Effects of Dietary Lysine Level on Growth Performance and Protein Metabolism in Juvenile Leopard Coral Grouper (Plectropomus leopardus)

An 8-week feeding trial was conducted to evaluate the effects of dietary lysine level on growth performance and protein metabolism of juvenile leopard coral grouper (Plectropomus leopardus) and thereby obtained the optimal dietary lysine requirement of P. leopardus. Six isoproteic and isolipidic experimental diets were formulated to contain 1.10%, 1.69%, 2.30%, 3.08%, 3.56%, and 4.36% lysine of diets, respectively. Each diet was assigned at random to triplicate groups of 25 juveniles (initial mean weight is 10.57 g) per tank in a flow-through mariculture system maintained at 27–30°C. Dietary inclusion of 2.30–3.08% lysine improved the weight gain rate (WGR) and specific growth rate and decreased the feed conversion ratio (FCR) of juveniles (P < 0.05). The intestinal digestive enzyme (trypsin, amylase, and lipase) activities were overall enhanced by dietary inclusion of 3.08–3.56% lysine (P < 0.05). The mammalian target of rapamycin (mTOR) signaling pathway was activated in fish fed diets with 1.69–2.30% lysine by upregulating the relative expression levels of hepatic TOR and S6K1 (p70 ribosomal protein S6 kinase 1) but downregulating the relative expression level of hepatic 4E-BP2 (eIF4E-binding protein 2). Conversely, the amino acid response signaling pathway was inhibited in fish fed diet with 2.30% lysine by downregulating the relative expression levels of hepatic GCN2 (general control nondepressible 2), ATF3 (activating transcription factor 3), ATF4a (activating transcription factor 4a), and ATF4b (activating transcription factor 4b). Additionally, dietary 1.69–3.08% lysine enhanced the plasma total protein level and hepatic lysine α-ketoglutarate reductase activity but depressed the blood urea nitrogen level and hepatic adenosine monophosphate deaminase activity (P < 0.05). Moreover, dietary 3.08% lysine increased the contents of whole-body crude protein and total amino acids, while 1.69%–4.36% lysine depressed the whole-body lipid content (P < 0.05). These results indicated that optimal dietary lysine increased the digestive enzyme activities, promoted protein synthesis but depressed protein degradation, and thereby improved the growth performance of P. leopardus. Based on the second-order polynomial model, the optimal lysine requirement of juvenile P. leopardus for WGR, FCR, and lysine deposition was 2.60%–2.97% of diets (4.91%–5.60% of dietary protein).


Introduction
Leopard coral grouper Plectropomus leopardus (Lacepède, 1802), a top carnivore in many coral reef communities, is mainly distributed from the Western Pacific to East Africa and the Red Sea [1]. Due to its good flavor and bright body color, P. leopardus is popular and marketable in China. The decline of wild resources and relatively slow growth rate have led to the increasing price of P. leopardus. Thus, this species has become one of the vital breeding varieties of grouper with a broad market prospect [2]. However, there is no specialized compound feed for P. leopardus due to the limited information on nutritional requirements [3]. So far, research on the requirement of essential amino acids of P. leopardus has not been reported.
Lysine is an alkaline amino acid with little content in cereals. It is an essential amino acid for fish and one of the most important limiting amino acids for most plant ingredients [4]. Decreased feeding, stunted growth, and reduced feed efficiency are the most responsive and apparent symptoms of lysine deficiency in fish. However, excess lysine also has negative effects on growth rate and feed utilization of fish [5,6] as it disrupts the balance of amino acids in diets and prevents the absorption and utilization of other amino acids [7]. Consequently, it is essential to determine the dietary lysine requirement of fish. To date, there is no research about the requirement of lysine for juvenile P. leopardus. Therefore, the current study was designed to assess the effects of dietary lysine level on the growth performance and protein metabolism of juvenile P. leopardus and thereby obtained the optimum dietary lysine requirement of P. leopardus.

