Black Sea Bream Acanthopagrus Schlegelii Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

An 8-week feeding trial was conducted to determine the effects of dietary carbohydrate raw corn starch level on growth performance and feed utilization of juvenile black sea bream, Acanthopagrus schlegelii. Six isonitrogenous and isoenergetic diets (400 g kg-1 crude protein and 20 KJ g−1 diet) using raw corn starch as the carbohydrate source, were formulated to contain six carbohydrate levels from 150 g kg-1 to 300 g kg-1 in 30 g kg-1 increments. Each diet was fed to triplicate groups of 20 fish (mean weight, 5.79 ± 0.07 g) twice a day for 8 weeks. Growth performance and feed utilization were significantly affected by dietary raw corn starch levels. Survival, moisture of whole body and crude lipid of muscle were not significantly different among diets. There were significant differences in FER, PER, PPV, protein and lipid contents in whole body, as well as triglyceride and cholesterol in serum and liver glycogen. Dietary starch level had no effect on starch and lipid digestibility, whereas significantly affected that of protein. The results indicated that the optimal dietary raw corn starch level for maximum WG of juvenile black sea bream was 184 g kg-1 of dry diet. However, black sea bream juvenile can use raw corn starch very well from 18% to 30% based on WG, FER, PER and starch digestibility. In conclusion, juvenile black sea bream can use high level of raw corn starch (300 g kg-1) with decreased dietary lipid without negative effect on growth but bring profit-benefit in aqua feed.


Black sea bream (Acanthopagrus schlegelii), a marine candidate species for intensive aquaculture, due to its high economic value, good meat quality, ability to tolerate a wide range of environmental conditions (such as salinity and temperature), high stocking densities and strong resistance to diseases (Shao et al., 2008; Zhou et al., 2011a), has been farmed on a large scale in China, Japan, Korea and other countries in Southeast Asia recent years (Gonzalez et al., 2008). Since limited nutritional requirement studies, which mostly focus on protein nutrition, have been conducted (Zhang et al., 2010; Zhou et al., 2011a; Zhou et al., 2011b; Zhou et al., 2011c), no commercial feed has been specifically formulated for this species yet, and as a result, trash fish is mainly used to feed farmed black sea bream in China, which brings many adverse effects on the aqua practice, such as limited supply, variable nutritional quality, poor feed conversion rate (FCR) and easily results in water pollution (Ma et al., 2008; Shao et al., 2008; Ngandzali et al., 2011). Hence, the development of commercial feeds formulated for black sea bream are urgent, which need more nutritional requirements investigated.

As is known, inclusion of non-protein energy sources such as carbohydrate or lipid in commercial feed can minimize economic cost of aquaculture (Wilson & Halver, 1986). It has been proved that carbohydrate, which is relatively inexpensive compared with lipid, is a kind of available energy source for fish diet (Erfanullah & Jafri, 1998; Peres & Oliva-Teles, 2002). Utilization of carbohydrate by fish varies greatly among different species (Hilton & Atkinson, 1982; Wilson, 1994; Stone et al., 2003c; Wang et al., 2005; Tan et al., 2009; Ren et al., 2011), which also showed that utilization of carbohydrate by most herbivorous and omnivorous fishes was generally better than that of carnivorous fishes. However, some carnivorous fishes showed improved growth when fed diet with an appropriate gelatinized starch level compared with a diet without starch as long as protein inclusion level was kept within a minimum adequate level (Hemre et al., 2002b). Certainly, excessive dietary starch in carnivorous fish diet also led to inhibited growth performance and disorder of some physiological function (Hutchins et al., 1998; Hemre et al., 2002b; Tan et al., 2009). Besides, carbohydrate utilization in diet by the same species of fish appears to be related to carbohydrate sources and inclusion levels (Bergot, 1979; Lee et al., 2003; Simon, 2009; Ren et al., 2011), environmental factors, including temperature (Medale et al., 1999; Moreira et al., 2008), light regime (Hemre et al., 2002a) and seasons (Thibault et al., 1997). Moreover, utilization of starch was also affected by its gelatinization degree (Bergot & Breque, 1983; Amirkolaie et al., 2006; Kumar et al., 2011). In some species, digestibility of gelatinized starch was higher than that of raw starch (Kaushik & de Oliva Teles, 1985; Peres & Oliva-Teles, 2002) due to the procedure of hydro-thermal treatment (gelatinization), besides it was also improved by supplementation of exogenous digestive enzyme (Stone et al., 2003b),

Therefore, the present experiment was conducted to elucidate the adequate level of cabohydrate to be incorporated into the diets of black sea bream juvenile for its maximum growth and nutrient utilization, and the results may be helpful in formulating cost-effective commercial feed for this fish.

