Metabolic Fate Is Defined by Amino Acid Nature in Gilthead Seabream Fed Different Diet Formulations

Simple Summary The maximisation of fish growth depends on amino acids’ availability in tissues at an optimum ratio since imbalances may lead to their utilisation for energy rather than growth. Amino acids may be catabolised in multiple pathways and be classified according to their metabolic fate: ketogenic and glucogenic. Ketogenic amino acids (e.g., lysine) are precursors of ketone bodies or long chain fatty acids and can be used in lipogenesis. Glucogenic amino acids (e.g., methionine) can be converted into glucose through gluconeogenesis. Some amino acids, such as tryptophan, can be ketogenic and glucogenic. This study aimed to evaluate how fish discriminate among different amino acids when they are fed different diets to attain the best utilisation of the feed. This trial was carried out in gilthead seabream juveniles fed experimental diets containing different levels of protein and/or distinct lipid levels. The metabolic fate and bioavailability of the indispensable amino acids lysine, methionine, and tryptophan were defined by their ketogenic and/or glucogenic nature rather than diet formulations. The optimisation of diets that consider the amino acids’ bioavailability will maximise protein retention in fish and is a viable solution to develop cost-effective fish diets. Abstract The sustainability of the Aquaculture industry relies on optimising diets to promote nitrogen retention and maximise fish growth. The aim of this study was to assess how different dietary formulations influence the bioavailability and metabolic fate of distinct amino acids in gilthead seabream juveniles. Amino acids (lysine, tryptophan, and methionine) were selected based on their ketogenic and/or glucogenic nature. Seabream were fed practical diets with different protein (44 and 40%) and lipid contents (21 and 18%): 44P21L, 44P18L, 40P21L, and 40P18L. After three weeks of feeding, the fish were tube-fed the correspondent diet labelled with 14C-lysine, 14C-tryptophan, or 14C-methionine. The amino acid utilisation was determined based on the evacuation, retention in gut, liver, and muscle, and the catabolism of the tracer. The metabolic fate of amino acids was mainly determined by their nature. Tryptophan was significantly more evacuated than lysine or methionine, indicating a lower availability for metabolic purposes. Methionine was more retained in muscle, indicating its higher availability. Lysine was mainly catabolised, suggesting that catabolism is preferentially ketogenic, even when this amino acid is deficient in diets. This study underpins the importance of optimising diets considering the amino acids’ bioavailability and metabolic fate to maximise protein retention in fish.


Introduction
The sustainability of the aquaculture sector is largely dependent on optimised diets that promote nitrogen retention and maximise fish growth performance. Growth is essentially protein deposition and improved growth requires the knowledge of the ideal fishmeal and incorporated 51-54% plant ingredients as protein sources. The lower protein diets (40P21L and 40P18L) contained 21% fishmeal and 57-60% plant protein sources. Fish oil to rapeseed oil ratio was kept at approximately 1.5 to 1.0 in all diets. Higher protein diets (44P21L and 44P18L) were formulated to meet lysine requirements for gilthead seabream juveniles, while lower protein diets (40P21L and 40P18L) were formulated to be deficient in lysine. Formulation, proximate composition, and amino acid analysis of diets are presented in Tables 1 and 2.  All experimental diets were manufactured by SPAROS Lda. (Olhão, Portugal). Diets (pellet size 2 mm) were produced via extrusion by means of a pilot-scale twin-screw extruder (CLEXTRAL BC45; Clextral, Firminy, France) with a screw diameter of 55.5 mm and temperature ranging from 105 • C to 110 • C. Upon extrusion, all batches of extruded feeds were dried in a vibrating fluid bed dryer (model DR100; TGC Extrusion, Roullet-Saint-Estèphe, France) and cooled at room temperature. Subsequently, the oil fraction was added under vacuum coating in a Pegasus vacuum mixer (PG-10VCLAB; DINNISSEN, Sevenum, The Netherlands). Experimental diets were kept in a cool and aerated storage room.

Fish Husbandry
Throughout the conditioning feeding period, gilthead seabream (Sparus aurata) juveniles were maintained at the Centre of Marine Sciences (CCMAR, Faro, Portugal). Fish were reared in 40 L cylinder-conical tanks in a recirculation aquaculture system at an initial density of 4.8 kg m −3 (45 fish per tank). Water temperature and salinity were 20.1 ± 1.2 • C and 32.2 ± 1.4 ppt, respectively. Fish were assigned one of the four experimental diets and fed by automatic feeders, six times per day at a feeding ration of 3% body weight day −1 , for three weeks (six meals of 0.5% body weight per day).

