Effects of dietary phosphorus and starch levels on growth performance, body composition and nutrient utilization of grass carp (Ctenopharyngodon idella Val.)
Abstract
Six iso-nitrogenous (410 g kg−1) diets with three levels of total phosphorus (P4, P10 and P18 g kg−1) and two levels of starch (S200 and S350 g kg−1) were fed to triplicate groups of 30 fish to evaluate whether the high level of dietary phosphorus could improve the utilization of starch. Over 8-week-growth trial, best weight gain (WG) and specific growth rate (SGR) (P < 0.05) were observed in fish fed the P10/S200 and P18/S200 diets. WG and SGR significantly decreased as starch levels increased whereas for P4, while lipid contents of liver and whole body, hepatosomatic index and intraperitoneal fat ratio (IPF) significantly increased. These results suggested that high dietary starch will depress the growth performance and cause lipid accumulation. Within both starch levels, fish fed diet with P4 tended to produce lower (P < 0.05) WG and SGR, and had higher (P < 0.05) values of IPF. The whole body lipid, ash, calcium, phosphorus and iron contents were significantly affected by dietary phosphorus levels. Supplied phosphorus could improve the growth and decrease the whole body lipid, but there is no more effect after the phosphorus requirement was met at 10 g kg−1.
Introduction
Production of grass carp (Ctenopharyngodon idella) constitutes the largest aquaculture industry of finfish in China. In 2008, the production of grass carp reached 3.71 million tons, which was 18% of freshwater aquaculture production in China (Ministry of Agriculture 2009). In production, feeding compound feed to grass carp frequently increases intraperitoneal fat (IPF) and largely deposits mesenteric fat, which may impair the growth performance and affect the body composition of grass carp. In our previous studies (L-X. Tian, Y-J. Liu, H-J. Yang, G-Y. Liang & J. Niu, unpublished data), we found that there was a positive correlation between the dietary starch level and IPF in grass carp.
Utilization of carbohydrate by fish depends on several factors, such as fish species, dietary carbohydrate level, origin, molecular complexity and physical state (Couto, Enes, Peres & Oliva-Teles 2008). Omnivorous or herbivorous warm-water fish generally tolerate higher carbohydrate levels, using them more effectively than carnivorous species as a source of energy, whereas dietary excess carbohydrate can be stored in the form of body lipids (Krogdahl, Hemre & Mommsen 2005). A positive correlation between dietary carbohydrate level and body lipid content was previously found in some fish species (Kaushik & de Oliva Teles 1985; Lanari, Poli, Ballestrazzi, Lupi, D'Agaro & Mecatti 1999).
Phosphorus is an important constituent of the fish endoskeleton and more than a third of body phosphorus is found in the phospholipids, nucleic acids, cell membranes and energy-rich compounds (Kaushik 2001). Therefore, phosphorus plays an important role in carbohydrate, lipid and nitrogen metabolism (Kaushik 2001; Lall 2002). The body lipid content was negatively correlated with the concentrations of dietary phosphorus. This was consistent with some previous studies, which showed increasing lipid content in the muscle, whole body, liver and viscera in response to low phosphorus diets (Takeuchi & Nakazoe 1981; Rodehutscord 1996; Eya & Lovell 1997; Roy & Lall 2003). From all these fact, how important role dietary phosphorus plays in utilization of lipid and phosphorus might accelerate the utilization of lipid.
However, there is no information about the effect of dietary phosphorus and starch levels, as well as their potential interaction, on grass carp until now. Therefore, the aim of the present experiment was to study the effect of dietary phosphorus and starch levels on growth, body composition and nutrient utilization of grass carp. We try to evaluate whether the high level of dietary phosphorus (P18) could improve the utilization of starch, as excess dietary starch will be stored in the form of body lipid, and phosphorus might accelerate the utilization of lipid.