Materials and methods

Experimental diets

The FM used in this study was purchased from Alaska Ocean Seafood Limited Partnership (Washington DC, USA), and the other feed ingredients were obtained locally from Zhejiang University. Prior to use, all feed ingredients were analysed for their proximate composition and the data obtained was used as the basis to formulate the experimental diets. Six isonitrogenous and isoenergetic diets were formulated to contain six levels (150 g kg-1, 180 g kg-1, 210 g kg-1, 240 g kg-1, 270 g kg-1 and 300 g kg-1, respectively) of raw corn starch. Six diets were maintained isoenergetic by decreasing the levels of corn oil as the raw corn starch level was increased. Experimental diets contained about 400 g kg-1 crude protein, as suggested in our preliminary experiment (Zhang et al., 2010) to meet anabolic requirements to gain a protein-sparing effect using carbohydrate (Stone et al., 2003a). Fish meal and gelatin supplied 400 g kg-1 dietary protein, while fish oil and corn oil provided different levels of dietary lipid in each diet to gain equal energy level in each diet. Yttrium oxide (Y2O3, ¼ž99.9% in purity, China) at an inclusion level of 1 g kg-1 was used as the external indicator for digestibility determination of nutrients in diets (Alan Ward et al., 2005). The formulation and proximate analysis of diets are presented in Table 1.

Drying ingredients were ground into particles with a diameter within 180-µm, weighed and mixed manually for 5 minutes and then transferred to a food mixer for another 15-minute mixing. During mixing, fish oil and corn oil were thoroughly mixed firstly and added slowly along with distilled water to form fine dough. Then the dough was extruded through a feed mill (Model HKJ-218, China) with a 2.5 mm diameter die, and pelleted to appropriate size after drying for about 72 hours in air-conditioning room with 23 °C. After air-conditioner drying and sieving, feeds were collected in air-tight polyethylene bag and stored at -20 °C until use.

Feeding trial

Black sea bream juvenile were obtained from the Research Institute of Zhejiang Marine Fisheries in Zhoushan (China) and were acclimated to laboratory conditions for two weeks. During acclimatization, fish were fed to satiation twice a day (08:00 and 16:00 h) with a commercial feed (crude protein: 425.1 g kg-1, crude lipid: 76.9 g kg-1, ash: 135.6 g kg-1, total phosphorus: 12.1 g kg-1; provided by Ningbo Tech-bank Co., Ltd, Ningbo, China). After the 2-week acclimation, 360 fish (initial mean weight: 5.79 ± 0.07 g, mean ± SD) in good health condition were randomly allocated into 18 350-L fibreglass tanks (with 300-L water volume) at a stocking density of 20 fish per tank. Each tank was supplied with flowing seawater filtered through sand at flow rate of 3 L min-1. Water quality was monitored daily and was within acceptable limits throughout the experiment. Temperature, ammonia-N concentration and salinity of the seawater in tanks were 27 ± 1 °C, 29 g L-1 and 0.02-0.04 mg L-1, respectively. Dissolved oxygen concentrations were controlled above 5.0 mg L-1 at any point during the experiment by using air stones with continuous aeration. Fish were maintained under a natural photoperiod (12-h dark / 12-h light). Dietary treatments were randomly assigned to triplicate tanks, and fish were fed twice daily (08:00 and 16:00) as during acclimatization to apparent satiation and feed consumption of each tank was recorded daily. Tanks were thoroughly cleaned as needed and mortality was checked daily. The growth trial lasted for 8 weeks.

