‐AOC and GSH concentration within tissues in each species, and to analyse the correlations of the two fishes in response to dietary supplementation with MGM and SBM proteins. spss 18.0 (SPSS PASW Statistics, SPSS Inc., Chicago, IL, USA) was used for the statistical analysis.

Effects of dietary supplementation with MGM and SBM proteins on biometric indices, growth and feed utilization of gibel carp and darkbarbel catfish are shown in Table [NaN] . For gibel carp, CF and CVf did not differ between the three groups of fish (P > 0.05). The feeding rate (FR) and feeding rate of protein (FRp) were lower in the MGM‐ and SBM‐supplemented fish when compared with the control fish (P < 0.05). Contrarily, feed conversion efficiency (FCE) and protein efficiency ratio (PER) of the MGM‐ and SBM‐supplemented fish were significantly higher than those of the control fish (P < 0.05).

Effects of dietary supplementation with MGM and SBM proteins on biometric indices, growth and feed utilization of gibel carp and darkbarbel catfish (mean ± SE, n  = 3)

• 6 GC, gibel carp; DC, darkbarbel catfish; IBW, Initial body weight; FBW, Final body weight.
• 7 Means with the different superscripts within the same column are significantly different at P < 0.05.
• 8 Condition factor (CF, g cm−3) = 100 × body weight (g)/FL3 (fork length, cm).
• 9 Hepatosomatic index (HSI, %) = 100 × hepatopancreas weight (g)/somatic weight (g).
• 10 Viscerosomatic index (VSI, %) = 100 × viscera weight (g)/body weight (g).
• 11 CVf: inter–individual variation coefficients of final body weight within tanks.
• 12 Feeding rate (FR, % bw/day) = 100 × feed intake (dry matter, g)/days/[(initial body weight (wet weight, g) + final body weight (wet weight, g))/2].
• 13 Specific growth rate (SGR, %/day) = 100 × [ln final body weight (wet weight, g) – ln initial body weight (wet weight, g)]/days.
• 14 Feed conversion efficiency (FCE,%) = 100 × [final body weight (wet weight, g) – initial body weight (wet weight, g)]/feed intake (dry matter, g).
• 15 Feeding rate of protein (FRp, % bw/day) = 100 × protein intake (dry matter, g)/days/[(initial body weight (wet weight, g) + final body weight (wet weight, g))/2].
• 16 Protein retention efficiency (PRE,%) = 100 × protein gain in fish/protein intake (dry matter, g).
• 17 Protein efficiency ratio (PER) = [final body weight (wet weight, g) – initial body weight (wet weight, g)]/protein intake (dry matter, g).

For darkbarbel catfish, significant decreases in CF, HSI and VSI were observed in the MGM‐ and SBM‐supplemented groups compared with the control (P < 0.05). Both MGM‐ and SBM‐supplemented fish had significantly higher CVf than that of the control fish (P < 0.05). The SGR, FCE and PER of the MGM‐ and SBM‐supplemented fish were significantly lower than those of the control fish (P < 0.05). The SBM‐supplemented fish had significantly higher FR and FRp than those of the control fish (P < 0.05).

Table [NaN] shows the effect of dietary supplementation with MGM and SBM proteins on whole body compostion of gibel carp and darkbarbel catfish. For gibel carp, there were no significant differences in crude protein and ash contents between the three groups (P > 0.05), but the MGM‐supplemented fish showed significantly higher content of crude lipid than that of the control fish (P < 0.05).

Effect of dietary supplementation with MGM and SBM proteins on whole body composition of gibel carp and darkbarbel catfish (on wet weight basis) (mean ± SE, n  = 3)

• 18 GC, gibel carp; DC, darkbarbel catfish.
• 19 Means with the different superscripts within the same column are significantly different at P < 0.05.

For darkbarbel catfish, the crude protein of the SBM‐supplemented fish was significantly higher than that of the control fish (P > 0.05). Both MGM‐ and SBM‐supplemented fish showed significantly lower crude lipid than that of the control fish (P < 0.05). There were no significant differences in ash and gross energy contents between the three groups (P > 0.05).

Table [NaN] shows the antioxidant responses of gibel carp and darkbarbel catfish in response to dietary supplementation with MGM and SBM proteins. For gibel carp, SOD, GPx, GR and γ‐GT activities, and T‐AOC and GSH concentration in liver of the MGM‐supplemented fish were significantly higher than those in fish fed with the diets control or SBM (P < 0.05), and the increased percentages of above indices in liver were 38.5%, 38.5%, 114.2%, 133.3%, 43.1% and 75.3%, respectively. When compared with the control groups, the increased percentages of GPx and γ‐GT activities in intestines of the MGM‐ and SBM‐supplemented fish were 18.8% and 31.0%, and 24.7% and 40.1%, respectively. The GR activity and T‐AOC in intestine of the MGM‐supplemented fish were significantly higher (61.31% and 45.6%, respectively) than those in the control groups (P < 0.05).