Materials and methods
Experimental diet and diet preparation
Prior to use, all feed ingredients were analysed for their proximate composition and the data obtained were used as a basis for the formulation (Table 1). Six semi-purified iso-nitrogenous [410 g kg−1 the optimum protein requirement (Dabrowski 1977)] diets with three levels of total phosphorus (P4, P10 and P18 g kg−1) and two levels of starch (S200 and S350 g kg−1) were fed to triplicate groups of 30 fish. Casein (Hulunbeier Sanyuan Milk Co., Ltd, Inner Mongolia, China) and gelatin (Rousselot Gelatin Co., Ltd, Guangdong, China) were used as protein sources. Corn starch (Langfang Starch Factory, Hebei, China) was used as a starch source. Fish oil (Gaolong Industrial Company Ltd, Fujian, China) and corn oil (Defeng Starch Sugar Company, Guangdong, China) were used as sources of lipid. Monocalcium phosphate was used as phosphorus source. Diet ingredients were ground through a 60-mm mesh. All the dry ingredients were weighed and mixed for 15 min, and then fish oil and corn oil were added and mixed for 15 min. Distilled water was added to the premixed dry ingredients and thoroughly mixed until homogenous in a Hobart-type mixer. The pellets were obtained (1.5 mm in diameter) using a pelletizer (Institute of Chemical Engineering, South China University of Technology, Guangzhou, China) and air dried to a moisture content of < 100 g kg−1. The noodle-like diets were ground, sieved and stored in plastic bags at −20°C until used.
| P4/S200 | P10/S200 | P18/S200 | P4/S350 | P10/S350 | P18/S350 | |
|---|---|---|---|---|---|---|
| Ingredients (g kg−1) | ||||||
| Casein | 420 | 420 | 420 | 420 | 420 | 420 |
| Gelatin | 40 | 40 | 40 | 40 | 40 | 40 |
| Cellulose | 245 | 220 | 180 | 95 | 70 | 30 |
| Corn starch | 200 | 200 | 200 | 350 | 350 | 350 |
| Monocalcium phosphate | 0 | 25 | 65 | 0 | 25 | 65 |
| Corn oil | 20 | 20 | 20 | 20 | 20 | 20 |
| Fish oil | 20 | 20 | 20 | 20 | 20 | 20 |
| Vitamin mixa | 10 | 10 | 10 | 10 | 10 | 10 |
| Mineral mixb | 30 | 30 | 30 | 30 | 30 | 30 |
| Choline chloride | 5 | 5 | 5 | 5 | 5 | 5 |
| Ascorbic phosphate ester | 5 | 5 | 5 | 5 | 5 | 5 |
| Taurine | 5 | 5 | 5 | 5 | 5 | 5 |
| Chemical composition (g kg−1) | ||||||
| Moisture | 75.0 | 72.8 | 72.9 | 75.5 | 73.1 | 76.4 |
| Crude protein | 406 | 409 | 407 | 414 | 414 | 407 |
| Crude lipid | 34.7 | 33.8 | 37.7 | 34.8 | 34.4 | 36.0 |
| Ash | 42.9 | 58.4 | 88.8 | 41.9 | 58.6 | 89.0 |
| Total phosphorus | 4.08 | 9.61 | 18.4 | 4.13 | 10.0 | 18.4 |
| Gross energy (kJg−1) | 16.4 | 16.4 | 16.4 | 20.1 | 20.0 | 20.1 |
| Ca/P | 0.93 | 0.74 | 0.69 | 0.95 | 0.72 | 0.66 |
- a Vitamin mix (mg kg−1 of diet): thiamine, 50; riboflavin, 50; vitamin A, 25 000 IU; vitamin E, 400; vitamin D3, 24 000 IU; menadione, 40; pyridoxine HCl, 40; cyanocobalamin, 0.1; biotin, 6; calcium pantothenate, 100; folic acid, 15; niacin, 200; inositol, 2000; and cellulose was used as a carrier.
- b Mineral mix (g kg−1 diet): sodium chloride, 2.6; potassium sulphate, 13.1; potassium chloride, 5.3; ferrous sulphate, 0.9; ferric citrate, 3.1; magnesium sulphate, 3.5; zinc sulphate, 0.04; manganese sulphate, 0.03; cupric sulphate, 0.02; cobalt chloride, 0.03; potassium iodide, 0.002; and cellulose was used as a carrier.
Fish and experimental conditions
Juvenile grass carp were obtained from a local hatchery. Before the experiment, the fish were acclimated to a basal diet (P4/S200) for 2 weeks. After the acclimatization, fish were sorted by weight and absence of physical abnormalities into uniform groups. The fish were randomly distributed to 18 experimental fibre glass tanks (98 L × 48 W × 42 H cm, water volume of 200 L) at an equal stocking rate of 30 fish per tank connected to a recirculation system. The initial body weight averaged 4.83 g (SEM = 0.02 g with n = 18). The feeding trial lasted for 8 weeks. The fish were fed with a daily ration of 40–50 g kg−1 of body weight divided into two meals day−1. The fish were weighed every 2 weeks for daily ratio estimated. During the trial period, the diurnal cycle was 12-h light/12-h dark. Water quality parameters monitored weekly was as follows (mean ± SEM): temperature, 27.5 ± 2.4°C; dissolved oxygen, 7.5 ± 0.26 mg L−1; total ammonia-nitrogen, 0.097 ± 0.005 mg L−1; pH, 7.8 ± 0.07, respectively.