Faeces were collected daily during the last 3 weeks of the trial, using a faecal collection device similar as used in rainbow trout (Cho & Kaushik, 1990). Enough care was taken as possible during the feeding to make feed losses avoided almost completely, which promised the quality of faeces collected. The drain pipe and faecal collection columns were thoroughly cleaned with a brush to remove residual faeces from the system 2 h after the final feeding of the day to avoid fragments of feeds produced by fish during feeding. Faeces were then allowed to settle overnight, and faecal samples were collected at 06:00 each morning before the next feeding. Faeces collected from the settling columns were immediately filtered with filter papers and stored at -20 °C for subsequent chemical analysis. Samples from the same tank were pooled over the sampling period to provide sufficient faecal matter for analysis.

Sampling collection and analytical methods

At the start of feeding trial, ten fish were sampled and stored at -20 °C for analysis of proximate carcass composition. At the termination of the 8-week feeding trial, all fish were starved for 24 h before sampling. Individual fish weight and body length were measured for calculation of final body weight and condition factor (CF). Three fish from each tank were anaesthetized (tricane methanesulphonate MS-222, 80 mg L-1) and then stored at -20 °C for subsequent whole body proximate analysis. Blood samples were drawn from the caudal vein of the remaining fish per tank with a 27-gauge needle and 1-mL syringe. Blood samples were immediately pooled and centrifuged at 2500 rpm for 15 min (4 °C) to obtain serum, and serum samples were then stored at -20 °C until use. Pooled samples of dorsal muscles were obtained from all the remaining fish in each tank and stored at -20 °C for subsequent proximate analysis. After dorsal muscle collection, liver and intraperitoneal fat were quickly sampled and weighed to calculate hepatosomatic index (HSI) and intraperitoneal fat ratio (IPR).

Survival, weight gain (WG), specific growth rate (SGR), CF, HSI, IPR, feed efficiency ratio (FER), protein efficiency ratio (PER) and protein productive value (PPV) were calculated after the experiment as follows:

Survival (%) = 100 Ã- final fish number/initial fish number

WG (%) = 100 Ã- (final body weight - initial body weight) / IBW

SGR (% day-1) = 100Ã- (Ln (final body weight) - Ln (initial body weight)) / day

CF (g cm-3) = 100 Ã- (body weight in g / (body length in cm) 3)

HSI = 100 Ã- liver weight in g / body weight in g

IPR = 100 Ã- intraperitoneal fat weight in g/body weight in g

FER = wet weight gain in g / dry diet fed in g

PER = weight gain in g / protein intake in dry basis in g

PPV = protein gain in g / protein fed in dry basis in g

Chemical analyses were conducted following standard laboratory procedures (AOAC, 1995). Moisture concentration was determined by drying minced samples for 6 h in a forced-air oven maintained at 105 °C. Ash content was analysed by incinerating samples at 550 °C for 12 h in a muffle furnace. Crude protein was estimated as Kjeldahl-nitrogen using factor 6.25. Lipid concentration of whole body was determined by soxhlet extraction with ether for 6 h. Starch content was determined as reported (Thivend et al., 1972).

Serum triglycerides, total protein and total cholesterol concentrations were assayed by enzymatic procedures using an automatic biochemical analyzer (Hitachi 7170, Japan) and attached kits (Daiichi Pure Chemicals Co., Japan). The concentration of glycogen in liver was measured within 3 days, using the diagnostic reagent kit purchased from the Nanjing Jiancheng Bioengineering Institute (Nanjing, China), according to the manufacturer's instructions. The content of yttrium in diet and fecal samples were determined by inductively coupled plasma emission spectrometer (model: IRIS Intrepid â…¡XSP, USA) after 0.5 g of sample was incinerated at 550 °C for 12 h in a muffle furnace and 5 drops of concentrated nitric acid and 10 mL HCl (1+1) was added. The solution was transferred to a 200 mL volumetric flask and diluted to volume with distilled water. Starch content was measured according to Thivend et al. (1972).

All data were presented as means ± SD subjected to one-way analysis of variance (ANOVA) to test the effects of experimental diets using the software of the SPSS for Windows (V. 16.0, SPSS Inc., Chicago, IL, USA). Turkey's multiple range test and the critical ranges were used to test the differences among individual means. The difference was regarded as significant when P ¼œ 0.05. Broken-line regression analysis was performed on WG to establish the adequate inclusion level of raw corn starch for black sea bream. The equation used in the model is Y = L + U (R - X), where Y is the parameter (WG) chosen to estimate the requirement, L is the ordinate and R is the abscissa of the breakpoint. R is taken as the estimated requirement and U is the slope of the line for X (Robbins et al., 1979).


