Black Soldier Fly Oil in Different Lipid Diets Could Regulate Tissue Lipid Metabolism and Fatty Acid Composition of Juvenile Mirror Carp

,


Introduction
Dietary lipid plays an important role in the growth and development of fish, as it promotes the absorption of lipophilic nutrients and provides energy and essential fatty acids [1].Mirror carp (Cyprinus carpio var.specularis), a common carp species, is widely farmed in China due to its taste, high nutritional value, and affordable price [2].It is generally considered that the optimal dietary lipid content for common carp is around 50-60 g kg −1 [3].However, with the development of the intensive breeding industry, high-lipid diets are often used in aquatic production to meet the demands of rapid growth [4].Increasing the dietary lipid level is a common strategy to promote fish growth and effectively spare dietary protein in many fish species [5,6].However, some studies indicate that high-lipid diets may lead to some negative effects, including excessive fat deposition, abnormal oxidative status, and even a decrease in nutritional value [7][8][9].
Although most of the research on black soldier fly (BSF) as an aqua-feed ingredient has focused on the use of BSF larvae meal [10], black soldier fly oil (BSFO) is gaining increasing attention [11][12][13].BSFO is interesting because it differs from most other insect and even plant oils in its fatty acid (FA) composition containing a large amount of the medium-chain fatty acids (MCFAs), for example, lauric acid (C12 : 0) typically represents 21%-50% of the total fatty acids (TFAs) [14,15].The uptake and transport of medium and short-chained FAs are quite different from the longerchained FAs commonly found in traditional lipids of the sense [16].Lauric acid may act like a direct source of energy, similar to sugar, than a lipid in the more traditional sense.A study by Li et al. [14] demonstrated that replacing soybean oil (SO) with BSFO (25 g kg −1 diet) reduced intra-peritoneal fat deposition and decreased adipocyte size in juvenile Jian carp.Xu et al. [13] also showed that BSFO (25 g kg −1 diet) had similar effects on lipid metabolism in juvenile mirror carp when compared to silkworm or yellow mealworm oils (YMOs).However, to our knowledge, little is known about the effects of BSFO on improving growth, lipid metabolism, and health status in fish fed diets with different lipid levels.The unique uptake and oxidation of lauric acid compared to other FAs commonly included in carp aqua-feeds may affect the tolerance to higher lipid levels when including BSFO.Therefore, this study was aimed to investigate the effect of high-lipid diets on juvenile mirror carp and whether BSFO could enhance the utilization of such diets, based on growth performance, FA composition, morphology observation, lipid metabolism, and related gene expressions in juvenile mirror carp.

Materials and Methods
2.1.Experimental Diets and Design.In this study, a 2 × 2 factorial design was used.Four isonitrogenous diets (320 g kg −1 crude protein) were formulated with two lipid sources (SO or BSFO) at two different lipid levels (60 or 90 g kg −1 crude lipid).Two diets were prepared with lipid levels of 60 and 90 g kg −1 , using 25 g kg −1 SO as the base oil source.These diets were named the control (CT) group and high lipid (HL) group, respectively.Additionally, the diets in the CT and HL groups, where the SO was replaced by 25 g kg −1 BSFO, were named CT + BSFO and HL + BSFO, respectively.BSF fed by wheat bran was obtained from Ankang Fisheries Experimental and Demonstration Station of Northwest A&F University (Ankang, Shaanxi Province, China).BSFO was obtained with an oil press machine at a temperature ranging between 80 and 90°C and a pressure between 5 and 20 KP (Dongdubao Electrical Technology Limited Company, Yueqing City, Zhejiang Province, China).The other feed ingredients were provided by Huaqin feed factory (Yangling, Shaanxi Province, China).The experimental diets were formulated with the 2.5 mm pellet diameter and collected after drying in a cool and well-ventilated place at room temperature for 24 hr in Ankang Fisheries Experimental and Demonstration Station of Northwest A&F University.Then, the diets were stored at −20°C until use.