Sampling and analytical methods
At the beginning of the feeding trial, 20 juveniles were randomly sampled from the initial fish and killed for analyses of whole body composition. At the termination of the 8-week feeding trail, fish in each tank were weighed and sampled for tissue analysis 24 h after the last feeding. Nine fish from each tank were randomly collected for proximate analysis, three for analysis of whole body composition and six were anaesthetized with tricaine methane sulphonate (MS222) (0.05 g L−1) for blood collection and to obtain weights of individual whole body, viscera, liver and mesenteric fat. White muscle from both sides of the fillets without skin was dissected. The plasma was separated using centrifugation (1790 g for 10 min) and stored at −20°C until analysed. Fish carcasses and another six fish were cooked in a microwave oven for 6 min, and the surrounding tissues were removed from the vertebrae.
Vertebrae were rinsed with deionized water, dried and ground for mineral analyses. To determine the mineral contents on a fat-free dry basis, ground vertebrae were extracted twice with 20 mL chloroform and methanol (1:1, v/v), air dried at room temperature in a fume hood and oven-dried at 105°C for 24 h. Approximately 0.10–0.15 g dried and finely ground samples (the feed, whole body, vertebrae and liver) were digested with 15 mL 65–68% nitric acid and 2 mL 72% perchloric acid using Kjeldahl flasks. After digestion, samples were diluted with deionized water to 50 mL and determined for mineral contents by inductively coupled plasma atomic-emission spectrophotometer [ICP; model: IRIS Advantage (HR); Thermo JarrelAsh Corporation, Boston, MA, USA].
Diets and fish samples (including white muscle and liver) were analysed in duplicate for proximate composition. Moisture, crude protein, crude lipid and ash were determined using standard methods (AOAC 1995). Moisture was determined by drying in an oven at 105°C for 24 h; crude protein (N × 6.25) was analysed using the Kjeldahl method after acid digestion (1030-Auto-analyzer; Tecator, Höganäs, Sweden); crude lipid was determined using the ether-extraction method by Soxtec System HT (Soxtec System HT6; Tecator, Sweden); crude ash by incineration in a muffle furnace (Heraeus instruments; Hanau, Hessen, Germany) at 550°C for 5 h. The concentrations of plasma triacyglyceride, total cholesterol (TC), low density lipoprotein cholesterol (C-LDL), high density lipoprotein cholesterol and phosphorus were assayed by enzymatic procedure using automatic biochemical analyser and attached kit (Hitachi 7170; DAICHI, Tokyo, Japan) from a clinical laboratory of the First Affiliated Hospital of Sun Yat-sen University.
Calculation and statistical analysis
The growth performance and morphometrical indices of the fish were assessed using the following parameters: weight gain (WG) = 100 × (final body weight−initial body weight)/initial body weight; specific growth ratio (SGR) = 100 × ln (final weight/initial weight)/days of the experiment; feed conversion ratio (FCR) = dry feed consumed/wet WG; protein efficiency ratio (PER) = fish wet WG/protein intake; protein production value (PPV) = 100 × (final body protein−initial body protein)/total protein fed; phosphorus retention (PR) = 100 × (final phosphorus fish content−initial phosphorus fish content)/phosphorus intake; lipid retention (LR) = 100 × (final body lipid−initial body lipid)/total lipid fed; condition factor = 100 × body weight (g)/body length (cm)3; viscerosomatic index = 100 × (viscera weight/whole-body weight); hepatosomatic index (HSI) = 100 × (hepatosomatic weight/whole-body weight); and IPF = 100 × (IPF weight/whole-body weight).
A 3 × 2 factorial trial was designed to evaluate effects of dietary phosphorus and starch levels, and also possible interaction between two factors. All data were presented as mean ± SEM and subjected to one-way analysis of variance (anova) to determine if significant differences occurred in treatments. If a significant difference was identified, differences among means were compared using Duncan's (1955) multiple range test by P < 0.05. Statistical analysis of data was done using two-way anova to see if a significant P and starch interaction exists. All statistical analysis was done using a spss 13.0 software package (SPSS 13.0 for Windows; SPSS Inc., Chicago, IL, USA).