Experimental Diets.
Six isoproteic (53% crude protein) and isolipidic (12% crude lipid) experimental diets were formulated to contain 1.10%, 1.69%, 2.30%, 3.08%, 3.56%, and 4.36% lysine of diet at the expense of alanine, respectively ( Table 1). The compositions of the essential amino acid (except for lysine) in the experimental diets were adjusted to the reference amino acid profile established by the previous study [8]. The proximate and amino acid compositions of the experimental diets are present in Tables 1 and 2, respectively.
All separately crushed ingredients (except for fish oil and soybean oil) were sieved through a sixty-mesh sieve and mixed thoroughly. After fish oil and soybean oil (soybean lecithin was predissolved in soybean oil) were added, all ingredients were remixed by a V-type vertical mixer (VI-1000L; Jiake Machinery Manufacturing Co., Ltd., Jiangyin, China). A suitable amount of distilled water (about 20-30%) was added to form a dough, and the dough was processed into 2 mm diameter pellets by a twin-screw extruder (F-26; South China University of Technology, Guangzhou, China). All diets were air-dried at 25°C for 24 h and stored at −20°C until used.
2.3. Feeding Management. Juvenile P. leopardus were obtained from a local commercial hatchery. After two weeks of cultivation in 1 m 3 tank, healthy and uniform juveniles (initial mean weight 10:57 ± 0:02 g) were randomly divided into 18 tanks (three replicates per group) with 25 fish per tank (0.3 m 3 ). All fish were fed twice daily (8:00 and 16:00) to apparent satiation (about 3% of body weight) and kept under natural photoperiod and continuous aeration. During the experimental period, the water temperature was maintained at 27−30°C, pH 7.7−8.0, ammonia nitrogen < 0:2 mg/L, and dissolved oxygen > 5:0 mg/L. The feeding trial lasted for 8 weeks.
2.4. Sample Collection. At the end of the feeding trial, all fish were fasted for 24 hours and anesthetized with MS-222 (Sigma-Aldrich Co., St. Louis, MO, USA). The final numbers and biomass of fish per tank were measured. Three fish per tank were randomly selected and stored at -20°C for analysis of body composition. Another four fish per tank were collected to obtain the weight of the whole body, viscera, liver, and intestine. Blood samples were taken from the caudal vein of six fish per tank with a heparinized syringe and then collected into a heparinized tube; the plasma was collected and stored at -80°C. Four fish per tank were aseptically sacrificed in an ice bath; then, the liver, midgut, and dorsal muscle samples were rapidly removed and frozen in liquid nitrogen and then stored at -80°C until used.

Analysis
2.5.1. Proximate Composition. Proximate composition of the experimental diets and whole-body samples was performed by the following AOAC method: moisture was dried in an oven at 105°C to constant weight; crude protein (nitrogen × 6:25) was determined by the regular Kjeldahl method using Dumas nitrogen analyzer (Primacs SN100; Skalar, Netherlands); crude lipid was reflux extracted with petroleum ether and calculated as the difference between weight of sample and residue; crude ash was incinerated at 550°C in a muffle furnace (SX-410; Beijing Ever Bright Medical Treatment Instrument Co., Ltd., Beijing, China) for 16 h. The amino acid profiles were measured using an automatic amino acid analyzer (Hitachi L8900; Hitachi, Tokyo, Japan) after acid hydrolysis using 6 N HCl at 110°C for 22 h.