Feeding Trial.
Juvenile mirror carp were provided by Ankang Fisheries Experimental and Demonstration Station of Northwest A&F University.Fish were acclimated to the indoor rearing conditions for 2 weeks while being fed commercial diets (Huaqin feed factory, Yangling, Shaanxi Province, China) three times a day.After the acclimation period, fish (initial mean body weight, 6.36 AE 0.18 g) were randomly distributed into 12 circular 215-L tanks at a density of 12 fish per tank (n = 3 per diet).The tanks were supplied with recycling water and aeration to maintain enough dissolved oxygen.All rearing tanks were maintained under fluorescent lamps (12-hr light/12-hr dark).The water temperature during the trial ranged between 25 and 28°C.Each experimental diet was randomly assigned to triplicate tanks, which was fed to apparent visual satiation (three times a day, at 8:30, 12:30, and 16:30) for 8 weeks.At the initial feeding, the juvenile mirror carp run upstream to water surface to feed aggressively.After a period, the feeding intensity decreased and when all fish settled on the bottom and did not come up to feed, this state was considered as the apparent satiation.The feeding process need ensure there was no feed residue during the entire feeding process.The dissolved oxygen, pH, and ammonia content maintained at 11.6 AE 0.20 mg L −1 , 7.28 AE 0.20, and 0.04 AE 0.01 mg L −1 , respectively.

Sampling.
Prior to the start of the feeding trial, six fish were sampled and stored at −20°C for proximate composition analysis.After 8 weeks, fish were starved for 24 hr prior to sampling, and then anesthetised with 0.1 g L −1 MS-222 (Sigma, St. Louis, MO, USA).Final body weight (FBW), body length, total length, and fish number of all fish were measured to calculate growth parameters and survival.
Blood samples (six fish per tank) were collected from the caudal vein using syringes (Zhengzhou Kangjia Medical Equipment Company, Zhengzhou Province, China).The serum was separated by centrifugation (825 g for 10 min) after 6 hr at 4°C, pooled for each tank, and frozen at −80°C.The hepatopancreas and intra-peritoneal fat (three fish per tank) were separated and stored at −80°C for quantitative polymerase chain reaction (qPCR) analyses.The hepatopancreas, intra-peritoneal fat, and muscle samples were sampled and stored at −20°C for the analysis of proximate composition (three fish per tank) and FA composition (two fish per tank).For the analysis of proximate composition, three whole fish per tank were pooled, homogenized, and stored at −20°C.Pieces measuring 0.5-1.0cm of intraperitoneal fat and hepatopancreas (one fish per tank) were cut and fixed in 4% formaldehyde in 1x-phosphate-buffered solution for histological analyses.

Proximate Composition of Diets and Fish Samples.
The proximate composition of diets and tissues was analyzed by the methods recommended by the Association of Official Analytical Chemists (AOAC, 2000).Crude protein was measured by Kjeldahl method.The crude lipid and moisture were measured by ethyl ether extraction and 105°C drying to constant weight.The ash content was measured at 550°C for 4 hr with a muffle furnace (TMF-3100, EYELA Co., Tokyo, Japan).The ingredients and proximate composition of the diets are presented in Table 1.
The FA compositions (diets and tissues) were determined by gas chromatography (Agilent Technologies, Santa Clara, CA, USA) as described previously by Tian et al [17].The results are presented as the percentage of TFA.The FA composition of the diets is presented in Table 2.
2.6.Histological Observations of Hepatopancreas and Intra-Peritoneal Fat Tissue.The fixed hepatopancreas and intraperitoneal fat samples were rinsed in running water for 24 hr.Subsequently, the samples were then dehydrated in ethanol, infiltrated in xylene, and embedded in paraffin, following standard histological procedures [18].The samples were then cut serially (5 μm thick) using a rotary microtome and stained with hematoxylin and eosin.One cross-sectional slice was prepared from each tissue sample and placed on a microscopy slide.The adipose tissue was observed and photographed using an inverted microscope (OPTEC, China).Five non-overlapping areas per piece of tissue were selected and calculated by TS View 7 (Softim aging solutions, Tucsen, China).
In each area, five measurements were taken [19], resulting in a total of 25 measurements per fish.The hepatopancreas was photographed using a light microscope (Eclipse 50i, Nikon, Tokyo) and a camera (Digital Sight DS2MV with control unit DS-L2, Nikon, Tokyo).due to dietary treatments were evaluated by using two-way ANOVA (variables: lipid sources and inclusion levels) followed by Duncan's post-hoc test.For data generated from histological analyses, which had several measurements per fish, a nested analysis was performed using nesting factors area, fish, and tank [21] (R core team., 2019).Result was considered significant at P <0:05.The graphs were created using Prism 7 (Graph Pad Software Inc., San Diego, USA).
Pearson correlation coefficient (PCC) was used to quantify the correlation between the tissue FA composition and dietary FA by using R value.The R value was computed using the equation: R = log 2 (Pt/Pd).R is the degree of variation between sampled tissue FA content and diet FA, Pt is the percentage of one FA in a specific tissue, and Pd is the percentage of the FA in diets.
Feed intake FI; g fish Feed conversion ratio ðFCRÞ ¼Feed intake=ðFinal weight of fish -initial weight of fishÞ; 3. Results