Growth performance, biological parameters and feed utilization of juvenile black sea bream fed with difference dietary raw corn starch levels for 8 weeks were presented in Table 2. Survival rates were not affected by experimental treatments, and were all up to 100%. An increasing dietary corn starch level from 150 g kg-1 to 210 g kg-1 significantly increased the WG of black sea bream (P ¼œ 0.05), but a further elevation to 300 g kg-1 could not produce any additional promoting effect. FER, PER and PPV had a similar change pattern with WG within the dietary treatments, though there were different decline extent and decline point for fish feed excessive raw corn starch level diets. MFI showed the same change tendency as above indices without significance among all the treatments. When it comes to SGR, it was significant lower in fish fed diet 1 than other treatments (P ¼œ 0.05), and there was no significant difference while raw corn starch level range from 180 g kg-1 to 300 g kg-1. Increasing dietary raw corn starch level decreased CF significantly (P ¼œ 0.05), whereas an opposite result was observed in HSI, which was significantly increased by the inclusion level of the starch (P ¼œ 0.05). IPR of fish fed diet 1 was significantly higher than that of fish fed diet 6, and there was no significant difference among diets 2-5. Based on WG, the optimum inclusion levels of raw corn starch in black sea bream feed was estimated to be 184 g kg-1 diet, using break-point regression method analysis (Fig. 1).

Apparent digestibility coefficients (ADCs) of dry matter, crude protein, crude lipid and starch of experimental diets for black sea bream are shown in Table 3. The values of ADCs of dry matter in 150-210 g kg-1 raw corn starch-containing diets were lower than that in 240-300 g kg-1 raw corn starch-containing diets. ADC of protein was significantly higher in fish fed T3-T6 (P ¼œ 0.05), but no significantly differences in that of lipid and starch were found between different groups (P ¼ž 0.05).

Body and dorsal muscle values are shown in Table 4. No significant difference was observed among treatments in carcass moisture. Crude protein and crude fat of body composition were significantly affected by raw corn starch level (P ¼œ 0.05). The crude protein content of fish fed T3 and T4 were significantly higher than those of fish fed T1, T2 and T6 (P ¼œ 0.05), respectively. Crude fat decreased with increasing levels of raw corn starch, namely increased with decreasing levels of dietary lipid (P ¼œ 0.05). Significant difference of crude fat was observed between fish fed T1 and T6, and the former was higher than the latter. Dorsal muscle composition except crude lipid were significantly affected by raw corn starch level among treatments (P ¼œ 0.05). No significant difference was observed among treatments in crude lipid. While related to moisture, it was the highest in fish fed diet 5, intermediate in fish fed T3, T4 and T6, and lowest in fish fed T1-T2 (P ¼œ 0.05). The crude protein content of fish fed T3 was significantly higher than that of fish fed T1 (P ¼œ 0.05), while fish fed the other diets showed no significant difference with neither T1 nor T3.

In the present experiment, no significant differences were found in serum glucose and total protein contents (Table 5), whereas serum triglyceride, serum cholesterol and liver glycogen of fish differed significantly between groups (P ¼œ 0.05). Triglyceride concentration reached the significant greatest values in T4 compared with T5 and T6, while that of cholesterol was in T2. Liver glycogen content in T3 was significantly higher than in other groups.


In aquaculture practice, carbohydrate due to its inexpensive economic value, was used to the greatest extent in aqua feed to reduce the costs while fish growth and feed utilization were not depressed (Tan et al., 2009) and to achieve extrusion in feed industry. As carbohydrate was used as an energy source in aqua feed due to its protein-sparing effect, determination of its adequate inclusion level in the feed should base on protein and the energy content. As a result, the experimental diets were formulated isonitrogenous and isoenergetic to evaluate the appropriate inclusion level of raw corn starch with an invariable and minimum adequate protein and energy level in the feed of juvenile black sea bream.