Metabolic Flux Assays
The metabolic fate of the indispensable amino acids as a function of their nature (ketogenic and/or glucogenic) and of the dietary treatment was determined after three weeks of feeding the experimental diets. The metabolic flux assays were performed according to the methodology previously published in detail by Teodósio et al. [18]. This method was adapted from Costas [19], which was modified for juvenile fish from the procedure described by Rønnestad et al. [20], a modification from the original method published by Rust et al. [21].
Briefly, after being fasted for 24 h, anaesthetised fish (50 mg L −1 2-phenoxyethanol, Merck KGaA, Darmstadt, Germany) from each dietary treatment (44P21L, 44P18L, 40P21L, and 40P18L) were tube-fed with the correspondent experimental diet labelled with one of the following tracers: 14  ). Fish were tube-fed the number of pellets corresponding to a single meal (approximately 0.5% of fish body weight). The tube-fed tracer presented no nutritional value and its metabolic fate was considered to represent the fate of the tracee of the diets [22]. Six fish per diet and tracer (average weight of 9.1 ± 2.9 g; mean ± standard deviation) were subjected to this procedure (6 fish × 4 diets × 3 tracers, n = 72 fish in total).
Tube-fed fish were transferred into individual incubation chambers containing 2 L of seawater at 20 • C. Chambers were hermitically sealed and each individual chamber was connected to a series of 14 CO 2 -metabolic traps to collect the 14 CO 2 produced by the fish from catabolism of 14 C-amino acids. After 18 h, fish were euthanised inside the metabolic chamber by a lethal dose of anaesthetic (1000 mg L −1 of MS-222 buffered with sodium bicarbonate) and removed for tissue sampling. After fish were removed, the chambers were resealed, and the incubation seawater was acidified gradually leading to any remaining 14 CO 2 in the seawater to be trapped in the 14 CO 2 -metabolic traps. The acidification procedure published by Rønnestad et al. [20] allowed for discrimination between unabsorbed nutrients evacuated from the gut and molecules originating from catabolism of the absorbed nutrient, both of which were present in the incubation water.

Metabolic Budget Determination
After acidification of the incubation water, seawater samples from each chamber (n = 5) and individual samples from each 14 CO 2 -metabolic trap (n = 3) were collected for radioactivity (disintegrations per minute, DPM) determination. The amount of radioactivity present in the seawater resulted from evacuated (unabsorbed) 14 C-amino acids or from a negligible amount of other metabolites containing the 14 C-skeleton, while the radioactivity present in the 14 CO 2 -metabolic traps resulted from 14 C-amino acid catabolism.
From each fish, the gastrointestinal tract (gut, from the oesophagus to the hindgut), whole liver, and whole left side skin-on fillet (muscle) were sampled to assess 14 C-amino acid retention in the body fractions. Tissues sampled in this trial represented between 40-45% of the total weight of fish. Fish digestive tract was previously washed in Ringer solution for marine fish to ensure that no alimentary bolus was present before DPM determination. All fish tissues were fully dissolved by adding an appropriate volume of Solvable™ (Perkin Elmer) at 50 • C for 24 h. Gut and liver were analysed as whole whereas two samples of dorsal muscle were used for DPM counting. Ultima Gold XR scintillation cocktail (Perkin Elmer) was added to all samples (incubation seawater, 14 CO 2 -metabolic traps, gut, liver, and muscle fractions) and DPM determined in a TriCarb 2910TR low activity liquid scintillation analyser (Perkin Elmer). Samples were corrected for quench and lumex.
Amino acid utilisation in seabream juveniles as a function of the amino acids' nature and diet formulation was determined based on the percentage of 14 C-amino acid evacuated, retained in the different tissues, or catabolised, as follows: where DPM Total is the sum of the radioactivity (DPM) determined in the incubation seawater (DPM SW ), gut (DPM Gut ), liver (DPM Liver ), muscle (DPM Muscle ), and 14 CO 2 -metabolic traps (DPM Traps ) fractions.