Effect of dietary supplementation with MGM and SBM proteins on antioxidant systems of gibel carp and darkbarbel catfish in liver and intestine (mean ± SE, n  = 3)

• 20 GC, gibel carp; DC, darkbarbel catfish.
• 21 Means with the different superscripts within the same column are significantly different at P < 0.05.

For darkbarbel catfish, the SOD, GPx, GR activities, T‐AOC and GSH concentrations in liver of the MGM‐ and SBM‐supplemented fish were significantly lower than those in the control groups (P < 0.05), and the decreased percentages of above indices in liver were 27.4% and 34.3%, 56.0% and 61.0%, 45.2% and 22.5%, 39.1% and 62.6%, and 46.4% and 58.2%, respectively. The SOD, GR and γ‐GT activities in intestine of the MGM‐ and SBM‐supplemented fish were significantly lower than those in the control groups (P < 0.05), while the reductions of above indices were 43.4 and 34.8%, 53.1 and 25.92, and 69.15 and 38.1%, respectively.

Principal components analysis analysis of the antioxidant enzymes, T‐AOC and GSH concentration in livers of gibel carp showed a clear separation between CAT activity and the other six indices (GR, GPx, γ‐GT, SOD, T‐AOC and GSH), but a higher similarity was observed among the other six indices (Fig. [NaN] a). Eigenvalues of the first principal component (PC1) and second principal component (PC2) axis are 5.3 and 1.1, and PC1 and PC2 explained 92.3% of the cumulative variance of the data, with 76.1% for the PC1 and 16.2% for the PC2 (Table [NaN] ). When compared with the gibel carp, PCA ordination of darkbarbel catfish in the above indices were somewhat more widely scattered (Fig. [NaN] b).

Eigenvalues and percentage variance explained of the PCA analysis of antioxidant indices, and dietary treatments for gibel carp and darkbarbel catfish

22 GC, gibel carp; DC, darkbarbel catfish.

Principal components analysis ordination of gibel carp and darkbarbel catfish, which was based on the antioxidant enzymes, T‐AOC and GSH concentration in intestines is shown in Fig. [NaN] . For gibel carp (Fig. [NaN] a), PCA ordination of the above indices is highly scattered, and all indices showed a clear separation, except for higher similarities between GR activity and T‐AOC, and GPx and γ‐GT activities. PCA ordination of darkbarbel catfish indicated that there is a higher correlation among CAT, GPx, SOD, GR and γ‐GT activities, while the PC2 space is dominated by T‐AOC and GSH concentration (Fig. [NaN] b).

A PCA was carried out to analyse the correlations of the two fishes in response to different dietary protein sources (Fig. [NaN] ). PCA ordination of gibel carp showed a higher similarity in response to dietary supplementation with MGM and SBM proteins between liver and intestine. However, a clear separation between liver and intestine was observed in darkbarbel catfish. The cumulative variance explained by the PC1 (eigenvalue: 8.60) and PC2 (eigenvalue: 2.76) is 93.46%, with 71.69% for the PC1 and 23.03% for the PC2 (Table [NaN] ).

In the current study, the average specific growth rate in wet weight (SGRw) of gibel carp (weighing 12.23 g) fed with the control diet (383.2 g kg−1 crude protein) was approximately 1.32% per day, which is lower than those (1.34–1.89% per day) reported for gibel carp (weighing 2.69–3.21 g) (Pan et al. [33] ; Chen et al. [5] ), because the smaller fish can obtain higher growth rate than the bigger fish. On the other hand, the average SGRw of darkbarbel catfish (weighing 11.40 g) fed with the control diet (383.2 g kg−1 crude protein) was approximately 1.80% per day, which is comparable to that (1.71–1.98% per day) reported for similar‐sized darkbarbel catfish (weighing 14.3 g) (Xue et al. [46] ). It indicates that the control diet prepared for the present study was nutritionally adequate to sustain good growth of juvenile gibel carp and darkbarbel catfish.