Results
Only six fish died during the experiment, and mortality was unrelated to treatments. No abnormal behaviours or body shapes of fish were observed in any treatment group. As shown in Table 2, both dietary level of phosphorus and starch had significant effects (P < 0.05) on FBW, WG, SGR, FCR, PER and PPV. At the same time, significant interaction between dietary phosphorus and starch was identified on the examined parameters. Best FBW, WG, SGR, FCR, PER and PPV (P < 0.05) were observed in fish fed diets had P10/S200 and P18/S200. When fish-fed diets had low level of phosphorus (P4), WG was not affected by the dietary starch level, but when fish-fed diets had P10 and P18, WG of fish-fed diets had S200 was significantly higher than fish-fed diets had S350. In diets with S200, WG increased by dietary phosphorus up to 10 g kg−1 (P < 0.05) and then leveled off beyond this level, but in diets had S350, WG increased by dietary phosphorus up to 10 g kg−1 and then decreased (P < 0.05). FBW, SGR, PER and PPV followed similar trend as WG. Both dietary starch levels, PR and LR decreased significantly (P < 0.05) as dietary phosphorus level increased. On the other hand, no significant interaction between dietary phosphorus and starch was identified on PR or LR.
| P/S | FBW | WG | SGR | FCR | PER | PPV | PR | LR |
|---|---|---|---|---|---|---|---|---|
| 4/200 | 19.01 ± 0.73a | 294 ± 14.7a | 2.45 ± 0.07a | 1.37 ± 0.06d | 1.80 ± 0.09a | 34.0 ± 0.9a | 73.2 ± 6.6e | 177 ± 22.8c |
| 10/200 | 29.73 ± 0.93c | 517 ± 19.1c | 3.25 ± 0.05c | 0.92 ± 0.02a | 2.60 ± 0.06c | 42.4 ± 0.5b | 59.3 ± 0.5d | 130 ± 15.5ab |
| 18/200 | 28.30 ± 1.05c | 486 ± 21.5c | 3.16 ± 0.07c | 0.96 ± 0.02ab | 2.51 ± 0.06c | 41.5 ± 0.7b | 35.1 ± 1.3b | 119 ± 0.8a |
| 4/350 | 18.67 ± 1.11a | 285 ± 20.9a | 2.41 ± 0.10a | 1.38 ± 0.05d | 1.76 ± 0.07a | 33.3 ± 1.0a | 67.6 ± 7.6de | 221 ± 17.5d |
| 10/350 | 24.69 ± 1.69b | 411 ± 35.3b | 2.91 ± 0.12b | 1.02 ± 0.09b | 2.43 ± 0.21c | 39.8 ± 2.7b | 48.7 ± 5.7c | 159 ± 16.7bc |
| 18/350 | 20.12 ± 0.32a | 315 ± 7.8a | 2.54 ± 0.03a | 1.22 ± 0.05c | 2.01 ± 0.08b | 35.7 ± 1.3a | 25.4 ± 1.2a | 119 ± 17.1a |
| Two-way anova | ||||||||
| S | 0.000 | 0.000 | 0.000 | 0.000 | 0.001 | 0.000 | 0.002 | 0.009 |
| P | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.001 | 0.000 | 0.000 |
| S × P | 0.000 | 0.000 | 0.000 | 0.005 | 0.025 | 0.009 | 0.642 | 0.106 |
- Data are presented as mean ± SE. Mean values within a row with unlike superscript letters were significantly different (P < 0.05).
- FBW, final body weight; WG, weigh gain; SGR, specific growth ratio; FCR, feed conversion ratio; PER, protein efficiency ratio; PPV, protein production value; PR, phosphorus retention; LR, lipid retention.
As shown in Table 3, dietary level of phosphorus had significant effects (P < 0.05) on moisture, protein contents and lipid contents of whole body and muscle, but not on liver. On the other hand, dietary level of starch had significant effects (P < 0.05) on moisture, protein contents and lipid contents of whole body and liver, but not on muscle. No significant interaction between dietary phosphorus and starch was identified on the examined parameters except whole body protein content. In both dietary starch levels, whole body lipid content was significantly decreased with increasing phosphorus level, but there was no significant difference between P10 and P18. Correspondingly, whole body protein content was significantly increased with increasing phosphorus level. The protein and lipid content in muscle followed similar trend. Interestingly, the lipid content in liver was not affected by dietary phosphorus level, but significantly increased with increasing starch level (P < 0.05). Accordingly, HSI performed the same trend. In each dietary phosphorus level, IPF significantly increased with increasing starch level (P < 0.05).