Chemical Analysis
Experimental diets were finely ground and analysed in duplicate for dry matter, ash, crude protein (N × 6.25), crude lipid, gross energy, and phosphorus contents, following standard procedures of the Association of Official Analytical Chemists [23]. Diets' total amino acid profile was determined by ultra-high-performance liquid chromatography (UPLC), after acid hydrolysis. Tryptophan content was not determined since it Animals 2022, 12, 1713 6 of 14 is partially destroyed by acid hydrolysis. All analyses were performed as described by Teodósio et al. [24].

Data Analysis
Data from the metabolic flux assays are presented as mean ± standard deviation. Data expressed as a percentage were arcsine square root transformed prior to the statistical analysis [25]. All data were checked for normal distribution and homogeneity of variances. The experimental design was randomised in a 4 × 3 factorial design with four dietary treatments (44P21L, 44P18L, 40P21L, and 40P18L) and three dietary components (lysine, tryptophan, and methionine). Main and interaction effects were identified by two-way analysis of variance (two-way ANOVA) followed by Tukey's multiple-comparison test at p < 0.05 level of significance. Statistical analyses were performed using the IBM SPSS Statistics 26 software (IBM Corp, Armonk, NY, USA).
Additionally, a principal component analysis (PCA) was performed to confirm differences between diet formulations and amino acids' nature and find potential clusters of observations. The standard prcomp R function in the auto-scaled matrices was used for PCA and score plots were produced for the two first principal components (PC1 and PC2) using the ggbiplot package for R. Loadings for PC1 and PC2 were calculated to determine the weight of each original variable in the corresponding PCs. All analyses were carried out using the open-source software R version 4.0.4 (R Core team).
To assess any dietary amino acid imbalances of the experimental diets, dietary A/E ratios [26] were calculated, on a weight basis, as: [each indispensable amino acid (IAA) content × (total IAA content) −1 × 1000], and plotted against previously published A/E ratios for gilthead seabream juveniles [27]. Cysteine and tyrosine were included with the IAA, since they can only be synthesised from methionine and phenylalanine, respectively. A deficiency or excess for a given IAA calculated as: × 100, was assumed to occur when the dietary A/E ratio was at least 10% lower or higher than that of the fish.