Digestive Enzyme Activities.
To obtain an adequate crude enzyme extract solution, the amount of physiological saline solution (0.9% NaCl) added to the wet midgut was determined by a preliminary study. The wet midgut plus fourfold volume (v/w) of ice-cold physiological saline solution was added to a 10 mL test tube and homogenized using an IKA homogenizer (IKA Works Asia, Bhd, Malaysia). The homogenate was centrifuged at 9 000 g for 30 min at 4°C using a high-speed refrigerated centrifuge (MX-160; Tomy Seiko Co., Ltd., Tokyo, Japan). The supernatant was diluted with physiological saline solution in a certain dilution ratio and used as a crude enzyme solution. The activities of intestinal trypsin, lipase, and amylase were detected by the UV colorimetric method, microplate method, and iodine-starch colorimetry using commercial kits (Shanghai Enzymelinked Biotechnology Co., Ltd., Shanghai, China) and expressed as enzyme activity per gram/microgram protein, respectively. The soluble protein content was determined by the Bradford method using commercial kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Using β-actin as internal reference, real-time quantitative PCR was performed using SYBR® Green Pro Taq HS Premix II qPCR kit and quantified on the LightCycler 480 (Roche Applied Science) using the following program: denaturation at 95°C for 30 s, followed by 40 cycles at 95°C of 5 s and 60°C for 30 s. The reaction system was 10 μL: 2x SYBR® Green Premix Pro Taq HS Kit II, 5 μL; 0.4 μL for positive and negative primers, respectively; cDNA, 1 μL; and RNase Free dH 2 O, 3.2 μL. Primers of target genes used for qPCR are shown in Table 3. The relative gene expressions were calculated by the method of 2 -ΔΔCt [9]. 1 Supplied by the Shaanxi Baichuan Biotechnology Co., Ltd., Xi'an, China. 2 Amino acid mixture included 1.02% DL-methionine, 1.60% L-threonine, 2.52% Larginine, 1.66% L-isoleucine, 1.97% L-leucine, 1.47% L-valine, 0.59% L-histidine, 1.18% L-phenylalanine, 0.25% L-cystine, 1.83% L-glycine, 1.03% L-serine, 1.53% L-alanine, 4.10% L-aspartate, 3.90% L-glutamate, and 1.07% L-tyrosine. 3 Supplied by the Qingdao Master Bio-Tech Co., Ltd., Qingdao, China. Vitamin premix (g/kg of mixture) included 1.55 g retinyl acetate (2 800 000 IU/g), 0.03 g cholecalciferol, 40 g DL-α-tocopheryl acetate, 8 g menadione, 6 g thiamine hydrochloride, 9 g riboflavin, 7 g pyridoxine hydrochloride, 0.05 g vitamin B 12 , 50 g ascorbic acid, 30 g calcium D-pantothenate, 45 g niacin, 2.5 g folic acid, 0.1 g D-biotin, and 100 g inositol.    Aquaculture Nutrition 2.6. Calculations and Statistical Analysis. The parameters were calculated using the following formulas: Hepatosomatic index HSI, % ð Þ= 100 × hepatic weight g ð Þ whole body weight g ð Þ , where W i , W f , and d stand for initial body weight (g), final body weight (g), and feeding day (d), respectively. The Leaven and Kolmogorov-Smirnov tests were used to confirm the homoscedasticity assumptions and normality, respectively. All data (means ± SEM) were analyzed by a one-way analysis of variance (ANOVA) followed by Duncan's test to compare the means between individual treatments. The significant difference was set at the level of P < 0:05. For detecting potential linear or quadratic effects of dietary lysine level, all data were also subjected to polynomial orthogonal contrasts analysis. Statistical analysis was performed using SPSS 21.0 (SPSS Inc., Chicago, IL, USA) for Windows.

Growth Performance and Body
Indexes. The final body weight, WGR, and SGR were improved with the increasing dietary lysine level up to 2.30% and depressed thereafter (Table 4). Conversely, the FCR depressed with the increasing dietary lysine level up to 2.30% and increased afterward. Interestingly, CF enhanced linearly with the increasing dietary lysine level, which was significantly higher in the 3.56% and 4.36% lysine groups compared to the 1.10% lysine group (P < 0:05). The VSI was significantly higher in the 2.30%, 3.08%, and 3.56% lysine groups compared to the 1.69% lysine group (P < 0:05). However, the feed intake, PER, HSI, and ISI were not affected by dietary lysine level (P > 0:05).
Based on the second-order polynomial (SOP) model, the optimal lysine requirement for final body weight, WGR, SGR, and FCR of juvenile P. leopardus was estimated to be 2.70%, 2.68%, 2.70%, and 2.60% of diets, respectively.

