Introduction

Protein is the major macronutrient in fish which provides essential and non-essential amino acids for protein synthesis and energy for maintenance and growth (Kim et al. 2002). However, protein is the most expensive component in fish feed (NRC 1993; Mohseni et al. 2013). For successful aquaculture practices, it is needed to determine the minimum level of protein at which fish can attain maximum growth as well as the operational costs can be saved (NRC 2011). It is well documented that dietary protein requirements for most of the fish species are found to be between 30 and 55 % of the diet; however, it depends on the fish species, fish size, dietary protein sources and environmental conditions (Hepher 1988; NRC 1993).

Parrot fish, Oplegnathus fasciatus, is one of the important commercial marine finfish species cultured in cages in Korea as well as in East Asia (Meng et al. 1995). In 2015, total production of parrot fish in Korea was approximately 1150 metric tons (mt) where its cage aquaculture contributed 1050 mt (National Statistical Office 2015). It has high market value and consumer demand. Despite the economic value, very little information is available on the nutritional requirements of parrot fish. This study was conducted to determine the optimum dietary protein level in diets for juvenile parrot fish at a specific dietary energy level.

Methods

Experimental diets

Five experimental diets using white fish meal and casein as the basic protein sources were prepared with protein levels of 35, 40, 45, 50 and 60 % which were designated as CP35, CP40, CP45, CP50 and CP60, respectively, at the expense of α-potato starch and squid liver oil (Table 1). The experimental diets were formulated to be isocaloric (16.7 kJ/g energy) based on calculation by Garling and Wilson (1976). The actual nutrient contents in experimental diets are shown in Table 2. The experimental diets preparation and storage have been done following Bai and Kim (1997). In brief, ingredients of the experimental diets were rigorously mixed with a mixer, and then, squid liver oil and EPA & DHA together with 30 % water were added and further mixed to make a mash. The mashed feeds were finally passed through a laboratory pelleting machine to get 2-mm-diameter pellets. The wet pellets were then stored at −20 °C until used.

Table 1 Composition of the experimental diets (% of dry matter basis)
Full size table
Table 2 Proximate composition of experimental diets (% of dry matter basis)
Full size table

Fish and husbandry and feeding

Juvenile parrot fish, O. fasciatus were transported from Geoje Marine Hatchery (Geoje, Korea) of National Institute of Fisheries Science (NIFS), Korea to Youngchang Fisheries Farm (Tongyoung, Korea). Before the start of the experiment, all fish were reared in a circular concrete tank with 5000 L well water and were fed a commercial diet for 2 weeks. For experimental purposes, 15 floating net cages (each size: 60 × 40 × 90 cm, W × L × H) were installed in a rectangular concrete tank (5 × 5 × 3 m, W × L × H) having flow through system. After a 2-week conditioning period, a group of 20 fish with an average initial weight of 7.1 ± 0.06 g (mean ± SD) was randomly distributed into the cages in triplicates according to the five experimental diets.

Fish were fed one of the five isocaloric diets twice (0900 and 1800 h) a day at a level of 4 % of wet body weight in the first 4 weeks and 3 % in the second 4 weeks, respectively, with apparent satiety. Total fish weight in each cage was determined every 2 weeks after anesthesia with 100 ppm of MS 222 (tricaine methanesulfonate), and the amount of feeds were adjusted accordingly. During the experimental period, water flow rate was maintained at 3 L/min and water temperature maintained between 19 and 22 °C due to natural fluctuations in seawater temperature. Supplemental aeration was provided to maintain dissolved oxygen levels near saturation.

