Fish Oil Replacement with Poultry Oil in the Diet of Tiger Puffer (Takifugu rubripes): Effects on Growth Performance, Body Composition, and Lipid Metabolism

Booming fish farming results in relative shortage of fish oil (FO), making it urgent to explore alternative lipid sources. This study comprehensively investigated the efficacy of FO replacement with poultry oil (PO) in diets of tiger puffer (average initial body weight, 12.28 g). An 8-week feeding trial was conducted with experimental diets, in which graded levels (0, 25, 50, 75, and 100%, named FO-C, 25PO, 50PO, 75PO, and 100PO, respectively) of FO were replaced with PO. The feeding trial was conducted in a flow-through seawater system. Each diet was fed to triplicate tanks. The results showed that FO replacement with PO did not significantly affect the growth performance of tiger puffer. FO replacement with PO at 50-100% even slightly increased the growth. PO feeding also had marginal effects on fish body composition, except that it increased the liver moisture content. Dietary PO tended to decrease the serum cholesterol and malondialdehyde content but increase the bile acid content. Increasing levels of dietary PO linearly upregulated the hepatic mRNA expression of the cholesterol biosynthesis enzyme, 3-hydroxy-3-methylglutaryl-CoA reductase, whereas high levels of dietary PO significantly upregulated the expression of the critical regulatory enzyme of bile acid biosynthesis, cholesterol 7-alpha-hydroxylase. In conclusion, poultry oil is a good substitution for fish oil in the diets of tiger puffer. Poultry oil could replace 100% added fish oil in the diet of tiger puffer, without adverse effects on growth and body composition.


Introduction
Fish are the main source of long chain polyunsaturated fatty acids (LC-PUFA) which are beneficial to the health of human consumers [1]. Aquaculture satisfies the growing global demand for fish but also consumes an increasing share of the world's wild fish resources via use of fishmeal and fish oil (FO) in fish feeds [2]. Therefore, increasing levels of alternative sources such as plant ingredients and livestock processing by-products are being used in fish feeds.
Poultry oil (PO) is a by-product of chicken processing, having a relatively low price and a large annual production.
The present study was aimed at comprehensively evaluating the efficacy of FO replacement with PO in an important aquaculture fish species, tiger puffer, in terms of growth, body composition, and lipid metabolism. Results of this study will be helpful to the lipid source management in the diets of tiger puffer and will also be inspiring to other farmed fish species.

Experimental Diets.
Five isonitrogenous (approximately 46% crude protein), isolipidic (approximately 10% crude lipid), and isoenergetic experimental diets were formulated. FO was used as the sole added oil in the control diet (FO-C). In other diets, FO in the control diet was replaced with PO (refined from duck skin) at different levels, namely, 25%, 50%, 75%, and 100%. The five experimental diets were designated as FO-C, 25PO, 50PO, 75PO, and 100PO, respectively. The formulation and proximate composition of the five experimental diets are presented in Table 1. Fishmeal, soybean meal, corn gluten meal, and brewer's yeast were used as the protein sources, and wheat meal was used as the binder. The diets were made with a laboratory-level single-screw pelleting machine and dried at 55°C [13]. The experimental diets were stored at -20°C prior to use. The fatty acid compositions of oils and diets are presented in Table 2.
2.2. Feeding Procedure and Sampling. Tiger puffer juveniles with an average initial body weight of 12.28 g were purchased from Hongqi Modern Fishery Industrial Park (Rizhao, China) and reared in Yellow Sea Aquaculture Co., Ltd. (Yantai, China) for the experimental use. Before the feeding trial, lower teeth of the experimental fish were cut short in order to prevent cannibalism, and the fish were temporarily raised in polyethylene tanks (2 m 3 ) with commercial feeds for 7 days to acclimate to the experimental conditions. The feeding trial was conducted in a flow-through seawater system. At the beginning of the experiment, 600 healthy fish were randomly selected and divided into 15 experimental tanks (0:7 × 0:7 × 0:4 m). Each diet was randomly fed to triplicate tanks, and each tank was stocked with 40 fish. Fish were hand-fed to apparent satiation three times daily (7 : 00, 12 : 00, and 18 : 00). The feeding trial lasted 8 weeks. Natural photoperiod was applied throughout the experiment. During the whole feeding trial, the water temperature ranged from 19 to 24°C; salinity, 28-32; pH, 7.6-7.8; dissolved oxygen > 6 mg/L; ammonia-N < 0:5 mg/L; and nitrite < 0:2 mg/L. At the end of the feeding trial, before sampling, fish were firstly fasted for 24 hours. Fish weight and survival in each tank were measured. After anesthetized with eugenol (1 eugenol: 10,000 water), 3 fish were randomly collected from each tank for the analysis of proximate composition. Four 2 Aquaculture Nutrition more fish were randomly selected from each tank, and the serum, muscle, liver, and intestine samples were collected. The blood was collected from the caudal vein and kept at room temperature for 2 h and then at 4°C for 6 h. Centrifugation (836 × g, 10 min, 4°C) was then conducted, and the straw-colored supernatants were collected as serum samples. From each fish, two pieces of dorsal muscles, two pieces of liver tip tissue, and one piece of midgut (about 1.0 cm) were collected. The samples for Real-Time quantitative Polymerase Chain Reaction (RT-qPCR) studies were immediately frozen with liquid nitrogen and then stored at -76°C before use. The tissue samples for proximate composition analysis were stored at -20°C before use. All sampling protocols, as well as all fish rearing practices, were reviewed and approved by the Animal Care and Use Committee of Yellow Sea Fisheries Research Institute.