In present growth trail, survival was not influenced by raw corn starch level (P ¼ž 0.05). High survival (high to 100%) of experimental fish in this growth trial ensures that analysis results obtained from growth and feeding performance are reliable. In terms of WG of experimental juvenile, it increased while raw corn starch inclusion level increased from 150 g kg-1 to 210 g kg-1 (P ¼œ 0.05), but showed decreased tendency without significant difference thereafter to a high level of 300 g kg-1. Generally, a level of 200 g kg-1 digestible carbohydrate appears to be optimal for marine or coldwater fish, whereas higher levels are used by fresh warm water fish (Wilson, 1994). However, this should be carefully confirmed for each species. Wang et al. (2005) reported the appropriate dietary carbohydrate level for juvenile tilapia, was 220 g kg-1 in the form of corn starch. Couto et al. (2008) consisted that gelatinized starch may be included up to 200 g kg-1 in diets for gilthead sea bream juveniles. Besides, growth and efficiency of feed utilization are depressed at higher dietary levels (300 g kg-1). Ren et al. (2011) reported that the appropriate dietary starch (gelatinized corn starch) supplementations of juvenile cobia based on SGR and FER were estimated to be 211 g kg-1 and 180 g kg-1 of diet respectively. However, the present study showed that dietary raw corn starch level of 210 g kg-1-300 g kg-1 did not reduce the growth and FER in juvenile black sea bream, which was in agreement with the results in European sea bass while included level of either raw or gelatinized starch up to 250 g kg-1 (Gouveia et al., 1995) and while included level of gelatinized starch was from 100 g kg-1 to 300 g kg-1, as well as in cobia which demonstrated to be able to utilize dietary carbohydrates up to at least 340 g kg-1 of dry diet with an optimal protein to energy ratio (JR Webb et al., 2010). As a carnivorous marine fish, it is marvelous for black sea bream to use such high level of raw corn starch level without compromising growth effect. Utilization of carbohydrate is species dependent, and is depends on their ability to oxidize the glucose from carbohydrate digestion, and to store the excess glucose into the forms of glycogen or fat (Guo et al., 2006). Growth performance in present study showed that black sea bream has great potential to utilize the glucose from carbohydrate digestion.

Starch is accumulated in granules in the endosperm, and the starch is deposited in layers with various contents of two kinds of macromolecules, namely amylose and amylopectin. Amylose has essentially a straight chain structure while amylopectin has a highly branched chain structure (Sá et al., 2008). In most common types of cereal endosperm starches (such as corn), it usually contains 18-33% amylose and 72-82% amylopectin, depending on the botanical origin (Buléon et al., 1998). High-amylose content is associated with reduced digestibility (Svihus et al., 2005). Gelatinization breaks down the starch granule exposing the bond amylase fraction and increases the surface area, which renders the starch more susceptible (French, 1973). However, Peres and Oliva-Teles (2002) found that performance of sea bass was better with a mixture of 125 g kg-1 raw starch plus 12.5 g kg-1 gelatinized starch than with either 250 g kg-1 gelatinized or 250 g kg-1 raw starch alone. Besides, results in yellowfin seabream demonstrated that diets with raw corn starch performed better growth compared with pre-gelatinization starch at an inclusion level of 200 g kg-1, which may be due to an increase in gastric evacuation time when fed the pre-gelatinized or gelatinized starch, or increased available energy due to improved digestibility after gelatinization (Wu et al., 2007). In the present study, black sea bream showed great ability to utilize high level of starch (300 g kg-1), mainly due to raw starch maintains its long gastric evacuation time as we found during other trial (about 16 h, unpublished), and make absorption of glucose digested from starch lasted much longer, which results in enhanced growth performance. And as cellulose content in feed may accelerate the feed get through the digestive tract, and thus reduced the absorption of glucose, which may be contributed to the decrease of growth in T1-T2 partially.

The natural diet of the carnivorous black sea bream consists of copepods, cladocerans and some plant materials such as algae (Nip et al., 2003), and according to that, carbohydrates will not be the main energy source for this species in nature. However, the present study suggested that black sea bream are able to adapt to a high carbohydrate diet. It appears that the sparid fishes such as black sea bream in present study, silver sea bream (Leung & Woo, 2012) and gilthead sea bream (Fernández et al., 2007) are among marine fish which are being able to utilize high level of dietary carbohydrate (210 g kg-1 raw corn starch, 200 g kg-1 dextrin and 200 g kg-1 gelatinized cornstarch, respectively) as energy source.