| P/S | 4/200 | 10/200 | 18/200 | 4/350 | 10/350 | 18/350 | Two-way anova | ||
|---|---|---|---|---|---|---|---|---|---|
| S | P | S × P | |||||||
| Whole body | |||||||||
| Moisture | 738 ± 4.7b | 774 ± 3.0cd | 781 ± 2.2d | 717 ± 10.0a | 768 ± 8.9c | 772 ± 5.7cd | 0.002 | 0.000 | 0.070 |
| Protein | 133 ± 1.1a | 137 ± 0.5b | 141 ± 1.3c | 133 ± 0.8a | 137 ± 1.2b | 142 ± 1.7c | 0.003 | 0.000 | 0.018 |
| Lipid | 88.6 ± 7.3c | 50.2 ± 1.1a | 48.9 ± 4.3a | 109 ± 8.5d | 64.4 ± 9.9b | 58.1 ± 5.6ab | 0.001 | 0.000 | 0.304 |
| Muscle | |||||||||
| Moisture | 790 ± 6.3a | 799 ± 2.8b | 805 ± 1.1c | 789 ± 1.6a | 800 ± 1.3bc | 799 ± 0.5b | 0.176 | 0.000 | 0.456 |
| Protein | 168 ± 1.2ab | 171 ± 0.9b | 175 ± 3.2 c | 166 ± 1.1a | 169 ± 1.6b | 175 ± 0.6c | 0.575 | 0.000 | 0.196 |
| Lipid | 12.5 ± 1.2b | 8.02 ± 1.41a | 7.91 ± 1.31a | 13.5 ± 1.3b | 9.72 ± 1.02a | 9.11 ± 1.21a | 0.052 | 0.000 | 0.888 |
| Liver | |||||||||
| Moisture | 561 ± 30b | 644 ± 26.9c | 634 ± 6.2c | 525 ± 40.2ab | 540 ± 14.5b | 483 ± 26.4a | 0.000 | 0.021 | 0.010 |
| Protein | 98.5 ± 1.2a | 110 ± 3.3b | 108 ± 4.1b | 95.3 ± 5.0a | 108 ± 5.7b | 106 ± 4.7ab | 0.030 | 0.218 | 0.157 |
| Lipid | 223 ± 42.3a | 173 ± 35.7a | 200 ± 10.0a | 292 ± 18.1b | 282 ± 13.4b | 289 ± 36.1b | 0.000 | 0.226 | 0.519 |
| Morphometry | |||||||||
| CF | 2.1 ± 0.08 | 2.2 ± 0.11 | 2.2 ± 0.06 | 2.2 ± 0.03 | 2.2 ± 0.03 | 2.2 ± 0.03 | 0.122 | 0.402 | 0.588 |
| VSI | 10.0 ± 0.45cd | 8.7 ± 0.24ab | 8.2 ± 0.37a | 10.8 ± 0.25d | 9.0 ± 0.82abc | 9.5 ± 1.18bc | 0.023 | 0.002 | 0.432 |
| HSI | 2.6 ± 0.10ab | 2.3 ± 0.28a | 2.4 ± 0.15a | 3.5 ± 0.30d | 3.0 ± 0.12bc | 3.1 ± 0.37cd | 0.000 | 0.035 | 0.756 |
| IPF | 2.4 ± 0.25c | 1.2 ± 0.21a | 1.2 ± 0.08a | 3.3 ± 0.21d | 1.6 ± 0.08b | 1.6 ± 0.08b | 0.000 | 0.000 | 0.046 |
- Data are presented as mean ± SE. Mean values within a row with unlike superscript letters were significantly different (P < 0.05).
- CF, condition factor; VSI, viscerosomatic index; HIS, hepatosomatic index; IPF, Intraperitoneal fat ratio.
Plasma TC, triacylglyceride (TG), C-LDL and phosphorus concentrations were significantly affected by dietary phosphorus level, but not by dietary starch level (Table 4). In S200, TC, TG and C-LDL concentrations were significantly decreased with increasing phosphorus level, but there is no significant difference between P10 and P18. On the other hand, in S350, there is no significant difference in all phosphorus levels except that C-LDL concentration was significantly decreased with increasing phosphorus level. Correspondingly, plasma phosphorus level significantly increased, but still no significant difference between P10 and P18.