3.2.
Intestinal Digestive Enzyme Activities. The activities of intestinal trypsin and amylase were increased linearly with the increasing dietary lysine level, and those were significantly higher in the 3.08%, 3.56%, and 4.36% lysine groups compared to the 1.10% lysine group (P < 0:05; Table 5). However, the intestinal lipase activity improved with the increasing dietary lysine level up to 2.30% and decreased thereafter.
3.3. Plasma Biochemical Indexes. The plasma TP content was enhanced with the increasing dietary lysine level up to 1.69% and plateaued thereafter (Table 6). Conversely, the BUN content significantly depressed with the increasing dietary lysine level up to 2.30% and increased afterward. However, no significant differences were observed in the plasma TAA content, AST, and ALT activities as well as AST/ALT ratio among the treatment groups (P > 0:05).

Lysine
Metabolism-Related Enzyme Activities. The hepatic LKR activity was significantly enhanced with the increasing dietary lysine level up to 1.69% and remained constant in the 2.30% and 3.08% lysine groups and then significantly reduced in the 3.56% and 4.36% lysine groups (P < 0:05; Table 7). No significant differences were found in the activities of LKR in muscle as well as AASS in the liver and muscle among the dietary treatments (P > 0:05).

Protein
Metabolism-Related Enzyme Activities. The hepatic AMPD activity was generally reduced with the increasing dietary lysine level, which was significantly higher in the 1.10% lysine group compared to the other groups (P < 0:05; Table 8). However, no significant differences were found in the activities of GDH in the liver as well as GDH and AMPD in the muscle among the treatment groups (P > 0:05).

Relative Expression Levels of Mammalian Target of Rapamycin (mTOR) and Amino Acid Response (AAR)
Pathway-Related Gene in the Liver. The relative expression level of hepatic TOR was upregulated with the increasing dietary lysine level up to 1.69% and plateaued thereafter, which was significantly lower in the 1.10% lysine group compared to the 1.69%, 2.30%, and 4.36% lysine groups (P < 0:05; Figure 1). The relative expression level of hepatic 4E-BP2 was downregulated with the increasing dietary lysine level up to 1.69% and then upregulated thereafter. Conversely, the relative expression of hepatic S6K1 was upregulated with the increasing dietary lysine level up to 1.69% and then downregulated afterward, which was significantly higher in the 1.69% lysine group compared to the 1.10%, 2.30%, 3.56%, and 4.36% lysine groups (P < 0:05).
The relative expression levels of GCN2, ATF3, and ATF4b in the liver were generally downregulated with the increasing dietary lysine level ( Figure 2); thereinto, the relative expression level of GCN2 was significantly higher in the 1.10% lysine group compared to the 2.3%, 3.56%, and 4.36% lysine groups, and the relative expression levels of ATF3 and ATF4b were significantly higher in the 1.10% lysine group compared to the other groups (P < 0:05). The relative expression level of ATF4a was downregulated with the 5 Aquaculture Nutrition increasing dietary lysine level up to 2.30% and then upregulated thereafter, which was significantly lower in the 1.69%, 2.30%, and 3.08% lysine groups compared to the 1.10% lysine group (P < 0:05).