Friedman ([14] ) indicated that the nutritional value or quality of structurally different proteins varies and is governed by amino acid composition, ratios of essential amino acids and susceptibility to hydrolysis during digestion. In the present experiment, the three protein sources could vary not only in amino acid composition but also to some extent in ratios of essential amino acids. In addition, soybean oil was not included in the MGM diet. However, amino acid and fatty acid profiles were not determined for the experimental diets in the present study. Therefore, the poor growth of both species fed the feeds containing SBM protein could be due to the presence of anti‐nutritional factors (Refstie et al. [37] ; Peres et al. [35] ), lower protein digestibility (Refstie et al. [37] ), essential amino acid deficiency (Chong et al. [6] ; Tantikitti et al. [40] ) and different fatty acid profiles in the SBM‐supplemented groups, and further studies are required to validate the possible reasons. In darkbarbel catfish, the lower growth of fish fed the diet containing MGM protein was observed; this result is contrary to those found in Nile tilapia by Ogunji et al. ([30] , [31] ), who reported that supplementing 250 g kg−1 to 550 g kg−1 MGM protein enhanced the growth rate of the fish, but supplementation with 150 g kg−1 and 680 g kg−1 MGM protein had no effect on growth rate of the fish. This different response between darkbarbel catfish and Nile tilapia could be attributed to differences in the amino acid and fatty acid profiles in the tested diets or the capability of MGM protein in different fishes.

From the comparison of FR, FCE and PER between the two fishes, the present results suggest that the influence of dietary supplementation with MGM and SBM proteins on the growth dynamics of gibel carp and darkbarbel catfish can differ substantially. Gibel carp consumed less feed/protein and utilized it more efficiently for growth as dietary supplementation with MGM or SBM protein. However, the compensatory increases in FCE and PER were not large enough to allow the MGM‐ and SBM‐supplemented gibel carp to reach the same growth rate of those fish in the control groups. The improved FCE and PER with decreasing dietary protein levels is in agreement with other studies in red tilapia, rockfish and yellow catfish (Lee et al. [21] ; Dong et al. [10] ). In darkbarbel catfish, the reduced FCE and PER due to dietary supplementation with MGM or SBM protein might, however, be explained by compensatory increase of FR/PER to obtain more amount of feed/protein for optimal growth of the fish. Therefore, darkbarbel catfish has much lower ability in using dietary MGM and SBM protein than those of gibel carp. Similar results were also found in red drum (McGoogan & Gatlin [25] ) and cuneate drum (Wang et al. [42] ) in using SBM as fish meal substitute.

Fish have developed antioxidant enzymes (such as CAT, SOD, GPx, GR) and use antioxidant substances (such as vitamin C, vitamin E, GSH and carotenoids) to alleviate adverse effects of oxidative stress (Rudneva [38] ; reviewed in Martínez‐Álvarez et al. [24] ). The present results show significant increases in the activities of SOD, GPx, GR and γ‐GT, and T‐AOC and GSH concentration in liver, and activities of GPx, GR, γ‐GT and T‐AOC in intestine of the MGM‐supplemented gibel carp. The endogenous scavenger, SOD, which catalyses the dismutation of the highly reactive superoxide anion to H2O2 (Dorval et al. [11] ), was significantly increased in liver of gibel carp. The observed increase in liver SOD activity may be a consequence of increased de novo synthesis of enzyme proteins or oxidative inactivation of enzyme protein (Dimitrova et al. [9] ; Kaushik & Kaur [20] ; Moreno et al. [29] ). GPx catalyses the reduction of H2O2 at the expense of GSH (Harris [17] ); thus, the higher GPx activity in liver may be attributable to an increase in GSH concentration, which is in good correspondence with the result found in higher GSH concentration in liver of gibel carp in the present study. GR, which regenerates GSH from its oxidized form (GSSG), glutathione disulphide, thus plays a crucial role in GSH turnover and in cellular antioxidant protection (Cossu et al. [8] ). Therefore, the observed increases in GR activities in both liver and intestine of the MGM‐supplemented gibel carp indicate an increase in the conversion of oxidized glutathione (GSSG) back to its reduced form (GSH), with high levels in GSH. γ‐GT is a central enzyme of the γ‐glutamyl cycle, the pathway for the synthesis and degradation of GSH (Meister et al. [27] ). The enhanced liver γ‐GT activity in MGM‐supplemented gibel carp seems to be due to an enhanced ability to synthesize GSH. On the other hand, this enhanced γ‐GT activity acts to degrade and recycle both GSSG and the products of glutathione S‐transferase activity (Meister et al. [27] ). Importantly, dietary supplementation with MGM protein increased T‐AOC in both liver and intestine of gibel carp, indicating the enhancement of whole‐body antioxidative function. On the other hand, the lipid peroxidation of the diets were not analysed in the present study, thus, fish fed with the MGM diet may partially increase the above parameters of oxidative stress in the liver and intestine compared to control. Therefore, the results indicate that supplementation with 390 g kg−1 MGM protein in basal diet cause an enhancement of the antioxidant capacity to some extent in gibel carp.