Sample collection, analyses and calculations

At the end of the feeding trial, all fish were weighed and counted to calculate growth parameters such as percent weight gain (WG), feed efficiency (FE) and specific growth rate (SGR), feed utilization parameters such as protein efficiency ratio (PER), protein retention efficiency (PRE), energy retention efficiency (ERE), biometrics such as hepatosomatic index (HSI) and condition factor (CF), blood parameters such as hematocrit (percentage of packed cell volume-PCV %) and hemoglobin (Hb); also, survival rate of juvenile parrot fish was determined (Table 3). After the final weighing, five fish were randomly collected from each aquarium, and blood samples were obtained using heparinized syringes from the caudal vein and pooled in the vials according to the number of diets. Fish blood hematocrit (PCV) was determined by the microhematocrit method (Brown 1980), and hemoglobin (Hb) was measured by the cyanmethemoglobin method using Drabkin’s solution. In this study, human blood hemoglobin standard (Sigma-Aldrich, St. Louis, MO, USA) was used for fish blood hemoglobin analysis. Fish liver weight was taken from dissected fish to determine the hepatosomatic index (HSI). Condition factor was measured after collection of fish weight and length data (Table 3). Crude protein, lipid, moisture and ash of whole-body samples were determined by the AOAC methods (1995). In brief, samples of diets and fish were dried to a constant weight at 135 °C for 2 h to determine moisture content. Ash was determined by incineration using muffle furnace at 550 °C for 3 h. Crude lipid was determined by Soxhlet extraction unit using Soxtec system 1046 (Foss, Hoganas, Sweden), and crude protein content was analyzed by Kjeldahl method (N × 6.25) after acid digestion.

Table 3 Growth, feed utilization, biometrics, hematology and survival of juvenile parrot fish Oplegnathus fasciatus fed with five experimental diets for 8 weeks
Full size table

Statistical analysis

All the data were analyzed by one-way ANOVA using SAS version 9.1 software (SAS Institute, Cary, NC, USA) to test the effects of dietary protein (Zar 1984). When a significant effect of the treatments was observed, Duncan’s test was used to compare the means. Treatment effects were considered significant at P < 0.05. Broken-line regression analysis (Robbins et al. 1979) was applied to determine the optimum inclusion level of protein in the diet of juvenile parrot fish.

Results

Growth performance of parrot fish fed with experimental diets at different protein levels for 8 weeks is shown in Table 3. At the end of feeding trial, WG of fish fed with CP50 and CP60 diets were significantly higher than those of fish fed with CP35, CP40 and CP45 diets (P < 0.05). However, there were no significant differences in WG between fish fed with CP50 and CP60 diets. Fish fed with CP45, CP50 and CP60 diets showed higher FE and SGR than those of fish fed with CP35 and CP40 diets. In contrast to WG, FE and SGR, protein efficiency ratio (PER) and protein retention efficiency (PRE) decreased with increasing dietary protein levels. The highest and the lowest ERE values were observed in fish fed with CP50 and CP35 diets, respectively. Hepatosomatic index (HSI) was highest in fish fed with CP35 diet, whereas lowest HSI was observed in fish fed with CP60 diet. Condition factor (CF) followed the same trend as FE and SGR of fish fed with the experimental diets. No significant differences were found in survival rate of fish fed with the diets.

In considering hematological characteristics of fish, juvenile parrot fish fed with the CP35 diet showed significantly lower hemoglobin levels than those of the fish fed with CP40, CP50 and CP60 diets (P < 0.05). Blood hematocrit level was lower in fish fed with the CP50 diet than those of the fish fed with CP40, CP45 and CP60 diets (P < 0.05).

Whole-body proximate composition of juvenile parrot fish is shown in Table 4. The table showed that crude protein (CP) and crude lipid (CL) content in whole body increased with the increase in dietary protein levels. Significantly higher whole-body CP and moisture contents were found in fish fed with CP50 and CP60 diets than those of the fish fed with CP35 and CP40 diets. Whole-body CL content was found to be highest in fish fed with CP50 diet and lowest in fish fed with CP35 diet. No significant differences were found in fish fed with the experimental diets in terms of whole-body ash contents.