Dietary starch level was reported to negatively affect carbohydrate digestibility and may interact with the digestibility of other dietary constituents (Dias et al., 1998; Stone et al., 2003a; Enes et al., 2006). The present experiment showed that raw corn starch was very well digested by black sea bream, and was not affected by the inclusion levels of raw corn starch, similar as the results in gilthead sea bream (Couto et al., 2012). ADC of protein were significantly lower in fish fed T1 and T2 than in other groups, and showed a decrease tendency without significance as the corn starch inclusion level increased, which was also detected in rainbow trout (Gumus & Ikiz, 2009). However, a decrease in protein digestibility with the increase of dietary carbohydrate level was recently reported in gilthead sea bream (Fountoulaki et al., 2005), European sea bass (Enes et al., 2006) and as well as in white sea bream (Sá et al., 2006). Spannhof and Plantikow (1983) imputed this phenomenon in rainbow trout to the adsorption of proteolytic enzymes to the starch molecules, which thus rendering them unavailable for enzymatic action and therefore decreasing nutrient absorption. Besides, as monosaccharide may inhibit amino acid transport in the intestine (Ferraris & Ahearn, 1984; Vinardell, 1990), increased glucose released in high-starch diets could be another explanation for this phenomenon, and the slowly released rate of glucose from raw starch granule make it not significant enough to gain significant decrease of protein digestibility. At last, as for the dry matter digestibility in the present study, the supplementation of dietary raw corn starch was adjusted by cellulose and corn oil, contents of cellulose were higher in T1, and this may be the reason that dry matter was lower in T1. Although Bergot and Breque found ADC of starch was only 38%-55% for raw starch in rainbow trout (Bergot & Breque, 1983), our result showed that black sea bream attained higher ADC values for raw corn starch. Data also suggest that digestibility of starch decreased with increasing levels of corn starch in the diet, which is in tune with the results in rainbow trout (Bergot & Breque, 1983) and European sea bass (Enes et al., 2006).

The values of ADCs of nutrients in this study were lower than that of our other researches (Ngandzali et al., 2011; Zhou et al., 2011b). Except for the protein sources, substitution of chromic oxide with yttrium oxide was the greatest difference between those researches. Chromic oxide was a traditional inert marker for feed nutrients apparent digestibility determination. Austreng et al. (2000) has comprehensively evaluated 15 trivalent metal oxides as inert markers to estimate the apparent digestibility in salmonids, and results showed that these trivalent metal oxides (including Y2O3) can substitute Cr2O3 in digestibility studies with salmonids, and can be used at lower concentrations without affecting accuracy. However, Bowen (1978) has described a potential problem with chromic oxide used in tilapia, with this compound being preferentially removed from digesta within the intestine, or even lost from branchial elimination of food particles during ingestion. Results observed by Davies and Gouveia (2006) showed that ADCs of nutrients obtained from Cr2O3 closely matched those obtained from Y2O3 holding the same relative trends. But obviously, ADCs values of protein, lipid and energy were all lower for Y2O3 compared with Cr2O3, though with different extent. That was in tune with our results in black sea bream researches, and we may impute that phenomenon to accumulation of Y2O3 in the faeces, due to its separation from the diet components. But, further research is required to better understand this phenomenon.

Results in this study showed a negative correlation between lipid content in body composition and dietary raw corn starch level, namely positive correlation with feed lipid, while no correlation was observed between lipid content in muscle and dietary raw corn starch level. Besides, IPR was decreased as dietary raw corn starch level increased and lipid level decreased, from which it can be concluded that excess lipid in diets with higher lipid level was mostly deposited as intraperitoneal fat, rather than providing protein-sparing effect. In other words, it may indicated that black sea bream can utilize carbohydrate (raw corn starch) more efficiency than lipid in dietary as an energy source in present dietary design at certain levels, and when dietary lipid was supplied in excess (such as in diet 1), a proportion of dietary lipid was deposited as intraperitoneal fat, resulting in limited protein-sparing effect, which was also can be concluded from the growth performance in the experimental fish. Protein content in whole body and muscle significantly increased with increase of the dietary raw corn starch level, which indicated dietary carbohydrate improved protein utilization, and was in agreement with the results reported for gilthead sea bream (Enes et al., 2008).