On the contrary, dietary supplementation with MGM and SBM proteins decreased the activities of SOD, GPx and GR, and T‐AOC and GSH concentrations in liver, and activities of SOD, GR and γ‐GT in intestine of darkbarbel catfish, while the decreased CAT activity was only observed in the MGM‐supplemented fish. However, when Nile tilapia were fed diets supplemented with 150‐680 g kg−1 MGM proteins, hepatic CAT activity did not differ significantly throughout the diets except for 350 g kg−1 supplementation level, which showed a significant lower value of CAT activity compared with the other treatments (Ogunji et al. [30] ). This discrepancy between the present study and the work of Ogunji et al. ([30] ) may be explained by the difference in experimental design, the duration of MGM supplementation, and the experimental fish used. According to Kaushik & Kaur ([20] ), the relative contributions of CAT and GPx are the decomposition of endogenously produced H2O2. Therefore, the decreased activities of CAT and GPx in both liver and intestine of the MGM‐supplemented darkbarbel catfish indicate the highly reduced capacity to scavenge H2O2 produced in these tissues, with an increase in ROS and oxidative stress in response to dietary supplementation with MGM protein. On the other hand, the lower GPx activity could also be due to a decline in GSH concentration, as previously demonstrated (Harris [17] ; Cossu et al. [8] ). In the current study, the observed decreases in GR activities in both liver and intestine in the MGM‐ or SBM‐supplemented darkbarbel catfish indicates a decrease in the conversion of GSSG back to its reduced form (GSH), with low levels in GSH. Moreover, the depletion in the activities of GR could be due to the decreases in NADPH levels, a secondary manifestation of cellular free radical stress (Chance & Boveris [4] ). From the present results, it appears conclusive that dietary supplementation with MGM and SBM proteins attenuates the GSH metabolism and therefore the antioxidant defences in darkbarbel catfish. In liver, this decreased GSH level could be due to the different enzymatic reactions involved in the GSH balance: GSH synthesis, GSH oxidation, GSH conjugation, cleavage of GSH‐conjugates and GSSG reduction by GR in darkbarbel catfish (Meister & Anderson [26] ). The obtained data seem to indicate that supplementation of 390 g kg−1 MGM and 450 g kg−1 SBM proteins in basal diets resulted in a significant attenuation of the antioxidant capacity in darkbarbel catfish.

In the present study, for both species, liver and intestine show different patterns in correlations between antioxidant enzymes (SOD, CAT, GPx, GR, γ‐GT)/antioxidant substances (GSH) and T‐AOC, especially for darkbarbel catfish, which may reflect the different defensive strategies in response to dietary proteins by different tissues within the same species (reviewed in Martínez‐Álvarez et al. [24] ). Martínez‐Álvarez et al. ([24] ) concluded that antioxidant defences in fish are dependent on feeding behaviour and nutritional factors. In the present study, the different patterns of gibel carp and darkbarbel catfish in response to supplementation with MGM or SBM protein in basal diets were also found. This discrepancy between the two fishes may be explained by the differences in their natural diet and the nutritional value of different dietary protein sources. Dietary supplementation with MGM and SBM proteins in basal diets are able to promote a stronger differentiation of antioxidative capacity for darkbarbel catfish.

In conclusion, this study demonstrates that dietary supplementation with 390 g kg−1 MGM and 450 g kg−1 SBM proteins in basal diets reduced the growth rate of the fish only in the SBM‐supplemented treatment, whereas supplementation with 390 g kg−1 MGM protein in basal diet cause an enhancement of the antioxidant capacity in gibel carp. However, supplementating 390 g kg−1 MGM and 450 g kg−1 SBM to basal diets reduced the growth rates and resulted in a significant attenuation of the antioxidant capacity in both MGM‐ and SBM‐supplemented darkbarbel catfish. Our results suggest that dietary proteins can affect the antioxidant capacity in the two fishes.

The authors wish to thank Rentong, Niu for his assistance in the study. This research was supported by International Science & Technology Cooperation Programme of China (grant no. 2011DFG33280), Development Fund for Doctor in Anhui Agricultural University (grant no. YJ2008‐22) and Undergraduates Scientific and Technological Innovation Project in Anhui Agricultural University (2010190). Thanks are also given to the anonymous reviewers for their helpful suggestions.

Graph: PCA analysis of the antioxidant enzymes, T‐ AOC and GSH concentrations in livers of gibel carp (a) and darkbarbel catfish (b).

Graph: PCA analysis of the antioxidant enzymes, T‐ AOC and GSH concentrations in intestines of gibel carp (a) and darkbarbel catfish (b).

Graph: PCA ordination of gibel carp and darkbarbel catfish to analyse the correlations in response to dietary supplementation with MGM and SBM proteins. L ‐ GC , livers of gibel carp; L ‐ DC , livers of darkbarbel catfish; GC , gibel carp; I ‐Control, intestines of the control fish; I ‐ SBM , intestines of the SBM ‐supplemented fish; I ‐ MGM , intestines of the MGM ‐supplemented fish; L ‐ C ontrol, livers of the control fish; L ‐ SBM , livers of the SBM ‐supplemented fish; L ‐ MGM , livers of the MGM ‐supplemented fish.