3.7.
Whole-Body Composition. The whole-body crude protein content was generally enhanced with the increasing dietary lysine level up to 3.08% and declined afterward (Table 9). However, the whole-body crude lipid content was significantly higher in the 1.10% lysine group compared to the other groups (P < 0:05). No significant differences were found in the whole-body moisture and crude ash contents among the treatment groups (P > 0:05).
The whole-body lysine, arginine, isoleucine, leucine, aspartate, alanine, serine, and total amino acid contents were firstly increased and then decreased with the raising dietary lysine level, and the highest values were observed in the 3.08% lysine group (Table 9). Based on the broken-line and SOP regressions between the dietary lysine level and the whole-body lysine content, the optimal lysine requirement of juvenile P. leopardus was estimated to be 2.82% and 2.97% of diets, respectively (Figure 3).

Discussion
As one of the essential amino acids, lysine plays an important role in the normal growth and development of fish. The deficiency of lysine will lead to a decline in appetite and feed intake, which is not conducive to the growth and development of fish [10,11]. However, excessive lysine also Values are means ± standard error (SE) of three replications. Means with different superscript letter(s) in the same row are significantly different (P < 0:05). SOP = second-order polynomial trend; L = linear trend; OIDL = optimal inclusion of dietary lysine level. Values are means ± standard error (SE) of three replications. Means with different superscript letter(s) in the same row are significantly different (P < 0:05). TP = total protein; TAA = total amino acids; BUN = blood urea nitrogen; AST = aspartate aminotransferase; ALT = alanine aminotransferase; SOP = secondorder polynomial trend; NR = no relationship; OIDL = optimal inclusion of dietary lysine level. 7 Aquaculture Nutrition has negative effects on the growth and health of fish [5,6] as it disrupts the balance of amino acids and prevents the absorption and utilization of other amino acids [7]. According to NRC [12], the lysine requirements of freshwater and marine fishes are 1.6-2.4% and 1.7-2.8%, respectively. In this study, dietary insufficient (1.10% and 1.69%) or excessive (3.56% and 4.36%) lysine retarded the growth rate and feed utilization, which reached the highest values in the 2.30-3.08% lysine groups. The lower growth rate of P. leopardus fed below the optimum dietary lysine may be due to the decline in lysine intake, while the depressed growth of fish fed diets with excessive lysine may be related to the lysine-arginine antagonism or the imbalanced amino acids. The dose-response trend is consistent with the finding for orange-spotted grouper (Epinephelus coioides) [13]. Based on the SOP regression model for growth rate and feed efficiency, the optimal dietary lysine requirement of P. leopardus was 2.60-2.68% of diet (4.91-5.06% of crude protein). The value was similar to that for juvenile orange-spotted grouper (2.83-3.04% of diet) [13,14]. Nevertheless, it was   [15], but lower than that for juvenile humpback grouper (Cromileptes altivelis) (3.99% of diet) [16]. Additionally, the previous study also showed that the dietary lysine requirements based on SGR were recommended as 3.04% and 2.61% of diets for juvenile (initial body weight 22.07 g) and subadult (initial body weight 102.51 g) orange-spotted grouper, respectively [13]. Consequently, the differences in dietary lysine requirements among these studies are possibly due to the species and size of groupers [17]. Additionally, body index (CF, VSI, HSI, and ISI) is also an important parameter reflecting the growth and health of fish.