Analysis of the Proximate Composition of Fish and Diets.
Proximate composition analysis of experimental diets and the whole body, muscle, and liver was performed according to the standard methods of Association of Official Analytical Chemists (AOAC). In brief, the moisture content was measured by drying the samples of diets and fish to a constant weight at 105°C; the protein content was assayed by measuring nitrogen content (N × 6:25) using the Kjeldahl method; the lipid content in the diet and whole body was assayed with petroleum ether extraction using the Soxhlet method (but the lipid in muscle and liver was extracted with the chloroform-methanol method), and the ash content was measured by incineration in a muffle furnace at 550°C for 8 h.  were used for the RT-qPCR. The specific primers for lipid metabolism genes and reference genes are presented in Table 3. The amplification efficiency for all primers was 95~105%, and the coefficients of linear regression were >0.99. The PCR reaction system consists of 1 μL cDNA template, 0.4 μL forward primer (10 μM), 0.4 L reverse primer (10 μM), 5 μL SYBR Green Pro Taq HS Premix II, and 3.2 μL sterilized water. The program was as follows: 95°C for 30 s followed by 40 cycles of "95°C for 5 s, 57°C for 30 s, and 72°C for 30 s." Melting curve analysis (1.85°C increment/min from 58°C to 95°C) was performed after the TFA: total fatty acid; SFA: saturated fatty acid; MUFA: mono-unsaturated fatty acid; n-6 PUFA: n-6 poly-unsaturated fatty acid; n-3 PUFA: n-3 polyunsaturated fatty acid; ND: nondetectable. Table 3: Sequence information of the primers used in this work.

Primer
Sequence (5′-3′) GenBank reference PL (bp) Lipid metabolism genes Lipogenesis Aquaculture Nutrition      (Table 3). β-Actin and EF1Α were used as the internal references. PCR amplification was performed as previously described in Section 2.5.

Statistical Analyses.
All data were analyzed with one-way ANOVA in SPSS 16.0. Prior to analysis, all data were tested for normal distribution using Shapiro-Wilk test, and the homogeneity of variance was tested with Levene's test. Multiple comparisons were performed using Tukey's test, and the significance level was decided when P < 0:05. The results were expressed as mean ± standard error.

Growth Performances, Somatic Indices, and Body
Compositions. No significant difference was observed in survival, feed efficiency, and weight gain of fish among different groups (P < 0:05, Table 4). However, the weight gain in groups 50PO, 75PO, and 100PO was slightly higher compared to other groups. Group 25PO showed a significantly lower VSI than group FO-C (P < 0:05), but no significant difference among groups was observed in other somatic indices.
The PO supplementation had marginal effects on the proximate composition of whole fish body, muscle, and liver (Table 5). Dietary PO supplementation significantly increased the moisture content of the liver (P < 0:05).

Serum Biochemical Parameters.
In general, the contents of cholesterol including TC, HDL-C, and LDL-C were decreased by the PO supplementation (Table 6). Increasing levels of dietary PO linearly decreased the serum MDA concentration (P < 0:05).

Hepatic mRNA Expression of Lipid Metabolism Genes.
The dietary PO supplementation had very little influence on the hepatic mRNA expression of most lipid metabolism genes (Table 7). Nevertheless, increasing dietary PO levels linearly upregulated the gene expression of hmgcr, and com-pared to group FO-C, group 100PO showed significantly higher gene expression of cyp7a1.
3.4. Mitochondrial DNA Copy Number. The PO supplementation did not significantly affect the relative gene expression of 16S rRNA and cytochrome B in the mitochondrial DHA of both muscle and liver (Figure 1).