Growth Performance and Biological
Parameters.At the end of the feeding trial, the HL group displayed significantly higher FBW, WGR, SGR, PER, PRR, and lower FCR compared to the CT group (P <0:05).Both the inclusion of lipid level (HL) and its source (HL + BSFO) decreased the LRR compared to fish fed CT or CT + BSFO (P <0:05).IFI was significantly influenced by lipid level and lipid source (P <0:05).The FI, VSI, HSI, and CF in all groups showed no significant differences.The growth performance and biological parameters were not influenced by the interaction of lipid level and BSFO (Table 4).

Tissue Proximate Composition.
The crude protein content of the whole body of fish fed the HL diets was significantly decreased while the crude lipid content was significantly increased compared to fish fed CT diets (P <0:05).The moisture and ash content of the whole body did not show significant differences between all groups (P <0:05).There were no significant differences in dietary effects on the moisture, crude protein, lipid, or ash in muscle.In the whole body, the protein content was lower in the HL fed fish compared to the two lowfat diets (P <0:05).The whole body lipid was higher in the two HL diets compared to the low-lipid diets (P <0:05).In the Notes: FAS = Fatty acid synthase; PPAR-γ = Peroxisome proliferator-activated receptor-γ; PPAR-α = Peroxisome proliferator-activated receptor-α; CPT-1 = Carnitine palmitoyltransferase−1.

4
Aquaculture Nutrition hepatopancreas, only the lipid content was affected by diets, being lowered by BSFO in the low-fat diets (P <0:05), but not in the HL diets, which were both higher than the low-fat diets (Table 5).
3.3.Tissue FA Composition.From Tables 6, 7, and 8, we observe that the use of BSFO had a strong effect on the FA composition in the analyzed tissues (muscle, hepatopancreas, and intra-peritoneal fat).In particular, the use of BSFO resulted in the increased levels of the C12 : 0 and C14 : 0. Other effects were primarily attributed to the reduction or removal of SO, leading to a significant decrease in n-6 FA content, across all analyzed tissues in the groups that contained BSFO (P <0:05).The C20 : 5n-3 (eicosapentaenoic acid) remained unaffected in all tissues, but the C22 : 6n-3 (docosahexaenoic acid, DHA) levels were higher in both groups that included BSFO.The low-fat groups exhibited the highest percentages of DHA (P <0:05).Aquaculture Nutrition  6 Aquaculture Nutrition intra-peritoneal fat) and diets.To quantify this relationship, we calculated the PCC of FA between the different tissues and diets, as displayed in Table 9.Among the three tissues, the CT group showed the weakest correlation with the tissues, while HL + BSFO displayed the strongest correlation.The PCC varied across the tissues.The results also showed that the hepatopancreas showed the weaker correlation with diets, followed by muscle; however, the intra-peritoneal fat was highest (P <0:05).

Serum Biochemical Health Markers. The serum T-cho and
TG content was significantly increased with higher dietary lipid levels (P <0:05).However, the levels of serum ALT, AST, TP,  ALB, GLO, GLU, LDL, and HDL content were not affected by the lipid levels, BSFO, or their interactions (Table 10).
3.7.Tissue Morphology.Figures 2 and 3 demonstrated that the hepatocytes in the CT and CT + BSFO groups showed a more rounded shape, centrally located nuclei, and clearer cell boundaries compared to the two high-fat groups.The presence of significant vacuoles in the HL fed fish indicated a greater lipid accumulation in the hepatopancreas with the higher dietary lipid levels.Moreover, the adipocytes were larger in the HL group than those in the CT group (P <0:05).The adipocyte sizes were significantly reduced in the low-fat diet with BSFO group (P <0:05).However, the same effect of BSFO was not observed in the HL diets.There was no influence from the interactions between lipid levels and BSFO.