In this study, the CF was linearly increased with the raising dietary lysine level, which indicates that the increased dietary lysine level resulted in the elevation of obesity degree. The positive correlation between CF and dietary lysine level was also observed in silver perch (Bidyanus bidyanus) [18], Chinese sucker (Myxocyprinus asiaticus) [19], dusky kob (Argyrosomus japonicus) [20], and Pseudobagrus ussuriensis [21]. Additionally, the present and previous studies [14] showed that fish fed diets with optimal lysine exhibited higher VSI and wholebody protein content, which indicates that the balanced amino acids promoted the protein deposition of visceral organ.
As is known to all, digestive enzymes play a key role in the digestibility and utilization of nutrients [22]. A previous    Aquaculture Nutrition study with Jian carp (Cyprinus carpio var. Jian) showed that the activities of intestinal protease and lipase significantly improved with the increase of dietary lysine level [23]. A similar observation was found in this study, and the activities of intestinal trypsin and amylase linearly were enhanced with the increasing dietary lysine level. The positive effects of dietary lysine on the intestinal digestive enzymes' activities could be explained by different assumptions: lysine has the ability to stimulate the synthesis of these enzymes in the pancreas and/ or lysine might affect the secretion and release of these enzymes in fish [24]. Additionally, the optimum dietary lysine significantly improved the intestinal lipase activity of P. leopardus, which was consistent with the results in large yellow croaker (Pseudosciaena crocea) [25] and grass carp (Ctenopharyngodon idellus) [26]. However, the intestinal lipase activity was depressed by excessive dietary lysine, indicating the digestive function of P. leopardus was adversely affected by the imbalance of dietary amino acids. Previous studies have also confirmed that excessive lysine level might inhibit lipase activity [24,27]. Hence, appropriate dietary lysine enhanced digestive enzyme activities and thereby might improve the digestion and growth of P. leopardus. Protein metabolism is a dynamic process involving the balance between the synthesis and degradation of protein [28]. It is well known that all amino acids are needed for protein synthesis, and a deficiency in one of these amino acids may diminish the protein synthetic response [29]. The TOR is a serine/threonine kinase, which regulates protein synthesis [30]. There are two structurally and functionally distinct TOR complexes (TORC). TORC1 is sensitive to rapamycin and nutrients, while TORC2 is insensitive [31,32]. After TORC1 was activated, the downstream main-acting factors 4EBPs and S6K were phosphorylated to activate protein translation [33][34][35]. In this study, dietary lysine inclusion of 1.69-2.30% upregulated the relative expression levels of TOR and S6K1 but downregulated the relative expression level of 4E-BP2, indicating that the optimum dietary lysine activated the mTOR signaling pathway and thereby promoted the protein synthesis in P. leopardus. Similarly, the optimum dietary lysine significantly upregulated the relative expression levels of hepatic TOR in juvenile Hemibagrus wyckioides [36] and hepatic TOR and S6K1 in hybrid grouper [15] and upregulated the relative expression level of intestinal TOR but downregulated the relative expression level of intestinal 4E-BP in grass carp [26]. Therefore, it is believed that a balanced dietary amino acid profile plays an important role in sustaining efficient protein synthesis [37].