Discussion
Growth performance is the most valuable indicator of fish nutritional status. The current study revealed that fish oil (FO) replacement with poultry oil (PO) had no significant effects on the weight gain and feed utilization of tiger puffer, indicating the high potential of PO as dietary lipid source. This finding was consistent with previous studies on other fish species such as rainbow trout (2/3 FO replacement) [4], Japanese seabass (50% replacement) [5], yellowtail kingfish (100% replacement) [7], largemouth bass (100% replacement) [6], barramundi (100% replacement) [8], and Florida pompano (75% replacement) [10]. However, other studies showed that the final body weight of gilthead sea bream fed PO was significantly lower than that in the FO control group [12], indicating this species may have a lower capacity to utilize PO.
When different oils were compared in tiger puffer diets, PO resulted in better fish growth than linseed oil, rapeseed oil, and beef tallow, which reduced the fish growth when replacing 100% FO [15]. Although other oils such as tiger puffer liver oil, soybean oil, and palm oil also resulted in comparable growth performance with the FO group, PO resulted in even slightly higher weight gain than the FO control group. The growth-promoting effects of PO could be due to more balanced contents of SFA and MUFA in PObased diets. These fatty acids, such as 18:1n-9, 16:1n-7, and 16 : 0, are the most preferred substrates for catabolism via β-oxidation in fish [16,17]. It has been demonstrated in many fish studies that a balanced dietary supply of SFA and MUFA limits the metabolic energy required for lipogenesis processes as well as the extent of n-3 LC-PUFA β-oxidation, which is called the "n-3 LC-PUFA sparing effect" [18][19][20][21][22][23][24][25][26][27]. In this study, the mitochondrial DNA copy number, which is indicative of basic energy supply status, in both muscle and liver of experimental tiger puffer was not significantly different between the FO and PO groups. Moreover, the hepatic expression of lipid metabolism genes was also   Apart from the growth performance, the somatic indices such as HSI, VSI, and condition factor were mildly affected by dietary PO too. Only the HSI was lowered in group 25PO compared to the FO control. Similar results were found when dietary FO was replaced by linseed oil in diets of tiger puffer [28]. The fish body composition was also very marginally affected by FO replacement with PO, similar to what observed in rainbow trout [4], European seabass [11], and gilthead sea bream [12]. In particular, very little change was observed in the proximate composition of muscle, indicating the little influence of dietary PO on fillet quality. For tiger puffer, the liver is also an edible organ. The liver moisture content of tiger puffer was significantly increased by dietary PO supplementation. Increased moisture content of liver is a favorable change for many consumers, due to the fact that tiger puffer stores lipid predominantly in the liver, leading to an already very high lipid content in the liver. Different from this study, the study on largemouth bass showed that dietary PO inclusion decreased the moisture content of fish liver [6]. This discrepancy may be related to different lipid contents in the liver of different fish species.
Regarding the hematological parameters, dietary inclusion of PO linearly reduced the concentrations of cholesterol and MDA. In general, PO contained lower cholesterol level than FO [29]. The decrease of MDA content by dietary PO could be due to the fact that PO had lower LC-PUFA con-tents than FO and consequently faced with lower peroxidation stress. Similar results were observed in other tiger puffer studies when linseed oil, soybean oil, rapeseed oil, palm oil, and beef tallow were used to replace FO [15,28]. Although dietary PO reduced the serum cholesterol content, complete FO replacement with PO increased the content of total bile acid, which is a product of cholesterol metabolism. The gene expression results showed that dietary PO upregulated the hepatic gene expression of both hmgcr, which is the rate-limiting enzyme in the synthesis of cholesterol [30,31], and cyp7a1, which functions as a critical regulatory enzyme of bile acid biosynthesis [32]. This indicates that dietary PO may stimulate the biosynthesis of cholesterol and bile acid in independent ways, similar to the simulating effects of dietary taurine observed in a recent study on tiger puffer [33]. Nevertheless, it remains unknown in what mechanisms dietary PO stimulated the biosynthesis of bile acid. This warrants further studies.
In conclusion, results of this study suggested that in terms of growth performance, poultry oil is a good potential lipid source in diets of farmed tiger puffer. Fish oil replacement with poultry oil also had marginal effects on the body composition and lipid metabolism of juvenile tiger puffer.

Data Availability
Raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher. Data in the same row not sharing the same superscript letter are significantly different (P < 0:05).