Expression of Lipid
Metabolism-Related Genes.The mRNA expression of FAS, ACC, and PPAR-γ in the intra-peritoneal fat tissue was significantly up-regulated in the HL fed fish compared to the diets fed low-lipid diet-fed fish (P <0:05), but not affected by BSFO.However, the mRNA expression of CPT-1 in the same fat tissue was significantly up-regulated in the low-lipid diets with BSFO.On the contrary, the hepatopancreas mRNA expressions of FAS, ACC, and CPT-1 were significantly down-regulated in the HL fed fish compared to in the fish fed low-lipid diets (P <0:05), but also not influenced by BSFO.
The hepatic mRNA expression of PPAR-γ was not affected by the diets.Only the mRNA expression of PPAR-α had a similar expression pattern in both tissues, being significantly down-regulated by the HL diets (P <0:05) and significantly up-regulated by BSFO in both the intra-peritoneal fat and hepatopancreas tissue (P <0:05) (Figures 4 and 5).

Discussion
By increasing the dietary lipid levels, the FBW, WGR, SGR, PER, and PRR were significantly increased, while FCR significantly decreased.These findings indicated a significant promotion of growth and protein-sparing effects through increased lipid levels in the diets.Wu et al. [22] found these similar results when using the 9% as the same lipid level for common carp.A lower LRR was observed in the HL groups, indicating that a substantial portion of these added lipids was utilized for energy supply,