Table 4 Proximate composition (%) of the whole body of juvenile parrot fish Oplegnathus fasciatus fed with five experimental diets for 8 weeks
Full size table

Broken-line analysis of weight gain indicated that the optimum dietary protein level was 48.5 % in juvenile parrot fish (Fig. 1).

Fig. 1
figure 1

Broken-line model of percent weight gain in parrot fish fed with five different levels of dietary protein for 8 weeks

Full size image

Discussion

After 8 weeks of the feeding trial, ANOVA showed that WG of fish fed with the 50 % CP diet was significantly higher than those of fish fed with the 35 and 40 % CP diets; however, there was no significant difference between fish fed with the 50 and 60 % CP diets (Table 3). Based on broken-line analysis of WG of parrot fish, the optimum dietary protein was 48.5 %. Similarly, Hossain et al. (2010) reported that the optimum dietary protein level for silver pomfret, Pampus argenteus, was 49 % CP. In line with our result, protein requirement of some other fish species such as in Olive flounder, Paralichthys olivaceus was found to be 46.4 % (Kim et al. 2002), 47.8 % for grouper, Epinephelus malabaricus (Chen and Tsai 1994). Generally, when dietary protein levels increase, growth of fishes also increases (NRC 1993). In this experiment, WG, FE, SGR and CF of fish improved with increasing dietary protein levels up to 50 % CP; then, no further improvements were observed in these parameters at higher protein levels (Table 2).

In the present study, PER and PRE decreased with increase of protein level in the treatment groups (Table 3). The result shows that, possibly, dietary protein was efficiently utilized by fish for protein synthesis which is in accordance with Berger and Halver (1987). Similar results have been reported in other fish species (Bai et al. 1999; Kim et al. 2004, 2005; Hossain et al. 2010; Zhang et al. 2010). In contrast to our study, Kikuchi et al. (1992) and Lee et al. (2000) reported that PER values of olive flounder increased with increasing dietary protein levels. However, Dabrowski (1979) reported that the relationship between dietary protein and PER differs from species to species. In the present study, ERE increased with the increase of dietary protein levels which means dietary protein could be spared by non-protein energy sources. Dietary protein sparing helps to reduce feed cost and nitrogen waste outputs (Wang et al. 2006). Ng et al. (2008) reported that lipid plays an important role for protein sparing when dietary protein level is low in relation to the requirement which might be reflected in our experiment as well.

Hematological parameters such as hemoglobin (Hb) and hematocrit (PCV) concentration levels were affected by dietary protein levels (Table 3). Higher level of Hb and lower level PCV were found in blood of fish fed with 50 % CP diet compared with other experimental diets which may indicate the healthy condition of fish. However, Kim et al. (2004) found that dietary protein levels have no significant effect on hematological and serological characteristics of juvenile Korean rockfish.

Fish biometrics in terms of hepatosomatic index (HSI) and condition factor (CF) indicate the body condition of fish. In this study, HSI was decreased and CF was increased with protein level increment in the diets which may indicate the higher utilization of protein levels from the diets by fish. These results are in agreement with Kim and Lall (2001). Survival rate was not significantly affected among fish fed with the experimental diets.

Proximate composition in terms of ash contents of fish fed with the experimental diets was not significantly affected by dietary protein levels (Table 4) which are in accordance with Okorie et al. (2007) for juvenile Japanese eel and Kim et al. (2004) for Korean rockfish. In this experiment, the whole-body CP content increased with the increasing dietary protein levels which agree with the results found by Kim et al. (2002). Similarly, the body lipid content generally increased as the dietary protein level increased which is in agreement with Shiau and Lan (1996) for grouper and Bai et al. (1999) for yellow puffer. On the contrary, Kim et al. (2002) reported that as the CP content of whole body increases, whole-body CL content decreases.

In fine, based on the broken-line analysis of weight gain, it can be corroborated that the optimum dietary protein level for juvenile parrot fish could be 48.5 % for its maximum growth at the gross energy level of 16.7 kJ/g diet.