Results
In all the dietary treatments, the proportion of the tube-fed 14 C-tryptophan that was evacuated was significantly higher compared to 14 C-lysine or 14 C-methionine (Figure 1;  see supplementary Table S1 for more details). The evacuated 14 C-tryptophan that was recovered in the incubation seawater varied from 57% to 67% for the fish that were fed the 40P18L and 44P21L diets, respectively. The proportion of 14 C-lysine that was evacuated ranged from 21% in fish that were fed the 44P21L diet to 33% in 44P18L fed fish. Concerning 14 C-methionine evacuation, the proportion that was not absorbed varied from 27% to 42% in the fish that were fed the 44P18L and 44P21L diets, respectively. No significant differences were detected between lysine and methionine evacuation in fish fed all the experimental diets. The assessment of 14 C-amino acid utilisation in gilthead seabream juveniles that were fed experimental diets with distinctive dietary protein and/or lipid content showed that the dietary treatments (44P21L, 44P18L, 40P21L, and 40P18L) had no influence on the evacuation of lysine, tryptophan, and methionine, but rather strongly dependent on their ketogenic and/or glucogenic nature.
The amino acid retention in the gut (Figure 2a) was affected by the diet formulations, although post-hoc tests did not determine which treatments were significantly different. In general, fish that were fed the 40P21L and the 40P18L diets presented higher retention values for all amino acids than the fish that were fed the high protein diets. Additionally, the nature of the amino acid influenced their retention in this tissue. Lysine was significantly more retained than methionine in all dietary treatments. Tryptophan retention in the gut was similar to lysine or methionine.
27% to 42% in the fish that were fed the 44P18L and 44P21L diets, respectively. No significant differences were detected between lysine and methionine evacuation in fish fed all the experimental diets. The assessment of 14 C-amino acid utilisation in gilthead seabream juveniles that were fed experimental diets with distinctive dietary protein and/or lipid content showed that the dietary treatments (44P21L, 44P18L, 40P21L, and 40P18L) had no influence on the evacuation of lysine, tryptophan, and methionine, but rather strongly dependent on their ketogenic and/or glucogenic nature. The amino acid retention in the gut (Figure 2a) was affected by the diet formulations, although post-hoc tests did not determine which treatments were significantly different. In general, fish that were fed the 40P21L and the 40P18L diets presented higher retention values for all amino acids than the fish that were fed the high protein diets. Additionally, the nature of the amino acid influenced their retention in this tissue. Lysine was significantly more retained than methionine in all dietary treatments. Tryptophan retention in the gut was similar to lysine or methionine.  The retention of the amino acids in the liver was influenced both by the amino acid nature and by the diet formulation (Figure 2b). Fish that were fed a higher amount of dietary protein (44P21L and 44P18L) presented lower amino acid retention in the liver, but only significantly different when compared to the diet with less protein and lipids  The retention of the amino acids in the liver was influenced both by the amino acid nature and by the diet formulation (Figure 2b). Fish that were fed a higher amount of dietary protein (44P21L and 44P18L) presented lower amino acid retention in the liver, but only significantly different when compared to the diet with less protein and lipids (40P18L). Concerning the nature of the amino acids, lysine was significantly more retained in the liver than tryptophan, while methionine presented intermediate levels.
Methionine was preferentially retained in the muscle (Figure 2c). Although no significant differences were found among fish fed the different diets, the proportion of 14 Cmethionine retained in the muscle ranged from 30% in fish fed the 44P21L to 40% in 40P18L fed fish. Moreover, the retention of methionine in the muscle was significantly higher than lysine and tryptophan that presented mean retention values of 26% and 8%, respectively, for all dietary treatments.
The amino acids' nature significantly influenced their catabolism, independently of the diet. Lysine was significantly more catabolised than tryptophan and methionine (Figure 2d). Catabolised lysine varied from approximately 21% in fish that were fed the 40P18L diet to 31% in 44P21L fed fish. The percentage of tube-fed 14 C-tryptophan and 14 C-methionine that was catabolised was around 12% and 14%, respectively, for all dietary treatments.
A principal component analysis (PCA) was used to reduce the complexity of the data from the metabolic flux assays. The PCA confirmed that the nature of the amino acids, not the diet formulations, was responsible for the observed differences ( Figure 3). PC1 and PC2 accounted for 49.5% and 22.7% of the total variability of the data. The analysis of the score plots indicated that the tryptophan data were separated from lysine and methionine along the PC1 axis, and that lysine and methionine were separated from each other along the PC2 axis. Evacuation and retention in the liver were the loadings that contributed the most for the dissimilarities observed between fish that were fed 14 C-tryptophan labelled diets and fish that were fed diets labelled either with 14 C-lysine or 14 C-methionine. On the other hand, catabolism and muscle retention were the loadings responsible for the differences between fish that were fed 14 C-lysine labelled diets and fish that were fed diets labelled with 14 C-methionine. A principal component analysis (PCA) was used to reduce the complexity of the data from the metabolic flux assays. The PCA confirmed that the nature of the amino acids, not the diet formulations, was responsible for the observed differences ( Figure 3). PC1 and PC2 accounted for 49.5% and 22.7% of the total variability of the data. The analysis of the score plots indicated that the tryptophan data were separated from lysine and methionine along the PC1 axis, and that lysine and methionine were separated from each other along the PC2 axis. Evacuation and retention in the liver were the loadings that contributed the most for the dissimilarities observed between fish that were fed 14 C-tryptophan labelled diets and fish that were fed diets labelled either with 14 C-lysine or 14 C-methionine. On the other hand, catabolism and muscle retention were the loadings responsible for the differences between fish that were fed 14 C-lysine labelled diets and fish that were fed diets labelled with 14 C-methionine. Although the diets were supplemented with crystalline indispensable amino acids, a comparison of the dietary A/E ratios with the A/E ratios of fish revealed that all diets presented some amino acid imbalances (Figure 4). Lysine was balanced in the high protein diets 44P21L and 44P18L (Figure 4a,b); however, the lower protein diets, 40P21L and 40P18L (Figure 4c,d), presented a deficiency of 20% and 24%, respectively. Phenylalanine Although the diets were supplemented with crystalline indispensable amino acids, a comparison of the dietary A/E ratios with the A/E ratios of fish revealed that all diets presented some amino acid imbalances (Figure 4). Lysine was balanced in the high protein diets 44P21L and 44P18L (Figure 4a,b); however, the lower protein diets, 40P21L and 40P18L (Figure 4c,d), presented a deficiency of 20% and 24%, respectively. Phenylalanine and tyrosine were found to be in excess in all diets (102% to 112%), while methionine and cysteine were deficient (~20% in the case of the higher protein diets and 14% for diets with lower protein content).