8
Aquaculture Nutrition thus promoting the protein-sparing effect seen.This has also been previously observed in grass carp (Ctenopharyngodon idellus) and some other fish species when fish fed with highlipid diets [23,24].Feeding high-lipid diets has been shown to increase VSI in several fish species, such as in grass carp [25], silver sillago (Sillago sihama) [23], cobia (Rachycentron canadum) [26], and Japanese seabass (Lateolabrax japonicus) [27].However, in our study, we observed only a non-significant tendency toward an increased VSI in HL-fed group.Some studies have reported no increase in VSI with different lipid levels in diets.For instance, Li et al. [28] demonstrated that the VSI and HSI of large-mouth bass (Micropterus salmoides) were not influenced by higher dietary lipid levels between 10% and 12.5%.Similarly, a study on common carp found that lipid levels of 6%-9% in diets did not significantly influence VSI, CF, and HSI [22].Our results suggested that the viscera did not make significant contributions to the weight gain in the HL groups, indicating potential benefits for fish health.A significant decrease in the IFI was observed in the fish fed diets with BSFO at both lipid levels.Xu et al. [13] also found a similar results in juvenile mirror carp fed a mixture of insect oils (containing 6.33 g kg −1 BSFO) compared to fish fed with silkworm pupa oil or YMO.This study indicated that lauric HL + BSFO HL CT + BSFO CT FIGURE 2: Effects of BSFO and lipid level on hepatopancreas morphology of juvenile mirror carp (20x) (n = 3).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table 1.   1.
Aquaculture Nutrition acid, as a MCFA, is known to be easily oxidized and can reduce lipid storage in fish [29][30][31].The dietary MCFAs (3.24% of TFAs) in the mixture of insect oils could therefore enhance the growth of mirror carp without increasing lipid deposition.In our current trial, the BSFO diets contained 7.36% and 6.21% lauric acid of TFAs in the low-and highlipid diets, respectively.Differently, the BSFO in the HL diets in our study did not cause lipid deposition when compared with the HL group.Upon comparison, we found that the HL + BSFO diet had a high content of C18 : 2n-6 (35.63% of TFAs, 9% lipid level).While the MIXO (YMO: BSFO: silkworm pupa oil = 1 : 1 : 1) diet used by Xu et al. [13] had a lower content of C18 : 2n-6 (30.5% of TFAs, 6% lipid level), the YMO diet had a high content of C18 : 2n-6 (33.9% of TFAs, 6% lipid level) which also lead to an increase in IFI.So, we speculated that the negative effect of high content C18 : 2n-6 maybe stronger, thereby resulting in the absence of a decrease in IFI despite the inclusion of BSFO in the diets.
The measured adipocyte sizes were larger in the HL groups than those in the fish fed the regular lipid level.Adipocyte sizes were also significantly smaller in the CT + BSFO group than that in the CT group.There was also a tendency for a smaller adipocyte size in the HL + BSFO group than in the HL group, but this was not significant.Moreover, the gene expression of FAS, ACC, and PPAR-γ in intra-peritoneal fat was significantly up-regulated, and the gene expression of PPAR-α was significantly down-regulated in the HL groups compared to the CT group.The gene expression results were consistent with the change in IFI.This indicates that high dietary lipid levels may lead to an increase in abdominal adipose tissue in mirror carp, which may mainly through the promotion of lipid synthesis and the suppression of lipid 10 Aquaculture Nutrition lipolysis.Similar research has been found in grass carp and Nile tilapia (Oreochromis niloticus) when fish are fed with high-lipid diets [24,32].It may also be the result of an adaptive regulatory mechanism of lipid metabolism in the abdominal adipose tissue of carp in response to the intake of a high-fat diet [33].Subsequently, the gene expression of PPAR-α and CPT-1 was significantly increased in CT + BSFO compared with that in the CT group.Similar results have been found in mirror carp fed with 25 g kg −1 BSFO, which is BSF fed by algae residue [34].Our research indicated that BSFO could increase the lipid lipolysis of fish when fed with 6% lipid diets.However, 25 g kg −1 BSFO in high-lipid diets could not activate the gene expression of lipid lipolysis, although it showed an increasing tendency.
The study of cobia indicated that the increasing of whole body lipid content is generally correlated with dietary lipid levels [26].In our study, the whole body lipid content also increased from 7.58 to 7.71 g kg −1 when the dietary lipid content changed from 6% to 9%.Similar results have been found in other fish species, including grass carp [24], European sea bass (Dicentrarchus labrax) [35], large-mouth bass [36], and cobia (R. canadum) [26].Additionally, the crude protein content was significantly decreased when fish were fed HL diets in our study.The crude lipid content in the hepatopancreas was significantly increased, but the crude lipid content in muscle was not influenced by the dietary lipid level.This indicates that lipid accumulation in fish is tissue-biased, and lipids seem to be mainly deposited in the viscera.Similar studies on fish have indicated that a high-lipid diet increases the abdominal fat index and hepatopancreas fat content, as observed in Russian sturgeon (Acipenser gueldenstaedtii) [37], grass carp [24,25], and rockfish (Sebastes schlegelii) [38].It is well-known that greater fat deposition occurring in the viscera (led by high-energy diets) can reduce the commercial value of the product [39].Therefore, this aspect must be considered when feeding fish with a high-lipid diet.In our study, the crude lipid content in the hepatopancreas was significantly decreased with the use of BSFO, regardless of the lipid diets were fed to the fish.In a previous study by our team, the author suggested that when fish are fed diets containing BSFO, there may be a faster β-oxidation of the dietary lipids or a more efficient transport out of the hepatopancreas to the tissues, resulting in decreased lipid content of the hepatopancreas [13].The histomorphological observation of the hepatopancreas supports this finding.Hepatocytes in CT and CT + BSFO groups showed rounded shapes with centrally located nuclei and clear cell boundaries in CT and CT + BSFO groups compared with the other two groups.The significant presence of vacuoles in the HL group indicates a large lipid accumulation in the hepatopancreas with the increasing lipid levels.The area of vacuoles was significantly smaller in HL + BSFO than that in the HL group.It indicated that with the use of BSFO it could effectively decrease the hepatopancreas lipid content without decreasing FBW, thereby improving the commercial value of the product.Subsequently, the results of gene expression relating to the metabolism of lipids in the hepatopancreas were further examined.Our results showed that the gene expression of lipid biosynthesis (FAS and ACC) and lipid catabolism (CPT-1 and PPAR-α) was significantly down-regulated in fish fed with HL diets.Similar studies about high-lipid diets suppressing lipogenesis or lipid catabolism have been observed in grass carp [24,40], black sea bream (Sparus macrocephalus) [41], and blunt snout bream (Megalobrama amblycephala) [42].In high-lipid diets, this effect may be due to the high content of C18 : 2n-6 in SO, which changes in the mitochondrial membrane composition, therefore decreasing the lipid catabolism gene expression [43].The down-regulation in lipid catabolism gene expression may lead to the increased hepatopancreas lipid content.The gene expression of PPAR-α in the CT + BSFO and HL + BSFO groups was significantly down-regulated compared with that in the CT and HL groups.It indicated that with the use of BSFO could promote the lipid metabolism of fish and help to decrease the hepatopancreas lipid content.Some researcheres [44] found in the study on freshwater Atlantic salmon (Salmo salar) that fish fed with BSF diets could decrease the hepatopancreas lipid content, accompanied by the down-regulation expression of CPT-1 and PPAR-γ.Moreover, some research on mammals indicated that mice fed with high-lipid diets containing MCFA could up-regulate the gene expression of PPAR-α when compared with a high-lipid control group [45].Coconut oil, which is also a kind of MCFA oil, could activate PPAR-α in the hepatopancreas in the trained state of mice when compared to those fed with SO [46].
Based on the correlation analysis between dietary FA composition and different tissues, we observed that the tissues in the HL + BSFO group had the highest correlation with diets, followed by the HL group, CT + BSFO group, and CT group.In low-lipid diets, the proportion of lipid in low-lipid diet was higher to maintain the body's normal needs, resulting in a lower correlation with tissues.The high-lipid diet contains a higher proportion of lipids used for deposition after absorption, hence maintaining a high correlation with tissues [47].We also found that with the use of BSFO, the correlation of diets to tissues was improved