| P/S | TC | TG | C-HDL | C-LDL | P |
|---|---|---|---|---|---|
| 4/200 | 9.61 ± 3.38c | 2.47 ± 0.90b | 1.41 ± 0.29b | 1.90 ± 0.62c | 1.46 ± 0.32a |
| 10/200 | 4.53 ± 0.12ab | 0.84 ± 0.09a | 0.92 ± 0.03a | 0.58 ± 0.06a | 2.43 ± 0.11b |
| 18/200 | 3.07 ± 1.36a | 0.95 ± 0.03a | 0.98 ± 0.03a | 0.49 ± 0.01a | 2.81 ± 0.10b |
| 4/350 | 6.58 ± 0.04b | 1.63 ± 0.41a | 1.14 ± 0.10ab | 1.44 ± 0.10bc | 1.49 ± 0.26a |
| 10/350 | 5.76 ± 0.74ab | 0.94 ± 0.23a | 0.97 ± 0.25a | 1.19 ± 0.28b | 2.41 ± 0.17b |
| 18/350 | 4.56 ± 0.60ab | 1.23 ± 0.31a | 0.80 ± 0.32a | 0.61 ± 0.25a | 2.68 ± 0.43b |
| Two-way anova | |||||
| S | 0.890 | 0.477 | 0.196 | 0.533 | 0.742 |
| P | 0.001 | 0.001 | 0.076 | 0.000 | 0.000 |
| S×P | 0.044 | 0.095 | 0.051 | 0.030 | 0.863 |
- Data are presented as mean ± SE. Mean values within a row with unlike superscript letters were significantly different (P < 0.05).
- TC, total cholesterol; TG, triacylglyceride; C-LDL, low density lipoprotein cholesterol; C-HDL, high density lipoprotein cholesterol; P, phosphorus.
As showed in Table 5, ash and phosphorus contents of whole body increased (P < 0.05) linearly as dietary phosphorus increased. Whole body Ca content was significantly increased with increasing phosphorus level, but no significant difference was found between P10 and P18, whereas Fe level decreased, but still no significant difference between P10 and P18. Phosphorus contents of vertebrae and liver were significantly higher in P18 than in P4. Correspondingly, Fe content of liver was significantly lower in P18 than in P4.
| P/S | 4/200 | 10/200 | 18/200 | 4/350 | 10/350 | 18/350 | Two-way anova | ||
|---|---|---|---|---|---|---|---|---|---|
| S | P | S × P | |||||||
| Whole body (wet weight) | |||||||||
| Ash (gkg−1) | 25.5 ± 0.7a | 30.3 ± 1.3b | 33.0 ± 1.6c | 25.4 ± 0.7a | 29.1 ± 1.2b | 33.8 ± 1.8c | 0.777 | 0.000 | 0.392 |
| Ca (gkg−1) | 7.41 ± 0.21a | 9.22 ± 0.31ab | 10.9 ± 0.3b | 7.22 ± 2.81a | 8.93 ± 0.71ab | 10.1 ± 0.4b | 0.477 | 0.002 | 0.897 |
| P (gkg−1) | 4.62 ± 0.22a | 5.61 ± 0.11bc | 6.42 ± 0.22d | 4.43 ± 1.71a | 5.33 ± 0.33b | 5.92 ± 0.11c | 0.009 | 0.000 | 0.512 |
| Fe (mgkg−1) | 64.7 ± 4.1cd | 55.9 ± 6.7bc | 51.7 ± 3.2ab | 66.2 ± 8.6d | 48.7 ± 3.2ab | 45.7 ± 3.6a | 0.146 | 0.000 | 0.341 |
| Vertebrae (dry matter weight) | |||||||||
| Ca (gkg−1) | 179 ± 14.4 | 180 ± 8.9 | 183 ± 13.4 | 171 ± 2.3 | 173 ± 5.9 | 183 ± 2.9 | |||
| P (gkg−1) | 84.0 ± 2.5a | 92.5 ± 4.8bc | 95.8 ± 7.1c | 87.3 ± 0.8ab | 92.8 ± 1.1bc | 95.7 ± 0.2c | 0.525 | 0.002 | 0.693 |
| Fe (mgkg−1) | 147 ± 16.3 | 132 ± 5.3 | 134 ± 10.7 | 150 ± 28.1 | 142 ± 16.4 | 114 ± 5.2 | |||
| Zn (mgkg−1) | 200 ± 4.5 | 197 ± 25.6 | 194 ± 13.5 | 193 ± 16.4 | 187 ± 29.4 | 182 ± 20.8 | |||
| Cu (mgkg−1) | 74.6 ± 24.5 | 75.0 ± 11.0 | 67.8 ± 6.2 | 84.2 ± 14.1 | 82.4 ± 12.7 | 64.3 ± 4.3 | |||
| Liver (wet weight) | |||||||||
| Ca (gkg−1) | 2.82 ± 1.21 | 2.11 ± 1.10 | 2.02 ± 0.81 | 2.62 ± 1.01 | 2.44 ± 0.92 | 2.52 ± 0.21 | |||
| P (gkg−1) | 1.81 ± 0.02ab | 2.23 ± 0.11c | 2.21 ± 0.11c | 1.72 ± 0.06a | 1.91 ± 0.08b | 2.11 ± 0.18c | 0.019 | 0.000 | 0.302 |
| Fe (mgkg−1) | 199 ± 1.9c | 168 ± 5.9bc | 143 ± 17ab | 177 ± 16.5c | 142 ± 8.1ab | 135 ± 25.7a | 0.033 | 0.001 | 0.647 |
| Zn (mgkg−1) | 68.0 ± 7.5 | 56.1 ± 13.4 | 51.5 ± 11.2 | 67.4 ± 19.0 | 54.0 ± 12.5 | 54.7 ± 10.0 | |||
| Cu (mgkg−1) | 39.6 ± 4.7c | 30.2 ± 4.5ab | 25.9 ± 1.3a | 42.5 ± 4.0c | 35.9 ± 3.1bc | 38.9 ± 3.5c | 0.001 | 0.003 | 0.83 |
- Data are presented as mean ± SE. Mean values within a row with unlike superscript letters were significantly different (P < 0.05).