Discussion
Metabolic flux assays for 14 C-lysine, 14 C-tryptophan, or 14 C-methionine were used to assess the bioavailability and metabolic fate of the selected indispensable amino acids as a function of their nature (ketogenic and/or glucogenic) and of the different dietary formulations in gilthead seabream juveniles. The metabolic budget of the selected amino acids was mainly affected by their nature rather than the diet formulation.
A principal component analysis (PCA) confirmed that the data were clustered by amino acid and not by dietary treatment. Additionally, the evacuation was the variable that contributed the most for the formation of two groups along the PC1 axis, one clustering observations from fish fed all diets labelled with 14 C-tryptophan and another that clustered data from fish fed 14 C-lysine and 14 C-methionine labelled diets. In the current study, tryptophan evacuation was substantially higher than the evacuation of lysine and methi-

Discussion
Metabolic flux assays for 14 C-lysine, 14 C-tryptophan, or 14 C-methionine were used to assess the bioavailability and metabolic fate of the selected indispensable amino acids as a function of their nature (ketogenic and/or glucogenic) and of the different dietary formulations in gilthead seabream juveniles. The metabolic budget of the selected amino acids was mainly affected by their nature rather than the diet formulation.
A principal component analysis (PCA) confirmed that the data were clustered by amino acid and not by dietary treatment. Additionally, the evacuation was the variable that contributed the most for the formation of two groups along the PC1 axis, one clustering observations from fish fed all diets labelled with 14 C-tryptophan and another that clustered data from fish fed 14 C-lysine and 14 C-methionine labelled diets. In the current study, tryptophan evacuation was substantially higher than the evacuation of lysine and methionine (mean values of 60% versus 26% and 32%, respectively). Similarly, the metabolic budget determination in Senegalese sole (Solea senegalensis) juveniles and in white seabream (Diplodus sargus) larvae also revealed that a higher proportion of 14 C-tryptophan was evacuated (41% and 30%, respectively) when compared to all the other indispensable amino acids [19] or to methionine and arginine [28]. The data on tryptophan digestibility is limited, yet juvenile rainbow trout (Oncorhynchus mykiss) showed a lower digestibility of tryptophan compared to all the other indispensable amino acids independently of the dietary treatment [29]. Free amino acids are transported across the brush-border membrane through carrier-mediated transport systems [4,30]. The same transporter may be responsible for the uptake of several amino acids and consequently some amino acids may interfere with the intestinal uptake of others [9,31]. TAT1 (Slc16a10) is responsible for the transport of the aromatic amino acids (phenylalanine, tyrosine, and tryptophan) in animal cells [32,33]. The calculation of the A/E ratios showed that in the present study all the experimental diets presented a phenylalanine and tyrosine content above the fish requirements. Therefore, it is reasonable to assume that a dietary excess of the two amino acids may interfere with the uptake of tryptophan since they all compete for the same transporter. The high levels of phenylalanine and tyrosine present in all the experimental diets might have potentially increased the evacuation of tryptophan resulting in its low availability for metabolic purposes. The current findings highlight the importance of feeding fish with balanced amino acid diets since not only deficiencies but also surpluses may have implications for amino acid absorption and utilisation.
The proportion of amino acids retained in the gut and liver of the seabream juveniles was affected not only by their nature but also by their dietary treatments. Fish that were fed the 40P18L diet retained more amino acids in the liver than fish that were fed both high protein diets (44P21L and 44P18L). These results suggest that when fish are fed the lower protein diets (40P21L and 40P18L), amino acids may be temporarily retained in the liver to ensure its availability for metabolic purposes. Furthermore, 18 h after feeding, lysine, which is considered an indicator of protein synthesis, was one of the most retained amino acids in the gut and liver of seabream juveniles. The relatively high retention in these tissues may be associated with the constant cellular turnover occurring in the former and with the crucial role of the latter in energy homeostasis [1]. In general, the tryptophan retention in the tissues was low: 13%, 6%, and 8% was found in the gut, liver, and muscle of juvenile seabream, respectively. It is worth mentioning that most of the retained tryptophan (±48%) was found in the gut, which was also observed in Senegalese sole juveniles [19]. This may be due to tryptophan being the precursor of 5-hydroxytryptophan, which in turn is the precursor of serotonin, a hormone mainly synthesised in the gastrointestinal tract [11,34], or due to the beneficial effects of tryptophan in the intestinal structural integrity of fish [35].