12
Aquaculture Nutrition when compared with the CT and HL groups, respectively.We speculated that this improvement may be due to the function of MCFA.The role of MCFA has been widely investigated in both livestock and human nutrition due to their rapid absorption and oxidation [48].MCFAs are more polar than other lipids and are less reliant on chylomicrons and lipoproteins for transport.Since they can be transported without the need for these systems, MCFAs can be absorbed (intact) into intestinal epithelial enterocytes and then hydrolyzed in cells by microsomal lipases [49].Diets containing BSFO could enhance the lipid utilization, leading to more FAs being deposited in tissues, ultimately improving the correlation of FA composition between tissues and diets.
On the flip side, we found that the intra-peritoneal fat tissue had the highest correlation with diets, followed by muscle and hepatopancreas.Tian et al. [50] also reported these similar results in Songpu mirror carp.Intra-peritoneal fat is the primary tissue that accumulates lipid or serves as an energy storage depot [51].FAs from the diet are deposited here as TG or membrane phospholipids after absorption and transportation [52], which may explain the reason for its high correlation with FA composition in the diet.Hepatopancreas is an important intermediary metabolic tissue with a higher capacity for FA metabolism, such as desaturation/elongation and β-oxidation [53].The FA composition of the hepatopancreas can be self-regulated and resulting in a lower correlation [54].
The absolute values of R represent the extent of variation in FAs [50].Considering the influence of dietary treatments, the variation trends of R were similar within a particular tissue, indicating the similar metabolism patterns of the FAs in the same tissue.There was clear tissue-specificity regarding FA composition.The C18 : 2n-6 was considered as the essential FA for carp [52].It should be noticed that the proportions of C18 : 2n-6 in the hepatopancreas (14.89%-25.03%)were relatively lower than those in IPF and muscle.The IPF showed the highest content (25.62%-35.26%).The intra-peritoneal fat showed the lowest C20 : 4n-6 (0.28%-0.70%) and DHA (0.50%-0.62%) contents among the tissues, and the highest C18 : 2n-6 contents, suggesting that the intra-peritoneal fat of mirror carp might prefer to retain C18 polyunsaturated fatty acids (PUFAs) but not HUFAs compared to other tissues, similar to our earlier studies in grass carp [54] and Songpu mirror carp [50].This indicated a tissue-specific manner of deposition for the retention of these FAs.FAs can affect the flavor and nutritional properties of muscles, and DHA deposition in muscles, as an edible part, helps to meet human nutritional needs.Our results indicated that with the use of BSFO, the DHA content was significantly increased in CT + BSFO and HL + BSFO groups.Our results align with the research of Li et al. [14], which found that feeding Jian carp with diets containing 25 g kg −1 could improve the DHA content in muscle.

Conclusion
This study indicated that a high-lipid level resulted in an increase in feed utilization, protein utilization, and growth performance of juvenile mirror carp.Diets with 25 g kg −1 BSFO contained in 60 g kg −1 lipid level diets could decrease the IFI, adipocyte size, and the hepatopancreas lipid content, while also up-regulating of lipolysis-related gene expression without negatively impacting growth performance.However, the effectiveness of lipid metabolism for BSFO contained in 90 g kg −1 lipid level diets was not as pronounced as in 60 g kg −1 lipid level diets.Moreover, diets containing BSFO could improve the content of total n-3 PUFA, while reducing the content of total n-6 PUFA in the muscle.