- Ca, calcium; P, phosphorus; Fe, iron; Zn, zinc; Cu, copper.
Discussion
In general, for freshwater or warm water fish species, such as the tilapia and the common carp, the dietary incorporation of carbohydrate can amount to 400 g kg−1 (Wilson 1994). In contrast, in the present study (Table 2), WG of fish fed diets with S200 were significantly higher than fish fed diets with S350 when fish fed diets were with P10 and P18. These results are similar to those reported for most fishes (Ballestrazzi, Lanari & D'Agaro 1998; Hutchins, Rawles & Gatlin 1998; Peres & Oliva-Teles 2002). In our previous study (L-X. Tian, Y-J. Liu, H-J. Yang, G-Y. Liang & J. Niu, unpublished data) on grass carp, there was no significant effect on growth performance and feed utilization as the dietary starch level from 200 to 330 g kg−1. In contrast, in the present study, high dietary starch (S350) would depress the growth performance of grass carp. The most difference between the two studies was dietary protein level. The optimum protein requirement for grass carp was 410 g kg−1 (Dabrowski 1977), so as in the present study. It was higher than the protein level used for our previous study (235 g kg−1). Grass carp fed with diets of much lower protein content (235 g kg−1), and then dietary starch could become an important source of energy. The protein level might be the main factor to decrease growth performance when the dietary starch concentration increased from 200 to 350 g kg−1. Whether lower protein content will change the result or not, this needs more trial to confirm.
In this study, the lipid contents of whole body and liver were significantly increased with increasing starch level, and the HSI and IPF followed the similar trend. An increase in whole body lipid at higher level of dietary carbohydrate has been reported in several fish species (Jafri 1995; Lanari et al. 1999; Stone 2003). Similarly, increase of fat deposition in the liver with the increase of available dietary carbohydrate was also observed in other studies (Lanari et al. 1999; Peres & Oliva-Teles 2002). As moisture in liver was significantly lower of fish fed diets with S350, although the lipid content was significantly higher, and protein content was similar between the two starch levels (Table 3), the lipid content was one of the reasons of the higher HSI in fish-fed diets with S350. In fish, excessive dietary carbohydrate can be stored in the form of body lipids (Krogdahl et al. 2005).
So, high dietary starch (S350) will depress the growth performance of grass carp and cause lipid accumulation.
Results of the present study showed that, signs of phosphorus deficiency in this experiment were characterized by reduced growth (Table 2), an increase in body lipid content (Table 3) and slightly reduced mineralization (Table 5). These signs have also been observed in several studies (Yang, Lin, Liu & Liou 2006; Zhang, Mai, Ai, Zhang, Duan, Tan, Ma, Xu, Liufu & Wang 2006; Shao, Ma, Xu, Hu, Xu & Xie 2008; Luo, Tan, Liu & Wang 2009; Yuan, Yang, Gong, Luo, Yu, Yan & Yang 2011). The lower WG and feed utilization under low phosphorus conditions (P4) might be due to insufficient phosphorus being available for growth after being allocated and utilized in other physiological processes (Brown, Jaramillo & Gatlin 1993). Unexpectedly, in the present study, high level of dietary phosphorus (P18) not only could not improve the utilization of starch but also significantly impaired the growth performance in S350 (Table 2). This might be due to the dietary phosphorus level (P18) being too high for grass carp. As in our previous studies (L-X. Tian, Y-J. Liu, H-J. Yang, G-Y. Liang & J. Niu, unpublished data), the optimum dietary phosphorus requirement of grass carp was about 10 g kg−1. These results were similar to those reported for most fishes (Eya & Lovell 1997; Kim, Kim, Song, Lee & Jeong 1998; Vielma, Koskela & Ruohonen 2002). In this study, Fe contents of whole body and liver were significantly lower in P18 than in P4 (Table 5). Lall (2002) and Vielma et al. (2002) suggested that a high phosphorus diet may chelate zinc and other trace elements and reduce their absorption and metabolism in fish, which resulted in reduced growth.