The present study showed that most of the absorbed methionine was retained in the muscle. This agrees with the results obtained with tracer studies in Senegalese sole juveniles and larvae, as well as in white seabream larvae [19,28,36,37]. Moreover, the current work revealed that the retention of methionine in the muscle was significantly higher than lysine and tryptophan retention. The PCA results corroborate these findings, demonstrating that fish that were fed all diets labelled with 14 C-methionine were clustered together according to the data from the muscle retention. Dietary methionine supplementation was shown to improve growth performance in several fish species [16,38,39] and its involvement in muscle growth and in the regulation of the expression of genes related to myogenesis has been established [40]. Additionally, as a precursor of S-adenosylmethionine (SAM) and of polyamines, methionine is involved in DNA and protein methylation as well as in cell proliferation and differentiation [41,42]. The fact that methionine was preferentially retained in the muscle of fish that were fed all experimental diets reinforces the importance of methionine availability in this tissue for growth and other metabolic purposes. A recent study in seabream juveniles that were fed diets differing in the protein to energy ratios revealed that the higher retention of methionine in the muscle, when compared with protein or lysine, was due to its presence in the amino acid free pool and not in the protein-bound fraction, reinforcing the crucial role of methionine in various metabolic functions in addition to protein synthesis [18].
Inevitably, a fraction of the absorbed amino acids is catabolised and used for energy rather than growth [8,15]. This is especially true if imbalances in the dietary amino acid profile occur. Amino acid catabolism occurs mainly in the liver where amino acids are either converted into others or directed to the tricarboxylic acid (TCA) cycle. Once they enter the TCA cycle, amino acids can either be oxidised to generate energy, or channelled towards fatty acid synthesis or gluconeogenesis [9]. Although gluconeogenesis occurs in the liver, in fish, this pathway is thought to be less significant for amino acids compared to their oxidation [8,15,43]. Previous metabolic studies in gilthead seabream [44,45] and Senegalese sole [46] using a 14 C-amino acid mixture demonstrated that only a small percentage of absorbed amino acids were converted into lipids or other metabolites, where glucose was included. Accordingly, the current findings revealed that 25% of the 14 C-lysine, an amino acid that is solely ketogenic, was detected in the CO 2 -metabolic traps, significantly more than methionine (glucogenic) or tryptophan (ketogenic and glucogenic). Similarly, a metabolic budget determination in Senegalese sole juveniles revealed that lysine was significantly more catabolised than all the other indispensable amino acids [19]. According to the A/E ratio analysis, all experimental diets exhibited a deficit in sulphur amino acids (methionine + cysteine). Moreover, this analysis confirmed that the higher protein diets, 44P21L and 44P18L, met gilthead seabream's requirements for lysine, and the low protein diets, 40P21L and 40P18L, were deficient in this amino acid. Nevertheless, lysine was significantly more catabolised than the other amino acids, independently of the diet. The PCA results confirmed that the data from the catabolism of the fish that were fed all the diets labelled with 14 C-methionine and 14 C-lysine grouped these amino acids in two different clusters along the PC2 axis. Fish have the ability to regulate amino acid metabolism with various enzymes that are responsible for adjusting the differential use of individual amino acids [43]. In fact, several studies have demonstrated that selective amino acid retention and catabolism occurs in fish [28,[47][48][49]. The present results revealed that gilthead seabream juveniles discriminate between the use of different amino acids and that catabolism seems to be preferentially ketogenic.

Conclusions
In this study, the assessment of 14 C-amino acid utilisation in gilthead seabream juveniles that were fed different diet formulations showed that amino acids' bioavailability and metabolic fate were mainly determined by their nature. The higher tryptophan evacuation may be related to its affinity to intestinal transporters and interaction among different amino acids. Therefore, the low availability of tryptophan should be considered when optimising fish diet formulations. This in vivo approach allowed for the unravelling of interesting particularities in tryptophan metabolism, which are usually difficult to untangle in nutritional studies, mainly due to technical difficulties in quantifying tryptophan content. Methionine was found to be preferentially retained in the muscle, most likely to be used for growth. Catabolism was mainly ketogenic, independently of the diet and of any dietary amino acid imbalances. The in vivo nutrient flux approach is a valuable tool that allows for the fine-tuning of diet formulations, since the optimisation of diets considering the amino acids' bioavailability will maximise protein retention in fish.