2. 9 .
Calculations of Growth and Nutritional Indices.Weight gain rate ðWGRÞ ¼ ðFinal weight of fish -initial weight of fishÞ Initial weight of fish ; ð1Þ Specific growth rate ðSGRÞ ¼ ½ln ðFinal weight of fishÞ -ln ðinitial weight of fishÞ Days of feeding × 100%;

7 FIGURE 1 :
FIGURE 1: The relationship between diet and FA composition of fish tissue using the R values.X-axis represented the FAs, and Y-axis represented the R values.Values were group means, with standard deviations represented by vertical bars (n = 3).

FIGURE 3 :
FIGURE 3: Effects of BSFO and lipid level on intra-peritoneal fat morphology of juvenile mirror carp (20x).(a) Morphological observation of adipocytes.(b) The statistics of adipocytes size; columns presented as means AE S.D. (n = 3).Bars bearing with different letters and symbols were significantly different among treatments (P <0:05).

FIGURE 4 :
FIGURE 4: Effects of BSFO and lipid level on relative gene expression in intra-peritoneal fat of juvenile mirror carp; columns presented as means AE S.D. (n = 3); bars bearing with different letters and symbols were significantly different among treatments (P <0:05).

FIGURE 5 :
FIGURE 5: Effects of BSFO and lipid level on relative gene expression in hepatopancreas of juvenile mirror carp; columns presented as means AE S.D. (n = 3); bars bearing with different symbols were significantly different among treatments (P <0:05).

TABLE 1 :
Ingredients and proximate composition of the experimental diets.
−1and only using SO as oil source; the CT + BSFO presented diet containing lipid level 60 g kg −1 and only using BSFO as oil source; the HL presented diet containing lipid level 90 g kg −1 and only using SO as oil source; the HL + BSFO presented diet containing lipid level 90 g kg −1 and accompanied by BSFO used as part of oil source.

TABLE 3 :
Primers used for qPCR analysis.

TABLE 4 :
Effects of BSFO and lipid level on growth performance and biological parameters of juvenile mirror carp (n = 3).
Notes: Values are means AE S.D. Values in the same line with different superscript letters are significantly different (P <0:05).IBW = Initial body weight; FBW = Final body weight; WGR = Weight gain rate; SGR = Special growth rate; FCR = Feed conversion ratio; FI = Feeding intake; PER = Protein retention ratio; PRR = Protein retention ratio; LRR = Lipid retention ratio; VSI = Viscera index; HSI = Hepatosomatic index; IFI = Intra-peritoneal fat index; CF = Conditional factor.The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1.

TABLE 5 :
Values are means AE S.D. Values in the same line with different superscript letters are significantly different (P <0:05).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1.
Effects of BSFO and lipid level on body composition of juvenile mirror carp (% wet weight, n = 3).Notes:

TABLE 6 :
Effects of BSFO and lipid level on muscle FA composition of juvenile mirror carp (% TFA, n = 3).
Notes: Values are means AE S.D. Values in the same line with different superscript letters are significantly different (P <0:05).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1.

TABLE 7 :
Effects of BSFO and lipid level on hepatopancreas FA composition of juvenile mirror carp (% TFA, n = 3).
Notes: Values are means AE S.D. Values in the same line with different superscript letters are significantly different (P <0:05).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1.

TABLE 8 :
Effects of BSFO and lipid level on intra-peritoneal fat FA composition of juvenile mirror carp (% TFA, n = 3).
Notes: Values are means AE S.D. Values in the same line with different superscript letters are significantly different (P <0:05).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1.

TABLE 9 :
PCC analysis of the relationship between the tissue and diet FA composition.Values are means AE S.D. Values in the same line with different lowercase letters are significantly different (P <0:05).Values in the same column with different capital letters are significantly different (P <0:05).The descriptions of CT, CT + BSFO, HL, and HL + BSFO were the same as Table1. Notes:

TABLE 10 :
Effects of BSFO and lipid level on serum indexes of juvenile mirror carp (n = 3).