In the present experiment, no significant difference was found in serum total protein and glucose (Table 5). In species such as cod, feeding high carbohydrate diets resulted in hyperglycemia (Hemre et al., 1989) which is a stress response in fish, but in other species, such as eel (Suarez et al., 2002), silver sea bream (Leung & Woo, 2012) and black sea bream in present study, serum glucose level was not affected, indicating that high level of raw corn starch (300 g kg-1) in black sea diet is unlikely stressful for it. Leung and Woo ( 2012) attributed the phenomenon that some fishes which do not display hyperglycemia in response to carbohydrate-rich diets to their ability to tightly regulate blood glucose levels by virtue of their ability to rapidly metabolize dietary carbohydrate intake, and marked induction of GK mRNA expression was also found in his study, which would subsequently lead to a higher rate of glucose uptake from the circulation into liver. The reasons above may also be responsible for the invariable serum glucose in our study, however, it need to be confirmed further. Moreover, the raw starch which prolonged the gastric evacuation time may also be contributed partially for the invariable serum glucose as it make this fish have much more time to eliminate the glucose. At last, the special structure of raw starch granule as described above along with the the long gastric evacuation, ensure there was no larger content fluctuation of serum glucose. Apparently, the black sea bream is able to utilize 300 g kg-1 raw corn starch as their energy source without exhibiting any hyperglycemic response.

The fate of excess hepatic dietary glucose is either towards glycogen synthesis or towards lipogenesis. In the present study, a positive effect of dietary carbohydrate on hepatic glycogen level was observed in T1-T3. Liver glycogen content did not increase with the inclusion level of raw corn starch in T4-T6 while HSI increased with dietary starch level, which appears that it is liver lipid that contributes to larger liver size much more than liver glycogen, especially in T4-T6. The reason for that may be that high starch level stimulate lipogenesis, resulting in increased liver weight (Hemre et al., 2002b; Stone, 2003).

In present study, serum cholesterol and triacylglyceride concentrations observed indicated a more active lipid transport, in response to the higher dietary lipid level with lower starch level. This could relate to the higher lipid contents in muscle and whole body of fish fed experimental diets. Besides, as we concluded that liver fat may attain high content in high starch level dietary, namely low lipid level, it may indicate that fat synthesis was more than lipolysis while starch level increased in low starch level (such as T1-T2 ), while it was opposite in high starch level. Kamalam et al. (2012) found that though the transcript abundance of SREBP1c (sterol regulatory element binding protein 1-like), a transcription factor involved in regulation of FAS (fatty acid synthase) gene expression, did not increase with carbohydrate intake, evidence for increased hepatic lipogenic capacities was still observed through the enhanced activity of FAS and the elevated mRNA levels of ACLY (ATP citrate lyase), the main lipogenic enzyme diverting glycolytic carbon flux into lipid biosynthesis (Lin et al., 1977b; Lin et al., 1977a). Besides, results observed by Kamalam et al. (2012) showed that no stimulatory effect of dietary carbohydrates on the expression of lipogenic enzymes in the adipose tissue. All the results of Kamalam et al confirmed the results of present study and our hypothesis that dietary starch evaluated liver fat deposition while has sparse effect on fat deposition in perivisceral adipose tissue, resulting in HSI values change with dietary lipid, as well as in CF alteration which may be contributed mainly by abdominal fat.

In conclusion, the present study provides some insight into the carbohydrate nutrition of juvenile black sea bream and indicates that the optimal incorporation levels of carbohydrate (raw corn starch) for juvenile black sea bream for maximum WG was 184 g kg-1. What's more, the most important result of this study was that black sea bream can utilize high raw corn starch (300 g kg-1) without compromising growth performance (SGR), feed utilization (FER and PER) and digestibility (protein). As a result, the ability of them to tolerate a carbohydrate-rich diet (with the decrease in lipid inclusion) brings a potential profit-benefit in the formulation of an artificial diet for sea bream culture.