The decreased lipid and increased protein contents in both the whole body and muscle with increasing dietary phosphorus (Table 3) was consistent with some previous studies including common carp Cyprinus carpio (Takeuchi & Nakazoe 1981), channel catfish Ictalurus punctatus (Eya & Lovell 1997), haddock Melanogrammus aeglefinus L (Roy & Lall 2003), Japanese seabass Lateolabrax japonicus (Zhang et al. 2006) and juvenile grouper Epinephelus coioides (Ye, Liu, Tian, Mai, Du, Yang & Niu 2006). In the present study, lower protein contents and PPV with higher lipid contents, LR, HSI and IPF in the fish fed a low phosphorus diet(P4) might explain that inhibition of β-oxidation of fatty acid resulted in a lower utilization of lipid as an energy source, then fish utilized protein for energy as an alternative to lipid (Roy & Lall 2003). On the other hand, fish fed diet with P18 showed the lowest lipid contents, LR, HSI and IPF, but was not significantly lower than fish fed diet with P10. So, when the dietary phosphorus met the need of grass carp, excess dietary phosphorus could not improve the utilization of lipid anymore.
Sugiura, Kelsey and Ferraris (2007) found that serum phosphorus concentration responded clearly and rapidly to dietary phosphorus levels, and it took 2 weeks for serum phosphorus to reach a stable saturation plateau. In this study, the concentration of plasma phosphorus was significantly affected by dietary phosphorus level (Table 4). The diminished plasma phosphorus of fish-fed phosphorus deficient diets was also reported in several other fishes (Brown et al. 1993; Rodehutscord 1996; Roy & Lall 2003; Yang et al. 2006). In this study, both TG and TC contents were decreased by increasing dietary phosphorus. Similar results were also observed in black seabream Sparus macrocephalus (Shao et al. 2008) and Chinese sucker Myxocyprinus asiaticus (Yuan et al. 2011). Perhaps the effect of dietary phosphorus on lipid metabolism could explain the reductions of TG and TC contents in plasma. Similarly, there still no significant difference in plasma phosphorus, TG or TC contents between P10 and P18.
In this study, a marked increase in crude ash, Ca and phosphorus contents of whole body was observed, so as the phosphorus contents of vertebrae and liver. Similar results were reported by several workers (Roy & Lall 2003; Schaefer, Koppe, Meyer-Burgdorff & Guenther 2007; Shao et al. 2008; Yuan et al. 2011). In this study, Fe contents of whole body and liver significantly decreased with increasing dietary phosphorus level. This might be caused by competitive inhibition of these cations with phosphorus during intestinal absorption. High level of dietary phosphorus also decreased magnesium and zinc levels in rainbow trout Oncorhynchus mykis vertebrae (Hardy & Shearer 1985), Atlantic salmon Salmo salar (Vielma & Lall 1998), orange-spotted grouper Epinephelus coioides (Zhou, Liu, Mai & Tian 2004).
In this study, PR was not only affected by dietary phosphorus level but also by dietary starch level. As excessive dietary starch could be stored in the form of body lipids, and dietary phosphorus could accelerate the utilization of lipid (Table 3), we suspected that the lower PR of fish fed higher starch level diet, not just because of the poor WG, but might also due to that the excessive dietary starch need more phosphorus to deal with. Such liver phosphorus content in fish fed with high starch was significantly lower than in fish fed with low starch. As liver is the location for lipid utilization, it might need more phosphorus to deal with the excessive body lipid that was stored by excess dietary starch.
The present study indicated that as high level of dietary starch (S350) will depress the growth performance of grass carp and cause lipid accumulation, a proper level of dietary phosphorus (P10) was essential to maintain normal physiology, growth and bone mineralization of juvenile grass carp. Supplied phosphorus could improve the growth and decrease the body lipid, but there was no more effect after the phosphorus was met at 10 g kg−1, and so high level of dietary phosphorus (P18) could not improve the utilization of starch.