Aquaculture Nutrition
On the other hand, ammonia nitrogen is a product of protein catabolism, which reflects the ability to use amino acids as energy source. In this study, optimal dietary lysine inclusion increased the plasma TP content but depressed the BUN content, indicating that protein utilization was improved in P. leopardus fed diets with 2.30-3.08% lysine. A similar observation was found in a previous study with dace (Leuciscus brandti) [38]. Lysine is converted to saccharopine in the liver mitochondria, and the catabolism of saccharopine is catalyzed by LKR and saccharopine dehydrogenase [39]. AASS is a bifunctional enzyme composed of both LKR and saccharopine dehydrogenase domains [40,41]. When dietary lysine is limited, the degradation of lysine is reduced so that lysine can be preserved in protein synthesis [42]. Accordingly, the hepatic LKR and AASS activities in dace were significantly increased with the increasing dietary lysine level [38]. A similar result was observed in this study, and the hepatic LKR activity was significantly increased by dietary 1.69%-3.08% lysine inclusion. However, the activity and mRNA level of hepatic LKR in rainbow trout (Oncorhynchus mykiss) were not affected by dietary lysine level [43,44]. GDH and AMPD are the key rate-limiting enzymes in the combined amino acid deamination reaction [45], which are usually considered evaluation indicators for using amino acids as an energy source [46]. In this study, the hepatic AMPD activity was linearly depressed with the increase of dietary lysine level, indicating that dietary lysine inclusion was conducive to reducing protein catabolism and thereby improving protein deposition. In the study of juvenile Hemibagrus wyckioides, optimal lysine also downregulated the relative expression levels of GDH in the liver and AMPD in the liver and muscle [36]. Additionally, dietary protein or amino acid deficiency will trigger a series of signal processes, collectively referred to as AAR [47,48]. GCN2 is an essential regulator of cellular response to amino acid limitation in all tissues [35]. When GCN2 kinase is activated in the absence of amino acids, its downstream ATF3 and ATF4 are also activated, thus inhibiting the initiation of the translation of protein in cells and reducing overall protein synthesis [49]. The present study showed that the relative expression levels of GCN2, ATF3, ATF4a, and ATF4b in the liver of P. leopardus were downregulated by dietary inclusion of 2.30% lysine, suggesting that dietary suitable lysine level and balanced amino acids inhibited the AAR signaling pathway and were more conducive to protein synthesis.
Appropriate dietary lysine level is conducive to the utilization of amino acids and the synthesis of body protein. The present study showed that dietary 3.08% lysine improved the whole-body crude protein content but depressed the wholebody lipid content of juvenile P. leopardus. Similarly, previous studies also showed that appropriate dietary lysine reduced body lipid content but increased body protein content of channel catfish (Ictalurus punctatus) [50], rainbow trout [51,52], Japanese seabass (Lateolabrax japonicus) [53], large yellow croaker [54], silver perch [18], and tilapia (Oreochromis niloticus) [55]. Meanwhile, lysine is also a precursor of carnitine, and carnitine involves in the entry of long-chain fatty acyl groups into the mitochondria for β-oxidation [43]. Hence, the lower protein content and higher lipid content in fish fed diets with scarce lysine may be explained by reduced β-oxidation of fatty acids, resulting in more utilization of protein rather than lipid as an energy source [56,57]. Additionally, the changing trend of the whole-body amino acid composition is consistent with the crude protein content. The contents of lysine, arginine, isoleucine, leucine, aspartate, alanine, and serine as well as total amino acids increased with the increasing dietary lysine level up to 3.08% and declined or remained constant afterward. A balanced dietary amino acid may increase the retention of amino acids, thus increasing protein retention. The lack and excess of dietary lysine may increase the consumption of amino acids as energy while decreasing the amino acids used for protein synthesis [37]. Likewise, Mai et al. [53] 11 Aquaculture Nutrition found that dietary lysine levels significantly impacted the amino acid contents in the muscle of Japanese bass, which was corresponding to the results of tiger puffer [55]. In the study with tilapia, the whole-body essential amino acid compositions were not changed with various dietary lysine levels, whereas the nonessential amino acid contents were increased [56]. However, some studies have shown that dietary lysine level had no effects on the whole-body composition [13,58,59] and amino acid profile [59,60] of fish. The discrepancy is possibly due to the differences in the feed composition, sensitivity, and tolerance to amino acids of different fish species.

Conclusion
Based on the SOP regression model, the optimal dietary lysine requirement of P. leopardus was 2.60%-2.97% of diets (4.91%-5.60% of diet protein) for growth performance, feed utilization, and lysine deposition. The optimal dietary lysine promoted the digestive enzymes' activities and protein synthesis, but depressed protein degradation thereby enhanced protein deposition and growth performance of juvenile P. leopardus. However, deficiency and excess dietary lysine resulted in amino acid imbalance and thereby repressed the growth of juvenile P. leopardus.

Data Availability
The data used to support the findings of this study are available from the corresponding authors upon request.

Conflicts of Interest
The authors declare no competing interests.

Authors' Contributions
Xiaomei Dou carried out the experiment, sample collection and determination, data analysis, and writing of the original draft, and wrote, reviewed, and edited the manuscript. Yu Liu, Yixiong Cao, Yumeng Zhang, and Xinlangji Fu were responsible for the conceptualization, methodology, and sample collection. Junming Deng and Beiping Tan were responsible for the funding acquisition and supervision and wrote, reviewed, and edited the manuscript. The manuscript was written or revised by all authors.